Citation
Searching molecular landscapes for the evolution of primal catalysts : in vitro selection of DNA-based ribonucleases

Material Information

Title:
Searching molecular landscapes for the evolution of primal catalysts : in vitro selection of DNA-based ribonucleases
Creator:
Carrigan, Matthew A
Publication Date:
Language:
English
Physical Description:
xi, 184 leaves : ill. ; 29 cm.

Subjects

Subjects / Keywords:
Catalysts ( jstor )
Enzyme substrates ( jstor )
Enzymes ( jstor )
Kinetics ( jstor )
Lasers ( jstor )
Libraries ( jstor )
Molecules ( jstor )
Nucleotides ( jstor )
Polymerase chain reaction ( jstor )
Population distributions ( jstor )
Dissertations, Academic -- Medical Sciences--Neuroscience -- UF ( lcsh )
Medical Sciences--Neuroscience thesis, Ph.D ( lcsh )
Genre:
bibliography ( marcgt )
theses ( marcgt )
non-fiction ( marcgt )

Notes

Thesis:
Thesis (Ph.D.)--University of Florida, 2002.
Bibliography:
Includes bibliographical references.
General Note:
Printout.
General Note:
Vita.
Statement of Responsibility:
by Matthew A. Carrigan.

Record Information

Source Institution:
University of Florida
Holding Location:
University of Florida
Rights Management:
Copyright [name of dissertation author]. Permission granted to the University of Florida to digitize, archive and distribute this item for non-profit research and educational purposes. Any reuse of this item in excess of fair use or other copyright exemptions requires permission of the copyright holder.
Resource Identifier:
029804347 ( ALEPH )
80511646 ( OCLC )

Downloads

This item has the following downloads:


Full Text











SEARCHING MOLECULAR LANDSCAPES FOR THE EVOLUTION OF PRIMAL CATALYSTS: IN VITRO SELECTION OF DNA-BASED RIBONUCLEASES















By

MATTHEW A. CARRIGAN















DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE
UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE
REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY

UNIVERSITY OF FLORIDA

2002























This work is dedicated to my son, Christian.















ACKNOWLEDGMENTS

I would like to thank my parents and family for their continuous support. I would like to thank several people for their early encouragement in my career in science, specifically my father, as well as Bill Crabtree, Tracy Bailey, Ken Cline, Ralph Henry, and Mike McCaffery. I am indebted to the scientific collaboration of Dr. Maury Swanson and his entire lab, specifically Carl Urbinati, Ron Hector, and Keith Nykamnp. Alonso Ricardo helped me tremendously to execute many of the experiments described in this dissertation. Mike Thomson offered invaluable and unending assistance. Without Steven Benner's patience and enthusiasm for science, this research endeavor would not have been possible.















TABLE OF CONTENTS

ACKNOWLEDGEMENTS ................................................................11i

ABBREVIATIONS ......................................................................... vii

ABSTRACT .................................................................................. ix

CHAPTER page

1 INTRODUCTION .........................................................................1I

Requirements of a "Living" SytmI...............
Chemical Challenges to the Prebiotic Synthesis of "Life"...............................4
Progress Towards Understanding the Genesis of Life:
Dual Function Biopolymers.................................................... 6
Progress Towards Understanding the Genesis of Life:
Prebiotic Synthesis of Potentially Useful Monomers and Polymers.............. 8
The Next Challenge: Obtaining Catalytic Function from Random Libraries 11...
Experimental Design..................................................................... 19


2 MATERIALS AND METHODS........................................................ 24

Preparation of Precursor DNAymes via PCR ......................................... 24
Preparation of Single-Stranded DNAzymes........................................... 26
5'-End Labeling of DNA................................................................ 28
DNAzyme Kinetic Assays .............................................................. 28
Cloning and Sequencing DNAzymes .................................................. 29
In Vitro Selection ........................................................................ 31

3 RESULTS OF 614 ANALYSIS......................................................... 36

Research Objectives ..................................................................... 36
Developing Methods for Preparing Single-stranded DNAzymes by
Exonuclease Degradation of 5'-Phosphorylated Complementary Strand ....37 Asymmetric PCR with "Tails," Followed by Gel Purification ................. 39
DNAzyme 614 with uncaged- and caged-ribose...................................... 41
Achieving maximal deprotection. of caged ribose ................................ 41
Kinetic profile of 614 with uncaged- and caged-ribose......................... 42



iv









Laser Does Not Damage Ribose-614 .............................................. 44
DNAzyme 614 is NaCi-dependent, and Mg9l2-independent................... 45
Understanding Reasons for Incomplete Cleavage of DNAzyme 614.............. 46
Incomplete De-protection of Caged-Ribose ...................................... 46
Incomplete Removal of the Complementary Strand.............................. 47
Approach to Chemical Equilibrium................................................ 48
Are the 27- and 79-nucleotide Fragments Acting as Catalysts or Inhibitors?. 50 Improperly Folded DNAzyme 614 ................................................. 51
Cloning and Sequencing Cleaved and Uncleaved 614 Near
Cleavage Plateau................................................................... 54
DNAzyme 614 Cleaves in cis and trans ............................................... 56
Cleavage Rate of Ribose-614 Varies with 614 Concentration ................. 58
Deoxyribose-614 Cleaves Various Ribose-Substrates........................... 62
Competition Studies of Ribose-614 Cleavage .................................... 63
d-614 Cleaves Faster in trans Than in cis: Single-turnover Kinetics ......... 64
d-614 Cleaves with Multiple-turnover............................................. 68
Testing 614 Rate of Association and Disassociation ................................. 68
Structural Analysis of 614 .............................................................. 74
Summary of Results ..................................................................... 81

4 STUDIES OF IN VITRO SELECTIONS............................................. 123

In Vitro Selection of Functionalized and Non-Functionalized Libraries
Using Caged-Ribose and Liquid-Phase Selection ................................... 123
In Vitro Selection .................................................................. 123
Testing Kinetics from Round 8 Pools Without Caged-ribose................. 127
Testing for Inadvertent Selection for Susceptibility to Laser ................. 127
Caged-ribose Does Not Induce Cleavage in trans............................... 128
Kinetic Analysis of Rounds 1-8 Pools for "H: ang + ribose"................. 128
Sequence Evolution During IVS .................................................. 131
Kinetic Analysis of Round 8 Clones.............................................. 133
Simulations of In Vitro Selection Experiments...................................... 134
Simulating In Vitro Selections .................................................... 135
Using Simulations to Determine the Impact of the Leakage (L)
Parameter on IVS ................................................................ 139
Examining the Impact of the Selection Time (St)................................ 145
Examining the Impact of the Rate of the Fastest Catalysts (kfastesr)........... 146
Examining the Impact of the Reactive Fraction................................. 147
Comparing Linearly and Exponentially Distributed Initial Populations ....148 Deciphering the Distribution Function for Catalysts........................... 149
Summary of Results .................................................................... 152

5 SUMMARY ............................................................................. 169

Catalytic Behavior of DNAzymes from IVS is More Complex
then Previously Assumed ............................................................ 170


v








Caged-ribose is Useful for Kinetics, But Increases Leakage
Too Much for IVS .............................................................................................. 173
IVS Isolates Sequence Variants of 614 with Wide Range of Catalytic Power ..... 174 Simulations Allow Estimation of Catalyst Distribution ....................................... 176
Nucleic Acid Catalysis and the Origins of Life .................................................... 177

REFERENCES ........................................................................................................... 179

BIOGRAPHICAL SKETCH ...................................................................................... 184









































vi














ABBREVIATIONS


IVS in vitro selection

bya billion years ago


AL Size of initial library

RF Reactive Fraction

NRF Non-reactive Fraction: NRF = SL*(l RE)

k. Intrinsic rate of molecule "n": each member of the initial library, n, is
assigned an intrinsic rate, k~, which relates to its ability to be cleaved
under selection conditions. The intrinsic rate is sequence-dependent and
therefore unchanging.

'Al1ife, The intrinsic rate of molecule "n" (Qs expressed in terms of half-life:
Y1ife., = 1n2 / nsoetThe rate constant for the slowest catalysts in the
library.

kfastest The rate constant of the fastest molecule in the library.

St Selection time: the amount of time allowed for a molecule to cleave itself
and therefore achieve "survival."

m or b Distribution function for reactive molecules. Two functions have been
tested thus far, a linear function with slope = in, and an exponential
function with decay rate = b.

L Leakage: the fraction of the pool that "survives" a round of selection
independent of St and intrinsic rate constant.










vii








caged-ribose a ribose-adenosine, with the 2'OH replaced with an ortho-nitrobenzyl
protecting group
r-primer 64ang + ribose" 49-nucleotide substrate containing ribose
r-X oligonucleotide "X" with ribose-adenosine
d-x oligonucleotide "X" with deoxy-adenosine

A Deoxyadenosine triphosphate
G Deoxyguanosine triphosphate
C Deoxycytidine tripbosphate
T Deoxythymidine triphosphate
E-base 5-(3-aminoallyl)-2'deoxyuridine triphosphate (see structure below)

0

H NH3

0

-0-ro-P-0-P-0-1;r


OH H





Vat(.i) chemical step for unimolecular catalysis
k 1 (uni) rate of folding for unimolecular catalysis k-l(uni) rate of unfolding for unimolecular catalysis
kcat(bi) chemical step for bimolecular catalysis
k I(bi) rate of association for bimolecular catalysis
k-I(bi) rate of disassociation for bimolecular catalysis

Birnolecular Kinetic Scheme:
kl(bi) kcat (bi)
E+S E-S E-P E = r-614 or d-614
k-l(bi) S = r-614, r-lib6lo2 or r-primer

Unimolecular Kinetic Scheme:
kl(uni) kcat (uni)
614u,,Iulueu -W- 614folded Products
k-,(uni)






viii















Abstract of Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy

SEARCHING MOLECULAR LANDSCAPES FOR THE EVOLUTION OF PRIMAL CATALYSTS: IN VITRO SELECTION OF DNA-BASED RIBONUCLEASES By

Matthew A. Carrigan

December 2002

Chair: Steven A. Benner
Cochair: Art Edison
Major Department: Neuroscience

Our current understanding of the origins of life suggests that catalytic nucleic

acids must arise from random polymers of nucleic acids. Searching random libraries of nucleic acid polymers has not produced the abundance and power of catalysis believed necessary to spawn life. Further, the addition of chemical functionality to random libraries has failed to improve the catalytic potential of random libraries as anticipated.

This apparent discrepancy between observation and expectation may be

attributable to several shortcomings of the experimental design. Active catalysts may be lost during the preparation of libraries, or during the enrichment for catalysts. Furthermore, only incomplete descriptions have been attempted for any, focusing only on the fastest catalysts in the library, rather than capturing completely the distribution of catalytic power. It is conceivable that added functionality, or alterations of the library or




ix








selection conditions, can dramatically alter the abundance and distribution of catalysts without altering the rate of the fastest catalyst.

The goal of this thesis is to develop methods for estimating the distribution of catalysts within a random library of DNA sequence. To aid in the development of realistic models, we began with an in depth analysis of the catalytic behavior of an individual DNAzyme, 614. This analysis revealed a number of surprising features of this DNAzyme that may be generalized to other DNAzymes created through in vitro Selection (IVS). Although selected with a protocol believed to enrich for Mg-dependent self-cleaving catalysts, 614 cleaves both in cis and trans independently of Mg'. The rate of trans-cleavage at substrate saturation is 6-fold higher than the rate of cis-cleavage. Further, both cis- and trans-cleavage rates are enhanced at temperatures below 25*C (the temperature at which the selections were performed), suggesting that a balance is reached between a slower chemical step (kcat(uni) or kcat(bi)) and a more stabilized association step (k I (uj) or k I (i)) at lower temperatures.

Mutagenesis of 614 and its substrates was performed to learn about the structure of 614 cleaving in cis and trans. Compensatory mutations revealed that one of the intermolecular helixes formed between 614 and its substrate is identical to an intramolecular helix formed in cis-acting 614. The trans-folded 614, however, has an additional 4 base-pair helix not found in cis-folded 614, and this may partially explain the faster rate of trans-cleavage at saturation.

We also addressed the potential loss of catalysts by "premature catalysis" during the preparation of DNAzyme pools in the IVS. Purification of single-stranded DNAzymes was accomplished using asymmetric PCR or exonuclease degradation,



x








eliminating the need for column-based methods for performing IVS. We also examined the utility of a protected ribose (caged-ribose) for both kinetic analysis and an WVS. Selections with caged-ribose failed to enrich for catalysts, whereas a selection with unprotected-ribose produce a number of highly active catalysts. Simulations revealed that the slight increase in background cleavage caused by laser photolysis explains the failure of the caged-ribose to enrich for catalysts.

Analysis of the catalysts isolated from this in vitro selection differed from

sequences isolated in previous selections by only a few nucleotides, but their catalytic rates varied by over an order of magnitude. This variation suggests that the sequence space near any particular active catalyst is well-populated with other active catalysts.

Simulations of in vitro selections examined the impact of various parameters on the outcome of selections. These simulations demonstrated that when the selection time is similar to the half life of the fastest catalysts in the initial library, only the fastest molecules are isolated. Increasing selection time such that it is significantly greater than the half life of the fastest catalysts in the initial population creates a stabilized population. The proportionality between the fastest and slower catalysts of this stabilized population remains similar to the initial population, thus allowing direct estimation of the initial population distribution. The model developed in this simulation can also be used to estimate the distribution of catalysts in the initial library for future in vitro selections.












xi














CHAPTER I
INTRODUCTION



Requirements of a "Living" Sys

In 1994, a committee empaneled by NASA defined life as "a self-sustaining chemical system capable of undergoing Darwinian evolution" (Joyce, 1994). "Nonliving" entities, such as natural phenomena like fire or crystal formation, often have physical properties commonly associated with life-forms such as "energy consuraption," 46 ordered structures," and "growth and replication." A living system is distinguished from non-living on the basis that living systems can evolve. The properties of nonliving phenomena such as fire or crystal formation follow directly from physical laws and the particular environmental conditions of the system. As physical laws presumably do not change, the behavior of nonliving systems cannot change. An evolving system has inherent physical properties as well, but in contrast to nonliving entities, the specific physical properties of the entire system are derived from information contained within the system and can change without destroying the ability to evolve. These physical properties are either intrinsic to the information-containing molecule or generated from the information).

As all life requires obtaining precursors and energy from an environment, the requirement that life be self-sustaining is always relative to a particular environment. The "NASA definition" also limits life to chemical systems, revealing a bias towards life as we know it and in which we are most interested. Evolving systems such as mernes


I





2


(cultural ideas) and in silicon evolving systems (software) are not formally recognized as living because they are not chemical. The information-storing structure for memes is poorly understood, and this may contribute to the biases against considering this evolving system a life-form. For in silicon evolving systems, the range of physical properties ("function space") is arbitrarily defined by the programmer and therefore not open-ended like a chemical system. It can be argued that the "function space" for a chemical evolving system is not in fact open-ended but rather limited by physical laws. Ultimately, both memes and in silicon evolving systems can be considered the product of and therefore not independent of, a living system. One may, however, conceive of life as any system capable of Darwinian evolution. The "NASA definition" of life might then be generalized to "any informational-containing system that undergoes Darwinian evolution."

Although the second law of thermodynamics dictates that the universe progress to a state of greater disorder, this does not necessarily result in the loss of all information. Within a system, order (information) can be created and maintained if it is matched by the creation of greater disorder elsewhere in the universe. Nonliving phenomena often result in ordered structures, such as ripples in the sand of a tidal basin, the lattice structure of a crystal, or polymer formation. This decrease in entropy within the system is coupled to free energy consumption. In order to persist indefinitely, an informational system must prevent information decay by coupling the maintenance of ordered structure to the release of available energy.

Two such mechanisms for preserving encoded information are easily imagined: repairing the information (requiring reference to a "correct" version), or duplicating the





2





3


information (requiring duplication efficiency/accuracy be greater than the rate of degradation).

When the physical properties of a system are deten-nined by the information it contains, it follows that the physical properties of the system can change according to changes in the encoded information. Such a system may never settle into equilibrium (as physical systems are expected to eventually reach). Instead, imperfect information duplication and repair changes the physical properties of a system. This eventually results in multiple variants of an information-containing system, each with variable physical properties. When two such systems utilize the same resources (energy rich molecules, elemental precursors, etc) to prevent information decay, competition exists among systems for more efficient use of the resource.

Despite competition among systems for common resources, the ultimate survival pressure is against the progression towards infonnation loss. A system can therefore persist by either out-competing rivals for a common resource, or by occupying a new niche (utilizing a different resource). Given the constant progressive force towards information decay, there exists selection and enrichment for systems that enhance the survival of their information. As multiple systems compete for resources, information systems linked to replication-enhancing functions will predominate.

From this one can see that Darwinian evolution (which is the fundamental trait of life) is the direct consequence of physical laws applied to an information-containing system. For this reason, some have argued that once a minimal system (hypercycle) capable of Darwinian evolution exists, it will generate higher orders of complexity







3





4


spontaneously and inevitably as long as minimal resources are available and a major global environmental change does not bring about extinction (Kaufman, 1993).

Two specific functional attributes follow from this definition of life, the ability to carry heritable information (genetics) which is linked to selectable traits (function). Selection is thus acting on two levels simultaneously, the system of storing information and the physical properties derivative of this information. Both "information-storage" and "function" can be accomplished using a myriad of chemical and physical formats. However, the requirements of information storage and duplication using chemical polymers are in many ways contradictory to the requirementsfor replicationenhancingfunctions (Benner and Switzer, 1999):

Changes in the information must not impede its ability to be duplicated, whereas
changes in information must allow changes in function, and thus, physical
properties. The selectable traits (function) of a polymer are determined by the
three dimensional structure (fold) of the polymer and the positioning of chemical
moieties, thus a functional polymer must fold to generate selectable chemical
traits. In contrast, an information storage system should not fold easily as this is
expected to impede generalized mechanisms for duplication.

An information storage system should have few subunits to ensure greater fidelity of information transfer, whereas a functional polymer should have many subunits
with diverse functionalities to enhance the chemical potential of the polymer.

For a system to explore new function through evolution, chemical properties must
be able to change rapidly (with few changes) and abruptly. Such abrupt changes in the chemical properties of a molecule following mutation are contradictory to
the requirement that the information carrying system maintain its ability to be
replicated in spite of significant change (mutation).


Chemical Challenges to the Prebiotic Svnthesis of "Life"

For a chemically based living system (which is the only undisputed form of life we know), it is difficult to imagine a mechanism for storing information and generating phenotypes other than through the structure of polymers. Given this restriction, a number




4





5


of requirements must be met for the prebiotic (pre-Darwinian) genesis of a living (evolving) system:

(a) Prebiotic synthesis of the monomers.
(b) Prebiotic formation of polymers from monomers.
(c) Obtaining catalysts from a library of random polymers.
(d) Genetics obtaining a polymer that acts as a heritable carrier of information for
the useful catalysts.
(e) Compartition linking the information storing system (genetics) to the catalytic
function(s) it encodes.

In modem biology, the contradiction between the requirements for a genetic

biopolymer and a catalytic biopolymer (see above) is ameliorated by (mostly) separating the functions of genetics and catalysis into two different bio-polymers. Nucleic acid is the information storing biopolymer, which in turn is used to produce proteins which generate most of the physical properties that are the subject of natural selection. Although the division of labor between two distinct bio-polymers resolves the contradictory demands of a life-form, it creates a significant paradox pertaining to the origins of life

In light of the challenges (a) and (b) to the prebiotic creation of life, it is difficult to imagine a non-biological mechanism for generating either of the contemporary biopolymer systems (nucleic acid or protein). It is astronomically more improbable that both bio-polymer systems would arise simultaneously and spontaneously, and furthermore in a manner that would allow their mutual integration/compatibility. In the 1960s, theoretical consideration of this dilemma led to the hypothesis that life must have originated as a single biopolymer systems (Rich, 1962). This proposal requires only one successful solution to the challenge of obtaining useful monomers and polymers. Furthermore, for a single gene/enzyme system, the challenge of compartition is resolved because the two





5






6


functional requirements are held simultaneously in a single molecule (the challenge of compartition remains when the system under selection contains multiple gene/enzymes). This solution, however, creates another challenge: Such a biopolymer must meet the contradictory challenges of (c)-catalysis and (d)-genetics simultaneously.



Progress Towards Understanding the Genesis of Life: Dual Function Biopolyners

Two approaches have since been pursued to gain insight about how to meet the challenges to the origins of life, one which begins with modem life-forms and works backwards to deduce information about its origins, while the other approach begins with prebiotic chemistry and attempts to work forwards towards life.

Much insight has been deduced about early life from an analysis of its modem

descendents. Although modem life-forms have nucleic acids specialized for genetics and proteins specialized for catalysis, there exist many idiosyncrasies about modem metabolism that are suggestive of a previous incarnation as a single-biopolymer life-form based on RNA: most biological coenzymes are based on nucleotides, nucleic acids are used to the generate proteins; histidine is biosynthesized from nucleotides (phosphoribosyl pyrophosphate and ATP); RNA is a key component of the spliceosome, ribosome, and RNAseP; RNA serves as the primer for DNA synthesis; DNA is synthesized by protein-based reduction of RNA (a chemical transformation that is quite difficult, and unlikely to occur via prebiotic mechanisms) (Joyce, 1989; Benner, 1989). The use of RNA for chemical purposes for which it is not intrinsically suited, particularly in the modem environment where chemically better-suited molecules could be made (proteins), argues against convergent evolution and instead suggests that these RNA





6





7


vestiges are "molecular fossils" leftover from a time in history when RNA was the sole biopolymer (as opposed to convergent evolution) (Gilbert, 1986). These considerations offer support for the previous theoretically-based proposition that modem life-forms passed through a single biopolymer, RNA-based life-form, a period of time known as the "RNA world" (Woese, 1967; Crick 1968; Orgel, 1968). Reconstructions of the RNA world indicate that RNA life-forms possessed quite complex metabolism: molecular fossils indicate that the RNA world developed the RNA cofactors ATP, coenzyme A, Sadenosylmethionine, and NADH. lit follows that the RNA world needed these, presumably for phosphorylations, Clasisen condensations, methyl transfers and oxidation-reduction reactions (Benner, 1989).

The discovery that modem life contains RNA that acts as a catalyst demonstrated that RNA is in fact a single bio-polymer capable of both genetics and catalysis, thus giving experimental support to the proposition that modem two-biopolymer life is descendent from a single-biopolymer (RNA) life-form (Usher, 1976; Cech, 1981, Geurrier-Takada, 1983). The existence of an "RNA world" does not necessitate, however, that life actually originated with RNA. Indeed, applying chemical knowledge to "work forward" from prebiotic earth to modem life suggests that ribonucleic acids may have been a very poor candidate as the primordial monomer in the origins of life. Instead, the RNA world may have served as a transitory life-form between the original single-polymer life-form and today's robust, two-biopolymer life-form. We now turn our attention to insights gained by "working forward" from basic chemical principles.









7






8


Progress Towards Understanding~ the Genesis of Life: Prebiotic Synthesis of Potentially Useful Monomers and Polymers

The challenge of non-biological synthesis of useful monomers and polymers

argues against the simultaneous creation of two interrelated polymeric systems, but the creation of even a single polymeric system for life (capable of genetics and catalysis) still must meet the need for prebiotic synthesis of monomers and polymers. Knowing that chemically based life does exist we must assume that monomers and polymers did form in the prebiotic world as result of the basic principles of chemistry. It has not, however, been easy to deduce how this might occur.

In light of the challenges to the prebiotic generation of life, what is so surprising about the origins of life (other than that it happened at all) is that it appeared relatively rapidly. The earth was formed -4.6 billion years ago (bya). The first -0O.5 billion years of earth's life, prior to the cooling of the earth's crust and the subsiding of severe meteorites bombardment, was inhospitable to the formation of any useful monomers, yet evidence of life in the fossil record is apparent as early as 3.6 bya (Walter, 1983). This provides a very short time for the generation of monomers, polymers and subsequent genesis of life.

In order to apply chemical principles to gain insight into the prebiotic creation of monomers, the prebiotic environent of the earth (approximately 4 billion years ago) must be understood. Although definitive information about the earth's early environment and composition is slow coming, it is widely assumed that H20, C02, H2, and CO existed. In a groundbreaking experiment, Stanley Miller demonstrated that combining these basic elements with an electric discharge (presumably also occurring with sufficient regularity on the prebiotic earth) produced a prebiotic soup containing a mixture of key




8





9


materials, most notably HCN (Miller and Urey, 1959; Sclesinger and Miller, 1983; Stibling and Miller, 1987). The prebiotic presence of HCN is further supported by its presence in intrastellar clouds, in comets, and in the atmosphere of Titan (Irvine, 1999; Huebner, et al, 1974; Hanel, 1981).

Base-catalysed tetramerization of HCN yields diaminomaleonitrile, which then leads to the production of adenine, hypoxanthine, guanine, xanthine, and diaminopurine. Hydrolysis of HCN oligomers yields amino acids, particularly glycine, alanine, aspartate, and diaminosuccinate, among other amino acids (reviewed in Joyce, 1989). Other electric discharge experiments under presumably prebiotic conditions generated purines, among other potentially useful materials. These experiments usually have low yields of pyrimidine synthesis, but recent experiments of Miller demonstrate that significant pyrimidine yield from frozen dilute samples of NH4CN (Miyakawa, et al 2002).

Another prebiotic molecule, H2CO, can lead to glyceraldehydes through

condensation in the presence of a catalyst such as calcium carbonate or alumina (Gabel, 1967; Reid, 1967). Glyceradehyde can then begin a cascade that converts formaldehyde into trioses, tertroses, and larger sugars through a process of aldol condensation and enolization (Joyce, 1989).

Joining nucleoside and sugar subunits has been accomplished by heating a

mixture of ribose and purine to dryness in the presence of inorganic salts; this process, however, leads to a mixture of a- and (3-nucleosides (Fuller, et al, 1972).

In summary, experimental evidence suggests that the chemical environment and composition of the prebiotic earth is sufficient to provide the necessary mechanisms and materials for the synthesis of many potentially useful monomers. These findings are





9





10


supported by the discovery of similar monomers on meteorites, which were also presumably synthesized by prebiotic mechanisms in space. The next step towards a living system requires that monomers be linked together by prebiotic chemistry to form polymers.

Limited polymerization of nucleic acids has been observed under prebiotic

conditions. Adenosine 2'-, 3'-cyclic phosphate can be converted to oligomers by heating to dryness, or by incubating at room temperature in the presence of a polyamine (Verlander, 1973). Nucleoside 5'-polyphosphates have also been produced by heating nucleotides and inorganic polyphosphates to dryness (Lohrmann, 1975). The random polymerization of monomers, however, remains a great challenge, specifically in light of the abundance of stereo-chemically related compounds that are expected to interfere with prebiotic polymerization. Taking ribose-based nucleotides as an example, nucleotide polymerization can result in a combinatorial mixture of 2'-, 5'-, 3'-, 5'-, and 5'-, 5'phosphodiester bonds, with variable numbers of phosphates between the sugar bases, combined with the D- and L- sugar stereoisomers and a- and 13-anomers of the glycosidic bonds. Most polymers generated are expected to have heterogeneous linkages, which is expected to make replication of the polymer exceedingly more difficult. Several mechanisms may have existed to overcome stereoisomeric interference, generally relying on some mechanism of enriching for specific isomeric forms of monomers or preferential incorporation of certain monomeric isoforms. Although prebiotic formation of both monomers and polymers is likely to result in a clutter of stereoisomers, a series of biased synthesis followed by fractionations or enrichments may have provided sufficient building blocks for creating life.





10





11


It is interesting to note that not all (potentially) prebiotic monomers suffer as

many of the stereoisomeric challenges as ribose-based nucleotides. Both threose nucleic acid (TNA) and peptide nucleic acid (PNA) use alternative bases attached to nucleic acids. Both monomers are more chemically simplistic, thus avoiding much of the stereoisomeric clutter of RNA. This has led to proposals of a pre-RNA world utilizing either polymer as potentially life-supporting polymers. Theoretical and experimental evidence suggest that information stored in the form of TNA or PNA polymers can be transferred to RNA, although it is unclear how much phenotypic function is preserved in the transfer (reviewed in Joyce, 2002).



The Next Challenge: Obtaining Catalytic Function from Random Libraries

Although monomer and polymer synthesis is expected to be difficult in a prebiotic world, it is the formation of polymers that act as genetic molecules with useful chemical properties that appears most formidable (useful from the perspective that it leads to the creation of life: an initially useful property would be the ability to enrich for specific monomers). This requires that polymers be of sufficient length to contain sufficient information, be sufficiently stable as to persist, be able to mutate yet still be replicated, and lastly, perform useful chemical function.

Although we do not know precisely what monomers were available in a priobiotic world or how they were synthesized and polymerized, it is clear that at some point 4 bya some randomly generated prebiotic polymers must have possessed "useful" properties (presumably a self-replicase) that permitted the jump from a collection of random polymers to a single-biopolymer life-form with a complex metabolism.





11





12


Although this is the most reasonable hypothesis, it is difficult to imagine that such a complex function could be generated from a random polymer. An experimental approach, in vitro selection (IVS), emerged in the 1 980s that served as a bridge between the "work forward" and "work backward" approaches by permitting a direct test of the catalytic power of random-polymer libraries (Szostak, 1992; Beaudry, 1992; Irvine, 1991).

IVS begins with a library containing _1013 unique and randomly generated polymers of DNA. This library is then challenged by a selection process (such as the ability to catalyze a particular reaction or bind a particular molecule). Molecules that can perform the selected task are separated from those that do not (Figure 1- 1). Finding molecules in a random library that perform the selected task is therefore a function the abundance of that activity, and the ability of the selection regime to separate active from inactive molecules.

Solid support
Combinatorial library Random region
of DNA molecules

Reaction Buffer
(Denaturation and elution Cofactor (Mg++) fl complementary strand
U Cleavage site
Binding of PCR AO 5r]
products to solid jJL h
support
FF~ Catalytic Motif



PCR amplification Elution of cleaved
Molecules
5' 3'

Figure 1-1. A schematic diagram describing in vitro selection for a DNA molecule that catalyzes a cleavage reaction.



12






13




In practice, the selection process is not perfect, allowing inactive molecules to

survive (and presumably losing some active molecules). When active molecules are rare and background is high, several sequential selections are required to enrich for active molecules. Experimentally, LYS accomplishes sequential rounds of selection by using the surviving molecules from the first round of selection as a template for PCR, thus amplifying the survivors and creating a second pool of DNA molecules (that are presumably enriched for catalysts) for a subsequent round of selection.

This experimental strategy is a close analogy to what is imagined to have led to the first life-form (with the exception that prebiotic mechanisms, as opposed to human intervention, lead to the development of the polymer library, and "selection for desired activity" was imposed by the Earth's environment and processes which lead to the decomposition of polymers). It was anticipated that IVS would now demonstrate that the chemical processes necessary for life could in fact be found in a prebiotic library of polymers. After a decade of experiments and great effort, many catalysts have indeed been isolated from random libraries, most notably a phosphodiesterase, a Diels-Alderase, a nucleotide ligase, and a limited polymerase (Breaker and Joyce, 1994; Johnston, 2001; Seeling, 1999; Tarasow, 1997).

Despite these successes, getting self-replicating nucleic acid polymers from IVS proved harder than anticipated/hoped. Bartel and others are coming close to achieving this goal (Johnston, 2001). It must be noted, however, that even the authors of this research contribute the success of the de novo generation of an RNA based RNA polymerase to a stepwise procedure guided by human chemical knowledge: "Thus far, efforts to select for polymerization activity in a single step directly from random-sequence



13





14


RNA have yielded only ribozymes that decorate themselves inappropriately with tagged nucleotides" (Johnston, 2001, page 1324). Furthermore, estimates from their work suggest that more than 1030 random RNA sequences might be required to get templatedirected RNA polymerization from a truly random pool. More directly, Bartel and Szostak noted that the odds of obtaining a ligase able to enhance a very simple template ligation reaction by four orders of magnitude were on the order of one in 10 13 (Bartel, 1993).

As mentioned earlier, it is not certain that life originated from RNA polymers, indeed many other candidate polymers are imaginable. In support of this, it is notable that catalytic activity has been isolated from various different types of libraries, including amino acid, ribonucleic acid, and deoxynucleic acid polymers. Extrapolations from a search of amino acid libraries estimate that in a fully randomized library, a library of 1025 members would be required to obtain an active catalyst (in that case, AroQ mutase) (Taylor, et al 2001). This search of amino acid libraries concluded that, although the diversity of functionality in proteins provides intrinsically greater catalytic potential, folding proteins may be inherently more difficult than folding nucleic acid polymers, therefore making an active enzyme from a random polymer of amino acids appears to substantially more difficult that obtaining a ribozyme from a random nucleic acid library.

Although natural amino acid polymers (proteins) function superbly as catalysts, and amino acid catalysts have been isolated from random libraries, amino acid polymers are not well suited as carriers of genetic information for reasons described earlier. In vitro selections using amino acid libraries are also much more technologically limited than searches of nucleic acid sequence space. For all these reasons, and the indication that, although life on earth may not have started with RNA, it clearly passed through a




14





15


one-biopolymer stage based on RNA, in vitro selections have largely focused on nucleic acid biochemistry.

More than a decade ago, Benner proposed what has come to be the standard

explanation for the "poor" catalytic power of nucleic acids (where "poor" reflects in part an unbalanced comparison with natural protein enzymes) (Benner, 1987; Benner, 1988b). Nucleic acids, he argued, have little organic functionality, at least compared to natural protein enzymes. DNA lacks in the scaffold cationic groups, imidazoles, thiols, and carboxylates, all of which play effective catalytic roles in protein enzymes near neutral pH. To relieve these limitations, Benner suggested expanding the genetic alphabet for the purpose of permitting DNA to carry more of these functional groups (see Switzer, 1989; Piccirilli, 1990).

This hypothesis gains support from the observation that many of the catalytic

RNAs believed to be vestiges of the RNA world (spliceosome, RNaseP, ribosome) contain post-transcriptional modifications; the conservation of this difficult task both throughout 4 billion years of evolution and across wide phylogenic taxa suggests that these modifications must play a critical role in the function of these catalytic RNA. Evidence for the prebiotic synthesis of C-5 functionalized uracil including ammonia, glycine, guanidine, hydrogen sulfide, hydrogen cyanide, imidazole, indole and phenol also lends credence to the hypothesis that functionalities appended to nucleic acids may have served a crucial role in the origins of catalytic nucleic acids and subsequent life (Robertson and Miller, 1995).

Several groups (Battersby, 1999; Latham, 1994; Sakthivel, 1998; Perrin, 1999;

Perrin, 2001) have now synthesized functionalized nucleic acids, mostly derivatives of 2'15





16

deoxyuridine, and used them in IVS experiments in an effort to demonstrate that adding functionality increases the odds of finding catalytically powerful sequences in a DNA library. Judging from published examples, it seems to be now accepted that adding functionality typically increases the catalytic potential of a DNA library perhaps by a factor of two to ten, not by orders of magnitude.

In a recent example, Joyce and co-workers selected a zinc-dependent RNAse

activity utilizing an imidazole-functionalized nucleotide in their library (Santoro, 2000). No significant catalytic boost was detected, however, over DNAzymes isolated earlier by the same researchers using non-functionalized nucleotides (Santoro, 1997). Together with other similar findings, this suggested that functional endowment offers less than an order of magnitude enhancement of the catalytic potential of the library (Ang, 1999; Ang, 2000). Analogous experiments where functionality was introduced as a cofactor, binding to but not covalently attached to the DNA molecules, generated analogous conclusions (Roth, 1998).

What appears to be absent from the literature is the sense of astonishment at these results. Functional groups are supposed to be important in organic reactivity. For example, in an experiment in which a polypeptide was generated that folded in solution and catalyzed the decarboxylation of oxaloacetate (Johnsson, 1990; Johnsson,93), the amino groups performed the catalysis. Having them mattered; not having them mattered even more, as catalysis went down by several orders of magnitude (to background) when they were removed. In chemical models, adding imidazole (for example) increases reactivity of catalytic models by orders of magnitude, not by a factor of two to ten (Zepik, 1999).





16





17


When organic chemical theory expects a large (orders of magnitude) impact of a change of structure on molecular behavior, and only a small impact is seen (a factor of 2 to 10), something must be wrong. Therefore, this "orders-of-magnitude" difference in expectation versus outcome needs to be pursued. Perhaps the most formidable challenge to our understanding of the origins of life lies in the experimental indication that function is excessively rare in random polymer libraries. Therefore, if a plausible model for the origin of life is ever to emerge based on an "RNA-world" scenario, we need to understand why functionality does not seem to significantly enhance the catalytic power of the library.

Although a key obstacle in the origins of life is the creation of function from

random libraries, none of the IVS experiments conducted thus far have been designed to directly estimate the frequency of function within a library, or determine directly how this frequency changes with various types of libraries and under varying selection schemes. Most in vitro selections are designed to determine if a desired activity exists in a library, and therefore are content to isolate a few molecules with the desired activity. The previously used methods of estimating the frequency of catalysts in a library are not comprehensive (Johnston, 2001, Bartel, 1993, Taylor, 2001), thus the few estimates of catalytic frequency provided are not to be considered realistic. Indeed it is difficult to imagine life emerging spontaneously from nucleic acids given these kinds of numbers, especially given the difficulties in obtaining RNA (or DNA) in polymeric form under any plausible prebiotic reaction conditions in water. Initially, the limited goal of isolating any activity from a library was understandable given that it was not known whether pAy such activity existed in a random library. Knowing that activity can be isolated from random





17






18


libraries, the next critical task of estimating the frequency of such activity in a random library, and determining what factors alter catalytic abundance and power, can be addressed. Understanding this is the next vital step to understanding how to make the transition from prebiotic polymers to a living system.

In order to understand how a single-biopolymer life-form is created by prebiotic mechanisms, we must understand the factors that alter the probability of creating function from random polymers, specifically the apparent contradication that added functionality doesn't appear to significantly improve catalysis. Five categories of hypotheses might be considered to account for this orders-of-magnitude difference between expectation and outcome when adding functionality:

(a) Our view of the role of functionality in catalysis may be naive. Functional groups
may not greatly enhance catalytic power.

(b) Nucleic acids are, of course, already functionalized (with phosphates and hydrogen
bond donating and accepting groups). Perhaps these are the only functionalities
needed for catalysis.

(c) Nucleic acids, although they do not have many functional groups intrinsically in
their covalent structure, may recruit sufficient numbers of these (in particular,
divalent cations such as Mg) as "cofactors" to make up for the lack of
functionality covalently linked to the scaffolding.

(d) The IVS experiments may have been designed in a way as to overlook or lose the
best functionalized catalysts.

(e) Perhaps adding a single type of functional group is not sufficient; one may need to
add two or more types of functional groups before the expected large benefit from
functionality is seen (Perrin, 1999; Perrin, 2001).

A priori, each hypothesis has its attractions and disadvantages. For example, we can easily identify ways in which the best endowed catalysts would be lost in a BreakerJoyce protocol (see below), suggestive of hypothesis (d). Divalent metal ions do bind to DNA, they can participate in catalysis, and they assist even more highly functionalized




19





19


protein catalysts, suggestive of hypothesis (c). Proteins do seem to exploit multiple types of functionality, and exploit hydrophobicity, something lacking in aliphatic form in DNA, suggestive of hypothesis (e). Last, recent experiments by Keefe and Szostak selecting for proteins that do things suggests that they do not do things orders of magnitude better than DNA (2001), suggestive of hypothesis (a) or (b).



Experimental Design

This thesis presents the first step towards addressing the hypotheses above. We utilized the experimental procedure developed by Breaker and Joyce (1994, 1995). This IVS procedure seeks DNA molecules that are catalytically active as ribonucleases; this protocol is the most widely utilized procedure for estimating catalytic performance under various conditions, allowing general comparisons such as the ones mentioned before that suggest functionality does not significantly improve catalytic potential. As mentioned above, one explanation for the apparent failure of functional endowment to generate very active deoxyribozymes may result from the loss of highly reactive catalysts during some step in the experiment before they had the opportunity to survive the selection step.

It is easy to find places in the set-up of the Breaker-Joyce selection procedure

where highly active catalysts might be lost. Standard IVS experimental systems assume that catalysts remain inactive during the work up procedure, which includes: PCR (with MgC12), ethanol precipitation (with high salts), binding to a column (again in high salts, but without MgCl2) and removing the complementary strand with high pH sodium hydroxide. After work up is completed, the reaction is "started" with the addition of MgCl2 and the eluate from a column is collected and presumed to contain the active





19






20


catalyst. There are several technical weaknesses of the Breaker-Joyce IVS protocol that may result in the loss of desired catalysts.

First, columns have poor resolving ability, in part because they have great

potential for residual retention (unbound material slowly filtering through the column) and leakage (bound material becoming unbound). It is also assumed that even the "best" catalysts remain inactive until the reaction is "started" by adding MgC12 (or other cofactor, such as histidine) and collection begins. Magnesium is in fact present during the preparation, including during the PCR, and other salts are also present during the ethanol precipitation potentially permitting catalysis during the preparation steps and resulting in loss of catalysis. DNAzymes have been isolated using the Breaker-Joyce protocol that are active without any added cofactor (requiring only NaCl), offering support to the hypothesis that active catalysts may be lost during the workup.

Second, the activity of a DNAzyme is inhibited by the complementary strand, and therefore must be removed. This is accomplished in the Breaker-Joyce protocol by denaturing the double-stranded DNA with a strong base solution (NaOH), and then washing the non-biotinylated complementary strand from the column. Strong base can itself cause sequence independent cleavage of RNA, and is known to increase the rate of many DNAzymes. It is therefore possible that many DNAzymes, in particular the fastest DNAzymes, are lost through this base-catalyzed cleavage of RNA. The strong base may also destabilize the binding of the catalysts via biotin to the streptavidin column, resulting in further loss of potential catalysts.

Lastly, DNAzyme folding is largely ignored. After flushing out the NaGH with

reaction buffer (minus the "required" cofactor), the reaction is "started" by the addition of





20






21


the cofactor. Correct folding is assumed to occur without the required cofactor, and quickly relative to catalysis.

Each of these problems arises because the Breaker-Joyce protocol has the 2'hydroxyl group of the RNA molecule to be free during the set-up. To resolve these problems, research in this disseration develops techniques to eliminate loss of catalysts during set up by using a caged-ribose. Caged-ribose was synthesized with the 2'-hydroxyl group protected via an ortho-nitrobenzyl protecting group. This protecting group can then be removed by UV photolysis (Figure 1-2).




Caged ribose



LASER




3' 3'

Caged is blocking the T-OH is free and able to 2-OH in ribose for attack the phospus
nucleophilic attack on atom with the help of a the phosphorus catlJ






Figure 1-2. Exposing the caged-ribose converts 2'-orthonitrobenzyl
protecting group to the standard hydroxyl group, thus generating a standard
RNA nucleotide.








21






22


Our eventual goal is to estimate catalytic distribution in a random library, and use this estimate to determine which factors enhance the catalytic power of a library. To make this estimate will require a realistic mathematical simulation of the selection system which, in turn, requires detailed knowledge of the catalytic process. Towards these goals, research in this dissertation seeks a detailed understanding of why the phosphodiesterase activity isolated using the Breaker-Joyce procedure routinely reaches a low plateau. It is also crucial to determine whether the reaction behaves as a first order chemical process (as predicted).

Conditions for the use of caged-ribose have been optimized, allowing reaction during the set up to be prevented. Preliminary selections with the caged-ribose indicate that the laser photolysis of the caged-ribose creates a slight increase in background cleavage, both experimental and simulated selections indicate that this leakage dramatically increases the number of rounds of selection required to isolate active catalysts. Although this limits the utility of caged-ribose for selections, the background does not significantly alter kinetic analysis, allowing its use for studying the pre-activation requirements of DNAzymes, such as folding.

To aid in the development of realistic models, we undertook a detailed study of an individual DNAzyme, 614. This revealed a number of significant findings. It is widely assumed that including a cofactor (such as Mg++ or other divalent cations) into the selection buffer will result in the selection for cofactor-dependent catalysts. Indeed, the Breaker-Joyce protocol presumes that potential catalysts are completely inactive without the added cofactor. We have determined that, although 614 was selected in the presence of Mg++, 614 retains complete catalytic activity in the absence of Mg++. It is also widely





22





23


assumed that the identity of the constant regions (arbitrarily chosen by researchers) surrounding the randomized region is largely irrelevant to the catalytic activity eventually selected, and it is instead only the catalytic potential of the random region that is being tested. The DNAzyme 614 was selected with 2 nucleotide changes to the standard Breaker-Joyce constant region; detailed analysis of 614 revealed that 614 can act in both a uni-molecular and bi-molecular fashion. This may be a direct consequence of the two nucleotide changes from the Breaker-Joyce motif (although this is not known for certain since few researchers have directly tested whether the isolated DNAzymes are in fact acting only in a uni-molecular fashion). This has given us insight into how to design and simulate future IVS.

We have also isolated several catalysts, some of which are close relatives of 614, 615 and 616 isolated in a previous selection, which each display catalytic activity. This is significant in two aspects, first in that the catalytic activity varies significantly with only a minimal sequence changes. This suggests that the landscape in the near vicinity (one or two nucleotides) of a particular active molecule contains many catalysts, presenting opportunities for continuous exploration of the functional landscape. Second, the abundance of near-relatives that retain activity suggests that most estimates of the catalytic frequency may be gross underestimations, as near-relatives may appear identical under the methods (restriction digest) used to distinguish them.













23















CHAPTER 2
MATERIALS AND METHODS


Preparation of Precursor DNAymes via PCR

DNAzymes for kinetics were prepared by PCR amplification of the template (either synthesized by IDT, or from a clone) using one of various primers to generate the catalytic strand and one of various primers to generate the complementary strand. All templates had a common 5' and 3' constant region to which the complementary and catalytic strand primers can bind. Between the primer binding sites is a 40 nucleotide region of varying sequence:

"library" template (complementary strand): 5 '-GTGCCAAGCTTACCGTCAC (n40)
GAGATGTCGCCATCTCTTCC

"614" template: 5' -CTGCAGAATTCTAATACGACTCACTATAGGAAGACATGGCGACTCTCACATCATGCGAGCACACGCAATAGCCTGATAAGGTTGGTAGTGACGGTAAGCTTGGCAC

Two sequence variants of the catalytic strand primer were commonly used, but both bound to the constant region of the various templates:

"ang + ribose" catalytic strand primer:
5' CTGCAGAATTCTAATACGACTCACTATrAGGAAGACATGGCGAC-TCTC)

"BJ + ribose" catalytic strand primer:
5' -F-GGGACGAATTCTAATACGACTCACTATrAGGAAGAGATGGCGACATCTC)

The differences between the "ang" and "BJ" primers are underlined; "-" represents an alignment gap; "N" is equimolar concentrations of each dNTP, "F" is a 5'Fluorescein; "rA" is a ribo-Adenosine. Primer variants were also used in which the ribo-adenosine was replaced with either deoxy-adenosine ("ang ribose" and "BJ -ribose") or caged24






25


ribose ("ang + cage" and "BJ + cage"). The complementary strand primer sequence was always the same (except "complementary +5'TAIL 2C9*G" in which the 5' G was changed to T), but varied in non-coding modifications appended to the 5'-end:

"complementary no 5'-P" primer:
(5' -GTGCCAAGCTTACCGTCA)

"complementary + 5'-Phosphate" primer:
(5' -P-GTGCCAAGCTTACCGTCA)

"complementary + 5'tail#1" primer:
(5'-GGTGGGTGGG-cl18-GTGCCAAGCTTACCGTCA)

"complementary +5'TAIL 2C9" primer:
(5' -AAAAAAAAAAAAAAAAAAAA-c9-c9-GTGCCAAGCTTACCGTCA)

"complementary +5'TAIL C 18#2" primer:
(5' -AAAAAAAAAAAAAAAAAAAA-cl8-GTGCCAAGCTTACCGTCA)

"complementary +5'TAIL C18#3" primer:
(5'- AAAAAAAAAAAAAA-cl8-GTGCCAAGCTTACCGTCA)

"complementary +5'TAIL 2C9*G" primer:
(5' -AAAAAAAAAAAAAAAAAAAA-c9-c9-TTGCCAAGCTTACCGTCA)

"complementary +5'TAIL 3C9" primer:
(5' -AAAAAAAAAAAAAAAAAAAA-c9-c9-c9-GTGCCAAGCTTACCGTCA)

"complementary +5'TAIL 4x2'OME" primer:
(5' -AAAAAAAAAAAAAAAAAAAA-mUmUmUmU-GTGCCAAGCTTACCGTCA)


"C 18" is an 18-atom hexaethyleneglycol-based spacer, "C9" is a 9-atom

triethyleneglycol-based spacer, "mU" is a 2'O-methyl-uridine, "P" is phosphate. Typical conditions for a 100 pL PCR contained up to 1 ng template, 100 nM catalytic strand primer, 100 nM complementary strand primer, 100 jIM dNTPS (either standard A, G, C, and T, or a "nonstandard" mixture in which thymidine is replaced by 5-(3-aminoallyl)2'deoxyuridine, henceforth referred to as "E-base"), 10 mM KCI, 20 mM Tris-HCI (pH 8.8), 10 mM (NH4)2SO4, 2 mM MgSO4, 0.1% Triton X-100, 3-4 units polymerase (Taq or Vent exo-), and 10 tCuries alpha-32P-CTP (for internally labeled samples). PCR



25





26

amplification consisted of 3 minutes at 96"C; followed by 20 cycles of 45 seconds at 96*C, 45 seconds at 50C, and 2 minutes at 720C.

PCR for in vitro selections (IVS) varied slightly: 400 nM of the catalytic strand primer was used (to compensate for hairpins formed by the "BJ +cage" primer), up to 40 cycles of PCR were used in the early rounds of IVS, Vent exo- was used (rather than Taq because it incorporates 5'-functionalized nucleotides such as E-base with greater efficiency), and primer binding was done at 57C rather than 500C (again, to compensate for hairpins formed by BJ primers).

Point mutations introduced within in the catalytic strand primer region but 3' of the ribose-adenosine, were incorporated using a two step PCR. In the first step, 10 cycles of PCR were conducted in which the catalytic strand primer was replaced with one containing the desired point mutation and a 5 nucleotide truncation from the 5'end. The product of this PCR was used as the template (diluted 1:10) for a second PCR containing a catalytic strand primer truncated two nucleotides 3' of the ribose-adenosine. This twostep PCR method allowed for the introduction of several point mutations into the catalytic primer region without necessitating unique DNA-RNA chimeric primers for each mutant.



Preparation of Single-Stranded DNAzymes

Double stranded DNAzymes generated via PCR using the 5'-phosphorylated form of the complementary strand primer ("complementary + 5'-phosphate) were converted to single-stranded DNA by digestion of the complementary strand using lambda exonuclease, an exonuclease enzyme specific for 5'-phosphorylated double stranded





26





27

DNA. Following ethanol precipitation of PCR products, DNA pellets were resuspended in 25 [tL of exonuclease solution (5 units lambda exonuclease enzyme, 67 mM glycineKOH (pH 9.4), 2.5 mM MgCl2, 50 jtg/mL BSA) per 100 RL PCR. Samples were mixed thoroughly and incubated at 37C for 30 minutes. Reactions were terminated by adding excess formamide stop dye (1 mg/mL xylene cyanol, 1 mg/mL bromophenol blue, 10 mM EDTA, in 98% formamide) and heating at 80'C for 10 minutes. The single-stranded products were loaded on an 8% PAGE/urea gel and full length ssDNA products excised using a sterile razor (thoroughly washed with ethanol between each sample). Each gel slice was crushed and eluted in 350 [tL elution buffer (500 mM NH40AC, 0.1 mM EDTA, 0.1% SDS) overnight. Gel purified samples were then extracted in phenol/chloroform/isoamyl alcohol, then extracteded choloroform/isoamyl alcohol, and precipitated in ammonium acetate and ethanol.

Asymmetric PCR was also used to generate single stranded DNAzymes. In this procedure, the complementary strand primer was synthesized with a 10-20 nucleotide extension (usually poly-adenosine) connected to the 5'-end of the primer by one of several linkers (C9, C18, 2'O-methyl-Uridine). Vent and Taq polymerase cannot efficiently read-through these linkers. Therefore, the catalytic strand generated by extension of the catalytic primer terminates at the linker, generating the correct size for full-length product. Complementary strand molecules generated by extension of a complementary strand primer containing the "5' tail" are longer than the full-length product by the length of the tail, and are easily separated by gel purification on an 8% PAGE/urea gel. Because Vent and Taq polymerase occasionally terminate extension (about 50% of the time) when encountering a caged-ribose in the template, PCR using a




27





28


caged-ribose in the catalytic primer and a 15 nucleotide tail on the complementary primer generated a complementary strand that was either 121 nucleotides (106 nucleotides plus a 15 nucleotide tail) or 94 nucleotides (106 nucleotides plus a 15 nucleotide tail minus 27nucleotides following the caged-ribose). The desired catalytic strand was easily identified as the middle band and cleanly excised on an 8% PAGE/urea gel. Gel purified samples were crushed, soaked overnight in elution buffer, and then extracted in phenol/chloroform/isoamyl alcohol, extracted again in choloroform/isoamyl alcohol, and precipitated in ammonium acetate and ethanol.


5'-End Labeling of DNA

Single-stranded DNA (20 jiM )was 5'-labeled with 20 jtCi gamma-32P-ATP using 10 units of T4 polynucleotide kinase, 70 mM Tris-HCl (pH 7.6), 10 mM MgC12, 5 mM dithiothreitol in a final volume of 10 pL. Reactions were incubated for 30 minutes at 370C, followed by addition of equal volume TE, and finally inactivated by incubating for 20 minutes at 700C. End-labeled DNA was separated from unincorporated nucleotides by spinning through a G-25 column at 600 g for 3 minutes.


DNAzyme Kinetic Assays

Following gel purification of single-stranded DNAzymes and substrates, samples were ethanol precipitated and re-suspended in 50 mM HEPES buffer. Samples were then transferred to a new tube and the concentration estimated by measuring specific radioactivity following a measurement of by Cherenkov radiation. Samples were then diluted to twice the desired final concentration with additional 50 mM HEPES buffer. For trans assays, enzyme and substrates were typically mixed together (unless otherwise





28





29

noted), and diluted to twice the desired final concentration. Samples were then mixed with equal volume 2X reaction buffer (typically 2M NaCl, 2 mM MgCI2, 50 mM HEPES pH 7). Reactions were then heat denatured and refolded using a thermocycler "slow cool" protocol (heating to 96*C for 3 minutes, and cooling to 230C over 10 minutes). For reactions with ribose-adenosine, "time zero" was considered to be the time at which the sample completed the "slow cool." For reactions with a caged-ribose, the reactions were initiated when the laser de-protected the caged-ribose (within one hour of the completion of the "slow cool"). Unless otherwise noted, reactions were conducted at 25C (incubated on a hot block set to 250C to prevent the temperature from dropping below 25*C). Reactions were terminated at various time points by diluting an aliquot of the reaction in excess formamide stop dye, and then frozen (-200C). Time points were run on an 8% PAGE/urea sequencing gels and the percent cleaved quantified using a Bio-Rad phosphorimager.


Cloning and Sequencing DNAzymes

Prior to cloning, single-stranded DNAzymes were first converted to double stranded DNA by PCR amplification, usually with only the complementary strand primer ("complementary no 5'-phosphate"). PCR conditions were: 96*C for 3 minutes; followed by three cycles of 45 seconds at 96C, four minute ramp cool to 50*C plus 45 seconds at 50*C, and 1 minute at 72*C; followed by 7 minutes at 72*C. When cloning cleaved and uncleaved 614, the desired single-stranded DNA was (usually) first isolated via gel purification. PCR amplifying the unpurified 614 reaction mixture containing both cleaved and uncleaved 614 with only the "complementary no 5'-phosphate" primer yielded only uncleaved molecule clones; cleaved 614 molecules were therefore cloned by



29





30


using the complementary strand primer "complementary no 5'-phosphate" with the 5'truncated catalytic primer "3'-cleaved oligo." Fresh double stranded PCR products were cloned using the TOPO TA Cloning System (Invitrogen) and plated onto Agarose plates containing ampicillin. When cloning sequences from the pool of survivors from the "H" selection condition, transformed cells were given only 15 minutes to recover in antibiotic free media prior to plating on antibiotic containing plates. This prevents recently transformed clones from doubling prior to plating and therefore favors sequencing only unique species from the original pool.

DNA was prepared from individual clones using a modified alkaline lysis

protocol. Individual clones from plates were used to inoculate 5 mL of TYGPN media containing freshly added ampicillin and grown overnight at 370C with vigorous shaking. Following overnight growth, 2 mL of cells were pelleted by centrifugation for 1 minute at 16,000 g. Growth media was removed and the cell pellet re-suspended in Solution 1 (300 [tL: 10 mM Tris pH 8.0, 1 mM EDTA, and 50 gg/ml Rnase A). Solution 11 (300 gL: 1 mL fresh 2 M NaOH, 1 mL 10% SDS, 8 mL water) was then added and the samples were then inverted gently to mix. Solution III (300 ttL: 5 M potassium acetate pH 4.8) was then added, followed by gentle mixing. The sample was centrifuged for 2 minutes at 16,000 g, after which 500 jiL of chloroform was added and gently mixed. The samples were then centrifuged again at 16,000 g and 750 gL of the aqueous supernatant was transferred to a new tube containing isopropanol (750 ttL). This sample was centrifuged for 6 minutes at 16,000 g. The isopropanol was completely removed, and the remaining pellet re-suspended immediately in 100 [tL water, followed by the addition of 100 JIL 15% PEG-8000. The re-suspended pellet was stored on ice for 30 minutes (or overnight),





30





31


after which the sample was centrifuged for 15 minutes at 16,000 g. The supernatant was removed and the pellet rinsed with 70% ethanol (200 pL). The ethanol was removed and the pellet was dried completely under vacuum. The resulting DNA pellet was resuspended in 20 iiL of water. This protocol routinely yielded 100 ng/uL of DNA of sufficient purity for sequencing reactions.

Sequences transformed into the TOPO TA Cloning Vector were sequenced

using 3.2 pM primer "1224" (5'-CGCCAGGGTTTTCCCAGTCACGAC) with 300-500 ng plasmid DNA template, 2X final concentration Big Dye reaction buffer and 2 gL Big Dye Terminator Sequencing mix in a final volume of 10 uL. Samples were overlayed with mineral oil and amplified using 25 cycles of PCR at 96C for 30 seconds, 500C for 15 seconds, and 60'C for 4 minutes. Unincorporated nucleotides were removed by spinning 9.5 ptL of the recovered PCR reaction through a column of 400 gL G-25 Sephadex at 800 g for 2 minutes. The recovered sample was then dried to completion, resuspended in 17 giL formamide sequencing buffer, heated to 95C for 2 minutes and quickly transferred to ice. Sequencing samples were analyzed on an Applied Biosystems Prism 310 Genetic Analyzer. Sequencing results were confirmed by examining the chromatograms manually using the "Sequencher" software package.


In Vitro Selection

DNAzyme libraries for the first round of selection were prepared by a single cycle of run-off PCR using the library template (5'-GTGCCAAGCTTACCGTCAC-N40GAGATGTCGCCATCTCTTCC (where N indicates equal molar concentrations of A, T, G, and C) and one of three different catalytic strand primers: "ang + ribose," "BJ + cage," or "BJ + de-protected cage" (the "BJ + de-protected cage" was the same primer



31





32

as "BJ + cage" but the caged-ribose had been de-protected by previous exposure to 300 pulses of laser (100 mJ/pulse)). Each primer was extended using either the standard dNTPs (equimolar T, A, G, and C) or with a non-standard, 5-position-functionalized nucleotide triphosphate (5-(3-aminoallyl)-2'-deoxyuridine 5'-triphosphate, referred to as E-base) substituted for T-triphosphate (equimolar E, A, G, C). This yielded six different IVS starting libraries,

libary #1: "BJ + cage" primer, standard dNTPS;
libary #2: "ang + ribose" primer, standard dNTPS;
libary #3: "BJ + cage" primer, AGC+E;
libary #4: "ang + ribose" primer, AGC+E;
libary #5: "BJ + de-protected cage" primer, standard dNTPS;
libary #6: "BJ + de-protected cage" primer, AGC+E)

Run-off reactions volumes ranged from three to 20 milliliters, each containing 1 ng/pL library template, 100 nM catalytic strand primer, 100 pM dNTPs, 10 mM KC1, 20 mM Tris-HCI (pH 8.8), 10 mM (NH4) 2SO4, 2 mM Mg2SO4, 0.1% Triton X-100, 20 units/mL Vent exo-, and 4.3 p.Curies/mL alpha-32P-CTP. One milliliter of each sample mixture (minus polymerase) was aliquoted into 1.5 mL eppendorf tubes, heated at 96*C for 8 minutes, and then slowly cooled to 55C over a period of 30 minutes. Polymerase was then added, and the samples were transferred to 720C for 15 minutes. The samples were then ethanol precipitated with ammonium acetate and with glycogen as a carrier by storing overnight at -80*C and then centrifuging for 40 minutes in Corex glass tubes at 10,000 rpms in a Beckman centrifuge at 4C. The ethanol was removed and the pellet was re-suspended in 200 p.L water. The samples were ethanol precipitated a second time in with ammonium acetate by centrifuging at 16,000 g in a desktop centrifuge for 20 minutes at 40C. After removing the ethanol, the pellet was resuspended in 80 p.L formamide stop dye and warmed at 37*C for 30 minutes to completely dissolve the pellet.



32





33


The samples were then run on an 8% PAGE-urea gel, and the full-length products excised using a sterile razor (rinsed thoroughly with ethanol between samples). Each gel slice was crushed and eluted in 350 fiL elution buffer (500 mM NHaOAC, 0.1 mM EDTA, 0.1% SDS) overnight. Samples were extracted in phenol/chloroform/isoamyl alcohol (25:24:1), then extracted in chloroform:isoamyl alcohol (24:1), and precipitated in ethanol with ammonium acetate. Pellets were resuspended in 50 mM HEPES (pH 7) and transferred to a new tube, at which point the final concentration was estimated using specific radioactivity. All samples were brought up to 327 nM, except library #1 ("BJ + cage" primer extended using standard dNTPS was brought up to 3270 nM). An aliquot (30 piL) of libary #1 was set aside for the IVS condition "E" (containing ten times the starting amount of library), and the remaining library #1 was diluted to 327 nM. An aliquot (30 pL) of each sample was then mixed with equal volume of the appropriate 2X reaction buffer as summarized in Table 2-1.

The samples were quickly heated to 96*C for 3 minutes, and cooled to 230C over 10 minutes. For samples without a caged-ribose (G, H, N and 0), the completion of the slow cool to 23C was considered the initiation of reaction. All samples were then exposed to 150 pulses of laser (90 mJ/pulse). The time of laser exposure was considered the time of reaction initiation for all samples containing a caged-ribose. Samples were allowed to react at 25*C for various times (20 to 720 minutes, see Table 2-1), and then stopped by adding 1.25 volume of formamide stop dye. The samples were then run on an 8% PAGE/urea gel next to a "cleaved library" marker. The area corresponding to the size of the cleaved library for each sample was excised, crushed, and eluted overnight (referred to as "survivors of round 1").





33





34




Table 2-1. Selection Conditions for Each of the Fifteen Parallel IVS Experiments.

Final Buffer:
50 mM HEPES Selection
pH7; 1M NaCl; Time
Sample Bases (AGC+) primer 1 mM of metal = (minutes)
A T-base BJ + cage --0-- 120
B T-base BJ+ cage Zn120
C T-base BJ + cage 20
D T-base BJ + cage 120
E BJ + cage 120
F T-base BJ + cage
G T-base BJ + de-protted
H T-base ang + ribose 120
I E-base BJ + cage --0120
J E-base BJ + cage Zn120
K E-base BJ + cage 20
L E-base BJ + cage 120
M E-base BJ + cage
N E-base BJ + d-protected cagh 120
O E-base ang + ribose 120



Following elution, samples were extracted in phenol/chloroform/isoamyl alcohol, then extracted in chloroform/isoamyl alcohol, and mixed with ammonium acetate and ethanol. Half of each sample was stored in ethanol, and the remaining half was precipitated and resuspended in 50 pL water. For subsequent rounds of selection, thirty microliters of the survivors of the previous round (in water) were used in a 100 pgL PCR containing 400 nM catalytic strand primer (either "ang +ribose," "BJ + cage," or "BJ + de-protected cage"), 100 nM complementary strand primer ("complementary + 5'34





35

phosphate"), 100 pM dNTPS (either AGCT or AGCE), 10 mM KCI, 20 mM Tris-HCI (pH 8.8), 10 mM (NH4)2SO4, 2 mM MgSO4, 0.1% Triton X-100, and 3 units Vent exopolymerase, and 7.5 plCuries alpha-32P-CTP. PCR amplification consisted of 3 minutes at 96C, followed by various numbers of cycles of 45 seconds at 96C, 45 seconds at 57C, and 2 minutes at 720C (a 2 minute extension was determined to be sufficiently long so as to yield approximately equal incorporation efficiency for both standard and nonstandard nucleotides).

Following PCR, samples were extracted in phenol/chloroform/isoamyl alcohol, then extracted chloroform/isoamyl alcohol, and precipitated with ammonium acetate and ethanol. Samples were re-suspended in exonuclease/buffer solution (25 gL final volume:

5 units lambda exonuclease enzyme, 67 mM glycine-KOH (pH 9.4), 2.5 mM MgCl2, 50 ptg/mL BSA) and incubated at 37*C for 30 minutes. Reactions were terminated by adding 60 pL formamide stop dye and heating at 80C for 10 minutes. The singlestranded products were loaded on an 8% PAGE/urea gel and full length ssDNA products excised. Gel purified samples were then extracted in phenol/chloroform/isoamyl alcohol, then extracted chloroform/isoamyl alcohol, and precipitated in ammonium acetate and ethanol. Samples were re-suspended in 50 mM HEPES buffer (pH 7) to a concentration of 25 30 nM. Each sample (25 jtL) was then mixed with equal volume of 2X reaction buffer (see Table I for buffer conditions for each sample), heated and slowly cooled, exposed to laser (150 pulses, 80-100 mJ/pulse), and stopped with formamide stop dye following various incubation times (see Table 1 for selection time used for each sample). This procedure was repeated for each round of selection.







35













CHAPTER 3
RESULTS OF 614 ANALYSIS

Research Objectives

Understanding why added functionality does not appearto significantly improve the catalytic power of nucleic acid enzymes under in vitro selection (IVS) schemes requires knowledge of how chemical functionality alters the distribution of catalytic potential within sequence space. This in turn requires that an experimental model system be comprehensively understood so that it can be modeled mathematically with sufficient accuracy to support quantitative analysis. Indeed, this depth of knowledge is important to understanding how molecular evolution in general affects phenotypic evolution in an evolutionary landscape over sequence space even up to organismal or ecosystem levels. The success of previous IVS has been only modest (described in the introduction), and apparently insufficient to explain the origins of life; this modest success of previous IVS may, in part, be attributable to several details of the IVS experimental system. One strong possibility is that the fastest catalysts are lost in the experimental work-up.

Our aim is two fold: (a) Examine the experimental systems in sufficient rigor so as to allow mathematical modeling, and (b) improve the selection procedure so as to minimize the potential loss of catalysts and allow quantitative analysis. To develop a system that supports detailed mathematical modeling of DNAzyme selections, we first examine in detail a model DNAzyme known as 614 (isolated from a previous IVS). To minimize potential loss of fast catalysts, we explore the use liquid phase, gel




36





37


electrophoresis-based selection, and the use of an ortho-nitrobenzyl-protected ribose ("caged-ribose") which can be quickly converted to the unprotected ribose by laser. An IVS is performed using these new procedures. Preliminary modeling of the experimental system is done to determine how various parameters of the experimental system have an impact on the mathematical model, and how the experimental system can be fine-tuned to improve modeling accuracy.


Developing Methods for Preparing Single-stranded DNAzymes by
Exonuclease Degradation of 5'-Phosphorylated Complementary Strand

Column purification of single-stranded DNAzymes present numerous problems, as outlined in the introduction. The ability to enzymatically remove the complementary strand was tested using lambda exonuclease. Lambda exonuclease is a processive 5' to 3' DNA exonuclease specific for 5'-phosphorylated DNA. Double stranded 614 DNAzyme was generated with PCR using a 5'-phosphorylated complementary strand primer ("complimentary + 5'-P") and a non-phosphorylated catalytic strand primer (either "ang + ribose" or "ang + cage"). Samples were internally labeled with alpha-32pCTP to allow visualization of both catalytic and complementary strands. Each of the two samples was then treated with or without exonuclease and/or strong base (which cleaves unprotected ribose) and run on a denaturing PAGE/urea gel. Figure 3-1 shows the results of this treatment (all figures at are at the end of the chapter).

Both untreated ribose- and caged-614 double stranded PCR products show two bands, the upper corresponding to the full size product, and the lower band corresponding roughly to the size of the cleaved product (lane 1 and 5). The amount of the lower band is significantly greater in the PCR product generated using the caged-ribose primer





37






38


(compare lane I and 5). This suggests that the lower band in lane 5, although the size of the catalytic strand cleaved at the ribose, is not actually cleaved product from the catalytic strand. This band is actually the result of polymerase termination when the complementary strand extension reaches the caged-ribose in the template. This is confirmed by the observation that this band is diminished by exonuclease treatment specific for the complementary strand (lane 7).

When the ribose-614 double stranded PCR product is treated with strong base, there is a significant increase in the lower band, corresponding to conversion of the fulllength catalytic strand to cleaved product. Comparing lane 1 with lane 2 shows that approximately half of the ribose-614 double stranded PCR product is converted to cleaved product following treatment with strong base (corresponding to the cleavage of the catalytic strand), while approximately half remains full-length (corresponding to the complementary strand). The catalytic strand, approximately half of originally double stranded ribose-614 PCR product, remains full-length following exonuclease treatment (lane 3), although there is no increase in the cleaved product (the complementary strand is degraded to single nucleotides). The full-length band remaining following treatment of ribose-614 with exonuclease disappears when also treated with strong base (lane 4). This full-length band remains for caged-614 treated with exonuclease and strong base (lane 8), demonstrating that conversion of the upper band to cleavage product seen in lane 4 corresponds to the catalytic strand (there is slight conversion of full-length caged-614 to cleaved product under strong base, possibly the result of partial de-protection under room light). The absence of any full-length product in lane 4 demonstrates that nearly the







39






39


entire catalytic strand has been converted to the cleaved product by the strong base and nearly the entire complementary strand has been degraded by exonuclease. Asymmetric PCR with "Tails," Followed by Gel Purification

A non-enzymatic method was also pursued for isolation of single-stranded

DNAzymes. In this procedure, termed "asymmetric PCR," the complementary strand primer was appended with a "tail" consisting of 10-20 nucleotides connected to the 5' end of the primer via a linker. Several linkers were tested for there ability to terminate polymerase extension when encountered on the template complementary strand. If polymerase extension of the catalytic strand in fact terminates when it encounters a linker on the complementary strand, all catalytic strands would be the normal full-length. However, all molecules generated by extension of the complementary strand primer would be increased in effective size by the length of the tail appended to the primer.

Figure 3-2 shows the product generated by PCR amplification of a 106-nt

fragment using 5 _32p-labeled primers and Taq polymerase. PCR amplifications were done in pairs: odd-numbered lanes show the product generated using the 5 _32p-labeled catalytic strand primer ("ang + ribose") with unlabeled complementary strand primer (designated in the subsequent lane), while even-numbered lanes show the product generated with unlabeled catalytic primer and 5'_32 P-labeled complementary strand primer. All products generated from complementary primers with tails are increased in size by the length of the tail, while products generated from the catalytic strand primer are of normal size (106 nucleotides) for each complementary strand primer except "Cl 18# 1 (lane 1 and 2). This demonstrates that Taq polymerase does in fact terminate effectively when it encounters a linker in the template for all linkers tested except




39





40


"C 18#1" (lane 1). It is believed that the complementary primer containing the "C 18#1" linker did not cause termination of the catalytic strand extension due to incorrect synthesis of the complementary strand primer (the same linker was used in C 18#2, and this successfully terminated extension).

The size differential between catalytic and complementary strands generated by asymmetric PCR was then used to isolate the two strands on an 8% PAGE/urea gel. Figure 3-3 shows the result of PCR in which both strands are labeled using alpha-32PCTP, demonstrating that all products are easily resolved. Lane 3 shows the full-length catalytic strand purified using exonuclease. Lane 4 shows two bands generated via asymmetric PCR, the 106-nucleotide band corresponding to the catalytic strand and the 121-nucleotide band corresponding to the complementary strand plus a 15-nucleotide tail. Treatment of the sample in lane 4 with strong base converts the majority of the 106nucleotide band to the 79-nucleotide cleavage product. A small amount of the 121nucleotide band is converted to the 94-nucleotide band by strong base treatment, indicating Taq polymerase can occasionally read through the linker of the complementary strand, thus generating a catalytic strand that is increased in size by the length of the tail.

As noted in the previous section, the presence of the caged-ribose in the catalytic strand occasionally causes strand termination, generating a complementary strand 27nucleotides shorter than full-length (the caged-ribose is at position 27- in the catalytic strand). Therefore, asymmetric PCR using a complementary strand primer with a 15nucleotide tail generates three bands, a 106-nt band corresponding to the normal sized catalytic strand, a 121-nt band corresponding to the full-length complementary strand





40





41


plus tail, and a 94-nt band resulting from termination of the complementary strand plus tail at the caged ribose.



DNAzyme 614 with uncaged- and caged-ribose Achieving maximal deprotection of caged ribose

A caged-ribose was synthesized and incorporated into the catalytic strand primer, known as "ang + cage" (an identical primer was synthesized with an unprotected ribose known as "ang + ribose". Using this caged-ribose, we are able to prevent cleavage at the ribose during the work up phase. This was demonstrated by incubating the caged-ribose primer, "ang + cage," in 0.5M NaOH at 80'C for one hour, resulting in only 6-10% conversion to the cleaved form (via base hydrolysis); the same conditions for the unprotected-ribose primer, "ang + ribose," resulted in 94% (90-97%) cleavage. The observation that about 6% of the unprotected-ribose primer does not convert to the cleaved form under base hydrolysis conditions suggest that as much as 6% of the "ang + ribose" primer may be missing the ribose-adenosine (due to imperfections in the DNA synthesis and purification).

To test the ability to de-protect the caged-ribose, "ang + cage" was exposed to increasing number of pulses of excimer laser (frequency = 308 nM, energy = 30-50 mJ/pulse) to induce conversion from the protected to unprotected ribose. The efficiency of conversion to unprotected ribose was assessed by incubating under base hydrolysis conditions (Figure 3-4). The amount of caged-primer cleaved by strong base increased with increasing laser, reaching a plateau of 80% cleaved with 100 pulses of laser. Added laser beyond 100 pulses did not further increase the amount of cleaved caged-614.





41





42

The amount of cleaved "ang + cage" (via base hydrolysis) at maximum de-protection (100 pulses of laser) was consistently about 10-15% lower than the cleavage obtained for "ang + ribose." This 10-15% reduction in cleavage for "ang + cage" must be the result of either two factors: incomplete de-protection or missing the caged-ribose adenosine. Although cleavage of caged-614 plateaus with 100 pulse of laser, showing no further cleavage even with 1000 pulses of laser, it is still possible that some caged-ribose molecules have not been hit by a laser photon.

The other possible explanation for reduced cleavage of de-protected caged-614 is the absence of the caged-ribose adenosine, again as a result of imperfect primer synthesis and purification. As much as 6% of the "ang + ribose" primer is believed to be missing the ribose-adenosine (above), but the caged-adenosine phosphoramidite may be incorporated into the primer more poorly than the ribose-adenosine phosphoramidite, resulting in as much as 20% of the "ang + ribose" primer missing the caged-adenosine. This issue is revisited in a following section "Cloning and Sequencing 614."


Kinetic profile of 614 with uncaged- and caged-ribose

The "ang + ribose" and "ang + cage" primers were incorporated into 614 via PCR. The catalytic strand of the PCR products, called "ribose-614" or "caged-614," were purified first by exonuclease degradation of the complementary strand (which was 5'-phosphorylated), followed by gel purification (8% PAGE/urea). The ability of "ribose-614" and caged-614 DNAzymes to self-cleave in reaction buffer (1 M NaC1, 1 mM MgCl2, 50 mM HEPES pH 7) was tested. Purified caged-614 was re-suspended in reaction buffer, heated to 97C for 3 minutes and cooled to 23C over 10 minutes (together this is referred to as slow cool), and incubated at room temperature for




42





43


approximately one hour. Cleavage was then initiated by exposing the sample to varying pulses of laser (0-1000 pulses). Cleavage of "ribose-614" was initiated by resuspending in reaction buffer and slow cooled. Figure 3-5 shows similar progress curves for both "ribose-614" and caged-614 cleavage. Both samples show identical initial cleavage kinetics, with cleavage plateaus near 50%. The caged-614 plateau is regularly about 1015% lower than "ribose-614." Several scenarios can explain the lower cleavage plateau seen with caged-614: (a) insufficient deprotection of caged-ribose, (b) the 614 sequence is missing the ribose-adenosine at position 27 (discussed above, and below "Cloning and Sequencing 614"), or (c) laser induced damage to the caged-614 DNA sequence. The reasons for both caged-614 and ribose-614 reaching cleavage plateau far below completion is explored below ("Understanding Reasons for Incomplete Cleavage of DNAzyme 614).

To confirm that the lower cleavage plateau of caged-614 was not due to

insufficient laser, kinetic experiments were conducted with additional pulses of laser. In one experiment, caged-614 cleavage was initiated with 150 pulses of laser and allowed to react for 73 hours, approaching cleavage plateau. The sample was divided into two aliquots, and one aliquot of the sample was exposed to an additional 150 pulses of laser. The added laser caused a slight increase in the amount cleaved, but this increase was immediate, and not increasing with time, suggesting against the possibility of additional de-protection of caged-ribose.

Another experiment was also performed in which caged-614 cleavage was

initiated with varying amounts of laser, from 0 to 1000 pulses. The cleavage plateau increased with increasing amounts of laser until 150 pulses of laser, after which





43





44


increasing amounts of laser did not cause an increase in cleavage plateau (Figure 3-6a and 7). When the percent of caged-614 product was normalized to account for the incomplete de-protection at low levels of laser (the fraction cleaved at each time point was divided by the fraction deprotected as determined by base hydrolysis), the kinetic profiles for caged-614 with varying amounts of laser overlap (except, of course, zero laser, which is not expected to cleave at all). This shows that they are all following the same kinetics parameters, with the plateau varying with the amount of laser (with maximum achieved at 150 pulses of laser).

The caged-614 exposed to increasing laser was also incubated under base

hydrolysis conditions. The amount of caged-614 cleaved by strong base increased with increasing laser, reaching a cleavage plateau with 150 pulses of laser. Added laser beyond 150 pulses did not further increase the amount of base-cleaved caged-614 (Figure 3-8). This directly parallels the results of caged primer, "ang + cage," plus laser cleaved by strong base condition, indicating that maximum de-protection has be obtained. Laser Does Not Damage Ribose-614

It is possible that caged-614 shows a lower cleavage plateau than ribose-614 because the laser (UV frequency) used to initiate the de-protection of the caged-ribose causes damage to the DNAzyme. Non-specific breaking of the phosphodiester backbone is expected to create a smear running from 1 to 106 nucleotides. Although slight smearing was noticed between the cleaved and full length product, there was no smearing below the full length product (79 nucleotides). Furthermore, this smearing was not present when ribose-614 was exposed to identical laser treatment.






44






45

The laser can potentially cause DNA damage which would inactive a DNAzyme without breaking the phosphodiester backbone (such as forming thymine-thymine dimers). We were unable to find evidence for thymine-thymine dimer formation when ribose-614 was exposed to laser (even though ribose-614 does contain two successive thymines at positions 82 and 83, and 100 and 101). If the laser is causing non-specific DNA damage resulting in a reduced cleavage plateau, this should be evident even without the caged-ribose. To test this, ribose-614 was exposed to increasing amounts of laser, from 0 to 300 pulses (100 mJ/pulse). No change was noticed in cleavage plateau or initial rates, suggesting laser does not cause damage to the DNA (Figure 3-9). DNAZvme 614 is NaCl-dependent. and MgCl2-independent

DNAzyme 614 was selected using the standard Breaker-Joyce IVS protocol and using a reaction buffer of 1 M NaC1, 1 mM MgC12, and 50 mM HEPES pH 7. This protocol is expected to create cis-cleaving enzymes that utilize the magnesium as a cofactor, while the sodium neutralizes the negative charge of the phosphodiester backbone and thus stabilizes folding. The Breaker-Joyce IVS protocol assumes that the DNAzymes generated may fold, but will not cleave without the added magnesium cofactor, and further, that the rate of cleavage will increase with increasing amounts of magnesium cofactor. Figure 3-10 shows that there is little difference in the kinetics of ribose-614 if is incubated with either 1 mM or 3 mM MgC12. Experiments in which the MgCI2 was completely removed and 3 mM EDTA was added also showed no change in cleavage rates. However, removing the NaCl from the reaction buffer completely prevents catalysis (Figure 3-11). DNAzyme 614 clearly is not magnesium-dependent, as expected under the Breaker-Joyce experimental set-up.




45





46


Supplementing the standard reaction buffer with 2 M ammonium acetate increases the rate of cleavage. Replacing the 1 M NaCl of the reaction buffer with IM ammonium acetate resulted in cleavage, although at a greatly reduced rate.



Understanding Reasons for Incomplete Cleavage of DNAzyme 614

Interpreting and modeling the results of an IVS requires a comprehensive

understanding of the mechanics of the model system. The previous section quantified the experimental parameters of using a caged-ribose that is de-protected by laser. Analysis of caged and ribose-614 both demonstrated a cleavage plateau far below "true completion" (defined as cleavage caused by strong base, which is -94% for "ang + ribose" and -80% for "ang +cage"). The explanation for the cleavage plateau must be common for both caged and ribose-614, and quite likely any other DNAzyme generated via this IVS procedure. Indeed, low cleavage plateau is an issue with many ribozymes and DNAzymes, but the specific causes for this reduced plateau is often ignored, presumably because the researchers either do not realize, or do not believe this phenomena is significant. We now seek to explain the failure for the reaction to reach "true completion," knowing that such information is crucial to both redesigning the IVS protocol and creating a reasonable model of the IVS experiment.



Incomplete De-protection of Caged-Ribose

We have already demonstrated that we have achieved maximum de-protection of the caged-ribose with 100 pulses of laser (meaning additional laser does not result in more de-protection). But is incomplete de-protection, even at maximum de-protection, a





46





47


plausible explanation for the failure to reach true completion? This possibility is ruled out on the basis that ribose-614 also reaches a plateau that is far below true completion. Although the maximal cleavage for caged-614 is consistently 10-15% lower than that of ribose-614 (both when self-cleaved and when cleaved by base hydrolysis), both reach cleavage plateaus that are as much as 35% below the amount of cleavage caused by strong base. The explanation for the reduced cleavage plateau must be common for both caged and ribose-614, and quite likely for any other DNAzyme generated via this IVS procedure.


Incomplete Removal of the Complementary Strand

Two methods were used for removing the complementary strand generated during the PCR synthesis of our DNAzymes, exonuclease degradation and asymmetric PCR. Asymmetric PCR has nearly no potential for complementary strand impurity since the complementary strand is 15 nucleotides longer than the catalytic strand, and thus easily separated from the catalytic strand during gel purification. Experiments in which "ribose-614" was prepared using asymmetric PCR followed by gel-purification, showed cleavage plateaus similar to ribose-614 generated with exonuclease, suggesting the cleavage plateau cannot be attributed to inhibition by contamination of the complementary strand.

When "ribose-614" generated via the exonuclease method was treated with strong base, similar amounts of conversion to the cleaved form were observed as seen for the ribose primer "ang + ribose" (85-95% for "ribose-614" vs. 90-97% for "ang + ribose). Since there is no possibility of a contaminating complementary strand in the "ang + ribose" primer, the -5% difference between base cleavage of the "ang + ribose" primer




47





48

and the full length DNAzyme "ribose-614" is the maximum amount attributable to incomplete degradation of the complementary strand by exonuclease.

Exonuclease efficiency was also tested by 5'-phosphorylating a single primer with gamma-32P-ATP. This radio-labeled primer was then used to synthesize full length "ribose-614." The double stranded product was then divided into two aliquots, and one was subjected to standard exonuclease treatment. Equal amounts of each sample were then electrophoresed on a PAGE-urea gel, and the amount of full-length product assessed. Exonuclease treated samples consistently showed less than 10% of the original material remained full length.

Experiments in which small amounts of the 614 complement (10% of the amount ribose-614) were added to ribose-614 had no impact on the cleavage plateau (Figure 3-12A and B). When cleavage experiments were performed with equal amounts of ribose-614 and its complement, however, the presence of the complementary strand dramatically reduced the cleavage plateau to nearly 30% (Figure 3-12 C and D). This demonstrates the importance of near-complete removal of the complementary strand to accurately examine DNAzyme kinetics. We believe both asymmetric PCR and exonuclease degradation are superior methods to column purification under strong base for removing the complementary strand.



Approach to Chemical Equilibrium

The failure of a substrate to completely transform itself to product may be the result of an approach to chemical equilibrium. In this case, the reverse reaction occurs fast enough to convert a significant fraction of the product back into substrate. We tested





48






49


to see if this was the case in our cleavage reaction by several methods. First, cleavage a 200 nM solution of caged-614 was initiated by 300 pulses of laser and allowed to react for 144 hours. At this time, near cleavage plateau, the reaction was split into three aliquots. One aliquot was unaltered. The second aliquot was diluted twenty-five fold in reaction buffer. Unlabeled d-614 was added to the third aliquot (to a final concentration of 200 nM). D-614 is the same sequence as ribose-614, but with a deoxy-adenosine used in the place of the ribose adenosine. In a later section we show that d-614 acts as an enzyme to cleave r-614 in trans. If the cleavage plateau is a result of equilibrium between the forward reaction and reverse reaction, the addition of additional enzyme (d614) should shift the equilibrium in favor the formation of more product over time. Similarly, diluting the reaction reduces the likelihood that an enzyme can find two substrates to ligate together, thus shifting the equilibrium in favor of cleavage (the cleavage reaction can occur in cis, and thus is not as strongly affected by dilution as ligation). Figure 3-13 shows that neither treatment significantly alters the cleavage plateau, arguing against equilibrium between the forward and reverse reaction.

In a second experiment, the 79-nucleotide cleavage product was isolated via gel purification. This cleavage product was radio-labeled, and following purification, was incubated (in excess) with full-length ribose-614 (not radio-labeled). Although unlabelled, previous studies have shown that the full-length 614 DNAzyme begins to cleave itself, generating both the 79 and 27-nucleotide products. If the reverse reaction (ligation) occurs, the excess of exogenous, radiolabeled 79-nucleotide fragment ensures that there is a greater chance for the radio-labeled 79-nucleotide fragment to become ligated to the 27-nucleotide fragment, thus causing an increase in size of the radio-labeled





49





50


79 nucleotide fragment over time. The reaction was monitored for 400 hours for a shift in size of the 79-nucleotide product to full-length (106 nucleotide), but no such conversion was observed. This argues against the possibility that a significant fraction of 614 cleavage products are ligated together. Are the 27- and 79-nucleotide Fragments Acting as Catalysts or Inhibitors?

Two products are created by the cleavage of 614 at the ribo-adenosine, a 27- and a 79-nucleotide fragment. The standard Breaker and Joyce IVS protocol was "designed" to select for molecules that are capable of self-cleavage (in cis), and cis cleavage is assumed to predominate. However, without making assumptions about the structure and mechanism of the 614 DNAzyme, we cannot rule out the possibility that either of the cleavage products continue to act as a ribonuclease DNAzyme, or as a ribonuclease DNAzyme inhibitor. We therefore set out to test this possibility using our model DNAzyme, 614.

A caged-614 cleavage assay was set up as normal and, following extended incubation in reaction buffer, the 27- and 79-nucleotide products were gel purified. These products were then added in trans to a new sample of full-length caged-614, as well as a caged-library sample (identical to caged-614, but containing a randomized 40nucleotide region between the catalytic and complementary primers). Following initiation with 150 pulses of laser and incubation in reaction buffer, cleavage of the new caged-614 sample occurred without change, even in the presence of equal concentration of the 27- and 79-nucleotide fragments (Figure 3-14). This suggests against the idea that the 27- and a 79-nucleotide fragments significantly affect 614 cleavage by acting as inhibitors or catalysts, at least when present in equal concentration (which is




50





51


approximately the highest concentration they reach under normal 614 cleavage conditions in which approximately half of 614 converts into cleavage products).

The caged-library alone did not show detectable catalysis under standard reaction conditions; this follows our expectation that an unselected, random library will have very few active catalysts. Addition of the 27-nucleotide product alone did not increase cagedlibrary cleavage, while addition of the 79-nucleotide product alone did slightly increase cleavage of the library. Cleavage levels were extremely low, however: while 614 had reached a cleavage plateau of 65% in about 50 hours, the library cleaved by the 79nucleotide fragment had reached only 15% cleavage in 300 hours. The addition of the 27-nucleotide fragment together with the 79-nucleotide fragment reduced cleavage of the library.

This shows that the 79 nucleotide fragment can act as a DNAzyme to cleave a random library substrate in trans, but the rate of cleavage is orders of magnitude slower that the full-length 614. The 27-nucleotide fragment reduces this already low activity of the 79-nucleotide fragment, suggesting it acts to inhibit cleavage by competing with the substrate (ribose-library) for binding to the 79-nucleotide fragment. These results are considered again in a later section when we explore potential structures of 614.



Improperly Folded DNAzvme 614

Having shown that the cleavage products do not significantly alter the cleavage plateau, we then set out to determine whether the plateau was due to the mis-folding of a fraction our DNAzyme into an inactive conformation. While most IVS protocols largely ignore the folding issue altogether (assuming folding happens quickly and efficiently, in





51





52


respect to cleavage, by re-suspending the DNAzyme in reaction buffer without cofactors), we included a re-folding step that is easily reproducible. Following resuspending the DNAzyme in reaction buffer, samples were heated to 97*C for three minutes to achieve complete denaturation. The samples were then cooled to 23*C over 10 minutes using a thermocycler. Although very consistent, even this folding procedure cannot guarantee that all DNAzymes find the same "active" conformation. Indeed, structure predictions of 614 show several potential structures, perhaps only one of which is active. Folding into inactive structures can potentially contribute to the low cleavage plateau.

When 614 cleavage assays are stopped by freezing to -80*C and run on a nondenaturing PAGE gel, two bands are present at initial time points. The first (highest) band represents '-15 % of the sample, and remains unchanged throughout the duration of the kinetic assay. The second (middle) band is initially 85% of the sample, but with time, is converted to a third (lowest) band (at 200 hours, the first band remains 15% of the total sample, while the second band is reduced to ~- 10%, and the third band is -75%). This result was observed when r-614 was purified via asymmetric PCR or by exonuclease degradation, suggesting that the unchanged upper band likely represents an alternative conformation of the catalytic strand rather than contaminating complementary strand.

We set out to test this possibility with two methods. First, ribose-614 was incubated under standard reaction conditions for 140 hours, allowing the reaction to approach the cleavage plateau. An aliquot of this sample was removed and denatured by adding 2 times the volume loading dye (95% formamide, EDTA, xylene cyanol, bromophenol blue) and heating at 90'C for 2 minutes. The sample was then





52






53


electrophoresed on a denaturing PAGE/urea gel, and the uncleaved ribose-614 was excised and purified. This previously uncleaved, gel-purified ribose-614 was ethanol precipitated, resuspended in reaction buffer, and folded using the slow cool protocol.

Cleavage progress following this gel-purification was then compared to an untreated sample. Twenty-five percent of the gel-purified, ribose-614 sample was cleaved in the 300 hours following gel-purification (Figure 3-15). The additional cleavage following the gel-purification suggests that some of the uncleaved sample is locked in an inactive structure. It is notable, however, that the gel-purified sample reached a cleavage plateau of approximately 25%, far lower than the cleavage plateau of the original sample (70%).

While part of this incomplete cleavage may again be attributed to mis-folded molecules, if mis-folding were the only cause of incomplete cleavage, a plateau of approximately 70% would be expected for the gel-purified sample. The fact that the gelpurified sample reaches a cleavage plateau far below 70% indicates that, while misfolding may account for as much as 25% of the uncleaved fraction, other reasons must account for the majority of the uncleaved fraction. Gel-purification allows mis-folded molecules the opportunity to refold and cleave, but it results in enrichment for molecules that are unable to cleave for reasons other than incorrect folding, such as missing the ribose-adenosine, or other mutations.

One can argue that gel-purification increases the cleavage of ribose-614 by

removing some inhibitory factor rather than allowing for proper refolding. Although the experiments described in previous sections demonstrated no significant inhibitory effect of the cleavage products on 614 cleavage, we tested the "mis-folding" possibility without





53





54


removing any potential inhibitors. We did this by again allowing 614 to reach cleavage plateau, and then denaturing an aliquot of the sample by heating it to 97C for 3 minutes and slowly cooling to 23C over 10 minutes. Reheating and slowly cooling allows any molecule "stuck" in an inactive conformation a second chance to adopt an active conformation. As in the gel-purification experiments, denaturing and refolding the sample via the slow cool procedure increases the amount of material cleaved (Figure 3-16). The increase in cleavage is significant (usually about 10%), but does not bring cleavage up to 94% (the amount cleaved by strong base and therefore considered "completion") and therefore cannot account entirely for the uncleaved 614 at plateau.


Cloning and Sequencing Cleaved and Uncleaved 614 Near Cleavage Plateau

The ability of strong base to cleave the ribose-primer, "ang + ribose," to only 9097% suggested that about 6% of the primer, and therefore any full length DNAzyme made from that primer, may be missing the ribose-adenosine. For the caged-primer, "ang + cage," base hydrolysis following laser deprotection resulted in only 80% cleavage, suggesting that the problem of missing ribose-adenosine may be even more severe for the caged-primer (this could also be due to incomplete deprotection, despite excess laser). Mutations may also occur throughout the DNAzyme, either as a result of mutations introduced during the synthesis of the primers, or mutations between the primers as a result of polymerase error. Together, these mutations may reduce or eliminate the catalytic power of a fraction of the DNAzyme pool. This possibility was examined by cloning and sequencing both cleaved and uncleaved 614 near the cleavage plateau (Figure 3-17). Of the 28 sequenced clones of 614 that were cleaved at the time of gel54





55

purification, only three (11%) were found to contain a mutation (all three were in the N40 region between the primers).

Of the 65 clones of 614 that were un-cleaved, 44 had at least one mutation (68%). Twelve clones (18% of those sequenced) had a mutation in the N40 region; thirty-one clones (46%) had a mutation in the "ang" primer; three clones (5%) had a mutation in the complementary primer. Of the uncleaved-614 clones sequenced, nine were from a 614 DNAzyme that originally contained a caged-ribose (cleavage initiated by laser). Of these nine clones, five (56%) were shown to be missing the ribose-adenosine, as opposed to seven of 56 clones (13%) isolated from uncleaved ribose-614. This confirms our original suspicion that a higher proportion of the "ang + cage" primer is missing the riboseadenosine as compared to the "ang + ribose" primer, and explains both the observation that about 15% less of the "ang + cage" primer with maximum de-protection is cleaved by strong base than the "ang + ribose" primer, and the observation that caged-614 reaches a cleavage plateau about 10% lower than ribose-614.

A caveat must be added, however, in that it has been observed that the polymerase tends to terminate primer elongation about 50% of the time when it encounters the cagedribose-adenosine on the template strand. Given this, it is possible that the polymerase skips over the caged-ribose-adenosine in the fraction that does not get terminated at the caged-ribose, and thus the full-length double stranded DNA that is required for cloning is missing the adenosine as a result of the polymerase skipping the caged-ribose. For this to occur, the caged-ribose must not have been fully de-protected by the initial laser treatment (polymerase does not terminate appreciably at an uncaged-ribose). Together, this confirms that the lower cleavage seen for caged-614, as compared to ribose-614, is a





55





56

result of only two possibilities, incomplete de-protection or a higher fraction of the primer missing the ribose-adenosine.

Several of the 614 clones with mutations were tested to see if their mutations in fact reduced the rate of cleavage. Figures 3-18 and 19 show the cleavage profiles of several mutants and demonstrate that nearly all of the mutations cause greatly reduced cleavage rates. Together with the observations above, we have shown that approximately 6% of the total ribose-614 sample does not cleave because it is missing the riboseadenosine (20% for the caged-614), approximately 10-15% of the total sample may be mis-folded and only slowly converts into the active conformation, and 68% of the uncleaved fraction may not cleave due to mutations.



DNAzyme 614 Cleaves in cis and trans

As is the case with all enzymes, a DNAzyme reaction can be thought of as

occurring in several steps: binding the substrate, the chemical step (where covalent bonds are made or broken), and release of the substrate. The IVS protocol was designed so that the DNAzyme contains the substrate (ribose-adenosine) linked to the enzyme (the random N40 region), and engineered such that the DNAzyme would fold back onto itself via two clamps of base pairing, one on each side of the ribose-adenosine, thus bringing the "enzymatic region" in proximity to the "substrate." Because of the design, the DNAzyme is expected to fold more rapidly with itself than bind other molecules at typical concentrations, and therefore the reaction is assumed be unimolecular. However, because all molecules in a library share the same 5'- and 3'-primers, over 60% of the







56





57

sequence is identical in all molecules (in particular, the regions surrounding the riboseadenosine).

This opens the possibility that the "enzymatic region" may bind to the "substrate region" of another molecule and either cleave in trans or inhibit cleavage. In the standard Breaker-Joyce protocol (which was used to generate 614), binding potential DNAzymes to a column is believed to prevent molecules from binding to and cleaving other molecules. Amplification and reselection of the survivors is also believed to preferentially enrich self-cleaving DNAzymes. It is possible, however, that an "enzymatic" DNAzyme can bind to a "substrate" molecule, and together they bind to the column via the biotin of one of the two molecules. This conjoined DNAzyme/substrate complex survives washing, but upon cleaving of the substrate, either both enzyme and substrate are released from the column (if the conjoined molecule was attached to the column via the biotin of the substrate), or only the "substrate" is washed off (if the conjoined molecule was attached to the column via the biotin of the enzyme). This may result in the inadvertent selection for DNAzymes that can cleave in trans, as well as molecules that are good substrates for trans-cleavage.

The "ang" primer used to generate DNAzyme 614 (as well as kinetic analysis) is nearly identical to the Breaker-Joyce primer, but with two single-nucleotide changes that reduce the strength of the base-pairing clamps designed to encourage cis folding. These changes further open up the possibility that cleavage can occur in trans. We are unaware of any published results that explicitly consider the impact of binding clamps on the outcome of selections, but based on the assumption that the protocol selects for cis57






58


cleaving activity of the random region, most research apparently assume the impact of these engineered sights is negligible.

However, some published accounts have reported that the rate of some

DNAzymes appears to decrease when they are generated with additional cycles of PCR (5 vs. 20 cycles too few cycles to expect a significant change in the amount of mutation for a 40-nucleotide product) (Geyer, 1997: Breaker, 1995). No explanation is offered for this result. Neither papers cited do not report the concentration of DNAzyme used in their kinetic analysis, presumably because they never considered the possibility of transcleavage (and it is therefore likely that the DNAzyme concentration was not controlled in their analysis). If the concentration is not directly controlled, additional cycles of PCR will obviously result in additional DNAzyme; it is plausible that the cleavage rates changed with additional cycles of PCR because the DNAzyme can act in trans and therefore the rate is concentration dependent. This next section explicitly explores this possibility.



Cleavage Rate of Ribose-614 Varies with 614 Concentration

If a DNAzyme exclusively self-cleaves, it generates product exclusively by a first order process, and the apparent first order rate constant should be independent of [DNAzyme]. A plot of apparent k~,bs versus [DNAzyme] should be a horizontal line. If product were generated exclusively by a second order process, then the apparent first order rate constant would be dependent upon [DNAzyme]. A plot of apparent kob versus [DNAzyme] should be a line that includes the (0,0) origin, with a slope equal to the second order rate constant. If product were generated exclusively by a second order





59





59


process with saturation (implying that, for the DNAzyme 614, one molecule of 614 must bind to another molecule of 614 for a reaction to occur, but that the 614-614 complex could disassociate before a reaction occurred, a condition well known in MichaelisMenten kinetics), then the apparent first order rate constant would be dependent upon [614] initially, and approach an asymptote at high [614]. Here, a plot of apparent 'cobs versus [614] should be a line that includes the (0,0) origin, sloping up and ultimately leveling to a plateau.

The concentration dependence of the rate of cleavage of ribose-614 was tested by examining the progress curve and the initial rates for ribose-614 cleavage at various different dilutions (430 nM to 0.3 n.M ribose-614). Results from this experiment (Figure

3 -20, 2 1, and 22) show a reality where several rate processes are operating at the same time.

Figure 3-21 shows that the initial rate of ribose-614 cleavage decreases with

decreasing concentration (a logarithmic abscissa is used to capture the full range of the experimental concentrations). The decrease in the initial rate of cleavage with decreasing concentration reaches a lower limit below which further dilution causes very little further reduction in the initial rate. This implies that the rate does have some concentration dependence and implies that substrate is being consumed by two paths: a first order rate process and, at least partly, by a process that is higher than first order. In the left portion of Figure 3-21 (1ow [614]), the first order process dominates the overall rate. In the right portion of Figure 3-21 (high [614]), second order processes dominate the overall rate. This experiment was repeated several times over with similar results.








59





60


In Figure 3-22A (the same data as Figure 3-21, but with a linear abscissa), the

intercept at [614] = 0 is not at the origin. This requires that at least some of the rate arise via a unimolecular process. Figure 3-22B shows a best-fit line extrapolated to zero (infinite dilution) for 614 concentrations below 3.5 nM (predominately unimolecular) yields an intercept corresponding to the unimolecular rate constant 0.0059 ht-' (half-life is 115 hours). The slope of this line is 8.9x105 M-Ihr', which for a second order process when [S]<
A plot of the natural log of uncleaved ribose-614 (substrate) versus time should be linear for a first order process; a linear relationship fit the data well for low ribose-614, but not for higher concentrations (Figure 3-23). The negative of the slope of the best-fit line for 0.3 nM ribose-614 is 0.0042x103 hr-<, which is in close agreement with the rate estimated by extrapolation of the initial rates of ribose-614 to infinite dilution (Figure 322A and B).

Further, it appears from Figure 3-22A that a plateau has not been reached. It may be that the [614] is simply not high enough in this experiment. However, this plot does not rule out the possibility that a purely bimolecular process (not involving saturation) is contributing to the overall rate of the reaction. Although this rate is low by normal ribozyme standards, and even the standard of some DNAzyme phosphodiesterases isolated via IVS, this unimolecular rate is comparable to the magnesium-independent catalyst isolated using IVS by Geyer (0.012 hour' for "G3 Family" DNAzyme before optimization).

We then tested to see if the progress curve for each concentration of ribose-614 fit best to one exponential (in which both the rate and plateau can vary) or the sum of two





60






61

exponential equations (in which both the rate and plateau can vary for each equation). At low ribose-614 concentrations (3.5 nM and below), the progress curves for cleavage fit well to a single exponential equation, whereas at higher concentrations the progress curves were fit better by the sum of two exponential equations. A single exponential curve fit to progress curve for 0.3 nM ribose-614 cleavage estimated a rate 0. 01 1X10-2 hr (with a predicted plateau of 6 1%). This agrees with the other estimations of the unimolecular rate made above (which, comparatively, are underestimations since they have not corrected for a plateau below 100%).

This suggests that at low concentrations (3.5 rim and below), the reaction behaves in unimolecular fashion while at higher concentrations the reaction is not well characterized by a single unimolecular rate law. Furthermore, we can conclude that, at low concentrations, where unimolecular processes dominate, a second conformer of cisfolded 614 with lower reactivity is not present at significant levels. In an earlier section ("Improperly Folded DNAzyme 614"), we already considered variants of 614 that were inactive (mutants, for example).

This is important because if these DNAzymes in general have a single active conformation, and react following a single exponential first order rate law, then the distribution of catalytic power within a population of related molecules can be modeled (at least as an approximation) by a transform that fits a spectrum of first order exponential processes to a progress curve.

The "non-unimolecular" component of ribose-614 cleavage is not purely bimolecular as tripling the concentration of ribose-614 does not increase the initial rate nine-fold as would be expected (but in fact it increases less than two-fold). Furthermore,





61





62

if the reaction where simply the sum of a unimolecular and bimolecular reaction, the progress curve should fit to the sum of two power functions (Y = AX + BX2, where X is the concentration of 614, Y is the initial rate (kobs), A is the first order rate constant and B is the second order rate constant). This model, however, was unable to provide a reasonable fit for the experimental results, suggesting that ribose-614 cleavage is not simply sum of a unimolecular and a bimolecular reaction.



Deoxyibose-614 Cleaves Various Ribose-Substrates

The observation that ribose-614 can cleave in a non-uni-molecular fashion suggests trans cleavage. This was tested directly by incubating 614 with various substrates, each containing the 5' catalytic motif and ribose-adenosine. Initially, 100 nM ribose-614 was incubated with 100 nM of 5'-32p-labelled "ang + ribose" primer. Figure 3-24 shows that both ribose-614 and "ang + ribose" are cleaved, clearly establishing trans cleavage ability of 614. The cleavage of ribose-614 in the presence of"ang + ribose" primer was lower than ribose-614 incubated alone, suggesting that the "ang + ribose" primer competes with ribose-614 for cleavage by ribose-614 (Figure 3-25).

We then tested whether a 614 DNAzyme synthesized with a deoxy-adenosine

instead of a ribose-adenosine (referred to as "d-614), and therefore unable to cleave itself, could cleave various substrates in trans. We again tested trans cleavage of the "ang + ribose" primer (Figure 3-24), as well as a mutant of 614 (known as "4up6," which has greatly reduced catalytic activity), a pool of molecules synthesized from the random library, and a individual clone from the random library (known as "lib6lo2," which has no detectable catalytic activity). Figure 3-26 shows that each substrate was cleaved by d62





63

614, although with varying degrees of efficiency. The observation that the singlenucleotide point mutation of 614, "4up6" (which is likely to fold similarly to 614), is cleaved the slowest by d-614, while the "ang + ribose" primer is cleaved faster than other full length molecules, agrees with the prediction that non-productive binding with "self' can reduce binding and catalysis by d-614.


Competition Studies of Ribose-614 Cleavage

We tested the ability of the same three substrates ("ang + ribose" primer, "ribose4up6," and "ribose-lib6lo2") to compete with r-614 for trans cleavage. The competitors were synthesized with a deoxy-adenosine in the place of the ribose-adenosine, and added in 9 fold excess over ribose-614. As a control, 9-fold excess of d-614 was added to r614. The addition of 270 nM d-614 to 30 nM r-614 increased cleavage to the same level as r-614 at 300 nM as a result of d-614 cleavage of r-614 in trans (Figure 3-27).

The cleavage of r-614 was reduced by the addition of"ang ribose" primer and "d-lib61o2," showing that these molecules compete for r-614 trans catalysis. However, the addition of 9-fold excess 614 mutant "4up6" to r-614 did not show the reduction in r614 cleavage seen by other competitors. Instead, the r-614 cleavage was increased, although the increase was far less than that seen with the addition of d-614. Two scenarios can explain this result. First, it is possible that although the catalytic activity of "4up6" is reduced, the reduction is offset by the great excess of trans acting "4up6." Alternatively, it is possible that the mutation in "4up6" does not alter its catalytic motif and activity, but instead the mutation prevents binding of the catalytic motif near its own substrate. Thus, incubating "4up6" alone shows almost no catalytic activity because the active catalytic motif cannot bind near its own substrate, but if incubated with another




63






64


molecule that does not have a mutation interfering with binding of the catalytic motif of 4up6 to the trans substrate, it may be able to cleave in trans.

The competition study described above was done at high concentrations of
"enzyme" (r-614) to favor inter-molecular association and trans cleavage. We performed a similar competition study, but with concentrations of r-614 and excess competitor sufficiently low as to favor uni-molecular cleavage. Under these conditions, association of the competitor to the enzyme which is inter-molecular is less favored than unimolecular association of the enzyme with its own substrate motif. As predicted, the addition of 3.0 nM d-lib6lo2 to 0.3 nM r-614 had no noticeable effect on r-614 cleavage (Figure 3-28).



d-614 Cleaves Faster in trans Than in cis: Single-tumover Kinetics

Substrate saturation experiments were performed with increasing concentration of "enzyme" (d-614) to determine the kcat(bi) for the bimolecular reaction. Figures 3-29, 30, and 31 show the full progress curves for cleavage, initial linear portion, and enzymesaturation profile for experiments with various substrates ("ang + ribose", r-lib6lo2, or r614) cleaved by d-614. The calculated Kd d is 39, 35, and 18 nM for r-primer, r-614 and r-lib6lo2 substrates, respectively. The kcat(bi), which is determined from the maximum rate of cleavage with saturating enzyme, is 0.040, 0.034, and 0.025 hr for r-primer, r614 and r-lib6lo2 substrates, respectively.

While the K for each substrate was relatively low (18-39 nM), the maximum rate was not attained until enzyme concentration reached approximately 2000 nM. Indeed, the initial rates appear to be increasing slightly even above 2000 nM, indicating that the





64






65


saturation may not be completely achieved (sufficient enzyme was not available to test kinetics at concentrations higher than 6000 nM). The slow approach to plateau is most notable for the r-lib6lo2 substrate, and to a lesser extent with r-6 14, suggesting that substrate or enzyme may exist in inactive, alternative conformations (an explaining the low W2 value for the association curve with r-lib6lo2).

A second experiment conducted with "ang + ribose" primer as the substrate and excess d-614 enzyme concentration increasing from 11I nM to 300 rnM also demonstrated progressively increasing rates up to 0.048 hr-1 at 300 nM (Figure 3-32). Similarly, r-lib6lo2 cleavage increased progressively when the enzyme (d-614) was in excess with concentrations ranging from 13 nM to 450 nM. Cleavage rates were always higher for "4ang + ribose" primer substrate than for r-lib6lo2 substrate at comparable d-614 concentrations, again suggesting that the shorter substrate may have less self-binding opportunities which compete with and therefore reduce binding and catalysis by d-614.

These enzyme excess saturation experiments reveal that the rate for trans

cleavage (kcat(bi) = 0.034 hr'1, for r-614) is distinctly higher than the rate for cis-cleavage (kcat(uni) = 0.006 hi). This 6-fold increase in rate is surprising given the expectation that tethering the "substrate" region to the "enzyme" region, as in the cis-cleaving 614, would significantly increase the effective concentration of the substrate relative to the enzyme, and thus dramatically increase the rate. When trans-acting enzyme reaches concentrations of 3.5 to 19 niM, the observed rate is twice the rate of cis-cleavage, and thus represents the concentration at which the cis-cleavage rate and trans-cleavage rate are equal (Figure 3-2 1). The effective molarity of linking the substrate to the enzyme, in







65






66


this case, is therfore approximately 3.5 19 nM; above this concentration, trans cleavage increasingly predominates.

The 6-fold increase in rate for trans cleavage is also surprising given the

expectation that both the "catalytic" and the "substrate" region of 614 is unchanged whether cleavage is in cis or trans. The implication is that the linker connecting the "substrate" and "enzyme" in the cis-acting 614 somehow interferes (steric hindrance) with binding in a way not present in the trans-acting 614.



d-614 Cleaves with Multiple-turnover

In order to be considered a true enzyme, the catalyst must be unchanged at the completion of its activity and thus be able to complete another round of catalysis. We have already demonstrated that cis-acting 614 cleaves itself, and that neither product retains significant catalytic activity, hence it behaves as an auto-catalytic molecule, but not classically as an enzyme. We have also demonstrated that d-614 can act catalytically in trans and is not altered in the process. We not proceed to test whether the catalytic activity of d-614 can function with multiple turnover.

Figure 3-33 and 34 show the result of incubating either "ang + ribose" or r-lib6lo2 substrate in four-fold excess over d-614 enzyme (either at 133:3 3 or 400: 100 nM substrate:enzyme). Multiple turnover was achieved, but rates were quite slow. At the highest concentration of enzyme and substrate, 2 turnovers of substrate was observed at 100 hours for "ang + ribose" primer substrate and 200 hours for the r-lib6lo2 substrate. As seen under single-turnover conditions, the "ang + ribose" substrate cleaved both faster, and at lower concentrations, than the full-length r-lib6lo2 substrate under multiple66






67

turnover conditions (compare the dramatic increase in multiple turnover cleavage for rlib6lo2 cleavage when d-614 concentration is increased from 33 to 100 nM with the modest increase for "ang + ribose" primer cleavage, indicating that the "ang + primer" cleavage is near its maximum rate of turnover). This may be attributable to the better ability of the "ang + ribose" substrate to bind with or disassociation from the enzyme, or its lower proclivity for forming nonproductive interactions.

Multiple turnover conditions were also tested by holding d-614 concentration constant at 20 nM and varying "ang + ribose" concentration from 100 to 2000 nM (Figure 3-35). Approximately two turnovers were observed in the first 100 hours, while only 5-6 turnovers were observed by 400 hours.

If the rate limiting step is disassociation of the substrate from the enzyme, we would expect to see a "burst phase" in the initial phase of multiple turnover kinetics in which all of the initially bound substrate is cleaved, followed by a slower phase in which the rate is limited by the product release. The duration of the burst phase corresponds to the fraction of the substrate that is initially bound to the enzyme. If the enzyme binds efficiently to the substrate, such that nearly all the enzyme has bound a substrate when the substrate is in excess, then this burst phase corresponds to the molar fraction of enzyme to substrate. However, if the enzyme does not bind substrate efficiently, or the substrate has competition for binding the enzyme, then the burst phase will be reduced. Figures 3-33, 34 and 35 show that the reaction remains linear well beyond the initial turnover, indicating that there is not a burst phase and that the overall rate is not limited by E-S disassociation.







67






68

Testing 614 Rate of Association and Disassociation


Values obtained above for cleavage rates of 614 are comparable to rates obtained for magnesium independent DNAzymes, but apparently slower than those reported for other ribonuclease DNAzymes generated by IVS (it is difficult to compare results directly since other DNAzymes are often characterized at higher temperatures and under alkaline conditions, both of which dramatically increase cleavage rates). One potential explanation for this may be the difference in the double straiid DNA clamps that were engineered to fold the DNAzyme onto itself and thus bring the presumably catalytic N40 region in proximity to the ribose substrate site. The lack of published studies on the impact of these engineered clamps presumably reflects an assumption by the IVS community that results obtained from in vitro selections should be independent of subtle changes in "starting conditions," and that the overall outcome is indicative of the inherent ability of nucleic acids to perform the selected behavior. This logic suggests that the two nucleotide changes made to the "BJ + ribose" primer to produce the "ang + ribose" primer should make little difference to the overall results. Nonetheless, these two "point mutations" have the potential to dramatically alter the ability of the DNAzyme to fold onto itself in the way engineered by Breaker and Joyce. The proclivity of 614 for acting in trans also suggests that DNA folding may be crucial our understanding of 614 activity.

In our standard protocol for initiating kinetics, enzyme and substrate are mixed together in reaction buffer, and then heated to 97T for 3 minutes, followed by a slow cool to 23*C over 10 minutes. The purpose of this "slow cool" is to ensure complete denaturation of substrate and enzyme at the onset of the experiment, and allow specific control of the folding conditions. Initially it was expected that catalysis was dominated




68






69

by cis-cleavage, but our studies above indicated a significant amount of trans-cleavage at concentrations higher than 3.5 19 nM. Whereas cis-cleavage entails only intramolecular folding, trans-cleavage entails two separate folding interactions, intramolecular folding (whether it is enzyme or substrate) and inter-molecular folding (whether it is between enzyme and substrate, or potentially between enzyme and inhibitor). Although phenomenalogically different, both types of folding occur simultaneously.

We tested the impact of our "slow cool" folding protocol on intra-molecular folding by comparing the cleavage of ribose-614 with and without the "slow cool" folding at a concentration low enough as to favor cis-cleavage (2 nM) (Figure 3-36). The overall cleavage of r-614 was higher when ribose-614 was completely denatured and slowly cooled, as opposed to simply re-suspending the pellet in HEPES buffer and initiating the reaction by adding NaCl and MgCI2, suggesting that the "slow cooling" procedure enhances intra-molecular folding. It is possible, however, that this rate enhancement is attributable to the denaturation induced by the high temperatures in the slow cool protocol, rather than enhancement of the intra-molecular folding, suggesting that a fraction of the ribose-614 is locked into an inactive conformation as a result of the ethanol precipitation. If this is the case, instead of improving intra-molecular folding, we must say that the "slow cool" procedure improves "intra-molecular re-folding."

When the same experiment was done at higher, trans-favoring concentrations, the initial rate of ribose-614 cleavage is unchanged, but the plateau is slightly higher with the "slow cooling" treatment (Figure 3-37). Similarly, cleavage of 1.2 nM r-614 by 150 nM d-614 was identical, regardless of whether the "enzyme" and "substrate" were mixed





69






70


prior to or subsequent to the "slow cool" folding protocol (data not shown). This suggests that, although the "slow cool" folding procedure enhances cis-cleavage, it has little effect on trans-cleavage (which, because it is faster, predominates the initial rate).

The impact of the "slow cool" folding protocol on inter-molecular folding was also tested directly by mixing the enzyme (d-614) and substrate (either r-lib6lo2 or "ang + ribose" primer) either before or immediately after the "slow cool" procedure (Figure 3-33 and 34). Multiple turnover experiments showed very little difference in cleavage, either in initial rates or cleavage plateau. In fact, completely omitting the denaturing/slow cool protocol had no effect on trans-cleavage (Figure 3-38). This further demonstrates that inter-molecular folding is not significantly effected by the "slow cool" protocol.

To test whether the rate-limiting step for ribose-614 trans-cleavage is the intermolecular association or the chemical step, a chase experiment was performed with cold competitor. Ribose-614 was incubated at sufficiently high concentration (200 nM) as to achieve significant trans-cleavage. The sample was allowed to fold for an additional thirty minutes after "slow cooling," at which point 15-fold excess unlabelled chase ("ang + ribose") was added. A significant decrease in cleavage was observed immediately following the addition of the cold chase (as compared to ribose-614 without added chase) (Figure 3-39). Comparing "200 nM ribose-614 plus chase" with "2 nM ribose-614 without chase" reveals that the chase has reduced cleavage levels below even that seen for a nearly all cis-cleavage reaction (Figure 3-40).

If the rate of substrate disassociation from enzyme is faster than the cleavage step (kcat(bi) k-1, the addition of a cold chase should completely eliminate subsequent





70






71


cleavage. If, however, kct(bi) > k-1, the addition of the chase should have no impact on subsequent cleavage (assuming sufficient time has been granted before the addition of the chase to ensure that all the substrate is bound by the enzyme). The addition of chase to r614 resulted in an intermediate result: cleavage was dramatically reduced, but cleavage continued to increase following the chase. Several scenarios may be put forward to explain this:

(a) The rate of disassociation of substrate from enzyme is within an order of
magnitude of the rate of cleavage (kvatQbi) k-).
(b) Cis-cleavage may not be completely quenched by the chase (or at least not at the
concentration of chase used). The residual cleavage following the addition of the chase could therefore be cis-cleavage. The apparent rate of cleavage with added
chase could be lower than the rate of cis-cleavage without chase if cis-cleavage
normally occurs through a process of folding, unfolding, and refolding, and at some point during the folded state, it self-cleaves. Given this scenario, adding
chase may reduce some cis-cleavage by competing with 614 for cis-folding. This
implies that ka(uf)kl) ni

(c) Not all of the substrate is bound by the enzyme at the time the chase is added, so
even though kcat(bi) >k1, the addition of the chase competes for association of
substrate to enzyme and subsequently eliminates cleavage of unbound substrate.
This implies that the rate limiting step is association (k.gb) kj).

(d) Insufficient chase has been added. If kca(bi)lk-1, the enzyme-substrate complex
will disassociate rapidly. If association is also relatively rapid, the enzyme may circulate through a number of chase molecules before it finds labeled substrate.
The reduction in cleavage is therefore proportional to the number of chase
molecules plus substrate molecules relative to enzyme molecules.


The chase experiment was therefore repeated with low concentrations of r-614

and high concentrations of d-614 (4n.M r-614, 2025 nM d-614 and 10 uM chase added at t = 4 hours, or 0.5 nM r-614, 64 nM d-614 and 1280 riM chase added at t = 45 minutes). Similar results were observed: "ribose-614 plus chase" was gradually cleaved, but always at levels far below the "minus chase" condition.





71






72


Previous enzyme-saturation experiments indicate that 2025 nM d-614 is sufficient to approach substrate saturation, therefore arguing against scenario (c) as a likely explanation for the fact that there is residual cleavage following the addition of the chase. The ratio of enzyme (d-614) to chase ("ang + ribose") in these latter two experiments was either 1:5 or 1:20. Slightly less cleavage was observed after the addition of chase in the 1:20 enzyme:chase experiment, suggesting scenario (d) as a possible explanation for residual cleavage following the chase.

Chase experiments conducted with ribose-614 cleaving in cis (2 nM) showed no significant change in cleavage following the addition of 30 nM chase. This suggests that either intra-molecular association is complete prior to the addition of the chase and it does not unfold prior to cleavage, or that intra-molecular folding is not affected by the presence of competitors at 30 nM. Unlike trans-acting catalysts, the absolute concentration not just the proportional dilution can alter the effectiveness of a chase. Previous experiment have demonstrated that low concentration of competitors (<3 nM) added at time = 0 do not compete with r-614 cleavage, while higher levels of competitor (270 nM) reduce, but do not eliminate, ribose-614 cleavage (30 nM) (Figure 3-27 and 28). Taken together, this suggests that at least some r-614 cis-cleavage can persist despite increasing concentrations of competitor, supporting the plausibility of scenario

(b).

A chase experiment was therefore performed using trace amounts of "ang + ribose" primer (4 riM ) as a substrate with near-saturating levels of d-614 (2000 nM) enzyme. This substrate has no intrinsic cis-cleaving potential, eliminating the possibility that cleavage following the chase is caused by cis-cleavage. Furthermore, the unlabelled





72





73

"6ang + ribose" chase was added in either 2.5- or 5-fold excess over the enzyme. The results indicate that cleavage was greatly reduced, but continued even after the addition of the chase (Figures 3-39 and 40). The reduction was slightly greater with additional chase. Together these experiments indicate that the rate of E-S disassociation is greater than the rate of chemical step (k.1 >> kcat(bi)) and the residual cleavage is a function of the ratio of substrate and chase to enzyme (scenario (d)).

The chase experiments indicate that inter-molecular association (folding) may be the rate-limiting step or that, once folded, the substrate disassociates from enzyme prior to cleavage. This was further explored by examining the temperature dependence of initial cleavage rates. If the chemical step is the rate-limiting step, lowering the temperature should lower the rate of cleavage, whereas raising the temperature will increase the rate (until the molecules denatures). However, when folding is rate limiting, lowering temperature is expected to stabilize folding and thus increase cleavage rate. Figures 3-41 and 42 shows that lowering incubation temperature increases the initial r-614 cleavage rate in both cis and trans conditions.

As the chemical step is expected to decrease approximately two-fold for every 10*C drop in temperature, the observation that a rate is unchanged by a 20*C change in temperature suggests that a balance is reached between a slower chemical step (kcat(uni) or kcat.bi)) and a more stabilized association step (kl(uni) or kl(b1)) at lower temperatures.

The enhancement of the folding/association step at lower temperatures is also seen with d-614 cleaving r-lib6lo2 in trans during single-turnover experiments (Figure 3-42B). Multiple-turnover experiments reveal a rate enhancement at lower temperatures during the first 20% of cleavage, following which the rate is reduced (Figure 3-3 8). This





73






74


"burst phase" seen under multiple turnover conditions at lower temperature (while not seen at higher temperatures) indicates lower temperatures increase the initial rate by stabilizing intermolecular association between enzyme and substrate; this stabilization also slows the disassociation of product from enzyme and therefore reduces the rate for multiple-turnover subsequent to the first catalysis.


Structural Analysis of 614

The ability of 614 to cleave in cis is not particularly surprising, both because this is the function for which it was selected, and because many ribozymes function to cleave in cis. It is, however, rather interesting that 614 also possesses the ability to cleave various substrates in trans, and furthermore, that the rate constant for trans cleavage is in fact 6-fold higher than the rate constant for cis-cleavage. The ability of 614 to cleave the "6ang + ribose" primer, as well as a library of molecules containing the "ang + ribose" primer, suggests that 614 has a "binding motif' which base-pairs with some part of the "6ang + ribose" primer common to all substrates. This binding positions a separate "catalytic motif' near the ribo-adenosine.

To understand how this may work in both cis and trans, the Stewart and Zuker

infold program was used to generate several structure predictions for 614 in cis and trans. Of the many structures proposed by the infold program, only those that positioned the ribo-adenosine in a "constrained but accessible" manner were selected structures were eliminated if the ribo-adenosine was either embedded in a lengthy helix (inaccessible) or if it was positioned in the middle of an extended loop (unconstrained).

Figure 3-43 shows eight of lowest free energy structure predictions. Despite the diversity of candidate structures, several common folds were present and used to group




74






75


structures into Families A, B, and C. The four structures of Family A each possess a 75GATA 78 stem loop (the superscript numbers refer to the position in the overall sequence), but vary in the fold of nucleotides 35 -70. Both structures in Family B share the same fold between nucleotides 50 and 77, but vary in the fold adopted by nucleotides 30 through 50 and 77 through 84. Structure AlI and Bi1 share the same fold for nucleotides 30 through 47. Both structures in the Family C share the same folding pattern for nucleotides 35 75.

Each of the three Families A, B and C can fold in such a way as to form basepairing between the '7CGACTCACTAT 17 (part of the 5'-"ang + ribose" primer) and 15 GTAGTGACG93 (part of the 3' primer). These 5' and 3' regions can also fold into separate hairpins, preventing interaction with each other, as seen in structure C2. The 3' hairpin formed between nucleotides 78 and 101 disrupts the defining elements of both Family A and B structures, but not the Family C structure (as seen in C2).

We have demonstrated that 614 can cleave "ang + ribose" primer (both in trans and when incorporated as nucleotides 1-49 of 614). If we assume that the structure formed when 614 cleaves in cis is similar to the structure formed when 614 cleaves "ang + ribose" primer in trans, then this suggests that the cis-folded 614 has some base-pairing between its first 49-nt and the remaining 3' end. The structure shown in C2 allows only minimal interaction between the "ang + ribose" primer region and the remaining 3' half (base pairing between 13 CATGGC40 and 71GCCTG 75) However, the base-pairing between nucleotides '7CGACTCACTAT 17 and 85GTAGTGACG9' ee in the structures A 1 -4, B 1, B2 and ClI allows significant interaction between the 5' and 3' halves of 614.







75





76

These latter structures therefore seem more probable for both cis-cleavage and transcleavage.

This was tested directly by mutating nucleotides T86 to A, G88 to C, and G90 to C in 614 (Figure 3-44). These introduced mutations greatly diminished the ability of 614 to cleave itself, as well as the substrate r-lib6lo2 in trans (compared to wild-type 614). Introducing the compensatory changes (C22G, C24G, A26T) predicted by the structures of A1-4, B1, B2 and C1 into the substrate r-lib6lo2 significantly improved trans cleavage. Introducing the compensatory changes C22G, C24G, A26T into the ribose-614 containing T86A, G88C, and G90C also dramatically improved catalysis. This demonstrates that the helix formed between 17CGACTCACTAT27 and 85GTAGTGACG93 shown in the structures A1-4, B1, B2 and Cl is probaby formed in cis and trans.

Attempts were made to eliminate some of the remaining candidate structures. Although structure C l appears quite "linear" and does not appear to position a folded "catalytic motif' near the ribo-adenosine, it does possess the requisite base-pairing between the 17CGACTCACTAT27 and 85GTAGTGACG93 regions. Because the 5' 49nucleotides of 614 correspond to the "ang + ribose" substrate, the essential elements of structure C1 could be formed by a 614/"ang + ribose" enzyme/substrate conjugate if the "ang + ribose" primer displaces the 5' half of 614. If structure Cl were indeed the correct structure, the 3' half of 614 (nucleotides 56 through 94) should be sufficient to bind and cleave the "ang + ribose" substrate. This was tested by generating the 3' half of 614 (both "614 56-94nt" and "614 56-106nt"). Truncating 614 at position 94 eliminates the possibility of forming structure C2, while still permitting structure C When mixed







76





77


with "ang + ribose" primer, neither "614 56-94nt" nor "614 56-106nf 'catalyzed cleavage of "ang + ribose", suggesting that structure C I is not correct.

Screening of mutants of 614 revealed that any single mutation of position 55, 65 or 72 greatly reduced catalytic activity. Either of the G55A or G65A mutations have the potential to disrupt the structure of A4. Each of these mutants was made with the compensatory mutations predicted by structure A4 (614:G55A/C47T, and 614:G65A/C40T). Neither compensatory mutation improved catalysis over the single mutant (Figure 3-45), arguing against the validity of structure A4.

The C72T mutant also had reduced catalytic activity; this mutation is predicted to disrupt a base-pair with G81 in all of the Family A structures. The validity of these structures was tested by making the compensatory G8 I A mutation. The double-mutant, however, cleaved even slower than the single mutant, suggesting against the validity of any of the Family A structures (Figure 3-45). It is, however, possible this stem-loop predicted in Family A is in fact part of the correct structure, but the GC base-pair cannot be replaced with an AT base-pair (either because the strength of the AT base-pair is lower, or because the original G or C played a catalytic function in addition to its structural role in the stem loop).

Position 81 of wild-type 614 was therefore converted from a G to a C (named "614:G8 I C"). The G8 I C mutation has the potential to disrupt the structures of both Family A and the structure B2, although in different manners. As mentioned above, G81 is predicted to base-pair with C72 in Family A structures; G81 is predicted to base pair with C43 in structure B2. The 614:G8 1 C mutant demonstrated greatly reduced selfcleavage, allowing us to test both Family A and B2 structures by making the predicted





77






78


compensatory mutations ("614 G8l1C, C72G" and "614 G8lIC, C4OG). Neither of the double-mutations restored catalytic activity, suggesting that neither structure B2 nor the structures of Family A are in fact the actual structure. It remains possible, however, that one of the structures of Family A or structure B2 is in fact correct, but G8 1 is crucial to catalysis for reasons beyond its ability to base-pair with either C72 or C43.

The evidence above argues against the validity of Family A and C structures, as well as structure B2, leaving only one candidate structure remaining (B 1). If structure B I represents the correct fold for 614 acting in cis, and the essential structural elements are preserved when 614 cleaves in trans, then nucleotides 41 through 91 should be sufficient for cleaving the "ang + ribose" substrate. Because this truncated 614 ("1614 trunc4l-9 I") can conceivably adopt the other structures of Family A and C, the "ang + ribose" substrate was altered to improve binding if it adopts the B 1 structure but not by other structures (this "B 1 substrate" was produced by truncating "ang + ribose" to include only nucleotides 2 1-35, with a G inserted between nucleotides 31 and 32, and A34 changed to T, resulting in 5'TCACTATrAGGAGAGTC"). The catalytic ability of "614 trunc4l-91" was then tested with both "ang + ribose" and "B 1 substrate." Neither substrate was cleaved by "614 trunc4l-9 I," arguing against our final proposed structures, AlI or BL

Structure predictions were made for Family A, B and C with 614 acting in trans (Figure 3-46). The ability of 614 to cleave the "ang + ribose" substrate demonstrates that nucleotides 50-106 are not a required element of the substrate for 614 trans-cleavage. The structure predictions shown for 614 cleaving in trans are those using "ang + ribose" substrate, but are essentially the same for full-length substrates.





79

Figure 3-46 shows that a four base-pair helix can be formed between nucleotides 35CATG38 of 614 and the identical nucleotides of the "ang + ribose" substrate. This helix is not formed when 614 cleaves in cis because the sequence CATG sequence occurs only once. This short helix presumably offers additional stabilization between the "ang + ribose" substrate and the 614 enzyme without disrupting potentially catalytic helixes of structures Al and C (although it does disrupt part of structures A2 and A3).

Interestingly, a mutational screening of 614 isolated a G38A mutant (called

"june#2 G38A") that had reduced activity (compared to wild-type 614) when tested at 100 nM (a concentration at which significant trans cleavage occurs).

The significance of the predicted 35CATG38 trans-helix was tested by introducing the compensatory mutation into the non-catalytic substrate r-lib6lo2 (the compensatory mutant is called "lib6lo2 C35T"). Previous trans-cleavage assays demonstrated that wild-type d-614 can cleave ribose-lib6lo2. The 614 mutant "june#2 G38A" was then tested for its ability to cleave either "r-lib6lo2" or "r-lib6lo2 C35T." Figure 3-47 shows that the introduction of the C35T mutation into the lib6lo2 substrate improves cleavage by the 614 mutant "june#2 G38A," suggesting that the 35CATG3 helix is indeed important for trans-cleavage. The reduction in cleavage seen when "june#2 G38A" cleaves "lib6lo2 C35T" as compared to wild-type 614 cleaving wild-type r-lib6lo2 is likely the result of the weaker AT base-pair as opposed to the GC base-pair.

Structure predictions were made with 614 bound to the "ang + ribose" substrate via the two experimentally supported interactions: base-pairing between 35CATG38 of the 614 and the 35CATG38 of the substrate, and base-pairing between the 17CGACTCACTAT27 region of the substrate and the 85GTAGTGACG93 region of the





79






80

enzyme. These trans-interactions interactions eliminate structures formed for cis-folded 614 in the first 38 nucleotides. Most of the structural elements predicted for cis-folded 614 are preserved in the trans-folded 614/substrate complex (Figure 3-46). The freeenergy predictions for the trans-folded 614/substrate complexes were all much lower than the cis-folded 614 structures (cis-61 4 dG = approximately 12 kcal/mol, trans 614:substrate dG = -20 kcal/mol). This agrees well with previous experimental results indicating that overall cleavage rate is affected by folding and occurs faster in trans.

Structural probing was also performed using chemical modifications specific for single-stranded DNA. The trans-folded conformation was favored by incubating high concentration of 5'-labeled d-614 (300 nM) in reaction buffer with and without excess unlabeled primer (10 pM). The samples were allowed to incubate overnight to maximize folding, after which they were treated with either potassium permanganate (specifically modifies T) or DMS (specifically modifies G at high salt concentrations) for two minutes and five minutes. Recovered samples were run on a PAGE-urea gel next to a l0-bp ladder. All T and G nucleotides demonstrated some sensitivity to the respective chemical modifications, indicating that either some 614 is not folded at all, or adopts multiple conformations. Nonetheless, contrast levels were adjusted to discern the most-protected nucleotides (suggesting they are base-paired). Figure 3-48 shows the result of this analysis, indicating that T44, T46, T74, G33, G41, G65, G71, and G75 all show some degree of differential protection. The pattern of protected and sensitive positions observed under chemical modifications is not simultaneously compatible with any single structure predicted in Figure 3-43 for cis-folded 614 or Figure 3-46 for trans-folded 614,







80






81


suggesting that either a mixture of multiple trans-folded conformers exists, or none of the predicted structures accurately reflect the true structure of 614.

Structural investigations of 614 are continuing. It is interesting to note that a vast majority of reported structures for in vitro selected DNAzymes are based on unverified predictions using the infold program. The infold program has been shown to be useful for providing plausible structures, but our experience with 614 demonstrates that unverified predictions of DNAzymes should be taken very lightly.



Summer of Results

Table 3-1 (on next page) summarizes the major results of experiments described in this chapter.






























81





82


Table 3-1. Summary of Results from Chapter 3


Reasons for cleavage plateau of 614
Missing ribose-A in -5% of primer (higher for caged ribose due to synthesis).
Alternative inactive conformation 15%.
Mutations that lower catalytic activity in both primer and N40 regions.
Not insufficient laser.
Not complementary strand inhibition.

Kinetic analysis
614 cleaves in cis (unimolecular) at low concentrations (< 20 nM).
614 cleaves in trans (biomolecular) at high concentrations.
614 cleaves various substrates in trans.
kcat(bi)= 0.035 hrI
kobs(uni)= 0.006 hr
The increase in the observed rate of cleavage at lower temperatures indicates
that the rate is a balance between the chemical step and the association step.
The rate of E*S disassociation (k1(bi)) for trans-cleavage is greater than the
chemical step (kcat(bi))

Structural Analysis
Many structures for cis-folded 614 are possible, but none have yet to be proven.
There is an interaction between 7CGACTCACTAT27 and 85GTAGTGACG93
that occurs in both cis and trans-folded 614.
There is an interaction between 35CATG38 of the Enzyme (614) and Substrate
that cannot occur in cis-folded 614.
























82





83

















Lamda exonuclease + + + +
strong base + + + +
ribose or caged-r R R R R C C C C
LANE # 1 2 3 4 5 6 7 8


106-nt full-length 614 > of





79-nt cleaved product > 0








Figure 3-1. Lamda exonuclease degrades the 5'-phosphorylated complementary stand of ribose-614 and caged-614. Ribose-614 and caged-614 were generated via PCR; both the catalytic and complementary strand were labeled with alpha-CTP. Treatment of double stranded ribose-614 with strong base converts the catalytic strand to the cleaved product, while treatment with exonuclease degrades the complementary strand to single-nucleotides (not shown on gel). PCR with the caged-ribose primer generates a complementary strand that is that is the same length as the cleaved catalytic strand, corresponding to the where the polymerase terminates elongation when encountering the caged-ribose in the template.





84





X '0


A A A A A A A 126 nt >
116 nt>
106 nt >


lane 1 2 3 4 5 6 7 8 9 10 11 12 13 14

106 nucleotide template extended with: A 5'-end labelled complimentary strand primer: "short" + linker + tail
5'-end labelled catalytic strand primer: "ang + ribose"



Figure 3-2. Asymmetric PCR using various linkers increases the size of the complementary strand without increasing the size of the catalytic strand. PCR reactions where done in pairs with the catalytic strand primer end labeled in the oddnumbered lanes, while the complementary strand primer was end-labeled in the evennumbered lanes.


+ + + tail on complimentary strand primer
+ + + caged ribose on catalytic strand primer
+ exonuclease
+ NaOH

< 121 nt: full length+ 15nttail

< 106 nt: full length product

< 92 nt: full length product + 15 nt tail 29 nt 5'of cage

< 79 nt: cleaved product

lane # 1 2 3 4 5

Figure 3-3. Size difference between catalytic and complementary strands generated by asymmetric PCR allow for gel-purification of the catalytic strand.
The presence of the cage-ribose in the template causes truncation of the complementary strand. Asymmetric PCR therefore generates a complementary strand that is either longer or shorter than the catalytic strand.





85



100- -100

90- -90

80- -0

70- -70

60- 40

50- -s

~40

30- -30

20- -20

10- -10

0 n I I 1 -0

caged primer plus var~n pulses of laser, +NaOH #








Figure 3-4. Maximum deprotection of caged-ribose primers occurs with 100 pulses of laser. End-labeled "ang + cage" primer was exposed to increasing pulses of laser (308 nM, 50 mJ/pulse) and the amount of deprotection determined by subsequent cleavage in 0.5 M NaOH for 60 minutes at 80*C. Cleavage reaches a plateau of approximately 80% with 100 pulses of laser. "Ang + cage" primer exposed to 1360 pulses of laser, but not treated with strong base, showed approximately 1% cleavage. Unprotected "ang + ribose" primer was 90-97% cleaved under strong base conditions. Exonuclease purified fuill-length ribose-614 was 85-95% cleaved under strong base conditions. indicates incubation with 0.25M NaOH rather than 0.5M.





86


















75
A A r-614, no lawe


so c-6144OOnM +Iae



i25


0I
0 25 50 75 100 125 150 175 200
hours












Figure 3-5. Caged-614 shows similar kinetics to ribose-614 when deprotected by 150 pulses of laser. For all kinetic analysis, error is estimated to be +/- 0.05 hour and less than +/- 5 percent cleaved. Identical experiments conducted in parallel fell within these error estimates (see Figure 3-30B as an example).






87



A
70



50-~ r-614 +O0laser
5 c-614 + 1000 laser
4- c-614 + 300 laser
30 --*c-614 + 150 laser
30
b. c-614 + 50 laser
20 c-614+ 1Olaser
C-614 + 0 laser
10


0 25 50 75 100 125 150 175
hours




B
75"


Sibos 614 + laser
Sc-614 + 1000 laser

Sc C-614 + 300 laser I c-614 + 150 laser
125- c-614 + 50 laser
A c-614 + 10 laser a c-614 + 0 Iaser


0o ; 160 ii5 150 175'

hours




Figure 3-6. Progress curves of caged-614 cleavage demonstrate maximum deprotection with 150 pulses of laser. (A) Caged-614 with less than 150 pulses of laser had greatly reduced cleavage plateau, indicating that the sample was not completely deprotected. (B) When cleavage was normalized by adjusting the percent cleaved out of only the fraction of sample that was deprotected (determined by NaOH treatment, the kinetic profiles overlap for all samples (except 0 laser).





88

60"

55040

30j20

1001

nunmr of lase pues Figure 3-7. The cleavage plateau of caged-614 reaches a maximum with 150 pulses of laser. Samples of caged-614 were initiated with increasing amounts of laser and allowed to self-cleave until the reaction reached plateau. Samples initiated with 150 pulses of laser and more reached equivalent cleavage plateaus of -50% (about
-10% lower than for an equivalent experiment with ribose-614.


100


75

I "I

25


0


number of laser pulm. Figure 3-8. Maximum deprotection of caged-614 occurs with 150 pulses of laser.
Samples of caged-614 were exposed to increasing amounts of laser and then incubated with strong base for 1 hour at 80"C to determine the amount of deprotection. Caged614 exposed to 150 pulses of laser is 75% cleaved by the strong base. Additional laser does not increase the amount cleaved by strong base. Ribose-614 has slightly higher cleavage under similar conditions.





89



A
50
'o 40- m r614 + 300 laser
I A r614 + 150 laser
30- r614 + 50 laser
r614 + 10 laser
0
r614 + 0 laser
10

0I I
0 100 200 300
hours

B
7.5


5.0- r-614 + 300 laser
>50r-614+ 150laser ., r-614 + 50 aser
# r-614 + 10 laser

2.5- r-614 + 0 laser



0.0- 1
0.0 2.5 5.0 7.5 10.0 12.5 15.0 hours


Figure 3-9. Ribose-614 is not damaged by laser. (A) Full time course, and (B) linear initial phase, of ribose-614 self cleavage following increasing amounts of laser pulses. The cleavage profile of ribose-614 is unchanged by exposure to increasing number of laser pulses.




Full Text
138
apparently chaotic. If the distribution of catalytic function can in fact be approximated
with a function, it seems reasonable that faster catalysts will be less abundant than slower
catalysts, and therefore a decreasing function should be used. The simulations described
here have examined only two of these functions: a linear and an exponential.
The final parameter considered in this model is leakage, defined as the fraction
of the pool at each round of selection that survives that round of selection independent of
both its intrinsic rate (sequence dependent) and the St. As a first approximation, leakage
can be considered the background rate of cleavage by the magnesium ion at pH 7.
However, experimental deficiencies are likely to be a more significant contributor to
leakage than the background, magnesium-dependent cleavage. Leakage is therefore
introduced to accommodate for deficiencies of the experimental system, and background
cleavage is considered separately.
In selections using biotinylated DNAzymes bound to a streptavidin column,
leakage results from molecules that are eluted from the column after the initiation of the
selection. This leakage may arise from molecules that never were never bound in the first
place (withheld in the matrix), molecules uncleaved but released due to incomplete
streptavidin/biotin interaction, or molecules cleaved by the NaOH wash. IVS based on
size separation using gel electrophoresis can have leakage resulting from incomplete size
separation such that some gel extraction inadvertently collects incorrect size molecules.
In our IVS system, leakage may also be the result of laser induced cleavage.
Survival of any particular molecule is based on its probability of being cleaved
given a Selection time (St), and is therefore independent of other molecules. The
probability of being cleaved, P(cleavage), is modeled assuming that each catalysts


12
Although this is the most reasonable hypothesis, it is difficult to imagine that such a
complex function could be generated from a random polymer. An experimental
approach, in vitro selection (IVS), emerged in the 1980s that served as a bridge between
the work forward and work backward approaches by permitting a direct test of the
catalytic power of random-polymer libraries (Szostak, 1992; Beaudry, 1992; Irvine,
1991).
IVS begins with a library containing ~10u unique and randomly generated polymers of
DNA. This library is then challenged by a selection process (such as the ability to
catalyze a particular reaction or bind a particular molecule). Molecules that can perform
the selected task are separated from those that do not (Figure 1-1). Finding molecules in
a random library that perform the selected task is therefore a function the abundance of
that activity, and the ability of the selection regime to separate active from inactive
molecules.
Combinatorial library
of DNA molecules
Solid support

- t
m
Random region
: s
: z
IT
^ mini nmiii in
Denaturation and elution
of complementary strand
3
5'
Binding of PCR
products to solid
support
PCR amplification
-H
'Mil
^ E h
Figure 1-1. A schematic diagram describing in vitro selection for a DNA molecule
that catalyzes a cleavage reaction.
12


105
c
[E = d614] nM
Figure 3-30. Single-turnover kinetics for d-614 cleaving r-614. (A) Fulltime
course. (B) Initial (linear) phase of cleavage. (C) Substrate saturation curve.
Substrate saturation experiments were conducted by incubating r-614 (4 nM) with
increasing amounts of d-614 enzyme.


13
In practice, the selection process is not perfect, allowing inactive molecules to
survive (and presumably losing some active molecules). When active molecules are rare
and background is high, several sequential selections are required to enrich for active
molecules. Experimentally, IVS accomplishes sequential rounds of selection by using
the surviving molecules from the first round of selection as a template for PCR, thus
amplifying the survivors and creating a second pool of DNA molecules (that are
presumably enriched for catalysts) for a subsequent round of selection.
This experimental strategy is a close analogy to what is imagined to have led to
the first life-form (with the exception that prebiotic mechanisms, as opposed to human
intervention, lead to the development of the polymer library, and selection for desired
activity was imposed by the Earths environment and processes which lead to the
decomposition of polymers). It was anticipated that IVS would now demonstrate that
the chemical processes necessary for life could in fact be found in a prebiotic library of
polymers. After a decade of experiments and great effort, many catalysts have indeed
been isolated from random libraries, most notably a phosphodiesterase, a Diels-Alderase,
a nucleotide ligase, and a limited polymerase (Breaker and Joyce, 1994; Johnston, 2001;
Seeling, 1999; Tarasow, 1997).
Despite these successes, getting self-replicating nucleic acid polymers from IVS
proved harder than anticipated/hoped. Bartel and others are coming close to achieving
this goal (Johnston, 2001). It must be noted, however, that even the authors of this
research contribute the success of the de novo generation of an RNA based RNA
polymerase to a stepwise procedure guided by human chemical knowledge: Thus far,
efforts to select for polymerization activity in a single step directly from random-sequence
13


115
A
B
0.07-]
0.06-
t3
ra 0.05-1
a>
o
c
.2
ts
re
£ 0.03-
0
co
k.
ro
C
0.04-
0.02-
0.01-
0.00
10 15 20 25 30
temperature
35
40
45
Figure 41. Ribose-614 fra/is-cleavage is increased by lowering the temperature.
(A) Full time course. (B) Plot of initial rates vs. temperature. Ribose-614 (200 nM)
was incubated at various temperatures from 4C to 42C. Temperatures below the
selection temperature of 25C increased the rate of cleavage, indicating that the rate
limiting step is the folding/association step rather than the chemical step.


106
c
Figure 3-31. Single-turnover kinetics for d-614 cleaving r-lib61o2. (A) Fulltime
course. (B) Initial (linear) phase of cleavage. (C) Substrate saturation curve.
Substrate saturation experiments were conducted by incubating r-lib61o2 (4 nM) with
increasing amounts of d-614 enzyme.


selection conditions, can dramatically alter the abundance and distribution of catalysts
without altering the rate of the fastest catalyst.
The goal of this thesis is to develop methods for estimating the distribution of
catalysts within a random library of DNA sequence. To aid in the development of
realistic models, we began with an in depth analysis of the catalytic behavior of an
individual DNAzyme, 614. This analysis revealed a number of surprising features of this
DNAzyme that may be generalized to other DNAzymes created through in vitro
Selection (IVS). Although selected with a protocol believed to enrich for Mg-dependent
self-cleaving catalysts, 614 cleaves both in cis and trans independently of Mg^+. The rate
of trans-cleavage at substrate saturation is 6-fold higher than the rate of c/s-cleavage.
Further, both cis- and trans-cleavage rates are enhanced at temperatures below 25C (the
temperature at which the selections were performed), suggesting that a balance is reached
between a slower chemical step (k^uni) or k^bi)) and a more stabilized association step
(ki(uni) or ki(bi)) at lower temperatures.
Mutagenesis of 614 and its substrates was performed to learn about the structure
of 614 cleaving in cis and trans. Compensatory mutations revealed that one of the
intermolecular helixes formed between 614 and its substrate is identical to an
intramolecular helix formed in cA-acting 614. The /rans-folded 614, however, has an
additional 4 base-pair helix not found in cA-folded 614, and this may partially explain the
faster rate of traws-cleavage at saturation.
We also addressed the potential loss of catalysts by premature catalysis during
the preparation of DNAzyme pools in the IVS. Purification of single-stranded
DNAzymes was accomplished using asymmetric PCR or exonuclease degradation,
x


47
plausible explanation for the failure to reach true completion? This possibility is ruled
out on the basis that ribose-614 also reaches a plateau that is far below true completion.
Although the maximal cleavage for caged-614 is consistently 10-15% lower than that of
ribose-614 (both when self-cleaved and when cleaved by base hydrolysis), both reach
cleavage plateaus that are as much as 35% below the amount of cleavage caused by
strong base. The explanation for the reduced cleavage plateau must be common for both
caged and ribose-614, and quite likely for any other DNAzyme generated via this IVS
procedure.
Incomplete Removal of the Complementary Strand
Two methods were used for removing the complementary strand generated during
the PCR synthesis of our DNAzymes, exonuclease degradation and asymmetric PCR.
Asymmetric PCR has nearly no potential for complementary strand impurity since the
complementary strand is 15 nucleotides longer than the catalytic strand, and thus easily
separated from the catalytic strand during gel purification. Experiments in which
ribose-614 was prepared using asymmetric PCR followed by gel-purification, showed
cleavage plateaus similar to ribose-614 generated with exonuclease, suggesting the
cleavage plateau cannot be attributed to inhibition by contamination of the
complementary strand.
When ribose-614 generated via the exonuclease method was treated with strong
base, similar amounts of conversion to the cleaved form were observed as seen for the
ribose primer ang + ribose (85-95% for ribose-614 vs. 90-97% for ang + ribose).
Since there is no possibility of a contaminating complementary strand in the ang +
ribose primer, the ~5% difference between base cleavage of the ang + ribose primer
47


75
structures into Families A, B, and C. The four structures of Family A each possess a
75GATA78 stem loop (the superscript numbers refer to the position in the overall
sequence), but vary in the fold of nucleotides 35 -70. Both structures in Family B share
the same fold between nucleotides 50 and 77, but vary in the fold adopted by nucleotides
30 through 50 and 77 through 84. Structure A1 and B1 share the same fold for
nucleotides 30 through 47. Both structures in the Family C share the same folding
pattern for nucleotides 35 -75.
Each of the three Families A, B and C can fold in such a way as to form base
pairing between the l7CGACTCACTAT27 (part of the 5-ang + ribose primer) and
85GTAGTGACG93 (part of the 3 primer). These 5 and 3 regions can also fold into
separate hairpins, preventing interaction with each other, as seen in structure C2. The 3
hairpin formed between nucleotides 78 and 101 disrupts the defining elements of both
Family A and B structures, but not the Family C structure (as seen in C2).
We have demonstrated that 614 can cleave ang + ribose primer (both in trans
and when incorporated as nucleotides 1-49 of 614). If we assume that the structure
formed when 614 cleaves in cis is similar to the structure formed when 614 cleaves ang
+ ribose primer in trans, then this suggests that the c/s-folded 614 has some base-pairing
between its first 49-nt and the remaining 3 end. The structure shown in C2 allows only
minimal interaction between the ang + ribose primer region and the remaining 3 half
(base pairing between 33CATGGC40 and 7IGCCTG75). However, the base-pairing
between nucleotides l7CGACTCACTAT"7 and 85GTAGTGACG93 seen in the structures
A1-4, Bl, B2 and Cl allows significant interaction between the 5 and 3 halves of 614.
75


Population Distribution After:
ROUND 3 ROUND 4 ROUND 5 ROUND 6 ROUND 7 ROUND 8 ROUND 9
Figure 4-11. Varying the selection time alters the range of rates surviving selection. Selections were simulated under varying selection times (St >15, 246, or 16922 minutes), while holding other
parameters of the model the same (parameters shown in figure). (B) Increasing the selection time from 15 minutes to 246 initially resulted in a larger fraction of the population being cleaved at the end of each round of
selection, corresponding to earlier initial detection, Increasing the selection time further to 16922 minutes reversed this trend. The fraction of the population cleaved at the end of each round reaches a plateau beginning after
round 4. indicating that the selections have reached stabilized populations When this cleavage plateau is below 95% (case 3. panel B), the stabilized population contains only the fastest catalysts (case 3, panel C). Because
there is no significant differential enrichment for catalysts with half-life (rate) 6-fold lower than the selection time, longer selection times result in cleavage profiles of -100% and enrich equally all molecules with a half-life (rate)
approximately 6-fold lower than the selection time (case 1 and 2. panel C) No catalysts with half-lives greater (lower rate, slower catalyst) than the selection time survive multiple rounds of selection Dashed vertical lines
represent rate corresponding to the three different selection times


ROUND 3 ROUND 4 ROUND 5 ROUND 6 ROUND 7 ROUND 8 ROUND 9
Progress Curve for New Population After:
three test cases varying in the rate of the fastest catalysts (kmaxn = 0.12, 0.012, or 0.0028 min-1) in the initial populations, while holding other parameters of the model constant (parameters shown in figure). The number of
catalysts with rates slower than the selection time was held constant at 1E-6 for all populations, while the distribution of catalysts with rates > St varied for each test case The catalysts for case 1 were distributed (for rates
between St and kmaxl) according to a linear function with slope of -1.2E-6. The catalysts for case 2 were distributed (for rates between St and kmax2) according to the proportional changes in catalysts in case 1 and holding
the reactive fraction the same (1E-6) All of the catalysts for case 3 were distributed into a single rate class with rate = kmax3 and reactive fraction equal to 1E-6. Because selections enrich for molecules with half lives 6-fold
less than selection time, if the half-life of the fastest catalyst is increased to within 6-fold of the selection time, only these fastest molecules survive.
On


146
selection time. This produces a stabilized population that includes catalysts with half
lives (rates) ranging from St/6 to kfastest. When the selection time is very short, only the
fastest molecules in the population have a high probability of surviving each round;
therefore, multiple rounds of selection eliminate all but the fastest molecules (rather than
isolating a population with a range of rates).
Examining the Impact of the Rate of the Fastest Catalysts
Another simulation was performed in which the half-life of the fastest catalyst in
the initial population was examined from 'life = 6 minutes to 60 and 246 minutes, while
holding other parameters of the model constant (parameters shown in Figure 4-12). The
number of catalysts with any given half-life greater than the selection time was held
constant at 10'6 for each test case, while the distribution of catalysts with half lives less
than the selection time varied. The population of catalysts in case 1 was distributed
according to a linear function with slope of-1.2 x 10'6 (for all catalysts with half lives
between St and kfastest!) The population of catalysts in case 2 was distributed according
to the proportionality established in case 1 (for all catalysts with half lives between St and
kfastest 1), while holding the reactive fraction constant (RF = 10'6). The catalysts for case 2
were distributed (for rates between St and kfastest^) according to the proportional changes
in catalysts in case 1 and holding the reactive fraction the same (1 O'6). All of the catalysts
for case 3 were distributed into a single rate class with rate = kfastest and reactive fraction
equal to 10'6.
Figure 4-12 shows that reducing the rate constant of the fastest catalyst in a
population (kfastest is lower, and therefore slower catalyst) has a similar effect as


135
catalysts were abundant in the initial library, catalysis could be detected above
background by electrophoretic size shift of the library itself, or by examining 100
individual molecules randomly selected from the library via cloning. Both methods,
however, could not yield a positive result if catalysis were less frequent than
approximately one part per hundred. In the first method, background would obscure less
abundant catalysts; in the second method, insufficient sampling would miss the catalysts.
Being that catalysts are much more rare than one per hundred, multiple rounds of
selection must be performed to enrich catalysts.
Most I VS experiments are designed to isolate the best catalysts in the library.
Isolating a few of the (presumably) best catalysts tells us very little about the likelihood
of generating function from a random library. Indeed, the question of greatest
importance is, How many catalysts were there in the initial pool, and what was the
distribution of their catalytic power (rate constants)? To do this we must extrapolate
from what we can observe following multiple rounds of selection.
Simulating In Vitro Selections
This section describes of a play forward model for I VS which can be used to
infer information about an initial library based on data gathered following multiple
rounds of selection. The model begins by estimating values for key parameters for an
initial random library, and then simulating multiple rounds of selection. This allows us to
determine the impact of various initial conditions on observable outcomes following
selection. The parameters of the model are on the following page.


127
Testing Kinetics from Round 8 Pools Without Caged-ribose
These data suggest that the laser may cause background cleavage that masks the
cleavage resulting from DNAzyme catalysis. This was tested by amplifying the survivors
of the eighth round of selection for each selection condition without a caged-ribose (using
either BJ + ribose or ang + ribose) and testing kinetics without the laser initiation.
Figure 4-2 shows the results. The pool of molecules generated from population H
demonstrates a dramatic increase in cleavage with increasing time. None of the other
selection conditions, however, show any detectable cleavage. This suggests that the
minimal cleavage seen in the kinetics profiles completed during the I VS is in fact
background caused by the combined presence of the caged-ribose and laser.
Testing for Inadvertent Selection for Susceptibility to Laser
In light of the slight increase in the amount of cleaved product observed with
increasing rounds of selection for all samples containing a caged ribose, it is conceivable
that the IVS with caged ribose and laser actually selected for molecules that are more
susceptible to cleavage by laser in the presence of a caged-ribose. To test this possibility,
pools from round 1, 4 and 8 for selection conditions D and H were re-amplified with
caged-ribose (using BJ + cage and ang + cage primers respectively). Each sample
was diluted to 15 nM and exposed to laser. Each of the six samples had less than 0.25%
cleavage prior to laser. Immediately following laser exposure, pools from H rounds 1,
4, and 8 showed 3.4, 3.5, and 3.0% cleavage; pools from D rounds 1, 4, 8 showed 5.6,
5.4, and 5.3% cleavage. The amount of each pool cleaved immediately following laser
did not increase from round 1 to 4 to 8, arguing against the possibility that there was


88
Figure 3-7. The cleavage plateau of caged-614 reaches a maximum with 150
pulses of laser. Samples of caged-614 were initiated with increasing amounts of laser
and allowed to self-cleave until the reaction reached plateau. Samples initiated with
150 pulses of laser and more reached equivalent cleavage plateaus of -50% (about
-10% lower than for an equivalent experiment with ribose-614.
Figure 3-8. Maximum deprotection of caged-614 occurs with 150 pulses of laser.
Samples of caged-614 were exposed to increasing amounts of laser and then incubated
with strong base for 1 hour at 80C to determine the amount of deprotection. Caged-
614 exposed to 150 pulses of laser is 75% cleaved by the strong base. Additional laser
does not increase the amount cleaved by strong base. Ribose-614 has slightly higher
cleavage under similar conditions.


60
In Figure 3-22A (the same data as Figure 3-21, but with a linear abscissa), the
intercept at [614] = 0 is not at the origin. This requires that at least some of the rate arise
via a unimolecular process. Figure 3-22B shows a best-fit line extrapolated to zero
(infinite dilution) for 614 concentrations below 3.5 nM (predominately unimolecular)
yields an intercept corresponding to the unimolecular rate constant 0.0059 hr 1 (half-life
is 115 hours). The slope of this line is 8.9xl05 Mhr'1, which for a second order process
when [S]Km, equals the second order rate constant (kcat/Km).
A plot of the natural log of uncleaved ribose-614 (substrate) versus time should be
linear for a first order process; a linear relationship fit the data well for low ribose-614,
but not for higher concentrations (Figure 3-23). The negative of the slope of the best-fit
line for 0.3 nM ribose-614 is 0.0042xl0'3 hr'1, which is in close agreement with the rate
estimated by extrapolation of the initial rates of ribose-614 to infinite dilution (Figure 3-
22A and B).
Further, it appears from Figure 3-22A that a plateau has not been reached. It may
be that the [614] is simply not high enough in this experiment. However, this plot does
not rule out the possibility that a purely bimolecular process (not involving saturation) is
contributing to the overall rate of the reaction. Although this rate is low by normal
ribozyme standards, and even the standard of some DNAzyme phosphodiesterases
isolated via IVS, this unimolecular rate is comparable to the magnesium-independent
catalyst isolated using IVS by Geyer (0.012 hour'1 for G3 Family DNAzyme before
optimization).
We then tested to see if the progress curve for each concentration of ribose-614 fit
best to one exponential (in which both the rate and plateau can vary) or the sum of two
60


134
in pool 8 is faster than that of the 614 sequence. This was explored by testing the rates of
c/s-cleavage of various clones isolated from the round 8 pool (Figure 4-9). Cleavage
assays were done at 12 nM, the same concentration used during the selections.
Clone Hr#8.161 differed from 614 by only one nucleotide, yet the rate constant
was 2-fold lower (0.0059 hr'1), whereas other clones (Hr#8.58, Hr#8.53, Hr#8.162)
differing from 614 and each other by only one or two nucleotides had rate constants ten
fold higher than 614 (-0.15 hr'1). Two Family gamma sequences (Hr#8.56 and
Hr#8.169) differed from each other by only three nucleotides, yet there rate constants
differed by an order of magnitude (0.12 vs. 0.017 hr'1). The clones from round 8 that did
not fall into one of the four major sequence Families all had very low rate constants
(Hr#8.170 was the fastest, with a rate of 0.0013 hr'1).
Simulations of In Vitro Selection Experiments
To understand the role that catalytic nucleic acids may have played in the origins
of life, we must be able to quantify the power and abundance of catalytic activity in a
random pool of nucleic acid polymers. Previous I VS experiments have supported the
expectation that catalytic nucleic acids (catalytic in regards to any specific reaction) are
rare. We are limited in our ability to directly estimate the frequency of rare catalysts
by limitations of experimental methods used to detect them. Detection of catalysis most
frequently relies on an electrophoretic size shift of a radio-labeled catalyst-substrate
conjugate. Alternatively, a sample of molecules could be isolated via cloning and
sequencing, and unique sequences could be tested individually for catalytic activity
(within an order of magnitude, 102 is largest practical sample size for this method). If


21
the cofactor. Correct folding is assumed to occur without the required cofactor, and
quickly relative to catalysis.
Each of these problems arises because the Breaker-Joyce protocol has the 2'-
hydroxyl group of the RNA molecule to be free during the set-up. To resolve these
problems, research in this disseration develops techniques to eliminate loss of catalysts
during set up by using a caged-ribose. Caged-ribose was synthesized with the 2'-hydroxyl
group protected via an ortho-nitrobenzyl protecting group. This protecting group can then
be removed by UV photolysis (Figure 1-2).
LASER
ww
Caged ribose
Caged is blocking the
2'-0H in ribose for
nucleophilic attack on
the phosphorus
2'-0H is free and able to
attack the phosphorus
atom with the help of a
catalyst
Figure 1-2. Exposing the caged-ribose converts 2-orthonitrobenzyl
protecting group to the standard hydroxyl group, thus generating a standard
RNA nucleotide.
21


86
hours
Figure 3-5. Caged-614 shows similar kinetics to ribose-614 when deprotected by
150 pulses of laser. For all kinetic analysis, error is estimated to be +/- 0.05 hour and
less than +/- 5 percent cleaved. Identical experiments conducted in parallel fell within
these error estimates (see Figure 3-30B as an example).


76
These latter structures therefore seem more probable for both ds-cleavage and trans-
cleavage.
This was tested directly by mutating nucleotides T86 to A, G88 to C, and G90 to
C in 614 (Figure 3-44). These introduced mutations greatly diminished the ability of 614
to cleave itself, as well as the substrate r-lib61o2 in trans (compared to wild-type 614).
Introducing the compensatory changes (C22G, C24G, A26T) predicted by the structures
of A1-4, Bl, B2 and Cl into the substrate r-lib61o2 significantly improved trans cleavage.
Introducing the compensatory changes C22G, C24G, A26T into the ribose-614
containing T86A, G88C, and G90C also dramatically improved catalysis. This
demonstrates that the helix formed between l7CGACTCACTAT27 and 8:>GTAGTGACG93
shown in the structures Al-4, Bl, B2 and Cl is probaby formed in cis and trans.
Attempts were made to eliminate some of the remaining candidate structures.
Although structure Cl appears quite linear and does not appear to position a folded
catalytic motif near the ribo-adenosine, it does possess the requisite base-pairing
between the l7CGACTCACTAT27 and 85GTAGTGACG93 regions. Because the 5 49-
nucleotides of 614 correspond to the ang + ribose substrate, the essential elements of
structure Cl could be formed by a 614/ang + ribose enzyme/substrate conjugate if the
ang + ribose primer displaces the 5 half of 614. If structure Cl were indeed the
correct structure, the 3 half of 614 (nucleotides 56 through 94) should be sufficient to
bind and cleave the ang + ribose substrate. This was tested by generating the 3 half of
614 (both 614 56-94nt and 614 56-106nt). Truncating 614 at position 94 eliminates
the possibility of forming structure C2, while still permitting structure Cl. When mixed
76


90
300 nM r614, 1mm MgCi2
100 nM r614, 1mm MgCI2
33nMr614, 1mm MgCI2
11.1 nM r614, 1mmMgCl2
300 nM r614, 3mm MgCI2
100 nM r614, 3mm MgCI2
33 nM r614, 3mm MgCI2
11.1 nM r614, 3mm MgCI2
Figure 3-10. Ribose-614 cleavage is not altered by 3-fold variation in MgCh
concentration. Various concentrations of r-614 were incubated in 50 mM HEPES
pH 7, 1 M NaCl, and either 1 mM or 3 mM MgCh. Increasing the MgCE
concentration three-fold had no significant effect on r-614 cleavage rate at any
concentration.
Figure 3-11. Ribose-614 is inactive without NaCl. Ribose-614 was incubated
with 50 mM HEPES pH 7, plus either 1M NaCl, 1 mM MgC12, or 50 mM
cacodylate.


92
c-614 (200 nM)
c-614 (200 nM)
+ unlabeled d-614
(400 nM) at 143.5 hrs
c-614 (200 nM) diluted
to 8 nM at 143.5 hours
hours
Figure 3-13. 614 cleavage plateau is not a result of an approach to chemical
equilibrium. 614 was allowed to react until it approach cleavage plateau, at which
point it was either diluted 25-fold, or incubated with 2-fold additional unlabeled d-
614. Neither condition altered the cleavage plateau of 614 in a time-dependent
manner.
r-614
r-614 + 27-nt product
r-614 + 79-nt product
r-614 + 27- & 79-nt product
r-library + 27-nt product
r-library + 79-nt product
r-library+27- & 79-nt product
Figure 3-14. Addition of the 27-nucleotide and the 79-nucleotide cleavage
products alone or together do not alter the cleavage profile of 614 (400 nM each).
Addition of the 79-nucleotide cleavage product to the r-library substrate resulted
increased cleavage, but at an extremely slow rate. This cleavage of the r-library
substrate in the presence of the 79-nt fragment is lowered by the addition of the 27-nt
cleavage product.


143
faster catalyst in order for both catalysts to be in equal abundance following one round of
selection. If three round of selection are required, the initial abundance of the slower
catalysts needs to be (7.5)3 = 416 times more abundant than the faster catalyst. The
situation is even more dramatic for very slow catalysts: if a very slow catalysts is 100
times slower ('AlifeverySioW = 100*5/) than the fast catalyst (Alife/ast = St), then this very
slow catalyst must be in 72-fold greater abundance in the initial pool to survive in equal
proportion as the faster catalyst following one round of selection, and 380,000-fold (=
72)3 greater abundance to survive three rounds of selection. Because I VS routinely
require 5 to 15 rounds of selection, molecules with half lives greater than St are likely to
be completely absent from the population by the time any cleavage above background is
detected. For the simulation shown in Figure 4-10, all molecules in zone A are
eliminated by round 3.
Zone B corresponds to an area of active enrichment where molecules with faster
rates are enhanced disproportionably greater than molecules with slower rates. With
sufficient rounds of selection, the slower molecules fall below the level of detection (and
hence zone B becomes more skewed towards faster catalysts and zone A enlarges to
include an area that was formerly part of zone B).
Zone C corresponds to the situation describe above in which all the molecules are
sufficiently fast relative to St (Alife < St/6) such that nearly all molecules are cleaved by
the completion of each round of selection and no selective enhancement exists.
Consequently, molecules in zone C continue to persist without any change in their
relative proportion to each other. These properties of the population distribution can be
seen in Figure 4-10, or in more detail in Table 4-1.


84
*
126 nt >
116 nt >
106 nt >
lane
2 3
4 5 6 7 8 9
10 11 12 13 14
106 nucleotide template extended with:
A 5'-end labelled complimentary strand primer: "short" + linker + tail
* 5-end labelled catalytic strand primer: "ang + ribose"
Figure 3-2. Asymmetric PCR using various linkers increases the size of the
complementary strand without increasing the size of the catalytic strand. PCR
reactions where done in pairs with the catalytic strand primer end labeled in the odd-
numbered lanes, while the complementary strand primer was end-labeled in the even-
numbered lanes.
+ - + + tail on complimentary strand primer
+ + + - caged ribose on catalytic strand primer
+ exonuclease
+
<
<
<
<
NaOH
121 nt:
full length + 15 nt tail
106 nt:
full length product
92 nt:
full length product
+ 15 nt tail 29 nt 5' of cage
79 nt:
cleaved product
lane # 1 2 3 4 5
Figure 3-3. Size difference between catalytic and complementary strands
generated by asymmetric PCR allow for gel-purification of the catalytic strand.
The presence of the cage-ribose in the template causes truncation of the
complementary strand. Asymmetric PCR therefore generates a complementary strand
that is either longer or shorter than the catalytic strand.


eliminating the need for column-based methods for performing IVS. We also examined
the utility of a protected ribose (caged-ribose) for both kinetic analysis and an IVS.
Selections with caged-ribose failed to enrich for catalysts, whereas a selection with
unprotected-ribose produce a number of highly active catalysts. Simulations revealed
that the slight increase in background cleavage caused by laser photolysis explains the
failure of the caged-ribose to enrich for catalysts.
Analysis of the catalysts isolated from this in vitro selection differed from
sequences isolated in previous selections by only a few nucleotides, but their catalytic
rates varied by over an order of magnitude. This variation suggests that the sequence
space near any particular active catalyst is well-populated with other active catalysts.
Simulations of in vitro selections examined the impact of various parameters on
the outcome of selections. These simulations demonstrated that when the selection time
is similar to the half life of the fastest catalysts in the initial library, only the fastest
molecules are isolated. Increasing selection time such that it is significantly greater than
the half life of the fastest catalysts in the initial population creates a stabilized population.
The proportionality between the fastest and slower catalysts of this stabilized population
remains similar to the initial population, thus allowing direct estimation of the initial
population distribution. The model developed in this simulation can also be used to
estimate the distribution of catalysts in the initial library for future in vitro selections.
xi


93
r-614 (222.2 nM)
r-614 (100 nM):
uncleaved r-614
gel purified
at 140 hours
Figure 3-15. Gel-purification of 614 at cleavage plateau results in additional
cleaveage, indicating a fraction of 614 is folded into an inactive conformation.
Ribose-614 was allowed to self-cleave for 140 hours, nearing cleavage plateau, at
which point half the sample was gel-purified on a denaturing gel. After resuspending
the gel-purified ribose-614 (gp614) in reaction buffer to 100 nM, additional
cleavage was detected.
Figure 3-16. Reheating ribose-614 results in additional cleavage, indicating a
fraction of 614 is folded into an inactive conformation. Ribose-614 was allowed to
react until it reached cleavage plateau, at which point half of the sample was denatured
by heating to 97C for 3 minutes, and slowly cooled to 23C over 10 minutes. This re
folding procedure resulted in an additional 10% cleavage.


102
r-614 (100nM), no r-primer
cleavage of r-614 (100 nM)
in the presence of
r-primer (100 nM)
cleavage of r-primer (100 nM)
in the presence of
r-614 (100 nM)
Figure 3-25. Ribose-primer competes with r-614 for cleavage by r-614. The
cleavage of r-614 is lowered when incubated with r-primer.
r-primer (7.5 nM)
+ d-614 (270 nM)
r-lib6lo2 (7.5 nM)
+ d-614 (270 nM)
r-library (7.5) nM
+ d-614 (270 nM)
r-4up6 (7.5) nM
+ d-614 (270 nM)
Figure 3-26. d-614 can cleave various substrates in trans. Unlabeled d-614 (270
nM) was incubated with various radio-labeled substrates (7.5 nM), resulting in
substrate cleavage. Each substrate contained the same 47-nucleotide r-primer region
with a ribose-adenosine at position 27.


180
Breaker, R.R., Joyce, G.F. (1995) A DNA enzyme with Mg-dependent RNA
phosphodiesterase activity. Chem. Biol. 2, 655-660
Cech, T.R., Zaug, A.J., Grabowski, P.J. (1981) In vitro splicing of the ribosomal RNA
precursor of Tetrahymena. Involvement of a guanosine nucleotide in the excision of the
intervening sequence.Cell 27, 487-496.
Crick, FHC. (1968) The origin of the genetic code. J Molec Biol, 38, 367-79.
Fuller WD, Sanchez RA, and Orgel LE. (1972) J Molec Biol, 67, 25-33.
Geyer CR, Sen D (1997). Evidence for the metal-cofactor independence of an RNA
phosphodiester-cleaving DNA enzyme, Chemistry and Biology 4, 579-593.
Gabel NW, Ponnamperuma C, (1967) Nature, 216, 453-455.
Guerrier-Takada, C., Gardiner, K., Marsh, T., Pace, N., Altman, S. (1983) The RNA
moiety of RNase P is the catalytic subunit of the enzyme.Cell 35, 849-857.
Gilbert, W. (1986) The RNA World. Nature, 319, 618.
Hampel A, Cowan JA. (1997) A unique mechanism for RNA catalysis: the role of metal
cofactors in hairpin ribozyme cleavage. Chem Biol, 4, 513-517.
Hanel R, Conrath B, Flasar FM, Kunde V, Maguir W, Pearl J, Pirraglia J, Samuelson R,
Herath L, Allison M, Cruikshank D, Gautier D, Gierasch P, Horn L, Koppany R, and
Ponnamperuma C. (1981) Infrared Observations of the Saturnian System from Voyager
1, Science 212, 192-200
Huebner, WF, Synder LE, Buhl D. (1974) HCN Radio Emission from Copmet Kohoutek
(1973), Icarus 23, 580-84.
Irvine, D., Tuerk, C., Gold, L. (1991) Selexion. Systematic evolution of ligands by
exponential enrichment with integrated optimization by non-linear analysis. J. Mol. Biol.
222, 739-761.
Irvine, WM. (1999) The Composition of Interstellar Molecular Clouds, Space Sci. Rev.
90, 203-218.
Johnsson, K., Allemann, R. K., Benner, S. A. (1990) Designed enzymes. New peptides
that fold in aqueous solution and catalyze reactions, in Molecular Mechanisms in
Bioorganic Processes, C. Bleasdale and B. T. Golding, eds. Cambridge, Royal Society of
Chemistry, 166-187
Johnsson, K., Allemann, R.K., Widmer, H., Benner, S.A. (1993) Synthesis, structure and
activity of artificial, rationally designed catalytic polypeptides. Nature 365, 530-532.


183
Switzer, C.Y., Moroney, S.E., Benner, S.A. (1989) Enzymatic incorporation of a new
base pair into DNA and RNA. J. Am. Chem. Soc. Ill, 8322-8323.
Szostak, J.W. (1992) In vitro genetics. Trends Biochem. Sci. 17, 89-93.
Tarasow, T, Tarasow, S, and Eaton B. (1997) Nature 389, 54-57.
Taylor S, Walter K, Kast P, and Hilvert D. (2001) PNAS 98, 10596-10601.
Usher, D. A., McHale, A. H. (1976) Hydrolytic stability of helical RNA: A selective
advantage for the natural 3',5'-bond. Proc. Natl. Acad. Sci.USA 73, 1149-1153.
Verlander MS, Lohrman R and Orgel LE. (1973) J Molec Biol, 2, 303-319.
Walter MR. (1983) Earth's Earliest Biosphere, Ed. by JW Schopf, Princeton University
Press, 187-213.
Woese, C. (1967) The Genetic Code, Ch. 7. Harper and Row, New York.
Young KJ, Gill F, Grasby JA. (1997) Metal ions play a passive role in the hairpin
ribozyme catalysed reaction. Nucleic Acids Research, 25, 3760-3766.
Zepik, H., Benner, S. A. (1999) Catalysts, anticatalysts, and receptors for unactivated
phosphate diesters in water. J. Org. Chem. 64, 8080-8083.


11
It is interesting to note that not all (potentially) prebiotic monomers suffer as
many of the stereoisomeric challenges as ribose-based nucleotides. Both threose nucleic
acid (TNA) and peptide nucleic acid (PNA) use alternative bases attached to nucleic
acids. Both monomers are more chemically simplistic, thus avoiding much of the
stereoisomeric clutter of RNA. This has led to proposals of a pre-RNA world utilizing
either polymer as potentially life-supporting polymers. Theoretical and experimental
evidence suggest that information stored in the form of TNA or PNA polymers can be
transferred to RNA, although it is unclear how much phenotypic function is preserved in
the transfer (reviewed in Joyce, 2002).
The Next Challenge: Obtaining Catalytic Function from Random Libraries
Although monomer and polymer synthesis is expected to be difficult in a prebiotic
world, it is the formation of polymers that act as genetic molecules with useful chemical
properties that appears most formidable (useful from the perspective that it leads to the
creation of life: an initially useful property would be the ability to enrich for specific
monomers). This requires that polymers be of sufficient length to contain sufficient
information, be sufficiently stable as to persist, be able to mutate yet still be replicated,
and lastly, perform useful chemical function.
Although we do not know precisely what monomers were available in a priobiotic
world or how they were synthesized and polymerized, it is clear that at some point ~ 4
bya some randomly generated prebiotic polymers must have possessed useful
properties (presumably a self-replicase) that permitted the jump from a collection of
random polymers to a single-biopolymer life-form with a complex metabolism.
11


109
A
400
300
O
a
c
o
c
o
o
200-
100-
K
100
B
65-i
60-
55-
50-
§ 45-
40-
&
35-
£ 30-
£ 25-
a 20-
15-
10-
0+
-i
10
I
20
r
30
*
200 300
hours
A

I 1 1 1-
40 50 60 70
hours
400 nM r-primer +
A 100 nM d-614, mixed
before slow cool
a 400 nM r-primer +
100 nM d-614, mixed
after slow cool
133 nM r-primer +
33 nM d-614, mixed
before slow cool
133 nM r-primer+
33 nM d-614, mixed
after slow cool
I l
400 500
400 nM r-primer +
100 nM d-614, mixed
before slow cool
a 400 nM r-primer +
100 nM d-614, mixed
after slow cool
133 nM r-primer +
33 nM d-614, mixed
before slow cool
133 nM r-primer +
33 nM d-614, mixed
after slow cool
80 90 100
Figure 3-34. Multi-turnover kinetics for d-614 cleaving r-primer is not affected
by the slow cool folding protocol, and indicates two turnovers in 100 hours. (A)
Full time course for r-primer cleavage by d-614. (B) Initial time course for r-primer
cleavage by d-614 reveals linear cleavage up to 50 hours. The enzyme and substrate
were mixed together either immediately before or immediately after the slow cool
protocol.


121
45-i
40-
35-
O
§ 30-
co
0)
o 25-
S 20-
i. 15-
10-

5-
0-f
0
10



20 30
hours

* r-lib6lo2 wt cleaved
by d-614 wt
a r-lib6lo2 C35T cleaved
by d-614 G38A
r-lib6lo2 wt cleaved
by d-614 G38A
40 50
Figure 3-47. The G38A mutation of d-614 is partially rescued by the
compensatory C35T mutation in the r-lib6lo2 substrate, substantiating the
predicted inter-molecular CATG association. Converting G38 of d-614 to A
greatly reduced /ram-cleavage of r-lib61o2, but tram-cleavage was partially recovered
by adding the compensatory C35T mutation to r-lib61o2. Concentrations used were d-
614 (100 nM), and r-lib61o2 (1.5 nM).


128
selection for molecules that are more susceptible to laser induced cleavage. When the
kinetics of these samples was examined with increasing time, cleavage of H round 1
did not increase significantly, whereas H round 4 and H round 8 showed increasing
cleavage similar to samples prepared with ang + ribose and without laser initiation
(Figure 4-3). Pools from D rounds 1, 4, and 8 did not show significant increase in
cleavage with time after the immediate increase after laser exposure.
Caged-ribose Does Not Induce Cleavage in trans
The results described in the previous sections suggest that, although the laser
alone does not cause significant cleavage of ribose-614, it appears as though the presence
of a caged-ribose in addition to laser causes a low level of sequence-independent
cleavage at the caged-ribose (no mechanism for this reaction is known). It remains
unclear, however, if the combination of laser and caged-ribose is causing strand breakage
of the strand to which the caged-ribose is attached (in cis), or whether cleavage is in
trans. We tested this possibility by mixing unlabeled caged-ribose primer with
internally-labeled ribose-lib61o2, and then exposing the combined sample to laser. The
immediate cleavage seen previously when the caged-ribose was internal to the labeled
strand was not detected when the caged-ribose was in trans to the labeled ribose-
substrate, demonstrating that the cleavage effect of caged-ribose and laser is localized in
cis to its own phosphodiester backbone.
Kinetic Analysis of Rounds 1-8 Pools for H: ang + ribose
Pools for H were re-amplified from the stock of survivors following each
round of selection and used for a more detailed kinetic analysis. Samples were


152
particularly disheartening to the person who individually tested the rates of 100 clones
(particularly if the majority of molecules sample have rates greater than St/6, requiring
that more individual catalysts be sampled).
An alternative technique for estimating the initial population distribution would
be useful. Rather than working backward from the stabilized population, the
simulation model developed here can be played forward with varying starting
parameters to optimize fit to the experimentally observed data (most notably, the round at
which cleavage is first detected, the percent of the population cleaved at selection
plateau, and the rate at which the selection approaches selection plateau). Only two
parameters need to be varied, the distribution and size of the reactive fraction (leakage
and kfastes, can be determined experimentally and therefore do not need to be estimated).
This method offers the advantage that it considers the impact of catalysts with half lives
greater than St/6, and uses data from every round of selection (rather than just the
stabilized population of a single round).
Summary of Results
Table 4-1 (next page) summarizes the major results of the experiments and
simulations described in this chapter.


173
nucleic acids had to rely on sequence tags to distinguish molecules within the hypercycle
from selfish (competitor) molecules outside the hypercycle. Without overt selection, 614
possessed qualities that allow for this type of distinct regulation, acting differentially in
cis and trans (even for closely related molecules) at various concentrations. These
complex patterns of regulation are possible because of the generality of base pairing, and
thus appears to be intrinsic and abundant to nucleic acids (and therefore do not need to be
engineered). The complex RNA-based regulatory mechanisms now appreciated within
modem organisms may be analogous to the complex regulation achieved by the single
biopolymer progenitor.
Caged-ribose is Useful for Kinetics. But Increases Leakage Too Much for IVS
There are several places in the IVS protocol where catalysts may be lost prior to
the selection step. We examined the utility of a caged-ribose to guard against this. The
caged-ribose proved useful in protecting the sample from premature catalysis, yet when
deprotected via photolysis, produced kinetic results similar to those generated with
unprotected ribose. Unfortunately, the laser photolysis generates a higher amount of
background cleavage (leakage) which resulted in the failure to enrich for active catalysts
when used during an IVS. Simulations of IVS revealed that even slight increases in this
sequence-independent cleavage parameter (analogous to other forms of leakage
inherent in the column-selection protocol) significantly delay enrichment of catalysts.
We have also developed alternatives to the standard Breaker-Joyce method for
isolating catalytic strand DNAzymes and for performing solid-phase selections.
Asymmetric PCR and exonuclease degradation were effective methods for purifying the


10
supported by the discovery of similar monomers on meteorites, which were also
presumably synthesized by prebiotic mechanisms in space. The next step towards a
living system requires that monomers be linked together by prebiotic chemistry to form
polymers.
Limited polymerization of nucleic acids has been observed under prebiotic
conditions. Adenosine 2-, 3-cyclic phosphate can be converted to oligomers by heating
to dryness, or by incubating at room temperature in the presence of a polyamine
(Verlander, 1973). Nucleoside 5-polyphosphates have also been produced by heating
nucleotides and inorganic polyphosphates to dryness (Lohrmann, 1975). The random
polymerization of monomers, however, remains a great challenge, specifically in light of
the abundance of stereo-chemically related compounds that are expected to interfere with
prebiotic polymerization. Taking ribose-based nucleotides as an example, nucleotide
polymerization can result in a combinatorial mixture of 2-, 5-, 3-, 5-, and 5-, 5-
phosphodiester bonds, with variable numbers of phosphates between the sugar bases,
combined with the D- and L- sugar stereoisomers and a- and P-anomers of the glycosidic
bonds. Most polymers generated are expected to have heterogeneous linkages, which is
expected to make replication of the polymer exceedingly more difficult. Several
mechanisms may have existed to overcome stereoisomeric interference, generally relying
on some mechanism of enriching for specific isomeric forms of monomers or preferential
incorporation of certain monomeric isoforms. Although prebiotic formation of both
monomers and polymers is likely to result in a clutter of stereoisomers, a series of biased
synthesis followed by fractionations or enrichments may have provided sufficient
building blocks for creating life.
10


103
d-614
d-4up6
d-lib6lo2
d-primer
Figure 3-27. Various substrates can compete with r-614 for cleavage by d-614.
Addition of a 9-fold excess of unlabeled d-lib61o2 or d-primer to radio-labeled r-614
(33 nM) at time zero lowered cleavage of r-614. Addition of the same amount of
unlabelled d-614 to labeled r-614 increased r-614 cleavage to levels similar to that
seen with 300 nM r-614. Addition of d-4up6 increased r-614 cleavage, but not as
much as the addition of d-614.
0.3 nM r-614 + 3 nM d-614
0 3 r-614 + 3 nM d-lib6b2
0.33 nM r-614
3.3 nM r-614
260 nM r-614
Figure 3-28. The cis-cleavage rate of r-614 is not affected by excess competitors.
Radio-labeled r-614 was incubated at a low concentration (0.33 nM) so as to minimize
trans-cleavage. Addition of 9-fold excess d-lib61o2 competitor did not affect the
cleavage rate of r-614.


SEARCHING MOLECULAR LANDSCAPES FOR THE EVOLUTION OF PRIMAL
CATALYSTS: IN VITRO SELECTION OF DNA-BASED RIBONUCLEASES
By
MATTHEW A. CARRIGAN
DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE
UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE
REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY
UNIVERSITY OF FLORIDA
2002


23
assumed that the identity of the constant regions (arbitrarily chosen by researchers)
surrounding the randomized region is largely irrelevant to the catalytic activity eventually
selected, and it is instead only the catalytic potential of the random region that is being
tested. The DNAzyme 614 was selected with 2 nucleotide changes to the standard
Breaker-Joyce constant region; detailed analysis of 614 revealed that 614 can act in both a
uni-molecular and bi-molecular fashion. This may be a direct consequence of the two
nucleotide changes from the Breaker-Joyce motif (although this is not known for certain
since few researchers have directly tested whether the isolated DNAzymes are in fact
acting only in a uni-molecular fashion). This has given us insight into how to design and
simulate future I VS.
We have also isolated several catalysts, some of which are close relatives of 614,
615 and 616 isolated in a previous selection, which each display catalytic activity. This is
significant in two aspects, first in that the catalytic activity varies significantly with only a
minimal sequence changes. This suggests that the landscape in the near vicinity (one or
two nucleotides) of a particular active molecule contains many catalysts, presenting
opportunities for continuous exploration of the functional landscape. Second, the
abundance of near-relatives that retain activity suggests that most estimates of the catalytic
frequency may be gross underestimations, as near-relatives may appear identical under the
methods (restriction digest) used to distinguish them.
23


181
Johnston, W.K., Unrau, P.J., Lawrence, M.S., Glasner, M.E., Bartel, D.P. (2001) RNA-
catalyzed RNA polymerization: Accurate and general RNA-templated primer extension.
Science 292, 1319-1325.
Joyce, GF. (1989) RNA evolution and the origins of life. Nature 338, 217-224.
Joyce, GF. (1994) Forward in: Origins of Life, The Central Concepts, ed. Deamer DW
and Fleischaker GR. Boxton: Jones and Bartlett.
Joyce, GF. (2002) The antiquity of RNA-based evolution. Nature 418, 214-221.
Kauffman, S. (1993) The origins of order : self organization and selection in evolution.
New York : Oxford University Press.
Keefe, A.D., Szostak, J.W. (2001) Functional proteins from a random-sequence library.
Nature 410, 715-718.
Latham, J.A., Johnson, R., Toole, J.J. (1994) The Application Of A modified nucleotide
in aptamer selection Novel thrombin aptamers containing 5-(l-pentynyl)-2'-
deoxyuridine. Nucleic Acids Res. 22, 2817-2822.
Lohrman RJ. (1975) JMolec Biol, 6, 237-252.
Matsuura T, Yamaguchi M, Ko-Mitamura EP, Shima Y, Urabe I, Yomo T. (2002)
Importance of compartment formation for a self-encoding system. PNAS 99, 7514-7517
Miller SL and Urey HC. (1959) Science 130, 245-51.
Miyakawa S, Cleaves, HJ, Miller SL. (2002) The cold origins of life: B. Implications
based on pyrimidines and purines produced from frozen ammonium cyanide solutions.
Origins of Life and Evolution of the Biosphere 32, 209-218.
Nakano S, Chadalavada DM, Bevilacqua PC. (2000) General acid-base catalysis in the
mechanism of a hepatitis delta virus ribozyme. Science, 287, 1493-1497.
Orgel, LE. (1968) Evolution of the genetic apparatus. J Molec Biol, 38, 381-93.
Ozkan SB, Bahar I, Dill KA (2001) Transition states and the meaning of Phi-values in
protein folding kinetics. NATURE STRUCTURAL BIOLOGY 8,765-769.
Perrin, D.M., Garestier, T., Helene, C. (2001) Bridging the gap between proteins and
nucleic acids: A metal-independent RNAseA mimic with two protein-like functionalities.
J. Amer Chem Soc. 123, 1556-1563.


4
spontaneously and inevitably as long as minimal resources are available and a major
global environmental change does not bring about extinction (Kaufman, 1993).
Two specific functional attributes follow from this definition of life, the ability to
carry heritable information (genetics) which is linked to selectable traits (function).
Selection is thus acting on two levels simultaneously, the system of storing information
and the physical properties derivative of this information. Both information-storage
and function can be accomplished using a myriad of chemical and physical formats.
However, the requirements of information storage and duplication using chemical
polymers are in many ways contradictory to the requirements for replication
enhancing functions (Benner and Switzer, 1999):
Changes in the information must not impede its ability to be duplicated, whereas
changes in information must allow changes in function, and thus, physical
properties. The selectable traits (function) of a polymer are determined by the
three dimensional structure (fold) of the polymer and the positioning of chemical
moieties, thus a functional polymer must fold to generate selectable chemical
traits. In contrast, an information storage system should not fold easily as this is
expected to impede generalized mechanisms for duplication.
An information storage system should have few subunits to ensure greater fidelity
of information transfer, whereas a functional polymer should have many subunits
with diverse functionalities to enhance the chemical potential of the polymer.
For a system to explore new function through evolution, chemical properties must
be able to change rapidly (with few changes) and abruptly. Such abrupt changes
in the chemical properties of a molecule following mutation are contradictory to
the requirement that the information carrying system maintain its ability to be
replicated in spite of significant change (mutation).
Chemical Challenges to the Prebiotic Synthesis of Life
For a chemically based living system (which is the only undisputed form of life
we know), it is difficult to imagine a mechanism for storing information and generating
phenotypes other than through the structure of polymers. Given this restriction, a number
4


147
increasing the selection time on the population distribution following selection. Because
selections enrich for molecules with half lives (rates) 6-fold less than selection time,
when the kfastest is reduced to within 6-fold of the selection time, only molecules with rate
= kfastest survive. This indicates that given sufficient rounds of selection as to attain a
stabilized population, the distribution of catalysts in the stabilized population is
determined in part by a combination of factors including both the fastest catalysts in the
initial library and the selection time.
Examining the Impact of the Reactive Fraction
Simulations were performed for initial populations that varied in either the
reactive fraction, the slope of the distribution of catalysts, or both Figure 4-13). The
catalysts in case 1 and case 2 were both distributed according to the same linear function
(with m = -4.5 x 10'9), but case 2 had 100 times more reactive molecules in its initial
population (RF = 10'6 rather than 108). Not only did the selections for case 2 reveal
catalysts at an earlier round of selection, but the distribution of catalysts in the stabilized
populations differed from the distribution for case 1.
Both stabilized populations for case 1 and 2 have a decreasing trend (less catalytic
abundance with increasing catalytic rate) as in the initial library, but case 1 decreased
more rapidly than case 2. Reasoning that it was not the slope of the initial distribution
function, but rather the ratio between faster and slower molecules established by the
initial distribution function and the reactive fraction, that determined the distribution of
catalysts in the stabilized population, a third case was tested. This third case differed
from case 1 in both the number of reactive catalysts (RF = 10'6vs 108) and in the


6
functional requirements are held simultaneously in a single molecule (the challenge of
compartition remains when the system under selection contains multiple gene/enzymes).
This solution, however, creates another challenge: Such a biopolymer must meet the
contradictory challenges of (c)-catalysis and (d)-genetics simultaneously.
Progress Towards Understanding the Genesis of Life: Dual Function Biopolymers
Two approaches have since been pursued to gain insight about how to meet the
challenges to the origins of life, one which begins with modem life-forms and works
backwards to deduce information about its origins, while the other approach begins with
prebiotic chemistry and attempts to work forwards towards life.
Much insight has been deduced about early life from an analysis of its modem
descendents. Although modem life-forms have nucleic acids specialized for genetics and
proteins specialized for catalysis, there exist many idiosyncrasies about modem
metabolism that are suggestive of a previous incarnation as a single-biopolymer life-form
based on RNA: most biological coenzymes are based on nucleotides, nucleic acids are
used to the generate proteins; histidine is biosynthesized from nucleotides
(phosphoribosyl pyrophosphate and ATP); RNA is a key component of the spliceosome,
ribosome, and RNAseP; RNA serves as the primer for DNA synthesis; DNA is
synthesized by protein-based reduction of RNA (a chemical transformation that is quite
difficult, and unlikely to occur via prebiotic mechanisms) (Joyce, 1989; Benner, 1989).
The use of RNA for chemical purposes for which it is not intrinsically suited, particularly
in the modem environment where chemically better-suited molecules could be made
(proteins), argues against convergent evolution and instead suggests that these RNA
6


5
of requirements must be met for the prebiotic (pre-Darwinian) genesis of a living
(evolving) system:
(a) Prebiotic synthesis of the monomers.
(b) Prebiotic formation of polymers from monomers.
(c) Obtaining catalysts from a library of random polymers.
(d) Genetics obtaining a polymer that acts as a heritable carrier of information for
the useful catalysts.
(e) Compartition linking the information storing system (genetics) to the catalytic
function(s) it encodes.
In modem biology, the contradiction between the requirements for a genetic
biopolymer and a catalytic biopolymer (see above) is ameliorated by (mostly) separating
the functions of genetics and catalysis into two different bio-polymers. Nucleic acid is
the information storing biopolymer, which in turn is used to produce proteins which
generate most of the physical properties that are the subject of natural selection.
Although the division of labor between two distinct bio-polymers resolves the
contradictory demands of a life-form, it creates a significant paradox pertaining to the
origins of life
In light of the challenges (a) and (b) to the prebiotic creation of life, it is difficult
to imagine a non-biological mechanism for generating either of the contemporary bio
polymer systems (nucleic acid or protein). It is astronomically more improbable that both
bio-polymer systems would arise simultaneously and spontaneously, and furthermore in a
manner that would allow their mutual integration/compatibility. In the 1960s, theoretical
consideration of this dilemma led to the hypothesis that life must have originated as a
single biopolymer systems (Rich, 1962). This proposal requires only one successful
solution to the challenge of obtaining useful monomers and polymers. Furthermore, for a
single gene/enzyme system, the challenge of compartition is resolved because the two
5


62
if the reaction where simply the sum of a unimolecular and bimolecular reaction, the
progress curve should fit to the sum of two power functions (Y = AX + BX where X is
the concentration of 614, Y is the initial rate (kobs), A is the first order rate constant and B
is the second order rate constant). This model, however, was unable to provide a
reasonable fit for the experimental results, suggesting that ribose-614 cleavage is not
simply sum of a unimolecular and a bimolecular reaction.
Deoxyribose-614 Cleaves Various Ribose-Substrates
The observation that ribose-614 can cleave in a non-uni-molecular fashion
suggests trans cleavage. This was tested directly by incubating 614 with various
substrates, each containing the 5 catalytic motif and ribose-adenosine. Initially, 100 nM
ribose-614 was incubated with 100 nM of 5-32P-labelled ang + ribose primer. Figure
3-24 shows that both ribose-614 and ang + ribose are cleaved, clearly establishing trans
cleavage ability of 614. The cleavage of ribose-614 in the presence of ang + ribose
primer was lower than ribose-614 incubated alone, suggesting that the ang + ribose
primer competes with ribose-614 for cleavage by ribose-614 (Figure 3-25).
We then tested whether a 614 DNAzyme synthesized with a deoxy-adenosine
instead of a ribose-adenosine (referred to as d-614), and therefore unable to cleave itself,
could cleave various substrates in trans. We again tested trans cleavage of the ang +
ribose primer (Figure 3-24), as well as a mutant of 614 (known as 4up6, which has
greatly reduced catalytic activity), a pool of molecules synthesized from the random
library, and a individual clone from the random library (known as lib61o2, which has
no detectable catalytic activity). Figure 3-26 shows that each substrate was cleaved by d-
62


67
turnover conditions (compare the dramatic increase in multiple turnover cleavage for r-
lib61o2 cleavage when d-614 concentration is increased from 33 to 100 nM with the
modest increase for ang + ribose primer cleavage, indicating that the ang + primer
cleavage is near its maximum rate of turnover). This may be attributable to the better
ability of the ang + ribose substrate to bind with or disassociation from the enzyme, or
its lower proclivity for forming nonproductive interactions.
Multiple turnover conditions were also tested by holding d-614 concentration
constant at 20 nM and varying ang + ribose concentration from 100 to 2000 nM
(Figure 3-35). Approximately two turnovers were observed in the first 100 hours, while
only 5-6 turnovers were observed by 400 hours.
If the rate limiting step is disassociation of the substrate from the enzyme, we
would expect to see a burst phase in the initial phase of multiple turnover kinetics in
which all of the initially bound substrate is cleaved, followed by a slower phase in which
the rate is limited by the product release. The duration of the burst phase corresponds to
the fraction of the substrate that is initially bound to the enzyme. If the enzyme binds
efficiently to the substrate, such that nearly all the enzyme has bound a substrate when
the substrate is in excess, then this burst phase corresponds to the molar fraction of
enzyme to substrate. However, if the enzyme does not bind substrate efficiently, or the
substrate has competition for binding the enzyme, then the burst phase will be reduced.
Figures 3-33, 34 and 35 show that the reaction remains linear well beyond the initial
turnover, indicating that there is not a burst phase and that the overall rate is not limited
by ES disassociation.
67


50
79 nucleotide fragment over time. The reaction was monitored for 400 hours for a shift
in size of the 79-nucleotide product to full-length (106 nucleotide), but no such
conversion was observed. This argues against the possibility that a significant fraction of
614 cleavage products are ligated together.
Are the 27- and 79-nucleotide Fragments Acting as Catalysts or Inhibitors?
Two products are created by the cleavage of 614 at the ribo-adenosine, a 27- and a
79-nucleotide fragment. The standard Breaker and Joyce IVS protocol was designed to
select for molecules that are capable of self-cleavage (in cis), and cis cleavage is assumed
to predominate. However, without making assumptions about the structure and
mechanism of the 614 DNAzyme, we cannot rule out the possibility that either of the
cleavage products continue to act as a ribonuclease DNAzyme, or as a ribonuclease
DNAzyme inhibitor. We therefore set out to test this possibility using our model
DNAzyme, 614.
A caged-614 cleavage assay was set up as normal and, following extended
incubation in reaction buffer, the 27- and 79-nucleotide products were gel purified.
These products were then added in trans to a new sample of full-length caged-614, as
well as a caged-library sample (identical to caged-614, but containing a randomized 40-
nucleotide region between the catalytic and complementary primers). Following
initiation with 150 pulses of laser and incubation in reaction buffer, cleavage of the new
caged-614 sample occurred without change, even in the presence of equal concentration
of the 27- and 79-nucleotide fragments (Figure 3-14). This suggests against the idea that
the 27- and a 79-nucleotide fragments significantly affect 614 cleavage by acting as
inhibitors or catalysts, at least when present in equal concentration (which is
50


136
SL Size of initial library
RF Reactive Fraction
NRF Non-reactive Fraction: NRF = SL*(l RF)
k Intrinsic rate of molecule each member of the initial library, n, is
assigned an intrinsic rate, k, which relates to its ability to be cleaved
under selection conditions. The intrinsic rate is sequence-dependent and
therefore unchanging.
kslowest The rate constant for the slowest catalysts in the library.
kfastest The rate constant of the fastest molecule in the library.
Mifen The intrinsic rate of molecule (k) expressed in terms of half-life:
Vilife* = ln2 / k
St Selection time: the amount of time allowed for a molecule to cleave itself
and therefore achieve survival.
m or b Distribution function for reactive molecules. Two functions have been
tested thus far, a linear function with slope = m, and an exponential
function with decay rate = b.
L Leakage: the fraction of the pool that survives a round of selection
independent of St and intrinsic rate constant.
Model of survival: first order chemical rate based on a molecules intrinsic rate
constant (kn) and the selection time (St), P(cleavage) = exp(-£ St)
During an actual in vitro selection, the only data that can (easily) be obtained
experimentally are fraction cleaved vs. time for the entire pool at each round of
selection. Once the progress curve is determined for each round of selection, data can be
extracted to make other useful plots that illustrate the progress of the entire selection
process. One such plot compares the fraction cleaved at time = t for increasing rounds of
selection. If time equals the selection time, this plot shows the amount of sample
cleaved at the completion of each round of selection (the surviving fraction). It may also
be useful to plot the fraction cleaved at a time less than the selection time, particularly
when comparing I VS experiments performed with various selection times, or when
comparing IVS experiments in which nearly 100% of the fraction has been cleaved at the
completion of each round (t = St). From these data we would like to estimate the


17
When organic chemical theory expects a large (orders of magnitude) impact of a
change of structure on molecular behavior, and only a small impact is seen (a factor of 2
to 10), something must be wrong. Therefore, this "orders-of-magnitude" difference in
expectation versus outcome needs to be pursued. Perhaps the most formidable challenge
to our understanding of the origins of life lies in the experimental indication that function
is excessively rare in random polymer libraries. Therefore, if a plausible model for the
origin of life is ever to emerge based on an "RNA-world" scenario, we need to understand
why functionality does not seem to significantly enhance the catalytic power of the
library.
Although a key obstacle in the origins of life is the creation of function from
random libraries, none of the IVS experiments conducted thus far have been designed to
directly estimate the frequency of function within a library, or determine directly how this
frequency changes with various types of libraries and under varying selection schemes.
Most in vitro selections are designed to determine if a desired activity exists in a library,
and therefore are content to isolate a few molecules with the desired activity. The
previously used methods of estimating the frequency of catalysts in a library are not
comprehensive (Johnston, 2001, Bartel, 1993, Taylor, 2001), thus the few estimates of
catalytic frequency provided are not to be considered realistic. Indeed it is difficult to
imagine life emerging spontaneously from nucleic acids given these kinds of numbers,
especially given the difficulties in obtaining RNA (or DNA) in polymeric form under any
plausible prebiotic reaction conditions in water. Initially, the limited goal of isolating any
activity from a library was understandable given that it was not known whether any such
activity existed in a random library. Knowing that activity can be isolated from random
17


130
Trans-cleavage was tested directly by incubating 5-labeled ang + ribose primer
with the pool derived from H round 7 (Figure 4-7). The presence of equimolar ang +
ribose primer substrate slightly reduced cleavage of molecules in the H.Round#7 pool,
while resulting in nearly 75% primer cleavage. Nine-fold excess of ang + ribose over
H pool reduced the cleavage of molecules in the H pool by half, while resulting in
five times more primer cleavage (the percent of primer cleaved was reduced in half, but
since the total amount of primer present was 9-fold higher, the net result is an increased
amount of product formation).
Selection condition H contained 1 M NaCl and 1 mM MgC^ in the reaction
buffer. It was assumed by Breaker and Joyce, and most subsequent researchers, that the
molecules generated using IVS with MgCh in the buffer would utilize the magnesium as
a cofactor (it was never overtly stated, but it is possible molecules that are Mg-
independent would be lost during the work-up phase of selection and therefore selected
against). Following this assumption, molecules would remain inactive until the
magnesium cofactor was added. Our experience with 614 demonstrated that molecules
selected with magnesium as a cofactor can in fact be magnesium-independent.
Pools generated from H round 7 showed similar kinetics regardless of whether
Zn was substituted for magnesium, or in fact if no divalent cation was present (Figure
4-8A). To remove the possibility that trace metal contamination contributed to catalysis
in the absence of magnesium or zinc in the reaction buffer, new buffers were made with
double-distilled water and ultra-pure NaCl. Cleavage was unchanged by the removal of
MgCl2, the addition of EDTA, or the addition of DEPC (used in previous buffers) to the


101
time (hours)
0 2 4 7 10 24 32 48 72 96
time (hours)
0 3 9 16 23 40 69 91
< d-614 (full length) r-614>
cleaved r-614
>
full-length r-primer
*
0 < cleaved r-primer >
%
Figure 3-24. Both r-614 and d-614 cleave r-primer (ang + ribose) in trans. 614
was internally labeled and r-primer was 5-end labeled.


144
TABLE 4-1 A. Probability of survival following successive rounds
of selection (based on intrinsic rate and selection time)
St = half life (min)
300
X round
of selection:
0
1200 600
intial rate (min'1)
6E-
04 0.001
300
0.002
150
0.005
75
0.009
50
0.014
37.5
0.018
18.75
0.037
9.375
0.074
4.688
0.148
Probability of survival following round X of selection:
1111111
1
1
1
1
0.16
0.29
0.50
0.75
0.94
0.98
1.00
1.00
1.00
1.00
2
0.03
0.09
0.25
0.56
0.88
0.97
0.99
1.00
1.00
1.00
3
0.00
0.03
0.13
0.42
0.82
0.95
0.99
1.00
1.00
1.00
4
0.00
0.01
0.06
0.32
0.77
0.94
0.98
1.00
1.00
1.00
5
0.00
0.00
0.03
0.24
0.72
0.92
0.98
1.00
1.00
1.00
6
0.00
0.00
0.02
0.18
0.68
0.91
0.98
1.00
1.00
1.00
7
0.00
0.00
0.01
0.13
0.64
0.90
0.97
1.00
1.00
1.00
8
0.00
0.00
0.00
0.10
0.60
0.88
0.97
1.00
1.00
1.00
9
0.00
0.00
0.00
0.08
0.56
0.87
0.97
1.00
1.00
1.00
10
0.00
0.00
0.00
0.06
0.52
0.85
0.96
1.00
1.00
1.00
TABLE 4-1B. Population distribution following successive
rounds of selection.
St = half life (min)
300
1200
600
300
150
75
50
37.5
18.75
9.375
4.688
X round
intial rate (min )
6E-
04 0.001
0.002
0.005
0.009
0.014
0.018
0.037
0.074
0.148
of selection:
percent of population after X rounds of selection:
0
10.00
10.00
10.00
10.00
10.00
10.00
10.00
10.00
10.00
10.00
1
2.09
3.84
6.56
9.84
12.30
12.92
13.07
13.12
13.12
13.12
2
0.37
1.27
3.70
8.32
12.99
14.33
14.67
14.78
14.78
14.78
3
0.06
0.40
1.97
6.65
12.99
15.04
15.58
15.77
15.77
15.77
4
0.01
0.12
1.03
5.20
12.70
15.44
16.18
16.44
16.44
16.44
5
0.00
0.04
0.53
4.02
12.27
15.67
16.62
16.95
16.95
16.95
6
0.00
0.01
0.27
3.09
11.79
15.80
16.96
17.36
17.36
17.36
7
0.00
0.00
0.14
2.36
11.27
15.86
17.23
17.71
17.71
17.71
8
0.00
0.00
0.07
1.80
10.75
15.88
17.46
18.01
18.01
18.01
9
0.00
0.00
0.04
1.37
10.23
15.87
17.65
18.28
18.28
18.28
10
0.00
0.00
0.02
1.04
9.72
15.83
17.82
18.52
18.53
18.53
= initial population distribution
zone A = below level of detection (2%)
zone B = population in flux
zone C = initial proportionality is unchanged (within 90%)


124
were stopped by the addition of formamide stop dye and freezing (-20C). Aliquots were
also removed from each pool and stopped at various times (from 2 to 170 hours) to assess
the progress curve for cleavage at each round of selection. The majority of each pool was
run on a thick sequencing gel and the cleaved fraction of the pool was excised and
purified. The purified survivors of each round were amplified with PCR and subjected
to an additional round of selection. This procedure was repeated for a total of eight
rounds of selection.
Figure 4-1 shows the percent of full-length substrate for each selection condition
that was converted to the cleaved product at the time each selection was stopped at each
round of selection. The first selection is termed round zero because it is the selection
of the initial random library and has therefore not undergone any rounds of selection. As
expected from a random library, inactive molecules dominate the library in the early
rounds of selection, and cleaved substrate is therefore nearly undetectable. Of particular
interest are samples E and G. Sample E contained ten times the starting material as other
samples, and sample G contained more radiolabel than other samples, and for these
reasons, both samples have a higher signal-to-noise ratio. Both samples contained the
BJ + cage primer with standard nucleotides, but the primer in sample G was de-
pro tected by laser prior to the initiation of the I VS experiments (and was therefore
effectively BJ + ribose), whereas sample E was protected until exposure to laser. Both
samples were exposed to laser at the initiation of each round of selection. Both samples
were allowed to react for two hours prior to the addition of stop buffer, and both showed
low levels of conversion to the cleaved product. This low level of cleavage (less than
1%) suggests a low level for laser damage.


157
a
*
c
8
I
A
60
50
40
30
20
L
10:-,

oU
25
75
100 125
hours
150
175 200
B
H R#1 60nM
HR#2 60nM
HR#3 60nM
HR#4 60nM
HR#5 60nM
H R#6 60nM
H R#7 60nM
HR#8 60nM
225
o HR#1 15nM
o HR#215nM
H R#3 15nM
HR#4 15nM
. HR#5 15nM
HR#6 15nM
H FW7 15nM
HR#8 15nM
HF3 3nM
H R#5 3nM
H ra7 3nM
Figure 4-4. Kinetic profiles for Selection H at various concentrations following
each round of selection shows detectable cleavage following Round 3 of selection.
DNA from the survivors of each round of Selection H was re-amplified from stocks
and progress curves for cleavage assayed at 3, 15, and 60 nM.


percent cleaved percent cleaved
110
A
75-\
60
45-
30-
15-
iA

s
B
35-i
30-
25-
20-
15-
10-
5-
i
0-
100
r~
10
200
300
hours
20 30
40
1
50
hours
-i
60
i
70
100 nM r-primer +
20 nM d614
a 200 nM r-primer +
20 nM d614
300 nM r-primer +
20 nM d614
500 nM r-primer +
20 nM d614
1000 nM r-primer +
20 nM d614
2000 nM r-primer +
20 nM d614
i 1
400 500
100 nM r-primer+
20 nM d614
* 200 nM r-primer +
20 nM d614
300 nM r-primer +
20 nM d614
a 500 nM r-primer +
20 nM d614
1000 nM r-primer+
20 nM d614
2000 nM r-primer +
20 nM d614

1 1
80 90 100
Figure 3-35. Multi-turnover kinetics for d-614 cleaving r-primer suggest the first
cleavage. (A) Full time course for r-primer cleavage by d-614. (B) Initial time course
for r-primer cleavage by d-614 reveals linear cleavage up to 50 hours. The enzyme
and substrate were mixed together either immediately before or immediately after the
slow cool protocol.


160
H.R#7 pool (30 nM) cleavage in presence of 270 nM r-primer
a H.R#7 pool (30 nM) cleavage in presence of 270 nM 30 nM r-primer
a r-primer (270 nM) cleaved by H.R#7 pool (30 nM)
o r-primer (30 nM) cleaved by H.R#7 pool (30 nM)
H. R#7 self-cleavage (15 nM)
H.R#8 self cleavage (30 nM)
Figure 4-7. Selection H following round 7 cleaves r-primer in trans. Selection H
(30 nM) following round 7 was incubated with either 30 nM or 270 nM r-primer.
Cleavage of both Selection H DNAzymes and the r-primer substrate was determined.


71
cleavage. If, however, kcat(bi) k_i, the addition of the chase should have no impact on
subsequent cleavage (assuming sufficient time has been granted before the addition of the
chase to ensure that all the substrate is bound by the enzyme). The addition of chase to r-
614 resulted in an intermediate result: cleavage was dramatically reduced, but cleavage
continued to increase following the chase. Several scenarios may be put forward to
explain this:
(a) The rate of disassociation of substrate from enzyme is within an order of
magnitude of the rate of cleavage (kcat(bi) ~ k_i).
(b) C/s-cleavage may not be completely quenched by the chase (or at least not at the
concentration of chase used). The residual cleavage following the addition of the
chase could therefore be cis-cleavage. The apparent rate of cleavage with added
chase could be lower than the rate of cis-cleavage without chase if czs-cleavage
normally occurs through a process of folding, unfolding, and refolding, and at
some point during the folded state, it self-cleaves. Given this scenario, adding
chase may reduce some c/s-cleavage by competing with 614 for czs-folding. This
implies that kcat(uni) ~ k-i(uni)
(c) Not all of the substrate is bound by the enzyme at the time the chase is added, so
even though kcat(bi)>:>k-i, the addition of the chase competes for association of
substrate to enzyme and subsequently eliminates cleavage of unbound substrate.
This implies that the rate limiting step is association (kcat(bi)k|).
(d) Insufficient chase has been added. If kcat(bi)k-i, the enzyme-substrate complex
will disassociate rapidly. If association is also relatively rapid, the enzyme may
circulate through a number of chase molecules before it finds labeled substrate.
The reduction in cleavage is therefore proportional to the number of chase
molecules plus substrate molecules relative to enzyme molecules.
The chase experiment was therefore repeated with low concentrations of r-614
and high concentrations of d-614 (4nM r-614, 2025 nM d-614 and 10 uM chase added at
t = 4 hours, or 0.5 nM r-614, 64 nM d-614 and 1280 nM chase added at t = 45 minutes).
Similar results were observed: ribose-614 plus chase was gradually cleaved, but
always at levels far below the minus chase condition.
71


151
library, to estimate the distribution of catalysts in the initial library. The proximity of the
peak for catalytic distribution is also indicative of the relative proportion of catalysts with
half lives between St and St/6. This method relies on determining the population
distribution at a least one round (preferably a round shortly following population
stabilization). The population distribution at a round can be determined experimentally
by sampling individual catalysts (by cloning) and individually assessing their individual
rates. Sampling approximately 100 individuals should give a reasonable estimation of
the entire population distribution.
It may also be possible to estimate population distribution using statistical
analysis of the cleavage progress curve. This population progress curve is the sum of the
progress curves for each catalyst in the population. The Prony technique is a statistical
approach to deconvoluting the sum of many exponential curves into its individual
components. This approach is similar (in concept) to the way a Fourier transform
deconvolutes a complex signal into the sum of many sine functions. The Prony technique
varies both the exponential rate and scaling parameters of multiple exponential functions
to achieve an optimal the fit to the experimental data.
Both statistical and experimental methods can be used to estimate the population
distribution for the stabilized population. The distribution of catalysts in the initial
population is directly proportional to the catalysts in the stabilized population for all
catalysts with half lives less than ~St/6. Although both techniques will estimate the
abundance of catalysts with half lives greater than St/6 in the stabilized population, this
unfortunately provides no information about the distribution of these catalysts in the
initial population. The irrelevance of catalysts with half-lives greater than St/6 is


82
Table 3-1. Summary of Results from Chapter 3
Reasons for cleavage plateau of 614
Missing ribose-A in ~5% of primer (higher for caged ribose due to synthesis).
Alternative inactive conformation -15%.
Mutations that lower catalytic activity in both primer and N40 regions.
Not insufficient laser.
Not complementary strand inhibition.
Kinetic analysis
614 cleaves in cis (unimolecular) at low concentrations (< 20 nM).
614 cleaves in trans (biomolecular) at high concentrations.
614 cleaves various substrates in trans.
kcabi)= 0.035 hr"1
kobs(uni) 0.006 hr
The increase in the observed rate of cleavage at lower temperatures indicates
that the rate is a balance between the chemical step and the association step.
The rate of ES disassociation (k_i(bi)) for trans-cleavage is greater than the
chemical step (kcat(bi))
Structural Analysis
Many structures for c/s-folded 614 are possible, but none have yet to be proven.
There is an interaction between l7CGACTCACTAT27 and 85GTAGTGACG93
that occurs in both cis and tras-folded 614.
j c io
There is an interaction between CATG of the Enzyme (614) and Substrate
that cannot occur in m-folded 614.
82


16
deoxyuridine, and used them in I VS experiments in an effort to demonstrate that adding
functionality increases the odds of finding catalytically powerful sequences in a DNA
library. Judging from published examples, it seems to be now accepted that adding
functionality typically increases the catalytic potential of a DNA library perhaps by a
factor of two to ten, not by orders of magnitude.
In a recent example, Joyce and co-workers selected a zinc-dependent RNAse
activity utilizing an imidazole-functionalized nucleotide in their library (Santoro, 2000).
No significant catalytic boost was detected, however, over DNAzymes isolated earlier by
the same researchers using non-functionalized nucleotides (Santoro, 1997). Together with
other similar findings, this suggested that functional endowment offers less than an order
of magnitude enhancement of the catalytic potential of the library (Ang, 1999; Ang,
2000). Analogous experiments where functionality was introduced as a cofactor, binding
to but not covalently attached to the DNA molecules, generated analogous conclusions
(Roth, 1998).
What appears to be absent from the literature is the sense of astonishment at these
results. Functional groups are supposed to be important in organic reactivity. For
example, in an experiment in which a polypeptide was generated that folded in solution
and catalyzed the decarboxylation of oxaloacetate (Johnsson, 1990; Johnsson,93), the
amino groups performed the catalysis. Having them mattered; not having them mattered
even more, as catalysis went down by several orders of magnitude (to background) when
they were removed. In chemical models, adding imidazole (for example) increases
reactivity of catalytic models by orders of magnitude, not by a factor of two to ten (Zepik,
1999).
16


27
DNA. Following ethanol precipitation of PCR products, DNA pellets were resuspended
in 25 pL of exonuclease solution (5 units lambda exonuclease enzyme, 67 mM glycine-
KOH (pH 9.4), 2.5 mM MgC^, 50 pg/mL BSA) per 100 pL PCR. Samples were mixed
thoroughly and incubated at 37C for 30 minutes. Reactions were terminated by adding
excess formamide stop dye (1 mg/mL xylene cyanol, 1 mg/mL bromophenol blue, 10
mM EDTA, in 98% formamide) and heating at 80C for 10 minutes. The single-stranded
products were loaded on an 8% PAGE/urea gel and full length ssDNA products excised
using a sterile razor (thoroughly washed with ethanol between each sample). Each gel
slice was crushed and eluted in 350 pL elution buffer (500 mM NH4OAC, 0.1 mM
EDTA, 0.1% SDS) overnight. Gel purified samples were then extracted in
phenol/chloroform/isoamyl alcohol, then extracteded choloroform/isoamyl alcohol, and
precipitated in ammonium acetate and ethanol.
Asymmetric PCR was also used to generate single stranded DNAzymes. In this
procedure, the complementary strand primer was synthesized with a 10-20 nucleotide
extension (usually poly-adenosine) connected to the 5-end of the primer by one of
several linkers (C9, Cl8, 2O-methyl-Uridine). Vent and Taq polymerase cannot
efficiently read-through these linkers. Therefore, the catalytic strand generated by
extension of the catalytic primer terminates at the linker, generating the correct size for
full-length product. Complementary strand molecules generated by extension of a
complementary strand primer containing the 5 tail are longer than the full-length
product by the length of the tail, and are easily separated by gel purification on an 8%
PAGE/urea gel. Because Vent and Taq polymerase occasionally terminate extension
(about 50% of the time) when encountering a caged-ribose in the template, PCR using a
27


7
vestiges are molecular fossils leftover from a time in history when RNA was the sole
biopolymer (as opposed to convergent evolution) (Gilbert, 1986). These considerations
offer support for the previous theoretically-based proposition that modem life-forms
passed through a single biopolymer, RNA-based life-form, a period of time known as the
RNA world (Woese, 1967; Crick 1968; Orgel, 1968). Reconstructions of the RNA
world indicate that RNA life-forms possessed quite complex metabolism: molecular
fossils indicate that the RNA world developed the RNA cofactors ATP, coenzyme A, S-
adenosylmethionine, and NADH. lit follows that the RNA world needed these,
presumably for phosphorylations, Clasisen condensations, methyl transfers and
oxidation-reduction reactions (Benner, 1989).
The discovery that modem life contains RNA that acts as a catalyst demonstrated
that RNA is in fact a single bio-polymer capable of both genetics and catalysis, thus
giving experimental support to the proposition that modem two-biopolymer life is
descendent from a single-biopolymer (RNA) life-form (Usher, 1976; Cech, 1981,
Geurrier-Takada, 1983). The existence of an RNA world does not necessitate,
however, that life actually originated with RNA. Indeed, applying chemical knowledge
to work forward from prebiotic earth to modem life suggests that ribonucleic acids may
have been a very poor candidate as the primordial monomer in the origins of life.
Instead, the RNA world may have served as a transitory life-form between the original
single-polymer life-form and todays robust, two-biopolymer life-form. We now turn our
attention to insights gained by working forward from basic chemical principles.
7


145
Examining the Impact of the Selection Time (St)
Figure 4-11 shows the results of another simulation in which the selection time
(St) was varied (St = 15, 246, and 16922 minutes for case 1, 2, and 3 respectively) while
all other parameters were held constant. Increasing the selection time from 15 minutes to
246 initially resulted in a larger fraction of the population being cleaved at the end of
each round of selection, corresponding to earlier initial detection. Increasing the
selection time further to 16922 minutes reversed this trend, demonstrating that the round
at which catalysts are first detected (at t = St) can vary according to the selection time.
When the selection time is very short, very few molecules meet the selection
requirement, and thus more rounds are required to isolate these infrequent molecules.
Extending the selection time increases the number of molecules that may survive, and
this may reduce the number or rounds required to isolate catalysts. However, as the
selection time is increased further, the average rate of the surviving population remains
low, and more rounds are required to detect cleavage in the population.
As seen in the simulation with varying leakage rates, the fraction of the
population cleaved at a given time reaches a plateau with sufficient rounds of selection,
indicating that the selections isolated stabilized populations (Figure 4-11, Panel B).
When the selection time is short relative to the half life of the fastest catalyst, a
population of only the fastest catalysts is isolated. This short selection time also results in
a plateau below 95% for the fraction of the population cleaved at the end of the round of
selection (cleavage at time t = St). Because there is no significant differential enrichment
for catalysts with half-life 6-fold slower than the selection time, longer selection times
equally enrich all molecules with a half-life (rate) approximately 6-fold lower than the


CHAPTER 1
INTRODUCTION
Requirements of a Living System
In 1994, a committee empaneled by NASA defined life as a self-sustaining
chemical system capable of undergoing Darwinian evolution (Joyce, 1994). Non
living entities, such as natural phenomena like fire or crystal formation, often have
physical properties commonly associated with life-forms such as energy consumption,
ordered structures, and growth and replication. A living system is distinguished from
non-living on the basis that living systems can evolve. The properties of nonliving
phenomena such as fire or crystal formation follow directly from physical laws and the
particular environmental conditions of the system. As physical laws presumably do not
change, the behavior of nonliving systems cannot change. An evolving system has
inherent physical properties as well, but in contrast to nonliving entities, the specific
physical properties of the entire system are derived from information contained within
the system and can change without destroying the ability to evolve. These physical
properties are either intrinsic to the information-containing molecule or generated from
the information).
As all life requires obtaining precursors and energy from an environment, the
requirement that life be self-sustaining is always relative to a particular environment.
The NASA definition also limits life to chemical systems, revealing a bias towards life
as we know it and in which we are most interested. Evolving systems such as memes
1


100
430.0 r-614
a 108.0 r-614
* 19.4 r-614
3.5 r-614
1.1 r-614
0.3 r-614
[r14fnM
slope
R2
430.0 108.0 19.4
-0.003219 -0.003537 -0.003047
0.7812 0.8898 0.9193
3.5
-0.003122
0.9034
1.1
-0.003026
0.9675
0.3
-0.004160
0.9668
Figure 3-23. Cleavage rates for low concentrations of r-614 (but not high) fit well
to a linear equation when the natural log of substrate concentration is plotted vs
time, indicating first order exponential rate.


Ill
50-1
40-
1
| 30-
u

i
50
A
a 2 nM r-614,
without slow cool folding
A 2 nM r-614,
with slow cool folding
i 1 1 1 1 1
75 100 125 150 175 200
hours
Figure 3-36. Intra-moiecular folding of r-614 is reduced by the omission of slow
cooling. Ribose-614 was incubated at sufficiently low concentration (2 nM ) as to
favor m-cleavage with and without the slow cool folding protocol.
Figure 3-37. Inter-molecular folding of r-614 is not affected by slow cooling.
Ribose-614 was incubated at sufficiently high concentration (200 nM ) as to favor
/ram-cleavage with and without the slow cool folding.


56
result of only two possibilities, incomplete de-protection or a higher fraction of the
primer missing the ribose-adenosine.
Several of the 614 clones with mutations were tested to see if their mutations in
fact reduced the rate of cleavage. Figures 3-18 and 19 show the cleavage profiles of
several mutants and demonstrate that nearly all of the mutations cause greatly reduced
cleavage rates. Together with the observations above, we have shown that approximately
6% of the total ribose-614 sample does not cleave because it is missing the ribose-
adenosine (20% for the caged-614), approximately 10-15% of the total sample may be
mis-folded and only slowly converts into the active conformation, and 68% of the
uncleaved fraction may not cleave due to mutations.
DNAzyme 614 Cleaves in cis and trans
As is the case with all enzymes, a DNAzyme reaction can be thought of as
occurring in several steps: binding the substrate, the chemical step (where covalent bonds
are made or broken), and release of the substrate. The I VS protocol was designed so that
the DNAzyme contains the substrate (ribose-adenosine) linked to the enzyme (the
random N40 region), and engineered such that the DNAzyme would fold back onto itself
via two clamps of base pairing, one on each side of the ribose-adenosine, thus bringing
the enzymatic region in proximity to the substrate. Because of the design, the
DNAzyme is expected to fold more rapidly with itself than bind other molecules at
typical concentrations, and therefore the reaction is assumed be unimolecular. However,
because all molecules in a library share the same 5- and 3-primers, over 60% of the
56


CHAPTER 5
SUMMARY
Our current understanding of the origins of life suggests that catalytic nucleic
acids must arise from random polymers of nucleic acids. Searching random libraries of
nucleic acid polymers has not produced the abundance and power of catalysis believed
necessary to spawn life. Further, the addition of chemical functionality to random
libraries has failed to improve the catalytic potential of random libraries anticipated.
This apparent discrepancy between observation and expectation may be
attributable to several shortcomings of the experimental design. Active catalysts may be
lost during the preparation of libraries, or during the enrichment for catalysts.
Furthermore, only incomplete descriptions have been attempt for any, focusing on only
the fastest catalysts in the library, rather than capturing completely the distribution of
catalytic power. It is conceivable that added functionality, or alterations of the library or
selection conditions, can dramatically alter the abundance and distribution of catalysts
without altering the rate of the fastest catalyst.
The goal of this thesis is to develop methods for estimating the distribution of
catalysts within a random library of DNA sequence. To aid in the development of
realistic models, we began with an in depth analysis of the catalytic behavior of an
individual DNAzyme, 614. We also addressed the problem of losing catalysts during the
preparation of DNAzyme pools by examining the utility of a protected ribose (caged-
ribose) in both kinetic analysis and an IVS. Simulations of IVS were also produced and
169


38
(compare lane 1 and 5). This suggests that the lower band in lane 5, although the size of
the catalytic strand cleaved at the ribose, is not actually cleaved product from the
catalytic strand. This band is actually the result of polymerase termination when the
complementary strand extension reaches the caged-ribose in the template. This is
confirmed by the observation that this band is diminished by exonuclease treatment
specific for the complementary strand (lane 7).
When the ribose-614 double stranded PCR product is treated with strong base,
there is a significant increase in the lower band, corresponding to conversion of the full-
length catalytic strand to cleaved product. Comparing lane 1 with lane 2 shows that
approximately half of the ribose-614 double stranded PCR product is converted to
cleaved product following treatment with strong base (corresponding to the cleavage of
the catalytic strand), while approximately half remains full-length (corresponding to the
complementary strand). The catalytic strand, approximately half of originally double
stranded ribose-614 PCR product, remains full-length following exonuclease treatment
(lane 3), although there is no increase in the cleaved product (the complementary strand
is degraded to single nucleotides). The full-length band remaining following treatment of
ribose-614 with exonuclease disappears when also treated with strong base (lane 4). This
full-length band remains for caged-614 treated with exonuclease and strong base (lane 8),
demonstrating that conversion of the upper band to cleavage product seen in lane 4
corresponds to the catalytic strand (there is slight conversion of full-length caged-614 to
cleaved product under strong base, possibly the result of partial de-protection under room
light). The absence of any full-length product in lane 4 demonstrates that nearly the
38


80
enzyme. These /ra^-interactions interactions eliminate structures formed for c/s-folded
614 in the first 38 nucleotides. Most of the structural elements predicted for c/s-folded
614 are preserved in the /raws-folded 614/substrate complex (Figure 3-46). The free-
energy predictions for the /ra/is-folded 614/substrate complexes were all much lower
than the c/s-folded 614 structures (cis-614 dG = approximately -12 kcal/mol, trans
614:substrate dG = -20 kcal/mol). This agrees well with previous experimental results
indicating that overall cleavage rate is affected by folding and occurs faster in trans.
Structural probing was also performed using chemical modifications specific for
single-stranded DNA. The /raws-folded conformation was favored by incubating high
concentration of 5-labeled d-614 (300 nM) in reaction buffer with and without excess
unlabeled primer (10 pM). The samples were allowed to incubate overnight to maximize
folding, after which they were treated with either potassium permanganate (specifically
modifies T) or DMS (specifically modifies G at high salt concentrations) for two minutes
and five minutes. Recovered samples were run on a PAGE-urea gel next to a 10-bp
ladder. All T and G nucleotides demonstrated some sensitivity to the respective chemical
modifications, indicating that either some 614 is not folded at all, or adopts multiple
conformations. Nonetheless, contrast levels were adjusted to discern the mosi-protected
nucleotides (suggesting they are base-paired). Figure 3-48 shows the result of this
analysis, indicating that T44, T46, T74, G33, G41, G65, G71, and G75 all show some
degree of differential protection. The pattern of protected and sensitive positions
observed under chemical modifications is not simultaneously compatible with any single
structure predicted in Figure 3-43 for c/s-folded 614 or Figure 3-46 for /ra/is-folded 614,
80


CHAPTER 2
MATERIALS AND METHODS
Preparation of Precursor DNAymes via PCR
DNAzymes for kinetics were prepared by PCR amplification of the template
(either synthesized by IDT, or from a clone) using one of various primers to generate the
catalytic strand and one of various primers to generate the complementary strand. All
templates had a common 5 and 3 constant region to which the complementary and
catalytic strand primers can bind. Between the primer binding sites is a 40 nucleotide
region of varying sequence:
library template (complementary strand): 5' -gtgccaagcttaccgtcac (n40) -
GAGATGTCGCCATCTCTTCC
614template: 5'-ctgcagaattctaatacgactcactataggaagacatggcgactctc-
ACATCATGCGAGCACACGCAATAGCCTGATAAGGTTGGTAGTGACGGTAAGCTTGGCAC
Two sequence variants of the catalytic strand primer were commonly used, but both
bound to the constant region of the various templates:
ang + ribose catalytic strand primer:
5'- CTGCAGAATTCTAATACGACTCACTATrAGGAAGACATGGCGAC-TCTC)
BJ + ribose catalytic strand primer:
5'-F-GGGACGAATTCTAATACGACTCACTATrAGGAAGAGATGGCGACATCTC)
The differences between the ang and BJ primers are underlined; represents an
alignment gap; N is equimolar concentrations of each dNTP, F is a 5Fluorescein;
rA is a ribo-Adenosine. Primer variants were also used in which the ribo-adenosine
was replaced with either deoxy-adenosine (ang ribose and BJ -ribose) or caged-
24


72
Previous enzyme-saturation experiments indicate that 2025 nM d-614 is sufficient
to approach substrate saturation, therefore arguing against scenario (c) as a likely
explanation for the fact that there is residual cleavage following the addition of the chase.
The ratio of enzyme (d-614) to chase (ang + ribose) in these latter two experiments was
either 1:5 or 1:20. Slightly less cleavage was observed after the addition of chase in the
1:20 enzymexhase experiment, suggesting scenario (d) as a possible explanation for
residual cleavage following the chase.
Chase experiments conducted with ribose-614 cleaving in cis (2 nM) showed no
significant change in cleavage following the addition of 30 nM chase. This suggests that
either intra-molecular association is complete prior to the addition of the chase and it
does not unfold prior to cleavage, or that intra-molecular folding is not affected by the
presence of competitors at 30 nM. Unlike trans-acting catalysts, the absolute
concentration not just the proportional dilution can alter the effectiveness of a chase.
Previous experiment have demonstrated that low concentration of competitors (<3 nM)
added at time = 0 do not compete with r-614 cleavage, while higher levels of competitor
(270 nM) reduce, but do not eliminate, ribose-614 cleavage (30 nM) (Figure 3-27 and
28). Taken together, this suggests that at least some r-614 c/s-cleavage can persist
despite increasing concentrations of competitor, supporting the plausibility of scenario
(b).
A chase experiment was therefore performed using trace amounts of ang +
ribose primer (4 nM ) as a substrate with near-saturating levels of d-614 (2000 nM )
enzyme. This substrate has no intrinsic c/s-cleaving potential, eliminating the possibility
that cleavage following the chase is caused by c/s-cleavage. Furthermore, the unlabelled
72


30
using the complementary strand primer complementary no 5-phosphate with the 5-
truncated catalytic primer 3-cleaved oligo. Fresh double stranded PCR products were
cloned using the TOPO TA Cloning System (Invitrogen) and plated onto Agarose plates
containing ampicillin. When cloning sequences from the pool of survivors from the H
selection condition, transformed cells were given only 15 minutes to recover in antibiotic
free media prior to plating on antibiotic containing plates. This prevents recently
transformed clones from doubling prior to plating and therefore favors sequencing only
unique species from the original pool.
DNA was prepared from individual clones using a modified alkaline lysis
protocol. Individual clones from plates were used to inoculate 5 mL of TYGPN media
containing freshly added ampicillin and grown overnight at 37C with vigorous shaking.
Following overnight growth, 2 mL of cells were pelleted by centrifugation for 1 minute at
16,000 g. Growth media was removed and the cell pellet re-suspended in Solution I (300
pL: 10 mM Tris pH 8.0, 1 mM EDTA, and 50 pg/ml Rnase A). Solution II (300 pL: 1
mL fresh 2 M NaOH, 1 mL 10% SDS, 8 mL water) was then added and the samples were
then inverted gently to mix. Solution III (300 pL: 5 M potassium acetate pH 4.8) was
then added, followed by gentle mixing. The sample was centrifuged for 2 minutes at
16,000 g, after which 500 pL of chloroform was added and gently mixed. The samples
were then centrifuged again at 16,000 g and 750 pL of the aqueous supernatant was
transferred to a new tube containing isopropanol (750 pL). This sample was centrifuged
for 6 minutes at 16,000 g. The isopropanol was completely removed, and the remaining
pellet re-suspended immediately in 100 pL water, followed by the addition of 100 pL
15% PEG-8000. The re-suspended pellet was stored on ice for 30 minutes (or overnight),
30


83
Lamda exonuclease
strong base
ribose or caged-r
LANE #
+ +
- + +
R R R R
12 3 4
- + +
+ +
c c c c
5 6 7 8
106-nt full-length 614
* ,*
79-nt cleaved product >
m #
Figure 3-1. Lamda exonuclease degrades the 5-phosphorylated complementary
stand of ribose-614 and caged-614. Ribose-614 and caged-614 were generated via
PCR; both the catalytic and complementary strand were labeled with alpha-CTP.
Treatment of double stranded ribose-614 with strong base converts the catalytic strand
to the cleaved product, while treatment with exonuclease degrades the complementary
strand to single-nucleotides (not shown on gel). PCR with the caged-ribose primer
generates a complementary strand that is that is the same length as the cleaved
catalytic strand, corresponding to the where the polymerase terminates elongation
when encountering the caged-ribose in the template.


Family
a
or
a
a
a
a
a
a
a
P
P
P
P
P
P
P
P
P
P
P
P
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
6
8
5
8
8
.119 (0)-
Hr#3.1
Hr#3.13
Hr#3.4
Hr#3.14
Hr#4.139
Hr#4.25
Hr#4.135
Hr#4.133
Hr#3.3
Hr # 4.131
Hr#4.27
Hr#5.148
Hr#5.159
Hr#5.154
Hr#5.151
Hr#8.161
Hr#8.58
Hr#8.162
Hr#8.53
614
Hr#4.137
Hr#5.160
Hr#5.153
Hr#5.150
Hr#5.149
Hr#5.145
Hr#5.144
Hr#5.147
Hr#5.143
Hr#8.166
Hr#8.56
Hr#8.47
Hr#8.169
615
Hr#4.26
Hr#5.142
Hr#5.146
Hr#5.158
616
GCGTGTACTTGGAGAACTGGCGCATTACTACGTCATGCTCG
GCGTGTACTTGGAGAACTGGCGCATTACTACGTCATGCTCG
GCGTGTACTTGGAGAACTGGCGCATTACTACGTCATGCTCG
GCGTGTACTTGGAGAACTGGCGCATTACTACGTCATGCTCG
GCGTGTACTTGGAGAACTAGCGCAT-ATTACGTCATGCTCG
GCGTGTACTTGGAGAACTGGCGCATTACTACGTCATGCTCG
GCGTGTACTTGGAGAACTGGCGCATTACTACGTCATGCTTG
GCGTGTACTTGGAGAACTGGCGCATTACTACGTCATGCTCG
GCGTGTACTTGGAGAACTGGCGCATTACTACGTCATGCTCG
CATCATGCGAGCACAC- GC -7
-CATCATGCGAGCACAC-GC-
-CATCATGCGAGCACAC-GC-
CATCATGCGAGC ACAC GC -
-CATCATGCGAGCACAC-GC-.
CATCATGCGAGC AC AC GCT.
-CATCATGCGAGCGCAT-GC-
-CATCATGCGAGCACAC-GC-
C ATCATGCG AGCAC AC GC
CAGCATGCGAGCACAC-GC
-CATCATGCGAGCACAC-AC
TGA
TGA
T-TGA
i-ATGA
ATGA
-TGA
-TGA
T-TGAl
A-TGA|m^^B|-"
CATC ATGCGAGCACAC GC AATAGCC TGATAAGGTTGGTAG -
a-ctgc--acaatccaac-accgattgct--gcaaaggttgttaggg
Hr#8
Hr#8
Hr#8
Hr #8
Hr#8
Hr#8
Hr#5
Hr# 5
Hr#5
Hr #4
Hr #4
Hr #4
Hr #4
Hr#3
Hr#3,
Hr#3,
Hr#3 .
Hr#3
Hr#3 .
Hr#3 .
Hr#3
Hr# 3 .
Hr#3 .
Hr#3
Hr#3 .
Hr#3.
Hr# 3 .
16 5 (0.2 ) A CA-TGA- -GCTGGATGCAGGTCTGCAACACCGGGACGCGTGTA
167 T TTGGTC GATACGTTGGTATTGACACTGGTGTTGTTTCTAG
170 (1.3) GACGCGACCTAT--GTCCACCTCAGGA--CATGGTGCGCACGGGG
55 GCGACCAAATGAGG- -GTCCGGACTGGAC- -CCT- -TCGGTACGAGG
168(0.1)TACACGACGTGCACTTTGGTAGCTAGT--CCTG--TGTTGGGTGG
163(0.1)CAT--CCTGTGGCT-AGACAACCTTGAGG--TGTCGCGTGCTCATCA
157
155
141
30
132
140
17
126
11
15
5
129
122
124
12
127
123
10
121
2
128
111
107
112
101
120
108
110
106
114
115
118
117
116
103
113
104
109
A CGGTGATAGCTTGCTGTGGGATTCCCAT GAGCCTGTCAAA
A CTATCTAATCCGGTGGTCAAATACACTCGTAGACCGTG -TG
C - TTAGCGTGTTCTCATAAAGGCTG AGCAT CTCACCTTGC CG
GACCCTGGAACAGGCATTTTTGAGCCCTTCCCTGCCGTTAG-TGACG
GATACCCCTCTGATCCCCCTGTCGGCTGGGTTGATCATTG-G
T TCGGC-AATCGGGTGCACAT-CATATTGGCCGGCTGAGCATG
A TGTCCC-AGTTTCTTTTGGA--TAGGACGATCGAAAAAGTTTG
C CATCCCGTTTGATATCTCCCTACCTAGTATCTGCTTAGGGG G
G CAGTAT-CGCGAAGGCATTTTTTTCATTAA--GCTAAGGTTTG
T CGGTTAAAGTTTCC CTTGATTTATTAGGCCAGTGGTGGG - G
C T -TTGACTTGCCCACTAGATCCGGTTGATTAGGGCCAGAC G
T AATACCACACTCAGTAATAQCCGTGTCCTATTTTACTGTG
C ACGAGG TCAGATTCTAGTCGGGGTATTTGTGCCCTACTTG
C GACTCTGGTTTTAAAACGGGAAAGATCTAGCCGACGTCAG- -
G TAAACGCCA- GC CTGTAGCAG -TAGC-- GATGGTCGGTATAAAG
TATTACCCTCAAGT CAATTATTATCGC AGT -GT TCCTTTGTG
T-TGAACGACGGGC-CGCTAGCATTTATT--AGT-GGGTCCGCGGTG
G GTCAAAAGAACGCTTCGAAATCCCCTTCTCCCCCCAGTCG -
C ACGAGG TCAGATTCTAGTCGGGGTATTTGTGCCCTACTTG
G TAGTTA TTTTGGCACGGTTTAGGGGTCTCTCTATTCCGTG
T GTGCTA- GGTGTTCTCTGAG - CCAGACGTTAGTGTAGTTAAG
GGAA- TTTAAGAACTAAACAC - GC ATTGGTGAAAGTGTGCG
CTGAGTAATGTGAAGATACATGCCATGATGACTCG GCCCA G
GG AC TAGGAAAATC AATTCCGTT AAAGTGGCGG AATTG ACG -
TGTGTGTTTGCTAC-TCGCAACCCCTAGT--ATCGA--TGTCCTATG
CACGAG-GTCAGATTCTAGTCGGGGTATT- -TGTGC -CCTACTTG
AGTACG CAGATGGTGGGATATTTATTCACGG CACGAC AGTG
A TTCAAAGCTGGTTGTCT-TCATGTTG--TTCGGGATAOCGAGG
Figure 4-9. Sequence alignment of clones isolated from pools of survivors after round
for some sequences are shown in red and in parenthesis following the sequence name.
TCCAAAATTACTGTTAT- TGCTCTTAATTTTGTGCGTGGAGCG
G CCGATTTTTACGGCGAACTACCTTAGC CAGTATTCGTTGG
T CTAATT- -GGAGTAAATATTCCGTAGCT- -CGATGCCTGACGG
GATC--CAGTCGATCTCTTAGAGGTCGATGCTCT-TCCGTACCGT-G
ACGGATATCAGTAAAGTATGTATCGAGGACTATAA ACCAGCG
ACGG ATATCAGTAAAGTATGTATCGAGGAC ATAA ACCAGCG
CATCGTGTATGCATAATCTTTA--TGGTGTCACTTATCCGGGG
AATC CTGTTTACTCT-CGTGTAGTTGCCCCACAC-GGGTTGCG
AACC GAATCCTGTCAATATGCTGTAGTTCCATC AGTAGAG
TGTG CTAGGTGTTCTCTGAGCCAGACGTTAGTGTAGTTAAG- -
I, 3,4, 5, and 8 of selection (I VS H). Rate constants (1000*hr ')
Sequences 614, 615, and 616 are from a previous IVS.
Os
K>


140
Figure 4-10 shows the results of simulations for three test cases, each differing in
the value of the leakage parameter (L, ranging from zero to 4%). Panel A shows the
initial distribution for each population (the population was distributed according to a
linear function with m = 0 for all three test cases). Panel E plots experimentally
observable data the progress curve for each test case following successive rounds of
selection (the progress curve is a plot of the fraction of the population cleaved with
increasing time). From the progress curve, the amount of each population cleaved at time
= selection time (St), or time = 81 minutes, was extracted for each round and plotted in
Panel B and C. Although during the experimental implementation of a selection we can
only observe the progress curve, the simulations allow us to observe the population
distribution at each round. These data are plotted in Panel D (the intrinsic rate, kn, for
each molecule has been converted into its corresponding half life, so a rate of .00231
min'1 is plotted as a '/¡life of 300 minutes).
Most of the molecules in the initial library are non-reactive, and therefore are
cleaved at the background rate of only 10'7 min'1. The fraction of the population cleaved
at the end of the first round of selection (t = St) referred to as survivors is therefore
extremely low, and mostly a reflection of leakage. With each successive round of
selection, the catalytic molecules are enhanced while the non-reactive fraction is
diminished. Eventually the catalytic molecules reach sufficient abundance relative to the
non-reactive molecules that they surpass the threshold of detection, defined as 1% above
background cleavage (at time = St).
Although each case shown in Figure 4-10 had the same initial distribution and
abundance of catalysts, increasing the amount of leakage from zero to 0.2% delayed first


39
entire catalytic strand has been converted to the cleaved product by the strong base and
nearly the entire complementary strand has been degraded by exonuclease.
Asymmetric PCR with Tails. Followed by Gel Purification
A non-enzymatic method was also pursued for isolation of single-stranded
DNAzymes. In this procedure, termed asymmetric PCR, the complementary strand
primer was appended with a tail consisting of 10-20 nucleotides connected to the 5
end of the primer via a linker. Several linkers were tested for there ability to terminate
polymerase extension when encountered on the template complementary strand. If
polymerase extension of the catalytic strand in fact terminates when it encounters a linker
on the complementary strand, all catalytic strands would be the normal full-length.
However, all molecules generated by extension of the complementary strand primer
would be increased in effective size by the length of the tail appended to the primer.
Figure 3-2 shows the product generated by PCR amplification of a 106-nt
fragment using 5- P-labeled primers and Taq polymerase. PCR amplifications were
done in pairs: odd-numbered lanes show the product generated using the 5- P-labeled
catalytic strand primer (ang + ribose) with unlabeled complementary strand primer
(designated in the subsequent lane), while even-numbered lanes show the product
generated with unlabeled catalytic primer and 5- P-labeled complementary strand
primer. All products generated from complementary primers with tails are increased in
size by the length of the tail, while products generated from the catalytic strand primer
are of normal size (106 nucleotides) for each complementary strand primer except
C18#l (lane 1 and 2). This demonstrates that Taq polymerase does in fact terminate
effectively when it encounters a linker in the template for all linkers tested except
39


74
burst phase seen under multiple turnover conditions at lower temperature (while not
seen at higher temperatures) indicates lower temperatures increase the initial rate by
stabilizing intermolecular association between enzyme and substrate; this stabilization
also slows the disassociation of product from enzyme and therefore reduces the rate for
multiple-turnover subsequent to the first catalysis.
Structural Analysis of 614
The ability of 614 to cleave in cis is not particularly surprising, both because this
is the function for which it was selected, and because many ribozymes function to cleave
in cis. It is, however, rather interesting that 614 also possesses the ability to cleave
various substrates in trans, and furthermore, that the rate constant for trans cleavage is in
fact 6-fold higher than the rate constant for cis-cleavage. The ability of 614 to cleave the
ang + ribose primer, as well as a library of molecules containing the ang + ribose
primer, suggests that 614 has a binding motif which base-pairs with some part of the
ang + ribose primer common to all substrates. This binding positions a separate
catalytic motif near the ribo-adenosine.
To understand how this may work in both cis and trans, the Stewart and Zuker
mfold program was used to generate several structure predictions for 614 in cis and trans.
Of the many structures proposed by the mfold program, only those that positioned the
ribo-adenosine in a constrained but accessible manner were selected structures were
eliminated if the ribo-adenosine was either embedded in a lengthy helix (inaccessible) or
if it was positioned in the middle of an extended loop (unconstrained).
Figure 3-43 shows eight of lowest free energy structure predictions. Despite the
diversity of candidate structures, several common folds were present and used to group
74


78
compensatory mutations (614 G81C, C72G and 614 G81C, C43G). Neither of the
double-mutations restored catalytic activity, suggesting that neither structure B2 nor the
structures of Family A are in fact the actual structure. It remains possible, however, that
one of the structures of Family A or structure B2 is in fact correct, but G81 is crucial to
catalysis for reasons beyond its ability to base-pair with either C72 or C43.
The evidence above argues against the validity of Family A and C structures, as
well as structure B2, leaving only one candidate structure remaining (Bl). If structure B1
represents the correct fold for 614 acting in cis, and the essential structural elements are
preserved when 614 cleaves in trans, then nucleotides 41 through 91 should be sufficient
for cleaving the ang + ribose substrate. Because this truncated 614 (614 trunc41-91)
can conceivably adopt the other structures of Family A and C, the ang + ribose
substrate was altered to improve binding if it adopts the B1 structure but not by other
structures (this Bl substrate was produced by truncating ang + ribose to include only
nucleotides 21-35, with a G inserted between nucleotides 31 and 32, and A34 changed to
T, resulting in 5TCACTATrAGGAGAGTC). The catalytic ability of 614 trunc41-91
was then tested with both ang + ribose and B1 substrate. Neither substrate was
cleaved by 614 trunc41-91, arguing against our final proposed structures, A1 or Bl.
Structure predictions were made for Family A, B and C with 614 acting in trans
(Figure 3-46). The ability of 614 to cleave the ang + ribose substrate demonstrates that
nucleotides 50-106 are not a required element of the substrate for 614 rraws-cleavage.
The structure predictions shown for 614 cleaving in trans are those using ang + ribose
substrate, but are essentially the same for full-length substrates.
78


34
Table 2-1. Selection Conditions for Each of the Fifteen Parallel IVS Experiments.
Final Buffer:
Sample
Bases (AGC+)
primer
50 mM HEPES
pH7; 1M NaCI;
1mM of metal =
Selection
Time
(minutes)
A
T-base
BJ + cage
-0-
120
B
T-base
BJ + cage
Zn++
120
C
T-base
BJ + cage
Mg++
20
D
T-base
BJ + cage
Mg~
120
E
T-base lOx [library]
BJ + cage
Mg++
120
F
T-base
BJ + cage
Mg++
720
G
T-base
BJ + de-protected cage
Mg^
120
H
T-base
ang + ribose
Mg++
120
I
E-base
BJ + cage
-0-
120
J
E-base
BJ + cage
Zn++
120
K
E-base
BJ + cage
Mg"1*"
20
L
E-base
BJ + cage
Mg'**"
120
M
E-base
BJ + cage
Mg^
720
N
E-base
BJ + de-protected cage
Mg"**"
120
O
E-base
ang + ribose
Mg'
120
Following elution, samples were extracted in phenol/chloroform/isoamyl alcohol,
then extracted in chloroform/isoamyl alcohol, and mixed with ammonium acetate and
ethanol. Half of each sample was stored in ethanol, and the remaining half was
precipitated and resuspended in 50 pL water. For subsequent rounds of selection, thirty
microliters of the survivors of the previous round (in water) were used in a 100 pL PCR
containing 400 nM catalytic strand primer (either ang +ribose, BJ + cage, or BJ +
de-protected cage), 100 nM complementary strand primer (complementary + 5-
34


155
A R#8
B R#8
C R#8
E R#8
F R#8
G R#8
12456124561245612456 1245612456

[I Ml
llllilllllllll .........
H R#8
1 2 3 4 5 6
I Re JR*8 K/L R#8 MR8 N R#8 0 R#8
146146146 146146146
IMM.IIMIImI
Figure 4-2. Cleavage assays for all selections conditions after 8 rounds of
selection reveal that only selection H has abundant catalysts. Pools from all
selection conditions were prepared with ribose-A primer (either BJ + ribose or ang
+ ribose). Cleavage assays were initiated by adding 2X reaction buffer and the slow
cool protocol (rather than laser initiation). Only selection condition H had any
detectable cleavage following 49 hours. Time points 1-6 are 0, 2, 3, 11, 23.3, and 48.7
hours.


8
Progress Towards Understanding the Genesis of Life: Prebiotic Synthesis
of Potentially Useful Monomers and Polymers
The challenge of non-biological synthesis of useful monomers and polymers
argues against the simultaneous creation of two interrelated polymeric systems, but the
creation of even a single polymeric system for life (capable of genetics and catalysis) still
must meet the need for prebiotic synthesis of monomers and polymers. Knowing that
chemically based life does exist we must assume that monomers and polymers did form
in the prebiotic world as result of the basic principles of chemistry. It has not, however,
been easy to deduce how this might occur.
In light of the challenges to the prebiotic generation of life, what is so surprising
about the origins of life (other than that it happened at all) is that it appeared relatively
rapidly. The earth was formed -4.6 billion years ago (bya). The first ~0.5 billion years
of earths life, prior to the cooling of the earths crust and the subsiding of severe
meteorites bombardment, was inhospitable to the formation of any useful monomers, yet
evidence of life in the fossil record is apparent as early as 3.6 bya (Walter, 1983). This
provides a very short time for the generation of monomers, polymers and subsequent
genesis of life.
In order to apply chemical principles to gain insight into the prebiotic creation of
monomers, the prebiotic environment of the earth (approximately 4 billion years ago)
must be understood. Although definitive information about the earths early environment
and composition is slow coming, it is widely assumed that H2O, CO2, H2, and CO
existed. In a groundbreaking experiment, Stanley Miller demonstrated that combining
these basic elements with an electric discharge (presumably also occurring with sufficient
regularity on the prebiotic earth) produced a prebiotic soup containing a mixture of key
8


35
phosphate), 100 pM dNTPS (either AGCT or AGCE), 10 mM KC1, 20 mM Tris-HCl
(pH 8.8), 10 mM (NH^SO^ 2 mM MgS04, 0.1% Triton X-100, and 3 units Vent exo-
polymerase, and 7.5 pCuries alpha-32P-CTP. PCR amplification consisted of 3 minutes
at 96C, followed by various numbers of cycles of 45 seconds at 96C, 45 seconds at
57C, and 2 minutes at 72C (a 2 minute extension was determined to be sufficiently long
so as to yield approximately equal incorporation efficiency for both standard and non
standard nucleotides).
Following PCR, samples were extracted in phenol/chloroform/isoamyl alcohol,
then extracted chloroform/isoamyl alcohol, and precipitated with ammonium acetate and
ethanol. Samples were re-suspended in exonuclease/buffer solution (25 pL final volume:
5 units lambda exonuclease enzyme, 67 mM glycine-KOH (pH 9.4), 2.5 mM MgCl2, 50
pg/mL BSA) and incubated at 37C for 30 minutes. Reactions were terminated by
adding 60 pL formamide stop dye and heating at 80C for 10 minutes. The single-
stranded products were loaded on an 8% PAGE/urea gel and full length ssDNA products
excised. Gel purified samples were then extracted in phenol/chloroform/isoamyl alcohol,
then extracted chloroform/isoamyl alcohol, and precipitated in ammonium acetate and
ethanol. Samples were re-suspended in 50 mM HEPES buffer (pH 7) to a concentration
of 25 30 nM. Each sample (25 pL) was then mixed with equal volume of 2X reaction
buffer (see Table 1 for buffer conditions for each sample), heated and slowly cooled,
exposed to laser (150 pulses, 80-100 mJ/pulse), and stopped with formamide stop dye
following various incubation times (see Table 1 for selection time used for each sample).
This procedure was repeated for each round of selection.
35


40
C18#l (lane 1). It is believed that the complementary primer containing the C18#l
linker did not cause termination of the catalytic strand extension due to incorrect
synthesis of the complementary strand primer (the same linker was used in Cl 8#2, and
this successfully terminated extension).
The size differential between catalytic and complementary strands generated by
asymmetric PCR was then used to isolate the two strands on an 8% PAGE/urea gel.
Figure 3-3 shows the result of PCR in which both strands are labeled using alpha-32P-
CTP, demonstrating that all products are easily resolved. Lane 3 shows the full-length
catalytic strand purified using exonuclease. Lane 4 shows two bands generated via
asymmetric PCR, the 106-nucleotide band corresponding to the catalytic strand and the
121-nucleotide band corresponding to the complementary strand plus a 15-nucleotide tail.
Treatment of the sample in lane 4 with strong base converts the majority of the 106-
nucleotide band to the 79-nucleotide cleavage product. A small amount of the 121-
nucleotide band is converted to the 94-nucleotide band by strong base
treatment, indicating Taq polymerase can occasionally read through the linker of the
complementary strand, thus generating a catalytic strand that is increased in size by the
length of the tail.
As noted in the previous section, the presence of the caged-ribose in the catalytic
strand occasionally causes strand termination, generating a complementary strand 27-
nucleotides shorter than full-length (the caged-ribose is at position 27- in the catalytic
strand). Therefore, asymmetric PCR using a complementary strand primer with a 15-
nucleotide tail generates three bands, a 106-nt band corresponding to the normal sized
catalytic strand, a 121-nt band corresponding to the full-length complementary strand
40


85
^
d*' o*
~ 3>
$? S? X
caged primer plus varying pulses of laser, +NaOH o ,* sn*
J" ¡>T
#w
v+
Figure 3-4. Maximum deprotection of caged-ribose primers occurs with 100
pulses of laser. End-labeled ang + cage primer was exposed to increasing pulses of
laser (308 nM, 50 mJ/pulse) and the amount of deprotection determined by subsequent
cleavage in 0.5 M NaOH for 60 minutes at 80C. Cleavage reaches a plateau of
approximately 80% with 100 pulses of laser. Ang + cage primer exposed to 1360
pulses of laser, but not treated with strong base, showed approximately 1% cleavage.
Unprotected ang + ribose primer was 90-97% cleaved under strong base conditions.
Exonuclease purified full-length ribose-614 was 85-95% cleaved under strong base
conditions. indicates incubation with 0.25M NaOH rather than 0.5M.


A
B
C
D
Figure 4-15. The upper row shows the population distribution following successive rounds of selection for catalysts initially distributed according to (A) linear, m=0.
(B) linear, m=-4x10E-7. (C) exponential. b= 2.5 (D) exponential, b= 25. The lower panel compares the population distribution before selection following the stabilized
population distribution following selection. The population distribution for catalysts following increasing rounds of selections eventually reaches a stabilized-population,
beyond which further rounds of selection do not significantly alter the distribution of catalysts (upper panel). The proportionality of catalysts with half lives 6-fold lower
than St remains nearly identical to their proportionality in the initial population (lower panels). The proximity of the stabilized distribution maxima to the selection time
is related to the initial distribution bias for catalysts with half lives between 6*St and St/6.
On
OO


161
A
60-
50-
*
> 40H
I
o
£ 30-
0)
o
V.
&
20-
HR#7Mg++
a H R#7 no metal
* HR#7Zn++
0 25 50 75 100 125 150 175 200 225
hours
B
80-i

T
100
T H R#7 pool cleavage ¡n
j vari us buffers:
8 HEPES -NaCI -MgCI
a HEPES-NaCI+MgCI
HEPES + NaCI-MgCI
HEPES + NaCI -MgCI
+DEPC
HEPES+NaCI-MgCI
+EDTA
HEPES+NaCI+MgCI
A
A
*1 1 1
150 200 250
hours
Figure 4-8. Cleavage of Selection H pool DNAzymes following round 7 is metal-
independent. (A) Selection H cleavage was examined in 1M NaCI and 50 mM
HEPES pH 7 plus either Zn++, Mg++, or no divalent metal. (B) Selection H cleavage
was examined in 50 mM HEPES buffer pH 7, plus 1 M NaCI, 1 mM MgC12, 1 pL
DEPC per mL buffer, or 3 mM EDTA.


77
with ang + ribose primer, neither 614 56-94nt nor 614 56-106nt catalyzed cleavage
of ang + ribose, suggesting that structure Cl is not correct.
Screening of mutants of 614 revealed that any single mutation of position 55, 65
or 72 greatly reduced catalytic activity. Either of the G55A or G65A mutations have the
potential to disrupt the structure of A4. Each of these mutants was made with the
compensatory mutations predicted by structure A4 (614:G55A/C47T, and
614:G65A/C40T). Neither compensatory mutation improved catalysis over the single
mutant (Figure 3-45), arguing against the validity of structure A4.
The C72T mutant also had reduced catalytic activity; this mutation is predicted to
disrupt a base-pair with G81 in all of the Family A structures. The validity of these
structures was tested by making the compensatory G81A mutation. The double-mutant,
however, cleaved even slower than the single mutant, suggesting against the validity of
any of the Family A structures (Figure 3-45). It is, however, possible this stem-loop
predicted in Family A is in fact part of the correct structure, but the GC base-pair cannot
be replaced with an AT base-pair (either because the strength of the AT base-pair is
lower, or because the original G or C played a catalytic function in addition to its
structural role in the stem loop).
Position 81 of wild-type 614 was therefore converted from a G to a C (named
614:G81C). The G81C mutation has the potential to disrupt the structures of both
Family A and the structure B2, although in different manners. As mentioned above, G81
is predicted to base-pair with C72 in Family A structures; G81 is predicted to base pair
with C43 in structure B2. The 614:G81C mutant demonstrated greatly reduced self
cleavage, allowing us to test both Family A and B2 structures by making the predicted
77


73
ang + ribose chase was added in either 2.5- or 5-fold excess over the enzyme. The
results indicate that cleavage was greatly reduced, but continued even after the addition
of the chase (Figures 3-39 and 40). The reduction was slightly greater with additional
chase. Together these experiments indicate that the rate of E*S disassociation is greater
than the rate of chemical step (k_i kcat(bi)) and the residual cleavage is a function of the
ratio of substrate and chase to enzyme (scenario (d)).
The chase experiments indicate that inter-molecular association (folding) may be
the rate-limiting step or that, once folded, the substrate disassociates from enzyme prior
to cleavage. This was further explored by examining the temperature dependence of
initial cleavage rates. If the chemical step is the rate-limiting step, lowering the
temperature should lower the rate of cleavage, whereas raising the temperature will
increase the rate (until the molecules denatures). However, when folding is rate limiting,
lowering temperature is expected to stabilize folding and thus increase cleavage rate.
Figures 3-41 and 42 shows that lowering incubation temperature increases the initial
r-614 cleavage rate in both cis and trans conditions.
As the chemical step is expected to decrease approximately two-fold for every
10C drop in temperature, the observation that a rate is unchanged by a 20C change in
temperature suggests that a balance is reached between a slower chemical step (kcat(uni) or
kcat(bi)) and a more stabilized association step (ki(uni) or k|(b¡)) at lower temperatures.
The enhancement of the folding/association step at lower temperatures is also
seen with d-614 cleaving r-lib61o2 in trans during single-turnover experiments (Figure
3-42B). Multiple-turnover experiments reveal a rate enhancement at lower temperatures
during the first 20% of cleavage, following which the rate is reduced (Figure 3-38). This
73


37
electrophoresis-based selection, and the use of an ortho-nitrobenzyl-protected ribose
(caged-ribose) which can be quickly converted to the unprotected ribose by laser. An
IVS is performed using these new procedures. Preliminary modeling of the experimental
system is done to determine how various parameters of the experimental system have an
impact on the mathematical model, and how the experimental system can be fine-tuned to
improve modeling accuracy.
Developing Methods for Preparing Single-stranded DNAzymes by
Exonuclease Degradation of 5-Phosphorylated Complementary Strand
Column purification of single-stranded DNAzymes present numerous problems,
as outlined in the introduction. The ability to enzymatically remove the complementary
strand was tested using lambda exonuclease. Lambda exonuclease is a processive 5 to
3 DNA exonuclease specific for 5-phosphorylated DNA. Double stranded 614
DNAzyme was generated with PCR using a 5-phosphorylated complementary strand
primer (complimentary + 5-P) and a non-phosphorylated catalytic strand primer
(either ang + ribose or ang + cage). Samples were internally labeled with alpha- P-
CTP to allow visualization of both catalytic and complementary strands. Each of the two
samples was then treated with or without exonuclease and/or strong base (which cleaves
unprotected ribose) and run on a denaturing PAGE/urea gel. Figure 3-1 shows the results
of this treatment (all figures at are at the end of the chapter).
Both untreated ribose- and caged-614 double stranded PCR products show two
bands, the upper corresponding to the full size product, and the lower band corresponding
roughly to the size of the cleaved product (lane 1 and 5). The amount of the lower band
is significantly greater in the PCR product generated using the caged-ribose primer
37


172
Enzyme folding is particularly problematic for nucleic acid enzymes because
folding requires perpendicular face-to-face interactions (more constrained), and
alternative conformations are relatively common (as structure predictions for 614 reveal).
Nucleic acid-polymer folding (with standard, non-functionalize nucleotides) may be a
slow search for optimal (active) structures, full of dead-end (inactive) traps. This is in
contrast to the funnel-shaped folding landscape predicted for proteins (Ozkan, 2001)
(recognizing that natural proteins may have evolved into regions of sequence space that
have this property). Functionalities added to nucleic acids that improve folding strength
and improve search properties may be critical.
Stanley Miller recently proposed that prebiotic synthesis of nucleic acids is
enhanced in cold environments, and postulated this may be evidence for cold origins of
single-biopolymer life-form (Miyawaka, 2001). To our knowledge, our research is the
first demonstration of better catalysis at temperatures lower than those used during the
selection. In a prebiotic world in which nucleic acid folding may be problematic, a cold
environment may have aided synthesis of monomers and folding of polymers in a nucleic
acid-based life.
The complex nature of 614 inter- and intra-molecular cleavage gives us insight
into how a single-biopolymer hypercycle may have originated. The earliest stage of life
requires that catalysts (specifically a replicase) act on various different substrates.
Studies of the impact of compartition on evolution have demonstrated that some
mechanism must exist to keep selfish molecules from overwhelming critical functions
(Matsuura, 2002). One method of achieving compartition relies on cellular membranes.
It is plausible, however, that before hypercycles were enclosed in membranes, catalytic


114
4 nM r-614 +
2025 nM d-614 +
10 uM chase at
t = 4 hours
v 200 nM r-614 +
3000 nM chase
at t = 1 hour
o 0.5 nM r-614 +
64 nM d-614 +
1280 nM chase
at t = 45 min
o 4 nM r-primer +
2025 nM d-614 +
5 uM chase at
t = 4 hours
o 4 nM r-primer +
2025 nM d-614 +
10 uM chase at
t = 4 hours
0.3 nM r-614
1.1 nM r-614
Figure 3-40. The residual cleavage following the addition of the chase is less than
614 as-cleavage, and is dependent on the amount of chase added. The amount of
cleavage following the addition of chase was re-plotted by subtracting the time at
which the chase was added and the amount of cleavage at the time chase was added
from all time/cleavage data point following the chase. Increasing the amount of chase
added to the r-primer(4 nM)/d-614(2025 nM) condition from 5 pM to 10 pM
decreased the residual r-primer cleavage.


177
leakage parameter and kfastest parameters can be experimentally determined. The
experimentally measurable outcome of selection experiments (progress curve for
cleavage at each round of selection) can be fit to the simulation model in which
the RF and distribution function are varied. Simulations have revealed that model
parameters such as RF and population distribution have a significant impact on
the outcome of selection experiments, suggesting that relatively small range of
parameters should fit the experimental data.
Nucleic Acid Catalysis and the Origins of Life
Given the apparent infrequency of nucleic acid enzymes in random libraries, is it
still reasonable to assume that life originated with nucleic acids as the basis of a single
biopolymer life-form? Based on the catalytic power of proteins, functionalized
nucleotides were expected to dramatically improve both power and frequency of nucleic
acid catalyst in random libraries. To date, this approach has not yet bridged the gap
between what is generated in the lab and what is expected to be sufficient for the genesis
of life. Nonetheless, abundant evidence argues that nucleic acids can in fact catalyze
complex reactions, as they currently do in the spliceosome and ribosome, and as
reconstructions of the RNA world predict (Benner, 1989).
Perhaps nucleic acids are not just limited by the absence of the chemical
functionality that enhance catalysis (as common in proteins), but also by the precise base
pairing that is required of nucleic acids as the genetic system. The standard nucleic acids
are the product of stringent selection for polymers that fold well in a linear fashion but do
not fold into three dimension structures. Catalytic nucleic acids such as the ribosome and


81
suggesting that either a mixture of multiple irans-folded conformers exists, or none of the
predicted structures accurately reflect the true structure of 614.
Structural investigations of 614 are continuing. It is interesting to note that a vast
majority of reported structures for in vitro selected DNAzymes are based on unverified
predictions using the mfold program. The mfold program has been shown to be useful for
providing plausible structures, but our experience with 614 demonstrates that unverified
predictions of DNAzymes should be taken very lightly.
Summary of Results
Table 3-1 (on next page) summarizes the major results of experiments described
in this chapter.
81



PAGE 1

6($5&+,1* 02/(&8/$5 /$1'6&$3(6 )25 7+( (92/87,21 2) 35,0$/ &$7$/<676 ,1 9,752 6(/(&7,21 2) '1$%$6(' 5,%218&/($6(6 %\ 0$77+(: $ &$55,*$1 ',66(57$7,21 35(6(17(' 72 7+( *5$'8$7( 6&+22/ 2) 7+( 81,9(56,7< 2) )/25,'$ ,1 3$57,$/ )8/),//0(17 2) 7+( 5(48,5(0(176 )25 7+( '(*5(( 2) '2&725 2) 3+,/2623+< 81,9(56,7< 2) )/25,'$

PAGE 2

7KLV ZRUN LV GHGLFDWHG WR P\ VRQ &KULVWLDQ

PAGE 3

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fV SDWLHQFH DQG HQWKXVLDVP IRU VFLHQFH WKLV UHVHDUFK HQGHDYRU ZRXOG QRW KDYH EHHQ SRVVLEOH LLL

PAGE 4

7$%/( 2) &217(176 $&.12:/('*(0(176 LLL $%%5(9,$7,216 YLL $%675$&7 L[ &+$37(5 SDJH ,1752'8&7,21 5HTXLUHPHQWV RI D f/LYLQJf 6\VWHP &KHPLFDO &KDOOHQJHV WR WKH 3UHELRWLF 6\QWKHVLV RI f/LIHf 3URJUHVV 7RZDUGV 8QGHUVWDQGLQJ WKH *HQHVLV RI /LIH 'XDO )XQFWLRQ %LRSRO\PHUV 3URJUHVV 7RZDUGV 8QGHUVWDQGLQJ WKH *HQHVLV RI /LIH 3UHELRWLF 6\QWKHVLV RI 3RWHQWLDOO\ 8VHIXO 0RQRPHUV DQG 3RO\PHUV 7KH 1H[W &KDOOHQJH 2EWDLQLQJ &DWDO\WLF )XQFWLRQ IURP 5DQGRP /LEUDULHV ([SHULPHQWDO 'HVLJQ 0$7(5,$/6 $1' 0(7+2'6 3UHSDUDWLRQ RI 3UHFXUVRU '1$\PHV YLD 3&5 3UHSDUDWLRQ RI 6LQJOH6WUDQGHG '1$]\PHV f(QG /DEHOLQJ RI '1$ '1$]\PH .LQHWLF $VVD\V &ORQLQJ DQG 6HTXHQFLQJ '1$]\PHV ,Q 9LWUR 6HOHFWLRQ 5(68/76 2) $1$/<6,6 5HVHDUFK 2EMHFWLYHV 'HYHORSLQJ 0HWKRGV IRU 3UHSDULQJ 6LQJOHVWUDQGHG '1$]\PHV E\ ([RQXFOHDVH 'HJUDGDWLRQ RI f3KRVSKRU\ODWHG &RPSOHPHQWDU\ 6WUDQG $V\PPHWULF 3&5 ZLWK f7DLOVf )ROORZHG E\ *HO 3XULILFDWLRQ '1$]\PH ZLWK XQFDJHG DQG FDJHGULERVH $FKLHYLQJ PD[LPDO GHSURWHFWLRQ RI FDJHG ULERVH .LQHWLF SURILOH RI ZLWK XQFDJHG DQG FDJHGULERVH ,9

PAGE 5

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f+ DQJ ULERVHf 6HTXHQFH (YROXWLRQ 'XULQJ 96 .LQHWLF $QDO\VLV RI 5RXQG &ORQHV 6LPXODWLRQV RI ,Q 9LWUR 6HOHFWLRQ ([SHULPHQWV 6LPXODWLQJ ,Q 9LWUR 6HOHFWLRQV 8VLQJ 6LPXODWLRQV WR 'HWHUPLQH WKH ,PSDFW RI WKH /HDNDJH /f 3DUDPHWHU RQ 96 ([DPLQLQJ WKH ,PSDFW RI WKH 6HOHFWLRQ 7LPH 6Wf ([DPLQLQJ WKH ,PSDFW RI WKH 5DWH RI WKH )DVWHVW &DWDO\VWV NIDVWHVWf ([DPLQLQJ WKH ,PSDFW RI WKH 5HDFWLYH )UDFWLRQ &RPSDULQJ /LQHDUO\ DQG ([SRQHQWLDOO\ 'LVWULEXWHG ,QLWLDO 3RSXODWLRQV 'HFLSKHULQJ WKH 'LVWULEXWLRQ )XQFWLRQ IRU &DWDO\VWV 6XPPDU\ RI 5HVXOWV 6800$5< &DWDO\WLF %HKDYLRU RI '1$]\PHV IURP ,96 LV 0RUH &RPSOH[ WKHQ 3UHYLRXVO\ $VVXPHG Y

PAGE 6

&DJHGULERVH LV 8VHIXO IRU .LQHWLFV %XW ,QFUHDVHV /HDNDJH 7RR 0XFK IRU ,96 ,96 ,VRODWHV 6HTXHQFH 9DULDQWV RI ZLWK :LGH 5DQJH RI &DWDO\WLF 3RZHU 6LPXODWLRQV $OORZ (VWLPDWLRQ RI &DWDO\VW 'LVWULEXWLRQ 1XFOHLF $FLG &DWDO\VLV DQG WKH 2ULJLQV RI /LIH 5()(5(1&(6 %,2*5$3+,&$/ 6.(7&+ YL

PAGE 7

$%%5(9,$7,216 96 LQ YLWUR VHOHFWLRQ E\D ELOOLRQ \HDUV DJR 6/ 6L]H RI LQLWLDO OLEUDU\ 5) 5HDFWLYH )UDFWLRQ 15) 1RQUHDFWLYH )UDFWLRQ 15) 6/r? 5)f Nf ,QWULQVLF UDWH RI PROHFXOH HDFK PHPEHU RI WKH LQLWLDO OLEUDU\ Q LV DVVLJQHG DQ LQWULQVLF UDWH NQ ZKLFK UHODWHV WR LWV DELOLW\ WR EH FOHDYHG XQGHU VHOHFWLRQ FRQGLWLRQV 7KH LQWULQVLF UDWH LV VHTXHQFHGHSHQGHQW DQG WKHUHIRUH XQFKDQJLQJ $OLIHQ 7KH LQWULQVLF UDWH RI PROHFXOH fQf Nff H[SUHVVHG LQ WHUPV RI KDOIOLIH $OLIHf OQ NQrVLRZHVW 7KH UDWH FRQVWDQW IRU WKH VORZHVW FDWDO\VWV LQ WKH OLEUDU\ NIDVWHVW 7KH UDWH FRQVWDQW RI WKH IDVWHVW PROHFXOH LQ WKH OLEUDU\ 6W 6HOHFWLRQ WLPH WKH DPRXQW RI WLPH DOORZHG IRU D PROHFXOH WR FOHDYH LWVHOI DQG WKHUHIRUH DFKLHYH fVXUYLYDOf P RU E 'LVWULEXWLRQ IXQFWLRQ IRU UHDFWLYH PROHFXOHV 7ZR IXQFWLRQV KDYH EHHQ WHVWHG WKXV IDU D OLQHDU IXQFWLRQ ZLWK VORSH P DQG DQ H[SRQHQWLDO IXQFWLRQ ZLWK GHFD\ UDWH E / )HDNDJH WKH IUDFWLRQ RI WKH SRRO WKDW fVXUYLYHVf D URXQG RI VHOHFWLRQ LQGHSHQGHQW RI 6W DQG LQWULQVLF UDWH FRQVWDQW 9OO

PAGE 8

FDJHGULERVH D ULERVHDGHQRVLQH ZLWK WKH f2+ UHSODFHG ZLWK DQ RUWKRQLWUREHQ]\O SURWHFWLQJ JURXS USULPHU fDQJ ULERVHf QXFOHRWLGH VXEVWUDWH FRQWDLQLQJ ULERVH U; ROLJRQXFOHRWLGH f;f ZLWK ULERVHDGHQRVLQH G; ROLJRQXFOHRWLGH f;f ZLWK GHR[\DGHQRVLQH $ 'HR[\DGHQRVLQH WULSKRVSKDWH 'HR[\JXDQRVLQH WULSKRVSKDWH & 'HR[\F\WLGLQH WULSKRVSKDWH 7 'HR[\WK\PLGLQH WULSKRVSKDWH (EDVH DPLQRDOO\OffGHR[\XULGLQH WULSKRVSKDWH VHH VWUXFWXUH EHORZf 2 NFDWXQLf NOXQLf NOXQLf NFDWELf N OELf NOELf FKHPLFDO VWHS IRU XQLPROHFXODU FDWDO\VLV UDWH RI IROGLQJ IRU XQLPROHFXODU FDWDO\VLV UDWH RI XQIROGLQJ IRU XQLPROHFXODU FDWDO\VLV FKHPLFDO VWHS IRU ELPROHFXODU FDWDO\VLV UDWH RI DVVRFLDWLRQ IRU ELPROHFXODU FDWDO\VLV UDWH RI GLVDVVRFLDWLRQ IRU ELPROHFXODU FDWDO\VLV %LPROHFXODU .LQHWLF 6FKHPH ) T NEL/ )R NFDWEr ( 3 ( U RU G 6 U UOLEOR RU USULPHU NELf 8QLPROHFXODU .LQHWLF 6FKHPH NLXQLf NFDWXQLf XQILGHM I_MH&M 3URGXFWV NLXQLf YLLL

PAGE 9

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

PAGE 10

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f $OWKRXJK VHOHFWHG ZLWK D SURWRFRO EHOLHYHG WR HQULFK IRU 0JGHSHQGHQW VHOIFOHDYLQJ FDWDO\VWV FOHDYHV ERWK LQ FLV DQG WUDQV LQGHSHQGHQWO\ RI 0JA 7KH UDWH RI WUDQVFOHDYDJH DW VXEVWUDWH VDWXUDWLRQ LV IROG KLJKHU WKDQ WKH UDWH RI FVFOHDYDJH )XUWKHU ERWK FLV DQG WUDQVFOHDYDJH UDWHV DUH HQKDQFHG DW WHPSHUDWXUHV EHORZ r& WKH WHPSHUDWXUH DW ZKLFK WKH VHOHFWLRQV ZHUH SHUIRUPHGf VXJJHVWLQJ WKDW D EDODQFH LV UHDFKHG EHWZHHQ D VORZHU FKHPLFDO VWHS NAXQLf RU NAELff DQG D PRUH VWDELOL]HG DVVRFLDWLRQ VWHS NLXQLf RU NLELff DW ORZHU WHPSHUDWXUHV 0XWDJHQHVLV RI DQG LWV VXEVWUDWHV ZDV SHUIRUPHG WR OHDUQ DERXW WKH VWUXFWXUH RI FOHDYLQJ LQ FLV DQG WUDQV &RPSHQVDWRU\ PXWDWLRQV UHYHDOHG WKDW RQH RI WKH LQWHUPROHFXODU KHOL[HV IRUPHG EHWZHHQ DQG LWV VXEVWUDWH LV LGHQWLFDO WR DQ LQWUDPROHFXODU KHOL[ IRUPHG LQ F$DFWLQJ 7KH UDQVIROGHG KRZHYHU KDV DQ DGGLWLRQDO EDVHSDLU KHOL[ QRW IRXQG LQ F$IROGHG DQG WKLV PD\ SDUWLDOO\ H[SODLQ WKH IDVWHU UDWH RI WUDZVFOHDYDJH DW VDWXUDWLRQ :H DOVR DGGUHVVHG WKH SRWHQWLDO ORVV RI FDWDO\VWV E\ fSUHPDWXUH FDWDO\VLVf GXULQJ WKH SUHSDUDWLRQ RI '1$]\PH SRROV LQ WKH ,96 3XULILFDWLRQ RI VLQJOHVWUDQGHG '1$]\PHV ZDV DFFRPSOLVKHG XVLQJ DV\PPHWULF 3&5 RU H[RQXFOHDVH GHJUDGDWLRQ [

PAGE 11

HOLPLQDWLQJ WKH QHHG IRU FROXPQEDVHG PHWKRGV IRU SHUIRUPLQJ ,96 :H DOVR H[DPLQHG WKH XWLOLW\ RI D SURWHFWHG ULERVH FDJHGULERVHf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

PAGE 12

&+$37(5 ,1752'8&7,21 5HTXLUHPHQWV RI D f/LYLQJf 6\VWHP ,Q D FRPPLWWHH HPSDQHOHG E\ 1$6$ GHILQHG OLIH DV fD VHOIVXVWDLQLQJ FKHPLFDO V\VWHP FDSDEOH RI XQGHUJRLQJ 'DUZLQLDQ HYROXWLRQf -R\FH f f1RQn OLYLQJf HQWLWLHV VXFK DV QDWXUDO SKHQRPHQD OLNH ILUH RU FU\VWDO IRUPDWLRQ RIWHQ KDYH SK\VLFDO SURSHUWLHV FRPPRQO\ DVVRFLDWHG ZLWK OLIHIRUPV VXFK DV fHQHUJ\ FRQVXPSWLRQf fRUGHUHG VWUXFWXUHVf DQG fJURZWK DQG UHSOLFDWLRQf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f $V DOO OLIH UHTXLUHV REWDLQLQJ SUHFXUVRUV DQG HQHUJ\ IURP DQ HQYLURQPHQW WKH UHTXLUHPHQW WKDW OLIH EH VHOIVXVWDLQLQJ LV DOZD\V UHODWLYH WR D SDUWLFXODU HQYLURQPHQW 7KH f1$6$ GHILQLWLRQf DOVR OLPLWV OLIH WR FKHPLFDO V\VWHPV UHYHDOLQJ D ELDV WRZDUGV OLIH DV ZH NQRZ LW DQG LQ ZKLFK ZH DUH PRVW LQWHUHVWHG (YROYLQJ V\VWHPV VXFK DV PHPHV

PAGE 13

FXOWXUDO LGHDVf DQG LQ VLOLFR HYROYLQJ V\VWHPV VRIWZDUHf DUH QRW IRUPDOO\ UHFRJQL]HG DV OLYLQJ EHFDXVH WKH\ DUH QRW FKHPLFDO 7KH LQIRUPDWLRQVWRULQJ VWUXFWXUH IRU PHPHV LV SRRUO\ XQGHUVWRRG DQG WKLV PD\ FRQWULEXWH WR WKH ELDVHV DJDLQVW FRQVLGHULQJ WKLV HYROYLQJ V\VWHP D OLIHIRUP )RU LQ VLOLFR HYROYLQJ V\VWHPV WKH UDQJH RI SK\VLFDO SURSHUWLHV fIXQFWLRQ VSDFHff LV DUELWUDULO\ GHILQHG E\ WKH SURJUDPPHU DQG WKHUHIRUH QRW RSHQHQGHG OLNH D FKHPLFDO V\VWHP ,W FDQ EH DUJXHG WKDW WKH fIXQFWLRQ VSDFHf IRU D FKHPLFDO HYROYLQJ V\VWHP LV QRW LQ IDFW RSHQHQGHG EXW UDWKHU OLPLWHG E\ SK\VLFDO ODZV 8OWLPDWHO\ ERWK PHPHV DQG LQ VLOLFR HYROYLQJ V\VWHPV FDQ EH FRQVLGHUHG WKH SURGXFW RI DQG WKHUHIRUH QRW LQGHSHQGHQW RI D OLYLQJ V\VWHP 2QH PD\ KRZHYHU FRQFHLYH RI OLIH DV DQ\ V\VWHP FDSDEOH RI 'DUZLQLDQ HYROXWLRQ 7KH f1$6$ GHILQLWLRQf RI OLIH PLJKW WKHQ EH JHQHUDOL]HG WR fDQ\ LQIRUPDWLRQDOFRQWDLQLQJ V\VWHP WKDW XQGHUJRHV 'DUZLQLDQ HYROXWLRQf $OWKRXJK WKH VHFRQG ODZ RI WKHUPRG\QDPLFV GLFWDWHV WKDW WKH XQLYHUVH SURJUHVV WR D VWDWH RI JUHDWHU GLVRUGHU WKLV GRHV QRW QHFHVVDULO\ UHVXOW LQ WKH ORVV RI DOO LQIRUPDWLRQ :LWKLQ D V\VWHP RUGHU LQIRUPDWLRQf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fFRUUHFWf YHUVLRQf RU GXSOLFDWLQJ WKH

PAGE 14

LQIRUPDWLRQ UHTXLULQJ GXSOLFDWLRQ HIILFLHQF\DFFXUDF\ EH JUHDWHU WKDQ WKH UDWH RI GHJUDGDWLRQf :KHQ WKH SK\VLFDO SURSHUWLHV RI D V\VWHP DUH GHWHUPLQHG E\ WKH LQIRUPDWLRQ LW FRQWDLQV LW IROORZV WKDW WKH SK\VLFDO SURSHUWLHV RI WKH V\VWHP FDQ FKDQJH DFFRUGLQJ WR FKDQJHV LQ WKH HQFRGHG LQIRUPDWLRQ 6XFK D V\VWHP PD\ QHYHU VHWWOH LQWR HTXLOLEULXP DV SK\VLFDO V\VWHPV DUH H[SHFWHG WR HYHQWXDOO\ UHDFKf ,QVWHDG LPSHUIHFW LQIRUPDWLRQ GXSOLFDWLRQ DQG UHSDLU FKDQJHV WKH SK\VLFDO SURSHUWLHV RI D V\VWHP 7KLV HYHQWXDOO\ UHVXOWV LQ PXOWLSOH YDULDQWV RI DQ LQIRUPDWLRQFRQWDLQLQJ V\VWHP HDFK ZLWK YDULDEOH SK\VLFDO SURSHUWLHV :KHQ WZR VXFK V\VWHPV XWLOL]H WKH VDPH UHVRXUFHV HQHUJ\ ULFK PROHFXOHV HOHPHQWDO SUHFXUVRUV HWFf WR SUHYHQW LQIRUPDWLRQ GHFD\ FRPSHWLWLRQ H[LVWV DPRQJ V\VWHPV IRU PRUH HIILFLHQW XVH RI WKH UHVRXUFH 'HVSLWH FRPSHWLWLRQ DPRQJ V\VWHPV IRU FRPPRQ UHVRXUFHV WKH XOWLPDWH VXUYLYDO SUHVVXUH LV DJDLQVW WKH SURJUHVVLRQ WRZDUGV LQIRUPDWLRQ ORVV $ V\VWHP FDQ WKHUHIRUH SHUVLVW E\ HLWKHU RXWFRPSHWLQJ ULYDOV IRU D FRPPRQ UHVRXUFH RU E\ RFFXS\LQJ D QHZ QLFKH XWLOL]LQJ D GLIIHUHQW UHVRXUFHf *LYHQ WKH FRQVWDQW SURJUHVVLYH IRUFH WRZDUGV LQIRUPDWLRQ GHFD\ WKHUH H[LVWV VHOHFWLRQ DQG HQULFKPHQW IRU V\VWHPV WKDW HQKDQFH WKH VXUYLYDO RI WKHLU LQIRUPDWLRQ $V PXOWLSOH V\VWHPV FRPSHWH IRU UHVRXUFHV LQIRUPDWLRQ V\VWHPV OLQNHG WR UHSOLFDWLRQHQKDQFLQJ IXQFWLRQV ZLOO SUHGRPLQDWH )URP WKLV RQH FDQ VHH WKDW 'DUZLQLDQ HYROXWLRQ ZKLFK LV WKH IXQGDPHQWDO WUDLW RI OLIHf LV WKH GLUHFW FRQVHTXHQFH RI SK\VLFDO ODZV DSSOLHG WR DQ LQIRUPDWLRQFRQWDLQLQJ V\VWHP )RU WKLV UHDVRQ VRPH KDYH DUJXHG WKDW RQFH D PLQLPDO V\VWHP K\SHUF\FOHf FDSDEOH RI 'DUZLQLDQ HYROXWLRQ H[LVWV LW ZLOO JHQHUDWH KLJKHU RUGHUV RI FRPSOH[LW\

PAGE 15

VSRQWDQHRXVO\ DQG LQHYLWDEO\ DV ORQJ DV PLQLPDO UHVRXUFHV DUH DYDLODEOH DQG D PDMRU JOREDO HQYLURQPHQWDO FKDQJH GRHV QRW EULQJ DERXW H[WLQFWLRQ .DXIPDQ f 7ZR VSHFLILF IXQFWLRQDO DWWULEXWHV IROORZ IURP WKLV GHILQLWLRQ RI OLIH WKH DELOLW\ WR FDUU\ KHULWDEOH LQIRUPDWLRQ JHQHWLFVf ZKLFK LV OLQNHG WR VHOHFWDEOH WUDLWV IXQFWLRQf 6HOHFWLRQ LV WKXV DFWLQJ RQ WZR OHYHOV VLPXOWDQHRXVO\ WKH V\VWHP RI VWRULQJ LQIRUPDWLRQ DQG WKH SK\VLFDO SURSHUWLHV GHULYDWLYH RI WKLV LQIRUPDWLRQ %RWK fLQIRUPDWLRQVWRUDJHf DQG fIXQFWLRQf FDQ EH DFFRPSOLVKHG XVLQJ D P\ULDG RI FKHPLFDO DQG SK\VLFDO IRUPDWV +RZHYHU WKH UHTXLUHPHQWV RI LQIRUPDWLRQ VWRUDJH DQG GXSOLFDWLRQ XVLQJ FKHPLFDO SRO\PHUV DUH LQ PDQ\ ZD\V FRQWUDGLFWRU\ WR WKH UHTXLUHPHQWV IRU UHSOLFDWLRQn HQKDQFLQJ IXQFWLRQV %HQQHU DQG 6ZLW]HU f f &KDQJHV LQ WKH LQIRUPDWLRQ PXVW QRW LPSHGH LWV DELOLW\ WR EH GXSOLFDWHG ZKHUHDV FKDQJHV LQ LQIRUPDWLRQ PXVW DOORZ FKDQJHV LQ IXQFWLRQ DQG WKXV SK\VLFDO SURSHUWLHV 7KH VHOHFWDEOH WUDLWV IXQFWLRQf RI D SRO\PHU DUH GHWHUPLQHG E\ WKH WKUHH GLPHQVLRQDO VWUXFWXUH IROGf RI WKH SRO\PHU DQG WKH SRVLWLRQLQJ RI FKHPLFDO PRLHWLHV WKXV D IXQFWLRQDO SRO\PHU PXVW IROG WR JHQHUDWH VHOHFWDEOH FKHPLFDO WUDLWV ,Q FRQWUDVW DQ LQIRUPDWLRQ VWRUDJH V\VWHP VKRXOG QRW IROG HDVLO\ DV WKLV LV H[SHFWHG WR LPSHGH JHQHUDOL]HG PHFKDQLVPV IRU GXSOLFDWLRQ f $Q LQIRUPDWLRQ VWRUDJH V\VWHP VKRXOG KDYH IHZ VXEXQLWV WR HQVXUH JUHDWHU ILGHOLW\ RI LQIRUPDWLRQ WUDQVIHU ZKHUHDV D IXQFWLRQDO SRO\PHU VKRXOG KDYH PDQ\ VXEXQLWV ZLWK GLYHUVH IXQFWLRQDOLWLHV WR HQKDQFH WKH FKHPLFDO SRWHQWLDO RI WKH SRO\PHU f )RU D V\VWHP WR H[SORUH QHZ IXQFWLRQ WKURXJK HYROXWLRQ FKHPLFDO SURSHUWLHV PXVW EH DEOH WR FKDQJH UDSLGO\ ZLWK IHZ FKDQJHVf DQG DEUXSWO\ 6XFK DEUXSW FKDQJHV LQ WKH FKHPLFDO SURSHUWLHV RI D PROHFXOH IROORZLQJ PXWDWLRQ DUH FRQWUDGLFWRU\ WR WKH UHTXLUHPHQW WKDW WKH LQIRUPDWLRQ FDUU\LQJ V\VWHP PDLQWDLQ LWV DELOLW\ WR EH UHSOLFDWHG LQ VSLWH RI VLJQLILFDQW FKDQJH PXWDWLRQf &KHPLFDO &KDOOHQJHV WR WKH 3UHELRWLF 6\QWKHVLV RI f/LIHf )RU D FKHPLFDOO\ EDVHG OLYLQJ V\VWHP ZKLFK LV WKH RQO\ XQGLVSXWHG IRUP RI OLIH ZH NQRZf LW LV GLIILFXOW WR LPDJLQH D PHFKDQLVP IRU VWRULQJ LQIRUPDWLRQ DQG JHQHUDWLQJ SKHQRW\SHV RWKHU WKDQ WKURXJK WKH VWUXFWXUH RI SRO\PHUV *LYHQ WKLV UHVWULFWLRQ D QXPEHU

PAGE 16

RI UHTXLUHPHQWV PXVW EH PHW IRU WKH SUHELRWLF SUH'DUZLQLDQf JHQHVLV RI D OLYLQJ HYROYLQJf V\VWHP Df 3UHELRWLF V\QWKHVLV RI WKH PRQRPHUV Ef 3UHELRWLF IRUPDWLRQ RI SRO\PHUV IURP PRQRPHUV Ff 2EWDLQLQJ FDWDO\VWV IURP D OLEUDU\ RI UDQGRP SRO\PHUV Gf *HQHWLFV REWDLQLQJ D SRO\PHU WKDW DFWV DV D KHULWDEOH FDUULHU RI LQIRUPDWLRQ IRU WKH XVHIXO FDWDO\VWV Hf &RPSDUWLWLRQ OLQNLQJ WKH LQIRUPDWLRQ VWRULQJ V\VWHP JHQHWLFVf WR WKH FDWDO\WLF IXQFWLRQVf LW HQFRGHV ,Q PRGHP ELRORJ\ WKH FRQWUDGLFWLRQ EHWZHHQ WKH UHTXLUHPHQWV IRU D JHQHWLF ELRSRO\PHU DQG D FDWDO\WLF ELRSRO\PHU VHH DERYHf LV DPHOLRUDWHG E\ PRVWO\f VHSDUDWLQJ WKH IXQFWLRQV RI JHQHWLFV DQG FDWDO\VLV LQWR WZR GLIIHUHQW ELRSRO\PHUV 1XFOHLF DFLG LV WKH LQIRUPDWLRQ VWRULQJ ELRSRO\PHU ZKLFK LQ WXUQ LV XVHG WR SURGXFH SURWHLQV ZKLFK JHQHUDWH PRVW RI WKH SK\VLFDO SURSHUWLHV WKDW DUH WKH VXEMHFW RI QDWXUDO VHOHFWLRQ $OWKRXJK WKH GLYLVLRQ RI ODERU EHWZHHQ WZR GLVWLQFW ELRSRO\PHUV UHVROYHV WKH FRQWUDGLFWRU\ GHPDQGV RI D OLIHIRUP LW FUHDWHV D VLJQLILFDQW SDUDGR[ SHUWDLQLQJ WR WKH RULJLQV RI OLIH ,Q OLJKW RI WKH FKDOOHQJHV Df DQG Ef WR WKH SUHELRWLF FUHDWLRQ RI OLIH LW LV GLIILFXOW WR LPDJLQH D QRQELRORJLFDO PHFKDQLVP IRU JHQHUDWLQJ HLWKHU RI WKH FRQWHPSRUDU\ ELRn SRO\PHU V\VWHPV QXFOHLF DFLG RU SURWHLQf ,W LV DVWURQRPLFDOO\ PRUH LPSUREDEOH WKDW ERWK ELRSRO\PHU V\VWHPV ZRXOG DULVH VLPXOWDQHRXVO\ DQG VSRQWDQHRXVO\ DQG IXUWKHUPRUH LQ D PDQQHU WKDW ZRXOG DOORZ WKHLU PXWXDO LQWHJUDWLRQFRPSDWLELOLW\ ,Q WKH V WKHRUHWLFDO FRQVLGHUDWLRQ RI WKLV GLOHPPD OHG WR WKH K\SRWKHVLV WKDW OLIH PXVW KDYH RULJLQDWHG DV D VLQJOH ELRSRO\PHU V\VWHPV 5LFK f 7KLV SURSRVDO UHTXLUHV RQO\ RQH VXFFHVVIXO VROXWLRQ WR WKH FKDOOHQJH RI REWDLQLQJ XVHIXO PRQRPHUV DQG SRO\PHUV )XUWKHUPRUH IRU D VLQJOH JHQHHQ]\PH V\VWHP WKH FKDOOHQJH RI FRPSDUWLWLRQ LV UHVROYHG EHFDXVH WKH WZR

PAGE 17

IXQFWLRQDO UHTXLUHPHQWV DUH KHOG VLPXOWDQHRXVO\ LQ D VLQJOH PROHFXOH WKH FKDOOHQJH RI FRPSDUWLWLRQ UHPDLQV ZKHQ WKH V\VWHP XQGHU VHOHFWLRQ FRQWDLQV PXOWLSOH JHQHHQ]\PHVf 7KLV VROXWLRQ KRZHYHU FUHDWHV DQRWKHU FKDOOHQJH 6XFK D ELRSRO\PHU PXVW PHHW WKH FRQWUDGLFWRU\ FKDOOHQJHV RI FfFDWDO\VLV DQG Gf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f 51$ LV D NH\ FRPSRQHQW RI WKH VSOLFHRVRPH ULERVRPH DQG 51$VH3 51$ VHUYHV DV WKH SULPHU IRU '1$ V\QWKHVLV '1$ LV V\QWKHVL]HG E\ SURWHLQEDVHG UHGXFWLRQ RI 51$ D FKHPLFDO WUDQVIRUPDWLRQ WKDW LV TXLWH GLIILFXOW DQG XQOLNHO\ WR RFFXU YLD SUHELRWLF PHFKDQLVPVf -R\FH %HQQHU f 7KH XVH RI 51$ IRU FKHPLFDO SXUSRVHV IRU ZKLFK LW LV QRW LQWULQVLFDOO\ VXLWHG SDUWLFXODUO\ LQ WKH PRGHP HQYLURQPHQW ZKHUH FKHPLFDOO\ EHWWHUVXLWHG PROHFXOHV FRXOG EH PDGH SURWHLQVf DUJXHV DJDLQVW FRQYHUJHQW HYROXWLRQ DQG LQVWHDG VXJJHVWV WKDW WKHVH 51$

PAGE 18

YHVWLJHV DUH fPROHFXODU IRVVLOVf OHIWRYHU IURP D WLPH LQ KLVWRU\ ZKHQ 51$ ZDV WKH VROH ELRSRO\PHU DV RSSRVHG WR FRQYHUJHQW HYROXWLRQf *LOEHUW f 7KHVH FRQVLGHUDWLRQV RIIHU VXSSRUW IRU WKH SUHYLRXV WKHRUHWLFDOO\EDVHG SURSRVLWLRQ WKDW PRGHP OLIHIRUPV SDVVHG WKURXJK D VLQJOH ELRSRO\PHU 51$EDVHG OLIHIRUP D SHULRG RI WLPH NQRZQ DV WKH f51$ ZRUOGf :RHVH &ULFN 2UJHO f 5HFRQVWUXFWLRQV RI WKH 51$ ZRUOG LQGLFDWH WKDW 51$ OLIHIRUPV SRVVHVVHG TXLWH FRPSOH[ PHWDEROLVP PROHFXODU IRVVLOV LQGLFDWH WKDW WKH 51$ ZRUOG GHYHORSHG WKH 51$ FRIDFWRUV $73 FRHQ]\PH $ 6 DGHQRV\OPHWKLRQLQH DQG 1$'+ OLW IROORZV WKDW WKH 51$ ZRUOG QHHGHG WKHVH SUHVXPDEO\ IRU SKRVSKRU\ODWLRQV &ODVLVHQ FRQGHQVDWLRQV PHWK\O WUDQVIHUV DQG R[LGDWLRQUHGXFWLRQ UHDFWLRQV %HQQHU f 7KH GLVFRYHU\ WKDW PRGHP OLIH FRQWDLQV 51$ WKDW DFWV DV D FDWDO\VW GHPRQVWUDWHG WKDW 51$ LV LQ IDFW D VLQJOH ELRSRO\PHU FDSDEOH RI ERWK JHQHWLFV DQG FDWDO\VLV WKXV JLYLQJ H[SHULPHQWDO VXSSRUW WR WKH SURSRVLWLRQ WKDW PRGHP WZRELRSRO\PHU OLIH LV GHVFHQGHQW IURP D VLQJOHELRSRO\PHU 51$f OLIHIRUP 8VKHU &HFK *HXUULHU7DNDGD f 7KH H[LVWHQFH RI DQ f51$ ZRUOGf GRHV QRW QHFHVVLWDWH KRZHYHU WKDW OLIH DFWXDOO\ RULJLQDWHG ZLWK 51$ ,QGHHG DSSO\LQJ FKHPLFDO NQRZOHGJH WR fZRUN IRUZDUGf IURP SUHELRWLF HDUWK WR PRGHP OLIH VXJJHVWV WKDW ULERQXFOHLF DFLGV PD\ KDYH EHHQ D YHU\ SRRU FDQGLGDWH DV WKH SULPRUGLDO PRQRPHU LQ WKH RULJLQV RI OLIH ,QVWHDG WKH 51$ ZRUOG PD\ KDYH VHUYHG DV D WUDQVLWRU\ OLIHIRUP EHWZHHQ WKH RULJLQDO VLQJOHSRO\PHU OLIHIRUP DQG WRGD\fV UREXVW WZRELRSRO\PHU OLIHIRUP :H QRZ WXUQ RXU DWWHQWLRQ WR LQVLJKWV JDLQHG E\ fZRUNLQJ IRUZDUGf IURP EDVLF FKHPLFDO SULQFLSOHV

PAGE 19

3URJUHVV 7RZDUGV 8QGHUVWDQGLQJ WKH *HQHVLV RI /LIH 3UHELRWLF 6\QWKHVLV RI 3RWHQWLDOO\ 8VHIXO 0RQRPHUV DQG 3RO\PHUV 7KH FKDOOHQJH RI QRQELRORJLFDO V\QWKHVLV RI XVHIXO PRQRPHUV DQG SRO\PHUV DUJXHV DJDLQVW WKH VLPXOWDQHRXV FUHDWLRQ RI WZR LQWHUUHODWHG SRO\PHULF V\VWHPV EXW WKH FUHDWLRQ RI HYHQ D VLQJOH SRO\PHULF V\VWHP IRU OLIH FDSDEOH RI JHQHWLFV DQG FDWDO\VLVf VWLOO PXVW PHHW WKH QHHG IRU SUHELRWLF V\QWKHVLV RI PRQRPHUV DQG SRO\PHUV .QRZLQJ WKDW FKHPLFDOO\ EDVHG OLIH GRHV H[LVW ZH PXVW DVVXPH WKDW PRQRPHUV DQG SRO\PHUV GLG IRUP LQ WKH SUHELRWLF ZRUOG DV UHVXOW RI WKH EDVLF SULQFLSOHV RI FKHPLVWU\ ,W KDV QRW KRZHYHU EHHQ HDV\ WR GHGXFH KRZ WKLV PLJKW RFFXU ,Q OLJKW RI WKH FKDOOHQJHV WR WKH SUHELRWLF JHQHUDWLRQ RI OLIH ZKDW LV VR VXUSULVLQJ DERXW WKH RULJLQV RI OLIH RWKHU WKDQ WKDW LW KDSSHQHG DW DOOf LV WKDW LW DSSHDUHG UHODWLYHO\ UDSLGO\ 7KH HDUWK ZDV IRUPHG ELOOLRQ \HDUV DJR E\Df 7KH ILUVW a ELOOLRQ \HDUV RI HDUWKfV OLIH SULRU WR WKH FRROLQJ RI WKH HDUWKfV FUXVW DQG WKH VXEVLGLQJ RI VHYHUH PHWHRULWHV ERPEDUGPHQW ZDV LQKRVSLWDEOH WR WKH IRUPDWLRQ RI DQ\ XVHIXO PRQRPHUV \HW HYLGHQFH RI OLIH LQ WKH IRVVLO UHFRUG LV DSSDUHQW DV HDUO\ DV E\D :DOWHU f 7KLV SURYLGHV D YHU\ VKRUW WLPH IRU WKH JHQHUDWLRQ RI PRQRPHUV SRO\PHUV DQG VXEVHTXHQW JHQHVLV RI OLIH ,Q RUGHU WR DSSO\ FKHPLFDO SULQFLSOHV WR JDLQ LQVLJKW LQWR WKH SUHELRWLF FUHDWLRQ RI PRQRPHUV WKH SUHELRWLF HQYLURQPHQW RI WKH HDUWK DSSUR[LPDWHO\ ELOOLRQ \HDUV DJRf PXVW EH XQGHUVWRRG $OWKRXJK GHILQLWLYH LQIRUPDWLRQ DERXW WKH HDUWKfV HDUO\ HQYLURQPHQW DQG FRPSRVLWLRQ LV VORZ FRPLQJ LW LV ZLGHO\ DVVXPHG WKDW +2 &2 + DQG &2 H[LVWHG ,Q D JURXQGEUHDNLQJ H[SHULPHQW 6WDQOH\ 0LOOHU GHPRQVWUDWHG WKDW FRPELQLQJ WKHVH EDVLF HOHPHQWV ZLWK DQ HOHFWULF GLVFKDUJH SUHVXPDEO\ DOVR RFFXUULQJ ZLWK VXIILFLHQW UHJXODULW\ RQ WKH SUHELRWLF HDUWKf SURGXFHG D SUHELRWLF VRXS FRQWDLQLQJ D PL[WXUH RI NH\

PAGE 20

PDWHULDOV PRVW QRWDEO\ +&1 0LOOHU DQG 8UH\ 6FOHVLQJHU DQG 0LOOHU 6WLEOLQJ DQG 0LOOHU f 7KH SUHELRWLF SUHVHQFH RI +&1 LV IXUWKHU VXSSRUWHG E\ LWV SUHVHQFH LQ LQWUDVWHOODU FORXGV LQ FRPHWV DQG LQ WKH DWPRVSKHUH RI 7LWDQ ,UYLQH +XHEQHU HW DO +DQHO f %DVHFDWDO\VHG WHWUDPHUL]DWLRQ RI +&1 \LHOGV GLDPLQRPDOHRQLWULOH ZKLFK WKHQ OHDGV WR WKH SURGXFWLRQ RI DGHQLQH K\SR[DQWKLQH JXDQLQH [DQWKLQH DQG GLDPLQRSXULQH +\GURO\VLV RI +&1 ROLJRPHUV \LHOGV DPLQR DFLGV SDUWLFXODUO\ JO\FLQH DODQLQH DVSDUWDWH DQG GLDPLQRVXFFLQDWH DPRQJ RWKHU DPLQR DFLGV UHYLHZHG LQ -R\FH f 2WKHU HOHFWULF GLVFKDUJH H[SHULPHQWV XQGHU SUHVXPDEO\ SUHELRWLF FRQGLWLRQV JHQHUDWHG SXULQHV DPRQJ RWKHU SRWHQWLDOO\ XVHIXO PDWHULDOV 7KHVH H[SHULPHQWV XVXDOO\ KDYH ORZ \LHOGV RI S\ULPLGLQH V\QWKHVLV EXW UHFHQW H[SHULPHQWV RI 0LOOHU GHPRQVWUDWH WKDW VLJQLILFDQW S\ULPLGLQH \LHOG IURP IUR]HQ GLOXWH VDPSOHV RI 1+&1 0L\DNDZD HW DO f $QRWKHU SUHELRWLF PROHFXOH +&2 FDQ OHDG WR JO\FHUDOGHK\GHV WKURXJK FRQGHQVDWLRQ LQ WKH SUHVHQFH RI D FDWDO\VW VXFK DV FDOFLXP FDUERQDWH RU DOXPLQD *DEHO 5HLG f *O\FHUDGHK\GH FDQ WKHQ EHJLQ D FDVFDGH WKDW FRQYHUWV IRUPDOGHK\GH LQWR WULRVHV WHUWURVHV DQG ODUJHU VXJDUV WKURXJK D SURFHVV RI DOGRO FRQGHQVDWLRQ DQG HQROL]DWLRQ -R\FH f -RLQLQJ QXFOHRVLGH DQG VXJDU VXEXQLWV KDV EHHQ DFFRPSOLVKHG E\ KHDWLQJ D PL[WXUH RI ULERVH DQG SXULQH WR GU\QHVV LQ WKH SUHVHQFH RI LQRUJDQLF VDOWV WKLV SURFHVV KRZHYHU OHDGV WR D PL[WXUH RI D DQG 3QXFOHRVLGHV )XOOHU HW DO f ,Q VXPPDU\ H[SHULPHQWDO HYLGHQFH VXJJHVWV WKDW WKH FKHPLFDO HQYLURQPHQW DQG FRPSRVLWLRQ RI WKH SUHELRWLF HDUWK LV VXIILFLHQW WR SURYLGH WKH QHFHVVDU\ PHFKDQLVPV DQG PDWHULDOV IRU WKH V\QWKHVLV RI PDQ\ SRWHQWLDOO\ XVHIXO PRQRPHUV 7KHVH ILQGLQJV DUH

PAGE 21

VXSSRUWHG E\ WKH GLVFRYHU\ RI VLPLODU PRQRPHUV RQ PHWHRULWHV ZKLFK ZHUH DOVR SUHVXPDEO\ V\QWKHVL]HG E\ SUHELRWLF PHFKDQLVPV LQ VSDFH 7KH QH[W VWHS WRZDUGV D OLYLQJ V\VWHP UHTXLUHV WKDW PRQRPHUV EH OLQNHG WRJHWKHU E\ SUHELRWLF FKHPLVWU\ WR IRUP SRO\PHUV /LPLWHG SRO\PHUL]DWLRQ RI QXFOHLF DFLGV KDV EHHQ REVHUYHG XQGHU SUHELRWLF FRQGLWLRQV $GHQRVLQH f fF\FOLF SKRVSKDWH FDQ EH FRQYHUWHG WR ROLJRPHUV E\ KHDWLQJ WR GU\QHVV RU E\ LQFXEDWLQJ DW URRP WHPSHUDWXUH LQ WKH SUHVHQFH RI D SRO\DPLQH 9HUODQGHU f 1XFOHRVLGH fSRO\SKRVSKDWHV KDYH DOVR EHHQ SURGXFHG E\ KHDWLQJ QXFOHRWLGHV DQG LQRUJDQLF SRO\SKRVSKDWHV WR GU\QHVV /RKUPDQQ f 7KH UDQGRP SRO\PHUL]DWLRQ RI PRQRPHUV KRZHYHU UHPDLQV D JUHDW FKDOOHQJH VSHFLILFDOO\ LQ OLJKW RI WKH DEXQGDQFH RI VWHUHRFKHPLFDOO\ UHODWHG FRPSRXQGV WKDW DUH H[SHFWHG WR LQWHUIHUH ZLWK SUHELRWLF SRO\PHUL]DWLRQ 7DNLQJ ULERVHEDVHG QXFOHRWLGHV DV DQ H[DPSOH QXFOHRWLGH SRO\PHUL]DWLRQ FDQ UHVXOW LQ D FRPELQDWRULDO PL[WXUH RI f f f f DQG f f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

PAGE 22

,W LV LQWHUHVWLQJ WR QRWH WKDW QRW DOO SRWHQWLDOO\f SUHELRWLF PRQRPHUV VXIIHU DV PDQ\ RI WKH VWHUHRLVRPHULF FKDOOHQJHV DV ULERVHEDVHG QXFOHRWLGHV %RWK WKUHRVH QXFOHLF DFLG 71$f DQG SHSWLGH QXFOHLF DFLG 31$f XVH DOWHUQDWLYH EDVHV DWWDFKHG WR QXFOHLF DFLGV %RWK PRQRPHUV DUH PRUH FKHPLFDOO\ VLPSOLVWLF WKXV DYRLGLQJ PXFK RI WKH VWHUHRLVRPHULF FOXWWHU RI 51$ 7KLV KDV OHG WR SURSRVDOV RI D SUH51$ ZRUOG XWLOL]LQJ HLWKHU SRO\PHU DV SRWHQWLDOO\ OLIHVXSSRUWLQJ SRO\PHUV 7KHRUHWLFDO DQG H[SHULPHQWDO HYLGHQFH VXJJHVW WKDW LQIRUPDWLRQ VWRUHG LQ WKH IRUP RI 71$ RU 31$ SRO\PHUV FDQ EH WUDQVIHUUHG WR 51$ DOWKRXJK LW LV XQFOHDU KRZ PXFK SKHQRW\SLF IXQFWLRQ LV SUHVHUYHG LQ WKH WUDQVIHU UHYLHZHG LQ -R\FH f 7KH 1H[W &KDOOHQJH 2EWDLQLQJ &DWDO\WLF )XQFWLRQ IURP 5DQGRP /LEUDULHV $OWKRXJK PRQRPHU DQG SRO\PHU V\QWKHVLV LV H[SHFWHG WR EH GLIILFXOW LQ D SUHELRWLF ZRUOG LW LV WKH IRUPDWLRQ RI SRO\PHUV WKDW DFW DV JHQHWLF PROHFXOHV ZLWK XVHIXO FKHPLFDO SURSHUWLHV WKDW DSSHDUV PRVW IRUPLGDEOH XVHIXO IURP WKH SHUVSHFWLYH WKDW LW OHDGV WR WKH FUHDWLRQ RI OLIH DQ LQLWLDOO\ XVHIXO SURSHUW\ ZRXOG EH WKH DELOLW\ WR HQULFK IRU VSHFLILF PRQRPHUVf 7KLV UHTXLUHV WKDW SRO\PHUV EH RI VXIILFLHQW OHQJWK WR FRQWDLQ VXIILFLHQW LQIRUPDWLRQ EH VXIILFLHQWO\ VWDEOH DV WR SHUVLVW EH DEOH WR PXWDWH \HW VWLOO EH UHSOLFDWHG DQG ODVWO\ SHUIRUP XVHIXO FKHPLFDO IXQFWLRQ $OWKRXJK ZH GR QRW NQRZ SUHFLVHO\ ZKDW PRQRPHUV ZHUH DYDLODEOH LQ D SULRELRWLF ZRUOG RU KRZ WKH\ ZHUH V\QWKHVL]HG DQG SRO\PHUL]HG LW LV FOHDU WKDW DW VRPH SRLQW a E\D VRPH UDQGRPO\ JHQHUDWHG SUHELRWLF SRO\PHUV PXVW KDYH SRVVHVVHG fXVHIXOf SURSHUWLHV SUHVXPDEO\ D VHOIUHSOLFDVHf WKDW SHUPLWWHG WKH MXPS IURP D FROOHFWLRQ RI UDQGRP SRO\PHUV WR D VLQJOHELRSRO\PHU OLIHIRUP ZLWK D FRPSOH[ PHWDEROLVP

PAGE 23

$OWKRXJK WKLV LV WKH PRVW UHDVRQDEOH K\SRWKHVLV LW LV GLIILFXOW WR LPDJLQH WKDW VXFK D FRPSOH[ IXQFWLRQ FRXOG EH JHQHUDWHG IURP D UDQGRP SRO\PHU $Q H[SHULPHQWDO DSSURDFK LQ YLWUR VHOHFWLRQ ,96f HPHUJHG LQ WKH V WKDW VHUYHG DV D EULGJH EHWZHHQ WKH fZRUN IRUZDUGf DQG fZRUN EDFNZDUGf DSSURDFKHV E\ SHUPLWWLQJ D GLUHFW WHVW RI WKH FDWDO\WLF SRZHU RI UDQGRPSRO\PHU OLEUDULHV 6]RVWDN %HDXGU\ ,UYLQH f ,96 EHJLQV ZLWK D OLEUDU\ FRQWDLQLQJ aX XQLTXH DQG UDQGRPO\ JHQHUDWHG SRO\PHUV RI '1$ 7KLV OLEUDU\ LV WKHQ FKDOOHQJHG E\ D VHOHFWLRQ SURFHVV VXFK DV WKH DELOLW\ WR FDWDO\]H D SDUWLFXODU UHDFWLRQ RU ELQG D SDUWLFXODU PROHFXOHf 0ROHFXOHV WKDW FDQ SHUIRUP WKH VHOHFWHG WDVN DUH VHSDUDWHG IURP WKRVH WKDW GR QRW )LJXUH f )LQGLQJ PROHFXOHV LQ D UDQGRP OLEUDU\ WKDW SHUIRUP WKH VHOHFWHG WDVN LV WKHUHIRUH D IXQFWLRQ WKH DEXQGDQFH RI WKDW DFWLYLW\ DQG WKH DELOLW\ RI WKH VHOHFWLRQ UHJLPH WR VHSDUDWH DFWLYH IURP LQDFWLYH PROHFXOHV &RPELQDWRULDO OLEUDU\ RI '1$ PROHFXOHV 6ROLG VXSSRUW ‘‘ W P 5DQGRP UHJLRQ V ] ,7 A PLQL QPLLL LQ 'HQDWXUDWLRQ DQG HOXWLRQ RI FRPSOHPHQWDU\ VWUDQG f ‘n %LQGLQJ RI 3&5 SURGXFWV WR VROLG VXSSRUW 3&5 DPSOLILFDWLRQ + n0LO A ( K )LJXUH $ VFKHPDWLF GLDJUDP GHVFULELQJ LQ YLWUR VHOHFWLRQ IRU D '1$ PROHFXOH WKDW FDWDO\]HV D FOHDYDJH UHDFWLRQ

PAGE 24

,Q SUDFWLFH WKH VHOHFWLRQ SURFHVV LV QRW SHUIHFW DOORZLQJ LQDFWLYH PROHFXOHV WR VXUYLYH DQG SUHVXPDEO\ ORVLQJ VRPH DFWLYH PROHFXOHVf :KHQ DFWLYH PROHFXOHV DUH UDUH DQG EDFNJURXQG LV KLJK VHYHUDO VHTXHQWLDO VHOHFWLRQV DUH UHTXLUHG WR HQULFK IRU DFWLYH PROHFXOHV ([SHULPHQWDOO\ ,96 DFFRPSOLVKHV VHTXHQWLDO URXQGV RI VHOHFWLRQ E\ XVLQJ WKH VXUYLYLQJ PROHFXOHV IURP WKH ILUVW URXQG RI VHOHFWLRQ DV D WHPSODWH IRU 3&5 WKXV DPSOLI\LQJ WKH VXUYLYRUV DQG FUHDWLQJ D VHFRQG SRRO RI '1$ PROHFXOHV WKDW DUH SUHVXPDEO\ HQULFKHG IRU FDWDO\VWVf IRU D VXEVHTXHQW URXQG RI VHOHFWLRQ 7KLV H[SHULPHQWDO VWUDWHJ\ LV D FORVH DQDORJ\ WR ZKDW LV LPDJLQHG WR KDYH OHG WR WKH ILUVW OLIHIRUP ZLWK WKH H[FHSWLRQ WKDW SUHELRWLF PHFKDQLVPV DV RSSRVHG WR KXPDQ LQWHUYHQWLRQ OHDG WR WKH GHYHORSPHQW RI WKH SRO\PHU OLEUDU\ DQG fVHOHFWLRQ IRU GHVLUHG DFWLYLW\f ZDV LPSRVHG E\ WKH (DUWKfV HQYLURQPHQW DQG SURFHVVHV ZKLFK OHDG WR WKH GHFRPSRVLWLRQ RI SRO\PHUVf ,W ZDV DQWLFLSDWHG WKDW ,96 ZRXOG QRZ GHPRQVWUDWH WKDW WKH FKHPLFDO SURFHVVHV QHFHVVDU\ IRU OLIH FRXOG LQ IDFW EH IRXQG LQ D SUHELRWLF OLEUDU\ RI SRO\PHUV $IWHU D GHFDGH RI H[SHULPHQWV DQG JUHDW HIIRUW PDQ\ FDWDO\VWV KDYH LQGHHG EHHQ LVRODWHG IURP UDQGRP OLEUDULHV PRVW QRWDEO\ D SKRVSKRGLHVWHUDVH D 'LHOV$OGHUDVH D QXFOHRWLGH OLJDVH DQG D OLPLWHG SRO\PHUDVH %UHDNHU DQG -R\FH -RKQVWRQ 6HHOLQJ 7DUDVRZ f 'HVSLWH WKHVH VXFFHVVHV JHWWLQJ VHOIUHSOLFDWLQJ QXFOHLF DFLG SRO\PHUV IURP ,96 SURYHG KDUGHU WKDQ DQWLFLSDWHGKRSHG %DUWHO DQG RWKHUV DUH FRPLQJ FORVH WR DFKLHYLQJ WKLV JRDO -RKQVWRQ f ,W PXVW EH QRWHG KRZHYHU WKDW HYHQ WKH DXWKRUV RI WKLV UHVHDUFK FRQWULEXWH WKH VXFFHVV RI WKH GH QRYR JHQHUDWLRQ RI DQ 51$ EDVHG 51$ SRO\PHUDVH WR D VWHSZLVH SURFHGXUH JXLGHG E\ KXPDQ FKHPLFDO NQRZOHGJH f7KXV IDU HIIRUWV WR VHOHFW IRU SRO\PHUL]DWLRQ DFWLYLW\ LQ D VLQJOH VWHS GLUHFWO\ IURP UDQGRPVHTXHQFH

PAGE 25

51$ KDYH \LHOGHG RQO\ ULER]\PHV WKDW GHFRUDWH WKHPVHOYHV LQDSSURSULDWHO\ ZLWK WDJJHG QXFOHRWLGHVf -RKQVWRQ SDJH f )XUWKHUPRUH HVWLPDWHV IURP WKHLU ZRUN VXJJHVW WKDW PRUH WKDQ A UDQGRP 51$ VHTXHQFHV PLJKW EH UHTXLUHG WR JHW WHPSODWH GLUHFWHG 51$ SRO\PHUL]DWLRQ IURP D WUXO\ UDQGRP SRRO 0RUH GLUHFWO\ %DUWHO DQG 6]RVWDN QRWHG WKDW WKH RGGV RI REWDLQLQJ D OLJDVH DEOH WR HQKDQFH D YHU\ VLPSOH WHPSODWHG OLJDWLRQ UHDFWLRQ E\ IRXU RUGHUV RI PDJQLWXGH ZHUH RQ WKH RUGHU RI RQH LQ %DUWHO f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f 7D\ORU HW DO f 7KLV VHDUFK RI DPLQR DFLG OLEUDULHV FRQFOXGHG WKDW DOWKRXJK WKH GLYHUVLW\ RI IXQFWLRQDOLW\ LQ SURWHLQV SURYLGHV LQWULQVLFDOO\ JUHDWHU FDWDO\WLF SRWHQWLDO IROGLQJ SURWHLQV PD\ EH LQKHUHQWO\ PRUH GLIILFXOW WKDQ IROGLQJ QXFOHLF DFLG SRO\PHUV WKHUHIRUH PDNLQJ DQ DFWLYH HQ]\PH IURP D UDQGRP SRO\PHU RI DPLQR DFLGV DSSHDUV WR VXEVWDQWLDOO\ PRUH GLIILFXOW WKDW REWDLQLQJ D ULER]\PH IURP D UDQGRP QXFOHLF DFLG OLEUDU\ $OWKRXJK QDWXUDO DPLQR DFLG SRO\PHUV SURWHLQVf IXQFWLRQ VXSHUEO\ DV FDWDO\VWV DQG DPLQR DFLG FDWDO\VWV KDYH EHHQ LVRODWHG IURP UDQGRP OLEUDULHV DPLQR DFLG SRO\PHUV DUH QRW ZHOO VXLWHG DV FDUULHUV RI JHQHWLF LQIRUPDWLRQ IRU UHDVRQV GHVFULEHG HDUOLHU ,Q YLWUR VHOHFWLRQV XVLQJ DPLQR DFLG OLEUDULHV DUH DOVR PXFK PRUH WHFKQRORJLFDOO\ OLPLWHG WKDQ VHDUFKHV RI QXFOHLF DFLG VHTXHQFH VSDFH )RU DOO WKHVH UHDVRQV DQG WKH LQGLFDWLRQ WKDW DOWKRXJK OLIH RQ HDUWK PD\ QRW KDYH VWDUWHG ZLWK 51$ LW FOHDUO\ SDVVHG WKURXJK D

PAGE 26

RQHELRSRO\PHU VWDJH EDVHG RQ 51$ LQ YLWUR VHOHFWLRQV KDYH ODUJHO\ IRFXVHG RQ QXFOHLF DFLG ELRFKHPLVWU\ 0RUH WKDQ D GHFDGH DJR %HQQHU SURSRVHG ZKDW KDV FRPH WR EH WKH VWDQGDUG H[SODQDWLRQ IRU WKH SRRU FDWDO\WLF SRZHU RI QXFOHLF DFLGV ZKHUH SRRU UHIOHFWV LQ SDUW DQ XQEDODQFHG FRPSDULVRQ ZLWK QDWXUDO SURWHLQ HQ]\PHVf %HQQHU %HQQHU Ef 1XFOHLF DFLGV KH DUJXHG KDYH OLWWOH RUJDQLF IXQFWLRQDOLW\ DW OHDVW FRPSDUHG WR QDWXUDO SURWHLQ HQ]\PHV '1$ ODFNV LQ WKH VFDIIROG FDWLRQLF JURXSV LPLGD]ROHV WKLROV DQG FDUER[\ODWHV DOO RI ZKLFK SOD\ HIIHFWLYH FDWDO\WLF UROHV LQ SURWHLQ HQ]\PHV QHDU QHXWUDO S+ 7R UHOLHYH WKHVH OLPLWDWLRQV %HQQHU VXJJHVWHG H[SDQGLQJ WKH JHQHWLF DOSKDEHW IRU WKH SXUSRVH RI SHUPLWWLQJ '1$ WR FDUU\ PRUH RI WKHVH IXQFWLRQDO JURXSV VHH 6ZLW]HU 3LFFLULOOL f 7KLV K\SRWKHVLV JDLQV VXSSRUW IURP WKH REVHUYDWLRQ WKDW PDQ\ RI WKH FDWDO\WLF 51$V EHOLHYHG WR EH YHVWLJHV RI WKH 51$ ZRUOG VSOLFHRVRPH 51DVH3 ULERVRPHf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f 6HYHUDO JURXSV %DWWHUVE\ /DWKDP 6DNWKLYHO 3HUULQ 3HUULQ f KDYH QRZ V\QWKHVL]HG IXQFWLRQDOL]HG QXFOHLF DFLGV PRVWO\ GHULYDWLYHV RI n

PAGE 27

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f 1R VLJQLILFDQW FDWDO\WLF ERRVW ZDV GHWHFWHG KRZHYHU RYHU '1$]\PHV LVRODWHG HDUOLHU E\ WKH VDPH UHVHDUFKHUV XVLQJ QRQIXQFWLRQDOL]HG QXFOHRWLGHV 6DQWRUR f 7RJHWKHU ZLWK RWKHU VLPLODU ILQGLQJV WKLV VXJJHVWHG WKDW IXQFWLRQDO HQGRZPHQW RIIHUV OHVV WKDQ DQ RUGHU RI PDJQLWXGH HQKDQFHPHQW RI WKH FDWDO\WLF SRWHQWLDO RI WKH OLEUDU\ $QJ $QJ f $QDORJRXV H[SHULPHQWV ZKHUH IXQFWLRQDOLW\ ZDV LQWURGXFHG DV D FRIDFWRU ELQGLQJ WR EXW QRW FRYDOHQWO\ DWWDFKHG WR WKH '1$ PROHFXOHV JHQHUDWHG DQDORJRXV FRQFOXVLRQV 5RWK f :KDW DSSHDUV WR EH DEVHQW IURP WKH OLWHUDWXUH LV WKH VHQVH RI DVWRQLVKPHQW DW WKHVH UHVXOWV )XQFWLRQDO JURXSV DUH VXSSRVHG WR EH LPSRUWDQW LQ RUJDQLF UHDFWLYLW\ )RU H[DPSOH LQ DQ H[SHULPHQW LQ ZKLFK D SRO\SHSWLGH ZDV JHQHUDWHG WKDW IROGHG LQ VROXWLRQ DQG FDWDO\]HG WKH GHFDUER[\ODWLRQ RI R[DORDFHWDWH -RKQVVRQ -RKQVVRQf WKH DPLQR JURXSV SHUIRUPHG WKH FDWDO\VLV +DYLQJ WKHP PDWWHUHG QRW KDYLQJ WKHP PDWWHUHG HYHQ PRUH DV FDWDO\VLV ZHQW GRZQ E\ VHYHUDO RUGHUV RI PDJQLWXGH WR EDFNJURXQGf ZKHQ WKH\ ZHUH UHPRYHG ,Q FKHPLFDO PRGHOV DGGLQJ LPLGD]ROH IRU H[DPSOHf LQFUHDVHV UHDFWLYLW\ RI FDWDO\WLF PRGHOV E\ RUGHUV RI PDJQLWXGH QRW E\ D IDFWRU RI WZR WR WHQ =HSLN f

PAGE 28

:KHQ RUJDQLF FKHPLFDO WKHRU\ H[SHFWV D ODUJH RUGHUV RI PDJQLWXGHf LPSDFW RI D FKDQJH RI VWUXFWXUH RQ PROHFXODU EHKDYLRU DQG RQO\ D VPDOO LPSDFW LV VHHQ D IDFWRU RI WR f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f WKXV WKH IHZ HVWLPDWHV RI FDWDO\WLF IUHTXHQF\ SURYLGHG DUH QRW WR EH FRQVLGHUHG UHDOLVWLF ,QGHHG LW LV GLIILFXOW WR LPDJLQH OLIH HPHUJLQJ VSRQWDQHRXVO\ IURP QXFOHLF DFLGV JLYHQ WKHVH NLQGV RI QXPEHUV HVSHFLDOO\ JLYHQ WKH GLIILFXOWLHV LQ REWDLQLQJ 51$ RU '1$f LQ SRO\PHULF IRUP XQGHU DQ\ SODXVLEOH SUHELRWLF UHDFWLRQ FRQGLWLRQV LQ ZDWHU ,QLWLDOO\ WKH OLPLWHG JRDO RI LVRODWLQJ DQ\ DFWLYLW\ IURP D OLEUDU\ ZDV XQGHUVWDQGDEOH JLYHQ WKDW LW ZDV QRW NQRZQ ZKHWKHU DQ\ VXFK DFWLYLW\ H[LVWHG LQ D UDQGRP OLEUDU\ .QRZLQJ WKDW DFWLYLW\ FDQ EH LVRODWHG IURP UDQGRP

PAGE 29

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fW DSSHDU WR VLJQLILFDQWO\ LPSURYH FDWDO\VLV )LYH FDWHJRULHV RI K\SRWKHVHV PLJKW EH FRQVLGHUHG WR DFFRXQW IRU WKLV RUGHUVRIPDJQLWXGH GLIIHUHQFH EHWZHHQ H[SHFWDWLRQ DQG RXWFRPH ZKHQ DGGLQJ IXQFWLRQDOLW\ Df 2XU YLHZ RI WKH UROH RI IXQFWLRQDOLW\ LQ FDWDO\VLV PD\ EH QDLYH )XQFWLRQDO JURXSV PD\ QRW JUHDWO\ HQKDQFH FDWDO\WLF SRZHU Ef 1XFOHLF DFLGV DUH RI FRXUVH DOUHDG\ IXQFWLRQDOL]HG ZLWK SKRVSKDWHV DQG K\GURJHQ ERQG GRQDWLQJ DQG DFFHSWLQJ JURXSVf 3HUKDSV WKHVH DUH WKH RQO\ IXQFWLRQDOLWLHV QHHGHG IRU FDWDO\VLV Ff 1XFOHLF DFLGV DOWKRXJK WKH\ GR QRW KDYH PDQ\ IXQFWLRQDO JURXSV LQWULQVLFDOO\ LQ WKHLU FRYDOHQW VWUXFWXUH PD\ UHFUXLW VXIILFLHQW QXPEHUV RI WKHVH LQ SDUWLFXODU I GLYDOHQW FDWLRQV VXFK DV 0J f DV FRIDFWRUV WR PDNH XS IRU WKH ODFN RI IXQFWLRQDOLW\ FRYDOHQWO\ OLQNHG WR WKH VFDIIROGLQJ Gf 7KH 96 H[SHULPHQWV PD\ KDYH EHHQ GHVLJQHG LQ D ZD\ DV WR RYHUORRN RU ORVH WKH EHVW IXQFWLRQDOL]HG FDWDO\VWV Hf 3HUKDSV DGGLQJ D VLQJOH W\SH RI IXQFWLRQDO JURXS LV QRW VXIILFLHQW RQH PD\ QHHG WR DGG WZR RU PRUH W\SHV RI IXQFWLRQDO JURXSV EHIRUH WKH H[SHFWHG ODUJH EHQHILW IURP IXQFWLRQDOLW\ LV VHHQ 3HUULQ 3HUULQ f $ SULRUL HDFK K\SRWKHVLV KDV LWV DWWUDFWLRQV DQG GLVDGYDQWDJHV )RU H[DPSOH ZH FDQ HDVLO\ LGHQWLI\ ZD\V LQ ZKLFK WKH EHVW HQGRZHG FDWDO\VWV ZRXOG EH ORVW LQ D %UHDNHU -R\FH SURWRFRO VHH EHORZf VXJJHVWLYH RI K\SRWKHVLV Gf 'LYDOHQW PHWDO LRQV GR ELQG WR '1$ WKH\ FDQ SDUWLFLSDWH LQ FDWDO\VLV DQG WKH\ DVVLVW HYHQ PRUH KLJKO\ IXQFWLRQDOL]HG

PAGE 30

SURWHLQ FDWDO\VWV VXJJHVWLYH RI K\SRWKHVLV Ff 3URWHLQV GR VHHP WR H[SORLW PXOWLSOH W\SHV RI IXQFWLRQDOLW\ DQG H[SORLW K\GURSKRELFLW\ VRPHWKLQJ ODFNLQJ LQ DOLSKDWLF IRUP LQ '1$ VXJJHVWLYH RI K\SRWKHVLV Hf /DVW UHFHQW H[SHULPHQWV E\ .HHIH DQG 6]RVWDN VHOHFWLQJ IRU SURWHLQV WKDW GR WKLQJV VXJJHVWV WKDW WKH\ GR QRW GR WKLQJV RUGHUV RI PDJQLWXGH EHWWHU WKDQ '1$ f VXJJHVWLYH RI K\SRWKHVLV Df RU Ef ([SHULPHQWDO 'HVLJQ 7KLV WKHVLV SUHVHQWV WKH ILUVW VWHS WRZDUGV DGGUHVVLQJ WKH K\SRWKHVHV DERYH :H XWLOL]HG WKH H[SHULPHQWDO SURFHGXUH GHYHORSHG E\ %UHDNHU DQG -R\FH f 7KLV ,96 SURFHGXUH VHHNV '1$ PROHFXOHV WKDW DUH FDWDO\WLFDOO\ DFWLYH DV ULERQXFOHDVHV WKLV SURWRFRO LV WKH PRVW ZLGHO\ XWLOL]HG SURFHGXUH IRU HVWLPDWLQJ FDWDO\WLF SHUIRUPDQFH XQGHU YDULRXV FRQGLWLRQV DOORZLQJ JHQHUDO FRPSDULVRQV VXFK DV WKH RQHV PHQWLRQHG EHIRUH WKDW VXJJHVW IXQFWLRQDOLW\ GRHV QRW VLJQLILFDQWO\ LPSURYH FDWDO\WLF SRWHQWLDO $V PHQWLRQHG DERYH RQH H[SODQDWLRQ IRU WKH DSSDUHQW IDLOXUH RI IXQFWLRQDO HQGRZPHQW WR JHQHUDWH YHU\ DFWLYH GHR[\ULER]\PHV PD\ UHVXOW IURP WKH ORVV RI KLJKO\ UHDFWLYH FDWDO\VWV GXULQJ VRPH VWHS LQ WKH H[SHULPHQW EHIRUH WKH\ KDG WKH RSSRUWXQLW\ WR VXUYLYH WKH VHOHFWLRQ VWHS ,W LV HDV\ WR ILQG SODFHV LQ WKH VHWXS RI WKH %UHDNHU-R\FH VHOHFWLRQ SURFHGXUH ZKHUH KLJKO\ DFWLYH FDWDO\VWV PLJKW EH ORVW 6WDQGDUG ,96 H[SHULPHQWDO V\VWHPV DVVXPH WKDW FDWDO\VWV UHPDLQ LQDFWLYH GXULQJ WKH ZRUN XS SURFHGXUH ZKLFK LQFOXGHV 3&5 ZLWK 0J&/f HWKDQRO SUHFLSLWDWLRQ ZLWK KLJK VDOWVf ELQGLQJ WR D FROXPQ DJDLQ LQ KLJK VDOWV EXW ZLWKRXW 0J&/f DQG UHPRYLQJ WKH FRPSOHPHQWDU\ VWUDQG ZLWK KLJK S+ VRGLXP K\GUR[LGH $IWHU ZRUN XS LV FRPSOHWHG WKH UHDFWLRQ LV fVWDUWHGf ZLWK WKH DGGLWLRQ RI 0J&/ DQG WKH HOXDWH IURP D FROXPQ LV FROOHFWHG DQG SUHVXPHG WR FRQWDLQ WKH DFWLYH

PAGE 31

FDWDO\VW 7KHUH DUH VHYHUDO WHFKQLFDO ZHDNQHVVHV RI WKH %UHDNHU-R\FH 96 SURWRFRO WKDW PD\ UHVXOW LQ WKH ORVV RI GHVLUHG FDWDO\VWV )LUVW FROXPQV KDYH SRRU UHVROYLQJ DELOLW\ LQ SDUW EHFDXVH WKH\ KDYH JUHDW SRWHQWLDO IRU UHVLGXDO UHWHQWLRQ XQERXQG PDWHULDO VORZO\ ILOWHULQJ WKURXJK WKH FROXPQf DQG OHDNDJH ERXQG PDWHULDO EHFRPLQJ XQERXQGf ,W LV DOVR DVVXPHG WKDW HYHQ WKH fEHVWf FDWDO\VWV UHPDLQ LQDFWLYH XQWLO WKH UHDFWLRQ LV fVWDUWHGf E\ DGGLQJ 0J&K RU RWKHU FRIDFWRU VXFK DV KLVWLGLQHf DQG FROOHFWLRQ EHJLQV 0DJQHVLXP LV LQ IDFW SUHVHQW GXULQJ WKH SUHSDUDWLRQ LQFOXGLQJ GXULQJ WKH 3&5 DQG RWKHU VDOWV DUH DOVR SUHVHQW GXULQJ WKH HWKDQRO SUHFLSLWDWLRQ SRWHQWLDOO\ SHUPLWWLQJ FDWDO\VLV GXULQJ WKH SUHSDUDWLRQ VWHSV DQG UHVXOWLQJ LQ ORVV RI FDWDO\VLV '1$]\PHV KDYH EHHQ LVRODWHG XVLQJ WKH %UHDNHU-R\FH SURWRFRO WKDW DUH DFWLYH ZLWKRXW DQ\ DGGHG FRIDFWRU UHTXLULQJ RQO\ 1D&Of RIIHULQJ VXSSRUW WR WKH K\SRWKHVLV WKDW DFWLYH FDWDO\VWV PD\ EH ORVW GXULQJ WKH ZRUNXS 6HFRQG WKH DFWLYLW\ RI D '1$]\PH LV LQKLELWHG E\ WKH FRPSOHPHQWDU\ VWUDQG DQG WKHUHIRUH PXVW EH UHPRYHG 7KLV LV DFFRPSOLVKHG LQ WKH %UHDNHU-R\FH SURWRFRO E\ GHQDWXULQJ WKH GRXEOHVWUDQGHG '1$ ZLWK D VWURQJ EDVH VROXWLRQ 1D2+f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fUHTXLUHGf FRIDFWRUf WKH UHDFWLRQ LV fVWDUWHGf E\ WKH DGGLWLRQ RI

PAGE 32

WKH FRIDFWRU &RUUHFW IROGLQJ LV DVVXPHG WR RFFXU ZLWKRXW WKH UHTXLUHG FRIDFWRU DQG TXLFNO\ UHODWLYH WR FDWDO\VLV (DFK RI WKHVH SUREOHPV DULVHV EHFDXVH WKH %UHDNHU-R\FH SURWRFRO KDV WKH n K\GUR[\O JURXS RI WKH 51$ PROHFXOH WR EH IUHH GXULQJ WKH VHWXS 7R UHVROYH WKHVH SUREOHPV UHVHDUFK LQ WKLV GLVVHUDWLRQ GHYHORSV WHFKQLTXHV WR HOLPLQDWH ORVV RI FDWDO\VWV GXULQJ VHW XS E\ XVLQJ D FDJHGULERVH &DJHGULERVH ZDV V\QWKHVL]HG ZLWK WKH nK\GUR[\O JURXS SURWHFWHG YLD DQ RUWKRQLWUREHQ]\O SURWHFWLQJ JURXS 7KLV SURWHFWLQJ JURXS FDQ WKHQ EH UHPRYHG E\ 89 SKRWRO\VLV )LJXUH f /$6(5 ZZ &DJHG ULERVH &DJHG LV EORFNLQJ WKH n+ LQ ULERVH IRU QXFOHRSKLOLF DWWDFN RQ WKH SKRVSKRUXV n+ LV IUHH DQG DEOH WR DWWDFN WKH SKRVSKRUXV DWRP ZLWK WKH KHOS RI D FDWDO\VW )LJXUH ([SRVLQJ WKH FDJHGULERVH FRQYHUWV fRUWKRQLWUREHQ]\O SURWHFWLQJ JURXS WR WKH VWDQGDUG K\GUR[\O JURXS WKXV JHQHUDWLQJ D VWDQGDUG 51$ QXFOHRWLGH

PAGE 33

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f &RQGLWLRQV IRU WKH XVH RI FDJHGULERVH KDYH EHHQ RSWLPL]HG DOORZLQJ UHDFWLRQ GXULQJ WKH VHW XS WR EH SUHYHQWHG 3UHOLPLQDU\ VHOHFWLRQV ZLWK WKH FDJHGULERVH LQGLFDWH WKDW WKH ODVHU SKRWRO\VLV RI WKH FDJHGULERVH FUHDWHV D VOLJKW LQFUHDVH LQ EDFNJURXQG FOHDYDJH ERWK H[SHULPHQWDO DQG VLPXODWHG VHOHFWLRQV LQGLFDWH WKDW WKLV OHDNDJH GUDPDWLFDOO\ LQFUHDVHV WKH QXPEHU RI URXQGV RI VHOHFWLRQ UHTXLUHG WR LVRODWH DFWLYH FDWDO\VWV $OWKRXJK WKLV OLPLWV WKH XWLOLW\ RI FDJHGULERVH IRU VHOHFWLRQV WKH EDFNJURXQG GRHV QRW VLJQLILFDQWO\ DOWHU NLQHWLF DQDO\VLV DOORZLQJ LWV XVH IRU VWXG\LQJ WKH SUHDFWLYDWLRQ UHTXLUHPHQWV RI '1$]\PHV VXFK DV IROGLQJ 7R DLG LQ WKH GHYHORSPHQW RI UHDOLVWLF PRGHOV ZH XQGHUWRRN D GHWDLOHG VWXG\ RI DQ LQGLYLGXDO '1$]\PH 7KLV UHYHDOHG D QXPEHU RI VLJQLILFDQW ILQGLQJV ,W LV ZLGHO\ DVVXPHG WKDW LQFOXGLQJ D FRIDFWRU VXFK DV 0J RU RWKHU GLYDOHQW FDWLRQVf LQWR WKH VHOHFWLRQ EXIIHU ZLOO UHVXOW LQ WKH VHOHFWLRQ IRU FRIDFWRUGHSHQGHQW FDWDO\VWV ,QGHHG WKH %UHDNHU-R\FH SURWRFRO SUHVXPHV WKDW SRWHQWLDO FDWDO\VWV DUH FRPSOHWHO\ LQDFWLYH ZLWKRXW WKH DGGHG FRIDFWRU :H KDYH GHWHUPLQHG WKDW DOWKRXJK ZDV VHOHFWHG LQ WKH SUHVHQFH RI 0J UHWDLQV FRPSOHWH FDWDO\WLF DFWLYLW\ LQ WKH DEVHQFH RI 0J ,W LV DOVR ZLGHO\

PAGE 34

DVVXPHG WKDW WKH LGHQWLW\ RI WKH FRQVWDQW UHJLRQV DUELWUDULO\ FKRVHQ E\ UHVHDUFKHUVf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f 7KLV KDV JLYHQ XV LQVLJKW LQWR KRZ WR GHVLJQ DQG VLPXODWH IXWXUH 96 :H KDYH DOVR LVRODWHG VHYHUDO FDWDO\VWV VRPH RI ZKLFK DUH FORVH UHODWLYHV RI DQG LVRODWHG LQ D SUHYLRXV VHOHFWLRQ ZKLFK HDFK GLVSOD\ FDWDO\WLF DFWLYLW\ 7KLV LV VLJQLILFDQW LQ WZR DVSHFWV ILUVW LQ WKDW WKH FDWDO\WLF DFWLYLW\ YDULHV VLJQLILFDQWO\ ZLWK RQO\ D PLQLPDO VHTXHQFH FKDQJHV 7KLV VXJJHVWV WKDW WKH ODQGVFDSH LQ WKH QHDU YLFLQLW\ RQH RU WZR QXFOHRWLGHVf RI D SDUWLFXODU DFWLYH PROHFXOH FRQWDLQV PDQ\ FDWDO\VWV SUHVHQWLQJ RSSRUWXQLWLHV IRU FRQWLQXRXV H[SORUDWLRQ RI WKH IXQFWLRQDO ODQGVFDSH 6HFRQG WKH DEXQGDQFH RI QHDUUHODWLYHV WKDW UHWDLQ DFWLYLW\ VXJJHVWV WKDW PRVW HVWLPDWHV RI WKH FDWDO\WLF IUHTXHQF\ PD\ EH JURVV XQGHUHVWLPDWLRQV DV QHDUUHODWLYHV PD\ DSSHDU LGHQWLFDO XQGHU WKH PHWKRGV UHVWULFWLRQ GLJHVWf XVHG WR GLVWLQJXLVK WKHP

PAGE 35

&+$37(5 0$7(5,$/6 $1' 0(7+2'6 3UHSDUDWLRQ RI 3UHFXUVRU '1$\PHV YLD 3&5 '1$]\PHV IRU NLQHWLFV ZHUH SUHSDUHG E\ 3&5 DPSOLILFDWLRQ RI WKH WHPSODWH HLWKHU V\QWKHVL]HG E\ ,'7 RU IURP D FORQHf XVLQJ RQH RI YDULRXV SULPHUV WR JHQHUDWH WKH FDWDO\WLF VWUDQG DQG RQH RI YDULRXV SULPHUV WR JHQHUDWH WKH FRPSOHPHQWDU\ VWUDQG $OO WHPSODWHV KDG D FRPPRQ f DQG f FRQVWDQW UHJLRQ WR ZKLFK WKH FRPSOHPHQWDU\ DQG FDWDO\WLF VWUDQG SULPHUV FDQ ELQG %HWZHHQ WKH SULPHU ELQGLQJ VLWHV LV D QXFOHRWLGH UHJLRQ RI YDU\LQJ VHTXHQFH fOLEUDU\f WHPSODWH FRPSOHPHQWDU\ VWUDQGf n JWJFFDDJFWWDFFJWFDF Qf *$*$7*7&*&&$7&7&77&& ffWHPSODWH nFWJFDJDDWWFWDDWDFJDFWFDFWDWDJJDDJDFDWJJFJDFWFWF $&$7&$7*&*$*&$&$&*&$$7$*&&7*$7$$**77**7$*7*$&**7$$*&77**&$& 7ZR VHTXHQFH YDULDQWV RI WKH FDWDO\WLF VWUDQG SULPHU ZHUH FRPPRQO\ XVHG EXW ERWK ERXQG WR WKH FRQVWDQW UHJLRQ RI WKH YDULRXV WHPSODWHV fDQJ ULERVHf FDWDO\WLF VWUDQG SULPHU n &7*&$*$$77&7$$7$&*$&7&$&7$7U$**$$*$&$7**&*$&7&7&f f%ULERVHf FDWDO\WLF VWUDQG SULPHU n)***$&*$$77&7$$7$&*$&7&$&7$7U$**$$*$*$7**&*$&$7&7&f 7KH GLIIHUHQFHV EHWZHHQ WKH fDQJf DQG f%-f SULPHUV DUH XQGHUOLQHG UHSUHVHQWV DQ DOLJQPHQW JDS f1f LV HTXLPRODU FRQFHQWUDWLRQV RI HDFK G173 f)f LV D f)OXRUHVFHLQ fU$f LV D ULER$GHQRVLQH 3ULPHU YDULDQWV ZHUH DOVR XVHG LQ ZKLFK WKH ULERDGHQRVLQH ZDV UHSODFHG ZLWK HLWKHU GHR[\DGHQRVLQH fDQJ ULERVHf DQG f%ULERVHff RU FDJHG

PAGE 36

ULERVH fDQJ FDJHf DQG f%FDJHff 7KH FRPSOHPHQWDU\ VWUDQG SULPHU VHTXHQFH ZDV DOZD\V WKH VDPH H[FHSW fFRPSOHPHQWDU\ 7$,/ &r*f LQ ZKLFK WKH f ZDV FKDQJHG WR 7f EXW YDULHG LQ QRQFRGLQJ PRGLILFDWLRQV DSSHQGHG WR WKH fHQG fFRPSOHPHQWDU\ QR f3f SULPHU n*7*&&$$*&77$&&*7&$f fFRPSOHPHQWDU\ f3KRVSKDWHf SULPHU n3*7*&&$$*&77$&&*7&$f fFRPSOHPHQWDU\ fWDLOOf SULPHU n**7***7***FO*7*&&$$*&77$&&*7&$f fFRPSOHPHQWDU\ n7$,/ &f SULPHU n$$$$$$$$$$$$$$$$$$$$FF*7*&&$$*&77$&&*7&$f fFRPSOHPHQWDU\ 7$,/ &f SULPHU n$$$$$$$$$$$$$$$$$$$$&*7*&&$$*&77$&&*7&$f fFRPSOHPHQWDU\ 7$,/ &f SULPHU n $$$$$$$$$$$$$$$FO*7*&&$$*&77$&&*7&$f fFRPSOHPHQWDU\ 7$,/ &r*f SULPHU n$$$$$$$$$$$$$$$$$$$$FF77*&&$$*&77$&&*7&$f fFRPSOHPHQWDU\ 7$,/ &f SULPHU n$$$$$$$$$$$$$$$$$$$$FFF*7*&&$$*&77$&&*7&$f fFRPSOHPHQWDU\ 7$,/ [n20(f SULPHU n$$$$$$$$$$$$$$$$$$$$P8P8P8P8*7*&&$$*&77$&&*7&$f f&Of LV DQ DWRP KH[DHWK\OHQHJO\FROEDVHG VSDFHU f&f LV D DWRP WULHWK\OHQHJO\FROEDVHG VSDFHU fP8f LV D f2PHWK\OXULGLQH f3f LV SKRVSKDWH 7\SLFDO FRQGLWLRQV IRU D S/ 3&5 FRQWDLQHG XS WR QJ WHPSODWH Q0 FDWDO\WLF VWUDQG SULPHU Q0 FRPSOHPHQWDU\ VWUDQG SULPHU S0 G1736 HLWKHU VWDQGDUG $ & DQG 7 RU D fQRQVWDQGDUGf PL[WXUH LQ ZKLFK WK\PLGLQH LV UHSODFHG E\ DPLQRDOO\Of fGHR[\XULGLQH KHQFHIRUWK UHIHUUHG WR DV f(EDVHff P0 .& P0 7ULV+&O S+ f P0 1+A62A P0 0J6&/ b 7ULWRQ ; XQLWV SRO\PHUDVH 7DT RU 9HQW H[Rf DQG S&XULHV DOSKD3&73 IRU LQWHUQDOO\ ODEHOHG VDPSOHVf 3&5

PAGE 37

DPSOLILFDWLRQ FRQVLVWHG RI PLQXWHV DW r& IROORZHG E\ F\FOHV RI VHFRQGV DW r& VHFRQGV DW r& DQG PLQXWHV DW r& 3&5 IRU LQ YLWUR VHOHFWLRQV ,96f YDULHG VOLJKWO\ Q0 RI WKH FDWDO\WLF VWUDQG SULPHU ZDV XVHG WR FRPSHQVDWH IRU KDLUSLQV IRUPHG E\ WKH f%FDJHf SULPHUf XS WR F\FOHV RI 3&5 ZHUH XVHG LQ WKH HDUO\ URXQGV RI ,96 9HQW H[R ZDV XVHG UDWKHU WKDQ 7DT EHFDXVH LW LQFRUSRUDWHV fIXQFWLRQDOL]HG QXFOHRWLGHV VXFK DV (EDVH ZLWK JUHDWHU HIILFLHQF\f DQG SULPHU ELQGLQJ ZDV GRQH DW r& UDWKHU WKDQ r& DJDLQ WR FRPSHQVDWH IRU KDLUSLQV IRUPHG E\ %SULPHUVf 3RLQW PXWDWLRQV LQWURGXFHG ZLWKLQ LQ WKH FDWDO\WLF VWUDQG SULPHU UHJLRQ EXW f RI WKH ULERVHDGHQRVLQH ZHUH LQFRUSRUDWHG XVLQJ D WZR VWHS 3&5 ,Q WKH ILUVW VWHS F\FOHV RI 3&5 ZHUH FRQGXFWHG LQ ZKLFK WKH FDWDO\WLF VWUDQG SULPHU ZDV UHSODFHG ZLWK RQH FRQWDLQLQJ WKH GHVLUHG SRLQW PXWDWLRQ DQG D QXFOHRWLGH WUXQFDWLRQ IURP WKH fHQG 7KH SURGXFW RI WKLV 3&5 ZDV XVHG DV WKH WHPSODWH GLOXWHG f IRU D VHFRQG 3&5 FRQWDLQLQJ D FDWDO\WLF VWUDQG SULPHU WUXQFDWHG WZR QXFOHRWLGHV f RI WKH ULERVHDGHQRVLQH 7KLV WZR VWHS 3&5 PHWKRG DOORZHG IRU WKH LQWURGXFWLRQ RI VHYHUDO SRLQW PXWDWLRQV LQWR WKH FDWDO\WLF SULPHU UHJLRQ ZLWKRXW QHFHVVLWDWLQJ XQLTXH '1$51$ FKLPHULF SULPHUV IRU HDFK PXWDQW 3UHSDUDWLRQ RI 6LQJOH6WUDQGHG '1$]YPHV 'RXEOH VWUDQGHG '1$]\PHV JHQHUDWHG YLD 3&5 XVLQJ WKH fSKRVSKRU\ODWHG IRUP RI WKH FRPSOHPHQWDU\ VWUDQG SULPHU fFRPSOHPHQWDU\ fSKRVSKDWHf ZHUH FRQYHUWHG WR VLQJOHVWUDQGHG '1$ E\ GLJHVWLRQ RI WKH FRPSOHPHQWDU\ VWUDQG XVLQJ ODPEGD H[RQXFOHDVH DQ H[RQXFOHDVH HQ]\PH VSHFLILF IRU fSKRVSKRU\ODWHG GRXEOH VWUDQGHG

PAGE 38

'1$ )ROORZLQJ HWKDQRO SUHFLSLWDWLRQ RI 3&5 SURGXFWV '1$ SHOOHWV ZHUH UHVXVSHQGHG LQ S/ RI H[RQXFOHDVH VROXWLRQ XQLWV ODPEGD H[RQXFOHDVH HQ]\PH P0 JO\FLQH .2+ S+ f P0 0J&A SJP/ %6$f SHU S/ 3&5 6DPSOHV ZHUH PL[HG WKRURXJKO\ DQG LQFXEDWHG DW r& IRU PLQXWHV 5HDFWLRQV ZHUH WHUPLQDWHG E\ DGGLQJ H[FHVV IRUPDPLGH VWRS G\H PJP/ [\OHQH F\DQRO PJP/ EURPRSKHQRO EOXH P0 ('7$ LQ b IRUPDPLGHf DQG KHDWLQJ DW r& IRU PLQXWHV 7KH VLQJOHVWUDQGHG SURGXFWV ZHUH ORDGHG RQ DQ b 3$*(XUHD JHO DQG IXOO OHQJWK VV'1$ SURGXFWV H[FLVHG XVLQJ D VWHULOH UD]RU WKRURXJKO\ ZDVKHG ZLWK HWKDQRO EHWZHHQ HDFK VDPSOHf (DFK JHO VOLFH ZDV FUXVKHG DQG HOXWHG LQ S/ HOXWLRQ EXIIHU P0 1+2$& P0 ('7$ b 6'6f RYHUQLJKW *HO SXULILHG VDPSOHV ZHUH WKHQ H[WUDFWHG LQ SKHQROFKORURIRUPLVRDP\O DOFRKRO WKHQ H[WUDFWHGHG FKRORURIRUPLVRDP\O DOFRKRO DQG SUHFLSLWDWHG LQ DPPRQLXP DFHWDWH DQG HWKDQRO $V\PPHWULF 3&5 ZDV DOVR XVHG WR JHQHUDWH VLQJOH VWUDQGHG '1$]\PHV ,Q WKLV SURFHGXUH WKH FRPSOHPHQWDU\ VWUDQG SULPHU ZDV V\QWKHVL]HG ZLWK D QXFOHRWLGH H[WHQVLRQ XVXDOO\ SRO\DGHQRVLQHf FRQQHFWHG WR WKH fHQG RI WKH SULPHU E\ RQH RI VHYHUDO OLQNHUV & &O f2PHWK\O8ULGLQHf 9HQW DQG 7DT SRO\PHUDVH FDQQRW HIILFLHQWO\ UHDGWKURXJK WKHVH OLQNHUV 7KHUHIRUH WKH FDWDO\WLF VWUDQG JHQHUDWHG E\ H[WHQVLRQ RI WKH FDWDO\WLF SULPHU WHUPLQDWHV DW WKH OLQNHU JHQHUDWLQJ WKH FRUUHFW VL]H IRU IXOOOHQJWK SURGXFW &RPSOHPHQWDU\ VWUDQG PROHFXOHV JHQHUDWHG E\ H[WHQVLRQ RI D FRPSOHPHQWDU\ VWUDQG SULPHU FRQWDLQLQJ WKH ff WDLOf DUH ORQJHU WKDQ WKH IXOOOHQJWK SURGXFW E\ WKH OHQJWK RI WKH WDLO DQG DUH HDVLO\ VHSDUDWHG E\ JHO SXULILFDWLRQ RQ DQ b 3$*(XUHD JHO %HFDXVH 9HQW DQG 7DT SRO\PHUDVH RFFDVLRQDOO\ WHUPLQDWH H[WHQVLRQ DERXW b RI WKH WLPHf ZKHQ HQFRXQWHULQJ D FDJHGULERVH LQ WKH WHPSODWH 3&5 XVLQJ D

PAGE 39

FDJHGULERVH LQ WKH FDWDO\WLF SULPHU DQG D QXFOHRWLGH WDLO RQ WKH FRPSOHPHQWDU\ SULPHU JHQHUDWHG D FRPSOHPHQWDU\ VWUDQG WKDW ZDV HLWKHU QXFOHRWLGHV QXFOHRWLGHV SOXV D QXFOHRWLGH WDLOf RU QXFOHRWLGHV QXFOHRWLGHV SOXV D QXFOHRWLGH WDLO PLQXV QXFOHRWLGHV IROORZLQJ WKH FDJHGULERVHf 7KH GHVLUHG FDWDO\WLF VWUDQG ZDV HDVLO\ LGHQWLILHG DV WKH PLGGOH EDQG DQG FOHDQO\ H[FLVHG RQ DQ b 3$*(XUHD JHO *HO SXULILHG VDPSOHV ZHUH FUXVKHG VRDNHG RYHUQLJKW LQ HOXWLRQ EXIIHU DQG WKHQ H[WUDFWHG LQ SKHQROFKORURIRUPLVRDP\O DOFRKRO H[WUDFWHG DJDLQ LQ FKRORURIRUPLVRDP\O DOFRKRO DQG SUHFLSLWDWHG LQ DPPRQLXP DFHWDWH DQG HWKDQRO f(QG /DEHOLQJ RI '1$ 6LQJOHVWUDQGHG '1$ S0 fZDV fODEHOHG ZLWK S&L JDPPD3$73 XVLQJ XQLWV RI 7 SRO\QXFOHRWLGH NLQDVH P0 7ULV+&O S+ f P0 0J&( P0 GLWKLRWKUHLWRO LQ D ILQDO YROXPH RI S/ 5HDFWLRQV ZHUH LQFXEDWHG IRU PLQXWHV DW r& IROORZHG E\ DGGLWLRQ RI HTXDO YROXPH 7( DQG ILQDOO\ LQDFWLYDWHG E\ LQFXEDWLQJ IRU PLQXWHV DW r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

PAGE 40

QRWHGf DQG GLOXWHG WR WZLFH WKH GHVLUHG ILQDO FRQFHQWUDWLRQ 6DPSOHV ZHUH WKHQ PL[HG ZLWK HTXDO YROXPH ; UHDFWLRQ EXIIHU W\SLFDOO\ 0 1D&O P0 0J&E P0 +(3(6 S+ f 5HDFWLRQV ZHUH WKHQ KHDW GHQDWXUHG DQG UHIROGHG XVLQJ D WKHUPRF\FOHU fVORZ FRROf SURWRFRO KHDWLQJ WR r& IRU PLQXWHV DQG FRROLQJ WR r& RYHU PLQXWHVf )RU UHDFWLRQV ZLWK ULERVHDGHQRVLQH fWLPH ]HURf ZDV FRQVLGHUHG WR EH WKH WLPH DW ZKLFK WKH VDPSOH FRPSOHWHG WKH fVORZ FRROf )RU UHDFWLRQV ZLWK D FDJHGULERVH WKH UHDFWLRQV ZHUH LQLWLDWHG ZKHQ WKH ODVHU GHSURWHFWHG WKH FDJHGULERVH ZLWKLQ RQH KRXU RI WKH FRPSOHWLRQ RI WKH fVORZ FRROff 8QOHVV RWKHUZLVH QRWHG UHDFWLRQV ZHUH FRQGXFWHG DW r& LQFXEDWHG RQ D KRW EORFN VHW WR r& WR SUHYHQW WKH WHPSHUDWXUH IURP GURSSLQJ EHORZ r&f 5HDFWLRQV ZHUH WHUPLQDWHG DW YDULRXV WLPH SRLQWV E\ GLOXWLQJ DQ DOLTXRW RI WKH UHDFWLRQ LQ H[FHVV IRUPDPLGH VWRS G\H DQG WKHQ IUR]HQ r&f 7LPH SRLQWV ZHUH UXQ RQ DQ b 3$*(XUHD VHTXHQFLQJ JHOV DQG WKH SHUFHQW FOHDYHG TXDQWLILHG XVLQJ D %LR5DG SKRVSKRULPDJHU &ORQLQJ DQG 6HTXHQFLQJ '1$]\PHV 3ULRU WR FORQLQJ VLQJOHVWUDQGHG '1$]\PHV ZHUH ILUVW FRQYHUWHG WR GRXEOH VWUDQGHG '1$ E\ 3&5 DPSOLILFDWLRQ XVXDOO\ ZLWK RQO\ WKH FRPSOHPHQWDU\ VWUDQG SULPHU fFRPSOHPHQWDU\ QR fSKRVSKDWHff 3&5 FRQGLWLRQV ZHUH r& IRU PLQXWHV IROORZHG E\ WKUHH F\FOHV RI VHFRQGV DW r& IRXU PLQXWH UDPS FRRO WR r& SOXV VHFRQGV DW r& DQG PLQXWH DW r& IROORZHG E\ PLQXWHV DW r& :KHQ FORQLQJ FOHDYHG DQG XQFOHDYHG WKH GHVLUHG VLQJOHVWUDQGHG '1$ ZDV XVXDOO\f ILUVW LVRODWHG YLD JHO SXULILFDWLRQ 3&5 DPSOLI\LQJ WKH XQSXULILHG UHDFWLRQ PL[WXUH FRQWDLQLQJ ERWK FOHDYHG DQG XQFOHDYHG ZLWK RQO\ WKH fFRPSOHPHQWDU\ QR fSKRVSKDWHf SULPHU \LHOGHG RQO\ XQFOHDYHG PROHFXOH FORQHV FOHDYHG PROHFXOHV ZHUH WKHUHIRUH FORQHG E\

PAGE 41

XVLQJ WKH FRPSOHPHQWDU\ VWUDQG SULPHU fFRPSOHPHQWDU\ QR fSKRVSKDWHf ZLWK WKH f WUXQFDWHG FDWDO\WLF SULPHU ffFOHDYHG ROLJRf )UHVK GRXEOH VWUDQGHG 3&5 SURGXFWV ZHUH FORQHG XVLQJ WKH 7232 7$ &ORQLQJ 6\VWHP ,QYLWURJHQf DQG SODWHG RQWR $JDURVH SODWHV FRQWDLQLQJ DPSLFLOOLQ :KHQ FORQLQJ VHTXHQFHV IURP WKH SRRO RI VXUYLYRUV IURP WKH f+f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r& ZLWK YLJRURXV VKDNLQJ )ROORZLQJ RYHUQLJKW JURZWK P/ RI FHOOV ZHUH SHOOHWHG E\ FHQWULIXJDWLRQ IRU PLQXWH DW J *URZWK PHGLD ZDV UHPRYHG DQG WKH FHOO SHOOHW UHVXVSHQGHG LQ 6ROXWLRQ S/ P0 7ULV S+ P0 ('7$ DQG SJPO 5QDVH $f 6ROXWLRQ ,, S/ P/ IUHVK 0 1D2+ P/ b 6'6 P/ ZDWHUf ZDV WKHQ DGGHG DQG WKH VDPSOHV ZHUH WKHQ LQYHUWHG JHQWO\ WR PL[ 6ROXWLRQ ,,, S/ 0 SRWDVVLXP DFHWDWH S+ f ZDV WKHQ DGGHG IROORZHG E\ JHQWOH PL[LQJ 7KH VDPSOH ZDV FHQWULIXJHG IRU PLQXWHV DW J DIWHU ZKLFK S/ RI FKORURIRUP ZDV DGGHG DQG JHQWO\ PL[HG 7KH VDPSOHV ZHUH WKHQ FHQWULIXJHG DJDLQ DW J DQG S/ RI WKH DTXHRXV VXSHUQDWDQW ZDV WUDQVIHUUHG WR D QHZ WXEH FRQWDLQLQJ LVRSURSDQRO S/f 7KLV VDPSOH ZDV FHQWULIXJHG IRU PLQXWHV DW J 7KH LVRSURSDQRO ZDV FRPSOHWHO\ UHPRYHG DQG WKH UHPDLQLQJ SHOOHW UHVXVSHQGHG LPPHGLDWHO\ LQ S/ ZDWHU IROORZHG E\ WKH DGGLWLRQ RI S/ b 3(* 7KH UHVXVSHQGHG SHOOHW ZDV VWRUHG RQ LFH IRU PLQXWHV RU RYHUQLJKWf

PAGE 42

DIWHU ZKLFK WKH VDPSOH ZDV FHQWULIXJHG IRU PLQXWHV DW J 7KH VXSHUQDWDQW ZDV UHPRYHG DQG WKH SHOOHW ULQVHG ZLWK b HWKDQRO S/f 7KH HWKDQRO ZDV UHPRYHG DQG WKH SHOOHW ZDV GULHG FRPSOHWHO\ XQGHU YDFXXP 7KH UHVXOWLQJ '1$ SHOOHW ZDV UHn VXVSHQGHG LQ S/ RI ZDWHU 7KLV SURWRFRO URXWLQHO\ \LHOGHG QJX/ RI '1$ RI VXIILFLHQW SXULW\ IRU VHTXHQFLQJ UHDFWLRQV 6HTXHQFHV WUDQVIRUPHG LQWR WKH 7232 7$ &ORQLQJ 9HFWRU ZHUH VHTXHQFHG XVLQJ S0 SULPHU ff f&*&&$***7777&&&$*7&$&*$&f ZLWK QJ SODVPLG '1$ WHPSODWH ; ILQDO FRQFHQWUDWLRQ %LJ '\H UHDFWLRQ EXIIHU DQG S/ %LJ '\H 7HUPLQDWRU 6HTXHQFLQJ PL[ LQ D ILQDO YROXPH RI X/ 6DPSOHV ZHUH RYHUOD\HG ZLWK PLQHUDO RLO DQG DPSOLILHG XVLQJ F\FOHV RI 3&5 DW r& IRU VHFRQGV r& IRU VHFRQGV DQG r& IRU PLQXWHV 8QLQFRUSRUDWHG QXFOHRWLGHV ZHUH UHPRYHG E\ VSLQQLQJ S/ RI WKH UHFRYHUHG 3&5 UHDFWLRQ WKURXJK D FROXPQ RI S/ 6HSKDGH[ DW J IRU PLQXWHV 7KH UHFRYHUHG VDPSOH ZDV WKHQ GULHG WR FRPSOHWLRQ UHn VXVSHQGHG LQ S/ IRUPDPLGH VHTXHQFLQJ EXIIHU KHDWHG WR r& IRU PLQXWHV DQG TXLFNO\ WUDQVIHUUHG WR LFH 6HTXHQFLQJ VDPSOHV ZHUH DQDO\]HG RQ DQ $SSOLHG %LRV\VWHPV 3ULVP *HQHWLF $QDO\]HU 6HTXHQFLQJ UHVXOWV ZHUH FRQILUPHG E\ H[DPLQLQJ WKH FKURPDWRJUDPV PDQXDOO\ XVLQJ WKH f6HTXHQFKHUf VRIWZDUH SDFNDJH ,Q 9LWUR 6HOHFWLRQ '1$]\PH OLEUDULHV IRU WKH ILUVW URXQG RI VHOHFWLRQ ZHUH SUHSDUHG E\ D VLQJOH F\FOH RI UXQRII 3&5 XVLQJ WKH OLEUDU\ WHPSODWH f*7*&&$$*&77$&&*7&$&1 *$*$7*7&*&&$7&7&77&& ZKHUH 1 LQGLFDWHV HTXDO PRODU FRQFHQWUDWLRQV RI $ 7 DQG &f DQG RQH RI WKUHH GLIIHUHQW FDWDO\WLF VWUDQG SULPHUV fDQJ ULERVHf f%FDJHf RU f%GHSURWHFWHG FDJHf WKH f%GHSURWHFWHG FDJHf ZDV WKH VDPH SULPHU

PAGE 43

DV f%FDJHf EXW WKH FDJHGULERVH KDG EHHQ GHSURWHFWHG E\ SUHYLRXV H[SRVXUH WR SXOVHV RI ODVHU P-SXOVHff (DFK SULPHU ZDV H[WHQGHG XVLQJ HLWKHU WKH VWDQGDUG G173V HTXLPRODU 7 $ DQG &f RU ZLWK D QRQVWDQGDUG SRVLWLRQIXQFWLRQDOL]HG QXFOHRWLGH WULSKRVSKDWH DPLQRDOO\OffGHR[\XULGLQH fWULSKRVSKDWH UHIHUUHG WR DV (EDVHf VXEVWLWXWHG IRU 7WULSKRVSKDWH HTXLPRODU ( $ &f 7KLV \LHOGHG VL[ GLIIHUHQW 96 VWDUWLQJ OLEUDULHV OLEDU\ f%FDJHf SULPHU VWDQGDUG G1736 OLEDU\ fDQJ ULERVHf SULPHU VWDQGDUG G1736 OLEDU\ f%FDJHf SULPHU $*&( OLEDU\ fDQJ ULERVHf SULPHU $*&( OLEDU\ f%GHSURWHFWHG FDJHf SULPHU VWDQGDUG G1736 OLEDU\ f%GHSURWHFWHG FDJHf SULPHU $*&(f 5XQRII UHDFWLRQV YROXPHV UDQJHG IURP WKUHH WR PLOOLOLWHUV HDFK FRQWDLQLQJ QJS/ OLEUDU\ WHPSODWH Q0 FDWDO\WLF VWUDQG SULPHU S0 G173V P0 .& P0 7ULV+&O S+ f P0 1+f6 P0 0J6 b 7ULWRQ ; XQLWVP/ 9HQW H[R DQG S&XULHVP/ DOSKD 3&73 2QH PLOOLOLWHU RI HDFK VDPSOH PL[WXUH PLQXV SRO\PHUDVHf ZDV DOLTXRWHG LQWR P/ HSSHQGRUI WXEHV KHDWHG DW r& IRU PLQXWHV DQG WKHQ VORZO\ FRROHG WR r& RYHU D SHULRG RI PLQXWHV 3RO\PHUDVH ZDV WKHQ DGGHG DQG WKH VDPSOHV ZHUH WUDQVIHUUHG WR r& IRU PLQXWHV 7KH VDPSOHV ZHUH WKHQ HWKDQRO SUHFLSLWDWHG ZLWK DPPRQLXP DFHWDWH DQG ZLWK JO\FRJHQ DV D FDUULHU E\ VWRULQJ RYHUQLJKW DW r& DQG WKHQ FHQWULIXJLQJ IRU PLQXWHV LQ &RUH[ JODVV WXEHV DW USPV LQ D %HFNPDQ FHQWULIXJH DW r& 7KH HWKDQRO ZDV UHPRYHG DQG WKH SHOOHW ZDV UHVXVSHQGHG LQ S/ ZDWHU 7KH VDPSOHV ZHUH HWKDQRO SUHFLSLWDWHG D VHFRQG WLPH LQ ZLWK DPPRQLXP DFHWDWH E\ FHQWULIXJLQJ DW J LQ D GHVNWRS FHQWULIXJH IRU PLQXWHV DW r& $IWHU UHPRYLQJ WKH HWKDQRO WKH SHOOHW ZDV UHVXVSHQGHG LQ S/ IRUPDPLGH VWRS G\H DQG ZDUPHG DW r& IRU PLQXWHV WR FRPSOHWHO\ GLVVROYH WKH SHOOHW

PAGE 44

7KH VDPSOHV ZHUH WKHQ UXQ RQ DQ b 3$*(XUHD JHO DQG WKH IXOOOHQJWK SURGXFWV H[FLVHG XVLQJ D VWHULOH UD]RU ULQVHG WKRURXJKO\ ZLWK HWKDQRO EHWZHHQ VDPSOHVf (DFK JHO VOLFH ZDV FUXVKHG DQG HOXWHG LQ S/ HOXWLRQ EXIIHU P0 1+2$& P0 ('7$ b 6'6f RYHUQLJKW 6DPSOHV ZHUH H[WUDFWHG LQ SKHQROFKORURIRUPLVRDP\O DOFRKRO f WKHQ H[WUDFWHG LQ FKORURIRUUPLVRDP\O DOFRKRO f DQG SUHFLSLWDWHG LQ HWKDQRO ZLWK DPPRQLXP DFHWDWH 3HOOHWV ZHUH UHVXVSHQGHG LQ P0 +(3(6 S+ f DQG WUDQVIHUUHG WR D QHZ WXEH DW ZKLFK SRLQW WKH ILQDO FRQFHQWUDWLRQ ZDV HVWLPDWHG XVLQJ VSHFLILF UDGLRDFWLYLW\ $OO VDPSOHV ZHUH EURXJKW XS WR Q0 H[FHSW OLEUDU\ f%FDJHf SULPHU H[WHQGHG XVLQJ VWDQGDUG G1736 ZDV EURXJKW XS WR Q0f $Q DOLTXRW S/f RI OLEDU\ ZDV VHW DVLGH IRU WKH ,96 FRQGLWLRQ f(f FRQWDLQLQJ WHQ WLPHV WKH VWDUWLQJ DPRXQW RI OLEUDU\f DQG WKH UHPDLQLQJ OLEUDU\ ZDV GLOXWHG WR Q0 $Q DOLTXRW S/f RI HDFK VDPSOH ZDV WKHQ PL[HG ZLWK HTXDO YROXPH RI WKH DSSURSULDWH ; UHDFWLRQ EXIIHU DV VXPPDUL]HG LQ 7DEOH 7KH VDPSOHV ZHUH TXLFNO\ KHDWHG WR r& IRU PLQXWHV DQG FRROHG WR r& RYHU PLQXWHV )RU VDPSOHV ZLWKRXW D FDJHGULERVH + 1 DQG 2f WKH FRPSOHWLRQ RI WKH VORZ FRRO WR r& ZDV FRQVLGHUHG WKH LQLWLDWLRQ RI UHDFWLRQ $OO VDPSOHV ZHUH WKHQ H[SRVHG WR SXOVHV RI ODVHU P-SXOVHf 7KH WLPH RI ODVHU H[SRVXUH ZDV FRQVLGHUHG WKH WLPH RI UHDFWLRQ LQLWLDWLRQ IRU DOO VDPSOHV FRQWDLQLQJ D FDJHGULERVH 6DPSOHV ZHUH DOORZHG WR UHDFW DW r& IRU YDULRXV WLPHV WR PLQXWHV VHH 7DEOH f DQG WKHQ VWRSSHG E\ DGGLQJ YROXPH RI IRUPDPLGH VWRS G\H 7KH VDPSOHV ZHUH WKHQ UXQ RQ DQ b 3$*(XUHD JHO QH[W WR D fFOHDYHG OLEUDU\f PDUNHU 7KH DUHD FRUUHVSRQGLQJ WR WKH VL]H RI WKH FOHDYHG OLEUDU\ IRU HDFK VDPSOH ZDV H[FLVHG FUXVKHG DQG HOXWHG RYHUQLJKW UHIHUUHG WR DV fVXUYLYRUV RI URXQG ff

PAGE 45

7DEOH 6HOHFWLRQ &RQGLWLRQV IRU (DFK RI WKH )LIWHHQ 3DUDOOHO ,96 ([SHULPHQWV )LQDO %XIIHU 6DPSOH %DVHV $*&f SULPHU P0 +(3(6 S+ 0 1D&, P0 RI PHWDO 6HOHFWLRQ 7LPH PLQXWHVf $ 7EDVH %FDJH % 7EDVH %FDJH =Q & 7EDVH %FDJH 0J 7EDVH %FDJH 0Ja ( 7EDVH O2[ >OLEUDU\@ %FDJH 0J ) 7EDVH %FDJH 0J 7EDVH %GHSURWHFWHG FDJH 0JA + 7EDVH DQJ ULERVH 0J (EDVH %FDJH (EDVH %FDJH =Q (EDVH %FDJH 0Jfr / (EDVH %FDJH 0Jnrfr 0 (EDVH %FDJH 0JA 1 (EDVH %GHSURWHFWHG FDJH 0Jrfr 2 (EDVH DQJ ULERVH 0Jfn )ROORZLQJ HOXWLRQ VDPSOHV ZHUH H[WUDFWHG LQ SKHQROFKORURIRUPLVRDP\O DOFRKRO WKHQ H[WUDFWHG LQ FKORURIRUPLVRDP\O DOFRKRO DQG PL[HG ZLWK DPPRQLXP DFHWDWH DQG HWKDQRO +DOI RI HDFK VDPSOH ZDV VWRUHG LQ HWKDQRO DQG WKH UHPDLQLQJ KDOI ZDV SUHFLSLWDWHG DQG UHVXVSHQGHG LQ S/ ZDWHU )RU VXEVHTXHQW URXQGV RI VHOHFWLRQ WKLUW\ PLFUROLWHUV RI WKH VXUYLYRUV RI WKH SUHYLRXV URXQG LQ ZDWHUf ZHUH XVHG LQ D S/ 3&5 FRQWDLQLQJ Q0 FDWDO\WLF VWUDQG SULPHU HLWKHU fDQJ ULERVHf f%FDJHf RU f%GHSURWHFWHG FDJHff Q0 FRPSOHPHQWDU\ VWUDQG SULPHU fFRPSOHPHQWDU\ f

PAGE 46

SKRVSKDWHff S0 G1736 HLWKHU $*&7 RU $*&(f P0 .& P0 7ULV+&O S+ f P0 1+A62A P0 0J6 b 7ULWRQ ; DQG XQLWV 9HQW H[R SRO\PHUDVH DQG S&XULHV DOSKD3&73 3&5 DPSOLILFDWLRQ FRQVLVWHG RI PLQXWHV DW r& IROORZHG E\ YDULRXV QXPEHUV RI F\FOHV RI VHFRQGV DW r& VHFRQGV DW r& DQG PLQXWHV DW r& D PLQXWH H[WHQVLRQ ZDV GHWHUPLQHG WR EH VXIILFLHQWO\ ORQJ VR DV WR \LHOG DSSUR[LPDWHO\ HTXDO LQFRUSRUDWLRQ HIILFLHQF\ IRU ERWK VWDQGDUG DQG QRQn VWDQGDUG QXFOHRWLGHVf )ROORZLQJ 3&5 VDPSOHV ZHUH H[WUDFWHG LQ SKHQROFKORURIRUPLVRDP\O DOFRKRO WKHQ H[WUDFWHG FKORURIRUPLVRDP\O DOFRKRO DQG SUHFLSLWDWHG ZLWK DPPRQLXP DFHWDWH DQG HWKDQRO 6DPSOHV ZHUH UHVXVSHQGHG LQ H[RQXFOHDVHEXIIHU VROXWLRQ S/ ILQDO YROXPH XQLWV ODPEGD H[RQXFOHDVH HQ]\PH P0 JO\FLQH.2+ S+ f P0 0J&O SJP/ %6$f DQG LQFXEDWHG DW r& IRU PLQXWHV 5HDFWLRQV ZHUH WHUPLQDWHG E\ DGGLQJ S/ IRUPDPLGH VWRS G\H DQG KHDWLQJ DW r& IRU PLQXWHV 7KH VLQJOH VWUDQGHG SURGXFWV ZHUH ORDGHG RQ DQ b 3$*(XUHD JHO DQG IXOO OHQJWK VV'1$ SURGXFWV H[FLVHG *HO SXULILHG VDPSOHV ZHUH WKHQ H[WUDFWHG LQ SKHQROFKORURIRUPLVRDP\O DOFRKRO WKHQ H[WUDFWHG FKORURIRUPLVRDP\O DOFRKRO DQG SUHFLSLWDWHG LQ DPPRQLXP DFHWDWH DQG HWKDQRO 6DPSOHV ZHUH UHVXVSHQGHG LQ P0 +(3(6 EXIIHU S+ f WR D FRQFHQWUDWLRQ RI Q0 (DFK VDPSOH S/f ZDV WKHQ PL[HG ZLWK HTXDO YROXPH RI ; UHDFWLRQ EXIIHU VHH 7DEOH IRU EXIIHU FRQGLWLRQV IRU HDFK VDPSOHf KHDWHG DQG VORZO\ FRROHG H[SRVHG WR ODVHU SXOVHV P-SXOVHf DQG VWRSSHG ZLWK IRUPDPLGH VWRS G\H IROORZLQJ YDULRXV LQFXEDWLRQ WLPHV VHH 7DEOH IRU VHOHFWLRQ WLPH XVHG IRU HDFK VDPSOHf 7KLV SURFHGXUH ZDV UHSHDWHG IRU HDFK URXQG RI VHOHFWLRQ

PAGE 47

&+$37(5 5(68/76 2) $1$/<6,6 5HVHDUFK 2EMHFWLYHV 8QGHUVWDQGLQJ ZK\ DGGHG IXQFWLRQDOLW\ GRHV QRW DSSHDU WR VLJQLILFDQWO\ LPSURYH WKH FDWDO\WLF SRZHU RI QXFOHLF DFLG HQ]\PHV XQGHU LQ YLWUR VHOHFWLRQ ,96f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f DQG DSSDUHQWO\ LQVXIILFLHQW WR H[SODLQ WKH RULJLQV RI OLIH WKLV PRGHVW VXFFHVV RI SUHYLRXV ,96 PD\ LQ SDUW EH DWWULEXWDEOH WR VHYHUDO GHWDLOV RI WKH ,96 H[SHULPHQWDO V\VWHP 2QH VWURQJ SRVVLELOLW\ LV WKDW WKH IDVWHVW FDWDO\VWV DUH ORVW LQ WKH H[SHULPHQWDO ZRUNXS 2XU DLP LV WZR IROG Df ([DPLQH WKH H[SHULPHQWDO V\VWHPV LQ VXIILFLHQW ULJRU VR DV WR DOORZ PDWKHPDWLFDO PRGHOLQJ DQG Ef LPSURYH WKH VHOHFWLRQ SURFHGXUH VR DV WR PLQLPL]H WKH SRWHQWLDO ORVV RI FDWDO\VWV DQG DOORZ TXDQWLWDWLYH DQDO\VLV 7R GHYHORS D V\VWHP WKDW VXSSRUWV GHWDLOHG PDWKHPDWLFDO PRGHOLQJ RI '1$]\PH VHOHFWLRQV ZH ILUVW H[DPLQH LQ GHWDLO D PRGHO '1$]\PH NQRZQ DV LVRODWHG IURP D SUHYLRXV ,96f 7R PLQLPL]H SRWHQWLDO ORVV RI IDVW FDWDO\VWV ZH H[SORUH WKH XVH OLTXLG SKDVH JHO

PAGE 48

HOHFWURSKRUHVLVEDVHG VHOHFWLRQ DQG WKH XVH RI DQ RUWKRQLWUREHQ]\OSURWHFWHG ULERVH fFDJHGULERVHff ZKLFK FDQ EH TXLFNO\ FRQYHUWHG WR WKH XQSURWHFWHG ULERVH E\ ODVHU $Q ,96 LV SHUIRUPHG XVLQJ WKHVH QHZ SURFHGXUHV 3UHOLPLQDU\ PRGHOLQJ RI WKH H[SHULPHQWDO V\VWHP LV GRQH WR GHWHUPLQH KRZ YDULRXV SDUDPHWHUV RI WKH H[SHULPHQWDO V\VWHP KDYH DQ LPSDFW RQ WKH PDWKHPDWLFDO PRGHO DQG KRZ WKH H[SHULPHQWDO V\VWHP FDQ EH ILQHWXQHG WR LPSURYH PRGHOLQJ DFFXUDF\ 'HYHORSLQJ 0HWKRGV IRU 3UHSDULQJ 6LQJOHVWUDQGHG '1$]\PHV E\ ([RQXFOHDVH 'HJUDGDWLRQ RI f3KRVSKRU\ODWHG &RPSOHPHQWDU\ 6WUDQG &ROXPQ SXULILFDWLRQ RI VLQJOHVWUDQGHG '1$]\PHV SUHVHQW QXPHURXV SUREOHPV DV RXWOLQHG LQ WKH LQWURGXFWLRQ 7KH DELOLW\ WR HQ]\PDWLFDOO\ UHPRYH WKH FRPSOHPHQWDU\ VWUDQG ZDV WHVWHG XVLQJ ODPEGD H[RQXFOHDVH /DPEGD H[RQXFOHDVH LV D SURFHVVLYH f WR f '1$ H[RQXFOHDVH VSHFLILF IRU fSKRVSKRU\ODWHG '1$ 'RXEOH VWUDQGHG '1$]\PH ZDV JHQHUDWHG ZLWK 3&5 XVLQJ D fSKRVSKRU\ODWHG FRPSOHPHQWDU\ VWUDQG SULPHU fFRPSOLPHQWDU\ f3ff DQG D QRQSKRVSKRU\ODWHG FDWDO\WLF VWUDQG SULPHU HLWKHU fDQJ ULERVHf RU fDQJ FDJHff 6DPSOHV ZHUH LQWHUQDOO\ ODEHOHG ZLWK DOSKD 3 &73 WR DOORZ YLVXDOL]DWLRQ RI ERWK FDWDO\WLF DQG FRPSOHPHQWDU\ VWUDQGV (DFK RI WKH WZR VDPSOHV ZDV WKHQ WUHDWHG ZLWK RU ZLWKRXW H[RQXFOHDVH DQGRU VWURQJ EDVH ZKLFK FOHDYHV XQSURWHFWHG ULERVHf DQG UXQ RQ D GHQDWXULQJ 3$*(XUHD JHO )LJXUH VKRZV WKH UHVXOWV RI WKLV WUHDWPHQW DOO ILJXUHV DW DUH DW WKH HQG RI WKH FKDSWHUf %RWK XQWUHDWHG ULERVH DQG FDJHG GRXEOH VWUDQGHG 3&5 SURGXFWV VKRZ WZR EDQGV WKH XSSHU FRUUHVSRQGLQJ WR WKH IXOO VL]H SURGXFW DQG WKH ORZHU EDQG FRUUHVSRQGLQJ URXJKO\ WR WKH VL]H RI WKH FOHDYHG SURGXFW ODQH DQG f 7KH DPRXQW RI WKH ORZHU EDQG LV VLJQLILFDQWO\ JUHDWHU LQ WKH 3&5 SURGXFW JHQHUDWHG XVLQJ WKH FDJHGULERVH SULPHU

PAGE 49

FRPSDUH ODQH DQG f 7KLV VXJJHVWV WKDW WKH ORZHU EDQG LQ ODQH DOWKRXJK WKH VL]H RI WKH FDWDO\WLF VWUDQG FOHDYHG DW WKH ULERVH LV QRW DFWXDOO\ FOHDYHG SURGXFW IURP WKH FDWDO\WLF VWUDQG 7KLV EDQG LV DFWXDOO\ WKH UHVXOW RI SRO\PHUDVH WHUPLQDWLRQ ZKHQ WKH FRPSOHPHQWDU\ VWUDQG H[WHQVLRQ UHDFKHV WKH FDJHGULERVH LQ WKH WHPSODWH 7KLV LV FRQILUPHG E\ WKH REVHUYDWLRQ WKDW WKLV EDQG LV GLPLQLVKHG E\ H[RQXFOHDVH WUHDWPHQW VSHFLILF IRU WKH FRPSOHPHQWDU\ VWUDQG ODQH f :KHQ WKH ULERVH GRXEOH VWUDQGHG 3&5 SURGXFW LV WUHDWHG ZLWK VWURQJ EDVH WKHUH LV D VLJQLILFDQW LQFUHDVH LQ WKH ORZHU EDQG FRUUHVSRQGLQJ WR FRQYHUVLRQ RI WKH IXOO OHQJWK FDWDO\WLF VWUDQG WR FOHDYHG SURGXFW &RPSDULQJ ODQH ZLWK ODQH VKRZV WKDW DSSUR[LPDWHO\ KDOI RI WKH ULERVH GRXEOH VWUDQGHG 3&5 SURGXFW LV FRQYHUWHG WR FOHDYHG SURGXFW IROORZLQJ WUHDWPHQW ZLWK VWURQJ EDVH FRUUHVSRQGLQJ WR WKH FOHDYDJH RI WKH FDWDO\WLF VWUDQGf ZKLOH DSSUR[LPDWHO\ KDOI UHPDLQV IXOOOHQJWK FRUUHVSRQGLQJ WR WKH FRPSOHPHQWDU\ VWUDQGf 7KH FDWDO\WLF VWUDQG DSSUR[LPDWHO\ KDOI RI RULJLQDOO\ GRXEOH VWUDQGHG ULERVH 3&5 SURGXFW UHPDLQV IXOOOHQJWK IROORZLQJ H[RQXFOHDVH WUHDWPHQW ODQH f DOWKRXJK WKHUH LV QR LQFUHDVH LQ WKH FOHDYHG SURGXFW WKH FRPSOHPHQWDU\ VWUDQG LV GHJUDGHG WR VLQJOH QXFOHRWLGHVf 7KH IXOOOHQJWK EDQG UHPDLQLQJ IROORZLQJ WUHDWPHQW RI ULERVH ZLWK H[RQXFOHDVH GLVDSSHDUV ZKHQ DOVR WUHDWHG ZLWK VWURQJ EDVH ODQH f 7KLV IXOOOHQJWK EDQG UHPDLQV IRU FDJHG WUHDWHG ZLWK H[RQXFOHDVH DQG VWURQJ EDVH ODQH f GHPRQVWUDWLQJ WKDW FRQYHUVLRQ RI WKH XSSHU EDQG WR FOHDYDJH SURGXFW VHHQ LQ ODQH FRUUHVSRQGV WR WKH FDWDO\WLF VWUDQG WKHUH LV VOLJKW FRQYHUVLRQ RI IXOOOHQJWK FDJHG WR FOHDYHG SURGXFW XQGHU VWURQJ EDVH SRVVLEO\ WKH UHVXOW RI SDUWLDO GHSURWHFWLRQ XQGHU URRP OLJKWf 7KH DEVHQFH RI DQ\ IXOOOHQJWK SURGXFW LQ ODQH GHPRQVWUDWHV WKDW QHDUO\ WKH

PAGE 50

HQWLUH FDWDO\WLF VWUDQG KDV EHHQ FRQYHUWHG WR WKH FOHDYHG SURGXFW E\ WKH VWURQJ EDVH DQG QHDUO\ WKH HQWLUH FRPSOHPHQWDU\ VWUDQG KDV EHHQ GHJUDGHG E\ H[RQXFOHDVH $V\PPHWULF 3&5 ZLWK f7DLOVf )ROORZHG E\ *HO 3XULILFDWLRQ $ QRQHQ]\PDWLF PHWKRG ZDV DOVR SXUVXHG IRU LVRODWLRQ RI VLQJOHVWUDQGHG '1$]\PHV ,Q WKLV SURFHGXUH WHUPHG fDV\PPHWULF 3&5f WKH FRPSOHPHQWDU\ VWUDQG SULPHU ZDV DSSHQGHG ZLWK D fWDLOf FRQVLVWLQJ RI QXFOHRWLGHV FRQQHFWHG WR WKH f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f 3ODEHOHG SULPHUV DQG 7DT SRO\PHUDVH 3&5 DPSOLILFDWLRQV ZHUH GRQH LQ SDLUV RGGQXPEHUHG ODQHV VKRZ WKH SURGXFW JHQHUDWHG XVLQJ WKH f 3ODEHOHG FDWDO\WLF VWUDQG SULPHU fDQJ ULERVHff ZLWK XQODEHOHG FRPSOHPHQWDU\ VWUDQG SULPHU GHVLJQDWHG LQ WKH VXEVHTXHQW ODQHf ZKLOH HYHQQXPEHUHG ODQHV VKRZ WKH SURGXFW JHQHUDWHG ZLWK XQODEHOHG FDWDO\WLF SULPHU DQG f 3ODEHOHG FRPSOHPHQWDU\ VWUDQG SULPHU $OO SURGXFWV JHQHUDWHG IURP FRPSOHPHQWDU\ SULPHUV ZLWK WDLOV DUH LQFUHDVHG LQ VL]H E\ WKH OHQJWK RI WKH WDLO ZKLOH SURGXFWV JHQHUDWHG IURP WKH FDWDO\WLF VWUDQG SULPHU DUH RI QRUPDO VL]H QXFOHRWLGHVf IRU HDFK FRPSOHPHQWDU\ VWUDQG SULPHU H[FHSW f&Of ODQH DQG f 7KLV GHPRQVWUDWHV WKDW 7DT SRO\PHUDVH GRHV LQ IDFW WHUPLQDWH HIIHFWLYHO\ ZKHQ LW HQFRXQWHUV D OLQNHU LQ WKH WHPSODWH IRU DOO OLQNHUV WHVWHG H[FHSW

PAGE 51

f&Of ODQH f ,W LV EHOLHYHG WKDW WKH FRPSOHPHQWDU\ SULPHU FRQWDLQLQJ WKH f&Of OLQNHU GLG QRW FDXVH WHUPLQDWLRQ RI WKH FDWDO\WLF VWUDQG H[WHQVLRQ GXH WR LQFRUUHFW V\QWKHVLV RI WKH FRPSOHPHQWDU\ VWUDQG SULPHU WKH VDPH OLQNHU ZDV XVHG LQ &O DQG WKLV VXFFHVVIXOO\ WHUPLQDWHG H[WHQVLRQf 7KH VL]H GLIIHUHQWLDO EHWZHHQ FDWDO\WLF DQG FRPSOHPHQWDU\ VWUDQGV JHQHUDWHG E\ DV\PPHWULF 3&5 ZDV WKHQ XVHG WR LVRODWH WKH WZR VWUDQGV RQ DQ b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f 7KHUHIRUH DV\PPHWULF 3&5 XVLQJ D FRPSOHPHQWDU\ VWUDQG SULPHU ZLWK D QXFOHRWLGH WDLO JHQHUDWHV WKUHH EDQGV D QW EDQG FRUUHVSRQGLQJ WR WKH QRUPDO VL]HG FDWDO\WLF VWUDQG D QW EDQG FRUUHVSRQGLQJ WR WKH IXOOOHQJWK FRPSOHPHQWDU\ VWUDQG

PAGE 52

SOXV WDLO DQG D QW EDQG UHVXOWLQJ IURP WHUPLQDWLRQ RI WKH FRPSOHPHQWDU\ VWUDQG SOXV WDLO DW WKH FDJHG ULERVH '1$]\PH ZLWK XQFDJHG DQG FDJHGULERVH $FKLHYLQJ PD[LPDO GHSURWHFWLRQ RI FDJHG ULERVH $ FDJHGULERVH ZDV V\QWKHVL]HG DQG LQFRUSRUDWHG LQWR WKH FDWDO\WLF VWUDQG SULPHU NQRZQ DV fDQJ FDJHf DQ LGHQWLFDO SULPHU ZDV V\QWKHVL]HG ZLWK DQ XQSURWHFWHG ULERVH NQRZQ DV fDQJ ULERVHf 8VLQJ WKLV FDJHGULERVH ZH DUH DEOH WR SUHYHQW FOHDYDJH DW WKH ULERVH GXULQJ WKH ZRUN XS SKDVH 7KLV ZDV GHPRQVWUDWHG E\ LQFXEDWLQJ WKH FDJHGULERVH SULPHU fDQJ FDJHf LQ 0 1D2+ DW r& IRU RQH KRXU UHVXOWLQJ LQ RQO\ b FRQYHUVLRQ WR WKH FOHDYHG IRUP YLD EDVH K\GURO\VLVf WKH VDPH FRQGLWLRQV IRU WKH XQSURWHFWHGULERVH SULPHU fDQJ ULERVHf UHVXOWHG LQ b bf FOHDYDJH 7KH REVHUYDWLRQ WKDW DERXW b RI WKH XQSURWHFWHGULERVH SULPHU GRHV QRW FRQYHUW WR WKH FOHDYHG IRUP XQGHU EDVH K\GURO\VLV FRQGLWLRQV VXJJHVW WKDW DV PXFK DV b RI WKH fDQJ ULERVHf SULPHU PD\ EH PLVVLQJ WKH ULERVHDGHQRVLQH GXH WR LPSHUIHFWLRQV LQ WKH '1$ V\QWKHVLV DQG SXULILFDWLRQf 7R WHVW WKH DELOLW\ WR GHSURWHFW WKH FDJHGULERVH fDQJ FDJHf ZDV H[SRVHG WR LQFUHDVLQJ QXPEHU RI SXOVHV RI H[FLPHU ODVHU IUHTXHQF\ Q0 HQHUJ\ P-SXOVHf WR LQGXFH FRQYHUVLRQ IURP WKH SURWHFWHG WR XQSURWHFWHG ULERVH 7KH HIILFLHQF\ RI FRQYHUVLRQ WR XQSURWHFWHG ULERVH ZDV DVVHVVHG E\ LQFXEDWLQJ XQGHU EDVH K\GURO\VLV FRQGLWLRQV )LJXUH f 7KH DPRXQW RI FDJHGSULPHU FOHDYHG E\ VWURQJ EDVH LQFUHDVHG ZLWK LQFUHDVLQJ ODVHU UHDFKLQJ D SODWHDX RI b FOHDYHG ZLWK SXOVHV RI ODVHU $GGHG ODVHU EH\RQG SXOVHV GLG QRW IXUWKHU LQFUHDVH WKH DPRXQW RI FOHDYHG FDJHG

PAGE 53

7KH DPRXQW RI FOHDYHG fDQJ FDJHf YLD EDVH K\GURO\VLVf DW PD[LPXP GHSURWHFWLRQ SXOVHV RI ODVHUf ZDV FRQVLVWHQWO\ DERXW b ORZHU WKDQ WKH FOHDYDJH REWDLQHG IRU fDQJ ULERVHf 7KLV b UHGXFWLRQ LQ FOHDYDJH IRU fDQJ FDJHf PXVW EH WKH UHVXOW RI HLWKHU WZR IDFWRUV LQFRPSOHWH GHSURWHFWLRQ RU PLVVLQJ WKH FDJHGULERVH DGHQRVLQH $OWKRXJK FOHDYDJH RI FDJHG SODWHDXV ZLWK SXOVH RI ODVHU VKRZLQJ QR IXUWKHU FOHDYDJH HYHQ ZLWK SXOVHV RI ODVHU LW LV VWLOO SRVVLEOH WKDW VRPH FDJHGULERVH PROHFXOHV KDYH QRW EHHQ KLW E\ D ODVHU SKRWRQ 7KH RWKHU SRVVLEOH H[SODQDWLRQ IRU UHGXFHG FOHDYDJH RI GHSURWHFWHG FDJHG LV WKH DEVHQFH RI WKH FDJHGULERVH DGHQRVLQH DJDLQ DV D UHVXOW RI LPSHUIHFW SULPHU V\QWKHVLV DQG SXULILFDWLRQ $V PXFK DV b RI WKH fDQJ ULERVHf SULPHU LV EHOLHYHG WR EH PLVVLQJ WKH ULERVHDGHQRVLQH DERYHf EXW WKH FDJHGDGHQRVLQH SKRVSKRUDPLGLWH PD\ EH LQFRUSRUDWHG LQWR WKH SULPHU PRUH SRRUO\ WKDQ WKH ULERVHDGHQRVLQH SKRVSKRUDPLGLWH UHVXOWLQJ LQ DV PXFK DV b RI WKH fDQJ ULERVHf SULPHU PLVVLQJ WKH FDJHGDGHQRVLQH 7KLV LVVXH LV UHYLVLWHG LQ D IROORZLQJ VHFWLRQ f&ORQLQJ DQG 6HTXHQFLQJ f .LQHWLF SURILOH RI ZLWK XQFDJHG DQG FDJHGULERVH 7KH fDQJ ULERVHf DQG fDQJ FDJHf SULPHUV ZHUH LQFRUSRUDWHG LQWR YLD 3&5 7KH FDWDO\WLF VWUDQG RI WKH 3&5 SURGXFWV FDOOHG fULERVHf RU fFDJHGf ZHUH SXULILHG ILUVW E\ H[RQXFOHDVH GHJUDGDWLRQ RI WKH FRPSOHPHQWDU\ VWUDQG ZKLFK ZDV fSKRVSKRU\ODWHGf IROORZHG E\ JHO SXULILFDWLRQ b 3$*(XUHDf 7KH DELOLW\ RI fULERVHf DQG FDJHG '1$]\PHV WR VHOIFOHDYH LQ UHDFWLRQ EXIIHU 0 1D&O P0 0J&E P0 +(3(6 S+ f ZDV WHVWHG 3XULILHG FDJHG ZDV UHVXVSHQGHG LQ UHDFWLRQ EXIIHU KHDWHG WR r& IRU PLQXWHV DQG FRROHG WR r& RYHU PLQXWHV WRJHWKHU WKLV LV UHIHUUHG WR DV VORZ FRROf DQG LQFXEDWHG DW URRP WHPSHUDWXUH IRU

PAGE 54

DSSUR[LPDWHO\ RQH KRXU &OHDYDJH ZDV WKHQ LQLWLDWHG E\ H[SRVLQJ WKH VDPSOH WR YDU\LQJ SXOVHV RI ODVHU SXOVHVf &OHDYDJH RI fULERVHf ZDV LQLWLDWHG E\ UHVXVSHQGLQJ LQ UHDFWLRQ EXIIHU DQG VORZ FRROHG )LJXUH VKRZV VLPLODU SURJUHVV FXUYHV IRU ERWK fULERVHf DQG FDJHG FOHDYDJH %RWK VDPSOHV VKRZ LGHQWLFDO LQLWLDO FOHDYDJH NLQHWLFV ZLWK FOHDYDJH SODWHDXV QHDU b 7KH FDJHG SODWHDX LV UHJXODUO\ DERXW b ORZHU WKDQ fULERVHf 6HYHUDO VFHQDULRV FDQ H[SODLQ WKH ORZHU FOHDYDJH SODWHDX VHHQ ZLWK FDJHG Df LQVXIILFLHQW GHSURWHFWLRQ RI FDJHGULERVH Ef WKH VHTXHQFH LV PLVVLQJ WKH ULERVHDGHQRVLQH DW SRVLWLRQ GLVFXVVHG DERYH DQG EHORZ f&ORQLQJ DQG 6HTXHQFLQJ ff RU Ff ODVHU LQGXFHG GDPDJH WR WKH FDJHG '1$ VHTXHQFH 7KH UHDVRQV IRU ERWK FDJHG DQG ULERVH UHDFKLQJ FOHDYDJH SODWHDX IDU EHORZ FRPSOHWLRQ LV H[SORUHG EHORZ f8QGHUVWDQGLQJ 5HDVRQV IRU ,QFRPSOHWH &OHDYDJH RI '1$]\PH f 7R FRQILUP WKDW WKH ORZHU FOHDYDJH SODWHDX RI FDJHG ZDV QRW GXH WR LQVXIILFLHQW ODVHU NLQHWLF H[SHULPHQWV ZHUH FRQGXFWHG ZLWK DGGLWLRQDO SXOVHV RI ODVHU ,Q RQH H[SHULPHQW FDJHG FOHDYDJH ZDV LQLWLDWHG ZLWK SXOVHV RI ODVHU DQG DOORZHG WR UHDFW IRU KRXUV DSSURDFKLQJ FOHDYDJH SODWHDX 7KH VDPSOH ZDV GLYLGHG LQWR WZR DOLTXRWV DQG RQH DOLTXRW RI WKH VDPSOH ZDV H[SRVHG WR DQ DGGLWLRQDO SXOVHV RI ODVHU 7KH DGGHG ODVHU FDXVHG D VOLJKW LQFUHDVH LQ WKH DPRXQW FOHDYHG EXW WKLV LQFUHDVH ZDV LPPHGLDWH DQG QRW LQFUHDVLQJ ZLWK WLPH VXJJHVWLQJ DJDLQVW WKH SRVVLELOLW\ RI DGGLWLRQDO GHSURWHFWLRQ RI FDJHGULERVH $QRWKHU H[SHULPHQW ZDV DOVR SHUIRUPHG LQ ZKLFK FDJHG FOHDYDJH ZDV LQLWLDWHG ZLWK YDU\LQJ DPRXQWV RI ODVHU IURP WR SXOVHV 7KH FOHDYDJH SODWHDX LQFUHDVHG ZLWK LQFUHDVLQJ DPRXQWV RI ODVHU XQWLO SXOVHV RI ODVHU DIWHU ZKLFK

PAGE 55

LQFUHDVLQJ DPRXQWV RI ODVHU GLG QRW FDXVH DQ LQFUHDVH LQ FOHDYDJH SODWHDX )LJXUH D DQG f :KHQ WKH SHUFHQW RI FDJHG SURGXFW ZDV QRUPDOL]HG WR DFFRXQW IRU WKH LQFRPSOHWH GHSURWHFWLRQ DW ORZ OHYHOV RI ODVHU WKH IUDFWLRQ FOHDYHG DW HDFK WLPH SRLQW ZDV GLYLGHG E\ WKH IUDFWLRQ GHSURWHFWHG DV GHWHUPLQHG E\ EDVH K\GURO\VLVf WKH NLQHWLF SURILOHV IRU FDJHG ZLWK YDU\LQJ DPRXQWV RI ODVHU RYHUODS H[FHSW RI FRXUVH ]HUR ODVHU ZKLFK LV QRW H[SHFWHG WR FOHDYH DW DOOf 7KLV VKRZV WKDW WKH\ DUH DOO IROORZLQJ WKH VDPH NLQHWLFV SDUDPHWHUV ZLWK WKH SODWHDX YDU\LQJ ZLWK WKH DPRXQW RI ODVHU ZLWK PD[LPXP DFKLHYHG DW SXOVHV RI ODVHUf 7KH FDJHG H[SRVHG WR LQFUHDVLQJ ODVHU ZDV DOVR LQFXEDWHG XQGHU EDVH K\GURO\VLV FRQGLWLRQV 7KH DPRXQW RI FDJHG FOHDYHG E\ VWURQJ EDVH LQFUHDVHG ZLWK LQFUHDVLQJ ODVHU UHDFKLQJ D FOHDYDJH SODWHDX ZLWK SXOVHV RI ODVHU $GGHG ODVHU EH\RQG SXOVHV GLG QRW IXUWKHU LQFUHDVH WKH DPRXQW RI EDVHFOHDYHG FDJHG )LJXUH f 7KLV GLUHFWO\ SDUDOOHOV WKH UHVXOWV RI FDJHG SULPHU fDQJ FDJHf SOXV ODVHU FOHDYHG E\ VWURQJ EDVH FRQGLWLRQ LQGLFDWLQJ WKDW PD[LPXP GHSURWHFWLRQ KDV EH REWDLQHG /DVHU 'RHV 1RW 'DPDJH 5LERVH ,W LV SRVVLEOH WKDW FDJHG VKRZV D ORZHU FOHDYDJH SODWHDX WKDQ ULERVH EHFDXVH WKH ODVHU 89 IUHTXHQF\f XVHG WR LQLWLDWH WKH GHSURWHFWLRQ RI WKH FDJHGULERVH FDXVHV GDPDJH WR WKH '1$]\PH 1RQVSHFLILF EUHDNLQJ RI WKH SKRVSKRGLHVWHU EDFNERQH LV H[SHFWHG WR FUHDWH D VPHDU UXQQLQJ IURP WR QXFOHRWLGHV $OWKRXJK VOLJKW VPHDULQJ ZDV QRWLFHG EHWZHHQ WKH FOHDYHG DQG IXOO OHQJWK SURGXFW WKHUH ZDV QR VPHDULQJ EHORZ WKH IXOO OHQJWK SURGXFW QXFOHRWLGHVf )XUWKHUPRUH WKLV VPHDULQJ ZDV QRW SUHVHQW ZKHQ ULERVH ZDV H[SRVHG WR LGHQWLFDO ODVHU WUHDWPHQW

PAGE 56

7KH ODVHU FDQ SRWHQWLDOO\ FDXVH '1$ GDPDJH ZKLFK ZRXOG LQDFWLYH D '1$]\PH ZLWKRXW EUHDNLQJ WKH SKRVSKRGLHVWHU EDFNERQH VXFK DV IRUPLQJ WK\PLQHWK\PLQH GLPHUVf :H ZHUH XQDEOH WR ILQG HYLGHQFH IRU WK\PLQHWK\PLQH GLPHU IRUPDWLRQ ZKHQ ULERVH ZDV H[SRVHG WR ODVHU HYHQ WKRXJK ULERVH GRHV FRQWDLQ WZR VXFFHVVLYH WK\PLQHV DW SRVLWLRQV DQG DQG DQG f ,I WKH ODVHU LV FDXVLQJ QRQVSHFLILF '1$ GDPDJH UHVXOWLQJ LQ D UHGXFHG FOHDYDJH SODWHDX WKLV VKRXOG EH HYLGHQW HYHQ ZLWKRXW WKH FDJHGULERVH 7R WHVW WKLV ULERVH ZDV H[SRVHG WR LQFUHDVLQJ DPRXQWV RI ODVHU IURP WR SXOVHV P-SXOVHf 1R FKDQJH ZDV QRWLFHG LQ FOHDYDJH SODWHDX RU LQLWLDO UDWHV VXJJHVWLQJ ODVHU GRHV QRW FDXVH GDPDJH WR WKH '1$ )LJXUH f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f '1$]\PH FOHDUO\ LV QRW PDJQHVLXPGHSHQGHQW DV H[SHFWHG XQGHU WKH %UHDNHU-R\FH H[SHULPHQWDO VHWXS

PAGE 57

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fWUXH FRPSOHWLRQf GHILQHG DV FOHDYDJH FDXVHG E\ VWURQJ EDVH ZKLFK LV b IRU fDQJ ULERVHf DQG b IRU fDQJ FDJHff 7KH H[SODQDWLRQ IRU WKH FOHDYDJH SODWHDX PXVW EH FRPPRQ IRU ERWK FDJHG DQG ULERVH DQG TXLWH OLNHO\ DQ\ RWKHU '1$]\PH JHQHUDWHG YLD WKLV 96 SURFHGXUH ,QGHHG ORZ FOHDYDJH SODWHDX LV DQ LVVXH ZLWK PDQ\ ULER]\PHV DQG '1$]\PHV EXW WKH VSHFLILF FDXVHV IRU WKLV UHGXFHG SODWHDX LV RIWHQ LJQRUHG SUHVXPDEO\ EHFDXVH WKH UHVHDUFKHUV HLWKHU GR QRW UHDOL]H RU GR QRW EHOLHYH WKLV SKHQRPHQD LV VLJQLILFDQW :H QRZ VHHN WR H[SODLQ WKH IDLOXUH IRU WKH UHDFWLRQ WR UHDFK fWUXH FRPSOHWLRQf NQRZLQJ WKDW VXFK LQIRUPDWLRQ LV FUXFLDO WR ERWK UHGHVLJQLQJ WKH ,96 SURWRFRO DQG FUHDWLQJ D UHDVRQDEOH PRGHO RI WKH ,96 H[SHULPHQW ,QFRPSOHWH 'HSURWHFWLRQ RI &DJHG5LERVH :H KDYH DOUHDG\ GHPRQVWUDWHG WKDW ZH KDYH DFKLHYHG PD[LPXP GHSURWHFWLRQ RI WKH FDJHGULERVH ZLWK SXOVHV RI ODVHU PHDQLQJ DGGLWLRQDO ODVHU GRHV QRW UHVXOW LQ PRUH GHSURWHFWLRQf %XW LV LQFRPSOHWH GHSURWHFWLRQ HYHQ DW PD[LPXP GHSURWHFWLRQ D

PAGE 58

SODXVLEOH H[SODQDWLRQ IRU WKH IDLOXUH WR UHDFK WUXH FRPSOHWLRQ" 7KLV SRVVLELOLW\ LV UXOHG RXW RQ WKH EDVLV WKDW ULERVH DOVR UHDFKHV D SODWHDX WKDW LV IDU EHORZ WUXH FRPSOHWLRQ $OWKRXJK WKH PD[LPDO FOHDYDJH IRU FDJHG LV FRQVLVWHQWO\ b ORZHU WKDQ WKDW RI ULERVH ERWK ZKHQ VHOIFOHDYHG DQG ZKHQ FOHDYHG E\ EDVH K\GURO\VLVf ERWK UHDFK FOHDYDJH SODWHDXV WKDW DUH DV PXFK DV b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fULERVHf ZDV SUHSDUHG XVLQJ DV\PPHWULF 3&5 IROORZHG E\ JHOSXULILFDWLRQ VKRZHG FOHDYDJH SODWHDXV VLPLODU WR ULERVH JHQHUDWHG ZLWK H[RQXFOHDVH VXJJHVWLQJ WKH FOHDYDJH SODWHDX FDQQRW EH DWWULEXWHG WR LQKLELWLRQ E\ FRQWDPLQDWLRQ RI WKH FRPSOHPHQWDU\ VWUDQG :KHQ fULERVHf JHQHUDWHG YLD WKH H[RQXFOHDVH PHWKRG ZDV WUHDWHG ZLWK VWURQJ EDVH VLPLODU DPRXQWV RI FRQYHUVLRQ WR WKH FOHDYHG IRUP ZHUH REVHUYHG DV VHHQ IRU WKH ULERVH SULPHU fDQJ ULERVHf b IRU fULERVHf YV b IRU fDQJ ULERVHf 6LQFH WKHUH LV QR SRVVLELOLW\ RI D FRQWDPLQDWLQJ FRPSOHPHQWDU\ VWUDQG LQ WKH fDQJ ULERVHf SULPHU WKH ab GLIIHUHQFH EHWZHHQ EDVH FOHDYDJH RI WKH fDQJ ULERVHf SULPHU

PAGE 59

DQG WKH IXOO OHQJWK '1$]\PH fULERVHf LV WKH PD[LPXP DPRXQW DWWULEXWDEOH WR LQFRPSOHWH GHJUDGDWLRQ RI WKH FRPSOHPHQWDU\ VWUDQG E\ H[RQXFOHDVH ([RQXFOHDVH HIILFLHQF\ ZDV DOVR WHVWHG E\ fSKRVSKRU\ODWLQJ D VLQJOH SULPHU ZLWK JDPPD3$73 7KLV UDGLRODEHOHG SULPHU ZDV WKHQ XVHG WR V\QWKHVL]H IXOO OHQJWK fULERVHf 7KH GRXEOH VWUDQGHG SURGXFW ZDV WKHQ GLYLGHG LQWR WZR DOLTXRWV DQG RQH ZDV VXEMHFWHG WR VWDQGDUG H[RQXFOHDVH WUHDWPHQW (TXDO DPRXQWV RI HDFK VDPSOH ZHUH WKHQ HOHFWURSKRUHVHG RQ D 3$*(XUHD JHO DQG WKH DPRXQW RI IXOOOHQJWK SURGXFW DVVHVVHG ([RQXFOHDVH WUHDWHG VDPSOHV FRQVLVWHQWO\ VKRZHG OHVV WKDQ b RI WKH RULJLQDO PDWHULDO UHPDLQHG IXOO OHQJWK ([SHULPHQWV LQ ZKLFK VPDOO DPRXQWV RI WKH FRPSOHPHQW b RI WKH DPRXQW ULERVHf ZHUH DGGHG WR ULERVH KDG QR LPSDFW RQ WKH FOHDYDJH SODWHDX )LJXUH $ DQG %f :KHQ FOHDYDJH H[SHULPHQWV ZHUH SHUIRUPHG ZLWK HTXDO DPRXQWV RI ULERVH DQG LWV FRPSOHPHQW KRZHYHU WKH SUHVHQFH RI WKH FRPSOHPHQWDU\ VWUDQG GUDPDWLFDOO\ UHGXFHG WKH FOHDYDJH SODWHDX WR QHDUO\ b )LJXUH & DQG 'f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

PAGE 60

WR VHH LI WKLV ZDV WKH FDVH LQ RXU FOHDYDJH UHDFWLRQ E\ VHYHUDO PHWKRGV )LUVW FOHDYDJH D Q0 VROXWLRQ RI FDJHG ZDV LQLWLDWHG E\ SXOVHV RI ODVHU DQG DOORZHG WR UHDFW IRU KRXUV $W WKLV WLPH QHDU FOHDYDJH SODWHDX WKH UHDFWLRQ ZDV VSOLW LQWR WKUHH DOLTXRWV 2QH DOLTXRW ZDV XQDOWHUHG 7KH VHFRQG DOLTXRW ZDV GLOXWHG WZHQW\ILYH IROG LQ UHDFWLRQ EXIIHU 8QODEHOHG G ZDV DGGHG WR WKH WKLUG DOLTXRW WR D ILQDO FRQFHQWUDWLRQ RI Q0f LV WKH VDPH VHTXHQFH DV ULERVH EXW ZLWK D GHR[\DGHQRVLQH XVHG LQ WKH SODFH RI WKH ULERVH DGHQRVLQH ,Q D ODWHU VHFWLRQ ZH VKRZ WKDW G DFWV DV DQ HQ]\PH WR FOHDYH U LQ WUDQV ,I WKH FOHDYDJH SODWHDX LV D UHVXOW RI HTXLOLEULXP EHWZHHQ WKH IRUZDUG UHDFWLRQ DQG UHYHUVH UHDFWLRQ WKH DGGLWLRQ RI DGGLWLRQDO HQ]\PH G f VKRXOG VKLIW WKH HTXLOLEULXP LQ IDYRU WKH IRUPDWLRQ RI PRUH SURGXFW RYHU WLPH 6LPLODUO\ GLOXWLQJ WKH UHDFWLRQ UHGXFHV WKH OLNHOLKRRG WKDW DQ HQ]\PH FDQ ILQG WZR VXEVWUDWHV WR OLJDWH WRJHWKHU WKXV VKLIWLQJ WKH HTXLOLEULXP LQ IDYRU RI FOHDYDJH WKH FOHDYDJH UHDFWLRQ FDQ RFFXU LQ FLV DQG WKXV LV QRW DV VWURQJO\ DIIHFWHG E\ GLOXWLRQ DV OLJDWLRQf )LJXUH VKRZV WKDW QHLWKHU WUHDWPHQW VLJQLILFDQWO\ DOWHUV WKH FOHDYDJH SODWHDX DUJXLQJ DJDLQVW HTXLOLEULXP EHWZHHQ WKH IRUZDUG DQG UHYHUVH UHDFWLRQ ,Q D VHFRQG H[SHULPHQW WKH QXFOHRWLGH FOHDYDJH SURGXFW ZDV LVRODWHG YLD JHO SXULILFDWLRQ 7KLV FOHDYDJH SURGXFW ZDV UDGLRODEHOHG DQG IROORZLQJ SXULILFDWLRQ ZDV LQFXEDWHG LQ H[FHVVf ZLWK IXOOOHQJWK ULERVH QRW UDGLRODEHOHGf $OWKRXJK XQODEHOOHG SUHYLRXV VWXGLHV KDYH VKRZQ WKDW WKH IXOOOHQJWK '1$]\PH EHJLQV WR FOHDYH LWVHOI JHQHUDWLQJ ERWK WKH DQG QXFOHRWLGH SURGXFWV ,I WKH UHYHUVH UHDFWLRQ OLJDWLRQf RFFXUV WKH H[FHVV RI H[RJHQRXV UDGLRODEHOHG QXFOHRWLGH IUDJPHQW HQVXUHV WKDW WKHUH LV D JUHDWHU FKDQFH IRU WKH UDGLRODEHOHG QXFOHRWLGH IUDJPHQW WR EHFRPH OLJDWHG WR WKH QXFOHRWLGH IUDJPHQW WKXV FDXVLQJ DQ LQFUHDVH LQ VL]H RI WKH UDGLRODEHOHG

PAGE 61

QXFOHRWLGH IUDJPHQW RYHU WLPH 7KH UHDFWLRQ ZDV PRQLWRUHG IRU KRXUV IRU D VKLIW LQ VL]H RI WKH QXFOHRWLGH SURGXFW WR IXOOOHQJWK QXFOHRWLGHf EXW QR VXFK FRQYHUVLRQ ZDV REVHUYHG 7KLV DUJXHV DJDLQVW WKH SRVVLELOLW\ WKDW D VLJQLILFDQW IUDFWLRQ RI FOHDYDJH SURGXFWV DUH OLJDWHG WRJHWKHU $UH WKH DQG QXFOHRWLGH )UDJPHQWV $FWLQJ DV &DWDO\VWV RU ,QKLELWRUV" 7ZR SURGXFWV DUH FUHDWHG E\ WKH FOHDYDJH RI DW WKH ULERDGHQRVLQH D DQG D QXFOHRWLGH IUDJPHQW 7KH VWDQGDUG %UHDNHU DQG -R\FH ,96 SURWRFRO ZDV fGHVLJQHGf WR VHOHFW IRU PROHFXOHV WKDW DUH FDSDEOH RI VHOIFOHDYDJH LQ FLVf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f )ROORZLQJ LQLWLDWLRQ ZLWK SXOVHV RI ODVHU DQG LQFXEDWLRQ LQ UHDFWLRQ EXIIHU FOHDYDJH RI WKH QHZ FDJHG VDPSOH RFFXUUHG ZLWKRXW FKDQJH HYHQ LQ WKH SUHVHQFH RI HTXDO FRQFHQWUDWLRQ RI WKH DQG QXFOHRWLGH IUDJPHQWV )LJXUH f 7KLV VXJJHVWV DJDLQVW WKH LGHD WKDW WKH DQG D QXFOHRWLGH IUDJPHQWV VLJQLILFDQWO\ DIIHFW FOHDYDJH E\ DFWLQJ DV LQKLELWRUV RU FDWDO\VWV DW OHDVW ZKHQ SUHVHQW LQ HTXDO FRQFHQWUDWLRQ ZKLFK LV

PAGE 62

DSSUR[LPDWHO\ WKH KLJKHVW FRQFHQWUDWLRQ WKH\ UHDFK XQGHU QRUPDO FOHDYDJH FRQGLWLRQV LQ ZKLFK DSSUR[LPDWHO\ KDOI RI FRQYHUWV LQWR FOHDYDJH SURGXFWVf 7KH FDJHGOLEUDU\ DORQH GLG QRW VKRZ GHWHFWDEOH FDWDO\VLV XQGHU VWDQGDUG UHDFWLRQ FRQGLWLRQV WKLV IROORZV RXU H[SHFWDWLRQ WKDW DQ XQVHOHFWHG UDQGRP OLEUDU\ ZLOO KDYH YHU\ IHZ DFWLYH FDWDO\VWV $GGLWLRQ RI WKH QXFOHRWLGH SURGXFW DORQH GLG QRW LQFUHDVH FDJHG OLEUDU\ FOHDYDJH ZKLOH DGGLWLRQ RI WKH QXFOHRWLGH SURGXFW DORQH GLG VOLJKWO\ LQFUHDVH FOHDYDJH RI WKH OLEUDU\ &OHDYDJH OHYHOV ZHUH H[WUHPHO\ ORZ KRZHYHU ZKLOH KDG UHDFKHG D FOHDYDJH SODWHDX RI b LQ DERXW KRXUV WKH OLEUDU\ FOHDYHG E\ WKH QXFOHRWLGH IUDJPHQW KDG UHDFKHG RQO\ b FOHDYDJH LQ KRXUV 7KH DGGLWLRQ RI WKH QXFOHRWLGH IUDJPHQW WRJHWKHU ZLWK WKH QXFOHRWLGH IUDJPHQW UHGXFHG FOHDYDJH RI WKH OLEUDU\ 7KLV VKRZV WKDW WKH QXFOHRWLGH IUDJPHQW FDQ DFW DV D '1$]\PH WR FOHDYH D UDQGRP OLEUDU\ VXEVWUDWH LQ WUDQV EXW WKH UDWH RI FOHDYDJH LV RUGHUV RI PDJQLWXGH VORZHU WKDW WKH IXOOOHQJWK 7KH QXFOHRWLGH IUDJPHQW UHGXFHV WKLV DOUHDG\ ORZ DFWLYLW\ RI WKH QXFOHRWLGH IUDJPHQW VXJJHVWLQJ LW DFWV WR LQKLELW FOHDYDJH E\ FRPSHWLQJ ZLWK WKH VXEVWUDWH ULERVHOLEUDU\f IRU ELQGLQJ WR WKH QXFOHRWLGH IUDJPHQW 7KHVH UHVXOWV DUH FRQVLGHUHG DJDLQ LQ D ODWHU VHFWLRQ ZKHQ ZH H[SORUH SRWHQWLDO VWUXFWXUHV RI ,PSURSHUO\ )ROGHG '1$]\PH +DYLQJ VKRZQ WKDW WKH FOHDYDJH SURGXFWV GR QRW VLJQLILFDQWO\ DOWHU WKH FOHDYDJH SODWHDX ZH WKHQ VHW RXW WR GHWHUPLQH ZKHWKHU WKH SODWHDX ZDV GXH WR WKH PLVIROGLQJ RI D IUDFWLRQ RXU '1$]\PH LQWR DQ LQDFWLYH FRQIRUPDWLRQ :KLOH PRVW ,96 SURWRFROV ODUJHO\ LJQRUH WKH IROGLQJ LVVXH DOWRJHWKHU DVVXPLQJ IROGLQJ KDSSHQV TXLFNO\ DQG HIILFLHQWO\ LQ

PAGE 63

UHVSHFW WR FOHDYDJH E\ UHVXVSHQGLQJ WKH '1$]\PH LQ UHDFWLRQ EXIIHU ZLWKRXW FRIDFWRUVf ZH LQFOXGHG D UHIROGLQJ VWHS WKDW LV HDVLO\ UHSURGXFLEOH )ROORZLQJ UHn VXVSHQGLQJ WKH '1$]\PH LQ UHDFWLRQ EXIIHU VDPSOHV ZHUH KHDWHG WR r& IRU WKUHH PLQXWHV WR DFKLHYH FRPSOHWH GHQDWXUDWLRQ 7KH VDPSOHV ZHUH WKHQ FRROHG WR r& RYHU PLQXWHV XVLQJ D WKHUPRF\FOHU $OWKRXJK YHU\ FRQVLVWHQW HYHQ WKLV IROGLQJ SURFHGXUH FDQQRW JXDUDQWHH WKDW DOO '1$]\PHV ILQG WKH VDPH fDFWLYHf FRQIRUPDWLRQ ,QGHHG VWUXFWXUH SUHGLFWLRQV RI VKRZ VHYHUDO SRWHQWLDO VWUXFWXUHV SHUKDSV RQO\ RQH RI ZKLFK LV DFWLYH )ROGLQJ LQWR LQDFWLYH VWUXFWXUHV FDQ SRWHQWLDOO\ FRQWULEXWH WR WKH ORZ FOHDYDJH SODWHDX :KHQ FOHDYDJH DVVD\V DUH VWRSSHG E\ IUHH]LQJ WR r& DQG UXQ RQ D QRQn GHQDWXULQJ 3$*( JHO WZR EDQGV DUH SUHVHQW DW LQLWLDO WLPH SRLQWV 7KH ILUVW KLJKHVWf EDQG UHSUHVHQWV b RI WKH VDPSOH DQG UHPDLQV XQFKDQJHG WKURXJKRXW WKH GXUDWLRQ RI WKH NLQHWLF DVVD\ 7KH VHFRQG PLGGOHf EDQG LV LQLWLDOO\ b RI WKH VDPSOH EXW ZLWK WLPH LV FRQYHUWHG WR D WKLUG ORZHVWf EDQG DW KRXUV WKH ILUVW EDQG UHPDLQV b RI WKH WRWDO VDPSOH ZKLOH WKH VHFRQG EDQG LV UHGXFHG WR b DQG WKH WKLUG EDQG LV bf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b IRUPDPLGH ('7$ [\OHQH F\DQRO EURPRSKHQRO EOXHf DQG KHDWLQJ DW r& IRU PLQXWHV 7KH VDPSOH ZDV WKHQ

PAGE 64

HOHFWURSKRUHVHG RQ D GHQDWXULQJ 3$*(XUHD JHO DQG WKH XQFOHDYHG ULERVH ZDV H[FLVHG DQG SXULILHG 7KLV SUHYLRXVO\ XQFOHDYHG JHOSXULILHG ULERVH ZDV HWKDQRO SUHFLSLWDWHG UHVXVSHQGHG LQ UHDFWLRQ EXIIHU DQG IROGHG XVLQJ WKH VORZ FRRO SURWRFRO &OHDYDJH SURJUHVV IROORZLQJ WKLV JHOSXULILFDWLRQ ZDV WKHQ FRPSDUHG WR DQ XQWUHDWHG VDPSOH 7ZHQW\ILYH SHUFHQW RI WKH JHOSXULILHG ULERVH VDPSOH ZDV FOHDYHG LQ WKH KRXUV IROORZLQJ JHOSXULILFDWLRQ )LJXUH f 7KH DGGLWLRQDO FOHDYDJH IROORZLQJ WKH JHOSXULILFDWLRQ VXJJHVWV WKDW VRPH RI WKH XQFOHDYHG VDPSOH LV ORFNHG LQ DQ LQDFWLYH VWUXFWXUH ,W LV QRWDEOH KRZHYHU WKDW WKH JHOSXULILHG VDPSOH UHDFKHG D FOHDYDJH SODWHDX RI DSSUR[LPDWHO\ b IDU ORZHU WKDQ WKH FOHDYDJH SODWHDX RI WKH RULJLQDO VDPSOH bf :KLOH SDUW RI WKLV LQFRPSOHWH FOHDYDJH PD\ DJDLQ EH DWWULEXWHG WR PLVIROGHG PROHFXOHV LI PLVIROGLQJ ZHUH WKH RQO\ FDXVH RI LQFRPSOHWH FOHDYDJH D SODWHDX RI DSSUR[LPDWHO\ b ZRXOG EH H[SHFWHG IRU WKH JHOSXULILHG VDPSOH 7KH IDFW WKDW WKH JHO SXULILHG VDPSOH UHDFKHV D FOHDYDJH SODWHDX IDU EHORZ b LQGLFDWHV WKDW ZKLOH PLV IROGLQJ PD\ DFFRXQW IRU DV PXFK DV b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fPLVIROGLQJf SRVVLELOLW\ ZLWKRXW

PAGE 65

UHPRYLQJ DQ\ SRWHQWLDO LQKLELWRUV :H GLG WKLV E\ DJDLQ DOORZLQJ WR UHDFK FOHDYDJH SODWHDX DQG WKHQ GHQDWXULQJ DQ DOLTXRW RI WKH VDPSOH E\ KHDWLQJ LW WR r& IRU PLQXWHV DQG VORZO\ FRROLQJ WR r& RYHU PLQXWHV 5HKHDWLQJ DQG VORZO\ FRROLQJ DOORZV DQ\ PROHFXOH fVWXFNf LQ DQ LQDFWLYH FRQIRUPDWLRQ D VHFRQG FKDQFH WR DGRSW DQ DFWLYH FRQIRUPDWLRQ $V LQ WKH JHOSXULILFDWLRQ H[SHULPHQWV GHQDWXULQJ DQG UHIROGLQJ WKH VDPSOH YLD WKH VORZ FRRO SURFHGXUH LQFUHDVHV WKH DPRXQW RI PDWHULDO FOHDYHG )LJXUH f 7KH LQFUHDVH LQ FOHDYDJH LV VLJQLILFDQW XVXDOO\ DERXW bf EXW GRHV QRW EULQJ FOHDYDJH XS WR b WKH DPRXQW FOHDYHG E\ VWURQJ EDVH DQG WKHUHIRUH FRQVLGHUHG fFRPSOHWLRQff DQG WKHUHIRUH FDQQRW DFFRXQW HQWLUHO\ IRU WKH XQFOHDYHG DW SODWHDX &ORQLQJ DQG 6HTXHQFLQJ &OHDYHG DQG 8QFOHDYHG 1HDU &OHDYDJH 3ODWHDX 7KH DELOLW\ RI VWURQJ EDVH WR FOHDYH WKH ULERVHSULPHU fDQJ ULERVHf WR RQO\ b VXJJHVWHG WKDW DERXW b RI WKH SULPHU DQG WKHUHIRUH DQ\ IXOO OHQJWK '1$]\PH PDGH IURP WKDW SULPHU PD\ EH PLVVLQJ WKH ULERVHDGHQRVLQH )RU WKH FDJHGSULPHU fDQJ FDJHf EDVH K\GURO\VLV IROORZLQJ ODVHU GHSURWHFWLRQ UHVXOWHG LQ RQO\ b FOHDYDJH VXJJHVWLQJ WKDW WKH SUREOHP RI PLVVLQJ ULERVHDGHQRVLQH PD\ EH HYHQ PRUH VHYHUH IRU WKH FDJHGSULPHU WKLV FRXOG DOVR EH GXH WR LQFRPSOHWH GHSURWHFWLRQ GHVSLWH H[FHVV ODVHUf 0XWDWLRQV PD\ DOVR RFFXU WKURXJKRXW WKH '1$]\PH HLWKHU DV D UHVXOW RI PXWDWLRQV LQWURGXFHG GXULQJ WKH V\QWKHVLV RI WKH SULPHUV RU PXWDWLRQV EHWZHHQ WKH SULPHUV DV D UHVXOW RI SRO\PHUDVH HUURU 7RJHWKHU WKHVH PXWDWLRQV PD\ UHGXFH RU HOLPLQDWH WKH FDWDO\WLF SRZHU RI D IUDFWLRQ RI WKH '1$]\PH SRRO 7KLV SRVVLELOLW\ ZDV H[DPLQHG E\ FORQLQJ DQG VHTXHQFLQJ ERWK FOHDYHG DQG XQFOHDYHG QHDU WKH FOHDYDJH SODWHDX )LJXUH f 2I WKH VHTXHQFHG FORQHV RI WKDW ZHUH FOHDYHG DW WKH WLPH RI JHO

PAGE 66

SXULILFDWLRQ RQO\ WKUHH bf ZHUH IRXQG WR FRQWDLQ D PXWDWLRQ DOO WKUHH ZHUH LQ WKH 1 UHJLRQ EHWZHHQ WKH SULPHUVf 2I WKH FORQHV RI WKDW ZHUH XQFOHDYHG KDG DW OHDVW RQH PXWDWLRQ bf 7ZHOYH FORQHV b RI WKRVH VHTXHQFHGf KDG D PXWDWLRQ LQ WKH 1 UHJLRQ WKLUW\RQH FORQHV bf KDG D PXWDWLRQ LQ WKH fDQJf SULPHU WKUHH FORQHV bf KDG D PXWDWLRQ LQ WKH FRPSOHPHQWDU\ SULPHU 2I WKH XQFOHDYHG FORQHV VHTXHQFHG QLQH ZHUH IURP D '1$]\PH WKDW RULJLQDOO\ FRQWDLQHG D FDJHGULERVH FOHDYDJH LQLWLDWHG E\ ODVHUf 2I WKHVH QLQH FORQHV ILYH bf ZHUH VKRZQ WR EH PLVVLQJ WKH ULERVHDGHQRVLQH DV RSSRVHG WR VHYHQ RI FORQHV bf LVRODWHG IURP XQFOHDYHG ULERVH 7KLV FRQILUPV RXU RULJLQDO VXVSLFLRQ WKDW D KLJKHU SURSRUWLRQ RI WKH fDQJ FDJHf SULPHU LV PLVVLQJ WKH ULERVH DGHQRVLQH DV FRPSDUHG WR WKH fDQJ ULERVHf SULPHU DQG H[SODLQV ERWK WKH REVHUYDWLRQ WKDW DERXW b OHVV RI WKH fDQJ FDJHf SULPHU ZLWK PD[LPXP GHSURWHFWLRQ LV FOHDYHG E\ VWURQJ EDVH WKDQ WKH fDQJ ULERVHf SULPHU DQG WKH REVHUYDWLRQ WKDW FDJHG UHDFKHV D FOHDYDJH SODWHDX DERXW b ORZHU WKDQ ULERVH $ FDYHDW PXVW EH DGGHG KRZHYHU LQ WKDW LW KDV EHHQ REVHUYHG WKDW WKH SRO\PHUDVH WHQGV WR WHUPLQDWH SULPHU HORQJDWLRQ DERXW b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f 7RJHWKHU WKLV FRQILUPV WKDW WKH ORZHU FOHDYDJH VHHQ IRU FDJHG DV FRPSDUHG WR ULERVH LV D

PAGE 67

UHVXOW RI RQO\ WZR SRVVLELOLWLHV LQFRPSOHWH GHSURWHFWLRQ RU D KLJKHU IUDFWLRQ RI WKH SULPHU PLVVLQJ WKH ULERVHDGHQRVLQH 6HYHUDO RI WKH FORQHV ZLWK PXWDWLRQV ZHUH WHVWHG WR VHH LI WKHLU PXWDWLRQV LQ IDFW UHGXFHG WKH UDWH RI FOHDYDJH )LJXUHV DQG VKRZ WKH FOHDYDJH SURILOHV RI VHYHUDO PXWDQWV DQG GHPRQVWUDWH WKDW QHDUO\ DOO RI WKH PXWDWLRQV FDXVH JUHDWO\ UHGXFHG FOHDYDJH UDWHV 7RJHWKHU ZLWK WKH REVHUYDWLRQV DERYH ZH KDYH VKRZQ WKDW DSSUR[LPDWHO\ b RI WKH WRWDO ULERVH VDPSOH GRHV QRW FOHDYH EHFDXVH LW LV PLVVLQJ WKH ULERVH DGHQRVLQH b IRU WKH FDJHGf DSSUR[LPDWHO\ b RI WKH WRWDO VDPSOH PD\ EH PLVIROGHG DQG RQO\ VORZO\ FRQYHUWV LQWR WKH DFWLYH FRQIRUPDWLRQ DQG b RI WKH XQFOHDYHG IUDFWLRQ PD\ QRW FOHDYH GXH WR PXWDWLRQV '1$]\PH &OHDYHV LQ FLV DQG WUDQV $V LV WKH FDVH ZLWK DOO HQ]\PHV D '1$]\PH UHDFWLRQ FDQ EH WKRXJKW RI DV RFFXUULQJ LQ VHYHUDO VWHSV ELQGLQJ WKH VXEVWUDWH WKH FKHPLFDO VWHS ZKHUH FRYDOHQW ERQGV DUH PDGH RU EURNHQf DQG UHOHDVH RI WKH VXEVWUDWH 7KH 96 SURWRFRO ZDV GHVLJQHG VR WKDW WKH '1$]\PH FRQWDLQV WKH VXEVWUDWH ULERVHDGHQRVLQHf OLQNHG WR WKH HQ]\PH WKH UDQGRP 1 UHJLRQf DQG HQJLQHHUHG VXFK WKDW WKH '1$]\PH ZRXOG IROG EDFN RQWR LWVHOI YLD WZR FODPSV RI EDVH SDLULQJ RQH RQ HDFK VLGH RI WKH ULERVHDGHQRVLQH WKXV EULQJLQJ WKH fHQ]\PDWLF UHJLRQf LQ SUR[LPLW\ WR WKH fVXEVWUDWHf %HFDXVH RI WKH GHVLJQ WKH '1$]\PH LV H[SHFWHG WR IROG PRUH UDSLGO\ ZLWK LWVHOI WKDQ ELQG RWKHU PROHFXOHV DW W\SLFDO FRQFHQWUDWLRQV DQG WKHUHIRUH WKH UHDFWLRQ LV DVVXPHG EH XQLPROHFXODU +RZHYHU EHFDXVH DOO PROHFXOHV LQ D OLEUDU\ VKDUH WKH VDPH f DQG fSULPHUV RYHU b RI WKH

PAGE 68

VHTXHQFH LV LGHQWLFDO LQ DOO PROHFXOHV LQ SDUWLFXODU WKH UHJLRQV VXUURXQGLQJ WKH ULERVH DGHQRVLQHf 7KLV RSHQV WKH SRVVLELOLW\ WKDW WKH fHQ]\PDWLF UHJLRQf PD\ ELQG WR WKH fVXEVWUDWH UHJLRQf RI DQRWKHU PROHFXOH DQG HLWKHU FOHDYH LQ WUDQV RU LQKLELW FOHDYDJH ,Q WKH VWDQGDUG %UHDNHU-R\FH SURWRFRO ZKLFK ZDV XVHG WR JHQHUDWH f ELQGLQJ SRWHQWLDO '1$]\PHV WR D FROXPQ LV EHOLHYHG WR SUHYHQW PROHFXOHV IURP ELQGLQJ WR DQG FOHDYLQJ RWKHU PROHFXOHV $PSOLILFDWLRQ DQG UHVHOHFWLRQ RI WKH VXUYLYRUV LV DOVR EHOLHYHG WR SUHIHUHQWLDOO\ HQULFK VHOIFOHDYLQJ '1$]\PHV ,W LV SRVVLEOH KRZHYHU WKDW DQ fHQ]\PDWLFf '1$]\PH FDQ ELQG WR D fVXEVWUDWHf PROHFXOH DQG WRJHWKHU WKH\ ELQG WR WKH FROXPQ YLD WKH ELRWLQ RI RQH RI WKH WZR PROHFXOHV 7KLV FRQMRLQHG '1$]\PHVXEVWUDWH FRPSOH[ VXUYLYHV ZDVKLQJ EXW XSRQ FOHDYLQJ RI WKH VXEVWUDWH HLWKHU ERWK HQ]\PH DQG VXEVWUDWH DUH UHOHDVHG IURP WKH FROXPQ LI WKH FRQMRLQHG PROHFXOH ZDV DWWDFKHG WR WKH FROXPQ YLD WKH ELRWLQ RI WKH VXEVWUDWHf RU RQO\ WKH fVXEVWUDWHf LV ZDVKHG RII LI WKH FRQMRLQHG PROHFXOH ZDV DWWDFKHG WR WKH FROXPQ YLD WKH ELRWLQ RI WKH HQ]\PHf 7KLV PD\ UHVXOW LQ WKH LQDGYHUWHQW VHOHFWLRQ IRU '1$]\PHV WKDW FDQ FOHDYH LQ WUDQV DV ZHOO DV PROHFXOHV WKDW DUH JRRG VXEVWUDWHV IRU WUD]VFOHDYDJH 7KH fDQJf SULPHU XVHG WR JHQHUDWH '1$]\PH DV ZHOO DV NLQHWLF DQDO\VLVf LV QHDUO\ LGHQWLFDO WR WKH %UHDNHU-R\FH SULPHU EXW ZLWK WZR VLQJOHQXFOHRWLGH FKDQJHV WKDW UHGXFH WKH VWUHQJWK RI WKH EDVHSDLULQJ FODPSV GHVLJQHG WR HQFRXUDJH FLV IROGLQJ 7KHVH FKDQJHV IXUWKHU RSHQ XS WKH SRVVLELOLW\ WKDW FOHDYDJH FDQ RFFXU LQ WUDQV :H DUH XQDZDUH RI DQ\ SXEOLVKHG UHVXOWV WKDW H[SOLFLWO\ FRQVLGHU WKH LPSDFW RI ELQGLQJ FODPSV RQ WKH RXWFRPH RI VHOHFWLRQV EXW EDVHG RQ WKH DVVXPSWLRQ WKDW WKH SURWRFRO VHOHFWV IRU FLV

PAGE 69

FOHDYLQJ DFWLYLW\ RI WKH UDQGRP UHJLRQ PRVW UHVHDUFK DSSDUHQWO\ DVVXPH WKH LPSDFW RI WKHVH HQJLQHHUHG VLJKWV LV QHJOLJLEOH +RZHYHU VRPH SXEOLVKHG DFFRXQWV KDYH UHSRUWHG WKDW WKH UDWH RI VRPH '1$]\PHV DSSHDUV WR GHFUHDVH ZKHQ WKH\ DUH JHQHUDWHG ZLWK DGGLWLRQDO F\FOHV RI 3&5 YV F\FOHV WRR IHZ F\FOHV WR H[SHFW D VLJQLILFDQW FKDQJH LQ WKH DPRXQW RI PXWDWLRQ IRU D QXFOHRWLGH SURGXFWf *H\HU %UHDNHU f 1R H[SODQDWLRQ LV RIIHUHG IRU WKLV UHVXOW 1HLWKHU SDSHUV FLWHG GR QRW UHSRUW WKH FRQFHQWUDWLRQ RI '1$]\PH XVHG LQ WKHLU NLQHWLF DQDO\VLV SUHVXPDEO\ EHFDXVH WKH\ QHYHU FRQVLGHUHG WKH SRVVLELOLW\ RI WUDQVn FOHDYDJH DQG LW LV WKHUHIRUH OLNHO\ WKDW WKH '1$]\PH FRQFHQWUDWLRQ ZDV QRW FRQWUROOHG LQ WKHLU DQDO\VLVf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f RULJLQ ZLWK D VORSH HTXDO WR WKH VHFRQG RUGHU UDWH FRQVWDQW ,I SURGXFW ZHUH JHQHUDWHG H[FOXVLYHO\ E\ D VHFRQG RUGHU

PAGE 70

SURFHVV ZLWK VDWXUDWLRQ LPSO\LQJ WKDW IRU WKH '1$]\PH RQH PROHFXOH RI PXVW ELQG WR DQRWKHU PROHFXOH RI IRU D UHDFWLRQ WR RFFXU EXW WKDW WKH FRPSOH[ FRXOG GLVDVVRFLDWH EHIRUH D UHDFWLRQ RFFXUUHG D FRQGLWLRQ ZHOO NQRZQ LQ 0LFKDHOLV 0HQWHQ NLQHWLFVf WKHQ WKH DSSDUHQW ILUVW RUGHU UDWH FRQVWDQW ZRXOG EH GHSHQGHQW XSRQ >@ LQLWLDOO\ DQG DSSURDFK DQ DV\PSWRWH DW KLJK >@ +HUH D SORW RI DSSDUHQW NRE6 YHUVXV >@ VKRXOG EH D OLQH WKDW LQFOXGHV WKH f RULJLQ VORSLQJ XS DQG XOWLPDWHO\ OHYHOLQJ WR D SODWHDX 7KH FRQFHQWUDWLRQ GHSHQGHQFH RI WKH UDWH RI FOHDYDJH RI ULERVH ZDV WHVWHG E\ H[DPLQLQJ WKH SURJUHVV FXUYH DQG WKH LQLWLDO UDWHV IRU ULERVH FOHDYDJH DW YDULRXV GLIIHUHQW GLOXWLRQV Q0 WR Q0 ULERVHf 5HVXOWV IURP WKLV H[SHULPHQW )LJXUH DQG f VKRZ D UHDOLW\ ZKHUH VHYHUDO UDWH SURFHVVHV DUH RSHUDWLQJ DW WKH VDPH WLPH )LJXUH VKRZV WKDW WKH LQLWLDO UDWH RI ULERVH FOHDYDJH GHFUHDVHV ZLWK GHFUHDVLQJ FRQFHQWUDWLRQ D ORJDULWKPLF DEVFLVVD LV XVHG WR FDSWXUH WKH IXOO UDQJH RI WKH H[SHULPHQWDO FRQFHQWUDWLRQVf 7KH GHFUHDVH LQ WKH LQLWLDO UDWH RI FOHDYDJH ZLWK GHFUHDVLQJ FRQFHQWUDWLRQ UHDFKHV D ORZHU OLPLW EHORZ ZKLFK IXUWKHU GLOXWLRQ FDXVHV YHU\ OLWWOH IXUWKHU UHGXFWLRQ LQ WKH LQLWLDO UDWH 7KLV LPSOLHV WKDW WKH UDWH GRHV KDYH VRPH FRQFHQWUDWLRQ GHSHQGHQFH DQG LPSOLHV WKDW VXEVWUDWH LV EHLQJ FRQVXPHG E\ WZR SDWKV D ILUVW RUGHU UDWH SURFHVV DQG DW OHDVW SDUWO\ E\ D SURFHVV WKDW LV KLJKHU WKDQ ILUVW RUGHU ,Q WKH OHIW SRUWLRQ RI )LJXUH ORZ >@f WKH ILUVW RUGHU SURFHVV GRPLQDWHV WKH RYHUDOO UDWH ,Q WKH ULJKW SRUWLRQ RI )LJXUH KLJK >@f VHFRQG RUGHU SURFHVVHV GRPLQDWH WKH RYHUDOO UDWH 7KLV H[SHULPHQW ZDV UHSHDWHG VHYHUDO WLPHV RYHU ZLWK VLPLODU UHVXOWV

PAGE 71

,Q )LJXUH $ WKH VDPH GDWD DV )LJXUH EXW ZLWK D OLQHDU DEVFLVVDf WKH LQWHUFHSW DW >@ LV QRW DW WKH RULJLQ 7KLV UHTXLUHV WKDW DW OHDVW VRPH RI WKH UDWH DULVH YLD D XQLPROHFXODU SURFHVV )LJXUH % VKRZV D EHVWILW OLQH H[WUDSRODWHG WR ]HUR LQILQLWH GLOXWLRQf IRU FRQFHQWUDWLRQV EHORZ Q0 SUHGRPLQDWHO\ XQLPROHFXODUf \LHOGV DQ LQWHUFHSW FRUUHVSRQGLQJ WR WKH XQLPROHFXODU UDWH FRQVWDQW KU KDOIOLIH LV KRXUVf 7KH VORSH RI WKLV OLQH LV [O 0fKUn ZKLFK IRU D VHFRQG RUGHU SURFHVV ZKHQ >6@m.P HTXDOV WKH VHFRQG RUGHU UDWH FRQVWDQW NFDW.Pf $ SORW RI WKH QDWXUDO ORJ RI XQFOHDYHG ULERVH VXEVWUDWHf YHUVXV WLPH VKRXOG EH OLQHDU IRU D ILUVW RUGHU SURFHVV D OLQHDU UHODWLRQVKLS ILW WKH GDWD ZHOO IRU ORZ ULERVH EXW QRW IRU KLJKHU FRQFHQWUDWLRQV )LJXUH f 7KH QHJDWLYH RI WKH VORSH RI WKH EHVWILW OLQH IRU Q0 ULERVH LV [On KUn ZKLFK LV LQ FORVH DJUHHPHQW ZLWK WKH UDWH HVWLPDWHG E\ H[WUDSRODWLRQ RI WKH LQLWLDO UDWHV RI ULERVH WR LQILQLWH GLOXWLRQ )LJXUH $ DQG %f )XUWKHU LW DSSHDUV IURP )LJXUH $ WKDW D SODWHDX KDV QRW EHHQ UHDFKHG ,W PD\ EH WKDW WKH >@ LV VLPSO\ QRW KLJK HQRXJK LQ WKLV H[SHULPHQW +RZHYHU WKLV SORW GRHV QRW UXOH RXW WKH SRVVLELOLW\ WKDW D SXUHO\ ELPROHFXODU SURFHVV QRW LQYROYLQJ VDWXUDWLRQf LV FRQWULEXWLQJ WR WKH RYHUDOO UDWH RI WKH UHDFWLRQ $OWKRXJK WKLV UDWH LV ORZ E\ QRUPDO ULER]\PH VWDQGDUGV DQG HYHQ WKH VWDQGDUG RI VRPH '1$]\PH SKRVSKRGLHVWHUDVHV LVRODWHG YLD ,96 WKLV XQLPROHFXODU UDWH LV FRPSDUDEOH WR WKH PDJQHVLXPLQGHSHQGHQW FDWDO\VW LVRODWHG XVLQJ ,96 E\ *H\HU KRXUn IRU f* )DPLO\f '1$]\PH EHIRUH RSWLPL]DWLRQf :H WKHQ WHVWHG WR VHH LI WKH SURJUHVV FXUYH IRU HDFK FRQFHQWUDWLRQ RI ULERVH ILW EHVW WR RQH H[SRQHQWLDO LQ ZKLFK ERWK WKH UDWH DQG SODWHDX FDQ YDU\f RU WKH VXP RI WZR

PAGE 72

H[SRQHQWLDO HTXDWLRQV LQ ZKLFK ERWK WKH UDWH DQG SODWHDX FDQ YDU\ IRU HDFK HTXDWLRQf $W ORZ ULERVH FRQFHQWUDWLRQV Q0 DQG EHORZf WKH SURJUHVV FXUYHV IRU FOHDYDJH ILW ZHOO WR D VLQJOH H[SRQHQWLDO HTXDWLRQ ZKHUHDV DW KLJKHU FRQFHQWUDWLRQV WKH SURJUHVV FXUYHV ZHUH ILW EHWWHU E\ WKH VXP RI WZR H[SRQHQWLDO HTXDWLRQV $ VLQJOH H[SRQHQWLDO FXUYH ILW WR SURJUHVV FXUYH IRU Q0 ULERVH FOHDYDJH HVWLPDWHG D UDWH [ KU ZLWK D SUHGLFWHG SODWHDX RI bf 7KLV DJUHHV ZLWK WKH RWKHU HVWLPDWLRQV RI WKH XQLPROHFXODU UDWH PDGH DERYH ZKLFK FRPSDUDWLYHO\ DUH XQGHUHVWLPDWLRQV VLQFH WKH\ KDYH QRW FRUUHFWHG IRU D SODWHDX EHORZ bf 7KLV VXJJHVWV WKDW DW ORZ FRQFHQWUDWLRQV Q0 DQG EHORZf WKH UHDFWLRQ EHKDYHV LQ XQLPROHFXODU IDVKLRQ ZKLOH DW KLJKHU FRQFHQWUDWLRQV WKH UHDFWLRQ LV QRW ZHOO FKDUDFWHUL]HG E\ D VLQJOH XQLPROHFXODU UDWH ODZ )XUWKHUPRUH ZH FDQ FRQFOXGH WKDW DW ORZ FRQFHQWUDWLRQV ZKHUH XQLPROHFXODU SURFHVVHV GRPLQDWH D VHFRQG FRQIRUPHU RI HQn IROGHG ZLWK ORZHU UHDFWLYLW\ LV QRW SUHVHQW DW VLJQLILFDQW OHYHOV ,Q DQ HDUOLHU VHFWLRQ f,PSURSHUO\ )ROGHG '1$]\PH ff ZH DOUHDG\ FRQVLGHUHG YDULDQWV RI WKDW ZHUH LQDFWLYH PXWDQWV IRU H[DPSOHf 7KLV LV LPSRUWDQW EHFDXVH LI WKHVH '1$]\PHV LQ JHQHUDO KDYH D VLQJOH DFWLYH FRQIRUPDWLRQ DQG UHDFW IROORZLQJ D VLQJOH H[SRQHQWLDO ILUVW RUGHU UDWH ODZ WKHQ WKH GLVWULEXWLRQ RI FDWDO\WLF SRZHU ZLWKLQ D SRSXODWLRQ RI UHODWHG PROHFXOHV FDQ EH PRGHOHG DW OHDVW DV DQ DSSUR[LPDWLRQf E\ D WUDQVIRUP WKDW ILWV D VSHFWUXP RI ILUVW RUGHU H[SRQHQWLDO SURFHVVHV WR D SURJUHVV FXUYH 7KH fQRQXQLPROHFXODUf FRPSRQHQW RI ULERVH FOHDYDJH LV QRW SXUHO\ EL PROHFXODU DV WULSOLQJ WKH FRQFHQWUDWLRQ RI ULERVH GRHV QRW LQFUHDVH WKH LQLWLDO UDWH QLQHIROG DV ZRXOG EH H[SHFWHG EXW LQ IDFW LW LQFUHDVHV OHVV WKDQ WZRIROGf )XUWKHUPRUH

PAGE 73

LI WKH UHDFWLRQ ZKHUH VLPSO\ WKH VXP RI D XQLPROHFXODU DQG ELPROHFXODU UHDFWLRQ WKH SURJUHVV FXUYH VKRXOG ILW WR WKH VXP RI WZR SRZHU IXQFWLRQV < $; %; ZKHUH ; LV WKH FRQFHQWUDWLRQ RI < LV WKH LQLWLDO UDWH NREVf $ LV WKH ILUVW RUGHU UDWH FRQVWDQW DQG % LV WKH VHFRQG RUGHU UDWH FRQVWDQWf 7KLV PRGHO KRZHYHU ZDV XQDEOH WR SURYLGH D UHDVRQDEOH ILW IRU WKH H[SHULPHQWDO UHVXOWV VXJJHVWLQJ WKDW ULERVH FOHDYDJH LV QRW VLPSO\ VXP RI D XQLPROHFXODU DQG D ELPROHFXODU UHDFWLRQ 'HR[\ULERVH &OHDYHV 9DULRXV 5LERVH6XEVWUDWHV 7KH REVHUYDWLRQ WKDW ULERVH FDQ FOHDYH LQ D QRQXQLPROHFXODU IDVKLRQ VXJJHVWV WUDQV FOHDYDJH 7KLV ZDV WHVWHG GLUHFWO\ E\ LQFXEDWLQJ ZLWK YDULRXV VXEVWUDWHV HDFK FRQWDLQLQJ WKH f FDWDO\WLF PRWLI DQG ULERVHDGHQRVLQH ,QLWLDOO\ Q0 ULERVH ZDV LQFXEDWHG ZLWK Q0 RI f3ODEHOOHG fDQJ ULERVHf SULPHU )LJXUH VKRZV WKDW ERWK ULERVH DQG fDQJ ULERVHf DUH FOHDYHG FOHDUO\ HVWDEOLVKLQJ WUDQV FOHDYDJH DELOLW\ RI 7KH FOHDYDJH RI ULERVH LQ WKH SUHVHQFH RI fDQJ ULERVHf SULPHU ZDV ORZHU WKDQ ULERVH LQFXEDWHG DORQH VXJJHVWLQJ WKDW WKH fDQJ ULERVHf SULPHU FRPSHWHV ZLWK ULERVH IRU FOHDYDJH E\ ULERVH )LJXUH f :H WKHQ WHVWHG ZKHWKHU D '1$]\PH V\QWKHVL]HG ZLWK D GHR[\DGHQRVLQH LQVWHDG RI D ULERVHDGHQRVLQH UHIHUUHG WR DV fGf DQG WKHUHIRUH XQDEOH WR FOHDYH LWVHOI FRXOG FOHDYH YDULRXV VXEVWUDWHV LQ WUDQV :H DJDLQ WHVWHG WUDQV FOHDYDJH RI WKH fDQJ ULERVHf SULPHU )LJXUH f DV ZHOO DV D PXWDQW RI NQRZQ DV fXSf ZKLFK KDV JUHDWO\ UHGXFHG FDWDO\WLF DFWLYLW\f D SRRO RI PROHFXOHV V\QWKHVL]HG IURP WKH UDQGRP OLEUDU\ DQG D LQGLYLGXDO FORQH IURP WKH UDQGRP OLEUDU\ NQRZQ DV fOLERf ZKLFK KDV QR GHWHFWDEOH FDWDO\WLF DFWLYLW\f )LJXUH VKRZV WKDW HDFK VXEVWUDWH ZDV FOHDYHG E\ G

PAGE 74

DOWKRXJK ZLWK YDU\LQJ GHJUHHV RI HIILFLHQF\ 7KH REVHUYDWLRQ WKDW WKH VLQJOHn QXFOHRWLGH SRLQW PXWDWLRQ RI fXSf ZKLFK LV OLNHO\ WR IROG VLPLODUO\ WR f LV FOHDYHG WKH VORZHVW E\ G ZKLOH WKH fDQJ ULERVHf SULPHU LV FOHDYHG IDVWHU WKDQ RWKHU IXOO OHQJWK PROHFXOHV DJUHHV ZLWK WKH SUHGLFWLRQ WKDW QRQSURGXFWLYH ELQGLQJ ZLWK fVHOIf FDQ UHGXFH ELQGLQJ DQG FDWDO\VLV E\ G &RPSHWLWLRQ 6WXGLHV RI 5LERVH &OHDYDJH :H WHVWHG WKH DELOLW\ RI WKH VDPH WKUHH VXEVWUDWHV fDQJ ULERVHf SULPHU fULERVH XSf DQG fULERVHOLERff WR FRPSHWH ZLWK U IRU WUDQV FOHDYDJH 7KH FRPSHWLWRUV ZHUH V\QWKHVL]HG ZLWK D GHR[\DGHQRVLQH LQ WKH SODFH RI WKH ULERVHDGHQRVLQH DQG DGGHG LQ IROG H[FHVV RYHU ULERVH $V D FRQWURO IROG H[FHVV RI G ZDV DGGHG WR U 7KH DGGLWLRQ RI Q0 G WR Q0 U LQFUHDVHG FOHDYDJH WR WKH VDPH OHYHO DV U DW Q0 DV D UHVXOW RI G FOHDYDJH RI U LQ WUDQV )LJXUH f 7KH FOHDYDJH RI U ZDV UHGXFHG E\ WKH DGGLWLRQ RI fDQJ ULERVHf SULPHU DQG fGOLERf VKRZLQJ WKDW WKHVH PROHFXOHV FRPSHWH IRU U WUDQV FDWDO\VLV +RZHYHU WKH DGGLWLRQ RI IROG H[FHVV PXWDQW fXSf WR U GLG QRW VKRZ WKH UHGXFWLRQ LQ U FOHDYDJH VHHQ E\ RWKHU FRPSHWLWRUV ,QVWHDG WKH U FOHDYDJH ZDV LQFUHDVHG DOWKRXJK WKH LQFUHDVH ZDV IDU OHVV WKDQ WKDW VHHQ ZLWK WKH DGGLWLRQ RI G 7ZR VFHQDULRV FDQ H[SODLQ WKLV UHVXOW )LUVW LW LV SRVVLEOH WKDW DOWKRXJK WKH FDWDO\WLF DFWLYLW\ RI fXSf LV UHGXFHG WKH UHGXFWLRQ LV RIIVHW E\ WKH JUHDW H[FHVV RI WUDQV DFWLQJ fXSf $OWHUQDWLYHO\ LW LV SRVVLEOH WKDW WKH PXWDWLRQ LQ fXSf GRHV QRW DOWHU LWV FDWDO\WLF PRWLI DQG DFWLYLW\ EXW LQVWHDG WKH PXWDWLRQ SUHYHQWV ELQGLQJ RI WKH FDWDO\WLF PRWLI QHDU LWV RZQ VXEVWUDWH 7KXV LQFXEDWLQJ fXSf DORQH VKRZV DOPRVW QR FDWDO\WLF DFWLYLW\ EHFDXVH WKH DFWLYH FDWDO\WLF PRWLI FDQQRW ELQG QHDU LWV RZQ VXEVWUDWH EXW LI LQFXEDWHG ZLWK DQRWKHU

PAGE 75

PROHFXOH WKDW GRHV QRW KDYH D PXWDWLRQ LQWHUIHULQJ ZLWK ELQGLQJ RI WKH FDWDO\WLF PRWLI RI XS WR WKH WUDQV VXEVWUDWH LW PD\ EH DEOH WR FOHDYH LQ WUDQV 7KH FRPSHWLWLRQ VWXG\ GHVFULEHG DERYH ZDV GRQH DW KLJK FRQFHQWUDWLRQV RI fHQ]\PHf Uf WR IDYRU LQWHUPROHFXODU DVVRFLDWLRQ DQG WUDQV FOHDYDJH :H SHUIRUPHG D VLPLODU FRPSHWLWLRQ VWXG\ EXW ZLWK FRQFHQWUDWLRQV RI U DQG H[FHVV FRPSHWLWRU VXIILFLHQWO\ ORZ DV WR IDYRU XQLPROHFXODU FOHDYDJH 8QGHU WKHVH FRQGLWLRQV DVVRFLDWLRQ RI WKH FRPSHWLWRU WR WKH HQ]\PH ZKLFK LV LQWHUPROHFXODU LV OHVV IDYRUHG WKDQ XQL PROHFXODU DVVRFLDWLRQ RI WKH HQ]\PH ZLWK LWV RZQ VXEVWUDWH PRWLI $V SUHGLFWHG WKH DGGLWLRQ RI Q0 GOLER WR Q0 U KDG QR QRWLFHDEOH HIIHFW RQ U FOHDYDJH )LJXUH f G &OHDYHV )DVWHU LQ WUDQV 7KDQ LQ FLV 6LQJOHWXUQRYHU .LQHWLFV 6XEVWUDWH VDWXUDWLRQ H[SHULPHQWV ZHUH SHUIRUPHG ZLWK LQFUHDVLQJ FRQFHQWUDWLRQ RI fHQ]\PHf Gf WR GHWHUPLQH WKH NFDWELf IRU WKH ELPROHFXODU UHDFWLRQ )LJXUHV DQG VKRZ WKH IRLO SURJUHVV FXUYHV IRU FOHDYDJH LQLWLDO OLQHDU SRUWLRQ DQG HQ]\PH VDWXUDWLRQ SURILOH IRU H[SHULPHQWV ZLWK YDULRXV VXEVWUDWHV fDQJ ULERVHf UOLER RU U f FOHDYHG E\ G 7KH FDOFXODWHG G LV DQG Q0 IRU USULPHU U DQG UOLER VXEVWUDWHV UHVSHFWLYHO\ 7KH NFDWELf ZKLFK LV GHWHUPLQHG IURP WKH PD[LPXP UDWH RI FOHDYDJH ZLWK VDWXUDWLQJ HQ]\PH LV DQG KUn IRU USULPHU U DQG UOLER VXEVWUDWHV UHVSHFWLYHO\ :KLOH WKH .G IRU HDFK VXEVWUDWH ZDV UHODWLYHO\ ORZ Q0f WKH PD[LPXP UDWH ZDV QRW DWWDLQHG XQWLO HQ]\PH FRQFHQWUDWLRQ UHDFKHG DSSUR[LPDWHO\ Q0 ,QGHHG WKH LQLWLDO UDWHV DSSHDU WR EH LQFUHDVLQJ VOLJKWO\ HYHQ DERYH Q0 LQGLFDWLQJ WKDW WKH

PAGE 76

VDWXUDWLRQ PD\ QRW EH FRPSOHWHO\ DFKLHYHG VXIILFLHQW HQ]\PH ZDV QRW DYDLODEOH WR WHVW NLQHWLFV DW FRQFHQWUDWLRQV KLJKHU WKDQ Q0f 7KH VORZ DSSURDFK WR SODWHDX LV PRVW QRWDEOH IRU WKH UOLER VXEVWUDWH DQG WR D OHVVHU H[WHQW ZLWK U VXJJHVWLQJ WKDW VXEVWUDWH RU HQ]\PH PD\ H[LVW LQ LQDFWLYH DOWHUQDWLYH FRQIRUPDWLRQV DQ H[SODLQLQJ WKH ORZ 5 YDOXH IRU WKH DVVRFLDWLRQ FXUYH ZLWK UOLERf $ VHFRQG H[SHULPHQW FRQGXFWHG ZLWK fDQJ ULERVHf SULPHU DV WKH VXEVWUDWH DQG H[FHVV G HQ]\PH FRQFHQWUDWLRQ LQFUHDVLQJ IURP Q0 WR Q0 DOVR GHPRQVWUDWHG SURJUHVVLYHO\ LQFUHDVLQJ UDWHV XS WR KUn DW Q0 )LJXUH f 6LPLODUO\ UOLER FOHDYDJH LQFUHDVHG SURJUHVVLYHO\ ZKHQ WKH HQ]\PH Gf ZDV LQ H[FHVV ZLWK FRQFHQWUDWLRQV UDQJLQJ IURP Q0 WR Q0 &OHDYDJH UDWHV ZHUH DOZD\V KLJKHU IRU fDQJ ULERVHf SULPHU VXEVWUDWH WKDQ IRU UOLER VXEVWUDWH DW FRPSDUDEOH G FRQFHQWUDWLRQV DJDLQ VXJJHVWLQJ WKDW WKH VKRUWHU VXEVWUDWH PD\ KDYH OHVV VHOIELQGLQJ RSSRUWXQLWLHV ZKLFK FRPSHWH ZLWK DQG WKHUHIRUH UHGXFH ELQGLQJ DQG FDWDO\VLV E\ G 7KHVH HQ]\PH H[FHVV VDWXUDWLRQ H[SHULPHQWV UHYHDO WKDW WKH UDWH IRU WUDQV FOHDYDJH NFDWELf KUn IRU Uf LV GLVWLQFWO\ KLJKHU WKDQ WKH UDWH IRU FVFOHDYDJH NFDWXQLf KUnf 7KLV IROG LQFUHDVH LQ UDWH LV VXUSULVLQJ JLYHQ WKH H[SHFWDWLRQ WKDW WHWKHULQJ WKH fVXEVWUDWHf UHJLRQ WR WKH fHQ]\PHf UHJLRQ DV LQ WKH FVFOHDYLQJ ZRXOG VLJQLILFDQWO\ LQFUHDVH WKH HIIHFWLYH FRQFHQWUDWLRQ RI WKH VXEVWUDWH UHODWLYH WR WKH HQ]\PH DQG WKXV GUDPDWLFDOO\ LQFUHDVH WKH UDWH :KHQ WUDQVDFWLQJ HQ]\PH UHDFKHV FRQFHQWUDWLRQV RI WR Q0 WKH REVHUYHG UDWH LV WZLFH WKH UDWH RI FVFOHDYDJH DQG WKXV UHSUHVHQWV WKH FRQFHQWUDWLRQ DW ZKLFK WKH FVFOHDYDJH UDWH DQG UUDmVFOHDYDJH UDWH DUH HTXDO )LJXUH f 7KH HIIHFWLYH PRODULW\ RI OLQNLQJ WKH VXEVWUDWH WR WKH HQ]\PH LQ

PAGE 77

WKLV FDVH LV WKHUIRUH DSSUR[LPDWHO\ Q0 DERYH WKLV FRQFHQWUDWLRQ WUDQV FOHDYDJH LQFUHDVLQJO\ SUHGRPLQDWHV 7KH IROG LQFUHDVH LQ UDWH IRU WUDQV FOHDYDJH LV DOVR VXUSULVLQJ JLYHQ WKH H[SHFWDWLRQ WKDW ERWK WKH fFDWDO\WLFf DQG WKH fVXEVWUDWHf UHJLRQ RI LV XQFKDQJHG ZKHWKHU FOHDYDJH LV LQ FLV RU WUDQV 7KH LPSOLFDWLRQ LV WKDW WKH OLQNHU FRQQHFWLQJ WKH fVXEVWUDWHf DQG fHQ]\PHf LQ WKH FLVDFWLQJ VRPHKRZ LQWHUIHUHV VWHULF KLQGUDQFHf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fDQJ ULERVHf RU UOLER VXEVWUDWH LQ IRXUIROG H[FHVV RYHU G HQ]\PH HLWKHU DW RU Q0 VXEVWUDWHHQ]\PHf 0XOWLSOH WXUQRYHU ZDV DFKLHYHG EXW UDWHV ZHUH TXLWH VORZ $W WKH KLJKHVW FRQFHQWUDWLRQ RI HQ]\PH DQG VXEVWUDWH WXUQRYHUV RI VXEVWUDWH ZDV REVHUYHG DW KRXUV IRU fDQJ ULERVHf SULPHU VXEVWUDWH DQG KRXUV IRU WKH UOLER VXEVWUDWH $V VHHQ XQGHU VLQJOHWXUQRYHU FRQGLWLRQV WKH fDQJ ULERVHf VXEVWUDWH FOHDYHG ERWK IDVWHU DQG DW ORZHU FRQFHQWUDWLRQV WKDQ WKH IXOOOHQJWK UOLER VXEVWUDWH XQGHU PXOWLSOH

PAGE 78

WXUQRYHU FRQGLWLRQV FRPSDUH WKH GUDPDWLF LQFUHDVH LQ PXOWLSOH WXUQRYHU FOHDYDJH IRU U OLER FOHDYDJH ZKHQ G FRQFHQWUDWLRQ LV LQFUHDVHG IURP WR Q0 ZLWK WKH PRGHVW LQFUHDVH IRU fDQJ ULERVHf SULPHU FOHDYDJH LQGLFDWLQJ WKDW WKH fDQJ SULPHUf FOHDYDJH LV QHDU LWV PD[LPXP UDWH RI WXUQRYHUf 7KLV PD\ EH DWWULEXWDEOH WR WKH EHWWHU DELOLW\ RI WKH fDQJ ULERVHf VXEVWUDWH WR ELQG ZLWK RU GLVDVVRFLDWLRQ IURP WKH HQ]\PH RU LWV ORZHU SURFOLYLW\ IRU IRUPLQJ QRQSURGXFWLYH LQWHUDFWLRQV 0XOWLSOH WXUQRYHU FRQGLWLRQV ZHUH DOVR WHVWHG E\ KROGLQJ G FRQFHQWUDWLRQ FRQVWDQW DW Q0 DQG YDU\LQJ fDQJ ULERVHf FRQFHQWUDWLRQ IURP WR Q0 )LJXUH f $SSUR[LPDWHO\ WZR WXUQRYHUV ZHUH REVHUYHG LQ WKH ILUVW KRXUV ZKLOH RQO\ WXUQRYHUV ZHUH REVHUYHG E\ KRXUV ,I WKH UDWH OLPLWLQJ VWHS LV GLVDVVRFLDWLRQ RI WKH VXEVWUDWH IURP WKH HQ]\PH ZH ZRXOG H[SHFW WR VHH D fEXUVW SKDVHf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}6 GLVDVVRFLDWLRQ

PAGE 79

7HVWLQJ 5DWH RI $VVRFLDWLRQ DQG 'LVDVVRFLDWLRQ 9DOXHV REWDLQHG DERYH IRU FOHDYDJH UDWHV RI DUH FRPSDUDEOH WR UDWHV REWDLQHG IRU PDJQHVLXP LQGHSHQGHQW '1$]\PHV EXW DSSDUHQWO\ VORZHU WKDQ WKRVH UHSRUWHG IRU RWKHU ULERQXFOHDVH '1$]\PHV JHQHUDWHG E\ ,96 LW LV GLIILFXOW WR FRPSDUH UHVXOWV GLUHFWO\ VLQFH RWKHU '1$]\PHV DUH RIWHQ FKDUDFWHUL]HG DW KLJKHU WHPSHUDWXUHV DQG XQGHU DONDOLQH FRQGLWLRQV ERWK RI ZKLFK GUDPDWLFDOO\ LQFUHDVH FOHDYDJH UDWHVf 2QH SRWHQWLDO H[SODQDWLRQ IRU WKLV PD\ EH WKH GLIIHUHQFH LQ WKH GRXEOH VWUDQG '1$ FODPSV WKDW ZHUH HQJLQHHUHG WR IROG WKH '1$]\PH RQWR LWVHOI DQG WKXV EULQJ WKH SUHVXPDEO\ FDWDO\WLF 1 UHJLRQ LQ SUR[LPLW\ WR WKH ULERVH VXEVWUDWH VLWH 7KH ODFN RI SXEOLVKHG VWXGLHV RQ WKH LPSDFW RI WKHVH HQJLQHHUHG FODPSV SUHVXPDEO\ UHIOHFWV DQ DVVXPSWLRQ E\ WKH ,96 FRPPXQLW\ WKDW UHVXOWV REWDLQHG IURP LQ YLWUR VHOHFWLRQV VKRXOG EH LQGHSHQGHQW RI VXEWOH FKDQJHV LQ fVWDUWLQJ FRQGLWLRQVf DQG WKDW WKH RYHUDOO RXWFRPH LV LQGLFDWLYH RI WKH LQKHUHQW DELOLW\ RI QXFOHLF DFLGV WR SHUIRUP WKH VHOHFWHG EHKDYLRU 7KLV ORJLF VXJJHVWV WKDW WKH WZR QXFOHRWLGH FKDQJHV PDGH WR WKH f%ULERVHf SULPHU WR SURGXFH WKH fDQJ ULERVHf SULPHU VKRXOG PDNH OLWWOH GLIIHUHQFH WR WKH RYHUDOO UHVXOWV 1RQHWKHOHVV WKHVH WZR fSRLQW PXWDWLRQVf KDYH WKH SRWHQWLDO WR GUDPDWLFDOO\ DOWHU WKH DELOLW\ RI WKH '1$]\PH WR IROG RQWR LWVHOI LQ WKH ZD\ HQJLQHHUHG E\ %UHDNHU DQG -R\FH 7KH SURFOLYLW\ RI IRU DFWLQJ LQ WUDQV DOVR VXJJHVWV WKDW '1$ IROGLQJ PD\ EH FUXFLDO RXU XQGHUVWDQGLQJ RI DFWLYLW\ ,Q RXU VWDQGDUG SURWRFRO IRU LQLWLDWLQJ NLQHWLFV HQ]\PH DQG VXEVWUDWH DUH PL[HG WRJHWKHU LQ UHDFWLRQ EXIIHU DQG WKHQ KHDWHG WR r& IRU PLQXWHV IROORZHG E\ D VORZ FRRO WR r& RYHU PLQXWHV 7KH SXUSRVH RI WKLV fVORZ FRROf LV WR HQVXUH FRPSOHWH GHQDWXUDWLRQ RI VXEVWUDWH DQG HQ]\PH DW WKH RQVHW RI WKH H[SHULPHQW DQG DOORZ VSHFLILF FRQWURO RI WKH IROGLQJ FRQGLWLRQV ,QLWLDOO\ LW ZDV H[SHFWHG WKDW FDWDO\VLV ZDV GRPLQDWHG

PAGE 80

E\ FLVFOHDYDJH EXW RXU VWXGLHV DERYH LQGLFDWHG D VLJQLILFDQW DPRXQW RI WUDQVFOHDYDJH DW FRQFHQWUDWLRQV KLJKHU WKDQ Q0 :KHUHDV FLVFOHDYDJH HQWDLOV RQO\ LQWUDn PROHFXODU IROGLQJ UDIOVFOHDYDJH HQWDLOV WZR VHSDUDWH IROGLQJ LQWHUDFWLRQV LQWUDn PROHFXODU IROGLQJ ZKHWKHU LW LV HQ]\PH RU VXEVWUDWHf DQG mWHUPROHFXODU IROGLQJ ZKHWKHU LW LV EHWZHHQ HQ]\PH DQG VXEVWUDWH RU SRWHQWLDOO\ EHWZHHQ HQ]\PH DQG LQKLELWRUf $OWKRXJK SKHQRPHQDORJLFDOO\ GLIIHUHQW ERWK W\SHV RI IROGLQJ RFFXU VLPXOWDQHRXVO\ :H WHVWHG WKH LPSDFW RI RXU fVORZ FRROf IROGLQJ SURWRFRO RQ LQWUDPROHFXODU IROGLQJ E\ FRPSDULQJ WKH FOHDYDJH RI ULERVH ZLWK DQG ZLWKRXW WKH fVORZ FRROf IROGLQJ DW D FRQFHQWUDWLRQ ORZ HQRXJK DV WR IDYRU FVFOHDYDJH Q0f )LJXUH f 7KH RYHUDOO FOHDYDJH RI U ZDV KLJKHU ZKHQ ULERVH ZDV FRPSOHWHO\ GHQDWXUHG DQG VORZO\ FRROHG DV RSSRVHG WR VLPSO\ UHVXVSHQGLQJ WKH SHOOHW LQ +(3(6 EXIIHU DQG LQLWLDWLQJ WKH UHDFWLRQ E\ DGGLQJ 1D&O DQG 0J&K VXJJHVWLQJ WKDW WKH fVORZ FRROLQJf SURFHGXUH HQKDQFHV LQWUDPROHFXODU IROGLQJ ,W LV SRVVLEOH KRZHYHU WKDW WKLV UDWH HQKDQFHPHQW LV DWWULEXWDEOH WR WKH GHQDWXUDWLRQ LQGXFHG E\ WKH KLJK WHPSHUDWXUHV LQ WKH VORZ FRRO SURWRFRO UDWKHU WKDQ HQKDQFHPHQW RI WKH LQWUDPROHFXODU IROGLQJ VXJJHVWLQJ WKDW D IUDFWLRQ RI WKH ULERVH LV ORFNHG LQWR DQ LQDFWLYH FRQIRUPDWLRQ DV D UHVXOW RI WKH HWKDQRO SUHFLSLWDWLRQ ,I WKLV LV WKH FDVH LQVWHDG RI LPSURYLQJ LQWUDPROHFXODU IROGLQJ ZH PXVW VD\ WKDW WKH fVORZ FRROf SURFHGXUH LPSURYHV fLQWUDPROHFXODU UHIROGLQJf :KHQ WKH VDPH H[SHULPHQW ZDV GRQH DW KLJKHU WUDZVIDYRULQJ FRQFHQWUDWLRQV WKH LQLWLDO UDWH RI ULERVH FOHDYDJH LV XQFKDQJHG EXW WKH SODWHDX LV VOLJKWO\ KLJKHU ZLWK WKH fVORZ FRROLQJf WUHDWPHQW )LJXUH f 6LPLODUO\ FOHDYDJH RI Q0 U E\ Q0 G ZDV LGHQWLFDO UHJDUGOHVV RI ZKHWKHU WKH fHQ]\PHf DQG fVXEVWUDWHf ZHUH PL[HG

PAGE 81

SULRU WR RU VXEVHTXHQW WR WKH fVORZ FRROf IROGLQJ SURWRFRO GDWD QRW VKRZQf 7KLV VXJJHVWV WKDW DOWKRXJK WKH fVORZ FRROf IROGLQJ SURFHGXUH HQKDQFHV FVFOHDYDJH LW KDV OLWWOH HIIHFW RQ WUDQVFOHDYDJH ZKLFK EHFDXVH LW LV IDVWHU SUHGRPLQDWHV WKH LQLWLDO UDWHf 7KH LPSDFW RI WKH fVORZ FRROf IROGLQJ SURWRFRO RQ LQWHUPROHFXODU IROGLQJ ZDV DOVR WHVWHG GLUHFWO\ E\ PL[LQJ WKH HQ]\PH Gf DQG VXEVWUDWH HLWKHU UOLER RU fDQJ ULERVHf SULPHUf HLWKHU EHIRUH RU LPPHGLDWHO\ DIWHU WKH fVORZ FRROf SURFHGXUH )LJXUH DQG f 0XOWLSOH WXUQRYHU H[SHULPHQWV VKRZHG YHU\ OLWWOH GLIIHUHQFH LQ FOHDYDJH HLWKHU LQ LQLWLDO UDWHV RU FOHDYDJH SODWHDX ,Q IDFW FRPSOHWHO\ RPLWWLQJ WKH GHQDWXULQJVORZ FRRO SURWRFRO KDG QR HIIHFW RQ WUDQVFOHDYDJH )LJXUH f 7KLV IXUWKHU GHPRQVWUDWHV WKDW LQWHUPROHFXODU IROGLQJ LV QRW VLJQLILFDQWO\ HIIHFWHG E\ WKH fVORZ FRROf SURWRFRO 7R WHVW ZKHWKHU WKH UDWHOLPLWLQJ VWHS IRU ULERVH WUDQVFOHDYDJH LV WKH LQWHU PROHFXODU DVVRFLDWLRQ RU WKH FKHPLFDO VWHS D FKDVH H[SHULPHQW ZDV SHUIRUPHG ZLWK FROG FRPSHWLWRU 5LERVH ZDV LQFXEDWHG DW VXIILFLHQWO\ KLJK FRQFHQWUDWLRQ Q0f DV WR DFKLHYH VLJQLILFDQW WUDQVFOHDYDJH 7KH VDPSOH ZDV DOORZHG WR IROG IRU DQ DGGLWLRQDO WKLUW\ PLQXWHV DIWHU fVORZ FRROLQJf DW ZKLFK SRLQW IROG H[FHVV XQODEHOOHG FKDVH fDQJ ULERVHff ZDV DGGHG $ VLJQLILFDQW GHFUHDVH LQ FOHDYDJH ZDV REVHUYHG LPPHGLDWHO\ IROORZLQJ WKH DGGLWLRQ RI WKH FROG FKDVH DV FRPSDUHG WR ULERVH ZLWKRXW DGGHG FKDVHf )LJXUH f &RPSDULQJ f Q0 ULERVH SOXV FKDVHf ZLWK f Q0 ULERVH ZLWKRXW FKDVHf UHYHDOV WKDW WKH FKDVH KDV UHGXFHG FOHDYDJH OHYHOV EHORZ HYHQ WKDW VHHQ IRU D QHDUO\ DOO FLVFOHDYDJH UHDFWLRQ )LJXUH f ,I WKH UDWH RI VXEVWUDWH GLVDVVRFLDWLRQ IURP HQ]\PH LV IDVWHU WKDQ WKH FOHDYDJH VWHS NFDWELfm NLf WKH DGGLWLRQ RI D FROG FKDVH VKRXOG FRPSOHWHO\ HOLPLQDWH VXEVHTXHQW

PAGE 82

FOHDYDJH ,I KRZHYHU NFDWELf} NBL WKH DGGLWLRQ RI WKH FKDVH VKRXOG KDYH QR LPSDFW RQ VXEVHTXHQW FOHDYDJH DVVXPLQJ VXIILFLHQW WLPH KDV EHHQ JUDQWHG EHIRUH WKH DGGLWLRQ RI WKH FKDVH WR HQVXUH WKDW DOO WKH VXEVWUDWH LV ERXQG E\ WKH HQ]\PHf 7KH DGGLWLRQ RI FKDVH WR U UHVXOWHG LQ DQ LQWHUPHGLDWH UHVXOW FOHDYDJH ZDV GUDPDWLFDOO\ UHGXFHG EXW FOHDYDJH FRQWLQXHG WR LQFUHDVH IROORZLQJ WKH FKDVH 6HYHUDO VFHQDULRV PD\ EH SXW IRUZDUG WR H[SODLQ WKLV Df 7KH UDWH RI GLVDVVRFLDWLRQ RI VXEVWUDWH IURP HQ]\PH LV ZLWKLQ DQ RUGHU RI PDJQLWXGH RI WKH UDWH RI FOHDYDJH NFDWELf a NBLf Ef &VFOHDYDJH PD\ QRW EH FRPSOHWHO\ TXHQFKHG E\ WKH FKDVH RU DW OHDVW QRW DW WKH FRQFHQWUDWLRQ RI FKDVH XVHGf 7KH UHVLGXDO FOHDYDJH IROORZLQJ WKH DGGLWLRQ RI WKH FKDVH FRXOG WKHUHIRUH EH FLVFOHDYDJH 7KH DSSDUHQW UDWH RI FOHDYDJH ZLWK DGGHG FKDVH FRXOG EH ORZHU WKDQ WKH UDWH RI FLVFOHDYDJH ZLWKRXW FKDVH LI F]VFOHDYDJH QRUPDOO\ RFFXUV WKURXJK D SURFHVV RI IROGLQJ XQIROGLQJ DQG UHIROGLQJ DQG DW VRPH SRLQW GXULQJ WKH IROGHG VWDWH LW VHOIFOHDYHV *LYHQ WKLV VFHQDULR DGGLQJ FKDVH PD\ UHGXFH VRPH FVFOHDYDJH E\ FRPSHWLQJ ZLWK IRU F]VIROGLQJ 7KLV LPSOLHV WKDW NFDWXQLf a NLXQLf Ff 1RW DOO RI WKH VXEVWUDWH LV ERXQG E\ WKH HQ]\PH DW WKH WLPH WKH FKDVH LV DGGHG VR HYHQ WKRXJK NFDWELf!!NL WKH DGGLWLRQ RI WKH FKDVH FRPSHWHV IRU DVVRFLDWLRQ RI VXEVWUDWH WR HQ]\PH DQG VXEVHTXHQWO\ HOLPLQDWHV FOHDYDJH RI XQERXQG VXEVWUDWH 7KLV LPSOLHV WKDW WKH UDWH OLPLWLQJ VWHS LV DVVRFLDWLRQ NFDWELf}N_f Gf ,QVXIILFLHQW FKDVH KDV EHHQ DGGHG ,I NFDWELfm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f 6LPLODU UHVXOWV ZHUH REVHUYHG fULERVH SOXV FKDVHf ZDV JUDGXDOO\ FOHDYHG EXW DOZD\V DW OHYHOV IDU EHORZ WKH fPLQXV FKDVHf FRQGLWLRQ

PAGE 83

3UHYLRXV HQ]\PHVDWXUDWLRQ H[SHULPHQWV LQGLFDWH WKDW Q0 G LV VXIILFLHQW WR DSSURDFK VXEVWUDWH VDWXUDWLRQ WKHUHIRUH DUJXLQJ DJDLQVW VFHQDULR Ff DV D OLNHO\ H[SODQDWLRQ IRU WKH IDFW WKDW WKHUH LV UHVLGXDO FOHDYDJH IROORZLQJ WKH DGGLWLRQ RI WKH FKDVH 7KH UDWLR RI HQ]\PH Gf WR FKDVH fDQJ ULERVHff LQ WKHVH ODWWHU WZR H[SHULPHQWV ZDV HLWKHU RU 6OLJKWO\ OHVV FOHDYDJH ZDV REVHUYHG DIWHU WKH DGGLWLRQ RI FKDVH LQ WKH HQ]\PH[KDVH H[SHULPHQW VXJJHVWLQJ VFHQDULR Gf DV D SRVVLEOH H[SODQDWLRQ IRU UHVLGXDO FOHDYDJH IROORZLQJ WKH FKDVH &KDVH H[SHULPHQWV FRQGXFWHG ZLWK ULERVH FOHDYLQJ LQ FLV Q0f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f DGGHG DW WLPH GR QRW FRPSHWH ZLWK U FOHDYDJH ZKLOH KLJKHU OHYHOV RI FRPSHWLWRU Q0f UHGXFH EXW GR QRW HOLPLQDWH ULERVH FOHDYDJH Q0f )LJXUH DQG f 7DNHQ WRJHWKHU WKLV VXJJHVWV WKDW DW OHDVW VRPH U FVFOHDYDJH FDQ SHUVLVW GHVSLWH LQFUHDVLQJ FRQFHQWUDWLRQV RI FRPSHWLWRU VXSSRUWLQJ WKH SODXVLELOLW\ RI VFHQDULR Ef $ FKDVH H[SHULPHQW ZDV WKHUHIRUH SHUIRUPHG XVLQJ WUDFH DPRXQWV RI fDQJ ULERVHf SULPHU Q0 f DV D VXEVWUDWH ZLWK QHDUVDWXUDWLQJ OHYHOV RI G Q0 f HQ]\PH 7KLV VXEVWUDWH KDV QR LQWULQVLF FVFOHDYLQJ SRWHQWLDO HOLPLQDWLQJ WKH SRVVLELOLW\ WKDW FOHDYDJH IROORZLQJ WKH FKDVH LV FDXVHG E\ FVFOHDYDJH )XUWKHUPRUH WKH XQODEHOOHG

PAGE 84

fDQJ ULERVHf FKDVH ZDV DGGHG LQ HLWKHU RU IROG H[FHVV RYHU WKH HQ]\PH 7KH UHVXOWV LQGLFDWH WKDW FOHDYDJH ZDV JUHDWO\ UHGXFHG EXW FRQWLQXHG HYHQ DIWHU WKH DGGLWLRQ RI WKH FKDVH )LJXUHV DQG f 7KH UHGXFWLRQ ZDV VOLJKWO\ JUHDWHU ZLWK DGGLWLRQDO FKDVH 7RJHWKHU WKHVH H[SHULPHQWV LQGLFDWH WKDW WKH UDWH RI (r6 GLVDVVRFLDWLRQ LV JUHDWHU WKDQ WKH UDWH RI FKHPLFDO VWHS NBL } NFDWELff DQG WKH UHVLGXDO FOHDYDJH LV D IXQFWLRQ RI WKH UDWLR RI VXEVWUDWH DQG FKDVH WR HQ]\PH VFHQDULR Gff 7KH FKDVH H[SHULPHQWV LQGLFDWH WKDW LQWHUPROHFXODU DVVRFLDWLRQ IROGLQJf PD\ EH WKH UDWHOLPLWLQJ VWHS RU WKDW RQFH IROGHG WKH VXEVWUDWH GLVDVVRFLDWHV IURP HQ]\PH SULRU WR FOHDYDJH 7KLV ZDV IXUWKHU H[SORUHG E\ H[DPLQLQJ WKH WHPSHUDWXUH GHSHQGHQFH RI LQLWLDO FOHDYDJH UDWHV ,I WKH FKHPLFDO VWHS LV WKH UDWHOLPLWLQJ VWHS ORZHULQJ WKH WHPSHUDWXUH VKRXOG ORZHU WKH UDWH RI FOHDYDJH ZKHUHDV UDLVLQJ WKH WHPSHUDWXUH ZLOO LQFUHDVH WKH UDWH XQWLO WKH PROHFXOHV GHQDWXUHVf +RZHYHU ZKHQ IROGLQJ LV UDWH OLPLWLQJ ORZHULQJ WHPSHUDWXUH LV H[SHFWHG WR VWDELOL]H IROGLQJ DQG WKXV LQFUHDVH FOHDYDJH UDWH )LJXUHV DQG VKRZV WKDW ORZHULQJ LQFXEDWLRQ WHPSHUDWXUH LQFUHDVHV WKH LQLWLDO U FOHDYDJH UDWH LQ ERWK FLV DQG WUDQV FRQGLWLRQV $V WKH FKHPLFDO VWHS LV H[SHFWHG WR GHFUHDVH DSSUR[LPDWHO\ WZRIROG IRU HYHU\ r& GURS LQ WHPSHUDWXUH WKH REVHUYDWLRQ WKDW D UDWH LV XQFKDQJHG E\ D r& FKDQJH LQ WHPSHUDWXUH VXJJHVWV WKDW D EDODQFH LV UHDFKHG EHWZHHQ D VORZHU FKHPLFDO VWHS NFDWXQLf RU NFDWELff DQG D PRUH VWDELOL]HG DVVRFLDWLRQ VWHS NLXQLf RU N_Ecff DW ORZHU WHPSHUDWXUHV 7KH HQKDQFHPHQW RI WKH IROGLQJDVVRFLDWLRQ VWHS DW ORZHU WHPSHUDWXUHV LV DOVR VHHQ ZLWK G FOHDYLQJ UOLER LQ WUDQV GXULQJ VLQJOHWXUQRYHU H[SHULPHQWV )LJXUH %f 0XOWLSOHWXUQRYHU H[SHULPHQWV UHYHDO D UDWH HQKDQFHPHQW DW ORZHU WHPSHUDWXUHV GXULQJ WKH ILUVW b RI FOHDYDJH IROORZLQJ ZKLFK WKH UDWH LV UHGXFHG )LJXUH f 7KLV

PAGE 85

fEXUVW SKDVHf VHHQ XQGHU PXOWLSOH WXUQRYHU FRQGLWLRQV DW ORZHU WHPSHUDWXUH ZKLOH QRW VHHQ DW KLJKHU WHPSHUDWXUHVf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fDQJ ULERVHf SULPHU DV ZHOO DV D OLEUDU\ RI PROHFXOHV FRQWDLQLQJ WKH fDQJ ULERVHf SULPHU VXJJHVWV WKDW KDV D fELQGLQJ PRWLIf ZKLFK EDVHSDLUV ZLWK VRPH SDUW RI WKH fDQJ ULERVHf SULPHU FRPPRQ WR DOO VXEVWUDWHV 7KLV ELQGLQJ SRVLWLRQV D VHSDUDWH fFDWDO\WLF PRWLIf QHDU WKH ULERDGHQRVLQH 7R XQGHUVWDQG KRZ WKLV PD\ ZRUN LQ ERWK FLV DQG WUDQV WKH 6WHZDUW DQG =XNHU PIROG SURJUDP ZDV XVHG WR JHQHUDWH VHYHUDO VWUXFWXUH SUHGLFWLRQV IRU LQ FLV DQG WUDQV 2I WKH PDQ\ VWUXFWXUHV SURSRVHG E\ WKH PIROG SURJUDP RQO\ WKRVH WKDW SRVLWLRQHG WKH ULERDGHQRVLQH LQ D fFRQVWUDLQHG EXW DFFHVVLEOHf PDQQHU ZHUH VHOHFWHG VWUXFWXUHV ZHUH HOLPLQDWHG LI WKH ULERDGHQRVLQH ZDV HLWKHU HPEHGGHG LQ D OHQJWK\ KHOL[ LQDFFHVVLEOHf RU LI LW ZDV SRVLWLRQHG LQ WKH PLGGOH RI DQ H[WHQGHG ORRS XQFRQVWUDLQHGf )LJXUH VKRZV HLJKW RI ORZHVW IUHH HQHUJ\ VWUXFWXUH SUHGLFWLRQV 'HVSLWH WKH GLYHUVLW\ RI FDQGLGDWH VWUXFWXUHV VHYHUDO FRPPRQ IROGV ZHUH SUHVHQW DQG XVHG WR JURXS

PAGE 86

VWUXFWXUHV LQWR )DPLOLHV $ % DQG & 7KH IRXU VWUXFWXUHV RI )DPLO\ $ HDFK SRVVHVV D *$7$ VWHP ORRS WKH VXSHUVFULSW QXPEHUV UHIHU WR WKH SRVLWLRQ LQ WKH RYHUDOO VHTXHQFHf EXW YDU\ LQ WKH IROG RI QXFOHRWLGHV %RWK VWUXFWXUHV LQ )DPLO\ % VKDUH WKH VDPH IROG EHWZHHQ QXFOHRWLGHV DQG EXW YDU\ LQ WKH IROG DGRSWHG E\ QXFOHRWLGHV WKURXJK DQG WKURXJK 6WUXFWXUH $ DQG % VKDUH WKH VDPH IROG IRU QXFOHRWLGHV WKURXJK %RWK VWUXFWXUHV LQ WKH )DPLO\ & VKDUH WKH VDPH IROGLQJ SDWWHUQ IRU QXFOHRWLGHV (DFK RI WKH WKUHH )DPLOLHV $ % DQG & FDQ IROG LQ VXFK D ZD\ DV WR IRUP EDVHn SDLULQJ EHWZHHQ WKH O&*$&7&$&7$7 SDUW RI WKH ffDQJ ULERVHf SULPHUf DQG *7$*7*$&* SDUW RI WKH f SULPHUf 7KHVH f DQG f UHJLRQV FDQ DOVR IROG LQWR VHSDUDWH KDLUSLQV SUHYHQWLQJ LQWHUDFWLRQ ZLWK HDFK RWKHU DV VHHQ LQ VWUXFWXUH & 7KH f KDLUSLQ IRUPHG EHWZHHQ QXFOHRWLGHV DQG GLVUXSWV WKH GHILQLQJ HOHPHQWV RI ERWK )DPLO\ $ DQG % VWUXFWXUHV EXW QRW WKH )DPLO\ & VWUXFWXUH DV VHHQ LQ &f :H KDYH GHPRQVWUDWHG WKDW FDQ FOHDYH fDQJ ULERVHf SULPHU ERWK LQ WUDQV DQG ZKHQ LQFRUSRUDWHG DV QXFOHRWLGHV RI f ,I ZH DVVXPH WKDW WKH VWUXFWXUH IRUPHG ZKHQ FOHDYHV LQ FLV LV VLPLODU WR WKH VWUXFWXUH IRUPHG ZKHQ FOHDYHV fDQJ ULERVHf SULPHU LQ WUDQV WKHQ WKLV VXJJHVWV WKDW WKH FVIROGHG KDV VRPH EDVHSDLULQJ EHWZHHQ LWV ILUVW QW DQG WKH UHPDLQLQJ f HQG 7KH VWUXFWXUH VKRZQ LQ & DOORZV RQO\ PLQLPDO LQWHUDFWLRQ EHWZHHQ WKH fDQJ ULERVHf SULPHU UHJLRQ DQG WKH UHPDLQLQJ f KDOI EDVH SDLULQJ EHWZHHQ &$7**& DQG ,*&&7*f +RZHYHU WKH EDVHSDLULQJ EHWZHHQ QXFOHRWLGHV O&*$&7&$&7$7 DQG *7$*7*$&* VHHQ LQ WKH VWUXFWXUHV $ %O % DQG &O DOORZV VLJQLILFDQW LQWHUDFWLRQ EHWZHHQ WKH f DQG f KDOYHV RI

PAGE 87

7KHVH ODWWHU VWUXFWXUHV WKHUHIRUH VHHP PRUH SUREDEOH IRU ERWK GVFOHDYDJH DQG WUDQV FOHDYDJH 7KLV ZDV WHVWHG GLUHFWO\ E\ PXWDWLQJ QXFOHRWLGHV 7 WR $ WR & DQG WR & LQ )LJXUH f 7KHVH LQWURGXFHG PXWDWLRQV JUHDWO\ GLPLQLVKHG WKH DELOLW\ RI WR FOHDYH LWVHOI DV ZHOO DV WKH VXEVWUDWH UOLER LQ WUDQV FRPSDUHG WR ZLOGW\SH f ,QWURGXFLQJ WKH FRPSHQVDWRU\ FKDQJHV &* &* $7f SUHGLFWHG E\ WKH VWUXFWXUHV RI $ %O % DQG &O LQWR WKH VXEVWUDWH UOLER VLJQLILFDQWO\ LPSURYHG WUDQV FOHDYDJH ,QWURGXFLQJ WKH FRPSHQVDWRU\ FKDQJHV &* &* $7 LQWR WKH ULERVH FRQWDLQLQJ 7$ *& DQG *& DOVR GUDPDWLFDOO\ LPSURYHG FDWDO\VLV 7KLV GHPRQVWUDWHV WKDW WKH KHOL[ IRUPHG EHWZHHQ O&*$&7&$&7$7 DQG !*7$*7*$&* VKRZQ LQ WKH VWUXFWXUHV $O %O % DQG &O LV SUREDE\ IRUPHG LQ FLV DQG WUDQV $WWHPSWV ZHUH PDGH WR HOLPLQDWH VRPH RI WKH UHPDLQLQJ FDQGLGDWH VWUXFWXUHV $OWKRXJK VWUXFWXUH &O DSSHDUV TXLWH fOLQHDUf DQG GRHV QRW DSSHDU WR SRVLWLRQ D IROGHG fFDWDO\WLF PRWLIf QHDU WKH ULERDGHQRVLQH LW GRHV SRVVHVV WKH UHTXLVLWH EDVHSDLULQJ EHWZHHQ WKH O&*$&7&$&7$7 DQG *7$*7*$&* UHJLRQV %HFDXVH WKH f QXFOHRWLGHV RI FRUUHVSRQG WR WKH fDQJ ULERVHf VXEVWUDWH WKH HVVHQWLDO HOHPHQWV RI VWUXFWXUH &O FRXOG EH IRUPHG E\ D fDQJ ULERVHf HQ]\PHVXEVWUDWH FRQMXJDWH LI WKH fDQJ ULERVHf SULPHU GLVSODFHV WKH f KDOI RI ,I VWUXFWXUH &O ZHUH LQGHHG WKH FRUUHFW VWUXFWXUH WKH f KDOI RI QXFOHRWLGHV WKURXJK f VKRXOG EH VXIILFLHQW WR ELQG DQG FOHDYH WKH fDQJ ULERVHf VXEVWUDWH 7KLV ZDV WHVWHG E\ JHQHUDWLQJ WKH f KDOI RI ERWK f QWf DQG f QWff 7UXQFDWLQJ DW SRVLWLRQ HOLPLQDWHV WKH SRVVLELOLW\ RI IRUPLQJ VWUXFWXUH & ZKLOH VWLOO SHUPLWWLQJ VWUXFWXUH &O :KHQ PL[HG

PAGE 88

ZLWK fDQJ ULERVHf SULPHU QHLWKHU f QWf QRU f QWf FDWDO\]HG FOHDYDJH RI fDQJ ULERVHf VXJJHVWLQJ WKDW VWUXFWXUH &O LV QRW FRUUHFW 6FUHHQLQJ RI PXWDQWV RI UHYHDOHG WKDW DQ\ VLQJOH PXWDWLRQ RI SRVLWLRQ RU JUHDWO\ UHGXFHG FDWDO\WLF DFWLYLW\ (LWKHU RI WKH *$ RU *$ PXWDWLRQV KDYH WKH SRWHQWLDO WR GLVUXSW WKH VWUXFWXUH RI $ (DFK RI WKHVH PXWDQWV ZDV PDGH ZLWK WKH FRPSHQVDWRU\ PXWDWLRQV SUHGLFWHG E\ VWUXFWXUH $ *$&7 DQG *$&7f 1HLWKHU FRPSHQVDWRU\ PXWDWLRQ LPSURYHG FDWDO\VLV RYHU WKH VLQJOH PXWDQW )LJXUH f DUJXLQJ DJDLQVW WKH YDOLGLW\ RI VWUXFWXUH $ 7KH &7 PXWDQW DOVR KDG UHGXFHG FDWDO\WLF DFWLYLW\ WKLV PXWDWLRQ LV SUHGLFWHG WR GLVUXSW D EDVHSDLU ZLWK LQ DOO RI WKH )DPLO\ $ VWUXFWXUHV 7KH YDOLGLW\ RI WKHVH VWUXFWXUHV ZDV WHVWHG E\ PDNLQJ WKH FRPSHQVDWRU\ *$ PXWDWLRQ 7KH GRXEOHPXWDQW KRZHYHU FOHDYHG HYHQ VORZHU WKDQ WKH VLQJOH PXWDQW VXJJHVWLQJ DJDLQVW WKH YDOLGLW\ RI DQ\ RI WKH )DPLO\ $ VWUXFWXUHV )LJXUH f ,W LV KRZHYHU SRVVLEOH WKLV VWHPORRS SUHGLFWHG LQ )DPLO\ $ LV LQ IDFW SDUW RI WKH FRUUHFW VWUXFWXUH EXW WKH *& EDVHSDLU FDQQRW EH UHSODFHG ZLWK DQ $7 EDVHSDLU HLWKHU EHFDXVH WKH VWUHQJWK RI WKH $7 EDVHSDLU LV ORZHU RU EHFDXVH WKH RULJLQDO RU & SOD\HG D FDWDO\WLF IXQFWLRQ LQ DGGLWLRQ WR LWV VWUXFWXUDO UROH LQ WKH VWHP ORRSf 3RVLWLRQ RI ZLOGW\SH ZDV WKHUHIRUH FRQYHUWHG IURP D WR D & QDPHG f*&ff 7KH *& PXWDWLRQ KDV WKH SRWHQWLDO WR GLVUXSW WKH VWUXFWXUHV RI ERWK )DPLO\ $ DQG WKH VWUXFWXUH % DOWKRXJK LQ GLIIHUHQW PDQQHUV $V PHQWLRQHG DERYH LV SUHGLFWHG WR EDVHSDLU ZLWK & LQ )DPLO\ $ VWUXFWXUHV LV SUHGLFWHG WR EDVH SDLU ZLWK & LQ VWUXFWXUH % 7KH *& PXWDQW GHPRQVWUDWHG JUHDWO\ UHGXFHG VHOIn FOHDYDJH DOORZLQJ XV WR WHVW ERWK )DPLO\ $ DQG % VWUXFWXUHV E\ PDNLQJ WKH SUHGLFWHG

PAGE 89

FRPSHQVDWRU\ PXWDWLRQV f *& &*f DQG f *& &*ff 1HLWKHU RI WKH GRXEOHPXWDWLRQV UHVWRUHG FDWDO\WLF DFWLYLW\ VXJJHVWLQJ WKDW QHLWKHU VWUXFWXUH % QRU WKH VWUXFWXUHV RI )DPLO\ $ DUH LQ IDFW WKH DFWXDO VWUXFWXUH ,W UHPDLQV SRVVLEOH KRZHYHU WKDW RQH RI WKH VWUXFWXUHV RI )DPLO\ $ RU VWUXFWXUH % LV LQ IDFW FRUUHFW EXW LV FUXFLDO WR FDWDO\VLV IRU UHDVRQV EH\RQG LWV DELOLW\ WR EDVHSDLU ZLWK HLWKHU & RU & 7KH HYLGHQFH DERYH DUJXHV DJDLQVW WKH YDOLGLW\ RI )DPLO\ $ DQG & VWUXFWXUHV DV ZHOO DV VWUXFWXUH % OHDYLQJ RQO\ RQH FDQGLGDWH VWUXFWXUH UHPDLQLQJ %Of ,I VWUXFWXUH % UHSUHVHQWV WKH FRUUHFW IROG IRU DFWLQJ LQ FLV DQG WKH HVVHQWLDO VWUXFWXUDO HOHPHQWV DUH SUHVHUYHG ZKHQ FOHDYHV LQ WUDQV WKHQ QXFOHRWLGHV WKURXJK VKRXOG EH VXIILFLHQW IRU FOHDYLQJ WKH fDQJ ULERVHf VXEVWUDWH %HFDXVH WKLV WUXQFDWHG f WUXQFff FDQ FRQFHLYDEO\ DGRSW WKH RWKHU VWUXFWXUHV RI )DPLO\ $ DQG & WKH fDQJ ULERVHf VXEVWUDWH ZDV DOWHUHG WR LPSURYH ELQGLQJ LI LW DGRSWV WKH % VWUXFWXUH EXW QRW E\ RWKHU VWUXFWXUHV WKLV f%O VXEVWUDWHf ZDV SURGXFHG E\ WUXQFDWLQJ fDQJ ULERVHf WR LQFOXGH RQO\ QXFOHRWLGHV ZLWK D LQVHUWHG EHWZHHQ QXFOHRWLGHV DQG DQG $ FKDQJHG WR 7 UHVXOWLQJ LQ f7&$&7$7U$**$*$*7&ff 7KH FDWDO\WLF DELOLW\ RI f WUXQFf ZDV WKHQ WHVWHG ZLWK ERWK fDQJ ULERVHf DQG f% VXEVWUDWHf 1HLWKHU VXEVWUDWH ZDV FOHDYHG E\ f WUXQFf DUJXLQJ DJDLQVW RXU ILQDO SURSRVHG VWUXFWXUHV $ RU %O 6WUXFWXUH SUHGLFWLRQV ZHUH PDGH IRU )DPLO\ $ % DQG & ZLWK DFWLQJ LQ WUDQV )LJXUH f 7KH DELOLW\ RI WR FOHDYH WKH fDQJ ULERVHf VXEVWUDWH GHPRQVWUDWHV WKDW QXFOHRWLGHV DUH QRW D UHTXLUHG HOHPHQW RI WKH VXEVWUDWH IRU UUDZVFOHDYDJH 7KH VWUXFWXUH SUHGLFWLRQV VKRZQ IRU FOHDYLQJ LQ WUDQV DUH WKRVH XVLQJ fDQJ ULERVHf VXEVWUDWH EXW DUH HVVHQWLDOO\ WKH VDPH IRU IXOOOHQJWK VXEVWUDWHV

PAGE 90

)LJXUH VKRZV WKDW D IRXU EDVHSDLU KHOL[ FDQ EH IRUPHG EHWZHHQ QXFOHRWLGHV &$7* RI DQG WKH LGHQWLFDO QXFOHRWLGHV RI WKH fDQJ ULERVHf VXEVWUDWH 7KLV KHOL[ LV QRW IRUPHG ZKHQ FOHDYHV LQ FLV EHFDXVH WKH VHTXHQFH &$7* VHTXHQFH RFFXUV RQO\ RQFH 7KLV VKRUW KHOL[ SUHVXPDEO\ RIIHUV DGGLWLRQDO VWDELOL]DWLRQ EHWZHHQ WKH fDQJ ULERVHf VXEVWUDWH DQG WKH HQ]\PH ZLWKRXW GLVUXSWLQJ SRWHQWLDOO\ FDWDO\WLF KHOL[HV RI VWUXFWXUHV $ DQG & DOWKRXJK LW GRHV GLVUXSW SDUW RI VWUXFWXUHV $ DQG $f ,QWHUHVWLQJO\ D PXWDWLRQDO VFUHHQLQJ RI LVRODWHG D *$ PXWDQW FDOOHG fMXQH *$ff WKDW KDG UHGXFHG DFWLYLW\ FRPSDUHG WR ZLOGW\SH f ZKHQ WHVWHG DW Q0 D FRQFHQWUDWLRQ DW ZKLFK VLJQLILFDQW WUDQV FOHDYDJH RFFXUVf ,I LR 7KH VLJQLILFDQFH RI WKH SUHGLFWHG &$7* UDmVKHOL[ ZDV WHVWHG E\ LQWURGXFLQJ WKH FRPSHQVDWRU\ PXWDWLRQ LQWR WKH QRQFDWDO\WLF VXEVWUDWH UOLER WKH FRPSHQVDWRU\ PXWDQW LV FDOOHG fOLER &7ff 3UHYLRXV UDQVFOHDYDJH DVVD\V GHPRQVWUDWHG WKDW ZLOGW\SH G FDQ FOHDYH ULERVHOLER 7KH PXWDQW fMXQH *$f ZDV WKHQ WHVWHG IRU LWV DELOLW\ WR FOHDYH HLWKHU fUOLERf RU fUOLER &7f )LJXUH VKRZV WKDW WKH LQWURGXFWLRQ RI WKH &7 PXWDWLRQ LQWR WKH +ER VXEVWUDWH LPSURYHV FOHDYDJH E\ WKH PXWDQW fMXQH *$f VXJJHVWLQJ WKDW WKH &$7* KHOL[ LV LQGHHG LPSRUWDQW IRU UDQVFOHDYDJH 7KH UHGXFWLRQ LQ FOHDYDJH VHHQ ZKHQ fMXQH *$f FOHDYHV f+ER &7f DV FRPSDUHG WR ZLOGW\SH FOHDYLQJ ZLOGW\SH UOLER LV OLNHO\ WKH UHVXOW RI WKH ZHDNHU $7 EDVHSDLU DV RSSRVHG WR WKH *& EDVHSDLU 6WUXFWXUH SUHGLFWLRQV ZHUH PDGH ZLWK ERXQG WR WKH fDQJ ULERVHf VXEVWUDWH YLD WKH WZR H[SHULPHQWDOO\ VXSSRUWHG LQWHUDFWLRQV EDVHSDLULQJ EHWZHHQ &$7* RI WKH DQG WKH &$7* RI WKH VXEVWUDWH DQG EDVHSDLULQJ EHWZHHQ WKH O&*$&7&$&7$7 UHJLRQ RI WKH VXEVWUDWH DQG WKH *7$*7*$&* UHJLRQ RI WKH

PAGE 91

HQ]\PH 7KHVH UDALQWHUDFWLRQV LQWHUDFWLRQV HOLPLQDWH VWUXFWXUHV IRUPHG IRU FVIROGHG LQ WKH ILUVW QXFOHRWLGHV 0RVW RI WKH VWUXFWXUDO HOHPHQWV SUHGLFWHG IRU FVIROGHG DUH SUHVHUYHG LQ WKH UDZVIROGHG VXEVWUDWH FRPSOH[ )LJXUH f 7KH IUHH HQHUJ\ SUHGLFWLRQV IRU WKH UDLVIROGHG VXEVWUDWH FRPSOH[HV ZHUH DOO PXFK ORZHU WKDQ WKH FVIROGHG VWUXFWXUHV FLV G* DSSUR[LPDWHO\ NFDOPRO WUDQV VXEVWUDWH G* NFDOPROf 7KLV DJUHHV ZHOO ZLWK SUHYLRXV H[SHULPHQWDO UHVXOWV LQGLFDWLQJ WKDW RYHUDOO FOHDYDJH UDWH LV DIIHFWHG E\ IROGLQJ DQG RFFXUV IDVWHU LQ WUDQV 6WUXFWXUDO SURELQJ ZDV DOVR SHUIRUPHG XVLQJ FKHPLFDO PRGLILFDWLRQV VSHFLILF IRU VLQJOHVWUDQGHG '1$ 7KH UDZVIROGHG FRQIRUPDWLRQ ZDV IDYRUHG E\ LQFXEDWLQJ KLJK FRQFHQWUDWLRQ RI fODEHOHG G Q0f LQ UHDFWLRQ EXIIHU ZLWK DQG ZLWKRXW H[FHVV XQODEHOHG SULPHU S0f 7KH VDPSOHV ZHUH DOORZHG WR LQFXEDWH RYHUQLJKW WR PD[LPL]H IROGLQJ DIWHU ZKLFK WKH\ ZHUH WUHDWHG ZLWK HLWKHU SRWDVVLXP SHUPDQJDQDWH VSHFLILFDOO\ PRGLILHV 7f RU '06 VSHFLILFDOO\ PRGLILHV DW KLJK VDOW FRQFHQWUDWLRQVf IRU WZR PLQXWHV DQG ILYH PLQXWHV 5HFRYHUHG VDPSOHV ZHUH UXQ RQ D 3$*(XUHD JHO QH[W WR D ES ODGGHU $OO 7 DQG QXFOHRWLGHV GHPRQVWUDWHG VRPH VHQVLWLYLW\ WR WKH UHVSHFWLYH FKHPLFDO PRGLILFDWLRQV LQGLFDWLQJ WKDW HLWKHU VRPH LV QRW IROGHG DW DOO RU DGRSWV PXOWLSOH FRQIRUPDWLRQV 1RQHWKHOHVV FRQWUDVW OHYHOV ZHUH DGMXVWHG WR GLVFHUQ WKH PRVLSURWHFWHG QXFOHRWLGHV VXJJHVWLQJ WKH\ DUH EDVHSDLUHGf )LJXUH VKRZV WKH UHVXOW RI WKLV DQDO\VLV LQGLFDWLQJ WKDW 7 7 7 * * DQG DOO VKRZ VRPH GHJUHH RI GLIIHUHQWLDO SURWHFWLRQ 7KH SDWWHUQ RI SURWHFWHG DQG VHQVLWLYH SRVLWLRQV REVHUYHG XQGHU FKHPLFDO PRGLILFDWLRQV LV QRW VLPXOWDQHRXVO\ FRPSDWLEOH ZLWK DQ\ VLQJOH VWUXFWXUH SUHGLFWHG LQ )LJXUH IRU FVIROGHG RU )LJXUH IRU UDLVIROGHG

PAGE 92

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f VXPPDUL]HV WKH PDMRU UHVXOWV RI H[SHULPHQWV GHVFULEHG LQ WKLV FKDSWHU

PAGE 93

7DEOH 6XPPDU\ RI 5HVXOWV IURP &KDSWHU 5HDVRQV IRU FOHDYDJH SODWHDX RI f 0LVVLQJ ULERVH$ LQ ab RI SULPHU KLJKHU IRU FDJHG ULERVH GXH WR V\QWKHVLVf f $OWHUQDWLYH LQDFWLYH FRQIRUPDWLRQ b f 0XWDWLRQV WKDW ORZHU FDWDO\WLF DFWLYLW\ LQ ERWK SULPHU DQG 1 UHJLRQV f 1RW LQVXIILFLHQW ODVHU f 1RW FRPSOHPHQWDU\ VWUDQG LQKLELWLRQ .LQHWLF DQDO\VLV f FOHDYHV LQ FLV XQLPROHFXODUf DW ORZ FRQFHQWUDWLRQV Q0f f FOHDYHV LQ WUDQV ELRPROHFXODUf DW KLJK FRQFHQWUDWLRQV f FOHDYHV YDULRXV VXEVWUDWHV LQ WUDQV f NFDmELf KU f NREVXQLf KU f 7KH LQFUHDVH LQ WKH REVHUYHG UDWH RI FOHDYDJH DW ORZHU WHPSHUDWXUHV LQGLFDWHV WKDW WKH UDWH LV D EDODQFH EHWZHHQ WKH FKHPLFDO VWHS DQG WKH DVVRFLDWLRQ VWHS f 7KH UDWH RI (}6 GLVDVVRFLDWLRQ NBLELff IRU WUDQVFOHDYDJH LV JUHDWHU WKDQ WKH FKHPLFDO VWHS NFDWELff 6WUXFWXUDO $QDO\VLV f 0DQ\ VWUXFWXUHV IRU FVIROGHG DUH SRVVLEOH EXW QRQH KDYH \HW WR EH SURYHQ f 7KHUH LV DQ LQWHUDFWLRQ EHWZHHQ O&*$&7&$&7$7 DQG *7$*7*$&* WKDW RFFXUV LQ ERWK FLV DQG WUDmVIROGHG fM F LR f 7KHUH LV DQ LQWHUDFWLRQ EHWZHHQ &$7* RI WKH (Q]\PH f DQG 6XEVWUDWH WKDW FDQQRW RFFXU LQ PIROGHG

PAGE 94

/DPGD H[RQXFOHDVH VWURQJ EDVH ULERVH RU FDJHGU /$1( 5 5 5 5 F F F F QW IXOOOHQJWK r fr } QW FOHDYHG SURGXFW P )LJXUH /DPGD H[RQXFOHDVH GHJUDGHV WKH fSKRVSKRU\ODWHG FRPSOHPHQWDU\ VWDQG RI ULERVH DQG FDJHG 5LERVH DQG FDJHG ZHUH JHQHUDWHG YLD 3&5 ERWK WKH FDWDO\WLF DQG FRPSOHPHQWDU\ VWUDQG ZHUH ODEHOHG ZLWK DOSKD&73 7UHDWPHQW RI GRXEOH VWUDQGHG ULERVH ZLWK VWURQJ EDVH FRQYHUWV WKH FDWDO\WLF VWUDQG WR WKH FOHDYHG SURGXFW ZKLOH WUHDWPHQW ZLWK H[RQXFOHDVH GHJUDGHV WKH FRPSOHPHQWDU\ VWUDQG WR VLQJOHQXFOHRWLGHV QRW VKRZQ RQ JHOf 3&5 ZLWK WKH FDJHGULERVH SULPHU JHQHUDWHV D FRPSOHPHQWDU\ VWUDQG WKDW LV WKDW LV WKH VDPH OHQJWK DV WKH FOHDYHG FDWDO\WLF VWUDQG FRUUHVSRQGLQJ WR WKH ZKHUH WKH SRO\PHUDVH WHUPLQDWHV HORQJDWLRQ ZKHQ HQFRXQWHULQJ WKH FDJHGULERVH LQ WKH WHPSODWH

PAGE 95

r QW QW QW ODQH QXFOHRWLGH WHPSODWH H[WHQGHG ZLWK $ nHQG ODEHOOHG FRPSOLPHQWDU\ VWUDQG SULPHU VKRUW OLQNHU WDLO r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n RI FDJH QW FOHDYHG SURGXFW ODQH )LJXUH 6L]H GLIIHUHQFH EHWZHHQ FDWDO\WLF DQG FRPSOHPHQWDU\ VWUDQGV JHQHUDWHG E\ DV\PPHWULF 3&5 DOORZ IRU JHOSXULILFDWLRQ RI WKH FDWDO\WLF VWUDQG 7KH SUHVHQFH RI WKH FDJHULERVH LQ WKH WHPSODWH FDXVHV WUXQFDWLRQ RI WKH FRPSOHPHQWDU\ VWUDQG $V\PPHWULF 3&5 WKHUHIRUH JHQHUDWHV D FRPSOHPHQWDU\ VWUDQG WKDW LV HLWKHU ORQJHU RU VKRUWHU WKDQ WKH FDWDO\WLF VWUDQG

PAGE 96

A Grn Rr a ‘f‘ ‘" 6" ; FDJHG SULPHU SOXV YDU\LQJ SXOVHV RI ODVHU 1D2+ Rr mr VQr c!7 Z Yp )LJXUH 0D[LPXP GHSURWHFWLRQ RI FDJHGULERVH SULPHUV RFFXUV ZLWK SXOVHV RI ODVHU (QGODEHOHG fDQJ FDJHf SULPHU ZDV H[SRVHG WR LQFUHDVLQJ SXOVHV RI ODVHU Q0 P-SXOVHf DQG WKH DPRXQW RI GHSURWHFWLRQ GHWHUPLQHG E\ VXEVHTXHQW FOHDYDJH LQ 0 1D2+ IRU PLQXWHV DW r& &OHDYDJH UHDFKHV D SODWHDX RI DSSUR[LPDWHO\ b ZLWK SXOVHV RI ODVHU f$QJ FDJHf SULPHU H[SRVHG WR SXOVHV RI ODVHU EXW QRW WUHDWHG ZLWK VWURQJ EDVH VKRZHG DSSUR[LPDWHO\ b FOHDYDJH 8QSURWHFWHG fDQJ ULERVHf SULPHU ZDV b FOHDYHG XQGHU VWURQJ EDVH FRQGLWLRQV ([RQXFOHDVH SXULILHG IXOOOHQJWK ULERVH ZDV b FOHDYHG XQGHU VWURQJ EDVH FRQGLWLRQV r LQGLFDWHV LQFXEDWLRQ ZLWK 0 1D2+ UDWKHU WKDQ 0

PAGE 97

KRXUV )LJXUH &DJHG VKRZV VLPLODU NLQHWLFV WR ULERVH ZKHQ GHSURWHFWHG E\ SXOVHV RI ODVHU )RU DOO NLQHWLF DQDO\VLV HUURU LV HVWLPDWHG WR EH KRXU DQG OHVV WKDQ SHUFHQW FOHDYHG ,GHQWLFDO H[SHULPHQWV FRQGXFWHG LQ SDUDOOHO IHOO ZLWKLQ WKHVH HUURU HVWLPDWHV VHH )LJXUH % DV DQ H[DPSOHf

PAGE 98

$ U ODVHU F ODVHU F ODVHU F ODVHU F ODVHU F ODVHU F ODVHU % ULERVH ODVHU F ODVHU F ODVHU F ODVHU F ODVHU F ODVHU F ODVHU )LJXUH 3URJUHVV FXUYHV RI FDJHG FOHDYDJH GHPRQVWUDWH PD[LPXP GHSURWHFWLRQ ZLWK SXOVHV RI ODVHU $f &DJHG ZLWK OHVV WKDQ SXOVHV RI ODVHU KDG JUHDWO\ UHGXFHG FOHDYDJH SODWHDX LQGLFDWLQJ WKDW WKH VDPSOH ZDV QRW FRPSOHWHO\ GHSURWHFWHG %f :KHQ FOHDYDJH ZDV QRUPDOL]HG E\ DGMXVWLQJ WKH SHUFHQW FOHDYHG RXW RI RQO\ WKH IUDFWLRQ RI VDPSOH WKDW ZDV GHSURWHFWHG GHWHUPLQHG E\ 1D2+ WUHDWPHQW WKH NLQHWLF SURILOHV RYHUODS IRU DOO VDPSOHV H[FHSW ODVHUf

PAGE 99

)LJXUH 7KH FOHDYDJH SODWHDX RI FDJHG UHDFKHV D PD[LPXP ZLWK SXOVHV RI ODVHU 6DPSOHV RI FDJHG ZHUH LQLWLDWHG ZLWK LQFUHDVLQJ DPRXQWV RI ODVHU DQG DOORZHG WR VHOIFOHDYH XQWLO WKH UHDFWLRQ UHDFKHG SODWHDX 6DPSOHV LQLWLDWHG ZLWK SXOVHV RI ODVHU DQG PRUH UHDFKHG HTXLYDOHQW FOHDYDJH SODWHDXV RI b DERXW b ORZHU WKDQ IRU DQ HTXLYDOHQW H[SHULPHQW ZLWK ULERVH )LJXUH 0D[LPXP GHSURWHFWLRQ RI FDJHG RFFXUV ZLWK SXOVHV RI ODVHU 6DPSOHV RI FDJHG ZHUH H[SRVHG WR LQFUHDVLQJ DPRXQWV RI ODVHU DQG WKHQ LQFXEDWHG ZLWK VWURQJ EDVH IRU KRXU DW f& WR GHWHUPLQH WKH DPRXQW RI GHSURWHFWLRQ &DJHG H[SRVHG WR SXOVHV RI ODVHU LV b FOHDYHG E\ WKH VWURQJ EDVH $GGLWLRQDO ODVHU GRHV QRW LQFUHDVH WKH DPRXQW FOHDYHG E\ VWURQJ EDVH 5LERVH KDV VOLJKWO\ KLJKHU FOHDYDJH XQGHU VLPLODU FRQGLWLRQV

PAGE 100

U ODVHU U ODVHU U ODVHU U ODVHU U ODVHU KRXUV U ODVHU U ODVHU U ODVHU U ODVHU U ODVHU KRXUV )LJXUH 5LERVH LV QRW GDPDJHG E\ ODVHU $f )XOO WLPH FRXUVH DQG %f OLQHDU LQLWLDO SKDVH RI ULERVH VHOI FOHDYDJH IROORZLQJ LQFUHDVLQJ DPRXQWV RI ODVHU SXOVHV 7KH FOHDYDJH SURILOH RI ULERVH LV XQFKDQJHG E\ H[SRVXUH WR LQFUHDVLQJ QXPEHU RI ODVHU SXOVHV

PAGE 101

Q0 U PP 0J&L Q0 U PP 0J&, Q0U PP 0J&, Q0 U PP0J&O Q0 U PP 0J&, Q0 U PP 0J&, Q0 U PP 0J&, Q0 U PP 0J&, )LJXUH 5LERVH FOHDYDJH LV QRW DOWHUHG E\ IROG YDULDWLRQ LQ 0J&K FRQFHQWUDWLRQ 9DULRXV FRQFHQWUDWLRQV RI U ZHUH LQFXEDWHG LQ P0 +(3(6 S+ 0 1D&O DQG HLWKHU P0 RU P0 0J&K ,QFUHDVLQJ WKH 0J&( FRQFHQWUDWLRQ WKUHHIROG KDG QR VLJQLILFDQW HIIHFW RQ U FOHDYDJH UDWH DW DQ\ FRQFHQWUDWLRQ )LJXUH 5LERVH LV LQDFWLYH ZLWKRXW 1D&O 5LERVH ZDV LQFXEDWHG ZLWK P0 +(3(6 S+ SOXV HLWKHU 0 1D&O P0 0J& RU P0 FDFRG\ODWH

PAGE 102

$ % )LJXUH 5LERVH FOHDYDJH LV XQDIIHFWHG E\ WKH DGGLWLRQ RI b FRPSOHPHQWDU\ VWUDQG EXW DGGLWLRQ RI HTXLPRODU DPRXQWV JUHDWO\ ORZHUV FOHDYDJH UDWH DQG SODWHDX $f )XOO WLPH FRXUVH DQG %f LQLWLDO WLPH FRXUVH IRU ULERVH ZLWK b DGGHG FRPSOHPHQWDU\ VWUDQG &f )XOO WLPH FRXUVH DQG 'f LQLWLDO WLPH FRXUVH IRU ULERVH ZLWK HTXLPRODU DGGHG FRPSOHPHQWDU\ VWUDQG 5LERVH ZDV LQFXEDWHG ZLWK H[RJHQRXVO\ DGGHG FRPSOHPHQWDU\ VWUDQG DW HLWKHU b RI WKH WRWDO DPRXQW RI ULERVH RU DW HTXLPRODU DPRXQW (DFK UHDFWLRQ ZDV GRQH DW KLJK Q0f DQG ORZ f Q0 ULERVH FRQFHQWUDWLRQ DQG FRPSDUHG DJDLQVW SXUH ULERVH DW D UDQJH RI FRQFHQWUDWLRQV

PAGE 103

F Q0f F Q0f XQODEHOHG G Q0f DW KUV F Q0f GLOXWHG WR Q0 DW KRXUV KRXUV )LJXUH FOHDYDJH SODWHDX LV QRW D UHVXOW RI DQ DSSURDFK WR FKHPLFDO HTXLOLEULXP ZDV DOORZHG WR UHDFW XQWLO LW DSSURDFK FOHDYDJH SODWHDX DW ZKLFK SRLQW LW ZDV HLWKHU GLOXWHG IROG RU LQFXEDWHG ZLWK IROG DGGLWLRQDO XQODEHOHG G 1HLWKHU FRQGLWLRQ DOWHUHG WKH FOHDYDJH SODWHDX RI LQ D WLPHGHSHQGHQW PDQQHU U U QW SURGXFW U QW SURGXFW U t QW SURGXFW UOLEUDU\ QW SURGXFW UOLEUDU\ QW SURGXFW UOLEUDU\ t QW SURGXFW )LJXUH $GGLWLRQ RI WKH QXFOHRWLGH DQG WKH QXFOHRWLGH FOHDYDJH SURGXFWV DORQH RU WRJHWKHU GR QRW DOWHU WKH FOHDYDJH SURILOH RI Q0 HDFKf $GGLWLRQ RI WKH QXFOHRWLGH FOHDYDJH SURGXFW WR WKH UOLEUDU\ VXEVWUDWH UHVXOWHG LQFUHDVHG FOHDYDJH EXW DW DQ H[WUHPHO\ VORZ UDWH 7KLV FOHDYDJH RI WKH UOLEUDU\ VXEVWUDWH LQ WKH SUHVHQFH RI WKH QW IUDJPHQW LV ORZHUHG E\ WKH DGGLWLRQ RI WKH QW FOHDYDJH SURGXFW

PAGE 104

U Q0f U Q0f XQFOHDYHG U JHO SXULILHG DW KRXUV )LJXUH *HOSXULILFDWLRQ RI DW FOHDYDJH SODWHDX UHVXOWV LQ DGGLWLRQDO FOHDYHDJH LQGLFDWLQJ D IUDFWLRQ RI LV IROGHG LQWR DQ LQDFWLYH FRQIRUPDWLRQ 5LERVH ZDV DOORZHG WR VHOIFOHDYH IRU KRXUV QHDULQJ FOHDYDJH SODWHDX DW ZKLFK SRLQW KDOI WKH VDPSOH ZDV JHOSXULILHG RQ D GHQDWXULQJ JHO $IWHU UHVXVSHQGLQJ WKH JHOSXULILHG ULERVH fJSff LQ UHDFWLRQ EXIIHU WR Q0 DGGLWLRQDO FOHDYDJH ZDV GHWHFWHG )LJXUH 5HKHDWLQJ ULERVH UHVXOWV LQ DGGLWLRQDO FOHDYDJH LQGLFDWLQJ D IUDFWLRQ RI LV IROGHG LQWR DQ LQDFWLYH FRQIRUPDWLRQ 5LERVH ZDV DOORZHG WR UHDFW XQWLO LW UHDFKHG FOHDYDJH SODWHDX DW ZKLFK SRLQW KDOI RI WKH VDPSOH ZDV GHQDWXUHG E\ KHDWLQJ WR r& IRU PLQXWHV DQG VORZO\ FRROHG WR r& RYHU PLQXWHV 7KLV UHn IROGLQJ SURFHGXUH UHVXOWHG LQ DQ DGGLWLRQDO b FOHDYDJH

PAGE 105

FORQH QDPH ZW FRSLHV f§ $$ $ &$7 MDQ3 $$ $&$7 MDQ5 r $$ $&$7 VHSWRILFR f§Ur $$ $&$7 & $&7&7&$&$7&$7 ; $ &$&$& &$$7$ &&7 $7$$ 77 7$ & $&7&7&$&$7&$7$& $ &$&$& &$$7$ &&7 $7$$ 77 7$ & $&7&7&$&$7&$7 ; $ &$&$& &$$7$ $&7 $7$$ 77 7$ & $&7&7&$&$7&$7; $ ;$&$& &$$7$ &&7 $7$$ 7 7$ $& $& $& $& MDQ MDQ/n MDQ MDQ MDQOr MDQ r MDQn MDQIIOO VHSWXS OE PFO OE OEt PH MXQH MXQH MXQH M$Q$ MXQH MXQH MXQH *$ PH MDQ MXQH *$ 0& MDQ*n MXQH XS OE VHSWOEO ZW &7 &7 &7 &7 &7 &7I &7 &7 &7 &7r &7r &7 &7 &7 &7$ &7 &7n &7r L &7rn &7 &7n &7 &7 &7 &7 &77 &7 &7r &7 &7 &$$$77&7$$7$& &$ $$77&7$$7$& &$ $$77&7$$7$& &$ n$$77&7$$7$& &$ $$77&7$$7$& &$ $$77&7$$7$& &$ $$77&7$$7$& &$ $$77&7$$7$& &$ $$77&7$$7$& &$ $$77&7$$7$& &$ $$77&7$$7$& &$ $$77&7$$7$& &$ $$77&7$$7$& &$ $$7717$$7$& &$ $$77&7$$7$& &$ $$77&7$$7$& &$ $$77&7$$7$& &$ $$77&7$$7$& &$ $$77&7$$7$& &$$$77&7$$7$& &$ $$77&7$$7$& &$ $$77&7$$7$& &$ $$77&7$7$& &$ $$77 &7$&7$ & $$$77&7$$7$& &$ $$77&7$$7$&& &$ $$77&7$$7$& &$ $$77&7$$7$& &$ $$77&7$$7$& &$ $$77&7$$7$& &7 &$ &7 &$ &7 &$ &7 &$ &7 &$ &7 &$ &7 &$ &7&$ &7&$ &7 &$ &,9n&$ &7&$ &7 &$ &77 &$ FRSLHV &7 &$ $$77&7$$7$& $$77&7$$7$& $$77&7$$7$& $$77&7$$7$& $$77&7$$7$& $$77&7$$7$& $$77&7$$7$& $$77&7$$7$& $$77&7$$7$& $$77&7$$7$& $$77&7$$7$& $$77&7$$7$& $$77&7$$7$& $$77&7$$7$&& $$77&7$$7$& $&7&$ $&7&$ $&7&$ $&7&$ $&7&$ $&7&$ $&7&$ $&7& $&7&$ $&7&$ $&7&$ $&7&$ $&7&$ $&7&$ $&7&$ $&7&$ $&7&$ $&7&$ $&7&$ $&7&$ $&7&$ $&7&$ $&7&$ $&7&$ $&7&$ $&1&$ $&7&$ $&7&$ $&7&$ $&7&$ $&7&$ $&7&$ $&7&$ $&7&$ $&7&$ $&7&$ $&7&$ $&7&$ $&7&$ $&7&$ $&7&$ $&7&$ $&7&$ $&1&$ $&7&$ &7$$7$ &7 $ &7$7$ &7$7 &7$7 &7$7 &7$7 &7$7 7$7r &7$7 &7$7rn &7$7 &7$ &7$7 &7$7 &77$ &77$ &77$ &7$$7$ &7$$ &7$7$ &7$7$ &7$7$ f &7$7$ &7$7$ &7$7$ &7$7$ &7$7$ &7$7$ &7$7$ $$ $&$7 $$ $&$7 $$ $&$7 $$n$&$7 $$ $&$7 $$ $&$7 $$$&$7 $$ $&$7 $ $&$7 $$ $&$7 $$ $&$7 $$ $&$7 $$ $&$7 $$ $&$7 $$ $&$7 $$ $$ $$ $$ $$ $$ $$ f$$ $$ $$ 1$$ $$ $$ $$ $$ &7$ &7$&7$ &7$ &7$ &7$ &7$ &7$ &7$ &7$ &7$ &7$ &7$ &7$ &7$ 7$ $$ 7$ $$ 7$ $$ 7$ $$ 7$ $$ 7$ f $$ 7$6 $$ 7$ $$ 7$n $$ 7$ $$ 7$ $$ 7$ $$ 7$ f $$ 7$ 1$$ 7$b $$ $&$7 $&$7 $&$7 $&$7 $&$7 $&$7 $&$7 $&$7 $&$7 $&$7 $&$7 $&$7$ $&$7 $&$7 7&$7 $&$7 $&$7 $&$7 $&$7 $&$7 $&$7 $&$7 $&$7 $&$7 $&$7 $&$7 $&$7 $&$7 $&$7 $&$7 & $&7&7&$&$7&$7 L& $ &$&$& & $&7&7&$&$7&$7*&* $ &$&$& f & $&7&7&$&$7&$7*&I $ &$&$&$ & $&7&7&$&$7&$7 &*$ &$&$& ‘ & $&7&7&$&$7&$7 m‘ $ &$&$& & $&7&7&$&$7&$7*& $ &$&$& & $&7&7&$&$7&$7 ; $ &$&$&$ & $&7&7&$&$7&$7 ; $ ;$&$& ‘ & $&7&7&$&$7&$7 & $ c&$&$& & $&7&7&$&$7&$7 ; $ ;$&$& & $&7&7&$&$7&$7 & $ &$&$& & $&7&7&$&$7&$7 ;A$ &$&$& & $&7&7&$&$7&$7 ; n$ &$&$& & $&7&7&$&$7&$7 & $ &$&$& & $&7&7&$&$7&$7 m*$& &$&$& & $&7&7&$&$7&$7 ; $ &$&$& & $&7&7&$&$7&$7;*$ &$&$& & $&7&7&$&$7&$7 ; $*&$&$& & $&7&7&$&$7&$7 ; $ ;$&$& & $&7&7&$&$7&$7 & $ &$&$& & $&7&7&$&$7&$& ; $ &$&$& &$$&7&7&$&$7&$7*&*$*&$&$& & $&7&7&$&$7&$7 ;*$ ;$&$& &$&7&7&$&$7&$7 ; $ &$&$& & $&7&7&$&$7&$7*&*$*&$&$& & $&7&7&$&$7&$7; $ ;$&$& & $&7&7&$&$7&$7 ; *$ &$&$& & $&7&7&$&$7&$7 ;*$&&$&$& 1 $1717 &$ &$7&$7 ; *$ ;$&$& & $&7&7&$&$7&$7 ; $ ;$&$& f&$$7$ &&7 $7$$ 77 7$ f&$$7$ &&7 $7$$ Y 77 7$ &$$7$ 7&7$7$$**77**7$ &$$7$ &&7 $7$$ 77 7$ f&$$7$ &&7 $7$$ 77 7$ ‘&$$7$ &&7 $7$$ 77 7$ &$$7$ &&7 $7$$ 77**7$ &$$7$ ;&7 $7$$ 77 7$ &$$7$ &&7 $7$$ 77 7$ &$$7$ &&7 $7$$ 77**7$ &$$7$ &&7 $7$$ 777$ &$$7$ ;&7 $7$$**77 *7$ &$$7$ &&7 $7$$ 77**7$ &$$7$ &&7 $7$$77 -7$ &$$7$ &&7 $7$$ 77 ‘ L7$ &$$7$ &&7 $7$$ L 77 f 7$ &$$7$ &&7 $7$$ 77 7$ &$$7$ &&7 $7$$ 77 7$ &$$7$ &&7 $7$$ 77**7$ &$$7$ &&7 $7$$ 77 U 7$ &$$7$ &&7 $7$$ 77 7$ &$$7$ &&7 $7$$**77 7$ &$$7$ &&7 $7$$ 77 7$ &$$7$ &&7 $7$$ m77**7$ &$$7$ &&7 $7$$77**7$ &$$7$ ;&7 $7$$ L 77**7$ &$$7$ FFW DWDD ,77 7$ &$$7$7&&7 $7$$ ; 77**7$ &$$7$ &&7 $7$$ 77 27$ &$$7$ &&7 c$7$$**77**7$ 7 $& 7 $& 7 $& 7 $& 7 $& 7 $& 7 $& 7 $& 7 $& 7 $& 7 $& 7 $& 7 $& 7 & 7 $& $& $& $& $& $& $& $& $& $& $& $& $& $& $& $& $&717&$&$7&$7 ; $ &$&$& $&7&7&$&$7&$7; $7&$&$& $&7&7&$&$7&$7 ; $7&$&$& $&7&7&$&$7&$7*& $ &$&$& $&7&7&$&$7&$7 ;*$ &$&$& $&7&7&$&$7&$7; *$ &$&$& $&7&7&$&$7&$7 ; $ &$&$& $&7&7&$&$7&$7*& $ &7&$& $&717&$&$7&$7 ;*$ &$&$& & $&7&7&$&$7&$7 &$ &$&$& & $&7&7&$&$7&$7 ; $ &$&$7 & $&7&7&$&$7&$7 ; $ &$&$& & $&7&7&$&$7&$7;$&$&$& & $&7&7&$&$7&$7; $*&$&$&& & $&7&7&$&$7&$7 ; $ ;$&$& &$$7$ &&7 7&$$7$ &&7 &$$7$ &&7 &$$7$ &&7 &$77$ &&7 &$$7$ &&7 &$$7$ &&7 &$$7$ &&7 &$$7$ &&7 &$$7$ 7&7 &$$7$ &&7 &$$77 &&7 &$$7$ &&7 &$$7$ &&7 &$$7$ &&7 $7$$;77*;$ $7$$ &77 7$ $7$$ ;77* 7$ $7$$*&77*7$ DWDDJJWWJJWD $7$$**77**7$ $$7$$**77**7$ A$7$$**77**7$ $$7$$**77**7$ $7$$&$&L77**7$ $7$$ 77 7$ $$7$$**77**7$ $$7$$**77**7$ $$7$$**77**7$ $$7$$**77**7$ 7 $& 7 $& 7 $& 7 $& 7 $& 7 $& 7 $& 7 $& 7 $& 7 $& 7 $& 7 $& 7 & 7 $& 7 $& WR ZW LQ FLL 7$$ &77 &$& 7$$ &77 &$& f§ 7$$ &77 &$& 7$$ &77 &$& 7$$ &77 &$& 7$$ &77 &$& 7$$ &77 &$& f§ 7$$ &77 &$& 7$$ &77 &$& 7$$ &77 &$& 7$$ &77 &$& 7$$ &77 &$& 7$$ &77 &$& 7$$ &77 &$& 7$$ &77 &$& 7$$ &77 &$& 7$$ &77 &$& 7$$ &77 && 1$$ &77 &$& 7$$ &77 &$& 7$$ &77 &$& 7$$ &77 &$& 7$$ &77 &$& 7$$ &77 &$& 7$$ &77 &$& 7$$ &77 &$& f§ 7$$ &77 &$& 7$$ &77 &$& 7$$ &77 &$& 7$$ &77 &1& 7$$ &77 &$& 7$$ &77 &$& 7$$ &77 &$& 7$$ &77 &$& 7$$ &77 &$& 7$$ &77 &$& 7$$ &77 &$& 7$$ &77 &$& f§ 7$$ &77 &$& 7$$ &77 &$& f§ $$$ &77 &$& 7$$ &77 &$& 7$$ &77 &$& 7$$ &77 &$& f§ 7$$ &77 &$& 7$$ &77 &$& 7$$ &77 &$& 7$$ &77 &1& 7$$ &77 &$& )LJXUH 6HTXHQFH DOLJQPHQW RI FOHDYHG DQG XQFOHDYHG FORQHG VHTXHQFH YDULDQWV 0XWDWLRQV RI DUH KLJKOLJKWHG LQ \HOORZ f1f UHJLRQ EHWZHHQ SULPHUV LV XQGHUOLQHG LQ WKH ILUVW DQG ODVW VHTXHQFH &DWDO\WLF UDWH UHODWLYH WR ZWf VKRZQ WR ULJKW 92 3

PAGE 106

$ U ZWf >QP@ UPXW =f 7&f UPXW 5n &$f UPXW Nf 7f UPXW Ef $7f UPXW f *$f UPXW 3n *$f UPXW LQV$ &7f UPXWXS &7f UPXW* *f UPXW *$f UPXW *$f UPXW 7f % U ZWf >QP@ UPXW f 7&f UPXW 5n &$f UPXW OR 7f UPXW E $7f UPXW *$f UPXW 5n *$f UPXW LQV$ &7f UPXW XS &7f UPXW *f UPXW *$f UPXW *$f UPXW 7f KRXUV )LJXUH 0XWDQWV RI KDYH UHGXFHG FDWDO\WLF DFWLYLW\ $f &RPSOHWH WLPH FRXUVH %f /LQHDU LQLWLDO SKDVH RI WLPH FRXUVH 0XWDWLRQV RI ZHUH FRPPRQ LQ WKH XQFOHDYHG IUDFWLRQ RI IROORZLQJ FOHDYDJH SODWHDX .LQHWLF DQDO\VLV ZDV GRQH RI PDQ\ RI WKHVH FORQHV DOO DW a Q0f 0RVW PXWDWLRQV JUHDWO\ UHGXFHG FDWDO\WLF DFWLYLW\ ZKHQ FRPSDUHG WR ZLOGW\SH

PAGE 107

D ‘ GZO >Q0@ 0LEOR >Q0@ GPXW $ ER[ m GPXW r *& *& 7$f R GPXW 7f $ GPXW r $f 9 GPXW XS & $ $f m GPXW PF$n LQV$f A ; GPXW *$f > Q0@ UOLEOR > Q0@ R UPXW *& *& 7$f Q0f UOLEN! > Q0@ D GZW A OLEORPXW &* &* $7f Q0@ $ % GZW >Q0@ 0LEN! >Q0@ GPXW $ ER[ GPXW *& *& 7$f GQQXW 7f GPXW $f GP XW XS & $ $f GPXW PF$n LQV$f GPXW *$f > Q0@ 1LEOR > Q0@ UPXW *& *& 7$f >Q0@ A UWWEOR > Q0@ GZW A 0EORPXW &* &* $7f > Q0@ )LJXUH 0XWDQWV RI KDYH UHGXFHG DFWLYLW\ ZKHQ FOHDYLQJ UOLEOR LQ WUDQV $f &RPSOHWH WLPH FRXUVH %f /LQHDU LQLWLDO SKDVH RI WLPH FRXUVH 7KH DELOLW\ RI VHYHUDO PXWDQWV WR FOHDYH LQ WUDQV ZDV WHVWHG E\ LQFXEDWLQJ YDULDQWV Q0 XQOHVV QRWHG GLIIHUHQWO\f ZLWK UOLERVXEVWUDWH Q0 XQOHVV QRWHG GLIIHUHQWO\f 0RVW PXWDWLRQV JUHDWO\ UHGXFHG FDWDO\WLF DFWLYLW\ ZKHQ FRPSDUHG WR ZLOGW\SH 1XFOHRWLGHV ZHUH FKDQJHG WR $*7$&7 WR FUHDWHfGPXW $ ER[f

PAGE 108

SHUFHQW FOHDYHG H[S ILW ‘ Q0 U Q0 U Q0 U Q0 U Q0 U Q0 U % P mIUDFWLRQ FOHDYHG$WRXU ‘ Q0 U r Q0 U Q0 U } Q0 U A Q0 U r‘ Q0 U )LJXUH FOHDYDJH UDWH LV FRQFHQWUDWLRQ GHSHQGHQW $f &RPSOHWH WLPH FRXUVH IRU U DW YDULRXV FRQFHQWUDWLRQV 5DWHV DUH HVWLPDWHG EDVHG RQ DQ H[SRQHQWLDO ILW WR D VLQJOH ILUVWRUGHU H[SRQHQWLDO SODWHDX EHORZ b %f /LQHDU LQLWLDO SKDVH RI WLPH FRXUVH 5DWHV DUH HVWLPDWHG EDVHG RQ VORSH RI EHVW ILW OLQH

PAGE 109

LQLWLDO UDWH IUDFWLRQ FOHDYHGKRXUf ‘ ‘ ‘ >U@ Q0 )LJXUH 7KH UDWH RI U FOHDYDJH LQFUHDVHV DW KLJKHU FRQFHQWUDWLRQV EXW UHDFKHV D ORZHU OLPLW DW ORZ GLOXWLRQV /RZHULQJ U FRQFHQWUDWLRQ LQLWLDOO\ UHGXFHV FOHDYDJH UDWH EXW WKH FOHDYDJH UDWH UHDFKHV D ORZHU OLPLW RI DSSUR[LPDWHO\ KRXU DW FRQFHQWUDWLRQV RI Q0 DQG EHORZ

PAGE 110

$ % >@ Q0 >@ T0
PAGE 111

‘ U D U r U A U f U ’ U >UfIQ0 VORSH 5 )LJXUH &OHDYDJH UDWHV IRU ORZ FRQFHQWUDWLRQV RI U EXW QRW KLJKf ILW ZHOO WR D OLQHDU HTXDWLRQ ZKHQ WKH QDWXUDO ORJ RI VXEVWUDWH FRQFHQWUDWLRQ LV SORWWHG YV WLPH LQGLFDWLQJ ILUVW RUGHU H[SRQHQWLDO UDWH

PAGE 112

WLPH KRXUVf WLPH KRXUVf G IXOO OHQJWKf U! FOHDYHG U IXOOOHQJWK USULPHU rfffffff FOHDYHG USULPHU b )LJXUH %RWK U DQG G FOHDYH USULPHU fDQJ ULERVHff LQ WUDQV ZDV LQWHUQDOO\ ODEHOHG DQG USULPHU ZDV fHQG ODEHOHG

PAGE 113

U Q0f QR USULPHU FOHDYDJH RI U Q0f LQ WKH SUHVHQFH RI USULPHU Q0f FOHDYDJH RI USULPHU Q0f LQ WKH SUHVHQFH RI U Q0f )LJXUH 5LERVHSULPHU FRPSHWHV ZLWK U IRU FOHDYDJH E\ U 7KH FOHDYDJH RI U LV ORZHUHG ZKHQ LQFXEDWHG ZLWK USULPHU USULPHU Q0f G Q0f UOLEOR Q0f G Q0f UOLEUDU\ f Q0 G Q0f UXS f Q0 G Q0f )LJXUH G FDQ FOHDYH YDULRXV VXEVWUDWHV LQ WUDQV 8QODEHOHG G Q0f ZDV LQFXEDWHG ZLWK YDULRXV UDGLRODEHOHG VXEVWUDWHV Q0f UHVXOWLQJ LQ VXEVWUDWH FOHDYDJH (DFK VXEVWUDWH FRQWDLQHG WKH VDPH QXFOHRWLGH USULPHU UHJLRQ ZLWK D ULERVHDGHQRVLQH DW SRVLWLRQ

PAGE 114

G GXS GOLEOR GSULPHU )LJXUH 9DULRXV VXEVWUDWHV FDQ FRPSHWH ZLWK U IRU FOHDYDJH E\ G $GGLWLRQ RI D IROG H[FHVV RI XQODEHOHG GOLER RU GSULPHU WR UDGLRODEHOHG U Q0f DW WLPH ]HUR ORZHUHG FOHDYDJH RI U $GGLWLRQ RI WKH VDPH DPRXQW RI XQODEHOOHG G WR ODEHOHG U LQFUHDVHG U FOHDYDJH WR OHYHOV VLPLODU WR WKDW VHHQ ZLWK Q0 U $GGLWLRQ RI GXS LQFUHDVHG U FOHDYDJH EXW QRW DV PXFK DV WKH DGGLWLRQ RI G Q0 U Q0 G U Q0 GOLEE Q0 U Q0 U Q0 U )LJXUH 7KH FLVFOHDYDJH UDWH RI U LV QRW DIIHFWHG E\ H[FHVV FRPSHWLWRUV 5DGLRODEHOHG U ZDV LQFXEDWHG DW D ORZ FRQFHQWUDWLRQ Q0f VR DV WR PLQLPL]H WUDQVFOHDYDJH $GGLWLRQ RI IROG H[FHVV GOLER FRPSHWLWRU GLG QRW DIIHFW WKH FOHDYDJH UDWH RI U

PAGE 115

F >( G@ Q0 )LJXUH 6LQJOHWXUQRYHU NLQHWLFV IRU G FOHDYLQJ USULPHU $f )XOO WLPH FRXUVH %f ,QLWLDO OLQHDUf SKDVH RI FOHDYDJH &f 6XEVWUDWH VDWXUDWLRQ FXUYH 6XEVWUDWH VDWXUDWLRQ H[SHULPHQWV ZHUH FRQGXFWHG E\ LQFXEDWLQJ USULPHU Q0f ZLWK LQFUHDVLQJ DPRXQWV RI G HQ]\PH

PAGE 116

F >( G@ Q0 )LJXUH 6LQJOHWXUQRYHU NLQHWLFV IRU G FOHDYLQJ U $f )XOOWLPH FRXUVH %f ,QLWLDO OLQHDUf SKDVH RI FOHDYDJH &f 6XEVWUDWH VDWXUDWLRQ FXUYH 6XEVWUDWH VDWXUDWLRQ H[SHULPHQWV ZHUH FRQGXFWHG E\ LQFXEDWLQJ U Q0f ZLWK LQFUHDVLQJ DPRXQWV RI G HQ]\PH

PAGE 117

F )LJXUH 6LQJOHWXUQRYHU NLQHWLFV IRU G FOHDYLQJ UOLER $f )XOOWLPH FRXUVH %f ,QLWLDO OLQHDUf SKDVH RI FOHDYDJH &f 6XEVWUDWH VDWXUDWLRQ FXUYH 6XEVWUDWH VDWXUDWLRQ H[SHULPHQWV ZHUH FRQGXFWHG E\ LQFXEDWLQJ UOLER Q0f ZLWK LQFUHDVLQJ DPRXQWV RI G HQ]\PH

PAGE 118

% H[S ILW Q0 USULPHU Q0 G Q0 USULPHU Q0 G Q0 USULPHU Q0 G Q0 USULPHU Q0 G Q0 UOLEOR FOHDYHG E\ Q0 G Q0 UOLEOR FOHDYHG E\ Q0 G Q0 UOLEOR FOHDYHG E\ Q0 G Q0 UOLEWR FOHDYHG E\ Q0 G )LJXUH 6LQJOHWXUQRYHU NLQHWLFV IRU G FOHDYLQJ USULPHU RU UOLER EHORZ VDWXUDWLRQ

PAGE 119

V m R F R N D! D ‘ Lf§,f§ Q0 UOLEWR Q0 G PL[HG EHIRUH VWRZ FRRO Q0 UOLEWR Q0 G PL[HG DIWHU VWRZ FRRO Q0 UOLEOR Q0 G PL[HG EHIRUH VWRZ FRRO Q0 0LEOR Q0 G PL[HG DIWHU VWRZ FRRO fc KRXUV % Q0 UOLEN! Q0 G PL[HG EHIRUH VORZ FRRO Q0 UOLEOR Q0 G PL[HG DIWHU VORZ FRRO Q0 UOLEOR Q0 G PL[HG EHIRUH VORZ FRRO Q0 UOLEOR Q0 G PL[HG DIWHU VORZ FRRO )LJXUH 0XOWLWXUQRYHU NLQHWLFV IRU G FOHDYLQJ UOLEOR LV QRW DIIHFWHG E\ WKH VORZ FRRO IROGLQJ SURWRFRO DQG LQGLFDWHV WZR WXUQRYHUV LQ KRXUV $f 3HUFHQW RI UOLER VXEVWUDWH FOHDYHG E\ G %f 7KH DEVROXWH DPRXQW RI VXEVWUDWH FOHDYHG 7KH HQ]\PH DQG VXEVWUDWH ZHUH PL[HG WRJHWKHU HLWKHU LPPHGLDWHO\ EHIRUH RU LPPHGLDWHO\ DIWHU WKH VORZ FRRO SURWRFRO

PAGE 120

$ 2 D F X R F R R % L ‘i m t m e e D Lf§ f§,f§ f§Uf§ r KRXUV $ ‘ f, KRXUV ‘ Q0 USULPHU $ Q0 G PL[HG EHIRUH VORZ FRRO D Q0 USULPHU Q0 G PL[HG DIWHU VORZ FRRO } Q0 USULPHU Q0 G PL[HG EHIRUH VORZ FRRO A Q0 USULPHU Q0 G PL[HG DIWHU VORZ FRRO O ‘ Q0 USULPHU Q0 G PL[HG EHIRUH VORZ FRRO D Q0 USULPHU Q0 G PL[HG DIWHU VORZ FRRO Q0 USULPHU Q0 G PL[HG EHIRUH VORZ FRRO A Q0 USULPHU Q0 G PL[HG DIWHU VORZ FRRO )LJXUH 0XOWLWXUQRYHU NLQHWLFV IRU G FOHDYLQJ USULPHU LV QRW DIIHFWHG E\ WKH VORZ FRRO IROGLQJ SURWRFRO DQG LQGLFDWHV WZR WXUQRYHUV LQ KRXUV $f )XOO WLPH FRXUVH IRU USULPHU FOHDYDJH E\ G %f ,QLWLDO WLPH FRXUVH IRU USULPHU FOHDYDJH E\ G UHYHDOV OLQHDU FOHDYDJH XS WR KRXUV 7KH HQ]\PH DQG VXEVWUDWH ZHUH PL[HG WRJHWKHU HLWKHU LPPHGLDWHO\ EHIRUH RU LPPHGLDWHO\ DIWHU WKH VORZ FRRO SURWRFRO

PAGE 121

SHUFHQW FOHDYHG SHUFHQW FOHDYHG $ ? Ls$s A V % L L f§Ua KRXUV KRXUV Lf§ f§L ‘ Q0 USULPHU Q0 G D Q0 USULPHU Q0 G Q0 USULPHU Q0 G A Q0 USULPHU Q0 G f Q0 USULPHU Q0 G ’ Q0 USULPHU Q0 G f§L ‘ Q0 USULPHU Q0 G r Q0 USULPHU Q0 G Q0 USULPHU Q0 G D A Q0 USULPHU Q0 G f Q0 USULPHU Q0 G ’ Q0 USULPHU A Q0 G ’ )LJXUH 0XOWLWXUQRYHU NLQHWLFV IRU G FOHDYLQJ USULPHU VXJJHVW WKH ILUVW FOHDYDJH $f )XOO WLPH FRXUVH IRU USULPHU FOHDYDJH E\ G %f ,QLWLDO WLPH FRXUVH IRU USULPHU FOHDYDJH E\ G UHYHDOV OLQHDU FOHDYDJH XS WR KRXUV 7KH HQ]\PH DQG VXEVWUDWH ZHUH PL[HG WRJHWKHU HLWKHU LPPHGLDWHO\ EHIRUH RU LPPHGLDWHO\ DIWHU WKH VORZ FRRO SURWRFRO

PAGE 122

,OO X L $ D Q0 U ZLWKRXW VORZ FRRO IROGLQJ $ Q0 U ZLWK VORZ FRRO IROGLQJ f§L KRXUV )LJXUH ,QWUDPRLHFXODU IROGLQJ RI U LV UHGXFHG E\ WKH RPLVVLRQ RI VORZ FRROLQJ 5LERVH ZDV LQFXEDWHG DW VXIILFLHQWO\ ORZ FRQFHQWUDWLRQ Q0 f DV WR IDYRU PFOHDYDJH ZLWK DQG ZLWKRXW WKH VORZ FRRO IROGLQJ SURWRFRO )LJXUH ,QWHUPROHFXODU IROGLQJ RI U LV QRW DIIHFWHG E\ fVORZ FRROLQJf 5LERVH ZDV LQFXEDWHG DW VXIILFLHQWO\ KLJK FRQFHQWUDWLRQ Q0 f DV WR IDYRU UDPFOHDYDJH ZLWK DQG ZLWKRXW WKH fVORZ FRROf IROGLQJ

PAGE 123

&+ Q0 UOLEOR Q0 G r PL[ WKHQ VORZFRRO &f r PL[ WKHQ VORZFRRO &f ‘ VORZFRRO WKHQ PL[ &f r PL[ QR VORZFRRO &f L KRXU )LJXUH ,QWHUPROHFXODU IROGLQJ EHWZHHQ U DQG UOLEOR LV QRW DIIHFWHG E\ fVORZ FRROLQJf $ IRXUIROG H[FHVV RI VXEVWUDWH UOLER ZDV PL[HG ZLWK G HLWKHU EHIRUH RU DIWHU WKH VORZ FRRO IROGLQJ SURWRFRO ,Q D WKLUG FRQGLWLRQ VDPSOHV ZHUH PL[HG EXW ZHUH QRW GHQDWXUHG DQG IROGHG XVLQJ WKH VORZFRRO SURWRFRO 7KH fPL[ WKHQ VORZ FRROf FRQGLWLRQ ZDV SHUIRUPHG LQ SDUDOOHO ZLWK RQH FRQGLWLRQ DW r& ZKLOH DOO RWKHU VDPSOHV ZHUH LQFXEDWHG DW r& 7KH VORZ FRRO IROGLQJ SURWRFRO KDG QRW HIIHFW RQ FOHDYDJH UDWHV ,QFXEDWLQJ DW WKH VDPSOHV DW r& LQLWLDOO\ LQFUHDVHG VXEVWUDWH FOHDYDJH FRPSDUHG WR WKH VDPH FRQGLWLRQ DW r&

PAGE 124

‘ Q0 U Q0 G ’ Q0 U Q0 G X0 FKDVH DW W KRXUV f Q0 U Y Q0 U Q0 FKDVH DW W KRXU f Q0 U Q0 G R Q0 U Q0 G Q0 FKDVH DW W PLQ f Q0 USULPHU Q0 G R Q0 USULPHU Q0 G X0 FKDVH DW W KRXUV R Q0 USULPHU Q0 G X0 FKDVH DW W KRXUV KRXUV )LJXUH !RLVFOHDYDJH RI U LV UHGXFHG E\ WKH DGGLWLRQ RI FROG FKDVH LQGLFDWLQJ WKDW WKH UDWH RI (r6 GLVDVVRFLDWLRQ LV IDVWHU WKDQ WKH FKHPLFDO VWHS RI FOHDYDJH 9DULRXV VXEVWUDWHV ZHUH LQFXEDWHG ZLWK H[FHVV HQ]\PH f )ROORZLQJ YDULRXV LQFXEDWLRQ WLPHV H[FHVV FROG FKDVH ZDV DGGHG WR WKH UHDFWLRQ 7KH DGGLWLRQ RI FKDVH JUHDWO\ UHGXFHG FOHDYDJH RI WKH VXEVWUDWH DV FRPSDUHG WR DQ LGHQWLFDO VDPSOH ZLWKRXW DGGHG FKDVHf UHJDUGOHVV RI WKH WLPH DW ZKLFK WKH FKDVH ZDV DGGHG

PAGE 125

’ Q0 U Q0 G X0 FKDVH DW W KRXUV Y Q0 U Q0 FKDVH DW W KRXU R Q0 U Q0 G Q0 FKDVH DW W PLQ R Q0 USULPHU Q0 G X0 FKDVH DW W KRXUV R Q0 USULPHU Q0 G X0 FKDVH DW W KRXUV f Q0 U ‘ Q0 U )LJXUH 7KH UHVLGXDO FOHDYDJH IROORZLQJ WKH DGGLWLRQ RI WKH FKDVH LV OHVV WKDQ DVFOHDYDJH DQG LV GHSHQGHQW RQ WKH DPRXQW RI FKDVH DGGHG 7KH DPRXQW RI FOHDYDJH IROORZLQJ WKH DGGLWLRQ RI FKDVH ZDV UHSORWWHG E\ VXEWUDFWLQJ WKH WLPH DW ZKLFK WKH FKDVH ZDV DGGHG DQG WKH DPRXQW RI FOHDYDJH DW WKH WLPH FKDVH ZDV DGGHG IURP DOO WLPHFOHDYDJH GDWD SRLQW IROORZLQJ WKH FKDVH ,QFUHDVLQJ WKH DPRXQW RI FKDVH DGGHG WR WKH USULPHU Q0fG Q0f FRQGLWLRQ IURP S0 WR S0 GHFUHDVHG WKH UHVLGXDO USULPHU FOHDYDJH

PAGE 126

$ % @  W UD D! R F WV UH e } FR N UR & WHPSHUDWXUH )LJXUH 5LERVH IUDLVFOHDYDJH LV LQFUHDVHG E\ ORZHULQJ WKH WHPSHUDWXUH $f )XOO WLPH FRXUVH %f 3ORW RI LQLWLDO UDWHV YV WHPSHUDWXUH 5LERVH Q0f ZDV LQFXEDWHG DW YDULRXV WHPSHUDWXUHV IURP r& WR r& 7HPSHUDWXUHV EHORZ WKH VHOHFWLRQ WHPSHUDWXUH RI r& LQFUHDVHG WKH UDWH RI FOHDYDJH LQGLFDWLQJ WKDW WKH UDWH OLPLWLQJ VWHS LV WKH IROGLQJDVVRFLDWLRQ VWHS UDWKHU WKDQ WKH FKHPLFDO VWHS

PAGE 127

mM m R F X R $ L 7 f f§U % 6 R F R D $ ‘ f Q0 U &f } Q0 U &f L KRXUV r Q0 UOLEN! Q0 G &f ‘ Q0 UOLENf Q0 G &f aL KRXU )LJXUH /RZHULQJ WKH LQFXEDWLRQ WHPSHUDWXUH GRHV QRW UHGXFH WKH UDWH RI FOHDYDJH DFWLQJ LQ HLWKHU FLV RU WUDQV $f /RZ FRQFHQWUDWLRQ RI U Q0 IDYRULQJ FYFOHDYDJHf ZDV LQFXEDWHG DW HLWKHU r& RU r& %f /RZ FRQFHQWUDWLRQ RI UOLER VXEVWUDWH ZDV LQFXEDWHG ZLWK G Q0f DW HLWKHU r& RU r&

PAGE 128

'RXEOH PXWDQWLRQV WKDW GLG QRW UHFRYHUIXQFWLRQ &7 *$ &7 *$ &* *& &7 *$ &* *& 7UXQFDWLRQV RI WKDW GLG QRW FOHDYH LQ WUDQV )LJXUH 6WUXFWXUDO SUHGLFWLRQV IRU IROGLQJ LQ FLV &RPPRQ IROGV VKDUHG E\ VHYHUDO VWUXFWXUHV DUH ER[HG LQ FRORU 7KH UHG DUURZ LQGLFDWHV WKH ULERVH$ VLWH RI FOHDYDJH 7KH UHG ; PDUNV GRXEOH PXWDQWV WKDW GLG QRW UHFRYHU IXQFWLRQ

PAGE 129

f UOLEOR ZW Q0f FOHDYHG E\ G ZW Q0f ‘ UOLEWR ZW FOHDYHG E\ G *& *& 7$f k k R rr F f R X t K KRXU 2 UOLEWR &* &* $7f FOHDYHG E\ G *& *& 7$f r Q0 UZW VHOIFOHDYDJH r Q0 U ZW VHOIFOHDYDJH ‘ Q0 U *& *& 7$f VHOIFOHDYDJH R Q0 U &* &* $7f *& *& 7$f VHOIFOHDYDJH )LJXUH 1XFOHRWLGHV RI EDVHSDLU ZLWK QXFOHRWLGHV RI HLWKHU LWVHOI LQ FLVf RU ZLWK D VXEVWUDWH LQ WUDQVf 0XWDWLRQV ZHUH LQWURGXFHG LQWR QXFOHRWLGHV DQG RI UHVXOWLQJ LQ UHGXFHG FLV DQG WUDQV FOHDYDJH UDWHV ,QWURGXFLQJ WKH FRPSOHPHQWDU\ PXWDWLRQV LQWR QXFOHRWLGHV DQG RI WKH VXEVWUDWH LPSURYHV VXEVWUDWH FOHDYDJH ERWK LQ FLV DQG WUDQV

PAGE 130

KRXUV f U ZWf >QP@ ‘ UPXW &7f D UPXW &7 t *$f DXJ A UPXW *$f R U *$ t &7f r UPXW *$f $ U *$ t &7f )LJXUH &RPSHQVDWRU\ PXWDWLRQV SUHGLFWHG E\ YDULRXV VWUXFWXUHV GR QRW LPSURYH FDWDO\VLV RI PXWDWHG DW QXFOHRWLGH RU 6HYHUDO VLQJOH QXFOHRWLGH PXWDWLRQV UHGXFHG FOHDYDJH UDWHV $WWHPSWV WR LPSURYH FOHDYDJH UDWHV E\ PDNLQJ FRPSHQVDWRU\ PXWDWLRQV SUHGLFWHG E\ VHYHUDO VWUXFWXUDO PRGHOV ZHUH XQVXFFHVVIXO

PAGE 131

$ USUcPHU IY b 0 ` 9m R" WnnH m ‘ D WUD0 fm +0UW m D 0L $$ WUDQV USUcPHU DW [ AU r ‘ }f" WL L YL L6nV 9 U $& USULPHU 4 n( U % USUcPHU R 6$ 2 6; ^ nn6 Y! 9 f D 9 ff 2 UI O R Or I % USULPHU &$7* LQWHUDFWLRQ f2[ $ [2 ‘; DƒVn U /  n6n( X m ‘ WUP D : ,UDQ )LJXUH 6HYHUDO VWUXFWXUDO PRGHOV IRU U FOHDYLQJ USULPHU LQ WUDQV 7KH WZR H[SHULPHQWDOO\ VXSSRUWHG WUDQVELQGLQJ VWUXFWXUHV DUH VKRZQ LQ FRORUHG ER[HV LU LR 7KH UHG ER[ KLJKOLJKWV WKH LQWHUDFWLRQ EHWZHHQ WKH &$7* UHJLRQ RI WKH SULPHU VXEVWUDWH 6f DQG WKH VDPH VHTXHQFH PRWLI RI (f 7KH \HOORZ ER[ KLJKOLJKWV WKH LQWHUDFWLRQ EHWZHHQ O&*$&7&$&7$7 RI WKH SULPHU VXEVWUDWH DQG V*7$*7*$&* RI 7KH UHG DUURZ PDUNV WKH ULERVH$ VLWH RI FOHDYDJH

PAGE 132

L ‘2 i FR f R 6 L Imf§ £ £ KRXUV r UOLEOR ZW FOHDYHG E\ G ZW D UOLEOR &7 FOHDYHG E\ G *$ ‘ UOLEOR ZW FOHDYHG E\ G *$ )LJXUH 7KH *$ PXWDWLRQ RI G LV SDUWLDOO\ fUHVFXHGf E\ WKH FRPSHQVDWRU\ &7 PXWDWLRQ LQ WKH UOLEOR VXEVWUDWH VXEVWDQWLDWLQJ WKH SUHGLFWHG LQWHUPROHFXODU f&$7*f DVVRFLDWLRQ &RQYHUWLQJ RI G WR $ JUHDWO\ UHGXFHG UDPFOHDYDJH RI UOLER EXW WUDPFOHDYDJH ZDV SDUWLDOO\ UHFRYHUHG E\ DGGLQJ WKH FRPSHQVDWRU\ &7 PXWDWLRQ WR UOLER &RQFHQWUDWLRQV XVHG ZHUH G Q0f DQG UOLER Q0f

PAGE 133

'06 *f ODQH / L , , OLOW 3RWDVVLXP SHUPDQJDQDWH7f ODQH / f A )LJXUH 6WUXFWXUDO SURELQJ XVLQJ '06 DQG SRWDVVLXP SHUPDQJDQDWH WR FKHPLFDOO\ PRGLI\ VLQJOHVWUDQGHG '1$ VKRZV VRPH SURWHFWHG SRVLWLRQV (QG ODEHOHG G Q0f ZDV LQFXEDWHG ZLWK ODQH DQG f RU ZLWKRXW ODQH DQG f XQODEHOHG USULPHU VXEVWUDWH Q0f RYHUQLJKW LQ UHDFWLRQ EXIIHU WR DOORZ IRU IROGLQJ WR UHDFK HTXLOLEULXP 6DPSOHV ZHUH WKHQ WUHDWHG ZLWK HLWKHU '06 RU SRWDVVLXP SHUPDQJDQDWH IRU PLQXWHV RU PLQXWHV WR PRGLI\ HLWKHU VLQJOHVWUDQGHG RU 7 QXFOHRWLGHV UHVSHFWLYHO\ 6DPSOHV ZHUH WKHQ HWKDQRO SUHFLSLWDWHG DQG WUHDWHG ZLWK 0 SLSHULGLQH IRU KRXU DW r& 6DPSOHV ZHUH DJDLQ HWKDQRO SUHFLSLWDWHG UHVXVSHQVGHG LQ IRUPDPLGH ORDGLQJ EXIIHU DQG UXQ RQ D b 3$*(XUHD JHO QH[W WR D ES ODGGHU ODQH /f 1XPEHUV LQGLFDWH WKH SRVLWLRQV RI *fV RU 7fV LQ WKH FRPSOHWH VHTXHQFH EHWZHHQ QXFOHRWLGH DQG f 3RVLWLRQV WKDW GHPRQVWUDWHG VRPH GHJUHH RI SURWHFWLRQ IURP FKHPLFDO PRGLILFDWLRQ DUH LQGLFDWHG ZLWK WKH UHG ER[

PAGE 134

&+$37(5 678',(6 2) ,1 9,752 6(/(&7,216 ,Q 9LWUR 6HOHFWLRQ RI )XQFWLRQDOL]HG DQG 1RQ)XQFWLRQDOL]HG /LEUDULHV 8VLQJ &DJHG5LERVH DQG /LTXLG3KDVH 6HOHFWLRQ ,Q 9LWUR 6HOHFWLRQ )LIWHHQ LQ YLWUR VHOHFWLRQV ZHUH SHUIRUPHG LQ SDUDOOHO HDFK ZLWK D YDULDWLRQ LQ HLWKHU WKH SULPHU VHTXHQFH fDQJf YV f%-ff ULERVH SURWHFWLRQ fFDJHGf YV XQSURWHFWHG IUHH ULERVHf PHWDO FRIDFWRU LQ WKH EXIIHU QRQH ]LQF RU PDJQHVLXPf WKH QXPEHU RI PROHFXOHV VDPSOHG LQ WKH VWDUWLQJ OLEUDU\ [ YV [ f RU WKH SUHVHQFH RI IXQFWLRQDO JURXS RQ D QXFOHLF DFLG (EDVH YV VWDQGDUG 7EDVHf 7KH VHOHFWLRQ ZDV SHUIRUPHG DV GHVFULEHG LQ 0DWHULDOV DQG 0HWKRGV 2QO\ RQH FRQGLWLRQ VWDQGDUG QXFOHRWLGHV XVLQJ fDQJ ULERVHf SULPHU ZLWK XQSURWHFWHGULERVHf JHQHUDWHG DFWLYH '1$]\PHV FDSDEOH RI SKRVSKRGLHVWHU FOHDYDJH )RU WKH LQLWLDO URXQG RI VHOHFWLRQ VL[ VWDUWLQJ OLEUDULHV ZHUH JHQHUDWHG YDU\LQJ LQ FDWDO\WLF SULPHU VHTXHQFH ULERVH SURWHFWLRQ DQG QXFOHRWLGHVf YLD D VLQJOH F\FOH RI SULPHU H[WHQVLRQ LQWHUQDOO\ ODEHOHG XVLQJ DOSKD3&73f 7KHVH OLEUDULHV ZHUH GLYLGHG LQWR WKH ILIWHHQ VHOHFWLRQ FRQGLWLRQV YDU\LQJ LQ EXIIHU FRPSRVLWLRQ DQG VHOHFWLRQ WLPH VHH 7DEOH f $OO VDPSOHV ZHUH WKHQ H[SRVHG WR SXOVHV RI ODVHU P-SXOVH WKH QXPEHU RI ODVHU SXOVHV ZDV YDULHG ZLWK ODVHU SRZHU VR WKDW WKH WRWDO HQHUJ\ ZDV KHOG FRQVWDQW DW P-f HIIHFWLYHO\ GHSURWHFWLQJ VDPSOHV FRQWDLQLQJ FDJHGULERVH 3RROV ZHUH WKHQ DOORZHG WR UHDFW IRU RU PLQXWHV VHH 7DEOH f DW ZKLFK SRLQW WKH\

PAGE 135

ZHUH VWRSSHG E\ WKH DGGLWLRQ RI IRUPDPLGH VWRS G\H DQG IUHH]LQJ r&f $OLTXRWV ZHUH DOVR UHPRYHG IURP HDFK SRRO DQG VWRSSHG DW YDULRXV WLPHV IURP WR KRXUVf WR DVVHVV WKH SURJUHVV FXUYH IRU FOHDYDJH DW HDFK URXQG RI VHOHFWLRQ 7KH PDMRULW\ RI HDFK SRRO ZDV UXQ RQ D WKLFN VHTXHQFLQJ JHO DQG WKH FOHDYHG IUDFWLRQ RI WKH SRRO ZDV H[FLVHG DQG SXULILHG 7KH SXULILHG fVXUYLYRUVf RI HDFK URXQG ZHUH DPSOLILHG ZLWK 3&5 DQG VXEMHFWHG WR DQ DGGLWLRQDO URXQG RI VHOHFWLRQ 7KLV SURFHGXUH ZDV UHSHDWHG IRU D WRWDO RI HLJKW URXQGV RI VHOHFWLRQ )LJXUH VKRZV WKH SHUFHQW RI IXOOOHQJWK VXEVWUDWH IRU HDFK VHOHFWLRQ FRQGLWLRQ WKDW ZDV FRQYHUWHG WR WKH FOHDYHG SURGXFW DW WKH WLPH HDFK VHOHFWLRQ ZDV VWRSSHG DW HDFK URXQG RI VHOHFWLRQ 7KH ILUVW VHOHFWLRQ LV WHUPHG fURXQG ]HURf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f%FDJHf SULPHU ZLWK VWDQGDUG QXFOHRWLGHV EXW WKH SULPHU LQ VDPSOH ZDV GH SUR WHFWHG E\ ODVHU SULRU WR WKH LQLWLDWLRQ RI WKH 96 H[SHULPHQWV DQG ZDV WKHUHIRUH HIIHFWLYHO\ f%ULERVHff ZKHUHDV VDPSOH ( ZDV SURWHFWHG XQWLO H[SRVXUH WR ODVHU %RWK VDPSOHV ZHUH H[SRVHG WR ODVHU DW WKH LQLWLDWLRQ RI HDFK URXQG RI VHOHFWLRQ %RWK VDPSOHV ZHUH DOORZHG WR UHDFW IRU WZR KRXUV SULRU WR WKH DGGLWLRQ RI VWRS EXIIHU DQG ERWK VKRZHG ORZ OHYHOV RI FRQYHUVLRQ WR WKH FOHDYHG SURGXFW 7KLV ORZ OHYHO RI FOHDYDJH OHVV WKDQ bf VXJJHVWV D ORZ OHYHO IRU ODVHU GDPDJH

PAGE 136

$OWKRXJK FOHDYHG SURGXFWV ZHUH XQGHWHFWDEOH VDPSOHV ZHUH UXQ RQ D 3$*(XUHD JHO DQG WKH VL]H FRUUHVSRQGLQJ WR FOHDYHG SURGXFW H[FLVHG 7KHVH JHOSXULILHG IUDJPHQWV ZHUH XVHG DV WHPSODWHV WR UHJHQHUDWH IXOO OHQJWK '1$]\PHV IRU WKH QH[W URXQG RI VHOHFWLRQ LQGLFDWLQJ WKDW VPDOO DPRXQWV RI FOHDYHG SURGXFW KDG LQ IDFW EHHQ SURGXFHG DQG LVRODWHGf 0RUH UDGLRODEHO ZDV XVHG IRU HDFK VDPSOH LQ DOO VXEVHTXHQW URXQGV LQFUHDVLQJ WKH VLJQDO WR QRLVH UDWLR 6DPSOHV $ ) DOO VKRZ D QRWLFHDEOH LQFUHDVH LQ SHUFHQW RI IXOO OHQJWK VDPSOH FRQYHUWHG WR FOHDYHG SURGXFW bf IROORZLQJ MXVW RQH URXQG RI VHOHFWLRQ $OO WKHVH VDPSOHV FRQWDLQHG WKH f%FDJHf SULPHU ZLWK VWDQGDUG QXFOHRWLGHV DQG YDULHG RQO\ LQ HLWKHU WKH EXIIHU FRPSRVLWLRQ RU WKH VHOHFWLRQ WLPH 2WKHU VHOHFWLRQ FRQGLWLRQV LQFOXGLQJ VWDQGDUG QXFOHRWLGHV ZLWK f%GHSURWHFWHG FDJHf RU fDQJ ULERVHf SULPHUV DQG DOO VDPSOHV FRQWDLQLQJ (EDVH KDG XQGHWHFWDEOH OHYHOV RI FOHDYHG SURGXFW $IWHU WKH FRPSOHWLRQ RI WZR URXQGV RI VHOHFWLRQ KRZHYHU VHOHFWLRQ FRQGLWLRQV FRQWDLQLQJ (EDVH ZLWK f%FDJHf SULPHU VDPSOHV 0f EHJDQ WR VKRZ b FOHDYHG SURGXFW SDUDOOHO VHOHFWLRQ FRQGLWLRQV FRQWDLQLQJ VWDQGDUG QXFOHRWLGHV KDG VOLJKWO\ KLJKHU OHYHOV RI FOHDYHG SURGXFW b WKLV VOLJKW LQFUHDVH LV SRVVLEO\ WKH UHVXOW RI OHVV fVPHDULQJf RI WKH VDPSOHV FRQWDLQLQJ VWDQGDUG QXFOHRWLGHVf 6HOHFWLRQ FRQGLWLRQV ZLWK GHSURWHFWHG FDJHGULERVH DQG XQFDJHGULERVH ZLWK HLWKHU VWDQGDUG RU QRQVWDQGDUG EDVHV VWLOO VKRZHG QR VLJQLILFDQW FOHDYDJH + 1 2f 7KH SHUFHQW RI IXOO OHQJWK VXEVWUDWH FRQYHUWHG WR FOHDYHG SURGXFW IRU HDFK VDPSOH $ ) FRQWDLQLQJ f%FDJHf ZLWK VWDQGDUG EDVHV UHJDUGOHVV RI WKH EXIIHU FRPSRVLWLRQ DQG VHOHFWLRQ WLPHf FRQWLQXHG WR LQFUHDVH VOLJKWO\ ZLWK HDFK SURJUHVVLYH URXQG RI VHOHFWLRQ %\ WKH FRPSOHWLRQ RI URXQGV RI VHOHFWLRQ DV PXFK DV b RI WKHVH VDPSOHV ZHUH FRQYHUWHG WR IXOOOHQJWK SURGXFW 'XULQJ WKH VDPH WLPH VHOHFWLRQ FRQGLWLRQV ZLWK

PAGE 137

WKH f%GHSURWHFWHG FDJHf SULPHU ZLWK VWDQGDUG DQG QRQVWDQGDUG EDVHVf FRQWLQXHG WR KDYH LQVLJQLILFDQW FRQYHUVLRQ WR FOHDYHG SURGXFW 6HOHFWLRQ FRQGLWLRQ f+f FRQWDLQLQJ VWDQGDUG QXFOHRWLGHV ZLWK WKH fDQJ ULERVHf SULPHU EHJDQ WR VKRZ b FOHDYDJH DIWHU IRXU URXQGV RI VHOHFWLRQ 7KH DPRXQW RI FOHDYHG SURGXFW IRU VHOHFWLRQ FRQGLWLRQ f+f LQFUHDVHG GUDPDWLFDOO\ IURP b DIWHU URXQGV RI VHOHFWLRQ WR b DIWHU URXQGV DQG WR b DIWHU VHYHQ URXQGV RI VHOHFWLRQ 7KH SDUDOOHO VHOHFWLRQ WR SRSXODWLRQ f+f LQ ZKLFK (EDVH ZDV VXEVWLWXWHG IRU 7EDVH KDG QR GHWHFWDEOH FOHDYDJH HYHQ DIWHU VHYHQ URXQGV RI VHOHFWLRQ 7KH UHVXOWV UHSRUWHG DERYH WHOO XV RQO\ ZKDW IUDFWLRQ RI WKH IXOO OHQJWK PDWHULDO ZDV FOHDYHG DW WKH WLPH HDFK URXQG RI VHOHFWLRQ ZDV VWRSSHG $ PRUH GHWDLOHG XQGHUVWDQGLQJ RI WKH SURJUHVV RI HDFK VHOHFWLRQ FDQ EH JDLQHG E\ H[DPLQLQJ WKH SURJUHVV FXUYH IRU FOHDYDJH 7KLV ZDV GRQH E\ UHPRYLQJ DQ DOLTXRW RI HDFK VHOHFWLRQ FRQGLWLRQ DW YDULRXV WLPHV IURP KRXUV WR KRXUVf DQG VWRSSLQJ WKH UHDFWLRQ E\ IRUPDPLGH VWRS G\H $ ORZ VLJQDOWRQRLVH UDWLR PDGH EDVHOLQH GHWHUPLQDWLRQ GLIILFXOW DQG WKHUHIRUH HVWLPDWLRQ RI WKH SHUFHQW FOHDYHG GLIILFXOW 1RQHWKHOHVV WKH SUHVHQFH RI '1$]\PH HQULFKPHQW LQ RXU OLEUDU\ VKRXOG EH PDQLIHVW E\ LQ DQ LQFUHDVH LQ FOHDYHG SURGXFW ZLWK WLPH ZKHUHDV QRLVH VKRXOG UHPDLQ UHODWLYHO\ FRQVWDQW RYHU WLPH 2I WKH ILIWHHQ VHOHFWLRQ FRQGLWLRQV RQO\ VHOHFWLRQ FRQGLWLRQ f+f VKRZHG D FRQVLVWHQW LQFUHDVH LQ FOHDYHG SURGXFW ZLWK WLPH 2WKHU VHOHFWLRQ FRQGLWLRQV KDG YDU\LQJ OHYHOV RI FOHDYDJH EXW FOHDYDJH GLG QRW FRQVLVWHQWO\ LQFUHDVH ZLWK WLPH ,W LV QRWDEOH WKDW FOHDYDJH ZDV QRW GHWHFWHG LQ VDPSOHV 1 DQG 2 DOO RI ZKLFK KDG DQ XQSURWHFWHG ULERVH DW WKH WLPH RI ODVHU H[SRVXUH HLWKHU f%GHSURWHFWHG FDJHf RU fDQJ ULERVHf SULPHUf

PAGE 138

7HVWLQJ .LQHWLFV IURP 5RXQG 3RROV :LWKRXW &DJHGULERVH 7KHVH GDWD VXJJHVW WKDW WKH ODVHU PD\ FDXVH EDFNJURXQG FOHDYDJH WKDW PDVNV WKH FOHDYDJH UHVXOWLQJ IURP '1$]\PH FDWDO\VLV 7KLV ZDV WHVWHG E\ DPSOLI\LQJ WKH VXUYLYRUV RI WKH HLJKWK URXQG RI VHOHFWLRQ IRU HDFK VHOHFWLRQ FRQGLWLRQ ZLWKRXW D FDJHGULERVH XVLQJ HLWKHU f%ULERVHf RU fDQJ ULERVHff DQG WHVWLQJ NLQHWLFV ZLWKRXW WKH ODVHU LQLWLDWLRQ )LJXUH VKRZV WKH UHVXOWV 7KH SRRO RI PROHFXOHV JHQHUDWHG IURP SRSXODWLRQ f+f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f'f DQG f+f ZHUH UHDPSOLILHG ZLWK FDJHGULERVH XVLQJ f%FDJHf DQG fDQJ FDJHf SULPHUV UHVSHFWLYHO\f (DFK VDPSOH ZDV GLOXWHG WR Q0 DQG H[SRVHG WR ODVHU (DFK RI WKH VL[ VDPSOHV KDG OHVV WKDQ b FOHDYDJH SULRU WR ODVHU ,PPHGLDWHO\ IROORZLQJ ODVHU H[SRVXUH SRROV IURP f+f URXQGV DQG VKRZHG DQG b FOHDYDJH SRROV IURP f'f URXQGV VKRZHG DQG b FOHDYDJH 7KH DPRXQW RI HDFK SRRO FOHDYHG LPPHGLDWHO\ IROORZLQJ ODVHU GLG QRW LQFUHDVH IURP URXQG WR WR DUJXLQJ DJDLQVW WKH SRVVLELOLW\ WKDW WKHUH ZDV

PAGE 139

VHOHFWLRQ IRU PROHFXOHV WKDW DUH PRUH VXVFHSWLEOH WR ODVHU LQGXFHG FOHDYDJH :KHQ WKH NLQHWLFV RI WKHVH VDPSOHV ZDV H[DPLQHG ZLWK LQFUHDVLQJ WLPH FOHDYDJH RI f+ URXQG f GLG QRW LQFUHDVH VLJQLILFDQWO\ ZKHUHDV f+ URXQG f DQG f+ URXQG f VKRZHG LQFUHDVLQJ FOHDYDJH VLPLODU WR VDPSOHV SUHSDUHG ZLWK fDQJ ULERVHf DQG ZLWKRXW ODVHU LQLWLDWLRQ )LJXUH f 3RROV IURP f'f URXQGV DQG GLG QRW VKRZ VLJQLILFDQW LQFUHDVH LQ FOHDYDJH ZLWK WLPH DIWHU WKH LPPHGLDWH LQFUHDVH DIWHU ODVHU H[SRVXUH &DJHGULERVH 'RHV 1RW ,QGXFH &OHDYDJH LQ WUDQV 7KH UHVXOWV GHVFULEHG LQ WKH SUHYLRXV VHFWLRQV VXJJHVW WKDW DOWKRXJK WKH ODVHU DORQH GRHV QRW FDXVH VLJQLILFDQW FOHDYDJH RI ULERVH LW DSSHDUV DV WKRXJK WKH SUHVHQFH RI D FDJHGULERVH LQ DGGLWLRQ WR ODVHU FDXVHV D ORZ OHYHO RI VHTXHQFHLQGHSHQGHQW FOHDYDJH DW WKH FDJHGULERVH QR PHFKDQLVP IRU WKLV UHDFWLRQ LV NQRZQf ,W UHPDLQV XQFOHDU KRZHYHU LI WKH FRPELQDWLRQ RI ODVHU DQG FDJHGULERVH LV FDXVLQJ VWUDQG EUHDNDJH RI WKH VWUDQG WR ZKLFK WKH FDJHGULERVH LV DWWDFKHG LQ FLVf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f+ DQJ ULERVHf 3RROV IRU f+f ZHUH UHDPSOLILHG IURP WKH VWRFN RI fVXUYLYRUVf IROORZLQJ HDFK URXQG RI VHOHFWLRQ DQG XVHG IRU D PRUH GHWDLOHG NLQHWLF DQDO\VLV 6DPSOHV ZHUH

PAGE 140

V\QWKHVL]HG ZLWK DQ XQSURWHFWHG ULERVH DQG WKHUHIRUH ODVHU ZDV QRW XVHG WR LQLWLDWH FOHDYDJH 3RROV IURP HDFK RI WKH HLJKW URXQGV RI VHOHFWLRQ ZHUH SUHSDUHG DW WKUHH FRQFHQWUDWLRQV Q0 Q0 DSSUR[LPDWHO\ WKH FRQFHQWUDWLRQ XVHG GXULQJ WKH ,96f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f+ URXQG f ZKHUH WHVWHG RYHU D WHPSHUDWXUH UDQJH WR GHWHUPLQH ZKHWKHU WKH UDWH OLPLWLQJ VWHS ZDV WHPSHUDWXUH GHSHQGHQW )LJXUH VKRZV WKDW FOHDYDJH UDWHV DUH XQFKDQJHG E\ D GURS RI WHPSHUDWXUH IURP r& WR r& )XUWKHU ORZHULQJ WKH WHPSHUDWXUHV ORZHUV WKH UDWH RI SURGXFW IRUPDWLRQ DOWKRXJK OHVV WKDQ H[SHFWHG IRU D UHDFWLRQ ZKLFK LV UDWHOLPLWHG E\ WKH FKHPLFDO VWHS /LNH WKH UDWH RI FOHDYDJH IRU WKH SRRO RI PROHFXOHV IURP VHOHFWLRQ + IROORZLQJ URXQGV RI VHOHFWLRQ DSSHDUV WR UHIOHFW D EDODQFH EHWZHHQ WKH DVVRFLDWLRQ VWHS DQG WKDQ WKH FKHPLFDO VWHS

PAGE 141

7UDQVFOHDYDJH ZDV WHVWHG GLUHFWO\ E\ LQFXEDWLQJ fODEHOHG fDQJ ULERVHf SULPHU ZLWK WKH SRRO GHULYHG IURP f+ URXQG f )LJXUH f 7KH SUHVHQFH RI HTXLPRODU fDQJ ULERVHf SULPHU VXEVWUDWH VOLJKWO\ UHGXFHG FOHDYDJH RI PROHFXOHV LQ WKH f+5RXQGf SRRO ZKLOH UHVXOWLQJ LQ QHDUO\ b SULPHU FOHDYDJH 1LQHIROG H[FHVV RI fDQJ ULERVHf RYHU f+f SRRO UHGXFHG WKH FOHDYDJH RI PROHFXOHV LQ WKH f+f SRRO E\ KDOI ZKLOH UHVXOWLQJ LQ ILYH WLPHV PRUH SULPHU FOHDYDJH WKH SHUFHQW RI SULPHU FOHDYHG ZDV UHGXFHG LQ KDOI EXW VLQFH WKH WRWDO DPRXQW RI SULPHU SUHVHQW ZDV IROG KLJKHU WKH QHW UHVXOW LV DQ LQFUHDVHG DPRXQW RI SURGXFW IRUPDWLRQf 6HOHFWLRQ FRQGLWLRQ f+f FRQWDLQHG 0 1D&O DQG P0 0J&A LQ WKH UHDFWLRQ EXIIHU ,W ZDV DVVXPHG E\ %UHDNHU DQG -R\FH DQG PRVW VXEVHTXHQW UHVHDUFKHUV WKDW WKH PROHFXOHV JHQHUDWHG XVLQJ ,96 ZLWK 0J&K LQ WKH EXIIHU ZRXOG XWLOL]H WKH PDJQHVLXP DV D FRIDFWRU LW ZDV QHYHU RYHUWO\ VWDWHG EXW LW LV SRVVLEOH PROHFXOHV WKDW DUH 0J LQGHSHQGHQW ZRXOG EH ORVW GXULQJ WKH ZRUNXS SKDVH RI VHOHFWLRQ DQG WKHUHIRUH VHOHFWHG DJDLQVWf )ROORZLQJ WKLV DVVXPSWLRQ PROHFXOHV ZRXOG UHPDLQ LQDFWLYH XQWLO WKH PDJQHVLXP FRIDFWRU ZDV DGGHG 2XU H[SHULHQFH ZLWK GHPRQVWUDWHG WKDW PROHFXOHV VHOHFWHG ZLWK PDJQHVLXP DV D FRIDFWRU FDQ LQ IDFW EH PDJQHVLXPLQGHSHQGHQW 3RROV JHQHUDWHG IURP f+ URXQG f VKRZHG VLPLODU NLQHWLFV UHJDUGOHVV RI ZKHWKHU =Q ZDV VXEVWLWXWHG IRU PDJQHVLXP RU LQ IDFW LI QR GLYDOHQW FDWLRQ ZDV SUHVHQW )LJXUH $f 7R UHPRYH WKH SRVVLELOLW\ WKDW WUDFH PHWDO FRQWDPLQDWLRQ FRQWULEXWHG WR FDWDO\VLV LQ WKH DEVHQFH RI PDJQHVLXP RU ]LQF LQ WKH UHDFWLRQ EXIIHU QHZ EXIIHUV ZHUH PDGH ZLWK GRXEOHGLVWLOOHG ZDWHU DQG XOWUDSXUH 1D&O &OHDYDJH ZDV XQFKDQJHG E\ WKH UHPRYDO RI 0J&O WKH DGGLWLRQ RI ('7$ RU WKH DGGLWLRQ RI '(3& XVHG LQ SUHYLRXV EXIIHUVf WR WKH

PAGE 142

UHDFWLRQ EXIIHU 1R FDWDO\VLV ZDV REVHUYHG KRZHYHU ZKHQ WKH 1D&O ZDV RPLWWHG IURP WKH UHDFWLRQ EXIIHU HYHQ ZLWK WKH DGGLWLRQ RI ,P0 0J&A )LJXUH %f 6HTXHQFH (YROXWLRQ 'XULQJ 96 7KH NLQHWLF SURILOHV RI SRRO f+f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f IHOO LQWR IRXU )DPLOLHV WHUPHG DOSKD EHWD JDPPD DQG GHOWD )LJXUH f )DPLOLHV EHWD JDPPD DQG GHOWD VKDUH WZR FRPPRQ VHTXHQFH PRWLIV ERWK LQ WKH f HQG RI WKH 1 UHJLRQ 7KH KLJK GHJUHH RI VHTXHQFH YDULDWLRQ VHHQ LQ WKH fURXQG RQH SRROf LV H[SHFWHG DQG ZLWK WKH REVHUYDWLRQ WKDW WKLV SRRO VKRZV YHU\ OLWWOH FDWDO\WLF DFWLYLW\ LV FRQVLVWHQW ZLWK WKH LGHD WKDW fVXUYLYRUVf LQ WKLV SRRO DUH GRPLQDWHG E\ QRQFDWDO\WLF PROHFXOHV WKDW fOHDNHGf WKURXJK WKH VHOHFWLRQ V\VWHP HLWKHU E\ QRQVSHFLILF FOHDYDJH WUDQVFOHDYDJH RU DEQRUPDOO\ IDVW PLJUDWLRQ GXULQJ JHO HOHFWURSKRUHVLVf $V FDWDO\WLF DFWLYLW\ RI WKH SRRO EHJLQV WR LQFUHDVH IROORZLQJ WKH WKLUG URXQG RI VHOHFWLRQ H[DPSOHV RI WKH VHTXHQFH )DPLO\ DOSKD DQG EHWD EHJLQ WR GRPLQDWH ([DPSOHV

PAGE 143

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f $OWKRXJK VHTXHQFHV FOHDUO\ FOXVWHU LQWR )DPLOLHV IHZ RI WKH VHTXHQFHV DUH H[DFWO\ LGHQWLFDO WR DQ\ RWKHU VHTXHQFH 7KLV VLPLODULW\ RI VHTXHQFHV ZLWKLQ D )DPLO\ FDQ HLWKHU EH WKH UHVXOW RI FRQYHUJHQFH RQWR VLPLODU VHTXHQFHV DVVXPLQJ HDFK VHTXHQFH ZDV RULJLQDOO\ SUHVHQW LQ WKH LQLWLDO UDQGRP OLEUDU\ RU WKURXJK GLYHUJHQFH IURP D VLQJOH VHTXHQFH YLD PXWDWLRQV DFFUXHG GXULQJ 3&5 DPSOLILFDWLRQ 7KH VHTXHQFHV RI )DPLO\ EHWD JDPPD DQG GHOWD DUH DOVR YHU\ VLPLODU WR VHTXHQFHV DQG UHVSHFWLYHO\f LVRODWHG IURP D SUHYLRXV VHOHFWLRQ H[SHULPHQW 7KH LQIUHTXHQF\ RI WKHVH ODUJH fPRWLIVf LQ f PDNHV LW H[WUHPHO\ XQOLNHO\ WKDW VXFK PRWLIV ZRXOG EH VDPSOHG IURP D UDQGRP OLEUDU\ FRQWDLQLQJ SRWHQWLDO YDULDQWVf PXOWLSOH WLPHV HLWKHU ZLWKLQ D VLQJOH H[SHULPHQW RU LQ WZR VHSDUDWH H[SHULPHQWVf ZKHQ HDFK VDPSOH FRQWDLQV RQO\ PROHFXOHV 6HYHUDO SRVVLELOLWLHV FRXOG H[SODLQ WKLV UHVXOW Df 7KH OLEUDU\ LV QRW WUXO\ UDQGRP ,I UDWKHU WKDQ FRQWDLQLQJ WKH H[SHFWHG XQLTXH PROHFXOHV D ELDV LQ WKH V\QWKHVLV RU SXULILFDWLRQ RI WKH OLEUDU\ JUHDWO\ UHGXFHV WKH DFWXDO OLEUDU\ VL]H WR IRU H[DPSOHf RQO\ GLIIHUHQW PROHFXOHV WKDQ WKH RGGV RI UHFRYHULQJ WKH VDPH VHTXHQFH LQ HDFK RI WZR LQGHSHQGHQW VDPSOLQJV RI PROHFXOHV LV UHODWLYHO\ KLJK Ef 3&5 JHQHUDWHG ELDV &UHDWLQJ WKH IXOOOHQJWK SURGXFW UHTXLUHV WKDW RQO\ QXFOHRWLGHV EH V\QWKHVL]HG EHWZHHQ WKH WZR SULPHUV *LYHQ WKDW RQH SULPHU LV QXFOHRWLGHV ORQJ DQG WKH RWKHU QXFOHRWLGHV ORQJ LW LV FRQFHLYDEOH DOWKRXJK QRW OLNHO\f WKDW WKH SULPHUV PD\ PLVSULPH RII RI HLWKHU WKH UDQGRPL]HG WHPSODWH

PAGE 144

RU RII RI RQH DQRWKHU WR JHQHUDWH SURGXFWV RI DSSUR[LPDWHO\ QXFOHRWLGHV 6XFFHVVLYH URXQGV RI 3&5 DQG VHOHFWLRQ EDVHG RQ VL]H DQG FDWDO\WLF DELOLW\ PD\ LVRODWH PROHFXOHV WKDW ZHUH QHYHU SUHVHQW LQ WKH LQLWLDO VDPSOH Ff &RQWDPLQDWLRQ IURP DQ H[WHUQDO VRXUFH '1$ IURP HQYLURQPHQWDO VRXUFHV VXFK DV SROOHQ PLFUREHV RU WKH KXPDQV FRQGXFWLQJ WKH H[SHULPHQWVf PD\ KDYH FRQWDPLQDWHG ERWK 96 H[SHULPHQWV Gf &RQWDPLQDWLRQ IURP DQ LQWHUQDO VRXUFH &DWDO\WLF VHTXHQFHV LVRODWHG IURP WKH SUHYLRXV ,96 PD\ KDYH FRQWDPLQDWHG HTXLSPHQW RU EXIIHUV WKHUHIRUH ELDVLQJ WKH OLEUDU\ RI WKH VHFRQG ,96 %HFDXVH LW LV GLIILFXOW WR LPDJLQH D PHFKDQLVP WKDW ZRXOG UHVXOW LQ HLWKHU VFHQDULR Df RU Ef FRQWDPLQDWLRQ VHHPV WR EH WKH PRVW OLNHO\ H[SODQDWLRQ IRU ILQGLQJ H[WUHPHO\ UDUH FDWDO\WLF VHTXHQFHV $ IHZ IDFWRUV KRZHYHU DUJXH DJDLQVW FRQWDPLQDWLRQ )LUVW WKH VHFRQG ,96 ZDV SHUIRUPHG E\ GLIIHUHQW UHVHDUFKHUV DW GLIIHUHQW WLPHV LQ GLIIHUHQW EXLOGLQJV ZLWK GLIIHUHQW HTXLSPHQW 6HFRQG RXU VHTXHQFLQJ VWXGLHV RI UHYHDOHG WKDW PXWDWLRQV RFFXUUHG YHU\ LQIUHTXHQWO\ OHVV WKDQ RXW RI VHTXHQFHV FRQWDLQHG D PXWDWLRQf DIWHU F\FOHV RI 3&5 ,I WKH )DPLO\ EHWD VHTXHQFHV ZHUH GHULYHG IURP FRQWDPLQDWLRQ E\ RQH ZRXOG H[SHFW WR ILQG PDQ\ H[DFW FRSLHV RI SDUWLFXODUO\ LQ HDUO\ URXQGV RI WKH VHOHFWLRQ
PAGE 145

LQ SRRO LV IDVWHU WKDQ WKDW RI WKH VHTXHQFH 7KLV ZDV H[SORUHG E\ WHVWLQJ WKH UDWHV RI FVFOHDYDJH RI YDULRXV FORQHV LVRODWHG IURP WKH URXQG SRRO )LJXUH f &OHDYDJH DVVD\V ZHUH GRQH DW Q0 WKH VDPH FRQFHQWUDWLRQ XVHG GXULQJ WKH VHOHFWLRQV &ORQH +U GLIIHUHG IURP E\ RQO\ RQH QXFOHRWLGH \HW WKH UDWH FRQVWDQW ZDV IROG ORZHU KUnf ZKHUHDV RWKHU FORQHV +U +U +Uf GLIIHULQJ IURP DQG HDFK RWKHU E\ RQO\ RQH RU WZR QXFOHRWLGHV KDG UDWH FRQVWDQWV WHQn IROG KLJKHU WKDQ KUnf 7ZR )DPLO\ JDPPD VHTXHQFHV +U DQG +Uf GLIIHUHG IURP HDFK RWKHU E\ RQO\ WKUHH QXFOHRWLGHV \HW WKHUH UDWH FRQVWDQWV GLIIHUHG E\ DQ RUGHU RI PDJQLWXGH YV KUnf 7KH FORQHV IURP URXQG WKDW GLG QRW IDOO LQWR RQH RI WKH IRXU PDMRU VHTXHQFH )DPLOLHV DOO KDG YHU\ ORZ UDWH FRQVWDQWV +U ZDV WKH IDVWHVW ZLWK D UDWH RI KUnf 6LPXODWLRQV RI ,Q 9LWUR 6HOHFWLRQ ([SHULPHQWV 7R XQGHUVWDQG WKH UROH WKDW FDWDO\WLF QXFOHLF DFLGV PD\ KDYH SOD\HG LQ WKH RULJLQV RI OLIH ZH PXVW EH DEOH WR TXDQWLI\ WKH SRZHU DQG DEXQGDQFH RI FDWDO\WLF DFWLYLW\ LQ D UDQGRP SRRO RI QXFOHLF DFLG SRO\PHUV 3UHYLRXV 96 H[SHULPHQWV KDYH VXSSRUWHG WKH H[SHFWDWLRQ WKDW FDWDO\WLF QXFOHLF DFLGV fFDWDO\WLFf LQ UHJDUGV WR DQ\ VSHFLILF UHDFWLRQf DUH fUDUHf :H DUH OLPLWHG LQ RXU DELOLW\ WR GLUHFWO\ HVWLPDWH WKH IUHTXHQF\ RI fUDUHf FDWDO\VWV E\ OLPLWDWLRQV RI H[SHULPHQWDO PHWKRGV XVHG WR GHWHFW WKHP 'HWHFWLRQ RI FDWDO\VLV PRVW IUHTXHQWO\ UHOLHV RQ DQ HOHFWURSKRUHWLF VL]H VKLIW RI D UDGLRODEHOHG FDWDO\VWVXEVWUDWH FRQMXJDWH $OWHUQDWLYHO\ D VDPSOH RI PROHFXOHV FRXOG EH LVRODWHG YLD FORQLQJ DQG VHTXHQFLQJ DQG XQLTXH VHTXHQFHV FRXOG EH WHVWHG LQGLYLGXDOO\ IRU FDWDO\WLF DFWLYLW\ ZLWKLQ DQ RUGHU RI PDJQLWXGH LV ODUJHVW SUDFWLFDO VDPSOH VL]H IRU WKLV PHWKRGf ,I

PAGE 146

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fEHVWf FDWDO\VWV LQ WKH OLEUDU\ ,VRODWLQJ D IHZ RI WKH SUHVXPDEO\f fEHVWf FDWDO\VWV WHOOV XV YHU\ OLWWOH DERXW WKH OLNHOLKRRG RI JHQHUDWLQJ IXQFWLRQ IURP D UDQGRP OLEUDU\ ,QGHHG WKH TXHVWLRQ RI JUHDWHVW LPSRUWDQFH LV f+RZ PDQ\ FDWDO\VWV ZHUH WKHUH LQ WKH LQLWLDO SRRO DQG ZKDW ZDV WKH GLVWULEXWLRQ RI WKHLU FDWDO\WLF SRZHU UDWH FRQVWDQWVf"f 7R GR WKLV ZH PXVW H[WUDSRODWH IURP ZKDW ZH FDQ REVHUYH IROORZLQJ PXOWLSOH URXQGV RI VHOHFWLRQ 6LPXODWLQJ ,Q 9LWUR 6HOHFWLRQV 7KLV VHFWLRQ GHVFULEHV RI D fSOD\ IRUZDUGf PRGHO IRU 96 ZKLFK FDQ EH XVHG WR LQIHU LQIRUPDWLRQ DERXW DQ LQLWLDO OLEUDU\ EDVHG RQ GDWD JDWKHUHG IROORZLQJ PXOWLSOH URXQGV RI VHOHFWLRQ 7KH PRGHO EHJLQV E\ HVWLPDWLQJ YDOXHV IRU NH\ SDUDPHWHUV IRU DQ LQLWLDO UDQGRP OLEUDU\ DQG WKHQ VLPXODWLQJ PXOWLSOH URXQGV RI VHOHFWLRQ 7KLV DOORZV XV WR GHWHUPLQH WKH LPSDFW RI YDULRXV LQLWLDO FRQGLWLRQV RQ REVHUYDEOH RXWFRPHV IROORZLQJ VHOHFWLRQ 7KH SDUDPHWHUV RI WKH PRGHO DUH RQ WKH IROORZLQJ SDJH

PAGE 147

6/ 6L]H RI LQLWLDO OLEUDU\ 5) 5HDFWLYH )UDFWLRQ 15) 1RQUHDFWLYH )UDFWLRQ 15) 6/rO 5)f Nf ,QWULQVLF UDWH RI PROHFXOH HDFK PHPEHU RI WKH LQLWLDO OLEUDU\ Q LV DVVLJQHG DQ LQWULQVLF UDWH Nf ZKLFK UHODWHV WR LWV DELOLW\ WR EH FOHDYHG XQGHU VHOHFWLRQ FRQGLWLRQV 7KH LQWULQVLF UDWH LV VHTXHQFHGHSHQGHQW DQG WKHUHIRUH XQFKDQJLQJ NVORZHVW 7KH UDWH FRQVWDQW IRU WKH VORZHVW FDWDO\VWV LQ WKH OLEUDU\ NIDVWHVW 7KH UDWH FRQVWDQW RI WKH IDVWHVW PROHFXOH LQ WKH OLEUDU\ f0LIHQ 7KH LQWULQVLF UDWH RI PROHFXOH Nff H[SUHVVHG LQ WHUPV RI KDOIOLIH 9LOLIHr OQ Nf 6W 6HOHFWLRQ WLPH WKH DPRXQW RI WLPH DOORZHG IRU D PROHFXOH WR FOHDYH LWVHOI DQG WKHUHIRUH DFKLHYH fVXUYLYDOf P RU E 'LVWULEXWLRQ IXQFWLRQ IRU UHDFWLYH PROHFXOHV 7ZR IXQFWLRQV KDYH EHHQ WHVWHG WKXV IDU D OLQHDU IXQFWLRQ ZLWK VORSH P DQG DQ H[SRQHQWLDO IXQFWLRQ ZLWK GHFD\ UDWH E / /HDNDJH WKH IUDFWLRQ RI WKH SRRO WKDW fVXUYLYHVf D URXQG RI VHOHFWLRQ LQGHSHQGHQW RI 6W DQG LQWULQVLF UDWH FRQVWDQW 0RGHO RI VXUYLYDO ILUVW RUGHU FKHPLFDO UDWH EDVHG RQ D PROHFXOHV LQWULQVLF UDWH FRQVWDQW NQf DQG WKH VHOHFWLRQ WLPH 6Wf 3FOHDYDJHf H[Sef r 6Wf 'XULQJ DQ DFWXDO LQ YLWUR VHOHFWLRQ WKH RQO\ GDWD WKDW FDQ HDVLO\f EH REWDLQHG H[SHULPHQWDOO\ DUH fIUDFWLRQ FOHDYHGf YV fWLPHf IRU WKH HQWLUH SRRO DW HDFK URXQG RI VHOHFWLRQ 2QFH WKH SURJUHVV FXUYH LV GHWHUPLQHG IRU HDFK URXQG RI VHOHFWLRQ GDWD FDQ EH H[WUDFWHG WR PDNH RWKHU XVHIXO SORWV WKDW LOOXVWUDWH WKH SURJUHVV RI WKH HQWLUH VHOHFWLRQ SURFHVV 2QH VXFK SORW FRPSDUHV WKH IUDFWLRQ FOHDYHG DW WLPH W IRU LQFUHDVLQJ URXQGV RI VHOHFWLRQ ,I fWLPHf HTXDOV WKH fVHOHFWLRQ WLPHf WKLV SORW VKRZV WKH DPRXQW RI VDPSOH FOHDYHG DW WKH FRPSOHWLRQ RI HDFK URXQG RI VHOHFWLRQ WKH VXUYLYLQJ IUDFWLRQf ,W PD\ DOVR EH XVHIXO WR SORW WKH IUDFWLRQ FOHDYHG DW D WLPH OHVV WKDQ WKH VHOHFWLRQ WLPH SDUWLFXODUO\ ZKHQ FRPSDULQJ 96 H[SHULPHQWV SHUIRUPHG ZLWK YDULRXV VHOHFWLRQ WLPHV RU ZKHQ FRPSDULQJ ,96 H[SHULPHQWV LQ ZKLFK QHDUO\ b RI WKH IUDFWLRQ KDV EHHQ FOHDYHG DW WKH FRPSOHWLRQ RI HDFK URXQG W 6Wf )URP WKHVH GDWD ZH ZRXOG OLNH WR HVWLPDWH WKH

PAGE 148

GLVWULEXWLRQ RI UDWHV DW HDFK URXQG DQG LI SRVVLEOH H[WUDSRODWH IURP WKHVH PHDVXUHPHQWV WKH GLVWULEXWLRQ RI UDWHV LQ WKH LQLWLDO OLEUDU\ 7KH PRGHO EHJLQV E\ VLPXODWLQJ DQ LQLWLDO SRRO FRPSRVHG RI XQLTXH PROHFXOHV URXJKO\ WKH VL]H RI WKH LQLWLDO OLEUDU\ VFUHHQHG LQ D W\SLFDO ,96f 7KLV LQLWLDO SRRO LV WKHQ GLYLGHG LQWR D f5HDFWLYH )UDFWLRQf 5)f DQG D f1RQUHDFWLYH )UDFWLRQf 15)f 7KH 15) LV GHILQHG DV WKH SRUWLRQ RI WKH OLEUDU\ WKDW GRHV QRW KDYH DQ\ LQWULQVLF FDWDO\WLF DFWLYLW\ DERYH WKH EDFNJURXQG FOHDYDJH UDWH IRU RXU H[SHULPHQWDO V\VWHP WKH EDFNJURXQG FOHDYDJH UDWH LV an PLQn EDVHG RQ WKH H[SHULPHQWDOO\ GHWHUPLQHG FOHDYDJH UDWH RI ULERVH E\ PDJQHVLXP LRQ %UHDNHU DQG -R\FH f $OO UHDFWLYH PROHFXOHV LQ WKH LQLWLDO OLEUDU\ 5)f DUH DVVLJQHG UDQGRPO\ D GLVFUHWH UDWH FDWHJRU\ LQ WKLV VLPXODWLRQ UDWH FDWHJRULHV ZHUH XVHGf EHWZHHQ NVc2:fVW DQG NIDVWHVW WKH LQWULQVLF UDWH IRU HDFK FDWHJRU\ LQFUHDVHV SURSRUWLRQDOO\ EHWZHHQ NVLRZHVW DQG NIDVWHVWf‘ 7KH ORZHU OLPLW IRU FDWDO\WLF UDWH FRQVWDQWV NVL2:HVL LV VHW WR WLPHV IDVWHU WKDQ WKH EDFNJURXQG UDWH WKLV ORZHU OLPLW LV VHW ERWK E\ WKH PHWKRGV RI GHWHFWLQJ D GLIIHUHQFH EHWZHHQ EDFNJURXQG DQG VLJQDO DQG WKH GXUDWLRQ RI D NLQHWLFV H[SHULPHQW ZH PXVW GHWHFW b FOHDYDJH DERYH EDFNJURXQG LQ PLQXWHV FRUUHVSRQGLQJ WR D ORZHU OLPLW RI an PLQnf 7KH UDWH FRQVWDQW RI WKH IDVWHVW PROHFXOH UHSUHVHQWHG LQ WKH LQLWLDO OLEUDU\ NIDVWHVW PXVW EH GHWHUPLQHG H[SHULPHQWDOO\ DV LW LV H[SHFWHG WR GHSHQG RQ WKH SDUWLFXODU FDWDO\WLF UHDFWLRQ DQG OLEUDU\ EHLQJ WHVWHG 3UHYLRXV UHVHDUFK VXJJHVWV DQ XSSHU OLPLW IRU '1$]\PH FOHDYDJH RI D SKRVSKRGLHVWHU ERQG LQ WKH UHJLRQ RI PLQn 8QOHVV RWKHUZLVH VWDWHG NIDVWHVW ZDV WKHUHIRUH VHW WR PLQn 9HU\ OLWWOH LV NQRZQ DERXW WKH OLNHO\ GLVWULEXWLRQ RI FDWDO\WLF UDWHV LQ D UDQGRP OLEUDU\ LQGHHG WKH GLVWULEXWLRQ RI FDWDO\WLF DEXQGDQFH YHUVXV FDWDO\WLF UDWH DQG PD\ EH

PAGE 149

DSSDUHQWO\ FKDRWLF ,I WKH GLVWULEXWLRQ RI FDWDO\WLF IXQFWLRQ FDQ LQ IDFW EH DSSUR[LPDWHG ZLWK D IXQFWLRQ LW VHHPV UHDVRQDEOH WKDW IDVWHU FDWDO\VWV ZLOO EH OHVV DEXQGDQW WKDQ VORZHU FDWDO\VWV DQG WKHUHIRUH D GHFUHDVLQJ IXQFWLRQ VKRXOG EH XVHG 7KH VLPXODWLRQV GHVFULEHG KHUH KDYH H[DPLQHG RQO\ WZR RI WKHVH IXQFWLRQV D OLQHDU DQG DQ H[SRQHQWLDO 7KH ILQDO SDUDPHWHU FRQVLGHUHG LQ WKLV PRGHO LV fOHDNDJHf GHILQHG DV WKH IUDFWLRQ RI WKH SRRO DW HDFK URXQG RI VHOHFWLRQ WKDW VXUYLYHV WKDW URXQG RI VHOHFWLRQ LQGHSHQGHQW RI ERWK LWV LQWULQVLF UDWH VHTXHQFH GHSHQGHQWf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f PROHFXOHV XQFOHDYHG EXW UHOHDVHG GXH WR LQFRPSOHWH VWUHSWDYLGLQELRWLQ LQWHUDFWLRQ RU PROHFXOHV FOHDYHG E\ WKH 1D2+ ZDVK ,96 EDVHG RQ VL]H VHSDUDWLRQ XVLQJ JHO HOHFWURSKRUHVLV FDQ KDYH OHDNDJH UHVXOWLQJ IURP LQFRPSOHWH VL]H VHSDUDWLRQ VXFK WKDW VRPH JHO H[WUDFWLRQ LQDGYHUWHQWO\ FROOHFWV LQFRUUHFW VL]H PROHFXOHV ,Q RXU ,96 V\VWHP OHDNDJH PD\ DOVR EH WKH UHVXOW RI ODVHU LQGXFHG FOHDYDJH 6XUYLYDO RI DQ\ SDUWLFXODU PROHFXOH LV EDVHG RQ LWV SUREDELOLW\ RI EHLQJ FOHDYHG JLYHQ D f6HOHFWLRQ WLPHf 6Wf DQG LV WKHUHIRUH LQGHSHQGHQW RI RWKHU PROHFXOHV 7KH SUREDELOLW\ RI EHLQJ FOHDYHG 3FOHDYDJHf LV PRGHOHG DVVXPLQJ WKDW HDFK FDWDO\VWV

PAGE 150

DSSUR[LPDWHO\ IROORZV ILUVW RUGHU FKHPLFDO NLQHWLFV DQG WKHUHIRUH LV GHWHUPLQHG VROHO\ E\ D PROHFXOHfV LQWULQVLF UDWH NQ DV DVVLJQHG E\ WKH LQLWLDO GLVWULEXWLRQ IXQFWLRQ HLWKHU OLQHDU RU H[SRQHQWLDOf DQG WLPH Wf IROORZLQJ WKH HTXDWLRQ 3FOHDYDJHf H[S$f r 6Wf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f 8VLQJ 6LPXODWLRQV WR 'HWHUPLQH WKH ,PSDFW RI WKH /HDNDJH /f 3DUDPHWHU RQ ,96 ,QLWLDO VLPXODWLRQV ZHUH GRQH WR GHWHUPLQH WKH LQIOXHQFH RI IDFWRUV VXFK DV OHDNDJH /f VHOHFWLRQ WLPH 6Wf PD[LPXP UDWH NIDVWHVWf DQG WKH LQLWLDO DEXQGDQFH RI UHDFWLYH PROHFXOHV 5)f RQ WKH FOHDYDJH SURJUHVV FXUYH DQG SRSXODWLRQ GLVWULEXWLRQ 0RVW RI WKHVH LQLWLDO VLPXODWLRQV ZHUH GRQH XVLQJ WKH VLPSOHVW GLVWULEXWLRQ IXQFWLRQ D OLQHDU GLVWULEXWLRQ ZLWK P EXW WKH FRQFOXVLRQV FDQ EH JHQHUDOL]HG WR RWKHU GLVWULEXWLRQ IXQFWLRQV

PAGE 151

)LJXUH VKRZV WKH UHVXOWV RI VLPXODWLRQV IRU WKUHH WHVW FDVHV HDFK GLIIHULQJ LQ WKH YDOXH RI WKH OHDNDJH SDUDPHWHU / UDQJLQJ IURP ]HUR WR bf 3DQHO $ VKRZV WKH LQLWLDO GLVWULEXWLRQ IRU HDFK SRSXODWLRQ WKH SRSXODWLRQ ZDV GLVWULEXWHG DFFRUGLQJ WR D OLQHDU IXQFWLRQ ZLWK P IRU DOO WKUHH WHVW FDVHVf 3DQHO ( SORWV H[SHULPHQWDOO\ REVHUYDEOH GDWD WKH SURJUHVV FXUYH IRU HDFK WHVW FDVH IROORZLQJ VXFFHVVLYH URXQGV RI VHOHFWLRQ WKH SURJUHVV FXUYH LV D SORW RI WKH IUDFWLRQ RI WKH SRSXODWLRQ FOHDYHG ZLWK LQFUHDVLQJ WLPHf )URP WKH SURJUHVV FXUYH WKH DPRXQW RI HDFK SRSXODWLRQ FOHDYHG DW WLPH VHOHFWLRQ WLPH 6Wf RU WLPH PLQXWHV ZDV H[WUDFWHG IRU HDFK URXQG DQG SORWWHG LQ 3DQHO % DQG & $OWKRXJK GXULQJ WKH H[SHULPHQWDO LPSOHPHQWDWLRQ RI D VHOHFWLRQ ZH FDQ RQO\ REVHUYH WKH SURJUHVV FXUYH WKH VLPXODWLRQV DOORZ XV WR REVHUYH WKH SRSXODWLRQ GLVWULEXWLRQ DW HDFK URXQG 7KHVH GDWD DUH SORWWHG LQ 3DQHO WKH LQWULQVLF UDWH NQ IRU HDFK PROHFXOH KDV EHHQ FRQYHUWHG LQWR LWV FRUUHVSRQGLQJ KDOI OLIH VR D UDWH RI PLQn LV SORWWHG DV D ncOLIH RI PLQXWHVf 0RVW RI WKH PROHFXOHV LQ WKH LQLWLDO OLEUDU\ DUH QRQUHDFWLYH DQG WKHUHIRUH DUH FOHDYHG DW WKH EDFNJURXQG UDWH RI RQO\ n PLQn 7KH IUDFWLRQ RI WKH SRSXODWLRQ FOHDYHG DW WKH HQG RI WKH ILUVW URXQG RI VHOHFWLRQ W 6Wf UHIHUUHG WR DV fVXUYLYRUVf LV WKHUHIRUH H[WUHPHO\ ORZ DQG PRVWO\ D UHIOHFWLRQ RI OHDNDJH :LWK HDFK VXFFHVVLYH URXQG RI VHOHFWLRQ WKH FDWDO\WLF PROHFXOHV DUH HQKDQFHG ZKLOH WKH QRQUHDFWLYH IUDFWLRQ LV GLPLQLVKHG (YHQWXDOO\ WKH FDWDO\WLF PROHFXOHV UHDFK VXIILFLHQW DEXQGDQFH UHODWLYH WR WKH QRQUHDFWLYH PROHFXOHV WKDW WKH\ VXUSDVV WKH WKUHVKROG RI GHWHFWLRQ GHILQHG DV b DERYH EDFNJURXQG FOHDYDJH DW WLPH 6Wf $OWKRXJK HDFK FDVH VKRZQ LQ )LJXUH KDG WKH VDPH LQLWLDO GLVWULEXWLRQ DQG DEXQGDQFH RI FDWDO\VWV LQFUHDVLQJ WKH DPRXQW RI OHDNDJH IURP ]HUR WR b GHOD\HG ILUVW

PAGE 152

GHWHFWLRQ RI VXUYLYRUV IURP URXQG WR URXQG ,QFUHDVLQJ WKH OHDNDJH SDUDPHWHU HYHQ IXUWKHU WR b GHOD\HG WKH HDUOLHVW GHWHFWLRQ RI VXUYLYRUV WR URXQG 7KLV LQGLFDWHV WKDW WKH OHDNDJH SDUDPHWHU SOD\V D VLJQLILFDQW UROH LQ GHWHUPLQLQJ WKH QXPEHU RI URXQGV RI VHOHFWLRQ UHTXLUHG WR DFKLHYH GHWHFWDEOH OHYHOV RI FDWDO\VLV :LWK VXIILFLHQW URXQGV RI VHOHFWLRQ HDFK WHVW FDVH HYHQWXDOO\ UHDFKHV D fVXUYLYRUVKLS SODWHDXf D SRLQW EH\RQG ZKLFK IXUWKHU URXQGV RI VHOHFWLRQ QR ORQJHU VLJQLILFDQWO\ LQFUHDVHV WKH IUDFWLRQ RI WKH SRSXODWLRQ VXUYLYLQJ HDFK URXQG 3DQHO %f 3DQHO & VKRZV D VLPLODU SODWHDX IRU WKH DPRXQW FOHDYHG DW W PLQXWHV LW LV XVHIXO WR SORW WKH DPRXQW FOHDYHG DW D VHW WLPH IRU FRQGLWLRQV ZKHUH SRSXODWLRQV DUH VHOHFWHG ZLWK YDULRXV VHOHFWLRQ WLPHVf ,Q WKLV VLPXODWLRQ HDFK WHVW FDVH UHDFKHV D VXUYLYRUVKLS SODWHDX QHDU b FOHDYHG 3DQHO VKRZV WKH GLVWULEXWLRQ RI FDWDO\VWV IRU HDFK WHVW FDVH DW VXFFHVVLYH URXQGV RI VHOHFWLRQ &RPSDULQJ 3DQHO % RU &f ZLWK 3DQHO LOOXVWUDWHV WKDW DW WKH VDPH URXQG WKH VXUYLYRUVKLS EHJLQV WR SODWHDX WKH GLVWULEXWLRQ RI FDWDO\VWV LQ WKH VHOHFWHG SRSXODWLRQV UHDFK D VWDEOH VWDWH UHIHUUHG WR DV WKH fVWDELOL]HG SRSXODWLRQf ZKHUH WKH SURSRUWLRQDOLW\ EHWZHHQ FDWDO\WLF UDWHV RI VXUYLYLQJ PROHFXOHV UHPDLQV FRQVWDQW HYHQ ZLWK DGGLWLRQDO URXQGV RI VHOHFWLRQ 2QFH WKLV VXUYLYRUVKLS UHDFKHV D SODWHDX WKHUH LV QR VHOHFWLYH HQKDQFHPHQW RI IDVWHU PROHFXOHV RYHU WKH UHPDLQLQJ VORZHU PROHFXOHV UHVXOWLQJ LQ WKH VWDELOL]DWLRQ RI WKH VXUYLYLQJ SRSXODWLRQ &DVH DFKLHYHV D VXUYLYRUVKLS SODWHDX RI b E\ URXQG SDQHO %f DQG SDQHO VKRZV WKDW WKH SRSXODWLRQ GLVWULEXWLRQ IRU FDVH UHPDLQV UHODWLYHO\ FRQVWDQW IURP URXQG WKURXJK WKH HQG RI WKH VLPXODWLRQ URXQG f &DVH DQG FDVH DFKLHYH D VXUYLYRUVKLS SODWHDX RI b PXFK ODWHU WKDQ FDVH URXQG DQG UHVSHFWLYHO\f DV D UHVXOW RI D KLJKHU OHDNDJH SDUDPHWHU / b DQG b UHVSHFWLYHO\f 'HVSLWH WKH

PAGE 153

DGGLWLRQDO URXQGV UHTXLUHG WR UHDFK D VXUYLYRUVKLS SODWHDX WKH SRSXODWLRQ GLVWULEXWLRQ IRU DOO WHVW FDVHV FRQYHUJHV WR WKH VDPH SRSXODWLRQ GLVWULEXWLRQ WKLV RFFXUV ZKHQ WKHUH LV QR ORQJHU VHOHFWLYH HQKDQFHPHQW IRU IDVWHU PROHFXOHV DQG FRUUHVSRQGV WR WKH VXUYLYRUVKLS SODWHDXf 7KLV LV PDQLIHVWHG E\ WKH FRQYHUJHQFH RI SURJUHVV FXUYHV IRU DOO WHVW FDVHV RQFH HDFK WHVW FDVH UHDFKHV D VXUYLYRUVKLS SODWHDXf 7KLV GHPRQVWUDWHV WKDW DOWKRXJK WKH OHDNDJH SDUDPHWHU GUDPDWLFDOO\ DIIHFWV WKH QXPEHU RI URXQGV QHFHVVDU\ WR LVRODWH FDWDO\VWV LW GRHV QRW DOWHU WKH GLVWULEXWLRQ RI VHOHFWHG FDWDO\VWV IROORZLQJ VXIILFLHQW URXQGV RI VHOHFWLRQf 7KH GLVWULEXWLRQ RI VHOHFWHG FDWDO\VWV LQ WKH VWDELOL]HG SRSXODWLRQ EHDUV DQ LQWHUHVWLQJ UHODWLRQVKLS WR WKH VHOHFWLRQ WLPH 7KHUH DUH WKUHH GLVWLQFW SRSXODWLRQ ]RQHV GHILQHG LQ UHODWLRQ WR WKH 6W ]RQH $ LV FRPSULVHG RI PROHFXOHV ZLWK KDOIOLYHV ODUJHU VORZHUf WKDQ 6W ]RQH % LV FRPSULVHG RI PROHFXOHV ZLWK KDOIOLYHV EHWZHHQ 6W DQG DSSUR[LPDWHO\ 6W IROG IDVWHUf DQG ]RQH & LV FRPSULVHG RI PROHFXOHV ZLWK KDOIOLYHV EHWZHHQ 6W DQG NIDVWHVW $OWKRXJK LQ WKLV VLPXODWLRQ DOO PROHFXOHV UHJDUGOHVV RI WKHLU UDWH FRQVWDQW ZHUH LQLWLDOO\ SUHVHQW LQ HTXDO DPRXQWV PROHFXOHV LQ ]RQH $ VXUYLYH ZLWK LQFUHDVLQJO\ ORZHU SUREDELOLW\ WKDQ IDVWHU PROHFXOHV ]RQH % DQG &f (YHQWXDOO\ WKH VXUYLYLQJ SRSXODWLRQ UHDFKHV D SRLQW ZKHUH PROHFXOHV IURP ]RQH $ DUH HIIHFWLYHO\ DEVHQW 8QOHVV WKHLU LQLWLDO DEXQGDQFH LV H[FHHGLQJO\ KLJK UHODWLYH WR IDVWHU PROHFXOHV PROHFXOHV RI ]RQH $ DUH TXLFNO\ HOLPLQDWHG IURP WKH VHOHFWHG SRSXODWLRQ ,Q IDFW JLYHQ DQ LQLWLDO SRSXODWLRQ FRQVLVWLQJ FDWDO\VWV ZLWK RQH RI RQO\ WZR SRVVLEOH LQWULQVLF UDWHV RQH ZLWK UDWH HTXDO WR 6W n$OLIHDVW 6Wf DQG WKH RWKHU WHQ WLPHV VORZHU n$OLIHVLRW rf WKH VORZHU FDWDO\VWV PXVW EH IROG PRUH DEXQGDQW WKDQ WKH

PAGE 154

IDVWHU FDWDO\VW LQ RUGHU IRU ERWK FDWDO\VWV WR EH LQ HTXDO DEXQGDQFH IROORZLQJ RQH URXQG RI VHOHFWLRQ ,I WKUHH URXQG RI VHOHFWLRQ DUH UHTXLUHG WKH LQLWLDO DEXQGDQFH RI WKH VORZHU FDWDO\VWV QHHGV WR EH f WLPHV PRUH DEXQGDQW WKDQ WKH IDVWHU FDWDO\VW 7KH VLWXDWLRQ LV HYHQ PRUH GUDPDWLF IRU fYHU\ VORZf FDWDO\VWV LI D YHU\ VORZ FDWDO\VWV LV WLPHV VORZHU n$OLIHYHU\6LR: rf WKDQ WKH IDVW FDWDO\VW f$OLIHDVW 6Wf WKHQ WKLV YHU\ VORZ FDWDO\VW PXVW EH LQ IROG JUHDWHU DEXQGDQFH LQ WKH LQLWLDO SRRO WR VXUYLYH LQ HTXDO SURSRUWLRQ DV WKH IDVWHU FDWDO\VW IROORZLQJ RQH URXQG RI VHOHFWLRQ DQG IROG f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f =RQH & FRUUHVSRQGV WR WKH VLWXDWLRQ GHVFULEH DERYH LQ ZKLFK DOO WKH PROHFXOHV DUH VXIILFLHQWO\ IDVW UHODWLYH WR 6W f$OLIH 6Wf VXFK WKDW QHDUO\ DOO PROHFXOHV DUH FOHDYHG E\ WKH FRPSOHWLRQ RI HDFK URXQG RI VHOHFWLRQ DQG QR VHOHFWLYH HQKDQFHPHQW H[LVWV &RQVHTXHQWO\ PROHFXOHV LQ ]RQH & FRQWLQXH WR SHUVLVW ZLWKRXW DQ\ FKDQJH LQ WKHLU UHODWLYH SURSRUWLRQ WR HDFK RWKHU 7KHVH SURSHUWLHV RI WKH SRSXODWLRQ GLVWULEXWLRQ FDQ EH VHHQ LQ )LJXUH RU LQ PRUH GHWDLO LQ 7DEOH

PAGE 155

7$%/( $ 3UREDELOLW\ RI VXUYLYDO IROORZLQJ VXFFHVVLYH URXQGV RI VHOHFWLRQ EDVHG RQ LQWULQVLF UDWH DQG VHOHFWLRQ WLPHf 6W KDOI OLIH PLQf ; URXQG RI VHOHFWLRQ LQWLDO UDWH PLQnf ( 3UREDELOLW\ RI VXUYLYDO IROORZLQJ URXQG ; RI VHOHFWLRQ 7$%/( % 3RSXODWLRQ GLVWULEXWLRQ IROORZLQJ VXFFHVVLYH URXQGV RI VHOHFWLRQ 6W KDOI OLIH PLQf ; URXQG LQWLDO UDWH PLQ ff ( RI VHOHFWLRQ SHUFHQW RI SRSXODWLRQ DIWHU ; URXQGV RI VHOHFWLRQ LQLWLDO SRSXODWLRQ GLVWULEXWLRQ ]RQH $ EHORZ OHYHO RI GHWHFWLRQ bf ]RQH % SRSXODWLRQ LQ IOX[ ]RQH & LQLWLDO SURSRUWLRQDOLW\ LV XQFKDQJHG ZLWKLQ bf

PAGE 156

([DPLQLQJ WKH ,PSDFW RI WKH 6HOHFWLRQ 7LPH 6Wf )LJXUH VKRZV WKH UHVXOWV RI DQRWKHU VLPXODWLRQ LQ ZKLFK WKH VHOHFWLRQ WLPH 6Wf ZDV YDULHG 6W DQG PLQXWHV IRU FDVH DQG UHVSHFWLYHO\f ZKLOH DOO RWKHU SDUDPHWHUV ZHUH KHOG FRQVWDQW ,QFUHDVLQJ WKH VHOHFWLRQ WLPH IURP PLQXWHV WR LQLWLDOO\ UHVXOWHG LQ D ODUJHU IUDFWLRQ RI WKH SRSXODWLRQ EHLQJ FOHDYHG DW WKH HQG RI HDFK URXQG RI VHOHFWLRQ FRUUHVSRQGLQJ WR HDUOLHU LQLWLDO GHWHFWLRQ ,QFUHDVLQJ WKH VHOHFWLRQ WLPH IXUWKHU WR PLQXWHV UHYHUVHG WKLV WUHQG GHPRQVWUDWLQJ WKDW WKH URXQG DW ZKLFK FDWDO\VWV DUH ILUVW GHWHFWHG DW W 6Wf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f :KHQ WKH VHOHFWLRQ WLPH LV VKRUW UHODWLYH WR WKH KDOI OLIH RI WKH IDVWHVW FDWDO\VW D SRSXODWLRQ RI RQO\ WKH IDVWHVW FDWDO\VWV LV LVRODWHG 7KLV VKRUW VHOHFWLRQ WLPH DOVR UHVXOWV LQ D SODWHDX EHORZ b IRU WKH IUDFWLRQ RI WKH SRSXODWLRQ FOHDYHG DW WKH HQG RI WKH URXQG RI VHOHFWLRQ FOHDYDJH DW WLPH W 6Wf %HFDXVH WKHUH LV QR VLJQLILFDQW GLIIHUHQWLDO HQULFKPHQW IRU FDWDO\VWV ZLWK KDOIOLIH IROG VORZHU WKDQ WKH VHOHFWLRQ WLPH ORQJHU VHOHFWLRQ WLPHV HTXDOO\ HQULFK DOO PROHFXOHV ZLWK D KDOIOLIH UDWHf DSSUR[LPDWHO\ IROG ORZHU WKDQ WKH

PAGE 157

VHOHFWLRQ WLPH 7KLV SURGXFHV D VWDELOL]HG SRSXODWLRQ WKDW LQFOXGHV FDWDO\VWV ZLWK KDOI OLYHV UDWHVf UDQJLQJ IURP 6W WR NIDVWHVW :KHQ WKH VHOHFWLRQ WLPH LV YHU\ VKRUW RQO\ WKH IDVWHVW PROHFXOHV LQ WKH SRSXODWLRQ KDYH D KLJK SUREDELOLW\ RI VXUYLYLQJ HDFK URXQG WKHUHIRUH PXOWLSOH URXQGV RI VHOHFWLRQ HOLPLQDWH DOO EXW WKH IDVWHVW PROHFXOHV UDWKHU WKDQ LVRODWLQJ D SRSXODWLRQ ZLWK D UDQJH RI UDWHVf ([DPLQLQJ WKH ,PSDFW RI WKH 5DWH RI WKH )DVWHVW &DWDO\VWV $QRWKHU VLPXODWLRQ ZDV SHUIRUPHG LQ ZKLFK WKH KDOIOLIH RI WKH IDVWHVW FDWDO\VW LQ WKH LQLWLDO SRSXODWLRQ ZDV H[DPLQHG IURP nOLIH PLQXWHV WR DQG PLQXWHV ZKLOH KROGLQJ RWKHU SDUDPHWHUV RI WKH PRGHO FRQVWDQW SDUDPHWHUV VKRZQ LQ )LJXUH f 7KH QXPEHU RI FDWDO\VWV ZLWK DQ\ JLYHQ KDOIOLIH JUHDWHU WKDQ WKH VHOHFWLRQ WLPH ZDV KHOG FRQVWDQW DW n IRU HDFK WHVW FDVH ZKLOH WKH GLVWULEXWLRQ RI FDWDO\VWV ZLWK KDOI OLYHV OHVV WKDQ WKH VHOHFWLRQ WLPH YDULHG 7KH SRSXODWLRQ RI FDWDO\VWV LQ FDVH ZDV GLVWULEXWHG DFFRUGLQJ WR D OLQHDU IXQFWLRQ ZLWK VORSH RI [ n IRU DOO FDWDO\VWV ZLWK KDOI OLYHV EHWZHHQ 6W DQG NIDVWHVWff 7KH SRSXODWLRQ RI FDWDO\VWV LQ FDVH ZDV GLVWULEXWHG DFFRUGLQJ WR WKH SURSRUWLRQDOLW\ HVWDEOLVKHG LQ FDVH IRU DOO FDWDO\VWV ZLWK KDOI OLYHV EHWZHHQ 6W DQG NIDVWHVW f ZKLOH KROGLQJ WKH UHDFWLYH IUDFWLRQ FRQVWDQW 5) nf 7KH FDWDO\VWV IRU FDVH ZHUH GLVWULEXWHG IRU UDWHV EHWZHHQ 6W DQG NIDVWHVWAf DFFRUGLQJ WR WKH SURSRUWLRQDO FKDQJHV LQ FDWDO\VWV LQ FDVH DQG KROGLQJ WKH UHDFWLYH IUDFWLRQ WKH VDPH 2nf $OO RI WKH FDWDO\VWV IRU FDVH ZHUH GLVWULEXWHG LQWR D VLQJOH UDWH FODVV ZLWK UDWH NIDVWHVW DQG UHDFWLYH IUDFWLRQ HTXDO WR n )LJXUH VKRZV WKDW UHGXFLQJ WKH UDWH FRQVWDQW RI WKH IDVWHVW FDWDO\VW LQ D SRSXODWLRQ NIDVWHVW LV ORZHU DQG WKHUHIRUH VORZHU FDWDO\VWf KDV D VLPLODU HIIHFW DV

PAGE 158

LQFUHDVLQJ WKH VHOHFWLRQ WLPH RQ WKH SRSXODWLRQ GLVWULEXWLRQ IROORZLQJ VHOHFWLRQ %HFDXVH VHOHFWLRQV HQULFK IRU PROHFXOHV ZLWK KDOI OLYHV UDWHVf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f 7KH FDWDO\VWV LQ FDVH DQG FDVH ZHUH ERWK GLVWULEXWHG DFFRUGLQJ WR WKH VDPH OLQHDU IXQFWLRQ ZLWK P [ nf EXW FDVH KDG WLPHV PRUH UHDFWLYH PROHFXOHV LQ LWV LQLWLDO SRSXODWLRQ 5) n UDWKHU WKDQ ff 1RW RQO\ GLG WKH VHOHFWLRQV IRU FDVH UHYHDO FDWDO\VWV DW DQ HDUOLHU URXQG RI VHOHFWLRQ EXW WKH GLVWULEXWLRQ RI FDWDO\VWV LQ WKH VWDELOL]HG SRSXODWLRQV GLIIHUHG IURP WKH GLVWULEXWLRQ IRU FDVH %RWK VWDELOL]HG SRSXODWLRQV IRU FDVH DQG KDYH D GHFUHDVLQJ WUHQG OHVV FDWDO\WLF DEXQGDQFH ZLWK LQFUHDVLQJ FDWDO\WLF UDWHf DV LQ WKH LQLWLDO OLEUDU\ EXW FDVH GHFUHDVHG PRUH UDSLGO\ WKDQ FDVH 5HDVRQLQJ WKDW LW ZDV QRW WKH VORSH RI WKH LQLWLDO GLVWULEXWLRQ IXQFWLRQ EXW UDWKHU WKH UDWLR EHWZHHQ IDVWHU DQG VORZHU PROHFXOHV HVWDEOLVKHG E\ WKH LQLWLDO GLVWULEXWLRQ IXQFWLRQ DQG WKH UHDFWLYH IUDFWLRQ WKDW GHWHUPLQHG WKH GLVWULEXWLRQ RI FDWDO\VWV LQ WKH VWDELOL]HG SRSXODWLRQ D WKLUG FDVH ZDV WHVWHG 7KLV WKLUG FDVH GLIIHUHG IURP FDVH LQ ERWK WKH QXPEHU RI UHDFWLYH FDWDO\VWV 5) nYV ff DQG LQ WKH

PAGE 159

GLVWULEXWLRQ RI WKRVH FDWDO\VWV P [ n YV [ nf EXW WKH UDWLR fSURSRUWLRQDOLW\ff RI FDWDO\VWV ZLWK UDWH NQ WR FDWDO\VWV ZLWK UDWH NQL ZKHUH Q HTXDOV DQ\ JLYHQ UDWHf ZDV KHOG FRQVWDQW IRU D OLQHDU GLVWULEXWLRQ WKH fSURSRUWLRQDOLW\f UHODWLRQVKLS LV PA5)Vf 5)cf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f &DWDO\VWV LQ FDVH DQG ZHUH GLVWULEXWHG DFFRUGLQJ WR D OLQHDU IXQFWLRQ ZLWK P DQG [ n UHVSHFWLYHO\f DV PRGHOHG LQ HDUOLHU VLPXODWLRQ $OWKRXJK FDVH DQG FDVH DSSURDFK SRSXODWLRQ VWDELOLW\ LQ QHDUO\ LGHQWLFDO PDQQHUV WKH GLIIHUHQFH EHWZHHQ WKHLU SRSXODWLRQ GLVWULEXWLRQV FDQ EH GHWHFWHG E\ H[DPLQLQJ FOHDYDJH SURILOH DW HDUO\ WLPHV 7KH GLVWULEXWLRQ RI FDWDO\VWV LQ WKH VWDELOL]HG SRSXODWLRQ GLIIHUV GUDPDWLFDOO\ IRU HDFK WHVW FDVH ,Q HDFK FDVH WKH SURSRUWLRQDOLW\ EHWZHHQ FDWDO\VWV ZLWK KDOI OLYHV OHVV WKDQ 6W

PAGE 160

PLUURUV WKH SURSRUWLRQDOLW\ LQ WKH LQLWLDO SRSXODWLRQ 7KLV UHODWLRQVKLS FDQ EH VHHQ LQ )LJXUH DQG 7DEOH 7KH DSSURDFK WR D VWDELOL]HG SRSXODWLRQ VHOHFWLRQ SODWHDXf LV VORZHU DQG WKH SHDN RI FDWDO\WLF GLVWULEXWLRQV IRU VHOHFWHG SRSXODWLRQV LV VKLIWHG FORVHU WR WKH 6W IRU WKH H[SRQHQWLDO GLVWULEXWLRQ %RWK IDFWRUV LQGLFDWH WKDW WKH SRSXODWLRQ LV VWLOO HQULFKLQJ IRU IDVWHU PROHFXOHV WKH VORZ HQULFKPHQW LV D UHVXOW RI DQ LQFUHDVHG GLVWULEXWLRQ RI FDWDO\VWV ZLWK KDOI OLYHV EHWZHHQ r6W DQG 6W FRPSDUHG WR FDWDO\VWV ZLWK KDOI OLIH 6W LQ WKH LQLWLDO GLVWULEXWLRQ $OWKRXJK WKH VWDELOL]HG SRSXODWLRQ JLYHV OLWWOH LQIRUPDWLRQ DERXW WKH GLVWULEXWLRQ RI FDWDO\VWV ZLWK KDOI OLYHV EHWZHHQ r DQG 6W WKH UDSLGLW\ DW ZKLFK LQFUHDVLQJ URXQGV RI VHOHFWLRQ DSSURDFKHV VHOHFWLRQ SODWHDX PD\ EH XVHG DV DQ LQGLFDWRU RI WKH LQLWLDO GLVWULEXWLRQ RI FDWDO\VWV LQ WKLV UDQJH 'HFLSKHULQJ WKH 'LVWULEXWLRQ )XQFWLRQ IRU &DWDO\VWV *LYHQ WKDW GXULQJ DQ DFWXDO LQ YLWUR VHOHFWLRQ WKH RQO\ GDWD WKDW FDQ HDVLO\f EH REWDLQHG H[SHULPHQWDOO\ DUH fIUDFWLRQ FOHDYHGf YV fWLPHf IRU WKH HQWLUH SRRO DW HDFK URXQG RI VHOHFWLRQ KRZ WKHQ FDQ ZH HVWLPDWH WKH LQLWLDO GLVWULEXWLRQ RI FDWDO\WLF SRZHU LQ OLEUDU\" 7KH RQO\ H[SHULPHQWDOO\ PHDVXUDEOH YDULDEOH LV WKH SURJUHVV FXUYH IRU FOHDYDJH DW HDFK URXQG RI WKH VHOHFWLRQ SURFHVV )URP WKLV ZH FDQ GHWHUPLQH WKH SURJUHVV RI WKH HQWLUH VHOHFWLRQ E\ SORWWLQJ WKH DPRXQW FOHDYHG DW WKH FRPSOHWLRQ RI HDFK URXQG W 6Wf :KHQ WKLV VHOHFWLRQSURJUHVV SORW UHDFKHV D SODWHDX ZH NQRZ ZH KDYH LVRODWHG D VWDELOL]HG SRSXODWLRQ )XUWKHU URXQGV RI VHOHFWLRQ ZLOO QRW VLJQLILFDQWO\ DOWHU WKH SURSRUWLRQ RI FDWDO\VWV LQ WKLV VWDELOL]HG SRSXODWLRQ DQG RI HTXDO PRUH LPSRUWDQFH WKH SURSRUWLRQDOLW\ RI WKLV SRSXODWLRQ GLUHFWO\ PLUURUV WKH LQLWLDO OLEUDU\ IRU FDWDO\VWV ZLWK

PAGE 161

KDOI OLYHV 6Wf ,I WKH VHOHFWLRQSURJUHVV SORW LQGLFDWHV D SODWHDX EHORZ b FOHDYDJH WKH VWDELOL]HG SRSXODWLRQ FRQWDLQV RQO\ WKH IDVWHVW PROHFXOHV +RZ GR ZH XVH WKH LQIRUPDWLRQ JDLQHG IURP WKHVH VLPXODWLRQV WR HVWLPDWH WKH SRSXODWLRQ GLVWULEXWLRQ IRU D UHDO OLEUDU\ XQGHUJRLQJ VHOHFWLRQ" 7KH ILUVW WKLQJ ZH PXVW FRQVLGHU LV WKH UDQJH RI UDWHV ZH DUH LQWHUHVWHG LQ HVWLPDWLQJ 8QOHVV VLJQLILFDQW DFWLYLW\ FDQ EH GHWHFWHG LQ RQO\ D IHZ URXQGV OHVV WKDQ WKUHHf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fSOD\ IRUZDUGf PRGHO GHVFULEHG EHORZ FRQVLGHUV WKLV PLQRU HIIHFWf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

PAGE 162

OLEUDU\ WR HVWLPDWH WKH GLVWULEXWLRQ RI FDWDO\VWV LQ WKH LQLWLDO OLEUDU\ 7KH SUR[LPLW\ RI WKH SHDN IRU FDWDO\WLF GLVWULEXWLRQ LV DOVR LQGLFDWLYH RI WKH UHODWLYH SURSRUWLRQ RI FDWDO\VWV ZLWK KDOI OLYHV EHWZHHQ 6W DQG 6W 7KLV PHWKRG UHOLHV RQ GHWHUPLQLQJ WKH SRSXODWLRQ GLVWULEXWLRQ DW D OHDVW RQH URXQG SUHIHUDEO\ D URXQG VKRUWO\ IROORZLQJ SRSXODWLRQ VWDELOL]DWLRQf 7KH SRSXODWLRQ GLVWULEXWLRQ DW D URXQG FDQ EH GHWHUPLQHG H[SHULPHQWDOO\ E\ VDPSOLQJ LQGLYLGXDO FDWDO\VWV E\ FORQLQJf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f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a6W $OWKRXJK ERWK WHFKQLTXHV ZLOO HVWLPDWH WKH DEXQGDQFH RI FDWDO\VWV ZLWK KDOI OLYHV JUHDWHU WKDQ 6W LQ WKH VWDELOL]HG SRSXODWLRQ WKLV XQIRUWXQDWHO\ SURYLGHV QR LQIRUPDWLRQ DERXW WKH GLVWULEXWLRQ RI WKHVH FDWDO\VWV LQ WKH LQLWLDO SRSXODWLRQ 7KH LUUHOHYDQFH RI FDWDO\VWV ZLWK KDOIOLYHV JUHDWHU WKDQ 6W LV

PAGE 163

SDUWLFXODUO\ GLVKHDUWHQLQJ WR WKH SHUVRQ ZKR LQGLYLGXDOO\ WHVWHG WKH UDWHV RI FORQHV SDUWLFXODUO\ LI WKH PDMRULW\ RI PROHFXOHV VDPSOH KDYH UDWHV JUHDWHU WKDQ 6W UHTXLULQJ WKDW PRUH LQGLYLGXDO FDWDO\VWV EH VDPSOHGf $Q DOWHUQDWLYH WHFKQLTXH IRU HVWLPDWLQJ WKH LQLWLDO SRSXODWLRQ GLVWULEXWLRQ ZRXOG EH XVHIXO 5DWKHU WKDQ fZRUNLQJ EDFNZDUGf IURP WKH VWDELOL]HG SRSXODWLRQ WKH VLPXODWLRQ PRGHO GHYHORSHG KHUH FDQ EH fSOD\HG IRUZDUGf ZLWK YDU\LQJ VWDUWLQJ SDUDPHWHUV WR RSWLPL]H ILW WR WKH H[SHULPHQWDOO\ REVHUYHG GDWD PRVW QRWDEO\ WKH URXQG DW ZKLFK FOHDYDJH LV ILUVW GHWHFWHG WKH SHUFHQW RI WKH SRSXODWLRQ FOHDYHG DW VHOHFWLRQ SODWHDX DQG WKH UDWH DW ZKLFK WKH VHOHFWLRQ DSSURDFKHV VHOHFWLRQ SODWHDXf 2QO\ WZR SDUDPHWHUV QHHG WR EH YDULHG WKH GLVWULEXWLRQ DQG VL]H RI WKH UHDFWLYH IUDFWLRQ OHDNDJH DQG NIDVWHV FDQ EH GHWHUPLQHG H[SHULPHQWDOO\ DQG WKHUHIRUH GR QRW QHHG WR EH HVWLPDWHGf 7KLV PHWKRG RIIHUV WKH DGYDQWDJH WKDW LW FRQVLGHUV WKH LPSDFW RI FDWDO\VWV ZLWK KDOI OLYHV JUHDWHU WKDQ 6W DQG XVHV GDWD IURP HYHU\ URXQG RI VHOHFWLRQ UDWKHU WKDQ MXVW WKH VWDELOL]HG SRSXODWLRQ RI D VLQJOH URXQGf 6XPPDU\ RI 5HVXOWV 7DEOH QH[W SDJHf VXPPDUL]HV WKH PDMRU UHVXOWV RI WKH H[SHULPHQWV DQG VLPXODWLRQV GHVFULEHG LQ WKLV FKDSWHU

PAGE 164

7DEOH 6XPPDU\ RI 5HVXOWV IURP &KDSWHU 96 H[SHULPHQW f &OHDYHG SURGXFWV ZHUH OHVV WKDQ b RI WKH WRWDO DIWHU WKH ILUVW URXQG RI VHOHFWLRQ f 7KH DPRXQW RI FOHDYHG SURGXFWV LQ VHOHFWLRQV ZLWK FDJHGULERVH LQFUHDVHG VWHDGLO\ ZLWK VXFFHVVLYH URXQGV RI VHOHFWLRQ WR b E\ URXQG f 7KH DPRXQW RI 6HOHFWLRQ + LQFUHDVHG GUDPDWLFDOO\ DIWHU WKH IRXUWK URXQG RI VHOHFWLRQ f (LJKW URXQGV RI VHOHFWLRQV IRU VDPSOHV FRQWDLQLQJ WKH FDJHGULERVH GLG QRW VLJQLILFDQWO\ HQULFK IRU FDWDO\VWV 6HOHFWLRQV ZLWK FDJHGULERVH GLG QRW VKRZ WLPHGHSHQGHQW FOHDYDJH DQG ZKHQ SRROV IURP 5RXQG DQG ZHUH UHn V\QWKHVL]HG ZLWK FDJHGULERVH WKH RQO\ FOHDYDJH RFFXUUHG LPPHGLDWHO\ DIWHU ODVHU SKRWRO\VLV3RROV IURP VHOHFWLRQV ZLWK FDJHGULERVH IROORZLQJ 5RXQG ZHUH UHV\QWKHVL]HG ZLWK XQSURWHFWHG ULERVH EXW GLG QRW VKRZ DQ\ WLPH GHSHQGHQW FOHDYDJH f 6HOHFWLRQ + UHVXOWHG LQ VLJQLILFDQW HQULFKPHQW IRU FDWDO\VWV f 6HYHUDO RI WKH FDWDO\VWV LVRODWHG IURP 6HOHFWLRQ + ZHUH QHDUO\ LGHQWLFDO WR FDWDO\VWV LVRODWHG LQ SUHYLRXV VHOHFWLRQ H[SHULPHQWV EXW VLQJOHQXFOHRWLGH YDULDQWV SRVVHVVHG UDWHV WKDW GLIIHUHQW E\ DQ RUGHU RI PDJQLWXGH f 7KH EXON SRRO RI FDWDO\VWV IURP 6HOHFWLRQ + LV 0J LQGHSHQGHQW ,96 6LPXODWLRQV f 7KH OHDNDJH SDUDPHWHU GUDPDWLFDOO\ DIIHFWV WKH QXPEHU RI URXQGV RU VHOHFWLRQ QHFHVVDU\ WR DFKLHYH HQULFKPHQW IRU DFWLYH FDWDO\VWV f 7KH UHODWLYH DEXQGDQFH RI FDWDO\VWV FRPSDUHG WR QRQUHDFWLYH PROHFXOHV WKH 5HDFWLYH )UDFWLRQf LQIOXHQFHV WKH QXPEHU RI URXQGV QHFHVVDU\ WR GHWHFW FDWDO\VWV f 7KH VHOHFWLRQ WLPH SDUDPHWHU LQIOXHQFHV WKH QXPEHU RI URXQGV RI VHOHFWLRQ UHTXLUHG WR ILUVW GHWHFW DFWLYH FDWDO\VWV f:KHQ WKH DPRXQW RI FOHDYDJH RI D SRRO DW WKH HQG RI HDFK URXQG RI VHOHFWLRQ UHDFKHV D SODWHDX WKH SRSXODWLRQ KDV UHDFKHG D VWDEOH SRLQW ZKHUH WKH GLVWULEXWLRQ RI FDWDO\WLF UDWHV UHPDLQV ODUJHO\ XQFKDQJHG ZLWK PRUH URXQGV RI VHOHFWLRQ f ,I WKH VHOHFWLRQ WLPH LV VKRUW UHODWLYH WR WKH KDOIOLIH RI WKH IDVWHVW FDWDO\VWV LQ D OLEUDU\ WKH VWDELOL]HG SRSXODWLRQ ZLOO UHDFK D FOHDYDJH SODWHDX EHORZ b FOHDYHG DW WKH HQG RI VXFFHVVLYH URXQGV RI VHOHFWLRQ DQG WKH SRRO ZLOO EH HQULFKHG IRU RQO\ WKH IDVWHVW FDWDO\VWV RI WKH LQLWLDO OLEUDU\ f ,I WKH VHOHFWLRQ WLPH LV ODUJH UHODWLYH WR WKH KDOIOLIH RI WKH IDVWHVW FDWDO\VWV LQ D OLEUDU\ WKH VWDELOL]HG SRSXODWLRQ ZLOO FRQWDLQ D GLVWULEXWLRQ RI FDWDO\WLF UDWHV WKDW LV SURSRUWLRQDO WR WKH GLVWULEXWLRQ RI WKH LQLWLDO OLEUDU\

PAGE 165

URXQG URXQG /012$%&' ()* P r mc) P6 cL Y Y Y Y a Y G k 4 4 4 4 4 2 4 ] ] ] ] ] ] ] 1 f§ 2 LU UM [ U & WQFLPQQGR T D D D D D R ] ] ] ] ] ] ] IW URXQG $%&' ()*+ URXQG A A A A 2 I1 22 1 UI 7I UW : URXQG URXQG $ %& '()*+ -./012D %&'()*+ -./012 K } P LV 7WTUTT1URUnTTRTTUV f§ UfTUATLaTTRTTLaT m7WLGLQGLQULQUcQQLGGLQn6LQLULLGLQGRMPGnLn6QGG URXQG URXQG $%&'()*+ -./012 $%&'()*+ -./ 012 A A A P}OILLcUfcAIAI!A f§ G 2n  f§ 2n : }I/ R Yc k GGULUnUnLnGRRGLUOnWVL7WGG GR!G1PLQL GGG )LJXUH &OHDYDJH DW WKH FRPSOHWLRQ RI HDFK URXQG RI VHOHFWLRQ WLPH YDULHV DFFRUGLQJ WR VHOHFWLRQ FRQGLWLRQf 5RXQG UHIHUV WR WKH LQLWLDO OLEUDU\ EHIRUH DQ\ URXQGV RI VHOHFWLRQ $OO VHOHFWLRQV ZHUH H[SRVHG WR ODVHU DW WKH EHJLQQLQJ RI HDFK URXQG RI VHOHFWLRQ 7KH VXUYLYRUV RI HDFK URXQG ZHUH JHOSXULILHG DQG DPSOLILHG IRU WKH QH[W URXQGRI VHOHFWLRQ 7KH DPRXQW FOHDYHG IRU VDPSOHV ZLWK FDJHGULERVH $) O0f LQFUHDVHG SURJUHVVLYHO\ ZLWK HDFK URXQG 6HOHFWLRQV 1 DQG 2 QHYHU LQFUHDVHG 6HOHFWLRQ + LQFUHDVHG GUDPDWLFDOO\ IROORZLQJ WKH IRXUWK URXQG RI VHOHFWLRQ 7KH SHUFHQW FOHDYHG DW WKH FRPSOHWLRQ RI HDFK URXQG LV VKRZQ EHORZ HDFK VDPSOH 5HIHU WR 7DEOH SDJH f IRU D D GHVFULSWLRQ RI HDFK VHOHFWLRQ FRQGLWLRQ

PAGE 166

$ 5 % 5 & 5 ( 5 ) 5 5 f >, 0O OOOOLOOOOOOOOO} + 5 5H -5r ./ 5 05 1 5 5 f r ,00,,0,,P, )LJXUH &OHDYDJH DVVD\V IRU DOO VHOHFWLRQV FRQGLWLRQV DIWHU URXQGV RI VHOHFWLRQ UHYHDO WKDW RQO\ VHOHFWLRQ + KDV DEXQGDQW FDWDO\VWV 3RROV IURP DOO VHOHFWLRQ FRQGLWLRQV ZHUH SUHSDUHG ZLWK ULERVH$ SULPHU HLWKHU f%ULERVHf RU fDQJ ULERVHff &OHDYDJH DVVD\V ZHUH LQLWLDWHG E\ DGGLQJ ; UHDFWLRQ EXIIHU DQG WKH VORZ FRRO SURWRFRO UDWKHU WKDQ ODVHU LQLWLDWLRQf 2QO\ VHOHFWLRQ FRQGLWLRQ + KDG DQ\ GHWHFWDEOH FOHDYDJH IROORZLQJ KRXUV 7LPH SRLQWV DUH DQG KRXUV

PAGE 167

)LJXUH 6HOHFWLRQ + VKRZV VLPLODU FOHDYDJH UDWHV ZLWK ULERVH RU FDJHGULERVH SULPHU 3RROV RI FDWDO\VWV ZHUH JHQHUDWHG XVLQJ HLWKHU ULERVHDGHQRVLQH RU FDJHG DGHQRVLQH IURP WKH VXUYLYRUV RI 6HOHFWLRQ + IROORZLQJ URXQG DQG 6DPSOHV FRQWDLQLQJ WKH FDJHGDGHQRVLQH ZHUH LQLWLDWHG ZLWK ODVHU

PAGE 168

D r F $ / } R8 KRXUV % + 5 Q0 +5 Q0 +5 Q0 +5 Q0 +5 Q0 + 5 Q0 + 5 Q0 +5 Q0 R +5 Q0 R +5Q0 + 5 Q0 f +5 Q0 +5 Q0 f +5 Q0 f + ): Q0 f +5 Q0 +)} Q0 } + 5 Q0 + UD} Q0 )LJXUH .LQHWLF SURILOHV IRU 6HOHFWLRQ + DW YDULRXV FRQFHQWUDWLRQV IROORZLQJ HDFK URXQG RI VHOHFWLRQ VKRZV GHWHFWDEOH FOHDYDJH IROORZLQJ 5RXQG RI VHOHFWLRQ '1$ IURP WKH VXUYLYRUV RI HDFK URXQG RI 6HOHFWLRQ + ZDV UHDPSOLILHG IURP VWRFNV DQG SURJUHVV FXUYHV IRU FOHDYDJH DVVD\HG DW DQG Q0

PAGE 169

$ +5 Q0 + 5 Q0 + 5 Q0 + 5. Q0 r +5} Q0 r +5r Q0 F f +)} Q0 +51Q0 f + 5. Q0 )LJXUH 7KH LQLWLDO UDWH RI FOHDYDJH IRU 6HOHFWLRQ + IROORZLQJ URXQGV LQFUHDVHV VOLJKWO\ ZLWK LQFUHDVHG FRQFHQWUDWLRQV RI '1$ '1$ IURP WKH VXUYLYRUV RI HDFK URXQG RI 6HOHFWLRQ + ZDV UHDPSOLILHG IURP VWRFNV DQG SURJUHVV FXUYHV IRU FOHDYDJH DVVD\HG DW DQG Q0

PAGE 170

$ WHPSHUDWXUH &f )LJXUH 7KH FOHDYDJH UDWH IRU 6HOHFWLRQ + IROORZLQJ URXQGV RI VHOHFWLRQ LV XQFKDQJHG E\ LQFXEDWLQJ DW r& EXW LV VOLJKWO\ ORZHU DW WHPSHUDWXUHV EHORZ r& 7KH UDWH GURSV TXLFNO\ ZKHQ WKH WHPSHUDWXUH LV LQFUHDVHG DERYH r& $f )XOO WLPH FRXUVH %f 3ORW RI LQLWLDO UDWH YV WHPSHUDWXUH

PAGE 171

‘ +5 SRRO Q0f FOHDYDJH LQ SUHVHQFH RI Q0 USULPHU D +5 SRRO Q0f FOHDYDJH LQ SUHVHQFH RI Q0 Q0 USULPHU D USULPHU Q0f FOHDYHG E\ +5 SRRO Q0f R USULPHU Q0f FOHDYHG E\ +5 SRRO Q0f f + 5 VHOIFOHDYDJH Q0f ’ +5 VHOI FOHDYDJH Q0f )LJXUH 6HOHFWLRQ + IROORZLQJ URXQG FOHDYHV USULPHU LQ WUDQV 6HOHFWLRQ + Q0f IROORZLQJ URXQG ZDV LQFXEDWHG ZLWK HLWKHU Q0 RU Q0 USULPHU &OHDYDJH RI ERWK 6HOHFWLRQ + '1$]\PHV DQG WKH USULPHU VXEVWUDWH ZDV GHWHUPLQHG

PAGE 172

$ r + m, R e f R 9 t ‘ +50J D + 5 QR PHWDO r +5=Q r KRXUV % L 7 7 + 5 SRRO FOHDYDJH cQ M YDUL XV EXIIHUV +(3(6 1D&, 0J&, D +(3(61D&,0J&, } +(3(6 1D&,0J&, A +(3(6 1D&, 0J&, '(3& f +(3(61D&,0J&, ('7$ ’ +(3(61D&,0J&, $ $ ‘r ‘ KRXUV )LJXUH &OHDYDJH RI 6HOHFWLRQ + SRRO '1$]\PHV IROORZLQJ URXQG LV PHWDO LQGHSHQGHQW $f 6HOHFWLRQ + FOHDYDJH ZDV H[DPLQHG LQ 0 1D&, DQG P0 +(3(6 S+ SOXV HLWKHU =Q 0J RU QR GLYDOHQW PHWDO %f 6HOHFWLRQ + FOHDYDJH ZDV H[DPLQHG LQ P0 +(3(6 EXIIHU S+ SOXV 0 1D&, P0 0J& S/ '(3& SHU P/ EXIIHU RU P0 ('7$

PAGE 173

)DPLO\ D RU D D D D D D D 3 3 3 3 3 3 3 3 3 3 3 3 < < < < < < < < < < < < < < f +U +U +U +U +U +U +U +U +U +U +U +U +U +U +U +U +U +U +U +U +U +U +U +U +U +U +U +U +U +U +U +U +U +U +U +U *&*7*7$&77**$*$$&7**&*&$77$&7$&*7&$7*&7&* *&*7*7$&77**$*$$&7**&*&$77$&7$&*7&$7*&7&* *&*7*7$&77**$*$$&7**&*&$77$&7$&*7&$7*&7&* *&*7*7$&77**$*$$&7**&*&$77$&7$&*7&$7*&7&* *&*7*7$&77**$*$$&7$*&*&$7$77$&*7&$7*&7&* *&*7*7$&77**$*$$&7**&*&$77$&7$&*7&$7*&7&* *&*7*7$&77**$*$$&7**&*&$77$&7$&*7&$7*&77* *&*7*7$&77**$*$$&7**&*&$77$&7$&*7&$7*&7&* *&*7*7$&77**$*$$&7**&*&$77$&7$&*7&$7*&7&* f &$7&$7*&*$*&$&$& *& &$7&$7*&*$*&$&$&*& &$7&$7*&*$*&$&$&*& ‘ &$7&$7*&*$*& $&$& *& &$7&$7*&*$*&$&$&*& ‘ &$7&$7*&*$*& $& $& *&7 &$7&$7*&*$*&*&$7*& &$7&$7*&*$*&$&$&*& f & $7&$7*&* $*&$& $& *& &$*&$7*&*$*&$&$&*& &$7&$7*&*$*&$&$&$& 7*$ 7*$ 77*$ L$7*$ $7*$ 7*$ 7*$ 77*$O $7*$_PAA%_ &$7& $7*&*$*&$&$& *& $$7$*&& 7*$7$$**77**7$* DFWJFDFDDWFFDDFDFFJDWWJFWJFDDDJJWWJWWDJJJ +U +U +U +U +U +U +U +U +U +U +U +U +U +U +U +U +U +U +U +U +U +U +U +U +U +U +U f $ &$7*$ *&7**$7*&$**7&7*&$$&$&&***$&*&*7*7$ 7 77**7& *$7$&*77**7$77*$&$&7**7*77*777&7$* f *f§$&*&*$&&7$7*7&&$&&7&$**$&$7**7*&*&$&**** *&*$&&$$$7*$** *7&&**$&7**$& &&7 7&**7$&*$** ff§7$&$&*$&*7*&$&777**7$*&7$*7&&7*7*77***7** f&$7&&7*7**&7$*$&$$&&77*$**7*7&*&*7*&7&$7&$ $ &**7*$7$*&77*&7*7***$77&&&$7 *$*&&7*7&$$$ $ &7$7&7$$7&&**7**7&$$$7$&$&7&*7$*$&&*7* 7* & 77$*&*7*77&7&$7$$$**&7* $*&$7 &7&$&&77*& &* *$&&&7**$$&$**&$77777*$*&&&77&&&7*&&*77$*7*$&* *$7$&&&&7&7*$7&&&&&7*7&**&7***77*$7&$77** 7 7&**&$$7&***7*&$&$7&$7$77**&&**&7*$*&$7* $ 7*7&&&$*777&7777**$7$**$&*$7&*$$$$$*777* & &$7&&&*777*$7$7&7&&&7$&&7$*7$7&7*&77$**** * &$*7$7&*&*$$**&$7777777&$77$$*&7$$**777* 7 &**77$$$*777&& &77*$777$77$**&&$*7**7*** & 7 77*$&77*&&&$&7$*$7&&**77*$77$***&&$*$& 7 $$7$&&$&$&7&$*7$$7$4&&*7*7&&7$77f§77$&7*7* & $&*$** 7&$*$77&7$*7&****7$777*7*&&&7$&77* & *$&7&7**7777$$$$&***$$$*$7&7$*&&*$&*7&$* 7$$$&*&&$ *& &7*7$*&$* 7$*& *$7**7&**7$7$$$* 7$77$&&&7&$$*7 &$$77$77$7&*& $*7 *7 7&&777*7* 77*$$&*$&***&&*&7$*&$777$77$*7***7&&*&**7* *7&$$$$*$$&*&77&*$$$7&&&&77&7&&&&&&$*7&* & $&*$** 7&$*$77&7$*7&****7$777*7*&&&7$&77* 7$*77$ 7777**&$&**777$****7&7&7&7$77&&*7* 7 *7*&7$ **7*77&7&7*$* &&$*$&*77$*7*7$*77$$* **$$ 777$$*$$&7$$$&$& *& $77**7*$$$*7*7*&*f§ &7*$*7$$7*7*$$*$7$&$7*&&$7*$7*$&7&* *&&&$ ** $& 7$**$$$$7& $$77&&f§*77 $$$*7**&** $$77* $&* 7*7*7*777*&7$&7&*&$$&&&&7$*7$7&*$7*7&&7$7* &$&*$**7&$*$77&7$*7&****7$77 7*7*& &&7$&7f§7* $*7$&* &$*$7**7***$7$777$77f§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f, 96 +ff 5DWH FRQVWDQWV rKU nf 6HTXHQFHV DQG DUH IURP D SUHYLRXV ,96 2V .!

PAGE 174

3RSXODWLRQ 'LVWULEXWLRQ $IWHU 5281' 5281' 5281' 5281' 5281' 5281' 5281' )LJXUH ,QFUHDVLQJ WKH DPRXQW RI OHDNDJH GHOD\V WKH LVRODWLRQ RI FDWDO\WLF PROHFXOHV ,Q YLWUR VHOHFWLRQV ZHUH VLPXODWHG XVLQJ D VHOHFWLRQ WLPH 6Wf RI PLQXWHV IRU D SRSXODWLRQ RI FDWDO\WLF PROHFXOHV ZLWK LQLWLDO DEXQGDQFH VHW WR RQH DFWLYH PROHFXOH SHU f LQDFWLYH PROHFXOHV 5) ff 7KH DFWLYH PROHFXOHV ZHUH GLVFUHWHO\ GLVWULEXWHG RYHU D UDQJH RI LQLWLDO UDWHV DFFRUGLQJ WR D OLQHDU UHODWLRQVKLS ZLWK P HDFK GLVFUHWH UDWH FDWHJRU\ KDG WKH VDPH QXPEHU RI PROHFXOHV LQ WKH LQLWLDO SRSXODWLRQf 6LPXODWLRQV UHYHDO WKDW YDU\LQJ WKH DPRXQW RI OHDNDJH FOHDYDJH LQGHSHQGHQW RI LQWULQVLF UDWHf IURP WR b GUDPDWLFDOO\ GHOD\V LVRODWLRQ RI PROHFXOHV 1RQHWKHOHVV UHSHDWHG VHOHFWLRQ HYHQWXDOO\ OHDGV WR QHDUO\ LGHQWLFDO VHOHFWHG SRSXODWLRQV LQ WKLV FDVH E\ URXQG f UHJDUGOHVV RI WKH DPRXQW RI OHDNDJH 'DVKHG YHUWLFDO OLQH FRUUHVSRQGV WR WKH VHOHFWLRQ WLPH $f ,QLWLDO SRSXODWLRQ GLVWULEXWLRQ %f 3HUFHQW FOHDYHG DW WKH HQG RI HDFK URXQG RI VHOHFWLRQ &f 3HUFHQW FOHDYHG DW W PLQXWHV 'f 3RSXODWLRQ IROORZLQJ HDFK URXQG RI VHOHFWLRQ (f 3URJUHVV FXUYH IROORZ HDFK URXQG RI VHOHFWLRQ 2Q

PAGE 175

3RSXODWLRQ 'LVWULEXWLRQ $IWHU 5281' 5281' 5281' 5281' 5281' 5281' 5281' )LJXUH 9DU\LQJ WKH VHOHFWLRQ WLPH DOWHUV WKH UDQJH RI UDWHV VXUYLYLQJ VHOHFWLRQ 6HOHFWLRQV ZHUH VLPXODWHG XQGHU YDU\LQJ VHOHFWLRQ WLPHV 6W RU PLQXWHVf ZKLOH KROGLQJ RWKHU SDUDPHWHUV RI WKH PRGHO WKH VDPH SDUDPHWHUV VKRZQ LQ ILJXUHf %f ,QFUHDVLQJ WKH VHOHFWLRQ WLPH IURP PLQXWHV WR LQLWLDOO\ UHVXOWHG LQ D ODUJHU IUDFWLRQ RI WKH SRSXODWLRQ EHLQJ FOHDYHG DW WKH HQG RI HDFK URXQG RI VHOHFWLRQ FRUUHVSRQGLQJ WR HDUOLHU LQLWLDO GHWHFWLRQ ,QFUHDVLQJ WKH VHOHFWLRQ WLPH IXUWKHU WR PLQXWHV UHYHUVHG WKLV WUHQG 7KH IUDFWLRQ RI WKH SRSXODWLRQ FOHDYHG DW WKH HQG RI HDFK URXQG UHDFKHV D SODWHDX EHJLQQLQJ DIWHU URXQG LQGLFDWLQJ WKDW WKH VHOHFWLRQV KDYH UHDFKHG VWDELOL]HG SRSXODWLRQV :KHQ WKLV FOHDYDJH SODWHDX LV EHORZ b FDVH SDQHO %f WKH VWDELOL]HG SRSXODWLRQ FRQWDLQV RQO\ WKH IDVWHVW FDWDO\VWV FDVH SDQHO &f %HFDXVH WKHUH LV QR VLJQLILFDQW GLIIHUHQWLDO HQULFKPHQW IRU FDWDO\VWV ZLWK KDOIOLIH UDWHf IROG ORZHU WKDQ WKH VHOHFWLRQ WLPH ORQJHU VHOHFWLRQ WLPHV UHVXOW LQ FOHDYDJH SURILOHV RI b DQG HQULFK HTXDOO\ DOO PROHFXOHV ZLWK D KDOIOLIH UDWHf DSSUR[LPDWHO\ IROG ORZHU WKDQ WKH VHOHFWLRQ WLPH FDVH DQG SDQHO &f 1R FDWDO\VWV ZLWK KDOIOLYHV JUHDWHU ORZHU UDWH VORZHU FDWDO\VWf WKDQ WKH VHOHFWLRQ WLPH VXUYLYH PXOWLSOH URXQGV RI VHOHFWLRQ 'DVKHG YHUWLFDO OLQHV UHSUHVHQW UDWH FRUUHVSRQGLQJ WR WKH WKUHH GLIIHUHQW VHOHFWLRQ WLPHV

PAGE 176

5281' 5281' 5281' 5281' 5281' 5281' 5281' 3URJUHVV &XUYH IRU 1HZ 3RSXODWLRQ $IWHU WKUHH WHVW FDVHV YDU\LQJ LQ WKH UDWH RI WKH IDVWHVW FDWDO\VWV NPD[Q RU PLQf LQ WKH LQLWLDO SRSXODWLRQV ZKLOH KROGLQJ RWKHU SDUDPHWHUV RI WKH PRGHO FRQVWDQW SDUDPHWHUV VKRZQ LQ ILJXUHf 7KH QXPEHU RI FDWDO\VWV ZLWK UDWHV VORZHU WKDQ WKH VHOHFWLRQ WLPH ZDV KHOG FRQVWDQW DW ( IRU DOO SRSXODWLRQV ZKLOH WKH GLVWULEXWLRQ RI FDWDO\VWV ZLWK UDWHV 6W YDULHG IRU HDFK WHVW FDVH 7KH FDWDO\VWV IRU FDVH ZHUH GLVWULEXWHG IRU UDWHV EHWZHHQ 6W DQG NPD[Of DFFRUGLQJ WR D OLQHDU IXQFWLRQ ZLWK VORSH RI ( 7KH FDWDO\VWV IRU FDVH ZHUH GLVWULEXWHG IRU UDWHV EHWZHHQ 6W DQG NPD[f DFFRUGLQJ WR WKH SURSRUWLRQDO FKDQJHV LQ FDWDO\VWV LQ FDVH DQG KROGLQJ WKH UHDFWLYH IUDFWLRQ WKH VDPH (f $OO RI WKH FDWDO\VWV IRU FDVH ZHUH GLVWULEXWHG LQWR D VLQJOH UDWH FODVV ZLWK UDWH NPD[ DQG UHDFWLYH IUDFWLRQ HTXDO WR ( %HFDXVH VHOHFWLRQV HQULFK IRU PROHFXOHV ZLWK KDOI OLYHV IROG OHVV WKDQ VHOHFWLRQ WLPH LI WKH KDOIOLIH RI WKH IDVWHVW FDWDO\VW LV LQFUHDVHG WR ZLWKLQ IROG RI WKH VHOHFWLRQ WLPH RQO\ WKHVH IDVWHVW PROHFXOHV VXUYLYH 2Q

PAGE 177

'LVWULEXWLRQ /LQHDU P r YDQHV 5) YDULHV / r b 6W +((' PLQ 5) ‘ P}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r (f EXW FDVH KDG WLPHV PRUH UHDFWLYH PROHFXOHV LQ LWV LQLWLDO SRSXODWLRQ 1RW RQO\ GLG WKH VHOHFWLRQV IRU FDVH UHYHDO FDWDO\VWV HDUOLHU %f EXW WKH GLVWULEXWLRQ RI FDWDO\VWV LQ WKH VWDEDOL]HG SRSXODWLRQV GLIIHUHG IURP WKH GLVWULEXWLRQ IRU FDVH &DVH GLIIHUHG IURP FDVH ,Q ERWK WKH QXPEHU RI UHDFWLYH FDWDO\VWV 5) (r YV (f DQG LQ WKH GLVWULEXWLRQ RI WKRVH FDWDO\VWV P ( YV (f EXW WKH UDWLR SURSRUWLRQDOLW\f RI WKH DPRXQW RI FDWDO\VWV ZLWK UDWH NQf WR FDWDO\VWV ZLWK UDWH NQf ZKHUH Q HTXDOV DQ\ JLYHQ UDWHf ZDV KHOG FRQVWDQW IRU D OLQHDU GLVWULEXWLRQ WKH SURSRUWLRQDOLW\ UHODWLRQVKLS LV P PO f5)5)f &DVH UHDFKHG D VWDEDOL]HG SRSXODWLRQ HDUOLHU WKDQ FDVH RQ DFFRXQW RI D IROG JUHDWHU UHDFWLYH IUDFWLRQ EXW RQFH VWDEDOL]HG ERWK FDVH DQG KDG LGHQWLFDO GLVWULEXWLRQV RI FDWDO\WLF UDWHV 21 2Q

PAGE 178

5281' 5281' 5281' 3URJUHVV &XUYH IRU 1HZ 3RSXODWLRQ $IWHU 5281' 5281' )LJXUH 9DULRXV LQLWLDO GLVWULEXWLRQV \LHOG GLIIHUHQW VWDEDOL]HG SRSXODWLRQV IROORZLQJ VHOHFWLRQV 7KH SHDN IRU WKH VWDELOL]HG SRSXODWLRQ GLVWULEXWLRQ LV VKLIWHG FORVHU WR WKH 6W ZKHQ WKH SURSRUWLRQDO DEXQGDQFH RI VORZHU FDWDO\VWV LV JUHDHU LQ FRPSDULVRQ WR IDVWHU FDWDO\VWV

PAGE 179

$ % & )LJXUH 7KH XSSHU URZ VKRZV WKH SRSXODWLRQ GLVWULEXWLRQ IROORZLQJ VXFFHVVLYH URXQGV RI VHOHFWLRQ IRU FDWDO\VWV LQLWLDOO\ GLVWULEXWHG DFFRUGLQJ WR $f OLQHDU P %f OLQHDU P [( &f H[SRQHQWLDO E 'f H[SRQHQWLDO E 7KH ORZHU SDQHO FRPSDUHV WKH SRSXODWLRQ GLVWULEXWLRQ EHIRUH VHOHFWLRQ IROORZLQJ WKH VWDELOL]HG SRSXODWLRQ GLVWULEXWLRQ IROORZLQJ VHOHFWLRQ 7KH SRSXODWLRQ GLVWULEXWLRQ IRU FDWDO\VWV IROORZLQJ LQFUHDVLQJ URXQGV RI VHOHFWLRQV HYHQWXDOO\ UHDFKHV D VWDELOL]HGSRSXODWLRQ EH\RQG ZKLFK IXUWKHU URXQGV RI VHOHFWLRQ GR QRW VLJQLILFDQWO\ DOWHU WKH GLVWULEXWLRQ RI FDWDO\VWV XSSHU SDQHOf 7KH SURSRUWLRQDOLW\ RI FDWDO\VWV ZLWK KDOI OLYHV IROG ORZHU WKDQ 6W UHPDLQV QHDUO\ LGHQWLFDO WR WKHLU SURSRUWLRQDOLW\ LQ WKH LQLWLDO SRSXODWLRQ ORZHU SDQHOVf 7KH SUR[LPLW\ RI WKH VWDELOL]HG GLVWULEXWLRQ PD[LPD WR WKH VHOHFWLRQ WLPH LV UHODWHG WR WKH LQLWLDO GLVWULEXWLRQ ELDV IRU FDWDO\VWV ZLWK KDOI OLYHV EHWZHHQ r6W DQG 6W 2Q 22

PAGE 180

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f LQ ERWK NLQHWLF DQDO\VLV DQG DQ ,96 6LPXODWLRQV RI ,96 ZHUH DOVR SURGXFHG DQG

PAGE 181

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f ([DPLQDWLRQ RI ERWK WKH KDLUSLQ ULER]\PH DQG KHSDWLWLV GHOWD YLUXV ULER]\PH KDV UHYHDOHG WKDW 0Ja SOD\V D RQO\ SDVVLYH UROH OLNHO\ VWUXFWXUDOf DQG LQVWHDG LPSOLFDWH DQ LQWHUQDO QXFOHRWLGH LQ JHQHUDO DFLGEDVH FDWDO\VLV
PAGE 182

WKH UHDFWLYH IROGHG IRUP LV JUHDWHU WKDQ WKH LPSDFW RQ WKH UDWH FRQVWDQW IRU WKH FKHPLFDO VWHS 7KH IDFW WKDW WKH EDODQFH EHWZHHQ XQL DQG ELPROHFXODU LQWHUDFWLRQ LV FRQFHQWUDWLRQ GHSHQGHQW PD\ H[SODLQ WKH REVHUYDWLRQ E\ *H\HU DQG 6HQ f DQG %UHDNHU DQG -R\FH f WKDW FOHDYDJH UDWHV GHFUHDVHG ZKHQ '1$]\PHV SURGXFHG ZLWK DGGLWLRQDO F\FOHV RI 3&5 YV f ZHUH H[DPLQHG 3UHVXPDEO\ EHFDXVH WKH\ FRQVLGHUHG RQO\ XQLPROHFXODU LQWHUDFWLRQV WKHVH UHVHDUFKHUV GLG QRW UHSRUW WKH FRQFHQWUDWLRQV RI '1$]\PHV LQ WKHLU DVVD\V DQG SUHVXPDEO\ WKH\ GLG QRW FRQWURO '1$]\PH FRQFHQWUDWLRQf :H SUHGLFW WKHUHIRUH WKDW WKH GHFUHDVH LQ UDWH ZLWK LQFUHDVLQJ F\FOHV RI 3&5 UHVXOWHG IURP DQ LQFUHDVHG FRQFHQWUDWLRQ RI WKHLU '1$]\PHV ,I NEL NXQL IRU WKHLU SDUWLFXODU '1$]\PHV KLJKHU FRQFHQWUDWLRQV ZRXOG IDYRU D VORZHU UDWH SURFHVV $OWKRXJK GLYDOHQW FDWLRQV FDQ DIIHFW WKH IROGLQJ SURSHUWLHV RI QXFOHLF DFLGV FRIDFWRUV VXFK DV 0Jf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f WR LPSURYH WKH UDWH RI FDWDO\VLV WKH UDWH OLPLWLQJ VWHS PXVW QRW EH WKH IROGLQJ RI WKH '1$]\PH EXW LQVWHDG WKH FKHPLFDO VWHS 7KH LPSDFW RI WKH FKHPLFDO VWHS FDQ EH LQFUHDVHG UHODWLYH WR WKH IROGLQJ VWHS E\ LQFUHDVLQJ WKH FRIDFWRU FRQFHQWUDWLRQ RU LQFUHDVLQJ WKH ELQGLQJ VWUHQJWK RI (}6 RU (rFRIDFWRU

PAGE 183

(Q]\PH IROGLQJ LV SDUWLFXODUO\ SUREOHPDWLF IRU QXFOHLF DFLG HQ]\PHV EHFDXVH IROGLQJ UHTXLUHV SHUSHQGLFXODU fIDFHWRIDFHf LQWHUDFWLRQV PRUH FRQVWUDLQHGf DQG DOWHUQDWLYH FRQIRUPDWLRQV DUH UHODWLYHO\ FRPPRQ DV VWUXFWXUH SUHGLFWLRQV IRU UHYHDOf 1XFOHLF DFLGSRO\PHU IROGLQJ ZLWK VWDQGDUG QRQIXQFWLRQDOL]H QXFOHRWLGHVf PD\ EH D VORZ VHDUFK IRU RSWLPDO DFWLYHf VWUXFWXUHV IXOO RI GHDGHQG LQDFWLYHf WUDSV 7KLV LV LQ FRQWUDVW WR WKH IXQQHOVKDSHG IROGLQJ ODQGVFDSH SUHGLFWHG IRU SURWHLQV 2]NDQ f UHFRJQL]LQJ WKDW QDWXUDO SURWHLQV PD\ KDYH HYROYHG LQWR UHJLRQV RI VHTXHQFH VSDFH WKDW KDYH WKLV SURSHUW\f )XQFWLRQDOLWLHV DGGHG WR QXFOHLF DFLGV WKDW LPSURYH IROGLQJ VWUHQJWK DQG LPSURYH VHDUFK SURSHUWLHV PD\ EH FULWLFDO 6WDQOH\ 0LOOHU UHFHQWO\ SURSRVHG WKDW SUHELRWLF V\QWKHVLV RI QXFOHLF DFLGV LV HQKDQFHG LQ FROG HQYLURQPHQWV DQG SRVWXODWHG WKLV PD\ EH HYLGHQFH IRU FROG RULJLQV RI VLQJOHELRSRO\PHU OLIHIRUP 0L\DZDND f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f DFW RQ YDULRXV GLIIHUHQW VXEVWUDWHV 6WXGLHV RI WKH LPSDFW RI FRPSDUWLWLRQ RQ HYROXWLRQ KDYH GHPRQVWUDWHG WKDW VRPH PHFKDQLVP PXVW H[LVW WR NHHS VHOILVK PROHFXOHV IURP RYHUZKHOPLQJ FULWLFDO IXQFWLRQV 0DWVXXUD f 2QH PHWKRG RI DFKLHYLQJ FRPSDUWLWLRQ UHOLHV RQ FHOOXODU PHPEUDQHV ,W LV SODXVLEOH KRZHYHU WKDW EHIRUH K\SHUF\FOHV ZHUH HQFORVHG LQ PHPEUDQHV FDWDO\WLF

PAGE 184

QXFOHLF DFLGV KDG WR UHO\ RQ VHTXHQFH WDJV WR GLVWLQJXLVK PROHFXOHV ZLWKLQ WKH K\SHUF\FOH IURP VHOILVK FRPSHWLWRUf PROHFXOHV RXWVLGH WKH K\SHUF\FOH :LWKRXW RYHUW VHOHFWLRQ SRVVHVVHG TXDOLWLHV WKDW DOORZ IRU WKLV W\SH RI GLVWLQFW UHJXODWLRQ DFWLQJ GLIIHUHQWLDOO\ LQ FLV DQG WUDQV HYHQ IRU FORVHO\ UHODWHG PROHFXOHVf DW YDULRXV FRQFHQWUDWLRQV 7KHVH FRPSOH[ SDWWHUQV RI UHJXODWLRQ DUH SRVVLEOH EHFDXVH RI WKH JHQHUDOLW\ RI EDVH SDLULQJ DQG WKXV DSSHDUV WR EH LQWULQVLF DQG DEXQGDQW WR QXFOHLF DFLGV DQG WKHUHIRUH GR QRW QHHG WR EH fHQJLQHHUHGff 7KH FRPSOH[ 51$EDVHG UHJXODWRU\ PHFKDQLVPV QRZ DSSUHFLDWHG ZLWKLQ PRGHP RUJDQLVPV PD\ EH DQDORJRXV WR WKH FRPSOH[ UHJXODWLRQ DFKLHYHG E\ WKH VLQJOH ELRSRO\PHU SURJHQLWRU &DJHGULERVH LV 8VHIXO IRU .LQHWLFV %XW ,QFUHDVHV /HDNDJH 7RR 0XFK IRU ,96 7KHUH DUH VHYHUDO SODFHV LQ WKH ,96 SURWRFRO ZKHUH FDWDO\VWV PD\ EH ORVW SULRU WR WKH VHOHFWLRQ VWHS :H H[DPLQHG WKH XWLOLW\ RI D FDJHGULERVH WR JXDUG DJDLQVW WKLV 7KH FDJHGULERVH SURYHG XVHIXO LQ SURWHFWLQJ WKH VDPSOH IURP fSUHPDWXUH FDWDO\VLVf \HW ZKHQ GHSURWHFWHG YLD SKRWRO\VLV SURGXFHG NLQHWLF UHVXOWV VLPLODU WR WKRVH JHQHUDWHG ZLWK XQSURWHFWHG ULERVH 8QIRUWXQDWHO\ WKH ODVHU SKRWRO\VLV JHQHUDWHV D KLJKHU DPRXQW RI EDFNJURXQG FOHDYDJH OHDNDJHf ZKLFK UHVXOWHG LQ WKH IDLOXUH WR HQULFK IRU DFWLYH FDWDO\VWV ZKHQ XVHG GXULQJ DQ ,96 6LPXODWLRQV RI ,96 UHYHDOHG WKDW HYHQ VOLJKW LQFUHDVHV LQ WKLV VHTXHQFHLQGHSHQGHQW FOHDYDJH SDUDPHWHU DQDORJRXV WR RWKHU IRUPV RI fOHDNDJHf LQKHUHQW LQ WKH FROXPQVHOHFWLRQ SURWRFROf VLJQLILFDQWO\ GHOD\ HQULFKPHQW RI FDWDO\VWV :H KDYH DOVR GHYHORSHG DOWHUQDWLYHV WR WKH VWDQGDUG %UHDNHU-R\FH PHWKRG IRU LVRODWLQJ FDWDO\WLF VWUDQG '1$]\PHV DQG IRU SHUIRUPLQJ VROLGSKDVH VHOHFWLRQV $V\PPHWULF 3&5 DQG H[RQXFOHDVH GHJUDGDWLRQ ZHUH HIIHFWLYH PHWKRGV IRU SXULI\LQJ WKH

PAGE 185

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fDQJ ULERVHf ZLWK VWDQGDUG QXFOHRWLGHVf VHOHFWLRQ IDOO LQWR VHYHUDO JURXSV Df f6LQJOHWRQVf ZKLFK DUH XQOLNH DQ\ RWKHU VHTXHQFH (DFK RI WKH VLQJOHWRQV WHVWHG KDG DQ H[WUHPHO\ VORZ UDWH RI FOHDYDJH LQ FRPSDULVRQ WR RWKHU FRPSHWLQJ PROHFXOHVf 7KH IDFW WKDW WKHVH PROHFXOHV SHUVLVW WKURXJK URXQGV RI VHOHFWLRQ GHVSLWH H[WUHPHO\ VORZ UDWHV RI FOHDYDJH VXJJHVWV WKDW WKH\ PD\ EH VXUYLYLQJ EHFDXVH WKH\ DUH JRRG VXEVWUDWHV IRU RWKHU FDWDO\VWV ,I WKLV VFHQDULR LV WUXH WKH\ PD\ EH SDUDVLWLF WKH\ GLUHFWO\ FRPSHWH ZLWK WKH DFWLYH FDWDO\VW IRU VXUYLYDOf RU V\PELRWLF WKH\ HQKDQFH WKH VHOIFOHDYDJH RI WKH HQ]\PH WKDW DOVR FOHDYHV WKHPf 6HOHFWLRQ LV H[SHFWHG WR FRQWLQXRXVO\ PLQLPL]H WKH SRSXODWLRQ RI SDUDVLWLF VXEVWUDWHV ,I WKHVH fVLQJOHWRQVf DUH LQ IDFW V\PELRWLF PROHFXOHV WKHQ WKH VXUYLYDO RI ERWK WKH HQ]\PH DQG VXEVWUDWH IROORZV D fNLQVHOHFWLRQ PRGHOf Ef )DPLO\ DOSKD PROHFXOHV ILUVW DSSHDU DIWHU 5RXQG RI VHOHFWLRQ DQG UHODWLYHV SHUVLVW LQ WKH SRSXODWLRQ IRU VHYHUDO URXQGV XQWLO WKH IDPLO\ DSSDUHQWO\ JRHV H[WLQFW DIWHU URXQG 2QH UHSUHVHQWDWLYH ZDV WHVWHG DQG LW KDG D UDWH RI FOHDYDJH QHDU ]HUR

PAGE 186

VXJJHVWLQJ WKDW OLNH WKH VLQJOHWRQV VXUYLYDO PD\ KDYH EHHQ EDVHG RQ EHLQJ D JRRG VXEVWUDWH UDWKHU WKDQ D JRRG FDWDO\VW Ff )DPLO\ EHWD JDPPD DQG GHOWD DUH DOO YHU\ VLPLODU WR SUHYLRXVO\ VHOHFWHG FDWDO\VWV EXW QHDUO\ DOO GLIIHU E\ DW OHDVW RQH QXFOHRWLGHf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f ODUJHU WKDQ WKH RULJLQDO VDPSOLQJ f :LWKLQ DQ RUGHU RI PDJQLWXGH ZH NQRZ RI DW OHDVW DFWLYH FDWDO\VWV IURP WKH OLEUDU\ LPSO\LQJ WKDW FDWDO\VWV DUH SUHVHQW QR OHVV WKDQ DQG FRXOG EH PXFK PRUH IUHTXHQW LI WKH OLEUDU\ LV VPDOOHU WKDQ f *LYHQ WKH DEXQGDQFH RI FDWDO\VWV GLIIHULQJ E\ RQO\ D IHZ QXFOHRWLGHV PHWKRGV RI HVWLPDWLQJ FDWDO\WLF DEXQGDQFH EDVHG RQ UHVWULFWLRQ GLJHVWLRQ %DUWHO f ZLOO JURVVO\ XQGHUHVWLPDWH WKH WUXH FDWDO\WLF DEXQGDQFH VLQFH WKH\ FDQ QRW GLVWLQJXLVK EHWZHHQ WKHVH VLPLODU FDWDO\VWV

PAGE 187

6LPXODWLRQV $OORZ (VWLPDWLRQ RI &DWDO\VW 'LVWULEXWLRQ 7KH LPSDFW RI YDULRXV VHOHFWLRQ SDUDPHWHUV DQG OLEUDU\ SDUDPHWHUV ZDV H[DPLQHG ZLWK VLPXODWLRQV RI WKH ,96 SURFHGXUH :KHQ WKH VHOHFWLRQ WLPH 6Wf LV YHU\ VKRUW FRPSDUHG WR WKH KDOIOLIH RI WKH IDVWHVW FDWDO\VWV LQ WKH LQLWLDO OLEUDU\ WKH VWDELOL]HG SRSXODWLRQ LV FOHDYHG OHVV WKDQ b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f 3HUIRUPLQJ VHOHFWLRQV ZLWK D UDQJH RI VHOHFWLRQ WLPHV DOORZV HVWLPDWLRQ RI ERWK IDVW DQG VORZ FDWDO\VW :KHQ WKH 6Wa NIDVWHVW RQO\ WKH IDVWHVW PROHFXOHV DUH LVRODWHG ,QFUHDVLQJ VHOHFWLRQ WLPH VXFK WKDW 6W} NIDVWHVW FUHDWHV D VWDELOL]HG SRSXODWLRQ VXFK WKDW WKH SURSRUWLRQDOLW\ EHWZHHQ IDVWHVW DQG VORZHU FDWDO\VWV DIWHU VHOHFWLRQ UHPDLQV VLPLODU WR WKH LQLWLDO SRSXODWLRQ 7KLV DOORZV HVWLPDWLRQ RI WKH LQLWLDO SRSXODWLRQ GLVWULEXWLRQ E\ GLUHFW H[DPLQDWLRQ RI WKH VHOHFWHG SRSXODWLRQ f 7KH SRSXODWLRQ RI FDWDO\VWV LQ LQLWLDO OLEUDU\ FDQ EH GHVFULEHG VXIILFLHQWO\ ZLWK WZR SDUDPHWHUV WKH 5HDFWLYH )UDFWLRQ 5)f DQG WKH 'LVWULEXWLRQ )XQFWLRQ 7KH

PAGE 188

OHDNDJH SDUDPHWHU DQG NIDVWHVW SDUDPHWHUV FDQ EH H[SHULPHQWDOO\ GHWHUPLQHG 7KH H[SHULPHQWDOO\ PHDVXUDEOH RXWFRPH RI VHOHFWLRQ H[SHULPHQWV SURJUHVV FXUYH IRU FOHDYDJH DW HDFK URXQG RI VHOHFWLRQf FDQ EH ILW WR WKH VLPXODWLRQ PRGHO LQ ZKLFK WKH 5) DQG GLVWULEXWLRQ IXQFWLRQ DUH YDULHG 6LPXODWLRQV KDYH UHYHDOHG WKDW PRGHO SDUDPHWHUV VXFK DV 5) DQG SRSXODWLRQ GLVWULEXWLRQ KDYH D VLJQLILFDQW LPSDFW RQ WKH RXWFRPH RI VHOHFWLRQ H[SHULPHQWV VXJJHVWLQJ WKDW UHODWLYHO\ VPDOO UDQJH RI SDUDPHWHUV VKRXOG ILW WKH H[SHULPHQWDO GDWD 1XFOHLF $FLG &DWDO\VLV DQG WKH 2ULJLQV RI /LIH *LYHQ WKH DSSDUHQW LQIUHTXHQF\ RI QXFOHLF DFLG HQ]\PHV LQ UDQGRP OLEUDULHV LV LW VWLOO UHDVRQDEOH WR DVVXPH WKDW OLIH RULJLQDWHG ZLWK QXFOHLF DFLGV DV WKH EDVLV RI D VLQJOHn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f 3HUKDSV QXFOHLF DFLGV DUH QRW MXVW OLPLWHG E\ WKH DEVHQFH RI WKH FKHPLFDO IXQFWLRQDOLW\ WKDW HQKDQFH FDWDO\VLV DV FRPPRQ LQ SURWHLQVf EXW DOVR E\ WKH SUHFLVH EDVHn SDLULQJ WKDW LV UHTXLUHG RI QXFOHLF DFLGV DV WKH JHQHWLF V\VWHP 7KH VWDQGDUG QXFOHLF DFLGV DUH WKH SURGXFW RI VWULQJHQW VHOHFWLRQ IRU SRO\PHUV WKDW IROG ZHOO LQ D OLQHDU IDVKLRQ EXW GR QRW IROG LQWR WKUHH GLPHQVLRQ VWUXFWXUHV &DWDO\WLF QXFOHLF DFLGV VXFK DV WKH ULERVRPH DQG

PAGE 189

VSOLFHRVRPH RIWHQ FRQWDLQ PRGLILHG QXFOHRWLGHV ZKLFK DORQJ ZLWK DVVRFLDWHG SURWHLQV FDQ LPSURYH VWUXFWXUDO DUUDQJHPHQW ,Q WKH IXWXUH WKH LQFRUSRUDWLRQ RI IXQFWLRQDOL]HG QXFOHRWLGHV WKDW HQDEOH FDWDO\VLV DQG IROGLQJ LQWR LQ YLWUR VHOHFWLRQV PD\ GUDPDWLFDOO\ LPSURYH WKH FDWDO\WLF SRWHQWLDO RI UDQGRP QXFOHLF DFLG OLEUDULHV

PAGE 190

5()(5(1&(6 $QJ 1 6XK % :HVWHUPDQQ&ODUN ( %DWWHUVE\ 7 %HQQHU 6 $ f $Q DOWHUQDWLYH WR WKH RULJLQV RI OLIH WKHRULHV $PLQR DFLGOLNH '1$ PROHFXOHV ZLWK FDWDO\WLF DFWLYLW\ )$6(% $ 6XSSO 6 $QJ 1 f $Q $OWHUQDWLYH WR WKH 2ULJLQV RI /LIH 7KHRULHV $PLQR $FLGOLNH '1$ 0ROHFXOHV &DSDEOH RI ,PSURYHG &DWDO\VLV 'LVVHUWDWLRQ 8QLYHUVLW\ RI )ORULGD *DLQHVYLOOH )/ %DUWHO 3 6]RVWDN -: f ,VRODWLRQ RI QHZ ULER]\PHV IURP D ODUJH SRRO RI UDQGRP VHTXHQFHV 6FLHQFH %DWWHUVE\ 7 5 $QJ 1 %XUJVWDOOHU 3 -XUF]\N 6 %RZVHU 0 7 %XFKDQDQ ' .HQQHG\ 5 7 %HQQHU 6 $ f ,Q YLWUR VHOHFWLRQ RI DQ DGHQRVLQH UHFHSWRU IURP D OLEUDU\ LQFRUSRUDWLQJ D FDWLRQLF QXFOHRWLGH DQDORJ $P &KHP 6RF %HDXGU\ $ $ -R\FH ) f 'LUHFWHG HYROXWLRQ RI DQ 51$ HQ]\PH 6FLHQFH %HQQHU 6 $ f 5HGHVLJQLQJ OLIH 2UJDQLF FKHPLVWU\ DQG WKH HYROXWLRQ RI SURWHLQV &KLPLD %HQQHU 6$ (OOLQJWRQ $' Df ,QWHUSUHWLQJ WKH EHKDYLRU RI HQ]\PHV 3XUSRVH RU SHGLJUHH" &5& &ULW 5HY %LRFKHP %HQQHU 6$ Ef 5HFRQVWUXFWLQJ WKH HYROXWLRQ RI SURWHLQV LQ 5HGHVLJQLQJ WKH 0ROHFXOHV RI /LIH %HQQHU 6$ HGLWRU 6SULQJHU9HUODJ +HLGHOEHUJ %HQQHU 6$ (OOLQJWRQ $' 7DXHU $ f 0RGHP PHWDEROLVP DV D SDOLPSVHVW RI WKH 51$ ZRUOG 3URF 1DW $FDG 6FL %HQQHU 6$ 6ZLW]HU &< f 6LPSOLFLW\ DQG &RPSOH[LW\ LQ 3URWHLQV DQG 1XFOHLF $FLGV &K (G E\ )UDXHQIHOGHU + 'HLVHQKRIHU :RO\QHV 3* 'DKOHP 8QLYHUVLW\ 3UHVV %UHDNHU 55 -R\FH *) f $ '1$ HQ]\PH WKDW FOHDYHV 51$ &KHP %LRO

PAGE 191

%UHDNHU 55 -R\FH *) f $ '1$ HQ]\PH ZLWK 0JGHSHQGHQW 51$ SKRVSKRGLHVWHUDVH DFWLYLW\ &KHP %LRO &HFK 75 =DXJ $*UDERZVNL 3f ,Q YLWUR VSOLFLQJ RI WKH ULERVRPDO 51$ SUHFXUVRU RI 7HWUDK\PHQD ,QYROYHPHQW RI D JXDQRVLQH QXFOHRWLGH LQ WKH H[FLVLRQ RI WKH LQWHUYHQLQJ VHTXHQFH&HOO &ULFN )+& f 7KH RULJLQ RI WKH JHQHWLF FRGH 0ROHF %LRO )XOOHU :' 6DQFKH] 5$ DQG 2UJHO /( f 0ROHF %LRO *H\HU &5 6HQ f (YLGHQFH IRU WKH PHWDOFRIDFWRU LQGHSHQGHQFH RI DQ 51$ SKRVSKRGLHVWHUFOHDYLQJ '1$ HQ]\PH &KHPLVWU\ DQG %LRORJ\ *DEHO 1: 3RQQDPSHUXPD & f 1DWXUH *XHUULHU7DNDGD & *DUGLQHU 0DUVK 7 3DFH 1 $OWPDQ 6 f 7KH 51$ PRLHW\ RI 51DVH 3 LV WKH FDWDO\WLF VXEXQLW RI WKH HQ]\PH&HOO *LOEHUW : f 7KH 51$ :RUOG 1DWXUH +DPSHO $ &RZDQ -$ f $ XQLTXH PHFKDQLVP IRU 51$ FDWDO\VLV WKH UROH RI PHWDO FRIDFWRUV LQ KDLUSLQ ULER]\PH FOHDYDJH &KHP %LRO +DQHO 5 &RQUDWK % )ODVDU )0 .XQGH 9 0DJXLU : 3HDUO 3LUUDJOLD 6DPXHOVRQ 5 +HUDWK / $OOLVRQ 0 &UXLNVKDQN *DXWLHU *LHUDVFK 3 +RUQ / .RSSDQ\ 5 DQG 3RQQDPSHUXPD & f ,QIUDUHG 2EVHUYDWLRQV RI WKH 6DWXUQLDQ 6\VWHP IURP 9R\DJHU 6FLHQFH +XHEQHU :) 6\QGHU /( %XKO f +&1 5DGLR (PLVVLRQ IURP &RSPHW .RKRXWHN f ,FDUXV ,UYLQH 7XHUN & *ROG / f 6HOH[LRQ 6\VWHPDWLF HYROXWLRQ RI OLJDQGV E\ H[SRQHQWLDO HQULFKPHQW ZLWK LQWHJUDWHG RSWLPL]DWLRQ E\ QRQOLQHDU DQDO\VLV 0RO %LRO ,UYLQH :0 f 7KH &RPSRVLWLRQ RI ,QWHUVWHOODU 0ROHFXODU &ORXGV 6SDFH 6FL 5HY -RKQVVRQ $OOHPDQQ 5 %HQQHU 6 $ f 'HVLJQHG HQ]\PHV 1HZ SHSWLGHV WKDW IROG LQ DTXHRXV VROXWLRQ DQG FDWDO\]H UHDFWLRQV LQ 0ROHFXODU 0HFKDQLVPV LQ %LRRUJDQLF 3URFHVVHV & %OHDVGDOH DQG % 7 *ROGLQJ HGV &DPEULGJH 5R\DO 6RFLHW\ RI &KHPLVWU\ -RKQVVRQ $OOHPDQQ 5. :LGPHU + %HQQHU 6$ f 6\QWKHVLV VWUXFWXUH DQG DFWLYLW\ RI DUWLILFLDO UDWLRQDOO\ GHVLJQHG FDWDO\WLF SRO\SHSWLGHV 1DWXUH

PAGE 192

-RKQVWRQ :. 8QUDX 3/DZUHQFH 06 *ODVQHU 0( %DUWHO '3 f 51$ FDWDO\]HG 51$ SRO\PHUL]DWLRQ $FFXUDWH DQG JHQHUDO 51$WHPSODWHG SULPHU H[WHQVLRQ 6FLHQFH -R\FH *) f 51$ HYROXWLRQ DQG WKH RULJLQV RI OLIH 1DWXUH -R\FH *) f )RUZDUG LQ 2ULJLQV RI /LIH 7KH &HQWUDO &RQFHSWV HG 'HDPHU ': DQG )OHLVFKDNHU *5 %R[WRQ -RQHV DQG %DUWOHWW -R\FH *) f 7KH DQWLTXLW\ RI 51$EDVHG HYROXWLRQ 1DWXUH .DXIIPDQ 6 f 7KH RULJLQV RI RUGHU VHOI RUJDQL]DWLRQ DQG VHOHFWLRQ LQ HYROXWLRQ 1HZ
PAGE 193

3HUULQ '0 *DUHVWLHU 7 +HOHQH & f ([SDQGLQJ WKH FDWDO\WLF UHSHUWRLUH RI QXFOHLF DFLG FDWDO\VWV 6LPXOWDQHRXV LQFRUSRUDWLRQ RI WZR PRGLILHG GHR[\ULERQXFOHRVLGH WULSKRVSKDWHV EHDULQJ DPPRQLXP DQG LPLGD]RO\O IXQFWLRQDOLWLHV 1FOHRV 1XFOHRW 3LFFLULOOL -$ .UDXFK 7 0RURQH\ 6( %HQQHU 6$ f ([WHQGLQJ WKH JHQHWLF DOSKDEHW (Q]\PDWLF LQFRUSRUDWLRQ RI D QHZ EDVH SDLU LQWR '1$ DQG .1$1DWXUH 5HLG & DQG 2UJHO /( f 1DWXUH 5LFK $ f 2Q WKH SUREOHPV RI HYROXWLRQ DQG ELRFKHPLFDO LQIRUPDWLRQ WUDQVIHU LQ +RUL]RQV LQ %LRFKHPLVWU\ .DVKD 0 DQG 3XOOPDQ % HGLWRUV 1< $FDGHPLF 3UHVV 5REHUWVRQ 0 DQG 0LOOHU 6 f 6FLHQFH 5RWK $ %UHDNHU 5 5 f $Q DPLQR DFLG DV D FRIDFWRU IRU D FDWDO\WLF SRO\QXFOHRWLGH 3URF 1DW $FDG 6FL 86$ 5XSHUW 3% 0DVVH\ $3 6LJXUGVVRQ 67 )HUUHf'f$PDUHf $5 f 7UDQVLWLRQ VWDWH VWDELOL]DWLRQ E\ D FDWDO\WLF 51$ 6FLHQFH 6DNWKLYHO %DUEDV &) f ([SDQGLQJ WKH SRWHQWLDO RI '1$ IRU ELQGLQJ DQG FDWDO\VLV +LJKO\ IXQFWLRQDOL]HG G873 GHULYDWLYHV WKDW DUH VXEVWUDWHV IRU WKHUPRVWDEOH '1$ SRO\PHUDVHV $QJHZ &KHP ,QW (GLW 6DQWRUR 6: -R\FH *) f $ JHQHUDO SXUSRVH 51$FOHDYLQJ '1$ HQ]\PH 31$6 6DQWRUR 6: -R\FH *) 6DNWKLYHO *UDPDWLNRYD 6 %DUEDV &) f 51$ FOHDYDJH E\ D '1$ HQ]\PH ZLWK H[WHQGHG FKHPLFDO IXQFWLRQDOLW\ -$0 &+(062& 6FOHVLQJHU DQG 0LOOHU 6/ f 3UHELRWLF 6\QWKHVLV LQ $WPRVSKHUHV &RQWDLQLQJ &+ &2 DQG & +\GRJHQ &\QDQLGH )RUPDOGHK\GH DQG $PPRQLD 0RO (YRO 6HHOLJ % DQG -DVFKNH $ f &KHP %LRO 6KLK ,+ %HHQ 0' f &DWDO\WLF VWUDWHJLHV RI WKH KHSDWLWLV GHOWD YLUXV ULER]\PHV $QQXDO 5HYLHZ RI %LRFKHPLVWU\ 6WULEOLQJ 5 DQG 0LOOHU 6/ f 2ULJLQV RI /LIH

PAGE 194

6ZLW]HU &< 0RURQH\ 6( %HQQHU 6$ f (Q]\PDWLF LQFRUSRUDWLRQ RI D QHZ EDVH SDLU LQWR '1$ DQG 51$ $P &KHP 6RF ,OO 6]RVWDN -: f ,Q YLWUR JHQHWLFV 7UHQGV %LRFKHP 6FL 7DUDVRZ 7 7DUDVRZ 6 DQG (DWRQ % f 1DWXUH 7D\ORU 6 :DOWHU .DVW 3 DQG +LOYHUW f 31$6 8VKHU $ 0F+DOH $ + f +\GURO\WLF VWDELOLW\ RI KHOLFDO 51$ $ VHOHFWLYH DGYDQWDJH IRU WKH QDWXUDO nnERQG 3URF 1DWO $FDG 6FL86$ 9HUODQGHU 06 /RKUPDQ 5 DQG 2UJHO /( f 0ROHF %LRO :DOWHU 05 f (DUWKnV (DUOLHVW %LRVSKHUH (G E\ -: 6FKRSI 3ULQFHWRQ 8QLYHUVLW\ 3UHVV :RHVH & f 7KH *HQHWLF &RGH &K +DUSHU DQG 5RZ 1HZ
PAGE 195

%,2*5$3+,&$/ 6.(7&+ 0DWWKHZ $ &DUULJDQ ZDV ERP LQ 5RFKHVWHU 1HZ
PAGE 196

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

PAGE 197

7KLV GLVVHUWDWLRQ ZDV VXEPLWWHG WR WKH *UDGXDWH )DFXOW\ RI WKH &ROOHJH RI 0HGLFLQH DQG WR WKH *UDGXDWH 6FKRRO DQG ZDV DFFHSWHG DV SDUWLDO IXOILOOPHQW RI WKH GHJUHH RI 'RFWRU RI 3KLORVRSK\ 'HFHPEHU 'HDQ FROOHJH RI 0HGLFLQH 'HDQ *UDGXDWH 6FKRRO

PAGE 198

81,9(56,7< 2) )/25,'$


87
A
r-614 + 0 laser
c-614 + 1000 laser
c-614 + 300 laser
c-614 + 150 laser
c-614 + 50 laser
c-614 + 10 laser
c-614 + 0 laser
B
ribose 614 + laser
c-614 + 1000 laser
c-614 + 300 laser
c-614 + 150 laser
c-614 + 50 laser
c-614 + 10 laser
c-614 + 0 laser
Figure 3-6. Progress curves of caged-614 cleavage demonstrate maximum
deprotection with 150 pulses of laser. (A) Caged-614 with less than 150 pulses of
laser had greatly reduced cleavage plateau, indicating that the sample was not
completely deprotected. (B) When cleavage was normalized by adjusting the percent
cleaved out of only the fraction of sample that was deprotected (determined by NaOH
treatment, the kinetic profiles overlap for all samples (except 0 laser).


68
Testing 614 Rate of Association and Disassociation
Values obtained above for cleavage rates of 614 are comparable to rates obtained
for magnesium independent DNAzymes, but apparently slower than those reported for
other ribonuclease DNAzymes generated by IVS (it is difficult to compare results directly
since other DNAzymes are often characterized at higher temperatures and under alkaline
conditions, both of which dramatically increase cleavage rates). One potential
explanation for this may be the difference in the double strand DNA clamps that were
engineered to fold the DNAzyme onto itself and thus bring the presumably catalytic N40
region in proximity to the ribose substrate site. The lack of published studies on the
impact of these engineered clamps presumably reflects an assumption by the IVS
community that results obtained from in vitro selections should be independent of subtle
changes in starting conditions, and that the overall outcome is indicative of the inherent
ability of nucleic acids to perform the selected behavior. This logic suggests that the two
nucleotide changes made to the BJ + ribose primer to produce the ang + ribose
primer should make little difference to the overall results. Nonetheless, these two point
mutations have the potential to dramatically alter the ability of the DNAzyme to fold
onto itself in the way engineered by Breaker and Joyce. The proclivity of 614 for acting
in trans also suggests that DNA folding may be crucial our understanding of 614 activity.
In our standard protocol for initiating kinetics, enzyme and substrate are mixed
together in reaction buffer, and then heated to 97C for 3 minutes, followed by a slow
cool to 23 C over 10 minutes. The purpose of this slow cool is to ensure complete
denaturation of substrate and enzyme at the onset of the experiment, and allow specific
control of the folding conditions. Initially it was expected that catalysis was dominated
68


53
electrophoresed on a denaturing PAGE/urea gel, and the uncleaved ribose-614 was
excised and purified. This previously uncleaved, gel-purified ribose-614 was ethanol
precipitated, resuspended in reaction buffer, and folded using the slow cool protocol.
Cleavage progress following this gel-purification was then compared to an
untreated sample. Twenty-five percent of the gel-purified, ribose-614 sample was
cleaved in the 300 hours following gel-purification (Figure 3-15). The additional
cleavage following the gel-purification suggests that some of the uncleaved sample is
locked in an inactive structure. It is notable, however, that the gel-purified sample
reached a cleavage plateau of approximately 25%, far lower than the cleavage plateau of
the original sample (70%).
While part of this incomplete cleavage may again be attributed to mis-folded
molecules, if mis-folding were the only cause of incomplete cleavage, a plateau of
approximately 70% would be expected for the gel-purified sample. The fact that the gel-
purified sample reaches a cleavage plateau far below 70% indicates that, while mis-
folding may account for as much as 25% of the uncleaved fraction, other reasons must
account for the majority of the uncleaved fraction. Gel-purification allows mis-folded
molecules the opportunity to refold and cleave, but it results in enrichment for molecules
that are unable to cleave for reasons other than incorrect folding, such as missing the
ribose-adenosine, or other mutations.
One can argue that gel-purification increases the cleavage of ribose-614 by
removing some inhibitory factor rather than allowing for proper refolding. Although the
experiments described in previous sections demonstrated no significant inhibitory effect
of the cleavage products on 614 cleavage, we tested the mis-folding possibility without
53


REFERENCES
Ang, D. N., Suh, B., Westermann-Clark, E., Battersby, T., Benner, S. A. (1999) An
alternative to the origins of life theories: Amino acid-like DNA molecules with catalytic
activity. FASEB J. 13, A1415 Suppl. S
Ang, D. N. (2000) An Alternative to the Origins of Life Theories: Amino Acid-like DNA
Molecules Capable of Improved Catalysis. Dissertation, University of Florida,
Gainesville FL.
Bartel, D. P., Szostak, J.W. (1993) Isolation of new ribozymes from a large pool of
random sequences. Science 261, 1411-1418.
Battersby, T. R., Ang, D. N., Burgstaller, P., Jurczyk, S., Bowser, M. T., Buchanan, D.
D., Kennedy, R. T., Benner, S. A. (1999) In vitro selection of an adenosine receptor from
a library incorporating a cationic nucleotide analog. J. Am. Chem. Soc. 121, 9781-9789.
Beaudry, A. A., Joyce, G. F. (1992) Directed evolution of an RNA enzyme. Science 257,
635-641.
Benner, S. A. (1987) Redesigning life. Organic chemistry and the evolution of proteins.
Chimia 41, 142-148.
Benner, S.A., Ellington, A.D. (1988a) Interpreting the behavior of enzymes: Purpose or
pedigree? CRC Crit. Rev. Biochem. 23, 369-426.
Benner, SA (1988b) Reconstructing the evolution of proteins, in Redesigning the
Molecules of Life, Benner, SA, editor, Springer-Verlag, Heidelberg, 115-175.
Benner, S.A., Ellington, A.D., Tauer, A. (1989) Modem metabolism as a palimpsest of
the RNA world. Proc. Nat. Acad. Sci. 86, 7054-7058.
Benner SA, Switzer, CY. (1999) Simplicity and Complexity in Proteins and Nucleic
Acids, Ch. 19, Ed. by Frauenfelder H, Deisenhofer J, Wolynes PG. Dahlem University
Press.
Breaker, R.R., Joyce, G.F. (1994) A DNA enzyme that cleaves RNA. Chem. Biol. 1, 223-
229.
179


122
DMS (G)
lane: 1 2 3 4 L
84,85
80,81
75
71
65
55,57,59
41
i
38,39 I I I I
33
29,30 lilt
Potassium permanganate(T)
lane: L 1 2 3 4
18 . 15^..
4
Figure 3-48. Structural probing using DMS and potassium permanganate to
chemically modify single-stranded DNA shows some protected positions. End-
labeled d-614 (300 nM) was incubated, with (lane 3 and 4) or without (lane 1 and 2)
unlabeled r-primer substrate (900 nM) overnight in reaction buffer to allow for folding
to reach equilibrium. Samples were then treated with either DMS or potassium
permanganate for 2 minutes or 5 minutes to modify either single-stranded G- or T-
nucleotides, respectively. Samples were then ethanol precipitated and treated with 1M
piperidine for 1 hour at 80C. Samples were again ethanol precipitated resuspensded
in formamide loading buffer, and run on a 10% PAGE/urea gel next to a 10-bp ladder
(lane L). Numbers indicate the positions of Gs or Ts in the complete sequence
(between nucleotide 10 and 90). Positions that demonstrated some degree of
protection from chemical modification are indicated with the red box.


175
suggesting that, like the singletons, survival may have been based on being a good
substrate rather than a good catalyst.
(c) Family beta, gamma, and delta are all very similar to previously selected catalysts (but
nearly all differ by at least one nucleotide), suggesting contamination, a non-
random library, or PCR bias. Each of these possibilities is being examined
further.
The sequences of Family beta, gamma, and delta all show a great deal of sequence
similarity within each Family, as well as between each Family. Within a Family, a
change of a few nucleotides was shown to change the rate up or down by an order of
magnitude. If the occurrence of similar catalysts from two independent selections arose
from contamination, the sequence variation within the families from the second selection
arose through divergence, suggesting that the landscape surrounding the original parent
strains is well populated with active sequences, providing a relatively smooth landscape
and allowing evolution to proceed easily.
If the occurrence of similar catalysts from two independent selections arose
through convergence, this implies that the library must not have been (much) larger than
the original sampling (10 ). Within an order of magnitude, we know of at least 10
active catalysts from the library, implying that catalysts are present no less than 102/1012
= 1010 (and could be much more frequent if the library is smaller than 1012).
Given the abundance of catalysts differing by only a few nucleotides, methods of
estimating catalytic abundance based on restriction digestion (Bartel, 1993) will grossly
underestimate the true catalytic abundance since they can not distinguish between these
similar catalysts.


139
approximately follows first order chemical kinetics and therefore is determined solely by
a molecules intrinsic rate (kn, as assigned by the initial distribution function, either linear
or exponential) and time (t) following the equation P(cleavage) = exp(-A: St).
Our previous analysis of 614 demonstrated that DNAzymes do not necessarily
follow first order kinetic models. Trans-cleavage is in fact possible. Therefore, some
cleavage is second order and concentration dependent. Does this invalidate our model
which assumes first order kinetics? Only slightly. Tra/is-cleavage for 614 was
significant only at high concentrations; the effective molarity of cis-cleavage was
approximately 25 nM, suggesting that c/s-cleavage predominates below 25 nM.
Therefore, if selections are performed at low concentrations, m-cleavage will
predominate, and our first order kinetic approximation is acceptable. Future models may
be improved to consider bi-molecular reactions. Alternatively, I VS experiments may be
designed to eliminate trans-cleavage (perhaps by re-engineering the binding motif in the
constant region of the library).
Using Simulations to Determine the Impact of the Leakage (L) Parameter on IVS
Initial simulations were done to determine the influence of factors such as leakage
(L), selection time (St), maximum rate (kfastest), and the initial abundance of reactive
molecules (RF) on the cleavage progress curve and population distribution. Most of
these initial simulations were done using the simplest distribution function, a linear
distribution with m = 0, but the conclusions can be generalized to other distribution
functions.


116
1
j

o
c
o
I
A
55-i
50-
45-
40-
35-
30-
25-
20-
15-
10-
5
T

r-
60
20 40
B
50-1
40-
S 30
o
<->
c
o
a 20
8.
10-
A
10
2 nM r-614 (4C)
2 nM r-614 (25C)
-i 1 1 1 1 1 1
80 100 120 140 160 180 200
hours
* 1.5 nM r-lib6k>2
+ 100 nM d614
(15C)
1.5 nM r-lib6k)2
+ 100 nM d614
(25C)
~i 1 1 1
20 30 40 50
hour
Figure 3-42. Lowering the incubation temperature does not reduce the rate of
614 cleavage acting in either cis or trans. (A) Low concentration of r-614 (2 nM,
favoring c/.v-cleavage) was incubated at either 4C or 25C. (B) Low concentration of
r-lib61o2 substrate was incubated with d-614 (100 nM) at either 15C or 25C.


149
mirrors the proportionality in the initial population. This relationship can be seen in
Figure 4-15 and Table 4-1.
The approach to a stabilized population (selection plateau) is slower, and the peak
of catalytic distributions for selected populations is shifted closer to the St, for the
exponential distribution. Both factors indicate that the population is still enriching for
faster molecules; the slow enrichment is a result of an increased distribution of catalysts
with half lives between 6*St and St/6 compared to catalysts with half life initial distribution. Although the stabilized population gives little information about the
distribution of catalysts with half lives between 6*5/ and St/6, the rapidity at which
increasing rounds of selection approaches selection plateau may be used as an indicator
of the initial distribution of catalysts in this range.
Deciphering the Distribution Function for Catalysts
Given that during an actual in vitro selection, the only data that can (easily) be
obtained experimentally are fraction cleaved vs. time for the entire pool at each
round of selection, how then can we estimate the initial distribution of catalytic power in
library? The only experimentally measurable variable is the progress curve for cleavage
at each round of the selection process. From this we can determine the progress of the
entire selection by plotting the amount cleaved at the completion of each round (t = St).
When this selection-progress plot reaches a plateau, we know we have isolated a
stabilized population. Further rounds of selection will not significantly alter the
proportion of catalysts in this stabilized population, and of equal more importance, the
proportionality of this population directly mirrors the initial library (for catalysts with


57
sequence is identical in all molecules (in particular, the regions surrounding the ribose-
adenosine).
This opens the possibility that the enzymatic region may bind to the substrate
region of another molecule and either cleave in trans or inhibit cleavage. In the standard
Breaker-Joyce protocol (which was used to generate 614), binding potential DNAzymes
to a column is believed to prevent molecules from binding to and cleaving other
molecules. Amplification and reselection of the survivors is also believed to
preferentially enrich self-cleaving DNAzymes. It is possible, however, that an
enzymatic DNAzyme can bind to a substrate molecule, and together they bind to the
column via the biotin of one of the two molecules. This conjoined DNAzyme/substrate
complex survives washing, but upon cleaving of the substrate, either both enzyme and
substrate are released from the column (if the conjoined molecule was attached to the
column via the biotin of the substrate), or only the substrate is washed off (if the
conjoined molecule was attached to the column via the biotin of the enzyme). This may
result in the inadvertent selection for DNAzymes that can cleave in trans, as well as
molecules that are good substrates for tra/zs-cleavage.
The ang primer used to generate DNAzyme 614 (as well as kinetic analysis) is
nearly identical to the Breaker-Joyce primer, but with two single-nucleotide changes that
reduce the strength of the base-pairing clamps designed to encourage cis folding. These
changes further open up the possibility that cleavage can occur in trans. We are unaware
of any published results that explicitly consider the impact of binding clamps on the
outcome of selections, but based on the assumption that the protocol selects for cis-
57


ACKNOWLEDGMENTS
I would like to thank my parents and family for their continuous support. I would
like to thank several people for their early encouragement in my career in science,
specifically my father, as well as Bill Crabtree, Tracy Bailey, Ken Cline, Ralph Elenry,
and Mike McCaffery. I am indebted to the scientific collaboration of Dr. Maury
Swanson and his entire lab, specifically Carl Urbinati, Ron Hector, and Keith Nykamp.
Alonso Ricardo helped me tremendously to execute many of the experiments described in
this dissertation. Mike Thomson offered invaluable and unending assistance. Without
Steven Benners patience and enthusiasm for science, this research endeavor would not
have been possible.
iii


89
r614 + 300 laser
r614 + 150 laser
r614 + 50 laser
r614 + 10 laser
r614 + 0 laser
hours
r-614 + 300 laser
r-614 + 150 laser
r-614 + 50 laser
r-614 + 10 laser
r-614 + 0 laser
hours
Figure 3-9. Ribose-614 is not damaged by laser. (A) Full time course, and (B)
linear initial phase, of ribose-614 self cleavage following increasing amounts of laser
pulses. The cleavage profile of ribose-614 is unchanged by exposure to increasing
number of laser pulses.


95
A
r-614 (wt) [108nm]
r-614mut "6/10#1Z (T54C)
r-614mut "1/20/R' (C72A)
r-614mut "4k)3" (T82-)
r-614mut "1b1 (A70T)
r-614mut "1/20#6 (G65A)
r-614mut "1/20/P' (G54A)
r-614mut" 1/20/3" (64insA65, C72T)
r-614mut"4up6" (C72T)
r-614mut"1/20/G" (G84-)
r-614mut"6/10#2" (G38A)
r-614mut"6/10#14" (G41A)
r-614mut"6/10#16" (T44-)
B
r-614 (wt) [108nm]
r-614mut "6/10#12 (T54C)
r-614mut "1/20/R' (C72A)
r-614mut "4lo3" (T82-)
r-614mut "1b1" (A70T)
r-614mut "1/20#6" (G65A)
r-614mut "1/20/R' (G54A)
r-614mut "1/20/3" (64insA65, C72T)
r-614mut "4up6" (C72T)
r-614mut "1/20/G" (G84-)
r-614mut "6/10#2" (G38A)
r-614mut "6/10#14" (G41A)
r-614mut "6/10#16" (T44-)
hours
Figure 3-18. Mutants of 614 have reduced catalytic activity. (A) Complete time
course; (B) Linear initial phase of time course. Mutations of 614 were common in the
uncleaved fraction of 614 following cleavage plateau. Kinetic analysis was done of
many of these clones (all at ~100 nM). Most mutations greatly reduced catalytic
activity when compared to wildtype 614.


137
distribution of rates at each round, and if possible, extrapolate from these measurements
the distribution of rates in the initial library.
The model begins by simulating an initial pool composed of 10 unique
molecules (roughly the size of the initial library screened in a typical IVS). This initial
pool is then divided into a Reactive Fraction (RF) and a Non-reactive Fraction
(NRF). The NRF is defined as the portion of the library that does not have any intrinsic
catalytic activity above the background cleavage rate (for our experimental system, the
background cleavage rate is ~10'7 min'1, based on the experimentally determined
cleavage rate of ribose by magnesium ion (Breaker and Joyce, 1995).
All reactive molecules in the initial library (RF) are assigned randomly a discrete
rate category (in this simulation 21 rate categories were used) between ks¡OWst and kfastest
(the intrinsic rate for each category increases proportionally between ksiowest and kfastest)
The lower limit for catalytic rate constants, ksiOWesi, is set to 100-times faster than the
background rate (this lower limit is set both by the methods of detecting a difference
between background and signal and the duration of a kinetics experiment: we must detect
1% cleavage above background in 1000 minutes, corresponding to a lower limit of ~10'5
min'1). The rate constant of the fastest molecule represented in the initial library, kfastest,
must be determined experimentally as it is expected to depend on the particular catalytic
reaction and library being tested. Previous research suggests an upper limit for
DNAzyme cleavage of a phosphodiester bond in the region of 0.1 min'1. Unless
otherwise stated, kfastest was therefore set to 0.1 min'1.
Very little is known about the likely distribution of catalytic rates in a random
library; indeed the distribution of catalytic abundance versus catalytic rate and may be


61
exponential equations (in which both the rate and plateau can vary for each equation). At
low ribose-614 concentrations (3.5 nM and below), the progress curves for cleavage fit
well to a single exponential equation, whereas at higher concentrations the progress
curves were fit better by the sum of two exponential equations. A single exponential
-2
curve fit to progress curve for 0.3 nM ribose-614 cleavage estimated a rate 0.011x10"
hr"1 (with a predicted plateau of 61%). This agrees with the other estimations of the
unimolecular rate made above (which, comparatively, are underestimations since they
have not corrected for a plateau below 100%).
This suggests that at low concentrations (3.5 nM and below), the reaction behaves
in unimolecular fashion while at higher concentrations the reaction is not well
characterized by a single unimolecular rate law. Furthermore, we can conclude that, at
low concentrations, where unimolecular processes dominate, a second conformer of en
folded 614 with lower reactivity is not present at significant levels. In an earlier section
(Improperly Folded DNAzyme 614), we already considered variants of 614 that were
inactive (mutants, for example).
This is important because if these DNAzymes in general have a single active
conformation, and react following a single exponential first order rate law, then the
distribution of catalytic power within a population of related molecules can be modeled
(at least as an approximation) by a transform that fits a spectrum of first order
exponential processes to a progress curve.
The non-unimolecular component of ribose-614 cleavage is not purely bi-
molecular as tripling the concentration of ribose-614 does not increase the initial rate
nine-fold as would be expected (but in fact it increases less than two-fold). Furthermore,
61


This work is dedicated to my son, Christian.
11


131
reaction buffer. No catalysis was observed, however, when the NaCl was omitted from
the reaction buffer, even with the addition of ImM MgC^ (Figure 4-8B).
Sequence Evolution During I VS
The kinetic profiles of pool H show a progressive increase in catalytic power
following the third, fourth, and fifth rounds of selection, with no further increase
following subsequent rounds of selection. The evolution of the populations during the
selections was examined by following the progression of sequence variation. Individual
species from pools of survivors following rounds 1, 3, 4, 5, and 8 of selection were
cloned and sequenced.
None of the eighteen clones sequenced from the round one pool were similar to
each other, and only one was similar to sequences isolated from later rounds. In contrast,
most of the sequences isolated from later rounds of selection (pools following rounds 3,
4, 5, and 8) fell into four Families, termed alpha, beta, gamma, and delta (Figure 4-9).
Families beta, gamma, and delta share two common sequence motifs, both in the 3 end
of the N40 region.
The high degree of sequence variation seen in the round one pool is expected,
and, with the observation that this pool shows very little catalytic activity, is consistent
with the idea that survivors in this pool are dominated by non-catalytic molecules that
leaked through the selection system (either by nonspecific cleavage, trans-cleavage, or
abnormally fast migration during gel electrophoresis).
As catalytic activity of the pool begins to increase following the third round of
selection, examples of the sequence Family alpha and beta begin to dominate. Examples


79
Figure 3-46 shows that a four base-pair helix can be formed between nucleotides
35CATG38 of 614 and the identical nucleotides of the ang + ribose substrate. This helix
is not formed when 614 cleaves in cis because the sequence CATG sequence occurs only
once. This short helix presumably offers additional stabilization between the ang +
ribose substrate and the 614 enzyme without disrupting potentially catalytic helixes of
structures A1 and C (although it does disrupt part of structures A2 and A3).
Interestingly, a mutational screening of 614 isolated a G38A mutant (called
june#2 G38A) that had reduced activity (compared to wild-type 614) when tested at
100 nM (a concentration at which significant trans cleavage occurs).
If io
The significance of the predicted CATG ras-helix was tested by introducing
the compensatory mutation into the non-catalytic substrate r-lib61o2 (the compensatory
mutant is called lib61o2 C35T). Previous /rans-cleavage assays demonstrated that
wild-type d-614 can cleave ribose-lib61o2. The 614 mutant june#2 G38A was then
tested for its ability to cleave either r-lib61o2 or r-lib61o2 C35T. Figure 3-47 shows
that the introduction of the C35T mutation into the Hb61o2 substrate improves cleavage
by the 614 mutant june#2 G38A, suggesting that the 35CATG38 helix is indeed
important for /rans-cleavage. The reduction in cleavage seen when june#2 G38A
cleaves Hb61o2 C35T as compared to wild-type 614 cleaving wild-type r-lib61o2 is
likely the result of the weaker AT base-pair as opposed to the GC base-pair.
Structure predictions were made with 614 bound to the ang + ribose substrate
via the two experimentally supported interactions: base-pairing between 35CATG38 of the
614 and the CATG of the substrate, and base-pairing between the
l7CGACTCACTAT27 region of the substrate and the 85GTAGTGACG93 region of the
79


2
(cultural ideas) and in silico evolving systems (software) are not formally recognized as
living because they are not chemical. The information-storing structure for memes is
poorly understood, and this may contribute to the biases against considering this evolving
system a life-form. For in silico evolving systems, the range of physical properties
(function space) is arbitrarily defined by the programmer and therefore not open-ended
like a chemical system. It can be argued that the function space for a chemical
evolving system is not in fact open-ended but rather limited by physical laws.
Ultimately, both memes and in silico evolving systems can be considered the product of
and therefore not independent of, a living system. One may, however, conceive of life as
any system capable of Darwinian evolution. The NASA definition of life might then
be generalized to any informational-containing system that undergoes Darwinian
evolution.
Although the second law of thermodynamics dictates that the universe progress to
a state of greater disorder, this does not necessarily result in the loss of all information.
Within a system, order (information) can be created and maintained if it is matched by
the creation of greater disorder elsewhere in the universe. Nonliving phenomena often
result in ordered structures, such as ripples in the sand of a tidal basin, the lattice structure
of a crystal, or polymer formation. This decrease in entropy within the system is coupled
to free energy consumption. In order to persist indefinitely, an informational system
must prevent information decay by coupling the maintenance of ordered structure to the
release of available energy.
Two such mechanisms for preserving encoded information are easily imagined:
repairing the information (requiring reference to a correct version), or duplicating the
2


170
used to examine the impact of various selection parameters on the outcome of IVS
experiments.
Catalytic Behavior of DNAzymes from IVS is More Complex then Previously Assumed
Catalysts isolated from a random library using the Breaker-Joyce protocol are
assumed to be dependent on the cofactor used in the selection. Further, cleavage is
assumed to follow a unimolecular mechanism. Analysis of 614 revealed that both of
these assumptions are violated.
DNAzyme 614 was generated with 1 mM MgCh in the reaction buffer but is in
fact active without MgC^. Geyer and Sen directly selected for metal-independent
DNAzymes using the Breaker/Joyce IVS procedure without metal cofactors in the buffer,
demonstrating that divalent cations are not absolutely required (1997). Examination of
both the hairpin ribozyme and hepatitis delta virus ribozyme has revealed that Mg+~ plays
a only passive role (likely structural), and instead implicate an internal nucleotide in
general acid-base catalysis (Young et al 1997; Hampel and Cowan, 1997; Nakano et al,
2000; Rupert et al, 2002; Shih and Been, 2002). It is likely that 614 utilizes an internal
nucleotide in a similar manner.
It was also observed that 614 cleavage is concentration-dependent, prompting a
study of its bimolecular cleavage process. This revealed that 614 can cleave a number of
substrates in trans, and that the rate of tras-cleavage at saturation is 6-fold higher than
cis cleavage. The balance between cA-cleavage and trans-cleavage is concentration-
dependent. The cleavage rate of 614 acting in cis and trans is enhanced by lowering the
temperature, likely indicating that the impact of lower temperature on the formation of


15
one-biopolymer stage based on RNA, in vitro selections have largely focused on nucleic
acid biochemistry.
More than a decade ago, Benner proposed what has come to be the standard
explanation for the "poor" catalytic power of nucleic acids (where "poor" reflects in part
an unbalanced comparison with natural protein enzymes) (Benner, 1987; Benner, 1988b).
Nucleic acids, he argued, have little organic functionality, at least compared to natural
protein enzymes. DNA lacks in the scaffold cationic groups, imidazoles, thiols, and
carboxylates, all of which play effective catalytic roles in protein enzymes near neutral
pH. To relieve these limitations, Benner suggested expanding the genetic alphabet for the
purpose of permitting DNA to carry more of these functional groups (see Switzer, 1989;
Piccirilli, 1990).
This hypothesis gains support from the observation that many of the catalytic
RNAs believed to be vestiges of the RNA world (spliceosome, RNaseP, ribosome) contain
post-transcriptional modifications; the conservation of this difficult task both throughout 4
billion years of evolution and across wide phylogenic taxa suggests that these
modifications must play a critical role in the function of these catalytic RNA. Evidence
for the prebiotic synthesis of C-5 functionalized uracil including ammonia, glycine,
guanidine, hydrogen sulfide, hydrogen cyanide, imidazole, indole and phenol also lends
credence to the hypothesis that functionalities appended to nucleic acids may have served
a crucial role in the origins of catalytic nucleic acids and subsequent life (Robertson and
Miller, 1995).
Several groups (Battersby, 1999; Latham, 1994; Sakthivel, 1998; Perrin, 1999;
Perrin, 2001) have now synthesized functionalized nucleic acids, mostly derivatives of 2'-
15


158
A
HR#4 60nM
H R#4 15nM
H R#4 3nM
D
- H RK7 60nM
* HR7 15nM
* HR*7 3nM
c
HF6 60nM
4 HRN615nM
H RK6 3nM
Figure 4-5. The initial rate of cleavage for Selection H following rounds 4-8
increases slightly with increased concentrations of DNA. DNA from the survivors
of each round of Selection H was re-amplified from stocks and progress curves for
cleavage assayed at 3, 15, and 60 nM.


107
B
K exp fit =
0.04213
003780
0.04260
0 02690
30 nM r-primer,
300 nM d-614
10 nM r-primer,
100 nM d-614
3.3 nM r-primer,
33 nM d-614
1.1 nM r-primer,
11 nM d-614
22.5 nM r-lib6lo2
cleaved by
450 nM d-614
7.5 nM r-lib6lo2
cleaved by
150 nM d-614
2.5 nM r-lib6lo2
cleaved by
50 nM d-614
0.8 nM r-lib6to2
cleaved by
13 nM d-614
Figure 3-32. Single-turnover kinetics for d-614 cleaving r-primer or r-lib61o2
below saturation.


TABLE OF CONTENTS
ACKNOWLEDGEMENTS iii
ABBREVIATIONS vii
ABSTRACT ix
CHAPTER page
1 INTRODUCTION 1
Requirements of a Living System 1
Chemical Challenges to the Prebiotic Synthesis of Life 4
Progress Towards Understanding the Genesis of Life:
Dual Function Biopolymers 6
Progress Towards Understanding the Genesis of Life:
Prebiotic Synthesis of Potentially Useful Monomers and Polymers 8
The Next Challenge: Obtaining Catalytic Function from Random Libraries 11
Experimental Design 19
2 MATERIALS AND METHODS 24
Preparation of Precursor DNAymes via PCR 24
Preparation of Single-Stranded DNAzymes 26
5-End Labeling of DNA 28
DNAzyme Kinetic Assays 28
Cloning and Sequencing DNAzymes 29
In Vitro Selection 31
3 RESULTS OF 614 ANALYSIS 36
Research Objectives 36
Developing Methods for Preparing Single-stranded DNAzymes by
Exonuclease Degradation of 5-Phosphorylated Complementary Strand 37
Asymmetric PCR with Tails, Followed by Gel Purification 39
DNAzyme 614 with uncaged- and caged-ribose 41
Achieving maximal deprotection of caged ribose 41
Kinetic profile of 614 with uncaged- and caged-ribose 42
IV


99
A
B
[614] nM
[614] Yint = 0 005893 /hr = k(unimolecular)
m = 8.9x10+5 M-1hr-1
Figure 3-22. Extrapolating r-614 cleavage rates at low concentrations to infinite
dilution estimates a unimolecular self-cleavage rate of 0.0059/hr. (A) The initial
rates of cleavage for 614 at various concentrations are plotted vs. initial r-614
concentration using a linear scale. (B) A best fit line of the initial rates for 614
cleavage at low concentration (<= 3.5 nM) is extrapolated to zero, yielding a cis-
cleavaee rate of 0.0059/hr.


112
4CH
400nM r-lib6lo2 + 100 nM d-614:
* mix, then slowcool (15C)
* mix, then slowcool (25C)
slowcool, then mix (25C)
* mix, no slowcool (25C)
-i 1 1 1
20 30 40 50
hour
Figure 3-38. Inter-molecular folding between r-614 and r-lib6lo2 is not affected
by slow cooling. A four-fold excess of substrate r-lib61o2 was mixed with d-614
either before or after the slow cool folding protocol. In a third condition, samples
were mixed but were not denatured and folded using the slowcool protocol. The mix,
then slow cool condition was performed in parallel, with one condition at 15C while
all other samples were incubated at 25C. The slow cool folding protocol had not
effect on cleavage rates. Incubating at the samples at 15C initially increased substrate
cleavage compared to the same condition at 25C.


142
additional rounds required to reach a survivorship plateau, the population distribution for
all test cases converges to the same population distribution (this occurs when there is no
longer selective enhancement for faster molecules and corresponds to the survivorship
plateau). This is manifested by the convergence of progress curves for all test cases
(once each test case reaches a survivorship plateau). This demonstrates that, although the
leakage parameter dramatically affects the number of rounds necessary to isolate
catalysts, it does not alter the distribution of selected catalysts (following sufficient
rounds of selection).
The distribution of selected catalysts in the stabilized population bears an
interesting relationship to the selection time. There are three distinct population zones
defined in relation to the St: zone A is comprised of molecules with half-lives larger
(slower) than St; zone B is comprised of molecules with half-lives between St and
approximately St/6 (6-fold faster); and zone C is comprised of molecules with half-lives
between St/6 and kfastest.
Although in this simulation all molecules, regardless of their rate constant, were
initially present in equal amounts, molecules in zone A survive with increasingly lower
probability than faster molecules (zone B and C). Eventually the surviving population
reaches a point where molecules from zone A are effectively absent. Unless their initial
abundance is exceedingly high relative to faster molecules, molecules of zone A are
quickly eliminated from the selected population.
In fact, given an initial population consisting catalysts with one of only two
possible intrinsic rates, one with rate equal to St ('Alife/ast = St), and the other ten times
slower ('Alifesiot = 10*5/), the slower catalysts must be 7.5-fold more abundant than the


initial rate (fraction cleaved/hour)
98
0.035-.

0.030-
0.025-
0.020-
0.015-
0.010- .


0.005-
0.000 1 1 1 1
0.1 1 10 100 1000
[r-614] nM
Figure 3-21. The rate of r-614 cleavage increases at higher concentrations, but
reaches a lower limit at low dilutions. Lowering r-614 concentration initially
reduces cleavage rate, but the cleavage rate reaches a lower limit of approximately
0.005/hour at concentrations of 20 nM and below.


141
detection of survivors from round 3 to round 4. Increasing the leakage parameter even
further to 4% delayed the earliest detection of survivors to round 7. This indicates that
the leakage parameter plays a significant role in determining the number of rounds of
selection required to achieve detectable levels of catalysis.
With sufficient rounds of selection, each test case eventually reaches a
survivorship plateau a point beyond which further rounds of selection no longer
significantly increases the fraction of the population surviving each round (Panel B).
Panel C shows a similar plateau for the amount cleaved at t = 81 minutes (it is useful to
plot the amount cleaved at a set time for conditions where populations are selected with
various selection times). In this simulation, each test case reaches a survivorship plateau
near 100% cleaved. Panel D shows the distribution of catalysts for each test case at
successive rounds of selection. Comparing Panel B (or C) with Panel D illustrates that at
the same round the survivorship begins to plateau, the distribution of catalysts in the
selected populations reach a stable state referred to as the stabilized population -
where the proportionality between catalytic rates of surviving molecules remains constant
even with additional rounds of selection. Once this survivorship reaches a plateau, there
is no selective enhancement of faster molecules over the remaining slower molecules,
resulting in the stabilization of the surviving population.
Case 1 achieves a survivorship plateau of 100% by round 4 (panel B), and panel
D shows that the population distribution for case 1 remains relatively constant from
round 4 through the end of the simulation (round 9). Case 2 and case 3 achieve a
survivorship plateau of 100% much later than case 1 (round 6 and 9, respectively) as a
result of a higher leakage parameter (L = 0.2% and 4%, respectively). Despite the


14
RNA have yielded only ribozymes that decorate themselves inappropriately with tagged
nucleotides (Johnston, 2001, page 1324). Furthermore, estimates from their work
suggest that more than 10^0 random RNA sequences might be required to get template-
directed RNA polymerization from a truly random pool. More directly, Bartel and Szostak
noted that the odds of obtaining a ligase able to enhance a very simple templated ligation
reaction by four orders of magnitude were on the order of one in 1013 (Bartel, 1993).
As mentioned earlier, it is not certain that life originated from RNA polymers,
indeed many other candidate polymers are imaginable. In support of this, it is notable
that catalytic activity has been isolated from various different types of libraries, including
amino acid, ribonucleic acid, and deoxynucleic acid polymers. Extrapolations from a
25
search of amino acid libraries estimate that in a fully randomized library, a library of 10
members would be required to obtain an active catalyst (in that case, AroQ mutase)
(Taylor, et al 2001). This search of amino acid libraries concluded that, although the
diversity of functionality in proteins provides intrinsically greater catalytic potential,
folding proteins may be inherently more difficult than folding nucleic acid polymers,
therefore making an active enzyme from a random polymer of amino acids appears to
substantially more difficult that obtaining a ribozyme from a random nucleic acid library.
Although natural amino acid polymers (proteins) function superbly as catalysts,
and amino acid catalysts have been isolated from random libraries, amino acid polymers
are not well suited as carriers of genetic information for reasons described earlier. In
vitro selections using amino acid libraries are also much more technologically limited
than searches of nucleic acid sequence space. For all these reasons, and the indication
that, although life on earth may not have started with RNA, it clearly passed through a
14


25
ribose (ang + cage and BJ + cage). The complementary strand primer sequence
was always the same (except complementary +5TAIL 2C9*G in which the 5 G was
changed to T), but varied in non-coding modifications appended to the 5-end:
complementary no 5-P primer:
(5'-GTGCCAAGCTTACCGTCA)
complementary + 5-Phosphate primer:
(5'-P-GTGCCAAGCTTACCGTCA)
complementary + 5tail#l primer:
(5'-GGTGGGTGGG-cl8-GTGCCAAGCTTACCGTCA)
complementary +5'TAIL 2C9 primer:
(5'-AAAAAAAAAAAAAAAAAAAA-c9-c9-GTGCCAAGCTTACCGTCA)
complementary +5TAIL C18#2 primer:
(5'-AAAAAAAAAAAAAAAAAAAA-C18-GTGCCAAGCTTACCGTCA)
complementary +5TAIL C18#3 primer:
(5'- AAAAAAAAAAAAAAA-cl8-GTGCCAAGCTTACCGTCA)
complementary +5TAIL 2C9*G primer:
(5'-AAAAAAAAAAAAAAAAAAAA-c9-c9-TTGCCAAGCTTACCGTCA)
complementary +5TAIL 3C9 primer:
(5'-AAAAAAAAAAAAAAAAAAAA-c9-c9-c9-GTGCCAAGCTTACCGTCA)
complementary +5TAIL 4x2'OME primer:
(5'-AAAAAAAAAAAAAAAAAAAA-mUmUmUmU-GTGCCAAGCTTACCGTCA)
Cl8 is an 18-atom hexaethyleneglycol-based spacer, C9 is a 9-atom
triethyleneglycol-based spacer, mU is a 2O-methyl-uridine, P is phosphate. Typical
conditions for a 100 pL PCR contained up to 1 ng template, 100 nM catalytic strand
primer, 100 nM complementary strand primer, 100 pM dNTPS (either standard A, G, C,
and T, or a nonstandard mixture in which thymidine is replaced by 5-(3-aminoallyl)-
2deoxyuridine, henceforth referred to as E-base), 10 mM KC1, 20 mM Tris-HCl (pH
8.8), 10 mM (NH^SO^ 2 mM MgSCL, 0.1% Triton X-100, 3-4 units polymerase (Taq
or Vent exo-), and 10 pCuries alpha-3"P-CTP (for internally labeled samples). PCR
25


ABBREVIATIONS
I VS in vitro selection
bya billion years ago
SL Size of initial library
RF Reactive Fraction
NRF Non-reactive Fraction: NRF= SL*(\ RF)
k Intrinsic rate of molecule each member of the initial library, n, is
assigned an intrinsic rate, kn, which relates to its ability to be cleaved
under selection conditions. The intrinsic rate is sequence-dependent and
therefore unchanging.
Alifen The intrinsic rate of molecule n (k) expressed in terms of half-life:
Alife = ln2 / kn*siowest The rate constant for the slowest catalysts in the
library.
kfastest The rate constant of the fastest molecule in the library.
St Selection time: the amount of time allowed for a molecule to cleave itself
and therefore achieve survival.
m or b Distribution function for reactive molecules. Two functions have been
tested thus far, a linear function with slope = m, and an exponential
function with decay rate = b.
L Feakage: the fraction of the pool that survives a round of selection
independent of St and intrinsic rate constant.
Vll


178
spliceosome often contain modified nucleotides which, along with associated proteins,
can improve structural arrangement. In the future, the incorporation of functionalized
nucleotides that enable catalysis and folding into in vitro selections may dramatically
improve the catalytic potential of random nucleic acid libraries.


UNIVERSITY OF FLORIDA
3 1262 08555 2890


caged-ribose a ribose-adenosine with the 2OH replaced with an ortho-nitrobenzyl
protecting group
r-primer ang + ribose 49-nucleotide substrate containing ribose
r-X oligonucleotide X with ribose-adenosine
d-X oligonucleotide X with deoxy-adenosine
A Deoxyadenosine triphosphate
G Deoxyguanosine triphosphate
C Deoxycytidine triphosphate
T Deoxythymidine triphosphate
E-base 5-(3-aminoallyl)-2deoxyuridine triphosphate (see structure below)
O
kcat(uni)
kl(uni)
k-l(uni)
kcat(bi)
k l(bi)
k-l(bi)
chemical step for unimolecular catalysis
rate of folding for unimolecular catalysis
rate of unfolding for unimolecular catalysis
chemical step for bimolecular catalysis
rate of association for bimolecular catalysis
rate of disassociation for bimolecular catalysis
Bimolecular Kinetic Scheme:
F + q k.1- S = r-614, r-lib6lo2 or r-primer
k-1(bi)
Unimolecular Kinetic Scheme:
k-i(uni) kcat(uni)
614unf0ide(j ,, 614f0|(jeCj Products
k-i(uni)
viii


113
4 nM r-614 +
2025 nM d-614
4 nM r-614 +
2025 nM d-614 +
10 uM chase at
t = 4 hours
200 nM r-614
v 200 nM r-614 +
3000 nM chase
at t = 1 hour
0.5 nM r-614 +
64 nM d-614
o 0.5 nM r-614 +
64 nM d-614 +
1280 nM chase
at t = 45 min
4 nM r-primer +
2025 nM d-614
o 4 nM r-primer +
2025 nM d-614 +
5 uM chase at
t = 4 hours
o 4 nM r-primer +
2025 nM d-614
+ 10 uM chase
at t = 4 hours
hours
Figure 3-39. 7>o/is-cleavage of r-614 is reduced by the addition of cold chase,
indicating that the rate of E*S disassociation is faster than the chemical step of
cleavage. Various substrates were incubated with excess enzyme (614). Following
various incubation times, excess cold chase was added to the reaction. The addition of
chase greatly reduced cleavage of the substrate (as compared to an identical sample
without added chase) regardless of the time at which the chase was added.


18
libraries, the next critical task of estimating the frequency of such activity in a random
library, and determining what factors alter catalytic abundance and power, can be
addressed. Understanding this is the next vital step to understanding how to make the
transition from prebiotic polymers to a living system.
In order to understand how a single-biopolymer life-form is created by prebiotic
mechanisms, we must understand the factors that alter the probability of creating function
from random polymers, specifically the apparent contradication that added functionality
doesnt appear to significantly improve catalysis. Five categories of hypotheses might be
considered to account for this orders-of-magnitude difference between expectation and
outcome when adding functionality:
(a) Our view of the role of functionality in catalysis may be naive. Functional groups
may not greatly enhance catalytic power.
(b) Nucleic acids are, of course, already functionalized (with phosphates and hydrogen
bond donating and accepting groups). Perhaps these are the only functionalities
needed for catalysis.
(c) Nucleic acids, although they do not have many functional groups intrinsically in
their covalent structure, may recruit sufficient numbers of these (in particular,
I f
divalent cations such as Mg ) as "cofactors" to make up for the lack of
functionality covalently linked to the scaffolding.
(d) The I VS experiments may have been designed in a way as to overlook or lose the
best functionalized catalysts.
(e) Perhaps adding a single type of functional group is not sufficient; one may need to
add two or more types of functional groups before the expected large benefit from
functionality is seen (Perrin, 1999; Perrin, 2001).
A priori, each hypothesis has its attractions and disadvantages. For example, we
can easily identify ways in which the best endowed catalysts would be lost in a Breaker-
Joyce protocol (see below), suggestive of hypothesis (d). Divalent metal ions do bind to
DNA, they can participate in catalysis, and they assist even more highly functionalized
18


156
Figure 4-3. Selection H shows similar cleavage rates with ribose- or caged-ribose
primer. Pools of catalysts were generated using either ribose-adenosine or caged-
adenosine from the survivors of Selection H following round 1, 4, and 8. Samples
containing the caged-adenosine were initiated with laser.


52
respect to cleavage, by re-suspending the DNAzyme in reaction buffer without
cofactors), we included a re-folding step that is easily reproducible. Following re
suspending the DNAzyme in reaction buffer, samples were heated to 97C for three
minutes to achieve complete denaturation. The samples were then cooled to 23C over
10 minutes using a thermocycler. Although very consistent, even this folding procedure
cannot guarantee that all DNAzymes find the same active conformation. Indeed,
structure predictions of 614 show several potential structures, perhaps only one of which
is active. Folding into inactive structures can potentially contribute to the low cleavage
plateau.
When 614 cleavage assays are stopped by freezing to -80C and run on a non
denaturing PAGE gel, two bands are present at initial time points. The first (highest)
band represents -15% of the sample, and remains unchanged throughout the duration of
the kinetic assay. The second (middle) band is initially 85% of the sample, but with time,
is converted to a third (lowest) band (at 200 hours, the first band remains 15% of the total
sample, while the second band is reduced to -10%, and the third band is -75%). This
result was observed when r-614 was purified via asymmetric PCR or by exonuclease
degradation, suggesting that the unchanged upper band likely represents an alternative
conformation of the catalytic strand rather than contaminating complementary strand.
We set out to test this possibility with two methods. First, ribose-614 was
incubated under standard reaction conditions for 140 hours, allowing the reaction to
approach the cleavage plateau. An aliquot of this sample was removed and denatured by
adding 2 times the volume loading dye (95% formamide, EDTA, xylene cyanol,
bromophenol blue) and heating at 90C for 2 minutes. The sample was then
52


174
catalytic strand of the double stranded DNAzyme. Both methods avoid many of the
experimental weaknesses of the Breaker-Joyce protocol, and permit selections to be
perfomed in solution. Even without the benefit of a protected ribose, these improvements
should dramatically improve future selections by minimizing the loss of catalysts due to
column leakage and NaOH-based DNAzyme duplex separation.
IVS Geneates Sequence Variants of 614 with Wide Range of Catalytic Power
Molecules isolated from round 8 of Selection H (ang + ribose with standard
nucleotides) selection fall into several groups:
(a) Singletons, which are unlike any other sequence. Each of the singletons tested
had an extremely slow rate of cleavage (in comparison to other competing
molecules). The fact that these molecules persist through 8 rounds of selection
despite extremely slow rates of cleavage suggests that they may be surviving
because they are good substrates for other catalysts. If this scenario is true, they
may be parasitic (they directly compete with the active catalyst for survival), or
symbiotic (they enhance the self-cleavage of the enzyme that also cleaves them).
Selection is expected to continuously minimize the population of parasitic
substrates. If these singletons are in fact symbiotic molecules, then the
survival of both the enzyme and substrate follows a kin-selection model.
(b) Family alpha molecules first appear after Round 1 of selection, and relatives persist in
the population for several rounds until the family apparently goes extinct after
round 4. One representative was tested and it had a rate of cleavage near zero,


118
r-lib6lo2 wt (7.5nM), cleaved by
d-614 wt (200nM)
r-lib6to2 wt cleaved by
d-614 (G86C, G88C, T90A)

>
(0

o
**
c
0)
o
u.
&
40-
35-
30-
25-
20-
15-
10-
5-
0h
10 15 20 25 30 35 40 45 50
hour
O r-lib6to2 (C22G, C24G, A26T)
cleaved by
d-614 (G86C, G88C, T90A)
* 3.5 nM r-614wt self-cleavage
* 19.4nM r-614 wt self-cleavage
12.5 nM r-614
(G86C, G88C, T90A)
self-cleavage
o 12.5 nM r-614
(C22G, C24G, A26T) +
(G86C, G88C, T90A)
self-cleavage
Figure 3-44. Nucleotides 86-90 of 614 base-pair with nucleotides 22-26 of either
itself (in cis), or with a substrate (in trans). .Mutations were introduced into
nucleotides 86, 88, and 90 of 614, resulting in reduced cis and trans cleavage rates.
Introducing the complementary mutations into nucleotides 22, 24, and 26 of the
substrate improves substrate cleavage both in cis and trans.


BIOGRAPHICAL SKETCH
Matthew A. Carrigan was bom in Rochester, New York, on April 18, 1974. He is
the third of four children bom to James Carrigan and Leslie Malone. Matthew attended
the University of Florida from 1992-1997 where he completed his undergraduate degree
in Biochemistry and Molecular Biology, along with minors in Psychology, Chemistry,
Religion, and Anthropology. As part of his undergraduate studies, Matthew performed
research with Dr. Ken Cline and Dr. Ralph Henry of the University of Florida. Matthew
started graduate school in the Department of Neuroscience in 1997. Matthew and
Cristina Mazpule were blessed with a baby son, Christian Curtis Mazpule-Carrigan, in
February of 2001. Matthew completed his thesis under the guidance of Dr. Steven A.
Benner, and received his doctorate in December 2002.
184


66
this case, is therfore approximately 3.5 19 nM; above this concentration, trans cleavage
increasingly predominates.
The 6-fold increase in rate for trans cleavage is also surprising given the
expectation that both the catalytic and the substrate region of 614 is unchanged
whether cleavage is in cis or trans. The implication is that the linker connecting the
substrate and enzyme in the cis-acting 614 somehow interferes (steric hindrance)
with binding in a way not present in the trans-acting 614.
d-614 Cleaves with Multiple-turnover
In order to be considered a true enzyme, the catalyst must be unchanged at the
completion of its activity and thus be able to complete another round of catalysis. We
have already demonstrated that cis-acting 614 cleaves itself, and that neither product
retains significant catalytic activity, hence it behaves as an auto-catalytic molecule, but
not classically as an enzyme. We have also demonstrated that d-614 can act catalytically
in trans and is not altered in the process. We not proceed to test whether the catalytic
activity of d-614 can function with multiple turnover.
Figure 3-33 and 34 show the result of incubating either ang + ribose or r-lib61o2
substrate in four-fold excess over d-614 enzyme (either at 133:33 or 400:100 nM
substrate:enzyme). Multiple turnover was achieved, but rates were quite slow. At the
highest concentration of enzyme and substrate, 2 turnovers of substrate was observed at
100 hours for ang + ribose primer substrate and 200 hours for the r-lib61o2 substrate.
As seen under single-turnover conditions, the ang + ribose substrate cleaved both
faster, and at lower concentrations, than the full-length r-lib61o2 substrate under multiple-
66


59
process with saturation (implying that, for the DNAzyme 614, one molecule of 614 must
bind to another molecule of 614 for a reaction to occur, but that the 614-614 complex
could disassociate before a reaction occurred, a condition well known in Michaelis-
Menten kinetics), then the apparent first order rate constant would be dependent upon
[614] initially, and approach an asymptote at high [614], Here, a plot of apparent kobS
versus [614] should be a line that includes the (0,0) origin, sloping up and ultimately
leveling to a plateau.
The concentration dependence of the rate of cleavage of ribose-614 was tested by
examining the progress curve and the initial rates for ribose-614 cleavage at various
different dilutions (430 nM to 0.3 nM ribose-614). Results from this experiment (Figure
3-20, 21, and 22) show a reality where several rate processes are operating at the same
time.
Figure 3-21 shows that the initial rate of ribose-614 cleavage decreases with
decreasing concentration (a logarithmic abscissa is used to capture the full range of the
experimental concentrations). The decrease in the initial rate of cleavage with decreasing
concentration reaches a lower limit below which further dilution causes very little further
reduction in the initial rate. This implies that the rate does have some concentration
dependence and implies that substrate is being consumed by two paths: a first order rate
process and, at least partly, by a process that is higher than first order. In the left portion
of Figure 3-21 (low [614]), the first order process dominates the overall rate. In the right
portion of Figure 3-21 (high [614]), second order processes dominate the overall rate.
This experiment was repeated several times over with similar results.
59


32
as BJ + cage but the caged-ribose had been de-protected by previous exposure to 300
pulses of laser (100 mJ/pulse)). Each primer was extended using either the standard
dNTPs (equimolar T, A, G, and C) or with a non-standard, 5-position-functionalized
nucleotide triphosphate (5-(3-aminoallyl)-2-deoxyuridine 5-triphosphate, referred to as
E-base) substituted for T-triphosphate (equimolar E, A, G, C). This yielded six different
I VS starting libraries,
libary #1: BJ + cage primer, standard dNTPS;
libary #2: ang + ribose primer, standard dNTPS;
libary #3: BJ + cage primer, AGC+E;
libary #4: ang + ribose primer, AGC+E;
libary #5: BJ + de-protected cage primer, standard dNTPS;
libary #6: BJ + de-protected cage primer, AGC+E)
Run-off reactions volumes ranged from three to 20 milliliters, each containing 1 ng/pL
library template, 100 nM catalytic strand primer, 100 pM dNTPs, 10 mM KC1, 20 mM
Tris-HCl (pH 8.8), 10 mM (NH4)2S04, 2 mM Mg2S04, 0.1% Triton X-100, 20 units/mL
Vent exo-, and 4.3 pCuries/mL alpha- P-CTP. One milliliter of each sample mixture
(minus polymerase) was aliquoted into 1.5 mL eppendorf tubes, heated at 96C for 8
minutes, and then slowly cooled to 55C over a period of 30 minutes. Polymerase was
then added, and the samples were transferred to 72C for 15 minutes. The samples were
then ethanol precipitated with ammonium acetate and with glycogen as a carrier by
storing overnight at -80C and then centrifuging for 40 minutes in Corex glass tubes at
10,000 rpms in a Beckman centrifuge at 4C. The ethanol was removed and the pellet
was re-suspended in 200 pL water. The samples were ethanol precipitated a second time
in with ammonium acetate by centrifuging at 16,000 g in a desktop centrifuge for 20
minutes at 4C. After removing the ethanol, the pellet was resuspended in 80 pL
formamide stop dye and warmed at 37C for 30 minutes to completely dissolve the pellet.
32


CHAPTER 3
RESULTS OF 614 ANALYSIS
Research Objectives
Understanding why added functionality does not appear to significantly improve
the catalytic power of nucleic acid enzymes under in vitro selection (IVS) schemes
requires knowledge of how chemical functionality alters the distribution of catalytic
potential within sequence space. This in turn requires that an experimental model system
be comprehensively understood so that it can be modeled mathematically with sufficient
accuracy to support quantitative analysis. Indeed, this depth of knowledge is important to
understanding how molecular evolution in general affects phenotypic evolution in an
evolutionary landscape over sequence space even up to organismal or ecosystem levels.
The success of previous IVS has been only modest (described in the introduction), and
apparently insufficient to explain the origins of life; this modest success of previous IVS
may, in part, be attributable to several details of the IVS experimental system. One
strong possibility is that the fastest catalysts are lost in the experimental work-up.
Our aim is two fold: (a) Examine the experimental systems in sufficient rigor so
as to allow mathematical modeling, and (b) improve the selection procedure so as to
minimize the potential loss of catalysts and allow quantitative analysis. To develop a
system that supports detailed mathematical modeling of DNAzyme selections, we first
examine in detail a model DNAzyme known as 614 (isolated from a previous IVS). To
minimize potential loss of fast catalysts, we explore the use liquid phase, gel
36


64
molecule that does not have a mutation interfering with binding of the catalytic motif of
4up6 to the trans substrate, it may be able to cleave in trans.
The competition study described above was done at high concentrations of
enzyme (r-614) to favor inter-molecular association and trans cleavage. We performed
a similar competition study, but with concentrations of r-614 and excess competitor
sufficiently low as to favor uni-molecular cleavage. Under these conditions, association
of the competitor to the enzyme which is inter-molecular is less favored than uni-
molecular association of the enzyme with its own substrate motif. As predicted, the
addition of 3.0 nM d-lib61o2 to 0.3 nM r-614 had no noticeable effect on r-614 cleavage
(Figure 3-28).
d-614 Cleaves Faster in trans Than in cis: Single-turnover Kinetics
Substrate saturation experiments were performed with increasing concentration of
enzyme (d-614) to determine the kcat(bi) for the bimolecular reaction. Figures 3-29, 30,
and 31 show the foil progress curves for cleavage, initial linear portion, and enzyme-
saturation profile for experiments with various substrates (ang + ribose, r-lib61o2, or r-
614) cleaved by d-614. The calculated d is 39, 35, and 18 nM for r-primer, r-614 and
r-lib61o2 substrates, respectively. The kcat(bi), which is determined from the maximum
rate of cleavage with saturating enzyme, is 0.040, 0.034, and 0.025 hr'1 for r-primer, r-
614 and r-lib61o2 substrates, respectively.
While the Kd for each substrate was relatively low (18-39 nM), the maximum rate
was not attained until enzyme concentration reached approximately 2000 nM. Indeed,
the initial rates appear to be increasing slightly even above 2000 nM, indicating that the
64


182
Perrin, D.M., Garestier, T., Helene, C. (1999) Expanding the catalytic repertoire of
nucleic acid catalysts: Simultaneous incorporation of two modified deoxyribonucleoside
triphosphates bearing ammonium and imidazolyl functionalities. Ncleos. Nucleot. 18,
377-391.
Piccirilli, J.A., Krauch, T., Moroney, S.E., Benner, S.A. (1990) Extending the genetic
alphabet: Enzymatic incorporation of a new base pair into DNA and KNA.Nature 343,
33-37.
Reid, C and Orgel, LE. (1967) Nature 216,455.
Rich, A. (1962) On the problems of evolution and biochemical information transfer, in
Horizons in Biochemistry, Kasha, M. and Pullman, B. editors, N.Y., Academic Press,
103-126.
Robertson M, and Miller S. (1995) Science 268, 702-705.
Roth, A., Breaker, R. R. (1998) An amino acid as a cofactor for a catalytic
polynucleotide. Proc. Nat. Acad. Sci. USA 95, 6027-6031 .
Rupert PB, Massey AP, Sigurdsson ST, Ferre-DAmare AR. (1997) Transition state
stabilization by a catalytic RNA. Science, 298, 1421-1424.
Sakthivel, K., Barbas, C.F. (1998) Expanding the potential of DNA for binding and
catalysis: Highly functionalized dUTP derivatives that are substrates for thermostable
DNA polymerases Angew Chem Int. Edit. 37, 2872-2875.
Santoro SW, Joyce GF. (1997) A general purpose RNA-cleaving DNA enzyme, PNAS
94, 4262-4266 .
Santoro SW, Joyce GF, Sakthivel K, Gramatikova S, Barbas CF (2000) RNA cleavage
by a DNA enzyme with extended chemical functionality, JAM CHEMSOC 122, 2433-
2439.
Sclesinger G and Miller SL. (1983) Prebiotic Synthesis in Atmospheres Containing CH4,
CO, and C02, Hydogen Cynanide, Formaldehyde and Ammonia, J. Mol. Evol. 19, 383-
90.
Seelig, B and Jaschke A. (1999) Chem. Biol. 6 167-76.
Shih IH, Been MD. (2002) Catalytic strategies of the hepatitis delta virus ribozymes.
Annual Review of Biochemistry, 71, 887-917.
Stribling, R, and Miller, SL. (1987) Origins of Life 17, 261-73.


42
The amount of cleaved ang + cage (via base hydrolysis) at maximum de-protection
(100 pulses of laser) was consistently about 10-15% lower than the cleavage obtained for
ang + ribose. This 10-15% reduction in cleavage for ang + cage must be the result of
either two factors: incomplete de-protection or missing the caged-ribose adenosine.
Although cleavage of caged-614 plateaus with 100 pulse of laser, showing no further
cleavage even with 1000 pulses of laser, it is still possible that some caged-ribose
molecules have not been hit by a laser photon.
The other possible explanation for reduced cleavage of de-protected caged-614 is
the absence of the caged-ribose adenosine, again as a result of imperfect primer synthesis
and purification. As much as 6% of the ang + ribose primer is believed to be missing
the ribose-adenosine (above), but the caged-adenosine phosphoramidite may be
incorporated into the primer more poorly than the ribose-adenosine phosphoramidite,
resulting in as much as 20% of the ang + ribose primer missing the caged-adenosine.
This issue is revisited in a following section Cloning and Sequencing 614.
Kinetic profile of 614 with uncaged- and caged-ribose
The ang + ribose and ang + cage primers were incorporated into 614 via
PCR. The catalytic strand of the PCR products, called ribose-614 or caged-614,
were purified first by exonuclease degradation of the complementary strand (which was
5-phosphorylated), followed by gel purification (8% PAGE/urea). The ability of
ribose-614 and caged-614 DNAzymes to self-cleave in reaction buffer (1 M NaCl, 1
mM MgCb, 50 mM HEPES pH 7) was tested. Purified caged-614 was re-suspended in
reaction buffer, heated to 97C for 3 minutes and cooled to 23 C over 10 minutes
(together this is referred to as slow cool), and incubated at room temperature for
42


Distribution Linear
m vanes
RF = varies
L 0 20%
St = HEED min
RF m
case 1 lOOE-oe -4 soe-09
case 2 i ooe-oe -s.soe-os
case 3 i ooe-oe -4 soe-07
D
Population Distribution After:
Progress Curve for New Population After:
depend on the initial reactive fraction. Instead, the distribution of catalysts for stabalized populations mirrors the proportional relationship betweem catalysts in the initial population for all catalysts with rates significantly
faster than the selection time Simulations were performed for initial populations that varied In either the reactive fraction, the slope of the distribution of catalysts, or both The catalysts in case 1 and case 2 were distributed
according to identical linear functions (m -4.5E-9). but case 2 had 100 times more reactive molecules in its initial population. Not only did the selections for case 2 reveal catalysts earlier (B). but the distribution of catalysts in the
stabalized populations differed from the distribution for case 1 Case 3 differed from case 1 In both the number of reactive catalysts (RF = 1E*6 vs 1E-8) and in the distribution of those catalysts (m = -4.5E-7 vs -4.5E-9). but the
ratio (proportionality) of the amount of catalysts with rate = k(n) to catalysts with rate = k(n+1) (where n equals any given rate) was held constant (for a linear distribution, the proportionality relationship is m3 = ml RF3/RF1)
Case 3 reached a stabalized population earlier than case 1 on account of a 100-fold greater reactive fraction, but once stabalized. both case 1 and 3 had identical distributions of catalytic rates.
ON
On


Laser Does Not Damage Ribose-614 44
DNAzyme 614 is NaCl-dependent, and MgCh-independent 45
Understanding Reasons for Incomplete Cleavage of DNAzyme 614 46
Incomplete De-protection of Caged-Ribose 46
Incomplete Removal of the Complementary Strand 47
Approach to Chemical Equilibrium 48
Are the 27- and 79-nucleotide Fragments Acting as Catalysts or Inhibitors?. 50
Improperly Folded DNAzyme 614 51
Cloning and Sequencing Cleaved and Uncleaved 614 Near
Cleavage Plateau 54
DNAzyme 614 Cleaves in cis and trans 56
Cleavage Rate of Ribose-614 Varies with 614 Concentration 58
Deoxyribose-614 Cleaves Various Ribose-Substrates 62
Competition Studies of Ribose-614 Cleavage 63
d-614 Cleaves Faster in trans Than in cis: Single-turnover Kinetics 64
d-614 Cleaves with Multiple-turnover 68
Testing 614 Rate of Association and Disassociation 68
Structural Analysis of 614 74
Summary of Results 81
4 STUDIES OF IN VITRO SELECTIONS 123
In Vitro Selection of Functionalized and Non-Functionalized Libraries
Using Caged-Ribose and Liquid-Phase Selection 123
In Vitro Selection 123
Testing Kinetics from Round 8 Pools Without Caged-ribose 127
Testing for Inadvertent Selection for Susceptibility to Laser 127
Caged-ribose Does Not Induce Cleavage in trans 128
Kinetic Analysis of Rounds 1-8 Pools for H: ang + ribose 128
Sequence Evolution During I VS 131
Kinetic Analysis of Round 8 Clones 133
Simulations of In Vitro Selection Experiments 134
Simulating In Vitro Selections 135
Using Simulations to Determine the Impact of the Leakage (L)
Parameter on I VS 139
Examining the Impact of the Selection Time (St) 145
Examining the Impact of the Rate of the Fastest Catalysts (kfastest) 146
Examining the Impact of the Reactive Fraction 147
Comparing Linearly and Exponentially Distributed Initial Populations 148
Deciphering the Distribution Function for Catalysts 149
Summary of Results 152
5 SUMMARY 169
Catalytic Behavior of DNAzymes from IVS is More Complex
then Previously Assumed 170
v


154
round 0
round 1
LMNOABCD EFG
m ¡F mS ¡i
v v v v ~ v d
Q Q Q Q Q O Q
z z z z z z z
(N O ir, rj x r, C
"tncimnndo
q a a a a a o
z z z z z z z
ft
round 2
ABCD EFGH
round 3
^ ^ ^ ^
O fN OO (N
rf Tf rt
W
round 4 round 5
A BC DEFGH I JKLMNOa BCDEFGH I JKLMNO
h m is Ttqr-;qq Ttidindin-rinr¡nni:ddin'S:iniriidindo round 6 round 7
ABCDEFGH I JKLMNO ABCDEFGH I JK/L MNO
^ ^ ^
m3lfii!¡r)5¡^;0;5;f^f>^ d O' O' W fL o v¡
ddrir'r'i'doodirl'tsiTtdd -do>--dNmin/i ddd
Figure 4-1 Cleavage at the completion of each round of selection (time varies according to selection
condition). "Round 0" refers to the initial library before any rounds of selection. All selections were
exposed to laser at the beginning of each round of selection. The survivors of each round were
gel-purified and amplified for the next roundof selection. The amount cleaved for samples with
caged-ribose (A-F, l-M) increased progressively with each round. Selections G, N. and O never
increased. Selection H increased dramatically following the fourth round of selection. The percent
cleaved at the completion of each round is shown below each sample. Refer to Table 2-1 (page 34) for a
a description of each selection condition.


126
the BJ + de-protected cage primer (with standard and non-standard bases) continued to
have insignificant conversion to cleaved product.
Selection condition H, containing standard nucleotides with the ang + ribose
primer, began to show 3.7% cleavage after four rounds of selection. The amount of
cleaved product for selection condition H increased dramatically from 3.7% after 4
rounds of selection, to 8% after 5 rounds, and to 24.3% after seven rounds of selection.
The parallel selection to population H in which E-base was substituted for T-base had
no detectable cleavage even after seven rounds of selection.
The results reported above tell us only what fraction of the full length material
was cleaved at the time each round of selection was stopped. A more detailed
understanding of the progress of each selection can be gained by examining the progress
curve for cleavage. This was done by removing an aliquot of each selection condition at
various times (from 2 hours to 170 hours) and stopping the reaction by formamide stop
dye. A low signal-to-noise ratio made base-line determination difficult, and therefore
estimation of the percent cleaved difficult. Nonetheless, the presence of DNAzyme
enrichment in our library should be manifest by in an increase in cleaved product with
time, whereas noise should remain relatively constant over time.
Of the fifteen selection conditions, only selection condition H showed a
consistent increase in cleaved product with time. Other selection conditions had varying
levels of cleavage, but cleavage did not consistently increase with time. It is notable that
cleavage was not detected in samples G, N, and O, all of which had an unprotected ribose
at the time of laser exposure (either BJ + de-protected cage or ang + ribose primer).


Caged-ribose is Useful for Kinetics, But Increases Leakage
Too Much for IVS 173
IVS Isolates Sequence Variants of 614 with Wide Range of Catalytic Power 174
Simulations Allow Estimation of Catalyst Distribution 176
Nucleic Acid Catalysis and the Origins of Life 177
REFERENCES 179
BIOGRAPHICAL SKETCH 184
vi


63
614, although with varying degrees of efficiency. The observation that the single
nucleotide point mutation of 614, 4up6 (which is likely to fold similarly to 614), is
cleaved the slowest by d-614, while the ang + ribose primer is cleaved faster than other
full length molecules, agrees with the prediction that non-productive binding with self
can reduce binding and catalysis by d-614.
Competition Studies of Ribose-614 Cleavage
We tested the ability of the same three substrates (ang + ribose primer, ribose-
4up6, and ribose-lib61o2) to compete with r-614 for trans cleavage. The competitors
were synthesized with a deoxy-adenosine in the place of the ribose-adenosine, and added
in 9 fold excess over ribose-614. As a control, 9-fold excess of d-614 was added to r-
614. The addition of 270 nM d-614 to 30 nM r-614 increased cleavage to the same level
as r-614 at 300 nM as a result of d-614 cleavage of r-614 in trans (Figure 3-27).
The cleavage of r-614 was reduced by the addition of ang ribose primer and
d-lib61o2, showing that these molecules compete for r-614 trans catalysis. However,
the addition of 9-fold excess 614 mutant 4up6 to r-614 did not show the reduction in r-
614 cleavage seen by other competitors. Instead, the r-614 cleavage was increased,
although the increase was far less than that seen with the addition of d-614. Two
scenarios can explain this result. First, it is possible that although the catalytic activity of
4up6 is reduced, the reduction is offset by the great excess of trans acting 4up6.
Alternatively, it is possible that the mutation in 4up6 does not alter its catalytic motif
and activity, but instead the mutation prevents binding of the catalytic motif near its own
substrate. Thus, incubating 4up6 alone shows almost no catalytic activity because the
active catalytic motif cannot bind near its own substrate, but if incubated with another
63


108
s

o
c
4>
o
k-
a>
a.
70
60-
50-
40-
30
20-
10-
iJ
100
I
200
300
400
400 nM r-lib6to2 +
100 nM d-614,
mixed before stow cool
400 nM r-lib6to2 +
100 nM d-614,
mixed after stow cool
133 nM r-lib6lo2 +
33 nM d-614,
mixed before stow
cool
400 nM Mib6lo2 +
100 nM d-614,
mixed after stow cool
¡00
hours
B
400 nM r-lib6k>2 +
100 nM d-614,
mixed before slow cool
400 nM r-lib6lo2 +
100 nM d-614,
mixed after slow cool
133 nM r-lib6lo2 +
33 nM d-614,
mixed before slow
cool
400 nM r-lib6lo2 +
100 nM d-614,
mixed after slow cool
Figure 3-33. Multi-turnover kinetics for d-614 cleaving r-lib6lo2 is not affected
by the slow cool folding protocol, and indicates two turnovers in 200 hours. (A)
Percent of r-lib61o2 substrate cleaved by d-614. (B) The absolute amount of substrate
cleaved. The enzyme and substrate were mixed together either immediately before or
immediately after the slow cool protocol.


70
prior to or subsequent to the slow cool folding protocol (data not shown). This
suggests that, although the slow cool folding procedure enhances c/s-cleavage, it has
little effect on trans-cleavage (which, because it is faster, predominates the initial rate).
The impact of the slow cool folding protocol on inter-molecular folding was
also tested directly by mixing the enzyme (d-614) and substrate (either r-lib61o2 or ang
+ ribose primer) either before or immediately after the slow cool procedure (Figure
3-33 and 34). Multiple turnover experiments showed very little difference in cleavage,
either in initial rates or cleavage plateau. In fact, completely omitting the
denaturing/slow cool protocol had no effect on trans-cleavage (Figure 3-38). This further
demonstrates that inter-molecular folding is not significantly effected by the slow cool
protocol.
To test whether the rate-limiting step for ribose-614 trans-cleavage is the inter-
molecular association or the chemical step, a chase experiment was performed with cold
competitor. Ribose-614 was incubated at sufficiently high concentration (200 nM) as to
achieve significant trans-cleavage. The sample was allowed to fold for an additional
thirty minutes after slow cooling, at which point 15-fold excess unlabelled chase (ang
+ ribose) was added. A significant decrease in cleavage was observed immediately
following the addition of the cold chase (as compared to ribose-614 without added chase)
(Figure 3-39). Comparing 200 nM ribose-614 plus chase with 2 nM ribose-614
without chase reveals that the chase has reduced cleavage levels below even that seen
for a nearly all cis-cleavage reaction (Figure 3-40).
If the rate of substrate disassociation from enzyme is faster than the cleavage step
(kcat(bi) k-i), the addition of a cold chase should completely eliminate subsequent
70


91
A B
Figure 3-12. Ribose-614 cleavage is unaffected by the addition of 10%
complementary strand, but addition of equimolar amounts greatly lowers
cleavage rate and plateau. (A) Full time course, and (B) initial time course, for
ribose-614 with 10% added complementary strand. (C) Full time course, and (D)
initial time course, for ribose-614 with equimolar added complementary strand.
Ribose-614 was incubated with exogenously added complementary strand at either
10% of the total amount of ribose 614, or at equimolar amount. Each reaction was
done at high (115 nM) and low (11.5) nM ribose-614 concentration and compared
against pure ribose-614 at a range of concentrations.


46
Supplementing the standard reaction buffer with 2 M ammonium acetate increases
the rate of cleavage. Replacing the 1 M NaCl of the reaction buffer with 1M ammonium
acetate resulted in cleavage, although at a greatly reduced rate.
Understanding Reasons for Incomplete Cleavage of DNAzyme 614
Interpreting and modeling the results of an I VS requires a comprehensive
understanding of the mechanics of the model system. The previous section quantified the
experimental parameters of using a caged-ribose that is de-protected by laser. Analysis
of caged and ribose-614 both demonstrated a cleavage plateau far below true
completion (defined as cleavage caused by strong base, which is -94% for ang +
ribose and -80% for ang +cage). The explanation for the cleavage plateau must be
common for both caged and ribose-614, and quite likely any other DNAzyme generated
via this I VS procedure. Indeed, low cleavage plateau is an issue with many ribozymes
and DNAzymes, but the specific causes for this reduced plateau is often ignored,
presumably because the researchers either do not realize, or do not believe this
phenomena is significant. We now seek to explain the failure for the reaction to reach
true completion, knowing that such information is crucial to both redesigning the IVS
protocol and creating a reasonable model of the IVS experiment.
Incomplete De-protection of Caged-Ribose
We have already demonstrated that we have achieved maximum de-protection of
the caged-ribose with 100 pulses of laser (meaning additional laser does not result in
more de-protection). But is incomplete de-protection, even at maximum de-protection, a
46


171
the reactive folded form is greater than the impact on the rate constant for the chemical
step.
The fact that the balance between uni- and bi-molecular interaction is
concentration dependent may explain the observation by Geyer and Sen (1997) and
Breaker and Joyce (1995) that cleavage rates decreased when DNAzymes produced with
additional cycles of PCR (20 vs. 5) were examined. Presumably because they considered
only unimolecular interactions, these researchers did not report the concentrations of
DNAzymes in their assays (and presumably they did not control DNAzyme
concentration). We predict, therefore, that the decrease in rate with increasing cycles of
PCR resulted from an increased concentration of their DNAzymes. If kbi < kuni for their
particular DNAzymes, higher concentrations would favor a slower rate process.
Although divalent cations can affect the folding properties of nucleic acids,
cofactors (such as Mg++) and functionalized nucleotides are often assumed to enhance the
rate of the chemical step. The observation that the rate limiting step for 614-catalysis
includes a folding/association step may explain why Mg++ is not required for catalysis. It
remains curious that a selection process would generate Mg-independent catalysts given
the availability of Mg++. This suggest that the probability of find a better catalyst that
uses Mg-H- is less than the probability of find a catalyst that does not. This also suggests
that, in order for cofactors (or covalently attached functionalities) to improve the rate of
catalysis, the rate limiting step must not be the folding of the DNAzyme, but instead the
chemical step. The impact of the chemical step can be increased relative to the folding
step by increasing the cofactor concentration, or increasing the binding strength of ES or
E*cofactor.


150
half lives < St/6). If the selection-progress plot indicates a plateau below 95% cleavage,
the stabilized population contains only the fastest molecules. How do we use the
information gained from these simulations to estimate the population distribution for a
real library undergoing selection?
The first thing we must consider is the range of rates we are interested in
estimating. Unless significant activity can be detected in only a few rounds (less than
three), nearly all molecules with half lives greater than the St will be completely lost and
have little effect on the outcome of the selection procedure. (Although slow catalysts
may be completely lost, their initial abundance will compete with faster catalysts for PCR
amplification, thus diluting the relative abundance of the faster catalysts and slightly
delaying their detection. Unless these catalysts with half lives close to, but above St, are
in enormous excess over catalysts with half lives just below St, this delay is unlikely to
make any difference. Nonetheless, the play forward model described below considers
this minor effect).
Choosing selection time near the half life of the fastest molecules will isolate only
the fastest molecules, while choosing a longer selection time results in a stabilized
population which contains a broad range of catalysts with rates less than the St.
Practically speaking, multiple selections should be performed with various selection
times to determine both the fastest catalysts in the population and to isolate a range of
catalytic rates.
We can take advantage of the observation that there is no selective enrichment for
catalysts with half lives greater than selection time, and therefore the distribution of
catalysts with half lives less than St/6 is proportional to their proportion in the initial


9
materials, most notably HCN (Miller and Urey, 1959; Sclesinger and Miller, 1983;
Stibling and Miller, 1987). The prebiotic presence of HCN is further supported by its
presence in intrastellar clouds, in comets, and in the atmosphere of Titan (Irvine, 1999;
Huebner, et al, 1974; Hanel, 1981).
Base-catalysed tetramerization of HCN yields diaminomaleonitrile, which then
leads to the production of adenine, hypoxanthine, guanine, xanthine, and diaminopurine.
Hydrolysis of HCN oligomers yields amino acids, particularly glycine, alanine, aspartate,
and diaminosuccinate, among other amino acids (reviewed in Joyce, 1989). Other
electric discharge experiments under presumably prebiotic conditions generated purines,
among other potentially useful materials. These experiments usually have low yields of
pyrimidine synthesis, but recent experiments of Miller demonstrate that significant
pyrimidine yield from frozen dilute samples of NH4CN (Miyakawa, et al 2002).
Another prebiotic molecule, H2CO, can lead to glyceraldehydes through
condensation in the presence of a catalyst such as calcium carbonate or alumina (Gabel,
1967; Reid, 1967). Glyceradehyde can then begin a cascade that converts formaldehyde
into trioses, tertroses, and larger sugars through a process of aldol condensation and
enolization (Joyce, 1989).
Joining nucleoside and sugar subunits has been accomplished by heating a
mixture of ribose and purine to dryness in the presence of inorganic salts; this process,
however, leads to a mixture of a- and P-nucleosides (Fuller, et al, 1972).
In summary, experimental evidence suggests that the chemical environment and
composition of the prebiotic earth is sufficient to provide the necessary mechanisms and
materials for the synthesis of many potentially useful monomers. These findings are
9


48
and the full length DNAzyme ribose-614 is the maximum amount attributable to
incomplete degradation of the complementary strand by exonuclease.
Exonuclease efficiency was also tested by 5-phosphorylating a single primer with
gamma-32P-ATP. This radio-labeled primer was then used to synthesize full length
ribose-614. The double stranded product was then divided into two aliquots, and one
was subjected to standard exonuclease treatment. Equal amounts of each sample were
then electrophoresed on a PAGE-urea gel, and the amount of full-length product
assessed. Exonuclease treated samples consistently showed less than 10% of the original
material remained full length.
Experiments in which small amounts of the 614 complement (10% of the amount
ribose-614) were added to ribose-614 had no impact on the cleavage plateau (Figure
3-12A and B). When cleavage experiments were performed with equal amounts of
ribose-614 and its complement, however, the presence of the complementary strand
dramatically reduced the cleavage plateau to nearly 30% (Figure 3-12 C and D). This
demonstrates the importance of near-complete removal of the complementary strand to
accurately examine DNAzyme kinetics. We believe both asymmetric PCR and
exonuclease degradation are superior methods to column purification under strong base
for removing the complementary strand.
Approach to Chemical Equilibrium
The failure of a substrate to completely transform itself to product may be the
result of an approach to chemical equilibrium. In this case, the reverse reaction occurs
fast enough to convert a significant fraction of the product back into substrate. We tested
48


44
increasing amounts of laser did not cause an increase in cleavage plateau (Figure 3-6a
and 7). When the percent of caged-614 product was normalized to account for the
incomplete de-protection at low levels of laser (the fraction cleaved at each time point
was divided by the fraction deprotected as determined by base hydrolysis), the kinetic
profiles for caged-614 with varying amounts of laser overlap (except, of course, zero
laser, which is not expected to cleave at all). This shows that they are all following the
same kinetics parameters, with the plateau varying with the amount of laser (with
maximum achieved at 150 pulses of laser).
The caged-614 exposed to increasing laser was also incubated under base
hydrolysis conditions. The amount of caged-614 cleaved by strong base increased with
increasing laser, reaching a cleavage plateau with 150 pulses of laser. Added laser
beyond 150 pulses did not further increase the amount of base-cleaved caged-614 (Figure
3-8). This directly parallels the results of caged primer, ang + cage, plus laser cleaved
by strong base condition, indicating that maximum de-protection has be obtained.
Laser Does Not Damage Ribose-614
It is possible that caged-614 shows a lower cleavage plateau than ribose-614
because the laser (UV frequency) used to initiate the de-protection of the caged-ribose
causes damage to the DNAzyme. Non-specific breaking of the phosphodiester backbone
is expected to create a smear running from 1 to 106 nucleotides. Although slight
smearing was noticed between the cleaved and full length product, there was no smearing
below the full length product (79 nucleotides). Furthermore, this smearing was not
present when ribose-614 was exposed to identical laser treatment.
44


120
A4 + r-pr¡mer
fv %
M }
V 4,
o? t',,5'-e
-a. 614 traM
I. HMrt 4 a. Mi-
A2/A3 trans + r-pr¡mer
-at. .x0
,--^r
( ;
* --? 614 ti
i-
. vi- iS'-s
V- /r
A/C2 + r-primer
,/Q,
5'-E
.r B2 + r-pr¡mer
o,,1
SA
O -
SX
{
;5''S
v> -J V "
a V<
/O
rf l .
o
-l*.4 (14 f
B2 + r-primer
-CATG interaction
Ox "A
xO
X-
as'-
r
J
L 'S'-E
u
-14.2 (14 trm
46 a -W.4 614 Iran
Figure 3-46. Several structural models for r-614 cleaving r-primer in trans. The
two experimentally supported trans-binding structures are shown in colored boxes.
ir io
The red box highlights the interaction between the CATG region of the primer
substrate (S) and the same sequence motif of 614 (E). The yellow box highlights the
interaction between l7CGACTCACTAT27 of the primer substrate and
s5GTAGTGACG43 of 614. The red arrow marks the ribose-A site of cleavage.


Abstract of Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy
SEARCHING MOLECULAR LANDSCAPES FOR THE EVOLUTION OF PRIMAL
CATALYSTS: IN VITRO SELECTION OF DNA-BASED RIBONUCLEASES
By
Matthew A. Carrigan
December 2002
Chair: Steven A. Benner
Cochair: Art Edison
Major Department: Neuroscience
Our current understanding of the origins of life suggests that catalytic nucleic
acids must arise from random polymers of nucleic acids. Searching random libraries of
nucleic acid polymers has not produced the abundance and power of catalysis believed
necessary to spawn life. Further, the addition of chemical functionality to random
libraries has failed to improve the catalytic potential of random libraries as anticipated.
This apparent discrepancy between observation and expectation may be
attributable to several shortcomings of the experimental design. Active catalysts may be
lost during the preparation of libraries, or during the enrichment for catalysts.
Furthermore, only incomplete descriptions have been attempted for any, focusing only on
the fastest catalysts in the library, rather than capturing completely the distribution of
catalytic power. It is conceivable that added functionality, or alterations of the library or
IX


55
purification, only three (11%) were found to contain a mutation (all three were in the N40
region between the primers).
Of the 65 clones of 614 that were un-cleaved, 44 had at least one mutation (68%).
Twelve clones (18% of those sequenced) had a mutation in the N40 region; thirty-one
clones (46%) had a mutation in the ang primer; three clones (5%) had a mutation in the
complementary primer. Of the uncleaved-614 clones sequenced, nine were from a 614
DNAzyme that originally contained a caged-ribose (cleavage initiated by laser). Of these
nine clones, five (56%) were shown to be missing the ribose-adenosine, as opposed to
seven of 56 clones (13%) isolated from uncleaved ribose-614. This confirms our original
suspicion that a higher proportion of the ang + cage primer is missing the ribose-
adenosine as compared to the ang + ribose primer, and explains both the observation
that about 15% less of the ang + cage primer with maximum de-protection is cleaved
by strong base than the ang + ribose primer, and the observation that caged-614 reaches
a cleavage plateau about 10% lower than ribose-614.
A caveat must be added, however, in that it has been observed that the polymerase
tends to terminate primer elongation about 50% of the time when it encounters the caged-
ribose-adenosine on the template strand. Given this, it is possible that the polymerase
skips over the caged-ribose-adenosine in the fraction that does not get terminated at the
caged-ribose, and thus the full-length double stranded DNA that is required for cloning is
missing the adenosine as a result of the polymerase skipping the caged-ribose. For this to
occur, the caged-ribose must not have been fully de-protected by the initial laser
treatment (polymerase does not terminate appreciably at an uncaged-ribose). Together,
this confirms that the lower cleavage seen for caged-614, as compared to ribose-614, is a
55


119
hours
r-614 (wt) [108nm]
r-614mut (C72T)
a r-614mut (C72T & G81A) aug14
r-614mut (G55A)
o r-614 (G55A & C47T)
* r-614mut (G65A)
A r-614 (G65A & C40T)
Figure 3-45. Compensatory mutations predicted by various structures do not
improve catalysis of 614 mutated at nucleotide 55, 65, or 72.
Several single nucleotide mutations reduced cleavage rates. Attempts to improve
cleavage rates by making compensatory mutations predicted by several structural
models were unsuccessful.


Population Distribution After:
ROUND 3 ROUND 4 ROUND 5 ROUND 6 ROUND 7 ROUND 8 ROUND 9
Figure 4-10. Increasing the amount of "leakage" delays the isolation of catalytic molecules. In vitro selections were simulated using a selection time (St) of 300 minutes for a
population of catalytic molecules with initial abundance set to one active molecule per 10(7) inactive molecules (RF = 10(-7)). The active molecules were discretely distributed over a
range of initial rates according to a linear relationship with m = 0 (each discrete rate category had the same number of molecules in the initial population). Simulations reveal that varying
the amount of leakage (cleavage independent of intrinsic rate) from 0 to 4% dramatically delays isolation of molecules Nonetheless, repeated selection eventually leads to nearly identical
selected populations (in this case, by round 8), regardless of the amount of leakage. Dashed vertical line corresponds to the selection time. (A) Initial population distribution. (B) Percent
cleaved at the end of each round of selection. (C) Percent cleaved at t = 81 minutes (D) Population following each round of selection. (E) Progress curve follow each round of selection.
On


3
information (requiring duplication efficiency/accuracy be greater than the rate of
degradation).
When the physical properties of a system are determined by the information it
contains, it follows that the physical properties of the system can change according to
changes in the encoded information. Such a system may never settle into equilibrium (as
physical systems are expected to eventually reach). Instead, imperfect information
duplication and repair changes the physical properties of a system. This eventually
results in multiple variants of an information-containing system, each with variable
physical properties. When two such systems utilize the same resources (energy rich
molecules, elemental precursors, etc) to prevent information decay, competition exists
among systems for more efficient use of the resource.
Despite competition among systems for common resources, the ultimate survival
pressure is against the progression towards information loss. A system can therefore
persist by either out-competing rivals for a common resource, or by occupying a new
niche (utilizing a different resource). Given the constant progressive force towards
information decay, there exists selection and enrichment for systems that enhance the
survival of their information. As multiple systems compete for resources, information
systems linked to replication-enhancing functions will predominate.
From this one can see that Darwinian evolution (which is the fundamental trait of
life) is the direct consequence of physical laws applied to an information-containing
system. For this reason, some have argued that once a minimal system (hypercycle)
capable of Darwinian evolution exists, it will generate higher orders of complexity
3


22
Our eventual goal is to estimate catalytic distribution in a random library, and use
this estimate to determine which factors enhance the catalytic power of a library. To make
this estimate will require a realistic mathematical simulation of the selection system
which, in turn, requires detailed knowledge of the catalytic process. Towards these goals,
research in this dissertation seeks a detailed understanding of why the phosphodiesterase
activity isolated using the Breaker-Joyce procedure routinely reaches a low plateau. It is
also crucial to determine whether the reaction behaves as a first order chemical process (as
predicted).
Conditions for the use of caged-ribose have been optimized, allowing reaction
during the set up to be prevented. Preliminary selections with the caged-ribose indicate
that the laser photolysis of the caged-ribose creates a slight increase in background
cleavage, both experimental and simulated selections indicate that this leakage
dramatically increases the number of rounds of selection required to isolate active
catalysts. Although this limits the utility of caged-ribose for selections, the background
does not significantly alter kinetic analysis, allowing its use for studying the pre-activation
requirements of DNAzymes, such as folding.
To aid in the development of realistic models, we undertook a detailed study of an
individual DNAzyme, 614. This revealed a number of significant findings. It is widely
assumed that including a cofactor (such as Mg++ or other divalent cations) into the
selection buffer will result in the selection for cofactor-dependent catalysts. Indeed, the
Breaker-Joyce protocol presumes that potential catalysts are completely inactive without
the added cofactor. We have determined that, although 614 was selected in the presence
of Mg++, 614 retains complete catalytic activity in the absence of Mg++. It is also widely
22


159
A
temperature (C)
Figure 4-6. The cleavage rate for Selection H following 8 rounds of selection is
unchanged by incubating at 15C, but is slightly lower at temperatures below
15C. The rate drops quickly when the temperature is increased above 25C. (A) Full
time course. (B) Plot of initial rate vs. temperature.


176
Simulations Allow Estimation of Catalyst Distribution
The impact of various selection parameters and library parameters was examined
with simulations of the IVS procedure. When the selection time (St) is very short
compared to the half-life of the fastest catalysts in the initial library, the stabilized
population is cleaved less than 95% at the completion of each round of selection,
meaning that the population contains only the fastest catalysts. This scenario describes
the selection regime of most in vitro selections, and thus they have focused only on the
rate of the fastest molecules.
If functionality were to dramatically improve the abundance of catalysts with
rates slower than kfastest, then these previous experiments only allow a comparison of the
fastest molecules. The origin of life does not likely depend on the spontaneous
generation of a relatively rare, fast catalyst, but rather on the spontaneous and relatively
frequent generation of any catalysts. Thus, a full description of the distribution of fast
and slow catalysts in a random library is needed if we are to understand how to improve
the catalytic power of the library is. The simulation reveals a method for doing just this:
Performing selections with a range of selection times allows estimation of both
fast and slow catalyst. When the St~ kfastest, only the fastest molecules are
isolated. Increasing selection time such that St kfastest creates a stabilized
population such that the proportionality between fastest and slower catalysts after
selection remains similar to the initial population. This allows estimation of the
initial population distribution by direct examination of the selected population.
The population of catalysts in initial library can be described sufficiently with two
parameters, the Reactive Fraction (RF) and the Distribution Function. The


129
synthesized with an unprotected ribose, and therefore laser was not used to initiate
cleavage. Pools from each of the eight rounds of selection were prepared at three
concentrations: 3 nM, 15 nM (approximately the concentration used during the IVS), and
60 nM. Figure 4-4 shows the full kinetic profile for the pool of molecules following each
round of selection. From this it can be seen there is no detectable activity in the pool of
molecules following only one round of selection. A slight increase in pool activity is
seen following only the second round of selection, and this activity dramatically increases
progressively after third, fourth, and fifth round of selection. By the completion of the
fifth round of selection, the activity of the pool had reached an apparent maximum, after
which further selection no longer increased the rate of product formation from the pool.
Figure 4-5 shows the initial rates of cleavage for each of the three dilutions. In all
cases, a slight increase in cleavage was seen with each successive increase in
concentration. This indicates that cleavage rate of these pools is occurring in trans, but
the amount of trans cleavage must be low as the rate increases no more than 3 fold when
the concentration increases 20 fold.
Pools from H round 8 where tested over a temperature range to determine
whether the rate limiting step was temperature dependent. Figure 4-6 shows that
cleavage rates are unchanged by a drop of temperature from 25C to 15C. Further
lowering the temperatures lowers the rate of product formation, although less than
expected for a reaction which is rate-limited by the chemical step. Like 614, the rate of
cleavage for the pool of molecules from selection H following 8 rounds of selection
appears to reflect a balance between the association step and than the chemical step.


41
plus tail, and a 94-nt band resulting from termination of the complementary strand plus
tail at the caged ribose.
DNAzyme 614 with uncaged- and caged-ribose
Achieving maximal deprotection of caged ribose
A caged-ribose was synthesized and incorporated into the catalytic strand primer,
known as ang + cage (an identical primer was synthesized with an unprotected ribose
known as ang + ribose. Using this caged-ribose, we are able to prevent cleavage at the
ribose during the work up phase. This was demonstrated by incubating the caged-ribose
primer, ang + cage, in 0.5M NaOH at 80C for one hour, resulting in only 6-10%
conversion to the cleaved form (via base hydrolysis); the same conditions for the
unprotected-ribose primer, ang + ribose, resulted in 94% (90-97%) cleavage. The
observation that about 6% of the unprotected-ribose primer does not convert to the
cleaved form under base hydrolysis conditions suggest that as much as 6% of the ang +
ribose primer may be missing the ribose-adenosine (due to imperfections in the DNA
synthesis and purification).
To test the ability to de-protect the caged-ribose, ang + cage was exposed to
increasing number of pulses of excimer laser (frequency = 308 nM, energy = 30-50
mJ/pulse) to induce conversion from the protected to unprotected ribose. The efficiency
of conversion to unprotected ribose was assessed by incubating under base hydrolysis
conditions (Figure 3-4). The amount of caged-primer cleaved by strong base increased
with increasing laser, reaching a plateau of 80% cleaved with 100 pulses of laser. Added
laser beyond 100 pulses did not further increase the amount of cleaved caged-614.
41


31
after which the sample was centrifuged for 15 minutes at 16,000 g. The supernatant was
removed and the pellet rinsed with 70% ethanol (200 pL). The ethanol was removed and
the pellet was dried completely under vacuum. The resulting DNA pellet was re
suspended in 20 pL of water. This protocol routinely yielded 100 ng/uL of DNA of
sufficient purity for sequencing reactions.
Sequences transformed into the TOPO TA Cloning Vector were sequenced
using 3.2 pM primer 1224 (5-CGCCAGGGTTTTCCCAGTCACGAC) with 300-500
ng plasmid DNA template, 2X final concentration Big Dye reaction buffer and 2 pL Big
Dye Terminator Sequencing mix in a final volume of 10 uL. Samples were overlayed
with mineral oil and amplified using 25 cycles of PCR at 96C for 30 seconds, 50C for
15 seconds, and 60C for 4 minutes. Unincorporated nucleotides were removed by
spinning 9.5 pL of the recovered PCR reaction through a column of 400 pL G-25
Sephadex at 800 g for 2 minutes. The recovered sample was then dried to completion, re
suspended in 17 pL formamide sequencing buffer, heated to 95C for 2 minutes and
quickly transferred to ice. Sequencing samples were analyzed on an Applied Biosystems
Prism 310 Genetic Analyzer. Sequencing results were confirmed by examining the
chromatograms manually using the Sequencher software package.
In Vitro Selection
DNAzyme libraries for the first round of selection were prepared by a single cycle
of run-off PCR using the library template (5-GTGCCAAGCTTACCGTCAC-N40-
GAGATGTCGCCATCTCTTCC (where N indicates equal molar concentrations of A, T,
G, and C) and one of three different catalytic strand primers: ang + ribose, BJ +
cage, or BJ + de-protected cage (the BJ + de-protected cage was the same primer
31


125
Although cleaved products were undetectable, samples were run on a PAGE/urea
gel and the size corresponding to cleaved product excised. These gel-purified fragments
were used as templates to re-generate full length DNAzymes for the next round of
selection (indicating that small amounts of cleaved product had in fact been produced and
isolated). More radiolabel was used for each sample in all subsequent rounds, increasing
the signal to noise ratio. Samples A F all show a noticeable increase in percent of full
length sample converted to cleaved product (-3-4%) following just one round of
selection. All these samples contained the BJ + cage primer with standard nucleotides,
and varied only in either the buffer composition or the selection time. Other selection
conditions, including standard nucleotides with BJ + de-protected cage or ang +
ribose primers, and all samples containing E-base, had undetectable levels of cleaved
product. After the completion of two rounds of selection, however, selection conditions
containing E-base with BJ + cage primer (samples I M) began to show 2-4% cleaved
product; parallel selection conditions containing standard nucleotides had slightly higher
levels of cleaved product, -5% (this slight increase is possibly the result of less
smearing of the samples containing standard nucleotides). Selection conditions with
de-protected caged-ribose and uncaged-ribose, with either standard or non-standard
bases, still showed no significant cleavage (G, H, N, O).
The percent of full length substrate converted to cleaved product for each sample
A -F (containing BJ + cage with standard bases, regardless of the buffer composition
and selection time) continued to increase slightly with each progressive round of
selection. By the completion of 7 rounds of selection, as much as 11% of these samples
were converted to full-length product. During the same time, selection conditions with


20
catalyst. There are several technical weaknesses of the Breaker-Joyce I VS protocol that
may result in the loss of desired catalysts.
First, columns have poor resolving ability, in part because they have great
potential for residual retention (unbound material slowly filtering through the column)
and leakage (bound material becoming unbound). It is also assumed that even the best
catalysts remain inactive until the reaction is started by adding MgCh (or other
cofactor, such as histidine) and collection begins. Magnesium is in fact present during
the preparation, including during the PCR, and other salts are also present during the
ethanol precipitation potentially permitting catalysis during the preparation steps and
resulting in loss of catalysis. DNAzymes have been isolated using the Breaker-Joyce
protocol that are active without any added cofactor (requiring only NaCl), offering
support to the hypothesis that active catalysts may be lost during the workup.
Second, the activity of a DNAzyme is inhibited by the complementary strand, and
therefore must be removed. This is accomplished in the Breaker-Joyce protocol by
denaturing the double-stranded DNA with a strong base solution (NaOH), and then
washing the non-biotinylated complementary strand from the column. Strong base can
itself cause sequence independent cleavage of RNA, and is known to increase the rate of
many DNAzymes. It is therefore possible that many DNAzymes, in particular the fastest
DNAzymes, are lost through this base-catalyzed cleavage of RNA. The strong base may
also destabilize the binding of the catalysts via biotin to the streptavidin column, resulting
in further loss of potential catalysts.
Lastly, DNAzyme folding is largely ignored. After flushing out the NaOH with
reaction buffer (minus the required cofactor), the reaction is started by the addition of
20


28
caged-ribose in the catalytic primer and a 15 nucleotide tail on the complementary primer
generated a complementary strand that was either 121 nucleotides (106 nucleotides plus a
15 nucleotide tail) or 94 nucleotides (106 nucleotides plus a 15 nucleotide tail minus 27-
nucleotides following the caged-ribose). The desired catalytic strand was easily
identified as the middle band and cleanly excised on an 8% PAGE/urea gel. Gel purified
samples were crushed, soaked overnight in elution buffer, and then extracted in
phenol/chloroform/isoamyl alcohol, extracted again in choloroform/isoamyl alcohol, and
precipitated in ammonium acetate and ethanol.
5-End Labeling of DNA
Single-stranded DNA (20 pM )was 5-labeled with 20 pCi gamma-32P-ATP using
10 units of T4 polynucleotide kinase, 70 mM Tris-HCl (pH 7.6), 10 mM MgCE, 5 mM
dithiothreitol in a final volume of 10 pL. Reactions were incubated for 30 minutes at
37C followed by addition of equal volume TE, and finally inactivated by incubating for
20 minutes at 70C. End-labeled DNA was separated from unincorporated nucleotides by
spinning through a G-25 column at 600 g for 3 minutes.
DNAzyme Kinetic Assays
Following gel purification of single-stranded DNAzymes and substrates, samples
were ethanol precipitated and re-suspended in 50 mM HEPES buffer. Samples were then
transferred to a new tube and the concentration estimated by measuring specific
radioactivity following a measurement of by Cherenkov radiation. Samples were then
diluted to twice the desired final concentration with additional 50 mM HEPES buffer.
For trans assays, enzyme and substrates were typically mixed together (unless otherwise
28


148
distribution of those catalysts (m = -4.5 x 10'7 vs. -4.5 x 10'9), but the ratio
(proportionality) of catalysts with rate = kn to catalysts with rate = kn+i (where n equals
any given rate) was held constant (for a linear distribution, the proportionality
relationship is = (m^RFs) / RF¡).
Case 3 reached a stabilized population earlier than case 1 on account of a 100-fold
greater reactive fraction, but once stable, both populations had identical distributions of
catalytic rates. The reactive fraction therefore alters the number of rounds of selection
necessary to achieve stabilized populations, but the distribution of catalysts for these
stabilized populations does not depend on the initial reactive fraction. Instead, the
distribution of catalysts for a stabilized population mirrors the proportional relationship
between catalysts in the initial population for all catalysts with rates 6-fold faster than the
selection time.
Comparing Linearly and Exponentially Distributed Initial Populations
This prompted another simulation in which the RF in the intial library was
distributed according to a decreasing exponential function with b = 25 (Figure 4-14, case
3). Catalysts in case 1 and 2 were distributed according to a linear function (with m = 0
and -1.2 x 10'6, respectively) as modeled in earlier simulation. Although case 1 and case
2 approach population stability in nearly identical manners, the difference between their
population distributions can be detected by examining cleavage profile at early times.
The distribution of catalysts in the stabilized population differs dramatically for each test
case. In each case, the proportionality between catalysts with half lives less than St/6


96
a d-614wl [200nM] + Mib6lo2 [7.5nM]
- d-614mut A box
d-614mut 3* (G86C, G88C, T90A)
4 o d-614mut 6/10#6 (T27-)
A d-614mut 6/10*9 (A26-)
V d-614mut 4up7 (C24-, A28-. A32-)
d-614mut mcA' (26insA27)
X d-614mut 6/10/#2 (G38A) [100 nM] +
r-lib6lo2 [1.5 nM]
o r-614mut (G86C, G88C. T90A) (100nM) +
r-lib6k>2 [1.5 nM]
a d614wt
lib6lo2mut (C22G. C24G, A26T) (20 nM]
A
250
B
d-614wt [200nM] + Mib6k>2 [7.5nM]
d-614mut A box
d-614mut 3 (G86C, G88C, T90A)
d-614nnut 6/1006 (T27-)
d-614mut 6/1009 (A26-)
d-614m ut 4up7 (C24-, A 28-, A32-)
d-614mut mcA' (26insA27)
d-614mut 6/10/#2 (G38A) [100 nM] +
Nib6lo2 [1.5 nM]
r-614mut (G86C, G88C, T90A) [100nM]
r-ttb6lo2 [1.5 nM]
d614wt
Mb6lo2mut (C22G, C24G. A26T) [20 nM]
Figure 3-19. Mutants of 614 have reduced activity when cleaving r-lib6lo2 in
trans. (A) Complete time course; (B) Linear initial phase of time course. The ability
of several 614 mutants to cleave in trans was tested by incubating 614 variants (200
nM, unless noted differently) with r-lib61o2substrate (7.5 nM, unless noted
differently). Most mutations greatly reduced catalytic activity when compared to
wildtype 614. Nucleotides 6-11 were changed to AGTACT to created-614mut A
box.


CHAPTER 4
STUDIES OF IN VITRO SELECTIONS
In Vitro Selection of Functionalized and Non-Functionalized Libraries Using
Caged-Ribose and Liquid-Phase Selection
In Vitro Selection
Fifteen in vitro selections were performed in parallel, each with a variation in
either the primer sequence (ang vs. BJ), ribose protection (caged vs. unprotected
free ribose), metal cofactor in the buffer (none, zinc, or magnesium), the number of
molecules sampled in the starting library (5x10 vs. 5 x 10 ), or the presence of
functional group on a nucleic acid (E-base vs. standard T-base). The selection was
performed as described in Materials and Methods. Only one condition, standard
nucleotides using ang + ribose primer (with unprotected-ribose) generated active
DNAzymes capable of phosphodiester cleavage.
For the initial round of selection, six starting libraries were generated (varying in
catalytic primer sequence, ribose protection, and nucleotides) via a single cycle of primer
extension (internally labeled using alpha-32P-CTP). These libraries were divided into the
fifteen selection conditions varying in buffer composition and selection time (see Table
2-1). All samples were then exposed to 150-200 pulses of laser (80-100 mJ/pulse, the
number of laser pulses was varied with laser power so that the total energy was held
constant at 15,000 mJ), effectively de-protecting samples containing caged-ribose. Pools
were then allowed to react for 20, 120, or 720 minutes (see Table 1), at which point they
123


29
noted), and diluted to twice the desired final concentration. Samples were then mixed
with equal volume 2X reaction buffer (typically 2M NaCl, 2 mM MgCb, 50 mM HEPES
pH 7). Reactions were then heat denatured and refolded using a thermocycler slow
cool protocol (heating to 96C for 3 minutes, and cooling to 23C over 10 minutes). For
reactions with ribose-adenosine, time zero was considered to be the time at which the
sample completed the slow cool. For reactions with a caged-ribose, the reactions were
initiated when the laser de-protected the caged-ribose (within one hour of the completion
of the slow cool). Unless otherwise noted, reactions were conducted at 25C
(incubated on a hot block set to 25C to prevent the temperature from dropping below
25C). Reactions were terminated at various time points by diluting an aliquot of the
reaction in excess formamide stop dye, and then frozen (-20C). Time points were run on
an 8% PAGE/urea sequencing gels and the percent cleaved quantified using a Bio-Rad
phosphorimager.
Cloning and Sequencing DNAzymes
Prior to cloning, single-stranded DNAzymes were first converted to double
stranded DNA by PCR amplification, usually with only the complementary strand primer
(complementary no 5-phosphate). PCR conditions were: 96C for 3 minutes;
followed by three cycles of 45 seconds at 96C, four minute ramp cool to 50C plus 45
seconds at 50C, and 1 minute at 72C; followed by 7 minutes at 72C. When cloning
cleaved and uncleaved 614, the desired single-stranded DNA was (usually) first isolated
via gel purification. PCR amplifying the unpurified 614 reaction mixture containing
both cleaved and uncleaved 614 with only the complementary no 5-phosphate primer
yielded only uncleaved molecule clones; cleaved 614 molecules were therefore cloned by
29


69
by cis-cleavage, but our studies above indicated a significant amount of trans-cleavage at
concentrations higher than 3.5 19 nM. Whereas cis-cleavage entails only intra
molecular folding, /rafls-cleavage entails two separate folding interactions, intra
molecular folding (whether it is enzyme or substrate) and /ter-molecular folding
(whether it is between enzyme and substrate, or potentially between enzyme and
inhibitor). Although phenomenalogically different, both types of folding occur
simultaneously.
We tested the impact of our slow cool folding protocol on intra-molecular
folding by comparing the cleavage of ribose-614 with and without the slow cool
folding at a concentration low enough as to favor c/s-cleavage (2 nM) (Figure 3-36). The
overall cleavage of r-614 was higher when ribose-614 was completely denatured and
slowly cooled, as opposed to simply re-suspending the pellet in HEPES buffer and
initiating the reaction by adding NaCl and MgCh, suggesting that the slow cooling
procedure enhances intra-molecular folding. It is possible, however, that this rate
enhancement is attributable to the denaturation induced by the high temperatures in the
slow cool protocol, rather than enhancement of the intra-molecular folding, suggesting
that a fraction of the ribose-614 is locked into an inactive conformation as a result of the
ethanol precipitation. If this is the case, instead of improving intra-molecular folding, we
must say that the slow cool procedure improves intra-molecular re-folding.
When the same experiment was done at higher, traws-favoring concentrations, the
initial rate of ribose-614 cleavage is unchanged, but the plateau is slightly higher with the
slow cooling treatment (Figure 3-37). Similarly, cleavage of 1.2 nM r-614 by 150 nM
d-614 was identical, regardless of whether the enzyme and substrate were mixed
69


104
c
[E = d614] nM
Figure 3-29. Single-turnover kinetics for d-614 cleaving r-primer. (A) Full time
course. (B) Initial (linear) phase of cleavage. (C) Substrate saturation curve.
Substrate saturation experiments were conducted by incubating r-primer (4 nM) with
increasing amounts of d-614 enzyme.


132
of the Family gamma and delta sequences begin to show up following the fourth round of
selection. All representatives of the Family alpha are eliminated from our sampling in
rounds five and subsequent rounds, and all of the representatives of the delta Family are
eliminated by the eighth round of selection. The extinction of Family alpha and Family
delta may suggest that these sequence Families have low catalytic activity compared to
other sequences with which they are in competition (or perhaps Families went extinct
because they were poor substrates for PCR).
Although sequences clearly cluster into Families, few of the sequences are exactly
identical to any other sequence. This similarity of sequences within a Family can either
be the result of convergence onto similar sequences, assuming each sequence was
originally present in the initial random library, or through divergence from a single
sequence via mutations accrued during PCR amplification.
The sequences of Family beta, gamma, and delta are also very similar to
sequences 614, 615, and 616 (respectively) isolated from a previous selection experiment.
The infrequency of these large motifs (1 in 1021) makes it extremely unlikely that such
motifs would be sampled from a random library (containing 1024 potential variants)
multiple times (either within a single experiment or in two separate experiments) when
each sample contains only 1012 molecules. Several possibilities could explain this result.
(a) The library is not truly random: If, rather than containing the expected 1024
unique molecules, a bias in the synthesis or purification of the library greatly
reduces the actual library size to (for example) only 1013 different molecules, than
the odds of recovering the same sequence in each of two independent samplings
of 1012 molecules is relatively high.
(b) PCR generated bias. Creating the full-length product requires that only 40
nucleotides be synthesized between the two primers. Given that one primer is 49-
nucleotides long, and the other 18-nucleotides long, it is conceivable (although
not likely) that the primers may mis-prime off of either the randomized template


54
removing any potential inhibitors. We did this by again allowing 614 to reach cleavage
plateau, and then denaturing an aliquot of the sample by heating it to 97C for 3 minutes
and slowly cooling to 23C over 10 minutes. Reheating and slowly cooling allows any
molecule stuck in an inactive conformation a second chance to adopt an active
conformation. As in the gel-purification experiments, denaturing and refolding the
sample via the slow cool procedure increases the amount of material cleaved (Figure
3-16). The increase in cleavage is significant (usually about 10%), but does not bring
cleavage up to 94% (the amount cleaved by strong base and therefore considered
completion) and therefore cannot account entirely for the uncleaved 614 at plateau.
Cloning and Sequencing Cleaved and Uncleaved 614 Near Cleavage Plateau
The ability of strong base to cleave the ribose-primer, ang + ribose, to only 90-
97% suggested that about 6% of the primer, and therefore any full length DNAzyme
made from that primer, may be missing the ribose-adenosine. For the caged-primer, ang
+ cage, base hydrolysis following laser deprotection resulted in only 80% cleavage,
suggesting that the problem of missing ribose-adenosine may be even more severe for the
caged-primer (this could also be due to incomplete deprotection, despite excess laser).
Mutations may also occur throughout the DNAzyme, either as a result of mutations
introduced during the synthesis of the primers, or mutations between the primers as a
result of polymerase error. Together, these mutations may reduce or eliminate the
catalytic power of a fraction of the DNAzyme pool. This possibility was examined by
cloning and sequencing both cleaved and uncleaved 614 near the cleavage plateau
(Figure 3-17). Of the 28 sequenced clones of 614 that were cleaved at the time of gel-
54


19
protein catalysts, suggestive of hypothesis (c). Proteins do seem to exploit multiple types
of functionality, and exploit hydrophobicity, something lacking in aliphatic form in DNA,
suggestive of hypothesis (e). Last, recent experiments by Keefe and Szostak selecting for
proteins that do things suggests that they do not do things orders of magnitude better than
DNA (2001), suggestive of hypothesis (a) or (b).
Experimental Design
This thesis presents the first step towards addressing the hypotheses above. We
utilized the experimental procedure developed by Breaker and Joyce (1994, 1995). This
IVS procedure seeks DNA molecules that are catalytically active as ribonucleases; this
protocol is the most widely utilized procedure for estimating catalytic performance under
various conditions, allowing general comparisons such as the ones mentioned before that
suggest functionality does not significantly improve catalytic potential. As mentioned
above, one explanation for the apparent failure of functional endowment to generate very
active deoxyribozymes may result from the loss of highly reactive catalysts during some
step in the experiment before they had the opportunity to survive the selection step.
It is easy to find places in the set-up of the Breaker-Joyce selection procedure
where highly active catalysts might be lost. Standard IVS experimental systems assume
that catalysts remain inactive during the work up procedure, which includes: PCR (with
MgCL), ethanol precipitation (with high salts), binding to a column (again in high salts,
but without MgCL) and removing the complementary strand with high pH sodium
hydroxide. After work up is completed, the reaction is started with the addition of
MgCL and the eluate from a column is collected and presumed to contain the active
19


33
The samples were then run on an 8% PAGE-urea gel, and the full-length products
excised using a sterile razor (rinsed thoroughly with ethanol between samples). Each gel
slice was crushed and eluted in 350 pL elution buffer (500 mM NH4OAC, 0.1 mM
EDTA, 0.1% SDS) overnight. Samples were extracted in phenol/chloroform/isoamyl
alcohol (25:24:1), then extracted in chloroforrmisoamyl alcohol (24:1), and precipitated
in ethanol with ammonium acetate. Pellets were resuspended in 50 mM HEPES (pH 7)
and transferred to a new tube, at which point the final concentration was estimated using
specific radioactivity. All samples were brought up to 327 nM, except library #1 (BJ +
cage primer extended using standard dNTPS was brought up to 3270 nM). An aliquot
(30 pL) of libary #1 was set aside for the IVS condition E (containing ten times the
starting amount of library), and the remaining library #1 was diluted to 327 nM. An
aliquot (30 pL) of each sample was then mixed with equal volume of the appropriate 2X
reaction buffer as summarized in Table 2-1.
The samples were quickly heated to 96C for 3 minutes, and cooled to 23C over
10 minutes. For samples without a caged-ribose (G, H, N and O), the completion of the
slow cool to 23 C was considered the initiation of reaction. All samples were then
exposed to 150 pulses of laser (90 mJ/pulse). The time of laser exposure was considered
the time of reaction initiation for all samples containing a caged-ribose. Samples were
allowed to react at 25C for various times (20 to 720 minutes, see Table 2-1), and then
stopped by adding 1.25 volume of formamide stop dye. The samples were then run on an
8% PAGE/urea gel next to a cleaved library marker. The area corresponding to the
size of the cleaved library for each sample was excised, crushed, and eluted overnight
(referred to as survivors of round 1).
33


Double mutantions
that did not
recoverfunction:
C47T + G55A
C40T + G65A
C72G + G81C
C72T + G81A
C43G + G81C
Truncations of
614 that did not
cleave in trans:
28-106
41-91
56-94
56-106
Figure 3-43. Structural predictions for 614 folding in cis. Common folds shared by several structures are boxed in color. The red arrow
indicates the ribose-A site of cleavage. The red "X" marks double mutants that did not recover function.


51
approximately the highest concentration they reach under normal 614 cleavage conditions
in which approximately half of 614 converts into cleavage products).
The caged-library alone did not show detectable catalysis under standard reaction
conditions; this follows our expectation that an unselected, random library will have very
few active catalysts. Addition of the 27-nucleotide product alone did not increase caged-
library cleavage, while addition of the 79-nucleotide product alone did slightly increase
cleavage of the library. Cleavage levels were extremely low, however: while 614 had
reached a cleavage plateau of 65% in about 50 hours, the library cleaved by the 79-
nucleotide fragment had reached only 15% cleavage in 300 hours. The addition of the
27-nucleotide fragment together with the 79-nucleotide fragment reduced cleavage of the
library.
This shows that the 79 nucleotide fragment can act as a DNAzyme to cleave a
random library substrate in trans, but the rate of cleavage is orders of magnitude slower
that the full-length 614. The 27-nucleotide fragment reduces this already low activity of
the 79-nucleotide fragment, suggesting it acts to inhibit cleavage by competing with the
substrate (ribose-library) for binding to the 79-nucleotide fragment. These results are
considered again in a later section when we explore potential structures of 614.
Improperly Folded DNAzyme 614
Having shown that the cleavage products do not significantly alter the cleavage
plateau, we then set out to determine whether the plateau was due to the mis-folding of a
fraction our DNAzyme into an inactive conformation. While most IVS protocols largely
ignore the folding issue altogether (assuming folding happens quickly and efficiently, in
51


I certify that I have read this study and that in my opinion it conforms to the
acceptable standard of scholarly presentation and is fully adequate, in scope and quality,
as a dissertation for the degree of Doctor of Philosophy.
Steven A. Benner, Chair
Professor of Chemistry
I certify that I have read this study and that in my opinion it conforms to the
acceptable standard of scholarly presentation and is fully adequate, in scope and quality,
as a dissertation for the degree of Doctor of Philosophy.
Arthur S. Edison, Co-chair
Associate Professor of Biochemistry
and Molecular Biology
I certify that I have read this study and that in my opinion it conforms to the
acceptable standard of scholarly presentation and is fully adequate, in scope and quality,
as a dissertation for the degree of Doctor of Philosophy.
_A
Gerry Shaw
Professor of Neuroscience
I certify that I have read this study and that in my opinion it conforms to the
acceptable standard of scholarly presentation and is fully adequate, in scope ai^quality,
as a dissertation for the degree of Doctor of Philosophy.
Alfred S.
Professor pfMolecular Genetics
Microbiology


49
to see if this was the case in our cleavage reaction by several methods. First, cleavage a
200 nM solution of caged-614 was initiated by 300 pulses of laser and allowed to react
for 144 hours. At this time, near cleavage plateau, the reaction was split into three
aliquots. One aliquot was unaltered. The second aliquot was diluted twenty-five fold in
reaction buffer. Unlabeled d-614 was added to the third aliquot (to a final concentration
of 200 nM). D-614 is the same sequence as ribose-614, but with a deoxy-adenosine used
in the place of the ribose adenosine. In a later section we show that d-614 acts as an
enzyme to cleave r-614 in trans. If the cleavage plateau is a result of equilibrium
between the forward reaction and reverse reaction, the addition of additional enzyme (d-
614) should shift the equilibrium in favor the formation of more product over time.
Similarly, diluting the reaction reduces the likelihood that an enzyme can find two
substrates to ligate together, thus shifting the equilibrium in favor of cleavage (the
cleavage reaction can occur in cis, and thus is not as strongly affected by dilution as
ligation). Figure 3-13 shows that neither treatment significantly alters the cleavage
plateau, arguing against equilibrium between the forward and reverse reaction.
In a second experiment, the 79-nucleotide cleavage product was isolated via gel
purification. This cleavage product was radio-labeled, and following purification, was
incubated (in excess) with full-length ribose-614 (not radio-labeled). Although
unlabelled, previous studies have shown that the full-length 614 DNAzyme begins to
cleave itself, generating both the 79 and 27-nucleotide products. If the reverse reaction
(ligation) occurs, the excess of exogenous, radiolabeled 79-nucleotide fragment ensures
that there is a greater chance for the radio-labeled 79-nucleotide fragment to become
ligated to the 27-nucleotide fragment, thus causing an increase in size of the radio-labeled
49


clone name:
614wt 25 copies -- AA A CAT
jan#P 9 AA ACAT
jan#R AA ACAT
sept41o2fic41o3 r---*- AA ACAT
C ACTCTCACATCAT X A CACAC -CAATA CCT ATAA TT TA
C ACTCTCACATCATAC A CACAC -CAATA CCT ATAA TT TA
C ACTCTCACATCAT X A CACAC -CAATA ACT ATAA TT TA
C ACTCTCACATCATX A XACAC -CAATA CCT ATAA T- TA
AC
AC
AC
AC
jan#13
jan#L'
jan#3
jan# 7
jan#l*
jan#2 *
jan#6'
janffll
sept4up7
lb3
mcl
lb4
lb7&8
76
me 5
june#10
june#15
june#9
jAn#A
june#6
june#12
june#14 G41A
me 3
jan#0
64
65
june#2 G38A
71
81
61
MC74
93
94
janG'
63
june#16
79
78
74
4up6
lb2
septlbl
69
65
614wt 21
CT-
CT-
CT-
CT-
CT-
CTf
CT
CT-
CT- .
CT*
CT*
CT-
CT-
CT-
CTA
CT-
CT-'
CT* i
CT*'
CT- ,
CT-'
CT-.
CT-
CT-
CT-;
CTT
CT-
CT*
CT-:
CT-
CA.AATTCTAATAC
CA AATTCTAATAC
;CA AATTCTAATAC
:CA 'AATTCTAATAC
CA AATTCTAATAC
CA AATTCTAATAC
CA AATTCTAATAC
CA AATTCTAATAC
CA AATTCTAATAC
CA AATTCTAATAC
CA AATTCTAATAC
CA AATTCTAATAC
CA AATTCTAATAC
CA AATTNTAATAC
CA AATTCTAATAC
CA AATTCTAATAC
CA AATTCTAATAC
CA AATTCTAATAC
CA AATTCTAATAC
CA.AATTCTAATAC
CA AATTCTAATAC
CA AATTCTAATAC
CA AATTCTA-TAC
CA AATT CTACTA C
A.AATTCTAATAC
CA AATTCTAATACC
CA AATTCTAATAC
CA AATTCTAATAC
CA AATTCTAATAC
CA AATTCTAATAC
CT- CA
CT- CA
CT- CA
CT- CA
CT- CA
CT- CA
CT- CA
CT--CA
CT-7CA
CT- CA
CIV'CA
CT-.CA
CT- CA
CTT CA
copies CT- CA
AATTCTAATAC
AATTCTAATAC
AATTCTAATAC
AATTCTAATAC
AATTCTAATAC
AATTCTAATAC
AATTCTAATAC
AATTCTAATAC
AATTCTAATAC
AATTCTAATAC
AATTCTAATAC
AATTCTAATAC
AATTCTAATAC
AATTCTAATACC
AATTCTAATAC
1
-ACTCA-
- ACTCA-
-ACTCA-
-ACTCA-
-ACTCA-
-ACTCA-
-ACTCA-
-ACTC--
-ACTCA-
-ACTCA-
-ACTCA-
-ACTCA-
-ACTCA-
-ACTCA-
-ACTCA-
-ACTCA-
-ACTCA-
-ACTCA-
-ACTCA-
-ACTCA-
-ACTCA-
- ACTCA-
-ACTCA-
-ACTCA-
-ACTCA-
-ACNCA
-ACTCA-
-ACTCA-
-ACTCA-
- ACTCA-
-ACTCA-
-ACTCA-
-ACTCA-
-ACTCA -
-ACTCA-
-ACTCA-
-ACTCA-
.ACTCA-
-ACTCA-
- ACTCA-
-ACTCA-
-ACTCA-
-ACTCA-
-ACNCA
ACTCA-
2
CTAATA
CT A
CTA-TA
CTA-T-
CTA-T--
CTA-T-
CTA-T4-,
CTA-T-
-TA-T* ,
CTA-T-
CTA-T*.'-
CTA-T-
CTA-
CTA-T-
CTA-T-
CT--TA.
CT--TA
CT--TA
CTAATA
CTA--A
CTA-TA
CTA-TA
CTA-TA
CTA-TA
CTA-TA7
CTA-TA
CTA-TA
CTA-TA
CTA-TA
CTA-TA
AA ACAT
AA ACAT
AA ACAT
AA'ACAT
AA ACAT
AA ACAT
AA-ACAT
AA ACAT
A- ACAT
AA ACAT
AA ACAT
AA ACAT
AA ACAT
AA ACAT
AA ACAT
AA
AA
AA
AA
AA
AA
AA
AA
AA
AA
NAA
AA
AA
AA
AA
CTA-
CTAJ
CTA-
CTA-
CTA-
CTA-
CTA-
CTA-
CTA-
CTA-
CTA-
CTA-
CTA-
CTA-
CTA-
TA AA
TA AA
TA AA
TA AA
TA AA
TA AA
TAS AA
TA., AA
TA', AA
TA AA
TA AA
TA AA
TA AA
TA NAA
TA.% AA
3
ACAT
ACAT
ACAT
ACAT
ACAT
ACAT
ACAT
ACAT
ACAT
ACAT
ACAT
ACATA
ACAT
ACAT
TCAT
ACAT
ACAT
ACAT
ACAT
ACAT
ACAT
ACAT
ACAT
ACAT
ACAT
ACAT
ACAT
ACAT
ACAT
ACAT
C ACTCTCACATCAT iC A CACAC -
C ACTCTCACATCATGCG A CACAC
C ACTCTCACATCATGCf A CACACA
C ACTCTCACATCAT 3CGA< -CACAC
C ACTCTCACATCAT : A CACAC -
C ACTCTCACATCATGC -A1 -CACAC
C -ACTCTCACATCAT X A CACACA-
C ACTCTCACATCAT X A XACAC
C ACTCTCACATCAT ,C A ¡CACAC -
C ACTCTCACATCAT X ;A XACAC --
C ACTCTCACATCAT C A CACAC -
C ACTCTCACATCAT X^A CACAC -
C ACTCTCACATCAT1 X ',A< CACAC -
C ACTCTCACATCAT C A CACAC -
C ACTCTCACATCAT GAC CACAC -
C ACTCTCACATCAT X A CACAC -
C ACTCTCACATCATXGA CACAC -
C ACTCTCACATCAT X AGCACAC -
C ACTCTCACATCAT X A XACAC -
C ACTCTCACATCAT C A CACAC -
C ACTCTCACATCAC X A CACAC -
CAACTCTCACATCATGCGAGCACAC -
C ACTCTCACATCAT XGA XACAC -
C:ACTCTCACATCAT X A- CACAC- -
C ACTCTCACATCATGCGAGCACAC -
C ACTCTCACATCATX -A XACAC -
C ACTCTCACATCAT X GA CACAC -
C ACTCTCACATCAT XGACCACAC -
N ANTNT CA CATCAT X GA XACAC G -
C ACTCTCACATCAT X A XACAC -
CAATA CCT ATAA TT :TA
CAATA CCT ATAA v -TT ;TA
CAATA TCTATAAGGTTGGTA
CAATA CCT ATAA TT ; TA
CAATA CCT ATAA TT TA
CAATA CCT ATAA TT TA
CAATA CCT ATAA TTGGTA
CAATA XCT ATAA TT TA
CAATA CCT ATAA TT TA
CAATA CCT ATAA TTGGTA
CAATA CCT ATAA TTTA
CAATA XCT ATAAGGTT -GTA
CAATA CCT ATAA- TTGGTA
CAATA CCT ATAATT JTA
CAATA CCT ATAA TT ; iTA
CAATA CCT ATAA i TT ,TA
CAATA CCT ATAA TT -TA
CAATA CCT ATAA TT TA
CAATA -CCT ATAA TTGGTA
CAATA CCT ATAA TT r TA
CAATA CCT ATAA TT ; -TA
CAATA CCT ATAAGGTT TA
CAATA CCT ATAA TT -TA
CAATA CCT ATAA TTGGTA
CAATA CCT ATAA:-TTGGTA
CAATA XCT ATAA i 3TTGGTA
CAATA cct ataa : ITT : 7TA
CAATATCCT ATAA X -TTGGTA
CAATA CCT ATAA TT OTA
CAATA CCT ¡ATAAGGTTGGTA
T AC
T AC
T AC
T AC
T AC
T AC
T AC
T AC
T AC
T AC
T AC
T AC
T AC
T -C
T AC
AC
AC
AC
AC
AC
AC
AC
AC
AC
AC
AC
AC
AC
AC
AC
ACTNTCACATCAT X A CACAC
ACTCTCACATCATX ATCACAC
ACTCTCACATCAT X ATCACAC
ACTCTCACATCATGC A CACAC
ACTCTCACATCAT XGA CACAC
ACTCTCACATCATX GA CACAC
ACTCTCACATCAT X -A CACAC
ACTCTCACATCATGC- A CTCAC
ACTNTCACATCAT XGA CACAC
C ACTCTCACATCAT CA CACAC
C ACTCTCACATCAT X A CACAT
C ACTCTCACATCAT X A CACAC
C ACTCTCACATCATX-A-CACAC
C ACTCTCACATCATX AGCACACC
C ACTCTCACATCAT X A XACAC
4 5 6
-CAATA CCT
TCAATA CCT
-CAATA CCT
-CAATA CCT
-CATTA CCT
-CAATA CCT
-CAATA CCT
-CAATA CCT
-CAATA CCT
-CAATA TCT
-CAATA CCT
-CAATT CCT
-CAATA CCT
-CAATA CCT
-CAATA CCT
ATAAXTTGXA
ATAA CTT TA
ATAA XTTG TA
ATAAGCTTG-TA
;ataaggttggta
ATAAGGTTGGTA
AATAAGGTTGGTA
^ATAAGGTTGGTA
AATAAGGTTGGTA
ATAACACiTTGGTA
ATAA : -TT -TA
AATAAGGTTGGTA
AATAAGGTTGGTA
AATAAGGTTGGTA
AATAAGGTTGGTA
8
T AC
T AC
T AC
T AC
T AC
T AC
T AC
T AC
T AC
T AC
T AC
T AC-
T -C
T AC
T AC
9
to 614wt
in:
cii
TAA
CTT
CAC
+
TAA
CTT
CAC

TAA
CTT
CAC
+ -
TAA
CTT
CAC
-
TAA
CTT
CAC
TAA
CTT
CAC
TAA
CTT
CAC

TAA
CTT
CAC
TAA
CTT
CAC
TAA
CTT
CAC
TAA CTT
CAC
-
TAA
CTT
CAC
TAA
CTT
CAC
TAA
CTT
CAC
TAA
CTT
CAC
TAA
CTT
CAC
TAA
CTT
CAC
TAA
CTT
C-C
NAA
CTT
CAC
TAA
CTT
CAC
TAA
CTT
CAC
TAA
CTT
CAC
TAA
CTT
CAC
TAA
CTT
CAC
TAA
CTT
CAC
+
TAA
CTT
CAC

TAA
CTT
CAC
TAA
CTT
CAC
TAA
CTT
CAC
TAA
CTT
CNC
TAA
CTT
CAC
-
TAA
CTT
CAC
TAA
CTT
CAC
TAA
CTT
CAC
TAA
CTT
CAC
TAA
CTT
CAC
TAA
CTT
CAC
TAA
CTT
CAC

TAA
CTT
CAC
-TAA
CTT
CAC

AAA
CTT
CAC
TAA
CTT
CAC
TAA
CTT
CAC
TAA
CTT
CAC

TAA
CTT
CAC
TAA
CTT
CAC
-
TAA
CTT
CAC
TAA
CTT
CNC
TAA
CTT
CAC
Figure 3-17. Sequence alignment of cleaved and un-cleaved cloned 614 sequence variants. Mutations of 614 are highlighted in
yellow; N40 region between primers is underlined in the first and last sequence. Catalytic rate (relative to 614wt) shown to right.
VO
-P-


58
cleaving activity of the random region, most research apparently assume the impact of
these engineered sights is negligible.
However, some published accounts have reported that the rate of some
DNAzymes appears to decrease when they are generated with additional cycles of PCR
(5 vs. 20 cycles too few cycles to expect a significant change in the amount of mutation
for a 40-nucleotide product) (Geyer, 1997: Breaker, 1995). No explanation is offered for
this result. Neither papers cited do not report the concentration of DNAzyme used in
their kinetic analysis, presumably because they never considered the possibility of trans
cleavage (and it is therefore likely that the DNAzyme concentration was not controlled in
their analysis). If the concentration is not directly controlled, additional cycles of PCR
will obviously result in additional DNAzyme; it is plausible that the cleavage rates
changed with additional cycles of PCR because the DNAzyme can act in trans and
therefore the rate is concentration dependent. This next section explicitly explores this
possibility.
Cleavage Rate of Ribose-614 Varies with 614 Concentration
If a DNAzyme exclusively self-cleaves, it generates product exclusively by a first
order process, and the apparent first order rate constant should be independent of
[DNAzyme]. A plot of apparent kobs versus [DNAzyme] should be a horizontal line. If
product were generated exclusively by a second order process, then the apparent first
order rate constant would be dependent upon [DNAzyme], A plot of apparent kobs versus
[DNAzyme] should be a line that includes the (0,0) origin, with a slope equal to the
second order rate constant. If product were generated exclusively by a second order
58


45
The laser can potentially cause DNA damage which would inactive a DNAzyme
without breaking the phosphodiester backbone (such as forming thymine-thymine
dimers). We were unable to find evidence for thymine-thymine dimer formation when
ribose-614 was exposed to laser (even though ribose-614 does contain two successive
thymines at positions 82 and 83, and 100 and 101). If the laser is causing non-specific
DNA damage resulting in a reduced cleavage plateau, this should be evident even without
the caged-ribose. To test this, ribose-614 was exposed to increasing amounts of laser,
from 0 to 300 pulses (100 mJ/pulse). No change was noticed in cleavage plateau or
initial rates, suggesting laser does not cause damage to the DNA (Figure 3-9).
DNAzyme 614 is NaCl-dependent, and MgCl?-independent
DNAzyme 614 was selected using the standard Breaker-Joyce I VS protocol and
using a reaction buffer of 1 M NaCl, 1 mM MgCh, and 50 mM HEPES pH 7. This
protocol is expected to create cw-cleaving enzymes that utilize the magnesium as a
cofactor, while the sodium neutralizes the negative charge of the phosphodiester
backbone and thus stabilizes folding. The Breaker-Joyce I VS protocol assumes that the
DNAzymes generated may fold, but will not cleave without the added magnesium
cofactor, and further, that the rate of cleavage will increase with increasing amounts of
magnesium cofactor. Figure 3-10 shows that there is little difference in the kinetics of
ribose-614 if is incubated with either 1 mM or 3 mM MgC^. Experiments in which the
MgCh was completely removed and 3 mM EDTA was added also showed no change in
cleavage rates. However, removing the NaCl from the reaction buffer completely
prevents catalysis (Figure 3-11). DNAzyme 614 clearly is not magnesium-dependent, as
expected under the Breaker-Joyce experimental set-up.
45


ROUND 3
ROUND 4
ROUND 5
Progress Curve for New Population After:
ROUND 6 ROUND 7
Figure 4-14. Various initial distributions yield different stabalized populations following selections.
The peak for the stabilized population distribution is shifted closer to the St when the proportional abundance of slower catalysts is greaer in comparison to faster catalysts.


133
or off of one another to generate products of approximately 100 nucleotides.
Successive rounds of PCR and selection based on size and catalytic ability may
isolate molecules that were never present in the initial sample.
(c) Contamination from an external source. DNA from environmental sources (such
as pollen, microbes, or the humans conducting the experiments) may have
contaminated both I VS experiments.
(d) Contamination from an internal source. Catalytic sequences isolated from the
previous IVS may have contaminated equipment or buffers, therefore biasing the
library of the second IVS.
Because it is difficult to imagine a mechanism that would result in either scenario (a) or
(b), contamination seems to be the most likely explanation for finding extremely rare
catalytic sequences. A few factors, however, argue against contamination. First, the
second IVS was performed by different researchers, at different times, in different
buildings, with different equipment. Second, our sequencing studies of 614 revealed that
mutations occurred very infrequently (less than 1 out of 8 sequences contained a
mutation) after 20 cycles of PCR. If the Family beta sequences were derived from
contamination by 614, one would expect to find many exact copies of 614, particularly in
early rounds of the selection. Yet of the 12 Family beta clones, eleven differed from 614
by at least one nucleotide. Finally, the event that led to contamination is itself rare; the
chance of it occurring three separate times (one for each sequence) is considerably more
unlikely.
Kinetic Analysis of Round 8 Clones
Over half the clones isolated following round 8 were very similar to 614 (Family
beta and gamma), yet the rate constant for cleavage of the round 8 pool was significantly
faster than 614. The difference could be because the average rate of all sequence variants


153
Table 4-1. Summary of Results from Chapter 4
I VS experiment
Cleaved products were less than 1% of the total after the first round of selection.
The amount of cleaved products in selections with caged-ribose increased
steadily with successive rounds of selection to -10% by round 7.
The amount of Selection H increased dramatically after the fourth round of
selection.
Eight rounds of selections for samples containing the caged-ribose did not
significantly enrich for catalysts. Selections with caged-ribose did not show
time-dependent cleavage, and when pools from Round 1,4, and 8 were re
synthesized with caged-ribose, the only cleavage occurred immediately after
laser photolysis.Pools from selections with caged-ribose following Round 8
were re-synthesized with unprotected ribose, but did not show any time-
dependent cleavage.
Selection H resulted in significant enrichment for catalysts.
Several of the catalysts isolated from Selection H were nearly identical to
catalysts isolated in previous selection experiments, but single-nucleotide
variants possessed rates that different by an order of magnitude.
The bulk pool of catalysts from Selection H is Mg++ independent.
IVS Simulations
The leakage parameter dramatically affects the number of rounds or selection
necessary to achieve enrichment for active catalysts.
The relative abundance of catalysts compared to non-reactive molecules (the
Reactive Fraction) influences the number of rounds necessary to detect
catalysts.
The selection time parameter influences the number of rounds of selection
required to first detect active catalysts.
When the amount of cleavage of a pool at the end of each round of selection
reaches a plateau, the population has reached a stable point where the
distribution of catalytic rates remains largely unchanged with more rounds of
selection
If the selection time is short relative to the half-life of the fastest catalysts in a
library, the stabilized population will reach a cleavage plateau below -95%
cleaved at the end of successive rounds of selection, and the pool will be
enriched for only the fastest catalysts of the initial library.
If the selection time is large relative to the half-life of the fastest catalysts in a
library, the stabilized population will contain a distribution of catalytic rates
that is proportional to the distribution of the initial library.


65
saturation may not be completely achieved (sufficient enzyme was not available to test
kinetics at concentrations higher than 6000 nM). The slow approach to plateau is most
notable for the r-lib61o2 substrate, and to a lesser extent with r-614, suggesting that
substrate or enzyme may exist in inactive, alternative conformations (an explaining the
low R value for the association curve with r-lib61o2).
A second experiment conducted with ang + ribose primer as the substrate and
excess d-614 enzyme concentration increasing from 11 nM to 300 nM also demonstrated
progressively increasing rates up to 0.048 hr'1 at 300 nM (Figure 3-32). Similarly,
r-lib61o2 cleavage increased progressively when the enzyme (d-614) was in excess with
concentrations ranging from 13 nM to 450 nM. Cleavage rates were always higher for
ang + ribose primer substrate than for r-lib61o2 substrate at comparable d-614
concentrations, again suggesting that the shorter substrate may have less self-binding
opportunities which compete with and therefore reduce binding and catalysis by d-614.
These enzyme excess saturation experiments reveal that the rate for trans
cleavage (kcat(bi) = 0.034 hr'1, for r-614) is distinctly higher than the rate for c/s-cleavage
(kcat(uni) = 0.006 hr'1). This 6-fold increase in rate is surprising given the expectation that
tethering the substrate region to the enzyme region, as in the c/s-cleaving 614, would
significantly increase the effective concentration of the substrate relative to the enzyme,
and thus dramatically increase the rate. When trans-acting enzyme reaches
concentrations of 3.5 to 19 nM, the observed rate is twice the rate of c/s-cleavage, and
thus represents the concentration at which the c/s-cleavage rate and rras-cleavage rate
are equal (Figure 3-21). The effective molarity of linking the substrate to the enzyme, in
65


43
approximately one hour. Cleavage was then initiated by exposing the sample to varying
pulses of laser (0-1000 pulses). Cleavage of ribose-614 was initiated by resuspending
in reaction buffer and slow cooled. Figure 3-5 shows similar progress curves for both
ribose-614 and caged-614 cleavage. Both samples show identical initial cleavage
kinetics, with cleavage plateaus near 50%. The caged-614 plateau is regularly about 10-
15% lower than ribose-614. Several scenarios can explain the lower cleavage plateau
seen with caged-614: (a) insufficient deprotection of caged-ribose, (b) the 614 sequence
is missing the ribose-adenosine at position 27 (discussed above, and below Cloning and
Sequencing 614), or (c) laser induced damage to the caged-614 DNA sequence. The
reasons for both caged-614 and ribose-614 reaching cleavage plateau far below
completion is explored below (Understanding Reasons for Incomplete Cleavage of
DNAzyme 614).
To confirm that the lower cleavage plateau of caged-614 was not due to
insufficient laser, kinetic experiments were conducted with additional pulses of laser. In
one experiment, caged-614 cleavage was initiated with 150 pulses of laser and allowed to
react for 73 hours, approaching cleavage plateau. The sample was divided into two
aliquots, and one aliquot of the sample was exposed to an additional 150 pulses of laser.
The added laser caused a slight increase in the amount cleaved, but this increase was
immediate, and not increasing with time, suggesting against the possibility of additional
de-protection of caged-ribose.
Another experiment was also performed in which caged-614 cleavage was
initiated with varying amounts of laser, from 0 to 1000 pulses. The cleavage plateau
increased with increasing amounts of laser until 150 pulses of laser, after which
43


26
amplification consisted of 3 minutes at 96C; followed by 20 cycles of 45 seconds at
96C, 45 seconds at 50C, and 2 minutes at 72C.
PCR for in vitro selections (IVS) varied slightly: 400 nM of the catalytic strand
primer was used (to compensate for hairpins formed by the BJ +cage primer), up to 40
cycles of PCR were used in the early rounds of IVS, Vent exo- was used (rather than Taq
because it incorporates 5-functionalized nucleotides such as E-base with greater
efficiency), and primer binding was done at 57C rather than 50C (again, to compensate
for hairpins formed by BJ primers).
Point mutations introduced within in the catalytic strand primer region but 3 of
the ribose-adenosine, were incorporated using a two step PCR. In the first step, 10 cycles
of PCR were conducted in which the catalytic strand primer was replaced with one
containing the desired point mutation and a 5 nucleotide truncation from the 5end. The
product of this PCR was used as the template (diluted 1:10) for a second PCR containing
a catalytic strand primer truncated two nucleotides 3 of the ribose-adenosine. This two-
step PCR method allowed for the introduction of several point mutations into the
catalytic primer region without necessitating unique DNA-RNA chimeric primers for
each mutant.
Preparation of Single-Stranded DNAzvmes
Double stranded DNAzymes generated via PCR using the 5-phosphorylated form
of the complementary strand primer (complementary + 5-phosphate) were converted to
single-stranded DNA by digestion of the complementary strand using lambda
exonuclease, an exonuclease enzyme specific for 5-phosphorylated double stranded
26


This dissertation was submitted to the Graduate Faculty of the College of
Medicine and to the Graduate School and was accepted as partial fulfillment of
the degree of Doctor of Philosophy.
December, 2002
Dean, college of Medicine
Dean, Graduate School


percent cleaved
97
K exp fit
0061
0031
0.023
0.020
0.011
0.011
430.0 nM r-614
108 0 nM r-614
19.4 nM r-614
3.5 nM r-614
1.1 nM r-614
0.3 nM r-614
B
m fraction
cleavedAtour
0 03233
430.0 nM r-614
0 01627 *.
108.0 nM r-614
0 00959
19.4 nM r-614
0 00899
3.5 nM r-614
0.00695
1.1 nM r-614
0 00610 *
0.3 nM r-614
Figure 3-20. 614 cleavage rate is concentration dependent. (A) Complete time
course for r-614 at various concentrations. Rates are estimated based on an
exponential fit to a single first-order exponential plateau below 100%. (B) Linear
initial phase of time course. Rates are estimated based on slope of best fit line.