Citation
Novel analytical approaches for the determination of leucine enkephalin as a model for opioid peptides

Material Information

Title:
Novel analytical approaches for the determination of leucine enkephalin as a model for opioid peptides
Creator:
Larsimont, Veronique, 1967-
Publication Date:
Language:
English
Physical Description:
vii, 145 leaves : ill. ; 29 cm.

Subjects

Subjects / Keywords:
Antibodies ( jstor )
Calibration ( jstor )
Fluorescence ( jstor )
Immunoassay ( jstor )
Incubation ( jstor )
Opioid peptides ( jstor )
Phosphates ( jstor )
Reagents ( jstor )
Receptors ( jstor )
Tracer bullets ( jstor )
Chromatography, High Pressure Liquid ( mesh )
Department of Pharmaceutics thesis Ph.D ( mesh )
Dissertations, Academic -- College of Pharmacy -- Department of Pharmaceutics -- UF ( mesh )
Enkephalin, Leucine ( mesh )
Enzyme-Linked Immunosorbent Assay ( mesh )
Immunoassay ( mesh )
Models, Molecular ( mesh )
Monophenol Monooxygenase ( mesh )
Opioid Peptides ( mesh )
City of Gainesville ( local )
Genre:
bibliography ( marcgt )
theses ( marcgt )
non-fiction ( marcgt )

Notes

Thesis:
Thesis (Ph. D.)--University of Florida, 1994.
Bibliography:
Includes bibliographical references (leaves 138-144).
Additional Physical Form:
Also available online.
General Note:
Typescript.
General Note:
Vita.
Statement of Responsibility:
by Veronique Larsimont.

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:
028818701 ( ALEPH )
52005731 ( OCLC )

Downloads

This item has the following downloads:


Full Text










NOVEL ANALYTICAL APPROACHES FOR THE DETERMINATION OF LEUCINE
ENKEPHALIN AS A MODEL FOR OPIOID PEPTIDES













By
VERONIQUE LARSIMONT














A 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

1994


























To my parents, Charles Larsimont and Kruawan Kanjanasuwan-Larsimont for their unfailing support and encouragement of all my endeavors.













ACKNOWLEDGMENTS


My thanks go to my advisor, Dr. Gunther Hochhaus and the members of my supervisory committee, Dr. Hartmut Derendorf, Dr. Paul Klein, Dr. Laszlo Prokai and Dr. Ian Tebbett for their guidance and support during the course of my doctoral research. Special thanks go to Dr. Prokai for the mass spectrometry analysis of hydroxylated leucine enkephalin derivatives. I am also grateful to Dr. Richard Prankerd for his help in the early stages of this work.

I acknowledge the P.D.A. Foundation for Pharmaceutical Sciences, Inc. and Schering-Plough Corporation for partial funding of the research presented in this dissertation.

I would also like to recognize the services provided by the Hybridoma Core and the Protein Chemistry Core of the Interdisciplinary Center for Biotechnology Research at the University of Florida.

There are many others who are too numerous to mention, who have been instrumental in enabling me to complete this work. I hope to be able to thank each of them personally.











iiin













TABLE OF CONTENTS

pne

A C K N O W L E D G M [E N T S ........................... ... ....................... --- ... .... .................... ........... ............ ............................................... ............ iii

A B S T R A C T ............................... ................. ................................ ......................... ... .............. ... .... .................................................. ............... v i

CHAPTERS

I IN T R O D U C T IO N ........................................................................... -- .......................................... ..................................... I

E n d o g e n o u s O p io id P ep tid e s ..................................................................................................... .............. ........................................ I
O p io id R e c e p to rs ..................................................................... ...................................................................................... 3
P h y sio lo g y a n d P h a rm a co lo g y .................... ................................................................................................................ 5
R atio n a le .......... ...................................................... .......... --.- ........................................................................................................................... 12
O bje ctiv e s ............. ................. --- ......................................................................................... ........................................ .............................................. 19

2 APPROACHES TO THE DEVELOPMENT OF AN IMMUNOASSAY
F O R L E U C IN E E N K E P H A L IN .................................................... .................................................. -- ............................... 2 3

In tro d u ctio n ..................... --- .......................................................................................... -- ........... .................... ........... --- 2 3
M a te ria ls... ........................ -- ...................................................................................................................................................... ........... ............ 2 8
M e th o d s ............................................................................................ .......................... .................... ............................. ..................... 2 8
R e su lts a n d D iscu ssio n ................. ............................................................................................................................................................. 4 1

3 A HIGH PERFORMANCE LIQUID CHROMATOGRAPHY ASSAY FOR
OPIOID PEPTIDES USING ELECTROCHEMICAL DETECTION. --- 66

In tro d u ctio n ........................ -- ................................... ............................. ..................... --- .......................... --- ......................................... 6 6
M ate ria ls ............................................................................................................................................................................................................................. 6 9
M eth o d s .......................... ......................................................................................................................................................................... 7 0
R e su lts ............ ................................................ ................. ............. --- ............................................................................................. 7 5
D isc u ssio n ...................................................................................................................... -- ............... ---- ................................................. 8 2

4 A HIGH PERFORMANCE LIQUID CHROMATOGRAPHY ASSAY
FOR OPIOID PEPTIDES USING FLUORESCENCE DETECTION ...................... 87

In tro d u c tio n ........................................................................................ ........................ .................................................... 8 7
M ate ria ls --- ............................................................... ........... ........................................................................................................... 8 9
M e th o d s ........................................................................ ............................. ....................................................................................... 9 0


iv








R e su lts ..................... ............. ... ........................ -- ...................... .. ... ....................... ..................... -- ............................................... ................ 9 3
D isc u ssio n ................... .. .... ............. .......... ..... ...... ....... ... ... .. ... ........................ ................................ --- .............. ........................ 10 0

5 LEUCINE ENKEPHALIN-TYROSINASE REACTION PRODUCTS
IDENTIFICATION AND BIOLOGICAL ACTIVITY .................................... .................................. 104

In tro d u c tio n ......................................................................... -- ............................ ........................ -.- .............................. -- ......................... 10 4
M ate ria ls ................... ......................................................................................................................................................................................................... 10 5
M e th o d s ........................... ........................................................................................................... ...................... ............................................ 10 5
R e su lts ................................................................................................. ........................... ............................ .......................................... ............... 1 10
D isc u ssio n ......................... .................................. .................................... -.- ............................................................................... ................................. 1 1 8

6 C O N C L U S IO N S ............ ............................................................................................................... .............................................................. 12 1

APPENDICES

A D A T A FO R H PL C -E D A PPR O A C H ...................................................................... .......................................... 126

B D A TA FO R H PLC -FL A PPR O A C H ................................................. ....................................... 131

R E F E R E N C E S .......................... ..................................... -- ............................................ ........................................................ ........................... 1 3 8

B IO G R A P 14 1C A L S K E T C H ............................................................................. .................................................................... 14 5


























v













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

NOVEL APPROACHES FOR THE DETERMINATION OF LEUCINE ENKEPHALIN AS A MODEL FOR OPIOID PEPTIDES
By
Vronique Larsimont

August 1994

Chairman: Ginther Hochhaus
Major Department: Pharmaceutics

The focus of this dissertation was the evaluation of novel analytical approaches for opioid peptides by immunoassay or high performance liquid chromatography (HPLC) using leucine enkephalin (LE) as a model peptide.

The proposed immunoassays are based on the high affinity exhibited by avidin for biotin (Kd=10'5 mol/l). The successful development of the enzyme-linked immunosorbent assay relied on the formation of a sandwich between anti-LE antibody, a biotinylated LE derivative and avidin, whereas the successful development of the homogeneous fluorescence immunoassay depended on a lack of sandwich formation. The formation of a sandwich was not achieved using any combination of the two anti-LE antibody preparations and several biotinylated LE derivatives tested, and therefore efforts in this direction were abandoned. However, using a polyclonal antibody produced in this laboratory, an N-terminal biotinylated LE derivative without a spacer arm and fluorescein





vi








isothiocyanate avidin, an homogeneous fluorescence immunoassay for LE was developed which was operational in a narrow concentration range (1 10-9 to 1*10"8 moles LE/mi).

Two HPLC assays for opioid peptides were evaluated. One was based on tyrosinespecific pre-column hydroxylation using tyrosinase, specific sample clean-up using a boronate gel and HPLC with electrochemical detection. The other involved tyrosinespecific pre-column hydroxylation using tyrosinase followed by fluorogenic derivatization using 1,2-diamino-1,2-diphenylethane and HPLC with fluorescence detection. These assays yielded limits of detection for LE of 170 fmol/inj and 500 fmnollinj respectively in buffer samples and 360 fmol/inj and 500 fmollinj respectively in spiked cerebrospinal fluid samples.

Using electrospray ionization mass spectrometry, the structure of the products of the reaction between LE and tyrosinase were found to be monohydroxylated LE ([HOTyr']-LE) and dihydroxylated LE ([(HO)2-Tyr']-LE). Compared to LE, the affinity of [HO-Tyrl]-LE to both p and 8 opioid receptor sites in rat brain homogenate was found to be lower by a factor of about 20. Since enkephalins and tyrosinase have been found to coexist in vivo, we speculate that tyrosinase may play role in the metabolic pathway of these compounds.














vii















CHAPTER 1
INTRODUCTION


Endogenous Opioid Peptides


There exists three separate families of natural or endogenous opioid peptides:

enkephalins, endorphins and dynorphins. These three families are derived from three

different prohormones, proenkephalin, pro-opiomelanocortin and prodynorphin,

respectively and are coded by messenger RNAs from three separate genes (Figure 1. 1)

[Pleuvry 1991].

Aipha-melanocyte stimulating hormone*
Corticotrophin-like intermediat Beta-melanocyte stimulating
lobe peptide* Beaedrhnhormone*

Adrenocorticotrophin* I:

eta-lipotrophin* B 4
PRO 1-01PlO LAN TIN Leumorphin
4 Beta-neo-endorphin
m-RNA
GENES Dynorphin A (1-17)
P *-mRAa. OPI
m-RNA
=-ENKEPHALINDynorphln B (1-13) (Melekepah Pepid E Dynorphin (1-8)
[Metlek finPeptide F tLeulenkphan (Metionkephalln-Arg-Gly-Leu (Met jenkephalin-Arg-Phe

Figure 1.1. Opioid peptides, and their precursors. (* Denotes no opioid receptor activity)
[Pleuvry 1991]




1





2


In the proenkephalin family, proenkephalin A is the precursor of the pentapeptides methionine enkephalin (ME, Tyr-Gly-Gly-Phe-Met) and leucine enkephalin (LE, Tyr-GlyGly-Phe-Leu) which were the first opioid peptides to be characterized by Hughes and coworkers in 1975 (Figure 1.2) [Hughes et al. 1975]. Proenkephalin A has been shown to contain ME and LE in a fixed ratio of six ME sequences to one LE sequence. Proenkephalin-expressing cells are widespread throughout the brain and spinal cord as well as in more peripheral sites such as the adrenal medulla and the gastrointestinal tract.


PRO-OPOMELANOCORTIN
-LPH -END
I I I Ie--I IIIh


a-MSH #-MSH MET-ENK
PROENKEPHAUN
MET-ENK LEU-ENK
wIN4 I I N I



MET-ENK MET-ENK
ARGO-GLY'-LEU* ARG-PHE'
PRODYNORPHIN
LEU-ENK




e-NEOENDORPHIN- DYN DYN
DYN DYN
P-NEOENDORPHIN

Figure 1.2. Schematic representation of the structures of opioid peptide precursors [Jaffe
and Martin 1990].


Pro-opiomelanocortin (POMC) is the precursor of the opioid peptide 1-endorphin as well as the non-opioid hormones adrenocorticotrophic hormone (ACTH) and ct- and 3-





3


melanocyte simulating hormone (MSH, Figure 1.2). The term endorphin applies to opioid peptides derived from POMC. POMC is synthesized in the pituitary but is also present in the hypothalamus and in the periphery. The ME amino acid sequence is present at the Nterminal end of O-endorphin.

The dynorphins as well as 03-neoendorphin and leumorphin are derived from prodynorphin which is also known as proenkephalin B. Prodynorphin contains LE sequences but no ME sequences and is synthesized throughout the central nervous system (Figure 1.2).

When the opioid peptide precursors are processed to give the various different opioid peptides, the N-terminal end of the molecules is highly conserved with tyrosine in the I position. However, tyrosine is often absent in this position in non-related peptides. This phenomenon is exploited in two of the analytical approaches described in this dissertation.


Opioid Receptors


Opioid receptors are widely distributed throughout the central nervous system of all vertebrates and have also been found in a number of peripheral tissues including the intestinal tract, the adrenal and pituitary glands and the vasa diferentia of several species [Simon 1984].

At present, three classes of opioid receptors are firmly recognized, predominantly on the basis of in vivo studies of opioid action (agonists and antagonists), in vitro bioassays and binding experiments with selective ligands [Simonds 1988]. These three






4


classes of receptors are defined as mu, kappa and delta. The existence of sigma and epsilon receptors has also been postulated; however, there is now some doubt as to whether the sigma receptor is truly "opioid" in nature and epsilon receptors have not been detected with any degree of certainty in any tissue except rat vas deferens. The cloning of opioid receptors using recombinant DNA techniques can be expected to yield information about the differences in the primary amino acid sequence of mu, delta and kappa receptors, and may also uncover new opioid receptor subtypes thus elucidating the structural features responsible for receptor specificity.




.cingulate gyrus
subcollosal striatur
,.......- striatum
medial thalamic nuclei
occipital lobe hebenutola
perltu ventral antertior nucleus
substantla ora septet re ion
interpeduncu a ft globus pattidus
.ld'f -- ref-c Form hypothalamus
reticular formation inferior frontal
re**a postrma tion olfactory tri ne
cerebellum l*a*myg M CI
substantial tpoltobe
gelatinos high hippocampus
medium
I ow


Figure 1.3. Distribution of opioid receptors in the human brain [Simon 1984].


No selective endogenous ligand has yet been isolated for the mu receptor as opioid peptides from all three families bind to this receptor. It has therefore been suggested that the mu receptor may be a "universal" opioid receptor [Akil et al. 1988]. Mu opioid receptor sites are widespread but are found in particular in the regions of the brain associated with pain regulation and sensorimotor integration [Mansour et al. 1988]. It has been proposed that mu receptors may be subdivided into high affinity sites (mu) which





5


are thought to mediate supraspinal analgesia and a low affinity sites (mu2) which are thought to be responsible for respiratory depression and gastrointestinal effects [Pasternak 1982].

Leucine enkephalin and other derivatives of proenkephalin A interact with the delta receptor, although not selectively. Opiate alkaloids, on the other hand, have low affinity for this receptor. Delta receptors are less widespread than mu receptors but are concentrated in neural areas involved with olfaction and motor integration and have been implicated in pain pathways [Mansour et al. 1988].

The derivatives of prodynorphin have selectivity for kappa receptors. Although LE is selective for the delta receptor, as the molecule is lengthened, its preference for the delta receptor is reduced and its affinity for the kappa receptor increases. Kappa receptors are found predominantly in brain areas associated with pain perception and the regulation of water balance and food intake [Mansour et al. 1988].


Physiology and Pharmacology


The binding of an opioid to its receptor triggers a number of complex processes which occur before leading to the ultimate opioid effect. Opioid effects are believed to mediated through guanine nucleotide regulatory proteins (G proteins) which are involved in signal transduction to a variety of effector systems including adenylate cyclase, phospholipidase C and ion channels. Presently, adenylate cyclase inhibition is the best characterized opioid effect mediated by G proteins [Simon 1984].





6

Adenylate cyclase is an enzyme that synthesizes cyclic AMP from ATP so that it can then go on to act as a "second messenger" in a number of biochemical systems. Gproteins cause adenylate cyclase to convert from the active form of the enzyme which is coupled to GTP to the inactive form which is coupled to GDP and vice versa. Therefore, a compound that stimulates adenylate cyclase production would do so through the interaction of its receptor with a G-protein that converts adenylate cyclase to the active GTP-coupled form, whereas inhibition of adenylate cyclase, as by opioids for example, would be mediated through a G-protein that favors the formation of inactive GDP-coupled adenylate cyclase (Figure 1.4).


GDP GTP

exchangeg
R s

AC-GTP AC-GDP


Ri
hydrolysis

Figure 1.4. Scheme for receptor mediated stimulation and inhibition of adenylate cyclase.
Rs receptors which produce stimulation of adenylate cyclase on binding, Ri receptors which produce inhibition of adenylate cyclase on binding, AC
adenylate cyclase [Simon 1984].


Opioid peptides are distributed widely throughout the central and peripheral nervous system, suggesting that these compounds play a part in a variety of physiological functions [Olson et al. 1991]. The physiological roles of endogenous opioid peptides have been attributed largely on the basis of effects seen on administration of the opioid





7


antagonist naloxone. These effects are assumed to be the result of opioid receptor blockade and are widely accepted as indirect evidence for endogenous opioid involvement in the physiological function under observation.

One of the most important functions of opioid peptides is in pain modulation. The mu receptor is the opiate receptor most associated with pain relief, although delta and kappa receptor agonists also have analgesic properties. It is thought that under some conditions, mu and delta receptors are functionally coupled as delta agonists given in subanalgesic doses have been found to either potentiate or inhibit the analgesic effects of morphine (a mu agonist) at different sites. It has therefore been postulated that mu and delta receptors may exist either separately or in a complexed form [Rapaka and Porreca 1991]. Acupuncture analgesia is also thought to be mediated by endogenous opioid peptides [Clement-Jones and Rees 1982].

Animal studies have indicated that endogenous opioid peptides play a role in the development of opiate dependence, a side effect common to opiate analgesics. This is a syndrome whereby distress is caused upon withdrawal of an opiate following chronic administration. It has been hypothesized that chronic narcotic abuse leads to the suppression of endogenous opioid production through a negative feedback mechanism so that sudden withdrawal of the narcotic leads to a deficiency in endogenous opioids which causes the classical physical withdrawal symptoms [Clement-Jones and Besser 1983]. Recently, Wang et al. [Wang et al. 1994] have proposed that stimulation by an opiate agonist causes gradual constitutive mu receptor activation so that an agonist is no longer required for signal transduction, and a dependent state is established, consisting of an





8

upregulated cAMP system, counterbalanced by constitutively active mu receptors. In this model, opiate tolerance results from fewer mu receptors remaining activatable by agonists and the enhanced activity of the cAMP system. In other words, in this scenario dependence occurs because an upregulated cAMP system is established which needs to be counterbalanced by opiate agonist activity and tolerance occurs because fewer mu receptors can now be activated by opiate agonists.

A lower degree of dependence is seen with agonists at delta receptors than with agonists at mu receptors. As mentioned above, the action of mu receptor agonists can be modulated by the co-administration of delta receptor agonists so that the potency and efficacy of analgesia is increased without a corresponding increase in side effects such as physical dependency, respiratory depression and gastrointestinal effects. This effect could be exploited to allow for the use of mu agonists of lower efficacy and increased safety while still providing adequate pain relief without the risk of side effects. Eventually, a delta agonist may be developed which is able to provide effective pain relief without side effects [Rapaka and Porreca 1991 ].

Opioid peptides play a part in the regulation of the immune system, particularly during periods of stress, as they modulate the functions of a number of cell types involved in the immune response [Murgo et al. 1986]. Generally, endogenous opioid peptides are immunostimulant as they enhance T-cell function and stimulate phagocyte function, thus increasing resistance to infection. There is also evidence that suggests a role for endogenous opioids in the growth and development of lymphoid tissue [Plotnikoff et al. 1985].





9

Opioid peptides are thought to depress the responsiveness of the chemosensors to carbon dioxide and may therefore play a physiological role in the control of respiration [McQueen 1983]. This applies in particular to neonates and to adults in stressful situations [Pleuvry 1991].

Opioid peptides may also be involved in blood pressure regulation as they are present in nerve fibers in areas of the brain stem responsible for the regulation of blood pressure and the secretion of vasopressin. Biochemical evidence suggests that opioid peptides interact with neurohormones to regulate blood pressure. Opioid peptides have been implicated in the dramatic changes in blood pressure which occur during sleep and in hypotension due to various states of shock [Rubin 1984]. It has also been suggested that endogenous opioid peptides may be involved in the pathogenesis of hypertension [Szilagyi 1989].

The presence of opioid peptides within limbic structures suggests their involvement in the regulation of mood and behavior. Endogenous opioid peptides are also known to interact with the central catecholamines implicated in psychiatric disease. However, the results of studies carried out to determine the role of opioid peptides in psychiatric disease have been contradictory [Clement-Jones and Besser 1983, Koob and Bloom 1983], and therefore no conclusions can be drawn at present.

The very high concentrations of opioid peptides present in the hypothalamus suggests a role for these substances in neuroendocrine regulation. Opioid peptides may control the secretion of anterior pituitary hormones by modifying the release of hypothalamic anterior pituitary regulating substances. This may be the mechanism by





10

which opioid peptides cause an increase in the secretion of prolactin, growth hormone and thyrotrophin and inhibit the release of luteinizing hormone, follicle stimulating hormone, adrenocorticotrophic hormone and beta- and gamma-lipotropin [Clement-Jones and Besser 1983, Grossman and Rees 1983]. Further possible functions of opioid peptides can be found in Table 1.1.


Table 1.1. Possible physiological functions of endogenous opioid peptides [Imura et al.
1985].

1. Defense against noxious stimuli
Activation of the pituitary-adrenocortical axis Regulation of the sympatho-adrenal system Inhibition of pain perception

2. Modulation of vegetative nervous system Cardiovascular and respiratory system Gastrointestinal tract and pancreas Genito-urinary tract

3. Modulation of neuroendocrine function
Anterior and posterior pituitary hormones Gastrointestinal and pancreatic hormones Catecholamines

4. Behavioral action
Mood and locomotor activity Food and water intake Sexual behavior


The design of opioid peptides as therapeutic agents has several advantages. Firstly, these substances are endogenous so that their metabolites are likely to be non-toxic and not to cause renal or hepatic damage, depending on the doses administered. Secondly, a large number of analogs can be synthesized from a few basic amino acid building blocks as synthesis has been simplified and automated and simple modifications can be used to





11

develop different analogs with desirable biological activities. Opioid peptides are unable to cross the placental barrier as they are subject to placental enzymatic deactivation and would therefore be ideal for obstetric use [Rapaka 1986, Rapaka and Porreca 1991 ].

At present, the therapeutic development of opioid peptides is focused on their potential as analgesic agents and in the treatment of opiate addiction. Research efforts are directed towards the design of analgesic peptides which can be administered orally, have a long duration of action and reduced potential for dependence and abuse. Peptides of interest include enkephalins, endorphins and related opioid peptides.

Opioid peptides are easily degraded by aminopeptidases which hydrolyze the Tyr'Gly2 bond, carboxypeptidases which cause cleavage at the C-terminal end of the molecule, relatively non-specific enzymes such as trypsin and angiotensin converting enzyme (ACE) and more specific enzymes such as enkephalinases which hydrolyze the Gly3-Phe4 in enkephalins. This has led to the suggestion that enkephalin degrading enzymes be used as an alternative therapeutic approach [Rapaka 1986, Rapaka and Porreca 1991]. Examples of these enzyme inhibitors include bestatin, thiorphan and captopril which inhibit aminopeptidase, enkephalinase and ACE, respectively. These inhibitors prolong the duration of action of endogenously released enkephalins and it is therefore hoped that they are free of the side effects produced by narcotic analgesics. The clinical use of these substances may however be limited due to their limited bioavailability.

Another approach which has been used to increase both the stability and the selectivity of opioid peptides is the introduction of synthetic modifications to the molecule [Shimohigashi 1986]. For example, stability can be increased by substituting the





12

corresponding D-amino acid for the naturally occurring L-amino acid in the peptide molecule. As peptides are conformationally labile, the selectivity of opioid peptides has been increased by stabilizing the peptide molecule in a conformation which prefers the desired receptor. This can be achieved through the incorporation of conformational restrictions such as the introduction of unnatural bulky synthetic amino acids (e.g./ penicillamine residues) or the cyclization of the peptide chain. Highly selective opioid peptide analogs such as [D-Pen2, D-Pen5]-enkephalin which is selective for the delta receptor [Mosberg et al. 1983] and [Tyr-D-Ala-Gly-MePhe-NH(CH2)20H] which is selective for the mu receptor [Handa et al. 1981] have been developed using these techniques.

Further research involving opioid peptides and their receptors is of importance in elucidating the precise physiological roles of these entities, particularly in the areas of pain and immunomodulatory pathways.


Rationale


As discussed above, a considerable body research has been focused on the development of opioid peptides as therapeutic agents. The low physiological concentrations of endogenous opioid peptides and those to be expected for therapeutically administered derivatives necessitate the development of specific and ultra-sensitive analytical methods, in the fmol per ml range, for these entities in biological fluids to support clinical studies. The need for new analytical approaches for the measurement of opioid peptides has been stressed by the National Institute on Drug Abuse [Rapaka 1986].





13

Current analytical methods for opioid peptides are summarized in Table 1.2 and are reviewed in the following pages.


Immunological Methods


At present, for the most part, opioid peptides are analyzed by radioimmunoassay (RIA) [Sato 1984, Venn 1987]. Although immunoassays are highly sensitive and reproducible, they are hampered by several disadvantages. Firstly, an antiserum to the analyte of interest has to be raised, which involves lengthy incubation times. Secondly, immunoassays suffer from low selectivity due to the cross-reactivity of the antiserum to structurally similar compounds. To overcome the problem of cross-reactivity, high performance liquid chromatography (HPLC)-immunoassay methods have been developed whereby collected eluted fractions from an HPLC system are analyzed by RIA [Defrutos and Regnier 1993, McDermott et al. 1981]. However, HPLC-immunoassay methods are time consuming, labor intensive and chromatographic solutions must be volatile or compatible with the immunoassay. Radioimmunoassays also have the added disadvantage that the use of radioactive isotopes as tracers calls for special considerations in the handling of the assays and the disposal of the radioactive waste produced.

Non-radiation immunological techniques have not been exploited to a great extent for opioid peptides. An enzyme-linked immunosorbent assay (ELISA) has been described for LE and ME [Zamboni et al. 1983] but its limits of detection are only in the 1 pmol per assay range. Only four enzyme immunoassays with high sensitivity have been developed for P3-endorphin and dynorphin [Hochhaus and Hu 1990, Hochhaus and Sadee 1988,





14

Kuhling et al. 1989, Sarma et al. 1986]. These assays have sensitivities ranging from 0.3 to

3.2 femtomole per assay.

The immunoassays developed by Hochhaus and Sadee [Hochhaus and Sadee 1988] and Hochhaus and Hu [Hochhaus and Hu 1990] are ELISAs based on the avidinbiotin system whereby the peptide of interest in the sample or standard and its biotinylated derivative compete for antibody binding sites. The antibody-bound biotinylated species is subsequently detected by enzymatic detection through the use of an avidin-enzyme complex. For this dissertation, an attempt was made at the development of an extremely sensitive avidin-biotin based ELISA for enkephalins, using LE as a model peptide, to compliment the existing ELISA tests for P3-endorphin and dynorphin. If this type of assay can be developed for LE, it can be expected to be also applicable to ME.

In addition, for this dissertation, an attempt was also made at the development of an homogenous or non-separation immunoassay for opioid peptides. Homogeneous immunoassays differ from traditional immunoassays in that the labor intensive separation of the bound and free fraction of the analyte (e.g./ by washing, precipitation or adsorbance) is not necessary prior to quantitation as the property being measured is characteristic of either the bound or the free analyte or label.

The immunological techniques proposed in this dissertation have the disadvantage of low specificity, but they were intended for use as "immunological HPLC detectors" in the hope that they would provide fast, ultra-sensitive assays which were highly suitable for processing large numbers of samples such as HPLC fractions.





15

Table 1.2. Summary of analytical methods for opioid peptides.

Reference Method Sensitivity Analyte Comments

Fleming and HPLC-ED 1 pmol/inj enkephalin tedious sample clean up,
Reynolds high oxidation potential
1988
Kim et al. HPLC-ED 1 pmol/inj enkephalin tedious sample clean up,
1989 high oxidation potential
Shibanoki et HPLC-ED 550 fmol/inj enkephalin tedious sample clean up, al. 1990 high oxidation potential
Monger and HPLC-ED 75 fmol/ ml 3-endorphin tedious sample clean up,
Olliff 1992 plasma high oxidation potential

Muck and HPLC-MS 100 fmol/inj dynorphin microbore LC system
Henion 1989

Mifune et al. HPLC-FL 100 fmollinj enkephalin complicated column 1989 switching
Nakano et al. HPLC-FL 140 fmol/inj enkephalin harsh reaction conditions 1987
Kai et al. HPLC-FL 500 fmol/inj enkephalin harsh reaction conditions
1988
van den Beld HPLC-FL 50 fmollinj -endorphin laser induced
et al. 1990 fluorescence
Dave et al. HPLC-FL 36 fmol/inj LE microbore LC system
1992

Hochhaus ELISA <1 fmollassay -endorphin
and Sadee
1988
Hochhaus ELISA 1 fmollassay dynorphin
and Hu 1990
Kuhling et al. ELISA 3 fmol/assay 3-endorphin
1989

de Ceballos HPLC-RIA 1.5 fmol/assay LE, ME et al. 1991
Maidment et RIA <1 fmollassay ME
al. 1989
HPLC-ED = high performance liquid chromatography with electrochemical detection, HPLC-FL = high performance liquid chromatography with fluorescence detection, HPLC-MS = high performance liquid chromatography with mass spectrometry, ELISA = enzyme-linked immunosorbent assay.





16


Instrumental Approaches


Other methods which have been used in the analysis of opioid peptides include IIPLC combined with electrochemical detection (I-PLC-ED), fluorescence detection (HIPLC-FL) or mass spectrometry (HPLC-MS). Electrochemical detection

HPLC in conjunction with electrochemical detection is an analytical method which offers several advantages. It provides selectivity, as only those compounds which are oxidizable or reducible at the applied potential will be detected. Multiple electrode detectors can be used to pre-oxidize contaminants in the mobile phase and the sample prior to detection of the analyte of interest, thus increasing sensitivity by improving signal to noise ratios. The method is versatile and the cost of instrumentation and reagents is relatively low. In electrochemical detection, peak current ratios are obtained from the ratio of the peak heights obtained (i.e. current generated) when two different voltages are applied to the same amount of sample. These peak current ratios are characteristic for each compound, much like absorbance ratios in ultraviolet detection, and therefore qualitative information about the analyte can be derived from them.

A review of the literature reveals that, to date, HPLC-ED methods for opioid peptides are less sensitive than other methods such as RIA [Fleming and Reynolds 1988, Kim et al. 1989, Mousa and Couri 1983]. This may be attributed to the fact that the high potentials necessary to detect opioid peptides using these methods (+0.9-1.25 V compared to +0.3-0.4 V used in HPLC-ED assays for catechols with sensitivities in the fmollinj range [Higa et al 1977, Koike et al. 1982]) compromise sensitivity as background current





17

and baseline noise are increased significantly. Selectivity is also compromised as more compounds are oxidized at these high potentials, thus necessitating extensive sample clean up.

In the HPLC-ED assay for LE described in this dissertation, the enzymatic derivatization used increased HPLC-ED selectivity and sensitivity by pre-column ohydroxylation of the highly conserved N-terminal tyrosine groups of the peptide resulting in easily oxidizable derivatives.

Fluorescence detection

Pre-column fluorescence derivatization of analytes can be achieved by fluorophoric labeling using a fluorescent precursor, or by fluorogenic derivatization using a nonfluorescent precursor. Fluorogenic derivatization is usually preferred as fluorophoric labeling often requires excess fluorescent reagent and subsequent extensive clean up to minimize background interference.

Fluorogenic derivatization reactions have been carried out by using fluorogenic reagents such as o-phthalaldehyde [Roth 1971], fluorescamine [Udenfriend et al. 1972] and naphthalene-2,3-dialdehyde in the presence of cyanide [Lunte and Wong 1989, Mifune et al. 1989]. However, these procedures derivatize N-terminal amino groups and consequently are not specific for any particular peptide so that subsequent chromatographic procedures are often extremely complex (e.g./ multidimensional HPLC systems using column switching) to allow for the selective determination of opioid peptides as the derivatives of other peptides may interfere with the signal.





18

A derivatization reaction which is specific for tyrosine groups such as the one described in this dissertation could be expected to increase selectivity in the determination of opioid peptides as tyrosine is highly conserved in the 1 position of these compounds but is often missing in unrelated peptides. Tyrosine specific HPLC methods with fluorescence detection such as those using 1,2-diamino-4,5-dimethoxybenzene [Ishida et al. 1986, Kai et al. 1988] and hydroxylamine, cobalt (II) ion and borate [Nakano et al. 1987, Zhang et al. 1991] do exist for peptide analysis. Although these methods have allowed the determination of opioid peptides with sensitivities of 100-500 femtomole per injection, the harsh reaction conditions of these derivatizations are expected to lead to diminished recoveries and reduced assay reproducibility of the fragile peptide analytes. Mass spectrometry

HPLC with mass spectrometry offers the highest level of molecular specificity compared to other analytical methods and is the only method with which unambiguous confirmation of the structure of the target peptide can be achieved. However, MS is costly and requires specialized instrumentation and therefore, it is an impractical method for the average analytical laboratory. The analysis of peptides by mass spectrometry has been facilitated in recent years by the advent of several "soft" ionization and sample introduction techniques such as fast atom bombardment, matrix-assisted laser desorption, electrospray and ionspray mass spectrometry (see Arnott et al. 1993, Biemann 1992 and Carr 1990 for reviews) which allow the production of intact molecular ions from these fragile species. Another advantage of these techniques is that they are often compatible with on-line microbore HPLC. A microbore HPLC-ionspray MS method has allowed the





19

determination of enkephalins in the 100 fmol/inj range [Muck and Henion 1989]. Levels as low as 5 fmol of 13-endorphin and 1 pmol of ME have been quantified by electrospray mass spectrometry with off-line HPLC [Dass and Kusmierz 1991] and pmol amounts of ME have been detected by fast atom bombardment mass spectrometry with off-line HPLC [Kusmierz and Sumrada 1990].


Objectives


Appropriate analytical methodology for opioid peptides is required for use in clinical, pharmacokinetic and formulation studies as well as in physiological studies to allow the successful development of these entities as therapeutic agents and the continuation of research to elucidate further the physiological role of these compounds. The need for accurate, specific, sensitive and reproducible analytical methods for opioid peptides has been stressed by representatives of the National Institute on Drug Abuse [Rapaka 1986]. However, to date no analytical procedure has emerged which adequately meets all of these criteria.

Therefore, the focus of the work carried out for this dissertation has been the evaluation of several novel analytical approaches for the determination of leucine enkephalin (LE, Figure 1.5) as a model for opioid peptides. Leucine enkephalin was chosen as a model peptide as it contains the same initial sequence common to all opioid peptides.

The first approach evaluated was a non-homogeneous enzyme-linked immunosorbent assay (ELISA) which employed the same strategy as ELISAs which have





20

been previously developed for dynorphin [Hochhaus and Hu 1990] and P-endorphin [Hochhaus and Sadee 1988]. This assay was based on the avidin-biotin system whereby avidin exhibits an extremely high affinity for biotin and biotinylated species. Here, LE and biotinylated LE derivative compete for antibody binding sites and subsequently, on separation of the antibody-bound and free fractions, the antibody-bound biotinylated species is detected through the use of an avidin-enzyme complex. The success of this type of assay depends on the formation of a "sandwich" between the antibody, the biotinylated LE derivative and enzyme-labeled avidin.

OHil


CH3 CH3
CH
I
CH2 0 0 CH
I II II 11
CH NH C CH H C CH / OH
NHCI\ /NH /C\ /CH2\ ,NH\ /C\ /CH\ ,O
NH2 C CH2 NI C CH NH C
II II I II
0 0 CH2 0





Tyr Gly Gly Phe Leu


Figure 1.5. Leucine enkephalin.


The second approach was an homogeneous or separation-free fluorescence immunoassay which also made use of the avidin-biotin system, and was based on an observation by AI-Hakiem and co-workers [Al Hakiem et al. 1981] that the binding of





21

biotin to fluorescence-labeled avidin produced an increase in fluorescence intensity. In contrast to the ELISA described above, this approach relied on a lack of sandwich formation between antibody, biotinylated LE derivative and fluorescence-labeled avidin, as the success of this approach depended on the inability of antibody-bound biotinylated LE to interact with fluorescence-labeled avidin and produce an increase in fluorescence intensity. Therefore, here LE and biotinylated LE compete for antibody binding sites and after equilibrium has been achieved, free biotinylated LE is detected through the increase in fluorescence intensity produced on addition of fluorescence-labeled avidin.

Two high-performance liquid chromatography (HPLC) assays were also evaluated, both making use of tyrosine-specific pre-column derivatization using the enzyme tyrosinase. In the high performance liquid chromatography assay with electrochemical detection (HPLC-ED), specific hydroxylation of the tyrosine group in the I position of LE, which is highly conserved in all opioid peptides, presents two analytical advantages. Firstly, the derivatization results in the formation of a catechol which is amenable to specific clean-up using boronate gels, and secondly, this catechol is more easily oxidizable than the parent peptide, thus facilitating electrochemical detection. In the high performance liquid chromatography assay with fluorescence detection (HPLC-FL), enzymatic derivatization by tyrosinase renders our peptide amenable to fluorogenic derivatization using 1,2-diamino- 1,2-diphenylethane.

In addition, having established in the enzymatic derivatization for the HPLC assays that products are formed from the reaction between LE and tyrosinase, the identity of these products was determined by mass spectrometry. Furthermore, since tyrosinase and





22

enkephalins have been found to co-exist in vivo [Merchenthaler 1993, Sesack and Pickel 1992, Zhuo et al. 1992], the biological activity in rat brain homogenate of the major product of the reaction between these two entities was investigated.














CHAPTER 2
APPROACHES TO THE DEVELOPMENT OF AN IMMUNOASSAY FOR LEUCINE ENKEPHALIN


Introduction


Immunoassay is an analytical method which exploits the binding of a ligand (antigen) to specific sites on an antibody. In most cases (e.g./ for radioimmunoassays), labeled ligand competes with unlabeled ligand (analyte) for antibody sites and the extent of binding of the labeled ligand is determined through the measurement of some physical or chemical property associated with the label. A standard calibration curve of the signal produced by the label with respect to the concentration of analyte present can then be constructed, thus allowing the estimation of unknown ligand concentrations from this curve.

Traditionally, immunoassays have involved the use of radioisotopes as labels. Although radioimmunoassays (RIA) have the advantages of being highly sensitive and invulnerable to environmental interferences (e.g.! by components of the assay), their use is accompanied by several disadvantages. These disadvantages include the emotive bias against the use of radioisotopes, the costly disposal of waste, special requirements for assay handling and training of staff, the lack of stability of radioisotopes and the consequent limited shelf-life of reagents [Gosling 1990]. Therefore, there has been considerable interest in developing non-isotopic labels, such as enzyme labels and



23





24


luminescent labels for use in inmmunoassay [Kricka 1993, Porstmann and Kiessig 1992, Schulman et al. 19901.

Immunoassays can be divided into those methods which require separation of the antibody-bound and free fractions prior to quantitation, and those in which the signal being measured is a function of antibody binding and therefore do not require separation of antibody-bound and free fractions prior to quantitation. These assays are referred to as heterogeneous and homogeneous immunoassays, respectively. The development of homogeneous immunoassays was prompted by the fact that the separation step in heterogeneous immunoassays is labor intensive, complicates automation and introduces inaccuracy and error into the assay as the equilibrium which exists between the bound and free fraction in the sample is disturbed. Enzyme- and luminescence- labeling techniques are used in the majority of the homogeneous immunoassays developed thus far [Coty et al. 1992, Garcia et al. 1993, Jenkins 1992]; however, other approaches involving the use of liposomes [Bowden et al. 1986, Ho and Huang 1985, Umeda et al. 1986], bilayer membranes [Ihara et al. 1988] and reversed micellar systems [Kabanov et al. 1989] have also been investigated. Although homogeneous immunoassays are convenient and relatively simple to perform, to date, they have suffered from lack of sensitivity compared to heterogeneous immunoassays, due in particular to interference with endpoint determination by components of biological samples [Jenkins 1992].

As mentioned in Chapter 1, one of the major disadvantages of immunoassays in general is their relative lack of specificity due to cross-reactivity of the antibody in use to structurally similar compounds. In order to circumvent this problem, analytical methods





25

have been developed whereby structurally similar analytes are first separated by highperformance liquid chromatography (HPLC) prior to quantitation by immunoassay. To this end, the goal in the development of the immunoassays proposed in this dissertation was the design of rapid and convenient assays for opioid peptides with "detector-like" properties, intended for the determination of opioid peptide concentrations in HPLC fractions. These assays should therefore allow the sensitive and straightforward analysis of opioid peptides, be practicable on a large scale and easily amenable to intensive automation.

Two different approaches to the development of an immunoassay for opioid peptides were conceived, using leucine enkephalin (LE) as a model peptide. Both of these approaches were based on the exploitation of the extremely high affinity exhibited by avidin for biotin (Ka= 10" l/mol). Each avidin molecule has four high affinity binding sites for biotin so that the central strategy of these approaches lies in the knowledge that when avidin is conjugated to an enzyme or fluorescent label, it will still bind to biotin or biotinylated species [Wilchek and Bayer 1988].

The first approach, an enzyme-linked immunosorbent assay (ELISA) for LE is similar to assays previously developed for dynorphin [Hochhaus and Hu 1990] and 3endorphin [Hochhaus and Sadee 1988] which have sensitivities in the lower fmol/assay range. The development of a similar assay for LE would have provided a battery of enzyme immunoassays for opioid peptides based on identical principles. This assay involves competition between LE in the sample or standard and biotinylated LE derivative for immobilized antibody binding sites. The antibody-bound biotinylated species is





26

subsequently detected through the use of an avidin-enzyme complex (Figure 2.1). The successful design of this type of assay for LE depends on the synthesis of a suitable biotinylated LE derivative that will allow the formation of a "sandwich" between the antibody, the biotinylated LE derivative and the enzyme-labeled avidin, thus enabling quantification upon addition of the enzyme substrate.



46 -- _] _7-- a" + EI






+E Substrate Detection
tE





Antibody Biotinylated analyte


Analyte Enzyme-labelled avidin


Figure 2.1. Avidin-biotin based ELISA


The second approach, which is an homogeneous or separation-free fluorescence immunoassay is based on an observation by AI-Hakiem et al. [Al-Hakiem et al. 1981] that the binding of biotin to fluorescein-labeled avidin leads to an increase in fluorescence intensity. This phenomenon is exploited in an attempt to develop a new class of homogeneous immunoassay which depends on the inability of antibody-bound biotinylated





27

LE to interact with fluorescein-labeled avidin. Therefore, in this assay, biotinylated LE and LE in the sample or standard compete for antibody binding sites and after equilibrium has been achieved, fluorescence-labeled avidin, which acts as a "detector" molecule, is added. Unbound biotinylated LE is determined by the increase of fluorescence intensity due to the interaction of the unbound biotinylated LE and the fluorescence-labeled avidin (Figure 2.2). In contrast to the ELISA described above, the success of this assay depends on the synthesis of a biotinylated LE derivative which will not allow fluorescence-labeled avidin to bind to antibody-bound biotinylated LE (i.e. lack of "sandwich" formation).











+ ,F _D _F


Increased fluorescence intensity



SAntibody Biotinylated analyte
_] --,, F -LF




F
Analyte Fluorescence-labelled avidin



Figure 2.2. Avidin-biotin-based homogeneous fluorescence immunoassay





28

Materials


Leucine enkephalin, leucine enkephalin-Lys6, Freund's complete adjuvant, Freund's incomplete adjuvant, bovine serum albumin, porcine thyroglobulin, glutaraldehyde, 1-ethyl-3-(3-dimethylaminopropyl)carbodiimide, sodium acetate, activated charcoal, dextran (average molecular weight 70,800), avidin, N-hydroxysuccinimidobiotin, biotinamidocaproate N-hydroxysuccinimide ester, triethanolamine, 2-4'-hydroxyazobenzene-benzoic acid, di-methylsulfoxide and fluorescein isothiocyanate avidin were obtained from Sigma Chemical Company, St. Louis, MO, USA. Methanol, acetonitrile, trifluoroacetic acid, sodium dihydrogen phosphate, disodium hydrogen phosphate, sodium chloride, potassium dihydrogen phosphate, dipotassium hydrogen phosphate and potassium chloride were purchased from Fisher Scientific, Pittsburgh, PA, USA. [tyrosyl3,5-3H(N)]-leucine enkephalin was procured from NEN Research Products, Dupont Company, Wilmington, DE, USA. Cytoscint scintillation cocktail was obtained from ICN Biomedicals Inc., Irvine, CA, USA and anti-LE antiserum was purchased from Peninsula Laboratories Inc., Belmont, CA, USA.


Methods


Antibody Production


Leucine enkephalin (LE) was conjugated to either porcine thyroglobulin or bovine serum albumin (BSA) by reaction with either glutaraldehyde or 1-ethyl-3-(3-dimethylaminopropyl)carbodiimide (EDC).





29

For the glutaraldehyde method (G. Adamus, personal communication), a 25 % v/v stock solution of glutaraldehyde was freshly prepared on ice and diluted 65-fold in 0.1 M sodium phosphate buffer pH 7. Equal amounts of LE and protein carrier (BSA or porcine thyroglobulin) were weighed out and dissolved in 0.1 M sodium phosphate buffer pH 7 to give a solution containing 1 mg/ml each of LE and protein carrier. One hundred and twenty four microliters of the final dilution of glutaraldehyde were added for each milliliter of protein-peptide solution and the reaction was allowed to proceed at room temperature overnight with constant stirring. The conjugate was then dialyzed against deionized water for 24 hours using pre-hydrated Spectrapor cellulose ester membranes with molecular weight cut off of 15,000 (Spectrum Medical Industries Inc., Los Angeles, CA, USA). After dialysis, aliquots of the conjugates were stored at -20'C prior to lyophilization.

For the EDC method [Harlow and Lane 1988], a 1 mg/ml solution of LE was prepared in water and EDC was weighed out and added to give a final concentration of 10 mg/ml. The pH of the reaction mixture was adjusted and maintained at pH 5 with 1 N NaOH throughout the 5 minute incubation time at room temperature. An equal volume of an 11 mg/ml solution of protein carrier (BSA or porcine thyroglobulin) was added and the reaction was allowed to proceed at room temperature for 4 hours. The reaction was then stopped by the addition of a sodium acetate solution (1 M, pH 4.2) to give a final concentration of 100 mM. After an additional incubation of 1 hour at room temperature, the conjugate was dialyzed against 0.1 M phosphate buffer pH 7 for 24 hours using prehydrated SpectraPor cellulose ester membranes with molecular weight cut off of 15,000 (Spectrum Medical Industries Inc., Los Angeles, CA, USA). After dialysis, aliquots of the





30

conjugates were stored at -20'C prior to lyophilization. Lyophilization of the conjugates was carried out at the Drug Delivery Laboratory, University of Florida, Progress Center, Alachua, FL, USA using a Model 12K Super Modulyo Lyophilizer (Edwards, West Sussex, England).

After lyophilization, samples of LE, BSA, porcine thyroglobulin and the conjugates were sent for amino acid analysis at the Peptide Core of the Interdisciplinary Center Biotechnology Research, University of Florida, Gainesville, FL, USA. The loading ratios of LE to carrier protein were calculated from the percentage composition of asparagine, threonine, serine and glutamine in the carriers and conjugates. The LEthyroglobulin conjugate obtained by the glutaraldehyde method showed the highest peptide loading ratio (see Results and Discussion) and therefore this conjugate was selected as the inoculum for the purposes of this work.

For the initial inoculation, complete Freund's adjuvant was placed in a test tube and an equal volume of 0.1 M sodium phosphate buffer pH 7.4 containing 2 mg/ml of the LE-thyroglobulin conjugate was added while vortexing, to give a thick emulsion containing 1 mg/ml of the conjugate. Three young adult male New Zealand White rabbits each received a total of 1 ml of this emulsion subcutaneously at 6-8 different sites in the back. Thereafter, booster injections with the same dose of conjugate were given at monthly intervals in the same manner, except that incomplete Freund's adjuvant was used instead of complete Freund's adjuvant. Test bleeds were taken 7-10 days after each boost and after serum separation, the antibody titer was tested by radioimmunoassay. Briefly, to test total binding, [tyrosyl-3,5-3H(N)]-leucine enkephalin (3H-LE, 1 nM) was incubated





31

overnight at 4'C with various dilutions of the antiserum (1/10, 1/100, 1/1,000, 1/10,000) in 0. 1 M sodium phosphate buffer pH 7. To test non-specific binding, 3 H-LE (1 nM) was incubated overnight at 4'C with LE (500 nM) and various dilutions of the antiserum (1/10, 1/100, 1/1,000, 1/10,000) in 0.1 M sodium phosphate buffer pH 7. An ice-cold suspension containing 1. 5 % w/v activated charcoal and 0. 15 % w/v dextran in water was then added to each sample and after a 5 minute incubation at 4 'C, the samples were centrifu~ged at 12,000 g for 3 minutes. An aliquot of the supernatant was then removed, Cytoscint scintillation cocktail (4 ml) was added and the radioactivity (CPM) representing the antibody-bound tracer (3H-LE) was determined using a Beckman LS 5,000 TD scintillation counter (Fullerton, CA, USA). After the third boost, the antibody titer of one of the rabbits was deemed sufficient and therefore, this rabbit was exsanguinated, the serum was separated from whole blood by centrifugation at 5,000g for 10 minutes (Dynac 11 Centrifuge, Clay Adams, Sparks, MID, USA) and stored in aliquots at -80'C.

A 10 ml aliquot of the anti-LE antiserumn was purified by the Hybridoma Core of the Interdisciplinary Center Biotechnology Research, University of Florida, Gainesville, FL. Briefly, a column was filled with 10 ml of Prot G-Gammabind Ultra matrix (Pharmacia LKB Biotechnology Inc., Piscataway, NJ, USA) and washed with 35 mld of elution buffer (0.1 M glycine pH 3.0). The column was then washed extensively with 200-300 ml of binding buffer (0.1 M phosphate buffered saline pH 7, 0.0 1% sodium azide). The antiserum was diluted two-fold with binding buffer, loaded onto the column and washed with binding buffer until the optical density of the eluent reached zero at X280. Elution buffer was then added to the column, the eluent was collected in 1 ml fractions and






32


neutralized immediately with 35 p.1 of 2 M Tris buffer pH 9. The fractions with high optical density at X280 were pooled and then desalted and concentrated by spinning down 5-6 times for 30 minutes in a C-PREP 10 centrifugal filter (Amicon, Danvers, NM, USA). This procedure yielded 8.5 ml of purified anti-LE antibody containing 6.1 mg IgG/ml, determined by measurement of optical density at X280.


Characterization of Antibody


The number of specific antibody sites in the purified antibody was determined by means of Scatchard analysis [Scatchard 1949]. Samples containing various concentrations of 3 H-LE as tracer (70-1000 pM) and a 1/1,000 dilution of the purified antibody were set up to determine total binding (Table 2.1). To determine non-specific binding, samples containing antibod y, 3 H-LE (70-1000 pM) and a high concentration of LE (1.2* 10 M) were also set up at the same time. After overnight incubation at 4'C, 200 PI1 of an ice cold suspension containing 1. 5% activated charcoal and 0. 15% dextran in water was added to each tube (except total counts) and after a further 5 minute incubation on ice, the samples were centrifuged at 12,000g for 3 minutes. Three hundred and fifty microliters of the resultant supernatant containing the antibody-bound fraction of the tracer were then removed and added to 4 ml of Cytoscint scintillation cocktail. The radioactivity (CPM) representing the antibody-bound tracer was determined using a Beckman LS 5,000 TD scintillation counter (Fullerton, CA, USA) and a 5 minute counting time with counting efficiency at about 50%.





33

Table 2.1. Samples set up for Scatchard analysis.

Total counts Total binding Non-specific binding Buffer 420 l 20 l ----3H-LE 20 pl 20 ll 20 pld
LE (1.4*10-5' M) ----- ----- 20 l
Antibody (1/1,000) ----- 200 pl 200 gl

Overnight at 4oC, then:
Charcoal (1.5%)/dextran (0.15%) 200 ptl 200 pl



A Scatchard plot was constructed by plotting the ratio of bound to free 3H-LE against bound 3H-LE and the Ka value was determined from the negative slope of the line drawn between the points. The number of specific antibody sites was determined from the intercept of this line with the x-axis.


Biotinylation of Leucine Enkephalin


N-terminal biotinylated LE (BLE) was synthesized by allowing 90 nmoles of LE, 180 nmoles of N-hydroxysuccinimidobiotin (BHS) and 120 nmoles of triethanolamine in 150 pl of dimethylsulfoxide (DMSO) to incubate overnight at room temperature. The product of the reaction was separated from the reagents by injecting the incubation mixture into a gradient HPLC system consisting of a Rainin Rabbit HP solvent delivery system (Rainin Instrument Company Inc., Woburn, MA) controlled by a Rainin Dynamax HPLC method manager (version 1.3, Rainin Instrument Company Inc., Woburn, MA), a Negretti & Zamba injector (Southampton, UK) fitted with a 100 pl loop, a tBondapak C18 column (10 um, 3.9 x 150 mm, Waters Associates, Milford, MA, USA) and a LDC





34

Milton Roy Spectromonitor 3100 variable wavelength detector (Riviera Beach, FL, USA). The detection wavelength was set at 210 nm. Mobile phase A consisted of 90 % v/v aqueous trifluoroacetic acid (0.02% v/v) and 10 % v/v acetonitrile and mobile phase B consisted of 10 % v/v aqueous trifluoroacetic acid (0.02% v/v) and 90 % v/v acetonitrile. A gradient of 90 % mobile phase A and 10 % mobile phase B to 50 % each of mobile phases A and B in 30 minutes was run at a flow rate of 1 ml/min. A control injection of the incubation mixture was also made into the HPLC system immediately after preparation to allow the calculation of the extent of conversion of LE to BLE. N-terminal biotinylated LE incorporating a spacer arm between the biotin group and the peptide (BXLE) was synthesized in the same manner except that biotinamidocaproate N-hydroxysuccinimide ester (BXHS) was used instead of BHS in the incubation mixture. The peaks corresponding to BLE and BXLE were collected from the HPLC eluent and the volatile components (acetonitrile and trifluoroacetic acid) were evaporated under a stream of nitrogen at 300C. The biotinylated LE derivatives were stored at 4oC and were used within two weeks of synthesis.

C-terminal biotinylated leucine enkephalin-Lys6 (LE-Lys6-B) was synthesized by allowing 73 nmoles of leucine enkephalin-Lys6 (LE-Lys6), 142 nmoles of BHS and 90 nmoles of triethanolamine in 150 pl of DMSO to incubate for two hours at room temperature. The product of the reaction was separated from the reagents by injecting this incubation mixture into the same gradient HPLC system described above except that a gradient of 90 % mobile phase A and 10 % mobile phase B to 70 % mobile phase A and 30% mobile phase B in 30 minutes was run at a flow rate of 1 ml/min. A control injection





35

of 73 nmoles of LE-Lys6 in 150 pl of DMSO was also made into the HPLC system to allow the calculation of the extent of conversion of LE-Lys6 to LE-Lys-B. C-terminal biotinylated LE incorporating a spacer arm between the biotin group and the peptide (LELys6-BX) was synthesized in the same manner except that BXHS was used instead of BHS in the incubation mixture. The peaks corresponding to LE-Lys6-B and LE-Lys6-BX were collected from the HPLC eluent and stored as described earlier for BLE and BXLE.

The successful biotinylation of LE by the methods described above was characterized by comparing the displacement of 2-(4'-hydroxyazobenzene)-benzoic acid (HABA) from avidin by the biotinylated LE derivatives and biotin itself according to the spectrophotometric method of Green [Green 1970]. A 0.25 mM solution of HABA was prepared in 0.1 M sodium phosphate buffer pH 7 and avidin was added to an aliquot of this solution to give a final concentration of 45 piM. One milliliter of this solution was placed in a UV quartz cuvette and upon serial additions of the biotinylated species or biotin as a positive control, the absorbance at Xo500. was monitored using a Cary 3E UVVisible spectrophotometer (Varian, Sugarland, TX, USA). The same experiment was repeated using each biotinylated LE derivative.

The structures of LE-Lys-B and LE-Lys-BX were confirmed by mass spectrometric analysis by the Protein Chemistry Core of the Interdisciplinary Center for Biotechnology Research, University of Florida, Gainesville, FL, USA. A Lasertech matrixassisted laser desorption (MALDI) time-of-flight (TOF) instrument operated at 10 kV acceleration was used with a 335 nm UV laser and oc-cyano-4-hydroxycinnamic acid as the matrix. The sample was dissolved in a solution (acetonitrile/water, 1/2 v/v) containing a





36


large excess of the matrix (5g1) and a 1 jtl aliquot was deposited onto the stainless steel laser target. Several spectra were averaged for each experiment.


Antibody-Binding of LE, Biotinylated LE Derivatives and Preformed Biotinylated LE-Avidin Complexes


The binding of LE, the biotinylated LE derivatives and preformed biotinylated LEavidin complexes to both a commercial anti-LE antiserum (Peninsula) and the antibody produced in this laboratory was tested using a standard radioimmunoassay procedure. For the assays using preformed biotinylated LE-avidin complexes, a ten-fold molar excess of avidin was added to the highest concentration of biotinylated LE in buffer and vortexed immediately. The complexes were allowed to form for 30 minutes prior to serial dilution to give the various concentrations of competitor required for the assay. The assays were set up in microcentrifuge tubes as shown in Table 2.2. The RIA buffer used consisted of 0.1 M sodium phosphate buffer pH 7 containing 0.1% w/v BSA. The tracer was 3H-LE (2.6* 10-10 M final concentration) and the competitor was various concentrations of either LE, the biotinylated LE derivatives or preformed biotinylated LE-avidin complexes. After overnight incubation at 4C, 200 p1 of an ice cold suspension containing 1.5% w/v activated charcoal and 0.15% w/v dextran in water was added to each tube (except total counts) and after a further 5 minute incubation on ice, the samples were centrifuged at 12,000g for 3 minutes. Three hundred and fifty microliters of the resultant supernatant containing the antibody-bound fraction of the tracer was then removed and added to 4 ml of Cytoscint scintillation cocktail. The radioactivity (CPM) representing the antibody-





37

bound tracer was determined using a Beckman LS 5,000 TD scintillation counter (Fullerton, CA, USA).


Table 2.2. Radioimmunoassay set up Total counts Total binding Sample RIA Buffer 420 pl 20 p1
Tracer (3.125* 10-9 M) 20 l1 20 pl 20 pl
Competitor ----- 20 ll
Antibody dilution 200 pl 200 pl

Overnight at 40C, then:
Charcoal (1. 5%)/dextran (0.15%) 200 [1 200 pl



The IC50 values (concentration of competitor displacing 50% of bound tracer) for LE, the biotinylated LE derivatives or preformed biotinylated LE-avidin complexes were determined using the MINSQ non-linear curve-fitting program (MicroMath Scientific Software, Salt Lake City, UT, USA). The data were fitted to the following model:


T *CN
B=T- +NS
CN + ICN


Where: B = CPM in the presence of competitor
T = CPM in the absence of competitor
C = competitor concentration
N = slope factor
NS = CPM under non-specific binding conditions


As far as possible, non-specific binding was determined in the presence of relatively high concentrations of competitor (2-3 orders of magnitude greater than the





38

IC50 value). When high enough concentrations of competitor could not be achieved to determine non-specific binding, the non-specific binding value obtained from the LE curve was used to fit the curve for the other competitors. Development of the Proposed Homogeneous Fluorescence Immunoassay Determination of excitation and emission maxima for fluorescence-labeled avidin

Fluorescein isothiocyanate avidin (FITC-avidin) was used as the "detector" molecule in this assay. To determine the excitation and emission maxima, 2 ml of a solution of FITC-avidin (7.5 pmol/ml) were dispensed into a quartz fluorescence cuvette and scans of the excitation and emission spectra were carried out using a Perkin Elmer LS-3B fluorescence spectrophotometer (Norwalk, CT, USA). The scans were then repeated upon addition of BLE (70 pmol/ml) to determine the fluorescence enhancement produced on interaction of these two reagents. Determination of reagent concentrations

The results of the binding experiments described above indicated that the antibody made in this laboratory in combination with the N-terminal biotinylated derivative without the spacer arm (BLE) should be used in the development of the proposed homogeneous fluorescence immunoassay for LE (see Results and Discussion). One of the first steps in the development of this assay lay in the determination of the concentrations of antibody, BLE and FITC-avidin to be used.

The decision as to the concentration of antibody to be used in the final assay (3.78*10-9 M, 1/50 dilution of purified antibody) was made based on practicality since it was anticipated that a large amount of antibody would be required for the development of





39

the homogeneous fluorescence immunoassay and a limited quantity of antibody, obtained through the exsanguination of a single rabbit was available.

To ensure low background fluorescence readings, the amount of BLE used in the final assay needed to be such that close to 100 % could be bound by the antibody present. When BLE and antibody (Ab) are present in the same solution, the equilibrium established can be described as:

BLE + Ab = BLE-Ab

According to the Law of Mass Action, the following equation can be set up: K, [BLE]- Ab]
-[BLEJ[AIJI

Which can be rearranged to give:

[BLE] 1
[BLE Ab] [Ab]Ka

If 95% of the BLE in the assay is to be bound by the amount of antibody chosen for the assay (3.78* 10-9 M, 1/50 dilution of purified antibody), the left side of the above equation can be set to equal 5. Since [Ab] = [Abtotj] [BLE-Ab], by using the K. value obtained 95

from the Scatchard plot described earlier (assuming that BLE and LE have the same affinity for the antibody), the above equation can be solved for [BLE-Ab]. [BLE-Ab] represents 95% of the concentration of BLE to be used in the assay and therefore, the total amount of BLE to be used in the final assay for 95% to be bound by the chosen antibody concentration could be calculated.

The concentration of FITC-avidin to be used in the assay was determined by finding the concentration of this reagent which will give maximum fluorescence





40

enhancement with the chosen concentration of BLE. Various concentrations of FITCavidin (200-1000 fmol/ml) in 0.1 M sodium phosphate buffer pH 7 were allowed to interact with various concentrations of BLE (0.1-20 pmol/ml) for 20 minutes, after which time, fluorescence readings were taken at X.., 482 nm and X.m 517 nm using a Perkin Elmer LS-3B fluorescence spectrophotometer (Norwalk, CT, USA). Homogeneous fluorescence immunoassay

Once the concentrations of the reagents to be used were determined, the complete homogeneous fluorescence immunoassay samples were set up as shown in Table 2.3. Total tracer samples (4 pmol BLE per ml of buffer) were also prepared to give an indication of the maximum fluorescence enhancement which could be expected due to the interaction of the total amount of tracer present with FITC-avidin. Fluorescence blank samples were set up to determine the fluorescence due to the native fluorescence of FITCavidin present in each sample. The buffer used consisted of 0.1 M sodium phosphate buffer pH 7 containing 0.1% w/v BSA. After overnight incubation at 40C, 20 .l of FITCavidin (13 pmol/ml) were added to each sample to give a final concentration of 500 fmol/ml. The interaction between the free tracer (BLE) and the "detector" molecule


Table 2.3. Homogeneous fluorescence immunoassay set-up.

Total tracer Total binding Sample Fl blank Buffer 450 pl 350 p.1 various 500 4.1
BLE (40 pmol/ml) 50 p.l 50 pt1 50 pl ---LE ---- ---- various ---Purified antibody (1/10) ---- 100 P.1 100 P 1--Total volume 500 pl 500 p1 500 p.l 500 p11





41

(FITC-avidin) was allowed to proceed for 20 minutes at room temperature in the dark, after which fluorescence readings were taken at X 482 nm and 517 nm using a Perkin Elmer LS-3B fluorescence spectrophotometer (Norwalk, CT, USA).



Results and Discussion


Antibody Production


It was anticipated that the development of the homogenous fluorescence immunoassay in particular would require large quantities of antibody. Therefore, it was decided to produce a polyclonal antibody to leucine enkephalin in this laboratory.

H H
R1-NH2 + O-=-(CH2)3- =O + H2N-R2

j 2H20


H H
R1-N=-(CH2)3.- -R2


Figure 2.3. Coupling of peptide to protein carrier using glutaraldehyde. R= peptide or
protein carrier, R2= peptide or protein carrier.


As LE is not antigenic itself, the peptide had to be conjugated to a protein carrier in order to stimulate an immune reaction in the host animal. Glutaraldehyde is a bifunctional coupling reagent that binds two compounds primarily through their amino groups. Since LE contains only one amino group at its N-terminal end, using the glutaraldehyde method of conjugation, the peptide is conjugated to the protein carrier via





42


the N-terminal end (Figure 2.3). Therefore, on inoculation with such a conjugate, one would expect the antibody that is produced to be directed against the C-terminal end of the peptide.



RIH
C=O + CH3-CH2-N=C=N-(CH2)3-N-CH3
I I
OH CH3



R1
I
C=O

II



RI
CH3





I H+
C=0 I
S + CH3-CH2-NH-C-NH-(CH2)3-N-CH3 NH I
I O CH3
R2

Figure 2.4. Coupling of peptide to protein carrier using EDC. Ri=peptide or protein
carrier, R2=peptide or protein carrier.


In contrast, EDC allows the conjugation of the peptide to the protein carrier via either the N-terminal or the C-terminal end as carbodiimides attack carboxylic groups to change them into reactive sites for free amino groups. Therefore, the carboxylic groups on





43

the protein carrier will conjugate to LE via the N-terminal amino group of the peptide and LE will also conjugate to the protein carrier through its C-terminal end by attacking the free amino groups on the protein carrier (Figure 2.4). As a result, on inoculation with a conjugate produced by the EDC method, one could expect antibodies directed against either or both the N-terminal or the C-terminal end of LE. Table 2.4. Loading ratios of LE to protein carrier by different conjugation methods. Protein carrier Conjugation method LE : protein ratio

Porcine thyroglobulin Glutaraldehyde 545 : 1
Porcine thyroglobulin EDC 70 : 1
Bovine serum albumin Glutaraldehyde 60: 1
Bovine serum albumin EDC 1.37: 1



The loading ratios of peptide to protein carrier calculated from the results of amino acid analysis are shown in Table 2.4. Due to the high immunogenicity of porcine thyroglobulin in rabbits and the fact that the LE-thyroglobulin conjugate obtained by the glutaraldehyde method showed the highest peptide loading ratio, this conjugate was selected as the inoculum for the purposes of this work.

The results of the radioimmunoassay carried out after the third boost showed that when a 1/1,000 dilution of the antiserum was used, 30% of the total radioactivity added was bound and 32% of this bound radioactivity was determined to be due to non-specific binding (Figure 2.5). At that time, this titer was deemed to be sufficient and therefore the antiserum was harvested.






44


18000
16000
1400012000 10000 U8000
6000
4000
2000
0
Total Total NonCounts Binding specific
Bindng


Figure 2.5. Bar graph showing total counts, total binding and non-specific binding of
3LFILE using 1/1,000 dilution of antibody produced in this laboratory.



Characterization of Antibody


The Scatchard plot of bound/free 3 H-LE versus bound 3 H-LE is shown in Figure 2.6. The K, value determined from the negative slope of the line drawn between the points was 3.8 7* 109 1/mol which converts to a Kd value of 2.5 8 *10-10 moll. The number of specific antibody sites present in the final incubation for this experiment was determined from the x-axis intercept to be 3.62*10-10 mol/l. Since 3. 18*108' mol/l of IgG was used in this preparation, the purified antibody contains only 1.14% specific antibody sites. By extrapolation, the stock solution of purified antibody was determined to contain 4.35* 1-7 molI of specific antibody sites. The affinity of the antibody made in this laboratory to LE compared favorably to the commercial antiserum obtained from Peninsula under the same assay conditions since IC50 values for LE using a 1/1,000 dilution of the antibody





45

produced in this laboratory (3.8*10"3 mol/assay) was similar to those obtained using a 1/3,000 dilution of the commercial antiserum (1.3 101'3 mol/assay). Peninsula claimed that IC50 values of 1.4* 1014 mol/assay could be achieved using a 1/15,000 dilution of their antiserum. However, their assay conditions involved the use of a tracer labeled with 125I instead of 3H. 1251 is a higher energy label than 3H, and therefore, Peninsula was able to use a lower concentration of tracer in their assay, which allowed them to use less antiserum, resulting in a lower IC50 value.



1A

UiU
12


-J 1
OB

kOA
02 0
0 I 1 I I I
0 5E41 1E-10 15E-10 2E-0 2 5E-10 3E-40
Bound 3H-LE (14)


Figure 2.6. Scatchard plot of bound/free 3H-LE versus bound 3H-LE


Biotinylation of Leucine Enkephalin


Since one of the proposed immunoassays required the formation of a sandwich between the anti-LE antibody, the biotinylated LE derivative and fluorescence- or enzymelabeled avidin, and the other did not, several different biotinylated derivatives were





46

synthesized and tested. Some of the biotinylated derivatives included a spacer arm between the biotin group and the peptide (Figure 2.8) as it was felt that this would increase the likelihood of the formation of a sandwich between the anti-LE antibody, the biotinylated LE derivative and enzyme-labeled avidin by reducing steric hindrance. Biotinylation took place at the N-terminal end of the peptide when LE was used as a starting material as the only free amino group available for the reaction is the N-terminal amino group of the peptide. When LE-Lys6 was used as a starting material, biotinylation took place primarily at the C-terminal end of the peptide as here, the amino group of the lysine moiety is more reactive than the N-terminal amino group under the reaction conditions used due to its greater basicity.

H
I
oN 0

Peptide-NH2 + N-O- -(CH2)4 NH

0

pH>7

H O
II II
H S N O + N-H
Pepfide-N-C-(CH2)4 NH

0Figure 2.7. Biotinylation of peptide using N-hydroxsuccinimidobiotin. Figure 2.7. Biotinylation of peptide using N-hydroxysuccinimidobiotin.





47

H
0 N 0
Peptide-NH2 + -0- -(CH,)5-N-C-(CH2 4 NH
I
H
-O

SpH>7

H O
N0
H 0 0 S + -H
Peptide-N--C-(CH2)5-N--(CH4H
H

Figure 2.8. Biotinylation of peptide using biotinamidocaproate N-hydroxysuccinimide
ester.


The reactants and products of the biotinylation reactions were successfully separated from each other using the HPLC systems described previously (Figures 2.9 and 2.10). In control injections containing LE or LE-Lys6 in concentrations equal to the initial concentrations of these compounds in the reaction mixture, the reactants were seen to elute after 9 minutes (Figure 2.9A) and 5.15 minutes (Figure 2.10 OA), respectively. The Nterminal biotinylated LE derivatives, BLE (no spacer arm) and BXLE (with spacer arm) were seen to elute distinct from the reactants at 15.3 and 16.3 minutes, respectively (Figures 2.9B and 2.9C). For the C-terminal biotinylation of LE-Lys6 to give LE-Lys6-B (no spacer arm) and LE-Lys6-BX (with spacer arm), two product peaks were seen on injection of the incubation mixtures. Based on the delayed formation in the time course of the reaction of the minor peak eluting at 24.8 minutes in Figure 2.10B and at 27.8 minutes in Figure 2.10C, these minor peaks were assumed to be di-biotinylated products biotinylated at both the N- and C-terminal ends of the peptide. Analysis by mass spectro-





48







..6
F




O0%













Figure 2.9. Chromatographs of control injection (A) showing LE eluting after 9 minutes
and reaction injections showing BLE (no spacer arm) eluting after 15.3
minutes (B) and BXLE (with spacer arm) eluting after 16.3 minutes (C).


metry (MALDI) of the major peak eluting at 14.9 minutes in Figure 2.10B and at 17.8 minutes in Figure 2.10C revealed molecular ion peaks ([M+H] ) at m/z 910.283 and 1022.1, respectively. The nominal molecular masses of the molecular ions of LE-Lys6-B and LE-Lysr-BX were calculated to be 910.11 and 1023.27, respectively. Since the mass determinations deviated from the nominal masses of mono-biotinylated derivatives by less than 0.1%, these products were confirmed as mono-biotinylated derivatives of LE-Lys6. Therefore, antibody binding experiments were carried out using these collected peaks. The





49











O O










Figure 2. 10. Chromatographs of control injection (A) showing LE-Lys6 eluting after 5.15
minutes and reaction injections showing LE-Lys6-B (no spacer arm) eluting after 14.9 minutes (B) and LE-Lyse-BX (with spacer arm) eluting after 17.8 minutes (C). Di-biotinylated products are seen eluting at 24.8 minutes (B)
and 27.8 minutes (Q).

extent of conversion from the starting materials (LE and LE-L ys) to the biotinylated derivatives was estimated from the difference in peak height of LE and LE-L ys6 in the control injection and in the injection of the reaction mixture. An extent of conversion of approximately 80% from the starting materials to the biotinylated derivatives was estimated. The results of the HABA displacement experiments confirmed that the products collected from the HPLC eluent were indeed biotinylated LE derivatives as they were seen to displace HABA from avidin (Figure 2.11). The results of this experiment confirm that






50


100
90 Bioi
80 --o-- BLE
70 -W- BXLE
60 LE-Lys-B
60 -LE-Lys-BX
50 40 30 20 10
0
0 2000 4000 6000 8000
pmoles displacer added


Figure 2.11. Displacement of HABA from avidin by biotin and biotinylated LE derivatives.


on conjugation to LE, biotin retains the ability to bind to avidin. The curvilinear nature of the displacement curves for BLE, BXLE and LE-Lys6-BX together with the observation that more of these biotinylated derivatives compared to biotin is required to displace the same amount of HABA from avidin indicates that these derivatives have a slightly lower affinity for avidin than biotin itself However, here, LE-Lys6-B is seems to have a slightly higher affinity to avidin than biotin. This effect can probably be attributed to experimental error as the concentrations of the biotinylated derivatives were calculated from estimated conversion rates based on the reduction of the peak heights of the starting materials in the biotinylation reaction. Previous investigations on the binding of biotinylated peptides or other macromolecules to avidin have shown that a spacer arm may be required to retain full affinity to avidin [Finn et al. 1984, Green et al. 1971], however this observation was not reflected in the results seen here as LE biotinylated with and without the spacer arm both showed similar affinity to avidin and in fact, LE-Lys6 biotinylated without the spacer





51

arm showed a higher affinity to avidin than LE-Lys6 biotinylated with the spacer arm (Figure 2.11). This may be attributed to the fact that LE and LE-Lys6 are rather small and therefore, on conjugation to biotin, they do not cause steric hindrance to the binding of biotin to avidin.


Antibody-Binding of LE, Biotinylated LE Derivatives and Preformed Biotinylated LE-Avidin Complexes


Using a 1/3,000 dilution of the commercial antiserum, binding curves were constructed for LE, the N-terminal biotinylated derivatives, the C-terminal biotinylated derivatives and the corresponding pre-formed complexes with avidin. The IC50 values obtained for LE, BLE, BXLE, BLE-avidin and BXLE-avidin are shown in Table 2.5. Representative binding curves for LE, BLE, BXLE, BLE-avidin and BXLE-avidin are shown in Figure 2.12.

Table 2.5. ICs50 values obtained for LE, BLE, BXLE, BLE-avidin and BXLE-avidin using
1/3,000 dilution of commercial antiserum.

Competitor IC50 (mol/assay) Mean Standard deviation

LE 1.15*10'" 1.3*10"13 2.7*10"14
1.04*10-"3
1.58*10"13

BLE 4.44*10-14 2.0*103" ----3.56*10-"


BXLE 4.42*10"5 1.3*10-14
2.26*10-14

BLE-avidin 1.89*10-10

BXLE-avidin 1.46*10"-






52


The C-terminal biotinylated LE derivatives LE-Lys6-B and LE-Lys6-BX showed no binding to the commercial antiserum in the concentrations used (data not shown) and therefore, using this preparation, they were unsuitable for the development of either of the immunoassays proposed in this study. It is possible that the inclusion of a longer spacer arm in the C-terminal biotinylated derivatives would have allowed binding of the commercial antiserum if the epitope to be recognized by the antibody remained intact on biotinylation. However, this avenue was not explored.


120

El
100 o o 9

cn U
80

60

ro I
6
.--

o
-- 40


20


0 I
10- is 10- 7 10-15 10-3 10-11 10-9 10-7

Competitor (mol/assay)
Figure 2.12. Representatives binding curves for LE (*), BLE (m), BXLE (*), BLEavidin (0) and BXLE-avidin (0) using 1/3,000 dilution of commercial
antiserum.


In these experiments, BXLE showed a ten-fold higher affinity to the commercial antiserum than LE (Table 2.5). It was postulated that since LE needs to be conjugated to a





53

protein carrier to render it immunogenic for antibody production, this biotinylated LE derivative including a spacer arm may resemble the epitope presented to the immune system more closely than LE itself and therefore, BXLE shows a higher affinity to the commercial antiserum than LE itself

Although the N-terminal biotinylated derivatives were seen to retain affinity for the commercial antiserum, a shift in affinity by 2 or 3 orders of magnitude was seen when complexes were formed between BLE or BXLE and avidin (Table 2.5), indicating that a sandwich was not formed between the commercial antiserum, BLE or BXLE and avidin. Therefore, this commercial antiserum was deemed to be unsuitable for the development of the proposed ELISA for which sandwich formation is a requirement. Although this lack of sandwich formation indicates that this preparation is suitable for the development of the proposed homogeneous fluorescence immunoassay, as it was anticipated that large quantities of antibody would be required for the development of this assay, this avenue was unfeasible. As a result, efforts involving the development of either an ELISA or an homogeneous fluorescence immunoassay using the commercial antiserum preparation were abandoned.

Using a 1/1,000 dilution of the antibody produced in this laboratory, binding curves were also constructed for LE, the N-terminal biotinylated LE derivatives, the Cterminal biotinylated LE derivatives and the corresponding pre-formed complexes with avidin. The IC50 values obtained for LE, BLE, BXLE and BXLE-avidin are shown in Table 2.6. Representative binding curves for LE, BLE, BXLE and BXLE-avidin are shown in Figure 2.13.





54


Table 2.6. ICs50 values obtained for LE, BLE, BLE-avidin, BXLE and BXLE-avidin using
1/1,000 dilution of antibody produced in this laboratory. fIC50 value for BLEavidin was estimated by fixing non-specific binding to 0% total binding and N to
1.

Competitor ICs50 (mol/assay) Mean Standard deviation

LE 4.95*103" 3.8*103" 7.3* 1014
3.70*10-3
3.78*10-13
3.44*10-13
2.79*10'"13
4.19* 10"13

BLE 6.22* 10-"13 1.1 10-12 4.0* 10"13
1.21*10-12
1.38*10-12

BLE-avidin t3.73*10-" --- --BXLE 3.53*1014 2.0*10-14
4.97*10"5

BXLE-avidin 1.70*10-" 1.4*10'"11 ----1.11*10-11


As with the commercial antiserum, the N-terminal biotinylated derivatives were seen to retain affinity for the antibody produced in this laboratory. However, a shift in affinity by almost three orders of magnitude is seen when complexes are formed between BXLE and avidin (Table 2.6), and pre-formed BLE-avidin complexes showed binding to this antibody at only the highest concentration tested (Figure 2.13).

Again, as with the commercial antiserum, BXLE showed a higher affinity than LE to the antibody produced in this laboratory. Here too, as before, it was postulated that since LE was conjugated to a porcine thyroglobulin to render it immunogenic for antibody production, this biotinylated LE derivative including a spacer arm may have resembled the





55


epitope presented to the immune system more closely than LE itself and as a result, BXLE also showed a higher affinity to this antibody than LE itself



120 i


120


980
LII



-0

60
-0
C



20

F0
H- 40


20



10-17 10-15 10-13 10-11 10-9 10-7
Competitor (molI/assay)

Figure 2.13. Representatives binding curves for LE (e), BLE (M), BXLE (*), BLEavidin (0) and BXLE-avidin (0) using 1/1,000 dilution of antibody produced
in this laboratory.


The C-terminal biotinylated LE derivatives LE-Lys -B and LE-Lys-BX showed no binding to the antibody produced in this laboratory in the concentrations used (data not shown) and therefore, in this preparation, they were also unsuitable for the development of either of the immunoassays proposed in this study. This result was expected since the antibody was produced in this laboratory following inoculation with LE conjugated to a carrier protein via the N-terminal end. Therefore, the C-terminal end of the peptide was





56

exposed as an epitope to produce an immune response in the host animal, resulting in the production of antibodies towards this end of the peptide. Since LE-Lys6-B and LE-Lys6BX are modified at the C-terminal end, one would naturally not expect them to bind to an antibody directed towards the C-terminal end of LE.

The formation of a sandwich between antibody, biotinylated LE derivative and avidin was not achieved using either the antibody produced in this laboratory or the commercial antiserum and any of the biotinylated LE derivatives synthesized. Since sandwich formation is a requirement for the successful development of the ELISA proposed in this study, efforts in this direction were abandoned at this point.

A lack of sandwich formation was observed between the antibody produced in this laboratory, BLE and avidin, and therefore, it was concluded that this combination of reagents was suitable for the development of the homogeneous fluorescence immunoassay proposed in this study.


Development of the Proposed Homogeneous Fluorescence Immunoassay Determination of excitation and emission maxima for fluorescence-labeled avidin

The excitation maximum for FITC-avidin was determined to be 482 nM. Figure 2.14 below shows the emission scans for FITC-avidin and FITC-avidin with BLE. Maximum emission was observed at 517 nm and fluorescence was seen to increase by a factor of 4 when FITC-avidin interacted with BLE. Therefore, Xc 482 nm and X., 517 nm were chosen for future fluorescence readings.






57


80



60



40
o
0

o 2 FITC-AV FITC-AV + BLE

04

0
EM 500 520 540 340 5i dI Ex 482

Figure 2.14. Emission scans with at 482 nm for FITC-avidin and FITC-avidin
with BLE.


Determination of reagent concentrations

An antibody dilution of 1/50 corresponding to a concentration of 8.7 pmol/ml of specific antibody sites was chosen for use in these experiments. Although a higher concentration of specific antibody sites in the final assay would have been preferred to afford binding of a greater number of tracer (BLE) molecules, the decision was based on practicality since a limited amount of antibody was available from the exsanguination of a single rabbit.

The calculations using the chosen antibody concentration, the K, value obtained from the Scatchard plot and the Law of Mass Action indicated that a concentration of






58


BLE of 4 pmol/ml would be 95 % bound by the chosen antibody concentration. Therefore, this concentration of tracer was chosen for use in all further experiments.



30
A
25

20S15
o
10 5
FITC-Av 200 finmoVl/ml
0
0.01 0.1 1 10 100
BLE(pmol/ml) loo- B

80

S60


40

20
FITC-Av 500 fnol/ml
0
0.01 0.1 1 10 100
BLE (pmol/ml) Figure 2.15. Plots of fluorescence versus amount of BLE added using 200 fmol/ml of
FITC-avidin (A) and 500 fmol/ml of FITC-avidin (B)



Although one would expect that about 4 times as much BLE as FITC-avidin would produce maximum fluorescence enhancement of FITC-avidin since it is known that 4 moles of biotin will bind to 1 mole of avidin, Figure 2.15A shows that a maximum fluorescence reading is obtained when 4 pmol/ml of BLE is added to 200 finol/ml of






59


FITC-avidin. The results of the 1-ABA displacement experiments (Figure 2. 11) indicated that BLE does have a lower affinity than biotin for avidin and in addition, since here FITCavidin is being used, one might expect BLE to exhibit a lower affinity for FITC-avidin than biotin exhibits for avidin. In the plot shown in Figure 2. 14A, the working scale is rather narrow and the curve obtained is not smooth as considerable noise interfered with the taking of the readings. Therefore, although a maximum fluorescence reading using 500 fmollml of FITC-avidin is seen at 10 prnol/ml rather than 4 pmollml of BLE (Figure 2.15SB), this concentration of FITC-avidin was chosen for use in this assay since less noise is observed in the readings and the working scale is more practical. Homogeneous fluorescence immunoassay

One of the first calibration curves obtained for LE using the complete homogeneous fluorescence immunoassay is shown in Figure 2.16. Here, as expected, an increase in fluorescence is seen with an increase of LE present in the sample. However, using the LE concentrations shown here, a plateau at the maximum theoretical fluorescence (total tracer value, no antibody present) was not reached. Therefore, another calibration curve was constructed using higher concentrations of LE in the samples in an attempt to reach this plateau (Figure 2.16).

Figure 2.17 shows a calibration curve for LE using the complete homogeneous fluorescence immunoassay where fluorescence readings above the maximum theoretical fluorescence were obtained at higher concentrations of LE (1I* 10-' moles/ml and above). In order to determine whether this unexpected effect could be attributed to either the







60



antibody, the tracer or the analyte, the assay was repeated omitting each of these reagents in turn.






100 --T otal tracer ---- --------------------------------- ---80




40


F1 blank 20


0
IE-Il I E-10 I E-9 I1E-8
LE (moles/mi)




Figure 2.16. Calibration curve for LE using homogeneous fluorescence immunoassay.




180

160

140

120 -Total tracer

G~100 o80


460 .. .. .. .. .. .
40 F1 blank
200
lE-IlI IE-10 IE-09 IE-08 I1E-07 I E-06
LE moles/nil




Figure 2.17. Calibration curve for LE using homogeneous fluorescence immunoassay
showing readings exceeding maximum theoretical fluorescence.







61



160
With antibody
140 120

100 Total tracer
." T a. .. .. .. .. ......... .. .


b 80

60

40 L................z



20

1E-11 IE-10 IE-09 IE-08 IE-07 IE-06
LE (mol eslml)



Figure 2.18. Calibration curve for LE using the homogeneous fluorescence immunoassay
set up with BLE (M) and without BLE(O).




160
No antibody 140 120 100
10--Total tracer 80

S60

40
Fl blank 20

0 IE-II IE-10 IE-09 IE-08 IE-07 IE-06
LE (tole s/ml)



Figure 2.19. Calibration curve for LE using the homogeneous fluorescence immunoassay
set up, omitting antibody but with BLE (U) and without BLE(O).



Figure 2.18 shows that at in the presence of the antibody, when the tracer (BLE) is


omitted from the assay, fluorescence readings began to increase above the fluorescence





62


blank levels at concentrations of LB in excess of I I O moles/mi. This effect was also observed when the antibody was omitted from the assay (Figure 2.19). When both tracer and antibody were included in the assay (Figure 2.18), fluorescence levels began to increase above fluorescence blank levels at concentrations of LB in excess of 1* 10-9 moles /mi. When antibody is omitted but tracer is included in the assay, fluorescence readings increased above total tracer levels at LB concentrations in excess of 1 *10QS moles/mI (Figure 2.19). This is the same concentration of LB at which the fluorescence readings increased above fluorescence blank levels when both tracer and antibody were omitted from the assay.


160
140 Des -Tyr LE

120 *~100
80 Total tracer
S60
40 Fl blank
20
0
I E-1lI IE-10 I E-09 IE-08 IE-07 I E-06 Des-lyr LE (moles/mi)

Figure 2.20. Calibration curve for des-Tyr' LB using the homogeneous fluorescence
immunoassay set up, omitting antibody but with BLB (0) and without
BLB(O).

These results led us to believe that at higher concentrations of LB, the analyte interacted directly with FITC-avidin to give increased fluorescence readings. It was hypothesized that this effect could be due to an interaction between the tyrosine moiety in






63


the 1 position of LE and the fluorescein groups of the "detector" molecule FITC-avidin. In order to test this hypothesis, assays were carried out in the absence of antibody, in the same way as described above. However, here des-Tyr' LE or pentaglycine were used as analytes in concentrations similar to those of LE which had been used previously.


160
Pentaglycine
140 120
100
0
60

40 ........ ... ........
20
0
IE-I IE-10 IE-09 IE-08 IE-07 IE-06 Pentaglycine (moles/ml)


Figure 2.21. Calibration curve for pentaglycine using the homogeneous fluorescence
immunoassay set up, omitting antibody but with BLE (U) and without
BLE(O3).


Figures 2.20 and 2.21 show that des-Tyr' LE and pentaglycine do not produce the game interaction with FITC-avidin as LE when used in the same concentrations in the homogeneous immunoassay set up. When BLE is included in the assay, the readings approach those obtained in the total tracer samples and when no BLE is present in the assay, the readings approach those in the fluorescence blank samples. It was therefore concluded that the effect seen at higher concentrations of LE in the homogenous






64


fluorescence immunoassay were probably due to an interaction between the tyrosine moiety of LE and the fluorescein groups of FITC-avidin.

The homogenous fluorescence immunoassay evaluated here operates as intended in the concentration range between 1* 10-9 moles/ml and 1 10-8 moles/ml of LE since at these concentrations, sufficient tracer (BLE) is displaced from the antibody by LE to produce detectable fluorescence enhancement on interaction with the detector molecule (FITCavidin), but the concentration of LE is not high enough to produce a direct interaction with FITC-avidin in the absence of BLE. Using the reagents tested, an homogeneous fluorescence immunoassay for LE operational over a wide concentration range could not be developed since at higher concentrations (1 *10O' moles/ml and above), LE interacted directly with FITC-avidin to produce fluorescence enhancement, thus interfering with the signal produced through the interaction of BLE with FITC-avidin. In this assay, the use of an antibody with a higher affinity for LE would present several advantages. Firstly, less antibody would be required to ensure maximal binding of the BLE used in the assay and consequent low background fluorescence. Secondly, less LE would be required to displace sufficient BLE from the antibody to produce a measurable signal on interaction with FITC-avidin and therefore, LE concentrations in the assay displacing the maximum amount of BLE may not reach the concentrations which were observed to produce a direct interaction with FITC-avidin and as a result interfered with the success of the assay. Thirdly, since lower concentrations of LE would be required to displace BLE from the antibody and produce a measurable signal, the sensitivity of the assay would be increased.





65

A fluorescence spectrophotometer set at close to maximum signal amplification was used to measure the fluorescence signal in the development of the homogenous fluorescence immunoassay. The use of laser-induced fluorescence may have allowed the successful development of the proposed homogeneous immunoassay as less BLE could have interacted with a lower concentration of FITC-avidin to produce a measurable signal. Therefore, less LE would have been required to displace BLE from antibody binding sites so that the higher concentrations of LE which were observed to produce a direct interaction with FITC-avidin would not have been attained.

In conclusion, although it has been demonstrated that the homogenous fluorescence immunoassay for LE proposed in this study works in theory, a fully operational assay could not be developed in practice as the concentrations of LE required in this assay to displace the maximum amount of BLE from the antibody interacted directly with FITC-avidin to produce fluorescence enhancement of the detector molecule. Existing homogeneous fluorescence immunoassays for haptens and proteins lie in the 10to 10-9 moles/mi range [Jenkins 1992] and homogeneous enzyme immunoassays are capable of limits of detection 10-" moles/nd of haptens [Engel and Khanna 1992]. It is proposed that the use of an antibody with higher affinity for LE would allow the successful development of the type of homogeneous fluorescence immunoassay for LE described in this study as lower concentrations of LE could be used and therefore the direct interaction between LE and FITC-avidin that was encountered would be avoided. It is anticipated that the successful development of this type of homogeneous immunoassay would allow the detection of LE in the 10-9 to 10-1' moles/mil range.












CHAPTER 3
A HIGH PERFORMANCE LIQUID CHROMATOGRAPHY ASSAY FOR OPIOID
PEPTIDES USING ELECTROCHEMICAL DETECTION


Introduction


The objective of this study was the evaluation of a tyrosine-specific clean-up and detection method for opioid peptides using leucine enkephalin (LE, Figure 3.1) as a model peptide. The assay described here is based on the derivatization of LE by specific enzymatic o-hydroxylation of the highly conserved tyrosine groups in the 1 position of opioid peptides by mushroom tyrosinase. This derivatization results in the formation of a catechol which is amenable to specific clean-up using boronate gels and is more easily oxidizable than the parent peptide, thus facilitating electrochemical detection.

OH


CH3 \ ,CH3
Ci
I
CH2 0 O CH2
III II I
/CH H/CCH2\ H C CH OH
NH /NII\ /C\ /NH\ / CH\ /OH
N2 C CH2 NH C CH NH C
II II I II
O O CH2 0





Tyr Gly Gly Phe Leu

Figure 3.1. Leucine enkephalin



66






67

Mushroom tyrosinase is a member of the monoxygenase class of enzymes which catalyses two successive reactions (Figure 3.2): the hydroxylation of mono-phenols (monophenolase activity) and the oxidation of o-diphenols (diphenolase activity) [Walsh 1979]. The o-quinones resulting from these two successive reactions often form high molecular weight polymerization products in vivo such as melanin. In this study, the formation of undesirable polymerization products was prevented by the addition of appropriate amounts of ascorbic acid as a reductant.


OH OH O
OH 0
/202 tyrosinase
tyrosinase reductants polymerization

R R R

Figure 3.2. Tyrosinase catalyzed reactions


Small molecules such as L- and D-tyrosine and L- and D-dopa are endogenous substrates for tyrosinase. However, enzyme activity has been shown with tyrosinecontaining di- and tri-peptides [Marumo and Waite 1986, Tellier et al. 1991] as well as LE and ME [Rosei et al. 1991, Rosei et al. 1989]. Larger proteins such as insulin, serum albumin and dehydrogenase enzymes have also been shown to be oxidized by tyrosinase [Cory et al. 1962, Cory and Frieden 1967a, Cory and Frieden 1967b, Gemant 1974, Ito et al. 1984].

In the assay described here, the enzymatic derivatization of LE by the means described above presents two analytical advantages. Firstly, the specific o-hydroxylation





68

of the tyrosine group of LE to give a more easily oxidizable catechol allows the use of a lower oxidation potentials for electrochemical detection, thus avoiding many of the disadvantages associated with high applied potentials such as high background current and baseline noise [Kim et al. 1989]. Selectivity is also compromised when high applied potentials are used as more compounds are oxidized at these high potentials. Therefore, extensive clean-up procedures to eliminate interfering peaks are often required when high applied potentials are used [Fleming and Reynolds 1988]. Secondly, the enzymatic derivatization increases the selectivity of this assay as the derivative is amenable to a specific boronate clean-up method which has previously been established for catecholamines [Eriksson and Wikstrom 1992, Higa et al. 1977, Koike et al. 1982].

R, HO
OH
/ \.LOH + HO _0/R2
6~H
innnobilized boronate gel hydroxylated LE H+1 OH"


R0H
OH



\ R2

complex

Figure 3.3. pH dependent complex formation between immobilized boronate gel and
hydroxylated LE.





69

The introduction of a 3-hydroxytyrosine group to the LE molecule by enzymatic derivatization allows the use of a specific clean-up procedure for 3,4-dihydroxyphenyl compounds using boronate gels and column chromatography. For this assay, the boronate clean-up method is based on the pH-dependent formation of a complex between immobilized boronate gel and the hydroxylated LE derivative (Figure 3.3). A complex is formed at weakly alkaline pH between ionized boronate affixed to a gel matrix and the hydroxylated LE derivative. Dissociation of the complex occurs at acidic pH.

Therefore, in the assay described here, LE in the sample is first derivatized enzymatically by mushroom tyrosinase and subsequently, the sample is subjected to cleanup through the use of a boronate gel column. Finally, the sample is quantified by high performance liquid chromatography with electrochemical detection (HPLC-ED).


Materials


Leucine enkephalin and mushroom tyrosinase were obtained from Sigma Chemical Company, St. Louis, MO, USA. Acetonitrile, methanol and trifluoroacetic acid were of HPLC grade and disodium hydrogen phosphate and citric acid were of reagent grade. These chemicals were procured from Fisher Scientific, Pittsburgh, PA, USA. Sodium dihydrogen phosphate was of molecular biology grade and was purchased from Fluka Chemie, Buchs, Switzerland. All other chemicals were of reagent grade. Double distilled water was used throughout.





70

Methods


Purification of Mushroom Tyrosinase


Mushroom tyrosinase was purified prior to use by ultra-filtration using Centricon membrane filters (molecular weight cut off 30,000, Amicon, Danvers, MA, USA). One milliliter of a solution of mushroom tyrosinase (1 mg/ml) in 0.1 M sodium phosphate buffer pH 7 was applied to the filter unit and centrifuged at 5,000 g until maximum concentration of the sample was achieved. This centrifugation step was repeated three times with the addition of an additional 2 ml of phosphate buffer prior to each centrifugation. The final concentrate was reconstituted in phosphate buffer to give a final concentration of 1 mg/mI of mushroom tyrosinase corresponding to 3870 units of activity per ml of solution. Aliquots were stored at -20'C and defrosted immediately prior to use. Enzymatic Derivatization


To characterize the enzymatic derivatization procedure, the following incubation mixture was set up in a microcentrifuge tube: 1 mM LE, 135 units/ml mushroom tyrosinase and 50 mM ascorbic acid in 0.5 M phosphate buffer pH 7.4. The reaction was allowed to proceed at room temperature with constant shaking for 60 minutes. Aliquots of this incubation mixture were then applied to an HPLC system consisting of an LDC/Milton Roy miniMetric II metering pump (Riviera Beach, FL, USA), a Negretti and Zamba injector (Southampton, UK) fitted with a 500 Id1 loop, a Perkin Elmer LC-75 spectrophotometric detector (Norwalk, CT, USA) and a Hewlett Packard HP 3394A





71

integrator (Avondale, PA, USA). The column was a Partisil 5 ODS-3 125 x 4.6 mm (Whatman Labsales, Hillsboro, OR, USA). The detection wavelength was 254 nm and the mobile phase consisted of 12.5% acetonitrile (v/v) in citrate/dipotassium phosphate buffer (pH 5) at a flow rate of 1 ml/minute.

The product peaks were collected from the HPLC eluent, organic solvent was removed by evaporating under a stream of nitrogen and the collected peaks were concentrated using a preconditioned Sep-Pak C18 preparative column (Waters Associates, Milford, MA, USA). The identity of the products was determined by electrospray ionization mass spectrometry (see Chapter 5). Electrochemical Detection


Electrochemical detection was effected using a Model 51 OOA Coulochem multielectrode electrochemical detector (ESA Inc., Bedford, MA, USA) fitted with a Model 5020 guard cell and a Model 5011 analytical cell. The complete HPLC-ED system was configured as shown in Figure 3.4.

In this system, the guard cell acts to pre-oxidize electroactive impurities in the mobile phase, thus reducing background current. The analytical cell consists of two working electrodes in series. The first (Det 1) acts to further reduce background current in the injected sample and to pre-oxidize co-eluting interferences that oxidize at potentials lower than the analyte of interest. The second working electrode (Det 2) is set to quantify the analyte. The solvent delivery system was an LDC/Milton Roy constaMetric III metering pump (Riviera Beach, FL, USA), and the injector was a Rheodyne Model 7125





72


injector (Cotati, CA, USA) fitted with a 100 jtl loop. A Spherisorb ODS2 5 pm 150 x 4.6 mm analytical column (Keystone Scientific Inc., Bellefonte, PA, USA) was used with a mobile phase of 20% acetonitrile (v/v) in monosodium phosphate buffer (100 mM, pH 5) containing 200 mg/l of sodium dodecyl sulfate. The mobile phase was freshly prepared, filtered through a 0.2 p.m membrane filter and degassed under vacuum with sonication daily, prior to use. Mobile phase flow rate was set at 1 ml/min. A Chromatopac C-R3A integrator (Shimadzu Corporation, Kyoto, Japan) was used to record the output from the control module.






Mobile
Phases
I Solvent Delivery System

inline Filter
Inlin Filt r Analytical Column odel 6010 Analytical Cell

Model 5020 Injector p* 'Other Detectors
Guard Cell



13 Chart recorder *Waste
Model 6100A Control Module

Figure 3.4. Chromatographic configuration of electrochemical detection system.


Electrochemical Characterization


LE and monohydroxylated leucine enkephalin ([HO-Tyr']-LE), the major product of the enzymatic reaction were characterized electrochemically using the HPLC-ED





73

system described above. Five nanomoles of analyte per injection were applied to the system and the responses (in LA) obtained at various potentials (+0.05 to +0.80V) at the analytical cell were recorded to allow the construction of current-voltage curves.


Boronate Clean-up


A hydrated boronate column packed with a 3 ml bed volume of Affigel 601 (BioRad Laboratories, Melville, NY, USA) was used for the clean-up of the enzymatically derived species prior to application to the HPLC-ED system. The column was pre-rinsed with 0.2 M phosphate buffer (pH 8.5) and an aliquot of the incubation mixture was applied. The column was washed with 20 ml of phosphate buffer (pH 8.5) and the analyte was eluted from the boronate column onto a pre-conditioned Sep-Pak Cl8 cartridge (Waters Associates, Milford, MA, USA) with 20 ml of aqueous 0.01% (v/v) trifluoroacetic acid (TFA). Here, the Sep-Pak C18 cartridge served to concentrate the sample as the analyte was eluted from the boronate column in a relatively large volume of aqueous 0.01% (v/v) TFA The Sep-Pak C18 column was then washed with 10 ml of aqueous 0.01% (v/v) TFA and the analyte was eluted in 2 ml of 0.01% TFA in acetonitrile. The sample was evaporated to dryness under a stream of nitrogen and the residue was reconstituted in 200 pll mobile phase prior to application to the HPLC-ED system.





74

Time Course of Enzymatic Derivatization


To determine the time course of the enzymatic derivatization at the analytical concentrations, the following incubation mixture was set up in microcentrifuge tubes: 8*10.8 M LE, 50 mM ascorbic acid and 135 units/ml mushroom tyrosinase in 300 gl of 0.5 M phosphate buffer pH 7.4. The reaction was stopped at various time points by adding 20 jtl of 1 N HCI. Two hundred an fifty micrometers of this incubation mixture was then subjected to the boronate clean-up procedure, the resulting sample was injected into the HPLC-ED system and the peak areas of the peak corresponding to the enzymatic derivative [HO-Tyr']-LE were recorded.


Extraction of Leucine Enkephalin from Cerebrospinal Fluid


Leucine enkephalin was extracted from human cerebrospinal fluid (CSF) through the use of Supelclean LC 18 solid phase extraction columns (Supelco Inc., Bellefonte, PA, USA). The columns were activated with 3 ml each of water and methanol and loaded with 100 li of spiked CSF. Subsequently, the columns were washed with 1 ml of water, 3 ml of 0.1 N HCI, 1 ml of water, 3 ml of 0.1 M borate buffer (pH 8.5) and 1 ml of water. The LE rich fraction was then eluted in 2 ml of methanol and evaporated to dryness under a stream of nitrogen.


Calibration Curves


Using the complete HPLC-ED method, including enzymatic derivatization and boronate clean-up, calibration curves were constructed for LE in both buffer and CSF. For





75

comparison, calibration curves for LE in CSF were also constructed using HPLC-ED without enzymatic derivatization or boronate clean-up.


Results


Enzymatic Derivatization


Preliminary experiments indicated that the concentration of mushroom tyrosinase (135 units/ml) used for the enzymatic derivatization of LE allowed the efficient conversion of LE to its hydroxylated derivative within 60 minutes (data not shown). A 50 mM concentration of ascorbic acid was found to be adequate to prevent or reverse the tyrosinase-induced formation of o-quinones in the enzymatic reaction mixture, thereby inhibiting the subsequent polymerization of the reaction products. Figure 3.5A shows the chromatograph of a control run where only LE and ascorbic acid are present at the incubation concentrations. Here, LE is seen eluting after 9.5 minutes, distinct from ascorbic acid eluting in the solvent front after 1.4 minutes. Two additional peaks at 5.7 minutes and 7.3 minutes are seen in the chromatograph of the incubation solution (Figure 3.5B). These two product peaks were identified by electrospray ionization mass spectroscopy (see Chapter 5) as the di- and mono-hydroxylated derivatives of LE ([(HO)2- Tyr']-LE and [HO-Tyrl]-LE, respectively). The relative intensity of the peaks obtained from mass spectrometric analysis showed that [(HO)2-Tyr']-LE is a minor product.






76


1.4 1.4
A






9.5
7.3









5.7


9.5


W

Figure 3.5. Chromatographs of a control run (A) showing LE eluting after 9.5 minutes,
distinct from ascorbic acid eluting in the solvent front after 1.4 minutes and incubation solution (B) showing the emergence of two new peaks with
retention times of 5.7 and 7.3 minutes. Electrochemical Characterization


Current-voltage curves for LE and [HO-Tyr']-LE are shown in Figure 3.6. When these curves were compared, [HO-Tyr']-LE was seen to be oxidized at considerably lower potentials than LE. As a result of this experiment, the following potentials were selected for use in the construction of calibration curves for LE using the complete HiPLC-ED






77


method (i.e. including enzymatic derivatization and boronate clean-up) and HPLC-ED without enzymatic derivatization or boronate clean-up: HPLC-ED with deriva- HPLC-ED, no derivatization and cleanup tization, no cleanup Guard cell +0.4 V +0.75 V
Analytical cell Det 1 -0.1 V +0.4 V
Analytical cell Det 2 +0.3 V +0.7 V


12
-4-- (HO-Tyrl)-LE ---O--LE
10


6 4

2 0
0 0.2 0.4 0.6 0.8
Voltage (V)

Figure 3.6. Current-voltage curves for (HO-Tyr')-LE and LE


Using these potentials, impurities in the mobile phase are pre-oxidized by the guard cell thus reducing background current and baseline noise. Impurities in the sample with relatively low oxidation potentials which might co-elute with the analyte and interfere with the signal produced by the analyte are preoxidized at Det 1. The Det 1 potential is set sufficiently low so that the analyte will not be pre-oxidized at this electrode, thus ensuring the production of a maximum signal at the analytical potential at Det 2.






78


Boronate Clean-up


By applying 35 pmoles to the boronate gel column, an average recovery of analyte of 68.67 % was achieved using the complete boronate clean-up method (SD = 6.1, n = 3). Time Course of Enzymatic Derivatization


A representative time course of the enzymatic derivatization at analytical concentrations (17 pmollinj) showing the peak area of [HO-Tyr']-LE plotted against time is shown in Figure 3.7. This experiment was carried out twice, showing the same trend each time. The plot indicates that the highest level of [HO-Tyr']-LE is seen after a 5 minute incubation and therefore, this incubation time was selected for future use in this study.

2.0E+55 1.5E+55 U 1.0E+55 5.0E+45


0.0E+05
0 20 40 60 80
Incubation time (minutes)

Figure 3.7. Time course of enzymatic derivatization showing peak height of [HO-Tyr']LE over incubation time.





79


Calibration Curves


Table 3.1 shows the limits of detection (LOD, defined as twice the baseline noise), average slope values and correlation coefficients (r2) obtained for the various calibration curves constructed for LE using this analytical approach. Raw data can be found in Appendix A. Representative calibration curves for LE in buffer and CSF using the complete HPLC-ED method are shown in Figures 3.8 and 3.9. Figure 3.10 shows a representative calibration curve for LE in CSF using HPLC-ED without enzymatic derivatization or boronate clean-up. The limits of detection for LE in CSF correspond to 8.8 pmol/ml of CSF for the complete HPLC-ED method and 176 pmol/ml of CSF for analysis by HPLC-ED without derivatization or boronate clean-up. A sample chromatograph of LE in CSF shows the analyte eluting after 6 minutes (Figure 3.11). Table 3.1. Limits of detection (LOD), average slopes and correlation coefficients (r2) for
LE in buffer or CSF using either complete HPLC-ED method or HPLC-ED
without derivatization or boronate clean-up. For raw data, see Appendix A.

Sample Method LOD Slope (SD, n) r2 (SD, n)
fmol/inj peak area/pmol/inj

Buffer complete 170 28071 (2381, 3) 0.9925 (0.005, 3)
HPLC-ED

CSF complete 360 22057 (1064, 3) 0.9944 (0.004, 3)
HPLC-ED

CSF HPLC-ED 8800 10217 (1163, 3) 0.9690 (0.024, 3)
no derivatization
no clean-up







80



7E+4 6E+4 5E+4

S4E+4 23E+4

2E+4 1 E+4

00E+0
0 0.5 1 1.5 2 2.5 3
LE(pmol/Inj) Figure 3.8. Representative calibration curve for LE in buffer using complete
HPLC-ED method.





20E+4



15 E+4



10E+4



5E+4



OOE+O
0 1 2 3 4 5 6 7 8
LE(pmol/Inj) Figure 3.9. Representative calibration curve for LE in CSF using complete
HPLC-ED method.







81



IE+6 8E+5 6E+5


c 4E+5


2E+5 0E+0O
0 20 40 60 80 100
LE (pmol/Inj) Figure 3.10. Representative calibration curve for LE in CSF using HPLC-ED without
derivatization or boronate clean-up.



I
































Figure 3.11. Sample chromatograph of LE in CSF showing peak of interest eluting after
6 minutes.






82


To give an indication of the inter-day variability associated with the complete HPLC-ED method incorporating enzymatic derivatization and boronate clean up, using the calibration curves obtained for LE in CSF, nominal concentrations in the samples were compared to found concentrations (Table 3.2). The found concentrations were determined from the calibration curve after regression analysis was repeated while omitting the data point under investigation. The relative standard deviation of the found concentrations was found to be <20% (n=3) and the accuracy was found to be within 6% of the nominal concentration.


Table 3.2. Nominal concentrations, found concentrations, relative standard deviation (SD)
and percentage accuracy calculated from 3 calibration curves for LE in CSF
using complete HPLC-ED method. For raw data, see Appendix A.

Nominal conc. Found conc, n=3 Relative SD (%) % Accuracy
(pmol/100 1.l CSF) (pmol/100 tl CSF)

1.76 1.84 20 104
3.46 3.34 12 97
6.67 6.41 12 96
10.48 11.13 1 106
17.14 17.65 9 103




Discussion


The HPLC-ED method we have evaluated for the analysis of LE in CSF compares favorably to existing HPLC methods for opioid peptides with on-line detection [de Montigny et al. 1990, Mifune et al. 1989, Muck and Henion 1989] and represent an improvement with respect to limit of detection and practicability when compared to





83

current HPLC-ED methods for enkephalins (Table 3.3) [Fleming and Reynolds 1988, Kim et al. 1989, Shibanoki et al. 1990]. The increased practicability of the analytical approach described here is characterized by the minimal precautions required in the preparation of the mobile phase (filtering and degassing under negative pressure with sonication) and is due to the fact that the enzymatic derivatization employed allowed the use of lower applied potentials for electrochemical detection (+0.3 V compared to +0.85-1.25 V for existing HPLC-ED methods). The low applied potentials used allowed the electrochemical detector to be operated at the maximum gain setting since background current and baseline noise were low. These factors, in addition to the minimal baseline drift observed, increased the ease of handling of this analytical approach since minimal precautions could be taken in the preparation of mobile phase and samples. This was demonstrated when a comparison was made between the analysis of LE in CSF using the complete HPLC-ED method incorporating enzymatic derivatization and boronate clean up and the HPLC-ED analysis of LE in CSF without derivatization or boronate clean-up. When an attempt was made to analyze LE in CSF without derivatization or boronate clean-up, considerably higher applied potentials had to be used (+0.7 V compared to +0.3 V) and as a consequence, background current, baseline noise and baseline drift were greatly increased. Therefore, the maximum gain setting of the instrument could not be used, sensitivity was compromised by a factor of 25 and the linearity of the calibration curves was reduced. In addition, the system was more difficult to handle as the mobile phase had to be continually degassed under a stream of helium in order to avoid unacceptable baseline drift.





84

Table 3.3. Comparison table of current analytical methods for opioid peptides. Reference Method Analyte Matrix Sensitivity

This study HPLC-ED LE Buffer 170 fmollinj

This study HPLC-ED LE CSF 360 fmol/inj

Fleming and HPLC-ED enkephalin rat brain 1 pmol/inj
Reynolds 1988

Shibanoki et al. HPLC-ED enkephalin plasma 550 fmollinj
1990

Kim et al. 1989 HPLC-ED enkephalin rat brain 1 pmol/inj

Muck and HPLC-MS dynorphin CSF 100 fmol/inj
Henion 1989

Mifune et al. HPLC-FL enkephalin rat brain 100 fmol/inj
1989

de Montigny et HPLC-FL LE plasma 7.7 pmollinj
al. 1990

HPLC-MS = high performance liquid chromatography with mass spectrometry, HPLC-FL = high performance liquid chromatography with fluorescence detection.


The boronate clean-up procedure used in the HPLC-ED assay described here proved to be an effective and efficient clean-up method for LE in CSF resulting in a clean chromatograph for the analyte (Figure 3.11). To avoid interfering peaks, existing HPLCED methods for enkephalins involve complex and relatively non-selective sample clean up procedures (e.g./ multiple precipitation, centrifugation and adsorption steps) prior to application of the sample to the HPLC system. By contrast, here the use of a boronate clean-up procedure, which had been previously established for use with catecholamines [Higa et al. 1977, Koike et al. 1982], increased the selectivity of this assay as only






85

derivatized peptide incorporating the dihydroxy group introduced by enzymatic derivatization was retained on the boronate gel column. The limits of detection achieved for LE using this analytical approach (170 fmol/inj in buffer and 360 fmol/inj in CSF) also compared favorably to assays for catecholamines using a boronate clean-up method and electrochemical detection (;200 fmol/inj) although the recovery of hydroxylated LE from the boronate gel (69%) was considerably lower than the recovery of catecholamines from the same matrix (80-100%) [Higa et al. 1977, Koike et al. 1982].

The HPLC-ED approach for the determination of LE in CSF described here was found to be reproducible and accurate since the relative standard deviations of various concentrations determined on different days was found to be < 20% and found concentrations were determined to be within 6% of nominal concentrations (Table 3.2).

In theory, the analytical method we have described here is applicable to a whole range of opioid peptides since the N-terminal tyrosine group which is derivatized is highly conserved throughout the entire family of opioid peptides. This method should also be applicable to the analysis of other tyrosine-containing proteins and peptides as it has been shown previously that the amino acid adjacent to the tyrosine group does not dramatically influence the tyrosinase reaction [Tellier et al. 1991]. Endogenous levels of opioid peptides in human CSF lie in the fmol/ml range as determined by radioimmunoassay [Eisenach et al. 1990, Hardebo et al. 1989, Samuelsson et al. 1993, Yaksh et al. 1990, Young et al. 1993]. Although the analytical approach described here is inadequate for the determination of endogenous levels of opioid peptides in human CSF, given sufficient





86


sample (>I "), it may be adequate for the analysis of the elevated physiological concentrations of opioid peptides to be expected in clinical studies.













CHAPTER 4
A HIGH PERFORMANCE LIQUID CHROMATOGRAPHY ASSAY FOR OPIOID
PEPTIDES USING FLUORESCENCE DETECTION


Introduction


The objective of this study was the evaluation of a tyrosine-specific analytical approach for opioid peptides using high performance liquid chromatography with fluorescence detection and leucine enkephalin (LE, Figure 4.1) as a model peptide.

OH


CH CH3
CH
I
CH2 O O H2
III II I 2

CH /N\N /CN /CH2\ /NH\ /C\ /CH\ /OH
NH2 CH2 NH C H NH C
II II I II
0 0 CH2 0





Tyr Gly Gly Phe Leu

Figure 4.1. Leucine enkephalin


As in Chapter 3, this analytical approach exploits the derivatization of LE by specific enzymatic o-hydroxylation of the highly conserved tyrosine groups in the 1




87






88


position of opioid peptides by mushroom tyrosinase. Here, the enzymatic derivatization renders the peptide suitable for subsequent fluorogenic derivatization using 1,2-diamino1,2-diphenylethane (DPE). Therefore, in this assay, the hydroxylated LE derivative obtained from the reaction with mushroom tyrosinase is oxidized in a controlled manner by potassium ferricyanide to give the corresponding quinone prior to a condensation reaction with DPE to give a fluorescent product (Figure 4.2).

Tynsinase reaction

OH OH O
OH 0
"'V02 tyrosilase
-0 1- I Polymerization
S tyrosinase ascorbic acid

R


DPE reaction

OH O
Oddation
R '~ Oil Potassium R.1iio
R OH fenrricyanide R O
Hydroxylated LE Corresponding quinone



+ CH-CHI
R"'][::i:iC: O N NH2


Condensation DPE








Ra N



Fluorescent product

Figure 4.2. Overview of derivatization reactions for HPLC-FL assay





89

Materials


Leucine enkephalin and mushroom tyrosinase were obtained from Sigma Chemical Company, St. Louis, MO, USA. Acetonitrile, methanol and trifluoroacetic acid were of HPLC grade and disodium hydrogen phosphate, sodium dihydrogen phosphate, citric acid, potassium chloride and potassium ferricyanide were of reagent grade. These chemicals as well as Scintiverse II scintillation cocktail were procured from Fisher Scientific, Pittsburgh, PA, USA. Tetrabutylammnonium (TBA) was purchased from the Eastman Kodak Company, Rochester, NY, USA and [tyrosyl-3,5-3H(N)]-leucine enkephalin (3HLE) was obtained from NEN Research Products, Dupont Company, Wilmington, DE, USA. 1,2-Diamino-1,2-diphenylethane (DPE) was synthesized according to Irving and Parkins [Irving and Parkins 1965]. Briefly, benzaldehyde (1.9 equivalents) was refluxed with ammonium acetate (1 equivalent) for 3 hours, the resulting precipitate was collected, and washed with ethanol. This precipitate was then hydrolyzed with 33 % v/v H2SO4, benzoic acid and benzaldehyde were removed by steam distillation, and DPE was precipitated by neutralization with ammonium hydroxide. After recrystallization from petroleum ether, the product had a melting point of 118-1190C (literature, 1200C) [Irving and Parkins 1965] and a 'H nuclear magnetic resonance spectrum which supported the assigned structure. All other chemicals were of reagent grade. Double distilled water was used throughout.






90

Methods


Purification of Mushroom Tyrosinase


Mushroom tyrosinase was purified prior to use by ultra-filtration using Centricon membrane filters (molecular weight cut off 30,000, Amicon, Danvers, MA, USA). One milliliter of a solution of mushroom tyrosinase (1 mg/ml) in 0.1 M phosphate buffer pH 7 was applied to the filter unit and centrifuged at 5,000 g until maximum concentration of the sample was achieved. This centrifugation step was repeated three times with the addition of an additional 2 ml of phosphate buffer prior to each centrifugation. The final concentrate was reconstituted in phosphate buffer to give a final concentration of 1 mg/ml of mushroom tyrosinase corresponding to 3870 units of activity per ml of solution. Aliquots were stored at -20'C and defrosted immediately prior to use. Enzymatic Derivatization


To characterize the derivatization procedure, LE (1 mM) in 0.5 M phosphate buffer pH 7.4 was reacted with mushroom tyrosinase (135 units/ml) in the presence of ascorbic acid (50 raM) at room temperature with constant shaking. The product peaks obtained from this enzymatic derivatization were collected from the HPLC eluent, organic solvent was removed by evaporation under a stream of nitrogen and the collected peaks were concentrated using a Sep-Pak C18 preparative column (Waters Associates, Milford, MA, USA). The identity of the products was determined by electrospray ionization mass spectrometry (see Chapter 5).






91

To determine the time course of the reaction, at various time points, aliquots of this incubation solution were applied to an HPLC system consisting of an LDC/Milton Roy miniMetric II metering pump (Riviera Beach, FL, USA), a Negretti and Zamba injector (Southampton, UK) fitted with a 500 [l loop, a Perkin Elmer LC-75 spectrophotometric detector (Norwalk, CT, USA) and a Hewlett Packard HP 3394A integrator (Avondale, PA, USA). The column was a Partisil 5 ODS-3 125 x 4.6 mm (Whatman Labsales, Hillsboro, OR, USA). The detection wavelength was 254 nm and the mobile phase consisted of 12.5% acetonitrile (v/v) in citrate/dipotassium phosphate buffer (pH 5) at a flow rate of I ml/minute.


Extraction of Leucine Enkephalin from Cerebrospinal Fluid


Leucine enkephalin was extracted from human cerebrospinal fluid (CSF) through the use of Supelclean LC 8 solid phase extraction columns (Supelco Inc., Bellefonte, PA, USA). The columns were activated with 3 ml each of water and methanol and loaded with 100 pl of spiked CSF. Subsequently, the columns were washed with 1 ml of water, 3 ml of 0.1 N HCI, 1 ml of water, 3 ml of 0.1 M borate buffer (pH 8.5) and 1 ml of water. The LE-rich fraction was then eluted in 2 ml of methanol and evaporated to dryness under a stream of nitrogen.

To test the recovery of LE from CSF using this extraction procedure, 100 pl of CSF spiked with 3H-LE (17,300 counts per minute (CPM)) was applied to an extraction column and the procedure described above was carried out. The final methanolic LE-rich fraction was collected in 1 ml aliquots, 4 ml of Scintiverse II scintillation cocktail were





92

added and the radioactivity (CPM) was measured using a Beckman LS 5,000 TD scintillation counter (Fullerton, CA, USA). As a control, the radioactivity in a sample containing the same amount of 3H-LE as the spiked CSF in 1 ml of methanol was also measured.


Calibration curves


Calibration curves for LE (0.5-7 pmol/inj) in both buffer and CSF were constructed using this HPLC-FL analytical approach.

For the tyrosinase reaction, various concentrations of LE in 0.5 M sodium phosphate buffer pH 7.4 or spiked CSF extracts reconstituted in 0.5 M sodium phosphate buffer pH 7.4 were reacted with mushroom tyrosinase (135 units/ml) in the presence of ascorbic acid (50 mM). The total volume of this incubation mixture was 100 gll. After a 60 minute incubation at room temperature with constant shaking, 449 p1 of an oxidizing solution containing 9.45 mg/ml potassium ferricyanide, 17.85 mg/ml potassium chloride and 55 % acetonitrile (v/v) was added followed by 45.5 pll of a solution of DPE containing 20 mg/mI in 0.1 N HCI. The fluorogenic reaction was allowed to proceed in the dark for 60 minutes at room temperature with constant shaking. A 250 tl aliquot of this reaction mixture, representing 42 % of the total final reaction mixture, was then injected into an HPLC system consisting of an LDC/Milton Roy miniMetric II metering pump (Riviera Beach, FL, USA), a Rheodyne Model 7125 injector (Cotati, CA, USA) fitted with a 500 [l loop, a Nucleosil C18 column (5 tm, 150 x 4.6 mm, Keystone Scientific, Bellefonte, PA, USA) and a Perkin Elmer 650S spectrofluorodetector (Norwalk, CT, USA) set at X 345






93

nm and ,, 480 nm with slit widths of 12.5 nm. A BD 41 chart recorder (Kipp and Zonen, Delft, The Netherlands) and a Hewlett Packard HP 3394A integrator (Avondale, PA, USA) recorded the output from the spectrofluorodetector. The mobile phase contained 32 % acetonitrile (v/v) and 67 mM TBA in citrate/dipotassium phosphate buffer (pH 5). The system was operated at a mobile phase flow rate of 1 ml/min.

An attempt was made to optimize the fluorogenic derivatization reaction by concentrating the fluorogenic reagents so that 100 jil of an oxidizing solution containing 40.1 mg/ml of potassium ferricyanide and 75.8 mg/ml of potassium chloride, 100 l of acetonitrile and 25 tl of a solution of DPE containing 20 mg/ml in 0.1 N HCI were added to the buffer samples or the CSF samples reconstituted in buffer. After a 60 minute incubation in the dark with constant shaking, a 250 pl aliquot of this mixture, representing 77 % of the total final reaction mixture, was injected into the HPLC system described above.

Since experiments conducted in parallel indicated that at low concentrations of analyte, the highest yield of enzymatically derived species was obtained after a 5 minute tyrosinase incubation time (see Chapter 3), an attempt was also made to optimize the derivatization reaction by reducing the tyrosinase incubation time to 5 minutes.


Results


Enzymatic Derivatization


Preliminary experiments indicated that the concentration of mushroom tyrosinase (135 units/ml) used for the enzymatic derivatization of LE allowed the efficient conversion




Full Text
51
arm showed a higher affinity to avidin than LE-Lys6 biotinylated with the spacer arm
(Figure 2.11). This may be attributed to the fact that LE and LE-Lysc are rather small and
therefore, on conjugation to biotin, they do not cause steric hindrance to the binding of
biotin to avidin.
Antibody-Binding of LE. Biotinylated LE Derivatives and Preformed Biotinylated
LE-Avidin Complexes
Using a 1/3,000 dilution of the commercial antiserum, binding curves were
constructed for LE, the N-terminal biotinylated derivatives, the C-terminal biotinylated
derivatives and the corresponding pre-formed complexes with avidin. The IC50 values
obtained for LE, BLE, BXLE, BLE-avidin and BXLE-avidin are shown in Table 2.5.
Representative binding curves for LE, BLE, BXLE, BLE-avidin and BXLE-avidin are
shown in Figure 2.12.
Table 2.5. IC50 values obtained for LE, BLE, BXLE, BLE-avidin and BXLE-avidin using
1/3,000 dilution of commercial antiserum.
Competitor
IC50 (mol/assay)
Mean
Standard deviation
LE
1.15* 10'13
1.04* 10'13
1.58*1 O'13
1.3*10'13
2.7*10'14
BLE
4.44* 10'14
3.56* 10"13
2.0* 10'13
BXLE
4.42* 1015
2.26* 10'14
1.3*1 O'14
BLE-avidin
1.89*10'10
BXLE-avidin
1.46*10'


APPENDIX A
DATA FOR HPLC-ED APPROACH
Curve 1
Tyrosinase derivatization, boronate clean up
Matrix: Buffer
Equation: y = 2761 lx + 2288, r2 = 0.9977
LE mol/inj
Peak area
0
0
5.33*1013
17597
1.07* 10"12
32566
2.13*1 O'12
64947
3.20* 10'12
87745
4.80*1 O'12
134842
Curve 2
Tyrosinase derivatization, boronate clean up
Matrix: Buffer
Equation: y = 30648x -1752, r2 = 0.9876
LE mol/inj
Peak area
0
0
3.37* 10'13
5077
5.06* 10"13
15565
8.44* 10'13
17240
1.18*10'12
40429
1.69*1 O'12
52711
2.53 1012
74861
3.37* 10'12
100722
126


85
derivatized peptide incorporating the dihydroxy group introduced by enzymatic
derivatization was retained on the boronate gel column. The limits of detection achieved
for LE using this analytical approach (170 fmol/inj in buffer and 360 fmol/inj in CSF) also
compared favorably to assays for catecholamines using a boronate clean-up method and
electrochemical detection (200 fmol/inj) although the recovery of hydroxylated LE from
the boronate gel (69%) was considerably lower than the recovery of catecholamines from
the same matrix (80-100%) [Higa et al. 1977, Koike et al. 1982],
The HPLC-ED approach for the determination of LE in CSF described here was
found to be reproducible and accurate since the elative standard deviations of various
concentrations determined on different days was found to be < 20% and found
concentrations were determined to be within 6% of nominal concentrations (Table 3.2).
In theory, the analytical method we have described here is applicable to a whole
range of opioid peptides since the N-terminal tyrosine group which is derivatized is highly
conserved throughout the entire family of opioid peptides. This method should also be
applicable to the analysis of other tyrosine-containing proteins and peptides as it has been
shown previously that the amino acid adjacent to the tyrosine group does not dramatically
influence the tyrosinase reaction [Tellier et al. 1991], Endogenous levels of opioid
peptides in human CSF lie in the fmol/ml range as determined by radioimmunoassay
[Eisenach et al. 1990, Hardebo et al. 1989, Samuelsson et al. 1993, Yaksh et al. 1990,
Young et al. 1993], Although the analytical approach described here is inadequate for the
determination of endogenous levels of opioid peptides in human CSF, given sufficient


100
not shown). Attempts to improve the linearity of a calibration curve for LE in spiked CSF
samples using a 5 minute tyrosinase reaction time by improving the extraction procedure
or the incubation conditions were not pursued.
Figure 4.8. Calibration curve for LE in buffer samples using 5 minute tyrosinase reaction
time in HPLC-FL method.
Discussion
The FIPLC-FL analytical approach for LE developed in this study yielded detection
limits of 500 fmol per injection in both buffer and spiked CSF samples, corresponding to
12 pmoles of LE per ml of CSF. This approach shows similar sensitivity or represents an
improvement in the limit of detection by one order of magnitude when compared to
existing tyrosine-specific HPLC-FL methods with pre-column fluorescence derivatization
[Ishida et al. 1986, Kai et al. 1988, Nakano et al. 1987, Zhang et al. 1991], The


16
Instrumental Approaches
Other methods which have been used in the analysis of opioid peptides include
HPLC combined with electrochemical detection (HPLC-ED), fluorescence detection
(HPLC-FL) or mass spectrometry (HPLC-MS).
Electrochemical detection
HPLC in conjunction with electrochemical detection is an analytical method which
offers several advantages. It provides selectivity, as only those compounds which are
oxidizable or reducible at the applied potential will be detected. Multiple electrode
detectors can be used to pre-oxidize contaminants in the mobile phase and the sample
prior to detection of the analyte of interest, thus increasing sensitivity by improving signal
to noise ratios. The method is versatile and the cost of instrumentation and reagents is
relatively low. In electrochemical detection, peak current ratios are obtained from the ratio
of the peak heights obtained (i.e. current generated) when two different voltages are
applied to the same amount of sample. These peak current ratios are characteristic for
each compound, much like absorbance ratios in ultraviolet detection, and therefore
qualitative information about the analyte can be derived from them.
A review of the literature reveals that, to date, HPLC-ED methods for opioid
peptides are less sensitive than other methods such as RIA [Fleming and Reynolds 1988,
Kim et al. 1989, Mousa and Couri 1983], This may be attributed to the fact that the high
potentials necessary to detect opioid peptides using these methods (+0.9-1.25 V compared
to +0.3-0.4 V used in HPLC-ED assays for catechols with sensitivities in the fmol/inj
range [Higa et al 1977, Koike et al. 1982]) compromise sensitivity as background current


BIOGRAPHICAL SKETCH
Veronique Larsimont was born in Taipei, Taiwan in 1967 and spent her childhood
traveling the world with her parents until she went to boarding school at Dollar Academy,
Dollar, Scotland in 1979. She earned a B.Sc. (Honours) in pharmacy from Heriot-Watt
University, Edinburgh, Scotland in 1988 and worked for Lilly Research and Development
in Windlesham, Surrey, England and Richard Clitherow Ltd., Liverpool, England during
her pre-registration year prior to becoming a member of the Royal Pharmaceutical Society
of Great Britain in 1989. In the same year, she began her graduate studies in the
Department of Pharmaceutics at the University of Florida and completed her doctoral
dissertation in 1994. She now intends to pursue a career in the pharmaceutical industry in
Europe.
145


20
been previously developed for dynorphin [Hochhaus and Hu 1990] and P-endorphin
[Hochhaus and Sadee 1988], This assay was based on the avidin-biotin system whereby
avidin exhibits an extremely high affinity for biotin and biotinylated species. Here, LE and
biotinylated LE derivative compete for antibody binding sites and subsequently, on
separation of the antibody-bound and free fractions, the antibody-bound biotinylated
species is detected through the use of an avidin-enzyme complex. The success of this type
of assay depends on the formation of a sandwich between the antibody, the biotinylated
LE derivative and enzyme-labeled avidin.
OH
CH, v CH,
3\ / 3
CH
I
O
O CH,
/\/\/\/2\/\/ \ / \
r rH2 NH C CH NH C
O
O CH,
O
Q
Tyr Gly
Figure 1.5. Leucine enkephalin.
Gly
Phe
Leu
The second approach was an homogeneous or separation-free fluorescence
immunoassay which also made use of the avidin-biotin system, and was based on an
observation by Al-Hakiem and co-workers [A1 Hakiem et al. 1981] that the binding of


40
enhancement with the chosen concentration of BLE. Various concentrations of FITC-
avidin (200-1000 fmol/ml) in 0.1 M sodium phosphate buffer pH 7 were allowed to
interact with various concentrations of BLE (0.1-20 pmol/ml) for 20 minutes, after which
time, fluorescence readings were taken at Xexc 482 nm and Xem 517 nm using a Perkin
Elmer LS-3B fluorescence spectrophotometer (Norwalk, CT, USA).
Homogeneous fluorescence immunoassay
Once the concentrations of the reagents to be used were determined, the complete
homogeneous fluorescence immunoassay samples were set up as shown in Table 2.3.
Total tracer samples (4 pinol BLE per ml of buffer) were also prepared to give an
indication of the maximum fluorescence enhancement which could be expected due to the
interaction of the total amount of tracer present with FITC-avidin. Fluorescence blank
samples were set up to determine the fluorescence due to the native fluorescence of FITC-
avidin present in each sample. The buffer used consisted of 0.1 M sodium phosphate
buffer pH 7 containing 0.1% w/v BSA. After overnight incubation at 4C, 20 pi of FITC-
avidin (13 pmol/ml) were added to each sample to give a final concentration of
500 fmol/ml. The interaction between the free tracer (BLE) and the detector molecule
Table 2.3. Homogeneous fluorescence immunoassay set-up.
Total tracer
Total binding
Sample
FI blank
Buffer
450 pi
350 pi
various
500 pi
BLE (40 pmol/ml)
50 pi
50 pi
50 pi
LE
various
Purified antibody (1/10)
100 pi
100 pi
Total volume
500 pi
500 pi
500 pi
500 pi


37
bound tracer was determined using a Beckman LS 5,000 TD scintillation counter
(Fullerton, CA, USA).
Table 2.2. Radioimmunoassay set up
Total counts
Total binding
Sample
RIA Buffer
420 pi
20 pi
Tracer (3.125*1 O'9 M)
20 pi
20 pi
20 pi
Competitor
20 pi
Antibody dilution
200 pi
200 pi
Overnight at 4C, then:
Charcoal (1,5%)/dextran (0.15%)
200 pi
200 pi
The IC5o values (concentration of competitor displacing 50% of bound tracer) for
LE, the biotinylated LE derivatives or preformed biotinylated LE-avidin complexes were
determined using the MINSQ non-linear curve-fitting program (MicroMath Scientific
Software, Salt Lake City, UT, USA). The data were fitted to the following model:
T*CN
CN + /C5"
+ NS
Where: B = CPM in the presence of competitor
T = CPM in the absence of competitor
C = competitor concentration
N = slope factor
NS = CPM under non-specific binding conditions
As far as possible, non-specific binding was determined in the presence of
relatively high concentrations of competitor (2-3 orders of magnitude greater than the


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
NOVEL APPROACF1ES FOR THE DETERMINATION OF LEUCINE ENKEPHALIN
AS A MODEL FOR OPIOID PEPTIDES
By
Vronique Larsimont
August 1994
Chairman: Giinther Hochhaus
Major Department: Pharmaceutics
The focus of this dissertation was the evaluation of novel analytical approaches for
opioid peptides by immunoassay or high performance liquid chromatography (HPLC)
using leucine enkephalin (LE) as a model peptide.
The proposed immunoassays are based on the high affinity exhibited by avidin for
biotin (Kd=1015 mol/1). The successful development of the enzyme-linked immunosorbent
assay relied on the formation of a sandwich between anti-LE antibody, a biotinylated LE
derivative and avidin, whereas the successful development of the homogeneous
fluorescence immunoassay depended on a lack of sandwich formation. The formation of a
sandwich was not achieved using any combination of the two anti-LE antibody
preparations and several biotinylated LE derivatives tested, and therefore efforts in this
direction were abandoned. However, using a polyclonal antibody produced in this
laboratory, an N-terminal biotinylated LE derivative without a spacer arm and fluorescein
vi


103
As with the HPLC-EC method described in Chapter 3, the HPLC-FL analytical
approach described here should also be applicable to a whole range of opioid peptides
since the N-terminal tyrosine group which is initially derivatized is highly conserved
throughout the entire family of opioid peptides. This approach should also be applicable to
the analysis of other tyrosine-containing proteins and peptides as it has been shown
previously that the amino acid adjacent to the tyrosine group does not dramatically
influence the tyrosinase reaction [Tellier et al. 1991], Endogenous levels of opioid
peptides in human CSF, as determined by radioimmunoassay lie in the fmol/ml range
[Eisenach et al. 1990, Hardebo et al. 1989, Samuelsson et al. 1993, Yaksh et al. 1990,
Young et al. 1993], Therefore, here too, although this analytical approach in its present
form is inadequate for the determination of endogenous levels of opioid peptides in human
CSF, given sufficient sample (>lml) it may also be adequate for the analysis of the
elevated physiological concentrations of opioid peptides to be expected in clinical studies.


55
epitope presented to the immune system more closely than LE itself and as a result, BXLE
also showed a higher affinity to this antibody than LE itself.
Figure 2.13. Representatives binding curves for LE (), BLE (), BXLE (), BLE-
avidin () and BXLE-avidin (0) using 1/1,000 dilution of antibody produced
in this laboratory.
The C-terminal biotinylated LE derivatives LE-Lys6-B and LE-Lys6-BX showed
no binding to the antibody produced in this laboratory in the concentrations used (data not
shown) and therefore, in this preparation, they were also unsuitable for the development of
either of the immunoassays proposed in this study. This result was expected since the
antibody was produced in this laboratory following inoculation with LE conjugated to a
carrier protein via the N-terminal end. Therefore, the C-terminal end of the peptide was


76
A
Figure 3.5. Chromatographs of a control run (A) showing LE eluting after 9.5 minutes,
distinct from ascorbic acid eluting in the solvent front after 1.4 minutes and
incubation solution (B) showing the emergence of two new peaks with
retention times of 5.7 and 7.3 minutes.
Electrochemical Characterization
Current-voltage curves for LE and [HO-Tyr'j-LE are shown in Figure 3.6. When
these curves were compared, [HO-Tyr']-LE was seen to be oxidized at considerably lower
potentials than LE. As a result of this experiment, the following potentials were selected
for use in the construction of calibration curves for LE using the complete HPLC-ED


92
added and the radioactivity (CPM) was measured using a Beckman LS 5,000 TD
scintillation counter (Fullerton, CA, USA). As a control, the radioactivity in a sample
containing the same amount of 3H-LE as the spiked CSF in 1 ml of methanol was also
measured.
Calibration curves
Calibration curves for LE (0.5-7 pmol/inj) in both buffer and CSF were
constructed using this HPLC-FL analytical approach.
For the tyrosinase reaction, various concentrations of LE in 0.5 M sodium
phosphate buffer pH 7.4 or spiked CSF extracts reconstituted in 0.5 M sodium phosphate
buffer pH 7.4 were reacted with mushroom tyrosinase (135 units/ml) in the presence of
ascorbic acid (50 mM). The total volume of this incubation mixture was 100 pi. After a
60 minute incubation at room temperature with constant shaking, 449 pi of an oxidizing
solution containing 9.45 mg/ml potassium ferricyanide, 17.85 mg/ml potassium chloride
and 55 % acetonitrile (v/v) was added followed by 45.5 pi of a solution of DPE containing
20 mg/ml in 0.1 N HC1. The fluorogenic reaction was allowed to proceed in the dark for
60 minutes at room temperature with constant shaking. A 250 pi aliquot of this reaction
mixture, representing 42 % of the total final reaction mixture, was then injected into an
FLPLC system consisting of an LDC/Milton Roy miniMetric II metering pump (Riviera
Beach, FL, USA), a Rheodyne Model 7125 injector (Cotati, CA, USA) fitted with a 500
pi loop, aNucleosil Cig column (5pm, 150 x 4.6 mm, Keystone Scientific, Bellefonte, PA,
USA) and a Perkin Elmer 650S spectrofluorodetector (Norwalk, CT, USA) set at A.ex 345


89
Materials
Leucine enkephalin and mushroom tyrosinase were obtained from Sigma Chemical
Company, St. Louis, MO, USA. Acetonitrile, methanol and trifluoroacetic acid were of
HPLC grade and disodium hydrogen phosphate, sodium dihydrogen phosphate, citric acid,
potassium chloride and potassium ferricyanide were of reagent grade. These chemicals as
well as Scintiverse II scintillation cocktail were procured from Fisher Scientific,
Pittsburgh, PA, USA. Tetrabutylammonium (TBA) was purchased from the Eastman
Kodak Company, Rochester, NY, USA and [tyrosyl-3,5-3H(N)]-leucine enkephalin (3H-
LE) was obtained from NEN Research Products, Dupont Company, Wilmington, DE,
USA. 1,2-Diamino-1,2-diphenylethane (DPE) was synthesized according to Irving and
Parkins [Irving and Parkins 1965], Briefly, benzaldehyde (1.9 equivalents) was refluxed
with ammonium acetate (1 equivalent) for 3 hours, the resulting precipitate was collected,
and washed with ethanol. This precipitate was then hydrolyzed with 33 % v/v H2S04,
benzoic acid and benzaldehyde were removed by steam distillation, and DPE was
precipitated by neutralization with ammonium hydroxide. After recrystallization from
petroleum ether, the product had a melting point of 118-119C (literature, 120C) [Irving
and Parkins 1965] and a H nuclear magnetic resonance spectrum which supported the
assigned structure. All other chemicals were of reagent grade. Double distilled water was
used throughout.


91
To determine the time course of the reaction, at various time points, aliquots of
this incubation solution were applied to an HPLC system consisting of an LDC/Milton
Roy miniMetric II metering pump (Riviera Beach, FL, USA), a Negretti and Zamba
injector (Southampton, UK) fitted with a 500 pi loop, a Perkin Elmer LC-75
spectrophotometric detector (Norwalk, CT, USA) and a Hewlett Packard HP 3394A
integrator (Avondale, PA, USA). The column was a Partisil 5 ODS-3 125 x 4.6 mm
(Whatman Labsales, Hillsboro, OR, USA). The detection wavelength was 254 nm and the
mobile phase consisted of 12.5% acetonitrile (v/v) in citrate/dipotassium phosphate buffer
(pH 5) at a flow rate of 1 ml/minute.
Extraction of Leucine Enkephalin from Cerebrospinal Fluid
Leucine enkephalin was extracted from human cerebrospinal fluid (CSF) through
the use of Supelclean LCig solid phase extraction columns (Supelco Inc., Bellefonte, PA,
USA). The columns were activated with 3 ml each of water and methanol and loaded with
100 pi of spiked CSF. Subsequently, the columns were washed with 1 ml of water, 3 ml of
0.1 N HC1, 1 ml of water, 3 ml of 0.1 M borate buffer (pH 8.5) and 1 ml of water. The
LE-rich fraction was then eluted in 2 ml of methanol and evaporated to dryness under a
stream of nitrogen.
To test the recovery of LE from CSF using this extraction procedure, 100 pi of
CSF spiked with 3H-LE (17,300 counts per minute (CPM)) was applied to an extraction
column and the procedure described above was carried out. The final methanolic LE-rich
fraction was collected in 1 ml aliquots, 4 ml of Scintiverse II scintillation cocktail were


57
Figure 2.14. Emission scans with X.exc at 482 nm for FITC-avidin and FITC-avidin
with BLE.
Determination of reagent concentrations
An antibody dilution of 1/50 corresponding to a concentration of 8.7 pmol/ml of
specific antibody sites was chosen for use in these experiments. Although a higher
concentration of specific antibody sites in the final assay would have been preferred to
afford binding of a greater number of tracer (BLE) molecules, the decision was based on
practicality since a limited amount of antibody was available from the exsanguination of a
single rabbit.
The calculations using the chosen antibody concentration, the Ka value obtained
from the Scatchard plot and the Law of Mass Action indicated that a concentration of


34
Milton Roy Spectromonitor 3100 variable wavelength detector (Riviera Beach, FL, USA).
The detection wavelength was set at 210 nm. Mobile phase A consisted of 90 % v/v
aqueous trifluoroacetic acid (0.02% v/v) and 10 % v/v acetonitrile and mobile phase B
consisted of 10 % v/v aqueous trifluoroacetic acid (0.02% v/v) and 90 % v/v acetonitrile.
A gradient of 90 % mobile phase A and 10 % mobile phase B to 50 % each of mobile
phases A and B in 30 minutes was run at a flow rate of 1 ml/min. A control injection of the
incubation mixture was also made into the HPLC system immediately after preparation to
allow the calculation of the extent of conversion of LE to BLE. N-terminal biotinylated
LE incorporating a spacer arm between the biotin group and the peptide (BXLE) was
synthesized in the same manner except that biotinamidocaproate N-hydroxysuccinimide
ester (BXHS) was used instead of BHS in the incubation mixture. The peaks
corresponding to BLE and BXLE were collected from the HPLC eluent and the volatile
components (acetonitrile and trifluoroacetic acid) were evaporated under a stream of
nitrogen at 30C. The biotinylated LE derivatives were stored at 4C and were used within
two weeks of synthesis.
C-terminal biotinylated leucine enkephalin-Lys6 (LE-Lys6-B) was synthesized by
allowing 73 nmoles of leucine enkephalin-Lys6 (LE-Lys6), 142 nmoles of BHS and 90
nmoles of triethanolamine in 150 pi of DMSO to incubate for two hours at room
temperature. The product of the reaction was separated from the reagents by injecting this
incubation mixture into the same gradient HPLC system described above except that a
gradient of 90 % mobile phase A and 10 % mobile phase B to 70 % mobile phase A and
30% mobile phase B in 30 minutes was run at a flow rate of 1 ml/min. A control injection


CHAPTER 5
LEUCINE ENKEPHALIN-TYROSINASE REACTION PRODUCTS -
IDENTIFICATION AND BIOLOGICAL ACTIVITY
Introduction
Tyrosinase is a copper containing enzyme which catalyses the ortho-hydroxylation
of phenols and the subsequent oxidation of the resulting catechols to o-quinones. It is
common throughout nature and plays a central role in the biosynthesis of both
norepinephrine and melanin. In Chapters 3 and 4 of this dissertation, we have shown that
tyrosinase will react with the tyrosine-containing peptide leucine enkephalin (LE).
Previously, our group and others have shown that tyrosinase also reacts with other
tyrosine-containing peptides to give hydroxylated products [Marumo and Waite 1986,
Rosei et al. 1991, Rosei et al. 1989, Tellier et al. 1991].
Numerous studies have revealed a loss of activity in enkephalins and other opioid
peptides when the Tyr1 moiety is absent (for reviews, see Hansen and Morgan 1984 and
Shimohigashi 1986), however, few have shown the effect of modification of the aromatic
side-chain of this residue. Therefore, as tyrosinase is known to react with tyrosine-
containing peptides, our goal in this study was to determine the structure of the products
formed when tyrosinase reacts with LE and to define the affinity of the products modified
at the Tyr1 residue to opioid receptors in rat brain homogenate.
104


144
Tellier, M, R. J. Prankerd and G. Hochhaus (1991). Journal of Pharmaceutical and
Biomedical Analysis 9: 557.
Udenfriend, S., S. Stein, P. Bohlen, W. Dairman, W. Leimgruber and M. Weigele (1972).
Science 178: 871.
Umeda, M, Y. Ishimori, K. Yoshikawa, M. Takada and T. Yasuda (1986). Journal of
Immunological Methods 95: 15.
van den Beld, C. M. B., U. R. B. Tjaden, N. J. Reinhoud, D. S. Stegehuis and J. van der
Greef (1990). Journal of Controlled Release 13: 129.
Venn, R. F. (1987). Journal of Chromatography 423: 93.
Venturelli, F., G. Roscetti, R. Possenti, F. Vita and G. Roda (1985). Neurochemistry
Research 10: 333.
Walsh, C. (1979). in Enzymatic Reaction Mechanisms (pg. 461) San Francisco: W. H.
Freeman & Co.
Wang, Z., E. J. Bilsky, F. Porreca and W. Sadee (1994). Life Sciences 54: 339.
Whitehouse, C. M., R. N. Dreyer, M. Yamashita and J. B. Fenn (1985). Analytical
Chemistry 57: 675.
Wilchek, M. and E. A Bayer (1988). Analytical Biochemistry 171: 1.
Yaksh, T. L., S. W. Carmichael, S. L. Stoddard, G. M. Tyce, P. J. Kelly, D. Lucas, J. van
Heerden, J. E. Ahlskog and D. E. Byer (1990). Journal of Laboratory and Clinical
Medicine 115: 346.
Young, R. F., F. W. Bach, N. A. Van and T. L. Yaksh (1993). Journal of Neurosurgery
79: 816.
Zamboni, G., C. A. Jones and J. Hughes (1983). Analytical Biochemistry 130: 83.
Zhang, G., M. Kai, M. Nakano and Y. Ohkura (1991). Chemical and Pharmaceutical
Bulletin 39: 126.
Zhuo, H., S. J. Fung, V. K. Reddy and C. D. Barnes (1992). Journal of Chemical
Neuroanatomy 5: 1.


CHAPTER 3
A HIGH PERFORMANCE LIQUID CHROMATOGRAPHY ASSAY FOR OPIOID
PEPTIDES USING ELECTROCHEMICAL DETECTION
Introduction
The objective of this study was the evaluation of a tyrosine-specific clean-up and
detection method for opioid peptides using leucine enkephalin (LE, Figure 3.1) as a model
peptide. The assay described here is based on the derivatization of LE by specific
enzymatic o-hydroxylation of the highly conserved tyrosine groups in the 1 position of
opioid peptides by mushroom tyrosinase. This derivatization results in the formation of a
catechol which is amenable to specific clean-up using boronate gels and is more easily
oxidizable than the parent peptide, thus facilitating electrochemical detection.
OH
CH
ch.
o
o
CH.
II
II
CH NH C
II
II
II
o
o
CH.
O
Tyr Gly Gly Phe Leu
Figure 3.1. Leucine enkephalin
66


132
Curve 3
Non-concentrated reagents
Matrix: buffer
Equation: y = 3.40* 1013x 1.3522, r2 = 0.9971
LE mol/inj
Peak height (mm)
0
0
2.81*10"13
9
3.65* 10*13
lost
5.62* 10'13
15
1.12* 10"12
37
2.23 1 O'12
75
Curve 4
Non-concentrated reagents
Matrix: CSF
Equation: y = 3.89* 10,2x 0.6900, r2 = 0.9975
LE mol/inj
Peak area
0
0
4.97*1013
0.6753
8.86* 1013
2.7875
2.12* 1 O'12
7.7407
3.96*1 O'12
14.0380
7.24*10'12
27.8690


21
biotin to fluorescence-labeled avidin produced an increase in fluorescence intensity. In
contrast to the ELISA described above, this approach relied on a lack of sandwich
formation between antibody, biotinylated LE derivative and fluorescence-labeled avidin, as
the success of this approach depended on the inability of antibody-bound biotinylated LE
to interact with fluorescence-labeled avidin and produce an increase in fluorescence
intensity. Therefore, here LE and biotinylated LE compete for antibody binding sites and
after equilibrium has been achieved, free biotinylated LE is detected through the increase
in fluorescence intensity produced on addition of fluorescence-labeled avidin.
Two high-performance liquid chromatography (HPLC) assays were also evaluated,
both making use of tyrosine-specific pre-column derivatization using the enzyme
tyrosinase. In the high performance liquid chromatography assay with electrochemical
detection (HPLC-ED), specific hydroxylation of the tyrosine group in the 1 position of
LE, which is highly conserved in all opioid peptides, presents two analytical advantages.
Firstly, the derivatization results in the formation of a catechol which is amenable to
specific clean-up using boronate gels, and secondly, this catechol is more easily oxidizable
than the parent peptide, thus facilitating electrochemical detection. In the high
performance liquid chromatography assay with fluorescence detection (HPLC-FL),
enzymatic derivatization by tyrosinase renders our peptide amenable to fluorogenic
derivatization using 1,2-diamino- 1,2-diphenylethane.
In addition, having established in the enzymatic derivatization for the HPLC assays
that products are formed from the reaction between LE and tyrosinase, the identity of
these products was determined by mass spectrometry. Furthermore, since tyrosinase and


97
Table 4.1 shows a summary of the results obtained using the various modifications
attempted in our HPLC-FL assay. Some of the results using the same method and the
same matrix could not be averaged since the output from the fluorescence detector was
monitored using different equipment (integrator or chart recorder).
Table 4.1. Summary of results for HPLC-FL assay. For raw data, see Appendix B.
Method
Matrix
Limit of
detection
N
slope
Non-concentrated
reagents
buffer
562 fmol/inj
1
0.9966
1.79*1012
area/mol
Non-concentrated
reagents
buffer
281 fmol/inj
2
0.9836-0.9971
1.14-3.40* 1013
mm/mol
Non-concentrated
reagents
CSF
500 fmol/inj
1
0.9975
3.89* 1012
area/mol
Non-concentrated
reagents
CSF
891 fmol/inj
1
0.9670
1.08* 1013
mm/mol
Concentrated
reagents
Buffer
500 fmol/inj
2
0.9926-0.9984
2.25-5.09* 1012
area/mol
Concentrated
reagents
CSF
500 fmol/inj
3
0.99170.0061
2.0711.26* 1012
area/mol
Concentrated
reagents, 5 min
tyrosinase
Buffer
277 fmol/inj
3
0.989510.0086
6.6913.80* 1012
mm/mol
Figures 4.5 and 4.6 show representative calibration curves obtained for LE in
buffer samples and spiked CSF, respectively, using the HPLC-FL method with non
concentrated reagents. Limits of detection for LE of 500 fmol/injection could be obtained


29
For the glutaraldehyde method (G. Adamus, personal communication), a 25 % v/v
stock solution of glutaraldehyde was freshly prepared on ice and diluted 65-fold in 0.1 M
sodium phosphate buffer pH 7. Equal amounts of LE and protein carrier (BSA or porcine
thyroglobulin) were weighed out and dissolved in 0.1 M sodium phosphate buffer pH 7 to
give a solution containing 1 mg/ml each of LE and protein carrier. One hundred and
twenty four microliters of the final dilution of glutaraldehyde were added for each milliliter
of protein-peptide solution and the reaction was allowed to proceed at room temperature
overnight with constant stirring. The conjugate was then dialyzed against deionized water
for 24 hours using pre-hydrated Spectrapor cellulose ester membranes with molecular
weight cut off of 15,000 (Spectrum Medical Industries Inc., Los Angeles, CA, USA).
After dialysis, aliquots of the conjugates were stored at -20C prior to lyophilization.
For the EDC method [Harlow and Lane 1988], a I mg/ml solution of LE was
prepared in water and EDC was weighed out and added to give a final concentration of 10
mg/ml. The pH of the reaction mixture was adjusted and maintained at pH 5 with 1 N
NaOH throughout the 5 minute incubation time at room temperature. An equal volume of
an 11 mg/ml solution of protein carrier (BSA or porcine thyroglobulin) was added and the
reaction was allowed to proceed at room temperature for 4 hours. The reaction was then
stopped by the addition of a sodium acetate solution (1 M, pH 4.2) to give a final
concentration of 100 mM. After an additional incubation of 1 hour at room temperature,
the conjugate was dialyzed against 0.1 M phosphate buffer pH 7 for 24 hours using pre-
hydrated SpectraPor cellulose ester membranes with molecular weight cut off of 15,000
(Spectrum Medical Industries Inc., Los Angeles, CA, USA). After dialysis, aliquots of the


APPENDIX B
DATA FOR HPLC-FL APPROACH
Curve 1
Non-concentrated reagents
Matrix: Buffer
Equation: y = 1.79*1012x + 0.4064, r2 = 0.9966
LE mol/inj
Peak area
0
0
5.62* 10'13
1.5338
1.12*10"12
2.7748
2.23* 10'12
4.0617
4.38*10'12
8.5962
8.76* 10'12
15.9292
Curve 2
Non-concentrated reagents
Matrix: buffer
Equation: y = 1.14*1013x + 2.2749, r2 = 0.9836
LE mol/inj
Peak height (mm)
0
0
8.91 10"13
13
1.80* 10"12
26
3.59* 1012
39
5.39* 10'12
70
7.18*1012
81
131


27
LE to interact with fluorescein-labeled avidin. Therefore, in this assay, biotinylated LE
and LE in the sample or standard compete for antibody binding sites and after equilibrium
has been achieved, fluorescence-labeled avidin, which acts as a detector molecule, is
added. Unbound biotinylated LE is determined by the increase of fluorescence intensity
due to the interaction of the unbound biotinylated LE and the fluorescence-labeled avidin
(Figure 2.2). In contrast to the ELISA described above, the success of this assay depends
on the synthesis of a biotinylated LE derivative which will not allow fluorescence-labeled
avidin to bind to antibody-bound biotinylated LE (i.e. lack of "sandwich" formation).
+
+
Increased fluorescence intensity
Antibody
Biotinylated analyte
Analyte
Fluorescence-labelled avidm
nr
Figure 2.2. Avidin-biotin-based homogeneous fluorescence immunoassay


75
comparison, calibration curves for LE in CSF were also constructed using HPLC-ED
without enzymatic derivatization or boronate clean-up.
Results
Enzymatic Derivatization
Preliminary experiments indicated that the concentration of mushroom tyrosinase
(135 units/ml) used for the enzymatic derivatization of LE allowed the efficient conversion
of LE to its hydroxylated derivative within 60 minutes (data not shown). A 50 mM
concentration of ascorbic acid was found to be adequate to prevent or reverse the
tyrosinase-induced formation of o-quinones in the enzymatic reaction mixture, thereby
inhibiting the subsequent polymerization of the reaction products. Figure 3.5A shows the
chromatograph of a control run where only LE and ascorbic acid are present at the
incubation concentrations. Here, LE is seen eluting after 9.5 minutes, distinct from
ascorbic acid eluting in the solvent front after 1.4 minutes. Two additional peaks at 5.7
minutes and 7.3 minutes are seen in the chromatograph of the incubation solution (Figure
3.5B). These two product peaks were identified by electrospray ionization mass
spectroscopy (see Chapter 5) as the di- and mono-hydroxylated derivatives of LE
([(HO)2- Tyr']-LE and [HO-Tyr']-LE, respectively). The relative intensity of the peaks
obtained from mass spectrometric analysis showed that [(HO)2-Tyr']-LE is a minor
product.


24
luminescent labels for use in immunoassay [Kricka 1993, Porstmann and Kiessig 1992,
Schulman et al. 1990],
Immunoassays can be divided into those methods which require separation of the
antibody-bound and free fractions prior to quantitation, and those in which the signal
being measured is a function of antibody binding and therefore do not require separation
of antibody-bound and free fractions prior to quantitation. These assays are referred to as
heterogeneous and homogeneous immunoassays, respectively. The development of
homogeneous immunoassays was prompted by the fact that the separation step in
heterogeneous immunoassays is labor intensive, complicates automation and introduces
inaccuracy and error into the assay as the equilibrium which exists between the bound and
free fraction in the sample is disturbed. Enzyme- and luminescence- labeling techniques are
used in the majority of the homogeneous immunoassays developed thus far [Coty et al.
1992, Garcia et al. 1993, Jenkins 1992]; however, other approaches involving the use of
liposomes [Bowden et al. 1986, Ho and Huang 1985, Umeda et al. 1986], bilayer
membranes [Ihara et al. 1988] and reversed micellar systems [Kabanov et al. 1989] have
also been investigated. Although homogeneous immunoassays are convenient and
relatively simple to perform, to date, they have suffered from lack of sensitivity compared
to heterogeneous immunoassays, due in particular to interference with endpoint
determination by components of biological samples [Jenkins 1992],
As mentioned in Chapter 1, one of the major disadvantages of immunoassays in
general is their relative lack of specificity due to cross-reactivity of the antibody in use to
structurally similar compounds. In order to circumvent this problem, analytical methods


128
Curve 5
Tyrosinase reaction, boronate clean up
Matrix: CSF
Equation: y = 21179x 16355, r2 = 0.9904
LE mol/inj
Peak area
0
21412
3.60* 10'13
7743f
7.22* 1013
29865
1.42*1012
42426
2.73*1012
65917
4.29* 10'12
111943
7.01 1 O12
176539
9.63* 10'12
212826
toullier, calculated concentration deviates from nominal concentration
by >20% (H. T. Karnes, personal communication)
Curve 6
Tyrosinase reaction, boronate clean up
Matrix: CSF
Equation: y = 23240x 10701, r2 = 0.9971
LE mol/inj
Peak area
0
10509
3.60*10'13
15452
7.22* 10'13
28020
1.42* 10'12
44913
2.73 1 O'12
73904
4.29*1012
116036
7.01 10"12
170143


95
Figure 4.3 A shows the chromatograph of a control run where only LE and
ascorbic acid are present at the incubation concentrations. Here, LE is seen eluting after
9.5 minutes, distinct from ascorbic acid eluting in the solvent front after 1.4 minutes. Two
additional peaks at 5.7 minutes and 7.3 minutes are seen in the chromatograph of the
incubation solution (Figure 4.3B). These two product peaks were identified by
electrospray ionization mass spectroscopy (see Chapter 5) as the di- and mono-
hydroxylated derivatives of LE ([(HO)2- Tyr']-LE and [HO-Tyr']-LE, respectively).
Figure 4.4. Time course for the enzymatic derivatization of LE (1 mM) by mushroom
tyrosinase (135 units/ml) in the presence of ascorbic acid (50 mM) showing
the disappearance of LE () as [HO-Tyr']-LE (A) and (HO)2-Tyr1 (0) are
produced.
A time course of the enzymatic derivatization of LE by mushroom tyrosinase in the
presence of ascorbic acid is shown in Figure 4.4. This time course shows the
disappearance of the peak corresponding to LE as [HO-Tyr^-LE and [(HO)2- Tyr]-LE


107
i d. deactivated fused silica capillary (Scientific Glass Engineering, Victoria, Australia) at 2
to 5 pl/min flow rate by a medical infusion pump (SAGE Instruments, Model 34IB,
Boston, MA, USA). A 0.005 in i d. x 0.010 in o.d. flat tipped hypodermic needle held at
2.4 kV potential produced spray current in the range of 130 to 180 pA, when the tip of
the needle to nozzle orifice distance was about 10 mm. The source block was heated to
250 C, and the spray chamber temperature was around 55 to 60 C. A Vector/One data
system (Teknivent, St. Louis, MO, USA) was used to control the quadrupole analyzer
(2,000 Da mass range), and to collect mass spectra in the 100 to 1,000 Da mass range at 3
ms/Da scan rate. For molecular weight determination, the repeller to collimator voltage
was held at 18 V, and collision-induced dissociation in the skimmer to collimator region
was obtained at 50 V. At least ten spectra were averaged for each experiment.
Radioreceptor Assay
The enzymatic reaction and radioreceptor assay were carried out on the same day.
Two hundred microliters of the incubation solution from the enzymatic reaction were
injected into the HPLC system and the peak corresponding to [HO-Tyr'J-LE was
collected in 1.9 ml of mobile phase. A 90% conversion from LE to [HO-Tyr'j-LE was
assumed based on the disappearance of the peak corresponding to LE after a 1 hour
incubation with mushroom tyrosinase (Figure 5.1) and the relative intensities of the
product peaks obtained from mass spectrometric analysis. Therefore, this solution
contained 180 nmoles of [HO-Tyr^-LE and was used directly in the radioreceptor assays.
Two hundred microliters of a 1 mM solution of LE were also injected into the HPLC


78
Boronate Clean-up
By applying 35 pmoles to the boronate gel column, an average recovery of analyte
of 68.67 % was achieved using the complete boronate clean-up method (SD = 6.1, n = 3).
Time Course of Enzymatic Derivatization
A representative time course of the enzymatic derivatization at analytical
concentrations (17 pmol/inj) showing the peak area of [HO-Tyr]-LE plotted against time
is shown in Figure 3.7. This experiment was carried out twice, showing the same trend
each time. The plot indicates that the highest level of [HO-Tyr']-LE is seen after a 5
minute incubation and therefore, this incubation time was selected for future use in this
study.
Figure 3.7. Time course of enzymatic derivatization showing peak height of [HO-Tyr1]-
LE over incubation time.


109
The radioreceptor assays were carried out in duplicate in microcentrifuge tubes.
Each tube contained 30 pM bestatin, 0.6 pM thiorphan and 10 pM captopril as a
peptidase inhibitor cocktail, 6.25 mg or 12.5 mg of rat brain membranes, various
concentrations of LE or [HO-Tyr'j-LE as competitor, and either 0.2 nM of [3H]-
diprenorphine, 1.65 nM of [3H]-DAGO or 1 nM [3H]-DPDPE as tracer, to assay for total
opioid receptor, p or 5 sites, respectively, in 1 ml of 50 mM Tris-HCl buffer (pH 7.4). The
tubes were incubated at room temperature with constant shaking for 1 hour. When using
[3H]-diprenorphine as tracer, the tubes were then centrifuged at 12,000g for 10 minutes to
bring down the pellet. The pellets were washed three times with ice-cold 50 mM Tris-HCl
buffer (without resuspending) and then were dissolved in 1 ml of Scintiverse II scintillation
cocktail. When using [3H]-DAGO or [3H]-DPDPE as tracer, the rat brain membranes
were separated from the supernatant and washed with ice-cold 50 mM Tris-HCl buffer by
means of a rapid filtration technique. The filters retaining the rat brain membranes and
bound tracer were placed in 4 ml of Scintiverse II scintillation cocktail and after being
allowed to stand overnight, the radioactivity (CPM) in either the pellets or on the filters
was determined using a Beckmann LS 5,000 TD scintillation counter (Fullerton, CA,
USA). In a control experiment, HPLC analysis of the supernatant showed that both LE
and [HO-Tyr^-LE were stable under radioreceptor assay conditions. Non-specific binding
was determined in the presence of high concentrations of competitor (1 x 10'5 M). The
IC5o values of LE and [HO-Tyr']-LE (concentration of competitor displacing 50% of
bound tracer) were determined using the MINSQ non-linear curve-fitting program


CHAPTER 4
A HIGH PERFORMANCE LIQUID CHROMATOGRAPHY ASSAY FOR OPIOID
PEPTIDES USING FLUORESCENCE DETECTION
Introduction
The objective of this study was the evaluation of a tyrosine-specific analytical
approach for opioid peptides using high performance liquid chromatography with
fluorescence detection and leucine enkephalin (LE, Figure 4.1) as a model peptide.
OH
Tyr Gly Gly Phe Leu
Figure 4.1. Leucine enkephalin
As in Chapter 3, this analytical approach exploits the derivatization of LE by
specific enzymatic o-hydroxylation of the highly conserved tyrosine groups in the 1
87


86
sample (>1 ml), it may be adequate for the analysis of the elevated physiological
concentrations of opioid peptides to be expected in clinical studies.


8
upregulated cAMP system, counterbalanced by constitutively active mu receptors. In this
model, opiate tolerance results from fewer mu receptors remaining activatable by agonists
and the enhanced activity of the cAMP system. In other words, in this scenario
dependence occurs because an upregulated cAMP system is established which needs to be
counterbalanced by opiate agonist activity and tolerance occurs because fewer mu
receptors can now be activated by opiate agonists.
A lower degree of dependence is seen with agonists at delta receptors than with
agonists at mu receptors. As mentioned above, the action of mu receptor agonists can be
modulated by the co-administration of delta receptor agonists so that the potency and
efficacy of analgesia is increased without a corresponding increase in side effects such as
physical dependency, respiratory depression and gastrointestinal effects. This effect could
be exploited to allow for the use of mu agonists of lower efficacy and increased safety
while still providing adequate pain relief without the risk of side effects. Eventually, a delta
agonist may be developed which is able to provide effective pain relief without side effects
[Rapaka and Porreca 1991],
Opioid peptides play a part in the regulation of the immune system, particularly
during periods of stress, as they modulate the functions of a number of cell types involved
in the immune response [Murgo et al. 1986], Generally, endogenous opioid peptides are
immunostimulant as they enhance T-cell function and stimulate phagocyte function, thus
increasing resistance to infection. There is also evidence that suggests a role for
endogenous opioids in the growth and development of lymphoid tissue [Plotnikoff et al.
1985],


108
system and the peak corresponding to LE was collected and used in the radioreceptor
assays. Mobile phase components were found not to interfere with the radioreceptor
assay.
Figure 5.1. Time course for the enzymatic derivatization of LE (1 mM) by mushroom
tyrosinase (135 units/ml) in the presence of ascorbic acid (50 mM) showing
the disappearance of LE () as [HO-Tyr]-LE (A) and (HO)2-Tyr (0) are
produced.
Rat brain membranes were prepared as described by Hochhaus et al [Hochhaus et
al. 1988], Briefly, the whole brain, without cerebellum, of male Sprague-Dawley rats
(120-140 g) was homogenized in 60 volumes of ice cold 50 mM Tris-HCl buffer (pH 7.4)
containing 100 mM NaCl. The homogenate was incubated for 1 hour at 20C and
centrifuged for 20 minutes at 4C. The pellet was then resuspended, washed twice with 50
mM Tris-HCl and diluted in 50 mM Tris-HCl to give 400 mg of rat brain membranes
per ml. Aliquots of this suspension were stored at -80C and used within one week.


I certify that I have read this study and that in my opinion it conforms to
acceptable standards of scholarly presentation and is fully adequate, in scope and quality,
as a dissertation for the degree of Doctor of Philosophy.
c.
Giinther Hochhaus, Chair
Associate Professor of Pharmaceutics
I certify that I have read this study and that in my opinion it conforms to
acceptable standards of scholarly presentation and is fully adequate, in scope and quality,
as a dissertation for the degree of Doctor of Philosophy.
Hartmut Derendorf
Professor of Pharmaceutics
I certify that I have read this study and that in my opinion it conforms to
acceptable standards of scholarly presentation and is fully adequate, in scope and quality,
as a dissertation for the degree of Doctor of Philosophy.
Paul Klein
Professor of Pathology and
Laboratory Medicine
I certify that I have read this study and that in my opinion it conforms to
acceptable standards of scholarly presentation and is fully adequate, in scope and quality,
as a dissertation for the degree of Doctor of Philosophy.
Laszlo Prokai
Assistant Professor of Pharmaceutics
I certify that I have read this study and that in my opinion it conforms to
acceptable standards of scholarly presentation and is fully adequate, in scope and quality,
as a dissertation for the degree of Doctor of Philosophy.
3 /frUT-
Ian Tebbett
Associate Professor of Pharmaceutics


120
acts as a driving force for the electrophilic substitution. Therefore, in conjunction with the
observations of Judd et al. [Judd et al. 1985], our findings indicate that the presence of a
hydroxy group in the meta position rather than the absence of one in the para position
gives rise to a decrease in opioid receptor binding in enkephalins.
Tyrosinase and enkephalin immunoreactivities have been identified in isolated cells
[Kimura et al. 1992] and the spinal and brain regions in some species, namely the locus
coeruleus complex in the cat [Zhuo et al. 1992] and the ventral tegmental area in the rat
[Sesack and Pickel 1992], Although aminopeptidases, carboxypeptidases and
enkephalinases are thought to be mainly responsible for the metabolic deactivation of LE
[Venturelli et al. 1985], given that tyrosinase and LE have been found to co-exist in vivo,
and that the product of the reaction between these two entities shows decreased affinity to
both p and 8 opioid receptors compared to the parent enkephalin, we speculate that
tyrosinase may also contribute to the metabolic fate of LE in vivo.


105
Materials
The following materials were procured from the sources indicated: Leucine
enkephalin, mushroom tyrosinase (3870 units/mg, E.C. 1.14.18.1), bestatin, thiorphan and
captopril from Sigma, St. Louis, MO, USA, ascorbic acid from Mallinckrodt, Paris, KY,
USA, [3H]-diprenorphine from Amersham International, Arlington Heights, IL, USA, and
Scintiverse II scintillation cocktail from Fisher Scientific, Pittsburgh, PA, USA. All other
chemicals were of reagent grade. Double distilled water was used throughout.
The chromatographic system used consisted of an LDC/Milton Roy miniMetric II
metering pump (Riviera Beach, FL, USA), a Negretti and Zamba injector (Southampton,
UK) fitted with a 500 pi loop, a Perkin Elmer LC-75 spectrophotometric detector
(Norwalk, CT, USA) and a Hewlett Packard HP 3394A integrator (Avondale, PA, USA).
The column was a Partisil 5 ODS-3 125 x 4.6 mm (Whatman Labsales, Hillsboro, OR,
USA). The detection wavelength was 254 nm and the mobile phase consisted of 12.5%
acetonitrile (v/v) in citrate/dipotassium phosphate buffer (pH 5) at a flow rate of
1 ml/minute.
Methods
Enzymatic Reaction
The reaction between LE and mushroom tyrosinase was carried out in
microcentrifuge tubes at the following concentrations: 1 mM LE and 50 mM ascorbic acid
in 0.1 M potassium phosphate buffer (pH 7). The reaction was started by adding


TABLE OF CONTENTS
page
ACKNOWLEDGMENTS iii
ABSTRACT vi
CHAPTERS
1 INTRODUCTION 1
Endogenous Opioid Peptides 1
Opioid Receptors 3
Physiology and Pharmacology 5
Rationale 12
Objectives 19
2 APPROACHES TO THE DEVELOPMENT OF AN IMMUNOASSAY
FOR LEUCINE ENKEPHALIN 23
Introduction 23
Materials 28
Methods 28
Results and Discussion. 41
3 A HIGH PERFORMANCE LIQUID CHROMATOGRAPHY ASSAY FOR
OPIOID PEPTIDES USING ELECTROCHEMICAL DETECTION 66
Introduction 66
Materials 69
Methods 70
Results 75
Discussion 82
4 A HIGH PERFORMANCE LIQUID CHROMATOGRAPHY ASSAY
FOR OPIOID PEPTIDES USING FLUORESCENCE DETECTION 87
Introduction 87
Materials 89
Methods 90
iv


xml version 1.0 encoding UTF-8
REPORT xmlns http:www.fcla.edudlsmddaitss xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.fcla.edudlsmddaitssdaitssReport.xsd
INGEST IEID E2ZTQ5YY0_78R2UQ INGEST_TIME 2014-10-07T21:10:22Z PACKAGE AA00025794_00001
AGREEMENT_INFO ACCOUNT UF PROJECT UFDC
FILES


7
antagonist naloxone. These effects are assumed to be the result of opioid receptor
blockade and are widely accepted as indirect evidence for endogenous opioid involvement
in the physiological function under observation.
One of the most important functions of opioid peptides is in pain modulation. The
mu receptor is the opiate receptor most associated with pain relief, although delta and
kappa receptor agonists also have analgesic properties. It is thought that under some
conditions, mu and delta receptors are functionally coupled as delta agonists given in sub
analgesic doses have been found to either potentiate or inhibit the analgesic effects of
morphine (a mu agonist) at different sites. It has therefore been postulated that mu and
delta receptors may exist either separately or in a complexed form [Rapaka and Porreca
1991], Acupuncture analgesia is also thought to be mediated by endogenous opioid
peptides [Clement-Jones and Rees 1982],
Animal studies have indicated that endogenous opioid peptides play a role in the
development of opiate dependence, a side effect common to opiate analgesics. This is a
syndrome whereby distress is caused upon withdrawal of an opiate following chronic
administration. It has been hypothesized that chronic narcotic abuse leads to the
suppression of endogenous opioid production through a negative feedback mechanism so
that sudden withdrawal of the narcotic leads to a deficiency in endogenous opioids which
causes the classical physical withdrawal symptoms [Clement-Jones and Besser 1983],
Recently, Wang et al. [Wang et al. 1994] have proposed that stimulation by an opiate
agonist causes gradual constitutive mu receptor activation so that an agonist is no longer
required for signal transduction, and a dependent state is established, consisting of an


CHAPTER 2
APPROACHES TO THE DEVELOPMENT OF AN IMMUNOASSAY FOR
LEUCINE ENKEPHALIN
Introduction
Immunoassay is an analytical method which exploits the binding of a ligand
(antigen) to specific sites on an antibody. In most cases (e g./ for radioimmunoassays),
labeled ligand competes with unlabeled ligand (analyte) for antibody sites and the extent of
binding of the labeled ligand is determined through the measurement of some physical or
chemical property associated with the label. A standard calibration curve of the signal
produced by the label with respect to the concentration of analyte present can then be
constructed, thus allowing the estimation of unknown ligand concentrations from this
curve.
Traditionally, immunoassays have involved the use of radioisotopes as labels.
Although radioimmunoassays (RIA) have the advantages of being highly sensitive and
invulnerable to environmental interferences (e g./ by components of the assay), their use is
accompanied by several disadvantages. These disadvantages include the emotive bias
against the use of radioisotopes, the costly disposal of waste, special requirements for
assay handling and training of staff, the lack of stability of radioisotopes and the
consequent limited shelf-life of reagents [Gosling 1990], Therefore, there has been
considerable interest in developing non-isotopic labels, such as enzyme labels and
23


52
The C-terminal biotinylated LE derivatives LE-Lys6-B and LE-Lys6-BX showed
no binding to the commercial antiserum in the concentrations used (data not shown) and
therefore, using this preparation, they were unsuitable for the development of either of the
immunoassays proposed in this study. It is possible that the inclusion of a longer spacer
arm in the C-terminal biotinylated derivatives would have allowed binding of the
commercial antiserum if the epitope to be recognized by the antibody remained intact on
biotinylation. However, this avenue was not explored.
Figure 2.12. Representatives binding curves for LE (), BLE (), BXLE (), BLE-
avidin () and BXLE-avidin (0) using 1/3,000 dilution of commercial
antiserum.
In these experiments, BXLE showed a ten-fold higher affinity to the commercial
antiserum than LE (Table 2.5). It was postulated that since LE needs to be conjugated to a


63
the 1 position of LE and the fluorescein groups of the detector molecule FITC-avidin. In
order to test this hypothesis, assays were carried out in the absence of antibody, in the
same way as described above. However, here des-Tyr1 LE or pentaglycine were used as
analytes in concentrations similar to those of LE which had been used previously.
Figure 2.21. Calibration curve for pentaglycine using the homogeneous fluorescence
immunoassay set up, omitting antibody but with BLE () and without
BLE(D).
Figures 2.20 and 2.21 show that des-Tyr1 LE and pentaglycine do not produce the
same interaction with FITC-avidin as LE when used in the same concentrations in the
homogeneous immunoassay set up. When BLE is included in the assay, the readings
approach those obtained in the total tracer samples and when no BLE is present in the
assay, the readings approach those in the fluorescence blank samples. It was therefore
concluded that the effect seen at higher concentrations of LE in the homogenous


22
enkephalins have been found to co-exist in vivo [Merchenthaler 1993, Sesack and Pickel
1992, Zhuo et al. 1992], the biological activity in rat brain homogenate of the major
product of the reaction between these two entities was investigated.


77
method (i.e. including enzymatic derivatization and boronate clean-up) and HPLC-ED
without enzymatic derivatization or boronate clean-up:
Guard cell
Analytical cell Det 1
Analytical cell Det 2
HPLC-ED with deriva
tization and cleanup
+0.4 V
-0.1 V
+0.3 V
HPLC-ED, no deriva
tization, no cleanup
+0.75 V
+0.4 V
+0.7 V
Figure 3.6. Current-voltage curves for (HO-Tyr^-LE and LE
Using these potentials, impurities in the mobile phase are pre-oxidized by the guard
cell thus reducing background current and baseline noise. Impurities in the sample with
relatively low oxidation potentials which might co-elute with the analyte and interfere with
the signal produced by the analyte are preoxidized at Det 1. The Det 1 potential is set
sufficiently low so that the analyte will not be pre-oxidized at this electrode, thus ensuring
the production of a maximum signal at the analytical potential at Det 2.


12
corresponding D-amino acid for the naturally occurring L-amino acid in the peptide
molecule. As peptides are conformationally labile, the selectivity of opioid peptides has
been increased by stabilizing the peptide molecule in a conformation which prefers the
desired receptor. This can be achieved through the incorporation of conformational
restrictions such as the introduction of unnatural bulky synthetic amino acids (e g./
penicillamine residues) or the cyclization of the peptide chain. Highly selective opioid
peptide analogs such as [D-Pen2, D-Pen5]-enkephalin which is selective for the delta
receptor [Mosberg et al. 1983] and [Tyr-D-Ala-Gly-MePhe-NH(CH2)20H] which is
selective for the mu receptor [Handa et al. 1981] have been developed using these
techniques.
Further research involving opioid peptides and their receptors is of importance in
elucidating the precise physiological roles of these entities, particularly in the areas of pain
and immunomodulatory pathways.
Rationale
As discussed above, a considerable body research has been focused on the
development of opioid peptides as therapeutic agents. The low physiological
concentrations of endogenous opioid peptides and those to be expected for therapeutically
administered derivatives necessitate the development of specific and ultra-sensitive
analytical methods, in the fmol per ml range, for these entities in biological fluids to
support clinical studies. The need for new analytical approaches for the measurement of
opioid peptides has been stressed by the National Institute on Drug Abuse [Rapaka 1986],


53
protein carrier to render it immunogenic for antibody production, this biotinylated LE
derivative including a spacer arm may resemble the epitope presented to the immune
system more closely than LE itself and therefore, BXLE shows a higher affinity to the
commercial antiserum than LE itself.
Although the N-terminal biotinylated derivatives were seen to retain affinity for the
commercial antiserum, a shift in affinity by 2 or 3 orders of magnitude was seen when
complexes were formed between BLE or BXLE and avidin (Table 2.5), indicating that a
sandwich was not formed between the commercial antiserum, BLE or BXLE and avidin.
Therefore, this commercial antiserum was deemed to be unsuitable for the development of
the proposed ELISA for which sandwich formation is a requirement. Although this lack of
sandwich formation indicates that this preparation is suitable for the development of the
proposed homogeneous fluorescence immunoassay, as it was anticipated that large
quantities of antibody would be required for the development of this assay, this avenue
was unfeasible. As a result, efforts involving the development of either an ELISA or an
homogeneous fluorescence immunoassay using the commercial antiserum preparation
were abandoned.
Using a 1/1,000 dilution of the antibody produced in this laboratory, binding
curves were also constructed for LE, the N-terminal biotinylated LE derivatives, the C-
terminal biotinylated LE derivatives and the corresponding pre-formed complexes with
avidin. The IC5o values obtained for LE, BLE, BXLE and BXLE-avidin are shown in
Table 2.6. Representative binding curves for LE, BLE, BXLE and BXLE-avidin are
shown in Figure 2.13.


44
Figure 2.5. Bar graph showing total counts, total binding and non-specific binding of
3FLLE using 1/1,000 dilution of antibody produced in this laboratory.
Characterization of Antibody
The Scatchard plot of bound/free 3H-LE versus bound 3H-LE is shown in
Figure 2.6. The Ka value determined from the negative slope of the line drawn between the
points was 3.87* 1091/mol which converts to a Ka value of 2.58* 1010 mol/1. The number
of specific antibody sites present in the final incubation for this experiment was determined
from the x-axis intercept to be 3.62*10'10 mol/1. Since 3.18*1 O'8 mol/1 of IgG was used in
this preparation, the purified antibody contains only 1.14% specific antibody sites. By
extrapolation, the stock solution of purified antibody was determined to contain 4.35* 107
mol/1 of specific antibody sites. The affinity of the antibody made in this laboratory to LE
compared favorably to the commercial antiserum obtained from Peninsula under the same
assay conditions since 1C50 values for LE using a 1/1,000 dilution of the antibody


ACKNOWLEDGMENTS
My thanks go to my advisor, Dr. Giinther Hochhaus and the members of my
supervisory committee, Dr. Hartmut Derendorf, Dr. Paul Klein, Dr. Laszlo Prokai and Dr.
Ian Tebbett for their guidance and support during the course of my doctoral research.
Special thanks go to Dr. Prokai for the mass spectrometry analysis of hydroxylated leucine
enkephalin derivatives. I am also grateful to Dr. Richard Prankerd for his help in the early
stages of this work.
I acknowledge the PDA. Foundation for Pharmaceutical Sciences, Inc. and
Schering-Plough Corporation for partial funding of the research presented in this
dissertation.
I would also like to recognize the services provided by the Hybridoma Core and
the Protein Chemistry Core of the Interdisciplinary Center for Biotechnology Research at
the University of Florida.
There are many others who are too numerous to mention, who have been
instrumental in enabling me to complete this work. I hope to be able to thank each of them
personally.
in


46
synthesized and tested. Some of the biotinylated derivatives included a spacer arm
between the biotin group and the peptide (Figure 2.8) as it was felt that this would
increase the likelihood of the formation of a sandwich between the anti-LE antibody, the
biotinylated LE derivative and enzyme-labeled avidin by reducing steric hindrance.
Biotinylation took place at the N-terminal end of the peptide when LE was used as a
starting material as the only free amino group available for the reaction is the N-terminal
amino group of the peptide. When LE-Lys6 was used as a starting material, biotinylation
took place primarily at the C-terminal end of the peptide as here, the amino group of the
lysine moiety is more reactive than the N-terminal amino group under the reaction
conditions used due to its greater basicity.
PeptideNH2 +
H O
I II
PeptideN C(CH,)4
+
Figure 2.7. Biotinylation of peptide using N-hydroxysuccinimidobiotin.


18
A derivatization reaction which is specific for tyrosine groups such as the one
described in this dissertation could be expected to increase selectivity in the determination
of opioid peptides as tyrosine is highly conserved in the 1 position of these compounds but
is often missing in unrelated peptides. Tyrosine specific HPLC methods with fluorescence
detection such as those using l,2-diamino-4,5-dimethoxybenzene [Ishida et al. 1986, Kai
et al. 1988] and hydroxylamine, cobalt (II) ion and borate [Nakano et al. 1987, Zhang et
al. 1991] do exist for peptide analysis. Although these methods have allowed the
determination of opioid peptides with sensitivities of 100-500 femtomole per injection, the
harsh reaction conditions of these derivatizations are expected to lead to diminished
recoveries and reduced assay reproducibility of the fragile peptide analytes.
Mass spectrometry
HPLC with mass spectrometry offers the highest level of molecular specificity
compared to other analytical methods and is the only method with which unambiguous
confirmation of the structure of the target peptide can be achieved. However, MS is costly
and requires specialized instrumentation and therefore, it is an impractical method for the
average analytical laboratory. The analysis of peptides by mass spectrometry has been
facilitated in recent years by the advent of several soft ionization and sample
introduction techniques such as fast atom bombardment, matrix-assisted laser desorption,
electrospray and ionspray mass spectrometry (see Arnott et al. 1993, Biemann 1992 and
Carr 1990 for reviews) which allow the production of intact molecular ions from these
fragile species. Another advantage of these techniques is that they are often compatible
with on-line microbore HPLC. A microbore HPLC-ionspray MS method has allowed the


50
Figure 2.11. Displacement of HABA from avidin by biotin and biotinylated LE derivatives.
on conjugation to LE, biotin retains the ability to bind to avidin. The curvilinear nature of
the displacement curves for BLE, BXLE and LE-Lys-BX together with the observation
that more of these biotinylated derivatives compared to biotin is required to displace the
same amount of HABA from avidin indicates that these derivatives have a slightly lower
affinity for avidin than biotin itself. However, here, LE-Lys6-B is seems to have a slightly
higher affinity to avidin than biotin. This effect can probably be attributed to experimental
error as the concentrations of the biotinylated derivatives were calculated from estimated
conversion rates based on the reduction of the peak heights of the starting materials in the
biotinylation reaction. Previous investigations on the binding of biotinylated peptides or
other macromolecules to avidin have shown that a spacer arm may be required to retain
full affinity to avidin [Finn et al. 1984, Green et al. 1971], however this observation was
not reflected in the results seen here as LE biotinylated with and without the spacer arm
both showed similar affinity to avidin and in fact, LE-Lys6 biotinylated without the spacer


71
integrator (Avondale, PA, USA). The column was a Partisil 5 ODS-3 125 x 4.6 mm
(Whatman Labsales, Hillsboro, OR, USA). The detection wavelength was 254 nm and the
mobile phase consisted of 12.5% acetonitrile (v/v) in citrate/dipotassium phosphate buffer
(pH 5) at a flow rate of 1 ml/minute.
The product peaks were collected from the HPLC eluent, organic solvent was
removed by evaporating under a stream of nitrogen and the collected peaks were
concentrated using a preconditioned Sep-Pak Ct8 preparative column (Waters Associates,
Milford, MA, USA). The identity of the products was determined by electrospray
ionization mass spectrometry (see Chapter 5).
Electrochemical Detection
Electrochemical detection was effected using a Model 5100A Coulochem multi
electrode electrochemical detector (ESA Inc., Bedford, MA, USA) fitted with a Model
5020 guard cell and a Model 5011 analytical cell. The complete HPLC-ED system was
configured as shown in Figure 3.4.
In this system, the guard cell acts to pre-oxidize electroactive impurities in the
mobile phase, thus reducing background current. The analytical cell consists of two
working electrodes in series. The first (Det 1) acts to further reduce background current
in the injected sample and to pre-oxidize co-eluting interferences that oxidize at potentials
lower than the analyte of interest. The second working electrode (Det 2) is set to quantify
the analyte. The solvent delivery system was an LDC/Milton Roy constaMetric III
metering pump (Riviera Beach, FL, USA), and the injector was a Rheodyne Model 7125


80
Figure 3.8. Representative calibration curve for LE in buffer using complete
HPLC-ED method.
Figure 3.9. Representative calibration curve for LE in CSF using complete
HPLC-ED method.


28
Materials
Leucine enkephalin, leucine enkephalin-Lys6, Freunds complete adjuvant,
Freunds incomplete adjuvant, bovine serum albumin, porcine thyroglobulin, glutar-
aldehyde, l-ethyl-3-(3-dimethylaminopropyl)carbodiimide, sodium acetate, activated
charcoal, dextran (average molecular weight 70,800), avidin, N-hydroxysuccinimidobiotin,
biotinamidocaproate N-hydroxysuccinimide ester, triethanolamine, 2-4-hydroxyazo-
benzene-benzoic acid, di-methylsulfoxide and fluorescein isothiocyanate avidin were
obtained from Sigma Chemical Company, St. Louis, MO, USA. Methanol, acetonitrile,
trifluoroacetic acid, sodium dihydrogen phosphate, disodium hydrogen phosphate, sodium
chloride, potassium dihydrogen phosphate, dipotassium hydrogen phosphate and
potassium chloride were purchased from Fisher Scientific, Pittsburgh, PA, USA. [tyrosyl-
3,5-?H(N)]-leucine enkephalin was procured from NEN Research Products, Dupont
Company, Wilmington, DE, USA. Cytoscint scintillation cocktail was obtained from ICN
Biomedicals Inc., Irvine, CA, USA and anti-LE antiserum was purchased from Peninsula
Laboratories Inc., Belmont, CA, USA.
Methods
Antibody Production
Leucine enkephalin (LE) was conjugated to either porcine thyroglobulin or bovine
serum albumin (BSA) by reaction with either glutaraldehyde or 1-ethyl-3-(3-dimethyl-
aminopropyl)carbodiimide (EDC).


134
Curve 7
Concentrated reagents
Matrix: Buffer
Equation: y = 5.09*1012x 0.8542, r2 = 0.9984
LE mol/inj
Peak area
0
0
5.01*10'13
1.1262
l*10-i2
4.1923
2*1 O'12
8.9492
3*10-12
14.35
4.67* 10'12
28.442f
6.68* 1012
33.319
¡outlier, calculated concentration deviates from nominal concentration
by >20% (H. T. Karnes, personal communication)
Curve 8
Concentrated reagents
Matrix: CSF
Equation: y = 9.91 10ux + 0.5776, r2 = 0.9925
LE mol/inj
Peak area
0
0.79768
5*1 O'13
1.0174
1*10*2
1.3522
2*1012
lost
3*1 O'12
5.2685
4.68* 1012
3.0910f
foullier


139
Dave, K. J., J. F. Stobaugh, T. M. Rossi and C. M. Riley (1992). Journal of
Pharmaceutical and Biomedical Analysis 10: 965.
de Ceballos, M, M. D. Taylor and P. Jenner (1991). Neuropeptides 20: 201.
de Montigny, P., C. M. Riley, L. A. Sternson and J. F. Stobaugh (1990). Journal of
Pharmaceutical and Biomedical Analysis 8: 419.
Defrutos, M. and F. E. Regnier (1993). Analytical Chemistry 65: A17.
Eisenach, J. C., C. 2. Dobson, C. E. Inturrisi, D. D. Hood and P. B. Agner (1990). Pain
43: 149.
Engel, W. D. and P. Khanna (1992). Journal of Immunological Methods 150: 99.
Eriksson, B. and M. Wikstrom (1992). Journal of Chromatography 593: 185.
Finn, F. M., G. Titus and K. Hofmann (1984). Biochemistry 23: 2554.
Fleming, L. H. and N. J. Reynolds (1988). Journal of Chromatography 431: 65.
Garcia, S. F., A. Navas and J. Lovillo (1993). Analytical Biochemistry 214: 359.
Gemant, A. (1974). Molecular Biology Reports 1: 423.
Gosling, J. P. (1990). Clinical Chemistry 36: 1408.
Green, N. M. (1970). Methods inEnzymology 18A: 418.
Green, N. M., L. Konieczny, E. J. Toms and R. C. Valentine (1971). Proceedings of the
National Academy of Sciences 74: 2697.
Grossman, A. and L. H. Rees (1983). British Medical Bulletin 39: 83.
Handa, B. K., A. C. Lane, J. A. H. Lord, B. A. Morgan, M. J. Ranee and C. F. C. Smith
(1981). European Journal of Pharmacology 70: 531.
Hansen, D. W., R. H. Mazur and M. Clare (1985). The Synthesis and SAR of Orally
Active Enkephalin Analogs with Modified N-terminal Tyrosine Residues, in Peptides:
Structure and Function (pg. 491) D. M. Deber, V. J. Hruby, & K. D. Kopple (Eds.).
Rockford, IL: Pierce Chemical Company.
Hansen, P. E. and B. A. Morgan (1984). Structure-Activity Relationships in Enkephalin
Peptides, in Opioid Peptides: Biology, Chemistry and Genetics (vol. 6, pg. 269) S.
Udenfriend, & J. Meienhofer (Eds.). Orlando: Academic Press Inc.


42
the N-terminal end (Figure 2.3). Therefore, on inoculation with such a conjugate, one
would expect the antibody that is produced to be directed against the C-terminal end of
the peptide.
c=o + ch3ch2n=c=n
I
OH
H+
(CH2)3-N-CH3
ch3
T*
c=o
/
O H+
I I
ch3ch2-nh-c=n(CH2)3n -ch3
ch3
+ h2n-r2
T-
c=o
I
NH
I
r2
H+
I
+ ch3ch2nh-c-nh(CH2)3n-ch3
0 ch3
Figure 2.4. Coupling of peptide to protein carrier using EDC. Ri=peptide or protein
carrier, R2=peptide or protein carrier.
In contrast, EDC allows the conjugation of the peptide to the protein carrier via
either the N-terminal or the C-terminal end as carbodiimides attack carboxylic groups to
change them into reactive sites for free amino groups. Therefore, the carboxylic groups on


6
Adenylate cyclase is an enzyme that synthesizes cyclic AMP from ATP so that it
can then go on to act as a "second messenger" in a number of biochemical systems. G-
proteins cause adenylate cyclase to convert from the active form of the enzyme which is
coupled to GTP to the inactive form which is coupled to GDP and vice versa. Therefore, a
compound that stimulates adenylate cyclase production would do so through the
interaction of its receptor with a G-protein that converts adenylate cyclase to the active
GTP-coupled form, whereas inhibition of adenylate cyclase, as by opioids for example,
would be mediated through a G-protein that favors the formation of inactive GDP-coupled
adenylate cyclase (Figure 1.4).
GDP GTP
hydrolysis
Figure 1.4. Scheme for receptor mediated stimulation and inhibition of adenylate cyclase.
Rs receptors which produce stimulation of adenylate cyclase on binding, Ri -
receptors which produce inhibition of adenylate cyclase on binding, AC -
adenylate cyclase [Simon 1984],
Opioid peptides are distributed widely throughout the central and peripheral
nervous system, suggesting that these compounds play a part in a variety of physiological
functions [Olson et al. 1991], The physiological roles of endogenous opioid peptides have
been attributed largely on the basis of effects seen on administration of the opioid


41
(FITC-avidin) was allowed to proceed for 20 minutes at room temperature in the dark,
after which fluorescence readings were taken at 482 nm and 517 nm using a
Perkin Elmer LS-3B fluorescence spectrophotometer (Norwalk, CT, USA).
Results and Discussion
Antibody Production
It was anticipated that the development of the homogenous fluorescence
immunoassay in particular would require large quantities of antibody. Therefore, it was
decided to produce a polyclonal antibody to leucine enkephalin in this laboratory.
H H
RNH2 + 0=C(CH2)3CO + h2nR2
- 2H20
T
H H
R,N=(CH2>3=NR2
Figure 2.3. Coupling of peptide to protein carrier using glutaraldehyde. Ri= peptide or
protein carrier, R2= peptide or protein carrier.
As LE is not antigenic itself, the peptide had to be conjugated to a protein carrier
in order to stimulate an immune reaction in the host animal. Glutaraldehyde is a
bilunctional coupling reagent that binds two compounds primarily through their amino
groups. Since LE contains only one amino group at its N-terminal end, using the
glutaraldehyde method of conjugation, the peptide is conjugated to the protein carrier via


39
the homogeneous fluorescence immunoassay and a limited quantity of antibody, obtained
through the exsanguination of a single rabbit was available.
To ensure low background fluorescence readings, the amount of BLE used in the
final assay needed to be such that close to 100 % could be bound by the antibody present.
When BLE and antibody (Ab) are present in the same solution, the equilibrium established
can be described as:
BLE + Ab ^ BLE-Ab
According to the Law of Mass Action, the following equation can be set up:
[BLE-Ab]
[BLE\[Ab]
Which can be rearranged to give:
[BLE] 1
[BLE Ab] ~ [ Ab]Ka
If 95% of the BLE in the assay is to be bound by the amount of antibody chosen for the
assay (3.78* 10'9 M, 1/50 dilution of purified antibody), the left side of the above equation
can be set to equal A. Since [Ab] = [Ab,otai] [BLE-Ab], by using the Ka value obtained
from the Scatchard plot described earlier (assuming that BLE and LE have the same
affinity for the antibody), the above equation can be solved for [BLE-Ab], [BLE-Ab]
represents 95% of the concentration of BLE to be used in the assay and therefore, the
total amount of BLE to be used in the final assay for 95% to be bound by the chosen
antibody concentration could be calculated.
The concentration of FITC-avidin to be used in the assay was determined by
finding the concentration of this reagent which will give maximum fluorescence


25
have been developed whereby structurally similar analytes are first separated by high-
performance liquid chromatography (HPLC) prior to quantitation by immunoassay. To
this end, the goal in the development of the immunoassays proposed in this dissertation
was the design of rapid and convenient assays for opioid peptides with detector-like
properties, intended for the determination of opioid peptide concentrations in HPLC
fractions. These assays should therefore allow the sensitive and straightforward analysis of
opioid peptides, be practicable on a large scale and easily amenable to intensive
automation.
Two different approaches to the development of an immunoassay for opioid
peptides were conceived, using leucine enkephalin (LE) as a model peptide. Both of these
approaches were based on the exploitation of the extremely high affinity exhibited by
avidin for biotin (Ka=10151/mol). Each avidin molecule has four high affinity binding sites
for biotin so that the central strategy of these approaches lies in the knowledge that when
avidin is conjugated to an enzyme or fluorescent label, it will still bind to biotin or
biotinylated species [Wilchek and Bayer 1988],
The first approach, an enzyme-linked immunosorbent assay (ELISA) for LE is
similar to assays previously developed for dynorphin [Hochhaus and Hu 1990] and P-
endorphin [Hochhaus and Sadee 1988] which have sensitivities in the lower fmol/assay
range. The development of a similar assay for LE would have provided a battery of
enzyme immunoassays for opioid peptides based on identical principles. This assay
involves competition between LE in the sample or standard and biotinylated LE derivative
for immobilized antibody binding sites. The antibody-bound biotinylated species is


106
mushroom tyrosinase (previously purified as described in Chapters 3 and 4) to give a final
concentration of 135 units/ml. The reaction mixture was incubated at room temperature
with constant shaking and the reaction was stopped after 1 hour by adding 66 (1 of IN
HC1 per ml of incubation solution. A control incubation was carried out in parallel,
omitting the addition of mushroom tyrosinase to insure the integrity of LE under the
reaction conditions.
Product Identification
For the isolation of the reaction products, a total of 4 ml of incubation solution
was injected into the HPLC system in 500 pi aliquots. The product peaks were collected
and pooled according to their retention times. The collected peaks were then concentrated
under a stream of nitrogen to remove organic solvents and prepared for analysis by mass
spectrometry as follows: A Sep-Pak C18 cartridge (Waters Associates, Milford, MA,
USA) was activated with 2 ml of methanol and washed with 3 ml each of methanol/3%
acetic acid 70/30 v/v and 3% acetic acid v/v. The concentrated materials were applied to
the cartridge and the cartridge was washed with 3 ml of 3% acetic acid v/v. The analytes
were then eluted in 500 pi of methanol/3% acetic acid 70/30 v/v and stored at -20C prior
to mass spectrometric analysis.
A Vestec 200ES instrument (Vestec Corp., Houston, TX, USA) was used to
obtain the electrospray ionization (ESI) mass spectra [Allen and Vestal 1992], The sample
solution was drawn into a standard laboratory syringe (250 pi, Model 1710, Hamilton Co.,
Reno, Nevada, USA) and supplied into the electrospray probe through a 50 cm x 0.1 mm


143
Roepstorff, P. and J. Fohlman (1984). Biomedical Mass Spectrometry 11: 601.
Rosei, M. A., A. Antonilli, R. Coccia and C. Blarzino (1991). Physiological, Chemical,
Physical and Medical NMR 23: 59.
Rosei, M. A., L. Antonilli, R. Coccia and C. Foppoli (1989). Biochemistry International
19: 1183.
Roth, M. (1971). Analytical Chemistry 43: 880.
Rubin, P. C. (1984). Clinical Science 66: 625.
Samuelsson, H., R. Ekman and T. Hedner (1993). Acta Anaesthesiologica Scandinavica
37: 502.
Sarma, J. K. S. R., Hoffmann and R. A. Houghten (1986). Life Sciences 38: 1723.
Sato, H. (1984). Life Sciences 35: 1051.
Scatchard, G. (1949). Annals of the New York Academy of Sciences 51: 660.
Schulman, S., G. Hochhaus and H. T. Karnes (1990). Fluorescence Immunoassay, in
Luminescence Spectroscopy (pg. 341) W. R. G. Baeyens (Ed.). New York: Marcel
Dekker.
Sesack, S. R. and V. M. Pickel (1992). Journal ofNeuroscience 12: 1335.
Shibanoki, S., S. B. Weinberger, K. Ishikawa and J. Martinez (1990). Journal of
Chromatography 532: 249.
Shimohigashi, Y. (1986 ). NIDA Research Monograph 69: 65.
Simon, E. J. (1984). Opiate Receptors, in Receptors in the Nervous System (vol. 6, pg.
331) A. Lajtha (Ed.). New York: Plenum Press.
Simonds, W. F. (1988). Endocrine Reviews 9: 200.
Smith, R. D., J. A. Loo, C. J. Barinaga, C. G. Edmonds and H. R. Udseth (1990). Journal
of the American Society of Mass Spectrometry 1: 53.
Szilagyi, J. E. (1989). Clinical and Experimental Hypertension Theory and Practice
All: 1.


67
Mushroom tyrosinase is a member of the monoxygenase class of enzymes which
catalyses two successive reactions (Figure 3.2): the hydroxylation of mono-phenols
(monophenolase activity) and the oxidation of o-diphenols (diphenolase activity) [Walsh
1979], The o-quinones resulting from these two successive reactions often form high
molecular weight polymerization products in vivo such as melanin. In this study, the
formation of undesirable polymerization products was prevented by the addition of
appropriate amounts of ascorbic acid as a reductant.
Figure 3.2. Tyrosinase catalyzed reactions
Small molecules such as L- and D-tyrosine and L- and D-dopa are endogenous
substrates for tyrosinase. However, enzyme activity has been shown with tyrosine-
containing di- and tri-peptides [Marumo and Waite 1986, Tellier et al. 1991] as well as LE
and ME [Rosei et al. 1991, Rosei et al. 1989], Larger proteins such as insulin, serum
albumin and dehydrogenase enzymes have also been shown to be oxidized by tyrosinase
[Cory et al. 1962, Cory and Frieden 1967a, Cory and Frieden 1967b, Gemant 1974, Ito et
al. 1984],
In the assay described here, the enzymatic derivatization of LE by the means
described above presents two analytical advantages. Firstly, the specific o-hydroxylation


REFERENCES
Akil, H., D. Bronstein and A. Mansour (1988). Overview of the Endogenous Opioid
Systems: Anatomical, Biochemical and Functional Issues, in Endorphins, Opiates and
Behavioural Processes (pg. 1) R. J. Rogers, & S. J. Cooper (Eds.). Chichester: John Wiley
& Sons.
Al-Hakiem, M. H. H., J. Landon, D. S. Smith and R. D. Nargessi (1981). Analytical
Biochemistry 116: 264.
Allen, M. H. and M. L. Vestal (1992). Journal of the American Society of Mass
Spectrometry 3: 18.
Arnott, D., J. Shabanowitz and D. F. Hunt (1993). Clinical Chemistry 39: 2005.
Biemann, K. (1992). Annual Reviews in Biochemistry 61: 977.
Bowden, D. W., M. Rising, G. Akots, A. Myles and R. J. Broeze (1986). Clinical
Chemistry 32: 275.
Carr, S. A. (1990). Advanced Drug Delivery Reviews 4: 113.
Clement-Jones, V. and G. M. Besser (1983). British Medical Bulletin 39: 95.
Clement-Jones, V. and L. H. Rees (1982). in Clinical Neuroendocrinology (vol. II, pg.
139) G. M. Besser, & L. Martin (Eds ). New York: Academic Press.
Cory, J. G. and E. Frieden (1967a). Biochemistry 6: 121.
Cory, J. G. and E. Frieden (1967b). Biochemistry 6: 116.
Cory, J. G., C. G. Bigelow and E. Frieden (1962). Biochemistry 1: 419.
Coty, W. A., R. Loor, N. Bellet, P. L. Khanna, P. Kaspar and M. Baier (1992). Wiener
Klinische Wochenschrift Supplement 191: 5.
Dass, C., J. J. Kusmierz, D. M. Desiderio, S. A. Jarvis and B. N. Green (1991). Journal of
the American Society for Mass Spectrometry 2. 149.
138


129
Curve 7
No derivatization, no boronate clean up
Matrix: CSF
Equation: y = 10319x 44657, r2 = 0.9770
LE mol/inj
Peak area
0
0
8.82* 10'12
93356
17.28* 1 O'12
113333
25.56* 1012
156482
33.30* 10'12
230196
43.2*1012
422738
69.29*1 O'12
714440
85.67* 10'12
841176
Curve 8
No derivatization, no boronate clean up
Matrix: CSF
Equation: y = 9006x 10275, r2 = 0.9570
LE mol/inj
Peak area
0
0
17.28* 1 O12
149435
25.56* 1 O'12
217441
33.30*10"12
193110
43.2* 10'12
463885
69.29* 1012
660135
85.67* 10'12
720965


13
Current analytical methods for opioid peptides are summarized in Table 1.2 and are
reviewed in the following pages.
Immunological Methods
At present, for the most part, opioid peptides are analyzed by radioimmunoassay
(RIA) [Sato 1984, Venn 1987], Although immunoassays are highly sensitive and
reproducible, they are hampered by several disadvantages. Firstly, an antiserum to the
analyte of interest has to be raised, which involves lengthy incubation times. Secondly,
immunoassays suffer from low selectivity due to the cross-reactivity of the antiserum to
structurally similar compounds. To overcome the problem of cross-reactivity, high
performance liquid chromatography (HPLC)-immunoassay methods have been developed
whereby collected eluted fractions from an HPLC system are analyzed by RIA [Defrutos
and Regnier 1993, McDermott et al. 1981], However, HPLC-immunoassay methods are
time consuming, labor intensive and chromatographic solutions must be volatile or
compatible with the immunoassay. Radioimmunoassays also have the added disadvantage
that the use of radioactive isotopes as tracers calls for special considerations in the
handling of the assays and the disposal of the radioactive waste produced.
Non-radiation immunological techniques have not been exploited to a great extent
for opioid peptides. An enzyme-linked immunosorbent assay (ELISA) has been described
for LE and ME [Zamboni et al. 1983] but its limits of detection are only in the 1 pmol per
assay range. Only four enzyme immunoassays with high sensitivity have been developed
for [3-endorphin and dynorphin [Hochhaus and Hu 1990, Hochhaus and Sadee 1988,


To my parents, Charles Larsimont and Kruawan Kanjanasuwan-Larsimont for their
unfailing support and encouragement of all my endeavors.


35
of 73 nmoles of LE-Lys6 in 150 pi of DMSO was also made into the HPLC system to
allow the calculation of the extent of conversion of LE-Lys6 to LE-Lys6-B. C-terminal
biotinylated LE incorporating a spacer arm between the biotin group and the peptide (LE-
Lys6-BX) was synthesized in the same manner except that BXHS was used instead of
BHS in the incubation mixture. The peaks corresponding to LE-Lysf-B and LE-Lys6-BX
were collected from the HPLC eluent and stored as described earlier for BLE and BXLE.
The successful biotinylation of LE by the methods described above was
characterized by comparing the displacement of 2-(4-hydroxyazobenzene)-benzoic acid
(HABA) from avidin by the biotinylated LE derivatives and biotin itself according to the
spectrophotometric method of Green [Green 1970], A 0.25 mM solution of HABA was
prepared in 0.1 M sodium phosphate buffer pH 7 and avidin was added to an aliquot of
this solution to give a final concentration of 45 pM. One milliliter of this solution was
placed in a UV quartz cuvette and upon serial additions of the biotinylated species or
biotin as a positive control, the absorbance at A.5oonm was monitored using a Cary 3E UV-
Visible spectrophotometer (Varan, Sugarland, TX, USA). The same experiment was
repeated using each biotinylated LE derivative.
The structures of LE-Lys6-B and LE-Lys6-BX were confirmed by mass
spectrometric analysis by the Protein Chemistry Core of the Interdisciplinary Center for
Biotechnology Research, University of Florida, Gainesville, FL, USA. A Lasertech matrix-
assisted laser desorption (MALDI) time-of-flight (TOF) instrument operated at 10 kV
acceleration was used with a 335 nm UV laser and a-cyano-4-hydroxycinnamic acid as the
matrix. The sample was dissolved in a solution (acetonitrile/water, 1/2 v/v) containing a


36
large excess of the matrix (5g/l) and a 1 pi aliquot was deposited onto the stainless steel
laser target. Several spectra were averaged for each experiment.
Antibody-Binding of LE. Biotinylated LE Derivatives and Preformed Biotinylated
LE-Avidin Complexes
The binding of LE, the biotinylated LE derivatives and preformed biotinylated LE-
avidin complexes to both a commercial anti-LE antiserum (Peninsula) and the antibody
produced in this laboratory was tested using a standard radioimmunoassay procedure. For
the assays using preformed biotinylated LE-avidin complexes, a ten-fold molar excess of
avidin was added to the highest concentration of biotinylated LE in buffer and vortexed
immediately. The complexes were allowed to form for 30 minutes prior to serial dilution
to give the various concentrations of competitor required for the assay. The assays were
set up in microcentrifuge tubes as shown in Table 2.2. The RIA buffer used consisted of
0.1 M sodium phosphate buffer pH 7 containing 0.1% w/v BSA. The tracer was 3H-LE
(2.6* 10'10 M final concentration) and the competitor was various concentrations of either
LE, the biotinylated LE derivatives or preformed biotinylated LE-avidin complexes. After
overnight incubation at 4C, 200 pi of an ice cold suspension containing 1.5% w/v
activated charcoal and 0.15% w/v dextran in water was added to each tube (except total
counts) and after a further 5 minute incubation on ice, the samples were centrifuged at
12,000g for 3 minutes. Three hundred and fifty microliters of the resultant supernatant
containing the antibody-bound fraction of the tracer was then removed and added to 4 ml
of Cytoscint scintillation cocktail. The radioactivity (CPM) representing the antibody-


CHAPTER 1
INTRODUCTION
Endogenous Opioid Peptides
There exists three separate families of natural or endogenous opioid peptides:
enkephalins, endorphins and dynorphins. These three families are derived from three
different prohormones, proenkephalin, pro-opiomelanocortin and prodynorphin,
respectively and are coded by messenger RNAs from three separate genes (Figure 1.1)
[Pleuvry 1991],
Alpha-melanocyte stimulating
hormone*
Corticotrophln-like intermediate
lobe peptide
Beta-endorphin
4
Adrenocorticotrophin*
Beta-lipotrophin*

IPRO-OPIOMELANOCORfl
4
m-RNA
A
GENES
Beta-melanocyte stimulating
hormone*
Leumorphin
Beta-neo-endorphin
m-RNA
*
IPRO-ENKEPHALINI
. Peptide E
t
[Metlenkephalin
[Leulenkephalin
[Metlenkephalin-Arg-Phe
Dynorphin A 0-17)
m-RNA-^[po.DY[goRPH)
X" Dynorphin B (1-13)
Dynorphin (1-8)
Peptide F
|Met]enkephalin-Arg-Gly-Leu
Figure 1.1. Opioid peptides and their precursors. (* Denotes no opioid receptor activity)
[Pleuvry 1991]
1


90
Methods
Purification of Mushroom Tyrosinase
Mushroom tyrosinase was purified prior to use by ultra-filtration using Centricon
membrane filters (molecular weight cut off 30,000, Amicon, Danvers, MA, USA). One
milliliter of a solution of mushroom tyrosinase (1 mg/ml) in 0.1 M phosphate buffer pH 7
was applied to the filter unit and centrifuged at 5,000 g until maximum concentration of
the sample was achieved. This centrifugation step was repeated three times with the
addition of an additional 2 ml of phosphate buffer prior to each centrifugation. The final
concentrate was reconstituted in phosphate buffer to give a final concentration of 1 mg/ml
of mushroom tyrosinase corresponding to 3870 units of activity per ml of solution.
Aliquots were stored at -20C and defrosted immediately prior to use.
Enzymatic Derivatization
To characterize the derivatization procedure, LE (1 niM) in 0.5 M phosphate
buffer pH 7.4 was reacted with mushroom tyrosinase (135 units/ml) in the presence of
ascorbic acid (50 mM) at room temperature with constant shaking. The product peaks
obtained from this enzymatic derivatization were collected from the HPLC eluent, organic
solvent was removed by evaporation under a stream of nitrogen and the collected peaks
were concentrated using a Sep-Pak Ci8 preparative column (Waters Associates, Milford,
MA, USA). The identity of the products was determined by electrospray ionization mass
spectrometry (see Chapter 5).


3
melanocyte simulating hormone (MSH, Figure 1.2). The term endorphin applies to opioid
peptides derived from POMC. POMC is synthesized in the pituitary but is also present in
the hypothalamus and in the periphery. The ME amino acid sequence is present at the N-
terminal end of (3-endorphin.
The dynorphins as well as P-neoendorphin and leumorphin are derived from
prodynorphin which is also known as proenkephalin B. Prodynorphin contains LE
sequences but no ME sequences and is synthesized throughout the central nervous system
(Figure 1.2).
When the opioid peptide precursors are processed to give the various different
opioid peptides, the N-terminal end of the molecules is highly conserved with tyrosine in
the 1 position. However, tyrosine is often absent in this position in non-related peptides.
This phenomenon is exploited in two of the analytical approaches described in this
dissertation.
Opioid Receptors
Opioid receptors are widely distributed throughout the central nervous system of
all vertebrates and have also been found in a number of peripheral tissues including the
intestinal tract, the adrenal and pituitary glands and the vasa diferentia of several species
[Simon 1984],
At present, three classes of opioid receptors are firmly recognized, predominantly
on the basis of in vivo studies of opioid action (agonists and antagonists), in vitro
bioassays and binding experiments with selective ligands [Simonds 1988], These three


CHAPTER 6
CONCLUSIONS
In the work carried out for this dissertation, the feasibilty of both an enzyme-linked
immunosorbent assay (ELISA) and an homogeneous fluorescence immunoassay for
leucine enkephalin (LE) was evaluated. Two high-performance liquid chromatography
(HPLC) approaches for opioid peptides using LE as a model peptide and enzymatic
derivatization by tyrosinase were also evaluated. One of these HPLC approaches used
electrochemical detection as a means of quantitation and the other used fluorescence
detection following a second fluorogenic derivatization step with 1,2-diamino-1,2-
diphenylethane (DPE). In addition, since hydroxylated products were produced from the
reaction between tyrosinase and leucine enkephalin in the development of the HPLC
approaches, and since tyrosinase and enkephalins have been found to co-exist in vivo, the
identity of these products was determined by mass spectrometry and their biological
activity in rat brain homogenate was investigated.
The formation of a sandwich between an anti-LE antibody, a biotinylated LE
derivative and avidin conjugated to an enzyme was a requirement for the successful
development of the ELISA proposed for this dissertation since in this assay, LE and
biotinylated LE should compete for antibody binding sites and antibody-bound biotinylated
LE should subsequently be detected through the use of enzyme-labeled avidin on addition
of the enzyme substrate. The formation of a sandwich could not be achieved using either
121


122
the antibody produced in this laboratory or the commercial antiserum tested and any of the
biotinylated derivatives synthesized. Therefore, a workable ELISA for LE could not be
developed using the proposed approach. ELIS As for |3-endorphin and dynorphin Al-13
have already been developed using the same strategy proposed here for the development
of an ELISA for LE [Hochhaus and Sadee 1988, Hochhaus and Hu 1990], However,
these opioid peptides consist of longer amino acid chains than LE (31 amino acids for (3-
endorphin and 13 amino acids for dynorphin compared to only 5 amino acids for LE) and
therefore, it is suggested that if a biotinylated LE derivative including a longer spacer arm
can be synthesized, a sandwich between anti-LE antibody, this biotinylated LE derivative
and avidin could be formed, leading to the successful development of an ELISA for LE
based on the same principles used in the ELISAs for fl-endorphin and dynorphin Al-13
which have been described earlier [Hochhaus and Sadee 1988, Hochhaus and Hu 1990],
The successliil development of the homogeneous fluorescence immunoassay
proposed in this dissertation relied on a lack of sandwich formation between an anti-LE
antibody, a biotinylated derivative and fluorescence-labeled avidin since here, LE and
biotinylated LE compete for antibody binding sites, and subsequently, free biotinylated LE
is detected on addition of fluorescence-labeled avidin due to the increase in fluorescence
observed on interaction between biotinylated LE and fluorescence-labeled avidin. A lack
of sandwich formation was seen using the anti-LE antibody produced in this laboratory, a
biotinylated derivative biotinylated at the N-terminal end of the peptide without a spacer
arm (BLE) and fluorescein isothiocyanate-labeled avidin (FITC-avidin). Although as
expected, an increase in fluorescence was seen in our homogenous fluorescence


65
A fluorescence spectrophotometer set at close to maximum signal amplification
was used to measure the fluorescence signal in the development of the homogenous
fluorescence immunoassay. The use of laser-induced fluorescence may have allowed the
successful development of the proposed homogeneous immunoassay as less BLE could
have interacted with a lower concentration of FITC-avidin to produce a measurable signal.
Therefore, less LE would have been required to displace BLE from antibody binding sites
so that the higher concentrations of LE which were observed to produce a direct
interaction with FITC-avidin would not have been attained.
In conclusion, although it has been demonstrated that the homogenous
fluorescence immunoassay for LE proposed in this study works in theory, a fully
operational assay could not be developed in practice as the concentrations of LE required
in this assay to displace the maximum amount of BLE from the antibody interacted
directly with FITC-avidin to produce fluorescence enhancement of the detector molecule.
Existing homogeneous fluorescence immunoassays for haptens and proteins lie in the 10'8
to 109 moles/ml range [Jenkins 1992] and homogeneous enzyme immunoassays are
capable of limits of detection 10'14 moles/ml of haptens [Engel and Khanna 1992], It is
proposed that the use of an antibody with higher affinity for LE would allow the
successful development of the type of homogeneous fluorescence immunoassay for LE
described in this study as lower concentrations of LE could be used and therefore the
direct interaction between LE and FITC-avidin that was encountered would be avoided. It
is anticipated that the successful development of this type of homogeneous immunoassay
would allow the detection of LE in the 10'9 to 10'11 moles/ml range.


88
position of opioid peptides by mushroom tyrosinase. Here, the enzymatic derivatization
renders the peptide suitable for subsequent fluorogenic derivatization using 1,2-diamino-
1,2-diphenylethane (DPE). Therefore, in this assay, the hydroxylated LE derivative
obtained from the reaction with mushroom tyrosinase is oxidized in a controlled manner
by potassium ferricyanide to give the corresponding quinone prior to a condensation
reaction with DPE to give a fluorescent product (Figure 4.2).
Tyrosinase reaction
OH
OH
O
V1O2
T
tyrosinase
OH
Polymerization
DPE reaction
Oxidation

Potassium
ferricyanide
Hydroxylated LE Corresponding quinone
CH-CH-
I I
NHj NHj
Condensation
'1
DPE
Fluorescent product
Figure 4.2. Overview of derivatization reactions for HPLC-FL assay


26
subsequently detected through the use of an avidin-enzyme complex (Figure 2.1). The
successful design of this type of assay for LE depends on the synthesis of a suitable
biotinylated LE derivative that will allow the formation of a sandwich between the
antibody, the biotinylated LE derivative and the enzyme-labeled avidin, thus enabling
quantification upon addition of the enzyme substrate.
Substrate Detection
\ Antibody
Biotinylated analyte
<^|||| Analyte
JL/E
nr
Enzyme-labelled avidin
Figure 2.1. Avidin-biotin based ELISA
The second approach, which is an homogeneous or separation-free fluorescence
immunoassay is based on an observation by Al-Hakiem et al. [Al-Hakiem et al. 1981] that
the binding of biotin to fluorescein-labeled avidin leads to an increase in fluorescence
intensity. This phenomenon is exploited in an attempt to develop a new class of
homogeneous immunoassay which depends on the inability of antibody-bound biotinylated


4
classes of receptors are defined as mu, kappa and delta. The existence of sigma and
epsilon receptors has also been postulated; however, there is now some doubt as to
whether the sigma receptor is truly "opioid" in nature and epsilon receptors have not been
detected with any degree of certainty in any tissue except rat vas deferens. The cloning of
opioid receptors using recombinant DNA techniques can be expected to yield information
about the differences in the primary amino acid sequence of mu, delta and kappa
receptors, and may also uncover new opioid receptor subtypes thus elucidating the
structural features responsible for receptor specificity.
^.cingulate gyrus
subcallosal striatur
- striatum
medial thalamic nuclei
- habenula
- ventral anterior nucleus
~ septal region
" globus pallidus
~~ hypothalamus
^inferior frontal
olfactory trigone
olfactory bulb
amygdala
temporal lobe
*** hippocampus
Figure 1.3. Distribution of opioid receptors in the human brain [Simon 1984],
No selective endogenous ligand has yet been isolated for the mu receptor as opioid
peptides from all three families bind to this receptor. It has therefore been suggested that
the mu receptor may be a "universal" opioid receptor [Akil et al. 1988], Mu opioid
receptor sites are widespread but are found in particular in the regions of the brain
associated with pain regulation and sensorimotor integration [Mansour et al. 1988], It has
been proposed that mu receptors may be subdivided into high affinity sites (mui) which
occipital lobe
corpora quadrlgemi
perioQuaductal gra
nm<
l gray
tubstantla nigra
interpeduncular n.
locus coeruleu
mid-line reitic form
reticular formation
orea postrema
cerebellum
substantia
gelatinosa


83
current HPLC-ED methods for enkephalins (Table 3.3) [Fleming and Reynolds 1988, Kim
et al. 1989, Shibanoki et al. 1990], The increased practicability of the analytical approach
described here is characterized by the minimal precautions required in the preparation of
the mobile phase (filtering and degassing under negative pressure with sonication) and is
due to the fact that the enzymatic derivatization employed allowed the use of lower
applied potentials for electrochemical detection (+0.3 V compared to +0.85-1.25 V for
existing HPLC-ED methods). The low applied potentials used allowed the electrochemical
detector to be operated at the maximum gain setting since background current and
baseline noise were low. These factors, in addition to the minimal baseline drift observed,
increased the ease of handling of this analytical approach since minimal precautions could
be taken in the preparation of mobile phase and samples. This was demonstrated when a
comparison was made between the analysis of LE in CSF using the complete HPLC-ED
method incorporating enzymatic derivatization and boronate clean up and the HPLC-ED
analysis of LE in CSF without derivatization or boronate clean-up. When an attempt was
made to analyze LE in CSF without derivatization or boronate clean-up, considerably
higher applied potentials had to be used (+0.7 V compared to +0.3 V) and as a
consequence, background current, baseline noise and baseline drift were greatly increased.
Therefore, the maximum gain setting of the instrument could not be used, sensitivity was
compromised by a factor of 25 and the linearity of the calibration curves was reduced. In
addition, the system was more difficult to handle as the mobile phase had to be continually
degassed under a stream of helium in order to avoid unacceptable baseline drift.


no
(MicroMath Scientific software, Salt Lake City, UT, USA). The data were fitted to the
following model:
B=T-
j1 j)c
CN+IC:
-+NS
50
Where: B = CPM in the presence of competitor
T = CPM in the absence of competitor
C = competitor concentration
N = slope factor
NS = CPM under non-specific binding conditions
Results
Chromatography
As in Chapters 3 and 4, leucine enkephalin and the products of the enzyme
reaction were separated successfully from each other using the chromatographic system
described above (Figure 5.2). In a control incubation where LE and ascorbic acid were
injected into the system in concentrations equal to the initial concentrations of these
reagents in the incubation mixture, LE had a retention time of 9.5 minutes and was well
separated from ascorbic acid which eluted in a large solvent front at 1.4 minutes (Figure
5.2A). When the test incubation was applied to the system, two distinct additional peaks
were seen on the chromatogram with retention times of 5.7 minutes and 7.3 minutes,
respectively (Figure 5.2B). These peaks were collected and subjected to mass
spectrometry as described earlier. An incubation time of one hour was chosen for our
experiments as at this time point, most of the LE present was seen to have been consumed
in the reaction and the levels of the major product were seen to plateau (Figure 5.1).


45
produced in this laboratory (3.8* 10'13 mol/assay) was similar to those obtained using a
1/3,000 dilution of the commercial antiserum (1.3* 101, mol/assay). Peninsula claimed that
IC50 values of 1.4*1 O'14 mol/assay could be achieved using a 1/15,000 dilution of their
antiserum. However, their assay conditions involved the use of a tracer labeled with l25I
instead of 3H. 125I is a higher energy label than 3H, and therefore, Peninsula was able to use
a lower concentration of tracer in their assay, which allowed them to use less antiserum,
resulting in a lower IC50 value.
Figure 2.6. Scatchard plot of bound/free 3H-LE versus bound 3H-LE
Biotinylation of Leucine Enkephalin
Since one of the proposed immunoassays required the formation of a sandwich
between the anti-LE antibody, the biotinylated LE derivative and fluorescence- or enzyme-
labeled avidin, and the other did not, several different biotinylated derivatives were


NOVEL ANALYTICAL APPROACHES FOR THE DETERMINATION OF LEUCINE
ENKEPHALIN AS A MODEL FOR OPIOID PEPTIDES
By
VERONIQUE LARSIMONT
A 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
1994

To my parents, Charles Larsimont and Kruawan Kanjanasuwan-Larsimont for their
unfailing support and encouragement of all my endeavors.

ACKNOWLEDGMENTS
My thanks go to my advisor, Dr. Giinther Hochhaus and the members of my
supervisory committee, Dr. Hartmut Derendorf, Dr. Paul Klein, Dr. Laszlo Prokai and Dr.
Ian Tebbett for their guidance and support during the course of my doctoral research.
Special thanks go to Dr. Prokai for the mass spectrometry analysis of hydroxylated leucine
enkephalin derivatives. I am also grateful to Dr. Richard Prankerd for his help in the early
stages of this work.
I acknowledge the PDA. Foundation for Pharmaceutical Sciences, Inc. and
Schering-Plough Corporation for partial funding of the research presented in this
dissertation.
I would also like to recognize the services provided by the Hybridoma Core and
the Protein Chemistry Core of the Interdisciplinary Center for Biotechnology Research at
the University of Florida.
There are many others who are too numerous to mention, who have been
instrumental in enabling me to complete this work. I hope to be able to thank each of them
personally.
in

TABLE OF CONTENTS
page
ACKNOWLEDGMENTS iii
ABSTRACT vi
CHAPTERS
1 INTRODUCTION 1
Endogenous Opioid Peptides 1
Opioid Receptors 3
Physiology and Pharmacology 5
Rationale 12
Objectives 19
2 APPROACHES TO THE DEVELOPMENT OF AN IMMUNOASSAY
FOR LEUCINE ENKEPHALIN 23
Introduction 23
Materials 28
Methods 28
Results and Discussion. 41
3 A HIGH PERFORMANCE LIQUID CHROMATOGRAPHY ASSAY FOR
OPIOID PEPTIDES USING ELECTROCHEMICAL DETECTION 66
Introduction 66
Materials 69
Methods 70
Results 75
Discussion 82
4 A HIGH PERFORMANCE LIQUID CHROMATOGRAPHY ASSAY
FOR OPIOID PEPTIDES USING FLUORESCENCE DETECTION 87
Introduction 87
Materials 89
Methods 90
iv

Results 93
Discussion 100
5 LEUCINE ENKEPHALIN-TYROSINASE REACTION PRODUCTS -
IDENTIFICATION AND BIOLOGICAL ACTIVITY 104
Introduction 104
Materials 105
Methods 105
Results 110
Discussion 118
6 CONCLUSIONS 121
APPENDICES
A DATA FOR HPLC-ED APPROACH 126
B DATA FOR HPLC-FL APPROACH 131
REFERENCES 138
BIOGRAPHICAL SKETCH 145
v

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
NOVEL APPROACF1ES FOR THE DETERMINATION OF LEUCINE ENKEPHALIN
AS A MODEL FOR OPIOID PEPTIDES
By
Vronique Larsimont
August 1994
Chairman: Giinther Hochhaus
Major Department: Pharmaceutics
The focus of this dissertation was the evaluation of novel analytical approaches for
opioid peptides by immunoassay or high performance liquid chromatography (HPLC)
using leucine enkephalin (LE) as a model peptide.
The proposed immunoassays are based on the high affinity exhibited by avidin for
biotin (Kd=1015 mol/1). The successful development of the enzyme-linked immunosorbent
assay relied on the formation of a sandwich between anti-LE antibody, a biotinylated LE
derivative and avidin, whereas the successful development of the homogeneous
fluorescence immunoassay depended on a lack of sandwich formation. The formation of a
sandwich was not achieved using any combination of the two anti-LE antibody
preparations and several biotinylated LE derivatives tested, and therefore efforts in this
direction were abandoned. However, using a polyclonal antibody produced in this
laboratory, an N-terminal biotinylated LE derivative without a spacer arm and fluorescein
vi

isothiocyanate avidin, an homogeneous fluorescence immunoassay for LE was developed
which was operational in a narrow concentration range (1*1 O'9 to 1*10"8 moles LE/ml).
Two HPLC assays for opioid peptides were evaluated. One was based on tyrosine-
specific pre-column hydroxylation using tyrosinase, specific sample clean-up using a
boronate gel and HPLC with electrochemical detection. The other involved tyrosine-
specific pre-column hydroxylation using tyrosinase followed by fluorogenic derivatization
using 1,2-diamino-1,2-diphenylethane and HPLC with fluorescence detection. These
assays yielded limits of detection for LE of 170 fmol/inj and 500 fmol/inj respectively in
buffer samples and 360 fmol/inj and 500 fmol/inj respectively in spiked cerebrospinal fluid
samples.
Using electrospray ionization mass spectrometry, the structure of the products of
the reaction between LE and tyrosinase were found to be monohydroxylated LE ([HO-
Tyrj-LE) and dihydroxylated LE ([(HO)2-Tyr']-LE). Compared to LE, the affinity of
[HO-Tyr*]-LE to both p and 5 opioid receptor sites in rat brain homogenate was found to
be lower by a factor of about 20. Since enkephalins and tyrosinase have been found to co
exist in vivo, we speculate that tyrosinase may play role in the metabolic pathway of these
compounds.

CHAPTER 1
INTRODUCTION
Endogenous Opioid Peptides
There exists three separate families of natural or endogenous opioid peptides:
enkephalins, endorphins and dynorphins. These three families are derived from three
different prohormones, proenkephalin, pro-opiomelanocortin and prodynorphin,
respectively and are coded by messenger RNAs from three separate genes (Figure 1.1)
[Pleuvry 1991],
Alpha-melanocyte stimulating
hormone*
Corticotrophln-like intermediate
lobe peptide
Beta-endorphin
4
Adrenocorticotrophin*
Beta-lipotrophin*

IPRO-OPIOMELANOCORfl
4
m-RNA
A
GENES
Beta-melanocyte stimulating
hormone*
Leumorphin
Beta-neo-endorphin
m-RNA
*
IPRO-ENKEPHALINI
. Peptide E
t
[Metlenkephalin
[Leulenkephalin
[Metlenkephalin-Arg-Phe
Dynorphin A 0-17)
m-RNA-^[po.DY[goRPH)
X" Dynorphin B (1-13)
Dynorphin (1-8)
Peptide F
|Met]enkephalin-Arg-Gly-Leu
Figure 1.1. Opioid peptides and their precursors. (* Denotes no opioid receptor activity)
[Pleuvry 1991]
1

2
In the proenkephalin family, proenkephalin A is the precursor of the pentapeptides
methionine enkephalin (ME, Tyr-Gly-Gly-Phe-Met) and leucine enkephalin (LE, Tyr-Gly-
Gly-Phe-Leu) which were the first opioid peptides to be characterized by Hughes and
coworkers in 1975 (Figure 1.2) [Hughes et al. 1975], Proenkephalin A has been shown to
contain ME and LE in a fixed ratio of six ME sequences to one LE sequence.
Proenkephalin-expressing cells are widespread throughout the brain and spinal cord as
well as in more peripheral sites such as the adrenal medulla and the gastrointestinal tract.
PRO-OPIOMELANOCORTIN
r-MSHj
r H i_r .
( -LPH *
*-/9-END-i
'l'
If1
in
a-MSH 0-MSH MET-ENK
PROENKEPHALIN
MET-ENK LEU-ENK
gl g
W HI
II 1
1 I
hrl
hrl
MET-ENK MET-ENK
ARG*- 6LY7-LEU* ARG*-PHE7
PRODYNORPHIN
LEU-ENK
Figure 1.2. Schematic representation of the structures of opioid peptide precursors [Jaflfe
and Martin 1990],
Pro-opiomelanocortin (POMC) is the precursor of the opioid peptide P-endorphin
as well as the non-opioid hormones adrenocorticotrophic hormone (ACTH) and a- and P-

3
melanocyte simulating hormone (MSH, Figure 1.2). The term endorphin applies to opioid
peptides derived from POMC. POMC is synthesized in the pituitary but is also present in
the hypothalamus and in the periphery. The ME amino acid sequence is present at the N-
terminal end of (3-endorphin.
The dynorphins as well as P-neoendorphin and leumorphin are derived from
prodynorphin which is also known as proenkephalin B. Prodynorphin contains LE
sequences but no ME sequences and is synthesized throughout the central nervous system
(Figure 1.2).
When the opioid peptide precursors are processed to give the various different
opioid peptides, the N-terminal end of the molecules is highly conserved with tyrosine in
the 1 position. However, tyrosine is often absent in this position in non-related peptides.
This phenomenon is exploited in two of the analytical approaches described in this
dissertation.
Opioid Receptors
Opioid receptors are widely distributed throughout the central nervous system of
all vertebrates and have also been found in a number of peripheral tissues including the
intestinal tract, the adrenal and pituitary glands and the vasa diferentia of several species
[Simon 1984],
At present, three classes of opioid receptors are firmly recognized, predominantly
on the basis of in vivo studies of opioid action (agonists and antagonists), in vitro
bioassays and binding experiments with selective ligands [Simonds 1988], These three

4
classes of receptors are defined as mu, kappa and delta. The existence of sigma and
epsilon receptors has also been postulated; however, there is now some doubt as to
whether the sigma receptor is truly "opioid" in nature and epsilon receptors have not been
detected with any degree of certainty in any tissue except rat vas deferens. The cloning of
opioid receptors using recombinant DNA techniques can be expected to yield information
about the differences in the primary amino acid sequence of mu, delta and kappa
receptors, and may also uncover new opioid receptor subtypes thus elucidating the
structural features responsible for receptor specificity.
^.cingulate gyrus
subcallosal striatur
- striatum
medial thalamic nuclei
- habenula
- ventral anterior nucleus
~ septal region
" globus pallidus
~~ hypothalamus
^inferior frontal
olfactory trigone
olfactory bulb
amygdala
temporal lobe
*** hippocampus
Figure 1.3. Distribution of opioid receptors in the human brain [Simon 1984],
No selective endogenous ligand has yet been isolated for the mu receptor as opioid
peptides from all three families bind to this receptor. It has therefore been suggested that
the mu receptor may be a "universal" opioid receptor [Akil et al. 1988], Mu opioid
receptor sites are widespread but are found in particular in the regions of the brain
associated with pain regulation and sensorimotor integration [Mansour et al. 1988], It has
been proposed that mu receptors may be subdivided into high affinity sites (mui) which
occipital lobe
corpora quadrlgemi
perioQuaductal gra
nm<
l gray
tubstantla nigra
interpeduncular n.
locus coeruleu
mid-line reitic form
reticular formation
orea postrema
cerebellum
substantia
gelatinosa

5
are thought to mediate supraspinal analgesia and a low affinity sites (imi2) which are
thought to be responsible for respiratory depression and gastrointestinal effects [Pasternak
1982],
Leucine enkephalin and other derivatives of proenkephalin A interact with the delta
receptor, although not selectively. Opiate alkaloids, on the other hand, have low affinity
for this receptor. Delta receptors are less widespread than mu receptors but are
concentrated in neural areas involved with olfaction and motor integration and have been
implicated in pain pathways [Mansour et al. 1988],
The derivatives of prodynorphin have selectivity for kappa receptors. Although LE
is selective for the delta receptor, as the molecule is lengthened, its preference for the delta
receptor is reduced and its affinity for the kappa receptor increases. Kappa receptors are
found predominantly in brain areas associated with pain perception and the regulation of
water balance and food intake [Mansour et al. 1988],
Physiology and Pharmacology
The binding of an opioid to its receptor triggers a number of complex processes
which occur before leading to the ultimate opioid effect. Opioid effects are believed to
mediated through guanine nucleotide regulatory proteins (G proteins) which are involved
in signal transduction to a variety of effector systems including adenylate cyclase,
phospholipidase C and ion channels. Presently, adenylate cyclase inhibition is the best
characterized opioid effect mediated by G proteins [Simon 1984],

6
Adenylate cyclase is an enzyme that synthesizes cyclic AMP from ATP so that it
can then go on to act as a "second messenger" in a number of biochemical systems. G-
proteins cause adenylate cyclase to convert from the active form of the enzyme which is
coupled to GTP to the inactive form which is coupled to GDP and vice versa. Therefore, a
compound that stimulates adenylate cyclase production would do so through the
interaction of its receptor with a G-protein that converts adenylate cyclase to the active
GTP-coupled form, whereas inhibition of adenylate cyclase, as by opioids for example,
would be mediated through a G-protein that favors the formation of inactive GDP-coupled
adenylate cyclase (Figure 1.4).
GDP GTP
hydrolysis
Figure 1.4. Scheme for receptor mediated stimulation and inhibition of adenylate cyclase.
Rs receptors which produce stimulation of adenylate cyclase on binding, Ri -
receptors which produce inhibition of adenylate cyclase on binding, AC -
adenylate cyclase [Simon 1984],
Opioid peptides are distributed widely throughout the central and peripheral
nervous system, suggesting that these compounds play a part in a variety of physiological
functions [Olson et al. 1991], The physiological roles of endogenous opioid peptides have
been attributed largely on the basis of effects seen on administration of the opioid

7
antagonist naloxone. These effects are assumed to be the result of opioid receptor
blockade and are widely accepted as indirect evidence for endogenous opioid involvement
in the physiological function under observation.
One of the most important functions of opioid peptides is in pain modulation. The
mu receptor is the opiate receptor most associated with pain relief, although delta and
kappa receptor agonists also have analgesic properties. It is thought that under some
conditions, mu and delta receptors are functionally coupled as delta agonists given in sub
analgesic doses have been found to either potentiate or inhibit the analgesic effects of
morphine (a mu agonist) at different sites. It has therefore been postulated that mu and
delta receptors may exist either separately or in a complexed form [Rapaka and Porreca
1991], Acupuncture analgesia is also thought to be mediated by endogenous opioid
peptides [Clement-Jones and Rees 1982],
Animal studies have indicated that endogenous opioid peptides play a role in the
development of opiate dependence, a side effect common to opiate analgesics. This is a
syndrome whereby distress is caused upon withdrawal of an opiate following chronic
administration. It has been hypothesized that chronic narcotic abuse leads to the
suppression of endogenous opioid production through a negative feedback mechanism so
that sudden withdrawal of the narcotic leads to a deficiency in endogenous opioids which
causes the classical physical withdrawal symptoms [Clement-Jones and Besser 1983],
Recently, Wang et al. [Wang et al. 1994] have proposed that stimulation by an opiate
agonist causes gradual constitutive mu receptor activation so that an agonist is no longer
required for signal transduction, and a dependent state is established, consisting of an

8
upregulated cAMP system, counterbalanced by constitutively active mu receptors. In this
model, opiate tolerance results from fewer mu receptors remaining activatable by agonists
and the enhanced activity of the cAMP system. In other words, in this scenario
dependence occurs because an upregulated cAMP system is established which needs to be
counterbalanced by opiate agonist activity and tolerance occurs because fewer mu
receptors can now be activated by opiate agonists.
A lower degree of dependence is seen with agonists at delta receptors than with
agonists at mu receptors. As mentioned above, the action of mu receptor agonists can be
modulated by the co-administration of delta receptor agonists so that the potency and
efficacy of analgesia is increased without a corresponding increase in side effects such as
physical dependency, respiratory depression and gastrointestinal effects. This effect could
be exploited to allow for the use of mu agonists of lower efficacy and increased safety
while still providing adequate pain relief without the risk of side effects. Eventually, a delta
agonist may be developed which is able to provide effective pain relief without side effects
[Rapaka and Porreca 1991],
Opioid peptides play a part in the regulation of the immune system, particularly
during periods of stress, as they modulate the functions of a number of cell types involved
in the immune response [Murgo et al. 1986], Generally, endogenous opioid peptides are
immunostimulant as they enhance T-cell function and stimulate phagocyte function, thus
increasing resistance to infection. There is also evidence that suggests a role for
endogenous opioids in the growth and development of lymphoid tissue [Plotnikoff et al.
1985],

9
Opioid peptides are thought to depress the responsiveness of the chemosensors to
carbon dioxide and may therefore play a physiological role in the control of respiration
[McQueen 1983], This applies in particular to neonates and to adults in stressful situations
[Pleuvry 1991],
Opioid peptides may also be involved in blood pressure regulation as they are
present in nerve fibers in areas of the brain stem responsible for the regulation of blood
pressure and the secretion of vasopressin. Biochemical evidence suggests that opioid
peptides interact with neurohormones to regulate blood pressure. Opioid peptides have
been implicated in the dramatic changes in blood pressure which occur during sleep and in
hypotension due to various states of shock [Rubin 1984], It has also been suggested that
endogenous opioid peptides may be involved in the pathogenesis of hypertension [Szilagyi
1989],
The presence of opioid peptides within limbic structures suggests their
involvement in the regulation of mood and behavior. Endogenous opioid peptides are also
known to interact with the central catecholamines implicated in psychiatric disease.
However, the results of studies carried out to determine the role of opioid peptides in
psychiatric disease have been contradictory [Clement-Jones and Besser 1983, Koob and
Bloom 1983], and therefore no conclusions can be drawn at present.
The very high concentrations of opioid peptides present in the hypothalamus
suggests a role for these substances in neuroendocrine regulation. Opioid peptides may
control the secretion of anterior pituitary hormones by modifying the release of
hypothalamic anterior pituitary regulating substances. This may be the mechanism by

10
which opioid peptides cause an increase in the secretion of prolactin, growth hormone and
thyrotrophin and inhibit the release of luteinizing hormone, follicle stimulating hormone,
adrenocorticotrophic hormone and beta- and gamma-lipotropin [Clement-Jones and
Besser 1983, Grossman and Rees 1983], Further possible functions of opioid peptides can
be found in Table 1.1.
Table 1.1. Possible physiological functions of endogenous opioid peptides [Imura et al.
1985],
1. Defense against noxious stimuli
Activation of the pituitary-adrenocortical axis
Regulation of the sympatho-adrenal system
Inhibition of pain perception
2. Modulation of vegetative nervous system
Cardiovascular and respiratory system
Gastrointestinal tract and pancreas
Genito-urinary tract
3. Modulation of neuroendocrine function
Anterior and posterior pituitary hormones
Gastrointestinal and pancreatic hormones
Catecholamines
4. Behavioral action
Mood and locomotor activity
Food and water intake
Sexual behavior
The design of opioid peptides as therapeutic agents has several advantages. Firstly,
these substances are endogenous so that their metabolites are likely to be non-toxic and
not to cause renal or hepatic damage, depending on the doses administered. Secondly, a
large number of analogs can be synthesized from a few basic amino acid building blocks as
synthesis has been simplified and automated and simple modifications can be used to

11
develop different analogs with desirable biological activities. Opioid peptides are unable to
cross the placental barrier as they are subject to placental enzymatic deactivation and
would therefore be ideal for obstetric use [Rapaka 1986, Rapaka and Porreca 1991],
At present, the therapeutic development of opioid peptides is focused on their
potential as analgesic agents and in the treatment of opiate addiction. Research efforts are
directed towards the design of analgesic peptides which can be administered orally, have a
long duration of action and reduced potential for dependence and abuse. Peptides of
interest include enkephalins, endorphins and related opioid peptides.
Opioid peptides are easily degraded by aminopeptidases which hydrolyze the Tyr1-
Gly2 bond, carboxypeptidases which cause cleavage at the C-terminal end of the molecule,
relatively non-specific enzymes such as trypsin and angiotensin converting enzyme (ACE)
and more specific enzymes such as enkephalinases which hydrolyze the Gly3-Phe4 in
enkephalins. This has led to the suggestion that enkephalin degrading enzymes be used as
an alternative therapeutic approach [Rapaka 1986, Rapaka and Porreca 1991], Examples
of these enzyme inhibitors include bestatin, thiorphan and captopril which inhibit
aminopeptidase, enkephalinase and ACE, respectively. These inhibitors prolong the
duration of action of endogenously released enkephalins and it is therefore hoped that they
are free of the side effects produced by narcotic analgesics. The clinical use of these
substances may however be limited due to their limited bioavailability.
Another approach which has been used to increase both the stability and the
selectivity of opioid peptides is the introduction of synthetic modifications to the molecule
[Shimohigashi 1986], For example, stability can be increased by substituting the

12
corresponding D-amino acid for the naturally occurring L-amino acid in the peptide
molecule. As peptides are conformationally labile, the selectivity of opioid peptides has
been increased by stabilizing the peptide molecule in a conformation which prefers the
desired receptor. This can be achieved through the incorporation of conformational
restrictions such as the introduction of unnatural bulky synthetic amino acids (e g./
penicillamine residues) or the cyclization of the peptide chain. Highly selective opioid
peptide analogs such as [D-Pen2, D-Pen5]-enkephalin which is selective for the delta
receptor [Mosberg et al. 1983] and [Tyr-D-Ala-Gly-MePhe-NH(CH2)20H] which is
selective for the mu receptor [Handa et al. 1981] have been developed using these
techniques.
Further research involving opioid peptides and their receptors is of importance in
elucidating the precise physiological roles of these entities, particularly in the areas of pain
and immunomodulatory pathways.
Rationale
As discussed above, a considerable body research has been focused on the
development of opioid peptides as therapeutic agents. The low physiological
concentrations of endogenous opioid peptides and those to be expected for therapeutically
administered derivatives necessitate the development of specific and ultra-sensitive
analytical methods, in the fmol per ml range, for these entities in biological fluids to
support clinical studies. The need for new analytical approaches for the measurement of
opioid peptides has been stressed by the National Institute on Drug Abuse [Rapaka 1986],

13
Current analytical methods for opioid peptides are summarized in Table 1.2 and are
reviewed in the following pages.
Immunological Methods
At present, for the most part, opioid peptides are analyzed by radioimmunoassay
(RIA) [Sato 1984, Venn 1987], Although immunoassays are highly sensitive and
reproducible, they are hampered by several disadvantages. Firstly, an antiserum to the
analyte of interest has to be raised, which involves lengthy incubation times. Secondly,
immunoassays suffer from low selectivity due to the cross-reactivity of the antiserum to
structurally similar compounds. To overcome the problem of cross-reactivity, high
performance liquid chromatography (HPLC)-immunoassay methods have been developed
whereby collected eluted fractions from an HPLC system are analyzed by RIA [Defrutos
and Regnier 1993, McDermott et al. 1981], However, HPLC-immunoassay methods are
time consuming, labor intensive and chromatographic solutions must be volatile or
compatible with the immunoassay. Radioimmunoassays also have the added disadvantage
that the use of radioactive isotopes as tracers calls for special considerations in the
handling of the assays and the disposal of the radioactive waste produced.
Non-radiation immunological techniques have not been exploited to a great extent
for opioid peptides. An enzyme-linked immunosorbent assay (ELISA) has been described
for LE and ME [Zamboni et al. 1983] but its limits of detection are only in the 1 pmol per
assay range. Only four enzyme immunoassays with high sensitivity have been developed
for [3-endorphin and dynorphin [Hochhaus and Hu 1990, Hochhaus and Sadee 1988,

14
Kuhling et al. 1989, Sarma et al. 1986], These assays have sensitivities ranging from 0.3 to
3.2 femtomole per assay.
The immunoassays developed by Hochhaus and Sadee [Hochhaus and Sadee
1988] and Hochhaus and Hu [Hochhaus and Hu 1990] are ELlSAs based on the avidin-
biotin system whereby the peptide of interest in the sample or standard and its biotinylated
derivative compete for antibody binding sites. The antibody-bound biotinylated species is
subsequently detected by enzymatic detection through the use of an avidin-enzyme
complex. For this dissertation, an attempt was made at the development of an extremely
sensitive avidin-biotin based ELISA for enkephalins, using LE as a model peptide, to
compliment the existing ELISA tests for P-endorphin and dynorphin. If this type of assay
can be developed for LE, it can be expected to be also applicable to ME.
In addition, for this dissertation, an attempt was also made at the development of
an homogenous or non-separation immunoassay for opioid peptides. Homogeneous
immunoassays differ from traditional immunoassays in that the labor intensive separation
of the bound and free fraction of the analyte (e g./ by washing, precipitation or
adsorbance) is not necessary prior to quantitation as the property being measured is
characteristic of either the bound or the free analyte or label.
The immunological techniques proposed in this dissertation have the disadvantage
of low specificity, but they were intended for use as "immunological HPLC detectors" in
the hope that they would provide fast, ultra-sensitive assays which were highly suitable for
processing large numbers of samples such as HPLC fractions.

15
Table 1.2. Summary of analytical methods for opioid peptides.
Reference
Method
Sensitivity
Analyte
Comments
Fleming and
Reynolds
1988
HPLC-ED
1 pmol/inj
enkephalin
tedious sample clean up,
high oxidation potential
Kim et al.
1989
HPLC-ED
1 pmol/inj
enkephalin
tedious sample clean up,
high oxidation potential
Shibanoki et
al. 1990
HPLC-ED
550 fmol/inj
enkephalin
tedious sample clean up,
high oxidation potential
Monger and
Olliif 1992
HPLC-ED
75 fmol/ ml
plasma
(3-endorphin
tedious sample clean up,
high oxidation potential
Muck and
Henion 1989
HPLC-MS
100 fmol/inj
dynorphin
microbore LC system
Mifune et al.
1989
HPLC-FL
100 fmol/inj
enkephalin
complicated column
switching
Nakano et al.
1987
HPLC-FL
140 fmol/inj
enkephalin
harsh reaction conditions
Kai et al.
1988
HPLC-FL
500 fmol/inj
enkephalin
harsh reaction conditions
van den Beld
et al. 1990
HPLC-FL
50 fmol/inj
P-endorphin
laser induced
fluorescence
Dave et al.
1992
HPLC-FL
36 fmol/inj
LE
microbore LC system
Hochhaus
and Sadee
1988
ELISA
<1 fmol/assay
P-endorphin
Hochhaus
and Hu 1990
ELISA
1 fmol/assay
dynorphin
Kuhling et al.
1989
ELISA
3 fmol/assay
P-endorphin
de Ceballos
et al. 1991
HPLC-R1A
1.5 fmol/assay
LE, ME
Maidment et
al. 1989
RIA
<1 fmol/assay
ME
HPLC-ED = high performance liquid chromatography with electrochemical detection, HPLC-FL = high
performance liquid chromatography with fluorescence detection, HPLC-MS = high performance liquid
chromatography with mass spectrometry, ELISA = enzyme-linked immunosorbent assay.

16
Instrumental Approaches
Other methods which have been used in the analysis of opioid peptides include
HPLC combined with electrochemical detection (HPLC-ED), fluorescence detection
(HPLC-FL) or mass spectrometry (HPLC-MS).
Electrochemical detection
HPLC in conjunction with electrochemical detection is an analytical method which
offers several advantages. It provides selectivity, as only those compounds which are
oxidizable or reducible at the applied potential will be detected. Multiple electrode
detectors can be used to pre-oxidize contaminants in the mobile phase and the sample
prior to detection of the analyte of interest, thus increasing sensitivity by improving signal
to noise ratios. The method is versatile and the cost of instrumentation and reagents is
relatively low. In electrochemical detection, peak current ratios are obtained from the ratio
of the peak heights obtained (i.e. current generated) when two different voltages are
applied to the same amount of sample. These peak current ratios are characteristic for
each compound, much like absorbance ratios in ultraviolet detection, and therefore
qualitative information about the analyte can be derived from them.
A review of the literature reveals that, to date, HPLC-ED methods for opioid
peptides are less sensitive than other methods such as RIA [Fleming and Reynolds 1988,
Kim et al. 1989, Mousa and Couri 1983], This may be attributed to the fact that the high
potentials necessary to detect opioid peptides using these methods (+0.9-1.25 V compared
to +0.3-0.4 V used in HPLC-ED assays for catechols with sensitivities in the fmol/inj
range [Higa et al 1977, Koike et al. 1982]) compromise sensitivity as background current

17
and baseline noise are increased significantly. Selectivity is also compromised as more
compounds are oxidized at these high potentials, thus necessitating extensive sample clean
up.
In the HPLC-ED assay for LE described in this dissertation, the enzymatic
derivatization used increased HPLC-ED selectivity and sensitivity by pre-column o-
hydroxylation of the highly conserved N-terminal tyrosine groups of the peptide resulting
in easily oxidizable derivatives.
Fluorescence detection
Pre-column fluorescence derivatization of analytes can be achieved by fluorophoric
labeling using a fluorescent precursor, or by fluorogenic derivatization using a non-
fluorescent precursor. Fluorogenic derivatization is usually preferred as fluorophoric
labeling often requires excess fluorescent reagent and subsequent extensive clean up to
minimize background interference.
Fluorogenic derivatization reactions have been carried out by using fluorogenic
reagents such as o-phthalaldehyde [Roth 1971], fluorescamine [Udenfriend et al. 1972]
and naphthalene-2,3-dialdehyde in the presence of cyanide [Lunte and Wong 1989, Mifune
et al. 1989], However, these procedures derivatize N-terminal amino groups and
consequently are not specific for any particular peptide so that subsequent
chromatographic procedures are often extremely complex (e.g./ multidimensional HPLC
systems using column switching) to allow for the selective determination of opioid
peptides as the derivatives of other peptides may interfere with the signal.

18
A derivatization reaction which is specific for tyrosine groups such as the one
described in this dissertation could be expected to increase selectivity in the determination
of opioid peptides as tyrosine is highly conserved in the 1 position of these compounds but
is often missing in unrelated peptides. Tyrosine specific HPLC methods with fluorescence
detection such as those using l,2-diamino-4,5-dimethoxybenzene [Ishida et al. 1986, Kai
et al. 1988] and hydroxylamine, cobalt (II) ion and borate [Nakano et al. 1987, Zhang et
al. 1991] do exist for peptide analysis. Although these methods have allowed the
determination of opioid peptides with sensitivities of 100-500 femtomole per injection, the
harsh reaction conditions of these derivatizations are expected to lead to diminished
recoveries and reduced assay reproducibility of the fragile peptide analytes.
Mass spectrometry
HPLC with mass spectrometry offers the highest level of molecular specificity
compared to other analytical methods and is the only method with which unambiguous
confirmation of the structure of the target peptide can be achieved. However, MS is costly
and requires specialized instrumentation and therefore, it is an impractical method for the
average analytical laboratory. The analysis of peptides by mass spectrometry has been
facilitated in recent years by the advent of several soft ionization and sample
introduction techniques such as fast atom bombardment, matrix-assisted laser desorption,
electrospray and ionspray mass spectrometry (see Arnott et al. 1993, Biemann 1992 and
Carr 1990 for reviews) which allow the production of intact molecular ions from these
fragile species. Another advantage of these techniques is that they are often compatible
with on-line microbore HPLC. A microbore HPLC-ionspray MS method has allowed the

19
determination of enkephalins in the 100 fmol/inj range [Muck and Henion 1989], Levels as
low as 5 fmol of P-endorphin and 1 pinol of ME have been quantified by electrospray
mass spectrometry with off-line HPLC [Dass and Kusmierz 1991] and pmol amounts of
ME have been detected by fast atom bombardment mass spectrometry with off-line HPLC
[Kusmierz and Sumrada 1990],
Objectives
Appropriate analytical methodology for opioid peptides is required for use in
clinical, pharmacokinetic and formulation studies as well as in physiological studies to
allow the successful development of these entities as therapeutic agents and the
continuation of research to elucidate further the physiological role of these compounds.
The need for accurate, specific, sensitive and reproducible analytical methods for opioid
peptides has been stressed by representatives of the National Institute on Drug Abuse
[Rapaka 1986], However, to date no analytical procedure has emerged which adequately
meets all of these criteria.
Therefore, the focus of the work carried out for this dissertation has been the
evaluation of several novel analytical approaches for the determination of leucine
enkephalin (LE, Figure 1.5) as a model for opioid peptides. Leucine enkephalin was
chosen as a model peptide as it contains the same initial sequence common to all opioid
peptides.
The first approach evaluated was a non-homogeneous enzyme-linked
immunosorbent assay (ELISA) which employed the same strategy as ELISAs which have

20
been previously developed for dynorphin [Hochhaus and Hu 1990] and P-endorphin
[Hochhaus and Sadee 1988], This assay was based on the avidin-biotin system whereby
avidin exhibits an extremely high affinity for biotin and biotinylated species. Here, LE and
biotinylated LE derivative compete for antibody binding sites and subsequently, on
separation of the antibody-bound and free fractions, the antibody-bound biotinylated
species is detected through the use of an avidin-enzyme complex. The success of this type
of assay depends on the formation of a sandwich between the antibody, the biotinylated
LE derivative and enzyme-labeled avidin.
OH
CH, v CH,
3\ / 3
CH
I
O
O CH,
/\/\/\/2\/\/ \ / \
r rH2 NH C CH NH C
O
O CH,
O
Q
Tyr Gly
Figure 1.5. Leucine enkephalin.
Gly
Phe
Leu
The second approach was an homogeneous or separation-free fluorescence
immunoassay which also made use of the avidin-biotin system, and was based on an
observation by Al-Hakiem and co-workers [A1 Hakiem et al. 1981] that the binding of

21
biotin to fluorescence-labeled avidin produced an increase in fluorescence intensity. In
contrast to the ELISA described above, this approach relied on a lack of sandwich
formation between antibody, biotinylated LE derivative and fluorescence-labeled avidin, as
the success of this approach depended on the inability of antibody-bound biotinylated LE
to interact with fluorescence-labeled avidin and produce an increase in fluorescence
intensity. Therefore, here LE and biotinylated LE compete for antibody binding sites and
after equilibrium has been achieved, free biotinylated LE is detected through the increase
in fluorescence intensity produced on addition of fluorescence-labeled avidin.
Two high-performance liquid chromatography (HPLC) assays were also evaluated,
both making use of tyrosine-specific pre-column derivatization using the enzyme
tyrosinase. In the high performance liquid chromatography assay with electrochemical
detection (HPLC-ED), specific hydroxylation of the tyrosine group in the 1 position of
LE, which is highly conserved in all opioid peptides, presents two analytical advantages.
Firstly, the derivatization results in the formation of a catechol which is amenable to
specific clean-up using boronate gels, and secondly, this catechol is more easily oxidizable
than the parent peptide, thus facilitating electrochemical detection. In the high
performance liquid chromatography assay with fluorescence detection (HPLC-FL),
enzymatic derivatization by tyrosinase renders our peptide amenable to fluorogenic
derivatization using 1,2-diamino- 1,2-diphenylethane.
In addition, having established in the enzymatic derivatization for the HPLC assays
that products are formed from the reaction between LE and tyrosinase, the identity of
these products was determined by mass spectrometry. Furthermore, since tyrosinase and

22
enkephalins have been found to co-exist in vivo [Merchenthaler 1993, Sesack and Pickel
1992, Zhuo et al. 1992], the biological activity in rat brain homogenate of the major
product of the reaction between these two entities was investigated.

CHAPTER 2
APPROACHES TO THE DEVELOPMENT OF AN IMMUNOASSAY FOR
LEUCINE ENKEPHALIN
Introduction
Immunoassay is an analytical method which exploits the binding of a ligand
(antigen) to specific sites on an antibody. In most cases (e g./ for radioimmunoassays),
labeled ligand competes with unlabeled ligand (analyte) for antibody sites and the extent of
binding of the labeled ligand is determined through the measurement of some physical or
chemical property associated with the label. A standard calibration curve of the signal
produced by the label with respect to the concentration of analyte present can then be
constructed, thus allowing the estimation of unknown ligand concentrations from this
curve.
Traditionally, immunoassays have involved the use of radioisotopes as labels.
Although radioimmunoassays (RIA) have the advantages of being highly sensitive and
invulnerable to environmental interferences (e g./ by components of the assay), their use is
accompanied by several disadvantages. These disadvantages include the emotive bias
against the use of radioisotopes, the costly disposal of waste, special requirements for
assay handling and training of staff, the lack of stability of radioisotopes and the
consequent limited shelf-life of reagents [Gosling 1990], Therefore, there has been
considerable interest in developing non-isotopic labels, such as enzyme labels and
23

24
luminescent labels for use in immunoassay [Kricka 1993, Porstmann and Kiessig 1992,
Schulman et al. 1990],
Immunoassays can be divided into those methods which require separation of the
antibody-bound and free fractions prior to quantitation, and those in which the signal
being measured is a function of antibody binding and therefore do not require separation
of antibody-bound and free fractions prior to quantitation. These assays are referred to as
heterogeneous and homogeneous immunoassays, respectively. The development of
homogeneous immunoassays was prompted by the fact that the separation step in
heterogeneous immunoassays is labor intensive, complicates automation and introduces
inaccuracy and error into the assay as the equilibrium which exists between the bound and
free fraction in the sample is disturbed. Enzyme- and luminescence- labeling techniques are
used in the majority of the homogeneous immunoassays developed thus far [Coty et al.
1992, Garcia et al. 1993, Jenkins 1992]; however, other approaches involving the use of
liposomes [Bowden et al. 1986, Ho and Huang 1985, Umeda et al. 1986], bilayer
membranes [Ihara et al. 1988] and reversed micellar systems [Kabanov et al. 1989] have
also been investigated. Although homogeneous immunoassays are convenient and
relatively simple to perform, to date, they have suffered from lack of sensitivity compared
to heterogeneous immunoassays, due in particular to interference with endpoint
determination by components of biological samples [Jenkins 1992],
As mentioned in Chapter 1, one of the major disadvantages of immunoassays in
general is their relative lack of specificity due to cross-reactivity of the antibody in use to
structurally similar compounds. In order to circumvent this problem, analytical methods

25
have been developed whereby structurally similar analytes are first separated by high-
performance liquid chromatography (HPLC) prior to quantitation by immunoassay. To
this end, the goal in the development of the immunoassays proposed in this dissertation
was the design of rapid and convenient assays for opioid peptides with detector-like
properties, intended for the determination of opioid peptide concentrations in HPLC
fractions. These assays should therefore allow the sensitive and straightforward analysis of
opioid peptides, be practicable on a large scale and easily amenable to intensive
automation.
Two different approaches to the development of an immunoassay for opioid
peptides were conceived, using leucine enkephalin (LE) as a model peptide. Both of these
approaches were based on the exploitation of the extremely high affinity exhibited by
avidin for biotin (Ka=10151/mol). Each avidin molecule has four high affinity binding sites
for biotin so that the central strategy of these approaches lies in the knowledge that when
avidin is conjugated to an enzyme or fluorescent label, it will still bind to biotin or
biotinylated species [Wilchek and Bayer 1988],
The first approach, an enzyme-linked immunosorbent assay (ELISA) for LE is
similar to assays previously developed for dynorphin [Hochhaus and Hu 1990] and P-
endorphin [Hochhaus and Sadee 1988] which have sensitivities in the lower fmol/assay
range. The development of a similar assay for LE would have provided a battery of
enzyme immunoassays for opioid peptides based on identical principles. This assay
involves competition between LE in the sample or standard and biotinylated LE derivative
for immobilized antibody binding sites. The antibody-bound biotinylated species is

26
subsequently detected through the use of an avidin-enzyme complex (Figure 2.1). The
successful design of this type of assay for LE depends on the synthesis of a suitable
biotinylated LE derivative that will allow the formation of a sandwich between the
antibody, the biotinylated LE derivative and the enzyme-labeled avidin, thus enabling
quantification upon addition of the enzyme substrate.
Substrate Detection
\ Antibody
Biotinylated analyte
<^|||| Analyte
JL/E
nr
Enzyme-labelled avidin
Figure 2.1. Avidin-biotin based ELISA
The second approach, which is an homogeneous or separation-free fluorescence
immunoassay is based on an observation by Al-Hakiem et al. [Al-Hakiem et al. 1981] that
the binding of biotin to fluorescein-labeled avidin leads to an increase in fluorescence
intensity. This phenomenon is exploited in an attempt to develop a new class of
homogeneous immunoassay which depends on the inability of antibody-bound biotinylated

27
LE to interact with fluorescein-labeled avidin. Therefore, in this assay, biotinylated LE
and LE in the sample or standard compete for antibody binding sites and after equilibrium
has been achieved, fluorescence-labeled avidin, which acts as a detector molecule, is
added. Unbound biotinylated LE is determined by the increase of fluorescence intensity
due to the interaction of the unbound biotinylated LE and the fluorescence-labeled avidin
(Figure 2.2). In contrast to the ELISA described above, the success of this assay depends
on the synthesis of a biotinylated LE derivative which will not allow fluorescence-labeled
avidin to bind to antibody-bound biotinylated LE (i.e. lack of "sandwich" formation).
+
+
Increased fluorescence intensity
Antibody
Biotinylated analyte
Analyte
Fluorescence-labelled avidm
nr
Figure 2.2. Avidin-biotin-based homogeneous fluorescence immunoassay

28
Materials
Leucine enkephalin, leucine enkephalin-Lys6, Freunds complete adjuvant,
Freunds incomplete adjuvant, bovine serum albumin, porcine thyroglobulin, glutar-
aldehyde, l-ethyl-3-(3-dimethylaminopropyl)carbodiimide, sodium acetate, activated
charcoal, dextran (average molecular weight 70,800), avidin, N-hydroxysuccinimidobiotin,
biotinamidocaproate N-hydroxysuccinimide ester, triethanolamine, 2-4-hydroxyazo-
benzene-benzoic acid, di-methylsulfoxide and fluorescein isothiocyanate avidin were
obtained from Sigma Chemical Company, St. Louis, MO, USA. Methanol, acetonitrile,
trifluoroacetic acid, sodium dihydrogen phosphate, disodium hydrogen phosphate, sodium
chloride, potassium dihydrogen phosphate, dipotassium hydrogen phosphate and
potassium chloride were purchased from Fisher Scientific, Pittsburgh, PA, USA. [tyrosyl-
3,5-?H(N)]-leucine enkephalin was procured from NEN Research Products, Dupont
Company, Wilmington, DE, USA. Cytoscint scintillation cocktail was obtained from ICN
Biomedicals Inc., Irvine, CA, USA and anti-LE antiserum was purchased from Peninsula
Laboratories Inc., Belmont, CA, USA.
Methods
Antibody Production
Leucine enkephalin (LE) was conjugated to either porcine thyroglobulin or bovine
serum albumin (BSA) by reaction with either glutaraldehyde or 1-ethyl-3-(3-dimethyl-
aminopropyl)carbodiimide (EDC).

29
For the glutaraldehyde method (G. Adamus, personal communication), a 25 % v/v
stock solution of glutaraldehyde was freshly prepared on ice and diluted 65-fold in 0.1 M
sodium phosphate buffer pH 7. Equal amounts of LE and protein carrier (BSA or porcine
thyroglobulin) were weighed out and dissolved in 0.1 M sodium phosphate buffer pH 7 to
give a solution containing 1 mg/ml each of LE and protein carrier. One hundred and
twenty four microliters of the final dilution of glutaraldehyde were added for each milliliter
of protein-peptide solution and the reaction was allowed to proceed at room temperature
overnight with constant stirring. The conjugate was then dialyzed against deionized water
for 24 hours using pre-hydrated Spectrapor cellulose ester membranes with molecular
weight cut off of 15,000 (Spectrum Medical Industries Inc., Los Angeles, CA, USA).
After dialysis, aliquots of the conjugates were stored at -20C prior to lyophilization.
For the EDC method [Harlow and Lane 1988], a I mg/ml solution of LE was
prepared in water and EDC was weighed out and added to give a final concentration of 10
mg/ml. The pH of the reaction mixture was adjusted and maintained at pH 5 with 1 N
NaOH throughout the 5 minute incubation time at room temperature. An equal volume of
an 11 mg/ml solution of protein carrier (BSA or porcine thyroglobulin) was added and the
reaction was allowed to proceed at room temperature for 4 hours. The reaction was then
stopped by the addition of a sodium acetate solution (1 M, pH 4.2) to give a final
concentration of 100 mM. After an additional incubation of 1 hour at room temperature,
the conjugate was dialyzed against 0.1 M phosphate buffer pH 7 for 24 hours using pre-
hydrated SpectraPor cellulose ester membranes with molecular weight cut off of 15,000
(Spectrum Medical Industries Inc., Los Angeles, CA, USA). After dialysis, aliquots of the

30
conjugates were stored at -20C prior to lyophilization. Lyophilization of the conjugates
was carried out at the Drug Delivery Laboratory, University of Florida, Progress Center,
Alachua, FL, USA using a Model 12K Super Modulyo Lyophilizer (Edwards, West
Sussex, England).
After lyophilization, samples of LE, BSA, porcine thyroglobulin and the
conjugates were sent for amino acid analysis at the Peptide Core of the Interdisciplinary
Center Biotechnology Research, University of Florida, Gainesville, FL, USA. The loading
ratios of LE to carrier protein were calculated from the percentage composition of
asparagine, threonine, serine and glutamine in the carriers and conjugates. The LE-
thyroglobulin conjugate obtained by the glutaraldehyde method showed the highest
peptide loading ratio (see Results and Discussion) and therefore this conjugate was
selected as the inoculum for the purposes of this work.
For the initial inoculation, complete Freunds adjuvant was placed in a test tube
and an equal volume of 0.1 M sodium phosphate buffer pH 7.4 containing 2 mg/ml of the
LE-thyroglobulin conjugate was added while vortexing, to give a thick emulsion
containing 1 mg/ml of the conjugate. Three young adult male New Zealand White rabbits
each received a total of 1 ml of this emulsion subcutaneously at 6-8 different sites in the
back. Thereafter, booster injections with the same dose of conjugate were given at
monthly intervals in the same manner, except that incomplete Freunds adjuvant was used
instead of complete Freunds adjuvant. Test bleeds were taken 7-10 days after each boost
and after serum separation, the antibody titer was tested by radioimmunoassay. Briefly, to
test total binding, [tyrosyl-3,5-3H(N)]-leucine enkephalin (3H-LE, 1 nM) was incubated

31
overnight at 4C with various dilutions of the antiserum (1/10, 1/100, 1/1,000, 1/10,000)
in 0.1 M sodium phosphate buffer pH 7. To test non-specific binding, ?H-LE (1 nM) was
incubated overnight at 4C with LE (500 nM) and various dilutions of the antiserum
(1/10, 1/100, 1/1,000, 1/10,000) in 0.1 M sodium phosphate buffer pH 7. An ice-cold
suspension containing 1.5 % w/v activated charcoal and 0.15 % w/v dextran in water was
then added to each sample and after a 5 minute incubation at 4 C, the samples were
centrifuged at 12,000 g for 3 minutes. An aliquot of the supernatant was then removed,
Cytoscint scintillation cocktail (4 ml) was added and the radioactivity (CPM) representing
the antibody-bound tracer (3H-LE) was determined using a Beckman LS 5,000 TD
scintillation counter (Fullerton, CA, USA). After the third boost, the antibody titer of one
of the rabbits was deemed sufficient and therefore, this rabbit was exsanguinated, the
serum was separated from whole blood by centrifugation at 5,000g for 10 minutes (Dynac
II Centrifuge, Clay Adams, Sparks, MD, USA) and stored in aliquots at -80C.
A 10 ml aliquot of the anti-LE antiserum was purified by the Hybridoma Core of
the Interdisciplinary Center Biotechnology Research, University of Florida, Gainesville,
FL. Briefly, a column was filled with 10 ml of Prot G-Gammabind Ultra matrix (Pharmacia
LKB Biotechnology Inc., Piscataway, NJ, USA) and washed with 35 ml of elution buffer
(0.1 M glycine pH 3.0). The column was then washed extensively with 200-300 ml of
binding buffer (0.1 M phosphate buffered saline pH 7, 0.01% sodium azide). The
antiserum was diluted two-fold with binding buffer, loaded onto the column and washed
with binding buffer until the optical density of the eluent reached zero at X28o- Elution
buffer was then added to the column, the eluent was collected in 1 ml fractions and

32
neutralized immediately with 35 pi of 2 M Tris buffer pH 9. The fractions with high
optical density at taso were pooled and then desalted and concentrated by spinning down
5-6 times for 30 minutes in a C-PREP 10 centrifugal filter (Amicon, Danvers, MA, USA).
This procedure yielded 8.5 ml of purified anti-LE antibody containing 6.1 mg IgG/ml,
determined by measurement of optical density at taso-
Characterization of Antibody
The number of specific antibody sites in the purified antibody was determined by
means of Scatchard analysis [Scatchard 1949], Samples containing various concentrations
of 3H-LE as tracer (70-1000 pM) and a 1/1,000 dilution of the purified antibody were set
up to determine total binding (Table 2.1). To determine non-specific binding, samples
containing antibody, ?H-LE (70-1000 pM) and a high concentration of LE (1.2* 10"6 M)
were also set up at the same time. After overnight incubation at 4C, 200 pi of an ice cold
suspension containing 1.5% activated charcoal and 0.15% dextran in water was added to
each tube (except total counts) and after a further 5 minute incubation on ice, the samples
were centrifuged at 12,000g for 3 minutes. Three hundred and fifty microliters of the
resultant supernatant containing the antibody-bound fraction of the tracer were then
removed and added to 4 ml of Cytoscint scintillation cocktail. The radioactivity (CPM)
representing the antibody-bound tracer was determined using a Beckman LS 5,000 TD
scintillation counter (Fullerton, CA, USA) and a 5 minute counting time with counting
efficiency at about 50%.

33
Table 2.1. Samples set up for Scatchard analysis.
Total counts
Total binding
Non-specific binding
Buffer
420 pi
20 pi
3H-LE
20 pi
20 pi
20 pi
LE( 1.4*10'5 M)
20 pi
Antibody (1/1,000)
200 pi
200 pi
Overnight at 4C, then:
Charcoal (1.5%)/dextran (0.15%)
200 pi
200 pi
A Scatchard plot was constructed by plotting the ratio of bound to free 3H-LE
against bound 3H-LE and the Ka value was determined from the negative slope of the line
drawn between the points. The number of specific antibody sites was determined from the
intercept of this line with the x-axis.
Biotinylation of Leucine Enkephalin
N-terminal biotinylated LE (BLE) was synthesized by allowing 90 nmoles of LE,
180 nmoles of N-hydroxysuccinimidobiotin (BHS) and 120 nmoles of triethanolamine in
150 pi of dimethylsulfoxide (DMSO) to incubate overnight at room temperature. The
product of the reaction was separated from the reagents by injecting the incubation
mixture into a gradient HPLC system consisting of a Rainin Rabbit HP solvent delivery
system (Rainin Instrument Company Inc., Woburn, MA) controlled by a Rainin Dynamax
HPLC method manager (version 1.3, Rainin Instrument Company Inc., Woburn, MA), a
Negretti & Zamba injector (Southampton, UK) fitted with a 100 pi loop, a pBondapak
Ci8 column (10 pm, 3.9 x 150 mm, Waters Associates, Milford, MA, USA) and a LDC

34
Milton Roy Spectromonitor 3100 variable wavelength detector (Riviera Beach, FL, USA).
The detection wavelength was set at 210 nm. Mobile phase A consisted of 90 % v/v
aqueous trifluoroacetic acid (0.02% v/v) and 10 % v/v acetonitrile and mobile phase B
consisted of 10 % v/v aqueous trifluoroacetic acid (0.02% v/v) and 90 % v/v acetonitrile.
A gradient of 90 % mobile phase A and 10 % mobile phase B to 50 % each of mobile
phases A and B in 30 minutes was run at a flow rate of 1 ml/min. A control injection of the
incubation mixture was also made into the HPLC system immediately after preparation to
allow the calculation of the extent of conversion of LE to BLE. N-terminal biotinylated
LE incorporating a spacer arm between the biotin group and the peptide (BXLE) was
synthesized in the same manner except that biotinamidocaproate N-hydroxysuccinimide
ester (BXHS) was used instead of BHS in the incubation mixture. The peaks
corresponding to BLE and BXLE were collected from the HPLC eluent and the volatile
components (acetonitrile and trifluoroacetic acid) were evaporated under a stream of
nitrogen at 30C. The biotinylated LE derivatives were stored at 4C and were used within
two weeks of synthesis.
C-terminal biotinylated leucine enkephalin-Lys6 (LE-Lys6-B) was synthesized by
allowing 73 nmoles of leucine enkephalin-Lys6 (LE-Lys6), 142 nmoles of BHS and 90
nmoles of triethanolamine in 150 pi of DMSO to incubate for two hours at room
temperature. The product of the reaction was separated from the reagents by injecting this
incubation mixture into the same gradient HPLC system described above except that a
gradient of 90 % mobile phase A and 10 % mobile phase B to 70 % mobile phase A and
30% mobile phase B in 30 minutes was run at a flow rate of 1 ml/min. A control injection

35
of 73 nmoles of LE-Lys6 in 150 pi of DMSO was also made into the HPLC system to
allow the calculation of the extent of conversion of LE-Lys6 to LE-Lys6-B. C-terminal
biotinylated LE incorporating a spacer arm between the biotin group and the peptide (LE-
Lys6-BX) was synthesized in the same manner except that BXHS was used instead of
BHS in the incubation mixture. The peaks corresponding to LE-Lysf-B and LE-Lys6-BX
were collected from the HPLC eluent and stored as described earlier for BLE and BXLE.
The successful biotinylation of LE by the methods described above was
characterized by comparing the displacement of 2-(4-hydroxyazobenzene)-benzoic acid
(HABA) from avidin by the biotinylated LE derivatives and biotin itself according to the
spectrophotometric method of Green [Green 1970], A 0.25 mM solution of HABA was
prepared in 0.1 M sodium phosphate buffer pH 7 and avidin was added to an aliquot of
this solution to give a final concentration of 45 pM. One milliliter of this solution was
placed in a UV quartz cuvette and upon serial additions of the biotinylated species or
biotin as a positive control, the absorbance at A.5oonm was monitored using a Cary 3E UV-
Visible spectrophotometer (Varan, Sugarland, TX, USA). The same experiment was
repeated using each biotinylated LE derivative.
The structures of LE-Lys6-B and LE-Lys6-BX were confirmed by mass
spectrometric analysis by the Protein Chemistry Core of the Interdisciplinary Center for
Biotechnology Research, University of Florida, Gainesville, FL, USA. A Lasertech matrix-
assisted laser desorption (MALDI) time-of-flight (TOF) instrument operated at 10 kV
acceleration was used with a 335 nm UV laser and a-cyano-4-hydroxycinnamic acid as the
matrix. The sample was dissolved in a solution (acetonitrile/water, 1/2 v/v) containing a

36
large excess of the matrix (5g/l) and a 1 pi aliquot was deposited onto the stainless steel
laser target. Several spectra were averaged for each experiment.
Antibody-Binding of LE. Biotinylated LE Derivatives and Preformed Biotinylated
LE-Avidin Complexes
The binding of LE, the biotinylated LE derivatives and preformed biotinylated LE-
avidin complexes to both a commercial anti-LE antiserum (Peninsula) and the antibody
produced in this laboratory was tested using a standard radioimmunoassay procedure. For
the assays using preformed biotinylated LE-avidin complexes, a ten-fold molar excess of
avidin was added to the highest concentration of biotinylated LE in buffer and vortexed
immediately. The complexes were allowed to form for 30 minutes prior to serial dilution
to give the various concentrations of competitor required for the assay. The assays were
set up in microcentrifuge tubes as shown in Table 2.2. The RIA buffer used consisted of
0.1 M sodium phosphate buffer pH 7 containing 0.1% w/v BSA. The tracer was 3H-LE
(2.6* 10'10 M final concentration) and the competitor was various concentrations of either
LE, the biotinylated LE derivatives or preformed biotinylated LE-avidin complexes. After
overnight incubation at 4C, 200 pi of an ice cold suspension containing 1.5% w/v
activated charcoal and 0.15% w/v dextran in water was added to each tube (except total
counts) and after a further 5 minute incubation on ice, the samples were centrifuged at
12,000g for 3 minutes. Three hundred and fifty microliters of the resultant supernatant
containing the antibody-bound fraction of the tracer was then removed and added to 4 ml
of Cytoscint scintillation cocktail. The radioactivity (CPM) representing the antibody-

37
bound tracer was determined using a Beckman LS 5,000 TD scintillation counter
(Fullerton, CA, USA).
Table 2.2. Radioimmunoassay set up
Total counts
Total binding
Sample
RIA Buffer
420 pi
20 pi
Tracer (3.125*1 O'9 M)
20 pi
20 pi
20 pi
Competitor
20 pi
Antibody dilution
200 pi
200 pi
Overnight at 4C, then:
Charcoal (1,5%)/dextran (0.15%)
200 pi
200 pi
The IC5o values (concentration of competitor displacing 50% of bound tracer) for
LE, the biotinylated LE derivatives or preformed biotinylated LE-avidin complexes were
determined using the MINSQ non-linear curve-fitting program (MicroMath Scientific
Software, Salt Lake City, UT, USA). The data were fitted to the following model:
T*CN
CN + /C5"
+ NS
Where: B = CPM in the presence of competitor
T = CPM in the absence of competitor
C = competitor concentration
N = slope factor
NS = CPM under non-specific binding conditions
As far as possible, non-specific binding was determined in the presence of
relatively high concentrations of competitor (2-3 orders of magnitude greater than the

38
IC50 value). When high enough concentrations of competitor could not be achieved to
determine non-specific binding, the non-specific binding value obtained from the LE curve
was used to fit the curve for the other competitors.
Development of the Proposed Homogeneous Fluorescence Immunoassay
Determination of excitation and emission maxima for fluorescence-labeled avidin
Fluorescein isothiocyanate avidin (FITC-avidin) was used as the detector
molecule in this assay. To determine the excitation and emission maxima, 2 ml of a
solution of FITC-avidin (7.5 pmol/ml) were dispensed into a quartz fluorescence cuvette
and scans of the excitation and emission spectra were carried out using a Perkin Elmer
LS-3B fluorescence spectrophotometer (Norwalk, CT, USA). The scans were then
repeated upon addition of BLE (70 pmol/ml) to determine the fluorescence enhancement
produced on interaction of these two reagents.
Determination of reagent concentrations
The results of the binding experiments described above indicated that the antibody
made in this laboratory in combination with the N-terminal biotinylated derivative without
the spacer arm (BLE) should be used in the development of the proposed homogeneous
fluorescence immunoassay for LE (see Results and Discussion). One of the first steps in
the development of this assay lay in the determination of the concentrations of antibody,
BLE and FITC-avidin to be used.
The decision as to the concentration of antibody to be used in the final assay
(3.78* 10'9 M, 1/50 dilution of purified antibody) was made based on practicality since it
was anticipated that a large amount of antibody would be required for the development of

39
the homogeneous fluorescence immunoassay and a limited quantity of antibody, obtained
through the exsanguination of a single rabbit was available.
To ensure low background fluorescence readings, the amount of BLE used in the
final assay needed to be such that close to 100 % could be bound by the antibody present.
When BLE and antibody (Ab) are present in the same solution, the equilibrium established
can be described as:
BLE + Ab ^ BLE-Ab
According to the Law of Mass Action, the following equation can be set up:
[BLE-Ab]
[BLE\[Ab]
Which can be rearranged to give:
[BLE] 1
[BLE Ab] ~ [ Ab]Ka
If 95% of the BLE in the assay is to be bound by the amount of antibody chosen for the
assay (3.78* 10'9 M, 1/50 dilution of purified antibody), the left side of the above equation
can be set to equal A. Since [Ab] = [Ab,otai] [BLE-Ab], by using the Ka value obtained
from the Scatchard plot described earlier (assuming that BLE and LE have the same
affinity for the antibody), the above equation can be solved for [BLE-Ab], [BLE-Ab]
represents 95% of the concentration of BLE to be used in the assay and therefore, the
total amount of BLE to be used in the final assay for 95% to be bound by the chosen
antibody concentration could be calculated.
The concentration of FITC-avidin to be used in the assay was determined by
finding the concentration of this reagent which will give maximum fluorescence

40
enhancement with the chosen concentration of BLE. Various concentrations of FITC-
avidin (200-1000 fmol/ml) in 0.1 M sodium phosphate buffer pH 7 were allowed to
interact with various concentrations of BLE (0.1-20 pmol/ml) for 20 minutes, after which
time, fluorescence readings were taken at Xexc 482 nm and Xem 517 nm using a Perkin
Elmer LS-3B fluorescence spectrophotometer (Norwalk, CT, USA).
Homogeneous fluorescence immunoassay
Once the concentrations of the reagents to be used were determined, the complete
homogeneous fluorescence immunoassay samples were set up as shown in Table 2.3.
Total tracer samples (4 pinol BLE per ml of buffer) were also prepared to give an
indication of the maximum fluorescence enhancement which could be expected due to the
interaction of the total amount of tracer present with FITC-avidin. Fluorescence blank
samples were set up to determine the fluorescence due to the native fluorescence of FITC-
avidin present in each sample. The buffer used consisted of 0.1 M sodium phosphate
buffer pH 7 containing 0.1% w/v BSA. After overnight incubation at 4C, 20 pi of FITC-
avidin (13 pmol/ml) were added to each sample to give a final concentration of
500 fmol/ml. The interaction between the free tracer (BLE) and the detector molecule
Table 2.3. Homogeneous fluorescence immunoassay set-up.
Total tracer
Total binding
Sample
FI blank
Buffer
450 pi
350 pi
various
500 pi
BLE (40 pmol/ml)
50 pi
50 pi
50 pi
LE
various
Purified antibody (1/10)
100 pi
100 pi
Total volume
500 pi
500 pi
500 pi
500 pi

41
(FITC-avidin) was allowed to proceed for 20 minutes at room temperature in the dark,
after which fluorescence readings were taken at 482 nm and 517 nm using a
Perkin Elmer LS-3B fluorescence spectrophotometer (Norwalk, CT, USA).
Results and Discussion
Antibody Production
It was anticipated that the development of the homogenous fluorescence
immunoassay in particular would require large quantities of antibody. Therefore, it was
decided to produce a polyclonal antibody to leucine enkephalin in this laboratory.
H H
RNH2 + 0=C(CH2)3CO + h2nR2
- 2H20
T
H H
R,N=(CH2>3=NR2
Figure 2.3. Coupling of peptide to protein carrier using glutaraldehyde. Ri= peptide or
protein carrier, R2= peptide or protein carrier.
As LE is not antigenic itself, the peptide had to be conjugated to a protein carrier
in order to stimulate an immune reaction in the host animal. Glutaraldehyde is a
bilunctional coupling reagent that binds two compounds primarily through their amino
groups. Since LE contains only one amino group at its N-terminal end, using the
glutaraldehyde method of conjugation, the peptide is conjugated to the protein carrier via

42
the N-terminal end (Figure 2.3). Therefore, on inoculation with such a conjugate, one
would expect the antibody that is produced to be directed against the C-terminal end of
the peptide.
c=o + ch3ch2n=c=n
I
OH
H+
(CH2)3-N-CH3
ch3
T*
c=o
/
O H+
I I
ch3ch2-nh-c=n(CH2)3n -ch3
ch3
+ h2n-r2
T-
c=o
I
NH
I
r2
H+
I
+ ch3ch2nh-c-nh(CH2)3n-ch3
0 ch3
Figure 2.4. Coupling of peptide to protein carrier using EDC. Ri=peptide or protein
carrier, R2=peptide or protein carrier.
In contrast, EDC allows the conjugation of the peptide to the protein carrier via
either the N-terminal or the C-terminal end as carbodiimides attack carboxylic groups to
change them into reactive sites for free amino groups. Therefore, the carboxylic groups on

43
the protein carrier will conjugate to LE via the N-terminal amino group of the peptide and
LE will also conjugate to the protein carrier through its C-terminal end by attacking the
free amino groups on the protein carrier (Figure 2.4). As a result, on inoculation with a
conjugate produced by the EDC method, one could expect antibodies directed against
either or both the N-terminal or the C-terminal end of LE.
Table 2.4. Loading ratios of LE to protein carrier by different conjugation methods.
Protein carrier
Conjugation method
LE : protein ratio
Porcine thyroglobulin
Glutaraldehyde
545 : 1
Porcine thyroglobulin
EDC
70 : 1
Bovine serum albumin
Glutaraldehyde
60 : 1
Bovine serum albumin
EDC
1.37 : 1
The loading ratios of peptide to protein carrier calculated from the results of amino
acid analysis are shown in Table 2.4. Due to the high immunogenicity of porcine
thyroglobulin in rabbits and the fact that the LE-thyroglobulin conjugate obtained by the
glutaraldehyde method showed the highest peptide loading ratio, this conjugate was
selected as the inoculum for the purposes of this work.
The results of the radioimmunoassay carried out after the third boost showed that
when a 1/1,000 dilution of the antiserum was used, 30% of the total radioactivity added
was bound and 32% of this bound radioactivity was determined to be due to non-specific
binding (Figure 2.5). At that time, this titer was deemed to be sufficient and therefore the
antiserum was harvested.

44
Figure 2.5. Bar graph showing total counts, total binding and non-specific binding of
3FLLE using 1/1,000 dilution of antibody produced in this laboratory.
Characterization of Antibody
The Scatchard plot of bound/free 3H-LE versus bound 3H-LE is shown in
Figure 2.6. The Ka value determined from the negative slope of the line drawn between the
points was 3.87* 1091/mol which converts to a Ka value of 2.58* 1010 mol/1. The number
of specific antibody sites present in the final incubation for this experiment was determined
from the x-axis intercept to be 3.62*10'10 mol/1. Since 3.18*1 O'8 mol/1 of IgG was used in
this preparation, the purified antibody contains only 1.14% specific antibody sites. By
extrapolation, the stock solution of purified antibody was determined to contain 4.35* 107
mol/1 of specific antibody sites. The affinity of the antibody made in this laboratory to LE
compared favorably to the commercial antiserum obtained from Peninsula under the same
assay conditions since 1C50 values for LE using a 1/1,000 dilution of the antibody

45
produced in this laboratory (3.8* 10'13 mol/assay) was similar to those obtained using a
1/3,000 dilution of the commercial antiserum (1.3* 101, mol/assay). Peninsula claimed that
IC50 values of 1.4*1 O'14 mol/assay could be achieved using a 1/15,000 dilution of their
antiserum. However, their assay conditions involved the use of a tracer labeled with l25I
instead of 3H. 125I is a higher energy label than 3H, and therefore, Peninsula was able to use
a lower concentration of tracer in their assay, which allowed them to use less antiserum,
resulting in a lower IC50 value.
Figure 2.6. Scatchard plot of bound/free 3H-LE versus bound 3H-LE
Biotinylation of Leucine Enkephalin
Since one of the proposed immunoassays required the formation of a sandwich
between the anti-LE antibody, the biotinylated LE derivative and fluorescence- or enzyme-
labeled avidin, and the other did not, several different biotinylated derivatives were

46
synthesized and tested. Some of the biotinylated derivatives included a spacer arm
between the biotin group and the peptide (Figure 2.8) as it was felt that this would
increase the likelihood of the formation of a sandwich between the anti-LE antibody, the
biotinylated LE derivative and enzyme-labeled avidin by reducing steric hindrance.
Biotinylation took place at the N-terminal end of the peptide when LE was used as a
starting material as the only free amino group available for the reaction is the N-terminal
amino group of the peptide. When LE-Lys6 was used as a starting material, biotinylation
took place primarily at the C-terminal end of the peptide as here, the amino group of the
lysine moiety is more reactive than the N-terminal amino group under the reaction
conditions used due to its greater basicity.
PeptideNH2 +
H O
I II
PeptideN C(CH,)4
+
Figure 2.7. Biotinylation of peptide using N-hydroxysuccinimidobiotin.

47
Figure 2.8. Biotinylation of peptide using biotinamidocaproate N-hydroxysuccinimide
ester.
The reactants and products of the biotinylation reactions were successfully
separated from each other using the HPLC systems described previously (Figures 2.9 and
2.10). In control injections containing LE or LE-Lys6 in concentrations equal to the initial
concentrations of these compounds in the reaction mixture, the reactants were seen to
elute after 9 minutes (Figure 2.9A) and 5.15 minutes (Figure 2.10A), respectively. The N-
terminal biotinylated LE derivatives, BLE (no spacer arm) and BXLE (with spacer arm)
were seen to elute distinct from the reactants at 15.3 and 16.3 minutes, respectively
(Figures 2.9B and 2.9C). For the C-terminal biotinylation of LE-Lys6 to give LE-Lys6-B
(no spacer arm) and LE-Lys6-BX (with spacer arm), two product peaks were seen on
injection of the incubation mixtures. Based on the delayed formation in the time course of
the reaction of the minor peak eluting at 24.8 minutes in Figure 2.10B and at 27.8 minutes
in Figure 2.10C, these minor peaks were assumed to be di-biotinylated products
biotinylated at both the N- and C-terminal ends of the peptide. Analysis by mass spectro-

48
Figure 2.9. Chromatographs of control injection (A) showing LE eluting after 9 minutes
and reaction injections showing BLE (no spacer arm) eluting after 15.3
minutes (B) and BXLE (with spacer arm) eluting after 16.3 minutes (C).
metry (MALDI) of the major peak eluting at 14.9 minutes in Figure 2.10B and at 17.8
minutes in Figure 2.IOC revealed molecular ion peaks ([M+H]+) at m/z 910.283 and
1022.1, respectively. The nominal molecular masses of the molecular ions of LE-Lys6-B
and LE-Lys6-BX were calculated to be 910.11 and 1023.27, respectively. Since the mass
determinations deviated from the nominal masses of mono-biotinylated derivatives by less
than 0.1%, these products were confirmed as mono-biotinylated derivatives of LE-Lys6.
Therefore, antibody binding experiments were carried out using these collected peaks. The

49
Figure 2.10. Chromatographs of control injection (A) showing LE-Lys6 eluting after 5.15
minutes and reaction injections showing LE-Lys6-B (no spacer arm) eluting
after 14.9 minutes (B) and LE-Lys6-BX (with spacer arm) eluting after 17.8
minutes (C). Di-biotinylated products are seen eluting at 24.8 minutes (B)
and 27.8 minutes (C).
extent of conversion from the starting materials (LE and LE-Lys6) to the biotinylated
derivatives was estimated from the difference in peak height of LE and LE-Lys6 in the
control injection and in the injection of the reaction mixture. An extent of conversion of
approximately 80% from the starting materials to the biotinylated derivatives was
estimated. The results of the HABA displacement experiments confirmed that the products
collected from the HPLC eluent were indeed biotinylated LE derivatives as they were seen
to displace HABA from avidin (Figure 2.11). The results of this experiment confirm that

50
Figure 2.11. Displacement of HABA from avidin by biotin and biotinylated LE derivatives.
on conjugation to LE, biotin retains the ability to bind to avidin. The curvilinear nature of
the displacement curves for BLE, BXLE and LE-Lys-BX together with the observation
that more of these biotinylated derivatives compared to biotin is required to displace the
same amount of HABA from avidin indicates that these derivatives have a slightly lower
affinity for avidin than biotin itself. However, here, LE-Lys6-B is seems to have a slightly
higher affinity to avidin than biotin. This effect can probably be attributed to experimental
error as the concentrations of the biotinylated derivatives were calculated from estimated
conversion rates based on the reduction of the peak heights of the starting materials in the
biotinylation reaction. Previous investigations on the binding of biotinylated peptides or
other macromolecules to avidin have shown that a spacer arm may be required to retain
full affinity to avidin [Finn et al. 1984, Green et al. 1971], however this observation was
not reflected in the results seen here as LE biotinylated with and without the spacer arm
both showed similar affinity to avidin and in fact, LE-Lys6 biotinylated without the spacer

51
arm showed a higher affinity to avidin than LE-Lys6 biotinylated with the spacer arm
(Figure 2.11). This may be attributed to the fact that LE and LE-Lysc are rather small and
therefore, on conjugation to biotin, they do not cause steric hindrance to the binding of
biotin to avidin.
Antibody-Binding of LE. Biotinylated LE Derivatives and Preformed Biotinylated
LE-Avidin Complexes
Using a 1/3,000 dilution of the commercial antiserum, binding curves were
constructed for LE, the N-terminal biotinylated derivatives, the C-terminal biotinylated
derivatives and the corresponding pre-formed complexes with avidin. The IC50 values
obtained for LE, BLE, BXLE, BLE-avidin and BXLE-avidin are shown in Table 2.5.
Representative binding curves for LE, BLE, BXLE, BLE-avidin and BXLE-avidin are
shown in Figure 2.12.
Table 2.5. IC50 values obtained for LE, BLE, BXLE, BLE-avidin and BXLE-avidin using
1/3,000 dilution of commercial antiserum.
Competitor
IC50 (mol/assay)
Mean
Standard deviation
LE
1.15* 10'13
1.04* 10'13
1.58*1 O'13
1.3*10'13
2.7*10'14
BLE
4.44* 10'14
3.56* 10"13
2.0* 10'13
BXLE
4.42* 1015
2.26* 10'14
1.3*1 O'14
BLE-avidin
1.89*10'10
BXLE-avidin
1.46*10'

52
The C-terminal biotinylated LE derivatives LE-Lys6-B and LE-Lys6-BX showed
no binding to the commercial antiserum in the concentrations used (data not shown) and
therefore, using this preparation, they were unsuitable for the development of either of the
immunoassays proposed in this study. It is possible that the inclusion of a longer spacer
arm in the C-terminal biotinylated derivatives would have allowed binding of the
commercial antiserum if the epitope to be recognized by the antibody remained intact on
biotinylation. However, this avenue was not explored.
Figure 2.12. Representatives binding curves for LE (), BLE (), BXLE (), BLE-
avidin () and BXLE-avidin (0) using 1/3,000 dilution of commercial
antiserum.
In these experiments, BXLE showed a ten-fold higher affinity to the commercial
antiserum than LE (Table 2.5). It was postulated that since LE needs to be conjugated to a

53
protein carrier to render it immunogenic for antibody production, this biotinylated LE
derivative including a spacer arm may resemble the epitope presented to the immune
system more closely than LE itself and therefore, BXLE shows a higher affinity to the
commercial antiserum than LE itself.
Although the N-terminal biotinylated derivatives were seen to retain affinity for the
commercial antiserum, a shift in affinity by 2 or 3 orders of magnitude was seen when
complexes were formed between BLE or BXLE and avidin (Table 2.5), indicating that a
sandwich was not formed between the commercial antiserum, BLE or BXLE and avidin.
Therefore, this commercial antiserum was deemed to be unsuitable for the development of
the proposed ELISA for which sandwich formation is a requirement. Although this lack of
sandwich formation indicates that this preparation is suitable for the development of the
proposed homogeneous fluorescence immunoassay, as it was anticipated that large
quantities of antibody would be required for the development of this assay, this avenue
was unfeasible. As a result, efforts involving the development of either an ELISA or an
homogeneous fluorescence immunoassay using the commercial antiserum preparation
were abandoned.
Using a 1/1,000 dilution of the antibody produced in this laboratory, binding
curves were also constructed for LE, the N-terminal biotinylated LE derivatives, the C-
terminal biotinylated LE derivatives and the corresponding pre-formed complexes with
avidin. The IC5o values obtained for LE, BLE, BXLE and BXLE-avidin are shown in
Table 2.6. Representative binding curves for LE, BLE, BXLE and BXLE-avidin are
shown in Figure 2.13.

54
Table 2.6. IC50 values obtained for LE, BLE, BLE-avidin, BXLE and BXLE-avidin using
1/1,000 dilution of antibody produced in this laboratory. ^050 value for BLE-
avidin was estimated by fixing non-specific binding to 0% total binding and N to
1.
Competitor
IC50 (mol/assay)
Mean
Standard deviation
LE
4.95* 10'13
3.70* 10'13
3.78* 10'13
3.44* 1013
2.79* 10'13
4.19* 10"13
3.8* 10'13
7.3* 1014
BLE
6.22*10'13
1.21*1 O12
1.38* 10'12
1.1*1012
4.0* 1013
BLE-avidin
*3.73*10
BXLE
3.53 10"14
4.97*10'15
2.0* 1014
BXLE-avidin
1.70*10
1.11*10
1.4*10
As with the commercial antiserum, the N-terminal biotinylated derivatives were
seen to retain affinity for the antibody produced in this laboratory. However, a shift in
affinity by almost three orders of magnitude is seen when complexes are formed between
BXLE and avidin (Table 2.6), and pre-formed BLE-avidin complexes showed binding to
this antibody at only the highest concentration tested (Figure 2.13).
Again, as with the commercial antiserum, BXLE showed a higher affinity than LE
to the antibody produced in this laboratory. Here too, as before, it was postulated that
since LE was conjugated to a porcine thyroglobulin to render it immunogenic for antibody
production, this biotinylated LE derivative including a spacer arm may have resembled the

55
epitope presented to the immune system more closely than LE itself and as a result, BXLE
also showed a higher affinity to this antibody than LE itself.
Figure 2.13. Representatives binding curves for LE (), BLE (), BXLE (), BLE-
avidin () and BXLE-avidin (0) using 1/1,000 dilution of antibody produced
in this laboratory.
The C-terminal biotinylated LE derivatives LE-Lys6-B and LE-Lys6-BX showed
no binding to the antibody produced in this laboratory in the concentrations used (data not
shown) and therefore, in this preparation, they were also unsuitable for the development of
either of the immunoassays proposed in this study. This result was expected since the
antibody was produced in this laboratory following inoculation with LE conjugated to a
carrier protein via the N-terminal end. Therefore, the C-terminal end of the peptide was

56
exposed as an epitope to produce an immune response in the host animal, resulting in the
production of antibodies towards this end of the peptide. Since LE-Lys6-B and LE-Lys6-
BX are modified at the C-terminal end, one would naturally not expect them to bind to an
antibody directed towards the C-terminal end ofLE.
The formation of a sandwich between antibody, biotinylated LE derivative and
avidin was not achieved using either the antibody produced in this laboratory or the
commercial antiserum and any of the biotinylated LE derivatives synthesized. Since
sandwich formation is a requirement for the successful development of the ELISA
proposed in this study, efforts in this direction were abandoned at this point.
A lack of sandwich formation was observed between the antibody produced in this
laboratory, BLE and avidin, and therefore, it was concluded that this combination of
reagents was suitable for the development of the homogeneous fluorescence immunoassay
proposed in this study.
Development of the Proposed Homogeneous Fluorescence Immunoassay
Determination of excitation and emission maxima for fluorescence-labeled avidin
The excitation maximum for FITC-avidin was determined to be 482 nm.
Figure 2.14 below shows the emission scans for FITC-avidin and FITC-avidin with BLE.
Maximum emission was observed at 517 nm and fluorescence was seen to increase by a
factor of 4 when FITC-avidin interacted with BLE. Therefore, Xexc 482 nm and Xem 517
nm were chosen for future fluorescence readings.

57
Figure 2.14. Emission scans with X.exc at 482 nm for FITC-avidin and FITC-avidin
with BLE.
Determination of reagent concentrations
An antibody dilution of 1/50 corresponding to a concentration of 8.7 pmol/ml of
specific antibody sites was chosen for use in these experiments. Although a higher
concentration of specific antibody sites in the final assay would have been preferred to
afford binding of a greater number of tracer (BLE) molecules, the decision was based on
practicality since a limited amount of antibody was available from the exsanguination of a
single rabbit.
The calculations using the chosen antibody concentration, the Ka value obtained
from the Scatchard plot and the Law of Mass Action indicated that a concentration of

58
BLE of 4 pmol/ml would be 95 % bound by the chosen antibody concentration. Therefore,
this concentration of tracer was chosen for use in all further experiments.
Figure 2.15. Plots of fluorescence versus amount of BLE added using 200 fmol/ml of
FITC-avidin (A) and 500 fmol/ml of FITC-avidin (B)
Although one would expect that about 4 times as much BLE as FITC-avidin
would produce maximum fluorescence enhancement of FITC-avidin since it is known that
4 moles of biotin will bind to 1 mole of avidin, Figure 2.15A shows that a maximum
fluorescence reading is obtained when 4 pmol/ml of BLE is added to 200 fmol/ml of

59
FITC-avidin. The results of the HABA displacement experiments (Figure 2.11) indicated
that BLE does have a lower affinity than biotin for avidin and in addition, since here FITC-
avidin is being used, one might expect BLE to exhibit a lower affinity for FITC-avidin than
biotin exhibits for avidin. In the plot shown in Figure 2.14A, the working scale is rather
narrow and the curve obtained is not smooth as considerable noise interfered with the
taking of the readings. Therefore, although a maximum fluorescence reading using 500
fmol/ml of FITC-avidin is seen at 10 pmol/ml rather than 4 pmol/ml of BLE (Figure
2.15B), this concentration of FITC-avidin was chosen for use in this assay since less noise
is observed in the readings and the working scale is more practical.
Homogeneous fluorescence immunoassay
One of the first calibration curves obtained for LE using the complete
homogeneous fluorescence immunoassay is shown in Figure 2.16. Here, as expected, an
increase in fluorescence is seen with an increase of LE present in the sample. However,
using the LE concentrations shown here, a plateau at the maximum theoretical
fluorescence (total tracer value, no antibody present) was not reached. Therefore, another
calibration curve was constructed using higher concentrations of LE in the samples in an
attempt to reach this plateau (Figure 2.16).
Figure 2.17 shows a calibration curve for LE using the complete homogeneous
fluorescence immunoassay where fluorescence readings above the maximum theoretical
fluorescence were obtained at higher concentrations of LE (1*1 O'8 moles/ml and above).
In order to determine whether this unexpected effect could be attributed to either the

60
antibody, the tracer or the analyte, the assay was repeated omitting each of these reagents
in turn.
100
80
60
c
3
E
40
20
Total tracer
FI blank
o 1 1
IE-11 IE-10 IE-9 IE-8
LE(moles/niI)
Figure 2.16. Calibration curve for LE using homogeneous fluorescence immunoassay.
180
160
140
120
Q
8 loo
80
60
40
20
0


Total tracer




FI blank



-+-
-+-
IE-11 IE-10 IE-09 IE-08
LF moles/ml
i
IE-07
IE-06
Figure 2.17. Calibration curve for LE using homogeneous fluorescence immunoassay
showing readings exceeding maximum theoretical fluorescence.

61
IE-11 IE-10 IE-09 IE-08 IE-07 IE-06
LE (inoles/ml)
Figure 2.18. Calibration curve for LE using the homogeneous fluorescence immunoassay
set up with BLE () and without BLE(D).
IE-11 IE-10 IE-09 IE-08 IE-07 IE-06
LE(molcs/nil)
Figure 2.19. Calibration curve for LE using the homogeneous fluorescence immunoassay
set up, omitting antibody but with BLE () and without BLE(D).
Figure 2.18 shows that at in the presence of the antibody, when the tracer (BLE) is
omitted from the assay, fluorescence readings began to increase above the fluorescence

62
blank levels at concentrations of LE in excess of 1*1 O'8 moles/ml. This effect was also
observed when the antibody was omitted from the assay (Figure 2.19). When both tracer
and antibody were included in the assay (Figure 2.18), fluorescence levels began to
increase above fluorescence blank levels at concentrations of LE in excess of 1*10 9 moles
/ml. When antibody is omitted but tracer is included in the assay, fluorescence readings
increased above total tracer levels at LE concentrations in excess of 1*1 O'8 moles/ml
(Figure 2.19). This is the same concentration of LE at which the fluorescence readings
increased above fluorescence blank levels when both tracer and antibody were omitted
from the assay.
Des-'tyr LIC(moles/ml)
Figure 2.20. Calibration curve for des-Tyr1 LE using the homogeneous fluorescence
immunoassay set up, omitting antibody but with BLE () and without
BLE(D).
These results led us to believe that at higher concentrations of LE, the analyte
interacted directly with FITC-avidin to give increased fluorescence readings. It was
hypothesized that this effect could be due to an interaction between the tyrosine moiety in

63
the 1 position of LE and the fluorescein groups of the detector molecule FITC-avidin. In
order to test this hypothesis, assays were carried out in the absence of antibody, in the
same way as described above. However, here des-Tyr1 LE or pentaglycine were used as
analytes in concentrations similar to those of LE which had been used previously.
Figure 2.21. Calibration curve for pentaglycine using the homogeneous fluorescence
immunoassay set up, omitting antibody but with BLE () and without
BLE(D).
Figures 2.20 and 2.21 show that des-Tyr1 LE and pentaglycine do not produce the
same interaction with FITC-avidin as LE when used in the same concentrations in the
homogeneous immunoassay set up. When BLE is included in the assay, the readings
approach those obtained in the total tracer samples and when no BLE is present in the
assay, the readings approach those in the fluorescence blank samples. It was therefore
concluded that the effect seen at higher concentrations of LE in the homogenous

64
fluorescence immunoassay were probably due to an interaction between the tyrosine
moiety of LE and the fluorescein groups of FITC-avidin.
The homogenous fluorescence immunoassay evaluated here operates as intended in
the concentration range between 1*1 O'9 moles/ml and 1*1 O'8 moles/ml of LE since at these
concentrations, sufficient tracer (BLE) is displaced from the antibody by LE to produce
detectable fluorescence enhancement on interaction with the detector molecule (FITC-
avidin), but the concentration of LE is not high enough to produce a direct interaction
with FITC-avidin in the absence of BLE. Using the reagents tested, an homogeneous
fluorescence immunoassay for LE operational over a wide concentration range could not
be developed since at higher concentrations (1*1 O'8 moles/ml and above), LE interacted
directly with FITC-avidin to produce fluorescence enhancement, thus interfering with the
signal produced through the interaction of BLE with FITC-avidin. In this assay, the use of
an antibody with a higher affinity for LE would present several advantages. Firstly, less
antibody would be required to ensure maximal binding of the BLE used in the assay and
consequent low background fluorescence. Secondly, less LE would be required to displace
sufficient BLE from the antibody to produce a measurable signal on interaction with
FITC-avidin and therefore, LE concentrations in the assay displacing the maximum
amount of BLE may not reach the concentrations which were observed to produce a
direct interaction with FITC-avidin and as a result interfered with the success of the assay.
Thirdly, since lower concentrations of LE would be required to displace BLE from the
antibody and produce a measurable signal, the sensitivity of the assay would be increased.

65
A fluorescence spectrophotometer set at close to maximum signal amplification
was used to measure the fluorescence signal in the development of the homogenous
fluorescence immunoassay. The use of laser-induced fluorescence may have allowed the
successful development of the proposed homogeneous immunoassay as less BLE could
have interacted with a lower concentration of FITC-avidin to produce a measurable signal.
Therefore, less LE would have been required to displace BLE from antibody binding sites
so that the higher concentrations of LE which were observed to produce a direct
interaction with FITC-avidin would not have been attained.
In conclusion, although it has been demonstrated that the homogenous
fluorescence immunoassay for LE proposed in this study works in theory, a fully
operational assay could not be developed in practice as the concentrations of LE required
in this assay to displace the maximum amount of BLE from the antibody interacted
directly with FITC-avidin to produce fluorescence enhancement of the detector molecule.
Existing homogeneous fluorescence immunoassays for haptens and proteins lie in the 10'8
to 109 moles/ml range [Jenkins 1992] and homogeneous enzyme immunoassays are
capable of limits of detection 10'14 moles/ml of haptens [Engel and Khanna 1992], It is
proposed that the use of an antibody with higher affinity for LE would allow the
successful development of the type of homogeneous fluorescence immunoassay for LE
described in this study as lower concentrations of LE could be used and therefore the
direct interaction between LE and FITC-avidin that was encountered would be avoided. It
is anticipated that the successful development of this type of homogeneous immunoassay
would allow the detection of LE in the 10'9 to 10'11 moles/ml range.

CHAPTER 3
A HIGH PERFORMANCE LIQUID CHROMATOGRAPHY ASSAY FOR OPIOID
PEPTIDES USING ELECTROCHEMICAL DETECTION
Introduction
The objective of this study was the evaluation of a tyrosine-specific clean-up and
detection method for opioid peptides using leucine enkephalin (LE, Figure 3.1) as a model
peptide. The assay described here is based on the derivatization of LE by specific
enzymatic o-hydroxylation of the highly conserved tyrosine groups in the 1 position of
opioid peptides by mushroom tyrosinase. This derivatization results in the formation of a
catechol which is amenable to specific clean-up using boronate gels and is more easily
oxidizable than the parent peptide, thus facilitating electrochemical detection.
OH
CH
ch.
o
o
CH.
II
II
CH NH C
II
II
II
o
o
CH.
O
Tyr Gly Gly Phe Leu
Figure 3.1. Leucine enkephalin
66

67
Mushroom tyrosinase is a member of the monoxygenase class of enzymes which
catalyses two successive reactions (Figure 3.2): the hydroxylation of mono-phenols
(monophenolase activity) and the oxidation of o-diphenols (diphenolase activity) [Walsh
1979], The o-quinones resulting from these two successive reactions often form high
molecular weight polymerization products in vivo such as melanin. In this study, the
formation of undesirable polymerization products was prevented by the addition of
appropriate amounts of ascorbic acid as a reductant.
Figure 3.2. Tyrosinase catalyzed reactions
Small molecules such as L- and D-tyrosine and L- and D-dopa are endogenous
substrates for tyrosinase. However, enzyme activity has been shown with tyrosine-
containing di- and tri-peptides [Marumo and Waite 1986, Tellier et al. 1991] as well as LE
and ME [Rosei et al. 1991, Rosei et al. 1989], Larger proteins such as insulin, serum
albumin and dehydrogenase enzymes have also been shown to be oxidized by tyrosinase
[Cory et al. 1962, Cory and Frieden 1967a, Cory and Frieden 1967b, Gemant 1974, Ito et
al. 1984],
In the assay described here, the enzymatic derivatization of LE by the means
described above presents two analytical advantages. Firstly, the specific o-hydroxylation

68
of the tyrosine group of LE to give a more easily oxidizable catechol allows the use of a
lower oxidation potentials for electrochemical detection, thus avoiding many of the
disadvantages associated with high applied potentials such as high background current and
baseline noise [Kim et al. 1989], Selectivity is also compromised when high applied
potentials are used as more compounds are oxidized at these high potentials. Therefore,
extensive clean-up procedures to eliminate interfering peaks are often required when high
applied potentials are used [Fleming and Reynolds 1988], Secondly, the enzymatic
derivatization increases the selectivity of this assay as the derivative is amenable to a
specific boronate clean-up method which has previously been established for
catecholamines [Eriksson and Wikstrom 1992, Higa et al. 1977, Koike et al. 1982],
immobilized boronale gel
hydroxylated LE
k
H+ OH-
complex
Figure 3.3. pH dependent complex formation between immobilized boronate gel and
hydroxylated LE.

69
The introduction of a 3-hydroxytyrosine group to the LE molecule by enzymatic
derivatization allows the use of a specific clean-up procedure for 3,4-dihydroxyphenyl
compounds using boronate gels and column chromatography. For this assay, the boronate
clean-up method is based on the pH-dependent formation of a complex between
immobilized boronate gel and the hydroxylated LE derivative (Figure 3.3). A complex is
formed at weakly alkaline pH between ionized boronate affixed to a gel matrix and the
hydroxylated LE derivative. Dissociation of the complex occurs at acidic pH.
Therefore, in the assay described here, LE in the sample is first derivatized
enzymatically by mushroom tyrosinase and subsequently, the sample is subjected to clean
up through the use of a boronate gel column. Finally, the sample is quantified by high
performance liquid chromatography with electrochemical detection (HPLC-ED).
Materials
Leucine enkephalin and mushroom tyrosinase were obtained from Sigma Chemical
Company, St. Louis, MO, USA. Acetonitrile, methanol and trifluoroacetic acid were of
HPLC grade and disodium hydrogen phosphate and citric acid were of reagent grade.
These chemicals were procured from Fisher Scientific, Pittsburgh, PA, USA. Sodium
dihydrogen phosphate was of molecular biology grade and was purchased from Fluka
Chemie, Buchs, Switzerland. All other chemicals were of reagent grade. Double distilled
water was used throughout.

70
Methods
Purification of Mushroom Tyrosinase
Mushroom tyrosinase was purified prior to use by ultra-filtration using Centricon
membrane filters (molecular weight cut off 30,000, Amicon, Danvers, MA, USA). One
milliliter of a solution of mushroom tyrosinase (1 mg/ml) in 0.1 M sodium phosphate
buffer pH 7 was applied to the filter unit and centrifuged at 5,000 g until maximum
concentration of the sample was achieved. This centrifugation step was repeated three
times with the addition of an additional 2 ml of phosphate buffer prior to each
centrifugation. The final concentrate was reconstituted in phosphate buffer to give a final
concentration of 1 mg/ml of mushroom tyrosinase corresponding to 3870 units of activity
per ml of solution. Aliquots were stored at -20C and defrosted immediately prior to use.
Enzymatic Derivatization
To characterize the enzymatic derivatization procedure, the following incubation
mixture was set up in a microcentrifuge tube: 1 inM LE, 135 units/ml mushroom
tyrosinase and 50 niM ascorbic acid in 0.5 M phosphate buffer pH 7.4. The reaction was
allowed to proceed at room temperature with constant shaking for 60 minutes. Aliquots of
this incubation mixture were then applied to an HPLC system consisting of an
LDC/Milton Roy miniMetric II metering pump (Riviera Beach, FL, USA), a Negretti and
Zamba injector (Southampton, UK) fitted with a 500 pi loop, a Perkin Elmer LC-75
spectrophotometric detector (Norwalk, CT, USA) and a Hewlett Packard HP 3394A

71
integrator (Avondale, PA, USA). The column was a Partisil 5 ODS-3 125 x 4.6 mm
(Whatman Labsales, Hillsboro, OR, USA). The detection wavelength was 254 nm and the
mobile phase consisted of 12.5% acetonitrile (v/v) in citrate/dipotassium phosphate buffer
(pH 5) at a flow rate of 1 ml/minute.
The product peaks were collected from the HPLC eluent, organic solvent was
removed by evaporating under a stream of nitrogen and the collected peaks were
concentrated using a preconditioned Sep-Pak Ct8 preparative column (Waters Associates,
Milford, MA, USA). The identity of the products was determined by electrospray
ionization mass spectrometry (see Chapter 5).
Electrochemical Detection
Electrochemical detection was effected using a Model 5100A Coulochem multi
electrode electrochemical detector (ESA Inc., Bedford, MA, USA) fitted with a Model
5020 guard cell and a Model 5011 analytical cell. The complete HPLC-ED system was
configured as shown in Figure 3.4.
In this system, the guard cell acts to pre-oxidize electroactive impurities in the
mobile phase, thus reducing background current. The analytical cell consists of two
working electrodes in series. The first (Det 1) acts to further reduce background current
in the injected sample and to pre-oxidize co-eluting interferences that oxidize at potentials
lower than the analyte of interest. The second working electrode (Det 2) is set to quantify
the analyte. The solvent delivery system was an LDC/Milton Roy constaMetric III
metering pump (Riviera Beach, FL, USA), and the injector was a Rheodyne Model 7125

72
injector (Cotati, CA, USA) fitted with a 100 pi loop. A Spherisorb ODS2 5 pm 150 x 4.6
mm analytical column (Keystone Scientific Inc., Bellefonte, PA, USA) was used with a
mobile phase of 20% acetonitrile (v/v) in monosodium phosphate buffer (100 mM, pH 5)
containing 200 mg/1 of sodium dodecyl sulfate. The mobile phase was freshly prepared,
filtered through a 0.2 pm membrane filter and degassed under vacuum with sonication
daily, prior to use. Mobile phase flow rate was set at 1 ml/min. A Chromatopac C-R3A
integrator (Shimadzu Corporation, Kyoto, Japan) was used to record the output from the
control module.
Figure 3.4. Chromatographic configuration of electrochemical detection system.
Electrochemical Characterization
LE and monohydroxylated leucine enkephalin ([HO-Tyr]-LE), the major product
of the enzymatic reaction were characterized electrochemically using the HPLC-ED

73
1
system described above. Five nanomoles of analyte per injection were applied to the
system and the responses (in pA) obtained at various potentials (+0.05 to +0.80V) at the
analytical cell were recorded to allow the construction of current-voltage curves.
Boronate Clean-up
A hydrated boronate column packed with a 3 ml bed volume of Affigel 601
(BioRad Laboratories, Melville, NY, USA) was used for the clean-up of the enzymatically
derived species prior to application to the HPLC-ED system. The column was pre-rinsed
with 0.2 M phosphate buffer (pH 8.5) and an aliquot of the incubation mixture was
applied. The column was washed with 20 ml of phosphate buffer (pH 8.5) and the analyte
was eluted from the boronate column onto a pre-conditioned Sep-Pak Cig cartridge
(Waters Associates, Milford, MA, USA) with 20 ml of aqueous 0.01% (v/v)
trifluoroacetic acid (TFA). Here, the Sep-Pak Cig cartridge served to concentrate the
sample as the analyte was eluted from the boronate column in a relatively large volume of
aqueous 0.01% (v/v) TFA The Sep-Pak Cig column was then washed with 10 ml of
aqueous 0.01% (v/v) TFA and the analyte was eluted in 2 ml of 0.01% TFA in
acetonitrile. The sample was evaporated to dryness under a stream of nitrogen and the
residue was reconstituted in 200 pi mobile phase prior to application to the HPLC-ED
system.

74
Time Course of Enzymatic Derivatization
To determine the time course of the enzymatic derivatization at the analytical
concentrations, the following incubation mixture was set up in microcentriiuge tubes:
8*10'8 M LE, 50 mM ascorbic acid and 135 units/ml mushroom tyrosinase in 300 pi of
0.5 M phosphate buffer pH 7.4. The reaction was stopped at various time points by adding
20 pi of 1 N HC1. Two hundred an fifty micrometers of this incubation mixture was then
subjected to the boronate clean-up procedure, the resulting sample was injected into the
HPLC-ED system and the peak areas of the peak corresponding to the enzymatic
derivative [HO-Tyr'j-LE were recorded.
Extraction of Leucine Enkephalin from Cerebrospinal Fluid
Leucine enkephalin was extracted from human cerebrospinal fluid (CSF) through
the use of Supelclean LC 18 solid phase extraction columns (Supelco Inc., Bellefonte, PA,
USA). The columns were activated with 3 ml each of water and methanol and loaded with
100 pi of spiked CSF. Subsequently, the columns were washed with 1 ml of water, 3 ml of
0.1 N HC1, 1 ml of water, 3 ml of 0.1 M borate buffer (pH 8.5) and 1 ml of water. The LE
rich fraction was then eluted in 2 ml of methanol and evaporated to dryness under a stream
of nitrogen.
Calibration Curves
Using the complete HPLC-ED method, including enzymatic derivatization and
boronate clean-up, calibration curves were constructed for LE in both buffer and CSF. For

75
comparison, calibration curves for LE in CSF were also constructed using HPLC-ED
without enzymatic derivatization or boronate clean-up.
Results
Enzymatic Derivatization
Preliminary experiments indicated that the concentration of mushroom tyrosinase
(135 units/ml) used for the enzymatic derivatization of LE allowed the efficient conversion
of LE to its hydroxylated derivative within 60 minutes (data not shown). A 50 mM
concentration of ascorbic acid was found to be adequate to prevent or reverse the
tyrosinase-induced formation of o-quinones in the enzymatic reaction mixture, thereby
inhibiting the subsequent polymerization of the reaction products. Figure 3.5A shows the
chromatograph of a control run where only LE and ascorbic acid are present at the
incubation concentrations. Here, LE is seen eluting after 9.5 minutes, distinct from
ascorbic acid eluting in the solvent front after 1.4 minutes. Two additional peaks at 5.7
minutes and 7.3 minutes are seen in the chromatograph of the incubation solution (Figure
3.5B). These two product peaks were identified by electrospray ionization mass
spectroscopy (see Chapter 5) as the di- and mono-hydroxylated derivatives of LE
([(HO)2- Tyr']-LE and [HO-Tyr']-LE, respectively). The relative intensity of the peaks
obtained from mass spectrometric analysis showed that [(HO)2-Tyr']-LE is a minor
product.

76
A
Figure 3.5. Chromatographs of a control run (A) showing LE eluting after 9.5 minutes,
distinct from ascorbic acid eluting in the solvent front after 1.4 minutes and
incubation solution (B) showing the emergence of two new peaks with
retention times of 5.7 and 7.3 minutes.
Electrochemical Characterization
Current-voltage curves for LE and [HO-Tyr'j-LE are shown in Figure 3.6. When
these curves were compared, [HO-Tyr']-LE was seen to be oxidized at considerably lower
potentials than LE. As a result of this experiment, the following potentials were selected
for use in the construction of calibration curves for LE using the complete HPLC-ED

77
method (i.e. including enzymatic derivatization and boronate clean-up) and HPLC-ED
without enzymatic derivatization or boronate clean-up:
Guard cell
Analytical cell Det 1
Analytical cell Det 2
HPLC-ED with deriva
tization and cleanup
+0.4 V
-0.1 V
+0.3 V
HPLC-ED, no deriva
tization, no cleanup
+0.75 V
+0.4 V
+0.7 V
Figure 3.6. Current-voltage curves for (HO-Tyr^-LE and LE
Using these potentials, impurities in the mobile phase are pre-oxidized by the guard
cell thus reducing background current and baseline noise. Impurities in the sample with
relatively low oxidation potentials which might co-elute with the analyte and interfere with
the signal produced by the analyte are preoxidized at Det 1. The Det 1 potential is set
sufficiently low so that the analyte will not be pre-oxidized at this electrode, thus ensuring
the production of a maximum signal at the analytical potential at Det 2.

78
Boronate Clean-up
By applying 35 pmoles to the boronate gel column, an average recovery of analyte
of 68.67 % was achieved using the complete boronate clean-up method (SD = 6.1, n = 3).
Time Course of Enzymatic Derivatization
A representative time course of the enzymatic derivatization at analytical
concentrations (17 pmol/inj) showing the peak area of [HO-Tyr]-LE plotted against time
is shown in Figure 3.7. This experiment was carried out twice, showing the same trend
each time. The plot indicates that the highest level of [HO-Tyr']-LE is seen after a 5
minute incubation and therefore, this incubation time was selected for future use in this
study.
Figure 3.7. Time course of enzymatic derivatization showing peak height of [HO-Tyr1]-
LE over incubation time.

79
Calibration Curves
Table 3.1 shows the limits of detection (LOD, defined as twice the baseline noise),
average slope values and correlation coefficients (r2) obtained for the various calibration
curves constructed for LE using this analytical approach. Raw data can be found in
Appendix A. Representative calibration curves for LE in buffer and CSF using the
complete HPLC-ED method are shown in Figures 3.8 and 3.9. Figure 3.10 shows a
representative calibration curve for LE in CSF using FIPLC-ED without enzymatic
derivatization or boronate clean-up. The limits of detection for LE in CSF correspond to
8.8 pmol/ml of CSF for the complete HPLC-ED method and 176 pmol/ml of CSF for
analysis by HPLC-ED without derivatization or boronate clean-up. A sample
chromatograph of LE in CSF shows the analyte eluting after 6 minutes (Figure 3.11).
Table 3.1. Limits of detection (LOD), average slopes and correlation coefficients (r2) for
LE in buffer or CSF using either complete HPLC-ED method or HPLC-ED
without derivatization or boronate clean-up. For raw data, see Appendix A.
Sample
Method
LOD
fmol/inj
Slope (SD, n)
peak area/pmol/inj
r2 (SD, n)
Buffer
complete
HPLC-ED
170
28071 (2381, 3)
0.9925 (0.005, 3)
CSF
complete
HPLC-ED
360
22057(1064, 3)
0.9944 (0.004, 3)
CSF
HPLC-ED
no derivatization
no clean-up
8800
10217 (1163, 3)
0.9690 (0.024, 3)

80
Figure 3.8. Representative calibration curve for LE in buffer using complete
HPLC-ED method.
Figure 3.9. Representative calibration curve for LE in CSF using complete
HPLC-ED method.

81
Figure 3.10. Representative calibration curve for LE in CSF using HPLC-ED without
derivatization or boronate clean-up.
Figure 3.11. Sample chromatograph ofLE in CSF showing peak of interest eluting after
6 minutes.

82
To give an indication of the inter-day variability associated with the complete
HPLC-ED method incorporating enzymatic derivatization and boronate clean up, using
the calibration curves obtained for LE in CSF, nominal concentrations in the samples were
compared to found concentrations (Table 3.2). The found concentrations were determined
from the calibration curve after regression analysis was repeated while omitting the data
point under investigation. The relative standard deviation of the found concentrations was
found to be <20% (n=3) and the accuracy was found to be within 6% of the nominal
concentration.
Table 3.2. Nominal concentrations, found concentrations, relative standard deviation (SD)
and percentage accuracy calculated from 3 calibration curves for LE in CSF
using complete HPLC-ED method. For raw data, see Appendix A.
Nominal cone.
(pmol/100 pi CSF)
Found cone, n=3
(pmol/100 pi CSF)
Relative SD (%)
% Accuracy
1.76
1.84
20
104
3.46
3.34
12
97
6.67
6.41
12
96
10.48
11.13
1
106
17.14
17.65
9
103
Discussion
The HPLC-ED method we have evaluated for the analysis of LE in CSF compares
favorably to existing HPLC methods for opioid peptides with on-line detection [de
Montigny et al. 1990, Mifune et al. 1989, Muck and Henion 1989] and represent an
improvement with respect to limit of detection and practicability when compared to

83
current HPLC-ED methods for enkephalins (Table 3.3) [Fleming and Reynolds 1988, Kim
et al. 1989, Shibanoki et al. 1990], The increased practicability of the analytical approach
described here is characterized by the minimal precautions required in the preparation of
the mobile phase (filtering and degassing under negative pressure with sonication) and is
due to the fact that the enzymatic derivatization employed allowed the use of lower
applied potentials for electrochemical detection (+0.3 V compared to +0.85-1.25 V for
existing HPLC-ED methods). The low applied potentials used allowed the electrochemical
detector to be operated at the maximum gain setting since background current and
baseline noise were low. These factors, in addition to the minimal baseline drift observed,
increased the ease of handling of this analytical approach since minimal precautions could
be taken in the preparation of mobile phase and samples. This was demonstrated when a
comparison was made between the analysis of LE in CSF using the complete HPLC-ED
method incorporating enzymatic derivatization and boronate clean up and the HPLC-ED
analysis of LE in CSF without derivatization or boronate clean-up. When an attempt was
made to analyze LE in CSF without derivatization or boronate clean-up, considerably
higher applied potentials had to be used (+0.7 V compared to +0.3 V) and as a
consequence, background current, baseline noise and baseline drift were greatly increased.
Therefore, the maximum gain setting of the instrument could not be used, sensitivity was
compromised by a factor of 25 and the linearity of the calibration curves was reduced. In
addition, the system was more difficult to handle as the mobile phase had to be continually
degassed under a stream of helium in order to avoid unacceptable baseline drift.

84
Table 3.3. Comparison table of current analytical methods for opioid peptides.
Reference
Method
Analyte
Matrix
Sensitivity
This study
HPLC-ED
LE
Buffer
170 fmol/inj
This study
HPLC-ED
LE
CSF
360 fmol/inj
Fleming and
Reynolds 1988
HPLC-ED
enkephalin
rat brain
1 pmol/inj
Shibanoki et al.
1990
HPLC-ED
enkephalin
plasma
550 fmol/inj
Kimetal. 1989
HPLC-ED
enkephalin
rat brain
1 pmol/inj
Muck and
Henion 1989
HPLC-MS
dynorphin
CSF
100 fmol/inj
Mifune et al.
1989
HPLC-FL
enkephalin
rat brain
100 fmol/inj
de Montigny et
al. 1990
HPLC-FL
LE
plasma
7.7 pmol/inj
HPLC-MS = high performance liquid chromatography with mass spectrometry, HPLC-FL
= high performance liquid chromatography with fluorescence detection.
The boronate clean-up procedure used in the HPLC-ED assay described here
proved to be an effective and efficient clean-up method for LE in CSF resulting in a clean
chromatograph for the analyte (Figure 3.11). To avoid interfering peaks, existing HPLC-
ED methods for enkephalins involve complex and relatively non-selective sample clean up
procedures (e.g./ multiple precipitation, centrifugation and adsorption steps) prior to
application of the sample to the HPLC system. By contrast, here the use of a boronate
clean-up procedure, which had been previously established for use with catecholamines
[Higa et al. 1977, Koike et al. 1982], increased the selectivity of this assay as only

85
derivatized peptide incorporating the dihydroxy group introduced by enzymatic
derivatization was retained on the boronate gel column. The limits of detection achieved
for LE using this analytical approach (170 fmol/inj in buffer and 360 fmol/inj in CSF) also
compared favorably to assays for catecholamines using a boronate clean-up method and
electrochemical detection (200 fmol/inj) although the recovery of hydroxylated LE from
the boronate gel (69%) was considerably lower than the recovery of catecholamines from
the same matrix (80-100%) [Higa et al. 1977, Koike et al. 1982],
The HPLC-ED approach for the determination of LE in CSF described here was
found to be reproducible and accurate since the elative standard deviations of various
concentrations determined on different days was found to be < 20% and found
concentrations were determined to be within 6% of nominal concentrations (Table 3.2).
In theory, the analytical method we have described here is applicable to a whole
range of opioid peptides since the N-terminal tyrosine group which is derivatized is highly
conserved throughout the entire family of opioid peptides. This method should also be
applicable to the analysis of other tyrosine-containing proteins and peptides as it has been
shown previously that the amino acid adjacent to the tyrosine group does not dramatically
influence the tyrosinase reaction [Tellier et al. 1991], Endogenous levels of opioid
peptides in human CSF lie in the fmol/ml range as determined by radioimmunoassay
[Eisenach et al. 1990, Hardebo et al. 1989, Samuelsson et al. 1993, Yaksh et al. 1990,
Young et al. 1993], Although the analytical approach described here is inadequate for the
determination of endogenous levels of opioid peptides in human CSF, given sufficient

86
sample (>1 ml), it may be adequate for the analysis of the elevated physiological
concentrations of opioid peptides to be expected in clinical studies.

CHAPTER 4
A HIGH PERFORMANCE LIQUID CHROMATOGRAPHY ASSAY FOR OPIOID
PEPTIDES USING FLUORESCENCE DETECTION
Introduction
The objective of this study was the evaluation of a tyrosine-specific analytical
approach for opioid peptides using high performance liquid chromatography with
fluorescence detection and leucine enkephalin (LE, Figure 4.1) as a model peptide.
OH
Tyr Gly Gly Phe Leu
Figure 4.1. Leucine enkephalin
As in Chapter 3, this analytical approach exploits the derivatization of LE by
specific enzymatic o-hydroxylation of the highly conserved tyrosine groups in the 1
87

88
position of opioid peptides by mushroom tyrosinase. Here, the enzymatic derivatization
renders the peptide suitable for subsequent fluorogenic derivatization using 1,2-diamino-
1,2-diphenylethane (DPE). Therefore, in this assay, the hydroxylated LE derivative
obtained from the reaction with mushroom tyrosinase is oxidized in a controlled manner
by potassium ferricyanide to give the corresponding quinone prior to a condensation
reaction with DPE to give a fluorescent product (Figure 4.2).
Tyrosinase reaction
OH
OH
O
V1O2
T
tyrosinase
OH
Polymerization
DPE reaction
Oxidation

Potassium
ferricyanide
Hydroxylated LE Corresponding quinone
CH-CH-
I I
NHj NHj
Condensation
'1
DPE
Fluorescent product
Figure 4.2. Overview of derivatization reactions for HPLC-FL assay

89
Materials
Leucine enkephalin and mushroom tyrosinase were obtained from Sigma Chemical
Company, St. Louis, MO, USA. Acetonitrile, methanol and trifluoroacetic acid were of
HPLC grade and disodium hydrogen phosphate, sodium dihydrogen phosphate, citric acid,
potassium chloride and potassium ferricyanide were of reagent grade. These chemicals as
well as Scintiverse II scintillation cocktail were procured from Fisher Scientific,
Pittsburgh, PA, USA. Tetrabutylammonium (TBA) was purchased from the Eastman
Kodak Company, Rochester, NY, USA and [tyrosyl-3,5-3H(N)]-leucine enkephalin (3H-
LE) was obtained from NEN Research Products, Dupont Company, Wilmington, DE,
USA. 1,2-Diamino-1,2-diphenylethane (DPE) was synthesized according to Irving and
Parkins [Irving and Parkins 1965], Briefly, benzaldehyde (1.9 equivalents) was refluxed
with ammonium acetate (1 equivalent) for 3 hours, the resulting precipitate was collected,
and washed with ethanol. This precipitate was then hydrolyzed with 33 % v/v H2S04,
benzoic acid and benzaldehyde were removed by steam distillation, and DPE was
precipitated by neutralization with ammonium hydroxide. After recrystallization from
petroleum ether, the product had a melting point of 118-119C (literature, 120C) [Irving
and Parkins 1965] and a H nuclear magnetic resonance spectrum which supported the
assigned structure. All other chemicals were of reagent grade. Double distilled water was
used throughout.

90
Methods
Purification of Mushroom Tyrosinase
Mushroom tyrosinase was purified prior to use by ultra-filtration using Centricon
membrane filters (molecular weight cut off 30,000, Amicon, Danvers, MA, USA). One
milliliter of a solution of mushroom tyrosinase (1 mg/ml) in 0.1 M phosphate buffer pH 7
was applied to the filter unit and centrifuged at 5,000 g until maximum concentration of
the sample was achieved. This centrifugation step was repeated three times with the
addition of an additional 2 ml of phosphate buffer prior to each centrifugation. The final
concentrate was reconstituted in phosphate buffer to give a final concentration of 1 mg/ml
of mushroom tyrosinase corresponding to 3870 units of activity per ml of solution.
Aliquots were stored at -20C and defrosted immediately prior to use.
Enzymatic Derivatization
To characterize the derivatization procedure, LE (1 niM) in 0.5 M phosphate
buffer pH 7.4 was reacted with mushroom tyrosinase (135 units/ml) in the presence of
ascorbic acid (50 mM) at room temperature with constant shaking. The product peaks
obtained from this enzymatic derivatization were collected from the HPLC eluent, organic
solvent was removed by evaporation under a stream of nitrogen and the collected peaks
were concentrated using a Sep-Pak Ci8 preparative column (Waters Associates, Milford,
MA, USA). The identity of the products was determined by electrospray ionization mass
spectrometry (see Chapter 5).

91
To determine the time course of the reaction, at various time points, aliquots of
this incubation solution were applied to an HPLC system consisting of an LDC/Milton
Roy miniMetric II metering pump (Riviera Beach, FL, USA), a Negretti and Zamba
injector (Southampton, UK) fitted with a 500 pi loop, a Perkin Elmer LC-75
spectrophotometric detector (Norwalk, CT, USA) and a Hewlett Packard HP 3394A
integrator (Avondale, PA, USA). The column was a Partisil 5 ODS-3 125 x 4.6 mm
(Whatman Labsales, Hillsboro, OR, USA). The detection wavelength was 254 nm and the
mobile phase consisted of 12.5% acetonitrile (v/v) in citrate/dipotassium phosphate buffer
(pH 5) at a flow rate of 1 ml/minute.
Extraction of Leucine Enkephalin from Cerebrospinal Fluid
Leucine enkephalin was extracted from human cerebrospinal fluid (CSF) through
the use of Supelclean LCig solid phase extraction columns (Supelco Inc., Bellefonte, PA,
USA). The columns were activated with 3 ml each of water and methanol and loaded with
100 pi of spiked CSF. Subsequently, the columns were washed with 1 ml of water, 3 ml of
0.1 N HC1, 1 ml of water, 3 ml of 0.1 M borate buffer (pH 8.5) and 1 ml of water. The
LE-rich fraction was then eluted in 2 ml of methanol and evaporated to dryness under a
stream of nitrogen.
To test the recovery of LE from CSF using this extraction procedure, 100 pi of
CSF spiked with 3H-LE (17,300 counts per minute (CPM)) was applied to an extraction
column and the procedure described above was carried out. The final methanolic LE-rich
fraction was collected in 1 ml aliquots, 4 ml of Scintiverse II scintillation cocktail were

92
added and the radioactivity (CPM) was measured using a Beckman LS 5,000 TD
scintillation counter (Fullerton, CA, USA). As a control, the radioactivity in a sample
containing the same amount of 3H-LE as the spiked CSF in 1 ml of methanol was also
measured.
Calibration curves
Calibration curves for LE (0.5-7 pmol/inj) in both buffer and CSF were
constructed using this HPLC-FL analytical approach.
For the tyrosinase reaction, various concentrations of LE in 0.5 M sodium
phosphate buffer pH 7.4 or spiked CSF extracts reconstituted in 0.5 M sodium phosphate
buffer pH 7.4 were reacted with mushroom tyrosinase (135 units/ml) in the presence of
ascorbic acid (50 mM). The total volume of this incubation mixture was 100 pi. After a
60 minute incubation at room temperature with constant shaking, 449 pi of an oxidizing
solution containing 9.45 mg/ml potassium ferricyanide, 17.85 mg/ml potassium chloride
and 55 % acetonitrile (v/v) was added followed by 45.5 pi of a solution of DPE containing
20 mg/ml in 0.1 N HC1. The fluorogenic reaction was allowed to proceed in the dark for
60 minutes at room temperature with constant shaking. A 250 pi aliquot of this reaction
mixture, representing 42 % of the total final reaction mixture, was then injected into an
FLPLC system consisting of an LDC/Milton Roy miniMetric II metering pump (Riviera
Beach, FL, USA), a Rheodyne Model 7125 injector (Cotati, CA, USA) fitted with a 500
pi loop, aNucleosil Cig column (5pm, 150 x 4.6 mm, Keystone Scientific, Bellefonte, PA,
USA) and a Perkin Elmer 650S spectrofluorodetector (Norwalk, CT, USA) set at A.ex 345

93
nm and Xem 480 nm with slit widths of 12.5 nm. A BD 41 chart recorder (Kipp and Zonen,
Delft, The Netherlands) and a Hewlett Packard HP 3394A integrator (Avondale, PA,
USA) recorded the output from the spectrofluorodetector. The mobile phase contained 32
% acetonitrile (v/v) and 67 mM TBA in citrate/dipotassium phosphate buffer (pH 5). The
system was operated at a mobile phase flow rate of 1 ml/min.
An attempt was made to optimize the fluorogenic derivatization reaction by
concentrating the fluorogenic reagents so that 100 pi of an oxidizing solution containing
40.1 mg/ml of potassium ferricyanide and 75.8 mg/ml of potassium chloride, 100 pi of
acetonitrile and 25 pi of a solution of DPE containing 20 mg/ml in 0.1 N HC1 were added
to the buffer samples or the CSF samples reconstituted in buffer. After a 60 minute
incubation in the dark with constant shaking, a 250 pi aliquot of this mixture, representing
77 % of the total final reaction mixture, was injected into the HPLC system described
above.
Since experiments conducted in parallel indicated that at low concentrations of
analyte, the highest yield of enzymatically derived species was obtained after a 5 minute
tyrosinase incubation time (see Chapter 3), an attempt was also made to optimize the
derivatization reaction by reducing the tyrosinase incubation time to 5 minutes.
Results
Enzymatic Derivatization
Preliminary experiments indicated that the concentration of mushroom tyrosinase
(135 units/ml) used for the enzymatic derivatization of LE allowed the efficient conversion

94
of LE to its hydroxylated derivative within 60 minutes. A 50 mM concentration of
ascorbic acid was found to be adequate to prevent or reverse the tyrosinase-induced
formation of o-quinones in our enzymatic reaction mixture, thereby inhibiting the
subsequent polymerization of the reaction products.
Figure 4.3. Chromatographs of a control run (A) showing LE eluting after 9.5 minutes,
distinct from ascorbic acid eluting in the solvent front after 1.4 minutes and
incubation solution (B) showing the emergence of two new peaks with
retention times of 5.7 and 7.3 minutes.

95
Figure 4.3 A shows the chromatograph of a control run where only LE and
ascorbic acid are present at the incubation concentrations. Here, LE is seen eluting after
9.5 minutes, distinct from ascorbic acid eluting in the solvent front after 1.4 minutes. Two
additional peaks at 5.7 minutes and 7.3 minutes are seen in the chromatograph of the
incubation solution (Figure 4.3B). These two product peaks were identified by
electrospray ionization mass spectroscopy (see Chapter 5) as the di- and mono-
hydroxylated derivatives of LE ([(HO)2- Tyr']-LE and [HO-Tyr']-LE, respectively).
Figure 4.4. Time course for the enzymatic derivatization of LE (1 mM) by mushroom
tyrosinase (135 units/ml) in the presence of ascorbic acid (50 mM) showing
the disappearance of LE () as [HO-Tyr']-LE (A) and (HO)2-Tyr1 (0) are
produced.
A time course of the enzymatic derivatization of LE by mushroom tyrosinase in the
presence of ascorbic acid is shown in Figure 4.4. This time course shows the
disappearance of the peak corresponding to LE as [HO-Tyr^-LE and [(HO)2- Tyr]-LE

96
are produced. An incubation time of 60 minutes was chosen for the enzymatic
derivatization in our HPLC-FL assay as at this time point, the peak corresponding to LE
had disappeared and the peak corresponding to our enzymatically derived species of
interest ([HO-Tyr]-LE) was seen to plateau. This time course was repeated several times
and the same trend was observed each time.
Recovery of Leucine Enkephalin from Cerebrospinal Fluid
Recovery of LE from CSF was estimated at >90% when the counts per minute in
the methanolic eluent of the extraction column following the extraction procedure was
compared to the control sample.
Calibration Curves
For the fluorogenic reaction, the concentration of reagents used was the same as
those used previously for fluorogenic derivatization of tyrosine-containing peptides
following enzymatic hydroxylation by mushroom tyrosinase [Tellier et al. 1991], Here, as
before, the concentrations of DPE and KC1 used were taken directly from corresponding
catechol assays [Mitsui et al. 1985, Nohta et al. 1984] but the concentration of potassium
ferricyanide was increased to ensure the formation of the quinone from the hydroxylated
LE derivative despite the presence of high concentrations of ascorbic acid. Acetonitrile is
included in the reaction mixture to facilitate the condensation reaction by reducing the
thermodynamic activity of water.

97
Table 4.1 shows a summary of the results obtained using the various modifications
attempted in our HPLC-FL assay. Some of the results using the same method and the
same matrix could not be averaged since the output from the fluorescence detector was
monitored using different equipment (integrator or chart recorder).
Table 4.1. Summary of results for HPLC-FL assay. For raw data, see Appendix B.
Method
Matrix
Limit of
detection
N
slope
Non-concentrated
reagents
buffer
562 fmol/inj
1
0.9966
1.79*1012
area/mol
Non-concentrated
reagents
buffer
281 fmol/inj
2
0.9836-0.9971
1.14-3.40* 1013
mm/mol
Non-concentrated
reagents
CSF
500 fmol/inj
1
0.9975
3.89* 1012
area/mol
Non-concentrated
reagents
CSF
891 fmol/inj
1
0.9670
1.08* 1013
mm/mol
Concentrated
reagents
Buffer
500 fmol/inj
2
0.9926-0.9984
2.25-5.09* 1012
area/mol
Concentrated
reagents
CSF
500 fmol/inj
3
0.99170.0061
2.0711.26* 1012
area/mol
Concentrated
reagents, 5 min
tyrosinase
Buffer
277 fmol/inj
3
0.989510.0086
6.6913.80* 1012
mm/mol
Figures 4.5 and 4.6 show representative calibration curves obtained for LE in
buffer samples and spiked CSF, respectively, using the HPLC-FL method with non
concentrated reagents. Limits of detection for LE of 500 fmol/injection could be obtained

98
in both buffer samples and spiked CSF samples. This limit of detection corresponds to 12
pmol of LE per ml of CSF. Figure 4.7 shows a representative chromatograph for LE in
CSF using our HPLC-FL method.
Figure 4.5. Calibration curve for LE in buffer using HPLC-FL method with non
concentrated reagents.
Figure 4.6. Calibration curve for LE in CSF using HPLC-FL method and non
concentrated reagents.

99
Figure 4.7. Chromatograph of LE in CSF using HPLC-FL method. LE is shown eluting
after 22.5 minutes.
An analytical advantage was not gained through the concentration of the
fluorogenic reagents and subsequent injection of a greater portion of the total final
reaction mixture as the same limit of detection of 500 fmol/injection could be obtained
when this modification in our method was attempted.
Although a lower limit of detection in buffer samples (277 fmol/injection) was
obtained when a 5 minute reaction time instead of a 60 minute reaction time was used for
the tyrosinase reaction (Figure 4.8), when this modification of the method was attempted
in spiked CSF samples, the linearity of the calibration curve could not be preserved (data

100
not shown). Attempts to improve the linearity of a calibration curve for LE in spiked CSF
samples using a 5 minute tyrosinase reaction time by improving the extraction procedure
or the incubation conditions were not pursued.
Figure 4.8. Calibration curve for LE in buffer samples using 5 minute tyrosinase reaction
time in HPLC-FL method.
Discussion
The FIPLC-FL analytical approach for LE developed in this study yielded detection
limits of 500 fmol per injection in both buffer and spiked CSF samples, corresponding to
12 pmoles of LE per ml of CSF. This approach shows similar sensitivity or represents an
improvement in the limit of detection by one order of magnitude when compared to
existing tyrosine-specific HPLC-FL methods with pre-column fluorescence derivatization
[Ishida et al. 1986, Kai et al. 1988, Nakano et al. 1987, Zhang et al. 1991], The

101
improvement in limit of detection seen with the approach described here may be attributed
in part to the fact that in some of these existing tyrosine-specific HPLC-FL methods
[Ishida et al. 1986, Kai et al. 1988], fluorogenic derivatization necessitates heating of the
reaction mixture to 100C for 3 minutes which may result in peptide instability and
consequent reduced recovery of intact fluorescence-labeled analyte. Although more
selective, the HPLC-FL approach described here does not achieve limits of detection for
LE as low as those obtained by investigators using methods involving post-column
fluorescence derivatization of the N-terminal primary amino group of opioid peptides (36-
100 fmol/injection) [Dave et al. 1992, Mifune et al. 1989, van den Beld et al. 1990],
However, due to lack of selectivity, these methods often involve the use of extensive
sample clean-up procedures and complex multi-dimensional chromatographic systems
(column switching) to minimize the occurrence of interfering peaks [Mifune et al. 1989],
The higher sensitivity obtained using these methods can be attributed to the use of laser-
induced fluorescence detection [Dave et al. 1992, van den Beld et al. 1990] and a
microbore chromatographic system [Dave et al. 1992], Incorporating a microbore
chromatographic system and laser-induced fluorescence into the analytical approach
described here could be expected to reduce considerably the limits of detection obtainable
by this approach.
An attempt to improve the approached described in this study by concentrating the
reagents in the fluorogenic reaction and thereby injecting a larger portion of the total
reaction mixture did not result in an analytical advantage. This may be due to a decrease in
the efficiency of the fluorogenic reaction in the presence of higher concentrations of

102
fluorogenic reagents. Although a second attempt to improve the analytical approach by
decreasing the tyrosinase reaction time resulted in a lower limit of detection in buffer
samples, when this modification was attempted with spiked CSF samples, the linearity of
the calibration curve was severely compromised. It is postulated that matrix components
in the spiked CSF extracts may have altered the efficiency of the tyrosinase reaction, thus
affecting the reproducibility of the yield of hydroxylated LE derivative after this short
incubation time.
Table 4.2. Comparison table of HPLC-FL analytical methods for opioid peptides.
Reference
Sensitivity
Analvte
Matrix
Comments
This study
500 fmol/inj
12 pmol/ml CSF
LE
Buffer
CSF
tyrosine specific
Ishida et al.
1986
7 pmol/inj
enkephalin
water
tyrosine specific
Nakano et al.
1987
270 fmol/inj
enkephalin
water
tyrosine specific
Kai et al. 1988
500 fmol/inj
enkephalin
rat brain
tyrosine specific
Zhang et al.
1991
0.33-1.21 pmol/inj
opioid peptides
rat brain
tyrosine specific
van den Beld et
al. 1990
80 fmol/inj
(3-endorphin
plasma
laser-induced
fluorescence
Mifune et al.
1989
100 fmol/inj
enkephalin
rat brain
column-switching
Dave et al.
1992
36 fmol/inj
LE
water
microbore FtPLC,
laser-induced
fluorescence

103
As with the HPLC-EC method described in Chapter 3, the HPLC-FL analytical
approach described here should also be applicable to a whole range of opioid peptides
since the N-terminal tyrosine group which is initially derivatized is highly conserved
throughout the entire family of opioid peptides. This approach should also be applicable to
the analysis of other tyrosine-containing proteins and peptides as it has been shown
previously that the amino acid adjacent to the tyrosine group does not dramatically
influence the tyrosinase reaction [Tellier et al. 1991], Endogenous levels of opioid
peptides in human CSF, as determined by radioimmunoassay lie in the fmol/ml range
[Eisenach et al. 1990, Hardebo et al. 1989, Samuelsson et al. 1993, Yaksh et al. 1990,
Young et al. 1993], Therefore, here too, although this analytical approach in its present
form is inadequate for the determination of endogenous levels of opioid peptides in human
CSF, given sufficient sample (>lml) it may also be adequate for the analysis of the
elevated physiological concentrations of opioid peptides to be expected in clinical studies.

CHAPTER 5
LEUCINE ENKEPHALIN-TYROSINASE REACTION PRODUCTS -
IDENTIFICATION AND BIOLOGICAL ACTIVITY
Introduction
Tyrosinase is a copper containing enzyme which catalyses the ortho-hydroxylation
of phenols and the subsequent oxidation of the resulting catechols to o-quinones. It is
common throughout nature and plays a central role in the biosynthesis of both
norepinephrine and melanin. In Chapters 3 and 4 of this dissertation, we have shown that
tyrosinase will react with the tyrosine-containing peptide leucine enkephalin (LE).
Previously, our group and others have shown that tyrosinase also reacts with other
tyrosine-containing peptides to give hydroxylated products [Marumo and Waite 1986,
Rosei et al. 1991, Rosei et al. 1989, Tellier et al. 1991].
Numerous studies have revealed a loss of activity in enkephalins and other opioid
peptides when the Tyr1 moiety is absent (for reviews, see Hansen and Morgan 1984 and
Shimohigashi 1986), however, few have shown the effect of modification of the aromatic
side-chain of this residue. Therefore, as tyrosinase is known to react with tyrosine-
containing peptides, our goal in this study was to determine the structure of the products
formed when tyrosinase reacts with LE and to define the affinity of the products modified
at the Tyr1 residue to opioid receptors in rat brain homogenate.
104

105
Materials
The following materials were procured from the sources indicated: Leucine
enkephalin, mushroom tyrosinase (3870 units/mg, E.C. 1.14.18.1), bestatin, thiorphan and
captopril from Sigma, St. Louis, MO, USA, ascorbic acid from Mallinckrodt, Paris, KY,
USA, [3H]-diprenorphine from Amersham International, Arlington Heights, IL, USA, and
Scintiverse II scintillation cocktail from Fisher Scientific, Pittsburgh, PA, USA. All other
chemicals were of reagent grade. Double distilled water was used throughout.
The chromatographic system used consisted of an LDC/Milton Roy miniMetric II
metering pump (Riviera Beach, FL, USA), a Negretti and Zamba injector (Southampton,
UK) fitted with a 500 pi loop, a Perkin Elmer LC-75 spectrophotometric detector
(Norwalk, CT, USA) and a Hewlett Packard HP 3394A integrator (Avondale, PA, USA).
The column was a Partisil 5 ODS-3 125 x 4.6 mm (Whatman Labsales, Hillsboro, OR,
USA). The detection wavelength was 254 nm and the mobile phase consisted of 12.5%
acetonitrile (v/v) in citrate/dipotassium phosphate buffer (pH 5) at a flow rate of
1 ml/minute.
Methods
Enzymatic Reaction
The reaction between LE and mushroom tyrosinase was carried out in
microcentrifuge tubes at the following concentrations: 1 mM LE and 50 mM ascorbic acid
in 0.1 M potassium phosphate buffer (pH 7). The reaction was started by adding

106
mushroom tyrosinase (previously purified as described in Chapters 3 and 4) to give a final
concentration of 135 units/ml. The reaction mixture was incubated at room temperature
with constant shaking and the reaction was stopped after 1 hour by adding 66 (1 of IN
HC1 per ml of incubation solution. A control incubation was carried out in parallel,
omitting the addition of mushroom tyrosinase to insure the integrity of LE under the
reaction conditions.
Product Identification
For the isolation of the reaction products, a total of 4 ml of incubation solution
was injected into the HPLC system in 500 pi aliquots. The product peaks were collected
and pooled according to their retention times. The collected peaks were then concentrated
under a stream of nitrogen to remove organic solvents and prepared for analysis by mass
spectrometry as follows: A Sep-Pak C18 cartridge (Waters Associates, Milford, MA,
USA) was activated with 2 ml of methanol and washed with 3 ml each of methanol/3%
acetic acid 70/30 v/v and 3% acetic acid v/v. The concentrated materials were applied to
the cartridge and the cartridge was washed with 3 ml of 3% acetic acid v/v. The analytes
were then eluted in 500 pi of methanol/3% acetic acid 70/30 v/v and stored at -20C prior
to mass spectrometric analysis.
A Vestec 200ES instrument (Vestec Corp., Houston, TX, USA) was used to
obtain the electrospray ionization (ESI) mass spectra [Allen and Vestal 1992], The sample
solution was drawn into a standard laboratory syringe (250 pi, Model 1710, Hamilton Co.,
Reno, Nevada, USA) and supplied into the electrospray probe through a 50 cm x 0.1 mm

107
i d. deactivated fused silica capillary (Scientific Glass Engineering, Victoria, Australia) at 2
to 5 pl/min flow rate by a medical infusion pump (SAGE Instruments, Model 34IB,
Boston, MA, USA). A 0.005 in i d. x 0.010 in o.d. flat tipped hypodermic needle held at
2.4 kV potential produced spray current in the range of 130 to 180 pA, when the tip of
the needle to nozzle orifice distance was about 10 mm. The source block was heated to
250 C, and the spray chamber temperature was around 55 to 60 C. A Vector/One data
system (Teknivent, St. Louis, MO, USA) was used to control the quadrupole analyzer
(2,000 Da mass range), and to collect mass spectra in the 100 to 1,000 Da mass range at 3
ms/Da scan rate. For molecular weight determination, the repeller to collimator voltage
was held at 18 V, and collision-induced dissociation in the skimmer to collimator region
was obtained at 50 V. At least ten spectra were averaged for each experiment.
Radioreceptor Assay
The enzymatic reaction and radioreceptor assay were carried out on the same day.
Two hundred microliters of the incubation solution from the enzymatic reaction were
injected into the HPLC system and the peak corresponding to [HO-Tyr'J-LE was
collected in 1.9 ml of mobile phase. A 90% conversion from LE to [HO-Tyr'j-LE was
assumed based on the disappearance of the peak corresponding to LE after a 1 hour
incubation with mushroom tyrosinase (Figure 5.1) and the relative intensities of the
product peaks obtained from mass spectrometric analysis. Therefore, this solution
contained 180 nmoles of [HO-Tyr^-LE and was used directly in the radioreceptor assays.
Two hundred microliters of a 1 mM solution of LE were also injected into the HPLC

108
system and the peak corresponding to LE was collected and used in the radioreceptor
assays. Mobile phase components were found not to interfere with the radioreceptor
assay.
Figure 5.1. Time course for the enzymatic derivatization of LE (1 mM) by mushroom
tyrosinase (135 units/ml) in the presence of ascorbic acid (50 mM) showing
the disappearance of LE () as [HO-Tyr]-LE (A) and (HO)2-Tyr (0) are
produced.
Rat brain membranes were prepared as described by Hochhaus et al [Hochhaus et
al. 1988], Briefly, the whole brain, without cerebellum, of male Sprague-Dawley rats
(120-140 g) was homogenized in 60 volumes of ice cold 50 mM Tris-HCl buffer (pH 7.4)
containing 100 mM NaCl. The homogenate was incubated for 1 hour at 20C and
centrifuged for 20 minutes at 4C. The pellet was then resuspended, washed twice with 50
mM Tris-HCl and diluted in 50 mM Tris-HCl to give 400 mg of rat brain membranes
per ml. Aliquots of this suspension were stored at -80C and used within one week.

109
The radioreceptor assays were carried out in duplicate in microcentrifuge tubes.
Each tube contained 30 pM bestatin, 0.6 pM thiorphan and 10 pM captopril as a
peptidase inhibitor cocktail, 6.25 mg or 12.5 mg of rat brain membranes, various
concentrations of LE or [HO-Tyr'j-LE as competitor, and either 0.2 nM of [3H]-
diprenorphine, 1.65 nM of [3H]-DAGO or 1 nM [3H]-DPDPE as tracer, to assay for total
opioid receptor, p or 5 sites, respectively, in 1 ml of 50 mM Tris-HCl buffer (pH 7.4). The
tubes were incubated at room temperature with constant shaking for 1 hour. When using
[3H]-diprenorphine as tracer, the tubes were then centrifuged at 12,000g for 10 minutes to
bring down the pellet. The pellets were washed three times with ice-cold 50 mM Tris-HCl
buffer (without resuspending) and then were dissolved in 1 ml of Scintiverse II scintillation
cocktail. When using [3H]-DAGO or [3H]-DPDPE as tracer, the rat brain membranes
were separated from the supernatant and washed with ice-cold 50 mM Tris-HCl buffer by
means of a rapid filtration technique. The filters retaining the rat brain membranes and
bound tracer were placed in 4 ml of Scintiverse II scintillation cocktail and after being
allowed to stand overnight, the radioactivity (CPM) in either the pellets or on the filters
was determined using a Beckmann LS 5,000 TD scintillation counter (Fullerton, CA,
USA). In a control experiment, HPLC analysis of the supernatant showed that both LE
and [HO-Tyr^-LE were stable under radioreceptor assay conditions. Non-specific binding
was determined in the presence of high concentrations of competitor (1 x 10'5 M). The
IC5o values of LE and [HO-Tyr']-LE (concentration of competitor displacing 50% of
bound tracer) were determined using the MINSQ non-linear curve-fitting program

no
(MicroMath Scientific software, Salt Lake City, UT, USA). The data were fitted to the
following model:
B=T-
j1 j)c
CN+IC:
-+NS
50
Where: B = CPM in the presence of competitor
T = CPM in the absence of competitor
C = competitor concentration
N = slope factor
NS = CPM under non-specific binding conditions
Results
Chromatography
As in Chapters 3 and 4, leucine enkephalin and the products of the enzyme
reaction were separated successfully from each other using the chromatographic system
described above (Figure 5.2). In a control incubation where LE and ascorbic acid were
injected into the system in concentrations equal to the initial concentrations of these
reagents in the incubation mixture, LE had a retention time of 9.5 minutes and was well
separated from ascorbic acid which eluted in a large solvent front at 1.4 minutes (Figure
5.2A). When the test incubation was applied to the system, two distinct additional peaks
were seen on the chromatogram with retention times of 5.7 minutes and 7.3 minutes,
respectively (Figure 5.2B). These peaks were collected and subjected to mass
spectrometry as described earlier. An incubation time of one hour was chosen for our
experiments as at this time point, most of the LE present was seen to have been consumed
in the reaction and the levels of the major product were seen to plateau (Figure 5.1).

Ill
Ascorbic acid was included in the reaction mixture to prevent further oxidation of our
hydroxylated products to the corresponding quiones.
1.4
A
1.4
B
Figure 5.2. Chromatographs of control incubation (A) showing LE eluting after 9.5
minutes, distinct form ascorbic acid eluting in the solvent front after 1.4
minutes, and a test incubation after 30 minutes (B) showing the emergence of
two new peaks with retention times of 5.7 and 7.3 minutes.

112
Mass spectrometry
ESI is a soft ionization technique that provides intact molecular ions from peptide
solutions [Whitehouse et al. 1985], However, selective fragmentation can also be effected
by collision-induced dissociation (CID) in certain regions of the ion source [Allen and
Vestal 1992, Katta et al. 1991, Smith et al. 1990], The nomenclature scheme, as proposed
by Roepstoff and Fohlman [Roepstorff and Fohlman 1984], used to label the sequence
ions in the mass spectra of pentapeptides is shown in Figure 5.3.
Figure 5.3. Nomenclature scheme for the labeling of sequence ions in the mass spectra of
pentapeptides.
The ESI mass spectra obtained from LE are shown in Figure 5.4. Using soft
ionization (repeller at 18 V, Figure 5.4A), only molecular ions ([M+H]+, m/z 556, and
[M+Na]+, m/z 578) were observed. Although the formation of protonated molecules
through acid-base equilibria in solution is preferred, the generation of sodiated species was
unavoidable upon use of commercially available HPLC-grade solvents [Lehman 1982],
From the spectrum obtained under CID conditions (repeller at 50 V, Figure 5.4B), several
important sequence ions (m/z 136, 221, 278, 397, and 425 for ai, b2, b3, a,}, and b4,

113
respectively) directly derived from Figure 5.3 were assigned. Two intense internal
fragments, [a4*y2]i at m/z 120 and [a4*y3]2 at m/z 177, were also observed. These were the
products of the dissociation of the protonated molecule (m/z 556). CID of the sodiated
enkephalin was limited, only the loss of the C-terminal leucine moiety was prominent (m/z
465).
556
Figure 5.4. Electrospray ionization mass spectra of LE. (A) Repeller at 18 V; (B) Repeller
at 50 V.

114
[M+HI*
[M+Naf
572 594
Figure 5.5. Electrospray ionization mass spectra of [HO-Tyr^-LE. (A) Repeller at 18 V;
(B) Repeller at 50 V.
The identification of the hydroxylated products of LE was achieved by comparing
the ESI mass spectra obtained to those of the parent peptide. The mono-hydroxylated
enkephalin, [(HO)-Tyr1]-LE, gave molecular ions at m/z 572 and 594 for [M+H]+ and
[M+Na]+, respectively (Figure 5.5A). As shown in Figure 5.5B, a 16 Da increase was
observed in the m/z value of the a and b sequence ions, starting with the aj ion of the
series (m/z 152). On the other hand, the internal fragments m/z 120 and 177 were not

115
affected. This is evidence that the phenylalanine moiety was not modified. Therefore, the
oxygen was incorporated, as expected, into the [Tyr1] residue. The ESI mass spectra
recorded from the second product of the enzymatic reaction (Figure 5.6) indicated, based
on the 32 Da increment of the molecular ions, and also of the a and b sequence ions from
aj to a4, m/z 168 to 429, as compared to the parent enkephalin, that additional
hydroxylation also took place at the N-terminal tyrosine, to give a di-hydroxylated
derivative, [(HO)2-Tyr1]-LE. The relative intensity of the peaks obtained from mass
spectrometric analysis showed that [(HO^-Tyr^-LE was a minor product and therefore,
sufficient quantities could not be collected for use in receptor binding studies.
ni/z
Figure 5.6. Electrospray ionization mass spectra of [(HO)2-Tyr1]-LE. (A) Repeller at
18 V; (B) Repeller at 50 V.

116
Receptor Binding Affinity
Representative displacement curves showing the binding of LE and [HO-Tyr']-LE
to receptors in rat brain homogenate using [3H]-diprenorphine, [3H]-DAGO (p sites) and
[3H]-DPDPE (5 sites) as tracer are shown in Figs 5.7, 5.8 and 5.9, respectively. The IC50
values obtained are shown in Table 5.1. The loss of affinity to total opioid receptors in rat
brain homogenate observed when LE is hydroxylated by mushroom tyrosinase is mirrored
by a loss of affinity to both p and 8 sites.
Figure 5.7. Displacement curves for LE () and [HO-Tyr^-LE (o) in rat brain
homogenate using [3H]-diprenorphine as tracer.

117
Figure 5.8. Displacement curves for LE () and [HO-Tyr^-LE (o) in rat brain
homogenate using [3H]-DAGO as tracer.
Figure 5.9. Displacement curves for LE () and [HO-Tyr^-LE (o) in rat brain
homogenate using [3H]-DPDPE as tracer.

118
Table 5.1. Summary of IC50 values obtained in rat brain homogenate receptor binding
studies using [3H]-diprenorphine, [3H]-DAGO and [3H]-DPDPE as tracers
IC50 (nM) SD, n
[3H]-diprenorphine
[3Hl-DAGO
[3H]-DPDPE
LE
26 11, 4
18 4, 3
1.2 0.25, 3
[HO-Tyr]-LE
440 150,4
381 106, 3
34 2, 3
Discussion
Using mass spectrometry, the products of the reaction between leucine enkephalin
and mushroom tyrosinase in the presence of ascorbic acid were positively identified as
[HO-Tyr]-LE and [(HO^-TyUj-LE. The identification of [HO-Tyr']-LE confirms the
findings of previous workers [Rosei et al. 1991, Rosei et al. 1989], however, the
dihydroxylated derivative, [(HO^-TyUj-LE, was not isolated by these investigators as its
presence probably could not be detected by means of ultraviolet spectrometry without the
benefit of separation of reaction components by HPLC. Tyrosinase has been shown to
produce 5-OH-dopa from dopa in the presence of ascorbic acid [Hansson et al. 1981],
Similarly, [HO-Tyrj-LE seems to act as a substrate for tyrosinase to give [(HO^-Tyr1]-
LE. This stepwise reaction mechanism is supported by the slower and delayed formation
of the dihydroxylated derivative compared to the formation of [(HO)2-Tyr1]-LE.
The biological activity of opioid peptides with modified Tyr1 residues has not been
studied to the same extent as [des-Tyr1] opioid peptides which generally show a loss of
activity [Hansen and Morgan 1984, Shimohigashi 1986] (Table 5.2). Hansen et al.
[Hansen et al. 1985] showed that the introduction of methyl groups in the ortho position

119
Table 5.2. Effect of modification of Tyr1 moiety of opioid peptides on opioid binding
affinity.
Modification at Tvr1
Effect
Reference
[des-Tyr1] opioid peptide
loss of opioid binding
affinity
Hansen and Morgan 1984
Shimohigashi 1986
, /0H
HO^ ^R
increased opioid binding
affinity
Hansen et al. 1985
H\ V
reduced opioid binding
affinity
Judd et al. 1985
HO
ho-)^r
reduced opioid binding
affinity
This study
R= opioid peptide analog
of the Tyr1 ring of enkephalin analogs greatly increased opioid binding affinity. By
contrast, Judd et al. [Judd et al. 1985] demonstrated that D-Ala2 methionine enkephalin
amide analogs with a hydroxy group in the meta position rather than the para position of
the Tyr1 ring showed reduced affinity to (.ij, p2> 5 and k opioid receptors. In our studies, a
decrease in opioid binding affinity to both p and 6 receptors was observed when a hydroxy
group was introduced at the meta position of the Tyr1 ring of LE, in addition to the
hydroxy group already present at the para position. We presumed that the first
hydroxylation took place at the meta position as the hydroxy group in the para position

120
acts as a driving force for the electrophilic substitution. Therefore, in conjunction with the
observations of Judd et al. [Judd et al. 1985], our findings indicate that the presence of a
hydroxy group in the meta position rather than the absence of one in the para position
gives rise to a decrease in opioid receptor binding in enkephalins.
Tyrosinase and enkephalin immunoreactivities have been identified in isolated cells
[Kimura et al. 1992] and the spinal and brain regions in some species, namely the locus
coeruleus complex in the cat [Zhuo et al. 1992] and the ventral tegmental area in the rat
[Sesack and Pickel 1992], Although aminopeptidases, carboxypeptidases and
enkephalinases are thought to be mainly responsible for the metabolic deactivation of LE
[Venturelli et al. 1985], given that tyrosinase and LE have been found to co-exist in vivo,
and that the product of the reaction between these two entities shows decreased affinity to
both p and 8 opioid receptors compared to the parent enkephalin, we speculate that
tyrosinase may also contribute to the metabolic fate of LE in vivo.

CHAPTER 6
CONCLUSIONS
In the work carried out for this dissertation, the feasibilty of both an enzyme-linked
immunosorbent assay (ELISA) and an homogeneous fluorescence immunoassay for
leucine enkephalin (LE) was evaluated. Two high-performance liquid chromatography
(HPLC) approaches for opioid peptides using LE as a model peptide and enzymatic
derivatization by tyrosinase were also evaluated. One of these HPLC approaches used
electrochemical detection as a means of quantitation and the other used fluorescence
detection following a second fluorogenic derivatization step with 1,2-diamino-1,2-
diphenylethane (DPE). In addition, since hydroxylated products were produced from the
reaction between tyrosinase and leucine enkephalin in the development of the HPLC
approaches, and since tyrosinase and enkephalins have been found to co-exist in vivo, the
identity of these products was determined by mass spectrometry and their biological
activity in rat brain homogenate was investigated.
The formation of a sandwich between an anti-LE antibody, a biotinylated LE
derivative and avidin conjugated to an enzyme was a requirement for the successful
development of the ELISA proposed for this dissertation since in this assay, LE and
biotinylated LE should compete for antibody binding sites and antibody-bound biotinylated
LE should subsequently be detected through the use of enzyme-labeled avidin on addition
of the enzyme substrate. The formation of a sandwich could not be achieved using either
121

122
the antibody produced in this laboratory or the commercial antiserum tested and any of the
biotinylated derivatives synthesized. Therefore, a workable ELISA for LE could not be
developed using the proposed approach. ELIS As for |3-endorphin and dynorphin Al-13
have already been developed using the same strategy proposed here for the development
of an ELISA for LE [Hochhaus and Sadee 1988, Hochhaus and Hu 1990], However,
these opioid peptides consist of longer amino acid chains than LE (31 amino acids for (3-
endorphin and 13 amino acids for dynorphin compared to only 5 amino acids for LE) and
therefore, it is suggested that if a biotinylated LE derivative including a longer spacer arm
can be synthesized, a sandwich between anti-LE antibody, this biotinylated LE derivative
and avidin could be formed, leading to the successful development of an ELISA for LE
based on the same principles used in the ELISAs for fl-endorphin and dynorphin Al-13
which have been described earlier [Hochhaus and Sadee 1988, Hochhaus and Hu 1990],
The successliil development of the homogeneous fluorescence immunoassay
proposed in this dissertation relied on a lack of sandwich formation between an anti-LE
antibody, a biotinylated derivative and fluorescence-labeled avidin since here, LE and
biotinylated LE compete for antibody binding sites, and subsequently, free biotinylated LE
is detected on addition of fluorescence-labeled avidin due to the increase in fluorescence
observed on interaction between biotinylated LE and fluorescence-labeled avidin. A lack
of sandwich formation was seen using the anti-LE antibody produced in this laboratory, a
biotinylated derivative biotinylated at the N-terminal end of the peptide without a spacer
arm (BLE) and fluorescein isothiocyanate-labeled avidin (FITC-avidin). Although as
expected, an increase in fluorescence was seen in our homogenous fluorescence

123
immunoassay when increased amounts of LE were present in the sample, we observed that
at higher concentrations of LE (1*1 O'8 moles/ml and above), the analyte interacted directly
with FITC-avidin to produce an increase in fluorescence in the absence of BLE. [des-
Tyr']-LE and pentaglycine at the same concentrations did not produce an increase in
fluorescence in the presence of FITC-avidin, suggesting that the effect observed in the
homogeneous immunoassay at higher concentrations of LE was due to an interaction
between the tyrosine moiety of LE and the fluorescein groups of FITC-avidin. The use of
an antibody with higher affinity for LE may allow the successful development of this type
of homogenous fluorescence immunoassay for LE since less LE could be used in the assay
and therefore the direct interaction observed at high concentrations between LE and
FITC-avidin would be avoided. The use of laser-induced fluorescence may also contribute
to the succesfiil development of this type of assay since less FITC-avidin would be
required to produce a measurable signal on interaction with smaller quantities of BLE. As
a consequence, less LE would be required to displace sufficient BLE to produce a signal
on interaction with FITC-avidin and therefore, once again, high concentrations of LE
which interfere with the success of the assay would be avoided.
Two HPLC approaches for the determination opioid peptides were developed for
this dissertation, one with electrochemical detection and the other with fluorescence
detection. Both of these approaches involved the specific enzymatic derivatization of the
Tyr1 moiety of the model peptide LE by mushroom tyrosinase. This enzymatic
derivatization selectively introduced an hydroxy group to the Tyr1 moiety of the peptide,
thus presenting several analytical advantages. Firstly, the hydroxylation of LE rendered it

124
more easily oxidizable than the parent peptide, thus allowing electrochemical detection at
much lower oxidation potentials and therefore avoiding many of the disadvantages
associated with the use of high oxidation potentials such as reduced selectivity, high
background current, baseline drift and baseline noise. Secondly, the production of a
catechol permitted the selective clean-up of the analyte using boronate gels. Thirdly,
controlled oxidation of the hydroxylated derivative to give the corresponding quinone
enabled the use a secondary fluorogenic condensation reaction with DPE and subsequent
quantification the analyte by fluorescence detection.
The HPLC-ED and HPLC-FL approaches described in Chapters 3 and 4 yielded
limits of detection for LE in buffer samples of 170 fmol/inj and 500 fmol/inj, respectively
and 360 fmol/inj and 500 fmol/inj for LE in CSF, respectively. These limits of detection
compared favorably to existing F1PLC assays for opioid peptides with on-line detection.
The limits of detection for LE in CSF corresponded to 8.8 pmol/ml for the HPLC-ED
assay and 12 pmol/ml for the HPLC-FL assay. Endogenous levels of opioid peptides in
human CSF lie in the fmol/ml range and therefore, in their present state of development,
the HPLC approaches described here are inadequate for such determinations. However,
given sufficient sample (>1 ml), these approaches may be suitable for the determination of
elevated physiological levels to be expected in clinical studies.
The HPLC-ED approach described in Chapter 3 involved a single 5 minute
derivatization step prior to injection of the sample into the HPLC system compared to two
one hour derivatization steps for the HPLC-FL assay described in Chapter 4. The HPLC-
ED method also resulted in a slightly lower limit of detection for LE than the FIPLC-FL

125
method (170 fmol/inj in buffer and 360 fmol/inj in CSF compared to 500 fmol/inj in both
buffer and CSF) and yields a cleaner chromatograph with the peak of interest eluting after
a shorter retention time (6 minutes compared to 22.5 minutes). Therefore, of the two
HPLC approaches described in this dissertation, based on sensitivity, convenience and
practicality, the HPLC-ED assay is the method of choice for the analysis of opioid
peptides in CSF. It is proposed that increased sensitivity could be achieved using the
HPLC-FL approach by incorporating microbore HPLC and laser-induced fluorescence
into the procedure.
Hydroxylated derivatives of LE were produced using mushroom tyrosinase in the
HPLC assays for opioid peptides described in Chapters 3 and 4. In Chapter 5 the structure
of the major product of the reaction between LE and mushroom tyrosinase was
determined by electrospray ionization mass spectrometry to be [HO-Tyrl]-LE and the
minor product of the reaction was identified as [(HO)2-Tyr1]-LE. The affinity of [HO-
Tyr']-LE to opioid receptors in rat brain homogenate was determined by radioreceptor
assay. Hydroxylation of LE was found to decrease receptor affinity to both p and 5 opioid
receptor sites by a factor of about 20. Tyrosinase and enkephalin have been found to co
exist in isolated cells and in the spinal and brain regions of some species. Therefore, since
we have demonstrated that the product of the reaction between these two entities shows
decreased affinity to opioid receptors compared to the parent enkephalin, we speculate
that tyrosinase may play a role in the metabolic pathway of LE in vivo.

APPENDIX A
DATA FOR HPLC-ED APPROACH
Curve 1
Tyrosinase derivatization, boronate clean up
Matrix: Buffer
Equation: y = 2761 lx + 2288, r2 = 0.9977
LE mol/inj
Peak area
0
0
5.33*1013
17597
1.07* 10"12
32566
2.13*1 O'12
64947
3.20* 10'12
87745
4.80*1 O'12
134842
Curve 2
Tyrosinase derivatization, boronate clean up
Matrix: Buffer
Equation: y = 30648x -1752, r2 = 0.9876
LE mol/inj
Peak area
0
0
3.37* 10'13
5077
5.06* 10"13
15565
8.44* 10'13
17240
1.18*10'12
40429
1.69*1 O'12
52711
2.53 1012
74861
3.37* 10'12
100722
126

127
Curve 3
Tyrosinase derivatization, boronate clean up
Matrix: Buffer
Equation: y = 25953x + 3341, r2 = 0.9922
LE mol/inj
Peak height (mm)
0
0
1.68* 10'13
9573
3.37* 1 O'13
14456
5.06* 1013
15197
1.69*1 O12
48918
2.53*10"12
67658
Curve 4
Tyrosinase reaction, boronate clean up
Matrix: CSF
Equation: y = 21753x 12089, r2 = 0.9957
LE mol/inj
Peak area
0
11724
3.60* 10'13
11205
7.22*10'13
30931
1.42*1 O'12
43348
2.73* 1012
74283
4.29* 10'12
109291
7.01 10"12
169232
9.63* 10'12
215784

128
Curve 5
Tyrosinase reaction, boronate clean up
Matrix: CSF
Equation: y = 21179x 16355, r2 = 0.9904
LE mol/inj
Peak area
0
21412
3.60* 10'13
7743f
7.22* 1013
29865
1.42*1012
42426
2.73*1012
65917
4.29* 10'12
111943
7.01 1 O12
176539
9.63* 10'12
212826
toullier, calculated concentration deviates from nominal concentration
by >20% (H. T. Karnes, personal communication)
Curve 6
Tyrosinase reaction, boronate clean up
Matrix: CSF
Equation: y = 23240x 10701, r2 = 0.9971
LE mol/inj
Peak area
0
10509
3.60*10'13
15452
7.22* 10'13
28020
1.42* 10'12
44913
2.73 1 O'12
73904
4.29*1012
116036
7.01 10"12
170143

129
Curve 7
No derivatization, no boronate clean up
Matrix: CSF
Equation: y = 10319x 44657, r2 = 0.9770
LE mol/inj
Peak area
0
0
8.82* 10'12
93356
17.28* 1 O'12
113333
25.56* 1012
156482
33.30* 10'12
230196
43.2*1012
422738
69.29*1 O'12
714440
85.67* 10'12
841176
Curve 8
No derivatization, no boronate clean up
Matrix: CSF
Equation: y = 9006x 10275, r2 = 0.9570
LE mol/inj
Peak area
0
0
17.28* 1 O12
149435
25.56* 1 O'12
217441
33.30*10"12
193110
43.2* 10'12
463885
69.29* 1012
660135
85.67* 10'12
720965

130
Curve 9
No derivatization, no boronate clean up
Matrix: CSF
Equation: y = 11326x + 59177, r2 = 0.9529
LE mol/inj
Peak area
0
0
8.82*10'12
259029
17.28* 10'12
2408431
25.56* 1 O'12
265387
33.30* 1012
231334f
43.2* 1012
646006
69.29*1 O'12
762035
85.67* 1012
1064494
f outlier

APPENDIX B
DATA FOR HPLC-FL APPROACH
Curve 1
Non-concentrated reagents
Matrix: Buffer
Equation: y = 1.79*1012x + 0.4064, r2 = 0.9966
LE mol/inj
Peak area
0
0
5.62* 10'13
1.5338
1.12*10"12
2.7748
2.23* 10'12
4.0617
4.38*10'12
8.5962
8.76* 10'12
15.9292
Curve 2
Non-concentrated reagents
Matrix: buffer
Equation: y = 1.14*1013x + 2.2749, r2 = 0.9836
LE mol/inj
Peak height (mm)
0
0
8.91 10"13
13
1.80* 10"12
26
3.59* 1012
39
5.39* 10'12
70
7.18*1012
81
131

132
Curve 3
Non-concentrated reagents
Matrix: buffer
Equation: y = 3.40* 1013x 1.3522, r2 = 0.9971
LE mol/inj
Peak height (mm)
0
0
2.81*10"13
9
3.65* 10*13
lost
5.62* 10'13
15
1.12* 10"12
37
2.23 1 O'12
75
Curve 4
Non-concentrated reagents
Matrix: CSF
Equation: y = 3.89* 10,2x 0.6900, r2 = 0.9975
LE mol/inj
Peak area
0
0
4.97*1013
0.6753
8.86* 1013
2.7875
2.12* 1 O'12
7.7407
3.96*1 O'12
14.0380
7.24*10'12
27.8690

133
Curve 5
Non-concentrated reagents
Matrix: CSF
Equation: y = 1.08* 1013x 0.7695, r2 = 0.9670
LE mol/inj
Peak height (mm)
0
0
8.91 10'13
12
1.78* 10"12
17
5.38* 1012
48
7.18*10"12
84
Curve 6
Concentrated reagents
Matrix: Buffer
Equation: y = 2.25* 1012x 0.8243, r2 = 0.9926
LE mol/inj
Peak area
0
0
4.47* 10'13
1.14
1.79* 10"12
7.151
6.68*10'12
14.916
1.67*10'12
38.496

134
Curve 7
Concentrated reagents
Matrix: Buffer
Equation: y = 5.09*1012x 0.8542, r2 = 0.9984
LE mol/inj
Peak area
0
0
5.01*10'13
1.1262
l*10-i2
4.1923
2*1 O'12
8.9492
3*10-12
14.35
4.67* 10'12
28.442f
6.68* 1012
33.319
¡outlier, calculated concentration deviates from nominal concentration
by >20% (H. T. Karnes, personal communication)
Curve 8
Concentrated reagents
Matrix: CSF
Equation: y = 9.91 10ux + 0.5776, r2 = 0.9925
LE mol/inj
Peak area
0
0.79768
5*1 O'13
1.0174
1*10*2
1.3522
2*1012
lost
3*1 O'12
5.2685
4.68* 1012
3.0910f
foullier

135
Curve 9
Concentrated reagents
Matrix: CSF
Equation: y = 1.76*1012x + 3.0077, r2 = 0.9853
LE mol/inj
Peak area
0
2.2941
4.69*10'13
3.5183
8.85* 10'13
2.4436|
1.95* 1012
7.0437
2.88* 1012
3.7482f
4.41*1 O12
10.505
toutlier
Curve 10
Concentrated reagents
Matrix: CSF
Equation: y = 3.45*1012x + 1.0393, r2 = 0.9975
LE mol/inj
Peak area
0
1.4506
4.69* 10'13
2.6145
8.85* 1013
4.3134
1.95*1 O'12
4.0743f
2.88* 10'12
9.3891
4.41 10"12
17.243
foutlier

136
Curve 11
Concentrated reagents, 5 minute tyrosinase reaction
Matrix: Buffer
Equation: y = 7.67* 1012x + 4.9244, r2 = 0.9922
LE mol/inj
Peak height (mm)
0
0
2.77* 10'13
10.5
5.54* 10"13
11.5
1.38*10"12
18
2.77*10'12
24
5.54* 10"12
44.2
8.31*10"12
64.5
1.38*10'"
115
Curve 12
Concentrated reagents, 5 minute tyrosinase reaction
Matrix: Buffer
Equation: y = 2.50* 1012x + 4.449, r2 = 0.9779
LE mol/inj
Peak height (mm)
0
3.5
2.77*1013
4.5
5.54* 10'13
7
1.38* 10'12
8.5
2.77*1012
11.5
5.54* 1012
18

137
Curve 13
Concentrated reagents, 5 minute tyrosinase reaction
Matrix: Buffer
Equation: y = 9.91*1012x + 6.4066, r2 = 0.9985
LE mol/inj
Peak height (mm)
0
6
2.77* 10'13
10.5
5.54*1 O13
13
1.38*1012
16.5
2.77* 1012
lit
5.54* 1 O12
63.5
6.64* 1012
43.5|
1.11*10'"
83 f
1.38*10'"
143
toutlier

REFERENCES
Akil, H., D. Bronstein and A. Mansour (1988). Overview of the Endogenous Opioid
Systems: Anatomical, Biochemical and Functional Issues, in Endorphins, Opiates and
Behavioural Processes (pg. 1) R. J. Rogers, & S. J. Cooper (Eds.). Chichester: John Wiley
& Sons.
Al-Hakiem, M. H. H., J. Landon, D. S. Smith and R. D. Nargessi (1981). Analytical
Biochemistry 116: 264.
Allen, M. H. and M. L. Vestal (1992). Journal of the American Society of Mass
Spectrometry 3: 18.
Arnott, D., J. Shabanowitz and D. F. Hunt (1993). Clinical Chemistry 39: 2005.
Biemann, K. (1992). Annual Reviews in Biochemistry 61: 977.
Bowden, D. W., M. Rising, G. Akots, A. Myles and R. J. Broeze (1986). Clinical
Chemistry 32: 275.
Carr, S. A. (1990). Advanced Drug Delivery Reviews 4: 113.
Clement-Jones, V. and G. M. Besser (1983). British Medical Bulletin 39: 95.
Clement-Jones, V. and L. H. Rees (1982). in Clinical Neuroendocrinology (vol. II, pg.
139) G. M. Besser, & L. Martin (Eds ). New York: Academic Press.
Cory, J. G. and E. Frieden (1967a). Biochemistry 6: 121.
Cory, J. G. and E. Frieden (1967b). Biochemistry 6: 116.
Cory, J. G., C. G. Bigelow and E. Frieden (1962). Biochemistry 1: 419.
Coty, W. A., R. Loor, N. Bellet, P. L. Khanna, P. Kaspar and M. Baier (1992). Wiener
Klinische Wochenschrift Supplement 191: 5.
Dass, C., J. J. Kusmierz, D. M. Desiderio, S. A. Jarvis and B. N. Green (1991). Journal of
the American Society for Mass Spectrometry 2. 149.
138

139
Dave, K. J., J. F. Stobaugh, T. M. Rossi and C. M. Riley (1992). Journal of
Pharmaceutical and Biomedical Analysis 10: 965.
de Ceballos, M, M. D. Taylor and P. Jenner (1991). Neuropeptides 20: 201.
de Montigny, P., C. M. Riley, L. A. Sternson and J. F. Stobaugh (1990). Journal of
Pharmaceutical and Biomedical Analysis 8: 419.
Defrutos, M. and F. E. Regnier (1993). Analytical Chemistry 65: A17.
Eisenach, J. C., C. 2. Dobson, C. E. Inturrisi, D. D. Hood and P. B. Agner (1990). Pain
43: 149.
Engel, W. D. and P. Khanna (1992). Journal of Immunological Methods 150: 99.
Eriksson, B. and M. Wikstrom (1992). Journal of Chromatography 593: 185.
Finn, F. M., G. Titus and K. Hofmann (1984). Biochemistry 23: 2554.
Fleming, L. H. and N. J. Reynolds (1988). Journal of Chromatography 431: 65.
Garcia, S. F., A. Navas and J. Lovillo (1993). Analytical Biochemistry 214: 359.
Gemant, A. (1974). Molecular Biology Reports 1: 423.
Gosling, J. P. (1990). Clinical Chemistry 36: 1408.
Green, N. M. (1970). Methods inEnzymology 18A: 418.
Green, N. M., L. Konieczny, E. J. Toms and R. C. Valentine (1971). Proceedings of the
National Academy of Sciences 74: 2697.
Grossman, A. and L. H. Rees (1983). British Medical Bulletin 39: 83.
Handa, B. K., A. C. Lane, J. A. H. Lord, B. A. Morgan, M. J. Ranee and C. F. C. Smith
(1981). European Journal of Pharmacology 70: 531.
Hansen, D. W., R. H. Mazur and M. Clare (1985). The Synthesis and SAR of Orally
Active Enkephalin Analogs with Modified N-terminal Tyrosine Residues, in Peptides:
Structure and Function (pg. 491) D. M. Deber, V. J. Hruby, & K. D. Kopple (Eds.).
Rockford, IL: Pierce Chemical Company.
Hansen, P. E. and B. A. Morgan (1984). Structure-Activity Relationships in Enkephalin
Peptides, in Opioid Peptides: Biology, Chemistry and Genetics (vol. 6, pg. 269) S.
Udenfriend, & J. Meienhofer (Eds.). Orlando: Academic Press Inc.

140
Hansson, C., H. Rorsman and E. Rosengren (1981). Acta Dermatovenerologica
(Stockholm) 61: 147.
Hardebo, J. E., R. Ekman and M. Eriksson (1989). Headache 29: 494.
Harlow, E. and D. Lane (1988), Antibodies: A Laboratory Manual. Cold Spring Harbor:
Cold Spring Harbor Laboratory.
Heerden, J. E. Ahlskog and D. E. Byer (1990). Journal of Laboratory and Clinical
Medicine 115: 346.
Higa, S., T. Suzuki, A. Hayashi, I. Tsuge and Y. Yamamura (1977). Analytical
Biochemistry 77: 18.
Ho, R. J. Y. and L. Huang (1985). Journal of Immunology 134: 4035.
Hochhaus, G. and L. Hu (1990). Journal of Pharmaceutical and Biomedical Analysis 8:
541.
Hochhaus, G. and W. Sadee (1988). Pharmaceutical Research 5: 232.
Hochhaus, G., B. W. Gibson and W. Sadee (1988). Journal of Biological Chemistry 263:
92.
Hughes, J., T. W. Smith, H. W. Kosterlitz, L. A. Fothergill, B. A. Morgan and H. R.
Morris (1975). Nature 258: 577.
Ihara, H., T. Imamura, T. Kunitake and C. Hirayama (1988). Clnica Chimica Acta 175:
51.
Imura, H., Y. Kato, Y. Nakai, K. Nakao, I. Tanaka, H. Jingami, T. Koh, T. Yoshimasa, T.
Tsukada, M. Suda, M. Sakamoto, N. Morii, H. Takahashi, K. Tojo and A. Sugawara
(1985). Journal of Endocrinology 107: 147.
Irving, M. N. H. and R. M. Parkins (1965). Journal of Inorganic and Nuclear Chemistry
27: 270.
Ishida, J., M. Kai and Y. Ohkura (1986). Journal of Chromatography 356: 171.
Ito, S., T. Kato, K. Shinpo and K. Fujita (1984). Biochemistry Journal 222: 407.
Jaflfe, J. H. and W. R. Martin (1990). Opioid Analgesics and Antagonists, in The
Pharmacological Basis of Therapeutics (pp. 486) A. G. Goodman-Gilman (Ed.). New
York: Pergamon Press.

141
Jenkins, S. H. (1992). Journal of Immunological Methods 150: 91.
Judd, A. K., L. R. Toll, J. A. Lawson, E. T. Uyeno, W. E. Polgar and G. H. Loew (1985).
Putative Opioid antagonists: Synthesis and Biological properties of D-Ala2 Met-Enk-
Amide Analogs with Unusual Tyr Residues, in Peptides: Structure and Function (pg. 499)
D. M. Deber, V. J. Hruby, & K. D. Kopple (Eds ). Rockford, IL: Pierce Chemical
Company.
Kabanov, A. V., M. M. Khrutskaya, S. A. Eremin, N. L. Klyachko and A. V. Levashov
(1989). Analytical Biochemistry 181: 145.
Kai, M., J. Ishida and Y. Ohkura (1988). Journal of Chromatography 430: 271.
Katta, V., S. K. Chowdhury and B. T. Chait (1991). Analytical Chemistry 63: 174.
Kim, C., R. Cheng and S. R. George (1989). Journal of Chromatography 494: 67.
Kimura, N., H. Yamamoto, H. Okamoto, K. Gotoh, M. Son, T. Mouri, K. Ota, T.
Kimura, T. Ohzeki and Y. Miura (1992). Virchows Archiv B Cell Pathology Including
Molecular Pathology 62: 321.
Koike, K., T. Aono, F. Chatani, T. Takemura and K. Kurachi (1982). Life Sciences 30:
2221.
Koob, G. F. and F. E. Bloom (1983). British Medical Bulletin 39: 89.
Kricka, L. J. (1993). Clinical Biochemistry 26: 325.
Kuhling, P., B. Siegfried, H. R. Frischknecht, F. S. Messiha and A. Pasi (1989).
Physiological Behaviour 46: 25.
Kusmierz, J. J., R. Sumrada and D. M. Desiderio (1990). Analytical Chemistry 62: 2395.
Lehman, W. D. (1982). Analytical Chemistry 54: 299.
Lunte, S. and O. Wong (1989). LC-GC 7: 908.
Maidment, N. T., D. R. Brumbaugh, V. D. Rudolph, E. Erdelyi and C. J. Evans (1989).
Neuroscience 33: 549.
Mansour, A., H. Khachaturian, M. E. Lewis, H. Akil and J. Watson (1988). Trends in
Neurosciences 11: 308.
Marumo, K. and J. H. Waite (1986). Biochimica et Biophysica Acta 872: 98.

142
McDermott, J. R., A. I. Smith, J. A. Biggins, M. C. Al-Noaemi and J. A. Edwardson
(1981). Journal of Chromatography 222: 371.
McQueen, D. S. (1983). British Medical Bulletin 39: 77.
Merchenthaler, I. (1993). Endocrinology 133: 2645.
Mifune, M., D. K. Krehbiel, J. F. Stobaugh and C. M. Riley (1989). Journal of
Chromatography 496: 55.
Mitsui, A., H. Nohta and Y. Ohkura (1985). Journal of Chromatography 344: 61.
Monger, L. S. and C. J. OllifT (1992). Journal of Chromatography 577: 239.
Mosberg, H. I., R. Hurst, V. J. Hruby, K. Gee, H. I. Yamamura, J. J. Galligan and T. F.
Burks (1983). Proceedings of the National Academy of Sciences 80: 5871.
Mousa, S. and D. Couri (1983). Journal of Chromatography 267: 191.
Muck, W. M. and J. D. Henion (1989). Journal of Chromatography 495: 41.
Murgo, A. J., R. E. Faith and N. P. Plotnikoff (1986). Enkephalins: Mediators of Stress
Induced Immunomodulation, in Enkephalins and Endorphins, Stress and the Immune
System (pg. 221) A. J. Murgo, R. E. Faith, N. P. Plotnikoff, & R. A. Good (Eds.). New
York: Plenum Press.
Nakano, M., M. Olmo, M. Kai and Y. Ohkura (1987). Journal of Chromatography 411:
305.
Nohta, H., A. Mitsui and Y. Ohkura (1984). Analytica Chimica Acta 165: 171.
Olson, G. A., R. D. Olson and A. J. Kastin (1991). Peptides 13: 1247.
Pasternak, G. W. (1982). Life Sciences 31: 1301.
Pleuvry, B. J. (1991). British Journal of Anaesthesia 66: 370.
Plotnikoff, N. P., A. J. Murgo, G. C. Miller, C. N. Corder and R. E. Faith (1985). Federal
Proceedings 44: 118.
Porstmann, T. and S. T. Kiessig (1992). Journal of Immunological Methods 150: 5.
Rapaka, R. S. (1986). Life Sciences 39: 1825.
Rapaka, R. S. and F. Porreca (1991). Pharmaceutical Research 8: 1.

143
Roepstorff, P. and J. Fohlman (1984). Biomedical Mass Spectrometry 11: 601.
Rosei, M. A., A. Antonilli, R. Coccia and C. Blarzino (1991). Physiological, Chemical,
Physical and Medical NMR 23: 59.
Rosei, M. A., L. Antonilli, R. Coccia and C. Foppoli (1989). Biochemistry International
19: 1183.
Roth, M. (1971). Analytical Chemistry 43: 880.
Rubin, P. C. (1984). Clinical Science 66: 625.
Samuelsson, H., R. Ekman and T. Hedner (1993). Acta Anaesthesiologica Scandinavica
37: 502.
Sarma, J. K. S. R., Hoffmann and R. A. Houghten (1986). Life Sciences 38: 1723.
Sato, H. (1984). Life Sciences 35: 1051.
Scatchard, G. (1949). Annals of the New York Academy of Sciences 51: 660.
Schulman, S., G. Hochhaus and H. T. Karnes (1990). Fluorescence Immunoassay, in
Luminescence Spectroscopy (pg. 341) W. R. G. Baeyens (Ed.). New York: Marcel
Dekker.
Sesack, S. R. and V. M. Pickel (1992). Journal ofNeuroscience 12: 1335.
Shibanoki, S., S. B. Weinberger, K. Ishikawa and J. Martinez (1990). Journal of
Chromatography 532: 249.
Shimohigashi, Y. (1986 ). NIDA Research Monograph 69: 65.
Simon, E. J. (1984). Opiate Receptors, in Receptors in the Nervous System (vol. 6, pg.
331) A. Lajtha (Ed.). New York: Plenum Press.
Simonds, W. F. (1988). Endocrine Reviews 9: 200.
Smith, R. D., J. A. Loo, C. J. Barinaga, C. G. Edmonds and H. R. Udseth (1990). Journal
of the American Society of Mass Spectrometry 1: 53.
Szilagyi, J. E. (1989). Clinical and Experimental Hypertension Theory and Practice
All: 1.

144
Tellier, M, R. J. Prankerd and G. Hochhaus (1991). Journal of Pharmaceutical and
Biomedical Analysis 9: 557.
Udenfriend, S., S. Stein, P. Bohlen, W. Dairman, W. Leimgruber and M. Weigele (1972).
Science 178: 871.
Umeda, M, Y. Ishimori, K. Yoshikawa, M. Takada and T. Yasuda (1986). Journal of
Immunological Methods 95: 15.
van den Beld, C. M. B., U. R. B. Tjaden, N. J. Reinhoud, D. S. Stegehuis and J. van der
Greef (1990). Journal of Controlled Release 13: 129.
Venn, R. F. (1987). Journal of Chromatography 423: 93.
Venturelli, F., G. Roscetti, R. Possenti, F. Vita and G. Roda (1985). Neurochemistry
Research 10: 333.
Walsh, C. (1979). in Enzymatic Reaction Mechanisms (pg. 461) San Francisco: W. H.
Freeman & Co.
Wang, Z., E. J. Bilsky, F. Porreca and W. Sadee (1994). Life Sciences 54: 339.
Whitehouse, C. M., R. N. Dreyer, M. Yamashita and J. B. Fenn (1985). Analytical
Chemistry 57: 675.
Wilchek, M. and E. A Bayer (1988). Analytical Biochemistry 171: 1.
Yaksh, T. L., S. W. Carmichael, S. L. Stoddard, G. M. Tyce, P. J. Kelly, D. Lucas, J. van
Heerden, J. E. Ahlskog and D. E. Byer (1990). Journal of Laboratory and Clinical
Medicine 115: 346.
Young, R. F., F. W. Bach, N. A. Van and T. L. Yaksh (1993). Journal of Neurosurgery
79: 816.
Zamboni, G., C. A. Jones and J. Hughes (1983). Analytical Biochemistry 130: 83.
Zhang, G., M. Kai, M. Nakano and Y. Ohkura (1991). Chemical and Pharmaceutical
Bulletin 39: 126.
Zhuo, H., S. J. Fung, V. K. Reddy and C. D. Barnes (1992). Journal of Chemical
Neuroanatomy 5: 1.

BIOGRAPHICAL SKETCH
Veronique Larsimont was born in Taipei, Taiwan in 1967 and spent her childhood
traveling the world with her parents until she went to boarding school at Dollar Academy,
Dollar, Scotland in 1979. She earned a B.Sc. (Honours) in pharmacy from Heriot-Watt
University, Edinburgh, Scotland in 1988 and worked for Lilly Research and Development
in Windlesham, Surrey, England and Richard Clitherow Ltd., Liverpool, England during
her pre-registration year prior to becoming a member of the Royal Pharmaceutical Society
of Great Britain in 1989. In the same year, she began her graduate studies in the
Department of Pharmaceutics at the University of Florida and completed her doctoral
dissertation in 1994. She now intends to pursue a career in the pharmaceutical industry in
Europe.
145

I certify that I have read this study and that in my opinion it conforms to
acceptable standards of scholarly presentation and is fully adequate, in scope and quality,
as a dissertation for the degree of Doctor of Philosophy.
c.
Giinther Hochhaus, Chair
Associate Professor of Pharmaceutics
I certify that I have read this study and that in my opinion it conforms to
acceptable standards of scholarly presentation and is fully adequate, in scope and quality,
as a dissertation for the degree of Doctor of Philosophy.
Hartmut Derendorf
Professor of Pharmaceutics
I certify that I have read this study and that in my opinion it conforms to
acceptable standards of scholarly presentation and is fully adequate, in scope and quality,
as a dissertation for the degree of Doctor of Philosophy.
Paul Klein
Professor of Pathology and
Laboratory Medicine
I certify that I have read this study and that in my opinion it conforms to
acceptable standards of scholarly presentation and is fully adequate, in scope and quality,
as a dissertation for the degree of Doctor of Philosophy.
Laszlo Prokai
Assistant Professor of Pharmaceutics
I certify that I have read this study and that in my opinion it conforms to
acceptable standards of scholarly presentation and is fully adequate, in scope and quality,
as a dissertation for the degree of Doctor of Philosophy.
3 /frUT-
Ian Tebbett
Associate Professor of Pharmaceutics

This dissertation was submitted to the Graduate Faculty of the College of
Pharmacy and to the Graduate School and was accepted a partial fulfillment of the
requirements for the degree of Doctor of Philosophy.
August 1994
Dean, Graduate School




Ill
Ascorbic acid was included in the reaction mixture to prevent further oxidation of our
hydroxylated products to the corresponding quiones.
1.4
A
1.4
B
Figure 5.2. Chromatographs of control incubation (A) showing LE eluting after 9.5
minutes, distinct form ascorbic acid eluting in the solvent front after 1.4
minutes, and a test incubation after 30 minutes (B) showing the emergence of
two new peaks with retention times of 5.7 and 7.3 minutes.


43
the protein carrier will conjugate to LE via the N-terminal amino group of the peptide and
LE will also conjugate to the protein carrier through its C-terminal end by attacking the
free amino groups on the protein carrier (Figure 2.4). As a result, on inoculation with a
conjugate produced by the EDC method, one could expect antibodies directed against
either or both the N-terminal or the C-terminal end of LE.
Table 2.4. Loading ratios of LE to protein carrier by different conjugation methods.
Protein carrier
Conjugation method
LE : protein ratio
Porcine thyroglobulin
Glutaraldehyde
545 : 1
Porcine thyroglobulin
EDC
70 : 1
Bovine serum albumin
Glutaraldehyde
60 : 1
Bovine serum albumin
EDC
1.37 : 1
The loading ratios of peptide to protein carrier calculated from the results of amino
acid analysis are shown in Table 2.4. Due to the high immunogenicity of porcine
thyroglobulin in rabbits and the fact that the LE-thyroglobulin conjugate obtained by the
glutaraldehyde method showed the highest peptide loading ratio, this conjugate was
selected as the inoculum for the purposes of this work.
The results of the radioimmunoassay carried out after the third boost showed that
when a 1/1,000 dilution of the antiserum was used, 30% of the total radioactivity added
was bound and 32% of this bound radioactivity was determined to be due to non-specific
binding (Figure 2.5). At that time, this titer was deemed to be sufficient and therefore the
antiserum was harvested.


99
Figure 4.7. Chromatograph of LE in CSF using HPLC-FL method. LE is shown eluting
after 22.5 minutes.
An analytical advantage was not gained through the concentration of the
fluorogenic reagents and subsequent injection of a greater portion of the total final
reaction mixture as the same limit of detection of 500 fmol/injection could be obtained
when this modification in our method was attempted.
Although a lower limit of detection in buffer samples (277 fmol/injection) was
obtained when a 5 minute reaction time instead of a 60 minute reaction time was used for
the tyrosinase reaction (Figure 4.8), when this modification of the method was attempted
in spiked CSF samples, the linearity of the calibration curve could not be preserved (data


93
nm and Xem 480 nm with slit widths of 12.5 nm. A BD 41 chart recorder (Kipp and Zonen,
Delft, The Netherlands) and a Hewlett Packard HP 3394A integrator (Avondale, PA,
USA) recorded the output from the spectrofluorodetector. The mobile phase contained 32
% acetonitrile (v/v) and 67 mM TBA in citrate/dipotassium phosphate buffer (pH 5). The
system was operated at a mobile phase flow rate of 1 ml/min.
An attempt was made to optimize the fluorogenic derivatization reaction by
concentrating the fluorogenic reagents so that 100 pi of an oxidizing solution containing
40.1 mg/ml of potassium ferricyanide and 75.8 mg/ml of potassium chloride, 100 pi of
acetonitrile and 25 pi of a solution of DPE containing 20 mg/ml in 0.1 N HC1 were added
to the buffer samples or the CSF samples reconstituted in buffer. After a 60 minute
incubation in the dark with constant shaking, a 250 pi aliquot of this mixture, representing
77 % of the total final reaction mixture, was injected into the HPLC system described
above.
Since experiments conducted in parallel indicated that at low concentrations of
analyte, the highest yield of enzymatically derived species was obtained after a 5 minute
tyrosinase incubation time (see Chapter 3), an attempt was also made to optimize the
derivatization reaction by reducing the tyrosinase incubation time to 5 minutes.
Results
Enzymatic Derivatization
Preliminary experiments indicated that the concentration of mushroom tyrosinase
(135 units/ml) used for the enzymatic derivatization of LE allowed the efficient conversion


19
determination of enkephalins in the 100 fmol/inj range [Muck and Henion 1989], Levels as
low as 5 fmol of P-endorphin and 1 pinol of ME have been quantified by electrospray
mass spectrometry with off-line HPLC [Dass and Kusmierz 1991] and pmol amounts of
ME have been detected by fast atom bombardment mass spectrometry with off-line HPLC
[Kusmierz and Sumrada 1990],
Objectives
Appropriate analytical methodology for opioid peptides is required for use in
clinical, pharmacokinetic and formulation studies as well as in physiological studies to
allow the successful development of these entities as therapeutic agents and the
continuation of research to elucidate further the physiological role of these compounds.
The need for accurate, specific, sensitive and reproducible analytical methods for opioid
peptides has been stressed by representatives of the National Institute on Drug Abuse
[Rapaka 1986], However, to date no analytical procedure has emerged which adequately
meets all of these criteria.
Therefore, the focus of the work carried out for this dissertation has been the
evaluation of several novel analytical approaches for the determination of leucine
enkephalin (LE, Figure 1.5) as a model for opioid peptides. Leucine enkephalin was
chosen as a model peptide as it contains the same initial sequence common to all opioid
peptides.
The first approach evaluated was a non-homogeneous enzyme-linked
immunosorbent assay (ELISA) which employed the same strategy as ELISAs which have


61
IE-11 IE-10 IE-09 IE-08 IE-07 IE-06
LE (inoles/ml)
Figure 2.18. Calibration curve for LE using the homogeneous fluorescence immunoassay
set up with BLE () and without BLE(D).
IE-11 IE-10 IE-09 IE-08 IE-07 IE-06
LE(molcs/nil)
Figure 2.19. Calibration curve for LE using the homogeneous fluorescence immunoassay
set up, omitting antibody but with BLE () and without BLE(D).
Figure 2.18 shows that at in the presence of the antibody, when the tracer (BLE) is
omitted from the assay, fluorescence readings began to increase above the fluorescence


124
more easily oxidizable than the parent peptide, thus allowing electrochemical detection at
much lower oxidation potentials and therefore avoiding many of the disadvantages
associated with the use of high oxidation potentials such as reduced selectivity, high
background current, baseline drift and baseline noise. Secondly, the production of a
catechol permitted the selective clean-up of the analyte using boronate gels. Thirdly,
controlled oxidation of the hydroxylated derivative to give the corresponding quinone
enabled the use a secondary fluorogenic condensation reaction with DPE and subsequent
quantification the analyte by fluorescence detection.
The HPLC-ED and HPLC-FL approaches described in Chapters 3 and 4 yielded
limits of detection for LE in buffer samples of 170 fmol/inj and 500 fmol/inj, respectively
and 360 fmol/inj and 500 fmol/inj for LE in CSF, respectively. These limits of detection
compared favorably to existing F1PLC assays for opioid peptides with on-line detection.
The limits of detection for LE in CSF corresponded to 8.8 pmol/ml for the HPLC-ED
assay and 12 pmol/ml for the HPLC-FL assay. Endogenous levels of opioid peptides in
human CSF lie in the fmol/ml range and therefore, in their present state of development,
the HPLC approaches described here are inadequate for such determinations. However,
given sufficient sample (>1 ml), these approaches may be suitable for the determination of
elevated physiological levels to be expected in clinical studies.
The HPLC-ED approach described in Chapter 3 involved a single 5 minute
derivatization step prior to injection of the sample into the HPLC system compared to two
one hour derivatization steps for the HPLC-FL assay described in Chapter 4. The HPLC-
ED method also resulted in a slightly lower limit of detection for LE than the FIPLC-FL


33
Table 2.1. Samples set up for Scatchard analysis.
Total counts
Total binding
Non-specific binding
Buffer
420 pi
20 pi
3H-LE
20 pi
20 pi
20 pi
LE( 1.4*10'5 M)
20 pi
Antibody (1/1,000)
200 pi
200 pi
Overnight at 4C, then:
Charcoal (1.5%)/dextran (0.15%)
200 pi
200 pi
A Scatchard plot was constructed by plotting the ratio of bound to free 3H-LE
against bound 3H-LE and the Ka value was determined from the negative slope of the line
drawn between the points. The number of specific antibody sites was determined from the
intercept of this line with the x-axis.
Biotinylation of Leucine Enkephalin
N-terminal biotinylated LE (BLE) was synthesized by allowing 90 nmoles of LE,
180 nmoles of N-hydroxysuccinimidobiotin (BHS) and 120 nmoles of triethanolamine in
150 pi of dimethylsulfoxide (DMSO) to incubate overnight at room temperature. The
product of the reaction was separated from the reagents by injecting the incubation
mixture into a gradient HPLC system consisting of a Rainin Rabbit HP solvent delivery
system (Rainin Instrument Company Inc., Woburn, MA) controlled by a Rainin Dynamax
HPLC method manager (version 1.3, Rainin Instrument Company Inc., Woburn, MA), a
Negretti & Zamba injector (Southampton, UK) fitted with a 100 pi loop, a pBondapak
Ci8 column (10 pm, 3.9 x 150 mm, Waters Associates, Milford, MA, USA) and a LDC


141
Jenkins, S. H. (1992). Journal of Immunological Methods 150: 91.
Judd, A. K., L. R. Toll, J. A. Lawson, E. T. Uyeno, W. E. Polgar and G. H. Loew (1985).
Putative Opioid antagonists: Synthesis and Biological properties of D-Ala2 Met-Enk-
Amide Analogs with Unusual Tyr Residues, in Peptides: Structure and Function (pg. 499)
D. M. Deber, V. J. Hruby, & K. D. Kopple (Eds ). Rockford, IL: Pierce Chemical
Company.
Kabanov, A. V., M. M. Khrutskaya, S. A. Eremin, N. L. Klyachko and A. V. Levashov
(1989). Analytical Biochemistry 181: 145.
Kai, M., J. Ishida and Y. Ohkura (1988). Journal of Chromatography 430: 271.
Katta, V., S. K. Chowdhury and B. T. Chait (1991). Analytical Chemistry 63: 174.
Kim, C., R. Cheng and S. R. George (1989). Journal of Chromatography 494: 67.
Kimura, N., H. Yamamoto, H. Okamoto, K. Gotoh, M. Son, T. Mouri, K. Ota, T.
Kimura, T. Ohzeki and Y. Miura (1992). Virchows Archiv B Cell Pathology Including
Molecular Pathology 62: 321.
Koike, K., T. Aono, F. Chatani, T. Takemura and K. Kurachi (1982). Life Sciences 30:
2221.
Koob, G. F. and F. E. Bloom (1983). British Medical Bulletin 39: 89.
Kricka, L. J. (1993). Clinical Biochemistry 26: 325.
Kuhling, P., B. Siegfried, H. R. Frischknecht, F. S. Messiha and A. Pasi (1989).
Physiological Behaviour 46: 25.
Kusmierz, J. J., R. Sumrada and D. M. Desiderio (1990). Analytical Chemistry 62: 2395.
Lehman, W. D. (1982). Analytical Chemistry 54: 299.
Lunte, S. and O. Wong (1989). LC-GC 7: 908.
Maidment, N. T., D. R. Brumbaugh, V. D. Rudolph, E. Erdelyi and C. J. Evans (1989).
Neuroscience 33: 549.
Mansour, A., H. Khachaturian, M. E. Lewis, H. Akil and J. Watson (1988). Trends in
Neurosciences 11: 308.
Marumo, K. and J. H. Waite (1986). Biochimica et Biophysica Acta 872: 98.


70
Methods
Purification of Mushroom Tyrosinase
Mushroom tyrosinase was purified prior to use by ultra-filtration using Centricon
membrane filters (molecular weight cut off 30,000, Amicon, Danvers, MA, USA). One
milliliter of a solution of mushroom tyrosinase (1 mg/ml) in 0.1 M sodium phosphate
buffer pH 7 was applied to the filter unit and centrifuged at 5,000 g until maximum
concentration of the sample was achieved. This centrifugation step was repeated three
times with the addition of an additional 2 ml of phosphate buffer prior to each
centrifugation. The final concentrate was reconstituted in phosphate buffer to give a final
concentration of 1 mg/ml of mushroom tyrosinase corresponding to 3870 units of activity
per ml of solution. Aliquots were stored at -20C and defrosted immediately prior to use.
Enzymatic Derivatization
To characterize the enzymatic derivatization procedure, the following incubation
mixture was set up in a microcentrifuge tube: 1 inM LE, 135 units/ml mushroom
tyrosinase and 50 niM ascorbic acid in 0.5 M phosphate buffer pH 7.4. The reaction was
allowed to proceed at room temperature with constant shaking for 60 minutes. Aliquots of
this incubation mixture were then applied to an HPLC system consisting of an
LDC/Milton Roy miniMetric II metering pump (Riviera Beach, FL, USA), a Negretti and
Zamba injector (Southampton, UK) fitted with a 500 pi loop, a Perkin Elmer LC-75
spectrophotometric detector (Norwalk, CT, USA) and a Hewlett Packard HP 3394A


5
are thought to mediate supraspinal analgesia and a low affinity sites (imi2) which are
thought to be responsible for respiratory depression and gastrointestinal effects [Pasternak
1982],
Leucine enkephalin and other derivatives of proenkephalin A interact with the delta
receptor, although not selectively. Opiate alkaloids, on the other hand, have low affinity
for this receptor. Delta receptors are less widespread than mu receptors but are
concentrated in neural areas involved with olfaction and motor integration and have been
implicated in pain pathways [Mansour et al. 1988],
The derivatives of prodynorphin have selectivity for kappa receptors. Although LE
is selective for the delta receptor, as the molecule is lengthened, its preference for the delta
receptor is reduced and its affinity for the kappa receptor increases. Kappa receptors are
found predominantly in brain areas associated with pain perception and the regulation of
water balance and food intake [Mansour et al. 1988],
Physiology and Pharmacology
The binding of an opioid to its receptor triggers a number of complex processes
which occur before leading to the ultimate opioid effect. Opioid effects are believed to
mediated through guanine nucleotide regulatory proteins (G proteins) which are involved
in signal transduction to a variety of effector systems including adenylate cyclase,
phospholipidase C and ion channels. Presently, adenylate cyclase inhibition is the best
characterized opioid effect mediated by G proteins [Simon 1984],


10
which opioid peptides cause an increase in the secretion of prolactin, growth hormone and
thyrotrophin and inhibit the release of luteinizing hormone, follicle stimulating hormone,
adrenocorticotrophic hormone and beta- and gamma-lipotropin [Clement-Jones and
Besser 1983, Grossman and Rees 1983], Further possible functions of opioid peptides can
be found in Table 1.1.
Table 1.1. Possible physiological functions of endogenous opioid peptides [Imura et al.
1985],
1. Defense against noxious stimuli
Activation of the pituitary-adrenocortical axis
Regulation of the sympatho-adrenal system
Inhibition of pain perception
2. Modulation of vegetative nervous system
Cardiovascular and respiratory system
Gastrointestinal tract and pancreas
Genito-urinary tract
3. Modulation of neuroendocrine function
Anterior and posterior pituitary hormones
Gastrointestinal and pancreatic hormones
Catecholamines
4. Behavioral action
Mood and locomotor activity
Food and water intake
Sexual behavior
The design of opioid peptides as therapeutic agents has several advantages. Firstly,
these substances are endogenous so that their metabolites are likely to be non-toxic and
not to cause renal or hepatic damage, depending on the doses administered. Secondly, a
large number of analogs can be synthesized from a few basic amino acid building blocks as
synthesis has been simplified and automated and simple modifications can be used to


49
Figure 2.10. Chromatographs of control injection (A) showing LE-Lys6 eluting after 5.15
minutes and reaction injections showing LE-Lys6-B (no spacer arm) eluting
after 14.9 minutes (B) and LE-Lys6-BX (with spacer arm) eluting after 17.8
minutes (C). Di-biotinylated products are seen eluting at 24.8 minutes (B)
and 27.8 minutes (C).
extent of conversion from the starting materials (LE and LE-Lys6) to the biotinylated
derivatives was estimated from the difference in peak height of LE and LE-Lys6 in the
control injection and in the injection of the reaction mixture. An extent of conversion of
approximately 80% from the starting materials to the biotinylated derivatives was
estimated. The results of the HABA displacement experiments confirmed that the products
collected from the HPLC eluent were indeed biotinylated LE derivatives as they were seen
to displace HABA from avidin (Figure 2.11). The results of this experiment confirm that


119
Table 5.2. Effect of modification of Tyr1 moiety of opioid peptides on opioid binding
affinity.
Modification at Tvr1
Effect
Reference
[des-Tyr1] opioid peptide
loss of opioid binding
affinity
Hansen and Morgan 1984
Shimohigashi 1986
, /0H
HO^ ^R
increased opioid binding
affinity
Hansen et al. 1985
H\ V
reduced opioid binding
affinity
Judd et al. 1985
HO
ho-)^r
reduced opioid binding
affinity
This study
R= opioid peptide analog
of the Tyr1 ring of enkephalin analogs greatly increased opioid binding affinity. By
contrast, Judd et al. [Judd et al. 1985] demonstrated that D-Ala2 methionine enkephalin
amide analogs with a hydroxy group in the meta position rather than the para position of
the Tyr1 ring showed reduced affinity to (.ij, p2> 5 and k opioid receptors. In our studies, a
decrease in opioid binding affinity to both p and 6 receptors was observed when a hydroxy
group was introduced at the meta position of the Tyr1 ring of LE, in addition to the
hydroxy group already present at the para position. We presumed that the first
hydroxylation took place at the meta position as the hydroxy group in the para position


This dissertation was submitted to the Graduate Faculty of the College of
Pharmacy and to the Graduate School and was accepted a partial fulfillment of the
requirements for the degree of Doctor of Philosophy.
August 1994
Dean, Graduate School


125
method (170 fmol/inj in buffer and 360 fmol/inj in CSF compared to 500 fmol/inj in both
buffer and CSF) and yields a cleaner chromatograph with the peak of interest eluting after
a shorter retention time (6 minutes compared to 22.5 minutes). Therefore, of the two
HPLC approaches described in this dissertation, based on sensitivity, convenience and
practicality, the HPLC-ED assay is the method of choice for the analysis of opioid
peptides in CSF. It is proposed that increased sensitivity could be achieved using the
HPLC-FL approach by incorporating microbore HPLC and laser-induced fluorescence
into the procedure.
Hydroxylated derivatives of LE were produced using mushroom tyrosinase in the
HPLC assays for opioid peptides described in Chapters 3 and 4. In Chapter 5 the structure
of the major product of the reaction between LE and mushroom tyrosinase was
determined by electrospray ionization mass spectrometry to be [HO-Tyrl]-LE and the
minor product of the reaction was identified as [(HO)2-Tyr1]-LE. The affinity of [HO-
Tyr']-LE to opioid receptors in rat brain homogenate was determined by radioreceptor
assay. Hydroxylation of LE was found to decrease receptor affinity to both p and 5 opioid
receptor sites by a factor of about 20. Tyrosinase and enkephalin have been found to co
exist in isolated cells and in the spinal and brain regions of some species. Therefore, since
we have demonstrated that the product of the reaction between these two entities shows
decreased affinity to opioid receptors compared to the parent enkephalin, we speculate
that tyrosinase may play a role in the metabolic pathway of LE in vivo.


Results 93
Discussion 100
5 LEUCINE ENKEPHALIN-TYROSINASE REACTION PRODUCTS -
IDENTIFICATION AND BIOLOGICAL ACTIVITY 104
Introduction 104
Materials 105
Methods 105
Results 110
Discussion 118
6 CONCLUSIONS 121
APPENDICES
A DATA FOR HPLC-ED APPROACH 126
B DATA FOR HPLC-FL APPROACH 131
REFERENCES 138
BIOGRAPHICAL SKETCH 145
v


69
The introduction of a 3-hydroxytyrosine group to the LE molecule by enzymatic
derivatization allows the use of a specific clean-up procedure for 3,4-dihydroxyphenyl
compounds using boronate gels and column chromatography. For this assay, the boronate
clean-up method is based on the pH-dependent formation of a complex between
immobilized boronate gel and the hydroxylated LE derivative (Figure 3.3). A complex is
formed at weakly alkaline pH between ionized boronate affixed to a gel matrix and the
hydroxylated LE derivative. Dissociation of the complex occurs at acidic pH.
Therefore, in the assay described here, LE in the sample is first derivatized
enzymatically by mushroom tyrosinase and subsequently, the sample is subjected to clean
up through the use of a boronate gel column. Finally, the sample is quantified by high
performance liquid chromatography with electrochemical detection (HPLC-ED).
Materials
Leucine enkephalin and mushroom tyrosinase were obtained from Sigma Chemical
Company, St. Louis, MO, USA. Acetonitrile, methanol and trifluoroacetic acid were of
HPLC grade and disodium hydrogen phosphate and citric acid were of reagent grade.
These chemicals were procured from Fisher Scientific, Pittsburgh, PA, USA. Sodium
dihydrogen phosphate was of molecular biology grade and was purchased from Fluka
Chemie, Buchs, Switzerland. All other chemicals were of reagent grade. Double distilled
water was used throughout.


84
Table 3.3. Comparison table of current analytical methods for opioid peptides.
Reference
Method
Analyte
Matrix
Sensitivity
This study
HPLC-ED
LE
Buffer
170 fmol/inj
This study
HPLC-ED
LE
CSF
360 fmol/inj
Fleming and
Reynolds 1988
HPLC-ED
enkephalin
rat brain
1 pmol/inj
Shibanoki et al.
1990
HPLC-ED
enkephalin
plasma
550 fmol/inj
Kimetal. 1989
HPLC-ED
enkephalin
rat brain
1 pmol/inj
Muck and
Henion 1989
HPLC-MS
dynorphin
CSF
100 fmol/inj
Mifune et al.
1989
HPLC-FL
enkephalin
rat brain
100 fmol/inj
de Montigny et
al. 1990
HPLC-FL
LE
plasma
7.7 pmol/inj
HPLC-MS = high performance liquid chromatography with mass spectrometry, HPLC-FL
= high performance liquid chromatography with fluorescence detection.
The boronate clean-up procedure used in the HPLC-ED assay described here
proved to be an effective and efficient clean-up method for LE in CSF resulting in a clean
chromatograph for the analyte (Figure 3.11). To avoid interfering peaks, existing HPLC-
ED methods for enkephalins involve complex and relatively non-selective sample clean up
procedures (e.g./ multiple precipitation, centrifugation and adsorption steps) prior to
application of the sample to the HPLC system. By contrast, here the use of a boronate
clean-up procedure, which had been previously established for use with catecholamines
[Higa et al. 1977, Koike et al. 1982], increased the selectivity of this assay as only


74
Time Course of Enzymatic Derivatization
To determine the time course of the enzymatic derivatization at the analytical
concentrations, the following incubation mixture was set up in microcentriiuge tubes:
8*10'8 M LE, 50 mM ascorbic acid and 135 units/ml mushroom tyrosinase in 300 pi of
0.5 M phosphate buffer pH 7.4. The reaction was stopped at various time points by adding
20 pi of 1 N HC1. Two hundred an fifty micrometers of this incubation mixture was then
subjected to the boronate clean-up procedure, the resulting sample was injected into the
HPLC-ED system and the peak areas of the peak corresponding to the enzymatic
derivative [HO-Tyr'j-LE were recorded.
Extraction of Leucine Enkephalin from Cerebrospinal Fluid
Leucine enkephalin was extracted from human cerebrospinal fluid (CSF) through
the use of Supelclean LC 18 solid phase extraction columns (Supelco Inc., Bellefonte, PA,
USA). The columns were activated with 3 ml each of water and methanol and loaded with
100 pi of spiked CSF. Subsequently, the columns were washed with 1 ml of water, 3 ml of
0.1 N HC1, 1 ml of water, 3 ml of 0.1 M borate buffer (pH 8.5) and 1 ml of water. The LE
rich fraction was then eluted in 2 ml of methanol and evaporated to dryness under a stream
of nitrogen.
Calibration Curves
Using the complete HPLC-ED method, including enzymatic derivatization and
boronate clean-up, calibration curves were constructed for LE in both buffer and CSF. For


136
Curve 11
Concentrated reagents, 5 minute tyrosinase reaction
Matrix: Buffer
Equation: y = 7.67* 1012x + 4.9244, r2 = 0.9922
LE mol/inj
Peak height (mm)
0
0
2.77* 10'13
10.5
5.54* 10"13
11.5
1.38*10"12
18
2.77*10'12
24
5.54* 10"12
44.2
8.31*10"12
64.5
1.38*10'"
115
Curve 12
Concentrated reagents, 5 minute tyrosinase reaction
Matrix: Buffer
Equation: y = 2.50* 1012x + 4.449, r2 = 0.9779
LE mol/inj
Peak height (mm)
0
3.5
2.77*1013
4.5
5.54* 10'13
7
1.38* 10'12
8.5
2.77*1012
11.5
5.54* 1012
18


17
and baseline noise are increased significantly. Selectivity is also compromised as more
compounds are oxidized at these high potentials, thus necessitating extensive sample clean
up.
In the HPLC-ED assay for LE described in this dissertation, the enzymatic
derivatization used increased HPLC-ED selectivity and sensitivity by pre-column o-
hydroxylation of the highly conserved N-terminal tyrosine groups of the peptide resulting
in easily oxidizable derivatives.
Fluorescence detection
Pre-column fluorescence derivatization of analytes can be achieved by fluorophoric
labeling using a fluorescent precursor, or by fluorogenic derivatization using a non-
fluorescent precursor. Fluorogenic derivatization is usually preferred as fluorophoric
labeling often requires excess fluorescent reagent and subsequent extensive clean up to
minimize background interference.
Fluorogenic derivatization reactions have been carried out by using fluorogenic
reagents such as o-phthalaldehyde [Roth 1971], fluorescamine [Udenfriend et al. 1972]
and naphthalene-2,3-dialdehyde in the presence of cyanide [Lunte and Wong 1989, Mifune
et al. 1989], However, these procedures derivatize N-terminal amino groups and
consequently are not specific for any particular peptide so that subsequent
chromatographic procedures are often extremely complex (e.g./ multidimensional HPLC
systems using column switching) to allow for the selective determination of opioid
peptides as the derivatives of other peptides may interfere with the signal.




112
Mass spectrometry
ESI is a soft ionization technique that provides intact molecular ions from peptide
solutions [Whitehouse et al. 1985], However, selective fragmentation can also be effected
by collision-induced dissociation (CID) in certain regions of the ion source [Allen and
Vestal 1992, Katta et al. 1991, Smith et al. 1990], The nomenclature scheme, as proposed
by Roepstoff and Fohlman [Roepstorff and Fohlman 1984], used to label the sequence
ions in the mass spectra of pentapeptides is shown in Figure 5.3.
Figure 5.3. Nomenclature scheme for the labeling of sequence ions in the mass spectra of
pentapeptides.
The ESI mass spectra obtained from LE are shown in Figure 5.4. Using soft
ionization (repeller at 18 V, Figure 5.4A), only molecular ions ([M+H]+, m/z 556, and
[M+Na]+, m/z 578) were observed. Although the formation of protonated molecules
through acid-base equilibria in solution is preferred, the generation of sodiated species was
unavoidable upon use of commercially available HPLC-grade solvents [Lehman 1982],
From the spectrum obtained under CID conditions (repeller at 50 V, Figure 5.4B), several
important sequence ions (m/z 136, 221, 278, 397, and 425 for ai, b2, b3, a,}, and b4,


101
improvement in limit of detection seen with the approach described here may be attributed
in part to the fact that in some of these existing tyrosine-specific HPLC-FL methods
[Ishida et al. 1986, Kai et al. 1988], fluorogenic derivatization necessitates heating of the
reaction mixture to 100C for 3 minutes which may result in peptide instability and
consequent reduced recovery of intact fluorescence-labeled analyte. Although more
selective, the HPLC-FL approach described here does not achieve limits of detection for
LE as low as those obtained by investigators using methods involving post-column
fluorescence derivatization of the N-terminal primary amino group of opioid peptides (36-
100 fmol/injection) [Dave et al. 1992, Mifune et al. 1989, van den Beld et al. 1990],
However, due to lack of selectivity, these methods often involve the use of extensive
sample clean-up procedures and complex multi-dimensional chromatographic systems
(column switching) to minimize the occurrence of interfering peaks [Mifune et al. 1989],
The higher sensitivity obtained using these methods can be attributed to the use of laser-
induced fluorescence detection [Dave et al. 1992, van den Beld et al. 1990] and a
microbore chromatographic system [Dave et al. 1992], Incorporating a microbore
chromatographic system and laser-induced fluorescence into the analytical approach
described here could be expected to reduce considerably the limits of detection obtainable
by this approach.
An attempt to improve the approached described in this study by concentrating the
reagents in the fluorogenic reaction and thereby injecting a larger portion of the total
reaction mixture did not result in an analytical advantage. This may be due to a decrease in
the efficiency of the fluorogenic reaction in the presence of higher concentrations of


135
Curve 9
Concentrated reagents
Matrix: CSF
Equation: y = 1.76*1012x + 3.0077, r2 = 0.9853
LE mol/inj
Peak area
0
2.2941
4.69*10'13
3.5183
8.85* 10'13
2.4436|
1.95* 1012
7.0437
2.88* 1012
3.7482f
4.41*1 O12
10.505
toutlier
Curve 10
Concentrated reagents
Matrix: CSF
Equation: y = 3.45*1012x + 1.0393, r2 = 0.9975
LE mol/inj
Peak area
0
1.4506
4.69* 10'13
2.6145
8.85* 1013
4.3134
1.95*1 O'12
4.0743f
2.88* 10'12
9.3891
4.41 10"12
17.243
foutlier


isothiocyanate avidin, an homogeneous fluorescence immunoassay for LE was developed
which was operational in a narrow concentration range (1*1 O'9 to 1*10"8 moles LE/ml).
Two HPLC assays for opioid peptides were evaluated. One was based on tyrosine-
specific pre-column hydroxylation using tyrosinase, specific sample clean-up using a
boronate gel and HPLC with electrochemical detection. The other involved tyrosine-
specific pre-column hydroxylation using tyrosinase followed by fluorogenic derivatization
using 1,2-diamino-1,2-diphenylethane and HPLC with fluorescence detection. These
assays yielded limits of detection for LE of 170 fmol/inj and 500 fmol/inj respectively in
buffer samples and 360 fmol/inj and 500 fmol/inj respectively in spiked cerebrospinal fluid
samples.
Using electrospray ionization mass spectrometry, the structure of the products of
the reaction between LE and tyrosinase were found to be monohydroxylated LE ([HO-
Tyrj-LE) and dihydroxylated LE ([(HO)2-Tyr']-LE). Compared to LE, the affinity of
[HO-Tyr*]-LE to both p and 5 opioid receptor sites in rat brain homogenate was found to
be lower by a factor of about 20. Since enkephalins and tyrosinase have been found to co
exist in vivo, we speculate that tyrosinase may play role in the metabolic pathway of these
compounds.


11
develop different analogs with desirable biological activities. Opioid peptides are unable to
cross the placental barrier as they are subject to placental enzymatic deactivation and
would therefore be ideal for obstetric use [Rapaka 1986, Rapaka and Porreca 1991],
At present, the therapeutic development of opioid peptides is focused on their
potential as analgesic agents and in the treatment of opiate addiction. Research efforts are
directed towards the design of analgesic peptides which can be administered orally, have a
long duration of action and reduced potential for dependence and abuse. Peptides of
interest include enkephalins, endorphins and related opioid peptides.
Opioid peptides are easily degraded by aminopeptidases which hydrolyze the Tyr1-
Gly2 bond, carboxypeptidases which cause cleavage at the C-terminal end of the molecule,
relatively non-specific enzymes such as trypsin and angiotensin converting enzyme (ACE)
and more specific enzymes such as enkephalinases which hydrolyze the Gly3-Phe4 in
enkephalins. This has led to the suggestion that enkephalin degrading enzymes be used as
an alternative therapeutic approach [Rapaka 1986, Rapaka and Porreca 1991], Examples
of these enzyme inhibitors include bestatin, thiorphan and captopril which inhibit
aminopeptidase, enkephalinase and ACE, respectively. These inhibitors prolong the
duration of action of endogenously released enkephalins and it is therefore hoped that they
are free of the side effects produced by narcotic analgesics. The clinical use of these
substances may however be limited due to their limited bioavailability.
Another approach which has been used to increase both the stability and the
selectivity of opioid peptides is the introduction of synthetic modifications to the molecule
[Shimohigashi 1986], For example, stability can be increased by substituting the



PAGE 1

129(/ $1$/<7,&$/ $3352$&+(6 )25 7+( '(7(50,1$7,21 2) /(8&,1( (1.(3+$/,1 $6 $ 02'(/ )25 23,2,' 3(37,'(6 %\ 9(521,48( /$56,0217 $ ',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

7R P\ SDUHQWV &KDUOHV /DUVLPRQW DQG .UXDZDQ .DQMDQDVXZDQ/DUVLPRQW IRU WKHLU XQIDLOLQJ VXSSRUW DQG HQFRXUDJHPHQW RI DOO P\ HQGHDYRUV

PAGE 3

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

PAGE 4

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

PAGE 5

5HVXOWV 'LVFXVVLRQ /(8&,1( (1.(3+$/,17<526,1$6( 5($&7,21 352'8&76 ,'(17,),&$7,21 $1' %,2/2*,&$/ $&7,9,7< ,QWURGXFWLRQ 0DWHULDOV 0HWKRGV 5HVXOWV 'LVFXVVLRQ &21&/86,216 $33(1',&(6 $ '$7$ )25 +3/&(' $3352$&+ % '$7$ )25 +3/&)/ $3352$&+ 5()(5(1&(6 %,2*5$3+,&$/ 6.(7&+ Y

PAGE 6

$EVWUDFW RI 'LVVHUWDWLRQ 3UHVHQWHG WR WKH *UDGXDWH 6FKRRO RI WKH 8QLYHUVLW\ RI )ORULGD LQ 3DUWLDO )XOILOOPHQW RI WKH 5HTXLUHPHQWV IRU WKH 'HJUHH RI 'RFWRU RI 3KLORVRSK\ 129(/ $3352$&)(6 )25 7+( '(7(50,1$7,21 2) /(8&,1( (1.(3+$/,1 $6 $ 02'(/ )25 23,2,' 3(37,'(6 %\ 9URQLTXH /DUVLPRQW $XJXVW &KDLUPDQ *LLQWKHU +RFKKDXV 0DMRU 'HSDUWPHQW 3KDUPDFHXWLFV 7KH IRFXV RI WKLV GLVVHUWDWLRQ ZDV WKH HYDOXDWLRQ RI QRYHO DQDO\WLFDO DSSURDFKHV IRU RSLRLG SHSWLGHV E\ LPPXQRDVVD\ RU KLJK SHUIRUPDQFH OLTXLG FKURPDWRJUDSK\ +3/&f XVLQJ OHXFLQH HQNHSKDOLQ /(f DV D PRGHO SHSWLGH 7KH SURSRVHG LPPXQRDVVD\V DUH EDVHG RQ WKH KLJK DIILQLW\ H[KLELWHG E\ DYLGLQ IRU ELRWLQ .G f PROf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

PAGE 7

LVRWKLRF\DQDWH DYLGLQ DQ KRPRJHQHRXV IOXRUHVFHQFH LPPXQRDVVD\ IRU /( ZDV GHYHORSHG ZKLFK ZDV RSHUDWLRQDO LQ D QDUURZ FRQFHQWUDWLRQ UDQJH r 2n WR r PROHV /(POf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fM/(f DQG GLK\GUR[\ODWHG /( >+2f7\Un@/(f &RPSDUHG WR /( WKH DIILQLW\ RI >+27\Ur@/( WR ERWK S DQG RSLRLG UHFHSWRU VLWHV LQ UDW EUDLQ KRPRJHQDWH ZDV IRXQG WR EH ORZHU E\ D IDFWRU RI DERXW 6LQFH HQNHSKDOLQV DQG W\URVLQDVH KDYH EHHQ IRXQG WR FRn H[LVW LQ YLYR ZH VSHFXODWH WKDW W\URVLQDVH PD\ SOD\ UROH LQ WKH PHWDEROLF SDWKZD\ RI WKHVH FRPSRXQGV 9OO

PAGE 8

&+$37(5 ,1752'8&7,21 (QGRJHQRXV 2SLRLG 3HSWLGHV 7KHUH H[LVWV WKUHH VHSDUDWH IDPLOLHV RI QDWXUDO RU HQGRJHQRXV RSLRLG SHSWLGHV HQNHSKDOLQV HQGRUSKLQV DQG G\QRUSKLQV 7KHVH WKUHH IDPLOLHV DUH GHULYHG IURP WKUHH GLIIHUHQW SURKRUPRQHV SURHQNHSKDOLQ SURRSLRPHODQRFRUWLQ DQG SURG\QRUSKLQ UHVSHFWLYHO\ DQG DUH FRGHG E\ PHVVHQJHU 51$V IURP WKUHH VHSDUDWH JHQHV )LJXUH f >3OHXYU\ @ $OSKDPHODQRF\WH VWLPXODWLQJ KRUPRQHr &RUWLFRWURSKOQOLNH LQWHUPHGLDWH OREH SHSWLGH %HWDHQGRUSKLQ $GUHQRFRUWLFRWURSKLQr %HWDOLSRWURSKLQr A ,35223,20(/$12&25I“O P51$ $ *(1(6 %HWDPHODQRF\WH VWLPXODWLQJ KRUPRQHr /HXPRUSKLQ %HWDQHRHQGRUSKLQ P51$ r ,352(1.(3+$/,1, 3HSWLGH ( W ,0HWOHQNHSKDOLQ >/HXOHQNHSKDOLQ >0HW@HQNHSKDOcQ$UJ3KH '\QRUSKLQ $ f P51$A>S'<1453+“f ; '\QRUSKLQ % f '\QRUSKLQ f 3HSWLGH ) _0HW@HQNHSKDOLQ$UJ*O\/HX )LJXUH 2SLRLG SHSWLGHV DQG WKHLU SUHFXUVRUV r 'HQRWHV QR RSLRLG UHFHSWRU DFWLYLW\f >3OHXYU\ @

PAGE 9

,Q WKH SURHQNHSKDOLQ IDPLO\ SURHQNHSKDOLQ $ LV WKH SUHFXUVRU RI WKH SHQWDSHSWLGHV PHWKLRQLQH HQNHSKDOLQ 0( 7\U*O\*O\3KH0HWf DQG OHXFLQH HQNHSKDOLQ /( 7\U*O\ *O\3KH/HXf ZKLFK ZHUH WKH ILUVW RSLRLG SHSWLGHV WR EH FKDUDFWHUL]HG E\ +XJKHV DQG FRZRUNHUV LQ )LJXUH f >+XJKHV HW DO @ 3URHQNHSKDOLQ $ KDV EHHQ VKRZQ WR FRQWDLQ 0( DQG /( LQ D IL[HG UDWLR RI VL[ 0( VHTXHQFHV WR RQH /( VHTXHQFH 3URHQNHSKDOLQH[SUHVVLQJ FHOOV DUH ZLGHVSUHDG WKURXJKRXW WKH EUDLQ DQG VSLQDO FRUG DV ZHOO DV LQ PRUH SHULSKHUDO VLWHV VXFK DV WKH DGUHQDO PHGXOOD DQG WKH JDVWURLQWHVWLQDO WUDFW 35223,20(/$12&257,1 U06+M Uf§ + LBU m/3+ r r(1'L nOn ,I LQ D06+ 06+ 0(7(1. 352(1.(3+$/,1 0(7(1. /(8(1. JO J : +, ,, KUO KUO 0(7(1. 0(7(1. $5r/-DIOIH DQG 0DUWLQ @ 3URRSLRPHODQRFRUWLQ 320&f LV WKH SUHFXUVRU RI WKH RSLRLG SHSWLGH 3HQGRUSKLQ DV ZHOO DV WKH QRQRSLRLG KRUPRQHV DGUHQRFRUWLFRWURSKLF KRUPRQH $&7+f DQG D DQG 3

PAGE 10

PHODQRF\WH VLPXODWLQJ KRUPRQH 06+ )LJXUH f 7KH WHUP HQGRUSKLQ DSSOLHV WR RSLRLG SHSWLGHV GHULYHG IURP 320& 320& LV V\QWKHVL]HG LQ WKH SLWXLWDU\ EXW LV DOVR SUHVHQW LQ WKH K\SRWKDODPXV DQG LQ WKH SHULSKHU\ 7KH 0( DPLQR DFLG VHTXHQFH LV SUHVHQW DW WKH 1 WHUPLQDO HQG RI HQGRUSKLQ 7KH G\QRUSKLQV DV ZHOO DV 3QHRHQGRUSKLQ DQG OHXPRUSKLQ DUH GHULYHG IURP SURG\QRUSKLQ ZKLFK LV DOVR NQRZQ DV SURHQNHSKDOLQ % 3URG\QRUSKLQ FRQWDLQV /( VHTXHQFHV EXW QR 0( VHTXHQFHV DQG LV V\QWKHVL]HG WKURXJKRXW WKH FHQWUDO QHUYRXV V\VWHP )LJXUH f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f LQ YLWUR ELRDVVD\V DQG ELQGLQJ H[SHULPHQWV ZLWK VHOHFWLYH OLJDQGV >6LPRQGV @ 7KHVH WKUHH

PAGE 11

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a VHSWDO UHJLRQ JOREXV SDOOLGXV aa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f ZKLFK RFFLSLWDO OREH FRUSRUD TXDGUOJHPL SHQR4XDGXFWDO JUD QP O JUR\ VXEVWDQWLD QLJUD f§ LQWHUSHGXQFXODU Q ORFXV FRHUXOHXV PLGOLQH UHLWLF IRUP UHWLFXODU IRUPDWLRQ RUHD SRVWUHPD FHUHEHOOXP VXEVWDQWLD JHODWLQRVD

PAGE 12

DUH WKRXJKW WR PHGLDWH VXSUDVSLQDO DQDOJHVLD DQG D ORZ DIILQLW\ VLWHV PXf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f ZKLFK DUH LQYROYHG LQ VLJQDO WUDQVGXFWLRQ WR D YDULHW\ RI HIIHFWRU V\VWHPV LQFOXGLQJ DGHQ\ODWH F\FODVH SKRVSKROLSLGDVH & DQG LRQ FKDQQHOV 3UHVHQWO\ DGHQ\ODWH F\FODVH LQKLELWLRQ LV WKH EHVW FKDUDFWHUL]HG RSLRLG HIIHFW PHGLDWHG E\ SURWHLQV >6LPRQ @

PAGE 13

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f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

PAGE 14

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n DQDOJHVLF GRVHV KDYH EHHQ IRXQG WR HLWKHU SRWHQWLDWH RU LQKLELW WKH DQDOJHVLF HIIHFWV RI PRUSKLQH D PX DJRQLVWf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

PAGE 15

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

PAGE 16

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

PAGE 17

ZKLFK RSLRLG SHSWLGHV FDXVH DQ LQFUHDVH LQ WKH VHFUHWLRQ RI SURODFWLQ JURZWK KRUPRQH DQG WK\URWURSKLQ DQG LQKLELW WKH UHOHDVH RI OXWHLQL]LQJ KRUPRQH IROOLFOH VWLPXODWLQJ KRUPRQH DGUHQRFRUWLFRWURSKLF KRUPRQH DQG EHWD DQG JDPPDOLSRWURSLQ >&OHPHQW-RQHV DQG %HVVHU *URVVPDQ DQG 5HHV @ )XUWKHU SRVVLEOH IXQFWLRQV RI RSLRLG SHSWLGHV FDQ EH IRXQG LQ 7DEOH 7DEOH 3RVVLEOH SK\VLRORJLFDO IXQFWLRQV RI HQGRJHQRXV RSLRLG SHSWLGHV >,PXUD HW DO @ 'HIHQVH DJDLQVW QR[LRXV VWLPXOL $FWLYDWLRQ RI WKH SLWXLWDU\DGUHQRFRUWLFDO D[LV 5HJXODWLRQ RI WKH V\PSDWKRDGUHQDO V\VWHP ,QKLELWLRQ RI SDLQ SHUFHSWLRQ 0RGXODWLRQ RI YHJHWDWLYH QHUYRXV V\VWHP &DUGLRYDVFXODU DQG UHVSLUDWRU\ V\VWHP *DVWURLQWHVWLQDO WUDFW DQG SDQFUHDV *HQLWRXULQDU\ WUDFW 0RGXODWLRQ RI QHXURHQGRFULQH IXQFWLRQ $QWHULRU DQG SRVWHULRU SLWXLWDU\ KRUPRQHV *DVWURLQWHVWLQDO DQG SDQFUHDWLF KRUPRQHV &DWHFKRODPLQHV %HKDYLRUDO DFWLRQ 0RRG DQG ORFRPRWRU DFWLYLW\ )RRG DQG ZDWHU LQWDNH 6H[XDO EHKDYLRU 7KH GHVLJQ RI RSLRLG SHSWLGHV DV WKHUDSHXWLF DJHQWV KDV VHYHUDO DGYDQWDJHV )LUVWO\ WKHVH VXEVWDQFHV DUH HQGRJHQRXV VR WKDW WKHLU PHWDEROLWHV DUH OLNHO\ WR EH QRQWR[LF DQG QRW WR FDXVH UHQDO RU KHSDWLF GDPDJH GHSHQGLQJ RQ WKH GRVHV DGPLQLVWHUHG 6HFRQGO\ D ODUJH QXPEHU RI DQDORJV FDQ EH V\QWKHVL]HG IURP D IHZ EDVLF DPLQR DFLG EXLOGLQJ EORFNV DV V\QWKHVLV KDV EHHQ VLPSOLILHG DQG DXWRPDWHG DQG VLPSOH PRGLILFDWLRQV FDQ EH XVHG WR

PAGE 18

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f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

PAGE 19

FRUUHVSRQGLQJ 'DPLQR DFLG IRU WKH QDWXUDOO\ RFFXUULQJ /DQLLQR DFLG LQ WKH SHSWLGH PROHFXOH $V SHSWLGHV DUH FRQIRUPDWLRQDOO\ ODELOH WKH VHOHFWLYLW\ RI RSLRLG SHSWLGHV KDV EHHQ LQFUHDVHG E\ VWDELOL]LQJ WKH SHSWLGH PROHFXOH LQ D FRQIRUPDWLRQ ZKLFK SUHIHUV WKH GHVLUHG UHFHSWRU 7KLV FDQ EH DFKLHYHG WKURXJK WKH LQFRUSRUDWLRQ RI FRQIRUPDWLRQDO UHVWULFWLRQV VXFK DV WKH LQWURGXFWLRQ RI XQQDWXUDO EXON\ V\QWKHWLF DPLQR DFLGV HJ SHQLFLOODPLQH UHVLGXHVf RU WKH F\FOL]DWLRQ RI WKH SHSWLGH FKDLQ +LJKO\ VHOHFWLYH RSLRLG SHSWLGH DQDORJV VXFK DV >'3HQ '3HQ@HQNHSKDOLQ ZKLFK LV VHOHFWLYH IRU WKH GHOWD UHFHSWRU >0RVEHUJ HW DO @ DQG >7\U'$OD*O\0H3KH1+&+f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

PAGE 20

&XUUHQW DQDO\WLFDO PHWKRGV IRU RSLRLG SHSWLGHV DUH VXPPDUL]HG LQ 7DEOH DQG DUH UHYLHZHG LQ WKH IROORZLQJ SDJHV ,PPXQRORJLFDO 0HWKRGV $W SUHVHQW IRU WKH PRVW SDUW RSLRLG SHSWLGHV DUH DQDO\]HG E\ UDGLRLPPXQRDVVD\ 5,$f >6DWR 9HQQ @ $OWKRXJK LPPXQRDVVD\V DUH KLJKO\ VHQVLWLYH DQG UHSURGXFLEOH WKH\ DUH KDPSHUHG E\ VHYHUDO GLVDGYDQWDJHV )LUVWO\ DQ DQWLVHUXP WR WKH DQDO\WH RI LQWHUHVW KDV WR EH UDLVHG ZKLFK LQYROYHV OHQJWK\ LQFXEDWLRQ WLPHV 6HFRQGO\ LPPXQRDVVD\V VXIIHU IURP ORZ VHOHFWLYLW\ GXH WR WKH FURVVUHDFWLYLW\ RI WKH DQWLVHUXP WR VWUXFWXUDOO\ VLPLODU FRPSRXQGV 7R RYHUFRPH WKH SUREOHP RI FURVVUHDFWLYLW\ KLJK SHUIRUPDQFH OLTXLG FKURPDWRJUDSK\ +3/&f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f KDV EHHQ GHVFULEHG IRU /( DQG 0( >=DPERQL HW DO @ EXW LWV OLPLWV RI GHWHFWLRQ DUH RQO\ LQ WKH SPRO SHU DVVD\ UDQJH 2QO\ IRXU HQ]\PH LPPXQRDVVD\V ZLWK KLJK VHQVLWLYLW\ KDYH EHHQ GHYHORSHG IRU >HQGRUSKLQ DQG G\QRUSKLQ >+RFKKDXV DQG +X +RFKKDXV DQG 6DGHH

PAGE 21

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f LV QRW QHFHVVDU\ SULRU WR TXDQWLWDWLRQ DV WKH SURSHUW\ EHLQJ PHDVXUHG LV FKDUDFWHULVWLF RI HLWKHU WKH ERXQG RU WKH IUHH DQDO\WH RU ODEHO 7KH LPPXQRORJLFDO WHFKQLTXHV SURSRVHG LQ WKLV GLVVHUWDWLRQ KDYH WKH GLVDGYDQWDJH RI ORZ VSHFLILFLW\ EXW WKH\ ZHUH LQWHQGHG IRU XVH DV LPPXQRORJLFDO +3/& GHWHFWRUV LQ WKH KRSH WKDW WKH\ ZRXOG SURYLGH IDVW XOWUDVHQVLWLYH DVVD\V ZKLFK ZHUH KLJKO\ VXLWDEOH IRU SURFHVVLQJ ODUJH QXPEHUV RI VDPSOHV VXFK DV +3/& IUDFWLRQV

PAGE 22

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

PAGE 23

,QVWUXPHQWDO $SSURDFKHV 2WKHU PHWKRGV ZKLFK KDYH EHHQ XVHG LQ WKH DQDO\VLV RI RSLRLG SHSWLGHV LQFOXGH +3/& FRPELQHG ZLWK HOHFWURFKHPLFDO GHWHFWLRQ +3/&('f IOXRUHVFHQFH GHWHFWLRQ +3/&)/f RU PDVV VSHFWURPHWU\ +3/&06f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f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f FRPSURPLVH VHQVLWLYLW\ DV EDFNJURXQG FXUUHQW

PAGE 24

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f WR DOORZ IRU WKH VHOHFWLYH GHWHUPLQDWLRQ RI RSLRLG SHSWLGHV DV WKH GHULYDWLYHV RI RWKHU SHSWLGHV PD\ LQWHUIHUH ZLWK WKH VLJQDO

PAGE 25

$ GHULYDWL]DWLRQ UHDFWLRQ ZKLFK LV VSHFLILF IRU W\URVLQH JURXSV VXFK DV WKH RQH GHVFULEHG LQ WKLV GLVVHUWDWLRQ FRXOG EH H[SHFWHG WR LQFUHDVH VHOHFWLYLW\ LQ WKH GHWHUPLQDWLRQ RI RSLRLG SHSWLGHV DV W\URVLQH LV KLJKO\ FRQVHUYHG LQ WKH SRVLWLRQ RI WKHVH FRPSRXQGV EXW LV RIWHQ PLVVLQJ LQ XQUHODWHG SHSWLGHV 7\URVLQH VSHFLILF +3/& PHWKRGV ZLWK IOXRUHVFHQFH GHWHFWLRQ VXFK DV WKRVH XVLQJ OGLDPLQRGLPHWKR[\EHQ]HQH >,VKLGD HW DO .DL HW DO @ DQG K\GUR[\ODPLQH FREDOW ,,f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fVRIWf LRQL]DWLRQ DQG VDPSOH LQWURGXFWLRQ WHFKQLTXHV VXFK DV IDVW DWRP ERPEDUGPHQW PDWUL[DVVLVWHG ODVHU GHVRUSWLRQ HOHFWURVSUD\ DQG LRQVSUD\ PDVV VSHFWURPHWU\ VHH $UQRWW HW DO %LHPDQQ DQG &DUU IRU UHYLHZVf ZKLFK DOORZ WKH SURGXFWLRQ RI LQWDFW PROHFXODU LRQV IURP WKHVH IUDJLOH VSHFLHV $QRWKHU DGYDQWDJH RI WKHVH WHFKQLTXHV LV WKDW WKH\ DUH RIWHQ FRPSDWLEOH ZLWK RQOLQH PLFURERUH +3/& $ PLFURERUH +3/&LRQVSUD\ 06 PHWKRG KDV DOORZHG WKH

PAGE 26

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f DV D PRGHO IRU RSLRLG SHSWLGHV /HXFLQH HQNHSKDOLQ ZDV FKRVHQ DV D PRGHO SHSWLGH DV LW FRQWDLQV WKH VDPH LQLWLDO VHTXHQFH FRPPRQ WR DOO RSLRLG SHSWLGHV 7KH ILUVW DSSURDFK HYDOXDWHG ZDV D QRQKRPRJHQHRXV HQ]\PHOLQNHG LPPXQRVRUEHQW DVVD\ (/,6$f ZKLFK HPSOR\HG WKH VDPH VWUDWHJ\ DV (/,6$V ZKLFK KDYH

PAGE 27

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fVDQGZLFKf EHWZHHQ WKH DQWLERG\ WKH ELRWLQ\ODWHG /( GHULYDWLYH DQG HQ]\PHODEHOHG DYLGLQ 2+ &+ Y &+ ? R &+ 2 &+ ?????? ? Z U U+ 1+ & &+ 1+ & 2 2 &+ 2 4 7\U *O\ )LJXUH /HXFLQH HQNHSKDOLQ *O\ 3KH /HX 7KH VHFRQG DSSURDFK ZDV DQ KRPRJHQHRXV RU VHSDUDWLRQIUHH IOXRUHVFHQFH LPPXQRDVVD\ ZKLFK DOVR PDGH XVH RI WKH DYLGLQELRWLQ V\VWHP DQG ZDV EDVHG RQ DQ REVHUYDWLRQ E\ $O+DNLHP DQG FRZRUNHUV >$ +DNLHP HW DO @ WKDW WKH ELQGLQJ RI

PAGE 28

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f DVVD\V ZHUH DOVR HYDOXDWHG ERWK PDNLQJ XVH RI W\URVLQHVSHFLILF SUHFROXPQ GHULYDWL]DWLRQ XVLQJ WKH HQ]\PH W\URVLQDVH ,Q WKH KLJK SHUIRUPDQFH OLTXLG FKURPDWRJUDSK\ DVVD\ ZLWK HOHFWURFKHPLFDO GHWHFWLRQ +3/&('f VSHFLILF K\GUR[\ODWLRQ RI WKH W\URVLQH JURXS LQ WKH SRVLWLRQ RI /( ZKLFK LV KLJKO\ FRQVHUYHG LQ DOO RSLRLG SHSWLGHV SUHVHQWV WZR DQDO\WLFDO DGYDQWDJHV )LUVWO\ WKH GHULYDWL]DWLRQ UHVXOWV LQ WKH IRUPDWLRQ RI D FDWHFKRO ZKLFK LV DPHQDEOH WR VSHFLILF FOHDQXS XVLQJ ERURQDWH JHOV DQG VHFRQGO\ WKLV FDWHFKRO LV PRUH HDVLO\ R[LGL]DEOH WKDQ WKH SDUHQW SHSWLGH WKXV IDFLOLWDWLQJ HOHFWURFKHPLFDO GHWHFWLRQ ,Q WKH KLJK SHUIRUPDQFH OLTXLG FKURPDWRJUDSK\ DVVD\ ZLWK IOXRUHVFHQFH GHWHFWLRQ +3/&)/f HQ]\PDWLF GHULYDWL]DWLRQ E\ W\URVLQDVH UHQGHUV RXU SHSWLGH DPHQDEOH WR IOXRURJHQLF GHULYDWL]DWLRQ XVLQJ GLDPLQR GLSKHQ\OHWKDQH ,Q DGGLWLRQ KDYLQJ HVWDEOLVKHG LQ WKH HQ]\PDWLF GHULYDWL]DWLRQ IRU WKH +3/& DVVD\V WKDW SURGXFWV DUH IRUPHG IURP WKH UHDFWLRQ EHWZHHQ /( DQG W\URVLQDVH WKH LGHQWLW\ RI WKHVH SURGXFWV ZDV GHWHUPLQHG E\ PDVV VSHFWURPHWU\ )XUWKHUPRUH VLQFH W\URVLQDVH DQG

PAGE 29

HQNHSKDOLQV KDYH EHHQ IRXQG WR FRH[LVW LQ YLYR >0HUFKHQWKDOHU 6HVDFN DQG 3LFNHO =KXR HW DO @ WKH ELRORJLFDO DFWLYLW\ LQ UDW EUDLQ KRPRJHQDWH RI WKH PDMRU SURGXFW RI WKH UHDFWLRQ EHWZHHQ WKHVH WZR HQWLWLHV ZDV LQYHVWLJDWHG

PAGE 30

&+$37(5 $3352$&+(6 72 7+( '(9(/230(17 2) $1 ,00812$66$< )25 /(8&,1( (1.(3+$/,1 ,QWURGXFWLRQ ,PPXQRDVVD\ LV DQ DQDO\WLFDO PHWKRG ZKLFK H[SORLWV WKH ELQGLQJ RI D OLJDQG DQWLJHQf WR VSHFLILF VLWHV RQ DQ DQWLERG\ ,Q PRVW FDVHV H J IRU UDGLRLPPXQRDVVD\Vf ODEHOHG OLJDQG FRPSHWHV ZLWK XQODEHOHG OLJDQG DQDO\WHf IRU DQWLERG\ VLWHV DQG WKH H[WHQW RI ELQGLQJ RI WKH ODEHOHG OLJDQG LV GHWHUPLQHG WKURXJK WKH PHDVXUHPHQW RI VRPH SK\VLFDO RU FKHPLFDO SURSHUW\ DVVRFLDWHG ZLWK WKH ODEHO $ VWDQGDUG FDOLEUDWLRQ FXUYH RI WKH VLJQDO SURGXFHG E\ WKH ODEHO ZLWK UHVSHFW WR WKH FRQFHQWUDWLRQ RI DQDO\WH SUHVHQW FDQ WKHQ EH FRQVWUXFWHG WKXV DOORZLQJ WKH HVWLPDWLRQ RI XQNQRZQ OLJDQG FRQFHQWUDWLRQV IURP WKLV FXUYH 7UDGLWLRQDOO\ LPPXQRDVVD\V KDYH LQYROYHG WKH XVH RI UDGLRLVRWRSHV DV ODEHOV $OWKRXJK UDGLRLPPXQRDVVD\V 5,$f KDYH WKH DGYDQWDJHV RI EHLQJ KLJKO\ VHQVLWLYH DQG LQYXOQHUDEOH WR HQYLURQPHQWDO LQWHUIHUHQFHV H J E\ FRPSRQHQWV RI WKH DVVD\f WKHLU XVH LV DFFRPSDQLHG E\ VHYHUDO GLVDGYDQWDJHV 7KHVH GLVDGYDQWDJHV LQFOXGH WKH HPRWLYH ELDV DJDLQVW WKH XVH RI UDGLRLVRWRSHV WKH FRVWO\ GLVSRVDO RI ZDVWH VSHFLDO UHTXLUHPHQWV IRU DVVD\ KDQGOLQJ DQG WUDLQLQJ RI VWDII WKH ODFN RI VWDELOLW\ RI UDGLRLVRWRSHV DQG WKH FRQVHTXHQW OLPLWHG VKHOIOLIH RI UHDJHQWV >*RVOLQJ @ 7KHUHIRUH WKHUH KDV EHHQ FRQVLGHUDEOH LQWHUHVW LQ GHYHORSLQJ QRQLVRWRSLF ODEHOV VXFK DV HQ]\PH ODEHOV DQG

PAGE 31

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

PAGE 32

KDYH EHHQ GHYHORSHG ZKHUHE\ VWUXFWXUDOO\ VLPLODU DQDO\WHV DUH ILUVW VHSDUDWHG E\ KLJK SHUIRUPDQFH OLTXLG FKURPDWRJUDSK\ +3/&f SULRU WR TXDQWLWDWLRQ E\ LPPXQRDVVD\ 7R WKLV HQG WKH JRDO LQ WKH GHYHORSPHQW RI WKH LPPXQRDVVD\V SURSRVHG LQ WKLV GLVVHUWDWLRQ ZDV WKH GHVLJQ RI UDSLG DQG FRQYHQLHQW DVVD\V IRU RSLRLG SHSWLGHV ZLWK fGHWHFWRUOLNHf SURSHUWLHV LQWHQGHG IRU WKH GHWHUPLQDWLRQ RI RSLRLG SHSWLGH FRQFHQWUDWLRQV LQ +3/& IUDFWLRQV 7KHVH DVVD\V VKRXOG WKHUHIRUH DOORZ WKH VHQVLWLYH DQG VWUDLJKWIRUZDUG DQDO\VLV RI RSLRLG SHSWLGHV EH SUDFWLFDEOH RQ D ODUJH VFDOH DQG HDVLO\ DPHQDEOH WR LQWHQVLYH DXWRPDWLRQ 7ZR GLIIHUHQW DSSURDFKHV WR WKH GHYHORSPHQW RI DQ LPPXQRDVVD\ IRU RSLRLG SHSWLGHV ZHUH FRQFHLYHG XVLQJ OHXFLQH HQNHSKDOLQ /(f DV D PRGHO SHSWLGH %RWK RI WKHVH DSSURDFKHV ZHUH EDVHG RQ WKH H[SORLWDWLRQ RI WKH H[WUHPHO\ KLJK DIILQLW\ H[KLELWHG E\ DYLGLQ IRU ELRWLQ .D PROf (DFK DYLGLQ PROHFXOH KDV IRXU KLJK DIILQLW\ ELQGLQJ VLWHV IRU ELRWLQ VR WKDW WKH FHQWUDO VWUDWHJ\ RI WKHVH DSSURDFKHV OLHV LQ WKH NQRZOHGJH WKDW ZKHQ DYLGLQ LV FRQMXJDWHG WR DQ HQ]\PH RU IOXRUHVFHQW ODEHO LW ZLOO VWLOO ELQG WR ELRWLQ RU ELRWLQ\ODWHG VSHFLHV >:LOFKHN DQG %D\HU @ 7KH ILUVW DSSURDFK DQ HQ]\PHOLQNHG LPPXQRVRUEHQW DVVD\ (/,6$f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

PAGE 33

VXEVHTXHQWO\ GHWHFWHG WKURXJK WKH XVH RI DQ DYLGLQHQ]\PH FRPSOH[ )LJXUH f 7KH VXFFHVVIXO GHVLJQ RI WKLV W\SH RI DVVD\ IRU /( GHSHQGV RQ WKH V\QWKHVLV RI D VXLWDEOH ELRWLQ\ODWHG /( GHULYDWLYH WKDW ZLOO DOORZ WKH IRUPDWLRQ RI D fVDQGZLFKf EHWZHHQ WKH DQWLERG\ WKH ELRWLQ\ODWHG /( GHULYDWLYH DQG WKH HQ]\PHODEHOHG DYLGLQ WKXV HQDEOLQJ TXDQWLILFDWLRQ XSRQ DGGLWLRQ RI WKH HQ]\PH VXEVWUDWH 6XEVWUDWH 'HWHFWLRQ ? $QWLERG\ f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

PAGE 34

/( WR LQWHUDFW ZLWK IOXRUHVFHLQODEHOHG DYLGLQ 7KHUHIRUH LQ WKLV DVVD\ ELRWLQ\ODWHG /( DQG /( LQ WKH VDPSOH RU VWDQGDUG FRPSHWH IRU DQWLERG\ ELQGLQJ VLWHV DQG DIWHU HTXLOLEULXP KDV EHHQ DFKLHYHG IOXRUHVFHQFHODEHOHG DYLGLQ ZKLFK DFWV DV D fGHWHFWRUf PROHFXOH LV DGGHG 8QERXQG ELRWLQ\ODWHG /( LV GHWHUPLQHG E\ WKH LQFUHDVH RI IOXRUHVFHQFH LQWHQVLW\ GXH WR WKH LQWHUDFWLRQ RI WKH XQERXQG ELRWLQ\ODWHG /( DQG WKH IOXRUHVFHQFHODEHOHG DYLGLQ )LJXUH f ,Q FRQWUDVW WR WKH (/,6$ GHVFULEHG DERYH WKH VXFFHVV RI WKLV DVVD\ GHSHQGV RQ WKH V\QWKHVLV RI D ELRWLQ\ODWHG /( GHULYDWLYH ZKLFK ZLOO QRW DOORZ IOXRUHVFHQFHODEHOHG DYLGLQ WR ELQG WR DQWLERG\ERXQG ELRWLQ\ODWHG /( LH ODFN RI VDQGZLFK IRUPDWLRQf ,QFUHDVHG IOXRUHVFHQFH LQWHQVLW\ $QWLERG\ %LRWLQ\ODWHG DQDO\WH $QDO\WH )OXRUHVFHQFHODEHOOHG DYLGP QU )LJXUH $YLGLQELRWLQEDVHG KRPRJHQHRXV IOXRUHVFHQFH LPPXQRDVVD\

PAGE 35

0DWHULDOV /HXFLQH HQNHSKDOLQ OHXFLQH HQNHSKDOLQ/\V )UHXQGfV FRPSOHWH DGMXYDQW )UHXQGfV LQFRPSOHWH DGMXYDQW ERYLQH VHUXP DOEXPLQ SRUFLQH WK\URJOREXOLQ JOXWDU DOGHK\GH OHWK\OGLPHWK\ODPLQRSURS\OfFDUERGLLPLGH VRGLXP DFHWDWH DFWLYDWHG FKDUFRDO GH[WUDQ DYHUDJH PROHFXODU ZHLJKW f DYLGLQ 1K\GUR[\VXFFLQLPLGRELRWLQ ELRWLQDPLGRFDSURDWH 1K\GUR[\VXFFLQLPLGH HVWHU WULHWKDQRODPLQH fK\GUR[\D]R EHQ]HQHEHQ]RLF DFLG GLPHWK\OVXOIR[LGH DQG IOXRUHVFHLQ LVRWKLRF\DQDWH DYLGLQ ZHUH REWDLQHG IURP 6LJPD &KHPLFDO &RPSDQ\ 6W /RXLV 02 86$ 0HWKDQRO DFHWRQLWULOH WULIOXRURDFHWLF DFLG VRGLXP GLK\GURJHQ SKRVSKDWH GLVRGLXP K\GURJHQ SKRVSKDWH VRGLXP FKORULGH SRWDVVLXP GLK\GURJHQ SKRVSKDWH GLSRWDVVLXP K\GURJHQ SKRVSKDWH DQG SRWDVVLXP FKORULGH ZHUH SXUFKDVHG IURP )LVKHU 6FLHQWLILF 3LWWVEXUJK 3$ 86$ >W\URV\O "+1f@OHXFLQH HQNHSKDOLQ ZDV SURFXUHG IURP 1(1 5HVHDUFK 3URGXFWV 'XSRQW &RPSDQ\ :LOPLQJWRQ '( 86$ &\WRVFLQW VFLQWLOODWLRQ FRFNWDLO ZDV REWDLQHG IURP ,&1 %LRPHGLFDOV ,QF ,UYLQH &$ 86$ DQG DQWL/( DQWLVHUXP ZDV SXUFKDVHG IURP 3HQLQVXOD /DERUDWRULHV ,QF %HOPRQW &$ 86$ 0HWKRGV $QWLERG\ 3URGXFWLRQ /HXFLQH HQNHSKDOLQ /(f ZDV FRQMXJDWHG WR HLWKHU SRUFLQH WK\URJOREXOLQ RU ERYLQH VHUXP DOEXPLQ %6$f E\ UHDFWLRQ ZLWK HLWKHU JOXWDUDOGHK\GH RU HWK\OGLPHWK\O DPLQRSURS\OfFDUERGLLPLGH ('&f

PAGE 36

)RU WKH JOXWDUDOGHK\GH PHWKRG $GDPXV SHUVRQDO FRPPXQLFDWLRQf D b YY VWRFN VROXWLRQ RI JOXWDUDOGHK\GH ZDV IUHVKO\ SUHSDUHG RQ LFH DQG GLOXWHG IROG LQ 0 VRGLXP SKRVSKDWH EXIIHU S+ (TXDO DPRXQWV RI /( DQG SURWHLQ FDUULHU %6$ RU SRUFLQH WK\URJOREXOLQf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f $IWHU GLDO\VLV DOLTXRWV RI WKH FRQMXJDWHV ZHUH VWRUHG DW r& SULRU WR O\RSKLOL]DWLRQ )RU WKH ('& PHWKRG >+DUORZ DQG /DQH @ D PJPO VROXWLRQ RI /( ZDV SUHSDUHG LQ ZDWHU DQG ('& ZDV ZHLJKHG RXW DQG DGGHG WR JLYH D ILQDO FRQFHQWUDWLRQ RI PJPO 7KH S+ RI WKH UHDFWLRQ PL[WXUH ZDV DGMXVWHG DQG PDLQWDLQHG DW S+ ZLWK 1 1D2+ WKURXJKRXW WKH PLQXWH LQFXEDWLRQ WLPH DW URRP WHPSHUDWXUH $Q HTXDO YROXPH RI DQ PJPO VROXWLRQ RI SURWHLQ FDUULHU %6$ RU SRUFLQH WK\URJOREXOLQf ZDV DGGHG DQG WKH UHDFWLRQ ZDV DOORZHG WR SURFHHG DW URRP WHPSHUDWXUH IRU KRXUV 7KH UHDFWLRQ ZDV WKHQ VWRSSHG E\ WKH DGGLWLRQ RI D VRGLXP DFHWDWH VROXWLRQ 0 S+ f WR JLYH D ILQDO FRQFHQWUDWLRQ RI QL0 $IWHU DQ DGGLWLRQDO LQFXEDWLRQ RI KRXU DW URRP WHPSHUDWXUH WKH FRQMXJDWH ZDV GLDO\]HG DJDLQVW 0 SKRVSKDWH EXIIHU S+ IRU KRXUV XVLQJ SUH K\GUDWHG 6SHFWUD3RU FHOOXORVH HVWHU PHPEUDQHV ZLWK PROHFXODU ZHLJKW FXW RII RI 6SHFWUXP 0HGLFDO ,QGXVWULHV ,QF /RV $QJHOHV &$ 86$f $IWHU GLDO\VLV DOLTXRWV RI WKH

PAGE 37

FRQMXJDWHV ZHUH VWRUHG DW r& SULRU WR O\RSKLOL]DWLRQ /\RSKLOL]DWLRQ RI WKH FRQMXJDWHV ZDV FDUULHG RXW DW WKH 'UXJ 'HOLYHU\ /DERUDWRU\ 8QLYHUVLW\ RI )ORULGD 3URJUHVV &HQWHU $ODFKXD )/ 86$ XVLQJ D 0RGHO 6XSHU 0RGXO\R /\RSKLOL]HU (GZDUGV :HVW 6XVVH[ (QJODQGf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f DQG WKHUHIRUH WKLV FRQMXJDWH ZDV VHOHFWHG DV WKH LQRFXOXP IRU WKH SXUSRVHV RI WKLV ZRUN )RU WKH LQLWLDO LQRFXODWLRQ FRPSOHWH )UHXQGfV DGMXYDQW ZDV SODFHG LQ D WHVW WXEH DQG DQ HTXDO YROXPH RI 0 VRGLXP SKRVSKDWH EXIIHU S+ FRQWDLQLQJ PJPO RI WKH /(WK\URJOREXOLQ FRQMXJDWH ZDV DGGHG ZKLOH YRUWH[LQJ WR JLYH D WKLFN HPXOVLRQ FRQWDLQLQJ PJPO RI WKH FRQMXJDWH 7KUHH \RXQJ DGXOW PDOH 1HZ =HDODQG :KLWH UDEELWV HDFK UHFHLYHG D WRWDO RI PO RI WKLV HPXOVLRQ VXEFXWDQHRXVO\ DW GLIIHUHQW VLWHV LQ WKH EDFN 7KHUHDIWHU ERRVWHU LQMHFWLRQV ZLWK WKH VDPH GRVH RI FRQMXJDWH ZHUH JLYHQ DW PRQWKO\ LQWHUYDOV LQ WKH VDPH PDQQHU H[FHSW WKDW LQFRPSOHWH )UHXQGfV DGMXYDQW ZDV XVHG LQVWHDG RI FRPSOHWH )UHXQGfV DGMXYDQW 7HVW EOHHGV ZHUH WDNHQ GD\V DIWHU HDFK ERRVW DQG DIWHU VHUXP VHSDUDWLRQ WKH DQWLERG\ WLWHU ZDV WHVWHG E\ UDGLRLPPXQRDVVD\ %ULHIO\ WR WHVW WRWDO ELQGLQJ >W\URV\O+1f@OHXFLQH HQNHSKDOLQ +/( Q0f ZDV LQFXEDWHG

PAGE 38

RYHUQLJKW DW r& ZLWK YDULRXV GLOXWLRQV RI WKH DQWLVHUXP f LQ 0 VRGLXP SKRVSKDWH EXIIHU S+ 7R WHVW QRQVSHFLILF ELQGLQJ "+/( Q0f ZDV LQFXEDWHG RYHUQLJKW DW r& ZLWK /( Q0f DQG YDULRXV GLOXWLRQV RI WKH DQWLVHUXP f LQ 0 VRGLXP SKRVSKDWH EXIIHU S+ $Q LFHFROG VXVSHQVLRQ FRQWDLQLQJ b ZY DFWLYDWHG FKDUFRDO DQG b ZY GH[WUDQ LQ ZDWHU ZDV WKHQ DGGHG WR HDFK VDPSOH DQG DIWHU D PLQXWH LQFXEDWLRQ DW r& WKH VDPSOHV ZHUH FHQWULIXJHG DW J IRU PLQXWHV $Q DOLTXRW RI WKH VXSHUQDWDQW ZDV WKHQ UHPRYHG &\WRVFLQW VFLQWLOODWLRQ FRFNWDLO POf ZDV DGGHG DQG WKH UDGLRDFWLYLW\ &30f UHSUHVHQWLQJ WKH DQWLERG\ERXQG WUDFHU +/(f ZDV GHWHUPLQHG XVLQJ D %HFNPDQ /6 7' VFLQWLOODWLRQ FRXQWHU )XOOHUWRQ &$ 86$f $IWHU WKH WKLUG ERRVW WKH DQWLERG\ WLWHU RI RQH RI WKH UDEELWV ZDV GHHPHG VXIILFLHQW DQG WKHUHIRUH WKLV UDEELW ZDV H[VDQJXLQDWHG WKH VHUXP ZDV VHSDUDWHG IURP ZKROH EORRG E\ FHQWULIXJDWLRQ DW J IRU PLQXWHV '\QDF ,, &HQWULIXJH &OD\ $GDPV 6SDUNV 0' 86$f DQG VWRUHG LQ DOLTXRWV DW r& $ PO DOLTXRW RI WKH DQWL/( DQWLVHUXP ZDV SXULILHG E\ WKH +\EULGRPD &RUH RI WKH ,QWHUGLVFLSOLQDU\ &HQWHU %LRWHFKQRORJ\ 5HVHDUFK 8QLYHUVLW\ RI )ORULGD *DLQHVYLOOH )/ %ULHIO\ D FROXPQ ZDV ILOOHG ZLWK PO RI 3URW **DPPDELQG 8OWUD PDWUL[ 3KDUPDFLD /.% %LRWHFKQRORJ\ ,QF 3LVFDWDZD\ 186$f DQG ZDVKHG ZLWK PO RI HOXWLRQ EXIIHU 0 JO\FLQH S+ f 7KH FROXPQ ZDV WKHQ ZDVKHG H[WHQVLYHO\ ZLWK PO RI ELQGLQJ EXIIHU 0 SKRVSKDWH EXIIHUHG VDOLQH S+ b VRGLXP D]LGHf 7KH DQWLVHUXP ZDV GLOXWHG WZRIROG ZLWK ELQGLQJ EXIIHU ORDGHG RQWR WKH FROXPQ DQG ZDVKHG ZLWK ELQGLQJ EXIIHU XQWLO WKH RSWLFDO GHQVLW\ RI WKH HOXHQW UHDFKHG ]HUR DW ;R (OXWLRQ EXIIHU ZDV WKHQ DGGHG WR WKH FROXPQ WKH HOXHQW ZDV FROOHFWHG LQ PO IUDFWLRQV DQG

PAGE 39

QHXWUDOL]HG LPPHGLDWHO\ ZLWK LO RI 0 7ULV EXIIHU S+ 7KH IUDFWLRQV ZLWK KLJK RSWLFDO GHQVLW\ DW WDVR ZHUH SRROHG DQG WKHQ GHVDOWHG DQG FRQFHQWUDWHG E\ VSLQQLQJ GRZQ WLPHV IRU PLQXWHV LQ D &35(3 FHQWULIXJDO ILOWHU $PLFRQ 'DQYHUV 0$ 86$f 7KLV SURFHGXUH \LHOGHG PO RI SXULILHG DQWL/( DQWLERG\ FRQWDLQLQJ PJ ,J*PO GHWHUPLQHG E\ PHDVXUHPHQW RI RSWLFDO GHQVLW\ DW WDVR &KDUDFWHUL]DWLRQ RI $QWLERG\ 7KH QXPEHU RI VSHFLILF DQWLERG\ VLWHV LQ WKH SXULILHG DQWLERG\ ZDV GHWHUPLQHG E\ PHDQV RI 6FDWFKDUG DQDO\VLV >6FDWFKDUG @ 6DPSOHV FRQWDLQLQJ YDULRXV FRQFHQWUDWLRQV RI +/( DV WUDFHU S0f DQG D GLOXWLRQ RI WKH SXULILHG DQWLERG\ ZHUH VHW XS WR GHWHUPLQH WRWDO ELQGLQJ 7DEOH f 7R GHWHUPLQH QRQVSHFLILF ELQGLQJ VDPSOHV FRQWDLQLQJ DQWLERG\ "+/( S0f DQG D KLJK FRQFHQWUDWLRQ RI /( OAO2 0f ZHUH DOVR VHW XS DW WKH VDPH WLPH $IWHU RYHUQLJKW LQFXEDWLRQ DW r& SL RI DQ LFH FROG VXVSHQVLRQ FRQWDLQLQJ b DFWLYDWHG FKDUFRDO DQG b GH[WUDQ LQ ZDWHU ZDV DGGHG WR HDFK WXEH H[FHSW WRWDO FRXQWVf DQG DIWHU D IXUWKHU PLQXWH LQFXEDWLRQ RQ LFH WKH VDPSOHV ZHUH FHQWULIXJHG DW J IRU PLQXWHV 7KUHH KXQGUHG DQG ILIW\ PLFUROLWHUV RI WKH UHVXOWDQW VXSHUQDWDQW FRQWDLQLQJ WKH DQWLERG\ERXQG IUDFWLRQ RI WKH WUDFHU ZHUH WKHQ UHPRYHG DQG DGGHG WR PO RI &\WRVFLQW VFLQWLOODWLRQ FRFNWDLO 7KH UDGLRDFWLYLW\ &30f UHSUHVHQWLQJ WKH DQWLERG\ERXQG WUDFHU ZDV GHWHUPLQHG XVLQJ D %HFNPDQ /6 7' VFLQWLOODWLRQ FRXQWHU )XOOHUWRQ &$ 86$f DQG D PLQXWH FRXQWLQJ WLPH ZLWK FRXQWLQJ HIILFLHQF\ DW DERXW b

PAGE 40

7DEOH 6DPSOHV VHW XS IRU 6FDWFKDUG DQDO\VLV 7RWDO FRXQWV 7RWDO ELQGLQJ 1RQVSHFLILF ELQGLQJ %XIIHU SL SL +/( SL SL SL /( rn 0f SL $QWLERG\ f SL SL 2YHUQLJKW DW r& WKHQ &KDUFRDO bfGH[WUDQ bf SL SL $ 6FDWFKDUG SORW ZDV FRQVWUXFWHG E\ SORWWLQJ WKH UDWLR RI ERXQG WR IUHH +/( DJDLQVW ERXQG +/( DQG WKH .D YDOXH ZDV GHWHUPLQHG IURP WKH QHJDWLYH VORSH RI WKH OLQH GUDZQ EHWZHHQ WKH SRLQWV 7KH QXPEHU RI VSHFLILF DQWLERG\ VLWHV ZDV GHWHUPLQHG IURP WKH LQWHUFHSW RI WKLV OLQH ZLWK WKH [D[LV %LRWLQ\ODWLRQ RI /HXFLQH (QNHSKDOLQ 1WHUPLQDO ELRWLQ\ODWHG /( %/(f ZDV V\QWKHVL]HG E\ DOORZLQJ QPROHV RI /( QPROHV RI 1K\GUR[\VXFFLQLPLGRELRWLQ %+6f DQG QPROHV RI WULHWKDQRODPLQH LQ SL RI GLPHWK\OVXOIR[LGH '062f WR LQFXEDWH RYHUQLJKW DW URRP WHPSHUDWXUH 7KH SURGXFW RI WKH UHDFWLRQ ZDV VHSDUDWHG IURP WKH UHDJHQWV E\ LQMHFWLQJ WKH LQFXEDWLRQ PL[WXUH LQWR D JUDGLHQW +3/& V\VWHP FRQVLVWLQJ RI D 5DLQLQ 5DEELW +3 VROYHQW GHOLYHU\ V\VWHP 5DLQLQ ,QVWUXPHQW &RPSDQ\ ,QF :REXUQ 0$f FRQWUROOHG E\ D 5DLQLQ '\QDPD[ +3/& PHWKRG PDQDJHU YHUVLRQ 5DLQLQ ,QVWUXPHQW &RPSDQ\ ,QF :REXUQ 0$f D 1HJUHWWL t =DPED LQMHFWRU 6RXWKDPSWRQ 8.f ILWWHG ZLWK D SL ORRS D S%RQGDSDN &L FROXPQ SP [ PP :DWHUV $VVRFLDWHV 0LOIRUG 0$ 86$f DQG D /'&

PAGE 41

0LOWRQ 5R\ 6SHFWURPRQLWRU YDULDEOH ZDYHOHQJWK GHWHFWRU 5LYLHUD %HDFK )/ 86$f 7KH GHWHFWLRQ ZDYHOHQJWK ZDV VHW DW QP 0RELOH SKDVH $ FRQVLVWHG RI b YY DTXHRXV WULIOXRURDFHWLF DFLG b YYf DQG b YY DFHWRQLWULOH DQG PRELOH SKDVH % FRQVLVWHG RI b YY DTXHRXV WULIOXRURDFHWLF DFLG b YYf DQG b YY DFHWRQLWULOH $ JUDGLHQW RI b PRELOH SKDVH $ DQG b PRELOH SKDVH % WR b HDFK RI PRELOH SKDVHV $ DQG % LQ PLQXWHV ZDV UXQ DW D IORZ UDWH RI POPLQ $ FRQWURO LQMHFWLRQ RI WKH LQFXEDWLRQ PL[WXUH ZDV DOVR PDGH LQWR WKH +3/& V\VWHP LPPHGLDWHO\ DIWHU SUHSDUDWLRQ WR DOORZ WKH FDOFXODWLRQ RI WKH H[WHQW RI FRQYHUVLRQ RI /( WR %/( 1WHUPLQDO ELRWLQ\ODWHG /( LQFRUSRUDWLQJ D VSDFHU DUP EHWZHHQ WKH ELRWLQ JURXS DQG WKH SHSWLGH %;/(f ZDV V\QWKHVL]HG LQ WKH VDPH PDQQHU H[FHSW WKDW ELRWLQDPLGRFDSURDWH 1K\GUR[\VXFFLQLPLGH HVWHU %;+6f ZDV XVHG LQVWHDG RI %+6 LQ WKH LQFXEDWLRQ PL[WXUH 7KH SHDNV FRUUHVSRQGLQJ WR %/( DQG %;/( ZHUH FROOHFWHG IURP WKH +3/& HOXHQW DQG WKH YRODWLOH FRPSRQHQWV DFHWRQLWULOH DQG WULIOXRURDFHWLF DFLGf ZHUH HYDSRUDWHG XQGHU D VWUHDP RI QLWURJHQ DW r& 7KH ELRWLQ\ODWHG /( GHULYDWLYHV ZHUH VWRUHG DW r& DQG ZHUH XVHG ZLWKLQ WZR ZHHNV RI V\QWKHVLV &WHUPLQDO ELRWLQ\ODWHG OHXFLQH HQNHSKDOLQ/\V /(/\V%f ZDV V\QWKHVL]HG E\ DOORZLQJ QPROHV RI OHXFLQH HQNHSKDOLQ/\V /(/\Vf QPROHV RI %+6 DQG QPROHV RI WULHWKDQRODPLQH LQ SL RI '062 WR LQFXEDWH IRU WZR KRXUV DW URRP WHPSHUDWXUH 7KH SURGXFW RI WKH UHDFWLRQ ZDV VHSDUDWHG IURP WKH UHDJHQWV E\ LQMHFWLQJ WKLV LQFXEDWLRQ PL[WXUH LQWR WKH VDPH JUDGLHQW +3/& V\VWHP GHVFULEHG DERYH H[FHSW WKDW D JUDGLHQW RI b PRELOH SKDVH $ DQG b PRELOH SKDVH % WR b PRELOH SKDVH $ DQG b PRELOH SKDVH % LQ PLQXWHV ZDV UXQ DW D IORZ UDWH RI POPLQ $ FRQWURO LQMHFWLRQ

PAGE 42

RI QPROHV RI /(/\V LQ SL RI '062 ZDV DOVR PDGH LQWR WKH +3/& V\VWHP WR DOORZ WKH FDOFXODWLRQ RI WKH H[WHQW RI FRQYHUVLRQ RI /(/\V WR /(/\V% &WHUPLQDO ELRWLQ\ODWHG /( LQFRUSRUDWLQJ D VSDFHU DUP EHWZHHQ WKH ELRWLQ JURXS DQG WKH SHSWLGH /( /\V%;f ZDV V\QWKHVL]HG LQ WKH VDPH PDQQHU H[FHSW WKDW %;+6 ZDV XVHG LQVWHDG RI %+6 LQ WKH LQFXEDWLRQ PL[WXUH 7KH SHDNV FRUUHVSRQGLQJ WR /(/\VIf% DQG /(/\V%; ZHUH FROOHFWHG IURP WKH +3/& HOXHQW DQG VWRUHG DV GHVFULEHG HDUOLHU IRU %/( DQG %;/( 7KH VXFFHVVIXO ELRWLQ\ODWLRQ RI /( E\ WKH PHWKRGV GHVFULEHG DERYH ZDV FKDUDFWHUL]HG E\ FRPSDULQJ WKH GLVSODFHPHQW RI fK\GUR[\D]REHQ]HQHfEHQ]RLF DFLG +$%$f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f 7KH VDPH H[SHULPHQW ZDV UHSHDWHG XVLQJ HDFK ELRWLQ\ODWHG /( GHULYDWLYH 7KH VWUXFWXUHV RI /(/\V% DQG /(/\V%; ZHUH FRQILUPHG E\ PDVV VSHFWURPHWULF DQDO\VLV E\ WKH 3URWHLQ &KHPLVWU\ &RUH RI WKH ,QWHUGLVFLSOLQDU\ &HQWHU IRU %LRWHFKQRORJ\ 5HVHDUFK 8QLYHUVLW\ RI )ORULGD *DLQHVYLOOH )/ 86$ $ /DVHUWHFK PDWUL[ DVVLVWHG ODVHU GHVRUSWLRQ 0$/',f WLPHRIIOLJKW 72)f LQVWUXPHQW RSHUDWHG DW N9 DFFHOHUDWLRQ ZDV XVHG ZLWK D QP 89 ODVHU DQG RFF\DQRK\GUR[\FLQQDPLF DFLG DV WKH PDWUL[ 7KH VDPSOH ZDV GLVVROYHG LQ D VROXWLRQ DFHWRQLWULOHZDWHU YYf FRQWDLQLQJ D

PAGE 43

ODUJH H[FHVV RI WKH PDWUL[ JOf DQG D SL DOLTXRW ZDV GHSRVLWHG RQWR WKH VWDLQOHVV VWHHO ODVHU WDUJHW 6HYHUDO VSHFWUD ZHUH DYHUDJHG IRU HDFK H[SHULPHQW $QWLERG\%LQGLQJ RI /( %LRWLQ\ODWHG /( 'HULYDWLYHV DQG 3UHIRUPHG %LRWLQ\ODWHG /($YLGLQ &RPSOH[HV 7KH ELQGLQJ RI /( WKH ELRWLQ\ODWHG /( GHULYDWLYHV DQG SUHIRUPHG ELRWLQ\ODWHG /( DYLGLQ FRPSOH[HV WR ERWK D FRPPHUFLDO DQWL/( DQWLVHUXP 3HQLQVXODf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b ZY %6$ 7KH WUDFHU ZDV +/( r n 0 ILQDO FRQFHQWUDWLRQf DQG WKH FRPSHWLWRU ZDV YDULRXV FRQFHQWUDWLRQV RI HLWKHU /( WKH ELRWLQ\ODWHG /( GHULYDWLYHV RU SUHIRUPHG ELRWLQ\ODWHG /(DYLGLQ FRPSOH[HV $IWHU RYHUQLJKW LQFXEDWLRQ DW r& SL RI DQ LFH FROG VXVSHQVLRQ FRQWDLQLQJ b ZY DFWLYDWHG FKDUFRDO DQG b ZY GH[WUDQ LQ ZDWHU ZDV DGGHG WR HDFK WXEH H[FHSW WRWDO FRXQWVf DQG DIWHU D IXUWKHU PLQXWH LQFXEDWLRQ RQ LFH WKH VDPSOHV ZHUH FHQWULIXJHG DW J IRU PLQXWHV 7KUHH KXQGUHG DQG ILIW\ PLFUROLWHUV RI WKH UHVXOWDQW VXSHUQDWDQW FRQWDLQLQJ WKH DQWLERG\ERXQG IUDFWLRQ RI WKH WUDFHU ZDV WKHQ UHPRYHG DQG DGGHG WR PO RI &\WRVFLQW VFLQWLOODWLRQ FRFNWDLO 7KH UDGLRDFWLYLW\ &30f UHSUHVHQWLQJ WKH DQWLERG\

PAGE 44

ERXQG WUDFHU ZDV GHWHUPLQHG XVLQJ D %HFNPDQ /6 7' VFLQWLOODWLRQ FRXQWHU )XOOHUWRQ &$ 86$f 7DEOH 5DGLRLPPXQRDVVD\ VHW XS 7RWDO FRXQWV 7RWDO ELQGLQJ 6DPSOH 5,$ %XIIHU SL SL 7UDFHU rf 0f SL SL SL &RPSHWLWRU SL $QWLERG\ GLOXWLRQ SL SL 2YHUQLJKW DW r& WKHQ &KDUFRDO bfGH[WUDQ bf SL SL 7KH ,&R YDOXHV FRQFHQWUDWLRQ RI FRPSHWLWRU GLVSODFLQJ b RI ERXQG WUDFHUf IRU /( WKH ELRWLQ\ODWHG /( GHULYDWLYHV RU SUHIRUPHG ELRWLQ\ODWHG /(DYLGLQ FRPSOH[HV ZHUH GHWHUPLQHG XVLQJ WKH 0,164 QRQOLQHDU FXUYHILWWLQJ SURJUDP 0LFUR0DWK 6FLHQWLILF 6RIWZDUH 6DOW /DNH &LW\ 87 86$f 7KH GDWD ZHUH ILWWHG WR WKH IROORZLQJ PRGHO 7r&1 &1 & 16 :KHUH % &30 LQ WKH SUHVHQFH RI FRPSHWLWRU 7 &30 LQ WKH DEVHQFH RI FRPSHWLWRU & FRPSHWLWRU FRQFHQWUDWLRQ 1 VORSH IDFWRU 16 &30 XQGHU QRQVSHFLILF ELQGLQJ FRQGLWLRQV $V IDU DV SRVVLEOH QRQVSHFLILF ELQGLQJ ZDV GHWHUPLQHG LQ WKH SUHVHQFH RI UHODWLYHO\ KLJK FRQFHQWUDWLRQV RI FRPSHWLWRU RUGHUV RI PDJQLWXGH JUHDWHU WKDQ WKH

PAGE 45

,& YDOXHf :KHQ KLJK HQRXJK FRQFHQWUDWLRQV RI FRPSHWLWRU FRXOG QRW EH DFKLHYHG WR GHWHUPLQH QRQVSHFLILF ELQGLQJ WKH QRQVSHFLILF ELQGLQJ YDOXH REWDLQHG IURP WKH /( FXUYH ZDV XVHG WR ILW WKH FXUYH IRU WKH RWKHU FRPSHWLWRUV 'HYHORSPHQW RI WKH 3URSRVHG +RPRJHQHRXV )OXRUHVFHQFH ,PPXQRDVVD\ 'HWHUPLQDWLRQ RI H[FLWDWLRQ DQG HPLVVLRQ PD[LPD IRU IOXRUHVFHQFHODEHOHG DYLGLQ )OXRUHVFHLQ LVRWKLRF\DQDWH DYLGLQ ),7&DYLGLQf ZDV XVHG DV WKH fGHWHFWRUf PROHFXOH LQ WKLV DVVD\ 7R GHWHUPLQH WKH H[FLWDWLRQ DQG HPLVVLRQ PD[LPD PO RI D VROXWLRQ RI ),7&DYLGLQ SPROPOf ZHUH GLVSHQVHG LQWR D TXDUW] IOXRUHVFHQFH FXYHWWH DQG VFDQV RI WKH H[FLWDWLRQ DQG HPLVVLRQ VSHFWUD ZHUH FDUULHG RXW XVLQJ D 3HUNLQ (OPHU /6% IOXRUHVFHQFH VSHFWURSKRWRPHWHU 1RUZDON &7 86$f 7KH VFDQV ZHUH WKHQ UHSHDWHG XSRQ DGGLWLRQ RI %/( SPROPOf WR GHWHUPLQH WKH IOXRUHVFHQFH HQKDQFHPHQW SURGXFHG RQ LQWHUDFWLRQ RI WKHVH WZR UHDJHQWV 'HWHUPLQDWLRQ RI UHDJHQW FRQFHQWUDWLRQV 7KH UHVXOWV RI WKH ELQGLQJ H[SHULPHQWV GHVFULEHG DERYH LQGLFDWHG WKDW WKH DQWLERG\ PDGH LQ WKLV ODERUDWRU\ LQ FRPELQDWLRQ ZLWK WKH 1WHUPLQDO ELRWLQ\ODWHG GHULYDWLYH ZLWKRXW WKH VSDFHU DUP %/(f VKRXOG EH XVHG LQ WKH GHYHORSPHQW RI WKH SURSRVHG KRPRJHQHRXV IOXRUHVFHQFH LPPXQRDVVD\ IRU /( VHH 5HVXOWV DQG 'LVFXVVLRQf 2QH RI WKH ILUVW VWHSV LQ WKH GHYHORSPHQW RI WKLV DVVD\ OD\ LQ WKH GHWHUPLQDWLRQ RI WKH FRQFHQWUDWLRQV RI DQWLERG\ %/( DQG ),7&DYLGLQ WR EH XVHG 7KH GHFLVLRQ DV WR WKH FRQFHQWUDWLRQ RI DQWLERG\ WR EH XVHG LQ WKH ILQDO DVVD\ r n 0 GLOXWLRQ RI SXULILHG DQWLERG\f ZDV PDGH EDVHG RQ SUDFWLFDOLW\ VLQFH LW ZDV DQWLFLSDWHG WKDW D ODUJH DPRXQW RI DQWLERG\ ZRXOG EH UHTXLUHG IRU WKH GHYHORSPHQW RI

PAGE 46

WKH KRPRJHQHRXV IOXRUHVFHQFH LPPXQRDVVD\ DQG D OLPLWHG TXDQWLW\ RI DQWLERG\ REWDLQHG WKURXJK WKH H[VDQJXLQDWLRQ RI D VLQJOH UDEELW ZDV DYDLODEOH 7R HQVXUH ORZ EDFNJURXQG IOXRUHVFHQFH UHDGLQJV WKH DPRXQW RI %/( XVHG LQ WKH ILQDO DVVD\ QHHGHG WR EH VXFK WKDW FORVH WR b FRXOG EH ERXQG E\ WKH DQWLERG\ SUHVHQW :KHQ %/( DQG DQWLERG\ $Ef DUH SUHVHQW LQ WKH VDPH VROXWLRQ WKH HTXLOLEULXP HVWDEOLVKHG FDQ EH GHVFULEHG DV %/( $E %/($E $FFRUGLQJ WR WKH /DZ RI 0DVV $FWLRQ WKH IROORZLQJ HTXDWLRQ FDQ EH VHW XS >%/($E@ f >%/(?>$E@ :KLFK FDQ EH UHDUUDQJHG WR JLYH >%/(@ >%/( $E@ a > $E@.D ,I b RI WKH %/( LQ WKH DVVD\ LV WR EH ERXQG E\ WKH DPRXQW RI DQWLERG\ FKRVHQ IRU WKH DVVD\ r n 0 GLOXWLRQ RI SXULILHG DQWLERG\f WKH OHIW VLGH RI WKH DERYH HTXDWLRQ FDQ EH VHW WR HTXDO $ 6LQFH >$E@ >$ERWDL@ >%/($E@ E\ XVLQJ WKH .D YDOXH REWDLQHG IURP WKH 6FDWFKDUG SORW GHVFULEHG HDUOLHU DVVXPLQJ WKDW %/( DQG /( KDYH WKH VDPH DIILQLW\ IRU WKH DQWLERG\f WKH DERYH HTXDWLRQ FDQ EH VROYHG IRU >%/($E@ >%/($E@ UHSUHVHQWV b RI WKH FRQFHQWUDWLRQ RI %/( WR EH XVHG LQ WKH DVVD\ DQG WKHUHIRUH WKH WRWDO DPRXQW RI %/( WR EH XVHG LQ WKH ILQDO DVVD\ IRU b WR EH ERXQG E\ WKH FKRVHQ DQWLERG\ FRQFHQWUDWLRQ FRXOG EH FDOFXODWHG 7KH FRQFHQWUDWLRQ RI ),7&DYLGLQ WR EH XVHG LQ WKH DVVD\ ZDV GHWHUPLQHG E\ ILQGLQJ WKH FRQFHQWUDWLRQ RI WKLV UHDJHQW ZKLFK ZLOO JLYH PD[LPXP IOXRUHVFHQFH

PAGE 47

HQKDQFHPHQW ZLWK WKH FKRVHQ FRQFHQWUDWLRQ RI %/( 9DULRXV FRQFHQWUDWLRQV RI ),7& DYLGLQ IPROPOf LQ 0 VRGLXP SKRVSKDWH EXIIHU S+ ZHUH DOORZHG WR LQWHUDFW ZLWK YDULRXV FRQFHQWUDWLRQV RI %/( SPROPOf IRU PLQXWHV DIWHU ZKLFK WLPH IOXRUHVFHQFH UHDGLQJV ZHUH WDNHQ DW ;H[F QP DQG ;HP QP XVLQJ D 3HUNLQ (OPHU /6% IOXRUHVFHQFH VSHFWURSKRWRPHWHU 1RUZDON &7 86$f +RPRJHQHRXV IOXRUHVFHQFH LPPXQRDVVD\ 2QFH WKH FRQFHQWUDWLRQV RI WKH UHDJHQWV WR EH XVHG ZHUH GHWHUPLQHG WKH FRPSOHWH KRPRJHQHRXV IOXRUHVFHQFH LPPXQRDVVD\ VDPSOHV ZHUH VHW XS DV VKRZQ LQ 7DEOH 7RWDO WUDFHU VDPSOHV SLQRO %/( SHU PO RI EXIIHUf ZHUH DOVR SUHSDUHG WR JLYH DQ LQGLFDWLRQ RI WKH PD[LPXP IOXRUHVFHQFH HQKDQFHPHQW ZKLFK FRXOG EH H[SHFWHG GXH WR WKH LQWHUDFWLRQ RI WKH WRWDO DPRXQW RI WUDFHU SUHVHQW ZLWK ),7&DYLGLQ )OXRUHVFHQFH EODQN VDPSOHV ZHUH VHW XS WR GHWHUPLQH WKH IOXRUHVFHQFH GXH WR WKH QDWLYH IOXRUHVFHQFH RI ),7& DYLGLQ SUHVHQW LQ HDFK VDPSOH 7KH EXIIHU XVHG FRQVLVWHG RI 0 VRGLXP SKRVSKDWH EXIIHU S+ FRQWDLQLQJ b ZY %6$ $IWHU RYHUQLJKW LQFXEDWLRQ DW r& SL RI ),7& DYLGLQ SPROPOf ZHUH DGGHG WR HDFK VDPSOH WR JLYH D ILQDO FRQFHQWUDWLRQ RI IPROPO 7KH LQWHUDFWLRQ EHWZHHQ WKH IUHH WUDFHU %/(f DQG WKH fGHWHFWRUf PROHFXOH 7DEOH +RPRJHQHRXV IOXRUHVFHQFH LPPXQRDVVD\ VHWXS 7RWDO WUDFHU 7RWDO ELQGLQJ 6DPSOH ), EODQN %XIIHU SL SL YDULRXV SL %/( SPROPOf SL SL SL /( YDULRXV 3XULILHG DQWLERG\ f SL SL 7RWDO YROXPH SL SL SL SL

PAGE 48

),7&DYLGLQf ZDV DOORZHG WR SURFHHG IRU PLQXWHV DW URRP WHPSHUDWXUH LQ WKH GDUN DIWHU ZKLFK IOXRUHVFHQFH UHDGLQJV ZHUH WDNHQ DW QP DQG QP XVLQJ D 3HUNLQ (OPHU /6% IOXRUHVFHQFH VSHFWURSKRWRPHWHU 1RUZDON &7 86$f 5HVXOWV DQG 'LVFXVVLRQ $QWLERG\ 3URGXFWLRQ ,W ZDV DQWLFLSDWHG WKDW WKH GHYHORSPHQW RI WKH KRPRJHQRXV IOXRUHVFHQFH LPPXQRDVVD\ LQ SDUWLFXODU ZRXOG UHTXLUH ODUJH TXDQWLWLHV RI DQWLERG\ 7KHUHIRUH LW ZDV GHFLGHG WR SURGXFH D SRO\FORQDO DQWLERG\ WR OHXFLQH HQNHSKDOLQ LQ WKLV ODERUDWRU\ + + 5f§1+ &f§ &+ff§& KQf§5 + + + 5f§ 1 f§ &+ff§ 1f§5 )LJXUH &RXSOLQJ RI SHSWLGH WR SURWHLQ FDUULHU XVLQJ JOXWDUDOGHK\GH 5L SHSWLGH RU SURWHLQ FDUULHU 5 SHSWLGH RU SURWHLQ FDUULHU $V /( LV QRW DQWLJHQLF LWVHOI WKH SHSWLGH KDG WR EH FRQMXJDWHG WR D SURWHLQ FDUULHU LQ RUGHU WR VWLPXODWH DQ LPPXQH UHDFWLRQ LQ WKH KRVW DQLPDO *OXWDUDOGHK\GH LV D ELOXQFWLRQDO FRXSOLQJ UHDJHQW WKDW ELQGV WZR FRPSRXQGV SULPDULO\ WKURXJK WKHLU DPLQR JURXSV 6LQFH /( FRQWDLQV RQO\ RQH DPLQR JURXS DW LWV 1WHUPLQDO HQG XVLQJ WKH JOXWDUDOGHK\GH PHWKRG RI FRQMXJDWLRQ WKH SHSWLGH LV FRQMXJDWHG WR WKH SURWHLQ FDUULHU YLD

PAGE 49

WKH 1WHUPLQDO HQG )LJXUH f 7KHUHIRUH RQ LQRFXODWLRQ ZLWK VXFK D FRQMXJDWH RQH ZRXOG H[SHFW WKH DQWLERG\ WKDW LV SURGXFHG WR EH GLUHFWHG DJDLQVW WKH &WHUPLQDO HQG RI WKH SHSWLGH F R FKf§FKf§Q F Q 2+ + &+f1&+ FK 7r F R 2 + , FKf§FKQKF Qf§&+ff§Q FK FK KQU 7 F R 1+ U + FKf§FKf§QKFQKf§&+ff§QFK FK )LJXUH &RXSOLQJ RI SHSWLGH WR SURWHLQ FDUULHU XVLQJ ('& 5L SHSWLGH RU SURWHLQ FDUULHU 5 SHSWLGH RU SURWHLQ FDUULHU ,Q FRQWUDVW ('& DOORZV WKH FRQMXJDWLRQ RI WKH SHSWLGH WR WKH SURWHLQ FDUULHU YLD HLWKHU WKH 1WHUPLQDO RU WKH &WHUPLQDO HQG DV FDUERGLLPLGHV DWWDFN FDUER[\OLF JURXSV WR FKDQJH WKHP LQWR UHDFWLYH VLWHV IRU IUHH DPLQR JURXSV 7KHUHIRUH WKH FDUER[\OLF JURXSV RQ

PAGE 50

WKH SURWHLQ FDUULHU ZLOO FRQMXJDWH WR /( YLD WKH 1WHULQLQDO DPLQR JURXS RI WKH SHSWLGH DQG /( ZLOO DOVR FRQMXJDWH WR WKH SURWHLQ FDUULHU WKURXJK LWV &WHUPLQDO HQG E\ DWWDFNLQJ WKH IUHH DPLQR JURXSV RQ WKH SURWHLQ FDUULHU )LJXUH f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b RI WKH WRWDO UDGLRDFWLYLW\ DGGHG ZDV ERXQG DQG b RI WKLV ERXQG UDGLRDFWLYLW\ ZDV GHWHUPLQHG WR EH GXH WR QRQVSHFLILF ELQGLQJ )LJXUH f $W WKDW WLPH WKLV WLWHU ZDV GHHPHG WR EH VXIILFLHQW DQG WKHUHIRUH WKH DQWLVHUXP ZDV KDUYHVWHG

PAGE 51

)LJXUH %DU JUDSK VKRZLQJ WRWDO FRXQWV WRWDO ELQGLQJ DQG QRQVSHFLILF ELQGLQJ RI )//( XVLQJ GLOXWLRQ RI DQWLERG\ SURGXFHG LQ WKLV ODERUDWRU\ &KDUDFWHUL]DWLRQ RI $QWLERG\ 7KH 6FDWFKDUG SORW RI ERXQGIUHH +/( YHUVXV ERXQG +/( LV VKRZQ LQ )LJXUH 7KH .D YDOXH GHWHUPLQHG IURP WKH QHJDWLYH VORSH RI WKH OLQH GUDZQ EHWZHHQ WKH SRLQWV ZDV r PRO ZKLFK FRQYHUWV WR D .D YDOXH RI r PRO 7KH QXPEHU RI VSHFLILF DQWLERG\ VLWHV SUHVHQW LQ WKH ILQDO LQFXEDWLRQ IRU WKLV H[SHULPHQW ZDV GHWHUPLQHG IURP WKH [D[LV LQWHUFHSW WR EH rn PRO 6LQFH r 2n PRO RI ,J* ZDV XVHG LQ WKLV SUHSDUDWLRQ WKH SXULILHG DQWLERG\ FRQWDLQV RQO\ b VSHFLILF DQWLERG\ VLWHV %\ H[WUDSRODWLRQ WKH VWRFN VROXWLRQ RI SXULILHG DQWLERG\ ZDV GHWHUPLQHG WR FRQWDLQ r f PRO RI VSHFLILF DQWLERG\ VLWHV 7KH DIILQLW\ RI WKH DQWLERG\ PDGH LQ WKLV ODERUDWRU\ WR /( FRPSDUHG IDYRUDEO\ WR WKH FRPPHUFLDO DQWLVHUXP REWDLQHG IURP 3HQLQVXOD XQGHU WKH VDPH DVVD\ FRQGLWLRQV VLQFH & YDOXHV IRU /( XVLQJ D GLOXWLRQ RI WKH DQWLERG\

PAGE 52

SURGXFHG LQ WKLV ODERUDWRU\ r n PRODVVD\f ZDV VLPLODU WR WKRVH REWDLQHG XVLQJ D GLOXWLRQ RI WKH FRPPHUFLDO DQWLVHUXP r nn PRODVVD\f 3HQLQVXOD FODLPHG WKDW ,& YDOXHV RI r 2n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

PAGE 53

V\QWKHVL]HG DQG WHVWHG 6RPH RI WKH ELRWLQ\ODWHG GHULYDWLYHV LQFOXGHG D VSDFHU DUP EHWZHHQ WKH ELRWLQ JURXS DQG WKH SHSWLGH )LJXUH f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f§1+ + 2 ,, 3HSWLGHf§1 f§ &f§&+f )LJXUH %LRWLQ\ODWLRQ RI SHSWLGH XVLQJ 1K\GUR[\VXFFLQLPLGRELRWLQ

PAGE 54

)LJXUH %LRWLQ\ODWLRQ RI SHSWLGH XVLQJ ELRWLQDPLGRFDSURDWH 1K\GUR[\VXFFLQLPLGH HVWHU 7KH UHDFWDQWV DQG SURGXFWV RI WKH ELRWLQ\ODWLRQ UHDFWLRQV ZHUH VXFFHVVIXOO\ VHSDUDWHG IURP HDFK RWKHU XVLQJ WKH +3/& V\VWHPV GHVFULEHG SUHYLRXVO\ )LJXUHV DQG f ,Q FRQWURO LQMHFWLRQV FRQWDLQLQJ /( RU /(/\V LQ FRQFHQWUDWLRQV HTXDO WR WKH LQLWLDO FRQFHQWUDWLRQV RI WKHVH FRPSRXQGV LQ WKH UHDFWLRQ PL[WXUH WKH UHDFWDQWV ZHUH VHHQ WR HOXWH DIWHU PLQXWHV )LJXUH $f DQG PLQXWHV )LJXUH $f UHVSHFWLYHO\ 7KH 1 WHUPLQDO ELRWLQ\ODWHG /( GHULYDWLYHV %/( QR VSDFHU DUPf DQG %;/( ZLWK VSDFHU DUPf ZHUH VHHQ WR HOXWH GLVWLQFW IURP WKH UHDFWDQWV DW DQG PLQXWHV UHVSHFWLYHO\ )LJXUHV % DQG &f )RU WKH &WHUPLQDO ELRWLQ\ODWLRQ RI /(/\V WR JLYH /(/\V% QR VSDFHU DUPf DQG /(/\V%; ZLWK VSDFHU DUPf WZR SURGXFW SHDNV ZHUH VHHQ RQ LQMHFWLRQ RI WKH LQFXEDWLRQ PL[WXUHV %DVHG RQ WKH GHOD\HG IRUPDWLRQ LQ WKH WLPH FRXUVH RI WKH UHDFWLRQ RI WKH PLQRU SHDN HOXWLQJ DW PLQXWHV LQ )LJXUH % DQG DW PLQXWHV LQ )LJXUH & WKHVH PLQRU SHDNV ZHUH DVVXPHG WR EH GLELRWLQ\ODWHG SURGXFWV ELRWLQ\ODWHG DW ERWK WKH 1 DQG &WHUPLQDO HQGV RI WKH SHSWLGH $QDO\VLV E\ PDVV VSHFWUR

PAGE 55

)LJXUH &KURPDWRJUDSKV RI FRQWURO LQMHFWLRQ $f VKRZLQJ /( HOXWLQJ DIWHU PLQXWHV DQG UHDFWLRQ LQMHFWLRQV VKRZLQJ %/( QR VSDFHU DUPf HOXWLQJ DIWHU PLQXWHV %f DQG %;/( ZLWK VSDFHU DUPf HOXWLQJ DIWHU PLQXWHV &f PHWU\ 0$/'f RI WKH PDMRU SHDN HOXWLQJ DW PLQXWHV LQ )LJXUH % DQG DW PLQXWHV LQ )LJXUH ,2& UHYHDOHG PROHFXODU LRQ SHDNV >0+@f DW P] DQG UHVSHFWLYHO\ 7KH QRPLQDO PROHFXODU PDVVHV RI WKH PROHFXODU LRQV RI /(/\V% DQG /(/\V%; ZHUH FDOFXODWHG WR EH DQG UHVSHFWLYHO\ 6LQFH WKH PDVV GHWHUPLQDWLRQV GHYLDWHG IURP WKH QRPLQDO PDVVHV RI PRQRELRWLQ\ODWHG GHULYDWLYHV E\ OHVV WKDQ b WKHVH SURGXFWV ZHUH FRQILUPHG DV PRQRELRWLQ\ODWHG GHULYDWLYHV RI /(/\V 7KHUHIRUH DQWLERG\ ELQGLQJ H[SHULPHQWV ZHUH FDUULHG RXW XVLQJ WKHVH FROOHFWHG SHDNV 7KH

PAGE 56

)LJXUH &KURPDWRJUDSKV RI FRQWURO LQMHFWLRQ $f VKRZLQJ /(/\V HOXWLQJ DIWHU PLQXWHV DQG UHDFWLRQ LQMHFWLRQV VKRZLQJ /(/\V% QR VSDFHU DUPf HOXWLQJ DIWHU PLQXWHV %f DQG /(/\V%; ZLWK VSDFHU DUPf HOXWLQJ DIWHU PLQXWHV &f 'LELRWLQ\ODWHG SURGXFWV DUH VHHQ HOXWLQJ DW PLQXWHV %f DQG PLQXWHV &f H[WHQW RI FRQYHUVLRQ IURP WKH VWDUWLQJ PDWHULDOV /( DQG /(/\Vf WR WKH ELRWLQ\ODWHG GHULYDWLYHV ZDV HVWLPDWHG IURP WKH GLIIHUHQFH LQ SHDN KHLJKW RI /( DQG /(/\V LQ WKH FRQWURO LQMHFWLRQ DQG LQ WKH LQMHFWLRQ RI WKH UHDFWLRQ PL[WXUH $Q H[WHQW RI FRQYHUVLRQ RI DSSUR[LPDWHO\ b IURP WKH VWDUWLQJ PDWHULDOV WR WKH ELRWLQ\ODWHG GHULYDWLYHV ZDV HVWLPDWHG 7KH UHVXOWV RI WKH +$%$ GLVSODFHPHQW H[SHULPHQWV FRQILUPHG WKDW WKH SURGXFWV FROOHFWHG IURP WKH +3/& HOXHQW ZHUH LQGHHG ELRWLQ\ODWHG /( GHULYDWLYHV DV WKH\ ZHUH VHHQ WR GLVSODFH +$%$ IURP DYLGLQ )LJXUH f 7KH UHVXOWV RI WKLV H[SHULPHQW FRQILUP WKDW

PAGE 57

)LJXUH 'LVSODFHPHQW RI +$%$ IURP DYLGLQ E\ ELRWLQ DQG ELRWLQ\ODWHG /( GHULYDWLYHV RQ FRQMXJDWLRQ WR /( ELRWLQ UHWDLQV WKH DELOLW\ WR ELQG WR DYLGLQ 7KH FXUYLOLQHDU QDWXUH RI WKH GLVSODFHPHQW FXUYHV IRU %/( %;/( DQG /(/\Vr%; WRJHWKHU ZLWK WKH REVHUYDWLRQ WKDW PRUH RI WKHVH ELRWLQ\ODWHG GHULYDWLYHV FRPSDUHG WR ELRWLQ LV UHTXLUHG WR GLVSODFH WKH VDPH DPRXQW RI +$%$ IURP DYLGLQ LQGLFDWHV WKDW WKHVH GHULYDWLYHV KDYH D VOLJKWO\ ORZHU DIILQLW\ IRU DYLGLQ WKDQ ELRWLQ LWVHOI +RZHYHU KHUH /(/\V% LV VHHPV WR KDYH D VOLJKWO\ KLJKHU DIILQLW\ WR DYLGLQ WKDQ ELRWLQ 7KLV HIIHFW FDQ SUREDEO\ EH DWWULEXWHG WR H[SHULPHQWDO HUURU DV WKH FRQFHQWUDWLRQV RI WKH ELRWLQ\ODWHG GHULYDWLYHV ZHUH FDOFXODWHG IURP HVWLPDWHG FRQYHUVLRQ UDWHV EDVHG RQ WKH UHGXFWLRQ RI WKH SHDN KHLJKWV RI WKH VWDUWLQJ PDWHULDOV LQ WKH ELRWLQ\ODWLRQ UHDFWLRQ 3UHYLRXV LQYHVWLJDWLRQV RQ WKH ELQGLQJ RI ELRWLQ\ODWHG SHSWLGHV RU RWKHU PDFURPROHFXOHV WR DYLGLQ KDYH VKRZQ WKDW D VSDFHU DUP PD\ EH UHTXLUHG WR UHWDLQ IXOO DIILQLW\ WR DYLGLQ >)LQQ HW DO *UHHQ HW DO @ KRZHYHU WKLV REVHUYDWLRQ ZDV QRW UHIOHFWHG LQ WKH UHVXOWV VHHQ KHUH DV /( ELRWLQ\ODWHG ZLWK DQG ZLWKRXW WKH VSDFHU DUP ERWK VKRZHG VLPLODU DIILQLW\ WR DYLGLQ DQG LQ IDFW /(/\V ELRWLQ\ODWHG ZLWKRXW WKH VSDFHU

PAGE 58

DUP VKRZHG D KLJKHU DIILQLW\ WR DYLGLQ WKDQ /(/\V ELRWLQ\ODWHG ZLWK WKH VSDFHU DUP )LJXUH f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f 0HDQ 6WDQGDUG GHYLDWLRQ /( r n r n r 2f rn rn %/( r n r r n %;/( r f r n r 2n %/(DYLGLQ rn %;/(DYLGLQ r n

PAGE 59

7KH &WHUPLQDO ELRWLQ\ODWHG /( GHULYDWLYHV /(/\V% DQG /(/\V%; VKRZHG QR ELQGLQJ WR WKH FRPPHUFLDO DQWLVHUXP LQ WKH FRQFHQWUDWLRQV XVHG GDWD QRW VKRZQf DQG WKHUHIRUH XVLQJ WKLV SUHSDUDWLRQ WKH\ ZHUH XQVXLWDEOH IRU WKH GHYHORSPHQW RI HLWKHU RI WKH LPPXQRDVVD\V SURSRVHG LQ WKLV VWXG\ ,W LV SRVVLEOH WKDW WKH LQFOXVLRQ RI D ORQJHU VSDFHU DUP LQ WKH &WHUPLQDO ELRWLQ\ODWHG GHULYDWLYHV ZRXOG KDYH DOORZHG ELQGLQJ RI WKH FRPPHUFLDO DQWLVHUXP LI WKH HSLWRSH WR EH UHFRJQL]HG E\ WKH DQWLERG\ UHPDLQHG LQWDFW RQ ELRWLQ\ODWLRQ +RZHYHU WKLV DYHQXH ZDV QRW H[SORUHG f 4,, ,4 &RPSHWLWRU PRODVVD\f )LJXUH 5HSUHVHQWDWLYHV ELQGLQJ FXUYHV IRU /( ff %/( ‘f %;/( Af %/( DYLGLQ ’f DQG %;/(DYLGLQ f XVLQJ GLOXWLRQ RI FRPPHUFLDO DQWLVHUXP ,Q WKHVH H[SHULPHQWV %;/( VKRZHG D WHQIROG KLJKHU DIILQLW\ WR WKH FRPPHUFLDO DQWLVHUXP WKDQ /( 7DEOH f ,W ZDV SRVWXODWHG WKDW VLQFH /( QHHGV WR EH FRQMXJDWHG WR D

PAGE 60

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f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

PAGE 61

7DEOH ,& YDOXHV REWDLQHG IRU /( %/( %/(DYLGLQ %;/( DQG %;/(DYLGLQ XVLQJ GLOXWLRQ RI DQWLERG\ SURGXFHG LQ WKLV ODERUDWRU\ A YDOXH IRU %/( DYLGLQ ZDV HVWLPDWHG E\ IL[LQJ QRQVSHFLILF ELQGLQJ WR b WRWDO ELQGLQJ DQG 1 WR &RPSHWLWRU ,& PRODVVD\f 0HDQ 6WDQGDUG GHYLDWLRQ /( r f r n r n r f r n r r n r f %/( rn r 2f r n rf r f %/(DYLGLQ rrnf %;/( r rn rf %;/(DYLGLQ rff rff rnf $V ZLWK WKH FRPPHUFLDO DQWLVHUXP WKH 1WHUPLQDO ELRWLQ\ODWHG GHULYDWLYHV ZHUH VHHQ WR UHWDLQ DIILQLW\ IRU WKH DQWLERG\ SURGXFHG LQ WKLV ODERUDWRU\ +RZHYHU D VKLIW LQ DIILQLW\ E\ DOPRVW WKUHH RUGHUV RI PDJQLWXGH LV VHHQ ZKHQ FRPSOH[HV DUH IRUPHG EHWZHHQ %;/( DQG DYLGLQ 7DEOH f DQG SUHIRUPHG %/(DYLGLQ FRPSOH[HV VKRZHG ELQGLQJ WR WKLV DQWLERG\ DW RQO\ WKH KLJKHVW FRQFHQWUDWLRQ WHVWHG )LJXUH f $JDLQ DV ZLWK WKH FRPPHUFLDO DQWLVHUXP %;/( VKRZHG D KLJKHU DIILQLW\ WKDQ /( WR WKH DQWLERG\ SURGXFHG LQ WKLV ODERUDWRU\ +HUH WRR DV EHIRUH LW ZDV SRVWXODWHG WKDW VLQFH /( ZDV FRQMXJDWHG WR D SRUFLQH WK\URJOREXOLQ WR UHQGHU LW LPPXQRJHQLF IRU DQWLERG\ SURGXFWLRQ WKLV ELRWLQ\ODWHG /( GHULYDWLYH LQFOXGLQJ D VSDFHU DUP PD\ KDYH UHVHPEOHG WKH

PAGE 62

HSLWRSH SUHVHQWHG WR WKH LPPXQH V\VWHP PRUH FORVHO\ WKDQ /( LWVHOI DQG DV D UHVXOW %;/( DOVR VKRZHG D KLJKHU DIILQLW\ WR WKLV DQWLERG\ WKDQ /( LWVHOI )LJXUH 5HSUHVHQWDWLYHV ELQGLQJ FXUYHV IRU /( ff %/( ‘f %;/( Af %/( DYLGLQ ’f DQG %;/(DYLGLQ f XVLQJ GLOXWLRQ RI DQWLERG\ SURGXFHG LQ WKLV ODERUDWRU\ 7KH &WHUPLQDO ELRWLQ\ODWHG /( GHULYDWLYHV /(/\V% DQG /(/\V%; VKRZHG QR ELQGLQJ WR WKH DQWLERG\ SURGXFHG LQ WKLV ODERUDWRU\ LQ WKH FRQFHQWUDWLRQV XVHG GDWD QRW VKRZQf DQG WKHUHIRUH LQ WKLV SUHSDUDWLRQ WKH\ ZHUH DOVR XQVXLWDEOH IRU WKH GHYHORSPHQW RI HLWKHU RI WKH LPPXQRDVVD\V SURSRVHG LQ WKLV VWXG\ 7KLV UHVXOW ZDV H[SHFWHG VLQFH WKH DQWLERG\ ZDV SURGXFHG LQ WKLV ODERUDWRU\ IROORZLQJ LQRFXODWLRQ ZLWK /( FRQMXJDWHG WR D FDUULHU SURWHLQ YLD WKH 1WHUPLQDO HQG 7KHUHIRUH WKH &WHUPLQDO HQG RI WKH SHSWLGH ZDV

PAGE 63

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

PAGE 64

)LJXUH (PLVVLRQ VFDQV ZLWK ;H[F DW QP IRU ),7&DYLGLQ DQG ),7&DYLGLQ ZLWK %/( 'HWHUPLQDWLRQ RI UHDJHQW FRQFHQWUDWLRQV $Q DQWLERG\ GLOXWLRQ RI FRUUHVSRQGLQJ WR D FRQFHQWUDWLRQ RI SPROPO RI VSHFLILF DQWLERG\ VLWHV ZDV FKRVHQ IRU XVH LQ WKHVH H[SHULPHQWV $OWKRXJK D KLJKHU FRQFHQWUDWLRQ RI VSHFLILF DQWLERG\ VLWHV LQ WKH ILQDO DVVD\ ZRXOG KDYH EHHQ SUHIHUUHG WR DIIRUG ELQGLQJ RI D JUHDWHU QXPEHU RI WUDFHU %/(f PROHFXOHV WKH GHFLVLRQ ZDV EDVHG RQ SUDFWLFDOLW\ VLQFH D OLPLWHG DPRXQW RI DQWLERG\ ZDV DYDLODEOH IURP WKH H[VDQJXLQDWLRQ RI D VLQJOH UDEELW 7KH FDOFXODWLRQV XVLQJ WKH FKRVHQ DQWLERG\ FRQFHQWUDWLRQ WKH .D YDOXH REWDLQHG IURP WKH 6FDWFKDUG SORW DQG WKH /DZ RI 0DVV $FWLRQ LQGLFDWHG WKDW D FRQFHQWUDWLRQ RI

PAGE 65

%/( RI SPROPO ZRXOG EH b ERXQG E\ WKH FKRVHQ DQWLERG\ FRQFHQWUDWLRQ 7KHUHIRUH WKLV FRQFHQWUDWLRQ RI WUDFHU ZDV FKRVHQ IRU XVH LQ DOO IXUWKHU H[SHULPHQWV )LJXUH 3ORWV RI IOXRUHVFHQFH YHUVXV DPRXQW RI %/( DGGHG XVLQJ IPROPO RI ),7&DYLGLQ $f DQG IPROPO RI ),7&DYLGLQ %f $OWKRXJK RQH ZRXOG H[SHFW WKDW DERXW WLPHV DV PXFK %/( DV ),7&DYLGLQ ZRXOG SURGXFH PD[LPXP IOXRUHVFHQFH HQKDQFHPHQW RI ),7&DYLGLQ VLQFH LW LV NQRZQ WKDW PROHV RI ELRWLQ ZLOO ELQG WR PROH RI DYLGLQ )LJXUH $ VKRZV WKDW D PD[LPXP IOXRUHVFHQFH UHDGLQJ LV REWDLQHG ZKHQ SPROPO RI %/( LV DGGHG WR IPROPO RI

PAGE 66

),7&DYLGLQ 7KH UHVXOWV RI WKH +$%$ GLVSODFHPHQW H[SHULPHQWV )LJXUH f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f WKLV FRQFHQWUDWLRQ RI ),7&DYLGLQ ZDV FKRVHQ IRU XVH LQ WKLV DVVD\ VLQFH OHVV QRLVH LV REVHUYHG LQ WKH UHDGLQJV DQG WKH ZRUNLQJ VFDOH LV PRUH SUDFWLFDO +RPRJHQHRXV IOXRUHVFHQFH LPPXQRDVVD\ 2QH RI WKH ILUVW FDOLEUDWLRQ FXUYHV REWDLQHG IRU /( XVLQJ WKH FRPSOHWH KRPRJHQHRXV IOXRUHVFHQFH LPPXQRDVVD\ LV VKRZQ LQ )LJXUH +HUH DV H[SHFWHG DQ LQFUHDVH LQ IOXRUHVFHQFH LV VHHQ ZLWK DQ LQFUHDVH RI /( SUHVHQW LQ WKH VDPSOH +RZHYHU XVLQJ WKH /( FRQFHQWUDWLRQV VKRZQ KHUH D SODWHDX DW WKH PD[LPXP WKHRUHWLFDO IOXRUHVFHQFH WRWDO WUDFHU YDOXH QR DQWLERG\ SUHVHQWf ZDV QRW UHDFKHG 7KHUHIRUH DQRWKHU FDOLEUDWLRQ FXUYH ZDV FRQVWUXFWHG XVLQJ KLJKHU FRQFHQWUDWLRQV RI /( LQ WKH VDPSOHV LQ DQ DWWHPSW WR UHDFK WKLV SODWHDX )LJXUH f )LJXUH VKRZV D FDOLEUDWLRQ FXUYH IRU /( XVLQJ WKH FRPSOHWH KRPRJHQHRXV IOXRUHVFHQFH LPPXQRDVVD\ ZKHUH IOXRUHVFHQFH UHDGLQJV DERYH WKH PD[LPXP WKHRUHWLFDO IOXRUHVFHQFH ZHUH REWDLQHG DW KLJKHU FRQFHQWUDWLRQV RI /( r 2n PROHVPO DQG DERYHf ,Q RUGHU WR GHWHUPLQH ZKHWKHU WKLV XQH[SHFWHG HIIHFW FRXOG EH DWWULEXWHG WR HLWKHU WKH

PAGE 67

DQWLERG\ WKH WUDFHU RU WKH DQDO\WH WKH DVVD\ ZDV UHSHDWHG RPLWWLQJ HDFK RI WKHVH UHDJHQWV LQ WXUQ D ( 7RWDO WUDFHU ), EODQN R ,( ,( ,( ,( /(PROHVQL,f )LJXUH &DOLEUDWLRQ FXUYH IRU /( XVLQJ KRPRJHQHRXV IOXRUHVFHQFH LPPXQRDVVD\ 4 ORR A A 7RWDO WUDFHU A A A A ), EODQN A A A ,( ,( ,( ,( /)PROFVPO f§L ,( ,( )LJXUH &DOLEUDWLRQ FXUYH IRU /( XVLQJ KRPRJHQHRXV IOXRUHVFHQFH LPPXQRDVVD\ VKRZLQJ UHDGLQJV H[FHHGLQJ PD[LPXP WKHRUHWLFDO IOXRUHVFHQFH

PAGE 68

,( ,( ,( ,( ,( ,( /( LQROHVPOf )LJXUH &DOLEUDWLRQ FXUYH IRU /( XVLQJ WKH KRPRJHQHRXV IOXRUHVFHQFH LPPXQRDVVD\ VHW XS ZLWK %/( ‘f DQG ZLWKRXW %/('f ,( ,( ,( ,( ,( ,( /(QLXOFVLQOf )LJXUH &DOLEUDWLRQ FXUYH IRU /( XVLQJ WKH KRPRJHQHRXV IOXRUHVFHQFH LPPXQRDVVD\ VHW XS RPLWWLQJ DQWLERG\ EXW ZLWK %/( ‘f DQG ZLWKRXW %/('f )LJXUH VKRZV WKDW DW LQ WKH SUHVHQFH RI WKH DQWLERG\ ZKHQ WKH WUDFHU %/(f LV RPLWWHG IURP WKH DVVD\ IOXRUHVFHQFH UHDGLQJV EHJDQ WR LQFUHDVH DERYH WKH IOXRUHVFHQFH

PAGE 69

EODQN OHYHOV DW FRQFHQWUDWLRQV RI /( LQ H[FHVV RI r 2n PROHVPO 7KLV HIIHFW ZDV DOVR REVHUYHG ZKHQ WKH DQWLERG\ ZDV RPLWWHG IURP WKH DVVD\ )LJXUH f :KHQ ERWK WUDFHU DQG DQWLERG\ ZHUH LQFOXGHG LQ WKH DVVD\ )LJXUH f IOXRUHVFHQFH OHYHOV EHJDQ WR LQFUHDVH DERYH IOXRUHVFHQFH EODQN OHYHOV DW FRQFHQWUDWLRQV RI /( LQ H[FHVV RI r PROHV PO :KHQ DQWLERG\ LV RPLWWHG EXW WUDFHU LV LQFOXGHG LQ WKH DVVD\ IOXRUHVFHQFH UHDGLQJV LQFUHDVHG DERYH WRWDO WUDFHU OHYHOV DW /( FRQFHQWUDWLRQV LQ H[FHVV RI r 2n PROHVPO )LJXUH f 7KLV LV WKH VDPH FRQFHQWUDWLRQ RI /( DW ZKLFK WKH IOXRUHVFHQFH UHDGLQJV LQFUHDVHG DERYH IOXRUHVFHQFH EODQN OHYHOV ZKHQ ERWK WUDFHU DQG DQWLERG\ ZHUH RPLWWHG IURP WKH DVVD\ 'FVnW\U /,&PROHVPOf )LJXUH &DOLEUDWLRQ FXUYH IRU GHV7\U /( XVLQJ WKH KRPRJHQHRXV IOXRUHVFHQFH LPPXQRDVVD\ VHW XS RPLWWLQJ DQWLERG\ EXW ZLWK %/( ‘f DQG ZLWKRXW %/('f 7KHVH UHVXOWV OHG XV WR EHOLHYH WKDW DW KLJKHU FRQFHQWUDWLRQV RI /( WKH DQDO\WH LQWHUDFWHG GLUHFWO\ ZLWK ),7&DYLGLQ WR JLYH LQFUHDVHG IOXRUHVFHQFH UHDGLQJV ,W ZDV K\SRWKHVL]HG WKDW WKLV HIIHFW FRXOG EH GXH WR DQ LQWHUDFWLRQ EHWZHHQ WKH W\URVLQH PRLHW\ LQ

PAGE 70

WKH SRVLWLRQ RI /( DQG WKH IOXRUHVFHLQ JURXSV RI WKH fGHWHFWRUf PROHFXOH ),7&DYLGLQ ,Q RUGHU WR WHVW WKLV K\SRWKHVLV DVVD\V ZHUH FDUULHG RXW LQ WKH DEVHQFH RI DQWLERG\ LQ WKH VDPH ZD\ DV GHVFULEHG DERYH +RZHYHU KHUH GHV7\U /( RU SHQWDJO\FLQH ZHUH XVHG DV DQDO\WHV LQ FRQFHQWUDWLRQV VLPLODU WR WKRVH RI /( ZKLFK KDG EHHQ XVHG SUHYLRXVO\ )LJXUH &DOLEUDWLRQ FXUYH IRU SHQWDJO\FLQH XVLQJ WKH KRPRJHQHRXV IOXRUHVFHQFH LPPXQRDVVD\ VHW XS RPLWWLQJ DQWLERG\ EXW ZLWK %/( ‘f DQG ZLWKRXW %/('f )LJXUHV DQG VKRZ WKDW GHV7\U /( DQG SHQWDJO\FLQH GR QRW SURGXFH WKH VDPH LQWHUDFWLRQ ZLWK ),7&DYLGLQ DV /( ZKHQ XVHG LQ WKH VDPH FRQFHQWUDWLRQV LQ WKH KRPRJHQHRXV LPPXQRDVVD\ VHW XS :KHQ %/( LV LQFOXGHG LQ WKH DVVD\ WKH UHDGLQJV DSSURDFK WKRVH REWDLQHG LQ WKH WRWDO WUDFHU VDPSOHV DQG ZKHQ QR %/( LV SUHVHQW LQ WKH DVVD\ WKH UHDGLQJV DSSURDFK WKRVH LQ WKH IOXRUHVFHQFH EODQN VDPSOHV ,W ZDV WKHUHIRUH FRQFOXGHG WKDW WKH HIIHFW VHHQ DW KLJKHU FRQFHQWUDWLRQV RI /( LQ WKH KRPRJHQRXV

PAGE 71

IOXRUHVFHQFH LPPXQRDVVD\ ZHUH SUREDEO\ GXH WR DQ LQWHUDFWLRQ EHWZHHQ WKH W\URVLQH PRLHW\ RI /( DQG WKH IOXRUHVFHLQ JURXSV RI ),7&DYLGLQ 7KH KRPRJHQRXV IOXRUHVFHQFH LPPXQRDVVD\ HYDOXDWHG KHUH RSHUDWHV DV LQWHQGHG LQ WKH FRQFHQWUDWLRQ UDQJH EHWZHHQ r 2n PROHVPO DQG r 2n PROHVPO RI /( VLQFH DW WKHVH FRQFHQWUDWLRQV VXIILFLHQW WUDFHU %/(f LV GLVSODFHG IURP WKH DQWLERG\ E\ /( WR SURGXFH GHWHFWDEOH IOXRUHVFHQFH HQKDQFHPHQW RQ LQWHUDFWLRQ ZLWK WKH GHWHFWRU PROHFXOH ),7& DYLGLQf EXW WKH FRQFHQWUDWLRQ RI /( LV QRW KLJK HQRXJK WR SURGXFH D GLUHFW LQWHUDFWLRQ ZLWK ),7&DYLGLQ LQ WKH DEVHQFH RI %/( 8VLQJ WKH UHDJHQWV WHVWHG DQ KRPRJHQHRXV IOXRUHVFHQFH LPPXQRDVVD\ IRU /( RSHUDWLRQDO RYHU D ZLGH FRQFHQWUDWLRQ UDQJH FRXOG QRW EH GHYHORSHG VLQFH DW KLJKHU FRQFHQWUDWLRQV r 2n PROHVPO DQG DERYHf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

PAGE 72

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n WR f PROHVPO UDQJH >-HQNLQV @ DQG KRPRJHQHRXV HQ]\PH LPPXQRDVVD\V DUH FDSDEOH RI OLPLWV RI GHWHFWLRQ n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n WR n PROHVPO UDQJH

PAGE 73

&+$37(5 $ +,*+ 3(5)250$1&( /,48,' &+520$72*5$3+< $66$< )25 23,2,' 3(37,'(6 86,1* (/(&752&+(0,&$/ '(7(&7,21 ,QWURGXFWLRQ 7KH REMHFWLYH RI WKLV VWXG\ ZDV WKH HYDOXDWLRQ RI D W\URVLQHVSHFLILF FOHDQXS DQG GHWHFWLRQ PHWKRG IRU RSLRLG SHSWLGHV XVLQJ OHXFLQH HQNHSKDOLQ /( )LJXUH f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

PAGE 74

0XVKURRP W\URVLQDVH LV D PHPEHU RI WKH PRQR[\JHQDVH FODVV RI HQ]\PHV ZKLFK FDWDO\VHV WZR VXFFHVVLYH UHDFWLRQV )LJXUH f WKH K\GUR[\ODWLRQ RI PRQRSKHQROV PRQRSKHQRODVH DFWLYLW\f DQG WKH R[LGDWLRQ RI RGLSKHQROV GLSKHQRODVH DFWLYLW\f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

PAGE 75

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

PAGE 76

7KH LQWURGXFWLRQ RI D K\GUR[\W\URVLQH JURXS WR WKH /( PROHFXOH E\ HQ]\PDWLF GHULYDWL]DWLRQ DOORZV WKH XVH RI D VSHFLILF FOHDQXS SURFHGXUH IRU GLK\GUR[\SKHQ\O FRPSRXQGV XVLQJ ERURQDWH JHOV DQG FROXPQ FKURPDWRJUDSK\ )RU WKLV DVVD\ WKH ERURQDWH FOHDQXS PHWKRG LV EDVHG RQ WKH S+GHSHQGHQW IRUPDWLRQ RI D FRPSOH[ EHWZHHQ LPPRELOL]HG ERURQDWH JHO DQG WKH K\GUR[\ODWHG /( GHULYDWLYH )LJXUH f $ FRPSOH[ LV IRUPHG DW ZHDNO\ DONDOLQH S+ EHWZHHQ LRQL]HG ERURQDWH DIIL[HG WR D JHO PDWUL[ DQG WKH K\GUR[\ODWHG /( GHULYDWLYH 'LVVRFLDWLRQ RI WKH FRPSOH[ RFFXUV DW DFLGLF S+ 7KHUHIRUH LQ WKH DVVD\ GHVFULEHG KHUH /( LQ WKH VDPSOH LV ILUVW GHULYDWL]HG HQ]\PDWLFDOO\ E\ PXVKURRP W\URVLQDVH DQG VXEVHTXHQWO\ WKH VDPSOH LV VXEMHFWHG WR FOHDQn XS WKURXJK WKH XVH RI D ERURQDWH JHO FROXPQ )LQDOO\ WKH VDPSOH LV TXDQWLILHG E\ KLJK SHUIRUPDQFH OLTXLG FKURPDWRJUDSK\ ZLWK HOHFWURFKHPLFDO GHWHFWLRQ +3/&('f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

PAGE 77

0HWKRGV 3XULILFDWLRQ RI 0XVKURRP 7\URVLQDVH 0XVKURRP W\URVLQDVH ZDV SXULILHG SULRU WR XVH E\ XOWUDILOWUDWLRQ XVLQJ &HQWULFRQ PHPEUDQH ILOWHUV PROHFXODU ZHLJKW FXW RII $PLFRQ 'DQYHUV 0$ 86$f 2QH PLOOLOLWHU RI D VROXWLRQ RI PXVKURRP W\URVLQDVH PJPOf LQ 0 VRGLXP SKRVSKDWH EXIIHU S+ ZDV DSSOLHG WR WKH ILOWHU XQLW DQG FHQWULIXJHG DW J XQWLO PD[LPXP FRQFHQWUDWLRQ RI WKH VDPSOH ZDV DFKLHYHG 7KLV FHQWULIXJDWLRQ VWHS ZDV UHSHDWHG WKUHH WLPHV ZLWK WKH DGGLWLRQ RI DQ DGGLWLRQDO PO RI SKRVSKDWH EXIIHU SULRU WR HDFK FHQWULIXJDWLRQ 7KH ILQDO FRQFHQWUDWH ZDV UHFRQVWLWXWHG LQ SKRVSKDWH EXIIHU WR JLYH D ILQDO FRQFHQWUDWLRQ RI PJPO RI PXVKURRP W\URVLQDVH FRUUHVSRQGLQJ WR XQLWV RI DFWLYLW\ SHU PO RI VROXWLRQ $OLTXRWV ZHUH VWRUHG DW r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f D 1HJUHWWL DQG =DPED LQMHFWRU 6RXWKDPSWRQ 8.f ILWWHG ZLWK D SL ORRS D 3HUNLQ (OPHU /& VSHFWURSKRWRPHWULF GHWHFWRU 1RUZDON &7 86$f DQG D +HZOHWW 3DFNDUG +3 $

PAGE 78

LQWHJUDWRU $YRQGDOH 3$ 86$f 7KH FROXPQ ZDV D 3DUWLVLO 2'6 [ PP :KDWPDQ /DEVDOHV +LOOVERUR 25 86$f 7KH GHWHFWLRQ ZDYHOHQJWK ZDV QP DQG WKH PRELOH SKDVH FRQVLVWHG RI b DFHWRQLWULOH YYf LQ FLWUDWHGLSRWDVVLXP SKRVSKDWH EXIIHU S+ f DW D IORZ UDWH RI POPLQXWH 7KH SURGXFW SHDNV ZHUH FROOHFWHG IURP WKH +3/& HOXHQW RUJDQLF VROYHQW ZDV UHPRYHG E\ HYDSRUDWLQJ XQGHU D VWUHDP RI QLWURJHQ DQG WKH FROOHFWHG SHDNV ZHUH FRQFHQWUDWHG XVLQJ D SUHFRQGLWLRQHG 6HS3DN &W SUHSDUDWLYH FROXPQ :DWHUV $VVRFLDWHV 0LOIRUG 0$ 86$f 7KH LGHQWLW\ RI WKH SURGXFWV ZDV GHWHUPLQHG E\ HOHFWURVSUD\ LRQL]DWLRQ PDVV VSHFWURPHWU\ VHH &KDSWHU f (OHFWURFKHPLFDO 'HWHFWLRQ (OHFWURFKHPLFDO GHWHFWLRQ ZDV HIIHFWHG XVLQJ D 0RGHO $ &RXORFKHP PXOWLn HOHFWURGH HOHFWURFKHPLFDO GHWHFWRU (6$ ,QF %HGIRUG 0$ 86$f ILWWHG ZLWK D 0RGHO JXDUG FHOO DQG D 0RGHO DQDO\WLFDO FHOO 7KH FRPSOHWH +3/&(' V\VWHP ZDV FRQILJXUHG DV VKRZQ LQ )LJXUH ,Q WKLV V\VWHP WKH JXDUG FHOO DFWV WR SUHR[LGL]H HOHFWURDFWLYH LPSXULWLHV LQ WKH PRELOH SKDVH WKXV UHGXFLQJ EDFNJURXQG FXUUHQW 7KH DQDO\WLFDO FHOO FRQVLVWV RI WZR ZRUNLQJ HOHFWURGHV LQ VHULHV 7KH ILUVW 'HW f DFWV WR IXUWKHU UHGXFH EDFNJURXQG FXUUHQW LQ WKH LQMHFWHG VDPSOH DQG WR SUHR[LGL]H FRHOXWLQJ LQWHUIHUHQFHV WKDW R[LGL]H DW SRWHQWLDOV ORZHU WKDQ WKH DQDO\WH RI LQWHUHVW 7KH VHFRQG ZRUNLQJ HOHFWURGH 'HW f LV VHW WR TXDQWLI\ WKH DQDO\WH 7KH VROYHQW GHOLYHU\ V\VWHP ZDV DQ /'&0LOWRQ 5R\ FRQVWD0HWULF ,,, PHWHULQJ SXPS 5LYLHUD %HDFK )/ 86$f DQG WKH LQMHFWRU ZDV D 5KHRG\QH 0RGHO

PAGE 79

LQMHFWRU &RWDWL &$ 86$f ILWWHG ZLWK D _QO ORRS $ 6SKHULVRUE 2'6 SP [ PP DQDO\WLFDO FROXPQ .H\VWRQH 6FLHQWLILF ,QF %HOOHIRQWH 3$ 86$f ZDV XVHG ZLWK D PRELOH SKDVH RI b DFHWRQLWULOH YYf LQ PRQRVRGLXP SKRVSKDWH EXIIHU P0 S+ f FRQWDLQLQJ PJ RI VRGLXP GRGHF\O VXOIDWH 7KH PRELOH SKDVH ZDV IUHVKO\ SUHSDUHG ILOWHUHG WKURXJK D SP PHPEUDQH ILOWHU DQG GHJDVVHG XQGHU YDFXXP ZLWK VRQLFDWLRQ GDLO\ SULRU WR XVH 0RELOH SKDVH IORZ UDWH ZDV VHW DW POPLQ $ &KURPDWRSDF &5$ LQWHJUDWRU 6KLPDG]X &RUSRUDWLRQ .\RWR -DSDQf ZDV XVHG WR UHFRUG WKH RXWSXW IURP WKH FRQWURO PRGXOH )LJXUH &KURPDWRJUDSKLF FRQILJXUDWLRQ RI HOHFWURFKHPLFDO GHWHFWLRQ V\VWHP (OHFWURFKHPLFDO &KDUDFWHUL]DWLRQ /( DQG PRQRK\GUR[\ODWHG OHXFLQH HQNHSKDOLQ >+27\Uf@/(f WKH PDMRU SURGXFW RI WKH HQ]\PDWLF UHDFWLRQ ZHUH FKDUDFWHUL]HG HOHFWURFKHPLFDOO\ XVLQJ WKH +3/&('

PAGE 80

V\VWHP GHVFULEHG DERYH )LYH QDQRPROHV RI DQDO\WH SHU LQMHFWLRQ ZHUH DSSOLHG WR WKH V\VWHP DQG WKH UHVSRQVHV LQ S$f REWDLQHG DW YDULRXV SRWHQWLDOV WR 9f DW WKH DQDO\WLFDO FHOO ZHUH UHFRUGHG WR DOORZ WKH FRQVWUXFWLRQ RI FXUUHQWYROWDJH FXUYHV %RURQDWH &OHDQXS $ K\GUDWHG ERURQDWH FROXPQ SDFNHG ZLWK D PO EHG YROXPH RI $IILJHO %LR5DG /DERUDWRULHV 0HOYLOOH 1< 86$f ZDV XVHG IRU WKH FOHDQXS RI WKH HQ]\PDWLFDOO\ GHULYHG VSHFLHV SULRU WR DSSOLFDWLRQ WR WKH +3/&(' V\VWHP 7KH FROXPQ ZDV SUHULQVHG ZLWK 0 SKRVSKDWH EXIIHU S+ f DQG DQ DOLTXRW RI WKH LQFXEDWLRQ PL[WXUH ZDV DSSOLHG 7KH FROXPQ ZDV ZDVKHG ZLWK PO RI SKRVSKDWH EXIIHU S+ f DQG WKH DQDO\WH ZDV HOXWHG IURP WKH ERURQDWH FROXPQ RQWR D SUHFRQGLWLRQHG 6HS3DN &LJ FDUWULGJH :DWHUV $VVRFLDWHV 0LOIRUG 0$ 86$f ZLWK PO RI DTXHRXV b YYf WULIOXRURDFHWLF DFLG 7)$f +HUH WKH 6HS3DN &LJ FDUWULGJH VHUYHG WR FRQFHQWUDWH WKH VDPSOH DV WKH DQDO\WH ZDV HOXWHG IURP WKH ERURQDWH FROXPQ LQ D UHODWLYHO\ ODUJH YROXPH RI DTXHRXV b YYf 7)$ 7KH 6HS3DN &LJ FROXPQ ZDV WKHQ ZDVKHG ZLWK PO RI DTXHRXV b YYf 7)$ DQG WKH DQDO\WH ZDV HOXWHG LQ PO RI b 7)$ LQ DFHWRQLWULOH 7KH VDPSOH ZDV HYDSRUDWHG WR GU\QHVV XQGHU D VWUHDP RI QLWURJHQ DQG WKH UHVLGXH ZDV UHFRQVWLWXWHG LQ SL PRELOH SKDVH SULRU WR DSSOLFDWLRQ WR WKH +3/&(' V\VWHP

PAGE 81

7LPH &RXUVH RI (Q]\PDWLF 'HULYDWL]DWLRQ 7R GHWHUPLQH WKH WLPH FRXUVH RI WKH HQ]\PDWLF GHULYDWL]DWLRQ DW WKH DQDO\WLFDO FRQFHQWUDWLRQV WKH IROORZLQJ LQFXEDWLRQ PL[WXUH ZDV VHW XS LQ PLFURFHQWULIXJH WXEHV rn 0 /( P0 DVFRUELF DFLG DQG XQLWVPO PXVKURRP W\URVLQDVH LQ SL RI 0 SKRVSKDWH EXIIHU S+ 7KH UHDFWLRQ ZDV VWRSSHG DW YDULRXV WLPH SRLQWV E\ DGGLQJ SL RI 1 +& 7ZR KXQGUHG DQ ILIW\ PLFURPHWHUV RI WKLV LQFXEDWLRQ PL[WXUH ZDV WKHQ VXEMHFWHG WR WKH ERURQDWH FOHDQXS SURFHGXUH WKH UHVXOWLQJ VDPSOH ZDV LQMHFWHG LQWR WKH +3/&(' V\VWHP DQG WKH SHDN DUHDV RI WKH SHDN FRUUHVSRQGLQJ WR WKH HQ]\PDWLF GHULYDWLYH >+27\UnM/( ZHUH UHFRUGHG ([WUDFWLRQ RI /HXFLQH (QNHSKDOLQ IURP &HUHEURVSLQDO )OXLG /HXFLQH HQNHSKDOLQ ZDV H[WUDFWHG IURP KXPDQ FHUHEURVSLQDO IOXLG &6)f WKURXJK WKH XVH RI 6XSHOFOHDQ /& VROLG SKDVH H[WUDFWLRQ FROXPQV 6XSHOFR ,QF %HOOHIRQWH 3$ 86$f 7KH FROXPQV ZHUH DFWLYDWHG ZLWK PO HDFK RI ZDWHU DQG PHWKDQRO DQG ORDGHG ZLWK SL RI VSLNHG &6) 6XEVHTXHQWO\ WKH FROXPQV ZHUH ZDVKHG ZLWK PO RI ZDWHU PO RI 1 +& PO RI ZDWHU PO RI 0 ERUDWH EXIIHU S+ f DQG PO RI ZDWHU 7KH /( ULFK IUDFWLRQ ZDV WKHQ HOXWHG LQ PO RI PHWKDQRO DQG HYDSRUDWHG WR GU\QHVV XQGHU D VWUHDP RI QLWURJHQ &DOLEUDWLRQ &XUYHV 8VLQJ WKH FRPSOHWH +3/&(' PHWKRG LQFOXGLQJ HQ]\PDWLF GHULYDWL]DWLRQ DQG ERURQDWH FOHDQXS FDOLEUDWLRQ FXUYHV ZHUH FRQVWUXFWHG IRU /( LQ ERWK EXIIHU DQG &6) )RU

PAGE 82

FRPSDULVRQ FDOLEUDWLRQ FXUYHV IRU /( LQ &6) ZHUH DOVR FRQVWUXFWHG XVLQJ +3/&(' ZLWKRXW HQ]\PDWLF GHULYDWL]DWLRQ RU ERURQDWH FOHDQXS 5HVXOWV (Q]\PDWLF 'HULYDWL]DWLRQ 3UHOLPLQDU\ H[SHULPHQWV LQGLFDWHG WKDW WKH FRQFHQWUDWLRQ RI PXVKURRP W\URVLQDVH XQLWVPOf XVHG IRU WKH HQ]\PDWLF GHULYDWL]DWLRQ RI /( DOORZHG WKH HIILFLHQW FRQYHUVLRQ RI /( WR LWV K\GUR[\ODWHG GHULYDWLYH ZLWKLQ PLQXWHV GDWD QRW VKRZQf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f 7KHVH WZR SURGXFW SHDNV ZHUH LGHQWLILHG E\ HOHFWURVSUD\ LRQL]DWLRQ PDVV VSHFWURVFRS\ VHH &KDSWHU f DV WKH GL DQG PRQRK\GUR[\ODWHG GHULYDWLYHV RI /( >+2f 7\Un@/( DQG >+27\Un@/( UHVSHFWLYHO\f 7KH UHODWLYH LQWHQVLW\ RI WKH SHDNV REWDLQHG IURP PDVV VSHFWURPHWULF DQDO\VLV VKRZHG WKDW >+2f7\Un@/( LV D PLQRU SURGXFW

PAGE 83

$ )LJXUH &KURPDWRJUDSKV RI D FRQWURO UXQ $f VKRZLQJ /( HOXWLQJ DIWHU PLQXWHV GLVWLQFW IURP DVFRUELF DFLG HOXWLQJ LQ WKH VROYHQW IURQW DIWHU PLQXWHV DQG LQFXEDWLRQ VROXWLRQ %f VKRZLQJ WKH HPHUJHQFH RI WZR QHZ SHDNV ZLWK UHWHQWLRQ WLPHV RI DQG PLQXWHV (OHFWURFKHPLFDO &KDUDFWHUL]DWLRQ &XUUHQWYROWDJH FXUYHV IRU /( DQG >+27\Un@/( DUH VKRZQ LQ )LJXUH :KHQ WKHVH FXUYHV ZHUH FRPSDUHG >+27\Un@/( ZDV VHHQ WR EH R[LGL]HG DW FRQVLGHUDEO\ ORZHU SRWHQWLDOV WKDQ /( $V D UHVXOW RI WKLV H[SHULPHQW WKH IROORZLQJ SRWHQWLDOV ZHUH VHOHFWHG IRU XVH LQ WKH FRQVWUXFWLRQ RI FDOLEUDWLRQ FXUYHV IRU /( XVLQJ WKH FRPSOHWH +3/&('

PAGE 84

PHWKRG LH LQFOXGLQJ HQ]\PDWLF GHULYDWL]DWLRQ DQG ERURQDWH FOHDQXSf DQG +3/&(' ZLWKRXW HQ]\PDWLF GHULYDWL]DWLRQ RU ERURQDWH FOHDQXS *XDUG FHOO $QDO\WLFDO FHOO 'HW $QDO\WLFDO FHOO 'HW +3/&(' ZLWK GHULYDn WL]DWLRQ DQG FOHDQXS 9 9 9 +3/&(' QR GHULYDn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

PAGE 85

%RURQDWH &OHDQXS %\ DSSO\LQJ SPROHV WR WKH ERURQDWH JHO FROXPQ DQ DYHUDJH UHFRYHU\ RI DQDO\WH RI b ZDV DFKLHYHG XVLQJ WKH FRPSOHWH ERURQDWH FOHDQXS PHWKRG 6' Q f 7LPH &RXUVH RI (Q]\PDWLF 'HULYDWL]DWLRQ $ UHSUHVHQWDWLYH WLPH FRXUVH RI WKH HQ]\PDWLF GHULYDWL]DWLRQ DW DQDO\WLFDO FRQFHQWUDWLRQV SPROLQMf VKRZLQJ WKH SHDN DUHD RI >+27\UfM/( SORWWHG DJDLQVW WLPH LV VKRZQ LQ )LJXUH 7KLV H[SHULPHQW ZDV FDUULHG RXW WZLFH VKRZLQJ WKH VDPH WUHQG HDFK WLPH 7KH SORW LQGLFDWHV WKDW WKH KLJKHVW OHYHO RI >+27\Un@/( LV VHHQ DIWHU D PLQXWH LQFXEDWLRQ DQG WKHUHIRUH WKLV LQFXEDWLRQ WLPH ZDV VHOHFWHG IRU IXWXUH XVH LQ WKLV VWXG\ )LJXUH 7LPH FRXUVH RI HQ]\PDWLF GHULYDWL]DWLRQ VKRZLQJ SHDN KHLJKW RI >+27\U@ /( RYHU LQFXEDWLRQ WLPH

PAGE 86

&DOLEUDWLRQ &XUYHV 7DEOH VKRZV WKH OLPLWV RI GHWHFWLRQ /2' GHILQHG DV WZLFH WKH EDVHOLQH QRLVHf DYHUDJH VORSH YDOXHV DQG FRUUHODWLRQ FRHIILFLHQWV Uf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f 7DEOH /LPLWV RI GHWHFWLRQ /2'f DYHUDJH VORSHV DQG FRUUHODWLRQ FRHIILFLHQWV Uf IRU /( LQ EXIIHU RU &6) XVLQJ HLWKHU FRPSOHWH +3/&(' PHWKRG RU +3/&(' ZLWKRXW GHULYDWL]DWLRQ RU ERURQDWH FOHDQXS )RU UDZ GDWD VHH $SSHQGL[ $ 6DPSOH 0HWKRG /2' IPROLQM 6ORSH 6' Qf SHDN DUHDSPROLQM U 6' Qf %XIIHU FRPSOHWH +3/&(' f f &6) FRPSOHWH +3/&(' f f &6) +3/&(' QR GHULYDWL]DWLRQ QR FOHDQXS f f

PAGE 87

)LJXUH 5HSUHVHQWDWLYH FDOLEUDWLRQ FXUYH IRU /( LQ EXIIHU XVLQJ FRPSOHWH +3/&(' PHWKRG )LJXUH 5HSUHVHQWDWLYH FDOLEUDWLRQ FXUYH IRU /( LQ &6) XVLQJ FRPSOHWH +3/&(' PHWKRG

PAGE 88

)LJXUH 5HSUHVHQWDWLYH FDOLEUDWLRQ FXUYH IRU /( LQ &6) XVLQJ +3/&(' ZLWKRXW GHULYDWL]DWLRQ RU ERURQDWH FOHDQXS )LJXUH 6DPSOH FKURPDWRJUDSK RI /( LQ &6) VKRZLQJ SHDN RI LQWHUHVW HOXWLQJ DIWHU PLQXWHV

PAGE 89

7R JLYH DQ LQGLFDWLRQ RI WKH LQWHUGD\ YDULDELOLW\ DVVRFLDWHG ZLWK WKH FRPSOHWH +3/&(' PHWKRG LQFRUSRUDWLQJ HQ]\PDWLF GHULYDWL]DWLRQ DQG ERURQDWH FOHDQ XS XVLQJ WKH FDOLEUDWLRQ FXUYHV REWDLQHG IRU /( LQ &6) QRPLQDO FRQFHQWUDWLRQV LQ WKH VDPSOHV ZHUH FRPSDUHG WR IRXQG FRQFHQWUDWLRQV 7DEOH f 7KH IRXQG FRQFHQWUDWLRQV ZHUH GHWHUPLQHG IURP WKH FDOLEUDWLRQ FXUYH DIWHU UHJUHVVLRQ DQDO\VLV ZDV UHSHDWHG ZKLOH RPLWWLQJ WKH GDWD SRLQW XQGHU LQYHVWLJDWLRQ 7KH UHODWLYH VWDQGDUG GHYLDWLRQ RI WKH IRXQG FRQFHQWUDWLRQV ZDV IRXQG WR EH b Q f DQG WKH DFFXUDF\ ZDV IRXQG WR EH ZLWKLQ b RI WKH QRPLQDO FRQFHQWUDWLRQ 7DEOH 1RPLQDO FRQFHQWUDWLRQV IRXQG FRQFHQWUDWLRQV UHODWLYH VWDQGDUG GHYLDWLRQ 6'f DQG SHUFHQWDJH DFFXUDF\ FDOFXODWHG IURP FDOLEUDWLRQ FXUYHV IRU /( LQ &6) XVLQJ FRPSOHWH +3/&(' PHWKRG )RU UDZ GDWD VHH $SSHQGL[ $ 1RPLQDO FRQH SPRO SL &6)f )RXQG FRQH Q SPRO SL &6)f 5HODWLYH 6' bf b $FFXUDF\ 'LVFXVVLRQ 7KH +3/&(' PHWKRG ZH KDYH HYDOXDWHG IRU WKH DQDO\VLV RI /( LQ &6) FRPSDUHV IDYRUDEO\ WR H[LVWLQJ +3/& PHWKRGV IRU RSLRLG SHSWLGHV ZLWK RQOLQH GHWHFWLRQ >GH 0RQWLJQ\ HW DO 0LIXQH HW DO 0XFN DQG +HQLRQ @ DQG UHSUHVHQW DQ LPSURYHPHQW ZLWK UHVSHFW WR OLPLW RI GHWHFWLRQ DQG SUDFWLFDELOLW\ ZKHQ FRPSDUHG WR

PAGE 90

FXUUHQW +3/&(' PHWKRGV IRU HQNHSKDOLQV 7DEOH f >)OHPLQJ DQG 5H\QROGV .LP HW DO 6KLEDQRNL HW DO @ 7KH LQFUHDVHG SUDFWLFDELOLW\ RI WKH DQDO\WLFDO DSSURDFK GHVFULEHG KHUH LV FKDUDFWHUL]HG E\ WKH PLQLPDO SUHFDXWLRQV UHTXLUHG LQ WKH SUHSDUDWLRQ RI WKH PRELOH SKDVH ILOWHULQJ DQG GHJDVVLQJ XQGHU QHJDWLYH SUHVVXUH ZLWK VRQLFDWLRQf DQG LV GXH WR WKH IDFW WKDW WKH HQ]\PDWLF GHULYDWL]DWLRQ HPSOR\HG DOORZHG WKH XVH RI ORZHU DSSOLHG SRWHQWLDOV IRU HOHFWURFKHPLFDO GHWHFWLRQ 9 FRPSDUHG WR 9 IRU H[LVWLQJ +3/&(' PHWKRGVf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f DQG DV D FRQVHTXHQFH EDFNJURXQG FXUUHQW EDVHOLQH QRLVH DQG EDVHOLQH GULIW ZHUH JUHDWO\ LQFUHDVHG 7KHUHIRUH WKH PD[LPXP JDLQ VHWWLQJ RI WKH LQVWUXPHQW FRXOG QRW EH XVHG VHQVLWLYLW\ ZDV FRPSURPLVHG E\ D IDFWRU RI DQG WKH OLQHDULW\ RI WKH FDOLEUDWLRQ FXUYHV ZDV UHGXFHG ,Q DGGLWLRQ WKH V\VWHP ZDV PRUH GLIILFXOW WR KDQGOH DV WKH PRELOH SKDVH KDG WR EH FRQWLQXDOO\ GHJDVVHG XQGHU D VWUHDP RI KHOLXP LQ RUGHU WR DYRLG XQDFFHSWDEOH EDVHOLQH GULIW

PAGE 91

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f 7R DYRLG LQWHUIHULQJ SHDNV H[LVWLQJ +3/& (' PHWKRGV IRU HQNHSKDOLQV LQYROYH FRPSOH[ DQG UHODWLYHO\ QRQVHOHFWLYH VDPSOH FOHDQ XS SURFHGXUHV HJ PXOWLSOH SUHFLSLWDWLRQ FHQWULIXJDWLRQ DQG DGVRUSWLRQ VWHSVf SULRU WR DSSOLFDWLRQ RI WKH VDPSOH WR WKH +3/& V\VWHP %\ FRQWUDVW KHUH WKH XVH RI D ERURQDWH FOHDQXS SURFHGXUH ZKLFK KDG EHHQ SUHYLRXVO\ HVWDEOLVKHG IRU XVH ZLWK FDWHFKRODPLQHV >+LJD HW DO .RLNH HW DO @ LQFUHDVHG WKH VHOHFWLYLW\ RI WKLV DVVD\ DV RQO\

PAGE 92

GHULYDWL]HG SHSWLGH LQFRUSRUDWLQJ WKH GLK\GUR[\ JURXS LQWURGXFHG E\ HQ]\PDWLF GHULYDWL]DWLRQ ZDV UHWDLQHG RQ WKH ERURQDWH JHO FROXPQ 7KH OLPLWV RI GHWHFWLRQ DFKLHYHG IRU /( XVLQJ WKLV DQDO\WLFDO DSSURDFK IPROLQM LQ EXIIHU DQG IPROLQM LQ &6)f DOVR FRPSDUHG IDYRUDEO\ WR DVVD\V IRU FDWHFKRODPLQHV XVLQJ D ERURQDWH FOHDQXS PHWKRG DQG HOHFWURFKHPLFDO GHWHFWLRQ m IPROLQMf DOWKRXJK WKH UHFRYHU\ RI K\GUR[\ODWHG /( IURP WKH ERURQDWH JHO bf ZDV FRQVLGHUDEO\ ORZHU WKDQ WKH UHFRYHU\ RI FDWHFKRODPLQHV IURP WKH VDPH PDWUL[ bf >+LJD HW DO .RLNH HW DO @ 7KH +3/&(' DSSURDFK IRU WKH GHWHUPLQDWLRQ RI /( LQ &6) GHVFULEHG KHUH ZDV IRXQG WR EH UHSURGXFLEOH DQG DFFXUDWH VLQFH WKH HODWLYH VWDQGDUG GHYLDWLRQV RI YDULRXV FRQFHQWUDWLRQV GHWHUPLQHG RQ GLIIHUHQW GD\V ZDV IRXQG WR EH b DQG IRXQG FRQFHQWUDWLRQV ZHUH GHWHUPLQHG WR EH ZLWKLQ b RI QRPLQDO FRQFHQWUDWLRQV 7DEOH f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
PAGE 93

VDPSOH POf LW PD\ EH DGHTXDWH IRU WKH DQDO\VLV RI WKH HOHYDWHG SK\VLRORJLFDO FRQFHQWUDWLRQV RI RSLRLG SHSWLGHV WR EH H[SHFWHG LQ FOLQLFDO VWXGLHV

PAGE 94

&+$37(5 $ +,*+ 3(5)250$1&( /,48,' &+520$72*5$3+< $66$< )25 23,2,' 3(37,'(6 86,1* )/825(6&(1&( '(7(&7,21 ,QWURGXFWLRQ 7KH REMHFWLYH RI WKLV VWXG\ ZDV WKH HYDOXDWLRQ RI D W\URVLQHVSHFLILF DQDO\WLFDO DSSURDFK IRU RSLRLG SHSWLGHV XVLQJ KLJK SHUIRUPDQFH OLTXLG FKURPDWRJUDSK\ ZLWK IOXRUHVFHQFH GHWHFWLRQ DQG OHXFLQH HQNHSKDOLQ /( )LJXUH f DV D PRGHO SHSWLGH 2+ 7\U *O\ *O\ 3KH /HX )LJXUH /HXFLQH HQNHSKDOLQ $V LQ &KDSWHU WKLV DQDO\WLFDO DSSURDFK H[SORLWV WKH GHULYDWL]DWLRQ RI /( E\ VSHFLILF HQ]\PDWLF RK\GUR[\ODWLRQ RI WKH KLJKO\ FRQVHUYHG W\URVLQH JURXSV LQ WKH

PAGE 95

SRVLWLRQ RI RSLRLG SHSWLGHV E\ PXVKURRP W\URVLQDVH +HUH WKH HQ]\PDWLF GHULYDWL]DWLRQ UHQGHUV WKH SHSWLGH VXLWDEOH IRU VXEVHTXHQW IOXRURJHQLF GHULYDWL]DWLRQ XVLQJ GLDPLQR GLSKHQ\OHWKDQH '3(f 7KHUHIRUH LQ WKLV DVVD\ WKH K\GUR[\ODWHG /( GHULYDWLYH REWDLQHG IURP WKH UHDFWLRQ ZLWK PXVKURRP W\URVLQDVH LV R[LGL]HG LQ D FRQWUROOHG PDQQHU E\ SRWDVVLXP IHUULF\DQLGH WR JLYH WKH FRUUHVSRQGLQJ TXLQRQH SULRU WR D FRQGHQVDWLRQ UHDFWLRQ ZLWK '3( WR JLYH D IOXRUHVFHQW SURGXFW )LJXUH f 7\URVLQDVH UHDFWLRQ 2+ 2+ 2 92 7 W\URVLQDVH 2+ 3RO\PHUL]DWLRQ '3( UHDFWLRQ 2[LGDWLRQ 3RWDVVLXP IHUULF\DQLGH +\GUR[\ODWHG /( &RUUHVSRQGLQJ TXLQRQH &+&+ , 1+M 1+M &RQGHQVDWLRQ n '3( )OXRUHVFHQW SURGXFW )LJXUH 2YHUYLHZ RI GHULYDWL]DWLRQ UHDFWLRQV IRU +3/&)/ DVVD\

PAGE 96

0DWHULDOV /HXFLQH HQNHSKDOLQ DQG PXVKURRP W\URVLQDVH ZHUH REWDLQHG IURP 6LJPD &KHPLFDO &RPSDQ\ 6W /RXLV 02 86$ $FHWRQLWULOH PHWKDQRO DQG WULIOXRURDFHWLF DFLG ZHUH RI +3/& JUDGH DQG GLVRGLXP K\GURJHQ SKRVSKDWH VRGLXP GLK\GURJHQ SKRVSKDWH FLWULF DFLG SRWDVVLXP FKORULGH DQG SRWDVVLXP IHUULF\DQLGH ZHUH RI UHDJHQW JUDGH 7KHVH FKHPLFDOV DV ZHOO DV 6FLQWLYHUVH ,, VFLQWLOODWLRQ FRFNWDLO ZHUH SURFXUHG IURP )LVKHU 6FLHQWLILF 3LWWVEXUJK 3$ 86$ 7HWUDEXW\ODPPRQLXP 7%$f ZDV SXUFKDVHG IURP WKH (DVWPDQ .RGDN &RPSDQ\ 5RFKHVWHU 1< 86$ DQG >W\URV\O+1f@OHXFLQH HQNHSKDOLQ + /(f ZDV REWDLQHG IURP 1(1 5HVHDUFK 3URGXFWV 'XSRQW &RPSDQ\ :LOPLQJWRQ '( 86$ 'LDPLQRGLSKHQ\OHWKDQH '3(f ZDV V\QWKHVL]HG DFFRUGLQJ WR ,UYLQJ DQG 3DUNLQV >,UYLQJ DQG 3DUNLQV @ %ULHIO\ EHQ]DOGHK\GH HTXLYDOHQWVf ZDV UHIOX[HG ZLWK DPPRQLXP DFHWDWH HTXLYDOHQWf IRU KRXUV WKH UHVXOWLQJ SUHFLSLWDWH ZDV FROOHFWHG DQG ZDVKHG ZLWK HWKDQRO 7KLV SUHFLSLWDWH ZDV WKHQ K\GURO\]HG ZLWK b YY +6 EHQ]RLF DFLG DQG EHQ]DOGHK\GH ZHUH UHPRYHG E\ VWHDP GLVWLOODWLRQ DQG '3( ZDV SUHFLSLWDWHG E\ QHXWUDOL]DWLRQ ZLWK DPPRQLXP K\GUR[LGH $IWHU UHFU\VWDOOL]DWLRQ IURP SHWUROHXP HWKHU WKH SURGXFW KDG D PHOWLQJ SRLQW RI r& OLWHUDWXUH r&f >,UYLQJ DQG 3DUNLQV @ DQG D f+ QXFOHDU PDJQHWLF UHVRQDQFH VSHFWUXP ZKLFK VXSSRUWHG WKH DVVLJQHG VWUXFWXUH $OO RWKHU FKHPLFDOV ZHUH RI UHDJHQW JUDGH 'RXEOH GLVWLOOHG ZDWHU ZDV XVHG WKURXJKRXW

PAGE 97

0HWKRGV 3XULILFDWLRQ RI 0XVKURRP 7\URVLQDVH 0XVKURRP W\URVLQDVH ZDV SXULILHG SULRU WR XVH E\ XOWUDILOWUDWLRQ XVLQJ &HQWULFRQ PHPEUDQH ILOWHUV PROHFXODU ZHLJKW FXW RII $PLFRQ 'DQYHUV 0$ 86$f 2QH PLOOLOLWHU RI D VROXWLRQ RI PXVKURRP W\URVLQDVH PJPOf LQ 0 SKRVSKDWH EXIIHU S+ ZDV DSSOLHG WR WKH ILOWHU XQLW DQG FHQWULIXJHG DW J XQWLO PD[LPXP FRQFHQWUDWLRQ RI WKH VDPSOH ZDV DFKLHYHG 7KLV FHQWULIXJDWLRQ VWHS ZDV UHSHDWHG WKUHH WLPHV ZLWK WKH DGGLWLRQ RI DQ DGGLWLRQDO PO RI SKRVSKDWH EXIIHU SULRU WR HDFK FHQWULIXJDWLRQ 7KH ILQDO FRQFHQWUDWH ZDV UHFRQVWLWXWHG LQ SKRVSKDWH EXIIHU WR JLYH D ILQDO FRQFHQWUDWLRQ RI PJPO RI PXVKURRP W\URVLQDVH FRUUHVSRQGLQJ WR XQLWV RI DFWLYLW\ SHU PO RI VROXWLRQ $OLTXRWV ZHUH VWRUHG DW r& DQG GHIURVWHG LPPHGLDWHO\ SULRU WR XVH (Q]\PDWLF 'HULYDWL]DWLRQ 7R FKDUDFWHUL]H WKH GHULYDWL]DWLRQ SURFHGXUH /( P0f LQ 0 SKRVSKDWH EXIIHU S+ ZDV UHDFWHG ZLWK PXVKURRP W\URVLQDVH XQLWVPOf LQ WKH SUHVHQFH RI DVFRUELF DFLG P0f DW URRP WHPSHUDWXUH ZLWK FRQVWDQW VKDNLQJ 7KH SURGXFW SHDNV REWDLQHG IURP WKLV HQ]\PDWLF GHULYDWL]DWLRQ ZHUH FROOHFWHG IURP WKH +3/& HOXHQW RUJDQLF VROYHQW ZDV UHPRYHG E\ HYDSRUDWLRQ XQGHU D VWUHDP RI QLWURJHQ DQG WKH FROOHFWHG SHDNV ZHUH FRQFHQWUDWHG XVLQJ D 6HS3DN &L SUHSDUDWLYH FROXPQ :DWHUV $VVRFLDWHV 0LOIRUG 0$ 86$f 7KH LGHQWLW\ RI WKH SURGXFWV ZDV GHWHUPLQHG E\ HOHFWURVSUD\ LRQL]DWLRQ PDVV VSHFWURPHWU\ VHH &KDSWHU f

PAGE 98

7R GHWHUPLQH WKH WLPH FRXUVH RI WKH UHDFWLRQ DW YDULRXV WLPH SRLQWV DOLTXRWV RI WKLV LQFXEDWLRQ VROXWLRQ ZHUH DSSOLHG WR DQ +3/& V\VWHP FRQVLVWLQJ RI DQ /'&0LOWRQ 5R\ PLQL0HWULF ,, PHWHULQJ SXPS 5LYLHUD %HDFK )/ 86$f D 1HJUHWWL DQG =DPED LQMHFWRU 6RXWKDPSWRQ 8.f ILWWHG ZLWK D SL ORRS D 3HUNLQ (OPHU /& VSHFWURSKRWRPHWULF GHWHFWRU 1RUZDON &7 86$f DQG D +HZOHWW 3DFNDUG +3 $ LQWHJUDWRU $YRQGDOH 3$ 86$f 7KH FROXPQ ZDV D 3DUWLVLO 2'6 [ PP :KDWPDQ /DEVDOHV +LOOVERUR 25 86$f 7KH GHWHFWLRQ ZDYHOHQJWK ZDV QP DQG WKH PRELOH SKDVH FRQVLVWHG RI b DFHWRQLWULOH YYf LQ FLWUDWHGLSRWDVVLXP SKRVSKDWH EXIIHU S+ f DW D IORZ UDWH RI POPLQXWH ([WUDFWLRQ RI /HXFLQH (QNHSKDOLQ IURP &HUHEURVSLQDO )OXLG /HXFLQH HQNHSKDOLQ ZDV H[WUDFWHG IURP KXPDQ FHUHEURVSLQDO IOXLG &6)f WKURXJK WKH XVH RI 6XSHOFOHDQ /&L VROLG SKDVH H[WUDFWLRQ FROXPQV 6XSHOFR ,QF %HOOHIRQWH 3$ 86$f 7KH FROXPQV ZHUH DFWLYDWHG ZLWK PO HDFK RI ZDWHU DQG PHWKDQRO DQG ORDGHG ZLWK SL RI VSLNHG &6) 6XEVHTXHQWO\ WKH FROXPQV ZHUH ZDVKHG ZLWK PO RI ZDWHU PO RI 1 +& PO RI ZDWHU PO RI 0 ERUDWH EXIIHU S+ f DQG PO RI ZDWHU 7KH /(ULFK IUDFWLRQ ZDV WKHQ HOXWHG LQ PO RI PHWKDQRO DQG HYDSRUDWHG WR GU\QHVV XQGHU D VWUHDP RI QLWURJHQ 7R WHVW WKH UHFRYHU\ RI /( IURP &6) XVLQJ WKLV H[WUDFWLRQ SURFHGXUH SL RI &6) VSLNHG ZLWK +/( FRXQWV SHU PLQXWH &30ff ZDV DSSOLHG WR DQ H[WUDFWLRQ FROXPQ DQG WKH SURFHGXUH GHVFULEHG DERYH ZDV FDUULHG RXW 7KH ILQDO PHWKDQROLF /(ULFK IUDFWLRQ ZDV FROOHFWHG LQ PO DOLTXRWV PO RI 6FLQWLYHUVH ,, VFLQWLOODWLRQ FRFNWDLO ZHUH

PAGE 99

DGGHG DQG WKH UDGLRDFWLYLW\ &30f ZDV PHDVXUHG XVLQJ D %HFNPDQ /6 7' VFLQWLOODWLRQ FRXQWHU )XOOHUWRQ &$ 86$f $V D FRQWURO WKH UDGLRDFWLYLW\ LQ D VDPSOH FRQWDLQLQJ WKH VDPH DPRXQW RI +/( DV WKH VSLNHG &6) LQ PO RI PHWKDQRO ZDV DOVR PHDVXUHG &DOLEUDWLRQ FXUYHV &DOLEUDWLRQ FXUYHV IRU /( SPROLQMf LQ ERWK EXIIHU DQG &6) ZHUH FRQVWUXFWHG XVLQJ WKLV +3/&)/ DQDO\WLFDO DSSURDFK )RU WKH W\URVLQDVH UHDFWLRQ YDULRXV FRQFHQWUDWLRQV RI /( LQ 0 VRGLXP SKRVSKDWH EXIIHU S+ RU VSLNHG &6) H[WUDFWV UHFRQVWLWXWHG LQ 0 VRGLXP SKRVSKDWH EXIIHU S+ ZHUH UHDFWHG ZLWK PXVKURRP W\URVLQDVH XQLWVPOf LQ WKH SUHVHQFH RI DVFRUELF DFLG P0f 7KH WRWDO YROXPH RI WKLV LQFXEDWLRQ PL[WXUH ZDV SL $IWHU D PLQXWH LQFXEDWLRQ DW URRP WHPSHUDWXUH ZLWK FRQVWDQW VKDNLQJ SL RI DQ R[LGL]LQJ VROXWLRQ FRQWDLQLQJ PJPO SRWDVVLXP IHUULF\DQLGH PJPO SRWDVVLXP FKORULGH DQG b DFHWRQLWULOH YYf ZDV DGGHG IROORZHG E\ SL RI D VROXWLRQ RI '3( FRQWDLQLQJ PJPO LQ 1 +& 7KH IOXRURJHQLF UHDFWLRQ ZDV DOORZHG WR SURFHHG LQ WKH GDUN IRU PLQXWHV DW URRP WHPSHUDWXUH ZLWK FRQVWDQW VKDNLQJ $ SL DOLTXRW RI WKLV UHDFWLRQ PL[WXUH UHSUHVHQWLQJ b RI WKH WRWDO ILQDO UHDFWLRQ PL[WXUH ZDV WKHQ LQMHFWHG LQWR DQ )/3/& V\VWHP FRQVLVWLQJ RI DQ /'&0LOWRQ 5R\ PLQL0HWULF ,, PHWHULQJ SXPS 5LYLHUD %HDFK )/ 86$f D 5KHRG\QH 0RGHO LQMHFWRU &RWDWL &$ 86$f ILWWHG ZLWK D SL ORRS D1XFOHRVLO &LJ FROXPQ SP [ PP .H\VWRQH 6FLHQWLILF %HOOHIRQWH 3$ 86$f DQG D 3HUNLQ (OPHU 6 VSHFWURIOXRURGHWHFWRU 1RUZDON &7 86$f VHW DW $H[

PAGE 100

QP DQG ;HP QP ZLWK VOLW ZLGWKV RI QP $ %' FKDUW UHFRUGHU .LSS DQG =RQHQ 'HOIW 7KH 1HWKHUODQGVf DQG D +HZOHWW 3DFNDUG +3 $ LQWHJUDWRU $YRQGDOH 3$ 86$f UHFRUGHG WKH RXWSXW IURP WKH VSHFWURIOXRURGHWHFWRU 7KH PRELOH SKDVH FRQWDLQHG b DFHWRQLWULOH YYf DQG P0 7%$ LQ FLWUDWHGLSRWDVVLXP SKRVSKDWH EXIIHU S+ f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b RI WKH WRWDO ILQDO UHDFWLRQ PL[WXUH ZDV LQMHFWHG LQWR WKH +3/& V\VWHP GHVFULEHG DERYH 6LQFH H[SHULPHQWV FRQGXFWHG LQ SDUDOOHO LQGLFDWHG WKDW DW ORZ FRQFHQWUDWLRQV RI DQDO\WH WKH KLJKHVW \LHOG RI HQ]\PDWLFDOO\ GHULYHG VSHFLHV ZDV REWDLQHG DIWHU D PLQXWH W\URVLQDVH LQFXEDWLRQ WLPH VHH &KDSWHU f DQ DWWHPSW ZDV DOVR PDGH WR RSWLPL]H WKH GHULYDWL]DWLRQ UHDFWLRQ E\ UHGXFLQJ WKH W\URVLQDVH LQFXEDWLRQ WLPH WR PLQXWHV 5HVXOWV (Q]\PDWLF 'HULYDWL]DWLRQ 3UHOLPLQDU\ H[SHULPHQWV LQGLFDWHG WKDW WKH FRQFHQWUDWLRQ RI PXVKURRP W\URVLQDVH XQLWVPOf XVHG IRU WKH HQ]\PDWLF GHULYDWL]DWLRQ RI /( DOORZHG WKH HIILFLHQW FRQYHUVLRQ

PAGE 101

RI /( WR LWV K\GUR[\ODWHG GHULYDWLYH ZLWKLQ PLQXWHV $ P0 FRQFHQWUDWLRQ RI DVFRUELF DFLG ZDV IRXQG WR EH DGHTXDWH WR SUHYHQW RU UHYHUVH WKH W\URVLQDVHLQGXFHG IRUPDWLRQ RI RTXLQRQHV LQ RXU HQ]\PDWLF UHDFWLRQ PL[WXUH WKHUHE\ LQKLELWLQJ WKH VXEVHTXHQW SRO\PHUL]DWLRQ RI WKH UHDFWLRQ SURGXFWV )LJXUH &KURPDWRJUDSKV RI D FRQWURO UXQ $f VKRZLQJ /( HOXWLQJ DIWHU PLQXWHV GLVWLQFW IURP DVFRUELF DFLG HOXWLQJ LQ WKH VROYHQW IURQW DIWHU PLQXWHV DQG LQFXEDWLRQ VROXWLRQ %f VKRZLQJ WKH HPHUJHQFH RI WZR QHZ SHDNV ZLWK UHWHQWLRQ WLPHV RI DQG PLQXWHV

PAGE 102

)LJXUH $ VKRZV WKH FKURPDWRJUDSK RI D FRQWURO UXQ ZKHUH RQO\ /( DQG DVFRUELF DFLG DUH SUHVHQW DW WKH LQFXEDWLRQ FRQFHQWUDWLRQV +HUH /( LV VHHQ HOXWLQJ DIWHU PLQXWHV GLVWLQFW IURP DVFRUELF DFLG HOXWLQJ LQ WKH VROYHQW IURQW DIWHU PLQXWHV 7ZR DGGLWLRQDO SHDNV DW PLQXWHV DQG PLQXWHV DUH VHHQ LQ WKH FKURPDWRJUDSK RI WKH LQFXEDWLRQ VROXWLRQ )LJXUH %f 7KHVH WZR SURGXFW SHDNV ZHUH LGHQWLILHG E\ HOHFWURVSUD\ LRQL]DWLRQ PDVV VSHFWURVFRS\ VHH &KDSWHU f DV WKH GL DQG PRQR K\GUR[\ODWHG GHULYDWLYHV RI /( >+2f 7\Un@/( DQG >+27\Un@/( UHVSHFWLYHO\f )LJXUH 7LPH FRXUVH IRU WKH HQ]\PDWLF GHULYDWL]DWLRQ RI /( P0f E\ PXVKURRP W\URVLQDVH XQLWVPOf LQ WKH SUHVHQFH RI DVFRUELF DFLG P0f VKRZLQJ WKH GLVDSSHDUDQFH RI /( ff DV >+27\Un@/( $f DQG +2f7\U f DUH SURGXFHG $ WLPH FRXUVH RI WKH HQ]\PDWLF GHULYDWL]DWLRQ RI /( E\ PXVKURRP W\URVLQDVH LQ WKH SUHVHQFH RI DVFRUELF DFLG LV VKRZQ LQ )LJXUH 7KLV WLPH FRXUVH VKRZV WKH GLVDSSHDUDQFH RI WKH SHDN FRUUHVSRQGLQJ WR /( DV >+27\UA/( DQG >+2f 7\Uf@/(

PAGE 103

DUH SURGXFHG $Q LQFXEDWLRQ WLPH RI PLQXWHV ZDV FKRVHQ IRU WKH HQ]\PDWLF GHULYDWL]DWLRQ LQ RXU +3/&)/ DVVD\ DV DW WKLV WLPH SRLQW WKH SHDN FRUUHVSRQGLQJ WR /( KDG GLVDSSHDUHG DQG WKH SHDN FRUUHVSRQGLQJ WR RXU HQ]\PDWLFDOO\ GHULYHG VSHFLHV RI LQWHUHVW >+27\Uf@/(f ZDV VHHQ WR SODWHDX 7KLV WLPH FRXUVH ZDV UHSHDWHG VHYHUDO WLPHV DQG WKH VDPH WUHQG ZDV REVHUYHG HDFK WLPH 5HFRYHU\ RI /HXFLQH (QNHSKDOLQ IURP &HUHEURVSLQDO )OXLG 5HFRYHU\ RI /( IURP &6) ZDV HVWLPDWHG DW !b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

PAGE 104

7DEOH VKRZV D VXPPDU\ RI WKH UHVXOWV REWDLQHG XVLQJ WKH YDULRXV PRGLILFDWLRQV DWWHPSWHG LQ RXU +3/&)/ DVVD\ 6RPH RI WKH UHVXOWV XVLQJ WKH VDPH PHWKRG DQG WKH VDPH PDWUL[ FRXOG QRW EH DYHUDJHG VLQFH WKH RXWSXW IURP WKH IOXRUHVFHQFH GHWHFWRU ZDV PRQLWRUHG XVLQJ GLIIHUHQW HTXLSPHQW LQWHJUDWRU RU FKDUW UHFRUGHUf 7DEOH 6XPPDU\ RI UHVXOWV IRU +3/&)/ DVVD\ )RU UDZ GDWD VHH $SSHQGL[ % 0HWKRG 0DWUL[ /LPLW RI GHWHFWLRQ 1 VORSH 1RQFRQFHQWUDWHG UHDJHQWV EXIIHU IPROLQM r DUHDPRO 1RQFRQFHQWUDWHG UHDJHQWV EXIIHU IPROLQM r PPPRO 1RQFRQFHQWUDWHG UHDJHQWV &6) IPROLQM r DUHDPRO 1RQFRQFHQWUDWHG UHDJHQWV &6) IPROLQM r PPPRO &RQFHQWUDWHG UHDJHQWV %XIIHU IPROLQM r DUHDPRO &RQFHQWUDWHG UHDJHQWV &6) IPROLQM s r DUHDPRO &RQFHQWUDWHG UHDJHQWV PLQ W\URVLQDVH %XIIHU IPROLQM r PPPRO )LJXUHV DQG VKRZ UHSUHVHQWDWLYH FDOLEUDWLRQ FXUYHV REWDLQHG IRU /( LQ EXIIHU VDPSOHV DQG VSLNHG &6) UHVSHFWLYHO\ XVLQJ WKH +3/&)/ PHWKRG ZLWK QRQn FRQFHQWUDWHG UHDJHQWV /LPLWV RI GHWHFWLRQ IRU /( RI IPROLQMHFWLRQ FRXOG EH REWDLQHG

PAGE 105

LQ ERWK EXIIHU VDPSOHV DQG VSLNHG &6) VDPSOHV 7KLV OLPLW RI GHWHFWLRQ FRUUHVSRQGV WR SPRO RI /( SHU PO RI &6) )LJXUH VKRZV D UHSUHVHQWDWLYH FKURPDWRJUDSK IRU /( LQ &6) XVLQJ RXU +3/&)/ PHWKRG )LJXUH &DOLEUDWLRQ FXUYH IRU /( LQ EXIIHU XVLQJ +3/&)/ PHWKRG ZLWK QRQn FRQFHQWUDWHG UHDJHQWV )LJXUH &DOLEUDWLRQ FXUYH IRU /( LQ &6) XVLQJ +3/&)/ PHWKRG DQG QRQn FRQFHQWUDWHG UHDJHQWV

PAGE 106

)LJXUH &KURPDWRJUDSK RI /( LQ &6) XVLQJ +3/&)/ PHWKRG /( LV VKRZQ HOXWLQJ DIWHU PLQXWHV $Q DQDO\WLFDO DGYDQWDJH ZDV QRW JDLQHG WKURXJK WKH FRQFHQWUDWLRQ RI WKH IOXRURJHQLF UHDJHQWV DQG VXEVHTXHQW LQMHFWLRQ RI D JUHDWHU SRUWLRQ RI WKH WRWDO ILQDO UHDFWLRQ PL[WXUH DV WKH VDPH OLPLW RI GHWHFWLRQ RI IPROLQMHFWLRQ FRXOG EH REWDLQHG ZKHQ WKLV PRGLILFDWLRQ LQ RXU PHWKRG ZDV DWWHPSWHG $OWKRXJK D ORZHU OLPLW RI GHWHFWLRQ LQ EXIIHU VDPSOHV IPROLQMHFWLRQf ZDV REWDLQHG ZKHQ D PLQXWH UHDFWLRQ WLPH LQVWHDG RI D PLQXWH UHDFWLRQ WLPH ZDV XVHG IRU WKH W\URVLQDVH UHDFWLRQ )LJXUH f ZKHQ WKLV PRGLILFDWLRQ RI WKH PHWKRG ZDV DWWHPSWHG LQ VSLNHG &6) VDPSOHV WKH OLQHDULW\ RI WKH FDOLEUDWLRQ FXUYH FRXOG QRW EH SUHVHUYHG GDWD

PAGE 107

QRW VKRZQf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

PAGE 108

LPSURYHPHQW LQ OLPLW RI GHWHFWLRQ VHHQ ZLWK WKH DSSURDFK GHVFULEHG KHUH PD\ EH DWWULEXWHG LQ SDUW WR WKH IDFW WKDW LQ VRPH RI WKHVH H[LVWLQJ W\URVLQHVSHFLILF +3/&)/ PHWKRGV >,VKLGD HW DO .DL HW DO @ IOXRURJHQLF GHULYDWL]DWLRQ QHFHVVLWDWHV KHDWLQJ RI WKH UHDFWLRQ PL[WXUH WR r& IRU PLQXWHV ZKLFK PD\ UHVXOW LQ SHSWLGH LQVWDELOLW\ DQG FRQVHTXHQW UHGXFHG UHFRYHU\ RI LQWDFW IOXRUHVFHQFHODEHOHG DQDO\WH $OWKRXJK PRUH VHOHFWLYH WKH +3/&)/ DSSURDFK GHVFULEHG KHUH GRHV QRW DFKLHYH OLPLWV RI GHWHFWLRQ IRU /( DV ORZ DV WKRVH REWDLQHG E\ LQYHVWLJDWRUV XVLQJ PHWKRGV LQYROYLQJ SRVWFROXPQ IOXRUHVFHQFH GHULYDWL]DWLRQ RI WKH 1WHUPLQDO SULPDU\ DPLQR JURXS RI RSLRLG SHSWLGHV IPROLQMHFWLRQf >'DYH HW DO 0LIXQH HW DO YDQ GHQ %HOG HW DO @ +RZHYHU GXH WR ODFN RI VHOHFWLYLW\ WKHVH PHWKRGV RIWHQ LQYROYH WKH XVH RI H[WHQVLYH VDPSOH FOHDQXS SURFHGXUHV DQG FRPSOH[ PXOWLGLPHQVLRQDO FKURPDWRJUDSKLF V\VWHPV FROXPQ VZLWFKLQJf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

PAGE 109

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

PAGE 110

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
PAGE 111

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f 3UHYLRXVO\ RXU JURXS DQG RWKHUV KDYH VKRZQ WKDW W\URVLQDVH DOVR UHDFWV ZLWK RWKHU W\URVLQHFRQWDLQLQJ SHSWLGHV WR JLYH K\GUR[\ODWHG SURGXFWV >0DUXPR DQG :DLWH 5RVHL HW DO 5RVHL HW DO 7HOOLHU HW DO @ 1XPHURXV VWXGLHV KDYH UHYHDOHG D ORVV RI DFWLYLW\ LQ HQNHSKDOLQV DQG RWKHU RSLRLG SHSWLGHV ZKHQ WKH 7\U PRLHW\ LV DEVHQW IRU UHYLHZV VHH +DQVHQ DQG 0RUJDQ DQG 6KLPRKLJDVKL f KRZHYHU IHZ KDYH VKRZQ WKH HIIHFW RI PRGLILFDWLRQ RI WKH DURPDWLF VLGHFKDLQ RI WKLV UHVLGXH 7KHUHIRUH DV W\URVLQDVH LV NQRZQ WR UHDFW ZLWK W\URVLQH FRQWDLQLQJ SHSWLGHV RXU JRDO LQ WKLV VWXG\ ZDV WR GHWHUPLQH WKH VWUXFWXUH RI WKH SURGXFWV IRUPHG ZKHQ W\URVLQDVH UHDFWV ZLWK /( DQG WR GHILQH WKH DIILQLW\ RI WKH SURGXFWV PRGLILHG DW WKH 7\U UHVLGXH WR RSLRLG UHFHSWRUV LQ UDW EUDLQ KRPRJHQDWH

PAGE 112

0DWHULDOV 7KH IROORZLQJ PDWHULDOV ZHUH SURFXUHG IURP WKH VRXUFHV LQGLFDWHG /HXFLQH HQNHSKDOLQ PXVKURRP W\URVLQDVH XQLWVPJ (& f EHVWDWLQ WKLRUSKDQ DQG FDSWRSULO IURP 6LJPD 6W /RXLV 02 86$ DVFRUELF DFLG IURP 0DOOLQFNURGW 3DULV .< 86$ >+@GLSUHQRUSKLQH IURP $PHUVKDP ,QWHUQDWLRQDO $UOLQJWRQ +HLJKWV ,/ 86$ DQG 6FLQWLYHUVH ,, VFLQWLOODWLRQ FRFNWDLO IURP )LVKHU 6FLHQWLILF 3LWWVEXUJK 3$ 86$ $OO RWKHU FKHPLFDOV ZHUH RI UHDJHQW JUDGH 'RXEOH GLVWLOOHG ZDWHU ZDV XVHG WKURXJKRXW 7KH FKURPDWRJUDSKLF V\VWHP XVHG FRQVLVWHG RI DQ /'&0LOWRQ 5R\ PLQL0HWULF ,, PHWHULQJ SXPS 5LYLHUD %HDFK )/ 86$f D 1HJUHWWL DQG =DPED LQMHFWRU 6RXWKDPSWRQ 8.f ILWWHG ZLWK D SL ORRS D 3HUNLQ (OPHU /& VSHFWURSKRWRPHWULF GHWHFWRU 1RUZDON &7 86$f DQG D +HZOHWW 3DFNDUG +3 $ LQWHJUDWRU $YRQGDOH 3$ 86$f 7KH FROXPQ ZDV D 3DUWLVLO 2'6 [ PP :KDWPDQ /DEVDOHV +LOOVERUR 25 86$f 7KH GHWHFWLRQ ZDYHOHQJWK ZDV QP DQG WKH PRELOH SKDVH FRQVLVWHG RI b DFHWRQLWULOH YYf LQ FLWUDWHGLSRWDVVLXP SKRVSKDWH EXIIHU S+ f DW D IORZ UDWH RI POPLQXWH 0HWKRGV (Q]\PDWLF 5HDFWLRQ 7KH UHDFWLRQ EHWZHHQ /( DQG PXVKURRP W\URVLQDVH ZDV FDUULHG RXW LQ PLFURFHQWULILLJH WXEHV DW WKH IROORZLQJ FRQFHQWUDWLRQV P0 /( DQG P0 DVFRUELF DFLG LQ 0 SRWDVVLXP SKRVSKDWH EXIIHU S+ f 7KH UHDFWLRQ ZDV VWDUWHG E\ DGGLQJ

PAGE 113

PXVKURRP W\URVLQDVH SUHYLRXVO\ SXULILHG DV GHVFULEHG LQ &KDSWHUV DQG f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f ZDV DFWLYDWHG ZLWK PO RI PHWKDQRO DQG ZDVKHG ZLWK PO HDFK RI PHWKDQROb DFHWLF DFLG YY DQG b DFHWLF DFLG YY 7KH FRQFHQWUDWHG PDWHULDOV ZHUH DSSOLHG WR WKH FDUWULGJH DQG WKH FDUWULGJH ZDV ZDVKHG ZLWK PO RI b DFHWLF DFLG YY 7KH DQDO\WHV ZHUH WKHQ HOXWHG LQ SL RI PHWKDQROb DFHWLF DFLG YY DQG VWRUHG DW r& SULRU WR PDVV VSHFWURPHWULF DQDO\VLV $ 9HVWHF (6 LQVWUXPHQW 9HVWHF &RUS +RXVWRQ 7; 86$f ZDV XVHG WR REWDLQ WKH HOHFWURVSUD\ LRQL]DWLRQ (6,f PDVV VSHFWUD >$OOHQ DQG 9HVWDO @ 7KH VDPSOH VROXWLRQ ZDV GUDZQ LQWR D VWDQGDUG ODERUDWRU\ V\ULQJH SL 0RGHO +DPLOWRQ &R 5HQR 1HYDGD 86$f DQG VXSSOLHG LQWR WKH HOHFWURVSUD\ SUREH WKURXJK D FP [ PP

PAGE 114

L G GHDFWLYDWHG IXVHG VLOLFD FDSLOODU\ 6FLHQWLILF *ODVV (QJLQHHULQJ 9LFWRULD $XVWUDOLDf DW WR SOPLQ IORZ UDWH E\ D PHGLFDO LQIXVLRQ SXPS 6$*( ,QVWUXPHQWV 0RGHO ,% %RVWRQ 0$ 86$f $ LQ L G [ LQ RG IODW WLSSHG K\SRGHUPLF QHHGOH KHOG DW N9 SRWHQWLDO SURGXFHG VSUD\ FXUUHQW LQ WKH UDQJH RI WR S$ ZKHQ WKH WLS RI WKH QHHGOH WR QR]]OH RULILFH GLVWDQFH ZDV DERXW PP 7KH VRXUFH EORFN ZDV KHDWHG WR r& DQG WKH VSUD\ FKDPEHU WHPSHUDWXUH ZDV DURXQG WR r& $ 9HFWRU2QH GDWD V\VWHP 7HNQLYHQW 6W /RXLV 02 86$f ZDV XVHG WR FRQWURO WKH TXDGUXSROH DQDO\]HU 'D PDVV UDQJHf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n-/( ZDV FROOHFWHG LQ PO RI PRELOH SKDVH $ b FRQYHUVLRQ IURP /( WR >+27\UnM/( ZDV DVVXPHG EDVHG RQ WKH GLVDSSHDUDQFH RI WKH SHDN FRUUHVSRQGLQJ WR /( DIWHU D KRXU LQFXEDWLRQ ZLWK PXVKURRP W\URVLQDVH )LJXUH f DQG WKH UHODWLYH LQWHQVLWLHV RI WKH SURGXFW SHDNV REWDLQHG IURP PDVV VSHFWURPHWULF DQDO\VLV 7KHUHIRUH WKLV VROXWLRQ FRQWDLQHG QPROHV RI >+27\UA/( DQG ZDV XVHG GLUHFWO\ LQ WKH UDGLRUHFHSWRU DVVD\V 7ZR KXQGUHG PLFUROLWHUV RI D P0 VROXWLRQ RI /( ZHUH DOVR LQMHFWHG LQWR WKH +3/&

PAGE 115

V\VWHP DQG WKH SHDN FRUUHVSRQGLQJ WR /( ZDV FROOHFWHG DQG XVHG LQ WKH UDGLRUHFHSWRU DVVD\V 0RELOH SKDVH FRPSRQHQWV ZHUH IRXQG QRW WR LQWHUIHUH ZLWK WKH UDGLRUHFHSWRU DVVD\ )LJXUH 7LPH FRXUVH IRU WKH HQ]\PDWLF GHULYDWL]DWLRQ RI /( P0f E\ PXVKURRP W\URVLQDVH XQLWVPOf LQ WKH SUHVHQFH RI DVFRUELF DFLG P0f VKRZLQJ WKH GLVDSSHDUDQFH RI /( ff DV >+27\Un@/( $f DQG +2A7\U f DUH SURGXFHG 5DW EUDLQ PHPEUDQHV ZHUH SUHSDUHG DV GHVFULEHG E\ +RFKKDXV HW DO >+RFKKDXV HW DO @ %ULHIO\ WKH ZKROH EUDLQ ZLWKRXW FHUHEHOOXP RI PDOH 6SUDJXH'DZOH\ UDWV Jf ZDV KRPRJHQL]HG LQ YROXPHV RI LFH FROG P0 7ULV+&O EXIIHU S+ f FRQWDLQLQJ P0 1D&O 7KH KRPRJHQDWH ZDV LQFXEDWHG IRU KRXU DW r& DQG FHQWULIXJHG IRU PLQXWHV DW r& 7KH SHOOHW ZDV WKHQ UHVXVSHQGHG ZDVKHG WZLFH ZLWK P0 7ULV+&O DQG GLOXWHG LQ P0 7ULV+&O WR JLYH PJ RI UDW EUDLQ PHPEUDQHV SHU PO $OLTXRWV RI WKLV VXVSHQVLRQ ZHUH VWRUHG DW r& DQG XVHG ZLWKLQ RQH ZHHN

PAGE 116

7KH UDGLRUHFHSWRU DVVD\V ZHUH FDUULHG RXW LQ GXSOLFDWH LQ PLFURFHQWULIXJH WXEHV (DFK WXEH FRQWDLQHG S0 EHVWDWLQ S0 WKLRUSKDQ DQG S0 FDSWRSULO DV D SHSWLGDVH LQKLELWRU FRFNWDLO PJ RU PJ RI UDW EUDLQ PHPEUDQHV YDULRXV FRQFHQWUDWLRQV RI /( RU >+27\UnM/( DV FRPSHWLWRU DQG HLWKHU Q0 RI >+@ GLSUHQRUSKLQH Q0 RI >+@'$*2 RU Q0 >+@'3'3( DV WUDFHU WR DVVD\ IRU WRWDO RSLRLG UHFHSWRU S RU VLWHV UHVSHFWLYHO\ LQ PO RI P0 7ULV+&O EXIIHU S+ f 7KH WXEHV ZHUH LQFXEDWHG DW URRP WHPSHUDWXUH ZLWK FRQVWDQW VKDNLQJ IRU KRXU :KHQ XVLQJ >+@GLSUHQRUSKLQH DV WUDFHU WKH WXEHV ZHUH WKHQ FHQWULIXJHG DW J IRU PLQXWHV WR EULQJ GRZQ WKH SHOOHW 7KH SHOOHWV ZHUH ZDVKHG WKUHH WLPHV ZLWK LFHFROG P0 7ULV+&O EXIIHU ZLWKRXW UHVXVSHQGLQJf DQG WKHQ ZHUH GLVVROYHG LQ PO RI 6FLQWLYHUVH ,, VFLQWLOODWLRQ FRFNWDLO :KHQ XVLQJ >+@'$*2 RU >+@'3'3( DV WUDFHU WKH UDW EUDLQ PHPEUDQHV ZHUH VHSDUDWHG IURP WKH VXSHUQDWDQW DQG ZDVKHG ZLWK LFHFROG P0 7ULV+&O EXIIHU E\ PHDQV RI D UDSLG ILOWUDWLRQ WHFKQLTXH 7KH ILOWHUV UHWDLQLQJ WKH UDW EUDLQ PHPEUDQHV DQG ERXQG WUDFHU ZHUH SODFHG LQ PO RI 6FLQWLYHUVH ,, VFLQWLOODWLRQ FRFNWDLO DQG DIWHU EHLQJ DOORZHG WR VWDQG RYHUQLJKW WKH UDGLRDFWLYLW\ &30f LQ HLWKHU WKH SHOOHWV RU RQ WKH ILOWHUV ZDV GHWHUPLQHG XVLQJ D %HFNPDQQ /6 7' VFLQWLOODWLRQ FRXQWHU )XOOHUWRQ &$ 86$f ,Q D FRQWURO H[SHULPHQW +3/& DQDO\VLV RI WKH VXSHUQDWDQW VKRZHG WKDW ERWK /( DQG >+27\UA/( ZHUH VWDEOH XQGHU UDGLRUHFHSWRU DVVD\ FRQGLWLRQV 1RQVSHFLILF ELQGLQJ ZDV GHWHUPLQHG LQ WKH SUHVHQFH RI KLJK FRQFHQWUDWLRQV RI FRPSHWLWRU [ n 0f 7KH ,&R YDOXHV RI /( DQG >+27\Un@/( FRQFHQWUDWLRQ RI FRPSHWLWRU GLVSODFLQJ b RI ERXQG WUDFHUf ZHUH GHWHUPLQHG XVLQJ WKH 0,164 QRQOLQHDU FXUYHILWWLQJ SURJUDP

PAGE 117

QR 0LFUR0DWK 6FLHQWLILF VRIWZDUH 6DOW /DNH &LW\ 87 86$f 7KH GDWD ZHUH ILWWHG WR WKH IROORZLQJ PRGHO % 7 Mn MfF &1,& 16 :KHUH % &30 LQ WKH SUHVHQFH RI FRPSHWLWRU 7 &30 LQ WKH DEVHQFH RI FRPSHWLWRU & FRPSHWLWRU FRQFHQWUDWLRQ 1 VORSH IDFWRU 16 &30 XQGHU QRQVSHFLILF ELQGLQJ FRQGLWLRQV 5HVXOWV &KURPDWRJUDSK\ $V LQ &KDSWHUV DQG OHXFLQH HQNHSKDOLQ DQG WKH SURGXFWV RI WKH HQ]\PH UHDFWLRQ ZHUH VHSDUDWHG VXFFHVVIXOO\ IURP HDFK RWKHU XVLQJ WKH FKURPDWRJUDSKLF V\VWHP GHVFULEHG DERYH )LJXUH f ,Q D FRQWURO LQFXEDWLRQ ZKHUH /( DQG DVFRUELF DFLG ZHUH LQMHFWHG LQWR WKH V\VWHP LQ FRQFHQWUDWLRQV HTXDO WR WKH LQLWLDO FRQFHQWUDWLRQV RI WKHVH UHDJHQWV LQ WKH LQFXEDWLRQ PL[WXUH /( KDG D UHWHQWLRQ WLPH RI PLQXWHV DQG ZDV ZHOO VHSDUDWHG IURP DVFRUELF DFLG ZKLFK HOXWHG LQ D ODUJH VROYHQW IURQW DW PLQXWHV )LJXUH $f :KHQ WKH WHVW LQFXEDWLRQ ZDV DSSOLHG WR WKH V\VWHP WZR GLVWLQFW DGGLWLRQDO SHDNV ZHUH VHHQ RQ WKH FKURPDWRJUDP ZLWK UHWHQWLRQ WLPHV RI PLQXWHV DQG PLQXWHV UHVSHFWLYHO\ )LJXUH %f 7KHVH SHDNV ZHUH FROOHFWHG DQG VXEMHFWHG WR PDVV VSHFWURPHWU\ DV GHVFULEHG HDUOLHU $Q LQFXEDWLRQ WLPH RI RQH KRXU ZDV FKRVHQ IRU RXU H[SHULPHQWV DV DW WKLV WLPH SRLQW PRVW RI WKH /( SUHVHQW ZDV VHHQ WR KDYH EHHQ FRQVXPHG LQ WKH UHDFWLRQ DQG WKH OHYHOV RI WKH PDMRU SURGXFW ZHUH VHHQ WR SODWHDX )LJXUH f

PAGE 118

,OO $VFRUELF DFLG ZDV LQFOXGHG LQ WKH UHDFWLRQ PL[WXUH WR SUHYHQW IXUWKHU R[LGDWLRQ RI RXU K\GUR[\ODWHG SURGXFWV WR WKH FRUUHVSRQGLQJ TXLRQHV $ % )LJXUH &KURPDWRJUDSKV RI FRQWURO LQFXEDWLRQ $f VKRZLQJ /( HOXWLQJ DIWHU PLQXWHV GLVWLQFW IRUP DVFRUELF DFLG HOXWLQJ LQ WKH VROYHQW IURQW DIWHU PLQXWHV DQG D WHVW LQFXEDWLRQ DIWHU PLQXWHV %f VKRZLQJ WKH HPHUJHQFH RI WZR QHZ SHDNV ZLWK UHWHQWLRQ WLPHV RI DQG PLQXWHV

PAGE 119

0DVV VSHFWURPHWU\ (6, LV D VRIW LRQL]DWLRQ WHFKQLTXH WKDW SURYLGHV LQWDFW PROHFXODU LRQV IURP SHSWLGH VROXWLRQV >:KLWHKRXVH HW DO @ +RZHYHU VHOHFWLYH IUDJPHQWDWLRQ FDQ DOVR EH HIIHFWHG E\ FROOLVLRQLQGXFHG GLVVRFLDWLRQ &,'f LQ FHUWDLQ UHJLRQV RI WKH LRQ VRXUFH >$OOHQ DQG 9HVWDO .DWWD HW DO 6PLWK HW DO @ 7KH QRPHQFODWXUH VFKHPH DV SURSRVHG E\ 5RHSVWRII DQG )RKOPDQ >5RHSVWRUII DQG )RKOPDQ @ XVHG WR ODEHO WKH VHTXHQFH LRQV LQ WKH PDVV VSHFWUD RI SHQWDSHSWLGHV LV VKRZQ LQ )LJXUH )LJXUH 1RPHQFODWXUH VFKHPH IRU WKH ODEHOLQJ RI VHTXHQFH LRQV LQ WKH PDVV VSHFWUD RI SHQWDSHSWLGHV 7KH (6, PDVV VSHFWUD REWDLQHG IURP /( DUH VKRZQ LQ )LJXUH 8VLQJ VRIW LRQL]DWLRQ UHSHOOHU DW 9 )LJXUH $f RQO\ PROHFXODU LRQV >0+@ P] DQG >01D@ P] f ZHUH REVHUYHG $OWKRXJK WKH IRUPDWLRQ RI SURWRQDWHG PROHFXOHV WKURXJK DFLGEDVH HTXLOLEULD LQ VROXWLRQ LV SUHIHUUHG WKH JHQHUDWLRQ RI VRGLDWHG VSHFLHV ZDV XQDYRLGDEOH XSRQ XVH RI FRPPHUFLDOO\ DYDLODEOH +3/&JUDGH VROYHQWV >/HKPDQ @ )URP WKH VSHFWUXP REWDLQHG XQGHU &,' FRQGLWLRQV UHSHOOHU DW 9 )LJXUH %f VHYHUDO LPSRUWDQW VHTXHQFH LRQV P] DQG IRU DL E E D` DQG E

PAGE 120

UHVSHFWLYHO\f GLUHFWO\ GHULYHG IURP )LJXUH ZHUH DVVLJQHG 7ZR LQWHQVH LQWHUQDO IUDJPHQWV >Dr\@L DW P] DQG >Dr\@ DW P] ZHUH DOVR REVHUYHG 7KHVH ZHUH WKH SURGXFWV RI WKH GLVVRFLDWLRQ RI WKH SURWRQDWHG PROHFXOH P] f &,' RI WKH VRGLDWHG HQNHSKDOLQ ZDV OLPLWHG RQO\ WKH ORVV RI WKH &WHUPLQDO OHXFLQH PRLHW\ ZDV SURPLQHQW P] f )LJXUH (OHFWURVSUD\ LRQL]DWLRQ PDVV VSHFWUD RI /( $f 5HSHOOHU DW 9 %f 5HSHOOHU DW 9

PAGE 121

>01DI )LJXUH (OHFWURVSUD\ LRQL]DWLRQ PDVV VSHFWUD RI >+27\UA/( $f 5HSHOOHU DW 9 %f 5HSHOOHU DW 9 7KH LGHQWLILFDWLRQ RI WKH K\GUR[\ODWHG SURGXFWV RI /( ZDV DFKLHYHG E\ FRPSDULQJ WKH (6, PDVV VSHFWUD REWDLQHG WR WKRVH RI WKH SDUHQW SHSWLGH 7KH PRQRK\GUR[\ODWHG HQNHSKDOLQ >+2f7\U@/( JDYH PROHFXODU LRQV DW P] DQG IRU >0+@ DQG >01D@ UHVSHFWLYHO\ )LJXUH $f $V VKRZQ LQ )LJXUH % D 'D LQFUHDVH ZDV REVHUYHG LQ WKH P] YDOXH RI WKH D DQG E VHTXHQFH LRQV VWDUWLQJ ZLWK WKH DM LRQ RI WKH VHULHV P] f 2Q WKH RWKHU KDQG WKH LQWHUQDO IUDJPHQWV P] DQG ZHUH QRW

PAGE 122

DIIHFWHG 7KLV LV HYLGHQFH WKDW WKH SKHQ\ODODQLQH PRLHW\ ZDV QRW PRGLILHG 7KHUHIRUH WKH R[\JHQ ZDV LQFRUSRUDWHG DV H[SHFWHG LQWR WKH >7\U@ UHVLGXH 7KH (6, PDVV VSHFWUD UHFRUGHG IURP WKH VHFRQG SURGXFW RI WKH HQ]\PDWLF UHDFWLRQ )LJXUH f LQGLFDWHG EDVHG RQ WKH 'D LQFUHPHQW RI WKH PROHFXODU LRQV DQG DOVR RI WKH D DQG E VHTXHQFH LRQV IURP DM WR D P] WR DV FRPSDUHG WR WKH SDUHQW HQNHSKDOLQ WKDW DGGLWLRQDO K\GUR[\ODWLRQ DOVR WRRN SODFH DW WKH 1WHUPLQDO W\URVLQH WR JLYH D GLK\GUR[\ODWHG GHULYDWLYH >+2f7\U@/( 7KH UHODWLYH LQWHQVLW\ RI WKH SHDNV REWDLQHG IURP PDVV VSHFWURPHWULF DQDO\VLV VKRZHG WKDW >+2A7\UA/( ZDV D PLQRU SURGXFW DQG WKHUHIRUH VXIILFLHQW TXDQWLWLHV FRXOG QRW EH FROOHFWHG IRU XVH LQ UHFHSWRU ELQGLQJ VWXGLHV $ OO 2 V D O >01D@ >0+@ G V QL] O LfN )LJXUH (OHFWURVSUD\ LRQL]DWLRQ PDVV VSHFWUD RI >+2f7\U@/( $f 5HSHOOHU DW 9 %f 5HSHOOHU DW 9

PAGE 123

5HFHSWRU %LQGLQJ $IILQLW\ 5HSUHVHQWDWLYH GLVSODFHPHQW FXUYHV VKRZLQJ WKH ELQGLQJ RI /( DQG >+27\Un@/( WR UHFHSWRUV LQ UDW EUDLQ KRPRJHQDWH XVLQJ >+@GLSUHQRUSKLQH >+@'$*2 S VLWHVf DQG >+@'3'3( VLWHVf DV WUDFHU DUH VKRZQ LQ )LJV DQG UHVSHFWLYHO\ 7KH ,& YDOXHV REWDLQHG DUH VKRZQ LQ 7DEOH 7KH ORVV RI DIILQLW\ WR WRWDO RSLRLG UHFHSWRUV LQ UDW EUDLQ KRPRJHQDWH REVHUYHG ZKHQ /( LV K\GUR[\ODWHG E\ PXVKURRP W\URVLQDVH LV PLUURUHG E\ D ORVV RI DIILQLW\ WR ERWK S DQG VLWHV )LJXUH 'LVSODFHPHQW FXUYHV IRU /( ff DQG >+27\UA/( Rf LQ UDW EUDLQ KRPRJHQDWH XVLQJ >)L@GLSUHQRUSKLQH DV WUDFHU

PAGE 124

)LJXUH 'LVSODFHPHQW FXUYHV IRU /( ff DQG >+27\UA/( Rf LQ UDW EUDLQ KRPRJHQDWH XVLQJ >+@'$*2 DV WUDFHU )LJXUH 'LVSODFHPHQW FXUYHV IRU /( ff DQG >+27\UA/( Rf LQ UDW EUDLQ KRPRJHQDWH XVLQJ >+@'3'3( DV WUDFHU

PAGE 125

7DEOH 6XPPDU\ RI ,& YDOXHV REWDLQHG LQ UDW EUDLQ KRPRJHQDWH UHFHSWRU ELQGLQJ VWXGLHV XVLQJ >+@GLSUHQRUSKLQH >+@'$*2 DQG >+@'3'3( DV WUDFHUV ,& Q0f s 6' Q >+@GLSUHQRUSKLQH >+O'$*2 >+@'3'3( /( s s s >+27\Un@/( s s s 'LVFXVVLRQ 8VLQJ PDVV VSHFWURPHWU\ WKH SURGXFWV RI WKH UHDFWLRQ EHWZHHQ OHXFLQH HQNHSKDOLQ DQG PXVKURRP W\URVLQDVH LQ WKH SUHVHQFH RI DVFRUELF DFLG ZHUH SRVLWLYHO\ LGHQWLILHG DV >+27\Uf@/( DQG >+2A7\8M/( 7KH LGHQWLILFDWLRQ RI >+27\Un@/( FRQILUPV WKH ILQGLQJV RI SUHYLRXV ZRUNHUV >5RVHL HW DO 5RVHL HW DO @ KRZHYHU WKH GLK\GUR[\ODWHG GHULYDWLYH >+2A7\8M/( ZDV QRW LVRODWHG E\ WKHVH LQYHVWLJDWRUV DV LWV SUHVHQFH SUREDEO\ FRXOG QRW EH GHWHFWHG E\ PHDQV RI XOWUDYLROHW VSHFWURPHWU\ ZLWKRXW WKH EHQHILW RI VHSDUDWLRQ RI UHDFWLRQ FRPSRQHQWV E\ +3/& 7\URVLQDVH KDV EHHQ VKRZQ WR SURGXFH 2+GRSD IURP GRSD LQ WKH SUHVHQFH RI DVFRUELF DFLG >+DQVVRQ HW DO @ 6LPLODUO\ >+27\Uf@/( VHHPV WR DFW DV D VXEVWUDWH IRU W\URVLQDVH WR JLYH >+2A7\U@ /( 7KLV VWHSZLVH UHDFWLRQ PHFKDQLVP LV VXSSRUWHG E\ WKH VORZHU DQG GHOD\HG IRUPDWLRQ RI WKH GLK\GUR[\ODWHG GHULYDWLYH FRPSDUHG WR WKH IRUPDWLRQ RI >+2f7\U@/( 7KH ELRORJLFDO DFWLYLW\ RI RSLRLG SHSWLGHV ZLWK PRGLILHG 7\U UHVLGXHV KDV QRW EHHQ VWXGLHG WR WKH VDPH H[WHQW DV >GHV7\U@ RSLRLG SHSWLGHV ZKLFK JHQHUDOO\ VKRZ D ORVV RI DFWLYLW\ >+DQVHQ DQG 0RUJDQ 6KLPRKLJDVKL @ 7DEOH f +DQVHQ HW DO >+DQVHQ HW DO @ VKRZHG WKDW WKH LQWURGXFWLRQ RI PHWK\O JURXSV LQ WKH RUWKR SRVLWLRQ

PAGE 126

7DEOH (IIHFW RI PRGLILFDWLRQ RI 7\U PRLHW\ RI RSLRLG SHSWLGHV RQ RSLRLG ELQGLQJ DIILQLW\ 0RGLILFDWLRQ DW 7YU (IIHFW 5HIHUHQFH >GHV7\U@ RSLRLG SHSWLGH ORVV RI RSLRLG ELQGLQJ DIILQLW\ +DQVHQ DQG 0RUJDQ 6KLPRKLJDVKL + +2f§A Af§5 LQFUHDVHG RSLRLG ELQGLQJ DIILQLW\ +DQVHQ HW DO +r? 9 UHGXFHG RSLRLG ELQGLQJ DIILQLW\ -XGG HW DO +2 KRff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

PAGE 127

DFWV DV D GULYLQJ IRUFH IRU WKH HOHFWURSKLOLF VXEVWLWXWLRQ 7KHUHIRUH LQ FRQMXQFWLRQ ZLWK WKH REVHUYDWLRQV RI -XGG HW DO >-XGG HW DO @ RXU ILQGLQJV LQGLFDWH WKDW WKH SUHVHQFH RI D K\GUR[\ JURXS LQ WKH PHWD SRVLWLRQ UDWKHU WKDQ WKH DEVHQFH RI RQH LQ WKH SDUD SRVLWLRQ JLYHV ULVH WR D GHFUHDVH LQ RSLRLG UHFHSWRU ELQGLQJ LQ HQNHSKDOLQV 7\URVLQDVH DQG HQNHSKDOLQ LPPXQRUHDFWLYLWLHV KDYH EHHQ LGHQWLILHG LQ LVRODWHG FHOOV >.LPXUD HW DO @ DQG WKH VSLQDO DQG EUDLQ UHJLRQV LQ VRPH VSHFLHV QDPHO\ WKH ORFXV FRHUXOHXV FRPSOH[ LQ WKH FDW >=KXR HW DO @ DQG WKH YHQWUDO WHJPHQWDO DUHD LQ WKH UDW >6HVDFN DQG 3LFNHO @ $OWKRXJK DPLQRSHSWLGDVHV FDUER[\SHSWLGDVHV DQG HQNHSKDOLQDVHV DUH WKRXJKW WR EH PDLQO\ UHVSRQVLEOH IRU WKH PHWDEROLF GHDFWLYDWLRQ RI /( >9HQWXUHOOL HW DO @ JLYHQ WKDW W\URVLQDVH DQG /( KDYH EHHQ IRXQG WR FRH[LVW LQ YLYR DQG WKDW WKH SURGXFW RI WKH UHDFWLRQ EHWZHHQ WKHVH WZR HQWLWLHV VKRZV GHFUHDVHG DIILQLW\ WR ERWK S DQG RSLRLG UHFHSWRUV FRPSDUHG WR WKH SDUHQW HQNHSKDOLQ ZH VSHFXODWH WKDW W\URVLQDVH PD\ DOVR FRQWULEXWH WR WKH PHWDEROLF IDWH RI /( LQ YLYR

PAGE 128

&+$37(5 &21&/86,216 ,Q WKH ZRUN FDUULHG RXW IRU WKLV GLVVHUWDWLRQ WKH IHDVLELOW\ RI ERWK DQ HQ]\PHOLQNHG LPPXQRVRUEHQW DVVD\ (/,6$f DQG DQ KRPRJHQHRXV IOXRUHVFHQFH LPPXQRDVVD\ IRU OHXFLQH HQNHSKDOLQ /(f ZDV HYDOXDWHG 7ZR KLJKSHUIRUPDQFH OLTXLG FKURPDWRJUDSK\ +3/&f DSSURDFKHV IRU RSLRLG SHSWLGHV XVLQJ /( DV D PRGHO SHSWLGH DQG HQ]\PDWLF GHULYDWL]DWLRQ E\ W\URVLQDVH ZHUH DOVR HYDOXDWHG 2QH RI WKHVH +3/& DSSURDFKHV XVHG HOHFWURFKHPLFDO GHWHFWLRQ DV D PHDQV RI TXDQWLWDWLRQ DQG WKH RWKHU XVHG IOXRUHVFHQFH GHWHFWLRQ IROORZLQJ D VHFRQG IOXRURJHQLF GHULYDWL]DWLRQ VWHS ZLWK GLDPLQR GLSKHQ\OHWKDQH '3(f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

PAGE 129

WKH DQWLERG\ SURGXFHG LQ WKLV ODERUDWRU\ RU WKH FRPPHUFLDO DQWLVHUXP WHVWHG DQG DQ\ RI WKH ELRWLQ\ODWHG GHULYDWLYHV V\QWKHVL]HG 7KHUHIRUH D ZRUNDEOH (/,6$ IRU /( FRXOG QRW EH GHYHORSHG XVLQJ WKH SURSRVHG DSSURDFK (/,6 $V IRU 3HQGRUSKLQ DQG G\QRUSKLQ $O KDYH DOUHDG\ EHHQ GHYHORSHG XVLQJ WKH VDPH VWUDWHJ\ SURSRVHG KHUH IRU WKH GHYHORSPHQW RI DQ (/,6$ IRU /( >+RFKKDXV DQG 6DGHH +RFKKDXV DQG +X @ +RZHYHU WKHVH RSLRLG SHSWLGHV FRQVLVW RI ORQJHU DPLQR DFLG FKDLQV WKDQ /( DPLQR DFLGV IRU 3 HQGRUSKLQ DQG DPLQR DFLGV IRU G\QRUSKLQ FRPSDUHG WR RQO\ DPLQR DFLGV IRU /(f DQG WKHUHIRUH LW LV VXJJHVWHG WKDW LI D ELRWLQ\ODWHG /( GHULYDWLYH LQFOXGLQJ D ORQJHU VSDFHU DUP FDQ EH V\QWKHVL]HG D VDQGZLFK EHWZHHQ DQWL/( DQWLERG\ WKLV ELRWLQ\ODWHG /( GHULYDWLYH DQG DYLGLQ FRXOG EH IRUPHG OHDGLQJ WR WKH VXFFHVVIXO GHYHORSPHQW RI DQ (/,6$ IRU /( EDVHG RQ WKH VDPH SULQFLSOHV XVHG LQ WKH (/,6$V IRU 3HQGRUSKLQ DQG G\QRUSKLQ $O ZKLFK KDYH EHHQ GHVFULEHG HDUOLHU >+RFKKDXV DQG 6DGHH +RFKKDXV DQG +X @ 7KH VXFFHVVOLLO GHYHORSPHQW RI WKH KRPRJHQHRXV IOXRUHVFHQFH LPPXQRDVVD\ SURSRVHG LQ WKLV GLVVHUWDWLRQ UHOLHG RQ D ODFN RI VDQGZLFK IRUPDWLRQ EHWZHHQ DQ DQWL/( DQWLERG\ D ELRWLQ\ODWHG GHULYDWLYH DQG IOXRUHVFHQFHODEHOHG DYLGLQ VLQFH KHUH /( DQG ELRWLQ\ODWHG /( FRPSHWH IRU DQWLERG\ ELQGLQJ VLWHV DQG VXEVHTXHQWO\ IUHH ELRWLQ\ODWHG /( LV GHWHFWHG RQ DGGLWLRQ RI IOXRUHVFHQFHODEHOHG DYLGLQ GXH WR WKH LQFUHDVH LQ IOXRUHVFHQFH REVHUYHG RQ LQWHUDFWLRQ EHWZHHQ ELRWLQ\ODWHG /( DQG IOXRUHVFHQFHODEHOHG DYLGLQ $ ODFN RI VDQGZLFK IRUPDWLRQ ZDV VHHQ XVLQJ WKH DQWL/( DQWLERG\ SURGXFHG LQ WKLV ODERUDWRU\ D ELRWLQ\ODWHG GHULYDWLYH ELRWLQ\ODWHG DW WKH 1WHUPLQDO HQG RI WKH SHSWLGH ZLWKRXW D VSDFHU DUP %/(f DQG IOXRUHVFHLQ LVRWKLRF\DQDWHODEHOHG DYLGLQ ),7&DYLGLQf $OWKRXJK DV H[SHFWHG DQ LQFUHDVH LQ IOXRUHVFHQFH ZDV VHHQ LQ RXU KRPRJHQRXV IOXRUHVFHQFH

PAGE 130

LPPXQRDVVD\ ZKHQ LQFUHDVHG DPRXQWV RI /( ZHUH SUHVHQW LQ WKH VDPSOH ZH REVHUYHG WKDW DW KLJKHU FRQFHQWUDWLRQV RI /( r 2n PROHVPO DQG DERYHf WKH DQDO\WH LQWHUDFWHG GLUHFWO\ ZLWK ),7&DYLGLQ WR SURGXFH DQ LQFUHDVH LQ IOXRUHVFHQFH LQ WKH DEVHQFH RI %/( >GHV 7\Un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

PAGE 131

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f WKHVH DSSURDFKHV PD\ EH VXLWDEOH IRU WKH GHWHUPLQDWLRQ RI HOHYDWHG SK\VLRORJLFDO OHYHOV WR EH H[SHFWHG LQ FOLQLFDO VWXGLHV 7KH +3/&(' DSSURDFK GHVFULEHG LQ &KDSWHU LQYROYHG D VLQJOH PLQXWH GHULYDWL]DWLRQ VWHS SULRU WR LQMHFWLRQ RI WKH VDPSOH LQWR WKH +3/& V\VWHP FRPSDUHG WR WZR RQH KRXU GHULYDWL]DWLRQ VWHSV IRU WKH +3/&)/ DVVD\ GHVFULEHG LQ &KDSWHU 7KH +3/& (' PHWKRG DOVR UHVXOWHG LQ D VOLJKWO\ ORZHU OLPLW RI GHWHFWLRQ IRU /( WKDQ WKH ),3/&)/

PAGE 132

PHWKRG IPROLQM LQ EXIIHU DQG IPROLQM LQ &6) FRPSDUHG WR IPROLQM LQ ERWK EXIIHU DQG &6)f DQG \LHOGV D FOHDQHU FKURPDWRJUDSK ZLWK WKH SHDN RI LQWHUHVW HOXWLQJ DIWHU D VKRUWHU UHWHQWLRQ WLPH PLQXWHV FRPSDUHG WR PLQXWHVf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f7\Uf@/( 7KH DIILQLW\ RI >+2 7\Un@/( WR RSLRLG UHFHSWRUV LQ UDW EUDLQ KRPRJHQDWH ZDV GHWHUPLQHG E\ UDGLRUHFHSWRU DVVD\ +\GUR[\ODWLRQ RI /( ZDV IRXQG WR GHFUHDVH UHFHSWRU DIILQLW\ WR ERWK S DQG RSLRLG UHFHSWRU VLWHV E\ D IDFWRU RI DERXW 7\URVLQDVH DQG HQNHSKDOLQ KDYH EHHQ IRXQG WR FRn H[LVW LQ LVRODWHG FHOOV DQG LQ WKH VSLQDO DQG EUDLQ UHJLRQV RI VRPH VSHFLHV 7KHUHIRUH VLQFH ZH KDYH GHPRQVWUDWHG WKDW WKH SURGXFW RI WKH UHDFWLRQ EHWZHHQ WKHVH WZR HQWLWLHV VKRZV GHFUHDVHG DIILQLW\ WR RSLRLG UHFHSWRUV FRPSDUHG WR WKH SDUHQW HQNHSKDOLQ ZH VSHFXODWH WKDW W\URVLQDVH PD\ SOD\ D UROH LQ WKH PHWDEROLF SDWKZD\ RI /( LQ YLYR

PAGE 133

$33(1',; $ '$7$ )25 +3/&(' $3352$&+ &XUYH 7\URVLQDVH GHULYDWL]DWLRQ ERURQDWH FOHDQ XS 0DWUL[ %XIIHU (TXDWLRQ \ O[ U /( PROLQM 3HDN DUHD rf r r r n r 2n &XUYH 7\URVLQDVH GHULYDWL]DWLRQ ERURQDWH FOHDQ XS 0DWUL[ %XIIHU (TXDWLRQ \ [ U /( PROLQM 3HDN DUHD r n r r n rn r 2n r f r n

PAGE 134

&XUYH 7\URVLQDVH GHULYDWL]DWLRQ ERURQDWH FOHDQ XS 0DWUL[ %XIIHU (TXDWLRQ \ [ U /( PROLQM 3HDN KHLJKW PPf r n r 2n r f r 2f r &XUYH 7\URVLQDVH UHDFWLRQ ERURQDWH FOHDQ XS 0DWUL[ &6) (TXDWLRQ \ [ U /( PROLQM 3HDN DUHD r n rn r f r f r n r r n

PAGE 135

&XUYH 7\URVLQDVH UHDFWLRQ ERURQDWH FOHDQ XS 0DWUL[ &6) (TXDWLRQ \ [ U /( PROLQM 3HDN DUHD r n I r n rf rf r n r 2f r n WRXOOLHU FDOFXODWHG FRQFHQWUDWLRQ GHYLDWHV IURP QRPLQDO FRQFHQWUDWLRQ E\ !b + 7 .DUQHV SHUVRQDO FRPPXQLFDWLRQf &XUYH 7\URVLQDVH UHDFWLRQ ERURQDWH FOHDQ XS 0DWUL[ &6) (TXDWLRQ \ [ U /( PROLQM 3HDN DUHD rn r n r n r 2n rf r

PAGE 136

&XUYH 1R GHULYDWL]DWLRQ QR ERURQDWH FOHDQ XS 0DWUL[ &6) (TXDWLRQ \ [ U /( PROLQM 3HDN DUHD r n r 2n r f r n rf r 2n r n &XUYH 1R GHULYDWL]DWLRQ QR ERURQDWH FOHDQ XS 0DWUL[ &6) (TXDWLRQ \ [ U /( PROLQM 3HDN DUHD r 2f r 2n rn r n r f r n

PAGE 137

&XUYH 1R GHULYDWL]DWLRQ QR ERURQDWH FOHDQ XS 0DWUL[ &6) (TXDWLRQ \ [ U /( PROLQM 3HDN DUHD r r n r 2n r f I r f r f r f ‘I RXWOLHU

PAGE 138

$33(1',; % '$7$ )25 +3/&)/ $3352$&+ &XUYH 1RQFRQFHQWUDWHG UHDJHQWV 0DWUL[ %XIIHU (TXDWLRQ \ r[ U /( PROLQM 3HDN DUHD r n r r n rn r n &XUYH 1RQFRQFHQWUDWHG UHDJHQWV 0DWUL[ EXIIHU (TXDWLRQ \ r[ U /( PROLQM 3HDN KHLJKW PPf r r r r n rf

PAGE 139

&XUYH 1RQFRQFHQWUDWHG UHDJHQWV 0DWUL[ EXIIHU (TXDWLRQ \ r [ U /( PROLQM 3HDN KHLJKW PPf r r r ORVW r n r 2n r f &XUYH 1RQFRQFHQWUDWHG UHDJHQWV 0DWUL[ &6) (TXDWLRQ \ r [ U /( PROLQM 3HDN DUHD rf r f r f r 2n rn

PAGE 140

&XUYH 1RQFRQFHQWUDWHG UHDJHQWV 0DWUL[ &6) (TXDWLRQ \ r [ U /( PROLQM 3HDN KHLJKW PPf r n r r f r &XUYH &RQFHQWUDWHG UHDJHQWV 0DWUL[ %XIIHU (TXDWLRQ \ r [ U /( PROLQM 3HDN DUHD r n r rn rn

PAGE 141

&XUYH &RQFHQWUDWHG UHDJHQWV 0DWUL[ %XIIHU (TXDWLRQ \ r[ U /( PROLQM 3HDN DUHD rn OrL r 2n rL r n I r f fcfRXWOLHU FDOFXODWHG FRQFHQWUDWLRQ GHYLDWHV IURP QRPLQDO FRQFHQWUDWLRQ E\ !b + 7 .DUQHV SHUVRQDO FRPPXQLFDWLRQf &XUYH &RQFHQWUDWHG UHDJHQWV 0DWUL[ &6) (TXDWLRQ \ r X[ U /( PROLQM 3HDN DUHD r 2n rn rf ORVW r 2n r f W IRXOOLHU

PAGE 142

&XUYH &RQFHQWUDWHG UHDJHQWV 0DWUL[ &6) (TXDWLRQ \ r[ U /( PROLQM 3HDN DUHD rn rn r r f I r 2f WRXWOLHU &XUYH &RQFHQWUDWHG UHDJHQWV 0DWUL[ &6) (TXDWLRQ \ r[ U /( PROLQM 3HDN DUHD r n rf r 2n I r n r IRXWOLHU

PAGE 143

&XUYH &RQFHQWUDWHG UHDJHQWV PLQXWH W\URVLQDVH UHDFWLRQ 0DWUL[ %XIIHU (TXDWLRQ \ r [ U /( PROLQM 3HDN KHLJKW PPf r n r r rn r r r n &XUYH &RQFHQWUDWHG UHDJHQWV PLQXWH W\URVLQDVH UHDFWLRQ 0DWUL[ %XIIHU (TXDWLRQ \ r [ U /( PROLQM 3HDN KHLJKW PPf rf r n r n rf r

PAGE 144

&XUYH &RQFHQWUDWHG UHDJHQWV PLQXWH W\URVLQDVH UHDFWLRQ 0DWUL[ %XIIHU (TXDWLRQ \ r[ U /( PROLQM 3HDN KHLJKW PPf r n r 2f rf r f OLW r 2f r f rn I rn WRXWOLHU

PAGE 145

5()(5(1&(6 $NLO + %URQVWHLQ DQG $ 0DQVRXU f 2YHUYLHZ RI WKH (QGRJHQRXV 2SLRLG 6\VWHPV $QDWRPLFDO %LRFKHPLFDO DQG )XQFWLRQDO ,VVXHV LQ (QGRUSKLQV 2SLDWHV DQG %HKDYLRXUDO 3URFHVVHV SJ f 5 5RJHUV t 6 &RRSHU (GVf &KLFKHVWHU -RKQ :LOH\ t 6RQV $O+DNLHP 0 + + /DQGRQ 6 6PLWK DQG 5 1DUJHVVL f $QDO\WLFDO %LRFKHPLVWU\ $OOHQ 0 + DQG 0 / 9HVWDO f -RXUQDO RI WKH $PHULFDQ 6RFLHW\ RI 0DVV 6SHFWURPHWU\ $UQRWW 6KDEDQRZLW] DQG ) +XQW f &OLQLFDO &KHPLVWU\ %LHPDQQ f $QQXDO 5HYLHZV LQ %LRFKHPLVWU\ %RZGHQ : 0 5LVLQJ $NRWV $ 0\OHV DQG 5 %URH]H f &OLQLFDO &KHPLVWU\ &DUU 6 $ f $GYDQFHG 'UXJ 'HOLYHU\ 5HYLHZV &OHPHQW-RQHV 9 DQG 0 %HVVHU f %ULWLVK 0HGLFDO %XOOHWLQ &OHPHQW-RQHV 9 DQG / + 5HHV f LQ &OLQLFDO 1HXURHQGRFULQRORJ\ YRO ,, SJ f 0 %HVVHU t / 0DUWLQ (GV f 1HZ
PAGE 146

'DYH ) 6WREDXJK 7 0 5RVVL DQG & 0 5LOH\ f -RXUQDO RI 3KDUPDFHXWLFDO DQG %LRPHGLFDO $QDO\VLV GH &HEDOORV 0 0 7D\ORU DQG 3 -HQQHU f 1HXURSHSWLGHV GH 0RQWLJQ\ 3 & 0 5LOH\ / $ 6WHUQVRQ DQG ) 6WREDXJK f -RXUQDO RI 3KDUPDFHXWLFDO DQG %LRPHGLFDO $QDO\VLV 'HIUXWRV 0 DQG ) ( 5HJQLHU f $QDO\WLFDO &KHPLVWU\ $ (LVHQDFK & & 'REVRQ & ( ,QWXUULVL ' +RRG DQG 3 % $JQHU f 3DLQ (QJHO : DQG 3 .KDQQD f -RXUQDO RI ,PPXQRORJLFDO 0HWKRGV (ULNVVRQ % DQG 0 :LNVWURP f -RXUQDO RI &KURPDWRJUDSK\ )LQQ ) 0 7LWXV DQG +RIPDQQ f %LRFKHPLVWU\ )OHPLQJ / + DQG 1 5H\QROGV f -RXUQDO RI &KURPDWRJUDSK\ *DUFLD 6 ) $ 1DYDV DQG /RYLOOR f $QDO\WLFDO %LRFKHPLVWU\ *HPDQW $ f 0ROHFXODU %LRORJ\ 5HSRUWV *RVOLQJ 3 f &OLQLFDO &KHPLVWU\ *UHHQ 1 0 f 0HWKRGV LQ(Q]\PRORJ\ $ *UHHQ 1 0 / .RQLHF]Q\ ( 7RPV DQG 5 & 9DOHQWLQH f 3URFHHGLQJV RI WKH 1DWLRQDO $FDGHP\ RI 6FLHQFHV *URVVPDQ $ DQG / + 5HHV f %ULWLVK 0HGLFDO %XOOHWLQ +DQGD % $ & /DQH $ + /RUG % $ 0RUJDQ 0 5DQHH DQG & ) & 6PLWK f (XURSHDQ -RXUQDO RI 3KDUPDFRORJ\ +DQVHQ : 5 + 0D]XU DQG 0 &ODUH f 7KH 6\QWKHVLV DQG 6$5 RI 2UDOO\ $FWLYH (QNHSKDOLQ $QDORJV ZLWK 0RGLILHG 1WHUPLQDO 7\URVLQH 5HVLGXHV LQ 3HSWLGHV 6WUXFWXUH DQG )XQFWLRQ SJ f 0 'HEHU 9 +UXE\ t .RSSOH (GVf 5RFNIRUG ,/ 3LHUFH &KHPLFDO &RPSDQ\ +DQVHQ 3 ( DQG % $ 0RUJDQ f 6WUXFWXUH$FWLYLW\ 5HODWLRQVKLSV LQ (QNHSKDOLQ 3HSWLGHV LQ 2SLRLG 3HSWLGHV %LRORJ\ &KHPLVWU\ DQG *HQHWLFV YRO SJ f 6 8GHQIULHQG t 0HLHQKRIHU (GVf 2UODQGR $FDGHPLF 3UHVV ,QF

PAGE 147

+DQVVRQ & + 5RUVPDQ DQG ( 5RVHQJUHQ f $FWD 'HUPDWRYHQHURORJLFD 6WRFNKROPf +DUGHER ( 5 (NPDQ DQG 0 (ULNVVRQ f +HDGDFKH +DUORZ ( DQG /DQH f $QWLERGLHV $ /DERUDWRU\ 0DQXDO &ROG 6SULQJ +DUERU &ROG 6SULQJ +DUERU /DERUDWRU\ +HHUGHQ ( $KOVNRJ DQG ( %\HU f -RXUQDO RI /DERUDWRU\ DQG &OLQLFDO 0HGLFLQH +LJD 6 7 6X]XNL $ +D\DVKL 7VXJH DQG <
PAGE 148

-HQNLQV 6 + f -RXUQDO RI ,PPXQRORJLFDO 0HWKRGV -XGG $ / 5 7ROO $ /DZVRQ ( 7 8\HQR : ( 3ROJDU DQG + /RHZ f 3XWDWLYH 2SLRLG DQWDJRQLVWV 6\QWKHVLV DQG %LRORJLFDO SURSHUWLHV RI '$OD 0HW(QN $PLGH $QDORJV ZLWK 8QXVXDO 7\U 5HVLGXHV LQ 3HSWLGHV 6WUXFWXUH DQG )XQFWLRQ SJ f 0 'HEHU 9 +UXE\ t .RSSOH (GV f 5RFNIRUG ,/ 3LHUFH &KHPLFDO &RPSDQ\ .DEDQRY $ 9 0 0 .KUXWVND\D 6 $ (UHPLQ 1 / .O\DFKNR DQG $ 9 /HYDVKRY f $QDO\WLFDO %LRFKHPLVWU\ .DL 0 ,VKLGD DQG < 2KNXUD f -RXUQDO RI &KURPDWRJUDSK\ .DWWD 9 6 &KRZGKXU\ DQG % 7 &KDLW f $QDO\WLFDO &KHPLVWU\ .LP & 5 &KHQJ DQG 6 5 *HRUJH f -RXUQDO RI &KURPDWRJUDSK\ .LPXUD 1 +
PAGE 149

0F'HUPRWW 5 $ 6PLWK $ %LJJLQV 0 & $O1RDHPL DQG $ (GZDUGVRQ f -RXUQDO RI &KURPDWRJUDSK\ 0F4XHHQ 6 f %ULWLVK 0HGLFDO %XOOHWLQ 0HUFKHQWKDOHU f (QGRFULQRORJ\ 0LIXQH 0 .UHKELHO ) 6WREDXJK DQG & 0 5LOH\ f -RXUQDO RI &KURPDWRJUDSK\ 0LWVXL $ + 1RKWD DQG < 2KNXUD f -RXUQDO RI &KURPDWRJUDSK\ 0RQJHU / 6 DQG & 2OOLI7 f -RXUQDO RI &KURPDWRJUDSK\ 0RVEHUJ + 5 +XUVW 9 +UXE\ *HH +
PAGE 150

5RHSVWRUII 3 DQG )RKOPDQ f %LRPHGLFDO 0DVV 6SHFWURPHWU\ 5RVHL 0 $ $ $QWRQLOOL 5 &RFFLD DQG & %ODU]LQR f 3K\VLRORJLFDO &KHPLFDO 3K\VLFDO DQG 0HGLFDO 105 5RVHL 0 $ / $QWRQLOOL 5 &RFFLD DQG & )RSSROL f %LRFKHPLVWU\ ,QWHUQDWLRQDO 5RWK 0 f $QDO\WLFDO &KHPLVWU\ 5XELQ 3 & f &OLQLFDO 6FLHQFH 6DPXHOVVRQ + 5 (NPDQ DQG 7 +HGQHU f $FWD $QDHVWKHVLRORJLFD 6FDQGLQDYLFD 6DUPD 6 5 +RIIPDQQ DQG 5 $ +RXJKWHQ f /LIH 6FLHQFHV 6DWR + f /LIH 6FLHQFHV 6FDWFKDUG f $QQDOV RI WKH 1HZ
PAGE 151

7HOOLHU 0 5 3UDQNHUG DQG +RFKKDXV f -RXUQDO RI 3KDUPDFHXWLFDO DQG %LRPHGLFDO $QDO\VLV 8GHQIULHQG 6 6 6WHLQ 3 %RKOHQ : 'DLUPDQ : /HLPJUXEHU DQG 0 :HLJHOH f 6FLHQFH 8PHGD 0 < ,VKLPRUL
PAGE 152

%,2*5$3+,&$/ 6.(7&+ 9HURQLTXH /DUVLPRQW ZDV ERUQ LQ 7DLSHL 7DLZDQ LQ DQG VSHQW KHU FKLOGKRRG WUDYHOLQJ WKH ZRUOG ZLWK KHU SDUHQWV XQWLO VKH ZHQW WR ERDUGLQJ VFKRRO DW 'ROODU $FDGHP\ 'ROODU 6FRWODQG LQ 6KH HDUQHG D %6F +RQRXUVf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

PAGE 153

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i W87 ,DQ 7HEEHWW $VVRFLDWH 3URIHVVRU RI 3KDUPDFHXWLFV

PAGE 154

7KLV GLVVHUWDWLRQ ZDV VXEPLWWHG WR WKH *UDGXDWH )DFXOW\ RI WKH &ROOHJH RI 3KDUPDF\ DQG WR WKH *UDGXDWH 6FKRRO DQG ZDV DFFHSWHG D SDUWLDO IXOILOOPHQW RI WKH UHTXLUHPHQWV IRU WKH GHJUHH RI 'RFWRU RI 3KLORVRSK\ $XJXVW 'HDQ *UDGXDWH 6FKRRO

PAGE 155

81,9(56,7< r) ,OO8MMMOOOOO OLOL


32
neutralized immediately with 35 pi of 2 M Tris buffer pH 9. The fractions with high
optical density at taso were pooled and then desalted and concentrated by spinning down
5-6 times for 30 minutes in a C-PREP 10 centrifugal filter (Amicon, Danvers, MA, USA).
This procedure yielded 8.5 ml of purified anti-LE antibody containing 6.1 mg IgG/ml,
determined by measurement of optical density at taso-
Characterization of Antibody
The number of specific antibody sites in the purified antibody was determined by
means of Scatchard analysis [Scatchard 1949], Samples containing various concentrations
of 3H-LE as tracer (70-1000 pM) and a 1/1,000 dilution of the purified antibody were set
up to determine total binding (Table 2.1). To determine non-specific binding, samples
containing antibody, ?H-LE (70-1000 pM) and a high concentration of LE (1.2* 10"6 M)
were also set up at the same time. After overnight incubation at 4C, 200 pi of an ice cold
suspension containing 1.5% activated charcoal and 0.15% dextran in water was added to
each tube (except total counts) and after a further 5 minute incubation on ice, the samples
were centrifuged at 12,000g for 3 minutes. Three hundred and fifty microliters of the
resultant supernatant containing the antibody-bound fraction of the tracer were then
removed and added to 4 ml of Cytoscint scintillation cocktail. The radioactivity (CPM)
representing the antibody-bound tracer was determined using a Beckman LS 5,000 TD
scintillation counter (Fullerton, CA, USA) and a 5 minute counting time with counting
efficiency at about 50%.


137
Curve 13
Concentrated reagents, 5 minute tyrosinase reaction
Matrix: Buffer
Equation: y = 9.91*1012x + 6.4066, r2 = 0.9985
LE mol/inj
Peak height (mm)
0
6
2.77* 10'13
10.5
5.54*1 O13
13
1.38*1012
16.5
2.77* 1012
lit
5.54* 1 O12
63.5
6.64* 1012
43.5|
1.11*10'"
83 f
1.38*10'"
143
toutlier


NOVEL ANALYTICAL APPROACHES FOR THE DETERMINATION OF LEUCINE
ENKEPHALIN AS A MODEL FOR OPIOID PEPTIDES
By
VERONIQUE LARSIMONT
A 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
1994


48
Figure 2.9. Chromatographs of control injection (A) showing LE eluting after 9 minutes
and reaction injections showing BLE (no spacer arm) eluting after 15.3
minutes (B) and BXLE (with spacer arm) eluting after 16.3 minutes (C).
metry (MALDI) of the major peak eluting at 14.9 minutes in Figure 2.10B and at 17.8
minutes in Figure 2.IOC revealed molecular ion peaks ([M+H]+) at m/z 910.283 and
1022.1, respectively. The nominal molecular masses of the molecular ions of LE-Lys6-B
and LE-Lys6-BX were calculated to be 910.11 and 1023.27, respectively. Since the mass
determinations deviated from the nominal masses of mono-biotinylated derivatives by less
than 0.1%, these products were confirmed as mono-biotinylated derivatives of LE-Lys6.
Therefore, antibody binding experiments were carried out using these collected peaks. The


68
of the tyrosine group of LE to give a more easily oxidizable catechol allows the use of a
lower oxidation potentials for electrochemical detection, thus avoiding many of the
disadvantages associated with high applied potentials such as high background current and
baseline noise [Kim et al. 1989], Selectivity is also compromised when high applied
potentials are used as more compounds are oxidized at these high potentials. Therefore,
extensive clean-up procedures to eliminate interfering peaks are often required when high
applied potentials are used [Fleming and Reynolds 1988], Secondly, the enzymatic
derivatization increases the selectivity of this assay as the derivative is amenable to a
specific boronate clean-up method which has previously been established for
catecholamines [Eriksson and Wikstrom 1992, Higa et al. 1977, Koike et al. 1982],
immobilized boronale gel
hydroxylated LE
k
H+ OH-
complex
Figure 3.3. pH dependent complex formation between immobilized boronate gel and
hydroxylated LE.


116
Receptor Binding Affinity
Representative displacement curves showing the binding of LE and [HO-Tyr']-LE
to receptors in rat brain homogenate using [3H]-diprenorphine, [3H]-DAGO (p sites) and
[3H]-DPDPE (5 sites) as tracer are shown in Figs 5.7, 5.8 and 5.9, respectively. The IC50
values obtained are shown in Table 5.1. The loss of affinity to total opioid receptors in rat
brain homogenate observed when LE is hydroxylated by mushroom tyrosinase is mirrored
by a loss of affinity to both p and 8 sites.
Figure 5.7. Displacement curves for LE () and [HO-Tyr^-LE (o) in rat brain
homogenate using [3H]-diprenorphine as tracer.


130
Curve 9
No derivatization, no boronate clean up
Matrix: CSF
Equation: y = 11326x + 59177, r2 = 0.9529
LE mol/inj
Peak area
0
0
8.82*10'12
259029
17.28* 10'12
2408431
25.56* 1 O'12
265387
33.30* 1012
231334f
43.2* 1012
646006
69.29*1 O'12
762035
85.67* 1012
1064494
f outlier


115
affected. This is evidence that the phenylalanine moiety was not modified. Therefore, the
oxygen was incorporated, as expected, into the [Tyr1] residue. The ESI mass spectra
recorded from the second product of the enzymatic reaction (Figure 5.6) indicated, based
on the 32 Da increment of the molecular ions, and also of the a and b sequence ions from
aj to a4, m/z 168 to 429, as compared to the parent enkephalin, that additional
hydroxylation also took place at the N-terminal tyrosine, to give a di-hydroxylated
derivative, [(HO)2-Tyr1]-LE. The relative intensity of the peaks obtained from mass
spectrometric analysis showed that [(HO^-Tyr^-LE was a minor product and therefore,
sufficient quantities could not be collected for use in receptor binding studies.
ni/z
Figure 5.6. Electrospray ionization mass spectra of [(HO)2-Tyr1]-LE. (A) Repeller at
18 V; (B) Repeller at 50 V.


117
Figure 5.8. Displacement curves for LE () and [HO-Tyr^-LE (o) in rat brain
homogenate using [3H]-DAGO as tracer.
Figure 5.9. Displacement curves for LE () and [HO-Tyr^-LE (o) in rat brain
homogenate using [3H]-DPDPE as tracer.


82
To give an indication of the inter-day variability associated with the complete
HPLC-ED method incorporating enzymatic derivatization and boronate clean up, using
the calibration curves obtained for LE in CSF, nominal concentrations in the samples were
compared to found concentrations (Table 3.2). The found concentrations were determined
from the calibration curve after regression analysis was repeated while omitting the data
point under investigation. The relative standard deviation of the found concentrations was
found to be <20% (n=3) and the accuracy was found to be within 6% of the nominal
concentration.
Table 3.2. Nominal concentrations, found concentrations, relative standard deviation (SD)
and percentage accuracy calculated from 3 calibration curves for LE in CSF
using complete HPLC-ED method. For raw data, see Appendix A.
Nominal cone.
(pmol/100 pi CSF)
Found cone, n=3
(pmol/100 pi CSF)
Relative SD (%)
% Accuracy
1.76
1.84
20
104
3.46
3.34
12
97
6.67
6.41
12
96
10.48
11.13
1
106
17.14
17.65
9
103
Discussion
The HPLC-ED method we have evaluated for the analysis of LE in CSF compares
favorably to existing HPLC methods for opioid peptides with on-line detection [de
Montigny et al. 1990, Mifune et al. 1989, Muck and Henion 1989] and represent an
improvement with respect to limit of detection and practicability when compared to


114
[M+HI*
[M+Naf
572 594
Figure 5.5. Electrospray ionization mass spectra of [HO-Tyr^-LE. (A) Repeller at 18 V;
(B) Repeller at 50 V.
The identification of the hydroxylated products of LE was achieved by comparing
the ESI mass spectra obtained to those of the parent peptide. The mono-hydroxylated
enkephalin, [(HO)-Tyr1]-LE, gave molecular ions at m/z 572 and 594 for [M+H]+ and
[M+Na]+, respectively (Figure 5.5A). As shown in Figure 5.5B, a 16 Da increase was
observed in the m/z value of the a and b sequence ions, starting with the aj ion of the
series (m/z 152). On the other hand, the internal fragments m/z 120 and 177 were not


79
Calibration Curves
Table 3.1 shows the limits of detection (LOD, defined as twice the baseline noise),
average slope values and correlation coefficients (r2) obtained for the various calibration
curves constructed for LE using this analytical approach. Raw data can be found in
Appendix A. Representative calibration curves for LE in buffer and CSF using the
complete HPLC-ED method are shown in Figures 3.8 and 3.9. Figure 3.10 shows a
representative calibration curve for LE in CSF using FIPLC-ED without enzymatic
derivatization or boronate clean-up. The limits of detection for LE in CSF correspond to
8.8 pmol/ml of CSF for the complete HPLC-ED method and 176 pmol/ml of CSF for
analysis by HPLC-ED without derivatization or boronate clean-up. A sample
chromatograph of LE in CSF shows the analyte eluting after 6 minutes (Figure 3.11).
Table 3.1. Limits of detection (LOD), average slopes and correlation coefficients (r2) for
LE in buffer or CSF using either complete HPLC-ED method or HPLC-ED
without derivatization or boronate clean-up. For raw data, see Appendix A.
Sample
Method
LOD
fmol/inj
Slope (SD, n)
peak area/pmol/inj
r2 (SD, n)
Buffer
complete
HPLC-ED
170
28071 (2381, 3)
0.9925 (0.005, 3)
CSF
complete
HPLC-ED
360
22057(1064, 3)
0.9944 (0.004, 3)
CSF
HPLC-ED
no derivatization
no clean-up
8800
10217 (1163, 3)
0.9690 (0.024, 3)


140
Hansson, C., H. Rorsman and E. Rosengren (1981). Acta Dermatovenerologica
(Stockholm) 61: 147.
Hardebo, J. E., R. Ekman and M. Eriksson (1989). Headache 29: 494.
Harlow, E. and D. Lane (1988), Antibodies: A Laboratory Manual. Cold Spring Harbor:
Cold Spring Harbor Laboratory.
Heerden, J. E. Ahlskog and D. E. Byer (1990). Journal of Laboratory and Clinical
Medicine 115: 346.
Higa, S., T. Suzuki, A. Hayashi, I. Tsuge and Y. Yamamura (1977). Analytical
Biochemistry 77: 18.
Ho, R. J. Y. and L. Huang (1985). Journal of Immunology 134: 4035.
Hochhaus, G. and L. Hu (1990). Journal of Pharmaceutical and Biomedical Analysis 8:
541.
Hochhaus, G. and W. Sadee (1988). Pharmaceutical Research 5: 232.
Hochhaus, G., B. W. Gibson and W. Sadee (1988). Journal of Biological Chemistry 263:
92.
Hughes, J., T. W. Smith, H. W. Kosterlitz, L. A. Fothergill, B. A. Morgan and H. R.
Morris (1975). Nature 258: 577.
Ihara, H., T. Imamura, T. Kunitake and C. Hirayama (1988). Clnica Chimica Acta 175:
51.
Imura, H., Y. Kato, Y. Nakai, K. Nakao, I. Tanaka, H. Jingami, T. Koh, T. Yoshimasa, T.
Tsukada, M. Suda, M. Sakamoto, N. Morii, H. Takahashi, K. Tojo and A. Sugawara
(1985). Journal of Endocrinology 107: 147.
Irving, M. N. H. and R. M. Parkins (1965). Journal of Inorganic and Nuclear Chemistry
27: 270.
Ishida, J., M. Kai and Y. Ohkura (1986). Journal of Chromatography 356: 171.
Ito, S., T. Kato, K. Shinpo and K. Fujita (1984). Biochemistry Journal 222: 407.
Jaflfe, J. H. and W. R. Martin (1990). Opioid Analgesics and Antagonists, in The
Pharmacological Basis of Therapeutics (pp. 486) A. G. Goodman-Gilman (Ed.). New
York: Pergamon Press.


47
Figure 2.8. Biotinylation of peptide using biotinamidocaproate N-hydroxysuccinimide
ester.
The reactants and products of the biotinylation reactions were successfully
separated from each other using the HPLC systems described previously (Figures 2.9 and
2.10). In control injections containing LE or LE-Lys6 in concentrations equal to the initial
concentrations of these compounds in the reaction mixture, the reactants were seen to
elute after 9 minutes (Figure 2.9A) and 5.15 minutes (Figure 2.10A), respectively. The N-
terminal biotinylated LE derivatives, BLE (no spacer arm) and BXLE (with spacer arm)
were seen to elute distinct from the reactants at 15.3 and 16.3 minutes, respectively
(Figures 2.9B and 2.9C). For the C-terminal biotinylation of LE-Lys6 to give LE-Lys6-B
(no spacer arm) and LE-Lys6-BX (with spacer arm), two product peaks were seen on
injection of the incubation mixtures. Based on the delayed formation in the time course of
the reaction of the minor peak eluting at 24.8 minutes in Figure 2.10B and at 27.8 minutes
in Figure 2.10C, these minor peaks were assumed to be di-biotinylated products
biotinylated at both the N- and C-terminal ends of the peptide. Analysis by mass spectro-


81
Figure 3.10. Representative calibration curve for LE in CSF using HPLC-ED without
derivatization or boronate clean-up.
Figure 3.11. Sample chromatograph ofLE in CSF showing peak of interest eluting after
6 minutes.


133
Curve 5
Non-concentrated reagents
Matrix: CSF
Equation: y = 1.08* 1013x 0.7695, r2 = 0.9670
LE mol/inj
Peak height (mm)
0
0
8.91 10'13
12
1.78* 10"12
17
5.38* 1012
48
7.18*10"12
84
Curve 6
Concentrated reagents
Matrix: Buffer
Equation: y = 2.25* 1012x 0.8243, r2 = 0.9926
LE mol/inj
Peak area
0
0
4.47* 10'13
1.14
1.79* 10"12
7.151
6.68*10'12
14.916
1.67*10'12
38.496


2
In the proenkephalin family, proenkephalin A is the precursor of the pentapeptides
methionine enkephalin (ME, Tyr-Gly-Gly-Phe-Met) and leucine enkephalin (LE, Tyr-Gly-
Gly-Phe-Leu) which were the first opioid peptides to be characterized by Hughes and
coworkers in 1975 (Figure 1.2) [Hughes et al. 1975], Proenkephalin A has been shown to
contain ME and LE in a fixed ratio of six ME sequences to one LE sequence.
Proenkephalin-expressing cells are widespread throughout the brain and spinal cord as
well as in more peripheral sites such as the adrenal medulla and the gastrointestinal tract.
PRO-OPIOMELANOCORTIN
r-MSHj
r H i_r .
( -LPH *
*-/9-END-i
'l'
If1
in
a-MSH 0-MSH MET-ENK
PROENKEPHALIN
MET-ENK LEU-ENK
gl g
W HI
II 1
1 I
hrl
hrl
MET-ENK MET-ENK
ARG*- 6LY7-LEU* ARG*-PHE7
PRODYNORPHIN
LEU-ENK
Figure 1.2. Schematic representation of the structures of opioid peptide precursors [Jaflfe
and Martin 1990],
Pro-opiomelanocortin (POMC) is the precursor of the opioid peptide P-endorphin
as well as the non-opioid hormones adrenocorticotrophic hormone (ACTH) and a- and P-


14
Kuhling et al. 1989, Sarma et al. 1986], These assays have sensitivities ranging from 0.3 to
3.2 femtomole per assay.
The immunoassays developed by Hochhaus and Sadee [Hochhaus and Sadee
1988] and Hochhaus and Hu [Hochhaus and Hu 1990] are ELlSAs based on the avidin-
biotin system whereby the peptide of interest in the sample or standard and its biotinylated
derivative compete for antibody binding sites. The antibody-bound biotinylated species is
subsequently detected by enzymatic detection through the use of an avidin-enzyme
complex. For this dissertation, an attempt was made at the development of an extremely
sensitive avidin-biotin based ELISA for enkephalins, using LE as a model peptide, to
compliment the existing ELISA tests for P-endorphin and dynorphin. If this type of assay
can be developed for LE, it can be expected to be also applicable to ME.
In addition, for this dissertation, an attempt was also made at the development of
an homogenous or non-separation immunoassay for opioid peptides. Homogeneous
immunoassays differ from traditional immunoassays in that the labor intensive separation
of the bound and free fraction of the analyte (e g./ by washing, precipitation or
adsorbance) is not necessary prior to quantitation as the property being measured is
characteristic of either the bound or the free analyte or label.
The immunological techniques proposed in this dissertation have the disadvantage
of low specificity, but they were intended for use as "immunological HPLC detectors" in
the hope that they would provide fast, ultra-sensitive assays which were highly suitable for
processing large numbers of samples such as HPLC fractions.


15
Table 1.2. Summary of analytical methods for opioid peptides.
Reference
Method
Sensitivity
Analyte
Comments
Fleming and
Reynolds
1988
HPLC-ED
1 pmol/inj
enkephalin
tedious sample clean up,
high oxidation potential
Kim et al.
1989
HPLC-ED
1 pmol/inj
enkephalin
tedious sample clean up,
high oxidation potential
Shibanoki et
al. 1990
HPLC-ED
550 fmol/inj
enkephalin
tedious sample clean up,
high oxidation potential
Monger and
Olliif 1992
HPLC-ED
75 fmol/ ml
plasma
(3-endorphin
tedious sample clean up,
high oxidation potential
Muck and
Henion 1989
HPLC-MS
100 fmol/inj
dynorphin
microbore LC system
Mifune et al.
1989
HPLC-FL
100 fmol/inj
enkephalin
complicated column
switching
Nakano et al.
1987
HPLC-FL
140 fmol/inj
enkephalin
harsh reaction conditions
Kai et al.
1988
HPLC-FL
500 fmol/inj
enkephalin
harsh reaction conditions
van den Beld
et al. 1990
HPLC-FL
50 fmol/inj
P-endorphin
laser induced
fluorescence
Dave et al.
1992
HPLC-FL
36 fmol/inj
LE
microbore LC system
Hochhaus
and Sadee
1988
ELISA
<1 fmol/assay
P-endorphin
Hochhaus
and Hu 1990
ELISA
1 fmol/assay
dynorphin
Kuhling et al.
1989
ELISA
3 fmol/assay
P-endorphin
de Ceballos
et al. 1991
HPLC-R1A
1.5 fmol/assay
LE, ME
Maidment et
al. 1989
RIA
<1 fmol/assay
ME
HPLC-ED = high performance liquid chromatography with electrochemical detection, HPLC-FL = high
performance liquid chromatography with fluorescence detection, HPLC-MS = high performance liquid
chromatography with mass spectrometry, ELISA = enzyme-linked immunosorbent assay.


127
Curve 3
Tyrosinase derivatization, boronate clean up
Matrix: Buffer
Equation: y = 25953x + 3341, r2 = 0.9922
LE mol/inj
Peak height (mm)
0
0
1.68* 10'13
9573
3.37* 1 O'13
14456
5.06* 1013
15197
1.69*1 O12
48918
2.53*10"12
67658
Curve 4
Tyrosinase reaction, boronate clean up
Matrix: CSF
Equation: y = 21753x 12089, r2 = 0.9957
LE mol/inj
Peak area
0
11724
3.60* 10'13
11205
7.22*10'13
30931
1.42*1 O'12
43348
2.73* 1012
74283
4.29* 10'12
109291
7.01 10"12
169232
9.63* 10'12
215784


73
1
system described above. Five nanomoles of analyte per injection were applied to the
system and the responses (in pA) obtained at various potentials (+0.05 to +0.80V) at the
analytical cell were recorded to allow the construction of current-voltage curves.
Boronate Clean-up
A hydrated boronate column packed with a 3 ml bed volume of Affigel 601
(BioRad Laboratories, Melville, NY, USA) was used for the clean-up of the enzymatically
derived species prior to application to the HPLC-ED system. The column was pre-rinsed
with 0.2 M phosphate buffer (pH 8.5) and an aliquot of the incubation mixture was
applied. The column was washed with 20 ml of phosphate buffer (pH 8.5) and the analyte
was eluted from the boronate column onto a pre-conditioned Sep-Pak Cig cartridge
(Waters Associates, Milford, MA, USA) with 20 ml of aqueous 0.01% (v/v)
trifluoroacetic acid (TFA). Here, the Sep-Pak Cig cartridge served to concentrate the
sample as the analyte was eluted from the boronate column in a relatively large volume of
aqueous 0.01% (v/v) TFA The Sep-Pak Cig column was then washed with 10 ml of
aqueous 0.01% (v/v) TFA and the analyte was eluted in 2 ml of 0.01% TFA in
acetonitrile. The sample was evaporated to dryness under a stream of nitrogen and the
residue was reconstituted in 200 pi mobile phase prior to application to the HPLC-ED
system.


62
blank levels at concentrations of LE in excess of 1*1 O'8 moles/ml. This effect was also
observed when the antibody was omitted from the assay (Figure 2.19). When both tracer
and antibody were included in the assay (Figure 2.18), fluorescence levels began to
increase above fluorescence blank levels at concentrations of LE in excess of 1*10 9 moles
/ml. When antibody is omitted but tracer is included in the assay, fluorescence readings
increased above total tracer levels at LE concentrations in excess of 1*1 O'8 moles/ml
(Figure 2.19). This is the same concentration of LE at which the fluorescence readings
increased above fluorescence blank levels when both tracer and antibody were omitted
from the assay.
Des-'tyr LIC(moles/ml)
Figure 2.20. Calibration curve for des-Tyr1 LE using the homogeneous fluorescence
immunoassay set up, omitting antibody but with BLE () and without
BLE(D).
These results led us to believe that at higher concentrations of LE, the analyte
interacted directly with FITC-avidin to give increased fluorescence readings. It was
hypothesized that this effect could be due to an interaction between the tyrosine moiety in


118
Table 5.1. Summary of IC50 values obtained in rat brain homogenate receptor binding
studies using [3H]-diprenorphine, [3H]-DAGO and [3H]-DPDPE as tracers
IC50 (nM) SD, n
[3H]-diprenorphine
[3Hl-DAGO
[3H]-DPDPE
LE
26 11, 4
18 4, 3
1.2 0.25, 3
[HO-Tyr]-LE
440 150,4
381 106, 3
34 2, 3
Discussion
Using mass spectrometry, the products of the reaction between leucine enkephalin
and mushroom tyrosinase in the presence of ascorbic acid were positively identified as
[HO-Tyr]-LE and [(HO^-TyUj-LE. The identification of [HO-Tyr']-LE confirms the
findings of previous workers [Rosei et al. 1991, Rosei et al. 1989], however, the
dihydroxylated derivative, [(HO^-TyUj-LE, was not isolated by these investigators as its
presence probably could not be detected by means of ultraviolet spectrometry without the
benefit of separation of reaction components by HPLC. Tyrosinase has been shown to
produce 5-OH-dopa from dopa in the presence of ascorbic acid [Hansson et al. 1981],
Similarly, [HO-Tyrj-LE seems to act as a substrate for tyrosinase to give [(HO^-Tyr1]-
LE. This stepwise reaction mechanism is supported by the slower and delayed formation
of the dihydroxylated derivative compared to the formation of [(HO)2-Tyr1]-LE.
The biological activity of opioid peptides with modified Tyr1 residues has not been
studied to the same extent as [des-Tyr1] opioid peptides which generally show a loss of
activity [Hansen and Morgan 1984, Shimohigashi 1986] (Table 5.2). Hansen et al.
[Hansen et al. 1985] showed that the introduction of methyl groups in the ortho position


30
conjugates were stored at -20C prior to lyophilization. Lyophilization of the conjugates
was carried out at the Drug Delivery Laboratory, University of Florida, Progress Center,
Alachua, FL, USA using a Model 12K Super Modulyo Lyophilizer (Edwards, West
Sussex, England).
After lyophilization, samples of LE, BSA, porcine thyroglobulin and the
conjugates were sent for amino acid analysis at the Peptide Core of the Interdisciplinary
Center Biotechnology Research, University of Florida, Gainesville, FL, USA. The loading
ratios of LE to carrier protein were calculated from the percentage composition of
asparagine, threonine, serine and glutamine in the carriers and conjugates. The LE-
thyroglobulin conjugate obtained by the glutaraldehyde method showed the highest
peptide loading ratio (see Results and Discussion) and therefore this conjugate was
selected as the inoculum for the purposes of this work.
For the initial inoculation, complete Freunds adjuvant was placed in a test tube
and an equal volume of 0.1 M sodium phosphate buffer pH 7.4 containing 2 mg/ml of the
LE-thyroglobulin conjugate was added while vortexing, to give a thick emulsion
containing 1 mg/ml of the conjugate. Three young adult male New Zealand White rabbits
each received a total of 1 ml of this emulsion subcutaneously at 6-8 different sites in the
back. Thereafter, booster injections with the same dose of conjugate were given at
monthly intervals in the same manner, except that incomplete Freunds adjuvant was used
instead of complete Freunds adjuvant. Test bleeds were taken 7-10 days after each boost
and after serum separation, the antibody titer was tested by radioimmunoassay. Briefly, to
test total binding, [tyrosyl-3,5-3H(N)]-leucine enkephalin (3H-LE, 1 nM) was incubated


58
BLE of 4 pmol/ml would be 95 % bound by the chosen antibody concentration. Therefore,
this concentration of tracer was chosen for use in all further experiments.
Figure 2.15. Plots of fluorescence versus amount of BLE added using 200 fmol/ml of
FITC-avidin (A) and 500 fmol/ml of FITC-avidin (B)
Although one would expect that about 4 times as much BLE as FITC-avidin
would produce maximum fluorescence enhancement of FITC-avidin since it is known that
4 moles of biotin will bind to 1 mole of avidin, Figure 2.15A shows that a maximum
fluorescence reading is obtained when 4 pmol/ml of BLE is added to 200 fmol/ml of


56
exposed as an epitope to produce an immune response in the host animal, resulting in the
production of antibodies towards this end of the peptide. Since LE-Lys6-B and LE-Lys6-
BX are modified at the C-terminal end, one would naturally not expect them to bind to an
antibody directed towards the C-terminal end ofLE.
The formation of a sandwich between antibody, biotinylated LE derivative and
avidin was not achieved using either the antibody produced in this laboratory or the
commercial antiserum and any of the biotinylated LE derivatives synthesized. Since
sandwich formation is a requirement for the successful development of the ELISA
proposed in this study, efforts in this direction were abandoned at this point.
A lack of sandwich formation was observed between the antibody produced in this
laboratory, BLE and avidin, and therefore, it was concluded that this combination of
reagents was suitable for the development of the homogeneous fluorescence immunoassay
proposed in this study.
Development of the Proposed Homogeneous Fluorescence Immunoassay
Determination of excitation and emission maxima for fluorescence-labeled avidin
The excitation maximum for FITC-avidin was determined to be 482 nm.
Figure 2.14 below shows the emission scans for FITC-avidin and FITC-avidin with BLE.
Maximum emission was observed at 517 nm and fluorescence was seen to increase by a
factor of 4 when FITC-avidin interacted with BLE. Therefore, Xexc 482 nm and Xem 517
nm were chosen for future fluorescence readings.


31
overnight at 4C with various dilutions of the antiserum (1/10, 1/100, 1/1,000, 1/10,000)
in 0.1 M sodium phosphate buffer pH 7. To test non-specific binding, ?H-LE (1 nM) was
incubated overnight at 4C with LE (500 nM) and various dilutions of the antiserum
(1/10, 1/100, 1/1,000, 1/10,000) in 0.1 M sodium phosphate buffer pH 7. An ice-cold
suspension containing 1.5 % w/v activated charcoal and 0.15 % w/v dextran in water was
then added to each sample and after a 5 minute incubation at 4 C, the samples were
centrifuged at 12,000 g for 3 minutes. An aliquot of the supernatant was then removed,
Cytoscint scintillation cocktail (4 ml) was added and the radioactivity (CPM) representing
the antibody-bound tracer (3H-LE) was determined using a Beckman LS 5,000 TD
scintillation counter (Fullerton, CA, USA). After the third boost, the antibody titer of one
of the rabbits was deemed sufficient and therefore, this rabbit was exsanguinated, the
serum was separated from whole blood by centrifugation at 5,000g for 10 minutes (Dynac
II Centrifuge, Clay Adams, Sparks, MD, USA) and stored in aliquots at -80C.
A 10 ml aliquot of the anti-LE antiserum was purified by the Hybridoma Core of
the Interdisciplinary Center Biotechnology Research, University of Florida, Gainesville,
FL. Briefly, a column was filled with 10 ml of Prot G-Gammabind Ultra matrix (Pharmacia
LKB Biotechnology Inc., Piscataway, NJ, USA) and washed with 35 ml of elution buffer
(0.1 M glycine pH 3.0). The column was then washed extensively with 200-300 ml of
binding buffer (0.1 M phosphate buffered saline pH 7, 0.01% sodium azide). The
antiserum was diluted two-fold with binding buffer, loaded onto the column and washed
with binding buffer until the optical density of the eluent reached zero at X28o- Elution
buffer was then added to the column, the eluent was collected in 1 ml fractions and


142
McDermott, J. R., A. I. Smith, J. A. Biggins, M. C. Al-Noaemi and J. A. Edwardson
(1981). Journal of Chromatography 222: 371.
McQueen, D. S. (1983). British Medical Bulletin 39: 77.
Merchenthaler, I. (1993). Endocrinology 133: 2645.
Mifune, M., D. K. Krehbiel, J. F. Stobaugh and C. M. Riley (1989). Journal of
Chromatography 496: 55.
Mitsui, A., H. Nohta and Y. Ohkura (1985). Journal of Chromatography 344: 61.
Monger, L. S. and C. J. OllifT (1992). Journal of Chromatography 577: 239.
Mosberg, H. I., R. Hurst, V. J. Hruby, K. Gee, H. I. Yamamura, J. J. Galligan and T. F.
Burks (1983). Proceedings of the National Academy of Sciences 80: 5871.
Mousa, S. and D. Couri (1983). Journal of Chromatography 267: 191.
Muck, W. M. and J. D. Henion (1989). Journal of Chromatography 495: 41.
Murgo, A. J., R. E. Faith and N. P. Plotnikoff (1986). Enkephalins: Mediators of Stress
Induced Immunomodulation, in Enkephalins and Endorphins, Stress and the Immune
System (pg. 221) A. J. Murgo, R. E. Faith, N. P. Plotnikoff, & R. A. Good (Eds.). New
York: Plenum Press.
Nakano, M., M. Olmo, M. Kai and Y. Ohkura (1987). Journal of Chromatography 411:
305.
Nohta, H., A. Mitsui and Y. Ohkura (1984). Analytica Chimica Acta 165: 171.
Olson, G. A., R. D. Olson and A. J. Kastin (1991). Peptides 13: 1247.
Pasternak, G. W. (1982). Life Sciences 31: 1301.
Pleuvry, B. J. (1991). British Journal of Anaesthesia 66: 370.
Plotnikoff, N. P., A. J. Murgo, G. C. Miller, C. N. Corder and R. E. Faith (1985). Federal
Proceedings 44: 118.
Porstmann, T. and S. T. Kiessig (1992). Journal of Immunological Methods 150: 5.
Rapaka, R. S. (1986). Life Sciences 39: 1825.
Rapaka, R. S. and F. Porreca (1991). Pharmaceutical Research 8: 1.


59
FITC-avidin. The results of the HABA displacement experiments (Figure 2.11) indicated
that BLE does have a lower affinity than biotin for avidin and in addition, since here FITC-
avidin is being used, one might expect BLE to exhibit a lower affinity for FITC-avidin than
biotin exhibits for avidin. In the plot shown in Figure 2.14A, the working scale is rather
narrow and the curve obtained is not smooth as considerable noise interfered with the
taking of the readings. Therefore, although a maximum fluorescence reading using 500
fmol/ml of FITC-avidin is seen at 10 pmol/ml rather than 4 pmol/ml of BLE (Figure
2.15B), this concentration of FITC-avidin was chosen for use in this assay since less noise
is observed in the readings and the working scale is more practical.
Homogeneous fluorescence immunoassay
One of the first calibration curves obtained for LE using the complete
homogeneous fluorescence immunoassay is shown in Figure 2.16. Here, as expected, an
increase in fluorescence is seen with an increase of LE present in the sample. However,
using the LE concentrations shown here, a plateau at the maximum theoretical
fluorescence (total tracer value, no antibody present) was not reached. Therefore, another
calibration curve was constructed using higher concentrations of LE in the samples in an
attempt to reach this plateau (Figure 2.16).
Figure 2.17 shows a calibration curve for LE using the complete homogeneous
fluorescence immunoassay where fluorescence readings above the maximum theoretical
fluorescence were obtained at higher concentrations of LE (1*1 O'8 moles/ml and above).
In order to determine whether this unexpected effect could be attributed to either the


72
injector (Cotati, CA, USA) fitted with a 100 pi loop. A Spherisorb ODS2 5 pm 150 x 4.6
mm analytical column (Keystone Scientific Inc., Bellefonte, PA, USA) was used with a
mobile phase of 20% acetonitrile (v/v) in monosodium phosphate buffer (100 mM, pH 5)
containing 200 mg/1 of sodium dodecyl sulfate. The mobile phase was freshly prepared,
filtered through a 0.2 pm membrane filter and degassed under vacuum with sonication
daily, prior to use. Mobile phase flow rate was set at 1 ml/min. A Chromatopac C-R3A
integrator (Shimadzu Corporation, Kyoto, Japan) was used to record the output from the
control module.
Figure 3.4. Chromatographic configuration of electrochemical detection system.
Electrochemical Characterization
LE and monohydroxylated leucine enkephalin ([HO-Tyr]-LE), the major product
of the enzymatic reaction were characterized electrochemically using the HPLC-ED


38
IC50 value). When high enough concentrations of competitor could not be achieved to
determine non-specific binding, the non-specific binding value obtained from the LE curve
was used to fit the curve for the other competitors.
Development of the Proposed Homogeneous Fluorescence Immunoassay
Determination of excitation and emission maxima for fluorescence-labeled avidin
Fluorescein isothiocyanate avidin (FITC-avidin) was used as the detector
molecule in this assay. To determine the excitation and emission maxima, 2 ml of a
solution of FITC-avidin (7.5 pmol/ml) were dispensed into a quartz fluorescence cuvette
and scans of the excitation and emission spectra were carried out using a Perkin Elmer
LS-3B fluorescence spectrophotometer (Norwalk, CT, USA). The scans were then
repeated upon addition of BLE (70 pmol/ml) to determine the fluorescence enhancement
produced on interaction of these two reagents.
Determination of reagent concentrations
The results of the binding experiments described above indicated that the antibody
made in this laboratory in combination with the N-terminal biotinylated derivative without
the spacer arm (BLE) should be used in the development of the proposed homogeneous
fluorescence immunoassay for LE (see Results and Discussion). One of the first steps in
the development of this assay lay in the determination of the concentrations of antibody,
BLE and FITC-avidin to be used.
The decision as to the concentration of antibody to be used in the final assay
(3.78* 10'9 M, 1/50 dilution of purified antibody) was made based on practicality since it
was anticipated that a large amount of antibody would be required for the development of


9
Opioid peptides are thought to depress the responsiveness of the chemosensors to
carbon dioxide and may therefore play a physiological role in the control of respiration
[McQueen 1983], This applies in particular to neonates and to adults in stressful situations
[Pleuvry 1991],
Opioid peptides may also be involved in blood pressure regulation as they are
present in nerve fibers in areas of the brain stem responsible for the regulation of blood
pressure and the secretion of vasopressin. Biochemical evidence suggests that opioid
peptides interact with neurohormones to regulate blood pressure. Opioid peptides have
been implicated in the dramatic changes in blood pressure which occur during sleep and in
hypotension due to various states of shock [Rubin 1984], It has also been suggested that
endogenous opioid peptides may be involved in the pathogenesis of hypertension [Szilagyi
1989],
The presence of opioid peptides within limbic structures suggests their
involvement in the regulation of mood and behavior. Endogenous opioid peptides are also
known to interact with the central catecholamines implicated in psychiatric disease.
However, the results of studies carried out to determine the role of opioid peptides in
psychiatric disease have been contradictory [Clement-Jones and Besser 1983, Koob and
Bloom 1983], and therefore no conclusions can be drawn at present.
The very high concentrations of opioid peptides present in the hypothalamus
suggests a role for these substances in neuroendocrine regulation. Opioid peptides may
control the secretion of anterior pituitary hormones by modifying the release of
hypothalamic anterior pituitary regulating substances. This may be the mechanism by


123
immunoassay when increased amounts of LE were present in the sample, we observed that
at higher concentrations of LE (1*1 O'8 moles/ml and above), the analyte interacted directly
with FITC-avidin to produce an increase in fluorescence in the absence of BLE. [des-
Tyr']-LE and pentaglycine at the same concentrations did not produce an increase in
fluorescence in the presence of FITC-avidin, suggesting that the effect observed in the
homogeneous immunoassay at higher concentrations of LE was due to an interaction
between the tyrosine moiety of LE and the fluorescein groups of FITC-avidin. The use of
an antibody with higher affinity for LE may allow the successful development of this type
of homogenous fluorescence immunoassay for LE since less LE could be used in the assay
and therefore the direct interaction observed at high concentrations between LE and
FITC-avidin would be avoided. The use of laser-induced fluorescence may also contribute
to the succesfiil development of this type of assay since less FITC-avidin would be
required to produce a measurable signal on interaction with smaller quantities of BLE. As
a consequence, less LE would be required to displace sufficient BLE to produce a signal
on interaction with FITC-avidin and therefore, once again, high concentrations of LE
which interfere with the success of the assay would be avoided.
Two HPLC approaches for the determination opioid peptides were developed for
this dissertation, one with electrochemical detection and the other with fluorescence
detection. Both of these approaches involved the specific enzymatic derivatization of the
Tyr1 moiety of the model peptide LE by mushroom tyrosinase. This enzymatic
derivatization selectively introduced an hydroxy group to the Tyr1 moiety of the peptide,
thus presenting several analytical advantages. Firstly, the hydroxylation of LE rendered it


96
are produced. An incubation time of 60 minutes was chosen for the enzymatic
derivatization in our HPLC-FL assay as at this time point, the peak corresponding to LE
had disappeared and the peak corresponding to our enzymatically derived species of
interest ([HO-Tyr]-LE) was seen to plateau. This time course was repeated several times
and the same trend was observed each time.
Recovery of Leucine Enkephalin from Cerebrospinal Fluid
Recovery of LE from CSF was estimated at >90% when the counts per minute in
the methanolic eluent of the extraction column following the extraction procedure was
compared to the control sample.
Calibration Curves
For the fluorogenic reaction, the concentration of reagents used was the same as
those used previously for fluorogenic derivatization of tyrosine-containing peptides
following enzymatic hydroxylation by mushroom tyrosinase [Tellier et al. 1991], Here, as
before, the concentrations of DPE and KC1 used were taken directly from corresponding
catechol assays [Mitsui et al. 1985, Nohta et al. 1984] but the concentration of potassium
ferricyanide was increased to ensure the formation of the quinone from the hydroxylated
LE derivative despite the presence of high concentrations of ascorbic acid. Acetonitrile is
included in the reaction mixture to facilitate the condensation reaction by reducing the
thermodynamic activity of water.


94
of LE to its hydroxylated derivative within 60 minutes. A 50 mM concentration of
ascorbic acid was found to be adequate to prevent or reverse the tyrosinase-induced
formation of o-quinones in our enzymatic reaction mixture, thereby inhibiting the
subsequent polymerization of the reaction products.
Figure 4.3. Chromatographs of a control run (A) showing LE eluting after 9.5 minutes,
distinct from ascorbic acid eluting in the solvent front after 1.4 minutes and
incubation solution (B) showing the emergence of two new peaks with
retention times of 5.7 and 7.3 minutes.


64
fluorescence immunoassay were probably due to an interaction between the tyrosine
moiety of LE and the fluorescein groups of FITC-avidin.
The homogenous fluorescence immunoassay evaluated here operates as intended in
the concentration range between 1*1 O'9 moles/ml and 1*1 O'8 moles/ml of LE since at these
concentrations, sufficient tracer (BLE) is displaced from the antibody by LE to produce
detectable fluorescence enhancement on interaction with the detector molecule (FITC-
avidin), but the concentration of LE is not high enough to produce a direct interaction
with FITC-avidin in the absence of BLE. Using the reagents tested, an homogeneous
fluorescence immunoassay for LE operational over a wide concentration range could not
be developed since at higher concentrations (1*1 O'8 moles/ml and above), LE interacted
directly with FITC-avidin to produce fluorescence enhancement, thus interfering with the
signal produced through the interaction of BLE with FITC-avidin. In this assay, the use of
an antibody with a higher affinity for LE would present several advantages. Firstly, less
antibody would be required to ensure maximal binding of the BLE used in the assay and
consequent low background fluorescence. Secondly, less LE would be required to displace
sufficient BLE from the antibody to produce a measurable signal on interaction with
FITC-avidin and therefore, LE concentrations in the assay displacing the maximum
amount of BLE may not reach the concentrations which were observed to produce a
direct interaction with FITC-avidin and as a result interfered with the success of the assay.
Thirdly, since lower concentrations of LE would be required to displace BLE from the
antibody and produce a measurable signal, the sensitivity of the assay would be increased.


98
in both buffer samples and spiked CSF samples. This limit of detection corresponds to 12
pmol of LE per ml of CSF. Figure 4.7 shows a representative chromatograph for LE in
CSF using our HPLC-FL method.
Figure 4.5. Calibration curve for LE in buffer using HPLC-FL method with non
concentrated reagents.
Figure 4.6. Calibration curve for LE in CSF using HPLC-FL method and non
concentrated reagents.


60
antibody, the tracer or the analyte, the assay was repeated omitting each of these reagents
in turn.
100
80
60
c
3
E
40
20
Total tracer
FI blank
o 1 1
IE-11 IE-10 IE-9 IE-8
LE(moles/niI)
Figure 2.16. Calibration curve for LE using homogeneous fluorescence immunoassay.
180
160
140
120
Q
8 loo
80
60
40
20
0


Total tracer




FI blank



-+-
-+-
IE-11 IE-10 IE-09 IE-08
LF moles/ml
i
IE-07
IE-06
Figure 2.17. Calibration curve for LE using homogeneous fluorescence immunoassay
showing readings exceeding maximum theoretical fluorescence.


54
Table 2.6. IC50 values obtained for LE, BLE, BLE-avidin, BXLE and BXLE-avidin using
1/1,000 dilution of antibody produced in this laboratory. ^050 value for BLE-
avidin was estimated by fixing non-specific binding to 0% total binding and N to
1.
Competitor
IC50 (mol/assay)
Mean
Standard deviation
LE
4.95* 10'13
3.70* 10'13
3.78* 10'13
3.44* 1013
2.79* 10'13
4.19* 10"13
3.8* 10'13
7.3* 1014
BLE
6.22*10'13
1.21*1 O12
1.38* 10'12
1.1*1012
4.0* 1013
BLE-avidin
*3.73*10
BXLE
3.53 10"14
4.97*10'15
2.0* 1014
BXLE-avidin
1.70*10
1.11*10
1.4*10
As with the commercial antiserum, the N-terminal biotinylated derivatives were
seen to retain affinity for the antibody produced in this laboratory. However, a shift in
affinity by almost three orders of magnitude is seen when complexes are formed between
BXLE and avidin (Table 2.6), and pre-formed BLE-avidin complexes showed binding to
this antibody at only the highest concentration tested (Figure 2.13).
Again, as with the commercial antiserum, BXLE showed a higher affinity than LE
to the antibody produced in this laboratory. Here too, as before, it was postulated that
since LE was conjugated to a porcine thyroglobulin to render it immunogenic for antibody
production, this biotinylated LE derivative including a spacer arm may have resembled the


102
fluorogenic reagents. Although a second attempt to improve the analytical approach by
decreasing the tyrosinase reaction time resulted in a lower limit of detection in buffer
samples, when this modification was attempted with spiked CSF samples, the linearity of
the calibration curve was severely compromised. It is postulated that matrix components
in the spiked CSF extracts may have altered the efficiency of the tyrosinase reaction, thus
affecting the reproducibility of the yield of hydroxylated LE derivative after this short
incubation time.
Table 4.2. Comparison table of HPLC-FL analytical methods for opioid peptides.
Reference
Sensitivity
Analvte
Matrix
Comments
This study
500 fmol/inj
12 pmol/ml CSF
LE
Buffer
CSF
tyrosine specific
Ishida et al.
1986
7 pmol/inj
enkephalin
water
tyrosine specific
Nakano et al.
1987
270 fmol/inj
enkephalin
water
tyrosine specific
Kai et al. 1988
500 fmol/inj
enkephalin
rat brain
tyrosine specific
Zhang et al.
1991
0.33-1.21 pmol/inj
opioid peptides
rat brain
tyrosine specific
van den Beld et
al. 1990
80 fmol/inj
(3-endorphin
plasma
laser-induced
fluorescence
Mifune et al.
1989
100 fmol/inj
enkephalin
rat brain
column-switching
Dave et al.
1992
36 fmol/inj
LE
water
microbore FtPLC,
laser-induced
fluorescence


113
respectively) directly derived from Figure 5.3 were assigned. Two intense internal
fragments, [a4*y2]i at m/z 120 and [a4*y3]2 at m/z 177, were also observed. These were the
products of the dissociation of the protonated molecule (m/z 556). CID of the sodiated
enkephalin was limited, only the loss of the C-terminal leucine moiety was prominent (m/z
465).
556
Figure 5.4. Electrospray ionization mass spectra of LE. (A) Repeller at 18 V; (B) Repeller
at 50 V.