Citation
Molecular mechanisms of host specific virulence and avirulence caused by cotton blight and citrus canker pathogens

Material Information

Title:
Molecular mechanisms of host specific virulence and avirulence caused by cotton blight and citrus canker pathogens
Creator:
Yang, Yinong, 1962-
Publication Date:
Language:
English
Physical Description:
v, 129 leaves : ill., photos ; 29 cm.

Subjects

Subjects / Keywords:
Cells ( jstor )
Cotton ( jstor )
Host range ( jstor )
Pathogens ( jstor )
Plant cells ( jstor )
Plasmids ( jstor )
Proteins ( jstor )
Signals ( jstor )
Symptomatology ( jstor )
Xanthomonas ( jstor )
City of Madison ( local )
Genre:
bibliography ( marcgt )
theses ( marcgt )
non-fiction ( marcgt )

Notes

Thesis:
Thesis (Ph. D.)--University of Florida, 1994.
Bibliography:
Includes bibliographical references (leaves 115-128).
General Note:
Typescript.
General Note:
Vita.
Statement of Responsibility:
by Yinong Yang.

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:
021783152 ( ALEPH )
AKL5729 ( NOTIS )
33032693 ( OCLC )

Downloads

This item has the following downloads:


Full Text








MOLECULAR MECHANISMS OF HOST SPECIFIC VIRULENCE
AND AVIRULENCE CAUSED BY COTTON BLIGHT
AND CITRUS CANKER PATHOGENS







By

YINONG YANG


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



UNIVERSITY OF FLORIDA


1994











ACKNOWLEDGMENTS


I wish to express my appreciation and gratitude to Dr. Dean W. Gabriel,

chairman of the supervisory committee, for his time, guidance and support during the

course of this study. Appreciation is also extended to Drs. D. R. Pring, H. C. Kistler,

L. C. Hannah, and R. A. Jensen for serving on the supervisory committee and critical

reading of this manuscript. Special thanks go to Dr. W. B. Gurley for reading my

dissertation and participating in my final examination.

I wish to thank Drs. Robert De Feyter, Mark T. Kingsley and Mr. Gary Marlow

for stimulating scientific discussions and technical assistance. I gratefully acknowledge

the help of Dr. Michael Paddy in fluorescence microscopy and Dr. Suzy Cocciolone in

microprojectile bombardment. I also greatly appreciate the technical assistance of

Qiaoping Yuan, Ruhui Li and Blanca Garagorry.

I would like to thank my parents for their understanding and encouragement.

Finally, I must thank my wife, Qin Wang, and daughter, Catherine, for their

companionship and strong support.












TABLE OF CONTENTS


ACKNOWLEDGMENTS ................................... ii


ABSTRACT

CHAPTERS


1. INTRODUCTION ..............................


2. HOST-SPECIFIC SYMPTOMS AND INCREASED
RELEASE OF XANTHOMONAS CITRI AND
X. CAMPESTRIS PV. MALVACEARUM
FROM LEAVES ARE DETERMINED BY
THE 102 BP TANDEM REPEATS OF
pthA AND avrb6, RESPECTIVELY .....


. 9


3. INACTIVATION OF MULTIPLE AVIRULENCE
GENES IN XANTHOMONAS CAMPESTRIS
PV. MALVACEARUM CREATED A
NONPATHOGENIC ENDOPHYTE
OF COTTON ........................... 37

4. INTRAGENIC RECOMBINATION OF A SINGLE
PLANT PATHOGEN GENE PROVIDES A
MECHANISM FOR THE EVOLUTION
OF NEW HOST SPECIFICITIES ............... 61

5. PLANT NUCLEAR TARGETING SIGNALS ENCODED
BY A FAMILY OF XANTHOMONAS
AVIRULENCE/PATHOGENICITY GENES ........ 87

6. SUMMARY AND CONCLUSIONS .................. 111

LITERATURE CITED ................................... 115

BIOGRAPHICAL SKETCH ................................... 129











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

MOLECULAR MECHANISMS OF HOST-SPECIFIC VIRULENCE
AND AVIRULENCE CAUSED BY COTTON BLIGHT
AND CITRUS CANKER PATHOGENS

By

Yinong Yang

August 1994


Chairman: Dean W. Gabriel
Major Department: Plant Molecular and Cellular Biology

Several members of a Xanthomonas avirulence (avr) gene family were shown to

play an important role in determining host-specific pathogenicity (pth) of cotton blight

and citrus canker pathogens. X. campestris pv. malvacearum strain XcmH1005, the

cotton blight pathogen, carries twelve members of what is now known as an avr/pth gene

family. Six plasmid-bome members were previously shown to confer avirulence on

cotton in a gene-for-genes manner. Four out of six newly cloned chromosomal members

also conferred avirulence on cotton. To investigate possible pleiotropic functions of these

avr genes in XcmH1005, marker exchange mutagenesis and complementation analyses

were carried out. Gene avrb6 was found to be a major pth gene of XcmH1005 and was

required for severe pathogenic symptoms on cotton. Other avr genes in XcmH1005 also

had an additive effect on pathogenicity. These avr genes did not contribute to bacterial

iv






growth in plant, but were essential for the induction of host-specific symptoms and

release of the pathogen onto the leaf surface, strongly indicating their role in pathogen

dispersal. Another member of the same gene family, pthA of X. citri, was previously

shown to be essential for symptoms of citrus canker disease and release of the pathogen

to the leaf surface. By constructing chimeric genes among pthA and six other avr/pth

genes, both avirulence and pathogenic specificities were found to be determined by the

102 bp tandem repeats present in the central region of all members of the gene family.

Furthermore, intragenic recombination was demonstrated to provide a mechanism for the

generation of new avr/pth genes with novel host specificities. Sequence analysis revealed

that pthA and avrb6 encoded proteins containing three nuclear localization sequences in

the C-terminal region. The predicted PthA and Avrb6 proteins were constitutively

expressed in Xanthomonas as detected by Western analysis using anti-PthA antibody.

When transiently expressed in plant cells, p-glucuronidase fusion proteins containing the

C-terminal region of PthA or Avrb6 were observed to localize specifically in the nucleus.

A working hypothesis is proposed that Avr/Pth proteins enter the plant cell nucleus and

directly affect gene expression, leading to the induction of host-specific symptoms.











CHAPTER 1
INTRODUCTION


Disease resistance of plants and virulence of microbial pathogens are governed

by their combined genotypes. During the process of reciprocal evolution, many plant

and microbial genes have coevolved to actively engage in pathogenic interactions between

host and pathogen, thus determine the final outcome of plant diseases. Successful

infection by microbial pathogens may require surface attachment, degradation of host

chemical and physical barriers, production of toxins, and inactivation of plant defenses

(Lamb et al. 1989). Overall, up to 100 genes may be required for bacterial pathogenesis

(Panopoulos and Peet 1985, Daniels et al. 1988), including those involved in bacterial

growth in plant, induction of pathogenic symptoms, host range and avirulence (Gabriel

1986). On the other hand, host plant resistance to infection involves preformed barriers,

inactivation of microbial toxins, recognition of pathogen signals and induction of active

plant defenses. In incompatible plant-bacteria interaction, induced resistance is usually

associated with a hypersensitive response (HR), characterized by the rapid, localized

death of plant cells at the site of infection. This process requires activation of plant

resistance genes involving signal perception and transduction and plant defense genes

involving biosynthesis of pathogenesis-related proteins, phytoalexins and other antibiotics

(Dixon and Harrison 1990, Dixon and Lamb 1990).






2

Bacterial plant pathogens require virulence genes to infect and to colonize plants.

A large group of genes, called hypersensitive response and pathogenicity (hrp) genes,

have been found generally conserved in all phytopathogenic bacteria examined (e. g.,

Erwinia, Pseudomonas and Xanthomonas) (Willis et al. 1991). Whatever the bacterial

pathogens studied, typical hrp- mutants are completely nonpathogenic and elicit neither

pathogenic symptoms on host plants nor hypersensitive response on nonhost plants. Most

hrp genes are clustered on 20-30 kb DNA fragments and are conserved at the genus or

family level. High levels of DNA homology are found among many hrp genes from

different genera of phytobacteria (Boucher et al. 1992). Several biological functions are

encoded by different hrp genes. Some hrp genes (e.g., hrpS of P. syringae pv.

phaseolicola) are clearly regulatory, responding to plant signals to induce transcription

of hrp operon in a manner that may be analogous to the two-component signaling systems

(Rahme et al. 1991). Other hrp genes are involved in the synthesis and export of

pathogenicity factors that also induce an HR on nonhost plants. DNA sequence analysis

reveals that several hrp genes of Pseudomonas and Xanthomonas are remarkably similar

to key virulence genes of animal bacteria such as Yersinia and Shigella, which are

involved in export of proteinaceous pathogenicity factors (Gough et al. 1992, 1993;

Fenselau et al. 1992). Some hrp genes of Erwinia and Pseudomonas were shown to

encode proteinaceous HR elicitors that are secreted via Hrp-dependent pathway (Wei et

al. 1992; He et al. 1993). Since HR on host plants (race specificity) conferred by

avirulence (avr) genes is dependent on hrp function, it was proposed that Hrp proteins






3
may also involve export of avirulence elicitors (Fenselau et al. 1992). Therefore, hrp

genes may control or condition the functions encoded by virulence and avirulence genes.

Bacteria belonging to the genus Xanthomonas are plant-associated and infect 392

monocot and dicot species in 68 families (Leyns et al. 1984). Although the genus as a

whole has wide host range, individual strains have host ranges limited to only certain

plant species. This phenomenon of host specialization indicates that establishment of

Xanthomonas infection involves host-specific recognition in addition to general virulence

function of pathogen conferred by hrp genes. Host-specific virulence (hsv) genes

encoding positive function have been shown to be a major determinant for host range and

pathovar status of Xanthomonas (Gabriel et al. 1993). These genes are required for

bacterial pathogens to grow and/or to induce pathogenic symptoms on specific host

plants, but not in other plant species. Several host-specific virulence genes have been

isolated from X. campestris pv. citrumelo (Kingsley et al. 1993), X. c. pv. translucens

(Waney et al. 1992) and X. citri (Swarup et al. 1991). For example, opsX of X. c. pv.

citrumelo is involved in biosynthesis of lipopolysaccharide and extracellular

polysaccharide and is required for virulence on citrus, but not on bean (Kingsley et al.

1993). These hsv genes are analogous to host-specific nodulation (hsn) genes of

Rhizobium. The hsn genes of Rhizobium is involved in host-specific modification of

lipooligosaccharide Nod factors and function only for nodulation on specific hosts (Fisher

and Long 1992; Denarie et al. 1992).

Besides hsv genes, avr genes have been shown to play a role in determining host

specificity at the species level (Keen 1990; Heath 1991). Single cloned avr genes from






4
a pathogen of one host species can cause an otherwise virulent pathogen of another host

species to become avirulent on its own host (Kobayashi et al. 1989; Whalen et al. 1988).

Therefore, host species specificity is conditioned by both host-specific virulence and

avirulence genes with the former as the major determinant.

Although avr genes may condition host range at the species level in some cases,

they act as negative factors to confer avirulence and to limit host range only if the host

plants have specific resistance (R) genes. The interaction of pathogens with avr genes

and host plants with specific R genes results in the plant defense response, often known

as hypersensitive response. This genetic requirement for specific plant R genes and

specific microbial avr genes is called a gene-for-gene interaction. The gene-for-gene

interactions were first demonstrated by Flor (1942, 1946, 1947) using flax and flax rust

and have since been observed in many plant-parasite interactions between fungi, bacteria,

virus, nematodes and their respective hosts (Thompson and Burdon 1992).

During the past decade, many avr genes have been isolated from different

bacterial and fungal plant pathogens (Gabriel et al. 1993; Keen 1990). In some

instances, avr genes encode proteinaceous elicitors that directly induce the HR on

resistance plants. For example, avr4 and avr9 of the fungal pathogen Cladosporium

fulvum encode race-specific peptides which are excreted into the apoplast of infected

leaves and induce the HR on tomato cultivars that carry resistance genes Cf4 and Cf9,

respectively (Joosten et al. 1994; van Kan et al. 1991). In another case, avrD of

Pseudomonas syringae pv. tomato appears to encode a enzyme involved in synthesis of

a race-specific glycolipid elicitor (Keen et al. 1990; Kobayashi et al. 1990, Midland et






5
al. 1993). In most cases, however, the biochemical identities of avr gene products are

unknown and the sequence analyses of avr genes are uninformative as to how they might

interact with R gene products to trigger the HR and specify gene-for-gene interaction.

Several models (including elicitor/receptor model, dimer model and ion channel

defense model) have been proposed to explain the molecular and biochemical bases of

gene-for-gene recognition between plant and pathogen (Gabriel and Rolfe 1990). But

very little experimental evidence is available to prove any of the hypotheses. Recently,

the first plant R gene, Pto, has been isolated from tomato by map-based positional

cloning (Martin et al. 1993). This gene encodes a serine/threonine protein kinase that

is homologous to the receptor protein kinase SRK6 involving Brassica pollen-stigma

recognition and to the Raf protein kinase involving the mammalian Ras signaling

pathway. The Ras pathway is the best known circuit of signal transduction found in

yeast, insect, as well as mammalian systems (Culotta and Koshland 1993). The Ras

pathway begins with extracellular signals (e. g. growth factors or cytokines) that interact

with receptor protein kinases, which in turn trigger phosphorylation cascades and finally

expression of targeted genes (Crews and Erikson 1993).

Although Ras-like signaling pathway has not been demonstrated in plants,

homologues of mammalian signal transduction components were found in many plants.

Those include homologues of G-protein subunits (e. g. auxin-regulated ArcA of tobacco,

Ishida et al. 1993), receptor protein kinases (e.g. pollen-specific SRK of Brassica,

Goring and Rothstein 1992), Ras-related proteins (e.g. Rgpl of rice, Kamada et al.

1992), Raf (e.g., Pto of tomato, Martin et al. 1993; and ethylene-regulated CTR1 of






6

Arabidoposis, Kieber et al. 1993) and mitogen-activated protein kinase (e.g., MsERK1

of alfalfa, Duerr et al. 1993). It is possible that virulence and avirulence gene products

may activate the similar signal transduction pathway in plants, leading to disease

symptoms or hypersensitive response. For example, transcriptional activation of plant

defense genes is modulated by phosphorylation (Felix et al. 1991; Yu et al. 1993) and

can be blocked by inhibitors of mammalian protein kinases (Raz and Fluhr 1993). It was

proposed that recognition of avr signals by R gene products would trigger a

phosphorylation cascade, leading to plant defense responses (Lamb 1994).

In this study, two economically important, world-wide diseases caused by

Xanthomonas have been used as model systems. Cotton bacterial blight is caused by X.

campestris pv. malvacearum; symptoms include angular, watersoaked spots on leaves,

black arm on stems and bollrot (Verma 1986). Citrus canker is caused by X. citri;

symptoms include erumpent, corky hyperplasia on leaves, stems and fruit (Schoulties et

al. 1987; Stall and Civerolo 1991). Recently, members of a Xanthomonas avirulence

(avr) and pathogenicity (pth) gene family were found to play an important role in

determining host-specific virulence and avirulence caused by these two pathogens (De

Feyter and Gabriel 1991a; De Feyter et al. 1993; Swarup et al. 1991, 1992). For

example, a pathogenicity gene, pthA, is required for X. citri to elicit hyperplastic lesions

on citrus (Swarup et al 1991). It also functions as an avr gene to confer the ability to

elicit an HR on cotton and bean (Swarup et al. 1992). Members of this gene family also

include avrB4, avrb6, avrb7, avrBIn, avrBlO1 and avrB102 of X. campestris pv.

malvacearum (De Feyter and Gabriel 1991a; De Feyter et al. 1993), avrBs3, avrBs3-2






7
and avrBsP of X. campestris pv. vesicatoria (Bonas et al. 1989, 1993; Canteros et al.

1991), and avrxa5, avrXa7 and avrXalO of X. oryzae (Hopkins et al. 1992). The most

conspicuous feature of this gene family is the nearly identical, tandemly arranged, 102bp

repeats in the central portion of genes (Fig. 1-1). Deletion analyses of avrBs3 have

shown that the 102bp repeated motifs determines the avirulence specificity of the gene

(Herbers et al. 1992). However, many questions remain regarding to their pleiotropic

functions, specificity, evolution, and biochemical mechanism of gene-for-gene

recognition.

The primary objectives of this work were to determine (1) why avirulence genes

are present in the cotton blight pathogen and whether they confer any pleiotropic

functions, (2) what determines avirulence and pathogenic specificity in cotton blight and

citrus canker pathogens, (3) how the specificity is created and evolved in this avr/pth

gene family, and (4) how the signals encoded by the avr/pth genes are perceived and

transduced in plant cells.























Gene Structure


# 102bp
Repeats


B P St H BS
I I I I I II

B P St H BS


BP St H BS
iI I 1 I II


H BS


17.5


13.5

17.5


15.5


Pleiotropic
Function


cankering


watersoaking


Figure 1-1. A Xanthomonas avr/pth gene family. Listed are four representive members:
pthA of X. citri, avrb6 of X. campestris pv. malvacearum, avrBs3 of X. campestris pv.
vesicatoria, and avrXalO of X. oryzae pv. oryzae. The tandemly arranged blocks
represent the 102-bp repeats in the central portion of genes. Restriction enzyme cleavage
sites, relevant to this work and found in most members of the gene family, are: B,
BamHI; P, PstI; St, StuI; H, HincII; S, Sstl.


Name


pthA


avrb6

avrBs3


avrXa 10


BP St


I I
1


s











CHAPTER 2
HOST-SPECIFIC SYMPTOMS AND INCREASED RELEASE OF XANTHOMONAS
CITRI AND X. CAMPESTRIS PV. MALVACEARUM FROM LEAVES
ARE DETERMINED BY THE 102BP TANDEM REPEATS
OF pthA AND avrb6, RESPECTIVELY


Introduction



Microbial genes involved in plant-microbe interactions may be functionally

classified into four broad categories: parasitic, pathogenic, host range and avirulence

(Gabriel 1986). Genes involved in parasitism are absolutely required for growth in

plant and are widely conserved at the family or genus level. Examples include most

hrp (hypersensitive response and pathogenicity) genes of Erwinia, Pseudomonas and

Xanthomonas, and the common nod nodulationn) genes of Rhizobium (Boucher et al.

1992, Denarie et al. 1992, Willis et al. 1991). Genes involved in pathogenicity are

required for induction of symptoms. Examples include pectate lyase, polygalacturonase

and endoglucanase genes (Collmer and Keen 1986, Schell et al. 1988, Roberts et al.

1988), dsp (disease specific, Arlat and Boucher 1991), wts (watersoaking, Coplin et al.

1992) genes, phytohormone biosynthetic genes (Smidt and Kosuge 1978) and toxin

biosynthetic genes (Mitchell 1984). Genes involved in conditioning host range are host

specific and required for growth on specific hosts. Examples include hsv (host-specific

virulence), pth pathogenicityy), and hsn (host-specific nodulation) genes of Pseudomonas,






10
Xanthomonas and Rhizobiwn (Denarie et al. 1992, Kingsley et al. 1993, Ma et al. 1988,

Salch and Shaw 1988, Swarup et al. 1991, Waney et al. 1991, Gabriel et al., 1993).

Information on conservation of these genes within species, pathovars and biovars is

scarce, but these genes appear to determine biovar and pathovar status (Djordjevic et al,

1987; Gabriel et al, 1993). The fourth group are termed avirulence (avr) genes because

they negatively affect virulence. The avr genes are superimposed on basic compatibility

(Ellingboe 1976), are not highly conserved, and determine pathogenic races below the

species, biovar or pathovar levels. These four broad categories are not mutually

exclusive.

The interaction of microbes having avr genes and host plants having resistance

(R) genes can result in plant defense responses, often observed visually as a

hypersensitive reaction (HR) and characterized by the rapid necrosis of plant cells at the

site of infection and the accumulation of phytoalexins. Most avr genes do not appear to

confer selective advantage to the pathogen (Gabriel 1989, Keen and Staskawicz 1988).

Pleiotropic functions have been identified for only 3 of the 30 avr genes cloned to date

(Gabriel et al, 1993). Furthermore, the DNA sequences of the cloned avr genes have

been remarkably uninformative in terms of function (Keen, 1990). The presence of most

avr genes in plant pathogens therefore remains enigmatic.

Recently, an avr gene family has been discovered in many different

xanthomonads; members include avrBs3, avrBs3-2 and avrBsP (Bonas et al. 1989, 1993;

Canteros et al. 1991) of X. campestris pv. vesicatoria, avrB4, avrb6, avrb7, avrBIn,

avrBlOl and avrBl02 of X. campestris pv. malvacearum (De Feyter and Gabriel 1991a,






11
De Feyter et al. 1993) and avrxa5, avrXa7 and avrXalO of X. oryzae (Hopkins et al.

1992). Interestingly, this gene family includes a gene, pthA, that is required for

pathogenicity of X. citri on citrus. This gene is not known to function for avirulence in

X. citri (Swarup et al. 1992), but is required for X. citri to induce cell divisions in the

leaf mesophyll of citrus, leading to epidermal rupture and subsequent release of the

bacteria onto the leaf surface. The gene also confers the ability to induce cell divisions

on citrus to X. campestris strains from several different pathovars (Swarup et al. 1991,

1992).

De Feyter and Gabriel (1991a) observed that avrb6 and avrb7 enhanced the

watersoaking ability of several X. campestris pv. malvacearum strains on cotton, but the

role of these genes in pathogenicity was not determined. The family therefore consists

primarily of avr genes, but includes at least one, and perhaps more, host-specific

pathogenicity genes. The most conspicuous feature of this highly homologous gene

family is the presence of nearly identical, tandemly arranged, 102bp repeats in the central

region of the genes. These repeats are known to determine the gene-for-gene specificity

of avrBs3 (Herbers et al. 1992). The purpose of this study was to characterize the

watersoaking functions of avrb6 and to investigate the role of the 102bp repeats of pthA,

avrB4, avrb6, avrb7, avrBIn, avrBlO1 and avrB102 in pathogenicity and avirulence.








Materials and Methods



Bacterial strains, plasmids. and culture media. The bacterial strains and plasmids

used in this study are listed in Table 2-1. Strains of Escherichia coli were grown in

Luria-Bertani (LB) medium (Sambrook et al. 1989) at 37*C. Strains of Xanthomonas

were grown in PYGM (peptone-yeast extract-glycerol-MOPS) medium at 30*C (De

Feyter et al. 1990). For culture on solid medium, agar was added at 15g/L. Antibiotics

were used at the following final concentrations (mg/L): ampicillin (Ap), 25; kanamycin

(Km), 20; gentamycin (Gm), 3; spectinomycin (Spc), 50; tetracycline (Tc), 15; and

rifampicin (Rif), 75.

Recombinant DNA techniques. Total DNA isolation from Xanthomonas was as

described (Gabriel and De Feyter 1992). Plasmids were isolated from E. coli by alkaline

lysis methods (Sambrook et al. 1989). Restriction enzyme digestion, alkaline

phosphatase treatment, DNA ligation and random priming reactions were performed as

recommended by the manufacturers. Southern hybridization was performed by using

nylon membranes as described (Lazo et al. 1987). Otherwise, standard recombinant

DNA procedures were used (Sambrook et al. 1989).

Construction of chimeric genes. To construct BamHI fragment-swapped chimeric

genes among avrB4, avrb6, avrb7, avrBIn, avrBlO1, avrB102, the BamHI fragments

from these genes were cloned into an avrb7 shell (containing 5' and 3' ends of avrb7,

but deleted for its BamHI fragment, pUFR163ABam) and an avrBIn shell (containing 5'

and 3' ends of avrBIn, but deleted for its BamHI fragment, pUFR186ABam) on the









Table 2-1. Bacterial strains and plasmids used in this study.


Strain or Plasmid Relevant Characteristics Reference or Source


E. coil


DH5-


HBI01


ED8767


Xandwhownas citi


3213
B21.2


X phaseoUl


G27
G27Sp


X campesinos pv. citrumelo


3048
3048Sp


X coampestris pv. alfalfa


KX-ISp


F-, endAl. hsdI7(rk(mk), supE44,
ihi-1, recAl, gyrA, reLAl, 80dlacZ
AM15, A(acZYA-arwF)U169
supE44, hsdS20(rmk+), vcAI3,
ara-14, proA2, lacYl, galK2, rpsL20,
xyl-5, ma-I
supE44, supF8, sdS3(rk-mk+), recAS6,
galK2, galT22, metBl



ATCC49118, citrus canker type strain
pthA::Ta5-gusA, marker-exchange
mutant of 3213


ATCC49119, bean blight type strain
Spc' derivative of 027



ATCC49120, citrus leaf spot pathotype strain
Spc' derivative of 3048



Spc' derivative of KX-1, isolated from
alfalfa, causing citrus leaf spot


Gibco-BRL,Gaithersburg,MD


Boyer & Roulland-Dussoix 1969



Murray et al. 1977


Gabriel ea a. 1989
Swamp et at. 1991



Gabriel et al. 1989
Swarup et al. 1991



Gabriel et al. 1989
Swarup et al. 1991



Swarup et ai. 1991


X campesris pv. malvacearum


Natural isolate from cotton from
Oklahoma; carries six avr genes used in this
study on pXcmH, plus additional avr genes
Spontaneous Rif derivative of XcmH
avrb6::TnS-gutA, marker-exchange
mutant of XcmHIO05
avrb7::Tan-gusA, marker-exchange
mutant of XcmHI005
avrBln::TnS-gusA, marker-exchange
mutant of XcmHI005
Natural isolate from cotton from
Upper Volta, Africa
Spc' Rif derivative of XcmN



ColEl, Kn', Tra+, helper plasmid
pRK2013 derivative, npt::Tn7,Kmn, Sp',
Tra+, helper plasmid
IncW, Kin', Gmn, Mob+, acZa+, Par+,
IncW, Gmn, Apr, Mob+, lacZa+, Par+


De Feyter & Gabriel, 1991a


This study
This study

This study

This study

Gabriel et al. 1986

DeFeyter & Gabriel 1991a



Figurski & Helinski, 1979
Leong et al. 1982

DeFeyter & Gabriel 1991a
De Feyter et al. 1993


XcmH


XcmHI005
XcmHI407

XcmHl427

XcmHl431

XcmN

Xcml003


Plasmid


pRK2013
pRK2073

pUFRO42
pUFRO47








Table 2-1--continued.


Strain or Plasmid Relevant Characteristics Reference or Source


pUFRO49

pUFRO34

pUFR1 15
pUFR127
pUFR135
pUFR142
pUFRI56
pUFR1S?
pUFRl63
pUFRI63ABam
pUFR171
pUFRl72
pUFR173
pUFR174
pUFRI75
pUFRI76
pUFR17S

pUFR179

pUFRiSO
pUFR186

pUFRI86A~ain
pUFRI9O-194


pUFRI96-200


pUFR20S-209



pUFR211-215



pUFR2I7
pUFR220
pUFR227
PUFY019


pUFY020


pZit45
pGiEM7Zf(+)
pGEMI lZf(+)


RSF1010 replicon, Cmf, Smo, lacW+,
Mob' displacement vector
IncP, Tc', Mob', containing methylases
Xmal and Xmaml
7.5kb fragment containing avrB4 in pUFRO42
5kb fragment containing awrb6 in pUFRO42
lacZ::avrb6 fusion in pUFRO42
9kb fragment containing avrBOl in pUFRO47
12.9 kb fragment containing avrBIn in pUFRO42
11kb fragment containing avrB102 in pUFRO42
10kb fragment containing aw*bin pUFRO42
pUFRI63 deleted for 3.4 kb BamHI fragment
Internal BamHI fragment of avrB4, in pGem lZf(+)
Internal BamHI fragment of avrb6, in pGeml IZf(+)
Internal BamHI fragment of avrBlOI, in pGeml IZf(+)
Internal BamHI fragment of avrBIn, in pGeml lZf(+)
Internal BamHI fragment of avrBI02, in pGeml IZf(+)
Internal BamHI fragment of avrb7, in pGemllZf(+)
Internal BamHI fragment of avrBIn,
in pGemI lZf(+), Sal site filled
Internal BamHI fragment of avrb7, in pGemllZf(+),
Sall site filled
9.5 kb fragment containing avrb6 in pUFRO42
10.3 kb Bgll-EcoRI fragment containing avrBln
from pUFR156 in pUFRO47
pUFRI86 deleted for 3.7 kb BamHI fragment
BamHI-swapped chimeric genes with avrb7
5' and 3' ends plus BamHI fragment of avrB4,
avrb6, avrBIn, avrBlOl, or avrBI02 in pUFRO47
BaimHI-swapped chimeric genes with avrBIn
5' and 3' ends plus &amHI fragment of avrB4,
avrb6, avrb7, avrBlOl, or avrBI02 in pUFRO47
Saul/Hincl-swapped chimeric genes with avrBIn
5' and 3' regions plus Sal/Hincll fragment of
avrBlOl, avrB4, awrb6, awrBJ02, or avrb7
in pUFRO47
SAuI//incU-swapped chimeric genes with avrb7
5' and 3' regions plus Saul/Hinc fragment of
avrBlOI, avrB4, awvr6, avrBln, or avrBI02
in pUFRO47
pUFR156 derivative, avrMBn::Ta5-gu4sA
pUFR163 derivative, avrb7::Tn5-gusA
pUFRISO derivative, avrb6::Tn5-gusA
3.7 kb Saul/Hinc-swapped fragment with phAA
5' and 3' regions of pZit45 plus avrb6 internal
repeat region of pUFR135 in pUFRO47
3.8 kb Saul/Hincl-swapped fragment with avrb6
5' and 3' regions of pUFRI35 plus pthA internal
repeat region of pZit45 in pUFRO47
4.5kb fragment containing pthA in pUFRO47
ColEl, Ap', lacZa*
ColEl, Ap, lacZa


Swamp et at. 1991

DeFeyter & Gabriel, 199 1b

De Feyter & Gabriel 1991a
De Feyter & Gabriel 1991a
De Feyter et at. 1993
De Feyter et al. 1993
De Feyter et al. 1993
De Feyter et al. 1993
De Feyter el al. 1993
This study
De Feyter e: al. 1993
De Feyter et al. 1993
De Feyter et al. 1993
De Feyter et at. 1993
De Feyter et al. 1993
De Feyter el al. 1993
This study

This study

De Feyter et al. 1993
This study

This study
This study


This study


This study



This study



This study
This study
This study
This study


This study


Swamp et al. 1992
Promega Co., Madison, WI
Promega Co., Madison, WI






15
pUFRO47 vector. The resulting BamHI fragment-swapped chimeric genes formed

pUFR190-200. The avr genes from pXcmH and pthA from X. citri all have unique StuI

and HincI sites delimiting the 102bp repeated regions, which allow the swapping of the

internal repeated regions among these genes. To construct StuI/HinclI fragment-swapped

chimeras using the pXcmH avr genes, pUFR174 and pUFR176 were cut with Sall, blunt

ended using the Klenow fragment and religated to destroy the HindI sites on the

pGEM1 lZf(+) portions of the plasmids, forming pUFR178 and pUFR179, respectively.

These were then cut with StuI and HincH to delete the internal fragments, and used as

recipients for the Stul/HincdI internal fragments from all pXcmH avr genes on pUFR171-

176. The BamHI fragments from pUFR178 and its StuI/HinclI chimeras were recloned

into pUFR186ABam, forming pUFR205-209. The BamHI fragments from pUFR179 and

its Stul/Hincll chimeras were recloned into pUFR163ABam, forming pUFR211-215. To

facilitate construction of chimeric genes between pthA and avrb6, the 4.1 kb Sall

fragment containing pthA from pZit45 and the 3.0 kb EcoRIISall fragment containing

avrb6 from pUFR135 were inserted into pGEM7Zf(+). The Stul/HinclI internal

fragments were swapped between the two pGEM derivatives to create chimeric genes.

The chimeric genes were then recloned as single EcoRIIHindSl fragments into pUFRO47,

forming pUFY019 and pUFY020.

Bacterial conjugation. Triparental matings were carried out to transfer broad host

range plasmids from E.coli DH5a to various Xanthomonas strain by using pRK2013 or

pRK2073 as helper plasmids as described (De Feyter and Gabriel 1991a). To transfer






16
plasmids into Xcm, the modifier plasmid pUFR054 carrying XcmI and XcmlII methylase

genes was used to increase the transfer frequency (De Feyter and Gabriel 1991b).

Marker-exchange mutagenesis. Marker-exchange mutagenesis of strain

XcmH1005 was accomplished by introducing the displacement vector pUFRO49 into

XcmH1005 transconjugants harboring pUFR227 (avrb6::Tn5-gusA), pUFR220

(avrb7: :Tn-gusA) or pUFR217 (avrBIn::Tn5-gusA) derivatives. The procedure was the

same as described (Swarup et al. 1991), except that the modifier plasmid pUFR054 was

used to facilitate plasmid transfer into XcmH1005.

Plant inoculations. Cotton (Gossypiwn hirsutum L.) lines used were Acala-44

(Ac44) and its congenic resistance lines Bl, B2, B4, b6, b7, Bin and BIn3 as described

(Swarup et al. 1992, De Feyter et al. 1993). Cotton plants were grown in the

greenhouse, transferred to growth chambers before inoculation, and maintained under

conditions as described (De Feyter and Gabriel 1991a). Bacterial suspensions of X.

campestris pv. malvacearum (108 cfu/ml) in sterile tap water were gently pressure

infiltrated into leaves of 4-5 week old cotton plants. Pathogenic symptoms were

observed periodically 2-7 days after inoculation.

All citrus (Citrus paradise 'Duncan', grapefruit) and bean (Phaseolus vulgaris

'California Light Red') plants were grown under greenhouse conditions. Plant

inoculations involving X. citri or pthA or derivatives of pthA were carried out in BL-3P

level containment (refer to Federal Register Vol.52, No.154, 1987) at the Division of

Plant Industry, Florida Department of Agriculture, Gainesville. Bacterial suspensions

were standardized in sterile tap water to 108 cfu/ml and pressure infiltrated into the






17
abaxial leaf surface of the plants. All plant inoculations on cotton, citrus or bean were

repeated at least three times.

Bacterial growth in plant. To determine the growth of X. campestris pv.

malvacearum in the susceptible cotton line Ac44, bacterial suspensions were adjusted to

106 cfu/ml and inoculated into cotton leaves by pressure infiltration. Leaf discs (1 cm2)

were taken by using a sterilized cork borer, then macerated in 1 ml sterile tap water. At

least three samples were taken at each time point for each strain inoculated. Viable

counts were determined by serial dilutions on plates containing appropriate antibiotics.

Data shown in Fig. 2-2 are the mean and standard error from three experiments.

To quantify the amount of bacteria present on the surface of watersoaked leaf

spots, leaves were inoculated with 107 cfu/ml bacterial suspensions from overnight

cultures. Five days later, 100 p~ sterile tap water was dispensed onto the watersoaked

leaf surface and spread to an area of approximately 1 cm2. After mixing with the

bacteria and slime on the leaf surface, the bacterial suspension was collected with a pipet.

Each 1 cm2 watersoaked leaf area was washed ten times, and a total of 1 ml of bacterial

suspension was collected. To quantify the bacteria remaining inside the leaf, leaf discs

(1 cm2) were taken with a sterilized cork borer, then macerated in 1 ml sterile tap water.

Viable counts were determined by serial dilutions on plates containing appropriate

antibiotics. Data reported in the results (external and total counts) were mean and

standard error of six replicates from two experiments.








Results



Pleiotropic pathogenicity functions of avrb6. X. campestris pv. malvacearum

strain XcmH carries avrB4, avrb6, avrb7, avrBIn, avrBlO1 and avrB102 on a single

plasmid, pXcmH, and elicits an HR on cotton lines carrying any one of many different

resistance (R) genes (De Feyter et al., 1993). Strain XcmH and a spontaneous

rifamycin-resistant derivative of XcmH, XcmH1005, were virulent on susceptible cotton

line Acala-44 (Ac44), and both elicited severe watersoaking and necrosis associated with

growth in plant (XcmHl005 on Ac44 is shown in Figs. 2-1 & 2-2). Mutations of

avrb6, avrb7 and avrBIn were individually generated in XcmH1005 by marker-exchange

mutagenesis and each mutant was confirmed to carry a single TnS-gusA insertion in the

appropriate DNA fragment by Southern blot hybridization (refer Fig. 2-3; some data not

shown). As predicted by gene-for-gene theory, marker-exchange mutants XcmH1407

(avrb6::Tn5-gusA), XcmH1427 (avrb7: :Tn5-gusA) and XcmH1431 (avrBIn: :Tn-gusA)

gained virulence on cotton lines with the resistance genes b6, b7 and BIn, respectively

(Table 2-2). Plasmids pUFR127 (avrb6+), pUFR163 (avrb7+) and pUFR156 (avrBIn+)

were able to fully complement the specific avirulence defects of XcmH 1407, XcmH1427

and XcmH1431, respectively.

Neither XcmH1427 nor XcmH1431 showed any change in watersoaking ability

as compared with XcmH1005. However, marker-exchange mutant XcmH1407 not only

lost avirulence on a cotton line with the b6 resistance gene, but also elicited significantly

less watersoaking and necrosis on susceptible cotton line Ac44 (Table 2-2 and Fig. 2-1A
























Figure 2-1. Watersoaked lesions caused by X. campestris pv. malvacearum strains on
the susceptible cotton line Acala-44. A. Leaf inoculated five days previously with: 1,
XcmH1005; 2, XcmH1407; 3, XcmH1407/pUFR127; 4, XcmH1431. B. Leaf
inoculated seven days previously with: 1, XcmH1005; 2, XcmH1407. C. Leaf
inoculated five days previously with: 1, XcmH1005; 2, Xcml003/pUFR127; 3,
Xcml003/pUFR156; 4, Xcml003.






20




























a-4


0





































n-4











C4.
tj '0)









































































0n o r-


C


o o
N0

0
C0


c









to
(I)

10 0

Cl)

0


(N)


So D
^- uC
0
C



00








(0 LIo 4 rfl


wzu-bs/n;o 5o4


on o3 r Q0 LO .:










1 2 3 4


O
12kb-
0@



iee





3kb- 4



Figure 2-3. Southern blot hybridization of total DNA from XcmH1005 and marker-
exchanged mutants of XcmH1005 after EcoRIISstI digestion. The blot was probed with
the internal 2.9kb BamHI fragment from avrb6. The expected positions of fragments
carrying each plasmid-borne member of the avr/pth gene family were determined from
the restriction map of pXcmH [De Feyter and Gabriel, 1991]. The upper arrow indicates
the expected position of the DNA fragment carrying avrBIn; the lower arrow indicates
the position of avrb6. Lane 1, 1 kb DNA ladder; Lane 2, XcmH1005; Lane 3,
XcmH1407; Lane 4, XcmH1431.









Table 2-2. Phenotypes of marker-exchange mutants of Xanthomonas campestris pv.
malvacearum strain XcmH1005 on cotton cultivar Acala-44 (Ac44) and congenic lines.


Xcm strains Ac44 Acb6 Acb7 AcBIn


XcmH1005 ++ -
XcmH1407 (avrb6::Tn5-gusA) + + -
XcmH1427 (avrb7::Tn5-gusA) ++ ++ -
XcmH1431 (avrBIn::TnS-gusA) ++ ++


aAcb6, Acb7, and AcBIn are congenic lines of Ac44 containing resistance gene b6, b7,
and BIn, respectively. + + = strong watersoaking symptoms; + = weak watersoaking
symptoms; = hypersensitive response.






25
& 2-1B). Plasmid pUFR127 was able to fully complement the pathogenicity defects)

of XcmH1407 on susceptible cotton plants, in addition to complementing the specific

avirulence defect. Despite its reduced ability to elicit pathogenic symptoms on

susceptible cotton lines, mutant XcmH1407 exhibited the same growth rate and yield as

that of the wild type XcmH1005 and XcmH1431 on Ac44 (Fig. 2-2A).

X. campestris pv. malvacearum strain Xcm1003 carries no known avr genes (De

Feyter and Gabriel 1991a), caused less watersoaking on Ac44 than XcmH1005 and

exhibited a lower growth rate on Ac44 than XcmH1005. Introduction of pUFR127

(avrb6+) into Xcml003 conferred increased watersoaking ability (Fig. 2-1C), but growth

rate and yield inplanta of the transconjugant containing pUFR127 were not increased in

comparison with that of Xcml003 or the transconjugant Xcml003/pUFR156 (avrBln+)

(Fig. 2-2B). Therefore pUFR127 affected pathogenic symptoms in both XcmH1005 and

Xcml003, but not bacterial growth rate or yield in plant of either strain.

Strains containing avrb6 (e.g. XcmH1005 and Xcm1003/pUFR127) elicited more

severe watersoaking and necrosis and were associated with much more slime oozing from

the watersoaked areas in comparison with strains lacking avrb6 (e.g. XcmH1407 and

Xcm1003; Fig. 2-1). The peak number of total colony forming units (cfu) per square

centimeter of watersoaked leaf inoculated with XcmH1005 or XcmH1407 were basically

the same (1.41 0.09 x 109 cfu/cm2 vs. 1.48 0.17 x 109 cfu/cm2). However,

14.1% of the XcmH1005 bacteria present in the lesion (2.32 0.56 x 108 cfu/cm2)

were released onto the surface of the leaf, whereas only 0.06% of the XcmH1407

bacteria (9.64 4.42 x 105 cfu/cm2) were released onto the surface of the leaf.






26
Therefore, more than 240 times more bacteria were present on the external surface of

the watersoaked lesions caused by XcmH1005 than by XcmH1407. In experiments

where the leaves were moistened periodically after inoculation with a hand-held mist

sprayer, X. campestris pv. malvacearum strains carrying avrb6 always exhibited many

secondary infections around the original inoculation site. In contrast, those strains

lacking avrb6 rarely exhibited secondary infections.

The pathogenicity functions encoded by pthA and avrb6 are host specific and

determined by 102bp repeats. Single StuI and Hincdl sites are found in avrb6 at positions

1281 and 2860, respectively, that closely flank the 102 bp tandemly repeated region of

the gene (De Feyter et al. 1993). Unique StuI and HincH sites are also found at the

same relative positions in pthA, and the DNA sequences of avrb6 and pthA are identical

in the flanking regions from the 102 bp tandem repeats to these restriction sites (refer to

Chapter 4). Chimeric genes were constructed by swapping the Stul/HincII fragments

containing the internal 102bp repeated regions between pthA and avrb6. At least three

individual clones of both chimeric genes were introduced into a mutant X. citri strain

B21.2 (pthA::Tn5-gusA) and wild type strains of the following pathogens: X. phaseoli

strain G27 (host range on bean), X. campestris pv. citrumelo strain 3048 (host range on

bean and citrus), X. campestris pv. alfalfae strain KX-1 (host range on alfalfa, bean and

citrus) and Xcml003. Strains and transconjugants were inoculated on citrus, bean, and

cotton leaves. As shown in Figure 2-4, pUFY020, carrying a chimeric gene containing

5' and 3' ends of avrb6 and the internal repeats of pthA, complemented the mutant B21.2

to full virulence on citrus. The citrus canker symptoms caused by B21.2/pUFY020 were





































Figure 2-4. Phenotypes of X. citri wild type and mutant strains on a grapefruit (Citrus
paradisi cv.'Duncan') leaf. 1, wild-type strain 3213; 2, marker-exchange mutant of
3213, B21.2; 3, B21.2/pZit45; 4, B21.2/pUFY020; 5, B21.2/pUFR135; 6,
B21.2/pUFY019.






28
indistinguishable from the symptoms caused by B21.2/pZit45 (pthA +). When pUFY020

was present in 3048 or KX-1, both of which cause watersoaked leaf spots on citrus,

canker symptoms were induced by the transconjugants that were indistinguishable from

those conferred by pZit45 in the same strains. Also like pZit45 (Swarup et al, 1992),

pUFY020 conferred avirulence on bean to 3048, KX-1 and G27, and on cotton to

Xcml003.

A chimeric gene containing the 5' and 3' ends of pthA and the internal repeats of

avrb6 on pUFY019 enhanced the ability of Xcml003 and XcmH1407 to watersoak cotton

and cause necrosis. The watersoaking symptoms caused by Xcml003/pUFY019 were

indistinguishable from those caused by Xcml003/pUFR127. This chimeric gene on

pUFY019 behaved like avrb6 on pUFR127 and conferred no detectable symptoms to

B21.2, 3048 or G27 on citrus or bean.

Cultivar-specific avirulence is determined by 102bp repeats. Each of the seven

avr/pth genes (avrB4, avrb6, avrb7, avrBIn, avrBl01, avrB102, and pthA) exhibits

unique avirulence specificity in Xcml003 on cotton resistance lines differing by single

R genes (De Feyter et al. 1993, Swarup et al. 1992). Each of these seven genes

contains two BamHI sites, one near the 5' end and one near the 3' end of each gene, and

single StuI and Hincd sites flanking the 102 bp direct repeat region (De Feyter et al.

1993, Swarup et al. 1992). To localize the region that determines the specificity of the

reactions, a series of chimeric genes were constructed by swapping BamHI and

StuI/HinclI internal fragments among these seven members of the gene family. One to

three individual clones of each chimeric gene were introduced into Xcml003 and








Table 2-3. Avirulence specificity of BamHI-fragment swapped chimeric genes in
Xcm1003 on cotton.



Chimeric Gene
Plasmid Avirulence Specificity
5' and 3' end BamHI fragment


pUFR190 avrb7 avrB4 avrB4
pUFR191 avrb7 avrb6 avrb6
pUFR192 avrb7 avrBIn avrBIn
pUFR193 avrb7 avrB1O1 avrBlOl
pUFR194 avrb7 avrB102 avrB102
pUFR196 avrBIn avrB4 avrB4
pUFR197 avrBIn avrb6 avrb6
pUFR198 avrBIn avrb7 avrb7
pUFR199 avrBIn avrBlO1 avrBlOl
pUFR200 avrBIn avrB102 avrB102

Note: Chimeric genes were introduced into Xcml003 and tested on cotton cv. Ac44 and
its congenic resistance lines AcB1, AcB2, AcB4, Acb6, Acb7, AcBIn and AcBIn3 for
avirulence specificity.









Table 2-4. Avirulence specificity of Stul/HincdI-fragment swapped chimeric genes in
Xcml003 on cotton.



Chimeric Gene
Plasmid Avirulence Specificity
5' and 3' region Stul/HincII fragment


pUFR205 avrBIn avrB4 avrB4*
pUFR206 avrBIn avrb6 avrb6*
pUFR207 avrBIn avrb7 avrb7*
pUFR208 avrBIn avrBlOl1 avrBlOl
pUFR209 avrBIn avrB102 avrB102
pUFR211 avrb7 avrB4 avrB4
pUFR212 avrb7 avrb6 avrb6
pUFR213 avrb7 avrBIn avrBIn*
pUFR214 avrb7 avrBlO1 avrBlOl
pUFR215 avrb7 avrB102 ND
pUFY019 pthA avrb6 avrb6
pUFY020 lacZ::avrb6 pthA pthA +

Note: Chimeric genes were introduced into Xcml003 and tested on cotton cv. Ac44 and
its congenic resistance lines AcB1, AcB2, AcB4, Acb6, Acb7, AcBIn and AcBIn3 for
avirulence specificity. = weak avirulence; + = strong avirulence; ND = no
avirulence detected.






31
inoculated on cotton cultivar Ac44 and its congenic resistance lines, each differing by a

single R gene (Bl, B2, B4, b6, b7, Bin or BIn3). The avirulence specificities of the

seven avr/pth genes were first localized within the internal BamHI fragments (Table 2-3),

and then were further localized within the internal Stul/HinclI fragments (Table 2-4).

In all cases, the avirulence specificity of a given gene was determined inside the

StuV/HinclI (tandem repeat) region.



Discussion



Plant pathologists have long been puzzled by the presence of avirulence genes in

pathogens. These genes act as negative factors to limit virulence and in most cases do

not appear to provide selective advantage to the pathogens (Ellingboe, 1976; Gabriel,

1989; Keen and Staskawicz 1988). Rare exceptions have been reported. For example,

avrBs2 from X. campestris pv. vesicatoria is required for optimal growth in plant

(Kearney and Staskawicz 1990). Both avrb6 and avrb7 from X. campestris pv.

malvacearum strain XcmH are known to enhance the watersoaking ability of X.

campestris pv. malvacearum strain Xcml003 on cotton (De Feyter and Gabriel 1991a).

In this study we demonstrated that the ability of strain XcmH1005 to cause strong

watersoaking and necrosis on cotton requires the presence of avrb6 but not avrb7.

Although avrb6 increased symptom elicitation by both Xcm1003 and XcmH1005, neither

their growth rates nor their maximum bacterial counts per square centimeter of leaf were

affected by the presence or absence of avrb6. Therefore avrb6 functions as a






32
pathogenicity gene and increases symptoms of cotton blight, but without eliciting an HR

on cotton lines lacking b6.

The strong watersoaking ability conferred by avrb6 was correlated with much

higher (240-fold) levels of bacterial cells released from inside the plant leaf to the

surface. Since bacterial blight of cotton is usually spread by rain splash, the presence

of large numbers of bacteria on the leaf surface would undoubtedly contribute to the

dissemination of the population. Strains carrying avrb6 would thereby have a selective

advantage on cotton plants lacking the b6 gene. Similarly, pthA appears to aid in the

dissemination of X. citri by rupturing leaf epidermis and releasing bacteria, although it

does so by inducing tissue hyperplasia (Swarup et al. 1991). Therefore, pthA and avrb6

not only contribute to the amount of damage caused by their respective xanthomonads

to their hosts, but may also contribute to the ecological fitness of their respective

bacterial populations as pathogenicity genes.

Both pthA and avrb6 may help determine host range in a positive manner, and not

as avr genes. When pthA was transferred to X. campestris pv. malvacearum Xcml003

and X. phaseoli G27, it conferred avirulence on the respective hosts, and did not induce

tissue hyperplasia (Swarup et al, 1992). In the present study, when avrb6 was

transferred to X. citri B21.2, X. campestris pv. citrumelo 3048 and X. phaseoli G27, it

conferred no detectable effect when these strains were inoculated onto their respective

hosts. Therefore, the pathogenicity functions of avrb6 and pthA are host specific. If

release of the pathogen to the leaf surface is host specific and it contributes to the

ecological fitness of the pathogen as we propose, then avrb6 and pthA function to






33
determine host range. Furthermore, the avirulence conferred by pthA appeared to be

gratuitous in terms of restricting host range (Swarup et al., 1992) and in the present

study, avrb6 failed to confer avirulence to three other pathogens. Therefore if avrb6 and

pthA help determine host range, it is not because of their function as avr genes. In a

formal genetic sense, these pleiotropic avr/pth genes resemble some Rhizobium host-

specific nodulation (hsn) genes which are required for host range on some hosts, but

which can also confer avirulence when transferred to other Rhizobium strains with a

different host range (Debelle et al. 1988; Faucher et al. 1989; Lewis-Henderson and

Djordjevic 1991).

By swapping the tandemly repeated regions, the avirulence specificities of avrB4,

avrb6, avrb7, avrBIn, avrBlOl, avrB102 and pthA were shown to be determined by the

102bp tandem repeats. These results are consistent with and extend the findings of

Herbers et al. (1992), who showed that the avirulence specificity of avrBs3 is determined

by the 102bp repeats of that gene. In addition, the swapping experiments clearly

demonstrated that the pathogenicity functions of pthA and avrb6 were distinct (cankers

vs. watersoaking), host specific, and in both cases the pathogenic specificity was

determined by their 102bp tandemly repeated regions. Furthermore, the use of chimeric

genes also ruled out the possibility that the host-specific pathogenicity on cotton and

citrus were result of additional, unidentified pathogenicity genes encoded on the plasmids

used (pUFR127 and pZit45).

Extensive deletion analyses of both the 5' and 3' ends of the pXcmH avr genes

showed that all sections of these genes are required for avirulence activity (De Feyter et






34
al. 1993). From the present study, all sections of pthA and avrb6 were required to

confer pathogenicity functions. However, the 5' and 3' ends of the genes outside of the

repeats appeared to be isofunctional among members of the gene family.

Besides the members of this Xanthomonas avr/pth gene family, internal tandem

repeats are found in some pathogenicity genes of animal pathogens. Examples include

the outer membrane protein A gene (ompA) of some Rickettsia species (Anderson et al.

1990, Gilmore 1993), internalin gene (in)) of Listeria monocytogenes (Gaillard et al.

1991), toxin A gene of Clostridium difficile (Dove et al. 1990), and M protein genes

(emm) of Streptococcus (Hollingshead et al. 1987). Tandem repeats were also found in

many genes from protozoan and metazoan parasites such as Plasmodium (McConkey et

al. 1990), Trypanosoma (Hoft et al. 1989), Leishmania (Wallis and McMaster 1987), and

Meloidogyne (Okimoto et al. 1991). Genes from Trypanosoma cruzi (the protozoan

agent of American trypanosomiasis) and mitochondria of Meloidogyne javanica (plant

root knot nematode) contain 102 bp tandem repeats. Most of these genes encode surface

proteins and the distinctive arrangement of the tandem repeats in these genes are thought

to encode a protective, strain-specific conformational epitope for evasion of host

immunity (Gilmore 1993, Hoft et al. 1989, McConkey et al. 1990). By contrast,

AvrBs3 is mainly located in cytosol (Knoop et al. 1991, Brown et al. 1993) and it is not

clear how it may interact with the plant cell.

Southern hybridization has shown that potential members of the avr/pth gene

family exist, often in multiple copies, in 9 of 12 Xanthomonas species or pathovars

examined (Bonas et al. 1989, Swarup et al. 1992, De Feyter et al. 1993). In all X. citri







35
and X. campestris pv. malvacearum cotton strains tested to date, multiple DNA fragments

hybridizing to pthA and avrb6 are found. The pleiotropic pathogenicity functions of pthA

and avrb6 and their potential fitness value may explain why these genes are maintained

in plant pathogens. Based on these data and knowledge of the functions of avrb6 and

pthA, we assume that all strains of X. campestris pv. malvacearum capable of strongly

watersoaking cotton carry an avrb6 gene or homologue that functions for pathogenicity.

Similarly, we assume that all strains of X. citri capable of causing cankers on citrus carry

a pthA gene. A number of pathogenic strains and pathovars in the genus Xanthomonas

do not carry members of the gene family, and therefore these genes are not required for

Xanthomonas virulence generally. Within pathovars where members of the avr/pth gene

family are found in some, but not all strains tested (such as X. campestris pv.

vesicatoria), there may be no pleiotropic pathogenicity function; for example, there is no

evidence of a pathogenicity function of avrBs3, avrBs3-2 and avrBsP (Bonas et al. 1989,

1993; Canteros et al. 1991).

The mechanisms) by which leaf spotting pathogens elicit watersoaking and

necrosis is unknown, but appears to involve both damage to leaf cell membranes (without

eliciting an HR) and production of extracellular polysaccharide (EPS). The EPS does

not appear to be involved in suppressing a potential HR, since production levels of EPS

by X. campestris pv. malvacearum are similar in both susceptible and resistant cotton

lines (Pierce et al. 1993). Instead, mutational analyses of Xanthomonas EPS biosynthetic

genes and inoculations with purified EPS have shown that the EPS contributes to

watersoaking by trapping water and nutrients in the intercellular spaces after they are






36
released (reviewed by Leigh and Coplin 1992). Coplin et al. (1992) have proposed that

watersoaking and pathogenicity of Erwinia stewartii involves the EPS plus a cell leakage

factor encoded by wts (watersoaking) genes. This hypothesis may well apply to X.

campestris pv. malvacearum, with avrb6 encoding a cell leakage factor. Like some wts

genes of E. stewartii, avrb6 was not required for bacterial growth in plant, but strongly

affected watersoaking on its host.

The ion-channel defense model of the gene-for-gene hypothesis invokes a cell

leakage factor as the product of an avr gene (Gabriel et al. 1988, Gabriel and Rolfe

1990). In this model, avr genes produce a protein or compound that opens an ion

channel in the plant cell membrane that rapidly depolarizes the membrane, causing

electrolyte leakage and host cell death. This gene-for-gene model is not inconsistent with

the idea that the product of an avr gene might induce slower cell leakage on susceptible

hosts. The only difference might be the allelic form of the host R gene. If the leakage

were slow enough to avoid cascade amplification of a wound-response signal, changes

in the osmotic gradient could cause a net loss of water from the cell and redistribution

to the apoplast, thereby increasing the fluidity of the EPS (M. Essenberg and M. Pierce,

personal communication). Increased fluidity or amounts of EPS may increase bacteria

exuding onto the leaf surface through stomata (Thiers and Blank 1951). Another possible

function is that the increased levels of necrosis induced by avrb6 may serve to collapse

the palisade layer and physically squeeze more bacteria onto the leaf surface.











CHAPTER 3
INACTIVATION OF MULTIPLE AVIRULENCE GENES IN XANTHOMONAS
CAMPESTRIS PV. MALVACEARUM CREATED A NONPATHOGENIC
ENDOPHYTE OF COTTON



Introduction



Avirulence (avr) genes act as negative factors to confer avirulence and to limit

host range of the pathogen on host plants carrying specific resistance (R) gene. As a

result, most avr genes reported to date appear to be gratuitous and of no obvious

selective value to the microbial plant pathogens (Gabriel et al. 1993). However, a few

of them have demonstrated pleiotropic functionss. For example, avrBs2 is required for

full virulence of Xanthomonas campestris pv. vesicatoria on susceptible hosts (Kearney

and Staskawicz 1990). More recently, pthA of X. citri and avrb6 of X. campestris pv.

malvacearum were also shown to have host-specific pathogenicity function in addition to

avirulence function (Swarup et al. 1991, 1992; Chapter 2). Gene pthA is essential for

X. citri strain 3213 to induce hyperplastic cankers on citrus specifically and confers that

function to several other X. campestris pathovars. Similarly, avrb6 is important for X.

campestris pv. malvacearum strain XcmH to induce watersoaking on cotton specifically,

and confers that function to other X. campestris pv. malvacearum strains. Both pthA and

avrb6 appear to contribute to the fitness of their respective pathogens on their hosts by






38
increasing release of the pathogens from plant leaves, thus affecting host range through

increased dispersal.

Both pthA and avrb6 are members of a large Xanthomonas avirulence (avr) and

pathogenicity (pth) gene family, and both confer gene-for-gene avirulence to X.

campestris pv. malvacearum on cotton (Swarup et al. 1993; De Feyter et al. 1993).

Other members of the avr/pth gene family include avrB4, avrb7, avrBIn, avrBlOl and

avrB102 of X. campestris pv. malvacearum (De Feyter and Gabriel 1991a, De Feyter et

al. 1993), avrBs3, avrBs3-2 and avrBsP of X. campestris pv. vesicatoria (Bonas et al.

1989, 1993; Canteros et al. 1991) and avrXa5, avrXa7 and avrXalO of X. oryzae

(Hopkins et al. 1992). Family members sequenced to date are 95%-98% identical to

each other (Bonas et al. 1993; Hopkins et al. 1992; De Feyter et al. 1993). Most

members of this gene family were isolated as avirulence genes with no evidence of

pathogenicity function.

Xanthomonas campestris pv. malvacearum (Smith) Dye is the causal agent of

bacterial blight of cotton, an economically important, world-wide disease. Six members

of this gene family were previously isolated from pXcmH, a 90-kb plasmid of X.

campestris pv. malvacearum strain XcmH. Southern hybridization analysis reveals at

least five additional hybridizing fragments in XcmH besides the six plasmid-borne avr

genes (De Feyter et al. 1993). Furthermore, all strains examined of Xanthomonas

campestris pv. malvacearum that attack cotton carry at least 4-11 DNA fragments that

hybridize with members of the avrlpth gene family, including many strains which carry

no known avr genes (De Feyter et al. 1993). The purpose of this research was to clone






39
and to mutagenize all avr genes in X. campestris pv. malvacearum strain XcmH and to

identify potential pleiotropic functions.



Materials and Methods



Bacterial strains, plasmids, and matings. The bacterial strains and plasmids used

in this study are listed Table 3-1. Strains of Escherichia coli were grown in Luria-

Bertani (LB) medium (Sambrook et al. 1989) at 370C. Strains of Xanthomonas were

grown in PYGM (peptone-yeast extract-glycerol-MOPS) medium at 30*C (De Feyter et

al. 1990). For culture on solid media, agar was added at 15g/L. Antibiotics were used

as previously described (refer to Chapter 2). To transfer plasmids from E.coli strains

HB101 or DH5a to X. campestris pv. malvacearum, helper plasmid pRK2073 and

modifier plasmid pUFR054 were used as described (Feyter and Gabriel 1991a,b).

Recombinant DNA techniques. Total DNA isolation from Xanthomonas was as

described (Gabriel and De Feyter 1992). Plasmids were isolated from E. coli by alkaline

lysis methods (Sambrook et al. 1989). Restriction enzyme digestion, alkaline

phosphatase treatment, DNA ligation and random priming reactions were performed as

recommended by the manufacturers. Southern hybridization was performed by using

nylon membranes as described (Lazo et al. 1987). Otherwise, standard recombinant

DNA procedures were used (Sambrook et al. 1989).

Gene replacement and marker eviction. To carry out marker-exchange

mutagenesis, a 3.8 kb BamHI fragment containing a nptl-sac cartridge from pUM24








Table 3-1. Bacterial strains and plasmids used in this study


Strain or Plasmid Relevant Characteristics Reference or Source


F-, endA1, hsdRl 7 (r-mk+), supE44,
thi-1, recAl, gyrA, relAl, #8OdlacZ
AM15, A(lacZYA-argF)U169
supE44, hsdS20 (r-m ), recAl3,
ara-14, proA2, lacYl, galK2, rpsL20,
xyl-5, mdl-I
supE44, supF58, hsdS3 (r-mk+), recA56,
galK2, galT22, metBI


Gibco-BRL,
Gaithersburg, MD

Boyer &
Roulland-Dussoix 1969


Murray et al. 1977


X campestris pv. malvacearum


Natural isolate from cotton from
Oklahoma; carries six avr gene on pXcmH,
and additional six potential avr genes.
Spontaneous Rif derivative of XcmH
avrBIn::nptl-sac, marker exchange
mutant of XcmHl005
avrb6::nptl-sac, marker exchange
mutant of XcmHl005
avrB5::nptl-sac, awvrb6), marker
exchange mutant of XcmH1005
avrB5-, avrb6- marker-evicted
mutant derived from HM 1.20
Hc l: :nptl-sac, marker-exchange
mutant of XcmH1005
Hcl-, avrBIn', avrBl04- marker
exchange mutant of XcmH1005
avrBn: :nptl-sac, marker exchange
mutant of XcmH1005
avrBlO3::nptl-sac, marker exchange
mutant of XcmH1005
avrB4-, avrbT, avrBIn-, avrBlOl',
avrl02- marker exchange mutant of XcmHl005
avrHc6~, avrB4", avrb6-, avrbT,
avrBIn-, avrBlOl', avrl02"), marker exchange
mutant of HM1.20S.
XcmH1005 (avrHc6-, avrB4-, avrb6", avrbT,
avrBIn-, avrBlOl', avrl02"), marker-evicted
mutant derived from HM2.2S.
Natural isolate from cotton from
Upper Volta, Africa.
Spc' Rif derivative of XcmN


De Feyter & Gabriel, 1991a


Chapter 2
This study

This study

This study

This study

This study

This study

This study

This study

This study

This study


This study


Gabriel et al. 1986

De Feyter & Gabriel 1991a


E. coli

DH5-,


HB101


ED8767


XcmH


XcmHl005
HM1.10

HM1.15

HM1.20

HM1.20S

HM1.26

HM1.32

HM1.34

HMI.36

HMI.38

HM2.2


HM2.2S


XcmN

Xcml003








Table 3-1--continued.

Strain or Plasmid Relevant Characteristics Reference or Source


Plasmid


pRK2073

pUFR004
pUFR034
pUFR042
pUFRO47
pUFRO54

pUFR103
pUFR106
pUFR107
pUFR109
pUFRI11
pUFRI12
pUFR113
pUFRI115
pUFR127
pUFRI42
pUFR156
pUFRI57
pUFR163
pUFYI.48

pUFY10.1

pUFY31.46
pUFY33.19
pUFY36.8
pUFY36.26
pUFY37.62

pUFY38.1
pUM24
pXcml.12
pXcml.21
pXcm2.23
pXcm2.12
pXcm2.11
pXcml.22
pZit45


pRK2013 derivative, npt::Tn7,Knm, Sp',
Tra+, helper plasmid
ColEl, Mob*, Cm'
IncW, Km', Mob*, lacZa*, Par*, cos
IncW, Knm, Gm', Mob*, lacZa*, Par+
IncW, Gmn, Ap', Mob+, lacZa+, Par+
IncP, Tc', Mob*, containing methylases
Xmal and Xmanl
Cosmid clone carrying avrB4, avrb6.
Cosmid clone carrying avrB4, avrb6, avrBlO1
Cosmid clone carrying avrb6, avrBlO1, avrBIn
Cosmid clone carrying avrBIOl, avrBIn
Cosmid clone carrying avrBlOl, avrBIn, avrBl02
Cosmid clone carrying avrBIn, avrB102, avrb7
Cosmid clone carrying avrBI02, avrb7
7.5-kb fragment containing avrB4 in pUFRO42
5-kb fragment containing avrb6 in pUFRO42
9-kb fragment containing avrBlOl in pUFRO47
12.9-kb fragment containing avrBIn in pUFRO42
11-kb fragment containing avrB102 in pUFRO42
10-kb fragment containing avrb7 in pUFRO42
8.4-kb fragment carrying pthA::npt-sac
in pUFRO47
8.1-kb fragment carrying pthA: :npt-sac
in pUFR004
8.5-kb fragment carrying avrBI03 in pUFRO47
5-kb fragment carrying avrB5 in pUFRO47
8.6-kb fragment carrying Hcl in pUFRO47
5-kb fragment carrying avrBn in pUFRO47
11-kb fragment carrying avrBl04
and avrBS on pUFR047
8-kb fragment carrying avrB104 in pUFRO47
nptl-sac marker in a 3.8-kb BamHl fragment
Cosmid done carrying Hcl and avrBn
Cosmid clone carrying avrBl03
Cosmid clone carrying avrBn
Cosmid clone carrying avrB104 and avrB5
Cosmid clone carrying avrB5
Cosmid clone carrying Hc6
4.5-kb fragment containing pthA in pUFRO47


Leong et al. 1982

De Feyter et al. 1990
De Feyter et al. 1990
De Feyter & Gabriel 1991a
De Feyter et al. 1993
De Feyter & Gabriel 1991b

De Feyter & Gabriel 1991a
De Feyter & Gabriel 1991a
De Feyter & Gabriel 1991a
De Feyter & Gabriel 1991a
De Feyter & Gabriel 1991a
De Feyter & Gabriel 1991a
De Feyter & Gabriel 1991a
De Feyter & Gabriel 1991a
De Feyter & Gabriel 1991a
De Feyter et al. 1993
De Feyter et al. 1993
De Feyter et al. 1993
De Feyter et al. 1993
This study

This study

This study
This study
This study
This study
This study

This study
Ried and Collmer 1987
This study
This study
This study
This study
This study
This study
Swarup et al. 1992






42
(Ried and Collmer 1987) was randomly ligated into a Ball site of pthA on pZit45.

Recombinant plasmids were screened for an insertion in the middle of the tandemly

repeated region, and pUFYl.48 was selected, in which the nptl-sac cartridge was found

in the Ball site of repeat number 10 of pthA. An SstI fragment carrying the pthA::nptl-

sac fusion from pUFY1.48 was recloned into the suicide vector pUFR004 (De Feyter et

al. 1990), forming pUFY10.1. Marker-exchange mutants (avr::nptl-sac) were created

by transferring pUFR10. 1 to XcmH 1005 and selecting for colonies resistant to kanamycin

(15 pg/ml) and sensitive to chloramphenicol (35 ;tg/ml) and sucrose (5%). Some

marker-exchange mutants were plated on the PYGM medium containing 5 % sucrose and

nptl-sac marker evicted strains were selected to allow further rounds of marker exchange

mutagenesis.

Plant inoculations and bacterial growth in plant. Cotton (Gossypiwn hirsutum

L.) lines used were Acala-44 (Ac44) and its congenic resistance lines AcB1, AcB2,

AcB4, AcB5a, AcB5b, Acb6, Acb7, AcBIn and AcBIn3 as described (Swarup et al.

1992, De Feyter et al. 1993). Cotton plants were grown in the greenhouse, transferred

to growth chambers before inoculation, and maintained under conditions as described (De

Feyter and Gabriel 1991a). Bacterial suspensions of X. campestris pv. malvacearum (108

cfu/ml) in sterile tap water were gently pressure infiltrated into leaves of 4-6 week old

cotton plants. Pathogenic symptoms were observed periodically 2-7 days after

inoculation.

The growth of X. campestris pv. malvacearum in the susceptible cotton line Ac44

was determined as described (refer to Chapter 2). In plant growth experiments were






43

performed three times. Data points shown in Fig. 3-4 were the mean and standard error

of three samples from one experiment. Bacterial populations released onto the leaf

surface were quantified as described (refer to Chapter 2). These data are reported as the

mean and standard error of three samples. Similar results were obtained in two

independent experiments.



Results



Cloning of six potential avr genes from XcmH by hybridization. In addition to

the six members of the Xanthomonas avr/pth gene family previously cloned from plasmid

pXcmH of strain XcmH (De Feyter and Gabriel, 1991a), Southern blot analyses revealed

the presence of six more DNA fragments in XcmH that hybridized to an internal BamHI

fragment from avrb6 (Fig. 3-1). All of the fragments appeared to be large enough to

carry functional members of the avr/pth gene family. Based on the intensity of

hybridization and the fact that XcmH appears to carry only a single plasmid (pXcmH;

De Feyter and Gabriel, 1991a), the additional hybridizing fragments appeared to be

chromosomal. Colony hybridizations were carried out to isolate these chromosomal

DNA fragments from a genomic library of XcmH (De Feyter and Gabriel 1991a) by

probing with an internal BamHI fragment from avrb6. Twenty-three out of seventy-two

cosmid clones identified by colony hybridization were found to carry DNA fragments not

found on pXcmH. These new, presumably chromosomal DNA fragments were readily

distinguished from the DNA fragments that form pXcmH by restriction fragment length

















rd14







Co~0 to



Fo

I4-R. .













oP.- 0 ce0
OA ~ I-*









E E) E ~
,uU

&,- -..

m b


Ga~







45









* .










e 0-M A
S1.




ate-mal -a a- m


I I I I
mn r'I V* 0
M da 4 A4







46
polymorphisms (Fig. 3-1). Six different chromosomal DNA fragments hybridizing with

avrb6 had been cloned and identified. These DNA fragments were arbitrarily named

according to size as Hcl 6. Six cosmid clones were selected for further study:

pXcml.12 carried Hcl and Hc3, pXcml.21 carried Hc2, pXcm2.23 carried Hc3,

pXcm2.12 carried Hc4 and Hc5, pXcm2.11 carried Hc5 only, and pXcml.22 carried

Hc6 (Fig. 3-1). Restriction digests of these clones revealed that Hcl, Hc2 and Hc3 were

closely linked, and that Hc4 and Hc5 were closely linked.

Avirulence activity of hybridizing fragments. All cosmid clones were conjugally

transferred to the widely virulent X. campestris pv. malvacearum strain Xcml003 and

tested on cotton cv. Acala 44 (Ac44) and 9 cotton lines, congenic with Ac44 and each

carrying one of the following resistance genes: Bl, B2, B4, B5a, B5b, b6, b7, Bin and

BIn3. To localize members of the avr/pth gene family, all cosmid clones with avirulence

activity were directionally subcloned into pUFRO47 by EcoRI and Hindull digestion

(these enzymes are not known cut within members of the family published to date).

pXcml.12 was subcloned to separate fragments Hcl and Hc3, generating pUFY36.8

(Hcl) and pUFY36.26 (Hc3). Fragment Hc2 on pXcml.21 was subcloned to form

pUFY31.46. pXcm2.12 was subcloned to generate pUFY37.62, still carrying Hc4 and

Hc5 on a single, 11 kb EcoRI/HindII DNA fragment. Fragment Hc4 on pUFY37.62

was further subcloned to form pUFY38.1. Fragment Hc5 on pXcm2.11 was subcloned

to form pUFY33.19. Since Hc6 on pUFY1.22 conferred no avirulence to Xcml003, it

was not evaluated further.






47
The indicated subclones were transferred to Xcm 1003 and tested on the congenic

cotton lines and on cotton line 20-3, carrying resistance gene Bn. The Hcl and Hc6

DNA fragments conferred no detected phenotype to Xcml003. The Hc2 and Hc4 DNA

fragments conferred exactly the same avirulence specificity to Xcml003 as did clones

carrying avrBIO1 (De Feyter et at 1993): avirulence on AcB5a and AcBIn3. Therefore

we conclude that Hc2 and Hc4 carry two new members of the gene family, named

avrB103 and avrB104, respectively. The Hc3 DNA fragment on pXcm2.23 conferred

avirulence to Xcml003 on cotton lines containing resistance genes Bn or BIn3. Gene

avrBn had previously been cloned from XcmH on pUFA-HI and shown to confer

avirulence on cotton lines containing Bn (Gabriel et al, 1986), but avrBn was not known

to be a member of this avr/pth gene family. Restriction and Southern hybridization

analyses (not shown) revealed that avrBn is a member of this gene family, and is carried

on both pUFA-HI and pXcm2.23. The Hc5 DNA fragment conferred avirulence to

Xcml003 on AcB5b and AcBIn3. Therefore we conclude that Hc5 carried a new

member of the gene family, named avrB5.

Inactivation of avr genes in XcmH1005. To study the pleiotropic function of all

members of the avr/pth gene family in XcmH1005 (a rifamycin mutant of XcmH),

marker exchange-eviction mutagenesis was carried out to inactivate or delete each of the

avr genes, singly or in combination. More than 140 mutants were generated and 60 of

them were tested on 10 congenic cotton resistance lines and analyzed by Southern

hybridization. Marker exchange-eviction mutants affecting all 10 avr gene were

obtained. In Table 3-2 are listed selected mutants which exhibited altered phenotypes on











I I I


+
I +


I I I


+
I I I +


+
I I + + + +


+
I I I +1 +1 I 0


+
I I I I +


+ +
+ + + + +


r-r

C13
C0 i


'to (
I

o5 re o ,--
1~~ ~ 1100'i?!i
Co CU (0 0 ^ s I
I o i
cu M O Oa
M co cc
to a
I I ^ (U1 01


1 LO0 c (j) -


x XI I 3m I X mm t


0
C0
0
C

0








C
I 8
0
0






0
0


C





Co






0+
0









o.









S-



o4)
0:

0 +1



0 0







S-
O







49
cotton. Southern analysis of 11 mutants was shown in Fig. 3-2. As predicted by gene-

for-gene theory, mutations of avrBIn (HM1.10), avrb6 (HM1.15), or avrB4 plus avrb7

(HM1.38) in XcmH1005 resulted in the loss of avr specificity for cotton blight resistance

genes Bin, b6, B4, and b7, respectively. As reported in the previous chapter, a single

mutation of avrb6 reduces the watersoaking ability of XcmH1005 on susceptible cotton

lines. In contrast, mutation of avrB4, avrb7 or avrBIn did not affect watersoaking ability

of XcmH1005 on Ac44. A mutation in avrB5 (HM1.20) was shown to result in the same

phenotypic defects as the mutation in avrb6, i. e., a loss of avr specificity for resistance

gene b6 and reduction of watersoaking on the susceptible cotton line Ac44. However,

this mutation could not be complemented by the cloned avrB5 gene, instead, was

complemented by the cloned avrb6, indicating avrb6 in HM1.20 was likely to be

affected. Mutation of both avrB5 and avrb6 in HM1.20 and HM1.20S resulted a partial

loss of avirulence specificity for cotton blight resistance gene B5b. Mutation of avrBlO1,

avrB102, avrB103 (HM1.36), avrBIO4 (HM1.32), avrBn (HM1.34) or Hcl (HM1.26)

fragment resulted in no alteration of avirulence phenotype. This is consistent with the

fact that the cotton resistance genes that react with these avr genes also react with other

avr genes (gene-for-genes; De Feyter et al. 1993). No mutations affecting fragment Hc6

were recovered.

To inactivate a large number of avr genes in XcmH1005, the nptl-sac marker-

evicted mutant HM1.20S (avrB5-, avrb6-) was subjected to a second round of marker

exchange mutagenesis. The resulting mutant HM2.2 and marker evicted progeny

HM2.2S had at least 7 avr genes (avrB4", avrB5", avrb6', avrb7, avrBIn', avrBlOl',












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


a- -


Figure 3-2. Southern hybridization of EcoRI and SstI double digested genomic DNAs of
XcmH mutants. The blot was probed with 32-P labelled internal BamHI fragment of
avrb6. Lane 1, XcmH; lane 2, HM1.10; lane 3, HM1.15; lane 4, HM1.38; lane 5,
HM1.26; lane 6, HM1.36; lane 7, HM1.34; lane 8, HM1.32; lane 9, HM1.20; lane 10,
HM1.20S; lane 11, HM2.2; lane 12, HM2.2S. a, Hcl; b, avrB4; c, avrBl01; d,
avrB102, e, avrBl03; f, avrBn; g, avrBIn; h, avrB104; i, avrb7; j, avrBS; k, avrb6; 1,
Hc6.


t-5


!~Sw


b-
c,d-
e-
g-

I-
h-

jr


.0..






51
avrl02") inactivated. These two mutants not only lost avr specificity on most of cotton

resistance lines, but also lost nearly all watersoaking ability on the susceptible line Ac44

(Tab. 3-2, Fig. 3-3).

Despite repeated attempts to inactivate all members of the gene family in XcmH,

we were unable to reduce the number of members below five. In all cases, members of

the family appeared to rearrange to form new members, as evidenced by the appearance

of new DNA fragments of sufficient size to encode a member, and often accompanied

by the appearance of a new avirulence specificity phenotype (data not shown). Also in

all cases, at least one member remained plasmid-borne, as evidenced by comparatively

strong hybridization intensities characteristic of the plasmid-borne genes.

Effect of loss of watersoaking on bacterial growth in plant. Since avrb6'

mutants of XcmH1005 that had lost significant watersoaking ability on Ac44 were shown

to be unaffected in growth in plant (refer to Chapter 2), we evaluated growth of

HM2.2S on Ac44 in comparison with the parental strain XcmH1005. While inactivation

of 7 avr genes resulted in a loss of nearly all watersoaking ability on cotton, in plant

growth rate and yield of HM2.2S was not affected (Fig. 3-4). However, the amount

of HM2.2S released onto the leaf surface was 1600 times less than that of XcmH1005.

Only 0.007% (8.8 3.0 X 104 cfu/cm2 out of 1.18 0.24 X 109 cfu/cm2) of HM2.2S

bacteria present in the infected leaf tissue were released onto the leaf surface, whereas

9.2% (1.42 0.15 X 10' cfu/cm2 out of 1.55 0.23 X 109 cfu/cm2) of the XcmH1005

bacteria were released onto the surface. Therefore, the mutant HM2.2S was basically

a nonpathogenic endophyte of cotton.






































Figure 3-3. Effect of avr genes on watersoaking symptoms caused by X. campestris pv.
malvacearum on cotton susceptible line Ac44.































0 2 4 6 8 10 12

Day post-inoculation





Figure 3-4. Inactivation of multiple avr genes in X. campestris pv. malvacearum did not
affect bacterial growth in plant on cotton susceptible line Ac44. Filled circle,
XcmH1005; open circle, HM2.2S.






54

Additive effect of different avr genes on watersoaking ability of XcmH1005.

Complementation tests were carried out to further analyze the potential pleiotropic

pathogenicity function of 12 avr genes isolated from XcmH. As shown in Fig. 3-5,

when plasmids carrying single avr genes isolated from XcmH were introduced into the

mutant HM2.2S, only avrb6 was able to partially complement the watersoaking defect

of HM2.2S. None of the other 9 avr genes or Hcl or Hc6 fragments were able to even

partially complement the watersoaking defect in HM2.2S when introduced individually.

However, when plasmids carrying two or three of these same avr genes (not necessarily

including avrb6) were introduced into HM2.2S, many were able to partially complement

the watersoaking defect (Figs. 3-3 and 3-5). For example, plasmids containing avrBIn

or avrBl01 alone conferred no watersoaking activity, but plasmids containing both

avrBIn and avrBl01 together exhibited watersoaking activity. Cosmid clones (pUFR106

and pUFR107) carrying avrb6 and at least two other avr genes were able to fully

complement the watersoaking defects of HM2.2S (Fig. 3-3). Therefore, at least three

of the pXcmH avr/pth genes were redundant in terms of pathogenicity function.

To exclude the possibility that some watersoaking genes were linked to avr genes

on these complementing cosmid clones, subcloning of pUFR107 was carried out.

Activity assays of regions between the avr genes did not reveal any watersoaking activity

when introduced into HM2.2S.


















avrB4 avrb6 avrBl0O avrBIn

1 I II I I I


20.5


avrBI02

I I


12.9


avrB 102 avrb7


avrBlOt avrBIn


avrwBIO avrBIn avrB02


avrBIn avrBI02


avrwBO avrBIn


avrB1Of


Figure 3-5. Pleiotropic watersoaking function of cloned XcmH avr genes in mutant
HM2.2S. HM2.2S was nearly asymptomatic on susceptible cotton line Ac44; + weak
watersoaking activity; ++ intermediate level watersoaking activity; +++ full
watersoaking activity (as wild type XcmH).


pXcmH



pUFR127


11.6 5.0 4.5 4.8


avrb7


avrb6


28.4


pUFR103 avrB4 avrb6


Activity

in HM2.2S


pUFR113


pUFRI09


pUFR111


pUFR112


pUFR107


pUFR106


+


+


++


++


++


avrb7


avrb6


avrB4 avrb6








Discussion



Nearly all of the avr genes reported from the genus Xanthomonas are members

of a single, highly homologous gene family, and with the exception of pthA

pathogenicityy) of X. citri (Swarup et al. 1991, 1992), all of the other published members

were identified and cloned as avr genes. Southern analyses have shown that members

of this avr/pth gene family exist in multiple copies (from 4 to 11) in all tested X.

campestris pv. malvacearum strains isolated from cotton, but that some strains,

particularly those from Africa, appear to lack functional avirulence (De Feyter et al.

1993). For example, the African strain Xcml003 used in this study contains at least five

hybridizing fragments, but is virulent on all tested cotton resistance lines. Of course it

is possible that the hybridizing fragments in a single strain are all nonfunctional, but it

is more difficult to explain the presence of from 4 11 hybridizing DNA fragments

present in all known X. campestris pv. malvacearum strains if they encoded no function

beneficial to the pathogen.

All 12 DNA fragments that hybridize with an internal fragment from the gene

family have now been isolated from a single X. campestris pv. malvacearum strain,

XcmH. Six of them are clustered on a 90-kb indigenous plasmid, pXcmH, and were

originally isolated as functional avr genes, although two of them (avrb6 and avrb7) were

noted to confer pleiotropic pathogenicity (watersoaking) function (De Feyter and Gabriel

1991a; De Feyter et al. 1993). We report here the cloning of the other six hybridizing

fragments that appear to be from the chromosome. One of these fragments encodes the






57
previously described avrBn (Gabriel et al 1986). Three of these encode previously

undescribed avr genes (avrBS, avrB103 and avrB104), and the remaining two fragments

conferred no avirulence on cotton and were presumably non-functional.

Mutations of all ten avr genes in XcmH1005 were achieved by marker exchange-

eviction mutagensis. With the exception of avrb6, mutations of individual members of

the avr/pth gene family in XcmH1005 had no obvious effect on watersoaking on

susceptible cotton lines. Mutation of avrb6 resulted in a significant loss of watersoaking

ability of XcmH1005. Although avrb6 appeared to be the major pathogenicity gene

among 10 homologous avr genes in the strain XcmH, many XcmH1005 avr genes also

contributed to watersoaking function of XcmH1005 on cotton. Therefore these avr genes

also function pleiotropically as pth genes. Mutant HM2.2S had at least seven avr genes

(including avrb6) inactivated, and completely lost the ability to induce watersoaking on

mature cotton leaves. Since HM2.2S suffered deletions of avr genes and adjacent DNA

fragments, mutational analysis alone did not rule out the potential involvement of closely-

linked virulence genes. However, complementation tests confirmed that avr/pth family

members fully complemented the watersoaking defects) of HM2.2S. Individually, none

of the 10 avr/pth genes, other than avrb6, could even partially complement the

nonpathogenic mutant strain HM2.2S in visual assays. However, DNA fragments

carrying multiple avr/pth genes from plasmid pXcmH were able to partially complement

the pathogenic defect of HM2.2S, even if they did not carry avrb6, demonstrating that

other avr/pth genes from XcmH also contributed to its pathogenicity.






58
Some avr/pth genes appeared to have stronger pathogenicity function than others

in HM2.2S, and all six pXcmH genes appeared able to contribute to pathogenicity on

cotton, when tested in combinations. However, no individual avr/pth gene, except

avrb6, exhibited pth function in HM2.2S. Full restoration of the watersoaking defect of

HM2.2S required avrb6 plus two other pXcmH avr/pth genes. Thus, the cotton pathogen

XcmH contained one major and multiple minor genes from the same avr/pth gene family.

Some of the minor avr/pth genes appeared to be redundant in terms of watersoaking

function. Therefore mutation of one or few of minor avr/pth genes would not affect the

watersoaking and release of the pathogen, and presumably, the fitness of the pathogen.

In contrast to avrBs2 of X. campestris pv. vesicatoria (Kearney and Staskawicz

1990), there was no evidence that the avr/pth genes of XcmH are even slightly involved

in bacterial growth in plant. Instead, they were involved in the induction of

watersoaking symptoms and associated release of 9.2% of the total bacteria of the leaf

onto the leaf surface, presumably aiding in dispersal. Mutant HM2.2S, with at least

seven avr/pth genes destroyed, released 1,600 times less bacteria to the leaf surface than

the wild type. Although HM2.2S was asymptomatic on cotton, it maintained the same

level of bacterial growth in plant as the wild type strain. Therefore, inactivation of

multiple avr/pth genes in X. campestris pv. malvacearum created a nonpathogenic

endophyte of cotton. In the case of X. campestris pv. malvacearum, pathogenicity and

host tissue destruction appears to be beneficial for the reproductive fitness of the

pathogen. Since X. campestris pv. malvacearum spreads primarily by rain splash, it

seems doubtful that an endophyte could disseminate efficiently in natural field situations.






59
The lack of correlation between disease symptoms and growth in plant has been

observed with other plant pathogens. Indole acetic acid-deficient strains of Pseudomonas

syringae pv. savastanoi failed to elicit gall on oleander but exhibited similar growth

patterns as the gall-eliciting wild-type strain (Smidt and Kosuge 1978). The lemA gene,

which encodes a two-component regulator, is required by P. syringae pv. syringae for

disease lesion formation on bean plants, but not bacterial growth within or on the leave

of bean (Hrabak and Willis 1994; Willis et al. 1990). Mutation of the fungal pathogen

Colletotrichum magna, resulted in a strain which is not able to induce pathogenic

symptoms, but grows in host tissue as an endophyte and retains the wild-type host range

(Freeman and Rodriguez 1993).

A growing number of Xanthomonas virulence / pathogenicity genes are not hrp

(hypersensitive response and pathogenicity) genes, but instead are host-specific

determinants of disease and/or host range (Gabriel et al, 1993). For example, several

genes affecting host range were identified and complementing DNA clones isolated from

X. campestris pv. citrumelo (Kingsley et al, 1993) and X. camnpestris pv. translucens

(Waney et al, 1992). These genes are host specific and affect growth in some hosts, but

not in other hosts. Other non-hrp Xanthomonas pathogenicity genes, such as pthA of X.

citri and avrb6 of X. campestris pv. malvacearum are host specific and strongly affect

elicitation of pathogenic symptoms. From the present study, it appears that many

members of the avr/pth gene family of X. campestris pv. malvacearum are host-specific

determinants of disease and host range on cotton. Genetically, pthA of X. citri and the

avr/pth genes of X. campestris pv. malvacearum resemble some Rhizobiwn host-specific






60
nodulation genes which are required for host range on some hosts, but which can also

confer avirulence when transferred to other Rhizobium strains with a different host range

(Debelle et al. 1988; Faucher et al. 1989; Lewis-Henderson and Djordjevic 1991).

The pleiotropic function and apparent fitness value of some avr genes do not

imply that most avr genes have selective value for a specific plant pathogenic strain. In

fact, most avr genes reported to date appear to be dispensible and have no obvious

selective value to the pathogen (Gabriel et al. 1993). At least some members of the

Xanthomonas avr/pth gene family are found on self-mobilizing plasmids (for example,

Bonas et al. 1989), and their presence in some strains of a pathovar may be due to a

coincidental linkage with another (selected) factor, such as copper resistance, on the same

plasmid. Once horizontally transferred to a different strain, these genes may not remain

stable enough to retain pth function specific for some ancestral host. In both X.

campestris pv. malvacearum and X. citri, members of the avr/pth gene family mutate at

very high frequency (approximately 10-4) through intergenic or intragenic recombination

(De Feyter et al. 1993; Chapter 4). High frequency recombination provides a powerful

mechanism to generate variation and horizontal gene transfer provides a powerful

mechanism to disseminate that variation in the genus. It seems likely that many avr

genes carried by bacterial plant pathogens may result from variation in redundant and/or

unnecessary virulence/pathogenicity genes which are selectively neutral in natural

populations.











CHAPTER 4
INTRAGENIC RECOMBINATION OF A SINGLE PLANT PATHOGEN GENE
PROVIDES A MECHANISM FOR THE EVOLUTION
OF NEW HOST SPECIFICITIES



Introduction



In gene-for-gene interactions between plants and microbial plant pathogens, host

plant resistance results from the genetic recognition of resistance (R) genes in the plant

and avirulence (avr) genes in the pathogen (Keen 1990; Gabriel and Rolfe 1990). It is

unknown why microbial plant pathogens carry genes that function to limit virulence,

since the majority of these genes are dispensable (Gabriel et al. 1993). For example, to

avoid plant resistance, plant pathogens evade R gene recognition by deletion (De Feyter

et al. 1993), mutation (Kobayashi et al. 1990; Joosten et al. 1994) or transposon

insertion (Kearney et al. 1988) of avr genes. These reports all involve natural

mechanisms whereby avirulence may be lost. Nevertheless, mutational analyses reveal

that gains in avirulence, including new avirulence specificities, are generated in at least

some pathogens at approximately the same frequency as they are lost (Statler 1985).

Although presumably an avr gene may be reactivated by reverse mutation, there are no

reports of natural mechanisms whereby new avirulence specificities or new avr genes are

generated. The mechanisms by which new avr genes evolve are unknown.






62

Xanthomonas citri is the causal agent of citrus canker disease, which occurs

worldwide and is subject to eradication and quarantine regulations in many countries.

A pathogenicity gene, pthA, is essential for the pathogen to cause hyperplastic canker

symptoms on citrus (Swarup et al. 1991). Furthermore, when transferred to other

xanthomonads, pthA confers ability to induce hyperplastic cankers on citrus and a

hypersensitive response (HR) on other hosts (Swarup et al. 1992). Therefore, pthA

exhibits pleiotropic pathogenicity and avirulence functions. Southern hybridization,

restriction analysis and partial DNA sequencing has shown that pthA belongs to a major

avr gene family widespread in the genus Xanthomonas (Swarup et al. 1992). Members

of this gene family include avrBs3, avrBs3-2, and avrBsP (a trancated form of avrBs3-2)

of X. c. pv. vesicatoria (Bonas et al. 1989, 1993; Canteros et al. 1991), avrB4, avrb6,

avrb7, avrBIn, avrBlO1 and avrB102 of X. c. pv. malvacearum (De Feyter and Gabriel

1991; De Feyter et al. 1993), and avrXalO and avrXa7 ofX. oryzae pv. oryzae (Hopkins

et al. 1992). Among these, only avrb6 and pthA are known to have pleiotropic

pathogenicity functions (Swarup et al. 1992; De Feyter and Gabriel 1991; also refer to

Chapter 2). The most striking feature of this gene family is the nearly identical, 102bp

tandem repeats in the central portion of the genes. Deletion analyses have shown that

the avirulence specificity of avrBs3 is determined by the 102bp repetitive motifs of that

gene (Herbers et al. 1992). By swapping the internal repeated regions between pthA and

avrb6, the 102 bp tandem repeats are found to determine both avirulence and pathogenic

specificities (refer to Chapter 2).






63
In most organisms, repetitive regions of DNA are known to be active sites for

homologous recombination (Albertini et al. 1982; Petes and Hill 1988). The high

frequency of such events causes them to be major genome modifying forces in evolution.

Intergenic recombination among members of this avr/pth gene family was proposed to

explain the high frequency of race change mutation found in X. c. pv. malvacearum (De

Feyter et al. 1993). Since both pathogenic and avirulence specificities are determined

by the 102bp tandem repeats, it seemed reasonable to suppose that homologous

recombination between genes and among repeats of the same gene might lead to an

acceleration in the evolution of this Xanthomonas avr/pth gene family, resulting in new

host specificities for the pathogen. In this study, we analyzed the nucleotide sequence

of pthA and demonstrate that intragenic recombination of a single gene can provide a

genetic mechanism for the creation of new avr/pth genes.



Materials and Methods



Bacterial strains. plasmids and matings. Escherichia coli strains DH5a (Gibco-

BRL, Gaithersburg, MD), JM83 (Yanisch-Perron et al. 1985) and derivatives were

grown in Luria-Bertani (LB) medium (Sambrook et al. 1989) at 370C. Xanthomonas

citri strain 3213 (causes hyperplastic cankers on citrus; Swarup et al. 1991, 1992), X.

campestris pv. citrumelo strain 3048 (causes water-soaked leaf spots on citrus and

common bean; Swarup et al. 1991, 1992), X. campestris pv. malvacearum strain

Xcml003 (causes water-soaked leaf spots on cotton; De Feyter and Gabriel 1991) and






64
derivatives were grown in PYGM (peptone-yeast extract-glycerol-MOPS) medium at

300C (De Feyter et al. 1990). Antibiotics were used as described in previous chapters.

Triparental matings were carried out to transfer plasmids from E.coli DH5a to various

Xanthomonas strains by using pRK2013 or pRK2073 as helper plasmids as described

(Swamp et al. 1991; De Feyter and Gabriel 1991a). To transfer plasmids into Xcml003,

the modifier plasmid pUFR054 carrying XcmI and XcmnIII methylase genes was used to

increase the transfer frequency (De Feyter and Gabriel 1991b).

DNA sequence analysis. The 4.1 kb Sall fragment carrying pthA in pZit45

(Swarup et al. 1992) was recloned into the pGEM 7Zf(+) vector (Promega, Madison,

WI) to yield pUFY14.5. Sets of overlapping, unidirectional deletion subclones were

generated in pUFY14.5 using exonuclease HI and mung bean nuclease as described by

Promega (Madison, WI). DNA sequencing was performed by the DNA Sequencing Core

Laboratory of the Interdisplinary Center for Biotechnology Research, University of

Florida, Gainesville. Some sequencing was determined by using the dideoxy nucleotide

chain termination technique with the Amersham (Arlington Heights, IL) multiwell

microtiter plate DNA sequencing system RPN1590. The sequence was analyzed using

the GCG package (version 7.00) by Genetics Computer Group, Inc., Madison, WI.

Gene replacement and selection of marker-evicted strains. Gene pthA on pZit45

was mutated by lighting a 3.9 kb BamHI fragment containing a nptl-sac cartridge (Ried

and Collmer 1987) into a Ball site in one of the 102bp tandem repeats of pthA. One of

the resulting transcriptional fusions, pUFY1.48, carried the nptl-sac cartridge inserted

into repeat #10 of pthA and was selected for further study. An SstI fragment carrying






65
the pthA::nptI-sac fusion from pUFYl.48 was recloned into "suicide" vector pUFR004

(colEl, mobP+, cat+; De Feyter et al. 1990), forming pUFY10.1. Marker-exchange

mutant Xcl.2 (pthA::nptI-sac) was created by transferring pUFR10. 1 to X. citri 3213 and

selecting for colonies resistant to kanamycin (15 pg/ml) and sensitive to chloramphenicol

(35 /g/ml) and sucrose (5%). Marker-evicted strains of X. citri were selected by plating

Xcl.2 on PYGM medium containing 5% sucrose and selecting colonies resistant to

sucrose. Colonies were then screened for sensitivity to kanamycin. Recombinant,

marker-evicted derivatives of pUFY1.48 were selected in X. campestris pv. citrumelo

3048, E. coli DH5a (recA~) and E. coli JM83 (recA+) by plating on PYGM or LB

medium containing 5% sucrose. Sucrose-resistant colonies were screened for sensitivity

to kanamycin.

To calculate recombination frequencies, XcMI.2, 3048/pUFY1.48,

DH5a/pUFY1.48 and JM83/pUFY1.48 were grown in liquid media in the presence of

appropriate antibiotics to densities of 109 cells/ml. Serial dilutions were then plated on

media with and without 5% sucrose. Recombination frequencies were calculated by

dividing the number of sucrose-resistant (also kanamycin-sensitive) colonies by the

number of colonies grown on the medium without sucrose selection (Xu et al. 1988).

Data shown in the results were the mean and standard error of four replicates obtained

from two independent experiments.

Plant inoculations. Citrus (Citrus paradise 'Duncan', grapefruit), common bean

(Phaseolus vulgaris 'California Light Red') and cotton (Gossypium hirsutum L.) plants

were grown under greenhouse conditions. Cotton lines used were Acala-44 (Ac44) and






66
its congenic resistance lines AcB1, AcB2, AcB4, AcB5, Acb6, Acb7, AcBIn and AcBIn3

as described (De Feyter et al. 1993, Swarup et al. 1992). Plant inoculations involving

X. citri or pthA or derivatives ofpthA were carried out in BL-3P level containment (refer

to Federal Register Vol.52, No.154, 1987) at the Division of Plant Industry, Florida

Department of Agriculture, Gainesville. Bacterial suspensions were standardized in

sterile tap water to 108 cfu/ml and pressure infiltrated into the abaxial leaf surface of the

plants. Plant inoculations were repeated at least three times on citrus, and two times on

bean and cotton.



Results



Sequence analysis of pt/A. The complete DNA sequence of pthA was determined

and shown in Fig. 4-1. DNA sequence comparisons revealed that pthA of X. citri is 98%

identical to avrBs3 and avrBs3-2 of X. campestris pv. vesicatoria (Bonas et al. 1989,

1993), 97% identical to avrb6 of X. campestris pv. malvacearum (De Feyter et al. 1993),

and 95% identical to avrXalO of X. oryzae (Hopkins et al. 1992) along its entire length.

As observed with all previously sequenced members of the gene family (Bonas et al.

1993; De Feyter et al. 1993), pthA is flanked by nearly identical 62bp terminal inverted

repeats that precisely define the limits of homology. The predicted amino acid sequence

encoded by pthA is presented in Fig. 4-2. PthA has a calculated molecular weight of

122kD and an isoelectric point of 7.72. The primary region which differentiates the

predicted PthA sequence from the predicted sequences of other members of the family















.- y)
~




0


~o 0
.I~ .~



'b-bc,,
0
.~ 0





1*
o ~





00

0






If~ **~









~


















0 0




> 0




















OK E-
40E
E- 0
13 <
0 0


00[-
@000

00 [
1) C
0 UE-.
000








E5E-'E


u-'Co
000
[-00




00
sa

L)
0 'C
0 E.
U L) E-
U U U





0 0
E. a
0 0

0'C








00-



VE-C
00
00E-
U F <












0E-0
U U E-

0< E0
0 0 E
<-00











0 0
0 0 1
U <


















HE-'E
3 U E-
u0 rl
404























Q 'E-4 Q
U00



E- N



[U E-

Ot- U
O t-'
0'00






0 5-'0
000
[-'00

00E-
0'00
000
000





O- 000
0t-0

[-00

'OBO
000
0 <

S00




'1 N
00 U
<0 EU
U U U E
u <



0 U

< U

U 0U


400
040 M


E-U< 6-E0UU( 00
E- 0 EE-0 0 O 0 <0 0
uu 0 )Cu u 9U L)
0 0 U O UE 2 U O 00 uE-
80~ 00 000F

E-'00 E U-'E.'0'C0
000 00000'C00

OOE- [-000 000 '
'1C' 0000 0'C't-
O6-0 U UE 0oE- 0
OU E 0 00 0U
OE- ED) Sr) 00000 Q 0


O 00 000 0
[- O000'O 0'
00 '0'C00 0 0
t- U0 000 0 0 E
H UE-L)6U< U-O
< 0 QE E- 0 E-[U E
[-00 OI- '0000 E
Du<
000 [-'O E 0000
I 0f- 0: U U U a E- 0
g U -()U E <5 u U [-00 UU0000 0-
E- UUOE E E-0
ou u 0u u E-o3!
t0 0 C- 0'UU0 0 <
EOU 0 0 0 E-E' U0 t- 0
00 000 00EE-U -0<
F. F. 0 DuUO->O 00 aU u 0 UUE.UU uU
00 0E- E- E'UUU- t 0
E-E-' -- E- U 'C0U C0
00 '0'C0 0000t-
00 00000'000
C


000 'C 'CCt''00'0 0
ZL')U< OUOOU u-E- 00000-E--UU U
OE- U < 6E- 0 < < <0 U
L'CO 000 .000 U 00000 E-00 0




'C0'C U 'C u U -E-tE-oo'a o
-[-0U E-'0
0 E- OUt-' E O 0 'C
4000 0t--00U0
[[0 00 00000
000 0t-oo -'Co <() U U L .U 9c ) EDL) L
U U .* <<(- 6uO<6U u
00 E- E- 0'-00-60
0 0 0 0 U-U E-







0[[- OO -'0'-00 0
0-' U'0 LL) C E- 0U00
I., UU00H U0000










UUU UUUUCft-OUU
0000>0'C0000
40000 000'CE''
O'CO [ '0 O 'C0040
000 [-0000 'C000
40 000000'000

E.' t-. 000 00 00000
LOUE U U a U 000
< U E- O<< *UsCo{U<0


0OK uEiOU E- H 0 0
0 o <-E 0 U c- 00
L)E-O u -3 U 0(u 0u L)U< <0 U

H0 U )0 L00 U 90 u
U U0 H U 0 E-

ou u f J Euou 06
E.0u 0oD 00 < 0
V0J U 0 60 U E .0E U

UE-U UU-O EUU UE-U EU 0 0 <

0 6E0 00 000 P
0 U E"U a 0
E- 00 U 0 L)E-UaE

SE-i 0 0UE OU <-000
HE-U6E < 0t 0t< UE-O0
. 0 .UU) UU 0E.0


E- E-; U E E 0 E- 9U 0E
00 U )9 0 000
4 E m OUE 000


E-U<~ < frU-E' E"'U
0QtU ) 0 0 UU 0 u
0 0 000 F 0 EUC u


U t< Ut0 0uu OuU
0 Ll I-'. L)4 u


L UUi, 6 o q ( U <
0 o < E- 0E-'UL) 0 0U E-


000 000000000
0 ~~ 0 0 N0 m
mwiU Wf'O c o0 0C0f
00t O *bU 000 *~


00 E-E- 00U 00-UOE-00 000 U 000UE-E-0
0 0 U00ED EOg- < ~00 <) 0 U0 < -
0 -<- 00 gOE-0 -UO 000 0LE U00<0
06 00 Q0.OE-U OO 0 Qf E QO 0
E 0 0 U 00DE-. 0 0 uN 0 u-0 E- C0 0 '
0 ) 0000 0 000000 u
<6 00Ot3400E0ttS ( U C >U E OUU<(OaDB
000 000~- ~00 00 O 0-
3 00UO 0E0UU' 0004000 00E-
000 .00U0 E S0'C<-.00E-.000E-'C 000
000 000EO00'U0000000'C 000000
UU E-0'C 0 0 C OOOOOOE-.E-E.000O0E-'CO
-0000E
t g 000 ooooC-o00 '0000 0U< OE<
u< uU 300 u00 0000'U 0 0 0E -'
UUOE-U0 0 0< Q0 000UU o u UE00 o 000 B00 0 0 0Uo00 C'0< o0
SOOE-0 E- 000 U E-U E00- ) 0 3E- (u0 0
pE UOUE-g(

1 8o0 C000 00'00000 0 00
o00-. 00 Ua40 UUU 0u-00
0000E.0 0 0000000 0'C00040
FOOEHOUO U u [U o UUU< e E- [" u

HEC 0 0<0 E.0 UE u 0 0 -E- 0 .00uUOE-UO E-
OEOE000 00 < 0000E-E-40004
g 000 0 000oUmE'OU O 0U U00U30U
000000 00000UE-E-E-00 U *u0'CEE00 00 u 0u0000-00000t'C0E
3C0000 000UE- U oO UE0 E00 E-'CE000 E-<



0000-000 0000~00E-000 0000 OE-
000 'C 0 'C E-E.O OOEE-00< 'c 000 ve0
00-000 O E-E-.0 a ooo -o 0e-0 00
S OUU E-UOU 0000E-00-u00 000
cUu UU0 -uo OO OE-UE-UE- U O 00U

uuuouo 6U600U0CHO Co0


080 0800000 00
U00 UOOUOOEuHE-OOEUUH 90E-. E
000U00-00E00000UE-0U 0u-0U00
E-O DO E-'E- 00U OU CE-O0 OE u E00
UE-uo Uo 000000-0O00000'U 00U-

uO 00'UU00000E-OUUE-D0' OE <0
Uo OMO-UE00E-H0' 00 -DOe<00C00

S0 0 U 00 EOU3 0t < E-00


OUU HUO 6-0EUUU00 E 00000 .00 0 E' E-U E-U E-00
E0-O0000 E-0 -0


00 08000@00000080 O
U0E-U00'U00000 -E0~00 0 000E-E
O UE UE-OUE E0000'CUU0000- -0OHE-

0UUO EEOuOCEC 00o0000' 0E-0OU



00&O0000uu0o0000 000E-000
UUUUUIUOU CONEOHOU 0 H H-U UU<<
HHOOUEgH O U UEO
S 0 EO00 0060O' 0<- U-0000
u EEO Ou' E- a 00 0 000 <0u0u





0'C00000 'C000 00 '0 0000 0000'00
U 00 O' < 0'C uu- 00 O E'C 0 uE-E-iE00 00



O0000H0 O 00 0 o0H000'00<00
OU6E-.0OU OU00 O6-. C< O -OUEO04E-*U U0<






00000000 0 00000000400
E COEEOE-.


0 'C0E- 'F E- 'U-O ' u00 00000 H 0o00U< 00 E-E-000


010000000 000 OE-.0 O0000
E D O :04 0ro(-) UQ g

19O-E0-00 00 OE--00 U00'U00
E- 0 UU E- 0 0 0< < E< 0 0 00 00
U 00 00 0'C00 04E-E00'<


0 E U- 0 000u U 0 0E-40 OOE- 3U E-0 'U0
0 0 00E- U 00 O EE- 0 0 'C0 E-E- 00
UE0 0 UO 0 0 0'U00 < I-00 E- 000<< 0 00E-.<00'C0' E-U0E0-
Q0 1- 0 0 L O OL)


000 0000 000 0 00000000
E-OU UU OE 000 O 000' 80U E O
UU^O <:U(OOU








0f 6D E- W 0 0 4 ED 0~ H 0Dr 0 6 e0 () (D r- .' 09 0
E-40UUOH>HONNU<0 HUENN N<-NNN NNt( UUUC m
UE-00 O 0 )U0 < U0 000 0 *104 < u uu L E uE-04
< 0U006 00E0 E 0'U
0 u 0 0 0E.0 U U r000 u E L) 0 E-UE
0 00 0 Q U 0 U 0 0< Q C U U U 0 0 < E- 0 L) U U
< UO L 006 Ca-) 000E. 86UU < a
U< 00 0 u U()0 E0U00E-HUU UE-OU6
u 0 E- 0 0 00 00U
LI ) 0E0 E4 0 uL)0 0 0 E- E- E- E4 E-E- 0
00 0<006 0 0 E- 90 No 000uO EU
1-0 0 E0- u0 u4 40 04 0E0 S 0 -0 400
<< OO U< < U tE- U 0kE r. 0E m 0




00000000000000000000000000
tnOW O ~ -tM ^l^r QDOOowMM TtwOr mQDIT'mem O


























Figure 4-2. Predicted amino acid sequence of PthA. The 17.5 internal tandem repeats
are aligned and numbered for comparison. Letters A through K represent different types
of repeats. indicates the repeats unique for PthA protein. Bold and underlined letters
indicate an amino acid deletion, substitution or specific combinations of amino acid
residues within a given repeat that is unique to PthA (and the combination is not found
in any repeat in any previously published member of the gene family). Three nuclear
localization consensus sequences are double underlined.











MDPIRSRTPS

SPAFSAGSFS


121 MRVAVTAARP

181 EALVGHGFTH

241 LLTVAGELRG


PARELLPGPQ

DLLRQFDPSL

PRAKPAPRRR

AHIVALSQHP

PPLQLDTGQL


PDGVQPTADR

FNTSLFDSLP

AAQPSDASPA

AALGTVAVKY

LKIAKRGGVT


GVSPPAGGPL

PFGAHHTEAA

AQVDLRTLGY

QDMIAALPEA

AVEAVHAWRN


DGLPARRTMS

TGEWDEVQSG

SQQQQEKIKP

THEAIVGVGK

ALTGAPLN


RTRLPSPPAP

LRAADAPPPT

KVRSTVAQHH

QWSGARALEA


LTPEQVVAIA

LTPEQVVAIA

LTPEQWAIA

LTPEQVVAIA

LTPEQVVAIA

LTPEQVVAIA

LTPDQVVAIA

LTPQQVVAIA

LTPEQVVAIA

LTPEQVVAIA

LTPEQVVAIA

LTPEQVVAIA

LTLDQVVAIA

LTPEQVVAIA

LTPDQVVAIA

LTPEQVVAIA

LTPEQVVAIA

LTPEQVVAIA


SNIGGKQALE

SN GGKQALE

SNIGGKQALE

SNIGGKQALE

SNIGGKQALE

SNGGGKQALE

SHDGGKQALE

SNGGGKQALE

SHDGGKQALE

SNGGGKQALE

SNGGGKQALE

SNGGGKQALE

SNGGGKQALE

SNSGGKQALE

SHDGGKQALE

SHDGGKQALE

CNGGGKQALE

SNGGGRPALE


TVQRLLPVLC

TVQRLLPVLC

TVQRLLPVLC

TVQRLLPVLC

TVQALLPVLC

TVQRLLPVLC

TVQRLLPVLC

TVQRLLPVLC

TVQRLLPVLC

TVQRLLPVLC

TVQRLLPVLC

TVQRLLPVLC

TVQRLLPVLC

TVQRLLPVLC

TVQRLLPVLC

TVQRLLPVLC

TVQRLLPVLC

SIVAQLSRPD


A LTNDHLVALA CLGGRPALDA VKKGLPHAPA

AQVVRVLGFF QCDSHPAQAF DDAMTQFGMS RHGLLQLFRR

DRILQASGMK RAKPSPTSTQ TPDQASLHAF ADSLERDLDA

DRAVTGPSAQ QSFEVRAPEQ RDALHLPLSW RVKRPRTSIG

MREQDEDPFA GAADDFPAFN EEELAWLMEL LPQ*


LIKRTNRRIP ERTSHRVADH

VGVTELEARS GTLPPASQRW

PSPTHEGDQR RASSRKRSRS

GGLPDPGTPT AADLAASSTV


QAHG

QAHG

QAHG

QAHG

QAHG

QAHG

QAHG

QAHG

QAHG

QAHG

QAHG

QAHG

QAHG

QAHG

QAHG

QAHG

QAHG

PALA


900

951

1011

1071

1131






71
is in the central portion of the protein, characterized by 17.5 nearly identical, tandemly

arranged, 34 amino acid repeats. As shown in Fig. 4-2, these repeats may be classified

into 11 types (A-K) depending on slight differences in sequence. Among them, repeat

types B, E, H, J are only found in PthA, and not in the predicted peptide sequences of

any other published family member. Repeat 2 (type B) resulted from an amino acid

deletion and repeat 17 (type J) resulted from an amino acid substitution; neither this

deletion nor this substition has been observed in any other member of the gene family.

Repeat types E (repeats 7 and 15) and H (repeat 13) resulted from unique combinations

of amino acids in specific positions. In addition to the presence of unique repeats, PthA

differs from its homologues in the arrangement of repeat types within the region.

Gene replacement and selection of pthA recombinants in X. citri. X. citri mutant

Xcl.2 (pthA::nptI-sac) was generated from wild type strain 3213 by marker exchange,

using pUFY10.1. Marker exchange was confirmed by Southern hybridization analyses

(Fig. 4-3, compare lanes 1 and 2). When inoculated on plants, Xcl.2 was unable to

induce cankers on citrus or to induce an HR on bean (Table 4-1). By plating Xcl.2 on

a sucrose-containing medium and screening for sucrose resistant and kanamycin sensitive

colonies, many marker-evicted strains of X. citri were obtained at a frequency of 5.22

1.8 x 10-5.

To examine the recombination events, total DNAs were extracted from twenty-

four marker-evicted derivatives of Xcl.2, digested with BamHI and probed with the 35P-

labelled internal Stul/HinclI fragment (containing the tandemly repeated region) ofpthA.


















1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18

4Kb-






2Kb-


Figure 4-3. Southern hybridization of BamHI digested, total DNA from X. citri strain
3213 and its pthA mutant derivatives. The blot was probed with the internal 2kb
Stul/HincH fragment (containing the tandemly repeated region) of pthA. Strain 3213
carries four BamHI fragments which hybridize to pthA; the largest fragment, indicated
by the arrow, encodes pthA. Not shown in lane 2 are two additional hybridizing
fragments (1.8 and 1.6 kb in size), resulting from the additional BamHI site in nptl-sac.
Lane 1, 3213; lane 2, Xcl.2 (pthA::nptl-sac); lanes 3-18, nptl-sac-evicted derivatives of
Xcl.2: XcS1, XcS15, XcS19, XcS24, XcS27, XcS30, XcS12, XcS13, XcS17, XcS18,
XcS26, XcS29, XcS32, XcS23, XcS34, XcS26 and XcS21, respectively.










Table 4-1. Phenotypes of Xanthomonas citri marker-evicted strains on citrus and bean.



Host
Strains -----------------------------------------------------
citrus bean


Xc3213 C HR
Xcl.2 0 0
XcS1, XcS9, XcS14 & XcS15 C HR
XcS5 C 0
XcS17 & XcS20 wC wHR
XcS12, XcS13 & XcS24 HR wHR
XcS7 wHR wHR
XcS6 & XcS11 wHR 0
XcS2* 0 0

*: Additional 11 strains (XcS3, XcS4, XcS8, XcS10O, XcS16, Xcsl8, XcS19, XcS20,
XcS21, XcS22 and XcS23); C, canker lesions; wC, weak canker; HR, hypersensitive
reaction; wHR, weak hypersensitive reaction; 0, no symptoms.






74
As shown in Fig. 4-3, the different derivatives of Xcl.2 exhibited different sized bands

(including those similar in size to pthA), indicating that new pthA homologues were

regenerated by homologous recombination among the tandem repeats of pthA. There was

no evidence of recombination between pthA and the other three DNA fragments that

hybridize with pthA, but the presence of the other three fragments raised the possibility

of intergenic recombination.

The twenty-four marker-evicted derivatives of Xcl.2 were inoculated on citrus

and bean plants. They exhibited different plant reaction phenotypes, including some

entirely new phenotypes, distinct from the wild type 3213 strain (Table 4-1). Five

derivatives reverted back to the parental 3213 phenotype, inducing strong canker lesions

on citrus (XcS1, XcS9, XcS14 and XcS15). Two derivatives caused weak canker (XcS17

and XcS20). Six derivatives elicited a previously unreported HR on grapefruit (XcS12,

XcS13 and XcS24). Eleven derivatives caused no pathogenic symptoms. In contrast to

wild type strain 3213, XcS5 retained the ability to elicit hyperplastic cankers on citrus,

but lost the ability to elicit HR on bean, indicating that these two pleiotropic functions

are separately encoded on pthA.

Intragenic recombination of pthA in X. campestris and E. coli. To demonstrate

the occurrence of intragenic recombination, plasmid pUFY 1.48 was introduced into three

bacterial strains which carry no DNA that hybridizes with pthA: X. campestris pv.

citrumelo strain 3048, E. coli strain DH5a (recA-) and E. coli strain JM83 (recA+).

The transconjugants 3048/pUFY1.48, DH5a/pUFY 1.48 or JM83/pUFY1.48 were plated

on sucrose-containing medium and screened for marker-eviction. Again many marker-






75
evicted derivatives were obtained. The frequencies of marker-eviction of pUFY1.48 in

3048, DH5a and JM83 were 4.05 0.44 x 10-4, 1.06 0.13 x 10-2 and 0.97 0.14

x 10-2, respectively.

To examine the recombination events, plasmid DNAs were analyzed from 24

marker-evicted strains from 3048 and 14 strains from DH5a. As shown in Fig. 4-4,

different sized BamHI fragments were observed, demonstrating the generation of new

pthA homologues by intragenic recombination. Southern hybridizations of total DNAs

extracted from these strains probed with the internal Stul/HincIL fragment ofpthA showed

only the presence of the same bands as those generated from the plasmid digestions (data

not shown).

The plasmids derived from the 24 3048 (pCS series) and 14 DH5a (pDS series)

marker evicted strains were individually introduced into X. citri strain B21.2 (a

nonpathogenic, pthA::TnS-gusA derivative of 3213, which is virulent only on citrus),

3048 (virulent on citrus and bean) and X. campestris pv. malvacearum Xcml003 (virulent

only on cotton). The resulting transconjugants were inoculated onto citrus, bean and

cotton plants (Fig. 4-5). Similar to the results observed with derivatives of Xcl.2, these

pthA recombinants exhibited altered phenotypes and/or host specificities. For example,

pCS23 conferred to B21.2 and to 3048 the ability to cause stronger canker lesions on

citrus than that conferred by pZit45 (pthA), and a weak HR often proceeded the canker

symptoms. Unlike pthA on pZit45, however, the recombinant gene on pCS13 conferred

to B21.2 or 3048 the ability to elicit canker lesions on citrus, but not HR on bean.

These results confirmed those indicated by recombinant X. citri mutant XcS5 (refer

























Figure 4-4. A. Intragenic recombinants of pthA generated in X. campestris pv. citrumelo
3048. Plasmid DNAs were digested with BamHl and fractionated in 0.7% agarose gel.
Lane 1, pZit45 (pth,4); lane 2, pUFY1.48 (pthA,::npt-sac); lane 3-19, pCS23, pCS13,
pCS16, pCS2, pCS17, pCS6, pCS1, pCS14, pCS5, pCS7, pCS11, pCS12, pCS15,
pCS19, pCS21, pCS18 and pCS22, respectively. B. Intragenic recombinants of pthA
generated in E. coli. Plasmid DNAs were digested with BamHI and fractionated in 0.7%
agarose gel. Lane 1, pZit45; lane 2, pUFY1.48; lanes 3-16, pDS2, pDS5, pDS4,
pDS13, pDS1, pDS10, pDS14, pDS9, pDS3, pDS6, pDS7, pDS8, pDSll and pDS12,
respectively.














M I 2 3 4 5 6 7 8 9 10111213141516171819


9 kb-


2 kb.


I kb.


M 1 2 3 4 5 6 7 8 9 1011 12 1314 1516


9 kb


4kb-

3 kb-


2 kb-,


1 kb-

















.44-


>bU g4*








00









4- 00
o C4 C









0 lz As



. 0 4) Or












C4.




00
.4J
*0 ac
>,. N
C4 .

042) 0


















0o

8 o



co


n 0 0












C












U'
C1 U







E 8
S> 0 0o CO (0 C) 0 0 0 0 on
a. a. a a. Q. a. a. a. a Q. Q. aQ a.







80
above): the ability to confer hyperplastic cankers on citrus and the HR on bean are

independently encoded on pthA.

Plasmids pCS1 and pCS6 conferred ability to elicit an HR on citrus and bean to

B21.2 and 3048, and ability to elicit an HR on 9 tested cotton lines (Ac44, AcB1, AcB2,

AcB4, AcB5, Acb6, Acb7, AcBIn and AcBIn3) to Xcml003. All pthA derivatives tested

which conferred the ability to induce HR to Xcml003 on normally susceptible cotton line

Ac44, also elicited HR on all tested cotton resistance lines. All pthA derivatives tested

were able to confer ability to induce an HR to Xcm1003 on one or more resistance cotton

lines, indicating that all pthA derivatives tested were functional genes.

Most of the variation within the predicted 34 amino acid repeat sequence of pthA

occurred at positions 12 and 13, and most repeats encoded either asparagine/glycine,

asparagine/isoleucine or histidine/aspartate at these positions. The latter two amino acid

combinations in the pthA repeats could be detected at the DNA level by presence of SspI

or BglI restriction sites, respectively. The SspI restriction enzyme cut specifically after

the 36th nucleotide of the 1st, 3rd, 4th and 5th repeats of pthA, while the BglI enzyme

cut specifically after the 38th nucleotide of the 7th, 9th, 15th, 16th repeats. By

comparing the BgII and SspI fragments of intragenic recombinants with those of pthA,

the specific repeats deleted or affected in 12 recombinants were determined (Fig. 4-5).

Deletions involving the 7th, 8th, and 9th repeats always reduced or abolished the canker-

eliciting ability of pthA (pCS16, pCS17, pDS5, pCS1, pCS6, pCS11, pCS22). Deletions

involving the 4th, 5th, and 6th repeats abolished all avirulence specificity; these

derivatives conferred ability to induce an HR on all citrus, bean and cotton lines tested.






81
Although recombinants with deletions in the 1st, 2nd, 16th or 17th repeats were not

obtained, these results indicated that the specific order of the first nine repeats of pthA

was important for ability to elicit cankers on citrus and for specificity of avirulence.

Deletions of the 10th through the 15th repeats appeared to alter avirulence specificity.



Discussion



The complete DNA sequence of pthA revealed that pthA is highly homologous to

avrb6, avrBs3, avrBs3-2 and avrXalO along its entire length. The pthA gene encoded

a protein of 1163 amino acids with calculated molecular weight of 122 kD. Western blot

analyses similar to those previously published for AvrBs3 (Knoop et al. 1991), but using

anti-PthA serum, revealed that a 122 kD protein was constitutively expressed in X. citri

and X. campestris pv. citrumelo transconjugants containing pthA (refer to Chapter 5).

Despite few differences in the predicted amino acid sequences of all members of the

avr/pth gene family, pthA has a unique ability to induce cell divisions on citrus leaf

mesophyll in addition to its pleiotropic ability to confer avirulence to pathogens of cotton

and bean (Swarup et al. 1992). Both functions were previously found to be localized to

the repeated region of pthA (refer to Chapter 2).

DNA sequence analysis revealed that there were five specific repeats only found

in pthA, but not in other published members of the avr/pth gene family sequenced to date

(avrb6, avrBs3, avrBs3-2 and avrXalO). By analyzing 12 intragenic recombinants with

BglI and SspI restriction enzyme digestion, we were able to approximately locate the






82
regions important for avirulence and pathogenic specificities. The first six repeats of

pthA appear to be essential for the pathogenicity function and any specificity (avirulence

or pathogenicity) at all. Deletions involving repeats 7 to 9 reduced or abolished canker-

eliciting ability of pthA on citrus. Deletions involving repeats 10 to 16 determined

avirulence specificity. These findings may also help to explain previously published data

on avrBs3 deletion derivatives (Herbers et al. 1992; Bonas et al. 1993). From the

published data, all deletions affecting repeats 1 through 6 of avrBs3 abolished all

specificity, although not avirulence per se.

The role of the tandem repeats in determining pathogenic or avirulence specificity

is striking, but it is not known how the repeats determine these functions. One of the

important biological functions for repetitive domains in prokaryotic and eukaryotic

proteins is ligand binding. In a family of clostridial and streptococcal ligand-binding

protein, for example, conserved C-terminal repeat sequences function as binding sites

(Wren 1991). Similarly, 34 amino acid repetitive motifs called tetratricopeptides (TPR)

found in proteins encoded by many mitotic genes (e.g., CDC16, CDC23, nuc2+ of

yeast, BimA of Aspergillus) are implicated to pair with WD-40 repeats found in many

proteins including the P-subunit of G-proteins and some transcriptional factors (Goebl

and Yanagida 1991; van der Voom and Ploegh 1992). It is possible that the repetitive

domain of Pth or Avr proteins may also function as a ligand binding site, specifically

interacting with corresponding plant receptors.

In this study, a nptl-sac cartridge was inserted into the tandemly repeated region

of pthA to serve as a selectable marker to detect homologous recombination between






83
intragenic repeats. Since the production of levansucrase encoded by sac is lethal to gram-

negative bacteria in the presence of sucrose, marker evicted strains can be selected by

the nptl-sac marker (Ried and Collmer 1987). The regeneration of functional genes from

the pthA::nptl-sac fragment on pUFY1.48 in 3048 or DH5a, which contains no DNA

hybridizing to pthA, demonstrated that intragenic recombination occurred. This was

confirmed by restriction analysis of internal tandem repeats of recombinants.

Intragenic recombinants of pthA were generated by two possible mechanisms.

The first mechanism involves a single crossover event among intragenic repeats on the

same plasmid, resulting in the deletion of nptl-sac marker and restoration of a functional

gene. The second mechanism requires a double crossover between plasmids. The

frequency of pthA recombination in E. coli was observed to be recA independent. Strain

DH5a (recA) yielded recombinants at a frequency of 1.06 0.13 x 10-2, while strain

JM83 (recA+) yielded recombinants at a frequency of 0.97 0.14 x 10-2. Since

recombination involving small repeated sequences (less than 1kb) is affected only slightly

by recA mutations (Petes and Hill 1988), recombination of pthA was expected to be recA

independent.

The intragenic recombination frequencies of pthA observed in the E. coli were

higher than those observed in X. campestris pv. citrumelo (4.05 0.44 x 10-4) and X.

citri (5.22 1.8 x 10-5). This may be due to differences in the copy number of

plasmids in which intragenic recombination occurred. Recombination of pthA in X. citri

strain Xcl.2 involved a native plasmid, while recombination in E. coli and X. campestris

pv. citrumelo involved pUFY1.48, which appeared (based on our plasmid DNA






84
extractions) to replicate at a higher copy number in E. coli than in Xanthomonas.

However, since we did not determine plasmid copy number, these results may be due to

different genetic backgrounds in each strain.

In animal pathogens, some pathogenicity genes encode surface proteins with a

distinctive arrangement of tandem repeats which may provide a protective, strain-specific

conformational epitope for evasion of host immunity (Hoft et al. 1989; McConkey et al.

1990; Hollingshead et al. 1987). Intragenic recombination between homologous repeats

has been found to be responsible for antigenic variations in the pathogens such as

Streptococcus, Neisseria, Salmonella, Plasmodiwn and Trypanosoma (McConkey et al.

1990; Hollingshead et al. 1987; Hagblom et al. 1985; Frankel et al. 1989). The

resulting antigenic variability allows pathogens to escape the host immune response. In

this study, we demonstrated that intragenic recombination of a single plant bacterial gene

can: 1) alter host specificity, 2) generate new pathogenic phenotypes, and 3) evade host

plant defenses. The evolutionary success of many plant pathogens may rely on their

ability to avoid host recognition. Tandemly repeated motifs in pthA and other members

of the gene family provide hot spots for recombination, and thus may accelerate evolution

of this gene family. Intragenic recombination between inexact repeats not only

introduces genetic variation, but also has the potential to amplify mutagenic effects. As

shown in the results, homologous recombination can lead to either duplication or deletion

of repeat blocks. A mechanism to generate genetic variation at high frequency may also

have the negative side effect of generating unnecessary, gratuitous avr genes.






85
Previously reported natural mechanisms for race change included deletions (De

Feyter et al. 1993), point mutations (Kobayashi et al. 1990; Joosten et al. 1994) or

transposon inactivation (Kearney et al. 1988) of avr genes and always resulted in a loss

of avirulence on a specific host plant, but not gains of specific avirulence. Because of

the unique 102bp tandem repeats and their role in determining both pathogenic and

avirulence specificities, intragenic recombination of pthA not only caused the loss of

pathogenic or avirulence phenotypes on a specific host, but also simultaneously resulted

in the gain of new pathogenic phenotypes and avirulence specificities. Many intragenic

recombinants of pthA lost the ability to confer canker-like lesions on citrus to 3048, but

conferred a new phenotype by reducing the watersoaking symptoms of 3048. Some

intragenic recombinants lost the ability to confer cankers on citrus and instead elicited

an HR on citrus, indicating that normsensitive (canker) and hypersensitive reactions may

share a very similar signal transduction pathway as initially suggested by Klement (1982).

Other recombinants lost the ability to elicit HR on bean, but retained the ability to induce

cankers on citrus. Therefore, the pleiotropic pathogenic and avirulence functions ofpthA

were separated by the rearrangement of tandem repeats through intragenic recombination.

Members of this avr/pth gene family have been found in many Xanthomonas

species and pathvars. When present in appropriate xanthomonads, many are able to elicit

an HR on diverse plant species including cotton, citrus, bean, tomato, pepper and rice.

As shown in Fig. 4-5, one intragenic recombinant of pthA (pCS6) has the ability to elicit

a strong HR on plants from taxonomically distinct plant families such as Rutaceae

(citrus), Legwuminosae (bean) and Malvaceae (cotton), suggesting that the corresponding






86
resistance genes may be widespread in different plant species. Tomato disease resistance

gene Pto, which encodes a protein kinase, hybridizes with multiple DNA fragments

found in a variety of different plants, indicating the presence of Pto homologues (Martin

et al. 1993). During the evolutionary process, members of the avr/pth gene family might

have moved horizontally among diverse xanthomonads pathogenic to a wide range of

plants (Gabriel et al. 1993; De Feyter et al. 1993), thereby enabling interactions with

members of the same plant gene family in a wide range of plants. As a result, members

of this avr/pth gene family may have accelerated the evolution and diversity of the

corresponding plant R gene family.

The gene-for-gene concept is a major model for research on the reciprocal

evolution between plants and parasites (Thompson and Burdon 1992). In the pathogen,

avirulence genes may be mutated to overcome resistance by evading host recognition.

In the plant, resistance genes may also evolve to recognize new avirulence genes,

resulting in new resistance genes. Recent genetic mapping data have shown that disease

resistance loci (especially those corresponding to different races of the same pathogen)

are often clustered within small genetic intervals, suggesting that intralocus recombination

may be responsible for generating resistance specific for new races of plant pathogens

(Dickinson et al. 1993; Martin et al. 1993; Pryor 1987). Intragenic recombination of

pthA (a member of an avr gene family) demonstrated in this research provides compelling

evidence that intragenic and intralocus recombination may play an important role during

the reciprocal evolution of gene-for-gene interactions.











CHAPTER 5
PLANT NUCLEAR TARGETING SIGNALS ENCODED BY A FAMILY
XANTHOMONAS AVIRULENCE/PATHOGENICITY GENES



Introduction



Interactions of microbial plant pathogens carrying avirulence (avr) genes with host

plants carrying specific resistance (R) genes results in plant defense responses, often

observed as a hypersensitive reaction (HR) that is characterized by rapid plant cell death

at the site of infection. Several models have been proposed to explain the genetically

specific recognition between microbial avr genes and plant R genes in these gene-for-gene

interaction (Gabriel and Rolfe 1990), and different models may be applicable to different

systems. In the elicitor/receptor model, avr genes encode a low molecular weight signal

molecule elicitorr) which is perceived by a sensor encoded by the corresponding R genes

and results in an HR. In Pseudomonas syringae pv. tomato, avrD encodes an enzyme

involved in the synthesis of a low molecular weight glycolipid elicitor (Midland et al.

1993; Smith et al. 1993). This elicitor induces an HR specifically on soybean plants

containing the resistance gene Rpg4 (Keen et al. 1990; Keen and Buzzel 1991; Kobayashi

et al. 1990). An alternative model is the dimer hypothesis in which the protein product

of an avr gene interacts directly with the protein product of a plant R gene or with the

R gene itself (Ellingboe 1982). The extracellular peptide encoded by avr4 and avr9 of

87






88
the fungal pathogen Cladosporium fulvum can directly induce a hypersensitive response

(HR) on tomato cultivars carrying the resistance gene Cf4 or CJ9 (Joosten et al. 1994;

van Kan et al. 1991). However, other than these three examples, the microbial signal

molecules that determine avirulence remain unidentified. In all cases, it is unknown how

the signals encoded by avr genes are specifically recognized by host cells with R genes,

triggering plant defense responses.

The first plant R gene, isolated from tomato, was shown to encode a

serine/threonine protein kinase that is homologous to the receptor protein kinase involved

in Brassica pollen-stigma recognition and to the Raf protein kinase involved in the

mammalian Ras signaling pathway (Martin et al. 1993). Since transcriptional activation

of plant defense genes is modulated by phosphorylation (Felix et al. 1991; Yu et al.

1993), and can be blocked by inhibitors of mammalian protein kinases (Raz and Fluhr

1993), it was proposed that recognition of avr signals by R gene products triggers

phosphorylation cascades, leading to plant defense responses (Lamb 1994).

In mammalian systems, two nearly complete signal transduction pathways used

by growth factors and/or cytokines have been elucidated (Culotta and Koshland 1993).

The best known is the Ras pathway, centered around the protein product of the prototype

oncogene ras (Hall 1993, Schlessinger 1993). Upon binding to extracellular signals, the

activated receptor protein kinase transduces signals to the GRB2, SOS, Ras and Raf

proteins. Raf (a Pto homolog) then initiates the mitogen-activated protein (MAP) kinase

cascade (Crews and Erikson 1993; Nishida and Gotoh 1993), which brings protein

kinases into the nucleus, activating nuclear transcriptional factors. The other route from






89
membrane to nucleus is more direct. In these cases, tyrosine kinases pick up signals

from receptors in the membrane and activate transcription factor subunits in the

cytoplasm, which are transported into the nucleus and induce transcription (Hunter 1993,

Kishimoto et al. 1994). In both cases, activated protein kinases and/or transcriptional

factors need to enter the nucleus in order to induce gene expression which may lead to

different physiological outcomes, including oncogenesis and apoptosis.

Viral pathogens encode some transcriptional factors that can directly enter the

nucleus, modulate gene expression of host cells. The bacterial plant pathogen

Agrobacteriwn tumafaciens also encodes several proteins targeted to the plant cell nucleus

(Citovsky and Zambryski 1993). Transport of proteins into the nucleus is an active

process and requires that proteins contain suitable nuclear localization sequences (NLSs,

Nigg et al 1991). NLSs are usually short stretches of 8-10 amino acids characterized by

basic amino acid residues and proline. NLSs are retained in the mature protein, may

be located at any position as long as they exposed on the protein surface, and can be

present in multiple copies. NLSs are recognized by the receptor, called NLS-binding

proteins, which direct NLS-containing proteins into nucleus through nuclear pores (Silver

1991). NLSs from different organisms, including the prokaryote Agrobacteriwn, are

functional throughout the eukaryotic kingdom. Most of the characterized NLS-containing

proteins were reported from mammalian virus and animal cells. Many are oncogene

products, hormone receptors and transcriptional factors, including those involving later

steps of the Ras signaling pathway such as the transcriptional factor AP-1 (Jun/Fos) (Kerr

et al. 1992). A few NLS-containing proteins from Agrobacterium, plant and plant virus






90
have also been characterized (Citovsky and Zambryski 1993; Restrepo et al. 1990; van

der Krol and Chua 1991; Varagona et al., 1991, 1992). To the best of my knowledge,

however, NLSs encoded by avr genes of plant pathogens have not been reported.

Recently, a large family of avirulence (avr)/pathogenicity (pth) genes has been

identified in many species and pathovars of Xanthomonas, a major group of plant

bacterial pathogens. Members includes avrBs3, avrBs3-2 and avrBsP of X. campestris

pv. vesicatoria (Bonas et al. 1989, 1993; Canteros et al. 1991), avrB4,avrb6, avrb7,

avrBIn, avrBlOl, and avrB102 of X. c. pv. malvacearum (De Feyter and Gabriel 1991;

De Feyter et al. 1993), pthA of X. citri (Swarup et al 1991, 1992), and avrxa5, avrXa7

and avrXalO of X. oryzae pv. oryzae (Hopkin et al. 1992). This avr/pth gene family

comprises the majority of described Xanthomonas avr genes and constitutes a large

portion of all avr genes cloned to date (Gabriel et al 1993). The pthA gene of X. citri

and avrb6 of X. c. pv. malvacearum are particularly intriguing because of their

pathogenicity functions (Swarup et al. 1992, Chapter 2). Gene pthA is essential for X.

citri to induce cell division specifically on citrus plants leading to hyperplastic canker

lesions. Gene avrb6 is important for X. c. pv. malvacearum to induce watersoaking

symptoms specifically on cotton. These two genes also function pleiotropically as avr

genes.

Despite the diverse avirulence and pathogenic specificities encoded by these

avr/pth genes, their DNA sequences are nearly identical (95-98%) (De Feyter et al,

1993, Chapter 4). Both pathogenic and avirulence specificities of the avr/pth genes are

determined by the nearly identical, 34 amino acid tandem repeats encoded in the central






91
region of the genes (Herbers et al. 1992; Chapter 2). To account for the precise

recognition and diverse specificities, it has been speculated that the avr/pth protein

product may directly mediate the plant hypersensitive response (Herbers et al. 1992), as

initially suggested in the dimer hypothesis.

In this study, we demonstrated that pthA conferred to X. campestris pv.

malvacearum (which is nonpathogenic on citrus) the ability to elicit hyperplastic canker

symptoms on citrus, independently of bacterial growth in plant. Although the

proteinaceous elicitors encoded by pthA or avrb6 were not detected three nuclear

localization sequences were identified to be encoded by all members of the Xanthomonas

avr/pth gene family sequenced to date. Furthermore, the C-termini of PthA and Avrb6,

containing the NLSs were shown to be capable of directing P-glucuronidase into the

nuclei of plant cells.



Materials and Methods



Bacterial strains and plasmids. Escherichia coli strains used were DH5a (Gibco-

BRL, Gaithersburg, MD) and BL21(DE3)pLysS (Novagen, Madison, WI). Xanthomonas

strains used were X. citri 3213 (causes hyperplastic cankers on citrus, Swarup et al.

1991, 1992), its marker exchange mutant Xcl.2 (pthA::npt-sac, causes no symptom on

citrus), X. campestris pv. citrumelo 3048 (causes water-soaked leaf spots on citrus and

common bean, Swarup et al. 1991, 1992), X. campestris pv. malvacearum XcmH1005

(causes water-soaked leaf spots on cotton) and its mutants XcmH1407 (avrb6::Tn5-gusA,






92
refer to Chapter 2), HM2.2S (seven avr genes deleted, refer to Chapter 3). E. coli

strains were grown in Luria-Bertani (LB) medium (Sambrook et al. 1989) at 370C or

30C. Xanthomonas strains were grown in PYGM (peptone-yeast extract-glycerol-

MOPS) medium at 300C as described (De Feyter et al. 1990). Plasmids pUC19

(Yanisch-Perron et al. 1985) and pUFRO47 (De Feyter et al. 1993) were used as cloning

vectors. Plasmid pUFR127 (De Feyter and Gabriel 1991a) contains avrb6 on a 5-kb

DNA fragment. Plasmid pZit45 (Swarup et al. 1992) contains pthA on a 4.5-kb DNA

fragment. Plasmid pUFY20 contains a chimeric gene consisting of the 5' and 3' regions

of avrb6 and internal repeated region of pthA (refer to Chapter 2). Plasmids pCS6 and

pCS23 contain intragenic recombinants of pthA (refer to Chapter 4). Triparental

matings were carried out to transfer plasmids from E. coli DH5a to Xanthomonas strains

by using pRK2013 or pRK2073 as helper plasmids (Swarup et al. 1991, De Feyter and

Gabriel 1991a). To transfer plasmids into X. campestris pv. malvacearum strains, the

modifier plasmid pUFY054 carrying XcmI and XcmIII methylase genes was used to

increase the transfer frequency (De Feyter and Gabriel 1991b).

Preparation of potential elicitors and plant inoculations. To detect elicitor

activity, crude lysates were prepared as described (He et al. 1993) from Xanthomonas

and E. coli cells carrying pthA or its derivatives. Basically, bacterial cultures were

grown overnight, harvested by centrifugation and resuspended in 10mM Tris-HCl

(pH8.0) solution at an ODr00 of 0.5-1. Bacterial cells were lysed by lysozyme

treatment (2mg/ml lysozyme in 10mM Tris-HCl,pH8.0) or disrupted by sonication in the

presence of 1 mM PMSF (phenylmethylsulfonyl fluoride). PthA protein was also






93
overexpressed in E. coli cells and purified by affinity chromatography (described later).

Cell-free suspensions (in 10mM Tris-HCl, pH8.0) or affinity column purified PthA were

immediately infiltrated into leaves of citrus and cotton plants.

Citrus (Citrus paradise 'Duncan', grapefruit) and cotton (Gossypium hirsutum)

plants were grown under greenhouse conditions. Cotton lines used were Acala-44 (Ac44)

and its congenic resistance lines Acb5b, Acb6, AcBIn3 as described (De Feyter et al.

1993; Swamp et al. 1992). X. campestris pv. malvacearum transconjugants carrying

pthA or its derivatives were inoculated into leaves of citrus plants in BL-3P level

containment (refer to Federal Register Vol.52, No. 154,1987) at the Division of Plant

Industry, Florida Department of Agriculture, Gainesville.

Preparation of antisera and immunoblotting. A 3.7kb BamHI fragment carrying

intact pthA was generated from partial digestion of pZit45 (Swarup et al. 1992), and

cloned into expression vector pET-19b (Novagen, Madison, WI). The fusion sites of the

resulting construct pUFY50.13 were determined by DNA sequence analysis. PthA protein

was overexpressed in E. coli strain BL21(DE3)/pLysS containing pUFY50.13 in the

presence of 1mM IPTG (isopropyl-P-D-thiogalactopyranoside). Since proteins expressed

on pET-19b are designed to contain 10 histidine residues at the amino terminus as an

affinity handle, overexpressed PthA was readily purified by His-tag affinity

chromatography (Novagen, Madison, WI). For immunization of rabbits, PthA was

further purified by SDS-polyacrylamide gel electrophoresis. Antiserum against PthA was

prepared from a New Zealand rabbit by Cocalico Biologicals, Inc. (Reamstown, PA).

For immunoblotting, E. coli or Xanthomonas cells were harvested by centrifugation and






94
resuspended in sample buffer (50 mM Tris-HC1, pH6.8, 10% glycerol, 2% SDS, 0.1%

bromophenol blue). After samples were boiled for 2 min and centrifuged for 5 min, they

were separated on 8% SDS-polyacrylamide gels and transferred to nitrocellulose by

electroblotting (Towbin et al. 1979). Western blots were probed with anti-PthA antibody

and visualized with goat anti-rabbit antibody conjugated with alkaline phosphatase.

DNA constructs for GUS fusion proteins. The DNA coding sequences for the C-

terminal region (191 amino acids) of PthA and Avrb6 were amplified by PCR from the

plasmid pZit45 (Swarup et al. 1992) by using the following primers:

1. CTCTAGAGCCATGACGCAGTI'C; 2. CAGATCTCTGAGGCAATAGCTC.

The amplified fragments were first cloned into pGEM-T (Promega Inc. Madison,

WI), then cut out with Xbal/BglII and recloned into the XbaI/BamHI site of plant

expression vector pBI221 (Clonetech, CA). The resulting plasmids pUFY082 and

pUFY083 contain the chimeric gene with the C-terminal coding sequence of PthA and

Avrb6, respectively, translationally fused with 5' end of GUS gene. Both pUFY082 and

pUFY083 were checked by sequence analysis at the fusion site.

Onion transformation system. The Helium Biolistic gene transformation system

(Du Pont) was used to transform epidermal cell layers of white onion as described

(Varagona et al. 1992). Inner epidermal layers were peeled and placed inside up on

Petri dishes containing MS (Murashige and Skoog 1962) basal media (4.4 g/L MS salt

mixture, Sigma, M5519, 30 g/L sucrose, pH 5.7) with 2.5 mg/L amphotericin B (Sigma)

and 6% agar. DNA samples were prepared as described (Taylor and Vasil 1991). Five

jug of column-purified (Magic Prep, Promega) plasmid DNA was precipitated onto 1 mg






95
of 1.6 pim gold particles using 50 4L of 2.5 M CaC!2 and 10 ,L of 100 mM spermidine.

DNA-coated gold particles were washed with 50 pd of 70% and 100% ethanol and finally

resuspended in 60 /L of 100% ethanol. The samples were dispersed by submerging in

a waterbath sonicator for a few seconds. Five pL samples were pipetted onto particle

delivery discs and delivered into onion epidermal cell layers at pressures of 1100 or 1300

p.s.i.. After microprojectile bombardment, Petri dishes were sealed with parafilm and

incubated overnight at 280C in the dark. The assays for nuclear localization activity of

the fusion protein were repeated in three independent transformation tests.

Histochemical analysis. The histochemical GUS assay was used to determine the

location of GUS fusion proteins in onion cells as described previously (Jefferson, 1987,

Varagona et al 1992). Onion epidermal layers were incubated at room temperature in

X-glu solution (50 mM phosphate buffer, pH7.0, ImM EDTA, 0.001% Triton X-100.

0.05mM sodium ferricyanide and ferrocyanide, 2mM X-glu). After detection of blue

color, onion epidermal layers were mounted on a glass slide with a solution containing

20/tg/ml of the nucleus-specific dye 4',6-diamidino-2-phenylindole (DAPI, 0.1 X

phosphate buffer, 10mM sodium azide, 90% glycerol). Cellular location of the blue

indigo dye produced by oxidative dimerization of the GUS product was determined under

bright-field optics using a Zeiss Axiophot microscope and compared with the location of

DAPI-stained nuclei under fluorescence optics.




Full Text

MOLECULAR MECHANISMS OF HOST SPECIFIC VIRULENCE
AND AVIRULENCE CAUSED BY COTTON BLIGHT
AND CITRUS CANKER PATHOGENS
By
YINONG YANG
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

ACKNOWLEDGMENTS
I wish to express my appreciation and gratitude to Dr. Dean W. Gabriel,
chairman of the supervisory committee, for his time, guidance and support during the
course of this study. Appreciation is also extended to Drs. D. R. Pring, H. C. Kistler,
L. C. Hannah, and R. A. Jensen for serving on the supervisory committee and critical
reading of this manuscript. Special thanks go to Dr. W. B. Gurley for reading my
dissertation and participating in my final examination.
I wish to thank Drs. Robert De Feyter, Mark T. Kingsley and Mr. Gary Marlow
for stimulating scientific discussions and technical assistance. I gratefully acknowledge
the help of Dr. Michael Paddy in fluorescence microscopy and Dr. Suzy Cocciolone in
microprojectile bombardment. I also greatly appreciate the technical assistance of
Qiaoping Yuan, Ruhui Li and Blanca Garagorry.
I would like to thank my parents for their understanding and encouragement.
Finally, I must thank my wife, Qin Wang, and daughter, Catherine, for their
companionship and strong support.
u

TABLE OF CONTENTS
ACKNOWLEDGMENTS ü
ABSTRACT iv
CHAPTERS
1.INTRODUCTION 1
2. HOST-SPECIFIC SYMPTOMS AND INCREASED
RELEASE OF XANTHOMONAS CITR1 AND
X. CAMPESTR1S PV. MALVACEARUM
FROM LEAVES ARE DETERMINED BY
THE 102 BP TANDEM REPEATS OF
pthA AND avrb6, RESPECTIVELY 9
3. INACTIVATION OF MULTIPLE A VIRULENCE
GENES IN XANTHOMONAS CAMPESTR1S
PV. MALVACEARUM CREATED A
NONPATHOGENIC ENDOPHYTE
OF COTTON 37
4. INTRAGENIC RECOMBINATION OF A SINGLE
PLANT PATHOGEN GENE PROVIDES A
MECHANISM FOR THE EVOLUTION
OF NEW HOST SPECIFICITIES 61
5. PLANT NUCLEAR TARGETING SIGNALS ENCODED
BY A FAMILY OF XANTHOMONAS
A VIRULENCE/PATHOGENICITY GENES 87
6. SUMMARY AND CONCLUSIONS Ill
LITERATURE CITED 115
BIOGRAPHICAL SKETCH 129
in

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
MOLECULAR MECHANISMS OF HOST-SPECIFIC VIRULENCE
AND AVIRULENCE CAUSED BY COTTON BLIGHT
AND CITRUS CANKER PATHOGENS
By
Yinong Yang
August 1994
Chairman: Dean W. Gabriel
Major Department: Plant Molecular and Cellular Biology
Several members of a Xanthomonas avirulence (avr) gene family were shown to
play an important role in determining host-specific pathogenicity (pth) of cotton blight
and citrus canker pathogens. X. campestris pv. malvacearum strain XcmH1005, the
cotton blight pathogen, carries twelve members of what is now known as an avr/pth gene
family. Six plasmid-borne members were previously shown to confer avirulence on
cotton in a gene-for-genes manner. Four out of six newly cloned chromosomal members
also conferred avirulence on cotton. To investigate possible pleiotropic functions of these
avr genes in XcmH1005, marker exchange mutagenesis and complementation analyses
were carried out. Gene avrb6 was found to be a major pth gene of XcmH1005 and was
required for severe pathogenic symptoms on cotton. Other avr genes in XcmH1005 also
had an additive effect on pathogenicity. These avr genes did not contribute to bacterial
IV

growth in planta, but were essential for the induction of host-specific symptoms and
release of the pathogen onto the leaf surface, strongly indicating their role in pathogen
dispersal. Another member of the same gene family, pthA of X. citri, was previously
shown to be essential for symptoms of citrus canker disease and release of the pathogen
to the leaf surface. By constructing chimeric genes among pthA and six other avr/pth
genes, both avirulence and pathogenic specificities were found to be determined by the
102 bp tandem repeats present in the central region of all members of the gene family.
Furthermore, intragenic recombination was demonstrated to provide a mechanism for the
generation of new avr/pth genes with novel host specificities. Sequence analysis revealed
that pthA and avrb6 encoded proteins containing three nuclear localization sequences in
the C-terminal region. The predicted PthA and Avrb6 proteins were constitutively
expressed in Xanthomonas as detected by Western analysis using anti-PthA antibody.
When transiently expressed in plant cells, p-glucuronidase fusion proteins containing the
C-terminal region of PthA or Avrb6 were observed to localize specifically in the nucleus.
A working hypothesis is proposed that Avr/Pth proteins enter the plant cell nucleus and
directly affect gene expression, leading to the induction of host-specific symptoms.
v

CHAPTER 1
INTRODUCTION
Disease resistance of plants and virulence of microbial pathogens are governed
by their combined genotypes. During the process of reciprocal evolution, many plant
and microbial genes have coevolved to actively engage in pathogenic interactions between
host and pathogen, thus determine the final outcome of plant diseases. Successful
infection by microbial pathogens may require surface attachment, degradation of host
chemical and physical barriers, production of toxins, and inactivation of plant defenses
(Lamb et al. 1989). Overall, up to 100 genes may be required for bacterial pathogenesis
(Panopoulos and Peet 1985, Daniels et al. 1988), including those involved in bacterial
growth in planta, induction of pathogenic symptoms, host range and avirulence (Gabriel
1986). On the other hand, host plant resistance to infection involves preformed barriers,
inactivation of microbial toxins, recognition of pathogen signals and induction of active
plant defenses. In incompatible plant-bacteria interaction, induced resistance is usually
associated with a hypersensitive response (HR), characterized by the rapid, localized
death of plant cells at the site of infection. This process requires activation of plant
resistance genes involving signal perception and transduction and plant defense genes
involving biosynthesis of pathogenesis-related proteins, phytoalexins and other antibiotics
(Dixon and Harrison 1990, Dixon and Lamb 1990).
1

2
Bacterial plant pathogens require virulence genes to infect and to colonize plants.
A large group of genes, called hypersensitive response and pathogenicity {hrp) genes,
have been found generally conserved in all phytopathogenic bacteria examined (e. g.,
Erwinia, Pseudomonas and Xanthomonas) (Willis et al. 1991). Whatever the bacterial
pathogens studied, typical hrp- mutants are completely nonpathogenic and elicit neither
pathogenic symptoms on host plants nor hypersensitive response on nonhost plants. Most
hrp genes are clustered on 20-30 kb DNA fragments and are conserved at the genus or
family level. High levels of DNA homology are found among many hrp genes from
different genera of phytobacteria (Boucher et al. 1992). Several biological functions are
encoded by different hrp genes. Some hrp genes (e.g., hrpS of P. syringae pv.
phaseolicola) are clearly regulatory, responding to plant signals to induce transcription
of hrp operon in a manner that may be analogous to the two-component signaling systems
(Rahme et al. 1991). Other hrp genes are involved in the synthesis and export of
pathogenicity factors that also induce an HR on nonhost plants. DNA sequence analysis
reveals that several hrp genes of Pseudomonas and Xanthomonas are remarkably similar
to key virulence genes of animal bacteria such as Yersinia and Shigella, which are
involved in export of proteinaceous pathogenicity factors (Gough et al. 1992, 1993;
Fenselau et al. 1992). Some hrp genes of Erwinia and Pseudomonas were shown to
encode proteinaceous HR elicitors that are secreted via Hrp-dependent pathway (Wei et
al. 1992; He et al. 1993). Since HR on host plants (race specificity) conferred by
avirulence (avr) genes is dependent on hrp function, it was proposed that Hrp proteins

3
may also involve export of avirulence elicitors (Fenselau et al. 1992). Therefore, hrp
genes may control or condition the functions encoded by virulence and avirulence genes.
Bacteria belonging to the genus Xanthomonas are plant-associated and infect 392
monocot and dicot species in 68 families (Leyns et al. 1984). Although the genus as a
whole has wide host range, individual strains have host ranges limited to only certain
plant species. This phenomenon of host specialization indicates that establishment of
Xanthomonas infection involves host-specific recognition in addition to general virulence
function of pathogen conferred by hrp genes. Host-specific virulence (hsv) genes
encoding positive function have been shown to be a major determinant for host range and
pathovar status of Xanthomonas (Gabriel et al. 1993). These genes are required for
bacterial pathogens to grow and/or to induce pathogenic symptoms on specific host
plants, but not in other plant species. Several host-specific virulence genes have been
isolated from X. campestris pv. citrumelo (Kingsley et al. 1993), X. c. pv. translucens
(Waney et al. 1992) and X. citri (Swarup et al. 1991). For example, opsX of X. c. pv.
citrumelo is involved in biosynthesis of lipopolysaccharide and extracellular
polysaccharide and is required for virulence on citrus, but not on bean (Kingsley et al.
1993). These hsv genes are analogous to host-specific nodulation (hsn) genes of
Rhizobium. The hsn genes of Rhizobium is involved in host-specific modification of
lipooligosaccharide Nod factors and function only for nodulation on specific hosts (Fisher
and Long 1992; Denarie et al. 1992).
Besides hsv genes, avr genes have been shown to play a role in determining host
specificity at the species level (Keen 1990; Heath 1991). Single cloned avr genes from

4
a pathogen of one host species can cause an otherwise virulent pathogen of another host
species to become avirulent on its own host (Kobayashi et al. 1989; Whalen et al. 1988).
Therefore, host species specificity is conditioned by both host-specific virulence and
avirulence genes with the former as the major determinant.
Although avr genes may condition host range at the species level in some cases,
they act as negative factors to confer avirulence and to limit host range only if the host
plants have specific resistance (R) genes. The interaction of pathogens with avr genes
and host plants with specific R genes results in the plant defense response, often known
as hypersensitive response. This genetic requirement for specific plant R genes and
specific microbial avr genes is called a gene-for-gene interaction. The gene-for-gene
interactions were first demonstrated by Flor (1942, 1946, 1947) using flax and flax rust
and have since been observed in many plant-parasite interactions between fungi, bacteria,
virus, nematodes and their respective hosts (Thompson and Burdon 1992).
During the past decade, many avr genes have been isolated from different
bacterial and fungal plant pathogens (Gabriel et al. 1993; Keen 1990). In some
instances, avr genes encode proteinaceous elicitors that directly induce the HR on
resistance plants. For example, avr4 and avr9 of the fungal pathogen Cladosporium
fulvum encode race-specific peptides which are excreted into the apoplast of infected
leaves and induce the HR on tomato cultivars that carry resistance genes Cf4 and Cf9,
respectively (Joosten et al. 1994; van Kan et al. 1991). In another case, avrD of
Pseudomonas syringae pv. tomato appears to encode a enzyme involved in synthesis of
a race-specific glycolipid elicitor (Keen et al. 1990; Kobayashi et al. 1990, Midland et

5
al. 1993). In most cases, however, the biochemical identities of aw gene products are
unknown and the sequence analyses of aw genes are uninformative as to how they might
interact with R gene products to trigger the HR and specify gene-for-gene interaction.
Several models (including elicitor/receptor model, dimer model and ion channel
defense model) have been proposed to explain the molecular and biochemical bases of
gene-for-gene recognition between plant and pathogen (Gabriel and Rolfe 1990). But
very little experimental evidence is available to prove any of the hypotheses. Recently,
the first plant R gene, Pto, has been isolated from tomato by map-based positional
cloning (Martin et al. 1993). This gene encodes a serine/threonine protein kinase that
is homologous to the receptor protein kinase SRK6 involving Brassica pollen-stigma
recognition and to the Raf protein kinase involving the mammalian Ras signaling
pathway. The Ras pathway is the best known circuit of signal transduction found in
yeast, insect, as well as mammalian systems (Culotta and Koshland 1993). The Ras
pathway begins with extracellular signals (e. g. growth factors or cytokines) that interact
with receptor protein kinases, which in turn trigger phosphorylation cascades and finally
expression of targeted genes (Crews and Erikson 1993).
Although Ras-like signaling pathway has not been demonstrated in plants,
homologues of mammalian signal transduction components were found in many plants.
Those include homologues of G-protein subunits (e. g. auxin-regulated ArcA of tobacco,
Ishida et al. 1993), receptor protein kinases (e.g. pollen-specific SRK of Brassica,
Goring and Rothstein 1992), Ras-related proteins (e.g. Rgpl of rice, Kamada et al.
1992), Raf (e.g., Pto of tomato, Martin et al. 1993; and ethylene-regulated CTR1 of

6
Arabidoposis, Kieber et al. 1993) and mitogen-activated protein kinase (e.g., MsERKl
of alfalfa, Duerr et al. 1993). It is possible that virulence and avirulence gene products
may activate the similar signal transduction pathway in plants, leading to disease
symptoms or hypersensitive response. For example, transcriptional activation of plant
defense genes is modulated by phosphorylation (Felix et al. 1991; Yu et al. 1993) and
can be blocked by inhibitors of mammalian protein kinases (Raz and Fluhr 1993). It was
proposed that recognition of aw signals by R gene products would trigger a
phosphorylation cascade, leading to plant defense responses (Lamb 1994).
In this study, two economically important, world-wide diseases caused by
Xanthomonas have been used as model systems. Cotton bacterial blight is caused by X.
campestris pv. malvacearum; symptoms include angular, watersoaked spots on leaves,
black arm on stems and bollrot (Verma 1986). Citrus canker is caused by X. citri;
symptoms include erumpent, corky hyperplasia on leaves, stems and fruit (Schoulties et
al. 1987; Stall and Civerolo 1991). Recently, members of a Xanthomonas avirulence
(aw) and pathogenicity (pth) gene family were found to play an important role in
determining host-specific virulence and avirulence caused by these two pathogens (De
Feyter and Gabriel 1991a; De Feyter et al. 1993; Swamp et al. 1991, 1992). For
example, a pathogenicity gene, pthA, is required for X. citri to elicit hyperplastic lesions
on citrus (Swamp et al 1991). It also functions as an avr gene to confer the ability to
elicit an HR on cotton and bean (Swamp et al. 1992). Members of this gene family also
include awB4, awb6, avrb7, avrBIn, awBlOl and awB102 of X. campestris pv.
malvacearum (De Feyter and Gabriel 1991a; De Feyter et al. 1993), awBs3, avrBs3-2

7
and avrBsP of X. campestris pv. vesicatoria (Bonas et al. 1989, 1993; Canteros et al.
1991), and avrxa5, avrXa7 and avrXalO of X. oryzae (Hopkins et al. 1992). The most
conspicuous feature of this gene family is the nearly identical, tandemly arranged, 102bp
repeats in the central portion of genes (Fig. 1-1). Deletion analyses of avrBs3 have
shown that the 102bp repeated motifs determines the avirulence specificity of the gene
(Herbers et al. 1992). However, many questions remain regarding to their pleiotropic
functions, specificity, evolution, and biochemical mechanism of gene-for-gene
recognition.
The primary objectives of this work were to determine (1) why avirulence genes
are present in the cotton blight pathogen and whether they confer any pleiotropic
functions, (2) what determines avirulence and pathogenic specificity in cotton blight and
citrus canker pathogens, (3) how the specificity is created and evolved in this avr/pth
gene family, and (4) how the signals encoded by the avr/pth genes are perceived and
transduced in plant cells.

8
Name
Gene Structure
# 102bp Pleiotropic
Repeats Function
pthA
avrb6
avrBs3
avrXalO
B P
Li¬
st
L
H B S
17.5
cankering
13.5
watersoaking
17.5
—
15.5
Figure 1-1. A Xanthomonas avr/pth gene family. Listed are four representive members:
pthA of X. citri, avrbó of X. campestris pv. malvacearum, avrBs3 of X. campestris pv.
vesicatoria, and avrXalO of X. oryzae pv. oryzae. The tandemly arranged blocks
represent the 102-bp repeats in the central portion of genes. Restriction enzyme cleavage
sites, relevant to this work and found in most members of the gene family, are: B,
BamHl\ P, Pstl; St, Stul; H, Hindi', S, Sstl.

CHAPTER 2
HOST-SPECIFIC SYMPTOMS AND INCREASED RELEASE OF XANTHOMONAS
CITR1 AND X. CAMPESTR1S PV. MALVACEARUM FROM LEAVES
ARE DETERMINED BY THE 102BP TANDEM REPEATS
OF pthA AND awb6, RESPECTIVELY
Introduction
Microbial genes involved in plant-microbe interactions may be functionally
classified into four broad categories: parasitic, pathogenic, host range and avirulence
(Gabriel 1986). Genes involved in parasitism are absolutely required for growth in
planta and are widely conserved at the family or genus level. Examples include most
hrp (hypersensitive response and pathogenicity) genes of Erwinia, Pseudomonas and
Xanthomonas, and the common nod (nodulation) genes of Rhizobium (Boucher et al.
1992, Denarie et al. 1992, Willis et al. 1991). Genes involved in pathogenicity are
required for induction of symptoms. Examples include pectate lyase, polygalacturonase
and endoglucanase genes (Collmer and Keen 1986, Schell et al. 1988, Roberts et al.
1988), dsp (disease specific, Arlat and Boucher 1991), wts (watersoaking, Coplin et al.
1992) genes, phytohormone biosynthetic genes (Smidt and Kosuge 1978) and toxin
biosynthetic genes (Mitchell 1984). Genes involved in conditioning host range are host
specific and required for growth on specific hosts. Examples include hsv (host-specific
virulence), pth (pathogenicity), and hsn (host-specific nodulation) genes of Pseudomonas,
9

10
Xanthomonas and Rhizobium (Denarie et al. 1992, Kingsley et al. 1993, Ma et al. 1988,
Saleh and Shaw 1988, Swarup et al. 1991, Waney et al. 1991, Gabriel et al., 1993).
Information on conservation of these genes within species, pathovars and biovars is
scarce, but these genes appear to determine biovar and pathovar status (Djordjevic et al,
1987; Gabriel et al, 1993). The fourth group are termed avirulence (avr) genes because
they negatively affect virulence. The avr genes are superimposed on basic compatibility
(Ellingboe 1976), are not highly conserved, and determine pathogenic races below the
species, biovar or pathovar levels. These four broad categories are not mutually
exclusive.
The interaction of microbes having avr genes and host plants having resistance
(/?) genes can result in plant defense responses, often observed visually as a
hypersensitive reaction (HR) and characterized by the rapid necrosis of plant cells at the
site of infection and the accumulation of phytoalexins. Most avr genes do not appear to
confer selective advantage to the pathogen (Gabriel 1989, Keen and Staskawicz 1988).
Pleiotropic functions have been identified for only 3 of the 30 avr genes cloned to date
(Gabriel et al, 1993). Furthermore, the DNA sequences of the cloned avr genes have
been remarkably uninformative in terms of function (Keen, 1990). The presence of most
avr genes in plant pathogens therefore remains enigmatic.
Recently, an avr gene family has been discovered in many different
xanthomonads; members include avrBs3, avrBs3-2 and avrBsP (Bonas et al. 1989, 1993;
Canteros et al. 1991) of X. campestris pv. vesicatoria, avrB4, avrbó, avrb7, avrBIn,
avrBlOl and avrBl02 ofX. campestris pv. malvacearum (De Feyter and Gabriel 1991a,

11
De Fey ter et al. 1993) and avrxa5, avrXa7 and avrXalO of X. oryzae (Hopkins et al.
1992). Interestingly, this gene family includes a gene, pthA, that is required for
pathogenicity of X. citri on citrus. This gene is not known to function for avirulence in
X. citri (Swarup et al. 1992), but is required for X. citri to induce cell divisions in the
leaf mesophyll of citrus, leading to epidermal rupture and subsequent release of the
bacteria onto the leaf surface. The gene also confers the ability to induce cell divisions
on citrus to X. campestris strains from several different pathovars (Swarup et al. 1991,
1992).
De Feyter and Gabriel (1991a) observed that avrb6 and avrb7 enhanced the
watersoaking ability of several X. campestris pv. malvacearum strains on cotton, but the
role of these genes in pathogenicity was not determined. The family therefore consists
primarily of avr genes, but includes at least one, and perhaps more, host-specific
pathogenicity genes. The most conspicuous feature of this highly homologous gene
family is the presence of nearly identical, tandemly arranged, 102bp repeats in the central
region of the genes. These repeats are known to determine the gene-for-gene specificity
of avrBs3 (Herbers et al. 1992). The purpose of this study was to characterize the
watersoaking functions of avrbó and to investigate the role of the 102bp repeats of pthA,
avrB4, avrbó, avrb7, avrBIn, avrBlOl and avrB102 in pathogenicity and avirulence.

12
Materials and Methods
Bacterial strains, plasmids, and culture media. The bacterial strains and plasmids
used in this study are listed in Table 2-1. Strains of Escherichia coli were grown in
Luria-Bertani (LB) medium (Sambrook et al. 1989) at 37°C. Strains of Xanthomonas
were grown in PYGM (peptone-yeast extract-glycerol-MOPS) medium at 30°C (De
Feyter et al. 1990). For culture on solid medium, agar was added at 15g/L. Antibiotics
were used at the following final concentrations (mg/L): ampicillin (Ap), 25; kanamycin
(Km), 20; gentamycin (Gm), 3; spectinomycin (Spc), 50; tetracycline (Tc), 15; and
rifampicin (Rif), 75.
Recombinant DNA techniques. Total DNA isolation from Xanthomonas was as
described (Gabriel and De Feyter 1992). Plasmids were isolated from E. coli by alkaline
lysis methods (Sambrook et al. 1989). Restriction enzyme digestion, alkaline
phosphatase treatment, DNA ligation and random priming reactions were performed as
recommended by the manufacturers. Southern hybridization was performed by using
nylon membranes as described (Lazo et al. 1987). Otherwise, standard recombinant
DNA procedures were used (Sambrook et al. 1989).
Construction of chimeric genes. To construct BamRl fragment-swapped chimeric
genes among avrB4, avrbó, avrb7, avrBln, avrBlOl, avrB102, the BamHl fragments
from these genes were cloned into an awb7 shell (containing 5’ and 3’ ends of avrb7,
but deleted for its BamWl fragment, pUFR163ABam) and an avrBln shell (containing 5’
and 3’ ends of avrBln, but deleted for its BamHl fragment, pUFR186ABam) on the

13
Table 2-1. Bacterial strains and plasmids used in this study.
Strain or Plasmid
Relevant Characteristics
Reference or Source
E. coli
DH5«
F", endAl, hsdR17 (rk“mk+), supE44,
thi-1, recAl, gyrA, relAI, $80dlacZ
AMI5. A (lacZYA-argF)U 169
Gibco-BRL,Gaithersburg,MD
HB101
supE44, AsdS20(rk~mk+), recA13,
ara-14, proA2, lacYl, galK2, rpsL20,
xyl-5, mtl-1
Boyer & Roulland-Dussoix 1969
ED8767
supE44, supFS8, hsdS3(rk~mk+), recA56,
galK2, galT22, metBl
Murray et al. 1977
Xanthomonas citri
3213
ATCC49118, citrus canker type strain
Gabriel et al. 1989
B21.2
pthA::Tn5-gusA, marker-exchange
mutant of 3213
Swamp et al. 1991
X. phaseoli
G27
ATCC49119, bean blight type strain
Gabriel et al. 1989
G27Sp
Spcr derivative of G27
Swamp et al. 1991
X. campestris pv. citrumelo
3048
ATCC49120, citrus leaf spot pathotype strain
Gabriel el al. 1989
3048Sp
Spcr derivative of 3048
Swamp et al. 1991
X. campestris pv. alfalfae
KX-ISp
Spcr derivative of KX-1, isolated from
alfalfa, causing citrus leaf spot
Swamp et al. 1991
X. campestris pv. malvacearum
XcmH
Natural isolate from cotton from
Oklahoma; carries six avr genes used in this
study on pXcmH, plus additional avr genes
De Feyter & Gabriel, 1991a
XcmH1005
Spontaneous Rif derivative of XcmH
This study
XcmH 1407
avrb6::TnS-gusA, marker-exchange
mutant of XcmH 1005
This study
XcmH 1427
avrb7::Ta5-gusA, marker-exchange
mutant of XcmH 1005
This study
XcmH1431
avrBIn\\Yt\5-gusA, marker-exchange
mutant of XcmH1005
This study
XcmN
Natural isolate from cotton from
Upper Volta, Africa
Gabriel et al. 1986
Xcml003
Spcr Rif derivative of XcmN
DeFeyter & Gabriel 1991a
Plasmid
pRK2013
ColEl, Km', Tra + , helper plasmid
Figurski & Helinski, 1979
pRK2073
pRK2013 derivative, npt::Tn7,Km‘, Spr,
Tra + , helper plasmid
Leong et al. 1982
pUFR042
IncW, Kmr, Gmr, Mob + , lacZa\ Par+,
DeFeyter & Gabriel 1991a
pUFR047
IncW, Gmr, Apr, Mob+, lacZa + , Par+
De Feyter et al. 1993

Table 2-1—continued.
14
Strain or Plasmid
Relevant Characteristics
Reference or Source
pUFR049
RSF1010 replicón, Cm1, Smr, IncW+,
Mob + displacement vector
Swarup el al. 1991
pUFR054
IncP, Tcr, Mob + , containing methylases
XmaI and XmaUl
DeFeyter & Gabriel, 1991b
pUFR115
7.5kb fragment containing avrB4 in pUFR042
De Feyter & Gabriel 1991a
pUFR127
5kb fragment containing avrbó in pUFR042
De Feyter & Gabriel 1991a
PÃœFR135
lacZ::avrbó fusion in pUFR042
De Feyter et al. 1993
pUFR142
9kb fragment containing avrBlOl in pUFR047
De Feyter et al. 1993
pUFR156
12.9 kb fragment containing avrBIn in pUFR042
De Feyter et al. 1993
pUFR157
1 lkb fragment containing avrB102 in pUFR042
De Feyter et al. 1993
pUFR163
lOkb fragment containing avrb7 in pUFR042
De Feyter et al. 1993
pUFR163ABam
pUFR163 deleted for 3.4 kb Bum HI fragment
This study
pUFR171
Internal Bum HI fragment of avrB4, in pGeml lZf(+)
De Feyter et al. 1993
pUFR172
Internal BomHI fragment of avrbó, in pGeml lZf(+)
De Feyter et al. 1993
pUFR173
Internal BamWl fragment of avrBlOl, in pGeml lZf(+)
De Feyter et al. 1993
pUFR174
Internal BomHI fragment of avrBIn, in pGeml lZf(+)
De Feyter et al. 1993
pUFR175
Internal BomHI fragment of avrBI02, in pGeml lZf(+)
De Feyter et al. 1993
pUFR176
Internal BomHI fragment of avrb7, in pGeml lZf(+)
De Feyter et al. 1993
pUFR178
Internal BamWl fragment of avrBIn,
in pGeml lZf( +), Safi site filled
This study
pUFR179
Internal BomHI fragment of avrb7, in pGeml lZf(+),
Sail site filled
This study
pUFR180
9.5 kb fragment containing avrbó in pUFR042
De Feyter et al. 1993
pUFR186
10.3 kb BglU-EcoBl fragment containing avrBIn
from pUFR156 in pUFR047
This study
pUFR186ABam
pUFR186 deleted for 3.7 kb BomHI fragment
This study
pUFR190-194
BumHI-swapped chimeric genes with avrb7
5’ and 3’ ends plus BomHI fragment of avrB4,
avrbó, avrBIn, avrBlOl, or avrB102 in pUFR047
This study
pUFR 196-200
BomHI-swapped chimeric genes with avrBIn
5’ and 3’ ends plus BomHI fragment of avrB4,
avrbó, avrb7, avrBlOl, or avrB102 in pUFR047
This study
pUFR205-209
StullHinc I r swapped chimeric genes with avrBIn
5’ and 3' regions plus StuVHincU fragment of
avrBlOl, avrB4, avrbó, avrB102, or avrb7
in pUFR047
This study
pUFR211-215
StuVHincU-swapped chimeric genes with avrb7
5’ and 3’ regions plus Siul/HincB fragment of
avrBlOl, avrB4, avrbó, avrBIn, or avrB102
in pUFR047
This study
pUFR217
pUFR156 derivative, avrBIn::TnS-gusA
This study
pUFR220
pUFR163 derivative, avrb7::Tti5-gusA
This study
pUFR227
pUFR180 derivative, avrb6::TnS-gusA
This study
pUFY019
3.7 kb 5iul//fineII-swapped fragment with pthA
5’ and 3’ regions of pZit45 plus avrbó internal
repeat region of pUFR135 in pUFR047
This study
pUFY020
3.8 kb BruI/Z/incII-swapped fragment with avrbó
S' and 3’ regions of pUFR135 plus pthA internal
repeat region of pZit45 in pUFR047
This study
pZit45
4.5kb fragment containing pthA in pUFR047
Swamp el al. 1992
pGEM7Zf(+)
ColEl, Apr, lacZa+
Promega Co., Madison, WI
pGEMllZf(+)
ColEl, Apr, lacZa*
Promega Co., Madison, Wl

15
pUFR047 vector. The resulting BamUl fragment-swapped chimeric genes formed
pUFR190-200. The avr genes from pXcmH and pthA from X. citri all have unique Stul
and Hindi sites delimiting the 102bp repeated regions, which allow the swapping of the
internal repeated regions among these genes. To construct Stul/Hindl fragment-swapped
chimeras using the pXcmH avr genes, pUFR174 and pUFR176 were cut with Sail, blunt
ended using the Klenow fragment and religated to destroy the Hindi sites on the
pGEMl lZf(+) portions of the plasmids, forming pUFR178 and pUFR179, respectively.
These were then cut with Stul and Hindi to delete the internal fragments, and used as
recipients for the Stul/Hindi internal fragments from all pXcmH avr genes on pUFR171-
176. The BamUl fragments from pUFR178 and its Stul/Hindl chimeras were recloned
into pUFR186ABam, forming pUFR205-209. The BamUl fragments from pUFR179 and
its Stul/Hindl chimeras were recloned into pUFR163ABam, forming pUFR211-215. To
facilitate construction of chimeric genes between pthA and avrbó, the 4.1 kb Sail
fragment containing pthA from pZit45 and the 3.0 kb EcoRl/SaB fragment containing
avrbó from pUFR135 were inserted into pGEM7Zf(+). The Stul/Hindl internal
fragments were swapped between the two pGEM derivatives to create chimeric genes.
The chimeric genes were then recloned as single EcoKl/Hindlll fragments into pUFR047,
forming pUFY019 and pUFY020.
Bacterial conjugation. Triparental matings were carried out to transfer broad host
range plasmids from E.coli DH5a to various Xanthomonas strain by using pRK2013 or
pRK2073 as helper plasmids as described (De Feyter and Gabriel 1991a). To transfer

16
plasmids into Xcm, the modifier plasmid pUFR054 carrying Xcml and XcmWl methylase
genes was used to increase the transfer frequency (De Feyter and Gabriel 1991b).
Marker-exchange mutagenesis. Marker-exchange mutagenesis of strain
XcmH1005 was accomplished by introducing the displacement vector pUFR049 into
XcmH1005 transconjugants harboring pUFR227 (avrb6::Tn5-gusA), pUFR220
(avrb7::Tn5-gusA) or pUFR217 (avrBIn::Tn5-gusA) derivatives. The procedure was the
same as described (Swarup et al. 1991), except that the modifier plasmid pUFR054 was
used to facilitate plasmid transfer into XcmH1005.
Plant inoculations. Cotton (Gossypium hirsutum L.) lines used were Acala-44
(Ac44) and its congenie resistance lines Bl, B2, B4, b6, b7, Bln and BIn3 as described
(Swarup et al. 1992, De Feyter et al. 1993). Cotton plants were grown in the
greenhouse, transferred to growth chambers before inoculation, and maintained under
conditions as described (De Feyter and Gabriel 1991a). Bacterial suspensions of X.
campestris pv. malvacearum (108 cfu/ml) in sterile tap water were gently pressure
infiltrated into leaves of 4-5 week old cotton plants. Pathogenic symptoms were
observed periodically 2-7 days after inoculation.
All citrus (Citrus paradisi ‘Duncan’, grapefruit) and bean (Phaseolus vulgaris
‘California Light Red’) plants were grown under greenhouse conditions. Plant
inoculations involving X. citri or pthA or derivatives of pthA were carried out in BL-3P
level containment (refer to Federal Register Vol.52, No. 154, 1987) at the Division of
Plant Industry, Florida Department of Agriculture, Gainesville. Bacterial suspensions
were standardized in sterile tap water to 108 cfu/ml and pressure infiltrated into the

17
abaxial leaf surface of the plants. All plant inoculations on cotton, citrus or bean were
repeated at least three times.
Bacterial growth in planta. To determine the growth of X. campestris pv.
malvacearum in the susceptible cotton line Ac44, bacterial suspensions were adjusted to
106 cfu/ml and inoculated into cotton leaves by pressure infiltration. Leaf discs (1 cm2)
were taken by using a sterilized cork borer, then macerated in 1 ml sterile tap water. At
least three samples were taken at each time point for each strain inoculated. Viable
counts were determined by serial dilutions on plates containing appropriate antibiotics.
Data shown in Fig. 2-2 are the mean and standard error from three experiments.
To quantify the amount of bacteria present on the surface of watersoaked leaf
spots, leaves were inoculated with 107 cfu/ml bacterial suspensions from overnight
cultures. Five days later, 100 /¿I sterile tap water was dispensed onto the watersoaked
leaf surface and spread to an area of approximately 1 cm2. After mixing with the
bacteria and slime on the leaf surface, the bacterial suspension was collected with a pipet.
Each 1 cm2 watersoaked leaf area was washed ten times, and a total of 1 ml of bacterial
suspension was collected. To quantify the bacteria remaining inside the leaf, leaf discs
(1 cm2) were taken with a sterilized cork borer, then macerated in 1 ml sterile tap water.
Viable counts were determined by serial dilutions on plates containing appropriate
antibiotics. Data reported in the results (external and total counts) were mean and
standard error of six replicates from two experiments.

18
Results
Pleiotropic pathogenicity functions of avrbó. X. campestris pv. malvacearum
strain XcmH carries avrB4, avrbó, avrb7, avrBIn, avrBlOl and avrB102 on a single
plasmid, pXcmH, and elicits an HR on cotton lines carrying any one of many different
resistance (/?) genes (De Feyter et al., 1993). Strain XcmH and a spontaneous
rifamycin-resistant derivative of XcmH, XcmH1005, were virulent on susceptible cotton
line Acala-44 (Ac44), and both elicited severe watersoaking and necrosis associated with
growth in planta (XcmH1005 on Ac44 is shown in Figs. 2-1 & 2-2). Mutations of
avrbó, avrb7 and avrBIn were individually generated in XcmH 1005 by marker-exchange
mutagenesis and each mutant was confirmed to carry a single Tn5-gusA insertion in the
appropriate DNA fragment by Southern blot hybridization (refer Fig. 2-3; some data not
shown). As predicted by gene-for-gene theory, marker-exchange mutants XcmH1407
(avrb6::Tn5-gusA), XcmH 1427 (avrb7::Tn5-gusA) and XcmH 1431 {avrBIn::Tn5-gusA)
gained virulence on cotton lines with the resistance genes b6, b7 and Bln, respectively
(Table 2-2). Plasmids pUFR127 {avrb6+), pUFR163 (avrb7+) and pUFR156 (avrBIn+)
were able to fully complement the specific avirulence defects of XcmH1407, XcmH1427
and XcmH 1431, respectively.
Neither XcmH1427 nor XcmH1431 showed any change in watersoaking ability
as compared with XcmH 1005. However, marker-exchange mutant XcmH 1407 not only
lost avirulence on a cotton line with the b6 resistance gene, but also elicited significantly
less watersoaking and necrosis on susceptible cotton line Ac44 (Table 2-2 and Fig. 2-1A

Figure 2-1. Watersoaked lesions caused by X. campestris pv. malvacearum strains on
the susceptible cotton line Acala-44. A. Leaf inoculated five days previously with: 1,
XcmH1005; 2, XcmH1407; 3, XcmH1407/pUFR127; 4, XcmH1431. B. Leaf
inoculated seven days previously with: 1, XcmH1005; 2, XcmH1407. C. Leaf
inoculated five days previously with: 1, XcmH1005; 2, Xcml003/pUFR127; 3,
Xcml003/pUFR156; 4, Xcml003.

20

Figure 2-2. Growth kinetics of X. campestris pv. malvacearum strains on the susceptible cotton line Acala-44.
A. • XcmH1005; a XcmH1407; BXcmH1431; O Xcml003. B. • XcmH1005; v Xcml003/pUFR127;
Xcml003/pUFR156; O Xcml003.

Days post-inoculation

23
Figure 2-3. Southern blot hybridization of total DNA from XcmH1005 and marker-
exchanged mutants of XcmH1005 after EcoRl/Sstl digestion. The blot was probed with
the internal 2.9kb BamHl fragment from avrb6. The expected positions of fragments
carrying each plasmid-borne member of the avr/pth gene family were determined from
the restriction map of pXcmH [De Fey ter and Gabriel, 1991]. The upper arrow indicates
the expected position of the DNA fragment carrying avrBIn; the lower arrow indicates
the position of avrb6. Lane 1, 1 kb DNA ladder; Lane 2, XcmH1005; Lane 3,
XcmH1407; Lane 4, XcmH1431.

24
Table 2-2. Phenotypes of marker-exchange mutants of Xanthomonas campestris pv.
malvacearum strain XcmH1005 on cotton cultivar Acala-44 (Ac44) and congenie lines3.
Xcm strains
Ac44
Acb6
Acb7
AcBIn
XcmH1005
+ +
_
_
—
XcmH1407 (avrb6::Tn5-gusA)
+
+
-
-
XcmH1427 (avrb7::Tn5-gusA)
+ +
-
+ +
-
XcmH1431 (avrBIn::Tn5-gusA)
+ +
-
-
+ +
aAcb6, Acb7, and AcBIn are congenie lines of Ac44 containing resistance gene b6, b7,
and Bln, respectively. + + = strong watersoaking symptoms; + = weak watersoaking
symptoms; - = hypersensitive response.

25
& 2-IB). Plasmid pUFR127 was able to fully complement the pathogenicity defect(s)
of XcmH1407 on susceptible cotton plants, in addition to complementing the specific
avirulence defect. Despite its reduced ability to elicit pathogenic symptoms on
susceptible cotton lines, mutant XcmH1407 exhibited the same growth rate and yield as
that of the wild type XcmH1005 and XcmH1431 on Ac44 (Fig. 2-2A).
X. campestris pv. malvacearum strain Xcml003 carries no known avr genes (De
Feyter and Gabriel 1991a), caused less watersoaking on Ac44 than XcmH1005 and
exhibited a lower growth rate on Ac44 than XcmH1005. Introduction of pUFR127
(avrb6+) into Xcml003 conferred increased watersoaking ability (Fig. 2-1C), but growth
rate and yield in planta of the transconjugant containing pUFR127 were not increased in
comparison with that of Xcml003 or the transconjugant Xcml003/pUFR156 (avrBIn+)
(Fig. 2-2B). Therefore pUFR127 affected pathogenic symptoms in both XcmH1005 and
Xcml003, but not bacterial growth rate or yield in planta of either strain.
Strains containing avrbó (e.g. XcmH1005 and Xcml003/pUFR127) elicited more
severe watersoaking and necrosis and were associated with much more slime oozing from
the watersoaked areas in comparison with strains lacking avrbó (e.g. XcmH1407 and
Xcml003; Fig. 2-1). The peak number of total colony forming units (cfu) per square
centimeter of watersoaked leaf inoculated with XcmH1005 or XcmH1407 were basically
the same (1.41 ± 0.09 x 109 cfu/cm2 vs. 1.48 ± 0.17 x 109 cfu/cm2). However,
14.1% of the XcmH1005 bacteria present in the lesion (2.32 ± 0.56 x 108 cfu/cm2)
were released onto the surface of the leaf, whereas only 0.06% of the XcmH1407
bacteria (9.64 ± 4.42 x 105 cfu/cm2) were released onto the surface of the leaf.

26
Therefore, more than 240 times more bacteria were present on the external surface of
the watersoaked lesions caused by XcmH1005 than by XcmH1407. In experiments
where the leaves were moistened periodically after inoculation with a hand-held mist
sprayer, X. campestris pv. malvacearum strains carrying avrbó always exhibited many
secondary infections around the original inoculation site. In contrast, those strains
lacking avrbó rarely exhibited secondary infections.
The pathogenicity functions encoded by pthA and avrbó are host specific and
determined by 102bp repeats. Single Stul and Hindi sites are found in avrbó at positions
1281 and 2860, respectively, that closely flank the 102 bp tandemly repeated region of
the gene (De Feyter et al. 1993). Unique Stul and Hindi sites are also found at the
same relative positions in pthA, and the DNA sequences of avrbó and pthA are identical
in the flanking regions from the 102 bp tandem repeats to these restriction sites (refer to
Chapter 4). Chimeric genes were constructed by swapping the Stul/Hindi fragments
containing the internal 102bp repeated regions between pthA and avrbó. At least three
individual clones of both chimeric genes were introduced into a mutant X. citri strain
B21.2 (pthA::Tr\5-gusA) and wild type strains of the following pathogens: X. phaseoli
strain G27 (host range on bean), X. campestris pv. citrumelo strain 3048 (host range on
bean and citrus), X. campestris pv. alfalfae strain KX-1 (host range on alfalfa, bean and
citrus) and Xcml003. Strains and transconjugants were inoculated on citrus, bean, and
cotton leaves. As shown in Figure 2-4, pUFY020, carrying a chimeric gene containing
5’ and 3’ ends of avrbó and the internal repeats of pthA, complemented the mutant B21.2
to full virulence on citrus. The citrus canker symptoms caused by B21.2/pUFY020 were

27
Figure 2-4. Phenotypes of X. citri wild type and mutant strains on a grapefruit (Citrus
parodist cv.‘Duncan’) leaf. 1, wild-type strain 3213; 2, marker-exchange mutant of
3213, B21.2; 3, B21.2/pZit45; 4, B21.2/pUFY020; 5, B21.2/pUFR135; 6,
B21.2/pUFY019.

28
indistinguishable from the symptoms caused by B21.2/pZit45 (pthA +). When pUFY020
was present in 3048 or KX-1, both of which cause watersoaked leaf spots on citrus,
canker symptoms were induced by the transconjugants that were indistinguishable from
those conferred by pZit45 in the same strains. Also like pZit45 (Swarup et al, 1992),
pUFY020 conferred avirulence on bean to 3048, KX-1 and G27, and on cotton to
Xcml003.
A chimeric gene containing the 5’ and 3’ ends of pthA and the internal repeats of
avrb6 on pUFY019 enhanced the ability of Xcml003 and XcmH1407 to watersoak cotton
and cause necrosis. The watersoaking symptoms caused by Xcml003/pUFY019 were
indistinguishable from those caused by Xcml003/pUFR127. This chimeric gene on
pUFY019 behaved like avrbó on pUFR127 and conferred no detectable symptoms to
B21.2, 3048 or G27 on citrus or bean.
Cultivar-specific avirulence is determined by 102bp repeats. Each of the seven
avr/pth genes (avrB4, avrbó, avrb7, avrBln, avrBlOl, avrB102, and pthA) exhibits
unique avirulence specificity in Xcml003 on cotton resistance lines differing by single
R genes (De Feyter et al. 1993, Swarup et al. 1992). Each of these seven genes
contains two BamUl sites, one near the 5’ end and one near the 3’ end of each gene, and
single Stul and Hindi sites flanking the 102 bp direct repeat region (De Feyter et al.
1993, Swarup et al. 1992). To localize the region that determines the specificity of the
reactions, a series of chimeric genes were constructed by swapping BamHl and
Stul/Hindl internal fragments among these seven members of the gene family. One to
three individual clones of each chimeric gene were introduced into Xcml003 and

29
Table 2-3. Avirulence specificity of ZtowHI-fragment swapped chimeric genes in
Xcml003 on cotton.
Plasmid
5’ and 3’ end
Chimeric Gene
BamHl fragment
Avirulence Specificity
pUFR190
avrb7
avrB4
avrB4
pUFR191
avrb7
avrbó
avrbó
pUFR192
avrb7
avrBln
avrBln
pUFR193
avrb7
avrBlOl
avrBlOl
pUFR194
avrb7
avrB102
avrB102
pUFR196
avrBln
avrB4
avrB4
pUFR197
avrBln
avrbó
avrbó
pUFR198
avrBln
avrb7
avrb7
pUFR199
avrBln
avrBlOl
avrBlOl
pUFR200
avrBln
avrB102
avrB102
Note: Chimeric genes were introduced into Xcml003 and tested on cotton cv. Ac44 and
its congenie resistance lines AcBl, AcB2, AcB4, Acb6, Acb7, AcBIn and AcBIn3 for
avirulence specificity.

30
Table 2-4. Avirulence specificity of SM/7/mcII-fragment swapped chimeric genes in
Xcml003 on cotton.
Plasmid
Chimeric Gene
Avirulence Specificity
5’ and 3’ region
Stull Hindi fragment
pUFR205
avrBIn
avrB4
avrB4*
pUFR206
avrBln
avrbó
avrbó*
pUFR207
avrBIn
avrb7
avrb7*
pUFR208
avrBln
avrBlOl
avrBlOl
pUFR209
avrBln
avrB102
avrB102
pUFR211
avrb7
avrB4
avrB4
pUFR212
avrb7
avrbó
avrbó
pUFR213
avrb7
avrBln
avrBln*
pUFR214
avrb7
avrBlOl
avrBlOl
pUFR215
avrb7
avrB102
ND
pUFY019
pthA
avrbó
avrbó
pUFY020
lacZ::avrb6
pthA
pthA-\-
Note: Chimeric genes were introduced into Xcml003 and tested on cotton cv. Ac44 and
its congenie resistance lines AcBl, AcB2, AcB4, Acb6, Acb7, AcBIn and AcBIn3 for
avirulence specificity. * = weak avirulence; -I- = strong avirulence; ND = no
avirulence detected.

31
inoculated on cotton cultivar Ac44 and its congenie resistance lines, each differing by a
single R gene (Bl, B2, B4, b6, b7, Bln or BIn3). The avirulence specificities of the
seven avr/pth genes were first localized within the internal BamWl fragments (Table 2-3),
and then were further localized within the internal Stull Hindi fragments (Table 2-4).
In all cases, the avirulence specificity of a given gene was determined inside the
Stull Hindi (tandem repeat) region.
Discussion
Plant pathologists have long been puzzled by the presence of avirulence genes in
pathogens. These genes act as negative factors to limit virulence and in most cases do
not appear to provide selective advantage to the pathogens (Ellingboe, 1976; Gabriel,
1989; Keen and Staskawicz 1988). Rare exceptions have been reported. For example,
avrBs2 from X. campestris pv. vesicatoria is required for optimal growth in planta
(Kearney and Staskawicz 1990). Both avrbó and avrb7 from X. campestris pv.
malvacearum strain XcmH are known to enhance the watersoaking ability of X.
campestris pv. malvacearum strain Xcml003 on cotton (De Feyter and Gabriel 1991a).
In this study we demonstrated that the ability of strain XcmH 1005 to cause strong
watersoaking and necrosis on cotton requires the presence of avrbó but not avrb7.
Although avrbó increased symptom elicitation by both Xcml003 and XcmH 1005, neither
their growth rates nor their maximum bacterial counts per square centimeter of leaf were
affected by the presence or absence of avrbó. Therefore avrbó functions as a

32
pathogenicity gene and increases symptoms of cotton blight, but without eliciting an HR
on cotton lines lacking b6.
The strong watersoaking ability conferred by avrb6 was correlated with much
higher (240-fold) levels of bacterial cells released from inside the plant leaf to the
surface. Since bacterial blight of cotton is usually spread by rain splash, the presence
of large numbers of bacteria on the leaf surface would undoubtedly contribute to the
dissemination of the population. Strains carrying avrbó would thereby have a selective
advantage on cotton plants lacking the b6 gene. Similarly, pthA appears to aid in the
dissemination of X. citri by rupturing leaf epidermis and releasing bacteria, although it
does so by inducing tissue hyperplasia (Swarup et al. 1991). Therefore, pthA and avrbó
not only contribute to the amount of damage caused by their respective xanthomonads
to their hosts, but may also contribute to the ecological fitness of their respective
bacterial populations as pathogenicity genes.
Both pthA and avrbó may help determine host range in a positive manner, and not
as avr genes. When pthA was transferred to X. campestris pv. malvacearum Xcml003
and X. phaseoli G27, it conferred avirulence on the respective hosts, and did not induce
tissue hyperplasia (Swarup et al, 1992). In the present study, when avrbó was
transferred to X. citri B21.2, X. campestris pv. citrumelo 3048 and X. phaseoli G27, it
conferred no detectable effect when these strains were inoculated onto their respective
hosts. Therefore, the pathogenicity functions of avrbó and pthA are host specific. If
release of the pathogen to the leaf surface is host specific and it contributes to the
ecological fitness of the pathogen as we propose, then avrbó and pthA function to

33
determine host range. Furthermore, the avirulence conferred by pthA appeared to be
gratuitous in terms of restricting host range (Swamp et al., 1992) and in the present
study, avrb6 failed to confer avirulence to three other pathogens. Therefore if avrbó and
pthA help determine host range, it is not because of their function as avr genes. In a
formal genetic sense, these pleiotropic avr/pth genes resemble some Rhizobium host-
specific nodulation (hsn) genes which are required for host range on some hosts, but
which can also confer avirulence when transferred to other Rhizobium strains with a
different host range (Debelle et al. 1988; Faucher et al. 1989; Lewis-Henderson and
Djordjevic 1991).
By swapping the tandemly repeated regions, the avirulence specificities of avrB4,
avrbó, avrb7, avrBIn, avrBlOl, avrB102 and pthA were shown to be determined by the
102bp tandem repeats. These results are consistent with and extend the findings of
Herbers et al. (1992), who showed that the avirulence specificity of avrBs3 is determined
by the 102bp repeats of that gene. In addition, the swapping experiments clearly
demonstrated that the pathogenicity functions of pthA and avrbó were distinct (cankers
vs. watersoaking), host specific, and in both cases the pathogenic specificity was
determined by their 102bp tandemly repeated regions. Furthermore, the use of chimeric
genes also ruled out the possibility that the host-specific pathogenicity on cotton and
citrus were result of additional, unidentified pathogenicity genes encoded on the plasmids
used (pUFR127 and pZit45).
Extensive deletion analyses of both the 5’ and 3’ ends of the pXcmH avr genes
showed that all sections of these genes are required for avirulence activity (De Feyter et

34
al. 1993). From the present study, all sections of pthA and avrb6 were required to
confer pathogenicity functions. However, the 5’ and 3’ ends of the genes outside of the
repeats appeared to be isofunctional among members of the gene family.
Besides the members of this Xanthomonas avr/pth gene family, internal tandem
repeats are found in some pathogenicity genes of animal pathogens. Examples include
the outer membrane protein A gene (ompA) of some Rickettsia species (Anderson et al.
1990, Gilmore 1993), intemalin gene (ini) of Listeria monocytogenes (Gaillard et al.
1991), toxin A gene of Clostridium difficile (Dove et al. 1990), and M protein genes
(emm) of Streptococcus (Hollingshead et al. 1987). Tandem repeats were also found in
many genes from protozoan and metazoan parasites such as Plasmodium (McConkey et
al. 1990), Trypanosoma (Hoft et al. 1989), Leishmania (Wallis and McMaster 1987), and
Meloidogyne (Okimoto et al. 1991). Genes from Trypanosoma cruzi (the protozoan
agent of American trypanosomiasis) and mitochondria of Meloidogyne javanica (plant
root knot nematode) contain 102 bp tandem repeats. Most of these genes encode surface
proteins and the distinctive arrangement of the tandem repeats in these genes are thought
to encode a protective, strain-specific conformational epitope for evasion of host
immunity (Gilmore 1993, Hoft et al. 1989, McConkey et al. 1990). By contrast,
AvrBs3 is mainly located in cytosol (Knoop et al. 1991, Brown et al. 1993) and it is not
clear how it may interact with the plant cell.
Southern hybridization has shown that potential members of the avr/pth gene
family exist, often in multiple copies, in 9 of 12 Xanthomonas species or pathovars
examined (Bonas et al. 1989, Swarup et al. 1992, De Feyter et al. 1993). In all X. citri

35
and X. campestris pv. malvacearum cotton strains tested to date, multiple DNA fragments
hybridizing to pthA and avrb6 are found. The pleiotropic pathogenicity functions of pthA
and avrb6 and their potential fitness value may explain why these genes are maintained
in plant pathogens. Based on these data and knowledge of the functions of avrbó and
pthA, we assume that all strains of X. campestris pv. malvacearum capable of strongly
watersoaking cotton carry an avrbó gene or homologue that functions for pathogenicity.
Similarly, we assume that all strains of X. citri capable of causing cankers on citrus carry
a pthA gene. A number of pathogenic strains and pathovars in the genus Xanthomonas
do not carry members of the gene family, and therefore these genes are not required for
Xanthomonas virulence generally. Within pathovars where members of the avr/pth gene
family are found in some, but not all strains tested (such as X. campestris pv.
vesicatoria), there may be no pleiotropic pathogenicity function; for example, there is no
evidence of a pathogenicity function of avrBs3, avrBs3-2 and avrBsP (Bonas et al. 1989,
1993; Canteros et al. 1991).
The mechanism(s) by which leaf spotting pathogens elicit watersoaking and
necrosis is unknown, but appears to involve both damage to leaf cell membranes (without
eliciting an HR) and production of extracellular polysaccharide (EPS). The EPS does
not appear to be involved in suppressing a potential HR, since production levels of EPS
by X. campestris pv. malvacearum are similar in both susceptible and resistant cotton
lines (Pierce et al. 1993). Instead, mutational analyses of Xanthomonas EPS biosynthetic
genes and inoculations with purified EPS have shown that the EPS contributes to
watersoaking by trapping water and nutrients in the intercellular spaces after they are

36
released (reviewed by Leigh and Coplin 1992). Coplin et al. (1992) have proposed that
watersoaking and pathogenicity of Erwinia stewartii involves the EPS plus a cell leakage
factor encoded by wts (watersoaking) genes. This hypothesis may well apply to X.
campestris pv. malvacearum, with avrb6 encoding a cell leakage factor. Like some wts
genes of E. stewartii, avrbó was not required for bacterial growth in planta, but strongly
affected watersoaking on its host.
The ion-channel defense model of the gene-for-gene hypothesis invokes a cell
leakage factor as the product of an avr gene (Gabriel et al. 1988, Gabriel and Rolfe
1990). In this model, avr genes produce a protein or compound that opens an ion
channel in the plant cell membrane that rapidly depolarizes the membrane, causing
electrolyte leakage and host cell death. This gene-for-gene model is not inconsistent with
the idea that the product of an avr gene might induce slower cell leakage on susceptible
hosts. The only difference might be the allelic form of the host R gene. If the leakage
were slow enough to avoid cascade amplification of a wound-response signal, changes
in the osmotic gradient could cause a net loss of water from the cell and redistribution
to the apoplast, thereby increasing the fluidity of the EPS (M. Essenberg and M. Pierce,
personal communication). Increased fluidity or amounts of EPS may increase bacteria
exuding onto the leaf surface through stomata (Thiers and Blank 1951). Another possible
function is that the increased levels of necrosis induced by avrbó may serve to collapse
the palisade layer and physically squeeze more bacteria onto the leaf surface.

CHAPTER 3
INACTIVATION OF MULTIPLE AVIRULENCE GENES IN XANTHOMONAS
CAMPESTR1S PV. MALVACEARUM CREATED A NONPATHOGENIC
ENDOPHYTE OF COTTON
Introduction
Avirulence (avr) genes act as negative factors to confer avirulence and to limit
host range of the pathogen on host plants carrying specific resistance (R) gene. As a
result, most avr genes reported to date appear to be gratuitous and of no obvious
selective value to the microbial plant pathogens (Gabriel et al. 1993). However, a few
of them have demonstrated pleiotropic function(s). For example, avrBs2 is required for
full virulence of Xanthomonas campestris pv. vesicatoria on susceptible hosts (Kearney
and Staskawicz 1990). More recently, pthA of X. citri and avrb6 of X. campestris pv.
malvacearum were also shown to have host-specific pathogenicity function in addition to
avirulence function (Swamp et al. 1991, 1992; Chapter 2). Gene pthA is essential for
X. citri strain 3213 to induce hyperplastic cankers on citrus specifically and confers that
function to several other X. campestris pathovars. Similarly, avrb6 is important for X.
campestris pv. malvacearum strain XcmH to induce watersoaking on cotton specifically,
and confers that function to other X. campestris pv. malvacearum strains. Both pthA and
avrbó appear to contribute to the fitness of their respective pathogens on their hosts by
37

38
increasing release of the pathogens from plant leaves, thus affecting host range through
increased dispersal.
Both pthA and avrbó are members of a large Xanthomonas avirulence (avr) and
pathogenicity (pth) gene family, and both confer gene-for-gene avirulence to X.
campestris pv. malvacearum on cotton (Swarup et al. 1993; De Feyter et al. 1993).
Other members of the avr/pth gene family include avrB4, avrb7, avrBIn, avrBlOl and
avrB102 of X. campestris pv. malvacearum (De Feyter and Gabriel 1991a, De Feyter et
al. 1993), avrBs3, avrBs3-2 and avrBsP of X. campestris pv. vesicatoria (Bonas et al.
1989, 1993; Canteros et al. 1991) and avrXa5, avrXa7 and avrXalO of X. oryzae
(Hopkins et al. 1992). Family members sequenced to date are 95%-98% identical to
each other (Bonas et al. 1993; Hopkins et al. 1992; De Feyter et al. 1993). Most
members of this gene family were isolated as avirulence genes with no evidence of
pathogenicity function.
Xanthomonas campestris pv. malvacearum (Smith) Dye is the causal agent of
bacterial blight of cotton, an economically important, world-wide disease. Six members
of this gene family were previously isolated from pXcmH, a 90-kb plasmid of X.
campestris pv. malvacearum strain XcmH. Southern hybridization analysis reveals at
least five additional hybridizing fragments in XcmH besides the six plasmid-borne avr
genes (De Feyter et al. 1993). Furthermore, all strains examined of Xanthomonas
campestris pv. malvacearum that attack cotton carry at least 4-11 DNA fragments that
hybridize with members of the avrlpth gene family, including many strains which carry
no known avr genes (De Feyter et al. 1993). The purpose of this research was to clone

39
and to mutagenize all avr genes in X. campestris pv. malvacearum strain XcmH and to
identify potential pleiotropic functions.
Materials and Methods
Bacterial strains, plasmids, and matings. The bacterial strains and plasmids used
in this study are listed Table 3-1. Strains of Escherichia coli were grown in Luria-
Bertani (LB) medium (Sambrook et al. 1989) at 37°C. Strains of Xanthomonas were
grown in PYGM (peptone-yeast extract-glycerol-MOPS) medium at 30°C (De Feyter et
al. 1990). For culture on solid media, agar was added at 15g/L. Antibiotics were used
as previously described (refer to Chapter 2). To transfer plasmids from E.coli strains
HB101 or DH5a to X. campestris pv. malvacearum, helper plasmid pRK2073 and
modifier plasmid pUFR054 were used as described (Feyter and Gabriel 199la,b).
Recombinant DNA techniques. Total DNA isolation from Xanthomonas was as
described (Gabriel and De Feyter 1992). Plasmids were isolated from E. coli by alkaline
lysis methods (Sambrook et al. 1989). Restriction enzyme digestion, alkaline
phosphatase treatment, DNA ligation and random priming reactions were performed as
recommended by the manufacturers. Southern hybridization was performed by using
nylon membranes as described (Lazo et al. 1987). Otherwise, standard recombinant
DNA procedures were used (Sambrook et al. 1989).
Gene replacement and marker eviction. To carry out marker-exchange
mutagenesis, a 3.8 kb BamHl fragment containing a nptl-sac cartridge from pUM24

40
Table 3-1. Bacterial strains and plasmids used in this study
Strain or Plasmid
Relevant Characteristics
Reference or Source
E. coli
DH5«
F", endAl, hsdR17 (rk'mk+), supE44,
Gibco-BRL,
thi-1, recAl, gyrA, relAl, <¡>80dlacZ
Gaithersburg, MD
HB101
AM15, A(lacZYA-argF)U 169
supE44, hsdS20 (rk'mk+), recA13,
Boyer &
ara-14, proA2, lacYl, galK2, rpsL20,
Roulland-Dussoix 1969
ED8767
xyl-5, mtl-1
supE44, supF58, hsdS3 (rk'mk+), recA56,
galK2, galT22, metBl
Murray et al. 1977
X. campestris pv.
malvacearum
XcmH
Natural isolate from cotton from
De Feyter & Gabriel, 1991a
XcmH1005
Oklahoma; carries six avr gene on pXcmH,
and additional six potential avr genes.
Spontaneous Rif derivative of XcmH
Chapter 2
HM1.10
avrBInv.nptl-sac, marker exchange
This study
HM1.15
mutant of XcmH 1005
avrb6::nptl-sac, marker exchange
This study
HM1.20
mutant of XcmH 1005
avrB5::nptI-sac, avrbó), marker
This study
HM1.20S
exchange mutant of XcmH 1005
avrBS', avrbó' marker-evicted
This study
HM1.26
mutant derived from HM1.20
He 1:: nptl-sac, marker-exchange
This study
HM1.32
mutant of XcmH 1005
Hcl", avrBln' , avrB104~ marker
This study
HM1.34
exchange mutant of XcmH 1005
avrBn:: nptl-sac, marker exchange
This study
HM1.36
mutant of XcmH 1005
avrB103::nptI-sac, marker exchange
This study
HM1.38
mutant of XcmH 1005
avrB4', avrbT, avrBln", avrBlOl',
This study
HM2.2
avr 102' marker exchange mutant of XcmH 1005
avrHcó', avrB4~, avrbó', avrbT,
This study
HM2.2S
avrBln', avrBlOT, avrl02'), marker exchange
mutant of HM1.20S.
XcmH 1005 (avrHcó', avrB4~, avrbó', avrbT,
This study
XcmN
avrBln', avrBlOT, avrl02'), marker-evicted
mutant derived from HM2.2S.
Natural isolate from cotton from
Gabriel et al. 1986
Xcml003
Upper Volta, Africa.
Spcr Rif derivative of XcmN
De Feyter «fe Gabriel 1991a

Table 3-1-continued.
41
Strain or Plasmid
Relevant Characteristics
Reference or Source
Plasmid
pRK2073
pRK2013 derivative, npt::Tn7,Km*, Spr,
Tra+, helper plasmid
Leong et al. 1982
pUFR004
ColEl, Mob\ Cm'
De Feyter et al. 1990
pUFR034
IncW, Km', Mob+, lacZa+, Par+, cos
De Feyter et al. 1990
pUFR042
IncW, Kmr, Gm', Mob+, lacZtt +, Par+
De Feyter & Gabriel 1991a
pUFR047
IncW, Gmr, Ap', Mob+, lacZa+, Par+
De Feyter et al. 1993
pUFR054
IncP, Tc', Mob+, containing methylases
Xmal and XmaWl
De Feyter & Gabriel 1991b
pUFR103
Cosmid clone carrying avrB4, avrb6.
De Feyter & Gabriel 1991a
pUFR106
Cosmid clone carrying avrB4, avrbó, avrBlOl
De Feyter & Gabriel 1991a
pUFR107
Cosmid clone carrying avrbó, avrBlOl, avrBln
De Feyter & Gabriel 1991a
pUFR109
Cosmid clone carrying avrBlOl, avrBln
De Feyter & Gabriel 1991a
pUFRlll
Cosmid clone carrying avrBlOl, avrBln, avrB102
De Feyter & Gabriel 1991a
pUFR112
Cosmid clone carrying avrBln, avrB102, avrb7
De Feyter & Gabriel 1991a
pUFR113
Cosmid clone carrying avrB102, avrb7
De Feyter & Gabriel 1991a
pUFR115
7.5-kb fragment containing avrB4 in pUFR042
De Feyter & Gabriel 1991a
pUFR127
5-kb fragment containing avrbó in pUFR042
De Feyter & Gabriel 1991a
pUFR142
9-kb fragment containing avrBlOl in pUFR047
De Feyter et al. 1993
pUFR156
12.9-kb fragment containing avrBln in pUFR042
De Feyter et al. 1993
pUFR157
11-kb fragment containing avrB102 in pUFR042
De Feyter et al. 1993
pUFR163
10-kb fragment containing avrb7 in pUFR042
De Feyter et al. 1993
pUFY1.48
8.4-kb fragment carrying pthAv.npt-sac
in pUFR047
This study
pUFYlO.l
8.1-kb fragment carrying pthAwnpt-sac
in pUFR004
This study
pUFY31.46
8.5-kb fragment carrying avrB103 in pUFR047
This study
pUFY33.19
5-kb fragment carrying avrB5 in pUFR047
This study
pUFY36.8
8.6-kb fragment carrying Hcl in pUFR047
This study
pUFY36.26
5-kb fragment carrying avrBn in pUFR047
This study
pUFY37.62
11-kb fragment carrying avrB104
and avrB5 on pUFR047
This study
pUFY38.1
8-kb fragment carrying avrB104 in pUFR047
This study
pUM24
npt 1-sac marker in a 3.8-kb BamHi fragment
Ried and Collmer 1987
pXcml.12
Cosmid clone carrying Hcl and avrBn
This study
pXcml.21
Cosmid clone carrying avrB103
This study
pXcm2.23
Cosmid clone carrying avrBn
This study
pXcm2.12
Cosmid clone carrying avrB104 and avrB5
This study
pXcm2.11
Cosmid clone carrying avrB5
This study
pXcml.22
Cosmid clone carrying Hc6
This study
pZit45
4.5-kb fragment containing pthA in pUFR047
Swarup et al. 1992

42
(Ried and Collmer 1987) was randomly ligated into a Ball site of pthA on pZit45.
Recombinant plasmids were screened for an insertion in the middle of the tandemly
repeated region, and pUFY1.48 was selected, in which the nptl-sac cartridge was found
in the Ball site of repeat number 10 of pthA. An SstI fragment carrying the ptliA:-.nptl-
sac fusion from pUFY1.48 was recloned into the suicide vector pUFR004 (De Feyter et
al. 1990), forming pUFYlO.l. Marker-exchange mutants (avr::nptl-sac) were created
by transferring pUFRIO. 1 to XcmH1005 and selecting for colonies resistant to kanamycin
(15 /xg/ml) and sensitive to chloramphenicol (35 pg/mY) and sucrose (5%). Some
marker-exchange mutants were plated on the PYGM medium containing 5% sucrose and
nptl-sac marker evicted strains were selected to allow further rounds of marker exchange
mutagenesis.
Plant inoculations and bacterial growth in planta. Cotton (Gossypium hirsutum
L.) lines used were Acala-44 (Ac44) and its congenie resistance lines AcBl, AcB2,
AcB4, AcB5a, AcB5b, Acb6, Acb7, AcBIn and AcBIn3 as described (Swarup et al.
1992, De Feyter et al. 1993). Cotton plants were grown in the greenhouse, transferred
to growth chambers before inoculation, and maintained under conditions as described (De
Feyter and Gabriel 1991a). Bacterial suspensions ofX. campestris pv. malvacearum (108
cfu/ml) in sterile tap water were gently pressure infiltrated into leaves of 4-6 week old
cotton plants. Pathogenic symptoms were observed periodically 2-7 days after
inoculation.
The growth of X. campestris pv. malvacearum in the susceptible cotton line Ac44
was determined as described (refer to Chapter 2). In planta growth experiments were

43
performed three times. Data points shown in Fig. 3-4 were the mean and standard error
of three samples from one experiment. Bacterial populations released onto the leaf
surface were quantified as described (refer to Chapter 2). These data are reported as the
mean and standard error of three samples. Similar results were obtained in two
independent experiments.
Results
Cloning of six potential avr genes from XcmH by hybridization. In addition to
the six members of the Xanthomonas avr/pth gene family previously cloned from plasmid
pXcmH of strain XcmH (De Fey ter and Gabriel, 1991a), Southern blot analyses revealed
the presence of six more DNA fragments in XcmH that hybridized to an internal Bam HI
fragment from avrb6 (Fig. 3-1). All of the fragments appeared to be large enough to
carry functional members of the avr/pth gene family. Based on the intensity of
hybridization and the fact that XcmH appears to carry only a single plasmid (pXcmH;
De Fey ter and Gabriel, 1991a), the additional hybridizing fragments appeared to be
chromosomal. Colony hybridizations were carried out to isolate these chromosomal
DNA fragments from a genomic library of XcmH (De Feyter and Gabriel 1991a) by
probing with an internal BamYil fragment from avrbó. Twenty-three out of seventy-two
cosmid clones identified by colony hybridization were found to carry DNA fragments not
found on pXcmH. These new, presumably chromosomal DNA fragments were readily
distinguished from the DNA fragments that form pXcmH by restriction fragment length

Figure 3-1. Southern hybridization of cosmid DNAs containing six chromosomal avr genes after EcoRI and S.v/I
double digestion. The blot was probed with 32~P labelled internal BamH\ fragment of avrbó. A, Ethidium bromide
stained gel; B, autoradiograph resulting from Southern hybridization. Lane 1, XcmH genomic DNA; lane 2, pXcmH;
lane 3, pXcml.12 (Hcl, upper band; and avrBn, lower band); lane 4, pXcml.21 (avrB103); lane 5, pXcm2.23
(avrBn); lane 6, pXcm2.12 (avrB104, upper band; and avrB5, lower band); lane 7, pXcm2.11 (avrB5)\ lane 8,
pXcml.22 (Hc6); M, molecular marker, a, Hcl; b, avrB4\ c, avrBlOl; d, avrB102, e, avrB103; f, avrBn; g, avrBIn\
h, avrB104\ i, avrb7\ j, avrB5\ k, avrbó', 1, Hc6.


46
polymorphisms (Fig. 3-1). Six different chromosomal DNA fragments hybridizing with
avrb6 had been cloned and identified. These DNA fragments were arbitrarily named
according to size as Hcl - 6. Six cosmid clones were selected for further study:
pXcml.12 carried Hcl and Hc3, pXcml.21 carried Hc2, pXcm2.23 carried Hc3,
pXcm2.12 carried Hc4 and Hc5, pXcm2.11 carried Hc5 only, and pXcml.22 carried
Hc6 (Fig. 3-1). Restriction digests of these clones revealed that Hcl, Hc2 and Hc3 were
closely linked, and that Hc4 and Hc5 were closely linked.
A virulence activity of hybridizing fragments. All cosmid clones were conjugally
transferred to the widely virulent X. campestñs pv. malvacearum strain Xcml003 and
tested on cotton cv. Acala 44 (Ac44) and 9 cotton lines, congenie with Ac44 and each
carrying one of the following resistance genes: Bl, B2, B4, B5a, B5b, b6, b7, Bln and
BIn3. To localize members of the avr/pth gene family, all cosmid clones with avirulence
activity were directionally subcloned into pUFR047 by EcoRl and Hindlll digestion
(these enzymes are not known cut within members of the family published to date).
pXcml.12 was subcloned to separate fragments Hcl and Hc3, generating pUFY36.8
(Hcl) and pUFY36.26 (Hc3). Fragment Hc2 on pXcml.21 was subcloned to form
pUFY31.46. pXcm2.12 was subcloned to generate pUFY37.62, still carrying Hc4 and
Hc5 on a single, 11 kb EcoRl/Hindlll DNA fragment. Fragment Hc4 on pUFY37.62
was further subcloned to form pUFY38.1. Fragment Hc5 on pXcm2.11 was subcloned
to form pUFY33.19. Since Hc6 on pUFY1.22 conferred no avirulence to Xcml003, it
was not evaluated further.

47
The indicated subclones were transferred to Xcml003 and tested on the congenie
cotton lines and on cotton line 20-3, carrying resistance gene Bn. The Hcl and Hc6
DNA fragments conferred no detected phenotype to Xcml003. The Hc2 and Hc4 DNA
fragments conferred exactly the same avirulence specificity to Xcml003 as did clones
carrying avrBlOl (De Feyter et al 1993): avirulence on AcB5a and AcBIn3. Therefore
we conclude that Hc2 and Hc4 carry two new members of the gene family, named
avrB103 and avrB104, respectively. The Hc3 DNA fragment on pXcm2.23 conferred
avirulence to Xcml003 on cotton lines containing resistance genes Bn or Bln3. Gene
awBn had previously been cloned from XcmH on pUFA-Hl and shown to confer
avirulence on cotton lines containing Bn (Gabriel et al, 1986), but avrBn was not known
to be a member of this avr/pth gene family. Restriction and Southern hybridization
analyses (not shown) revealed that avrBn is a member of this gene family, and is carried
on both pUFA-Hl and pXcm2.23. The Hc5 DNA fragment conferred avirulence to
Xcml003 on AcB5b and AcBIn3. Therefore we conclude that Hc5 carried a new
member of the gene family, named avrB5.
Inactivation of avr genes in XcmH 1005. To study the pleiotropic function of all
members of the avr/pth gene family in XcmH 1005 (a rifamycin mutant of XcmH),
marker exchange-eviction mutagenesis was carried out to inactivate or delete each of the
avr genes, singly or in combination. More than 140 mutants were generated and 60 of
them were tested on 10 congenie cotton resistance lines and analyzed by Southern
hybridization. Marker exchange-eviction mutants affecting all 10 avr gene were
obtained. In Table 3-2 are listed selected mutants which exhibited altered phenotypes on

Table 3-2. Phenotypes of marker-exchange mutants of Xanthomonas campestris pv. malvacearum strain XcmH1005 on cotton.
Xcm strains
Cotton cv. Acala-44 congenie lines
Ac44
AcB4
AcB5a
Acb6
Acb7
AcBln
AcBln3
XcmH1005
+ +
-
-
-
-
-
-
HM1.10 (avrBln::nptl-sac)
+ +
-
-
-
-
4- 4-
-
HM1.15 (avrb6::nptl-sac)
+
-
-
4-
-
-
-
HM1.20 (avrB5~, avrb6“)
+
-
±
4-
-
-
-
HM1.20S (avrB5~, avrb6~)
+
-
±
4-
-
-
-
HM1.38 (avrB4~, avrb7~,
avrBln~, avrB101~, avr102~)
+ +
4- 4-
-
4- 4-
4- 4-
-
-
HM2.2 (avrB4~, avrB5~, avrb6~,
avrb7~, avrBln~, avrB101~, avr102~)
0/+
0/4-
0/4-
0/+
0/4-
0/+
-
HM2.2S (avrB4~, avrB5~, avrb6~,
avrb7~, avrBln", avrB101~, avr102~)
0/4-
0/4-
0/4-
0/4-
0/4-
0/+
-
Note: "+ +" strong watersoaking ; "+" weak watersoaking; "0/+" extremely weak watersoaklng;
hypersensitive response (HR); "±" weak HR.
4^
OO

49
cotton. Southern analysis of 11 mutants was shown in Fig. 3-2. As predicted by gene-
for-gene theory, mutations of avrBIn (HM1.10), avrbó (HM1.15), or avrB4 plus avrb7
(HM1.38) in XcmH1005 resulted in the loss of avr specificity for cotton blight resistance
genes Bln, b6, B4, and b7, respectively. As reported in the previous chapter, a single
mutation of avrbó reduces the watersoaking ability of XcmH1005 on susceptible cotton
lines. In contrast, mutation of awB4, avrb7 or avrBIn did not affect watersoaking ability
of XcmH1005 on Ac44. A mutation in avrB5 (HM1.20) was shown to result in the same
phenotypic defects as the mutation in avrbó, i. e., a loss of avr specificity for resistance
gene b6 and reduction of watersoaking on the susceptible cotton line Ac44. However,
this mutation could not be complemented by the cloned avrB5 gene, instead, was
complemented by the cloned avrbó, indicating avrbó in HM1.20 was likely to be
affected. Mutation of both avrB5 and avrbó in HM1.20 and HM1.20S resulted a partial
loss of avirulence specificity for cotton blight resistance gene B5b. Mutation of avrBlOl,
avrB102, avrB103 (HM1.36), avrB104 (HM1.32), avrBn (HM1.34) or Hcl (HM1.26)
fragment resulted in no alteration of avirulence phenotype. This is consistent with the
fact that the cotton resistance genes that react with these avr genes also react with other
avr genes (gene-for-genes; De Feyter et al. 1993). No mutations affecting fragment Hc6
were recovered.
To inactivate a large number of avr genes in XcmH1005, the nptl-sac marker-
evicted mutant HM1.20S (avrB5~, avrbó') was subjected to a second round of marker
exchange mutagenesis. The resulting mutant HM2.2 and marker evicted progeny
HM2.2S had at least 7 avr genes (avrBf, avrB5~, avrbó', avrbT, avrBIn , avrBlOl',

50
Figure 3-2. Southern hybridization of jEcoRI and Ss/I double digested genomic DNAs of
XcmH mutants. The blot was probed with 32'P labelled internal BamUl fragment of
avrb6. Lane 1, XcmH; lane 2, HM1.10; lane 3, HM1.15; lane 4, HM1.38; lane 5,
HM1.26; lane 6, HM1.36; lane 7, HM1.34; lane 8, HM1.32; lane 9, HM1.20; lane 10,
HM1.20S; lane 11, HM2.2; lane 12, HM2.2S. a, Hcl; b, avrB4\ c, avrB101\ d,
avrB102, e, awB103\ f, avrBn\ g, avrBln\ h, avrB104\ i, avrb7; j, avrB5; k, avrb6\ 1,
Hc6.

51
avrl02~) inactivated. These two mutants not only lost aw specificity on most of cotton
resistance lines, but also lost nearly all watersoaking ability on the susceptible line Ac44
(Tab. 3-2, Fig. 3-3).
Despite repeated attempts to inactivate all members of the gene family in XcmH,
we were unable to reduce the number of members below five. In all cases, members of
the family appeared to rearrange to form new members, as evidenced by the appearance
of new DNA fragments of sufficient size to encode a member, and often accompanied
by the appearance of a new avirulence specificity phenotype (data not shown). Also in
all cases, at least one member remained plasmid-bome, as evidenced by comparatively
strong hybridization intensities characteristic of the plasmid-bome genes.
Effect of loss of watersoaking on bacterial growth in planta. Since avrbS
mutants of XcmH 1005 that had lost significant watersoaking ability on Ac44 were shown
to be unaffected in growth in planta (refer to Chapter 2), we evaluated growth of
HM2.2S on Ac44 in comparison with the parental strain XcmH 1005. While inactivation
of 7 aw genes resulted in a loss of nearly all watersoaking ability on cotton, in planta
growth rate and yield of HM2.2S was not affected (Fig. 3-4). However, the amount
of HM2.2S released onto the leaf surface was 1600 times less than that of XcmH 1005.
Only 0.007% (8.8 ± 3.0 X 104 cfu/cm2 out of 1.18 ± 0.24 X 109 cfu/cm2) of HM2.2S
bacteria present in the infected leaf tissue were released onto the leaf surface, whereas
9.2% (1.42 ± 0.15 X 108 cfu/cm2 out of 1.55 ± 0.23 X 109 cfu/cm2) of the XcmH1005
bacteria were released onto the surface. Therefore, the mutant HM2.2S was basically
a nonpathogenic endophyte of cotton.

52
Figure 3-3. Effect of avr genes on watersoaking symptoms caused by X. campestris pv
malvacearum on cotton susceptible line Ac44.

53
Day post—inoculation
Figure 3-4. Inactivation of multiple avr genes in X. campestris pv. malvacearum did not
affect bacterial growth in planta on cotton susceptible line Ac44. Filled circle,
XcmH1005; open circle, HM2.2S.

54
Additive effect of different avr genes on watersoaking ability of XcmH1005.
Complementation tests were carried out to further analyze the potential pleiotropic
pathogenicity function of 12 avr genes isolated from XcmH. As shown in Fig. 3-5,
when plasmids carrying single avr genes isolated from XcmH were introduced into the
mutant HM2.2S, only avrbó was able to partially complement the watersoaking defect
of HM2.2S. None of the other 9 avr genes or Hcl or Hc6 fragments were able to even
partially complement the watersoaking defect in HM2.2S when introduced individually.
However, when plasmids carrying two or three of these same avr genes (not necessarily
including avrbó) were introduced into HM2.2S, many were able to partially complement
the watersoaking defect (Figs. 3-3 and 3-5). For example, plasmids containing avrBIn
or avrBlOl alone conferred no watersoaking activity, but plasmids containing both
avrBIn and avrBlOl together exhibited watersoaking activity. Cosmid clones (pUFR106
and pUFR107) carrying avrbó and at least two other avr genes were able to fully
complement the watersoaking defects of HM2.2S (Fig. 3-3). Therefore, at least three
of the pXcmH avr/pth genes were redundant in terms of pathogenicity function.
To exclude the possibility that some watersoaking genes were linked to avr genes
on these complementing cosmid clones, subcloning of pUFR107 was carried out.
Activity assays of regions between the avr genes did not reveal any watersoaking activity
when introduced into HM2.2S.

55
avrB4 avrb6
! I
avrB101 avrBln
avrB102
avrb7
pXcmH ^
i ii i i
!
1 1
Activity
11.6 5.0 4.5 4.8
20.5
12.9
28.4
in HM2.2S
PUFR127
avrb6
+
PUFR103
avrB4 avrb6
+
pUFR113
avr B102
avrb7
+
PUFR109
avrBIOI avrBln
+ +
pUFR111
avrBIOI avrBln
avr B102
+ +
pUFR112
avrBln
avr B102
avrb7
+ +
pUFR107
avrb6
avrBIOI avrBln
+ + +
PUFR106
avrB4 avrb6
avrB101
+ + +
Figure 3-5. Pleiotropic watersoaking function of cloned XcmH avr genes in mutant
HM2.2S. * HM2.2S was nearly asymptomatic on susceptible cotton line Ac44; + weak
watersoaking activity; ++ intermediate level watersoaking activity; + + + full
watersoaking activity (as wild type XcmH).

56
Discussion
Nearly all of the aw genes reported from the genus Xanthomonas are members
of a single, highly homologous gene family, and with the exception of pthA
(pa/Ztogenicity) of X. citri (Swarup etal. 1991, 1992), all of the other published members
were identified and cloned as aw genes. Southern analyses have shown that members
of this aw/pth gene family exist in multiple copies (from 4 to 11) in all tested X.
campestris pv. malvacearum strains isolated from cotton, but that some strains,
particularly those from Africa, appear to lack functional avirulence (De Feyter et al.
1993). For example, the African strain Xcml003 used in this study contains at least five
hybridizing fragments, but is virulent on all tested cotton resistance lines. Of course it
is possible that the hybridizing fragments in a single strain are all nonfunctional, but it
is more difficult to explain the presence of from 4-11 hybridizing DNA fragments
present in all known X. campestris pv. malvacearum strains if they encoded no function
beneficial to the pathogen.
All 12 DNA fragments that hybridize with an internal fragment from the gene
family have now been isolated from a single X. campestris pv. malvacearum strain,
XcmH. Six of them are clustered on a 90-kb indigenous plasmid, pXcmH, and were
originally isolated as functional aw genes, although two of them {avrbó and avrbT) were
noted to confer pleiotropic pathogenicity (watersoaking) function (De Feyter and Gabriel
1991a; De Feyter et al. 1993). We report here the cloning of the other six hybridizing
fragments that appear to be from the chromosome. One of these fragments encodes the

57
previously described avrBn (Gabriel et al 1986). Three of these encode previously
undescribed avr genes (avrB5, avrB103 and avrB104), and the remaining two fragments
conferred no avirulence on cotton and were presumably non-functional.
Mutations of all ten avr genes in XcmH1005 were achieved by marker exchange-
eviction mutagensis. With the exception of avrbó, mutations of individual members of
the avr/pth gene family in XcmH1005 had no obvious effect on watersoaking on
susceptible cotton lines. Mutation of avrbó resulted in a significant loss of watersoaking
ability of XcmH1005. Although avrbó appeared to be the major pathogenicity gene
among 10 homologous avr genes in the strain XcmH, many XcmH1005 avr genes also
contributed to watersoaking function of XcmH 1005 on cotton. Therefore these avr genes
also function pleiotropically as pth genes. Mutant HM2.2S had at least seven avr genes
(including avrbó) inactivated, and completely lost the ability to induce watersoaking on
mature cotton leaves. Since HM2.2S suffered deletions of avr genes and adjacent DNA
fragments, mutational analysis alone did not rule out the potential involvement of closely-
linked virulence genes. However, complementation tests confirmed that avr/pth family
members fully complemented the watersoaking defect(s) of HM2.2S. Individually, none
of the 10 avr/pth genes, other than avrbó, could even partially complement the
nonpathogenic mutant strain HM2.2S in visual assays. However, DNA fragments
carrying multiple avr/pth genes from plasmid pXcmH were able to partially complement
the pathogenic defect of HM2.2S, even if they did not carry avrbó, demonstrating that
other avr/pth genes from XcmH also contributed to its pathogenicity.

58
Some avr/pth genes appeared to have stronger pathogenicity function than others
in HM2.2S, and all six pXcmH genes appeared able to contribute to pathogenicity on
cotton, when tested in combinations. However, no individual avr/pth gene, except
avrb6, exhibited pth function in HM2.2S. Full restoration of the watersoaking defect of
HM2.2S required avrbó plus two other pXcmH avr/pth genes. Thus, the cotton pathogen
XcmH contained one major and multiple minor genes from the same avr/pth gene family.
Some of the minor avr/pth genes appeared to be redundant in terms of watersoaking
function. Therefore mutation of one or few of minor avr/pth genes would not affect the
watersoaking and release of the pathogen, and presumably, the fitness of the pathogen.
In contrast to avrBs2 of X. campestris pv. vesicatoria (Kearney and Staskawicz
1990), there was no evidence that the avr/pth genes of XcmH are even slightly involved
in bacterial growth in planta. Instead, they were involved in the induction of
watersoaking symptoms and associated release of 9.2% of the total bacteria of the leaf
onto the leaf surface, presumably aiding in dispersal. Mutant HM2.2S, with at least
seven avr/pth genes destroyed, released 1,600 times less bacteria to the leaf surface than
the wild type. Although HM2.2S was asymptomatic on cotton, it maintained the same
level of bacterial growth in planta as the wild type strain. Therefore, inactivation of
multiple avr/pth genes in X. campestris pv. malvacearum created a nonpathogenic
endophyte of cotton. In the case of X. campestris pv. malvacearum, pathogenicity and
host tissue destruction appears to be beneficial for the reproductive fitness of the
pathogen. Since X. campestris pv. malvacearum spreads primarily by rain splash, it
seems doubtful that an endophyte could disseminate efficiently in natural field situations.

59
The lack of correlation between disease symptoms and growth in planta has been
observed with other plant pathogens. Indole acetic acid-deficient strains of Pseudomonas
syringae pv. savastanoi failed to elicit gall on oleander but exhibited similar growth
patterns as the gall-eliciting wild-type strain (Smidt and Kosuge 1978). The lemA gene,
which encodes a two-component regulator, is required by P. syringae pv. syringae for
disease lesion formation on bean plants, but not bacterial growth within or on the leave
of bean (Hrabak and Willis 1994; Willis et al. 1990). Mutation of the fungal pathogen
Colletotrichum magna, resulted in a strain which is not able to induce pathogenic
symptoms, but grows in host tissue as an endophyte and retains the wild-type host range
(Freeman and Rodriguez 1993).
A growing number of Xanthomonas virulence / pathogenicity genes are not hrp
(hypersensitive response and pathogenicity) genes, but instead are host-specific
determinants of disease and/or host range (Gabriel et al, 1993). For example, several
genes affecting host range were identified and complementing DNA clones isolated from
X. campestris pv. citrumelo (Kingsley et al, 1993) and X. campestris pv. translucens
(Waney et al, 1992). These genes are host specific and affect growth in some hosts, but
not in other hosts. Other non -hrp Xanthomonas pathogenicity genes, such as pthA of X.
citri and avrb6 of X. campestris pv. malvacearum are host specific and strongly affect
elicitation of pathogenic symptoms. From the present study, it appears that many
members of the avr/pth gene family of X. campestris pv. malvacearum are host-specific
determinants of disease and host range on cotton. Genetically, pthA of X. citri and the
avr/pth genes of X. campestris pv. malvacearum resemble some Rhizobium host-specific

60
nodulation genes which are required for host range on some hosts, but which can also
confer avirulence when transferred to other Rhizobium strains with a different host range
(Debelle et al. 1988; Faucher et al. 1989; Lewis-Henderson and Djordjevic 1991).
The pleiotropic function and apparent fitness value of some avr genes do not
imply that most avr genes have selective value for a specific plant pathogenic strain. In
fact, most avr genes reported to date appear to be dispensible and have no obvious
selective value to the pathogen (Gabriel et al. 1993). At least some members of the
Xanthomonas avr/pth gene family are found on self-mobilizing plasmids (for example,
Bonas et al. 1989), and their presence in some strains of a pathovar may be due to a
coincidental linkage with another (selected) factor, such as copper resistance, on the same
plasmid. Once horizontally transferred to a different strain, these genes may not remain
stable enough to retain pth function specific for some ancestral host. In both X.
campestris pv. malvacearum and X. citri, members of the avr/pth gene family mutate at
very high frequency (approximately 1(T4) through intergenic or intragenic recombination
(De Fey ter et al. 1993; Chapter 4). High frequency recombination provides a powerful
mechanism to generate variation and horizontal gene transfer provides a powerful
mechanism to disseminate that variation in the genus. It seems likely that many avr
genes carried by bacterial plant pathogens may result from variation in redundant and/or
unnecessary virulence/pathogenicity genes which are selectively neutral in natural
populations.

CHAPTER 4
INTRAGENIC RECOMBINATION OF A SINGLE PLANT PATHOGEN GENE
PROVIDES A MECHANISM FOR THE EVOLUTION
OF NEW HOST SPECIFICITIES
Introduction
In gene-for-gene interactions between plants and microbial plant pathogens, host
plant resistance results from the genetic recognition of resistance (R) genes in the plant
and avirulence (avr) genes in the pathogen (Keen 1990; Gabriel and Rolfe 1990). It is
unknown why microbial plant pathogens carry genes that function to limit virulence,
since the majority of these genes are dispensable (Gabriel et al. 1993). For example, to
avoid plant resistance, plant pathogens evade R gene recognition by deletion (De Feyter
et al. 1993), mutation (Kobayashi et al. 1990; Joosten et al. 1994) or transposon
insertion (Kearney et al. 1988) of avr genes. These reports all involve natural
mechanisms whereby avirulence may be lost. Nevertheless, mutational analyses reveal
that gains in avirulence, including new avirulence specificities, are generated in at least
some pathogens at approximately the same frequency as they are lost (Statler 1985).
Although presumably an avr gene may be reactivated by reverse mutation, there are no
reports of natural mechanisms whereby new avirulence specificities or new avr genes are
generated. The mechanisms by which new avr genes evolve are unknown.
61

62
Xanthomonas citri is the causal agent of citrus canker disease, which occurs
worldwide and is subject to eradication and quarantine regulations in many countries.
A paz/iogenicity gene, pthA, is essential for the pathogen to cause hyperplastic canker
symptoms on citrus (Swarup et al. 1991). Furthermore, when transferred to other
xanthomonads, pthA confers ability to induce hyperplastic cankers on citrus and a
hypersensitive response (HR) on other hosts (Swarup et al. 1992). Therefore, pthA
exhibits pleiotropic pathogenicity and avirulence functions. Southern hybridization,
restriction analysis and partial DNA sequencing has shown that pthA belongs to a major
aw gene family widespread in the genus Xanthomonas (Swarup et al. 1992). Members
of this gene family include awBs3, awBs3-2, and awBsP (a trancated form of awBs3-2)
of X. c. pv. vesicatoria (Bonas et al. 1989, 1993; Canteros et al. 1991), avrB4, avrb6,
avrb7, awBln, awBlOl and awB102 of X. c. pv. malvacearum (De Feyter and Gabriel
1991; De Feyter et al. 1993), and avrXalO and awXa7 of X. oryzae pv. oryzae (Hopkins
et al. 1992). Among these, only awb6 and pthA are known to have pleiotropic
pathogenicity functions (Swarup et al. 1992; De Feyter and Gabriel 1991; also refer to
Chapter 2). The most striking feature of this gene family is the nearly identical, 102bp
tandem repeats in the central portion of the genes. Deletion analyses have shown that
the avirulence specificity of awBs3 is determined by the 102bp repetitive motifs of that
gene (Herbers et al. 1992). By swapping the internal repeated regions between pthA and
awb6, the 102 bp tandem repeats are found to determine both avirulence and pathogenic
specificities (refer to Chapter 2).

63
In most organisms, repetitive regions of DNA are known to be active sites for
homologous recombination (Albertini et al. 1982; Petes and Hill 1988). The high
frequency of such events causes them to be major genome modifying forces in evolution.
Intergenic recombination among members of this avr/pth gene family was proposed to
explain the high frequency of race change mutation found in X. c. pv. malvacearum (De
Feyter et al. 1993). Since both pathogenic and avirulence specificities are determined
by the 102bp tandem repeats, it seemed reasonable to suppose that homologous
recombination between genes and among repeats of the same gene might lead to an
acceleration in the evolution of this Xanthomonas avr/pth gene family, resulting in new
host specificities for the pathogen. In this study, we analyzed the nucleotide sequence
of pthA and demonstrate that intragenic recombination of a single gene can provide a
genetic mechanism for the creation of new avr/pth genes.
Materials and Methods
Bacterial strains, plasmids and matings. Escherichia coli strains DH5a (Gibco-
BRL, Gaithersburg, MD), JM83 (Yanisch-Perron et al. 1985) and derivatives were
grown in Luria-Bertani (LB) medium (Sambrook et al. 1989) at 37°C. Xanthomonas
citri strain 3213 (causes hyperplastic cankers on citrus; Swarup et al. 1991, 1992), X.
campestris pv. citrumelo strain 3048 (causes water-soaked leaf spots on citrus and
common bean; Swarup et al. 1991, 1992), X. campestris pv. malvacearum strain
Xcml003 (causes water-soaked leaf spots on cotton; De Feyter and Gabriel 1991) and

64
derivatives were grown in PYGM (peptone-yeast extract-glycerol-MOPS) medium at
30 °C (De Fey ter et al. 1990). Antibiotics were used as described in previous chapters.
Triparental matings were carried out to transfer plasmids from E.coli DH5a to various
Xanthomonas strains by using pRK2013 or pRK2073 as helper plasmids as described
(Swarup et al. 1991; De Feyter and Gabriel 1991a). To transfer plasmids into Xcml003,
the modifier plasmid pUFR054 carrying Xcml and XcmlW methylase genes was used to
increase the transfer frequency (De Feyter and Gabriel 1991b).
DNA sequence analysis. The 4.1 kb Sa/I fragment carrying pthA in pZit45
(Swamp et al. 1992) was recloned into the pGEM 7Zf(+) vector (Promega, Madison,
WI) to yield pUFY14.5. Sets of overlapping, unidirectional deletion subclones were
generated in pUFY14.5 using exonuclease III and mung bean nuclease as described by
Promega (Madison, WI). DNA sequencing was performed by the DNA Sequencing Core
Laboratory of the Interdisplinary Center for Biotechnology Research, University of
Florida, Gainesville. Some sequencing was determined by using the dideoxy nucleotide
chain termination technique with the Amersham (Arlington Heights, IL) multiwell
microtiter plate DNA sequencing system RPN1590. The sequence was analyzed using
the GCG package (version 7.00) by Genetics Computer Group, Inc., Madison, WI.
Gene replacement and selection of marker-evicted strains. Gene pthA on pZit45
was mutated by ligating a 3.9 kb BamHl fragment containing a nptl-sac cartridge (Ried
and Collmer 1987) into a Ball site in one of the 102bp tandem repeats of pthA. One of
the resulting transcriptional fusions, pUFY1.48, carried the nptl-sac cartridge inserted
into repeat #10 of pthA and was selected for further study. An Sstl fragment carrying

65
thepthAr.nptl-sac fusion from pUFY1.48 was recloned into "suicide" vector pUFR004
(colEl, mobP+, cat+\ De Feyter et al. 1990), forming pUFYlO.l. Marker-exchange
mutant Xcl.2 (pthA::nptI-sac) was created by transferring pUFRIO. 1 to X. citri 3213 and
selecting for colonies resistant to kanamycin (15 /ig/ml) and sensitive to chloramphenicol
(35 /ig/ml) and sucrose (5%). Marker-evicted strains of X. citri were selected by plating
Xcl.2 on PYGM medium containing 5% sucrose and selecting colonies resistant to
sucrose. Colonies were then screened for sensitivity to kanamycin. Recombinant,
marker-evicted derivatives of pUFY1.48 were selected in X. campestris pv. citrumelo
3048, E. coli DH5a (recA ~) and E. coli JM83 (recA+) by plating on PYGM or LB
medium containing 5 % sucrose. Sucrose-resistant colonies were screened for sensitivity
to kanamycin.
To calculate recombination frequencies, XcM1.2, 3048/pUFY1.48,
DH5a/pUFY1.48 and JM83/pUFY1.48 were grown in liquid media in the presence of
appropriate antibiotics to densities of 109 cells/ml. Serial dilutions were then plated on
media with and without 5% sucrose. Recombination frequencies were calculated by
dividing the number of sucrose-resistant (also kanamycin-sensitive) colonies by the
number of colonies grown on the medium without sucrose selection (Xu et al. 1988).
Data shown in the results were the mean and standard error of four replicates obtained
from two independent experiments.
Plant inoculations. Citrus (Citrus paradisi ‘Duncan’, grapefruit), common bean
(Phaseolus vulgaris ‘California Light Red’) and cotton (Gossypium hirsutum L.) plants
were grown under greenhouse conditions. Cotton lines used were Acala-44 (Ac44) and

66
its congenie resistance lines AcBl, AcB2, AcB4, AcB5, Acb6, Acb7, AcBIn and AcBIn3
as described (De Feyter et al. 1993, Swarup et al. 1992). Plant inoculations involving
X. citri or pthA or derivatives of pthA were carried out in BL-3P level containment (refer
to Federal Register Vol.52, No. 154, 1987) at the Division of Plant Industry, Florida
Department of Agriculture, Gainesville. Bacterial suspensions were standardized in
sterile tap water to 108 cfu/ml and pressure infiltrated into the abaxial leaf surface of the
plants. Plant inoculations were repeated at least three times on citrus, and two times on
bean and cotton.
Results
Sequence analysis of pthA. The complete DNA sequence of pthA was determined
and shown in Fig. 4-1. DNA sequence comparisons revealed that pthA of X. citri is 98%
identical to a\rBs3 and avrBs3-2 of X. campestris pv. vesicatoria (Bonas et al. 1989,
1993), 97% identical to avrbó of X. campestris pv. malvacearum (De Feyter et al. 1993),
and 95% identical to avrXalO of X. oryzae (Hopkins et al. 1992) along its entire length.
As observed with all previously sequenced members of the gene family (Bonas et al.
1993; De Feyter et al. 1993), pthA is flanked by nearly identical 62bp terminal inverted
repeats that precisely define the limits of homology. The predicted amino acid sequence
encoded by pthA is presented in Fig. 4-2. PthA has a calculated molecular weight of
122kD and an isoelectric point of 7.72. The primary region which differentiates the
predicted PthA sequence from the predicted sequences of other members of the family

Figure 4-1. The complete DNA sequence of pthA. Indicated in the figure are the left and right 62-bp terminal
inverted repeats (IRL and IRR) flanking pthA gene, -35 and -10 elements, Shine-Dalgamo (SD) ribosome-binding site,
methionine (M) start codon, the first 102-bp tandem repeat, and stop codon (*).

1
101
201
301
401
501
601
701
801
901
1001
1101
1201
1301
1401
1501
1601
1701
1801
1901
2001
2101
2201
2301
2401
2501
2601
2701
2801
2901
3001
3101
3201
3301
3401
3501
3601
3701
3801
3901
4001
4101
4201
AGCACGGTAC CCGGCGGTCG CATGACGGGG TTTATGCATT CTTGCATCGC CCGGGCGCAA CCGCTATTGT GCACATGCAC CACCGACCGG ACCCTGGAGA
CAGGTGTGGT GCCTCGATAG ATCGATAGCT TCACCCGATC TATCCCTTGC CTGGAGCGGG ATACGGCAGA AATGCCCAAA GAGCCTACCG AGTAGTGAAT
ACCGAAAGCG CTCCAGCGAT GGGGCGCTTT CTTTTTATAG GGCAATGGTC TGTTCGGCCA AGCGCCGCCG GTACACAGTC TCAACCTTGT AGAGCTTCCC
GCGCGGGAGT TTATGGATGG TTGTCCCGTC GACGGGGTAG CAGGAGATGA GGGTCGGCAG GGATTGGTGT AAAAAACAGC CAAAAGTGAG CTAACTCGCT 62bp IRL
GTCAGCACAG AAATTTTTCA CAACCTTCTG CCGATCCTCC ATGCGGGTCC GTGATCGCCT TCATGTCTGC GCCTCACCCT GGTCGTCGAG GGTTGCCAGG
ATCACCCGAA GTTGTGTACT GCCATGCGGC CTCGGAAGCT ATGTAGGAAC CACAGACCGC TAGTCTGGAG GCGACCATGT AAAGAGGTAT GCCTGATGGA
- 35 - 10 SO M
TCCCATTCGT TCGCGCACAC CAAGTCCTGC CCGCGAGCTT CTGCCCGGAC CCCAACCCGA TGGGGTTCAG CCGACTGCAG ATCGTGGGGT GTCTCCGCCT
GCCGGCGGCC CCCTGGATGG CTTGCCCGCT CGGCGGACGA TGTCCCGGAC CCGGCTACCA TCTCCCCCTG CCCCCTCACC TGCGTTCTCG GCGGGCAGCT
TCAGTGACCT GTTACGTCAG TTCGATCCGT CACTTTTTAA TACATCGCTT TTTGATTCAT TGCCTCCCTT CGGCGCTCAC CATACAGAGG CTGCCACAGG
CGAGTGGGAT GAGGTGCAAT CGGGTCTGCG GGCAGCCGAC GCCCCCCCAC CCACCATGCG CGTGGCTGTC ACTGCCGCGC GGCCGCCGCG CGCCAAGCCG
GCGCCGCGAC GACGTGCTGC GCAACCCTCC GACGCTTCGC CGGCCGCGCA GGTGGATCTA CGCACGCTCG GCTACAGCCA GCAGCAACAG GAGAAGATCA
AACCGAAGGT TCGTTCGACA GTGGCGCAGC ACCACGAGGC ACTGGTCGGC CATGGGTTTA CACACGCGCA CATCGTTGCG CTCAGCCAAC ACCCGGCAGC
GTTAGGGACC GTCGCTGTCA AGTATCAGGA CATGATCGCA GCGTTGCCAG AGGCGACACA CGAAGCGATC GTTGGCGTCG GCAAACAGTG GTCCGGCGCA
CGCGCCCTGG AGGCCTTGCT CACGGTGGCG GGAGAGTTGA GAGGTCCACC GTTACAGTTG GACACAGGCC AACTTCTCAA GATTGCAAAA CGTGGCGGCG
TGACCGCAGT GGAGGCAGTG CATGCATGGC GCAATGCACT GACGGGTGCC CCCCTGAACC TGACCCCGGA GCAGGTGGTG GCCATCGCCA GCAATATTGG
1st 102bp Internal tandem repeat
TGGCAAGCAG GCGCTGGAGA CGGTGCAGCG GCTGTTGCCG GTGCTGTGCC AGGCCCATGG CCTGACCCCG GAGCAGGTGG TGGCCATCGC CAGCAATGGC
GGCAAGCAGG CGCTGGAGAC GGTGCAGCGG CTGTTGCCGG TGCTGTGCCA GGCCCATGGC CTGACCCCGG AGCAGGTGGT GGCCATCGCC AGCAATATTG
GTGGCAAGCA GGCGCTGGAG ACGGTGCAGC GGCTGTTGCC GGTGCTGTGC CAGGCCCATG GCCTGACCCC GGAGCAGGTG GTGGCCATCG CCAGCAATAT
TGGTGGCAAG CAGGCGCTGG AGACGGTGCA GCGGCTGTTG CCGGTGCTGT GCCAGGCCCA TGGCCTGACC CCGGAACAGG TGGTGGCCAT CGCCAGCAAT
ATTGGTGGCA AGCAGGCGCT GGAGACGGTG CAGGCGCTGT TGCCGGTGCT GTGCCAGGCC CATGGCCTGA CCCCGGAGCA GGTGGTGGCC ATCGCCAGCA
ATGGCGGTGG CAAGCAGGCG CTGGAGACGG TGCAGCGGCT GTTGCCGGTG CTGTGCCAGG CCCATGGCCT GACCCCGGAC CAGGTGGTGG CCATCGCCAG
CCACGATGGC GGCAAGCAGG CGCTGGAGAC GGTGCAGCGG CTGTTGCCGG TGCTGTGCCA GGCCCATGGC CTGACCCCGC AGCAGGTGGT GGCCATCGCC
AGCAATGGCG GTGGCAAGCA GGCGCTGGAG ACGGTGCAGC GGCTGTTGCC GGTGCTGTGC CAGGCCCATG GCCTGACCCC GGAGCAGGTG GTGGCCATCG
CCAGCCACGA TGGCGGCAAG CAGGCGCTGG AGACGGTGCA GCGGCTGTTG CCGGTGCTGT GCCAGGCCCA TGGCCTGACC CCGGAGCAGG TGGTGGCCAT
CGCCAGCAAT GGCGGTGGCA AGCAGGCGCT GGAGACGGTG CAGCGGCTGT TGCCGGTGCT GTGCCAGGCC CATGGCCTGA CCCCGGAGCA GGTGGTGGCC
ATCGCCAGCA ATGGCGGTGG CAAGCAGGCG CTGGAGACGG TGCAGCGGCT GTTGCCGGTG CTGTGCCAGG CCCATGGCCT GACCCCGGAG CAGGTGGTGG
CCATCGCCAG CAATGGCGGT GGCAAGCAGG CGCTGGAGAC GGTGCAGCGG CTGTTGCCGG TGCTGTGCCA GGCCCATGGC CTGACCCTGG ACCAGGTCGT
GGCCATCGCC AGCAATGGCG GTGGCAAGCA GGCGCTGGAG ACGGTGCAGC GGCTGTTGCC GGTGCTGTGC CAGGCCCATG GCCTGACCCC GGAGCAGGTG
GTGGCCATCG CCAGCAATAG CGGTGGCAAG CAGGCGCTGG AGACGGTGCA GCGGCTGTTG CCGGTGCTGT GCCAGGCCCA TGGCCTGACC CCGGACCAGG
TGGTGGCCAT CGCCAGCCAC GATGGCGGCA AGCAGGCGCT GGAGACGGTG CAGCGGCTGT TGCCGGTGCT GTGCCAGGCC CATGGCCTGA CCCCGGAGCA
GGTCGTGGCC ATCGCCAGCC ACGATGGCGG CAAGCAGGCG CTGGAGACGG TGCAGCGGCT GTTGCCGGTG CTGTGCCAGG CCCATGGCCT GACCCCGGAG
CAGGTGGTGG CCATCGCCTG CAATGGCGGT GGCAAGCAGG CGCTGGAGAC GGTGCAGCGG CTGTTGCCGG TGCTGTGCCA GGCCCATGGC CTGACCCCGG
AGCAGGTGGT GGCCATCGCC AGCAATGGCG GCGGCAGGCC GGCGCTGGAG AGCATTGTTG CCCAGTTATC TCGCCCTGAT CCGGCGTTGG CCGCGTTGAC
CAACGACCAC CTCGTCGCCT TGGCCTGCCT CGGCGGACGT CCTGCGCTGG ATGCAGTGAA AAAGGGATTG CCGCACGCnC CGGCCTTGAT CAAAAGAACC
AATCGCCGTA TTCCCGAACG CACATCCCAT CGCGTTGCCG ACCACGCGCA AGTGGTTCGC GTGCTGGGTT TTTTCCAGTG CGACTCCCAC CCAGCGCAAG
CATTTGATGA CGCCATGACG CAGTTCGGGA TGAGCAGGCA CGGGTTGTTA CAGCTCTTTC GCAGAGTGGG CGTCACCGAA CTCGAAGCCC GCAGTGGAAC
GCTCCCCCCA GCCTCGCAGC GTTGGGACCG TATCCTCCAG GCATCAGGGA TGAAAAGGGC CAAACCGTCC CCTACTTCAA CTCAAACGCC GGACCAGGCG
TCTTTGCATG CATTCGCCGA TTCGCTGGAG CGTGACCTTG ATGCGCCCAG CCCAACGCAC GAGGGAGATC AGAGGCGGGC AAGCAGCCGT AAACGGTCCC
GATCGGATCG TGCTGTCACC GGTCCCTCCG CACAGCAATC GTTCGAGGTG CGCGCTCCCG AACAGCGCGA TGCGCTGCAT TTGCCCCTCA GTTGGAGGGT
AAAACGCCCG CGTACCAGTA TCGGGGGCGG CCTCCCGGAT CCTGGTACGC CCACGGCTGC CGACCTGGCA GCGTCCAGCA CCGTGATGCG GGAACAAGAT
GAGGACCCCT TCGCAGGGGC AGCGGATGAT TTCCCGGCAT TCAACGAAGA GGAGCTCGCA TGGTTGATGG AGCTATTGCC TCAGTGAGGC TCAGTCGGTG
ACTACCTGAG CGTCGGCAGG GATTGGTGTA AGTAACCTTT ACTGACAGCG AGTTAGCCCA CTTTTGGCTG TTTTTTACAC AAATCCCTGC CTCCCCTCAC 62bp IRR
CTGGGCACGC AGTCCCTCGG TGCGCATCAA TCGATGCACC GATGGCGACT GCAACGTCAC CAGATCGGTA GATCC
O'
00

Figure 4-2. Predicted amino acid sequence of PthA. The 17.5 internal tandem repeats
are aligned and numbered for comparison. Letters A through K represent different types
of repeats, indicates the repeats unique for PthA protein. Bold and underlined letters
indicate an amino acid deletion, substitution or specific combinations of amino acid
residues within a given repeat that is unique to PthA (and the combination is not found
in any repeat in any previously published member of the gene family). Three nuclear
localization consensus sequences are double underlined.

70
1 MDPIRSRTPS PARELLPGPQ PDGVQPTADR GVSPPAGGPL DGLPARRTMS RTRLPSPPAP
61 SPAFSAGSFS DLLRQFDPSL FNTSLFDSLP PFGAHHTEAA TGEWDEVQSG LRAADAPPPT
121 MRVAVTAARP PRAKPAPRRR AAQPSDASPA AQVDLRTLGY SQQQQEKIKP KVRSTVAQHH
181 EALVGHGFTH AHIVALSQHP AALGTVAVKY QDMIAALPEA THEAIVGVGK QWSGARALEA
241 LLTVAGELRG PPLQLDTGQL LKIAKRGGVT AVEAVHAWRN ALTGAPLN
LTPEQWAIA
SNIGGKQALE
TVQRLLPVLC
QAHG
1
(A)
LTPEQWAIA
SN_GGKQALE
TVQRLLPVLC
QAHG
2
(B)
LTPEQWAIA
SNIGGKQALE
TVQRLLPVLC
QAHG
3
(A)
LTPEQWAIA
SNIGGKQALE
TVQRLLPVLC
QAHG
4
(A)
LTPEQWAIA
SNIGGKQALE
TVQALLPVLC
QAHG
5
(C)
LTPEQWAIA
SNGGGKQALE
TVQRLLPVLC
QAHG
6
(D)
LTPDQWAIA
SHDGGKQALE
TVQRLLPVLC
QAHG
7
(E)
LTPQQWAIA
SNGGGKQALE
TVQRLLPVLC
QAHG
8
(F)
LTPEQWAIA
SHDGGKQALE
TVQRLLPVLC
QAHG
9
(G)
LTPEQWAIA
SNGGGKQALE
TVQRLLPVLC
QAHG
10
(D)
LTPEQWAIA
SNGGGKQALE
TVQRLLPVLC
QAHG
11
(D)
LTPEQWAIA
SNGGGKQALE
TVQRLLPVLC
QAHG
12
(D)
LTLDQWAIA
SNGGGKQALE
TVQRLLPVLC
QAHG
13
(H)
LTPEQWAIA
SNSGGKQALE
TVQRLLPVLC
QAHG
14
(I)
LTPDQWAIA
SHDGGKQALE
TVQRLLPVLC
QAHG
15
(E)
LTPEQWAIA
SHDGGKQALE
TVQRLLPVLC
QAHG
16
(G)
LTPEQWAIA
CNGGGKQALE
TVQRLLPVLC
QAHG
17
(J)
LTPEQWAIA
SNGGGRPALE
SIVAQLSRPD
PALA
18
(K)
900
A
LTNDHLVALA
CLGGRPALDA
VKKGLPHAPA
LIKRTNRRIP
ERTSHRVADH
951
AQWRVLGFF
QCDSHPAQAF
DDAMTQFGHS
RHGLLQLFRR
VGVTELEARS
GTLPPASQRW
1011
DRILQASGMK
RAKPSPTSTQ
TPDQASLHAF
ADSLERDLDA
PSPTHEGDQR
RASSRKRSRS
1071
DRAVTGPSAQ
QSFEVRAPEQ
RDALHLPLSW
RVKRPRTSIG
GGLPDPGTPT
AADLAASSTV
1131
MREQDEDPFA
GAADDFPAFN
EEELAWLMEL
LPQ*

71
is in the central portion of the protein, characterized by 17.5 nearly identical, tandemly
arranged, 34 amino acid repeats. As shown in Fig. 4-2, these repeats may be classified
into 11 types (A-K) depending on slight differences in sequence. Among them, repeat
types B, E, H, J are only found in PthA, and not in the predicted peptide sequences of
any other published family member. Repeat 2 (type B) resulted from an amino acid
deletion and repeat 17 (type J) resulted from an amino acid substitution; neither this
deletion nor this substition has been observed in any other member of the gene family.
Repeat types E (repeats 7 and 15) and H (repeat 13) resulted from unique combinations
of amino acids in specific positions. In addition to the presence of unique repeats, PthA
differs from its homologues in the arrangement of repeat types within the region.
Gene replacement and selection of pthA recombinants in X. citri. X. citri mutant
Xcl.2 (pthA::nptI-sac) was generated from wild type strain 3213 by marker exchange,
using pUFYlO.l. Marker exchange was confirmed by Southern hybridization analyses
(Fig. 4-3, compare lanes 1 and 2). When inoculated on plants, Xcl.2 was unable to
induce cankers on citrus or to induce an HR on bean (Table 4-1). By plating Xcl.2 on
a sucrose-containing medium and screening for sucrose resistant and kanamycin sensitive
colonies, many marker-evicted strains of X. citri were obtained at a frequency of 5.22
± 1.8 x 10'5.
To examine the recombination events, total DNAs were extracted from twenty-
four marker-evicted derivatives of Xcl.2, digested with BamUl and probed with the 35P-
labelled internal Stull Hindi fragment (containing the tandemly repeated region) of pthA.

72
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18
4Kb-
3Kb- 5 - t ^ L 2 2— —
2Kb-
Figure 4-3. Southern hybridization of BamHl digested, total DNA from X. citri strain
3213 and its pthA mutant derivatives. The blot was probed with the internal 2kb
Stul/Hindi fragment (containing the tandemly repeated region) of pthA. Strain 3213
carries four BamHl fragments which hybridize to pthA; the largest fragment, indicated
by the arrow, encodes pthA. Not shown in lane 2 are two additional hybridizing
fragments (1.8 and 1.6 kb in size), resulting from the additional BamHl site in nptl-sac.
Lane 1, 3213; lane 2, Xcl.2 (pthA::nptI-sac); lanes 3-18, nptl-sac-evicled derivatives of
Xcl.2: XcSl, XcS15, XcS19, XcS24, XcS27, XcS30, XcS12, XcS13, XcS17, XcS18,
XcS26, XcS29, XcS32, XcS23, XcS34, XcS26 and XcS21, respectively.

73
Table 4-1. Phenotypes of Xanthomonas citri marker-evicted strains on citrus and bean.
Host
O Li dlllo
citrus
bean
Xc3213
C
HR
Xcl.2
0
0
XcSl, XcS9, XcS14 & XcS15
c
HR
XcS5
c
0
XcS17 & XcS20
wC
wHR
XcS12, XcS13 & XcS24
HR
wHR
XcS7
wHR
wHR
XcS6 & XcSll
wHR
0
XcS2*
0
0
*: Additional 11 strains (XcS3, XcS4, XcS8, XcSIO, XcS16, Xcsl8, XcS19, XcS20,
XcS21, XcS22 and XcS23); C, canker lesions; wC, weak canker; HR, hypersensitive
reaction; wHR, weak hypersensitive reaction; 0, no symptoms.

74
As shown in Fig. 4-3, the different derivatives of Xcl.2 exhibited different sized bands
(including those similar in size to pthA), indicating that new pthA homologues were
regenerated by homologous recombination among the tandem repeats of pthA. There was
no evidence of recombination between pthA and the other three DNA fragments that
hybridize with pthA, but the presence of the other three fragments raised the possibility
of intergenic recombination.
The twenty-four marker-evicted derivatives of Xcl.2 were inoculated on citrus
and bean plants. They exhibited different plant reaction phenotypes, including some
entirely new phenotypes, distinct from the wild type 3213 strain (Table 4-1). Five
derivatives reverted back to the parental 3213 phenotype, inducing strong canker lesions
on citrus (XcSl, XcS9, XcS14 and XcS15). Two derivatives caused weak canker (XcS17
and XcS20). Six derivatives elicited a previously unreported HR on grapefruit (XcS12,
XcS13 and XcS24). Eleven derivatives caused no pathogenic symptoms. In contrast to
wild type strain 3213, XcS5 retained the ability to elicit hyperplastic cankers on citrus,
but lost the ability to elicit HR on bean, indicating that these two pleiotropic functions
are separately encoded on pthA.
Intragenic recombination of pthA in X. campestris and E. coli. To demonstrate
the occurrence of intragenic recombination, plasmid pUFYl .48 was introduced into three
bacterial strains which carry no DNA that hybridizes with pthA: X. campestris pv.
citrumelo strain 3048, E. coli strain DH5a (recA~) and E. coli strain JM83 (recA+).
Thetransconjugants 3048/pUFY1.48, DH5a/pUFY1.48or JM83/pUFY1.48 were plated
on sucrose-containing medium and screened for marker-eviction. Again many marker-

75
evicted derivatives were obtained. The frequencies of marker-eviction of pUFY1.48 in
3048, DH5a and JM83 were 4.05 ± 0.44 x 10"4, 1.06 ± 0.13 x 10'2 and 0.97 ± 0.14
x 10"2, respectively.
To examine the recombination events, plasmid DNAs were analyzed from 24
marker-evicted strains from 3048 and 14 strains from DH5a. As shown in Fig. 4-4,
different sized BamHl fragments were observed, demonstrating the generation of new
pthA homologues by intragenic recombination. Southern hybridizations of total DNAs
extracted from these strains probed with the internal Stull Hindi fragment of pthA showed
only the presence of the same bands as those generated from the plasmid digestions (data
not shown).
The plasmids derived from the 24 3048 (pCS series) and 14 DH5a (pDS series)
marker evicted strains were individually introduced into X. citri strain B21.2 (a
nonpathogenic, pthA::Tn5-gusA derivative of 3213, which is virulent only on citrus),
3048 (virulent on citrus and bean) and X. campestris pv. malvacearum Xcml003 (virulent
only on cotton). The resulting transconjugants were inoculated onto citrus, bean and
cotton plants (Fig. 4-5). Similar to the results observed with derivatives of Xcl.2, these
pthA recombinants exhibited alterred phenotypes and/or host specificities. For example,
pCS23 conferred to B21.2 and to 3048 the ability to cause stronger canker lesions on
citrus than that conferred by pZit45 (pthA), and a weak HR often preceeded the canker
symptoms. Unlike pthA on pZit45, however, the recombinant gene on pCS13 conferred
to B21.2 or 3048 the ability to elicit canker lesions on citrus, but not HR on bean.
These results confirmed those indicated by recombinant X. citri mutant XcS5 (refer

Figure 4-4. A. Intragenic recombinants of pthA generated in X. campestris pv. citrumelo
3048. Plasmid DNAs were digested with BamHl and fractionated in 0.7% agarose gel.
Lane 1, pZit45 (pthA)\ lane 2, pUFY1.48 (pthAv.npt-sac); lane 3-19, pCS23, pCS13,
pCS16, pCS2, pCS17, pCS6, pCSl, pCS14, pCS5, pCS7, pCSll, pCS12, pCS15,
pCS19, pCS21, pCS18 and pCS22, respectively. B. Intragenic recombinants of pthA
generated in E. coli. Plasmid DNAs were digested with BamUl and fractionated in 0.7%
agarose gel. Lane 1, pZit45; lane 2, pUFY1.48; lanes 3-16, pDS2, pDS5, pDS4,
pDS13, pDSl, pDSlO, pDS14, pDS9, pDS3, pDS6, pDS7, pDS8, pDSll and pDS12,
respectively.

77

Fig. 4. Biological activity and restriction enzyme analysis of pthA intragenic recombinants. Only the 102bp repeated
region of the gene is shown, as represented by tandemly arranged blocks. Solid blocks indicate repeats which were
not likely affected by recombination. Open blocks indicate repeats which may be composite and/or located on the
opposite side of the deleted region. Missing blocks represent repeats which were deleted by recombination. The Bgll
and Sspl sites are indicated by upward and downward bars, respectively. The position of the npt-sac marker insertion
(in the 10th repeat) in pUFY1.48 is indicated by the inverted triangle. a: Plasmids pCS2, pCS3, pCS7, pCS12,
pCS15, pCS19, pCS20, pCS21, pDS4, pDS13 and pDS14 confered similar phenotypes; b: Plasmids pCS4, pCS9,
pCSIO and pDSlO confered similar phenotypes; c: Plasmids pCS18, pDS6, pDS7, pDS8 and pDSll confered similar
phenotypes. 0, no activity; C, canker lesions; mC, intermediate level canker; wC, weak canker; HR,
hypersensitive reaction; wHR, weak HR; rWS, reduced watersoaking (assayed in 3048 and Xcml003).

Plasmid
PUFY1-48 Ssp,
pZit45 (17.5)
pCS23 (17.5)
pDS2 (17.5)
pCS13 (10.5)
pCS16 (11.5)
pCS17 (12.5)
pDS5 (13.5)
pCS5 (16.5)
pCS1 (8.5)
pCS6 (10.5)
pCS11 (13.5)3
pDS9 (16.5)b
pCS22 (11.5)°
Activity in
B21.2 or 3048
Xcm1003
Citrus
Bean
Cotton
0
0
0
C
HR
rWS
C/wHR
HR
rWS
C
HR
rWS
c
0
rWS
mC
0
rWS
wC
0
HR
wC
0
HR
wC
0
wHR
wHR
HR
HR
HR
HR
HR
rWS
0
HR
0
0
wHR
0
0
rWS
««j
vo

80
above): the ability to confer hyperplastic cankers on citrus and the HR on bean are
independently encoded on pthA.
Plasmids pCSl and pCS6 conferred ability to elicit an HR on citrus and bean to
B21.2 and 3048, and ability to elicit an HR on 9 tested cotton lines (Ac44, AcBl, AcB2,
AcB4, AcB5, Acb6, Acb7, AcBIn and AcBIn3) to Xcml003. All pthA derivatives tested
which conferred the ability to induce HR to Xcml003 on normally susceptible cotton line
Ac44, also elicited HR on all tested cotton resistance lines. All pthA derivatives tested
were able to confer ability to induce an HR to Xcml003 on one or more resistance cotton
lines, indicating that all pthA derivatives tested were functional genes.
Most of the variation within the predicted 34 amino acid repeat sequence of pthA
occurred at positions 12 and 13, and most repeats encoded either asparagine/glycine,
asparagine/isoleucine or histidine/aspartate at these positions. The latter two amino acid
combinations in the pthA repeats could be detected at the DNA level by presence of Sspl
or Bgll restriction sites, respectively. The Sspl restriction enzyme cut specifically after
the 36th nucleotide of the 1st, 3rd, 4th and 5th repeats of pthA, while the Bgll enzyme
cut specifically after the 38th nucleotide of the 7th, 9th, 15th, 16th repeats. By
comparing the Bgll and Sspl fragments of intragenic recombinants with those of pthA,
the specific repeats deleted or affected in 12 recombinants were determined (Fig. 4-5).
Deletions involving the 7th, 8th, and 9th repeats always reduced or abolished the canker-
eliciting ability of pthA (pCS16, pCS17, pDS5, pCSl, pCS6, pCSl 1, pCS22). Deletions
involving the 4th, 5th, and 6th repeats abolished all avirulence specificity; these
derivatives conferred ability to induce an HR on all citrus, bean and cotton lines tested.

81
Although recombinants with deletions in the 1st, 2nd, 16th or 17th repeats were not
obtained, these results indicated that the specific order of the first nine repeats of pthA
was important for ability to elicit cankers on citrus and for specificity of avirulence.
Deletions of the 10th through the 15th repeats appeared to alter avirulence specificity.
Discussion
The complete DNA sequence of pthA revealed that pthA is highly homologous to
avrbó, avrBs3, avrBs3-2 and avrXalO along its entire length. The pthA gene encoded
a protein of 1163 amino acids with calculated molecular weight of 122 kD. Western blot
analyses similar to those previously published for AvrBs3 (Knoop et al. 1991), but using
anti-PthA serum, revealed that a 122 kD protein was constitutively expressed in X. citri
and X. campestris pv. citrumelo transconjugants containing pthA (refer to Chapter 5).
Despite few differences in the predicted amino acid sequences of all members of the
avr/pth gene family, pthA has a unique ability to induce cell divisions on citrus leaf
mesophyll in addition to its pleiotropic ability to confer avirulence to pathogens of cotton
and bean (Swarup et al. 1992). Both functions were previously found to be localized to
the repeated region of pthA (refer to Chapter 2).
DNA sequence analysis revealed that there were five specific repeats only found
in pthA, but not in other published members of the avr/pth gene family sequenced to date
{avrbó, avrBs3, avrBs3-2 and avrXalO). By analyzing 12 intragenic recombinants with
BglI and Sspl restriction enzyme digestion, we were able to approximately locate the

82
regions important for avirulence and pathogenic specificities. The first six repeats of
pthA appear to be essential for the pathogenicity function and any specificity (avirulence
or pathogenicity) at all. Deletions involving repeats 7 to 9 reduced or abolished canker-
eliciting ability of pthA on citrus. Deletions involving repeats 10 to 16 determined
avirulence specificity. These findings may also help to explain previously published data
on avrBs3 deletion derivatives (Herbers et al. 1992; Bonas et al. 1993). From the
published data, all deletions affecting repeats 1 through 6 of avrBs3 abolished all
specificity, although not avirulence per se.
The role of the tandem repeats in determining pathogenic or avirulence specificity
is striking, but it is not known how the repeats determine these functions. One of the
important biological functions for repetitive domains in prokaryotic and eukaryotic
proteins is ligand binding. In a family of clostridial and streptococcal ligand-binding
protein, for example, conserved C-terminal repeat sequences function as binding sites
(Wren 1991). Similarly, 34 amino acid repetitive motifs called tetratricopeptides (TPR)
found in proteins encoded by many mitotic genes (e.g., CDC16, CDC23, nuc2+ of
yeast, BimA of Aspergillus) are implicated to pair with WD-40 repeats found in many
proteins including the /3-subunit of G-proteins and some transcriptional factors (Goebl
and Yanagida 1991; van der Voom and Ploegh 1992). It is possible that the repetitive
domain of Pth or Avr proteins may also function as a ligand binding site, specifically
interacting with corresponding plant receptors.
In this study, a nptl-sac cartridge was inserted into the tandemly repeated region
of pthA to serve as a selectable marker to detect homologous recombination between

83
intragenic repeats. Since the production of levansucrase encoded by sac is lethal to gram¬
negative bacteria in the presence of sucrose, marker evicted strains can be selected by
the nptl-sac marker (Ried and Collmer 1987). The regeneration of functional genes from
the pthAwnptl-sac fragment on pUFY1.48 in 3048 or DH5a, which contains no DNA
hybridizing to pthA, demonstrated that intragenic recombination occurred. This was
confirmed by restriction analysis of internal tandem repeats of recombinants.
Intragenic recombinants of pthA were generated by two possible mechanisms.
The first mechanism involves a single crossover event among intragenic repeats on the
same plasmid, resulting in the deletion of nptl-sac marker and restoration of a functional
gene. The second mechanism requires a double crossover between plasmids. The
frequency of pthA recombination in E. coli was observed to be recA independent. Strain
DH5a (recA) yielded recombinants at a frequency of 1.06 ± 0.13 x 10'2, while strain
JM83 (recA+) yielded recombinants at a frequency of 0.97 ± 0.14 x 10'2. Since
recombination involving small repeated sequences (less than lkb) is affected only slightly
by recA mutations (Petes and Hill 1988), recombination of pthA was expected to be recA
independent.
The intragenic recombination frequencies of pthA observed in the E. coli were
higher than those observed in X. campestris pv. citrumelo (4.05 ± 0.44 x 10-4) and X.
citri (5.22 + 1.8 x 10‘5). This may be due to differences in the copy number of
plasmids in which intragenic recombination occurred. Recombination of pthA in X. citri
strain Xcl.2 involved a native plasmid, while recombination in E. coli and X. campestris
pv. citrumelo involved pUFY1.48, which appeared (based on our plasmid DNA

84
extractions) to replicate at a higher copy number in E. coli than in Xanthomonas.
However, since we did not determine plasmid copy number, these results may be due to
different genetic backgrounds in each strain.
In animal pathogens, some pathogenicity genes encode surface proteins with a
distinctive arrangement of tandem repeats which may provide a protective, strain-specific
conformational epitope for evasion of host immunity (Hoft et al. 1989; McConkey et al.
1990; Hollingshead et al. 1987). Intragenic recombination between homologous repeats
has been found to be responsible for antigenic variations in the pathogens such as
Streptococcus, Neisseria, Salmonella, Plasmodium and Trypanosoma (McConkey et al.
1990; Hollingshead et al. 1987; Hagblom et al. 1985; Frankel et al. 1989). The
resulting antigenic variability allows pathogens to escape the host immune response. In
this study, we demonstrated that intragenic recombination of a single plant bacterial gene
can: 1) alter host specificity, 2) generate new pathogenic phenotypes, and 3) evade host
plant defenses. The evolutionary success of many plant pathogens may rely on their
ability to avoid host recognition. Tandemly repeated motifs in pthA and other members
of the gene family provide hot spots for recombination, and thus may accelerate evolution
of this gene family. Intragenic recombination between inexact repeats not only
introduces genetic variation, but also has the potential to amplify mutagenic effects. As
shown in the results, homologous recombination can lead to either duplication or deletion
of repeat blocks. A mechanism to generate genetic variation at high frequency may also
have the negative side effect of generating unnecessary, gratuitous avr genes.

85
Previously reported natural mechanisms for race change included deletions (De
Feyter et al. 1993), point mutations (Kobayashi et al. 1990; Joosten et al. 1994) or
transposon inactivation (Kearney et al. 1988) of avr genes and always resulted in a loss
of avirulence on a specific host plant, but not gains of specific avirulence. Because of
the unique 102bp tandem repeats and their role in determining both pathogenic and
avirulence specificities, intragenic recombination of pthA not only caused the loss of
pathogenic or avirulence phenotypes on a specific host, but also simultaneously resulted
in the gain of new pathogenic phenotypes and avirulence specificities. Many intragenic
recombinants of pthA lost the ability to confer canker-like lesions on citrus to 3048, but
conferred a new phenotype by reducing the watersoaking symptoms of 3048. Some
intragenic recombinants lost the ability to confer cankers on citrus and instead elicited
an HR on citrus, indicating that normsensitive (canker) and hypersensitive reactions may
share a very similar signal transduction pathway as initially suggested by Klement (1982).
Other recombinants lost the ability to elicit HR on bean, but retained the ability to induce
cankers on citrus. Therefore, the pleiotropic pathogenic and avirulence functions of pthA
were separated by the rearrangement of tandem repeats through intragenic recombination.
Members of this avr/pth gene family have been found in many Xanthomonas
species and pathvars. When present in appropriate xanthomonads, many are able to elicit
an HR on diverse plant species including cotton, citrus, bean, tomato, pepper and rice.
As shown in Fig. 4-5, one intragenic recombinant of pthA (pCS6) has the ability to elicit
a strong HR on plants from taxonomically distinct plant families such as Rutaceae
(citrus), Leguminosae (bean) and Malvaceae (cotton), suggesting that the corresponding

86
resistance genes may be widespread in different plant species. Tomato disease resistance
gene Pto, which encodes a protein kinase, hybridizes with multiple DNA fragments
found in a variety of different plants, indicating the presence of Pto homologues (Martin
et al. 1993). During the evolutionary process, members of the avr/pth gene family might
have moved horizontally among diverse xanthomonads pathogenic to a wide range of
plants (Gabriel et al. 1993; De Feyter et al. 1993), thereby enabling interactions with
members of the same plant gene family in a wide range of plants. As a result, members
of this avr/pth gene family may have accelerated the evolution and diversity of the
corresponding plant R gene family.
The gene-for-gene concept is a major model for research on the reciprocal
evolution between plants and parasites (Thompson and Burdon 1992). In the pathogen,
avirulence genes may be mutated to overcome resistance by evading host recognition.
In the plant, resistance genes may also evolve to recognize new avirulence genes,
resulting in new resistance genes. Recent genetic mapping data have shown that disease
resistance loci (especially those corresponding to different races of the same pathogen)
are often clustered within small genetic intervals, suggesting that intralocus recombination
may be responsible for generating resistance specific for new races of plant pathogens
(Dickinson et al. 1993; Martin et al. 1993; Pryor 1987). Intragenic recombination of
pthA (a member of an avr gene family) demonstrated in this research provides compelling
evidence that intragenic and intralocus recombination may play an important role during
the reciprocal evolution of gene-for-gene interactions.

CHAPTER 5
PLANT NUCLEAR TARGETING SIGNALS ENCODED BY A FAMILY
XANTHOMONAS AVIRULENCE/PATHOGENICITY GENES
Introduction
Interactions of microbial plant pathogens carrying avirulence (avr) genes with host
plants carrying specific resistance (R) genes results in plant defense responses, often
observed as a hypersensitive reaction (HR) that is characterized by rapid plant cell death
at the site of infection. Several models have been proposed to explain the genetically
specific recognition between microbial avr genes and plant R genes in these gene-for-gene
interaction (Gabriel and Rolfe 1990), and different models may be applicable to different
systems. In the elicitor/receptor model, avr genes encode a low molecular weight signal
molecule (elicitor) which is perceived by a sensor encoded by the corresponding R genes
and results in an HR. In Pseudomonas syringae pv. tomato, avrD encodes an enzyme
involved in the synthesis of a low molecular weight glycolipid elicitor (Midland et al.
1993; Smith et al. 1993). This elicitor induces an HR specifically on soybean plants
containing the resistance gene Rpg4 (Keen et al. 1990; Keen and Buzzel 1991; Kobayashi
et al. 1990). An alternative model is the dimer hypothesis in which the protein product
of an avr gene interacts directly with the protein product of a plant R gene or with the
R gene itself (Ellingboe 1982). The extracellular peptide encoded by avr4 and avr9 of
87

88
the fungal pathogen Cladosporium fulvum can directly induce a hypersensitive response
(HR) on tomato cultivars carrying the resistance gene Cf4 or Cf9 (Joosten et al. 1994;
van Kan et al. 1991). However, other than these three examples, the microbial signal
molecules that determine avirulence remain unidentified. In all cases, it is unknown how
the signals encoded by aw genes are specifically recognized by host cells with R genes,
triggering plant defense responses.
The first plant R gene, isolated from tomato, was shown to encode a
serine/threonine protein kinase that is homologous to the receptor protein kinase involved
in Brassica pollen-stigma recognition and to the Raf protein kinase involved in the
mammalian Ras signaling pathway (Martin et al. 1993). Since transcriptional activation
of plant defense genes is modulated by phosphorylation (Felix et al. 1991; Yu et al.
1993), and can be blocked by inhibitors of mammalian protein kinases (Raz and Fluhr
1993), it was proposed that recognition of aw signals by R gene products triggers
phosphorylation cascades, leading to plant defense responses (Lamb 1994).
In mammalian systems, two nearly complete signal transduction pathways used
by growth factors and/or cytokines have been elucidated (Culotta and Koshland 1993).
The best known is the Ras pathway, centered around the protein product of the prototype
oncogene ras (Hall 1993, Schlessinger 1993). Upon binding to extracellular signals, the
activated receptor protein kinase transduces signals to the GRB2, SOS, Ras and Raf
proteins. Raf (a Pto homolog) then initiates the mitogen-activated protein (MAP) kinase
cascade (Crews and Erikson 1993; Nishida and Gotoh 1993), which brings protein
kinases into the nucleus, activating nuclear transcriptional factors. The other route from

89
membrane to nucleus is more direct. In these cases, tyrosine kinases pick up signals
from receptors in the membrane and activate transcription factor subunits in the
cytoplasm, which are transported into the nucleus and induce transcription (Hunter 1993,
Kishimoto et al. 1994). In both cases, activated protein kinases and/or transcriptional
factors need to enter the nucleus in order to induce gene expression which may lead to
different physiological outcomes, including oncogenesis and apoptosis.
Viral pathogens encode some transcriptional factors that can directly enter the
nucleus, modulate gene expression of host cells. The bacterial plant pathogen
Agrobacterium tumafaciens also encodes several proteins targeted to the plant cell nucleus
(Citovsky and Zambryski 1993). Transport of proteins into the nucleus is an active
process and requires that proteins contain suitable nuclear localization sequences (NLSs,
Nigg et al 1991). NLSs are usually short stretches of 8-10 amino acids characterized by
basic amino acid residues and proline. NLSs are retained in the mature protein, may
be located at any position as long as they exposed on the protein surface, and can be
present in multiple copies. NLSs are recognized by the receptor, called NLS-binding
proteins, which direct NLS-containing proteins into nucleus through nuclear pores (Silver
1991). NLSs from different organisms, including the prokaryote Agrobacterium, are
functional throughout the eukaryotic kingdom. Most of the characterized NLS-containing
proteins were reported from mammalian virus and animal cells. Many are oncogene
products, hormone receptors and transcriptional factors, including those involving later
steps of the Ras signaling pathway such as the transcriptional factor AP-1 (Jun/Fos) (Kerr
et al. 1992). A few NLS-containing proteins from Agrobacterium, plant and plant virus

90
have also been characterized (Citovsky and Zambryski 1993; Restrepo et al. 1990; van
der Krol and Chua 1991; Varagona et al., 1991, 1992). To the best of my knowledge,
however, NLSs encoded by aw genes of plant pathogens have not been reported.
Recently, a large family of avirulence (avr)/pathogenicity (pth) genes has been
identified in many species and pathovars of Xanthomonas, a major group of plant
bacterial pathogens. Members includes awBs3, avrBs3-2 and avrBsP of X. campestris
pv. vesicatoria (Bonas et al. 1989, 1993; Canteros et al. 1991), awB4,avrb6, awb7,
awBln, awBlOl, and awB102 of X. c. pv. malvacearum (De Feyter and Gabriel 1991;
De Feyter et al. 1993), pthA of X. citri (Swarup et al 1991, 1992), and avrxa5, awXa7
and avrXalO of X. oryzae pv. oryzae (Hopkin et al. 1992). This avr/pth gene family
comprises the majority of described Xanthomonas aw genes and constitutes a large
portion of all aw genes cloned to date (Gabriel et al 1993). The pthA gene of X. citri
and awb6 of X. c. pv. malvacearum are particularly intriguing because of their
pathogenicity functions (Swamp et al. 1992, Chapter 2). Gene pthA is essential for X.
citri to induce cell division specifically on citms plants leading to hyperplastic canker
lesions. Gene awb6 is important for X. c. pv. malvacearum to induce watersoaking
symptoms specifically on cotton. These two genes also function pleiotropically as aw
genes.
Despite the diverse avirulence and pathogenic specificities encoded by these
aw/pth genes, their DNA sequences are nearly identical (95-98%) (De Feyter et al,
1993, Chapter 4). Both pathogenic and avimlence specificities of the aw/pth genes are
determined by the nearly identical, 34 amino acid tandem repeats encoded in the central

91
region of the genes (Herbers et al. 1992; Chapter 2). To account for the precise
recognition and diverse specificities, it has been speculated that the avr/pth protein
product may directly mediate the plant hypersensitive response (Herbers et al. 1992), as
initially suggested in the dimer hypothesis.
In this study, we demonstrated that pthA conferred to X. campestris pv.
malvacearum (which is nonpathogenic on citrus) the ability to elicit hyperplastic canker
symptoms on citrus, independently of bacterial growth in planta. Although the
proteinaceous elicitors encoded by pthA or avrb6 were not detected , three nuclear
localization sequences were identified to be encoded by all members of the Xanthomonas
avr/pth gene family sequenced to date. Furthermore, the C-termini of PthA and Avrb6,
containing the NLSs were shown to be capable of directing /3-glucuronidase into the
nuclei of plant cells.
Materials and Methods
Bacterial strains and plasmids. Escherichia coli strains used were DH5a (Gibco-
BRL, Gaithersburg, MD) and BL21(DE3)pLysS (Novagen, Madison, WI). Xanthomonas
strains used were X. citri 3213 (causes hyperplastic cankers on citrus, Swarup et al.
1991, 1992), its marker exchange mutant Xcl.2 {pthA:\npt-sac, causes no symptom on
citrus), X. campestris pv. citrumelo 3048 (causes water-soaked leaf spots on citrus and
common bean, Swarup et al. 1991, 1992), X. campestris pv. malvacearum XcmH1005
(causes water-soaked leaf spots on cotton) and its mutants XcmH1407 (avrb6::Tn5-gusA,

92
refer to Chapter 2), HM2.2S (seven avr genes deleted, refer to Chapter 3). E. coli
strains were grown in Luria-Bertani (LB) medium (Sambrook et al. 1989) at 37°C or
30°C. Xanthomonas strains were grown in PYGM (peptone-yeast extract-glycerol -
MOPS) medium at 30 °C as described (De Feyter et al. 1990). Plasmids pUC19
(Yanisch-Perron et al. 1985) and pUFR047 (De Feyter et al. 1993) were used as cloning
vectors. Plasmid pUFR127 (De Feyter and Gabriel 1991a) contains avrb6 on a 5-kb
DNA fragment. Plasmid pZit45 (Swamp et al. 1992) contains pthA on a 4.5-kb DNA
fragment. Plasmid pUFY20 contains a chimeric gene consisting of the 5’ and 3’ regions
of a\rb6 and internal repeated region of pthA (refer to Chapter 2). Plasmids pCS6 and
pCS23 contain intragenic recombinants of pthA (refer to Chapter 4). Triparental
matings were carried out to transfer plasmids from E. coli DH5a to Xanthomonas strains
by using pRK2013 or pRK2073 as helper plasmids (Swamp et al. 1991, De Feyter and
Gabriel 1991a). To transfer plasmids into X. campestris pv. malvaceamm strains, the
modifier plasmid pUFY054 carrying Xcml and Xcmlll methylase genes was used to
increase the transfer frequency (De Feyter and Gabriel 1991b).
Preparation of potential elicitors and plant inoculations. To detect elicitor
activity, cmde lysates were prepared as described (He et al. 1993) from Xanthomonas
and E. coli cells carrying pthA or its derivatives. Basically, bacterial cultures were
grown overnight, harvested by centrifugation and resuspended in lOmM Tris-HCl
(pH8.0) solution at an OD600 of 0.5-1. Bacterial cells were lysed by lysozyme
treatment (2mg/ml lysozyme in lOmM Tris-HCl,pH8.0) or dismpted by sonication in the
presence of 1 mM PMSF (phenylmethylsulfonyl fluoride). PthA protein was also

93
overexpressed in E. coli cells and purified by affinity chromatography (described later).
Cell-free suspensions (in lOmM Tris-HCl, pH8.0) or affinity column purified PthA were
immediately infiltrated into leaves of citrus and cotton plants.
Citrus (Citrus paradisi ‘Duncan’, grapefruit) and cotton (Gossypiwn hirsutwn)
plants were grown under greenhouse conditions. Cotton lines used were Acala-44 (Ac44)
and its congenie resistance lines Acb5b, Acb6, AcBIn3 as described (De Feyter et al.
1993; Swarup et al. 1992). X. campestris pv. malvacearum transconjugants carrying
pthA or its derivatives were inoculated into leaves of citrus plants in BL-3P level
containment (refer to Federal Register Vol.52, No. 154,1987) at the Division of Plant
Industry, Florida Department of Agriculture, Gainesville.
Preparation of antisera and immunoblotting. A 3.7kb BamRl fragment carrying
intact pthA was generated from partial digestion of pZit45 (Swarup et al. 1992), and
cloned into expression vector pET-19b (Novagen, Madison, WI). The fusion sites of the
resulting construct pUFY50.13 were determined by DNA sequence analysis. PthA protein
was overexpressed in E. coli strain BL21(DE3)/pLysS containing pUFY50.13 in the
presence of ImM IPTG (isopropyl-/3-D-thiogalactopyranoside). Since proteins expressed
on pET-19b are designed to contain 10 histidine residues at the amino terminus as an
affinity handle, overexpressed PthA was readily purified by His-tag affinity
chromatography (Novagen, Madison, WI). For immunization of rabbits, PthA was
further purified by SDS-polyacrylamide gel electrophoresis. Antiserum against PthA was
prepared from a New Zealand rabbit by Cocalico Biologicals, Inc. (Reamstown, PA).
For immunoblotting, E. coli or Xanthomonas cells were harvested by centrifugation and

94
resuspended in sample buffer (50 mM Tris-HCl, pH6.8, 10% glycerol, 2% SDS, 0.1%
bromophenol blue). After samples were boiled for 2 min and centrifuged for 5 min, they
were separated on 8% SDS-polyacrylamide gels and transferred to nitrocellulose by
electroblotting (Towbin etal. 1979). Western blots were probed with anti-PthA antibody
and visualized with goat anti-rabbit antibody conjugated with alkaline phosphatase.
DNA constructs for GUS fusion proteins. The DNA coding sequences for the C-
terminal region (191 amino acids) of PthA and Avrb6 were amplified by PCR from the
plasmid pZit45 (Swarup et al. 1992) by using the following primers:
1. CTCTAGAGCCATGACGCAGTTC; 2. CAGATCTCTGAGGCAATAGCTC.
The amplified fragments were first cloned into pGEM-T (Promega Inc. Madison,
WI), then cut out with XballBgtll and recloned into the Xbal/BamHl site of plant
expression vector pBI221 (Clonetech, CA). The resulting plasmids pUFY082 and
pUFY083 contain the chimeric gene with the C-terminal coding sequence of PthA and
Avrb6, respectively, translationally fused with 5’ end of GUS gene. Both pUFY082 and
pUFY083 were checked by sequence analysis at the fusion site.
Onion transformation system. The Helium Biolistic gene transformation system
(Du Pont) was used to transform epidermal cell layers of white onion as described
(Varagona et al. 1992). Inner epidermal layers were peeled and placed inside up on
Petri dishes containing MS (Murashige and Skoog 1962) basal media (4.4 g/L MS salt
mixture, Sigma, M5519, 30 g/L sucrose, pH 5.7) with 2.5 mg/L amphotericin B (Sigma)
and 6% agar. DNA samples were prepared as described (Taylor and Vasil 1991). Five
gLg of column-purified (Magic Prep, Promega) plasmid DNA was precipitated onto 1 mg

95
of 1.6 /xm gold particles using 50 /xL of 2.5 M CaCl2 and 10 ¿xL of 100 mM spermidine.
DNA-coated gold particles were washed with 50 /xl of 70% and 100% ethanol and finally
resuspended in 60 /xL of 100% ethanol. The samples were dispersed by submerging in
a waterbath sonicator for a few seconds. Five /xL samples were pipetted onto particle
delivery discs and delivered into onion epidermal cell layers at pressures of 1100 or 1300
p.s.i.. After microprojectile bombardment, Petri dishes were sealed with parafilm and
incubated overnight at 28°C in the dark. The assays for nuclear localization activity of
the fusion protein were repeated in three independent transformation tests.
Histochemical analysis. The histochemical GUS assay was used to determine the
location of GUS fusion proteins in onion cells as described previously (Jefferson, 1987,
Varagona et al 1992). Onion epidermal layers were incubated at room temperature in
X-glu solution (50 mM phosphate buffer, pH7.0, ImM EDTA, 0.001% Triton X-100.
0.05mM sodium ferricyanide and ferrocyanide, 2mM X-glu). After detection of blue
color, onion epidermal layers were mounted on a glass slide with a solution containing
20/xg/ml of the nucleus-specific dye 4’,6-diamidino-2-phenylindole (DAPI, 0.1 X
phosphate buffer, lOmM sodium azide, 90% glycerol). Cellular location of the blue
indigo dye produced by oxidative dimerization of the GUS product was determined under
bright-field optics using a Zeiss Axiophot microscope and compared with the location of
DAPI-stained nuclei under fluorescence optics.

96
Results
Citrus canker symptoms elicited by X. campestris pv. malvacearum carrying
pthA. Gene pthA is required for pathogenicity of X. citri on citrus and confers canker-
eliciting ability to many citrus-compatible xanthomonads (Swarup et al. 1991). In this
study, pthA and its derivatives were introduced into X. campestris pv. malvacearum, a
cotton pathogen that does not grow on citrus plants (Kingsley et al. 1993). When
inoculated on citrus, transconjugants of X. c. pv. malvacearum carrying pthA elicited
hyperplastic canker-like lesions on citrus (Fig. 5-1). The transconjugant
XcmH1407/pUFY20, which differs from XcmH only by replacing the repeated region
of avrb6 with that of pthA, also elicited canker-like lesions. However, pthA did not
enable X. c. pv. malvacearum to grow on nonhost citrus, and high initial inoculum levels
( > 107 bacteria/ml) were required to elicit visible canker-like symptoms.
Attempt to detect elicitor activity encoded by pthA. The diverse specificities
encoded by tandem repeats and the fact that a few amino acid changes in the tandem
repeats of pthA enabled a cotton pathogen to elicit canker symptoms on nonhost citrus
suggest that pthA and other members of the gene family may encode protein elicitors
directly interacting with plant receptors. In repeated attempts, however, cell-free lysates
and affinity column purified PthA protein from X. campestris pv. malvacearum and/or
E. coli cells carrying pthA or its derivatives failed to elicit canker or HR symptoms on
citrus and cotton plants.

97
Figure 5-1. Canker symptoms elicited by X. campestris pv. malvacearum
transconjugants on nonhost citrus. 1, XcmH/pUFR047; 2, XcmH1407/pUFR047;
3, XcmHM2.2S/pUFR047; 4, XcmH/pZit45; 5, XcmH1407/pUFY20; 6,
XcmHM2.2S/pZit45.

98
Constitutive expression of Pth A protein. Western blot analysis revealed that the
predicted 122 kD PthA protein was constitutively expressed in Xanthomonas cells
carrying pthA (Fig. 5-2). The antibody specifically recognized 3-4 proteins in protein
extracts of wild type X. citri strain 3213, which contains pthA plus three putative
homologues. As predicted by the size of these homologous gene, the largest protein was
PthA (indicated by an arrow in lane 1, Fig. 5-2). This was confirmed by Western
analysis of the marker exchange mutant Xcl.2 {pthAwnpt-sac). Gene pthA was disrupted
in Xc 1.2 by a nptl-sac marker inserted at the 10th tandem repeat (Chapter 4), which was
expected to generate a truncated protein of 68kD upon translation. As shown in lane 2,
Fig. 5-2, this mutant lost the PthA protein and instead expressed a protein close to the
size of predicted truncated protein (68 kD). The anti-PthA antibody did not react with
extracts from X. campestris pv. citrumelo 3048 which contains no genes homologous to
pthA. However, PthA protein was constitutively expressed in rich medium by the 3048
transconjugant carrying pthA (indicated by an arrow in lane 4, Fig. 5-2).
The C-terminal region of Avr/Pth proteins contain nuclear localization consensus
sequences. The proteins encoded by all sequenced members of the Xanthomonas avr/pth
such as avrbó (De Fey ter et al. 1993), pthA (chapter 4), a\rBs3 (Bonas et al. 1989,
1993) and avrXalO (Hopkins et al. 1992) are highly homologous and only vary in the
internal repetitive region which determines avirulence and pathogenic specificities.
Analysis of the amino acid sequences encoded by these genes revealed the presence of
three stretches of basic residues with complete homology with the nuclear localization

99
kD
105
69
12 3 4
Figure 5-2. Western blot analysis of protein extracts from Xanthomonas strains. 1,
Xanthomonas citri strain 3213; 2, marker exchange mutant Xcl.2 (pthA:\npt-sac)\ 3, X.
c. pv. citrumelo strain 3048; 4, 3048 transconjugant carrying pZit45 (pthA). Asterisk
indicate the band of PthA protein or its truncated derivative.

100
Protein
Nuclear Localization Sequences
Consensus sequence
KR/KXK/R
SV40-T (monopartite)
p K K K R K v
Nucleoplasmin (bipartite)
KRpaatKKaggaKKKKl
VirD2
KRpRedddgepseRKReRd
PthA and Avrb6
K R a K p-40 aa-R K R s R-31 aa-R v K R p R
Figure 5-3. Nuclear localization sequences from SV40 (T-antigen), Xenopus
(nucleoplasmin), Agrobacterium (VirD2) and Xanthomonas (PthA and Avrb6). Capital
letters denotes basic residues presumed important for nuclear targeting.

101
consensus sequences (K-R/K-X-R/K) found in many characterized nuclear localized
proteins (Chelsky et al. 1989) (Fig. 5-3). These three putative nuclear localization
sequences (NLSs) are located at the carboxyl-terminal part of the proteins, at positions
1020-1024 (K-R-A-R-P), 1065-1069 (R-K-R-S-R), and 1101-1106 (R-V-K-R-P-R) in
PthA. The first and third putative NLSs also contains proline residues which are often
found in NLSs. In addition, three putative casein kinase II sites were found in proximity
to these three nuclear localization consensus sequences. They are located at positions
994-997 (TELE), 1068-1071 (SRSD), 1120-1123 (TAAD) in PthA. This feature is
common to many nuclear localized proteins which often have casein kinase II sites (S/T-
X-X-D/E) at a distance of 10-30 amino acids up or downstream from the NLSs (Rihs et
al. 1991).
The C-terminus of PthA and Avrb6 proteins directed fl-glucuronidase into the
plant nucleus. The 68 kD p-glucuronidase (GUS) protein has been shown to localize in
the cytoplasm (Restrepo et al. 1990; van der Krol and Chua 1991) and is too large to
enter the nucleus without an NLS (Varagona et al. 1992). As shown in Fig. 5-4A,
when pBI221 (encoding GUS) was introduced into onion epidermal cells, GUS activity
was found in the cytoplasm and not in the nucleus. To determine whether or not the
putative NLSs are functional in plant cells, the DNA coding sequences for the C-terminal
region (191 amino acids) of PthA and Avrb6 were translationally fused with the 5’ end
of a GUS reporter gene. The resulting constructs were introduced into onion epidermal
cells by microprojectile bombardment and expressed transiently. Intracellular localization
of GUS fusion proteins (approximately 90-kD) containing the C-terminal region of PthA

Figure 5-4. Histochemical localization of GUS and GUS fusion proteins containing the
C-terminal region of PthA or Avrb6 in transformed onion epidermal tissues. Tissue were
simultaneously analyzed using X-glu staining (A, B, C) and nuclei-specific DAPI staining
(B’, C’). A, transformed with pBI221 encoding GUS protein; B and B’, transformed
with pUFY082 encoding GUS fusion protein containing the C-terminus of PthA; C and
C\ transformed with pUFY083 encoding GUS fusion protein containing the C-terminus
of Avrb6.

103

104
or Avrb6 was determined histochemically, with GUS activity localized specifically to
the nucleus of plant cells (Fig. 5-4 B and C). In repeated experiments, more than one
hundred nuclei were observed to be GUS positive. Rarely, some diffuse GUS activity
was observed within the cytoplasm. The location of nuclei was confirmed by staining
with the nucleus-specific dye 4’,6-diamidino-2-phenylindole (DAPI) and observed under
fluorescence microscope (Fig. 5-4 B’ and C’).
Discussion
The pathogenicity gene, pthA, is required for X. citri to elicit hyperplastic canker
symptoms on citrus and confers canker-eliciting ability to several xanthomonads that are
mildly pathogenic on citrus (Swarup et al. 1991). X. c. pv. malvacearum, a cotton
pathogen, is nonpathogenic on citrus and can not grow on citrus plants (Kingsley et al.
1993), but it has high homology (90%) with X. citri as shown by DNA reassociation
(Egel et al. 1991). In this study, introduction of pthA into X. campestris pv.
malvacearum strain XcmH1005, or the replacement of the tandem repeats of avrb6 in
XcmH1005 with those of pthA, conferred the ability to induce hyperplastic canker lesions
on citrus. The host-specific canker symptoms induced by X. campestris pv. malvacearum
transconjugants carrying pthA were independent of bacterial growth in planta since high
inoculum levels (> 107 cells/ml) were required to elicit visible symptoms. Therefore,
additional virulence genes are needed to extend the host range of X. campestris pv.
malvacearum to include citrus plants. Gene pthA belongs to a Xanthomonas avr/pth gene

105
family and differs only slightly from other members in the repeated region of genes
(Swarup et al. 1992; also refer to Chapter 4). It is striking that only a few amino acid
changes in the tandem repeats of a single avr gene enabled XcmH1005 to elicit canker
symptoms on citrus. Since the repetitive domains of prokaryotic and eukaryotic proteins
often function as ligand binding sites and determine specificity (Wren 1991; McConkey
et al. 1990), pthA may encode a proteinaceous elicitor capable of interacting with a plant
receptor and inducing host-specific symptoms.
Despite repeated attempts, elicitor activity was not detected in cell-free crude
extracts from X. c. pv. malvacearum and E. coli cells carrying pthA or its derivatives.
Affinity column purified PthA also failed to elicit pathogenic symptoms. The failure
could result from a variety of potential problems. Protein elicitors may be unstable and
may need post-translational processing. Protein elicitors may not be able to gain entry
into plant cell in the absence of living pathogens which secrete extracellular enzymes and
weaken plant cell wall.
Western blot analysis revealed that the predicted 122 kD PthA protein was
constitutively expressed in Xamhomonas cells carrying pthA. AvrBs3, a homolog of
PthA, is also constitutively expressed in X. campestris pv. vesicatoria grown on rich
media (Knoop et al. 1991). Despite this constitutive expression, hrp (hypersensitive
response and pathogenicity) gene function is absolutely required for Xamhomonas
carrying avrBs3 to elicit an HR on pepper (Knoop et al. 1991) and for Xamhomonas
carrying pthA to elicit cankers on citrus (unpublished data, Gabriel). Since hrp genes are
known to be involved in the export of proteinaceous HR elicitors (He et al 1993), and

106
since hip' mutations abolish activity of two members of the gene family, it seems likely
that the proteins encoded by these avr/pth genes are exported from the bacteria in their
interactions with plants (Fenslau 1993).
No direct evidence has been found for the export of proteins encoded by this gene
family. Cell fractionation indicates that the major portion of AvrBs3 protein is in the
bacterial cytoplasm and immunoelectron microscopy fails to detect AvrBs3 in the
bacterial outer membrane or evidence of secretion (Knoop et al. 1991; Brown et al.
1993). The failure to detect protein secretion could have resulted from protein
instability or very low levels of the antigen in the extracellular environment. For
example, VirE2 of Agrobacterium tumafaciens was localized in the bacterial cytoplasm
by cell fractionation and by immunoelectron microscopy (Christie et al. 1988;
Thorstenson and Zambryski 1994), however, it is known to be exported together with
T-DNA and enters the nucleus of plant cells (Citovsky et al. 1989, 1992)
The translocation of the Agrobacterium T-DNA complex into the nucleus of plant
cells can be viewed as an unique signaling pathway (Citovsky and Zambryski 1993). The
T-DNA is coated by the single strand binding protein VirE2 and piloted by protein VirD2
(Christie et al. 1988; Ward and Barnes 1988). Both VirE2 and VirD2 contain nuclear
localization sequences (NLSs) which could facilitate the translocation of T-DNA complex
into the nucleus (Citovsky et al. 1992; Howard et al. 1992; Shurvinton et al. 1992;
Tinland et al. 1992). Nuclear localization sequences are short stretches of basic amino
acid residues conserved in many nuclear localized proteins. VirD2 and VirE2 are the
only bacterial proteins that have been reported to contain nuclear localization sequences.

107
Analyses of the predicted amino acid sequences of PthA and its homologs encoded by
members of the Xanthomonas avr/pth gene family revealed three putative nuclear
localization sequences located in the C-terminal region of these proteins. By transiently
expressing GUS fusion proteins in onion epidermal cells, the C-terminal region of PthA
or Avrb6 was shown to direct the GUS fusion protein into the nucleus of plant cells.
Therefore, functional nuclear localization sequences were identified in the proteins
encoded by members of the Xanthomonas avr/pth gene family.
We have examined amino acid sequences encoded by some other avr genes, and
at least one of them, AVR4 of Cladosporium fulvum (Joosten et al. 1994), also contains
a putative bipartite nuclear localization sequence (KKWCDYPNLSTCPVKTPGPKPKK)
in the C-terminal region. AVR4 and AVR9 of Cladosporium fulvum are the only known
race-specific protein encoded by avr genes that are capable of directly eliciting an HR
on plants. It will be interesting to see if this putative NLS are functional in plant cells.
Changes in the cell division and differentiation status of eukaryotic cells may be
accompanied by the selective relocalization of proteins between cytoplasm and nucleus
(Nigg et al. 1991). This holds for fos and rel encoded oncoproteins, certain
transcriptional factors, steroid hormone receptors, the Drosophila morphogen Dorsal and
the Agrobacterium T-complex. For example, the activity of the Rel/NFicB/Dorsal family
of proteins is controlled by sequestration in the cytoplasm in association with inhibitory
proteins called IkB (Kerr et al. 1992). In unstimulated cells, transcription factor NFkB
is associated with IkB and retained in the cytoplasm. However, external signals activate
protein kinases which phosphorylate IkB, resulting in the dissociation of the NFkB-IkB

108
complex. Subsequent translocation of NFkB into the nucleus triggers gene expression.
In mammalian systems, NFkB is activated by oxygen radicals induced by diverse signals
through distinct intracellular pathways (Schreck et al. 1991). In plants, active oxygen
species are known to be involved in hypersensitive response and activation of plant
defense genes (Bradley et al. 1992; Chen et al. 1993; Legendre et al. 1992). However,
plant homologues of NFkB and IkB are not known. The inhibitory IkB proteins contain
5 to 8 ankyrin repeats. The ankyrin repeat motif is a 33 amino acid unit that is found
in a number of transcription factors, receptors and ion transporters and is important in
protein-protein interactions (Blank et al. 1992; Sentenac et al. 1992). Interestingly, two
short stretches of conserved amino acids (G-X-T-P and V-X-X-L-L) found in the ankyrin
repeat also exist in the 34 amino acid repeats of PthA and its homologs. However, it is
not known whether this is just coincidental or has functional meaning.
Based on the known data and previous models (Gabriel and Rolfe 1990), a
testable hypothesis was proposed to explain the possible mechanism by which specific
signals encoded by the Xanthomonas avr/pth gene are perceived and transduced (Fig. 5-
5). After entering intercellular space of plant tissues, the pathogen releases the Avr/Pth
proteins by an /i/p-dependent export pathway activated by plant signals (Fenselau et al.
1992). At the same time, the pathogen also secretes extracellular enzymes to weaken
the plant cell wall. Avr/Pth proteins enter plant cell through a process similar to the
receptor-mediated endocytosis of some fungal elicitors (Horn et al. 1989). Once
internalized, Avr/Pth proteins interact specifically with R gene products to form the
dimers or multimers which are then transported into the nucleus of plant cells.

109
Bacterial cell
(released by secretion or cell lysis)
Specific recognition
Gene expression
Nucleus
Extracellular enzymes
weakening plant cell wall
Avr/Pth proteins
(endocytosis)
Hyperplasia
(cell division)
Watersoaking & necrosis
(delayed cell death)
Hypersensitive response
(rapid cell death)
Figure 5-5. A working hypothesis for the transduction of Xanthomonas Avr/Pth signals
during the plant-pathogen interaction.

110
Alternatively, Avr/Pth proteins are imported alone into the nucleus. The nuclear
transport process can be activated or inhibited by many factors such as phosphorylation
or dephosphorylation at specific casein kinase II sites of proteins (Rihs et al. 1989,
1991). After entering the nucleus, Avr/Pth proteins may act on nuclear transcriptional
factors encoded by R genes or even directly on the R gene themselves. Mediated by the
34 amino acid tandem repeats, the avirulence and pathogenic specificity can be
determined during the protein-protein or protein-DNA interaction. Activation of specific
targeted genes may lead to different physiological outcomes such as hyperplasia (cell
division) on citrus, hypersensitive response (rapid cell death) on cotton or citrus,
watersoaking and subsequent necrosis on cotton (delayed cell death). This may be
similar to oncogenesis (cell division) and apoptosis (cell death) of mammalian cells that
can be altered by the ratio and interactions among transcription factors in specific cells
(Martin et al. 1994).
This hypothesis may apply to other model systems of specific recognition such as
the Cladosporium Julvum AVR4 protein containing potential NLSs. However, diverse
signal transduction mechanisms may have been exploited by hosts and pathogens to
determine the specific recognitions involving plant disease resistance and susceptibility.
Although gene-for-gene interactions can appear to be similar genetically, they may be
quite different in terms of biochemical mechanisms.

CHAPTER 6
SUMMARY AND CONCLUSIONS
Members of a Xanthomonas avirulence (avr) gene family were shown to play an
important role in determining host-specific pathogenic symptoms of cotton blight and
citrus canker. This host-specific pathogenicity (pth) function may be critical in
conditioning the host range of Xanthomonas. All cotton blight strains examined to date
of X. campestris pv. malvacearum carry 4 to 12 DNA fragments that hybridize with
members of this gene family. Twelve hybridizing fragments have now been isolated
from a single X. campestris pv. malvacearum strain XcmH1005. Six of them (avrB4,
avrb6, avrb7, avrBln, avrBlOl and avrBlOT) are clustered on a 90-kb indigeneous
plasmid and were previously isolated as functional avr genes (De Feyter and Gabriel
1991a). The other six appear to be chromosomal fragments and were isolated in this
study. One of these encodes the previously described avrBn (Gabriel et al. 1986), and
three of these encode previously undescribed avr genes (avrB5, avrB103, and avrB104).
To investigate possible pleiotropic functions of these avr genes in XcmH1005,
marker exchange mutagenesis and complementation analyses were carried out. A single
mutation of avrb6, but not other members in XcmH1005, resulted in significant reduction
of the pathogen’s ability to induce watersoaking symptoms on cotton susceptible line
Acala-44. An average of 240 times more bacteria were released onto the cotton leaf
111

112
surface from watersoaked spots caused by XcmH1005 than from those caused by an
isogenic avrb6~ mutant. Another XcmH1005 mutant, in which seven avr genes were
inactivated by repeated cycles of mutagenesis, was almost asymptomatic on cotton and
had 1600 times less bacteria released onto the leaf surface in comparison with that of
XcmH1005. However, in planta growth rates and yields of these mutants were identical
to that of the parental XcmH1005 strain. Therefore, inactivation of multiple avr genes
in X. campestris pv. malvacearum basically created a nonpathogenic endophyte of cotton.
The pleiotropic pathogenicity function of these avr genes was further confirmed by
complementation tests. Introduction of a DNA fragment carrying only avrbó into the
avrb6~ mutant fully complemented the phenotypic defects for both avirulence and
pathogenicity. Introduction of DNA fragments carrying avrbó plus two other avr genes
into the latter mutant fully restored its pathogenicity, indicating an additive pth function
of these avr genes. The role of these avr genes in eliciting host-specific symptoms and
aiding pathogen release and dispersal strongly indicates their fitness value in cotton blight
pathogen. The pleiotropic pathogenicity function and fitness value may partly explain
why plant pathogens carry avr genes which usually act as negative factors to limit
virulence on plants carrying specific R genes. However, the presence of many other avr
genes in plant pathogens may simply results from mutations neutral to the selection.
In citrus canker disease, the host-specific canker symptoms are known to be
determined by pthA of X. citri, another member of what is now known as the avr/pth
gene family. Gene pthA also confers host specific pathogenicty and avirulence function
to many xanthomonads on their respective hosts. DNA sequence comparisons revealed

113
that pthA is 98% identical to awBs3 and avrBs3-2 of X. campestris pv. vesicatoria, 97%
identical to avrb6 of X. campestris pv. malvacearum and 95% identical to avrXalO of X.
oryzae (Hopkins et al. 1992) along its entire length. The primary sequence differences
among these genes lie in the central region of the genes, characterized by nearly
identical, 102 bp tandem repeats. By constructing chimeric genes among pthA, avrB4,
avrbó, avrb7, avrBln, avrBlOl and avrB102, the 102 bp tandem repeats were found to
determine the gene-for-gene specificity of the avirulence reactions on cotton. In
addition, the tandem repeats of avrbó and pthA determined their specificity in enhancing
watersoaking on cotton and causing cankers on citrus, respectively. Replacement of
tandem repeats of avrbó in XcmH1005 with those of pthA conferred to the cotton
pathogen the ability to induce hyperplastic canker lesions on nonhost citrus at high
inoculum level.
By inserting a npt-sac cartridge into the tandemly repeated region of pthA as a
selective marker, intragenic recombination among homologous repeats was observed
when the gene was present in both Xanthomonas and Escherichia coli strains. Intragenic
recombination within pthA created new genes with novel host specificities and altered
pathogenicity and/or avirulence phenotypes. Many pthA recombinants gained or lost
avirulence function in many different combinations in pathogenicity assays on bean,
citrus and cotton cultivars. Intragenic recombination therefore provides a genetic
mechanism for the evolution of this avr/pth gene family.
It is striking that only a few amino acid changes in the tandem repeats of a single
gene altered avirulence and pathogenic specificities. To account for the precise

114
recognition and diverse specificities, it is likely that these avr/pth genes encode protein
elicitors capable of interacting with plant receptors and inducing host-specific symptoms.
Unfortunately, elicitor activities encoded by pthA or avrbó were not detected despite
repeated attempts. By analysis of the predicted amino acid sequences encoded by pthA,
avrbó, and their homologues, however, three nuclear localization consensus sequences
were found in the C-terminal region of these proteins. The predicted PthA and Avrbó
proteins were constitutively expressed in Xanthomonas as detected by Western analysis
using anti-PthA antibody. By creating fusions of the C-terminus coding region of pthA
and avrbó to a gus reporter gene and transiently expressing these fusion genes in onion
epidermal cells, the C-terminal region of PthA or Avrbó was observed to direct the GUS
fusion protein into the nuclei. A working hypothesis is proposed that Avr/Pth proteins
are secreted, taken up by the plant cell, enter the nucleus and directly affect gene
expression, leading to the induction of host-specific symptoms.

LITERATURE CITED
Albertini, A.M., Hofer, M., Calos, M.P. and Miller, J.H. 1982. On the formation of
spontaneous deletions: the importance of short sequence homologies in the generation of
large deletions. Cell 29:319-328.
Anderson, B.E., McDonald, G.A., Jones, D.C. and Regnery, R.L. 1990. A protective
protein antigen of Rickettsia rickettsii has tandemly repeated, near-identical sequences.
Infect. Immun. 58:2760-2769.
Arlat, M., and Boucher, C. A. 1991. Identification of a dsp DNA region controlling
aggressiveness of Pseudomonas solanacearum. Mol. Plant-Microbe Interact. 4:211-213.
Blank, V., Kourilsky, P., and Israel, A. 1992. NF-kappa B and related proteins:
Rel/dorsal homologies meet ankyrin-like repeats. Trends Biochem. Sci. 17:135-140.
Bonas, U., Conrads Strauch, J. and Balbo, I. 1993. Resistance in tomato to Xanthomonas
campestris pv. vesicatoria is determined by alleles of the pepper-specific avirulence gene
avrBs3. Mol. Gen. Genet. 238:261-269.
Bonas, U., Schulte, R., Fenselau, S., Minsavage, G.V., Staskawicz, B.J. and Stall, R.E.
1991. Isolation of a gene cluster from Xanthomonas campestris pv. vesicatoria that
determines pathogenicity and the hypersensitive response on pepper and tomato. Molec.
Plant-Microbe Interact. 4:81-88.
Bonas, U., Stall, R. E., and Staskawicz, B. 1989. Genetic and structural characterization
of the avirulence gene, avrBs3 from Xanthomonas campestris pv. vesicatoria. Mol. Gen.
Genet. 218:127-136.
Boucher, C.A., Gough, C.L., and Arlat, M. 1992. Molecular genetics of pathogenicity
determinants of Pseudomonas solanacearum with special emphasis on hrp genes. Annu.
Rev. Phytopathol. 30:443-461.
Boyer, H. W., and Roulland-Dussoix, D. 1969. A complementation analysis of the
restriction and modification of DNA in Escherichia coli. J. Mol. Biol. 41:459-465.
115

116
Bradley, D.J., Kjellbom, and Lamb, C.J. 1992. Elicitor- and wound-induced oxidative
cross-linking of a proline-rich plant cell wall protein: a novel, rapid defense response.
Cell 70: 21-30.
Brown, I., Mansfield, J., Irlam, I., Conrads-Strauch, L, and Bonas, U. 1993.
Ultrastructure of interactions between Xaruhomonas campestris pv. vesicatoria and
pepper, including immunocytochemical localization of extracellular polysaccharides and
the AvrBs3 protein. Mol. Plant-Microbe Interact. 6:376-386.
Canteros, B., Minsavage, G., Bonas, U., Pring, D. and Stall, R. 1991. A gene from
Xanthomonas campestris pv. vesicatoria that determines avirulence in tomato is related
to avrBs3. Molec. Plant-Microbe Interact. 4:628-632.
Chelsky, D., Ralph, R., and Jonak, G. 1989. Sequence requirements for synthetic
peptide-mediated translocation to the nucleus. Mol. Cell. Biol. 9: 2487-2492.
Chen, Z., Silva, H., and Klessig, D.F. 1993. Active oxygen species in the induction of
plant systemic acquired resistance by salicylic acid. Science 262: 1883-1886.
Christie, P.J., Ward, J.E., Winans, S.C., Nester, E.W. 1988. The Agrobacteriwn
tumafaciens virE2 gene product is a single-stranded-DNA-binding protein that associates
with T-DNA. J. Bacteriol. 170:2659-2667.
Citovsky, V., Wong, M.L., and Zambryski, P. 1991. Cooperative interaction of
Agrobacterium VirE2 protein with single stranded DNA: implications for the T-DNA
transfer process. Proc. Natl. Acad. Sci. USA 86:1193-1197.
Citovsky, V., and Zambryski, P. 1993. Transport of nucleic acids through membrane
channels: snaking through small holes. Annu. Rev. Microbiol. 47:167-197.
Citovsky, V., Zupan, J., Wamick, D. and Zambryski, P. 1992. Nuclear localization of
Agrobacterium VirE2 protein in plant cells. Science. 256:1802-1805.
Collmer, A., and Keen, N. T. 1986. The role of pectic enzymes in plant pathogenesis.
Ann. Rev. Phytopathol. 24:383-409.
Coplin, D.L., Frederick, R.D., Majerczak, D.R. and Tuttle, L.D. 1992.
Characterization of a gene cluster that specifies pathogenicity in Erwinia stewartii. Mol.
Plant-Microbe Interact. 5:81-88.
Crews, C.M. and Erikson, R.L. 1993. Extracellular signals and reversible protein
phosphorylation: what to Mek of it all. Cell. 74:215-217.

117
Culotta, E., and Koshland, D.E. 1993. p53 sweeps through cancer research. Science
262: 1958-1961.
Daniels, M.J., Dow, J.M., Osbourn, A.E. 1988. Molecular genetics of pathogenicity in
phytopathogenic bacteria. Annu. Rev. Phytopathol. 26: 285-312.
Debelle, F., Maillet, F., Vasse, J., Rosenberg, C., de Billy, F., Truchet, G., Denarie,
J. and Ausubel, F.M. 1988. Interference between Rhizobium meliloti and Rhizobium
trifolii nodulation genes: genetic basis of R. meliloti dominance. J. Bacteriol.
170:5718-5727.
De Feyter, R., and Gabriel, D. W. 1991a. At least six avirulence genes are clustered on
a 90-kilobase plasmid in Xanthomonas campestris pv. malvacearum. Mol. Plant-Microbe —
Interact. 4:423-432.
De Feyter, R., and Gabriel, D.W. 1991b. Use of cloned DNA methylase genes to
increase the transfer frequency of foreign genes into Xanthomonas campestris pv.
malvacearum. J. Bacteriol. 173:6421-6427.
De Feyter, R., Kado, C.I., and Gabriel, D.W. 1990. Small, stable shuttle vectors for
use in Xanthomonas. Gene 88:65-72.
De Feyter, R., Yang, Y., and Gabriel, D.W. 1993. Gene-for-genes interactions between
cotton R genes and Xanthomonas campestris pv. malvacearum avr genes. Mol.
Plant-Microbe Interact. 6:225-237.
Denarie, J., Debelle, F., and Rosenberg, C. 1992. Signalling and host range variation
in nodulation. Ann. Rev. Microbiol. 46:497-531.
Dickinson, M. J., Jones, D. A., and Jones, J. D. G. 1993. Close linkage between the
Cf-2/Cf-5 and Mi resistance loci in tomato. Mol. Plant-Microbe Interact. 6: 341-347.
Dixon, R.A., and Harrison, M.J. 1990. Activation, structure, and organization of genes
involved in microbial defense in plants. Adv. Genet. 28:165-232.
Dixon, R.A., and Lamb, C.J. 1990. Molecular communication in interactions between
plants and microbial pathogens. Annu. Rev. Plant Physiol. Plant Mol. Biol. 41:339-67.
Djordjevic, M.A., Gabriel, D.W. and Rolfe, B.G. 1987. Rhizobium the refined parasite
of legumes. Ann. Rev. Phytopath. 25:145-168.
Dove, C.H., Wang, S.-Z., Price, S.B. Phelps, C.J., Lyerly, D.M., Wilkins, T.D., and
Johnson, J.L. 1990. Molecular characterization of the Clostridium difficile Toxin A gene.
Infect. Immun. 58:480-488.

118
Duerr, B., Gawienowski, M., Ropp, T., and Jacobs, T. 1993. MsERKl: a mitogen-
activated protein kinase from a flowering plant. Plant Cell 5: 87-96.
Egel, D.S. Graham, J.H., and Stall, R.E. 1991. Genomic relatedness of Xanthomonas
campestris strains causing disease of citrus. Appl. Eviron. Microbiol. 57: 2724-2730.
Ellingboe, A.H. 1976. Genetics of host-parasite interactions. Pages 761-778 in:
Encyclopedia of plant pathology, new series, Vol 4: Physiological Plant Pathology. R.
Heitefuss and P.H. Williams, eds. Springer-Verlag, Heidelberg,
Ellingboe, A.H. 1982. Genetical aspects of active defense. In "Active Defense
Mechanisms of Plants", Proc. NATO Conf., Cape Sounion, Greece, ed. R.K.S. Wood,
pp. 179-192. New York, Plenum.
Faucher, C., Camut, S., Denarie, J. and Truchet, G. 1989. The nodH and nodQ host
range genes of Rhizobium meliloti behave as avirulence genes in R. leguminosarum bv.
viciae and determine changes in the production of plant-specific extracellular signals.
Molec. Plant-Microbe Interact. 2:291-300.
Felix, G., Grosskopf, D.G., Regenass, M., and Boiler, T. 1991. Rapid changes of
protein phosphorylation are involved in transduction of the elicitor signal in plant cells.
Proc. Natl. Acad. Sci. USA 88: 8831-8834.
Fenselau, S., Balbo, I., and Bonas, U. 1992. Determinants of pathogenicity in
Xanthomonas campestris pv. vesicatoria are related to proteins involved in secretion in
bacterial pathogens of animal. Mol. Plant-Microbe Interact. 5: 390-396.
Figurski, D. H., and Helinski, D. R. 1979. Replication of an origin-containing
derivatives of plasmid RK2 dependent on a plasmid function provided in trans. Proc.
Natl. Acad. Sci. USA 76:1648-1652.
Fisher, R. F., and Long, S. R. 1992. Rhizobium-plant signal exchange. Nature 357: 655-
660.
Flor, H.H. 1942. Inheritance of pathogenicity in Melampsora lini. Phytopathol. 32:653-
669.
Flor, H.H. 1946. Genetics of pathogenicity in Melampsora lini. J. Agrie. Res. 73: 335-
357.
Flor, H.H. 1947. Inheritance of reaction to rust in flax. J. Agrie. Res. 74:241-262.

119
Frankel, G., Newton, S. M., Schoolnik, G. K., and Stocker, B. A. 1989. Intragenic
recombination in a flagellin gene: characterization of the HI-j gene of Salmonella typhi.
EMBO J. 8:3149-3152.
Freeman, S., and Rodriguez, R.J. 1993. Genetic conversion of a fungal plant pathogen
to a nonpathogenic, endophytic mutualist. Science 260: 75-78.
Gabriel, D.W. 1986. Specificity and gene function in plant/pathogen interactions. ASM
News 52:19-25.
Gabriel, D.W. 1989. The genetics of plant pathogen population structure and
host-parasite specificity. Pages 343-379 in: Plant-Microbe Interactions: Molecular and
Genetic Perspectives, Vol 3. T. Kosuge and E.W. Nester, eds. Macmillan Publishing
Co., New York. .
Gabriel, D.W., Burges, A., and Lazo, G.R. 1986. Gene-for-gene recognition of five
cloned avirulence genes from Xanthomonas campestris pv. malvacearum by specific
resistance genes in cotton. Proc. Natl. Acad. Sci. USA. 83:6415-6419.
Gabriel, D.W. and De Feyter, R. 1992. RFLP analyses and gene tagging for bacterial
identification and taxonomy. Pages 51-66 in: Molecular Plant Pathology, Vol. I, A
Practical Approach. S.J. Gurr, M.J. McPherson and D.J. Bowles, eds. IRL Press,
Oxford.
Gabriel, D.W., Kingsley, M. T., Hunter, J.E., and Gottwald, T. R. 1989. Reinstatement
of Xanthomonas citri (ex Hasse) and X. phaseoli (ex Smith) and reclassification of all X.
campestris pv. citri strains. Int. J. Syst. Bacterid. 39:14-22.
Gabriel, D.W., Kingsley, M.T., Yang, Y., Chen, J. and Roberts, P. 1993. Host-specific
virulence genes of Xanthomonas. Pages 141-158 in: Molecular mechanisms of bacterial
virulence. C.I. Kado and J. Crosa, eds. Kluwer Academic Publishers, Dordrecht. ^—
Gabriel, D.W., Loschke, D.C. and Rolfe, B.G. 1988. Gene-for-gene recognition: the ion
channel defense model. Pages 3-14 in: Molecular Plant-Microbe Interactions,
Proceedings of the 4th International Symposium. D.P. Verma and R. Palacios, eds. APS
Press, St. Paul.
Gabriel, D.W. and Rolfe, B.G. 1990. Working models of specific recognition in
plant-microbe interactions. Annu. Rev. Phytopathol. 28:365-391.
Gaillard, J.-L. Berche, P., Frehel, C., Gouin, E., and Cossart, P. 1991. Entry of L.
monocytogenes into cells is mediated by intemalin, a repeat protein reminiscent of
surface antigens from gram-posistive cocci. Cell 65:1127-1141.

120
Gilmore Jr., R. D. 1993. Comparison of the rompA gene repeat regions of Rickettsiae
reveals species-specific arrangements of individual repeating units. Gene 125:97-102.
Goebl, M. and Yanagida, M. 1991. The TPR snap helix: a novel protein repeat motif
from mitosis to transcription. Trends Biochem. Sci. 16:173-177.
Goring, D.R., Rothstein, S.J. 1992. The S-locus receptor kinase gene in a self¬
incompatible Brassica napus line encodes a functional serine/threonine kinase. Plant Cell
4: 1273-1281.
Gough, C.L., Genin, S., Lopes, V. and Boucher, C.A. 1993. Homology between the
HrpO protein of Pseudomonas solanacearum and bacterial proteins implicated in a signal
peptide-independent secretion mechanism. Mol. Gen. Genet. 239:378-392.
Gough, C.L., Genin, S., Zischek, C., and Boucher, C.A. 1992. hrp genes of
Pseudomonas solanacearum are homologous to pathogenicity determinants of animal
pathogenic bacteria and are conserved among plant pathogenic bacteria. Mol. Plant-
Microbe Interact. 5:384-389.
Hagblom, P., Segal, E., Billyard, E., So, M. 1985. Intragenic recombination leads to
pilus antigenic variation in Neisseria gonorrhoeae. Nature 315:156-158.
Hall, A. 1993. Ras-related proteins. Current Opinion Cell Biol. 5: 265-268.
He, S.Y., Huang, H.C., and Collmer, A. 1993. Pseudomonas syringae pv. syringae
harpiiipSS: A protein that is secreted via the Hrp pathway and elicits the hypersensitive
response in plants. Cell 73: 1255-1266.
Heath, M. C. 1991. The role of gene-for-gene interactions in the determination of host
species specificity. Phytopathol. 81:127-130.
Herbers, K., Conrads-Strauch, J. and Bonas, U. 1992. Race-specificity of plant
resistance to bacterial spot disease determined by repetitive motifs in a bacterial
avirulence protein. Nature 356:172-174.
Herrera Estrella, A., Van Montagu, M. and Wang, K. 1990. A bacterial peptide acting
as a plant nuclear targeting signal: the amino-terminal portion of Agrobacterium VirD2
protein directs a beta-galactosidase fusion protein into tobacco nuclei. Proc. Natl. Acad.
Sci. USA 87:9534-9537.
Hoft, D.F., Kim, K.S., Otsu K., Moser, D.R., Yost, W.J., Blumin, J.H., Donelson,
J.E. and Kirchhoff, L.V. 1989. Trypanosoma cruzi expresses diverse repetitive protein
antigens. Infect. Immun. 57:1959-1967.

121
Hollingshead, S.K., Fischetti, V.A., and Scott. J.R. 1987. Size variation in group A
streptococcal M protein is generated by homologous recombination between intragenic
repeats. Mol. Gen. Genet. 207:196-203.
Hopkins, C.M., White, F.F., Choi, S.-H., Guo, A., and Leach, J.E. 1992. Identification
of a family of avirulence genes from Xanthomonas oryzae pv. oryzae. Mol.
Plant-Microbe Interact. 5:451-459.
Horn, M.A., Heinstein, P.F., Low, P.S. 1989. Receptor-mediated endocytosis in plant
cells. Plant Cell 1: 1003-1009.
Howard, E.A., Zupan, J.R., Citovsky, V. and Zambryski, P.C. 1992. The VirD2
protein of A. tumefaciens contains a C-terminal bipartite nuclear localization signal:
implications for nuclear uptake of DNA in plant cells. Cell. 68:109-118.
Hrabak, E.M., and Willis, D.K. 1992. The lemA gene required for pathogenicity of
Pseudomonas syringae pv. syringae on bean is a member of a family of two-component
regulator. J. Bacteriol. 174: 3011-3020.
Hunter, T. 1993. Cytokine connections. Nature 366: 114-116.
Ishida, S., Takahashi, Y., and Nagata, T. 1993. Isolation of cDNA of an auxin-regulated
gene encoding a G protein (3 subunit-like protein from tobacco BY-2 cells. Proc. Natl.
Acad. Sci. USA 90:11152-11156.
Jefferson, R. A. 1987. Assaying chimeric genes in plants: the GUS gene fusion system.
Plant Mol. Biol. Rep. 5: 387-405.
Joosten, M.H.A.J, Cozijnsen, T.J., and De Wit,P.J.G.M. 1994. Host resistance to a
fungal tomato pathogen lost by a single base-pair change in an avirulence gene. Nature
367:384-386.
Kamada, L, Yamauchi, S., Youssefian, S., and Sano, H. 1992. Transgenic tobacco
plants expressing rgpl, a gene encoding a ray-related GTP-binding protein from rice,
show distinct morphological characteristics. Plant J. 2:799-807.
Kearney, B., Ronald, P.C., Dahlbeck, D. and Staskawicz, B.J. 1988. Molecular basis
for evasion of plant host defence in bacterial spot disease of pepper. Nature 332:541-543.
Kearney, B., and Staskawicz, B.J. 1990b. Widespread distribution and fitness
contribution of Xanthomonas campestris avirulence gene avrBs2. Nature 346:385-386.
Keen, N.T. 1990. Gene-for-gene complementarity in plant-pathogen interactions. Ann.
Rev. Genet. 24:447-463.

122
Keen, N.T., and Buzzel, R.I. 1991. New disease resistance genes in soybean against
Pseudomonas syringae pv. glycinea: Evidence that one of them interacts with a bacterial
elicitor. Theor. Appl. Genet. 81: 133-138.
Keen, N.T., and Staskawicz, B.J. 1988. Host range determinants in plant pathogens and
symbionts. Annu. Rev. Microbiol. 42:421-440.
Keen, N.T., Tamaki, S., Kobayashi, D., Gerhold, D., Stayton, M., Shen, H., Gold, S.,
Lorang, J., Thordal-christensen, H. Dahlbeck, D. and Staskawicz, B. 1990. Bacterial
expressing avirulence gene D produce a specific elicitor of the soybean hypersensitive
reation. Mol. Plant-Microbe Interact. 3: 112-121.
Kerr, L.D., Inoue, J., and Verma I.M. 1992. Signal transduction: the nuclear target.
Current Opinion Cell Biol. 4: 496-501.
Kieber, J.J., Rothenberg, M., Roman, G., Feldmann, K.A. and Ecker, J.R. 1993.
CTR1, a negative regulator of the ethylene response pathway in Arabidopsis, encodes a
member of the raf family of protein kinases. Cell. 72:427-441.
Kingsley, M.T., Gabriel, D.W., Marlow, G.C. and Roberts, P.D. 1993. The opsXlocus
of Xanthomonas campestris affects host range and biosynthesis of lipopolysaccharide and
extracellular polysaccharide. J. Bacteriol. 175:5839-5850.
Kishimoto, T., Taga, T., and Akira, S. 1994. Cytokine signal transduction. Cell 76:
253-262.
Klement, Z. 1982. Hypersensitivity. Pages 149-177 in: Phytopathogenic Prokaryotes, Vol
II. M.S. Mount and G.H. Lacy, eds. Academic Press, New York.
Knoop, V., Staskawicz, B. and Bonas, U. 1991. Expression of the avirulence gene
avrBs3 from Xanthomonas campestris pv. vesicatoria is not under the control of hrp
genes and is independent of plant factors. J. Bacteriol. 173:7142-7150.
Kobayashi, D.Y., Tamaki, S.J., and Keen, N.T. 1989. Cloned avirulence genes from
the tomato pathogen Pseudomonas syringae pv. tomato confer cultivar specificity on
soybean. Proc. Natl. Acad. Sci. USA 86:157-61.
Kobayashi, D.Y., Tamaki, S.J., Trollinger, D.J., Gold, S. and Keen, N.T. 1990. A
gene from Pseudomonas syringae pv. glycinea with homology to avirulence gene D from
P. s. pv. tomato but devoid of the avirulence phenotype. Mol. Plant Microbe. Interact.
3:103-111.
Lamb, C. J. 1994. Plant disease resistance genes in signal perception and transduction.
Cell 76: 419-422.

123
Lamb, C. J., Lawton, M. A., Dron, M., and Dixon, R.A. 1989. Signals and
Transduction mechanisms for activation of plant defenses against microbial attack. Cell
56: 215-224.
Lazo, G.R., and Gabriel, D.W. 1987. Conservation of plasmid DNA sequences and
pathovar identification of strains of Xanthomonas campestris. Phytopathology 77:448-
453.
Legendre, L., Heinstein, P.F., and Low, P.S. 1992. Evidence for participation of GTP-
binding proteins in elicitaiton of the rapid oxidative burst in cultured soybean cells. J.
Biol. Chem. 267:20140-20147.
Leigh, J.A. and Coplin, D.L. 1992. Exopolysaccharides in plant-bacterial interactions.
Annu. Rev. Microbiol. 46:307-346.
Leong, S. A., Ditta, G. S., and Helinski, D. R. 1982. Heme biosynthesis in Rhizobium.
Identification of a cloned gene coding for aminolevulinic acid synthetase from Rhizobium
meliloti. J. Biol. Chem. 257:8724-8730.
Lewis-Henderson, W.R. and Djordjevic, M.A. 1991. A cultivar-specific interaction
between Rhizobium leguminosarum bv. trifolii and subterranean clover is controlled by
nodM, other bacterial cultivar specificity genes, and a single recessive host gene. J.
Bacteriol. 173:2791-2799.
Leyns, F., De Cleene, M., Swings, J., and De Ley, J. 1984. The host range of the
genus Xanthomonas. Bot. Rev. 50:308-355.
Ma, Q.S., Chang, M.F., Tang, J.L., Feng, J.X., Fan, M.J., Han, B., and Liu, T.
1988. Identification of DNA sequences involved in host specificity in the pathogenesis
of Pseudomonas solanacearum strain T2005. Mol. Plant-Microbe Interact. 4:169-174.
Martin, G. B., Brommonschenkel, S. H. Chunwongse, J. Frary, A. Ganal, M. W.
Spivey, R., Wu, T., Earle, E. D., and Tanksley, S. D. 1993. Map-based cloning of a
protein kinase gene conferring disease resistance in tomato. Science 262:1432-1436.
Martin, S.J., Green, D.R., and Cotter, T.G. 1994. Dicing with death: dissecting the
components of the apoptosis machinery. Trends Biochem. Sci. 19: 26-20.
McConkey, G.A., Waters, A.P., and McCutchan T.F. 1990. The generation of genetic
diversity in malaria parasites. Annu. Rev. Microbiol. 44:479-498.

124
Midland, S.L., Keen, N.T., Sims, J.J., Midland, M.M., Stayton, M.M., Burton, V.,
Smith, M.J., Mazzola, E.P., Graham, K.J. and Clardy, J. 1993. The structures of
syringolides 1 and 2, novel C-glycosidic elicitors from Pseudomonas syringae pv.
tomato. J. Org. Chem. 58:2940-2945.
Mitchell, R. E. 1984. The relevance of non host-specific toxins in the expression of
virulence by pathogens. Ann. Rev. Phytopathol. 22:215-245.
Murashige, T., and Skoog, F. 1962. A revised medium for rapid growth and bioassays
with tobacco tissue cultures. Physiol. Plant. 15: 473-497.
Murray, N. E., Brammar, W. J., and Murray, K. 1977. Lambdoid phages that simplify
the recovery of in vitro recombinants. Mol. Gen. Genet. 150:53-61.
Nigg, E.A., Baeuerle, P.A., and Luhrmann R. 1991. Nuclear import-export: In search
of signals and mechanisms. Cell 66: 15-22.
Nishida, E., and Gotoh, Y. 1993. The MAP kinase cascade is essential for diverse signal
transduction pathways. Trend Biochem. Sci. 18: 128-131.
Okimoto, R., Chamberlin, H., Macfarlane, J.L., and Wolstenholme D. R. 1991.
Repeated sequence sets in mitochondrial DNA molecules of root knot nematodes
(Meloidogyne): nucleotide sequences, genome location and potential for host-race
identification. Nucleic Acid Res. 19:1619-1626.
Osbourn, A.E., Clarke, B.R. and Daniels, M.J. 1990. Identification and DNA sequence
of a pathogenicity gene of Xanthomonas campestris pv. campestris. Mol. Plant Microbe.
Interact. 3:280-285.
Panopoulos, N.J. and Peet, R.C. 1985. The molecular genetics of plant pathogenic
bacteria and their plasmids. Annu. Rev. Phytopathol. 23: 381-419.
Petes, T.D., and Hill, C.W. 1988. Recombination between repeated genes in
microorganisms. Annu. Rev. Genet. 22: 147-168.
Pierce, M.L., Essenberg, M. and Mort, A.J. 1993. A comparison of the quantities of
exopolysaccharide produced by Xanthomonas campestris pv. malvacearum in susceptible
and resistant cotton cotyledons during early stages of infection. Phytopathology
83:344-349.
Pryor, T. 1987. The origin and structure of fungal disease resistance genes in plants.
Trends Genet. 3: 157-161.

125
Rahme, L.G., Mindrinos, M.N., and Panopoulos, N.J. 1991. The genetic and
transcriptional organization of the hrp cluster of Pseudomonas syringae pathovar
phaseolicola. J. Bacteriol. 173: 575-586.
Raz, V., and Fluhr, R. 1993. Ethylene signal is transduced via protein phosphorylation
events in plants. Plant Cell 5: 523-530.
Restrepo, M.A., Freed, D.D, and Carrington, J.C. 1990. Nuclear transport of plant
potyviral proteins. Plant Cell 2: 987-998.
Ried, J.L., and Collmer, A. 1987. An nptl-sacB-sacR cartridge for constructing
directed, unmarked mutations in Gram-negative bacteria by marker exchange-eviction
mutagenesis. Gene 57:239-246.
Rihs, H.-P., Jans, D.A., Fan, H., and Peters, R. 1991. The rate of nuclear cytoplasmic
protein transport is determined by the casein kinase II site flanking the nuclear
localization sequence of the SV40 T-antigen. EMBO J. 10: 633-639.
Rihs, H.-P., and Peters, R. 1989. Nuclear transport kinetics depend on phosphorylation-
site-containing sequences flanking the karyophilic signal of the Simian virus 40 T-
antigen. EMBO J. 8: 1479-1484.
Roberts, D.P., Denny, T.P., and Schell, M.A. 1988. Cloning of the egl gene of
Pseudomonas solanacearwm and analysis of its role in pathogenicity. J. Baceriol. 170:
1445-1451.
Rossi, L., Hohn, B. and Tinland, B. 1993. The VirD2 protein of Agrobacterium
tumefaciens carries nuclear localization signals important for transfer of T-DNA to plant.
Mol. Gen. Genet. 239:345-353.
Saleh, Y.P., and Shaw, P.D. 1988. Isolation and characterization of pathogenicity genes
of Pseudomonas syringae pv. tabaci. J. Bacteriol. 170:2584-2591.
Sambrook, J., Fritsch, E.F., and Maniatis, T. 1989. Molecular cloning: a laboratory
manual. Ed. 2nd, Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY..
Schell, M.A., Rpberts, D.P., and Denny, T.P. 1988. Analysis of the Pseudomonas
solanacearum polygalacturonase encoded by pglA and its involvement in pathogenicity.
J. Bacteriol. 170: 4501-4508.
Schlessinger, J. 1993. How receptor tyrosine kinases activate Ras. Trends Biochem. Sci.
18: 273-275.

126
Schoulties, C.L., Civerolo, E.L., Miller, J.W., Stall, R.E., Krass, C.J., Poe, S.R., and
DuCharme, E.P. 1987. Citrus canker in Florida. Plant Dis. 71:388-395. —
Schreck, R., Rieber, P., and Baeuerle, P.A. 1991. Reactive oxygen intermediates as
apparently widely used messengers in the activation of the NF-kappa B transcription
factor and HIV-1. EMBO J. 10: 2247-2258.
Schulte, R. and Bonas, U. 1992. Expression of the Xanthomonas campestris pv.
vesicatoria hrp gene cluster, which determines pathogenicity and hypersensitivity on
pepper and tomato, is plant inducible. J. Bacteriol. 174:815-823.
Sentenac, H., Bonneaud, N., Minet, M., Lacroute, F., Salmon, J.M., Gaymard, F.,
Grignon, C. 1992. Cloning and expression in yeast of a plant potassium ion transport
system. Science 256: 663-665.
Shurviton, C.E., Hodges, L., and Ream, W. 1992. A nuclear localization signal in the
Agrobacterium VirD2 endonuclease is essential for tumor formation. Proc. Natl. Acad.
Sci. USA 89: 11837-11841.
Silver, P.A. 1991. How proteins enter the nucleus. Cell 64: 489-497.
Smith, M.J., Mazzola, E.P., Sims, J.J., Midland, S.L., Keen, N.T., Burton, V., and
Stayton, M.M. 1993. The syringolides: Bacterial C-glycosyl lipids that trigger plant
disease resistance. Tetrahedron Lett. 34:223-226.
Smidt, J., and Kosuge, T. 1978. The role of indole-3-acetic acid accumulation by alpha
methyl tryptophan-resistant mutants of Pseudomonas savastanoi in gall formation on
oleanders. Physiol. Plant Pathol. 13:203-214.
Stall, R.E., and Civerolo, E.L. 1991. Research relating to the recent outbreak of citrus
canker in Florida. Annu. Rev. Phytopathol. 29: 399-420.
Statler, G.D. 1985. Mutations to virulence and avirulence in Melampsora lini.
Phytopathol. 75:771-773.
Swamp, S., De Feyter, R., Brlansky, R.H. and Gabriel, D.W. 1991. A pathogenicity
locus from Xanthomonas citri enables strains from several pathovars of X. campestris to
elicit cankerlike lesions on citrus. Phytopathology 81:802-809.
Swamp, S., Yang, Y., Kingsley, M.T. and Gabriel, D.W. 1992. A Xanthomonas citri
pathogenicity gene, pthA, pleiotropically encodes gratuitous avimlence on nonhosts.
Molec. Plant-Microbe Interact. 5:204-213.

127
Tang, J.L., Liu, Y.N., Barber, C.E., Dow, J.M., Wootton, J.C. and Daniels, MJ.
1991. Genetic and molecular analysis of a cluster of rpf genes involved in positive
regulation of synthesis of extracellular enzymes and polysaccharide in Xanthomonas
campestris pathovar campestris. Mol. Gen. Genet. 226:409-417.
Taylor, M.G., and Vasil, I.K. 1991. Histology of, and physical factors affecting, GUS
expression in pearl millet {Pennisetum glaucum (L.) R.Br.) embryos following
microprojectile bombardment. Plant Cell Rep. 10:120-125.
Thiers, H. D., and Bank, L. M. 1951. A histological study of bacterial blight of cotton.
Phytopathol. 41:499-510.
Thompson, J. N., and Burdon, J. J. 1992. Gene-for-gene coevolution between plants and
parasites. Nature 360: 121-125.
Tinland, B., Koukolikova-Nicola, Z., Hall, M.N., and Hohn, B. 1992. The T-DNA
linked VirD2 protein contains two distinct functional nuclear localization signals. Proc.
Nad. Acad. Sci. USA 89: 7442-7446.
Towbin, H., Staehelin, T., and Gordon, J. 1979. Electrophoretic transfer of proteins
from polyacrylamide gels to nitrocellulose sheets: procedure and some applications. Proc.
Natl. Acad. Sci. USA 76: 4350-4354.
Truchet, G., Roche, P., Lerouge, P., Vasse, J., Camut, S., de Billy, F., Prome, J. and
Denarie, J. 1991. Sulphated lipo-oligosaccharide signals of Rhizobium meliloti elicit root
nodule organogenesis in alfalfa. Nature 351:670-673.
van der Krol, A.R., and Chua, N.-H. 1991. The basic domain of plant B-ZIP proteins
facilitates import of a reporter protein into plant nuclei. Plant Cell 3: 667-675.
van der Voom, L. and Ploegh H. L. 1992. The WD-40 repeat. FEBS Letter 307:131-
134.
van Kan, J.A.L., van den Ackerveken, G.F.J.M. and de Wit, P.J.G.M. 1991. Cloning
and characterization of cDNA of avirulence avr9 of the fungal pathogen Cladosporium
fulvum, causal agent of tomato leaf mold. Molec. Plant-Microbe Interact. 4:52-59.
Varagona, M.J., Schmidt, R.J., and Raikhel, N.V. 1991. Monocot regulatory protein
opaque-2 is localized in the nucleus of maize endosperm and transformed tobacco plants.
Plant Cell 3: 105-113.
Varagona, M.J., Schmidt, R.J., and Raikhel, N.V. 1992. Nuclear localization signals
required for nuclear targeting of the maize regulatory protein opaque-2. Plant Cell 4:
1213-1227.

128
Verma, J.P. 1986. Bacterial blight of cotton. CRC Press, Inc., Boca Raton, Florida.
Wallis A. E., and McMaster W. R. 1987. Identification of Leishmania genes encoding
proteins containing tandemly repeating peptides. J. Exp. Med. 166:1814-1824.
Waney, V. R., Kingsley, M. T., and Gabriel D. W. 1991. Xanthomonas campestris pv.
translucens genes determining host-specific virulence on cereals identified by Tn5-gu&4
insertion mutagenesis. Mol. Plant-Microbe Interact. 4:623-627.
Ward, E., and Barnes, W. 1988. VirD2 protein of Agrobacterium tumafaciens very
tightly linked to the 5’ end of T-strand DNA. Science 242: 927-930.
Wei, Z.M., Laby, R.J., Zumoff, C.H., Bauer, D.W., He, S.Y., Collmer, A. and Beer,
S.V. 1992. Harpin, elicitor of the hypersensitive response produced by the plant
pathogen Erwinia amylovora. science 257:85-88.
Whalen, M.C., Stall, R.E. and Staskawicz, B.J. 1988. Characterization of a gene from
a tomato pathogen determining hypersensitive resistance in non-host species and genetic
analysis of this resistance in bean. Proc. Natl. Acad. Sci. USA 85:6743-6747.
Willis, D.K., Hrabak, E.M., Rich, J.J., Barta, T.M., Lindow, S.E., and Panopoulos,
N J. 1990. Isolation and characterization of a Pseudomonas syringae pv. syringae mutant
deficient in lesion formation on bean. Mol. Plant-Microbe Interact. 3: 149-156.
Willis, D.K., Rich, J.J. and Hrabak, E.M. 1991. hrp genes of phytopathogenic bacteria.
Molec. Plant-Microbe Interact. 4:132-138.
Wren, B.W. 1991. A family of clostridial and streptococcal ligand-binding proteins with
conserved C-terminal repeat sequences. Mol. Microbiol. 5: 797-803.
Xu, B., Paszty, C., and Lurquin, P.F. 1988. A plasmid-based method to quantitate
homologous recombination frequencies in gram-negative bacteria. BioTechniques 6:752-
759.
Yanisch-Perron, C., Viera, J., and Messing, J. 1985. Improved M13 phage cloning
vectors and host strains: Nucleotide sequences of the M13mpl8 and pUC19 vectors.
Gene 33: 103-119.
Yu, L. M, Lamb, C.J., and Dixon, R.A. 1993. Purification and biochemical
characterization of proteins which bind to the H-box c/s-element implicated in
transcriptional activation of plant defense genes. Plant J. 3:805-816.

BIOGRAPHICAL SKETCH
Yinong Yang was bom on January 29, 1962, to Meizhen and Zhengqin Yang in
Huangyan, Zhejiang, China. He received a Bachelor of Science degree in biology from
Hangzhou University in China in 1982. From 1982 to 1987, he worked as a junior
research scientist at the Citrus Research Institute, Zhejiang Academy of Sciences in his
hometown. He came to the United States in 1988 and attended the University of South
Florida, where he earned a Master of Science degree in botany in 1990. He continued
his graduate studies at the University of Florida towards a Doctor of Philosophy degree
in plant molecular and cellular biology. He has accepted a postdoctoral fellowship from
the Waksman Institute, Rutgers University, New Jersey, to work with Daniel F. Klessig
on molecular aspects of plant disease resistance.
129

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.
Dean W. Gabriel, Chair
Associate Professor of
Plant Pathology
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.
L. Curtis Hannah
Professor of Horticultural
Sciences
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.
Roy A. wen sen f
Professor of Milrobiology
and Cell Science
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.
H. Corby Kistler
Associate Professor of
Plant Pathology

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.
Daryl R. Pring
Professor of Plant Pathology
This dissertation was submitted to the Graduate Faculty of the College of
Agriculture and to the Graduate School and was accepted as partial fulfillment of the

LD
ligo
i -y^3
o 111111 '•mi ii urn i
3 1262 08557 0793




PAGE 2

02/(&8/$5 0(&+$1,606 2) +267 63(&,),& 9,58/(1&( $1' $9,58/(1&( &$86(' %< &27721 %/,*+7 $1' &,7586 &$1.(5 3$7+2*(16 %\ <,121* <$1* $ ',66(57$7,21 35(6(17(' 72 7+( *5$'8$7( 6&+22/ 2) 7+( 81,9(56,7< 2) )/25,'$ ,1 3$57,$/ )8/),//0(17 2) 7+( 5(48,5(0(176 )25 7+( '(*5(( 2) '2&725 2) 3+,/2623+< 81,9(56,7< 2) )/25,'$

PAGE 3

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
PAGE 4

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

PAGE 5

$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\ 02/(&8/$5 0(&+$1,606 2) +26763(&,),& 9,58/(1&( $1' $9,58/(1&( &$86(' %< &27721 %/,*+7 $1' &,7586 &$1.(5 3$7+2*(16 %\
PAGE 6

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

PAGE 7

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f 2YHUDOO XS WR JHQHV PD\ EH UHTXLUHG IRU EDFWHULDO SDWKRJHQHVLV 3DQRSRXORV DQG 3HHW 'DQLHOV HW DO f LQFOXGLQJ WKRVH LQYROYHG LQ EDFWHULDO JURZWK LQ SODQWD LQGXFWLRQ RI SDWKRJHQLF V\PSWRPV KRVW UDQJH DQG DYLUXOHQFH *DEULHO f 2Q WKH RWKHU KDQG KRVW SODQW UHVLVWDQFH WR LQIHFWLRQ LQYROYHV SUHIRUPHG EDUULHUV LQDFWLYDWLRQ RI PLFURELDO WR[LQV UHFRJQLWLRQ RI SDWKRJHQ VLJQDOV DQG LQGXFWLRQ RI DFWLYH SODQW GHIHQVHV ,Q LQFRPSDWLEOH SODQWEDFWHULD LQWHUDFWLRQ LQGXFHG UHVLVWDQFH LV XVXDOO\ DVVRFLDWHG ZLWK D K\SHUVHQVLWLYH UHVSRQVH +5f FKDUDFWHUL]HG E\ WKH UDSLG ORFDOL]HG GHDWK RI SODQW FHOOV DW WKH VLWH RI LQIHFWLRQ 7KLV SURFHVV UHTXLUHV DFWLYDWLRQ RI SODQW UHVLVWDQFH JHQHV LQYROYLQJ VLJQDO SHUFHSWLRQ DQG WUDQVGXFWLRQ DQG SODQW GHIHQVH JHQHV LQYROYLQJ ELRV\QWKHVLV RI SDWKRJHQHVLVUHODWHG SURWHLQV SK\WRDOH[LQV DQG RWKHU DQWLELRWLFV 'L[RQ DQG +DUULVRQ 'L[RQ DQG /DPE f

PAGE 8

%DFWHULDO SODQW SDWKRJHQV UHTXLUH YLUXOHQFH JHQHV WR LQIHFW DQG WR FRORQL]H SODQWV $ ODUJH JURXS RI JHQHV FDOOHG K\SHUVHQVLWLYH UHVSRQVH DQG SDWKRJHQLFLW\ ^KUSf JHQHV KDYH EHHQ IRXQG JHQHUDOO\ FRQVHUYHG LQ DOO SK\WRSDWKRJHQLF EDFWHULD H[DPLQHG H J (UZLQLD 3VHXGRPRQDV DQG ;DQWKRPRQDVf :LOOLV HW DO f :KDWHYHU WKH EDFWHULDO SDWKRJHQV VWXGLHG W\SLFDO KUS PXWDQWV DUH FRPSOHWHO\ QRQSDWKRJHQLF DQG HOLFLW QHLWKHU SDWKRJHQLF V\PSWRPV RQ KRVW SODQWV QRU K\SHUVHQVLWLYH UHVSRQVH RQ QRQKRVW SODQWV 0RVW KUS JHQHV DUH FOXVWHUHG RQ NE '1$ IUDJPHQWV DQG DUH FRQVHUYHG DW WKH JHQXV RU IDPLO\ OHYHO +LJK OHYHOV RI '1$ KRPRORJ\ DUH IRXQG DPRQJ PDQ\ KUS JHQHV IURP GLIIHUHQW JHQHUD RI SK\WREDFWHULD %RXFKHU HW DO f 6HYHUDO ELRORJLFDO IXQFWLRQV DUH HQFRGHG E\ GLIIHUHQW KUS JHQHV 6RPH KUS JHQHV HJ KUS6 RI 3 V\ULQJDH SY SKDVHROLFRODf DUH FOHDUO\ UHJXODWRU\ UHVSRQGLQJ WR SODQW VLJQDOV WR LQGXFH WUDQVFULSWLRQ RI KUS RSHURQ LQ D PDQQHU WKDW PD\ EH DQDORJRXV WR WKH WZRFRPSRQHQW VLJQDOLQJ V\VWHPV 5DKPH HW DO f 2WKHU KUS JHQHV DUH LQYROYHG LQ WKH V\QWKHVLV DQG H[SRUW RI SDWKRJHQLFLW\ IDFWRUV WKDW DOVR LQGXFH DQ +5 RQ QRQKRVW SODQWV '1$ VHTXHQFH DQDO\VLV UHYHDOV WKDW VHYHUDO KUS JHQHV RI 3VHXGRPRQDV DQG ;DQWKRPRQDV DUH UHPDUNDEO\ VLPLODU WR NH\ YLUXOHQFH JHQHV RI DQLPDO EDFWHULD VXFK DV
PAGE 9

PD\ DOVR LQYROYH H[SRUW RI DYLUXOHQFH HOLFLWRUV )HQVHODX HW DO f 7KHUHIRUH KUS JHQHV PD\ FRQWURO RU FRQGLWLRQ WKH IXQFWLRQV HQFRGHG E\ YLUXOHQFH DQG DYLUXOHQFH JHQHV %DFWHULD EHORQJLQJ WR WKH JHQXV ;DQWKRPRQDV DUH SODQWDVVRFLDWHG DQG LQIHFW PRQRFRW DQG GLFRW VSHFLHV LQ IDPLOLHV /H\QV HW DO f $OWKRXJK WKH JHQXV DV D ZKROH KDV ZLGH KRVW UDQJH LQGLYLGXDO VWUDLQV KDYH KRVW UDQJHV OLPLWHG WR RQO\ FHUWDLQ SODQW VSHFLHV 7KLV SKHQRPHQRQ RI KRVW VSHFLDOL]DWLRQ LQGLFDWHV WKDW HVWDEOLVKPHQW RI ;DQWKRPRQDV LQIHFWLRQ LQYROYHV KRVWVSHFLILF UHFRJQLWLRQ LQ DGGLWLRQ WR JHQHUDO YLUXOHQFH IXQFWLRQ RI SDWKRJHQ FRQIHUUHG E\ KUS JHQHV +RVWVSHFLILF YLUXOHQFH KVYf JHQHV HQFRGLQJ SRVLWLYH IXQFWLRQ KDYH EHHQ VKRZQ WR EH D PDMRU GHWHUPLQDQW IRU KRVW UDQJH DQG SDWKRYDU VWDWXV RI ;DQWKRPRQDV *DEULHO HW DO f 7KHVH JHQHV DUH UHTXLUHG IRU EDFWHULDO SDWKRJHQV WR JURZ DQGRU WR LQGXFH SDWKRJHQLF V\PSWRPV RQ VSHFLILF KRVW SODQWV EXW QRW LQ RWKHU SODQW VSHFLHV 6HYHUDO KRVWVSHFLILF YLUXOHQFH JHQHV KDYH EHHQ LVRODWHG IURP ; FDPSHVWULV SY FLWUXPHOR .LQJVOH\ HW DO f ; F SY WUDQVOXFHQV :DQH\ HW DO f DQG ; FLWUL 6ZDUXS HW DO f )RU H[DPSOH RSV; RI ; F SY FLWUXPHOR LV LQYROYHG LQ ELRV\QWKHVLV RI OLSRSRO\VDFFKDULGH DQG H[WUDFHOOXODU SRO\VDFFKDULGH DQG LV UHTXLUHG IRU YLUXOHQFH RQ FLWUXV EXW QRW RQ EHDQ .LQJVOH\ HW DO f 7KHVH KVY JHQHV DUH DQDORJRXV WR KRVWVSHFLILF QRGXODWLRQ KVQf JHQHV RI 5KL]RELXP 7KH KVQ JHQHV RI 5KL]RELXP LV LQYROYHG LQ KRVWVSHFLILF PRGLILFDWLRQ RI OLSRROLJRVDFFKDULGH 1RG IDFWRUV DQG IXQFWLRQ RQO\ IRU QRGXODWLRQ RQ VSHFLILF KRVWV )LVKHU DQG /RQJ 'HQDULH HW DO f %HVLGHV KVY JHQHV DYU JHQHV KDYH EHHQ VKRZQ WR SOD\ D UROH LQ GHWHUPLQLQJ KRVW VSHFLILFLW\ DW WKH VSHFLHV OHYHO .HHQ +HDWK f 6LQJOH FORQHG DYU JHQHV IURP

PAGE 10

D SDWKRJHQ RI RQH KRVW VSHFLHV FDQ FDXVH DQ RWKHUZLVH YLUXOHQW SDWKRJHQ RI DQRWKHU KRVW VSHFLHV WR EHFRPH DYLUXOHQW RQ LWV RZQ KRVW .RED\DVKL HW DO :KDOHQ HW DO f 7KHUHIRUH KRVW VSHFLHV VSHFLILFLW\ LV FRQGLWLRQHG E\ ERWK KRVWVSHFLILF YLUXOHQFH DQG DYLUXOHQFH JHQHV ZLWK WKH IRUPHU DV WKH PDMRU GHWHUPLQDQW $OWKRXJK DYU JHQHV PD\ FRQGLWLRQ KRVW UDQJH DW WKH VSHFLHV OHYHO LQ VRPH FDVHV WKH\ DFW DV QHJDWLYH IDFWRUV WR FRQIHU DYLUXOHQFH DQG WR OLPLW KRVW UDQJH RQO\ LI WKH KRVW SODQWV KDYH VSHFLILF UHVLVWDQFH 5f JHQHV 7KH LQWHUDFWLRQ RI SDWKRJHQV ZLWK DYU JHQHV DQG KRVW SODQWV ZLWK VSHFLILF 5 JHQHV UHVXOWV LQ WKH SODQW GHIHQVH UHVSRQVH RIWHQ NQRZQ DV K\SHUVHQVLWLYH UHVSRQVH 7KLV JHQHWLF UHTXLUHPHQW IRU VSHFLILF SODQW 5 JHQHV DQG VSHFLILF PLFURELDO DYU JHQHV LV FDOOHG D JHQHIRUJHQH LQWHUDFWLRQ 7KH JHQHIRUJHQH LQWHUDFWLRQV ZHUH ILUVW GHPRQVWUDWHG E\ )ORU f XVLQJ IOD[ DQG IOD[ UXVW DQG KDYH VLQFH EHHQ REVHUYHG LQ PDQ\ SODQWSDUDVLWH LQWHUDFWLRQV EHWZHHQ IXQJL EDFWHULD YLUXV QHPDWRGHV DQG WKHLU UHVSHFWLYH KRVWV 7KRPSVRQ DQG %XUGRQ f 'XULQJ WKH SDVW GHFDGH PDQ\ DYU JHQHV KDYH EHHQ LVRODWHG IURP GLIIHUHQW EDFWHULDO DQG IXQJDO SODQW SDWKRJHQV *DEULHO HW DO .HHQ f ,Q VRPH LQVWDQFHV DYU JHQHV HQFRGH SURWHLQDFHRXV HOLFLWRUV WKDW GLUHFWO\ LQGXFH WKH +5 RQ UHVLVWDQFH SODQWV )RU H[DPSOH DYU DQG DYU RI WKH IXQJDO SDWKRJHQ &ODGRVSRULXP IXOYXP HQFRGH UDFHVSHFLILF SHSWLGHV ZKLFK DUH H[FUHWHG LQWR WKH DSRSODVW RI LQIHFWHG OHDYHV DQG LQGXFH WKH +5 RQ WRPDWR FXOWLYDUV WKDW FDUU\ UHVLVWDQFH JHQHV &I DQG &I UHVSHFWLYHO\ -RRVWHQ HW DO YDQ .DQ HW DO f ,Q DQRWKHU FDVH DYU' RI 3VHXGRPRQDV V\ULQJDH SY WRPDWR DSSHDUV WR HQFRGH D HQ]\PH LQYROYHG LQ V\QWKHVLV RI D UDFHVSHFLILF JO\FROLSLG HOLFLWRU .HHQ HW DO .RED\DVKL HW DO 0LGODQG HW

PAGE 11

DO f ,Q PRVW FDVHV KRZHYHU WKH ELRFKHPLFDO LGHQWLWLHV RI DZ JHQH SURGXFWV DUH XQNQRZQ DQG WKH VHTXHQFH DQDO\VHV RI DZ JHQHV DUH XQLQIRUPDWLYH DV WR KRZ WKH\ PLJKW LQWHUDFW ZLWK 5 JHQH SURGXFWV WR WULJJHU WKH +5 DQG VSHFLI\ JHQHIRUJHQH LQWHUDFWLRQ 6HYHUDO PRGHOV LQFOXGLQJ HOLFLWRUUHFHSWRU PRGHO GLPHU PRGHO DQG LRQ FKDQQHO GHIHQVH PRGHOf KDYH EHHQ SURSRVHG WR H[SODLQ WKH PROHFXODU DQG ELRFKHPLFDO EDVHV RI JHQHIRUJHQH UHFRJQLWLRQ EHWZHHQ SODQW DQG SDWKRJHQ *DEULHO DQG 5ROIH f %XW YHU\ OLWWOH H[SHULPHQWDO HYLGHQFH LV DYDLODEOH WR SURYH DQ\ RI WKH K\SRWKHVHV 5HFHQWO\ WKH ILUVW SODQW 5 JHQH 3WR KDV EHHQ LVRODWHG IURP WRPDWR E\ PDSEDVHG SRVLWLRQDO FORQLQJ 0DUWLQ HW DO f 7KLV JHQH HQFRGHV D VHULQHWKUHRQLQH SURWHLQ NLQDVH WKDW LV KRPRORJRXV WR WKH UHFHSWRU SURWHLQ NLQDVH 65. LQYROYLQJ %UDVVLFD SROOHQVWLJPD UHFRJQLWLRQ DQG WR WKH 5DI SURWHLQ NLQDVH LQYROYLQJ WKH PDPPDOLDQ 5DV VLJQDOLQJ SDWKZD\ 7KH 5DV SDWKZD\ LV WKH EHVW NQRZQ FLUFXLW RI VLJQDO WUDQVGXFWLRQ IRXQG LQ \HDVW LQVHFW DV ZHOO DV PDPPDOLDQ V\VWHPV &XORWWD DQG .RVKODQG f 7KH 5DV SDWKZD\ EHJLQV ZLWK H[WUDFHOOXODU VLJQDOV H J JURZWK IDFWRUV RU F\WRNLQHVf WKDW LQWHUDFW ZLWK UHFHSWRU SURWHLQ NLQDVHV ZKLFK LQ WXUQ WULJJHU SKRVSKRU\ODWLRQ FDVFDGHV DQG ILQDOO\ H[SUHVVLRQ RI WDUJHWHG JHQHV &UHZV DQG (ULNVRQ f $OWKRXJK 5DVOLNH VLJQDOLQJ SDWKZD\ KDV QRW EHHQ GHPRQVWUDWHG LQ SODQWV KRPRORJXHV RI PDPPDOLDQ VLJQDO WUDQVGXFWLRQ FRPSRQHQWV ZHUH IRXQG LQ PDQ\ SODQWV 7KRVH LQFOXGH KRPRORJXHV RI *SURWHLQ VXEXQLWV H J DX[LQUHJXODWHG $UF$ RI WREDFFR ,VKLGD HW DO f UHFHSWRU SURWHLQ NLQDVHV HJ SROOHQVSHFLILF 65. RI %UDVVLFD *RULQJ DQG 5RWKVWHLQ f 5DVUHODWHG SURWHLQV HJ 5JSO RI ULFH .DPDGD HW DO f 5DI HJ 3WR RI WRPDWR 0DUWLQ HW DO DQG HWK\OHQHUHJXODWHG &75 RI

PAGE 12

$UDELGRSRVLV .LHEHU HW DO f DQG PLWRJHQDFWLYDWHG SURWHLQ NLQDVH HJ 0V(5.O RI DOIDOID 'XHUU HW DO f ,W LV SRVVLEOH WKDW YLUXOHQFH DQG DYLUXOHQFH JHQH SURGXFWV PD\ DFWLYDWH WKH VLPLODU VLJQDO WUDQVGXFWLRQ SDWKZD\ LQ SODQWV OHDGLQJ WR GLVHDVH V\PSWRPV RU K\SHUVHQVLWLYH UHVSRQVH )RU H[DPSOH WUDQVFULSWLRQDO DFWLYDWLRQ RI SODQW GHIHQVH JHQHV LV PRGXODWHG E\ SKRVSKRU\ODWLRQ )HOL[ HW DO
PAGE 13

DQG DYU%V3 RI ; FDPSHVWULV SY YHVLFDWRULD %RQDV HW DO &DQWHURV HW DO f DQG DYU[D6 DYU;D DQG DYU;DO2 RI ; RU\]DH +RSNLQV HW DO f 7KH PRVW FRQVSLFXRXV IHDWXUH RI WKLV JHQH IDPLO\ LV WKH QHDUO\ LGHQWLFDO WDQGHPO\ DUUDQJHG ES UHSHDWV LQ WKH FHQWUDO SRUWLRQ RI JHQHV )LJ f 'HOHWLRQ DQDO\VHV RI DYU%V KDYH VKRZQ WKDW WKH ES UHSHDWHG PRWLIV GHWHUPLQHV WKH DYLUXOHQFH VSHFLILFLW\ RI WKH JHQH +HUEHUV HW DO f +RZHYHU PDQ\ TXHVWLRQV UHPDLQ UHJDUGLQJ WR WKHLU SOHLRWURSLF IXQFWLRQV VSHFLILFLW\ HYROXWLRQ DQG ELRFKHPLFDO PHFKDQLVP RI JHQHIRUJHQH UHFRJQLWLRQ 7KH SULPDU\ REMHFWLYHV RI WKLV ZRUN ZHUH WR GHWHUPLQH f ZK\ DYLUXOHQFH JHQHV DUH SUHVHQW LQ WKH FRWWRQ EOLJKW SDWKRJHQ DQG ZKHWKHU WKH\ FRQIHU DQ\ SOHLRWURSLF IXQFWLRQV f ZKDW GHWHUPLQHV DYLUXOHQFH DQG SDWKRJHQLF VSHFLILFLW\ LQ FRWWRQ EOLJKW DQG FLWUXV FDQNHU SDWKRJHQV f KRZ WKH VSHFLILFLW\ LV FUHDWHG DQG HYROYHG LQ WKLV DYUSWK JHQH IDPLO\ DQG f KRZ WKH VLJQDOV HQFRGHG E\ WKH DYUSWK JHQHV DUH SHUFHLYHG DQG WUDQVGXFHG LQ SODQW FHOOV

PAGE 14

1DPH *HQH 6WUXFWXUH ES 3OHLRWURSLF 5HSHDWV )XQFWLRQ SWK$ DYUE DYU%V DYU;DO2 %3 6W + % 6 FDQNHULQJ ZDWHUVRDNLQJ f§ )LJXUH $ ;DQWKRPRQDV DYUSWK JHQH IDPLO\ /LVWHG DUH IRXU UHSUHVHQWLYH PHPEHUV SWK$ RI ; FLUUL DYUE RI ; FDPSHVWULV SY PDOYDFHDUXP DYU%V RI ; FDPSHVWULV SY YHVLFDWRULD DQG DYU;DO2 RI ; RU\]DH SY RU\]DH 7KH WDQGHPO\ DUUDQJHG EORFNV UHSUHVHQW WKH ES UHSHDWV LQ WKH FHQWUDO SRUWLRQ RI JHQHV 5HVWULFWLRQ HQ]\PH FOHDYDJH VLWHV UHOHYDQW WR WKLV ZRUN DQG IRXQG LQ PRVW PHPEHUV RI WKH JHQH IDPLO\ DUH % %DP+O 3 3VWO 6W 6WXO + +LQGLn 6 6VWO

PAGE 15

&+$37(5 +26763(&,),& 6<037206 $1' ,1&5($6(' 5(/($6( 2) ;$17+2021$6 &,75 $1' ; &$03(6756 39 0$/9$&($580 )520 /($9(6 $5( '(7(50,1(' %< 7+( %3 7$1'(0 5(3($76 2) SWK$ $1' DZE 5(63(&7,9(/< ,QWURGXFWLRQ 0LFURELDO JHQHV LQYROYHG LQ SODQWPLFUREH LQWHUDFWLRQV PD\ EH IXQFWLRQDOO\ FODVVLILHG LQWR IRXU EURDG FDWHJRULHV SDUDVLWLF SDWKRJHQLF KRVW UDQJH DQG DYLUXOHQFH *DEULHO f *HQHV LQYROYHG LQ SDUDVLWLVP DUH DEVROXWHO\ UHTXLUHG IRU JURZWK LQ SODQWD DQG DUH ZLGHO\ FRQVHUYHG DW WKH IDPLO\ RU JHQXV OHYHO ([DPSOHV LQFOXGH PRVW KUS K\SHUVHQVLWLYH UHVSRQVH DQG SDWKRJHQLFLW\f JHQHV RI (UZLQLD 3VHXGRPRQDV DQG ;DQWKRPRQDV DQG WKH FRPPRQ QRG QRGXODWLRQf JHQHV RI 5KL]RELXP %RXFKHU HW DO 'HQDULH HW DO :LOOLV HW DO f *HQHV LQYROYHG LQ SDWKRJHQLFLW\ DUH UHTXLUHG IRU LQGXFWLRQ RI V\PSWRPV ([DPSOHV LQFOXGH SHFWDWH O\DVH SRO\JDODFWXURQDVH DQG HQGRJOXFDQDVH JHQHV &ROOPHU DQG .HHQ 6FKHOO HW DO 5REHUWV HW DO f GVS GLVHDVH VSHFLILF $UODW DQG %RXFKHU f ZWV ZDWHUVRDNLQJ &RSOLQ HW DO f JHQHV SK\WRKRUPRQH ELRV\QWKHWLF JHQHV 6PLGW DQG .RVXJH f DQG WR[LQ ELRV\QWKHWLF JHQHV 0LWFKHOO f *HQHV LQYROYHG LQ FRQGLWLRQLQJ KRVW UDQJH DUH KRVW VSHFLILF DQG UHTXLUHG IRU JURZWK RQ VSHFLILF KRVWV ([DPSOHV LQFOXGH KVY KRVWVSHFLILF YLUXOHQFHf SWK SDWKRJHQLFLW\f DQG KVQ KRVWVSHFLILF QRGXODWLRQf JHQHV RI 3VHXGRPRQDV

PAGE 16

;DQWKRPRQDV DQG 5KL]RELXP 'HQDULH HW DO .LQJVOH\ HW DO 0D HW DO 6DOHK DQG 6KDZ 6ZDUXS HW DO :DQH\ HW DO *DEULHO HW DO f ,QIRUPDWLRQ RQ FRQVHUYDWLRQ RI WKHVH JHQHV ZLWKLQ VSHFLHV SDWKRYDUV DQG ELRYDUV LV VFDUFH EXW WKHVH JHQHV DSSHDU WR GHWHUPLQH ELRYDU DQG SDWKRYDU VWDWXV 'MRUGMHYLF HW DO *DEULHO HW DO f 7KH IRXUWK JURXS DUH WHUPHG DYLUXOHQFH DYUf JHQHV EHFDXVH WKH\ QHJDWLYHO\ DIIHFW YLUXOHQFH 7KH DZ JHQHV DUH VXSHULPSRVHG RQ EDVLF FRPSDWLELOLW\ (OOLQJERH f DUH QRW KLJKO\ FRQVHUYHG DQG GHWHUPLQH SDWKRJHQLF UDFHV EHORZ WKH VSHFLHV ELRYDU RU SDWKRYDU OHYHOV 7KHVH IRXU EURDG FDWHJRULHV DUH QRW PXWXDOO\ H[FOXVLYH 7KH LQWHUDFWLRQ RI PLFUREHV KDYLQJ DZ JHQHV DQG KRVW SODQWV KDYLQJ UHVLVWDQFH "f JHQHV FDQ UHVXOW LQ SODQW GHIHQVH UHVSRQVHV RIWHQ REVHUYHG YLVXDOO\ DV D K\SHUVHQVLWLYH UHDFWLRQ +5f DQG FKDUDFWHUL]HG E\ WKH UDSLG QHFURVLV RI SODQW FHOOV DW WKH VLWH RI LQIHFWLRQ DQG WKH DFFXPXODWLRQ RI SK\WRDOH[LQV 0RVW DZ JHQHV GR QRW DSSHDU WR FRQIHU VHOHFWLYH DGYDQWDJH WR WKH SDWKRJHQ *DEULHO .HHQ DQG 6WDVNDZLF] f 3OHLRWURSLF IXQFWLRQV KDYH EHHQ LGHQWLILHG IRU RQO\ RI WKH DZ JHQHV FORQHG WR GDWH *DEULHO HW DO f )XUWKHUPRUH WKH '1$ VHTXHQFHV RI WKH FORQHG DYU JHQHV KDYH EHHQ UHPDUNDEO\ XQLQIRUPDWLYH LQ WHUPV RI IXQFWLRQ .HHQ f 7KH SUHVHQFH RI PRVW DYU JHQHV LQ SODQW SDWKRJHQV WKHUHIRUH UHPDLQV HQLJPDWLF 5HFHQWO\ DQ DYU JHQH IDPLO\ KDV EHHQ GLVFRYHUHG LQ PDQ\ GLIIHUHQW [DQWKRPRQDGV PHPEHUV LQFOXGH DYU%V DYU%V DQG DYU%V3 %RQDV HW DO &DQWHURV HW DO f RI ; FDPSHVWULV SY YHVLFDWRULD DYU% DYUE DYUE DYU%,Q DYU%O2O DQG DYU% RI; FDPSHVWULV SY PDOYDFHDUXP 'H )H\WHU DQG *DEULHO D

PAGE 17

'H )H\ WHU HW DO f DQG DYU[D DYU;D DQG DYU;DO2 RI ; RU\]DH +RSNLQV HW DO f ,QWHUHVWLQJO\ WKLV JHQH IDPLO\ LQFOXGHV D JHQH SWK$ WKDW LV UHTXLUHG IRU SDWKRJHQLFLW\ RI ; FLWUL RQ FLWUXV 7KLV JHQH LV QRW NQRZQ WR IXQFWLRQ IRU DYLUXOHQFH LQ ; FLWUL 6ZDUXS HW DO f EXW LV UHTXLUHG IRU ; FLWUL WR LQGXFH FHOO GLYLVLRQV LQ WKH OHDI PHVRSK\OO RI FLWUXV OHDGLQJ WR HSLGHUPDO UXSWXUH DQG VXEVHTXHQW UHOHDVH RI WKH EDFWHULD RQWR WKH OHDI VXUIDFH 7KH JHQH DOVR FRQIHUV WKH DELOLW\ WR LQGXFH FHOO GLYLVLRQV RQ FLWUXV WR ; FDPSHVWULV VWUDLQV IURP VHYHUDO GLIIHUHQW SDWKRYDUV 6ZDUXS HW DO f 'H )H\WHU DQG *DEULHO Df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f 7KH SXUSRVH RI WKLV VWXG\ ZDV WR FKDUDFWHUL]H WKH ZDWHUVRDNLQJ IXQFWLRQV RI DYUE DQG WR LQYHVWLJDWH WKH UROH RI WKH ES UHSHDWV RI SWK$ DYU% DYUE DYUE DYU%,Q DYU%O2O DQG DYU% LQ SDWKRJHQLFLW\ DQG DYLUXOHQFH

PAGE 18

0DWHULDOV DQG 0HWKRGV %DFWHULDO VWUDLQV SODVPLGV DQG FXOWXUH PHGLD 7KH EDFWHULDO VWUDLQV DQG SODVPLGV XVHG LQ WKLV VWXG\ DUH OLVWHG LQ 7DEOH 6WUDLQV RI (VFKHULFKLD FROL ZHUH JURZQ LQ /XULD%HUWDQL /%f PHGLXP 6DPEURRN HW DO f DW r& 6WUDLQV RI ;DQWKRPRQDV ZHUH JURZQ LQ 3<*0 SHSWRQH\HDVW H[WUDFWJO\FHURO0236f PHGLXP DW r& 'H )H\WHU HW DO f )RU FXOWXUH RQ VROLG PHGLXP DJDU ZDV DGGHG DW J/ $QWLELRWLFV ZHUH XVHG DW WKH IROORZLQJ ILQDO FRQFHQWUDWLRQV PJ/f DPSLFLOOLQ $Sf NDQDP\FLQ .Pf JHQWDP\FLQ *Pf VSHFWLQRP\FLQ 6SFf WHWUDF\FOLQH 7Ff DQG ULIDPSLFLQ 5LIf 5HFRPELQDQW '1$ WHFKQLTXHV 7RWDO '1$ LVRODWLRQ IURP ;DQWKRPRQDV ZDV DV GHVFULEHG *DEULHO DQG 'H )H\WHU f 3ODVPLGV ZHUH LVRODWHG IURP ( FROL E\ DONDOLQH O\VLV PHWKRGV 6DPEURRN HW DO f 5HVWULFWLRQ HQ]\PH GLJHVWLRQ DONDOLQH SKRVSKDWDVH WUHDWPHQW '1$ OLJDWLRQ DQG UDQGRP SULPLQJ UHDFWLRQV ZHUH SHUIRUPHG DV UHFRPPHQGHG E\ WKH PDQXIDFWXUHUV 6RXWKHUQ K\EULGL]DWLRQ ZDV SHUIRUPHG E\ XVLQJ Q\ORQ PHPEUDQHV DV GHVFULEHG /D]R HW DO f 2WKHUZLVH VWDQGDUG UHFRPELQDQW '1$ SURFHGXUHV ZHUH XVHG 6DPEURRN HW DO f &RQVWUXFWLRQ RI FKLPHULF JHQHV 7R FRQVWUXFW %DP+O IUDJPHQWVZDSSHG FKLPHULF JHQHV DPRQJ DYU% DYUE DYUE DYU%OQ DYU%O2O DYU% WKH %DP+O IUDJPHQWV IURP WKHVH JHQHV ZHUH FORQHG LQWR DQ DYUE VKHOO FRQWDLQLQJ f DQG f HQGV RI DYUE EXW GHOHWHG IRU LWV %DP+O IUDJPHQW S8)5$%DPf DQG DQ DYU%OQ VKHOO FRQWDLQLQJ f DQG f HQGV RI DYU%OQ EXW GHOHWHG IRU LWV %DP+O IUDJPHQW S8)5$%DPf RQ WKH

PAGE 19

7DEOH %DFWHULDO VWUDLQV DQG SODVPLGV XVHG LQ WKLV VWXG\ 6WUDLQ RU 3ODVPLG 5HOHYDQW &KDUDFWHULVWLFV 5HIHUHQFH RU 6RXUFH ( FROO '+m ) HQG$O KVG5 UNfPNf VXS( WKL UHF$O J\U$ UHO$, GODF= $0, $ODF=<$DUJ)f9 *LEFR%5/*DLWKHUVEXUJ0' +% VXS( $VG6UNaPNf UHF$ DUD SUR$ ODF
PAGE 20

7DEOH f§FRQWLQXHG 6WUDLQ RU 3ODVPLG 5HOHYDQW &KDUDFWHULVWLFV 5HIHUHQFH RU 6RXUFH S8)5 56) UHSOLFQ &P 6PU ,QF: 0RE GLVSODFHPHQW YHFWRU 6ZDUXS HO DO S8)5 ,QF3 7FU 0RE FRQWDLQLQJ PHWK\ODVHV ;PD, DQG ;PDO8 'H)H\WHU t *DEULHO E S8)5 NE IUDJPHQW FRQWDLQLQJ DYU% LQ S8)5 'H )H\WHU t *DEULHO D S8)5 NE IUDJPHQW FRQWDLQLQJ DYUE LQ S8)5 'H )H\WHU t *DEULHO D 3)5 ODF=DYUE IXVLRQ LQ S8)5 'H )H\WHU HW DO S8)5 NE IUDJPHQW FRQWDLQLQJ DYU%O2O LQ S8)5 'H )H\WHU HW DO S8)5 NE IUDJPHQW FRQWDLQLQJ DYU%,Q LQ S8)5 'H )H\WHU HW DO S8)5 ONE IUDJPHQW FRQWDLQLQJ DYU% LQ S8)5 'H )H\WHU HW DO S8)5 O2NE IUDJPHQW FRQWDLQLQJ DYUE LQ S8)5 'H )H\WHU HW DO S8)5$%DP S8)5 GHOHWHG IRU NE %XP +, IUDJPHQW 7KLV VWXG\ S8)5 ,QWHUQDO %XP +, IUDJPHQW RI DYU% LQ S*HPO O=If 'H )H\WHU HW DO S8)5 ,QWHUQDO %RP+, IUDJPHQW RI DYUE LQ S*HPO O=If 'H )H\WHU HW DO S8)5 ,QWHUQDO %DP:O IUDJPHQW RI DYU%O2O LQ S*HPO O=If 'H )H\WHU HW DO S8)5 ,QWHUQDO %RP+, IUDJPHQW RI DYU%,Q LQ S*HPO O=If 'H )H\WHU HW DO S8)5 ,QWHUQDO %RP+, IUDJPHQW RI DYU% LQ S*HPO O=If 'H )H\WHU HW DO S8)5 ,QWHUQDO %RP+, IUDJPHQW RI DYUE LQ S*HPO O=If 'H )H\WHU HW DO S8)5 ,QWHUQDO %DP:O IUDJPHQW RI DYU%,Q LQ S*HPO O=I f 6DIL VLWH ILOOHG 7KLV VWXG\ S8)5 ,QWHUQDO %RP+, IUDJPHQW RI DYUE LQ S*HPO O=If 6DO@ VLWH ILOOHG 7KLV VWXG\ S8)5 NE IUDJPHQW FRQWDLQLQJ DYUE LQ S8)5 'H )H\WHU HW DO S8)5 NE %JO8(FR%O IUDJPHQW FRQWDLQLQJ DYU%,Q IURP S8)5 LQ S8)5 7KLV VWXG\ S8)5$%DP S8)5 GHOHWHG IRU NE %RP+, IUDJPHQW 7KLV VWXG\ S8)5 %XP+,VZDSSHG FKLPHULF JHQHV ZLWK DYUE 6n DQG f HQGV SOXV %RP+, IUDJPHQW RI DYU% DYUE DYU%,Q DYU%O2O RU DYU% LQ S8)5 7KLV VWXG\ S8)5 %RP+,VZDSSHG FKLPHULF JHQHV ZLWK DYU%,Q 6f DQG f HQGV SOXV %RP+, IUDJPHQW RI DYU% DYUE DYUE DYU%O2O RU DYU% LQ S8)5 7KLV VWXG\ S8)5 6WXOO +LQF,QVZDSSHG FKLPHULF JHQHV ZLWK DYU%,Q 6f DQG n UHJLRQV SOXV 6LX/LQF,, IUDJPHQW RI DYU%O2O DYU% DYUE DYU% RU DYUE LQ S8)5 7KLV VWXG\ S8)5 6WX9+PF8VZDSSHG FKLPHULF JHQHV ZLWK DYUE 6f DQG f UHJLRQV SOXV 6WXOO+LQF8 IUDJPHQW RI DYU%O2O DYU% DYUE DYU%,Q RU DYU% LQ S8)5 7KLV VWXG\ S8)5 S8)5 GHULYDWLYH DYU%,Q7QJXV$ 7KLV VWXG\ S8)5 S8)5 GHULYDWLYH DYUE7WLJXV$ 7KLV VWXG\ S8)5 S8)5 GHULYDWLYH DYUE7Q6JXV$ 7KLV VWXG\ S8)< NE $X,WILQF4VZDSSHG IUDJPHQW ZLWK SWK$ 6n DQG f UHJLRQV RI S=LW SOXV DYUE LQWHUQDO UHSHDW UHJLRQ RI S8)5 LQ S8)5 7KLV VWXG\ S8)< NE 6UX,7MQFO,VZDSSHG IUDJPHQW ZLWK DYUE 6n DQG f UHJLRQV RI S8)5 SOXV SWK$ LQWHUQDO UHSHDW UHJLRQ RI S=LW LQ S8)5 7KLV VWXG\ S=LW NE IUDJPHQW FRQWDLQLQJ SWK$ LQ S8)5 6ZDUXS HW DO S*(0=If &RO(O $SU ODF=D 3URPHJD &R 0DGLVRQ :, S*(0OO=If &RO(O $SU ODF=Dr 3URPHJD &R 0DGLVRQ :O

PAGE 21

S8)5 YHFWRU 7KH UHVXOWLQJ %DP+O IUDJPHQWVZDSSHG FKLPHULF JHQHV IRUPHG S8)5 7KH DYU JHQHV IURP S;FP+ DQG SWK$ IURP ; FLWUL DOO KDYH XQLTXH 6WXO DQG +LQGL VLWHV GHOLPLWLQJ WKH ES UHSHDWHG UHJLRQV ZKLFK DOORZ WKH VZDSSLQJ RI WKH LQWHUQDO UHSHDWHG UHJLRQV DPRQJ WKHVH JHQHV 7R FRQVWUXFW 6WXO+LQGO IUDJPHQWVZDSSHG FKLPHUDV XVLQJ WKH S;FP+ DYU JHQHV S8)5 DQG S8)5 ZHUH FXW ZLWK 6DLO EOXQW HQGHG XVLQJ WKH .OHQRZ IUDJPHQW DQG UHOLJDWHG WR GHVWUR\ WKH +LQGL VLWHV RQ WKH S*(0O O=If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f 7KH 6WXO+LQGO LQWHUQDO IUDJPHQWV ZHUH VZDSSHG EHWZHHQ WKH WZR S*(0 GHULYDWLYHV WR FUHDWH FKLPHULF JHQHV 7KH FKLPHULF JHQHV ZHUH WKHQ UHFORQHG DV VLQJOH (FR.O+LQGOOO IUDJPHQWV LQWR S8)5 IRUPLQJ S8)< DQG S8)< %DFWHULDO FRQMXJDWLRQ 7ULSDUHQWDO PDWLQJV ZHUH FDUULHG RXW WR WUDQVIHU EURDG KRVW UDQJH SODVPLGV IURP (FROL '+D WR YDULRXV ;DQWKRPRQDV VWUDLQ E\ XVLQJ S5. RU S5. DV KHOSHU SODVPLGV DV GHVFULEHG 'H )H\WHU DQG *DEULHO Df 7R WUDQVIHU

PAGE 22

SODVPLGV LQWR ;FP WKH PRGLILHU SODVPLG S8)5 FDUU\LQJ ;FPO DQG ;FP:O PHWK\ODVH JHQHV ZDV XVHG WR LQFUHDVH WKH WUDQVIHU IUHTXHQF\ 'H )H\WHU DQG *DEULHO Ef 0DUNHUH[FKDQJH PXWDJHQHVLV 0DUNHUH[FKDQJH PXWDJHQHVLV RI VWUDLQ ;FP+ ZDV DFFRPSOLVKHG E\ LQWURGXFLQJ WKH GLVSODFHPHQW YHFWRU S8)5 LQWR ;FP+ WUDQVFRQMXJDQWV KDUERULQJ S8)5 DYUE7QJXV$f S8)5 DYUE7QJXV$f RU S8)5 DYU%,Q7QJXV$f GHULYDWLYHV 7KH SURFHGXUH ZDV WKH VDPH DV GHVFULEHG 6ZDUXS HW DO f H[FHSW WKDW WKH PRGLILHU SODVPLG S8)5 ZDV XVHG WR IDFLOLWDWH SODVPLG WUDQVIHU LQWR ;FP+ 3ODQW LQRFXODWLRQV &RWWRQ *RVV\SLXP KLUVXWXP /f OLQHV XVHG ZHUH $FDOD $Ff DQG LWV FRQJHQLH UHVLVWDQFH OLQHV %O % % E E %OQ DQG %,Q DV GHVFULEHG 6ZDUXS HW DO 'H )H\WHU HW DO f &RWWRQ SODQWV ZHUH JURZQ LQ WKH JUHHQKRXVH WUDQVIHUUHG WR JURZWK FKDPEHUV EHIRUH LQRFXODWLRQ DQG PDLQWDLQHG XQGHU FRQGLWLRQV DV GHVFULEHG 'H )H\WHU DQG *DEULHO Df %DFWHULDO VXVSHQVLRQV RI ; FDPSHVWULV SY PDOYDFHDUXP FIXPOf LQ VWHULOH WDS ZDWHU ZHUH JHQWO\ SUHVVXUH LQILOWUDWHG LQWR OHDYHV RI ZHHN ROG FRWWRQ SODQWV 3DWKRJHQLF V\PSWRPV ZHUH REVHUYHG SHULRGLFDOO\ GD\V DIWHU LQRFXODWLRQ $OO FLWUXV &LWUXV SDUDGLVL f'XQFDQf JUDSHIUXLWf DQG EHDQ 3KDVHROXV YXOJDULV f&DOLIRUQLD /LJKW 5HGff SODQWV ZHUH JURZQ XQGHU JUHHQKRXVH FRQGLWLRQV 3ODQW LQRFXODWLRQV LQYROYLQJ ; FLWUL RU SWK$ RU GHULYDWLYHV RI SWK$ ZHUH FDUULHG RXW LQ %/3 OHYHO FRQWDLQPHQW UHIHU WR )HGHUDO 5HJLVWHU 9RO 1R f DW WKH 'LYLVLRQ RI 3ODQW ,QGXVWU\ )ORULGD 'HSDUWPHQW RI $JULFXOWXUH *DLQHVYLOOH %DFWHULDO VXVSHQVLRQV ZHUH VWDQGDUGL]HG LQ VWHULOH WDS ZDWHU WR FIXPO DQG SUHVVXUH LQILOWUDWHG LQWR WKH

PAGE 23

DED[LDO OHDI VXUIDFH RI WKH SODQWV $OO SODQW LQRFXODWLRQV RQ FRWWRQ FLWUXV RU EHDQ ZHUH UHSHDWHG DW OHDVW WKUHH WLPHV %DFWHULDO JURZWK LQ SODQWD 7R GHWHUPLQH WKH JURZWK RI ; FDPSHVWULV SY PDOYDFHDUXP LQ WKH VXVFHSWLEOH FRWWRQ OLQH $F EDFWHULDO VXVSHQVLRQV ZHUH DGMXVWHG WR FIXPO DQG LQRFXODWHG LQWR FRWWRQ OHDYHV E\ SUHVVXUH LQILOWUDWLRQ /HDI GLVFV FPf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, VWHULOH WDS ZDWHU ZDV GLVSHQVHG RQWR WKH ZDWHUVRDNHG OHDI VXUIDFH DQG VSUHDG WR DQ DUHD RI DSSUR[LPDWHO\ FP $IWHU PL[LQJ ZLWK WKH EDFWHULD DQG VOLPH RQ WKH OHDI VXUIDFH WKH EDFWHULDO VXVSHQVLRQ ZDV FROOHFWHG ZLWK D SLSHW (DFK FP ZDWHUVRDNHG OHDI DUHD ZDV ZDVKHG WHQ WLPHV DQG D WRWDO RI PO RI EDFWHULDO VXVSHQVLRQ ZDV FROOHFWHG 7R TXDQWLI\ WKH EDFWHULD UHPDLQLQJ LQVLGH WKH OHDI OHDI GLVFV FPf ZHUH WDNHQ ZLWK D VWHULOL]HG FRUN ERUHU WKHQ PDFHUDWHG LQ PO VWHULOH WDS ZDWHU 9LDEOH FRXQWV ZHUH GHWHUPLQHG E\ VHULDO GLOXWLRQV RQ SODWHV FRQWDLQLQJ DSSURSULDWH DQWLELRWLFV 'DWD UHSRUWHG LQ WKH UHVXOWV H[WHUQDO DQG WRWDO FRXQWVf ZHUH PHDQ DQG VWDQGDUG HUURU RI VL[ UHSOLFDWHV IURP WZR H[SHULPHQWV

PAGE 24

5HVXOWV 3OHLRWURSLF SDWKRJHQLFLW\ IXQFWLRQV RI DYUE ; FDPSHVWULV SY PDOYDFHDUXP VWUDLQ ;FP+ FDUULHV DYU% DYUE DYUE DYU%,Q DYU%O2O DQG DYU% RQ D VLQJOH SODVPLG S;FP+ DQG HOLFLWV DQ +5 RQ FRWWRQ OLQHV FDUU\LQJ DQ\ RQH RI PDQ\ GLIIHUHQW UHVLVWDQFH "f JHQHV 'H )H\WHU HW DO f 6WUDLQ ;FP+ DQG D VSRQWDQHRXV ULIDP\FLQUHVLVWDQW GHULYDWLYH RI ;FP+ ;FP+ ZHUH YLUXOHQW RQ VXVFHSWLEOH FRWWRQ OLQH $FDOD $Ff DQG ERWK HOLFLWHG VHYHUH ZDWHUVRDNLQJ DQG QHFURVLV DVVRFLDWHG ZLWK JURZWK LQ SODQWD ;FP+ RQ $F LV VKRZQ LQ )LJV t f 0XWDWLRQV RI DYUE DYUE DQG DYU%,Q ZHUH LQGLYLGXDOO\ JHQHUDWHG LQ ;FP+ E\ PDUNHUH[FKDQJH PXWDJHQHVLV DQG HDFK PXWDQW ZDV FRQILUPHG WR FDUU\ D VLQJOH 7QJXV$ LQVHUWLRQ LQ WKH DSSURSULDWH '1$ IUDJPHQW E\ 6RXWKHUQ EORW K\EULGL]DWLRQ UHIHU )LJ VRPH GDWD QRW VKRZQf $V SUHGLFWHG E\ JHQHIRUJHQH WKHRU\ PDUNHUH[FKDQJH PXWDQWV ;FP+ DYUE7QJXV$f ;FP+ DYUE7QJXV$f DQG ;FP+ DYU%,Q7U?JXV$f JDLQHG YLUXOHQFH RQ FRWWRQ OLQHV ZLWK WKH UHVLVWDQFH JHQHV E E DQG %OQ UHVSHFWLYHO\ 7DEOH f 3ODVPLGV S8)5 DYUEf S8)5 DYUEf DQG S8)5 DYU%,Qf ZHUH DEOH WR IXOO\ FRPSOHPHQW WKH VSHFLILF DYLUXOHQFH GHIHFWV RI ;FP+ ;FP+ DQG ;FP+ UHVSHFWLYHO\ 1HLWKHU ;FP+ QRU ;FP+ VKRZHG DQ\ FKDQJH LQ ZDWHUVRDNLQJ DELOLW\ DV FRPSDUHG ZLWK ;FP+ +RZHYHU PDUNHUH[FKDQJH PXWDQW ;FP+ QRW RQO\ ORVW DYLUXOHQFH RQ D FRWWRQ OLQH ZLWK WKH E UHVLVWDQFH JHQH EXW DOVR HOLFLWHG VLJQLILFDQWO\ OHVV ZDWHUVRDNLQJ DQG QHFURVLV RQ VXVFHSWLEOH FRWWRQ OLQH $F 7DEOH DQG )LJ $

PAGE 25

)LJXUH :DWHUVRDNHG OHVLRQV FDXVHG E\ ; FDPSHVWULV SY PDOYDFHDUXP VWUDLQV RQ WKH VXVFHSWLEOH FRWWRQ OLQH $FDOD $ /HDI LQRFXODWHG ILYH GD\V SUHYLRXVO\ ZLWK ;FP+ ;FP+ ;FP+S8)5 ;FP+ % /HDI LQRFXODWHG VHYHQ GD\V SUHYLRXVO\ ZLWK ;FP+ ;FP+ & /HDI LQRFXODWHG ILYH GD\V SUHYLRXVO\ ZLWK ;FP+ ;FPOS8)5 ;FPOS8)5 ;FPO

PAGE 27

)LJXUH *URZWK NLQHWLFV RI ; FDPSHVWULV SY PDOYDFHDUXP VWUDLQV RQ WKH VXVFHSWLEOH FRWWRQ OLQH $FDOD $ f ;FP+ D ;FP+ ‘;FP+ 2 ;FPO % f ;FP+ Y ;FPOS8)5 ;FPOS8)5 2 ;FPO

PAGE 28

'D\V SRVWLQRFXODWLRQ

PAGE 29

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

PAGE 30

7DEOH 3KHQRW\SHV RI PDUNHUH[FKDQJH PXWDQWV RI ;DQWKRPRQDV FDPSHVWULV SY PDOYDFHDUXP VWUDLQ ;FP+ RQ FRWWRQ FXOWLYDU $FDOD $Ff DQG FRQJHQLH OLQHV ;FP VWUDLQV $F $FE $FE $F%,Q ;FP+ B B f§ ;FP+ DYUE7QJXV$f ;FP+ DYUE7QJXV$f ;FP+ DYU%,Q7QJXV$f D$FE $FE DQG $F%,Q DUH FRQJHQLH OLQHV RI $F FRQWDLQLQJ UHVLVWDQFH JHQH E E DQG %OQ UHVSHFWLYHO\ VWURQJ ZDWHUVRDNLQJ V\PSWRPV ZHDN ZDWHUVRDNLQJ V\PSWRPV K\SHUVHQVLWLYH UHVSRQVH

PAGE 31

t ,%f 3ODVPLG S8)5 ZDV DEOH WR IXOO\ FRPSOHPHQW WKH SDWKRJHQLFLW\ GHIHFWVf RI ;FP+ RQ VXVFHSWLEOH FRWWRQ SODQWV LQ DGGLWLRQ WR FRPSOHPHQWLQJ WKH VSHFLILF DYLUXOHQFH GHIHFW 'HVSLWH LWV UHGXFHG DELOLW\ WR HOLFLW SDWKRJHQLF V\PSWRPV RQ VXVFHSWLEOH FRWWRQ OLQHV PXWDQW ;FP+ H[KLELWHG WKH VDPH JURZWK UDWH DQG \LHOG DV WKDW RI WKH ZLOG W\SH ;FP+ DQG ;FP+ RQ $F )LJ $f ; FDPSHVWULV SY PDOYDFHDUXP VWUDLQ ;FPO FDUULHV QR NQRZQ DYU JHQHV 'H )H\WHU DQG *DEULHO Df FDXVHG OHVV ZDWHUVRDNLQJ RQ $F WKDQ ;FP+ DQG H[KLELWHG D ORZHU JURZWK UDWH RQ $F WKDQ ;FP+ ,QWURGXFWLRQ RI S8)5 DYUEf LQWR ;FPO FRQIHUUHG LQFUHDVHG ZDWHUVRDNLQJ DELOLW\ )LJ &f EXW JURZWK UDWH DQG \LHOG LQ SODQWD RI WKH WUDQVFRQMXJDQW FRQWDLQLQJ S8)5 ZHUH QRW LQFUHDVHG LQ FRPSDULVRQ ZLWK WKDW RI ;FPO RU WKH WUDQVFRQMXJDQW ;FPOS8)5 DYU%,Qf )LJ %f 7KHUHIRUH S8)5 DIIHFWHG SDWKRJHQLF V\PSWRPV LQ ERWK ;FP+ DQG ;FPO EXW QRW EDFWHULDO JURZWK UDWH RU \LHOG LQ SODQWD RI HLWKHU VWUDLQ 6WUDLQV FRQWDLQLQJ DYUE HJ ;FP+ DQG ;FPOS8)5f HOLFLWHG PRUH VHYHUH ZDWHUVRDNLQJ DQG QHFURVLV DQG ZHUH DVVRFLDWHG ZLWK PXFK PRUH VOLPH RR]LQJ IURP WKH ZDWHUVRDNHG DUHDV LQ FRPSDULVRQ ZLWK VWUDLQV ODFNLQJ DYUE HJ ;FP+ DQG ;FPO )LJ f 7KH SHDN QXPEHU RI WRWDO FRORQ\ IRUPLQJ XQLWV FIXf SHU VTXDUH FHQWLPHWHU RI ZDWHUVRDNHG OHDI LQRFXODWHG ZLWK ;FP+ RU ;FP+ ZHUH EDVLFDOO\ WKH VDPH s [ FIXFP YV s [ FIXFPf +RZHYHU b RI WKH ;FP+ EDFWHULD SUHVHQW LQ WKH OHVLRQ s [ FIXFPf ZHUH UHOHDVHG RQWR WKH VXUIDFH RI WKH OHDI ZKHUHDV RQO\ b RI WKH ;FP+ EDFWHULD s [ FIXFPf ZHUH UHOHDVHG RQWR WKH VXUIDFH RI WKH OHDI

PAGE 32

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f 8QLTXH 6WXO DQG +LQGL VLWHV DUH DOVR IRXQG DW WKH VDPH UHODWLYH SRVLWLRQV LQ SWK$ DQG WKH '1$ VHTXHQFHV RI DYUE DQG SWK$ DUH LGHQWLFDO LQ WKH IODQNLQJ UHJLRQV IURP WKH ES WDQGHP UHSHDWV WR WKHVH UHVWULFWLRQ VLWHV UHIHU WR &KDSWHU f &KLPHULF JHQHV ZHUH FRQVWUXFWHG E\ VZDSSLQJ WKH 6WXO+LQGL IUDJPHQWV FRQWDLQLQJ WKH LQWHUQDO ES UHSHDWHG UHJLRQV EHWZHHQ SWK$ DQG DYUE $W OHDVW WKUHH LQGLYLGXDO FORQHV RI ERWK FKLPHULF JHQHV ZHUH LQWURGXFHG LQWR D PXWDQW ; FLWUL VWUDLQ % SWK$7QJXV$f DQG ZLOG W\SH VWUDLQV RI WKH IROORZLQJ SDWKRJHQV ; SKDVHROL VWUDLQ KRVW UDQJH RQ EHDQf ; FDPSHVWULV SY FLWUXPHOR VWUDLQ KRVW UDQJH RQ EHDQ DQG FLWUXVf ; FDPSHVWULV SY DOIDOIDH VWUDLQ .; KRVW UDQJH RQ DOIDOID EHDQ DQG FLWUXVf DQG ;FPO 6WUDLQV DQG WUDQVFRQMXJDQWV ZHUH LQRFXODWHG RQ FLWUXV EHDQ DQG FRWWRQ OHDYHV $V VKRZQ LQ )LJXUH S8)< FDUU\LQJ D FKLPHULF JHQH FRQWDLQLQJ f DQG f HQGV RI DYUE DQG WKH LQWHUQDO UHSHDWV RI SWK$ FRPSOHPHQWHG WKH PXWDQW % WR IXOO YLUXOHQFH RQ FLWUXV 7KH FLWUXV FDQNHU V\PSWRPV FDXVHG E\ %S8)< ZHUH

PAGE 33

)LJXUH 3KHQRW\SHV RI ; FLWUL ZLOG W\SH DQG PXWDQW VWUDLQV RQ D JUDSHIUXLW &LWUXV SDURGLVW FYf'XQFDQff OHDI ZLOGW\SH VWUDLQ PDUNHUH[FKDQJH PXWDQW RI % %S=LW %S8)< %S8)5 %S8)<

PAGE 34

LQGLVWLQJXLVKDEOH IURP WKH V\PSWRPV FDXVHG E\ %S=LW SWK$f :KHQ S8)< ZDV SUHVHQW LQ RU .; ERWK RI ZKLFK FDXVH ZDWHUVRDNHG OHDI VSRWV RQ FLWUXV FDQNHU V\PSWRPV ZHUH LQGXFHG E\ WKH WUDQVFRQMXJDQWV WKDW ZHUH LQGLVWLQJXLVKDEOH IURP WKRVH FRQIHUUHG E\ S=LW LQ WKH VDPH VWUDLQV $OVR OLNH S=LW 6ZDUXS HW DO f S8)< FRQIHUUHG DYLUXOHQFH RQ EHDQ WR .; DQG DQG RQ FRWWRQ WR ;FPO $ FKLPHULF JHQH FRQWDLQLQJ WKH f DQG f HQGV RI SWK$ DQG WKH LQWHUQDO UHSHDWV RI DYUE RQ S8)< HQKDQFHG WKH DELOLW\ RI ;FPO DQG ;FP+ WR ZDWHUVRDN FRWWRQ DQG FDXVH QHFURVLV 7KH ZDWHUVRDNLQJ V\PSWRPV FDXVHG E\ ;FPOS8)< ZHUH LQGLVWLQJXLVKDEOH IURP WKRVH FDXVHG E\ ;FPOS8)5 7KLV FKLPHULF JHQH RQ S8)< EHKDYHG OLNH DYUE RQ S8)5 DQG FRQIHUUHG QR GHWHFWDEOH V\PSWRPV WR % RU RQ FLWUXV RU EHDQ &XOWLYDUVSHFLILF DYLUXOHQFH LV GHWHUPLQHG E\ ES UHSHDWV (DFK RI WKH VHYHQ DYUSWK JHQHV DYU% DYUE DYUE DYU%,Q DYU%O2O DYU% DQG SWK$f H[KLELWV XQLTXH DYLUXOHQFH VSHFLILFLW\ LQ ;FPO RQ FRWWRQ UHVLVWDQFH OLQHV GLIIHULQJ E\ VLQJOH 5 JHQHV 'H )H\WHU HW DO 6ZDUXS HW DO f (DFK RI WKHVH VHYHQ JHQHV FRQWDLQV WZR %DP8O VLWHV RQH QHDU WKH f HQG DQG RQH QHDU WKH f HQG RI HDFK JHQH DQG VLQJOH 6WXO DQG +LQGL VLWHV IODQNLQJ WKH ES GLUHFW UHSHDW UHJLRQ 'H )H\WHU HW DO 6ZDUXS HW DO f 7R ORFDOL]H WKH UHJLRQ WKDW GHWHUPLQHV WKH VSHFLILFLW\ RI WKH UHDFWLRQV D VHULHV RI FKLPHULF JHQHV ZHUH FRQVWUXFWHG E\ VZDSSLQJ %DP5O DQG 6WXO+LQGO LQWHUQDO IUDJPHQWV DPRQJ WKHVH VHYHQ PHPEHUV RI WKH JHQH IDPLO\ 2QH WR WKUHH LQGLYLGXDO FORQHV RI HDFK FKLPHULF JHQH ZHUH LQWURGXFHG LQWR ;FPO DQG

PAGE 35

7DEOH $YLUXOHQFH VSHFLILFLW\ RI =WRZ+,IUDJPHQW VZDSSHG FKLPHULF JHQHV LQ ;FPO RQ FRWWRQ 3ODVPLG f DQG f HQG &KLPHULF *HQH %DP+O IUDJPHQW $YLUXOHQFH 6SHFLILFLW\ S8)5 DYUE DYU% DYU% S8)5 DYUE DYUE DYUE S8)5 DYUE DYU%,Q DYU%,Q S8)5 DYUE DYU%O2O DYU%O2O S8)5 DZE DYU% DYU% S8)5 DYU%,Q DYU% DYU% S8)5 DYU%,Q DYUE DYUE S8)5 DYU%,Q DYUE DYUE S8)5 DYU%,Q DYU%O2O DYU%O2O S8)5 DYU%,Q DYU% DYU% 1RWH &KLPHULF JHQHV ZHUH LQWURGXFHG LQWR ;FPO DQG WHVWHG RQ FRWWRQ FY $F DQG LWV FRQJHQLH UHVLVWDQFH OLQHV $F%O $F% $F% $FE $FE $F%,Q DQG $F%,Q IRU DYLUXOHQFH VSHFLILFLW\

PAGE 36

7DEOH $YLUXOHQFH VSHFLILFLW\ RI 607PF,,IUDJPHQW VZDSSHG FKLPHULF JHQHV LQ ;FPO RQ FRWWRQ 3ODVPLG &KLPHULF *HQH $YLUXOHQFH 6SHFLILFLW\ f DQG f UHJLRQ 6WXOO +LQGL IUDJPHQW S8)5 DYU%,Q DYU% DYU%r S8)5 DYU%,Q DYUE DYUEr S8)5 DYU%,Q DYUE DYUEr S8)5 DYU%,Q DYU%O2O DYU%O2O S8)5 DYU%,Q DYU% DYU% S8)5 DYUE DYU% DYU% S8)5 DYUE DYUE DYUE S8)5 DYUE DYU%,Q DYU%,Qr S8)5 DYUE DYU%O2O DYU%O2O S8)5 DYUE DYU% 1' S8)< SWK$ DYUE DYUE S8)< ODF=DYUE SWK$ SWK$? 1RWH &KLPHULF JHQHV ZHUH LQWURGXFHG LQWR ;FPO DQG WHVWHG RQ FRWWRQ FY $F DQG LWV FRQJHQLH UHVLVWDQFH OLQHV $F%O $F% $F% $FE $FE $F%,Q DQG $F%,Q IRU DYLUXOHQFH VSHFLILFLW\ r ZHDN DYLUXOHQFH VWURQJ DYLUXOHQFH 1' QR DYLUXOHQFH GHWHFWHG

PAGE 37

LQRFXODWHG RQ FRWWRQ FXOWLYDU $F DQG LWV FRQJHQLH UHVLVWDQFH OLQHV HDFK GLIIHULQJ E\ D VLQJOH 5 JHQH %O % % E E %OQ RU %,Qf 7KH DYLUXOHQFH VSHFLILFLWLHV RI WKH VHYHQ DYUSWK JHQHV ZHUH ILUVW ORFDOL]HG ZLWKLQ WKH LQWHUQDO %DP:O IUDJPHQWV 7DEOH f DQG WKHQ ZHUH IXUWKHU ORFDOL]HG ZLWKLQ WKH LQWHUQDO 6WXOO +LQGL IUDJPHQWV 7DEOH f ,Q DOO FDVHV WKH DYLUXOHQFH VSHFLILFLW\ RI D JLYHQ JHQH ZDV GHWHUPLQHG LQVLGH WKH 6WXOO +LQGL WDQGHP UHSHDWf UHJLRQ 'LVFXVVLRQ 3ODQW SDWKRORJLVWV KDYH ORQJ EHHQ SX]]OHG E\ WKH SUHVHQFH RI DYLUXOHQFH JHQHV LQ SDWKRJHQV 7KHVH JHQHV DFW DV QHJDWLYH IDFWRUV WR OLPLW YLUXOHQFH DQG LQ PRVW FDVHV GR QRW DSSHDU WR SURYLGH VHOHFWLYH DGYDQWDJH WR WKH SDWKRJHQV (OOLQJERH *DEULHO .HHQ DQG 6WDVNDZLF] f 5DUH H[FHSWLRQV KDYH EHHQ UHSRUWHG )RU H[DPSOH DYU%V IURP ; FDPSHVWULV SY YHVLFDWRULD LV UHTXLUHG IRU RSWLPDO JURZWK LQ SODQWD .HDUQH\ DQG 6WDVNDZLF] f %RWK DYUE DQG DYUE IURP ; FDPSHVWULV SY PDOYDFHDUXP VWUDLQ ;FP+ DUH NQRZQ WR HQKDQFH WKH ZDWHUVRDNLQJ DELOLW\ RI ; FDPSHVWULV SY PDOYDFHDUXP VWUDLQ ;FPO RQ FRWWRQ 'H )H\WHU DQG *DEULHO Df ,Q WKLV VWXG\ ZH GHPRQVWUDWHG WKDW WKH DELOLW\ RI VWUDLQ ;FP+ WR FDXVH VWURQJ ZDWHUVRDNLQJ DQG QHFURVLV RQ FRWWRQ UHTXLUHV WKH SUHVHQFH RI DYUE EXW QRW DYUE $OWKRXJK DYUE LQFUHDVHG V\PSWRP HOLFLWDWLRQ E\ ERWK ;FPO DQG ;FP+ QHLWKHU WKHLU JURZWK UDWHV QRU WKHLU PD[LPXP EDFWHULDO FRXQWV SHU VTXDUH FHQWLPHWHU RI OHDI ZHUH DIIHFWHG E\ WKH SUHVHQFH RU DEVHQFH RI DYUE 7KHUHIRUH DYUE IXQFWLRQV DV D

PAGE 38

SDWKRJHQLFLW\ JHQH DQG LQFUHDVHV V\PSWRPV RI FRWWRQ EOLJKW EXW ZLWKRXW HOLFLWLQJ DQ +5 RQ FRWWRQ OLQHV ODFNLQJ E 7KH VWURQJ ZDWHUVRDNLQJ DELOLW\ FRQIHUUHG E\ DYUE ZDV FRUUHODWHG ZLWK PXFK KLJKHU IROGf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f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f ,Q WKH SUHVHQW VWXG\ ZKHQ DYUE ZDV WUDQVIHUUHG WR ; FLWUL % ; FDPSHVWULV SY FLWUXPHOR DQG ; SKDVHROL LW FRQIHUUHG QR GHWHFWDEOH HIIHFW ZKHQ WKHVH VWUDLQV ZHUH LQRFXODWHG RQWR WKHLU UHVSHFWLYH KRVWV 7KHUHIRUH WKH SDWKRJHQLFLW\ IXQFWLRQV RI DYUE DQG SWK$ DUH KRVW VSHFLILF ,I UHOHDVH RI WKH SDWKRJHQ WR WKH OHDI VXUIDFH LV KRVW VSHFLILF DQG LW FRQWULEXWHV WR WKH HFRORJLFDO ILWQHVV RI WKH SDWKRJHQ DV ZH SURSRVH WKHQ DYUE DQG SWK$ IXQFWLRQ WR

PAGE 39

GHWHUPLQH KRVW UDQJH )XUWKHUPRUH WKH DYLUXOHQFH FRQIHUUHG E\ SWK$ DSSHDUHG WR EH JUDWXLWRXV LQ WHUPV RI UHVWULFWLQJ KRVW UDQJH 6ZDUXS HW DO f DQG LQ WKH SUHVHQW VWXG\ DYUE IDLOHG WR FRQIHU DYLUXOHQFH WR WKUHH RWKHU SDWKRJHQV 7KHUHIRUH LI DYUE DQG SWK$ KHOS GHWHUPLQH KRVW UDQJH LW LV QRW EHFDXVH RI WKHLU IXQFWLRQ DV DYU JHQHV ,Q D IRUPDO JHQHWLF VHQVH WKHVH SOHLRWURSLF DYUSWK JHQHV UHVHPEOH VRPH 5KL]RELXP KRVW VSHFLILF QRGXODWLRQ KVQf JHQHV ZKLFK DUH UHTXLUHG IRU KRVW UDQJH RQ VRPH KRVWV EXW ZKLFK FDQ DOVR FRQIHU DYLUXOHQFH ZKHQ WUDQVIHUUHG WR RWKHU 5KL]RELXP VWUDLQV ZLWK D GLIIHUHQW KRVW UDQJH 'HEHOOH HW DO )DXFKHU HW DO /HZLV+HQGHUVRQ DQG 'MRUGMHYLF f %\ VZDSSLQJ WKH WDQGHPO\ UHSHDWHG UHJLRQV WKH DYLUXOHQFH VSHFLILFLWLHV RI DYU% DYUE DYUE DYU%,Q DYU%O2O DYU% DQG SWK$ ZHUH VKRZQ WR EH GHWHUPLQHG E\ WKH ES WDQGHP UHSHDWV 7KHVH UHVXOWV DUH FRQVLVWHQW ZLWK DQG H[WHQG WKH ILQGLQJV RI +HUEHUV HWDO f ZKR VKRZHG WKDW WKH DYLUXOHQFH VSHFLILFLW\ RI DYU%V LV GHWHUPLQHG E\ WKH ES UHSHDWV RI WKDW JHQH ,Q DGGLWLRQ WKH VZDSSLQJ H[SHULPHQWV FOHDUO\ GHPRQVWUDWHG WKDW WKH SDWKRJHQLFLW\ IXQFWLRQV RI SWK$ DQG DYUE ZHUH GLVWLQFW FDQNHUV YV ZDWHUVRDNLQJf KRVW VSHFLILF DQG LQ ERWK FDVHV WKH SDWKRJHQLF VSHFLILFLW\ ZDV GHWHUPLQHG E\ WKHLU ES WDQGHPO\ UHSHDWHG UHJLRQV )XUWKHUPRUH WKH XVH RI FKLPHULF JHQHV DOVR UXOHG RXW WKH SRVVLELOLW\ WKDW WKH KRVWVSHFLILF SDWKRJHQLFLW\ RQ FRWWRQ DQG FLWUXV ZHUH UHVXOW RI DGGLWLRQDO XQLGHQWLILHG SDWKRJHQLFLW\ JHQHV HQFRGHG RQ WKH SODVPLGV XVHG S8)5 DQG S=LWf ([WHQVLYH GHOHWLRQ DQDO\VHV RI ERWK WKH f DQG f HQGV RI WKH S;FP+ DYU JHQHV VKRZHG WKDW DOO VHFWLRQV RI WKHVH JHQHV DUH UHTXLUHG IRU DYLUXOHQFH DFWLYLW\ 'H )H\WHU HW

PAGE 40

DO f )URP WKH SUHVHQW VWXG\ DOO VHFWLRQV RI SWK$ DQG DYUE ZHUH UHTXLUHG WR FRQIHU SDWKRJHQLFLW\ IXQFWLRQV +RZHYHU WKH f DQG f HQGV RI WKH JHQHV RXWVLGH RI WKH UHSHDWV DSSHDUHG WR EH LVRIXQFWLRQDO DPRQJ PHPEHUV RI WKH JHQH IDPLO\ %HVLGHV WKH PHPEHUV RI WKLV ;DQWKRPRQDV DYUSWK JHQH IDPLO\ LQWHUQDO WDQGHP UHSHDWV DUH IRXQG LQ VRPH SDWKRJHQLFLW\ JHQHV RI DQLPDO SDWKRJHQV ([DPSOHV LQFOXGH WKH RXWHU PHPEUDQH SURWHLQ $ JHQH RPS$f RI VRPH 5LFNHWWVLD VSHFLHV $QGHUVRQ HW DO *LOPRUH f LQWHPDOLQ JHQH LQLf RI /LVWHULD PRQRF\WRJHQHV *DLOODUG HW DO f WR[LQ $ JHQH RI &ORVWULGLXP GLIILFLOH 'RYH HW DO f DQG 0 SURWHLQ JHQHV HPPf RI 6WUHSWRFRFFXV +ROOLQJVKHDG HW DO f 7DQGHP UHSHDWV ZHUH DOVR IRXQG LQ PDQ\ JHQHV IURP SURWR]RDQ DQG PHWD]RDQ SDUDVLWHV VXFK DV 3ODVPRGLXP 0F&RQNH\ HW DO f 7U\SDQRVRPD +RIW HW DO f /HLVKPDQLD :DOOLV DQG 0F0DVWHU f DQG 0HORLGRJ\QH 2NLPRWR HW DO f *HQHV IURP 7U\SDQRVRPD FUX]L WKH SURWR]RDQ DJHQW RI $PHULFDQ WU\SDQRVRPLDVLVf DQG PLWRFKRQGULD RI 0HORLGRJ\QH MDYDQLFD SODQW URRW NQRW QHPDWRGHf FRQWDLQ ES WDQGHP UHSHDWV 0RVW RI WKHVH JHQHV HQFRGH VXUIDFH SURWHLQV DQG WKH GLVWLQFWLYH DUUDQJHPHQW RI WKH WDQGHP UHSHDWV LQ WKHVH JHQHV DUH WKRXJKW WR HQFRGH D SURWHFWLYH VWUDLQVSHFLILF FRQIRUPDWLRQDO HSLWRSH IRU HYDVLRQ RI KRVW LPPXQLW\ *LOPRUH +RIW HW DO 0F&RQNH\ HW DO f %\ FRQWUDVW $YU%V LV PDLQO\ ORFDWHG LQ F\WRVRO .QRRS HW DO %URZQ HW DO f DQG LW LV QRW FOHDU KRZ LW PD\ LQWHUDFW ZLWK WKH SODQW FHOO 6RXWKHUQ K\EULGL]DWLRQ KDV VKRZQ WKDW SRWHQWLDO PHPEHUV RI WKH DYUSWK JHQH IDPLO\ H[LVW RIWHQ LQ PXOWLSOH FRSLHV LQ RI ;DQWKRPRQDV VSHFLHV RU SDWKRYDUV H[DPLQHG %RQDV HW DO 6ZDUXS HW DO 'H )H\WHU HW DO f ,Q DOO ; FLWUL

PAGE 41

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f WKHUH PD\ EH QR SOHLRWURSLF SDWKRJHQLFLW\ IXQFWLRQ IRU H[DPSOH WKHUH LV QR HYLGHQFH RI D SDWKRJHQLFLW\ IXQFWLRQ RI DYU%V DYU%V DQG DYU%V3 %RQDV HW DO &DQWHURV HW DO f 7KH PHFKDQLVPVf E\ ZKLFK OHDI VSRWWLQJ SDWKRJHQV HOLFLW ZDWHUVRDNLQJ DQG QHFURVLV LV XQNQRZQ EXW DSSHDUV WR LQYROYH ERWK GDPDJH WR OHDI FHOO PHPEUDQHV ZLWKRXW HOLFLWLQJ DQ +5f DQG SURGXFWLRQ RI H[WUDFHOOXODU SRO\VDFFKDULGH (36f 7KH (36 GRHV QRW DSSHDU WR EH LQYROYHG LQ VXSSUHVVLQJ D SRWHQWLDO +5 VLQFH SURGXFWLRQ OHYHOV RI (36 E\ ; FDPSHVWULV SY PDOYDFHDUXP DUH VLPLODU LQ ERWK VXVFHSWLEOH DQG UHVLVWDQW FRWWRQ OLQHV 3LHUFH HW DO f ,QVWHDG PXWDWLRQDO DQDO\VHV RI ;DQWKRPRQDV (36 ELRV\QWKHWLF JHQHV DQG LQRFXODWLRQV ZLWK SXULILHG (36 KDYH VKRZQ WKDW WKH (36 FRQWULEXWHV WR ZDWHUVRDNLQJ E\ WUDSSLQJ ZDWHU DQG QXWULHQWV LQ WKH LQWHUFHOOXODU VSDFHV DIWHU WKH\ DUH

PAGE 42

UHOHDVHG UHYLHZHG E\ /HLJK DQG &RSOLQ f &RSOLQ HW DO f KDYH SURSRVHG WKDW ZDWHUVRDNLQJ DQG SDWKRJHQLFLW\ RI (UZLQLD VWHZDUWLL LQYROYHV WKH (36 SOXV D FHOO OHDNDJH IDFWRU HQFRGHG E\ ZWV ZDWHUVRDNLQJf JHQHV 7KLV K\SRWKHVLV PD\ ZHOO DSSO\ WR ; FDPSHVWULV SY PDOYDFHDUXP ZLWK DYUE HQFRGLQJ D FHOO OHDNDJH IDFWRU /LNH VRPH ZWV JHQHV RI ( VWHZDUWLL DYUE ZDV QRW UHTXLUHG IRU EDFWHULDO JURZWK LQ SODQWD EXW VWURQJO\ DIIHFWHG ZDWHUVRDNLQJ RQ LWV KRVW 7KH LRQFKDQQHO GHIHQVH PRGHO RI WKH JHQHIRUJHQH K\SRWKHVLV LQYRNHV D FHOO OHDNDJH IDFWRU DV WKH SURGXFW RI DQ DYU JHQH *DEULHO HW DO *DEULHO DQG 5ROIH f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f ,QFUHDVHG IOXLGLW\ RU DPRXQWV RI (36 PD\ LQFUHDVH EDFWHULD H[XGLQJ RQWR WKH OHDI VXUIDFH WKURXJK VWRPDWD 7KLHUV DQG %ODQN f $QRWKHU SRVVLEOH IXQFWLRQ LV WKDW WKH LQFUHDVHG OHYHOV RI QHFURVLV LQGXFHG E\ DYUE PD\ VHUYH WR FROODSVH WKH SDOLVDGH OD\HU DQG SK\VLFDOO\ VTXHH]H PRUH EDFWHULD RQWR WKH OHDI VXUIDFH

PAGE 43

&+$37(5 ,1$&7,9$7,21 2) 08/7,3/( $9,58/(1&( *(1(6 ,1 ;$17+2021$6 &$03(6756 39 0$/9$&($580 &5($7(' $ 1213$7+2*(1,& (1'23+<7( 2) &27721 ,QWURGXFWLRQ $YLUXOHQFH DYUf JHQHV DFW DV QHJDWLYH IDFWRUV WR FRQIHU DYLUXOHQFH DQG WR OLPLW KRVW UDQJH RI WKH SDWKRJHQ RQ KRVW SODQWV FDUU\LQJ VSHFLILF UHVLVWDQFH 5f JHQH $V D UHVXOW PRVW DYU JHQHV UHSRUWHG WR GDWH DSSHDU WR EH JUDWXLWRXV DQG RI QR REYLRXV VHOHFWLYH YDOXH WR WKH PLFURELDO SODQW SDWKRJHQV *DEULHO HW DO f +RZHYHU D IHZ RI WKHP KDYH GHPRQVWUDWHG SOHLRWURSLF IXQFWLRQVf )RU H[DPSOH DYU%V LV UHTXLUHG IRU IXOO YLUXOHQFH RI ;DQWKRPRQDV FDPSHVWULV SY YHVLFDWRULD RQ VXVFHSWLEOH KRVWV .HDUQH\ DQG 6WDVNDZLF] f 0RUH UHFHQWO\ SWK$ RI ; FLWUL DQG DYUE RI ; FDPSHVWULV SY PDOYDFHDUXP ZHUH DOVR VKRZQ WR KDYH KRVWVSHFLILF SDWKRJHQLFLW\ IXQFWLRQ LQ DGGLWLRQ WR DYLUXOHQFH IXQFWLRQ 6ZDPS HW DO &KDSWHU f *HQH SWK$ LV HVVHQWLDO IRU ; FLWUL VWUDLQ WR LQGXFH K\SHUSODVWLF FDQNHUV RQ FLWUXV VSHFLILFDOO\ DQG FRQIHUV WKDW IXQFWLRQ WR VHYHUDO RWKHU ; FDPSHVWULV SDWKRYDUV 6LPLODUO\ DYUE LV LPSRUWDQW IRU ; FDPSHVWULV SY PDOYDFHDUXP VWUDLQ ;FP+ WR LQGXFH ZDWHUVRDNLQJ RQ FRWWRQ VSHFLILFDOO\ DQG FRQIHUV WKDW IXQFWLRQ WR RWKHU ; FDPSHVWULV SY PDOYDFHDUXP VWUDLQV %RWK SWK$ DQG DYUE DSSHDU WR FRQWULEXWH WR WKH ILWQHVV RI WKHLU UHVSHFWLYH SDWKRJHQV RQ WKHLU KRVWV E\

PAGE 44

LQFUHDVLQJ UHOHDVH RI WKH SDWKRJHQV IURP SODQW OHDYHV WKXV DIIHFWLQJ KRVW UDQJH WKURXJK LQFUHDVHG GLVSHUVDO %RWK SWK$ DQG DYUE DUH PHPEHUV RI D ODUJH ;DQWKRPRQDV DYLUXOHQFH DYUf DQG SDWKRJHQLFLW\ SWKf JHQH IDPLO\ DQG ERWK FRQIHU JHQHIRUJHQH DYLUXOHQFH WR ; FDPSHVWULV SY PDOYDFHDUXP RQ FRWWRQ 6ZDUXS HW DO 'H )H\WHU HW DO f 2WKHU PHPEHUV RI WKH DYUSWK JHQH IDPLO\ LQFOXGH DYU% DYUE DYU%,Q DYU%O2O DQG DYU% RI ; FDPSHVWULV SY PDOYDFHDUXP 'H )H\WHU DQG *DEULHO D 'H )H\WHU HW DO f DYU%V DYU%V DQG DYU%V3 RI ; FDPSHVWULV SY YHVLFDWRULD %RQDV HW DO &DQWHURV HW DO f DQG DYU;D DYU;D DQG DYU;DO2 RI ; RU\]DH +RSNLQV HW DO f )DPLO\ PHPEHUV VHTXHQFHG WR GDWH DUH bb LGHQWLFDO WR HDFK RWKHU %RQDV HW DO +RSNLQV HW DO 'H )H\WHU HW DO f 0RVW PHPEHUV RI WKLV JHQH IDPLO\ ZHUH LVRODWHG DV DYLUXOHQFH JHQHV ZLWK QR HYLGHQFH RI SDWKRJHQLFLW\ IXQFWLRQ ;DQWKRPRQDV FDPSHVWULV SY PDOYDFHDUXP 6PLWKf '\H LV WKH FDXVDO DJHQW RI EDFWHULDO EOLJKW RI FRWWRQ DQ HFRQRPLFDOO\ LPSRUWDQW ZRUOGZLGH GLVHDVH 6L[ PHPEHUV RI WKLV JHQH IDPLO\ ZHUH SUHYLRXVO\ LVRODWHG IURP S;FP+ D NE SODVPLG RI ; FDPSHVWULV SY PDOYDFHDUXP VWUDLQ ;FP+ 6RXWKHUQ K\EULGL]DWLRQ DQDO\VLV UHYHDOV DW OHDVW ILYH DGGLWLRQDO K\EULGL]LQJ IUDJPHQWV LQ ;FP+ EHVLGHV WKH VL[ SODVPLGERUQH DYU JHQHV 'H )H\WHU HW DO f )XUWKHUPRUH DOO VWUDLQV H[DPLQHG RI ;DQWKRPRQDV FDPSHVWULV SY PDOYDFHDUXP WKDW DWWDFN FRWWRQ FDUU\ DW OHDVW '1$ IUDJPHQWV WKDW K\EULGL]H ZLWK PHPEHUV RI WKH DYUSWK JHQH IDPLO\ LQFOXGLQJ PDQ\ VWUDLQV ZKLFK FDUU\ QR NQRZQ DYU JHQHV 'H )H\WHU HW DO f 7KH SXUSRVH RI WKLV UHVHDUFK ZDV WR FORQH

PAGE 45

DQG WR PXWDJHQL]H DOO DYU JHQHV LQ ; FDPSHVWULV SY PDOYDFHDUXP VWUDLQ ;FP+ DQG WR LGHQWLI\ SRWHQWLDO SOHLRWURSLF IXQFWLRQV 0DWHULDOV DQG 0HWKRGV %DFWHULDO VWUDLQV SODVPLGV DQG PDWLQJV 7KH EDFWHULDO VWUDLQV DQG SODVPLGV XVHG LQ WKLV VWXG\ DUH OLVWHG 7DEOH 6WUDLQV RI (VFKHULFKLD FROL ZHUH JURZQ LQ /XULD %HUWDQL /%f PHGLXP 6DPEURRN HW DO f DW r& 6WUDLQV RI ;DQWKRPRQDV ZHUH JURZQ LQ 3<*0 SHSWRQH\HDVW H[WUDFWJO\FHURO0236f PHGLXP DW r& 'H )H\WHU HW DO f )RU FXOWXUH RQ VROLG PHGLD DJDU ZDV DGGHG DW J/ $QWLELRWLFV ZHUH XVHG DV SUHYLRXVO\ GHVFULEHG UHIHU WR &KDSWHU f 7R WUDQVIHU SODVPLGV IURP (FROL VWUDLQV +% RU '+D WR ; FDPSHVWULV SY PDOYDFHDUXP KHOSHU SODVPLG S5. DQG PRGLILHU SODVPLG S8)5 ZHUH XVHG DV GHVFULEHG )H\WHU DQG *DEULHO ODEf 5HFRPELQDQW '1$ WHFKQLTXHV 7RWDO '1$ LVRODWLRQ IURP ;DQWKRPRQDV ZDV DV GHVFULEHG *DEULHO DQG 'H )H\WHU f 3ODVPLGV ZHUH LVRODWHG IURP ( FROL E\ DONDOLQH O\VLV PHWKRGV 6DPEURRN HW DO f 5HVWULFWLRQ HQ]\PH GLJHVWLRQ DONDOLQH SKRVSKDWDVH WUHDWPHQW '1$ OLJDWLRQ DQG UDQGRP SULPLQJ UHDFWLRQV ZHUH SHUIRUPHG DV UHFRPPHQGHG E\ WKH PDQXIDFWXUHUV 6RXWKHUQ K\EULGL]DWLRQ ZDV SHUIRUPHG E\ XVLQJ Q\ORQ PHPEUDQHV DV GHVFULEHG /D]R HW DO f 2WKHUZLVH VWDQGDUG UHFRPELQDQW '1$ SURFHGXUHV ZHUH XVHG 6DPEURRN HW DO f *HQH UHSODFHPHQW DQG PDUNHU HYLFWLRQ 7R FDUU\ RXW PDUNHUH[FKDQJH PXWDJHQHVLV D NE %DP+O IUDJPHQW FRQWDLQLQJ D QSWOVDF FDUWULGJH IURP S80

PAGE 46

7DEOH %DFWHULDO VWUDLQV DQG SODVPLGV XVHG LQ WKLV VWXG\ 6WUDLQ RU 3ODVPLG 5HOHYDQW &KDUDFWHULVWLFV 5HIHUHQFH RU 6RXUFH ( FROL '+m ) HQG$O KVG5 UNnPNf VXS( *LEFR%5/ WKL UHF$O J\U$ UHO$O c!GODF= *DLWKHUVEXUJ 0' +% $0 $ODF=<$DUJ)f? VXS( KVG6 UNnPNf UHF$ %R\HU t DUD SUR$ ODF
PAGE 47

7DEOH FRQWLQXHG 6WUDLQ RU 3ODVPLG 5HOHYDQW &KDUDFWHULVWLFV 5HIHUHQFH RU 6RXUFH 3ODVPLG S5. S5. GHULYDWLYH QSW7Q.Pr 6SU 7UD KHOSHU SODVPLG /HRQJ HW DO S8)5 &RO(O 0RE? &Pn 'H )H\WHU HW DO S8)5 ,QF: .Pn 0RE ODF=D 3DU FRV 'H )H\WHU HW DO S8)5 ,QF: .PU *Pn 0RE ODF=WW 3DU 'H )H\WHU t *DEULHO D S8)5 ,QF: *PU $Sn 0RE ODF=D 3DU 'H )H\WHU HW DO S8)5 ,QF3 7Fn 0RE FRQWDLQLQJ PHWK\ODVHV ;PDO DQG ;PD:O 'H )H\WHU t *DEULHO E S8)5 &RVPLG FORQH FDUU\LQJ DYU% DYUE 'H )H\WHU t *DEULHO D S8)5 &RVPLG FORQH FDUU\LQJ DYU% DYUE DYU%,2O 'H )H\WHU t *DEULHO D S8)5 &RVPLG FORQH FDUU\LQJ DYUE DYU%,2O DYU%OQ 'H )H\WHU t *DEULHO D S8)5 &RVPLG FORQH FDUU\LQJ DYU%,2O DYU%OQ 'H )H\WHU t *DEULHO D S8)5OOO &RVPLG FORQH FDUU\LQJ DYU%,2O DYU%OQ DYU% 'H )H\WHU t *DEULHO D S8)5 &RVPLG FORQH FDUU\LQJ DYU%OQ DYU% DYUE 'H )H\WHU t *DEULHO D S8)5 &RVPLG FORQH FDUU\LQJ DYU% DYUE 'H )H\WHU t *DEULHO D S8)5 NE IUDJPHQW FRQWDLQLQJ DYU% LQ S8)5 'H )H\WHU t *DEULHO D S8)5 NE IUDJPHQW FRQWDLQLQJ DYUE LQ S8)5 'H )H\WHU t *DEULHO D S8)5 NE IUDJPHQW FRQWDLQLQJ DYU%,2O LQ S8)5 'H )H\WHU HW DO S8)5 NE IUDJPHQW FRQWDLQLQJ DYU%OQ LQ S8)5 'H )H\WHU HW DO S8)5 NE IUDJPHQW FRQWDLQLQJ DYU% LQ S8)5 'H )H\WHU HW DO S8)5 NE IUDJPHQW FRQWDLQLQJ DYUE LQ S8)5 'H )H\WHU HW DO S8)< NE IUDJPHQW FDUU\LQJ SWK$YQSWVDF LQ S8)5 7KLV VWXG\ S8)
PAGE 48

5LHG DQG &ROOPHU f ZDV UDQGRPO\ OLJDWHG LQWR D %DOO VLWH RI SWK$ RQ S=LW 5HFRPELQDQW SODVPLGV ZHUH VFUHHQHG IRU DQ LQVHUWLRQ LQ WKH PLGGOH RI WKH WDQGHPO\ UHSHDWHG UHJLRQ DQG S8)< ZDV VHOHFWHG LQ ZKLFK WKH QSWOVDF FDUWULGJH ZDV IRXQG LQ WKH %DOO VLWH RI UHSHDW QXPEHU RI SWK$ $Q 6VW, IUDJPHQW FDUU\LQJ WKH SWOL$QSWO VDF IXVLRQ IURP S8)< ZDV UHFORQHG LQWR WKH VXLFLGH YHFWRU S8)5 'H )H\WHU HW DO f IRUPLQJ S8)
PAGE 49

SHUIRUPHG WKUHH WLPHV 'DWD SRLQWV VKRZQ LQ )LJ ZHUH WKH PHDQ DQG VWDQGDUG HUURU RI WKUHH VDPSOHV IURP RQH H[SHULPHQW %DFWHULDO SRSXODWLRQV UHOHDVHG RQWR WKH OHDI VXUIDFH ZHUH TXDQWLILHG DV GHVFULEHG UHIHU WR &KDSWHU f 7KHVH GDWD DUH UHSRUWHG DV WKH PHDQ DQG VWDQGDUG HUURU RI WKUHH VDPSOHV 6LPLODU UHVXOWV ZHUH REWDLQHG LQ WZR LQGHSHQGHQW H[SHULPHQWV 5HVXOWV &ORQLQJ RI VL[ SRWHQWLDO DYU JHQHV IURP ;FP+ E\ K\EULGL]DWLRQ ,Q DGGLWLRQ WR WKH VL[ PHPEHUV RI WKH ;DQWKRPRQDV DYUSWK JHQH IDPLO\ SUHYLRXVO\ FORQHG IURP SODVPLG S;FP+ RI VWUDLQ ;FP+ 'H )H\ WHU DQG *DEULHO Df 6RXWKHUQ EORW DQDO\VHV UHYHDOHG WKH SUHVHQFH RI VL[ PRUH '1$ IUDJPHQWV LQ ;FP+ WKDW K\EULGL]HG WR DQ LQWHUQDO %DP +, IUDJPHQW IURP DYUE )LJ f $OO RI WKH IUDJPHQWV DSSHDUHG WR EH ODUJH HQRXJK WR FDUU\ IXQFWLRQDO PHPEHUV RI WKH DYUSWK JHQH IDPLO\ %DVHG RQ WKH LQWHQVLW\ RI K\EULGL]DWLRQ DQG WKH IDFW WKDW ;FP+ DSSHDUV WR FDUU\ RQO\ D VLQJOH SODVPLG S;FP+ 'H )H\ WHU DQG *DEULHO Df WKH DGGLWLRQDO K\EULGL]LQJ IUDJPHQWV DSSHDUHG WR EH FKURPRVRPDO &RORQ\ K\EULGL]DWLRQV ZHUH FDUULHG RXW WR LVRODWH WKHVH FKURPRVRPDO '1$ IUDJPHQWV IURP D JHQRPLF OLEUDU\ RI ;FP+ 'H )H\WHU DQG *DEULHO Df E\ SURELQJ ZLWK DQ LQWHUQDO %DP
PAGE 50

)LJXUH 6RXWKHUQ K\EULGL]DWLRQ RI FRVPLG '1$V FRQWDLQLQJ VL[ FKURPRVRPDO DYU JHQHV DIWHU (FR5, DQG 6Y, GRXEOH GLJHVWLRQ 7KH EORW ZDV SUREHG ZLWK a3 ODEHOOHG LQWHUQDO %DP+? IUDJPHQW RI DYUE $ (WKLGLXP EURPLGH VWDLQHG JHO % DXWRUDGLRJUDSK UHVXOWLQJ IURP 6RXWKHUQ K\EULGL]DWLRQ /DQH ;FP+ JHQRPLF '1$ ODQH S;FP+ ODQH S;FPO +FO XSSHU EDQG DQG DYU%Q ORZHU EDQGf ODQH S;FPO DYU%f ODQH S;FP DYU%Qf ODQH S;FP DYU% XSSHU EDQG DQG DYU% ORZHU EDQGf ODQH S;FP DYU%f ODQH S;FPO +Ff 0 PROHFXODU PDUNHU D +FO E DYU% F DYU%O2, G DYU%O H DYU% I DYU%Q J DYU%,Q K DYU%? L DYUE? M DYU%? N DYUE +F

PAGE 52

SRO\PRUSKLVPV )LJ f 6L[ GLIIHUHQW FKURPRVRPDO '1$ IUDJPHQWV K\EULGL]LQJ ZLWK DYUE KDG EHHQ FORQHG DQG LGHQWLILHG 7KHVH '1$ IUDJPHQWV ZHUH DUELWUDULO\ QDPHG DFFRUGLQJ WR VL]H DV +FO 6L[ FRVPLG FORQHV ZHUH VHOHFWHG IRU IXUWKHU VWXG\ S;FPO FDUULHG +FO DQG +F S;FPO FDUULHG +F S;FP FDUULHG +F S;FP FDUULHG +F DQG +F S;FP FDUULHG +F RQO\ DQG S;FPO FDUULHG +F )LJ f 5HVWULFWLRQ GLJHVWV RI WKHVH FORQHV UHYHDOHG WKDW +FO +F DQG +F ZHUH FORVHO\ OLQNHG DQG WKDW +F DQG +F ZHUH FORVHO\ OLQNHG $ YLUXOHQFH DFWLYLW\ RI K\EULGL]LQJ IUDJPHQWV $OO FRVPLG FORQHV ZHUH FRQMXJDOO\ WUDQVIHUUHG WR WKH ZLGHO\ YLUXOHQW ; FDPSHVWULV SY PDOYDFHDUXP VWUDLQ ;FPO DQG WHVWHG RQ FRWWRQ FY $FDOD $Ff DQG FRWWRQ OLQHV FRQJHQLH ZLWK $F DQG HDFK FDUU\LQJ RQH RI WKH IROORZLQJ UHVLVWDQFH JHQHV %O % % %D %E E E %OQ DQG %,Q 7R ORFDOL]H PHPEHUV RI WKH DYUSWK JHQH IDPLO\ DOO FRVPLG FORQHV ZLWK DYLUXOHQFH DFWLYLW\ ZHUH GLUHFWLRQDOO\ VXEFORQHG LQWR S8)5 E\ (FR5O DQG +LQGOOO GLJHVWLRQ WKHVH HQ]\PHV DUH QRW NQRZQ FXW ZLWKLQ PHPEHUV RI WKH IDPLO\ SXEOLVKHG WR GDWHf S;FPO ZDV VXEFORQHG WR VHSDUDWH IUDJPHQWV +FO DQG +F JHQHUDWLQJ S8)< +FOf DQG S8)< +Ff )UDJPHQW +F RQ S;FPO ZDV VXEFORQHG WR IRUP S8)< S;FP ZDV VXEFORQHG WR JHQHUDWH S8)< VWLOO FDUU\LQJ +F DQG +F RQ D VLQJOH NE (FR%O+LQGOOO '1$ IUDJPHQW )UDJPHQW +F RQ S8)< ZDV IXUWKHU VXEFORQHG WR IRUP S8)< )UDJPHQW +F RQ S;FP ZDV VXEFORQHG WR IRUP S8)< 6LQFH +F RQ S8)< FRQIHUUHG QR DYLUXOHQFH WR ;FPO LW ZDV QRW HYDOXDWHG IXUWKHU

PAGE 53

7KH LQGLFDWHG VXEFORQHV ZHUH WUDQVIHUUHG WR ;FPO DQG WHVWHG RQ WKH FRQJHQLH FRWWRQ OLQHV DQG RQ FRWWRQ OLQH FDUU\LQJ UHVLVWDQFH JHQH %Q 7KH +FO DQG +F '1$ IUDJPHQWV FRQIHUUHG QR GHWHFWHG SKHQRW\SH WR ;FPO 7KH +F DQG +F '1$ IUDJPHQWV FRQIHUUHG H[DFWO\ WKH VDPH DYLUXOHQFH VSHFLILFLW\ WR ;FPO DV GLG FORQHV FDUU\LQJ DYU%,2O 'H )H\WHU HW DO f DYLUXOHQFH RQ $F%D DQG $F%,Q 7KHUHIRUH ZH FRQFOXGH WKDW +F DQG +F FDUU\ WZR QHZ PHPEHUV RI WKH JHQH IDPLO\ QDPHG DYU% DQG DYU% UHVSHFWLYHO\ 7KH +F '1$ IUDJPHQW RQ S;FP FRQIHUUHG DYLUXOHQFH WR ;FPO RQ FRWWRQ OLQHV FRQWDLQLQJ UHVLVWDQFH JHQHV %Q RU %,Q *HQH DZ%Q KDG SUHYLRXVO\ EHHQ FORQHG IURP ;FP+ RQ S8)$+O DQG VKRZQ WR FRQIHU DYLUXOHQFH RQ FRWWRQ OLQHV FRQWDLQLQJ %Q *DEULHO HW DO f EXW DYU%Q ZDV QRW NQRZQ WR EH D PHPEHU RI WKLV DYUSWK JHQH IDPLO\ 5HVWULFWLRQ DQG 6RXWKHUQ K\EULGL]DWLRQ DQDO\VHV QRW VKRZQf UHYHDOHG WKDW DYU%Q LV D PHPEHU RI WKLV JHQH IDPLO\ DQG LV FDUULHG RQ ERWK S8)$+O DQG S;FP 7KH +F '1$ IUDJPHQW FRQIHUUHG DYLUXOHQFH WR ;FPO RQ $F%E DQG $F%,Q 7KHUHIRUH ZH FRQFOXGH WKDW +F FDUULHG D QHZ PHPEHU RI WKH JHQH IDPLO\ QDPHG DYU% ,QDFWLYDWLRQ RI DYU JHQHV LQ ;FP+ 7R VWXG\ WKH SOHLRWURSLF IXQFWLRQ RI DOO PHPEHUV RI WKH DYUSWK JHQH IDPLO\ LQ ;FP+ D ULIDP\FLQ PXWDQW RI ;FP+f PDUNHU H[FKDQJHHYLFWLRQ PXWDJHQHVLV ZDV FDUULHG RXW WR LQDFWLYDWH RU GHOHWH HDFK RI WKH DYU JHQHV VLQJO\ RU LQ FRPELQDWLRQ 0RUH WKDQ PXWDQWV ZHUH JHQHUDWHG DQG RI WKHP ZHUH WHVWHG RQ FRQJHQLH FRWWRQ UHVLVWDQFH OLQHV DQG DQDO\]HG E\ 6RXWKHUQ K\EULGL]DWLRQ 0DUNHU H[FKDQJHHYLFWLRQ PXWDQWV DIIHFWLQJ DOO DYU JHQH ZHUH REWDLQHG ,Q 7DEOH DUH OLVWHG VHOHFWHG PXWDQWV ZKLFK H[KLELWHG DOWHUHG SKHQRW\SHV RQ

PAGE 54

7DEOH 3KHQRW\SHV RI PDUNHUH[FKDQJH PXWDQWV RI ;DQWKRPRQDV FDPSHVWULV SY PDOYDFHDUXP VWUDLQ ;FP+ RQ FRWWRQ ;FP VWUDLQV &RWWRQ FY $FDOD FRQJHQLH OLQHV $F $F% $F%D $FE $FE $F%OQ $F%OQ ;FP+ +0 DYU%OQQSWOVDFf +0 DYUEQSWOVDFf +0 DYU%a DYUEff s +06 DYU%a DYUEaf s +0 DYU%a DYUEa DYU%OQa DYU%a DYUaf +0 DYU%a DYU%a DYUEa DYUEa DYU%OQa DYU%a DYUaf +06 DYU%a DYU%a DYUEa DYUEa DYU%OQ DYU%a DYUaf 1RWH VWURQJ ZDWHUVRDNLQJ ZHDN ZDWHUVRDNLQJ H[WUHPHO\ ZHDN ZDWHUVRDNOQJ K\SHUVHQVLWLYH UHVSRQVH +5f s ZHDN +5 A 22

PAGE 55

FRWWRQ 6RXWKHUQ DQDO\VLV RI PXWDQWV ZDV VKRZQ LQ )LJ $V SUHGLFWHG E\ JHQH IRUJHQH WKHRU\ PXWDWLRQV RI DYU%,Q +0f DYUE +0f RU DYU% SOXV DYUE +0f LQ ;FP+ UHVXOWHG LQ WKH ORVV RI DYU VSHFLILFLW\ IRU FRWWRQ EOLJKW UHVLVWDQFH JHQHV %OQ E % DQG E UHVSHFWLYHO\ $V UHSRUWHG LQ WKH SUHYLRXV FKDSWHU D VLQJOH PXWDWLRQ RI DYUE UHGXFHV WKH ZDWHUVRDNLQJ DELOLW\ RI ;FP+ RQ VXVFHSWLEOH FRWWRQ OLQHV ,Q FRQWUDVW PXWDWLRQ RI DYU% DYUERU DYU%,Q GLG QRW DIIHFW ZDWHUVRDNLQJ DELOLW\ RI ;FP+ RQ $F $ PXWDWLRQ LQ DYU% +0f ZDV VKRZQ WR UHVXOW LQ WKH VDPH SKHQRW\SLF GHIHFWV DV WKH PXWDWLRQ LQ DYUE L H D ORVV RI DYU VSHFLILFLW\ IRU UHVLVWDQFH JHQH E DQG UHGXFWLRQ RI ZDWHUVRDNLQJ RQ WKH VXVFHSWLEOH FRWWRQ OLQH $F +RZHYHU WKLV PXWDWLRQ FRXOG QRW EH FRPSOHPHQWHG E\ WKH FORQHG DYU% JHQH LQVWHDG ZDV FRPSOHPHQWHG E\ WKH FORQHG DYUE LQGLFDWLQJ DYUE LQ +0 ZDV OLNHO\ WR EH DIIHFWHG 0XWDWLRQ RI ERWK DYU% DQG DYUE LQ +0 DQG +06 UHVXOWHG D SDUWLDO ORVV RI DYLUXOHQFH VSHFLILFLW\ IRU FRWWRQ EOLJKW UHVLVWDQFH JHQH %E 0XWDWLRQ RI DYU%O2O DYU% DYU% +0f DYU% +0f DYU%Q +0f RU +FO +0f IUDJPHQW UHVXOWHG LQ QR DOWHUDWLRQ RI DYLUXOHQFH SKHQRW\SH 7KLV LV FRQVLVWHQW ZLWK WKH IDFW WKDW WKH FRWWRQ UHVLVWDQFH JHQHV WKDW UHDFW ZLWK WKHVH DYU JHQHV DOVR UHDFW ZLWK RWKHU DYU JHQHV JHQHIRUJHQHV 'H )H\WHU HW DO f 1R PXWDWLRQV DIIHFWLQJ IUDJPHQW +F ZHUH UHFRYHUHG 7R LQDFWLYDWH D ODUJH QXPEHU RI DYU JHQHV LQ ;FP+ WKH QSWOVDF PDUNHU HYLFWHG PXWDQW +06 DYU%a DYUEaf ZDV VXEMHFWHG WR D VHFRQG URXQG RI PDUNHU H[FKDQJH PXWDJHQHVLV 7KH UHVXOWLQJ PXWDQW +0 DQG PDUNHU HYLFWHG SURJHQ\ +06 KDG DW OHDVW DYU JHQHV DYU%I DYU%a DYUEn DYUE7 DYU%,Q DYU%O2On

PAGE 56

FGf§ Hf§ Jf§ Kf§ L"r L U )LJXUH 6RXWKHUQ K\EULGL]DWLRQ RI eFR5, DQG 6V, GRXEOH GLJHVWHG JHQRPLF '1$V RI ;FP+ PXWDQWV 7KH EORW ZDV SUREHG ZLWK n3 ODEHOOHG LQWHUQDO %DP8O IUDJPHQW RI DYUE /DQH ;FP+ ODQH +0 ODQH +0 ODQH +0 ODQH +0 ODQH +0 ODQH +0 ODQH +0 ODQH +0 ODQH +06 ODQH +0 ODQH +06 D +FO E DYU%? F DYU%? G DYU% H DYU% I DZ%UY J DYU%OQ? K DYU%O? L DYUE M DYU% N DYUE? +F

PAGE 57

DYU nf LQDFWLYDWHG 7KHVH WZR PXWDQWV QRW RQO\ ORVW DYU VSHFLILFLW\ RQ PRVW RI FRWWRQ UHVLVWDQFH OLQHV EXW DOVR ORVW QHDUO\ DOO ZDWHUVRDNLQJ DELOLW\ RQ WKH VXVFHSWLEOH OLQH $F 7DE )LJ f 'HVSLWH UHSHDWHG DWWHPSWV WR LQDFWLYDWH DOO PHPEHUV RI WKH JHQH IDPLO\ LQ ;FP+ ZH ZHUH XQDEOH WR UHGXFH WKH QXPEHU RI PHPEHUV EHORZ ILYH ,Q DOO FDVHV PHPEHUV RI WKH IDPLO\ DSSHDUHG WR UHDUUDQJH WR IRUP QHZ PHPEHUV DV HYLGHQFHG E\ WKH DSSHDUDQFH RI QHZ '1$ IUDJPHQWV RI VXIILFLHQW VL]H WR HQFRGH D PHPEHU DQG RIWHQ DFFRPSDQLHG E\ WKH DSSHDUDQFH RI D QHZ DYLUXOHQFH VSHFLILFLW\ SKHQRW\SH GDWD QRW VKRZQf $OVR LQ DOO FDVHV DW OHDVW RQH PHPEHU UHPDLQHG SODVPLGERPH DV HYLGHQFHG E\ FRPSDUDWLYHO\ VWURQJ K\EULGL]DWLRQ LQWHQVLWLHV FKDUDFWHULVWLF RI WKH SODVPLGERPH JHQHV (IIHFW RI ORVV RI ZDWHUVRDNLQJ RQ EDFWHULDO JURZWK LQ SODQWD 6LQFH DYUE6 PXWDQWV RI ;FP+ WKDW KDG ORVW VLJQLILFDQW ZDWHUVRDNLQJ DELOLW\ RQ $F ZHUH VKRZQ WR EH XQDIIHFWHG LQ JURZWK LQ SODQWD UHIHU WR &KDSWHU f ZH HYDOXDWHG JURZWK RI +06 RQ $F LQ FRPSDULVRQ ZLWK WKH SDUHQWDO VWUDLQ ;FP+ :KLOH LQDFWLYDWLRQ RI DYU JHQHV UHVXOWHG LQ D ORVV RI QHDUO\ DOO ZDWHUVRDNLQJ DELOLW\ RQ FRWWRQ LQ SODQWD JURZWK UDWH DQG \LHOG RI +06 ZDV QRW DIIHFWHG )LJ f +RZHYHU WKH DPRXQW RI +06 UHOHDVHG RQWR WKH OHDI VXUIDFH ZDV WLPHV OHVV WKDQ WKDW RI ;FP+ 2QO\ b s ; FIXFP RXW RI s ; FIXFPf RI +06 EDFWHULD SUHVHQW LQ WKH LQIHFWHG OHDI WLVVXH ZHUH UHOHDVHG RQWR WKH OHDI VXUIDFH ZKHUHDV b s ; FIXFP RXW RI s ; FIXFPf RI WKH ;FP+ EDFWHULD ZHUH UHOHDVHG RQWR WKH VXUIDFH 7KHUHIRUH WKH PXWDQW +06 ZDV EDVLFDOO\ D QRQSDWKRJHQLF HQGRSK\WH RI FRWWRQ

PAGE 58

)LJXUH (IIHFW RI DYU JHQHV RQ ZDWHUVRDNLQJ V\PSWRPV FDXVHG E\ ; FDPSHVWULV SY PDOYDFHDUXP RQ FRWWRQ VXVFHSWLEOH OLQH $F

PAGE 59

'D\ SRVWf§LQRFXODWLRQ )LJXUH ,QDFWLYDWLRQ RI PXOWLSOH DYU JHQHV LQ ; FDPSHVWULV SY PDOYDFHDUXP GLG QRW DIIHFW EDFWHULDO JURZWK LQ SODQWD RQ FRWWRQ VXVFHSWLEOH OLQH $F )LOOHG FLUFOH ;FP+ RSHQ FLUFOH +06

PAGE 60

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f ZHUH LQWURGXFHG LQWR +06 PDQ\ ZHUH DEOH WR SDUWLDOO\ FRPSOHPHQW WKH ZDWHUVRDNLQJ GHIHFW )LJV DQG f )RU H[DPSOH SODVPLGV FRQWDLQLQJ DYU%,Q RU DYU%O2O DORQH FRQIHUUHG QR ZDWHUVRDNLQJ DFWLYLW\ EXW SODVPLGV FRQWDLQLQJ ERWK DYU%,Q DQG DYU%O2O WRJHWKHU H[KLELWHG ZDWHUVRDNLQJ DFWLYLW\ &RVPLG FORQHV S8)5 DQG S8)5f FDUU\LQJ DYUE DQG DW OHDVW WZR RWKHU DYU JHQHV ZHUH DEOH WR IXOO\ FRPSOHPHQW WKH ZDWHUVRDNLQJ GHIHFWV RI +06 )LJ f 7KHUHIRUH DW OHDVW WKUHH RI WKH S;FP+ DYUSWK JHQHV ZHUH UHGXQGDQW LQ WHUPV RI SDWKRJHQLFLW\ IXQFWLRQ 7R H[FOXGH WKH SRVVLELOLW\ WKDW VRPH ZDWHUVRDNLQJ JHQHV ZHUH OLQNHG WR DYU JHQHV RQ WKHVH FRPSOHPHQWLQJ FRVPLG FORQHV VXEFORQLQJ RI S8)5 ZDV FDUULHG RXW $FWLYLW\ DVVD\V RI UHJLRQV EHWZHHQ WKH DYU JHQHV GLG QRW UHYHDO DQ\ ZDWHUVRDNLQJ DFWLYLW\ ZKHQ LQWURGXFHG LQWR +06

PAGE 61

DYU% DYUE DYU% DYU%OQ DYU% DYUE S;FP+ A L LL L L $FWLYLW\ LQ +06 38)5 DYUE6 38)5 DYU% DYUE S8)5 DYU % DYUE 38)5 DYU%,2, DYU%OQ S8)5 DYU%,2, DYU%OQ DYU % S8)5 DYU%OQ DYU % DYUE S8)5 DYUE DYU%,2, DYU%OQ 38)5 DYU% DYUE DYU%, )LJXUH 3OHLRWURSLF ZDWHUVRDNLQJ IXQFWLRQ RI FORQHG ;FP+ DYU JHQHV LQ PXWDQW +06 r +06 ZDV QHDUO\ DV\PSWRPDWLF RQ VXVFHSWLEOH FRWWRQ OLQH $F ZHDN ZDWHUVRDNLQJ DFWLYLW\ LQWHUPHGLDWH OHYHO ZDWHUVRDNLQJ DFWLYLW\ IXOO ZDWHUVRDNLQJ DFWLYLW\ DV ZLOG W\SH ;FP+f

PAGE 62

'LVFXVVLRQ 1HDUO\ DOO RI WKH DZ JHQHV UHSRUWHG IURP WKH JHQXV ;DQWKRPRQDV DUH PHPEHUV RI D VLQJOH KLJKO\ KRPRORJRXV JHQH IDPLO\ DQG ZLWK WKH H[FHSWLRQ RI SWK$ SD=WRJHQLFLW\f RI ; FLWUL 6ZDUXS HWDO f DOO RI WKH RWKHU SXEOLVKHG PHPEHUV ZHUH LGHQWLILHG DQG FORQHG DV DZ JHQHV 6RXWKHUQ DQDO\VHV KDYH VKRZQ WKDW PHPEHUV RI WKLV DZSWK JHQH IDPLO\ H[LVW LQ PXOWLSOH FRSLHV IURP WR f LQ DOO WHVWHG ; FDPSHVWULV SY PDOYDFHDUXP VWUDLQV LVRODWHG IURP FRWWRQ EXW WKDW VRPH VWUDLQV SDUWLFXODUO\ WKRVH IURP $IULFD DSSHDU WR ODFN IXQFWLRQDO DYLUXOHQFH 'H )H\WHU HW DO f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f ZHUH QRWHG WR FRQIHU SOHLRWURSLF SDWKRJHQLFLW\ ZDWHUVRDNLQJf IXQFWLRQ 'H )H\WHU DQG *DEULHO D 'H )H\WHU HW DO f :H UHSRUW KHUH WKH FORQLQJ RI WKH RWKHU VL[ K\EULGL]LQJ IUDJPHQWV WKDW DSSHDU WR EH IURP WKH FKURPRVRPH 2QH RI WKHVH IUDJPHQWV HQFRGHV WKH

PAGE 63

SUHYLRXVO\ GHVFULEHG DYU%Q *DEULHO HW DO f 7KUHH RI WKHVH HQFRGH SUHYLRXVO\ XQGHVFULEHG DYU JHQHV DYU% DYU% DQG DYU%f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f LQDFWLYDWHG DQG FRPSOHWHO\ ORVW WKH DELOLW\ WR LQGXFH ZDWHUVRDNLQJ RQ PDWXUH FRWWRQ OHDYHV 6LQFH +06 VXIIHUHG GHOHWLRQV RI DYU JHQHV DQG DGMDFHQW '1$ IUDJPHQWV PXWDWLRQDO DQDO\VLV DORQH GLG QRW UXOH RXW WKH SRWHQWLDO LQYROYHPHQW RI FORVHO\ OLQNHG YLUXOHQFH JHQHV +RZHYHU FRPSOHPHQWDWLRQ WHVWV FRQILUPHG WKDW DYUSWK IDPLO\ PHPEHUV IXOO\ FRPSOHPHQWHG WKH ZDWHUVRDNLQJ GHIHFWVf RI +06 ,QGLYLGXDOO\ QRQH RI WKH DYUSWK JHQHV RWKHU WKDQ DYUE FRXOG HYHQ SDUWLDOO\ FRPSOHPHQW WKH QRQSDWKRJHQLF PXWDQW VWUDLQ +06 LQ YLVXDO DVVD\V +RZHYHU '1$ IUDJPHQWV FDUU\LQJ PXOWLSOH DYUSWK JHQHV IURP SODVPLG S;FP+ ZHUH DEOH WR SDUWLDOO\ FRPSOHPHQW WKH SDWKRJHQLF GHIHFW RI +06 HYHQ LI WKH\ GLG QRW FDUU\ DYUE GHPRQVWUDWLQJ WKDW RWKHU DYUSWK JHQHV IURP ;FP+ DOVR FRQWULEXWHG WR LWV SDWKRJHQLFLW\

PAGE 64

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f WKHUH ZDV QR HYLGHQFH WKDW WKH DYUSWK JHQHV RI ;FP+ DUH HYHQ VOLJKWO\ LQYROYHG LQ EDFWHULDO JURZWK LQ SODQWD ,QVWHDG WKH\ ZHUH LQYROYHG LQ WKH LQGXFWLRQ RI ZDWHUVRDNLQJ V\PSWRPV DQG DVVRFLDWHG UHOHDVH RI b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

PAGE 65

7KH ODFN RI FRUUHODWLRQ EHWZHHQ GLVHDVH V\PSWRPV DQG JURZWK LQ SODQWD KDV EHHQ REVHUYHG ZLWK RWKHU SODQW SDWKRJHQV ,QGROH DFHWLF DFLGGHILFLHQW VWUDLQV RI 3VHXGRPRQDV V\ULQJDH SY VDYDVWDQRL IDLOHG WR HOLFLW JDOO RQ ROHDQGHU EXW H[KLELWHG VLPLODU JURZWK SDWWHUQV DV WKH JDOOHOLFLWLQJ ZLOGW\SH VWUDLQ 6PLGW DQG .RVXJH f 7KH OHP$ JHQH ZKLFK HQFRGHV D WZRFRPSRQHQW UHJXODWRU LV UHTXLUHG E\ 3 V\ULQJDH SY V\ULQJDH IRU GLVHDVH OHVLRQ IRUPDWLRQ RQ EHDQ SODQWV EXW QRW EDFWHULDO JURZWK ZLWKLQ RU RQ WKH OHDYH RI EHDQ +UDEDN DQG :LOOLV :LOOLV HW DO f 0XWDWLRQ RI WKH IXQJDO SDWKRJHQ &ROOHWRWULFKXP PDJQD UHVXOWHG LQ D VWUDLQ ZKLFK LV QRW DEOH WR LQGXFH SDWKRJHQLF V\PSWRPV EXW JURZV LQ KRVW WLVVXH DV DQ HQGRSK\WH DQG UHWDLQV WKH ZLOGW\SH KRVW UDQJH )UHHPDQ DQG 5RGULJXH] f $ JURZLQJ QXPEHU RI ;DQWKRPRQDV YLUXOHQFH SDWKRJHQLFLW\ JHQHV DUH QRW KUS K\SHUVHQVLWLYH UHVSRQVH DQG SDWKRJHQLFLW\f JHQHV EXW LQVWHDG DUH KRVWVSHFLILF GHWHUPLQDQWV RI GLVHDVH DQGRU KRVW UDQJH *DEULHO HW DO f )RU H[DPSOH VHYHUDO JHQHV DIIHFWLQJ KRVW UDQJH ZHUH LGHQWLILHG DQG FRPSOHPHQWLQJ '1$ FORQHV LVRODWHG IURP ; FDPSHVWULV SY FLWUXPHOR .LQJVOH\ HW DO f DQG ; FDPSHVWULV SY WUDQVOXFHQV :DQH\ HW DO f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

PAGE 66

QRGXODWLRQ JHQHV ZKLFK DUH UHTXLUHG IRU KRVW UDQJH RQ VRPH KRVWV EXW ZKLFK FDQ DOVR FRQIHU DYLUXOHQFH ZKHQ WUDQVIHUUHG WR RWKHU 5KL]RELXP VWUDLQV ZLWK D GLIIHUHQW KRVW UDQJH 'HEHOOH HW DO )DXFKHU HW DO /HZLV+HQGHUVRQ DQG 'MRUGMHYLF f 7KH SOHLRWURSLF IXQFWLRQ DQG DSSDUHQW ILWQHVV YDOXH RI VRPH DYU JHQHV GR QRW LPSO\ WKDW PRVW DYU JHQHV KDYH VHOHFWLYH YDOXH IRU D VSHFLILF SODQW SDWKRJHQLF VWUDLQ ,Q IDFW PRVW DYU JHQHV UHSRUWHG WR GDWH DSSHDU WR EH GLVSHQVLEOH DQG KDYH QR REYLRXV VHOHFWLYH YDOXH WR WKH SDWKRJHQ *DEULHO HW DO f $W OHDVW VRPH PHPEHUV RI WKH ;DQWKRPRQDV DYUSWK JHQH IDPLO\ DUH IRXQG RQ VHOIPRELOL]LQJ SODVPLGV IRU H[DPSOH %RQDV HW DO f DQG WKHLU SUHVHQFH LQ VRPH VWUDLQV RI D SDWKRYDU PD\ EH GXH WR D FRLQFLGHQWDO OLQNDJH ZLWK DQRWKHU VHOHFWHGf IDFWRU VXFK DV FRSSHU UHVLVWDQFH RQ WKH VDPH SODVPLG 2QFH KRUL]RQWDOO\ WUDQVIHUUHG WR D GLIIHUHQW VWUDLQ WKHVH JHQHV PD\ QRW UHPDLQ VWDEOH HQRXJK WR UHWDLQ SWK IXQFWLRQ VSHFLILF IRU VRPH DQFHVWUDO KRVW ,Q ERWK ; FDPSHVWULV SY PDOYDFHDUXP DQG ; FLWUL PHPEHUV RI WKH DYUSWK JHQH IDPLO\ PXWDWH DW YHU\ KLJK IUHTXHQF\ DSSUR[LPDWHO\ 7f WKURXJK LQWHUJHQLF RU LQWUDJHQLF UHFRPELQDWLRQ 'H )H\ WHU HW DO &KDSWHU f +LJK IUHTXHQF\ UHFRPELQDWLRQ SURYLGHV D SRZHUIXO PHFKDQLVP WR JHQHUDWH YDULDWLRQ DQG KRUL]RQWDO JHQH WUDQVIHU SURYLGHV D SRZHUIXO PHFKDQLVP WR GLVVHPLQDWH WKDW YDULDWLRQ LQ WKH JHQXV ,W VHHPV OLNHO\ WKDW PDQ\ DYU JHQHV FDUULHG E\ EDFWHULDO SODQW SDWKRJHQV PD\ UHVXOW IURP YDULDWLRQ LQ UHGXQGDQW DQGRU XQQHFHVVDU\ YLUXOHQFHSDWKRJHQLFLW\ JHQHV ZKLFK DUH VHOHFWLYHO\ QHXWUDO LQ QDWXUDO SRSXODWLRQV

PAGE 67

&+$37(5 ,175$*(1,& 5(&20%,1$7,21 2) $ 6,1*/( 3/$17 3$7+2*(1 *(1( 3529,'(6 $ 0(&+$1,60 )25 7+( (92/87,21 2) 1(: +267 63(&,),&,7,(6 ,QWURGXFWLRQ ,Q JHQHIRUJHQH LQWHUDFWLRQV EHWZHHQ SODQWV DQG PLFURELDO SODQW SDWKRJHQV KRVW SODQW UHVLVWDQFH UHVXOWV IURP WKH JHQHWLF UHFRJQLWLRQ RI UHVLVWDQFH 5f JHQHV LQ WKH SODQW DQG DYLUXOHQFH DYUf JHQHV LQ WKH SDWKRJHQ .HHQ *DEULHO DQG 5ROIH f ,W LV XQNQRZQ ZK\ PLFURELDO SODQW SDWKRJHQV FDUU\ JHQHV WKDW IXQFWLRQ WR OLPLW YLUXOHQFH VLQFH WKH PDMRULW\ RI WKHVH JHQHV DUH GLVSHQVDEOH *DEULHO HW DO f )RU H[DPSOH WR DYRLG SODQW UHVLVWDQFH SODQW SDWKRJHQV HYDGH 5 JHQH UHFRJQLWLRQ E\ GHOHWLRQ 'H )H\WHU HW DO f PXWDWLRQ .RED\DVKL HW DO -RRVWHQ HW DO f RU WUDQVSRVRQ LQVHUWLRQ .HDUQH\ HW DO f RI DYU JHQHV 7KHVH UHSRUWV DOO LQYROYH QDWXUDO PHFKDQLVPV ZKHUHE\ DYLUXOHQFH PD\ EH ORVW 1HYHUWKHOHVV PXWDWLRQDO DQDO\VHV UHYHDO WKDW JDLQV LQ DYLUXOHQFH LQFOXGLQJ QHZ DYLUXOHQFH VSHFLILFLWLHV DUH JHQHUDWHG LQ DW OHDVW VRPH SDWKRJHQV DW DSSUR[LPDWHO\ WKH VDPH IUHTXHQF\ DV WKH\ DUH ORVW 6WDWOHU f $OWKRXJK SUHVXPDEO\ DQ DYU JHQH PD\ EH UHDFWLYDWHG E\ UHYHUVH PXWDWLRQ WKHUH DUH QR UHSRUWV RI QDWXUDO PHFKDQLVPV ZKHUHE\ QHZ DYLUXOHQFH VSHFLILFLWLHV RU QHZ DYU JHQHV DUH JHQHUDWHG 7KH PHFKDQLVPV E\ ZKLFK QHZ DYU JHQHV HYROYH DUH XQNQRZQ

PAGE 68

;DQWKRPRQDV FLWUL LV WKH FDXVDO DJHQW RI FLWUXV FDQNHU GLVHDVH ZKLFK RFFXUV ZRUOGZLGH DQG LV VXEMHFW WR HUDGLFDWLRQ DQG TXDUDQWLQH UHJXODWLRQV LQ PDQ\ FRXQWULHV $ SD]LRJHQLFLW\ JHQH SWK$ LV HVVHQWLDO IRU WKH SDWKRJHQ WR FDXVH K\SHUSODVWLF FDQNHU V\PSWRPV RQ FLWUXV 6ZDUXS HW DO f )XUWKHUPRUH ZKHQ WUDQVIHUUHG WR RWKHU [DQWKRPRQDGV SWK$ FRQIHUV DELOLW\ WR LQGXFH K\SHUSODVWLF FDQNHUV RQ FLWUXV DQG D K\SHUVHQVLWLYH UHVSRQVH +5f RQ RWKHU KRVWV 6ZDUXS HW DO f 7KHUHIRUH SWK$ H[KLELWV SOHLRWURSLF SDWKRJHQLFLW\ DQG DYLUXOHQFH IXQFWLRQV 6RXWKHUQ K\EULGL]DWLRQ UHVWULFWLRQ DQDO\VLV DQG SDUWLDO '1$ VHTXHQFLQJ KDV VKRZQ WKDW SWK$ EHORQJV WR D PDMRU DZ JHQH IDPLO\ ZLGHVSUHDG LQ WKH JHQXV ;DQWKRPRQDV 6ZDUXS HW DO f 0HPEHUV RI WKLV JHQH IDPLO\ LQFOXGH DZ%V DZ%V DQG DZ%V3 D WUDQFDWHG IRUP RI DZ%Vf RI ; F SY YHVLFDWRULD %RQDV HW DO &DQWHURV HW DO f DYU% DZE DYUE DZ%OQ DZ%O2O DQG DZ% RI ; F SY PDOYDFHDUXP 'H )H\WHU DQG *DEULHO 'H )H\WHU HW DO f DQG DYU;DO2 DQG DZ;D RI ; RU\]DH SY RU\]DH +RSNLQV HW DO f $PRQJ WKHVH RQO\ DZE DQG SWK$ DUH NQRZQ WR KDYH SOHLRWURSLF SDWKRJHQLFLW\ IXQFWLRQV 6ZDUXS HW DO 'H )H\WHU DQG *DEULHO DOVR UHIHU WR &KDSWHU f 7KH PRVW VWULNLQJ IHDWXUH RI WKLV JHQH IDPLO\ LV WKH QHDUO\ LGHQWLFDO ES WDQGHP UHSHDWV LQ WKH FHQWUDO SRUWLRQ RI WKH JHQHV 'HOHWLRQ DQDO\VHV KDYH VKRZQ WKDW WKH DYLUXOHQFH VSHFLILFLW\ RI DZ%V LV GHWHUPLQHG E\ WKH ES UHSHWLWLYH PRWLIV RI WKDW JHQH +HUEHUV HW DO f %\ VZDSSLQJ WKH LQWHUQDO UHSHDWHG UHJLRQV EHWZHHQ SWK$ DQG DYUE WKH ES WDQGHP UHSHDWV DUH IRXQG WR GHWHUPLQH ERWK DYLUXOHQFH DQG SDWKRJHQLF VSHFLILFLWLHV UHIHU WR &KDSWHU f

PAGE 69

,Q PRVW RUJDQLVPV UHSHWLWLYH UHJLRQV RI '1$ DUH NQRZQ WR EH DFWLYH VLWHV IRU KRPRORJRXV UHFRPELQDWLRQ $OEHUWLQL HW DO 3HWHV DQG +LOO f 7KH KLJK IUHTXHQF\ RI VXFK HYHQWV FDXVHV WKHP WR EH PDMRU JHQRPH PRGLI\LQJ IRUFHV LQ HYROXWLRQ ,QWHUJHQLF UHFRPELQDWLRQ DPRQJ PHPEHUV RI WKLV DYUSWK JHQH IDPLO\ ZDV SURSRVHG WR H[SODLQ WKH KLJK IUHTXHQF\ RI UDFH FKDQJH PXWDWLRQ IRXQG LQ ; F SY PDOYDFHDUXP 'H )H\WHU HW DO f 6LQFH ERWK SDWKRJHQLF DQG DYLUXOHQFH VSHFLILFLWLHV DUH GHWHUPLQHG E\ WKH ES WDQGHP UHSHDWV LW VHHPHG UHDVRQDEOH WR VXSSRVH WKDW KRPRORJRXV UHFRPELQDWLRQ EHWZHHQ JHQHV DQG DPRQJ UHSHDWV RI WKH VDPH JHQH PLJKW OHDG WR DQ DFFHOHUDWLRQ LQ WKH HYROXWLRQ RI WKLV ;DQWKRPRQDV DYUSWK JHQH IDPLO\ UHVXOWLQJ LQ QHZ KRVW VSHFLILFLWLHV IRU WKH SDWKRJHQ ,Q WKLV VWXG\ ZH DQDO\]HG WKH QXFOHRWLGH VHTXHQFH RI SWK$ DQG GHPRQVWUDWH WKDW LQWUDJHQLF UHFRPELQDWLRQ RI D VLQJOH JHQH FDQ SURYLGH D JHQHWLF PHFKDQLVP IRU WKH FUHDWLRQ RI QHZ DYUSWK JHQHV 0DWHULDOV DQG 0HWKRGV %DFWHULDO VWUDLQV SODVPLGV DQG PDWLQJV (VFKHULFKLD FROL VWUDLQV '+D *LEFR %5/ *DLWKHUVEXUJ 0'f -0
PAGE 70

GHULYDWLYHV ZHUH JURZQ LQ 3<*0 SHSWRQH\HDVW H[WUDFWJO\FHURO0236f PHGLXP DW r& 'H )H\ WHU HW DO f $QWLELRWLFV ZHUH XVHG DV GHVFULEHG LQ SUHYLRXV FKDSWHUV 7ULSDUHQWDO PDWLQJV ZHUH FDUULHG RXW WR WUDQVIHU SODVPLGV IURP (FROL '+D WR YDULRXV ;DQWKRPRQDV VWUDLQV E\ XVLQJ S5. RU S5. DV KHOSHU SODVPLGV DV GHVFULEHG 6ZDUXS HW DO 'H )H\WHU DQG *DEULHO Df 7R WUDQVIHU SODVPLGV LQWR ;FPO WKH PRGLILHU SODVPLG S8)5 FDUU\LQJ ;FPO DQG ;FPO: PHWK\ODVH JHQHV ZDV XVHG WR LQFUHDVH WKH WUDQVIHU IUHTXHQF\ 'H )H\WHU DQG *DEULHO Ef '1$ VHTXHQFH DQDO\VLV 7KH NE 6D, IUDJPHQW FDUU\LQJ SWK$ LQ S=LW 6ZDPS HW DO f ZDV UHFORQHG LQWR WKH S*(0 =If YHFWRU 3URPHJD 0DGLVRQ :,f WR \LHOG S8)< 6HWV RI RYHUODSSLQJ XQLGLUHFWLRQDO GHOHWLRQ VXEFORQHV ZHUH JHQHUDWHG LQ S8)< XVLQJ H[RQXFOHDVH ,,, DQG PXQJ EHDQ QXFOHDVH DV GHVFULEHG E\ 3URPHJD 0DGLVRQ :,f '1$ VHTXHQFLQJ ZDV SHUIRUPHG E\ WKH '1$ 6HTXHQFLQJ &RUH /DERUDWRU\ RI WKH ,QWHUGLVSOLQDU\ &HQWHU IRU %LRWHFKQRORJ\ 5HVHDUFK 8QLYHUVLW\ RI )ORULGD *DLQHVYLOOH 6RPH VHTXHQFLQJ ZDV GHWHUPLQHG E\ XVLQJ WKH GLGHR[\ QXFOHRWLGH FKDLQ WHUPLQDWLRQ WHFKQLTXH ZLWK WKH $PHUVKDP $UOLQJWRQ +HLJKWV ,/f PXOWLZHOO PLFURWLWHU SODWH '1$ VHTXHQFLQJ V\VWHP 531 7KH VHTXHQFH ZDV DQDO\]HG XVLQJ WKH *&* SDFNDJH YHUVLRQ f E\ *HQHWLFV &RPSXWHU *URXS ,QF 0DGLVRQ :, *HQH UHSODFHPHQW DQG VHOHFWLRQ RI PDUNHUHYLFWHG VWUDLQV *HQH SWK$ RQ S=LW ZDV PXWDWHG E\ OLJDWLQJ D NE %DP+O IUDJPHQW FRQWDLQLQJ D QSWOVDF FDUWULGJH 5LHG DQG &ROOPHU f LQWR D %DOO VLWH LQ RQH RI WKH ES WDQGHP UHSHDWV RI SWK$ 2QH RI WKH UHVXOWLQJ WUDQVFULSWLRQDO IXVLRQV S8)< FDUULHG WKH QSWOVDF FDUWULGJH LQVHUWHG LQWR UHSHDW RI SWK$ DQG ZDV VHOHFWHG IRU IXUWKHU VWXG\ $Q 6VWO IUDJPHQW FDUU\LQJ

PAGE 71

WKHSWK$UQSWOVDF IXVLRQ IURP S8)< ZDV UHFORQHG LQWR VXLFLGH YHFWRU S8)5 FRO(O PRE3 FDW? 'H )H\WHU HW DO f IRUPLQJ S8)
PAGE 72

LWV FRQJHQLH UHVLVWDQFH OLQHV $F%O $F% $F% $F% $FE $FE $F%,Q DQG $F%,Q DV GHVFULEHG 'H )H\WHU HW DO 6ZDUXS HW DO f 3ODQW LQRFXODWLRQV LQYROYLQJ ; FLWUL RU SWK$ RU GHULYDWLYHV RI SWK$ ZHUH FDUULHG RXW LQ %/3 OHYHO FRQWDLQPHQW UHIHU WR )HGHUDO 5HJLVWHU 9RO 1R f DW WKH 'LYLVLRQ RI 3ODQW ,QGXVWU\ )ORULGD 'HSDUWPHQW RI $JULFXOWXUH *DLQHVYLOOH %DFWHULDO VXVSHQVLRQV ZHUH VWDQGDUGL]HG LQ VWHULOH WDS ZDWHU WR FIXPO DQG SUHVVXUH LQILOWUDWHG LQWR WKH DED[LDO OHDI VXUIDFH RI WKH SODQWV 3ODQW LQRFXODWLRQV ZHUH UHSHDWHG DW OHDVW WKUHH WLPHV RQ FLWUXV DQG WZR WLPHV RQ EHDQ DQG FRWWRQ 5HVXOWV 6HTXHQFH DQDO\VLV RI SWK$ 7KH FRPSOHWH '1$ VHTXHQFH RI SWK$ ZDV GHWHUPLQHG DQG VKRZQ LQ )LJ '1$ VHTXHQFH FRPSDULVRQV UHYHDOHG WKDW SWK$ RI ; FLWUL LV b LGHQWLFDO WR D?U%V DQG DYU%V RI ; FDPSHVWULV SY YHVLFDWRULD %RQDV HW DO f b LGHQWLFDO WR DYUE RI ; FDPSHVWULV SY PDOYDFHDUXP 'H )H\WHU HW DO f DQG b LGHQWLFDO WR DYU;DO2 RI ; RU\]DH +RSNLQV HW DO f DORQJ LWV HQWLUH OHQJWK $V REVHUYHG ZLWK DOO SUHYLRXVO\ VHTXHQFHG PHPEHUV RI WKH JHQH IDPLO\ %RQDV HW DO 'H )H\WHU HW DO f SWK$ LV IODQNHG E\ QHDUO\ LGHQWLFDO ES WHUPLQDO LQYHUWHG UHSHDWV WKDW SUHFLVHO\ GHILQH WKH OLPLWV RI KRPRORJ\ 7KH SUHGLFWHG DPLQR DFLG VHTXHQFH HQFRGHG E\ SWK$ LV SUHVHQWHG LQ )LJ 3WK$ KDV D FDOFXODWHG PROHFXODU ZHLJKW RI N' DQG DQ LVRHOHFWULF SRLQW RI 7KH SULPDU\ UHJLRQ ZKLFK GLIIHUHQWLDWHV WKH SUHGLFWHG 3WK$ VHTXHQFH IURP WKH SUHGLFWHG VHTXHQFHV RI RWKHU PHPEHUV RI WKH IDPLO\

PAGE 73

)LJXUH 7KH FRPSOHWH '1$ VHTXHQFH RI SWK$ ,QGLFDWHG LQ WKH ILJXUH DUH WKH OHIW DQG ULJKW ES WHUPLQDO LQYHUWHG UHSHDWV ,5/ DQG ,55f IODQNLQJ SWK$ JHQH DQG HOHPHQWV 6KLQH'DOJDPR 6'f ULERVRPHELQGLQJ VLWH PHWKLRQLQH 0f VWDUW FRGRQ WKH ILUVW ES WDQGHP UHSHDW DQG VWRS FRGRQ rf

PAGE 74

$*&$&**7$& &&**&**7&* &$7*$&**** 777$7*&$77 &77*&$7&*& &&***&*&$$ &&*&7$77*7 *&$&$7*&$& &$&&*$&&** $&&&7**$*$ &$**7*7**7 *&&7&*$7$* $7&*$7$*&7 7&$&&&*$7& 7$7&&&77*& &7**$*&*** $7$&**&$*$ $$7*&&&$$$ *$*&&7$&&* $*7$*7*$$7 $&&*$$$*&* &7&&$*&*$7 ****&*&777 &77777$7$* **&$$7**7& 7*77&**&&$ $*&*&&*&&* *7$&$&$*7& 7&$$&&77*7 $*$*&77&&& *&*&***$*7 77$7**$7** 77*7&&&*7& *$&****7$* &$**$*$7*$ ***7&**&$* **$77**7*7 $$$$$$&$*& &$$$$*7*$* &7$$&7&*&7 ES ,5/ *7&$*&$&$* $$$77777&$ &$$&&77&7* &&*$7&&7&& $7*&***7&& *7*$7&*&&7 7&$7*7&7*& *&&7&$&&&7 **7&*7&*$* **77*&&$** $7&$&&&*$$ *77*7*7$&7 *&&$7*&**& &7&**$$*&7 $7*7$**$$& &$&$*$&&*& 7$*7&7**$* *&*$&&$7*7 $$$*$**7$7 *&&7*$7**$ 62 0 7&&&$77&*7 7&*&*&$&$& &$$*7&&7*& &&*&*$*&77 &7*&&&**$& &&&$$&&&*$ 7****77&$* &&*$&7*&$* $7&*7****7 *7&7&&*&&7 *&&**&**&& &&&7**$7** &77*&&&*&7 &**&**$&*$ 7*7&&&**$& &&**&7$&&$ 7&7&&&&&7* &&&&&7&$&& 7*&*77&7&* *&***&$*&7 7&$*7*$&&7 *77$&*7&$* 77&*$7&&*7 &$&77777$$ 7$&$7&*&77 777*$77&$7 7*&&7&&&77 &**&*&7&$& &$7$&$*$** &7*&&$&$** &*$*7***$7 *$**7*&$$7 &***7&7*&* **&$*&&*$& *&&&&&&&$& &&$&&$7*&* &*7**&7*7& $&7*&&*&*& **&&*&&*&* &*&&$$*&&* *&*&&*&*$& *$&*7*&7*& *&$$&&&7&& *$&*&77&*& &**&&*&*&$ **7**$7&7$ &*&$&*&7&* *&7$&$*&&$ *&$*&$$&$* *$*$$*$7&$ $$&&*$$**7 7&*77&*$&$ *7**&*&$*& $&&$&*$**& $&7**7&**& &$7***777$ &$&$&*&*&$ &$7&*77*&* &7&$*&&$$& $&&&**&$*& *77$***$&& *7&*&7*7&$ $*7$7&$**$ &$7*$7&*&$ *&*77*&&$* $**&*$&$&$ &*$$*&*$7& *77**&*7&* *&$$$&$*7* *7&&**&*&$ &*&*&&&7** $**&&77*&7 &$&**7**&* **$*$*77*$ *$**7&&$&& *77$&$*77* *$&$&$**&& $$&77&7&$$ *$77*&$$$$ &*7**&**&* 7*$&&*&$*7 **$**&$*7* &$7*&$7**& *&$$7*&$&7 *$&***7*&& &&&&7*$$&& 7*$&&&&**$ *&$**7**7* *&&$7&*&&$ *&$$7$77** VW ES ,QWHUQDO WDQGHP UHSHDW 7**&$$*&$* *&*&7**$*$ &**7*&$*&* *&7*77*&&* *7*&7*7*&& $**&&&$7** &&7*$&&&&* *$*&$**7** 7**&&$7&*& &$*&$$7**& **&$$*&$** &*&7**$*$& **7*&$*&** &7*77*&&** 7*&7*7*&&$ **&&&$7**& &7*$&&&&** $*&$**7**7 **&&$7&*&& $*&$$7$77* *7**&$$*&$ **&*&7**$* $&**7*&$*& **&7*77*&& **7*&7*7*& &$**&&&$7* *&&7*$&&&& **$*&$**7* *7**&&$7&* &&$*&$$7$7 7**7**&$$* &$**&*&7** $*$&**7*&$ *&**&7*77* &&**7*&7*7 *&&$**&&&$ 7**&&7*$&& &&**$$&$** 7**7**&&$7 &*&&$*&$$7 $77**7**&$ $*&$**&*&7 **$*$&**7* &$**&*&7*7 7*&&**7*&7 *7*&&$**&& &$7**&&7*$ &&&&**$*&$ **7**7**&& $7&*&&$*&$ $7**&**7** &$$*&$**&* &7**$*$&** 7*&$*&**&7 *77*&&**7* &7*7*&&$** &&&$7**&&7 *$&&&&**$& &$**7**7** &&$7&*&&$* &&$&*$7**& **&$$*&$** &*&7**$*$& **7*&$*&** &7*77*&&** 7*&7*7*&&$ **&&&$7**& &7*$&&&&*& $*&$**7**7 **&&$7&*&& $*&$$7**&* *7**&$$*&$ **&*&7**$* $&**7*&$*& **&7*77*&& **7*&7*7*& &$**&&&$7* *&&7*$&&&& **$*&$**7* *7**&&$7&* &&$*&&$&*$ 7**&**&$$* &$**&*&7** $*$&**7*&$ *&**&7*77* &&**7*&7*7 *&&$**&&&$ 7**&&7*$&& &&**$*&$** 7**7**&&$7 &*&&$*&$$7 **&**7**&$ $*&$**&*&7 **$*$&**7* &$*&**&7*7 7*&&**7*&7 *7*&&$**&& &$7**&&7*$ &&&&**$*&$ **7**7**&& $7&*&&$*&$ $7**&**7** &$$*&$**&* &7**$*$&** 7*&$*&**&7 *77*&&**7* &7*7*&&$** &&&$7**&&7 *$&&&&**$* &$**7**7** &&$7&*&&$* &$$7**&**7 **&$$*&$** &*&7**$*$& **7*&$*&** &7*77*&&** 7*&7*7*&&$ **&&&$7**& &7*$&&&7** $&&$**7&*7 **&&$7&*&& $*&$$7**&* *7**&$$*&$ **&*&7**$* $&**7*&$*& **&7*77*&& **7*&7*7*& &$**&&&$7* *&&7*$&&&& **$*&$**7* *7**&&$7&* &&$*&$$7$* &**7**&$$* &$**&*&7** $*$&**7*&$ *&**&7*77* &&**7*&7*7 *&&$**&&&$ 7**&&7*$&& &&**$&&$** 7**7**&&$7 &*&&$*&&$& *$7**&**&$ $*&$**&*&7 **$*$&**7* &$*&**&7*7 7*&&**7*&7 *7*&&$**&& &$7**&&7*$ &&&&**$*&$ **7&*7**&& $7&*&&$*&& $&*$7**&** &$$*&$**&* &7**$*$&** 7*&$*&**&7 *77*&&**7* &7*7*&&$** &&&$7**&&7 *$&&&&**$* &$**7**7** &&$7&*&&7* &$$7**&**7 **&$$*&$** &*&7**$*$& **7*&$*&** &7*77*&&** 7*&7*7*&&$ **&&&$7**& &7*$&&&&** $*&$**7**7 **&&$7&*&& $*&$$7**&* *&**&$**&& **&*&7**$* $*&$77*77* &&&$*77$7& 7&*&&&7*$7 &&**&*77** &&*&*77*$& &$$&*$&&$& &7&*7&*&&7 7**&&7*&&7 &**&**$&*7 &&7*&*&7** $7*&$*7*$$ $$$***$77* &&*&$&*&Q& &**&&77*$7 &$$$$*$$&& $$7&*&&*7$ 77&&&*$$&* &$&$7&&&$7 &*&*77*&&* $&&$&*&*&$ $*7**77&*& *7*&7***77 7777&&$*7* &*$&7&&&$& &&$*&*&$$* &$777*$7*$ &*&&$7*$&* &$*77&***$ 7*$*&$**&$ &***77*77$ &$*&7&777& *&$*$*7*** &*7&$&&*$$ &7&*$$*&&& *&$*7**$$& *&7&&&&&&$ *&&7&*&$*& *77***$&&* 7$7&&7&&$* *&$7&$***$ 7*$$$$***& &$$$&&*7&& &&7$&77&$$ &7&$$$&*&& **$&&$**&* 7&777*&$7* &$77&*&&*$ 77&*&7**$* &*7*$&&77* $7*&*&&&$* &&&$$&*&$& *$***$*$7& $*$**&***& $$*&$*&&*7 $$$&**7&&& *$7&**$7&* 7*&7*7&$&& **7&&&7&&* &$&$*&$$7& *77&*$**7* &*&*&7&&&* $$&$*&*&*$ 7*&*&7*&$7 77*&&&&7&$ *77**$***7 $$$$&*&&&* &*7$&&$*7$ 7&*****&** &&7&&&**$7 &&7**7$&*& &&$&**&7*& &*$&&7**&$ *&*7&&$*&$ &&*7*$7*&* **$$&$$*$7 *$**$&&&&7 7&*&$****& $*&**$7*$7 77&&&**&$7 7&$$&*$$*$ **$*&7&*&$ 7**77*$7** $*&7$77*&& 7&$*7*$**& 7&$*7&**7* $&7$&&7*$* &*7&**&$** *$77**7*7$ $*7$$&&777 $&7*$&$*&* $*77$*&&&$ &7777**&7* 777777$&$& $$$7&&&7*& &7&&&&7&$& ES ,55 &7***&$&*& $*7&&&7&** 7*&*&$7&$$ 7&*$7*&$&& *$7**&*$&7 *&$$&*7&$& &$*$7&**7$ *$7&& 2n

PAGE 75

)LJXUH 3UHGLFWHG DPLQR DFLG VHTXHQFH RI 3WK$ 7KH LQWHUQDO WDQGHP UHSHDWV DUH DOLJQHG DQG QXPEHUHG IRU FRPSDULVRQ /HWWHUV $ WKURXJK UHSUHVHQW GLIIHUHQW W\SHV RI UHSHDWV LQGLFDWHV WKH UHSHDWV XQLTXH IRU 3WK$ SURWHLQ %ROG DQG XQGHUOLQHG OHWWHUV LQGLFDWH DQ DPLQR DFLG GHOHWLRQ VXEVWLWXWLRQ RU VSHFLILF FRPELQDWLRQV RI DPLQR DFLG UHVLGXHV ZLWKLQ D JLYHQ UHSHDW WKDW LV XQLTXH WR 3WK$ DQG WKH FRPELQDWLRQ LV QRW IRXQG LQ DQ\ UHSHDW LQ DQ\ SUHYLRXVO\ SXEOLVKHG PHPEHU RI WKH JHQH IDPLO\f 7KUHH QXFOHDU ORFDOL]DWLRQ FRQVHQVXV VHTXHQFHV DUH GRXEOH XQGHUOLQHG

PAGE 76

0'3,565736 3$5(//3*34 3'*9437$'5 *9633$**3/ '*/3$55706 575/3633$3 63$)6$*6)6 '//54)'36/ )176/)'6/3 3)*$++7($$ 7*(:'(946* /5$$'$3337 059$97$$53 35$.3$3555 $$436'$63$ $49'/57/*< 64444(.,.3 .95679$4++ ($/9*+*)7+ $+,9$/64+3 $$/*79$9.< 4'0,$$/3($ 7+($,9*9*. 4:6*$5$/($ //79$*(/5* 33/4/'7*4/ /.,$.5**97 $9($9+$:51 $/7*$3/1 /73(4:$,$ 61,**.4$/( 7945//39/& 4$+* $f /73(4:$,$ 61B**.4$/( 7945//39/& 4$+* %f /73(4:$,$ 61,**.4$/( 7945//39/& 4$+* $f /73(4:$,$ 61,**.4$/( 7945//39/& 4$+* $f /73(4:$,$ 61,**.4$/( 794$//39/& 4$+* &f /73(4:$,$ 61***.4$/( 7945//39/& 4$+* 'f /73'4:$,$ 6+'**.4$/( 7945//39/& 4$+* (f /7344:$,$ 61***.4$/( 7945//39/& 4$+* )f /73(4:$,$ 6+'**.4$/( 7945//39/& 4$+* *f /73(4:$,$ 61***.4$/( 7945//39/& 4$+* 'f /73(4:$,$ 61***.4$/( 7945//39/& 4$+* 'f /73(4:$,$ 61***.4$/( 7945//39/& 4$+* 'f /7/'4:$,$ 61***.4$/( 7945//39/& 4$+* +f /73(4:$,$ 616**.4$/( 7945//39/& 4$+* ,f /73'4:$,$ 6+'**.4$/( 7945//39/& 4$+* (f /73(4:$,$ 6+'**.4$/( 7945//39/& 4$+* *f /73(4:$,$ &1***.4$/( 7945//39/& 4$+* -f /73(4:$,$ 61***53$/( 6,9$4/653' 3$/$ .f $ /71'+/9$/$ &/**53$/'$ 9..*/3+$3$ /,.57155,3 (576+59$'+ $4:59/*)) 4&'6+3$4$) ''$074)*+6 5+*//4/)55 9*97(/($56 *7/33$645: '5,/4$6*0. 5$.3637674 73'4$6/+$) $'6/(5'/'$ 3637+(*'45 5$665.5656 '5$97*36$4 46)(95$3(4 5'$/+/3/6: 59.53576,* **/3'3*737 $$'/$$6679 05(4'('3)$ *$$'')3$)1 (((/$:/0(/ /34r

PAGE 77

LV LQ WKH FHQWUDO SRUWLRQ RI WKH SURWHLQ FKDUDFWHUL]HG E\ QHDUO\ LGHQWLFDO WDQGHPO\ DUUDQJHG DPLQR DFLG UHSHDWV $V VKRZQ LQ )LJ WKHVH UHSHDWV PD\ EH FODVVLILHG LQWR W\SHV $.f GHSHQGLQJ RQ VOLJKW GLIIHUHQFHV LQ VHTXHQFH $PRQJ WKHP UHSHDW W\SHV % ( + DUH RQO\ IRXQG LQ 3WK$ DQG QRW LQ WKH SUHGLFWHG SHSWLGH VHTXHQFHV RI DQ\ RWKHU SXEOLVKHG IDPLO\ PHPEHU 5HSHDW W\SH %f UHVXOWHG IURP DQ DPLQR DFLG GHOHWLRQ DQG UHSHDW W\SH -f UHVXOWHG IURP DQ DPLQR DFLG VXEVWLWXWLRQ QHLWKHU WKLV GHOHWLRQ QRU WKLV VXEVWLWLRQ KDV EHHQ REVHUYHG LQ DQ\ RWKHU PHPEHU RI WKH JHQH IDPLO\ 5HSHDW W\SHV ( UHSHDWV DQG f DQG + UHSHDW f UHVXOWHG IURP XQLTXH FRPELQDWLRQV RI DPLQR DFLGV LQ VSHFLILF SRVLWLRQV ,Q DGGLWLRQ WR WKH SUHVHQFH RI XQLTXH UHSHDWV 3WK$ GLIIHUV IURP LWV KRPRORJXHV LQ WKH DUUDQJHPHQW RI UHSHDW W\SHV ZLWKLQ WKH UHJLRQ *HQH UHSODFHPHQW DQG VHOHFWLRQ RI SWK$ UHFRPELQDQWV LQ ; FLWUL ; FLWUL PXWDQW ;FO SWK$QSW,VDFf ZDV JHQHUDWHG IURP ZLOG W\SH VWUDLQ E\ PDUNHU H[FKDQJH XVLQJ S8)
PAGE 78

.E .E A r "+ .E )LJXUH 6RXWKHUQ K\EULGL]DWLRQ RI %DP5O GLJHVWHG WRWDO '1$ IURP ; FLWUL VWUDLQ DQG LWV SWK$ PXWDQW GHULYDWLYHV 7KH EORW ZDV SUREHG ZLWK WKH LQWHUQDO NE 6WXO+LQGL IUDJPHQW FRQWDLQLQJ WKH WDQGHPO\ UHSHDWHG UHJLRQf RI SWK$ 6WUDLQ FDUULHV IRXU %DP+O IUDJPHQWV ZKLFK K\EULGL]H WR SWK$ WKH ODUJHVW IUDJPHQW LQGLFDWHG E\ WKH DUURZ HQFRGHV SWK$ 1RW VKRZQ LQ ODQH DUH WZR DGGLWLRQDO K\EULGL]LQJ IUDJPHQWV DQG NE LQ VL]Hf UHVXOWLQJ IURP WKH DGGLWLRQDO %DP+O VLWH LQ QSWOVDF /DQH ODQH ;FO SWK$QSW,VDFf ODQHV QSW,VDFH9?F?R£ GHULYDWLYHV RI ;FO ;F6O ;F6 ;F6 ;F6 ;F6 ;F6 ;F6 ;F6 ;F6 ;F6 ;F6 ;F6 ;F6 ;F6 ;F6 ;F6 DQG ;F6 UHVSHFWLYHO\

PAGE 79

7DEOH 3KHQRW\SHV RI ;DQWKRPRQDV FLWUL PDUNHUHYLFWHG VWUDLQV RQ FLWUXV DQG EHDQ +RVW 2 /L F/OOOR FLWUXV EHDQ ;F & +5 ;FO ;F6O ;F6 ;F6 t ;F6 F +5 ;F6 F ;F6 t ;F6 Z& Z+5 ;F6 ;F6 t ;F6 +5 Z+5 ;F6 Z+5 Z+5 ;F6 t ;F6OO Z+5 ;F6r r $GGLWLRQDO VWUDLQV ;F6 ;F6 ;F6 ;F6,2 ;F6 ;FVO ;F6 ;F6 ;F6 ;F6 DQG ;F6f & FDQNHU OHVLRQV Z& ZHDN FDQNHU +5 K\SHUVHQVLWLYH UHDFWLRQ Z+5 ZHDN K\SHUVHQVLWLYH UHDFWLRQ QR V\PSWRPV

PAGE 80

$V VKRZQ LQ )LJ WKH GLIIHUHQW GHULYDWLYHV RI ;FO H[KLELWHG GLIIHUHQW VL]HG EDQGV LQFOXGLQJ WKRVH VLPLODU LQ VL]H WR SWK$f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f )LYH GHULYDWLYHV UHYHUWHG EDFN WR WKH SDUHQWDO SKHQRW\SH LQGXFLQJ VWURQJ FDQNHU OHVLRQV RQ FLWUXV ;F6O ;F6 ;F6 DQG ;F6f 7ZR GHULYDWLYHV FDXVHG ZHDN FDQNHU ;F6 DQG ;F6f 6L[ GHULYDWLYHV HOLFLWHG D SUHYLRXVO\ XQUHSRUWHG +5 RQ JUDSHIUXLW ;F6 ;F6 DQG ;F6f (OHYHQ GHULYDWLYHV FDXVHG QR SDWKRJHQLF V\PSWRPV ,Q FRQWUDVW WR ZLOG W\SH VWUDLQ ;F6 UHWDLQHG WKH DELOLW\ WR HOLFLW K\SHUSODVWLF FDQNHUV RQ FLWUXV EXW ORVW WKH DELOLW\ WR HOLFLW +5 RQ EHDQ LQGLFDWLQJ WKDW WKHVH WZR SOHLRWURSLF IXQFWLRQV DUH VHSDUDWHO\ HQFRGHG RQ SWK$ ,QWUDJHQLF UHFRPELQDWLRQ RI SWK$ LQ ; FDPSHVWULV DQG ( FROL 7R GHPRQVWUDWH WKH RFFXUUHQFH RI LQWUDJHQLF UHFRPELQDWLRQ SODVPLG S8)
PAGE 81

HYLFWHG GHULYDWLYHV ZHUH REWDLQHG 7KH IUHTXHQFLHV RI PDUNHUHYLFWLRQ RI S8)< LQ '+D DQG -0 ZHUH s [ s [ n DQG s [ UHVSHFWLYHO\ 7R H[DPLQH WKH UHFRPELQDWLRQ HYHQWV SODVPLG '1$V ZHUH DQDO\]HG IURP PDUNHUHYLFWHG VWUDLQV IURP DQG VWUDLQV IURP '+D $V VKRZQ LQ )LJ GLIIHUHQW VL]HG %DP+O IUDJPHQWV ZHUH REVHUYHG GHPRQVWUDWLQJ WKH JHQHUDWLRQ RI QHZ SWK$ KRPRORJXHV E\ LQWUDJHQLF UHFRPELQDWLRQ 6RXWKHUQ K\EULGL]DWLRQV RI WRWDO '1$V H[WUDFWHG IURP WKHVH VWUDLQV SUREHG ZLWK WKH LQWHUQDO 6WXO+LQGL IUDJPHQW RI SWK$ VKRZHG RQO\ WKH SUHVHQFH RI WKH VDPH EDQGV DV WKRVH JHQHUDWHG IURP WKH SODVPLG GLJHVWLRQV GDWD QRW VKRZQf 7KH SODVPLGV GHULYHG IURP WKH S&6 VHULHVf DQG '+D S'6 VHULHVf PDUNHU HYLFWHG VWUDLQV ZHUH LQGLYLGXDOO\ LQWURGXFHG LQWR ; FLWUL VWUDLQ % D QRQSDWKRJHQLF SWK$7QJXV$ GHULYDWLYH RI ZKLFK LV YLUXOHQW RQO\ RQ FLWUXVf YLUXOHQW RQ FLWUXV DQG EHDQf DQG ; FDPSHVWULV SY PDOYDFHDUXP ;FPO YLUXOHQW RQO\ RQ FRWWRQf 7KH UHVXOWLQJ WUDQVFRQMXJDQWV ZHUH LQRFXODWHG RQWR FLWUXV EHDQ DQG FRWWRQ SODQWV )LJ f 6LPLODU WR WKH UHVXOWV REVHUYHG ZLWK GHULYDWLYHV RI ;FO WKHVH SWK$ UHFRPELQDQWV H[KLELWHG DOWHUUHG SKHQRW\SHV DQGRU KRVW VSHFLILFLWLHV )RU H[DPSOH S&6 FRQIHUUHG WR % DQG WR WKH DELOLW\ WR FDXVH VWURQJHU FDQNHU OHVLRQV RQ FLWUXV WKDQ WKDW FRQIHUUHG E\ S=LW SWK$f DQG D ZHDN +5 RIWHQ SUHFHHGHG WKH FDQNHU V\PSWRPV 8QOLNH SWK$ RQ S=LW KRZHYHU WKH UHFRPELQDQW JHQH RQ S&6 FRQIHUUHG WR % RU WKH DELOLW\ WR HOLFLW FDQNHU OHVLRQV RQ FLWUXV EXW QRW +5 RQ EHDQ 7KHVH UHVXOWV FRQILUPHG WKRVH LQGLFDWHG E\ UHFRPELQDQW ; FLWUL PXWDQW ;F6 UHIHU

PAGE 82

)LJXUH $ ,QWUDJHQLF UHFRPELQDQWV RI SWK$ JHQHUDWHG LQ ; FDPSHVWULV SY FLWUXPHOR 3ODVPLG '1$V ZHUH GLJHVWHG ZLWK %DP8O DQG IUDFWLRQDWHG LQ b DJDURVH JHO /DQH S=LW SWK$f? ODQH S8)< SWK$QSWVDFf ODQH S&6 S&6 S&6 S&6 S&6 S&6 S&6O S&6 S&6 S&6 S&6OO S&6 S&6 S&6 S&6 S&6 DQG S&6 UHVSHFWLYHO\ % ,QWUDJHQLF UHFRPELQDQWV RI SWK$ JHQHUDWHG LQ ( FROL 3ODVPLG '1$V ZHUH GLJHVWHG ZLWK %DP8O DQG IUDFWLRQDWHG LQ b DJDURVH JHO /DQH S=LW ODQH S8)< ODQHV S'6 S'6 S'6 S'6 S'6O S'6O2 S'6 S'6 S'6 S'6 S'6 S'6 S'6OO DQG S'6 UHVSHFWLYHO\

PAGE 84

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f LQ S8)< LV LQGLFDWHG E\ WKH LQYHUWHG WULDQJOH D 3ODVPLGV S&6 S&6 S&6 S&6 S&6 S&6 S&6 S&6 S'6 S'6 DQG S'6 FRQIHUHG VLPLODU SKHQRW\SHV E 3ODVPLGV S&6 S&6 S&6,2 DQG S'6O2 FRQIHUHG VLPLODU SKHQRW\SHV F 3ODVPLGV S&6 S'6 S'6 S'6 DQG S'6OO FRQIHUHG VLPLODU SKHQRW\SHV QR DFWLYLW\ & FDQNHU OHVLRQV P& LQWHUPHGLDWH OHYHO FDQNHU Z& ZHDN FDQNHU +5 K\SHUVHQVLWLYH UHDFWLRQ Z+5 ZHDN +5 U:6 UHGXFHG ZDWHUVRDNLQJ DVVD\HG LQ DQG ;FPOf

PAGE 85

3ODVPLG 38)< 6VS S=LW f S&6 f S'6 f S&6 f S&6 f S&6 f S'6 f S&6 f S&6 f S&6 f S&6 f S'6 fE S&6 fr $FWLYLW\ LQ % RU ;FP &LWUXV %HDQ &RWWRQ F +5 U:6 &Z+5 +5 U:6 & +5 U:6 F U:6 P& U:6 Z& +5 Z& +5 Z& Z+5 Z+5 +5 +5 +5 +5 +5 U:6 +5 Z+5 U:6 YR

PAGE 86

DERYHf WKH DELOLW\ WR FRQIHU K\SHUSODVWLF FDQNHUV RQ FLWUXV DQG WKH +5 RQ EHDQ DUH LQGHSHQGHQWO\ HQFRGHG RQ SWK$ 3ODVPLGV S&6O DQG S&6 FRQIHUUHG DELOLW\ WR HOLFLW DQ +5 RQ FLWUXV DQG EHDQ WR % DQG DQG DELOLW\ WR HOLFLW DQ +5 RQ WHVWHG FRWWRQ OLQHV $F $F%O $F% $F% $F% $FE $FE $F%,Q DQG $F%,Qf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f 'HOHWLRQV LQYROYLQJ WKH WK WK DQG WK UHSHDWV DOZD\V UHGXFHG RU DEROLVKHG WKH FDQNHU HOLFLWLQJ DELOLW\ RI SWK$ S&6 S&6 S'6 S&6O S&6 S&6O S&6f 'HOHWLRQV LQYROYLQJ WKH WK WK DQG WK UHSHDWV DEROLVKHG DOO DYLUXOHQFH VSHFLILFLW\ WKHVH GHULYDWLYHV FRQIHUUHG DELOLW\ WR LQGXFH DQ +5 RQ DOO FLWUXV EHDQ DQG FRWWRQ OLQHV WHVWHG

PAGE 87

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f EXW XVLQJ DQWL3WK$ VHUXP UHYHDOHG WKDW D N' SURWHLQ ZDV FRQVWLWXWLYHO\ H[SUHVVHG LQ ; FLWUL DQG ; FDPSHVWULV SY FLWUXPHOR WUDQVFRQMXJDQWV FRQWDLQLQJ SWK$ UHIHU WR &KDSWHU f 'HVSLWH IHZ GLIIHUHQFHV LQ WKH SUHGLFWHG DPLQR DFLG VHTXHQFHV RI DOO PHPEHUV RI WKH DYUSWK JHQH IDPLO\ SWK$ KDV D XQLTXH DELOLW\ WR LQGXFH FHOO GLYLVLRQV RQ FLWUXV OHDI PHVRSK\OO LQ DGGLWLRQ WR LWV SOHLRWURSLF DELOLW\ WR FRQIHU DYLUXOHQFH WR SDWKRJHQV RI FRWWRQ DQG EHDQ 6ZDUXS HW DO f %RWK IXQFWLRQV ZHUH SUHYLRXVO\ IRXQG WR EH ORFDOL]HG WR WKH UHSHDWHG UHJLRQ RI SWK$ UHIHU WR &KDSWHU f '1$ VHTXHQFH DQDO\VLV UHYHDOHG WKDW WKHUH ZHUH ILYH VSHFLILF UHSHDWV RQO\ IRXQG LQ SWK$ EXW QRW LQ RWKHU SXEOLVKHG PHPEHUV RI WKH DYUSWK JHQH IDPLO\ VHTXHQFHG WR GDWH ^DYUE DYU%V DYU%V DQG DYU;DO2f %\ DQDO\]LQJ LQWUDJHQLF UHFRPELQDQWV ZLWK %JO, DQG 6VSO UHVWULFWLRQ HQ]\PH GLJHVWLRQ ZH ZHUH DEOH WR DSSUR[LPDWHO\ ORFDWH WKH

PAGE 88

UHJLRQV LPSRUWDQW IRU DYLUXOHQFH DQG SDWKRJHQLF VSHFLILFLWLHV 7KH ILUVW VL[ UHSHDWV RI SWK$ DSSHDU WR EH HVVHQWLDO IRU WKH SDWKRJHQLFLW\ IXQFWLRQ DQG DQ\ VSHFLILFLW\ DYLUXOHQFH RU SDWKRJHQLFLW\f DW DOO 'HOHWLRQV LQYROYLQJ UHSHDWV WR UHGXFHG RU DEROLVKHG FDQNHU HOLFLWLQJ DELOLW\ RI SWK$ RQ FLWUXV 'HOHWLRQV LQYROYLQJ UHSHDWV WR GHWHUPLQHG DYLUXOHQFH VSHFLILFLW\ 7KHVH ILQGLQJV PD\ DOVR KHOS WR H[SODLQ SUHYLRXVO\ SXEOLVKHG GDWD RQ DYU%V GHOHWLRQ GHULYDWLYHV +HUEHUV HW DO %RQDV HW DO f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f 6LPLODUO\ DPLQR DFLG UHSHWLWLYH PRWLIV FDOOHG WHWUDWULFRSHSWLGHV 735f IRXQG LQ SURWHLQV HQFRGHG E\ PDQ\ PLWRWLF JHQHV HJ &'& &'& QXF RI \HDVW %LP$ RI $VSHUJLOOXVf DUH LPSOLFDWHG WR SDLU ZLWK :' UHSHDWV IRXQG LQ PDQ\ SURWHLQV LQFOXGLQJ WKH VXEXQLW RI *SURWHLQV DQG VRPH WUDQVFULSWLRQDO IDFWRUV *RHEO DQG
PAGE 89

LQWUDJHQLF UHSHDWV 6LQFH WKH SURGXFWLRQ RI OHYDQVXFUDVH HQFRGHG E\ VDF LV OHWKDO WR JUDPn QHJDWLYH EDFWHULD LQ WKH SUHVHQFH RI VXFURVH PDUNHU HYLFWHG VWUDLQV FDQ EH VHOHFWHG E\ WKH QSWOVDF PDUNHU 5LHG DQG &ROOPHU f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f \LHOGHG UHFRPELQDQWV DW D IUHTXHQF\ RI s [ n ZKLOH VWUDLQ -0 UHF$f \LHOGHG UHFRPELQDQWV DW D IUHTXHQF\ RI s [ n 6LQFH UHFRPELQDWLRQ LQYROYLQJ VPDOO UHSHDWHG VHTXHQFHV OHVV WKDQ ONEf LV DIIHFWHG RQO\ VOLJKWO\ E\ UHF$ PXWDWLRQV 3HWHV DQG +LOO f UHFRPELQDWLRQ RI SWK$ ZDV H[SHFWHG WR EH UHF$ LQGHSHQGHQW 7KH LQWUDJHQLF UHFRPELQDWLRQ IUHTXHQFLHV RI SWK$ REVHUYHG LQ WKH ( FROL ZHUH KLJKHU WKDQ WKRVH REVHUYHG LQ ; FDPSHVWULV SY FLWUXPHOR s [ f DQG ; FLWUL [ ff 7KLV PD\ EH GXH WR GLIIHUHQFHV LQ WKH FRS\ QXPEHU RI SODVPLGV LQ ZKLFK LQWUDJHQLF UHFRPELQDWLRQ RFFXUUHG 5HFRPELQDWLRQ RI SWK$ LQ ; FLWUL VWUDLQ ;FO LQYROYHG D QDWLYH SODVPLG ZKLOH UHFRPELQDWLRQ LQ ( FROL DQG ; FDPSHVWULV SY FLWUXPHOR LQYROYHG S8)< ZKLFK DSSHDUHG EDVHG RQ RXU SODVPLG '1$

PAGE 90

H[WUDFWLRQVf WR UHSOLFDWH DW D KLJKHU FRS\ QXPEHU LQ ( FROL WKDQ LQ ;DQWKRPRQDV +RZHYHU VLQFH ZH GLG QRW GHWHUPLQH SODVPLG FRS\ QXPEHU WKHVH UHVXOWV PD\ EH GXH WR GLIIHUHQW JHQHWLF EDFNJURXQGV LQ HDFK VWUDLQ ,Q DQLPDO SDWKRJHQV VRPH SDWKRJHQLFLW\ JHQHV HQFRGH VXUIDFH SURWHLQV ZLWK D GLVWLQFWLYH DUUDQJHPHQW RI WDQGHP UHSHDWV ZKLFK PD\ SURYLGH D SURWHFWLYH VWUDLQVSHFLILF FRQIRUPDWLRQDO HSLWRSH IRU HYDVLRQ RI KRVW LPPXQLW\ +RIW HW DO 0F&RQNH\ HW DO +ROOLQJVKHDG HW DO f ,QWUDJHQLF UHFRPELQDWLRQ EHWZHHQ KRPRORJRXV UHSHDWV KDV EHHQ IRXQG WR EH UHVSRQVLEOH IRU DQWLJHQLF YDULDWLRQV LQ WKH SDWKRJHQV VXFK DV 6WUHSWRFRFFXV 1HLVVHULD 6DOPRQHOOD 3ODVPRGLXP DQG 7U\SDQRVRPD 0F&RQNH\ HW DO +ROOLQJVKHDG HW DO +DJEORP HW DO )UDQNHO HW DO f 7KH UHVXOWLQJ DQWLJHQLF YDULDELOLW\ DOORZV SDWKRJHQV WR HVFDSH WKH KRVW LPPXQH UHVSRQVH ,Q WKLV VWXG\ ZH GHPRQVWUDWHG WKDW LQWUDJHQLF UHFRPELQDWLRQ RI D VLQJOH SODQW EDFWHULDO JHQH FDQ f DOWHU KRVW VSHFLILFLW\ f JHQHUDWH QHZ SDWKRJHQLF SKHQRW\SHV DQG f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

PAGE 91

3UHYLRXVO\ UHSRUWHG QDWXUDO PHFKDQLVPV IRU UDFH FKDQJH LQFOXGHG GHOHWLRQV 'H )H\WHU HW DO f SRLQW PXWDWLRQV .RED\DVKL HW DO -RRVWHQ HW DO f RU WUDQVSRVRQ LQDFWLYDWLRQ .HDUQH\ HW DO f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f DQG K\SHUVHQVLWLYH UHDFWLRQV PD\ VKDUH D YHU\ VLPLODU VLJQDO WUDQVGXFWLRQ SDWKZD\ DV LQLWLDOO\ VXJJHVWHG E\ .OHPHQW f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f KDV WKH DELOLW\ WR HOLFLW D VWURQJ +5 RQ SODQWV IURP WD[RQRPLFDOO\ GLVWLQFW SODQW IDPLOLHV VXFK DV 5XWDFHDH FLWUXVf /HJXPLQRVDH EHDQf DQG 0DOYDFHDH FRWWRQf VXJJHVWLQJ WKDW WKH FRUUHVSRQGLQJ

PAGE 92

UHVLVWDQFH JHQHV PD\ EH ZLGHVSUHDG LQ GLIIHUHQW SODQW VSHFLHV 7RPDWR GLVHDVH UHVLVWDQFH JHQH 3WR ZKLFK HQFRGHV D SURWHLQ NLQDVH K\EULGL]HV ZLWK PXOWLSOH '1$ IUDJPHQWV IRXQG LQ D YDULHW\ RI GLIIHUHQW SODQWV LQGLFDWLQJ WKH SUHVHQFH RI 3WR KRPRORJXHV 0DUWLQ HW DO f 'XULQJ WKH HYROXWLRQDU\ SURFHVV PHPEHUV RI WKH DYUSWK JHQH IDPLO\ PLJKW KDYH PRYHG KRUL]RQWDOO\ DPRQJ GLYHUVH [DQWKRPRQDGV SDWKRJHQLF WR D ZLGH UDQJH RI SODQWV *DEULHO HW DO 'H )H\WHU HW DO f WKHUHE\ HQDEOLQJ LQWHUDFWLRQV ZLWK PHPEHUV RI WKH VDPH SODQW JHQH IDPLO\ LQ D ZLGH UDQJH RI SODQWV $V D UHVXOW PHPEHUV RI WKLV DYUSWK JHQH IDPLO\ PD\ KDYH DFFHOHUDWHG WKH HYROXWLRQ DQG GLYHUVLW\ RI WKH FRUUHVSRQGLQJ SODQW 5 JHQH IDPLO\ 7KH JHQHIRUJHQH FRQFHSW LV D PDMRU PRGHO IRU UHVHDUFK RQ WKH UHFLSURFDO HYROXWLRQ EHWZHHQ SODQWV DQG SDUDVLWHV 7KRPSVRQ DQG %XUGRQ f ,Q WKH SDWKRJHQ DYLUXOHQFH JHQHV PD\ EH PXWDWHG WR RYHUFRPH UHVLVWDQFH E\ HYDGLQJ KRVW UHFRJQLWLRQ ,Q WKH SODQW UHVLVWDQFH JHQHV PD\ DOVR HYROYH WR UHFRJQL]H QHZ DYLUXOHQFH JHQHV UHVXOWLQJ LQ QHZ UHVLVWDQFH JHQHV 5HFHQW JHQHWLF PDSSLQJ GDWD KDYH VKRZQ WKDW GLVHDVH UHVLVWDQFH ORFL HVSHFLDOO\ WKRVH FRUUHVSRQGLQJ WR GLIIHUHQW UDFHV RI WKH VDPH SDWKRJHQf DUH RIWHQ FOXVWHUHG ZLWKLQ VPDOO JHQHWLF LQWHUYDOV VXJJHVWLQJ WKDW LQWUDORFXV UHFRPELQDWLRQ PD\ EH UHVSRQVLEOH IRU JHQHUDWLQJ UHVLVWDQFH VSHFLILF IRU QHZ UDFHV RI SODQW SDWKRJHQV 'LFNLQVRQ HW DO 0DUWLQ HW DO 3U\RU f ,QWUDJHQLF UHFRPELQDWLRQ RI SWK$ D PHPEHU RI DQ DYU JHQH IDPLO\f GHPRQVWUDWHG LQ WKLV UHVHDUFK SURYLGHV FRPSHOOLQJ HYLGHQFH WKDW LQWUDJHQLF DQG LQWUDORFXV UHFRPELQDWLRQ PD\ SOD\ DQ LPSRUWDQW UROH GXULQJ WKH UHFLSURFDO HYROXWLRQ RI JHQHIRUJHQH LQWHUDFWLRQV

PAGE 93

&+$37(5 3/$17 18&/($5 7$5*(7,1* 6,*1$/6 (1&2'(' %< $ )$0,/< ;$17+2021$6 $9,58/(1&(3$7+2*(1,&,7< *(1(6 ,QWURGXFWLRQ ,QWHUDFWLRQV RI PLFURELDO SODQW SDWKRJHQV FDUU\LQJ DYLUXOHQFH DYUf JHQHV ZLWK KRVW SODQWV FDUU\LQJ VSHFLILF UHVLVWDQFH 5f JHQHV UHVXOWV LQ SODQW GHIHQVH UHVSRQVHV RIWHQ REVHUYHG DV D K\SHUVHQVLWLYH UHDFWLRQ +5f WKDW LV FKDUDFWHUL]HG E\ UDSLG SODQW FHOO GHDWK DW WKH VLWH RI LQIHFWLRQ 6HYHUDO PRGHOV KDYH EHHQ SURSRVHG WR H[SODLQ WKH JHQHWLFDOO\ VSHFLILF UHFRJQLWLRQ EHWZHHQ PLFURELDO DYU JHQHV DQG SODQW 5 JHQHV LQ WKHVH JHQHIRUJHQH LQWHUDFWLRQ *DEULHO DQG 5ROIH f DQG GLIIHUHQW PRGHOV PD\ EH DSSOLFDEOH WR GLIIHUHQW V\VWHPV ,Q WKH HOLFLWRUUHFHSWRU PRGHO DYU JHQHV HQFRGH D ORZ PROHFXODU ZHLJKW VLJQDO PROHFXOH HOLFLWRUf ZKLFK LV SHUFHLYHG E\ D VHQVRU HQFRGHG E\ WKH FRUUHVSRQGLQJ 5 JHQHV DQG UHVXOWV LQ DQ +5 ,Q 3VHXGRPRQDV V\ULQJDH SY WRPDWR DYU' HQFRGHV DQ HQ]\PH LQYROYHG LQ WKH V\QWKHVLV RI D ORZ PROHFXODU ZHLJKW JO\FROLSLG HOLFLWRU 0LGODQG HW DO 6PLWK HW DO f 7KLV HOLFLWRU LQGXFHV DQ +5 VSHFLILFDOO\ RQ VR\EHDQ SODQWV FRQWDLQLQJ WKH UHVLVWDQFH JHQH 5SJ .HHQ HW DO .HHQ DQG %X]]HO .RED\DVKL HW DO f $Q DOWHUQDWLYH PRGHO LV WKH GLPHU K\SRWKHVLV LQ ZKLFK WKH SURWHLQ SURGXFW RI DQ DYU JHQH LQWHUDFWV GLUHFWO\ ZLWK WKH SURWHLQ SURGXFW RI D SODQW 5 JHQH RU ZLWK WKH 5 JHQH LWVHOI (OOLQJERH f 7KH H[WUDFHOOXODU SHSWLGH HQFRGHG E\ DYU DQG DYU RI

PAGE 94

WKH IXQJDO SDWKRJHQ &ODGRVSRULXP IXOYXP FDQ GLUHFWO\ LQGXFH D K\SHUVHQVLWLYH UHVSRQVH +5f RQ WRPDWR FXOWLYDUV FDUU\LQJ WKH UHVLVWDQFH JHQH &I RU &I -RRVWHQ HW DO YDQ .DQ HW DO f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f 6LQFH WUDQVFULSWLRQDO DFWLYDWLRQ RI SODQW GHIHQVH JHQHV LV PRGXODWHG E\ SKRVSKRU\ODWLRQ )HOL[ HW DO
PAGE 95

PHPEUDQH WR QXFOHXV LV PRUH GLUHFW ,Q WKHVH FDVHV W\URVLQH NLQDVHV SLFN XS VLJQDOV IURP UHFHSWRUV LQ WKH PHPEUDQH DQG DFWLYDWH WUDQVFULSWLRQ IDFWRU VXEXQLWV LQ WKH F\WRSODVP ZKLFK DUH WUDQVSRUWHG LQWR WKH QXFOHXV DQG LQGXFH WUDQVFULSWLRQ +XQWHU .LVKLPRWR HW DO f ,Q ERWK FDVHV DFWLYDWHG SURWHLQ NLQDVHV DQGRU WUDQVFULSWLRQDO IDFWRUV QHHG WR HQWHU WKH QXFOHXV LQ RUGHU WR LQGXFH JHQH H[SUHVVLRQ ZKLFK PD\ OHDG WR GLIIHUHQW SK\VLRORJLFDO RXWFRPHV LQFOXGLQJ RQFRJHQHVLV DQG DSRSWRVLV 9LUDO SDWKRJHQV HQFRGH VRPH WUDQVFULSWLRQDO IDFWRUV WKDW FDQ GLUHFWO\ HQWHU WKH QXFOHXV PRGXODWH JHQH H[SUHVVLRQ RI KRVW FHOOV 7KH EDFWHULDO SODQW SDWKRJHQ $JUREDFWHULXP WXPDIDFLHQV DOVR HQFRGHV VHYHUDO SURWHLQV WDUJHWHG WR WKH SODQW FHOO QXFOHXV &LWRYVN\ DQG =DPEU\VNL f 7UDQVSRUW RI SURWHLQV LQWR WKH QXFOHXV LV DQ DFWLYH SURFHVV DQG UHTXLUHV WKDW SURWHLQV FRQWDLQ VXLWDEOH QXFOHDU ORFDOL]DWLRQ VHTXHQFHV 1/6V 1LJJ HW DO f 1/6V DUH XVXDOO\ VKRUW VWUHWFKHV RI DPLQR DFLGV FKDUDFWHUL]HG E\ EDVLF DPLQR DFLG UHVLGXHV DQG SUROLQH 1/6V DUH UHWDLQHG LQ WKH PDWXUH SURWHLQ PD\ EH ORFDWHG DW DQ\ SRVLWLRQ DV ORQJ DV WKH\ H[SRVHG RQ WKH SURWHLQ VXUIDFH DQG FDQ EH SUHVHQW LQ PXOWLSOH FRSLHV 1/6V DUH UHFRJQL]HG E\ WKH UHFHSWRU FDOOHG 1/6ELQGLQJ SURWHLQV ZKLFK GLUHFW 1/6FRQWDLQLQJ SURWHLQV LQWR QXFOHXV WKURXJK QXFOHDU SRUHV 6LOYHU f 1/6V IURP GLIIHUHQW RUJDQLVPV LQFOXGLQJ WKH SURNDU\RWH $JUREDFWHULXP DUH IXQFWLRQDO WKURXJKRXW WKH HXNDU\RWLF NLQJGRP 0RVW RI WKH FKDUDFWHUL]HG 1/6FRQWDLQLQJ SURWHLQV ZHUH UHSRUWHG IURP PDPPDOLDQ YLUXV DQG DQLPDO FHOOV 0DQ\ DUH RQFRJHQH SURGXFWV KRUPRQH UHFHSWRUV DQG WUDQVFULSWLRQDO IDFWRUV LQFOXGLQJ WKRVH LQYROYLQJ ODWHU VWHSV RI WKH 5DV VLJQDOLQJ SDWKZD\ VXFK DV WKH WUDQVFULSWLRQDO IDFWRU $3 -XQ)RVf .HUU HW DO f $ IHZ 1/6FRQWDLQLQJ SURWHLQV IURP $JUREDFWHULXP SODQW DQG SODQW YLUXV

PAGE 96

KDYH DOVR EHHQ FKDUDFWHUL]HG &LWRYVN\ DQG =DPEU\VNL 5HVWUHSR HW DO YDQ GHU .URO DQG &KXD 9DUDJRQD HW DO f 7R WKH EHVW RI P\ NQRZOHGJH KRZHYHU 1/6V HQFRGHG E\ DZ JHQHV RI SODQW SDWKRJHQV KDYH QRW EHHQ UHSRUWHG 5HFHQWO\ D ODUJH IDPLO\ RI DYLUXOHQFH DYUfSDWKRJHQLFLW\ SWKf JHQHV KDV EHHQ LGHQWLILHG LQ PDQ\ VSHFLHV DQG SDWKRYDUV RI ;DQWKRPRQDV D PDMRU JURXS RI SODQW EDFWHULDO SDWKRJHQV 0HPEHUV LQFOXGHV DZ%V DYU%V DQG DYU%V3 RI ; FDPSHVWULV SY YHVLFDWRULD %RQDV HW DO &DQWHURV HW DO f DZ%DYUE DZE DZ%OQ DZ%O2O DQG DZ% RI ; F SY PDOYDFHDUXP 'H )H\WHU DQG *DEULHO 'H )H\WHU HW DO f SWK$ RI ; FLWUL 6ZDUXS HW DO f DQG DYU[D DZ;D DQG DYU;DO2 RI ; RU\]DH SY RU\]DH +RSNLQ HW DO f 7KLV DZSWK JHQH IDPLO\ FRPSULVHV WKH PDMRULW\ RI GHVFULEHG ;DQWKRPRQDV DZ JHQHV DQG FRQVWLWXWHV D ODUJH SRUWLRQ RI DOO DZ JHQHV FORQHG WR GDWH *DEULHO HW DO f 7KH SWK$ JHQH RI ; FLWUL DQG DZE RI ; F SY PDOYDFHDUXP DUH SDUWLFXODUO\ LQWULJXLQJ EHFDXVH RI WKHLU SDWKRJHQLFLW\ IXQFWLRQV 6ZDPS HW DO &KDSWHU f *HQH SWK$ LV HVVHQWLDO IRU ; FLWUL WR LQGXFH FHOO GLYLVLRQ VSHFLILFDOO\ RQ FLWPV SODQWV OHDGLQJ WR K\SHUSODVWLF FDQNHU OHVLRQV *HQH DZE LV LPSRUWDQW IRU ; F SY PDOYDFHDUXP WR LQGXFH ZDWHUVRDNLQJ V\PSWRPV VSHFLILFDOO\ RQ FRWWRQ 7KHVH WZR JHQHV DOVR IXQFWLRQ SOHLRWURSLFDOO\ DV DZ JHQHV 'HVSLWH WKH GLYHUVH DYLUXOHQFH DQG SDWKRJHQLF VSHFLILFLWLHV HQFRGHG E\ WKHVH DZSWK JHQHV WKHLU '1$ VHTXHQFHV DUH QHDUO\ LGHQWLFDO bf 'H )H\WHU HW DO &KDSWHU f %RWK SDWKRJHQLF DQG DYLPOHQFH VSHFLILFLWLHV RI WKH DZSWK JHQHV DUH GHWHUPLQHG E\ WKH QHDUO\ LGHQWLFDO DPLQR DFLG WDQGHP UHSHDWV HQFRGHG LQ WKH FHQWUDO

PAGE 97

UHJLRQ RI WKH JHQHV +HUEHUV HW DO &KDSWHU f 7R DFFRXQW IRU WKH SUHFLVH UHFRJQLWLRQ DQG GLYHUVH VSHFLILFLWLHV LW KDV EHHQ VSHFXODWHG WKDW WKH DYUSWK SURWHLQ SURGXFW PD\ GLUHFWO\ PHGLDWH WKH SODQW K\SHUVHQVLWLYH UHVSRQVH +HUEHUV HW DO f DV LQLWLDOO\ VXJJHVWHG LQ WKH GLPHU K\SRWKHVLV ,Q WKLV VWXG\ ZH GHPRQVWUDWHG WKDW SWK$ FRQIHUUHG WR ; FDPSHVWULV SY PDOYDFHDUXP ZKLFK LV QRQSDWKRJHQLF RQ FLWUXVf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f DQG %/'(fS/\V6 1RYDJHQ 0DGLVRQ :,f ;DQWKRPRQDV VWUDLQV XVHG ZHUH ; FLWUL FDXVHV K\SHUSODVWLF FDQNHUV RQ FLWUXV 6ZDUXS HW DO f LWV PDUNHU H[FKDQJH PXWDQW ;FO ^SWK$?QSWVDF FDXVHV QR V\PSWRP RQ FLWUXVf ; FDPSHVWULV SY FLWUXPHOR FDXVHV ZDWHUVRDNHG OHDI VSRWV RQ FLWUXV DQG FRPPRQ EHDQ 6ZDUXS HW DO f ; FDPSHVWULV SY PDOYDFHDUXP ;FP+ FDXVHV ZDWHUVRDNHG OHDI VSRWV RQ FRWWRQf DQG LWV PXWDQWV ;FP+ DYUE7QJXV$

PAGE 98

UHIHU WR &KDSWHU f +06 VHYHQ DYU JHQHV GHOHWHG UHIHU WR &KDSWHU f ( FROL VWUDLQV ZHUH JURZQ LQ /XULD%HUWDQL /%f PHGLXP 6DPEURRN HW DO f DW r& RU r& ;DQWKRPRQDV VWUDLQV ZHUH JURZQ LQ 3<*0 SHSWRQH\HDVW H[WUDFWJO\FHURO 0236f PHGLXP DW r& DV GHVFULEHG 'H )H\WHU HW DO f 3ODVPLGV S8&
PAGE 99

RYHUH[SUHVVHG LQ ( FROL FHOOV DQG SXULILHG E\ DIILQLW\ FKURPDWRJUDSK\ GHVFULEHG ODWHUf &HOOIUHH VXVSHQVLRQV LQ O2P0 7ULV+&O S+f RU DIILQLW\ FROXPQ SXULILHG 3WK$ ZHUH LPPHGLDWHO\ LQILOWUDWHG LQWR OHDYHV RI FLWUXV DQG FRWWRQ SODQWV &LWUXV &LWUXV SDUDGLVL f'XQFDQf JUDSHIUXLWf DQG FRWWRQ *RVV\SLZQ KLUVXWZQf SODQWV ZHUH JURZQ XQGHU JUHHQKRXVH FRQGLWLRQV &RWWRQ OLQHV XVHG ZHUH $FDOD $Ff DQG LWV FRQJHQLH UHVLVWDQFH OLQHV $FEE $FE $F%,Q DV GHVFULEHG 'H )H\WHU HW DO 6ZDUXS HW DO f ; FDPSHVWULV SY PDOYDFHDUXP WUDQVFRQMXJDQWV FDUU\LQJ SWK$ RU LWV GHULYDWLYHV ZHUH LQRFXODWHG LQWR OHDYHV RI FLWUXV SODQWV LQ %/3 OHYHO FRQWDLQPHQW UHIHU WR )HGHUDO 5HJLVWHU 9RO 1R f DW WKH 'LYLVLRQ RI 3ODQW ,QGXVWU\ )ORULGD 'HSDUWPHQW RI $JULFXOWXUH *DLQHVYLOOH 3UHSDUDWLRQ RI DQWLVHUD DQG LPPXQREORWWLQJ $ NE %DP5O IUDJPHQW FDUU\LQJ LQWDFW SWK$ ZDV JHQHUDWHG IURP SDUWLDO GLJHVWLRQ RI S=LW 6ZDUXS HW DO f DQG FORQHG LQWR H[SUHVVLRQ YHFWRU S(7E 1RYDJHQ 0DGLVRQ :,f 7KH IXVLRQ VLWHV RI WKH UHVXOWLQJ FRQVWUXFW S8)< ZHUH GHWHUPLQHG E\ '1$ VHTXHQFH DQDO\VLV 3WK$ SURWHLQ ZDV RYHUH[SUHVVHG LQ ( FROL VWUDLQ %/'(fS/\V6 FRQWDLQLQJ S8)< LQ WKH SUHVHQFH RI ,P0 ,37* LVRSURS\O'WKLRJDODFWRS\UDQRVLGHf 6LQFH SURWHLQV H[SUHVVHG RQ S(7E DUH GHVLJQHG WR FRQWDLQ KLVWLGLQH UHVLGXHV DW WKH DPLQR WHUPLQXV DV DQ DIILQLW\ KDQGOH RYHUH[SUHVVHG 3WK$ ZDV UHDGLO\ SXULILHG E\ +LVWDJ DIILQLW\ FKURPDWRJUDSK\ 1RYDJHQ 0DGLVRQ :,f )RU LPPXQL]DWLRQ RI UDEELWV 3WK$ ZDV IXUWKHU SXULILHG E\ 6'6SRO\DFU\ODPLGH JHO HOHFWURSKRUHVLV $QWLVHUXP DJDLQVW 3WK$ ZDV SUHSDUHG IURP D 1HZ =HDODQG UDEELW E\ &RFDOLFR %LRORJLFDOV ,QF 5HDPVWRZQ 3$f )RU LPPXQREORWWLQJ ( FROL RU ;DQWKRPRQDV FHOOV ZHUH KDUYHVWHG E\ FHQWULIXJDWLRQ DQG

PAGE 100

UHVXVSHQGHG LQ VDPSOH EXIIHU P0 7ULV+&O S+ b JO\FHURO b 6'6 b EURPRSKHQRO EOXHf $IWHU VDPSOHV ZHUH ERLOHG IRU PLQ DQG FHQWULIXJHG IRU PLQ WKH\ ZHUH VHSDUDWHG RQ b 6'6SRO\DFU\ODPLGH JHOV DQG WUDQVIHUUHG WR QLWURFHOOXORVH E\ HOHFWUREORWWLQJ 7RZELQ HWDO f :HVWHUQ EORWV ZHUH SUREHG ZLWK DQWL3WK$ DQWLERG\ DQG YLVXDOL]HG ZLWK JRDW DQWLUDEELW DQWLERG\ FRQMXJDWHG ZLWK DONDOLQH SKRVSKDWDVH '1$ FRQVWUXFWV IRU *86 IXVLRQ SURWHLQV 7KH '1$ FRGLQJ VHTXHQFHV IRU WKH & WHUPLQDO UHJLRQ DPLQR DFLGVf RI 3WK$ DQG $YUE ZHUH DPSOLILHG E\ 3&5 IURP WKH SODVPLG S=LW 6ZDUXS HW DO f E\ XVLQJ WKH IROORZLQJ SULPHUV &7&7$*$*&&$7*$&*&$*77& &$*$7&7&7*$**&$$7$*&7& 7KH DPSOLILHG IUDJPHQWV ZHUH ILUVW FORQHG LQWR S*(07 3URPHJD ,QF 0DGLVRQ :,f WKHQ FXW RXW ZLWK ;EDO%JO8 DQG UHFORQHG LQWR WKH ;EDO%DP+O VLWH RI SODQW H[SUHVVLRQ YHFWRU S%, &ORQHWHFK &$f 7KH UHVXOWLQJ SODVPLGV S8)< DQG S8)< FRQWDLQ WKH FKLPHULF JHQH ZLWK WKH &WHUPLQDO FRGLQJ VHTXHQFH RI 3WK$ DQG $YUE UHVSHFWLYHO\ WUDQVODWLRQDOO\ IXVHG ZLWK f HQG RI *86 JHQH %RWK S8)< DQG S8)< ZHUH FKHFNHG E\ VHTXHQFH DQDO\VLV DW WKH IXVLRQ VLWH 2QLRQ WUDQVIRUPDWLRQ V\VWHP 7KH +HOLXP %LROLVWLF JHQH WUDQVIRUPDWLRQ V\VWHP 'X 3RQWf ZDV XVHG WR WUDQVIRUP HSLGHUPDO FHOO OD\HUV RI ZKLWH RQLRQ DV GHVFULEHG 9DUDJRQD HW DO f ,QQHU HSLGHUPDO OD\HUV ZHUH SHHOHG DQG SODFHG LQVLGH XS RQ 3HWUL GLVKHV FRQWDLQLQJ 06 0XUDVKLJH DQG 6NRRJ f EDVDO PHGLD J/ 06 VDOW PL[WXUH 6LJPD 0 J/ VXFURVH S+ f ZLWK PJ/ DPSKRWHULFLQ % 6LJPDf DQG b DJDU '1$ VDPSOHV ZHUH SUHSDUHG DV GHVFULEHG 7D\ORU DQG 9DVLO f )LYH WLJ RI FROXPQSXULILHG 0DJLF 3UHS 3URPHJDf SODVPLG '1$ ZDV SUHFLSLWDWHG RQWR PJ

PAGE 101

RI [P JROG SDUWLFOHV XVLQJ [/ RI 0 &D&O DQG [/ RI P0 VSHUPLGLQH '1$FRDWHG JROG SDUWLFOHV ZHUH ZDVKHG ZLWK [O RI b DQG b HWKDQRO DQG ILQDOO\ UHVXVSHQGHG LQ [/ RI b HWKDQRO 7KH VDPSOHV ZHUH GLVSHUVHG E\ VXEPHUJLQJ LQ D ZDWHUEDWK VRQLFDWRU IRU D IHZ VHFRQGV )LYH [/ VDPSOHV ZHUH SLSHWWHG RQWR SDUWLFOH GHOLYHU\ GLVFV DQG GHOLYHUHG LQWR RQLRQ HSLGHUPDO FHOO OD\HUV DW SUHVVXUHV RI RU SVL $IWHU PLFURSURMHFWLOH ERPEDUGPHQW 3HWUL GLVKHV ZHUH VHDOHG ZLWK SDUDILOP DQG LQFXEDWHG RYHUQLJKW DW r& LQ WKH GDUN 7KH DVVD\V IRU QXFOHDU ORFDOL]DWLRQ DFWLYLW\ RI WKH IXVLRQ SURWHLQ ZHUH UHSHDWHG LQ WKUHH LQGHSHQGHQW WUDQVIRUPDWLRQ WHVWV +LVWRFKHPLFDO DQDO\VLV 7KH KLVWRFKHPLFDO *86 DVVD\ ZDV XVHG WR GHWHUPLQH WKH ORFDWLRQ RI *86 IXVLRQ SURWHLQV LQ RQLRQ FHOOV DV GHVFULEHG SUHYLRXVO\ -HIIHUVRQ 9DUDJRQD HW DO f 2QLRQ HSLGHUPDO OD\HUV ZHUH LQFXEDWHG DW URRP WHPSHUDWXUH LQ ;JOX VROXWLRQ P0 SKRVSKDWH EXIIHU S+ ,P0 ('7$ b 7ULWRQ ; P0 VRGLXP IHUULF\DQLGH DQG IHUURF\DQLGH P0 ;JOXf $IWHU GHWHFWLRQ RI EOXH FRORU RQLRQ HSLGHUPDO OD\HUV ZHUH PRXQWHG RQ D JODVV VOLGH ZLWK D VROXWLRQ FRQWDLQLQJ [JPO RI WKH QXFOHXVVSHFLILF G\H fGLDPLGLQRSKHQ\OLQGROH '$3, ; SKRVSKDWH EXIIHU O2P0 VRGLXP D]LGH b JO\FHUROf &HOOXODU ORFDWLRQ RI WKH EOXH LQGLJR G\H SURGXFHG E\ R[LGDWLYH GLPHUL]DWLRQ RI WKH *86 SURGXFW ZDV GHWHUPLQHG XQGHU EULJKWILHOG RSWLFV XVLQJ D =HLVV $[LRSKRW PLFURVFRSH DQG FRPSDUHG ZLWK WKH ORFDWLRQ RI '$3,VWDLQHG QXFOHL XQGHU IOXRUHVFHQFH RSWLFV

PAGE 102

5HVXOWV &LWUXV FDQNHU V\PSWRPV HOLFLWHG E\ ; FDPSHVWULV SY PDO YDF HDUQ P FDUU\LQJ SWK$ *HQH SWK$ LV UHTXLUHG IRU SDWKRJHQLFLW\ RI ; FLWUL RQ FLWUXV DQG FRQIHUV FDQNHU HOLFLWLQJ DELOLW\ WR PDQ\ FLWUXVFRPSDWLEOH [DQWKRPRQDGV 6ZDUXS HW DO f ,Q WKLV VWXG\ SWK$ DQG LWV GHULYDWLYHV ZHUH LQWURGXFHG LQWR ; FDPSHVWULV SY PDOYDFHDUXP D FRWWRQ SDWKRJHQ WKDW GRHV QRW JURZ RQ FLWUXV SODQWV .LQJVOH\ HW DO f :KHQ LQRFXODWHG RQ FLWUXV WUDQVFRQMXJDQWV RI ; F SY PDOYDFHDUXP FDUU\LQJ SWK$ HOLFLWHG K\SHUSODVWLF FDQNHUOLNH OHVLRQV RQ FLWUXV )LJ f 7KH WUDQVFRQMXJDQW ;FP+S8)< ZKLFK GLIIHUV IURP ;FP+ RQO\ E\ UHSODFLQJ WKH UHSHDWHG UHJLRQ RI DYUE ZLWK WKDW RI SWK$ DOVR HOLFLWHG FDQNHUOLNH OHVLRQV +RZHYHU SWK$ GLG QRW HQDEOH ; F SY PDOYDFHDUXP WR JURZ RQ QRQKRVW FLWUXV DQG KLJK LQLWLDO LQRFXOXP OHYHOV EDFWHULDPOf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

PAGE 103

)LJXUH &DQNHU V\PSWRPV HOLFLWHG E\ ; FDPSHVWULV SY PDOYDFHDUXP WUDQVFRQMXJDQWV RQ QRQKRVW FLWUXV ;FP+S8)5 ;FP+S8)5 ;FP+06S8)5 ;FP+S=LW ;FP+S8)< ;FP+06S=LW

PAGE 104

&RQVWLWXWLYH H[SUHVVLRQ RI 3WK $ SURWHLQ :HVWHUQ EORW DQDO\VLV UHYHDOHG WKDW WKH SUHGLFWHG N' 3WK$ SURWHLQ ZDV FRQVWLWXWLYHO\ H[SUHVVHG LQ ;DQWKRPRQDV FHOOV FDUU\LQJ SWK$ )LJ f 7KH DQWLERG\ VSHFLILFDOO\ UHFRJQL]HG SURWHLQV LQ SURWHLQ H[WUDFWV RI ZLOG W\SH ; FLWUL VWUDLQ ZKLFK FRQWDLQV SWK$ SOXV WKUHH SXWDWLYH KRPRORJXHV $V SUHGLFWHG E\ WKH VL]H RI WKHVH KRPRORJRXV JHQH WKH ODUJHVW SURWHLQ ZDV 3WK$ LQGLFDWHG E\ DQ DUURZ LQ ODQH )LJ f 7KLV ZDV FRQILUPHG E\ :HVWHUQ DQDO\VLV RI WKH PDUNHU H[FKDQJH PXWDQW ;FO SWK$ZQSWVDFf *HQH SWK$ ZDV GLVUXSWHG LQ ;F E\ D QSWOVDF PDUNHU LQVHUWHG DW WKH WK WDQGHP UHSHDW &KDSWHU f ZKLFK ZDV H[SHFWHG WR JHQHUDWH D WUXQFDWHG SURWHLQ RI N' XSRQ WUDQVODWLRQ $V VKRZQ LQ ODQH )LJ WKLV PXWDQW ORVW WKH 3WK$ SURWHLQ DQG LQVWHDG H[SUHVVHG D SURWHLQ FORVH WR WKH VL]H RI SUHGLFWHG WUXQFDWHG SURWHLQ N'f 7KH DQWL3WK$ DQWLERG\ GLG QRW UHDFW ZLWK H[WUDFWV IURP ; FDPSHVWULV SY FLWUXPHOR ZKLFK FRQWDLQV QR JHQHV KRPRORJRXV WR SWK$ +RZHYHU 3WK$ SURWHLQ ZDV FRQVWLWXWLYHO\ H[SUHVVHG LQ ULFK PHGLXP E\ WKH WUDQVFRQMXJDQW FDUU\LQJ SWK$ LQGLFDWHG E\ DQ DUURZ LQ ODQH )LJ f 7KH &WHUPLQDO UHJLRQ RI $YU3WK SURWHLQV FRQWDLQ QXFOHDU ORFDOL]DWLRQ FRQVHQVXV VHTXHQFHV 7KH SURWHLQV HQFRGHG E\ DOO VHTXHQFHG PHPEHUV RI WKH ;DQWKRPRQDV DYUSWK VXFK DV DYUE 'H )H\ WHU HW DO f SWK$ FKDSWHU f D?U%V %RQDV HW DO f DQG DYU;DO2 +RSNLQV HW DO f DUH KLJKO\ KRPRORJRXV DQG RQO\ YDU\ LQ WKH LQWHUQDO UHSHWLWLYH UHJLRQ ZKLFK GHWHUPLQHV DYLUXOHQFH DQG SDWKRJHQLF VSHFLILFLWLHV $QDO\VLV RI WKH DPLQR DFLG VHTXHQFHV HQFRGHG E\ WKHVH JHQHV UHYHDOHG WKH SUHVHQFH RI WKUHH VWUHWFKHV RI EDVLF UHVLGXHV ZLWK FRPSOHWH KRPRORJ\ ZLWK WKH QXFOHDU ORFDOL]DWLRQ

PAGE 105

N' r r f§ )LJXUH :HVWHUQ EORW DQDO\VLV RI SURWHLQ H[WUDFWV IURP ;DQWKRPRQDV VWUDLQV ;DQWKRPRQDV FLWUL VWUDLQ PDUNHU H[FKDQJH PXWDQW ;FO SWK$?QSWVDFf? ; F SY FLWUXPHOR VWUDLQ WUDQVFRQMXJDQW FDUU\LQJ S=LW SWK$f $VWHULVN LQGLFDWH WKH EDQG RI 3WK$ SURWHLQ RU LWV WUXQFDWHG GHULYDWLYH

PAGE 106

3URWHLQ 1XFOHDU /RFDOL]DWLRQ 6HTXHQFHV &RQVHQVXV VHTXHQFH .5.;.5 697 PRQRSDUWLWHf S . 5 Y 1XFOHRSODVPLQ ELSDUWLWHf .5SDDW..DJJD....O 9LU' .5S5HGGGJHSVH5.5H5G 3WK$ DQG $YUE 5 D S DD5 5 V 5 DD5 Y 5 S 5 )LJXUH 1XFOHDU ORFDOL]DWLRQ VHTXHQFHV IURP 69 7DQWLJHQf ;HQRSXV QXFOHRSODVPLQf $JUREDFWHULXP 9LU'f DQG ;DQWKRPRQDV 3WK$ DQG $YUEf &DSLWDO OHWWHUV GHQRWHV EDVLF UHVLGXHV SUHVXPHG LPSRUWDQW IRU QXFOHDU WDUJHWLQJ

PAGE 107

FRQVHQVXV VHTXHQFHV .5.;5.f IRXQG LQ PDQ\ FKDUDFWHUL]HG QXFOHDU ORFDOL]HG SURWHLQV &KHOVN\ HW DO f )LJ f 7KHVH WKUHH SXWDWLYH QXFOHDU ORFDOL]DWLRQ VHTXHQFHV 1/6Vf DUH ORFDWHG DW WKH FDUER[\OWHUPLQDO SDUW RI WKH SURWHLQV DW SRVLWLRQV .5$53f 5.565f DQG 59.535f LQ 3WK$ 7KH ILUVW DQG WKLUG SXWDWLYH 1/6V DOVR FRQWDLQV SUROLQH UHVLGXHV ZKLFK DUH RIWHQ IRXQG LQ 1/6V ,Q DGGLWLRQ WKUHH SXWDWLYH FDVHLQ NLQDVH ,, VLWHV ZHUH IRXQG LQ SUR[LPLW\ WR WKHVH WKUHH QXFOHDU ORFDOL]DWLRQ FRQVHQVXV VHTXHQFHV 7KH\ DUH ORFDWHG DW SRVLWLRQV 7(/(f 656'f 7$$'f LQ 3WK$ 7KLV IHDWXUH LV FRPPRQ WR PDQ\ QXFOHDU ORFDOL]HG SURWHLQV ZKLFK RIWHQ KDYH FDVHLQ NLQDVH ,, VLWHV 67 ;;'(f DW D GLVWDQFH RI DPLQR DFLGV XS RU GRZQVWUHDP IURP WKH 1/6V 5LKV HW DO f 7KH &WHUPLQXV RI 3WK$ DQG $YUE SURWHLQV GLUHFWHG IOJOXFXURQLGDVH LQWR WKH SODQW QXFOHXV 7KH N' SJOXFXURQLGDVH *86f SURWHLQ KDV EHHQ VKRZQ WR ORFDOL]H LQ WKH F\WRSODVP 5HVWUHSR HW DO YDQ GHU .URO DQG &KXD f DQG LV WRR ODUJH WR HQWHU WKH QXFOHXV ZLWKRXW DQ 1/6 9DUDJRQD HW DO f $V VKRZQ LQ )LJ $ ZKHQ S%, HQFRGLQJ *86f ZDV LQWURGXFHG LQWR RQLRQ HSLGHUPDO FHOOV *86 DFWLYLW\ ZDV IRXQG LQ WKH F\WRSODVP DQG QRW LQ WKH QXFOHXV 7R GHWHUPLQH ZKHWKHU RU QRW WKH SXWDWLYH 1/6V DUH IXQFWLRQDO LQ SODQW FHOOV WKH '1$ FRGLQJ VHTXHQFHV IRU WKH &WHUPLQDO UHJLRQ DPLQR DFLGVf RI 3WK$ DQG $YUE ZHUH WUDQVODWLRQDOO\ IXVHG ZLWK WKH f HQG RI D *86 UHSRUWHU JHQH 7KH UHVXOWLQJ FRQVWUXFWV ZHUH LQWURGXFHG LQWR RQLRQ HSLGHUPDO FHOOV E\ PLFURSURMHFWLOH ERPEDUGPHQW DQG H[SUHVVHG WUDQVLHQWO\ ,QWUDFHOOXODU ORFDOL]DWLRQ RI *86 IXVLRQ SURWHLQV DSSUR[LPDWHO\ N'f FRQWDLQLQJ WKH &WHUPLQDO UHJLRQ RI 3WK$

PAGE 108

)LJXUH +LVWRFKHPLFDO ORFDOL]DWLRQ RI *86 DQG *86 IXVLRQ SURWHLQV FRQWDLQLQJ WKH &WHUPLQDO UHJLRQ RI 3WK$ RU $YUE LQ WUDQVIRUPHG RQLRQ HSLGHUPDO WLVVXHV 7LVVXH ZHUH VLPXOWDQHRXVO\ DQDO\]HG XVLQJ ;JOX VWDLQLQJ $ % &f DQG QXFOHLVSHFLILF '$3, VWDLQLQJ %f &ff $ WUDQVIRUPHG ZLWK S%, HQFRGLQJ *86 SURWHLQ % DQG %f WUDQVIRUPHG ZLWK S8)< HQFRGLQJ *86 IXVLRQ SURWHLQ FRQWDLQLQJ WKH &WHUPLQXV RI 3WK$ & DQG &? WUDQVIRUPHG ZLWK S8)< HQFRGLQJ *86 IXVLRQ SURWHLQ FRQWDLQLQJ WKH &WHUPLQXV RI $YUE

PAGE 110

RU $YUE ZDV GHWHUPLQHG KLVWRFKHPLFDOO\ ZLWK *86 DFWLYLW\ ORFDOL]HG VSHFLILFDOO\ WR WKH QXFOHXV RI SODQW FHOOV )LJ % DQG &f ,Q UHSHDWHG H[SHULPHQWV PRUH WKDQ RQH KXQGUHG QXFOHL ZHUH REVHUYHG WR EH *86 SRVLWLYH 5DUHO\ VRPH GLIIXVH *86 DFWLYLW\ ZDV REVHUYHG ZLWKLQ WKH F\WRSODVP 7KH ORFDWLRQ RI QXFOHL ZDV FRQILUPHG E\ VWDLQLQJ ZLWK WKH QXFOHXVVSHFLILF G\H fGLDPLGLQRSKHQ\OLQGROH '$3,f DQG REVHUYHG XQGHU IOXRUHVFHQFH PLFURVFRSH )LJ %f DQG &ff 'LVFXVVLRQ 7KH SDWKRJHQLFLW\ JHQH SWK$ LV UHTXLUHG IRU ; FLWUL WR HOLFLW K\SHUSODVWLF FDQNHU V\PSWRPV RQ FLWUXV DQG FRQIHUV FDQNHUHOLFLWLQJ DELOLW\ WR VHYHUDO [DQWKRPRQDGV WKDW DUH PLOGO\ SDWKRJHQLF RQ FLWUXV 6ZDUXS HW DO f ; F SY PDOYDFHDUXP D FRWWRQ SDWKRJHQ LV QRQSDWKRJHQLF RQ FLWUXV DQG FDQ QRW JURZ RQ FLWUXV SODQWV .LQJVOH\ HW DO f EXW LW KDV KLJK KRPRORJ\ bf ZLWK ; FLWUL DV VKRZQ E\ '1$ UHDVVRFLDWLRQ (JHO HW DO f ,Q WKLV VWXG\ LQWURGXFWLRQ RI SWK$ LQWR ; FDPSHVWULV SY PDOYDFHDUXP VWUDLQ ;FP+ RU WKH UHSODFHPHQW RI WKH WDQGHP UHSHDWV RI DYUE LQ ;FP+ ZLWK WKRVH RI SWK$ FRQIHUUHG WKH DELOLW\ WR LQGXFH K\SHUSODVWLF FDQNHU OHVLRQV RQ FLWUXV 7KH KRVWVSHFLILF FDQNHU V\PSWRPV LQGXFHG E\ ; FDPSHVWULV SY PDOYDFHDUXP WUDQVFRQMXJDQWV FDUU\LQJ SWK$ ZHUH LQGHSHQGHQW RI EDFWHULDO JURZWK LQ SODQWD VLQFH KLJK LQRFXOXP OHYHOV FHOOVPOf ZHUH UHTXLUHG WR HOLFLW YLVLEOH V\PSWRPV 7KHUHIRUH DGGLWLRQDO YLUXOHQFH JHQHV DUH QHHGHG WR H[WHQG WKH KRVW UDQJH RI ; FDPSHVWULV SY PDOYDFHDUXP WR LQFOXGH FLWUXV SODQWV *HQH SWK$ EHORQJV WR D ;DQWKRPRQDV DYUSWK JHQH

PAGE 111

IDPLO\ DQG GLIIHUV RQO\ VOLJKWO\ IURP RWKHU PHPEHUV LQ WKH UHSHDWHG UHJLRQ RI JHQHV 6ZDUXS HW DO DOVR UHIHU WR &KDSWHU f ,W LV VWULNLQJ WKDW RQO\ D IHZ DPLQR DFLG FKDQJHV LQ WKH WDQGHP UHSHDWV RI D VLQJOH DYU JHQH HQDEOHG ;FP+ WR HOLFLW FDQNHU V\PSWRPV RQ FLWUXV 6LQFH WKH UHSHWLWLYH GRPDLQV RI SURNDU\RWLF DQG HXNDU\RWLF SURWHLQV RIWHQ IXQFWLRQ DV OLJDQG ELQGLQJ VLWHV DQG GHWHUPLQH VSHFLILFLW\ :UHQ 0F&RQNH\ HW DO f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f 'HVSLWH WKLV FRQVWLWXWLYH H[SUHVVLRQ KUS K\SHUVHQVLWLYH UHVSRQVH DQG SDWKRJHQLFLW\f JHQH IXQFWLRQ LV DEVROXWHO\ UHTXLUHG IRU ;DQWKRPRQDV FDUU\LQJ DYU%V WR HOLFLW DQ +5 RQ SHSSHU .QRRS HW DO f DQG IRU ;DQWKRPRQDV FDUU\LQJ SWK$ WR HOLFLW FDQNHUV RQ FLWUXV XQSXEOLVKHG GDWD *DEULHOf 6LQFH KUS JHQHV DUH NQRZQ WR EH LQYROYHG LQ WKH H[SRUW RI SURWHLQDFHRXV +5 HOLFLWRUV +H HW DO f DQG

PAGE 112

VLQFH KLSn PXWDWLRQV DEROLVK DFWLYLW\ RI WZR PHPEHUV RI WKH JHQH IDPLO\ LW VHHPV OLNHO\ WKDW WKH SURWHLQV HQFRGHG E\ WKHVH DYUSWK JHQHV DUH H[SRUWHG IURP WKH EDFWHULD LQ WKHLU LQWHUDFWLRQV ZLWK SODQWV )HQVODX f 1R GLUHFW HYLGHQFH KDV EHHQ IRXQG IRU WKH H[SRUW RI SURWHLQV HQFRGHG E\ WKLV JHQH IDPLO\ &HOO IUDFWLRQDWLRQ LQGLFDWHV WKDW WKH PDMRU SRUWLRQ RI $YU%V SURWHLQ LV LQ WKH EDFWHULDO F\WRSODVP DQG LPPXQRHOHFWURQ PLFURVFRS\ IDLOV WR GHWHFW $YU%V LQ WKH EDFWHULDO RXWHU PHPEUDQH RU HYLGHQFH RI VHFUHWLRQ .QRRS HW DO %URZQ HW DO f 7KH IDLOXUH WR GHWHFW SURWHLQ VHFUHWLRQ FRXOG KDYH UHVXOWHG IURP SURWHLQ LQVWDELOLW\ RU YHU\ ORZ OHYHOV RI WKH DQWLJHQ LQ WKH H[WUDFHOOXODU HQYLURQPHQW )RU H[DPSOH 9LU( RI $JUREDFWHULXP WXPDIDFLHQV ZDV ORFDOL]HG LQ WKH EDFWHULDO F\WRSODVP E\ FHOO IUDFWLRQDWLRQ DQG E\ LPPXQRHOHFWURQ PLFURVFRS\ &KULVWLH HW DO 7KRUVWHQVRQ DQG =DPEU\VNL f KRZHYHU LW LV NQRZQ WR EH H[SRUWHG WRJHWKHU ZLWK 7'1$ DQG HQWHUV WKH QXFOHXV RI SODQW FHOOV &LWRYVN\ HW DO f 7KH WUDQVORFDWLRQ RI WKH $JUREDFWHULXP 7'1$ FRPSOH[ LQWR WKH QXFOHXV RI SODQW FHOOV FDQ EH YLHZHG DV DQ XQLTXH VLJQDOLQJ SDWKZD\ &LWRYVN\ DQG =DPEU\VNL f 7KH 7'1$ LV FRDWHG E\ WKH VLQJOH VWUDQG ELQGLQJ SURWHLQ 9LU( DQG SLORWHG E\ SURWHLQ 9LU' &KULVWLH HW DO :DUG DQG %DUQHV f %RWK 9LU( DQG 9LU' FRQWDLQ QXFOHDU ORFDOL]DWLRQ VHTXHQFHV 1/6Vf ZKLFK FRXOG IDFLOLWDWH WKH WUDQVORFDWLRQ RI 7'1$ FRPSOH[ LQWR WKH QXFOHXV &LWRYVN\ HW DO +RZDUG HW DO 6KXUYLQWRQ HW DO 7LQODQG HW DO f 1XFOHDU ORFDOL]DWLRQ VHTXHQFHV DUH VKRUW VWUHWFKHV RI EDVLF DPLQR DFLG UHVLGXHV FRQVHUYHG LQ PDQ\ QXFOHDU ORFDOL]HG SURWHLQV 9LU' DQG 9LU( DUH WKH RQO\ EDFWHULDO SURWHLQV WKDW KDYH EHHQ UHSRUWHG WR FRQWDLQ QXFOHDU ORFDOL]DWLRQ VHTXHQFHV

PAGE 113

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f DOVR FRQWDLQV D SXWDWLYH ELSDUWLWH QXFOHDU ORFDOL]DWLRQ VHTXHQFH ..:&'<31/67&39.73*3.3..f LQ WKH &WHUPLQDO UHJLRQ $95 DQG $95 RI &ODGRVSRULXP IXOYXP DUH WKH RQO\ NQRZQ UDFHVSHFLILF SURWHLQ HQFRGHG E\ DYU JHQHV WKDW DUH FDSDEOH RI GLUHFWO\ HOLFLWLQJ DQ +5 RQ SODQWV ,W ZLOO EH LQWHUHVWLQJ WR VHH LI WKLV SXWDWLYH 1/6 DUH IXQFWLRQDO LQ SODQW FHOOV &KDQJHV LQ WKH FHOO GLYLVLRQ DQG GLIIHUHQWLDWLRQ VWDWXV RI HXNDU\RWLF FHOOV PD\ EH DFFRPSDQLHG E\ WKH VHOHFWLYH UHORFDOL]DWLRQ RI SURWHLQV EHWZHHQ F\WRSODVP DQG QXFOHXV 1LJJ HW DO f 7KLV KROGV IRU IRV DQG UHO HQFRGHG RQFRSURWHLQV FHUWDLQ WUDQVFULSWLRQDO IDFWRUV VWHURLG KRUPRQH UHFHSWRUV WKH 'URVRSKLOD PRUSKRJHQ 'RUVDO DQG WKH $JUREDFWHULXP 7FRPSOH[ )RU H[DPSOH WKH DFWLYLW\ RI WKH 5HO1)LF%'RUVDO IDPLO\ RI SURWHLQV LV FRQWUROOHG E\ VHTXHVWUDWLRQ LQ WKH F\WRSODVP LQ DVVRFLDWLRQ ZLWK LQKLELWRU\ SURWHLQV FDOOHG ,N% .HUU HW DO f ,Q XQVWLPXODWHG FHOOV WUDQVFULSWLRQ IDFWRU 1)N% LV DVVRFLDWHG ZLWK ,N% DQG UHWDLQHG LQ WKH F\WRSODVP +RZHYHU H[WHUQDO VLJQDOV DFWLYDWH SURWHLQ NLQDVHV ZKLFK SKRVSKRU\ODWH ,N% UHVXOWLQJ LQ WKH GLVVRFLDWLRQ RI WKH 1)N%,N%

PAGE 114

FRPSOH[ 6XEVHTXHQW WUDQVORFDWLRQ RI 1)N% LQWR WKH QXFOHXV WULJJHUV JHQH H[SUHVVLRQ ,Q PDPPDOLDQ V\VWHPV 1)N% LV DFWLYDWHG E\ R[\JHQ UDGLFDOV LQGXFHG E\ GLYHUVH VLJQDOV WKURXJK GLVWLQFW LQWUDFHOOXODU SDWKZD\V 6FKUHFN HW DO f ,Q SODQWV DFWLYH R[\JHQ VSHFLHV DUH NQRZQ WR EH LQYROYHG LQ K\SHUVHQVLWLYH UHVSRQVH DQG DFWLYDWLRQ RI SODQW GHIHQVH JHQHV %UDGOH\ HW DO &KHQ HW DO /HJHQGUH HW DO f +RZHYHU SODQW KRPRORJXHV RI 1)N% DQG ,N% DUH QRW NQRZQ 7KH LQKLELWRU\ ,N% SURWHLQV FRQWDLQ WR DQN\ULQ UHSHDWV 7KH DQN\ULQ UHSHDW PRWLI LV D DPLQR DFLG XQLW WKDW LV IRXQG LQ D QXPEHU RI WUDQVFULSWLRQ IDFWRUV UHFHSWRUV DQG LRQ WUDQVSRUWHUV DQG LV LPSRUWDQW LQ SURWHLQSURWHLQ LQWHUDFWLRQV %ODQN HW DO 6HQWHQDF HW DO f ,QWHUHVWLQJO\ WZR VKRUW VWUHWFKHV RI FRQVHUYHG DPLQR DFLGV *;73 DQG 9;;//f IRXQG LQ WKH DQN\ULQ UHSHDW DOVR H[LVW LQ WKH DPLQR DFLG UHSHDWV RI 3WK$ DQG LWV KRPRORJV +RZHYHU LW LV QRW NQRZQ ZKHWKHU WKLV LV MXVW FRLQFLGHQWDO RU KDV IXQFWLRQDO PHDQLQJ %DVHG RQ WKH NQRZQ GDWD DQG SUHYLRXV PRGHOV *DEULHO DQG 5ROIH f D WHVWDEOH K\SRWKHVLV ZDV SURSRVHG WR H[SODLQ WKH SRVVLEOH PHFKDQLVP E\ ZKLFK VSHFLILF VLJQDOV HQFRGHG E\ WKH ;DQWKRPRQDV DYUSWK JHQH DUH SHUFHLYHG DQG WUDQVGXFHG )LJ f $IWHU HQWHULQJ LQWHUFHOOXODU VSDFH RI SODQW WLVVXHV WKH SDWKRJHQ UHOHDVHV WKH $YU3WK SURWHLQV E\ DQ LSGHSHQGHQW H[SRUW SDWKZD\ DFWLYDWHG E\ SODQW VLJQDOV )HQVHODX HW DO f $W WKH VDPH WLPH WKH SDWKRJHQ DOVR VHFUHWHV H[WUDFHOOXODU HQ]\PHV WR ZHDNHQ WKH SODQW FHOO ZDOO $YU3WK SURWHLQV HQWHU SODQW FHOO WKURXJK D SURFHVV VLPLODU WR WKH UHFHSWRUPHGLDWHG HQGRF\WRVLV RI VRPH IXQJDO HOLFLWRUV +RUQ HW DO f 2QFH LQWHUQDOL]HG $YU3WK SURWHLQV LQWHUDFW VSHFLILFDOO\ ZLWK 5 JHQH SURGXFWV WR IRUP WKH GLPHUV RU PXOWLPHUV ZKLFK DUH WKHQ WUDQVSRUWHG LQWR WKH QXFOHXV RI SODQW FHOOV

PAGE 115

%DFWHULDO FHOO UHOHDVHG E\ VHFUHWLRQ RU FHOO O\VLVf 6SHFLILF UHFRJQLWLRQ *HQH H[SUHVVLRQ 1XFOHXV ([WUDFHOOXODU HQ]\PHV ZHDNHQLQJ SODQW FHOO ZDOO $ $YU3WK SURWHLQV HQGRF\WRVLVf +\SHUSODVLD FHOO GLYLVLRQf :DWHUVRDNLQJ t QHFURVLV GHOD\HG FHOO GHDWKf +\SHUVHQVLWLYH UHVSRQVH UDSLG FHOO GHDWKf )LJXUH $ ZRUNLQJ K\SRWKHVLV IRU WKH WUDQVGXFWLRQ RI ;DQWKRPRQDV $YU3WK VLJQDOV GXULQJ WKH SODQWSDWKRJHQ LQWHUDFWLRQ

PAGE 116

$OWHUQDWLYHO\ $YU3WK SURWHLQV DUH LPSRUWHG DORQH LQWR WKH QXFOHXV 7KH QXFOHDU WUDQVSRUW SURFHVV FDQ EH DFWLYDWHG RU LQKLELWHG E\ PDQ\ IDFWRUV VXFK DV SKRVSKRU\ODWLRQ RU GHSKRVSKRU\ODWLRQ DW VSHFLILF FDVHLQ NLQDVH ,, VLWHV RI SURWHLQV 5LKV HW DO f $IWHU HQWHULQJ WKH QXFOHXV $YU3WK SURWHLQV PD\ DFW RQ QXFOHDU WUDQVFULSWLRQDO IDFWRUV HQFRGHG E\ 5 JHQHV RU HYHQ GLUHFWO\ RQ WKH 5 JHQH WKHPVHOYHV 0HGLDWHG E\ WKH DPLQR DFLG WDQGHP UHSHDWV WKH DYLUXOHQFH DQG SDWKRJHQLF VSHFLILFLW\ FDQ EH GHWHUPLQHG GXULQJ WKH SURWHLQSURWHLQ RU SURWHLQ'1$ LQWHUDFWLRQ $FWLYDWLRQ RI VSHFLILF WDUJHWHG JHQHV PD\ OHDG WR GLIIHUHQW SK\VLRORJLFDO RXWFRPHV VXFK DV K\SHUSODVLD FHOO GLYLVLRQf RQ FLWUXV K\SHUVHQVLWLYH UHVSRQVH UDSLG FHOO GHDWKf RQ FRWWRQ RU FLWUXV ZDWHUVRDNLQJ DQG VXEVHTXHQW QHFURVLV RQ FRWWRQ GHOD\HG FHOO GHDWKf 7KLV PD\ EH VLPLODU WR RQFRJHQHVLV FHOO GLYLVLRQf DQG DSRSWRVLV FHOO GHDWKf RI PDPPDOLDQ FHOOV WKDW FDQ EH DOWHUHG E\ WKH UDWLR DQG LQWHUDFWLRQV DPRQJ WUDQVFULSWLRQ IDFWRUV LQ VSHFLILF FHOOV 0DUWLQ HW DO f 7KLV K\SRWKHVLV PD\ DSSO\ WR RWKHU PRGHO V\VWHPV RI VSHFLILF UHFRJQLWLRQ VXFK DV WKH &ODGRVSRZQ ILLOYXP $95 SURWHLQ FRQWDLQLQJ SRWHQWLDO 1/6V +RZHYHU GLYHUVH VLJQDO WUDQVGXFWLRQ PHFKDQLVPV PD\ KDYH EHHQ H[SORLWHG E\ KRVWV DQG SDWKRJHQV WR GHWHUPLQH WKH VSHFLILF UHFRJQLWLRQV LQYROYLQJ SODQW GLVHDVH UHVLVWDQFH DQG VXVFHSWLELOLW\ $OWKRXJK JHQHIRUJHQH LQWHUDFWLRQV FDQ DSSHDU WR EH VLPLODU JHQHWLFDOO\ WKH\ PD\ EH TXLWH GLIIHUHQW LQ WHUPV RI ELRFKHPLFDO PHFKDQLVPV

PAGE 117

&+$37(5 6800$5< $1' &21&/86,216 0HPEHUV RI D ;DQWKRPRQDV DYLUXOHQFH DYUf JHQH IDPLO\ ZHUH VKRZQ WR SOD\ DQ LPSRUWDQW UROH LQ GHWHUPLQLQJ KRVWVSHFLILF SDWKRJHQLF V\PSWRPV RI FRWWRQ EOLJKW DQG FLWUXV FDQNHU 7KLV KRVWVSHFLILF SDWKRJHQLFLW\ SWKf IXQFWLRQ PD\ EH FULWLFDO LQ FRQGLWLRQLQJ WKH KRVW UDQJH RI ;DQWKRPRQDV $OO FRWWRQ EOLJKW VWUDLQV H[DPLQHG WR GDWH RI ; FDPSHVWULV SY PDOYDFHDUXP FDUU\ WR '1$ IUDJPHQWV WKDW K\EULGL]H ZLWK PHPEHUV RI WKLV JHQH IDPLO\ 7ZHOYH K\EULGL]LQJ IUDJPHQWV KDYH QRZ EHHQ LVRODWHG IURP D VLQJOH ; FDPSHVWULV SY PDOYDFHDUXP VWUDLQ ;FP+ 6L[ RI WKHP DYU% DYUE DYUE DYU%OQ DYU%O2O DQG DYU%O27f DUH FOXVWHUHG RQ D NE LQGLJHQHRXV SODVPLG DQG ZHUH SUHYLRXVO\ LVRODWHG DV IXQFWLRQDO DYU JHQHV 'H )H\WHU DQG *DEULHO Df 7KH RWKHU VL[ DSSHDU WR EH FKURPRVRPDO IUDJPHQWV DQG ZHUH LVRODWHG LQ WKLV VWXG\ 2QH RI WKHVH HQFRGHV WKH SUHYLRXVO\ GHVFULEHG DYU%Q *DEULHO HW DO f DQG WKUHH RI WKHVH HQFRGH SUHYLRXVO\ XQGHVFULEHG DYU JHQHV DYU% DYU% DQG DYU%Of 7R LQYHVWLJDWH SRVVLEOH SOHLRWURSLF IXQFWLRQV RI WKHVH DYU JHQHV LQ ;FP+ PDUNHU H[FKDQJH PXWDJHQHVLV DQG FRPSOHPHQWDWLRQ DQDO\VHV ZHUH FDUULHG RXW $ VLQJOH PXWDWLRQ RI DYUE EXW QRW RWKHU PHPEHUV LQ ;FP+ UHVXOWHG LQ VLJQLILFDQW UHGXFWLRQ RI WKH SDWKRJHQfV DELOLW\ WR LQGXFH ZDWHUVRDNLQJ V\PSWRPV RQ FRWWRQ VXVFHSWLEOH OLQH $FDOD $Q DYHUDJH RI WLPHV PRUH EDFWHULD ZHUH UHOHDVHG RQWR WKH FRWWRQ OHDI

PAGE 118

VXUIDFH IURP ZDWHUVRDNHG VSRWV FDXVHG E\ ;FP+ WKDQ IURP WKRVH FDXVHG E\ DQ LVRJHQLF DYUEa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a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

PAGE 119

WKDW SWK$ LV b LGHQWLFDO WR DZ%V DQG DYU%V RI ; FDPSHVWULV SY YHVLFDWRULD b LGHQWLFDO WR DYUE RI ; FDPSHVWULV SY PDOYDFHDUXP DQG b LGHQWLFDO WR DYU;DO2 RI ; RU\]DH +RSNLQV HW DO f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

PAGE 120

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

PAGE 121

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

PAGE 122

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

PAGE 123

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f§ ,QWHUDFW 'H )H\WHU 5 DQG *DEULHO ': E 8VH RI FORQHG '1$ PHWK\ODVH JHQHV WR LQFUHDVH WKH WUDQVIHU IUHTXHQF\ RI IRUHLJQ JHQHV LQWR ;DQWKRPRQDV FDPSHVWULV SY PDOYDFHDUXP %DFWHULRO 'H )H\WHU 5 .DGR &, DQG *DEULHO ': 6PDOO VWDEOH VKXWWOH YHFWRUV IRU XVH LQ ;DQWKRPRQDV *HQH 'H )H\WHU 5
PAGE 124

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
PAGE 125

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
PAGE 126

*LOPRUH -U 5 &RPSDULVRQ RI WKH URPS$ JHQH UHSHDW UHJLRQV RI 5LFNHWWVLDH UHYHDOV VSHFLHVVSHFLILF DUUDQJHPHQWV RI LQGLYLGXDO UHSHDWLQJ XQLWV *HQH *RHEO 0 DQG
PAGE 127

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
PAGE 128

.HHQ 17 DQG %X]]HO 5, 1HZ GLVHDVH UHVLVWDQFH JHQHV LQ VR\EHDQ DJDLQVW 3VHXGRPRQDV V\ULQJDH SY JO\FLQHD (YLGHQFH WKDW RQH RI WKHP LQWHUDFWV ZLWK D EDFWHULDO HOLFLWRU 7KHRU $SSO *HQHW .HHQ 17 DQG 6WDVNDZLF] %+RVW UDQJH GHWHUPLQDQWV LQ SODQW SDWKRJHQV DQG V\PELRQWV $QQX 5HY 0LFURELRO .HHQ 17 7DPDNL 6 .RED\DVKL *HUKROG 6WD\WRQ 0 6KHQ + *ROG 6 /RUDQJ 7KRUGDOFKULVWHQVHQ + 'DKOEHFN DQG 6WDVNDZLF] % %DFWHULDO H[SUHVVLQJ DYLUXOHQFH JHQH SURGXFH D VSHFLILF HOLFLWRU RI WKH VR\EHDQ K\SHUVHQVLWLYH UHDWLRQ 0RO 3ODQW0LFUREH ,QWHUDFW .HUU /' ,QRXH DQG 9HUPD ,0 6LJQDO WUDQVGXFWLRQ WKH QXFOHDU WDUJHW &XUUHQW 2SLQLRQ &HOO %LRO .LHEHU -5RWKHQEHUJ 0 5RPDQ )HOGPDQQ .$ DQG (FNHU -5 &75 D QHJDWLYH UHJXODWRU RI WKH HWK\OHQH UHVSRQVH SDWKZD\ LQ $UDELGRSVLV HQFRGHV D PHPEHU RI WKH UDI IDPLO\ RI SURWHLQ NLQDVHV &HOO .LQJVOH\ 07 *DEULHO ': 0DUORZ *& DQG 5REHUWV 3' 7KH RSV;ORFXV RI ;DQWKRPRQDV FDPSHVWULV DIIHFWV KRVW UDQJH DQG ELRV\QWKHVLV RI OLSRSRO\VDFFKDULGH DQG H[WUDFHOOXODU SRO\VDFFKDULGH %DFWHULRO .LVKLPRWR 7 7DJD 7 DQG $NLUD 6 &\WRNLQH VLJQDO WUDQVGXFWLRQ &HOO .OHPHQW = +\SHUVHQVLWLYLW\ 3DJHV LQ 3K\WRSDWKRJHQLF 3URNDU\RWHV 9RO ,, 06 0RXQW DQG *+ /DF\ HGV $FDGHPLF 3UHVV 1HZ
PAGE 129

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

PAGE 130

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f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

PAGE 131

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

PAGE 132

6FKRXOWLHV &/ &LYHUROR (/ 0LOOHU -: 6WDOO 5( .UDVV &3RH 65 DQG 'X&KDUPH (3 &LWUXV FDQNHU LQ )ORULGD 3ODQW 'LV f§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
PAGE 133

7DQJ -/ /LX <1 %DUEHU &( 'RZ -0 :RRWWRQ -& DQG 'DQLHOV 0*HQHWLF DQG PROHFXODU DQDO\VLV RI D FOXVWHU RI USI JHQHV LQYROYHG LQ SRVLWLYH UHJXODWLRQ RI V\QWKHVLV RI H[WUDFHOOXODU HQ]\PHV DQG SRO\VDFFKDULGH LQ ;DQWKRPRQDV FDPSHVWULV SDWKRYDU FDPSHVWULV 0RO *HQ *HQHW 7D\ORU 0* DQG 9DVLO ,. +LVWRORJ\ RI DQG SK\VLFDO IDFWRUV DIIHFWLQJ *86 H[SUHVVLRQ LQ SHDUO PLOOHW ^3HQQLVHWXP JODXFXP /f 5%Uf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

PAGE 134

9HUPD -3 %DFWHULDO EOLJKW RI FRWWRQ &5& 3UHVV ,QF %RFD 5DWRQ )ORULGD :DOOLV $ ( DQG 0F0DVWHU : 5 ,GHQWLILFDWLRQ RI /HLVKPDQLD JHQHV HQFRGLQJ SURWHLQV FRQWDLQLQJ WDQGHPO\ UHSHDWLQJ SHSWLGHV ([S 0HG :DQH\ 9 5 .LQJVOH\ 0 7 DQG *DEULHO : ;DQWKRPRQDV FDPSHVWULV SY WUDQVOXFHQV JHQHV GHWHUPLQLQJ KRVWVSHFLILF YLUXOHQFH RQ FHUHDOV LGHQWLILHG E\ 7QJXt LQVHUWLRQ PXWDJHQHVLV 0RO 3ODQW0LFUREH ,QWHUDFW :DUG ( DQG %DUQHV : 9LU' SURWHLQ RI $JUREDFWHULXP WXPDIDFLHQV YHU\ WLJKWO\ OLQNHG WR WKH f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
PAGE 135

%,2*5$3+,&$/ 6.(7&+
PAGE 136

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

PAGE 137

, FHUWLI\ WKDW KDYH UHDG WKLV VWXG\ DQG WKDW LQ P\ RSLQLRQ LW FRQIRUPV WR DFFHSWDEOH VWDQGDUGV RI VFKRODUO\ SUHVHQWDWLRQ DQG LV IXOO\ DGHTXDWH LQ VFRSH DQG TXDOLW\ DV D GLVVHUWDWLRQ IRU WKH GHJUHH RI 'RFWRU RI 3KLORVRSK\ 'DU\O 5 3ULQJ 3URIHVVRU RI 3ODQW 3DWKRORJ\ 7KLV GLVVHUWDWLRQ ZDV VXEPLWWHG WR WKH *UDGXDWH )DFXOW\ RI WKH &ROOHJH RI $JULFXOWXUH DQG WR WKH *UDGXDWH 6FKRRO DQG ZDV DFFHSWHG DV SDUWLDO IXOILOOPHQW RI WKH

PAGE 138

/' L Q]R c: \A 2 A 9} m OPO


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 E983UCXGR_OSDIDH INGEST_TIME 2017-07-13T15:22:49Z PACKAGE AA00003644_00001
AGREEMENT_INFO ACCOUNT UF PROJECT UFDC
FILES