Molecular analysis of genetic diversity and variability in Colletotrichum gloeosporioides

MISSING IMAGE

Material Information

Title:
Molecular analysis of genetic diversity and variability in Colletotrichum gloeosporioides
Physical Description:
viii, 148 leaves : ill.,photos ; 29 cm.
Language:
English
Creator:
Liyanage, Hemachandra D., 1958-
Publication Date:

Subjects

Subjects / Keywords:
Colletotrichum gloeosporioides -- Analysis   ( lcsh )
Fungal diseases of plants   ( lcsh )
Citrus -- Diseases and pests   ( lcsh )
Genre:
bibliography   ( marcgt )
theses   ( marcgt )
non-fiction   ( marcgt )

Notes

Thesis:
Thesis (Ph. D.)--University of Florida, 1992.
Bibliography:
Includes bibliographical references (leaves 131-147).
Statement of Responsibility:
by Hemachandra D. Liyanage.
General Note:
Typescript.
General Note:
Vita.

Record Information

Source Institution:
University of Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
aleph - 001800203
notis - AJM3952
oclc - 27569324
System ID:
AA00003699:00001

Full Text












MOLECULAR ANALYSIS OF GENETIC DIVERSITY AND VARIABILITY IN
COLLETOTRICHUM GLOEOSPORIOIDES














BY

HEMACHANDRA D. LIYANAGE


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

1992


UF FL:Nf L!ES


































Dedicated
to
Mother















ACKNOWLEDGEMENTS


I sincerely thank Dr. Corby Kistler for offering me

this opportunity to work in his lab, for all the advice,

guidance, constructive criticism and help throughout the

research study and my career. I am very thankful to Dr. R.

T. McMillan, Jr., for providing financial arrangements and

for being helpful in many ways. I gratefully acknowledge

the willing assistance and advice given by Dr. Frank Martin

and Dr. Ron Sonoda. I thank my committee members, Dr. Daryl

Pring, Dr. James Kimbrough, and Dr. Curtis Hannah for their

assistance and guidance throughout this study.

A special word of thanks to my loving wife, Thamara,

who spent her time and effort helping me throughout the

period of study and in all my difficult times.


iii
















TABLE OF CONTENTS

Page

ACKNOWLEDGEMENTS................................... iii

ABSTRACT................. ............................ vii

CHAPTERS

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

2 OBSERVED VARIABILITY IN COLLETOTRICHUM
GLOEOSPORIOIDES CAUSING POST BLOOM FRUIT
DROP IN CITRUS ................................. 7

Introduction.................................... 7
Materials and Methods......................... 9
Strains of Colletotrichum gloeosporioides. 9
Pathogenicity............................. 10
Benomyl Tolerance......................... 10
Results.......................................... 11
Colletotrichum gloeosporioides Strains
from Citrus are Morphologically
Variable................................. 11
Colletotrichum gloeosporioides Strains
have Different Nuclear Numbers
in their Spores......................... 13
Both Type 1 and Type 2 Strains are
Pathogenic to Tahiti Lime Flowers....... 15
Type 1 and Type 2 Strains Differ in their
Tolerance to Benomyl.................... 15
Discussion....................................... 16

3 DNA POLYMORPHISMS FOUND AT MANY GENETIC LOCI
EXAMINED IN COLLETOTRICHUM GLOEOSPORIOIDES... 25

Introduction................... ....... ...... 25
Ribosomal DNA in Fungi.................... 25
Ribosomal DNA is Polymorphic in Many
Fungi................................... .27
Fungal Cutinase Genes and Cutinase
Isozymes................................ 31
Restriction Fragment Length Polymorphisms
(RFLP) in Fungi. .......... .. ........... 36










Page


Materials and Methods.......................... 39
Strains of Colletotrichum gloeosporioides. 39
DNA Extraction ........................... 40
DNA Cloning and Restriction Enzyme Mapping 41
Enzyme Assays and Electrophoresis of
Cutinase .............................. 42
Probes Containing Cutinase Gene Sequences. 43
Detection of Restriction Fragment Length
Polymorphisms............ .... .............. 44
Results.......................................... 45
Ribosomal DNA is Polymorphic in
Colletotrichum gloeosporioides........... 45
Ribosomal RNA Genes....................... 47
Diverse Cutinases and Cutinase Genes are
Found in Type 1 and Type 2 Strains of
Colletotrichum gloeosporioides.......... 49
Subgroups of Colletotrichum gloeosporioides
have Distinct RFLP Patterns Detected by
Many Genetic Markers................... 53
Discussion...................................... 73

4 VARIABILITY OF MOLECULAR KARYOTYPES AND
CHROMOSOMOL DNAS IN COLLETOTRICHUM
GLOEOSPORIOIDES.............................. 81

Introduction ...................... ..... .... 81
Pulsed Field Gel Electrophoresis.......... 81
Molecular Karyotypes of Fungi.............. 82
Materials and Methods......................... 85
Strains of Colletotrichum gloeosporioides. 85
Preparation of Protoplast Plugs............ 85
Electrophoresis and Southern Analysis..... 86
Results........................ ................ 87
Discussion............................... ...... 97

5 GENERAL DISCUSSION AND CONCLUSIONS............. 99

APPENDICES

A STRAINS OF COLLETOTRICHUM GLOEOSPORIOIDES..... 102

B ANALYSIS OF VARIANCE TABLES.................... 104

C PROCEDURES FOR DNA LABELLING AND SOUTHERN
HYBRIDIZATION...... ......................... 110

D CALCULATED MB SIZES FOR CHROMOSOMAL DNAS IN
COLLETOTRICHUM GLOEOSPORIOIDES............... 112











pace

LITERATURE CITED.................................... 131

BIOGRAPHICAL SKETCH................................. 148















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 ANALYSIS OF GENETIC DIVERSITY AND VARIABILITY IN
COLLETOTRICHUM GLOEOSPORIOIDES

By

Hemachandra D. Liyanage

August 1992

Chairman: Dr. R. T. McMillan, Jr.
Cochairman: Dr. Corby H. Kistler
Major Department: Plant Pathology

Results from this study suggest two distinct genetic

subpopulations of Colletotrichum gloeosporioides from citrus

based on DNA variation, cultural morphology, and growth.

Type 1 strains are slow growing, morphologically stable,

benomyl tolerant, and contain a single form of ribosomal DNA

(rDNA) as detected by common HindIII, PstI, SphI and SstI

fragments hybridizing to cloned Neurospora crassa rDNA. Type

2 strains are faster growing, morphologically less stable,

benomyl sensitive, and have rDNA distinct from type 1

strains. The rDNA from type 1 and type 2 strains were cloned

and mapped for 10 restriction enzyme sites and genes coding

for large subunit, small subunit and 5.8S rRNA. A subclone

constructed from the non-transcribed spacer region of type 1

rDNA clone hybridizes only to rDNA from type 1 strains. DNA

polymorphisms detected by heterologous hybridization with

vii









cloned N. crassa genes for glutamate dehydrogenase,

anthranilate synthetase, histidinol dehydrogenase, and S-

tubulin corresponded to type 1 or type 2 strains. All

strains liberate free fatty acid from [3H]-labelled cutin

and hydrolyze cutin model substrates. Serine esterases from

extracellular fluids of cutin-grown C. gloeosporioides

strains were detected by labelling proteins separated by

sodium dodesyl sulfate polyacrylamide gel electrophoresis

with 3H-diisofluorophosphate. The two major esterases from

type 1 strains have molecular weights of 26 and 20

kilodaltons (kd) whereas the type 2 esterases were 24 and 22

kd. A DNA probe containing the cloned cutinase gene from C.

gloeosporioides hybridized strongly to DNA from type 2

strains but poorly to type 1 strains. Distinct cutinase

genes may be present in the two types of C. gloeosporioides

strains from citrus. Chromosome-sized DNAs separated by

pulsed-field gel electrophoresis corresponded to type 1 or

type 2 strains. Type 1 strains had five large chromosomal

DNAs 7.6, 7.0, 4.7, 3.7, and 3.3 (or 2.8) million base pairs

(Mb) in size and one or two smaller chromosomes (1.6 to 0.63

Mb). Type 2 strains had three large chromosomal DNAs (7.8,

4.7, and 3.7 Mb) and two to four smaller chromosomal DNAs

(0.52-0.28 Mb).


viii















CHAPTER 1
INTRODUCTION


Historically, the study of plant diseases dates back to

Theophrastus (371-287 B.C.) who first described disease

conditions of plants, mostly cereal rusts, in Historia

plantarum and De causis plantarum (Ainsworth 1981).

Afterwards, studies of causal agents of plant diseases and

control measures played a major role in human survival. The

period from the mid-eighteenth to the mid-nineteenth century

was marked by the accumulation of experimental evidence for

the pathogenicity of fungi to plants. Almost all groups of

fungi include some plant pathogenic species but the greatest

number of plant pathogens is to be found among the imperfect

fungi (Ainsworth 1971).

The most important technique for the identification of

plant pathogenic fungi has always been macroscopic and

microscopic morphological examination. Morphology always

took precedence over other considerations in describing

genera and species of plant pathogenic fungi (Ainsworth

1981). The genus Colletotrichum Corda was established in

1831 and was characterized by having setose acervuli

containing hyaline, curved fusiform conidia (Baxter et al.

1985). However, there was always confusion in describing









2

fungi to this genus due to similar morphological characters

of Vermicularia Tode and Gloeosporium Desm. & Mont. (Dickson

1925; Duke 1928; Arx 1957; Baxter et al. 1985). Duke (1928)

suggested that type species of Vermicularia and

Colletotrichum represented the same fungus. Species in the

genus Gloeosporium probably represent the same fungi as in

the genus Colletotrichum because the Gloeosporium species,

which supposedly lack setae, were found to produce them on

certain substrates (Baker et al. 1940). Arx (1957) accepted

Colletotrichum and Vermicularia as separate genera while

rejecting the more heterogenous genus Gloeosporium.

The species concept of Colletotrichum gloeosporioides

(Penz.) is still uncertain (Van Der Aa et al. 1990).

Colletotrichum gloeosporioides was first described in 1882

by Penzig as Vermicularia gloeosporioides, and in 1887 it

was renamed Colletotrichum gloeosporioides (Burger 1921).

The presence of this fungus in the United States was first

observed in 1886 in Florida and was reported by Underwood

(1891). Arx (1957) recognized eleven species in the genus

Colletotrichum, and the name C. gloeosporioides with nearly

600 synonyms was maintained to designate the variable

anamorph of Glomerella cingulata (Stonem.) Spauld. & Schr.

He recognized nine forms within the species C.

gloeosporioides. Arx (1970, 1987) introduced the concept of

host forms of C. gloeosporioides but did not accept these

forms as species or intraspecific taxa with certainty.










3

Sutton (1980) considered C. gloeosporioides a group species

showing excessively wide variation.

The conidia of C. gloeosporioides are straight, obtuse

at the apex, 9-24 x 3-4.5 pm and appressoria are 6-20 x 4-12

jm, clavate or irregular, sometimes becoming complex (Sutton

1980).

Colletotrichum gloeosporioides is a ubiquitous fungus

and often causes a variety of diseases commonly known as

anthracnose on fruits, leaves and stems of a wide range of

host species. The host range of this fungus is so wide that

nearly 200 susceptible host species were listed under C.

gloeosporioides in the Index of Plant Diseases in Florida

(Alfieri et al. 1984). Many tropical fruit crops are

attacked by this fungus in the field and in post-harvest

condition (Nolla 1926; Simmonds 1965; Brown 1975). Citrus is

one of the major fruit crops attacked by this fungus, and

the diseases of citrus caused by C. gloeosporioides have

been known since 1886 when it was first isolated from citrus

plants in Florida, U.S.A. (Underwood 1891). Rolfs (1904 and

1905) described a group of citrus diseases (wither tip, leaf

spot, lemon spot, canker, and anthracnose) caused by C.

gloeosporioides. A more recently described citrus disease,

post bloom fruit drop, PFD (Fagan 1979; Sonoda and Pelosi

1988; McMillan and Timmer 1989) is caused by the same

species. The name PFD was suggested by Fagan (1979) to

distinguish this disease condition of citrus characterized









4

by premature fruit drop or blossom blight from normal

physiological thinning of fruits. The symptoms first appear

as small, brown spots on flower buds or light pink water-

soaked spots on open petals. These spots may enlarge and

rapidly cover the petals within 24 h. Afterwards, the petals

become brown and desiccated. Eventually, young fruitlets

become discolored, and they abscise, leaving the calyxes

behind as persistent buttons (Figure 1). The disease is

economically important in regions where citrus is grown

(Denham and Waller 1981; Fagan 1984a). In Florida the

disease has been reported from all commercially grown citrus

(Sonoda and Pelosi 1988; McMillan and Timmer 1989).

The causal agent of PFD of citrus was identified as C.

gloeosporioides (Fagan 1979; Sonoda and Pelosi 1988;

McMillan and Timmer 1989). Colletotrichum gloeosporioides

isolated from citrus diseases were reported to be variable

in morphology and pathogenicity (Burger 1921). Morphological

variability also has been observed in the strains of this

fungus causing PFD (Denham and Waller 1981; Sonoda and

Pelosi 1988). Three strains C. gloeosporioides varying in

morphology and pathogenicity were reported to be associated

with diseased plants by Fagan (1980). Because of the

inconsistency of the morphological characteristics, it is

uncertain whether the strains or forms of C. gloeosporioides

recognized by morphological criteria alone are really

different at genetic and molecular level. The morphological









5

changes could be caused by environmental effects, genetic

differences or both. Therefore, study of this group species

at the molecular level to understand the genetic and

molecular differences among the strains is important.

The present study was undertaken to investigate the

morphological variability and potential genetic variation of

C. gloeosporioides at the molecular level. The objectives of

this study are:

1. To examine the basis for morphological variability

of C. gloeosporioides causing PFD disease of citrus.

2. To investigate genetic variation of C.

gloeosporioides at the molecular level.

3. To examine variations in chromosome-size DNA and to

describe the molecular karyotypes.

Morphological and growth diversity arising from single

spore cultures of different C. gloeosporioides strains is

examined in Chapter 2. Molecular investigations using

genetic markers were carried out to study the variation

within citrus strains of C. gloeosporioides, and they are

reported in Chapter 3. Chapter 4 describes the chromosomal

variation of C. gloeosporioides strains and the molecular

karyotypes of the fungus. The results obtained in these

studies are reviewed comprehensively in Chapter 5, and a

concept of genetically distinct subpopulations of C.

gloeosporioides is proposed.











































Figure 1 Symptoms of post bloom fruit drop disease on sweet
orange (Citrus sinensis var. Valencia) caused by
Colletotrichum gloeosporioides.















CHAPTER 2
OBSERVED VARIABILITY IN COLLETOTRICHUM GLOEOSPORIOIDES
CAUSING POST BLOOM FRUIT DROP IN CITRUS


Introduction


More than six hundred synonyms for the fungal species

Colletotrichum gloeosporioides have been published (Arx

1957). This reflects the considerable amount of diversity

and variability observed for the fungus by different

investigators throughout the world. Sutton (1980) could not

give a standard morphological description of C.

gloeosporioides. He considered the different forms of this

fungus to be within a group species.

The association of C. gloeosporioides with citrus dates

back to 1886-1891 (Underwood 1891). The fungus has been

found to cause various diseases on this crop for the past

century (see Chapter 1). The most thorough studies on the

morphological variation of C. gloeosporioides were done by

Burger (1921). The fungus he studied was the causal agent of

bloom drops and leaf spots in citrus. Cultural

characteristics such as mycelial color, growth and

sporulation enabled him to classify C. gloeosporioides

strains into five groups. However, some strains did not fit

into any of the morphological classes due to inconsistency









8

of mycelial and sporulation characteristics in continuous

culture.

Mycelial sectors distinguished by growth and color

differences within single spore cultures of strains are

another type of variability observed in C. gloeosporioides.

Burger (1921) observed black and white mycelial sectors in

single spore cultures of the fungus. When single spored,

these black and white sectors were able to maintain their

identity in continuous culture.

Burger (1921), after studying cultural characteristics,

spore dimensions, and sectoring, concluded that C.

gloeosporioides is constantly giving off new types under

natural conditions as well as in artificial cultures. He

further suggested that these variabilities of C.

gloeosporioides may have arisen from environmental effects

as well as from high frequency mutations.

Morphological and pathogenic variability in C.

gloeosporioides causing PFD of citrus has been reported by

Fagan (1979,1980) and Denham and Waller (1981). Three

different forms of C. gloeosporioides were recognized by

Fagan (1980). Two forms, cgm with gray to dark gray mycelium

and cgc with light gray mycelium, were isolated from

senescent leaves and were nonpathogenic to citrus flowers.

The pathogenic form, cgp, had off-white to pink mycelium and

was isolated from floral parts of citrus. Fagan (1980)

concluded that at least two strains of C. gloeosporioides









9

causing PFD occurred in Belize. These strains corresponded

to morphological groups of C. gloeosporioides described by

Burger (1921).

The objective of this study is to examine the

morphological and phenotypic diversity of C. gloeosporioides

causing PFD of Tahiti lime (Citrus aurantifolia Swingle) and

Sweet orange (Citrus sinensis Osbek).

Materials and Methods


Strains of Colletotrichum gloeosporioides


Strain number, host, place and year of isolation are

tabulated in Appendix A. Isolation of C. gloeosporioides

from host plants was carried out as follows. Host plant

tissues were surface sterilized in 1% sodium hypochlorite

(Clorox Co., Oakland, CA) for 30-60 s, rinsed 3 times with

sterilized water and plated on potato dextrose agar (PDA,

Difco laboratories, Detroit, MI) plates. Edges from growing

mycelia were isolated and maintained in the laboratory as

strains. Strains were grown in 20% (w/v) V-8 juice (Campbell

Soup Co., Camden, NJ) for 7 d at 250 rpm on a Lab-Line orbit

shaker (Lab-Line Instruments Inc., Melrose Park, IL). Spores

were collected by centrifugation at 7000 x g for 5 min and

washed 2 times with sterilized water before storing in 50%

glycerol (in water) at -800C. To obtain single spore

cultures, spores were spread on PDA plates; 14-16 h later,

germinating spores were isolated under a dissecting









10

microscope (25x10 magnification) and plated on PDA plates.

Morphology of colony growth, mycelial color and sectoring

were examined in PDA culture and still liquid culture,

potato dextrose broth (PDB Difco laboratories, Detroit, MI).

To examine the nuclear number, spores were stained with

1% aniline blue (Sigma Chemical Co., St. Louis, MO) in 50%

glycerin in water (Tu and Kimbrough 1973). To stain nuclei,

a drop of spores in water was placed on a microscopic slide,

and a drop of stain was added. The slide was then heated

over a flame for 5-10 seconds. Approximately 1000 spores

were examined for each strain.


Pathogenicity


All the strains were tested for their ability to infect

flowers of Tahiti lime under natural conditions in the field

as well as in the laboratory. Strains were grown in 20% V-8

juice for 7 days and inoculum containing 107 spores ml'1

water were prepared. Tahiti Lime flowers were sprayed using

a hand sprayer to wetness with inoculum or water, and

symptom development was observed for 3 days. Each treatment

contained 10-15 flowers. The control was sprayed with water.


Benomvl Tolerance


The growth of C. gloeosporioides strains was examined

in PDA medium containing 0, 2 and 10 pg benomyl (methyl-

(butyl carbamoyl)-2-benzimidazolecarbamate, Sigma Chemical









11

Co., St. Louis, MO) ml-1. Radial growth of the mycelial

colony was measured every 24 h for a 10 day period. Growth

rates in mm h-' were estimated by the slopes obtained with

linear regression analysis of the growth curve. A comparison

of slopes was made using analysis of variance (Appendix B).

Each treatment was replicated 5 times, and the experiment

was repeated once with 2 replicates.


Results


Colletotrichum gloeosporioides Strains from Citrus are
Morpholocicallv Variable


The C. gloeosporioides strains examined can be grouped

into two major categories based on morphology and growth

characteristics. Type 1 strains (H-l, H-3, H-9, H-21, H-22,

H-25B, H-36, IMB-3, LP-1, Maran, OCO, and TUR-1) produce

morphologically stable and relatively slow-growing mycelial

colonies in PDB. The colonies are orange-colored and have

appressed mycelia with abundant sporodochia (Figure 2.1).

Type 2 strains (H-4, H-ll, H-12, H-23, H-24, H-46, H-47, H-

48, 180269 and 226802) grow faster and produce mostly gray,

fluffy mycelial colonies (Figure 2.1). The type 1 strains

grow at a significantly slower rate from 0.008 to 0.10 mm

h'1; type 2 strains grow significantly faster from 0.12 to

0.15 mm/ h"1 as calculated by slopes of linear regression

data (Table 2.1). The strain types also differ in culture












Table 2.1 Effect of benomyl concentration on the estimated
radial growth rates in mm h'1 of Colletotrichum
gloeosporioides type 1 and type 2 strains.


Strain Benomyl concentration jg/ml
0 2 10


Type 1

H-1

H-3

H-9

H-25B

H-3 6

IMB-3

LP-1

Maran

OCO

Type 2

H-4

H-1l

H-12

H-46

H-47

H-48

180269

226802


0.10

0.10

0.10

0.041

0.095

0.008

0.10

0.041

0.095


0.141

0.125

0.121

0.133

0.133

0.145

0.133

0.150


0.033

0.033

0.045

0.041

0.037

0.008

0.050

0.029

0.037



0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00


0.029

0.033

0.041

0.041

0.033

0.004

0.041

0.033

0.033



0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00









13

stability as determined by their ability to produce sectors

of different color, morphology and growth habit. To

quantitate these levels of instability, 100 conidia were

isolated from three type 1 strains and two type 2 strains

and tested for morphological stability. One hundred single

spore cultures from strains H-l, H-3, and H-25B (type 1)

grown in PDB were found to produce identical colonies. One

hundred single spore cultures from strain H-12 and H-48

(type 2) produced 100% sectoring colonies. These colonies

varied in colony color from dark gray to gray, white, and

orange with different growth rates (Figures 2.2, 2.3, and

2.4). Sporodochia production was scattered or inhibited but

could be stimulated by mycelial injury (Figure 2.5). One

hundred injury-induced spores from an H-48 gray mycelial

sector produced 50 sectoring colonies, 24 dark gray with no

sporodochia and 26 orange colonies with scattered

sporodochia production.


Colletotrichum gloeosporioides Strains have Different
Nuclear Numbers in their Spores


The nuclear number observed by aniline blue staining

varied from 1 to 3 per single spore (Table 2.2). All spores

examined from all isolates were single called. All the

isolates examined contained spores with more than one

nucleus as identified by dark stained objects distinguished

from the lightly stained cytoplasm under the high












Table 2.2 Percentage of spores carrying different numbers of
nuclei in Colletotrichum gloeosporioides strains.


Percentage of spores*
Strain Number of nuclei
1 2 3


H-l

H-3

H-4

H-9

H-12

H-25B

H-4 6

H-4 8

180269

226802

LP-1

Maran


92.6

97.8

94.3

96.4

98.8

96.9

93.4

92.8

92.9

94.7

99.6

98.6


6.8

2.2

5.4

3.4

1.2

2.9

6.0

6.6

6.6

5.2

0.4

1.4


0.6

<0.1

0.3

0.2

<0.1

0.2

0.6

0.6

0.5

0.1

<0.1

<0.1


*=Calculated from 1000 spores










15

magnification (46x10) of a light microscope. Spores

containing a single nucleus varied from 92.6 to 99.6% in the

15 isolates studied. The maximum number of nuclei observed

within a single spore was 3, and the percentage of spores

containing three nuclei varied from <0.1-0.6%. The

percentage of spores containing two nuclei varied from 0.4-

6.8%.


Both Type 1 and Type 2 Strains are Pathoqenic to Tahiti Lime
Flowers


Brown lesions developed in flowers individually

inoculated with all strains of the pathogen 24 h after

spraying. The petals were blighted completely at 36 h and

had dropped at 48 h. Flowers sprayed with water alone were

not blighted after 72 h. The fungal strains reisolated from

infected tissues were found to be morphologically like the

original strains. The relative virulence of strains was not

measured in this study.


Type 1 and Type 2 Strains Differ in their Tolerance to
Benomvl


All type 2 strains were completely inhibited by 2 or 10

Ag ml" benomyl in PDA, but type 1 strains were more

tolerant. Average growth rates for individual type 1 and

type 2 strains are listed in Table 2.1. Analysis of variance

showed that benomyl concentration had a significant effect

on type 1 strains. There was a significant interaction









16

between strains and concentration indicating that growth

rate of each strain may respond differently to different

concentrations of benomyl (Appendix B).



Discussion


Grouping of Colletotrichum gloeosporioides strains

based on morphological and physiological observations was

first attempted by Burger (1921). However, morphologically

based groups of C. gloeosporioides strains have been

inconsistently described in this and subsequent studies

(Burger 1921; Arx 1957; Sutton 1980). Fagan (1980), Denham

and Waller (1981), and Sonoda and Pelosi (1988) reported

morphological variations associated with this fungus

isolated from Sweet orange cultivars. The type 1 strains in

this study show similarities in morphology, growth and

sporodochia production to strains designated cgp by Fagan's

(1980) description and correspond to the orange colored,

slow-growing colonies described by Sonoda and Pelosi (1988).

The more variable type 2 strains show similarities to the

cgm and cgc strains of Fagan and correspond to faster

growing colonies described by Sonoda and Pelosi.

Both type 1 and type 2 strains were isolated from sweet

orange (C. sinensis) as well as Tahiti lime (Appendix A).

Both types were pathogenic to Tahiti lime flowers as










































Figure 2.1. Morphology of type 1 (left) and type 2 (right)
strains of Colletotrichum gloeosporioides grown
in potato dextrose broth.












































Figure 2.2 Fluffy and appressed mycelial sectors
produced by a single spore culture of
Colletotrichum gloeosporioides type 2 strain in
potato dextrose agar.












































Figure 2.3. Dark gray, light gray and orange mycelial
sectors produced by a single spore culture of a
Colletotrichum gloeosporioides type 2 strain in
potato dextrose broth.












































Figure 2.4. Multi-colored mycelial sectors produced by a
single spore culture of a Colletotrichum
gloeosporioides type 2 strain in potato dextrose
broth.











































Figure 2.5. Mycelial injury can induce type 2 (non-
sporodochia-forming) type 2 strains to form
sporodochia.









22

indicated by inoculation tests, confirming previous results

(Sonoda and Pelosi 1988). Variation in the ability of type 1

and type 2 strains to cause disease on Tahiti lime flowers

was not observed. However, only a single high inoculum

concentration was used for pathogenicity testing.

Differences in virulence perhaps could be found if a

dilution series of inoculum were used in assessing the

disease-causing potential of the strains. Sonoda and Pelosi

(1988) and Agostini et al (1992) suggested that slowly

growing, orange-colored strains (type 1 strains) were the

actual causal agent of PFD because only they could be

consistently isolated from diseased petals in the field

while the gray (type 2) strains were isolated primarily from

leaves, stems and fruit. This observation has been confirmed

by other workers (Agostini et al. 1992; Gantotti and Davis,

personal communication). Clearly, further pathogenicity

testing and field sampling are needed to confirm whether one

or both types of the pathogen are important in PFD

epidemics.

The genetic and molecular basis of the morphological

diversity caused by sectoring of C. gloeosporioides is yet

to be elucidated. One of the probable genetic explanations

for the observed morphological diversity and sectoring of

type 2 strains may be heterokaryosis followed by

parasexuality or nuclear sorting out. Heterokaryosis and

parasexuality have been found to contribute to the variation









23

of C. gloeosporioides as reviewed by Baxter et al. (1985).

The strains studied carried only <7% spores with 2-3 nuclei.

Therefore, heterokaryosis or nuclear sorting out may not be

a cause of 100% sectoring observed in type 2 strains.

Multinuclear spores have been reported in C. gloeosporioides

as well as many other fungi (Panaccione et al. 1989; Shirane

et al. 1989; TeBeest et al. 1989). The multinuclear

condition may arise from division of the nucleus without

division of the spore (Churchill 1982). Hence, it may

represent homokaryotic condition.

An interesting phenomenon observed between type 1 and

type 2 strains is the differential sensitivity to benomyl

(methyl-2-benzimidazole carbamate: active ingredient in the

commonly used fungicide benlate). Type 2 strains were

completely inhibited by the levels of benomyl tested while

type 1 strains were tolerant although their growth rates

were significantly reduced (Table 2.2). The benomyl

tolerance of type 1 strains may have practical consequences

to the control of this disease. Current control measures

include spraying benomyl to control PFD (Fagan 1984b). If

type 1 strains are the primary causal agent of PFD as

previously suggested (Sonoda and Pelosi 1988; Agostini et

al. 1992; Gantotti and Davis, personal communication),

spraying in the field may only partially inhibit the

virulent pathogen while completely eliminating the less









24

virulent form. While slowing the epidemic in the short run,

this practice may have the long-term effect of selecting for

the most virulent form of the fungus.
















CHAPTER 3
DNA POLYMORPHISMS FOUND AT MANY GENETIC LOCI EXAMINED
IN COLLETOTRICHUM GLOEOSPORIOIDES


Introduction


Ribosomal DNA in Fungi


The nuclear ribosomal RNA (rRNA) genes of eukaryotes

are clustered in tandemly repeating units known as ribosomal

DNA (rDNA) unit repeats. In fungi as in many other

eukaryotes, each rDNA unit repeat consists of coding regions

for small subunit, SSU (17-18S), 5.8S, and large subunit,

LSU (25-26S) rRNA and intervening internal and external

transcribed and non-transcribed spacer (NTS) regions

(Fedoroff 1979; Chambers et al. 1986). Each rDNA unit repeat

codes for a 35S rRNA precursor which gives rise to SSU and

LSU rRNAs. In Neurospora crassa and Saccharomyces cerevisiae

a 35-37S rRNA precursor cleaves into a 17-18S rRNA of the

small ribosomal subunit (37S), and the 5.8S and 25S rRNAS of

large ribosomal subunit (60S) required for building 80S

ribosomes (Russell et al. 1976; Bell et al. 1977; Planta et

al. 1980).

The number of times rRNA genes are repeated varies

depending on the species of the organism. It was estimated









26

that there are about 185-225 copies of rDNA unit repeats in

Neurospora crassa (Krumlauf and Marzluf 1980; Rodland and

Russell 1982), 140 copies in yeast, Saccharomyces cerevisiae

(Schweizer et al. 1969; Rubin and Sulston 1973), 59 copies

in Rhizoctonia solani (Thanatephorus praticola AG-4)

(Vilgalys and Gonzalez 1990) and 60-90 copies in Coprinus

cinereus (Cassidy et al. 1984) per haploid genome.

Another rRNA gene recognized in fungi is 5S rRNA gene.

The 5S rRNA gene may be present within the rDNA unit repeat

or dispersed elsewhere in the genome. The 5S rDNA sequences

are located within the same rDNA unit repeat in S.

cerevisiae (Bell et al. 1977), Mucor racemosus (Cihlar and

Sypherd 1980), S. rose and S. carlsbergensis (Verbeet et

al. 1983), C. cinereus (Cassidy et al. 1984), Schizophyllum

commune (Buckner et al. 1988), R. solani (Vilgalys and

Gonzalez 1990), the slime mold, Dictyostelium discoideum

(Maizels 1976), and water mold, Achlya ambisexualis, (Rozek

and Timberlake 1979). It is located elsewhere in the genome

in N. crassa (Free et al. 1979; Selker et al. 1981),

Schizosaccharomyces pombe (Tabata 1981), Aspergillus

nidulans (Borsuk et al. 1982), yeasts, Yarrowia lipolytica,

(Van Heerikhuizen et al. 1985), and Cochliobolus

heterostrophus (Garber et al. 1988).

In all known cases the 5S rRNA is transcribed

independently as a primary transcript separate from the 35-

37S rRNA precursor transcript (Udem and Warner 1972;









27

Miyazaki 1974). When the 5S rRNA gene is within the same

unit repeat the 5S rRNA gene could be located in the same

strand, transcribed in the same direction as the other rRNA

genes or in the opposite strand, and transcribed in an

antiparallel manner (Aarstad and Oyen 1975). In C. cinereus

the 5S rRNA gene is transcribed in the same direction as the

rest of the rRNA genes (Cassidy et al. 1984).


Ribosomal DNA is Polymorphic in Many Fungi


Ribosomal DNA is a unique genetic marker which can be

used in the study of relatedness among organisms. Generally

the number of rDNA unit repeats is maintained from

generation to generation of an organism. The meiotic

recombination is suppressed within the rDNA array. In N.

crassa (Russell et al. 1988) and in C. cinerus (Cassidy et

al. 1984) the rDNA was shown to be inherited in a simple

stable Mendelian fashion exhibiting an approximately 1:1

ratio of the two parental rDNA types. No meiotic

recombinants were detected among the progeny indicating that

non-sister chromatid crossing over was highly suppressed in

the rDNA region of these organisms. However, Butler and

Metzenberg (1989 and 1990) demonstrated that N. crassa rDNA

can undergo unequal sister chromatid exchange and that the

number of rDNA unit repeats does not segregate in a simple

Mendelian fashion. Their observations suggested that









28
although the same rDNA RFLP can be inherited, the number of

unit repeats can be different from either of the parents.

Within a given species, the members of the rRNA gene

family are reasonably homogeneous in sequence, as are their

associated spacer sequences, despite frequent length and

restriction site differences among the latter. Yet there are

interspecific differences in sequence and these appear to be

much more pronounced for spacers than for genes. Smith

(1973) suggested that the differences between genes and

spacers might be in the rate at which they accumulate

mutations. Chromosomes containing mutations deleterious to

gene function would be eliminated by natural selection while

neutral spacer mutations would be retained in the

population. Hence, spacers change more rapidly than genes

simply by retaining a larger fraction of mutation (Smith

1973).

Length heterogeneity and restriction site polymorphisms

in rDNA has been commonly observed in many fungi. These

polymorphisms were common among strains of S. cerevisiae

(Petes and Botstein 1977), N. crassa (Russell et al. 1984),

S. commune (Specht et al. 1984), C. cinerus (Wu et al.

1983), and Y. lipolytica (Clare et al. 1986). Both

restriction site and length polymorphisms have been also

observed among biological species of Armillaria (Anderson et

al. 1989). Polymorphisms of rDNA in many fungi are located

within the NTS region of the rDNA unit repeat (Cassidy et









29

al. 1984; Van Heerikhuizen et al. 1985; Rogers et al. 1989).

Chambers et al. (1986) compared the 8.4 kb rDNA unit repeat

of N. intermedia and N. sitophila with the 8.7 kb long rDNA

unit repeat of N. crassa and found that the 300 bp

difference was within the NTS region. Verbeet et al. (1983),

comparing S. rose and S. carlsbergensis by heteroduplex

analysis, concluded that the NTS regions are largely non-

homologous in sequence whereas the transcribed regions are

essentially homologous. Russell et al. (1984) studied the

organization of the rDNA unit repeat in the strains of N.

crassa, N. tetrasperma, N. sitophila, N. intermedia, and N.

discreta and found that the size of the unit repeat has been

highly conserved among the strains of Neurospora. However, a

restriction enzyme site polymorphism in the NTS region was

found between the strains. This restriction site

polymorphism was strain-specific and not species-specific.

Restriction enzyme mapping of rDNA in yeast, Kluyveromyces

species has shown a length variation, and the variability

was found to reside in the NTS region (Lachance 1989).

Martin (1990) reported the presence of restriction site and

length polymorphisms within single oospore isolates of the

Oomycete genus Pythium, and the differences were found

within the NTS region and 3' end of the 26S coding region.

The NTS region has also been useful to study the

phylogenetic relatedness among fungal species and other

organisms (Verma and Dutta 1987).









30

The transcribed intergenic spacer (ITS) has also been

shown to be variable in fungi. In S. commune location of

strain-dependent length polymorphisms resided in the ITS

region between 18S and 5.8S cistrons (Buckner et al. 1988).

Chambers et al. (1986) compared the sequences of ITS regions

for N. crassa and S. carlsbergensis, and found that there is

a general lack of homology between the internal transcribed

spacer regions between 5.8S and 26S rRNA genes of these two

species. Buchko and Klassen (1990), using PCR technique to

amplify the ITS region, demonstrated length heterogeneity in

strains of Pythium ultimum.

The locations of rDNA polymorphisms were not confined

to the ITS regions. Polymorphisms within the coding regions

of rRNA genes due to addition or deletion of restriction

enzyme sites were found in fungi (Chambers et al. 1986).

Another cause of rDNA polymorphism in eukaryotes is the

presence of introns in the coding regions. In fruit fly,

Drosophila melanogastor, the presence of an intron in the

coding region of the 28S rRNA gene has given rise to

polymorphism (Glover and Hogness 1977) of rDNA in this

organism. In fungi there are no conclusive reports for the

presence of introns in rRNA genes. However, Buckner et al.

(1988) examining the strain-dependent rDNA length

polymorphism in S. commune suggested the possibility of

having an intron in the coding region of the 18S rRNA gene.









31

Deletions of large fragments of rDNA may also occur in

organisms as reported by Malezka and Clark-Walker (1989). A

deletion of a 300 kb chromosomal fragment containing 35-40

rRNA cistrons has given rise to a new petite positive strain

of Kluyveromyces lactis.

One of the objectives of this study is to investigate

the variation of rDNA among the strains of C.

gloeosporioides causing PFD.


Fungal Cutinase Genes and Cutinase Isozvmes


Plant pathogenic fungi penetrate their hosts through

the cuticle of epidermal cells or through cutinized cells

below natural apertures. Penetration may take place by

mechanical pressure (Brown and Harvey 1927; Brown 1936;

Pristou and Gallegly 1954; Chakravarty 1957; Wood 1960;

Meredith 1964; Bonnen and Hammerschmidt 1989b), or by

enzymatic degradation of the cuticle (De Bary 1887; Miyoshi

1895; Linskens et al. 1965; Akai et al. 1968; Kunoh and Akai

1969; Shayakh et al. 1977; Kolattukudy 1985) or by both

(Ellingboe 1968; Shishiyama et al. 1970; Nicholson et al.

1972).

The plant cuticular barrier is composed of a

biopolymer, cutin and associated waxes providing a

protective covering against pathogen invasions and hazardous

effects of environment (Martin and Juniper 1970). The

structure of cuticle varies from one plant to another, and











it is influenced by genetic background as well as

environmental factors (Martin and Batt 1958; Martin 1964).

Almost all parts of the plant, surfaces of epidermal cells

of aerial plant parts, substomatal areas, mesophyll and

palisade cells (Martin and Juniper 1970; Sitholey 1971),

flower parts, seed coat (Kolattukudy et al. 1974) fruit

(Espelie et al. 1980), roots and tubers (Kolattukudy and

Agrawal 1974; Kolattukudy et al. 1975) contain a cuticular

layer.

The biopolymer, cutin is composed of C16 and C18 hydroxy

and hydroxy epoxy fatty acids (Van den Ende and Linskens

1974; Espelie et al. 1980; Kolattukudy 1980, 1981). The

composition of the cutin polymer and the proportions of C16

and C,1 fatty acid monomers may vary depending on plant

species or varieties, organs of the same plant, or on growth

conditions (Espelie et al. 1979; Kolattukudy 1980).

The enzyme, cutinase can facilitate the hydrolysis of

cutin into its components (Baker and Bateman 1978; Dickman

et al. 1982). These hydrolysis products of cutin are also

potent inducers of the cutinase gene of the penetrating

fungus (Woloshuk and Kolattukudy 1986). A small amount of

cutinase is constitutively expressed in the fungal spore

which senses the contact with the plant cuticle via the

unique cuticle monomers generated by this small amount of

cutinase. Consequently, these monomers trigger the

expression of the cutinase gene/ genes needed for the









33

production of cutinases which eventually degrade the cuticle

(Kbller et al. 1982; Kolattukudy 1985; Woloshuk and

Kolattukudy 1986; Podila et al. 1988; Kolattukudy et al.

1989).

Many plant pathogenic fungi examined have shown

production of different levels of cutinase isozymes (Purdy

and Kolattukudy 1975a; Lin and Kolattukudy 1980; Kolattukudy

et al. 1981). Direct observational, enzymological and

histochemical evidences have suggested that cutinase is

essential for the penetration of the plant by the fungal

pathogens. Specific antibodies prepared against cutinase

from Nectria haematococca (Fusarium solani f. sp. pisi,

Shaykh et al. 1977) and/ or diisopropylfluorophosphate

(DFP), a potent inhibitor of serine esterases, can prevent

infection of the host by this fungus indicating that

cutinase plays an essential role in the infection process

(Kolattukudy 1979). Dickman and Patil (1986) obtained

cutinase-deficient mutants of C. gloeosporioides, the causal

agent of papaya anthracnose, and found that they were

nonpathogenic to the intact papaya fruit. However, these

cutinase-deficient mutants produced normal lesions when

papaya surfaces were artificially wounded or treated with

purified cutinase enzyme. Dickman et al. (1989) were able to

introduce a Fusarium cutinase gene into a wound pathogen,

Mycosphaerella species through genetic transformation. These

transformants acquired the capacity to infect intact papaya









34

fruits, and the infection by them was prevented by the

treatment of antibodies against Fusarium cutinase.

Cutinolytic enzymes have been purified and

characterized from various plant pathogens (Kolattukudy

1980, 1985; K6ller 1991) including Colletotrichum

gloeosporioides. The single enzyme produced by a strain of

C. gloeosporioides isolated from papaya fruit had a

molecular weight of 24 kd (Dickman et al. 1982) which is

very similar in size to other fungal cutinases (Kolattukudy

1980, 1985; K6ller 1991).

There is considerable heterogeneity of molecular,

immunological and enzymological properties and primary

sequences of the cutinase enzymes (Kolattukudy 1985;

Ettinger et al. 1987; Trail and Kl11er 1990). Sequence

comparison of the cutinase genes cloned from C.

gloeosporioides and N. haematococca revealed considerable

dissimilarity. Even though both cutinase genes shared

homologous regions critical for activity and structural

integrity, only 43% of the amino acids were directly

conserved (Ettinger et al. 1987). Profound differences in

cutinase appear to exist even among Colletotrichum species

(Kolattukudy 1987). A cDNA clone of the cutinase gene from

C. capsici hybridized to genomic DNA from C. graminicola and

C. gloeosporioides, but not with C. orbiculare (syn. C.

lagenarium) or C. coccodes DNA. Though not extensively

investigated, the phenomenon of cutinase diversity is also









35

reflected in enzyme kinetics and activity. For example,

cutinolytic activity of esterases purified from N.

haematococca (Purdy and Kolattukudy 1975b) and F. roseum

culmorum (Soliday and Kolattukudy 1976) was highest at

alkaline conditions (pH 10), whereas an optimum of pH 6.5

was determined for the enzyme derived from Venturia

inaequalis (Kller and Parker 1989). Baker and Bateman

(1978) assayed sixteen plant pathogenic fungi, Botrytis

cinera, B. squamosa, Cladosporium cucumerinum, C.

graminicola, N. haematococca, F. roseum, Gloeocercospora

sorghi, Helminthosporium carbonum, H. maydis (race T),

Pythium aphanidermatum, P. arrhenomanes, P. ultimum, R.

solani, Stemphylum loti, and Sclerotium rolfsii and found

that they can produce various levels of cutinase isozymes

with acidic or alkaline pH optimas. Evidence has been

presented that these differences in enzymatic properties may

allow for the tissue specificity of pathogens. Trail and

K511er (1990) reported an acidic pH optimum for the leaf

pathogen, Cochliobolus heterostrophus, pH 6.5 and an

alkaline pH optimum for the stem pathogen, R. solani, pH

8.5. The leaf and stem pathogen, Alternaria brassicola,

produced two cutinases, one with acidic and the other with

alkaline pH optima, pH 7.0 and 9.0 respectively. Differences

also have been reported for the cutinases produced by N.

haematococca and C. gloeosporioides. Only the enzyme from

the latter accepted palmitate as a substrate and the









36

specific esterase activity with both p-nitrophenol butyrate

and polymeric cutin was reported to be substantially lower

(Dickman et al. 1982).

The ability of a pathogen to produce cutinase can be

used to measure the infecting capacity of the fungal

pathogen (Dickman et al. 1982; K6ller et al. 1982). Thus the

regulation of expression of the cutinase gene could be

highly relevant to pathogenesis. Therefore, the cutinase

gene may be a good genetic marker to examine polymorphisms

among populations of fungi with differing specificities and

capabilities of causing plant disease.

One of the goals in this study was to investigate if

differences in morphologically defined type 1 and type 2

strains (see Chapter 2) are also reflected in the cutinase

isozymes and genetic organization of cutinase gene or genes.


Restriction Fragment Length Polymorphisms (RFLP) in Fungi


When fungal nuclear DNA is digested with a restriction

enzyme an enormous number of fragments generally result. In

order to study the restriction fragment pattern of DNA from

a specific chromosomal locus these fragments are size

fractionated by gel-electrophoresis, and individual

fragments are identified by Southern hybridization to

labelled probes (Southern 1975; Bernatzky 1988). Each

restriction fragment that hybridizes to a given probe

constitute a discrete chromosomal locus. Alleles can be









37

differentiated by the variation in restriction sites.

Restriction fragment length polymorphisms result from

specific differences in DNA sequence such as single base

pair substitution, additions, deletions, or chromosomal

changes (inversions and translocations) that alter the

fragment size obtained by restriction enzyme digestion.

First demonstrated by Grodzicker et al. (1974) for mapping

temperature-sensitive mutations in adenovirus, RFLP analysis

has contributed significantly in genetic analysis of many

organisms.

Genetic studies of plant pathogenic fungi have been

difficult due to lack of easily assayed genetic markers.

Restriction fragment length polymorphism has become a

popular tool for studying genetics of fungi because RFLP

markers are precise, codominant, selectively neutral, easy

to assay, and provide an unlimited number of genetic markers

(Michelmore and Hulbert 1987). Restriction fragment length

polymorphism could provide sufficient markers for the

development of detailed linkage maps for the plant

pathogenic fungi. It is also useful in studying genetic

variation, genomic organization and population genetics of

fungi. Combined with pulsed field gel electrophoresis RFLP

analysis provides a powerful tool to monitor genetic changes

throughout the genome (Michelmore and Hulbert 1987).

Analysis of RFLP markers that flank genetic loci such as

virulence genes can provide information on the genetic basis









38

of any changes in phenotype. Closely linked RFLPs can be

used as tags for important traits. With RFLP markers it is

possible to create a molecular fingerprint of specific

individuals in a population. Hence, RFLPs provide a tool for

studying asexually reproducing populations of fungi. Engels

(1981) and Hudson (1982) presented mathematical models for

the genetic determination of variation among individuals in

a population using RFLP.

Use of RFLPs to measure genetic relatedness among

strains and closely related species of plant pathogenic

fungi is still in its beginning. Genetic variability of

several plant pathogenic fungi, Armillaria mellea (Anderson

et al. 1987), Sclerotinia species (Kohn et al. 1988),

Septoria tritici (McDonald and Martinez 1990) and

Aspergillus species (Someren et al. 1991) has been studied

using RFLP genetic markers. In C. gloeosporioides two

population subgroups were recognized by distinct RFLP

patterns detected by human minisatelite probes for

hypervariable regions within the genome (Braithwaite and

Manners 1989). Linkage maps have been developed for the

lettuce downey mildew fungus, Bremia lactucae using RFLPs as

genetic markers (Hulbert and Michelmore 1988; Hulbert et al.

1988). Hulbert et al. (1988) also reported the linkage of an

avirulance gene and a RFLP locus and suggested the

possibility of cloning the avirulance gene by chromosome

walking.









39

Castle et al. (1987) distinguished the commercial

mushroom, Agaricus brunnescens, from A. bitorquis using

distinct RFLP patterns. These RFLP patterns were used in the

identification of homokaryotic, heterokaryotic and hybrid

strains of this fungus (Castle et al. 1987) Summerbell et

al. (1989) followed the segregation of RFLPs in wild

collected and artificially synthesized heterokaryotic

strains of A. brunnescens to investigate meiosis and the

meiotic recombination in this fungus.

The objective of this study is to determine if type 1

and type 2 Colletotrichum gloeosporioides are genetically

distinguishable. This will be done by examining rDNA

polymorphisms, the diversity of cutinase enzymes at the

isozyme and molecular level, and other molecular markers to

examine RFLPs in type 1 and type 2 strains of C.

gloeosporioides causing PFD of Tahiti lime and sweet orange.


Materials and Methods


Strains of Colletotrichum qloeosporioides


Strains of C. gloeosporioides, host and geographic

location are listed in the appendix A. Each strain is a

single spore culture grown. The place, year, and the host

tissue of isolation are mentioned in appendix A.











DNA Extraction


Fungal mycelium was grown in potato dextrose broth

(PDB) for seven days, harvested, frozen at -80 oC overnight

and lyophilized to complete dryness. The mycelium was ground

into a powder in liquid nitrogen using a mortar and pestle.

The mycelium powder was mixed with extraction buffer (100 mM

Tris pH 8.0, 50 mM EDTA, 100 mM NaC1, 10 mM 8-

mercaptoethanol, 1% SDS, in H20) to make a slurry and

incubated at 65 OC for 30 min. One half volume of 5 M

potassium acetate (60 ml of 5 M potassium acetate, 11.5 ml

glacial acetic acid and 28.5 ml H20) was added to samples

and incubated on ice for 30 min. The supernatant was

collected by centrifugation at 12000 x g for 15 min and was

treated with 30-50 Ag/ml of DNase free RNase (30 min at 37

OC, Sigma Chemical Co., St. Louis, MO). After RNase

treatment 200-250 gg/ml Proteinase K (Sigma Chemical Co.,

St. Louis, MO) was added and incubated for an additional 20

min. Samples were purified by phenol:isoamyl alcohol:

chloroform (25:1:24 by volume) extraction and DNA was

precipitated by addition of a two-fold volume of absolute

ethanol. DNA pellets were dissolved in 100 Al of TE (10 mM

Tris pH 8.0, 1 mM EDTA) and further purified by

precipitation with 0.7 volume of PEG/NaCl ((20% PEG 8000 in

2.5 M NaC1, Sigma Chemical Co., St. Louis, MO ) for 20-30

min on ice. Precipitated DNAs were resuspended in TE and

stored at -20 OC.









41

DNA Cloning and Restriction Enzyme Mapping


The rDNA of C. gloeosporioides was identified by

heterologous hybridization with N. crassa rDNA unit repeat

(pMF2, Free et al. 1979). Total DNA from C. gloeosporioides

strains H-25B and H-48 was digested with restriction enzyme

PstI and fractionated on a 0.7% agarose (FMC BioProducts,

Rockland, ME) gel. The piece of the gel containing the 7 to

10 kb (H-25B) or 6-10 kb (H-48) DNA range was cut out and

the DNA eluted by the freeze squeeze method (Thuring et al.

1975). The DNA was ligated to PstI-cut pUC119 (Sambrook et

al. 1989; Yanisch-Perron et al. 1985) and transformed into

Escherichia coli strain ER1647 (E. coli K-12 mcrB', Ref.

Raleigh et al. 1989; Woodcock et al. 1989) or DH5-a

(Sambrook et al. 1989). Ligation, preparation of competent

cells and transformation was carried out according to

Sambrook et al. (1989). Clones hybridizing to pMF2 were

identified and restriction mapped. For constructing

restriction maps, single and double restriction enzyme

digestions of two presumptive rDNA clones, called pCGR1 (8.4

kb from strain H-25B), pCGR2 (6.8 kb from strain H-48), were

size fractionated in 1% agarose gels. Regions homologous to

the LSU and SSU rRNA of N. crassa were mapped by Southern

hybridization to heterologous probes subcloned from plasmid

pMF2 (Free et al. 1979; Martin 1990). The probe specific to

LSU was a 1.7 kb XbaI + BamHI fragment comprising all but









42

150 bp of the 5' end of the 17S coding region. A 2.9 kb

EcoRI fragment containing all but approximately 700 bp of

the 3' end of the 26S coding region in addition to 200 bp of

transcribed spacer sequences adjacent to 5' end was used as

the probe specific to LSU. These subclones were provided by

Dr. F. N. Martin, Plant Pathology Department, University of

Florida. A probe to detect the 5.8 rRNA gene was prepared by

polymerase chain reaction using primer flanking the gene.

The primers 5' TCCGTAGGTGAACCTGCGC 3' and 5'

GCTGCGTTCTTCATCGATGC 3' amplify a 290 bp fragment which

includes the transcribed spacer of the 3' end of the SSU

rRNA gene and the entire 5.8S gene (White et al. 1990).


Enzyme Assays and Electrophoresis of Cutinase


Colletotrichum gloeosporioides cultures were grown in a

mineral medium (Hankin and Kolattukudy 1968) amended with

tritiated cutin as the sole carbon source. Esterase

activities of strains were measured using p-nitrophenyl

butyrate (PNB) and p-nitrophenyl palmitate (PNP) as model

substrates (Kller and Kolattukudy 1982; K6ller and Parker

1989; Purdy and Kolattukudy 1975a). Assays for cutinase

activity, sodium dodecyl sulfate polyacrylamide gel

electrophoresis (SDS-PAGE) and detection of active serine

esterases by tritium-labelled diisopropyl fluorophosphate

(3H-DFP) were as previously described (K611er and Parker

1989; Trail and Kller 1990). These experiments were









43
conducted by Dr. Wolfram K6ller at the New York State

Agricultural Experiment Station, Geneva.


Probes Containing Cutinase Gene Sequences


Oligonucleotide primers, 5' TGCCCCAAGGTCATCTACATC 3'and

5'GAAGTTGGAGGCCAGGTCGGC 3' were synthesized to amplify a 220

bp fragment (intron and flanking sequences) of C.

gloeosporioides cutinase gene by polymerase chain reaction

(PCR). The PCR reaction mixture was prepared in a total of

100 jl containing 100 pM of each primer, 1.25 mM each of

dATP, dTTP, dCTP and dGTP, 2 gg of template DNA, and 10 Al

of reaction buffer (50mM KC1, 10 mM Tris-HCl pH 8.3, 1.5mM

MgCl2, 0.01% gelatin). The PCR mixture was denatured by

boiling 10 min and chilled on ice before adding 2 units of

Taq DNA Polymerase (Promega Inc. Madison, WI). The PCR

temperature cycles were programmed as following in a Coy

Temp Cycler II (Coy Corp., Grass Lake, MI). Denaturation

temperature was 94 OC, annealing temperature was 37 oC and

primer extension was at 72 OC. The first cycle was run 6 min

at 940C, 2 min at 37 oC and 3 min at 720C and the subsequent

30 cycles were run at 1, 2, and 3 min time intervals,

respectively, at these temperatures. Primer extension time

for the final cycle was 10 min. The fragments amplified by

PCR were labelled with 32P dCTP or digoxigenin (dig) dUTP

(Boehringer Mannheim Corp. Indianapolis, IN; see Appendix

C). Both a 220 and a 260 bp DNA fragment amplified by PCR









44

from strain H-48 hybridized to a genomic clone containing,

the cutinase gene, a 2.2 kb SphI DNA fragment, from C.

gloeosporioides (Ettinger et al. 1987). This clone was

provided by Dr. M. B. Dickman, Department of Plant

Pathology, University of Nebraska, Lincoln, NE.

High stringency Southern hybridization (Southern 1975)

using 32P-labelled probes was carried out according to

methods described by Sambrook et al. (1989). Hybridization

and washing of blots were carried out at 68 OC. First and

second washes were with 2X SSC, 0.1% SDS and 0.1X SSC, 0.1%

SDS respectively. The conditions of low stringency Southern

hybridization were as follows. DNA hybridization was at

65 OC and first and second washes were with 2X SSC at 55 oC.

Autoradiography was performed with Kodak X-OMAT AR5 film

(Eastman Kodak Co., Rochester, NY) and Dupont Hi-Plus

intensifying screens at -80 OC.


Detection of Restriction Fragment Length Polymorphisms


Plasmids containing N. crassa genes for anthranilate

synthetase (pNC2, Schechtman and Yanofsky 1983), glutamate

dehydrogenase (pJR2, Kinsey and Rambosek 1984), histidinol

dehydrogenase (pNH60, Legerton and Yanofsky 1985), and B-

tubulin (pSV50, Vollmer and Yanofsky 1986) were used to

detect DNA polymorphisms. Clones of N. crassa genes were

obtained from the Fungal Genetics Stock Center (Department

of Microbiology, University of Kansas Medical Center, Kansas









45

City, KS). Southern hybridization of 32P-labelled probes

were carried out according to methods described by Sambrook

et al. (1989) and Appendix C.


Results


Ribosomal DNA is Polymorphic in Colletotrichum
gloeosporioides


Southern hybridization of 32P labelled pMF2 to PstI-

digested total blots from C. gloeosporioides strains

detected polymorphic forms of rDNA (Figure 3.1 and 3.2). The

type 1 strains (see Chapter 2 and Appendix A) contained only

a single form (8.4 kb PstI fragment) of rDNA (Figures 3.1

and 3.2). Although the size of the rDNA fragment in type 1

strains H-36 and OCO appears to be slightly higher than 8.4

kb in Figure 3.1, other restriction enzyme digestion tests

concluded that it is 8.4 kb in size (compare figures 3.5,

3.6, and 3.7). The 8.4 kb PstI rDNA fragment was cloned from

Type 1 strain, H-25B and will be referred to as type 1 rDNA.

The restriction fragments obtained by digesting with 10

restriction enzymes and the restriction map of cloned type 1

rDNA unit (pCGR1) for these enzymes are illustrated in Table

3.1 and Figure 3.3 respectively. The restriction map of the

cloned rDNA unit was compared with the total rDNA

restriction fragments of the strain H-25B for 7 enzymes and

was identical (Figure 3.4). The map of cloned rDNA unit from

H-25B was tested against all type 1 strains for three









46

restriction enzymes, HindIII, SphI and SstI (Figures 3.5,

3.6, and 3.7 respectively). All the type 1 strains fell into

an identical group and the hybridization fragments for the

three enzymes agreed with restriction map of the cloned rDNA

fragment from Strain H-25B. Several subcloned fragments from

the NTS region were tested to determine if they can

specifically hybridize to type 1 strains. A 0.4 kb KpnI-PstI

subclone (pCGR1N) from the 3' end of NTS region (Figure 3.3)

was found to hybridize only to type 1 rDNA. The same total

DNA blot in Figure 3.1 was reprobed with the type 1-specific

subclone after removing the previous probe and only type 1

strains show hybridization to the subclone (compare Figures

3.1 and 3.8).

Ribosomal DNA among type 2 strains (see chapter 2) was

polymorphic for PstI (Figures 3.1 and 3.2), SphI (Figure

3.6), and SstI (Figure 3.7). However, HindIII digested DNA

shows a similar pattern of rDNA polymorphism among all type

2 strains (Figure 3.5). Two hybridizing PstI fragments were

detected in strains H-48, 180269, 226802 (8.4 and 6.8 kb),

H-11 (5.0 and 3.4 kb), and H-47 (8.4 and 7.8 kb) by 32P

labelled pMF2 (Figures 3.1 and 3.2). However, the

hybridization intensity of the 8.4 kb band in strains 180269

and 226802 was very low and almost undetectable compared to

6.8 kb hybridizing band (Figure 3.1). All other type 2

strains had only one 6.8 kb hybridizing band. The 6.8 kb

PstI fragment was cloned from strain H-48, and the









47

restriction map was identical when compared with the total

rDNA restriction fragments of the strain H-48 for 7 enzymes

(Figure 3.4). The cloned 6.8 kb PstI fragment from type 2

strain H-48 hybridizing to pMF2 will be referred to as pCGR2

or type 2 rDNA. The restriction fragments obtained by

digestion of clones of type 1 and type 2 rDNA unit with

various restriction enzymes and restriction enzyme maps are

listed in Table 3.1. The length of type 1 rDNA differs from

type 2 by the size of the NTS region of the unit (Figure

3.3). The restriction map of pCGR2 is distinct from that of

type 1 rDNA and the NTS is 1.6 kb shorter.


Ribosomal RNA Genes


In addition to the length heterogeneity, type 1 and

type 2 rDNA units differ by having restriction site

polymorphisms and addition and deletion of restriction sites

within coding regions for rRNA as well as intergenic

regions. Restriction sites for SmaI and SstI within the SSU

coding region and a EcoRI site within the 5.8 S coding

region were found in type 1 rDNA. For the 10 restriction

enzyme sites examined none was detected within the SSU rRNA

and 5.8S RNA coding regions of type 2 rDNA. The coding

region for the LSU lies within BamHI and EcoRI sites for

both type 1 and type 2 rDNA, and were detected as 3.1 and

3.0 kb hybridizing fragments respectively. Restriction site












TABLE 3.1 Restriction fragments obtained by complete
digestion of the three ribosomal DNA clones


No. Restriction Fragment size (kb)
enzyme pCGR1 pCGR2


PstI
HindIII


SphI


EcoRI


BamHI

KpnI




HincII


XbaI

SmaI


SstI


Clal


4.6


8.4
4.1
3.2
1.1
6.0
1.4
1.0
2.7
2.5
2.4
0.8
5.5
2.9
6.2
0.6
0.6
0.6
0.4
3.3
2.8
1.4
0.7
0.2
7.8
0.6
2.2
2.2
2.0
1.0
1.0

2.0
1.8
N


6.8
3.2
3.1
0.5
4.2
2.5
0.1
3.2
2.6
1.0

4.0
2.8
6.4
0.4


3.0
1.6
1.0
0.6
0.6
6.3
0.5
4.3
1.6
0.9


4.6


1.8
0.4
N


N = No restriction site detected









49

polymorphisms for enzymes HindIII, EcoRI, SstI, HincII, and

SmaI were detected within the coding region for LSU rRNA in

the type 1 and type 2 rDNA forms. The NTS region of type 1

rDNA has 4 KpnI sites with three present at equal distance

of 0.6 kb, whereas within the NTS region of type 2 rDNA

there is only a single KpnI site. The additional 1.6 kb NTS

region fragment in type 1 rDNA contains two KpnI sites, each

0.6 kb apart and a 0.4 kb PstI/KpnI type 1 rDNA specific

fragment.


Diverse Cutinases and Cutinase Genes are Found in Type 1 and
Type 2 Strains of Colletotrichum gloeosporioides


Cutinase production by fungal mycelium can be induced

by cutin monomers (Lin and Kolattukudy 1978). Similarly, all

Tahiti lime and Sweet orange strains of C. gloeosporioides

excreted esterases under these inductive conditions when

cutin was used as the sole carbon source. Although, both

model cutin substrates, p-nitrophenol-butyrate and -

palmitate, were hydrolyzed, the ratio of these two

activities was remarkably different (Table 3.2). Cutinolytic

activity was identified for all isolates and was

consistently higher at pH 6.0 than at the alkaline pH of

9.5. Extracellular proteins were labelled with 3H-DFP and

used as active site probe for serine esterase. Two esterases

in the molecular weight range common to many known fungal

cutinases (17kd-32kd, Kolattukudy 1980; Tanabe et al 1988;










50

TABLE 3.2 Extracellular enzyme activities of Colletotrichum
gloeosporioides



Strain PNBase PNPase Cutinase PNB/PNP PNB/
mg/ml mg/ml kBq/h/mg Cutinase pH 6
pH 6 pH9.5 ratio




H-1 4731 255 18.7 2.0 9.4 18.6 254

IMB-3 6475 351 25.4 2.8 9.1 18.5 255

H-3 3678 302 23.9 4.5 5.3 12.2 154

H-4 8784 752 57.5 8.0 7.2 11.7 153

LP-1 5844 499 28.6 8.3 3.4 11.7 204

H-46 477 50 5.9 1.5 3.9 9.6 81

H-12 4692 554 40.0 1.3 30.8 8.5 117

H-25B 3627 488 27.5 5.5 5.0 7.4 132

Maran 3575 484 23.1 10.3 2.2 7.4 155

H-9 2193 329 18.8 4.2 4.4 6.7 117

H-48 4099 662 27.0 6.1 4.4 6.2 152

180269 4668 771 28.1 4.9 5.7 6.1 166

226802 4695 782 49.3 10.7 4.6 6.0 95

H-36 1863 369 23.7 5.8 4.1 5.0 78

Control 1456 332 12.8 7.9 1.6 4.4 144


PNB=p-nitrophenyl butyrate
PNP=p-nitrophenyl palmitate
Control consisted of all treatment without a fungal strain.










51
Trail and Kller 1990) were present for all strains (Figure

3.9). The molecular weight of these proteins differed among

strains and was correlated to C. gloeosporioides RFLP-types.

All strains of type 1 contained bands of 24 and 21 kd,

whereas all strains of type 2 contained 26 and 19 kd bands.

An additional esterase with a molecular weight of about 70

kd, which was not reported for the papaya isolate of C.

gloeosporioides (Dickman et al. 1982), was present

throughout the set of isolates. The high molecular weight

esterase was slightly larger for type 1 strains. The

relative contribution of this high molecular weight esterase

to the total esterase and cutinase activities remains

unknown. The enzyme might be similar to the 60 kd alkaline

cutinolytic esterase isolated from C. lagenarium (Bonnen and

Hammerschmidt 1989a) or the 54 kd non-specific esterase of

N. haematococca (Purdy and Kolattukudy 1975a).

A genomic clone containing the cutinase gene from a

papaya strain of C. gloeosporioides (Ettinger et al 1987)

was 32P-labelled and used to probe SphI-digested DNA of C.

gloeosporioides from citrus. This clone contained a 2.2 kb

SphI fragment which included the cutinase gene (189 bp exon-

52 bp intron-486 bp exon) and 5' and 3' flanking sequences

(Ettinger et al 1987). The probe hybridized to a 2.2 kb SphI

fragment only in type 2 strains (Figure 3.10). DNA from type

1 strains showed no detectable level of hybridization at the

high level of stringency (see materials and methods) used









52

for Southern hybridization. Although the cutinase gene

sequence shows no polymorphism among type 2 strains for the

restriction enzyme SphI, a restriction fragment length

polymorphism can be detected among type 2 strains for

HindIII (Figure 3.11). A 9.0 kb HindIII fragment hybridized

to the probe in all type 2 strains except 226802 which show

hybridization to a 8.0 kb fragment. Type 1 strains did not

show any detectable level of hybridization to the probe at

this level of stringency used for Southern hybridization

(see Appendix C). The long exposure (>1 month) of this blot

resulted in appearance of 4.8, 5.4, 6.6, 7.4, and 9.0 kb

HindIII hybridizing fragments of low level homology (Figure

3.12). Non-radioactive hybridization (Genius, Boehringer

Mannheim Corp. Indianapolis, IN) under low stringency

conditions shows weak hybridization of the probe to a 7.4 kb

HindIII fragment from type 1 strains (Figure 3.13).

Polymerase chain reaction amplification using

oligonucleotide primers flanking an intron sequence in the

cutinase gene resulted in a single 220 bp fragment when H-3

(type 1 strain) or H-12 (type 2 strain) DNA was used as a

template. However, strains LP-1 (type 1) and H-48 (type 2)

produced two amplified fragments, 220 and 260 bp (Figure

3.14). The 220 and 260 bp fragment amplified by PCR from

strain LP-1, when used as probes for Southern hybridization,

also hybridized to numerous HindIII fragments. Approximately

5-10 restriction fragments, ranging in size from 0.5 to >10









53

kb were identified (Figure 3.15 and 3.16 respectively).

Restriction fragments from all type 1 strains were almost

entirely identical. Hybridizing fragments from the type 2

strains showed dissimilar patterns. Hybridization of these

two probes to total DNA resulted in distinct DNA fingerprint

for type 1 strains.


Subgroups of Colletotrichum gloeosporioides have Distinct
RFLP Patterns Detected by Many Genetic Markers


Four clones of N. crassa genes were used as

heterologous probes to identify additional genetic loci in

HindIII-digested DNA from C. gloeosporioides strains. The

probe pSV50, containing the gene for B-tubulin, hybridized

to a 3.2 kb fragment in type 1 strains but a 5.0 kb fragment

in type 2 strains (Figure 3.17). The probe, pJR2, containing

the gene for glutamate dehydrogenase, hybridized to a 3.2 kb

fragment only in type 2 strains, but only diffuse

hybridization was observed in type 1 strains (Figure 3.18).

The probe, pNH60, containing the gene for histidinol

dehydrogenase, hybridized to both a 3.8 and a 4.3 kb

fragment in type 1 strains but hybridized to 3.3 and 4.8 kb

fragments in type 2 strains (Figure 3.19).


























'00
CO N CO kO
I I CO I U
Sa H a0

OR0DcJ


Figure 3.1


Polymorphic forms of ribosomal DNA unit in C.
gloeosporioides strains. Total DNA was digested
with PstI and Southern hybridized with 32P
labelled pMF2 (N. crassa rDNA unit repeat).
Lanes H-12, 226802, H-36, and OCO shows slower
migration of DNA than expected. Lane H-4 DNA is
degraded. Numbers at the left indicate the size
of restriction fragments in kilobases (kb).


I I


go rg
Iv In f
a4 a


24


r
























kb l




5-0
3-4


Figure 3.2


kb TTT


Various restriction fragments contain ribosomal
DNA in type 2 strains of C. gloeosporioides.
Total DNA was digested with PstI and Southern
hybridized with P labelled pMF2 (N. crassa rDNA
unit repeat). Numbers at the left indicate the
size of the major restriction fragments in
kilobases (kb).


- *












1 8 3 109 3 475 2 7 10 9


88U 5.8 8


L8U


104 572 7 9 97 74 2 1


II II


8U 5.8 8
SaU 6.8 8


Figure 3.3


I I I II I CR2
LSU


Restriction enzyme maps for cloned rDNA units
(pCGR1 from Colletotrichum gloeosporioides type 1
strain, H-25B, and pCGR2 from type 2 strain, H-
48). Regions hybridizing to large subunit rRNA
(LSU), small subunit rRNA (SSU) from N.crassa and
the PCR amplified 5.8S rRNA gene from C.
gloeosporioides are indicated by solid boxes.
Restriction enzyme sites are as follows. PstI
(1), HindIII (2), SphI (3), EcoRI (4), BamHI (5),
KpnI (6), HincII (7), XbaI (8), SmaI (9), and
SstI (10).


1 3108


l III I i iI I I II_ I I I I I pCR1


7 4 9 46 9 8 26 671



























8 4




H H H
8b 4 ) Qu ^QQ0 (*Q

25 ~~cc>;q~iq
84 A ~ k *


I -T


Figure 3.4


e4 H

'UJ.J0 i"
--
C- -
r^ BQ-:
& rr 3
\n''q~
H H1- fg
(0 *U 4J 00


I Q(Q0*


The rDNA of C. gloeosporioides strains H-25B
(first 10 lanes from left) and H-48 (next 10
lanes) digested with various restriction enzymes
and detected by Southern hybridization using 32P
labelled pMF2 (N. crassa rDNA unit repeat) as a
probe. The numbers at the left indicate the size
of major restriction fragments in kilobases (kb).


n




















I v S I I I & I I I f0 I I 0 N I U







4-3 .......
4-1
3 -1 1. .E:...


"" ... ,,E ..
::!!


Figure 3.5


Ribosomal DNA polymorphism in C. gloeosporioides
strains. Total DNA was digested with HindIII
and Southern hybridized with 32P labelled pMF2
(N. crassa rDNA unit repeat). H-4 DNA is
degraded. The numbers at the left indicate the
size of restriction fragments in kilobases (kb).

























kb CJ IN



7.o0


2-5


1*4 o-


~I r 'c
Iou I~c


Figure 3.6


Ribosomal DNA polymorphism in C. gloeosporioides
strains. Total DNA was digested with SphI and
Southern hybridized with P labelled pMF2
(N. crassa rDNA unit repeat). The numbers at the
left indicate the size of major restriction
fragments in kilobases (kb).


\oh
~0
CO
0..Io


409M 4


MODO



















c) me 00
9 0 02 k m o 0uco\
I -'e If to I I O C-41 4U I
kb C-4 g 0a- Mcu--





6-4



2-0


Figure 3.7


Ribosomal DNA polymorphism in C. gloeosporioides
strains. Total DNA was digested with SstI and
Southern hybridized with P labelled pMF2
(N. crassa rDNA unit repeat). The numbers at the
left indicate the size of major restriction
fragments in kilobases (kb).























I w* I
kcb


04

Iin


OR
$40
EU


%0
~0
00
cEO


8-4
























Figure 3.8 A 0.4 kb PstI/KpnI fragment (pCGR1N) from the
non-transcribed spacer region of cloned rDNA
unit from C. gloeosporioides isolate H-25B
hybridizes only to the 8.4 kilobase (kb) rDNA
form in type 1 strains. Total DNA was digested
with PstI and Southern hybridized with 2P
labelled pCGR1N.


























kd i~:I10


M R
f-4 C4 1.4
rI iaa


N00
0 %D'
OD C4O
T-1 C


21 .- ,W
19 d


Figure 3.9


Fluorography of 3HDFP-treated proteins after SDS-
polyacrylamide gel electrophoresis of
extracellular fluid from C. gloeosporioides
cultures grown on cutin as the sole carbon
source. Numbers at the left indicate the
molecular weight in kilodaltons (kd).





















(A N
rlr~~M E W 0QU
I$ 04 W0 0 A g
kb ( H N 0- U
r((u30


- MO


Figure 3.10 Cutinase gene in C. gloeosporioides type 2
strains. Total DNA was digested with SphI and
hybridized to a 32P labelled probe containing a
cloned cutinase gene. Numbers at the left
indicate the size of restriction fragment in
kilobases (kb).




















0)
I 1-4 00C CO C4

kb 0 I 1 1 Iu I 1 0


Mq
0
G %0
%cD MO
C4 I
M = 0


- "


Figure 3.11 Restriction fragments hybridize to the cloned
cutinase gene only within type 2 strains. Total
DNA was digested with HindIII and hybridized to
a 32P labelled probe containing a cloned
cutinase gene sequence from C. gloeosporioides.
H-4 DNA is degraded. Numbers at the left
indicate the size of restriction fragments in
kilobases (kb).

























kb I Z 1 I1f 1 OC4 IU


Figure 3.12


Hybridization of total DNA from the indicated
strains to a cloned cutinase gene sequence. The
DNA was digested with HindIII and hybridized to
a 32P labelled probe containing a cloned
cutinase gene sequence from C. gloeosporioides.
The blot was overexposed by placing it next to
X-ray film for more than 4 weeks. H-4 DNA is
partially degraded. Numbers at the left indicate
the size of major restriction fragments in
kilobases (kb).





















IM N


kb #0 1P 1 0N 4 1


)-0
f-4 jr


Figure 3.13


Presumptive cutinase genes in C. gloeosporioides
type 1 and type 2 strains. Total DNA was
digested with HindIII and Southern hybridized to
a digoxigenin labelled probe containing a cloned
cutinase gene sequence from C. gloeosporioides.
H-4 DNA is degraded. Numbers at the left
indicate the size of restriction fragments in
kilobases (kb).























1-4
0


bp U










260
220


Figure 3.14


PCR Amplified fragments using oligonucleotide
primers flanking the intron in the
C. gloeosporioides cutinase gene. Lanes contain
DNA from PCR reactions using for the template
the cloned cutinase gene (CG), or total DNA from
the strains indicated. The control reaction lane
contained no template DNA. Numbers at the left
indicate the size of DNA restriction fragments
in basepairs (bp).










68











a\ C





10















Figure 3.15 RFLP patterns in total DNA from the indicated C.
gloeosporioides strains detected by Southern
hybridization to a 220 bp fragment amplified by
1 0 4 k- WO N























PCR. The total DNAs were digested with HindIII
and the probe was labelled with digoxigenin
dUTP. H-4 DNA was partially degraded. Numbers at
the left indicate the size of major restriction
fragments in kilobases (kb).
fragments in kilobases (kb).





























kb
~eH~ l~cfT 0 vf


CONDo %
v 0 4O
1 ON1U
r-4 C4 C


.a- L


Figure 3.16 RFLP patterns in total DNA from the indicated C.
gloeosporioides strains detected by Southern
hybridization to a 260 bp fragment amplified by
PCR. The total DNAs were digested with HindIII
and the probe was labelled with digoxigenin
dUTP. H-4 DNA was partially degraded. Numbers at
the left indicate the size of major restriction
fragments in kilobases (kb).


. t




























N (A
%0 C(VO M t in cv %o r-i
OMWO\t (A $4 C*4 V-1 V I V M(r CO -4
kb UO MlNN e 04 = =1 v =1=1 =


II I


Figure 3.17


RFLP patterns in total DNA from the indicated C.
gloeosporioides strains detected by Southern
hybridization to the 32P labelled 8-tubulin gene
from N. crassa plasmidd pSV50). The total DNAs
were digested with HindIIl. Numbers at the left
indicate the size of restriction fragments in
kilobases (kb).


*M 0





















%0 C4 In gIo
ri ^- "' "~


Figure 3.18


RFLP patterns in total DNA from the indicated C.
gloeosporioides strains detected by Southern
hybridization to the 32P labelled glutamate
dehydrogenase gene from N. crassa plasmidd
pJR2). The total DNAs were digested with
HindIII. Numbers at the left indicate the size
of the major restriction fragment in kilobases
(kb).


I
-4C
kb I1
r-


en N
'Do
.ugVOD
l Oon
I rON


000


i~i~i~i~i~i~i~i~i~i~i~i~i~i~i~i~i~i~i~i~

















%VI 4 0 oI 0
kb i Z i i P i i i to i i co c4 i u
=H==A===Z r-4 N_.a= 0_
Gi n iaiti


33 1
!+!: o+ \,


Figure 3.19


RFLP patterns in total DNA from the indicated C.
gloeosporioides strains detected by Southern
hybridization to the 32P labelled histidinol
dehydrogenase gene from N. crassa plasmidd
pNH60). The total DNAs were digested with
HindIII. Numbers at the left indicate the size
of the major restriction fragments in kilobases
(kb).


*9









73

Discussion


In Chapter 2, two types of Colletotrichum

gloeosporioides strains were described based on

morphological and growth characteristics. However,

morphological features vary with culture conditions and

time. Morphological variability always has been the case

with C. gloeosporioides (Burger 1921; Arx 1957). Despite

confusion associated with morphological inconsistency, at

the molecular level we see distinct differences between type

1 and type 2 strains.

The rDNA unit proved to be a molecular marker which can

be used to detect type 1 strains of C. gloeosporioides

(Figure 3.8). Based on the Southern hybridization results,

morphologically stable type 1 strains contain a single form

of rDNA. The restriction map of the cloned rDNA unit (Figure

3.3, pCGR1) from type 1 strain, H-25B was identical when

compared with the restriction map of the genomic rDNA

(compare Figures 3.3 and 3.4). The three enzyme sites

HindIII, SphI, and SstI examined for all type 1 strains

agree with the map of the rDNA clone from strain H-25B

indicating that the form of rDNA present in type 1 strains

has been cloned and mapped (compare figures 3.3, 3.4, 3.5,

3.6 and 3.7). The size of the type 1 rDNA unit, 8.4 kb in C.

gloeosporioides, is within the range of observed rDNA unit

repeat sizes for filamentous fungi such as N. crassa, 9.23

kb (Free et al. 1979), Aspergillus nidulans, 7.8 kb (Borsuk









74

et al. 1982), Schizophyllum commune 9.2-9.6 kb (Specht et

al. 1984), and Thanatephorus praticola, 8.8 kb (Vilgalys and

Gonzalez 1990). The given order of rRNA genes, 5' SSU-5.8S-

LSU 3' (Figure 3.3) within the rDNA unit repeat is similar

in all the fungi examined (Free et al. 1979; Cihlar and

Sypherd 1980; Borsuk et al. 1982; Cassidy et al. 1984;

Buckner et al 1988; Garber et al. 1988; Vilgalys and

Gonzalez 1990). The rDNA unit repeats in C. gloeosporioides

may also code for large, 35-37S, precursor rRNAs which give

rise to SSU, 5.8S, and LSU rRNAs required for the building

of 80S ribosomes (Russell et al. 1976; Bell et al. 1977;

Planta et al. 1980). The search for a specific fragment of

rDNA which can detect only type 1 rDNA was successful. The

sub-clone pCGR1N containing a 0.4 kb PstI-KpnI from the NTS

region was strain specific (compare Figures 3.1 and 3.8).

Therefore, type 1 strains can be defined as having a single

homogeneous form of rDNA as detected by a 8.4 kb Pst-1

fragment hybridizing to pMF2 and pCGR1N.

Morphologically variable type 2 strains (see Chapter 2)

were also diverse at DNA levels having different forms of

rDNA (Figures 3.2). The PstI-SphI-and SstI digested total

DNA blots, show RFLPs for rDNA within type 2 strains

(Figures 3.1, 3.2, 3.6, and 3.7). Only the HindIII-digested

blot shows a similar pattern of rDNA bands for type 2

strains (Figure 3.5). These results suggest that rDNA among

type two strains is heterogeneous. In addition, the specific









75

detection of only type 1 rDNA by pCGR1N suggest that

although some type 2 strains contain a rDNA form similar in

size to type 1 rDNA, the sequences may be different at least

at the NTS region. Several subclones from the NTS region of

type 2 rDNA hybridized to both type 1 and type 2 rDNA.

Therefore, a type 2 strain specific rDNA marker was not

found. Restriction site polymorphisms and length

heterogeneity in the rDNA unit repeat have been commonly

observed in many fungi. These polymorphisms were found

within the NTS region (Verbeet et al. 1983; Cassidy et al.

1984; Russell et al. 1984; Van Heerikhuizen et al 1985;

Chambers et al. 1986; Lachance 1989; Rogers et al. 1989),

ITS region (Chambers et al. 1986; Buckner et al. 1988) or

within coding regions (Chambers et al. 1986; Martin 1990).

In the fruit fly Drosophila melanogastor the presence of

introns has given rise to polymorphic forms of rDNA (Glover

and Hogness 1977). In fungi evidence for the presence of

introns in the rDNA coding regions is inconclusive (Buckner

et al. 1988).

Type 1 and type 2 C. gloeosporioides strains,

distinguished by rDNA polymorphisms were different both in

cutinase isozymes and molecular organization of relevant DNA

sequences. The slightly acidic pH optimum (pH 6.0) of

cutinolytic activity (Table 3.2) of these strains is a

reflection of their pathogenicity to aerial plant parts

(flowers or leaves) as observed for other aerial plant









76

pathogens such as V. inaequalis (KBller and Parker 1989),

Botrytis cinera (Salinas et al. 1986) and Cochliobolus

heterostrophus (Trail and K6ller 1990). This pH preference

is also congruent with the hypothesis that pathogens with

this type of cutinase are specialized for infecting aerial

plant surfaces rather than stem bases and roots (Trail and

Kal1er 1990).

All cutinases are serine esterases, and therefore they

can be detected by 3H-DFP which phosphorylates and inhibits

specific serine esterases (Kller and Kolattukudy 1982;

Kolattukudy 1985; K6ller and Parker 1989). The two 3H-DFP

labelled bands (Figure 3.9) which are within the range of

molecular weights of fungal cutinases (Kolattukudy 1980;

Tanabe et al. 1988; Trail and K11er 1990) were not

previously seen in Colletotrichum species including papaya

isolate of C. gloeosporioides (Dickman et al. 1982;

Kolattukudy 1985). Although molecular weight and culture

conditions prior to electrophoresis suggest that cutinase

enzymes are present, DFP binding is serine esterase but not

cutinase specific (K6ller and Kolattukudy 1982; Kolattukudy

1985; K1ller 1991). These bands may not represent two

different primary gene products but may be the result of

post-translational modification (K6ller 1991; Soliday et al.

1984). Some cutinase enzymes may undergo a proteolytic nick

and appear as two fragments after reduction of a disulfide

bridge and electrophoresis under denaturing conditions









77

(KSller 1991; Lin and Kolattukudy 1980; Purdy and

Kolattukudy 1975b; Soliday and Kolattukudy 1976). Conclusive

demonstration that these two bands are actually two

cutinases awaits further experiments such as purification

and characterization of the catalytic activity. Other fungal

pathogens, N. haematococca (Purdy and Kolattukudy 1975a,

1975b), A. brassicola, and R.solani (Trail and K11er 1990)

are known to produce at least two distinct cutinases as

detected by distinct isozyme bands and enzyme catalysis.

For type 2 strains, only one DNA restriction fragment

hybridizes to the cutinase gene probe but two cutinase

isozymes may be present. One possibility is that these may

be two forms of cutinases encoded by distinctly different

genes in the same organism. Only one form of cutinase has

been described biochemically from the papaya strain of C.

gloeosporioides (Dickman et al. 1982). The poor

hybridization of type 1 strains to the cutinase clone

despite the fact that they have abundant cutinase activity

suggests considerable evolutionary diversification of

cutinase gene sequences. Since the cDNA clone for cutinase

from C. capsici hybridizes readily with the total DNA from

C. graminicola and the C. gloeosporioides strain from

papaya, these species may be more related to type 2 than

type 1.


Another line of evidence for distinct genetic forms of









78

C. gloeosporioides from citrus comes from PCR amplification

of a cutinase gene sequence. The published DNA sequences for

cutinase genes from C. gloeosporioides and C. capsici

(Ettinger et al. 1987) show both conserved and diversified

regions of the gene. Two regions of identical sequence flank

a short stretch of DNA which includes the 52 bp intron of

the C. gloeosporioides cutinase and the 57 bp intron of the

C. capsici cutinase. The primer sequences are conserved in

cutinase genes of C. gloeosporioides, and C. capsici and

respectively should amplify a 220 bp or 222 bp fragment when

used as primers for PCR (Ettinger et al. 1987). Indeed, one

or two amplified fragments of about 220-260 bp were obtained

from the DNA of the four strains tested. When used as a

probe for Southern hybridization these fragments were

anticipated to hybridize to conserved elements of the

cutinase gene and divergent sequences expected within the

intron. However, these sequences hybridize to multiple

restriction fragments producing DNA finger prints that

correspond to RFLP types.

Hybridization to these probes must not be specific for

cutinase sequences. The exact nucleotide sequences of the

two amplified fragments were not determined. Therefore, it

is necessary to determine the nucleotide sequence of the 220

and 260 bp amplified fragments before reaching any

conclusions. However, a computer search of the GenBank

database (Release 70.0, December 15, 1991) indicated no









79

significant sequence similarity between the 220 bp targeted

sequence of the cutinase gene from C. gloeosporioides

(Ettinger et al 1987) and genes other than for C. capsici

and C. gloeosporioides cutinase. Hybridization may be to

sequences common to other introns such as putative splice

junction sequences identified previously (Ettinger et al.

1987). Hybridization to the repetitive sequences may be

detected using the 220 bp probe and not the entire cutinase

clone because the intron sequence represents <1% of the

clone but 25% of the PCR fragment.

Cutinase is not the only extracellular enzyme produced

by fungi in the process of plant infection. There are many

others such as cellulolytic and pectinolytic enzymes

produced by fungi in culture and in diseased leaves (Oke

1989; Prusky et al. 1989). The type 1 and type 2 strains

also produce different isozymes of pectinesterase that

correlate to type 1 and type 2 (personal communication

Gantotti and Davis, Homestead, FL).

DNA polymorphisms detected by hybridization to 3 of 4

"housekeeping" genes from N. crassa also correspond directly

to type 1 and type 2 strains. To detect DNA polymorphisms

correlated to type 1 and type 2 strains by Southern

hybridization using these molecular markers it was not

necessary to search for specific restriction enzymes or to

test multiple loci. The only enzyme used to digest total

DNA, HindIII was sufficient to provide RFLPs capable of









80

separating type 1 from type 2 strains. However, exceptions

to the strict correlation between polymorphism and strain

type was seen for hybridization to pNC2. In this case

variation was seen among type 1 strains. On the whole all

the molecular markers suggest that the two types are

different at the molecular level indicating they are indeed

genetically distinct populations of C. gloeosporioides.















CHAPTER 4
VARIABILITY OF MOLECULAR KARYOTYPES AND CHROMOSOMAL DNAS IN
COLLETOTRICHUM GLOEOSPORIOIDES


Introduction


Pulsed Field Gel Electrophoresis


Macromolecules such as nucleic acids and proteins can

be separated on the basis of size, charge or conformation by

gel-electrophoresis. Schwartz et al. (1982) made use of the

relaxation properties (Klotz and Zimm 1972) of large DNA

molecules for their separation in agarose gels by using two

alternating electric fields known as pulsed field gel

electrophoresis (PFGE). A major advance of the pulsed field

gel electrophoresis was achieved by Chu et al. (1986). They

applied the principles of electrostatics to calculate the

voltages needed to generate homogeneous electric fields

using multiple electrodes arranged around a closed contour.

In this system, contour clamped homogeneous electric field

(CHEF) gel electrophoresis, twenty four electrodes were

arranged in a hexagonal contour which offers reorientation

angles of 60 or 120.











Molecular Karvotypes of Fungi


Fungal chromosomes are too small to be observed readily

by conventional cytological methods using light microscopy.

However, electrophoretic karyotyping and molecular analysis

of chromosome-size DNA have become the new methods for

studying genomic structure of various organisms including

filamentous fungi. Many studies have conclusively

demonstrated that DNAs resolved by PFGE corresponds to

chromosomes (Carle and Olson 1985; Orbach et al. 1988; Brody

and Carbon 1989; Kayser and Wostemeyer 1991). However, the

number of bands need not be equal to the number of

chromosomes (Horton and Raper 1991). Electrophoretic

analysis of chromosomes provide a very convenient and rapid

way of assigning genes to chromosomes and for monitoring

entire genomes for any chromosomal rearrangements.

Pulsed field gel electrophoresis has been used to

separate chromosome-size DNA and analyze molecular

karyotypes of many fungi such as S. cerevisiae (Schwartz and

Cantor 1984; Chu et al. 1986), Candida albicans (Snell and

Wilkins 1986), Schizosaccharomyces pombe (Smith et al 1987;

Vollrath and Davis 1987), Neurospora crassa (Orbach et al.

1988), Ustilago maydis (Kinscherf and Leong 1988), Candida

stellatoidea (Kwon-Chung 1988, 1989: Wickes et al. 1991),

Aspergillus nidulans (Brody and Carbon 1989), C.

gloeosporioides (Masel et al. 1990), Ustilago hordei,

Tilletia caries, T. controversy (McCluskey et al. 1990),









83

Schizophyllum commune (Horton and Raper 1991), Absidia

glauca (Kayser and Wostemeyer 1991), Septoria tritici

(McDonald and Martinez 1991), Nectria haematococca (Miao et

al. 1991), Acremonium species (Smith et al. 1991; Walz and

Kuck 1991), Leptosphaeria maculans (Taylor et al. 1991), and

Fusarium oxysporum (Momol and Kistler 1992).

Pulsed field gel electrophoresis of chromosomal DNA

combined with Southern analysis using linkage group-specific

probes were key methods in defining molecular karyotypes of

N. crassa (Orbach et al. 1988) and Aspergillus nidulans

(Brody and Carbon 1989). Molecular karyotyping of N. crassa

by Orbach et al. (1988) confirmed the seven linkage groups

previously defined by genetic analysis (Perkins et al.

1982). The genome size of A. nidulans was estimated by Brody

and Carbon (1989) to be approximately 31 Mb with six

chromosome-sized DNA bands. Kayser and Wostemeyer (1991)

reported differences in electrophoretic karyotypes for

mating types of the Zygomucete Absidia glauca.

Molecular karyotypes of many plant pathogenic fungi

examined to date have been variable. Kinscherf and Leong

(1988) analyzed the molecular karyotype of U. maydis and

demonstrated that considerable chromosomal length

heterogeneity exists in this fungus. DNA hybridization

analysis suggested that stable large scale inter-chromosomal

exchange has given rise to novel chromosomes in one of the

strains. Taylor et al. (1991) using TAFE demonstrated that









84

the karyotypes of highly virulent and weakly virulant

strains of Leptosphaeria maculans (black leg of crucifers)

were polymorphic in both chromosome number and size. Highly

variable karyotypes for N. haematococca with unique

karyotypes for each strain were reported by Miao et al.

(1991). Deletions of large amounts of DNA from chromosomes

have given rise to karyotype variation as well as a

decreased frequency of the pisatin demethylase gene in N.

haematococca. Masel et al. (1990) suggested that chromosomal

rearrangements may play a role in generating variability of

karyotype of C. gloeosporioides. Distinct electrophoretic

karyotypes were reported for strains from two types of C.

gloeosporioides causing different anthracnose diseases in

Stylosanthus species in Australia. The strains showed

extensive chromosomal polymorphisms for both length and

number in the mini-chromosomes (molecules less than 2

million base pairs (Mb) in length) within each type.

The present study was undertaken to investigate the

variation of molecular karyotypes and chromosomal DNAs in

two types of C. gloeosporioides (see Chapter 2 and 3)

causing post bloom fruit drop of Tahiti lime and Sweet

orange.









85

Materials and Methods


Strains of Colletotrichum qloeosporioides


Strains used were obtained from several different areas

of Florida, Mexico, and from the Commonwealth Institute of

Mycology, England. They were isolated from diseased lime or

orange tissues. Details of host, place of collection and

date are tabulated in the appendix A.


Preparation of Protoplast Plugs


The strains of C. gloeosporioides were grown for 7 days

in 50 ml of 20% (w/v) V-8 juice (Campbell Co., Camden, NJ)

in at 250 rpm in Erlinmayer flasks on a Lab-Line orbit

shaker (Lab-Line Instruments Inc., Melrose Park, IL) at

ambient temperature (21-23 oC), and conidia were collected

by centrifugation at 7000 x g for 5 min. About 109 spores

per ml were resuspended in 50 ml potato dextrose broth (PDB)

and incubated at room temperature (23 to 25 OC) for 16-24 h

at 200 rpm. When over 90% of spores were germinated, the

germlings were pelletted by centrifuging at 7000 x g for 5

min. Protoplasts were made by adding germlings to a 10 ml

solution containing NovoZym 234 (Novo Industries, Bagsvaerd,

Denmark) a complex mixture of wall-degrading enzymes. The

NovoZym solution was prepared by mixing 1.5 ml of 1 M

sorbitol, 50 mM sodium citrate containing 0.2 g of NovoZym

234 with 8.5 ml of 1.4 M MgSo4, and 50 mM Sodium Citrate pH










86

5.8. Germlings were incubated in this solution with gentle

rocking on a Bellco rocker (Bellco Biotechnology, Vineland

NJ) at 4 rpm for 3 to 6 h at ambient temperature for 3 to 6

hours until most cells were protoplasts. The protoplasts

were filtered through 4 layers of cheese cloth in order to

remove cell debris and undigested germlings. The filtrate

was centrifuged at 3000 rpm for 25 min at room temperature.

Protoplasts were removed from the top and washed three times

with 1 M sorbitol-50 mM EDTA pH 8.0. Protoplast inserts for

PFGE were made as described by method 1 of Orbach et al.

(1988).


Electrophoresis and Southern Analysis


A commercially available apparatus (BioRad CHEF DRII,

Richmond, CA) using different pulse time combinations was

employed in order to separate chromosome-sized DNAs.

Electrophoresis was done with 0.6% FastLane agarose (FMC

BioProducts, Rockland, ME) gels in 0.25X Tris Borate EDTA

(TBE) buffer (Sambrook et al. 1989) at 4 OC with rapid

circulation of the buffer. The gels were run at 40 volts for

6-10 days. Pulse times were "ramped" for various times

ranging from 10 to 180 min. For the separation of smaller

chromosome sized DNA 1% SeaKem Agarose (FMC BioProducts,

Rockland, ME) in 0.5X TBE buffer was used. These gels were

run at 200 V for 24 h with pulse times of 30-60 s or 50-90s.









87

Southern hybridization (Appendix C) experiments with

32P labelled ribosomal DNA (pMF2), B-tubulin gene (pSV50)

and cutinase gene (see chapter 3) were carried out to assign

these sequences to chromosome-size DNAs separated in CHEF-

gels.

Results


Chromosome-size DNAs (henceforth called chromosomes)

from type 1 and type 2 strains of C. gloeosporioides were

separated by PFGE using Saccharomyces cerevisiae and

Schizosaccharomyces pombe chromosome size DNA as standards

(BioRad Laboratories, Richmond, CA). The sizes in Table 4.1

represent the average size calculated for each chromosome

size-DNA band from independent CHEF-gels. Calculated sizes

for individual chromosome size DNA bands and relevant

figures are compiled in the Appendix D. Type 1 strains have

chromosomes distinguishable from type 2 strains (Table 4.1).

The chromosomes of C. gloeosporioides isolated from

Stylosanthes have been classified by Masel et al. (1990)

into larger, similar-sized chromosomes (>2 Mb) and smaller

variable-sized elements called "minichromosomes" (<2 Mb). A

similar arrangement was noted for strains isolated from

Tahiti lime and Sweet orange. Type 1 strains possess 5

chromosomes (Figures 4.1 and 4.2) and an additional 1 or 2

minichromosomes (Figure 4.3 and 4.5). Type 2 strains possess

3 chromosomes (Figures 4.1 and 4.2) in addition to 2 to 4










88

Table 4.1 Estimated megabase sizes for chromosome-size DNA
from Colletotrichum gloeosporioides type 1 and
type 2 strains


Strain Estimated size" (Mb)
I II III IV V VI VII


Type-1 strains

H-1

H-3

H-9

H-25B

H-36

LP-1

Maran

IMB-3

OCO

Type-2 strains

H-4

H-12

H-46

H-48

180269

226802


7.6

7.6

7.6

7.6

7.6

7.6

7.6

7.6

7.6


7.8

7.8

7.8

7.8

7.8

7.8


7.0

7.0

7.0

7.0

7.0

7.0

7.0

7.0

7.0


4.7

4.7

4.7

4.7

4.7

4.7


4.7

4.7

4.7

4.7

4.7

4.7

4.7

4.7

4.7


3.7

3.7

3.7

3.7

3.7

3.7


3.7

3.7

3.7

3.7

3.7

3.7

3.7

3.7

3.7


0.42

0.46

0.52

0.46

0.43

0.44


3.3

3.3

3.3

3.3

3.3

3.3

2.8

3.3

2.8


0.38

0.38

0.47

0.43

0.41

0.42


1.1

1.1

1.1

1.1

0.77

1.6



1.1


0.42

0.40

0.39

0.39


0.63

0.63

0.63

0.63

0.63

0.63

0.63

0.63

0.65


0.27





0.37


Schizosaccharomyces pombe and Saccharomyces cerevisiae size
standard were used for the calculation of Mb sizes. Megabase
sizes greater than 5.6 were estimated by extending the
calibration curve and therefore may be considered
approximate sizes.
- = not detected in any of the gels.










89

minichromosomes (Figures 4.4 and 4.5) depending on the

strain. Within each type, strains show variations in

chromosome number and size. However type 2 strains show more

total variation in chromosome and minichromosome size

(Figures 4.4 and 4.5).

A Southern blot separating larger chromosome-size DNAs

was hybridized with a 32P labelled ribosomal DNA probe. The

rDNA is associated with the 4.7 Mb chromosome in type 1

strains and with the 7.8 Mb chromosome in type 2 strains

(Figure 4.6). The cutinase gene can be assigned to the 4.7

Mb chromosome-size DNA only in type-2 strains (Figure 4.7).

Homologous regions were not detected in chromosomes of type

1 strains. The B-tubulin gene hybridizes to both the 7.0 and

7.6 Mb chromosome doublet in type-1 strains and the 7.8 Mb

chromosome in type-2 strains (results are not shown due to

weak signals on X-ray film).
























0% N
%00
1O 00 N 1 1 13
MbUO4 0 1 0" .1
Mb'W T- CMTT


5-7
4.6
3-5
2"2










Figure 4.1 Chromosome-sized DNAs from C. gloeosporioides
compared to those for yeast (S. cerevisiae) and
fission yeast (S. pombe). DNAs were separated on
agarose gels using CHEF electrophoresis. Gels
were stained with ethidium bromide (0.5 gg/ml)
and fluorescence photographed with UV
transillumination. DNAs from the indicated
strains were run in 0.25X TBE, 0.6% agarose for
168 h at 40 V. Pulse switching times were ramped
from 40-70 min. Numbers at the left indicate size
standards in megabases (Mb).
























Ica 0 %D
cm Ln o % 0

Mba 04 1 1 0 $

5.7
4 6~ltP


Figure 4.2


Chromosome-sized DNAs from C. gloeosporioides
compared to those for fission yeast (S. pombe).
DNAs were separated on agarose gels using CHEF
electrophoresis. Gels were stained with ethidium
bromide (0.5 Ag/ml) and fluorescence photographed
with UV transillumination. DNAs from the
indicated strains were run in 0.25X TBE, 0.6%
agarose for 168 h at 40 V. Pulse switching times
were ramped from 20-60 min. Numbers at the left
indicate size standards in megabases (Mb).

















n3 0 e
I -I L 0 Uo
Sn oZ ,- I ( 4 0
li : X


Figure 4.3


Chromosome-sized DNAs from C. gloeosporioides
compared to those for yeast (S. cerevisiae). DNAs
were separated on agarose gels using CHEF
electrophoresis. Gels were stained with ethidium
bromide (0.5 jg/ml) and fluorescence photographed
with UV transillumination. DNAs from the
indicated strains were run in 0.5X TBE, 1.0%
agarose for 24 h at 200 V. Pulse switching times
were ramped from 60-90 s. Numbers at the left
indicate size standards in kilobases (kb).


kb 0






1125





630



245