. .
2
«uhww...
. . hawk—rubrh:
v 1 . .
5% $5
_ n... a.
. I.” {(42.5 a
un- h.. I. to

.. r
.FWHL... .
.

3.3:!

vb. .1.

.31...sz
2:. . 1......
’ I
do!!!
... .2135:
in: 5,?

1 . . .
v... in“... 7.1.5.135?
.,. «rgnapfrt:

y RB...“

.fxlk.

.fi
we
hang

L
a

can.

7

11"..

‘ ......irhxn.u!
2 Se; .r .
. , 3.2:. I. .n

 

WNW!!!“IHHIIWWIHIUlllHllHlHIHWNIHUflUl

301771 8242

 

LIBRARY

Michigan State
University

 

 

 

This is to certify that the

dissertation entitled

M04 EC 0444 Gen/tries on VIRULIWS'
15v doc/IL ice 0:. as 60.98 ova/M

presented by

fl MAS 7451.5 M M'KOLSKA V4

has been accepted towards fulfillment
of the requirements for

’04. a 3 degree in Gut-W; 7/.“

%0L (LO/x, MLJG :i7\

(/ “Wm

Date #91 AZ, /9%’

MS U i: an Affirmative Action/Equal Opportunity Institution 0-12771

 

PLACE IN RETURN BOX to remove this checkout from your record.
TO AVOID FINES return on or before date due.
MAY BE RECALLED with earlier due date if requested.

 

DATE DUE

DATE DUE

DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

ma animus-p14

MOLECULAR GENETICS OF VIRULENCE
IN C OCHLIOBOL US CARBON UM

By

Anastasia N. Nikolskaya

A DISSERTATION

Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of

DOCTOR OF PHILOSOPHY
Genetics Program

1999

ABSTRACT

MOLECULAR GENETICS OF VIRULENCE IN
C OCHLIOBOL US CA RBON UM

By

Anastasia N. Nikolskaya

Tox2+ isolates of the filamentous fungus C ochliobolus carbonum produce a cyclic
tetrapeptide, HC-toxin, that is necessary for virulence on certain genotypes of maize.
HC-toxin production is controlled by T 0X2 genes that are located on the supemumerary
T 0X2 chromosome. In this work, the roles of horizontal gene transfer and TOX2
chromosome instability in the evolution of the C. carbonum virulence were evaluated.

Four other filamentous fungi unrelated to C. carbonum produce cyclic peptides
closely related to HC-toxin, raising the possibility that the genes involved in the synthesis
of these compounds have moved between these fungi by horizontal gene transfer. In
order to test this hypothesis, genomic DNA sequences encoding non-ribosomal peptide
synthetases were amplified from different fungi by PCR. In addition, the genomic locus
encoding a peptide synthetase from Diheterospora chlamydosporia was cloned, and a
partial DNA sequence was obtained. The deduced amino acid sequences of the non-
ribosomal peptide synthetases obtained from these fungi were found to be closely related
to the C. carbonum HC-toxin synthetase (HTS), but the percent amino acid identity was
lower than expected if these genes have been recently transferred horizontally.

A study of alterations in the TOXZ chromosome of C. carbonum revealed

exceptional meiotic instability of this chromosome. Of 200 progeny analyzed in crosses

between Tox2+ and Tox2- isolates and between isolates in which the T 0X2 genes were
on chromosomes of different sizes, eight (4%) had lost at least one copy of one of the
T 0X2 genes. All of them still had at least one functional copy of each of the TOX2
genes. The deletion strains were characterized with respect to virulence, HC-toxin
production, TOX2 gene expression, and size of the TOXZ chromosome. Most deletions
could be explained by simple chromosome breaks resulting in the loss of major
contiguous portions (0.8 Mb to 1.4 Mb) of the 3.5-Mb TOXZ chromosome. Most strains
were still completely virulent, but two strains displayed a novel phenotype of reduced
virulence (RV), characterized by lesions that expanded at a reduced rate and an inability
to colonize plants systemically. Although the RV strains produced no detectable HC-
toxin in culture, the RV phenotype was dependent on the presence of a functional copy of
HT SI . We propose that the RV strains still make a low level of HC-toxin, at least in
planta, and that the RV strains are missing unknown gene(s) that play a role in, but are
not absolutely required for, HC-toxin production.

In addition, genomic and cDNA copies of EXGI, a gene encoding a novel exo-B-
1,3-g1ucanase from C. carbonum, were isolated and sequenced. The deduced amino acid
sequence of EXGI contains two imperfect copies of a 23-amino acid motif that is found
in several other proteins that interact with polysaccharides, including two exo-B-1,3-
glucanases, plant and bacterial polygalacturonases, PZA phage neck appendage protein,
K1 phage endoneuraminidase, and bacterial mannuronan epimerase.

In a study of the function of the amino-acid activating domains of HTS, domain A

was expressed separately in a Tox2_ strain and found to activate L-Proline.

But you yourself should not be able
To tell defeats from victories of yours.

- Boris Pastemack

ACKNOWLEDGMENTS

I thank Dr. Jonathan Walton for the opportunity to work in his laboratory, to learn
so much in the process, and to have the benefit of his guidance and advice. His patience
with me was truly unlimited, and his enthusiasm for various projects and for science in
general makes working in his lab a really enjoyable experience.

I consider myself lucky to have been a member of the Walton Lab Team, and I
would like to thank all the members past and present for their help and for creating a
friendly and fim atmosphere in the lab. I would especially like to thank John Scott-Craig
and John Pitkin for their unceasing advice and discussion of every ongoing project. It is
from them that I have learned most of the techniques used in this work, and whenever I
needed to ask a stupid question or complain about a failed experiment, I could count on
them. I do not think this work would have been possible without their help.

I thank my committee members, Dr. Frans de Bruijn, Dr. Richard Lenski, and Dr.
Barbara Sears for their guidance and important discussions of the projects I worked on. I
also thank the faculty and staff of the PRL for making this such a great place to work, and
the Genetics Program for the opportunity to be a student in this Program.

Many thanks go to my husband Alex (Sasha) and to the rest of my family for their
constant support through thick and thin. I would also like to thank all my new friends in

Michigan and my old friends in Russia, because their support has kept me going.

TABLE OF CONTENTS

LIST OF TABLES ........................................................................................................... viii
LIST OF FIGURES ............................................................................................................ ix
INTRODUCTION ................................................................................................................ 1
CHAPTER 1

IDENTIFICATION OF PEPTIDE SYNTHETASE-ENCODING GENES FROM
F ILAMENTOUS F UNGI PRODUCING HOST-SELECTIVE PHYTOTOXIN S OR

THEIR AN ALOGS ...................................................................... . ...................................... 1 1
Abstract ........................................................................................................................... 11
Introduction ..................................................................................................................... 12
Results and Discussion ................................................................................................... 16

PCR Strategy ............................................................................................................... 16
Analysis of Putative Non-Ribosomal Peptide Synthetase Gene Fragments ................ 18
Cloning and Mapping of the Non-Ribosomal Peptide Synthetase Gene Locus from
D. chlamydosporia ...................................................................................................... 24
Attempts to Disrupt the Peptide Synthetase Gene of D. Chlamydosporia .................. 26
Materials and Methods ................................................................................................... 28
Fungal Culture Growth and Maintenance ................................................................... 28
DNA Manipulations .................................................................................................... 28
Toxin Extraction and Epoxide Detection .................................................................... 29

CHAPTER 2

REDUCED VIRULENCE CAUSED BY MEIOTIC INSTABILITY OF THE T 0X2

CHROMOSOME OF C OCHLIOBOL US CARBON UM ................................................... 30
Abstract ........................................................................................................................... 30
Introduction ..................................................................................................................... 3 1
Results and Discussion ................................................................................................... 35

Disease Phenotype of 243-7 ........................................................................................ 35
Saprophytic Growth, Maize Leaf Penetration and Toxin Phenotype of Strain

243-7 ............................................................................................................................ 43
Disruption of HT S1 in Isolate 243-7 Eliminates the RV Phenotype ........................... 43
Isolation and Initial Characterization of Additional Strains with Chromosomal
Aberrations .................................................................................................................. 48
Co-Segregation of the RV Phenotype with the Chromosomal Aberration ................ 57
Discussion of T 0X2 Chromosome Segregation and Instability .. ................................ 59
Pathogenicity Phenotype and T 0X2 Gene Expression in Strains with Chromosomal
Aberrations .................................................................................................................. 69
Possible Mechanisms That May Cause the RV Phenotype ......................................... 71

vi

Cosmids Containing the Chromosomal Region Adjacent to HT S1 / TOXA-l Do Not

Complement the RV Phenotype .................................................................................. 78
Materials and Methods ................................................................................................... 8O
Fungal Culture Growth and Maintenance ................................................................... 8O
Pathogenicity Tests and Leaf Penetration Tests .......................................................... 81
Fungal Transformations .............................................................................................. 82
Nucleic Acid Manipulations ........................................................................................ 82
Pulsed-Field Gel Electophoresis ................................................................................. 82
Protein Manipulations and HTS Activity Assays ........................................................ 83
CHAPTER 3
EXGI, A NOVEL B-l-3-EXOGLUCANASE FROM COCHLIOBOL US
CARBON UM ...................................................................................................................... 85
Abstract ........................................................................................................................... 85
Introduction ..................................................................................................................... 86
Results and Discussion ................................................................................................... 87
Analysis of Genomic and cDNA Clones of EXGI ...................................................... 87
EXGI p is a Novel B-l,3-glucanase ............................................................................. 9O
EXGlp Has Two Copies of a Motif Shared with Other Proteins That Interact with
Polysaccharides ........................................................................................................... 94
Effect of Carbon Source on Expression of EXGI . ...................................................... 97
Materials and Methods ................................................................................................... 97
Fungal Culture Growth and Maintenance ................................................................... 97
Nucleic Acid Manipulations ........................................................................................ 99
DNA Sequencing ......................................................................................................... 99
APPENDIX A
AMINO ACID SPECIFICITY OF THE FUNCTIONAL DOMAIN A OF THE
C OCHLIOBOL US CARBON UM HC-TOXIN SYNTHETASE ...................................... 102
Introduction ................................................................................................................... 1 02
Results and discussion .................................................................................................. 103
Domain A of HTS Activates L-Proline ..................................................................... 103
Sequential Disruption of HTS] Functional Domains ................................................ 113
APPENDIX B
CC115, A TRANSPOSASE-LIKE SEQUENCE FROM C OCHLIOBOLUS
CARBON UM .................................................................................................................... 116
BIBLIOGRAPHY ............................................................................................................ 121

vii

Table 1.

Table 2.

Table 3.

Table 4.

Table 5.

LIST OF TABLES

Fungal species used in this study and peptide toxins that they produce. . . ..14
Summary of data on C. carbonum strains used in this study ...................... 52
HC-toxin synthetase (HTS) activity in isolates of C. carbonum .................. 74
ATP/PP, exchange activity in C. carbonum bearing pGPD18 expression

vector ...................................................................................... 1 1 1

Analysis of the transformants for the step-wise disruption of HT S1 ............ 114

viii

Figure 1.

Figure 2.

Figure 3.

Figure 4.

Figure 5.

Figure 6.

Figure 7.

Figure 8.

Figure 9.

Figure 10.
Figure 11.
Figure 12.
Figure 13.
Figure 14.
Figure 15.

Figure 16.

LIST OF FIGURES

The family of cyclic tetrapeptides from filamentous fungi: Aeo-containing
peptides and apicidin (after Walton et al., 1997; Darkin-Rattray et al., 1996). 5

TLC analysis of epoxide-containing toxin production by representatives of
the fungal species used in this study... ... ....................................................... 17

PCR strategy to amplify peptide synthetase sequences from several fungal
species ............................................................................................................. 19

Analysis of the products of the second round of PCR (reamplification) of
Cy]. macrosporum DNA by gel electrophoresis ............................................. 20

Alignment of the predicted amino acid sequences of the PCR products
containing conserved peptide synthetase sequences with the corresponding
sequences from the functional domains of HTS ............................................. 22

Restriction map of D. chlamydosporia locus that contains the putative
chlamydocin synthetase gene ............................................................................ 25

Alignment of amino acid sequences of the four conserved domains of C.
carbonum HTS and two conserved domains from D. chlamydosporia .......... 27

Physical maps of the TOXZ chromosome in SB1 11 and 243-7 (after Ahn

and Walton, 1996) ........................................................................................... 34
Strain 243-7 exhibits a new pathogenicity phenotype on maize .................... 36
Disease phenotype after inoculation of very young seedlings ........................ 39
Disease phenotype after inoculation of seeds ................................................. 41
Disease phenotype after soil inoculation ........................................................ 42
Growth of strain 243-7 compared to SB1 11 ................................................... 44
Analysis of HT S] disruption mutants ............................................................. 46
Disruption of HT S1 in 243-7 makes this strain avirulent (Tox2_) ................. 47
Crosses to generate deletion strains ................................................................ 49

Figure 17.
Figure 18.
Figure 19.

Figure 20.

Figure 21.

Figure 22.

Figure 23.

Figure 24.

Figure 25.
Figure 26.
Figure 27.

Figure 28.

Figure 29.

Figure 30.

Figure 31.

Figure 32.

Figure 33.

Figure 34.

Southern blot analysis of the strains used in this study .................................. 50

Southern blot analysis of T OXE RFLP in selected deletion strains ................ 53
Southern blot analysis of strain 512-98D ....................................................... 55
Physical mapping of TOXZ chromosomes in representatives of wild type

and deletion strains ......................................................................................... 56
Physical maps of the TOX2 chromosome ....................................................... 58

Proposed translocation involving two naturally occurring variants of T 0X2
chromosome .................................................................................................... 61

Northern blot analysis of T OXE expression in the T 0X2 deletion strains ..... 70
Western blot of HTS preparations from different strains used in this study .. 72
HC-toxin synthetase (HTS) activity in the RV isolates of C. carbonum ....... 73
Restriction map of EXGI genomic and cDNA clones .................................... 88
Nucleotide and deduced amino acid sequences of EXGI ............................... 89

Comparison of the predicted amino acid sequence of C. carbonum EXGlp
with other B-1,3-glucanases from the same novel family ............................... 92

Alignment of the amino acid sequences of the two 23-amino acid motifs of
EXGlp with the corresponding related sequences of other proteins .............. 95

Northern blot analysis of EXGI expression in C. carbonum grown on
different carbon sources .................................................................................. 98

Design of pGPDl 8, the vector for HTS domain A expression driven by
the GPD] promoter ....................................................................................... 104

Southern blot analysis of pGPD18 integration into the PGN] locus of the
Tox2_ strain 243-10 ..................................................................................... 107

Northern blot analysis of transformants bearing pGPD18 expression
vector ............................................................................................................. 110

Western blot analysis of protein extracts from transformants bearing
pGPD18 expression vector ........................................................................... 112

Figure 35. Nucleotide and partial deduced amino acid sequences of CC] 15 ................ 1 17

Figure 36. Comparison of the partial predicted amino acid sequence of C. carbonum
CC115 with transposase sequences from filamentous fungi ........................ 118

xi

INTRODUCTION

Interactions between plants and their microbial pathogens constitute an important
area of study both because of the basic biological mechanisms involved and because plant
diseases often pose a serious threat to agricultural productivity. Many plant-pathogen
interactions are very complex, being a result of co-evolution directed by the reciprocal
selection pressure that the pathogen and the host plant impose on each other. Pathogen
specificity and plant resistance are the results of this co-evolutionary process.

The concepts of host specificity and host range describe the restriction of a
particular pathogen to particular host plants. Both organisms, the pathogen and the plant
host, contribute to their interaction and to the outcome of this interaction. Consequently,
two genomes and two sets of biochemical and physiological traits are involved. The
presence of two genomes adds to the challenge of dissecting the genetic elements
controlling these interactions. If a particular pathogen can cause disease on a particular
plant, that plant is considered to be susceptible to that pathogen, and the interaction is
called “compatible”. If no disease results, the interaction is called “incompatible”. At the
start of both types of interactions, one or a few plant cells are invaded by the pathogen. In
an incompatible interaction, growth of the pathogen ceases without invading more than a
few plant cells, but in a compatible interaction the pathogen continues to spread fiom the
initial site of infection and causes disease symptoms (Walton, 1997).

All phytopathogenic microorganisms have a set of traits that allow them to
colonize their host. These traits, or factors, are generally divided into basic and host-

specific pathogenicity factors. Basic pathogenicity (or compatibility) factors are non-

specific; they can similarly affect a broad range of plants and allow a microorganism to
penetrate and colonize host tissue and to use it as a source of nutrition. Without some
minimum set of these factors, a microorganism is capable only of saprophytic growth and
can not be a plant pathogen. Pathogen-secreted cell wall degrading enzymes are thought
to be part of this attack “arsenal”, helping the pathogen to penetrate the cuticle and
epidermal cell walls of the host plant (Walton, 1994). Numerous studies of bacterial and
fungal cell wall degrading enzymes have been aimed at determining which ones are
important for pathogenicity. Of the many cell wall degrading enzymes studied, only three
(one cutinase and two polygalacturonases) have so far been shown to have a role in host
penetration and/or virulence. Introduction of a cutinase gene from Nectria haematococca
(Fusarium solam' f sp pisi) into the obligate wound pathogen Mycosphaerella sp. enabled
this fungus to penetrate the intact cuticle of papaya fruits (Dickman et al., 1989).
Recently, two polygalacturonases were shown to be required for full virulence of fungal
plant pathogens: P2c contributes to invasion and colonization of cotton bolls by
Aspergillus flavus, as determined by gene disruption and by introducing the gene into the
strain that previously lacked it (Shieh et al., 1997) and BcPGl contributes to the growth
of Botrytis cinerea lesions on tomato leaves and on apple and tomato fruits, as
determined by gene replacement experiment (ten Have et al., 1998).

In contrast to the basic pathogenicity factors, host-specific, or host-selective,
pathogenicity factors (e.g., host-selective toxins) are designed to break a defense of a
particular host. Thus, they are positive determinants of host range and disease specificity.
Plant species, varieties, cultivars or genotypes that are sensitive to a host-selective

pathogenicity factor are susceptible to the producing pathogen, and the resulting plant-

microbe interaction is compatible (Walton, 1996). In contrast to host-selective toxins,
host-selective elicitors are defined as pathogen-derived inducers of plant resistance
leading to an incompatible interaction (Walton, 1997).

Although many plant pathogenic bacteria and fungi produce phytotoxins, most of
these toxins are nonselective and could be considered to be basic pathogenicity factors.
All known host-selective toxins are produced by fungi. Most of them are produced by
fungi belonging to the genera Cochliobolus and Alternaria. Other well-studied fungi
producing host-selective toxins are Phyllosticta maydis (PM-toxin) and Periconia
circinata (peritoxin). All known host-selective toxins, except for Ptr-toxin from
Pyrenophora tritici-repentis, are low molecular weight compounds with diverse
structures. Ptr-toxin is a ribosomally synthesized polypeptide (Ballance et al., 1989;
Walton, 1996).

Fungi, as a group, are the most economically important plant pathogens. They are
also very diverse in their life cycles, morphology, and in the nature and complexity of
their interactions with plants, ranging from obligate pathogens to saprophytes, which can
cause opportunistic plant infections (Alexopoulos and Mims, 1979).

Cochliobolus carbonum Nelson (anamorph, Helminthosporium carbonum
Ullstrup, synonym Bipolaris zeicola (G. L. Stout) Shoemaker) is a filamentous fungus
belonging to the Ascomycetes (Ascomycota). It causes northern leaf spot and ear mold of
maize (Zea mays L.) and was first discovered when it appeared suddenly on susceptible
inbred cultivars of maize in the late 19308 (Ullstrup, 1941). Tox2+ (race 1) isolates of
this fungus produce a host-selective toxin known as HC-toxin, a cyclic tetrapeptide of the

structure cyclo(D-prolyl-L-alanyl-D-alanyl-L-Aeo), where Aeo is 2-amino-9,10-epoxi-8-

oxodecanoic acid (Figure 1). This toxin is required for the pathogenicity of C. carbonum
on susceptible cultivars of maize (Scheffer and Livingston, 1984). For the toxin to be
active, the epoxide group and the carbonyl group of the Aeo component are required
(Walton and Earle, 1983; Kim et al., 1987). Scheffer et a1. (1967) showed that toxin
production and virulence segregate 1:1 in the progeny of crosses between HC-toxin
producing (Tox2+ or race 1, virulent) and non-producing (Tox2— or race 2, non-virulent)
isolates of C. carbonum. This indicated that a single Mendelian locus, termed TOX2,
controls HC-toxin production.

Both race 1 and race 2 isolates of C. carbonum occur naturally in maize fields.
Maize cultivars that are homozygous for the hm] allele are sensitive to HC-toxin and
susceptible to race 1 isolates of C. carbonum (Nelson and Ullstrup, 1964; Meeley et al.,
1992). If the inoculum is large enough, the whole plant dies in several days (Scheffer et
al., 1967). If the maize cultivar harbors the dominant Hm] allele, it is resistant to race 1
isolates. Race 2 isolates of C. carbonum (non-toxin producing) cause only small, necrotic
flecks on both susceptible (hmI/hmI) and resistant (Hm]/-) maize cultivars (Ullstrup,
1941; Scheffer et al., 1967). If HC-toxin is added to race 2 spores before inoculation, the
fungus colonizes susceptible maize leaves with symptoms similar to those caused by race
1 (Comstock and Scheffer, 1973). Hm], a dominant maize gene responsible for
resistance to C. carbonum, was shown to encode HC-toxin reductase, an enzyme that
inactivates HC-toxin by reducing the carbonyl group of the Aeo component (Meeley et
al., 1992; Johal and Briggs, 1992).

The most important enzyme in HC-toxin biosynthesis is HC-toxin synthetase

(HTS) (Walton, 1987; Walton and Holden, 1988). This enzyme catalyzes the L-proline,

O
O
O
HC-toxin
(Cochliobolus carbonum
race I )

 

O

Trapoxin B

(Helicoma ambiens)

   

O
O
O
Cyl-2 WF-3 l 61
( C ylindrocladium scopan'um (Petriella guttulata)

Cylindrocladium macrosporum)

   

H O
O
O
O
Chlamydocin Apicidin

(Diheterospora chlamydosporia) ( F usarium sp. )

Figure 1. The family of cyclic tetrapeptides from filamentous fungi: Aeo-containing
peptides and apicidin (after Walton et al., 1997 ; Darkin-Rattray et al., 1996).

D-alanine, and L-alanine dependent ATP/PP; exchange and epimerizes L-proline to D-
proline and L-alanine to D-alanine (Walton and Holden, 1988). HTS is encoded by a
gene called HTS], which consists of a single 15.7-kb open reading frame (Scott-Craig et
al., 1992) and is present in two copies in most naturally occurring race 1 isolates of C.
carbonum (Ahn and Walton, 1996). Both copies are functional, since disruption of both
COpies results in the loss of HTS enzyme activity, toxin production and virulence, while
disruption of either copy alone results in a wild type phenotype (Panaccione et al., 1992).

In addition to HT S], there are other genes involved in HC-toxin production.
Adjacent to both copies of HT S1 , and transcribed in the opposite direction, is T 0201, a
gene predicted to encode a small molecule efflux pump. It has proven to be impossible to
recover mutants with both copies of T OXA disrupted (Pitkin et al., 1996). Thus, T OXA
may be essential for the protection of the fungus against HC-toxin, potentially by
transporting the toxic compound out of the cell. Another gene, TOXC, encodes a protein
that is highly similar to the [3 subunit of fatty acid synthetase and may be involved in the
biosynthesis of Aeo. Different isolates of C. carbonum have two or three copies of
TOXC. When all copies are disrupted, toxin production and virulence are lost (Ahn and
Walton, 1997). Another gene unique to Tox2+ strains, TOXD, has no known fimction and
its disruption has no effect on toxin production or virulence (Ahn and Walton, 1996; Y.
Cheng, unpublished data). TOXE, a gene encoding a DNA-binding protein, is required
for expression of TOXA, T OXC and TOXD, but not HTS] (Ahn and Walton, 1998).
TOXEp protein has been shown to bind the T OXA promoter (K. Pedley, unpublished
data).

The genes that have been identified so far are probably not sufficient for HC-toxin

biosynthesis. There may be more genes involved in Aeo biosynthesis (e.g., or subunit of
fatty acid synthetase interacting with T OXC, genes involved in the biosynthesis of the
epoxide group, genes for the regulation of toxin production).

The complexity of HC-toxin biosynthesis and the number of genes involved raised
the question of why toxin production appears to segregate genetically as a single
Mendelian locus. This was addressed by physical mapping combined with pulsed field
gel electrophoresis analysis (Ahn and Walton, 1996). All known HC-toxin biosynthetic
genes are unique to Tox2+ strains, being found only in isolates that produce HC-toxin and
completely lacking in naturally occurring avirulent, non-toxin producing isolates. All
known HC-toxin biosynthetic genes were found to reside on the same supemumerary
chromosome, called the T 0X2 chromosome. In different Tox2+ isolates, two partially
homologous T 0X2 chromosomes, one 3.5 Mb and another 2.2 Mb in length, were found.
The known TOX2 genes were physically mapped on the 3.5-Mb chromosome in strain
SB111 (Ahn and Walton, 1996). Tox2_ isolates lack the chromosome of corresponding
size, which can explain the Mendelian segregation of HC-toxin production. Since at least
part of this chromosome is missing in Tox2_ isolates, this part (called the TOX2 module)
is conditionally dispensable and is assumed to be essential. only for HC-toxin production
and thus for virulence on susceptible maize. Since C. carbonum competes well as a
facultative saprophyte, the presence or absence of this TOX2 module may be part of
natural genetic diversity within this species. It is also possible that TOXZ module has
been acquired from some other species at the time prior to the first corn leaf spot disease
outbreak. Isolates harboring the TOX2 module may be favored when a compatible host is

present, and in the absence of such a host, isolates without it may be favored because of a

decrease of parasitic fitness resulting from unnecessary toxin production. In fact, the
frequency of Tox2Jr isolates in the absence of susceptible maize varieties in the field is
less than 2% of the total population of C. carbonum, while in the presence of uniform
plantings of susceptible maize Tox2+ ( race 1) isolates are frequent (ca. 50% of all the
isolates from infected plants) and cause damage (Leonard, 1978).

The size polymorphism of different variants of the T 0X2 chromosome and the
hybridization pattern of different markers can be explained by a reciprocal translocation
involving four chromosomal segments (Ahn and Walton, 1996). This explains two facts:
at least some Tox2— isolates do contain a chromosome (2.0 Mb in size) that contains at
least two single-copy sequences (markers G242 and Q) that are also present on the 3.5-
Mb chromosome of 831 11 (Ahn and Walton, 1996; Canada and Dunkle, 1997), and in
some isolates the second copy of TOXE is located on the 0.7-Mb chromosome which is
assumed to be involved in the translocation (Ahn and Walton, 1998).

The mode of action of HC-toxin is not clearly understood. It does not cause cell
death. It is known to inhibit mammalian cell division (Walton et al., 1985) and to
increase the uptake of certain ions, amino acids and nitrate (Yoder and Scheffer, 1973a,
1973b). It has been proposed that HC-toxin blocks the ability of a susceptible maize
plant to develop a defense response to a pathogen attack, thus allowing the fungus to
colonize the host plant tissue. It was shown that HC-toxin significantly inhibits histone
deacetylase (Brosch et al., 1995). As histone deacetylase is considered to be involved in
global gene regulation, this may be a way in which HC-toxin affects expression of plant

defense genes.

Apart from the host-selective toxin HC-toxin, C. carbonum, like any other plant

pathogen, must produce and secrete a number of basic pathogenicity factors that allow
colonization of susceptible host tissue. The ability to penetrate the plant tissue is very
important in this respect. Germinating C. carbonum enters plants predominantly through
the intact cuticle and epidermis, rather than using wounds or stomata (90% of fungal
penetrations are at cell junctions and 10% at stomata] openings [Jennings and Ullstrup,
1957]). Based on the ultrastructural analysis of the initial penetration (Murray and
Maxwell, 1975), and on the data on melanin-deficient mutants (which are deficient in
hydrostatic pressure in the appressorium) (Kubo et al., 1989; Kubo et al., 1991; J. Pitkin,
unpublished data), plant cell wall dissolution is thought to be more important than
mechanical pressure for the C. carbonum penetration. Therefore, cell wall-degrading
enzymes must play an important role in C. carbonum pathogenicity. To date, 17 genes
encoding various cell wall degrading enzymes from C. carbonum have been cloned and
null mutants have been created for most of them via transformation-mediated gene
disruption (reviewed by Scott-Craig et al., 1998b). None of the genes known so far was
found to be essential for virulence. This may be due in part to their redundancy, since
most cell wall-degrading enzymes exist as multiple isoenzymes, encoded by different
genes. Studies of the mutants deficient in several of these genes (multiple gene
disruptions) are in progress and may reveal sets of enzymes at least one of which is
required to be present in order for the fungus to be virulent.

The mechanisms by which plant pathogenic fungi evolve and change the host
range and virulence are poorly understood. In this study, molecular mechanisms of C.
carbonum virulence and pathogenicity were investigated. The role of horizontal gene

transfer and supernumerary chromosome (TOX2 chromosome) instability in the evolution

of the C. carbonum virulence were evaluated. In order to assess the possible role of
horizontal gene transfer in the acquisition of non-ribosomal peptide synthetase-encoding
genes by C. carbonum and several other fungal species, the corresponding deduced amino
acid sequences were compared, and found to be not similar enough to support the
horizontal gene transfer hypothesis. A study of alterations in the TOX2 chromosome
revealed exceptional meiotic and mitotic instability of this chromosome, and it was
shown that alterations of this chromosome can result in a novel phenotype of reduced
virulence. In addition, EXGI, the gene encoding B-1,3-exoglucanases from C. carbonum,

was cloned and found to be a member of a novel glucanase family.

10

Chapter 1

IDENTIFICATION OF PEPTIDE SYNTHETASE-ENCODING GENES FROM
FILAMENTOUS F UNGI PRODUCING HOST-SELECTIVE PHYTOTOXINS OR

THEIR ANALOGS

Abstract

Part of this chapter was published as Nikolskaya et el., 1995.

Race 1 (Tox2+) isolates of Cochliobolus carbonum produce a cyclic tetrapeptide,
HC-toxin, that is necessary for their virulence on certain genotypes of maize. The
synthesis of HC-toxin is catalyzed by a 570-kDa, multifunctional enzyme, HC-toxin
synthetase (HTS). The gene encoding HTS (HT S1) is absent from other races of C.
carbonum and from other species of Cochliobolus. Four other unrelated filamentous
fungi produce cyclic peptides closely related to HC-toxin, raising the possibility that the
corresponding non-ribosomal peptide synthetase-encoding genes have moved between
these fungi by horizontal gene transfer. Degenerate PCR primers were designed based on
several highly conserved amino acid motifs common to known non-ribosomal peptide
synthetase domains and used to amplify genomic sequences from different fungi. PCR
products representing non-ribosomal peptide synthetase genes from Diheterospora
chlamydosporia, which produces the HC-toxin analog Chlamydocin, and Cylindrocladium
macrosporum, which produces the analog Cyl-2, were cloned and analyzed. The genomic
locus encoding the corresponding putative peptide synthetase from D. chlamydosporia

was cloned and mapped, and partial sequence was obtained. All putative non-ribosomal

ll

peptide synthetase amino acid sequences obtained from these fungi were found to be
closely related to HTS, but the percent amino acid identity was not consistent with a very

recent horizontal movement of these genes.

Introduction

Cyclic peptides (and some linear peptides) are synthesized by a class of enzymes
known as non—ribosomal peptide synthetases (reviewed by Kleinkauf and von Dohren,
1996; Cane et al., 1998). Multifunctional peptide synthetases, those that catalyze
activation of more than one amino acid, are organized into synthetase units (ca. 1000 -
1500 amino acids in length), one for each amino acid substrate. Each unit contains
conserved motifs known or believed to be involved in adenylate formation, ATP binding,
peptide bond formation, and sometimes epimerization and cyclization (Gocht and
Marahiel, 1994; Kleinkauf and von Dohren, 1996). The most highly conserved part of
each unit (ca. 600 amino acids), the amino acid activating domain, performs aminoacyl
adenylation and thioester binding.

HC-toxin (Figure 1, p. 5) is a cyclic tetrapeptide produced by the filamentous
fungus Cochliobolus carbonum. It is a critical pathogenicity determinant in the
interaction between C. carbonum and its host, maize, Zea mays (Panaccione et al., 1992).
A non-ribosomal peptide synthetase of 570 kDa called HTS, encoded by HT S1 (Scott-
Craig et al., 1992), catalyses activation of L-Pro, D-Ala and L-Ala, and epimerizes L-Pro
and L-Ala, and is presumed to activate the fourth amino acid, 2-amino-9,10-epoxi-8-
oxodecanoic acid (Aeo) or an Aeo precursor and to polymerize and cyclize the peptide.

HTS contains four conserved domains, each approximately 600 amino acids long, which

12

are similar to each other and to conserved amino-acid activating domains from other non-
ribosomal peptide synthetases. The protein sequences between domains are
approximately 1000, 550 and 550 amino acids long. Each of the four domains is believed
to activate one of the four amino acids, L-Pro, D-Ala, L-Ala and Aeo.

Naturally occurring isolates that do not produce HC-toxin completely lack the
locus encoding HTS, which raises the question of the origin of HC-toxin production in C.
carbonum. Non-toxin producing isolates might have lost HT S] or it might have been
acquired by the toxin-producing strains by horizontal gene transfer from another
organism. Four other unrelated filamentous fungi produce cyclic tetrapeptides
structurally similar to HC-toxin (Figure 1; Table 1). To explore the possibility that C.
carbonum might have acquired the capacity to synthesize HC-toxin by horizontal gene
transfer of HT S] from one of these other fungi, we isolated the corresponding non-
ribosomal peptide synthetase genes from these species and compared their sequences.
Prior to this study, the strategies used to isolate fungal non-ribosomal peptide synthetase
genes have involved cross-hybridization with bacterial non-ribosomal peptide synthetase
genes (Smith et al., 1990) or purification of the non-ribosomal peptide synthetases to raise
antibodies to screen expression libraries (Haese et al., 1993) or to obtain amino acid
sequences to be used for designing oligodeoxyribonucleotide probes (Scott-Craig et al.,
1992). While this work was in progress, another PCR strategy to isolate non-ribosomal
peptide synthetase from bacteria was described (Turgay and Marahiel, 1994), and since
the publication of the results described in this chapter (N ikolskaya et al., 1995), numerous
other peptide synthetase genes from different sources have been cloned (e.g.,

Schauwecker et al., 1998; Konz et al., 1997; Annis and Panaccione, 1997; Van

l3

£3 covo>
awe 58>
A53 6333
one .800
one .890
33 .800

oom- :3 .0 33a:
6w3 .800

22 .22:
at: .a a 52:

$2 :3 6 «35:5
”2&— .anzwsm Ea ammo—U

Eb— .thH use .3525
Hwo— datum

9mg 58—35

oae 5833

353.3%

5x8
A. 8355 8: meow .v0 9 oEowSTBoc

5x29 8268a .83.: moNEmfinn
98 monEman
oinqoaam
oinaoaa

83353

32853

853 kiwi. got? moNEmflaa

3335mm :8

$32.25: was E938 «3:683

_ _ Em 9 38:: 633$ 2cowofimqéoc

SEE moNEmEaa

26E Egg—com

565— use:
565. 28:
£883
535. 28:
:32; one:

5305. use:

A8<-._-o§-._-o§-._-o£-oCo_ca
5x095

824552-992Eaaégoa
23

A8<Jaid535153:23
.2 FE

Aco<-._-e<-u§-g-oi-oV205
583535

5522 one:

€~<a_a_<-o-a_<-._-o£-eo_oa
56.0:

A. 5x8 oEanoExU

mo §<Q9=ue§8< 0

v0 Exmehuoksmx 0

<3: octet; U

Gage 68$ seaweed .:
38S 08$ ”cases: r

638 8:; $8228 .:

39556 i

333.. DUES
$3.698;er .10

8353M Q

$23 8::
stemmeuAEeEQ .Q

E Em 58.33 9

$9.50 UHF—{V
27mm Ezzoaaco U

.333»

.835:— zo... :2: 258» 3:an ES .35: £5 E can: 830% .«wnah A 935—.

14

(Table 1 - continued)

a Isolates of D. chlamydosporia, Cy]. macrosporum, H. olivaceum, H. monilipes and H.
palmigenum were obtained from ATCC. H. ambiens RF-1023 was a kind gift of
Shionogi and Co., Japan; P. guttulata of F ujisawa Pharmaceutical Co., Japan; C. victoriae
of Dr. R.P.Scheffer, MSU; and C. heterostrophus of Dr. O.C. Yoder, Cornell University.
b Toxin production by the strains of D. chlamydosporia, Cy]. macrosporum, P. guttulata
and H. ambiens was confirmed for all species except for P. guttulata from which WF-
3161 could not be detected (Figure 2).

15

Wageningen et al., 1998). Here we describe a method to isolate fungal non-ribosomal
peptide synthetase genes using PCR. PCR primers based on highly conserved non-
ribosomal peptide synthetase motifs were used to isolate nucleotide sequences encoding

non-ribosomal peptide synthetase domains from several different fungi.

Results and discussion

PCR strategy
As potential sources of non—ribosomal peptide synthetase genes several
taxonomically unrelated species of filamentous fungi were examined (Table 1) that
produce Aeo-containing cyclic tetrapeptides. We also used several other species from the
genus Helicoma that are not known to produce any cyclic peptides. Data on the non-
ribosomal peptide synthetase sequences obtained by D. Panaccione (Nikolskaya et al.,
1995) from C ochliobolus victoriae using the same strategy are used in Figures 3 and 5 for
the purpose of comparison. C. victoriae is an oat pathogen that produces the cyclic
pentapeptide victorin. Victorin is not a member of the Aoe-containing cyclic tetrapeptide
family. Although they are in the same genus, C. victoriae does not produce HC-toxin and
C. carbonum does not produce Victorin (Walton et al., 1995).
Preparations made from Cy]. macrosporum, P. guttulata, D. chlamydosporia and
H. ambiens culture filtrates were analyzed by TLC for the presence of the epoxide groups.
Epoxide group-containing compounds were detected in Cy]. macrosporum and D.
chlamydosporia—produced culture filtrates, trace amounts were detected in a H. ambiens

preparation (Figure 2).

1234

 

 

Figure 2. TLC analysis of epoxide-containing toxin production by representatives
of the fungal species used in this study. Toxins were purified from 24-day culture
filtrates, fractions were lyophilized, resuspended in methanol and loaded onto silica gel
TLC plates. Epoxide-containing compounds were detected by spraying plates with the
epoxide indicator 4-(p-nitrobenzyl)-pyridine (p-NBP) and are visible as bright spots.
Lanes: 1, Cyl. macrosporum; 2, P. guttulata; 3, D. chlamydosporia; 4, H. ambiens.

Degenerate oligodeoxyribonucleotide PCR primers were designed based on highly
conserved amino acid motifs in the non-ribosomal peptide synthetase genes for which
sequence information was available (Turgay et al., 1992). Codon bias was used in the
design of some of the primers based on codon usage in HTS] in order to reduce
degeneracy (Figure 3B). The sequences of the oligodeoxyfibonucleotides and their
relative positions within the consensus sequence of the conserved domains of HTS from
C. carbonum are shown in Figure 3 (A and B).

Four primers were synthesized and used in different combinations for PCR with
the genomic DNA of each species as template (Figure 3C). All species tested except
Petriella guttulata, Helicoma monilipes and H. palmigenum yielded at least some PCR
products of the predicted sizes (Figure 3C). We were unable to detect WF-3161 (Table 1)
in our culture of P. guttulata, which may explain the negative PCR results for this
species. These PCR fragments were isolated from gels and used as template for a second
PCR reaction (reamplification) with the same primers in order to increase the amount of
material for cloning. Figure 4 shows reamplified PCR products from four PCR reactions

in which Cy]. macrosporum DNA was used as template.

Analysis of putative non-ribosoma] peptide synthetase gene fragments

Reamplified PCR fragments were isolated from gels and subcloned into the
plasmid pUC18. A number of independent plasmid clones were obtained from each
size-class of the PCR products and partial nucleotide sequences were determined: 18
clones from D. chlamydosporia yielded seven different sequences, six from Cy].

macrosporum yielded three, 12 from H. ambiens yielded eight, and three from H.

18

A.

 

 

 

 

 

AI--WDG--S Y-BL---S-- LA--L—--G- -------V—- —FEKS—-AVV -M-A—-KAGG 6O
—FVP-DP-—P -—RL--II-- --A---L -------- L -------------- K—I ------- 120
GKPKG
------------------------- AY-LF TSGSTG-PKG -V--H—---- -—----—--- 180
-—G----—R— LQF-SY—FD- SI-DIF—-L— -GGCLCIP-E E-R---NL --------- N-- 240
YGP TE
—LTPS-——-L -P-—-P---- -L--—GE—-- —S---4W—-- ---L-N-YGP TE ------ A- 300
RGYLN
------------ S-IG---- ---WV--P-- ---LVP—GA- GEL-IE---L AR—YL——P-R 360
YKTG DL

T---F ----- WL ------------ R-Y-TG DLVRY--DG- L---GRKD-— Q-K--GQR-E 420
LGE-E --------- DP ------ V—L -------- R ------------- A-L --G ------- 480
---------------- L--- ---------- ---LP--MVP ---------L P--—SGKLDR 540
—-—R ----- L ---------- T ------------------- -LR--W -------------- 600
L ---------- F---GG-SI -A--L—---- R--G--L-V- DIP 643
B.

Primer amino acid nucleotide sequence Length Degeneracy
number sequence

78 YGPTE 5'—TCTAGATAYGGNCCNACNGA l4mer 96-fold

79 YKTGDL 5'-TCTAGAARRTCNCCNGTYTTRTA 17mer 256-fold

114 GKPKG 5'-TATCTAGAGGNAARCCNAARGG l4mer 64-fold

115 RGYLN 5'—TATCTAGARTTNAGRTADCCHCG 15mer l44-fold
C.

Resulting size (bp)

Primer Predicted

combination size (bp) Dc Cm Ha Ho Hm Hp Pg CV

78 + 79 300 300 300 np np np np no 300
114 + 115 600 nd 600 np 800 np np np nd

78 + 115 180 nd 180 700 500 np np np nd
114 + 79 720 nd 700 700 650 np np np nd

 

Figure 3. PCR strategy to amplify peptide synthetase sequences from several fungal
species. (A) Consensus sequence of the four amino acid-activating domains of HTS
encoded by HTS] (accession # M98024). Consensus amino acids are indicated in normal
type when three of the four amino acids in the four HTS domains are identical and in bold
type when all four are identical. Sequences on which the PCR primers were based are
shown above the consensus sequence and highlighted. When there was no consensus
between HTS domains, amino acids for the primers were chosen based on other available
non-ribosomal peptide synthetase sequences (Turgay et al., 1992). (B) Design of the PCR
primers. The underlined linker sequence was added to each primer to facilitate cloning.
In the nucleotide sequence, R represents G or A; Y represents C or T; H represents A, C
or T; D represents G, A or T; N represents A, C, G or T. (C) PCR results with various
primer combinations. Dc, D. chlamydosporia; Cm, C y]. macrosporum; Ha, H. ambiens;
H0, H. olivaceum; Hm, H. monilipes; Hp, H. palmigenum; Pg, P. guttulata; Cv, C.
victoriae; nd, not done; np, no product generated by PCR.

19

DP

1900-,
1200-

700-

200-

 

Figure 4. Analysis of the products of the second round of PCR (reamplification) of
Cy]. macrosporum DNA by gel electrophoresis. Lanes: 1, primer combination 78+79; 2,
primer combination 1 14+l 15; 3, primer combination 78+ 1 15; 4, primer combination 1 l4+79.

20

olivaceum yielded three. Based on the presence of characteristic non-ribosomal peptide
synthetase amino acid motifs other than those used to design the PCR primers, one
sequence from D. chlamydosporia and three from Cy]. macrosporum encoded portions of
non-ribosomal peptide synthetase domains. Complete sequences of these clones were
obtained and their predicted amino acid sequences are shown in Figure 5. Three C.
victoriae sequences obtained by D. Panaccione are also shown in Figure 5. No PCR
products obtained from Cy]. macrosporum with primer combinations 78+115 and 114+79
were sequenced. None of the PCR products from H. ambiens and H. olivaceum
contained non-ribosomal peptide synthetase -like sequences.

Comparison of the predicted amino acid sequences of the PCR products from C.
carbonum, D. chlamydosporia and Cy]. macrosporum (Figure 5) indicates that these
products are clearly related. When compared against the GenBank DNA database, the
highest BLAST (Altschul et al., 1993; Altschul et al., 1997) scores for all of the PCR
products were to domain B of HTS, followed by other domains of HTS and then domains
from other non-ribosomal peptide synthetases. We conclude that degenerate
oligodeoxyribonucleotide PCR primers based on conserved non-ribosomal peptide
synthetase motifs can be used to identify non-ribosomal peptide synthetase-like
sequences in a variety of fungal species. Indeed, since the publication of this work,
several researchers have requested and used our primers to clone non-ribosomal peptide
synthetase genes.

The significance of the fact that the deduced amino acid sequences of the PCR
products always showed more similarity to domain B of HTS than to the other domains of

HTS is not clear. Percent amino acid identity as determined by BESTFIT (Program

21

 

 

HTSl-A GVPKLLVV ...unsuv

 

I‘lnuTFTnugxuusu

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

HTSl-B IlVlT'I TGDTWTTLAVGGCL
HTSl-C GKPKGVVMEHRAWSLGFTCHA.. I _ °TTFI I
HTSl-D unwnvav ‘A' T‘T‘P - “Mr“! ' u... - 'nT aGGCL
CYL—l GKPKGVVVHHRALCTSIEEHG....-_ inaiirLAEQGSV
HTSl-A PTPQPPDR ”““’ vnusiSLLIFTGEASREADTV
HTSl-B PI inn.uv.nnu.1PSLALHLDPDAVPT. LKALCVAGEPLSMSVVT
HTSl-C FIPSDKER.VNNLgurrnLNUi ““ ISFAGFIGEPMTRSLID
HTSl-D PTPKEFOR “°“”>'""' .LETLCLTGEANLQSDVD
CYL-l CIGSESDR. IEDLAGFFNRFKVNWTVLTPAIARMLNPEDVPT. LQTLICGGDAIGDLTPR
HTSl-A PVT *TTATRTRK ...... GKSSNIGYGVNTRTWVTDV.SGAC
HTSl-B VWS. KRLNLINMYGPTEATVACIANQVTCT ..... TTTVSDIGRGYRATTWVVQPDNHNS
HTSl-C ‘"T' vuvannnlaylfip .HDKPSSNIGHALGANIWVVEP. QRTA
HTSl-D RWA. PKLHLIAGYGPTETCIMSVSGELTPS ...... ““1“”W‘nnviuythTE
CYL-l KWS.SKLRFIQVYGPTETTIVVSISDRQNK ..... EVRPAMIGHMFTSAAWIVNPRNHDI
CYL-2 YGPTETTVIATCHVFQSTS.... ”*VNI

CYL-3 YGPTETTVICVARQFPEES. .ETDPTNIGKPVGCRAHVVDPTDYSS
DIH-l YGPTECTIWTSRYEVGGQS ..... LDHLNLUKAMbL

CV-2 YGPTECAVWCTSYTNGASG ..... YKPGIIGKPIASVSWVVDPDDCNK
CV-7 YGPTECACVATCNIMTPR ...... “"“ Vinny";

CV-19 YGPTEunv- Thu-Pl! I

HTSl-A LVPVCSTPVTTT”°""*‘“"V ""c“7°“Tva .n“9.RRFYRTGD

HTSl—B LVPIGAvawrrrraqurKhy ..ixsyswnHDLRPNST ...... LYKTGD

HTSl-C LVPIGAVGELCIEAPSLARCYLANPERTEYSFPSTVLDNWQTKK ...... GTRVYRTGD

HTSl-D LAPYGATGELYIQGPTVARGYLHDDVLTSKAFIVDPQWLTGYKTNENQW.SRRAYKTGD
CYL—l LMPVGAAGELLIEGPVLARGYLN
CYL-2 LAPIGCVGELVIQGPTLARGYFNDDSKTSESFIEGVNILPANLATGNP....RFYKTGD

 

 

 

 

 

 

CYL-3 LTSIGAIGELVTFCDNT'T'DCVY an“ SLFDTPSST. .AFNLYKTGD
DIH-l LVPAEGVGELLVTGPILSRGYLNKPEATQLAFVTDLEWAKEKDA ........ RFYKTQ

CV-2 LAPPGAIGELLVEGPIQARGYLNDIVKTEAAFINNPSWLVAG. SKTCAGRQGRLEKTGD
CV-7 Ln; __ ma T Duemnrrruvncw PLRIYKTGD
CV-19 umrvuivunuvlnurLJARGv LTNPLWAATIAARDGAFETVRMYKTGD

 

Figure 5. Alignment of the predicted amino acid sequences of the PCR products
containing conserved peptide synthetase sequences with the corresponding
sequences from the functional domains of HTS. Peptide synthetase sequences are as
follows: one amplified from D. chlamydosporia (DIH-l), three from Cy]. macrosporum
(CYL-l, CYL-2 and CYL-3), three from C. victoriae (CV-2, CV-7, CV-19) (sequenced
by D. Panaccione, see Nikolskaya et al., 1995), and corresponding conserved sequences
from the four functional domains of HTS (HTSl-A through D). Conserved amino acids
are highlighted, and amino acids that were used to design the PCR primers are
underlined. Alignments were done using PILEUP (Program Manual for the Wisconsin
Package, 1994). The corresponding nucleotide sequences have been deposited in
GenBank with accession numbers L42329 (CYL-l), L42330 (CYL-2), L4233l (CYL-3),
L42332 (DIH-l), L42312 (CV-l9), L42313 (CV-7), L42314 (CV-2).

22

Manual for the Wisconsin Package, 1994) was also slightly higher between domain B of
HTS and all PCR products. However, the percent amino acid identity between the
different PCR products and the different domains of HTS ranged from 34 to 48%, which
is only somewhat higher than, for example, the percent identity between the
corresponding fragments of the four domains of HTS itself (33 to 44%) and between the
domains of HTS and other non-ribosomal peptide synthetases of bacterial and fungal
origin (28 to 34%) (Scott-Craig et al., 1992).

Since the PCR primers were based on the amino acid sequence of HTS, the
products most likely represent the sequences in D. chlamydosporia and Cy].
macrosporum that are the most closely related to HTS. However, we do not know if the
PCR products from D. chlamydosporia and Cy]. macrosporum (Figure 5) are in fact parts
of the genes for the putative "chlamydocin synthetase" and "Cyl-2 synthetase",
respectively, because these sequences might be involved in the synthesis of other cyclic
peptides or might be nonfunctional. If these sequences do encode "chlamydocin
synthetase" and "Cyl-2 synthetase", then based on their low degree of similarity to HTS]
it seems unlikely that they derived from each other by horizontal gene transfer in the
recent evolutionary past.

When used as probes at high stringency, PCR products from the fungi producing
Aeo-containing tetrapeptides hybridized to genomic DNA of the species from which they
were derived but not to that of the other species listed in Table 1 (results not shown).
Likewise, HT S] does not cross-hybridize to DNA from the other fungi (results not
shown). Thus, among the fungi studied these genes are unique to their corresponding

species.

23

In contrast to the PCR products from D. chlamydosporia and Cy]. macrosporum,
PCR products obtained from C. victoriae by D. Panaccione (Figures 3 and 5) and by J.
Pitkin (not shown), hybridized not only to C. victoriae DNA but also to that of C.
carbonum and C. heterostrophus, although the degree of their similarity to HTS was
within the same range (CV-2, -7 and -19 are 43 to 48% identical to domain B of HTS). If
any of them represent the putative "victorin synthetase", then the gene for victorin
synthetase is unlike HTS] in being present even in species and isolates of Cochliobolus
that do not produce victorin. J. Pitkin has shown by targeted gene disruption that two of
the C. victoriae products are not parts of the victorin synthetase (J. Pitkin, unpublished
data). Alternatively, these sequences from C. victoriae might be nonfunctional or
represent one or more non-ribosomal peptide synthetase genes involved in the

biosynthesis of an unknown peptide common to all three species of Cochliobolus.

Cloning and mapping of the non-ribosoma] peptide synthetase gene locus from D.
chlamydosporia

In order to clone non-ribosomal peptide synthetase gene genes from D.
chlamydosporr'a we constructed a genomic DNA library of this species in phage XEMBL3
and screened it with the corresponding cloned PCR products. Three overlapping lambda
clones containing a genomic locus with a putative non-ribosomal peptide synthetase gene
gene from D. chlamydosporia were isolated (Figure 6). The locus was mapped by
restriction analysis of these lambda clones and plasmid subclones (Figure 6). The map
was confirmed by comparison to restriction enzyme-digested genomic DNA on Southern

blots probed with various regions of the locus (data not shown). Positions of the four

24

.Efiem .m “Boom .m ”new. .m .moxon x83 mm Esofi 2a
338% Biomcou .25» «32:5» Egan—«Eu 953.:— 2: £5.89 «a... 9.8. Egbes‘o .Q we 9:: 568.38% .9 0.59m

 

Nm&

 

mv&

25

 

 

mx

 

 

_ ___ _ ___—______ _

m mmm m m mm mmmmm m m m

e:

conserved domains (Figure 6) were determined by sequencing parts of the locus and
comparing the sequence to the consensus sequence of the known non-ribosomal peptide
synthetase gene domains. Two of the domains were sequenced completely on one strand.
Predicted amino acid sequences of domains A and B were compared to protein sequence
databases. They showed significant similarity to various peptide synthetases. One of the
best matches was to the sequences of HTS, surpassed only by the sequences of a putative
peptide synthetase from the entomopathogenic fungus Metarhizium anisopliae (Bailey et
al., 1996). Alignment of the sequences of D. chlamydosporia domains A and B and four
domains from HTS is shown in Figure 7.

Thioester binding core motifs, located in the C-terminal part of the peptide
synthetase domains, are SGGDSISAM in domain A and LGGDSLTAM in domain B
(Figure 7). They are in a reasonably good agreement with the consensus sequence
LGGXSIXAI deduced from other known non-ribosomal peptide synthetases (Stachelhaus
et al., 1995; Kleinkauf and von Dohren, 1996; Cane et al., 1998). The conserved motifs
involved in adenylate formation and ATP binding are also present in both domains
(Figure 7). They show some deviations from the previously reported consensus
sequences SGSTGXPKG, YGXTE, GELXIXGXXVAR, RLYRTGDL, and LXXYMVP

(Kleinkauf and von Dohren, 1996) deduced from other known non-ribosomal peptide

synthetases.

Attempts to disrupt the peptide synthetase gene of D. chlamydosporia

To test whether the cloned non-ribosomal peptide synthetase gene from D.

chlamydosporia was Chlamydocin synthetase, an attempt was made to develop a

26

C.carb.

C.carb.

C.carb.

(1

C.carb.

C.Carb.

C.carb.

C.carb.

.carb.

m w o 0 m y m > o n m > m y Utfi w > m y D O m y m r o n m > m r U 0 m > m r o O m y m p o n m y

313’00013’

AIDAWDGRMS
AICSWDGSLT
AIVAHDGQLS
AIZSWDGSI
AISSNDGLL
AVCSWDLDLT

TTNATIALVG
QVQAELLLCS
ESGAFLVL..
ETGssvrvr

QPNAHLLIAT
DVKAHIILAS

CIVVTHSQI
GAVATHQAY

GVVMEHRAW‘
GVVHEHHSVC
GVVISSFVAS
GIVLEHDAVA

O

WAVLTPTVSR
WAGITPSLAL
TVFLTPSIGK
TMLLSTSVSR
WAVFTPSELR
WALLTPTVAR

VNTRTWVTDV
YRATTWE'QP
LGANZWVVEP
VSCQAW‘JI NP
MGCSTYVVEA
VGGFMWLVDP

GSLICVGRSD
GKIIFIGRKD
GTLDFLGRKD
SNLYYVRRKD
GTVSFIGRK.
GSVSFVRRKD

PHNQQSLPKP
TUNSDEZFIA
HRSWDSMHVL
GPEDLEVI..
DSHNIRVMNT
PTTEYSTIRV

LGTTGLEVD.
..TSKALSTS
..TAPLDGNF
AFSLPQASAR
LVSRNL...

ALTGMLEDEQ

VKDLL
VRDIF
HEDIF
VAEIF
VRHVL
VQDVL

YTELERVSST
YAELSDLSQR
YNQMDRCADV
YHELDGLSSI
YQEIDELSSI
YAQLNDLSDR

AGKTAALFKS
PA.TSRMGAL
..... TLPES
AA.YSKLC..
A..TETLSMS
A..TYADQF.

TAVQAYK...
TGIYEHAVAC
LGFTCHA..

SALZALG...
LLPKSKTSLG
SSAKTHG...

FLDPGVVKDF
HLDPDAVPT.
LLNPKDLPN.
LMQPADTPS.
IIQPHQVPN.
MLPPEDVLHN

.SGACLVPVG
DNHNSLVPIG
.QRTALVPIG
LKETELAPYG
ANHNKLVPAE
NDHNVLVPVG

.TQIKLAGQR
.TQVKMNGQR
.GQIKLRGQR
SSQVKIRGQR
.TOVKLRGOR
N.QVKVRGNR

AFAQCPPCLV
TSTSSLSEFS
NIANTSENLQ

..GYMDDLK
IKRKEAEDLI
IDRKLQKQFN

PGGAASSVAS
VKDAETTDTV
PKVEQVLTTN
HPNQPTVTHT
PGGLRTTRR.
VSVDEPSTDL

WARQLQKQGI
LSIHLVSLGI
LARQIRKTNM
LAEHLSQLGV
FSLTLLKQHH
LALELRQNGV

ADTAVQTIDI
QNISTQMGTE
ANALATLSGL
...... LSLS
A ..... LDML
ADVVERVIVV

DRFGV.TSET
GMTSL.GAPP
EYMGF.NSCT
KRMGL.GPQS
RKLELPPNET
EAMMV.GKDS

ISTLIFTGEA
LKALCVAGEP
ISFAGFIGEP
LETLCLTGEA
LKTIIVGGEP
EHTI.ALGVE

SIGELLIESG
AVGELIIEGS
AVGELCIEAP
ATGELYIQGP
GVGELLVTGP
AVGEILIEGP

00......

VELGDVEAHL
FELGEVEHAL
IELGEIEHHI
VELAEIEEVI
VELTEIEIVL
IELGEIEFNL

KYAT..SSLQ
T'v'i K KLQLAQ
TLVTELKKSL
SHIIPALEA.
DNLR.AAAS.
TELQKIKDTL

DSDLRDMNDD
EDRLARIWEK
ESVLRQWWGT
QKLLRQLWCK
Q.VLLQTVAE
ENSLVQIWEH

...SQGSWVL
.KVGTK..IP
.ISAOSPFVC
.RPEAP..VC
PYLSKGDFVC
.EPEV..FVP

TKDIPHGLSD
FKIVELE.PE
TKVIPVSLSE
VRGIVCD.GS
..ILP...YI
SHTTLEDLKT

RVLQFSSYTF
RSLQFASYSF
RILQFSSLHF
R'v'l.QF.‘~.‘SYWF
RVLQHLSLTF
RVIQFAAYMF

SREADTVPWI
LSMSVVTVWS
MTRSLIDAWT
VLQSDVD.WA
VAADLIATWA
RVRKETVETL

HLADKYLNRP
ILCRGYLNDP
SLARCYLANP
TVARGYLHDD
ILSRGYLNKP
LLARHYLHEP

.00.

Q....SDPTT
QLQL..DPSD
RRLMSDDPRF
RQHI...PPD
......... A
....WNHPSV

...OFLPTYM
RAMEVLPLFN
R..GVLPEYM

.SLPliliM
R...VLAFYM
ANQ..LPSYM

.0...

SLLLTACSRV

FCFEKSRLAV
CFEKSIMWTI
IHLLRSATAV
LLFEKSKWAV
LCFEKSRWAV
LCFEKSGVHV

TVVQ ......
FIRSL..PLP
LVQQITDNTT
VFSSTKKPLP
T .........
AHFDESIVSR

DISIADTFTA
DASIGDIFTT
DLSILEIWAV
DVMLLDIFGT
DISILDIFTT
DVSIEDLFTT

EAGVNLYNVY
.KRLNLINMY
LPGRRLVNSY
.PKLHLIAGY
.DKVKLIISY
ADHTCLSLIY

'I

DRTEAAFLSD
ERTAEVFIRS
ERTEYSFPST
VLTSKAFIVD
EATQLAFVTD
EKTAASFIED

SQAAVVF...
GPIIVDLLKR
HEASVQLYNP
VTVCVDLLSS
STAAVFL...
RHGLAYMPYT

VPSUWLGIDF
VPQAYIPIEG
VPLHFVAVSR
IPEAYVPFVQ
VPTDWVVLNQ
IPSVWLIVHR

ti

LNLPAGKISY
..LGVKGVGR
LAMDPHSIQR
.QLDESAVNK
NKSAGTVNFK
.GNPRVSASR

VSMIAILKAG
VTI LAVVQAG
VSMLAVLKAG
VAMIGIIKAG
VSMVAV I KAG
VTMLAIIKAG

........ SN
PKPNHQPMVG
KKDEYCKSGD
STADSPPSFS

LFYGGTLCIP
LAVGGCLCIP
LYAGGCLFIP
LVYGGCLCIP
LEVGGCICIP
LMRGGTVCVP

GPAENTLITT
GPTEATVACI
GPTEACVLVT
GPTETCIMSV
GPTECTIWTS
GPAECCMTMS

O...

LPWIPNYEGD
PSNLHDLRPN
V..LDNHQT.
PQWLTGYKTN
LENAKE....
PAFVKGGG..

....PRSGPL
TQSGEPD...
ATDPDRDATV
DDQNTRI...
....PKVGPF
GVLSGK....

.LPMSVSGKL
GIPLTAAGKI
.LPTGSSGKL
.LPTLGSGKL
.LPSLASGKL
.VPLNTSGKH

SQSFIHAGGD
ESDFFSSGGN
GDDFFSLGGS
QDNFLGIGSD
.SSFIASGGD
..SFISLGGD

@@@@

GVCVPIDPRY
GVFVLLEPGH
GAFNPVDISQ
GCFVPLDPSY
CVFIPLDPKH
GVCVALDPSH

TKIDDPAFGL
L..NDDLYVV
TDPSSPAYLL
VRPNQAAYIL
..PGDDAYLL
VQPHNAAFVV

SEEDR.MSNL
REEDRNPAGI
SDKER.VNNL
KEEQR.MSNL

DHQRMEDLV
SESERMNHLV

ATRI ..... R
ANQVTCT...
AREISPTAPH
SGELTP....
RYEVGGQSLD
CN.VGV..AN

SVRRGRRFYR
ST ..... LYK
..KKGTRVYR
ENQWSRRAYK
...KDARFYK
..QPGRRMYL

fifiifi

EAR....LIA
...... LLIA
DVQMREPYLA
...... ILGA
..QDRLVAVV
....... IVT

DRAVLQDQLE
DRRHLRKLCE
DHAFVRACLR
DRKTVRRVAG
DRKFLREWLE
DRMKVKNWI N

SITAMQVSSW
SMAAIALRAE
SISAMRLVG.
SIAAIKLVAL
SISAMQLTAR
SLTAMQVVSQ
@@@@@

PVERIRDIIR
PESRLSGIIK
PRSRLQNLIE
PHERLEHIIS
PGGRLQKIVT
PHSRLVEIVK

FTSGSTGVPK
FTSGSTGVPK
YTSGTSGKPK
FTSGSTGKPK
YTSGSTGQPK
FTSGSTGKPK

tiffitiii
QDYMVSVRPN
TTFINRYGVT
QDFTRINDIN
SGWVQKFKVN
.GAIKLLRAN
.GALRSTVPN

KGKSSNIGYG
TTTVSDIGRG
DKPSSNIGHA
SSPANLIGKP
HT...NIGRA
NANPANFGHQ

TGDLVRYCDD
TGDLVRYSAD
TGDLVRYASD
TGDLVPWGPQ
TGDLVRHNSD
SGDLGRYNYD

i...

LLVTGNKDGT
FLFVGRANTG
GLLVLDLVSL
VLGIGDRALG
SLLQTLAEPQ
VLSL.RALDP

SLSPSDYAEI
PFNRNDLISF
EL ........
PL ........
IMITASIRKS
SIEEE .....

MKRFTGKRIG
AQR.SGFTLF
LARSSGHKLQ
LRQ.HGISLA
CA.SKGYRLM
SRR.MGIQVT

Figure 7. Alignment of amino acid sequences of the four conserved domains of C.
carbonum HTS and two conserved domains from D. chlamydosporia. Conserved amino
acids are highlighted; putative adenylate formation, ATP binding and thioester binding motifs
are shown by asterisks and @, respectively. C. carb., C. carbonum; D. ch], D.
chlamydosporia.

27

transformation system for D. chlamydosporia. Three potential markers were tested:
hygromycin B resistance, neomycin resistance and acetamide utilization. D.
chlamydosporia was found to be resistant to Hygromycin. Several transformation vectors
incorporating neomycin resistance and acetamide utilization markers were used but all

attempts at transformation failed (data not shown).

Materials and Methods

Fungal culture growth and maintenance

Fungal strains used in this study are described in Table 1. Conidia of all fungal
strains were stored at -80° C in 25% (v/v) glycerol and grown on V8 juice agar plates.
For DNA extraction from C. carbonum, mycelial agar plugs (0.5 cm2) were inoculated
into l-L Erlenmeyer flasks containing 125 ml modified Fries’ medium (Scheffer and
Ullstrup, 1965). Cultures were incubated at room temperature (21-23° C) without
shaking for four days. For DNA extraction from other species, spores and hyphae were
scraped off the Petri dish and the resulting suspension in 0.1% Tween-20 was used to
inoculate a flask. Cultures were then agitated at room temperature for two days (Cyl.

macrosporum) or four days (P. guttulata, D. chlamydosporia and all Helicoma species).

DNA manipulations

Genomic DNA was isolated from fungal mycelia as described (Pitkin et al., 1996).
Southern blot analysis was performed as described (Sarnbrook et al., 1989).

For PCR, 1 to 3 ug DNA was used as template. The reaction mixture contained

28

0.2 mM of each dNTP, 1 uM of each primer, and 0.5 to 1 units of Taq polymerase
(Promega, Madison, WI, USA) with buffer supplied by the manufacturer, in a total
volume of 100 pl. The reactions were carried out in a DNA Thermal Cycler model 480
(Perkin-Elmer Corp, Norwalk, CT, USA) programmed as follows: 94° C for 4 min; 25
cycles of 94° C for 1 min, 37° C for 2 min, 72° C for 3 min; 72° C for 6 min. Aliquots
from the completed reactions were fractionated on 2% agarose gels. PCR products of the
predicted sizes were isolated and used as templates to reamplify under the reaction
conditions outlined above.

D. chlamydosporia genomic DNA library in phage XEMBL3 was constructed
using the Undigested EMBL3 Cloning Kit (Stratagene, La Jolla, CA) and the Gigapack II
Gold Packaging Extract (Stratagene) according to the manufacturer’s instructions. The
library was screened as described (Sambrook et al., 1989).

The DNA sequence was determined by automated fluorescent sequencing
performed by the MSU-DOE-PRL Plant Biochemistry Facility using an ABI Catalyst 800
(Applied Biosystems Division, Foster City, CA) for Taq cycle sequencing and an ABI

373A Sequencer (Applied Biosystems Division) for analysis of the products.

Toxin extraction and epoxide detection

HC-toxin and other epoxide-containing toxins were purified from 24-day culture
filtrates by solvent extraction as described by Walton et a1. (1982). Fractions were
lyophilized, resuspended in methanol and loaded onto silica gel TLC. TLC plates were
developed in CHzClzzacetone (1:1, [v/v]) and epoxide groups were detected by spraying

plates with 4-(p-nitrobenzyl)-pyridine (p-NBP) (Hammock et al., 1974).

29

Chapter 2

REDUCED VIRULENCE CAUSED BY MEIOTIC INSTABILITY OF THE T 0X2

CHROMOSOME OF COCHLIOBOL US CARBONUM

Abstract

In some HC-toxin-producing (Tox2+) isolates of Cochliobolus carbonum, the
known HC-toxin biosynthetic genes are located on a chromosome of 3.5 Mb, whereas in
other isolates the genes are on a chromosome of 2.2 Mb. Crosses between Tox2+ and
Tox2_ isolates and between isolates in which the TOX2 genes were on chromosomes of
different size yielded progeny that had lost one or more copies of one or more T 0X2
genes. Of 200 progeny analyzed, eight (4%) had lost at least one TOXZ gene. All eight
still had at least one functional copy of HTS], T OXA, T OXC, and TOXE. The deletion
strains were characterized with respect to virulence, HC-toxin production, HTS enzyme
activity, and size of the TOX2 chromosome. Most deletion strains could be explained by
simple chromosome breaks resulting in the loss of major contiguous portions (0.8 Mb to
1.4 Mb) of the. 3.5-Mb T 0X2 chromosome, but at least one strain had apparently
undergone an internal deletion. Most strains were still completely virulent (Tox2+), but
two displayed a novel phenotype of reduced virulence (RV), defined as an ability to cause
small lesions that expanded at a reduced rate and an inability to colonize plants
systemically. Although the RV strains produced no detectable HC-toxin in culture, the
RV phenotype was dependent on the presence of a functional copy of H TS] . We propose

that the RV strains still produce a low level of HC-toxin, at least in planta, and that the

30

RV strains are missing one or more unknown genes that have a role in, but are not

absolutely required for, HC-toxin production.

Introduction

Isolates of C. carbonum that produce the cyclic tetrapeptide known as HC-toxin
are exceptionally virulent on maize of genotype hm] /hm1 . These strains are called Tox2+
or race 1. Race 2 (Tox2_) strains of C. carbonum are non-virulent and do not produce
HC-toxin. HC-toxin production is controlled by a complex locus, T 0X2, that contains
multiple copies of multiple genes (Walton, 1996; Walton et al., 1998). The genes of
TOX2 include HTS], encoding a large multifunctional non-ribosomal peptide synthetase;
T0264, encoding an HC-toxin efflux carrier; T OXC, encoding a fatty acid synthase [3
subunit; and TOXE, encoding a regulatory protein (Panaccione et al., 1992; Scott-Craig et
al., 1992; Pitkin et al., 1996; Ahn and Walton, 1997; Ahn and Walton, 1998). The genes
of TOX2 are absent from toxin non-producing (Tox2_) isolates.

The synthesis of HC-toxin is catalyzed in part by HTS, a non-ribosomal peptide
synthetase encoded by the HTS] gene (Walton and Holden, 1988; Panaccione et al.,
1992). HTS catalyzes the activation of the four amino acids found in HC-toxin and
probably polymerizes and cyclizes the HC-toxin tetrapeptide. HTS also epimerizes L-
Ala to D-Ala and L-Pro to D-Pro (Walton and Holden, 1988). HT S1 is absent from Tox2_
isolates and is, thus, Tox2+ -unique. Disruption of both, but not only one or the other,
copies of HTS] results in strains that do not produce HC-toxin and have lost their

virulence (Panaccione et al., 1992).

Three other Tox2+ -unique genes, T OXA, TOXC and T OXE, have been found to be

31

involved in HC-toxin biosynthesis and efflux (Pitkin et al., 1996; Ahn and Walton, 1997,
Ahn and Walton, 1998). The alkyl side chain of Aeo (2-amino-9,10-epoxy-8-
oxodecanoic acid) may be produced in part by the gene product of the TOXC gene, whose
predicted amino acid sequence is similar to the B subunit of fungal fatty acid synthases
(Ahn and Walton, 1997). Disruption of the TOXC gene results in strains that do not
produce HC-toxin and are non-virulent.

The predicted amino acid sequence of the protein encoded by T 0264 gene is very
hydrophobic and shows amino acid and secondary structure similarities to integral
membrane pumps of small molecules and antibiotics. TOXA probably encodes an HC-
toxin efilux pump (Pitkin et al., 1996). Each of the two copies of T 0XA is tightly linked
to a corresponding copy of HT S1 , forming a TOXA/HT S1 cluster. The transcriptional
start sites of T 0XA and HTS] (in both copy 1 and copy 2 of T 0XA/HTS1) are 386 bp
apart, and these two genes are transcribed in opposite directions (Pitkin et al., 1996).
T0264 may be essential for the protection of the fungus against HC-toxin by transporting
it out of the cell, because it has proven impossible to recover mutants with both copies of
TOXA disrupted (Pitkin et al., 1996).

T OXE, encoding a regulatory protein, is present in two copies in most naturally
occurring Tox2+ isolates. This gene is also essential for HC-toxin biosynthesis, its
disruption turns off transcription of T0264, TOXC and TOXD, another Tox2+ -unique
gene with unknown function (Ahn and Walton, 1998).

In some isolates, including the strain SB] 1 1, all of the TOX2 genes except one
copy of TOXE are located on a chromosome of 3.5 Mb (which is the largest known

chromosome in any isolate of C. carbonum). In SB111, the other copy of TOXE is on a

32

0.7-Mb chromosome. In other Tox2+ isolates, such as CC141, all of the known TOXZ
genes, including both copies of T OXE, are on a chromosome of 2.2 Mb (Ahn and Walton,
1996; Canada and Dunkle, 1997). The 3.5-Mb TOX2 chromosome of SB111 also
contains single-copy DNA that is common to Tox2+ and Tox2_ isolates. In Tox2—
isolates and in Tox2+ isolates such as CC141, this single-copy DNA is on a chromosome
of 2.0 Mb. Together, the data are consistent with the 3.5-Mb and 0.7-Mb chromosomes
of SB111 being related by a reciprocal translocation to the 2.2-Mb and 2.0-Mb
chromosomes of other Tox2+ isolates (Ahn and Walton, 1996).

T 0X2 genes have been physically mapped on the 3.5 Mb T 0x2 chromosome of strain
SB] 11 (Ahn and Walton, 1996) (Figure 8). A chromosome of this size is not found in
Tox2_ strains of C. carbonum, and at least part of it is completely missing from Tox2—
strains (Ahn and Walton, 1996).

Among the progeny from a cross between SB111 and 83114 (a related Tox2—
isolate), one progeny, called 243-7, was found to have a truncated 3.5-Mb chromosome
and to lack at least one copy of each of the known TOX2 genes. This chapter describes
the molecular genetic, biochemical and pathogenic phenotype of 243-7, as well as the
isolation and characterization of other isolates with deletions of the 3.5-Mb TOX2
chromosome.

The truncated TOX2 chromosome of 243-7 has been partially described (Figure 8)
(Ahn and Walton, 1996). By PFGE, the chromosome of 243-7 is 2.1 Mb instead of the
parental 3.5 Mb, and the parental l-Mb PacI fragment on which both copies of HT S] are
located is reduced to 0.28 Mb (Ahn and Walton, 1996). Isolate 243-7 is missing copy 1

of TOXA/HTS] (these two genes are tightly clustered; Pitkin et al., 1996); copy 3 of

33

Ft

 

 

PacIPacI
. 110 kb 60 50 220 kb 30 0 30 20
C3 D3 DI Al/Hl C2 D2A I112 C1
> e 6) <
A2/H2 '//
B. 1.3 Mb 03 1.9 Mb
PacI Al/Hl PacI .

3.5-Mb TOX2 chromosome of C. carbonum SB111

C HTSl-2 1 9Mb
Pacl

2.1-Mb chromosome of C. carbonum 243-7

Figure 8. Physical maps of the TOX2 chromosome in SBlll and 243-7 (after Ahn and
Walton, 1996). (A) Detailed map of the region of the TOX2 chromosome containing the
HTS] genes. Distances are given in kb. A1/I-Il indicates TOXA-1 (TOXA copy 1) and HTS]-
1 (HTSI copy 1). C1, C2 and C3 indicate copies 1, 2 and 3 of TOXC, respectively. D1, D2
and D3 indicate copies 1, 2 and 3 of TOXD, respectively. Arrows indicate the direction of
transcription of the genes. (B) Map of the entire 3.5-Mb HTS] -containing (TOXZ) chromo-
some of the Tox2+ (race 1) strain, SB1 1 1. (C) Deduced map of 243-7, a strain bearing
truncated TOX2 chromosome. Lines connect the corresponding sites in (A) and (B). Dis-
tances are given in Mb. PacI indicates known restriction sites of the rare-cutting restriction
endonuclease Pad.

34

T 0XC, and copies 1 and 3 of TOXD (Figure 8). The evidence is consistent with 243-7
having undergone a simple chromosome break, resulting in the loss of 1.4 Mb of
contiguous DNA extending from just to the right of copy 1 of T0264/HTS1 to the left-
hand end of the chromosome (Figure 8) (Ahn and Walton, 1996). Strain 243-7 still has at
least one flinctional copy of each of the known TOX2 genes, including the regulatory gene

TOXE (Ahn and Walton, 1998).

Results and discussion

Disease phenotype of 243-7
1. Inoculation of 2-week-old seedlings.

Pathogenicity of strain 243-7 was compared to 367-2, a typical Tox2+ isolate, and
243-10, a typical Tox2_ isolate, on maize of genotype hm1/hm] (Multani et al., 1998)
when the plants were 25 cm tall and had four true leaves. No obvious differences in
lesion morphology or size were observed two days post-inoculation, but at four days the
development of lesions on the leaves infected with 243-7 was clearly intermediate to that
caused by 367-2 and 243-10 (Figure 9A). The lesions caused by 243-7 were fewer than
those caused by 367-2 and more variable in size and shape. Isolate 243-10 caused only
small flecks, typical of a resistance response in this disease. At 7 days, the intermediate
disease phenotype of 243-7 was still clear. Plants infected with the Tox2+ isolate 367-2
were completely dead as the fungus colonized the entire plant, including the stalks,
whereas plants infected with the Tox2— isolate 243-10 were completely healthy (Figure

9B). The plants infected with 243-7, on the other hand, showed some lesions of highly

35

2 Days

243-10

4 Days

307—2

243—10

7 Days

: ...., . if? - -’ “was. :» fa.
It 4. .

'7‘... est-1.3' "V's-nos ., ..,
‘3:- r . a; .T_% ;i§..

243—10

 

Figure 9. Strain 243-7 exhibits a new pathogenicity phenotype on maize. Race 1—
susceptible maize leaves were inoculated with strains 367-2 (Tox2", race 1), 243-7 (RV) and
243-10 (Tox2‘, race 2) strains of C. carbonum. (A) Time course of infection. Individual
leaves are shown at two, four and seven days post inoculation. Note that the lesions on the
367-2 (Tox2*, race 1) infected leaf have not spread significantly after four days post inocu—
lation due to the desiccation of the leaf.

36

 

Figure 9 (B). Plants at seven days post inoculation.

variable size and irregular outline but the plants remained alive and continued to grow

Figure 9B). Isolate 243-7 was never observed to invade the stalks of infected plants or to

kill plants. We call this intermediate disease phenotype “RV” for “reduced virulence”.
Inoculation of plants that were 10 cm (two true leaves) or 16 cm tall produced the

same infection phenotype as older plants.

2. Inoculation of very young seedlings.

When seedlings with only the flag leaf and one true leaf were inoculated with the
same three isolates at conidial concentrations of either 104/ml or 5 x 104/ml, all of the
plants survived and grew until 6 days post-inoculation (Figure 10). The plants inoculated
with 367-2 or 243-7 developed lesions only on the tips of the first true leaves, which had
been present at the time of inoculation. Plants inoculated with 2437 had consistently
fewer lesions than those sprayed with 367-2. At 6 or 7 days post-inoculation, plants
infected with 367-2 but not 243-7 began to lose turgor and became grayish in color, and
they collapsed and died 2-3 days later without developing any lesions on the leaves that
had not been directly inoculated. Plants inoculated with 243-10 or 243-7 continued to

grow normally for at least 3 weeks.

3. Inoculation of seeds.

Maize seeds (nine per treatment planted three per 15-cm diameter pot) were
soaked in a suspension of conidia from strains 267-2, 243-10, and 243-7 (104/ml in 50 ml
0.1% Tween-20) for 1.5 hr. Plants in all treatments emerged at the same time. Plants

infected with 367-2 were brown in color from the first day of appearance, grew to a

38

 

367-2 243-7 243-10

Figure 10. Disease phenotype after inoculation of very young seedlings. Seedlings l in.
in height with only the flag leaf and one true leaf were inoculated with strains 367-2 (Tox2*,
race 1), 243-7 (RV) and 243-10 (Tox2', race 2) of C. carbonum at conidial concentration of

104/ml. Plants are shown at 5 and 1 1 days post inoculation. Note a few lesions on the lower
leaves of 243-7-infected plant.

39

height of only 1-2 cm, and died 3-4 days after appearance (Figure 11). Plants from seeds
inoculated with 243-7 or 243-10 had the same appearance as non-inoculated plants and

never developed any visible symptoms.

4. Soil inoculation.

After planting the seeds (three per 15-cm diameter pot), the soil was watered with
a suspension of conidia at concentrations of 104/ml, 5 x 104/ml, or 105/ml in a volume of
10 ml. The pots were covered with plastic bags overnight. Plants inoculated with conidia
of isolate 367-2 at 5 x 104/ml or 105/ml looked healthy for the first 3 days after their
appearance above ground, but then became greyish in color, wilted, and died by day 10
without the development of any visible leaf lesions (Figure 12). Plants inoculated with
367-2 at 104 conidia/ml remained healthy. Plants infected by soil inoculation with
isolates 243-10 or 243-7 showed no symptoms at any concentration at any time point.

In conclusion, the pathogenicity tests indicated that 243-7 has an intermediate
virulence between wild type Tox2+ and Tox2_ isolates. When inoculated directly onto
leaves, isolate 243-7 can cause disease, but the lesions are fewer, more irregular in
outline, and more variable than those caused by Tox2+ isolates (Figure 9 A, B). Isolate
243-7 never caused wilting or death of seedlings, and could not cause any disease when
inoculated onto seeds either directly or through the soil. The most probable explanation
for this is that isolate 243-7 cannot infect stems and therefore cannot infect or spread
systemically. It can cause disease only in leaf tissue that has come into direct contact

with the primary inoculum.

4O

 

Figure 11. Disease phenotype after inoculation of seeds. Plants are shown at 5 days after
emerging (12 days after infected seeds were planted). Note that 367-2-infected plant can be
barely seen because it grew only 1 in. in height before desiccation.

41

 

Day 19

 

 

Figure 12. Disease phenotype after soil inoculation. After planting the seeds, the soil
was watered with a suspension of conidiospores of strains 367-2 (Tox2", race 1), 243-7
(RV) and 243-10 (Tox2', race 2) at a concentration of 5 x 104/ml, in a volume of 10 ml.
Time course of infection is shown (the day of inoculation is the same day the seeds were
planted). Plants emerged at seven days after the seeds were planted and the soil inoculated.

42

Saprophytic growth, maize leaf penetration and toxin phenotype of strain 243-7

Strain 243-7 sporulated normally in culture. It also grew slightly faster on Petri
plates and in race tubes than wild type (wild type strain 367-2 had a linear growth rate of
~ 0.79 cm/day, strain 243-7, ~ 0.84 cm/day) (Figure 13). Therefore, the RV disease
phenotype of 243-7 is not due to reduced growth.

Appressoriurn formation and penetration of the maize leaf by strain 243-7 was
compared to that of the wild type. After inoculation of the susceptible maize plants or
individual excised leaves with strains 367-2 (Tox2+), 243-7 (RV) or 243-10 (Tox23,
leaves were collected at 18, 24 and 48 hr post-inoculation and stained in 0.1% cotton blue
in lactophenol. The number of germinated spores that formed appressoria and penetrated
the intact leaf surface, those that penetrated through stomata and those that failed to
penetrate the leaf were counted under the microscope. At 24 and 48 hr, all three strains
penetrated the leaf at the same rate (results not shown). Therefore, strain 243-7 is not
deficient in appressoriurn formation and leaf penetration.

By analyzing culture filtrates by HPLC followed by TLC for the presence of HC—
toxin, J. Pitkin showed that if 243-7 produces any HC-toxin in culture, it does so at a
level less than 1% of the wild type, which is undetectable by the TLC method (data not

shown).

Disruption of HTSl in isolate 243-7 eliminates the R V phenotype
Two explanations for the RV phenotype seemed possible. One, supported by the
in vitro HC-toxin analysis, was that strain 243-7 had completely lost the ability to

produce HC-toxin and its residual virulence was due to a different factor that is

43

 

growth (cm)

r_____
+SB111
2 -- . .—0—243—7

 

 

 

 

Figure 13. Growth of strain 243-7 compared to $8111. SB1 11 (Tox2+, racel, virulent)
and 243-7 (RV) strains were grown in horizontal race tubes containing potato dextrose agar.
Data points are averages of duplicate samples.

44

genetically linked to the T 0X2 locus. An alternative hypothesis was that 243-7 produces
a very low level of HC-toxin, at least in planta, that is sufficient to allow some
colonization of the plant. To test the involvement of HC-toxin in the RV phenotype, the
remaining copy of HT S1 (copy 2, see Figure 8, p. 34) in strain 243-7 was disrupted by
transformation-mediated homologous integration, creating an HT S1 -null strain.

Plasmid pCCl 19, which contains a 0.6-kb fragment from the 5’ end of HT S1 and
the hph gene for hygromycin resistance (Panaccione et al., 1992), was linearized at a
unique Xhol restriction site and transformed into C. carbonum strain 243-7. Five
hygromycin-resistant transformants were analyzed. The restriction map of the wild-type
HTS] locus (3’ end) and the predicted map resulting from integration of a single or
multiple copies of pCC119 are shown in Figure 14A, B and C, respectively. Southern
blot analysis (Figure 14D) indicates that the wild-type 17.1-kb band disappears in the
disruption mutants, and the new pattern of hybridization (12.5-, 10.6- and 6.0-kb bands)
corresponds to the integration of pCCl 19 at HT S] in single (560-2) or multiple (560-1, 3,
S, and 6) copies (Figure 14D).

Pathogenicity of the transformants was compared to that of 367-2 (wild type
Tox2+), 243-10 (wild type Tox2? and 243-7 (RV). All five transformants were
completely avirulent, that is, their disease phenotype was indistinguishable from that of
the wild type Tox2: as shown in Figure 15 for transformant 560-2. Thus, the residual
virulence of 243-7 is dependent on HT S] . The simplest explanation for this result is that
243-7 still produces a low level of HC-toxin, at least in planta, and that this low level of

HC-toxin production accounts for its residual virulence.

45

12 0.6 4.5

r
4
1
4
v

u:
><
{'11
___m

 

 

 

 

 

 

 

 

 

 

 

 

 

B.
A 12.5 -1 10.6 _
S B ES B E S
I l
X X
C.
# 12.5 -1 6 A- 10.6 A
S B E S B ES B E S
l H J
I
x x x
disruption mutants
1) I '7 s3 "9 "a ll?
s s s s g
In In In In
kb kb
. -17.l
1" II we =13?
8.4- '

Figure 14. Analysis of HTS] disruption mutants. (A) Restriction map of wild type
HTS] copy 1 locus. (B) and (C) Predicted restriction maps of HTS] with insertion of a
single (B) and double (C) copies of pCCl l9. Predicted fragment sizes are indicated in
kilobases. (D) Southern blot of genomic DNA from 243-7 and HTS] disruption mutants
560-1, 560-2, 560-3 and 560-5. DNA was cut with Soil and the filter was probed with the
insert of pCCl 19. The black boxes indicate the genomic copy of the BamHI/EcoRI frag-
ment used to construct pCCl l9, and the unshaded boxes indicate pCCl l9 bearing the same
fragment indicated by the shaded boxes. B, BamHI: E, EcoRI; S, SalI; X, XbaI.

46

 

367-2 243-7

 

Figure 15. Disruption of HTS] in 243-7 makes this strain avirulent (Tox2'). Race 1-
susceptible maize plants were inoculated with strains 367-2, 243-7, 243- 10 and 560-2 (HTS 1
disruption mutant). Five plants per strain were inoculated, and from these five plants, 10
leaves were collected at 7 days post inoculation.

Isolation and initial characterization of additional strains with chromosomal
aberrations

The molecular evidence suggests that the 1.4-Mb of DNA that is deleted from
243-7 contains genetic material that is necessary for wild type rates of synthesis of HC-
toxin and hence for full virulence. In order to obtain an indication of the rarity of the
chromosome break event that had occurred in 243-7, i.e., of the instability of the T OX2
chromosome, we examined progeny from additional crosses for deletions within the
TOX2 chromosome. Another rationale for searching for new isolates with deletions in the
T 0X2 locus is that the analysis of such strains could help determine the presence and
location of additional genes necessary for HC-toxin production or virulence. For these
crosses (Figure 16), parental strains were chosen that differed from each other with
respect to the T 0X2 chromosome, either in presence or size, on the presupposition that
this would promote abnormal behavior of the T 0X2 chromosome during meiosis. The
parents in cross 373 were Tox2+ isolates with TOXZ chromosomes of different size (3.5
and 2.5 Mb), and the parents in cross 512 were a Tox2+ isolate with a 3.5 Mb T 0X2
chromosome and a Tox2— isolate lacking any chromosome of 3.5-Mb and all of the
known TOX2 genes (Ahn and Walton, 1996; Canada and Dunkle, 1997).

Thirty-four progeny were screened in cross 373 and 110 in cross 512. Progeny
were initially screened by DNA blot analysis using TOXC as a probe (a subset is shown in
Figure 17A), because copies 1 and 3 of T OXC form the boundaries of the mapped ~540
kb region of the TOX2 chromosome (Figures 8, 21). Strains lacking one or more copies
of T OXC were analyzed further by Southern blotting for TOM and TOXD RFLPs (Figure

17B,C).

48

Crosses A, 243:

Cross 3 73:

Cross 512:

Cross 625:

$13111 (Tox2+, 3.5 Mb) X SB114 (TOXD

l

164R10 (Tox2+, race 1)
243-7 (RV)

$13111 (Tox2+, 3.5 Mb) x CC141 (Tox2+, 2.5 Mb)
373-39 (RV)
367-2 (Tox2+, 3.5 Mb) x 24310 (Tox2_)

512-3, -8, -98, 48 (Tox2+)
512-24 (slow grower, outcrossed to produce 643-9, Tox2+)
512-35 (slow grower, not analyzed)

243-1 (Tox2+, 3.5 Mb) x 2437 (RV, 2.1 Mb)

625-66 (Tox2+)

Figure 16. Crosses to generate deletion strains. Crosses were made between Tox2+
and Tox2_ C. carbonum strains and between Tox2+ strains with different TOX2
chromosome sizes. The size of the TOX2 chromosome in the Tox2+ parents is given in
parentheses. The relevant deletion-bearing progeny from the crosses are indicated below
the arrows with the race phenotype indicated in parentheses. Cross A was performed by
S. Briggs (Walton, 1987). Strains 243-1, 243-7 and 243-10 are progeny of a cross
between SB1 11 (Tox2+; ATCC 90305) and SB114 (Tox2-; Walton 1987). Strain 367-2
was derived from a progeny (243-1) of the same cross by backcrossing it three times to
SB1 11 (J. Pitkin, unpublished).

49

A.TOXC
12347891011

 

B.TOXA
1 2 3 4 56 7891011

 

C.TOXD
1234567891011
- e' a»

ovzwwsv-ua

50

  

fl <—copy 1

‘ «copy 2
Q «copy 3

<—copy 1

«copy 2

Figure 17. Southern blot analysis of the strains used in this study. Lanes: 1, 243-7; 2,
373-39; 3, 512-3; 4, 512-48; 5, 625-66; 6, 512-8; 7, 643—9; 8, 164R10; 9, 367-2A; 10, 512-
98; 11, 243-10a. (A) TOXC RFLP. Genomic DNA was cut with Xhol and the filter was
probed with TOXC internal fragment. TOXC copy I restriction fragment is 18 kb, copy 2,
14 kb; copy 3, 12 kb. (B) TOXA RFLP. Genomic DNA was cut with BglI and probed with
TOXA internal fragment. TOXA copy I restriction fragment is 15 kb, copy 2 restriction
fragment is 12 kb. (C) TOXD RFLP. Genomic DNA was cut with BamHI and probed with
TOXD internal fragment. TOXD copy 1 restriction fragment is 15 kb; copy 2, 9 kb; copy 3,

progeny (512-24, 512-35, 512-48) were found to lack copy 3 of T OXC. However, eight
additional progeny from cross 512 had an extremely faint TOXC-3 band on Southern
blots (data not shown), suggesting that they were not homogenous (either heterokaryons
or contaminated). An attempt was made to purify these deletion strains away from the
contaminating karyotype by two rounds of single conidiospore isolation, but all resulting
single spore isolates had either the wild type RFLP pattern (harboring all three TOXC
copies) or presented the same RFLP pattern as the original isolate (very faint TOXC-3
band). However, some of these isolates formed sectors frequently when grown on a Petri
plate, and these sectors were also examined for T OXC RFLP pattern after single spore
isolation. Through this process, three of the original isolates gave rise to nuclearly
homogenous strains bearing deletions (512-3, 512-8, 512-98).

In summary, two isolates (373-39 and 512-24) were found that lack copy 3 of
T OXC, copies 1 and 3 of TOXD, and copy 1 of T OM/HTSI (Table 2). Thus, these
strains resemble 243-7 in their TOX2 haplotype. All other strains that lacked copy 3 of
TOXC (512-3, 512-8, 512-35, 512-48, and 512-98) still had copy 1 of TOM/HTS].
Strain 512-48 lacks copies 1 and 3 of TOXD in addition to copy 3 of TOXC. TOXE is
present in one copy in all the strains tested (164R10, 512-3, 512-48, 243-7, 373-39, 643-
9), while wild-type strains SB111 and CC141 have two copies of this gene (Figure 18).

Growth tests measuring either radial growth on Petri plates or linear growth in
race tubes of various isolates in parallel indicated that the growth rates of 367-2, 243—7,
243-10, 373-39, 512-48, and 512-8 were statistically indistinguishable (data not shown).

Isolates 512-24 and 512-35, on the other hand, grew slowly in culture. To test

whether this slow growth was related to the loss of DNA from the TOX2 chromosome,

51

Table 2. Summary of data on C. carbonum strains used in this study.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Strain Genes Missingb Phenotypec HTS HTS HC- Pacl TOX2
activityd proteinc toxinf fragment chromosome
size (kfi size (Mb)h
S81 11 all present V + + + 1000 3.5
367-2
CC 1 41 all present V + + + ND 2.2
243-10 none present AV —— — - _ _
243-7 H1, A1, C3, RV + + — 280 2.1
D1, D3
560-2‘ H1, H2::hyg. AV — - — ND‘ ND
A1, C3, D1, D3
373-39 H1, A1, C3, RV + + - 190 ND
D1, D3
643-9 H1, A1, C3, V + + ND 290 ND
D1, D3
164R10 H1,A1,C3, V + + + 900 3
D1, D3
512-48 C3, D1, D3 V + + + 400 2.7
625-66 C3, D1, D3 V + + ND 490 ND
512-3 C3 V + + + 450 2.3
512-8 C3 V + + ND ND ND
512-98 C3 V + + ND ND ND
512-35 C3 ND, slow ND ND ND ND ND
grower
512-98D A1, A2, H1, AV ND ND ND ND ND
H2, C3, C1,
D1, D2, D3

 

 

 

 

 

 

 

 

 

a243-7 with remaining copy of HTS] mutated by targeted gene disruption
bH = HTS]; A= TOXA; c = TOXC; D=TOXD. Numbers indicate the particular copy (see

Figures 8, 21). Data on TOXE (see Figure 18) are not included in this Table.

cV, virulent; RV, reduced virulence; AV, avirulent (see Figure 9 A,B).
dHC-toxin synthetase activity (HTS) measured by D-Ala-depedent ATP/PP, exchange
(see Figure 25 and Table 3)
eHTS protein detected by irnmunoblotting (see Figure 24)
fHC-toxin production in culture, analyzed by TLC (J. Pitkin, unpublished data).

gsize of PacI fragment hybridizing to HT S] as measured by pulsed-field gel

electrophoresis (see Figure 20)
hsize of chromosome containing all copies of HT S1 , TOM, TOXC, and TOXD (J. Ahn,
unpublished)

'ND - not determined

52

 

 

Figure 18. Southern blot analysis of TOXE RFLP in selected deletion strains.
Genomic DNA was cut with BamHI and probed with TOXE internal fragment. Lanes: 1,
CC141; 2, 164R10; 3, 243-7; 4, 373-39; 5, 512-3; 6, 512-48; 7, 643-9.

53

512-24 was outcrossed to strain 243-6 (Tox2+, normal growth). Ten progeny were
isolated and their growth rates were assessed. Five of them grew at the same rate as the
wild type. These five were analyzed by Southern blotting and two (643-8 and 643-9) had
the same genetic composition as the parental strain 512-24. Therefore, in these two
isolates the chromosomal aberration segregated away from the slow-grth phenotype.

Slow-growing strain 512-35 was not outcrossed and not analyzed further, because
it belonged to the same deletion class as strains 512-3, 512-8 and 512-98 based on RFLP
analysis.

An additional strain, 512-98D, was isolated from a sector (derivative of strain
512-98). It lacked all the Tox2 genes except TOXC-1 (Figure 19).

The isolates with deletions of the TOX2 chromosome, including 643-9, were
analyzed further. Isolate 164R10, which, like 243-7, is a progeny of S81 11 and S81 14
(Figure 16) was also included. Isolate 164R10 has been used in other experiments from
this laboratory (Ahn and Walton, 1997); it lacks copy 3 of T OXC, copies 1 and 3 of
TOXD, and copy 1 of T OXA/HT S1 , but has a wild type Tox2+ virulence phenotype and
produces HC—toxin in culture.

To better estimate the sizes of deletions on T 0X2 chromosome, some of the
strains were analyzed by pulsed-field gel electrophoresis after digesting chromosomal
DNA with the rare-cutting restriction endonuclease PacI (Figure 20). Available markers
allow the determination of the size of only one relevant PacI fragment, the one containing
both copies of HT S] (Figures 8, p. 34 and 21, p. 58). Additional experiments performed
by J. Ahn (unpublished data) determined the T OX2 chromosome sizes for some of the

strains (summarized in Table 2).

54

A. TOXC B. TOXA C. TOXD

l 2 3 1 2 3 l 2 3
C 1 , an - Copy 1
w . - 0py .h- Copy 1
.. ' C01” 2 '. i . '0' ' C0” 2 ' - Copy 2
_ ' - ’ - Copy 3

Figure 19. Southern blot analysis of strain 512-98D. (A) TOXC RFLP. Genomic DNA
was cut with Xhol and the filter was probed with a TOX C internal fragment. Lanes: 1 and 2,
two different single-spore isolates of strain 512-98D; 3, strain 243-7 (harboring TOX C cop-
ies l and 2). (B) TOXA RFLP. Genomic DNA was cut with BglI and probed with TOXA
internal fragment. Lanes: 1 and 2, two different single spore isolates of strain 512-98D; 3,
wild type Tox2+ strain 367-2. (C) TOXD RFLP. Genomic DNA was cut with BamHI and
probed with TOXD internal fragment. Lanes: 1 and 2, two different single spore isolates of
strain 512-98D; 3, wild type Tox2+ strain 367-2.

55

12 345
Mb

1.0- *1

 

Figure 20. Physical mapping of TOX2 chromosomes in representatives of wild type
and deletion strains. DNA was cut with PacI, separated by pulsed field gel electrophore-
sis, blotted, and probed with a fragment of the HTS] gene. The locations of relevant Pacl
sites are shown in Figure 1. Lane 1, 373-39; lane 2, 243-7; lane 3, 367-2; lane 4, 635-66;
lane 5, 643-9.

56

Maps of T OX2 chromosomal deletions were constructed based on all available
data. They are shown in Figure 21. In some of the strains, chromosomal deletions may
be internal and in others, terminal; and some strains may have translocations in addition
to deletions. Strain 164R10 is an example of a putative translocation or large internal
deletion because it has a 3.0-Mb TOX2 chromosome and a PacI digestion fragment of
900 kb while missing all the known TOX2 genes to the lefi of TOXC-2. Another factor
contributing to the imprecision of the maps is that in the cases of simple truncations,
telomeres must be added to the end of the chromosome after the break occurs. Lack of
available markers does not allow us to tell how much repair DNA might have been added
and how much of the DNA between the markers is native DNA (Figure 21). The fact that
in strains 512-48 and 625-66, the PacI restriction fragments are larger than predicted

based on the genes missing in these strains, may be due to one or both of these factors.

Co-segregation of the R V phenotype with the chromosomal aberration

To test whether the RV phenotype is linked to the chromosomal deletion in strain
243-7, it was crossed to the wild type Tox2+ isolate 243-1, which is a sibling of 243-7
from cross 243. Fifty-three progeny of this cross were isolated and analyzed. Twenty-
seven were found to have three copies of TOXC and be fully virulent, thus resembling the
wild type parent, 243-1. Another 25 progeny were missing copy 3 of T OXC and had an
RV disease phenotype, thus resembling the RV parent, 243-7. One progeny (625-66) was
missing copy 3 of T OXC but had a fully virulent (Tox2+) disease phenotype. It was
subsequently found to have two copies of T OM/HT SI and one copy (copy 2) of T OXD

(Figure 17). Thus, it had a genetic composition different from both parents and had

57

 

 

1.3 Mb 0.3 Mb 1.9 Mb
1- 9k 4 al
1, 1.0 Mb ,1
1‘ 1:1
Pacl
C3 41/111 42/112“
B.
20 Kb ’acIC Pad
(2 113 D1 Al/Hl H (:2 112421112
110 so so 220 30 20 30 20

243-7, 373-39 (RV)

 

--1

164R10, 643-9 (Tox2+)

 

 

W

512-48, 625—66 (Tox2+)

 

——1
.—

 

512-3, 512-8, 512-9s (Tox2+)

 

_—
.—
1
4
\

 

512-981) (Tox2‘)
1 r: . _

 

Figure 21. Physical maps of the TOX2 chromosome. (A) Map of the entire 3.5 megabase
TOX2 chromosome of the wild type Tox2+ strain, SB1 11 (after Ahn and Walton, 1996).
Distances are given in megabases. PacI indicates known PacI restriction sites. (B) De-
tailed map of the region of the TOX2 chromosome containing TOX2 genes (after Ahn and
Walton, 1996, with the addition of TOXE). Distances are given in kilobases. Al/Hl indi-
cates TOXA-1 (TOXA c0py 1) and HTSI-I (HTSI copy 1). C1, C2 and C3 are copies 1, 2
and 3 of T OXC, respectively. D1, D2 and D3 are copies 1, 2 and 3 of TOXD. E indicates
TOXE copy 1. Lines connect the corresponding sites in (A) and (B). (C) Deduced maps of
deletions in strains bearing TOX2 chromosome aberrations in relation to the defined TOX2
region (~540 Kb from C3 to C 1) of SB1 l 1. Data were derived from Figures 17 - 20. The
maps are not meant to imply anything about the presence or absence of DNA outside the
region shown.

-= region of the chromosome remaining

1:]: region of the chromosome missing
E: region of the chromosome where the break point maps

58

apparently arisen as a de novo aberration during the cross. It belongs to the same class as
512-48 (Figure 21). Two versions of the TOX2 chromosome - wild type 3.5-Mb
(presumed when all three copies of TOXC were present) and truncated 2.1-Mb (presumed
when copy 3 of TOXC was missing) segregated as expected for homologous
chromosomes, at a ratio not statistically different from 1:1 (x2 = 0.077; df=1; 0.70< P<
0.90).

We also tested if any of the progeny from cross 373 (Figure 16) had an RV
phenotype despite not having T 0X2 deletions detectable by RFLP. All 34 progeny from
this cross were screened for pathogenicity. All except 373-39 were firlly virulent
(Tox2+). Strain 373-39 had an RV phenotype. These results show that the RV
phenotype cosegregates with deletion of the chromosomal region containing copy 1 of
TOM/HTS], copy 3 of TOXC, and copies 1 and 3 of TOXD, and is therefore presumed to

be due to the deletion.

Discussion of TOX2 chromosome segregation and instability

The TOX2 chromosome of C. carbonum is a large supemumerary chromosome,
and at least part of it is conditionally dispensable. Supemumerary chromosomes in
filamentous fungi are extra chromosomes found only in some isolates of a given species
and are composed primarily of DNA not found in all representatives of this species
(Covert, 1998). Sometimes they carry genes encoding detectable phenotypes. In Nectria
haematococca, the PDAI-I chromosome carries a number of genes contributing to high
virulence on garden pea: the PDA] -1 gene (Kistler and Van Etten, 1984a, 1984b; Miao et

al., 1991b; Wasmann and Van Etten, 1996), and the PEP genes (Van Etten et al., 1997;

59

Covert, 1998). The MAKIPDA6-1 chromosome of N. haematococca carries MAK], a
gene associated with high virulence on chickpea (Miao and Van Etten, 1992; Covert et
al., 1996; Enkerli et al., 1998), and PDA6-1, one of the pisatin detoxification genes (Miao
et al., 1991a). The TOX2 chromosome of C. carbonum carries genes conferring high
virulence on susceptible maize, an adaptive trait important in some, but not all, grth
conditions. By definition, supemumerary chromosomes do not carry any essential genes,
and the TOX2 chromosome is not known to carry any. It does, however, contain some
DNA common to Tox2+ and Tox2_ isolates (e.g., marker G242 [Ahn and Walton, 1996];
marker Q [Canada and Dunkle, 1997], EXGZ [J . Ahn, unpublished data]). The 3.5-Mb
T 0X2 chromosome may be a product of a translocation between a 2.2-Mb T 0X2
chromosome found in some wild-type isolates and a 2.0-Mb essential chromosome (Ahn
and Walton, 1996). The proposed translocation pattern is shown in Figure 22. It implies
that ca. 1.6 Mb of the TOX2 chromosome containing the TOX2 locus and everything to
the left of it (i.e., conditionally dispensable “TOX2 module”) may be joined to either a
1.9- or a 0.6-Mb chromosomal segment and thus form the 3.5- or 2.2-Mb TOX2
chromosome. The chromosomal 1.9-Mb segment contains markers G242, EXGZ (Ahn
and Walton, 1996 and unpublished data) and Q (Canada and Dunkle, 1997); and the 0.6-
Mb segment contains marker CC62 (Ahn and Walton, 1996). A reciprocal event joins a
putative 0.1-Mb segment to either the 0.6- or 1.9-Mb segment and results in
chromosomes of 0.7 or 2.0 Mb. The chromosomal segments of 1.9, 0.6 and 0.1 Mb may
or may not be dispensable and may carry essential genes. In the latter case, one or more
of them must be present in the genome either as part of the 3.5- or 2.2-Mb TOX2

chromosomes or as part of the 2.0- or 0.7-Mb chromosomes. This could be tested by

60

 

 

 

 

 

 

 

0.6 0 1
TOXE-7
CC62
(1 COPY)
1: 2 2 :1 ‘-= 2 0 “
CC141 1 6 0 6 1 9 0 1
(Tox2) " ' 3
L51 TOXE-2 0242
T02(2 CC62 EXGZ
genes (I copy) Q
4 2.0 =
Tox2' —)
( ) 6242
EXGZ
Q

   

mg?

Figure 22. Proposed translocation involving two naturally occurring variants of TOX2
chromosome. (A) Karyotypes of naturally occurring isolates. Known markers are indi-
cated below the corresponding chromosomal segments. Maps are derived from all available
data (Ahn and Walton, 1996; Canada and Dunkle, 1997; this work). (B) Cruciforrn struc-
ture that can form during meiosis I in the cross between SB1 11 and CC141 (cross 373).
Two outcomes are possible depending on the location of the centromeres, indicated by blue
ovals. Chromosome sizes and distances are given in Mb.

61

screening T 0X2_ cross progeny for the presence or absence of these chromosomes by
CHEF or by screening for corresponding markers on Southern blots. Markers available
on the 0.6-Mb segment are one of the copies of an internal fragment of CC62 (or A2B3)
and T 0XE-2. Based on the data from Southern blots (Figure 18 and DNA blots probed
with A2B3, results not shown), these markers are missing in some strains (164R10, 243-
7, 373-39, 512-3, 512-48, 643-9) and in independent Tox2- isolates (e.g., SB114). Thus,
the available data suggests that the 0.6-Mb segment is most likely dispensable.

All these data put together suggest an intriguing hypOthesis. The original T 0X2
chromosome may have been the 2.2-Mb chromosome (1.6-Mb T 0X2 module plus 0.6-
Mb segrnent). It contained all of the genes essential for HC-toxin biosynthesis and no
genes essential for growth, and was dispensable. It may have moved from another
species and may have contained genes that were essential for its organism of origin but
that had counterparts in C. carbonum and thus became dispensable. It is not clear,
though, what could have been the species of origin and why it might have produced HC-
toxin.

The T 0X2 chromosome contains much repetitive DNA (Ahn and Walton, 1996).
Some of the repetitive sequences are Tox2+-unique (e.g., CC115 transposase-like
sequence, see Appendix B), while others are also found elsewhere in the genome (e.g.,
Fccl transposon, Panaccione et al., 1996). Similarly, in two other well-studied species,
N. haematococca and Colletotrichum gloeosporioides, moderately and highly repeated
DNA sequences are found on the supemumerary chromosomes, and some of these
sequences are chromosome-specific, while others are shared with other chromosomes

(Enkerli et al., 1997; Masel et al., 1993; 1996). This fact suggests that the T OX2

62

chromosome and the supemumerary chromosomes from N. haematococca and C.
gloeosporioia’es have a different evolutionary history relative to the rest of the genome
and lends further evidence to the horizontal transfer hypothesis.

The mechanism of the proposed horizontal chromosomal transfer remains to be
elucidated, but there is evidence that in C. gloeosporioides, the supemumerary
chromosome is transferred between asexual isolates at a frequency of 10'7 (Masel et al.,
1996; Covert, 1998). It was suggested that such transfer may be analogous to the
conjugative transfer of plasmids in bacteria (Covert, 1998).

Unlike other supemumerary chromosomes from sexually reproducing fungal
species, for which the information is available (Miao et al., 1991a; Orbach et al., 1996;
Xu and Leslie, 1996; Tzeng et al., 1992), the T 0X2 chromosome of C. carbonum does
not exhibit non-Mendelian segregation ratios. In crosses between Tox2+ and Tox2"
isolates, a segregation ratio of 1:1 was first demonstrated by Nelson and Ullstrup (1961).
To confirm this in light of the data on the non-Mendelian segregation of the
supemumerary chromosomes in other fungal species, the segregation ratio in cross 512
with respect to the presence or absence of the T 0262 chromosome was determined and
compared with the Mendelian ratio of 1:1 which was expected based on earlier data
(Nelson and Ullstrup, 1961). For this purpose, we can assume that the TOX2
chromosome is missing in those progeny in which none of the copies of TOXC are
present (although it is possible that in some of them, the whole TOX2 locus was deleted
with the rest of the chromosome remaining). In 59 progeny, the TOX2 chromosome was
present, and in 51 it was missing, which is not statistically different from the expected

Mendelian ratio of 1:1 (x2 = 0.582; df—=1; 0.3< P< 0.5). In a cross between a wild type

63

Tox2+ strain and an RV strain (243-7) bearing a truncated but presumably otherwise
unaltered T OX2 chromosome (cross 625), the wild-type and truncated chromosomes (as
detected by Southern blot T OXC RFLP patterns) also segregated at a Mendelian ratio of
1:1, as expected for homologous chromosomes. In a cross between two wild type
parental strains with different naturally-occurring variants (3.5- and 2.2-Mb) of the T 0X2
chromosome (cross 373), none of the progeny lost the chromosome, as determined by
hybridization to TOXC probe. If the 3.5-Mb and 2.2-Mb T OX2 chromosomes segregate
independently of each other and also independently of the 2.0- and 0.7-Mb chromosomes
with which they have sequences in common (Ahn and Walton, 1996; Canada and Dunkle,
1997), 25% of the progeny would be expected to lose the T 0X2 locus altogether, and
another 25% to harbor both the 3.5-Mb and 2.2-Mb T 0X2 chromosomes. Although we
did not determine by CHEF chromosome separation whether these two variants of T 0X2
chromosome segregated at a 1:1 ratio, the absence of Tox2—progeny suggests that, despite
the fact that these two naturally occurring TOX2 chromosome variants are homologous
only in part, they pair up during meiosis as opposed to assorting independently. A similar
conclusion was reached by Canada and Dunkle (1997). Furthermore, seven progeny of
cross 373 were analyzed by PF GE (J. Ahn, unpublished data), and 11 progeny of a similar
cross were analyzed by Canada and Dunkle (1997), and none were found to harbor two
TOX2 chromosomes.

However, we have to consider the possibility that one or both of the non-parental
progeny classes (harboring none or both T 0X2 chromosomes) are inviable. We assume
that duplications of the chromosomal segments within the C. carbonum genome are not

lethal, similar to what is known for some other fungi (e.g., Perkins, 1974; Tzeng et al.,

64

1992). However, both of these classes may be missing some of the chromosomal
segments (the progeny containing both TOX2 chromosomes may be missing the 0.1-Mb
“tip” segment located on the other chromosomes involved in translocation).
In the case of the two TOX2 chromosomes segregating independently of each other, there
are three possible segregation patterns for the other chromosomes involved in the
translocation. These possibilities are:

1. All four chromosomes segregate independently of each other. In this case, %
of all progeny will have both TOX2 chromosomes and ‘A will have none. In each of these
classes, % will have all three of the other chromosomal segments present on the non-
TOX2 chromosomes. Thus, 1/ 16 of all progeny will be missing the TOX2 locus but be
viable and detectable by Southern blots. Of the class with two TOX2 chromosomes, %
will have all chromosomal segments (some in more than two copies) on different
chromosomes, and so presumably will be viable. Because neither of these are found
among the progeny, we may conclude that this scenario is not possible.

2. The 3.5-Mb chromosome pairs with the 2.0-Mb chromosome, and the 2.2-Mb
chromosome pairs with the 0.7-Mb chromosome (using the 0.6-Mb segment for a
homologous region). Because the progeny classes that lack both of the TOX2
chromosomes and that harbor both of the TOX2 chromosomes will be missing the “tip”
segment in addition to the TOX2 module, and this can be lethal, this scenario cannot be
excluded.

3. Only one or the other of the above pairs form, and the remaining two
chromosomes segregate independently. In this case, ‘A of all the progeny will have both

TOX2 chromosomes, and ‘/4 will have none. In each of these classes, 1/2 will have all

65

three of the other chromosomal segments present and so presumably be viable. Thus, 1/8
of all the progeny will be missing the TOX2 locus but be viable and detectable by
Southern blots. Because this is not the case, we may conclude that this scenario is not
possible.

The possible (# 2) scenario could be tested by using each of the available probes
(T 0X2 genes, G242, EXGZ and CC62). Assuming the progeny classes missing both
T 0X2 chromosomes or harboring both T 0X2 chromosomes are inviable, 1/2 of the
remaining progeny will contain all segments (having the 2.0- and 2.2-Mb chromosomes
present); another 1/2 will lack the 1.9-Mb segment (and thus will lack G242 and EXGZ
markers) while containing the 3.5-Mb T 0X2 chromosome (this class may also be
inviable). It would be possible to distinguish between this outcome and the scenario
where the two T 0X2 chromosomes act as homologs (with the other pair acting as
homologs or independently).

Finally, if these four chromosomes form a cruciform pattern typical for balanced
translocations in meiosis, the results would be indistinguishable from those with two
independent homologous pairs. The actual outcome will depend on the centromere
position. If centromeres are on the TOX2 module and on the “tip”, the resulting progeny
will be the same as in the case when two TOX2 chromosomes (3.5- and 2.2-Mb) form a
pair and two other chromosomes (2.0- and 0.7-Mb) form a pair. Alternatively, if the
centromeres are on the 1.9- and 0.6-Mb segments, the resulting progeny will be the same
as in the case when the 3.5-Mb chromosome pairs with 2.0-Mb chromosome (using the
1.9-Mb segment as a homologous region), and 2.2-Mb chromosome pairs with 0.7-Mb

chromosome (using 0.6-Mb segment for a homologous region).

66

Still, it is unlikely that both the progeny class lacking the T 0X2 chromosome
altogether and the one containing both T OX2 chromosomes are inviable (because scenario
#2 requires that the “tip” segment is indispensable), and so the 2.2- and 3.5-Mb T 0X2
chromosomes probably behave as homologs. This, in turn, would mean that the
centromere is located on the 1.6-Mb TOX2 module, and the one in the second pair of
homologs is on the 0.1-Mb “tip” (first option in Figure 22B, p. 60, as opposed to the
second option). On the other hand, from the data collected so far it appears that breaks in
the TOX2 chromosome are random, but usually to the left of T 0XC-2. Since at least
some of the strains (e.g., 243-7) appear to be missing the entire left segment of the T OX2
chromosome (1.4 Mb in case of 243-7, with a breakpoint between HTS] -l and HT S1 -2 at
~20 kb from HT S1 -1), the centromere must lie to the right of HT S1 -1, and probably to the
right of HT S1 -2 (since the T OX2 chromosome size in strain 512-98D was not determined,
we cannot be sure it has a simple truncation). Assuming that the ca. 1.3-Mb segment of
the TOX2 chromosome to the left of and including HT SI-l is dispensable, the centromere
is between HTSI-l and TOXC-1. Further, if the segment between Al/Hl and D2 is
dispensable, the centromere is very close to TOXC-1. This region is also the location of
DNA common to Tox2+ and Tox— isolates (Ahn and Walton, 1996).

We found that in some progeny truncated versions of this chromosome or other
rearrangement patterns arose. In this work, a total of ~200 progeny from three crosses
were analyzed by Southern blotting, and of these eight (4.0%) had deletions of one or
more genes within the ~540 kb region of T 0X2. Adding previously analyzed progeny
from crosses A and 243 (Figure 16), out of a total of ~220 progeny analyzed, 10 (4.5%)

had deletions within the TOX2 locus. Pulsed-field gel electophoresis analysis of T 0X2

67

chromosomes in these strains shows that major chromosomal rearrangements
(truncations, translocations or major deletions) had taken place. This form of meiotic
instability of supemumerary chromosomes has been also detected in Nectria
haematococca and Gibberellafujikuroi (Miao et al., 1991a; Xu and Leslie, 1996). It was
noted (Covert, 1998) that it is possible that such rearrangements are detected only in
supemumerary chromosomes and not in essential chromosomes simply because the
aberrations in the essential chromosomes are lethal. Interestingly, the data for yeast
show that about 7% of haploid meioses (or 3.5% of all haploid progeny) produced
chromosomes that differed by more than 10 kb from their wild-type counterparts (Loidl
and Nairz, 1997).

One possible mechanism that can generate these chromosomal aberrations is via
nonallelic recombination facilitated by repetitive sequences. This mechanism is well
studied in yeast (e.g., Jinks-Robertson and Petes, 1986; Loidl and Nairz, 1997). Such
recombination may occur between homologous as well as nonhomologous chromosomes
and may produce major deletions and translocations.

In our experiments, some of the progeny in cross 512 showed a pattern of
hybridization with the TOXC probe (two strong bands, one extremely faint band)
suggesting nuclear non-homogeneity. Since in some cases this pattern persisted through
several rounds of single-spore re-isolation, it may be due to the existence of
heterokaryons among the progeny. The observation that such strains sometimes formed
sectors when grown on plates and some of these sectors gave rise to pure isolates with
deletions or complete loss of the T OX2 chromosome firrther suggests that these

heterokaryons can spontaneously resolve. The existence of heterokaryons among single

68

conidiospore isolates of C. carbonum recently also was suggested by observations on C.
carbonum strains with disruptions of cell wall degrading enzyme genes (J. Scott-Craig,
unpublished data). One possible way for the heterokaryons observed in this work to arise
is that the TOX2 chromosome rearrangement occurs in the early mitotic divisions of a
germinating ascospore and the resulting heterogenous nuclei do not separate into different
daughter cells. The fact that sometimes C. carbonum strains form sectors with de novo
T 0X2 chromosome rearrangements (isolate 512-98D) proves that such rearrangements

can be produced during mitosis (mitotic instability).

Pathogenicity phenotype and TOX2 gene expression in strains with chromosomal

aberrations

Pathogenicity assays were done on strains with detectable deletions within the
defined region between the two flanking copies of T OXC (summarized in Table 2).
Strain 373-39 had an RV disease phenotype, and strains 512-3, 512-48, 643-9, 625-66
and 164R10 had a Tox2+ (fully virulent) phenotype.

Isolates 512-3, 512-48 and 164R10 were tested for HC-toxin production by J.
Pitkin (unpublished data summarized in Table 2). All produced HC-toxin in culture at
levels comparable to the wild type Tox2+ strains.

One possible reason for the putative low levels of HC-toxin in RV strains was that
one or more of the HC-toxin biosynthetic genes was underexpressed. To test this
hypothesis, we tested all the deletion strains for the presence of T OXC, TOM (J. Pitkin,
unpublished data) and TOXE (Figure 23) mRNA by Northern blotting and found that

these genes are expressed at a level similar to wild type in all the strains used in this

69

 

Figure 23. Northern blot analysis of TOXE expression in the TOX2 deletion strains.
(A) Total RNA was probed with TOXE internal fragment. (B) The same RNA gel stained
with ethidium bromide before transferring RNA to nitrocellulose filter. Lanes: 1, S81 1 1; 2,
CC141; 3, 164R10; 4, 243—7; 5, 373—39; 6, 512-3; 7, 512—48; 8, 560—2 (243-7 with remain-
ing copy of HTS] mutated by targeted gene disruption); 9, 243-10 (Tox2').

70

study.

Because the HTS mRNA is too large to be reliably detected by Northern blotting
(Scott-Craig et al., 1992), we tested the deletion strains for the presence of HTS and HTS
enzyme activity. HTS was partially purified by ammonium sulfate fractionation from
mycelial pads and analyzed by Western blotting using anti-HTS-l polyclonal antibodies
(Scott-Craig et al., 1992) (Figure 24). HTS protein was detected in all of the strains. To
compare HTS activity in RV and wild type Tox2+ strains, cultures of 367-2, 243-7, 373-
39 and 243-10 were grown in parallel, HTS was partially purified from mycelial pads by
ammonium sulfate fractionation and anion exchange chromatography, and the HTS
activity was compared in the fractions containing maximum activity (Figure 25). Tox2+
and RV isolates had comparable HTS activity, the differences between isolates being
considered within the range of experimental variation (Walton, 1987). Therefore, the
defect in HC-toxin production in RV strains is probably not related to a defect in HTS.
HTS activity was also measured in most other deletion strains (Tox2+) and was also
found to be comparable to that of the wild type (Table 3; summarized in Table 2, p. 52).

In conclusion, no changes in expression of any known TOX2 genes were detected

in RV strains or any other strains with chromosomal aberrations.

Possible mechanisms that may cause the R V phenotype

The reduced virulence phenotype in some isolates of C. carbonum was found to
be associated with large deletions in the T 0X2 chromosome. Because the disease
phenotype of the RV strains became race 2 (non-virulent) when the remaining copy of

HT S1 was disrupted in strain 243-7, we conclude that the RV strains produce at least

71

 

Figure 24. Western blot of HTS preparations from different strains used in this study.
Protein extracts were purified through ammonium sulfate precipitation, resolved on a 6%
acrylamide SDS—PAGE gel, and transferred to nitrocellulose as described by Scott—Craig et
al., 1992. The filter was probed with anti-HTS-l antiserum (raised against HTS-1, the 220-
kDa portion of HTS that contains L-proline-activating activity and is one of the HTS break-
down products, Scott-Craig et al., 1992). The difference in size of the protein to which the
antibody binds is the same as is normally observed when different preparations from the
same wild type strain are compared (Scott-Craig et al., 1992). It is explained by the break-
down of the native HTS (570 kDa) into “HTS—l” (~200—220 kDa), “HTS-2” (~160 kDa),
and some other breakdown products. HTS is extremely prone to this partial breakdown and
two major breakdown products were originally thought to be two different enzymes (Walton
and Holden, 1988). Lanes: 1, SB1 11; 2, 643—9; 3, 164R10; 4, 512-8; 5, 373-39; 6, 243-7; 7,
512—3; 8, 512—48; 9, 367-2; 10, 512-98; 11, 625-66; 12, CC141; 13, 243-10.

72

15

 

10-

 

cpm x 1000

    

 

 

$13111 243-7 373-39 243-10
(Tox2+) (RV) (RV) (Tox2')

Figure 25. HC-toxin synthetase (HTS) activity in the RV isolates of C. carbonum.
The protein extracts from isolates SB111 (Tox2+), 373-39 (RV), 243-7 (RV), and 243-10
(Tox2) were desalted, and nine mg protein from each preparation was used for anion
exchange HPLC (Walton and Holden, 1988). Ten 111 of each l-ml fraction was assayed for
alanine-dependent ATP/PP.l exchange. The reaction mixes contained 300,000 cpm

[’2P]pyrophosphate per reaction (125 pl). The results for the peak fractions (either fraction
20 or 21) are shown.

73

Table 3. HC-toxin synthetase (HTS) activity in isolates of C. carbonum.

D-alanine and L-proline-dependent ATP/PP, exchange was measured in HTS
preparations partially purified through ammonium sulfate precipitation and desalted.
Specific activity was calculated as pmol PPi/min/pg protein. For each reaction, 10 pg
protein in 25 pl of extraction buffer (Walton and Holden, 1988) was added to 100 pl
reaction buffer (Walton and Holden, 1988) containing 642,313 cpm of [32P]PP3, and
reaction mixes were incubated for 20 min at 30° C.

 

 

 

 

 

 

 

 

 

 

L-Pro -dependent specific D-Ala -dependent specific
activity activity

Strain

243-7 13.6 6.8

643-9 19.5 14.6

164R10 7.3 2.9

367-2A 18.5 10.2

512-8 16.1 6.8

625-66 12.2 4.9

 

 

 

74

some small amount of HC-toxin in planta which accounts for the residual virulence.
However, we could not detect HC-toxin in culture filtrates of RV strains. It is possible
that RV strains produce less than 1% of the wild type-level of HC-toxin. Alternatively,
these strains may be producing HC-toxin only in planta. We propose that reduced
amounts of HC-toxin are responsible for the reduced virulence phenotype.

It is known (Panaccione et al., 1992) that both copies of HT S] are active in wild-
type race 1 strains, and when one or the other is disrupted, the remaining one is sufficient
to sustain a race 1 (virulent) phenotype. Therefore, the RV phenotype is not due simply
to the loss of one of the copies of HT S] .

Several explanations are possible for the reduction in HC-toxin production in RV
strains. The presence and levels of activity of HC-toxin synthetase (HTS) in RV strains
were similar to those in wild-type virulent strains. Therefore, the low level of HC-toxin
production is not due to a low level of HTS. One possibility is that HTS in the RV strains
is defective in some step other than amino acid activation, such as thioesterification,
epimerization, or peptide bond formation. Such a defect would not be detectable by
ATP/PP, exchange assay or Western blotting. However, the fact that both copies of HT S]
are fully functional in the Tox2+ parents of RV strains and each of the copies alone
produce a fully virulent phenotype makes this explanation unlikely. It would require two
independent mutational events during or after meiosis in two different crosses to produce
two strains (243-7 and 373-39) with identical phenotypes.

Another possibility is the loss of a gene involved in some step of HC-toxin
biosynthesis other than those catalyzed by HT S] and T OXC. For the fungus to retain a

residual level of HC-toxin production there should be another gene whose product can

75

partially substitute for the missing enzyme. The absence of this gene may be partially
complemented by another enzyme, presumably encoded by a housekeeping gene. This
partial complementation would result in the low levels of HC-toxin production. For
example, the putative or subunit of TOXC may have a housekeeping homologue involved
in primary fatty acid synthesis that may be responsible for the residual levels of HC-toxin
if the TOX2+ specific gene (T OXC) is deleted. We also know that C. carbonum has
peptide synthetase genes other than HT S1 (Nikolskaya et al., 1995). These peptide
synthetases may have their own corresponding set of biosynthetic genes (e.g., alanine
racemase), and they could also play a role in partially substituting for the gene missing in
RV strains. Basal levels of HC-toxin may also be produced if one of the two copies of a
biosynthetic gene was partly defective, and the fully functional copy was lost. It is also
possible that the gene predicted to be missing in RV strains may actually be present but
inactivated by the proximity to the new telomere.

It is also possible that RV isolates have a defect in the regulation of expression of
genes other than HT S] or TOXC involved in HC-toxin biosynthesis. Changes in
regulation may be due to the loss of a positive activator gene that controls their
expression. This may result in basal expression of such HC-toxin biosynthetic gene(s),
decreased HC-toxin production and a reduction in virulence. So far one Tox2+ -unique
gene (T OXE) encoding a promoter-binding protein has been found (Ahn and Walton,
1998; K. Pedley, unpublished data). TOXE, however, is not responsible for the RV
phenotype because it is necessary for the transcription of TOM, TOXC and TOXD, is
present in RV strains, and is expressed in these strains at wild-type levels (Figures 18 and

23).

76

Alternatively, HC-toxin of much reduced activity (e.g., lacking an epoxide group)
may be produced by RV strains. In this case, it may be not detectable by the epoxide-
specific assay used after TLC separation. Both the epoxide group and the 8-carbonyl
group of the Aeo side chain are known to be important for biological activity of HC-toxin
(Ciuffetti et al., 1983, Kim et al., 1987). Current evidence indicates that the site of action
of HC-toxin is histone deacetylase, an enzyme that reversibly deacetylates core histones
while they are assembled in chromatin. HC-toxin was shown to inhibit histone
deacetylase (Brosch et al., 1995). Recently, apicidin, a novel cyclic tetrapeptide, was also
shown to inhibit histone deacetylase and to have strong cytostatic activity similar to HC-
toxin (Darkin-Rattray et al., 1996) (Figure 1, p. 5). Apicidin lacks the epoxide group but
has the 8-carbonyl group on the 2-amino-8-oxo-decanoic acid moiety, similar to HC-
toxin. This makes it plausible that an HC-toxin derivative without the epoxide group,
while lacking the biological activity in the standard root growth assay, may still possess
residual activity during the plant-pathogen interaction. Such altered HC-toxin can be
produced if the RV strains lack gene or genes required for the epoxide group
biosynthesis.

In conclusion, to account for the low level of HC-toxin produced by RV strains,
the gene or genes that are presumed missing could have a structural (biosynthetic) or
regulatory function. The deleted structural gene or genes could be partially substituted

for by other genes that encode enzymes with overlapping function.

77

Cosmids containing the chromosomal region adjacent to HTS 1/T OXA-I do not
complement the R V phenotype

Since it appears to be most likely that the RV phenotype is due to a missing gene
or genes, we attempted to narrow down the region where such gene(s) may be located by
comparing putative maps of all available strains with chromosome aberrations. From the
available data on strains 512-48 and 625-66 (Figure 21, p. 58; Table 2, p. 52), it appears
that insofar as at least one of these deletion isolates is missing all of its DNA to the left of
TOXD-1, this segment of the TOX2 chromosome is not required for HC-toxin
biosynthesis. Since all the data is consistent with strain 243-7 having a simple terminal
deletion of the entire segment to the left and including HTS] -1 (Ahn and Walton, 1996),
it is likely that there are no essential genes on this ca. 1.3 Mb of DNA. Thus, the left
segment of the T 0X2 chromosome appears to be conditionally dispensable.

The maps of the TOX2 chromosomes in the strains with chromosomal aberrations
are consistent with the hypothesis that a putative gene missing in RV strains maps to the
region immediately adjacent to HT SI/T OM-l (Figure 21). Based on the T 0X2
chromosome and PacI restriction fragment sizes, the breakpoint in strain 243-7 maps to
ca. 20 kb to the right of HT S1 -1 (Ahn and Walton, 1996). It is important to remember
that the exact breakpoint in this and other strains is unknown, because available markers
do not give sufficient resolution. The PacI restriction fi'agment size in the second RV
strain 373-39 (ca. 190 kb) is smaller than in 243-7 (ca. 280 kb), so this strain should be
missing the same region adjacent to HT SI/T OM-l plus 90 kb to the right. Strain 643-9
has a Pacl digestion fragment of ca. 290 kb and is fully virulent (Figures 20, p. 56; and

21, p. 58). The putative gene is predicted to be present in strain 643-9 and missing in

78

243-7 and 373-39, in which case it must be located within ca. 10 kb to the right of HTS] -
1. Allowing for the possibly imprecise distances determined by PFGE analysis, this
region can be expanded to ca. 20 kb. However, the T 0X2 chromosomes in some of these
strains may bear translocations and/or large internal deletions rather than having
undergone a single, simple chromosome break. (Strain 164R10 is an obvious example
and therefore was excluded from the above assessment of the location of the putative
missing gene).

To test the hypothesis that the missing gene is located in this region,
complementation experiments were conducted. They were designed to determine if the
RV phenotype could be restored to wild-type (virulent) by introducing the chromosomal
region in question back into 243-7. Chromosomal walking is ofien impossible in C.
carbonum because of the abundance of repetitive DNA. A C. carbonum cosmid library
constructed by J. Scott-Craig (unpublished data) was screened with two Tox2+-unique
DNA fragments, CC61 and a TOM internal fragment. These two fragments are located
in the rightmost and leftmost parts of the HT SI/T 0M cluster, respectively. By selecting
cosmid clones that hybridized to one of these probes and not the other, we obtained
cosmids that extended to the right or to the left of TOM/HTS], and that did not contain
the complete copy of TOM/HT S1. Among these cosmids, we selected by RFLP analysis
those that originated from copy 1 of HT SI/T 0M, and not from copy 2. The resulting
four cosmids (5E8, 7F6, 14F6, 15A7) covering the region immediately to the right of
HTSI/TOM-l were transformed into 243-7. Hygromycin-resistant transformants were
generated (5, 3, 20 and 6 transformants for 5E8, 7F6, 14F6 and 15A7, respectively).

They were isolated and tested for pathogenicity. All exhibited the RV phenotype and

79

none showed wild-type race 1 phenotype. Thus, these cosmids failed to complement the
RV phenotype.

These results indicate that either the putative missing gene is not located in this 20
kb region, or the cosmids used are chimeric. The map of the putative chromosomal
breaks is imprecise because the region between HT S] -1 and TOXC-2 in RV strains does
not necessarily correspond to this region in wild type. It may have originated by a
translocation, or a telomere may have added the critical “extra” DNA to the detectable
PacI fragment. In both of these cases, the actual chromosomal region missing in strain
243-7 may be much larger than 20 kb. More markers between T OXC-2 and HTSI-l

could help to determine if this is the case.

Materials and methods

Fungal culture growth and maintenance

Conidia of Cochliobolus carbonum S81 11 (Tox2+; ATCC 90305), SB114
(Tox2—; Walton, 1987), 367-2 (Gdrlach et al., 1998), CC141 (Ahn and Walton, 1996),
243-7 (Ahn and Walton, 1996), and others (this study: Figure 16; Table 2) were stored at
-80° C in 25% (v/v) glycerol and grown on V8 juice agar plates. For DNA or protein
extraction, mycelial agar plugs (0.5 cm2) were inoculated into l-L Erlenmeyer flasks
containing 125 ml modified Fries’ medium (Scheffer and Ullstrup, 1965). Cultures were
incubated at room temperature (21-23° C) without shaking for four days.

Fungal crosses were performed as described for Cochliobolus heterostrophus

(Yoder, 1988; Tzeng et al., 1992). Individual ascospores were picked and progeny were

80

put through two rounds of single-spore re-isolation to insure nuclear homogeneity.
For growth rate comparison, cultures were grown in horizontal race tubes
containing potato dextrose agar. Averages of duplicate samples were used to calculate

growth rate. Radial grth was measured on V8 juice agar plates.

Pathogenicity tests and leaf penetration tests

All pathogenicity tests were done in a greenhouse. Plants were grown in 30-cm
pots (three plants per pot) in soil composed of Bactomix (Michigan Peat Co., Houston,
TX) and sand, 4:1 (v/v). Leaves of maize inbred Pr (genotype hmI/hmI), typically at the
4-leaf stage (~25 cm tall) were sprayed with an atomizer with a suspension of conidia
(typically 5x104/ml) suspended in 0.1% Tween-20 and covered with plastic bags
overnight. Development of symptoms was observed each day for 7 to 21 days. Every
pathogenicity test was done at least three times, and in each of these experiments, one to
three pots were used to inoculate with each strain tested.

Seed and soil inoculation was done as described on pp. 38 and 40. Each of these
experiments was done twice, and 3 pots were used for each strain and each treatment.

For leaf penetration tests, maize plants at the 4-leaf stage or individual excised
leaves were inoculated with a suspension of conidia (5x104/ml) in 0.1% Tween-20. For
the whole plant inoculation, the general pathogenicity test protocol was followed and
leaves were examined at appropriate time points. For individual leaf inoculation, leaf
segments were wetted with the spore suspension, placed in closed Petri dishes on water-
saturated filter paper, and incubated at room temperature. To stain the fungal hyphae, a

solution of 0.1% cotton blue in lactophenol (20% phenol, 20% lactic acid, 40% glycerol,

81

20% water [v/v]) was diluted 1:3 (v/v) with 95% ethanol. Leaf segments were placed in
25 m1 of the resulting mix, boiled for 1 min, cooled and boiled again for 30 sec, and
incubated for 48 hr at room temperature. After washing in water several times, leaves
were mounted on glass slides with 50% glycerol in H20 and observed under a microscope

at the magnification of X100.

Fungal transformations

Preparation and transformation of protoplasts was as described (Scott-Craig et al.,
1990). Transforrnants able to grow on V8 juice agar containing 100 units/ml hygromycin
(Calbiochem, La Jolla, CA) were purified by two rounds of single-spore isolation to

obtain nuclear homogeneity.

Nucleic acid manipulations
Fungal genomic DNA and total RNA was isolated as described by Pitkin et a1.
(1996). Southern and Northern blot analysis was done as described by Sambrook et a1.

(1989).

Pulsed-field gel electophoresis

Chromosomal DNA was prepared as described by Orbach et a1. (1988). Agarose
plugs containing DNA were immersed in 1 ml of TE (10 mM Tris, 1 mM EDTA, pH 8.0)
and chilled on ice for 30 min. TE was replaced with 1 ml of NDS buffer (0.45 M EDTA,
10 mM Tris pH 7.5, 1% lauroyl sarcosine) containing 2 mg/ml Proteinase K and plugs

were incubated for 48 h at 50° C, then rinsed with 50 mM EDTA. Plugs containing 1 to 2

82

pg DNA were rinsed with 1 ml of TE, then TE was replaced with the appropriate
restriction endonuclease buffer and agarose plugs were incubated on ice for 1 hr. The
buffer was replaced with the fresh buffer containing 30 units of PacI restriction enzyme;
the plugs were incubated on ice overnight and then for 2 hr at 37°. After digestion,
enzyme and buffer were removed, the plugs were washed with TB and loaded onto the
CHEF gel.

Contour-clamped homogenous field (CHEF) electrophoresis (CHEF-DR II, Bio-
Rad) was performed as described by Ahn and Walton (1996). Chromosomal DNA was
fractionated by CHEF on a 1% chromosomal-grade agarose (Bio-Rad) gel at 170 V with a
2- to 25-sec switching time for 22 hr and then a 60- to 120-sec switching time for 24 hr.
After staining with ethidium bromide for photographing, the gels were destained and

transfer and hybridization were performed as described by Ahn and Walton (1996).

Protein manipulations and HTS activity assays

HTS enzyme extraction, partial purification by ammonium sulfate precipitation
and anion exchange HPLC, and assay by ATP/PP, exchange were done as described (Lee
and Lipmann, 1975; Walton, 1987; Walton and Holden, 1988). Protein was measured by
the method of Bradford (1976). For Western blot analysis, loading buffer (62.5mM Tris-
HCl, pH 6.8, 10% glycerol, 2% SDS, 5% [v/v] B-mercaptoethanol, 0.0025% Bromphenol
Blue) was added to the desalted samples and they were heated to 95° C for 5 min.
Resolving gels were prepared with an acrylamide concentration of 6%
(acrylamide/bisacrylamide ratio 30:0.8). Electrophoresis was performed in a mini-

PROTEAN minigel apparatus (Bio-Rad Laboratatories, Richmond, California) using high

83

molecular weight prestained standards (Bio-Rad Laboratories). After electrophoresis,
proteins were transferred to nitrocellulose using an LKB Multiphor H transfer unit. The
nitrocellulose was blocked with 0.3% Tween-20, incubated first with polyclonal
anti-HTS-l antibody (Scott-Craig et al., 1992) at 1:500 dilution and then with goat

anti-mouse IgG coupled with alkaline phosphatase (Sigma) at 1:1000 dilution.

84

Chapter 3

EXGI, A NOVEL B-l-3-EXOGLUCANASE FROM COCHLIOBOLUS

CARB 0N UM

Abstract

Results described in this chapter were published as Nikolskaya et al., 1998.

Genomic and cDNA copies of EXGI, a gene encoding an exo-B-l,3-glucanase
from the filamentous fungus Cochliobolus carbonum, were isolated. The gene contains
two introns of 50 and 53 bp, and the mRNA has a 3' untranslated region of 159 bases.
The deduced protein product, EXGlp, has a predicted 17-amino acid signal peptide but
apparently undergoes a second processing event resulting in the removal of a total of 42
amino acids. At the time (1995) of submission of the EXGI sequence to GenBank
(accession # L48994), the deduced amino acid sequence of EXGlp was not closely
related to any other known protein, thus being a novel glucanase. Later, three glucanase
gene sequences from mycoparasitic fungi became available that belong to the same
family: BGN13.1, an endo-B—1,3-glucanase from Trichoderma harzianum, and exo-B-1,3-
glucanases from T. harzianum and one from Ampelomyces quisqualis. BGN13.1, in
contrast to EXGlp and the other two exoglucanases from this family, is acidic rather than
basic, and its distribution of cysteine residues is biased towards the carboxy half of the
protein. EXGlp and the other two exo-B-1,3-glucanases from this family contain two
imperfect copies of a 23-amino acid motif that is found in several other proteins that

interact with polysaccharides, including plant and bacterial polygalacturonases, phage

85

neck appendage protein, phage endoneuramidase, and bacterial mannuronan epimerase.

Introduction

Extracellular cell-wall degrading enzymes are widespread among plant pathogenic
and saprophytic microorganisms. During plant pathogenesis, cell wall degrading
enzymes may contribute to penetration, ramification, and the acquisition of nutrients from
plant wall polymers (Walton, 1994).

B-1,3-Glucanases occur widely in bacteria, fungi and higher plants. In fungi, they
have been proposed to have one or more functions. The nutritional utilization of B-1,3-
glucans is one obvious possible function (e.g., Stahmann et al., 1992; Lorito et al., 1994),
but they have also been proposed to be involved in autodigestive loosening of the wall to
promote wall expansion and hence growth (Nobela et al., 1988). Although plants
normally contain only small amounts of B-l,3-glucan, during plant pathogenesis B-1,3-
glucanases could have a role in penetration through papillae, which are pathogen-induced
wall appositions containing the B-1,3-glucan known as callose (Schaeffer et al., 1994).

The filamentous fungus Cochliobolus carbonum produces many extracellular cell
wall degrading enzymes, including at least one enzyme with [3-1,3-glucanase activity
(Walton, 1994). The major B-1,3-glucanase activity of C. carbonum, EXGlp, an exo-
acting enzyme, has been purified (Van Hoof et al., 1991). A partial genomic clone of the
gene encoding EXGlp, EXGI, was isolated by H. Schaeffer and used to create, by
targeted gene disruption, a strain of C. carbonum mutated in EXGI . Disappearance of the
major chromatographic peak of exo-B-1,3-glucanase activity in the mutant strain

indicated that the correct protein had been purified and that the correct gene had been

86

disrupted (Schaeffer et al., 1994). The exg] mutant grew much less well than wild type
on B-1,3-glucan as sole carbon source but had unaltered pathogenicity on maize
(Schaeffer et al., 1994). The exg] mutant strain had significant residual [3-1,3-glucanase
activity due to MLGlp, an endo-acting enzyme that can degrade both B-1,3-B-1,4-
(mixed-linkage) and B-1,3-glucans (Van Hoof et al., 1991; Gdrlach et al., 1998), EXGZ

(J. Ahn, unpublished data), and perhaps to other [3-1 ,3-glucanases as well.

Results and discussion

Analysis of genomic and cDNA clones of EXGI

A partial genomic copy of EXGI was previously isolated using PCR primers
based on the experimentally determined sequences of N-terminal and internal tryptic
peptides of EXGlp (Schaeffer et al., 1994). To obtain a clone containing the entire
coding region, a 650-bp BgIII fragment internal to this open reading frame was used as a
probe to screen a genomic DNA library made in phage lambda. Two overlapping
fragments that hybridized to the probe were subcloned from a single lambda clone and a
total of 4048 bp was sequenced on both strands (Figure 26).

The same BgIII fragment was used to screen a cDNA library made from mRNA
from C. carbonum grown on maize cell walls (Pitkin et al., 1996). Two overlapping
cDNA clones were isolated and the sequences of both strands determined. Genomic and
cDNA sequences were entirely colinear with the exception of two introns of 50 and 53 bp
(Figure 27). The introns contain conserved 5' (consensus GTAMGH, where M = A or C,

and H = C, T or A) and 3' (consensus YAG, where Y = C or T) splice junctions as well as

87

le

 

 

 

 

 

A.
Bm S BSB E Bm
| l l | l
l 1H1 l
H CCH H C

 

 

Figure 26. Restriction map of EXGI genomic and cDNA clones. (A) Genomic
restriction map of the EXGI locus. Location of the two overlapping genomic fragments
and of the two overlapping cDNA clones are shown above and below the restriction map,
respectively. BglII fragment is shown as a black box. (B) Sequenced region with open
reading frame shown as a shaded box. Location of the two introns is shown by triangles.
B, BgIII; Bm, BamHI; C, ClaI; E, EcoRV; H, HindIII; S, SalI.

88

91
181
271
361
451
541

631

721
21
811
44
901
74
991
86
1081
116
1171
146
1261
176
1351
206
1441
231
1531
261
1621
291
1711
321
1801
351
1891
381
1981
411
2071
441
2161
471
2251
501
2341
531
2431
561
2521
591
2611
621
2701
651
2791
661
2881
711
2971
741
3061
771
3151
3241

3331
3421
3511
3601
3691
3781
3871
3961

Figure 27. Nucleotide and deduced amino acid sequences of EXGI. Intron sequences
are shown in lower case, and conserved splice sites are underlined. The conserved motifs
are highlighted. The start of the longest cDNA and the polyadenylation site are indicated
underneath the corresponding nucleotide by @ and &, respectively.
underlined regions are the peptide sequences obtained from amino acid sequencing of the
purified protein (Schaeffer et al., 1993), the first one corresponds to the amino terminus
of the mature protein. The sequence was submitted to GenBank in 1995 (accession

CTACTACCTCTACTATACCACCGGACTATCTACCAGCGATCTTCAGTACCGCATTTGTAGCCCTATGTTTCGGAAATGAAGCTCTATGCG
GATTAGATCGTAGCCTAGCCCGACATGCGCGATCACACTGGATATGGCATTGTCAGAGGCAAAGTCACACTACCCTGCCCTCCAAGATAT
CCCCTACAAGGGTTTTCTTTTAGGGGAAGGTGATGGATAACCGCCCTACCCCAGATTATGCCATTTGAACTAGGCGGTTCCAACCATACA
AAAAGACGCAAGACATGGAGCCACCCAAATGGCCGCATGTGTAGTCGCGCACCAAGAGTGCAAACCATCACTCCGTGTGACCTTTGAGCG
GGAGCCAGTAAGGAATCCTGACCGGGTCCTGCCTGTTTGTGAGCTGCTCCGGGGAGTGTTCATCCTCGATCCACCCGGTGTAGTGAACGT
CGCGTTTACACGATGATAAAACCGGGGCTACAAGGAATGTGTTGACAGAATACTTAAATAGAGAGCATCACCTGCCTTTGAGACATTCCC
ATCTCTACTACGCTCTACATTTCCTTGTATAAAGTGTCCTTCAATTCACTTTGCTGCTCAATCGGAAATCTGCTGGTCTGAAGTTCACGA
@
GCATGCGTTTTTCTTCTTTGCTCGCCTGCCTAGGTGCAGTCGGCATTCAAGCCGCTGCTATACgtacgaagatgtcttccqctttacaac
M R F S S L L A C L G A V G I Q A A A I
tcctggttactqacactttgtagCATTCCAAAGGCGTGTTGATAACACTACCGACAGTGGAAGTCTTGATGLIGLILAAGLIGLGGLIGL
P Q R R V D N T T D S G S L D A A Q A A A A
TATAGTCGATGGCTACTGGCTAAAC GATCTCTCCGGCAAAGGCAGAGCCCCTTTTAACAGCAACCCGAACTACAAGGTCTTCCGAAATGT
I_,¥::LL,G I ,HL,L_:§ D_.L 5. G K G R A P F N S N P N Y K V F R N V
CAAGGATTACGGAGCGAAGGgtaagcaattttttttcacattgatcttgaggatataactaaccgatttgtagGTGACGGTGTCACTGAC
K D Y 6 A K G D G V T D
GACTCTGATGCCTTCAACCGTGCCATCTCTGACGGCAGCCGTTGCGGCCCATGGGTTTGCGACTCGTCAACTGACAGCCCAGCTGTTGTT
D S D A P N R A I S D G S R C G P N V C D S S T D S P A V v
TACGTGCCTTCTGGAACCTATCTCATCAACAAGCCCATCATCTTCTACTACATGACTGCTCTCATCGGTAACCCCCGCGAACTTCCCGTC
Y V P S G T Y L I N K P I I F Y Y M T A L I G N P R E L P V
CTCAAGGCTGCATCTTCACTCCAAGCTCTTGCTCTGATCGACGGAAGCCCCTACAGCAACCAAAACGGTGAGCCCGGCTGGATCTCAACC
L K A A S S L Q A L A L I D G S P Y S N O N G E P G w I S T
AACTTGTTCTTGCGCCAAATCCGCAACTTGATCATCGATGGCACTGCTGTTGCACCAACATCGGGTTTCCAGGCTATCCATTGGCCCGCC
N L F L R Q I R N L I I D G T A V A P T S G F O A I H N P A
TCTCAACCIACCAC! KTC AAA‘T’TKAASATCCGCAT ACACAGGCGTCCAACTCTGTTCACGCTGGTATCTTTGTCGAGAATGGATCT
S Q A T T I Q N V K I R M T C A S N S V H A G I F V E N G S
GGCGGTCATATGGCCGACC TC GAC CATCACCGGTGGTCTGTACGGCATGAACATTGGCAATCAGCAGTTCACCATGCGTAACGTCAAGATC
G G H M A D L D I T G G L Y G M N I G N O O F T M R N V K I
TCCAAGGCTGTCGTCGGTATCTCACAAATCTGGAACTGGGGCTGGCTGTACTCTGGTCTCCAGATCAGCGACTGCGGCACTGCTTTCTCC
S K A V V G I S Q I H N N G w L Y S G L Q I S D C G T A F S
ATGGTTAACGGTGGCTCTGCTGGCAAACAGGAGGTTGGCTCCGCCGTCATCATCGATTCTGAGATTACCAACTGCCAAAAGTTTGTCGAC
M V N G G S A G K O E V G S A V I I D S E I T N C O K F V D
TCAGCATGGTCGCAGACCAGCAACCCTACCGGTTCCGGCCAGCTCGTCATTGAGAACATCAAGCTCACCAACGTTCCCGCTGCTGTTGTC
S A w 5 Q T S N P T G S G O L V I E N I K L T N V P A A V V
AGCAATGGCGCCACTGTCCTCGCTGGCGGCTCTCTTACCATCCAGACCTGGGGTCAGGGCAACAAGTACGCACCCAACGCATCTGGCCCA
S N G A T V L A G G S L T I Q T N G 0 G N K Y A P N A S G P
TCCAAGTTCCAGGGCGCCATCAGCGGTGCCACTCGTCCCACTGGTCTCCTCCAGAACGGCAAGTTCTACTCCAAGTCGAAGCCACAGTAC
S K F O G A I S G A T R P T G L L O N G K _E. g2 5:,K_ S K P Q_;X
GAGACTCTCAGCACTTCAAGCTTTATCAGTGCCCGCGGTGCAGGTGCAACCGGTGATGGTGTCACTGACGACACACGCGCCGTCCAGGCT
E T__L? S T S S F I s A R G A G A T G D G V T D D T R A V Q A
GCCGTCACTCAGGCCGCGTCTCAGAACAAGGTCCTCTTCTTCGAGCACGGCGTCTACAAGGTCACCAACACCATCTACGTTCCCCCCGGC
A V T Q A A S O N K V L F F E H G V Y K V T N T I Y V P P G
TCC CGCATGGTCGGTGAGATCTTCTCCGCCATCATGGGCTCTGGCAGCACCTTCGGCGACCAAGCCAACCCCGTCCCCATTATCCAAATC
R M V G E I F S A I M G S G S T F G D Q A N P V P I I Q I
GGCAAGCCC GGCGAGTCCGGCAGCATCGAGTGGTCCGACATGATTGTCCAGACCCAAGGCGCAACCCCAGGAGCCATCGTCATCCAGTAC
G K P G E S G S I E N S D M I V 0 T Q G A T P G A I V I O Y
AACCTCAACALG GLLLIIGGL1LLGGILIL1GGGACGTCCACACCCGCATCGGCGGCGCAAAGGGAACCAACCTCCAAGTCGCCCAGTGC
N L N T A L G S G L w D V H T R I G G A K G T N L O V A Q C
CCCGCGGTCCTCGGCCAAGTCAAGCCCGAATGCTTCTCTGCGCACACCAACGTGCACGTAACTAAGGGCGCCAACGGCGCCTACTTTGAA
P A V L G O V K P E C F S A H T N V H V T K G A N G A Y F E
AAC AACT GGT TC TGGA CCGCCGACCACGACCTCGACGACGCAGACTCGACCCGCATCAACATCTACACCGGCCGCGGCTTCCACGTCGAA
N N w F w T A D H D L D D A D s T R I N I Y T G R G F H V E
GCAAAC AACGTCT GG GATCTG GGCAAACCGCGCAGAGCACCACACCATGTACCAGTACCAATTCAACGCCGCCCAAGACATCTTCGCAGGC
ACN CN V Gw I w A N G A E H H T M Y Q Y 0 F N A A o D I F A G

 

 

 

U)

 

P Y P Q P T P I A P L P Y V S S S K Y S D P V Y
AGCCTGGGSCCTCCG<TTGCTCGACGCAAAAAACGTACTCATCTACGGCGGCGGCCTCTACTCCTTCTTCGAC
Q S A W G L R L L D A K N V L I Y 6 G G L Y S F F D
CGTCGGATGCTCTTCTCCCACCGCCCCCAACGGCTTCCGCGACTGCCAGACCCGCATTCTGAGCATCGAGGGATCGACGAGT
V G C S S P T A P N G F R D C O T R I L S I E G S T S
66 GTTTGGATTCAGCGAGGTTGGTGTCGAGTGGATGGTTACGGCCGCTGGTCAGGATAAGGCGAATTGGAAGGATAATCTAAGT
V O A F G F S E V G V E W M V T A A G O D K A N W K D N L S
GTTTATCCTACTACCATTGGGTATTTGAGCTATGGGTTTTAAGATGGTAAGTGAGGGGGGAAGGGAAGATGGGGTTTGGTATATATATAT

V Y P T T I G Y L S Y G F '
GTTTCACTTCTTTGGTTTGTGGGACAGGGGAGATTGCAAACGGTACATAGATGATGGATATGAGACATGACAGLGIIL1111111111G1
AGTTATCAI11111IILILIIILLIILbbILACCACTTACGGTCGCTTTCTCATTTCGCATAGCCAACTTCTCATTTGCACATATCTGGC

6

AAAGAGAAGGAAAGAAAAAAAAGCTTTACGTTGGTTGTTTTTATAGCAAAGACAAGTTAACGGGGAAAAAGGGCGAGCTTTCTTGGTAGC
GAAATACAATTGTTTACAACAAAAAAGGTGTTCCTACTGCTGCTATATAAGCATTTGCTATCAACGTTGTATTCCTTTGGGATGACCAAA
AGGTAAATTACCATCTGCCTTTTCCAACAAAAAAAAGCATATCACGTGCACCTAAAACGCCCGAGTAAAGAAAAAAAAGAAAGAACAACG
AAAAGTTTGACAACACAGAAACAAAAGCAAAGGAACATACAGAACCACGTCGTTATGTGCGACTTCCACATATACCAAAAAGTAAAAAAG
CAACAATCATAACGATGACGAATATCCTTTCATATATATGACCAGAGAGATGCGAGTTCGCGATGCCAGCATGAAAAATGGGCGCCCCTC
TCCATATCTTTCTCTTTCCTATATTTCCCGATGTACGTGGAAACAAGAAGGTATAGTCTCGAAGGATAGATAGGAAATAATAAAAGAAAA
AAAACATTGGAAGCGTTTGCCCTACGATGACAACGACAAGAGCAGAAATAGCGAAAGAAAGTTGCGAAGACAGCGATTATTCAAGGAAAA
ATCAACGAAGAAAAGAGTGCGCAGTCGAACGCCGGGAGTTCCCGAATATTCACAATAGCAACAACATAAAGGAAAAAAAAGAGAACGT

t—i HI
rtD><y>r

O

H)

m

._]
><rrion

O

1‘)
.E.48
v-1 (‘1 If)

)1

>4

OD

Er.
n (1
n a
>‘(3’MtDr‘Z
a

0
~32

 

 

number L48994).

89

The double

internal consensus sequences RCTRAC, where R = A or G (Orbach et al., 1986;
Ballance, 1986) (Figure 27). Based on the sequences of two independent cDNAs, the
polyadenylation site is 159 bp downstream of the stop codon, TAA. The first ATG codon
between the start of the longest cDNA and the experimentally determined N-terminus of
the mature protein (at nucleotide 633) is assumed to be the translational start site. This
conclusion is supported by the lack of any other in-frame Met residues, numerous stop
codons in all frames upstream of the cDNA start, and the presence of a strongly predicted
signal peptide from amino acids 1 through 17 (see below).

EX GI encodes three peptides derived directly from the purified protein, indicated
by double underlining in Figure 27 (Schaeffer et al., 1994). The predicted size of the
mature EXGlp protein is 79 kDa, in reasonable agreement with the size determined by
SDS-PAGE, 73 kDa (Van Hoof et al., 1991). A signal peptide cleavage site is predicted
between amino acids 17 and 18 by the SignalP program (Nielsen et al., 1997). Because
the mature protein starts at amino acid 42 (Van Hoof et al., 1991), EXGlp probably
undergoes a second processing event, but there is not a pair of basic residues at this
position indicative of processing by a KEXZ- like protease. There are three potential N-
glycosylation sites (NXS/T) at amino acids 233, 333, and 773, but glycosylation of the

mature protein was not detected (Van Hoof et al., 1991).

EXGlp is a novel ,6-1,3-glucanase
At the time when these results were submitted to GenBank, analysis using BLAST
indicated that EXGI p had some overall amino acid sequence similarity to BGN13.1, an

endo-acting [3-1,3-glucanase produced by the mycoparasitic fimgus Trichoderma

9O

harzianum (de la Cruz et al., 1995). The two sequences are only 29% identical overall,
and the largest contiguous stretch of identical amino acids is seven (AASQNKV, starting
at position 450) (Figure 28). Unlike BGN13.1, the distribution of the cysteine residues of
EXGlp is not biased toward the carboxy terminal half of the protein. Furthermore,
EXGlp is an acidic protein, with an experimentally determined as well as predicted pI of
6.3 (Van Hoof et al., 1991), whereas BGN13.1 is basic, with an experimentally
determined pl of 7.7 to 8.0 and a predicted pI of 7.5 (de la Cruz et al., 1995).

Recently, two new B-l,3-exoglucanases that belong to the same novel family were
sequenced: one from T richoderma harzianum (Cohen-Kupiec et al., 1998) and another
from Ampelomyces quisqualis (Rotem et al., 1998) (Figure 28). They are 42% and 38%
identical to EXGI p and have pI’s of 4.7 and 4.9, respectively. Thus, these proteins are
more similar to EXGlp than is BGN13.1. It was proposed that cysteine residues that
occur predominantly in the C-terminal portion of BGN13.1 may play a role in cell wall
binding (de la Cruz et al., 1995). Some of these residues are conserved throughout the
members of the EXGlp family (Figure 28), but others are not. Furthermore, in the three
exoglucanases, unlike in BGN13.1, the distribution of cysteine residues is not biased
towards the carboxy half of the protein. Whether the carboxy-terminal domain of any or
all of the glucanases from the EXGlp family is a cell wall-binding domain has to be
determined experimentally.

The deduced amino acid sequences of [3—1,3-glucanases from barley and yeast are
conserved around the nucleophilic active site, V(S/G)E(S/T)GWP(S/T) (Chen et al.,
1993). EXGlp contains a sequence, GEPGWIS, starting at amino acid 167, that is

somewhat similar (GEXGWXS) to this motif. The Glu, second Gly, and final Ser

91

Figure 28. Comparison of the predicted amino acid sequence of C. carbonum
EXGlp with other (LLB-glucanases from the same novel family. The sequences of
EXGlp, Trichoderma harzianum B-l,3-exoglucanase (Cohen-Kupiec, 1998),
Ampelomyces quisqualis [3-1,3-exoglucanase (Rotem et al., 1998), and T. harzianum B-
1,3-endoglucanase (BGN13.1, De la Cruz et al., 1995) were compared using
LASERGENE software (DNASTAR, Inc., Madison, WI, USA). Identical amino acids
are highlighted, conserved cysteine residues are indicated by asterisks, and the N-termini
of mature proteins are shown in bold.

92

$9.399.” FUN“? F3???)

9??“

F3???) 35”.”? 2‘3???) “5‘70

53.373.”

59.37)." 7*???“ 739’???” ”3'70

5’???

‘9

Figure 28. Comparison of the predicted amino acid sequence of C. carbonum

carbonum EXGlp
harzianum EXG
quisqualis EXG

harzianum BGN13.

carbonum EXG1p
harzianum EXG
quisqualis EXG

harzianum BGN13.

carbonum EXGlp
harzianum EXG
quisqualis EXG

harzianum BGN13.

carbonum EXGlp
harzianum EXG
quisqualis EXG

harzianum BGN13.

carbonum EXGlp
harzianum EXG
quisqualis EXG

harzianum BGN13.

carbonum EXGlp
harzianum EXG
quisqualis EXG

harzianum BGN13.

carbonum EXGlp
harzianum EXG
quisqualis EXG

harzianum BGN13.

carbonum EXGIp
harzianum EXG
quisqualis EXG

harzianum BGN13.

carbonum EXGlp
harzianum EXG
quisqualis EXG

harzianum BGN13.

carbonum EXGlp
harzianum EXG
quisqualis EXG

harzianum BGN13.

carbonum EXGlp
harzianum EXG
quisqualis EXG

harzianum BGN13.

carbonum EXGlp
harzianum EXG
quisqualis EXG

harzianum BGN13.

carbonum EXGlp
harzianum EXG
quisqualis EXG

harzianum BGN13.

carbonum EXGlp
harzianum EXG
quisqualis EXG

harzianum BGN13.

harzianum EXG

harzianum EXG

harzianum EXG

harzianum EXG

M.RF..SSLLACLGAVGIQAAAIPFORRVDNTTDSGSLDAAQAAAAIVDGYWLNDLSGKG

M.GFIRSAVLSALT .......... FAAACRGLATPGS.EAEPSVEKRASSYWYENIAHQG
MLAFSAGAFLLTLRV ......... FLTATPSAAAPVA.QAVEVPQAGASGYWFGNIKRQG
........................................... ATSFYYPNMDHVNAPRG

RAPFNSNP.NYKVFRNVKDYGAKGDGVTDDSDAFNRAISDGSRCGPWVCDSSTDSPAVVY
IAFFAPS..NYTVFRNVKDYGAKGDGVTDDTAAINNAILSGGRCG.RLCTSSTLTPAVVY
IAPYNENPAAYKVFRNVKLLGAKGDGVTDDTAAINAAIADGNRCGQ.GCDSTTTSPAIIY
FAPDLDGDFNYPIYQZVN...A.GDG ..... LQNAITTDG.KGGSRHPQWFASQPRVVY

VPSGTYLINKPIIFYYMTALIGNPRELPVLKAASSLQALALIDGSPY..SNQNGEPGWIS
FPAGTYVISTPIIDQYYTNIIGDPTNLPT1KATAGFSGIALIDGDTYYGDNNPNDPNWIS
FPAGTYLISEPIIOYYYTQFVGDATNPPTLKAKDTFEGMGLIDADPYIPGGDGAN..WYT
IPPGTYTISKTLRFNTDTILMGDPTNPPIIKAAAGFSG...DQTLISAQDPSTNBKGELS

T.NLFLRQIRNLIIDGTAVAPTSG.FQAIHWPASQATTIQNVKIRMTQA.SNSVHAGIFV
T.NVFYRQVRNFKLDMTSIPTSAPKIYGIHWPTAQATSLQNIQITMSTA.SGNSQVGLFI
NQNNFYRQIRNFVIDIKDTKAAA....GIHWQVSQATSLQNIRFEMATGEAGANQKGIFQ
....EAVAIKNVVLDTTAI.PGGNSFTALWWGVAQAAHLQNVRITMSSSSGGNGHTGIRM

ENGSGGHMADLDITGGLYGMNI.GNQQFTMRNVKISKAVVGISQIWNWGWLYSGLQISDC
ENGSAGFLTDMTFNGGLIGAAI.GNQQTTMRNLVFNNCAQPLSAASIGSGFTRAISINNC
DNGSGGFMSDLVFNGGAIGAFL.GSQQFTTRNMTFNNCGTAIFMNWNWLWTLKSIFINDC
GRGSTLGLADVRVERGQNGIWIDGHQQASFHNIYFFQNTIGMLISSGNTFSIFSSTFDTC

*

GTAFSMVNGGSAGKQEVGSAVIIDSEITNCQKFVDSAWSQTSNPTGSGQLVIENIKLTNV

GLGIDMTAA ........ ESITLIDSSISGTPVGIKTSFRRNQSPATSNSLIVENLSLNNV
KLGLDMAN..SPDNQTVGSVLLLDSKFTNTPIGINSSFTQDSVPHTGGTLIIDNVDFEGS
GTAFPTLAGSP ....... WIALIDAKSINSGV....TFTTNQFPS....FMIENLT.KDN

PAAV.VSNGATVLAGGSLTIQTWGQGNKYAPNASGPSKFQGAIS.GATRPTGLLQNGKFY
PVAIQSSSGSTILAGGTTTIAAWGQGHQYTPN..GPTTFQGSIT.PNSRPSSLLSGSNYY
NVAVQNVAGETLLAGKS.KVATWAOGNAMAAGQAQAGRVQGDVNNPPTKPQSLLGENGWF

GTPVVVVRGSTLV.GASSHVNTYSYGNTVGRNPTY ..... GDVTSSNTRPSALAPGGRYP
SKSKPQYETLSTSSFISARGAGATG ...... DGVTDDTRAVQAAVTQAASQNKVLFFEHG
TRSKPQYETLPVSSFRSVRSAGATG ...... NAVTDDTAALQSVINSATACGQIVYFDAG
ERSKPQYENIDVSKFVSLKDAGAVG ...... DGVTDDTAMIQKAID.GLQDGQILHADHG

YVAPPTYGDLPISSFLNVKDPAQNGNRQVKGDNTINEADTLNAILELAASQNKVAYFPFG

VYKVTNTIYVPPGS..RMVGEIFSAIMGSGSTFGDQANPVPIIOIGK.PGESGSIEWSDM
IYRITSTLSIPPGA..KIVGEEYPIIMSSGSFFNDQSNPKPVVQVGT.PGQTGQVEWSDM
AYLITKTIEIPAEKNIKIVGEIYTMFFITGKFFGNMDDPQPGFRVGKKSGDKGTFEMSDA
KYRVDSTLFIPKGS..RIVGEAWATITGNGNFFKNENSPQPVVSVG.RAGDVGIAQLQDL

IVQTQGATPGAIVIQYNLNTA..LGSGLWDVHTRIGGAKGTNLQVAQCPAVLGQ...VKP

IVSTQGTQAGAVLIEWNLATSG.TPSGMWDVHTRIGGFKGSNLQVAQCPVTASST.TVNT

IISTQGPAPGGILMEWNINAEA.GKAGLWDVHFRVGGFAGTNLQSSNCKKNPDTEHPPNE

RVTTNDVLPGAILVQFNMAGNNPGDVALWNSLVTVGGTRGAQALANACTNN ....... SN
fl

ECFSAHTNVHVTKGANGAYFENNWFWTADHDLDDADSTRINIYTGRGFHVEANNVWIW..
ACIGAYMSMHITASASNLYMENNWLWTADHDIDDSSNTQITIFSGRGLYVESTAGTFWFV
ECIGSFMQLHITKSSSG.YFENVWLWTADHELDQPDHAQIDIYNGRGMLVES.QGPVWLV
ECKGAFIGIHVAKGSSP.YIQNVWELGLRDHIAENFSGGTSHRRERWNFGPIRRNATCLY

ANGAEHHTMYQYQFNAAQDIFAGYIQTETPYFQPTPIAPLP....YVSSSKYSDPVYSSS
GTAVEHHTLYQYQFANTQNIYAGVIQTETPYYQPNPDAPTP....FNVNTALNDPNFATS
GTASEHSQLSQYQFQGAKDIWYGAIQTETPYYQPNPKANVP....FKKNDKFSDPDMSNT
PIGSGHWWLYQLNLHNAANVVVSLLQAETNYHQGANTQQIPPAPWVANVGTWGDPDFSW.

QTS ....... AWGLRLLDAKNVLIYGGGLYSFFDNYD..VGCSSPTAPNGFRDCQTRILS
CSGSSGRCAEAWGLRIVSSQNILIYAAGLYSFFENNDGNTGCDVALGPE...NCQNNIFD
TS ........ AWAVRIIDSSSIWNYGAGTYSFFDNYSQK..CVVG ...... QNCQEHINE
CNGGDKRCRMGPANFINGGSNIYTYASAAWAFFSGPGQ..GCAQ ...... F.5CQQTIHW
i i i i
IEGS.TSVQAFGFSEVGVEWMVTAAGQDKANWKDNLSVYPTTIGYLSYGF
LEGTLTNINVYNLGTVGVVNQITQNGNVLATSSSNVNAFADVIALFRLASGSGGVTPPPS
IENS.RNVNIFGLSTKASVNMISSGGVGLLKDEDNRSNFCATLGIFAQA
IASTPSNLQAFGLCSKDSVNTLRL.GDGTFINTQNGYTGGWTPGGGDVARYTT
STTKAQSTTFSTIITSSPPKQTGWNFLGCYSDNVNGRTLANQVQVAGGASAMSIEACETA
SESAGYTIAGVEYSGECWCDTKFQNGGGPASDGSAQCTMTCSGAPQETCGGPNRLDVYSL
ATATGSASPPAATGWNFRGCYTDSVNARALIAESVPNGPSSMTIEACQSVCKGLGYTLAG

LEYADECYCGNSLANGATIAPDGNAGCNMNCAGNAAETCGGPNRLDIYSYGQANGTQPL

EXGlp with other B-l,3-glucanases from the same novel family.

93

60

120

360

480

600

660

720

780

1020

1079

residues of this motif are also conserved in a comparable position (NEKGELS) in
BGN13.1 (de la Cruz et al., 1995). However, in the barley and yeast B-l,3-glucanases
this motif is located toward the C-terminus, whereas in EXGlp and BGN13.] it is closer
to the N-terminus. B-l,3-Glucanases from other plants, fungi and bacteria do not have

this motif at the active site (Keitel et al., 1993; Mackenzie et al., 1997).

EX GI p has two copies of a motif shared with other proteins that interact with
polysaccharides

EXGI p contains two imperfect copies of a 23-amino acid sequence
(NVKDYGAKGDGVTDDSDAFNRAI and SARGAGATGDGVTDDTRAVQAAV),
starting at amino acid 72 (equivalent to amino acid 29 in the mature protein) and 425,
respectively (Figures 26, 28). The two copies are 48% identical. A related amino acid
sequence is found in recently sequenced B-1,3-exoglucanases from T. harzianum and
from A. quisqualis, and in several other proteins, including viral endoneuroarninidase
(sialidase), viral neck appendage protein, bacterial mannuronan epimerase, and several
plant and bacterial polygalacturonases (Figure 29). In all enzymes that have a single copy
of this motif, it is found near the N-terminus. New members of the EXGlp family from
T. harzianum and from A. quisqualis contain two copies of this motif located at positions
similar to those of EXGlp (Figures 26, 27). Of the five known mannuronan epimerases
of Azotobacter vinelandii, two contain a 385-amino acid duplicated module, and each
module contains a copy of the EXGlp motif in its N-terminal part (Ertesvag et al., 1995;
Figure 29). Interestingly, B-1,3-glucanases of the EXGlp family are much bigger than

other [3-1,3-glucanases, and two copies of the conserved motifs may belong to two

94

EXGlp, C. carbonum, Motif l

EXGlp, C. carbonum, Motif 2

EXG, Trichoderma, Motif l (Cohen-Kupiec, 1998)
EXG, Trichoderma, Motif 2 (Cohen-Kupiec, 1998)
EXG, Ampelomyces, Motif l (Rotem et al., 1998)
EXG, Ampelomyces, Motif 2 (Rotem et al., 1998)
NE, Phage Kl (Gerardy-Schan et al., 1995)

NE, Phage Kl (Long et al., 1995)

NE, Phage KlF (Petter and Vimr, 1993)

NAP, Phage PZA (Paces et al., 1985)

NVKDYGAKGDGVTDDSDAFNRAI
SARGAGATGDGVTDDTRAVQAAV
NVKDYGAKGDGVTDDTAAINNAI
SVRSAGATGNAVTDDTAALQSVI
NVKLLGAKGDGVTDDTAAINNAI
SLKDAGAVGDGVTDDTAMIQKAI
SLKDFGAKGDGKTNDQDAVNRAI
SLKDFGAKGDGKTNDQDAVNAAM
DARGWGAKGDGVTDDTAALTSAL
SVKTYGAKGDGVTDDIKAFEKAI

ME, Azotobacter, Motif 1 (Ertesvag et al., 1995)NVKDFGALGDGVSDDTAAIQAAI
ME, Azotobacter, Motif 2 (Ertesvag et al., 1995)NAKDFGALGDGASDDRPAIQAAI

PG, Solanum (Kalaitzis et al., 1995)

PG, Prunus (Lester et al., 1994)

PG, Actinidia (Atkinson and Gardner, 1993)
PG, Arabidopsis (Quigley, 1993)

PG, Gossypium (John and Petersen, 1994)
PG, Brassica (Petersen et al., 1996)

PG, Erwinia (He and Collmer, 1990)

Consensus

NVQNYGAKSDGKTDSSKAFLNAW
NVASLGAKADGKTDSTKAFLSAW
NVDDFGAKGDGR-DDTKAFEKAW
DVKASGAKGDGKTDDSAAFAAAW
VVAKFGAKADGKTDLSKPFLDAW
SVSNFGAKGDGKTDDTQAFKKAW
NITQYGAKGDGTTLNTSAIQKAI

NVKXFGAKGDGKTDDTXAFXXAI
S R Y V W

Figure 29. Alignment of the amino acid sequences of the two 23-amino acid motifs
of EXGlp with the corresponding related sequences of other proteins. Conserved
amino acids are indicated by shading. Abbreviations: EXG, [3-1,3-exoglucanase; NE,
endo-N-acetylneuraminidase; NAP, neck appendage protein; ME, mannuronan C-5-
epimerase E1; PG, polygalacturonase. The dash in the sequence of Actim'dia PG

represents an introduced gap of one amino acid.

95

domains within these enzymes, each motif being at the N terminus of the corresponding
putative domain. However, we could not find any other internal similarities between the
putative first and second domains of EXGlp.

None of the other 16 known C. carbonum cell wall degrading enzymes, including
PGNlp (encoding endopolygalacturonase) (Scott-Craig et al., 1990), PGXl p
(exopolygalacturonase) (Scott—Craig et al., 1998a), and MLGl p (which has activity
against both B-1,3 and B-1,3-B-1,4 glucans) (Gdrlach et al., 1998), contains any
detectable sequences related to this motif. This motif is also not apparent in BGN13.1 of
T. harzianum, although amino acid boxes GDG and NVKD occur at positions
corresponding to the EXGlp motif first and second copies, respectively (Figure 28).

Most of the proteins with the EXGlp motif, including both repeats of EXGI p,
also have the sequence YVPXG seven to 22 amino acids downstream from the 23-amino
acid motif.

Although the proteins sharing the conserved 23-amino acid motif of EXGlp have
different enzymatic activities, they have in common the fact that they all interact with
polysaccharides. B-l,3-Glucanase, N-acetylneuraminidase, polygalacturonase, and
mannuronan epimerase catalyze the structural modification of polysaccharides, whereas
the viral neck appendage protein is involved in binding to cell surface carbohydrates
(Villanueva and Salas, 1981). Therefore, these conserved motifs may be important for
binding to polysaccharides. The sequence similarity between two of these types of
enzymes, polygalacturonases and mannuronan epimerase, has been noted in relation to
the fact that polygalacturonases can bind alginate, the substrate of mannuronan epimerase

(Ertesvag et al., 1995; Gupta et al., 1993).

96

B-1,3-Glucanase A1 from Bacillus circulans has two imperfect repeats of 100
amino acids at its N-terminus that are involved in binding to the substrate (Watanabe et
al., 1992). The sequences of these repeats appear to be unrelated to those of the EXGlp

repeats.

Effect of carbon source on expression of EXGI

An internal 650-bp BglII fragment was used to probe a blot of RNA isolated from
C. carbonum grown on media with different carbon sources. The 2.7-kb mRNA
corresponding to EXGI was detected when C. carbonum was grown on laminarin, a
laminarin-rich extract of extract of the seaweed Laminaria saccharina, oat bran, and
maize cell walls, but not when the fungus was grown on 2% sucrose as a carbon source
(Figure 30). Like other cell wall degrading enzyme genes of C. carbonum and other
fungi, EXGI mRNA is more abundant when grown on its substrate and is scarce when

grown on sucrose (e.g., Scott-Craig et al., 1990; Sposato et a1, 1995).

Materials and Methods

Fungal culture growth and maintenance

Conidia of the strains of C. carbonum race 1 isolate SB111 (ATCC #90305) were
stored at -80° C in 25% glycerol and grown on V8 juice agar plates. For DNA extraction,
mycelial agar plugs (ca. 0.5 cm2) were inoculated into 1-L Erlenmeyer flasks containing
125 ml modified Fries’ medium (Scheffer and Ullstrup, 1965). Cultures were incubated

at room temperature (21-23° C) without shaking for four days. For RNA extraction, the

97

 

2.7kb— ' w

 

Figure 30. Northern blot analysis of EXGI expression in C. carbonum grown on
different carbon sources. Total RNA was extracted from C. carbonum grown for 8 days
on 2% (w/v) sucrose (lane 1), 1% oat bran (lane 2) or for 12 days on 1% maize cell walls
(lane 3), 1% ground Laminaria saccharina (Sigma, St. Louis, MO) (lane 4), or 1%
laminarin (lane 5) as a sole carbon source. The blot was probed with Bng fragment
internal to EXGI (upper panel). The lower panel shows the ethidium bromide staining of
the ribosomal RNA band from the same gel. The age of culture for each substrate was
the time of maximum expression of EXGI for the given substrate based on a time course
experiment (data not shown).

98

same culture conditions were used except that the media contained 2% (w/v) sucrose, 1%
maize cell walls, 1% oat bran, 1% laminarin or 1% ground Laminaria saccharina (Sigma,
St. Louis, MO) as sole carbon source, and the cultures were incubated for 8 or 12 days.

Maize cell walls were prepared as previously described (Sposato et al., 1995).

Nucleic acid manipulations

Fungal genomic DNA was isolated as described by Pitkin et a1. (1996). Southern
blot analysis was done as described (Sambrook et al., 1989). Total C. carbonum RNA
was isolated as described (Pitkin et al., 1996), and RNA blotting was done following
standard techniques (Sambrook et al., 1989).

A C. carbonum genomic library (Scott-Craig et al., 1990) was screened using a
BglII fragment internal to the partial EXGI clone described by Schaeffer et al. (1994). A
C. carbonum cDNA library constructed from the fungus grown on maize cell walls as a
carbon source (Pitkin et al., 1996) was screened using the same BgIII fiagment.
Screening of the cDNA and genomic libraries was done by standard methods (Sambrook

etaL,l989)

DNA sequencing

Sequencing was done using nested exonuclease III deletions made using the
Erase-a-base kit (Promega, Madison, WI, USA). The sequence of both strands was
determined by automated fluorescent sequencing at the MSU-DOE-PRL Plant
Biochemistry Facility using an Applied Biosystems (Foster City, California) Catalyst 800

for Taq cycle sequencing and an Applied Biosystems 373A Sequencer for analysis of the

99

products. DNASIS software (Hitachi, San Bruno, CA) and the Wisconsin Computer
Genetics Software (Wisconsin Package, 1996) were used to analyze the data. Database

searching was done using BLAST (Altschul et al., 1993; Altschul et al., 1997).

100

APPENDICES

101

APPENDIX A

AMINO ACID SPECIFICITY OF THE FUNCTIONAL DOMAIN A OF THE

COCHLIOBOLUS CARBONUM HC-TOXIN SYNTHETASE

Introduction

Non-ribosomal peptides (and some linear peptides) are synthesized by a class of
enzymes known as non-ribosomal peptide synthetases. These enzymes catalyze
thioesterification of their amino acid substrates via ATP/PP; exchange and the formation
of amino acid adenylate intermediates (reviewed by Kleinkauf and von Dohren, 1990;
Kleinkauf and von Dohren, 1996; Cane et al., 1998). Multifunctional non-ribosomal
peptide synthetases, those that catalyze activation of more than one amino acid, are
organized into units, one for each amino acid substrate. Within these units, conserved
amino acid activating domains are approximately 600 amino acids in length and contain
highly conserved motifs known or believed to be involved in aminoacyl adenylation, ATP
binding, thioester binding, and epimerization (Stachelhaus and Marahiel, 1995; Kleinkauf
and von Dohren, 1996; Cane et al., 1998) (see also Chapter 1). HC-toxin synthetase
(HTS) from C. carbonum contains four such domains, which is consistent with the
number of amino acids in HC-toxin. The amino acid specificity of these putative
functional domains is not known, although they are presumed to contain motifs involved

in substrate selection.

102

Results and discussion

Domain A of H T S activates L-Proline

In order to understand the structure/function relationship of the domains of HTS
and, possibly, the putative Chlamydocin synthetase, expression of individual domains and
subsequent assaying of their amino acid-dependent ATP/PP, exchange activity was
attempted. In the experiments utilizing the E. coli/pET expression system and
Baculovirus system (K. Akimitsu, unpublished data), the desired protein domains were
expressed and their identity was confirmed by Western blotting. However, in this study it
was found that unlike the native HTS, the expressed proteins were in the insoluble
fraction and no ATP/PP, exchange activity could be detected (data not shown). The
expressed proteins could be solubilized only by harsh methods (6M urea or 1% SDS)
which caused them to lose enzymatic activity irreversibly (data not shown). Chaperonins
GroES and GroEL were expressed in E. coli simultaneously with the desired HTS
domains as an attempt to achieve native protein folding/activity, but no change in
solubility/activity was found (results not shown).

The expression system for individual HTS (and other possible peptide synthetase)
domains described here utilized as a host organism a Tox2- strain of C. carbonum (strain
243-10) that lacks the whole HT SI -containing TOX2 chromosome. Expression was
driven by a strong constitutive promoter from the Cochliobolus heterostrophus GPDI
gene encoding glyceraldehyde-3-phosphate dehydrogenase (Van Wert and Yoder, 1992).
The native HT SI promoter could not be used because it may be regulated by TOX2 -

specific regulators which may be missing in Tox2_ isolates. The entire 5’ S-kb region of

103

Figure 31. Design of pGPD18, the vector for HTS domain A expression driven by
the GPDI promoter. The resulting construct, pGPD18, contains the PGNI internal
fragment that can be linearized using a unique internal NotI site for homologous
integration into the Tox2' genome. Because SalI is the only site in pGPDA suitable for
cloning, pPGE#3 plasmid had to be made so that the PGNI fragment could be available
as a SalI/Sall fragment. Distances are given in kilobases. Restriction enzyme sites used
in constructing the plasmids are shown in bold. P, GPDI promoter from Cochliobolus
heterostrophus; PG, C. carbonum PGNI internal fragment (Scott-Craig et al., 1990);
HYGR, the cassette conferring hygromycin B resistance (C. heterostrophus promoter 1
driving the expression of the hph gene encoding hygromycin phosphotransferase; Schafer
et al. 1989); A, ApaI; B, BamHI; C, ClaI; E, EcoRI; H, HindIII; Hp, HpaI; K, Kpnl; N,
NotI; P, PstI; S, SalI; Sc, SacI; Sm, SmaI; Sn, SnaBI; Sp, SpeI; X, Xhol; Xb, XbaI.

104

lkb

Skb HTS] segment containing N-terminal
HTS translational start, Domain A, and the
interdomain sequence downstream was
blunt-end ligated into pGPDhyg plasmid.

 

 

KSmBXbS

cut out EcoRI/EcoRl fragment
Ligate into EcoRI-cut pBS KS

 

 

 

 

 

 

 

 

 

 

pPGE#3 pGPDA
0.9 1.2 0.7 5 2.5
4—94—9 3 :4 if :
pBS KS P r—Dfim’ A [Xb/SmHYG" pUCl9
: [EB/Hp]
K A x 8 CH E E 8 H

 

   
   
 

cut out Sall/Sall fragment cut with Sall

ligate
check orientation

pGPDl8
0.7 5 0.9 2.5

A LA LA ‘4 ‘
V 7‘

[KIT/Sn]
P l Dom.A PG HYGR pUC19
[BL-1p] N >

E SXbBSmK ScEHS H

 

 

 

 

 

 

Figure 31. Design of pGPD18, the vector for HTS domain A expression driven by
the CPD] promoter.

105

HTS] containing the native translational start site was fused to the GPDI promoter
(Figure 31, construct pGPDA, T. Black, unpublished results). For selection of
transformants, the cassette conferring hygromycin B resistance was used (Schafer et al.,
1989). This cassette is a SalI/HindIII fragment containing the C. heterostrophus promoter
1 driving the expression of the hph gene encoding hygromycin phosphotransferase. To
increase transformation efficiency, the HTS domain expression vector had to be
introduced into the Tox2— recipient strain via homologous integration. For this purpose,
a fragment of an unrelated non—essential C. carbonum gene (PGNI, encoding
endopolygalacturonase, Scott-Craig et al., 1990) was engineered into the pGPDA
construct (Figure 31) utilizing SalI, the only restriction site in pGPDA linkers that is
unique throughout the construct and thus available for cloning.

The resulting construct, pGPD18, was linearized by cutting with NotI, a unique
restriction enzyme site internal to the PGNI fragment and introduced into wild type C.
carbonum 243-10 (Tox2_) strain by transformation of protoplasts as described by Scott-
Craig et a1. (1990). Five hygromycin B resistant transformants (551-1 through -5) were
obtained and characterized by DNA blot analysis (Figure 32A). Figure 32B depicts the
wild type PGNI locus and Figure 32C and D shows the predicted maps for single or
multiple integration events of pGPD18 into PGNI, respectively. The DNA gel blot
analysis of transformants T551-3 and T551-5 was consistent with a single insertion event,
and transformants T551-1, T551-2, and T551-4 - with multiple insertion events.
Additional DNA blot analysis (not shown) confirmed that the integrated pGPDA18
construct was intact in all transformants, containing the entire GPDI/HT S1 (Domain A)

fusion.

106

Figure 32. Southern blot analysis of pGPD18 integration into the PGNI locus of the
Tox2_ strain 243-10. (A) DNA blot of wild type Tox2— strain 243-10 (lane 6) and five
transformants, 551-1 through 5 (lanes 1 - 5). Genomic DNA was digested with HindIII
and the filter was probed with the 0.9-kb KpnI/Sacl PGNI internal fragment. (B)
Restriction map of the wild type PGNI locus (Scott-Craig et al., 1990). (C) and (D)
Predicted restriction maps of the PGNI locus with a single and multiple (double)
insertion of the pGPD18 expression vector, respectively. Distances are given in
kilobases. GPDI promoter/HT S1 domain A fusion construct is indicated by shaded
boxes, PGNI fragment introduced as part of pGPD18 vector is indicated by solid black
boxes, and genomic copy of the same fragment is indicated by dark gray boxes. P, GPDI
promoter (Van Wert and Yoder, 1992); PG, C. carbonum PGNI gene internal fragment
(Scott-Craig et al., 1990); HYGR, the cassette conferring hygromycin B resistance
(containing C. heterostrophus promoter 1 driving the expression of the hph gene
encoding hygromycin phosphotransferase; Schafer et al. 1989); E, EcoRI; H, HindIH; K,
KpnI; N, Notl; S, Sal]; Sc, SacI.

107

are.” as; has 2: a 2.8. ~22 2.. as 8:385 ”Smog a are; 83 E235 .5. 25E

 

 

 

 

 

 

 

 

 

 

 

= a a. m m = .3 a. = m = as s. =
\ o.— <.§5 E AGE. «PE 0.— <.aoa .— 20:; «our: 0..
A 2 \2x A XSX Z 111 2 moat W Max 3 X 2 xaaxaov
_ _
m: _Iw.~|_ .131.
d
m a a. = m n ma 5. m
_ x . _ _ L 7; _
on €58 .— 262; :95 o.—
2 as m S Z 2 as so
_ _
mo. _Ilw._L
.0
m 8 m m . l 2
_ z__ _ .‘ D '2
t
1 I I r . led
2 no as 3.1 a. I E
_ a: L
.m e m 9 m N w e. .<

108

Transformants 551-1, -2, -3, and -5 were further analyzed by Northern blot
(Figure 33) and the 6.6 kb mRNA hybridizing to the HT S1 domain A internal fragment
was detected in all of them. To test these strains for the presence of HTS domain A
protein and any ATP/PP, exchange activity, cultures of T551-1, T551-3, SBlll (wild
type Tox2+), and 243-10 (wild type Tox2? were grown in parallel, and protein was
extracted from mycelial pads and partially purified by ammonium sulfate fractionation
and anion exchange chromatography (Walton and Holden, 1988). ATP/PP; exchange in
wild type strains and transformants T551-1 and T551-3 was compared in the fractions
containing maximum activity. The results are given in Table 4. Expressed protein was
also analyzed by Western blotting using anti-HTS-l polyclonal antibodies (Scott-Craig et
al., 1992) (Figure 34). These results show that HPLC fraction 17 of protein preparations
from Tox2+ wild type strain SB] 1 1, as well as from transformants T551-1 and T551-3,
contained protein recognized by anti-HTS-l antibodies. The same fraction contained L-
Pro-dependent and D-Ala-dependent ATP/PPi exchange activity in SE] 1 1, whereas in
both transformants it contains only L-Pro-dependent ATP/PPi exchange activity. No
ATP/PP, exchange activity or protein recognized by anti-HTS-l antibodies was detected
in strain 243-10 (wild type Tox2? which was used as a recipient strain for transformation
T551.

In conclusion, expression of domain A of HTS driven by the CPD] promoter
from C. heterostrophus in a Tox2” strain was confirmed by Northern and Western

blotting, and this domain was shown to have L-Pro-dependent ATP/PP, exchange
activity. This result is consistent with the fact that, according to the partial peptide

sequence obtained from proteolytic fragments of HTS (Panaccione et al., 1992),

109

 

Figure 33. Northern blot analysis of transformants bearing pGPD18 expression vec-
tor. Total RNA was extracted from four—day-old mats and separated on a 1% agarose gel
containing formaldehyde and transferred to nitrocellulose filter. Upper panel: The filter
was probed with a DNA fragment internal to domain A of HT SI . Lower panel: same gel
stained with ethidium bromide before the transfer, showing the ribosomal RNA bands. Lanes:
1 - 4, transformants 551-1, —2, -3, and -5, respectively; 5, 243-10 (Tox2' recipient strain).

110

Table 4. AT P/PPi exchange activity in C. carbonum bearing pGPD18 expression
vector. L-Proline and D-alanine-dependent ATP/PPi exchange was measured in the peak
fraction (fractions 15 and 17) of anion exchange HPLC (Walton and Holden 1988). One
ml of each protein preparation was chromatographed, and 25 pl of each l-ml fraction
assayed. The reaction mixes contained ca. 105,000 cpm [32P]pyrophosphate per reaction
(total volume 125 pl).

chm incorporated-

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Isolate Fraction L-Pro-dependent D-Ala—dependent
ATP/PP, exchange ATP/PPi exchange
SB 1 1 la 1 5 28 3 .4
1 7 80 61
243-10b 15 5 3
17 3 .5 0.8
T551-1 15 14.6 2.8
1 7 22 0.3 5
T551-3 1 5 8.8 2. 1
17 9.6 0.6
a wild type Tox2+ strain

b wild type Tox2-strain

111

 

T551-3 T551-l 243-10 $8111

123456789

 

kDa 1 2 3 4
—.7_
206—1“-

117—

 

 

 

Figure 34. Western blot analysis of protein extracts from transformants bearing
pGPD18 expression vector. (A) Protein extracts were purified through ammonium
sulfate precipitation, desalted, resolved on a 6% acrylamide SDS-PAGE gel, and
transferred to nitrocellulose as described by Scott-Craig et a1. (1992). The filter was
probed with anti—HTS-l antiserum (Scott-Craig et al., 1992). Lanes: 1, SB111 (wild type
Tox2+); 2, 243-10 (wild type Tox2_); 3, T551-3; 4, T551-1. (B) Anion exchange HPLC
(Walton and Holden, 1988) fractions containing ATP/PPi exchange activity (fractions 15
and 17) were resolved on a 6% acrylamide SDS-PAGE gel, transferred to nitrocellulose,
and treated with anti-HTS-l antiserum. Lanes: 1, S81 11 preparation not fractionated by
HPLC; 2 and 3, T551-3 fractions 17 and 15; 4 and 5, T551-l fractions 17 and 15; 6 and 7,
243-10 fractions 17 and 15; 8 and 9, SB111 fractions 17 and 15.

112

L-Pro-dependent activity resides in domains A or B, and L-Ala and D-Ala—dependent

activity - in domains C and/or D.

Sequential disruption of H TS] functional domains

Another approach undertaken to determine the amino acid specificity of HTS
functional domains was to disrupt HTS] gene in three different ways to eliminate
domains D, C and D, or B, C and D and assess the activity and amino acid specificity of
remaining domains.

Before Tox2+ strains with only one copy of HTS] were discovered (see chapter 2),
to create a strain with only one functional copy it was necessary to disrupt the second one
so that no protein is produced by it. To make the second copy disruption and subsequent
selection easier, for the disruption of the first copy acetamide utilization as sole nitrogen
source was used as a selectable marker, and for the second copy disruption hygromycin B
resistance was used as a selectable marker. Strain 553-11 was created by disrupting copy
1 of HTS] with the C. carbonum transformation vector pCC129 (Panaccione et al., 1992),
and transformants were selected for acetamide utilization. The insertion of pCC129
disrupts HT S1 at the 5' end (at the beginning of domain A).

Strain 553-11 was used as a recipient for three different step-wise disruptions of
the second copy of HT SI (Table 5). Three transformation vectors were designed to
contain the HTS] sequences for the homologous integration at the beginning of the
domains B, C, or D (vectors pHTSl, pHTSZ, and pHTS3, respectively). Surprisingly,
very few of the recovered transformants had homologous integration of pCC129 into

copy 2 of HT S1 (Table 5). These transformants were analyzed for the presence of the

113

Table 5. Analysis of the transformants for the step-wise disruption of HTS].

 

 

 

 

 

Transformation Domains Number of Transformants Antibody ATP/PP;
number and remaining Hng with homologous binding to exchange
vector transformants integrations into protein on activity

analyzed HTS] copy 2 Western
T593 A 20 3 none or very L-Pro °
pHTSl weak anti-

HTS-1'

T594 A and B 29 3 anti-HTS-la’b L—Pro °
pHTSZ
T595 A, B, and 26 none n/a n/a
pHTS3 C

 

 

 

 

 

 

3 Weak antibody hybridization to the ca. 150 kDa band.
b Weak antibody hybridization to the ca. 200 kDa band.

° Weak exchange activity detected by ATP/PP. assay (ca. 1/5 to M: of the wild type
control).

114

 

truncated HTS protein and for HTS activity (summarized in Table 5). When strain
164R10 was found to harbor only one copy of HTS] (see Chapter 2), the same
transformation vectors were used for the step-wise disruption in this strain, but none of
the transformants had homologous integration of the vector.

Overall, only very low levels of the truncated HTS protein products were
produced by these transformants, if at all. The transformants showed some L-Pro-
dependent ATP/PPi exchange activity, which is expected because this activity was shown
to reside in domain A by the expression experiments (see above), and this domain
remains intact in all of the transformants. Based on the sequence comparisons,
Stachelhaus and Marahiel (1995) and Kleinkauf and von Dohren (1996) have suggested
that domain A activates L-Pro and epimerizes it (converting to D-Pro), domain B
activates L-Ala, domain C activates D-Ala, and domain D activates Aeo. The results
described here confirm that domain A activates L-Pro. The L-Ala-dependent activity of
the second domain may be harder to detect, since there always is more background L-Ala-

dependent activity. The function of domains C and D remains unconfirmed.

l15

APPENDIX B

CC115, A TRANSPOSASE-LIKE SEQUENCE FROM COCHLIOBOLUS

CARBONUM

In the C. carbonum T 0X2 locus, immediately downstream from HT S1, there is a
segment of TOX2 -unique DNA that was subcloned as CC62 (Panaccione et al., 1992).
CC115 is a 1.2-kb cDNA clone obtained from a C. carbonum cDNA library by screening
it with CC62. CC115 was sequenced on both strands (Figure 35) and was not found to
contain a contiguous open reading frame. This means that either the corresponding gene
is nonfunctional and is not translated, or the cDNA is chimeric. However, the deduced
amino acid sequence of the part of one short open reading frame shows high similarity to
the transposase sequences from the transposons found in various filamentous fungi
(Figure 35). Another putative transposon from C. carbonum, named Fccl (Panaccione et
al., 1996), also shows sequence similarity to CC115. However, although Fccl appears
especially prevalent on the T 0X2 chromosome, it is also present on most of the
chromosomes in the C. carbonum genome. The CC115 DNA sequence, on the other
hand, is Tox2+-unique. Chromosome-specific repeated sequences from supemumerary
chromosomes are known in Nectria haematococca (Enkerli et al., 1997; Covert, 1998)
and Colletotrichum gloeosporioides (Masel et al., 1993; 1996). In both of these fungi,
there are also repeated sequences that are common to both essential and supemumerary
chromosomes. In addition, in C. gloeosporioides, repeated sequences that are present on

the essential chromosomes but not on the supemumerary chromosome have also been

116

l CGTTGCTGTCGGCCATTAGATCAAACTAGGTGGAAGGAATTGCTGTTGGTGAGCATGACG
* V E G I A V G E H D
61 TGTTTATAACATCAACCCCGTCAGGTTGGAGCAATAATGCCGTTAGGCTAGCATGGCTTG
V F I T S T P S G W S N N A V R L A W L
121 AGCAAGTTTTTGATCGCTGCACGAGGAACAAACCAGGGAGATGGCGATTACTTATCCTTG
E Q V F D R C T R N K P G R W R L L I L
181 ATGGCCATGGATCTCACGTCACGCCCGAGTTTATTGAGTATTGCAATCGCCATAGAATAC
D G H G S H V T P E F I E Y C N R H R I
241 TCCTTATGGTGTTTCCTCCTCATTCAACTCACACACTGCAACCGCTTGATGTGGTGATGT
L L M V F P P H S T H T L Q P L D V V M
301 TTAAACCACTCTCCACCAGCTACTCAAATGAGCTCACTAATCACCTCTACAACGCCCAAG
F K P L S T S Y S N E L T N H L Y N A Q
361 GCCTCGTCTCAGTCAAGAAGGGAGACTTTTTTCCGCTGTTCTGGCGAGCCTGGAGCTCAT
G L V S V K K G D F F P L F W R A W S S
421 CCTCTACTAAAAAATAATATCTTGAAGGCCTTTAGCGCCACTGGTATATGGCCAGCAGAT
S S T K K * Y L E G L *
481 CCCGACGTTATACTCAAAAGTTTAGTTCAACACCTGATAAGAGCCACCGCAAGCGATCTC
541 GGCTCTCTCCAAGTGATTGGAATCACCTGAGGCAGCTAGTATGAGAAGCTGCTGAAGATG
601 GAGCTGAGAGTGGAGTTAAAAAGCTTAGTGCCCTACTCCATCATCTCCAGGTTCAGAATG
661 AGCTATTGCGTCATGAGATGGAGGGATTGAGAGCAGCTCTTTCACAAAAACAGAAGCATA
721 AAGGCAAGGGCAAAGCTCTAAATCTTCAATAACGCAAGGAGTATCATGGCGGAGCGGTCT
781 TCTGGTCACCTCGTAAGTTCCGCGAAGCTCGAGCTCGAGAAGCAGTTCGTGAGCGCGAGG
841 AAGTAGAGGAGAAACTCCAGAAAGCACAGGCTAAGAAGGACTGCGAGGAGACTCAGTTGT
901 GGCGTCAAGTTGAGCGCGAGGAGAAGCGCACTGAACGATTGAGACTTAATGAGATGCGTA
961 AGCTTGAGCGAGCTGAGAAAGCAGCTGAACGCGCGCGTAAAAAAGAAGCTCGCAACACTG
1021 AAAAATCTCAACACCAAGCTCAAAAGCGTAAGCGTACAGCCTCACGAGTGCCCACCTCTA
1081 AGAACAAGCGTCAAAAACTATCAGTGCTGGATGGAGCTCGCGATGGAGCTGCATCTACTT
1141 CATCATCTATCCCGCCAAAGATCACGACGCGAGGCCGCAGCGTTAACGTCCCGCAGAAAT
1201 TTAGATAGCACAAACTAACCACAAGTATTTGGTTACTGTAAATCCAAGTTTTATATATAT
1261 TAAAAAAAAAAAAAAAA

Figure 35. Nucleotide and partial deduced amino acid sequences of CC115. The part
of the open reading frame that is similar to fungal transposase sequences is highlighted.

117

C. carbonum CC115 EGIAVGEHDV.FITSTPSGWSNNAVRLAWLEQVFDRCTRN.KPGRWRLLILDGHGSHVTP
Magnaporthe 1 PTDLSPFDNWQF.HATBNGWTNNQTAIEWLKKVFIPYTQPLTP.EKRLLVLDGHGSHITD
Magnaporthe 2 EGG.LP.DTWRL.KPTVNGWTDNETGLDWVQ.HFDNHTKSRTKGVYRMLVLDGHGSHRSP

A. nidulans TG..LP.PDWRF.EISTNGWTTNEISLRWLQKQFIPSTEHRTRGRYQLLVLDGHGSHLTP
Botrytis PIKLDNYEGWEF.TATDNGWTTDSTGLEWLKEVFIPQSAPTRPKEARLLVLDGHGSHETT

A. niger EGQSIP.PTWRF.EVSDNGWTTDKIGLRWLQKHFIPLIRGKSVGKYSLLVLDGHGSHLTP

C. carbonum Fot ..HSLP.PDWTI.GVSENDWKTDELGVEWV.KHFNQHTVTRTAGVYRLLILDGHSSHATP
Talaromyces ..D.LP.DDWRI.NISDNGWTTDQIGLEWLKTHFIPYINGRTVGKYRMLILDGHGSHLTP
Nectria EEFKEI.ADWYY.ITSPNGWTDDHIGVEWLERVYLPQTMPADDSDARLIILDGHGSHATD

C. carbonum CC115 EFIEYCNRHRILLMVFPPHSTHTLQPLDVVMFKPLSTSYSNELTNHLYNAQGLVSVKKGD
Magnaporthe 1 EFMLLCLQNNIQLLYLPPHSSHVLQPLDLSVFGPLKEAYRRQL.GFVSQFCCSTVIGKRN
Magnaporthe 2 EFEGYCKDYNIIPLYLPAHSSHLTQPLDVGVFNVLKRAYGQKI.NDFIRAH.ITNISKVD

A. nidulans EFDQICTDHNIIPLCMPAHSSHLLQPLDIGCFAVLKRSYASLV.DQKMRLG.ISHIDKLD
Botrytis QFMLECFKNNIHLLFLPPHTSHVLQPPDLSIFSPLKKEYRYHL.NTLDSLADSTPIDKRN

A. niger EFDQSCAENEVIPICMPAHSSHLLQPLDVGCFSVLKRTYGGMV.QKQMQYG.RNHIDKLD

C. carbonum Fot EFDQFCTENKIITLCMPSHTSHLLQPLDVSCYSTLKRAYGREI.EELARHG.VYHVDKID
Talaromyces EFDHICTENNIIPVCMPPHSSHLLQPLDVGCFAVLKRHYGQLV.EQRMRLG.FNHIDKMD
Nectria EWMATCFLNNVYCCYLPAHCSHGLQPLDNGVFNASKAAYRREL.ENFASLTDSTPMDKVN

C. carbonum CC115 FFPLFWRAWSSSSTKK'YLEGL*R
Magnaporthe 1 FLLCYRKARLKAFIAKTIQSGW.
Magnaporthe 2 FFLAFAAAYKKSMTKENMAGGF.

A. nidulans FLAAYPQARISTFKLDTIRNSF.
Botrytis FLACYQKARLKALTLRNITSGW.

A. niger FLEVYPKAHQCALSKSNIISGF.

C. carbonum Fot FLTVYTRIRPTAFTQQNIQAGF.
Talaromyces FLTAFPQARTVAYRAQTIRNSF.
Nectria FIRAYAKARRVGMTEKNILSGW.

wzuoauxwwau

Figure 36. Comparison of the partial predicted amino acid sequence of C. carbonum
CC115 with transposase sequences from filamentous fungi. Only the portion of
CC115 that shows similarity to the transposase sequences, and the corresponding amino
acid sequences of fungal transposases are shown. The sequences belong to the middle
parts of the corresponding transposases, and the N-terminal and C-terminal parts are not
shown. Highly conserved amino acids are highlighted. Magnaporthe l, Magnaporthe
grisea (Kachroo et al., 1994); Magnaporthe 2, Magnaporthe grisea (Farman et al., 1996);
A. nidulans, Aspergillus nidulans (Emericella nidulans) (Kupfer et al., 1997); Botrytis,
Botrytis cinerea (Botryotinia fuckeliana) (Levis et al., 1997); A. niger, Aspergillus niger
var. awamori (Amutan et al., 1996; Nyyssonen et al., 1996); C. carbonum Fot, C.
carbonum transposase (Fotl-like) sequence (Panaccione et al., 1996); Talaromyces,
T alaromyces stipitatus (Cummings et al., 1998); Nectria, Nectria haematococca (Enkerli
etaL,1997)

118

found, and at least one of them (Cng) appears to be a transposable element (Masel et al.,
1993; 1996; He et al., 1996). Thus, in this fungus there are three sets of repeated
transposon-like sequences: those that are unique to the supemumerary chromosomes,
those that are unique to the essential chromosomes, and those that are common
throughout the genome. The existence in C. carbonum, N. haematococca, and C.
gloeosporioides of the chromosome-specific repeated sequences and/or transposable
elements that are unique to the supemumerary chromosomes provides an additional
argument in favor of the horizontal movement of the supemumerary chromosomes

(horizontal movement of the TOX2 module in the case of C. carbonum).

119

 

BIBLIOGRAPHY

120

BIBLIOGRAPHY

Ahn, J.-H., and Walton, J. D. 1996. Chromosomal organization of TOX2, a complex
locus controlling host-selective toxin biosynthesis in Cochliobolus carbonum.
Plant Cell 8:887-897.

Ahn, J .-H., and Walton, J. D. 1997. A fatty acid synthase gene in Cochliobolus carbonum
required for production of HC-toxin, cyclo(D-prolyl-L-alany1-D-Alany1-L-2-
amino-9,10-epoxi-8-oxodecanoyl). Mol. Plant-Microbe Interact. 10:207-214.

Ahn, J .-H., and Walton, J. D. 1998. Regulation of cyclic peptide biosynthesis and
pathogenicity in Cochliobolus carbonum by T OXE, a gene encoding a novel
protein with a bZIP basic DNA binding motif and four ankyrin repeats. Mol. Gen.
Genet., in press.

Alexopoulos, C. J ., and Mims, C. W. 1979. Introductory mycology. Third edition. John
Wiley and Sons, NY.

Altschul, S. F ., Gish, W., Miller, W., Myers, E. W., and Lipman, D. J. 1993. Basic local
alignment search tool. Nature Genet. 3:266-272.

Altschul, S. F ., Madden, T. L., Schéiffer, A. A., Zhang, J., Zhang, Z., Miller, W., and
Lipman, D. J. 1997. Gapped BLAST and PSI-BLAST: a new generation of
protein database search programs. Nucleic Acids Res. 25:3389-3402.

Amutan, M., Nyyssonen, E., Stubbs, J., Diaz-Torres, M. R., and Dunn-Coleman, N. 1996.
Identification and cloning of a mobile transposon from Aspergillus niger var.
awamori. Curr. Genet. 29:468-473.

Annis, S. L, and Panaccione, D. G. 1998. Presence of peptide synthetase gene transcripts
and accumulation of ergopeptines in Claviceps purpurea and Neolyphodium
coenophialum. Can. J. Microbiol. 44:80-86.

Apel, P. C., Panaccione, D. G., Holden, F. R., and Walton, J. D. 1993. Cloning and
targeted gene disruption of XYL], a B-1,4-Xylanase gene from the maize pathogen
Cochliobolus carbonum. Mol. Plant-Microbe Interact. 6:467-473.

Atkinson, R. G., and Gardner, R. C. 1993. A polygalacturonase gene from kiwifruit
(Actinidia deliciosa). Plant Physiol. 103:669-670.

121

Bailey, A. M., Kershaw, M. J ., Hunt, B. A., Paterson, I. C., Charnley, A. K., Reynolds, S.
E., and Clarkson, J. M. 1996. Cloning and sequence analysis of an intron-
containing domain from a peptide synthetase-encoding gene of the
entomopathogenic fiingus Metarhizium anisopliae. Gene 173:195-197.

Ballance, D. J. 1986. Sequences important for gene expression in filamentous fungi.
Yeast 2:229-236.

Ballance, G. M., Larnari, L., and Bemier, C. C. 1989. Purification and characterization of

a host-selective necrosis toxin from Pyrenophora tritici-repentis. Physiol. Mol.
Plant Pathol. 35:203-213.

Barron, G. L. 1985. Fungal parasites of bdelloid rotifers: Diheterospora. Can. J. Bot.
63:211-222.

Bradford, M. M. 1976. A rapid and sensitive method for the quantitation of microgram
quantities of protein utilizing the principle of protein-dye binding. Anal. Biochem.
72:248-254.

Bronson, C. R. 1991. The genetics of phytotoxin production by plant pathogenic fungi.
Experientia 47:771-776.

Brosch, G., Ransom, R., Lechner, T., Walton, J. D., and Loidl, P. 1995. Inhibition of
maize histone deacetylases by HC-toxin, the host-selective toxin of Cochliobolus
carbonum. Plant Cell 7:1941-1950.

Bursnall, L. A., and Tribe, H. T. 1974. Fungal parasitism in cysts of Heterodera. Trans.
Br. Mycol. Soc. 62:595-601.

Canada, S. R., and Dunkle, L. D. 1997. Polymorphic chromosomes bearing the T 0x2
locus in Cochliobolus carbonum behave as homologs during meiosis. Appl.
Environ. Microbiol. 63 :996- 1 001 .

Cane, D. E., Walsh, C. T., and Khosla, C. 1998. Harnessing the biosynthetic code:
combinations, permutations, and mutations. Science 282:63-68

Chang, H.-R., and Bronson, C. R. 1996. A reciprocal translocation and possible
insertion(s) tightly associated with host-specific virulence in Cochliobolus
heterostrophus. Genetics 39:549-557.

Chen, L., Fincher, G. B., and Hoj, P. B. 1993. Evolution of polysaccharide hydrolase

substrate specificity. Catalytic amino acids are conserved in barley 1,3-1,4- and
1,3-B-glucanases. J. Biol. Chem. 268:13318-13326.

122

Ciuffetti, L. M., Pope, M. R., Dunkle, L. D., Daly, J. M., and Knoche, H. W. 1983.
Isolation and structure of an inactive product derived from the host-selective toxin
produced by Helminthosporium carbonum. Biochemistry 22:3507-3510.

Closse, A., and Huguenin, R. 1974. Isolierung und structuraufldarung von chlarnydocin.
Helv. Chim. Acta 57:533-545.

Cohen-Kupiec, R., Broglie, K., Friesem, D., Broglie, R., and Chet,I. 1998. Direct
sequence submission to the EMBL/GenBank data banks, accession # AJ 002397.

Comstock, J. C., and Scheffer, R. P. 1973. Role of host-selective toxin in colonization of
corn leaves by Helminthosporium carbonum. Phytopathology 63:24-29.

Covert, S. F. 1998. Supemumerary chromosomes in filamentous fungi. Curr. Genet.
33:311-319.

Covert, S. F ., Enkerli, J ., Miao, V. P., and Van Etten, H. D. 1996. A gene for maackiain
detoxification from a dispensable chromosome of Nectria haematococca. Mol.
Gen. Genet. 251 :397-406.

Cummings, N. J ., Faulds, C. B., Williamson, G., and Connerton, I. F. 1998. Direct
sequence submission to the EMBL/GenBank data banks, accession # AJ 010981

Darkin-Rattray, S. J., Gurnett, A. M., Myers, R. W., Dulski, P. M., Crumley, T. M.,
Allocco, J. J ., Cannova, C., Meinke, P. T., Colletti, S. L., Bednarek, M. A., Singh,
S. B., Goetz, M. A., Dombrowski, A. W., Polishook, J. D., and Schmatz, D. 1996.
Apicidin: A novel antiprotozoal agent that inhibits parasite histone deacetylase.
Proc. Natl. Acad. Sci. USA 93:13143-13147.

De la Cruz, J., Pintor-Toro, J. A., Benitez, T., Llobell, A., and Romero, L. C. 1995. A
novel endo-B-1,3-glucanase, BGN13.1, involved in the mycoparasitism of
T richoderma harzianum. J. Bacteriol. 1771693 7-6945.

Dickman, M. B., Podila, G. K., and Kolattukudy, P. E. 1989. Insertion of cutinase gene
into a wound pathogen enables it to infect intact host. Nature 342:446-448.

Drenth, J., Low, B. W., Richardson, J. S., and Wright, C. S. 1980. The toxin-agglutinin
fold. A new group of small protein structures organized around a four-disulfide
core. J.Biol. Chem. 255:2652-2655.

Edelmann, S. E., and Staben, C. 1994. A statistical analysis of sequence features within
genes from Neurospora crassa. Exp. Mycol. 18:70-81.

123

Enkerli, J ., Bhatt, G., and Covert, S. F. 1997. Nht], a transposable element cloned from a

dispensable chromosome in Nectria haematococca. Mol. Plant-Microbe Interact.
10:742-749.

Enkerli, J ., Bhatt, G., and Covert, S. F. 1998. Maackiain detoxification contributes to the
virulence of Nectria haematococca on chickpea (Cicer arietinum). Mol. Plant-
Microbe Interact. 1 1:317-326.

Ertesvag, H., Hoidal, H. K., Hals, I. K., Rian, A., Doseth, B., and Valla, S. 1995. A
family of modular type mannuronan C-S-epirnerase genes controls alginate

structure in Azotobacter vinelandii. Mol. Microbiol. 16:719-731.

Farrnan, M. L., Taura, S., and Leong, S. A. 1996. The Magnaporthe griseo DNA
fingerprinting probe MGRS 86 contains the 3' end of an inverted repeat transposon
Mol. Gen. Genet. 251:675-681.

Gavel, Y., and von Heijne, G. 1990. Sequence differences between glycosylated and non-
glycosylated Asn-X-Thr/Ser acceptor sites: implications for protein engineering.
Prot. Engin. 3:433-442.

Gerardy-Schahn, R., Bethe, A., Brennecke, T., Muhlenhoff, M., Eckhardt, M., Ziesing,
S., Lottspeich, F ., and Frosch, M. 1995. Molecular cloning and functional
expression of bacteriophage PKlE-encoded endoneuraminidase Endo NE. Mol.
Microbiol. 16:441-450.

Gocht, M., and Marahiel, M. A. 1994. Analysis of core sequences in the D-Phe activating
domain of the multifimctional peptide synthetase TycA by site-directed
mutagenesis. J. Bacteriol. 176:2654-2662.

Goos, R. D. 1986. A review of the anarnorph genus Helicoma. Mycologia 78:744-761.

Gorlach, J. M., Van der Knaap, 13., and Walton, J. D. 1998. Cloning and targeted
disruption of MLGI, a gene encoding two of three extracellular mixed-linked
glucanases of Cochliobolus carbonum. Appl. Environ. Microbiol. 64:385-391.

Gupta, M. N., Guoqiang, D., and Mattiasson, B. 1993. Purification of endo-
polygalacturonase by affinity precipitation using alginate. Biotechnol. Appl.
Biochem. 18:321-327.

Haese, A., Schubert, M., Herrmann, M., and Zocher, R. 1993. Molecular characterization

of the enniatin synthetase gene encoding a multifimctional enzyme catalysing N-
methyldepsipeptide formation in F usarium scirpi. Mol. Microbiol. 7:905-914.

124

Hammock, L. G., Hammock, B. D., Casida, J. E. 1974. Detection and analysis of
epoxides with 4-(p-nitrobenzyl)-pyridine. Bull. Environ. Contarnin. Toxicol.
12:759-764.

ten Have, A., Mulder, W., Visser, J., and van Kan, J. A. 1998. The endopolygalacturonase

gene Bcpg] is required for full virulence of Botrytis cinerea. Mol Plant Microbe
Interact. 11:1009—1016.

He, S. Y., and Collmer, A. 1990. Molecular cloning, nucleotide sequence, and marker
exchange mutagenesis of the exo-poly-alpha-D-galacturonosidase-encoding pth
gene of Erwinia chrysanthemi EC16. J. Bacteriol. 172:4988-4995.

He, C., Nourse, J. P., Kelemu, S., Irwin, J. A., and Manners, J. M. 1996. Cng: a non-
LTR retrotransposon with restricted distribution in the fungal phytopathogen
CoIletotrichum gloeosporioides. Mol. Gen. Genet. 252:320—331.

Hirota, A., Suzuki, A. and Tamura, S. 1973. Characterization of four amino acids

constituting Cyl-2, a metabolite from Cylindrocladium scoparium. Agr. Biol.
Chem. 37:1185-1189.

Itazaki, H., Nagashima, K., Sugita, K., Yoshida, H., Kawamura, Y., Yasuda, Y.,
Matsumoto, K., Ishii, K., Uotani, N., Nakai, H., Terui, A., Yoshimatsu, S.,
Ikenishi, Y. and Nakagawa, Y. 1990. Isolation and structural elucidation of new
cyclotetrapeptides, trapoxins A and B, having detransformation activities as
antitumor agents. J. Antibiot. 43: 1524-1 532.

Jennings, P. R., and Ullstrup, A. J. 1957. A histological study of three
Helminthosporioum leaf blights of corn. Phytopathology 47:707-714.

Jinks-Robertson, S., and Petes, T. D. 1986. Chromosomal translocations generated by
high-frequency meiotic recombination between repeated yeast genes. Genetics
114:731-752.

Johal, G. S., and Briggs, S. P. 1992. Reductase activity encoded by the HM] disease
resistance gene in maize. Science 258:985-987.

John, M. E., and Petersen, M. W. 1994. Cotton (Gossypium hirsutum L.) pollen-specific
polygalacturonase mRN A: tissue and temporal specificity of its promoter in
transgenic tobacco. Plant Mol. Biol. 26: 1989-1993.

Kachroo, P., Leong, S. A., and Chattoo, B. B. 1994. Pot2, an inverted repeat transposon
from the rice blast fungus Magnaporthe grisea. Mol. Gen. Genet. 245:339-348.

Kalaitzis, P., Koehler, S. M., and Tucker, M. L. 1995. Cloning of a tomato
polygalacturonase expressed in abscission. Plant Mol. Biol. 28:647-656.

125

Keitel, T., Simon, 0., Borris, R. and Heinemann, U. 1993. Molecular and active-site
structure of a Bacillus 1,3-1,4-D-B-glucanase. Proc. Natl. Acad. Sci. USA
90:5287-5291.

Keller, N. P., and Hohn, T. M. 1997. Metabolic pathway gene clusters in filamentous
fungi. Fungal Genet. Biol. 21:17-29.

Kim, S.-D., Knoche, H. W., and Dunkle, L. D. 1987. Essentiality of the ketone fimction
for toxicity of the host-selective toxin produced by Helminthosporium carbonum.
Physiol. Mol. Plant Pathol. 30:433-440.

Kistler, H. C., and Van Etten, H. D. 1984a. Regulation of pisatin demethylation in

Nectria haematococca and its influence on pisatin tolerance and virulence. J. Gen.
Microbiol. 130:2605-2613.

Kistler, H. C., and Van Etten, H. D. 1984b. Three non-allelic genes for pisatin
demethylation in the fungus Nectria haematococca. J. Gen. Microbiol. 130:2595-
2603.

Kleinkauf, H., and von Ddhren, H. 1990. Bioactive peptides - recent advances and trends.
In: Kleinkauf, H. and von Dohren, H. (eds.). Biochemistry of Peptide Antibiotics.
Walter de Gruyter, Berlin-NY, pp. 1-31.

Kleinkauf, H., and von Dohren, H. 1996. A nonribosomal system of peptide biosynthesis.
Eur. J. Biochem. 236:335-351.

Kofod, L.V., Kauppinen, S., Christgau, S., Andersen, L. N., Heldt-Hansen, H. P.,
Dorreich, K. and Dalboge, H. 1994. Cloning and characterization of two
structurally and functionally divergent rhamnogalacturonases from Aspergillus
aculeatus. J. Biol. Chem. 269:29182-29189.

Konz, D., Klens, A., Schorgendorfer, K., and Marahiel, M. A. 1997. The bacitracin
biosynthesis operon of Bacillus licheniformis ATCC 10716: molecular
characterization of three multi-modular peptide synthetases. Chem. Biol. 4:927-
937.

Kubo, Y., Tsuda, M., Furusawa, I., and Shishiyama, J. 1989. Genetic analysis of genes
involved in melanin biosynthesis of Cochliobolus miyabeanus. Exp. Mycol.
13:77-84.

Kubo, Y., Nakamura, H., Kobayashi, T., Okuno, T., and Furusawa, I. 1991. Cloning of a

melanin biosynthetic gene essential for appressoria] penetration of Colletotrichum
lagenarium. Mol. Plant-Microbe Interact. 4:440-445.

126

Kupfer, D. M., Reece, C. A., Clifton, S. W., Roe, B. A., and Prade, R. A. 1997. Direct
sequence submission to the EMBL/GenBank data banks, accession # AC000133.

Lee, S. G., and Lipmann, F. 1975. Tyrocidine synthetase system. Meth. Enzymol. 43:585-
602.

Leonard, K. J. 1978. Polymorphisms for lesion type, fiingicide tolerance, and mating
capacity in Cochliobolus carbonum isolates pathogenic to corn. Can. J. Bot.
56:1809-1815.

Lester, D. R., Speirs, J., Orr, G., and Brady, C. J. 1994. Peach (Prunus persica)
endopolygalacturonase cDNA isolation and mRNA analysis in melting and
nonmelting peach cultivars. Plant Physiol. 105:225-231.

Levis, C., Fortini, D., and Brygoo, Y. 1997. Flipper, a mobile Fotl-like transposable
element in Botrytis cinerea. Mol. Gen. Genet. 254:674-680.

Loidl, J ., and Nairz, K. 1997. Karyotype variability in yeast caused by nonallelic
recombination in haploid meiosis. Genetics 146:79—88.

Long, G. S., Bryant, J. M., Taylor, P. W., and Luzio, J. P. 1995. Complete nucleotide
sequence of the gene encoding bacteriophage E endosialidase: implications for
KlE endosialidase structure and function. Biochem. J. 309:543-550.

Lorito, M., Hayes, C.K., di Pietro, A., Woo, S. L., and Harman, G. E. 1994. Purification,
characterization and synergistic activity of a glucan 1,3-B-glucosidase and an N-
acetyl-B-glucosaminidase from T richoderma charzianum. Phytopathology,
84:398-405.

Mackenzie, L. F., Brooke, G. S., Cutfield, J. F., Sullivan, P. A., and Withers, S. G. 1997.
Identification of Glu-330 as catalytic nucleophile of Candida albicans exo-beta-
(1,3)-glucanase. J. Biol. Chem. 272:3161-3167.

Masel, A. M., Irwin, J. A., and Manners, J. M. 1993. DNA addition or deletion is
associated with a major karyotype polymorphism in the fungal phytopathogen
Colletotrichum gloeosporioides. Mol. Gen. Genet. 237:73-80.

Masel, A. M., He, C., Poplawski, A. M., Irwin, J. A. G., and Manners, J. M. 1996.
Molecular evidence for chromosome transfer between biotypes of Colletotrichum
gloeosporioides. Mol. Plant-Microbe Interact. 9:339-348.

Meeley, R. B., and Walton, J. D. 1991. Enzymatic detoxification of HC-toxin, the host-

selective cyclic peptide from Cochliobolus carbonum. Plant Physiol. 9721080-
1086.

127

Meeley, R. B., Johal, G. S., Briggs, S. P., and Walton, J. D. 1992. A biochemical
phenotype for a disease resistance gene of maize. Plant Cell 4:71-77.

Miao, V. P., Covert, S. F., and Van Etten, H. D. 1991a. A fungal gene for antibiotic
resistance on a dispensable (“B”) chromosome. Science 254: 1773-1776.

Miao, V. P., Matthews, D. E., and Van Etten, H. D. 1991b. Identification and
chromosomal locations of a family of cytochrome P-450 genes for pisatin

detoxification in the fungus Nectria haematococca. Mol. Gen. Genet. 226:214-
223.

Miao, V. P. W., and Van Etten, H. D. 1992. Genetic analysis of the role of phytoalexin
detoxification in virulence of the fungus Nectria haematococca on chickpea
(Cicer arietinum). Appl. Environ. Microbiol. 58:801-808.

Multani, D. S., Meeley, R. B., Paterson, A. H., Gray, J., Briggs, S. P., and Johal, G. S.
1998. Plant-pathogen microevolution: molecular basis for the origin of a fungal
disease in maize. Proc. Natl. Acad. Sci. U S A. 95:1686-1691.

Murray, G. M., and Maxwell, D. P. 1975. Penetration of Zea mays by Helminthosporium
carbonum. Can. J. Bot. 53:2872-2883.

Nelson, 0. E., and Ullstrup, A. J. 1961. The inheritance of pathogenicity in Cochliobolus
carbonum. Phytopathology 5 l : 1288-1291 .

Nelson, 0. E., and Ullstrup, A. J. 1964. Resistance to leafspot in maize. J. Hered. 55:195-
199.

Nielsen, H., Engelbrecht, J ., Brunak, S. and von Heijne, G. 1997. Identification of
prokaryotic and eukaryotic signal peptides and prediction of their cleavage sites.
Protein Engineering 10: 1-6.

Nikolskaya, A. N., Panaccione, D. G., and Walton, J. D. 1995. Identification of peptide
synthetase-encoding genes from filamentous fungi producing host-selective
phytotoxins or analogs. Gene 165:207-211.

Nikolskaya, A. N., Pitkin, J. W., Schaeffer, H. J., Ahn, J.-H., and Walton, J. D. 1998.
EXGlp, a novel exo-B-1,3-glucanase from the fungus Cochliobolus carbonum,
contains a repeated motif present in other proteins that interact with
polysaccharides. Biochimica et Biophysica Acta 1425:632-636.

Nobela, C., Molina, M., Cenamor, R. and Sanchez, M. 1988. Yeast B-glucanases: a
complex system of secreted enzymes. Microbiol. Sci. 5:328-332.

128

Nyyssonen, E., Amutan, M., Enfield, L., Stubbs, J ., and Dunn-Coleman, N. S. 1996. The
transposable element Tanl of Aspergillus niger var. awamori, a new member of
the Pot] family. Mol. Gen. Genet. 253:50—56.

Orbach, M. J ., Porro, E. B., and Yanofsky, C. 1986. Cloning and characterization of the
gene for beta-tubulin from a benomyl-resistant mutant of Neurospora crassa and
its use as a dominant selectable marker. Mol. Cell. Biol. 6:2452-2461.

Orbach, M. J., Chumley, F. G., and Valent, B. 1996. Electrophoretic karyotypes of

Magnaporthe grisea pathogens of diverse grasses. Mol. Plant—microbe Interact.
9:261-271.

Paces, V., Vlcek, C., Urbanek, P., and Hostomsky, Z. 1985. Nucleotide sequence of the
major early region of Bacillus subtilis phage PZA, a close relative of phi-29. Gene
38:45-56.

Panaccione, D.G., Scott-Craig, J .S., Pocard, J .A. and Walton, J .D. 1992. A cyclic peptide
synthetase gene required for pathogenicity of the fungus Cochliobolus carbonum
on maize. Proc. Natl. Acad. Sci. USA 89:6590-6594.

Panaccione, D. G., Pitkin, J. W., Walton, J. D., and Annis, S. L. 1996. Transposon-like
sequences at the T 0X2 locus of the plant-pathogenic fungus Cochliobolus
carbonum. Gene 176: 103-109.

Peberdy, J .F. 1990. Fungal cell wall - a review. In: Kuhn, P. J., Trinci, A. P. J., Jung, M.
J., Goosey, M. W., and Copping, L. G. (eds.). Biochemistry of Cell Walls and
Membranes in Fungi. Springer-Verlag, Heidelberg, Germany.

Perkins, D. D. 1974. The manifestation of chromosome rearrangements in unordered asci
of Neurospora. Adv. Genet. 19:133-285.

Petersen, M., Sander, L., Child, R., van Onchelen, H., Ulvskov, P. and Borkhardt, B.
1996. Isolation and characterization of a pod dehiscence zone-specific
polygalacturonase from Brassica napus. Plant Mol. Biol. 31 :517-527.

Petter, J. G. and Vimr, E. R. 1993. Complete nucleotide sequence of the bacteriophage
KlF tail gene encoding endo-N-acylneurarninidase (endo-N) and comparison to
an endo-N homolog in bacteriophage PKlE. J. Bacteriol. 175:43 54-4363.

Pirone, P. P. 1970. Diseases and pests of ornamental plants. Ronald Press, NY, p. 376.

Pitkin, J. W., Panacione, D. G., and Walton, J. D. 1996. A putative cyclic peptide efilux

pump encoded by the TOXA gene of the plant-pathogenic fimgus Cochliobolus
carbonum. Microbiology 142:557-1565.

129

Program manual for the Wisconsin Package, version 8. Genetics Computer Group,
Madison, WI 53711, 1994.

Quigley, F. R. 1993. Direct sequence submission to the EMBL/GenBank data banks,
accession # 1346702.

Rotem, Y., Yarden, O., and Sztejnberg, A. 1998. Direct sequence submission to the
GenBank data bank, accession # AF 029354.

Sambrook, J ., Fritsch, E. F., and Maniatis, T. A. 1989. Molecular Cloning: A Laboratory
Manual. 2nd ed. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY.

Schaeffer, H. J ., Leykam, J ., and Walton, J. D. 1994. Cloning and targeted gene
disruption of EXG] , encoding exo-B-1,3-glucanase, in the phytopathogenic fungus
Cochliobolus carbonum. Appl. Environ. Microbiol. 60:594-598.

Schafer, W., Straney, D., Cuiffetti, L., Van Etten, H. D., and Yoder, O. C. 1989. One
enzyme makes a fungal pathogen, but not a saprophyte, virulent on a new host
plant. Science 246:247-249.

Schauwecker, F., Pfennig, F, Schroder, W., and Keller, U. 1998. Molecular cloning of
the actinomycin synthetase gene cluster from Streptomyces chrysomallus and
functional heterologous expression of the gene encoding actinomycin synthetase
II. J. Bacteriol. 180:2468-2474.

Scheffer, R. P., and Ullstrup, A. J. 1965. A host-specific toxic metabolite from
Helminthosporium carbonum. Phytopathology 55: 1037-1038.

Scheffer, R. R, Nelson, R. R., and Ullstrup, A. J. 1967. Inheritance of toxin production
and pathogenicity in Cochliobolus carbonum and Cochliobolus victoriae.
Phytopathology 57:1288-1291.

Scheffer, R. P., and Livingston, R. S. 1984. Host selective toxins and their role in plant
diseases. Science 223 : 1 7-21 .

Scott-Craig, J. S., Panaccione, D. G., Cervone, F., and Walton, J. D. 1990.
Endopolygalacturonase is not required for pathogenicity of Cochliobolus
carbonum on maize. Plant Cell 2:1191-1200.

Scott-Craig, J. S., Panaccione, D. G., Pocard, J. A., and Walton, J. D. 1992. The cyclic
peptide synthetase catalyzing HC-toxin production in the filamentous fungus
Cochliobolus carbonum is encoded by a 15.7-kilobase open reading frame. J.
Biol. Chem. 267:26044-26049.

130

Scott-Craig, J. S., Cheng, Y.-Q., Cervone, F., De Lorenzo, G., Pitkin, J. W., and Walton,
J. D. 1998a. Targeted mutants of Cochliobolus carbonum lacking the two major
extracellular polygalacturonases. Appl. Environ. Microbiol. 64: 1497-1503.

Scott-Craig, J. S., Apel-Birkhold, P. C., Géirlach, J. M., Nikolskaya, A., Pitkin, J. W.,
Ransom, R. F., Sposato, P., Ahn, J.-H., Tonukari, N. J ., Wegener, S., and Walton,
J. D. 1998b. Cell wall degrading enzymes in HST-producing fungal pathogens. In:
Kohmoto, K., and Yoder, O. C. (eds.). Molecular Genetics of Host-Specific
Toxins in Plant Disease. Kluwer Academic Publishers, The Netherlands, pp. 245-
252.

Shen, S., Chretien, P., Bastien, L., and Slilaty, S. N. 1991. Primary structure of the
glucanase gene from Oerskovia xanthineolytica. Expression and purification of
the enzyme from Escherichia coli. J. Biol. Chem. 266:675-682.

Shieh, M. T., Brown, R. L., Whitehead, M. P., Cary, J. W., Cotty, P. J., Cleveland, T. E.,
and Dean, R. A. 1997. Molecular genetic evidence for the involvement of a
specific polygalacturonase, P2c, in the invasion and spread of Aspergillus flavus
in cotton bolls. Appl. Environ. Microbiol. 63:3548-3552.

Smith, D. J ., Earl, A. J. and Turner, G. 1990. The multifunctional peptide synthetase
performing the first step of penicillin biosynthesis in Penicillum chrysogenum is a
421,073 dalton protein similar to Bacillus brevis peptide antibiotic synthetases.
EMBO J. 9:2743-2750.

Sposato, P., Ahn, J .-H., and Walton, J. D. 1995. Characterization and disruption of a gene
in the maize pathogen Cochliobolus carbonum encoding a cellulase lacking a
cellulose binding domain and hinge region. M01. Plant-Microbe Interact. 8:602-
609.

Stachelhaus, T., and Marahiel, A. 1995. Modular structure of peptide synthetases revealed
by dissection of the multifunctional enzyme GrsA. J. Biol. Chem. 270:6163-6169.

Stahmann, K. P., Pielken, P., Schimz, K. L. and Sahm, H. 1992. Degradation of
extracellular B-(l,3)—(l,6)—D-glucan by Bohytis cinerea. Appl. Environ.
Microbiol. 58:3347—3354.

Stahmann, K. P., Schimz, K. L. and Sahm, H. 1993. purification and characterization of
four extracellular 1,3-B-glucanases from Bonytis cinerea. J. Gen. Microbiol.
139:2833-2840.

Turgay, K., Krause, M., and Marahiel, M. A. 1992. Four homologous domains in the

primary structure of GrsB are related to domains in a superfamily of adenylate-
forming enzymes. Mol. Microbiol. 62529-546.

131

Turgay, K., and Marahiel, M. A. 1994. A general approach for identifying and cloning
peptide synthetase genes. Pept. Res. 7:23 8-241.

Tzeng, T.-H., Lyngholm, L. K., Ford, C. F., and Bronson, C. R.. 1992. A restriction
fragment length polymorphism map and electrophoretic karyotype of the fungal
maize pathogen Cochliobolus heterostrophus. Genetics 130:81-96.

Ullstrup, A. J. 1941. Two physiological races of Helminthosporium maydis in the corn
belt. Phytopathology 57:1288-1291.

Umehara, K., Nakahara, K., Kiyoto, S., Iwami, M., Okamoto, M., Tanaka, H., Kohsaka,
M., Aoki, H., and Imanaka, H. 1983. Studies on WF-3161, a new antitumor
antibiotic. J. Antibiot. 36:478-483.

Van Etten, H. D., Han, Y., Liu, X., and Kistler, C. 1997. Identification of pea
pathogenicity genes on a dispensable chromosome in Nectria haematococca.
American Phytopathological Society Annual Meeting Program Book, Rochester,
NY, p. 80.

Van Hoof, A., Leykam, J., Schaeffer, H. J. and Walton, J. D. 1991. A single B-1,3
glucanase secreted by the maize pathogen Cochliobolus carbonum acts by an
exolytic mechanism. Physiol. Mol. Plant Pathol. 39:259-267.

Van Wageningen, A. M., Kirkpatrick, P. N., Williams, D. H., Harris, B. R., Kershaw, J.
K., Lennard, N. J., Jones, M., Jones, S. J., and Solenberg, P. J. 1998. Sequencing
and analysis of genes involved in the biosynthesis of a vancomycin group
antibiotic. Chem. Biol. 5:155-162.

Van Wert, S. L., and Yoder, O. C. 1992. Structure of the Cochliobolus heterostrophus
glyceraldehyde-3-phosphate dehydrogenase gene. Curr. Genet. 22:29-35.

Villanueva, N., and Salas, M. 1981. Adsorption of bacteriophage phi 29 to Bacillus
subtilis through the neck appendages of the viral particle. J. Virol. 38:15-19.

Walton, J. D. 1987. Two enzymes involved in biosynthesis of the host-selective
phytotoxin HC-toxin. Proc. Natl. Acad. Sci. USA 84:8444-8447.

Walton, J. D. 1990. Peptide phytotoxins from plant pathogenic fungi. In: Kleinkauf, H.
and von Dohren, H. (eds.). Biochemistry of Peptide Antibiotics. Walter de
Gruyter, Berlin - NY, pp. 179-203.

Walton, J. D. 1994. Deconstructing the cell wall. Plant Physiol. 104:1113-1118.

132

Walton, J. D. 1996. Host-selective toxins: agents of compatibility. Plant Cell. 8:1723-
1733.

Walton, J. D., Earle, E. D., and Gibson, B. W. 1982. Purification and structure of the
host-specific toxin from Helminthosporium carbonum Race 1. Biochem. Biophys.
Res. Commun. 107:785-794.

Walton, J. D., Earle, E. D., Stahelin, H., Grieder, A., Hirota, A., and Suzuki, A. 1985.
Reciprocal biological activities of the cyclic tetrapeptides Chlamydocin and HC-
toxin. Experientia 41:348-350.

Walton, J. D., and Holden, F. R. 1988. Properties of two enzymes involved in the
biosynthesis of the fungal pathogenicity factor HC-toxin. Mol. Plant-Microbe
Interact. 1:128-134.

Walton, J. D, Bronson, C. R., Panaccione, D. G., Braun, E. J., and Akimitsu, K. 1995.
Cochliobolus. In: Kohmoto, K., Singh, U. S., and Singh, R. P. (eds.). Pathogenesis
and Host Specificity in Plant Diseases 2. Pergamon/Elsevier, NY, pp. 65-81.

Walton, J. D., Ahn, J.-H., Pitkin, J. W., Cheng, Y., Nikolskaya, A. N., Ransom, R., and
Wegener, S. 1998. Enzymology, molecular genetics, and regulation of
biosynthesis of the host-selective toxin HC-toxin. In: Kohmoto, K., and Yoder, O.
C. (eds.). Molecular Genetics of Host-Specific Toxins in Plant Disease. Kluwer
Academic Publishers, The Netherlands, pp. 25-34.

Wasmann, C. C., and Van Etten, H. D. 1996. Transformation-mediated chromosome loss
and disruption of a gene for pisatin demethylase decrease the virulence of Nectria
haematococca on pea. Mol. Plant-Microbe Interact. 9:793-803.

Watanabe, T., Kasahara, N., Aida, K., and Tanaka, H. 1992. Three N-terrninal domains
of beta-1,3-glucanase A1 are involved in binding to insoluble beta-1,3-glucan. J.
Bacteriol. 174:186-190.

Xu, J. R., and Leslie, J. F. 1996. A genetic map of Gibberellafiijikuroi mating population
A (F usarium moniliforme). Genetics 143:175-189.

Yoder, DC. 1988. Cochliobolus heterostrophus, cause of southern corn leaf blight. Adv.
Plant Pathol. 6:93-112.

Yoder, O. C., and Scheffer, R. P. 1973a. Effects of Helminthosporium carbonum toxin on
absorption of solutes by com roots. Plant Physiol. 52:518-523.

Yoder, O. C., and Scheffer, R. P. 1973b. Effects of Helminthosporium carbonum toxin on
nitrate uptake and reduction by com tissues. Plant Physiol. 52:513-517.

133

HIGQN STRTE UNIV. LI

llllllllllllzllllllllllllllllll 7llll7| llll alllllLlllLllll