1.. f. . :1
2.2 ;. . .
15.1 c.-

vvaz It:
v.9..pv.l
uni-.5“.
r.l.'vA..u .
;:.'IE‘ 1
{ell v."

.3 n.)
. I. I'll.

.' r :
bindinhahw

III":

I. ..
4!:t!
.1 1 ‘

vh.

uwuflng

. .l: . 7 2r

0 :
aim?

. _..:.;.r:. 1‘ .i:

 

I—
tES

‘ llllimlmix

This is to certify that the
dissertation entitled
A BIOCHEMICAL BASIS FOR HERITABLE
RESISTANCE OF MAIZE TO THE
FUNGAL PATHOGEN COCHLIOBOLUS CARBONUM

presented by

Robert Brendan Meeley

has been accepted towards fulfillment
of the requirements for

__EhD_ degree in _Biochemistry

 

Date 3 September, 1992

 

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

 

 

{V

 

k

Michigan State

‘
LEEM‘RY

University

 

_,l

_____7

PLACE IN RETURN BOX to remove this checkout from your record.
TO AVOID FINES return on or before date due.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

fl

 

 

 

ll

 

MSU Is An Affirmative Action/Equal Opportunity Institution

ammo-9.1

A BIOCHEMICAL BASIS FOR HERITABLE RESISTANCE OF MAIZE
TO THE FUNGAL PATHOGEN COCHLIOBOLUS CARBONUM

By

Robert Brendan Meeley

A DISSERTATION

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

DOCTOR OF PHILOSOPHY

Department of Biochemistry

1992

ABSTRACT

A BIOCHEMICAL BASIS FOR HERITABLE RESISTANCE OF MAIZE
TO THE FUNGAL PATHOGEN, COCHLIOBOLUS CARBONUM

By

Robert Brendan Meeley

The fungus Cochliobolus carbonum Nelson race 1 is a foliar pathogen of maize
hosts that are homozygous recessive for the resistance gene locus hm. C. carbonum race
1 produces HC-toxin, a cyclic tetrapeptide required for successful colonization of hm/hm
maize. This dissertation addresses questions about the function of the Hm allele in
providing resistance to C. carbonum race 1. Previous workers have determined that
near-isogenic maize lines, differing only at the hm locus, are differentially sensitive to
the effects of HC-toxin. The working hypothesis of this dissertation proposes that the
Hm allele confers decreased sensitivity to HC-toxin by metabolic inactivation of this
compound. An inactive derivative of a radiolabeled form of HC-toxin was recovered
from resistant (Hm/-) maize and characterized by thin-layer chromatography, NMR, and
mass spectrometry. An enzymatic basis for metabolite formation was detected, partially
purified, and characterized. The enzyme HC-toxin reductase utilizes NAD(P)H for
stereoselective reduction of the 8-keto group of the amino acid 2-amino-9,10—epoxy-8-
oxodecanoic acid. A variety of genetic materials were examined for HC-toxin reductase

activity. The results establish HC-toxin reductase activity as the biochemical phenotype

for the Hm resistance allele. Best evidence suggests that the Hm allele encodes the gene
for HC-toxin reductase. Purification procedures enrich for three polypeptides of 49,
43.5, and 40 kD; an estimate for the HC-toxin reductase molecular weight is 42 kD.
HC-toxin reductase is induced in resistant maize by inoculation of etiolated mdlings with
a spore suspension of C. carbonum race 1, indicative of a defense-related role for this
enzyme. In addition, enzyme activities analogous to HC-toxin reductase were detected
in extracts from several other monocot species, but not in dicot extracts. Evidence
suggests that the alternate C. carbonum race 1 resistance gene of maize, Hm2, may
provide resistance to C. carbonum race 1 by a similar mechanism of HC-toxin
inactivation. This dissertation presents a description of the biochemical function of a
disease resistance gene from a major crop, and raises difficult but approachable questions

about host-pathogen coevolution.

To Ken and Holly

iv

 

ACKNOWLEDGEMENTS

Many, many thanks to Dr. Jon Walton for his patience and guidance. The past
few years have really been an exciting time to be in the Walton Lab, and I wish everyone
continued success. I am grateful to Drs. Kindel, Hausinger, McIntosh, Deits, and
Lamport (while free) for their encouragement and input as my guidance committee. Dr.
Steve Briggs and Dr. Guri J ohal from Pioneer Hi-Bred receive my enduring gratitude and
respect for their collaborative spirit. And to my uniquely insane family for all the love,
fun, and plane tickets.

Special bullshit is reserved here for my friends Jim and Todd; words are not
required because they always get the joke. Mickstah, Jack the Bear, and Mabel, the
Baby Kitty.

A ritual ”flip of the bird" goes to Mr. Robert Blanchard of Okemos, Bank-One
of E. Lansing (eighteen thousand people poised and ready to give you the shaft), and the
Meridian Township Police Department. Thanks to J SC, Dan P. , John P. , John Shanklin,
Doug Gage, Scott Peck, Vincent Arondel, Kermit Johnson, and David Wagner for
tremendous help along the way. And here’s to Woody’s Oasis (Woody, you are so

cool), The Peanut Barrel, Luigi’s Pizza, and Shaw’s Triangle Truck-Stop.

LIST OF TABLES
LIST OF FIGURES

TABLE OF CONTENTS

........................ viii
........................ ix

LIST OF ABBREVIATIONS ........................ xii
CHAPTER 1: INTRODUCTION ......................... 1
The Gene-for-Gene Hypothesis .......... 3

The Genetic System ................. 7

The Host ................... 8

The Pathogen ................ 9

HC-toxin ...................... 16

Structure .................. 16

Biosynthesis ................ 19

Biological Activity ............ 21

Structural Requirements ......... 28

The Working Hypothesis ............. 29

References ..................... 34

CHAPTER 2: ENZYMATIC DETOXIFICATION OF HC-TOXIN . . . . 39

Abstract ....................... 40
Introduction ..................... 41
Materials and Methods .............. 42
Results ........................ 48
Discussion ...................... 67
References ..................... 70

vi

CHAPTER 3: A BIOCHEMICAL PHENOTYPE FOR A

DISEASE RESISTANCE GENE OF MAIZE ........ 73
Abstract ....................... 74
Introduction ..................... 75
Materials and Methods .............. 77
Results ........................ 79
Discussion ...................... 94
Acknowledgements ................ 95
References ..................... 96

CHAPTER 4: STUDIES ON THE PURIFICATION AND
INDUCTION OF HC-TOXIN REDUCTASE,
AND ITS DISTRIBUTION AMONG
PLANT SPECIES ................

Introduction .............
Materials and Methods ......
Results ................
Discussion ..............

References .............

vii

........ 98

....... 100

....... 104

....... 119

 

 

LIST OF TABLES

CHAPTER 2:

Table 2.1. Incorporation of radiolabeled amino acid
precursors into HC-toxin in vivo. ............ 49

Table 2.2. Characterization of HCTR activity ............. 64

viii

CHAPTER 1:
Figure 1.1.

Figure 1.2.

Figure 1.3.
Figure 1.4.

Figure 1.5.

Figure 1.6.

CHAPTER 2:

Figure 2.1.

Figure 2.2.

Figure 2.3.

Figure 2.4.

Figure 2.5.

LIST OF FIGURES

A hypothetical complementary genic system ........ 6
A complementary genic interaction between

C. carbonum race 1 and maize .............. 15
The structure of HC-toxin .................. 18
Root growth bioassay of HC-toxin ............. 23

Other Aeo-containing
cyclic tetrapeptides from fungi .............. 27

Resistance of maize to C. carbonum defined
in context with the end-product of the
C. carbonum pathogenicity gene ............. 31

Formation of a metabolite of HC-toxin by
green maize leaf segments ................ 53

TLC analysis of HC-toxin metabolites

produced by green leaf segments and by

cell-free extracts of resistant maize ........... 55
Fast-atom bombardment mass spectra of

(A) the metabolite of HC-toxin produced

by maize cell-free extracts and

(B) HC-toxin reduced with NaBIL ........... 57
NMR analysis of the HC-toxin metabolite ........ 59

Anion exchange fractionation of HCTR activit ..... 62

ix

Figure 2.6.

CHAPTER 3:

Figure 3.1.

Figure 3.2.

Figure 3.3.

Figure 3.4.

Figure 3.5.

Figure 3.6.

CHAPTER 4:

Figure 4.1.

Figure 4.2.

Figure 4.3.

Figure 4.4.

Figure 4.5.

Time course of [3H]HC-toxin metabolism
by maize leaf segments during

uptake by transpiration ...........

HC-toxin reductase (HCTR) activity
in extracts of etiolated shoots of
maize lines susceptible or resistant

to C. carbonum race 1 ...........

Cosegregation of HCTR activity and
disease reaction to C. carbonum race 1

HCTR activity and disease phenotype
in extracts of transposon—induced

mutants of Hm ................

HCTR activity and disease phenotype
of revertants due to germinal excision

of Spm (or dSpm) from hut-1040 . . . .

Host-selective metabolism of HC-toxin

by green leaf discs .............

Host-selective metabolism of HC-toxin

by intact plumules .............

Fractionation of HCTR activity ......

SDS-PAGE analysis

of HCTR-containing fractions ......

Comparative SDS-PAGE analysis of

HCTR-containing fractions ........

Induction of HCTR activity

in response to fungal inoculation .....

Distribution of HCTR-like activity

among several different plant species . .

........ 66

........ 81

........ 83

........ 86

........ 88

........ 91

........ 93

....... 108

....... 111

....... 113

....... 116

Figure 4.6. HC-toxin metabolism and Hm2 .............. 118

CHAPTER 5:

Figure 5.1. The C. carbonum race 1/maize interaction
in biochemical terms .................. 130

Figure 5.2. Comparison of the deduced amino acid sequence
of the Hm gene with the A1 gene of maize ..... 133

Figure 5.3. Methods for intron removal and sequence

adaptation for expression of the Hm cDNA
in heterologous systems ................. 138

xi

LIST OF ABBREVIATIONS

Aeo .......... 2-amino—9,10-epoxy-8-oxodecanoic acid
bis-Tris . bis[2-hydroxyethyl]imino-tris[hydroxymethyl]methane
6m .............. chemical shift (parts per million)
DTI‘ .......................... dithiothreitol
HCTR ...................... HC-toxin reductase
p—HMB .................. p-hydroxymercuribenzoate
Mops ........... 3-[N-morpholino]propanesulfonic acid
p—NBP ................... 4-(p-nitrobenzyl)-pyridine
pers. commun. .................. personal communication
PMSF ................ phenylmethylsulfonyl fluoride
PVP ...................... polyvinylpyrrolidone

xii

CHAPTER 1

INTRODUCTION

2

Plant pathogens have evolved presumably because living plants offer a practical
niche for the growth, development, and reproduction of certain organisms. As an
evolutionary response, the development of specific defenses by plants might be considered
an essential step for their own productivity and survival. In the interest of understanding
the process of co—evolution between a host and parasite (reviewed by Frank, 1992), and
to eventually manipulate genetic traits for disease resistance in higher plants (reviewed by
Gasser and Fraley, 1989), scientists have studied the genetic principles that govern plant-
microbe interactions in a variety of model systems. These models typically involve major
crop species and their perennial microbial pests.

Despite great strides in our knowledge of host-pathogen genetics, very little is
known about the gene products that mediate plant-microbe interactions. Current molecular
and/or biochemical models largely represent extrapolations of genetic data (Gabriel and
Rolfe, 1990), particularly with regard to the biochemical nature of specific genes for
disease resistance in plants. For this reason a protracted review of specific models in
modern plant pathology will be avoided. A variety of detail and Opinion is expressed
regularly in review articles (Gabriel and Rolfe 1990; Keen, 1990). This introduction will
focus on the fundamentals of host-pathogen genetics, for virtually all of the models
expressed in plant pathology today have their roots in a single tenet known as the gene-for-

gene hypothesis.

The Gene-for-Gene Hypothesis

H. H. Flor (1900—1987), a plant pathologist from the USDA in North Dakota, is
widely acknowledged for his critical contribution to the knowledge of plant-pathogen
interactions. Flor examined the genetic principles that govern the interaction between flax
(Linum usitatissimum) and flax rust (Melampsora lint), developing a seminal hypothesis

which states:

”For each gene conditioning resistance in the host there is a specific gene

conditioning avirulence in the parasite.” (Flor, 1942)

and alternatively:

”For each gene conditioning resistance in the host there is a specific gene

conditioning pathogenicity in the parasite." (Flor, 1956)

As complementary genes, a gene in one organism is defined in context with a gene
from the other. The difference between the above hypotheses concerns the type of
pathogen being studied. In many interactions, pathogen resistance by the host occurs via
genes that complement, or recognize, characteristics for avirulence in the pathogen. The
product(s) of dominant avirulence genes are thought to interact with the product(s) of
dominant plant resistance genes at some molecular level. A physical interaction between
gene products is implied, that is somehow transduced into a resistance response in the

infected plant tissue (Gabriel and Rolfe, 1990; Keen, 1990). Hypersensitivity is the term

4

used to describe this response because it involves local necrosis of the plant tissue
surrounding the site of infection. This premature death of host cells restricts the growth
of pathogens to a few infected cells. Further infection is thus avoided (Gabriel and Rolfe,
1990). Diseases involving this type of complementarity are quite common, and this was

the type studied by Flor himself.

Avirulence and hypersensitivity are not central to the disease being studied in this
dissertation. Our host-pathogen system involves gene—for—gene complementarity with
respect to pathogenicity, so Flor’s 1956 wording of the gene-for-gene hypothesis will
serve as our starting point. Examples of this type involve the interaction of pathogenicity
gene product(s) with the product(s) of a complementary host resistance gene at some
molecular level.

Figure 1.1 illustrates a quadratic check for a hypothetical complementary genie
system involving host resistance in the context of traits for pathogenicity. In systems
where sexual manipulation of a plant pathogen can be accomplished in the laboratory, a
single gene for pathogenicity (denoted P) can be identified in context with its complement;
a gene that conditions resistance (R) or susceptibility (r) in the host. Host resistance is
based on specific "recognition” of the pathogenicity gene product(s). We have little or no
evidence to propose a general biochemical definition for "recognition". This remains an
ultimate objective in the field. In the pages that follow, this quadratic check will be
elaborated to communicate the historical and experimental details of our system, such that

a hypothesis about pathogen recognition can be developed for our specific case.

Figure 1.1. A hypothetical complementary genie system. This quadratic
check shows a plant gene for resistance (R) or susceptibility (r) defined in
context with a complementary pathogen gene pair for pathogenicity (P) or
non-pathogenicity (p). (+) indicates a compatible interaction (disease
favorable). (-) indicates an incompatible interaction (no disease).

 

 

Hypothetical Pathogen] Plant Interaction
Host Genotype

R/- r/ r

, — +

 

 

Pathogen
Genotype

 

 

 

 

Figure 1.1.

7

Another distinction to be made is that our model system involves an interaction at
the level of race-cultivar specificity. Plant pathogens are capable of distinguishing
potential hosts at a variety of levels, and vice versa. For example, a certain set of traits
common to a fungal genus may condition pathogenicity to monocotyledons in general,
while another set may be required for pathogenicity of a particular species on maize. The
same might be said of traits for avirulence, or even plant characteristics that limit the host
range of potential pathogens. These kinds of traits are important because they increase the
web of genetic complexity in host-pathogen interactions. They are regarded as polygenic
traits that comprise the elements of basic compatibility between two organisms (Heath,
1991). Race-cultivar specificity denotes specific compatibility, or discrimination between
a potential host and potential pathogen at the sub-species level. Like Flor, plant
pathologists frequently concentrate on the race-cultivar level of compatibility since it
appears to represent a high degree of specialization and can often be resolved into

complementary genic systems involving allelic differences at single genes.

The Genetic System

The subject of this dissertation is a complementary genic system involving the
fungus Cochliobolus carbonum Nelson race 1 and its specific host, Zea mays L. A
combatible or pathogenic interaction results when a spore from a pathogenic race of C.
carbonum encounters a susceptible maize line (cultivar) to produce a disease known as

Helminthosporium leaf spot.

 

 

 

17w

SUS

lon

int

The Host

Helminthosporium leaf spot came to the attention of pathologists and plant breeders
when it was discovered as a new foliar disease of maize in the midwestem Corn Belt
region of the United States. Severe fungal infection of a specific inbred line of corn was
noted during field trials in Indiana in 1938—39 (Ullstrup, 1941). In addition to the original
susceptible inbred, Pr, other susceptible inbred stocks were soon discovered. The F,
progeny of single crosses between these susceptible inbreds were infected with equal
severity, suggesting that a single recessive gene governed susceptibility to C. carbonum
race 1 (Ullstrup, 1944). This notion was confirmed when a single Mendelian locus for
susceptibility to Helminthosporium leaf spot, called hm, was genetically mapped to the
long arm of chromosome-l (Ullstrup and Brunson, 1947). The dominant allele Hm
conditioned full resistance to C. carbonum race 1, even in the heterozygote. It was
postulated that this pathogen went undetected in the field due to the cultivation of open-
pollinated com, a largely heterozygous population. The disease’s full severity was not
discovered until the development and cultivation of the susceptible inbred, Pr (Ullstrup,
1941). Fortunately, full resistance was detected in most other commercial maize lines,
and incorporation of Hm into new stocks became standard practice in maize brwding
programs.

It should be noted that additional genetic elements in maize have been implicated
in resistance to C. carbonum race 1. A separate locus, non-allelic to Hm, conditions
developmentally limited resistance to C. carbonum race 1. This locus, called Hm2 is
located on chromosome 9L, displays incomplete dominance (heterozygotes have

intermediate resistance), and exhibits a dependence on plant age for its expression (Nelson

9

and Ullstrup, 1964). Hm-mediated resistance to C. carbonum race 1 is regarded as
epistatic to Hm2, thus Hm is the primary source of heritable resistance to C. carbonum
race 1 (Nelson and Ullstrup, 1964). This dissertation is focused on the identity and

function of Hm, although Hm2 will be discussed later in Chapters 4 and 5.

The Pathogen

Cochliobolus carbonum Nelson race 1, is a filamentous ascomycete of the
Loculoascomycetes. This genus largely consists of species associated with grass hosts
(Kline and Nelson, 1969). These include C. heterostrophus, a maize pathogen, and C.
victoriae, a pathogen of oats. Cochliobolus is the genus name given to a teleomorph
(sexual) stage of Helminthosporium (Nelson, 1959). The species Cochliobolus carbonum
Nelson (anamorph, Helminthosporium carbonum Ullstrup) is presently distinguished into
three or possibly four separate races based upon their host range on maize (Leonard,
1978). This dissertation is concerned with races 1 and 2 which are distinguished by their
pathogenicity in context with the hm locus of maize. Race 1 of C. carbonum is pathogenic
solely to hm/hm maize, race 2 is a designation for isolates that are non-pathogenic to
hm/hm maize (Nelson and Ullstrup, 1961).

The genetics of C. carbonum pathogenicity were studied after the sexual stage of
the fungus’ life cycle was demonstrated in culture (Nelson, 1959). Segregation data from
crosses between race 1 and race 2 of C. carbonum suggested that pathogenicity against
hm/hm maize was under the control of a single gene (Nelson and Ullstrup, 1961).

In the years to follow, rapid progress was made toward identifying the biochemical

nature of race 1 pathogenicity. This progress was due, in part, to the fact that another

10

member of the genus Cochliobolus was known to produce a low-molecular-weight
phytotoxin that caused necrosis on host tissues (Meehan and Murphy, 1947). The
production of such compounds by pathogens was thought to have a direct implication in
disease development (Scheffer and Livingston, 1984).

For example, specific oat breeds carrying the single gene Vb are highly susceptible
to pathogenic isolates of Cochliobolus victoriae. Victoria Blight of oats occurred on a
major scale in the early 1930’s due to widespread cultivation of Vb-containing oats.
Researchers revealed the involvement of a phytotoxic molecule called victorin that was
produced by pathogenic C. victoriae (Meehan and Murphy, 1947; Luke and Wheeler,
1955). Victorin alone could initiate many of the symptoms of blight on susceptible (Vb-
containing) varieties of oat, with very little effect on the resistant variety (Yoder and
Scheffer, 1969). Victorin was one of the premier examples of a host-selective toxin: a
pathogen-derived metabolite that influences host-range and pathogenicity (Scheffer and
Livingston, 1984). Pathogenicity of C. victoriae was linked to the production of victorin
(Scheffer et al. , 1967), and in oat, susceptibility to pathogenic C. victoriae was
indistinguishable from sensitivity to victorin (Yoder and Scheffer, 1969).

Given its close relationship to C. victoriae, researchers began examining the
interaction between maize and C. carbonum race 1 for a similar phenomenon. Like the
Victoria blight of oats, the symptoms of Helminthosporium leaf spot included yellowing
of the leaf tissue well in advance of the actual lesion area; suggestive of an effect caused
by a diffusible substance.

Pathogenicity of C. carbonum, already shown to be under the control of a single

gene, was soon associated with the ability to produce a low-molecular-weight compound

1 1
that displayed selective toxicity toward hm/hm maize (Scheffer and Ullstrup, 1965). The

crude culture filtrate from C. carbonum race 1 was shown to differentially inhibit the
growth of susceptible (hm/hm) corn roots versus the growth of nearly isogenic (Hm/-)
resistant roots (Pringle and Scheffer, 1967). This toxin was given the name HC-toxin for
the fungus’ anamorph name Helminthosporium carbonum. N on-pathogenic, race 2 isolates
of the fungus did not produce this toxin (Scheffer and Ullstrup, 1965). Progeny from
crosses between race 1 and race 2 isolates of C. carbonum segregated 1:1 for
productionzno production of HC-toxin (Scheffer et al. , 1967). All isolates deficient in
toxin production, including mutants and non-pathogenic wild-types, fail to initiate disease
on hm/hm maize.

Matings were also carried out between pathogenic C. carbonum and pathogenic C.
victoriae. To a degree, these species are sexually compatible. The progeny were scored
for the production of toxins and evaluated for their disease phenotype on both susceptible
corn and oats. Progeny from interspecific crosses segregate 1: 1: 1 :1 for production of HC-
toxin or victorin (parental phenotypes), both toxins (fully pathogenic, non-parental
recombinant), or neither toxin (non-pathogenic, non-parental recombinant). As sexually
compatible anamorphs, these two fungi are thought to differ by two gene pairs, one locus
controlling the production of HC—toxin in C. carbonum, the other of victorin in C.
victoriae (Scheffer et al. , 1967).

The importance of HC-toxin to the race 1 infection process was demonstrated by
supplying purified HC-toxin to race 2 spores inoculated on hm/hm maize leaves. In the
presence of HC-toxin, race 2 inoculum was able to penetrate and infect hm/hm maize as

effectively as a race 1 isolate (Comstock and Scheffer, 1973). Similar results were

12

observed when HC-toxin was provided to inocula of C. victoriae on hm/hm leaves;
exogenously applied HC-toxin was able to extend the host-range of C. victoriae to include
hm/hm maize (Comstock and Scheffer, 1973). Production of HC—toxin is considered the
critical required element for pathogenicity against hm/hm maize. Genetic and
physiological data supported the notion that the single gene controlling pathogenicity of
C. carbonum race 1 was responsible for the production of HC-toxin. This genetic locus
was deemed TOX2 (Yoder et al. , 1989).

With a basis for race 1 pathogenicity established, the susceptibility of specific
maize lines to race 1 was purported to involve differential sensitivity to HC-toxin. A
survey of maize lines exhibiting different degrees of susceptibility to C. carbonum race
1 were analyzed for their sensitivity to HC-toxin. The maize lines most sensitive C.
carbonum race 1 were also most sensitive to HC-toxin in the root growth bioassay, while
lines most resistant to the fungus were most resistant to the toxin (Kuo and Scheffer,
1970a). Thus, HC-toxin represents a chemical entity critical to the infection process, that
also displays selective biological activity toward maize depending on the allelic condition
at the hm locus.

Returning to an updated version of our quadratic check (Figure 1.2), some of the
genetic details of Helminthosporium leaf spot can be entered. C. carbonum race 1 is
pathogenic in the context of hm/hm maize, and the genetic trait controlling pathogenicity
was associated with the ability to produce a toxin. The designation P from our
hypothetical interaction can be replaced by TOXZ, the single locus required for HC-toxin
production and pathogenicity against hm/hm maize. A single completely dominant gene

from maize, denoted Hm/-, exists as the complementary gene for resistance to TOX2.

13
Race 2 isolates are designated tox2 (replacing p) because they do not produce HC-toxin

and are non-pathogenic against either Hm/- or hm/hm maize.

A compatible interaction between C. carbonum and maize is that which results in
disease. On susceptible (hm/hm) leaf surfaces, C. carbonum race 1 spores, which contain
TOX2 and produce HC-toxin, form specialized structures for attachment and penetration
called appressoria. Once penetration is established, the fungus grows rapidly throughout
the leaf mesophyll resulting in the collapse of host cell walls and the breakdown of
organelles. As infectious hyphae advance, the boundaries of necrotic lesions expand,
creating large leaf spots (Jennings and Ullstrup, 1957; Comstock and Scheffer, 1973).

The interaction of resistant leaf tissues with C. carbonum race 1 is said to be
incompatible. Germinating race 1 spores form appressoria and penetrate the leaf
epidermis, but because of the resistance factor provided by Hm, infection is confined to
only one or two epidermal cells. The end result of an incompatible (resistant) response
is small necrotic flecks on the leaf surface, similar to the hypersensitive response common
to other diseases (Jennings and Ullstrup, 1957 ; Comstock and Scheffer, 1973).

C. carbonum race 2 spores, which are ton and do not produce HC-toxin, are non-
pathogenic on both Hm/- and hm/hm maize. This interaction is incompatible in either
Case; both host genotypes are resistant to race 2 and display the typical small necrotic
flecks around the sites of fungal penetration (Jennings and Ullstrup, 1957; Comstock and

Scheffer, 1973).

r-

 

14

Figure 1.2. A complementary genic interaction between C. carbonum
race 1 and maize. Gene P from Fig. 1.1 has been replaced by the C.
carbonum race 1 gene TOX2, which controls pathogenicity and the
production of HC-toxin. The complementary host resistance gene receives
the designation Hm, for resistance to race 1. hm confers susceptibility (+)
in context with TOX2. Race 2 fungi are ton, and non-pathogenic (-)
against either Hm or hm.

 

15

Cochliobolus carbonum/maize
Maize Genotype

Hml- hm/hm

TOX2 _ +

 

Fungus
Genotype

 

tox2 _' —

 

 

 

 

Figure 1.2.

16
HC—toxin

HC-toxin’s role in pathogenicity inspired efforts to identify its structure. Initial
efforts to purify HC-toxin from culture filtrates were aided extensively by the development
of an effective bioassay that evaluates the inhibition of root growth of susceptible (hm/hm)
corn seeds as a function of HC-toxin concentration (Pringle and Scheffer, 1967).
Qualitatively, toxic fractions will differentially inhibit the growth of susceptible roots
when compared to growth of resistant (Hm/-) controls. Quantitatively, toxin preparations
can be evaluated against a standard curve; typical homogeneous preparations of HC-toxin
have an EC,o (effective concentration inhibiting susceptible root growth by 50%) between

0.2 and 0.5 rig/m1 (Ciuffetti et al. , 1983).

Structure

Many years of effort in HC—toxin purification and structure elucidation can be
summarized by Figure 1.3 which shows the structure of HC-toxin (Walton et al. , 1982;
Pope et al. , 1983; Leisch er al. 1982). The compound is a cyclic tetrapeptide, cyclo[D-
prolyl-L-alanyl-D-alanyl-L—Aeo], where Aeo stands for the amino acid component 2-
amino-9,10—epoxy-8-oxodecanoic acid. Aeo is a peculiar long chain amino acid originally
described as a component of another fungal cyclic tetrapeptide called Cyl-2, produced by
the phytopathogenic fungus, Cylindrocladium scoparium (Hirota et al. , 1973).
Confirmation of HC-toxin’s structure was obtained when a completely synthetic HC-toxin
was shown to exhibit specific toxicity toward susceptible corn in the root growth bioassay

(Kawai and Rich, 1983).

17

 

Figure 1.3. The structure of HC-toxin. HC-toxin is a cyclic
tetrapeptide, cyclo-[D-prolyl~L-alanyl-D-alanyl-L-Aeo], where Aeo stands
for 2-amino-9,10—epoxy-8-oxodecanoyl.

 

18

D-Ala
O\ (EHa
\ C\N/C 0 L-A3 0
L-Ala é \c¢
CH — I

HC-toxin

Figure 1.3.

19
Biosynthesis

HC-toxin’s novel structure and relationship to disease spurred interest in its
biosynthesis. Cyclic peptides of varying size and character are common as antibiotics or
secondary metabolites from bacteria and fungi. Considerable research into the synthesis
of so-called non-ribosomal peptides has been conducted. Some of the best examples
describe the biosynthesis of the antibiotic gramicidin-S, produced by the bacterium,
Bacillus brevis (Kleinkauf and von Dohren, 1983), and the enniatins, cyclic depsipeptide
ionophores produced by the fungus, Fusarium oxysporum (Zocher et al. , 1982). In these
and all subsequent cases, cyclic peptides were found to be synthesized by large, cytosolic,
polyfunctional enzymes. Amino/imino acids are activated by ATP, forming an aminoacyl-
AMP and PP,. The activated amino acids are transferred to enzyme thioesters (4’-
phosphopantetheine groups) for condensation into cyclic peptides (Kleinkauf and von
Dohren, 1983). Purification and characterization of these synthetase enzymes can be
followed by monitoring amino acid-dependent ATP/PPi exchange; exploiting the
reversibility of the activation reaction using radiolabeled pyrophosphate (Zocher et al. ,
1982).

Based on ATP/PPi exchange activity, two C. carbonum race 1 enzymes were
implicated in the synthesis of HC-toxin. These large enzymes, called HC-toxin
synthetases (HTS), are found in cytosolic preparations of race 1 of C. carbonum, but are
lacking in race 2 (Walton, 1987). HTS-1, with an apparent molecular weight of 220 kD
based on SDS-PAGE, specifically recognizes and activates the L—isomer of proline and
epimerizes it to the D-isomer for incorporation into HC-toxin. The second enzyme, HTS-

2 (160 kD on SDS-PAGE), recognizes and activates both L— and D-alanine, and also

2O
epimerizes L—alanine to D-alanine (Walton and Holden, 1988). Both isomers of alanine

are found in HC-toxin, yet HTS-2 catalyzes epimerization only in the L- to D- direction.
Both enzyme activities are found only in toxin producing isolates of the fungus, and co-
segregate with toxin production in sexual crosses between race 1 and race 2 isolates
(Walton, 1987).

The TOX2 locus of C. carbonum race 1 was thought to encode these enzymes.
Molecular genetic analyses have identified a 15.7kb ORF that is required for HC-toxin
biosynthesis. Interestingly, this massive ORF appears to encode both HTS-1 and -2 as a
single gene. Separable HTS-1 and HTS-2 enzymes in race 1 extracts are perhaps products
of proteolytic cleavage following translation (Scott-Craig et al. , 1992). The function of
the TOX2 locus has been investigated by transformation-mediated gene disruption of C.
carbonum race 1. Importantly, both HC-toxin biosynthesis and race 1 pathogenicity are
nullified by disrupting the TOX2 gene. When inoculated on hm/hm maize, the TOX2
knockout mutants produce only small necrotic flecks similar to non-pathogenic race 2
isolates. These findings constitute the most substantial proof that HC-toxin is required for
pathogenicity against hm/hm maize (Panaccione et al. , 1992). In turn, the results also
support the hypothesis that Hm-mediated resistance to C. carbonum race 1 is synonymous
with resistance to HC-toxin.

To date, the DNA included in TOX2 does not provide any clues to the biosynthesis
or activation of Aeo. An epoxide-containing compound thought to be a precursor of Aeo
or HC-toxin has been detected as a by-product in culture filtrates from the TOX2 knockout
mutants (D. Panaccione, personal communication). Its characterization may provide

insight into Aeo biosynthesis.

2 1
Biological Activity

It is evident from a number of studies that production of HC-toxin must, in some
way, give a critical advantage to a germinating fungal spore for successful colonization
of susceptible leaf tissues. Many researchers have examined susceptible and resistant
maize tissues for a differential response to HC-toxin in attempts to explain the toxin’s
mode of action and/or the role of the Hm allele in disease resistance. Paradoxically, some
of the first studies documented stimulatory effects by this "toxin". For example, root
growth and general metabolism of both resistant and susceptible maize are actually
stimulated at low concentrations of HC-toxin (Kuo and Scheffer, 1970a,b). Evidence of
this can be seen in Figure 1.4. At low HC-toxin concentrations in the root growth
bioassay, the growth of resistant roots is increased in comparison to untreated controls.
In addition, stimulatory effects on leaf respiration, COz-fixation (Kuo and Scheffer,
1970b), and solute uptake (Yoder and Scheffer, 1973a,b) have been documented for toxin-
treated maize tissues.

Importantly, resistant maize is actually sensitive to HC-toxin when the
concentration is increased lOO-fold (Kuo and Scheffer, 1970a). This effect is also
illustrated in Figure 1.4. HC-toxin administered at 0.2 ug HC-toxin/ml inhibits
susceptible root growth by 50% . Root growth of resistant maize is 50% inhibited when
HC-toxin is given at 20 ug/ ml. These maize lines are near-isogenic, thus the data suggests
that a 100-fold differential in HC-toxin sensitivity is attributable to the Hm allele. This
differential has been observed repeatedly regardless of the biological response being

examined.

 

 

22

Figure 1.4. Root growth bioassay of HC-toxin. A repeat of the classic
experiment documented by Pringle and Scheffer (1967). Purified HC-toxin
is administered to germinated corn seedlings from the near-isogenic hybrids
Per61 (hm/hm, O) and PrlxK6l (Hm/hm, O). Dotted vertical lines
indicate toxin concentration causing 50% inhibition of root growth (ECSO).

23

160—

1’20- \ \‘+

 

 

Root Growth (mm)
00
C?

4).

O
l

/

\ ‘+
\
tr e
O.— l {P l T l 1 I l ' l 3 ' l
00 1.0 2.0 3.0 4.0 5.0
[HO—toxin] (log [ng/mID

\GD

 

 

Figure 1.4.

24

The following are documented biological effects of HC-toxin:

Perturbation of the plasma membrane. HC-toxin causes a rapid but transient
increase in negative electropotential across the plasma membrane of susceptible coleoptile
cells (Gardner et al. , 1974). Within 2—3 minutes after toxin exposure, increases from 10
to 40mV were recorded, followed by a gradual return to the original level. An increase
in negative electropotential is dramatically different from the effect of other toxins such
as victorin (on oats) and PC—toxin (on grain sorghum) which cause significant decreases
in negative electropotential across plasma membranes, K+ ion loss, and electrolyte
leakage. The increase in potential induced by HC-toxin is perhaps indicative of a direct
effect on the electrogenic ion pumps of the plasma membrane (Gardner et al. , 1974).

Eflects on solute uptake. Significant increases in nitrate uptake occur 2-3 h after
toxin treatment of susceptible tissues (Yoder and Scheffer, 1973a). Increases in nitrate
reductase activity stem from increased availability of N 03‘. Uptake of Na+ , Cl', leucine,
and methylglucose is also stimulated to a lesser extent by treatment of susceptible roots
with toxin, while uptake of N02} K“, P0433 Ca“, 803', and glutamate are unaffected
(Yoder and Scheffer, 1973b). In addition, HC-toxin appears to affect solute uptake only
on the plasma membrane; no effects on membrane permeability are observed when maize
mitochondria and chloroplasts are isolated from toxin-treated cells (Yoder, 1971).
Ionophoric compounds can be potent toxins, but the data for HC-toxin are not consistent
with the physical properties and ion selectivity of microbial ionophores (Dobler, 1981).

Difl’erential inhibition of chlorophyll synthesis in leaf tissue. Susceptible maize
leaves, when exposed to HC-toxin, begin to show inhibition of chlorophyll synthesis after

6 h in light. Protection from inhibition can be afforded by supplying leaves with the

25

chlorophyll precursor, 5-aminolevulinic acid. Resistant leaves show the same inhibition
at 100-fold higher HC-toxin concentration (Rasmussen and Scheffer, 1988a).

Efi’ects on chloroplast clustering and difi’erentiation. Maize leaf protoplasts display
a tendency to cluster their chloroplasts together after several hours in cell culture. HC—
toxin appears to inhibit this phenomenon (Earle and Gracen, 1982). Susceptible
protoplasts appear smaller, with uniformly distributed chloroplasts following treatment
with HC-toxin. Other observations of protoplasts indicate a cytostatic effect by HC-toxin
that actually helps extend the life of susceptible protoplasts in culture (Wolf and Earle,
1991). This effect is quite different from toxins like victorin, which is much more
pernicious. Victorin collapses susceptible oat protoplasts after only a few hours of toxin
treatment (Earle and Gracen, 1982).

Biological activity in mammalian systems. Other cyclic tetrapeptides produced by
fungi contain Aeo (Closse and Huguenin, 1974; Hirota et al. , 1973; Kawai et al. , 1986).
Several of these structures are shown in Figure 1.5. Aeo-containing cyclic tetrapeptides
have been shown to be biologically active in mammalian systems (Closse and Huguenin,
1974). Cytostasis and inhibition of both leucine and uridine incorporation in cultured
murine P-815 mastocytoma cells was reported in response to HC-toxin, chlamydocin, and
Cyl-2 (Walton et al., 1985). HC-toxin and chlamydocin were also found to inhibit
mitogenesis of Con A“ stimulated lymphocytes (Shute et al. , 1987). The discovery of
these antimitogenic properties generated interest in Aeo-containing cyclic peptides as
chemotherapeutic agents for cancer treatment (Shute et al. , 1987). Thus, the biological

activity of HC-toxin, or other Aeo-containing cyclic peptides, is not restricted to plants.

26

.eBNESm
335mm memee 2: 3 3260a .Hoof;-Eoooaafi-_>o=o_-q-_Ȥ_w$=o;a-a.293 . 38m?

Surgeon §N§~8€§6
Ewes... 2: .3 32605 .Hoo<fiabo8aaéaxo=28M-A-_>mpoifioE.O-B.e_o>o .Nénu

63983539 SonnegaefiQ
mswce 05 an 30:85 .Hoo<-q-58993533294-_.Cb:eom_oEE£.298 .Eoonzzafio

June 8?: 823333“ 9:93 u§5=e§< .550 .mA «Suw—

 

27

.2 ennui

5343
2E...
x w t
O// \ 2 / do
\o, a \u 29/
«519101393: 9.. use
,
ocdé 2h 2:
/ ER ”.3
asu/ z \ azurzoA
u u, n
/z\ 6 .3
33
2.3
«.39 58352.5
n
ear—35956.9 :8 a;
n
a o o,,o\u=z/ \5
z w N \ / __ mu \ u
z .8 3.3:? :35: n x /
Pd\ /=m \ .6 one
\a/ m 3 \ M“: 8.3 :2 _
£81510: _ :3qu / r “z
2:. a . =2 o\\ / z 29/
=2/ fix a o/zlu.,\ «.8
o x :9 o:... aha—é C a
o z \ sex

2:. .._

28

Structural Requirements

Relevant discoveries about HC-toxin’s structure were made during the analyses of
HC-toxin’s biological activity. For example, destruction of the terminal epoxide moiety
of Aeo via acid hydrolysis results in nearly complete loss of toxicity to susceptible roots
(Ciuffetti et al. , 1983; Walton and Earle, 1983). In addition, the 8-keto group of Aeo,
vicinal to the terminal epoxide, is required for biological activity of HC-toxin. Sodium
borohydride selectively reduces this carbonyl group to the corresponding alcohol. Both
8-hydroxy isomeric products of this reaction fail to inhibit susceptible root growth in the
bioassay (Kim et al. , 1987).

Aeo by itself is not biologically active against root growth or against mammalian
cells (Shute et al. , 1987). Therefore, in addition to Aeo, other structural elements must
be important for biological activity. Evidence suggests that a significant contribution is
made by the 12-membered cyclic peptide ring. In culture, C. carbonum race 1 produces
small amounts of alternate forms of HC-toxin, each of which contain Aeo (Rasmussen and
Scheffer, 1988b). The primary form of HC-toxin (toxin I) is produced in the greatest
amount, and has the greatest biological activity. The alternate forms have essentially the
same host-selectivity in the root growth bioassay (the 100-fold differential is maintained),
but the potency of the compounds is reduced. For instance, HC-toxin III, which contains
hydroxyproline in place of the proline residue, has an EC,0 of 2.0 ug/ ml against
susceptible roots. The addition of one hydroxyl group to the B—carbon of proline lowers
the potency of the compound by approximately 90% (Rasmussen and Scheffer, 1988b).
However, Aeo-containing cyclic peptides composed of other amino/imino acids have

different effects in the root growth bioassay. Chlamydocin contains phenylalanine and

29

aminoisobutyrate in addition to proline and Aeo (see Figure 1.5). Chlamydocin is not as
active as HC-toxin against susceptible root growth, but is more active than HC-toxin
against resistant roots (Walton et al. , 1985) such that the differential is less than the 100-
fold differential recorded for HC-toxin. Therefore, the peptide ring appears to contribute

to the potency and/or host-selectivity of Aeo-containing cyclic peptides.

The Working Hypothesis

Since the detection of Helminthosporium leaf spot disease in the field, research has
uncovered a complementary genetic relationship between fungal pathogenicity and host
resistance. The fungus Cochliobolus carbonum Nelson race 1, as part of a group of plant
pathogens that produce host-selective toxins, presents to us a level of detail not often
available in other host-pathogen systems. Here, a body of evidence suggests a scenario
in which a small molecule, specifically a secondary metabolite produced by a pathogenic
fungus, is absolutely required for the initiation and development of disease on the
susceptible host.

The TOX2 locus of C. carbonum race 1 confers the ability to synthesize this
defined substance. A last look at our quadratic check indicates a unique opportunity to
characterize resistance to this compound at the biochemical and molecular genetic levels.
A tremendous amount of detail can be conveyed by including HC-toxin into the equation
as the ultimate gene product of our hypothetical gene P (Fig. 1.6). The challenge remains
to describe how the Hm gene product interacts with HC-toxin. Data on this point will
serve to provide a biochemical and/or molecular genetic definition of "recognition" in a

host—pathogen interaction.

 

30

Figure 1.6. Resistance of maize to C. carbonum defined in context with
the end-product of the C. carbonum pathogenicity gene. The production
of HC-toxin is the defining element of C. carbonum race 1 pathogenicity
against hm/hm maize. Resistance to C. carbonum race 1, mediated by Hm,
is defined as resistance to the end product of the pathogenicity gene (HC-
toxin +). Race 2 isolates (tox2) do not make HC-toxin and are non-
pathogenic (HC-toxin-) against Hm/— or hm/hm.

31

Cochliobolus carbonum/maize

 

 

Maize Genotype
Hml- hm/hm
no... , — +
Fungus
Phenotype
HC-toxin- _ '—

 

 

 

 

Figure 1.6.

32

In the past, researchers have addressed Hm-mediated resistance to HC-toxin by
examining the effects of HC-toxin on maize tissues. The lOO-fold differential in
sensitivity between hm and Hm has traditionally been regarded as a critical difference in
toxin binding to a receptor, or a difference in transduction of a toxin binding signal to a
response (Scheffer and Livingston, 1984). A receptor model for toxin resistance is
perhaps the approach most consistent with a theory based on pathogen "recognition",
calling to mind images of "lock-and-key" models. But the array of responses to HC-toxin
is quite complex, perhaps suggesting that a receptor-mediated mechanism of differential
sensitivity is insufficient.

In general, HC-toxin’s biological activities are quite diverse and subtle, especially
when compared to powerful phytotoxins like victorin (Scheffer and Livingston, 1984).
Thus, since a specific mode of action for HC-toxin remains unresolved, the receptor
model is open to question. Can a single binding site account for sensitivity to HC-toxin
such that it results in differential inhibition of root growth as well as differential inhibition
of chlorophyll synthesis? Perhaps, but there remains a great difficulty in delineating the
primary from secondary effects of this toxin. As an alternative, it is reasonable to propose
that HC-toxin has more than one site of action, or more importantly, that specific toxin
resistance controlled by Hm occurs by some process independent of toxin action.

Rasmussen and Scheffer (1988b) have concluded that resistant maize has a toxin-
sensitive site similar to susceptible maize. While this putative site of action is important
to our understanding of the role of toxins in disease, the current hypothesis focuses on the
idea that resistant maize may preferentially modify this foreign compound. Given the

structural requirements for toxin activity, the question of Hm-specific metabolism of the

33

toxin must be addressed. Differential sensitivity to HC-toxin between Hm and hm might
be explained by a resistance mechanism based on detoxification. With HC-toxin as its
substrate, saturation of a detoxicative enzyme in resistant maize might explain the lOO—fold
differential in toxin sensitivity. HC-toxin above a putative saturation point represents
toxin free to act upon its site(s) in the resistant cell.

A fungal secondary metabolite produced in infected maize cells represents a
compound foreign to the host. The working hypothesis of this dissertation treats this as
a problem of xenobiotic metabolism in plants. Plants, including maize, produce several
families of enzymes whose role it is to detoxify foreign compounds (Shimabukuro et al. ,
1971; Lamoreaux and Rusness, 1986). An enzymatic basis for HC-toxin inactivation is

proposed to form the basis of Hm-mediated resistance to C. carbonum race 1.

34
REFERENCES

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

Closse, A., and Huguenin, R. (1974). Isolierung und strukturauflclarung von
chlamydocin. Helv. Chim. Acta. 57: 533-545.

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.

Dobler, M. (1981). Ionophores and Their Structures, Wiley & Sons, New York.

Earle, E.D., and Gracen, V.E. (1982). Effects of Helminthosporium phytotoxins on
cereal leaf protoplasts. In Plant Tissue Culture: 1982, Proc. Fifth Intl. Cong. Plant
Tissue and Cell Cult., 663-664.

Flor, 11.11. (1942). Inheritance of pathogenicity in Melampsora lini. Phytopathology 32,
65 3-669.

Flor, H.H. (1956). The complementary genic systems in flax and flax rust. Advan.
Genet. 8, 29-54.

Frank, S.A. (1992). Models of plant-pathogen coevolution. Trends Genet. 8, 213-219.

Gabriel D.W., and Rolfe, B.G. (1990). Working models of specific recognition in plant-
microbe interactions. Annu. Rev. Phytopath. 28, 365-391.

Gardner, J .M., Scheffer, R.P., and Higinbotham, N. (1974). Effects of host-specific
toxins on electropotentials of plant cells. Plant Physiol. 54, 246-249.

Gasser, C.S., and Fraley, R.T. (1989) Genetically engineering plants for crop
improvement. Science 244, 1293-1299.

Heath, M.C. (1991). The role of gene-for-gene interactions in the determination of host
species specificity. Phytopathology 81, 127-130.

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.

Jennings, P.R., and Ullstrup, AJ. (1957). A histological study of three
Helminthosporium leaf blights of corn. Phytopathology 47, 707

35

Kawai, M., and Rich, D.H. (1983). Total synthesis of the cyclic tetrapeptide, HC-toxin.
Tet. Lett. 24, 5309-5312.

Kawai, M., Pottorf, R.S., and Rich, DH. (1986). Structure and solution conformation
of the cyctostatic cyclic tetrapeptide WF-3161. J. Med. Chem. 29, 2409-2411.

Keen, N.T. (1990). Gene-for-gene complementarity in plant-pathogen interactions.
Annu. Rev. Genet. 24, 447-463.

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

Kleinkauf, H., and von Dbhren, H. (1983). Non-ribosomal peptide formation on
multifunctional proteins. Trends Biochem. Sci. 8, 281-283.

Kline, D.M., and Nelson, R.R. (1969). Inheritance of factors in Cochliobolus carbonum
conditioning symptom expression on grass hosts. Phytopathology 59, 1133-1135.

Kuo, M-S., and Scheffer, R.P. (1970a). Comparative specificity of the toxins of

Helminthosporium carbonum and Helminthosporium victoriae. Phytopathology 60,
365 -368.

Kuo, M-S., and Scheffer, R.P. (1970b). Comparative effects of host-specific toxins and
Helminthosporium infections on respiration and carboxylation by host tissue.
Phytopathology 60, 1391-1394.

Lamoreaux, G.L., and Rusness, D.G. (1986). Xenobiotic conjugation in higher plants.
In Xenobiotic Conjugation Chemistry (Paulson,G.D. et al., eds.). ACS Symp.
Series 299, 62-105.

Leisch, J.M., Sweeley, C.C., Staffeld, G.D., Anderson, M.S., Weber, DJ., and
Scheffer, R.P. (1982). Structure of HC-toxin, a cyclic tetrapeptide from
Helminthosporium carbonum. Tetrahedron 38, 45-48.

Leonard, KJ. (1978). Polymorphisms for lesion type, fungicide tolerance, and mating
capacity in Cochliobolus carbonum isolates pathogenic to corn. Can. J. Bot. 56,
1809-1815.

Meehan, F., and Murphy, H.C. (1947). Differential phytotoxicity of metabolic by-
products of Helminthosporium victoriae. Science 106, 270-271.

Nelson, O.E., and Ullstrup, A.,]. (1964). Resistance to leaf spot in maize; genetic
control of resistance to race 1 of Helminthosporium carbonum Ull. J. Heredity 55,
195-199.

 

'4
r.
In.
-l' .

36

Nelson, R.R. (1959). Cochliobolus carbonum, the perfect stage of Helminthosporium
carbonum. Phytopathology 49, 807-810.

Nelson, R.R., and Ullstrup, AJ. (1961). The inheritance of pathogenicity in
Cochliobolus carbonum. Phytopathology 51, 1-2.

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. Nat. Acad. Sci. USA 89, 6590-6594.

Pope, M.R., Ciuffetti, L.M., Knoche, H.W., McCreary, D., Daly, J .M., and Dunkle,
L.D. (1983). Structure of the host-specific toxin produced by Helminthosporium
carbonum. Biochemistry 22, 3502-3506.

Pringle, R.B., and Scheffer, R.P. (1967). Isolation of the host-specific toxin and a
related substance with nonspecific toxicity from Helminthosporium carbonum.
Phytopathology 57, 1169-1172.

Rassmussen, J .B. and Scheffer, R.P. (1988a). Effects of selective toxin from
Helminthosporium carbonum on chlorophyll synthesis in maize. Physiol. Mol.
Plant Path. 32, 283-291.

Rasmussen, J .B., and Scheffer, R.P. (1988b). Isolation and biological activities of four
selective toxins from Helminthosporium carbonum. Plant Physiol. 86, 187-191.

Scheffer, R.P., and Livingston, R.S. (1984). Host-selective toxin and their role in plant
diseases. Science 223, 17-21.

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

Scott-Craig, J .S., Panaccione, D.G., Pocard, J .-A., and Walton, JD. (1992). The
cyclic peptide synthetase catalyzing HC-toxin production in the filamentous fungus
Cochliobolus carbonum is encoded by a 15.7kb open reading frame. J. Biol.
Chem. (in press).

Shimabukuro, R.R., Frear, D.S., Swanson, H.R., and Walsh, W.C. (1971).
Glutathione conjugation: an enzymatic basis for atrazine resistance in corn. Plant
Physiol. 47, 10-14.

37

Shute, R.E., Dunlap, B., and Rich, D.H. (1987). Analogues of the cytostatic and
antimitogenic agents chlamydocin and HC-toxin: synthesis and biological activity
of chloromethyl ketone and diazomethyl ketone functionalized cyclic tetrapeptides.
J. Med. Chem. 30, 71-78.

Ullstrup, A.J. (1941). Two physiologic races of Helminthosporium maydis in the corn
belt. Phytopathology 31, 508-521.

Ullstrup, A.J. (1944). Further studies on a species of Helminthosporium parasitizing
corn. Phytopathology 34, 214-222.

Ullstrup, A.J., and Brunson, A.M. (1947). Linkage relationships of a gene in corn
determining susceptibility to a Helminthosporium leaf spot. J. Am. Soc. Agron.
39, 606-609.

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

Walton, J .D. (1991) Genetics and biochemistry of toxin synthesis in Cochliobolus
(Helminthosporium). In Molecular Industrial Mycology, Systems and Applications
for Filamentous Fungi (Leong, S.A., Berka, R.M., eds.). Marcel Dekker, New
York, 225-249.

Walton, J .D. and Earle, E.D. (1983). The epoxide in HC-toxin is required for activity
against susceptible maize. Physiol. Plant Path. 22, 371-376.

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. Comm. 107, 785-794.

Walton, J.D., Earle, E.D., Stiihelin, 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, ER. (1988). Properties of two enzymes involved in the
biosynthesis of the fungal pathogenicity factor HC-toxin. Molec. Plant Microbe
Interact. 1, 128-134.

Wolf, SJ., and Earle, E.D. (1991). Effects of Helminthosporium carbonum race 1 toxin
on host and non-host cereal protoplasts. Plant Science 70, 127-137.

Yoder, O.C. (1971). Selective effects of the host-specific toxin from Helminthosporium
carbonum on cellular organelles and on solute absorption. PhD. Thesis, Michigan
State University.

38

Yoder, O.C., and Scheffer, R.P. (1969). Role of toxin in early interaction of
Helminthosporium victoriae with susceptible and resistant oat tissue.
Phytopathology 59, 1954-1959.

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

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

Yoder, O.C., Turgeon, B.G., Ciuffetti, L., and Schafer, W. (1989) Genetic analysis
of toxin production by fungi. In Phytotoxins and Plant Pathogenesis (Graniti, A.
et al. , eds.) Springer-Verlag, Berlin, 43-60.

Zocher, R. Keller, U., and Kleinkauf, H. (1982). Enniatin synthetase, a novel type of
multifunctional enzyme catalyzing depsipeptide synthesis in Fusarium oxysporum.
Biochemistry 21, 43-48.

CHAPTER 2*

ENZYMATIC DETOXIFICATION OF HC-TOXIN

* originally published in Plant Physiology:
0 1991, American Society of Plant Physiologists

Meeley, R.B., and Walton, J .D. (1991). Enzymatic detoxification of

HC-toxin, the host selective cyclic peptide from Cochliobolus
carbonum. Plant Physiol. 97, 1080-1086.

39

40
ABSTRACT

Resistance to the fungal plant pathogen Cochliobolus carbonum race 1 and to its host-
selective toxin, HC-toxin, is determined by Hm, a single dominant gene in the host plant
Zea mays L. Radiolabeled HC-toxin of specific activity 70 mCi/mmol, prepared by
feeding tritiated D,L—alanine to the fungus, was used to study its fate in maize leaf
tissues. HC-toxin was converted by resistant leaf segments to a single compound,
identified by MS and NMR as the 8-hydroxy derivative of HC-toxin formed by reduction
of the 8-keto group of 2-amino-9,10—epoxy-8-oxodecanoic acid, one of the amino acids
in HC-toxin. Reduction of HC-toxin occurred in cell-free preparations from etiolated
(Hm/hm) maize shoots, and the activity was sensitive to heat and proteolytic digestion,
dependent on NADPH, and inhibited by p-hydroxymercuribenzoate and disulfrram. The
enzyme (from the Hm/hm genotype) was partially purified by ammonium sulfate
precipitation and DEAE-ion exchange chromatography. By gel-filtration chromatog-
raphy, the enzyme had an M, of 42,000. NADH was approximately 30% as effective
as N ADPH as a hydride donor, and flavin-containing co-factors had no effect on activity.
When HC-toxin was introduced to maize leaf segments through the transpiration stream,
leaf segments from both resistant and susceptible maize inactivated toxin equally well
over a time-course of nine hours. Although these data suggest no relationship between
toxin metabolism and host-selectivity, we discuss findings in apparent conflict with the
current data and describe why the relationship between enzymatic reduction of HC-toxin

and Hm remains unresolved.

 

4 1
INTRODUCTION

A number of phytopathogenic fungi, especially in the genera Alternaria and
Cochliobolus, produce low-molecular-weight compounds known as host-selective toxins
that determine their host range and contribute to their virulence (Scheffer, 1976). Race
1 of Cochliobolus carbonum Nelson [Helminthosporium carbonum Ullstrup or Bipolaris
zeicola (Stout) Shoem.] produces a cyclic tetrapeptide, called HC-toxin, that accounts for
its exceptional virulence on Zea mays L. varieties that are homozygous recessive at the
nuclear hm locus (Nelson and Ullstrup, 1964; Scheffer and Ullstrup, 1965). HC-toxin
has the structure cyclo- [D-prolyl-L—alanyl-D—alanyl-L-Aeo](Kawai and Rich, 1983; Kawai
et al., 1983; Leisch et al., 1982; Pope et al., 1983; Walton et al., 1982). Both the
terminal epoxide and vicinal ketone of Aeo are required for biological activity of HC-
toxin and of its naturally-occuring analogs (Ciuffetti et al. , 1983; Closse and Huguenin,
1974; Kim et al., 1987; Walton and Earle, 1983)).

Considerable research has been published on the differential effects of HC-toxin
on resistant and susceptible maize (see Scheffer, 1976), but the mode of action of this
compound remains unknown. As an alternative to studying the effect of HC-toxin on
maize tissues, we have taken an approach that examines the effect of maize tissues on
the biological activity of HC-toxin. Given the requirements for the epoxy-ketone moiety
of Aeo, we sought to determine if the integrity of these groups is maintained in planta.
Plants, including maize, are known to contain enzymes capable of inactivating xenobiotic
compounds (Lamoreaux and Rusness, 1986). For example, atrazine tolerance in maize

is due to elevated levels of glutathione-S-transferase enzymes that inactivate atrazine by

42
conjugation (Shimabukuro et al. , 1971). Knowledge of the biochemical fate of HC-toxin

within maize tissues lays the groundwork to ultimately address the hypothesis that host-
selectivity in maize to race 1 of C. carbonum is due to a difference in ability to detoxify
HC-toxin.

We introduce this topic by presenting a method to prepare tritiated HC-toxin for
use in metabolic studies. The biochemical fate of HC-toxin is described as we report the
conversion of HC-toxin by resistant maize leaves and cell-free extracts to a single non-
toxic compound. The enzyme responsible for detoxification is described, and our initial
comparison of HC-toxin metabolism between resistant and susceptible maize is critically

presented.

MATERIALS AND METHODS
Growth of the Fungus
Maintenance of stock cultures of Cochliobolus carbonum Nelson and liquid
growth conditions were as described (Walton and Holden, 1988). The toxin-producing
isolate SB111 of C. carbonum was originally provided by S.P. Briggs, Pioneer Hi-Bred

International, Johnston, IA.

43
Purification and Analysis of H C-toxin

Methods of toxin purification by solvent extraction and reverse-phase HPLC were
as given (Walton et al. , 1982). HC-toxin was quantified by HPLC and a Spectra-Physics
Model 4270 automatic integrator. HC-toxin identity was confirmed by root growth
bioassay (Walton et al. , 1982), fast-atom bombardment MS, and proton NMR. Mass
spectra were collected on a J EOL Model HXl 10-HF double focusing instrument
(accelerating voltage, 10 kV; ionization, 6 keV xenon beam; mass resolution, 1000).
NMR analyses were performed in CDCl3 on a Varian VXR-S 500 MHz instrument.

The 8-ketone group of Aeo in HC-toxin was specifically reduced to the
corresponding 8-alcohol with sodium borohydride (Kim et al. , 1987). The 9,10-epoxide
group of Aeo was hydrolyzed with TFA (Walton and Earle, 1983). Derivatives and
metabolites of HC-toxin were separated either by reverse-phase HPLC (Walton et al. ,
1982), by TLC (silica 60, Merck) developed in CH2C12zacetone (1: 1, v/v)(Rasmussen and
Scheffer, 1988), or by flash chromatography (Rasmussen and Scheffer, 1988). Epoxides

were detected after TLC by spraying plates with p-NBP (Hammock et al. , 1974).

In vivo Production of Radiolabeled HC-toxin

Static liquid cultures ( 125 mL) of C. carbonum were grown in 1-L Erlenmeyer
flasks. For testing incorporation of various amino acid precursors, 15 uCi of D-
[l‘C]alanine (specific activity 30 to 60 mCi/mmol, ICN); 40 uCi of L-[’H]alanine (30 to
50 Ci/mmol, ICN); or 40 uCi of L-[3H]proline (60 to 100 Ci/mmol, ICN) were added

per flask. In experiments conducted to maximize incorporation of radiolabeled alanine,

the original growth medium from 8- to 14-day-old cultures was replaced under sterile

 

44

conditions by fresh medium containing 5 pCi of [“C]D-alanine (46 mCi/mmol,
Amersham). The original medium, which contained any unlabeled HC-toxin, was
discarded. The length of incubation in the presence of radioactivity was varied from 24
to 96 h. Once favorable conditions for label incorporation and toxin yield were
established, the production of tritiated HC-toxin of high specific activity was initiated by
adding 20 mCi of D,L-[2,3-3H]alanine (59 Ci/mmol, Amersham) to a single flask.
Radioactivity was monitored during HPLC purification with an in-line scintillation-flow
detector (Radiomatic Model CT) at a scintillantzcolumn flow ratio of 3:1 (v/v).
Radioactivity was quantified by scintillation counting, corrected for 3H efficiency, and

HC-toxin mass was quantified by HPLC with absorbance monitoring at 230 nm.

GrOWth of Plant Materials

Caryopses of near-isogenic resistant (Prl x K61 or K61 x Prl, genotype Hm/hm)
or susceptible (Pr x K61 or K61 x Pr, genotype hm/hm) maize (Zea mays L.) were
surface sterilized with 0.5% hypochlorite plus 0.1% Tween-20 for 30 min, rinsed
thoroughly with sterile distilled water, and allowed to imbibe water for 2 to 4 h. For the
production of green leaves, imbibed Ms from both genotypes were sown in 8-inch
diameter clay pots containing a perlitezfine-vermiculite:sphagnum (1:1:1 w/w) mixture.
The pots were sub-irrigated with water. Plants were grown in a growth chamber under
the following conditions: daylength, 16 h; light intensity, 126 uE/mz-sec (PAR cool white
fluorescent lights); day temperature, 21°C; relative humidity, 72%; night length, 8 h;
temperature, 18°C; humidity, 80%.

For production of etiolated shoots for enzyme extraction, imbibed seeds of the

 

45

resistant genotype were sown in flats of vermiculite that had been saturated with half-
strength Hoagland’s solution. The flats were covered with lids and placed in a dark

cabinet for 5 to 6 d.

Metabolism of fH 1H C-toxin by Maize Leaf Segments

HC-toxin was administered to mature green leaves through the transpiration
stream. Segments of green leaves, ca. 10 cm in length, were cut (approximately 10 cm
from the leaf tip) from three- to five-week-old resistant and susceptible plants. The
segments were submerged in a beaker of water, evacuated of intercellular air with a
laboratory aspirator for 30 min, and then blotted dry with paper towels and placed in 18-
mm test tubes containing 5 mL H20, with or without 0.25 uCi of [3H]HC-toxin (0.3
ug/mL). The leaf segments were placed in a lighted laboratory fume hood during uptake
of [3H]HC-toxin.

For extraction, the leaf segments were rinsed thoroughly in deionized water,
frozen in liquid nitrogen, and ground to a powder in a mortar and pestle. Five mL of
methanol were added and the leaves were ground again. The methanolic extracts were
passed through glass fiber filters (Whatman GF/A) and the methanol was evaporated
under vacuum. The aqueous residues were transferred to 1.5 -mL polyethylene microfuge
tubes and centrifuged at 15,000g for 5 min. The supematants were transferred to fresh
microfuge tubes and the pellets discarded. Radioactivity recovered from leaf extracts
was analyzed by HPLC coupled to an in-line scintillation-flow detector or by TLC and

a scanning beta-detector (Bioscan). As a control, 0.25 uCi of [3H]HC-toxin were added

to leaf segments immediately prior to freezing and methanol extraction.

 

46
Preparation of Maize Crude Extracts

All steps were carried out at 4°C or on ice. Etiolated plumule tissue (from 5 to
25 g) from freshly germinated resistant (Hm/-) maize was cut 1 cm below the coleoptilar
node and ground in extraction buffer (0.5 mL per g FW) with a mortar and pestle. The
extraction buffer contained 0.1 M Mops (pH 7.4), 0.3 M sucrose, 5% (w/v) PVP, 10%
(v/v) glycerol, 5mM DTT, lmM EDTA, 15 mM ascorbate, and 0.2 mM PMSF. The
extract was filtered through four layers of cheesecloth, centrifuged at 3,000g for 10 min,

and the supernatant saved.

Enzyme Enrichment

The crude extract was initially fractionated with ammonium sulfate. Material
precipitating between 30% and 55% saturation (30% saturation equals 17.6 g/100 mL)
was collected by centrifugation (10,000g, 10 min) and de-salted by gel filtration (PD-10
column, Pharmacia) in 25 mM potassium phosphate (pH 7.5), 2.5 mM DTT, 1 mM
EDTA, and 1% (v/v) glycerol. The material was further fractionated on an anion
exchange HPLC column (TSK-DEAE-SPW, 7.5 mm x 7.5 cm, Beckman), with a 25-min
linear gradient from 0 to 0.5 M NaCl in the same buffer. The flow rate was 1 mL/min
and 2 mL fractions were collected.

The molecular weight of the enzyme was estimated by gel filtration HPLC on a
TSK-4000 column (30 cm x 7 .5 mm, Beckman) equilibrated with 0.15 M potassium
phosphate (pH 7.2), 1 mM EDTA, and 5 mM DTT. Approximately 1 mg of protein
from an ion exchange fraction containing HCTR activity was loaded onto the column,

and 2-mL fractions were collected. Immediately after elution, ascorbate to 10 mM was

47
added to each fraction. The column was calibrated with the following proteins (Sigma):

thyroglobulin (M, 670,000), IgG (M, 158,000), ovalbumin (M, 44,000), and myoglobin

(M, 17,000).

Enzyme Assay Conditions

Typical assay volumes were 125 uL and contained 115 uL enzyme solution, 4
mM NADPH (Sigma), and 0.25 uCi [3H]HC-toxin (23 uM). Reactions were run at 30°C
for 15 min and were stopped by the addition of an equal volume of chloroform and rapid
mixing. The reactions were extracted twice more with chloroform, and the organic
phases were combined and concentrated under vacuum. Concentrated extracts were
analyzed by TLC or HPLC. Tritiated substrates and products were quantified with a
scanning beta-detector or analyzed by spraying TLC plates with a fluorography enhancer
(enHance, DuPont) and exposing the plates to X-ray film (Kodak XAR-5) for several
days.

For production of large quantities of the HC-toxin metabolite, unlabeled HC-toxin
was used, and the reaction was scaled up to a volume of 4 mL. The HC-toxin metabolite
was purified by chloroform extraction, flash chromatography (Rasmussen and Scheffer,

1988), and reverse-phase HPLC (Walton et al. , 1982).

48
RESULTS

Production of Radiolabeled H C-toxin

Several radiolabeled amino acids present in native HC-toxin were evaluated as
precursors for in viva production of radiolabeled HC-toxin. As expected from the fact
that D-alanine is a substrate for HC-toxin synthetase in vitro (Walton and Holden, 1988),
D-alanine was incorporated into HC-toxin in vivo. D-alanine was incorporated 68-fold
more effectively than L-alanine and six-fold more effectively than L-proline (Table I).
Radiolabeled D-proline was not tested because it was not commercially available. The
racemic mixture of tritiated alanine used in the final experiment was not evaluated in the
manner presented in Table I prior to the final labeling experiment.

Although production of HC-toxin can vary from culture to culture, we attempted
to optimize the conditions for in vivo incorporation of alanine into HC-toxin by
evaluating the incorporation of D-[14C]alanine with respect to the time added to culture,
and the duration of incubation in the presence of the radioactivity. In our hands, eight
days of fungal growth prior to media exchange followed by 48 h incubation with
radiolabeled D-[“C]alanine gave the best balance between specific activity and toxin yield
(data not shown). Following this protocol, 20 mCi of D,L-[2,3-3H]alanine were added
to a single flask, and approximately 2 mg of pure, tritiated HC-toxin were recovered.
The chromatographic behavior of tritiated HC-toxin was identical to that of unlabeled
toxin during purification. Its biological activity in the root growth bioassay was the same

as that of unlabeled HC-toxin (ED,0 of 0.3 rig/mL). The specific activity of the tritiated

HC-toxin was 70.1 i 0.2 mCi/mmol.

 

49

 

Table 2.1. Incorporation of radiolabeled amino acid precursors into HC—toxin

in vivo. (by J .D. Walton)

The indicated amino acids were added to C. carbonum race 1 after 10 (1 growth, and the
culture filtrates were harvested after an additional 2 d.

 

i ’iR vr friin

Crude Chloroform Final

 

Amino Acid Amount Culture Phase” Recovery"
Added Filtrate'
(#Ci)
D-[I‘C]alanine 15 10.5 4.4 2.7
L-[3H]alanine 40 5 1 .0 0.06 0.04
L-[3H]proline 40 22.0 0.7 0.45

 

‘% of total radioactivity present in crude culture filtrate.

"% of total radioactivity extracted from culture filtrate with chloroform.
° % of total radioactivity remaining after chloroform was evaporated and
sample was redissolved in water.

 

50

Alteration of Toxin Structure Within Maize Leaf Segments

Native HC-toxin was altered following its uptake by transpiration into resistant
leaf segments (Fig. 2.1). An apparent metabolite (Peak 3) eluted with a polarity
intermediate to native HC-toxin (Peak 1) and its 9,10-diol form (Peak 2) (Fig. 2.1B).
The diol of HC-toxin is formed by hydrolysis of the epoxide of Aeo, and was present as
a minor contaminant in all chromatograms, including controls (Fig. 2.1A). Beyond a
minor increase in the amount of diol produced during extraction, no metabolites of HC-
toxin apart from 3 were detected. In addition, 3 was the only altered form of HC-toxin
observed when leaf extractions were performed in aqueous solvents (not shown), or when
leaf extracts were analyzed by TLC. Feeding partially purified 3 (0.05 uCi/mL) back
to maize leaves did not result in the formation of any additional tritium-containing
compounds (data not shown), suggesting that 3 is an end product of HC-toxin

metabolism.

Metabolite Identification

Figure 2.2 shows TLC analysis of native HC-toxin, NaBlL-reduced toxin, and
the HC-toxin metabolite recovered from resistant leaf segments (Peak 3). The HC-toxin
metabolite (lane 3) had the same Rf as NaBH4-reduced HC-toxin (lane 2). The metabolite
reacted with the epoxide indicator p—NBP (lane 7), indicating that the epoxide was still
intact. The FAB-mass spectra of chemically reduced HC-toxin and purified toxin
metabolite had molecular ions of m/e = 439, consistent with the addition of two atomic
mass units to HC-toxin (Fig. 2.3). The NMR spectrum, shown in Figure 2.4, indicates

the creation of a new proton signal vicinal to epoxide carbon-9 of the Aeo side chain.

51

This proton’s chemical shift, multiplicity, and influence on spin-spin coupling and
chemical shift for carbons 9 and 10 are in agreement with the data for the R-isomer of
carbon-8 as described by Kim et al. (1987). Rasmussen described a similar spectrum
and gave the name HC-toxin IV to this derivative (Rasmussen, 1987; Rasmussen and
Scheffer, 1988). Based on these results, we conclude that the metabolite formed in
maize leaves is HC-toxin in which the 8-carbonyl group of the Aeo side chain has been
reduced to the 8-alcohol. We have confirmed the results of Kim et al. (1987) that

showed this form of HC-toxin to be non-toxic.

52

Figure 2.1. Formation of a metabolite of HC-toxin by green maize
leaf segments. Following uptake of 0.25 uCi [3H]HC-toxin by
transpiration, the leaves were ground and analyzed by reverse-phase
HPLC. Radioactivity (shown) eluting from the HPLC column was
detected with an in-line scintillation counter. (A) Control: [3H]HC-toxin
added to leaf segments immediately prior to extraction, (B) Extract of
resistant leaves. Peak 1: native HC-toxin; Peak 2: 9,10-diol-HC-toxin;
Peak 3: HC-toxin metabolite.

 

 

 

 

 

 

 

 

 

 

 

 

 

0
4
10
L 3
\ 2ft0
2
am
“M I
o ..onan2. _ oo
5 at In
3
5 0
4
LI0
3
2 I
wro
2
#m
W _-
0 _ _ - 00
5. Anton. Eng fa

Minutes

Minutes

Figure 2.1.

54

Figure 2.2. TLC analysis of HC-toxin metabolites produced by green
leaf segments and by cell-free extracts of resistant maize. Native
[3H]HC-toxin (l), NaBI-L-reduced [3H]HC-toxin (2), Partially-purified
toxin metabolite recovered from maize leaves (3). Metabolites produced
by 10 min (4) and 20 min (5) incubation of cell-free extracts with [‘H]HC-
toxin, and extract boiled before incubation (6). Lanes 1 through 6 were
detected by fluorography. Metabolites produced in cell-free extracts
detected with the epoxide indicator p-NBP (7). Both native HC-toxin (R,
0.55) and the toxin metabolite (Rf 0.35) react with the epoxide indicator.

55

1-2 3 4 5 6 7
155-.

—-o.35- i

 

, a" -O.23"‘
, -_ ' ..
.e of. ‘ . '—o.0—

 

Figure 2.2.

56

Figure 2.3. Fast-atom bombardment mass spectra of (A) the
metabolite of HC-toxin produced by maize cell-free extracts and (B)
HC-toxin reduced with NaBH,. Samples were dissolved in a matrix of
glycerol and HCl. Both compounds produce molecular ions of m/e 439
(HC-toxin + 2H + H)+ and the following: m/e 369, [(glycerol)4 + H]+;
m/e 461, [(glycerol), + H]+; m/e 475, [M + HC1]+; m/e 531, [M +
glycerol]+.

 

57

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

201
R .
. l A
l
a 4
t 15"
i 1
v 4
e
4 i
D
U 1
n
d
a s
n l
C J
e 1' l ’
t l . l .
C lillrlrlririlii i iirlrvlrlrlrfiliilllrldrlriliiijrrlrlrirlrillrl.r.l.lriri|lliurr.r.ll iiinrln+hhlrirll il.r.l.lr lil “litigant 1.31.1. .1.1.r.l.....r.lrlilrl.n..v. “Yum
350 400 450 5:0 550
M/Z
100“ 4 ‘ 9
R .
+
1 4
a 80‘
t 1
i
v
e 60“
A
b .
U 40-i
n l
d J
" l
n 20‘ l
C
e , l ' l
G 1111 v..l llliil. v 11.;1 ...r .. . ”lit. ' .rlrnrl. lr .illlrl... . .“ii . iilr irrT..r.Jiiiljir v 1 iii iiii'rr .1 Y J 1 ll il V
350 400 450 500 550

M/Z

Figure 2.3.

58

Figure 2.4. NMR analysis of the HC-toxin metabolite‘. Shown in A
is the relevant region of a 500 MHz proton NMR spectrum for the HC-
toxin metabolite purified from a crude extract of resistant maize. The
region contains signals for carbons 8-10 of the Aeo side chain. Relative
to native HC-toxin, the metabolite contains an additional signal at 3.43
ppm. This signal influences the splitting pattern of G9 (the signal
changes from a sharp doublet-of-doublets to a multiplet due to spin-spin
coupling from both C-8 and C-10), and infleunces the chemical shifts for
both C-9 and 010 (am for C-9 is lowered from 3.38 to 2.96; 5pm, for C-
10 lowered from 2.96 to 2.59 (10a), and 2.83 to 2.72 (10b)). A selective
pulse at carbon-9 (3.0 ppm) decouples the signal at 3.43, indicating that
the signal is attributable to a proton at carbon-8 (shown in panel B).
Expected decoupling effects also occur to the protons of carbon-10. The
designation i indicates a detectable impurity. Selective decoupling of the
impurity signal has no effect on any of the HC-toxin resonances (not
shown).

 

1The data in Figure 2.4 were not shown in Meeley and Walton (1991, Plant Physiol. 97 ,
1080-1086), but are included here as additional proof.

 

 

59

A metabolite

PRO

PRO
PRO

HC -toxin 9

 

 

 

 

 

 

Figure 2.4.

 

 

 

HC

ICE

EX

at

p

H C-toxin Metabolism in vitro

In the presence of NADPH, extracts prepared from the etiolated shoots of
resistant maize catalyzed the same metabolic conversion of HC-toxin as intact leaves
(Fig. 2.2). The amount of reduced HC-toxin recovered from incubations with cell-free
extracts increased with time (Fig. 2.2, lanes 4 and 5), while activity was completely
abolished by boiling the extract for 10 min prior to incubation (lane 6).

This enzymatic activity, which we call HC-toxin reductase (HCTR), was partially
purified from etiolated resistant shoots. Ammonium sulfate fractionation and anion
exchange HPLC resulted in a five-fold enrichment of reductase activity with an 18%
recovery. TLC and fluorography of the products formed from [’H]HC-toxin by
individual fractions from an anion exchange separation (Fig. 2.5 , bottom) is shown below
the UV trace (Fig. 2.5, top). HCTR was eluted from the anion exchange column in a
single fraction, No. 10; 8-hydroxy HC-toxin is indicated by (a). A second product of
greater polarity (labeled b) was formed by Fraction 11. Product b reacted with p-NBP,
indicating an intact epoxide, but its formation was partially resistant to boiling and did
not require a hydride donor (data not shown). Its formation was never observed in crude
HCTR preparations (see Fig. 2.2). When we purged Fraction 11 with nitrogen or

included oxygen-scavenging compounds such as GSH, ascorbate, or DTT, in the reaction
mixture, the formation of product b was reduced substantially but no formation of 8-
hydroxy HC-toxin was observed (not shown). We conclude that product b is a form of
HC-toxin produced by a side reaction, perhaps oxygen-dependent, that occurs in solution

in Fraction 11. Importantly, this reaction is unrelated to HCTR activity.

61

Figure 2.5. Anion exchange fractionation of HCTR activity. The top
panel represents the elution of protein (A280). Each fraction was assayed
for HCTR activity, and the resulting products were separated by TLC and
detected by fluorography (bottom panel). (a) the 8-hydroxy derivative of
HC-toxin. (b) unknown compound, formed as an artifact of the ion-
exchange process (see text).

 

Figure 2.5.

Ab‘eorbance (280nm)

62

 

 

 

 

 

 

 

 

2.0 1.0
1.0-

v-—-h—h_L_L¥J l 1 1 l l .
°'°0 2 4 e e 10 12 14 1e 18 20 2:200

Fraction Number
2 4 6 91011121314 151617181920212223
eeee» coo—0.00.0000

(W) [ISBN]

 

 

63
Characterization of HCTR Activity

Table II shows the effects of various treatments and co-factors on HCTR activity.
Activity of partially-purified HCTR was completely abolished by boiling or by pre-
treatment with proteinase K. N ADPH was a better co-substrate than N ADH. A ten-fold
excess of NADP“ over NADPH inhibited HCTR activity by approximately 30% , and a
ten-fold excess of NAD+ had no effect. Approximately 35 % of the HCTR activity was
lost by simply incubating the partially-purified preparation for 60 min at 30°C.
Incubation for 60 min in the presence of Zn2+ or Fe“, and to a lesser extent Cu“,
further inhibited (or destabilized) HCTR activity (Table II). Two known inhibitors of
carbonyl reductases, p-HMB and disulfiram (Fischer et al., 1988; Iwata et al., 1990),
inhibited HCTR by 50% and 70%, respectively (Table II).

On gel-filtration HPLC, HCTR was eluted as a single, symmetrical peak with an

M, of 42,000 (data not shown).

H C-toxin Metabolism by Resistant and Susceptible Leaf Segments

When green leaf segments of equal weight were evacuated of intercellular air and
allowed to transpire water containing [3H]HC-toxin, significant production of 8-hydroxy-
HC-toxin occurred in both resistant and susceptible leaves (Fig. 2.6). The toxin
concentration used (0.3 rig/mL) in these experiments was equivalent to the ED,o for HC-
toxin in the root growth bioassay. Under these conditions, HC-toxin uptake and
metabolism occurred at the same rate in both resistant and susceptible leaves over a 9-h
transpiration period (Fig. 2.6). A time course extended over 48 h showed a similar lack

of host-selective detoxification (data not shown).

 

Table 2.2. Characterization of HCTR activity.

Concentration of NADPH was 4 mM unless otherwise indicated. All ions were 2 mM
sulfate salts. The concentration of [’H]HC-toxin in each case was 23 pM and assays
were run for 15 min at 30°C.

 

HCTR Activity

 

Treatment (% of Control)
Experiment 1
+ NADPH 100‘
boiled (10 min) 0
- NADPH 0
+ proteinase K° 0
+ NADH (4 mM) 31
+ NADPH, FAD (0.1 mM) 97
+ NADPH, FMN (0.1 mM) 71
+ NADPH (1 mM), NADP+ (10 mM) 70
+ NADPH (1 mM), NAD“ (10 mM) 112
+ NADPH + disulfiram (10 ptM) 52
+ NADPH + p-HMB (10 uM) 30
Experiment 2
Preincubation for 60 min with:
No divalent cations 100°
Fe2+ 0
Mg2+ 95
Zn2+ 0
Mn2+ 91
Co2+ 77
Cu2+ 42

 

‘100% activity = 659 pmol/min/mg protein.
I’100% activity = 427 pmol/min/mg protein.
c80 ug/mL, 10 min, 30°C.

 

65

Figure 2.6. Time course of [3H]HC-toxin metabolism by maize leaf
segments during uptake by transpiration. Results are the average of
duplicate samples from three independent experiments. Error bars
represent 1: 1 SD.

 

 

‘73 Conversion to Epoxy Alcohol (AHE)

Figure 2.6.

80

 

7o—
50-
50—
404
30—
20——

10-

 

 

 

 

 

H = Resistant (Pr—1)
e—e — Susceptible (Pr) __
fi/ __
.q
l I I I I I I
l 3 4 5 6 7 8 10

TIME (hours)

67
DISCUSSION

Previous work from this laboratory has described the purification and
characterization of two enzymes involved in HC-toxin biosynthesis (Walton, 1987 ;
Walton and Holden, 1988). One of these enzymes, HTS-2, activates both D- and L-
alanine for incorporation into HC-toxin. Perhaps because D-alanine, unlike L-alanine,
is not diverted into cellular primary metabolism, its efficiency of incorporation into HC-
toxin was relatively high (Table I) . Once favorable conditions for radiolabel
incorporation and toxin yield were determined, alanine incorporation was exploited to
make tritiated HC-toxin in vivo. The chromatographic behavior and biological activity
of HC-toxin were not affected by the incorporation of tritiated alanine into the peptide
ring. The radiolabeled HC-toxin produced by this method was satisfactory for studying
its fate in maize tissues.

The 8-hydroxy derivative of HC-toxin was the only metabolite recovered from
resistant maize leaves following uptake of [3H]HC-toxin by transpiration (Fig. 2.1).
Importantly, 8-hydroxy-HC-toxin is biologically inactive, and was the only toxin
metabolite recovered from any the tissues tested, including cut leaves, cell-free
preparations from etiolated shoots, excised roots (not shown), and whole leaves (not
shown). Notable is the observation that the other critical functional group, the epoxide
of Aeo, was not altered in planta or in vitro.

Reduction of the 8-keto group of HC-toxin is an enzymatic process. An enzyme,
referred to as HCTR, that catalyzes this reduction was partially purified from extracts

prepared from etiolated resistant shoot tissue. HCTR appears to be similar to other

68
NADPH-dependent carbonyl reductases found in plants and animals; it is soluble, uses

NADPH more effectively than NADH as a hydride donor, and has an M, in the range
of 32,000 to 45,000. Characterized NADPH-dependent carbonyl reductases have
biosynthetic functions in anthocyanidin production in plants (Fischer et al. , 1988; Welle
and Grisebach, 1988) and the interconversion of steroids in mammalian tissues (Iwata et
al. , 1990; Wolfe et al. , 1989). Carbonyl reductases often contain metal ions for catalytic
stability and are highly specific for their substrates. In the case of partially-purified
HCTR, metal ions did not contribute to enzyme stability, and certain divalent cations
inhibited activity (Table 11). Whether HCTR has an endogenous substrate is not known
at this time.

With the knowledge that HC-toxin is enzymatically metabolized to a single
inactive compound in resistant maize tissues, we have sought to determine if this
phenomenon is related to the hm locus of maize that governs host-selectivity of C.
carbonum. Our first comparative experiments are summarized by Fig. 2.5, which
indicated that both resistant and susceptible maize were capable of reducing and thereby
inactivating HC-toxin when toxin was delivered to green excised leaves through the
transpiration stream. These data suggest that reduction of HC-toxin is not the basis of
host-selectivity and hence resistance to race 1 of C. carbonum. However, we feel it
necessary to caution that delivery of toxin via transpiration is an artificial technique
devised to deliver toxin to leaf tissues as quickly as possible.

In maize, host-selective reaction to C. carbonum is expressed in etiolated shoots
(Heim et al. , 1983). Therefore, to address whether kinetic aspects of toxin metabolism

are related to host-selectivity, we have initiated a comparative study on HCTR from

69

etiolated resistant and susceptible maize. Interestingly, these experiments have failed
because repeated attempts to detect HCTR activity in extracts from etiolated susceptible
maize have been unsuccessful (unpublished results). This result is in apparent conflict
with the results of our transpiration experiments. Until we resolve this paradox, we can
not base our conclusions solely on data from transpiration experiments. It is possible that
HC-toxin encounters a reductase during transpiration that is different from HCTR.
Therefore, a relationship between HCTR and the hm locus remains an open question.
However, whether or not a relationship between HCTR and host-selectivity exists, the
discovery that maize tissues contain an enzyme capable of inactivating HC-toxin
establishes a novel facet of the interaction between C. carbonum and maize. Because
toxin production is so critical to the infection process, a completely characterized
enzymatic mechanism affecting the biological activity of HC-toxin may in time be

considered as an inteng part of this host/pathogen interaction.

ACKNOWLEDGEMENTS

The authors wish to thank Kermit Johnson of the MSU-Max T. Rogers NMR Facility for
training and expertise, David Wagner of the MSU-Mass Spectrometry Facility for FAB-
MS analyses, and Dr. Jack Rasmussen and Dr. Greg Dilworth (Division of Energy
Biosciences, Department of Energy) for advice on flash chromatography and oxygen-

dependent artifacts, respectively.

70
REFERENCES

Ciuffetti, L.M., Pope, M.R., Dunkle, L.D., Dal)’, J.M., and Knoche, H.W.
(1983). Isolation and su'ucture of an inactive product derived from the host-

specific toxin produced by Helminthosporium carbonum. Biochemistry 22,
3507-3510.

Closse, A., and Huguenin, R. (1974). Isolierung und strukturaufldarung von
chlamydocin. Helv. Chim. Acta. 57, 533-545.

Fischer, D., Stich, K., Britsch, L., and Grisebach, H. (1988). Purification and
characterization of (+)dihydroflavonol (3-hydroxyflavanone) 4-reductase from
flowers of Dahlia variabilis. Arch. Biochem. Biophys. 264, 40-47.

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

Heim, D., Nicholson, R.lL., Pascbolati, S.F., Hagerman, A.E., and Billet, W.
(1983). Etiolated maize mesocotyls: A tool for investigating disease
interactions. Phytopathology 73, 424-428.

Iwata, N., Inazu, N ., Takeo, S., and Satoh, T. (1990). Carbonyl reductases from rat
testis and vas deferens. Eur. J. Biochem. 193, 75-81.

Kawai, M., and Rich, D.H. (1983). Total synthesis of the cyclic tetrapeptide, HC-
toxin. Tet. Lett. 24, 5309-5312.

Kawai, M., Rich, D.H., and Walton, J .D. (1983). The structure and conformation of
HC-toxin. Biochem. Biophys. Res. Commun. 111, 398-403.

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

Lamoreaux, G.L., and Rusness, B.G. (1986). Xenobiotic conjugation in higher
plants. In GD Paulson, J Caldwell, DH Hutson, JJ Menn, eds, Xenobiotic
Conjugation Chemistry, ACS Symp Series, Vol 299. American Chemical
Society, Washington DC, pp 62-105.

Leisch, J .M., Sweeley, C.C., Staffeld, G.D., Anderson, M.S., Weber, DJ., and
Scheffer, R.P. (1982). Structure of HC-toxin, a cyclic tetrapeptide from
Helminthosporium carbonum. Tetrahedron 38, 45-48.

71

Nelson, O.E., and Ullstrup, A.J. (1964). Resistance to leaf spot in maize; genetic
control of resistance to race 1 of Helminthosporium carbonum U11. J. Hered. 55,
195-199.

Pope, M.R., Ciufetti, L.M., Knoche, H.W., McCrery, D., Daly, J.M., and
Dunkle, L.D. (1983). Structure of the host-specific toxin produced by
Helminthosporium carbonum. Biochemistry 22, 3502-3506.

Rasmussen, J .B. (1987). Host-selective toxins from Helminthosporium carbonum:
purification, chemistry, biological activities, and effect on chlorOphyll synthesis
in maize. Ph.D. thesis, Michigan State University, E. Lansing.

Rasmussen, J .B., and Scheffer, R.P. (1988). Isolation and biological activities of
four selective toxins from Helminthosporium carbonum. Plant Physiol. 86,
187-191.

Scheffer, R.P. (1976). Host-specific toxins in relation to pathogenesis and disease
resistance. Encycl. Plant Physiol. 4, 247-269.

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

Shimabukuro, R.H., Frear, D.S., Swanson, H.R., and Walsh, W.C. (1971).
Glutathione conjugation: an enzymatic basis for atrazine resistance in corn.
Plant Physiol. 47, 10-14.

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., and Earle, E.D. (1983). The epoxide in HC-toxin is required for
activity against susceptible maize. Physiol. Plant. Pathol. 22, 371-376.

Walton, J .D., and Holden, ER. (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., 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.

Welle, R., and Grisebach, H. (1988). Isolation of a novel NADPH-dependent
reductase which coacts with chalcone synthase in the biosynthesis of 6’-
deoxychalcone. FEBS Lett. 236, 221-225.

72

Wolfe, L.S., Rostworowski, K., Pellerin, L., and Sherwin, A. (1989). Metabolism
of prostaglandin D2 by human cerebal cortex into 90:, 118-prostaglandin F2 by an
active NADPH—dependent ll-ketoreductase. J. Neurochem. 53, 64-70.

CHAPTER 3*

A BIOCHEMICAL PHENOTYPE FOR A
DISEASE RESISTANCE GENE OF MAIZE

* originally published in The Plant Cell:
‘0 1992, American Society of Plant Physiologists

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.

73

74
ABSTRACT

In maize, major resistance to the pathogenic fungus Cochliobolus (Helminthosporium)
carbonum race 1 is determined by the dominant allele of the nuclear locus hm. The
interaction between C. carbonum race 1 and maize is mediated by a pathogen-produced,
low-molecular weight compound called HC-toxin. We recently described an enzyme
from maize, called HC-toxin reductase, that inactivates HC-toxin by pyridine nucleotide-
dependent reduction of an essential carbonyl group. We now report that this enzyme
activity is detectable only in extracts of maize that are resistant to C. carbonum race 1
(genotype Hm/Hm or Hm/hm). In several genetic analyses, in vitro HC-toxin reductase
activity was without exception associated with resistance to C. carbonum race 1. The
results indicate that detoxification of HC-toxin is the biochemical basis of Hm-specific

resistance of maize to infection by C. carbonum race 1.

75
INTRODUCTION

Resistance of plants to pathogens is frequently inherited in a simple Mendelian
fashion, and typically, resistance is dominant to susceptibility. Although specific
resistance has been intensively studied at the genetic level, in no case is the biochemical
process controlled by a specific resistance gene known. The fungal pathogen
Cochliobolus carbonum (Helminthosporium carbonum) race 1 causes leaf spot and ear
mold of maize. The dominant allele of the Mendelian gene Hm gives specific resistance
to C. carbonum race 1 at all stages of growth (Nelson and Ullstrup, 1964). A number
of studies have concluded that specific pathogenicity of Cochliobolus carbonum is
mediated by a low-molecular-weight compound called HC-toxin, and that Hm governs
resistance to C. carbonum by way of insensitivity to HC-toxin (Yoder, 1980). For
example, root growth of susceptible maize (genotype hm/hm) is inhibited by HC-toxin
whereas root growth of resistant maize (genotype Hm/Hm or Hm/hm) and of other plants
is not inhibited except at much higher concentrations (Scheffer and Ullstrup, 1965). In
the fungus, HC-toxin production genetically co-segregates with specific pathogenicity on
hm/hm maize (Scheffer and Nelson, 1967).

HC-toxin is a cyclic tetrapeptide of structure cyclo(D-prolyl-L-alanyl-D-alanyl-L-
Aeo), where Aeo stands for 2-amino-9,10-epoxy-8-oxodecanoic acid (Gross et al. , 1982;
Walton et al. , 1982; Kawai et al. , 1983). The epoxide and the 8-carbonyl groups of Aeo
are required for biological activity of HC-toxin (Ciuffetti et al. , 1983; Walton and Earle,
1983; Kim et al. , 1987). During studies on the biochemistry of HC-toxin synthesis

(Walton, 1987 ; Walton and Holden, 1988), it was found that D-alanine is recognized and

 

 

76

activated by one of the biosynthetic enzymes, HTS-2, in vitro, and that radiolabeled D-
alanine is incorporated into HC-toxin in vivo (Walton, 1991). We have exploited this
fact to prepare [’H]-HC-toxin (Meeley and Walton, 1991).

When fed to green maize leaves through the transpiration stream, [3H]-HC-toxin
is metabolized to a single, nontoxic derivative in which the 8-carbonyl group of Aeo is
reduced to the corresponding alcohol (Meeley and Walton, 1991). However, leaves of
both resistant (genotype Hm/hm) and susceptible (genotype hm/hm) maize convert HC-
toxin equally well. This originally suggested that detoxification by reduction is not the
basis of resistance controlled by Hm (Meeley and Walton, 1991).

Cell-free extracts of etiolated maize seedlings metabolize HC-toxin to the same
compound, 8-hydroxy-HC-toxin, as whole green leaves. The responsible enzyme, called
HC-toxin reductase (HCTR), uses NADPH as cosubstrate and has been partially purified
and characterized from resistant (Hm/hm) maize (Meeley and Walton, 1991). We sought
to compare the kinetics of HCTR from susceptible and resistant plants, but as shown in
this paper, HCTR activity is detectable only in extracts of resistant maize, genetically
segregates with the Hm allele, and is severely reduced in transposon-induced mutations
of Hm. In addition, differential detoxification of HC-toxin occurs in intact tissues when
[3H]-HC-toxin is delivered by means other than the transpiration stream. The results
argue that HCTR is the critical biochemical function controlled by Hm and is responsible

for specific resistance of maize to C. carbonum race 1.

77
MATERIALS AND METHODS

In Vitro HCTR Assay

Tritiated HC-toxin (specific activity 70 mCi/mmol) was prepared as described
(Meeley and Walton, 1991). Etiolated maize seedlings were grown as described (Meeley
and Walton, 1991). To assay HCTR, single plumules of equal fresh weight (150 mg)
were harvested and ground with a pestle in 200 uL extraction buffer [0.1 M Mops, pH
7.4, 0.3 M sucrose, 5% (v/v) polyvinylpyrrolidone, 10% (v/v) glycerol), 5 mM
dithiothreitol, 1 mM EDTA, 15 mM ascorbate, and 0.2 mM phenylmethylsulfonate] in
1.5 mL microfuge tubes. NADPH (final concentration 4 mM) and 0.25 uCi [3H]-HC-
toxin were added directly to the ground tissue and the mixture incubated for 30 min at
30°C. The reaction mixtures were extracted three times with chloroform, the chloroform
evaporated under reduced pressure, and the residue analyzed by thin-layer
chromatography (TLC) on silica gel 60 (Meeley and Walton, 1991). Radioactive
compounds on the TLC plates were detected either with an automated B-detector
(Bioscan) (Figures 3.1, 3.5, and 3.6) or by fiuorography (Figures 3.2, 3.3, and 3.4) as

described (Meeley and Walton, 1991).

Pathogenicity Tests

Maize seedlings were grown in darkness as described (Meeley and Walton, 1991).
After the plumule was removed for the HCTR assay, the mesocotyl, still attached to the
caryopsis, was wounded slightly with a razor blade and inoculated with approximately
100 conidia of C. carbonum race 1 in one uL of 0.1% Tween-20. The inoculated

seedlings were then returned to darkness for 3 to 6 days before being photographed.

78

Identification of Revertants

Independent ears (families) were harvested from selfed heterozygotes of mutant
hm-1040, containing either Spm or dSpm at the Hm locus. Batches of 50 kernels from
each car were grown in the dark and the plumules excised, ground in extraction buffer
as described above, and frozen at -80°C. The mesocotyls were inoculated with C.
carbonum race 1, and placed in the dark until disease symptoms developed. When
candidates for germinal revertants were identified by resistance to the fungus, the
appropriate extracts were thawed and assayed for HCTR activity as described above.

Extracts from susceptible progeny were chosen at random as negative controls.

In Vivo Metabolism Assays

For the in vivo green leaf disc assay (Figure 3.5), plants were grown in a growth
chamber as described (Meeley and Walton, 1991). Fourth true leaves were abraded on
both sides with carborundum using a watercolor brush and l-cm discs cut with a cork
borer. The discs were floated lower side down in 24-well microtiter plates in 200 uL of
5 mM CaClz, 5 mM KCl, 20 mM sucrose, 0.1% Tween-20, and 2 ug/mL [’H]-HC-toxin
(143,000 dpm). The plates were incubated 24 h under room lights. To calculate uptake
of [3H]-HC-toxin, the leaf discs were removed, rinsed, frozen, and extracted with
methanol. Radioactivity in the methanol extract was measured by scintillation counting.
To measure metabolism of [3H]-HC-toxin, the liquid remaining in the wells was extracted
three times with CHCl3 and the extracts analyzed by TLC as described above.

For the in vivo plumule assay (Figure 3.6), etiolated plants were grown as

described (Meeley and Walton, 1991). Five day old seedlings were excised just above

 

 

the

SR

79
the coleoptilar node and injected with a 22G hypodermic needle through the base into the

space between the coleoptile and leaves with [’H]-HC-toxin (concentration 200 ug/mL)
in a volume of 10 uL. The plumules were incubated horizontally on moist filter paper
in covered petri plates in the dark for 3 or 6 hr. They were then rinsed with distilled
water, ground in Eppendorf tubes in 0.5 mL distilled water, centrifuged to remove
cellular debris, and the supematants extracted three times with CHC13 and analyzed by

TLC as described above.

RESULTS

As shown in Figure 3.1, HCTR is detectable in several maize lines of genotype
Hm/Hm or Hm/hm (inbred Prl and hybrids K61 x Prl and Prl x K61) but not in maize
lines of genotype hm/hm (inbreds Pr, K61, Mo21a, and K44, and hybrid Pr x K61). To
test genetic cosegregation of Hm with in vitro HCTR activity, a method was developed
to simultaneously assay HCTR activity and disease reaction in single seedlings. In the
testcross Hm/hm x hm/hm, resistant and susceptible plants segregate 1: 1. In Figure 3.2,
a representative sample of progeny from such a cross is shown. From a total of 50
progeny, only the resistant plants (29 of 50) had HCTR activity, converting from 40%
to 70% of HC-toxin to the 8-hydroxy derivative. The susceptible plants (21 of 50)
lacked detectable enzyme activity. No recombination between the resistant disease

phenotype and presence of HCTR activity was observed. Statistical analysis of these data

80

Figure 3.1. HC-toxin reductase (HCTR) activity in extracts of etiolated
shoots of maize lines susceptible or resistant to C. carbonum race 1.
Extracts of single plumules (leaves plus coleoptile) were assayed for in
vitro HCTR activity and the reaction products separated by TLC and
detected with a Bioscan B-detector. HCTR activity is evidenced by
conversion of native HC-toxin (R, 0.5) to 8-hydroxy-HC-toxin (R, 0.3).
Inbred lines are indicated by ®. Prl, Prl x K61, and K61 x Prl are
resistant; Pr, K61, Mo21a, K44, and Pr x K61 are susceptible. Inbreds
Pr and Prl are near-isogenic, differing only at the hm locus (Scheffer and
Ullstrup, 1965).

81

Pr1® \A-MJ’ W

I . ._ a - ..
Pr; ® K/wvaww—j ,__._, _ ___,.__~___W,.,

l
K61 ® Why—Nap!)

 

 

 

M0218 ® NM)

K44 ® \NAW f 300
Pr x K61 mama/J j 200

l’ r
K61 x Prl ”NI-we" - 100

PH x K61 Nov-"VJ” L o
\ \
Or Sf

Figure 3.1.

 

3H (cpm)

82

Figure 3.2. Cosegregation of HCTR activity and disease reaction to
C. carbonum race 1. Shown is a representative sample of progeny from
a cross between maize of genotype Hm/hm and genotype hm/hm. Each
plant was assayed for in vitro HCTR activity and disease reaction.
(Top) Metabolites of [3H]-HC-toxin were separated by TLC and detected
by fluorography. The upper spot is native HC-toxin and the lower spot,
when present, is 8-hydroxy-HC-toxin.

(Bottom) In a resistant reaction (lanes 4, 5, 7, and 9-13) the fungus does
not spread from the site of inoculation and the mesocotyl remains white.
In a susceptible reaction (lanes 1-3, 6, 8, 14, and 15) the fungus colonizes
the tissue and the mesocotyl becomes water-soaked and necrotic.

 

 

83

 

Figure 3.2.

84

places the locus which is responsible for HCTR activity within 3 cM of Hm, at a 95%
confidence level (Allard, 1956).

These results were extended by examining specific mutants of Hm. Several
independent susceptible mutants, due to insertions at Hm by members of the Mutator
(Mu) and Spm/En families of transposable elements (Gierl et al. , 1989; Walbot, 1991),
have been identified by cosegregation with flanking markers that detect restriction-
fragment-length polymorphisms (G.S. Johal and SP. Briggs, unpublished, Maize
Genetics Newsletter 64, 36-37). A complete description of the isolation and structure
of these alleles will be published elsewhere. These mutants were tested for both HCTR
activity and reaction to the pathogen, as shown in Figure 3.3. Two mutants due to
insertion of Mu elements (hm-1369 and hm-656), one Spm/dSpm mutant (hm-1040), and
one mutant due to insertion of an uncharacterized element (hm-1062) had little or no
HCTR activity and were susceptible. The low levels of HCTR activity seen in hm-1040
and hm-656 may be the result of positional effects and/ or somatic reversions (Gierl et al. ,
1989; Walbot, 1991), but these levels of HCTR activity appear insufficient to confer
resistance to the fungus (Figure 3.3, bottom). Clearly, transposon-induced mutations of
Hm have striking effects on the levels of HCTR activity detectable in etiolated seedlings.

Germinal excision of a transposable element can restore the wild-type phenotype
(Gierl et al. , 1989). A screen to detect reversion events in a population containing the
hm-1040 mutant allele was initiated by inoculating the progeny of selfed hm-1040
(genotype Hm::Spm/hm) with the fungus. In a population of 300, two seedlings were

identified that had reverted from susceptibility to resistance. Figure 3.4 shows that

85

Figure 3.3. HCTR activity and disease phenotype in extracts of
transposon-induced mutants of Hm.

(Top) HCTR assays of two plumules are shown for each independent
transposon-induced mutant of Hm (hm-1369, hm-656, hm-1062, and hm-
1040), and resistant (K61 x Prl; Hm/hm) and susceptible (Pr x K61;
hm/hm) controls.

(Bottom) The disease phenotype of each mutant as compared to resistant
and susceptible controls.

 

 

86

I...”

.
E:\ E: O
Geo—1::
woo—1:5,,
omOL—E

OOMTE:

.N .

 

Figure 3 .3.

87

Figure 3.4. HCTR activity and disease phenotype of revertants due
to germinal excision of Spm (or dSpm) from hm-1040.

(Top) HCTR activity is detectable in extracts from two resistant plants
(lanes 3 and 6) identified as revertants from a susceptible hm-1040
background. The level of HCTR activity in these progeny is comparable
to that of a resistant (Hm/hm) control (lane 1). Three sibling progeny
(lanes 2,4,5) are included as representatives of the mutant HCTR
phenotype.

(Bottom) The disease phenotypes of the resistant (Hm/hm) control (lane
1), and two resistant revertants (lanes 3 and 6) as compared to susceptible
mutant siblings (lanes 2,4,5).

 

 

 

88

4O

3O

2.

 

Figure 3 .4.

89

extracts of these revertants contained HCTR activity comparable to that of a resistant
control, but HCTR was not detected in susceptible siblings.

There remains the paradox of why HC-toxin delivered through the transpiration
stream to green leaves of susceptible plants is metabolized, whereas HC-toxin incubated
with extracts of etiolated susceptible shoots is not. Several attempts to extract HCTR
from green leaves of either susceptible or resistant plants were unsuccessful. To deliver
[3I—I]-HC-toxin to green leaf tissues by means other than the transpiration stream, discs
were cut from green leaves and floated on a solution of [3H]-HC-toxin for 24 hr. Since
the recovery of [3H]-HC-toxin from the leaf discs was too low (approximately 10%) to
assay directly, we assayed HC-toxin conversion in the solution remaining in the wells.
Figure 3.5 shows that susceptible (Pr x K61) leaf discs did not produce significant levels
of 8-hydroxy-HC-toxin above background, whereas resistant (K61 x Prl) leaf discs
produced 8-hydroxy-HC-toxin at a minimum of 2 to 3 times the background signal
(Figure 3.5).

Figure 3.6 shows that intact plumules from etiolated plants metabolized HC-toxin
in a host-selective manner. After 3 hr, resistant plumules had metabolized over 50% of
total [3H]-HC-toxin (lane 1) while susceptible plumules failed to produce a detectable
amount of the toxin metabolite (lane 2). After 6 hr, etiolated resistant shoots had
metabolized nearly all of the supplied HC-toxin (lanes 3 and 4). In one case, etiolated
susceptible shoots afer 6 hr showed some metabolism of HC-toxin to a compound with

the same R, as 8-hydroxy-HC-toxin (Figure 3.6, iane 5).

Figure 3.5. Host-selective metabolism of HC-toxin by green leaf discs.
Green leaf discs were floated on a solution containing [3H]-HC-toxin for
24 hr, after which the discs were analyzed for radioactivity (as a measure
of uptake of 3H-HC-toxin), and the remaining solutions were analyzed for
metabolism. Lanes 1-3, genotype Hm/hm (resistant); lanes 4-6, genotype
hm/hm (susceptible). The numbers above the traces indicate integration
of the radioactivity in the peak of 8-hydroxy-HC-toxin, centered at R,
0.25, normalized to the same area in lane 6. Each lane represents one
leaf disc. Uptake for the three discs (given as the mean + 1 SD, n = 3,
with percent uptake in parentheses) from susceptible plants was 12,200 +
1660 dpm (8.5%) and from the resistant plants was 18,700 + 1570 dpm
(13. 1 %).

 

 

Figure 3.5.

91

\_
1 WM \.,~,_L ~Mw
0 fl

3.
/ I
2 W,/A\MMJ (”WWW

2.8

/ ‘1‘.
3 MIWM LA“ ,,,,, g,”
r

1.1 f/ \,
4 W’JWNVV’ \/W,-" \._.~—-—-I\. . _‘,.w_’~__

IA
5 __A’_~_A/\/~f\/-Wxx,/\.J~J/ A"_v___ —~w~u__..~

A'qx
1.0 ~/\/\/ \
l -

6 -Wv¢l’\’\ A—-,_Af~¢v~/"\M~¢a,r. ’\.-- ‘
0 0.25 0.5 ()3

Br

‘~. ‘RNVVVmw—A.1

92

Figure 3.6. Host-selective metabolism of HC-toxin by intact plumules.
Etiolated mdlings with unbroken coleoptiles were excised above the
coleoptilar node and injected with [3H]-HC-toxin. The plumules were
then incubated for 3 or 6 hr, after which they were ground, extracted with
chloroform, and analyzed by TLC. Lane 1, genotype Hm/hm, 3 hr; lane
2, genotype hm/hm, 3 hr; lanes 3 and 4, genotype Hm/hm, 6 hr; lanes 5
and 6, genotype hm/hm, 6 hr. Each lane represents one plumule.

 

 

93

1M

 

 

 

 

Figure 3.6.

94
DISCUSSION

The cyclic tetrapeptide HC-toxin has the central role in the interaction between
C. carbonum race 1 and maize. The results in this paper argue that the Hm gene confers
resistance to C. carbonum race 1 by controlling the presence of HCTR, an enzyme that
detoxifies PIC-toxin (Meeley and Walton, 1991).

The results with green leaf discs (Figure 3.5) and intact etiolated shoots (Figure
3.6) demonstrate that differential metabolism of HC-toxin due to Hm occurs not just in
extracts but also in intact green and etiolated tissues. The differential metabolism seen
in vitro (Figures 3.1-3.4) therefore is not due to differential extractability or differential
in vitro stability of HCTR. The in vitro studies show that HCTR activity in resistant
plants is constitutive, and that the lack of comparable toxin metabolism in susceptible
intact tissues (Figures 3.5 and 3.6) is not a secondary effect of HC-toxin treatment.

Our results raise three interesting questions. First, is Hm the structural gene for
HCTR or does Hm regulate the expression of HCTR? Work is in progress to answer this
question through the cloning of Hm and the purification of HCTR.

Second, what is the nature of the apparent HCTR activity seen when HC-toxin
is delivered through the transpiration stream to green leaves of susceptible plants (Meeley
and Walton, 1991) and at low levels in intact etiolated shoots (Figure 3.6, lane 5)?
Clearly, under at least some conditions, maize of genotype hm/hm has some HCTR-like
activity. One possibility is an unrelated enzyme with capacity to metabolize HC-toxin,
for example, one of the various N ADPH-dependent reductases involved in cellular

metabolism. The transpiration/ green leaf results could also be due to an HCTR-like

95

activity restricted to the vascular tissue. Dinucleotide-dependent reductases have been
found in xylem exudates (Biles et al. , 1989). A xylem-localized reductase could
metabolize HC-toxin in the transpiration stream but would be irrelevant during infection
by C. carbonum race 1, which does not, at least initially, infect the vascular tissue.
Third, does HCTR have a function in the plant in the absence of C. carbonum
race 1? hm has no known phenotype other than reaction to C. carbonum race 1. We
have detected HCTR activity in etiolated shoots of barley, wheat, oats, and sorghum, all
potential hosts of C. carbonum, but not in Arabidopsis, peas, or cucumber (R.B. Meeley
and J .D. Walton, unpublished results). Conceivably, HC-toxin or the five related cyclic
peptides made by other fungi (Walton et al. , 1985; Walton, 1990, Itazaki et al. , 1990)
have exerted a significant selective pressure on the evolution of the Poaceae. Relevant
to this question is whether the recessive alleles of hm are nulls or whether they result in
the production of HCTR at lower levels or with altered enzymatic properties. For
example, if HCTR is the product of the Hm gene, hm could encode an enzyme with
normal affinity for its putative endogenous substrate but reduced or no affinity for HC-

toxin.

ACKNOWLEDGMENTS
We thank Kurt Stepnitz for photographic services. Work in J.D.W.’s laboratory was
supported by the United States Department of Energy, the National Science Foundation,
a Michigan State All-University Research Initiation Grant, and the Michigan State
University Research Excellence Fund. S.P.B. ’s laboratory was supported by a grant

from Pioneer Hi-Bred International.

 

 

All

Bl

96
REFERENCES

Allard, R.W. (1956). Formulas and tables to facilitate the calculation of recombination
values in heredity. Hilgardia 24, 235-278.

Biles, C.L., Martyn, R.B., and Wilson, E.D. (1989). Isozymes and general proteins
from various watermelon cultivars and tissue types. HortScience 24, 810—812.

Briggs, S.P. (1987). Molecular tagging of a toxin-resistance gene. Curr. Topics Plant
Biochem. Physiol. 6, 69-67.

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

Gierl, A., Saedler, H., and Peterson, P.A. (1989). Maize transposable elements. Annu.
Rev. Genet. 23, 71-85.

Gross, M.L., McCrery, D., Crow, F., Tomer, K.B., Pope, M.R., Ciuffeti, L.M.,
Knoche, H.W., Daly, J.M., and Dunkle, L.D. (1982). The structure of the
toxin from Helminthosporiwn carbonum. Tetra. Lett. 23, 5381-5384.

Itazaki, H., Nagashima, K., Sugita, K., Yoshida, H., Kawamura, Y., Yasuda, Y.,
Matsumoto, K., Isbii, 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. Antibiotics 43, 1524-1532.

Kawai, M., Rich, D.H., and Walton, J .D. (1983). The structure and conformation of
HC-toxin. Biochem. Biophys. Res. Comm. 111, 398-403.

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

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

host-selective cyclic peptide from Cochliobolus carbonum. Plant Physiology, 97 ,
1080-1086.

Nelson, O.E., and Ullstrup, A.J. (1964). Resistance to leaf spot in maize. J. Heredity
55, 195-199.

Nelson, R.R., and Ullstrup, A.J. (1961). The inheritance of pathogenicity in
Cochliobolus carbonum. Phytopathology 51, 1-2.

97

Scheffer, R.P., Nelson, R.R., and Ullstrup AJ. (1967). Inheritance of toxin
production and pathogenicity in Cochliobolus carbonum and Cochliobolus
victoriae. Phytopathology 57, 1288-1291

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

Walbot, V. (1991). The mutator transposable element family of maize. In Genetic
Engineering, Vol. 13, J .K. Setlow, ed (New York: Plenum), pp. 1-37.

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 Biochemistry
of Peptide Antibiotics, H. Kleinkauf and H. von Dohren, eds (Berlin: deGruyter),
pp. 179-203.

Walton, J .D. (1991). Genetics and biochemistry of toxin synthesis in Cochliobolus
(Helminthosporium). In Molecular Industrial Mycology, S. Leong and RM.
Berka, eds (New York: M. Dekker), pp. 225-249.

Walton, J .D., and Earle, E.D. (1983). The epoxide in HC-toxin is required for activity
against susceptible maize. Physiol. Plant Pathol. 22, 371-376.

Walton, J .D., and Holden, ER. (1988). Properties of two enzymes involved in the
biosynthesis of the fungal pathogenicity factor HC-toxin. Mol. Plant-Microbe
Inter. 1, 128-134.

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

Walton, J.D., Earle, E.D., Stiihelin, H., Grieder, A., Hirota, A., and Suzuki, A.
(1985). Reciprocal biological activities of the cyclic tetrapeptides chlamydocin and
HC-toxin. Experientia 41, 1370-1374.

Yoder, O.C. (1980). Toxins in pathogenesis. Annu. Rev. Phytopathol. 18, 103-129.

CHAPTER 4

STUDIES ON THE PURIFICATION AND INDUCTION
OF HC-TOXIN REDUCTASE,
AND ITS DISTRIBUTION AMONG PLANT SPECIES

98

99
ABSTRACT

The maize enzyme HC-toxin reductase represents a known biochemical function
controlled by a plant disease resistance gene. The resistance gene Hm of maize confers
full resistance to the fungal pathogen Cochliobolus carbonum Nelson race 1.
Biochemical/ genetic evidence indicates that the Hm gene of maize either encodes or
controls the expression of HC-toxin reductase, an enzyme responsible for inactivation of
HC-toxin, the cyclic peptide required for pathogenicity of C. carbonum race 1. This
report demonstrates further purification of HC-toxin reductase beyond initial enrichment
procedures. Three candidate polypeptides from 40-50 kD are visualized by SDS-PAGE
when active HPLC fractions from a number of chromatographic steps are compared.
HC-toxin reductase activity is inducible. Significant increases in enzyme activity occur
in response to inoculation of resistant maize seedlings with race 1 of C. carbonum.
Enzyme activities analogous to HC-toxin reductase from maize are found in other plant
species. These activities are detected in extracts from other monocotyledonous crops,
but not in dicot extracts. Other evidence suggests that HC-toxin metabolism is associated
with the C. carbonum resistance allele Hm2. HC-toxin reductase activity is induced
following pathogen inoculation, appears specific for monocot species, and possibly shares
the same detoxification mechanism with an independent C. carbonum resistance gene in
maize. Collectively, these observations argue that the Hm gene may have a dedicated

function in maize for resistance to HC-toxin or similar natural products.

100
INTRODUCTION

The cyclic tetrapeptide HC-toxin has the central role in the interaction between
maize and the phytopathogenic fungus Cochliobolus carbonum Nelson race 1. Production
of HC-toxin is the principal requirement for pathogenicity of C. carbonum race 1 against
maize (Panaccione et a1. , 1992). In the plant, the resistance gene Hm confers race-
specific resistance to the fungus as well as decreased sensitivity to HC-toxin (Kuo and
Scheffer, 1970). A biochemical phenotype for the Hm gene has been established by
comparing HC-toxin metabolism in resistant (Hm/-) and susceptible (hm/hm) maize. The
Hm allele is thought to either encode or control the expression of an enzyme called HC-
toxin reductase (HCTR), which inactivates the toxin by carbonyl reduction. Susceptible
maize (hm/hm) are more sensitive to HC-toxin presumably because they are deficient in
this capacity (Meeley et al. , 1992).

That Hm is the structural gene for HCTR is the simplest interpretation of the
genetic data. To investigate this hypothesis, purification of the enzyme is sought in order
to obtain amino acid sequence data. Such data could be compared with deduced amino
acid sequences generated by others working to clone the Hm gene. HPLC procedures
presented in this paper detail HCTR purification beyond the initial enrichment procedures
reported previously (Meeley and Walton, 1991).

An important issue to be addressed is whether the gene for HCTR is maintained
in the maize genome specifically for defense against C. carbonum race 1. In this paper,
HCTR activity is examined in response to fungal inoculation, and its distribution among

other plants is tested.

101

Resistance to C. carbonum is functionally duplicated in the maize genome (Nelson
and Ullstrup, 1964). The Hm gene for C. carbonum race 1 resistance is epistatic to
Hm2, a gene that confers developmentally limited, and incompletely dominant resistance
to C. carbonum race 1 (Nelson and Ullstrup, 1964). Experiments examining HC-toxin
metabolism associated with the Hm2 allele are presented. The discussion of these results
pertains to the origin and specificity of HC-toxin reductase and to questions about the

process of co-evolution between a pathogen and its host.

MATERIALS AND METHODS

Plant Materials and Extraction Procedures

The growth, harvest, and initial extraction procedures for HCTR purification were
as described (Meeley and Walton, 1991). Etiolated seedling tissue (100 g) was
homogenized in extraction buffer, and HCTR was enriched by an ammonium sulfate
fractionation (30-55 %). Enzyme preparations were made from the resistant (Hm/-) maize
hybrid GL-582 (Great Lakes Hybrids Inc. , Ovid, MI). For fungal induction
experiments, seedlings of the resistant homozygote Prl (Hm/Hm) were used. For the
distribution survey, extracts were prepared from etiolated seedlings of Avena sativa (oat),
Hordeum vulgare (barley), Glycine max (soybean), Sorghum bicolor, Triticum aestivum
(wheat), Pisum sativum (pea), Cucumis sativus (cucumber), and Arabidopsis thaliana.
Tissues (150 mg) were extracted and assayed for HCTR activity as described (Meeley

and Walton, 1991).

102
Chromatography Columns and Bufi’ers

HPLC separations were performed on a Waters system equipped with two Model
501 pumps, a Model 440 absorbance detector (280 nm), and an automated fraction
collector (Pharmacia Frac- 100) . Columns used for anion exchange chromatography
included a Beckman TSK-DEAE-S-PW (10 am, 7 .5 x 75 mm), or a Rainin Dynamax-
Hydropore-S-AX (5 pm, 10 x 100 mm), in buffer A, eluted with buffer B. Hydrophobic
interaction chromatography was performed with a BioRad Bio-Gel TSK-phenyl-S-PW
column (10 pm, 7.5 x 75 mm), in buffer C, eluted with buffer D. Gel filtration
chromatography was performed on a tandem coupling of two Beckman Ultraspherogel
SEC 3000 (5 mm, 7 .5 x 300 mm), in buffer B. Chromatofocusing was performed on a
Pharmacia FPLC system with Pharmacia Mono-P HR-S/20 column (10 pm, 5 x 200
mm), in buffer F, eluted with buffer G.

Buffers: (A) 25 mM bis-Tris-HCl (pH 7.0 or pH 6.25), 10% glycerol, 2.5 mM
DTT. (B) same as A plus 1M NaCl. (C) 1M (NI-I4)2804, 25 mM bis-Tris-HCl (pH 7.0),
10% glycerol, 2.5 mM DTT. (D) 10% glycerol, 2.5 mM DTT. (E) 100 mM bis-Tris-
HCl (pH 7.0), 250 mM NazSO4, 10% glycerol, 2.5 mM DTT. (F) 25 mM bis-Tris (pH
7.1), 10% glycerol, 2.5 mM DTT. (G) Polybuffer 74 (pH 4.0, Pharmacia), 10%
glycerol, 2.5 mM DTT. Between steps, buffer exchange was performed on samples with
a Sephadex G-25 column (Pharmacia PD-lO). To all HPLC fractions, ascorbate (to 10
mM) was added for protection against dissolved oxygen. HCTR assays were 30 min

long.

103
Gel Electrophoresis

Aliquots of HCTR-containing HPLC fractions were analyzed by SDS-PAGE on
10% acrylamide gels. Silver staining of gels was performed using a Bio-Rad silver

staining kit according to the manufacturer’s directions.

Fungal Induction Experiments

Etiolated resistant (Prl , Hm/Hm) seedlings were inoculated with a heavy spore
suspension (approximately 50 conidia/pL) of C. carbonum race 1 in 0.1% Tween-20.
Cotton swabbing was used to ”paint" seedlings with the spore suspension. Control
mungs were treated with 0.1% Tween-20. Seedlings were placed back in the dark for
24 h prior to harvest. Crude extracts of seedling tissues (5 g) were prepared and assayed
for HCTR activity as described (Meeley and Walton, 1991). Product formation was
quantitated by TLC as described (Meeley and Walton, 1991). As starting material for
HCTR purification, large batches of etiolated seedlings were induced with C. carbonum

spores to maximize HCTR levels prior to extraction.

H C-toxin Metabolism and Hm2

Etiolated seedling materials containing the Hm2 resistance allele (Hm2/Hm2) were
assayed for disease phenotype and HC-toxin metabolism as described (Meeley et al. ,
1992). Susceptible (hm2/hm2, and hm/hm) and resistant (Hm/-) seedlings were analyzed

as controls. Assay times were 30 min.

104
RESULTS

HPLC Fractionation of HCTR Activity

Figure 4.1 illustrates enzyme assay data typical of all purification procedures.
This example shows fractionation of HCTR activity by anion exchange (pH 7.0),
followed by hydrophobic interaction chromatography (pH 7.0). The bulk of HCTR
activity eluted in fractions 16 and 17 during anion exchange. No artifactual product
(reported previously in Meeley and Walton, 1991) was formed during anion exchange
when precautions were taken to minimize oxygen in solutions. Fractions 16 and 17 were
combined and exchanged into high-salt buffer for hydrophobic interaction HPLC. HCTR
activity was eluted in fractions 14-16 from the hydrophobic interaction column. Without
an effective means to stabilize HCTR in vitro, the key to success in these procedures is
to run the columns as quickly as possible, but even when fraction HI-15 from
hydrophobic interaction was immediately taken on to another column, HCTR activity was
not recovered (data not shown).

Figure 4.2 shows the enrichment obtained from a series of column procedures.
Etiolated tissue (70 g, C. carbonum-induced) was extracted, taken through ammonium
sulfate fractionation, and a primary round of anion exchange (pH 7.0). Following
additional separations, HCTR-containing fractions were analyzed by SDS-PAGE. The
active fractions from hydrophobic interaction (lane 1), gel filtration (lanes 2 and 3), and
anion exchange (pH 6.25, lanes 4 and 5) are shown. The latter two steps produced an
enrichment for three polypeptides of 49, 43.5, and 40 kD. A previous estimate by gel

filtration determined HCTR to be Mr 42,000 (Meeley and Walton, 1991).

105

Figure 4.1. Fractionation of HCTR activity. Each lane represents the
assay of individual HPLC fractions from anion exchange, pH 7.0 (IEX
fractions 15-18) and hydrophobic interaction, pH 7.0 (HI fractions 11-20).
Tritiated HC-toxin was used as substrate; HCTR activity is denoted by
conversion of labeled HC-toxin (5.2 cm) to its 8-hydroxy derivative (3.5
cm).

 

106

 

 

 

 

 

 

_ CPM

 

L
x.
H

l
w
r}

Ex I?

‘ 4-

(«€13
33:.“
W777???
*l p l

. ,{r

. [r W
a l

 

'K-C'JAr—‘H 'flvla | IW‘MP 0

Figure 4.1.

107

Figure 4.2. SDS-PAGE analysis of HCTR-containing fractions. In this
example, the fractionation involved preparative anion exchange (Dynarnax,
pH 7 .0), hydrophobic interaction (pH 7 .0) (lane 1), gel filtration (pH 7 .0)
(lanes 2 and 3), and another round of anion exchange at pH 6.25 (lanes
4 and 5). Molecular weight standards are indicated by the dots at the left,
and correspond to 97 , 66, 45, 31, and 21.5 kD. These procedures enrich
for three polypeptides with molecular weights of 49, 43.5, and 40 kD, as
indicated by the dots.

 

 

108

 

 

Figure 4.2.

109

To date, the development of an HCTR purification protocol has involved trial
separations on a number of HPLC columns. Matrices found to retain or resolve HCTR
include anion exchange, hydrophobic interaction, gel filtration, and Chromatofocusing.
To identify a polypeptide by SDS-PAGE that corresponds to HCTR, the order of these
columns was varied. HCTR-active fractions were then compared by SDS-PAGE. An
example of this analysis is shown in Figure 4.3. By examining HCTR-containing
fractions from a number of purification attempts, certain polypeptides can be eliminated
from consideration. For example, two smaller bands that comprise a major portion of
the protein in lanes 5 and 6 are not present in lanes 1-4. HCTR thus does not appear to
be a dimer between these two polypeptides. Likewise, many of the larger polypeptides
(seen in lanes 2 and 3) are not present in other HCTR preparations. A definitive
designation for the HCTR polypeptide can not be made from this analysis, but one of the
strongest candidates (indicated by arrows) corresponds to the 49 kD protein from Fig.

4.2.

Fungal Induction of HCTR

As shown in Figure 4.4, a twenty-four hour treatment of etiolated resistant
swdlings with a spore suspension of C. carbonum race 1 results in an approximate 2.5-
fold induction of HCTR specific activity. Comstock and Scheffer (1973) have shown that
growth of C. carbonum race 1 is arrested during the first 24-48 h post-inoculation on
resistant (Hm/-) maize tissues. Induction of HCTR activity within this time period is

consistent with the enzyme’s role in disease resistance.

110

Figure 4.3. Comparative SDS-PAGE analysis of HCTR-containing
fractions. The column methods from a number of purification attempts
were as follows:
Lane: (1) Dynamax anion exchange (pH 7.0), gel-filtration (pH 7.0)
(2) Dynamax anion exchange (pH 7 .0), gel-filtration (pH 7 .0)
(3) Dynamax anion exchange (pH 7.0), hydrophobic interaction (pH 7.0)
(4) DEAE anion exchange (pH 7.5), Mono-P Chromatofocusing
(5) Dynamax anion exchange (pH 7.0), hydrophobic interaction (pH 7.0),
gel filtration (pH 7.0)
(6) Dynamax anion exchange (pH 6.25) of HCTR from lane 5.

 

 

111

 

 

Figure 4.3.

112

Figure 4.4. Induction of HCTR activity in response to fungal
inoculation. Pathogenic fungal spores (C. carbonum race 1) were painted
onto etiolated resistant (inbred Prl) seedlings. Seedlings were harvested
and extracted for HCTR activity 24 h post-inoculation. Tween-20-treated
seedlings served as the non-inoculated control. The specific activity of
HCTR was determined in crude extracts prepared from 5 g of inoculated
and control tissues.

 

 

 

113

 

 

 

 

 

 

 

 

 

 

 

m L. MU .zu to. m m. o
MAEBWQ @UEEOWBEQM asmg mm:

Seedling Treotment

Figure 4.4.

 

 

Distributio

Fig
number 0
seedlings
WMMJI
8-hydrox
other m<

wxwa

HC-toxi

Md sus
to Hm;
to C. .
Veryv
0bS€p
both
metal
hm2/.

8~hyc

114
Distribution of H C—toxin Metabolism

Figure 4.5 summarizes the results of a test for HC—toxin metabolism among a
number of different plant species. In these experiments, equal weights of etiolated
seedlings were extracted and assayed (30 min) for HCTR activity. Extracts from oat,
wheat, and barley converted more than 50% of the total [3H]-HC-toxin (0.25 uCi) to its
8-hydroxy derivative. Sorghum extract was also quite active, though less so than the
other monocot species. An interesting result was that HC-toxin metabolism was not

observed in any of the dicot extracts.

H C—toxin metabolism and Hm2

The locus hm2 (located on chromosome 9-L) is an alternate source of resistance
and susceptibility to C. carbonum race 1 (Nelson and Ullstrup, 1964). Hm is epistatic
to Hm2, thus the effects of Hm2 can be seen only in an hm/hm background. Resistance
to C. carbonum race 1 conferred by Hm2 is developmentally limited, and is expressed
very weakly in the seedling stage (Nelson and Ullstrup, 1964). Evidence of this can be
observed in Figure 4.6, which shows an essentially susceptible disease phenotype for
both Hm2/Hm2 and hm2/hm2 seedlings. However, a small amount of an HC-toxin
metabolite is formed in extracts of Hm2/Hm2 seedlings in comparison with either
hm2/hm2 or hm/hm susceptible controls. This metabolite has a polarity identical to the

8-hydroxy derivative of HC-toxin which is formed by Hm/- resistant maize.

115

Figure 4.5. Distribution of HCTR-like activity among several
different plant species. Etiolated seedling tissue (150 mg) from each
species was extracted and assayed for HCTR activity. The entire extract,
including insoluble material, was included in the assay. Results are
expressed simply as the percentage of HC-toxin converted. The top
panels compare TLC analysis of extracts from oat and cucumber.

 

116

 

 

 

 

 

 

 

 

 

 

. . . . 00 0.10 0.20 0.30 0.40
250:00 010 030 0430 050 3a: 4 . A g
. Oat ngcumber
2000 250%
L
2N4-
150+-
E » 150+-
O
1.3.. o
I lut-
I
1 r
5“" V 5% L
L >
0 Z») L4 . 1-- 0 -A 1 5‘ ._ 1

ca 0.0 270 4f0 5:0 830 ca 0.0 230 430 5.0 030

 

 

“ml Distance (cm)

30::

s 2::

:23: m % from converted
cucumber n.d.*
soybean n.d.
barley 75.4
sorghum 17.2
wheat 81.6
oat 67.2
pea n.d
Ara bidopsis n.d.

 

*n.d. - none detected

Figure 4.5.

 

117

Figure 4.6. HC-toxin metabolism and Hm2. These assays were
performed as described (Meeley et al. , 1992). Resistance to C. carbonum
race 1 by Hm2 is expressed poorly at the seedling stage (Nelson and
Ullstrup, 1964). The disease phenotype (bottom) is indicative of this, but
a low level of HCTR activity is evident in comparison with hm2/hm2
plants (top).

118

£55.. .”
EfE... ”
O

NEI\NEI

 

 

«£5 «E;

 

 

Figure 4.6.

l 19
DISCUSSION

A purified form of HC-toxin reductase is important for study of its substrate
specificity, and for other useful purposes such as antibody production. The data
presented in Figures 4.2 and 4.3 represent the best results obtained to date for
purification of HCTR. The purification procedures presented here enrich for three
polypeptides of 49, 43.5 , and 40 kD. From earlier gel filtration data (Meeley and
Walton, 1991), HCTR has an estimated molecular weight of 42 kD. Despite this
success, a definite polypeptide assignment can not be made at this time. The
chromatography steps presented here must be further developed.

HC-toxin reductase activity is quite unstable following extraction. Several
methods meant to stabilize in vitro HCTR activity (some described previously in Meeley
and Walton, 1991) have met with very little success. For example, storage at 4‘C or
-20'C appear to have no advantage over storage at -80'C. Storage with NADPH has
been recommended for other N ADPH-dependent oxidoreductases (De-Eknamkul and
Zenk, 1992), but this does not appear to improve HCTR’s stability. To date, the most
effective precaution is to prevent oxygen contamination of buffers and samples.
Ascorbate is added routinely to extraction buffer and to HPLC fractions immediately
after collection. HPLC buffers are degassed thoroughly prior to use. Still, the enzyme
has a half life of only 24-48 h under routine conditions.

HCTR activity is induced by fungal inoculation in a manner temporally consistent
with the resistance response of maize against C. carbonum race 1. Yet it remains

important to know how other fungi, pathogens and non-pathogens of maize alike, affect

 

120
HCTR activity. In particular, race 2 isolates of the fungus and perhaps HC-toxin itself

should be examined for any effect on HCTR activity.

In a survey of several monocot and dicot species, the evidence suggests that
HCTR, or at least an analogous activity, is present in monocots, but not in dicots.
Importantly, results from molecular genetic analyses of Hm are consistent with both
HCTR induction and its specificity to monocots. Johal and Briggs (pers. commun.),
have described transcriptional activation of the Hm gene within 24 h (post-inoculation)
of resistant leaf tissue with C. carbonum race 1. They also report that DNA sequences
in the monocots sorghum and Coix cross-hybridize with a probe specific for the maize
Hm gene. No hybridization with Arabidopsis DNA was observed (J ohal and Briggs,
1992).

Although HCTR activity is the critical biochemical factor involved in resistance
to C. carbonum, a separate issue is whether the gene for HCTR is intended solely for
defense against this (or other) pathogen(s). HCTR operates so slowly with HC-toxin as
a substrate that an effective spectrophotometric assay based on NADPH oxidation has
never been developed. In addition, barriers to enzyme purification have precluded a
meaningful analysis of HCTR’s substrate specificity. With these difficulties in mind, it
is possible that HC-toxin represents a fortuitous substrate for an enzyme with some other
function in the cell. The enzyme activity, "HC-toxin reductase" might be an anomaly,
but HCTR activity is induced by fungal inoculation, and appears exclusive to monocots;
attributes that place limits on the type of oxidoreductase that could fortuitously
metabolize HC-toxin. HC-toxin is certainly not a promiscuous substrate; a number of

commercially available oxidoreductase enzymes were unable to reduce HC-toxin (data

a n“:

 

121

not shown), and there is no evidence of HCTR-like activity in a number of non-plant
systems. For example, enzyme extracts from yeast, and even from C. carbonum race
1 fail to catabolize HC-toxin (Meeley, unpublished observations). In mammalian blood,
HC-toxin appears to be inactivated, but by a mechanism very different from the reduction
described in maize (Meeley, unpublished data). These details deserve attention because
they indicate that there is something rather exclusive about HC-toxin metabolism in maize
and other monocots.

Consider the mechanism of toxin inactivation. Reduction of the 8-keto moiety of
Aeo is perchance the most sophisticated route to toxin inactivation. If HC-toxin were
a fortuitous substrate, and no selection pressure to evolve resistance to it ever existed,
it is reasonable to expect that toxin metabolism would involve a more non-specific
mechanism, such as inactivation of the epoxide group of Aeo. Enzymes like glutathione-
S-transferases, that are able to conjugate epoxides, occur in maize (Shimabukuro et al. ,
1971). There are also a number of other non-specific enzymatic mechanisms for
xenobiotic detoxification in plants (Lamoreaux and Rusness, 1986).

Evidence in favor of HCTR as a dedicated function includes the fact that the
hm/hm genotype has no known phenotype other than susceptibility to C. carbonum race
1. The attribute of fungal induction reinforces a defense role for HCTR. The attribute
of monocot specificity raises the prospect that HCTR’s distribution among plants is
biased because of selective pressures among monocots to evolve resistance to this type

of compound.

122

Given that other fungi have been shown to produce Aeo-containing cyclic peptides
(see Walton, 1990, or Figure 1.5), ancestral selection pressure to evolve resistance to
these compounds may have been provided by a pathogen other than C. carbonum, and/or
in a monocotyledonous host other than maize. Thus, if C. carbonum race 1 acquired the
genes necessary for HC-toxin biosynthesis relatively recently, as has been proposed
(Panaccione et al. , 1992), susceptible hosts may represent recent inbreeds that carry
homozygous mutation(s) in the gene for HCTR (Hm). The natural history of Helmintho-
sporium leaf spot is at least consistent with this interpretation, since only a few genetic
sources of the hm/hm genotype exist. DNA sequence analysis of the hm allele(s) is in
progress to search for evidence of mutation (S.P. Briggs, pers. comm.). Evidence
already suggests that the hm alleles are transcriptionally defective (Johal and Briggs,
1992). Therefore, this allele does not appear to produce an enzyme with a function
unrelated to HC-toxin metabolism (Johal and Briggs, 1992).

How is metabolism by Hm2 relevant to these discussions? Maize contains two
independent sources of heritable resistance to C. carbonum race 1. The Hm locus is the
focus of this dissertation because it is the major source of resistance, being epistatic to
Hm2 (of chromosome 9L). Hm2-mediated resistance to C. carbonum race 1 exhibits
incomplete dominance and elements of developmental regulation. This paper presents
some preliminary evidence that Hm2-mediated resistance to C. carbonum race 1 involves
a mechanism of HC-toxin inactivation similar to that controlled by Hm. Since Hm2
expression increases with plant age (Nelson and Ullstrup, 1964), a reliable method to
assay HCTR activity in mature tissues is required. Attempts to extract and detect HCTR

activity in the post-anthesis tissue of corn silks have been unsuccessful (data not shown).

123

HC-toxin metabolism influenced by Hm2 may be relevant to some of our earlier
findings. In previous data showing low levels of HCTR activity in transposon-induced
mutants of Hm (Meeley et al. , 1992, or Figure 3.3), our interpretation was that these low
levels were due to positional effects of the inserted elements, or somatic excision of the
element. However, most maize lines that are naturally resistant to C. carbonum race 1
due to Hml- also have the genotype Hm2/- (Nelson and Ullstrup, 1964). It is possible
that Hm mutant lines contained low-levels of HCTR activity because they were present
in an Hm2/- background (G.S. Johal, pers. comm.). In turn, this may support a
mechanism of HC-toxin metabolism controlled by Hm2. Genetic manipulations to
resolve Hm mutant alleles from Hm2 are in progress (G.S. Johal, pers. comm.).

Further discussion of a common mechanism of toxin metabolism between Hm and
Hm2 considers the evidence for "large scale interchromosomal homology" throughout the
maize genome (Helentjaris et al. , 1988). Even though the origins of modern maize
remain nebulous and controversial, evidence in favor of ancestral duplication, as well as
a number of possible explanations for its occurrence have been forwarded (Helentjaris
et al. , 1988). This may explain how two sources of resistance to C. carbonum race 1
share the same mechanism. These two genes could be of common origin, and perhaps
Hm is epistatic to Hm2 because developmental regulations were either imposed on Hm2,
or relaxed on Hm. If this is the case, heritable resistance to HC-toxin could have its
origins in the ancestry of maize. A biased distribution of HCTR-like activity among the
monocots may support this notion.

As yet, no model system stands out as one amenable to tracing the biochemical

and molecular events associated with host/pathogen co—evolution, but a detailed picture

124

of HCTR’s origin, substrate specificity, and distribution among the plant kingdom is
tenable. In time, studies on the interaction of C. carbonum race 1 and maize may
contribute to our understanding of a particular evolutionary history between a host and

its pathogen.

REFERENCES

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.

De-Eknamkul, W., and Zenk, M.H. (1992). Purification and properties of 1,2-
dehydroreticuline reductase from Papaver somniferum seedlings. Phytochemistry
31 , 813-821.

Helentjaris, T., Weber, D., and Wright, S. (1988) Identification of the genomic
locations of duplicate nucleotide sequences in maize by analysis of restriction
fragment length polymorphisms. Genetics 118, 353-363.

J ohall, G.S., and Briggs, S.P. (1992). Isolation and characterization of a plant disease
resistance gene. Science (in press).

Kuo, M.-S., and Scheffer, R.P. (1970). Comparative specificity of the toxins of
Helminthosporium carbonum and Helminthosporium victoriae. Phytopathology
60, 365-368.

Lamoreaux, G.L., and Rusness, D.G. (1986). Xenobiotic conjugation in higher plants.
In Xenobiotic Conjugation Chemistry (Paulson,G.D. et al., eds.). ACS Symp.
Series 299, 62-105.

Meeley, R.B., and Walton, J .D. (1991). Enzymatic detoxification of HC—toxin, the
host-selective cyclic peptide from Cochliobolus carbonum. Plant Physiology, 97 ,
1080-1086.

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.

 

125

Nelson, O.E., and Ullstrup, A.J. (1964). Resistance to leaf spot in maize; genetic
control of resistance to race 1 of Helminthosporium carbonum U11. J. Heredity

55, 195-199.

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. Nat. Acad. Sci. USA 89, 6590-6594.

Shimabukuro, R.H., Frear, D.S., Swanson, H.R., and Walsh, W.C. (1971).
Glutathione conjugation: an enzymatic basis for atrazine resistance in corn. Plant

Physiol. 47, 10-14.

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

. ~“ "“'I. u—“rq

 

CHAPTER 5

SUMMARY AND RECOMMENDATIONS

126

 

127
SUMMARY

The fungal pathogen Cochliobolus carbonum Nelson race 1 exists in a
complementary genie interaction with its host, maize. At the biochemical level, the
cyclic tetrapeptide HC-toxin has the central role in this interaction. Race 1 isolates of
the fungus are the only isolates pathogenic to hm/hm maize, and they are the only isolates
that produce HC-toxin. Production of HC-toxin is essential for race 1 pathogenicity as
demonstrated by Panaccione et al. (1992). In maize, resistance to C. carbonum race 1
occurs via the Hm gene. This allele confers decreased sensitivity to HC-toxin (Kuo and
Scheffer, 1970) .

This dissertation is aimed at determining the function of the Hm gene of maize
as it has been defined; a gene for resistance to C. carbonum race 1, and decreased
sensitivity to HC-toxin. Associated with the resistance allele Hm is an enzymatic activity
called HC-toxin reductase, that detoxifies HC-toxin by metabolism of a critical functional
group (Meeley and Walton, 1991). Analyses of HCTR activity among a variety of
resistant and susceptible inbreds, segregating resistant and susceptible progeny, and
transposon-induced mutants, and revertants, establish that HCTR is the biochemical
phenotype for the Hm allele (Meeley et al. , 1992). This work, together with the work
of Drs. Guri Johal and Steve Briggs (Pioneer Hi-Bred Intl.), represent the premier
descriptions of the molecular identity and biochemical function of a plant disease
resistance gene.

Purification of HCTR was attempted principally for the purpose of confirming

whether Hm actually encodes HCTR. By SDS-PAGE, three polypeptide candidates were

‘.' "0.31“ 01-... _ ’ -

 

128

observed in HCTR-containing fractions from a number of HPLC steps. A definite
assignment of a polypeptide as HCTR can not be made without further experimentation.

A variety of experimental and historical information raises the question whether
the Hm gene is specifically intended to provide resistance to HC-toxin or similar natural
products. The hm/hm genotype has no phenotype other than susceptibility to C.
carbonum race 1, sensitivity to HC-toxin, and a characteristic lack of HCTR activity.
Mutations in Hm are defective in HCTR, but have no observable pleiotropic effects. The
hm/hm host appears to be transcriptionally defective, implying that no alternative enzyme
product without HC-toxin metabolic capability exists for this allele. The product of Hm
is HCTR. Both Hm transcription and HCTR activity are induced by fungal inoculation
which supports a disease resistance/defense role for this gene and its product. This
mechanism appears specific to monocots and may share a similar mechanism with Hm2.
Together, these points imply the existence of an evolutionary history involving HC-toxin
production by the fungus, and plant defense against it.

To return to the quadratic check that was used to illustrate this Introduction, our
final variation includes what we have learned about the function of the Hm gene. Figure
5.1 conveys the same concept of gene-for-gene complementarity, with new information
pertaining to the function of Hm. Our current model proposes that the Hm gene encodes
the enzyme HC-toxin reductase, a soluble, NAD(P)H-dependent oxidoreductase, that
inactivates HC-toxin by specific reduction of the 8-keto moiety of Aeo. This is the
biochemical definition of pathogen ”recognition" for this particular host-pathogen system.

Resistant maize is generally 100-fold less-sensitive to HC-toxin than susceptible (hm/hm)

129

Figure 5.1. The C. carbonum race 1/maize interaction in biochemical
terms. This quadratic check has been updated with the recent information
gained about the function of the Hm resistance gene. The incompatible
interaction between a race 1 (PIC-toxin“) fungus and a resistant (Hm/-)
plant can be explained by the action of HC-toxin reductase (HCTR). The
compatible interaction results because susceptible (hm/hm) plants do not
produce HCTR to counteract the toxin.

130

 

 

 

Cochliobolus carbonum/maize
Maize Phenotype
HCTR+ HCTR-
HC-toxin+ — +
Fungus
Phenotype
HC-toxin- — —

 

 

 

 

Figure 5.1.

131
maize (Kuo and Scheffer, 1970). This dissertation proposes that this differential can be

explained, at least in part, by the activity of HC-toxin reductase. Since these data do not
concern the mode of action of HC-toxin, differential sensitivity may also reside at the site
of toxin action.

The experimental work presented here tells nearly a complete story about the
function of the Hm gene in disease resistance. One question left unresolved is whether
Hm actually encodes HCTR. Purification of HCTR was intended to address this question
by generating amino acid sequence data that could serve as an independent source of
confirmation for our collaborators working to clone the Hm gene. HCTR appears
reticent to purification to the level necessary for amino acid analysis. Despite this
shortfall, many of the procedures were successful, and substantial progress was made.
Further development of HCTR purification is feasible and highly recommended.

Presently, the best evidence that Hm encodes HCTR is provided by our
collaborators at Pioneer (Johal and Briggs, 1992). Figure 5.2 shows a comparison of the
deduced amino acid sequence data for the Hm gene with that determined for the AI gene
of maize. These two genes are over 34% identical and 56% similar by this analysis.
Importantly, the AI gene encodes the enzyme dihydro-4-flavonol reductase (DFR), an
N ADPH-dependent oxidoreductase involved in anthocyanin biosynthesis (Reddy et al. ,
1987). The similarity between the enzymes, and their corresponding genes, indicates that
Hm, like Al, encodes an oxidoreductase enzyme. The enzyme encoded by Hm is the

resistance gene product, HC-toxin reductase.

"i‘ 2:776:11”

132

Figure 5.2. Comparison of the deduced amino acid sequence of the
Hm gene with the A1 gene of maize. The maize AI gene encodes the
enzyme dihydro-4-flavonol reductase (or dihydroquercetin reductase).
When compared with the deduced amino acid sequence for Hm, the level
of amino acid identity (indicated by |), is 34% , and the level of similarity
is 56% (indicated by .: j). The highest degree of identity occurs in the
amino terminal end of the primary sequence, which contains the consensus
sequences for the dinucleotide fold (Johal and Briggs, 1992). The
dinucleotide fold is required for coordination of the nicotinamide
cosubstrate.

133

A1 3 RGAGASEKGTVLVTGASGFVGSWLVMKLLQAGYTVRATVRDPANVGKTKP 52

. I. Illl.l|.l|lll l||.llll.ll.l. .: .l
°’° ‘l” I Ill"ll°lllll°lll° llll'll‘l’°' ‘I'

Hm 1 MAEKESNGVRVCVTGGAGFIGSWLVRKLLEKGYTVHATLRNTGDEAXAGL 50

53 LMDL. PGATERLSIWKADLAEEGSFHDAIRGCTGVFHVATPMDFLSKDPE 101
I

51 LRRLVPGAAERLRLFQADLFDAATFAPAIAGCQFVFLVATPFGLDSAGSQ 100

102 .NEVIKPTVEGMISIMRACKEAGTVRRIVFTSS..AGTVNLEERQ..... 143
I I.

II
“me 101 YRSTAEAVVDAVRAILRQCEESRTVKRVIHTASVAAASPLLEEEVSASGV 150

 

SEC 144 ..RPVYDEESWTDVDFCRRVKMTGW.MYFVSKTLAEKAALAYAAEH..GL 188
tax) ' ' " ee 00 e 'ee" ' "00 co

7H0 151 GYRDFIDESCWTSLNVDYPLRSAHFDKYILSKLRSEQELLSYNGGESPAF 200

lmW 189 DLYTIIPTL?VGPFISASMPPSLITALALITGNAPHYSILKQV?.....: 232
mi 201 éMipLéilAéIS-réLéiaééiénélélpéénéélcééiifixiéqmsn 250
me 233 await-:3???" %Mi:§¥?°sé“vvrtr<3%0€~mwr¥rmé 28°
ma plaaseesaacnatgaaanmgimammal 3..

281 VPQRFPGIQDDLQPVRFSSKKLQDLGFTSGTRRWRTCSTPPSGLARRRAS 330
II .I III I . I I :I

z O :0:.." OOOO'O:"' .0 IO. .0. O '|

301 ILKE....TEAVATVRPARDRLGELGFQVPSTAWEEILDSSVACAARLGS 346

331 SPSPLPPEGTALPRC 345

347 ....LDASKLGLQKG 357

Figure 5.2.

 

134

The prospect that the HCTR mechanism is stably inherited in maize for specific
defense against pathogens has very exciting evolutionary implications. The DNA for
both fungal pathogenicity (TOX2 for HC-toxin biosynthesis) and host resistance (Hm for
HC-toxin reductase) have been isolated, and perhaps can be used as molecular markers
for studying the co-evolution of a host with its parasite. Studies of the natural history
of Helminthosporium leaf spot, extended maize lineages, monocot evolution, and other
fungi that produce Aeo-containing cyclic peptides present new opportunities for inquiry
into the evolution of this particular host-pathogen interaction.

The prospect that HC-toxin is a fortuitous substrate for an enzyme with another
function has other implications. One can assume, probably with certainty now, that
hm/hm maize lines represent inbreeds with homozygous mutations at Hm. So C.
carbonum race 1 appears to have capitalized on this mutation. A race endowed with HC-
toxin production has created a niche for itself. But how? A credible answer to this

question requires an understanding of HC-toxin’s mode of action.

 

135
RECOMMENDATIONS

The mode of action of HC-toxin is perhaps best addressed by pursuing new
alternatives to obtain a radiolabeled toxin analog. High specific activity is desired for
binding studies with maize cell components. Information about HC-toxin biosynthesis
has already fostered a method for radiolabeling HC-toxin (Meeley and Walton, 1991).
A variation on in viva feeding could be attempted, substituting aminoacrylic acid as an
alanine analog. Aminoacrylic acid is an unsaturated analog of alanine which serves as
the direct precursor to alanine in catalytic tritiation. The HC-toxin synthetase enzyme
HTS-2, which activates both D- and L-alanine could be examined for an ability to
activate aminoacrylic acid. Incorporation of this analog into the peptide ring of HC-toxin
may provide one with an unsaturated substrate that could be purified and catalytically
tritiated to high specific activity. 9

To assist with binding studies, a maize cDNA expression library in a suitable
vector could be probed with radiolabeled HC-toxin to screen for specific sites of toxin

action. Such pursuits are planned for the future (G.S. Johal, pers. commun.).

For studies of HCTR’s localization and substrate specificity, heterologous
expression of the Hm gene may present some new opportunities. A considerable amount
of time and effort was recently expended trying to express an Hm cDNA in a
heterologous system such as yeast or E. coli. 1 received a cDNA clone from our
collaborators at Pioneer. While this project gave me some nwded exposure to molecular

biology, the cDNA supplied, called Hm2l-l , was quite problematic due to the presence

; ‘

g .

"Q

l ‘I

. ’3‘.

I II

 

 

 

 

136

of an unprocessed, 286 bp intron. Several methods were designed to remove this intron,
and are summarized in Figure 5 .3.

Attempts to isolate an intron-free Hm construct in the proper frame in the E. coli
expression plasmids pET-3a and pET-9a (Novagen), or the yeast vector pYE82.0
(Invitrogen) have not been successful. Alternative Hm constructs and expression
systems, such as bacculovirus expression, are being considered for the future (S.P.
Briggs, pers. commun.). Heterologous expression could be a very important alternative
to traditional protein purification methods. HCTR-specific antibodies and an amenable
source of active HCTR are feasible rewards to expect from these pursuits. Meaningful
substrate specificity studies will rely on a purified form of HCTR.

In the short term, the issue of whether HCTR activity is encoded by Hm can be
addressed by in vitro RNA transcription-translation experiments. In vitro transcription-
translation of plasmid encoded cDNA was the method used to link DFR activity with the
maize AI gene (Reddy et al. , 1987). Similar experiments can be performed in
pBluescript-II (Stratagene), which contains the T3 and T7 promoters for transcription.
pHm2l-l is an Hm cDNA clone that contains 5’ and 3’ untranslated regions, a poly A+
tail, and a 286 bp intron (Figure 5.3). The NcoI fragment from pHm21-1 can be
repaired in a vector that contains no XhoI or NotI restriction sites. The vector pET-3d
is suitable to sub—clone the N001 fragment from pHm21-l, and insert the 408 bp, intron-
free, XhoI/Notl fragment from pHM-XN (see Figure 5.3). The repaired Ncol fragment
can be cloned back into pBluescript, creating a full-length cDN A construct for

transcription-translation experiments.

 

137

. €909 Hum 95 mafia? 8m 3385 8.» D 28 U 88.“ 3coEw£
ghoul 2F .c.~mm>q .869 7.8.» 05 e8 3225 an R5 ZRANEE can .on AANEE 85:. .«o
flee—Lama.“ 583 2:. .AB anémem 9588 .5me wE:§=8-=obE ofi com—mo.— 9 com: 83 83 .ZX
.33 Eat “5893 53:35“ .3 wow 2F .9 “ENEE $58.8 .3385 mom 05 .3 82%: 95.253 .8.“
and: :68 9 8mm ==>m 35¢ a 83380 $255 82:. .85sz £55 05 mega: 325.5 mom Bo: 93
.m was 5:5 wean 25w 2: me 933: .m 28 .m 05 .8 seasocsaea .3 3.6an 33 £55 an own 05. Amy
8:85 mom 3:51 983 2: 3:32.12 mama: same 8 «Ema 95 383 5.88% .8 an... :62
BEBE SW 88.8 an ER .98 :08 :e 8% 553 888 Ad . LANE: Eat “Sauna 3mm; x >395 2 ”8
Bcwfioe 225 m use 58m .9889, communes 38» ER :8 .m 2: E wanes new ACE 85-3 .8255
.32 .o .3 828.3 Eugenia 25 8:20 seemed Hogxioom 963 a a sense cause <29 5:
2F $38.8 .3 @985: 8a 85st 05 he cocaine“ do.“ #3285 Enema mom £88m? gene—220:
E «.759 Si 2: he 56835 .8. £55.31: 8:258 ES 3.6.5.— 555 .5. $3522 .n.n unaut—

138

_Im\=oz-_ocx\_1m
33¢” 5 2X- FNEI

«on—.03

con _Im:_:>n_:Im
.nxfiw. S m- FNEI

.269... com
is; .32.: m- .«E:

ESQ—zoom
3300. S F- 381

6). © @

.3 2am

 

 

 

 

 

 

 

m |.
I j
to: .055
I I . h I I
r m _ m m Jr
\ _002 all 302 =a_>l —O£X _.‘z
.IEIn @ 58.0
T m m m m T.
.32 a... to: .2? 2.
.osx _ 2.125
£25.. “:8:

cube.

 

=3>m =33“.

139
REFERENCES

Johal, G.S., and Briggs, S.P. (1992). Isolation and characterization of a plant disease
resistance gene. Science (in press).

Kuo, M-S., and Scheffer, R.P. (1970). Comparative Specificity of the toxins from
Helminthosporiwn carbonum and Helminthosporium victoriae. Phytopathology
60, 365-368.

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

host-selective cyclic peptide from Cochliobon carbonum. Plant Physiology, 97 ,
1080-1086.

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.

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. Nat. Acad. Sci. USA 89, 6590-6594.

Reddy, A.R., Britsch, L., Salamini, F., Saedler, H., and Rohde, W. (1987). The

A1 (anthocyanin-1) locus in Zea mays encodes dihydroquercetin reductase. Plant
Sci. 52, 7-13.

 

"‘l‘lrlllullllll