.nmom.
Liwmuuuamtl.
. fibril". 1331...”;
. 3n z. JINQII. (‘90.

bfl i.

 

 

 

. I. I 10-..!

 

. I.l..r.:u.| - (I; .l. vtifl‘ifllleW In

 

 

, , . . . . ,...4..0..&..‘.a.....v

I. .3... . 1:: 1.: .

   

 

THES'S _lllllllllllllllllllllllllllllllllllllllllllllllllllllHIHIHI

29 01568 7654

,.. LIBRARY

Michigan State
University

 

 

 

This is to certify that the

dissertation entitled

Investigation of the Role of Plant Cell Wall
Degrading Enzymes in Host-Pathogen Interactions

presented by
Jenifer M. Go'rlach
has been accepted towards fulfillment

of the requirements for

Ph. D . degree in Plant Pathology

 

 

1%

Major professor

Date l-ZQf- [0,. I???

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

2 i _ ’_
PLACE IN ner‘uaupox to remove rm. checkout tram your record.
TO AVOID FINES return on or bdore date due.

  

DATE DUE DATE DUE . DATE DUE

ll

 

  

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

l IF—Tl I
L__l

 

 

 

 

 

 

 

 

 

MSU loAn Affirmative ActionlEquol Opportunity Intuition
Warns-oi

 

 

INVESTIGATION OF THE ROLE OF PLANT CELL WALL DEGRADING
ENZYMES IN HOST-PATHOGEN INTERACTIONS

By

Jenifer M. Gerlach

A DISSERTATION

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

DOCTOR OF PHILOSOPHY

Department of Botany and Plant Pathology

1 997

ABSTRACT

INVESTIGATION OF THE ROLE OF PLANT CELL WALL DEGRADING
ENZYMES IN HOST-PATHOGEN INTERACTIONS

By

Jenifer M. Gorlach

This research focused on determining whether cell wall degrading ezymes secreted by
the fungal pathogen, Cochliobolus carbonum, had an essential role in the colonization of
maize tissue. Three secreted proteases (Alpla, Alplb, Alp2) were purified and
characterizedand the corresponding genes (ALPI, ALPZ) were cloned and sequenced.
Biochemical features of Alpla and Alplb and analysis of a mutant deficient in the
production of ALP] indicated that Alpla and Alplb were difl‘erentially glycosylated forms
of the same gene product. The ability of alpI , a1p2, and alpI/alpz mutants to cause
disease on maize is discussed.

Similarly, three 1,3-1,4-B-glucanases (Mlgla, Mlglb, Mlg2) were purified and
characterized, the genes encoding these proteins (M61, MLGZ) were cloned and
sequenced, and mlgl mutants were generated. Again, the biochemical data and the
analysis of the mIgI mutants indicated that Mgla is the glycosylated form of Mlglb. The
capacity of a strain deficient in the synthesis of MG] gene products to colonize maize

tissue is discussed.

Redundancy is the root of all evil.

iii

ACKNOWLEDGMENTS

No one works on a project for four years without the help of others. Therefore, I
would like to thank all the people in the lab who have pitched in to help me accomplish my
goals. I would especially like to thank John Pitkin for literally standing by me for most of
the last four years even though I was typically ofl‘ by ten-fold. I learned most of my biting
sarcasm from John, while the remainder came from Robin Buell. It goes without saying
that everything I ever learned about the history of the PRL and my predecessors was
taught to me by John Scott-Craig. Ifthe PRL should be unfortunate enough to lose John
to another university, they should make him sign a secrecy agreement. I thank Ambro van
Hoof for being a very important part of my life. I also thank Fabienne Hamburger for
doing all those dreaded PABA’s with a smile on her face.

A great deal of thanks goes to my family for supporting me through the last nine years
of school. I promise some day I will get a “real job”.

I thank my committee members, Dr. Jonathan Walton, Dr. Kenneth Keegstra, Dr. John
Ohlrogge, and Dr. Gregory Zeikus for their guidance. A special thanks goes to the faculty
and stafl' of the PRL for giving me every opportunity to succeed in science. I feel that the
scientific standard set by the PRL is one of the highest and I am proud to say that I

received my Ph.D. while being a member of the PRL. I thank the NIH-Biotechnology

iv

Training Program for supporting me for most of my tenure at MSU. Thanks to John
Ryals, I was given the chance to fulfill my Industrial Internship at CIBA-GEIGY. During
my time there, I gained a new perspective on science and met J om, who as Jonathan said,
took care of me.

Finally, I would again like to thank Dr. Jonathan Walton for his patience with me even
though I had a tendency to redecorate his office and “foolishly” send letters for which I
deserved a sailboat ride to Alcatraz. I am now convinced that all of his great ideas come

fi'om the New York Times crossword puzzles and solitaire.

TABLE OF CONTENTS

LIST OF TABLES ......................................................................................................... ix
LIST OF FIGURES ......................................................................................................... x
INTRODUCTION ........................................................................................................... l
The Pathogen ....................................................................................................... 1
The Host-Pathogen Interaction ............................................................................. 4
CHAPTER 1
THREE EXTRACELLULAR PROTEASES FROM COCHLIOBOL US
CARBONW: CLONING AND TARGETED DISRUPTION OF ALP] ...................... 12
Abstract ............................................................................................................. 12
Introduction ....................................................................................................... 13
Results and Discussion ....................................................................................... 15
Characterization of Alpla, Alplb, and Alp2 ............................................ 15
Isolation and Characterization of ALP] ................................................... 19
N-terminal Sequencing of Alp2 ............................................................... 22
Transformation-Mediated Gene Disruption of ALP] ............................... 24
Materials and Methods ....................................................................................... 28
Fungal Culture Grth and Maintenance ................................................ 28
Enzyme Assays ....................................................................................... 29
Protein Purification and Characterization ................................................ 29
Nucleic Acid Manipulation and Sequencing ............................................ 31
Isolation of ALP] ................................................................................... 32
Disruption of ALP] ................................................................................ 33
CHAPTER 2

THREE MIXED-LINKED GLUCANASES FROM THE FILAMENTOUS
FUNGUS COCHLIOBOL US CARBON W : CLONING AND TARGETED

DISRUPTION OF MLGI .............................................................................................. 3 5
Abstract ............................................................................................................. 3 5
Introduction ....................................................................................................... 36
Results and Discussion ....................................................................................... 38

Characterization and Purification of Mlgla, Mlglb, and NIng ................. 38
Isolation and Characterization of MLGI ................................................. 43
Transformation-Mediated Gene Disruption of MLGI .............................. 48
Pathogenicity of mlgl Mutants ............................................................... SO

Materials and Methods ....................................................................................... 52

Fungal Culture and Maintenance ............................................................. 52

Enzyme Assays ....................................................................................... 52

Protein Purification ................................................................................. 53

Nucleic Acid Manipulation ...................................................................... 54

Cloning of MLGI ................................................................................... 55

Targeted Gene Disruption of A/ILGI ....................................................... 56
CHAPTER 3

INVESTIGATION OF THE INVOLVEMENT OF TWO PROTEASES IN
VIRULENCE OF THE FUNGAL PATHOGEN, C OCHLIOBOL US

CARBONUM, ON MAIZE ............................................................................................ 57
Abstract ............................................................................................................. 57
Introduction ....................................................................................................... 58
Results and Discussion ....................................................................................... 59

Isolation and Characterization of ALPZ ................................................... 59
Transformation-Mediated Gene Replacement of ALP2 ............................ 65
Materials and Methods ....................................................................................... 69
Fungal Growth and Maintenance ............................................................ 69
Protein Purification and Enzyme Assays .................................................. 70
Nucleic Acid Manipulation and Sequence Analysis .................................. 7O
Isolation of ALPZ ................................................................................... 70
Disruption of ALP2 ................................................................................ 71

CONCLUSION ............................................................................................................. 73
Future Objectives ............................................................................................... 73

APPENDIX A

PURIFICATION OF A PUTATIVE EXO-l,3-B-GLUCANASE AND MOLECULAR

CLONING OF THE GENE, EYGZ, FROM COCHLIOBOL US CARBONUM ............... 78
Introduction ....................................................................................................... 78
Results and Discussion ....................................................................................... 79

Protein Purification and Characterization ................................................ 79
Isolation and Characterization of the Gene Encoding p50, EXGZ ............ 80
Materials and Methods ....................................................................................... 85
Isolation of EXGZ ................................................................................... 85

APPENDIX B

CLONING OF THE GENE ENCODING A l,3-l,4-B-GLUCANASE, MLGZ, FROM

COCHLIOBOL US CARBON UM ................................................................................... 87
Introduction ....................................................................................................... 87
Results and Discussion ....................................................................................... 88
Materials and Methods ....................................................................................... 9O

Nucleic Acid Manipulation ...................................................................... 9O
Isolation of MLGZ .................................................................................. 91

vii

BIBLIOGRAPHY ..................
....................................................................................... 92

viii

LIST OF TABLES

Table 1 - Cell Wall Degrading Enzymes of C. carbonum .................................................. 9
Table 2 - Amino Acid Sequences from Purified Mlgla, Mlglb, and NIng Proteins ......... 43
Table 3 - Intron Features of C. carbonum Genes ............................................................ 62
Table 4 - Features of C. carbonum Cell Wall Degrading Enzyme N-Terrnini .................. 75

ix

LIST OF FIGURES

Figure l - Structure of HC-Toxin and Derivatives (Leisch et al 1982, Pope et al.

1983, Rasmussen and Schefi‘er 1988b, Walton et al. 1982) ............................... 1
Figure 2 - Efl‘ect of Difl‘erent Media Supplements on Total Protease Activity in

Culture Filtrates of C. carbonum .................................................................... 16
Figure 3 - Efl‘ect of Protease Inhibitors on Alpla, Alplb, and Alp2 ................................ 17
Figure 4 - Sequence of ALP] ......................................................................................... 21
Figure 5 - Comparison of the Predicted Amino Acid Sequences of ALP] and Four

Related Trypsin-Like Proteases Using PILEUP (Devereux et al. 1984) .......... 23
Figure 6 - A, Restriction Map of Wild type locus of ALP] Showing the Location of

the ALP] Transcript ....................................................................................... 25
Figure 7 - Cation Exchange HPLC Analysis of Proteins fi'om Culture Filtrates of (A)

Wild Type and (B) ALP] Mutant T515-1A .................................................... 27
Figure 8 - Purification of Mlgla, Mlglb, and Mng ........................................................ 40
Figure 9 - Glycosylation of Mlgla and Mlglb ................................................................ 42
Figure 10 - Nucleotide Sequence and Deduced Amino Acid Sequence of MLGI ............ 45
Figure 11 - Comparison of the Amino Acid Sequences of C. carbonum Mlgl

and Rhodothermus marinas B-Glucanase ...................................................... 47
Figure 12 - Analysis of the MLG] Locus in Wild Type and mlgl Mutants ...................... 49
Figure 13 - HPLC Analysis of MLGase from Mutant ..................................................... 51
Figure 14 - Sequence of ALPZ ....................................................................................... 61
Figure 15 - Comparison of ALPZ Predicted Amino Acid Sequence with Other Fungal
Subtilisin-Like Proteases ............................................................................... 64

Figure 16 - Restriction Maps of the Wild Type and Mutant Loci of ALP] and ALPZ ...... 66

Figure 17 - Protease Activity in Fractions Collected from Cation Exchange HPLC
Analysis of Proteins fiom Culture Filtrates of Wild Type (3 67-2A), an alp2

Mutant (T650-1A), and an alpI/aIpZ Mutant (T651-2A) .............................. 68
Figure 18 - Sequence of EYGZ ...................................................................................... 81
Figure 19 - Comparison of the Predicted Amino Acid Sequence of EYGZ with Other

Known Exo-1,3-B-Glucanases ...................................................................... 82
Figure 20 - Restriction Map of the Wild Type EXGZ Locus Showing the Location of

the EXGZ Transcript (Shaded Box) ............................................................. 83
Figure 21 - Sequences of Mlg2 Tryptic Peptide Fragments ............................................. 88
Figure 22 - Deduced Amino Acid Sequence of MLGZ Compared to Homologous

Protein Sequences ........................................................................................ 89

Figure 23 - Restriction Map of the MLGZ Locus Showing the Location of the
MGZ Transcript (Shaded Box) ................................................................... 90

INTRODUCTION

The Pathogen

Cochliobolus carbonum R. R. Nelson (anamorph, Bipolaris zeicola (G. L. Stout)
Shoemaker = Helminthosporium carbonum Ullstrup) is the causal agent of northern
leafspot of maize (Zea mays L.). This species of firngus can be found as several difi‘erent
races, which defines host specificity. Race 1, first described by Ullstrup (1941), produces
a host-selective phytotoxin called HC-toxin (Pringle and Scheffer 1967) which is required
for pathogenicity of this fimgus on maize (Schefi‘er and Livingston 1984). This toxin is a
cyclic tetrapeptide (Figure 1) which can be isolated as several forms, each with varying

activities (Rasmussen and Schefi’er 1988b, Liesch et al. 1982, Pope et al. 1983, Walton et

/N\
o=c HC-R
\\1
H3C-CH c=o
/ \
HN NH
o\c / H
\\H HC-(CH2)5-C-C-CH2
C / ll \/
Rz’gy I 330
CH2 ’CHZ

Toxin I R1-C83 R288
Toxin I I 3.1-3 Rz-B
Toxin III 1111-6113 33-08

Figure 1. Structure of HC-toxin and derivatives (Leisch et al. 1982, Pope et al. 1983,
Rasmussen and Scheffer 1988b, Walton et al. 1982).

al. 1982). However, toxicity absolutely requires the presence of the epoxide group and
the carbonyl group of Aeo (2-amino-9,10-epoxy-8-oxodecanoic acid) (Walton and Earle
1983, Ciufetti et al. 1983, Kim et al. 1987). Toxin production is conditioned by a single
genetic locus, T 0X2 (Nelson and Ullstrup 1961, Schefi‘er et al. 1967). Biosynthesis of
HC-toxin involves a cyclic peptide synthetase, HTS (Walton 1987, Walton and Holden
1988). The 15.7 kb open reading frame of the gene, H T S] , encoding this enzyme is found
in two copies in most race 1 isolates of C. carbonum (Ahn and Walton 1996).
Simultaneous disruption of both copies of H T S] results in a loss of HC-toxin production
and a concommitant loss in pathogenicity on maize (Panaccione et al. 1992).

In addition to the cyclic peptide synthetase, several other genes (T OXA, T OXC,
T 0m), unique to toxin producing isolates of C. carbonum, have been found which may
be involved in HC-toxin biosynthesis. The protein encoded by T 0X4 exhibits a high
degree of similarity to small molecule efilux pumps and may be involved in secreting HC-
toxin and/or protecting C. carbonum from the toxic effects of HC-toxin (Pitkin et a1.
1996). Supporting the argument that the T OXA gene product is essential for protection of
the fimgus fiom HC-toxin, Pitkin et al. (1996) were not able to recover mutants with
disruptions in the two copies of T OXA. T OXC, which appears to encode a fatty acid
synthetase, might be involved in the synthesis of Aeo (Figure 1) (Ahn and Walton, 1997).
Gene disruption of all three copies of T OXC result in a loss of toxin production and
pathogenicity. T 0H), found in three copies in C. carbonum race 1, has no known role in
biosynthesis of HC-toxin (unpublished data). Gene disruptions of the T 02a) loci have no

efi‘ect on toxin production or pathogenicity. In addition to the genes already identified,

there may be genes encoding proteins which participate in cyclization of the toxin and
regulation of toxin production.

Resistance to race 1 is governed by a single dominant gene, Hm (Nelson and Ullstrup
1964). Meeley and Walton (1991) found that a cell-free extract from a resistance maize
genotype (Hm/hm) was able to inactivate HC-toxin by the reduction of the carbonyl group
on Aeo (Figure 1). Later, Meeley et al. (1992) showed that this HC-toxin reductase
activity was found only in extracts from maize genotypes resistant to C. carbonum race 1
(Hm/hm, Hm/Hm) and not in those that were susceptible (hm/hm). These studies
demonstrated that the biochemical activity of the enzyme encoded by the Hm locus was an
HC—toxin reductase. Johal and Briggs (1992) subsequently cloned the Hm gene and
showed that it had sequence similiarity to other known reductases.

HC-toxin has a variety of efi‘ects on susceptable tissues and cells. It can inhibit the
biosynthesis of chlorophyll (Rasmussen and Schefi‘er 1988a), inhibit mammalian cell
division (Walton et al. 1985), and increase the uptake of certain amino acids, ions, and
nitrate (Yoder and Scheffer 1973a, 1973b). It seems that HC-toxin is not really a toxin
but rather a cytostatic agent which may in some way alter or block the ability of the host
to mount a defense response, thus allowing the fungus to ramify in the host tissue.
Towards defining the mode of action of HC-toxin, Brosch et al. (1995) have shown that
three forms of maize histone deacetylase were significantly inhibited by HC-toxin. This
may indicate that HC-toxin affects gene regulation by altering histone binding to DNA.

A second race of C. carbommr, race 2, does not produce HC-toxin. HTS], T OXA,

T OXC, and T OXD are missing in race 2 isolates of C. carbonum (Ahn and Walton 1996).

The morphology of race 2 lesions (small, necrotic flecks) on both susceptible (hmfltm) and
resistant (Hm/hm) genotypes of maize is similar to that of race 1 on Hml- maize (Nelson
and Ullstrup 1961). If HC-toxin is present in a race 2 inoculation, the fungus is able to
colonize susceptible maize leaves in a manner similar to race 1 (Comstock and Schefi‘er
1973). Collectively, the introduction of C. carbonum-resistant maize varieties and the
possible fitness disadvantage of race 1, due to the burden of toxin production (Leonard
1987), has made race 2 the predominant isolate found in maize fields (Leonard 1978).
Though C. carbonum is not an economic threat to maize growers today, it is an
excellent system to study plant-pathogen interactions. It grows well under most
conditions, it is genetically stable, there are two mating types which allow for Mendelian
genetic analyses, transformation-mediated gene disruptions can be performed routinely,
and most importantly the disease determinant (HC-toxin) has been defined and shown to

be the key factor for a successfirl disease phenotype on susceptible maize.

The Host-Pathogen Interaction

Pathogens have devised a diverse array of mechanisms to colonize host tissues.
Bacterial pathogens can invade plant tissue by taking advantage of natural openings due to
injury, hydathodes, stomata, and spaces where lateral roots have emerged. In addition,
cell wall degrading enzymes are required by several of the soft-rotting bacteria to macerate
tissue and cause disease (Liao et al. 1988, Boccara et al. 1988). Viruses can only be
introduced by vectors such as fimgi and insects or mechanically. Phytopathogenic fimgi,

on the other hand, use not only natural openings of plants but also generate structures

specialized for host penetration. Some rust fimgi search out stomatal openings by
orienting germ tubes (primary hyphae originating from the spore) at right angles to the cell
wall junctions until a stomatal apparatus is located (Staples and Macko 1980, Allen et al.
1991). Upon sensing a stomate, germ tube growth ceases, an appressorium (a specialized
structure which facilitates adhesion and penetration) is formed over the opening, and the
penetration peg pushes between the guard cells down into the substomatal cavity.

Howard and Ferrari (1989) have demonstrated that mechanical pressure is necessary
for Magnaporthe grisea to penetrate a surface as resistant as a cell wall. High internal
hydrostatic pressure is generated in the appressorium by an increase in solutes and water
influx. This pressure can be maintained because of a layer of melanin in the appressorium
which is impermeable to solutes (Rast et al. 1981). The pressure is then focused at the
appressorium pore, located at the point of attachment to the plant surface, and a
penetration peg is forced through the plant cell wall. This does not occur in melanin-
deficient M. grisea mutants, which are consequently nonpathogenic (Chumley and Valent
1990). At least two other fiingal species, Pyricularia and Colletotrichum, also require
melanized appressoria for penetration of the host plant (Kubo et al. 1991, Woloshuk et al.
1980, Rasmussen and Hanan 1989).

An alternative to mechanical pressure is cell wall dissolution. For many years, plant
pathologists have tried to use the knowledge gained from plant cell wall structural analysis
to determine what components of the plant cell wall are most important for cell wall
structural integrity and protection from the environment. As it is yet unkown what cell

wall macromolecules in any plant species maintain wall integrity, plant pathologists have

had to make educated guesses about which pathogen-secreted cell wall degrading enzymes
should be investigated first. Studies concerning the interactions between Leptosphaeria
maculans and oilseed rape (Easton and Rossall 1985), Colletotn‘chum lindemuthianum
and bean (Wijesundera et al. 1989), Verticillium albo-atrum and tomato (Cooper and
Wood 1980), and Gaeumannomyces graminis var. tritici and wheat (Dori et al. 1995)
have demonstrated a correlation between the levels of various cell wall degrading enzyme
activities and the formation of disease. Le Carn et al. (1994) have even shown that the
difi‘erence in timing and magnitude of secretion of cell wall degrading enzymes by four
difi‘erent isolates of Mycocentrospora acerina may determine the degree of aggressiveness
on carrot. Using monoclonal antibodies against major plant wall constituents, Xu and
Mendgen (1997) demonstrated that the density of carbohydrate epitopes decreased upon
infection with Uromyces vignae. The authors propose that the density decrease was due
to the action of degradative enzymes secreted by the fungus. Many more studies on the
identification of individual enzymes and their potential role in virulence of firngi have been
reviewed (Cooper 1984, Walton 1994).

Of these numerous studies, only two cell wall degrading enzymes, in a defined number
of host-pathogen interactions, have been shown to have a potential role in virulence. The
work of Dickman et al. (1989) demonstrated that the introduction of a cutinase gene fi'om
Nectria haematococca into the obligate wound pathogen Mycosphaerella enabled this
firngus to breach an intact cuticle and be pathogenic on papaya. Moreover, cutinases may
be important for attachment to the plant surface as the addition of cutinase to Uromyces

viciae-fabae spores increased the adhesion to bean leaves 2-fold (Deising et al. 1992). A

UV-induced, non-pathogenic mutant of Pyrenopeziza brassicae was characterized as also
being deficient in protease production (Ball et al. 1991). The non-pathogenic phenotype
co-segregated with protease deficiency and both could be complemented with a 40-kb
genomic cosmid clone. To date, the gene encoding this protease has not been cloned and
disrupted to demonstrate that pathogenicity is unequivocally due to a single protease.

In the case of C. carbonum, 90% of fungal penetrations are at cell junctions and 10%
are through stomatal openings (Jennings and Ullstnrp 1957, Murray and Maxwell 1975).
Therefore, host invasion is primarily by mechanical pressure and/or enzymatic
degradation. Melanin-deficient mutants of both C. miyabeanus, a pathogen of rice, and C.
carbonum are fully pathogenic (Kubo et al. 1989, J. Pitkin unpublished data). Therefore,
if mechanical pressure is the means by which these firngi penetrate the host tissue, then
there must be some other component, yet to be found, that allows for the increase in
hydrostatic pressure in the appressorium.

Supporting the hypothesis that cell wall degrading enzymes play a role in maize
colonization by C. carbonum, Murray and Maxwell (1975) concluded from an
ultrastructural analysis that the initial penetration by C. carbonum was probably due to
enzymatic degradation rather than mechanical means. They observed that the host cuticle
was not pushed inwards by the penetration peg, but instead appeared to be initially
dissolved and then ruptured by mechanical pressure. Furthermore, they suggested that the
swollen endoplasmic reticulum above the penetration site may be involved in the transport
of cell wall degrading enzymes and material for the grth of the penetration peg. In a

related firngus, C. sativus, penetration and grth of the hyphae between the host cells is

believed to involve both cell wall depolymerization and mechanical mechanisms (Huang
and Tinline 197 6).

The cell wall degrading enzymes that have been identified from Cochliobolus fall into
many difi‘erent classes. Deshpande and Deshpande (1968) and Nelson and Sherwood
(1968) first demonstrated that Helminothosporium appicum and C. carbonum,
respectively, secreted pectin degrading enzymes. Later, cellulase, cutinase, and xylanase
activities were characterized in C. sativum, C. carbonum, and C. heterostrophus (Muse et
al. 1972, Bateman et al. 1973, Baker and Bateman 1978). More than a decade later,
Walton and Cervone (1990) purified and characterized an endopolygalacturonase (PGNI)
from C. carbonum. The gene corresponding to this protein was next cloned and a null
mutant was generated to evaluate the role of pectin degradation on pathogenicity of maize
(Scott-Craig et al. 1990). Subsequently, twelve more secreted cell wall degrading
enzymes were purified and characterized from C. carbonum (Table 1). The genes
corresponding to eleven of these proteins have been cloned and five additional genes for
other cell wall degrading enzymes have been cloned by homology with sequences from
known genes (Table 1). Transformation-mediated gene disruptions have been performed
for most of these genes and virulence of the resulting null mutants evaluated (Table 1).

To date, no single cell wall degrading enzyme secreted by C. carbonum is essential for the
successful colonization of this fungus on maize. Even a strain deficient in multiple
xylanases has no efi‘ect on virulence; however, residual xylanase activity was detected in

vitro (Apel-Birkhold and Walton 1996).

TABLE 1. Cell Wall Degrading Enzymes of C. carbonum.

Mutant

Enzyme Activity Purified Gene Name Phenotype
a-Arabinosidase Xa n.d.
Serine Protease (Trypsin-Like) Xb ALP] b #b
Serine Protease (Subtilisin-Like) xb ALP2° #°
Endo-B-1,4-Glucanase CEL]d #d
Cellobiohydrolase CELZC n.d.
Cutinase CUTIf’B #9
Exo-B-1,3-Glucanase 1 Xi EXGIj Sk
Exo-B-l,3-Glucanase 2 xf EXG2f n.d.
Endo-B-l,3-1,4-Glucanase/1,3-Glucanase 1 X1 MLG]l itl
Endo-B-l,3-l,4-Glucanase/1,4-Glucanase 2 X1 MLGZf n.d.
Endopolygalacturonase Xm PGNI“ #“
Exopolygalacturonase X° PGX1° #°
Pectin Methylesterase Xg PME]g n. d.
Endo-B-l,4-Xylanase l Xp XYqu #‘1
Endo-B-1,4-Xylanase 2 X712” its
Endo-B-1,4-Xylanase 3 XYL38 #8
Endo-B-l,4-Xylanase 4 Xt XYL4g n.d.
B-Xylosidase xa XYPI“ n.d.

- 5, indicates an altered phenotype was observed in vitro due to mutation.

- #, indicates no difference in phenotype in vivo or in vitro due to mutation.

- n.d. = not determined.

- a, Ransom and Walton 1996; b, Murphy and Walton 1996; c, Gorlach et al. 1997b; (1,
Sposato et al. 1995; e, J .-H Ahn unpublished data; f, J .M. Gorlach unpublished data; g,
1.8. Scott-Craig unpublished data; h, J .W. Pitkin unpublished data; i, van Hoof et al. 1991;
j, Nikolskaya unpublished data; k, Schaeffer et al. 1994; l, Gorlach et al. 1997a; m, Walton
and Cervone 1990; n, Scott-Craig et al. 1990; o, Scott-Craig et al. 1996; p, Holden and
Walton, 1992; q, Apel et al. 1993; s, Apel-Birkhold and Walton 1996; t, P.C. Apel-
Birkhold unpublished data; u, R. Ransom unpublished data.

10

In light of these fi'ustrating results, several interesting observations have been made.
First, xylanase activities and gene expression patterns indicated that of the three xylanase
family G class xylanases, only XYLI and XYL3 are expressed in planta. In contrast, only
XIII and XYL2 are expressed in vitro, with enzyme activity only being detected for
Xyll(Apel-Birkhold and Walton 1996). Second, a mutant of EXG] is still pathogenic but
is impaired in its ability to grow on 1,3-B-glucan as the sole carbon source (Schaefi‘er et a1.
1994). Third, a double disruption in PGN] and PGX] diminishes nearly 99% of all
polygalacturonase activity, yet there is no measurable differences in the growth rate or dry
weight of this mutant compared to wild type when grown on pectin as the sole carbon
source (J .S. Scott-Craig, unpublished data). Support of grth may be due to detectable
pectin methylesterase activity. Multiple gene disruptions in all the presently identified
xylanases (XYLI through XYL4) and pectin-degrading enzymes (PGNI, PGX], PMEI) are
currently in progress.

Whether or not cell wall degrading enzymes have a direct impact on the penetration of
maize by C. carbonum is still unclear. One must realize, however, that cell wall
degrading enzymes may not only have a direct role in penetration, they may also be
important in fungal growth and differentiation, as the already mentioned role for cutinase
in adhesion (Deising et al. 1992). Independently, two groups of researchers demonstrated
that S. cerevisiae mutants deficient in a secreted exo-1,3-B-glucanase (SPRl or SSGl)
had reduced spore thennoresistance and delayed mature ascus formation (San Segundo et
al. 1993, Muthulcumar et al. 1993). They speculated that the exo-1,3-B-glucanase may

soften the pro-existing wall or, as seen in Candida albicans, firnction as a transferase to

ll

elongate 1,3-B-glucans being synthesized by glucan synthetase. Moreover, cell wall
degrading enzymes could have a direct or indirect role in warning the plant that the
pathogen is present. For instance, an endo-B-1,4-xylanase protein was identified as being
responsible for ethylene biosynthesis and necrosis (Sharon et al. 1993, Bailey et al. 1990).
Cell wall degrading enzymes might also release oligosaccharides from their own wall or
fi'om that of the plant wall which then act as potent elicitors of the hypersensitive
resistance response (C6te' and Hahn 1994). Finally, wall depolyrnerases undoubtedly
function as a means of saprophytic nutrient acquisition and this may be their only role in
some phytopathogenic organisms.

This study was undertaken to evaluate the role that several difi‘erent plant cell wall
degrading enzymes play in the colonization of maize by C. carbonum. Three proteases
were purified and characterized, the corresponding genes were cloned, and gene
disruption experiments were performed to investigate the involvement of proteases in the
growth and virulence of C. carbonum. The purification and biochemical charactization of
three enzymes able to degrade l,3-1,4-B-glucans was also accomplished. The genes
encoding these enzymes were obtained and a gene disruption in one was performed to
evaluate its role in the grth and virulence of C. carbonum. Furthermore, N-terminal
amino acid sequence of an exo-B-1,3-glucanase aided in the cloning of its corresponding

gene.

CHAPTER 1

Jenifer M. Murphy and Jonathan D. Walton. 1996. Three extracellular proteases from
Cochliobolus carbonum: cloning and targeted disruption of ALP] . Mol. Plant-Microbe
Interact. 9: 1091-1098.

Chapter 1
THREE EXTRACELLULAR PROTEASES FROM
COCHLIOBOL US CARBON UM: CLONING AND TARGETED
DISRUPTION OF ALP]
Abstract
Three extracellular serine proteases(Alp1a, Alplb, Alp2) from Cochliobolus
carbonum were purified and characterized. Of eight carbon/protein substrates tested, total
protease activity was highest when the fungus was grown on medium containing collagen.
Alpla and Alplb are members of the trypsin family (EC 3.4.21.4), and Alp2 is a member
of the subtilisin family (EC 3.4.21.62). Alpla, Alplb, and Alp2 have monomer molecular
masses of 25-kD, 30-kD, and 38-kD, respectively. Alplb is glycosylated whereas Alpla is
not. The gene encoding Alpla, ALP], was isolated using PCR primers based on two
amino acid sequences: one obtained directly fiom the N-terminus of Alpla and another
that is highly conserved in other trypsins. The transcriptional start site was determined
using RACE and the intron structure and polyadenylation site were determined from a
cDNA clone. An internal fragment of ALP] was used to create Alpla null-mutants by
transformation-mediated gene disruption. Total protease activity in the mutants was
reduced by 35 to 45%. By chromatographic analysis, the mutants had lost two peaks of
UV absorption and the two protease activities corresponding to Alpla and Alplb, which,

together with the biochemical data, indicates that Alpla and Alplb are products of the

12

13

same gene. The in vitro growth and disease phenotypes of the ALP] mutants were
indistinguishable fi'om the wild type strain; therefore, ALPl is not by itself required for

pathogenicity.

Introduction

The first barrier a phytopathogenic microbe encounters on its host is the plant cell
wall, and cellular pathogens secrete a variety of wall-depolymerizing enzymes. The role, if
any, of these enzymes in the process of pathogenesis has been the subject of intensive
research (Bateman and Basham 1976; Cooper 1983; Walton 1994).

Although the plant cell wall is mainly composed of polysaccharides, at least five
classes of structural proteins and numerous classes of enzymes are present (McNeil et al.
1984; Showalter 1993). During pathogenesis the expression of many structural proteins
and enzymes are up-regulated and secreted into the plant cell wall as part of a general
defense response (Showalter 1993; Alexander et al. 1994). Insofar as these wall proteins
are important to plant defense, efi‘ective pathogens may require extracellular proteases to
degrade them. The proteases secreted by a pathogen might also be important as activators
of any of its cell wall degrading enzymes that are secreted as zymogens (Drapeau 1978;
Rypniewski et al. 1993; Moormann et al. 1993), or as processors of toxins (Howard and
Buckley 1985) and elicitors (van den Ackerveken et al. 1993). Proteases might also have
a role in pathogenesis by increasing the permeability of the plant plasma membrane (Tseng
and Mount 1974). Many plants produce protease inhibitors, which suggests that plants

have evolved mechanisms to counter pathogen proteases (Ryan 1990).

14

Extracellular proteases are produced by many phytopathogenic bacteria and fungi,

e. g., Xanthomonas alfalfae (Reddy et al. 1971), Moniliniafiuctigena (Hislop 1982),
Colletotrichum lindemuthianum (Ries and Albersheirn 197 3), and others (Porter 1966).
An aspartic protease gene has been isolated fiom Cophonectria parasitica (Choi et al.
1993). A protease mutant of Cladosporium cucumerinum had wild type symptom
development on cucumber seedlings, but residual protease activity was present (Robertson
1984). No reduction in virulence was observed in a metalloprotease mutant of Erwinia
chrysanthemi EC 16 when inoculated on either potato tubers or chrysanthemum stems
(Dahler et al. 1990). A protease-deficient mutant of Xanthomonas campestris pv.
campestris had reduced virulence when introduced into the cut vein endings of turnip
leaves (Dow et al. 1990). A UV-induced mutant of the fungus Pyrenopeziza brassicae
that had lost both protease activity and pathogenicity could be complemented for both
traits by a 40-kb genomic cosmid clone (Ball et al. 1991).

One definitive test of the role of any particular gene in pathogenesis is the construction
of a null mutant using targeted gene disruption (e. g., Scott-Craig et al. 1990). Towards
the goal of testing the role of proteases in plant pathogenicity, we report here the
characterization of three proteases that C. carbonum secretes when grown on collagen,

and the sequence and disruption of ALP] , which encodes two of these proteases.

15

Results and Discussion
Characterization of Alp] a, Alp] b, and Alp2.

To facilitate the study of proteases, we investigated which growth conditions
maximized total protease activity. Collagen was tried as a substrate because it is a
hydroxyproline-rich glycoprotein like the extensins of the plant cell wall. Stimulation of
the de novo synthesis as well as cross-linking of structural wall proteins such as extensin in
response to pathogen attack is thought to be an important plant defense response
(Showalter 1993; Lawton and Lamb 1987). Therefore, fungal proteases that are induced
in planta to degrade wall structural proteins might also be induced in culture by collagen.
Preliminary experiments indicated that of the potential protease substrates, C. carbonum
produced most total protease on medium supplemented with collagen. Other protein
sources (casein, gelatin, and bovine serum albumin) yielded less total protease activity than
collagen (Figure 2). Low levels of protease activity were observed on media
supplemented with pectin, corn bran, or 2.0% sucrose (Figure 2). In the presence of
collagen, 0.2% sucrose stimulates protease production (Figure 2), an effect also seen with
polygalacturonase (Walton and Cervone 1990) and xylanaes (Holden and Walton 1992).
High protease production on protein substrates and low production on other substrates
indicate that total protease activity in C. carbonum is substrate-induced and partially
catabolite-repressed.

Culture filtrates were concentrated by rotary evaporation, dialyzed, and then passed
over a low-pressure anion-exchange column to remove acidic proteins and pigments.

Alpla, Alplb, and Alp2 were then separated by cation-exchange HPLC. Alpla was

16

 

Protease Activity ('/o)

 

 

 

+ + := .

53 s 83 a g E

s§ :9 3% f8: .5 3,1.» g gag g g

L"m = :1 o "to -—

8,. 8 8\e a a? £3 a .85 a i

so! .\° :22 .\° .\° .\° .\° .\° .\° .\°

"O F" F‘N N '— —- — w-na F-l r-n
Medium Supplement

Figure 2. Efi‘ect of different media supplements on total protease activity in culture
filtrates of C. carbonum. Protease activities are shown as a percentage of the highest
activity measured. Experiment was repeated three times with similar results. Data fiom
one experiment is presented in this figure.

l7

 

 

 

 

 

 

 

180
160 ~— IAlp1a
x 140 ~—
[:iAl 1b
.3» 120 —- p
g 100 -— IAIDZ
3 80 '— , ,f
is” 6° ‘" iii?
is’ 40 * :1; 12;: , ‘
S 20 T ‘ I‘ I I
0 ._. 7 a
'a a < e .s .g .
a s E ‘5 E E’ s g
E 2 m n e E § §
6 °° E E E E < 3 e-
2 ° 3 ._ :- ._. 2 E E
o' g; c

Inhibitor

Figure 3. Effect of protease inhibitors on Alpla, Alplb, and Alp2. Each enzyme was
incubated with azocasein in the presence of the inhibitor being tested for 30 min at 45°C.
Protease activities are expressed as a percentage of activity obtained in the control. Data
are the average of two experiments. DTT, dithiothreitol; EDTA, ethyelenediamine-
tetraacetic acid; BME, B-mercaptoethanol; PMSF, phenylmethylsulfonylfluoride.

18

further purified by hydrophobic-interaction HPLC and sequenced at the N-terminus. By

SDS-PAGE, Alpla, Alplb, and Alp2 had Mr’s of 25-kD, 30-kD, and 38-kD,

respectively, similar to serine proteases (North 1982). Alpla and Alplb activities were
inhibited strongly by aprotinin (77% and 85%, respectively) and leupeptin (76% and 96%,
respectively) and weakly by PMSF (37% and 37%, respectively) (Figure 3), suggesting
that these proteases are related to trypsin (Gebhard et al. 1986; Powers and Harper 1986).
Alp2 was more sensitive to PMSF (81% inhibition) than to aprotinin or leupeptin (40%
and 47% inhibition, respectively) (Figure 3), suggesting that it is related to the subtilisin
family of proteases (Ottensen and Svendsen 1970). Alpla, Alplb, and Alp2 were less
sensitive to other major classes of protease inhibitors (Figure 3).

Periodic acid/Schifi‘ staining indicated that Alplb is a glycoprotein whereas Alpla
lacks glycosylation (data not shown). All three enzymes were most active between pH 7
and pH 8, but each showed some activity over the pH range of 5 to 11. Temperature
optima were similar for all three enzymes. Each enzyme was as active at 45°C as at 55°C
but lost activity at 65°C and above. Alpla and Alplb seemed to be less stable than Alp2
based on the observation that after being stored at -20°C and then separated by SDS-
PAGE, Alpla and Alplb were degraded whereas Alp2 was not.

To test the possibility that plant cell wall structural proteins might be substrates for
these proteases, salt-extractable extensin was purified fi'om maize stylar tissue (Murphy
and Hood 1993). Neither crude culture filtrates nor the proteases individually or in
combination could degrade extensin, whereas in parallel experiments casein was degraded

to small peptide fragments (data not shown). Therefore, these proteases probably do not

19

have a role in degradation of this class of maize cell wall structural proteins during
pathogenesis.

Although no evidence for additional proteases was found, we cannot exclude the
possibility that C. carbonum makes other proteases, which might have been overlooked
because (1) they are unstable, (2) they are not produced on the substrates tested, (3) they
are acidic and therefore retained by the DEAE-cellulose pre-treatment, or (4) they are

active only below pH 5, under which conditions the substrate, azocasein, precipitates.

Isolation and characterization of ALP] .

The N—terminal amino acid sequence of Alpla was determined to be
IVGGTTAAAGEYPFIVS (indicated by double—underlining in Figure 4). A search of the
non-redundant databases using BLASTP (Gish and States 1993) identified a 76% identity
with the N-terminus of a 22-kD trypsin-like protease from F usarium oxysporum. A 48-
fold degenerate oligonucleotide based on the amino acid sequence EYPFIV was used in
conjunction with a 256-fold anti-sense degenerate oligonucleotide coding for the amino
acid sequence VAGWGA (also indicated by double-underlining in Figure 4), which is a
highly conserved internal amino acid sequence of many trypsins. Using these two primers
and DNA isolated fi'om a C. carbonum cDNA library as template in PCR, a 330-bp
product was generated. BLASTX analysis of the PCR product showed a high degree of
similarity with trypsin-like proteases. The PCR product was used as a probe to screen a
C. carbonum cDNA library. A 1.0-kb cDNA clone (pC8-6. 1) was isolated and

sequenced, and also used to screen a library of C. carbonum genomic DNA.

20

Oligonucleotide primers that had been used to sequence the cDNA copy of ALP] were
used to sequence, on both strands, approximately 1.5-kb of genomic DNA covering
AIJ’] .

Figure 4 shows the sequence and structure of ALP] and its deduced amino acid
sequence. The start of the ALP] message was determined by sequencing three
independent RACE products (F rohman et al. 1988). The context of the first ATG, 91-bp
downstream of the transcription start site (CACCflQCGT) (Figure 4), is in good
agreement with the 5' end of the consensus sequence for Neurospora translation initiation
(CAMMA_T_GGCT where M = C or A) (Edelrnann and Staben 1994). Sequences typical
of promoters of lower eukaryotes, TATAA and CAAC (Gurr et al. 1987), are located 34-
and 72-bp, respectively, upstream of the transcription start site (Figure 4). A single 74-bp
intron deduced by comparing the cDNA and genomic sequences of ALP] is indicated by
lower-case letters (Figure 4). The 5' (GAGTAAGTTCACTCA; consensus
GAGTAAGTNNYCNYY, where Y = T or C) and 3' (AACAG; consensus WACAG,
where W = A or T) donor sites as well as the splice branch site (AACTAACA; consensus
WRCTRACM, where R = A or G) and intron length are consistent with other introns of
C. carbonum and other fungi (Apel et al. 1993; Scott-Craig et al. 1990; Sposato et al.
1995; Edehnann and Staben 1994). No AATAAA polyadenylation signal sequence (Gurr
et al. 1987) could be identified before the polyadenylation site 229-bp downstream of the
stop codon (Figure 4).

ALP] is predicted to encode a mature protein of 261 amino acids and a mass of 24.5-

21

1 GTAGATAAGGCAGCTGCTAGGCTCGGGGTAATTGGTCTCCACTGCTTGAT
51 CTAAGGCACACGGCGGACTACAGGTTAAGACTTTGCCAAGCCATAATAGG
101 TCCCAAACTGGAAGGACAAATCGTACTCCTAGTAGATCAGTCTTGGTATA
151 CCCTAAAGCACGCCATGACAAGGTCGGCCTTCTACCATCTACACAACCAT
201 TGGAGTGTCCATGACCGACAACGACATTCACTATAAGTATCCAGCAATCT
5
251 GCCCTCTGTAATCATCATCAGCCAGCTCATCCAGTCGCTTGTCTCTTCAA
301 ACCATTCCATCGCTTCTCTCCAACAGTCGCTGCCTTTGCAAAGCCCATCA
351 TTCACCATGCGTTTCCAGTCTATGATCACTGCTGCGCTTCCTGCGCTCGT
M R P Q S M I T A A L P A L V
401 CCTCTCCGCTCCTACTCCCCAGTGGGATGATGTTCCTGAGGACTCCATTG
16 L S A P T P Q N D D V P E D S I
451 TTGGTGGAACCACCGCTGCTGCCGGCGAGTACCCCTTCATCGTCTCTATC
32 V G G T T A A A G E Y P F I V S I
501 CAGCTTGGCGGTCGCCACAACTGCGGTGGTACCCTCATCAACGGCAACAC
49 Q L G G R H N C G G T L I N G N T
551 CGTTGTCACTGCTGCCCACTGCTCCGTCAGCAGCGCCATTGGCGGCTCCA
66 V V T A A H C S V S S A I G G S
601 TCAACAACGTCGCTGTCCGCGTCGGCTCCTTGgtaagttcactcatctga
82 I N N V A V R V G S L
651 tacagtactttatgcacttggcaaaggacaaaqaaactaacacaagttct
701 aaacagAGCGCCAACTCTGGTGGCCAAGTCATCAAGGTCTCCAAAATCAT
93 S A N S G G Q V I K V S K I I
751 CATCCACCCCAGCTACCAGGCAAGCACCTCCAACAACGACATTGCCATCT
108 I H P S Y Q A S T S N N D I A I
HindIII
801 GGAAGCTTTCCAGCACCGTCACTGCCGGTGGCAACATCGGCTTTGCTTCC
124 W K L S S T V T A G G N I G F A S
BamHI
851 CTCGCCGCCTCTGGCTCTGATCCCGCCAGCGGATCCACCACCTCCGTTGC
141 L A A S G S D P A S G S T T S V A
901 TGGATGGGGAGCTACCCGTGAGGGTGGCGGCGCCAACAACGCTCTCCTCA
158 G w G A T R E G G G A N N A L L
951 AGGTCAGCGTCCCCATTGTTGCCCGCTCCACCTGCGTGTCCAACTACAAC
174 K V S V P I V A R S T C V S N Y N
1001 GCCGTCGGTCTCACCGTCACCACCAACATGGTCTGCGCTGGTGTCACTGC
191 A V G L T V T T N M V C A G V T A
SalI
1051 TGGTGGCCGCGACTCTTGCCAGGGCGACTCTGGCGGCCCTCTCGTCGACG
208 G G R D S C Q G D S G G P L V D
1101 CCAACAAGACCCTCATCGGCGTCGTCTCCTGGGGAACCGGCTGCGCTCGC
224 A N K T L I G V V S W G T G C A R
1151 CCCAACCTCCCCGGTGTCTACTCCCGCGTCGGCACCCTCCGCAGCTTCAT
241 P N L P G V Y S R V G T L R S F I
1201 CGACCAGAACGCTTAAGCGCGTACATCTTGAAAGCGAGTTGGATATGATT
258 D Q N A *
1251 TGGAAACGGTCGACTTTGGATATGAAAAGAGCAATGGCTTTGATGAGTAT
1301 GGTATGGGGGAGACCCTGAAGTTGGGAGGGAAAACGGTGATGATGGACTT
1351 TGCTTTTTTACTTTACCTCTTCTCCCTCCTTAATTTCGGTGACGGCATCT
1401 TGTAAATAGGTCTAGCCTCCCACGATTATTTTTTGTCTGTACCTTATTTT
+
1451 TTTCCTTTGTGTAGCTAGGAAATCGCATTGTGTGTTGGAACAACATCCCT
1501 TTTTTGTGCTCTC

 

Figure 4. Sequence of ALP] . The amino acid sequence of the mature N-terminus
obtained directly from Alpla, IVGGTTAAAGEYPFIVS, is indicated by double-
underlining. The sequence conserved among other trypsin-like proteases and used to
design the second oligonucleotide primer for PCR amplificatiomVAGWGA, is also
indicated by double-underlining The intron is indicated in lower case letters. The
indicated HindIII, SalI, and BamHI restriction sites are those used to construct and
linearize the disruption vector pJM9. The transcriptional start site is indicated by # and
the polyadenylation site by + (symbols refer to nucleotides underneath and amino acid
codes refer to nucleotides above). The predicted N-glycosylation site (amino acid 225) is
indicated in bold lettering.

22

kD, which is in agreement with the size of Alpla estimated by SDS-PAGE. A single
predicted N—glycosylation site (NKT) occurs at amino acid 225. Comparison of the
predicted N-terminal amino acid sequence with the experimentally-determined sequence of
the mature N-terminus (double-underlined in Figure 4) reveals a 30-amino acid signal
peptide with a potential signal peptide cleavage site that violates the (-3, -1) rule (von
Heijne 1986). Many trypsin—like proteases are synthesized as prepropeptides
(trypsinogens) that undergo a second proteolytic processing after removal of the signal
peptide, e.g., mammalian trypsins (Walsh and Wilcox 1970). Alpl might also be
synthesized as a prepropeptide.

The predicted amino acid sequence of Alpla has a high degree of similarity with other
trypsin-er proteases fiom a variety of species (Figure 5). Alpla contains the catalytic
triad common to all serine proteases (His-71, Asp-120, and Set-217) (Neurath 1984).
The sequence flanking the active site serine (CQGDSGGP) is completely conserved in the

other trypsin-like proteases (Figure 5).

N—terminal sequencing of A [[22

Fractions from the cation-exchange HPLC step that had Alp2 activity were run on
SDS-PAGE, blotted, and three proteins that were visible with Coomassie R—250 staining
of the blot were excised and sequenced from their N-terrnini. Two of the proteins had N-
terrninal sequences with no strong similarity to any sequences in the non-redundant
databases, whereas one of the proteins, which showed the best correlation between

staining intensity and Alp2 activity among the HPLC fiactions, gave the sequence

23

1 # 50
C. carbonum ...... MRFQ SMITAALPAL VLSAPTPQWD DVPEDSIVGG TTAAAGEYPF
F. oxysporum .......... ..MVKEASVV ALVAPLAAAA PQEIPNIVGG TSASAGDFPF
B. mori .......... MTNSLLICFT ILGLAASSPK PIGDIRIVGG EDIVITEAPY
B. taurus .............................. VDDDDKIVGG YTCGANTVPY
S. griseus MKHFLRALKR CSVAVATVAI AVVGLQPVTA SAAPNPVVGG TRAAQGEFP.
**** * * *** i
51 100
C. carbonum IVSIQLGGRH NCGGTLINGN TVVTAAHCSV SSAIGGSINN VAVRVGSLSA
F. oxysporum IVSISRNGGP WCGGSLLNAN TVLTAAHCVS GYAQSG.... FQIRAGSLSR
B. mori QVSVMFRGAH SCGGTLVAAD IVVTAAHCVM SEAPED.... YRIRVGSSFH
B. taurus QVSLN.SGYH FCGGSLINSQ WWSAAHCYK S ........ G IQVRLGEDNI
S. griseus MVRLSMG... .CGGALYAQD IVLTAAHCVS GSGNNTS... ..ITATGGVV
** i * *** i * ******* i t * *t
101 150
C. carbonum N. . .SGGQVI KVSKIIIHPS YQASTSNNDI AIWKLSSTVT AGGNIGFASL
F. oxysporum T. . .SGGITS SLSSVRVHPS Y. .SGNNNDL AILICLSTSIP SGGNIGYARL
B. mori Q...RDGMLY DVGDLAWHPD FNEASMDNDI AILWLPKPVM FGDTVEAIEM
B. taurus NVVEGNEQFI SASKSIVHPS YNSNTLNNDI MLIKLKSAAS LNSRVASISL
S. griseus DLQSSSAVKV RSTKVLQAPG YN..GTGKDW ALIKLAQP.. ....INQPTL
* i ** *** * **** ** ** i * *
151 200
C. carbonum AASGSDPASG STTS.VAGWG ATREGGGANN .ALLKVSVPI VARSTCVSNY
F. oxysporum AASGSDPVAG SSAT.VAGWG ATSEGGSSTP VNLLKVTVPI VSRATCRAQY
B. mori VETNSEIPDG DITI.VTGWG HMEEGGG.NP SVLQRVIVPK INEAACAEAY
B. taurus PTSCA..SAG TQCL.ISGWG NTKSSGTSYP DVLKCLKAPI LSDSSCKSAY
S. griseus KIAITTAYNQ GTFTGVAGWA NR.EGGSQQR Y.LLKANVPF VSDAACRSAY
* i * *itti *** **** *** t * i t
201 250
C. carbonum NAVGLTVTTN MVCAGV.TAG GRDSCQGDSG GPLVDANKT. ...LIGVVSW
F. oxysporum GTSA. .I‘I‘NQ MFCAGV.SSG GKDSCQGDSG GPIVDSSNT. . . .LIGAVSW
B. mori SPI.YAITPR MLCAGT.PEG GKDACQGDSG GPLV'HJCKK. . . .LAGIVSW
B. taurus PG...QITSN MFCAGY.LEG GKDSCQGDSG GPVV.CSGK. ...LQGIVSW
s. griseus GNE.LVANEE I.CAGYPDTG GVDTCQGDSG GPMFRKDNAD EWIQVGIVSW
* * *** * * ******** it * t * ***
251 290
C. carbonum GTGCARPNLP GVYSRVGTLR SFIDQNAS. . ..........
.F. oxysporum GNGCARPNYS GVYASVGALR SFIDTYA$.. ...... ....
B. mori GLGCARPEYP GVYTKVSALR EWVDENITNL RLKHILRRF$
B. taurus GSGCAQKNKP GVYTKVCNYV SWIKQTIASN $ .........
S. griseus GYGCARPGYP GVYTEVSTEA SAIASAARTL s... ......
* ****** * *t* i it i ** *
%Similarity %Identity
.F. oxysporum 80 57
B. mori 60 39
B. taurus 62 41
S. griseus 55 33

Figure 5. Comparison of the predicted amino acid sequences of ALP] and four related
trypsin-like proteases using PILEUP (Devereux et al. 1984). Sequence references:
Fusarium oxysporum, SwissProt P3 5049, Rypniewski et al. (1993); Bombyx mori, PIR
832794, Ikeda et al. (1991); Bos taurus, PIR A90164, Walsh and Neurath (1964);
Streptomyces griseus, PIR JQ1302, Olafson et al. (1975). Amino acids conserved
between Alpl and at least two of the other proteins are indicated by asterisks. Putative
signal peptides were not included in the PILEUP analysis. The sequence for B. taurus is
trypsinogen. The mature N-terminus of Alpla is indicated by # and the stop codons by $.

24

AYTTQSSAPWGLARISSQXRGTTGYXXDD, where X indicates an unknown amino
acid. Analysis of this sequence by TBLASTN (Gish and States 1993) showed that it has
strong similarity to numerous serine proteases, including subtilisin-like proteases from the
firngi T richoderma album (GenBank M54901) (69% identity), Paecilomyces lilacinus
(GenBank L29262) (57% identity) and Beauveria bassiana (GenBank U163 05) (58%
identity). We are currently testing the hypothesis that this protein is Alp2 by cloning and

disrupting the corresponding gene.

T ransformation—mediated gene disruption of ALP] .

Plasmid pJM9 containing an internal 250-bp HindIII/Sall fiagment of ALP] (Figures 3
and 6A) plus the gene for hygromycin resistance was linearized at a unique BamI-II
restriction site and transformed into C. carbonum wild type strain 367-2A. Two single-
spored, hygromycin-resistant transformants (T515-1A and T515-3 A) were analyzed. The
restriction map of the wild type ALP] locus and the predicted map resulting from
integration of a single copy of pJM9 at ALP] are shown in Figures 6A and 6B,
respectively. The pattern of hybridization of pC8-6.1 indicates that pJM9 has integrated at
ALP] in single and multiple copies in T515-1A and T515-3A, respectively (Figure 6C).

Both ALP] mutants were analyzed for their protease profiles and pathogenicity. Total
extracellular protease activity was reduced by 35 to 45% in the ALP] mutants. Growth of
the mutants on 1% collagen was not significantly altered based on appearance or yield of
DNA and total extracellular protein (data not shown). Therefore, we conclude that ALP]

by itself is not required for growth in vitro on collagen. Activities of other extracellular

25

 

 

 

 

 

 

Figure 6. A, Restriction map of wild type locus of ALP] showing the location of the
ALP] transcript. B, Predicted restriction map of ALP] with insertion of a single copy of
pJM9. Predicted EcoRI and Sphl fiagment sizes are indicated (in kb). C, DNA blot
comparing wild type (367-2A) and two transformants (T515-1A and T515-3A). Isolated
DNA was cut with SphI (lanes 1 to 3) or EcoRI (lanes 4 to 6) and the blot was probed
with a cDNA copy of ALP] (C8-6.1). The extra bands in the digests of T5 1 5-1A are as
predicted for tandem multiple insertions of pJM9. The shaded boxes indicate ALP]
sequences. RV, EcoRV; RI, EcoRI; X, XbaI; B, BamHI; Sp, SphI; H, HindIII; S, SalI.
Not all sites are shown. Additional sites within ALP] are shown in Figure 4.

26

enzymes of C. carbonum (cellulase, endo-1,4-xylanase, endo-polygalacturonase, 1,3-1,4-
glucanase and exo-l,3-glucanase - Scott-Craig et al. 1990; Apel et al. 1993; Sposato et al.
1995; unpublished results) were unaltered in the ALP] mutant, indicating that ALP] is
also not required for functional processing of these enzymes.

Protease activities from wild type and ALP] mutant strains were purified as described
above through cation exchange HPLC. The ALP] mutants lacked two peaks of activity
and UV absorption corresponding to Alpla and Alplb (shown for mutant T515-1A in
Figure 7B). Taking into account that (l) Alplb is ca. 5 kDa larger than Alpla, (2) Alplb
but not Alpla is glycosylated, (3) Alpla and Alplb are similarly inhibited by aprotinin and
leupeptin, (4) the product of ALP] has one predicted glycosylation site, and (5) disruption
of ALP] results in the disappearance of both Alpla and Alplb, we conclude that Alpla
and Alplb are products of the same gene, ALP]. Since Alp2 is not afi‘ected by disrupting
ALP] (Figure 7), we conclude that it is the product of another gene.

Pathogenicity of T5 1 5-1A and T515-3A were compared to 367-2A on both resistant
(cv. Great Lakes) and susceptible (Pr x K61) cultivars of maize in the greenhouse. Rate of
lesion development, lesion size, and lesion morphology were examined daily for 14 d, at
which point the plants had been killed by both the wild type and the ALP] mutants. No
differences in lesion morphology, size, color, or rate of formation between the wild type
and the two mutants were observed (data not shown). Leaves with both low and high
lesion densities were observed. The fungus was reisolated fiom the maize leaves and

tested for hygromycin sensitivity. Twenty-six isolates fi'om wild type lesions were all

27

 

 

 

 

 

 

500 -- Alp2 —~ 1
400 ~—

250 A 0.8
° -. 0.7
fl Alp2 Al 1.
200 ii p _ 0.6
150 —
100 ~
A
:2
50 ~ g
o 4 3*
'S
'4:
600 1.2
B. 3:
0
m
a
8
6
I-
an

300 *

Absorbance, 280 nm (mAU)

200 a
100 “

 

 

 

 

Retention Time (min)

Figure 7. Cation exchange HPLC analysis of proteins fi'om culture filtrates of (A) wild
type and (B) ALP] mutant T515-1A. Prior to HPLC analysis, crude culture filtrates were
passed over a low-pressure anion-exchange column to remove acidic proteins and
pigments. Solid lines: OD280; dashed lines: protease activity.

28

hygromycin-sensitive, 50 isolates fi'om T515-1A lesions were all resistant to hygromycin,
and 35 isolates from T515-3A lesions were all resistant to hygromycin. Thus we conclude
that the pathogenicity of T5 1 5-1A and T515-3A was not due to restoration of ALP]
activity by simple excision of pJM9.

We conclude that ALP] is neither an essential pathogenicity factor nor a major
virulence factor for C. carbonum race 1. Since this pathogenicity assay is not quantitative,
we cannot exclude the possibility that ALP] makes a small contribution to
virulence. Since C. carbonum makes at least one additional extracellular protease, it
cannot be concluded from this study that proteases have no role in pathogenicity or
virulence. A similar conclusion was drawn from studies of the role of ALP, encoding an
alkaline protease, in murine respiratory mycosis caused by Aspergillusfiimigatus; ALP
mutants were still firlly pathogenic but residual protease activity remained (Monod et al.

1993; Tang et al. 1993).

Materials and Methods
F unga] culture growth and maintenance

Conidia of C. carbonum race 1, strain 367-2A, were stored at -80°C in 25% glycerol

and grown on V-8 juice agar plates. For enzyme production, approximately 5 x 105
spores were inoculated into a l-L Erlenmeyer flask containing 200 ml of mineral salts
supplemented with 0.1% yeast extract and trace elements (van Hoof et al. 1991).
Substrate supplements were: Type I collagen (Sigma C-9879), casein (Sigma C-03 76),

maize cell walls (Sposato et al. 1995), Type A gelatin (Sigma G-2625), bovine serum

29

albumin (Sigma A-7906), corn bran (Country Life Natural Foods, Pullman, Michigan),
and pectin (Sigma P-9l35). For routine protease production, the fimgus was grown on
1.0% collagen. Cultures were incubated at 21 to 23°C with shaking at 125 rpm for 3.5 to

4d.

Enzyme assays

Proteases were assayed using azocasein (Sigma A-2765) (Ansari and Stevens 1983).
Azocasein (0.5 ml of a 25 mg/ml solution in 50 mM sodium phosphate, pH 7.5) was
incubated with 5 to 10 ul of enzyme fraction at 45°C for 30 min. To stop the enzyme
reaction, 20 pl of 50% (w/v) trichloroacetic acid was added and the solution vortexed
vigorously for ca. 5 sec. The sample was centrifirged for 5 min at 14,000 x g in a

microcentrifirge and the OD410 of 200 pl measured in an ELISA plate reader (Bio-Tek).
Protease inhibitors were from Sigma. Units of enzyme activity are defined as OD410 per

10 ul enzyme fraction under the conditions described.

Protein purification and characterization

Culture filtrates (typically 200 to 300 ml per batch) were concentrated to ca. 10% of
the original volume by rotary evaporation under vacuum at 37°C. After centrifugation to
remove insoluble material, the filtrates were dialyzed for 16 hr at 4°C in cellulose dialysis
tubing (SpectraPor, MWCO 12,000 to 14,000) against 25 mM sodium acetate, pH 5.0
and applied to a column (10 to 20 ml bed volume in a 60 cc disposable syringe) of DEAE

cellulose (Sigma) equilibrated in 25 mM sodium acetate, pH 5.0. The column was washed

30

with one bed volume of 25 mM sodium acetate, pH 5.0, and all material that was eluted
from the column was pooled. The samples were again concentrated by rotary evaporation
to ca. 5 ml and dialyzed against 25 mM sodium acetate, pH 5.0. After clarification by
centrifugation, the samples were fractionated on a cation exchange HPLC column
(polysulfoethylaspartamide, The Nest Group, Southboro, Massachusetts). Running
conditions were a linear gradient of bufi‘er A (25 mM sodium acetate, pH 5.0) to buffer B
(25 mM sodium acetate, pH 5.0, plus 0.4 M KCl) in 20 min at 1 ml/min. The peak of UV
(280 nm) absorbance corresponding to Alpla activity was collected and firrther purified by
hydrophobic interaction chromatography (Biogel TSK-Phenyl-SPW, Biorad, Richmond,
California) after adding ammonium sulfate to a final concentration of 1.7 M. Proteins

were eluted with a 20 min linear gradient of 0.1 M KH2P04, pH 7.0, plus 1.7 M

ammonium sulfate to water at a flowrate of 0.9 ml/min. Fractions containing Alpla were
desalted, lyophilized, and sequenced by automated Edman degradation. Alp2 was purified
using the same procedure as for Alpla through cation exchange HPLC, taking 0.2 ml
fi'actions (Figure 7), and then separated on SDS-PAGE. The proteins in the gel were
transferred to ProBlot (Applied Biosystems, Foster City, California) (Matsudaira 1987)
and the blot was stained briefly with 0.1% Coomassie R-250 in 40% methanol and
destained with 50% methanol. Three proteins were visible on the blot; all three were
excised and sequenced.

SDS-PAGE was carried out in 12% (w/v) acrylamide resolving gel with 5% (w/v)
stacking gels (Hames and Rickwood 1981). Glycoproteins were detected by periodic

acid/Schifi‘ staining (Stromqvist and Grufiinan 1992).

31

The pH optima for the purified enzymes was measured using bufi‘ers composed of 10
mM citric acid/20 mM sodium phosphate (pH range 5 to 7), 50 mM Tris-HCl (pH range 7
to 9), and 50 mM CAPS (3 -cyclohexylamino-l-propane-sulfonic acid)-HC1 (pH range 9 to
11).

Extensin was purified from maize stylar tissue as described (Murphy and Hood, 1993)
and incubated with Alpla, Alplb, and Alp2 alone and in combination for 60 min or
overnight at 45°C. Degradation was evaluated by SDS-PAGE. Casein was used as a

control.

Nucleic acid manipulations and sequencing

DNA was isolated as described by Pitkin et al. (1996) and RNA was isolated as
described by Chomczynski and Sacchi (1987). The transcription start site of ALP] was
determined using the Amplifinder RACE Kit (Clontech, Palo Alto, California) (Frohman et
al. 1988). Reverse transcription was primed with the oligonucleotide
CGTCGCTGTCCGCGTCGG (starting at nucleotide 608, see Figure 4). PCR
amplification was done using the primer sequence GTTGGTGGAACCACCGCT
GCTGCCG (starting at nucleotide 450, see Figure 4) and the "anchor" primer supplied
with the RACE Kit. ALP] was sequenced using specific oligonucleotides spaced ca. 250-
bp apart. Sequencing was performed by automated fluorescent sequencing at the MSU-
DOE-PRL Plant Biochemistry Facility using an Applied Biosystems (Foster City,
California) Catalyst 800 for Taq cycle sequencing and an Applied Biosystems 373A

Sequencer for analysis of the products.

32

Isolation of ALP]

PCR amplification of the 330-bp 5' region of ALP] was performed as follows: 1x PCR
reaction bufl‘er (Gibco-BRL, Gaithersburg, Maryland); 0.15 mM each of dATP, dCTP,
dATP, and dGTP; 1.5 mM MgC12;2 U T aq Polymerase (Gibco-BRL); 100 uM

oligonucleotide primer; 50 ng DNA isolated from phage lysate of a cDNA library prepared

fiom polyA+-RNA extracted from C. carbonum grown on maize cell walls (Pitkin et al.
1996). PCR reactions were performed in a Perkin-Elmer thennocycler model 480 under
the following conditions: one min denaturation at 94°C; 35 cyles of 1 min denaturation at
94°C, 2 min annealing at 55°C, and 3 min primer extension at 72°C; 7 min primer
extension at 72°C. PCR primers used were GAGTAYCCNTTYATHGT and
GTNGCNGGNTGGGGNGC (N=any nucleotide; Y=T or C; H=A,T, or C) corresponding
to the amino acid sequences EYPFIV and VAGWGA, respectively (Figure 4). After one
extraction with one volume chloroform, PCR products were precipitated with two
volumes ethanol plus one-tenth volume 3 M sodium acetate, pH 5.2, by incubating for 10
min at 20°C and centrifirging 10 min in a microcentrifirge (14,000 x g). The resulting
pellet was dried, redissolved in water, treated with T4 DNA Polymerase (Boehringer-
Mannheim, Indianapolis, Indiana) in the presence of 1 mM dNTP's for 5 min at 37°C, and
fractionated on a 1.0% agarose gel in TAE bufi‘er (Maniatis et al. 1982). The unique
product corresponding to ALP] was excised from the gel and purified using Gene-Clean
(Bio 101, Vista, California). The blunt-ended ALP] PCR product was ligated into

pBluescript H SK+ at the SmaI site and sequenced.

33

Screening of the cDNA and genomic libraries, DNA blotting, probe labelling, and
hybridization have been described (Scott-Craig et al. 1990; Sposato et al. 1995). DNA
blotting was done with Nytran (Schleicher & Schuell, Keene, New Hampshire) and
hybridizations were done in in 5x SSPE (Maniatis et al. 1982), 7% SDS, and 0.5% non-fat
dry milk at 65°C for 16 hr. Blots were washed twice at 22°C with 2x SSPE and 0.1%
SDS for 15 min each time, and twice at 65°C in 0.1x SSPE and 0.1% SDS for 15 min

each time.

Disruption of ALP]

The transformation vector was constructed by cloning the 2.5-kb SaII/HindIII
fragment of pHYGl (Apel et al. 1993) containing the C. heterostrophus promoter 1
driving the expression of the hph gene encoding hygromycin phosphotransferase (Schafer
et al. 1989) into the AatII site of pUCl 19 to create pHYG3. Concurrently, the plasmid
harboring the ALP] cDNA clone, pC8-6.1, was digested with HindIII, treated with T4
polymerase, and digested with SalI. The resulting 290-bp internal fragment of ALP]
(Figure 4) was ligated into pHYG3 that had been digested with SmaI and SalI to create
pJM9. This vector was linearized at the unique BamHI site located within the 290-bp
ALP] sequence (Figure 4) and used to transform strain 367-2A of C. carbonum.

Preparation and transformation of protoplasts was as described (Scott-Craig et al.
1990; Apel et al. 1993). Transfonnants able to grow on V-8 juice agar containing 100
units/ml hygromycin (Calbiochem) were single-spored twice to obtain nuclear

homogeneity. Pathogenicity was evaluated by inoculating leaves of 2-week-old maize

34

cultivars Pr x K61 (susceptible) and Great Lakes (resistant) with 104 conidia/rnl in 0.1%
Tween-20. The whole plants as well as different leaves with different infection densities
were observed daily for two weeks. Alter one week, individual lesions were excised and
after surface sterilization in 10% (v/v) commercial bleach plus 0.1% Tween-20 plated on
V-8 juice agar. Afier two days growth, the fungi were transferred to V-8 juice agar plates

containing hygromycin and evaluted for resistance to hygromycin.

CHAPTER 2

Jenifer M. Gérlach, Esther van der Knaap, and Jonathan D. Walton. 1997. Three
rnixed-linked glucanases from the filamentous fungus Cochliobolus carbonum: cloning
and targeted disruption of MLG]. Eur. J. Biochem. (submitted).

Chapter 2
THREE MIXED-LINKED GLUCANASES FROM THE FHAMENTOUS

FUN GUS COCHLIOBOL US CARBONUM‘ CLONING AND TARGETED
DISRUPTION OF MG]

Abstract

Three extracellular enzymes (MLGases) able to hydrolyze B-l,3-l,4-glucans (mixed-
linked glucans or B-glucans) were purified and characterized from culture filtrates of the
plant pathogenic firngus Cochliobolus carbonum. Total MLGase activity was stimulated
by substrates containing mixed-linked glucans and by 0.2% sucrose and was partially
repressed by 2.0% (w/v) sucrose. Maize bran induced higher total MLGase activity in
culture than oat bran. The three MLGases, called Mlgla, Mlglb, and Mlg2, were
resolved by cation exchange and hydrophobic-interaction HPLC. Mlgla and Mlglb also
hydrolyze B-1,3-glucan, whereas Mlg2 does not degrade B-l,3-glucan but does degrade
B-l,4-glucan to a slight extent. Mlgla, Nflglb, and Mlg2 have monomer molecular
masses of 33.5-kD, 31-kD, and 29.5-kD, respectively. The N-terminal amino acid
sequences of Mlgla and Mlglb are identical (AAYNLI). Mlgla is glycosylated whereas
Mlglb is not. The gene encoding Mlglb, MLG], was isolated using PCR primers based
on amino acid sequences of Nflglb obtained from the purified protein. The product of

MG] has no close similarity to any known protein but does contain a motif (EIDI)

35

36

that occurs at the active site of MLGases fi'om several prokaryotes. An internal fiagment
of MG] was used to create mlg] mutants by transformation-mediated gene disruption.
Total MLGase and B-l,3-glucanase activities in culture filtrates of the mutants are reduced
by approximately 50% and 40%, respectively. When analyzed by cation exchange HPLC,
the mutants are missing the two peaks of mixed-linked glucanase activity corresponding to
Mlgla and Mlglb. Together, the data indicate that Mlgla and Mlglb are products of the
same gene, MLG] . Growth of mlg] in culture medium supplemented with macerated

maize cell walls or maize bran and disease symptoms on maize are identical to wild type.

Introduction

Monocot cell walls are composed of a variety of macromolecules including cellulose,
arabinoxylan, xyloglucan, pectin, and proteins. One of the major hemicelluloses of the
walls of plants in the Poaceae is mixed-linked glucan (also called B-l,3-1,4-glucan or B-
glucan), in which unbranched chains of B-l,4-glucose are disrupted by periodic B-1,3-
linkages in the ratio of about 2: 1. Several lines of evidence suggest that this
polysaccharide is particularly important for the control of plant cell expansion, and
therefore it might have a critical role in maintaining the structural integrity of the wall
(Carpita, 1996).

Enzymes that can degrade mixed-linked glucan are called mixed-linked glucanases
(here abbreviated MLGase), B-1,3-1,4-glucanases, B-glucanases, or lichenases. Some
MLGases can also degrade other glucans, for example, B-1,3-glucans and B-1,4-glucans

(Sakellaris et al., 1993, Schimming et al., 1992, Spilliaert et al., 1994, Haj et al., 1989).

37

Genes encoding MLGases have been cloned from a number of bacteria (e. g., Schimming
et al., 1992; Teather and Erfle, 1990; Spilliaert et al., 1994) and higher plants (e.g.,
Slakeski et al., 1990; Yun et al., 1993). An MLGase has been purified from the fungus
Rhizopus arrhizus (Clark et al., 1978) but to our knowledge no genes encoding MLGases
have previously been isolated from fungi.

Cochliobolus carbonum, an ascomycetous pathogen of maize, penetrates into and
ramifies through intact leaves, in the process obtaining nutrients for its grth from the
plant cell cytoplasm and walls. For penetration, ramification, and nutrient assimilation,
both as a pathogen and during the saprophytic phase of its life cycle, C. carbonum
produces a variety of extracellular enzymes, including pectinases, xylanases, B-1,3-
glucanases, cellulases, B-xylosidase, or-arabinosidase, and proteases, that can degrade the
polymers of the plant cell wall.

A common feature of the extracellular degradative enzymes of microorganisms is
redundancy, that is, most microorganisms make two or more chromatographically
separable proteins that have the same or similar enzymatic activities. C. carbonum is no
exception to this rule, making, for example, at least four endo-B-l,4-xylanases (Apel-
Birkhold et al., 1996) and three proteases (Murphy and Walton, 1996). Enzymatic
redundancy can be due to multiple genes encoding proteins with similar or overlapping
enzymatic activities, differential RNA processing (e. g., Boel et al., 1984), and/or
differential post-translational modification. Apparent redundancy can also be caused by
artifacts such as proteolytic nicking during fermentation or purification. Although it is

difficult to establish the ontogenetic relationships between isozymes by purely biochemical

38

methods, it is feasible to establish the relationship between isozymes, and between
isozymes and their encoding genes, by comparing the enzymatic profiles of wild type
strains and strains that have been specifically mutated. Targeted mutation of genes is
facile in prokaryotes, but due to the greater technical difficulties of molecular genetic
manipulation of eukaryotes such as firngi much less is known about the relationship
between the multiple forms of extracellular degradative enzymes and their genes in these
organisms.

In this study, we describe the identification and characterization of three extracellular
enzymes that degrade B-glucan fiom the filamentous fungus C. carbonum and
demonstrate by cloning and targeted gene disruption experiments that one gene encodes
two of the three MLGases. The two MLGases that are derived from the same gene difl‘er

in glycosylation.

Results and Discussion
Characterization and purification ofMlgla, Mlgl b, and Mlg2

Several carbon sources were tested for optimal production of extracellular MLGase by
C. carbonum. Maize bran is a better inducer than two commercial oat bran products (data
not shown). Similar to endopolygalacturanase, exo-B-1,3-glucanase, xylanase, B-
xylosidase, or-arabinosidase, and protease, adding 0.2% sucrose enhances production of
MLGase. Unlike xylanase, exo-B-1,3-glucanase, or endopolygalacturanase, however,

some MLGase activity is still produced when C. carbonum is grown on 2% (w/v) sucrose

39

as sole carbon source (data not shown) (Holden and Walton, 1992; Murphy and Walton,
1996; Ransom and Walton, 1997; van Hoof et al., 1991; Walton and Cervone, 1990).

After concentration by rotary evaporation, dialysis, and passage through an anion-
exchange column to remove acidic proteins and pigments, MLGase activities were
fiactionated by cation-exchange HPLC. One major (Peak 1) and one minor (Peak 2) peak
of activity were resolved (Figure 8A). Both peaks were then separately applied to HI-
HPLC for firrther purification (Figs. 8B and 8C). Cation-exchange HPLC peak 1 (Figure
8A) was thereby resolved into two peaks of activity, called Mlgla and Mlglb (Figure 8B).
Peak 2 (Figure 8A) remained as a single peak of MLGase activity, called Mlg2 (Figure
8C). Mlgla and Mlglb were subsequently chromatographed by gel filtration to purify
them to electrophoretic homogeneity.

The molecular masses of Mlgla, Mlglb, and Mlg2, as determined by SDS-PAGE, are
33.5, 31, and 29.5-kD, respectively. Mlgla and Mlglb are endo-acting enzymes as
determined by their ability to rapidly reduce the viscosity of a B-glucan solution relative to
the simultaneous appearance of reducing sugars (data not shown). The temperature and
pH optima for all three enzymes are approximately 55°C and 5.0, respectively. The
activity of Mlgla and Mlglb against B-l,3-glucan is comparable to that against B-glucan.
Neither Mgla nor Mlglb has activity in long-term assays (17 h) against any of the B-1,4-
glucan substrates tested. Based on their HI-HPLC retention times and their ability to
degrade B-l,3-glucan, Mlgla and Mlglb are probably responsible for the two peaks of B-
1,3-g1ucanase activity remaining in culture filtrates of EXG] (exo-B-1,3-glucanase)

mutants of C. carbonum (Schaefi‘er et al., 1994). Mlg2 has no detectable activity against

 

 

 

 

 

 

 

 

 

 

1600 A 800
"3 Peakl
1200-» ’ .- 600 1}?
i ' 1..
1 ‘ 01-4
A 800" '1 \‘ PeakZ ,_ 400 5
D 400-- . {"1 200 v
E 0 “_ eras-“9’51 it“??? it“. 1 1 1 ’ 0 b
v 17 21 25 29 33 37 41 45 r;
E 60 Ml 1b 250 *3
.1}- g

50 B Mlgla -200 <

O 40» a 1“ ,_
'1 1‘ 150 Q)
00 301- §\ ,1, m
N 20 .. i z '1 9‘ “ 100 :6

cs \

o 10 -. I \ g " 50 g
Q 1 . u. . - - M A .e 1 o
G 0 vvvvvvvvvvvvvvvvvvv . . . . . 0 fi
:3 20 22 24 2628 30 32 3436 38 40 _.
"e 50 250 L?
8 40 -- C M132 «» 200 a.
'2 30-- -- 150 <1;
20« ~- 100 '7‘
10 I” _, 50 max
0‘ 0 "'1

25 27 29 31 33 35

 

 

Retention Time (min)

Figure 8. Purification of Mlgla, Mlglb, and Mlg2 (A) Cation exchange
chromatography of culture filtrates. (B) HI-HPLC of Peak 1 from (A). (C) HI-HPLC of
Peak 2 fiom (A). Solid lines, absorbance at 280 nm; dashed lines, MLGase activity.

41

B-1,3-glucan but in long (17h) incubations shows some ability to degrade several B-l,4-
glucans (low viscosity CM-cellulose, Whatman cellulose, and Avicel). Mlg2 is 180 to 340
times less active against these B-1,4-glucans than it is against B-glucan (data not shown).
In a 17 h incubation, Mlg2 shows no ability to degrade two other B-1,4-glucan substrates,
high viscosity CM-cellulose and a-cellulose.

On the basis of their substrate preferences, Mlgla and Mlglb can be considered as
bifunctional B-1,3-1,4/ -1,3-glucanases and Mlg2 as a B-l,3-l,4-glucanase.

Mlgla is glycosylated whereas Mlglb is not (Figure 9). Glycosylation can alter the pH
optima, temperature optima, or thermostability of enzymes (Doan and Fincher, 1992;
Meldgaard and Svendsen, 1994), but Mlgla and Mlglb have similar temperature optima
and thermostability (data not shown).

At least the first five amino acids of the mature Mlgla and Mlglb proteins are identical
(Table 2). Analysis of the N-terminal sequence of Mlglb (22 amino acids) by BLASTP
(Gish and States, 1993) found no strong similiarity to any sequence in the nonredundant
databases. Two internal tryptic peptides of Mlglb were sequenced. One (peptide 3)
overlaps with the N-terminal peptide (Table 2). Peptide 2 fiom Mlglb has 77% identity to
an MLGase from the prokaryote Rhodothermus marinas (PIR 848201).

For the final purification of Mlg2, proteins in the I-II-HPLC fiactions containing
MLGase activity were separated by SDS-PAGE, blotted, and Mlg2 was excised fi'om the
blot for sequencing (Matsudaira, 1987). As the N—terminus was blocked, internal amino
acid sequences were obtained from three tryptic peptides (Table 2). BLASTP analysis

indicates that peptide 2 of Mlg2 (Table 2) is 56% identical to a cellulase, Eng, from the

42

 

A B
| :2 al.! a
B 9 L59 9
kg 2 2 s s s

 

Figure 9. Glycosylation of Mlgla and Mlglb. (A) Protein blot from SDS-PAGE gel of
HI-HPLC purified Mlgla and Mlglb stained with periodic acid/Schifi‘ reagent (Stromqvist
and Gruflinan, 1992). (B) SDS-PAGE of the same samples in (A) stained with Coomassie
R-250. Mlgla and Mlglb are the bands at 33.5 and 31 kD, respectively. Molecular
weight standards are shown on the lefi; ovalbumin (45 kDa) is a glycoprotein.

43

TABLE 2. Experimentally determined amino acid sequences from Mlgla, Mlglb, and
Mlg2, and comparison to the sequences deduced fiom the nucleotide sequence of MLGl.

 

Protein Peptide sequence

 

Experimental Deducedfi'om DNA Sequence

 

Mlgla 1. AAYNLI l. AAYNLI

Mglb l. AAYNLIDTYDA (A) N (W) AAKFNFED l. AAYNLIDTYDASNWASKFNFED
2. GPNWP (A) QGE‘. (I) D (I) 2. GPNWPnQGEIDI
3. FNFEDIADPDT 3. FNFEDIADPT

Mlg2 1. FTVNQ (C) SANAY
2. YDVYPIGSSQGMVNVAGR
3. GFPINSQNLITYQFGTEAFTGGP

 

- Purified proteins were sequenced at the N-terminus (Mlgla peptide 1 and Mlglb peptide
1) or fiom tryptic peptides (all of the others).

- Amino acids in parentheses indicate uncertainty.

- Lower case letters in the deduced sequence indicate discrepancies between the
experimental and the deduced sequences.

prokaryote Streptomyces rochei (GenBank X73953) and peptide 3 is 78% identical to the
F1 CM-cellulase of Aspergillus acuIeatus (PIR S 12610). Eng (Perito et al., 1994) and
F1 CM-cellulase (Ooi et al., 1990) are members of cellulase family H (Gilkes et al., 1991)
or glycosyl hydrolase family 12 (Heniissat and Bairoch, 1993). The cloning and

sequencing of the gene for Mlg2 is in progress.

Isolation and characterization of MLGI
Two 96-fold degenerate oligonucleotides based on the amino acid sequences IDTYDA

and QGEIDI (Table 2) were used in PCR to amplify a fragment of the encoding gene.

44

This primer combination yielded a single, 340-bp PCR product when DNA from a C.
carbonum cDNA library was used as a template and a single, 460-bp PCR product when
C. carbonum genomic DNA was used as a template. To confirm that these PCR products
encoded Mlglb, they were blotted and probed with a third 36-fold degenerate
oligonucleotide based on the amino acid sequence KFNFED (Table 2, Figure 10). Both
products hybridized to the third oligonucleotide and were therefore cloned and sequenced.
Sequencing indicated that the PCR products are identical except for the presence of two
introns of 57- and 64-bp in the PCR product amplified from genomic DNA (see Figure
10).

Using the cDNA-derived PCR fragment as a probe, C. carbonum cDNA and
genomic libraries were screened. From the cDNA library, a 1.34-kb A/ILGI cDNA (C4-
2.1) was isolated and sequenced. A 7.0-kb BamHI fi'agment (see Figure 12) of DNA
(MLGl-ZB) containing the MLGI genomic locus was subcloned and sequenced on both
strands. Figure 10 shows the sequence of MG] and the deduced amino acid sequence.
The transcription start site of MLGI was determined by analyzing the sequence of three
independent RACE products (Frohman et al., 1988). The MLGI transcript has a 64-bp 5'
untranslated region. The context of the deduced translation start site (CACTCAlQTCT,
Figure 10) conforms with the consensus sequence for Neurospora crassa translation
initiation (CAMMA_T_GGCT where M = C or A) (Edelrnann and Staben, 1994). Three
introns (Figure 10) were identified by comparing the sequences of the cDNA and the
RACE products with the genomic clone. The 5’ and 3’ splice sites, splice branch sites,

and lengths of the introns are consistent with introns in N. crassa and in other genes of C.

45

l TGCTTTCAACGTACATGTCCATTGGGGATGGGCGGGTGGAGCAGTGCAGCAGACCTGGAT

61 GGGTCTGCGAGGAGGTTGTGGTGGGAGTGGGGCGCGCATAGGGAGGGTTTGGGTGTGTGA
121 TAAGGAGGGTTGTGAAATGTATAAGTTGGATAGATGTGACCGTTTTTTCAGGATTGAGGT

I

181 TTCTTGCTATCATCTGTTTGGTTGCATTTCCTGCACTTTTAGTTACTTGGATACGAATAT
241 CTGCCACTCATGTCTCTCAAGTCCCTCTTCGTTTCGGCACCGTTGCTTTGGCGCGGTGTT
M S L K S L F V S A P L L W R G V

301 ACCGCACTCCCTGCTGGCCCTGGTTCCTGGACACATGGAAACTCGACGATTGTGTCTTCT
18 T A L P A G P G S W T H G N S T I V S S
361 TCGGACTTCAGTGCTGCGGCGGCATACAACCTGATTGATACGTATGACGCTAGTAACTGG
38 S D F S A A A A Y N L I D T Y D A S N W
421 GCGAGCAAGTTCAACTTTGAAGATATCGCTGATCCCACGCgtatgtatcacttcactcta

 

 

58 A S K F N F E D I A D P T SalI
481 gcgtctgtgagttttgatttctgacctgatgatgcagACGGCTTCGTCGACTATGTCACT
71 H G F V D Y V T

541 TTGCAACAGGCGCAACAATATGGACTATTTAAGACGCAGAACAATCAGGTATACATGGGC
79 L Q Q A Q Q Y G L F K T Q N N Q V Y M G
SmaI
601 GTAGACTCCACCTCCACTCTCAACCCCAACGGCCCGGGAAGAAGGAGTGTCAGAATACAA
99 V D S T S T L N P N G P G R R S V R I Q
661 AGCAAGACAGCGTACAACCGCGCTCTCGTCATTGCAGACTTTGCCCATGTTCCTGGTAGT
119 S K T A Y N R A L V I A D F A H V P G S
721 GCATGTGGCTCTTGGCCCGCCTTgtatgtctccttttccacatccctcatcccatcccac
139 A C G S W P A F
781 acctcgctaacactcccttccgtccagCTGGATGGTCGGTCCCAACTGGCCTAACCAAGG
147 W M V G P N W P N Q G
XhoI
841 CGAAATCGATATCTACGAAGGCGTCCACCTCTCGAGCTCCAACCAAGTAACCCTGCATAC
158 B I D I Y E G V H L S S S N Q V T L H T
901 ATCCCCCGGTTGCAATCCCTCCATCGGCCCCGGCGGAGAAACCGGACGTCGTCTCGCAGG
178 S P G C N P S I G P G G E T G R R L A G
961 CGACTGCGGCGCCGACGGTGGCTTCAACGGCTGCGGCATCCAAGCCGACAACCCCGTCTC
198 D C G A D G G F N G C G I Q A D N P V S
1021 GTTCGGCACGCCCTTCAACGCCAACGGCGGCGGCGTCTACGCCACCCTCTGGACCAGCTC
218 F G T P F N A N G G G V Y A T L W T S S
1081 CGGCGTCAAAGTCTGGTACTTTGCAACTCGCAACATCCCTGCCAACATCAAGTCCGGGAA
238 G V K V W Y P A T R N I P A N I K S G N
1141 CCCAGACCCCTCGGCTTGGGGCACCCCGATTGCGAATTTCGGAAACAATGGATGCGACTT
258 P D P S A W G T P I A N F G N N G C D F
1201 CGATGCCAAGTTTCGCGACTTGAATATCGTGTTCGATGTTACGTTTTGTGGCGATTGGGC
278 D A K F R D L N I V F D V T F C G D W A
1261 GGGAGGGGTTTGGGGGTCCACGACTCGTGCGCAGGTTAATCCAAGTTGTGTGGCGTATGT
298 G G V W G S T T R A Q V N P S C V A Y V
1321 TGCGAGTCAGCCGCAGAACTTTTCGGAGgtgagttgcgttgtgaagtatttgaaaaggga
318 A S Q P Q N F S E
1381 attgttatgctaactttacatagTCGTACTGGCTCATCAACTCGGTCAAGGTCTACAGTG

 

327 S Y W L I N S V K V Y S
1441 TTTAGGCTTGAGAGTTATTGTTAGATGCTTCGAGATATCGTGTGCATCTGTTTATGCACA
339 V *

1501 CATAATATCTCGTACGCGTCGCTCTTTGTTTTTCGTTTCTCTGACTATATTCTTCTTCTC

1561 TTTTTTCTATGTATTTACTTCTTGGATGCAGGGTGTGATGTGCTAGAGGAGAGTCACATC

1621 ACATATGTTGCCCGATATCTACTTTGTATTTCCTATCCACACTCTATTCCCTTGTACAAC
+

1681 CCGCTTCATATACCCTCTATTCTCCTTTATAGAATCTATTGCTCATTCACCATCTTTTCC
1741 AGCGCCTCAACCTCACTCTTTTGCCACTCCGTGACCTTGCCCTGCCACTTTTCGCCCTTG

Figure 10. Nucleotide sequence and deduced amino acid sequence of AJLGI. Amino
acids are placed below the corresponding codons. The amino acid sequences of the
mature N-terminus and of the internal tryptic peptide are indicated by double underlining.
The three introns are indicated in lower case letters. The indicated SaII, SmaI, and XhoI
restriction sites were used to construct and linearize pJMS for the gene disruption
experiments. The transcription start site is indicated by # and the polyadenylation site by +
(symbols refer to the nucleotides below). The predicted N-glycosylation site (amino acid
323) is single-underlined. The nucleotide sequence of MLGI has been deposited in
GenBank with accession number U81606.

46

carbonum and N. crassa (Apel et al., 1993; Apel-Birkhold and Walton, 1996; Murphy and
Walton, 1996; Scott-Craig et al., 1990; Sposato et al., 1995; Pitkin et al., 1996; Edelrnann
and Staben, 1994). Like other C. carbonum genes, no obvious AATAAA polyadenlyation
signal sequence precedes the polyadenylation site (Gurr et al., 1987).

The open reading frame of the cDNA C4-2.1 is predicted to encode a mature protein
of 3 1 .8-kD which is in good agreement with the size of Nflglb, 31-kD, as determined by
SDS-PAGE. The predicted pI of the mature protein is 4.99. The program SignalP v1.1
(Nielsen et al., 1997) predicts a signal peptide cleavage site between amino acids 19 and
20 (VTA/LPA); if this is the true cleavage site then Mlgl must undergo additional
processing to generate the mature protein. One predicted N-glycosylation site (NF S),
which therefore probably accounts for the difl‘erential glycosylation of Mlgla and Mlglb,
occurs near the C-terminal end (Figure 10).

The predicted amino acid sequence of Mlglb has little overall sequence similarity to
that of any other gene in the nonredundant databases. The best matches are to two
prokaryotic glucanases, a B-1,3-glucanase from Oerskovia xanthineolytica (BLAST score
51, P=0.84) and an MLGase from R marinas (BLAST score 50, P=0.92) (Spilliaert et al.,
1994; F errer et al., 1996). The overall amino acid similarity and identity of Mlgl to the
MLGase of R. marinas are 50% and 22%, respectively. The longest stretch of identity is
a motif of five amino acids (GEIDI) surrounding a Glu residue which is at the active site
of MLGases fi'om Bacillus and other prokaryotes (Figure 11) (Keitel et al., 1993; Planas
et al., 1992; Spilliaert et al., 1994). This similarity to the bacterial MLGases classifies

Mlgll as a member of the family 16 glycosyl hydrolases (Henrissat and Bairoch, 1993).

47

1 50
C. carbonum ....... AAY NLI..DTYDA SNWASKFNFE DIADPTHGFV DYVTLQQAQQ
R. marinas SDRSDKAPHW ELVWSDEFDY SGLPDPEKW. DYDVGGHGWG N....QELQY
51 100
C. carbonum YGLFKTQNNQ VYMGVDSTST LNPNGPGRR. .SVRIQSKTA YNRALVIADF
R..marinus YTRARIENAR VGGGVLIIEA RHEPYEGREY TSARLVTRGK ASWTYGRFEI
101 # 150
C. carbonum.AHVPGSACGS WPAFWMV... ....GPNWPN QGEIDIYEGV HLSSS.NQVT
R. marinus RARLPSGRGT WPAIWMLPDR QTYGSAYWPD NGEIDIMEHV GFNPDVVHGT
151 200
C. carbonum LHTSPGCNPS IGPGGETGRR LAGDCGADGG FNGCGIQADN PVSFGTPFNA
R. marinas VHTK.AYNHL LGTQRGGSIR VP...TARTD FH ..... ... ..........
201 250
C. carbonum NGGGVYATLW TSSGVKVWYF ATRNIPANIK SGNPDPSAWG TPIANFGNNG
R. marinus ....VYAIEW TPEE..IRWF VDDSLYYRFP NER.....LT DPEADWRHWP
251 300
C. carbonum CDFDAKFRDL NIVFDVTFCG DWAGGVWGST TRAQVNPSCV.AYVASQPQNF
R. marinus FD ..... QPF HLIMNIAVGG AWGG...... ...QQGVDPE.AFPAQLVVDY
301 316
C. carbonum SESYWLINSV KVYSV*
R. marinas VRVYRWVE*. ......

Figure 11. Comparison of the amino acid sequences of C. carbonum Mlgl and

Rhodothermus marinas B-glucanase. Alignment was done using PILEUP (Program
Manual, 1994). The probable glutamic acid residue of the active site, deduced by
sequence comparison to the known active site of Bacillus MLGases (Keitel et al., 1993;
Planas et al., 1992), is indicated by #. Identical and similar amino acids are indicated by
colons and periods, respectively. Stop codons are indicated by *. Putative signal peptides
are not shown.

48

The sequence of MG] has no detectable similarity to any known plant MLGases nor to

any B-l,3-glucanases, including EXGI from C. carbonum (Schaefl‘er et al., 1994).

T ramformation—mediated gene disruption of MLGI

A 345-bp SaII/XhoI fragment, within the open reading frame of MG] (Figure 10),
was subcloned into the Cochliobolus transformation vector pHYGl (Sposato et al. 1995)
conferring hygromycin B resistance. The resulting plasmid, pJMS, was linearized with
SmaI and introduced into wild type C. carbonum strain 367-2A by transformation of
protoplasts. Hygromycin B-resistant transformants were characterized by DNA gel blot
analysis. Figure 12A depicts the wild type MLGI locus whereas Figure 12B shows the
predicted map for a single integration event of pJMS into AJLGI. The DNA gel blot
analysis of strain T503 -4A (Figure 12C, lanes 3 and 6) is consistent with the predicted
single insertion event shown in Figure 128. The pattern of hybridization seen for strain
T503-1A (Figure 12C, lanes 2 and 5) is consistent with a tandem integration event of the
6-kb pJMS vector. Total MLGase and B-1,3-glucanase activities in culture filtrates of
Mlglb mutants T503-1A and T503-4A grown on maize bran are reduced by
approximately 50 and 40%, respectively. However, growth (i.e., mycelial mat dry weight)
of T503-1A and T503-4A are similar to wild type (data not shown); apparently the
residual MLGase activity as well as other enzymes capable of degrading other substrates
in maize bran are sufficient to support normal growth of the mlgI mutants.

MLGase activities from wild type and the mlgI mutant T503-1A were purified in

parallel through HI-HPLC. When analyzed by cation exchange HPLC, the mlgl mutant is

49

 

r 7.0 . 1 In;
1—1

 

 

 

 

 

C ,——BamHI—..——Hlndm—-.
< < <
a. :6 n a. «'3 i
" S 8 1; S 3
kb 8 1- 1- n 1- 1-

    

5.0-

Figure 12. Analysis of the MG] locus in wild type and mIgI mutants. (A) Restriction
map of the wild type MG] locus showing the location of the MLGI transcript (shaded
box). (B) Predicted restriction map of the MG] locus with a single insertion of the
transforming plasmid pJMS. (C) DNA blot of wild type (367-2A) and two transformants
(T503-1A and T503 -4A). Total genomic DNA was digested with BamHI (lanes 1 to 3)
or HindIII (lanes 4 to 6), fractionated by agarose gel electrophoresis, blotted, and probed
with the MLGI cDNA (C4-2.1). The disappearance of a BamHI band of 7.0 kb and a
HindIII band of 5.0 kb, and the appearance of BamI-II bands of 9.7 kb and 2.5 kb and
HindIII bands of 2.1 kb and 8.5 kb, are as predicted from homologous integration of
pJMS. The additional bands in digests of T503-1A are as predicted for homologous
integration of more than one copy of pJM5. The shaded areas in (A) and (B) indicate
MG] sequences. B, BamI-H; S, SalI; P, PstI; X, XhoI; Sp, SphI; H, HindIII.

50

missing Peak 1; Peak 2, corresponding to Mlg2, is still present (Figs. 13A and 13B).
Thus, Mlg2 is not encoded by MLGI. HI-I-IPLC analysis (Figs. 13C and 13D) indicates
that both Mlgla and Mlglb are missing in the mlgl mutant (Figure 13D).

Because Mlgla and Mlglb have the same substrate specificities and the same N-
terminal amino acid sequences, and because mlg] mutants have neither Mlgla nor Mlglb
activity, we conclude that MLGI encodes both Mlgla and Mlglb. The different
chromatographic behavior of Mlgla and Mlglb is probably due to differential
glycosylation, because Mlgla is glycosylated whereas Mlglb is not, Mlgla is 2.5-kD
larger than Mlglb, and the MG] gene product has one predicted N-glycosylation site.
The two products of AJLGI are probably not due to difi‘erential intron splicing of the
MG] transcript (Boel et al., 1984) because all three introns of MLGI contain a stop

codon or frame shifi (Figure 10).

Pathogenicity of mlg] mutants

As the highest amounts of B-glucan are in young maize seedlings (Carpita, 1984), we
tested whether infection of young seedlings by C. carbonum was impeded by mutating
AJLGI . There were no measurable difi‘erences in lesion morphology and development or
percent of seedlings that germinated when inoculated with either wild type strain 367-2A
or mlgl mutant strains T503-1A or T503-4A. Wild type and mlgI mutants were also
indistinguishable in regard to lesion size, color, and rate of lesion formation when spray—
inoculated onto leaves of 14-day old maize seedlings. Thus, MG] does not by itself

make a significant contribution to the virulence of C. carbonum.

51

 

 

 

 

 

 

 

   

 

 

 

150 -_ ” 14 . 30
~ 10
100 r ._ 20

“”10

Absorbance, 280 nm (mAU)

   

 

 

50 A.—

 

 

  

00000000000000000000

01 1‘ i ii i 1 11-2

15 19 23 27 31 35 39 43 20 22 24 26 28 30 32 34 36 38 40

Retention Time (min)

 

 

0

1,3-1,4-B-glucanase Activity (Units)

Figure 13. HPLC analysis of MLGase from mlgI mutant. (A,B) cation exchange HPLC
analysis of MLGase fi'om wild type (A) and Mg] mutant T503-1A (B). (C,D) HI-HPLC
analysis of Peak 1 from cation exchange HPLC of wild type (C) and from cation exchange
HPLC of mIgI mutant (D). Solid lines, absorbance at 280 nm; dashed lines, MLGase
activity.

52

Materials and Methods
Fungal culture and maintenance

C. carbonum race 1 strain 367-2A, which is a progeny of strain SB] 11 (American
Type Culture Collection 90305) was grown on V8 juice agar plates. For MLGase
production, two fiingal plugs (5 m2) were inoculated into a 1000-ml Erlenmeyer flask
containing 125 ml mineral salts, 0.2% yeast extract, and trace elements (van Hoof et al.,
1991) and grown in still culture for 9 d at 21 to 23°C. Supplemental carbon sources
tested were Country Life maize bran (Country Life Natural Foods, Pullman, Michigan),
Mother’s Oat Bran cereal, and Quaker Oat Bran cereal (both from The Quaker Oats
Company, Chicago, Illinois). For routine enzyme production, cultures were grown on 1%

maize bran plus 0.2% sucrose.

Enzyme assays

Routine enzyme assays were done using a reducing sugar assay (Lever 1972) and
barley B-glucan (Sigma 66513) as substrate. Laminarin (Sigma L9634) was used to test
for B-l,3-glucanase activity and Avicel PH-101 (Fluka 11365), high viscosity CM-
cellulose (Sigma C5013), low viscosity CM-cellulose (Sigma C5678), a-cellulose (Sigma
C8002), and microgranular Whatman cellulose were used for 8-1,4-glucanase activities.
Assays were performed using 0.2% substrate, except laminarin which was used at 0.1%, in
50 mM sodium acetate buffer, pH 5.0, at 37°C for 30 min with 10 to 20 ul enzyme. When
cellulosic substrates were used the assay was performed for 17 h at 37°C. After heating

the reaction mixtures at 100°C for 10 min, 200 111 of each reaction was placed in a 96-well

53

microtiter plate, cooled to 22°C, and the absorbance at 410 nm read in an ELISA plate
reader (Bio-Tek). One unit of activity is defined as one nmol glucose released per pl
enzyme per min at 37°C.

Viscometric assays were performed with a number 200 tube viscometer and 0.5%
barley B-glucan in 50 mM sodium acetate, pH 5.0, at 37°C. Viscometry readings were

taken every 3 min for 20 min.

Protein purification

Concentration and purification of MLGase activities from culture filtrates through
low-pressure DEAE-cellulose chromatography and dialysis was by the method of Murphy
and Walton (1996) except that the 25 mM sodium acetate bufl‘er was adjusted to pH 4.0.
Fractionation on a polysulfylethyl aspartamide cation exchange HPLC column (The Nest
Group, Southboro, Massachusetts) was with a 30 min linear gradient fi'om buffer A (25
mM sodium acetate, pH 4.0) to buffer B (25 mM sodium acetate, pH 4.0, plus 0.4 M KCl)
at 1 ml/min. The peak of UV absorption (280 nm) containing Mlgla and Mlglb was
collected, adjusted to 1.7 M ammonium sulfate, and applied to a hydrophobic-interaction
HPLC (I-II-I-IPLC) column (Biogel TSK-Phenyl-SPW, BioRad, Richmond, California)
(Murphy and Walton, 1996). Fractions containing Mlgla and Mlglb activity were then
individually passed over a gel filtration HPLC column (Beckman Ultraspherogel
SEC3000, 7.5 x 300 mm). Purified Mlgla and Mlglb were lyophilized and sequenced
directly from the N-terminus, as well as afier digestion with trypsin and separation of

peptides by microbore HPLC, by automated Edman degradation. Mlg2 was purified using

54

the same methods as for Mlgla and Nflglb through HI-HPLC. The fractions containing
Mlg2 activity were then fractionated by SDS-PAGE (12% acrylamide), transferred to
ProBlot (Applied Biosystems, Foster City, California) (Matsudaira, 1987), stained with
0.1% Coomassie R—250 in 40% methanol, and destained with 50% methanol. Mg2 was
excised from the blot and digested with trypsin. Resulting peptides were separated by
HPLC and sequenced by automated Edman degradation.

Methods of SDS-PAGE and glycoprotein detection by periodic acid/Schifi‘ staining
were as described (Hames and Rickwood, 1981; Stromqvist and Grumnan, 1992).
Determination of pH optima for the three enzymes was as described (Murphy and Walton,

1996)

Nucleic acid manipulations

DNA and RNA were isolated as described by Pitkin et al. (1996) and Chomczynski
and Sacchi (1987), respectively. Genomic and cDNA library screening, probe labeling,
DNA blotting, and hybridization have been described (Scott-Craig et al., 1990; Murphy
and Walton, 1996). Sequencing with gene-specific primers was performed by automated
fluorescent sequencing at the MSU—DOE-PRL Plant Biochemistry Facility using an
Applied Biosystems 373A Sequencer for analysis of the products. The transcription start
site of MLGI was determined using the Amplifinder RACE kit (Clonetech, Palo Alto,
California) (F rohman et al., 1988). First strand cDNA synthesis was primed with the
reverse complement oligonucleotide GAAGGCGGGCCAAGAGCC (starting at

nucleotide 727, Figure 10). PCR primer CGTGCGTGGGATCAGCGATATCTTC

55

(reverse complement) (starting at nucleotide 43 9, Figure 10) and the “anchor” primer

provided with the RACE kit were used to amplify the 5’ end of the AJLGI transcript.

Cloning of [MLGI

PCR conditions used to amplify MLGI and the protocol used to clone the PCR
fragments were as described (Murphy and Walton, 1996). Template DNA that generated
the 340-bp product was DNA isolated from phage lysate of a cDNA library prepared from
mRNA from C. carbonum grown on maize cell walls (Pitkin et al., 1996). Total genomic
DNA was used as a template for the reaction that yielded the 460-bp product. The PCR
primers ATHGAYACNTAYGAYGC and ATRTCDATYTCNCCYTG (H = A,C, or T; Y
= C or T; N = any nucleotide; R = A or G; D = A,G, or T) corresponding to the sequences
IDTYDA and QGEIDI (Table 2 and Figure 10), respectively, were used at an annealing
temperature of 55°C for PCR amplification of a fragment of MG] . Oligonucleotide
sequence AARTTYAAYTTYGARGA, corresponding to amino acid sequence KFNFED
(Table 2, Figure 10), was end-labeled (Sambrook et al., 1989) and hybridized at 45°C to
the PCR products for confirmation that a fragment of MG! had been amplified. The
MG] PCR products were cloned into pBluescript II SK+ at the SmaI restriction site and
sequenced. A 7.0-kb BamI-II [MLGI genomic fragment from a EMBL3 phage that

hybridized to the MG] cDNA was subcloned into pBluescript II SK+.

56

Targeted gene disruption of A/ILGI

The transformation vector was made by digesting the MLGI cDNA clone, pC4-2. l,
with XhoI/SalI to liberate a 345-bp fragment internal to the MLGI locus (Figure 10),
treating the fi'agment with T4 DNA polymerase, and ligating it into the SmaI restriction
site of pHY G1 (Sposato et al., 1995). The resulting vector, pJMS, was linearized at the
unique SmaI restriction site (Figure 10) and used to transform wild type C. carbonum
strain 367-2A.

Protoplast isolation and transformation have been described (Scott-Craig et al., 1990;
Apel et al., 1993). Transforrnants were selected for their ability to grow on 100 units per
ml of hygromycin B (Calbiochem, La Jolla, California). Two rounds of single spores were
isolated to ensure nuclear homogeneity.

For pathogenicity tests on germinating young seedlings, ten seeds of susceptible maize
cultivar Pr (genotype hm/hm) and ten seeds of resistant cultivar Prl (genotype Hm/Hm)
were surface sterilized 10 min in 10% (v/v) commercial sodium hypochlorite (household
bleach), washed five times with water, imbibed with water for 17 h, and planted at a depth
of 2 cm in soil in 13-cm diameter clay pots. The pots were watered with 100 ml of 105
fresh conidia per ml. Germination and growth were monitored daily for 10 days.
Pathogenicity tests on 14-day old maize seedlings were performed by inoculating leaves of
susceptible hybrid Pr X K61 (genotype hm/hm) and resistant cultivar Great Lakes
(genotype Hm/Hm) with a fine mist of 104 conidia per ml suspended in 0.1% Tween 20.

Disease symptoms were observed daily until the plants were dead.

CHAPTER 3

Jenifer M. Gérlach, John W. Pitkin, and Jonathan D. Walton. 1997. Investigation of
the involvement of two proteases in virulence of the fungal pathogen, Cochliobolus
carbonum, on maize (in preparation).

Chapter 3
INVESTIGATION OF THE INVOLVEMENT OF TWO
PROTEASES IN VIRULENCE OF THE FUNGAL PATHOGEN,
COCHLIOBOL US CARBON UM, ON MAIZE

Abstract

This paper describes the cloning and disruption of ALP2, the gene encoding a
subtilisin-like protease secreted by Cochliobolus carbonum. The ALP2 gene was isolated
by PCR amplification using primers based on N-terminal amino acid sequence and
conserved amino acid sequences of subtilisins. A genomic and partial cDNA clone were
isolated and sequenced. Four introns were found which conform with 5’ and 3 ’ splice site
sequences from other C. carbonum genes. Transformation-mediated gene replacement of
ALP2 was performed in wild type C. carbonum and in an alpl mutant strain. Total
protease activity in crude culture filtrates of the alp2 and alpl/alp2 mutants grown on 1%
collagen were approximately 130% and 41%, respectively, of wild type activity, yet fungal
mat dry weights were similar to wild type. Chromatographic analysis of culture filtrates of
alp2 mutants indicated that one peak of protease activity, corresponding to Alp2, was
missing. The alpI/aIpZ double mutant lacked three peaks of protease activity,
corresponding to Alpla, Alplb, and Alp2. Disease symptoms on maize of the aIpZ and

alpI/alp2 were similiar to wild type.

57

58

Introduction

Proteases have been implicated in a number of plant-firngal (Rauscher et al 1995,
I-Iislop et al. 1982, Ries and Albersheim 1973, Choi et al. 1993, Ball et al. 1991), plant-
bacterial (Dahler et al. 1990, Dow et al. 1990), human-fungal (Monod et al. 1993, Tang et
al. 1993, Larcher et al. 1996), human-bacterial (Grenier 1996), and fungal-insect (St.
Leger 1995, Bonants et al. 1995) host-pathogen interactions. A possible role for
proteases during pathogenesis of these organisms on their respective hosts has been
demonstrated in only a few cases (Dow et al. 1990, Ball et al. 1991, St. Leger 1995).

The fimctions of these proteases and their proposed involvement in virulence may vary
depending on the pathogen and host. Proteases may degrade host proteinase inhibitors
(Grenier 1996) or proteinaceous substrates as a means of invasion or nutrient gathering
for proliferation of growth. Alternatively, proteases could have a role in the propagation
of the pathogen. Conidial discharge of the fungus Conidiobolus coronatus is regulated by
a secreted subtilisin-like protease (Phadatare et al. 1992, Phadatare et al. 1989). Protease
inhibitors afl‘ected conidial discharge, induction of the protease by casein promoted early
conidial discharge, and a mutant with reduced conidial discharge had reduced protease
activity (Phadatare et al. 1989).

To evaluate the role that proteases play in virulence of the firngal pathogen
Cochliobolus carbonum on maize, we have constructed null mutants deficient in protease
production. In a previous report, Murphy and Walton (1996) showed that there were at
least three alkaline proteases secreted by C. carbonum and that a gene disruption of ALP] ,

which encodes two of these proteases, has no significant effect on virulence or growth of

59

this fungus. As these proteases may have redundant functions during growth and host
invasion, we have cloned the gene encoding the third protease (Alp2) secreted by C.

carbonum and generated alp2 and alpI/aIpZ mutants.

Results and Discussion
Isolation and characterization of ALP2

Purification and characterization of the Alp2 subtilisin-like protein has been described
(Murphy and Walton 1996). The N—terminal amino acid sequence of Alp2 was determined
to be AYTTQSSAPWGLARISSQXRGTTGYXXDD (Figure 14), where X was an
unknown amino acid. Four degenerate oligonucleotides were designed to amplify by PCR
the DNA containing the gene encoding Alp2, ALP2. A 256-fold degenerate
oligonucleotide based on the N-terminal amino acid sequence AYTTQS (see Figure 14)
was used in combination with oligonucleotides based on highly conserved internal amino
acid sequences of fungal subtilisin-like proteases. The three additional 256-fold
degenerate oligonucleotides were based on the sequences TYGVAK (a sense and anti-
sense oligonucleotide were designed) and MATPHI/V (see Figure 14). The primers, in
appropriate combinations, were used in PCR with either genomic DNA or DNA isolated
from a C. carbonum cDNA library as templates. No PCR products were amplified when
the N-terminus-derived oligonucleotide was used in combination with any other primer.
However, a 450-bp cDNA-derived PCR product was obtained when the combination of
primers based on the internal amino acid sequences TYGVAK and MATPHI/V were used.

BLASTX analysis of the PCR product sequence showed a 78% identity to a Paecilomyces

60

lilacinus serine protease, 74% identity to a Fusarium alkaline protease, and 67% identity
to proteinase K. C. carbonum genomic and cDNA libraries were hybridized with the 450-
bp PCR fragment. A 1.45-kb partial cDNA clone (C2-1. 1) and a 8.0-kb PstI/EcoRI
genomic fragment (ALP2G1)(see Figure 16) were isolated and sequenced. Even though
the PCR product was derived from conserved internal amino acid sequences, the predicted
amino acid sequence of the cloned gene and the mature N-terminus of Alp2 were nearly
identical.

Figure 14 shows the nucleotide sequence of ALP2 and its deduced amino acid
sequence. Three independent RACE products (F roMan et al. 1988) were sequenced and
compared to determine the ALP2 transcription start site. The context of the codon for the
first methionine (CAACA'IEAAG) (Figure 14) is similar to the consensus for Neurospora
crassa translation initiation (CAMMALGGCG where M = C or A) (Edelrnann and Staben
1994). Typical lower eukaryotic promoter sequences, TATAA and CAAC (Gurr et al.
1987) are located 33 and 59 bp, respectively, upstream of the transcription start site
(Figure 14). Four introns were found when genomic and cDNA clones of ALP2 were
compared (Figure 14). The intron borders, splice junction sites and lengths are consistent
with those fiom other genes of C. carbonum and filamentous fungi (Figure 14 and Table
3). A characteristic polyadenylation signal sequence, AATAAA (Gurr et al. 1987), could
be found 33 bp upstream of the polyadenylation site (Figure 14).

The ALP2 gene is predicted to encode a 40.4-kD polypeptide, which is slightly

larger than the mature protein (3 8-kD) as determined by SDS-PAGE (Murphy and Walton

1996). Alp2 is predicted to contain one N-glycosylation site (N242-M-S) based on the

61

1 TGACAGCCGCCTTGCCACTAGGCTACTCATCGCTTCGCTCCGTTATCTGGATTTCGGCAA
61 GGGCTGTTGCGGGTGTAGATCGATTCATCCCACGGGCCTTCTAGGTTGGATCTCTACGCG
121 CACTTATTCACATCCCTAAACAAGTAAGCAATATTAATATAATATGCATCATGCCAAGAC
181 CCGACGTCTGACCTTCTACTAGCAGCCTTTCCTCCTTCTCCTAGACGCCCCGCCCACTCT
241 GTCACTGGTACAGCGGGTGCTTGGAAGCGCGTAGTGCCGGAATCACATGCTGTATCACCT
301 CCCAAACTAGGCATCAAAAGTCTACATACGCCATTTTGTGCCGCTTTGGCGACTGGCAAC
361 TCTACGCGCCATGGTGTAGGTACGCCGGCTCTGTACGCGATTGTCACATCGGTGATTGCT
421 AGTCCTGATTGGAAGGTGGGCTTGCCTCCTGTCATGTCATTGCAATTAGAGGCTACTGAG
I
4 8 1 TGTGTGTATATAAGAGGGACAGAATGCCTTGTTGAAACATCAGGCAACGAGCACAACAGC
5 4 1 ATCCAGCCCATCAACAGCAGCACTCATCAACCTTCGCTATATCAGCATCGTTCTTCGATC
601 GTTCTTCAACATGAAGCTCTCACTTCTCCTCGCTCTTCTGCCAGTGGCTCTTGCCCTTCC
M K L S L L L A L L P V A L A L P
661 TGCGCCAGTCATCGTTCCCCGTGCTGGTACTCCCATCCCAGGAAGGTACATTGTCAAGCT
18 A P V I V P R A G T P I P G R Y I V K L
721 CAAGAACCAGAACCTTGAGAACCTCATCAACACTGCCTTGAAGCTTCTCAAGAAGGACCC
38 K N Q N L E N L I N T A L K L L K K D P
781 CACCCACGTCTACAAGTTCGGTGGTTTCGGTGGTTTCTCCGCTGATATTACTGATGACAT
58 T H V Y K F G G F G G F S A D I T D D I
841 TGTTGAGCTGCTCCGCAACCTCCCCGGTgtaagcaattgatttccaacactacgagtcta
78 V E L L R N L P G
901 agcactaacaaagtacacacagGTCGACTACATCGAGCAGGATGCCGTTGTCCAAGCCAA
87 V D Y I E Q D A V V Q A N
961 CCTTGGTGTCGAGGTTGAGCTCGAGAAGAAGGCTTACACTACCCAGTCCTCTGCTCCTTG
100 L G V E V E L E K K A Y T T .Q, S S A P w
1021 GGGTCTCGCCCGTATCTCCAGCCAGAGCCGTGGCACCACCTCATACACCTACGACACCAG
120 G L A R I S S Q S R G T T S Y T Y D T S
1081 CGGTGGTGAGGGCACCTGCTCTTACGTCATCGACACTGGTATCCAGGTCGACCACCCAGA
140 G G E G T C S Y V I D T G I Q V D H P E
1141 GTTCGAGGGCCGTGCCACTTGGCTCGCCAACTTCGCTGACAGCTCGAACACTGACGGCAA
160 F E G R A T W L A N F A D S S N T D G N
1201 CthaatatacacttatcctcgaaagatgacaagtagactaacggtttctagGCCACGGC
180 G H G
1261 ACCCACTGTGCTGGTACCATCGGTTCCAAGACCTACGGAGTAGCCAAGAAGACTAAGCTG
183 T H C A G T I G S K T Y G V A K K T K L
1321 TACGCTGTCAAGGTCCTCGATGCCAGCGGCTCGgtatgtaaagacattgcctcttgttga
203 Y A V K V L D A S G S
1381 gatatgctgctaactatatcaagGGTACCAACTCCGGTGTTATTGCCGGTATCAACTTCG
214 G T N S G V I A G I N P
1441 TTGCCACCGATGCTAAGACCCGTAGCTGCCCCAACGGTGCCGTTGCCAACATGTCTCTTG
226 V A T D A K T R S C P N G A V A I .I. 8 L
1501 GTGGCAGCCGCTCCACCGCTGTCAACTCTGCTGCTGCCAATGCTGTTTCTGCTGGCGTCT
246 G G S R S T A V N S A A A N A V S A G V
1561 TCTTTGCCGTTGCTGCTGGTAACTCTGCCGCAAATGCTGCCAACTTCTCTCCCGCATCTG
266 F F A V A A G N S A A N A A N P S P A S
1621 AGCCCACTGTCTACACCGTCGGTGCCACCGACAGCTCCGACCGCCTGGCTACCTTCTCCA
286 E P T V Y T V G A T D S S D R L A T F S
1681 ACTTCGGCGCATCTGTCGACATCCTTGCCCCTGGTGTCTCTATCCTTTCCACCTGGATTG
306 N F G A S V D I L A P G V S I L S T W I
1741 GCGGCCGTACTgtaagttatatccaagaaggatattcatcaatttgtttttcatacggct
326 G G R T
1801 aacacttcaacagAACACCATCTCTGGTACCTCCATGGCTTCTCCCCACGTTGCTGGTCT
330 N T I S G T S M A S P H V A G L
1861 CGCTGCCTACATCCTGACTCTTGAGGGCAAGAAGACCCCTGCCGCTCTCTCTTCCCGCCT
346 A A Y I L T L E G K K T P A A L S S R L
1921 CACTGCTCTCTCCCTCAAGAGCAAGGTCACCGGTCTTCCCTCCGGCACCGTCAACAACCT
366 T A L S L K S K V T G L P S G T V N N L
1981 TGCCTTCAACGGCAACCCCTCCGCTACCTAAGCATGTTGCAAGCTGGTTCTAAGCAGGTC
386 A F N G N P S A T ‘
2041 GAGATGATGAGATGCCCTCTCTCTCTCCCTCTCTCCTTTGTGCTCTTTTTCCATTACAAC
2101 TGTATATATGATGATTGGGTTTCGCATAGGCTTTTGGGCTTGTTGCGCCGATGGTATTGG
+

 

 

2161 AAGAGATGGCATGATTGAGATTTAATAAACGATGTTGAGCAAATTCTTCTTGATACGTCT
2221 AGTGACCTTGAATCTTCTTTTCTCGAAAATACTCAATTTAGGTTGCATAAATCGTACTAT
2281 ACTAACATCGAGTTTGAAACCCCATCTCCCCAAGCATCCCCTTCAGAAAACGCACCCATC
2341 CGATCCCCGCACCCACTCTACCACCTCATACATCGCAAGGAAATAATTATACAGTAAAAC

Figure 14. Sequence of ALP2. The experimentally determined N-terrninal amino acid
sequence of Alp2 is double-underlined. Sequences conserved among other fungal
subtilisin-like proteases, used to design degenerate oligonucleotides for PCR amplification
of ALP2, are indicated by single-underlining. Introns are designated by lower case
lettering. The transcription start site is labeled with # and the polyadenylation site by +.
Symbols refer to nucleotides below and amino acid codes refer to the nucleotides above.
One predicted N-glycosylation site (starting at amino acid 242) is indicated by bold letters.

TABLE 3. Intron features C. carbonum genes.

62

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Gene Intron # 5 ' splice site branch site 3' splice site length (bp)
ALP2“ 1, T“GTAAGc CAC‘I‘AAC CAG 54
2. G ‘G‘I‘AATA GACTAAC TAG 50
3, G‘GTATGT TGCTAAC AAG 50
4, T‘GTAAGT GGCTAAC CAG 62
ALP 1b 1, G‘GTAAGT AACTAAC CAG 74
MG 1° 1 . C ‘ GTATGT TTCTGAC CAG 57
2. T‘GTATGT CGCTAAC CAG 64
3 . G ‘ GTGAGT TGCTAAC TAG 5 5
X17. 1d 1, G‘GTATGG CACTAAC CAG 53
mze 1, G“GTAGGT TACTAAT TAG 53
2. G‘GTAAGT GACTGTT TAG 56
M33 1 G ‘ GTAAGC TACTAAC CAG 60
CEL 1T 1, G“GTAAGT TTCTAAC CAG 55
PGNI‘ 1_ G‘GTAAGC GACTAAC CAG 57
PG X1“ 1, G‘GTTCGT TACTAAC CAG 51
2. G “GTGAGT AACTGAC TAG 63
3, G‘GTACGT TGCTAAC TAG 70
EXG? 1, c*GTAAGT TGCTGAC TAG 70
2_ C “GTAAGT TGCTAAC CAG 55
3, G ‘GTAAGC TGCTAAT CAG 47
4. GAGTATGT TACTGAC CAG 53
5. C‘GTGAGC AACTGAC CAG 55
TOXAJ' 1, A‘GTAAGT TACTAAT TAG 61
2, A“GTAAGT TTCTAAC TAG 74
3, T‘GTTAGT TACTGAT TAG 31
TOXCk 1 . G “ GTAGAG AGCTGAC CAG 52
Zigzag mgjny G‘GTAAGT TRCTAAC YAG 50-700

 

-R=AorG,Y=CorT
- a, Gorlach and Walton 1997b; b, Murphy and Walton 1996; c, Gerlach et al. 1997a; d,
Apel et al. 1993 ; e, Apel-Birkhold and Walton 1996; f, Sposato et al. 1995; g, Scott-Craig
et al. 1990; h, Scott-Craig et al. 1996; i, J .M. Gorlach unpublished data; j, Pitkin et al.

1996, k, Ahn and Walton 1997; l, Ballance 1991, Edelmann and Staben 1994.

 

63

information compiled by Gavel and von Heijne (1990) (N-X-T/s-Y, where X and Y are

typically not proline). Comparison of the experimentally determined N-terminus with the
predicted N-terminal amino acid sequence (Figure 14) reveals that the Alp2 precursor is
probably 109 amino acids longer than the mature protein. A signal peptide cleavage site,
as predicted by SignalP v1.1 (Neilsen et al. 1997), is ALAJvLP (residues 13 to 17, Figure
14).

The signal peptide is followed by a 94 amino acid propeptide sequence characteristic
of subtilisins. Interestingly, when the deduced amino acid sequence of ALP2 is compared
to other subtilisins (Figure 15), Alp2 has an additional 10 amino acids ending in K
preceding the mature N-terminus. Out of 27 fiingal and bacterial subtilisins retrieved fiom
GenBank, only three enzymes had a dibasic peptide preceding the experimentally
determined N-terminus (Geremia et al. 1993, Sato et al. 1994, Davidow et al. 1987).
Instead of, or in addition to, autocatalytic processing for activation (Power et al. 1986),
the dibasic amino acid (KK) sequence may indicate that Alp2 is further processed by an
enzyme similar to the endoproteases Mkcl, Kex2p, and Yap3p found in the secretory
pathway of Saccharomyces cerevisiae (Komamo and Fuller 1995, Bourbonnais et al.
1993, Julius et al. 1984). An additional KK processing site may lie between residues 55
and 56. Multiple precursor processing sites has been described by Matoba et al. (1988)
for a subtilisin secreted by Yarrowia lipolytica.

A comparison of the mature protein sequence of Alp2, deduced fiom the ALPZ
sequence, with subtilisins fi'om other filamentous fungi (Figure 15) showed 63% identity

(75.8% similarity) with a Metarhizium anisopliae cuticle-degrading protease (PIR

64

1 50
C. carbonum MKLSLLLALL PVALALP... ...... APVI VPRAGTPIPG RYIVKLKNQN
TL anisopliae MHLSALLTLL PAVLAAPATI GRRAEPAPLF TPQAESIIAD KYIVKFKDDI
Fusarium sp. MRLSI.IAVL PLALAAPV.L ....EPAPLL EARGSQPIAG KYIVKLKDTA
T. album MRLSVLLSLL PLALGAPA.V EQRSEAAPLI EARG.EMVAN KYIVKFKEGS
51 100
C. carbonum LENLINTALK LLKKDPTHVY KFGGFGGFSA DITDDIVELL RNLPGVDYIE
LL anisopliae ARIATDDTVS ALTSKADFVY E.HAFHGFAG SLTKEELKML REHPGVDFIE
Fusarium sp. KIGIMEATAK ..VANPERVY Q.NVIKGFSA SLCKEEVERL RHDPDVESIE
T. album ALSALDAAME KISGKPDHVY K.NVFSGFAA TLDENMVRVL RAHPDVEYIE
101 150
C. carbonum QDAVVQANLG VEVELEKKBY TTQSSAPWGL ARISSQSRGT TSYTYDTSGG
1L anisopliae KDAVMRIS.. ........§I TEQSGAPWGL GRISHRSKGS TTYRYDDSAG
Fusarium sp. QDAIISIN.. ........ AI TQQQGATWGL TRISHRQRGS TAYAYDTTAG
T. album QDAVVTIN.. ........ é. .AQTNAPWGL ARISSTSPGT STYYYDESAG
151 l l 200
C. carbonum EGTCSYVIDT GIQVDHPEFE GRATWLANFA DSSNTDGNGH GTHCAGTIGS
1L anisopliae QGTCVYIIDT GIEASHPEFE GRATFLKSFI SGQNTDGHGH GTHCAGTIGS
Fusarium sp. QGACAYVIDT GVEDTHPEFE GRAKQIKTFA ST.ARDGNGH GTHCSGTIGS
T. album QGSCVYVIDT GIEASHPEFE GRAQMVKTYY YS.SRDGNGH GTHCAGTVGS
201 250
C. carbonum KTYGVAKKTK LYAVKVLDAS GSGTNSGVIA GINFVATDAK TRSCPNGAVA
ML anisopliae KTYGVAKKAK LYGVKVLDNQ GSGSYSGIIS GMDYVAQDSK TRGCPNGAIA
Pusarium sp. KTYGVAKKVS IFGVKVLDDN GSGSLSNVIA GMDFVASDYR SRNCPRGVVA
T. album RTYGVAKKTQ LFGVKVLDDN GSGQYSTIIA GMDFVASDKN NRNCPKGVVA
251 300
C. carbonum NMSLGGSRST AVNSAAANAV SAGVFFAVAA GNSAANAANF SPASEPTVYT
TL anisopliae SMSLGGGYSA SVNQGAAALV NSGVFLAVAA GNDNRDAQNT SPASEPSACT
Fusarium sp. SMSLGGGYSA TVNQAAARLQ SSGVFVAVAA GNDNRDAANT SPASBPSVCT
T. album SLSLGGGYSS SVNSAAARLQ SSGVMVAVAA GNNNADARNY SPASEPSVCT
301 l 350
C. carbonum VGATDSSDRL ATFSNFGASV DILAPGVSIL STWIGGRTNT ISGTSMASPH
LL anisopliae VGASAENDSR SSFSNYGRVV DIFAPGSNVL STWIVGRTNS ISGTSMATPH
Fusarium sp. VGATDSSDRR SSFSNYGRAL DIFAPGTDIT STWIGGRTNT ISGTSMATPH
T. album VGASDRYDRR SSFSNYGSVL DIFGPGTSIL STWIGGSTRS ISGTSMATPH
351 400
C. carbonum VAGLAAYILT LEGKKTPAAL SSRLTALSLK SKVTGLPSGT VNNLAFNGNP
1L anisopliae IAGLAAYLSA LQGKTTPAAL CKKIQDTATK NVLTGVPSGT VNYLAYNGA*
Fusarium sp. IAGLGAYLLA LEG.GSASTI CARIQTLSTK NAISGVPSGT VNYLAFNNAT
T. album VAGLAAYLMT L.GKTTAASA CRYIADTANK GDLSNIPFGT VNLLAYNNYQ
401
C. carbonum SAT*
1L anisopliae .. .
Fusarium sp. ...
T. album A*..

Figure 15. Comparison of ALP2 predicted amino acid sequence with other fungal
subtilisin-like proteases. The sequences of C. carbonum, Metarhizium anisopliae (St.
Leger et al. 1992), Fusarium sp. (Morita et al. 1994), T ritirachium album (Gunkel and
Gassen 1989) were compared using PILEUP (Program Manual, 1994). Similar and
identical amino are indicated by . and :, respectively. The mature N-terminal amino acids
are underlined, the putative dibasic endopeptidase processing site is in italics, residues in
the catalytic triad are indicated by 1., and stop codons are designated with *.

65

8223 87, St. Leger et al. 1992), 62% identity (72.5% similarity) with a Fusarium sp. S-19-
5 alkaline protease (PIR JC2142, Morita et al. 1994), and 60% identity (71.7% similarity)
with T ritirachium album proteinase K (PIR S02142, Gunkel and Gassen 1989). The
catalytic triad found in serine proteases is conserved in Alp2 (Asp-159, His-190, Set-345;

Figure 15).

T ransformation—mediated gene replacement of ALP2

To investigate whether proteases have a role in the virulence of C. carbonum, we
generated alpZ and aIpI/alpZ mutants. A gene replacement plasmid, IMM80, containing
amdS, a gene required for acetamide utilization, flanked by 5’ and 3’ ALP2 sequences was
constructed and used to transform both C. carbonum wild type (3 67-2A) and aIpI mutant
(T515-3A, Murphy and Walton 1996) strains. Transformants able to use acetamide as the
sole nitrogen source were single-spore isolated. The mutant lacking Alp2, designated
T650-1A, and the alpI/alp2 mutant, designated T651-2A, were further analyzed. Figure
16 depicts the restriction map for the wild type ALP] locus and ALP} gene disruption
(Figure 16A) and the wild type ALP2 locus and ALP2 gene replacement (Figure 16B).
DNA blot analysis of genomic DNA isolated from wild type (3 67-2A), an alp] mutant
(T515-3A, Murphy and Walton 1996), an alp2 mutant (T650-1A), and an alpI/alp2
mutant (T651-2A) indicates that the ALP] locus is intact in 367-2A and T650-1A and
disrupted in T515-3A and T651-2A (Figure 16C). Likewise, the ALPZ locus is intact in
367-2A and T515-3A and replaced by amdS, as indicated by the absence of the internal

SalI fiagments, in T650-1A and T651-2A (Figure 16C).

 

 

 

 

 

 

 

 

 

 

A. Wild Type ALP] Locus B . Wild Type ALP2 Locus
(I: c
I - l l
RI 8 8 RI 8 s s s s
' 7.0— 0.1 0.5 2.2
ALP] Mutant Locus ALP2 Mutant Locus
W T | m!—

RI 6.5 R1 R1 6.0 31 S 7.0 _S— 7.0 ...—s

1 kb

 

 

 

EcoRI restriction map of the ALP] wild type and mutant loci depicting the result of a gene
disruption of ALP]. B, SalI restriction map of the ALP2 wild type and mutant loci
depicting the result of an ALP2 gene replacement with amdS. Predicted EcoRI and SalI
fragment sizes are indicated (in kb). ALP] and ALP2 open reading frames are indicated by
shaded boxes. C, DNA blot analyses comparing wild type (Lanes 1 and 5), alp] mutant
(Lanes 2 and 6), alpZ mutant (Lanes 3 and 7), and alpI/alp2 double mutant (Lanes 4 and
8). Identical DNA blots were hybridized with a cDNA of ALP] and the 4.5 kb ClaI
fragment of A1192. S, SaII; RI, EcoRI; C, ClaI.

67

Protease production in culture was measured for alp2 and alpI/alp2 mutants. Total
protease activity in crude culture filtrates of the alpZ and aIpI/alp2 (T651-2A) mutants
grown on 1% collagen was approximately 130% and 41%, respectively. The increase in
total protease activity in the alp2 mutant may be due to an increase in other protease
activities to compensate for the loss of Alp2. Growth of alpZ and alpI/alp2 mutants, as
determined by dry weight, in liquid culture containing 1% collagen or 1% macerated
maize cell walls was comparable to wild type; therefore, Alpl and Alp2 are not required
for in vitro growth on collagen or maize cell walls. There was also no observable delay in
sporulation in culture as a result of the mutations.

In addition to their involvement in catabolism, secreted proteases may be involved in
activating other secreted cell wall degrading enzymes (Rypniewski et al. 1993, Moormann
et al. 1993, Drapeau 1978). Therefore, for both the alp2 and alpI/alpz mutants we
investigated the activities of other C. carbonum-secreted enzymes. Like alp] mutants
(Murphy and Walton 1996), total exo-l,3-B-glucanase and polygalacturonase activities
were not significantly changed in alpZ and alpI/alp2 mutants. However, total cellulase,
endo-Bl,4-xylanase, and 1,3-l,4-B-glucanase activities were reduced to 8%, 85%, and
53%, respectively, in the alpI/aIpZ double mutant but not in alp] and alp2 single mutants
Therefore, Alpl and Alp2 may be fimctionally redundant enzymes involved in precursor
processing.

Wild type and alp2 and alpI/alpZ mutant strains were grown in liquid culture
supplemented with 1% collagen; protease activities from each were purified in parallel as

described (Murphy and Walton 1996). The alp2 mutants lacked one peak of protease

68

 

   
 
   

 

 

 

 

1.4

1.2 ‘-
3 +367-2A
g 1 -- —a—T650-1A
“E: 0 8 _- —o—T651-2A
a:
«‘E
3 0.
8
E
n.

 

 

‘_-_.. 1
‘-;!...':.
I

   

27 29 3 1 33
Retention Time (min)

Figure 17. Protease activity in fractions collected from cation exchange HPLC analysis of
proteins from culture filtrates of wild type (367-2A), an aIpZ mutant (T650-1A), and an
alpI/alpz mutant (T651-2A). Prior to HPLC analysis, proteins were passed over a low
pressure anion exchange column, concentrated, and dialyzed. One unit of protease
activity is defined as 1.0 OD410 per 15 11L enzyme fiaction at 45°C for 45 min.

69

activity corresponding to Alp2 (Figure 17) and therefore ALP2 encodes Alp2. The
alpI/alpz mutants lacked all three peaks of protease activity corresponding to Alp2,
Alpla, and Alplb (Figure 17). One substantial residual protease activity was detected
(Alp3, retention time 23.2 min, Figure 17) which, based on protease inhibition assays,
appears to be another subtilisin-like protease.

To test the potential role that proteases may play during pathogenesis of C. carbonum
on maize, 21-day-old maize seedlings were inoculated with 367-2A and alp], alpZ, and
alp1/alp2 mutants. There were no distinguishable differences in lesion onset or
morphology between wild type and any of the mutants on PR (resistant) or PR1
(susceptible) maize cultivars. As Alpla, Alplb, and Alp2 constitute the majority of
proteases secreted by C. carbonum in vitro we conclude that proteases alone probably do

not significantly contribute to the overall virulence of C. carbonum on maize.

Materials and Methods
Fungal growth and maintenance

Fungal strains were routinely grown on V8-juice agar for production of conidia. For
protease production, approximately 4.5 x 105 spores were inoculated into a 1-L
Erlenmeyer flask containing 200 mL mineral salts medium plus 0.1% yeast extract, trace
elements, and 1% type I collagen (Sigma C-9879) or 1% macerated maize cell walls
(Sposato et al. 1995) (van Hoof et al. 1991). For protein purification cultures were
incubated at 22 to 25°C with continuous shaking at 130 rpm for 3.5 days. For growth

studies cultures were incubated for seven days at 22 to 25°C without shaking.

70

Protein purification and enzyme assays
Purification of secreted proteases during mutant analysis was previously described

(Murphy and Walton 1996). Serine protease activity was measured as described (Murphy
and Walton 1996). Units of enzyme activity are defined as OD410 per 15 11L enzyme

fiaction at 45°C for 45 min.

Nucleic acid manipulation and sequence analysis

Fungal genomic DNA was isolated as described by Pitkin et al. (1996). RNA was
isolated as described (Chomczynski and Sacchi 1987). The 5’ end of the ALPZ transcript
was determined using the Amplifinder RACE kit (Clonetech, Palo Alto, CA) (F rohman et
al. 1988). The oligonucleotide used for reverse transcription was GCTCGAACACT
GACGGC (starting at nucleotide 1182, see Figure 14). The PCR amplification step was
performed using the “anchor” primer provided by the RACE kit in combination with the
primer CGGTGGTTTCGGTGGTT (starting at nucleotide 799, see Figure 14).

ALP2 sequence was generated by automated fluorescent sequencing at the MSU-
DOE-PRL Plant Biochemistry Facility using an Applied Biosystems (Foster City, CA)
Catalyst 800 for Taq cycle sequencing and an Applied Biosystems 373A Sequencer for

analysis of the products.

Isolation of ALP2
The reaction mixure and PCR conditions for amplifying DNA containing the ALP2

gene were as described (Murphy and Walton 1996), with the exception that annealing was

71

performed at 45°C. Template DNA was genomic DNA and DNA isolated from phage
lysate of a C. carbonum cDNA library (Pitkin et al. 1996). PCR primers were as follows:
GCNTAYACNACNCARTC, corresponding to the Alp2 N-tenninal amino acid sequence
AYTTQS, and ATGGCNACNCCNCAYRT and TAYGGNGTNGCNAARAA,
corresponding to the subtilisin internal amino acid sequences MAPTHI/v and TYGVAK,
respectively (where N = any nucleotide, R = A or G, and Y = T or C). PCR products
were cloned as described by Murphy and Walton (1996) into the SmaI restriction site of

pBluescriptII SK+.

Screening of the genomic and cDNA libraries, DNA blotting, probe labeling, and
hybridization protocols have been described (Scott-Craig et al. 1990, Sposato et al. 1995).

DNA was blotted to Nytran (Schleicher and Schuell, Keene, New Hampshire).

Disruption of ALP2

The ALP2 gene replacement vector was generated by first subcloning the 4.5-kb
ClaI fragment (Figure 16A) into Ach-digested pBluescriptII SK+ lacking KpnI, resulting

in pJMM78. The gene encoding acetamidase, amdS, was liberated fi'om pAMD-72 (Pitkin
et al. 1996) with KpnI and SalI and introduced into pJMM7 8 digested with KpnI and SalI

such that 900-bp of the ALP2 open reading frame was replaced with a 4.0-kb fragment of

amdS (see Figure 16A). The resulting plasmid, pJMMSO, was than digested with XhoI

and EcoRI to liberate vector sequence from the insert and the entire digest was introduced

72

into both a C. carbonum wild type strain, 367—2A, and a hygromycin B-resistant (Hng)

alp] mutant strain, T515-3A (Murphy and Walton 1996).
Isolation and transformation of protoplasts has been described (Scott-Craig et al.
1990, Apel et al. 1993). Transforrnants were selected for their ability to utilize acetamide

(Ace+) as a sole nitrogen source (Pitkin et al. 1996). The alpI/alpz mutants were selected

for their ability to grow on acetamide plates (Hynes et al. 1983) and V8-juice agar plates

containing 100 units/mL of hygromycin B (Calbiochem). To ensure nuclear homogeneity,
Ace+ (alpZ mutants) transformants and Ace+lHng (aIpI/alpz mutants) transformants
were single-spore isolated as described by Pitkin et al. (1996).

Pathogenicity was measured by inoculating 21-day-old PR1 (susceptible) and PR
(resistant) maize seedlings with 104 conidia/ml in 0.1% Tween-20. Disease symptoms

were analyzed until the plants were dead (approximately 4 days).

CONCLUSION

Microscopic and genetic evidence suggests that Cochliobolus carbonum may require
cell wall depolyrnerases to penetrate the maize cell wall. However, which cell wall
degrading enzyme is (are) essential to the process of cell wall degradation is not known.

This study focused on investigating the role that several different plant cell wall
degrading enzymes have during the colonization of maize by C. carbonum. Three
proteases (Alpla, Alplb, Alp2) and three 1,3-l,4-B-glucanases (Mlgla, Mlglb, Mlg2)
were purified and the corresponding genes cloned. Gene disruption experiments were
performed to generate alp], alp2, alpI/alp2, and mlg] mutants. Each mutant was tested
for its ability to cause lesions on maize leaves. The result of these experiments indicated
that each enzyme or enzyme class (for proteases only) is not required for this fungus to
penetrate the maize wall. One can not say, however, that these mutations do not have
some efi‘ect that we are unable to detect in the vinrlence assays we conduct under artificial
greenhouse conditions. These mutations may, over generations, decrease the fitness of

this organism in the field.
Future Objectives

One hurdle that has been encounted in the search for the key enzyme(s) involved in

host wall penetration is redundancy. The second has been the uncertainty of whether the

73

74

enzymes being investigated in in vitro studies are being expressed in planta. Several
approaches could be implemented to overcome these problems. One approach would be
to identify and dismpt genes encoding transcription factors required for cell wall
degrading enzyme expression. It is possible that one transcription factor may either
regulate all genes encoding one enzyme class or genes induced upon exposure to a
particular carbon source. As X113 from C. carbonum is expressed only in planta, one
could identify transcription factors specific for in planta gene expression. Gene
disruptions in several of these transcription factors could eliminate the expression of a
significant number of wall depolymerases.

A second approach is to alter the processing of the secreted enzymes such that
once secreted they are inactive. When one compares the experimentally determined
mature N-termini of the cell wall degrading enzymes purified from C. carbonum (Table 4,
indicated by #) to the mature N-terrnini predicted by SignalP v1.1 (Nielsen et al. 1997),
one sees that several of these enzymes do not have typical signal peptides sequences. In
addition, there are C. carbonum enzymes for which the predicted mature N-termini
indicate that they would also lack a typical signal peptide sequence (Table 4, indicated by
*). The extra amino acids may, however, correspond to processing signals required for
enzyme activation prior to secretion. The basic residues preceding the mature N-termini
of Alp2, Exg2, Pgnl, and Pmel and the putative N-terrnini of ngl, Xyll, Xy12, Xyl3, and
Xyl4 (Table 4) conform with the processing signal sequence recognized by the
Saccharomyces cerevisiae processing enzymes Kex2p, Yap3 p, and Mkc7p (Julius et al.

1984, Komano and Fuller 1995, Bourbonnais et al. 1993). Disruption of C. carbonum

75

TABLE 1. Features of C. carbonum cell wall degrading enzyme N-termini.

 

Protein Putative Signal ngtide Sequence N- Terminus
#
A11’1 aIIRI-‘QSMITAALPALELSAPTPQWDDVPEDS "3

Alp2 bMKLSLLLALLPVALALPAPVIVPRAGTI?IPGRYIVKLKNQNL
EN LINTALKLLQDPTHVYKFGGFGGFSADITDDIVELLR

NLPGVDYIEQDAVVQANLGVEVELEQ; A“
Exgl CMRFSSLLACLGAVGLCLAAAIPFQQVDNTTDSGSLDAAM #3“
15”‘92 dMILTKLVSTLSLCAAMAPAQQQ #3"
"191 °MSLKSLFVSAPLLWRGLTALPAGPGSWTHGNSTIvssSDFSAA #5“
P991 EMVAYALTSMLLSAGALflLAAPSGLDAg #DGC
Pmel 'Ml-IPTLVFFLSLVAAIMAPAENVLB; #3”
1’9"1 gMRVTDIISCALLQASIALSTPVEELGAKAWAQ *FPP
XYll hMVSFTSIITAAVAATGALAAPATDVSLVAB *9“
"Y12 i'MVSFKSLLLAAVATTSMAPFDFLRERDDVNATALLEQ *9”
XY13 iM'VAFTSVLLGLSAIGSAEAAPVADVPDFEFSGPKHLAAB *9“
XY14 jMI-(FSLITILSASALMSPFAEPEAFLEEE *9”
cell kMYRTLAFASLSLYGAfl *9”
Cum IMKFLTLSMMTALflASPITTRSETA *3“

# Indicates that the N-terminus was experimentally determined

- 1. Indicates that N-terminus was determined by sequence homology with other enzymes
having experimentally determined N—termini.

- Single-underlined amino acids precede putative signal peptide cleavage sites as predicted
by SignalP v1.1 (Nielsen et al. 1997).

- Double-underlined amino acids are residues possibly processed by Yap3 p, Mkc7p, and
Kex2p homologues (Julius et al. 1984, Bourbonnais et al. 1993, Komano and Fuller
1995)

- Bold letters indicate the first three amino acids in the (putative) mature form of the
enzyme.

- a, Murphy and Walton 1996; b, Gfirlach et al. 1997b; c, Nikolskaya et al. 1996; d, C.

Caprari unpublished data; e, G6rlach et al. 1997a; f, Scott-Craig et al. 1990; g, Scott-

Craig et al. 1996; h, Apel et al. 1993; i, Apel-Birkhold and Walton 1996; j, P.C. Apel-

Birkhold unpublished data; k, Sposato et al. 1995; l, J .S. Scott-Craig unpublished data.

76

KEXZ, MKC 7, and YAP3 homologues may block the activation of several secreted cell
wall degrading enzymes. Because of redundancy, not all enzyme activities in one class
may be eliminated but the overall reduction of active wall depolyrnerases may have an
afl'ect on host invasion by delaying infection or reducing the total number of successfiil
penetrations. One possible risk with this approach is that mutations in any one of these
three loci might be lethal. S. cerevisiae mutants deficient in Kex2p are viable (Rogers et
al. 1979); however, a mutation in the KHZ-homologue of Schizosaccharomyces pombe,
KRP, is lethal (Davey et al. 1994). Another consideration is that a mutation in KEXZ,
YAP3, or AJKC 7 may alter the processing of a protein, other than a cell wall degrading
enzyme, that is essential for growth and/or pathogenesis, making pathogenicity tests
dificult to interpret.

To address the concern about whether the genes encoding C. carbonum cell wall
degrading enzymes are expressed in planta, one could use these genes to probe RNA
blots of C. carbonum-infected maize tissue. With this information, it could be determined
which enzymes are being expressed in the compatible (susceptible) or incompatible
(resistance) interaction during the early and late stages of the infection process. If there is
temporal regulation, one could evaluate whether gene expression correlates with spore
germination or differentiation of hyphae into appressoria, penetration pegs, or conidia.
Possible common themes of expression may emerge revealing shared upstream regulatory
sequences which govern expression in planta.

As an alternative one could examine what plant or firngal-derived signals induce the

expression of genes such as XYL3. One could then identify the fungal receptor which

77

starts the signal transduction cascade leading ultimately to cell wall degrading enzyme
gene expression. Furthermore, one could investigate which proteins are involved in the
signal transduction cascade and whether the pathogenicity factor, HC-toxin, is also
regulated by this pathway.

As many phytopathology laboratories focus on one particular class of wall
depolymerase or have identified many degradative enzymes without obtaining the
corresponding genes, little is known about how cell wall degrading enzymes fit into the
process of disease. Completing one of the approaches outlined above would increase the
collective knowledge about C. carbonum wall depolyrnerases; getting us closer to the goal

of understanding the role that cell wall degrading enzymes play in pathogenicity

APPENDICES

APPENDIX A

APPENDIX A

PURIFICATION OF A PUTATIVE EXO—l,3-B—GLUCANASE AND
MOLECULAR CLONING OF THE GENE, EXGZ,
FROM COCHLIOBOL US CARBON UM

Introduction

The cell walls of ascomycetous firngi are composed of chitin, protein, 1,3-B-glucan,
1,6-glucan, and melanin. During growth, chitin and 1,3-B-glucan are extruded from the
cell and cross-linked together (for a review see Wessels 1994). Cell expansion or
differentiation might then involve the secretion of hydrolytic enzymes capable of degrading
these cell wall components. Both endo-l,3-B-glucanases and exo-l,3-B-glucanases have
been described from fungi (van Hoof et al. 1991, Chambers et al. 1993, Hien and Fleet
1983). In Saccharomyces cerevisiae these enzymes are synthesized during difi‘erent stages
of development (Laniba et al. 1995, Rey et al. 1979). EXGI and EXGII are constitutively
expressed exo-l,3-B-glucanase activities secreted by S. cerevisiae. SPRl (also called
SSGl) is a sporulation-specific exo-1,3-B-glucanase (Muthukumar et al. 1993, San
Segundo et al. 1993). The spr] mutants have mild phenotypes such as reduced ascospore
thermoresistance and delayed ascus maturation.

In phytopathogenic organisms, exo-l,3-B-glucanases might be involved in degrading
callose deposits that the plant has synthesized in defense against invading pathogens. In

an attempt to investigate this, Schaeffer et al. (1994) generated an exo-1,3-B-glucanase

71

79

APPENDIX A

mutant and evaluated its ability to penetrate maize tissue. The results indicated that the
mutant was as pathogenic as wild type C. carbonum. However, it must be noted that
residual activity remained in the mutant, probably due to a l,3-1,4-B-glucanase which was
able to hydrolyze 1,3-B-glucans.

This paper describes the identification and cloning of a second exo-l,3-B-glucanase
secreted by C. carbonum. This enzyme was identified as being a major protein in culture
filtrate (C. Caprari, unpublished data). N-terminal amino acid sequence identified it as a

homolog of EXGl from yeast. The gene was cloned and sequence data were obtained.

Results and Discussion
Protein purification and characterization

One method to investigate the role of cell wall degrading enzymes in host colonization
is to purify proteins that are secreted at high levels, obtain N-terminal amino acid
sequence, clone the corresponding gene, and generate null mutants which can be tested for
pathogenicity. Using this approach, C. carbonum was grown for nine days on macerated
corn cell walls, crude culture filtrate was harvested, and proteins were separated by
cation-exchange HPLC. A major peak of 1,3-B-glucanase activity was measured in the
column flow through. This fraction was filrther purified by hydrophobic interaction HPLC
resulting in three peaks of 1,3-B-glucanase activities. The third peak was identified from

its N-terminus as the already characterized Exgl protein (van Hoof et al. 1991, Schaefl‘er

80

APPENDIX A

et al. 1994). The second peak, in addition to having 1,3-B-glucanase activity, appeared to
contain B—glucosidase activity. Two abundant proteins, 50- and 3 l-kD, from the second
peak were resolved on SDS-PAGE and sequenced at the N-terminus. The N-terrninal
amino acid sequence of the 3 l-kD protein, AAYNLIDTYDASNWASKFNFEDIAD, was
identical to Mlgl (Gorlach et al. 1997a). The N-terrninal amino acid sequence of the 50-
kD protein, called p50, VGFNWGSEKIRGVNIGGXLVLEPXITPSI where X is
unknown, had 51% identity with an exo-l,3-B-glucanase, XOGl, from Candida albicans
(Chambers et al. 1993), 73% identity with the sporulation-specific exo-l,3-B-glucanase,
SPRl, from S. cerevisiae (Muthukumar et al. 1993), and 77% identity with EXGl from S.
cerevisiae (Vazquez de Aldana et al. 1991). The first peak of 1,3-B-glucanase activity
was a 65-kD protein as determined by SDS-PAGE. The N-terrninal amino acid sequence
was identified as FVGSATVSSTVLVIARDAISALN, which had no sequence homology

to any protein in the database.

Isolation and characterization of the gene encoding p5 0, EX 02

Based on part of the N-terminal amino acid sequence (VGFNWG, Figure 18) and
sequence fiom the homologous exo-l,3-B-glucanases (QNGFDN, Figure 19) described,
two degenerate PCR primers were synthesized. Using genomic DNA and DNA fi'om a C.
carbonum cDNA library as templates, two PCR products, 550-bp and 430-bp, were

amplified, respectively. Sequence analysis of the PCR products confirmed that they

81

APPENDIX A

1 GGTATGTTGGCATTTCTTTGTCTGCTCACGCTCGGATTCAGTATAAAAGTGACCTCCATG
61 ACCGCGTCTTGAACTCTCCACTCCAGCAACTCATCCTGACAACACAATCTGAGAGCAACA
121 TCATCTTTGCAACATGATTCTCACCAAACTCGTTTCAACCTTGTCGCTATGCGCTGCTGT
M I L T K L V S T L S L C A A V
181 CCTGGCTGCTCCTGCgtaagtgatgcttccagcgcatcactgccaaaacaaacatcacat
17 L A A P A
241 tgtgtccttgctgacttttctctagCCAGCAGAAGCGTGCAGTCGGCTTCAACTGGGGAT
22 Q Q K R A V G P N W G
301 CTGAGAAGATTCGCGGTGTAAACATTGGTGGATGGTTGGTCCTGGAGCCgtaagtgacat
33 S E K I R G V N I G G W L V L E P
361 gggttgtgatatgctttgcgacatctgctaacctccttccacagCTGGATCACCCCATCA
50 W I T P S
421 ATCTTCGACAACGCAAACCGAGGCCGTCCACAAAATGACTTTGTTGACGgtaagctgaat
55 I F D N A N R G R P Q N D F V D
481 gacttteatattgagactgatgctaataacgcgcagAGTACACATTGGGTGAAAAGCTTG
71 E Y T L G E K L
541 GCAGCCAAAATGCCCTAAACATCCTTCGTAACCATTGGGATACTTTCGTCACCTGGCAAG
79 G S Q N A L N 1 L R N H W D T F V T W Q
601 ACTTCAACAAAATCAAGCAGTCGGGTTTCAACGTTGTCCGTATCCCCGTTGGCTACTGGG
99 D F N K I K Q S G P N V V R I P V G Y W
661 CATACGATACCTTTGGCTCGCCCTACGTCAGTGGAGCAGCTGTCTACATTGATGCTGCCA
119 A Y D T P G S P Y V S G A A V Y I D A A
721 TCGACTGGGCTCGTAGCCTGGGCTTGAAGATTATCATTGACCTTCACGGTGCACCTGGAT
139 I D W A R S L G L K I I I D L H G A P G
781 CCCAGAACGGGTTTGACAACTCTGGTCAACGCATGGATCGCCCCACGTGGCAGCAAGGCG
159 S Q N G F D N S G Q R M D R P T W Q Q G
841 ACACCGTCCGAAGGACCCTTCAAGTTTTGCGCACAATCTCTCAGAAATACGCACAAACGA
179 D T V R R T L Q V L R T I S Q K Y A Q T
901 GCTACCAGGATGTCATCGTCGGTATTCAACTGCTCAACGAGCCCGCACTTTACAACGGCC
199 S Y Q 0 V 1 V G I Q L L N E P A L Y N G
961 TCAGTCGTGATGTTCTTGCACAGTTCTATCGCGATGGCTATGGCCAAGTGCGCGAGGTTT
219 L S R D V L A Q P Y R D G Y G Q V R E V
1021 CCGACACGCCGGTCATCATTTCTGACGGCTTCACTGCACCAAACTCCTGGAACGGCTTCC
239 S D T P V I I S D G F T A P N S W N G F
1081 TCACACCCTCAGATGCCAACGCCCAGAATGTAGCCATTGACAACCACCAATACCAAGTTT
259 L T P S D A N A Q N V A I D N H Q Y Q V
1141 TCGACTCTAATCTGCTCAAACTGTCACCCGCTGGACACGCCCAACAGGCCTGCAGGAACA
279 P D S N L L K L S P A G H A Q Q A C R N
1201 CTGGCGCATATGGCGGTGCAGACAAGTGGACCTTTGgtatgttcaagaagatgcgataca
299 T G A Y G G A D K W T F
1261 AtaagcccttactgacttatatgttgcagTTGGAGAGTGGACTTCCGCCATGACgtgagc

 

311 V G E W T S A M T
1321 atgatcctttcacacataacacaaaaagcacataaactgactgccggcagGGACTGCGCA
320 D C A

1381 CGTTACTTGAACGGCTACGGCCGTGGTGCCCGCTATGATGGTACCTACTTGGGCAACCCC
323 R Y L N G Y G R G A R Y D G T Y L G N P
1441 AAGTTGGGCGAGTGCGGCTGGCGAAACGACCTAGCGCAGTGGCCCGCCTCTTATAAGGAC
343 K L G E C G W R N D L A Q W P A S Y K D
1501 GACTCCAGGCGCTACATCGAGGCCCAGATCCGCGCTTTTGAGTCGACGACCCAAGGCTGG
363 D S R R Y I E A Q I R A F E S T T Q G W
1561 TTCTGGTGGAACTTCAAGACTGAGGGTGCGGCTGAGTGGGATGCTTTCAGGCTCATCGAC
383 F W W N F K T E G A A E W D A F R L I D
1621 GCTGGTGTCTTCCCCGCCATCAGGAACGGACAGGTAGAGTACAAGTTTGGCGCTGCCTGC
403 A G V F P A I R N G Q V E Y K F G A A C
1681 TAGAAACATGGCTTGACACCTGGCTTAAAATTTACTTTGTTGCACCTCTAGAAGGGGCTA

423 *

1741 GACGGGCTGAACGCAATCGCAAAGGAAAAGCAAAGCAGGCATGTGTGATTGCATAGAGGA

1801 GGCCTGACTTTTCATTGCTCAGTTCGATTAACAACTTATGATGATTTGGAATAACGAAGT
+

1861 ACACGATAGCAATGAGATTTTTTTTTCATACACTTTTTGTGCAAGCACAAGATAAGCTAC

Figure 18. Sequence of EXGZ. The N-terrninal amino acid sequence of the mature
protein is indicated by double-underlining. The amino acid residues conserved amongst
exo-1,3-B-glucanases and used to design a degenerate oligonucleotide for amplification of
EXGZ are indicated by single-underlining. The introns are designated by lower case
letters. The polyadenylation site is labeled with + (symbols refer to the nucleotides below
and amino acid codes refer to the nucleotide above). There are no predicted N-
glycosylation sites.

82

APPENDIX A
1 50
C. carbonum EXGZ .......... ..MILTKLVS TLSLCAAV.. ..... LAAPA QQKRAVGFNW
S. cerevisiae EXGI .MLSLKTLLC .TLLTVSSVL ATPV..PARD PSSIQPVHEE NKKRYYDYDH
S. cerevisiae SPRI .MVSPRGLTT LTLLFTKLVN CNPV..STKN RDSIQFIYKE KDSIYSAINN
C. albicans XOGI MQLSFILTSS VFILLLEPVK ASVISNPFKP NGNLKFKRGG GHNVAWDYDN
51 100
C. carbonum EXGZ GS..EKIRGV NIGGWLVLEP WITPSIFD.. NANRGRPQND FVDEYTLGEK
S. cerevisiae EXGI GSLGEPIRGV NIGGWLLLEP YITPSLFEAF RTNDDNDEGI PVDEYHFCQY
S. cerevisiae SPRI QAINEKIHGV NLGGWLVLEP YITPSLFETF RTNPYNDDGI PVDEYHPCEK
C. albicans X061 N....VIRGV NLGGWFVLEP YMTPSLFEPF Q.NGNDQSGV PVDEYHWTQT
101 150
C. carbonum EXGZ LGSQNALNIL RNHWDTFVTW QDFNKIKQSG FNVVRIPVGY WAYDTPG.SP
S. cerevisiae EXGI LGKDLAKSRL QSHWSTFYQB QDFANIASQG FNLVRIPIGY WAFQTLDDDP
S. cerevisiae SPRI LGYEKAKERL YSHWSTFYKE EDFAKIASQG PNLVRIPIGY WAFTTLSHDP
C. albicans XOGI LGKEAASRIL QKHWSTWITE QDFKQISNLG LNFVRIPIGY WAPQLLDNDP
151 200
C. carbonum EXGZ YVSGAAV.YI DAAIDWARSL GLKIIIDLHG APGSQNGPDN SGQRMDRPTW
S. cerevisiae EXGI YVSGLQESYL DQAIGWARNN SLKVWVDLHG AAGSQNGFDN SGLR.DSYKF
S. cerevisiae SPRI YVTAEQEYFL DRAIDWARKY GLKVWIDLHG AAGSQNGFDN SGLR.DSYKF
C. albicans XOGI YVQG.QVQYL EKALGWARKN NIRVWIDLHG APGSQNGFDN SGLR.DSYNP
201 250
C. carbonum EXGZ QQGDTVRRTL QVLRTISQKY AQTSYQDVIV GIQLLNEPAL YNGLSRDVLA
S. cerevisiae EXG] LEDSNLAVTT NVLNYILKKY SAEEYLDTVI GIELINEP.L GPVLDMDKMK
S. cerevisiae SPRI LEDENLSATM KALTYILSKY STDVYLDTVI GIELLNEP.L GPVIDMERLK
C. albicans XOGI QNGDNTQVTL NVLNTIPKKY GGNEYSDVVI GIELLNEP.L GPVLNMDKLK
251 300
C. carbonum EXGZ Q.FYRDGYGQ VRE..VSDTP VIISDGFTAP NSWNG FLTP SDANAQNVAI
S. cerevisiae EXGI NDYLAPAYEY LRNNIKSDQV IIIHDAFQPY NYWDD FMTE ND.GYWGVTI
S. cerevisiae SPR] NLLLKPAYDY LRNKINSNQI IVIHDAFQPY HYWDG FLND EK.NEYGVII
C. albicans XOGI Q.FFLDGYNS LRQT. GSVT PVIIHDAFQV FGYWNNFLTV AE.GQWNVVV
301 350
C. carbonum 3X62 DNHQYQVFDS NLLKLSPAGH AQQACRNTGA.YGGADKWTFV GEWTSAMTDC
S. cerevisiae EXGI DHHHYQVFAS DQLERSIDEH IKVACEWGTG.VLNESHWTVC GEPAAALTDC
S. cerevisiae SPRI DHHHYQVFSQ VELTRKMNER IKIACQWGKD.AVSEKHWSVA GEFSAALTDC
C. albicans XOGI DHHHYQVFSG GELSRNINDH ISVACNWGWD.AKKESHWNVA GEWSAALTDC
351 400
C. carbonum EXGZ ARYLNGYGRG ARYDGTYL.. ..GNPKLGEC GWRNDLAQWP ASYKDDSRRY
S. cerevisiae EXGJ TKWLNSVGFG ARYDGSWVNG DQTSSYIGSC ANNDDIAYWS DERKENTRRY
S. cerevisiae SPR1 TKWLNGVGLG ARYDGSWTKD NEKSHYINTC ANNENIALWP EERKQNTRKF
C. albicans XOGI AKWLNGVNRG ARYEGAY... .DNAPYIGSC QPLLDISQWS DEHKTDTRRY
401 450
C. carbonum EXGZ IEAQIRAFES TTQGWFWWNF KTEGAAEWDA FRLIDAGVFP AIRNGQVEYK
S. cerevisiae EXGI VEAQLDAFEM RG.GWIIWCY KTESSLEWDA QRLMPNGLFP QPLTDR...K
S. cerevisiae SPRI IEAQLDAFEM TG.GWIMWCY KTENSIEWDV EKLIQLNIFP QPINDR...K
C. albicans XOGI IBAQLDAFEY TG.GWVFWSW KTENAPEWSF QTLTYNGLPP QPVTDR...Q
451 460
C. carbonum EXGZ FGAAC .....
S. cerevisiae EXGI YPNQCGTISN
S. cerevisiae SPRI YPNQCH....
C. albicans X061

FPNQCGFH..

Figure 19. Comparison of the predicted amino acid sequence of EXGZ with other known
exo-1,3-B-glucanases. The sequences Exg2 of C. carbonum (this study), EXGI of S.
cerevisiae (SwissProt P23776, Vazquez de Aldana et al. 1991), SPRl of S. cerevisiae
(SwissProt P3 2603, Muthukumar et al. 1993), and XOGl of C. albicans (SwissProt
P29717, Chambers et al. 1993) were compared by PILEUP (Devereux et al. 1984).
Amino acid located at the mature N-terminus is underlined. Identical (:) and similar (.)
amino acids are indicated. The conserved region used to design a degenerate
oligonucleotide primer for PCR-amplification of EXGZ is indicate in bold letters. Stop
codons are designated by *.

83

APPENDIX A

encoded the purified p50 (Figure 18). The size difference between the two PCR products
was later established to be due to two introns (Figure 18).

Genomic and cDNA libraries were screened with the 43 0-bp cDNA-derived PCR
product. A 1.7-kb cDNA clone, C2-l . l, was isolated and sequenced. To obtain sequence
of the genomic copy of EXGZ, two PstI fi'agrnents, 2.4-kb and 1.2-kb, were subcloned
(Figure 20). For confirmation that the 2.4-kb and 1.2-kb fiagments overlapped at the
same PstI restriction site, sequence was obtained fi'om a 65 O-bp BamHI/Kpnl fragment

which spans the PstI restriction site (Figure 20). When cDNA and genomic sequences

 

 

1.0111:

1———1

1 9.0 kb 1
RI 9 x H p s 9 RI
1 I l

‘

~:
—
—

 

 

Figure 20. Restriction map of the wild type H02 locus showing the location of the
EXGZ transcript (shaded box). The PstI and BamHI/KpnI fiagments sequenced are
indicated (in kb). P, PstI; X, XhoI; RI, EcoRI ; H, HindIII; B, BamI-II; K, KpnI; S, SalI.

were compared, five introns were identified; three within the first 380-bp of the open

reading flame and two closer to the C-terrninal end of the protein (Figure 18). The 5' and

84

APPENDIX A

3' intron borders, splice branch sites, and lengths are consistent with introns in other genes
of C. carbonum (G6rlach et al. 1997b).

The deduced amino acid sequence (Figure 18) predicts that EXGZ encodes a mature
polypeptide of 44,525 Daltons, which is smaller than the SDS-PAGE estimated size of 50-
kD for this protein, and has a pI of 5.51. The context of the predicted translation start site
(CAACALGATT) (Figure 18) conforms with the Neurospora crassa transcription start
site consensus sequence (CAMMA_T_GGCT, where M = C or A, Edelmann and Staben
1994). The amino acids residues preceding the mature N-terminus (KR) do not agree
with the (-3, -1) rule described by von Heijne (1986). Instead these charged amino acids
are indicative of a Kex2p processing site (Julius et al. 1984). A comparison of the
deduced amino acid sequence of EXGZ with other exo-l,3-B-glucanase genes (Figure 19)
indicated that several of these enzymes have the dibasic peptide sequence KR preceding
the mature N-terminus. It has been shown that EXGI of S. cerevisiae is processed by
Kex2p (Basco et al. 1996). The significance of this processing is not known as the
enzyme is active without being processed . Ifthe exo-l,3-B-glucanases are involved in
fiingal cell wall metabolism, they may be regulated by this post-translational processing.

The similarity between the deduced amino acid sequence of protein encoded by EYGZ
and that of known exo-1,3-B-glucanases is 61 to 63% (Figure 19). However, this protein
is not similar to Exgl from C. carbonum which shares slight homology with an a-amylase

from Bacillus lichentformis (van Hoof et al. 1991). The homologies with the yeast

85

APPENDIX A

proteins indicate that the protein encoded by EXGZ is a member of family 5 of the
glycosylhydrolases (Henrissat and Bairoch 1993) As p50 co-purified with Mlgl, it is not.
clear whether the 1,3-B-glucanase activity detected was associated with p50 or Mlgl.
Therefore, it is not known whether the gene product of EXGZ of C. carbonum is an active

exo-l,3-B-glucanase or not.

Materials and Methods
Isolation of EXGZ

PCR conditions for EXGZ amplification and cloning of the resulting PCR fiagment
have been described (Murphy and Walton 1996). Template DNA was isolated C.
carbonum genomic DNA and DNA extracted fi'om phage lysate of a C. carbonum cDNA
library (Pitkin et al. 1996). The PCR primers used were GTNGGNTTYAAYTGGGG and
CARAAYGGNTTYGAYAA (where N = any nucleotide, R = A or G, Y = C or T) which
correspond to the amino acid sequences VGFNWG and QNGFDN, respectively. The

final concentration of Mng in the reaction mixture was 1.5 mM. The reaction conditions

for 35 cycles were: denaturation at 94°C for 1 min, annealing at 50°C for 2 min, and
primer extension at 72°C for 3 min. The PCR products were separated on a 1.2% agarose
gel and putative EXGZ-containing PCR fragments were excised fiom the gel and cloned in

the pBluescriptH SK+ at the SmaI restriction site as described (Murphy and Walton 1996).

Screening of the genomic and the cDNA libraries, genomic DNA isolation, DNA

86

APPENDIX A

blotting, probe labeling, and hybridization protocols have been described (Scott-Craig et
al. 1990, Pitkin et al. 1996, Sposato et al. 1995). DNA was blotted to Nytran (Schleicher
and Schuell, Keene, New Hampshire).

EYGZ sequence was generated by automated fluorescent sequencing at the MSU-

DOE-PRL Plant Biochemistry Facility.

APPENDIX B

APPENDIX B

CLONING OF THE GENE ENCODING A l,3-l,4-B-GLUCANASE,
MLGZ, FROM COCHLIOBOL US CARBON UM

Introduction

As redundancy contributes to the problem of interpreting gene disruption experiments
for any one enzyme activity, each overlapping activity should be identified, the
corresponding gene cloned, and a mutant generated lacking all overlapping activities.
Apel et al. (1996) and Murphy and Walton (1996) have taken this approach with xylanases
and proteases, respectively, from Cochliobolus carbonum. In addition, all pectin
hydrolyzing activities are being characterized fi'om C. carbonum with the ultimate goal of
a mutant deficient in pectin-degrading enzymes (Scott-Craig et al. 1990, 1996). As there
are two detectable l,3-1,4-B-glucanase activities secreted by C. carbonum (Gorlach et al.
1997a), it is necessary to clone the corresponding genes and generate null mutants to
eliminate all 1,3-1,4-B-glucanases. The major l,3-1,4-B-glucanase has already been
cloned and mlg] mutants generated (Gorlach et al. 1997a). In this study, a cDNA
encoding the second 1,3-1,4-B-glucanase (Mlg2) was cloned, sequence was obtained, and

the M62 locus was mapped.

87

88

APPENDIX B

Results and Discussion

Internal tryptic peptide sequence analysis of the purified Mlg2 protein yielded three
sequences (Figure 21) (Gorlach et al. 1997a). Mlg2 internal peptides 11 and III (Figure
21) shared homology with cellulases fiorn Streptomyces roche and Aspergillus aculeatus.
Utilizing the homologous sequences as a reference, the location of peptides H (starting at
amino acid 170, Figure 22) and III (starting at amino acid 231, Figure 22) in the Mlg2
protein were determined. Using DNA from a C. carbonum cDNA library as a template,

PCR primers corresponding to the amino

Peptide I. FTVNQ(C)SANAY
Peptide II. YDVYPIGSSQGMVNVAGR

Peptide III. GFPINSQNLITYQFGTEAFTGGP
Figure 21. Sequences of NIng tryptic peptide fragments. The identity of the amino acid

in parentheses was uncertain. Single-underlining indicates the location of the amino acids
used to design PCR primers for amplification of the MGZ gene.

acid sequences QGMVNV and ITYQFG were used to amplify a DNA fragment
containing the gene encoding Mlg2. The deduced amino acid sequence of the resulting
200 bp-cDNA-derived fragment had the peptide sequences used to make both primers and
flanking sequences of peptides H and III (compare Figure 21 and 22).

A C. carbonum cDNA library was screened using the 200-bp PCR fi'agment as a

C. carbonum
A. aculeatus
S. lividans
C. carbonum
A. aculeatus
S. lividans
C. carbonum
A. aculeatus
S. lividans
C. carbonum
A. aculeatus
S. lividans
C. carbonum
A. aculeatus
S. lividans
C. carbonum
A. aculeatus
S. lividans
C. carbonum
A. aculeatus
S. lividans
C. carbonum
A. aculeatus
S. lividans

MRTLRPQARA
51
TYTGGVYTIN
TTIQGRYVVQ
101
RDNVKSYVYS
ENSVKSYANS
FNGCHYTNCS
151
NVNHPTSSGD
DINHVTWSGD
ARTDGV..NQ
201
KVYSFVTPSG
KTYSFVAP.T
DVLSFVAPSA

251

GGPAKFTVNQ7WSANAY*....

89

APPENDIX B

PRGLLAALGA

NNIWGRSSAT
NNLWGKDAGS
NNRWGSTAPQ

GPSNHKKPVS
GLTFNKKLVS
PGTDLPVRLD

YELMVWLGRY

.MKAFHLLAA
VLAAEALVSS
SGSQCTY...

GSQCTT...
CVTATDTGFR

QYSNLETEAY
QISQIPTTAR
TVSAAPSSIS

LAGAAVAQQA
LVTAAAPAQA
VNSVSQTGAK
VNSASSAGTS
VTQADGSAPT

WVYDTSNIRC
WSYDNTGIRA
YGFVDGAVYN

.DVYPIGSSQ GMVNVAGRQW

 

YELMIWLARY
TEIMIWFNRV

PIYNFKASMK
PITSFQGDVN
.ISGWSFDVM

 

 

GGPATLSVSN
NG.AGLAVNS

301

00......

SHNARIAPGG

WSASVQ*...
FSSTVETGTP

GTAPVDGWQL

SLSFGFQGTY

GGVQPIGSQI
GPIQPIGSPV

DFFQYLANNK
DFFKYLTQNH
DFVR.ATVAR

....GGTDPG

AFTLPSGQRI
GGAFAEPTGF

ATATVDGQTW
GTASVGGRTW

GFPINSQNLI

50
Q..LCDQYA.
DTTICEPFGT

100
WTSNWQWQGG
WSTKWNWSGG
NGAPKSYPSV
150
NVAYDLFTSA
DVAYDLFTAA
ASYDIWLDPT

200
DFFYGLNGNM
ELWYGANGSQ
EVWSAANGSN

250
TYQFGTEAFT

 

GFPASSQYLI
GLAENDWYLT

DPGGPSACAV

TNAWNASLTP

390

RLNGTACTTV*

TLQFGTEPFT
SVQAGFEPWQ

300

SYGTNVWQDG

SSGSVTATGA

Figure 22. Deduced amino acid sequence of Mlg2 compared to homologous protein
sequences. PILEUP (Devereux et al. 1984) of C. carbonum Mlg2, A. aculeatus FI-

CMCase (EMBL X52525, Ooi et al. 1990), and S. lividans CelB (GenBank U04629,
Wittmann et al. 1994) Single-underlining indicates tryptic peptide sequences from the
Mlg2 protein. Similar (.) and identical (:) amino acids are labeled. Stop codons are

designated by *.

APPENDIX B

probe. A partial 900-bp MGZ cDNA clone, C2- 1, was isolated and sequenced obtained
for one strand. A comparison of the deduced amino acid sequence of lVflg2 with other
proteins in the databases, indicated that Mlg2 was highly similar to cellulases from
Streptomyces lividans and Aspergillus aculeatus (Figure 22).

To map the MLGZ locus, C. carbonum genomic DNA was digested with various
restriction enzymes, size-separated on an 0.8% agarose gel, blotted, and hybridized with

the 200-bp PCR fi'agment. The resulting map is shown in figure 23.

 

 

1.0 kb
l—-I
B Sm B
1 fl 1
l l l
H P X H X

Figure 23. Restriction map of the M62 locus showing the location of the MGZ
transcript (shaded box). P, PstI; Sm, SmaI; X, XhoI; B, BamI-II; H, HindIII.

Materials and Methods
Nucleic acid manipulation

Genomic DNA was isolated as described by Pitkin et al. (1996). The cDNA library
was synthesized fi'om PolyA+-RNA isolated fiom C. carbonum grown on macerated corn

cell walls (Pitkin et al. 1996). Sequencing was performed by automated fluorescent

91

APPENDIX B

sequencing at the MSU-DOE-PRL Plant Biochemistry Facility.

Isolation ofA/[1.62

PCR conditions for the amplification of MGZ by PCR and the method used to clone
the resulting PCR fragment were as described by Murphy and Walton (1996). Template
DNA was isolated from phage lysate of a C. carbonum cDNA library (Pitkin et al. 1996).
The PCR primers used were CCRAAYTGRTNAGTRAT (where R = A or G, Y = C or T,

N = any nucleotide), corresponding to ITYGFG, and CARGGNATGGTNAAYGT,

corresponding to QGMVNV. The final concentration of MgCl; in the PCR reaction was

4 mM. The reaction conditions for 35 cycles were as follows: denaturation for 1 min at
94°C, annealing for l min at 55°C, and primer extension for 2 min at 72°C. The PCR
products were size separated on a 1.2% agarose gel and potential M02 PCR fragments
were excised from the gel and cloned into the SmaI site of pBluescript II SK+ as described
(Murphy and Walton 1996).

Screening of the genomic and the cDNA libraries, genomic DNA isolation, DNA
blotting, probe labeling, and hybridization protocols have been described (Scott-Craig et
al. 1990, Pitkin et al. 1996, Sposato et al. 1995). DNA was blotted to Nytran (Schleicher

and Schuell, Keene, New Hampshire).

BIBLIOGRAPHY

BIBLIOGRAPHY

. Ahn, J. -H., and J. D. Walton. 1996. Chromosomal organization of T 0X2, a
complex locus controlling host-selective toxin biosynthesis in Cochliobolus carbonum.
Plant Cell 8:887-897.

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

. Alexander, D., K. Lawton, S. Uknes, E. Ward, J. and Ryals. 1994. Defense-related
gene induction in plants. Pages 195-212 in: Genetic engineering, Vol. 16, J .K. Setlow,
ed. Plenum Press, New York.

. Allen, E. A., B. E. Hazen, H. C. Hoch, Y. Kwon, G. M. E. Leinhos, R. C. Staples,
M. A. Stumpf, and B. T. Terhune. 1991. Appressorium formation in response to
topographical signals by 27 rust species. Phytopathology. 81:323-331.

. Ansari, H., and L. Stevens. 1983. Purification and properties of two neutral
proteinases from Aspergillus nidulans. J. Gen. Microbiol. 129: 1637-1644.

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

. Apel-Birkhold, P. C. and J. D. Walton. 1996. Cloning, disruption, and
expression of two endo-Bl,4-xylanase genes, XYL2 and XYL3, from
Cochliobolus carbonum. Appl. Environ. Microbiol. 62.

. Bailey, B. A., J. F. D. Dean, and J. D. Anderson. 1990. An ethylene biosynthesis-
inducing endoxylanase elicits electrolyte leakage and necrosis in Nicotiana tabacum
cv. Xanthi leaves. Plant Physiol. 94: 1849-1854.

. Baker, C. J., and D. F. Bateman. 1978. Cutin degradation by plant pathogenic fungi.
Phytopathology 68: 1 577-1 584.

92

10.

ll.

12.

l3.

14.

15.

16.

17.

18.

19.

93

Ball, A. M., A. M. Ashby, M. J. Daniels, D. S. Ingram, and K. Johnstone. 1991.
Evidence for the requirement of extracellular protease in the pathogenic interraction of
Pyrenopeziza brassicae with oilseed rape. Physio]. Mol. Plant Pathol. 38: 147-161.

Ballance, D. J. 1991. Transformation systems for filamentous fungi and an overview
of fungal gene structure. Pages 1-29 in: Molecular Industrial Mycology; Systems and
Applications for Filamentous Fungi. S. A. Leong and R. M. Berka, eds. Marcel
Dekker, New York.

Basco, R. D., R. Cueva, E. Andaluz, and G. Larriba. 1996. In vivo processing of
the precursor of the major exoglucanase by KEX2 endoprotease in the Saccharomyces
cerevisiae secretory pathway. Biochim. Biophys. Acta 1310:110-118.

Bateman, D. F., and H. G. Basham. 1976. Degradation of plant cell walls and
membranes by microbial enzymes. Pages 316-355 in: Encyclopedia of Plant
Physiology, New Series, Vol. 4, R. Heitefiiss and PH. Williams, eds. Springer-Verlag,
NY.

Bateman, D. F., T. M. Jones, and O. C. Yoder. 1973. Degradation of corn cell
walls by extracellular enzymes produced by Helminthosporium maydis race T.
Phytopathology 63: 1523-1529.

Becker, A., A. Kleickmann, W. Arnold, and A. Pfihler. 1993. Analysis of
the Rhizobium meliloti eon/eon/exoL fragment: Eon shows homology to

excreted endo-B-1,3-1,4-glucanases and Eon resembles membrane proteins.
Mol. Gen. Genet. 238:145-154.

Boccara, M., A. Diola, M. Rouve, A. Kotoujansky. 1988. The role of individual
pectate lyases of Erwinia chrysanthemi strain 3937 in pathogenicity on saintpaulia
plants. Physio]. Mol. Plant Pathol. 30:95-104.

Boel, E., M.T. Hansen, 1. Hjort, I. Hoegh, & N.P. Fiil. 1984. Two difi‘erent types
of intervening sequences in the glucoamylase gene from Aspergillus niger. EMBO J.
3:1581-1585.

Bonants, P. J. M., P. F. L. Fitters, H. Thijs, E. den Belder, C. Waalwijk, and J.
W. D. M. Henl'ling. 1995. A basic serine protease fiom Paecilomyces lilacinus with
biological activity against Meloidogyne hapla eggs. Microbiology 141:77 5-7 84.

Bourbonnais, Y., J. Ash, M. Daigle, and D. Y. Thomas. 1993. Isolation and
characterization of S. cerevisiae mutants defective in somatostatin expression: cloning
and functional role of a yeast gene encoding an asparytl protease in precursor
processing at monobasic cleavage sites. EMBO J. 12:285-294.

20.

21.

22.

23.

24.

25.

26.

27.

28.

29.

3O

31.

94

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

Carpita, N. C. 1984. Cell wall development in maize coleoptiles. Plant Physiol.
76:205-212.

Carpita, N. C. 1996. Structure and biogenesis of the cell walls of grasses.
Annu. Rev. Plant Physiol. Plant Mol. Biol. 47:445-476.

Choi, G. H., D. M. Pawlyk, B. Rae, R. Shapira, and D. L. N uss. 1993. Molecular
analysis and overexpression of the gene encoding endothiapepsin, and aspartic
protease from Cophonectria parasitica. Gene 125:135-141.

Chomczynski, P., and N. Sacchi. 1987. Single step method of RNA extraction by

acid guanidinium thiocyanate-phenol-chloroforrn extraction. Anal. Biochem. 162: 156-
1 59.

Chumley, F. G., and B. Valent. 1990. Genetic analysis of melanin-deficient
nonpathogenic mutants of Magnaporthe grisea. Mo]. Plant-Microbe Interact. 3:13 5-
143.

Ciufl'etti, L. M., M. R. Pope, L. D. Dunlde, J. M. Daly, and H. W. Knoche. 1983.
Isolation and structure of an inactive product derived fi'om the host-selective toxin
produced by Helminthosporium carbonum. Biochemistry 22:3 507 -35 10.

Clark, D. R., J. Johnson, Jr., K. H. Chung, and S. Kirkwood. 1978.
Purification, characterization, and action-pattem studies on the endo-(1—)3)-B-
D-glucanase from Rhizopus arrhizus QM1032. Carbohydrate Res. 61:457-
477.

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

Cooper, R. M. 1983. The mechanisms and significance of enzyrnic degradation of
host cell walls by parasites. Pages 101-135 in: Biochemical Plant Pathology, J .A.
Callow, ed. John Wiley & Sons, NY.

. Cooper, R. M. 1984. The role of cell wall-degrading enzymes in infection and

damage. In: Plant Diseases: Infection, Damage, and Loss. pp. 13-21. R. K. S. Wood
and G. J. Jellis (eds). Blackwell Scientific Publications, Oxford, UK.

Cooper, R. M., and R. K. S. Wood. 1980. Cell wall degrading enzymes of vascular
wilt fungi. III. Possible involvement of endo-pectin lyase in Verticillium wilt of
tomato. Physiol. Plant Pathol. 16:285-300.

32.

33.

34.

35.

36.

37.

38.

39.

40.

41.

42.

43.

95

C6té, F., and M. G. Hahn. 1994. Oligosaccharins: structures and signal transduction.
Plant Mol. Biol. 26: 1379-141 1.

Dahler, G. S., F. Dams, and N. T. Keen. 1990. Cloning of genes encoding
extracellular metalloproteases fi'om Erwinia chrysanthemi EC 16. J. Bacteriol.
172:5803-5815.

Davidow, L. S., M. M. O’Donnell, F. S. Kaczmarek, D. A. Pereira, J. R.
DeZeeuw, and A. E. Franke. 1987. Cloning and sequencing of the alkaline
extracellular protease gene of Yarrowia lipolytica. J. Bacteriol. 169:4621-4629.

Davey, J., K. Davis, Y. Imai, M. Yamamoto, and G. Matthews. 1994. Isolation
and characterization of krp, a dibasic endopeptidase required for cell viability in the
fission yeast Schizosaccharomyces pombe. EMBO J. 13:5910-5921.

Deising, H., R. L. Nicholson, M. Haug, R. J. Howard, K. Mendgen. 1992.
Adhesion pad formation and the involvement of cutinase and esterases in the
attachment of uredospores to the host cuticle. Plant Cell 4:1101-1111.

Deshpande, K. S., and K. B. Deshpande. 1968. Pectic enzyme production and
morphogenesis in Helminthosporium aopicum. Biol. Plant. (Praha) 10:245-250.

Devereux, J., P. Haeberli, and O. Smithies. 1984. A comprehensive set of sequence
analysis programs for the VAX. Nucleic Acids Res. 12:3 87-3 95.

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

Doan, D. N. P., and G. B. Fincher. 1992. Difi‘erences in the thermostability
of barley (1—)3, 1—)4)-beta-glucanases are only partly determined by N-
glycosylation. FEBS Lett. 309:265-271.

Dori, S., Z. Sole], and I. Barash. 1995. Cell wall-degrading enzymes produced by
Gaeumannomyces graminis var. tritici in vitro and in vivo. Physio]. Mol. Plant Pathol.
46: 189-198

Dow, J. M., B. R. Clarke, D. E. Milligan, J. -L. Tang, and M. J. Daniels. 1990.
Extracellular proteases from Xanthomonas campestris pv. campestris, the black rot
pathogen. Appl. Environ. Microbiol. 56:2994-2998.

Drapeau, G. R. 197 8. Role of metalloprotease in activation of the precursor of
Staphylococcal protease. J. Bacteriol. 136:607-613.

. Easton, C. J., and S. Rossall. 1985. The production of certain cell wall-degrading

enzymes by Leptosphaeria maculans in culture and in stem canker lesions of oilseed
rape. Physiol. Plant Pathol. 26: 185-197.

45.

46.

47.

48.

49.

50.

51.

52.

53.

54.

55.

96

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

Ferrer, P., T. Halkier, L. Hedegaard, D. Savva, L. Diers, & J. A. Asenjo. 1996.
Nucleotide sequence of a B-1,3-glucanase isoenzyme H A gene of Oersokovia

xanthineolytica LL G109 (Cellulomonas cellulans) and initial characterization of the
recombinant enzyme expressed in Bacillus subtilis. J. Bacteriol. 178:4751-4757.

Flint, H. J., J. Martin, C. A. McPherson, A. S. Daniel, and J.-X. Zhang.
1993. A bifunctional enzyme, with separate xylanase and B(1,3-1,4)-glucanase
domains, encoded by the xynD gene of Ruminococcus flavefaciens. J.
Bacteriol. 175: 2943-2951.

Frohman, M. A., M. K. Bush, and G. R. Martin. 1988. Rapid production
of full-length cDNAs from rare transcripts: Amplification using a single gene-
specific oligonucleotide primer. Proc. Natl. Acad. Sci. USA 85:8998-9002.

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

Gebhard, W., H. Tschesche, and H. Fritz. 1986. Biochemistry of aprotinin and
aprotinin-like inhibitors. Pages 375-388 in: Proteinase Inhibitors, A.J. Barrett and G.
Salvesen, eds. Elsevier Science, Amsterdam.

Geremia, R. A., G. H. Goldman, D. Jacobs, W. Ardiles, S. B. Vila, M. Van
Montague, and A. Herrera-Estrella. 1993. Molecular characterization of the
proteinase-encoding gene, prb], related to mycoparasitism by T richoderma
harzianum. Mol. Microbiol. 8:603-613.

Gilkes, N. R., B. Henrissat, D. G. Kilburn, R. C. Miller, Jr., and R. A. J.
Warren. 1991. Domains in microbial B-l,4-glycanases: Sequence
conservation, function, and enzyme families. Microbiol. Rev. 55:303-315.

Gish, W., and D. J. States. 1993. Identificationof protein coding regions by
database similarity search. Nature Genet. 3:266-272.

Gbrlach, J. M., E. van der Knaap, and J. D. Walton. 1997a. Three mixed-linked
glucanases fiom the filamentous fungus Cochliobolus carbonum: cloning and targeted
disruption of [MLGI (submitted).

Giirlach, J. M., J. W. Pitkin, and J. D. Walton. 1997b. Investigation of the
involvement of two proteases in virulence of the fungal pathogen, Cochliobolus
carbonum, on maize. (in preparation).

56.

57.

58.

59.

60.

61

62.

63.

65.

66.

67.

97

Grenier, D. 1996. Degradation of host protease inhibitors and activation of
plasminogen by proteolytic enzymes from Porphyromonas gingivalis and T reponema
denticola. Microbiol. 142:955-961.

Gunkel, F. A., and H. G. Gassen. 1989. Proteinase K from T ritirachium album
Lirnber. Characterization of the gene and expression of the cDNA in Escherichia coli.
Biochem. 179: 185-194.

Gurr, S. J., S. E. Unkles, and J. R. Kinghom. 1987 . The structure and organization

of mucear genes of filamentous fungi. Pages 93-139 in: Gene Structure in Eukaryotic
Microbes. J .R. Kinghorn, ed. IRL Press, London.

Hames, B. D., and D. Rickwood, D. (eds.) 1981. Gel electrophoresis, a practical
approach. IRL Press, Oxford.

Henrissat, B., and A. Bairoch. 1993. New families in the classification of
glycosyl hydrolases based on amino acid sequence similarities. Biochem. J.
293:781-788.

. Hislop, E. C., J. L. Paver, and J. P. R. Keon. 1982. An acid protease produced by

Moniliniafluctigena in vitro and in infected apple fi'uits, and its possible role in
pathogenesis. J. Gen. Microbiol. 128:799-807.

Haj, P. B., E. B. Rodriguez, R. V. Stick, B. A. Stone. 1989. Difl‘erences in
active site structure in a family of B-glucan endohydrolases deduced from the
kinetics of inactivation by epoxyalkyl B-oligoglucosides. J. Biol. Chem.
264:4939-4947.

Holden, F. R., and J. D. Walton. 1992. Xylanases fi'om the fungal maize pathogen
Cochliobolus carbonum. Physio]. Mol. Plant Pathol. 40:39-47.

. Howard, R. J., and M. A. Ferrari. 1989. Role of melanin in appressorium function.

Exp. Mycol. 13:403-418.

Howard, S. P., and J. T. Buckley. 1985. Activation of the hole-forming toxin
aerolysin by extracellular processing. J. Bacteriol. 163:336-340.

Huang, H. C., and R. D. Tinline. 1976. Histology of Cochliobolus sativus infection
in subcrown intemodes of wheat and barley. Can. J. Bot. 54: 1344-1354.

Hynes, M. J., C. M. Corrick, and J. A. King. 1983. Isolation of genomic clones
containing the amdS' gene of Aspergillus nidulans and their use in the analysis of
structural regulatory mutations. Mol. Cell Biol. 3: 1430-1439.

68.

69.

70.

71.

72.

73.

74.

75.

76.

77.

78.

79.

98

Ikeda, M., T. Yaginuma, M. Kobayashi, and O. Yamashita. 1991. cDNA cloning,
sequencing and temporal expression of the protease responsible for vitellin degradation
in the silkworm, Bombyx mori. Comp. Biochem. Physiol. 99B:405-411.

Jennings, P. R., and A. J. Ullstrup. 1957. A histological study of three
Helminthosporium leaf blights of corn. Phytopathology 47 :707-7 14.

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

Julius, D., A. Brake, L. Blair, R. Kunisawa, and J. Thorner. 1984. Isolation of the
putative structural gene for the lysine-arginine-cleaving endopeptidase required for
processing of yeast prepro-a-factor. Cell 37: 1075-1089.

Keitel, T., S. Ortwin, R. Borriss, and U. Heinemann. 1993. Molecular and
active-site structure of a Bacillus l,3-1,4-B-glucanase. Proc. Natl. Acad. Sci.
USA 90:5287-5291.

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

Komano, H, and R. S. Fuller. 1995. Shared firnctions in vivo of a glycosyl-
phosphatidylinositol-linked aspartyl protease, Mkc7, and the proprotein processing
protease Kex2 in yeast. Proc. Natl. Acad. Sci. USA 92: 107 52-10756.

Kubo, Y., H. Nakamura, K. Kobayashi, T. Okuno, and I. Furusawa. 1991.
Cloning of a melanin biosynthetic gene essential for appressoria] penetration of
Colletotrichum Iagenarium. Mo]. Plant-Microbe Interact. 4:440-445.

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

Laio, C. -H., H. -Y. Hung, A. K. Chatterjee. 1988. An extracellular pectate lyase is
the pathogenicity factor of the soft-rotting bacterium Pseudomonas viridiflava. Mo].
Plant-Microbe Interact. 1:199-206.

Larcher, G., B. Cimon, F. Symoens, G. Tronchin, D. Chabasse, and J. -P.
Bouchara. 1996. A 33 kDa serine proteinase fi'om Scedosporium apiospermum.
Biochem. J. 315:119-125.

Lawton, M. A., and C. J. Lamb. 1987. Transcriptional activation of plant defense
genes by firngal elicitor, wounding, and infection. Mol. Cell. Biol. 7:335-341.

80.

81.

82.

83.

84.

85.

86.

87.

88.

89.

90.

91.

92.

99

Le Cam, B., P. Massiot, and F. Rouxel. 1994. Cell wall Polysaccharide—degrading
enzymes produced by isolates of Mycocentrospora acerina differing in aggressiveness
on carrot. Physio]. Mol. Plant Pathol. 44: 187-198.

Leonard, K. J. 1978. Polymorphisms for lesion type, fimgicide tolerance, and mating
capacity in Cochliobolus carbonum isolates pathogenic to corn. Can. J. Bot. 56:1809-
1 81 5.

Leonard, K. J. 1987. The host populations as a selective factor. In: Populations of
Plant Pathogens: Their Dynamics and Their Genetics. pp. 163-179. M. S. Wolfe and
C. E. Caten (eds). Blackwell Sci. Publ., Oxford.

Lever, M. 1972. A new reaction for colorimetric determination of
carbohydrates. Anal. Biochem. 47:273-279.

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

Maniatis, T., E. F. Fritsch, and J. Sambrook. 1982. Molecular Cloning: A
Laboratory Manual. Cold Spring Harbor Laboratory, Cold Spring Harbor, NY.

Matoba, S., J. Fukayama, R. A. Wing, and D. M. Ogrydziak. 1988. Intracellular
precursors and secretion of alkaline extracellular protease of Yarrowia lipolytica. Mol.
Cell. Biol. 8:4904-4916.

Matsudaira, P. 1987. Sequence from picomole quantitites of proteins electroblotted
onto polyvinylidene difluoride membranes. J. Biol. Chem. 262: 1003 5-10038.

McNeil, M., A. G. Darvill, S. C. Fry, and P. Albersheim. 1984. Structure and
function of the primary cell walls of plants. Annu. Rev. Biochem. 53:625-663.

Meeley, R. B., and J. D. Walton. 1991. Enzymatic detoxification of HC-toxin, the
host-selective cyclic peptide fiom Cochliobolus carbonum. Plant Physiol. 97: 1080-
1086.

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

Meldgaard, M., and I. Svendsen. 1994. Difl‘erent effects of N-glycosylation
on the thermostability of highly homologous bacterial (l,3-1,4)-beta-
glucanases secreted from yeast. Microbiology 140:159-166.

Monod, M., S. Paris, J. Sarfati, K. Jaton-Ogay, P. Ave, and J. -P. Latgé. 1993.
Virulence of alkaline protease-deficient mutants of Aspergillus fumigatus. FEMS
Microbiol. Lett. 106:39-46.

100

93. Moormann, M., A. Schlochtermeier, and H. Schrempf. 1993. Biochemical
characterization of a protease involved in the processing of a Streptomyces reticuli
cellulase (Avicelase). Appl. Environ. Microbiol. 59:1573-1578.

94. Morita, S., M. Kuriyama, K. Maejima, and K. Kitano. 1994. Cloning and
nucleotide sequence of the alkaline protease gene from F usarium sp. S-19-5 and
expression in Sacchm'omyces cerevisiae. Biosci. Biotechnol. Biochem. 58:621-626.

95. Murao, S., R. Sakamoto, and M. Aral. 1988. Cellulases of Aspergillus
aculeatus. Meth. Enzymo]. 160:274-299.

96. Murphy, J. M., and E. E. Hood. 1993. Molecular basis for extensin size
heterogeneity in two maize varieties. Plant Mol. Biol. 21:885-893.

97. Murphy, J. M., and J. D. Walton. 1996. Three extracellular proteases from
Cochliobolus carbonum: cloning and targeted disruption of ALP] . Mol. Plant-
Microbe Interact. 9:290-297.

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

99. Muse, R. R., H. B. Couch, L. D. Moore, and B. D. Muse. 1972. Pectolytic and
cellulolytic enzymes associated with Helminthosporium leaf spot on Kentucky
bluegrass. Can. J. Microbiol. 18:1091-1098.

100.Muthukumar, G., S. -H. Suhng, P. T. Magee, R. D. Jewell, and D. A.
Primerano. 1993. The Saccharomyces cerevisiae SPR] gene encodes a sporulation-

specific exo-l,3-B-glucanase which contributes to ascospore thermoresistance. J.
Bacteriol. 175:386-394.

101.Nelson, O. E., and A. J. Ullstrup. 1961. The inheritance of pathogenicity in
Cochliobolus carbonum. Phytopathology 51:1288-1291.

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

103.Nelson, R. R., and R. T. Sherwood. 1968. Genetic control of
endopolygalacturonase production in Cochliobolus carbonum. Phytopathology
58: 1277-1280.

104.Neurath, H. 1984. Evolution of proteolytic enzymes. Science 224:3 50-3 57 .

105.Nielsen, H., J. Engelbrecht, S. Brunak, & G . von Heijne. 1997 . Identification of
prokaryotic and eukaryotic signal peptides and prediction of their cleavage sites.
Protein Engineer. 10 (in press).

lOl

106.Nikolskaya, A., J. W. Pitkin, H. J. Schaeffer, and J. D. Walton. 1996. (in
preparation).

107.North, M. J. 1982. Comparative biochemistry of the proteinases of eukaryotic
microorganisms. Microbiol. Rev. 46:308-340.

108.0lafson, R. W., L. Jurasek, M. R. Carpenter, and L. B. Smillie. 1975. Amino
acid sequence of Streptomyces griseus trypsin. Cyanogen bromide fragments and
complete sequence. Biochemistry 14:1168-1177.

109.001, T, A. Shinmyo, H. Okada, S. Hara, T. Ikenaka, S. Murao, and M.
Arai. 1990. Cloning and sequence analysis of a cDNA for cellulase (F I-
CMCase) from Aspergillus aculeatus. Curr. Genet. 18:217-222.

110.0ttensen, M., and I. Svendsen. 1970. The subtilisins. Page 199-215 in: Methods in
Enzymology: Proteolytic Enzymes, Vol. 19, GE. Perlrnann and L. Lorand, eds.
Academic Press, Orlando.

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

112.Perito, B., E. Hanhart, T. Irdani, M. Iqbal, A.J. McCarthy, & G. Mastromei.
1994. Characterization and sequence analysis of a Streptomyces rochei A2
endoglucanase-encoding gene. Gene 148:119-124.

113.Phadatare, S. U., M. C. Srinivasan, and V. V. Deshpande. 1992. Evidence for
controlled autoproteolysis of alkaline protease: a mechanism for physiological

regulation of conidial discharge in Conidiobolus coronatus. Eur. J. Biochem. 205:679-
686.

114.Phadatare, S., M. C. Srinivasan, and M. Deshpande. 1989. Evidence for the
involvement of serine protease in the conidial discharge of Conidiobolus coronatus.
Arch. Microbiol. 153:47-49.

115.Pitkin, J. W., D. G. Panaccione, and J. D. Walton. 1996. A putative cyclic peptide
efilux pump encoded by the T 0264 gene of the plant-pathogenic fiingus Cochliobolus
carbonum. Microbiology 142: 1557- l 565.

116.Planas, A., M. Juncosa, J. Lloberas, and E. Querol. 1992. Essential
catalytic role of Glum in endo-B-1,3-1,4-D-glucan 4-glucanohydrolase from B.
Iicheniformis as determined by site-directed mutagenesis. FEBS Lett. 308: 141-
145.

102

117 .Pope, M. R., L. M. Ciuffetti, H. W. Knoche, D. McCrery, J. M. Daly, and L. D.
Dunkle. 1983. Structure of the host-specific toxin produced by Helminthosporium
carbonum. Biochemistry 22:3502-3 506.

118.Porter, F. M. 1966. Protease activity in diseased fiuits. Phytopathology. 56:1424-
1425.

119.Power, S. D., R. M. Adams, and J. A. Wells. 1986. Secretion and autoproteolytic
maturation of subtilisin. Proc. Natl. Acad. Sci. USA 83:3096-3100.

120.Powers, J. C., and E. A. Harper. 1986. Inhibitors of serine proteases. Pages 55-152
in: Protease Inhibitors, A.J. Barrett and G. Salvesen, eds. Elsevier Science,
Amsterdam.

121.Pringle, R. B., and R. P. Schefl'er. 1967. Isolation of the host-specific toxin and a
related substance with nonspecific toxicity from Helminthosporium carbonum.
Phytopathology 57:1 169-1 172.

122.Prograrn Manual for the Wisconsin Package 1994. Version 8, Genetics Computer
Group, University of Wisconsin, Madison, Wiscosin, U.S.A.

123.Ransom, R., and J. D. Walton. 1996. Purification and characterization of
extracellular B-xylosidase and a-arabinosidase from the plant pathogenic fungus
Cochliobolus carbonum. Carbohyd. Res. (in press).

124.Rasmussen, J. B., and R. M. Hanau. 1989. Exogenous scytalone restores
appressoria] melanization and pathogenicity in albino mutants of Colletotrichum
graminicola. Can. J. Plant Pathol. 11:349-352.

125.Rasmussen, J. B., and R. P. Scheffer. 1988a. Effects of selective toxin fiom
Helminthosporium carbonum on chlorophyll synthesis in maize. Physio]. Mol. Plant
Pathol. 32:283-291.

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

127.Rast, D. M., H. Stiissi, H. Hegnauer, and L. E. Nyhlén. 1981. Melanins. In: The
Fungal Spore: Morphogenic Controls. pp. 507-531. G. Turian and H. R. Hohl (eds).
Academic Press, New York.

128.Rauscher, M., K. Mendgen, H. Deising. 1995. Extracellular proteases of the rust
firngus Uromyces viciae-fabae. Experiment. Mycol. 19:26-34.

129.Reddy, M. N ., D. L. Stuteville, and E. L. Sorensen. 1971. Protease production
during pathogenesis of bacterial leaf spot of alfalfa and by Xanthomonas alfalfae in
vitro. Phytopathology 61:361-365.

103

130.Ries, S. M., and P. Albersheim. 1973. Purification of a protease secreted by
Colletotrichum Iindemuthianum. Phytopathology 63:625-629.

131.Robertsen, B. 1984. An alkaline protease produced by Cladosporium cucumerinum
and its possible importance in the development of scab disease of cucumber seedlings.
Physiol. Plant Pathol. 24:83-92.

132.Rogers, D. T., D. Saville, and H. Bussey. 1979. Saccharomyces cerevisiae killer
expression mutant kex2 has altered secretory proteins and glycoproteins. Biochem.
Biophys. Res. Comm. 90:187-193.

133.Ryan, C.A. 1990. Protease inhibitors in plants: Genes for improving defenses against
insects and pathogens. Annu. Rev. Phytopathol. 28:425-429.

134.Rypniewski, W. R., S. Hastrup, C. Betzel, M. Dauter, Z. Dauter, G. Papendorl',
S. Branner, and K. S. Wilson. 1993. The sequence and X-ray structure of the trypsin
fiom F usarium oxysporum. Protein Engineer. 6:341-348.

135.Sakellaris, H., J. M. Pemberton, and J. M. Manners. 1993.
Characterization of an endo-l,3(4)-B-D-glucanase gene fiom Cellvibrio
mixtus. FEMS Microbiol. Lett. 109:269-272.

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

137.San Segundo, P., J. Correa, C. R. Vazquez de Aldana, and F. del Rey. 1993.
SSGI, a gene encoding a sporulation-specific 1,3-B-glucanase in Saccharomyces
cerevisiae. J. Bacteriol. 175:3823-3837.

138.Sato, S., H. Suzuki, U. Widyastuti, Y. Hotta, and S. Tabata. 1994. Identification
and characterization of genes induced during sexual differentiation in
Schizosaccharomyces pombe. Curr. Genet. 26:31-37.

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

140.Sch§fer, W., D. Straney, L. Cuifl'etti, H. D. Van Etten, and O. C. Yoder. 1989.
One enzyme makes a fungal pathogen, but not a saprophyte, virulent on a new host
plant. Science 246:247-249.

141.Scheffer, R. P., and R. S. Livingston. 1984. Host selective toxins and their role in
plant diseases. Science 223: 17-21.

104

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

143.Schimming, S., W. H. Schwarz, and W. L. Staudenbauer. 1992. Structure
of the Clostridium thermocellum gene licB and the encoded B-1,3-1,4-
glucanase. Eur. J. Biochem. 204: 13-19.

l44.Scott-Craig, J. S., D. G. Panaccione, F. Cervone, and J. D. Walton. 1990.
Endopolygalacturonase is not required for pathogenicity of Cochliobolus carbonum
on maize. Plant Cell 2:1191-1200.

145.Scott-Craig, J. S., F. Cervone, G. de Lorenzo, and J. D. Walton. 1996. (in
preparation).

146.Sharon, A., Y. Fuchs, and J. D. Anderson. 1993. The elicitation of ethylene
biosynthesis by T richoderma xylanase is not related to cell wall degradation activity of
the enyzrne. Plant Physio]. 102: 1325-1329.

147 .Showalter, A. M. 1993. Structure and function of plant cell wall proteins. Plant Cell
5:9-23.

148.Slakeski, N ., D.C. Baulcombe, K.M. Devos, D.N.P. Doan, & G.B. Fincher. 1990.
Structure and tissue-specific regulation of genes encoding barley (1 3,1 4)-B-g]ucan
endohydrolases. Mol. Gen. Genet. 224:437-449.

149.Spilliaert, R., G. O. Hreggvidsson, J. K. Kristjansson, G. Eggertsson,
and A. Palsdottir. 1994. Cloning and sequencing of a Rhodothermus marinus
gene, bglA, coding for a therrnostable B-glucanase and its expression in
Escherichia coli. Eur. J. Biochem. 224:923-930.

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

151.St. Leger, R. J. 1995. Role of cuticle-degrading proteases in fimgal pathogenesis of
insects. Can. J. Bot. 73:S1119-81125.

152.St. Leger, R. J., D. C. Frank, D. W. Roberts, and R. C. Staples. 1992. Molecular
cloning and regulatory analysis of the cuticle-degrading protease structural gene for

the entomopathogenic fungus Metarhizium anisopliae. Eur. J. Biochem. 204:991-
1001.

153.Staples, R. C. and V. Macko. 1980. Formation of infection structures as a
recognition response in fungi. Exp. Mycol. 4:2-16.

105

154.Str6mqvist, M., and H. Grufl‘man. 1992. Periodic acid/Schifl‘ staining of
glycoproteins immobilized on a blotting matrix. BioTechniques 13:744-746.

155.Tang, C. M., J. Cohen, T. Krausz, S. Van Noorden, and D. W. Holden. 1993.
The alkaline protease of Aspergillus flimigatus is not a virulence determinant in two
murine models of invasive pulmonary aspergillosis. Infect. Irnmun. 61: 1650-1656.

156.Teather, R., and J. D. Erl'le. 1990. DNA sequence of a Fibrobacter
succinogenes mixed linkage B-glucanase (1,3-l,4-B-D-glucan 4-
glucanohydrolase) gene. J. Bacteriol. 172:3837-3 841.

157.Tseng, T. C., and M. 8. Mount. 1974. Toxicity of endopolygalacturonate trans-

eliminase, phosphatidase, and protease to potato and cucumber tissue. Phytopathology
64:229-236.

158.Ullstrup, A. J. 1941. Two physiologic races ofHelminthosporium maydis in the corn
belt. Phytopathology 57:1288-1291.

159.van den Ackerveken, G. F. J. M., P. Vossen, and P. J. G. M. De Wit. 1993. The
AVR9 race-specific elicitor of Cladosporiumfidvum is processed by endogenous and
plant proteases. Plant Physiol. 103:91-96.

l60.van Hoof, A., J. Leykam, H. J. Schaeffer, and J. D. Walton. 1991. A single 81,3-
glucanase secreted by the maize pathogen Cochliobolus carbonum acts by an exolytic
mechanism. Physio]. Mol. Plant Pathol. 39:259-267.

161.von Heijne, G. 1986. A new method for predicting signal sequence cleavage sites.
Nuc. Acids Res. 14:4683-4690.

162.Walsh, K. A., and H. Neurath. 1964. Trypsinogen and chyrnotrypsinogen as
homologous proteins. Proc. Natl. Acad. Sci. USA. 52:884-889.

163.Walsh, K. A., and P. E. Wilcox. 1970. Serine proteases. Pages 31-63 in: Methods
in Enzymology: Proteolytic Enzymes, Vol. 19, G.B. Perlrnann and L. Lomad, eds.
Academic Press, Orlando.

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

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

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

167.Walton, J. D., and F. Cervone. 1990. Endopolygalacturonase from maize pathogen
Cochliobolus carbonum. Physio]. Mol. Plant Pathol. 36:351-359.

106

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

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

170.Walton, J. D., E. D. Earle, H. Stihelin, A. Grieder, A. Hirota, and A. Suzuki.
1985. Reciprocal biological activities of the cych tetrapeptides chlamydocin and HC
toxin. Experientia 41:348-3 50.

17 1.Wijesundera, R. L. C., J. A. Bailey, R. J. W. Byrde, and A. H. Fielding. 1989.
Cell wall degrading enzymes of Colletotrichum lindemuthianum: their role in the
development of bean anthracnose. Physio]. Mol. Plant Pathol. 34:403-413.

172.Woloshuk, C. P., H. D. Sisler, M. C. Tokousbalides, and S. R. Dutky. 1980.
Melanin biosynthesis in Pyricularia oryzae: site of tricyclazole inhibition and
pathogenicity of melanin deficient mutants. Pestic. Biochem. Physiol. 14:256-264.

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

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

175.Yun, S.J., D.J. Martin, B.G. Gengenbach, H.W. Rines, and D.A. Somers. 1993.
Sequence of a (1-3, 1-4)-B-g]ucanasc cDNA from oat. Plant Physiol. 103:295-296.

176.Xu, H., and K. Mendgen. 1997. Targeted cell wall degradation at the penetration
site of cowtpea rust basidiosporelings. Mo]. Plant-Microbe Interact. 10:87-94.