 

‘.
;.
a
f"
5'
" ﬁt»
"
f
-I
-|
Y‘
'_ a
‘.|
1 ,
C
I.V’
‘7
‘ _L
.,;a
I‘:
J.
9o‘5n
1. g
x-y'n'

 

 

I.”
a RI
:9 i".
7“" 1*.
§g.r;
9. .;
.;_ 7‘
‘7 .
l‘ 1",
‘l ‘2'!
x ,
v H;
> I
. n
r!~

 

 

 

 

 

‘
f

l

x

5
15 n

‘ E}

'1 .
. ‘ _c
SUI-h! ‘

a:

1' 1'5?

1 .
6" ,.
‘L

[M

it.

(I: :-
g1
‘ t
a '1
:t

".r
“f

z ‘5‘-

ma

3

\
95-

JV
{2

‘. _ 7‘

N" “a" "by: ﬁn.
“LL-2‘ '1 ’ f a”.
sif‘g- ‘iﬂ:

.

.. 'lt/"p‘" 7‘.
‘1‘".
u,-

l
m

m 5" -. .
419.33%!
" n p.

f

:0

5. u
r

la

 

,4;
/¢‘7
'THS

 

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

 

DATE DUE DATE DUE DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

5108 K:IProj/Aoc&Pros/ClRC/OetoDuo.indd

COMPARISON OF COLLETOTRICHUM SPP. CAUSING CROWN ROTTING
ANTHRACNOSE TO COLLETOTRICHUM GRAMINICOLA AND
COLLETOTRICHUM SUBLINEOLUM ISOLATES USING ISOZY ME MARKERS

BY

Brandon Joseph Horvath

A THESIS

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

MASTER OF SCIENCE

Department of Botany and Plant Pathology

1999

ABSTRACT

COMPARISON OF COLLETOTRICHUM SPP. CAUSING CROWN ROTTING
ANTHRACNOSE TO COLLETOTRICHUM GRAMINICOLA AND
COLLETOTRICHUM SUBLINEOLUM ISOLATES USING ISOZY ME MARKERS

By

Brandon Joseph Horvath

Crown rotting anthracnose (CRA) caused by Colletotrichum graminicola, is
a serious problem on annual bluegrass (Poa annua) and creeping bentgrass
(Agrostis pelustn's) under high maintenance conditions. The objectives for this
study were: 1) determine the genetic relatedness of isolates of C. graminicola
from annual bluegrass and creeping bentgrass, and 2) compare the genetic
relatedness of isolates from annual bluegrass and creeping bentgrass to isolates
from two economically important hosts, corn (Zea mays) and sorghum (Sorghum
spp.) using isozyme genetic markers. Our results show that isolates from annual
bluegrass and creeping bentgrass are related at the sub-species level, and that
isolates from the turfgrass hosts are more distantly related to the crop hosts, corn
and sorghum. These results are also consistent with a sympatric speciation
system where genetic isolation occurs because of niche specialization. These
results question the use of this pathogen as a biological control agent for annual

bluegrass.

Copyright by
Brandon Joseph Horvath
1 999

To my wife, Laura, whose undying love and support gives me the strength to
complete a project this large, and to my parents, Jean and Joe, who taught me

that if I put my mind to it, I can do anything.

ACKNOWLEDGMENTS

I wish to extend my thanks and appreciation to the multitude of people
who have helped me throughout my graduate career here at Michigan State
University. First, I would like to thank the members of my guidance committee,
Dr. Jarosz, for your valuable insights and getting me interested in epidemiology
and population genetics, and Dr. Hausbeck, for keeping me grounded in reality
and focused on my goal of ﬁnishing my thesis. My major professor, Dr. Vargas,
the thanks I owe you goes beyond a simple note. I feel like you have prepared
me to be an independent and resourceful researcher for which I will be eternally
grateful.

I would also like to extend my thanks to the people with whom I have
shared the lab with over the years. Nancy Dykema, your friendship and
discussions about science and life in general are a cherished part of my
experience and your help in solving problems that I had is greatly appreciated.
Dr. Jon Powell, our discussions about turfgrass pathology and science, and our
friendship are things that I hope to have well into the future. Phil Dwyer, your
friendship, wit, and humor are things that l have enjoyed throughout my time
here, and I know that you will be successful in whatever career you choose to
pursue. I am thankful to Ron Detweiler for his help in whatever area | asked for it
and for teaching me those “insider" tips that will help me solve problems in the
future. To the others that have been in the lab at one time or another, thank you

for your help, encouragement, and companionship.

To Karch, Goose, Rouge, Sorochan, Sharer, Jimmy, Susan, Beau, Thom,
Ron, Trey, Crum, and Dr. Rieke, I am especially grateful for all of our interactions
both professionally and personally without which this experience wouldn’t have
been as enjoyable nor as fulﬁlling as it has been. To my fellow graduate students
in the department, the support, camaraderie, and encouragement you have
provided is greatly appreciated.

A special note of thanks is due to Dr. David Douches and his students and
staff for welcoming a “turﬁe” into their lab. Without their support and assistance I
could not have completed my research. Dr. Scott Warnke, I am especially
indebted to you for teaching me the art and science of isozyme research.

Last, but certainly not least, I would like to thank those people to whom I
am closest. I would like to thank my Mom and Dad for their love and support
throughout my life in whatever I have chosen to do. Nate, your zest for life and
support will be a source of strength no matter where we end up. Finally, my wife
Laurie, through the good days and bad days you were there. No thank you could

ever be enough for the support you have provided me.

vi

TABLE OF CONTENTS

LIST OF TABLES ............................................................................................... viii
LIST OF FIGURES ............................................................................................... ix
Literature Review .................................................................................................. 1

Literature Cited .................................................................................................. 9
Chapter 1

Genetic comparison of crown rotting anthracnose isolates from amenity turfgrass
hosts to isolates from corn (Zea mays) and sorghum (Sorghum spp.) using

isozyme analysis ................................................................................................. 12
Introduction ...................................................................................................... 12
Materials and Methods .................................................................................... 16

Isolate Collection .......................................................................................... 16
Culture Preparation ...................................................................................... 16
Extraction and Electrophoresis .................................................................... 17
Visualization of enzyme activity ................................................................... 22
Genic Nomenclature .................................................................................... 22
Analysis of Data ........................................................................................... 23
Results ............................................................................................................ 24
Discussion ....................................................................................................... 29
Literature Cited ................................................................................................ 35

APPENDIX A ....................................................................................................... 38

APPENDIX B ....................................................................................................... 41

APPENDIX C ...................................................................................................... 42

vii

LIST OF TABLES

Table 1- Electrophoretic phenotypes (EP) of crown rotting anthracnose isolates
collected from various hosts and locations ................................................... 19

Table 2- Enzymes and buffer systems used for the detection of variation in
collected C. graminicola isolates. ................................................................. 21

Table 3- Nei’s corrected genetic distance (12) by host origin (A) and geographic
location (B). .................................................................................................. 26

Table 4- Allele frequencies for isozyme loci tested, subdivided by host origin and
geographic location ...................................................................................... 27

Table 5- Unique electrophoretic phenotypes (EP) of Colletotrichum isolates
sampled ....................................................................................................... 41

viii

LIST OF FIGURES

Figure 1- Figure showing the different alleles for the enzymes PGI (A), PGM (B),
and GOT (C). ............................................................................................... 42

Literature Review

Annual bluegrass (Poa annua L.) is one of the ﬁve most widely
disseminated weeds in the world (1) and is often troublesome in golf course turfs
because no control measures exist to selectively eliminate it from a mixed stand
of grasses (2). However, due to P. annua’s cosmopolitan nature and the lack of
effective controls, it is tolerated and managed as a turfgrass on golf courses.
There has been controversy over the life cycle of P. annua since 1965 when
Timm (3) separated P. annua into two distinct subgroups: an erect annual
subgroup, and a prostrate biennial or perennial subgroup. Johnson, et al. (4)
assert that Poa annua, as a species, is almost anything but annual. Others
report that “an endless array of genotypes can exist within the same golf green”
(5). The divergent reports in the literature suggest a very high level of genetic
diversity in P. annua populations. Because all of these studies have been based
on morphological and physical characteristics that may be plastic over time, the

true life cycle of P. annua is still in question.

Poe annua’s growth requirements as a turfgrass were ﬁrst characterized in
1937 (6). Its proliﬁc seedhead production and susceptibility to fungal pathogens
during periods of high temperature and humidity were also described (6). This
susceptibility is the primary obstacle to P.annua’s acceptance as a desirable
turfgrass, and suggests this turfgrass might be amenable to biological control.

Diseases that cause widespread damage on a single undesirable species in a

mixed stand of plants are often investigated as potential biological controls
because of their putative ability to selectively eliminate the undesirable species.
In the case of P. annua, crown rotting anthracnose (CRA), caused by
Colletotrichum graminicola (Ces.) G.W. Wils., has the potential to be a biological

control.

One reason for the lack of interest in biological control of grass species is
the fear that organisms used for biological control could severely damage
important non-target crop species. Therefore, a good biological control agent is
characterized by its host speciﬁcity. The literature presents a conﬂicting account
of the host speciﬁcity of C. graminicola. This pathogen infects many grass
species including, maize (Zea mays L.), sorghum (Sorghum spp.), creeping
bentgrass (Agrostis palustn's Huds.), and annual bluegrass (P. annua). Different
species of turfgrass from several geographic locations in Canada were
inoculated with an isolate of C. graminicola from P. annua and only the Poe
species was infected (16). Another study with 381 isolates of C. graminicola
from six hosts (primarily Sorghum spp., and sugarcane) revealed that isolates
only infected the host from which they were isolated (17). An evaluation of C.
graminicola as a possible biological control for johnsongrass (Sorghum
halepense (L.) Pers.) determined that isolates from johnsongrass were severely
damaging to all Sorghum spp. tested, but that other graminaceous hosts were
not damaged (18). These reports each demonstrate that host speciﬁcity may
exist in the C. graminicola pathosystem. However, Jamil and Nicholson (19)

showed that, depending on a plants physiological condition, isolates from other

hosts can be as damaging as isolates from the same host. This account shows

that host speciﬁcity may be affected by external factors such as plant health.

CRA was ﬁrst reported in 1954 in Great Britain, and symptoms occurring
on infected turfgrass were described as “measly” and “piebald” (7). The disease
manifests itself as small patches of chlorotic, dying turf that may increase in
diameter to 15 cm or more. Initially, older leaves are discolored, but eventually
the central leaf becomes discolored and plant death usually follows (8). The
detached crowns of infected plants appear black and rotted, and numerous

aoervuli are apparent upon close examination.

CRA is often associated with stressful growth conditions such as low
mowing heights, reduced N fertility, and wet soils. A possible explanation for this
association is that the onset of pathogenesis is triggered by stress-induced
signals from the plant. C. graminicola belongs to a group of Colletotrichum
species that exhibit hemibiotrophic disease cycles (9). Pathogens in this group
penetrate the host without triggering the plant defense mechanisms and remain
latent until the plant is stressed (9). Ethylene production by the plant has been
reported as a potential signal for pathogen invasion in this group of
hemibiotrophs (10,11). Ethylene production is triggered by wounding and other
mechanical stresses (12). The production of ethylene in older tissues has also
been shown to cause leaf senescence (14). In the 1980’s, Vargas suggested
that the death of P. annua was related to several factors described as HAS
(ﬂelminthosporium, Anthracnose, Senescence) decline (15). Coleman and

Hodges (12) found that when leaves of Kentucky bluegrass (Poa pratensis) were

wounded, ethylene production increased, and spores of Bipolaris sorokiniana
germinated and produced penetration structures on glass cover slides when the
spores were exposed to the wounded leaves. Maize plants can also be
predisposed to stalk rot as a result of wounding stress (13). Such predisposition
to infection by wounding or other stresses is hypothesized to play a similar role in

P. annua.

The current taxonomy of Colletotrichum spp. occurring on grass hosts is
not well deﬁned. Isolates that occur on graminaceous hosts are commonly
assigned to C. graminicola. The characterization of Colletotrichum spp. using
molecular techniques has greatly facilitated the delineation of many species in
this genus and these techniques have become the method of choice to identify
species that are difﬁcult to distinguish morphologically (20,21 ,22,23,24,25).
Isozyme analysis, Randomly Ampliﬁed Polymorphic DNAs (RAPDs), Restriction
Fragment Length Polymorphisms (RFLPs) and DNA sequence analysis have
been used to examine genetic diversity in the C. graminicola pathosystem
(20,21,22,23,24,25,26). Molecular techniques are particularly useful in this
genus because the morphological characteristics used for identiﬁcation (e.g.
conidia, perithecia, ascus, and ascospore sizes; setae length; and appressorium
diameter) have been demonstrated to be variable. Thus, without the use of
molecular techniques, meaningful identiﬁcations are extremely difﬁcult to obtain

(22.23.24).

The relationships between isolates of C. graminicola from different grass

hosts have been examined in maize (Zea mays) and sorghum (Sorghum bicolor)

(22,23). Taxonomically, these two hosts are very closely related, both belonging
to the family Andropogoneae, subfamily Panicoideae (27). Therefore, isolates of
C. graminicola infecting these hosts could also be closely related. However,
Hugenin found large differences in zymograms between isolates from maize,
sorghum, and sugarcane, but because the isozymes used were highly variable, it
was impossible to make inferences about the taxonomy of these isolates (23). In
another study, isolates from maize and sorghum from various geographic regions
were differentiated using RAPDs, and the proﬁles were separated according to
geography (21 ). The authors concluded that C. graminicola exhibits a high level
of genetic variation (21). Recently, Vaillancourt and Hanau conducted a
comprehensive study demonstrating the genetic separation of isolates from
maize and sorghum (22). Data generated from crossing studies, RAPDs, and
RFLPs of mtDNA demonstrated that isolates from the two hosts could not
interbreed, and RAPDs showed only 45% similarity between isolates from
different host origins (22). These data support the reclassiﬁcation of isolates of
C. graminicola from sorghum (Sorghum bicolor) to a new species (22). Sherriff,
et al. conﬁrmed these results using the ribosomal DNA internal transcribed
spacer region (ITS-1) sequence and suggested C. sublineolum as the name for
isolates from sorghum (28). This is signiﬁcant because isolates from corn and
sorghum have been shown to be distinct genetic lineages that are reproductively
isolated. Therefore, isolates from corn and sorghum cannot share genetic

information such as genes for pathogenicity on other hosts. Isolates from maize

and sorghum can now be considered sibling species that are deﬁned as being

morphologically nearly identical, but reproductively completely isolated (29).

A single isolate from P. annua was recently included in a study of
Colletotrichum species relationships using ITS-1 sequencing. The ITS-1
sequence from the P. annua isolate was nearly identical to sequences found in
isolates from Sorghum (20). This suggests that isolates from P. annua may be
accommodated under C. sublineolum. Backman (24) found signiﬁcant variation
between isolates from A. palustris and P. annua using RAPD markers. Cluster
and principal coordinate analyses grouped isolates from the same host together
(24). Browning, Rowley, Zeng, Chandlee, and Jackson (30) also studied
morphological and RAPD variation present in isolates of C. graminicola from P.
annua and A. palustn's. They found that morphological comparisons were useful
in distinguishing between isolates from the two hosts. They also found that
RAPD ﬁngerprints from isolates from A. palustn's and P. annua were very
different from those generated by the isolates from Z. mays and S. bicolor.
Because of the sampling limitations of these studies, determining what the actual
relationship is between isolates from the amenity turfgrasses and isolates from

corn and sorghum is difﬁcult to ascertain.

The relationship of isolates from amenity turfgrasses to isolates from
maize and sorghum has not been a subject of much study. Because the
turfgrass hosts P. annua and A. palustris are in a different plant family than the
crop hosts, and the two turfgrasses are in different subfamilies from each other,

is reasonable to hypothesize that isolates from these hosts would demonstrate

some specialization (27). Understanding the relationships between isolates from
maize, sorghum, creeping bentgrass, and annual bluegrass would allow us
develop information about the reservoirs of inoculum, gene ﬂow, genetic drift,

and taxonomy.

lsozymes are often used as an initial tool to examine the variation present
in an unstudied system. Isozyme analysis has been used to study variation in
other species of Colletotrichum (23.25.26). As with any technique, isozyme
analysis has advantages and disadvantages. The advantages of this technique
include: low cost, minimal use of toxic chemicals, rapid results, true genetic loci,
codominant markers, easy preparation, and the ability to screen for a large
number of markers in a single run (30,31 ). Since only 1/3 of the mutations in
isozymes result in a change of charge that would result in a change in migration
distance, isozyme analysis represents a conservative estimate of variation
present within genomes (30). The disadvantages include: the need for sufﬁcient
quantities of fungal tissue, a limited number of loci, and the possibility of cryptic,
post-translational modiﬁcations of functional enzymes (30,31). Overall, isozyme
data provide a broad view of the variation present in a system upon which further

studies using DNA-based technologies can be built.

Since most golf course managers consider P. annua a weed, a pathogen
that could selectively eliminate P. annua from the turf stand would be very useful.
The literature presents an inconsistent view as to whether biological control is
possible. One step in assessing the feasibility of using C. graminicola as a

biological control organism is to examine the genetic relationships of the

pathogen from P. annua to the closely related and well-studied Colletotrichum
pathogens in maize and sorghum and to isolates originating on the primary non-
target host, A. palustn's. Unless the isolates from P. annua can be demonstrated
to be genetically distinct from other hosts, then the possibility of using C.

graminicola as a biological control agent is at best a risky proposition.

Literature Cited
1) Fenner, M. 1985. Seed Ecology. Chapman and Hall, New York, NY.

2) Vargas, J.M., Jr. 1994. Management of Turfgrass Diseases. 2nd Ed. CRC
Press, Inc., Boca Raton, FL.

3) Timm, G. 1965. Beitrage zur biologie und syztematik von Poa annua L. 2.
Acker Pﬂanzenbau 122: 267-294.

4) Johnson, P.G., et al. 1993. An overview of P03 annua L. reproductive
biology. lnt Turfgrass Soc Res J. 7: 796-804.

5) Cline, V.W., et al. 1993. Observations of population dynamics on selected
annual bluegrass-creeping bentgrass golf greens in MN. Int Turfgrass
Soc Res J. 7: 839-844.

6) Sprague, H.B., and Burton, G.W. 1937. Annual bluegrass (Poa annua L.)
and its requirements for growth. NJ Agr Exp Sta Bull. 630, 24pp.

7) Smith, JD. 1954. A disease of Poa annua. J Sports Turf Res Inst 8: 344-
353.

8) Smith, J.D., Jackson, N., and Woolhouse, A.R. Fungal Diseases of Amenity
Turf Grasses. 1989. E.& F.N. Spon, London, p. 219-226.

9) Bailey, J.A., O’Connell, R.J., Pring, R.J., and Nash, C. 1992. “Infection
strategies of Colletotrichum species,” in: Colletotrichum: Biology,
Pathology, and Control, eds, Bailey, J.A., and Jeger, M.J. CAB,
Wallingford, U.K. p. 88-120.

10) Kolattukudy, P.E., Rogers, L.M., Li, D., Hwang, 0.8., and Flaishman, MA.
1995. Surface signaling in pathogenesis. Proc. Natl. Acad. Sci. 9224080-
4087.

11) Flaishman, MA. and Kolattukudy, RE. 1994. Timing of fungal invasion
using host’s ripening hormone as a signal. Proc. Natl. Acad. Sci.
91 :6579-6583.

12) Coleman, L.W. and Hodges, CF. 1987. Ethylene biosynthesis in Poe
pretensis leaves in response to injury or infection by Bipolaris sorokiniana.
Phytopathology 77: 1 280-1 283.

13) Muimba—Kankolongo, A., and Bergstrom, 6.0. 1992. Wound predisposition
of maize to anthracnose stalk rot as affected by intemode position and
inoculum concentration of Colletotrichum graminicola. Plant Dis 76: 188-
195.

14) Vargas,J.M. Jr. 1979. H.A.S.- Decline of Annual Bluegrass. Proc. Of
Northwest Turfgrass Conf. 33: 41-46.

15) Abeles, F. 8., Morgan, P.W., and Saltviet, M.E., Jr. 1992. Ethylene in Plant
Biology 2nd Ed. Academic Press, Inc., New York, NY.

16) Bolton, AT, and Cordurkes, W.E. 1981. Resistance to Colletotrichum
graminicola in strains of Poe annua and reaction of other turfgrasses. Can
J. Plant Pathology 3: 94-96.

17) LeBeau, F.J. 1949. Pathogenicity studies with Colletotrichum from different
hosts on sorghum and sugarcane. Phytopathology 40: 430-438.

18) Chiang, M.Y., Van Dyke, CG, and Leonard, K.J. 1989. Evaluation of
endemic foliar fungi for potential biological control of johnsongrass

(Sorghum halepense): screening and host range tests. Phytopathology
73: 459-464.

19) Jamil, F.F., and Nicholson, R.L. 1987. Susceptibility of corn to isolates of
Colletotrichum graminicola pathogenic to other grasses. Plant Disease
71: 809-810.

20) Sreenivasaprasad, 8., Mills, P.R., Meehan, B.M., and Brown, AB. 1996.
Phylogeny and systematics of 18 Colletotrichum species based on
ribosomal DNA spacer sequences. Genome 39: 499-512.

21) Guthrie, P.A.l., Magill, C.W., Frederiksen, RA, and Odvody, GM 1992.
Random ampliﬁed polymorphic DNA markers: A system for identifying

and differentiating isolates of Colletotrichum graminicola. Phytopathology
82: 832-835.

10

22) Vaillancourt, L.J., and Hanau, RM. 1992. Genetic and morphological
comparisons of Glomerelle (Colletotrichum) isolates from maize and from
sorghum. Exp Mycol 16: 219-229.

23) Huguenin, B., Lourd, M. and Geiger, JP. 1982. Compairaison entre isolets
de Colletotrichum falcatum et Colletotrichum gramincole sur la base de
leurs characteristiques morphologiques, physiologiques, et pathologiques.
Phytopath Z. 105: 293-304.

24) Backman, P. MS. Thesis. Pennsylvainia State University.

25) Bonde, M.R., Peterson, G.L., and Maes, J.L. 1991. Isozyme comparisons
for identiﬁcation of Colletotrichum species pathogenic to strawberry.
Phytopathology 81 : 1523-1528.

26) Lenne, J.M., and Burdon, J.J. 1990. Preliminary study of virulence and
isozymic variation in natural populations of Colletotrichum gloeospon'oides.
from Stylosenthes guienensis. Phytopathology 80: 728- 731.

27) Gould, F.W., and Shaw, RB. 1983. Grass Systematics 2nd Ed. Texas
A&M Press, Inc., College Station, TX

28) Sherriff, C., Whelen, M.J., Arnold, GM, and Bailey, J.A. 1995. DNA
sequence analysis conﬁrms the distinction between Colletotrichum
graminicola and C. sublineolum. Mycol. Res. 99: 475-478.

29) Braiser, CM. 1987. “The dynamics of fungal speciation,” In: Evolutionary
Biology of the Fungi eds. Rayner, A.D.M., Braiser, CM, and Moore, D.
Cambridge University Press, Cambridge, UK. p. 231-260.

30) Hoelzel, A.R., ed. 1992. Molecular Genetic Analysis of Populations-a
practical approach. Oxford University Press, NY.

31) Micales, J.A., Bonde, MR, and Peterson, G.L. 1986. The use of isozyme
analysis in fungal taxonomy and genetics. Mycotexon 27: 405-449.

11

Chapter 1

Genetic comparison of crown rotting anthracnose isolates from amenity
turfgrass hosts to isolates from corn (Zea mays) and sorghum (Sorghum
spp.) uslng isozyme analysis

Introduction

Crown rotting anthracnose (CRA) caused by Colletotrichum graminicola is
a serious disease of Poe annua L. CRA and other fungal diseases severely
diminish P. annua’s ability to survive; this high susceptibility to disease presents
a major obstacle to developing this grass as a desirable species (2). As one of
the ﬁve most widely disseminated weeds in the world (1 ), P. annua is often
troublesome in golf course turfs because no effective control measures exist at
this time to selectively eliminate it from the turf stand (2). For this reason, P.
annua is tolerated and managed as a turfgrass on golf courses. However, a
disease that selectively infects P. annua in a mixed stand has the potential to be
used as a biological control. Biological control of P. annua has been attempted
with several pathogens, but as yet no disease has provided economically
effective control. One of the problems that golf course superintendents face
when their golf course is infected with CRA is the unpredictable ball roll caused
by the small voids of bare earth left as a result of the disease. These voids also
negatively affect the appearance of the playing surface. Because aesthetics are
critically important to pleasing golfers, CRA can cause signiﬁcant problems for

golf course operators. There are basically two competing viewpoints on how to

12

deal with P. annua encroachment. The ﬁrst group views P. annua as good
turfgrass that tolerates low mowing, and should be managed as a desirable
turfgrass. In fact, some in this school would even argue for improving the wild
types of this species through directed breeding. The second group views P.
annua as a weed that should be eradicated. No matter which viewpoint one
subscribes to, diseases like CRA need to be understood either to better manage

P. annua, or to take advantage of one of its weaknesses.

CRA was ﬁrst reported in 1954 in Great Britain (3). The symptoms
occurring on an infected turfgrass area were described as “measly” and “piebald”
(3). The disease begins as small, diffuse patches of chlorotic, dying turf that may
increase in diameter to 15 cm or more. Initially, older leaves are discolored, but
eventually the central leaf becomes discolored, and plant death usually follows
(2,3,4). The base of a detached crown from an infected plant appears black and

rotted. Close inspection of the rotted tissue reveals numerous aoervuli.

The causal agent of CRA is presently identiﬁed as Colletotrichum
graminicola. Traditionally, all Colletotrichum isolates that infect graminaceous
hosts are identiﬁed as C. graminicola by default. This convention is due to the
assumption that if an isolate of Colletotrichum is morphologically similar to the
type isolate from corn (Zea mays L.), then it is identiﬁed as C. graminicola.
However, recent research has shown that multiple species of Colletotrichum
infect graminaceous hosts (5,6). Vaillancourt and Hanau used data generated
from mtDNA RFLPs, RAPDs, and crossing studies to demonstrate that

Colletotrichum isolates from maize and sorghum were probably different species

13

(5). Data from RAPD ﬁngerprinting showed only 45% similarity between isolates
from different host origins (5). These data supported the reclassiﬁcation of
isolates originating on sorghum from C. graminicola to a new species (5).
Sherriff, et al. conﬁrmed these results using the ribosomal DNA internal
transcribed spacer region (ITS-1) sequence and suggested placing those isolates
from sorghum into a new species, C. sublineolum (6). Currently, there is a lack
of information about the species composition of Colletotrichum occurring on

amenity turfgrasses.

Taxonomic identiﬁcation of isolates within this group is important for
practical reasons. The principal reason is to facilitate regulatory control of the
development of pathogens as biological control agents, which must be regulated
to reduce the possibility of destroying economically important crops due to cross
infection. Inaccurate identiﬁcations result in regulations that may prevent useful
organisms from becoming safe biological control agents. Once a correct
identiﬁcation has been obtained, additional information (e.g., host speciﬁcity, host
range, and morphological measurements) can be correctly referenced to the
appropriate species and informed regulatory decisions can be made. Therefore,
information about the genetic background of isolates as it relates to taxonomy is
a crucial part of evaluating the potential organisms have as biological control

agents.

In the case of Colletotrichum spp., using only morphological
characteristics for identiﬁcation is unreliable as evidenced by the recent

identiﬁcation of the new species, C. sublineolum (5,6). Therefore, molecular

14

genetic techniques should become the method of choice to identify species
within Colletotrichum, and unless these techniques are used, meaningful
identiﬁcations of isolates that infect graminaceous hosts will remain extremely

difﬁcult to obtain.

One of the ﬁrst steps in examining potential taxonomic variation in fungi is
isozyme research. Isozyme analysis provides a broad view of variation that may
be present, and the relative amount of genetic information that is generated
compared to other techniques makes isozyme analysis an efﬁcient and cost-
effective ﬁrst step. The lack of meaningful data regarding the amount of genetic
variation in isolates of Colletotrichum from Agrostis pelustn’s Huds. (creeping
bentgrass) and P. annua indicates the need to determine the amount of genetic
variation present. Isozyme analysis has been used to study variation in other
species of Colletotrichum (7,8,9). Bonde et al. used isozymes to examine
Colletotrichum spp. infecting strawberries (7). They found that for the 5 species
tested, all the isolates supported the current species designation with the
exception of C. acutetum and C. gloeosporioides that were more closely related
than the species level (7). Their ﬁndings were that isozymes provide an effective

method for distinguishing the Colletotrichum species tested (7).

Our objective for this study was to examine and compare the interspeciﬁc
and intraspeciﬁc variation present in isolates of C. graminicola originating on P.
annua, A. pelustris, Z. mays, and Sorghum spp. using isozyme techniques. Data
resulting from this study of Colletotrichum species on turfgrasses could be used

as a basis for further research using DNA-based techniques to examine species-

15

level relationships and help determine if host speciﬁcity exists in isolates that
infect graminaceous hosts. Ultimately, data from this study will begin to assess

the potential of using CRA as a biological control for P. annua.

Materials and Methods
I§QlaI§§QIJﬁcliQn

Isolates of C. graminicola were collected from several different geographic
locations (Table 1). Isolates were collected from the following hosts: P. annua
(25), A. palustn's (21), Z. mays (10), and Sorghum spp (4). The isolate collection
from the turfgrass hosts was developed by collecting isolates from areas that I
visited, and also from Dr. Noel Jackson at the University of Rhode Island, Dr.
Randy Kane at the Chicago District Golf Association/University of Illinois and Dr.
Peter Landschoot at Pennsylvania State University. The collection of isolates
from corn and sorghum were from the collections of Dr. Ralph Nicholson at
Purdue University, Dr. Gary Bergstrom at Cornell University, Dr. Don White at the
University of Illinois and Dr. David TeBeest at the University of Arkansas.
Isolates were single spored and stored as a spore suspension in 20% glycerol
solution at -80°C. Several isolates were also stored as potato dextrose agar
(PDA) (Difco, Detroit, MI) plugs in 1 mL sterile heavy mineral oil. Isolates were
maintained in storage for up to two years until they were needed for

electrophoresis.
CultuLeELenetetion
Isolates were removed from storage and prepared for electrophoresis by

seeding 15 mL V8 juice broth (200 mL low Na V8, 800 mL distilled H20, 6 g

16

CaCOa) with 200 pl of spore suspension and incubating them in a 25 cm2 tissue
culture ﬂask (Corning, Corning, NY) under cool-white ﬂuorescent lighting for a 24

hr photoperiod for four days.

El l' IEIII .

The techniques concerning the procedures for isozyme analysis were modiﬁed
as described below from Quiros (10). Proteins were extracted immediately prior
to electrophoresis. The contents of the tissue culture ﬂask were harvested and a
small portion of mycelium (~2-3 cmz) was excised and blotted dry. The mycelium
was placed in a sample well of a Plexiglas grinding tray that was placed on ice to
prevent protein degradation. Cold 2% glutathione extraction buffer (120ul) was
added to each sample to further prevent protein degradation. The mycelium was
macerated with a blunt Plexiglas rod to release the proteins into the extraction
buffer. Extracted proteins from one sample were absorbed into four ﬁlter paper
wicks (Whatman 3mm Chr, Maidstone, England) measuring approximately 3 x 8
mm. Both a pH 8.3 system, and a pH 5.7 system were used to detect enzymes
present. Horizontal starch gels (12% w/v) were prepared using the appropriate
gel buffers for each system (Table 2). The gel origin was sliced perpendicular to
the direction of the current, and two wicks from each sample were placed
vertically at the origin of the gel. The gels were maintained at 4°C, placed in
trays containing the appropriate running buffer (Table 2), and electrical current
was applied across the gel at the appropriate amperage for each pH system
(Table 2). The protein front was allowed to progress for about 40 minutes before

the wicks were removed from the origin. The gel was then covered with

17

transparency ﬁlm followed by a glass plate on top of which ice bags were placed
to cool the gel and prevent protein denaturation. Amperage (Table 2) was

monitored closely throughout each 3 hr run to ensure consistent migration of the

protein front.

18

Table 1- Electrophoretic phenotypes (EP) of crown rotting anthracnose isolates

collected from various hosts and locations.

 

 

Isolate ID Host Regiona EP Geographic Origin
lW 94-1 P. annua M 1 Detroit, MI

DB 95-1 P. annua M 1 Detroit, MI

WH 95-1 P. annua M 1 Lansing, MI

CCL 95-1 P. annua M 3 Lansing, MI

CL 95-1 P. annua M 4 Missouri

WGC 95-1 P. annua M 4 Weirton, WV

CW 95-1 P. annua M 2 Vienna, OH

LS 95—1 P. annua -- 1 Chicago, IL

KWIL 95-1 P. annua M 1 Chicago, IL

EV 95-1 P. annua M 1 Chicago, IL

BA 95-1 P. annua --- 3 Bel Aire, CA

WF 95—1 P. annua A 2 Wakeﬁeld, RI
CAM P. annua A 2 State College, PA
QR 95—1 P. annua A 1 Scarsdale, NY
RK 96-1 P. annua M 2 Detroit, MI

SUN 1 P. annua A 1 Johnston, PA
OAK 2 P. annua A 2 Pittsburgh, PA
OAK 3 P. annua A 2 Pittsburgh, PA
VAL 1 P. annua A 2 State College, PA
MCC 95—1 P. annua A 2 Madison, CT

AP 95-1 P. annua A 2 Cranston, RI

GR 95—1 P. annua M 2 Grand Rapids, MI
IV 95-1a P. annua M 2 Toledo, OH

IWIL 95-1 P. annua M 2 Chicago, IL

SPC 95-1 P. annua A 2 Sands Point, NY
CH 95-2 A. pelustris A 5 Southbridge, MA
SS 95—2 A. pelustris A 5 Ridgeﬁeld, CT
BOB 95-2 A. pelustris A 6 Brewster, NY

AV 95-2 A. pelustris A 6 Weston, CT

FCC 95-2 A. pelustn's A 7 Fermington, CT
GW 95-2 A. pelustris A 6 Greenwich, CT
PA 95-2 A. pelustris A 6 Bellingham, MA
SCC 95-2 A. pelustris A 6 Stonington, MA
VAL 2 A. pelustris A 6 State College, PA
WFCC 95—2 A. pelustris A 6 Wethersﬁeld, CT
ACC 95-2 A. pelustris A 8 Plymouth, MA

AT 96-2 A. pelustris --- 3 Atlanta, GA

CCN 95—2 A. pelustris --- 3 Pinehurst, NC

19

Table 1 (cont’d)

 

Isolate ID Host Region‘ EP Geographic Origin

FF 96-2 A. pelustris -- 3 Frankfort, KY

GM 95-2b A. pelustris -- 3 Missouri

ICC 95-2 A. pelustris A 3 Ipswich, MA

KCC 95-2 A. pelustris A 3 Yarmouthport, MA

KEN-1 A. pelustris --- 3 Lexington, KY

MA 95-2 A. pelustris --- 8 Kennesaw, GA

PCC 95-2 A. pelustris A 3 Litchﬁeld, NH

PQ 95—4 A. tenuis A 3 Bristol, ME

UNK 95-2 A. pelustris --- 9 Chicago, IL

COR 2 Z. mays --- 10 Rock Springs, PA

Cg 151 Z. mays --- 10 New York

COR 3 Z. mays --- 14 New York

Cg ASK 88 Z. mays --- 12 Unknownb

Cg M6 Z. mays --- 12 Missouri

Cg 46 Z. mays --- 11 Unknown”

Cg M9 M0 Z. mays --- 13 Missouri

Mo 940 D Z. mays --- 12 Indiana

Cg 17 Z. mays --- 12 Unknownb

Cg 122 Z. mays --- 12 Indiana

Cg 1042 (2-3-9) 8. halepense --- 15 Stuttgart, AR

Cg 1034 S. halepense --- 16 Devil’s Den, AR

Cg 1031 S. halepense -- 18 College Station, TX
W -- 17 Grimm GA

 

' Isolate region assignment for geographic data analysis (where M=midwest,
A=etlantic

" Unknown locations from Ohio to Nebraska

20

Table 2- Enzymes and buffer systems used for the detection of variation in

collected C. graminicola isolates.

 

 

Enzyme“ Abbreviation EC No. Activity” Electro- Buffer
morphs° Systemd
Aspartate AAT 2.6.1.1 A, B 4 1
aminotransferase
Isocitric dehydrogenase IDH 1.1.1.42 A 1 2
Malate dehydrogenase MDH 1.1.1.37 AB 3 2
Phosphoglucose PGI 5.3.1 .9 AB 4 2
isomerase
Phosphoglucomutase PGM 5.4.2.2 AB 2 1
6-Phosphogluconate 6PGDH 1 .1 .1 .44 A 1 2
dehydrogenase
Triosephosphate TPI 5.3.1.1 AB 3 1
isomerase

 

‘ All stain recipes from (11)

" Activity visualized in isolates from: A= amenity turfgrasses, B= crop plants

° Number of different activity bands visualized in each enzyme studied

“ Buffer systems used (11): 1= Lithium borate/ Tris-citrate, pH 8.3, 55 mA

2= Histidine-citrate, pH 5.7, 35 mA

21

1!. I' I' I ll

Following electrophoresis, gels were sliced into four approximately 1 mm
thick slabs and each slab was placed in a separate enzyme stain. Enzyme
activity was assayed using common histochemicel stains that produce an
insoluble dye where enzyme activity is present. The staining protocols were
modiﬁed from Vallejos (11) by optimizing the intensity of the staining through
adjustments to the substrate amount in the stain. To ensure the repeatability of
the zymograms, each isolate was run on two replicate gels and an isolate with a
deﬁned banding pattern was used as an internal reference standard for each gel.
A total of 16 enzyme systems were assayed for activity and resolution in two
buffer systems. Seven enzyme systems were resolved in isolates originating
from A. palustris and P. annua, while only ﬁve enzyme systems were resolved in
isolates from Z. mays and Sorghum spp. (Table 2). The nine enzyme systems
remaining that were not repeatedly resolved were not investigated further.

Genimmencletute

Alleles found in the different enzyme systems (Table 2) were numbered
sequentially from the anodal end of the gel using one number for locus and a
second number for allele (abbreviations with all capital letters refer to enzymes,
and those with the ﬁrst letter capitalized refer to the locus coding for the particular
enzyme). For example, the ﬁrst locus, third allele of the enzyme PGI would be

abbreviated as, Pgi13.

22

Analvslmﬂata

Date generated from each gel run were recorded in a matrix identifying the
presence (1) or absence (0) of a particular band. Allele frequencies were
calculated from the presence/absence matrices and subdivided based on host
origin or geographic location (Table 6). Nei’s unbiased genetic distance

(corrected for small sample sizes) (12) was calculated for each group of isolates,

and is deﬁned as follows:

Let p, and q, be the frequencies of the ith allele in populations X and Y,
respectfully, and x, and y,are the corresponding allele frequencies from the

sampled populations.

D = —ln[ny/ ,lGxGy], where 6,, Gy, and Q, are the means of Epf, 2g}, and

2p,q, over all loci in the genome. Replacing the population gene identities with
the sampled population gene identities J," J,, and JJ, which are the means of
2x3, zyf, and My, over r loci studied. However, with small sample sizes the

measure of genetic distance is biased. Therefore, Nei’s genetic distance (12)

can be corrected for sample size as follows:

Let n, and n, be the number of individuals sampled from populations X and
Y, respectively. Substituting an unbiased estimator for each frequency value,

yields an unbiased estimate of genetic distance. Namely,

23

15 = —ln[éxéy/ (16.53] where the caret (A) denotes an unbiased estimator

for the frequency means.
The estimators are deﬁned such that,

Gxand G, are the averages where, 22m!" '1 , and 2"»5 ‘1

nx—l Zny—l

over r loci

 

 

studied, respectively, and ny = A. Nei’s distance essentially is calculating the
probability that an allele came from one of two sampled populations such that a
large distance (> 1.000) indicates that one of the sampled populations is probably
similar to the other population. However, a small distance measure (< 0.300)
indicates that one of the populations isnot as closely related to the original
population. The small distances could be potential areas where speciation may
be occurring. Areas where the distance is between the two extreme values
(0.3< X < 1.0), typically indicates the possibility of sub-species organization

levels.

Results

Seven enzymes produced distinct and repeatable zymograms in isolates
from A. palustris and P. annua. Only ﬁve enzymes produced repeatable
zymograms in isolates from Z. mays or Sorghum spp. The enzymes 6-PGDH
(E.C. number 1.1.1.44) and IDH (E.C. number 1.1.1.42) were dropped from all
analyses because they were monomorphic in the A. palustris and P. annua
groups of isolates, but were not resolvable in the Z. mays or Sorghum spp.

isolate groups. Therefore, only the ﬁve enzymes that were resolved in all isolates

24

were used for analysis.

Sixteen putative allelles were resolved with these 5 enzymes. Eighteen
unique electrophoretic phenotypes (EPs) were observed in the 61 isolates
sampled. All isolates produced single banded phenotypes consistent with a
haploid genetic condition in all enzyme systems tested. Crossing studies were
not possible with these isolates because the teleomorph of C. graminicola has

not been formed using isolates from amenity turfgrasses.

Nei’s corrected genetic distance between isolates from P. annua and A.
palustn's was consistent with a relationship at the sub-species level (Table 4A).
Isolates from P. annua were more closely related to isolates from Sorghum spp.
than to isolates from Z. mays (Table 4A). In contrast, isolates from A. palustn's
were more closely related to isolates from Z. mays than to isolates from Sorghum
spp., and the distances between each of these groups of isolates are consistent

with species level relationships (Table 4A).

25

Table 3- Nei’s corrected genetic distance (12) by host origin (A) and geographic

location (B).

A)

 

Host P. annua A. pelustris Z. mays

 

A. palustn's 0.1870

 

 

Z. mays 0.5683 0.9133

Sorghum 4.4371 1 .1540 1 .6089
3)

Location Atlantic P. annua

 

Midwest P. annua 0.1 165
Atlantic A. pelustris 0.3242

26

0000.2 n...< 0:0 3032.2 ">22.

 

0000.0 0000.0 0000.? 0000.0 0000.0 50010 0000.0 ~000.0 0000.0 0050.0 0000.0 0000.0 0050.0 0000.0 0000.0 0000.0 «Swimq .< -=<
0000.0 0000.0 0000... 0000.0 0000.0 0000.0 000N.0 0000.0 0000.0 000N.0 0000.0 0000.0 0000.0 000N.0 0000.0 0000.0 «:55 .l ._.<

0000.0 0000.0 0000... 0000.0 0000.0 002.0 0000.0 000v.0 0000.0 0000.0 0000.0 0000.0 0000.0 00010 0000.0 0000.0 03:20 .1522

 

 

020.0 N20... 3.0... 3.00 0200 N200 .200 NrEdn. :Ema 02.05. N205. 2.50.2 3.00 0:00 N.._0n. 3.0m 20080..

 

.03 cE>Nom_

cosmos 9509080 >9 35:02. 92?. .0

0.3.0 000N.0 00050 0005.0 ..0N.0 0000.0 ..00.. 0000... .....0 :00... 0000.0 .....0 2.0.. 0000.. 0000.0 000N.0 Eu... 00
0000.0 0000.0 003.0 0000.0 0000.0 0000.0 0000... 0000.0 0000... 009.0 0000.0 0000.0 0000.0 0000.0 0000.0 0000.0 9.05 .N
0000.0 0000.0 0000... 0000.0 0000.0 0000.0 ..00¢.0 0000.0 500.0 0000.0 0000.0 0000.0 0.0.4.0 0000.0 0000.0 000.0 0.5033 .<

0000.0 0000.0 0000... 00N0.0 0000.0 0000.0 000v.0 00N0.0 0000.0 00N0.0 0000.0 0000.0 0000.0 00N0.0 0000.0 0000.0 035.5 .1

 

«:3 «:3 E3 360 «:8 «:8 :60 «EB :53 2:22 «222 3:22 Ema 2mm ~28 EB 5998:

 

_oo._ 0:332

E95 .8; 3 5:302“. 222 .<

5082 0302080 0cm 50:0 .8: B 822.53 .02me _oo_ mE>Now_ Lo. 385309.. 322 -v 03m...

27

The smell distance value between isolates from P. annua in the Atlantic region
and those from the midwest region (Table 1) was inconsistent with expected
distances resulting from geographic specialization (Table 4B). Distances for
eastern seaboard isolates of P. annua and A. pelustris (Table 1) were consistent
with isolates that were specialized based on host type (Table 48). Therefore,
host specialization appears to be a primary source of diversity in these two

groups of isolates.

Since the sample sizes for isolates from Z. mays and Sorghum spp. were
small, meaningful conclusions regarding the relationship of these two groups are
difﬁcult. However, distances of isolates from these two hosts were consistent
with the known species-level relationship that these two groups of fungi have with

each other (Table 4A).

Four EPs contained multiple members that all originated on a single host
species. EP-2 had eight members that were geographically diverse, but all
members originated on P. annua (Table 1). EP—3 had thirteen members that also
were geographically diverse and of P. annua origin (Table 1). EP-6 had two
members from NY and CT, and both isolates were from A. pelustris (Table 1).
EP-7 had seven members from three states that were all from A. pelustris (Table
1). Two EPs contained members from both P. annua and A. pelustris. EP-5 had
one member each from the two hosts (Table 1). EP-4 had ten members from A.
palustris and two members from P. annua. Isolates in EP—4 were also

geographically diverse thus reducing the signiﬁcance of geographic specialization

28

(Table 1). Zea mays isolates were made up of ﬁve EPs, and they were not
observed in any isolates from any other host sampled (Table 1). Likewise,
Sorghum isolates were all unique from each other in their EP as well as to all

other isolates EPs (Table 1).

Discussion

This is the ﬁrst report to compare variation in Colletotrichum spp. from
amenity turfgrass hosts to the variation observed in Colletotrichum spp. from corn
and sorghum at the genetic level. Based on these date, two general conclusions
arise: host specialization, not geography, appears to be the primary factor
affecting the pattern of variation in the isolates tested, and morphological
characteristics are of limited value when examining isolates of Colletotrichum
from graminaceous hosts. A limitation of this study was the small sample size of
isolates. Because of the small sample size, it is difﬁcult to draw conclusions
regarding the processes involved in the divergence of these fungi. However,
these data present a baseline to which future date can be compared. Both of the
conclusions that are supported by the data illustrate why understanding
taxonomy is an important part of assessing the potential of a biological control

agent

There was a signiﬁcant amount of variation observed in screened isolates
as 30% of them produced unique EPs. Our results are consistent with reports
that also show a large amount of genetic diversity in other species of
Colletotrichum (5,679.14). The diversity found in this genus contrasts with the

lack of diversity often found in other genera when using isozymes (8). The

29

diversity in our study is similar to that observed in the Bonde, et al. (7) study of
Colletotrichum spp. on strawberry, and indicates that there is enough variation
present to use isozymes to study diversity in populations of Colletotrichum

infecting graminaceous hosts (Figure 1).

Isolates from P. annua were geographically distributed evenly from the
midwest to the Atlantic seaboard, and this provided the opportunity to test the
possibility of geographic specialization. If geographic specialization existed in the
isolates tested, then the expected Nei’s distance would be large (i.e. >1.000).
However, Nei's corrected distance (12) revealed a very close relationship
between isolates from the midwest and those from the seaboard (0.1165), and

this distance was the smallest of all the comparisons made by either host type or

geography.

The possibility of host specialization was tested by comparing isolates
from P. annua and A. pelustris from the Atlantic seaboard. If host type
specialization was present, then the observed distance value would be large (i.e.
>1 .000). Although observed distances were consistent with a sub-species
relationship, this distance was larger than the distance observed between
isolates from these two hosts for the entire collection, indicative of host type

specialization.

The distances between isolates from A. palustris and the isolates from the
crop hosts indicated a strong divergence consistent with a species level

relationship. However, the same comparison between isolates from P. annua

30

and the crop hosts indicated a relationship below the species level. One possible
reason for this inconsistency is that P. annua is an ubiquitous weed found in
agricultural and turf areas while A. pelustris is usually only found in turf areas.
These inconsistent distances could reﬂect a possible geographic association
between isolates from P. annua and those from the two crop hosts that could not
be tested in our study. This is important if C. graminicola is to be used as a

biological control for P. annua.

Morphological differences have often been used to study speciation
processes in fungi (15). Isolates of Colletotrichum that infect corn have been
shown to be closely related, but distinct species from those isolates that infect
sorghum. This raises the possibility of other such species relationships within
Colletotrichum spp. that infect graminaceous hosts. Numerous reports have
shown that morphology is of limited value when studying closely related species
because the character measurements often overlap making determinations
difﬁcult, if not impossible (5,9,13). Traditionally, when samples of isolates lack
morphological differences, they are classiﬁed as a single species. Morphological
comparisons between selected isolates in this study showed no differences
between their measurements and those published previously (Horvath,

unpublished data, 5,9,13).

Sympatric speciation is a possible theoretical framework for the
differences observed both in this study and other published studies using
graminaceous infecting isolates of Colletotrichum (15). With sympatric

speciation, an organism diverges rapidly from its basal population, and eventually

31

becomes different enough to be reproductively isolated from the original basal
population (15). One of the major ways an organism can diverge from its basal
population is by exploiting a new niche (15). Niche divergence would explain the
observed differences in this group of Colletotrichum spp. Sympatric speciation
has been the subject of much study in the apple fruit ﬂy species group Rhegoletis
pomonella (16,17,18). Larvae of Rhegoletis from apple and hawthorn were
analyzed using isozyme loci (17). The allele frequencies of two of the six loci
tested in the different populations varied signiﬁcantly between host types for all
three years tested. In a sympatrically speciating population other inﬂuences like
host preference, mate selection, and survivorship can result in genetic barriers to
gene ﬂow. Collectively, these barriers restrict gene ﬂow in two isolated groups
such that these two groups sympatrically diverge into two sibling species with
alternative host preferences (16,17,18). In the Colletotn'chcum system, isolates
from corn could have taken advantage of the presence of sorghum nearby and
those isolates that could exploit this new niche eventually isolated themselves so
that now isolates from sorghum belong to a new species, C. sublineolum.
Isolates infecting the crop hosts could also capitalize on the presence of an
ubiquitous weed, P. annua, as another possible under-utilized niche. Like
isolates from corn and sorghum, isolates that infect the amenity turfgrass hosts
are also spatially associated, so there is the possibility of these isolates also
diverging into other isolated niches. This model would account for the variation
observed in this study and other studies with the grass infecting Colletotrichum

spp. Future work in this area would involve sampling local populations and using

32

DNA based techniques to determine how these groups of pathogens are

evolving.

Understanding the genetic relationships a potential biological control agent
has is an important part of assessing the biological control potential for the
organism of interest. Based on the data from this study, isolates from P. annua
share a large amount of genetic information with isolates from A. pelustris. This
close relationship would seem to limit the potential use of Colletotrichum isolates
from P. annua as biological control agents because of the potential risk of cross
infection. Isolates from P. annua are currently identiﬁed as C. graminicola as are
all isolates from Z. mays, and A. pelustris. However, our data indicate that
isolates from the amenity turfgrass hosts are more closely related to each other
than they are to isolates from corn. Isolates from sorghum are currently identiﬁed
as C. sublineolum, and have been shown to be reproductively isolated from
these other groups identiﬁed as C. graminicola (5,6). Isolates from P. annua
were more closely related to the isolates from sorghum than to isolates from
corn. The signiﬁcance of properly assigning taxonomic names is that since
isolates are currently named based on morphological characteristics, those that
have potential as biological control agents may not be investigated because of a
perceived threat to a major crop. Therefore, for isolates to be properly assigned
a correct taxonomic name, molecular genetic techniques should become the
primary method by which closely related isolates in Colletotrichum are identiﬁed

and classiﬁed.

The data from this study illustrate that the genetic relationships of

33

pathogens can be tremendously important in assessing the potential any
pathogen might have as a biological control agent. In this case, isolates from P.
annua are so closely related to isolates from A. pelustris that this relationship
signiﬁcantly decreases the potential of using CRA as a safe and effective
biological control against P. annua. The risk of non-target exposure to a
potentially damaging pathogen is signiﬁcant given our current understanding of

this pathogen.

Future work on this disease must include a study of host speciﬁcity,
population studies using DNA-based techniques, and taxonomic studies to

determine the species relationships in grass infecting Colletotrichums.

34

Literature Cited
1) Fenner, M. 1985. Seed Ecology. Chapman and Hall, New York, NY.

2) Vargas, J.M., Jr. 1994. Management of Turfgrass Diseases. 2nd Ed. CRC
Press, Inc., Boca Raton, FL.

3) Smith, JD. 1954. A disease of Poe annua. J Sports Turf Res Inst 8: 344-
353.

4) Smith, JD, Jackson, N., and Woolhouse, A.R. Fungal Diseases of
Amenity Turf Grasses. 1989. E.& F.N. Spon, London, p. 219-226.

5) Vaillancourt, L.J., and Hanau, RM. 1992. Genetic and morphological
comparisons of Glomerelle (Colletotrichum) isolates from maize and from
sorghum. Exp Mycol 16: 219-229.

6) Sherriff, C., Whelen, M.J., Arnold, GM, and Bailey, J.A. 1995. DNA
sequence analysis conﬁrms the distinction between Colletotrichum
graminicola and C. sublineolum. Mycol. Res. 99: 475-478.

7) Bonde, M.R., Peterson, G.L., and Maas, J.L. 1991. Isozyme comparisons for
identiﬁcation of Colletotrichum species pathogenic to strawberry.
Phytopathology. 81: 1523- 1528.

8) Lenne, J.M., and Burdon, J.J. 1990. Preliminary study of virulence and
isozymic variation in natural populations of Colletotrichum gloeosporioides.
from Stylosenthes guienensis. Phytopathology. 80: 728- 731.

9) Huguenin, B., Lourd, M. and Geiger, JP. 1982. Comparaison entre isolats
de Colletotrichum felcatum et Colletotrichum gramincole sur la base de
leurs characteristiques morphologiques, physiologiques, et pathologiques.
Phytopath Z. 105: 293-304.

10) Quiros, CF. 1981. Starch gel electrophoresis techniques used with alfalfa
and other Medicago species. Can. J. Plant Sci. 61: 745-749.

35

11) Vallejos, CE. 1983. “Enzyme activity staining,” In: Isozymes in Plant
Genetics and Breeding, eds., Tanksley, SD, and Orton, T.J., Elsevier,
N.Y. p. 469-516.

12) Nei, M. 1978. Estimation of average heterozygosity and genetic distance
from a small number of individuals. Genetics 89: 583-590.

13) Backman, P. MS. Thesis. Pennsylvainia State University.

14) Guthrie, P.A.I., Magill, C.W., Frederiksen, RA, and Odvody, G.N. 1992.
Random ampliﬁed polymorphic DNA markers: A system for identifying

and differentiating isolates of Colletotrichum graminicola. Phytopathology
82: 832-835.

15) Braiser, CM. 1987. “The dynamics of fungal speciation,” In: Evolutionary
Biology of the Fungi eds. Rayner, A.D.M., Braiser, CM, and Moore, D.
Cambridge University Press, Cambridge, UK. p. 231-260.

16) Feder, J.L., Chilcote, CA, and Bush, G.L. 1990. The geographic pattern of
genetic differentiation between host associated populations of Rhegoletis

pomonella (Diptera: Tephritidae) in the eastern United States. Evolution
44(3): 570-594.

17) Feder, J.L., Chilcote, CA, and Bush, G.L. 1990. The regional, local, and
microgeographic allele frequency variation between apple and hawthorn

populations of Rhegoletis pomonella (Diptera: Tephritidae) in western
Michigan. Evolution 44(3): 595-608.

18) Feder, J.L., and Bush, G.L. 1989. A ﬁeld test of differential host-plant usage
between two sibling species of Rhegoletis pomonella fruit ﬂies (Diptera:
Tephritidae) and its consequences for sympatric models of speciation.
Evolution 43(8): 1813-1819.

36

APPENDICES

37

APPENDIX A

EPILOGUE

These few pages at the end of this document are intended to highlight the
current state of the ﬁeld in this area of research and bring attention to techniques
and methods that were attempted during the course of this project that are not
mentioned in the formal text because they did not work out. I do this with the
hope that those reading this epilogue use the information contained within to

avoid some of the same trouble spots I encountered.

This project initially set out to study whether crown rotting anthracnose
(CRA) could effectively kill annual bluegrass (Poe annua L.) from within a healthy
stand of creeping bentgrass. The idea of using a fungus to selectively eliminate
one undesirable plant from a stand of desirable plants is attractive especially
since, in this case, P. annua currently can not be controlled through any other
means. In order to study the efﬁcacy of the pathogen, inoculation experiments
must be done. The problem with this pathogen is that it is a stress-related
disease that does not seem to be virulent against healthy, vigorous plants like
those usually maintained in the greenhouse. The inoculation methods that were
attempted included: sand/cornmeal around the base of the plant, wounding with
carborundum powder, wounding with needles, and infested soil. None of these
methods worked. There were several possible reasons for the lack of infection.

The ﬁrst is that the plants were not stressed enough to allow the pathogen to

38

gain entry and begin destroying tissue. This is probably not the likely because
the wounding methods did result in several plants being infected, but the
infection rate was very low. Another possible reason could be that the pathogen
was not able to colonize where it was inoculated and so could not establish high
enough inoculum levels to result in a successful infection rate. Again, this is
unlikely because with several of these methods an excess of inoculum was used
to avoid this very problem. The most likely problem with the methods that were
attempted is that a combination of the above factors coupled with the a lack of
control over environmental parameters caused the lack of success. One method
did work sufﬁciently to make it a small-scale inoculation teChnique useful for
working with this pathogen. This method involved injecting a spore suspension
directly into the crown region of the plant and then placing the injected plants into
a high humidity environment for 48-96 hrs. This method did result in successful
infections that were repeatable. This success however, was too late in my
project to be of use. Future work involving inoculations can use this information
as a starting point to develop a protocol that would allow large scale inoculations
to be used for screening plants for resistance to CRA and to screen isolates for

the possibility of altered virulence levels in populations.

Research often raises more questions than it answers. Colleagues often
comment that this principle is the sign of a good project. My research also raised
more questions than it raised. The two basic conclusions were; host type, not
geography was responsible for the variation observed in the isozyme patterns,

and that morphological characteristics are of limited value when trying to identify

39

isolates of Colletotrichum.

The questions that were raised by this my research include:

ls speciation occurring in this genus of Colletotrichum?

What are the evolutionary forces driving any observed speciation

events?

What organizational levels exist in these populations of C.

graminicola? Geographical? Host Type? Species? Environmental?

Is there any association between certain ecotypes of P. annua and

different EP types in the pathogen?

This is not a comprehensive list, but these questions could provide a much
improved understanding of the processes involved in this important group of
pathogens. The continuing use of molecular techniques to help study these

questions will only serve to make arriving at an answer to these questions easier.

It is my hope that these short words would help the reader understand
some of the thought processes involved in deciding on a project. I also hope that
this document is used to continue study on this diverse group of pathogens that
infect grasses, and I hope that ultimately someone beneﬁts from the information

that I developed.

40

APPENDIX B

Table 5- Unique electrophoretic phenotypes (EP) of Colletotrichum isolates

sampled

Isozyme Loci

 

....u_ .0
._..u_ .N
._..u_ ..
00... .0
GO... .0
GO... .0
00.. ..
.uOZ. .N
.0an ..
2:0... .0
2.01 .0
EU... ..
.uO. .0
.09 .0
.00; .N

.09 ..

20. o.
_mo_m.mm

m_u#

 

000000000010010000

000000000001100001

111111111100001110

010000000000000111

000000000000001000

001010010000000000

100101101111110000

100000110000001111

011111001111110000

100011101000011111

000000000111100000

011100010000000000

000011100000000000

100000000001000010

000000000110111100

011100011000000001

 

41

APPENDIX C

Figure 1- Figure showing the different alleles for the enzymes PGI (A), PGM (B),

and GOT (C) (NOTE: each set of two bands is from a single isolate).

A) PGI- 4 alleles

 

B) PGM- 2 alleles

1

 
   
   

C) GOT- 3 alleles

 

42

G

 

I(11111111111111ljllll'll

E

580

ill

 

 

—1
: -—____