 

. ‘ £3.
3%.. . , . , .

7;.
wt. 6..

 

 

in.
.

‘
h. .
rim.”
$33... I . x.
4... . . z A
— v -
w: ..

. Hm"?
, . . , L33”
7‘”... 11.
i5. ﬂung... .,
.4; «gm.
‘ a“! 535 z. . . LR“.
I ,..: a. , 3:. K a. . , A ”inn. .
i ‘ ‘ ‘ :3 7 ‘ _ , z m
J xzrl . . . , x: ‘ 4 .
.. an“ V , . . ‘ .
.. a, as“ . .
Waugh“...

 

 

~
iii; 7 Lira!
an...” 3....“ an;
3 :1! .J‘

.69

.-.

)l.(<.>‘u- I. LtD—l-‘u- aim

 

?‘

....--f

: ‘
5.1% r

. ‘

.... m.“ A .

e“! .

A .a
PI‘...
r

.1 .
”7.2.2
or,

 

   

. V ‘ ; «.7. . . , , us:
m? £1 . x .. f 7 . V an

  

av. i am.
ﬁxing». V .P

 

 

 

This is to certify that the

dissertation entitled

AN INVESTIGATION OF PHYTOPHTHORA CAPSICI ON
VEGETABLE HOSTS IN MICHIGAN: SURVIVAL, SPREAD,
AND RESPONSE TO THE PHENYLAMIDE FUNGICIDE MEFENOXAM

presented by

Kurt Haas Lamour

has been accepted towards fulﬁllment
of the requirements for

Ph . D . degree in Botany and
Plant Pathology

Ma?A/x/%LW%

Major professor

Date March 30, 2001

 

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

 

 

 

LIEHARY

Michigan State
University

 

 

 

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

 

DATE DUE DATE DUE DATE DUE

 

MAW; 0 19 gnii}

 

 

 

 

 

 

 

 

 

 

 

 

 

 

6’01 C‘JCIRC/DateDUGpGS-p. 15

 

AN INVESTIGATION OF PHYTOPHTHORA CAPSICI ON VEGETABLE HOSTS IN
MICHIGAN: SURVIVAL, SPREAD, AND RESPONSE TO THE PHENYLAMIDE
FUNGICIDE MEFENOXAM.

By

Kurt Haas Lamour

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

2001

ABSTRACT
AN INVESTIGATION OF PHYTOPHTHORA CAPSICI ON VEGETABLE HOSTS IN
MICHIGAN: SURVIVAL, SPREAD, AND RESPONSE To THE PHENYLAMIDE
FUNGICIDE MEFENOXAM. ‘
By

Kurt Haas Lamour

The incidence of root, crown, and fruit rot of pumpkins, squash, tomatoes, and
peppers caused by Phytophthora capsici Leonian has increased during the last two
decades in Michigan vegetable production ﬁelds. Currently recommended control
strategies include the use of well drained ﬁelds, crop rotation, and the use of preventative
fungicides. Growers employing all available management strategies have sustained
signiﬁcant losses and, at the behest of the cucurbit industry, an intensive investigation of
the life history of P. capsici throughout Michigan was initiated. The primary objective of
this study was to ascertain whether or not the sexual stage is active in natural populations
and, if so, to determine what effect this may have on the survival and evolution of
naturally occurring populations. The sensitivity to mefenoxam and compatibility type
were assayed in natural populations. In 1997 and 1998, 523 isolates of P. capsici were
recovered from infested ﬁelds throughout the state. In vitro crosses between sensitive
and insensitive isolates from diverse locations indicated that mefenoxam insensitivity is
inherited as a single incompletely dominant gene unlinked to compatibility type. All six
possible mefenoxam sensitivity/compatibility type combinations were recovered from a
single ﬁeld and both mating types were recovered from every ﬁeld sampled. Oospores

were observed in diseased host tissue from four locations and all six phenotypic

combinations were found in 223 oospore progeny recovered from a single naturally
infected cucumber fruit. In 1999 and 2000, a single ﬁeld from which only intermediately
or fully insensitive P. capsici isolates were recovered in 1998 was sampled intensively in
the absence of mefenoxam selection pressure. Isolates from 1998 and 1999 were
analyzed using ﬂuorescently labeled ampliﬁed fragment length polymorphism (AF LP)
markers. Clonal reproduction was signiﬁcant within a single season but no clones were
recovered between years. Approximately ﬁfty percent of the AF LP markers were
polymorphic and 199 of the 263 isolates analyzed had unique multi-locus AF LP
genotypes. Furthermore, the frequency of individual AF LP markers remained stable
between years and mefenoxam insensitivity did not decrease over time. AF LP markers
were then used to characterize an additional 383 isolates from 6 populations of P. capsici
at locations ranging from 1 to >200 km distant. A similarly high level of genotypic and
gene diversity was recorded in every population investigated. Cluster analysis indicated
discrete clusters based on location with no blurring of the groupings based on the year of
sampling. The overall picture presented by these results suggests that outcrossing occurs
frequently in populations of P. capsici, that populations are large enough to withstand
dramatic effects of genetic drift, and that migration between locations appears to be rare.
Sexual recombination appears to have played a signiﬁcant role in the integration of
mefenoxam insensitivity into populations under mefenoxam selection pressure and there
is no evidence that the frequency of mefenoxam insensitivity will decrease once selection

pressure is removed.

To my parents, Thomas and Sharon Lamour,
whose love, patience, and humor
I cherish deeply.
To Kierstyn G. Lamour
for giving me a dynamic growing love

which words could never capture.

To Iris Gloria Lamour,

our own little oospore,

who teaches me daily

to smile.

iv

ACKNOWLEDGEMENTS

In particular, I want to thank my major professor and mentor, Dr. Mary Hausbeck,
for her candid honesty, dedication, and unceasing intellectual curiosity. An attempt at
summarizing the impact of her tutelage upon my life would be, in the words of Ralph
Waldo Emerson, like painting the lightning with charcoal.

I extend sincere thanks to my committee members Dr. Andrew Jarosz, Dr. Ray
Hammerschmidt, and Dr. Francis Trail who always took the time to listen and critically
evaluate my thoughts. A Special thanks is due to Dr. J arosz for astutely challenging my
early hypotheses and pushing my work into realms I never anticipated and have
thoroughly enjoyed. Dr. Hausbeck’s generous support allowed me the privilege of
working with Elizabeth Webster, Matt Bour, Jason Jabara, Charles Hunter, and Jeff
Woodworth. Each of these individuals provided valuable assistance and were a joy to
work with.

Finally, I sincerely thank Pavani Tumbalam who worked many hours without
receiving any formal academic or monetary credit. Without her, only a fraction of this

work could have been accomplished.

TABLE OF CONTENTS

LIST OF TABLES ........................................................................................... ‘ .................... v
LIST OF FIGURES ............................................................................................................ vi
LITERATURE REVIEW .................................................................................................... 1
CHAPTER 1: Mefenoxam insensitivity and the sexual stage of Phytophthora capsici
in Michigan cucurbit ﬁelds .................................................................................. 10
ABSTRACT ........................................................................................................... 10
INTRODUCTION .................................................................................................. 10
MATERIALS AND METHODS ........................................................................... 12
RESULTS .............................................................................................................. 16
DISCUSSION ........................................................................................................ 23
LITERATURE CITED .......................................................................................... 26
CHAPTER 2: The dynamics of mefenoxam insensitivity in a recombining population
of Phytophthora capsici characterized with AF LP markers .................................. 29
ABSTRACT ........................................................................................................... 29
INTRODUCTION .................................................................................................. 29
MATERIALS AND METHODS ........................................................................... 32
RESULTS .............................................................................................................. 37
DISCUSSION ........................................................................................................ 42
LITERATURE CITED .......................................................................................... 45
CHAPTER 3: The spatiotemporal genetic structure of Phytophthora capsici in Michigan
and implications for disease management ............................................................. 48
ABSTRACT ........................................................................................................... 48
INTRODUCTION .................................................................................................. 48
MATERIALS AND METHODS ........................................................................... 52
RESULTS .............................................................................................................. 58
DISCUSSION ........................................................................................................ 68
LITERATURE CITED .......................................................................................... 74

vi

LIST OF TABLES

TABLE 1.1: Phytophthora capsici isolates described by collection in 1998. .................. 20

TABLE 1.2: Phenotypic diversity of 223 Phytophthora capsici oospores isolated from a
single naturally infected cucumber .................................................................................... 21

TABLE 1.3: Chi-square analysis of Phytophthora capsici crosses for segregation
and linkage of compatibility type (CT) and mefenoxam sensitivity (MS) ......................... 23

TABLE 2.1: Fixation indices (FST) for 37 AF LP loci from unique Phytophthora capsici
isolates collected from a single Michigan cucurbit ﬁeld during 1998 (N = 57) and 1999
(N = 141) ............................................................................................................................ 39

TABLE 2.2: Phenotypic diversity of Phytophthora capsici isolates recovered from the
same cucurbit ﬁeld in 1998, 1999, and 2000 ..................................................................... 40

TABLE 2.3: Clone contribution of ﬁfteen Phytophthora capsici isolates to the total
number of isolates collected in 1999 (N = 200) ................................................................. 41

TABLE 3.1: Inheritance of 17 AFLP markers, compatibility type (CT), and mefenoxam
sensitivity (MS) in 107 progeny of a cross between Phytophthora capsici isolates OP97
(Al/IS) and SFF3 (AZ/S) ................................................................................................... 59

TABLE 3.2: Population, number of isolates, total number of AF LP bands, number of
polymorphic bands, and estimated heterozygosity for populations of Phytophthora capsici
in Michigan ........................................................................................................................ 60

TABLE 3.3: Clonal component of genotypic diversity within sample sets of
Phytophthora capsici from Michigan ................................................................................ 61

TABLE 3.4: F -statisics (@ST) (below diagonal) and geographical distances (in km, above
diagonal) between Phytophthora capsici sample sets collected from single locations over
time and different locations in Michigan ........................................................................... 62

TABLE 3.5: Results of nested analysis of molecular variance (AMOVA) for
Phytophthora capsici isolates based on 94 AFLP markers. Variance is partitioned (A)
between 1998 and 1999 at SWl-A, (B) between 1999 and 2000 at SWI-B, (C) between
combined sample sets from SWI-A and SWl-B, and (D) within and between samples
sets from seven locations in Michigan ............................................................................... 63

TABLE 3.6: Location, year, hosts, compatibility type, and mefenoxam sensitivity of

genetically unique Phytophthora capsici isolates collected in Michigan between 1998 and
2000 .................................................................................................................................... 65

vii

LIST OF FIGURES

FIGURE 1.1 : Spatial distribution of Phytophthora capsici isolates illustrating
compatibility type ( open circles = A2, open stars = A1), ‘

infections with oospores (black diamonds),and mefenoxam sensitivity ( open circles/
open stars = sensitive, patterned circles/ patterned stars = intermediate sensitivity, and
black circles/ black stars = insensitive) for cucurbit ﬁelds in A, south central (SCIA); B,
northwestern (N W2B); and C, southwestern (SWlA) Michigan sampled during 1998.
Each symbol represents a single isolate from a single infected plant or cucurbit fruit ...... 18

FIGURE 1.2: Frequency of mefenoxam sensitivity in Phytophthora capsici for A, 1997
to 1998 ﬁeld isolates and in vitro progeny from crosses: B, SLCC-6B x 0P9; C, SSB98 x
OP97; D, SF F-3 x OP97; and E, 244 x 216. S = sensitive, <30% growth of control (GC);
IS = intermediate, 30 to 90% GC; I = insensitive, > 90% GC; and

N = number of isolates ....................................................................................................... 19

FIGURE 3.1: UPGMA cluster analysis of Phytophthora capsici isolates from location
SWI-B over 1999 and 2000 (N=58) and SWI-A in 1998 (N=57) based on the Jaccard
similarity coefﬁcient using 94 ampliﬁed fragment length polymorphism (AF LP) markers.
Nodes contain isolates exclusively from single locations. Location identiﬁers precede the
inclusive node and are indicated by region (S = south, N = north, W = west, and C =
central) and a farm identiﬁer (1 ,2,...n) prior to the hyphen with a ﬁeld indicator (A, B,...n)
and the year of sampling (eg; 00 = 2000) following .......................................................... 66

FIGURE 3.2: UPGMA cluster analysis of 255 Phytophthora capsici isolates from seven
locations in Michigan based on the Jaccard similarity coefficient using 94 ampliﬁed
fragment length polymorphism (AFLP) markers. Nodes contain isolates exclusively from
single locations. Location identiﬁers precede the inclusive node and are indicated by
region (S = south, N = north, W = west, and C = central) and a farm identiﬁer (1,2,...n)
prior to the hyphen with a ﬁeld indicator (A, B,...n) and the year of sampling (eg; 00 =
2000) following .................................................................................................................. 67

viii

LITERATURE REVIEW

Phytophthora capsici was ﬁrst described by Leon H. Leonian in 1922 (23).
Leonian reported considerable damage to chili pepper plants by a novel species of
Phytophthora in the fall of 1918 on the farm of the New Mexico Experiment Station. In
1919, the same disease reappeared in this and surrounding farms. The initial in vitro
characterization describes a highly torturous hyphal growth morphology, papillate
sporangia, and suggests that oospores are formed in single cultures. The modern
description of P. capsici as a species falls into Waterhouses Group II (39) and is
characterized by sporangia that are conspicuously papillate with amphigynous oospores
generally forming only when A1 and A2 mating types are paired. The ﬁrst literature
citing of P. capsici on a cucurbit host was reported by W. A. Kreutzer in 1937 (21).
Kreutzer reports that disease was conﬁned to an 8-acre cucumber ﬁeld with 100 percent
of the cucumber fruits infected (21). By 1940, in addition to pepper, P. capsici had been
described on eggplant, honeydew melon fruit, summer squash, and tomato fruit (22, 40).
During the thirty years following these initial reports there are occasional entries into the
literature describing P. capsici on additional cucurbit hosts as well as a range of more
exotic hosts (12), but overall there is little new information published (6, 38).

Information concerning the different spore types produced by members of the
genus Phytophthora accumulated slowly between 1940 and 1970. In 1970, Waterhouse
provided a useful, and still used, key for identifying isolates to species based on the
determination of whether or not an isolate could produce oospores in single culture and
the morphology of sporangia and oospores (3 9). Research with other Phytophthora

species established much of what is known about the three dominant spore types

produced by P. capsici (12). The thallus is composed of coenocytic mycelium which may
give rise to lemon-shaped sporangia born on long caducous pedicels (1). When sporangia
are immersed in free water they differentiate to produce 20-40 bi-motile swimming
zoospores (2). Zoospores exhibit negative geotropism and chemotactically follow
nutrient gradients while swimming (12). Once zoospores contact the plant surface they
encyst and germinate to produce germ tubes (17). Penetration of leaf surfaces by P.
capsici has been shown to occur directly and through natural openings such as stomata
(19). Phytophthora capsici produces an extracellular macerating enzyme which likely
plays a signiﬁcant role in breaching the host epidermis and ramifying through susceptible
host tissue (41 ). Approximately half of the sixty recognized species of Phytophthora are
self-fertile (homothallic) while the other half, including P. capsici, generally produce
oospores only when both mating types are present (heterothallic) (16). Oospores are
formed when Al and A2 compatibility types come into close association (20). Each of
the parent isolates make both male (antheridium) and female (oogonium) gametangia
once the sexual stage has been initiated and self-fertilization is possible in obligately
outcrossing species (20).

The asexual sporangia and zoospores proved to be much easier to manipulate and
study than the oospore and it is not surprising that the salient features of these spore types
were outlined relatively early (17, 25). The main impediment to detailed studies of
oospores and the inherent genetics therein was due primarily to difﬁculties in separating
and germinating the oospores (36). In 1968, Satour and Butler provide crucial
information concerning the generation and germination of P. capsici oospores (34). They

report that relatively young oospores produced in paired cultures of P. capsici germinated

to produce recombinant progeny after 30 days incubation. Prior to this it was generally
thought that 6-9 month incubation periods were necessary for oospore germination. The
progeny from their crosses were shown to differ from the parental types in both
morphology and pathogenicity. Signiﬁcantly, one progeny isolate exhibited increased
virulence on pepper compared to either of the parents. A number of important
milestones were reached in this investigation. A simple method for the production,
germination, and harvesting of oospore progeny for P. capsici was formally presented and
the authors convincingly argued that proper media containing ample nutrients as well as
genetically compatible parent isolates are required for successful matings. In addition,
this work provided convincing evidence for the potential role of oospores in generating
genetic variation (34). In 1971, Polach and Webster corroborated this ﬁnding using the
oospore techniques recently described (28). They investigated 391 single oospore
progeny from four mating reactions and report that the parent isolates differed in their
pathogenicity to cucurbit and solanaceous hosts and that segregation and recombination
were observed for all the characters studied.

Early investigators recognized that the genus Phytophthora exhibited some
striking dissimilarities to many other fungal organisms, but a full resolution of its
taxonomic and evolutionary standing would not be made until DNA sequence analysis
was completed by Forster et al. in 1990. They found that oomycetes are more closely
related to heterokont photosynthetic algae than to members of the kingdom Fungi (13).
Signiﬁcant investigations into the genetics of P. capsici do not appear in the literature
until the late 1980's and early 1990's when isozyrne and restriction fragment length

polymorphism (RFLP) analysis of both mitochondreal and nuclear DNA were conducted

on isolates from widely different geographical locations, years, and hosts located in a
worldwide Phytophthora culture collection at the University of California at Riverside
(14, 24, 26). An isozyme study involving 113 P. capsici isolates was interpreted as
revealing two subgroups within the P. capsici species (24). Subgroups are deﬁned as
being signiﬁcantly different based on sporangia morphology and ontogeny. RFLP
investigation of mitochondreal DNA revealed no patterns of similarity based on host or
geographical location (14). RF LP analysis of nuclear DNA’s using low copy number
probes of ﬁfteen P. capsici isolates indicated nuclear DNA diversity was high (14).
There are no published studies of isozyme or DNA level diversity within single
populations of P. capsici.

In 1981, the presence of Al and A2 isolates of P. capsici from single ﬁelds was
reported in New Jersey (27). The investigators provide the only report of naturally
occurring P. capsici oospores being observed in diseased host tissue in North America.
This investigation focused on the oospore as a means for survival and does not address
the potential role of the sexual stage in generating variability. In 1990, the presence of
A1 and A2 isolates of P. capsici was described in single ﬁelds in North Carolina (29).
The author characterized morphological variation in ﬁeld isolates in an attempt to
determine if there are signiﬁcant differences between isolates infecting cucurbit or
solanaceous hosts. The morphological characters studied exhibited continuous rather
than discrete variation within populations based on host type and the overall conclusion
was that a combination of molecular and classical taxonomic approaches may be
necessary to further delimit the species. The author reports that oospores may be formed

in the ﬁeld but does not discuss the possibility that the sexual stage may have contributed

to the observed variability (29).

Much of the published literature on P. capsici concerns the environmental
conditions favorable for infection and spread. The most prominent and recurring ﬁnding
is that excess moisture is the single most important component to the initial infection and
subsequent spread of P. capsici (3, 30, 31, 33, 35, 37). This follows in the wake of
similar ﬁndings for many species in the genus Phytophthora and is not surprising in light
of this organisms evolutionary ties to the algae (10, 11). Recommended control strategies
reﬂect our understanding of the importance of water in the epidemiology of P. capsici
and include planting into well drained ﬁelds and into raised beds whenever possible (32).
Rotation to non-susceptible hosts is recommended for at least 2-3 years and further
recommendations suggest that the highly active phenylamide fungicide mefenoxam may
be useful in preventing disease. Mefenoxam is a relatively new compound which has the
same mode of action as metalaxyl (9). Metalaxyl has been shown to speciﬁcally inhibit
the incorporation of uridine into RNA in sensitive oomycetes (9). Metalaxyl was ﬁrst
used on a large scale by potato growers to control epidemics caused by P. infestans in the
early 1980's in Europe (8). As early as 1981 researchers working with P. capsici
demonstrated that insensitivity to metalaxyl was readily selected for using sub-lethally
amended media (4, 5). Insensitivity soon developed in natural populations of oomycetous
organisms where metalaxyl was heavily relied upon (7-9, 15, 18).

When the following research was initiated in 1997 our understanding of P. capsici
was limited to a relatively good understanding of the asexual phase of disease
development and did not include many studies speciﬁcally exploring the impact, or lack

thereof, that the sexual stage may have on the life history of P. capsici. Mefenoxam was

being applied by some growers as a part of a Phytophthora management strategy and the

sensitivity of natural populations of P. capsici in Michigan to mefenoxam was unknown.

10.

ll.

LITERATURE CITED

Alconero, R., and Santiago, A. 1972. Characteristics of asexual sporulation in
Phytophthora palmivora and Phytophthora parasitica nicotianae. Phytopathology
62:993-997.

Bernhardt, E. A., and Grogan, R. G. 1982. Effect of soil matric potential on the
formation and indirect germination of sporangia of Phytophthora parasitica,
Phytophthora capsici, and Phytophthora cryptogea rots of tomatoes,
Lycopersicon esculentum. Phytopathology 72:507-51 1.

Bowers, J. H., and Mitchell, D. J. 1990. Effect of soil-water matric potential and
periodic ﬂooding on mortality of pepper caused by Phytophthora capsici.
Phytopathology 80:1447-1450.

Bruin, G. C., and Edgington, L. V. 1980. Induced resistance to ridomil of some
oomycetes. Phytopathology 70:459.

Bruin, G. C. A. 1981. Adaptive resistance in Peronosporales to metalaxyl. Can. J.
Plant Path. 3:201 -206.

Crossan, D. F., Haasis, F. A., and Ellis, D. E. 1954. Phytophthora blight of
summer squash. Plant Dis. Rep. 38:557-559.

Crute, I. R. 1987. The occurrence, characteristics, distribution, genetics, and
control of a metalaxyl-resistant pathotype of Bremia lactucae in the United
Kingdom. Plant Dis. 71:763-767.

Davidse, L. C., Henken, J., Van Dalen, A., Jespers, A. B. K., and Mantel, B. C.
1989. Nine years of practical experience with phenylamide resistance in
Phytophthora infestans in the Netherlands. Neth. J. P1. Path. 95:197-213.

Davidse, L. C., van den Berg-Vclthuis, G. C. M., Mantel, B. C., and Jespers, A. B.
K. 1991. Phenylamides and Phytophthora. Pages 349-360 in: Phytophthora. J. A.
Lucas, R. C. Shattock, D. S. Shaw, and L. R. Cooke, Eds. British Mycological
Society, Cambridge.

Duniway, J. M. 1979. Water relations of water molds. Ann. Rev. Phytopahtol.
1 7:43 1460.

Duniway, J. M. 1983. Role of Physical Factors in the Development of

6

12.

13.

14.

15.

16.

17.

18.

19.

20.

21.

22.

23.

24.

Phytophthora diseases. Pages 175-187 in: Phytophthora: Its Biology, Taxonomy,
Ecology, and Pathology. D. C. Erwin, S. Bartnicki-Garcia, and P. H. Tsao, Eds.
APS Press, St. Paul, MN.

Erwin, D. C., and Ribeiro, O. K. 1996. Phytophthora Diseases Worldwide. The
American Phytopathological Society, St. Paul, MN.

Forster, H., Coffey, M. D., Elwood, H., and Sogin, M. L. 1990. Sequence analysis
of the small subunit ribosomal RN As of three zoosporic fungi and implications
for fungal evolution. Mycologia 82:306-312.

Forster, H., Oudemans, P., and Coffey, M. D. 1989. Mitochondreal and nuclear
DNA diversity within six species of Phytophthora. Exp. Mycology 14:18-31.

Georgopoulos, S. G., and Grigoriu, A. C. 1981. Metalaxyl-resistant strains of
Pseudoperonospora cubensis in cucumber greenhouses of southern Greece. Plant
Dis. 65:729-731.

Goodwin, S. B. 1997. The population genetics of Phytophthora. Phytopathology
87:462-473.

Hickman, C. J. 1970. Biology of Phytophthora Zoospores. Phytopathology
60:1128-1135.

Katan, T., and Bashi, E. 1981. Resistance to metalaxyl in isolates of
Pseudoperonospora cubensis, the downy mildew pathogen of cucurbits. Plant Dis.
65:798-800.

Katsura, K., and Miyazaki, S. 1960. Leaf Penetration by Phytophthora capsici
Leonian. Sci. Rept. Kyoto Prefect. Univ. Agr. 12:65-70.

Ko, W. 1988. Hormonal heterothallism and homothallism in Phytophthora. Ann.
Rev. Phytopathol. 26:57-73.

Kreutzer, W. A. 1937. A Phytophthora rot of cucumber fruit. Phytopathology
27:955.

Kreutzer, W. A., Bodine, E. W., and Durrell, L. W. 1940. Cucurbit diseases and
rot of tomato fruit caused by Phytophthora capsici. Phytopathology 30:972-976.

Leonian, L. H. 1922. Stern and ﬁ'uit blight of peppers caused by Phytophthora
capsici sp. nov. Phytopathology 12:401-408.

Mchau, G. R. A., and Coffey, M. D. 1995. Evidence for the existence of two
subpopulations in Phytophthora capsici and a redescription of the species. Mycol.
Res. 99:89-102.

25.

26.

27.

28.

29.

30.

31.

32.

33.

34.

35.

36.

37.

Minogue, K. P., and Fry, W. E. 1981. Effect of temperature, relative humidity,
and rehydration rate on germination of dried sporangia of Phytophthora infestans.
Phytopathology 71:1181-1184.

Oudemans, P., and Coffey, M. D. 1991c. Relationships between Phytophthora

species: Evidence from isozyme analysis. Pages 184-203 in: Phytophthora. J. A.
Lucas, R. C. Shattock, D. S. Shaw, and L. R. Cooke, Eds. Cambridge Univ. Press,
Cambridge.

Papavizas, G. C., Bowers, J. H., and Johnston, S. A. 1981. Selective Isolation of
Phytophthora capsici from Soils. Phytopathology 71:129-133.

Polach, F. J ., and Webster, R. K. 1972. Identiﬁcation of strains and inheritance of
pathogenicity in Phytophthora capsici. Phytopathology 62:20-26.

Ristaino, J. B. 1990. Intraspeciﬁc variation among isolates of Phytophthora
capsici from pepper and cucurbit ﬁelds in North Carolina. Phytopathology
80: 1253-1259.

Ristaino, J. B. 1991. Inﬂuence of rainfall, drip irrigation, and inoculum density on
the development of Phytophthora root and crown rot epidemics and yield in bell
pepper. Phytopathology 81:922-929.

Ristaino, J. B., Hord, M. J., and Gumpertz, M. L. 1992. Population densities of
Phytophthora capsici in ﬁeld soils in relation to drip irrigation, rainfall, and
disease incidence. Plant Dis. 76: 1017-1024.

Ristaino, J. B., and Johnston, S. A. 1999. Ecologically based approaches to
management of Phytophthora blight on bell pepper. Plant Dis. 83: 1080-1089.

Ristaino, J. B., Larkin, R. P., and Campbell, C. L. 1993. Spatial and temporal
dynamics of Phytophthora epidemics in commercial bell pepper ﬁelds.
Phytopathology 83:1312-1320.

Satour, M. M., and Butler, E. E. 1968. Comparative morphological and
physiological studies of the progenies from intraspeciﬁec matings of
Phytophthora capsici. Phytopathology 58:183-192.

Schlub, R. L. 1983. Epidemiology of Phytophthora capsici on bell pepper. J.
agric. Sci., Camb. 100:7-11.

Shaw, D. S. 1967. A method of obtaining single-oospore cultures of
Phytophthora cactorum using live water snails. Phytopathology 57:454.

Springer, J. K., and Johnston, S. A. 1982. Black polyethylene mulch and
Phytophthora blight of pepper. Plant Dis. 66:281.

8

38.

39.

40.

41.

Tompkins, C. M. 1941. Root rot of pepper and pumpkin caused by Phytophthora
capsici. J.“ Agr. Res. 63:417-426.

Waterhouse, G. M. 1970. Taxonomy of Phytophthora. Phytopathology 60:1 141-
1143.

Wiant, J. S. 1940. A rot of winter queen watermelons caused by Phytophthora
capsici. J. Agr. Res. 60:73-88.

Yoshikawa, M., Tsukadaira, T., Masago, H., and Minoura, S. 1977. A non-
pectolytic protein from Phytophthora capsici that macerates plant tissue. Physiol.
Plant Pathol. 11:61-70.

Chapter 1: Mefenoxam insensitivity and the sexual stage of Phytophthora capsici
in Michigan cucurbit ﬁelds.

ABSTRACT

Lamour, K. H., and Hausbeck, M. K. 2000. Mefenoxam insensitivity and the sexual
stage of Phytophthora capsici in Michigan cucurbit ﬁelds. Phytopathology 90:396-400

The potential for outcrossing, occurrence of oospores, and inheritance of
mefenoxam sensitivity was assessed in naturally occurring populations of Phytopthora
capsici. Between 1997 and 1998 14 farms were sampled with 473 isolates recovered
from cucurbit hosts and 30 from bell pepper. The Al and A2 compatibility types were
recovered in a roughly 1:1 ratio in 8 of 14 farms with sample sizes greater than 15. In
1997, one isolate was designated as insensitive and four as sensitive to mefenoxam. In
1998, 55% of the 498 isolates sampled were sensitive, 32% were intermediate, and 13%
were fully insensitive to mefenoxam. In vitro characterization of mefenoxam sensitivity
was conducted by crossing ﬁeld isolates. Chi-square analysis of crosses between
sensitive, intermediately sensitive, and insensitive isolates indicate that mefenoxam
insensitivity segregate as an incompletely dominant trait unlinked to compatibility type (P
= 0.05). Oospores were observed in diseased cucurbit fruit from four farms in 1998 and
223 oospore progeny were recovered from a single diseased cucumber. All six
mefenoxam sensitivity/compatibility type combinations were present in these oospore
progeny and within single ﬁelds. Based on these ﬁndings we conclude that oospores
likely play a role in the survival of P. capsici and that sexual recombination may
signiﬁcantly inﬂuence population structure.

INTRODUCTION

Phytophthora capsici Leonian causes root, crown, and fruit rot on solanaceous

10

and cucurbit hosts worldwide (6, 11, 16). In Michigan, this pathogen causes serious
damage annually to peppers, cucumbers, pumpkins, and squash. The asexual phase of the
life cycle includes a mycelial thallus which produces extracellular enzymes capable of
macerating host tissue (31), papillate sporangia borne on long (>20um) caducous pedicels
(29), and biﬂagellate, chemotactic, negatively geotropic zoospores which are liberated
from sporangia under free water conditions (26). The sexual phase occurs when isolates
of opposite compatibility type (designated Al and A2) are in close proximity leading to
the formation of thick-walled oospores. Oospores require an indeterminate time period (2
weeks to 3 months) for maturation, remain viable for extended periods of time (years),
and are thought to be the primary survival structure (16). Because P. capsici’ is
heterothallic, oospores have the potential to represent new gene combinations in addition
to their role as long-terrn dormant inoculum (11).

The principal methods of controlling P. capsici include cultural practices and the
use of fungicides. Recommended cultural strategies are aimed at avoiding a build-up of
inoculum by rotating to non-susceptible hosts and planting in well-drained ﬁelds (25).
Oomycetes, although morphologically similar to true fungi, are genetically and
biochemically dissimilar (11) and are not susceptible to most broad-spectrurn fungicides
(8). For this reason growers tend to rely on a limited number of fungicides. The
phenylamide class of fungicides (PAF), speciﬁcally metalaxyl and the newest formulation
mefenoxam (Novartis, Greensboro, NC), have been used on cucurbits in Michigan.
Mefenoxam is the active enantiomer contained in the racemic fungicide metalaxyl (22).
The mode of action of metalaxyl is postulated to be site speciﬁc and it was not surprising

when resistance surfaced in populations of susceptible plant pathogens after the PAFs

11

were introduced in the late 19705 (8). Researchers investigating a range of obligate and
hemibiotrophic oomycetes suggest that tolerance to the PAFs is conditioned by a single
locus with major effect exhibiting incomplete dominance subject to modiﬁcation by
genes of minor effect (2, 5, 7, 12, 28). Insensitivity to mefenoxam, which also conferred
insensitivity to metalaxyl, has recently been reported from ﬁeld populations of P. capsici
on bell pepper (22).

The potential for outcrossing in epidemic populations of P. capsici has been
reported based on the recovery of both the A1 and A2 mating types ﬁ'om soil samples and
diseased plant material from single ﬁelds (20, 21, 24). Typical amphyginous oospores
have been reported in the US. on a single occasion from diseased bell pepper tissue (21).
In vitro analysis indicates that P. capsici oospore progeny are recombinant for
pathogenicity to various hosts and mating type (3, 23).

This paper reports on the initial phase of an investigation into the biology and
structure of P. capsici populations in Michigan vegetable production ﬁelds. The primary
objective was to test the hypothesis that the sexual stage of P. capsici occurs in Michigan
vegetable production ﬁelds. In addition, we tested the hypothesis that mefenoxam
resistance is inherited as a single dominant gene (19).

MATERIALS AND METHODS

Sampling strategy and pathogen isolation. Diseased plant material was
collected in a haphazard fashion from ﬁelds with limited disease (e.g., a single focus of
infection). In ﬁelds with widespread disease, the spatial distribution of isolates was
documented using permanent structures (e.g., telephone poles, houses, and barns) to

construct ﬁeld-speciﬁc grid patterns prior to sampling (Figure 1.1). Grid block areas

12

varied from approximately 40 m2 to 12 km2 (Table 1.1). Infected plant material was
selected at random from within blocks.

Diseased tissue was surface sterilized with 70% EtOH for approximately 10 s, and
small pieces of actively expanding lesions were plated onto 100 x 15 mm BARP
(Benomyl 25ppm, Ampicillin lOOppm, Rifarnpicin 30ppm, and Pentachloronitrobenzene
lOOppm) amended UCV8 (840 ml distilled water, 163 ml un-clariﬁed V8 juice, 3 g
CaCO3, 16 g Bacto agar) plates (27). Plates were wrapped with Paraﬁlm and incubated
for 3-10 days in the dark at 23 - 25°C. Once an 8 cm diameter or larger colony had
developed, the Paraﬁlm was removed from the petri dish and the plates were incubated
under lab lighting at 23 - 25°C for 2-3 days. This protocol stimulated ample sporangia
production for zoospore release. Single-zoospore isolates were obtained by ﬂooding the
plates with sterile distilled water (SDW), placing plates into a 10°C refrigerator for 30
min, incubating the plates at 23-25°C for 30 min, placing four drops of the zoospore
solution onto 100 x 15 cm water agar plates and tilting the plates to get streaks of
zoospores on the water agar surface. Plates were incubated 45-90 min at 23-25°C and
single germinated zoospores transferred to UCV8 using a dissecting microscope (25 X)
(11). Single-zoospore cultures were stored on UCV8 plates at 15°C and transferred
monthly or bi-monthly. In addition, two 7 mm plugs were placed into 20 ml screw-top
vials with 2 sterilized hemp seeds and 10 ml SDW water, incubated at 23-25°C for 2
weeks under laboratory lighting, and stored long term at 15°C (11).

Determination of compatibility type. Agar plugs from the edge of an expanding
single- zoospore derived colony were placed at the center of UCV8 plates approximately

2 cm from ATCC (American Type Culture Collection, Rockville, MD) isolate 15427 (A1

13

compatibility type) and ATCC 15399 (A2 compatibility type), incubated at 23-25°C in the
dark for 3-6 days, and compatibility type determined. Thereafter, all compatibility type
determinations were accomplished using the OP97 (A1) and SP98 (A2) ﬁeld isolates.

In vitro response to mefenoxam. Agar plugs from the edge of actively
expanding single-zoospore colonies were placed at the center of 100 x 15 cm UCV8
plates amended with 0 and 100 ppm mefenoxam (Ridomil Gold EC, Novartis,
Greensboro, NC; 48% AI, suspended in SDW; added to UCV8 cooled to 49°C).
Inoculated plates were incubated at 23-25°C for 3 days and colony diameters measured.
Percent growth of an isolate on amended media was calculated by subtracting the
inoculation plug diameter (7 mm) from the diameter of each colony and dividing the
average diameter of the amended plates by the average diameter of the unamended
control plates (9). All tests were conducted at least twice. Field isolates were assigned
putative mefenoxam sensitivities based on the percent growth of the control as
determined above. An isolate was scored as sensitive if growth at lOOppm was less than
30% of the control, intermediately sensitive if growth was between 30 and 90% of the
control, and insensitive if growth was greater than 90% of the control. Preliminary
cutoffs were determined by visual assessment of the frequency of mefenoxam sensitivity
in ﬁeld isolates (Figure 1.2A).

Segregation analysis. The validity of these mefenoxam sensitivity assignments
was tested by crossing isolates representative of the three groups from within and
between diverse geographical locations and comparing the ratios of progeny phenotypes
to those expected under Mendelian inheritance using chi-square analysis. Isolate OP97

(A1, sensitive, farm NW2B) was crossed with isolates SSB98 (A2, insensitive, farm

14

SWIA), SLCC-6B (A2, sensitive, farm SClB), and SPF-3 (A2, intermediate sensitivity,
farm SClA) and isolate 216 (Al, intermediate sensitivity, farm SClA ) was crossed with
244 (A2, intermediate sensitivity, farm SClA). Crossing was performed as described in
the compatibility type screen. Plates were incubated at 23-25°C in the dark for 3-4
months (15) and the surface of a lcm square area was scraped from the zone containing
oospores between the inoculation plugs. Scrapings were placed in 10 ml SDW and
homogenized in a Sorvall mixer (Ivan Sorvall, Inc., Norwalk, CT) on the highest setting
for 3 min. Novozyme (Sigma, St. Louis, MO) (5mg/ml) was added to the homogenate
and the solutions incubated on a shaker (200 rpm) ovemight (18 - 24 h). Solutions were
then diluted 1 to 10 with SDW and incubated in 15 x 100 petri dishes under ﬂourescent
lighting (1 ). After approximately 24 h, germinated oospores were retrieved from the
suspensions using a suction device constructed from a pasteur pippette (11). Gerrninated
oospores are characterized by dissolution of the electron dense thick-wall and the
presence of one to many germ tubes with or without terminal sporangia. Individual
oospores, were transferred to water agar plates and after 1-2 days single hyphal tips were
transferred to UCV8 plates. Single-oospore hyphal tip progeny were then screened for
mating type and sensitivity to mefonoxam as described above.

Naturally occurring oospores. Selected infected fruits were observed for the
presence or absence of oospores. Slides were prepared by excising a thin slice of suspect
tissue, staining with crystal violet (300 ppm), and inspecting for typical amphigynous
oospores under a light microscope (Leitz Laborlux S, Wetzlar, West Germany). Diseased
plant material containing amphigynous oospores typical for P. capsici was surface

sterilized with 70% EtOH for approximately 10 s and a 1 cm section embedded in BARP-

15

UCV8 media. Plates were incubated for 3-4 months in the dark. Following incubation
the initial diseased plant material was removed and subjected to the germination
procedure described in segregation analysis.

RESULTS

Mefenoxam sensitivity and compatibility type among isolates. Single isolates
of P. capsici were recovered from ﬁve farms in 1997; 1 from bell pepper (A2, resistant)
and 4 from cucurbit hosts (A1, sensitive). In 1998, 498 isolates were obtained from 11
farms; 468 from cucurbit hosts and 30 from bell pepper (Table 1.1). In 1998, 258 A1 and
240 A2 compatibility types were recovered with both compatibility types being found in
every ﬁeld sampled. Eight of ten ﬁelds with sample sizes greater than 15 had an
approximate 1:1 ratio of A1 and A2 compatibility types (Table 1.1). When ﬁelds were
sampled on a grid, both A1 and A2 compatibility types were located within and among
geographically diverse quadrants (Figure 1.1).

In 1997, one isolate was designated as insensitive and four as sensitive to
mefenoxam. In 1998, 55% (274) of the isolates were sensitive, 32% (161) were
intermediate, and 13% (63) were fully insensitive to mefenoxam (Table 1.1). Of the 14
farms sampled during 1997 and 1998, 43% had fully insensitive isolates, and 79% had
isolates of intermediate sensitivity (Table 1.1). Insensitive isolates were not recovered
from 6 of the farms sampled. With one exception, fully sensitive isolates were recovered
from all farms sampled (Figure 1.2A).

Three ﬁelds with sample sizes greater than 15 were of particular interest because
they represent predominantly sensitive, intermediately sensitive, or fully insensitive

populations of P. capsici (Figure 1.1). Within ﬁeld SWIA, in the southwestern region,

16

87% of the isolates (56) were fully insensitive to mefenoxam with no sensitive isolates
being recovered (Table 1.1). Both A1 and A2 isolates were recovered from 9 of the 13
quadrants sampled (Figure 1.1C). Whereas, in ﬁeld SW1 B,on the same farm, the
majority (62%) of the isolates (56) had intermediate sensitivity and 20% were fully
sensitive (Table 1.1). When four other ﬁelds in southwest Michigan were sampled in a
limited manner (i.e., fewer than 16 samples), an insensitive isolate was recovered in one
ﬁeld (Table 1.1). In south central Michigan, intermediately sensitive strains dominated in
all ﬁelds (Table 1.1). Isolates (145) collected on a Single day from a ﬁeld of pickling
cucumbers in south central Michigan represent all six combinations of mefenoxam
sensitivity and compatibility type with 17% sensitive A2, 20% sensitive A1, 28%
intermediately sensitive A2, 32% intermediately sensitive A1, 1% insensitive A2, and 2%
insensitive Al isolates (Figure 1.1A). In northwest Michigan, completely sensitive
strains dominated all ﬁelds sampled (Table 1.1). In one of the seven grids where two

samples were collected both A1 and A2 compatibility types were present (Figure 1.1B).

l7

Ag «3.303 8 Edi 3805: 03:6

a 80.: 820mm Bwﬁm a $5852 758.? comm .33 mats—G BEES gmﬁoaz $2 39 5883538 .U can ”Am—N336 53835.8:
.m 5209 .868 5:8 .< E mEom “5525 8» 83:28.: n maﬁa v.83 \mﬂemu x83 Pa 53383 233—825 H mam @2533
B286 38038 63:28 n Em some B286 some v 53238 Saxocouoﬁ 2235.56 x228 ween—moo 5MB maouoomcm

A _< n mam some .N< u v.28? some v 25 mam—3:888 33862: 833mm 8.2.58 REESE: mo 8:32:66 33an .2 0.8me

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

18

 

 

 

 

 

 

 

 

 

 

 

oo o... o 44 440 £04440
44 444444 0034
O O
0044 4444 Go 4 o c
44444
GO 004444ooww_o
044 o 44 O
e 000410444 0 00—9490
«Lu 0 BBiiﬁﬁO OOLOOOO.O
004 4444 4+
40 4 0400404004 4
0004 4 4 0 4444440444 00
o
3 :. E. f. o “Worms...“Umwmmnoo
04 4 04 4 O 44
400 4444400009000
4 4 4 4 4 4 4 44 O 4 O 4 4444—4444440444

 

 

 

 

 

 

 

 

 

 

 

 

   
 
 

 

 

 

 

 

 

 

 

 

 

 

200 A
150 _ Field Isolate?
Na 52
100
so
0
80 *- B
so - "3};
m 40 —
.9 20 _
2
O _
.2 °
'5 so
'~
320
m
0
- 10
D.
H-
O o
3 so
.n lsxs
E __ Isa-112i
3 4o
Z
20 —
o
E
2° ' ISXIS
NI106
1o —
o

 

o 20 40 so 80100 120
Percent growth of control
Figure 2. Frequency of mefenoxam sensitivity in Phytophthora capsici for A, 1997 to
1998 ﬁeld isolates and in vitro progeny from crosses: B, SLCC-6B x 0P9; C, SSB98
x OP97; D, SPF-3 x OP97; and E, 244 x 216. S = sensitive, <30% growth of control
(GC); IS = intermediate, 30 to 90% GC; I = insensitive, > 90% GC; and N = number
of isolates.

19

TABLE 1.1. Phytophthora capsici isolates described by collection in 1998

 

Compatibility Mefenoxam
Number of type (%) sensitivityg (%)

 

 

Locationa Host° Sampling isolates A1 A2 S _ IS I
SWIA S, P c"‘4 km2 / (l-5)° 56 61 39 0 13 87
SWIB S, C 40 In2 / (0-3) 56 45 55 20 62 18

SW2 P, M, S 12 km2 / (0-6) 16 56 44 81 19 0
SW3 P Hf 6 17 83 8O 20 O
SW4 C H 5 80 2O 2O 60 2O
SW5 P H 4 75 25 40 60 O
SCIA C 1 km2 / (0-4) 145 54 46 36 62 2
SCIB C 6 km2 / (0-6) 82 46 54 44 53 3
SC2 C H 20 9O 10 5 8O 15
C1 P, S H l6 19 81 88 12 0
NW] BP c4 km2 / (0-5) 30 63 37 93 3 4
NW2A S H 29 52 48 90 3
NW2B C, S 40 In2 / (0-3) 28 57 43 100 O 0
NW3 S H 5 2O 8O 8O 2O 0
Total 498 52 48 55 32 13

 

a First uppercase letters denote region, number denotes farm, and A or B represent ﬁelds
on a single farm.

b P = pumpkin, C = cucumber, M = melon, S = squash, BP = bell pepper.

° Refers to samples collected from ﬁrngicide trial plots with mefenoxam as one of the
treatments. d Approximate grid quadrant area in square meters.

° Number of samples recovered per quadrant.

f Haphazard acquisition of diseased material from ﬁelds.

9 S = sensitive, IS = intermediate sensitivity, and I = insensitive.

20

TABLE 1.2. Phenotypic diversity of 223 Phytophthora capsici
oospores isolated from a single naturallLinfected cucumber

 

Compatibility Mefenoxam sensitivity (%)°
type S IS I Total
A1 16 (35) 23 (52) 10 (23) 49 (110)
A2 19(41) 22 (49) 10(23) 51(113)

Total 35 (76) 45 (101) 20 (46)

° S = sensitive, IS = intermediate sensitivity, and I = insensitive.
Percentage of total followed by number of isolates
in parentheses.

 

 

 

Oospores in the ﬁeld. In 1998, typical amphigynous oospores were observed in
diseased cucurbit fruit collected from multiple locations on farms in south central
(cucumber, Figure 1.1A) and southwest Michigan ( butternut squash), and single
locations on additional farms (pumpkin, cucumber) located in these regions. Single-
zoospore isolates were cultured from each of the infected fruit. Two hundred and twenty-
three single-oospore progeny were recovered from a single cucumber from the southwest
region and identiﬁed as P. capsici based on the presence of typical papillate, caducous
sporangia borne on long (>20um) pedicels (11). These progeny exhibited diversity for
mating type and mefenoxam sensitivity (Table 1.2).

In vitro segregation analysis of mefenoxam insensitivity and compatibility
type. Each of the in vitro crosses resulted in greater than 60% germination of oospores.
A cross between sensitive isolates originating from the northern (OP97) and south central
(SLCC-6B) regions of Michigan (Figure 1.2B, Table 3) resulted in 48 A1 and 37 A2
sensitive progeny (Figure 1.2B). Crossing OP97 (sensitive) with an insensitive isolate
(SSB98) from the

southwestern region of Michigan resulted in 33 Al and 35 A2 isolates with 91% of the

21

progeny designated as intermediately sensitive,2% as sensitive, and 7% as insensitive
(Figure 1.2C). OP97 (sensitive) crossed with an intermediately sensitive isolate (SFF3)
from southwestern Michigan resulted in 51 Al and 61 A2 isolates with 53% of the Al
and 58% of the A2 designated as sensitive, and 47% of the A1 and 42% of the A2
designated as intermediately sensitive (Figure 1.2D). A cross between two intermediately
sensitive isolates (244 x 216) from the same ﬁeld in south central Michigan resulted in
10% A1 and 16% A2 sensitive isolates, 20% A1 and 28% A2 intermediately sensitive
isolates, and 11% A1 and 15% A2 insensitive isolates (Figure 1.2E). None of the
progeny sets deviated from the expected ratios for mefenoxam insensitivity segregating as
an incompletely dominant trait or compatibility type segregating in a 1:1 ratio (P = 0.05).
Chi-square tests for linkage between mefenoxam insensitivity and compatibility type in
the progeny of the in vitro sexual crosses SFF 3 x OP97 and 244 x 216 indicate that these

phenotypes are unlinked (P = 0.05) (Table 1.3).

22

TABLE 1.3. Chi-square analysis of Phytophthora capsici crosses for segregation
and linkage of compatibility type (CT) and mefenoxam sensitivity (MS)

Cross Parent Isolate° CT MSb CT MS CT and MS linkage
number isolates origin .

I SLCC-OB SC I B A2 S * 1 2 1c
48:37d
OP97 NW2B Al S
2 SSB98 SWIA A2 I *1 :1

 

 

33:3

OP97 NW2B A1 S 5
3 SFF3 SClA A2 IS *1:1 *1:1 *1:1:l:1
51:61 63:49° 27135225225g

OP97 NW2B Al S

4 244 SClA A2 IS *1:1 *l:2:1 *1:1:2:2:1:1
43:63 27:51:28f 10:17:21 :30:12:16h
216 SClA A1 IS

° First uppercase letters denote region, number denotes farm, and A or B indicate ﬁelds.
b S = sensitive, IS = intermediate sensitivity, and I = insensitive.

° Expected Mendelian ratio.

d A1:A2 progeny.

° S:IS progeny.

fS:IS:I progeny.

Table 1.3 (cont’d).

9 S/Al :S/A2:IS/A1:IS/A2 progeny.

" S/Al :S/A2:IS/Al:IS/A2:I/A1:I/A2 progeny.

* Chi-square value not signiﬁcant at P = 0.05.

 

DISCUSSION
In the past 10 years, Michigan has experienced a steady increase in the incidence
of root, fruit, and crown rot on cucurbits caused by P. capsici. Rotation to non-
susceptible hosts for up to four years, in conjunction with cultural and chemical control
strategies, have not provided economic control. The polycyclic nature of asexual
reproduction and the role of environmental factors such as free water in disease
development are well understood (4), but the role of the sexual stage in natural

populations has not been investigated. Due to the survival capabilities of the oospore and

23

the genetic implications inherent in outcrossing and recombination, sexual reproduction
has the potential to signiﬁcantly affect naturally occurring populations (10, 14).

A hallmark of sexually active populations is the presence of both compatibility
types in a one to one ratio (14, 17). This criterion was met in many of the ﬁelds sampled
during 1998. Sample sets which reﬂect the spatial distribution of isolates indicate that Al
and A2 compatibility types occur throughout epidemic populations.

Once the sexual stage of heterothallic spp. of Phytophthora has been stimulated
there is the potential for selﬁng, as well as outcrossing. This may explain the fully
sensitive and fully insensitive progeny recovered from our SSB98 (insensitive) x OP97
(sensitive) cross. Our initial assignment of three mefenoxam sensitivity classes was
conﬁrmed by chi-square analysis of progeny phenotypes compared to those expected
under Mendelian segregation for a single dominant gene. Our in vitro ﬁndings that
mefenoxam sensitivity in P. capsici is conditioned by a single incompletely dominant
gene and that this phenotype is unlinked to compatibility type provides useful information
for assessing population structure. Although the frequency of mefenoxam sensitivity is
inﬂuenced by the method, frequency, and rate of fungicide application and becomes less
informative in situations where populations are either fully resistant or fully sensitive, it
does provide useful information in ﬁelds with a mixture of phenotypes. The probability
of ﬁnding all six mefenoxam sensitivity/compatibility type combinations in the same ﬁeld
in the absence of sex is unlikely (13, 18).

The recovery and germination of typical amphigynous oospores from naturally
infected cucurbit fruit provides direct evidence for sexual reproduction. It is clear that

these progeny have the potential to serve as a diverse inoculum representing every

24

combination of compatibility type and mefenoxam sensitivity.

Based on the composite picture presented by these various lines of investigation
we conclude that sexual reproduction plays an active role in the survival of P. capsici in
Michigan and may play a signiﬁcant role in shaping population structure. Overwintering
and long-term viability of oospores represents an obvious advantage gained by sexually
active populations of P. capsici, but there may be an additional gain when considering
insensitivity to mefenoxam. A viable strategy for recovering the effectiveness of a
fungicidel management tool in the advent of widespread resistance is to stop using the
fungicide and allow populations to shift back to sensitivity. This strategy is based on the
phenomenon of a ﬁtness cost for resistance and hinges upon the idea that resistant isolates
are less ﬁt and will be out-competed by sensitive isolates with the removal of the
selective pressure (30). If sexual reproduction has played a role in the overwintering and
survival of P. capsici, concomitant with PAP usage, it is possible that outcrossing and
recombination in the population may generate a genotypically diverse array of resistant
isolates. In this scenario the negative impact of genetic hitchhiking which may occur
when resistance occurs within a single clone would be avoided and it would be less likely
that resistant isolates in total harbor ﬁtness disadvantages linked to mefenoxam
insensitivity. Even if the wild type sensitive allele is not pushed to complete extinction
there may be no reason for sensitive isolates to increase disproportionally when the
selection pressure is removed.

The ﬁndings reported in this paper may explain why rotational strategies (> 2
years) to a non-susceptible host have not provided control and may give some insight into

the failure of mefenoxam in controlling blight caused by P. capsici on cucurbits in

25

Michigan.
ACKNOWLEDGMENTS
This work was funded by the Michigan Agricultural Experiment Station (GREEN
initiative), Pickle and Pepper Research Committee, Pickle Packers International, Inc. and
the Pickle Seed Research Fund, Pickle Packers International, Inc. We thank A. M. Jarosz

for critical comments on the manuscript and valuable criticism during this project.

LITERATURE CITED
1. Berg, L. A., and Gallegly, M. E. 1966. Effect of light on oospore germination of
species of Phytophthora (Abstr.). Phytopathology 56: 583.

2. Bhat, R. G., McBlain, B. A., and Schmitthenner, A. F. 1993. The inheritance of
resistance to metalaxyl and to ﬂuorophenylalanine in matings of homothallic
Phytophthora sojae. Mycol. Res. 97: 865-870.

3. Bowers, J. H., and Mitchell, D. J. 1991. Relationship between inoculum level of
Phytophthora capsici and mortality of pepper. Phytopathology 81: 178-184.

4. Cafe Filho, A. C., Duniway, J. M., and Davis, R. M. 1995. Effects of the
frequency of furrow irrigation on root and fruit rots of squash caused by
Phytophthora capsici. Plant Dis. 79: 44-48.

5. Chang, T. T., and Ko, W. H. 1990. Resistance to fungicides and antibiotics in

Phytophthora parasitica: genetic nature and use in hybrid determination.
Phytopathology 80: 1414-1421.

6. Crossan, D. F., Haasis, F. A., and Ellis, D. E. 1954. Phytophthora blight of
summer squash. Plant Dis. Rptr. 38: 557-559.

7. Crute, I. R., and Harrison, J. M. 1988. Studies on the inheritance of resistance to
metalaxyl in Bremia lactucae and on the stability and ﬁtness of ﬁeld isolates.
Plant Pathol. 37: 231-250.

8. Davidse L. C., van den Berg-Velthuis, G. C. M., Mantel, B. C., and Jespers, A. B.
K., 1991. Phenylamides and Phytophthora. Pages 349-3 60 in Phytophthora, J. A.
Lucas, R. C. Shattock, D. S. Shaw, and L. R. Cooke, Eds.,. British Mycological
Society, Cambridge, UK.

9. Deahl, K. L., DeMuth, S. P., Sinden, S. L., and Rivera-Pena, A. 1995.
Identiﬁcation of mating types and metalaxyl resistance in North American

26

10.

11.

12.

13.

14.

15.

16.

17.

18.

19.

20.

21.

22.

populations of Phytophthora infestans. Amer. Pot. J. 72: 35-49.

Drenth, A., Janssen, E. M., and Govers, F. 1995. Formation and survival of
oospores of Phytophthora infestans under natural conditions. Plant Pathol. 44: 86-
94. '

Erwin, D. C., and Ribeiro, O. K. 1996. Phytophthora Diseases Worldwide. APS
Press, St. Paul, MN.

Fabritius, A., Shattock, R. C., and Judelson, H. S. 1997. Genetic analysis of
metalaxyl insenitivity loci in Phytophthora infestans using linked DNA markers.
Phytopathology 87: 1034-1040.

Georgopoulos, S. G. 1988. Genetics and population dynamics. Pages 12-13 in
F ungicide Resistance in North America, C. J. Delp, Ed., APS Press, St. Paul, MN.

Goodwin, S. B. 1997. The population genetics of Phytophthora. Phytopathology
87: 462-473.

Hamish, W. N. 1965. Effect of light on production of oospores and sporangia in
species of Phytophthora. Mycologia: 57: 85-90.

Hwang, B. K., and Kim, C. H. 1995. Phytophthora blight of pepper and its control
in Korea. Plant Dis. 79: 221-227.

McDonald, B. A., and McDermott, J. M. 1993. Population genetics of plant
pathogenic fungi. BioScience 43: 31 1-319.

Milgroom, M. G. 1995. Analysis of population structure in fungal plant
pathogens. Pages 203-219 in Biotechnology: Interdisciplinary Bridges to
Improved Sorghum and Millet Cr0ps. J. F. Leslie and R. A. Fredriksen, Eds., Iowa
State University Press, Ames, IA.

Milgroom, M. G. 1996. Recombination and the multilocus structure of fungal
populations. Annu. Rev. Phytopathol. 34: 457-477.

Miller, S. A., Bhat, R. G., and Schmitthenner, A. F. 1994. Detection of
Phytophthora capsici in pepper and cucurbit crops in Ohio with two commercial
immunoassay kits. Plant Dis. 78: 1042-1046.

Papavizas, G. C., Bowers, J. H., and Johnston, S. A. 1981. Selective isolation of
Phytophthora capsici from soils. Phytopathology 71: 129-133.

Parra, G., and Ristaino, J. 8.1998. Insensitivity to Ridomil Gold (Mefenoxam)

found among ﬁeld isolates of Phytophthora capsici causing Phytophthora blight
on bell pepper in North Carolina and New Jersey. Plant Dis. 82: 711.

27

23.

24.

25.

26.

27.

28.

29.

30.

31.

Polach, F. J ., and Webster, R. K. 1972. Identiﬁcation of strains and inheritance of
pathogenicity in Phytophthora capsici. Phytopathology 62: 20-26.

Ristaino, J. B. 1990. Intraspeciﬁc variation among isolates of Phytophthora
capsici from pepper and cucurbit ﬁelds in North Carolina. Phytopathology 80:
1253-1259.

Ristaino, J. B., Hord, M. J ., and Gumpertz, M. L. 1992. Population densities of

Phytophthora capsici in ﬁeld soils in relation to drip irrigation, rainfall, and
disease incidence. Plant Dis. 76: 1017-1024.

Schlub, R. L. 1983. Epidemiology of Phytophthora capsici on bell pepper. J.
Agric. Sci., Camb. 100: 7-11.

Schmitthenner, A. F., and Bhat, R. G. 1994. Useful Methods for Studying
Phytophthora in the Laboratory. OARDC, Wooster, OH.

Shattock, R. C. 1988. Studies on the inheritance of resistance to metalaxyl in
Phytophthora infestans. Plant Pathol. 37: 4-11.

Tucker, C. M. 1931. Taxonomy of the Genus Phytophthora De Bary. University of
Missouri, Columbia, MO.

Wade, M. 1988. Strategies for preventing or delaying the onset of resistance to
fungicides and for managing resistance occurrences. Pages 14-15 in Fungicide
Resistance in North America. J. Delp, Ed., APS Press, St. Paul, Mn.

Yoshikawa, M., Tsukadaira, T., Masago, H., and Minoura, S. 1977. A non-

pectolytic protein from Phytophthora capsici that macerates plant tissue. Physiol.
Plant Pathol. 11: 61-70.

28

Chapter 2: The dynamics of mefenoxam insensitivity in a recombining population
of Phytophthora capsici characterized with AFLP markers.

ABSTRACT
Lamour, K. H., and Hausbeck, M. K. 2000. The dynamics of mefenoxam insensitivity in
a recombining population of Phytophthora capsici characterized with AFLP markers.
Phytopathology (in press).

Recent ﬁndings from Michigan suggest that recombination may play a role in the
survival and evolution of sensitivity to the fungicide mefenoxam in populations of
Phytophthora capsici on cucurbit hosts. In 1998, 63 mefenoxam insensitive isolates were
recovered from a squash ﬁeld in which mefenoxam had been applied. Additional isolates
were recovered from untreated squash ﬁelds planted at this location in 1999 (200 isolates)
and the spring of 2000 (34 isolates). Isolates from 1998 and 1999 were characterized
using ﬂuorescent ampliﬁed fragment length polymorphism (AFLP) markers and all
isolates were screened for compatibility type and mefenoxam sensitivity. In 1998 and
1999, 92% and 71% of the isolates, respectively, had unique multilocus AF LP genotypes
with no identical isolates recovered between years. Seventy-two identical AF LP markers
were clearly resolved in both the 1998 and 1999 sample sets and ﬁxation indices for the
37 polymorphic AF LP loci indicates little differentiation between years. There was no
decrease in the frequency of resistant isolates during the 2 years without mefenoxam
selection. We conclude that oospores play a key role in overwintering and that the
frequency of mefenoxam insensitivity may not decrease in an agriculturally signiﬁcant
time period (2 years) once mefenoxam selection pressure is removed.

INTRODUCTION

Crown, root, and fruit rot caused by Phytophthora capsici is increasing in

29

Michigan cucurbit production ﬁelds, and uninfested land suitable for rotation is becoming
increasingly scarce, especially in areas undergoing rapid urban development. The
phenylamide fungicide (PAF) mefenoxam is a systemic fungicide which appears to be
acting at the level of DNA transcription, and is fungistatic to fully sensitive isolates of P.
capsici (2, 13). While mefenoxam has been considered by some growers to be helpful,
mefenoxam insensitive isolates were reported on bell peppers in North Carolina and New
Jersey by Parra and Ristaino in 1998 (18) and have since been recovered from 10 of 11
farms sampled in Michigan (13), as well as, in Georgia (15) and southern Italy (19).
Mefenoxam insensitivity in Michigan P. capsici isolates is inherited as a single gene
exhibiting incomplete dominance (13) which is consistent with the reports for a variety of
other oomycetous organisms (2). Investigations with P. infestans indicate that
insensitivity may be conferred by genes at different chromosomal positions (5) suggesting
that the basis of insensitivity in different populations may not be identical. Sexual
recombination, in particular, has the potential to impact management strategies that
employ PAF’s because the fully insensitive (two copies of the insensitivity allele)
phenotype may be directly generated. Phytophthora capsici is heterothallic and the
sexual stage is initiated when isolates of opposite compatibility type, designated A1 and
A2, come into close association to form thick-walled oospores (4). The asexual stage
includes the production of caducous sporangia borne on long pedicels which may release
motile zoospores if free water is present. Asexual spores are thought to be responsible
for the polycyclic nature of disease development (20).

Phenylamide fungicide resistance in the genus Phytophthora and, in particular, the

Phytophthora infestans/potato pathosystem, is well documented (2, 4, 8). Until recently,

30

the population structure of P. infestans appeared to be largely clonal outside of P.
infestans putative center of origin (6). The recent detection of both P. infestans
compatibility types along with increased genotypic diversity in some potato growing
regions indicates that the sexual stage is likely active and may signiﬁcantly impact control
strategies which have proved useful in the past (3, 9). When PAF resistance in European
P. infestans populations increased signiﬁcantly in the early 1980's, the efﬁcacy of the
PAF metalaxyl was only regained after the product was not made available to growers for
a period of time (2). This strategy apparently allowed the resistant populations to decline
or become extinct and depends on ephemeral populations or, in the case of resident
populations, upon a signiﬁcant cost for resistance outside of selection pressure. A recent
study of sensitive versus PAF resistant P. nicotianae isolates from citrus suggests
negligible ﬁtness costs for PAF resistance and reports that two years without PAF use did
not reduce the proportion of resistant isolates in groves (22). Kadish and Cohen report
that PAF resistant P. infestans isolates in Israel were more aggressive in colonizing tuber
tissue than sensitive isolates (12).

Novel techniques have been developed recently that allow characterization of
DNA level polymorphism in organisms for which little is known about the genome. An
example is the ampliﬁed fragment length polymorphism (AF LP) technique introduced by
Vos et al. in 1995 (24). This technique relies on restriction enzyme ﬁ‘agmentation of
genomic DNA with the concomitant ligation of synthetic adaptors to the DNA fragment
ends. Stringent PCR ampliﬁcation using adaptor-complementary primers with additional
selective nucleotides allow for the ampliﬁcation of fragment subsets. DNA fragment

subsets are termed ﬁngerprints and may be resolved using a range of techniques (1).

31

AF LP markers have been used on a variety of organisms (14, 23) and the procedure has
been Shown to generate a large number of reproducible markers (1, 23). The limitation
that these markers are generally scored as dominant markers (eg, either present or absent)
for diploid organisms requires the use of relatively large sample sets (11, 26).

Our null hypotheses are that sexual recombination has a signiﬁcant impact on the
population structure of P. capsici in Michigan and that mefenoxam insensitivity may not
decrease in the time frame of a typical 2 year rotation outside of mefenoxam selection
pressure.

Materials and Methods
Field plot: Research was conducted on a commercial farm in southwest Michigan with a
history (> 11 years) of P. capsici on bell peppers and squash and intensive use of PAF.
The 4.05 hectare ﬁeld sampled had previously been cropped to soybeans and corn with no
known record of P. capsici susceptible crops (e.g.; tomatoes, peppers, or cucurbits) prior
to 1997. During 1997 and 1998, yellow squash and zucchini grown in this ﬁeld became
diseased with Phytophthora crown, root, and fruit rot and the grower applied mefenoxarrr
as part of a disease management strategy (Novartis, Greensboro, NC). In 1998, all
isolates recovered were either intermediately or fully insensitive to mefenoxam. Both A1
and A2 compatibility types were present, and oospores were detected in diseased fruit. In

1999 and 2000, yellow squash was established in an 1,124 m2 experimental plot in this
ﬁeld, and mefenoxam was not applied. Diseased plants and/or fruit were sampled on 20
August 1998 (63 isolates from entire ﬁeld), June through August 1999 (200 isolates from
experimental plot), and 13 July 2000 (34 isolates from experimental plot). All isolates

were recovered from single diseased plants or fruit.

32

Isolate collection and maintenance: Isolation from diseased plant material was made
onto BARP (Benomyl 25ppm, Ampicillin lOOppm, Rifampicin 30ppm, and
Pentachloronitrobenzene lOOppm) amended UCV8 (840 ml distilled water, 163 m1
unclariﬁed V8 juice, 3 g CaCO3, and 16 g Bacto agar) plates. Procedures for obtaining
single zoospore isolates were as previously described (13). Single zoospore cultures were
maintained on RA (Rifampicin 30ppm, Ampicillin lOOppm)-UCV8 plates and transferred
bi-monthly. Long term storage consisted of a single 7 mm plug of expanding mycelium
from each single zoospore culture being placed into a 1.5m] microfuge tube with one
sterilized hemp seed and 1 ml of sterile distilled water, incubated for 2-3 weeks at 23 to
25°C, and stored at 15°C long term.

Phenotypic characterization: Isolates were screened for compatibility type as
previously described (13). Mefenoxam sensitivity was characterized using the in vitro
screening technique described by Lamour and Hausbeck (LH technique) for P. capsici
isolates in Michigan (13). Isolates are scored as sensitive (S) if growth on UC-V8 agar
amended with 100 ppm mefenoxam is less than 30% compared to a control, as
intermediately sensitive (IS) if between 30 and 90%, and fully insensitive (I) if greater
than 90% compared to the unamended control. These mefenoxam sensitivity categories
are based on a tri-modal distribution of 523 ﬁeld isolates of P. capsici. Clear modal
distributions were only attained when screening was conducted with a single high rate of
mefenoxam (100 ppm) amended media (K. Lamour, unpublished data). These putative
mefenoxam sensitivity categories were tested by in vitro crosses (I x S, IS x IS, IS x S,
and S x S) and chi-square analysis conﬁrmed that the observed progeny numbers were not

signiﬁcantly different than expected for Mendelian inheritance of an incompletely

33

dominant trait (13).

The LH technique differs from a commonly used method described by Goodwin,
Sujkowski, and Fry (GSF technique) (8) for P. infestans which uses two levels of
amended media (5 ppm and 100 ppm) to differentiate the three mefenoxam sensitivity
phenotypes and which has been used to characterize P. capsici isolates (15, 18, 19).
Unfortunately, analysis of our in vitro crosses and ﬁeld isolates using the GSF technique
did not resolve a clear modal distribution (K. Lamour, unpublished data). Assignment of
Michigan P. capsici isolates to the sensitive (S) category is the same whether using the
LH or GSF technique. The only difference is that some P. capsici isolates from Michigan
rated as fully insensitive using the GSF technique are rated as intermediately sensitive
using the LH technique.

DNA extraction and AF LP ﬁngerprinting: A technique for avoiding bacterial
contamination prior to growing isolates for DNA extraction was implemented using a
modiﬁed Van Teigham cell (4). The uppermost portion of a 7 mm plug of mycelium was
placed onto the surface of RA-WA plates (Rifampicin 30ppm, Ampicillin lOOppm, 1000
ml distilled water, and 16 g Bacto agar) and an autoclaved cap from a 1.5ml microfuge
tube was placed over the plug which forced the isolate to grow through the amended
media. Isolates were incubated in the dark for 2-3 days before two 7 mm plugs were
transferred to approximately 15ml of RA-UCV8 broth in 100 x 15mm Petri dishes and
incubated in the dark for three days at 23 to 25°C. Mycelial mats were washed with
distilled water and dried brieﬂy under vacuum before being frozen to -20°C and
lyophilized.

Lyophilized mats were ground with a sterile mortar and pestle. Whole genomic

34

DNA from approximately 50mg of ground mycelium was extracted using a QIAGEN
Dneasy Plant Mini Kit (QIAGEN Inc., Valencia, CA) according to the manufacturers
directions. DNA was quantiﬁed using Nucleic Acid QuickSticks (CLONTECH, Palo
Alto, CA) according to the manufacturers directions and approximately 100 ng of DNA
was then subjected to a restriction / ligation reaction, pre-selcctive ampliﬁcation, and
selective ampliﬁcations using the PCR core mix, adaptor sequences, core primer
sequences and ﬂuorescent labeled primers available in the Perkin-Elmer Applied
Biosystems AF LPTM Microbial Fingerprinting Kit (The Perkin-Elmer Corp., Foster City,
CA henceforth referred to as PE/ABI) and performed exactly as described in the PE/ABI
AF LP Microbial Fingerprinting protocol part # 402977 Rev A (24). All PCR reactions
were performed using an MJ Research Minicycler (MJ Research Inc., Waltharn, MA) in
0.2 ml tubes according to the cycling parameters outlined in the Microbial Fingerprinting
protocol.

An initial optimization set of reactions was performed using pre-selective
products from P. capsici isolate OP97 which was isolated from a cucumber fruit in 1997
(13). Selective ampliﬁcations with the selective primers EcoRI-AA, AC, AG and AT
were performed in all 16 combinations with the MseI-CA, CC, CG and CT selective
primers. EcoRI selective primers available from PE/ABI are labeled at the 5' end with
either carboxyﬂuorescein (F AM), carboxytetrarnethyrhodamine (TAMRA), or carboxy-
4',5'-dichloro-2',7'-dimethoxyﬂuorescein (JOE) ﬂuorescent dyes. The ﬂuorescent dyes
are excited by laser radiation and visualized by their characteristic absorption-emission
frequencies. Only the fragments containing an EcoRI restriction site are resolved.

Products from three reactions labeled with different colored dyes and a carboxy-

35

X-rhodarnine (ROX) size standard were loaded into each lane on a denaturing
polyacrylarnide gel and the fragments resolved in an ABI Prism 377 DNA Sequencer.
Results were prepared for analysis in the form of electropherograms using GeneScan
Analysis software (PE/ABI). AF LP fragments were scored manually as present (1) or
absent (0) using Genotyper (PE/ABI). Only DNA bands which consistently exhibited
unambiguous presence/absence proﬁles were scored.

A single isolate, OP97, was subjected to the aforementioned protocol using three
primer pair combinations which were chosen as optimal on 3 separate occasions
approximately 3 months apart to test for reproducibility of AF LP proﬁles.

Clone detection and cluster analysis: AF LP fragments were considered polymorphic if
the most common allele was present in less than 95% of the isolates from a given sample
set and scored for presence (1) or absence (0) (10). AF LP fragments present in more than
95% of the isolates from a given sample set were considered monomorphic. Analysis of
the resulting binary data matrix was performed using NTSYSpc version 2.02k (21).
Unweighted pair group method with arithmetic averages (UPGMA) cluster analysis was
performed on the matrix of similarity coefﬁcients calculated from all possible pairwise
comparisons of individuals within and among the 1998 and 1999 populations and a tree
generated. Isolates showing complete homology at all loci were considered to be clones
and except for a single representative isolate were excluded from frequency calculations.
Allele frequency and ﬁxation indices: Allele frequencies for AF LP markers were
estimated utilizing the expected relationship between gene and genotype frequencies in a
randomly mating population (i.e. Hardy-Weinberg proportions). The frequency of the

recessive (absent) allele (q) was calculated from the observed number of recessive

36

homozygote individuals (X) in a sample of n individuals using the formula for dominant

markers described by Jorde (11):

 

where x = X/n is the observed proportion of individuals that do not display the dominant
(present) marker phenotype. In order to test whether the composite genetic proﬁles from
1998 and 1999 were consistent with a single randomly mating population, the ﬁxation
index was calculated for each AF LP loci from the variance in allele frequencies according
to the following formula: FST = ((p,-p2)2/4)/(Avg p x Avg q), where p is the allele
frequency for the present state with pl and p2 indicating the two sample populations and q
is the allele frequency for the absent state (10). Fixation indices for individual loci were
interpreted according to the qualitative guidelines suggested by Wright (25) where the
range 0 to 0.05 may be considered as indicating little genetic differentiation, the range
0.05 to 0.15 indicates moderate genetic differentiation, and greater than 0.25 indicates
great genetic differentiation (10).

Results
AFLP band characterization: Evaluation of the 16 EcoRI + 2/MseI + 2 selective
primer pair combinations indicated that EcoRI + AC/ MseI + CA gave the most clearly
resolved fragment proﬁle and was used to amplify genomic DNA from all isolates in both
the 1998 and 1999 sample sets. This primer combination resulted in 72 clearly resolved
fragments of which 37 (51%) fragments were polymorphic in both 1998 and 1999 (Table
2.1). All 72 fragments were present in both 1998 and 1999 and no novel fragments were

detected between years. The following 35 fragments (size in basepairs) were

37

monomorphic in both the 1998 and 1999 sample sets: 41, 43, 47, 49, 58, 66, 70, 82, 85,
114, 118,123, 133, 135, 140, 159, 174, 235, 247, 249, 272, 278, 295, 298, 300, 341, 351,
355, 367, 402, 474, 488, 502, 519, and 527. AFLP proﬁles for isolate OP97, generated
from separate DNA extractions on three separate occasions over a one year period,
resulted in identical banding patterns with the only difference being minor changes in the
intensity of the electropherogram signal. Occasionally individual reactions resulted in
poorly resolved ﬁngerprint proﬁles (eg, low intensity of signal) and were repeated until
signals were deemed optimal.

Phenotypic, genotypic and gene diversity: No isolates sensitive to mefenoxam were
recovered in 1998 or 2000 and single A1 sensitive and A2 sensitive isolates were
recovered in 1999 (Table 2.2). In 1998, 18% of the isolates were intermediately sensitive
and 82% were insensitive, in 1999, 2% were sensitive, 28% were intermediately
sensitive and 70% were insensitive, and in 2000, 15% of the isolates were intermediately

sensitive and 85% were insensitive to mefenoxam (Table 2.2).

38

Table 2.1. Fixation indices (EST) for 37 AFLP loci from unique Phytophthora capsici
isolates collected from a single Michigan cucurbit ﬁeld during 1998 (N = 57) and 1999

(N: 141).

 

 

Fragmenta 1998 f(aa)b 1999 f(aa) Fmc
45 0.02 0.06 0.018
54 0.29 0.29 0.000
64 0.82 0.55 0.048
104 0.11 0.06 0.007
106 0.1 l 0.04 0.025
110 0.41 0.36 0.002
130 0.41 0.30 0.009
146 0.47 0.24 0.038
149 0.12 0.27 0.029
154 0.39 0.31 0.004
156 0.53 0.83 0.054
172 0.56 0.33 0.034
189 0.16 0.56 0.121
192 0.16 0.37 0.044
193 0.35 0.20 0.022

211 0.47 0.15 0.088
241 0.48 0.32 0.018
256 0.04 0.01 0.022
258 0.43 0.49 0.002
261 0.55 0.54 0.000
270 0.57 0.41 0.015
282 0.35 0.40 0.002
285 0.51 0.73 0.030
314 0.51 0.34 0.019
320 0.41 0.51 0.006
333 0.16 0.20 0.002
346 0.36 0.33 0.001
361 0.33 0.49 0.017
383 0.21 0.15 0.005
418 0.40 0.34 0.002
431 0.34 0.32 0.001
438 0.67 0.45 0.028
454 0.65 0.49 0.015
492 0.29 0.40 0.009
504 0.51 0.47 0.001
511 0.38 0.28 0.007
548 0.78 0.78 0.000

 

‘ EcoRl-AC/Msel -CA selectively ampliﬁed fragment size in base-pairs.

" Observed frequency of the absent state where ‘a’ represents the absence of a fragment.

° FST calculated from estimated allele frequencies. According to Wright’s qualitative
guidelines values between 0-0.05 indicate little genetic differentiation and values between

39

Table 2.1 (cont’d).
0.05-0.15 indicate moderate genetic differentiation.

Table 2.2. Phenotypic diversity of Phytophthora capsici isolates recovered from the same
cucurbit ﬁeld in 1998, 1999, and 2000. '

 

 

 

Numberb Compatibility type and mefenoxam sensitivity“d
of
, Al/S Al/IS Al/I A2/S A2/IS A2/I
Year“ Isolates
1998 57 - 4 31 - 6 16
1999 141 1 (2) 17 (20) 57 (53) 1 (1) 23 (18) 42 (47)
2000 34 - 2 8 - 3 21

 

' Mefenoxam was applied in 1998 but not in 1999 or 2000.

b Sample sets from 1998 and 1999 consist of unique multilocus genotypes as determined
with AF LP ﬁngerprinting. The 2000 sample set was recovered at the beginning of the
growing season and was not ﬁngerprinted.

° MS = mefenoxam sensitivity where S = sensitive, IS = intermediately sensitive and I =
insensitive as determined by in vitro screening on 100 ppm mefenoxam amended agar.

° Numbers in parentheses indicate the expected number of isolates if mefenoxam
insensitivity is assumed to be controlled by a single incompletely dominant gene in
Hardy-Weinberg equilibrium unlinked to compatibility type.

Fifty seven of the 63 isolates recovered in 1998, and 141 of the 200 isolates recovered in
1999 were unique based on multilocus AF LP proﬁles. No identical multilocus genotypes
were recovered between 1998 and 1999. Five isolates (2 A2/I, 2 A2/IS, and 1 AM ) of P.
capsici collected in 1998 had one clonal representative. Fourteen isolates collected in
1999 had between 2 and 4 clones (Table 2.3). A single A1 compatibility type insensitive
isolate had 40 clones recovered over the course of the 1999 season and comprised 3% of
the early, 15% of the mid, and 43% of the late sampling intervals (Table 2.3). The 1999
sampling intervals (early, mid, and late) are based on the dates of sampling and are not

intended to reﬂect stages of plant growth or the epidemiology of P. capsici. Cluster

analysis of AFLP ﬁngerprint variation indicated no signiﬁcant clustering of isolates

40

between 1998 and 1999.

Table 2.3. Clone contribution of ﬁfteen Phytophthora capsici isolates to the total number
of isolates collected in 1999 (N = 200).
CT No. of clones in early, mid and late season sample intervalsc

 

 

 

No. of
Isolate and 6/22 - 7/16 7/20 - 8/3 8/5 - 8/18

clonesa

MS” N=6O N=80 N=60

JP571 2 AM 2 - -
JP583 2 AM 2 - -
JP944 3 A1/I 2 1 -
JP999 3 AU] 2 1 -
JP1007 2 All] 1 1 -
JP1042 2 AM 1 1 -
JP1096 2 AM - 1 1
JP1102 2 AM - 2 -
JP1215 3 A2/I 3 - -
JP1342 2 A2/IS - 2 -
JP1369 2 AU] 1 1 -
JP1384 4 A2/I 3 1 -
JP1512 2 AM 1 - 1
JP1555 3 AM - - 3
JP1632 40 All] 2 12 26

 

' Total number of isolates with identical multilocus AF LP proﬁles.

° CT = compatibility type and MS = mefenoxam sensitivity where S = sensitive, IS =
intermediately sensitive and I = insensitive as determined by in vitro screening on 100
ppm mefenoxam amended agar.

° Sample intervals based on sampling dates only.

41

The majority (98%) of the 37 polymorphic AF LP fragments showed little genetic
differentiation (FST < 0.05) between 1998 and 1999 according to Wrights qualitative
criterion (Table 2.1) (25).

Discussion

Phytophthora capsici causes Signiﬁcant damage to cucurbit hosts in Michigan
each year. In an effort to prevent or control epidemics, many growers have used either
metalaxyl or the newer, but similarly acting compound, mefenoxam as a part of their
disease management strategy. This study was initiated in an effort to address the
concerns of growers who have high levels of mefenoxam insensitivity.

Phenotypic data (mefenoxam sensitivity and compatibility type) from a 1998
survey suggested that insensitivity to mefenoxam was common and that some level of
recombination is occurring in the ﬁeld (13) but without the application of additional
polymorphic markers our ability to assess population structure was severely restricted.
AF LP analysis proved to be a powerful tool for resolving the population dynamics of P.
capsici. A single selective primer combination, EcorRI-AC/MseI-CA, generated 72
bands of which 37 were polymorphic in our 1998 and 1999 sample sets. AF LP
ﬁngerprinting, in conjunction with temporal sampling, provided a useful characterization
of P. capsici from one season to the next and allowed us to track asexual disease
development over the course of a single season.

Our data suggests that sexual recombination signiﬁcantly impacts the structure of
this P. capsici population. The ﬁnding that 198 of the 262 isolates recovered between
1998 and 1999 had unique multilocus AF LP genotypes is consistent with the high level of

genotypic diversity expected in an outcrossing population (7, 16, 17). Even though clonal

42

reproduction occurred in 1998 and 1999, no identical genotypes were recovered between
years, suggesting that oospores are important for overwintering. The ﬁnding that 35 of
the 37 polymorphic fragments exhibited very little differentiation (ie: change in allele
frequency) based on the estimated ﬁxation indices between 1998 and 1999 is consistent
with the expectations for a recombining population large enough to avoid dramatic
changes due to genetic driﬁ.

In 1999 and 2000, sensitive and intermediately sensitive isolates (42 of 175) did
not increase in a manner suggesting selection in favor of mefenoxam sensitivity outside
of mefenoxam selection pressure. The fact that 14 of the 15 isolates with clonal
reproduction in 1999 were fully insensitive may be another indication that mefenoxam
insensitivity does not have signiﬁcant costs outside of mefenoxam selection pressure. If
we assume that there is only a single mefenoxam insensitivity gene in this population
unlinked to compatibility type, designated 1, and that this population is effectively free
from the effects of migration and genetic drift, some interesting speculations can be
made. For instance, in 1999, if the mefenoxam sensitivity phenotypes are assumed to
represent genotypes (eg; a ﬁrlly insensitive isolate has two copies of the I allele) then the
frequency of I can be estimated and the observed number of unique isolates that fall into
each of the six mefenoxam sensitivity/compatibility type categories can be compared to
the expectations under Hardy-Weinberg equilibrium. In 1999, the estimated frequency
of I is 0.84 and chi-square analysis using the data in Table 2 indicates that the observed
numbers do not differ from those expected under Hardy-Weinberg equilibria at P = 0.50
(X 2 of 3.09, df = 4). Although this is not a particularly powerﬁrl test due to the large

number of assumptions (10), it does lend support to the hypothesis that this population

43

meets the criterion for panmixia.

Our results do not allow us to reject the null hypothesis that sexual recombination
signiﬁcantly impacts the structure of this population. It appears that sexual .
recombination plays a signiﬁcant role in maintaining genotypic and gene diversity while
concomitantly producing overwintering inocultun. Our data also suggests that sexual
recombination may serve as a potent force for integrating a beneﬁcial allele based on the
ﬁnding that there were a total of 133 unique multi-locus genotypes fully insensitive to
mefenoxam between 1998 and 1999. An interesting question that can only be answered
by following a fully sensitive population as it shiﬁs to insensitivity is how much genetic
diversity is lost, if any, during the PAF selection process? The question of how long
mefenoxam resistance will remain in a population of P. capsici when selection pressure is
removed can only be answered in a tentative way. It appears that in this population
insensitivity will not decrease within the time frame of a typical 2 year rotation and , once
resistance to mefenoxam is established, the future usefulness of this fungicide may be
extremely limited.

Comparison of the population structure reported at this single location is currently
being compared to other locations in Michigan and the United States and should provide
useful insight into the amount of genetic diversity in sensitive vs. insensitive populations
as well as the contribution of migration to P. capsici population structure.

ACKNOWLEDGMENTS

This work was funded by the Michigan Agricultural Experiment Station,

Michigan State University Extension, Michigan Department of Agriculture, Michigan

Farm Bureau (GREEEN cooperative), Pickle and Pepper Research Committee, Pickle

44

Packers International, Inc. and the Pickle Seed Research Fund, Pickle Packers
International. We thank A. M. Jarosz for comments on the manuscript and valuable
criticism during this project, E. A. Webster for supervision of lab procedures, and M.
Bour, C. Hunter, J. Jabara and P. Tumbalam for competent lab assistance.

LITERATURE CITED

1. Blears, M. J., De Grandis, S. A., Lee, H., and Trevors, J. T. 1998. Ampliﬁed
fragment length polymorphism (AF LP): a review of the procedure and its
applications. J. of Ind. Microbiol. and Biotech. 21: 99-114.

2. Davidse L. C., van den Berg-Velthuis, G. C. M., Mantel, B. C., and Jespers, A. G.
K. 1991. Phenylamides and Phytophthora. Pages 349-360 in Phytophthora. J. A.
Lucas, R. C. Shattock, D. S. Shaw, and L. R. Cooke, Eds., British Mycological
Society, Cambridge.

3. Drenth, A., Tas, I. C. Q., and Govers, F. 1994. DNA ﬁngerprinting uncovers a
new sexually reproducing population of Phytophthora infestans in the
Netherlands. Euro. J. Plant Pathol. 100: 97-107.

4. Erwin, D. C., and Ribeiro, O. K. 1996. Phytophthora Diseases Worldwide. The
American Phytopathological Society, St. Paul, MN.

5. F abritius, A., Shattock, R. C., and Judelson, H. S. 1997. Genetic analysis of
metalaxyl insensitivity loci in Phytophthora infestans using linked DNA markers.
Phytopathology 87: 1034-1040.

6. Fry , W. E., Goodwin, S. B., Dyer, A. T., Matsuzak, J. M., Drenth, A., Tooley, P.
W., and Sandlan, K. P. 1993. Historical and recent migrations of Phytophthora
infestans: chronolog, pathways, and implications. Plant Dis. 77: 653-661.

7. Goodwin, S. B. 1997. The population genetics of Phytophthora. Phytopathology
87: 462-473.

8. Goodwin, S. B., Sujkowski, L. S., and Fry, W. E. 1996. Widespread distribution
and probable origin of resistance to metalaxyl in clonal genotypes of
Phytophthora infestans in the United States and Western Canada. Phytopathology
86: 793-800.

9. Goodwin, S. B., Sujkowski, L. S., Dyer, A. T., Fry, B. A., and Fry, W. E. 1995.

Direct detection of gene ﬂow and probable sexual reproduction of Phytophthora
infestans in northern North America. Phytopathology 85: 473-479.

45

10.

11.

12.

13.

14.

15.

16.

17.

18.

19.

20.

21.

22.

Hartl, D. L., and Clark, A. G. 1997. Principles of Population Genetics, Sinauer
Associates, Inc., Sunderland, MA.

Jorde, P. E., Palm, 8., and Ryman, N. 1999. Estimating genetic driﬁ and effective
population size from temporal shifts in dominant gene markers. Mol. Ecology 8:
1171-1178.

Kadish, D., and Cohen, Y. 1992. Overseasoning of metalaxyl-sensitive and
metalaxyl-resistant isolates of Phytophthora inﬁzstans in potato tubers.
Phytopathology 82: 887-889.

Lamour, K. H., and Hausbeck, M. K. 2000. Mefenoxam insensitivity and the

sexual stage of Phytophthora capsici in Michigan cucurbit ﬁelds. Phytopathology
90: 396-400.

Lopes, M., Rainieri, S., Henschke, P. A., and Langridge, P. 1999. AFLP
ﬁngerprinting for analysis of yeast genetic variation. Intl. J. of Syst. Bacteriol. 49:
915-924.

Mathis, W. L., Williams-Woodward, J ., and Csinos, A. S. 1999. Insensitivity to P.
capsici to mefenoxam in Georgia. Phytopathology 89: S49.

McDonald, B. A., and McDermott, J. M. 1993. Population genetics of plant
pathogenic fungi. BioScience 43: 311-319.

Milgroom, M. G. 1996. Recombination and the multilocus structure of fungal
populations. Annu. Rev. Phytopathol. 34: 457-477.

Parra, G., and Ristaino, J. 1998. Insensitivity to Ridomil Gold (Mefenoxam)
found among ﬁeld isolates of Phytophthora capsici causing Phytophthora blight
on bell pepper in North Carolina and New Jersey. Plant Dis. 82: 711.

Pennisi, A. M., Agosteo, G. E., Cacciola, S. O., Pane, A., and Faedda, R. 1998.
Insensitivity to metalaxyl among isolates of Phytophthora capsici causing root
and crown rot of pepper in southern Italy. Plant Dis. 82: 1283.

Ristaino, J. B. 1991. Inﬂuence of rainfall, drip irrigation, and inoculum density on
the development of Phytophthora root and crown rot epidemics and yield in bell
pepper. Phytopathology 81: 922-929.

Rohlf, F. J. 1993. NTSYS-pc - Numerical Taxonomy and Multivariate Analysis
System, Version 2.02k. Applied Biostatistics Inc.

Timmer, L. W., Graham, J. H., and Zitko, S. E. 1998. Metalaxyl-resistant isolates
of Phytophthora nicotianae: occurrence, sensitivity, and competitive parasitic

46

23.

24.

25.

26.

ability on citrus. Plant Dis. 82: 254-261.

Van der Lee, T., De Witte, I., Drenth, A., Alfonso, C., and Govers, F. 1997. AF LP
linkage map of the oomycete Phytophthora infestans. Fungal Genetics and
Biology 21: 278-291. '

Vos, P. e. a. 1995. AF LP: a new technique for DNA ﬁngerprinting. Nucleic Acids
Res. 23: 4407-4414.

Wright, S. 1978. Evolution and the genetics of populations. University of Chicago
Press, Chicago, IL.

Zhivotovsky, L. A. 1999. Estimating population structure in diploids with
multilocus dominant DNA markers. Mo]. Ecology 8: 907-913.

47

Chapter 3: The spatiotemporal genetic structure of Phytophthora capsici in
Michigan and implications for disease management.

ABSTRACT
Lamour, K. H. and Hausbeck, M. K. 2001. The spatiotemporal genetic structure of

Phytophthora capsici in Michigan and implications for disease management.
PhytOpathology (submitted).

Phytophthora capsici isolates were recovered from pepper and cucurbit hosts at
seven locations in Michigan from 1998 to 2000. Isolates were characterized for
compatibility type (CT), mefenoxam sensitivity (MS), and AF LP marker proﬁles. In
total, 94 AFLP bands were resolved. Individual populations were highly variable.
Within populations, 39 to 49% of the AF LP bands were polymorphic and estimated
heterozygosities ranged from 0.16 to 0.19. Of the 646 isolates fmgerprinted, 70% (454)
had unique AF LP proﬁles. No clones were recovered between years or locations.
Pairwise F-statistics (¢ST) between populations ranged from 0.18 to 0.40. There was no
correlation between @ST and geographical distance. A tree based on UPGMA cluster
analysis indicates discrete clusters based on location with no clustering based on year of
sampling. Analysis of molecular variance (AMOVA) partitioned variability among
(40%) and within populations (60%). The overall estimated @ST was 0.34 (sd = 0.03).
Ale2 CT ratio’s were 2 1:1and MS frequencies were similar between years. These data
suggest that long distance dispersal is rare, that the sexual stage plays a signiﬁcant role in
survival and maintaining high levels of genetic diversity, and that control strategies aimed
at preventing the introduction of P. capsici need to be investigated.

Introduction

Phytophthora capsici Leonian causes signiﬁcant damage to a variety of plant

48

hosts worldwide and in the United States seriously impacts the production of cucurbits
and peppers (10, 17, 23). In Michigan, P. capsici’s life history is divided between an
active growth phase in the presence of susceptible host tissue and a state of dormancy
over the winter. Overwintering survival is thought to be accomplished by thick walled
oospores which are produced during sexual reproduction (10, l 1). Phytophthora capsici
is heterothallic and completion of the sexual stage requires both the A1 and A2
compatibility types. Sexual reproduction is mediated by extracellular hormonal signals
and there is the potential for both self and cross-fertilization (9). Oospores generally
require a dormancy period prior to germination. Gerrninating oospores produce
coenocytic mycelium which can directly infect or, under suitable conditions, differentiate
into caducous sporangia. Sporangia can be dislodged and cause infection directly, or, in
the presence of free water, release 20 to 40 motile zoospores. Polycyclic asexual spread
of P. capsici between and down rows has been clearly documented in the pepper/P.
capsici pathosystem (24) but, to our knowledge, there have been no reports suggesting
that P. capsici has a signiﬁcant long distance mode of dispersal.

Ristaino has recently summarized management strategies useful for disease
control (23). The primary strategy is to manage soil water dynamics by providing the best
possible drainage for the host plants rhizosphere and the ﬁeld in general. Growers are
advised to rotate ﬁelds to non-susceptible hosts and when appropriate to apply fungicides.
The phenylamide fungicide (PAF) mefenoxam has been shown to be fungistatic to
sensitive isolates of P. capsici (19), but, as has occurred with many oomycetes,
insensitivity has developed in ﬁeld populations (10, 20, 21). Research using P. capsici

isolates from Michigan indicates that insensitivity is controlled by an incompletely

49

dominant gene of major effect (10) which is consistent with the ﬁndings for a number of
other oomycetes (3).

In Michigan, fruit, stem and root rots caused by P. capsici on cucurbit hosts have
increased in recent years and growers employing available management strategies have
experienced signiﬁcant losses. Over the last 4 years an investigation of P. capsici
populations in Michigan vegetable production ﬁelds has been conducted (10, 11). The
initial phase of this study was based on the distribution and frequency of compatibility
type (CT) and mefenoxam sensitivity (MS) phenotypes within and between ﬁelds. In
1998, an approximate 1:1 ratio of A1 :A2 isolates was discovered in the majority of ﬁelds
sampled and oospores were detected in diseased cucurbit ﬁ'uit on 4 separate farms. All
six CT/MS phenotypes were recovered as oospore progeny ﬁom a single diseased
cucumber fruit as well as from a single diseased cucumber ﬁeld (10). These initial
ﬁndings suggested that the sexual stage is active in populations of P. capsici in Michigan
and, based on the MS ﬁndings, that sexual recombination may play an important role in
directly generating the fully insensitive MS phenotype.

The ability to assess population dynamics using only CT and MS is limited by the
fact that only 6 phenotypic combinations are able to be resolved and is further limited
because some populations appear to have only sensitive or insensitive isolates (10).
Ampliﬁed ﬁagment length polymorphism (AFLP) markers are increasingly being used as
a tool to investigate population genetic structure in a wide variety of living organisms
including plants (27, 29), animals (22), insects (4), and microorganisms (12). A
molecular map of the P. infestans genome was constructed based on AF LP and RF LP

markers and corroborates the ﬁnding of other researchers that AF LP markers span the

50

genome (28). No prior sequencing or cloning of fragments is needed to utilize this
marker system and it has been shown to be highly reproducible between labs (1). AFLP
markers are generally scored as present or absent (cg; dominant markers) and the
conﬁdence with which population level inferences can be made is greatly increased by
sample sets that are approximately twice the size used for co-dominant markers (8, 13,
32).

A single ﬁeld (SW1-A) from which only intermediately or fully insensitive MS
phenotypes were collected in 1998 was sampled in the absence of mefenoxam selection
pressure over 1999 and 2000 (11). Characterization of the 1998 and 1999 samples with
AF LP markers revealed that genotypic and genic diversity were high, that clonal
reproduction was signiﬁcant within a single season but that members of the same clone
were not detected between years, that AF LP marker frequencies did not change
signiﬁcantly between years, and that the frequency of mefenoxam insensitivity did not
appear to decrease in the absence of mefenoxam selection pressure (11). These results
suggested that this population was large enough to withstand dramatic effects of genetic
drift and that there may not be a signiﬁcant cost for mefenoxam insensitivity. At the time
this information was reported it was not known what effects, if any, migration of P.
capsici from outside sources might have had in determining the observed population
structure.

In this paper we report on the genetic structure of P. capsici populations
throughout Michigan. Our objectives were to determine if the level of genetic diversity
found in the population at SW1-A is representative of P. capsici populations at different

geographical locations and to determine if long distance dispersal is a signiﬁcant

51

component in determining the genetic structure of populations in Michigan. A related
objective was to investigate how long P. capsici can survive in the absence of known
susceptible host material. We also report on the frequency of self-fertilized ‘vs. hybrid
progeny in a sexual cross between isolates from different geographical locations and the
inheritance of AFLP markers in this cross. Portions of the information in this paper have
been reported previously (10, 11).

Materials and Methods
Isolate collection and maintenance: Pepper, cucumber, pumpkin, tomato and squash
plant material (root, crown and fruit) with typical signs and symptoms of infection by P.
capsici were collected from 6 farms throughout Michigan between 1998 and 2000.
Sampling was conducted using ﬁeld speciﬁc grids with grid quadrants varying from 40m2
to 12km2 depending on the size of the ﬁeld. Sample sets are labeled according to the
following criterion; location (SW = southwest, SC = south central, C = central, and NW
= northwest) followed by a farm designator (1, 2,...n) with a hyphen separating a ﬁeld
designator (A, B, ...n) and the year sampling was conducted (1998 = 98, 1999 = 99, and
2000 = 00). Diseased plant material was collected from quadrants in a haphazard fashion.
Isolation from diseased plant material was made onto BARP (Benomyl 25ppm,
Ampicillin lOOppm, Rifampicin 30ppm, and Pentachloronitrobenzene lOOppm) amended
UCV8 (840 ml distilled water, 163 ml unclariﬁed V8 juice, 3 g CaCO3, and 16 g Bacto
agar) plates. Procedures for obtaining single zoospore isolates were as previously
described (10). Single zoospore cultures were maintained on RA (Rifampicin 30ppm,
Ampicillin lOOppm)-UCV8 plates and transferred bi-monthly. Long term storage

consisted of a single 7 mm plug of expanding mycelium from each single zoospore

52

culture being placed into a 1.5m] microfuge tube with one sterilized hemp seed and 1 ml
of sterile distilled water, incubated for 2-3 weeks at 23 to 25°C, and stored at 15°C.
Compatibility type and mefenoxam sensitivity determination: Agar plugs from the
edge of an expanding single- zoospore colony were placed at the center of UCV8 plates
approximately 2 cm from ATCC (American Type Culture Collection, Rockville, MD)
isolate 15427 (A1 compatibility type) and ATCC 15399 (A2 compatibility type),
incubated at 23-25°C in the dark for 3-6 days, and compatibility type determined.
Thereafter, all compatibility type determinations were accomplished using the OP97 (A1)
and SP98 (A2) ﬁeld isolates.

Agar plugs from the edge of actively expanding single-zoospore colonies were
placed at the center of 100 x 15 cm UCV8 plates amended with 0 and 100 ppm
mefenoxam (Ridomil Gold EC, Novartis, Greensboro, NC; 48% AI, suspended in SDW;
added to UCV8 cooled to 49°C). Inoculated plates were incubated at 23-25°C for 3 days
and colony diameters measured. Percent growth of an isolate on amended media was
calculated by subtracting the inoculation plug diameter (7 mm) from the diameter of each
colony and dividing the average diameter of the amended plates by the average diameter
of the unamended control plates. All tests were conducted at least twice. An isolate was
scored as sensitive (S) if growth at lOOppm was less than 30% of the control,
intermediately sensitive (IS) if grth was between 30 and 90% of the control, and
insensitive (I) if growth was greater than 90% of the control (10).

DNA extraction and AFLP ﬁngerprinting: A technique for avoiding bacterial
contamination prior to growing isolates for DNA extraction was implemented using a

modiﬁed Van Teigham cell (5). The uppermost portion of a 7 mm plug of mycelium was

53

placed onto the surface of RA-WA plates (Rifampicin 30ppm, Ampicillin lOOppm, 1000
ml distilled water, and 16 g Bacto agar) and an autoclaved cap from a 1.5m] microfuge
tube was placed over the plug which forced the isolate to grow through the amended
media. Isolates were incubated in the dark for 2-3 days before two 7 mm plugs were
transferred to approximately 15ml of RA—UCV8 broth in 100 x 15mm Petri dishes and
incubated in the dark for three days at 23 to 25°C. Mycelial mats were washed with
distilled water and dried brieﬂy under vacuum before being frozen to -20°C and
lyophilized.

Lyophilized mats were ground with a sterile mortar and pestle. Whole genomic
DNA from approximately 50mg of ground mycelium was extracted using a QIAGEN
Dneasy Plant Mini Kit (QIAGEN Inc., Valencia, CA) according to the manufacturers
directions or using a CTAB procedure in conjunction with an automated DNA extractor.
DNA was quantiﬁed using Nucleic Acid QuickSticks (CLONTECH, Palo Alto, CA)
according to the manufacturers directions or on 1.5% agarose gels and approximately 100
ng of DNA was then subjected to a restriction / ligation reaction, pre-selective
ampliﬁcation, and selective ampliﬁcations using the PCR core mix, adaptor sequences,
core primer sequences and ﬂuorescent labeled primers available in the Perkin-Elmer
Applied Biosystems AF LPTM Microbial Fingerprinting Kit (The Perkin-Ehner Corp.,
Foster City, CA henceforth referred to as PE/ABI) and performed exactly as described in
the PE/ABI AFLP Microbial Fingerprinting protocol part # 402977 Rev A (30). All PCR
reactions were performed using an MJ Research Minicycler (MJ Research Inc., Waltham,
MA) in 0.2 ml tubes according to the cycling parameters outlined in the Microbial

Fingerprinting protocol.

54

An initial optimization set of reactions was performed using pre-selective
products from P. capsici isolate OP97 which was isolated from a cucumber ﬁ'uit in 1997
(10). Selective ampliﬁcations with the selective primers EcoRI-AA, AC, AG and AT
were performed in all 16 combinations with the MseI-CA, CC, CG and CT selective
primers. EcoRI selective primers available from PE/ABI are labeled at the 5' end with
either carboxyﬂuorescein (F AM), carboxytetramethyrhodamine (TAMRA), or carboxy-
4‘,5'-dichloro-2',7'-dimethoxyﬂuorescein (JOE) ﬂuorescent dyes. The ﬂuorescent dyes
are excited by laser radiation and visualized by their characteristic absorption-emission
frequencies. Only the fragments containing an EcoRI restriction site are resolved.

Selective ampliﬁcation AF LP products and a carboxy-X-rhodarnine (ROX) size
standard were loaded into each lane on a denaturing polyacrylamide gel and the ﬁagments
resolved in an ABI Prism 377 DNA Sequencer. Results were prepared for analysis in the
form of electropherograms using GeneScan Analysis software (PE/ABI). AF LP
fragments were scored manually as present (1) or absent (0) using Genotyper (PE/ABI).
Only DNA bands which consistently exhibited unambiguous presence/absence proﬁles
were scored.

A single isolate, OP97, was subjected to the aforementioned protocol using three
primer pair combinations which were chosen as optimal on 3 separate occasions
approximately 3 months apart to test for reproducibility of AF LP proﬁles.

Marker inheritance: Oospore progeny (N = 107) resulting from a cross between isolate
OP97 (Al/IS) x SFF 3 (A2/S) were subjected to AF LP analysis as described above.
Protocols for the generation, germination, and phenotypic characterization of the F1

oospores from this cross have been reported previously (10). The inheritance of AF LP

55

bands present in one parent and absent in the other which were inherited consistent with
one parent being heterozygous were analyzed using chi-square analysis to compare
observed numbers to those expected under simple Mendelian inheritance (28). Individual
oospore isolates were checked for the co-presence of AFLP markers present in single
copies in each parent to determine if they were the products of self-fertilization or
hybridization between the parent isolates. Bands present in both parents or homozygous
present in one parent and absent in the other are not reported on in this study.

Clone detection: AFLP fragments were scored for presence or absence and the binary
data matrix was converted to a similiarity matrix using a simple matching coefﬁcient of
resemblance with the program NTSYSpc version 2.02k (25). Unweighted pair group
method with arithmetic averages (U PGMA) cluster analysis was performed on the
similarity matrix and a tree generated. Isolates showing complete homology at all loci
were considered to be members of the same clone and except for a single representative
isolate were excluded from population genetic analysis (15).

Population genetic analysis: Sample sets collected from single ﬁelds during a single
year were considered a population. Populations were assumed to be in Hardy-Weinberg
equilibrium and each AFLP locus was assumed to be unambiguously di-allelic. The
program ‘Tools for population genetic analysis’ (TFPGA) (l 6) was used to (i) assess
genetic diversity within each population on the basis of estimated average heterozygosity
(18) and the proportion of polymorphic loci at the 95% level (6), (ii) calculate pair-wise
and overall F-statistics according to the methods of Weir and Cockerham (31) and (iii), to
test the effect of spatial separation on genetic structure by performing a Mantel test (14)

on the pairwise F-statistic matrix and the matrix of geographical distance between

56

populations. The signiﬁcance of the correlation between the two matrices was tested by
1000 random permutations to generate a null distribution of correlation coefﬁcients (Z
values) and a signiﬁcant result inferred if 295% of the randomly generated statistics were
greater than the observed value. Conﬁdence intervals for F-statistics at the 95%
conﬁdence level were generated by boot-strapping using 1000 iterations.

Using the program NTSYS-pc (25) the combined 0/1 data matrix for isolates from
all populations was used to construct a genetic similarity matrix of all possible pairwise
comparisons of individuals within and among populations using Jaccard’s similarity
coefﬁcient: GS(ij) = a/(a + b + c). Where GS(ij) is the measure of genetic similarity
between individuals i and j, a is the number of polymorphic bands shared by i and j, b is
the number of bands present in i and absent in j, and c is the ntunber of bands present in j
but absent in i. Trees were constructed using UPGMA cluster analysis to provide a
graphic representation of the relationships among isolates. A cophenetic correlation
coefﬁcient was computed to assess the goodness of ﬁt of the tree to the similarity matrix.

Genetic structure was also examined by analysis of molecular variance (AMOVA)
using the ARLEQUIN software package (26). The AMOVA analysis was used to
partition the variance in banding patterns within and among populations from the same
geographical site over consecutive years, between sites on the same farm separated by
approximately 1 km, and between all the locations sampled in Michigan. Signiﬁcance
values are assigned to variance components on the basis of a set of null distributions
generated by a permutation process which randomly assigns individuals to populations
and draws 1000 independent samples. AMOVA results were compared to the patterns

and degrees of similarity revealed by ﬁxation indices and cluster analysis.

57

Results
AFLP band characterization: Evaluation of 16 EcoRI + 2/Msel + 2 selective primer
pair combinations indicated that EcoRI + AC/ MseI + CA (EAC/MCA) gave the most
clearly resolved fragment proﬁle and was used for AF LP analysis. AFLP proﬁles for
isolate OP97, generated from separate DNA extractions on three separate occasions over
a one year period, resulted in identical banding patterns with the only difference being
minor changes in the intensity of the electropherogram signal. Occasionally individual
reactions resulted in poorly resolved ﬁngerprint proﬁles (eg, low intensity of signal) and
were repeated until signals were deemed optimal. The EAC/MCA primer combination
resulted in 94 clearly resolved fragments between 40 and 550 bps when considering the
combined data from all the isolates recovered from Michigan. AF LP analysis of oospore
progeny from cross OP97 x SF F3 revealed that all 107 progeny had the co-presence of
bands which were present in only one of the parents indicating that each is a product of
hybridization between the parent isolates. A comparison of the observed ratios to the 1 :1
expected under Mendelian inheritance for 17 bands which were present in only one parent
indicates that only one band segregated in a manner signiﬁcantly different than expected
at P = 0.05 (Table 3.1). Chi-square analysis also indicated that the observed ratios of
Ale2 compatibility types and S:IS mefenoxam sensitivities were not signiﬁcantly

different than expected under Mendelian inheritance (Table 3.1).

58

Table 3.1: Inheritance of 17 AF LP markers, compatibility type (CT), and mefenoxam

sensitivity (MS) in 107 progeny of a cross between Phytophthora capsici isolates OP97
(Al/IS) and SFF3 (A2/S).

 

Markera Progeny ratiob X 2° - Pd

 

E+AC/M+CA-66 47:60 1 .58 0.20
E+AC/M+CA-97 51 :56 0.23 0.70
E+AC/M+CA-146 53 :54 0.01 0.90
E+AC/M+CA-l49 60:47 1 .58 0.20
E+AC/M+CA-156 64:43 4.12 0.04
E+AC/M+CA-159 56:51 0.23 0.70
E+AC/M+CA-244 46:61 2.10 0.17
E+AC/M+CA-25 8 52:55 0.08 0.80
E+AC/M+CA-270 53:54 0.01 0.98
E+AC/M+CA-282 56:51 0.23 0.70
E+AC/M+CA-290 62:45 2.70 0.13
E+AC/M+CA-328 55 :52 0.08 0.80
E+AC/M+CA-351 61 :46 2.10 0.15
E+AC/M-l-CA-398 55:52 0.08 0.80
E+AC/M+CA-431 55 :52 0.08 0.80
E+AC/M+CA-435 57:50 0.46 0.90
E+AC/M+CA-444 49:58 0.76 0.85
CT 53:54 0.01 0.98

MS 47:60 1.58 0.20

 

3 AF LP marker labels indicate the restriction enzymes (E = EcoRI, M=Msel), the two
selective nucleotides, and the size of the DNA fragment in basepairs.

° Presence:absence ratio’s for AF LP markers, A1:A2 for CT, and sensitive (S):
intermediately sensitive (IS) for mefenoxam sensitivity as determined by screening on
100 ppm AI mefenoxam amended media.

Table 3.1 (cont’d).

° X 2 value for testing 1:1 segregation (l d.f.).

° Probability of the observed ratio occurring by chance under the null hypothesis of 1 :1
segregation.

59

Gene and genotypic diversity: All 94 AF LP bands were scored for presence or absence
in every isolate. The number of AFLP bands present in each population ranged from 68
to 80 with an average of 72, the number of polymorphic bands ranged from 39 to 49 with
an average of 43, and the estimated average heterozygosity ranged from 0.16 to 0.19 with
an average of 0.17 (Table 3.2).
Table 3.2: Population, number of isolates, total number of AFLP bands, number of
polymorphic bands, and estimated heterozygosity for populations of Phytophthora capsici
in Michigan.

No. and percent

No. of No. of AF LP polymorphic Estimated average

 

Population‘l
isolatesb bands bands Heterozygosity
(N = 94)

SW1-A98 57 72 37 (39) 0.16
SW1-A99 141 72 37 (39) 0.16
SW1-B99 35 69 33 (40) 0.16
SW1-BOO 24 69 38 (40) 0.16
SCI-A98 50 68 42 (45) 0.17
SC2-B99 45 71 43 (46) 0.17
Cl-AOO 48 77 41 (44) 0.17
NW1-A99 3 7 80 44 (47) 0.19
NW2-B98 24 73 46 (49) 0.18

 

a First two capital letters indicate location in Michigan with S = south, W = west, C =
central, and N = north, the number following the location designator indicates the farm,
the capital letter following the hyphen is a ﬁeld designator, and the numbers following the
ﬁeld designator indicate year (eg; 00 = 2000).

b Total number of isolates with unique multilocus AF LP proﬁles.

60

Seventeen (18%) AF LP loci were ﬁxed for the present state in all populations, 12
(13%) were polymorphic in all populations, and 65 (69%) were ﬁxed for presence or
absence in some populations and polymorphic in others. Of the 646 isolates ﬁngerprinted
70% (454) had unique multilocus AFLP ﬁngerprints (Table 3.3). The number of clones
detected from single locations in Michigan varied from 3 to 15 and the number of isolates
within any single clonal lineage ranged from 2 to 40 (Table 3.3). In all cases isolates with
identical multilocus AF LP proﬁles had identical compatibility types and fell into the
same mefenoxam sensitivity category.

Table 3.3: Clonal component of genotypic diversity within sample sets of Phytophthora
capsici from Michigan.

Populationa Total no. of Unique AFLP No. of clonal Minimumzmaximum

 

isolates genotypes (%) lineages no.of isolates per

clone
SW1-A98 63 57 (0.94) 5 2:2

SW1-A99 200 141 (0.71) 15 2:40
SW1-B99 71 34 (0.48) 12 2:9
SW1-B00 36 24 (0.67) 5 2:8
SCI-A98 57 50 (0.88) 5 2:3
SC2-B99 56 45 (0.80) 5 2:5
Cl-AOO 51 48 (0.94) 3 2:2
NW1-A99 88 37 (0.42) 12 2: 12
NW2-B98 24 18 (0.75) 3 2:3

Totals 646 454 (0.70) 65 -

 

3' First two capital letters indicate location in Michigan with S = south, W = west, C =
central, and N = north, the number following the location designator indicates the farm,
the capital letter following the hyphen is a ﬁeld designator, and the numbers following the
ﬁeld designator indicate year (eg; 00 = 2000).

61

Temporal dynamics: F-statistics (@ST) for populations of P. capsici sampled from ﬁeld
SW1-A over 1998 and 1999, and ﬁeld SW1-B sampled over 1999 and 2000 were 0.04

and 0.03 respectively (Table 3.4).

Table 3.4: F-Statisics (@ST) (below diagonal) and geographical distances (in km, above
diagonal) between Phytophthora capsici sample sets collected from single locations over
time and different locations in Michigan.

SW1- SW1- SW1- SW1- SCl- SC2- Cl- NW1- NW2-

 

Populations“

A98 A99 B99 B00 A98 B99 A00 A99 B98
SW1-A98 - 0 1 1 165 169 150 180 185
SW1-A99 0.04 - 0 1 165 169 150 180 185
SW1-B99 0.18 0.25 - 0 166 170 150 180 185
SW1-BOO 0.25 0.24 0.03 - 166 170 150 180 185
SCl-A98 0.36 0.37 0.29 0.29 - 8 135 260 265
SC2-B99 0.33 0.35 0.32 0.33 0.28 - 130 255 260
Cl-AOO 0.36 0.37 0.33 0.32 0.38 0.40 - 140 145

NW1-A99 0.32 0.34 0.30 0.30 0.32 0.32 0.38 - 5

NW2-B98 0.36 0.37 0.31 0.32 0.33 0.33 0.33 0.27 -

 

° First two capital letters indicate location in Michigan with S = south, W = west, C =
central, and N = north, the number following the location designator indicates the farm,
the capital letter following the hyphen is a ﬁeld designator, and the numbers following the
ﬁeld designator indicate year (eg; 00 = 2000).

At both locations the number and identity of AF LP bands resolved remained
identical over time with 72 total bands recovered from populations at SW1-A and 69
bands recovered from populations at SW1-B (Table 3.2). The number and identity of
bands polymorphic at the 95% level (37 for SW1-A and 38 for SW1-B) and the estimated
average heterozygosity (0.16 for both locations) also remained constant over time (Table

3.2). AMOVA analysis of SW1-A and SW1-B over time partitioned 5% of the total

variability between years for SW1-A, and < 1% of the total variability between years at

62

SW1-B (Table 3.5).

Table 3.5: Results of nested analysis of molecular variance (AMOVA) for Phytophthora
capsici isolates based on 94 AF LP markers. Variance is partitioned (A) betWeen 1998
and 1999 at SW1-A, (B) between 1999 and 2000 at SW1-B, (C) between combined
sample sets from SW1-A and SW1-B, and (D) within and between samples sets from
seven locations in Michigan.

 

Source of variation“: ° Degrees Sum of Variance Percentage P °
of squares component of variation
freedom

(A) SW1-A (98-99)
Among populations 1 39.658 0.396 5.05 <0.0001
Within populations 197 1461.559 7.457 94.95

(B) SW1-B (99-00)
Among populations 1 6.678 0.016 0.27 .0029
Within populations 57 312.399 6.248 99.73

(C) SW1-A vs SW1-B
Among populations 1 234.790 2.762 27.34 <0.0001
Within populations 255 1820.294 7.340 72.66

(C) All locations
Among populations 6 1169.295 4.814 39.67 <0.0001
Within populations 273 1984.345 7.322 60.33

 

° First two capital letters indicate location in Michigan with S = south, W = west, C =
central, and N = north, the number following the location designator indicates the farm,
the capital letter following the hyphen is a ﬁeld designator, and the numbers following the
ﬁeld designator indicate year (eg; 00 = 2000).
b AMOVA analysis for all locations includes sample sets from a single year for locations

SW1-A and SW1-B.

° P = the probability of obtaining a more extreme component estimate by chance alone

based on 1000 sampling realizations.

63

Signiﬁcant clonal reproduction was detected at both ﬁeld sites within a given year
with genotypic diversity reduced by 6% for SW1-A98, 29% for SW1-A99, 52% SW1-
B99, and 33% for SW1-BOO (Table 3.3). No clonemates were detected between years for
either location. Cluster analysis of all the isolates combined showed that samples from
SW1-A and SW1-B branched from location speciﬁc nodes and that there was no
clustering within either of the location speciﬁc clusters based on year (Figure 3.1). The
ratio of A1 :A2 compatibility types at each location was 35:22 for SW1-A98, 75:66 for
SW1-A99, 18:16 for SW1-B99, and 12:12 for SW1-B00 (Table 3.6). The percentage of
isolates falling into the six mefenoxam sensitivity/compatibility type categories remained
relatively similar between years at each location with a breakdown of 0 and 1% AI/S, 7
and 12% A1/IS, 54 and 40% AM, 0 and 1% A2/S,11 and 16% A2/IS, and 28 and 30%
A2/I for location SW1-A in 1998 and 1999 respectively (Table 3.6). The percentage of
isolates in each of the six categories for SW1-B was 41 and 29% Al/S, 12 and 21%
Al/IS, 0 and 0% AM, 32 and 21% A2/S, 12 and 21% A2/IS, and 3 and 8% A2/I between
1999 and 2000 respectively (Table 3.6).

Spatial structure: Pairwise F-statistics (QT) between populations separated by
geographical distances ranging between 1 and 265 km had values ranging from 0.18 to
0.40 (Table 3.4). A Mantel test comparing the geographical distance matrix to the ¢ST
matrix showed essentially no correlation between geographical distance and genetic
differentiation with 8 of the 999 perrnutated data sets having Z-scores ztlre original Z-
score (r = 0.45, P = 0.009). Although there was not a direct correlation between
geographical distance and genetic differentiation, UPGMA cluster analysis based on

J accard’s similarity coefficient (Figures 3.1 and 3.2) indicates that populations located

64

within the same vegetable production region were more similar to each other than to
populations located within different regions.
Table 3.6: Location, year, hosts, compatibility type, and mefenoxam sensitivity of

genetically unique Phytophthora capsici isolates collected in Michigan between 1998 and
2000.

 

No. of Compatibility type and mefenoxam sensitivity"

 

Populationa Hostsb
isolates“ A“ S AI/IS Al/I A2/ S A2/IS A2/I

 

SW1-A98 S, PK 57 - 4 31 - 6 l6
SW1-A99 S 141 1 17 57 1 23 42
SW1-B99 S 34 14 4 - 11 4 1
SW1-B00 S 24 7 5 - 5 5 2
SCI-A98 C 50 10 17 2 10 11 -
SC2-B99 C 45 - 6 22 - 2 15
Cl-AOO P 48 20 - - 28 - -
NW1-A99 S, C 37 25 - - 12 - -
NW2-B98 P 18 10 - - 7 l -
Totals 454 87 53 l 12 74 52 76

 

a First two capital letters indicate location in Michigan with S = south, W = west, C =

central, and N = north, the number following the location designator indicates the farm,

the capital letter following the hyphen is a ﬁeld designator, and the numbers following the

ﬁeld designator indicate year (eg; 00 = 2000).

b S = squash, C = cucumber, PK = pumpkin, and P = pepper.

° Total number of isolates with unique multilocus AF LP proﬁles.

° Mefenoxam sensitivity determined by in vitro screening on 100 ppm AI amended media

with S = < 30% grth of control (GC), IS = between 30 and 90% GC and I = >90% GC.
The SW1-A and SW1-B populations which were located approximately 1 km

apart on the same farm had the lowest ¢ST scores ranging from 0.18 to 0.25 when
comparing each SW1-A population to each SW1-B population. AMOVA analysis based
on comparing combined sample sets from SW1—A to combined sample sets from SW1-B

partitioned the total variation into 73% within and 27% between locations (Table 3.5).

65

 

 

 

 

 

 

 

 

.__1
.——C—_——L

_,__r—r——L—.—

_q——l
.———L t g

f

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

SW1-B99
SW1-BOO

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

SWI- _j——(_,___. 1““.
A98 {__4
___f——_'
r I I T T f T I T I I I I I I I I I I I I I I I I l I I I I I I
.73 .78 .83 .88 .93 .98

Genetic similarity

Fig 1: UPGMA cluster analysis of Phytophthora capsici isolates from location SW1-B
over 1999 and 2000 (N=58) and SW1-A in 1998 (N=57) based on the Jaccard similarity
coefﬁcient using 94 ampliﬁed fragment length polymorphism (AF LP) markers. Nodes
contain isolates exclusively from single locations. Location identiﬁers precede the
inclusive node and are indicated by region (S = south, N = north, W = west, and C =
central) and a farm identiﬁer (1,2,...n) prior to the hyphen with a ﬁeld indicator (A,
B,...n) and the year of sampling (eg; 00 = 2000) following.

66

C1-A00

 

SC2-B99
I -———______
——————————:-

 

 

 

 

 

 

 

 

SW1-A98 I;—
.63 .88 .33 .98

Fig 2: UPGMA cluster analysis of 255 Phytophthora capsici isolates from seven
locations in Michigan based on the Jaccard similarity coefﬁcient using 94 ampliﬁed
fragment length polymorphism (AF LP) markers. Nodes contain isolates exclusively
from single locations. Location identiﬁers precede the inclusive node and are indicated
by region (S = south, N = north, W = west, and C = central) and a farm identiﬁer
(1,2,...n) prior to the hyphen with a ﬁeld indicator (A, B,...n) and the year of sampling

r1 I T 1 I I 1 I I I I 7 I I I l l I I I I I I
.68 .73 .78 -
Genetic similarity

(eg; 00 = 2000) following.

67

Genetic structure: The overall $57 value when analyzing sample sets from all 7
locations combined was 0.34 (s.d.= 0.03). An AMOVA analysis of sample sets from all
locations corroborated this ﬁnding and attributed 39.67% of the genetic variation between
populations and 60.33% within populations (Table 5). Cluster analysis was also in
agreement with the overall ﬁxation index and revealed that populations from different
geographical locations branched from speciﬁc nodes (Figures 3.1 and 3.2) with
population Speciﬁc clusters being between 63 and 75% similar. Genetic similarities
between individuals within each of the clusters showed similar patterns with individuals
ranging between 75-95% for SW1-A(98 and 99), 77-94% for SW1-B(99 and 00), 75-94%
for SCI-A98, 69-92% for SC2-B99, 76-95% for Cl-AOO, 71-97% for NW1-A99, and 72-
93% similar for NW2-B98 (Figures 3.1 and 3.2). The cophenetic correlation coefﬁcient
for the overall tree (Figure 3.2) was 0.84 indicating that the tree provided a good ﬁt to the
data matrix.
Discussion

The formulae, or tools, of population genetics provide an indirect methodology for
characterizing the evolutionary forces which contribute to the structure of a population.
The two evolutionary forces we focused on in this investigation are migration and
reproduction. In the case of P. capsici it would appear, a priori, that long distance aerial
dispersal similar to that described for P. infestans is not common (5). If this is true, then
the observed increase in disease at new locations in Michigan and recurrence of P. capsici
at previously diseased sites may be due to spread through waterways, the movement of
infected plant material, or in the case of previously infested sites, to survival of P. capsici

as dormant inoculum in the soil. The advantage of using population genetic theory to

68

unravel the history of a population is that an understanding of the mechanism of dispersal
is unnecessary. In theory, the history should be evident in the patterns of genetic
variation at the individual and group level. One very important point is that there must be
some level of genetic variability from which to make inferences. If there was only
asexual reproduction, or if P. capsici was exclusively self-fertile, then it would be very
difficult to draw conclusions concerning the evolutionary forces acting upon a population.
A 1998 survey of phenotypic diversity suggested that sexual reproduction is active
and that recombination via meiosis may play an important role in integrating mefenoxam
insensitivity into a population (10), especially because the mating between, or self-
fertilization of, intermediately sensitive isolates would directly generate the fully
insensitive phenotype. This study was followed by an intensive investigation of a single
geographical location (SW1-A) over time (1 1). During 1999 and 2000, location SW1-A
was chosen as an experimental site for testing alternative management strategies for the
production of squash, none of which included the use of mefenoxam. Sampling was
conducted over the course of the 1999 and 2000 growing seasons to assess whether there
is a signiﬁcant cost incurred by mefenoxam insensitivity. In order to measure the
frequency of MS genotypes unambiguously it was important to separate clonal from non-

clonal lineages. AF LP markers proved to be useful in this respect due to the large

 

number of loci resolved for a single isolate and also provided an opportunity to estimate
the amount of genetic diversity present within and between years at this location. Results
of this study indicated that the amount of genic and genotypic diversity was high and did
not decrease signiﬁcantly between years. Clonal reproduction within a single season was

found to be signiﬁcant, but no clones were recovered between years. The overall

69

conclusion was that this population appeared to be large enough to withstand the chance
effects of genetic drift and the frequency of mefenoxam insensitivity did not decrease
over time (11). Almost 100% of the isolates recovered from this site were either
intermediately or fully insensitive to mefenoxam between 1998 and 2000, whereas nearby
sites (as close as 1 km) had many sensitive isolates. The phenotypic data alone suggests
that extensive migration, or introduction, from geographically separated sites did not
contribute signiﬁcantly to population structure at SW1-A.

The inheritance of MS in sexual crosses between isolates with differing levels of
mefenoxam sensitivity was reported previously. It was not possible, based solely on
phenotype, to determine how many of the progeny were the products of self-fertilization
versus true hybridization between the parents. AF LP analysis of the progeny ﬁ'om the
cross between OP97 x SFF 3 indicates that all 107 progeny investigated have the co-
presence of bands only present in one parent and thus are all products of hybridization.
Chi-square analysis of the inheritance of 17 bands used in the overall population analysis
indicates that all but one AF LP marker was inherited as a simple Mendelian character.
This is consistent with research based on mapping of the P. infestans genome (28). It
also suggests that hybrid individuals may be more viable than self-fertilized, or inbred,
individuals. Genetically determined self-incompatibility is common in plants (7) and may
play some role in our observed results.

Location SW1-B, which is approximately 1 km from SW1-A, had a serious P.
capsici epidemic on squash in 1994 and was planted to corn or soybeans until 1999. In
1999 and 2000, the grower planted squash in this ﬁeld and P. capsici was isolated from

infected plants during both years. The overall trends based on phenotype and AF LP

70

analysis of the SW1-B isolates were consistent with the ﬁndings from SW1-A. Clonal
lineages were common within a single season, but no clones were recovered between
years. Genic and genotypic diversity remained high in both years with 40% of the AF LP
markers polymorphic in a given year and between 48 and 67% of the isolates having
unique multilocus AF LP proﬁles in 1999 and 2000 respectively. When comparing
sample sets between years, both SW1-A and SW1-B had ﬁxation indices approaching
zero (0.04 and 0.03) indicating that genie diversity was maintained between growing
seasons. AMOVA analysis corroborated this ﬁnding and partitioned 5% of the total
genetic variability between years for SW1-A and <l% between years for SW1-B. A
major difference between the two sites was that 81% of the isolates with unique
genotypes recovered at SW1-B were fully sensitive to mefenoxam, whereas 1% of the
genetically unique isolates recovered at SW1-A were sensitive to mefenoxam. The
structuring revealed by screening isolates for mefenoxam sensitivity was supported by
the estimated pairwise ﬁxation indices based on AF LP markers. Results indicate that
between 75 and 82% of the total genetic diversity present at these two sites is found in
any one population and that 18 to 25% is found at one site and not the other. AMOVA
analysis corroborated this ﬁnding and partitioned 27% of the total genetic variability
between populations. Cluster analysis clearly shows that isolates from SW1-B cluster
together from a single node with no blurring of the tree due to isolates from SW1-A.
SW1-B is not irrigated from a water source common to SW1-A, nor are there streams or
drainage ditches which connect the two sites or run near SW1-B. To our knowledge there
were not any ﬁelds planted to susceptible hosts, other than SW1-A, within at least 10 km

of SW1-B during 1999 or 2000. Without a sample set from the 1994 epidemic at SW1-B

71

 

it is not possible to say conclusively, but there is a strong probability that the 1999
epidemic at SW1-B originated from dormant inoculum generated ﬁve years earlier.

The considerable genetic variation observed within and between sites SW1-A and
SW1-B turned out to be representative for the entire state. Although populations within a
given region appear to be slightly more similar to each other than to populations ﬁom
different regions, there was no correlation between @ST and geographical distance.
Populations 1 to 8 km distant had only slightly lower ﬁxation indices than populations up
to 265 km apart. The populations in total were surprisingly uniform in their levels of
genetic diversity with the total number of AF LP markers resolved in any one population,
the number of polymorphic fragments, and the estimated average heterozygosities all
being surprisingly similar. Cluster analysis provided a useful visual representation of the
trends within and between populations. Isolates clustered unambiguously from location
speciﬁc nodes with populations from the same region showing slightly more similarity.
The range of genetic similarities between individual isolates within any one population
speciﬁc cluster was similar across all populations. The levels of genetic similarity
between populations as visualized by UPGMA cluster analysis were supported by both
the overall ﬁxation index of 0.34 and the overall AMOVA analysis which partitioned
total genetic variability into a 40% between and 60% within population component.

The null hypothesis that migration signiﬁcantly contributes to the genetic structure
of populations of P. capsici in Michigan is not supported by this investigation. If
signiﬁcant migration occurred between sites the expectation is that there would be less
diversity between populations. Certainly, migration ﬁom some source population(s) had

to occur at some point in time for each of these locations and there is no way to rule out

72

 

the possibility that migration via the same, or some alternative route, does not still occur.
What can be concluded is that migration does not exert enough of an equalizing force
between geographically separated populations to counterbalance what are likely a
combination of founder effects, genetic drift, mutation, and selection pressures. An
aspect of P. capsici’s life history which may also contribute to the apparent genetic
stability of populations over time is the survival and non-synchronous germination of
oospores. An interesting example of how dormant inoculum (eg; seeds) may act as a
stabilizing force against the effects of genetic drift is provided by Epling et al. who
followed small populations of Linanthus parryae over a 20 year period. It was found that
L. parryae had an extensive bank of dormant seeds that could remain viable for at least
six years and it was concluded that this dense seed store counterbalanced the effects of
genetic drift by limiting the effects of chance reductions in population size (2).

The ﬁnding that migration does not appear to play a signiﬁcant role in shaping or
maintaining population structure puts the amount, and apparent stability, of genetic
variability within individual populations of P. capsici in Michigan into perspective. The
implications are that the founding individuals in each of these separate geographical
locations contained what appears to be a high level of genetic variability and that,
according to the results from SW1-A and SW1-B, this variability is maintained over time.
A possible explanation for the population structure of P. capsici in Michigan is that
oospores formed by hybridization between genotypically diverse parents may have a
signiﬁcant survival or viability advantage compared to either oospores produced by self-
fertilization or clonal inoculum. This hypothesis is supported by the fact that all the

oospores recovered and analyzed ﬁ'om the cross between OP97 and SFF3 were hybrid,

73

r
I
I
l
I.

 

and our inability to detect clonal overwintering in natural populations.

An evaluation of currently employed management strategies in light of these
ﬁndings suggests that rotation to non-susceptible hosts for up to ﬁve years may not
protect against epidemics due to P. capsici. These results also suggest that factors which
may contribute to the initial colonization (eg; source of irrigation water, the disposal of
infected plant material, infected transplants) need to be carefully considered and, if
possible, every possible action taken to avoid the introduction of P. capsici into un-

infested ﬁeld sites.

ACKNOWLEDGMENTS

This work was funded by the Michigan Agricultural Experiment Station,
Michigan State University Extension, Michigan Department of Agriculture, Michigan
Farm Bureau (GREEEN cooperative), Pickle and Pepper Research Committee, Pickle
Packers International, Inc. and the Pickle Seed Research Fund, Pickle Packers
International. We thank A. M. Jarosz for comments on the manuscript and valuable
criticism during this project, E. A. Webster for supervision of lab procedures, and M.
Bour, C. Hunter, J. Jabara, P. Tumbalam, and J. Woodworth for competent lab assistance.

LITERATURE CITED

1. Blears, M. J ., De Grandis, S. A., Lee, H., and Trevors, J. T. 1998. Ampliﬁed

fragment length polymorphism (AF LP): a review of the procedure and its

applications. J. of Ind. Microbiol. and Biotech. 21 :99-1 14.

2. Coyne, J. A., Barton, N. H., and Turelli, M. 1997. Perspective: a critique of
Sewall Wright's shifting balance theory of evolution. Evolution 51 :643-671.

3. Davidse, L. C., van den Berg-Velthuis, G. C. M., Mantel, B. C., and Jespers, A. B.

K. 1991. Phenylamides and Phytophthora. Pages 349-360 in: Phytophthora. J. A.
Lucas, R. C. Shattock, D. S. Shaw, and L. R. Cooke, Eds. British Mycological

74

10.

11.

12.

13.

14.

15.

16.

Society, Cambridge.

Downie, D. A. 2000. Patterns of genetic variation in native grape phylloxera on
two syrnpatric host species. Mol. Ecology 9:505-514.

Erwin, D. C., and Ribeiro, O. K. 1996. Phytophthora Diseases Worldwide. The
American Phytopathological Society, St. Paul, MN.

Hartl, D. L., and Clark, A. G. 1997. Principles of Population Genetics, 3 ed.
Sinauer Associates, Inc., Sunderland, MA.

Holsinger, K. E. 2000. Reproductive systems and evolution in vascular plants.
Proc. Nat. Acad. Sci. USA 97:7037-7042.

Jorde, P. E., Palm, S., and Ryman, N. 1999. Estimating genetic drift and effective
population size from temporal shifts in dominant gene markers. Mol. Ecology
8:1171-1178.

Ko, W. 1988. Hormonal heterothallism and homothallism in Phytophthora. Ann.
Rev. Phytopathol. 26:57-73.

Lamour, K. H., and Hausbeck, M. K. 2000. Mefenoxam insensitivity and the
sexual stage of Phytophthora capsici in Michigan cucurbit ﬁelds. Phytopathology
90:396-400.

Lamour, K. H., and Hausbeck, M. K. 2001. The dynamics of mefenoxam
insensitivity in a recombining population of Phyt0phthora capsici characterized
with AF LP markers. Phytopathology In Press.

Lopes, M., Rainieri, S., Henschke, P. A., and Langridge, P. 1999. AFLP
ﬁngerprinting for analysis of yeast genetic variation. International Journal of
Systematic Bacteriology 49:915-924.

Lynch, M., and Milligan, B. G. 1994. Analysis of population genetic structure
with RAPD markers. Molecular Ecology 3:91-99.

Mantel, N., and Valand, R. S. 1970. A technique of nonparametric multivariate
analysis. Biometrics 26:547-558.

Milgroom, M. G. 1996. Recombination and the multilocus structure of fungal
populations. Annu. Rev. Phytopathol. 34:457-477.

Miller, M. P. 1997a. Tools for population genetic analysis (TFPGA); A windows

program for the analysis of allozyme and molecular population genetic data.
Department of biological sciences, Northern Arizona University, Flagstaff.

75

17.

18.

19.

20.

21.

22.

23.

24.

25.

26.

27.

28.

29.

Miller, S. A., Bhat, R. G., and Schmitthenner, A. F. 1994. Detection of
Phytophthora capsici in pepper and cucurbit crops in Ohio with two commercial
immunoassay kits. Plant Dis. 78: 1042-1046.

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

Papavizas, G. C., and Bowers, J. H. 1981. Comparative fungitoxicity of captafol
and metalaxyl to Phytophthora capsici. Phytopathology 71:123-128.

Parra, G., and Ristaino, J. 1998. Insensitivity to Ridomil Gold (mefenoxam) found
among ﬁeld isolates of Phytophthora capsici causing Phytophthora blight on bell
pepper in North Carolina and New Jersey. Plant Dis. 82:711.

Pennisi, A. M., Agosteo, G. E., Cacciola, S. O., Pane, A., and Faedda, R. 1998.
Insensitivity to metalaxyl among isolates of Phytophthora capsici causing root
and crown rot of pepper in southern Italy. Plant Dis. 82: 1283.

Questiau, S., Eybert, M. C., and Taberlet, P. 1999. Ampliﬁed fragment length
polymorphism (AF LP) markers reveal extra-pair parentage in a bird species: the
bluethroat (Lucinia svecica). Mol. Ecology 8:1331-1339.

Ristaino, J. B., and Johnston, S. A. 1999. Ecologically based approaches to
management of Phytophthora blight on bell pepper. Plant Dis. 83: 1080-1089.

Ristaino, J. B., Larkin, R. P., and Campbell, C. L. 1993. Spatial and temporal
dynamics of Phyt0phthora epidemics in commercial bell pepper ﬁelds.
Phytopathology 83:1312-1320.

Rohlf, F. J. 1993. NTSYS-pc - Numerical Taxonomy and Multivariate Analysis
System, Version 2.02k. Applied Biostatistics Inc.

Schneider, S., Kueffer, J.-M., Roessli, D., and Excofﬁer, L. 1997. Arlequin: A
software for population genetic data analysis. University of Geneva, Geneva.

Travis, S. E., Maschinski, J ., and Keim, P. 1996. An analysis of genetic variation
in Astragalus cremnophylax var. cremnophylax, a critically endangered plant,
using AF LP markers. Mol. Ecology 5:735-745.

Van der Lee, T., De Witte, 1., Drenth, A., Alfonso, C., and Govers, F. 1997. AFLP
linkage map of the oomycete Phyophthora infestans. Fungal Genetics and Biology
21:278-291.

Van Der Merwe, M., Winﬁeld, M. 0., Arnold, G. M., and Parker, J. S. 2000.

Spatial and temporal aspects of the genetic structure of Juniperus communis
populations. Mol. Ecology 9:379-386.

76

_ m 1....“1

30.

31.

32.

Vos, P. e. a. 1995. AF LP: a new technique for DNA ﬁngerprinting. Nucleic Acids
Research 23:4407-4414.

Weir, B. S., and Cockerham, C. C. 1984. Estimating F -statistics for the analysis of
population structure. Evolution 3 8: 1358-1370.

Zhivotovsky, L. A. 1999. Estimating population structure in diploids with
multilocus dominant DNA markers. Mol. Ecology 8:907-913.

77