MSU

 

 

RETURNING MATERIALS:
Place in book drop to

 

 

 

LIBRARIES remove this checkout from
“ your record. FINES win
be charged if book is
returned after the date
stamped below.
_..:.r- 9..- ‘i_".,‘s t‘ *1 ’~ '\
in w % «.4. .. t. {I M

 

 

DEPLOYMENT, DETECTION AND ANALYSIS
OF HYPOVIRULENT STRAINS OF
ENDOTHIA PARASITICA IN MICHIGAN

 

BY

Sally Westveer Garrod

' A THESIS

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

MASTER OF SCIENCE

Department of Botany and Plant Pathology

1984

ABSTRACT

DEPLOYMENT, DETECTION AND ANALYSIS
OF HYPOVIRULENT STRAINS OF
ENDOTHIA PARASITICA IN MICHIGAN

 

 

BY

Sally Westveer Garrod

The infection and spread of Endothia parasitica was

 

examined in an American chestnut ( Castanea dentata )

 

grove using genetically marked strains of the fungus and a
unique dsRNA banding pattern. Spread of virulent and
hypovirulent strains of the fungus was detected within and
among trees. There also appears to have been conversion in
situ of virulent to hypovirulent strains. Infection
studies suggest that nail and cork borer wounds were more
likely to become infected than were others, however,
cankers at branch scar wounds were largest in area. Wounds
located 10 to 110 cm from inoculum sources were equally
infected, but cankers nearer to the base of the tree were
larger than those higher. Sexual mating types (A, and a)
were determined for a number of Michigan isolates. IdsRNA
was not detected in ascospores isolated from perithecia

formed by hypovirulent cultures.

To the memory of my grandmother,

Olive Colvin Sheltraw

ii

ACKNOWLEDGEMENTS

I wish to express my gratitude and appreciation to my
major professor, Dr. Dennis Fulbright, for his guidance,
encouragement and financial support throughout my graduate
career. I would also like to extend my thanks to Drs.
Karen Baker and John Hart for participating as members on
my graduate committee, and to Karen for her financial
support. In addition, I wish to thank Al Ravenscroft for
his expert technical assistance in the laboratory; Molly
and Jim Rogers for graciously permitting me to carry out my
research in their chestnut grove; the people in the
Department of Botany and Plant Pathology, especially those
in my laboratory, for their friendship; and my parents for
their constant encouragement. Finally, for his patience,
assistance, encouragement and support, I thank my husband

Mark.

iii

List of Tables . .

List of Figures .

General Introduction and

Introduction . . .

Materials and Methods

Results . . . . .

Discussion . . . .

Introduction . . .

Materials and Methods

Results . . . . .

Discussion . . . .

Appendix . . . . .

List of References

TABLE OF CONTENTS

Literature Review

Part I

iv

PAGE

 

LIST OF TABLES

 

 

TABLE PAGE
Part I
l. Endothia parasitica strains used . . . . . . . l4
2. Endothia parasitica isolates obtained from

 

designated wounds or cankers on study trees

at Crystal Lake, MI in August 1982, and

their sensitivity to PCNB and the presence

of dsRNA . . . . . . . . . . . . . . . . . . . 22

3. Endothia parasitica isolates obtained from
designated wounds or cankers on study trees
at Crystal Lake, MI in October 1982, and
their sensitivity to PCNB and the presence
of dsRNA . . . . . . . . . . . . . . . . . . . 23

 

4. Endothia parasitica isolates obtained from
designated wounds or cankers on study trees
at Crystal Lake, MI in December 1982, and
their sensitivity to PCNB and the presence
of dsRNA . . . . . . . . . . . . . . . . . . . 24

 

5. Endothia parasitica isolates obtained from
designated wounds or cankers on study trees
at Crystal Lake, MI in May 1983, and their
sensitivity to PCNB and the presence of
dsRNA . . . . . . . . . . . . . . . . . . . . 25

 

6. Endothia parasitica isolates obtained from
designated wounds or cankers on study trees
at Crystal Lake, MI in July 1983, and their
sensitivity to PCNB and the presence of
dSRNA . . . . . . . . . . . . . . . . . . . . 27

 

7. Endothia parasitica isolates obtained from
designated wounds or cankers on study trees
at Crystal Lake, MI in September 1983, and
their sensitivity to PCNB and the presence
of dsRNA . . . . . . . . . . . . . . . . . . . 29

 

TABLE PAGE

 

8. Additional PCNB/virulence isolate types for
Endothia parasitica cultures isolated from
wounds and cankers . . . . . . . . . . . . . . 32

 

 

9. Isolates obtained from cankers on the study
trees at Crystal Lake, MI containing changes
in the expected GH2 dsRNA banding pattern . . 39

10. Chronological history of dsRNA banding
patterns found in isolates containing changes
in the expected GH2 dsRNA banding pattern . . 4O

11. Total number of cankers initiated by Endothia
parasitica found on American chestnut trees in
this study grouped in relationship to: A) The
inoculum source placed on the tree; B) The
phenotype of the isolated culture being the
same as that of the inoculum source; C) The
wound type made in the trees; and D) The
distance the wound was located from the
inoculum source . . . . . . . . . . . . . . . 41

 

12. X2 values for interactions due to inoculum
source/wound distance, inoculum source/wound
type and wound distance/wound type . . . . . . 44

Part II

1. Histories of cankers on American chestnut
trees from which bark samples were collected
June 1982 O O O O O O O O O O O O O O O O O O 68

2. Mating types of Endothia parasitica strains
involved in successful sexual crosses . . . . 72

 

vi

FIGURE

LIST OF FIGURES

Part I

Photograph of a representative g; parasitica
inoculum source tied over established wound
sites on an American chestnut tree trunk . . .

 

Map showing the relative position of trees
used in the study . . . . . . . . . . . . . . .

Diagram representing the dsRNA banding

pattern observed in the CL1(GH2) PCNB-R

strain in comparison to the banding patterns
observed in the isolates with changes in the
expected banding pattern . . . . . . . . . . .

Patterns of dsRNA segements from six E.
parasitica field isolates, the three E;

 

 

parasitica inoculum sources, and reovirus

 

mixed with a VLP from Hm9 separated by
electrophoresis on a 5% polyacrylamide gel . .

Average canker area calculated by combining
the various E; parasitica isolates based on
their PCNB sensitivity and virulence status . .

 

Average canker area associated with the
various wound distances from inoculum sources .

Average canker area found associated with

the various wound types . . . . . . . . . . . .
Part II

Perithecia formed by crossing two

hypovirulent strains, GH2 and GHU4 . . . . . .

vii

PAGE

 

16

17

36

37

45

47

49

73

GENERAL INTRODUCTION AND LITERATURE REVIEW

The American chestnut ( Castanea dentata [Marsh.]

 

Borkh.), a member of the beech family (Fagaceae), was once
a dominant or codominant species throughout the deciduous
forests of eastern North America (7, 19, 34, 86). Its
natural range extended from southern Maine south to
southwest Georgia, west to southeastern Michigan, Ohio,
Indiana, Kentucky, Tenessee and Mississippi (19, 34, 35,
65).

The American chestnut was a large tree with a massive
trunk and a broad, rounded, dense crown. Its height and

diameter ranged from 18-30 m and 0.6-1.2 m, respectively

(65). Many useful products were obtained from this
versatile tree. Wood was attractive, strong and unusually
resistant to decay. Tannins extracted from the bark and

wood were the basis of a large industry. Food for wildlife
and man was furnished through its abundant production of
nuts.

The average American chestnut tree of modern times is
very different than that of the past. 9; dentata now
exists within its natural range primarily as sprouts
growing from the root systems of trees that have been

attacked by a fungal disease known as chestnut blight

2

(chestnut bark disease) (2, 35, 41, 52, 85, 86). The
fungus was presumably introduced into North America around
the turn of the twentieth century on Oriental chestnut
stock imported from Asia. The pathogen, first found in
North America in 1904 on trees in the New York Zoological
Park in New York City (68), spread quickly throughout the
chestnut forests of North America, reducing g; dentata to
a minor understory shrub within fifty years (52).

The fungus, Endothia parasitica (Murr.) And. and And.,

 

an ascomycete (Diaporthales, Diaporthaceae), is the
pathogen responsible for the chestnut blight disease. It
also attacks oaks, maples, hickory and occasionally other
trees, but not nearly as severely as it attacks the

American chestnut (19, 69, 77). E; parasitica achieves

 

infection in the chestnut by penetrating the bark through
wounds (74). Aggregated in mycelial fans (buff-yellow in
color), the fungus advances rapidly through the host,
killing bark tissues, the vascular cambium and outer
sapwood (7, 19, 34, 74) (Changes produced in the host
tissues following fungal invasion have been described in
detail by Keefer [60]). Elliptical cankers, sunken below
or swollen above the surrounding bark become apparent at
infection sites. The cankers are orange-brown in color due
to fungal stromata which erupt through the surface of the
infected bark. Mycelia within the cankers spread until the
branch or stem is completely girdled, causing wilting and
death of the distal portion of the tree beyond the

infection (2, 14, 19, 35). Suckers (water sprouts) usually

arise below the cankered areas, stimulated by the
interference of nutrient transfer, but these suckers
eventually become blighted by new infections (14, 34, 62).
Drooping clusters of dead leaves (termed "flags") on dead
branches and/or undersized burrs may be additional symptoms
associated with the disease (14).

Perithecia and pycnidia of E; parasitica are borne in

 

the orange reddish-brown stromata which are scattered
thickly over the canker. The stromata average 1.8 mm
diameter, and are about 1.3 mm high. When wet,
uninucleate, oblong or cylindrical, hyaline thin-walled
conidia, 1.3 x 3.6 pm in size, exude in slender curving
yellow cirrhi (spore horns) from irregularly shaped
pycnidia embedded in the stromatal tissues (13). The
conidia, also refered to as pycndidiospores, asexual spores
or summer spores are dispersed during the spring, summer
and autumn months by rain, wind, insects, birds, and small
mammals (19, 40, 49, 77, 82, 87). Conidia germinate by two
germ tubes within 12 to 36 hours in chestnut wounds, on PDA
and almost any other nutrient media (13) (Bazziger has
noted that biotin, in combination with an inorganic
nitrogen source, and thiamine are essential for conidial
germination [15]). Light is necessary for conidial
production (13). Longevity of the spores has been
determined to be at least one year (13). Neither freezing
nor dessication was found to affect spore viability (49).

Perithecia of E. parasitica form in the autumn

 

months, deeply embedded at the base of the stromata in

4

which the pycnidia are contained. Perithecia are globose,
350-400 pm in diameter, with long black necks that
terminate at the surface of the stroma in an ostiole. When
wet, some of the club-shaped asci within the perithecia
expand, migrate towards the perithecial ostiole and
forcibly eject oval, hyaline, two-celled thick-walled
ascospores year-round (13, 50, 51, 72, 73, 74). The
ascospores, which average approximately 8.6 x 4.5 pm are
dispersed primarily by wind (12, 13, 74). Each of the two
cells of the ascospore carries two to four nuclei (13).
Ascospores will germinate within six to twelve hours (at
approximately 24 C) by two germ tubes per cell in chestnut
wounds, in water, or on any ordinary media (13). Longevity
of the ascospores was found to be the same as for conidia
(49).

Perithecia may also be produced in controlled
laboratory crosses on autoclaved chestnut stems (3).
Through some of these crosses it was determined that E;

parasitica is homothallic, but will outcross preferentially

 

(3, 4, 71). And, that more than one male nucleus
(conidium) may fertilize a single protoperithecium (8).
Conventional methods such as sanitation, chemical
application and breeding for resistance (through crosses
involving the moderately and highly resistant Chinese and

Japanese chestnut species, Castanea mollissima B1. and

 

 

Castanea crenata Sieb. and Zucc. have so far been
ineffective in controlling the chestnut blight disease (35,

39, 59). The goal of resistance breeding programs has been

5

to produce a tree with the form of the American chestnut
and the resistance of the Oriental species (the Oriental
species have forms resembling apple trees). Progress has
been made, but the desired endproduct has not yet been
achieved (29, 53, 54, 64). Evidence of the existence of
naturally occurring blight resistant American chestnuts has
not been observed until recently (46, 47).

In Europe, Endothia parasitica was discovered in Italy

 

 

in 1938 on Castanea sativa , Mill., the European chestnut
(21). This tree, like the American chestnut is highly
susceptible to chestnut blight (2). It is most likely that
imported Oriental chestnut stocks were the source of these
infections as was the case in the United States (35).
Within 25 years of its discovery, the pathogen had spread
to all major chestnut growing areas in Italy (66).
Attempts to control the disease failed, as they had in the
United States (35).

In 1950, Biraghi (22, 23) observed abnormal cankers on
chestnut sprouts. The cankers were abnormal in that the
fungus was restricted to the outer layer of the bark. As
time passed, the number of abnormal cankers increased as
the incidence of normal cankers decreased. Biraghi felt
that the cankers were abnormal because the host had aquired
a resistance to the disease (24).

Chestnut blight was discovered in France in 1965 (35).
Grente acquired bark samples from abnormal cankers from
Italy to study the phenomenon decribed by Biraghi (42).

Grente and Sauret (43) found that many of the strains

6

isolated from the Italian abnormal cankers were different
that those isolated from normal cankers. Cultures obtained
from abnormal cankers were white (instead of the normal
orange), sporulated less, and were less virulent than
cultures obtained from normal cankers. In addition, these
abnormal cultures slowed or prevented canker development
when inoculated into chestnut bark together with normal
cultures. Over a period of time, trees inoculated with the
less virulent cultures recovered. Grente and Sauret (43)
coined the term "exclusive hypovirulence" for the observed
phenomenon because the abnormal cultures excluded the
normal cultures by converting them to abnormal cultures.
The mechanism by which these hypovirulent strains prevented
the attack of more virulent strains was found to involve
the transfer of cytoplasmic agents through hyphal
anastomosis (20, 44, 86). Because of the existence of
naturally occurring hypovirulent strains, chestnut blight
is no longer a problem in Italy (66, 84) and is being
controlled in France by a biological control program which
involves the spread of artificially introduced hypovirulent
strains (45).

The first native American hypovirulent strains were
isolated from an abnormal canker on an American chestnut
tree growing in Michigan (36). Additional native
hypovirulent strains have been isolated from abnormal
cankers on chestnuts located in other areas in Michigan as
well as in Tennesse, Pennsylvania, Maryland, New York and

Virginia (35, 58). These native American hypovirulent

7
strains behave as European strains, but differ in that the
American strains are orange and the European are white when
grown in culture (36, 38).

Hypovirulent strains (but not virulent strains) of

Endothia parasitica , both EurOpean and American, were

 

found to contain double-stranded ribonucleic acid (dsRNA)

(30, 67), the genetic material of most fungal viruses (25).
European and North American hypovirulent strains were found
to contain one of at least three distinct types of complex
dsRNA banding patterns (30, 31). Fulbright et a1. (38),
have reported several additional dsRNA banding patterns,
different than those reported by Dodds, associated with
hypovirulent strains isolated from several locations in
Michigan.

Conversion of virulent strains to hypovirulent is
accompanied by the transmission of dsRNA after hyphal
anastomosis (5, 9, 30). Reversion back to full virulence
by single conidial spore selection may be associated with
loss of dsRNA (30). These findings suggested that dsRNA

may be responsible for hypovirulence in E; parasitica

 

(30, 31). However, evidence that dsRNA is the cause of
hypovirulence is correlative because cell-free transmission
of dsRNA into virulent strains has not yet been
accomplished (85).

Dodds (32) isolated and purified an extract of
pleomorphic, club-shaped particles from a European
hypovirulent isolate that contained dsRNA. The appearance

of the particles resembled virus-like particles (VLPs)

8

isolated from diseased mushrooms (63). Dodds suggested
that these particles were either a new fungal VLP, or a
site of accumulation of the dsRNA. Similar VLPs that also
contained dsRNA were extracted from another European
hypovirulent isolate (28). Newhouse, et a1. (70), using
transmission electron microscopy, has observed spherical,
membrane-bounded VLPs in thin sections of hyphal tips of a
European hypovirulent strain, but not in virulent isolates.
He suggested that they may have been responsible for
hypovirulence in the hypovirulent isolate.

Hypovirulent strains obtained from Grente and
hypovirulent strains derived from American virulent strains
(through hyphal anastomosis) were used to attempt to
control virulent cankers on American chestnut trees in the
United States (10, 55, 86). Cankers were more likely to be
controlled by hypovirulent strains which were related. For
example, French hypovirulent strains achieved control of
cankers initiated by French virulent strains and American
virulent strains converted to hypovirulent by European
dsRNA controlled cankers initiated by American virulent
strains. Cankers initiated by American virulent strains
were not controlled by French virulent strains.-

Vegetative incompatibility in E; parasitica has been

 

described by Anagnostakis (1). Vegetatively incompatible
strains might explain the failure of certain cankers to be
controlled due to the lack of transfer of cytoplasmic
determinants by the failure of hyphal anastomosis to occur

between incompatible strains (1, 11). Anagnostakis has

9

determined that vegetative incompatibility is heterogenic,
and controlled by at least 7 loci (with two alleles at each
locus) (4). However, evidence has been presented that
indicates vegetative incompatibility is not necessarily a
barrier for hyphal anastomosis. Hypovirulent and virulent
strains which belong to different vegetative compatibility
groups can still fuse and transfer dsRNA (1). In addition,
the problem of vegetative incompatibility in treating
virulent cankers may be overcome by applying mixtures of
hypovirulent strains that belong to many different
vegetative compatibility groups (57).

Control of individual cankers on American chestnut
trees has been achieved in many cases (11, 17, 55, 56, 57,
62, 86), however, due to secondary blight infections that
developed later, trees continued to die. In Michigan
however, American chestnut trees in some groves are
surviving (26, 37, 38). This appears to be due to the
presence of naturally occurring hypovirulent strains (37,
38).

Effective natural spread of hypovirulent strains of

Endothia parasitica is essential if the biological control

 

of chestnut blight is to succeed. Elliston has suggested
that persistent sources of hypovirulent strains would
increase the opportunity for spread (35). The strategy in
the past has been to eliminate virulent cankers through
treatment with hypovirulent strains. This process not only
eliminates the virulent canker, but also the hypovirulent

strains (because hypovirulent cankers are often healed over

10

time), thus removing the source of hypovirulent inoculum
which is necessary for spread (35). Willey (88) has
reported the spread of hypovirulence among cankers on trees
that had been previously inoculated with hypovirulent
strains. This procedure was successful in establishing
hypovirulent strains on the same tree, but presented no
evidence of spread to cankers on untreated trees.

The objectives of this research were: 1) to examine

the spread of E. parasitica within a blighted American

 

chestnut grove in Michigan using a genetically marked
strain; 2) to study the infection of the American chestnut

by E. parasitica based upon three physical factors

 

 

(inoculum strain type, wound type and wound distance from
inoculum source); 3) to determine the mating types of a

number of Michigan E. parasitica isolates; and 4) to

 

search for dsRNA in ascospores isolated from perithecia

formed by hypovirulent cultures.

PART I

DEPLOYMENT AND DETECTION OF ENDOTHIA PARASITICA
STRAINS WITHIN A BLIGHTED AMERICAN CHESTNUT GROVE
IN NORTHERN MICHIGAN.

INFECTION OF THE AMERICAN CHESTNUT BASED UPON
THREE FACTORS: (1) INOCULUM STRAIN TYPE,

(2) WOUND TYPE AND (3) WOUND DISTANCE FROM
INOCULUM SOURCE.

INTRODUCTION

Chestnut blight, caused by the fungus Endothia

 

parasitica (Murr.) And. and And., was responsible for the

 

demise of the once prevalent American chestnut ( Castanea
dentata [Marsh.] Borkh.). Today this blight continues
killing young stump sprouts throughout New England and the
Appalachian forest (34). In Europe and in certain groves
in Michigan, chestnut trees survive and produce nuts in
spite of the presence of chestnut blight (37, 38). It is

hypothesized that hypovirulent strains of E; parasitica

 

are responsible for the survival of these trees.
Hypovirulent strains of the fungus are less virulent than
normal strains, contain double-stranded ribonucleic acid
(dsRNA) and can arrest individual cankers on trees which
were caused by virulent strains.

Control of this disease may be possible through the
use of hypovirulence. However, natural dissemination of
the hypovirulent strains within groves is imperative if
control is to be effectively achieved.

Willey (88) has reported the spread of hypovirulence
among cankers on trees that had been previously inoculated
with hypovirulent strains. This procedure was successful

in establishing hypovirulent strains on the same tree, but

11

12
presented no evidence of spread to cankers on untreated
trees.
The objectives of this research were to (a) examine

the spread of E. parasitica , both hypovirulent and

 

virulent, within a blighted American chestnut grove using
genetically marked strains of the fungus and a known dsRNA
banding pattern and (b) to study the infection of the

American chestnut by E. parasitica based upon three

 

physical factors: 1) inoculum strain type, 2) wound type

and 3) wound distance from inoculum source.

MATERIALS AND METHODS

An American chestnut grove consisting of approximately
3000 trees was used to carry out the experiment. The
stand, located at Crystal Lake near Frankfort, Michigan
(83), is presently heavily infected with virulent strains

of E; parasitica .

 

Fungal strains

 

Table 1 lists Endothia parasitica strains used in the

 

study.

Deployment of the strains

Six by one cm sections of autoclaved chestnut wood
(with bark) were placed in 100 x 15 mm sterile plastic
plates with approximately 20 ml potato dextrose agar (PDA;
Difco; Detroit, MI). Each plate was inoculated with one of

three strains of E; parasitica :

 

(1) CL1 (virulent)
(2) CL1 PCNB-R (virulent and pentachloronitrobenzene

[PCNB; Terra-Coat LT-2] resistant)

13

14

Table 1. Endothia parasitica strains used.

 

 

 

Strain Virulencel PCNB2 Description

CL1 V S Isolated from a normal,
virulent canker at Crystal
Lake, MI in 1980.

CL1 PCNB-R V R Isolated from CL1 strain
growing on PDA with
100 ug PCNB/ml.

CL1(GH2) PCNB-R H R CL1 PCNB-R converted with

the dsRNA from the GH2
hypovirulent strain.

 

1V = Virulent; H = Hypovirulent

2R = Resistant to pentachloronitrobenzene;
S = Sensitive to pentachloronitrobenzene
3

A hypovirulent strain isolated from Grand Haven, MI in
1980. This isolate contains a unique dsRNA banding
pattern which was transfered to CL1 PCNB-R by hyphal
anastomosis.

15

(3) CL1(GH2) PCNB-R (hypovirulent and PCNB resistant)

Control plates were uninoculated. Cultures were incubated
at approximately 24 C under fluorscent lights with a 16-hr
photoperiod.

Two to three weeks later, after pycnidia were densely
covering the wood and agar surfaces, two small holes were
drilled on both sides of the wood through the agar and
plastic. The cover and sides of the plates were removed
and cotton string was threaded through the holes so that
the plates could'be tied to the tree trunks (Fig. 1).

In June 1982, twenty trees (Fig. 2) were selected and
divided into five blocks based upon trunk diameter at one
meter above soil level (trees chosen ranged from 12.13 to

29.71 cm in diameter). Five wound types:

(1) cork borer hole (5 mm diameter)

(2) nail hole (3 mm diameter)

(3) scratch (approximately 2 cm [made with hammer
tinesl)

(4) vertical scalpel slice (1.5 cm)

(5) natural or artificial branch scar (Artificial
branch scars were made by two scalpel slices
[1.5 cm each] at right angles to each other with
the apex of the angle pointing towards the top

of the tree.)

were made to the depth of the sapwood, 3 cm apart in a

16

 

Figure 1. Photograph of a representative E; parasitica
inoculum source tied over established wound
sites on an American chestnut tree trunk.

 

17

Figure 2. Map showing the relative position of trees used
in the study.

18

A,
,4
I)
x
“u. -a w

I)

 

LEGEND

{2cm (3}
.CL1(GH2) PCNB-R
{@4CHECK

-—-* = 5 FT

FIGURE 2

 

r _B
.CLl PCNB-R 9

ON

19

randomized order at five distances:

(1) 10 cm
(2) 35 cm
(3) 60 cm
(4) 85 cm
(5) 110 cm

from two meters above soil level on each tree trunk. One
of the four inoculum sources (CL1, CL1 PCNB-R, CL1(GH2)
PCNB—R, and control) was placed on the trunks over the
wound series at two meters above soil level (Fig. 1). The
inoculum sources were sprayed briefly with tap water to
initiate conidial spread. Inoculum sources were replaced
by fresh ones in August 1982 and October 1982, and removed
in December 1982.

Pieces of bark tissue approximately 2 mm2 were
aseptically collected 2, 4, 6, 11, 13 and 15 months later
from the center wound of the middle row and from any
visible cankers that had developed on each tree. At the
time of the final three samplings, canker areas were
determined (from length and width measurements using the
formula for an ellipse) and the presence of pycnidia and
perithecia was noted (as either "+" fruiting bodies present
or "-" fruiting bodies absent). Bark samples were immersed

in 15% commercial Chlorox (NaClO, 5.25%) solution 3-5 min.

and placed on PDA. If E; parasitica was isolated it was

 

subcultured onto fresh PDA. Cultures were incubated at

 

20
approximately 24 C under fluorescent lights with a 16-hr

photoperiod.

Detection of strains

 

Each time E. parasitica was isolated it was

 

subcultured on 2 plates of three different media 1)
Endothia complete medium (ECM) (71), modified by the
omission of glucose, 2) ECM with 100 pg PCNB/ml (Low-PCNB)
and 3) ECM with 1000 pg PCNB/ml (High-PCNB). Cultures were
incubated for seven days at approximately 24 C under
fluorescent lights with a 16-hr photoperiod. Resistance to
PCNB carried by isolates obtained from the trees was
determined by comparing the percent area growth on L—PCNB
and H-PCNB to ECM without PCNB, to percent area growth of
standards (CL1, CL1 PCNB-R and CL1(GH2) PCNB-R on the same
media.

Each isolate was analyzed for dsRNA after the method
of Day et al. (30) with modifications by Dodds (31) and

Fulbright (38).

RESULTS

Sixteen, fourteen, eighteen, thirty-five, fifty-eight
and eighty-seven field isolates were collected from wounds
and cankers after two, four, six, eleven, thirteen and
fifteen months, respectively. PCNB resistance and dsRNA
content was determined for all isolates that had not become
contaminated in the laboratory (Tables 2 - 7).

PCNB-R virulent strains were recovered from trees in
which CL1 PCNB-R was the source of inoculum. PCNB
sensitive (PCNB-S) virulent strains were recovered from
trees with CL1 sources of inoculum, and PCNB-R hypovirulent
strains were isolated 'from trees with CL1(GH2) PCNB-R
sources of inocula.

Isolates differing in PCNB sensitivity and virulence
from the inoculum sources placed on trees were also
detected. Fifty cultures, isolated from trees which had
PCNB-R/virulent and PCNB-R/hypovirulent inoculum sources,
were PCNB-S and virulent (Table 8). Four PCNB-R/virulent
cultures were isolated from trees with PCNB-S/virulent and
PCNB-R/hypovirulent inocula (Table 8). Seventeen
PCNB-R/hypovirulent strains were found on trees with CL1
PCNB-R and CL1 as inoculum sources (Table 8). And five

PCNB-S/hypovirulent strains were isolated from cankers on

21

22

Table 2. Endothia parasitica isolates obtained from
designated wounds or cankers on study trees at
Crystal Lake, MI in August 1982, and their
sensitivity to PCNB and the presence of dsRNA.

 

 

 

 

Inoculum Source1 Culture Isolated PCNB Resistance2 dsRNA3
CL1 T-C4 - -
T-E - -
T-K 5 - -
CL1 PCNB-R T-A(R1-3) + -
T-G + -
T-G(R2-1) + -
T-G(R5-4) + -
T-N + -
T-Q + _.
T-S + -
CL1(GH2) PCNB-R T-D + +
T-F - +
T—M + -
T-T 6 - -
T—T(c#l) - _
None T-U - -
1

§;_ parasitica strains CL1, CL1 PCNB—R or CL1(GH2) PCNB-R
were grown on autoclaved chestnut segments in a petri dish
with PDA. These were tied to designated trees within the
study area.

 

2

+

3+

Resistance to PCNB; - = Sensitive to 100 ug PCNB/ml

Presence of dsRNA; - = dsRNA not detected

4If parenthesis do not follow the tree letter then the
isolated culture originated from the third wound site of
the third row for the indicated tree (not a canker).

5Information enclosed in parenthesis after the tree letter
indicates location of canker on the tree. The canker on

this tree is located at the third wound site in the first
row.

6T-T(c#1) was isolated from a natural canker (canker #1)

that had begun development prior to the beginning of the
experiment.

23

Table 3. Endothia parasitica isolates obtained from
designated wounds or cankers on study trees at
Crystal Lake, MI in October 1982, and their
sensitivity to PCNB and the presence of dsRNA.

 

 

 

 

Inoculum Sourcel Culture Isolated PCNB Resistance2 dsRNA3
CL1 T-C4 - c
T-E - -
T-R - -
CL1 PCNB-R T-A - -
T-G 5 + c
T-G(R2-l) + -
T-N + -
T-S + -
CL1(GH2) PCNB-R T-D - C
T-F - c
T-M + c
T-T 6 - c
T-T(c#1) - -
None T-H - c
1

E; parasitica strains CL1, CL1 PCNB-R or CL1(GH2) PCNB-R
were grown on autoclaved chestnut segments in a petri dish
with PDA. These were tied to designated trees within the
study area.

 

2+

3+ = Presence of dsRNA; - = dsRNA not detected; 0 = Culture
was contaminated before the presence of dsRNA could be
tested.

Resistance to PCNB; - = Sensitive to 100 ug PCNB/ml

4If parenthesis do not follow the tree letter then the
isolated culture originated from the third wound site of
the third row for the indicated tree (not a canker).
5Information enclosed in parenthesis after the tree letter
indicates location of canker on the tree. The canker on
this tree is located at the first wound site in the second
row.

6T-T(c#1) was isolated from a natural canker (canker #1)
that had begun development prior to the initiation of the
experiment.

24

Table 4. Endothia parasitica isolates obtained from
designated wounds or cankers on study trees at
Crystal Lake, MI in December 1982, and their
sensitivity to PCNB and the presence of dsRNA.

 

Inoculum Source1 Culture Isolated PCNB Resistance2 dsRNA

 

CL1-éR2-l)4

CL1 PCNB-R (R1- 3)
(R2- 2)

(R2- 3)

Flat-388868
GIUSLS’IIL'JUMO

+

T- G(R2- l)
T- G(R5- 4)
T-S(Rl-4)
T-S(R3-5)
CL1(GH2) PCNB-R T- D(RZ- 3)
T- D(R2- 5)

I+ +4-+ |+ ++-+-++ +I
l

(RI-2)
(RS-3) -

IOI+++IOI

til-33688
WJBHHHI

None

 

l E; parasitica strains CL1, CL1 PCNB-R or CL1(GH2) PCNB-R

were grown on autoclaved chestnut segments in a petri dish
with PDA. These were tied to designated trees within the
study area.

 

2

+
3+ Presence of dsRNA; - = dsRNA not detected; c = Culture
was contaminated before the presence of dsRNA could be
tested.

Resistance to PCNB; - = Sensitive to 100 pg PCNB/ml

4Information enclosed in parenthesis after the tree letter

indicates location of canker on the tree. The canker on
this tree is located at the first wound site in the second
row.

51f parenthesis do not follow the tree letter then the
isolated culture originated from the third wound site of
the third row for the indicated tree (not a canker).

25

Table 5. Endothia parasitica isolates obtained from
designated wounds or cankers on study trees at
Crystal Lake, MI in May 1983, and their
sensitivity to PCNB and the presence of dsRNA.

 

 

Inoculum Sourcel Culture Isolated PCNB Resistance2 dsRNA

 

CL1

:37
23%
w
M
I
H
II
II

T-R(R3-4)
CL1 PCNB-R T-A(Rl-3)
T-A(R2-2)
T-A(R2-3)
T-A(R3-4)
T-G
T-G(R2-2)
T-G(R4-4)
T-G(R5-4)
T-N
T-N(R1-2)
T-Q(R1-2)
T-S
T-S(R1-4)
T-S(R3-5)
CL1(GH2) PCNB-R T-D
T-D(Rl-1)
T-D(Rl-5)
T-D(R2-3)
T-D(R2-5)
T-T
T-T(R1-2&
T-T(c#1) - _
T-T(c#2) - -
None T-B - -

l+‘+4-+-++ +4-+-+I +I
I + + l+ I

I++I
|++|

 

l E; parasitica strains CL1, CL1 PCNB-R or CL1(GH2) PCNB-R

were grown on autoclaved chestnut segments in a petri dish
with PDA. These were tied to designated trees within the
study area.

 

26

Table 5. (continued)

 

2

+

3

4..

Resistance to PCNB; - = Sensitive to 100 pg PCNB/ml

Presence of dsRNA; - = dsRNA not detected

4If parenthesis do not follow the tree letter then the
isolated culture originated from the third wound site of
the third row for the indicated tree (not a canker).
5Information enclosed in parenthesis after the tree letter
indicates location of canker on the tree. The canker on
this tree is located at the first wound site in the second
row.

6T-T(c#1) and T-T(c#2) were isolated from natural cankers
(numbered 1 and 2) that had begun develOpment prior to the
initiation of the experiment.

27

Table 6. Endothia parasitica isolates obtained from
designated wounds or cankers on study trees at
Crystal Lake, MI in July 1983, and their
sensitivity to PCNB and the presence of dsRNA.

 

 

Inoculum Sourcel Culture Isolated PCNB Resistance2 dsRNA

 

c4 — —

T" 5

T-C(R2-1) - -
T-C(R4-1) - -
T-E - -
T-E(Rl-l) - -
T-E(R1-2) - -
T-E(R2-3) - -
T-E(R2-4)
T-E(R3-2)
T-E(R4-l)
T-E(R5-5)

CL1

I + + II
+ I

A
w
M

I
H
V

I

I

CL1 PCNB-R

Pal-96330988
CUZUWNWNX
7']

DJ
I
4:.

I
I

I + +-+I + + I

I-B
I
0
id
N
I
I—A
v
I+ +4-+-++-+-++ ++-+-++ +I

T-S(Rl-4)
T-S(R3-5)
CL1(GH2) PCNB-R T-D
T-D(Rl-1)
T-D(Rl-5)
T-D(R2-3)
T-D(R2-5)
T-D(R3-5)
T-F(R3-5)
T-T -
T-T(Rl-l) -
T-T(Rl-2) - -
T-T(Rl-5) - -
T-T(R2-3) - -

II++II+I
l+l

I+I++I

28

Table 6. (continued)

 

Inoculum Sourcel Culture Isolated PCNB Resistance2 dsRNA3

 

CL1(GH2) PCNB-R T-T(R3-5) - '
T-T(R4-5é - -
T-T(c#1) ‘ ’
T-T(c#2) ‘ ’
T—T(c#3) ' ‘
None T-B - ’
T-B(R2-1) - -
T-B(R2-4) - -
T-B(R3-5) - -
T-B(R4-1) - -
T—B(R4-3) - -
T-U - -
T-U(R2-4) - -

 

1

 

E; parasitica strains CL1, CL1 PCNB-R or CL1(GH2) PCNB-R
were grown on autoclaved chestnut segments in a petri dish
with PDA. These were tied to designated trees within the
study area.

2+

Resistance to PCNB; - = Sensitive to 100 pg PCNB/ml
3

+ Presence of dsRNA; - = dsRNA not detected

41f parenthesis do not follow the tree letter then the
isolated culture originated from the third wound site of
the third row for the indicated tree (not a canker).

5Information enclosed in parenthesis after the tree letter
indicates location of canker on the tree. The canker on
this tree is located at the first wound site in the second
row.

6T-T(c#1), T-T(c#2) and T-T(c#3) were isolated from natural
cankers (numbered 1, 2 and 3) that had begun development
prior to the initiation of the experiment.

29

quakyle 7. Endothia parasitica isolates obtained from
designated wounds or cankers on study trees at
Crystal Lake, MI in September 1983, and their
sensitivity to PCNB and the presence of dsRNA.

 

jlricxzulum Source1 Culture Isolated PCNB Resistance2 dsRNA

 

CL1 T-C4 - -

T-C(R1-5)5 - -
T-C(R2-l) - -
T-C(R4-5) + —
T-E - -
T-E(R1-1) - -
T-E(R1-2) - -
T-E(R1-3) - -
T-E(R2-3) - -
T-E(R2-4) - -
T-E(R3-2) -
T-E(R4-1) -
T-E(R4-5) -
T-E(R5-5) - -
T-J(R1-5) - -
T-K - -
T-K(Rl—2) - -
T-R - —
T-R(R2-4) - -
T-R(R2-5) _ _
T-R(R4-1)
T-R(R5-2)

CL1 PCNB-R T-A (121-2)
T-A(Rl-3)
T-A(R2-l)
T-A(R2-2)
T-A(R2-3)
T-A(R3-3)
T-A(R3-4)
T-G
T-G(R1-1)
T-G(Rl-3)
T-G(R2-1)
T-G(R2-2)
T-G(R3-5)
T-G(R4-2)
T-G(R4-4)
T-G(R5-1)
T-G(R5-4)
T-N(R1-4)
T-N(R3-5)
T-N(R4-2)
T-N(R5-1)
T-Q(R1-2)
T-S

I+I

I-++ +I

I+++++I

I+++++++++++++I++++++++I

30

Table 7. (continued)

 

Inoculum Sourcel Culture Isolated PCNB Resistance2 dsRNA3

 

CL1 PCNB-R T-S(Rl-4) - -
T-S(R2-5)
T-S(R3-5)
T-S(R4-4)

CL1(GH2) PCNB-R T-D
T-D(R1-l)
T-D(Rl-5)
T-D(R2-3)
T-D(R2-5)
T-D(R3-5)
T-D(R4-2)
T—D(R4-3)
T-D(R5—2)
T-F
T-F(R3-5)
T-M(R3-5)
T-T(R1-l)
T-T(R1-2) — -
T-T(R1-5) - -
T-T(R2-3) — —
T-T(R3-5) - -
T-T(R4—5) - -
T-T(R5-3g - -
T-T(c#1) - -
T-T(C#2) — _
T-T(c#3) - -

None T-B - -
T-B(R1-4) - -
T-B(R2—l) - —
T-B(R2-4) - -
T-B(R2-5) - -
T-B(R3-5) - -
T-B(R4—1) - -
T-B(R4-3) - -
T-H - —
T—H(R2—5) - -

l++l|
+ l

I++I
|++I

I +-++ +I
I +-++ +I

T

31

able 7. (continued)

 

l

2

3

E; parasitica strains CL1, CL1 PCNB-R or CL1(GH2) PCNB-R
were grown on autoclaved chestnut segments in a petri dish
with PDA. These were tied to designated trees within the
study area.

 

+ Resistance to PCNB; - = Sensitive to 100 pg PCNB/ml

4..

Presence of dsRNA; - = dsRNA not detected

41f parenthesis do not follow the tree letter then the

isolated culture originated from the third wound site of
the third row for the indicated tree (not a canker).

5Information enclosed in parenthesis after the tree letter

indicates location of canker on the tree. The canker on
this tree is located at the fifth wound site in the first
row.

6T-T(c#1), T-T(c#2) and T-T(c#3) were isolated from natural

cankers (numbered 1, 2 and 3) that had begun development
prior to the initiation of the experiment.

32
Table 8. Additional PCNB/virulence isolate types for
Endothia parasitica cultures isolated from
wounds and cankers.

 

 

Inoculum Source PCNB/Virulence Status1 Isolated Culture,

 

 

of E; parasitica (Sampling Date)
Isolated
1) CL1 PCNB-R S/V T-AZ, 3 (10/82)
T-A(R2-2) , (5/83)
T-S(Rl-4), (5/83)
T-S(R3—5), (5/83)
T-S(R1-4), (7/83)
T-S(R3-5), (7/83)
T-S, (9/83)
T-S(R1-4), (9/83)
T-S(R2-5), (9/83)
T-S(R3-5), (9/83)
2) CL1(GH2) PCNB-R S/V T-T, 4 (8/82)
T-T(C#1) I (8/82)
T-T(C#l). (10/82)
T-T, (12/82)
T-T(R5-3), (12/82)
T—D, (5/83)
T-D(R1-1), (5/83)
T-D(Rl-5), (5/83)
T-Tl (5/83)
T-T(R1-2), (5/83)
T-T(c#l), (5/83)
T-T(C#2). (5/83)
T-D(R1-1), (7/83)
T—D(R1-5), (7/83)
T-D(R3-5), (7/83)
T-T, (7/83)
T-T(R1-1), (7/83)
T-T(R1-2), (7/83)
T-T(Rl-5), (7/83)
T-T(R2-3), (7/83)
T-T(R3-5), (7/83)
T-T(R4-5), (7/83)
T-T(C#l). (7/83)
T-T(C#2), (7/83)
T-T(c#3). (7/83)
T-D(Rl-1), (9/83)
T-D(Rl-5), (9/83)
T-D(R3-5), (9/83)
T-D(R4-2), (9/83)
T-M(R3-5), (9/83)
T-T(R1-1), (9/83)
T-T(R1-2), (9/83)

33

Table 8. (continued)

 

Inoculum Source

 

PCNB/Virulence Status1

Isolated Culture,

 

of E; parasitica (Sampling Date)
Isolated
2) CL1(GH2) PCNB-R S/V T-T(Rl-5), (9/83)
T-T(R2-3), (9/83)
T-T(R3-5), (9/83)
T-T(R4-5), (9/83)
T-T(R5-3), (9/83)
T-T(C#l), (9/83)
T-T(C#2): (9/83)
T—T(c#3)p (9/83)
3) CL1 R/V T-R, (12/82)
T-E(R5-5), (7/83)
T-C(R4-5), (9/83)
4) CL1(GH2) PCNB-R R/V T-M, (8/82)
5) CL1 PCNB-R R/H T-G, (12/82)
T-G, (5/83)
T-G(R4-4), (5/83)
T-G(R5-4), (5/83)
T-G, (7/83)
T-G(R1-3), (7/83)
T-G(R3-5), (7/83)
T-G(R4-4), (7/83)
T-G(R5-4), (7/83)
T-G, (9/83)
T-G(R1-3), (9/83)
T-G(R3-5), (9/83)
T-G(R4-2), (9/83)
T-G(R4-4), (9/83)
T-G(R5-1), (9/83)
T-G(R5-4), (9/83)
6) CL1 R/H T-E(R4-1), (7/83)
7) CL1 S/H T-E(R4-1), (9/83)
8) CL1(GH2) PCNB-R S/H T-F, (8/82)
T-F(R3-5), (7/83)
9) CL1 PCNB-R S/H T-G(R1-1), (9/83

34

Table 8. (continued)

 

1R

V

Resistant to PCNB; S = Sensitive to PCNB;
Virulent; H = Hypovirulent

2If parenthesis do not follow the tree letter then the
isolated culture originated from the third wound site of
the third row for the indicated tree (not a canker).

3Information enclosed in parenthesis indicates location of
canker on tree. The canker on this tree is located at the
second wound site in the second row.

4T-T(c#1), T-T(c#2) and T-T(c#3) were isolated from natural
cankers (numbered 1, 2 and 3) that had begun development
prior to the initiation of the experiment.

35
three trees with CL1, CL1(GH2) PCNB-R and CL1 PCNB-R as
inoculum sources (Table 8). Cultures which became
contaminated before the presence of dsRNA could be tested
were not included in Table 8.

The dsRNA detected in E; parasitica cultures isolated

 

during the first three sampling dates had dsRNA banding
patterns in polyacrylamide gels that were identical to the
unique dsRNA banding pattern of the hypovirulent inoculum
source [CL1(GH2) PCNB-R] (Figures 3 and 4). Beginning in
May 1983, new, different banding patterns began to appear,
however, they contained certain similarities to the
original banding patterns (Figs. 3 and 4, Table 9). Two,
five and nine isolates were obtained with new dsRNA banding
patterns from the fourth, fifth and sixth sampling dates,
respectively (Table 10). Three distinct banding patterns
were found when observed after electrophoresis on 5%
polyacrylamide gels (Figures 3 and 4).

Differences between the number of cankers initiated on
trees with respect to inoculum source, and wound distance
from the inoculum source were significant at P=0.05 by the
X2 (81) test in May, but not in July or September 1983
(Table 11). Differences between the number of cankers
initiated on a tree whose isolated cultures were the same
in regards to PCNB sensitivity and virulence as the
inoculum source on the same tree were not significant until
September of 1983 (Table 11). Differences between the

number of cankers that developed at the various wound types

Figure 3.

36

 

 

 

 

 

 

 

 

 

 

_T _T _T _T
_M _M

_LM _LM
_B _B _B _B
A B c D

 

 

Diagram representing the dsRNA banding pattern
observed in the CL1(GH2) PCNB-R strain in
comparison to the banding patterns observed in
the isolates with changes in the expected
banding pattern. Horizontal lines labeled T, M,
LM or B represent dsRNA segments separated by
electrophoresis on 5% polyacrylamide gels

(12 hrs, 40 V).

A.

dsRNA banding pattern of CL1(GH2) PCNB-R,
contains three main segments: top (T),
middle (M) and bottom (B).

dsRNA banding pattern B, contains four main
segments: top, middle, lower middle (LM),
and bottom.

dsRNA banding pattern C, contains three main
segments: top, lower middle and bottom.

dsRNA banding pattern D, contains two main
segments: top and bottom.

Figure 4.

37

Patterns of dsRNA segments from

six E; parasitica field isolates (all collected
9-83), the three E; parasitica inoculum
sources, and reovirus mixed with a VLP

from Helminthosporium maydis (Hm9) separated by
electrophoresis on a 5% polyacrylamide gel

(12 hr, 40 V) and stained with ethydium bromide
(0.05 mg/ml H20).

 

 

 

 

W = wells

T = top band

M = middle band

LM = lower middle band

B = bottom band

Lane Sample
1..........................CL1
2..........................reovirus/Hm9
3..........................CL1(GH2) PCNB-R
4..........................T-E(R4-1)
5..........................T-D(R2-5)
6..........................T-G(Rl-1)
7..........................T-G(R4—2)
8..........................T-FéR3-5)
9..........................T-F

10..........................CL1 PCNB-R

 

1Information enclosed in parenthesis after the

tree letter indicates location of canker on the
tree. The canker on this tree is located at
the first wound site in the fourth row.

2If parenthesis do not follow the tree letter
then the isolated culture originated from the
third wound site of the third row for the
indicated tree (not a canker).

38

 

FIGURE 4

39

Table 9. Isolates obtained from cankers on the study trees
at Crystal Lake, MI containing changes in the
expected GH2 dsRNA banding pattern.

Pattern B: New band, lower than middle

Tree (Canker position)

 

T-G(R4-4)1 (5-83)
T-G(R1-3) (7-83)
T-G(R4-4) (7-83)
T-G(R5-4) (9-83)
T-F(R3-5) (9-83)

Pattern C:

Lower middle band only

Tree (Canker position)

 

 

T-D(R2-3) (5-83)
T-F(R3-5) (7-83)
T-D(R2-3) (7-83)
T-G(R4-2) (9-83)
T-G(R4-4) (9-83)
T-G(R1-3) (9-83)
T-D(R2-3) (9—83)
T-G(R5-1) (9-83)
Pattern D: Missing middle band
Tree (Canker position)

T-GéR3-5) (7-83)
T-F (9-83)
T-G(R3-5) (9-83)

 

lInformation enclosed in parenthesis after the tree letter

indicates location of canker on the tree. The canker on
this tree is located at the fourth wound site in the
fourth row.

2If parenthesis do not follow the tree letter then the
isolated culture originated from the third wound site of
the third row for the indicated tree (not a canker).

40

Table 10. Chronological History of dsRNA banding patterns
found in isolates containing changes in the
expected GH2 dsRNA banding pattern.

Type of dsRNA
banding pattern observed

 

 

Isolated Culture 8/82 10/82 12/82 5/83 7/83 9/83
T-DéR2-3)l - - A c c c
T-F ~ A ? A - D
T-F(R3-5) - - - - C B
T-G(Rl-3) - - - - B C
T-G(R3-5) - - - - D D
T-G(R4-2) - - - - - C
T-G(R4-4) - - - B B C
T-G(R5-l) - - - - - C
T-G(R5-4) V - V A A B
1

Information enclosed in parenthesis after the tree letter
indicates location of canker on the tree. The canker on
this tree is located at the third wound site in the second
row.

2If parenthesis do not follow the tree letter then the
isolated culture originated from the third wound site of
the third row for the indicated tree (not a canker).

A = CL1(GH2) PCNB banding pattern. Original GH2 banding
pattern.

B = dsRNA banding pattern B, new band, lower than middle.
= dsRNA banding pattern C, lower middle band only.

dsRNA banding pattern D, missing middle band.

< D r)
n

= Virulent, no dsRNA.

'0
ll

dsRNA banding pattern not determined because culture
became contaminated.

- = No culture was isolated.

41

{raible 11. Total number of cankers initiated by Endothia
parasitica found on American chestnut trees in
this study grouped in relationship to: A) The
inoculum source placed on the tree; B) The
phenotype of the isolated culture being the same
as that of the inoculum source; C) The wound
type made in the trees; and D) The distance the
wound was located from the inoculum source.

A- Total number of cankers grouped by inoculum strain

 

 

 

 

 

 

 

 

 

 

 

Date CL1 CL1 PCNB-R CL1(GH2) PCNB-R None 5 2
5/ 83 8 11 5 0 11.000*
7/ 83 12 13 12 6 2.860
9/ 83 18 25 17 12 4.778
E3- Isolated culture same as inoculum source
IDEitze CL1 CL1 PCNB—R CL1(GH2) PCNB-R E’Z
5/ 83 8 6 2 1.333
7/ 83 10 7 2 5.158
9/ 83 16 15 5 6.167*
C- Wound type
Daitze cork borer nail scratch slice branch scar E 2
5/ 83 5 18 0 0 1 48.916*
7/ 83 8 25 5 1 4 42.000*
9/ 83 17 32 10 6 7 32.027*
D- Wound distance from inoculum source

2
Baxte 10 cm 35 cm 60 cm 85 cm 110 cm x
5/83 9 8 3 2 2 9.750*
7/83 12 12 10 6 3 7.349
9/83 20 19 12 14 7 7.861

 

 

Critical X2 values for Inoculum Strain: P(O.10)=6.25,
P(O.05)=7.81, P(0.025)=9.35, P(0.01)=11.3, P(0.005)=12.8.

Critical X2 values for Isolated Culture Same as Inoculum
SOurce: P(O.10)=4.6l, P(0.05)=5.99, P(0.025)=7.38,
P(O.01)=9.21, P(0.005)=10.6.

42

Table 11. (continued)

 

Critical X2 values for Wound Type and Wound Distance:
P(O.10)=7.78, P(0.05)=9.49, P(0.025)=11.1, P(0.01)=13.3,
P(0.005)=14.9.

*Differences between strains, wound types or wound
distances are significant at P=0.05 by X test.

43
remained significant throughout 1983 (Table 11). No
significant interactions between all combinations of the
three factors (inoculum source, wound distance and wound
type) were detected at any date (Table 12).
The average area of cankers calculated for the various

2 for

strains isolated in September 1983 were 112.28 cm
PCNB-S virulent strains, 59.45 cm2 for PCNB-R virulent
strains and 43.10 cm2 and 70.00 cm2 for PCNB-R and PCNB-S
hypovirulent strains, respectively (Fig. 5). Average
canker area at the different wound distances were 88.30

cm2 , 74.30 cm2 , 102.77 cmz, 77.09 cm2, and 125.38 cm2 at
10, 35, 60, 85 and 110 cm distances, respectively (Fig. 6).
Average canker area associated with the different wound
types were 59.25 cm2 for cork borer, 95.15 cm2 for nail,
80.29 cm2 for scratch, 81.19 cm2 for slice and 146.57 cm2
for branch scars (Fig. 7).

In May 1983, 8.33% of the cankers observed supported
pycnidia, and by July 1983, 100% of the cankers supported
pycnidia. Trees with confirmed hypovirulent cankers
(positive dsRNA content as determined in the laboratory)
contained approximately 75% of the sporulation observed in

virulent cankers. In all cases, only pycnidia were

observed (no perithecia).

44

Table 12. X2 values for interactions due to inoculum
source/wound distance, inoculum source/wound
type and wound distance/wound type.

A. Inoculum Source/Wound Distance Interaction

Date K 2
5/83 1.691
7/83 10.291
9/83 6.707

B. Inoculum Source/Wound Type Interaction

Date K 2
5/83 2.691
7/83 16.288
9/83 12.954

C. Wound Distance/Wound Type Interaction

Date 5 2
5/83 5.406
7/83 18.620
9/83 13.772

 

Critical X2 values for Inoculum Source/Wound Type and
Innoculum Source/Wound Distance Interactions:
P(0.10)=18.5, P(0.05)=21.0, P(0.025)=23.3, P(0.01)=26.2,
P(0.005)=28.3.

Critical X2 values for Wound Type/Wound Distance
Interactions: P(O.10)=23.5, P(0.05)=26.3, P(0.025)=28.8,
P(0.01)=32.0, P(0.005)=43.3.

Figure 5.

45

Average canker area calculated by combining the
various E; parasitica isolates based on their
PCNB sensitivity and virulence status.

 

PCNB-S/V = PCNB sensitive and virulent,
PCNB-R/V = PCNB resistant and virulent,
PCNB-R/H = PCNB resistant and hypovirulent,

and PCNB-S/H = PCNB sensitive and hypovirulent.

46

Hmmfixd. D

mmmfih E 3me I

NUZMJDM H31}... H D H ._. H mzmm wron—

INmImZUn. Ikmlmzum DREImIUn— 3.....mlmzum

 

 

 

 

   

 

 

IIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIII

 

 

 

 

.m mmflmnm

   

m
EN

   

\
\ .......
m E.

5......“ ....... mm
............. mm

............. Gm."

 

............... EN.“
............... EV.“
............... mm."
............... mm."

GEN

(HEHHFIUS N3) HBHH HEHHH'J

47

Figure 6. Average canker area associated with the
various wound distances from inoculum sources.

48

la

 

IIIII , L

 

Hmmfixm D

33.3.. E $me I
hand muztpmHn nzzo:
2|: Immuu lam): 11mm}. 1....an

       

 

 

 

m
EN
...... Ev

. ...... Gm
........... Gm

, ....................................................... Gm H
,. ................................................................................................... GNfi

‘N

 

 

 

 

 

 

.................................................................................................... EV."
.................................................................................................... mm."
.................................................................................................... mm.“

BEN
m NMDEHL

(IIBHHHEIS H3) HBHH HBHHHD

49

Figure 7. Average canker area found associated with the
various wound types. C = cork borer, N = nail,
S = scratch, SL = slice, and B = branch scar.

50

Hmmfi\m _.I.H_

nmmfik E 9me I
mmyh nzzoz
I...m.... I..._m.i slum). 1:12.... .i..u....

 

  

 

 

 

m. mmflmHn.

 

 

(USHHI'II‘JS H3) HEHH HEHNHU

DISCUSSION

The use of PCNB resistant strains in combination with
the ability to identify characteristic dsRNA banding
patterns of the hypovirulent inoculum source proved to be
effective tools in monitoring the movement of _E;

parasitica . Spread of virulent and, more importantly,

 

hypovirulent strains within and between trees in the grove
was detected. In addition, there appears to have been
natural conversion __ig 3139 of virulent strains to
hypovirulence by the passage of dsRNA from hypovirulent
strains present on the same tree. This is the first
experimental documentation of the natural spread and
infection of a specific hypovirulent strain of Endothia

parasitica in a blighted American chestnut grove in North

 

America as documented by using specific dsRNA banding
patterns and genetically marked strains.

Inoculum for the PCNB-R/virulent cultures isolated
from trees with PCNB-R/virulent inoculum sources most
likely originated from the PCNB-R virulent inoculum sources
placed on the same trees. PCNB-S/virulent cultures
obtained from trees with CL1 inoculum sources probably
originated from either the CL1 inoculum sources placed on

the trees or from the natural E4. parasitica population.

 

51

52
The dsRNA banding pattern of the hypovirulent inoculum

source appears to be very unique in that the bottom band

(B) (Fig. 3) of the pattern has been found only in this
particular hypovirulent isolate (38). Therefore,
PCNB-R/hypovirulent cultures isolated from trees with

hypovirulent inoculum sources most likely originated from
the CL1(GH2) PCNB-R inoculum sources since these isolates
contained dsRNA banding patterns identical, or very
similar, to that of the hypovirulent inoculum source (Figs.
3 and 4).

Inoculum for PCNB-S/virulent cultures which were
isolated from trees with PCNB-R/virulent and
PCNB-R/hypovirulent sources of inocula (Table 8, isolate
types one and two) could have originated from trees with
CL1 inoculum sources, or from natural PCNB sensitive
inoculum present in the grove.

The isolation of PCNB-R/virulent isolates from trees
with PCNB-S/virulent and PCNB-R/hypovirulent inocula (Table
8, isolate types three and four) indicated that there
appeared to be spread of PCNB-R/virulent inocula to these
trees. Because prevailing winds are from the west, E;

parasitica isolated from tree R may have originated from

 

tree Q (Fig. 2). Apparently tree G, D or F was the tree
from which inoculum for the canker on tree E originated.

If it originated from tree D or F, the inoculum would have
had to be a PCNB-R/hypovirulent strain which had lost its
dsRNA or from conidia without dsRNA. It may be likely that

this isolate was derived from a PCNB-R/hypovirulent strain

53

since a PCNB-R/hypovirulent isolate was collected from tree
E on the same date. Tree A was presumably the source of
inoculum for the isolate recovered from tree C. The
PCNB-R/virulent isolate on tree M may have originated from
the PCNB-R/hypovirulent inoculum source on tree M or from
tree N. If from tree M, the inoculum would have had to be
a PCNB-R/hypovirulent strain which had lost its dsRNA or
from conidia without dsRNA.

PCNB-R/hypovirulent strains found on trees with CL1
PCNB-R and CL1 as inoculum sources (Table 8, isolate types
five and six) probably originated from CL1(GH2) PCNB-R
inoculum sources on nearby trees. Presumably tree T was
the source of PCNB-R/hypovirulent inocula for the isolate
obtained from tree G in December of 1982 (Fig. 2). Once
established in tree G, it appears that hypovirulent inocula
spread to other wounds, up and down the tree (probably by
rain and insects), and was subsequently isolated upon
several occasions. The canker found on tree E most likely
originated from inoculum sources on trees F or D.

Because PCNB-S/hypovirulent E. parasitica isolates

 

were not introduced into the grove and because there is
essentially no reversion to wild type by conidia from
PCNB-R strains (D.W. Fulbright, personal communication) it
appears that virulent PCNB-S mycelia present in wounds may
have been converted to hypovirulence by hypovirulent
strains also on the trees. Mycelium in canker E(R4-1)
isolated in July 1983 (Table 8) presumably provided dsRNA

for the conversion of natural PCNB-S mycelia which resulted

54

in the PCNB-S hypovirulent culture isolated in September of
1983 (canker E[R4-1], Table 8). The CL1(GH2) PCNB-R
inoculum source located on tree F was most likely the
source of dsRNA involved in the conversion of
PCNB-S/virulent mycelia. Finally dsRNA found in a PCNB-S
strain on tree G probably originated from other
hypovirulent cankers on the same tree.

It is difficult to understand why PCNB-R virulent and
hypovirulent inocula did not spread to trees without
inoculum sources. It was suggested from infection studies
that wounds on trees with inoculum sources are infected
more quickly than are wounds on trees without inoculum
sources (Table 11). The absence of inoculum sources on
control trees may be one possible explanation of the
failure to detect spread of PCNB-R virulent and
hypovirulent inocula to these trees. Possibly the failure
to observe movement of these particular strains was
influenced by some unknown factor in the methods, or, it
could have been due purely to chance. With further
samplings from these trees, over time, dsRNA-containing
strains may be detected on trees without inoculum sources.
Other explanations for the lack of spread of PCNB-R
virulent and hypovirulent strains to these trees may become
apparent with further observations. However, these
findings do not appear to detract from the detection of
Spread of hypovirulent strains to trees with virulent
Sources, because, hypovirulent isolates obtained from trees

Witfli ‘virulent sources contained dsRNA banding patterns

55
identical, or very similar, to that of the hypovirulent
inoculum source (Figs. 3 and 4).

Transmission of hypovirulence from strain to strain
has been shown to occur by at least two methods: 1) to
progeny through the production of conidia and 2) by
hyphal anastomosis of compatible strains (85). Conidia
have been shown to be disseminated by vectors including
wind, rain, insects, birds, and small mammals (6, 19, 40,
49, 77, 82, 87). There is no reason to suspect that the
dissemination of hypovirulent strains may not occur in the
same manner. To substantiate this possibility, Fulbright
(personal communication) reported the isolation of dsRNA

infected conidia of E; parasitica from rain water dripping

 

down an American chestnut tree trunk below a natural
canker.

The observation that hypovirulent strains produce
fewer pycnidia has been suggested as a possible barrier in
achieving the natural dissemination of these strains (33,
85). For this reason, the CL1(GH2) PCNB-R strain was
chosen for the experiment. This hypovirulent strain was
selected because it sporulates well and is more virulent
than other hypovirulent strains (it has the capicity to
kill small suckers and seedlings when directly inoculated
into them, but has not been shown to kill larger trees).
Another desirable characteristic of this strain is that its
dsRNA banding pattern is unique when compared to other
dsRNA banding patterns of hypovirulent strains found, thus

making the dsRNA easily recognizable when isolated on

56

polyacrylamide gels. Even though hypovirulent cankers in
this study sporulated approximately 75% as much as virulent
cankers, results obtained do not indicate that the
production of fewer pycnidia by the hypovirulent inoculum
source was an effective barrier in the prevention of the
spread of the hypovirulent inocula. This statement is
specifically supported by the spread of hypovirulence
observed within tree G. It was dramaticlly demonstrated in
this tree that once established, hypovirulent cankers can
act as inoculum sources for subsequent hypovirulent
infections.

Survival of hypovirulent strains among high densities
of virulent inoculum has been discussed as a hinderance for
the establishment of hypovirulent populations of _§;

parasitica in heavily blighted chestnut groves (35, 85).

 

This may be an explanation of the failure to recover
hypovirulent strains on tree T (three sporulating virulent
natural cankers were overlooked when this tree was selected
for study). For hypovirulent strains to have an advantage
over virulent strains it may be necessary to leave inoculum
sources out for longer periods of time on trees which are
heavily infected. Results on tree E however tend to
contradict those found on tree T (tree E was also heavily
covered with virulent cankers). Hypovirulence found on
this tree, the first time, in July 1983 was apparently due
to a hypovirulent strain from a nearby inoculum source and
the: second time, in September 1983 was the result of a

INitural conversion within the tree. The success of

57

hypovirulent strains on tree E as compared to tree T may be
due to the possibility that virulent cankers on tree E were
not as well established as those on tree T. Nevertheless,
these data, along with data from tree F and tree G which
also provide evidence of natural conversion, are promising
in that they indicate that established virulent cankers may
be converted to hypovirulence naturally.

The dsRNA segments of CL1(GH2), the hypovirulent
strain from which the PCNB resistant strain used in the
field experiment was isolated, have been transferred
faithfully to numerous cultures derived from single
conidial isolates of this strain (Fulbright, personal
communication). It has also been shown (in European
hypovirulent strains) that dsRNA banding patterns will
remain intact through transfer (to virulent cultures) and
upon subculture (5). Therefore it appears that the new
dsRNA banding patterns detected in recovered field isolates
were probably not due to incomplete transfer of dsRNA to
conidia or through hyphal anastomosis.

When the chronological order of banding pattern
appearance is analyzed (Table 10) it is noted that pattern
D never preceeds A, B or C; pattern C does not preceed
patterns A or B except in T-F(R3-5) and pattern B never
preceeds pattern A. This suggests that in these
experiments a shift has occured from pattern A to B to C to
D. The hypovirulent culture obtained from the canker in
tree F is the result of a natural conversion of virulent

PCNB sensitive mycelium by dsRNA from the inoculum source

58
on the same tree. When this is taken into account, the
continuity of the pattern sequence is preserved, thus
explaining the appearence of pattern C before B in this
particular situation.

An exclusion principle has been proposed to explain
the loss of dsRNA segments found in Ustilago maydis , the
corn smut pathogen (61). In this system it was found that
certain dsRNA segments could not coexist in the same cell
protoplasm, either one or both of the segments were lost
from the fungal cells. Segments which were most affected
were of medium and low molecular weights. In cases where a
medium molecular weight segment was excluded from the
cytoplasm, leaving the low molecular weight segment, it was
hypothesized that the smaller molecules were retained
because they had a replicative advantage.

A similar principle may be acting on the Endothia
system, resulting in the observed alterations of the
original dsRNA banding pattern of the hypovirulent inoculum
source. All of the dsRNA banding patterns observed,
whether changed or not, had the original top (T) and bottom
(B) segments intact, thus providing evidence of the
CL1(GH2) PCNB-R origin of the dsRNA. However, the middle
segment (M) appeared to be unstable. Pattern B had an
additional dsRNA segment (LM), which was slightly smaller
than M; this segment may have been derived from M (M minus
approximately 200 base pairs). For a period of time both
bands could havev existed together, then, possibly because

the LM had a replicative advantage over M, M could have

59

been completely lost, giving rise to pattern C. Later,
pattern D could have appeared when LM was lost. Because T
and B were always retained, it indicates that they may have
been packaged together and therefore always transferred
together. Therefore, a possible explanation is that there
may have been at least two virus-like particles (VLPs)
infecting hypovirulent strains with the GH2 banding
pattern.

Most natural cankers found on American chestnut trees
at the Crystal Lake site were located primarily at wounds
created by branch sites or scars (Garrod, unpublished).
Wounds made by woodpeckers (which are similar to those made
by nail and cork borer) were also seen, however, no signs
of canker development were ever associated with these holes
(Fulbright and Garrod, unpublished). The wound type versus
infection results in this study were contrary to these
observations. It was found that nail and cork borer holes,
rather than branch scar wounds (whether artificial or
natural) were more likely to have become infected.

Bazzigher (16) found that in older wounds, there was
an increasing development of callus tissue, thus lessening
the chance for infection. He stated, "if a wound parasite

like E. parasitica cannot enter the wounded tissue early

 

enough, the infection fails or succeeds, but is greatly
delayed." When branches die and break away from tree
trunks a wound is created. These branch wounds expose all
tissues of the tree, new and old (78). When small branches

die, wounds close rapidly, but when large branches die, the

60

closing is slow (79). The mere physical structure of
branch sites may be conducive to infection. The sites may
trap and direct fungal spores into the crevices formed
between the branch stub and tree trunk, thereby providing a
direct pathway to susceptible tissues.‘ Because certain
branch sites are open to infection for a long period of
time, this could explain why there are so many naturally
existing cankers at these positions on American chestnut
trees. Woodpecker wounds however, could close much more
rapidly than branch site wounds and may therefore have a
shorter period of susceptibility to infection. In this
study, inoculum was provided to all wound types at the same
time, most likely before callus tissues could form in the
wounds. The data suggested that when exposed to inoculum,
before wound closing could take place, nail and cork borer
holes are more efficient (in becoming infected) than were
other wounds tested. If this experiment were to be
continued, based upon the conclusions drawn above, it would
be expected that, ultimately, an increasing number of
cankers would appear at natural branch scar wound positions
(because they may not close quickly) and that the number of
new cankers at the nail and cork borer sites would
diminish.

Support for the ideas presented above is provided by
average canker area data. Cankers located at branch scars
had the largest average area of all other wounds,
indicating that these wounds (branch sites and scars) are

most likely to be successful if infected (success being

61
defined in terms of area). The average area of cankers
located at slice wounds in September 1983 decreased due to

the failure to reisolate E. parasitica from existing

 

cankers.

Wounds at all distances were found to be equally
infected. Therefore, it has been shown that 1) conidia
can be carried in some manner at least 110 cm down a tree
trunk, encounter a wound and initiate a canker (if the
inoculum for these infections originated from the inoculum
source on the same tree), and 2) the susceptibility of the
host plant is homogeneous within the distance range of the
wounds on the trunk.

Bazzigher and Schmid (18) found that the
susceptibility to infection of four-year-old chestnut
plants differed within a range of 20 cm on the trunk.
Upper wounds were less likely to become infected than were
lower ones. In addition, the average length of cankers at
the upper wound spots were only half as large as were those
at lower positions. They believed that the reason for this
variation in susceptibility depended upon different healing
abilities of the tissue in different parts of the tree. It
must be noted that the trees used by Bazzigher and Schmid
were much younger than those used in this study, and,
therefore, tissues 20 cm apart in younger trees would be
more variable than would those the same distance apart in
older trees (such as those used in this study). This could
explain why they found differences in the number of

infections 20 cm apart and I did not. However, data

62

obtained in this study and that by Bazzigher and Schmid do
agree on the observations that lower cankers were larger
than those found further up the tree. For reasons stated
above in regards to tissue differences over distances in
younger versus older trees, an alternate explanation for
these data may be more reasonable. Perhaps cankers at
lower positions were larger and more numerous, in their
study and this one, because lower tissues were being
deprived of nutrients due to the girdling effect of higher
cankers. If this were true, lower cankers would be larger
because tissues below upper cankers would be weaker, and
therefore less able to defend against fungal invasion.

According to data collected in May 1983, wounds on
trees that had inoculum sources were more likely to become
infected than were wounds on trees that did not have
inoculum sources. These data show that wounds can become
readily infected if inoculum sources are available on
trees. Later in the study, data indicated otherwise. At
the July and September 1983 sampling dates, essentially the
same number of cankers were found on all trees regardless
of the previous presence of an inoculum source. If the
data are examined (Table 11) it becomes apparent that the
number of cankers increased equally on all trees. These
new cankers were initiated due to the spread of natural
inoculum that was equally available to all study trees.
The continuing increase of cankers masked differences that
had previously existed. If inoculum sources were

continually present during the entire study, from June 1982

63
until September 1983 (rather than from June 1982 until
December 1982), then it is likely that there may have been
a larger number of cankers on trees with inoculum sources
than on trees without.

Areas of cankers were compared according to the strain
type that was isolated from them. Canker area has been
used in other studies as a measure of virulence (33, 38,
86); strains which are more virulent produced larger canker
areas. It was found that virulent PCNB-S cultures were
isolated from cankers which were larger than those from
which virulent PCNB-R were isolated. This was expected
because it has been shown that PCNB-R strains are less
virulent than sensitive strains (Fulbright, personal
communication). The average canker area produced by
virulent PCNB-R strains decreased in September of 1983
because some of these cankers were converted to
hypovirulence and were therefore calculated with the
hypovirulent PCNB-R group. Canker areas in which
hypovirulent PCNB-S strains were isolated increased sharply
in September 1983 because these cankers were originally
initiated by virulent strains. It would be expected that
the areas of these cankers would not increase as rapidly as
they had because the cankers are now infected with

hypovirulent E; parasitica .

 

There were no significant interactions between factors
determined significant. This indicates that a particular
inoculum source strain does not prefer a particular wound

type at a specific wound distance.

64
Data obtained from this study may have useful
applications in achieving the natural dissemination of

hypovirulent Endothia parasitica . Control of virulent

 

cankers may be more rapidly and effectively reached with
use of hypovirulent inoculum sources, rather than the cork
borer, inoculum plug procedure (10). However, more
research must be done to determine the practicality of this

method.

PART II

THE DETERMINATION OF MATING TYPES OF AMERICAN VIRULENT
AND HYPOVIRULENT STRAINS OF ENDOTHIA PARASITICA .

THE SEARCH FOR dsRNA IN SINGLE ASCOSPORE ISOLATES
OF ENDOTHIA PARASITICA OBTAINED FROM (1) SEXUAL
CROSSES INVOLVING HYPOVIRULENT STRAINS AND (2)
PERITHECIA REMOVED FROM CANKERS WHICH WERE TREATED
WITH HYPOVIRULENT ISOLATES.

 

INTRODUCTION

Perithecia are rarely formed by European hypovirulent
strains in both controlled laboratory crosses and in
hypovirulent and/or hypovirulent-treated cankers (35, 58,
85). In the few cases in which perithecia are obtained
from these interactions, transmission of dsRNA to the
ascospores has not been observed (2, 30, 85).

Because the ascospores are probably the long distance
dispersal units of the fungus (2, 12, 13, 50, 51), it would
be most advantageous to find dsRNA in these spores to
facilitate the spread of hypovirulence.

Anagnostakis has identified mating type testers which
may be used in effecting controlled laboratory crosses of

Endothia parasitica . One mating type locus with two

 

alleles (A,a) has been found so far (4).
The objectives of this research were to (a) determine
the mating types (using established testers) of a number of

Endothia parasitica isolates collected from American

 

chestnut groves located in Michigan including the three
inoculum source strains used in part I of this thesis, and
to (b) detect dsRNA in ascospores isolated from perithecia
formed by the crosses above and from virulent cankers which

had been treated with American hypovirulent strains.

65

MATERI ALS AND METHODS

Isolates used

 

Thirty-five isolates of Endothia parasitica collected

 

from American chestnut groves located at Crystal Lake and
Grand Haven, Michigan and the three inoculum source strains
used in part I of this thesis (CL1, CL1 PCNB-R ,and

CL1(GH2) PCNB-R ) were paired with themselves, known mating
types (EP110 A and EP 329 a or EP 42 A and EP 339 a,
obtained from S. L. Anagnostakis [The Connecticut
Agricultural Experimental Station, New Haven, CT]) and each
other (in some cases). Field isolates analyzed were single
conidial, single ascospore and mass isolate bark cultures

of virulent and hypovirulent strains.

Sexual matings

 

Ascospores were obtained from perithecia from crosses
which involved hypovirulent strains. Representative single
ascospore cultures with hypovirulent-like morphologies were
analyzed for the presence of dsRNA according to the method
of Day et al. (30) with modifications by Dodds (31) and

Fulbright (38). In addition, ascospore cultures obtained

66

67
from the cross between CL1(GH2) PCNB-R and EP 110 A were
tested for resistance to pentachloronitrobenzene by
subculturing ascospore cultures onto ECA (71) with 100 pg
PCNB/ml. Control crosses between complementary mating
types accompianied each sexual mating trial.

Bark samples supporting perithecia were collected from
naturally and artificially inoculated cankers on fourteen
trees at the Crystal Lake site. Most cankers had been
treated with hypovirulent strains (see Table 1). Five
ascospores were randomly selected from each of five
perithecia that were isolated from each sample. Selected
single ascospore cultures with hypovirulent-like
morphologies were analyzed for the presence of dsRNA (using
methods described in part I of this thesis).

In vitro matings of E. parasitica were carried out

 

by the methods of Anagnostakis (3) except where modified as
below.

Sections of healthy chestnut stems, approximately 6 cm
x 1 cm x 0.7 cm were autoclaved twice at 15 psi for 30
minutes. The stems were transferred aseptically to 100 x
15 mm plastic petri plates and orientated so that the bark
surface was facing the top of the plate. Four percent
water agar (Difco; Detroit, MI), without biotin and
methionine, was autoclaved at 15 psi for 20 minutes, cooled
to approximately 40 C, and gently poured around the
chestnut stems to a depth which covered half the branch.

Inoculum was grown on potato dextrose agar (PDA;

Difco; Detroit, MI), which was not amended with biotin and

68

Table 1. Histories of cankers on American chestnut trees
from which bark samples were collected in June

 

1982.
Tree- canker canker 2 control
canker number origin treated achieved

5-2 artificial yes ?

5-8 artificial yes yes

6-1 artificial yes no
11-1 artificial no -
12-1 natural yes yes
13-2 natural yes no
14-3 artificial yes no
17-1 natural yes yes
18-2 natural yes no
20-1 natural yes yes
21-1 natural yes yes
21-2 natural yes yes
24-6 natural yes yes
24-1 natural no -
26-1 natural yes yes

 

lArtificial cankers were initiated in June 1981 by filling

a 7 mm diameter cork borer hole with a disc of

virulent E; parasitica mycelium, and the PDA it was
growing on. Inoculated wounds were covered with masking
tape to prevent dessication of the inoculum.

 

2Cankers were treated in July 1981 by filling 7 mm

diameter cork borer holes placed around the periphery of
established cankers with discs of

hypovirulent E; parasitica mycelium, and the PDA they
were growing on. Inoculated wounds were covered with
masking tape to prevent dessication of the inocula.

 

3Biological control of the treated cankers was considered
achieved if expansion of the cankers was prevented.

69
methionine. Seven mm diameter agar discs, cut from 5 to 10
day-old inoculum cultures, were placed on either side of
the chestnut stems. Petri plate perimeters were sealed
with parafilm to prevent contamination. Plates were
incubated at approximately 24 C under fluorescent lights
with a 16—hr photoperiod.

Ten to 20 days later, when conidia were issuing from
pycnidia, 3 ml sterile distilled H120 was added to each
plate. Conidia were spread over the stem surface by
shaking the plate vigorously for 30 seconds. Cultures were
incubated as described above.

Plates were checked every week for 16 weeks under a
dissecting microscope for signs of perithecial development

(usually 2-8 weeks).

Ascospore isolation

 

Ascospores were removed from perithecia and isolated
by the methods of Puhalla and Anagnostakis (71) except
where modified as follows.

Perithecia were aseptically dissected from the
chestnut bark with sterile tweezers. Individual perithecia
were macerated in a sterile test tube with 1 ml sterile
distilled water, vortexed for 1 minute to facilitate
ascospore release, and the resulting suspension was
distributed over 4% water agar, 0.1 ml per plate. Cultures
were incubated 24 hours at approximately 24 C under

fluorescent lights with a 16-hr photoperiod.

7O
Germinated ascospores were identified at 45X
magnification with a dissecting microsc0pe. Ascospores
were distinguished from conidial contaminants by the
presence of two germ tubes emerging from each cell of the
ascospore (48). Small pieces of agar, each containing a

single ascospore, were cut out and transferred to PDA, five

spores per plate. Cultures were incubated as described
above. Three to five days later, isolates were subcultured
onto fresh PDA. Morphologies of the single ascospore

colonies were rated according to mycelial appearance and

growth rate after seven days.

RESULTS

Mating types were assigned to 22 of the 35 isolates
tested. Seventeen were type A, five were type a and one
isolate formed perithecia when crossed with itself (Table
2). The remaining 12 field isolates did not form
perithecia. Seventeen of the 22 successful crosses
included hypovirulent strains. One of the seventeen was a
cross which was made between two hypovirulent strains (GH2
and GHU4). This particular mating was successfully
repeated four times (Fig. 1). Some single ascospore
cultures from the cross involving GH2 and GHU4 had unusual
culture morphologies on PDA which were typical of GHU4, the
more debilitated of the two hypovirulent strains. Nine of
these abnormal cultures from five separate perithecia were
analyzed for dsRNA upon two separate occasions. In all
cases dsRNA was not found to be present.

Mating types of the three inoculum sources were all
type a. Seven morpholigically abnormal single ascospore
cultures from two perithecia obtained from the CL1(GH2)
PCNB-R x EP 110 A cross were assayed for dsRNA. Again,
dsRNA was not found. Seventy-three single ascospore
cultures from the same two perithecia were tested for PCNB

resistance. Forty-three were resistant and 30 were not. A

71

72

 

 

Table 2. Mating types of Endothia parasitica strains
involved in successful sexual crosses.
Strain Location Isolate Type Mating Type
GH2 Grand Haven H, mass A
GH2B2 Grand Haven H, single C A
GH2F3 Grand Haven H, single C A
GH2K5 Grand Haven H, single C A
GH2L6 Grand Haven H, single C A
GH2N1 Grand Haven H, single C A
GHZIB Grand Haven H, mass A
GHA Grand Haven H, mass -
GHU2 Grand Haven H, mass A
GHU3 Grand Haven H, mass A
GHU4 Grand Haven H, mass a
GH6 Grand Haven H, mass A
GH7 Grand Haven H, mass A
GH8? Grand Haven H, mass A
GH14 Grand Haven H, mass A
TAclMSSI Grand Haven H, mass A
TACZMSSI Grand Haven H, mass a
CL1 Crystal Lake V, mass a
CL1-16 Crystal Lake V, single C a
CL4 Crystal Lake V, mass A
CL25 Crystal Lake V, mass A
TIZSSZ Crystal Lake V, single As a
T12SSS Crystal Lake V, single As A

 

1Mass isolates of E.

parasitica were obtained by

 

subculturing myceIIa which were growing on PDA from bark
samples that were taken from cankers on American chestnut

trees.
H = hypovirulent
V = virulent
C = conidial
As = ascospore

isolate formed perithecia with itself

73

 

Figure 1. Perithecia formed by the cross of two
hypovirulent strains, GH2 and GHU4.

74
X2 analysis (81) at P=0.05 found no significant difference
between these two values.

Abnormal morphology was also associated with a number
of the 300+ single ascospore cultures obtained from the
perithecia removed from field-bark samples. Six of the
cultures from a total of three perithecia were tested for
the presence of dsRNA (three of the isolates were checked

twice). dsRNA was not observed in any of the cultures.

DISCUSSION

Puhalla and Anagnostakis (71) found that perithecia of
Endothia parasitica could arise from the sexual fusion of
two genetically identical or different nuclei. This
suggested that the fungus was homothallic, but could
outcross preferentially. Evidence supporting these
conclusions was obtained from the present and other studies
(1, 4). Subsequent investigations have shown that the

ascogenous nuclei of E. parasitica can come from at least

 

three parents, one female and two males, thereby further
increasing the potential for genetic diversity provided by
sexual propagation (8).

The non-significant X2 value obtained in the
statistical analysis of PCNB resistance in progeny from the
CL1(GH2) PCNB-R, EP 110 A cross implies that resistance to
PCNB is carried on the fungal nuclear genome.

Mating methods described by Anagnostakis (3) proved to
be successful in 64 percent of the crosses performed.
Failure of the remaining matings to produce perithecia may
be partially attributed to (a) contamination of the

chestnut stems by Penicillium sp., which almost always

 

occured over the long incubation periods required by the

procedure and/or (b) the ability of some strains to

75

76
function only as males (4).

Hypovirulent strains involved in crosses were found to
be repeatedly successful in forming perithecia, contrary to
results reported by others (35, 85). Hypovirulent strains
used by others were primarily those obtained from European
and/or derivitives of European strains. This factor may
account at least partially for the observed differential
abilities of hypovirulent strains to form perithecia.
However, in this study as well as in others (30), dsRNA was
not found to be incorporated into the ascospores. This
phenomenon was observed in both single ascospore cultures
that were derived from crosses that involved one and two
hypovirulent strains and in those cultures that were
obtained from ascospores that were isolated from perithecia
which were present in hypovirulent cankers. Day et al.
(30) suggested that transmission of dsRNA through
ascospores derived from perithecia produced in healing
cankers was not observed because the perithecia were
collected from the central, older region of the canker
which may have been formed by islands of virulent mycelia.
This may be true, nevertheless, because dsRNA was never
found in controlled laboratory crosses involving
hypovirulent strains, it may be stated with confidence that
transmission of dsRNA to the sexual spores of Endothia

parasitica is very rare or does not occur.

 

Gaeumannomyces graminis , the fungal pathogen

 

responsible for the take-all diseases of cereals can be

infected by virus-like particles (VLPs) (76). VLPs in G;

77
graminis are transmitted inefficiently into the ascospores
(75). It has been suggested that because VLPs can pass
from mycelium into basidiospores (in Ustilago sp. infected
with VLPs) and conidia, a physical barrier preventing entry
into ascospores is unlikely (80). However, because
ascogenous hyphae grow very quickly, like hyphal tips, and
are quickly delimitated by septa (27), VLPs may not have
sufficient time to enter the developing asci (76). It has
also been suggested that the ascospore.environment could be
chemically unsuitable for virus-like particle inhabitation,
possibly due to an undesirable pH (76). Some, all or none
of these suggestions may be responsible for the apparent

exclusion of dsRNA from ascospores of E; parasitica .

 

Hypovirulence in E. parasitica is found in a number

 

of locations, great distances apart (38). Because dsRNA
has not been found to be incorporated into the ascospores
(the long distance disseminators of the fungus) it appears
that hypovirulence is not spread by these spores. The
question of the origin of dsRNA has remained unanswered.
It has been suggested that it was introduced into E;

parasitica by a closely related fungus, Endothia radicalis

 

 

(Schw.) DeNot. (35) or possibly by a unrelated species or
species' of fungi or other microorganisms (W. H. Weidlich,
personnal communication). It has also been suggested that
a proviral stage may be involved and that copies of the
dsRNA are integrated into the fungal chromosome as DNA
(85). No positive evidence for any of these hypotheses are

presently available.

APPENDIX

APPENDIX

ADDITIONAL FIELD EXPERIMENTS

Two additional field experiments were carried out at

the Crystal Lake site. The first was designed to study the

infection of the American chestnut by Endothia parasitica

 

based upon strain, wound type and time of inoculation after

wounding. E; parasitica strains and wound types were the

 

same as described in part I of this thesis. Eight sets of
wounds were made in a randomized order on two trees. Four
of the wound series were inoculated immediately after
wounding and the remaining four were inoculated 24 hours
later. Inoculations were made by directly placing a small
disc of fungal mycelium (growing on PDA) into the wounds.
Inoculation sites were covered with masking tape to prevent
dessication of the inoculum. Uninoculated wounds were not
covered.

Infections took place in all wound types that were
inoculated at both times. Uninoculated wounds were
infected equally well. These data indicate that the
inocula used could cause infection in wounds of these types
up to at least 24 hours after the wounding had occured.

The second field experiment involved a descriptive

78

79
study of existing natural cankers. Areas of 11 cankers on
7 trees were calculated four times over 14 months. The
presence of pycnidia and perithecia in the cankers were
also recorded as either "+" fruiting bodies present or "-"
fruiting bodies absent.

Data indicated that the average canker increased
approximately 2.5 times its original area over the 14 month
time span. Pycnidia were observed in all cankers at each
observation. Perithecia were observed in 18% of the
cankers at the initial observation and in 100% of the
cankers at the second observation (one month later) and in

all remaining observations.

LI ST OF REFERENCES

LIST OF REFERENCES

 

1. Anagnostakis, S. L. 1977. Vegetative incompatibility
in Endothia parasitica . Experimental Mycology 1:306-316.

2. Anagnostakis, S. L. 1978. The American chestnut: New
hope for a fallen giant. Connecticut Agricultural

Experiment Station Bulletin No. 777. 9 p.

3. Anagnostakis, S. L. 1979. Sexual reproduction of
Endothia parasitica in the laboratory. Mycologia
71:213-215.

 

 

 

 

4. Anagnostakis, S. L. 1980. Notes on the genetics of
Endothia parasitica . In Neurospora Newsletter No. 27.
p. 36.

5. Anagnostakis, S. L. 1981. Stability of
double-stranded RNA components of Endothia parasitica
through transfer and subculture. Experimental Mycology

6. Anagnostakis, S. L. 1982. Carpenter ants as carriers
of Endothia parasitica . In_ Proceedings of the USDA
Forest Service American Chestnut Cooperators' Meeting. H.
C. Smith and W. L. MacDonald, eds. West Virginia
University Books, Morgantown. pp. 111-113.

 

 

7. Anagnostakis, S. L. 1982. Biological control of
chestnut blight. Science 215:466-471.

8. Anagnostakis, S. L. 1982. The origin of ascogenous
nuclei in Endothia parasitica . Genetics 100:413-416.

 

9. Anagnostakis, S. L. and P. R. Day. 1979.
Hypovirulence conversion in Endothia parasitica .
Phytopathology 69:1226-1229.

 

 

10. Anagnostakis, S. L. and R. A. Jaynes. 1973. Chestnut
blight control: Use of hypovirulent cultures. Plant
Disease Reporter 57:225-226.

11. Anagnostakis, S. L. and P. E. Waggoner. 1981.
Hypovirulence, vegetative incompatibility, and the growth
of cankers of chestnut blight. Phytopathology
71:1198-1202.

80

81

12. Anderson, P. J. 1913. Wind dissemination of the
chestnut blight organism. Phytopathology 3:68.

13. Anderson, P. J. 1913. The morphology and life
history of the chestnut blight fungus. Pensylvania
Chestnut Tree Blight Commission Bulleten No. 7. 44 p.

14. Baxter, D. W. and F. C. Strong. 1931. Chestnut
blight in Michigan. Michigan Circular Bulletin No. 135.
18 p.

15. Bazzigher, G. 1958. Der wuchsstoffbedarf zweier
phytopathogener pilze. Phytopathol. Z. 32:352-358.

16. Bazzigher, G. 1981. Selection of blight-resistant
chestnut trees in Switzerland. European Journal of Forest
Pathology 11:199-207.

17. Bazzigher, G., E. Kanzler and T. Kubler. 1981.
Irreversible reduction in pathogenicity of Endothia
parasitica through transmissible hypovirulence. European
Journal of Forest Pathology 11:358-369.

 

 

18. Bazzigher, G. and P. Schmid. 1962. Methodik zur
prufungder Endothia - resistenz bei kastanien.
Phytopathol. Z. 45:169-189.

 

19. Beattie, R. K. and J. D. Diller. 1954. Fifty years
of chestnut blight in America. Journal of Forestry
52:323-329.

20. Berthelay-Sauret, S. 1973. Utilisation de mutants
auxotophes dans les recherches sur le determinisme de

"l'hypovirulence exclusive." Annales de Phytopathologie
5:318.
21. Biraghi, A. 1946. Il cancro del castagno causato da

Endothia parasitica . L'Italia Agricola 7:406-412.

 

22. Biraghi, A. 1950. Caratteri di resistanza in
Castanea sativa nei confronti di Endothia parasitica .
Boll. Staz. Patol. Veg., Anno VIII, Serie III, p. 167-171.

 

23. Biraghi, A. 1953. Possible active resistance to
Endothia parasitica in Castanea Sativa . Reports of 11th
Congress of the International Union of Forest Research
Organisation-Rome: 643-645.

 

24. Biraghi, A. 1953. Ulteriori notizie sulla resistenza
di Castanea sativa Mill. nei confronti di Endothia
parasitica (Murr.). Ann. Boll. Staz. Patol. Veg., Anno XI.
Serie III, p. 149-157.

 

 

 

82

25. Bozarth, R. F. 1979. The physio-chemical properties
of mycoviruses. In Viruses and Plasmids in Fungi. P A.
Lemke, ed. Dekker, New York. pp. 43-91.

26. Brewer, L. G. 1982. The present status and future
prospect for the American chestnut in Michigan. The
Michigan Botanist 21:117-128. ,

27. Carrol, G. C. 1967. The fine structure of the ascus
septum in Ascodesmis sphaerospora and Saccobolus
kerVerni. Mycologia 59:527-532.

 

 

 

 

28. Chmelo, R. A. and W. J. Kaczmarczyk. 1982. Isolation
and partial characterization of a virus-like particle and
its genome associated with Endothia parasitica strain 43.
In Proceedings of the USDA Forest Service American Chestnut
COOperators' Meeting. H. C. Smith and W. L. MacDonald,
eds. West Virginia University Books, Morgantown. pp.
165-169.

 

29. Clapper, R. B. 1952. Relative blight resistance of
some chestnut species and hybrids. Journal of Forestry
50:453-455.

30. Day, P. R., J. A. Dodds, J. E. Elliston, R. A. Jaynes
and S. L. Anagnostakis. 1977. Double-stranded RNA in
Endothia parasitica . Phytopathology 67:1393-1396.

 

31. Dodds, J. A. 1980. Revised estimates of the
molecular weights of dsRNA segments in hypovirulent strains
of Endothia parasitica . Phytopathology 70:1217-1220.

 

32. Dodds, J. A. 1980. Association of type 1 viral-like
dsRNA with club-shaped particles in hypovirulent strains of
Endothia parasitica . Virology 107:1-12.

 

33. Elliston, J. E. 1978. Pathogenicity and sporulation
of normal and diseased strains of Endothia parasitica in
American chestnut. In Proceedings of the American Chestnut
Symposium. W. L. MacDonald, F. C. Cech, J. Luchok and H.
C. Smith, eds. West Virginia University Books, Morgantown.
pp. 95-100.

 

 

34. Elliston, J. E. 1981. Hypovirulence and chestnut
blight research: Fighting disease with disease. Journal
of Forestry 79:657-660.

35. Elliston, J. E. 1982. Hypovirulence. In Advances in
Plant Pathology. P. H. Williams and D. Ingram, eds.
Academic Press, New York. pp. 1-33.

83

36. Elliston, J. E., R. A. Jaynes, P. R. Day and S. L.
Anagnostakis. 1977. A native American hypovirulent strain
of Endothia parasitica . Proceedings of the American
Phytopathological Society 4:83.

 

 

 

 

37. Fulbright, D. W. and W. H. Weidlich. 1983.
Interactions of American chestnut and Endothia parasitica
in Michigan. Annual Report Northern Nut Growers

Association 73:74-81.

38. Fulbright, D. W., W. H. Weidlich, K. Z. Haufler, C. S.
Thomas, and C. P. Paul. 1984. Chestnut blight and
recovering american chestnut trees in Michigan. Canadian
Journal of Botany 63: In press.

39. Gravatt, G. F. 1926. The future of the chestnut as
an orchard, ornamental and forest tree in the Southern
Appalachians. Phytopathology 16:643-644.

40. Graves, A. H. 1912. The chestnut bark disease in
Massachusetts. Phytopathology 2:99.

41. Graves, A. H. 1926. The cause of the persistent
development of basal shoots from blighted chestnut trees.
Phytopathology 16:615-621.

 

42. Grente, J. 1965. Les formes hypovirulentes d'
Endothia parasitica et les espoirs de lutte contre le
chancre du chataignier. C. R. Hebd. Seances Acad. Agr.

France 51:1033-1037.

43. Grente, J. and S. Sauret. 1969. L'hypovirulence
exclusive, phenomene original en pathologie vegetale. C.
R. Acad. Sci. Paris, D. 268:2347-2350.

44. Grente, J. and S. Sauret. 1969. L'"hypovirulence
exclusive' est-elle controllee par les determinants
cytoplasmiques? C. R. Acad. Sci. Paris, D. 268:3173-3176.

45. Grente, J. and S. Berthelay-Sauret. 1978.
Biological control of chestnut blight in France. _n
Proceedings of the American Chestnut Symposium. W. L.

MacDonald, F. C. Cech, J. Luchok and H. C. Smith, eds.
West Virginia University Books, Morgantown. pp. 30-34.

46. Griffin, G. J., F. V. Hebard and J. R. Elkins. 1983.
Blight resistance in American chestnut. Annual Report
Northern Nut Growers Association 73:66-73.

47. Griffin, G. J. , F. V. Hebard, R. W. Wendt and J. R.
Elkins. 1983. Survival of American chestnut trees:
Evaluation of blight resistance and virulence in Endothia
parasitica . Phytopathology 73:1084-1092.

 

84

48. Heald, F. D. 1913. A method of determining in
analytic work whether colonies of the chestnut blight
fungus originate from pycnospores or ascospores. Mycologia
5:274-277.

49. Heald, F. D. and M. W. Gardner. 1913. The relative,
prevalence of pycnospores and ascospores of the chestnut
blight fungus during the winter. Phytopathology 3:296-305.

50. Heald, F. D., M. W. Gardner and R. A. Studhalter.
1915. Air and wind dissemination of ascospores of the
chestnut-blight fungus. Journal of Agricultural Research
3:493-525.

51. Heald, F. D. and R. C. Walton. 1914. The expulsion
of ascospores from the perithecia of the chestnut blight
fungus, Endothia parasitica (Murr.) And. American Journal
of Botany 1:499-521.

 

 

52. Hepting, G. H. 1974. Death of the American chestnut.
Journal of Forest History 18:60-67.

53. Hodson, E. R. 1920. Is American chestnut developing
immunity to the blight? Journal of Forestry 18:693-700.

54. Jaynes, R. A. 1978. Selecting and breeding blight

resistant chestnut trees. In Proceedings of the American
chestnut symposium. W. L. MacDonald, F. C. Cech, J. Luchok
and H. C. Smith eds. West Virginia University Books,

Morgantown. pp. 4-6.

55. Jaynes, R. A., S. L. Anagnostakis and N. K. Van Alfen.

1976. Chestnut research and biological control of the
chestnut blight fungus. ‘In Perspectives in Forest
Entomology. J. F. Anderson and H. Kaya, eds. Academic

Press, New York. pp. 61-70.

56. Jaynes, R. A. and J. E. Elliston. 1978. Control of
Endothia parasitica cankers on American chestnut sprouts
with hypovirulent strains. In Proceedings of the American
chestnut symposium. W. L. MacDonald, F. C. Cech, J. Luchok
and H. C. Smith eds. West Virginia University Books,
Morgantown. pp. 110-114.

 

57. Jaynes, R. A. and J. E. Elliston. 1980.
Pathogenicity and canker control by mixtures of
hypovirulent strains of Endothia parasitica in American
chestnut. Phytopathology 70:453-456.

 

 

58. Jaynes, R. A. and J. E. Elliston. 1982. Hypovirulent
isolates of Endothia parasitica associated with large
American chestnut trees. Plant Disease 66:769-772.

 

 

85

59. Jaynes, R. A. and Van Alfen. 1977. Plant Disease
Reporter 61:1032.

60. Keefer, W. E. 1914. Pathological histology of the
endothia canker of chestnut. Phytopathology 4:191-200.

61. Koltin, Y. 1983. Exclusion mechanisms among the
Ustilago dsRNA viruses. In Double-Stranded RNA Viruses.
R. W. Compans and D. H. L. Bishop, eds. Elsevier

Biomedical, New York. pp. 495-501.

62. Kuhlman, E. G. 1983. Effects of hypovirulence in
Cryphonectria parasitica and of secondary blight
infections on dieback of American chestnut trees.
Phytopathology 73:1030-1034.

 

 

63. Lesemann, D. E. and R. Koenig. 1977. Association of
club-shaped virus-like particles with a severe disease of
Agaricus bisporus . Phytopathol. Z. 89:161-169.

64. Little, E. L. and J. D. Diller. 1964. Clapper
chestnut, a hybrid forest tree. Journal of Forestry
62:109-110.

65. Little, E. L. 1980. The Audubon Society Field Guide
to North American Trees. Alfred A. Knopf, New York. 714

p.

66. Mittempergher, L. 1978. The present status of
chestnut blight in Italy. In Proceedings of the American
chestnut symposium. W. L. MacDonald, F. C. Cech, J. Juchok
and H. C. Smith, eds. West Virginia University Books,
Morgantown. pp. 34-37.

67. Moffitt, E. M., and R. M. Lister. 1975. Application
of a serological screening test for detecting
double-stranded RNA mycoviruses. Phytopathology
65:851-859.

68. Murrill, W. A. 1906. A serious chestnut disease.
Journal of the New York Botanical Garden 7:143-153.

69. Nash, B. L. and W. J. Stambaugh. 1982. Disease
incidence, symptomatology, and vegetative compatibility
type distribution of Endothia parasitica on oak and
chestnut hosts in North Carolina. In Proceedings of the
USDA Forest Service American Chestnut COOperators' Meeting.
H. C. Smith and W. L. MacDonald, eds. West Virginia
University Books, Morgantown. pp. 74-82.

 

 

70. Newhouse, J. R., H. C. Hoch and W. L. MacDonald.
1983. The ultrastructure of Endothia parasitica .
Comparison of a virulent with a hypovirulent isolate.
Canadian Journal of Botany 61:389-399.

 

 

86

71. Puhalla, J. E. and S. L. Anagnostakis. 1971.
Genetics and nutritional requirements of Endothia
parasitica . Phytopathology 61:169-173.

 

 

72. Rankin, W. H. 1912. The chestnut tree canker
disease. Phytopathology 2:99.

73. Rankin, W. H. 1913. Some field experiments with the
chestnut canker fungus. Phytopathology 3:73.

74. Rankin, W. H. 1914. Field studies on the endothia
canker of chestnut in New York state. Phytopathology
4:232-260.

75. Rawlinson, C. J. and K. W. Buck. 1981. Viruses in
Gaeumannomyces and Phialophora spp. In Biology and Control
of Take-all. M. J. C. Asher and P. J. Shipton, eds.
Academic Press, London. 538 p.

 

 

76. Rawlinson, C. J., D. Hornby, V. Pearson and J. M.
Carpenter. 1973. Virus-like particles in the take-all
fungus, Gaeumannomyces graminis . Annals of Applied
Biology 74:197-209.

 

77. Russin, J. S., L. Shain and G. L. Nordin. 1982.
Summary research, sporulation and dissemination of
hypovirulent strains of the chestnut blight fungus at the
University of Kentucky. _In_Proceedings of the USDA Forest
Service American Chestnut Cooperators' Meeting. H. C.
Smith and W. L. MacDonald, eds. West Virginia University
Books, Morgantown. pp. 40-48.

78. Shigo, A. L. 1979. Tree decay an expanded concept.
USDA Forest Service Bulletin No. 419. 72 p.

79. Shigo, A. L. and H. G. Marx. 1977.
Compartmentalization of decay in trees. USDA Forest
Service Bulletin No. 405. 73 p.

80. Spire, D. 1971. Virus des Champignons. Physiologie
Vegetale 9:555-567.

81. Steel, R. G. D. and J. H. Torrie. 1980. Principles
and Procedures of Statistics. A Biometrical Approach.
McGraw-Hill, New York. 633 p.

82. Studhalter, R. A. 1914. Insects as carriers of the
chestnut blight fungus. Phytopathology 4:52.

83. Thompson, P. W. 1967. A unique American chestnut
grove. Michigan Acad. 1:175-178.

87

84. Turchetti, T. 1978. Some observations on the
"hypovirulence" of chestnut blight in Italy. In
Proceedings of the American chestnut Symposium. W. L.

MacDonald, F. C. Cech, J. Luchok and H. C. Smith, eds.
West Virginia University Books, Morgantown. pp. 92-94.

85. Van Alfen, N. K. 1982. Biology and potential for
disease control of hypovirulence of Endothia parasitica .
Ann. Rev. Phytopathol. 20:349-362.

86. Van Alfen, N. K., R. A. Jaynes, S. L. Anagnostakis and
P. R. Day. 1975. Chestnut blight: Biological control by
transmissible hypovirulence in Endothia parasitica .
Science 189:890-891

 

87. Wendt, R., J. Weidhaas, G. J. Griffin and J. R.
Elkins. 1983. Association of Endothia parasitica with
mites isolated from cankers on American chestnut trees.
Plant Disease 67:757-758.

88. Willey, R. L. 1982. Natural dissemination of
artificially inoculated hypovirulent strains of Endothia
parasitica . In Proceedings of the USDA Forest Service
American Chestnut Cooperators' Meeting. H. C. Smith and W.
L. MacDonald, eds. West Virginia University Books,
Morgantown. pp. 117-127.

 

MICHIGAN STATE UNI IVERSITIY LIB

IIII III IMIIIO WJIIII IIIIIII'IIII