3-“

2:3.
'fkff'if-‘Y

m}? {f‘fi‘x .
. ‘v”. ’%{I\ Il‘rfv,

‘Z-r‘ .

.2' 223;.
"\ 4:;‘<¢;.’“ 62‘:
"(ff/III.) V}- (""3"
J

‘V"

:1),
V‘ ::!:::v "V
4;: “.2234; ‘3‘ ' 7""-

xyj, -,
£351. 1

3.5;}?

.v
fi-f'ulx‘r
L: (“~- 05.3%;

“;‘y\:., 1 33'1"

., -. 2- 1,» ‘
L».’,WE 4’35 fact-£4: v '
‘4. 2"; .1 -
..v. ,- *
:;.::‘l'.?~‘I;B}“/flfii ; 3h" 3‘4"},
~ . |, 'V
v Ix ".33“ fi fI"

v ‘.

" .‘rn Wu}
W4

I"; 21%;? I\_)- $733.45.:

4 V. ’3'“

,x
‘I x?
J .J (RI?
': A: ”J n"

.?$é/4§‘L}I‘A i1, “”2

:""L.
10"
{lg/4‘“

‘,
“-4. . 2
‘. 2‘43" (:7 ".11 3:,“- ‘ ”1:43,,”
MI} “W, l\“vj‘[“(‘( Iv‘l‘
r‘fifl
.t‘"

x “.2
g}

.

2, 21;} 41',
If. ‘."' ‘6"-
rW-nufl' f I

r 4(

.f: "((1. g

inx‘ffir

I'u' I/
”I, '14:}. 47:. I"
,‘v.‘ ‘"‘ {mug}

f: 20311:; '42:?
f II

.5
5144'.”

‘ II)";
)‘(ivf'If'bl‘j {5:31

0"“?
;(}‘}‘_2f.‘

4",)"~
IJ’1I’I2I"I"” 35‘2"":

I"
.
x 3""

3rd;
1’)

~

“nu-2‘1")“ II"??? '
. : a u’frrdfit‘g‘gglfig , 1‘!
" MI
1;! SI! “‘1‘!“ i 39‘" "6‘91?“
"' .4’1"?
MP!"

(“'7'

“I $4.

"‘55 I” “ééyyfii‘

‘L

m .3; '5 ,3i
(4.1.x?! ‘3‘"; "IE”

V \*-¢.“L
1132‘ 22.2%

35'. . "3,:

("5 ~ ‘73:! 'I ‘

A 4“" 32? 3:;

. W *‘tfitgxfig; g} ""3233:

, ~

fg‘5W (“I"Cgflézxf‘fis‘v Ini‘is‘vfi'idf}
3153* ‘"fi'"‘: ‘

Ml\~~.f

A,” y. ‘9‘“

(3.. 5:54.
THEE-‘49 f: "2f
+1=f¢fi€fmz
«nfiudugyg ~.
. 3535"“;

. “£2

/ I
fix"? N 35‘3”“
{fig-(v, , “,4" If
"Sum/"l ““21 13L" 9
"I"; V:
fifl'.‘qa "I":

"ZN?"

/ 2" 1’
I’VI‘H,‘J/{I'

v9
-..'.“ . .
"i‘uf‘. ,

‘-‘.2}f‘¢: 2.‘

"’2‘; ”My:

41"]:X’Z"
732x

2 1‘
«(4'4"

”'6‘:
J’"|I"(
5.};

‘:II
23. f ‘
"kit’s": III."":;II1‘:
pudfll': ”Hf“;
JI‘I‘W (‘ E

'23} I“?
”W"

’ Hf}
Via/1:4" 4,’ «1:4, {4:2}: I
f"; "IV/:61“ '3)?!" "3/

{1x4

 

1:" a”,

Ell/N

fix. .4},
“3%

§,fa’%l! 1%

"£522

WHI‘ ' £31
'. '4 "
g???"

«i
:23» _ urn-9:"
95“ s

éofgqy‘.

17 31:22]"-

‘\"4.
V 'f‘ws“.
» "v ‘33,,“

22.52
7

‘52 vi"
"4'1. VV"? “EWEJIS‘

 

 

-nQ-v 11"

MICHIGAN STATE UNIV

1 l l RSI" UBR
um» v llllll : ' Hill}!

 

 

 

 

 

 

 

 

I'll lullzmmlllilll
Mm“ 3m _ 3 1293 00605 7_339
University

 

 

 

 

 

 

 

 

 

 

 

 

 

 

This is to certify that the

thesis entitled

THE EFFECT OF CRYPHONECTRIA PARASITICA
0N HATER RELATIONS AND XYLEM FUNCTION IN
AMERICAN CHESTNUT (CASTANEA DENTATA)

presented by

Patricia S. McManus
has been accepted towards fulfillment

of the requirements for

Masters degree in Botany 8: Plant Pathology

W fwd

Major professor

 

Date NOT/fimbcr Z (938'

0-7639 MS U is an Affirmative Action/Equal Opportunity Institution

 

-- c-

PLACE IN RETURN BOX to remove this checkout tram your record.
TO AVOID FINES rotum on or botoro duo duo.

'_—————————_=———_——11

DATE DUE DATE DUE DATE DUE

imgzlzmz
—.19-6—6—4-0,

lgm

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

l‘T‘r—fi‘f

 

 

MSU Is An Affirmdlvo ActioNEqual Opportunity lnditution

THE EFFECT OF CRYPHONECTRIA PARASITICA
ON WATER RELATIONS AND XYLEM FUNCTION IN
AMERICAN CHESTNUT (CASTANEA DENTATA)

BY

Patricia S. McManus

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

1988

ABSTRACT
THE EFFECT OF CRYPHONECTRIA PARASITICA

ON WATER RELATIONS AND XYLEM FUNCTION IN
AMERICAN CHESTNUT (CASTANEA DENTATA)

BY

Patricia S. McManus

Chestnut blight has commonly been regarded as a phloem
disease due to conspicuous stem cankers that result from
infection by the fungal pathogen Cryphonectria (Endothia)

parasitica (Murr.) Barr. The effect of g; parasitica on

 

water relations and xylem function in Castanea dentata was
examined. Wilting of leaves distal to cankers suggests that
water flow is obstructed and that xylem tissue is critically
damaged. Stomatal conductances (gs), transpiration rates
(E) and leaf water potentials were measured on a diurnal
basis for leaves distal to cankers initiated by virulent and
hypovirulent strains of g; parasitica. Hydraulic
conductance (Rh), g3 and B were reduced significantly (p <
0.05) for stems that were inoculated with a virulent strain
relative to control stems. These effects were less
pronounced in stems inoculated with a hyvairulent strain.
9; parasitica was observed microscopically in the xylem of
cankers initiated by a virulent strain. The pathogen was
isolated from all the xylem rings of stems with reduced Kh'

but not from stems that showed no such impairment.

To my mother, whose fascination with
nature and biology has always inspired me.

iii

ACKNOWLEDGEMENTS

I gratefully acknowledge the guidance and enthusiastic
support offered by my major professor, Dr. Frank Ewers. His
expertise in plant anatomy and physiology was absolutely
essential in formulating experiments.

I thank Dr. Dennis Fulbright for serving on my guidance
committee and for the many opportunities he provided. He
made possible my attendance at conferences so that I was
able to meet several of the chestnut blight researchers
cited in this thesis. He included me on field trips and
provided trees at Russ Forest. I also thank Dr. Robert
Scheffer for serving on my guidance committee and for many
insightful suggestions.

I wish to thank my colleagues in Drs. Ewers and
Fulbright's labs for their scientific input and constant
companionship. I learned a lot about science and loads
about life from these interactions.

I cannot thank my husband, Mark, enough for his
contributions, especially in the preparation of this thesis.
Besides the continual emotional support and extraordinary
acts of patience, he prevented me on numerous occasions from
putting my fist through the computer screen.

Finally, I thank my parents, Charles and Nora McManus,

for teaching me to recognize the hard work required in
obtaining anything of value---especially an education.

iv

LIST OF
LIST OF
LIST OF

CHAPTER

CHAPTER

CHAPTER

SUMMARY
LIST OF

TABLE OF CONTENTS

TABLES O O O O O O I O O O O O O O O O O O O
FIGURES O O O O O O O O O O O O O O O I O O O
ABBREVIATIONS O O C C O O O O O O O O O O O O

1.

INTRODUCTION, LITERATURE REVIEW, AND
OBJECTIVES . . . . . . . .
A. Introduction . . . .
B. Literature Review . .
C. Objectives . . . . .

THE EFFECTS OF VIRULENT AND HYPOVIRULENT
STRAINS OF CRYPHONECTRIA PARASITICA ON
WATER RELATIONS AND XYLEM FUNCTION IN
AMERICAN CHESTNUT . . . . . . . . .
A. Introduction . . . .
B. Materials and Method
C. Results . . . . . .
D. Discussion . . . . .

CHARACTERIZATION OF THE CHESTNUT BLIGHT
CANKER AND THE LOCALIZATION AND
ISOLATION OF THE PATHOGEN CRYPHONECTRIA
PARASITICA . . . . . . . . . . . . .
A. Introduction . . . .
B. Materials and Method
C. Results . . . . . . .
D. Discussion . . . . .

AND CONCLUS IONS O O O O O O O O O O O O O 0
REFERENCES 0 O O O O O O O O O O O O O O O 0

vi
vii
viii

wle-J

15
16
18
21
24

35
36
38
41
45

60
65

LIST OF TABLES

Data collected from 12 natural cankers and 6
control stems at Grand Haven, 1986, 1987 and
1988. Mean value 1 SE.

Data collected from induced cankers, girdled
stems and control stems at Russ Forest, 1988.
Mean values 1 SE; n=8.

Leaf specific conductivities (LSC) and
theoretical stem water potential gradients
(dP/dx) at Grand Haven, 1987.

Percentage of xylem chips yielding g;
parasitica, Grand Haven, 1987. Ring 1 is

 

outermost; Ax=anomalous xylem.

Percentage of xylem chips yielding g;
parasitica, Russ Forest, 1988. Ring 1 is
outermost.

 

Summary of data collected on samples from Grand
Haven, 1986, 1987 and 1988, and Russ Forest,
1987 and 1988.

vi

LIST OF FIGURES

Stomatal conductance, transpiration and leaf
water potential at Grand Haven, 1987. Each
point is the mean of 3 measurements; bars
indicate standard errors.

Stomatal conductance, transpiration and leaf
water potential at Russ Forest, 1988. Each
point is the mean of 7 or 8 measurements; bars
indicate standard errors.

Naturally occurring virulent (A) and
hypovirulent (B) cankers from Grand Haven.

Transverse sections of naturally occurring
virulent cankers (A), hypovirulent cankers (B)
and control stems (C).

Cankers induced by the virulent strain CLl-16
(A) and the hypovirulent strain GH2 (B).

Transverse sections revealing functional
vessels (marked with red dye) in a virulent
canker (A), a hypovirulent canker (B) and a
healthy stem (C).

Transverse sections revealing functional
vessels (marked with red dye) in cankers
induced by the virulent strain CL1-16 (A) and
the hypovirulent strain GH2 (B).

Longitudinal section of chestnut xylem tissue
infected with Q; parasitica. 360x (A); 900x
(B).

 

Model for the steps leading to canker formation
and the death or survival of branches of
Castanea dentaga infected by Cryphonectria
parasitica. Broken lines represent less
clearly documented steps.

 

 

 

vii

LIST OF ABBREVIATIONS

dP/dx = Theoretical stem water potential gradient (MPa m'l)
dsRNA = Double-stranded ribonucleic acid

E = Transpiration (mmol in-2 8'1)

98 = Stomatal Conductance (mmol In.2 3-1)

Kh = Hydraulic conductance per unit length (m4 MPa'1 s'l)

LSC = Leaf specific conductivity (m2 s"1 MPa-l)

viii

CHAPTER ONE

INTRODUCTION, LITERATURE REVIEW AND OBJECTIVES

A . I NTRODUCT ION

The American chestnut [Castanea dentata (Marsh)

 

Borkh.] was once the dominant tree species in eastern
North America, accounting for one quarter of the hardwood
trees in the deciduous forest extending from lower Ontario
to the southern Appalachian mountains. Chestnut timber is
strong and extremely decay resistant and was therefore
used in the manufacture of fences, railroad ties,
furniture, barns and houses by early American settlers.
The chestnut timber industry was worth millions of
dollars, and the nuts provided an important cash crop for
many rural families (Roane et al., 1986).

In the early 19003 the fungus Endothia parasitica

(Murr.) Anderson, later renamed Cryphonectria parasitica

 

(Murr.) Barr (Barr, 1978), as it will be referred to in
the ensuing text, was introduced into New York City,
probably on Chinese chestnut (Castanea mollissima Bl.)
nursery stock. While the Chinese chestnut trees tolerated
the fungus, it proved to be a devastating pathogen of
American chestnut. By the 19508 an estimated four billion
trees had been annihilated by the blight; the once
dominant species was but a minor component of the
understory, represented primarily by small suckers arising
from the rootstock of dead and dying trees. Considering
its former economic significance and ecological dominance,

the demise of the American chestnut is a great tragedy.

B. LITERATURE REVIEW

Following its discovery in 1904 (Merkel, 1905) and
the description of the pathogen in 1906 (Murrill, 1906),
the "chestnut bark disease” was the concern of several
pioneering studies (Anderson, 1913; Heald, 1910; Rankin,
1914; Rumbold 1920). Research was active through the
19203 but waned throughout the Great Depression and the
Second World War. During this quiescent period for
research, chestnut blight had spread throughout the
natural range of the host in North America and, in 1938,
was discovered in Italy on European chestnut trees

(Castanea sativa Mill) (Biraghi, 1946).

 

It was not until 1950, when surviving chestnut trees
were discovered in Europe (Biraghi, 1950a,b), that
chestnut blight research was revived. Certain trees,
though heavily infected with Cryphonectria parasitica,
were surviving, and cankers of abnormal morphology were
noticed on such trees. Rather than having smooth, sunken
bark blemished by abundant fructifications, these cankers
appeared swollen, and the bark was often cracked but
lacked stromata. The pathogen isolated from the abnormal
cankers showed unusual morphology, pale pigmentation and
reduced rate of growth in culture relative to isolates
from normal cankers (Grente, 1965). Such isolates were
termed "hypovirulent" and have been of great interest to

contemporary plant pathologists and molecular biologists

4
(Elliston, 1987; Fulbright et al., 1983; Paul and
Fulbright, 1988; Tarteglia et al., 1986; Van Alfen et al.,
1975). The presence of double-stranded deoxyribonucleic

acid (dsRNA) in the cytoplasm of Q; parasitica is strongly

 

correlated with hypovirulence (Fulbright, 1984). No study
has unequivocally correlated hypovirulence to the
occurrence of abnormally swollen cankers. However, in
situations where hypovirulent isolates have been
confirmed, it is convenient to refer to the abnormal
cankers as "hypovirulent cankers” and to the normal,
sunken lesions as "virulent cankers".

Though the presence of stem cankers is the most
definitive symptom of chestnut blight, perhaps more
striking is the yellowing and flagging of leaves distal to
cankers. The principal cause of this symptom in many
vascular wilt diseases is occlusion of the water transport
pathway (Ayres, 1978). However, chestnut blight is not a
wilt disease; its disease cycle is dramatically different
from that of a wilt disease. In wilt diseases the
pathogen is confined primarily to host xylem tissue. Some
examples of fungal wilt diseases are Dutch elm disease,

oak wilt, Verticillium wilts and Fusarium wilts. Canker

 

diseases are characterized by extensive colonization of
phloem tissue, the vascular cambium and outer xylem tissue
(Agrios, 1988). Some fungal canker diseases are chestnut

blight, Nectria canker, Cytospora canker, Hypoxylon canker

 

 

and black knot. In canker diseases, colonization of the

5
outer growth ring could prove detrimental to xylem
function. In addition, disruption of the cambium could
result in the unusual development of secondary vascular
tissues that has been reported for chestnut blight (Ewers
et al., in press).

Water relations in fungal diseases

 

The role of fungal pathogens in host water relations
and hydraulic conductance has been and continues to be an
important area of study in plant pathology. Water
relations parameters such as stomatal conductance (gs),
transpiration rate (E) and leaf water potential have often
been addressed in the literature without adequate
discussion of the movement of water in xylem tissue
(Duniway and Slayter, 1971; Helms et al., 1971). More
enlightening to our understanding of the water economy of
diseased plants are the results of studies that recognize
water in a plant as a continuous column, from absorption
at the roots to evapotranspiration at the leaf surfaces. A
This review and Chapter II will focus on research that
couples leaf water relations and hydraulic conductance of
the xylem.

Central to any discussion of xylem pathology are the
vascular wilt diseases. A wilt disease arises when a
pathogen invades and establishes itself in conductive
xylem elements, causing discoloration of xylem tissue and
wilting of the leaves. By the strictest definition, the

pathogen remains confined to the xylem until the host is

6
moribund, and then spreads to outer parenchymatous tissues
(Ayres, 1978). This review, however, will include fungal
diseases in which the pathogen is disruptive to xylem
function and water relations at any time during the
disease, whether the organism is confined to the xylem or
not.

This review will concern only vascular wilts and
canker diseases caused by fungi and not bacterial
diseases. In contrast to wilt diseases, in which the
pathogen resides primarily in the xylem tissue, canker
diseases caused by fungi originate when a propagule enters
its host via a wound or a branch scar and proliferates in
the phloem tissue (Agrios, 1988). In defense, the host
may produce callus and wound periderm in an attempt to
confine the pathogen. The result is a rough, often split
and swollen localized stem lesion. When the rate of
fungal growth exceeds the host's ability to produce
callus, the hyphae will penetrate the vascular cambium and
invade xylem tissue. If colonization is extensive,
hydraulic conductance of the xylem will be significantly
reduced (Chapter 2) producing symptoms similar to those of
wilt diseases.

The bulk of the literature favors the theory that
wilting caused by fungal pathogens either in wilt diseases
or advanced canker diseases, is due to reduced water
supply to the leaves rather than to excessive

transpiration. There is evidence, however, that when

7
toxins are involved, as in Cephalosporium stripe of wheat
and several Fusarium diseases, wilting may be due to
altered stomatal functioning (Creatura et al., 1981;
Scheffer and Livingston, 1984; Turner and Graniti, 1969).
Wilt diseases have been studied more frequently in this
regard than have canker diseases.

A series of studies on Verticillium wilt of

 

Chrysanthemum demonstrated the relationship of water flow

 

in the xylem and water loss at the leaves. MacHardy et a1.
(1976) reported an increase in stomatal resistance and a
decrease in relative water content (RWC) immediately
before the development of wilt symptoms. Infected plants
did not conduct dyes or 14C-mannitol to wilted regions
(MacHardy et a1. 1976, Hall et a1. 1975), and there was no
evidence that the fungus altered cell permeability in the
host prior to the appearance of symptoms (Hall and Busch,
1971; MacHardy et al., 1974). They concluded that
occlusion of vascular elements altered stomatal behavior
and RWC, resulting in leaf wilt.

Melching and Sinclair (1975) observed that hydraulic

conductance was reduced in elms harboring Ceratocystis

 

2131. In a separate study, MacHardy and Beckman (1973)
showed that a decrease in transpiration accompanied the
onset of symptoms and that stomatal behavior of elms
afflicted with Dutch elm disease was similar to that of
healthy trees under water stress. These results suggested

that the reduced transpiration rates were a response to

8
water stress which was probably the result of pathogen-
induced xylem impairment. This is consistent with the
results of Duniway who concluded that high xylem
resistance to water flow is the cause of leaf wilting in

Fusarium-infected tomato (1971) and Phytophthora-infected

 

safflower (1975). Similarly, Helms et a1. (1971) reported
that for Ponderosa pine, increased vascular colonization
by Verticladiella wagenerii was accompanied by decreased
host transpiration.

Mechanisms of Xylem Dysfunction

 

Possibly the most debated aspect of xylem pathology
is the mechanism by which water transport is obstructed.
Most studies favor the unlikely prospect that a single
mechanism acts in blocking xylem. In fact, the mechanisms
are not distinct but are interrelated and could act
simultaneously in disabling xylem tissue.

Masses (1895) and Smith (1899) theorized that vessels
of several crops were physically occluded by wefts of
mycelia. This theory was disputed by Gottlieb (1944) and
later by Talboys (1978). Talboys described a laboratory
model in which the physical presence of hyphae did not
adequately account for the significant increase in

resistance observed in the xylem of Verticillium-infected

 

tomato plants.
Brandes (1919) was the first of many to report "toxic
excretions" by fungi in culture. The recognition and

description of host-selective toxins permitted intensive

9
studies of disease physiology (Pringle and Scheffer, 1964;
Scheffer and Livingston, 1984). Host-selective toxins
affect host water relations by altering cell permeability,
protein synthesis, respiration and carbon dioxide

fixation. A toxic metabolite of Cephalosporium gramium

 

caused stomata on wheat leaves to respond in an abnormally
slow manner to changes in leaf water stress (Creatura,
Safir and Scheffer, 1981). Therefore, the stomata tended
to remain open and lose water under moisture stress
conditions. Toxins have been shown to induce changes in
guard cell permeability to potassium ion uptake and
thereby influence stomatal functioning (Turner and
Graniti, 1969).

Kitajima (1927) reported on the possible role of a
toxin, later described and named diaporthin (Hardeeger,

1966), that was produced by Cryphonectria parasitica and

 

which might be involved in chestnut blight. While many
workers isolated and identified toxins in culture
filtrates, there was no evidence that these substances
were produced lg 3339, and thus the toxin theory lacked
credence.

High molecular weight products of fungal pathogens,
such as polysaccharides, glycoproteins and enzymes, have
been implicated as factors in wilt diseases. Van Alfen
and Turner (1975) concluded that a glycopeptide produced
by Ceratocystis ulmi was responsible for reducing flow

through Dutch elm disease-afflicted stems to 20% of

10
normal. The macromolecule-induced xylem resistance was
instantaneous (Van Alfen and Allard-Turner, 1979). This
implied that reduced flow was due to physical plugging of
pit membranes, and not to toxic properties of the molecule
or to increased sap viscosity and gelation, which would
have taken periods of hours or days (VanderMolen et al.,
1977).

Cell wall degrading enzymes are produced by a number
of pathogens (Dimond 1970; Beckman, 1987). These enzymes
permit the rapid colonization of vascular tissue and the
liberation of carbon for pathogen nutrition. Enzymatic
activity can be detected in advance of the hyphal tip.
Thus, there could be damage to vessels even before fungal
invasion. Such damage could result in gas embolisms as
discussed below. In chestnut blight cankers, mycelial

fans or individual hyphae of g; parasitica killed host

 

phloem cells apparently before contacting them (Hebard et'
al., 1984). This suggests that a toxin is secreted at the
advancing fan margin or hyphal tip. Oxalic acid, which
was identified at the advancing margin of cankers, was
found to act synergistically in gitrg with
polygalacturonase activity in degradation of calcium
polypectate, and to display toxicity toward chestnut
protoplasts lg yitrg (McCarroll and Thor, 1978). Further
investigation showed that oxalic acid was produced in
yitgg by virulent but not hypovirulent strains of g;

pararsitica (Havir and Anagnostakis, 1983). The results

 

ll
of these studies could prove useful in studying host-
pathogen relationships lg glggl

Many workers have noticed abundant tyloses occluding
xylem elements in plants harboring pathogenic fungi
(Bramble, 1938; Sinclair and Campana, 1978; Bishop and
Cooper, 1984; Scheffer and Walker, 1953; Beckman et al.,
1953). Pioneering work on chestnut blight indicated the
abnormal appearance of tyloses in the outermost xylem ring
(Bramble, 1938). Powers (1954) observed tylosis formation
in tobacco plants afflicted with black shank. Tyloses
grew into not only air-filled vessels, but also into
vessels with intact water columns. Thus, it was
speculated that abundant tyloses, rather than a
discontinuous water column might obstruct water movement
in diseased plants.

The formation of gels and gums discussed below often
accompanies tylosis formation (Bishop and Cooper, 1984;
Bramble, 1938; Stanova, 1985). Therefore it is impossible
to distinguish their roles in xylem blockage. Gels and
gums arise as a general response to invasion by
microorganisms. They are commonly believed to be of plant
origin, and are thought to result from enzymatic
degradation of cell walls (Mishaghi et al., 1978; Scheffer
and Walker, 1953; VanderMolen et al., 1977). Only
recently have these substances been characterized
biochemically (Beckman, 1987). Gums in the xylem of peach

trees infected by Cytospora cincta inhibit water transport

 

12
and are considered to be the primary cause of leaf wilting
and senescence (Hampson and Sinclair, 1973; Stanova,
1985).

Though gas embolisms were suggested as a factor in
pathogen-induced xylem dysfunction over fifty years ago
(Linford, 1931), the topic received little attention until
the long-standing belief that vascular occlusions were the
primary cause of xylem dysfunction was challenged by
Newbanks et a1. (1983). In American elms afflicted with
Dutch elm disease, they identified non-conducting vessels
in infected stems before gums, tyloses or hyphal
penetration of vessels were detected. The authors
suggested that enzymatic degradation of vessel walls by
the pathogen permitted air-seeding and vapor blockage to
occur. Zimmermann (1983) emphasized the vulnerability of
ring-porous trees such as chestnut, elm and oak, to vapor
blockage because these trees depend almost exclusively
upon superfically located earlywood vessels for their
water. He speculated that vapor blockage could be the
most direct cause of leaf wilting and death in chestnut
trees afflicted with chestnut blight.

The conspicuous bark cankers that characterize
chestnut blight have attracted researchers to probe canker
morphology and canker histopathology. Indeed, chestnut
blight is commonly considered a disease of the phloem
(Hebard, 1982; Russin and Shain, 1984), even though phloem

dysfunction has never been demonstrated to cause leaf wilt

l3

and leaf senescence. To determine how 9; parasitica

 

debilitates chestnut trees requires an understanding of
host physiology, but unfortunately, little progress has
been made in this area. The wilting and senescence of
leaves distal to cankers implies that water transport is
impaired. The host suffers when leaves become unable to
photosynthesize, and thus it appears that xylem
dysfunction may initially be more devastating for the tree
than is phloem damage.

Two studies have pertained directly to xylem function
in blighted American chestnut trees. Bramble (1938)
reported reduced hydraulic conductance in diseased stems.
Tyloses were abundant in the outer xylem tissue of
cankered stems and were discussed as a factor impeding
flow. These conductivity results were corroborated by
Ewers et al. (in press). They concluded that xylem
dysfunction was strongly correlated with reductions in
hydraulic conductance and with death of the leaves distal

to cankers incited by virulent strains of 9; parasitica.

 

Similarly, xylem dysfunction has been implicated as a

major cause of leaf wilting in response to Cytospora

 

canker on Prunus spp. (Hampson and Sinclair, 1973;

Stanova, 1985).

III. OBJECTIVES
The objectives of the studies to be presented are:

(1) to test the effects of virulent and hypovirulent

14

strains of g; parasitica on water relations and xylem

 

function of American chestnut (Castanea dentata), (2) to
test the hypothesis that virulent cankers harbor the
fungus in xylem tissue whereas hypovirulent cankers do
not, and (3) to contribute to the existing definitions of
virulent and hypovirulent cankers based on xylem function,
xylem colonization by the fungus, and the presence or

absence of dsRNA.

CHAPTER TWO
THE EFFECTS OF VIRULENT AND HYPOVIRULENT STRAINS

OF CRYPHONECTRIA PARASITICA ON WATER RELATIONS
AND XYLEM FUNCTION IN AMERICAN CHESTNUT

15

16
A. INTRODUCTION

Cryphonectria parasitica (Murr.) Barr is the causal

 

agent of chestnut blight, a disease responsible for
decimating an estimated four billion American chestnut

(Castanea dentata (Marsh) Borkh.) trees in the Eastern

 

deciduous forest of North America. Upon the discovery of
the pathogen in 1904, research efforts were directed toward
describing the general biology and dissemination of the
pathogen (Murrill, 1906; Anderson, 1913; Rankin, 1914) and
the formation of cankers (Heald, 1913; Keefer, 1914). The
aim of numerous recent studies has been to clarify the
phenomenon of hypovirulence, a state of reduced pathogen
virulence that may be associated with the presence of
cytoplasmic double-stranded ribonucleic acid (dsRNA) in E;
parasitica (Fulbright, 1984). Hypovirulent strains of the
fungus grow more slowly and show abnormal culture morphology
compared to their normal virulent counterparts (Grente,
1965), and the cankers incited by hypovirulent strains are
notably different from those caused by virulent strains
(Ewers, et al., in press; Chapter 3).

In nature, cankers vary in morphlogy. Several studies
have attempted to relate canker morphology to pathogen
virulence (Turchetti, 1978; Jaynes and Elliston, 1982;
Kuhlman, 1982; Chapter 3). While no study has unequivocally
correlated pathogen virulence to canker morphology, it is
convenient to classify chestnut cankers into two general

categories. The virulent canker is characterized by bark

17
that is sunken and abundantly dotted with orange stromata.
Colonization is rapid and extends to the vascular cambium
which may be destroyed. Hypovirulent cankers include those
in which the surface is irregularly swollen, and the
rhytidome is cracked and sloughing due to alternating
regions of healthy and impaired cambium. In some
hypovirulent cankers colonization is superficial, being
confined to the bark which is cracked but intact and lacking
fructifications.

Chestnut blight is commonly considered a phloem disease
because of the symptomatic bark cankers. However, phloem
dysfunction has never been demonstrated to cause the leaf
wilting and senescence typical of this disease. In fact,
little progress has been made in the area of host
physiology. I am aware of only two studies that have
pertained specifically to xylem function in blighted
American chestnut trees. Bramble (1938) reported reduced
water flow through diseased stems compared to control stems.
Tyloses were abundant in the outer xylem tissue and were
discussed as a factor impeding the movement of water through
infected stems. The conductivity results were corroborated
by Ewers et al. (in press). They concluded that xylem
dysfunction is strongly correlated with the death of the
leaves distal to cankers incited by virulent but not

hypovirulent strains of g; parasitica. However, they did

 

not measure stomatal conductance (gs), transpiration (E) or

leaf water potential, nor did they isolate the fungus from

18
the xylem.
The purpose of this study was to test the effects of

virulent and hypovirulent strains of g; parasitia on the

 

water relations and xylem function of American chestnut
trees. In addition, girdling experiments were done to test
the possible influence of phloem dysfunction on water

relations.

B. MATERIALS AND METHODS

Plant material and field sites

 

Field studies were conducted at a grove (planted ca.
1910) near Grand Haven, MI and at an experimental plot
(planted 1982) located at Russ Forest Experiment Station,
Cass County, MI during the summers of 1987 and 1988.

The 1987 Grand Haven experiment involved three stems
arising from a supposed common rootstock. One stem was
afflicted with a virulent canker, a second stem bore a
hypovirulent canker near its base, and a third healthy stem
served as a control. In 1988, several virulent and
hypovirulent cankers and healthy stems were harvested from
trees throughout the Grand Haven grove.

In late May, 1988, sixteen 6-year-old trees at the Russ
Forest site were divided into groups of four trees. Each
group received one of the following treatments on two
branches per tree for a total of eight repetitions per
treatment: (1) inoculation with the virulent strain of Q;

parasitica, CL1-16, (2) inoculation with the hypovirulent

19
strain, GH2, (3) inoculation with potato dextrose agar
(control) or (4) girdling the branch to the vascular cambium
by removing a 1 cm wide band of bark. Girdles were cut five
weeks after the other branches were inoculated. By that
time, cankers had formed on the infected branches. The
girdles were intended to simulate the possible loss of
phloem tissue by cankers. Girdles were cut with a razor
blade and were coated with petroleum jelly to prevent
drying. Care was taken to avoid damaging the vascular
cambium. Inoculations were done by removing a small patch
of bark and smearing mycelia into the wounds which were
immediately covered with masking tape to prevent
desiccation. The site was visited weekly to monitor water
relations, canker development and leaf condition. This
experiment was modified slightly from a similar experiment
conducted in 1987 at the same site.

Water relations

 

Rates of 9s and B were monitored for three leaves per
stem at Grand Haven (1987) and for one leaf per branch at
Russ Forest (1987, 1988) using a Steady State Porometer
(Model LI 1600M, LI-COR, Inc., Lincoln, NE). Measurements
were taken at regular intervals on a diurnal basis. Leaf
water potentials were determined throughout the day for at
least one leaf per stem or branch at both sites with a
pressure chamber (PMS Instruments Co. Corvallis, OR).
Statistically significant differences among the treatments

were determined using Student's t-test.

20

Hydraulic conductance (Khl

Stems were harvested in the early morning the day after
final water relations data were collected. Cut ends were
quickly submerged to minimize the introduction of gas
bubbles into the vessels. Stems were cut to approximately
15 cm in length, and the cut surfaces were shaved smooth
with a razor blade. The stems were vacuum infiltrated at 50
cm Hg for one minute to remove bubbles at the cut surfaces.
Hydraulic conductance per unit stem length (Kh) was
determined by measuring the flow rate of 0.5 M oxalic acid
(pH 2) and dividing by the applied gravity gradient. For
the 1987 samples, Kh was first measured for a length of stem
that included regions distal and proximal to cankers or
treatments. The stems were then cut and Rh was obtained for
the cankered or treated portion and for the portions
immediately proximal and distal to the canker or girdle.
The stems were not trisected in 1988 because the 1987 data
indicated that most or all reductions in Kh were due to the
cankered portion of the stem.
Isolation of C. parasitica

To determine the extent of fungal infection into the
cankered stems, small chips of each xylem ring from various
locations in the canker were surface sterilized with a 20%
Chlorox solution and incubated on plates of potato dextrose
agar. The percentage of chips yielding E; parasitica was
recorded for each of the xylem rings of cankered regions.

For the natural cankers collected at Grand Haven, at least

21
40 chips from each of the three outermost xylem rings and at
least 20 chips from each subsequent inner ring and from
anomalous xylem tissue were plated. At least 40 chips from
each growth ring of eight stems per treatment were plated
for Russ Forest samples.

Leaf Specific Conductivity (LSC) and theortetical stem water

 

potential gradients

All leaves distal to cankers or treated regions on
experimental and control branches were collected and dried
to constant weight at 60°C. Before drying, the area of a
subsample was determined with a portable leaf area meter
(Model LI-3000, LAMBDA Instruments, Inc., Lincoln, NE) to
obtain a leaf area per dry weight relationship from which
leaf area could be calculated for all leaves harvested.
These values were used to obtain leaf specific conductivity
(LSC) which is the Kh of a stem divided.by the leaf area
supplied by that stem. The theoretical drop in pressure
along the length of a stem was determined from the
relationship dP/dx=E/LSC where dP/dx is the theoretical
pressure differential, E is maximum rate of
evapotranspiration, and LSC is leaf specific conductivity

(Tyree et al., 1983).

C . RESULTS
Water relations
The 1987 Grand Haven experiment revealed significantly

reduced 98 and E for leaves distal to cankers relative to

22
leaves on healthy stems. These differences were especially
noticable at midday and throughout the afternoon (Figure
la,b). Leaf water potentials of diseased stems were similar
to leaf water potentials of the control stem (Figure 1c),
and leaves on all of the stems were green at the time of the
study.

Differences in water relations among the Russ Forest
experimental groups were first detected at seven weeks after
inoculation in 1987 and at five weeks after inoculation in
1988. Rates of gs, E and leaf water potential were
generally greater in 1987 than in 1988, but relative
differences among the groups were similar both seasons. In
1988, final water relations data were collected at Russ
Forest seven weeks following inoculation (Figure 2), at
which time the experiment was terminated. Most striking
were the reduced 98 and E of leaves on stems infected with
CL1-16, the virulent strain, and similar reductions for
leaves on girdled stems (Figure 2a,b). The leaves of both
the CL1-16 and girdled treatments were pale green or yellow,
but their water potentials were not significantly different
from the control leaves at the 0.05 level of probability.
Hydraulic conductance (K51

Hydraulic conductance (Kh) was much lower in stems
afflicted with naturally occurring chestnut blight cankers
relative to healthy stems at Grand Haven (Table 1). Kh was
reduced more by virulent cankers than by hypovirulent

cankers. Among the treatment groups at Russ Forest, the

23
inoculated branches had Kh significantly lower than the
control branches (p < 0.05) (Table 2). The average Kh for
girdled stems was somewhat lower than that of the control
group, but the difference was not significant at the 0.05
level of probability.

Isolation of C. parasitica

 

The percentage of xylem tissue chips from natural

virulent cankers found to harbor Q; parasitica was about

 

double the percentage of infected chips from hypovirulent
cankers (Table 1). A much greater percentage of xylem chips
from cankers induced by CL1-16, the virulent strain, yielded
the pathogen than did the chips from cankers caused by the
hypovirulent strain, GH2 (Table 2). All of the control
stems were found to be free of the pathogen.
LSCs and theoretical water potential gradients

LSCs and theoretical stem water potential gradients
were calculated for the 1987 Grand Haven experiment (Table
3). LSCs were lower for infected branches compared to the
control. The largest drop in water potential, 1.274
MPa m'l, was calculated for the stem segment bearing the
virulent canker. The hypovirulent canker imposed a modest
drop in pressure, whereas the pressure differential along
the control stem was negligible.

Branches inoculated with CL1-16, the virulent strain of
‘Ql parasitica, had a mean LSC significantly lower than that
of control branches (p < 0.05) (Table 2). Branches infected

by the hypovirulent strain, GH2, and girdled branches had

24
mean LSCs somewhat lower than the control group mean, but
the differences were not significant at the 0.05 level of

probability.’

D. DISCUSSION

Yellowing and flagging of leaves occurs sometime after
stem infection. Though less diagnostic than canker
formation, leaf decline is a more advanced symptom of
chestnut blight. The principal reason for the wilting
symptom in many other diseases is occlusion of the water
transport pathway (Ayres, 1978). Since leaves are the site
of photosynthesis, leaf senescence and defoliation are the
ultimate cause of death in trees afflicted with chestnut
blight.

Colonization of the xylem tissue was correlated with a
decrease in Kh (Tables 1, 2). The xylem tissue of natural
virulent cankers and cankers induced by a virulent strain of
Q; parasitica were more heavily infected with the pathogen
and had lower Kh values than natural or induced hypovirulent
cankers. That Kh reduction is narrowly localized around
cankers has been demonstrated previously (Bramble, 1938;
Ewers et al., in press) and is supported by this study.

Xylem dysfunction has been cited as an important factor
in chestnut blight (Bramble, 1938; Zimmermann, 1983; Ewers,
et al., in press) and diseases of other ring-porous trees
(Hampson and Sinclair, 1973; Newbanks et al., 1983; Stanova,

1985; Kostka et al., 1986). Ring-porous species depend

25
almost exclusively on wide, superficially located earlywood
vessels for water transport. For example, Ewers and Ellmore
(1985) demonstrated that over 90% of the conduction of water
in elms is by the outermost xylem increment. The location
and size of these vessels makes them especially vulnerable
to invasion by pathogens which can result in dysfunction
(Zimmermann, 1983).

The mechanism of xylem dysfunction which causes
reductions in Kh is not known for chestnut blight. Bramble
(1938) reported abundant tyloses and gums in the outermost
xylem tissue of chestnut stems and speculated that these
occlusions were responsible for reduced rate of water flow.
Similarly, gums have been cited as the primary cause of

wilting and senescence of leaves of Cytospora-infected

 

Prunus (Hampson and Sinclair, 1973; Stanova, 1985).
Enzymatic degradation of vessel walls in advance of growing
hyphal tips has has been suggested to occur in American elms

infected by Ceratocystis ulmi (Newbanks et al., 1983). This

 

would permit air-seeding and embolism of vessels, rendering
them non-functional. Based on circumstantial evidence,
Zimmermann (1983) cited vapor blockage as the most direct
cause of leaf wilting and death in blighted American
chestnut.

Oxalic acid was identified at the margin of cankers and
displayed toxicity toward chestnut protoplasts lg glggg
(McCarroll and Thor, 1978b). Further lg ylggg studies

showed oxalic acid production by virulent but not

26

hypovirulent strains of C. parasitica (Havir and

 

Anagnostakis, 1983). The effect of oxalic acid on chestnut
xylem tissue has not been investigated. Recently, however,
Sperry and Tyree (1988) reported that perfusion of stems of

Acer saccharum with a solution of oxalic acid and calcium

 

increased the permability of intervessel pit membranes
thereby facilitating vessel embolism.

Decreased Rb and the colonizaton of xylem tissue by E;
parasitica corresponded to altered host water relations.
Unlike some vascular diseases, in which g8 and E are
increased (Creatura et al., 1981; Scheffer and Livingston,
1984), these parameters were significantly reduced on
cankered stems, especially those with virulent cankers.
This was most apparent at midday and throughout the
afternoon when trees have presumably transpired most of
their stored water and have begun to experience maximum
water stress.

Although Kh was only slightly affected by artificially
girdling branches to the vascular cambium, leaves of these
girdled stems were pale green to yellow and showed reduced
98 and E at the time of harvest (Figure 2a,b). It is not
clear whether these leaves were showing early signs of water
stress or if interruption of the phloem tissue is generally
deleterious to leaves.

It is interesting that leaf water potential did not
vary greatly among treatment groups at Russ Forest or in

natural conditions at Grand Haven. Only when leaves were

27
brown and no longer capable of gas exchange did leaf water
potential drop below the level of controls. Apparently,
yellowing leaves and leaves under water stress were able to
maintain near normal water potentials by closing stomata.
It is not clear how long leaves could exist in such a state,
but it would certainly depend on environmental factors such
as temperature and moisture as well as host vigor.

A severe drought during the course of the 1988 Russ
Forest experiment may be the reason that cankers developed
more quickly that year than in 1987, and undoubtedly caused
lower rates of 9s and E. However, the drought seemed to
affect the treatment groups uniformly, because similar
patterns in water relations were detected both seasons.

Despite recent advances (Dixon and Tyree, 1984)
measuring stem xylem water potential remains extremely
difficult. Leaf specific conductivity, the hydraulic
conductance per unit length of a stem divided by the amount
of leaf tissue distal to the stem segment, when combined
with transpiration data, is a convenient measurement for
predicting stem water potential differentials. Tyree et a1.
(1983) found that measured stem pressure gradients were
within about 3% of theoretical values. The theoretical
values obtained for the 1987 Grand Haven experiment (Table
3) showed that the virulent canker, distal to which leaves
had reduced transpiration and stomatal conductance, imposed
the greatest pressure gradient.

The data presented provide strong evidence that the

28

presence of E; parasitica in xylem tissue of American

 

chestnut results in xylem dysfunction. It appears that in
order to conserve water, the leaves close their stomata,
thereby reducing g8 and E. Hypovirulent cankers, both
naturally occurring and induced, had much less pronounced
effects of water relations than did virulent cankers.
Although the results of this study indicate that xylem
dysfunction may cause the death of the leaves in chestnut
blight, the possible involvement of phloem dysfunction
cannot be discounted. Artificially girdled stems, which
demonstrated no apparent xylem dysfunction, closely followed
the patterns of reduced 98 and E found in the cankered
stems. Thus, it remains difficult to distinguish between
effects generated by stem xylem dysfunction and those caused

by phloem dysfunction.

29

Table 1. Data collected from 12 natural cankers and 6
control stems at Grand Haven, 1986, 1987 and 1988. Mean
value 1 SE.

Kh % Xylem chips

(10'8m4MPa'1s'l) yielding E; parasitica

 

 

Virulent 1.6 11.3 84.5 17.3
Hypovirulent 5.7 12.6 42 120.8
Control 313.3 1158.8 0 +0

Table 2. Data collected from induced cankers, girdled stems
and control stems at Russ Forest, 1988. Mean values 1 SE;
n=8.

Stem Kh % xylem chips LSC
yielding _ _
(10'8m4MPa'ls'1) 91 parasitica (lo-emzs 1MPa 1)

 

 

CL1-l6 20.5 1 11.6 29.5 1 5.5 23.0 1 8.6
GH2 71 1 26.9 6.7 1 2.6 69.1 1 19.5
Girdled 90 1 13.4 0 1 0 ' 108.6 1 17.9
Control 125 1 14.4 0 1 0 137.3 1 20.3

30

Table 3. Leaf specific conductivities (LSC) and theoretical
stem water potential gradients (dP/dx) at Grand Haven, 1987.

 

Stem LSC dP/dx
(lo-amzs'lMPa-l) (MPa m’l)

Virulent 0.49 1.274

Hypovirulent 30 0.030

Control 230 0.006

31

Figure l. Stomatal conductance, transpiration and leaf
water potential at Grand Haven, 1987. Each point is the
mean of 3 measurements; bars indicate standard errors.

32

l—i thflent

o-eCeIm-el

 

 

200m

150—

monouodvdoo Hogan—no.5

 

H In NIE H088

nofiohmmqoue

   

32
35.68.61 6363

 

Time of day

Figure 1.

33

Figure 2. Stomatal conductance, transpiration and leaf water
potential at Russ Forest, 1988. Each point is the mean of 7
or 8 measurements; bars indicated standard errors.

34

 

 

 

 

 

 

 

200m
—— CLl-IG
b—l m
m
o
gfi 150~
Isl
d m
'U N
: 8 ' .1
U E 100
H
.:§ g
d
E g E3 504
4.3
‘ m
1 03
57 -— CIA-16
ratmm
eatflwhd
44 edramhd
H
d I
.9 w
E“? 3‘
in E
8 ._. 2
O "‘ a
E-*
13
0.1
—2.0-1 H CL1—16
. .rmcmi
eatwflbd
——1.8— H W
H
d
-H
“E:
36 —1.6—
O .
9%
5 —L44
4)
d
B
—1.24
—1.0—
r" ' T ‘ T ‘ l ' l ' T ‘ 118
a 8 10 12 14 16

Time of day
Figure 2.

CHAPTER THREE
CHARACTERIZATION OF THE CHESTNUT BLIGHT CANKER

AND THE LOCALIZATION AND ISOLATION OF THE
PATHOGEN CRYPHONECTRIA PARASITICA

35

36

A . INTRODUCTION

The earliest and most definitive symptom of chestnut
blight is canker formation on the stem, which occurs when
the fungal pathogen Cryphonectria parasitica (Murr.) Barr
invades via a wound or a branch scar. Canker morphology is
variable and has been described for American chestnut
(Bramble, 1936, Heald, 1913; Fulbright et al., 1983),
European chestnut (Biraghi, 1946; Bonifacio and Turchetti,
1973; Mittemperger, 1978), Chinese chestnut (Headland et
al., 1976; Jones et al., 1980) and Japanese chestnut
(Uchida, 1977). Several studies have attempted to relate
canker morphology to pathogen virulence (Turchetti, 1978;
Jaynes and Elliston, 1980; Kuhlman 1982). While no study
has unequivocally correlated pathogen virulence to canker
morphology, based on these works it is convenient to
classify chestnut cankers into two general categories. The
virulent canker is characterized by bark that is sunken and
abundantly dotted with orange stromata. Colonization is
rapid and extends to the vascular cambium which may be
destroyed. Hypovirulent cankers include those in which the
surface appears irregularly swollen, and bark is cracked and
sloughing due to alternating regions of healthy and impaired
cambium. In some hypovirulent cankers, colonization is
superficial, being confined to the bark which is cracked but
intact and lacking fructifications.

Chestnut blight is commonly considered to be a disease

of the phloem, and the descriptions of canker morphology

37
have been based largely on surface appearance of the

lesions. 9; parasitica has been reported to invade the

 

vascular cambium and outer xylem tissue of cankers (Keefer,
1914; Hebard, 1982; McCarroll and Thor, 1978a), but these
studies did not clearly document the invasion, nor did they
explore xylem colonization in terms of host xylem
development or water transport physiology. Cambial
disturbances can cause not only uneven phloem development,
giving rise to the alternating swollen and sunken regions on
canker surfaces, but also can alter xylem formation and
function (Ewers et al., in press).

Although the presence of Q1 parasitica in outer xylem

 

tissue has not been quantified, its inhabitation of these
regions is of paramount importance in the water economy of

the plant. 91 parasitica in xylem tissue of chestnut blight

 

cankers caused reductions in gs, E and stem Kh (Chapter II).
These effects were more pronounced in virulent cankers than
in hypovirulent cankers.

The goals of the present study were 1) to contribute
to the existing definitions of virulent and hypovirulent
cankers based on xylem function, xylem colonization by 91

parasitica, and the presence or absence of double-stranded

 

deoxyribonucleic acid (dsRNA) in the pathogen, and (2) to
test the hypothesis that virulent cankers harbor the fungus
in the xylem tissue, thereby rendering it non-functional,

whereas hypovirulent cankers do not.

38

B. MATERIALS AND METHODS
Plant material and inoculations

Twelve naturally occurring cankers were collected at
Grand Haven, MI during the summers of 1986, 1987 and 1988.
Additional studies were initiated in late May, 1988 at Russ
Forest Experimental Station, Cass County, MI. For the Russ
Forest experiments, eight branches ranging in diameter from
2.0cm to 3.5cm (approximately 3 or 4 years old) from four
different trees, were inoculated with the virulent strain of
Q1 parasitica CL1-16. Similarly, another set of branches
was treated with the hypovirulent (dsRNA-containing) strain
GH2 and eight control stems were treated with potato
dextrose agar. The inoculations were covered with masking
tape to prevent desiccation. The branches were monitored
weekly for canker development and leaf condition. Stems
were harvested for laboratory studies seven weeks after
inoculation. .

Canker morphology

 

Canker surfaces were observed, paying special attention
to bark characteristics such as depressions, swelling and
cracking and abundance of stromata. Each stem was viewed in
transverse section at several locations along the length of
the canker using a dissecting microscope. Abnormal xylem
appearance, such as irregularly shaped xylem rings, wood
discoloration, and tyloses-filled vessels was recorded. How
surface irregularities, such as sunken or swollen regions,

corresponded to anomalous growth of the underlying xylem

39

tissue was observed.
Demarcation of the conductive pathway

Upon harvesting, stems were kept under water to prevent
the introduction of gas bubbles into the vessels. The stems
were cut to size, and the cut surfaces were shaved smooth
with a razor blade and vacuum infiltrated at 50 cm Hg for
one minute to remove surface bubbles. Stem segments were
perfused with 0.5% safranin 0 to demarcate conductive
vessels, and the dye was chased with water to minimize
lateral diffusion. After drying, the stems were cut
transversely and viewed with a dissecting microscope.
Photographs were taken using a Nikon Microflex HFX II
photomicrographic attachment and fiber optics illumination
(Dolan-Jenner Fiber-Lite High Intensity Illuminator) with
35mm color film (Kodachrome 64 ASA).
Isolation of C. parasitica

To determine the extent of fungal infection into the
cankered stems, small chips of infected bark and xylem rings
from various locations in the canker were surface sterilized
with a 20% Chlorox solution and incubated on plates of
potato dextrose agar. The percentage of chips yielding 91
parasitica was recorded for the bark and each of the xylem
rings of cankered regions. For the Grand Haven samples, at
least 40 chips from the bark and each of the three outermost
xylem rings, and at least 20 chips from each subsequent
inner ring and from anomalous xylem tissue, were plated. At

least 40 chips from the bark and each growth ring of eight

40

branches per treatment were plated for the Russ Forest
samples.
Detection of dsRNA

Cultures obtained from bark samples were grown in
stationary liquid culture in Endothia-complete medium
(Puhalla and Anagnostakis, 1971) for 10-14 days. Molecules
of dsRNA were isolated as described by Morris and Dodds
(1979) and Fulbright et a1. (1983). The presence of dsRNA
was determined by means of polyacrylamide gel
electrophoresis (Morris and Dodds, 1979).

Localization of hyphae in xylem tissue

 

The methods of Morrell et a1. (1985) were modified
slightly for microscopic observations of hyphae in xylem
tissue. Cankers were sectioned transversely and
longitudinally. The 20 to 30 um thick sections were stained
with 0.5% safranin and then with fluorescein isothiocyanate-
coupled wheat germ agglutinin (FITC-WGA) (Vector
Laboratories Inc. Burlingame, CA) diluted 1:1000 in
phosphate buffered saline (PBS). The sections were
incubated in darkness for 15-20 minutes to prevent quenching
by fluorescent light, and then rinsed thoroughly with PBS
and placed in Bacto FA mounting medium (pH 9) (Difco
Laboratories, Detroit, MI) on glass slides. The sections
were viewed on a Leitz microscope using epifluorescence
optics (455 excitation, 510 split and 528 barrier filters
with a 50 watt mercury lamp). Photomicrographs were taken

using 35 mm color film (Kodachrome 400 ASA) with a Wild

41
photoautomat attachment.
C. RESULTS

Canker morphology

 

The naturally occurring cankers collected from Grand
Haven varied greatly in outward appearance (Figure 3a,b).
Virulent cankers (Figure 3a) were sunken lesions, the
borders of which were usually clearly delineated by orange
discoloration. The bark was generally intact and dotted
with stromata. Leaves distal to virulent cankers ranged in
appearance from dark green and healthy to yellow and
flagging, or the stem was defoliated. Often leaves were
healthy and green immediately proximal to the canker.

The rhytidome of hypovirulent cankers was bulging and
cracked, giving the canker a swollen silhouette (Figure 3b).
The bark was discolored, but the perimeters of cankers of
this type were poorly defined compared to virulent cankers.
Stromata were sparse or absent on the surface of
hypovirulent cankers, but when present were usually confined
to a small portion of the canker. Leaves distal to
hypovirulent cankers appeared healthy.

Transverse sections of the virulent and hypovirulent
cankers and control stems from Grand Haven showed dramatic
differences in xylem anatomy (Figure 4). The general
morphology of the xylem of virulent cankers looked similar
to that of healthy stems, but often the outermost growth
ring was discontinuous. The wood of the diseased stem was

darker than that of the healthy stem. Its vessels, even

42

those representing the current year's growth, were often
filled with tyloses. Unusual secondary growth was found in
hypovirulent cankers (Figure 4). Often the outermost xylem
ring was irregularly shaped being much wider in some regions
than in others. In other hypovirulent cankers there were
"islands" of anomalous xylem tissue external to the main
xylem ring axis, isolated by intervening callus and phloem
tissues. Though tissue of the main stem axis was discolored
in areas, resembling that found in virulent cankers, the
thickened xylem rings and ”islands" of xylem were usually
light in color, that is, healthy appearing, and vessels were
small and lacking tyloses. Anomalous secondary growth was
observed to some extent in all five of the naturally
occurring hypovirulent cankers that were carefully examined.

Inoculating stems with either a known virulent strain

or known hypovirulent strain of 91 parasitica produced

 

cankers of different morphologies. The cankers induced by
the virulent strain CL1-16 were sunken (Figure 5a), and
looked much like naturally occurring virulent cankers. The
cankers induced by GH2, the hypovirulent strain, were
swollen (Figure 5b), and bark was cracked, resembling
naturally occurring hypovirulent cankers. However, these
cankers lacked the anomalous secondary xylem that was

characteristic of hypovirulent cankers occurring in nature.

43

Demarcation of the conductive pathway

Functional xylem demarcated with safranin was located
primarily in the outermost growth ring in healthy stems
(Figure 6c). In the hypovirulent cankers, functional
vessels were irregularly distributed but never found in the
dark, discolored areas. Often it was the anomalous xylem
which harbored the only conductive vessels in the stem
section (Figure 6b, arrows). The virulent cankers had few
or no vessels that carried the dye (Figure 6a, arrow).
Transverse sections at 3 to 6 cm distal or proximal to
hypovirulent cankers showed dye patterns similar to those
seen in control stems. This was also usually the case with
virulent cankers; however, if the canker was advanced, as
indicated by brown or abscised leaves, the stem segment
distal to the canker often did not conduct dye, whereas the
proximal section, which bore green leaves, did pass dye.

Transverse sections of cankers induced by CLl-16 and
GH2 demonstrated the different effects these strains had on
xylem conductivity. The xylem of the CL1-16 cankers was
discolored. The few vessels that passed the red dye were
found where the wood was light in color (Figure 7a, arrows).
Discolored regions of xylem corresponded to sunken areas on
the canker surface. The effects of the hypovirulent strain
GH2 on the conductive pathway were variable. In some cases
there was no evidence that the vascular cambium or xylem
tissue had been disrupted, and the dye patterns were

indistinguishable from those found in control stems.

44
However, where the cambium appeared discontinuous, the
result was localized xylem dysfunction (Figure 7b, arrow).
The non-conducting vessels were located directly below the
point of inoculation. A similar situation was noted on the
opposite side of the stem, where vessels immediately below a
second inoculation point were non-functional.

Isolation of C. parasitica

 

The fungal isolation experiments showed that 91
parasitica colonized the xylem of naturally occurring
virulent cankers and cankers induced by the virulent strain
CL1-16 to a greater extent than it did naturally occurring
hypovirulent cankers and cankers induced by GH2, a
hypovirulent strain (Tables 4 and 5). For all canker types,
colonization decreased going inward toward the pith. No
fungus was isolated from the anomalous "islands" of xylem
tissue of hypovirulent cankers (Table 4). The percentage of
xylem chips harboring the virulent strain CL1-16 was roughly
twice that of the hypovirulent strain GH2 for artificially
induced cankers (Table 5).

Detection of dsRNA
No correlation was found between canker morphology and

the presence or absence of dsRNA in 91 parasitica isolated

 

from the bark of natural cankers. Of seven virulent cankers
tested, four were found to harbor dsRNA-containing strains.
Two of the five hypovirulent cankers screened yielded dsRNA-
containing strains. The isolates from CL1-16-induced

cankers contained no dsRNA. Isolations from GH2-induced

45

cankers yielded g1 parasitica with the banding pattern of
GH2.
Localization of hyphae in xylem tissue

Hyphae were observed via fluorescence microscopy in
the vessels and rays of advanced virulent and hypovirulent
cankers and from CL1-16 induced cankers (Figure 8a,b),
though colonization was typically not as extensive as shown.
Since the FITC-WGA label is specific for N-acetyl
glucosamine, it will adhere to any chitinous fungus, and it
is not certain that all hyphae viewed were Q1 parasitica.
The amount of visible hyphae invading xylem tissue was not
quantified but seemed to be greatest in advanced virulent
cankers, especially in the outermost xylem tissue with
deeper invasion via the vascular rays.
D. DISCUSSION

Previous workers have described chestnut blight as a
disease of the phloem, and cankers have been characterized
based on observations at the bark surface and via
histochemical methods (Heald, 1913; Keefer, 1914; Hebard,
1982). The surface descriptions of naturally occurring
cankers of various morphologies summarized in the present
study are consistent with those of Fulbright et a1. (1983)
who first observed abnormal canker morphology on trees that
appeared to be recovering from chestnut blight at the Grand
Haven site.

One aim of this work was to better define virulent and

hypovirulent cankers in terms of xylem appearance. The

46
xylem of virulent cankers was discolored relative to control
stems. Shigo (1985) cited the accumulation of phenolic
compounds as the cause of discoloration in xylem tissue
injured by pathogens. In chestnut blight, the presence of
gums and tyloses in vessels may also contribute to wood
discoloration (Rumbold, 1916; Bramble, 1938).

Viewing transverse sections of hypovirulent cankers
revealed anomalous xylem formation external to a sometimes
discolored main xylem ring axis. This phenomenon was
observed in blighted American chestnut (Ewers et al., in
press; Shigo, 1983) but has not been previously evaluated in
terms of xylem anatomy and physiology. Fulbright et a1.
(1983) speculated that certain abnormally swollen cankers
were at one time normal, sunken lesions that partially
healed by production of wound tissue and secondary vascular
tissue. This would explain darkened xylem tissue that
corresponded to sunken areas on the canker surface, and the
localized destruction of the vascular cambium which would.
disrupt the formation of secondary xylem tissue.

Aloni and Zimmermann (1984) reported that partial

girdles in stems of Acer rubra resulted in the production of

 

many narrow vessels. They suggested that damage to the
vascular cambium could obstruct the basipetal flow of auxin
thereby altering xylem differentiation. A similar situation
may occur in hypovirulent infections of chestnut. The
unusual xylem tissue seen in hypovirulent cankers had

vessels of narrrow diameter, but the xylem rings tended to

47
be wide; that is, great amounts of tissue arose in a single
season.

It is interesting to note that no g1 parasitica was

 

isolated from anomalous xylem tissue (Table 4), and that
these regions often harbored the only conductive tissues in
the cankered portion of the stem. This lends support to the
hypothesis that xylem tissue containing 91 parasitica is
non-functional.

Increased vascularization which may compensate for
xylem tissue rendered non-functional by pathogens was
reported in Fusarium-infected tomatoes (Scheffer and Walker,

1954) and Verticillium-infected hops (Talboys, 1958). The

 

anomalous xylem tissue observed in the present study may
indeed be "compensatory xylem" because in hypovirulent
cankers it conducted dye, whereas dark, discolored wood of
the main stem axis did not (Figure 6b).

The fact that dye patterns at regions distal and
proximal to the cankers were similar to those of healthy
stems indicates that xylem damage is localized in chestnut
blight cankers as has been previously reported (Bramble,
1938; Ewers et al., in press). After the stem distal to a
virulent canker loses its leaves, its vessels apparently
become vapor-blocked, tyloses-filled, and unable to conduct
dye. Thus the distal portion of the stem is rendered non-
functional even though canker damage was initially quite
localized.

It is difficult to compare naturally occurring cankers

48
with cankers induced by artificial means on relatively small
stems. Judging from the extensive colonization of xylem
tissue (Table 4), it seems that the natural cankers had been
developing for several seasons. Therefore, it is
understandable that the 7-week-old cankers from Russ Forest
did not display the unusual patterns of vascular tissue
noted in the Grand Haven cankers. Even so, stems infected
with the virulent strain CL1-l6 formed sunken lesions
(Figure 5a) which in cross-section showed discolored non-
conductive xylem tissue (Figure 7a) just as in naturally
occurring virulent cankers (Figures 4a, 6a).

These variations in canker morphology may be in part
due to differences in pathogen growth rate. The rapidly
growing virulent strain can perhaps overcome host defense
mechanisms, whereas the host is able to keep pace with the
more slowly growing hypovirulent pathogen by producing
callus, wound periderm and secondary vascular tissues.

My initial hypothesis that virulent cankers harbor the
pathogen in xylem tissue, whereas hypovirulent cankers never
do, is discredited by the isolation data (Tables 4, 5).
However, Table 4 shows that xylem colonization is more

extensive in the xylem of virulent than hypovirulent

 

cankers. That the percentage of xylem chips yielding 91

parasitica in CL1-16 infected stems was double that in GH2

 

infected stems after seven weeks (Table 5), may demonstrate
the differences in pathogenicity between virulent and

hypovirulent strains.

49
In agreement with the findings of Kuhlman (1982), I
detected no correlation between canker morphology and the
presence or absence of dsRNA in naturally occurring cankers.
Most likely the young, artificially induced cankers were

colonized by just one strain of E; parasitica, whereas the

 

natural cankers, which were older and had been exposed to
inocula for several seasons, might have harbored both
virulent and hypovirulent strains. Thus, any conclusion
relating canker morphology to the presence or absence of
dsRNA extracted from the canker would be untenable.

A major drawback in using the fluorescent label
technique described is lack of specificity. Since FITC-WGA
is specific for N-acetylglucosamine, it will adhere to any
chitinous hyphae that may inhabit chestnut. However, since

the only fungus isolated from the cankers was 91 parasitica,

 

it was assumed that it was the dominant if not sole fungal
species present.

Xylem parenchyma rays may serve as an avenue for the
pathogen to invade inner tissues since hyphae were more
abundant in rays than in vessels. Hyphae were not usually
visible 3g ggggg in xylem tissue, even in the outermost ring
of advanced virulent cankers. Figure 8 illustrates an
extreme situation. Therefore, it is unlikely that vessels
are occluded by the pathogen itself. More microscopic work
would be required to ascertain the mechanism of dysfunction.

The mechanism by which xylem dysfunction occurs in

diseased plants has been the debate of numerous pathologists

50
(Smith, 1899; Scheffer, 1953; Powers, 1954; Van Alfen and
Turner, 1975; Newbanks et al., 1983) but has not been
determined for chestnut blight. However, it is now clear
that physical invasion of the xylem is closely correlated
with xylem dysfunction.

Based on circumstantial evidence, Zimmermann (1983)
proposed that enzymatic degradation of the crucial outermost
vessels of a ring-porous tree could result in air-seeding
and embolism. Ring-porous trees, such as chestnut, which
depend almost exclusively upon superficially located
earlywood vessels for their water supply, are especially
vulnerable to this type of xylem interruption. Thus,
Zimmermann cited vapor-blockage as the most direct cause of
leaf wilting and death in blighted American chestnut trees.
If vessels become vapor-blocked, then xylem parenchyma cells
can expand through pits into the vessel lumens forming
tyloses.

The virulent cankers observed in this study contained
abundant tyloses in the current season's vessels. This is
unusual for vessels that have been functional for less than
one season. Tyloses have been suggested as a factor
reducing water movement in stems afflicted with chestnut
blight (Bramble, 1938) and in other vascular diseases of
ring-porous trees (Sinclair and Campana, 1978; Beckman et
al., 1953).

This study has shown that virulent and hypovirulent

cankers differ not only in outward appearance, but also in

51
terms of xylem anatomy, xylem function and colonization of
xylem tissue. Advanced hypovirulent cankers have regions of
anomalous xylem tissue that is functional and devoid of 91

parasitica. The xylem of virulent cankers harbors the

 

pathogen, is largely non-conductive and lacks anomalous
xylem development. These differences in the water
conducting tissue are important to host physiology and
probably are a factor in the reported survival and recovery

of American chestnut in Michigan (Fulbright et al., 1983).

52

Table 4. Percentage1 of xylem chips yielding 91 parasitica,
Grand Haven, 1987. Ring 1 is outermost; Ax=anomalous xylem.

Region of Stem

 

 

Stem Bark 1 2 3 4 5 6 7 AX
Virulent 100 98 88 73 68 67 50 50 --
Hypovirulent 96 76 60 63 29 67 -- -- 0
Control 0 0 0 0 0 0 0 -- --

lPercentage based on 40 chips for bark and rings 1, 2 and 3;
20 chips for rings 4-7 and AX for each of 18 stems or
cankers.

Table 5. Percentage1 of xylem chips yielding
g1 parasitica, Russ Forest, 1988. Ring 1
1s outermost.

 

Region_of stem

 

Treatment Bark l 2 3 4
CL1-16 92 49 42 6 10
GH2 79 24 23 2 0
Control 0 0 0 0 0

1Percentage based on 40 chips per stem region from
each of 8 stems per treatment.

53

 

Figure 3. Naturally occurring virulent (A) and hypovirulent
(B) cankers from Grand Haven.

54

 

Figure 4. Transverse section of healthy stems (A),
naturally occurring virulent cankers (B) and naturally
occurring hypovirulent cankers (C).

55

 

Figure 5. Cankers induced by the virulent strain CL1-16 (A)
and the hypovirulent strain GH2 (B).

56

Figure 6. Transverse sections revealing functional vessels
(marked with red dye) in a healthy stem (A), a hypovirulent
canker (B) and a virulent canker (C).

 

Figure 6.

S8

 

Figure 7. Transverse sections revealing functional vessels
(red dye) in cankers induced by the virulent strain CL1-16
(A) and the hypovirulent strain GH2 (B).

59

 

Figure 8. Longitudinal section of chestnut wood infected
with 91 parasitica. 430x (A); 1070x (B).

SUMMARY AND CONCLUSIONS

60

61

The results of the studies presented in this thesis are
summarized (Table 6), and a model for the steps leading to
canker formation and the subsequent death or survival of
branches of American chestnut is proposed (Figure 9). The

causal pathogen of chestnut blight, Cryphonectria

 

parasitica, invades the xylem causing reductions in Kh' gs

 

and E. Xylem dysfunction can be attributed as one of the

major causes of death of leaves distal to cankers.

Hypovirulent strains invade the xylem to a lesser extent

than virulent strains. In naturally occurring hypovirulent

cankers, anomalous "compensatory" xylem is produced and
appears to be important for conducting water to the distal
leaves. From the results of these studies I conclude:

1. Rates of g and E are reduced for leaves distal to
virulent cfiestnut blight cankers and mechanically
girdled stems.

2. Chestnut blight cankers have Kh lower than control
stems; this is more pronounced in virulent cankers
than in hypovirulent cankers.

3. Cryphonectria pargsitica is more abundant, i.e., can

be isolated more frequently, in the xylem of virulent
cankers than in hypovirulent cankers.

 

4. Advanced hypovirulent cankers have anomalous xylem
that is conductive, whereas virulent cankers lack
such tissue.

5. 91 parasitica can be isolated from non-conductive
xylem tissue but not from conductive, anomalous xylem
tissue of naturally occurring hypovirulent cankers.

 

6. There is no correlation between canker morphology and
the presence or absence of dsRNA isolated from
natually occurring cankers.

.>_co nopneon ce>oI ucoco hoop Eocw oueom
.oaop away wanton nuam >n ooze—.0m even Axov co>eI ucetuP

O o e O O a
km_ onw _ am No on 0 an «me o A.Ieaz.t «sous—V

um; ouoco><

mu. .m.m 0.1R R.m m.o~ 0.. A 6 ea: 5 opv
pl. —l£ V OI
x oaaLo>(

III + o_nn.to> I e_oosce> ousuvnetem 4w Dean—on.
to acoacou «2156

9. I I + + + Eo_>x c.
oeu._ennv> novareotem 4m

I + + + + Elp>x EOLL
Dene—on? ne'ernetem 4w

I + + + + Eco—ca Eocm
noun—on, euvuvnetem 4w

I I + I I 50—): nae—0Eoc<
>£u_eoz >£u_eo£ >zu_ooc o_ne.to> o_nevco> cowuvucou macs

III co__oxn cop—03¢ coxcan cexcan eaocn Lexcnu
may“. ........... wa...fl...m::_uw..u ......... . ..........._:........1..:--._....H.....fl_ ................................
.ocucou Loxcou «co—3L—>0Q>I Loxcou «co_3L*>

.mom— Ocn hump .uQOLou 003“
pco moan uce boa. .mmmp .co>oI pcoto Scum nopaeou co peace—.00 euou so >LoEEJm .o 0.96»

63

Figure 9. Model for the steps leading to canker formation
and the death or survival of branches of Castanea dentata
infected by Cryphonectria parasitica. Broken lines
represent less clearly defined steps.

 

64

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Cryphonectria parasitica
[Virulent strain] LHypovir ulent strain]
Castanea dentam
Infection of \] Infection of ]
hloem tissue phloem tissue \
Sunken ] / Swollen.
. , h ovirulent
virulent Infection of Infection of 4 ygk r
canker vascular vascular _,
* cambium cambium L
Maj or 1053 of ] £ ‘ i Minor loss
phloem tissue: of hl
cambium Infection of “I nfec t ion of tissgefglnmbium
destroyed xylem tissue xylem tissue not destroyed
. V ,
I ; orma on o
' . anomalous

 

 

[Vessel embolism] [Vessel embolisfl xylem tissue »

‘ r

 

 

 

 

‘ I
‘ |
. Q

 

 

 

 

 

 

 

 

 

 

 

 

  
     
   

  

 

 

 

 

 

 

 

 

 

 

 

I
I I
I I
' I
I I
‘ I
‘ I
‘ :
: ‘ ‘ Functional, :
I Ll‘ylosis formation] b‘ylosis formation] anomalous :
: ‘f ‘ xylem tissue .
' : : compensates I
: * ‘ r I , for reduced Kh :
: ]Xylem dysfunction] ]Xylem dysfunction] gillgam stem :
I $ 1
' 7 : i 7 * I
I
: main stem axis reduction less I
' 7 than in virulent ‘
: I Reduoedg s and E I canker :
I $ , .
' I
' I
I I Yellow. wilted lfmeen, turgid I I
1 leaves . eaves :
' I I :
I
' ' I
' """ ] Branch death I Branch survival] ' -‘

 

 

 

Figure 9.

LIST 01" REFERENCES

65

66
Agrios, G. N. 1988. Plant pathology. Academic Press, New York.

Aloni, R., and M. H. Zimmermann. 1984. Length, width, and
pattern of regenerative vessels along strips of vascular
tissue. Bot. Gaz. 145: 50-54.

Anderson, P. J. 1913. The morphology and life history of the
chestnut blight fungus. Pa. Chestnut Tree Blight Comm.
Bull. 7: 1-43

Ayres, P. G. 1978. Water relations of diseased plants. In:
Kozlowski, T. T. (ed.) Water deficits and plant growth.
Vol. 5. Academic Press, New York, London, pp. 1-60.

Barr, M. E. 1978. The Diaporthales in North America with
emphasis on Gnomonia and its segregates. Mycologia Mem.
7. J. Cramer, Lehre, Germany. 232 pp.

Bazzigher, G., and G. Miller. 1982. Chestnut blight resistance
breeding in Switzerland. North. Nut Grow. Assoc. Annu.

Beckman, C. H. 1987. The Nature of Wilt Diseases. American
Phytopath. Soc. Press. 175 pp.

Beckman, C. H., J. E. Kuntz, A. J. Riker, and J. G. Berbee.
1953. Host responses associated with the development of
oak wilt. PhytopathologY. 43: 448-454.

Biraghi, A. 1946. Il cancro del castagno causato da Endothia
parasitica. Ital. Agric. 7: 406-412.

Biraghi, A. 1950a. Carartteri di resistenze in Castanea sativa
nei confronti di gndothia parasitica. Boll. Stn. Patol.
Veg. Rome 7: 161-171.

 

Biraghi, A. 1950b. La distribuzione del cancro del castagno in
Italia. Ital. For. Mont. 5: 18-21.

Bishop, C. D., and R. M. Cooper. 1984. Ultrastructure of
vascular colonization by fungal wilt pathogens. II.
Invasion of resistant cultivars. Physiol. Plant Path. 24:
277-289.

Bonifacio, A., and T. Turchetti. 1973. Differenze morfologiche
e fisiologiche in isolati di Endothia parasitica (Murr.)
And. Annali. Accad. Ital. Sci. For., Florence. 22: 111-
131.

 

Bramble, W. C. 1936. Reactions of chestnut bark to invasion by
Endothia parasitica. Am. J. Bot. 23: 89-99.

 

Bramble, W. C. 1938. Effect of Endothia parasitica on
conduction. Am. J. Bot. 25: 61-65.

 

67

Brandes, E. W. 1919. Banana wilt. Phytopathology 338-389.

Creatura, P. J., G. R. Safir, R. P. Scheffer, T. D. Sharkey.
1981. Effects of Cephalosporium gramium and a toxic
metabolite on stomatal conductance of wheat. Physiol.
Plant Path. 19: 313-323.

Dimond, A. E. 1970. Biophysics and biochemistry of the
vascular wilt syndrome. Annu. Rev. Phytopathol. 8: 301-
316.

Dixon, M. A., and M. T. Tyree. 1984. A new stem hygrometer,
corrected for temperature gradients and calibrated against
the pressure bomb. Plant, Cell and Envir. 7: 693-697.

Duniway, J. M. 1971. Resistance to water movement in tomato
plants infected with Fusarium. Nature 230: 252-253.

Duniway, J. M. 1973. Pathogen-induced changes in host water
relations. Phytopathology 63: 458-472.

Duniway, J. M. 1975. Water relations in safflower during
wilting induced by Phytophthora root rot. Phytopathology
65: 886-891.

 

Duniway, J. M., and R. O. Slayter. 1971. Gas exchange studies
on the transpiration and photosynthesis of tomato leaves
affected by Fusarium oxyspgrum f. sp. lycopersici.
Phytopathology: 61: 1377-1381.

Ellmore, G. S., and F. W. Ewers. 1986. Fluid flow in the
outermost xylem increment of a ring-porous tree, Ulmus
americana. Am. J. Bot. 73: 1771-1774.

 

Elliston, J. E. 1987. A comparison of five similar cytoplasmic
hypovirulent strains of Endothia parasitica from the
southern Appalachian Mountains. Phytopathology 77: 467-
472.

 

Ewers, F. W., P. S. McManus, A. Goldman, R. Gucci, and D.
W.Fu1bright. The effect of virulent and hypovirulent
strains of Endothia parasitica on hydraulic conductance
in American chestnut. Can. J. Bot., in press.

Fulbright, D. W. 1984. Effect of eliminating dsRNA in
hypovirulent Endothia parasitica. Phytopathology 74: 722-
724.

 

Fulbright, D. W., W. H. Weidlich, X. z. Haufler, C. S. Thomas,
and C. P. Paul. 1983. Chestnut blight and recovering
American chestnut trees in Michigan. Can. J. Bot. 61:
3164-3171.

68

Gottlieb, D. 1944. The mechanism of wilting caused by Fusarium
bulbigenum var. lycopersici. Phytopathology 34: 41-59.

Grente, J. 1965. Les formes hypovirulentes d'Endothia
parasitica et les spoirs de lutte contre 1e chancre du
chataignier. C. R. Seances Acad. Agric. Fr. 51: 1033-1037.

Hall, R., A. Ali, and L. V. Busch. 1975. Verticillium wilt of
Chrysanthemum: development of wilt in relation to leaf
diffusive resistance and vascular conductivity. Can. J.
Bot. 53: 1200-1205.

 

Hall, R., and L. V. Busch. 1971. Verticillium wilt of
Chrysanthemum: colonization of leaves in relation to
symptom development. Can. J. Bot. 49: 181-185.

Hampson, M. C., and W. A. Sinclair. 1973. Xylem dysfunction in
peach caused by Cytospora leucostoma. Phytopathology 63:
676-681.

 

Hardegger, E., W. Rieder, A. Walser, and P. Kugler. 1966.
Welkstoff und Antibiotika 35. Konstitution des
Diaporthins und Synthese der Diaporthinsaure. Helv. Chim.
Acta. 49: 1283-1290.

Havir, E. A., and S. L. Anagnostakis. 1983. Oxalate production
by virulent but not by hypovirulent strains of Endothia
parasitica. Physiol. Plant Pathol. 23:369-376.

 

Headland, J. R., G. J. Griffin, R. J. Stipes, and J. R. Elkins.
1976. Severity of natural Endothia parasitgg infection of
Chinese chestnut. Plant Dis. Reptr. 60: 426-429.

Heald, F. D. 1913. The symptoms of the chestnut tree blight
and a brief description of the blight fungus. Pa.
Chestnut Tree Blight Comm. Bull. 5. 15 pp.

Hebard, F. V. 1982. Biology of virulent and hypovirulent
Endothia parasitica on American chestnut (Castanea
dentata). Ph.D. thesis, Virginia Polytechnic Institute
and State University, Blacksburg. 295 pp.

Hebard, F. V., G. J. Griffin, and J. R. Elkins. 1984.
Developmental histopathology of cankers incited by
hypovirulent and virulent isolates of Endothia parasitica
on susceptible and resistant chestnut trees.
Phytopathology 74: 140-149.

Helms, J. A., F. W. Cobb, Jr., and H. S. Whitney. 1971. Effect
of infection by Verticladiella wagenerii on the physiology
of Pinus ponderosa. Phytopathology, 61: 920-925.

 

 

Jaynes, R. A., and J. E. Elliston. 1982. Hypovirulent isolates
of Endothia parasitica associated with large American

69
chestnut trees. Plant Dis. 66: 769-772.

Jones, C. G., G. J. Griffin, and J. R. Elkins 1980.
Association of climatic stress with blight on Chinese
chestnut in the eastern United States. Plant Dis. 64:
1001-1004.

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

Kitajima, K. 1927. Studies on the chestnut blight fungus and
related fungi. For. Exp. Sta. Rep. 27: 29-81.

Kostka, S. J., T. A. Tattar, and J. L. Sherald. 1986. Elm leaf
scorch: abnormal physiology in American elms infected
with fastidious, xylem-inhabiting bacteria. Can. J. For.
Res. 16: 1088-1091.

Kuhlman, E. G. 1982. Summary of relationships among swollen
superficial cankers, survival of American chestnut trees,
and hypovirulence in Endothia parasitica at Southeastern
Forest Experiment Station. Pages 24-34 in:.Proc. U.S.
For. Serv. Am. Chestnut Cooperators' Meet. H.C. Smith and
W.L. MacDonald, eds. West Virginia University Books,
Morgantown. 229 pp.

Linford, M. B. 1931. Studies of pathogenesis and resistance in
pea wilt caused by Fusarium orthoceros var. pisi.
Phytopathology 21: 797-826.

MacHardy, W. E., and C. H. Beckman. 1973. Water relations in
American elm infected with Ceratocystis ulmi.
Phytopathology 63: 98-103.

MacHardy, W. E., L. V. Busch, and R. Hall. 1976. Verticillium
wilt of chrysanthemum: quantitative relationship between
increased stomatal resistance and local vascular
disfunction preceeding wilt. Can. J. Bot. 54: 1023-1034.

MacHardy, W. E., R. Hall, and L. V. Busch. 1974. Verticillium
wilt of chrysanthemum: Relative water content and protein,
R.N.A., and chlorophyll levels in leaves in relation to
visible wilt symptoms. Can. J. Bot. 52: 49-54.

Massee, G. 1895. A disease of tomatoes. Gard. Chron. Ser.
17:707-708

McCarroll, D. R., and E. Thor. 1978a. Death of a chestnut:
The host pathogen interaction. Pages 25-26 in: Proc. Am.
Chestnut Symp. W. L. MacDonald, F. C. Cech, J. Luchok, and
H. C. Smith, eds. West Virginia University Books,
Morgantown. 122 pp.

McCarroll, D. R., and E. Thor. 1978b. The role of oxalic acid

70

in the pathogenesis of Endothia parasitica. Pages 60- 63

in: Proc. Am. Chestnut Symp. W. L. MacDonald, F. C. Cech,
J. Luchok, and H. C. Smith, eds. West Virginia University
Books, Morgantown. 122 pp.

 

Melching, J. 8., and W. A. Sinclair. 1975. Hydraulic
conductivity of stem internodes related to resistance of
American elms to Ceratocystis ulmi. Phytopathology 65:
645-647.

Merkel, H. W. 1905. A deadly fungus on the American chestnut.
N. Y. 2001. Soc. 10th Ann. Rep. pp. 97-103.

Misaghi, I. J., J. E. Devay, and J. M. Duniway. 1978.
Relationship between occlusion of xylem elements and
disease symptoms in leaves of cotton plants infected with
Verticillium dahliae. Can. J. Bot. 56: 339-342.

 

Mittempherger, L. 1978. The present status of chestnut blight
in Italy. Pages 34-37 in: W. L. MacDonald, F. C. Cech, J.
Luchok, and C. Smith, eds., Proceedings of the American
Chestnut Symposium. Morgantown, West Virginia.

Morrell, J. J., D. G. Gibson, and R. L. Krahmer. 1985. Effect
of fluorescent-labeled lectins on visualization of decay
fungi in wood sections. Phytopathology 75: 329-332.

Morris, T. J., and J. A. Dodds. 1979. Isolation and analysis of
double-stranded RNA from virus-infected plant and fungal
tissue. Phytopathology 69: 854-858.

Murrill, W. A. 1906. A serious chestnut disease. Journ. New
York Bot. Gard. 7: 143-153.

Newbanks, D., A. Bosch, and M. H. Zimmermann. 1983. Evidence
for xylem dysfuction by embolization in Dutch elm disease.
Phytopathology 73: 1060-1063.

Paul, C. P., and D. W. Fulbright. 1988. Double-stranded RNA
molecules from Michigan hypovirulent isolates of Endothia
parasitica vary in size and sequence homology.
Phytopathology, 78: 751-755.

 

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

 

 

Powers, H. R., Jr. 1954. The mechanism of wilting in tobacco
plants affected by black shank. Phytopathology, 44: 513-
521.

Pringle, R. B., and R. P. Scheffer. 1964. Host-specific plant
toxins. Ann. Rev. Plant Pathol. 2: 133-156.

71

Rankin, W. H. 1914. Field studies on the Endothia canker of
chestnut in New York State. Phytopathology 4: 233-259.

Roane, M. R., G. J. Griffin, and J. R. Elkins. 1986. Chestnut
blight, other Endothia diseases, and the genus Endothia.
APS Monograph Series, APS Press, St. Paul, MN.

Rumbold, C. 1920. Effect on chestnuts of substances injected
into their trunks. Am. J. Bot. 2: 45-56.

Russin, J. S., and L. Shain. 1984. Initiation and development
of cankers caused by virulent and cytoplasmic hypovirulent
isolates of the chestnut blight fungus. Can. J. Bot. 62:
2660-2664.

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

Scheffer, R. P., and J. C. Walker. 1953. The physiology of
Fusarium wilt of tomato. Phytopathology 43: 116-125.

Shigo, A. L. 1985. Compartmentalization of decay in trees.
Scientific American 252: 96-103.

Shigo, A. L. 1983. Tree defects: a photo guide. USDA Forest
Service. General Technical Report, NE—82. 167 pp.

Sinclair, W. A., and R. J. Campana. 1978. Dutch elm disease:
perspectives after 60 years. Cornell Univ. Ag. Expt. St.
Bull. 52 pp.

Smith, E. F. 1899. Wilt disease of cotton, cowpea, and
watermelon. USDA Bull. 17: 1-50.

Sperry, J. S., and M. T. Tyree. 1988. Mechanism of water stress
induced xylem embolism. Am. J. Bot. 75: 49 (abstr).

Stanova, M. 1985. Structural changes in fruit tree tissues
infected with the fungus Cytospora cincta Sacc. Biologia
(Bratislava) 40: 867-871.

Talboys, P. W. 1958. Association of tylosis and hyperplasia of
the xylem with vascular invasion of the hop by
Verticillium albo-atrum. Trans. Br. Mycol. Soc. 41: 249-
260.

Talboys, P. W. 1978. Dysfunction of the water system. In:
Horsefall, J. G. and E. B. Cowling (eds.). Plant disease:
an advanced treatise. Vol 3. Academic Press, New York,
London. pp. 141-162.

Tartaglia, J., C. P. Paul, D. W. Fulbright and D. L. Nuss.
1986. Structural properties of doubled-stranded RNAs

72

associated with biological control of chestnut blight
fungus. Proc. Natl. Acad. Sci. 83: 9109-9113.

Turchetti, T. 1978. Some observations on the "hypovirulence"
of chestnut blight in Italy. Pages 92-94 in: W.L.
MacDonald, F.C. Cech, J. Luchok, and C. Smith, eds.,
Proceedings of the American chestnut symposium.
Morgantown, West Virginia.

Turner, N. C., and A. Graniti. 1969. Fusicoccin: a fungal toxin
that opens stomates. Nature 223: 1070-1071.

Tyree, M. T., M. E. D. Graham, K. E. Cooper, and L. J. Bazos.
The hydraulic architecture of Thuja occidentalis. Can. J.
Bot. 61: 2105-2111.

 

Uchida, K. 1977. Studies on Endothia canker of Japanese
chestnut trees caused by Endothia parasitica (Murrill)
P.J. et H.W. Anderson. Bulletin of the Ibaraki-ken
Horticultural Experiment Station, Special Issue No. 4. 65
pp. (In Japanese).

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

Van Alfen, N. K., and V. Allard-Turner. 1979. Susceptibility of
plants to vascular disruption by macromoleules. Plant
Physiol. 63: 1072-1075.

Van Alfen, N. R., 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.

Van Alfen, N. R., and N. C. Turner. 1975. Influence of a
Ceratocystis ulmi toxin on water relations of elm (Ulmus
americana). Plant Physiol. 55: 312-316.

 

 

VanderMolen, G. E., C. H. Beckman, and E. Rodehorst. 1977.
Vascular gelation: a general response phenomenon following
infection. Physiol. Plant Path. 11: 95-100.

Zimmermann, M. H. 1983. Xylem structure and the ascent of
sap. Springer-Verlag, Berlin. 143 pp.

"71111011111111]