I
23113:

 

 

 

5'Isa‘153‘i
—"‘ av
3“” fl " State This is to certify that the
{MIN :rSIty thesis entitled

 

THE IDENTIFICATION, DISTRIBUTION AND SPREAD OF
DOGWOOD ANTHRACNOSE IN MICHIGAN

presented by
ZACHARY JAMES BLANKENHEIM

has been accepted towards fulfillment
of the requirements for the

MS. degree in PLANT PATHOLOGY

I] /~// ‘l/IJ////1

'1 'JorPfi' e rs Sana?

Date

MSU is an Affirmative Action/Equal Opportunity Institution

 

PLACE IN RETURN BOX to remove this checkout from your record.

TO AVOID FINES return on or before date due.
MAY BE RECALLED with earlier due date if requested.

 

DATE DUE

DATE DUE

DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

6/01 c:/CIRC/DateOue.pG$-p.15

.. -_A‘ ””4

THE IDENTIFICATION, DISTRIBUTION AND SPREAD OF DOGWOOD
ANTHRACNOSE IN MICHIGAN

By

Zachary James Blankenheim

A THESIS
Submitted to
Michigan State University

in partial fulfillment of the requirements
for the degree of

MASTER OF SCIENCE
Department of Plant Pathology

2003

ABSTRACT

THE IDENTIFICATION, DISTRIBUTION AND SPREAD OF DOGWOOD
ANTHRACNOSE IN MICHIGAN

By

Zachary James Blankenheim

Dogwood anthracnose, caused by the fungal pathogen Discula destructiva, has led
to wide spread mortality of the native flowering dogwood tree, Cornusflorida. Since its
initial identification in the 19705, D. destructiva has spread throughout much of the range
of C. florida, but only one county in Michigan was identified as infected in 1993. The
objectives of this study were to determine the distribution of dogwood anthracnose in
Michigan forests, design species-specific PCR primers for detection of D. destructiva,
and survey imported flowering dogwood stock in Michigan nurseries by microscopic and
PCR methods. A total of 52 long-term monitoring plots recorded using global positioning
systems (GPS) were developed during surveys for dogwood anthracnose in Michigan.
Dogwood anthracnose was found at 13 sites in 7 counties in Michigan. PCR primers DdF
and DdR targeted the internal transcribed spacer (ITS) region of rDNA, amplifying a 460
bp fragment of DNA. PCR primers for DdF and DdR were found to amplify DNA from
both D. destructiva and D. fraxinae. A survey of five Michigan nurseries in 2001
examined 342 newly arrived C. florida and C. kousa trees from ten out—of-state nurseries.
Analysis of twig samples from nursery trees by microsc0pic methods yielded an infection
rate of 10.23%, with infections occurring in 8 of 10 out-of—state nurseries and all five
Michigan nurseries. PCR of nursery samples using DdF and DdR detected infected trees

in 2 of 10 out-of-state nurseries and 2 of 5 Michigan nurseries. for a 4.39% infection rate.

DEDICATION
I dedicate this to my grandparents Ralph and Beatrice Blankenheim, and Albert and

Helen Itzin. Your love throughout my life and support of my college career allowed me
to follow my dreams. Thank you.

iii

ACKNOWLEDGMENTS

I would like to thank those people that contributed to the successful completion of
these studies. Dr. Gerard Adams, my major professor, you provided me the opportunity
to work under your guidance in plant pathology and introduced me to the world of
molecular genetics. Dr. Janice Glime, your teaching made the realm of plants truly
interesting to me. Dr. John Adler, your support as my mentor and fiiend throughout my
college career has been inspiring.

To my mom, dad and brother —— your love and support over the years is more than
any son/brother could have asked for. You’ve always been behind my career choices
without hesitation and that means the world to me, thank you.

Tina your enthusiasm for the outdoors, interest in my work, love and friendship

throughout the years has kept me going during good and bad times. I love you.

TABLE OF CONTENTS

LIST OF TABLES ................................................................................... vi
LIST OF FIGURES ................................................................................. vii
LITERATURE REVIEW 1
Introduction 2
Symptoms and Signs 4
Culture Morphology and Sporulation 6
Dissemination of Dogwood Anthracnose 6
Spread and Impact 7
Use of PCR for Detection of Fungal Plant Pathogens 9
Genetics of Discula destructiva 1 1
Literature Cited 13
CHAPTER I. THE INTRODUCTION, DISTRIBUTION AND SPREAD OF
DOGWOOD ANTHRACNOSE IN MICHIGAN ............................. 19
Introduction 20
Materials and Methods 25
Results 34
Discussion 49
Literature Cited 57

LIST OF TABLES

Table Eggs;
CHAPTER I. THE INTRODUCTION, DISTRIBUTION AND SPREAD OF
DOGWOOD ANTHRACNOSE IN MICHIGAN
l. Fungal cultures used for development, specificity and sensitivity testing of
Discula destructiva PCR primers DdF and DdR ............................................. 30
2. Location of dogwood health monitoring plots in Michigan by county and
geographic coordinates developed using global positioning systems, and the
presence or absence of dogwood anthracnose ................................................ 36
3. Inspection of Cornusflorida and C. kousa nursery trees exported to Michigan
nurseries in 2001, for Discula destructiva, using microscopy and PCR primer
detection ........................................................................................... 38
4. Michigan forest sites identified as infected with Discula destructiva from
symptomatic twigs, using microscopy and PCR primers DdF and DdR ................. 45

vi

Fi

LIST OF FIGURES

re Page

CHAPTER I. THE INTRODUCTION, DISTRIBUTION AND SPREAD OF

1.

DOGWOOD ANTHRACNOSE IN MICHIGAN

Map of Lower Michigan showing counties containing Camus florida trees,
location of all 52 health monitoring transects including the 11 transects infected
with Discula destructiva, and one infected site without a transect ...................... 28

Results showing the number of imported Cornusflorida nursery trees in 2001,
identified microscopically as infected with Discula destructiva, based on the
presence, morphology and size of phialides and conidia ................................. 39

PCR amplification of nuclear rDNA from total DNA isolated from pure cultures
of various Discula species and a unknown Phoma sp. Electrophoresis in 1.5%
(wt/vol) agarose in 1x TAE. The two outer lanes contain lKB plus molecular
markers (Invitrogen). Odd numbered lanes contain samples amplified using
universal primers ITSlF and ITS4, and even numbered lanes contain samples
amplified using primers DdF and DdR. Lanes 1 and 2, Discula destructiva #23;
lanes 3 and 4, Discula destructiva #100; lanes 5 and 6, Discula campestris
Williams Hollow; lanes 7 and 8, Discula campestris Blessed Mountain; lanes 9
and 10, Disculafiaxinae 96001; lanes 11 and 12, Disculafi'axinae 96003; lanes
13 and 14, Phoma sp. BIO-t; lanes 15 and 16, Phoma sp. GR]; lanes 17 and 18,
no template DNA (i.e. negative controls) ................................................... 42

PCR amplification of rDNA from total DNA extracted from Michigan cultures

of Discula destructiva and woody twig extracts, both from symptomatic forest
samples. Electrophoresis in 1.5% (wt/vol) agarose in 1x TAE. The two outer

lanes contain lKB plus molecular markers (Invitrogen). Lanes 1 to 18 contain
samples amplified using primers DdF and DdR. Lanes 1 to 5, cultured Discula
destructiva; lanes 6 to 16, woody extracts. Lane 1, T3; lane 2, T26; lane 3, T5;

lane 4, T44; lane 5, T52; lane 6, T3; lane 7, T26; lane 8, T5; lane 9, T44; lane 10,
T52; lane 11, T24; lane 12, T48; lane 13, T47; lane 14, FCTC 25; lane 15, FCTC
40; lane 16, T22. Lane 17, Discula destructiva #23 (i.e. positive control); lane 18,
no DNA template (i.e. negative control) .................................................... 43

vii

Fi

5.

8.

LIST OF FIGURES (can’t)

re Page

PCR products of rDNA, testing sensitivity of primers DdF and DdR against
DNA extraction of infected and uninfected woody material. Electrophoresis in
1.5% (wt/vol) agarose in 1x TAE. The two outer lanes contain lKB plus
(Invitrogen) molecular marker. Lanes 1 to 3, samples amplified with universal
primers NS7 and NS8. Lane 1 and 2 contain uninfected live and dead twig
samples, respectively; lane 3, no DNA template (i.e. negative control). Lanes 4 to
18 contain samples amplified with primers DdF and DdR. Lanes 4 and 5, live and
dead uninfected twig samples, respectively; lane 6 to 13, extraction of decreasing
number of acervuli; lane 6, 110 acervuli; lane 7, 80 acervuli; lane 8, 60 acervuli;
lane 9, 40 acervuli; lane 10, 30 acervuli; lane 11, 20 acervuli; lane 12, 15 acervuli;
lane 13, 10 acervuli. Lanes 14 to 17, progressive ten-fold dilutions (100mg, 10ng,
lng and IOOpg) of woody extract containing Discula destructiva. Lane 18, no
DNA template (i.e. negative control) ........................................................ 44

Results showing the number of imported Cornusflorida nursery trees in 2001,
identified as infected with Discula destructiva using PCR primers DdF and DdR,
based on the presence of Discula nuclear DNA at a concentration greater than or
equal to lOng ................................................................................... 46

PCR amplification of dogwood woody extracts from nursery samples. Nursery
sample set: Cornusflorida var. cloud 9, from nursery #5, Oregon, individual
trees 1 to 8 tested. Electrophoresis in 1.5% (wt/vol) agarose in 1x TAE.
Amplification using primer combinations DdF and DdR (A), and NS7 and N88
(B). The two outer lanes contain lKB plus (Invitrogen) molecular markers. For
both gels, odd numbered lanes contain undiluted samples and even numbered
lanes contain 10 fold dilutions. For both gels (A & B) lanes 1 and 2, tree 1; lanes
3 and 4, tree 2; lanes 5 and 6, tree 3; lanes 7 and 8, tree 4; lanes 9 and 10, tree 5;
lanes 11 and 12, tree 6; lanes 13 and 14, tree 7; lanes 15 and 16, tree 8; lane 17,
Discula destructiva #23 DNA template (i.e. positive control); lane 18, no
template DNA (i.e. negative control) ....................................................... 47

Comparison of microscopic and PCR results positive for the presence of Discula
destructiva, from imported Cornusflorida nursery trees in 2001 ....................... 48

viii

LITERATURE REVIEW

Introduction

The native eastern flowering dogwood, Cornusflorr’da Link, is a popular
ornamental tree in the United States and an important forest sub-canopy species with a
wide geographical range. The natural range of flowering dogwood extends from the
Atlantic coast in the east, to mid—Michigan and southern Maine in the north, to Missouri,
eastern Oklahoma and Texas in the west, and to northern Florida and southern Alabama
in the south (McLemore 1990). C. florida is found growing in well drained, moderately
moist to dry, light textured soils with a pH of 6 to 7, with a shallow root system, which
makes this tree susceptible to drought (Vimmerstedt 1965, McLemore 1990). Due to the
broad geographical range and tolerance of soil conditions, the flowering dogwood has
been recorded in 22 forest cover types. However, it is most commonly associated with
oak-hickory forests and southern pine forests (Eyre 1980, McLemore 1990). Flowering
dogwood is tolerant to shade and reacts quickly to increased levels of sunlight, acting as a
gap closer in forest communities (Wallace and Dunn 1980, McGee 1986).

Flowering occurs between mid—March in the south to late May in the north, often
before leaf-out of most associated species (McLemore 1990). lnflorescences contain
approximately 30 perfect flowers, consisting of very large, white, showy bracts and yield
of about 1-5 fruit (Carr and Banas 2000). Fruit consist of thin, fleshy, red drupes that
contain a two-seeded stone (McLemore 1990).

Besides having an important role in gap dynamics, C. florida has other important
ecological roles. Stiles (1980) and Rossell et a1. (2001) reported that fruit from the
flowering dogwood tree are high in lipid content 04-17%), and Servello and Kirkpatrick

(1987) also reported crude protein to be 10%. This makes dogwood drupes a high quality

fruit in the fall for many mammals, as well as nearly 40 species of overwintering and
migratory birds (Stiles 1980, Mitchell et al. 1988, Rossell et al. 2001). Leaf litter plays an
important role in calcium cycling as it contains 2.0 —- 3.5% calcium by dry weight
(Thomas 1969). C. florida leaves decompose quickly returning calcium to the soil and is
an important pH buffering agent in acidic soils (Hiers and Evans 1997). Studies in other
temperate forests have shown that calcium cycling in acidic soils is critical for birds,
which depend on calcium obtained through invertebrates for proper eggshell formation
(Graveland et al. 1994).

In addition to the ecological significance, C. florida is economically important as
well. Flowering dogwood is an important ornamental tree due to its size and flowering
ability, with an estimated 20 to 30 million in sales in Tennessee in 1989 (Windham
1990). In the southeastern United States one acre of flowering dogwoods can generate up
to $60,000 in gross revenue (Badenhop et al. 1985).

During the mid 19703, the health of C. florida on the east coast and Camus
nuttallii Audubon, the Pacific dogwood, on the west coast declined. A new mysterious
disease named ‘dogwood anthracnose’ was reported in Washington State in 1976 and
1977 on C. nuttallii, and C. florida trees (Byther and Davidson 1979). In southeastern
New York and southwestern Connecticut in 1978 and 1979, Pirone (1980) reported a
massive decline of flowering dogwood, which was attributed to the fungus
Colletotrichum gloeosporioides (Penz.) Sacc., in conjunction with wet springs. In 1983,
this new disease affecting flowering dogwood in the cast was named ‘lower branch
dieback’ (Daughtrey and Hibben 1983) and was caused by an unknown species of

Discula (Salogga 1983). Hibben and Daughtrey (1988) concluded that ‘lower branch

dieback’ in the east and ‘dogwood anthracnose’ in the west, were caused by the same
Discula sp. and agreed on the common name ‘dogwood anthracnose.’ Redlin (1991)
described the pathogen as a new species, Discula destructiva Redlin, and named C.
florida, C. nuttallii and C. kousa as the hosts.

A second undescribed species of Discula isolated from 7-8% of the symptomatic
lesions was identified as causing dogwood anthracnose of forest trees (Windham et al.
1994). Discula destructiva (type 1) and the unknown Discula sp. (type 2) are similar in
colony morphology and cause similar symptoms on dogwood trees, however only D.
destructiva can grow on gallic acid medium (Trigiano et al. 1993a, Daughtrey and
Hibben 1994). Type 1 and 2 isolates can also be distinguished by DNA amplification
fingerprinting (DAF; Trigiano et al. 1993b).

Symptoms and signs

Leaf spots, usually found on the lower canopy, are the first symptoms of dogwood
anthracnose. Leaf spots consist of spots, necrotic blotches, and blight (Hibben and
Daughtrey 1988). Leaf spots are first observed as small (<7mm diameter) purple/red
rimmed lesions that expand slowly, and can take on a shot-hole appearance when the
necrotic tissue in the center breaks apart. Brown-olive leaf blotches are larger and usually
form along the edge of the blade, sometimes with a reddish-purple border, but are not
observed in Michigan. Infection progressing through the petiole leads to leaf blight, most
often of the terminal pair of leaves and especially on terminal leaves of epicormic shoots.
Leaf abscission can occur when infections resulting in leaf blight occur early in the
growing season as reported by Britton (1993). Otherwise blighted leaves will remain

attached until the following spring (Hibben and Daughtrey 1988). Rainy periods in the

spring can lead to red-brown spots and necrotic blotches on flowers. Progression of the
fungus through the petiole leads to tan-white colored twig cankers, with brown discolored
areas beneath the bark at the point of leaf attachment (Hibben and Daughtrey 1988).
Sunken lesions (<3mm diameter) varying from orange-red to black often form on the
small shoots. High incidence of twig dieback on the lower branches is what originally led
to the early disease name of ‘lower branch dieback’ (Daughtrey and Hibben 1983).
Epicormic shoots often form on the larger branches and trunk of diseased trees. Infection
occurs rapidly in the epicormic shoots and provides an infection avenue into the main
branches and trunk. Brown elliptical cankers are sometimes present underneath the bark
where infected epicormic shoots were attached. These annual cankers are often delimited
by callus beneath areas of sunken, cracked bark and can expand, hastening the decline of
the tree (Daughtrey et al. 1988, Hibben and Daughtrey 1988).

Signs of D. destructiva include production of conidia from conidiomata.
Conidiomata are found abundantly in the necrotic tissue on the abaxial side of leaves and
in the dead bark of twigs and epicormic shoots (Hibben and Daughtrey 198 8, Daughtrey
et al. 1988). Conidiomata on leaves are reddish brown to black, acervular, subcuticular,

30-135um wide and are often found below trichomes (Hibben and Daughtrey 1988,

Redlin 1991). On twigs, conidiomata are irregularly elliptical acervuli, 90-340um wide
and orange-black in color (Hibben and Daughtrey 1988, Redlin 1991). Conidia are
released in gelatinous masses or cirrhi, and are white-gray to orange-pink in color
(Hibben and Daughtrey 1988). Conidia are hyaline, single celled, elliptical-fusiforrn with

polar guttules, 6-12 x 2.5-4um in size, and formed from enteroblastic, phialidic cells 10-

18 x 2.5-4.5um in size (Redlin 1991). To date, no known sexual state has been identified,

however a teleomorph may be found in the genera Apiognomom'a based on phylogenetic
analysis of rDNA (Redlin 1991, Zhang and Blackwell 2001).

Culture morphology and Sporulation

D. destructiva colonies on potato dextrose agar (PDA) or malt agar (MA) are

initially white but turn gray-green to black with age. Colonies are appressed with an
undulate margin, sparse aerial mycelium, and form droplets of exudate near the center of
the colony, which upon drying give the colony a pitted appearance (Hibben and
Daughtrey 1988, Redlin 1991). Variances in colony morphology have been reported
including sectoring, pigmentation and growth rate, which is believed to be due to the
presence of double stranded RNA (Yao et al. 1993, McElreath and Tainter 1993).
Conidiomata formation on PDA or MA is sparse and typically forms toward the outer
edges of the colony, with white-pink masses of conidia being formed (Hibben and
Daughtrey 1988, Redlin 1991). Sporulation in culture is enhanced by adding dogwood
tissue (Hibben and Daughtrey 1988), ground up oak or maple leaves (McElreath and
Tainter 1993), or by using oatmeal agar (Redlin 1991). Optimal conidia production is
enhanced by light (McElreath and Tainter 1993) and germination occurs at 20-24°C
(Britton 1989), with no growth at 27°C (Salogga 1982).

Dissemination of dogwood anthracnose

Primary spread of D. destructiva over short distances occurs via wind and rain

splash of conidia during weather events (Daughtrey et al. 1988). Studies have also

suggested that birds may contribute to the spread of dogwood anthracnose. D. destructiva

has been successfully cultured from infected dogwood seeds. Birds may contribute by the

spread by consuming infected fruits and excreting the seeds, which they cannot digest
(Britton et al. 1993).

Insect mediated transportation of D. destructiva spores is possible in a laboratory
setting. The convergent lady beetle, Hippodamia convergens Guerin-Meneville can pick
up spores on infected leaves of C. florida. These spores remained viable for up to 16 days
when the temperature was close to 20°C with 51-60% relative humidity (Colby 1993,
Colby et al. 1995, Colby et al. 1996). More importantly, lady beetles carrying D.
destructiva spores were able to infect healthy flowering dogwood leaves in the laboratory
(Colby 1993). Research has shown that D. destructiva spores that are ingested by H.
convergens remain viable when excreted as frass, with a range of viability from 80% at
12 hours to 12.5% at 96 hours (Hed et al. 1999).

Several researchers have alluded to the possibility of dissemination of dogwood
anthracnose through infected nursery stock (Daughtrey et a1. 1996) or by propagation
with infested seeds (Britton et al. 1993). However, at this time there are no reports
documenting any infected material being shipped within and among states.

Spread and impact

After the initial 1983 report of dogwood anthracnose as a new disease in New
York, Pennsylvania, New Jersey and Connecticut, the disease expanded through much of
C. florida’s natural range (Daughtrey and Hibben 1983). By 1987, Maryland, Delaware,
Massachusetts, Georgia, Virginia and West Virginia had been added to the growing list
of states with dogwood anthracnose (Anderson 1990). In 1988, the disease was reported
in North Carolina, South Carolina and Tennessee, with Kentucky and Alabama being

added to the list in 1989 (Knighten and Anderson 1992, Daughtrey and Hibben 1994).

Other states that were confirmed as having infected trees included Ohio and the District
of Columbia (1990) (Daughtrey and Hibben 1994), New Hampshire (1991), Rhode Island
(1992) (Daughtrey et al. 1996), Vermont, Michigan (Daughtrey et al. 1996) and Indiana
(1993) (Rane 1993), and Kansas and Missouri (1994) (Daughtrey et al. 1996).

Early surveys in 1984 at Catoctin Mountain National Park, Maryland, revealed
that only 3% of flowering dogwoods were free of dogwood anthracnose symptoms and
nearly 33% were dead (Mielke and Langdon 1986). Schneeberger and Jackson (1989)
followed up this survey in 1988 and reported that mortality had increased to 89%, with
100% disease incidence, and little regeneration. A final survey of the same impact plots
in 1994 showed a 94% loss of flowering dogwood (Sherald et al. 1996). Another long-
term study by Knighten and Anderson (1992, 1993) reported that the number of hectares
of diseased dogwoods in 210 impact assessment plots increased from 0.2 million hectares
in 1988 to 7.0 million hectares in 1993, with an increase in mortality from 0% (1988) to
23% (1993). Hiers and Evans (1997) conducted a study at the Cumberland Plateau,
Tennessee, on dogwood stem density and disease incidence. Two sites were studied and
showed a 98% and 87% reduction in live stems, and 87% and 86% disease incidence,
respectively. Despite D. destructiva causing a high incidence of mortality, McEwan et al.
(2000), reported that long-term mortality trends were difficult to quantify. It has been
suggested that not all areas exhibiting dogwood decline suffer from dogwood
anthracnose. In a study conducted in Kentucky, old-growth forest plots had a 36% stem
reduction from 1989-1998, which corresponded to anthracnose infection. However, the
percent of decline was similar for the same sites during 1979-1989 when dogwood

anthracnose was not present.

Use of PCR for detection of fungal plant pathogens

Methods for detecting and identifying plant pathogens are not always quick or
reliable (Henson and French 1993). The polymerase chain reaction (PCR) has expanded
pathologists’ ability to detect disease by amplification of specific DNA sequences, which
allows for the identification and comparison of species on a molecular level, instead of
solely relying on traditional identification methods (Mullis and F aloona 1987). One
particular area of the fungal genome that has been targeted for both phylogenetic and
diagnostic studies has been the internal transcribed spacer (ITS) region, which contains
two variable non-coding segments that lie between the highly conserved ribosomal DNA
(rDNA) 18S, 5.8S and 283 subunits (White et al. 1990). The ITS region is a useful target
for molecular identification because of ‘universal primers’ that will semi-selectively
amplify the fungal ITS region, which is only 500-800 bp in length (White et al. 1990,
Gardes and Bruns 1993). The rDNA region is present in high quantity allowing for
amplification from small or degraded samples (Russel et al. 1984). Also the ITS is useful
for identification because the base sequence can be quite variable among morphologically
distinct species (Anderson and Stasovski 1992, Chen et al. 1992, Gardes et al. 1991) but
shows little intraspecific variation (Lee and Taylor 1992, Anderson and Stasovski 1992).
This allows for possible identification to the species level with the right oligonucleotide
primers. Species—specific oligonucleotide primers are often designed for diagnostic
purposes for use with diseased plant samples. These species-specific primers have
allowed for identification to the species level of several pathogenic fungi including.

Serpula lacrymans (Schmidt and Moreth 2000), Ophiosroma picea and 0. quercus (Kim

et al. 1999), Pythium ultimum (Kageyama et al.1997), and Verticillium albo—atrum and V.
dahliae (Nazar et al. 1991).

Using species-specific primers and PCR technology has allowed researchers to
identify fungi to the species level and has been used to overcome various difficulties. For
example Plasmodiophora brassicae, the causal agent of clubroot (Faggian et al. 1999),
and various ectomycorrhizal basidiomycetes that serve as symbionts to many plants
(Gardes and Bruns 1993). PCR can be used to identify these fungi that would otherwise
be difficult to identify since they cannot be grown in culture. The use of PCR with
species-specific primers has also reduced the time it takes to identify wood rot and stain
fungi, from months to hours, through bypassing tedious culturing (Kim et al. 1999,
Schmidt and Moreth 2000). PCR technology saves time, money and may help prevent the
movement of quarantined pathogens. Powdery scab of potato (Spongospora sublerranea)
is a quarantined pathogen that despite all efforts has spread worldwide due to its ability to
exist in an asymptomatic state. The development of species-specific primers has
improved detection capabilities, which may allow for better containment and reduced
economic loss (Bulman and Marchall 1998). PCR is also effective for assaying for
pathogens when other methods fail due to plant inhibitory compounds. Enzyme-linked
immunosorbent assay (ELISA) is a sensitive method for detection of wood decay fungi
but is often inhibited by compounds found in the wood. With the right DNA extraction
procedure and PCR species-specific primers, reliable detection has been achieved

(J asalavich et al. 2000).

10

Genetics of D. destructiva

Dogwood anthracnose appears to be an exotic fungus introduced during the
19705. This long-supported hypothesis (Salogga and Ammirati 1983, Hibben and
Daughtrey 1988), originates from the rapid appearance of D. destructiva at two coastal
port cities, its unchecked spread over much of the geographic range of C. florida
(Daughtrey et al. 1996), and lack of host resistance (Mielke and Langdon 1986, Brown et
al. 1996).

Recent research supports the concept of D. destructiva as an introduced pathogen.
Early molecular analysis using DNA amplification fingerprinting (DAF) on pooled
isolates from the western versus the eastern United States shows that the genome is
highly conserved (Trigiano et al. 1995). Further work utilizing arbitrary signatures from
amplification profiles (ASAP) distinguished fine population differences between western
and eastern isolates (Caetano-Anolles et al. 1996). Additional evidence for D. destructiva
being an introduced pathogen has come from the discovery of double-stranded RNA,
present in many isolates (McElreath et a1. 1994, Yao et al. 1997). Studies show dsRNA
are present in all eastern isolates (McElreath et al. 1994), but only 25% of western
isolates from the United States. Generation of agarose-banding profiles show differences
between the western and eastern isolates (Yao et al. 1997, Rong et al. 2001). While
genetic diversity of D. destructiva is low, especially when compared to other fungal
species (Caetano-Anolles et al. 2001), there are two main populations of isolates from the
western and the eastern United States. In addition, these latest findings used amplified

fragment length polymorphisms and sequencing of the intergenic spacer (IGS), B-tubulin

genes, and translation elongation factor-1a, which showed that out of 72 isolates there

11

were 20 genotypes present (Zhang and Blackwell 2002). However, of these 20 genotypes,
17 were found in the eastern United States. suggesting that the eastern population is more
diverse than the western population.

Since D. destructiva has limited genetic diversity, it may be adventitious to
develop species-specific primers. Species-specific PCR primers, coupled with a DNA
extraction procedure that avoids PCR inhibition when diagnosing D. destructiva from C.

florida, could be useful for diagnostic clinics and state nursery inspectors.

12

LITERATURE CITED

Anderson, J. B. and Stasovski, E. 1992. Molecular phylogeny of Northern Hemisphere
species of Armillaria. Mycologia 84:505-516.

Anderson, R. L., (compiler). 1990. Results of the 1989 dogwood anthracnose impact
assessment and pilot test in the southeastern United States. USDA For. Serv. South.
Region. Prot. Rep. R8-PR l8.

Badenhop, B. B., Witte, W. T., and Glasgow, T. E. 1985. Production systems and costs
for producing balled and burlapped trees of dogwood cultivars. Tennessee, 1984. Univ. of
Tenn. Agric. Expt. Sta. Bul. 637.

Britton, K. O. 1989. Temperature, pH and free water effects on in vitro germination of
conidia of a Discula sp. isolated from dogwood anthracnose lesions. Phytopathology
79:1203.

Britton, K. O. 1993. Anthracnose infection of dogwood seedlings exposed to natural
inoculum in western North Carolina. Plant Disease 77234-3 7.

Britton, K. O., Roncadori, R. W., and Hendrix, F. F. 1993. Isolation of Discula
destructiva and other fungi from seeds of dogwood trees. Plant Disease 77: 1026-1028.

Brown, D. A., Windham, M. T., and Trigiano, R. N. 1996. Resistance to dogwood
anthracnose among Cornus species. Journal of Arboriculture 22:83-85.

Bulman, S. R. and Marshall, J. W. 1998. Detection of Spongospora subterranean in
potato tuber lesions using the polymerase chain reaction (PCR). Plant Pathology 47:75 9-
766.

Burns, R. M., and Honkala, B. H. 1990. Silvics of North America. Volume 2.
Hardwoods. US. Forest Service, Washington, DC. pp278-283.

Byther, R. S. and Davidson, R. M., Jr. 1979. Dogwood anthracnose. Ornamental
Northwest Newletter 3(2):20-21.

Caetano-Anolles, G., Trigiano, R. N., and Windham, M. T. 1996. Sequence signatures
from DNA amplification fingerprints reveal fine population structure of the dogwood
pathogen Discula destructiva. FEMS Microbiology Letters 145:377-383.

Caetano-Anolles, G., Trigiano, R. N., and Windham, M. T. 2001. Patterns of evolution in

Discula fungi and the origin of dogwood anthracnose in North America, studied using
arbitrarily amplified and ribosomal DNA. Current Genetics 39:346-354.

13

Carr, D. E., and Banas, LE. 2000. Dogwood anthracnose (Discula destructiva): effects of

and consequences for host (Cornusflorida) demography. American Midland Naturalist
143:169-177.

Chen, W., Hoy, J. W., and Schneider, R. W. 1992. Species-specific polymorphisms in
transcribed ribosomal DNA of five Pythium species. Experimental Mycology 16:22-34.

Colby, D. M. 1993. Arthropod dissemination of Discula destructiva conidia on Camus
florida. M.S. thesis, University of Tennessee, Knoxville.

Colby, D. M., Windham, M. T., and Grant, J. F. 1995. Hippodamia convergens
(Coleoptera: Coccinellidae) dissemination of dogwood anthracnose fungus
(Melanconiales: Melanconeacae). Environmental Ecology 24: 1 075-1079.

Colby, D. M., Windham, M. T., and Grant, J. F. 1996. Transportation and viability of
conidia of Discula destructiva on Hippodamia convergens. Plant Disease 80:804-805.

Daughtrey, M. L. and Hibben, C. R. 1983. Lower branch dieback, a new disease of
northern dogwoods. (Abstr.) Phytopathology 73:365.

Daughtrey, M. L., Hibben, C. R., and Hudler, G. W. 1988. Cause and control of dogwood
anthracnose in northeastern United States. Journal of Aboriculture 14:159-164.

Daughtrey, M. L. and Hibben, C. R. 1994. Dogwood anthracnose: A new disease
threatens two native Cornus species. Annual Review of Phytopathology 32:61-73.

Daughtrey, M. L., Hibben, C. R., Britton, K. O., Windham, M. T., and Redlin, S. C.
1996. Dogwood anthracnose, understanding a disease new to North America. Plant
Disease 80:349-358.

Eyre, F. H., ed. 1980. Forest cover types of the United States and Canada. Society of
American Foresters, Washington, DC. p 148.

Faggian, R., Bulman, S. R., Lawrie, A. C., and Porter, 1. J. 1999. Specific polymerase
chain reaction primers for the detection of Plasmodiophora brassicae in soil and water.
Phytopathology 89:392-397.

Gardes, M., White, T. J ., Fortin, J. A., Bruns, T. D., and Taylor, J. W. 1991.
Identification of indigenous and introduced symbiotic fungi in ectomycorrhizae by

amplification of nuclear and mitochondrial ribosomal DNA. Canadian Journal of Botany
69:180-190.

Gardes, M. and Bruns, T. D. 1993. ITS primers with enhanced specificity for

basidiomycetes -— application to the identification of mycorrhizae and rusts. Molecular
Ecology 2:113-118.

14

Graveland, J. R., van der Wal, J. H., and Van Noordwijk, A. J. 1994. Poor reproduction
in forest passerines from decline of snail abundance on acidified soils. Nature 3682446-
448.

Hed, E. H., Windham, M. T., and Grant, J. F. 1999. Survival of conidia of Discula
destructiva in frass of the convergent lady beetle. Plant Disease 83:806-809.

Henson, J. M. and French, R. 1993. The polymerase chain reaction and plant disease
diagnosis. Annual Review of Phytopathology 31 :81-109.

Hibben, C. R. and Daughtrey, M. L. 1988. Dogwood anthracnose in northeastern United
States. Plant Disease 72:199-203.

Hiers, J. K. and Evans, J. P. 1997. Effects of anthracnose on dogwood mortality and
forest composition of the Cumberland Plateau (USA). Conservation Biology 1121430-
1435.

Jasalavich, C. A., Ostrofsky, A., and Jellison, J. 2000. Detection and identification of
decay fungi in spruce wood by restriction fragment length polymorphism analysis of

amplified genes encoding rRNA. Applied and Environmental Microbiology 66:4725-
4734.

Kageyama, K., Ohyama, A., and Hyakumachi, M. 1997. Detection of Pythium ultimum
using polymerase chain reaction with species-specific primers. Plant Disease 81 :1 15 5-
1160.

Kim, S. H., Uzunovic, A. and Breuil, C. 1999. Rapid detection of Ophiostoma piceae and
0. quercus in stained wood by PCR. Applied and Environmental Microbiology 65:287-
290.

Knighten, J. L. and Anderson, R. L., eds. 1992. Results of the 1991 dogwood anthracnose
impact assessment and pilot test in the southeastern United States. USDA For. Serv.
South. Reg. Prot. Rep. R8-PR 23.

Knighten, J. L. and Anderson, R. L., eds. 1993. Results of the 1992 dogwood anthracnose
impact assessment and pilot test in the southeastern United States. USDA For. Serv.
South. Reg. Prot. Rep. R8-PR 24.

Lee, S. B. and Taylor, J. W. 1992. Phylogeny of five fungus-like protoctistan
Phytophthora species, inferred from the internal transcribed spacers of ribosomal DNA.
Molecular Biology and Evolution 9:636-653.

McElreath, S. D. and Tainter, F. H. 1993. A sporulation medium for Discula destructiva,
the dogwood anthracnose fungus. Current Microbiology 26:1 17-121.

15

McElreath, S. D., Yao, J. M., Coker, P. S., Tainter, F. H. 1994. Double-stranded RNA in
isolates of Discula destructiva from the eastern United States. Current Microbiology
29:57-60.

McEwan, R. W., Muller, R. N., Arthur, M. A., and Housman, H. H. 2000. Temporal and
ecological patterns of flowering dogwood mortality in the mixed mesophytic forest of
eastern Kentucky. Journal of the Torrey Botanical Society 127:221—229.

McGee, C. E. 1986. Loss of Quercus spp. dominance in an undisturbed old-growth
forest. Journal of the Elisha Mitchell Scientific Society 102:10-15.

McLemore, B. F. 1990. Carnusflarida L.: Flowering Dogwood. Pages 278-283 in Silvics
of North America, Vol 2, Hardwoods. Burns, R. M. and Honkala, B. H., eds. Forest
Service, US. Dept. Agric., Washington, DC.

Mielke, M. E. and Langdon, K. 1986. Dogwood anthracnose threatens Catoctin Mountain
Park. Park Science (Winter):6-8.

Mitchell, W. A., Gibbs, P. A., and Martin, C. O. 1988. Flowering dogwood (Camus
flarida): section 7.5.9, US. Army Corps of Engineers Wildlife Resources Management

Manual, Tech Rept. EL-88-9, US. Army Engineer Waterways Experiment Station,
Vicksburg, MS. 25p.

Mullis, K. B. and Faloona, F. A. 1987. Specific synthesis of DNA in vitro via a
polymerase-catalysed chain reaction. Methods in Enzymology 155:335-350.

Nazar, R. N., Hu, X., Schmidt, J ., Culham, D., and Robb, J. 1991. Potential use of PCR-
amplified ribosomal intergenic sequences in the detection and differentiation of
verticillium wilt pathogens. Physiological and Molecular Plant Pathology 3921-11.

Pirone, P. P. 1980. Parasitic fungus affects region’s dogwood. New York Times. Feb. 24,
pp.34,37.

Rane, K. K. 1993. Dogwood anthrcnose, it’s here! Indiana Nursery News 54:1, 6, 15.

Redlin, S. C. 1991. Discula destructiva sp. nov., cause of dogwood anthracnose.
Mycologia 83:633-642.

Rong, R., Rao, 8., Scott, S. W., and Tainter, F. H. 2001. Common multiple dsRNAs are
present in populations of the fungus Discula destructiva originating from widely
separated geographic locations. Current Microbiology 42: 144-148.

Rossell, I. M., Rossell, C. R. Jr., Hining, K. J., and Anderson, R. L. 2001. Impacts of
dogwood anthracnose (Discula destructiva Redlin) on the fruits of flowering dogwood
(Carnusflarida L.): Implications for wildlife. American Midland Naturalist 146:3 79-387.

16

Russel, P. J ., Wagner, 8., Rodland, K. D., Feinbaum, R. L., Russel, J. P., Bret-Harte, M.
8., Free, S. J ., and Metzenberg, R. L. 1984. Organization of the ribosomal ribonucleic
acid genes in various wild-type strains and wild-collected strains of Neuraspara.
Molecular Genomic Genetics 1962275-282.

Salogga, D. S. 1982. Occurrence, symptoms and probable cause, Discula species
associated with anthracnose of dogwood in the Pacific Northwest. MS thesis. University
of Washington, Seattle. 167 pp.

Salogga, D. S. and Ammirati, J. F. 1983. Discula species associated with anthracnose of
dogwood in the Pacific Northwest. Plant Disease 67:1290.

Schmidt, 0. and Moreth, U. 2000. Species-specific PCR primers in the rDNA-ITS region
as a diagnostic tool for Serpula Iacrymans. Mycological Research 14:69-72.

Schneeberger, N. F. and Jackson, W. 1989. Dogwood anthracnose at Catoctin Mountain
Park. USDA Forest Service NE. Area State Priv. For. 16pp.

Servello, F. A., and Kirkpatrick, R. L. 1987. Regional variation in the nutritional ecology
of ruffed grouse. Journal of Wildlife Management 51:749-770.

Sherald, J. L., Stidham, T. M., Hadidian, J. M., and Hoeldtke, J. E. 1996. Progression of
the dogwood anthracnose epidemic and the status of flowering dogwood in Catoctin
Mountain Park. Plant Disease 80:310-312.

Stiles, E. W. 1980. Patterns of fruit presentation and seed dispersal in bird-disseminated
woody plants in the eastern deciduous forests. The American Naturalist 116:670-688.

Thomas, W. A. 1969. Accumulation and cycling of calcium by dogwood trees.
Ecological Monographs 39: 101 -1 20.

Trigiano, R. N., Gerhaty, N. E., and Windham, M. T. 1993a. Extracellular enzymes of
two fungi associated with dogwood anthracnose. Phytopathology 83:1410.

Trigiano, R. N., Bassam, B., Caetano-Anolles, G., and Windham, M. T. 1993b.
Characterization of isolates of Discula destructiva and Discula sp. using DNA
amplification fingerprinting. Phytopathology 83: 1 338.

Trigiano, R. N., Caetano-Anolles, G., Bassam, B. J., and Windham, M. T. 1995. DNA
amplification fingerprinting provides evidence that Discula destructiva, the cause of
dogwood anthracnose in North America, is an introduced pathogen. Mycologia 87:490-

500.

Vimmerstedt, J. P. 1965. Flowering dogwood (Carnusflorida L.). In Silvics of forest
trees of the United States. pp 162-166. H. A. Fowells, (compiler). US. Department of
Agriculture, Agriculture Handbook 271. Washington, DC.

17

Wallace, L. L., and Dunn, E. L. 1980. Comparative photosynthesis of three gap phase
successional tree species. Oecologia 45:331-340.

White, T. J ., Bruns, T., Lee, S., and Taylor, J. 1990. Amplification and direct sequencing
of fungal ribosomal RNA genes for phylogenetics. In: Innis M.A., Gelfand D.H., Sninsky
J .J ., and White T.J. (eds). PCR Protocols: a Guide to Methods and Applications, pp. 315-
322. Academic Press, New York.

Windham, M. T. 1990. Dogwood anthracnose research at the University of Tennessee at
Knoxville. Pp. 261-268. In: Proc. Annual Tennessee Nursery Short Course 23, Tennessee
Nurseryman’s Assoc., McMinnville.

Windham, M. T., Erbaugh, E. K., Montgomery-Dee, M. E., and Trigiano, R. N. 1994.
Frequency of Discula destructiva Redlin and an undescribed Discula species from
dogwood tissue. Phytopathology 84:778.

Yao, J. M., McElreath, S. D., and Tainter, F. H. 1993. Correlation of phenotypes of
Discula destructiva with dsRNA. Phytopathology 8321392.

Yao, J. M., McElreath, S. D., and Tainter, F. H. 1997. Double-stranded RNA in isolates
of Discula destructiva from the Pacific northwestern United States and British Columbia,
Canada. Current Microbiology 34:67-69.

Zhang, N. and Blackwell, M. 2001. Molecular phylogeny of dogwood anthracnose
fungus (Discula destructiva) and the Diaporthales. Mycologia 93: 355-365.

Zhang, N. and Blackwell, M. 2002. Population structure of dogwood anthracnose fungus.
Phytopathology 92: 1 276-1 283.

18

CHAPTER I.

THE IDENTIFICATION, DISTRIBUTION AND SPREAD OF DOGWOOD
ANTHRACNOSE IN MICHIGAN

19

INTRODUCTION
The native flowering dogwood, C arnus flarida Link, has a large natural range

covering nearly the eastern third of the United States. Flowering dogwoods are in highest
abundance in the Appalachian mountains, but also reach as far north as Michigan and
Maine, west to Missouri, Arkansas and east Texas, and as far south as northern Florida
(McLemore 1990). During the late 19705 C. flarida on the east coast and the Pacific
dogwood, Camus nuttallii Audubon, on the northwest coast, began declining at epidemic
proportions (Byther and Davidson 1979, Pirone 1980). This decline was initially termed
‘Iower branch dieback of flowering dogwood’ (Daughtrey and Hibben 1983) and was
attributed to an undescribed Discula species (Salogga 1983). Lower branch dieback was
renamed ‘dogwood anthracnose’ in 1988 (Hibben and Daughtrey 1988) and was
described as a new species, Discula destructiva Redlin in 1991 (Redlin).

Symptoms and signs of dogwood anthracnose first appear on the foliage of the
lower branches. Leaf symptoms include brown, necrotic, leaf spots and can include large
blotches or blight during severe outbreaks (Hibben and Daughtrey 1988). However,
purple margins are not readily apparent in leaf spots in Michigan. Light tan twig cankers
result from fungal penetration through the petiole or open wounds. Black fruiting bodies
(acervuli) are visible on cankered twigs and within leaf spots. These fruiting bodies
produce masses of white-gray to orange-pink conidia in a gelatinous cirrhi. Acervuli also
permits secondary infection to occur in wet seasons and allows the fungus to over winter,
acting as an inoculum reservoir for the following spring (Daughtrey et al. 1988, Hibben

and Daughtrey 1988). Limb dieback occurs as the disease progresses, often accompanied

by the production of epicormic shoots. Infection of epicormic shoots often results in the

20

formation of brown-black elliptical trunk cankers underneath the bark, which eventually
leads to girdling of the tree.

Dogwood anthracnose is spread in several ways. Rain splash is the primary
method of spread, allowing the conidia to disperse over short distances through forest
stands (Daughtrey et al. 1988). The spread of infected dogwood seeds by wildlife, more
specifically birds, has been suggested (Britton et al. 1993). D. destructiva has been
isolated from infected seeds; therefore it is possible that birds feeding on infected fruits
could spread the disease since they do not digest the seeds. Insect mediated transport of
conidia is another possibility as it has been shown in a laboratory setting (Colby 1993).
Hippadamia canvergens Guerin-Meneville, the convergent lady beetle, was able to pick
up conidia from infected leaves of C. flarida, and deposit them on healthy leaves causing
new infection (Colby 1993). These conidia remained viable for up to 16 days when the
temperature was approximately 20°C with 51-60% relative humidity (Colby 1993, Colby
et al. 1995, Colby et al. 1996). Additional studies showed spores of D. destructiva that
were ingested by H. canvergens remained viable (12.5%) when excreted as frass up to 96
hours later (Hed et al. 1999). The possibility of spreading dogwood anthracnose through
propagation of infected seeds (Britton et al. 1993) or by state-to-state movement of
infected nursery stock has also been suggested (Daughtrey et al. 1996). At this time
however, no studies have been conducted to determine if infected material is being
shipped within and among states.

The impact of dogwood anthracnose on native C. flarida mortality has been
significant over the past 25 years. Studies done at Catoctin Mountain National Park,

Maryland, showed an increase in mortality from 33% to 89% between 1983 (Mielke and

21

Langdon 1986) and 1988 (Schneeberger and Jackson 1989), with little regeneration
occurring. Knighten and Anderson (1992, 1993) reported that 210 impact plots surveyed
between 1988 and 1993 showed the number of dead dogwoods to have increased from
0.2 million hectares to 7.0 million, respectively.

The destruction of the flowering dogwood is important ecologically for several
reasons. One aspect is that C. flarida grows in a wide geographical range of soils and
forest cover types, from upland oak-hickory forests to southern pine forests (Eyre 1980,
McLemore 1990). Flowering dogwood constitutes an important component of these
different forest ecosystems for several reasons. First, while never acting as a dominant
canopy tree, flowering dogwood acts as a gap-closer when exposed to increased levels of
sunlight (Wallace and Dunn1980, McGee 1986). Second, C. flarida possesses a high
ecological value from a wildlife perspective. Fruit from the flowering dogwood have a
high lipid content of 14-17% (Rossell et a1. 2001) and contain 10% crude protein
(Servello and Kirkpatrick 1987). Due to their nutritional value dogwood fruits are an
important food sources for many mammals and nearly 40 species of migratory and
overwintering birds (Stiles 1980, Mitchell et al. 1988, Rossell et al. 2001). Finally, the
leaves of the flowering dogwood play an important role in cycling of calcium, with a
reported dry-weight of 2.0—3.5% (Thomas 1969). Dogwood leaves, which are high in
calcium, decompose readily, returning calcium to the soil, while acting as a pH buffer in
acidic soils (Heirs and Evans 1997). Studies have shown that calcium cycling is
important as invertebrates consume calcium rich debris and are in turn consumed by
various species of birds that depend on the calcium for preper eggshell formation

(Graveland et al. 1994).

22

The flowering dogwood is also valuable from an economic perspective. C. flarida
is one of the first forest trees to bloom in the spring, and its large showy white bracts,
make it aesthetically pleasing to the public (McLemore 1990). This unique characteristic,
along with its size has led to its utilization by the ornamental tree industry as a widely
planted landscape tree. One acre of flowering dogwood can bring up to $60,000 in gross
revenue and many do gwoods are shipped between states for retail nurseries (Badenhop et
al. 1985, Daughtrey et al. 1996).

Methods for identifying D. destructiva currently include culturing and
microscopic identification, both of which are especially difficult with this species.
Improvements in disease detection through the use of the polymerase chain reaction
(PCR), has allowed for accurate and quick diagnosis of many other plant pathogens that
are difficult to identify by traditional methods (Henson and French 1993, Gardes and
Bruns 1993, Dan et a1. 2001). PCR primers have been designed to detect specific fungal
species from various diseased sources such as Plasmadiaphara brassicae in soil (Faggian
et al. 1999), Xanthamanas campestris pv. citri on leaves (Hartung et a1. 1993 ), and
Ophiastama from wood (Kim et al. 1999). The internal transcribed spacer (ITS) region of
the ribosomal DNA (rDNA) is particularly useful for designing species-specific primers
for two reasons. First the ITS region lies between the highly conserved 18S, 5.88 and 28S
ribosomal subunits, which can be selected for with ‘universal primers’ (White et al. 1990,
Gardes and Bruns 1993). Second, the ITS region shows variability between
morphologically distinct species (Anderson and Stasovski 1992, Chen et al. 1992) but
also shows minimal intraspecific variation (Lee and Taylor 1992, Anderson and

Stasovski 1992). PCR using species-specific primers could prove useful for a disease

23

such as dogwood anthracnose, where culturing frequently fails and diagnosis based on
microscopic examination is unusually difficult.

In Michigan only one site in Kalamazoo County was identified as infected with
dogwood anthracnose by 1993, and no subsequent studies on its spread were pursued
(Daughtrey et a1. 1996). Unlike the Appalachian Mountains where flowering dogwood
grows in abundance, Michigan's populations of flowering dogwood may be protected
from infection as they are fragmented due to large areas of land utilized by agriculture.
However, many nurseries in Michigan import dogwoods from other heavily infected
states. Nursery importation of infected dogwood trees has been suggested as a means of
spread of dogwood anthracnose to states on the edge of its natural range, or to isolated
stands separated by large areas of agriculture (Daughtrey et al. 1996). To date, no studies
have been done to assess the amount of D. destructiva that may be entering Michigan via
the nursery trade.

Based on the lack of knowledge of dogwood anthracnose in Michigan forests and
nurseries, several research objectives were addressed, including: (i) determination of the
extent of spread of D. destructiva in Michigan since 1993 by surveying areas of C.
flarida throughout its natural growing range; (ii) establishment of forest health
monitoring plots which would allow for future monitoring of dogwood anthracnose
spread; (iii) determining what level, if any, of dogwood anthracnose was being imported
into Michigan by inspecting Michigan nurseries and collecting samples of flowering
dogwood for microscopic examination; (iv) development of PCR based detection
methods utilizing species-specific primers to both verify and expedite detection of the

pathogen in plant tissues.

24

MATERIALS AND METHODS

Survey for C. florida in Michigan

In 1997, surveys were initiated to determine areas with sufficient numbers of
flowering dogwood trees for disease inspection and to establish GPS (global positioning
system) health-monitoring plots. Letters explaining the importance of flowering
dogwoods, the impact of dogwood anthracnose and the research that was to be conducted
on the disease, were sent out to people in Michigan who were presumed to have interest
and knowledge of native trees and who might recognize C. florida in a natural setting.
Recipients included in the survey were members on mailing lists of the Michigan Nursery
and Landscape Association (MNLA), Michigan Christmas Tree Association, Michigan
Department of Natural Resources forestry division (MDNR), Michigan Department of
Agriculture (MDA), Michigan State University county extension agents and master
gardeners. In addition to the letter, a self-addressed stamped envelope and form
consisting of the survey with a map of their county was included. The survey form
requested whether they might indicate and describe locations of flowering dogwoods,
whether they had noticed any possible decline, dieback, death or obvious disease
symptoms, and their phone numbers for future reference. In addition to letter surveys,
phone surveys were conducted at nature centers, gardens, and headquarters in state park
and recreation areas. When possible, homeowners near sites showing disease symptoms
were asked if they had ever planted a dogwood tree purchased from a nursery in the area.
Based on survey information, state and private landowners that indicated knowledge of
locations having sufficient amounts of dogwoods present, diseased or healthy were

contacted for permission to enter their property. Suitable sites containing high numbers of

25

C. flarida (dead or alive) were investigated and sampled, as described below. The
establishment of three widely-spaced sample plots per county within the native range of
C. flarida was desired but this was not always possible due to lack of dense dogwood
stands in some counties.
Monitoring tree disease using GPS

Using a compass and hip chain, a 100m transect was laid out in suitably dense
stands of dogwood trees for establishment of long term GPS plots. The first ten trees
within one meter of either side of the transect having a minimum DBH (diameter at
breast height) of 3.8cm, were selected for disease monitoring. Preference was given to
live trees but recently killed trees in stands where mortality was occurring were also
utilized. A GeoExplorer 11 Global Positioning System (GPS) unit by Trimble (Trimble,
Sunnyvale, CA, USA) was used for marking the position of each tree in the transect. A
data dictionary was also created for the GPS system, which allowed for recording of
features of the target trees. Features included DBH, noting the presence or absence of
limb dieback, epicormic shoots, cankers and leaf spots. For each tree a minimum of 30
satellite positions were recorded for accuracy. Leaf and twig samples showing disease
symptoms were collected and placed in marked envelops from transect trees, in addition
to several other dogwood trees showing the best disease symptoms at the site. Transects
in forests with considerable canopy cover often contained a tree of a different species as
the first marked tree. This was done to allow the GPS unit to calibrate its location in order
to minimize error. Transect data was differentially corrected using Pathfinder Office
software (Trimble, Sunnyvale, CA, USA) and databases provided by the Basic Science

and Remote Sensing Institute (BSRSI), Michigan State University, East Lansing,

26

Michigan. A map showing transect placement was developed using ArcView GIS
software (ESRI, Redlands, CA, USA) (Figure 1).
Microscopic identification

Samples collected from infected sites were viewed under a dissecting microscope
to first determine the presence or absence of fungal fruiting bodies. When fruiting bodies
were present they were hand-sectioned and stained with an equal mixture of 95% ethanol
and 95% lactophenol cotton blue stain (Tuite 1969). Slides were examined at 1000X,
using phase contrast and oil immersion optics. This allowed for confirmation or rejection
of dogwood anthracnose based on conidia (6-12 x 2.5-4um) and conidiophore (10—18 x
2.5-4.5um) size and morphology, as previously described by Redlin (1991).
Nursery inspection for D. destructiva

In late April through early May in 2001, five Michigan nurseries were inspected
for dogwood anthracnose. Dormant trees without leaves, being shipped from nurseries in
Oregon, North Carolina, Alabama, Tennessee and Ohio were examined. C. flarida and
Camus kausa Hance were inspected, however hybrids (C. flarida x C. kausa) were not
included in the survey. C. kausa was included as it is partially resistant to D. destructiva
but is believed to be a carrier. Trees were inspected for dead twigs showing symptomatic
cankers and/or fruiting bodies, often with the aid of a 10X hand lens. Distinct groups of
dogwoods designated first by cultivar type and second by the nursery size class, shipped
from one location, were sampled. These groups were utilized in order to look at trends of
tree size and origin, in relation to the amount of disease that might be imported from
states and nurseries. An average of ten trees were sampled per group, though this was

dictated by what was being ordered by the Michigan nurseries. Twigs were placed into

27

    
         

.
C-OIOOOOOCUOOO...
O

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

location of all 52 health monitoring transects including the 11 transects infected with

Figure 1. Map of Lower Michigan showing counties containing Camusflarida trees,
Discula destructiva, and one infected site without a transect.

- Counties with Camus flarida

-

thout Carnusflorida

ICS W1

+ Infected site without transect

Infected transects

28

Uninfected transects

envelopes for each individual tree, allowing a tree—by-tree diagnosis in the lab. Samples
were examined microscopically in the lab as previously described and infection totals
were calculated.

Inspection of remaining unsold C. flarida trees in nurseries were carried out in
July through August of 1997 to check for leaf spots in order to observe if D. destructiva
was spreading throughout the nursery.
Fungal culture

Twigs with fruiting bodies from forest plots were placed on a glass slide in moist
chambers at approximately 22-25°C, with 10/14 hrs light/dark, for 2-3 days (Britton
1989, Hibben and Daughtrey 1988). Moist masses of conidia that formed on individual
acervuli in the moist chamber were picked up using the tip of a sterile scalpel blade and
transferred to potato dextrose agar containing 150ppm streptomycin sulfate. Discula
cultures received from other sources and isolates recovered from Michigan forest samples
are given in Table 1.
DNA isolation

Genomic DNA was isolated from cultured mycelia, in addition to twigs collected
from both nursery and forest plots. Extraction was performed using a
cetyltrimethylammonium bromide (CTAB) extraction buffer following the methods of
Jasalavich et al. (2000), Taylor et al. (1993), and Wilson et al. (1987). For extraction
from mycelial cultures, mycelia was scraped from petri plates and ground in liquid
nitrogen with a mortar and pestle. Approximately 200 to 300p] of ground mycelium was

then transferred to a sterile 1.5m] microfuge tube.

29

.Eqfiflmfl—z Co 3653:: 4:5 consumed 5:355 333E .2262 mm<-<om3 amen? 225 um _

 

- NE .5. .8 28. E0 .8 mecca
- m: =2 :00 commons: worm do 2205
+ mm... :2 :00 88538. 9.5330 Stuntman 328.5
+ v: =2 ..oo :95 cm> cmSm cm> «.285me «.385
+ m... =2 :00 some? cmmo=< méogwon 3:85
+ map :2 :00 332. .am €82. am $533.8 «.335
+ 2 =2 :00 :95 cm> co>mI .mx méogmmu «.335
+ $2 Fcase‘s 223 moowm $5an $385
+ $2 Fe963 E>mo Boom 335$ $38.5
- $2 .333 225 59522 3805 osmoquo 3:85
- $2 romeo; 250 26:01 mEm____>> 33388 @385
+ $2 $3nt 223 8 E «58$an $38.5
+ $2 cameo; Ema mun 38‘:me 338.5
-1. «to x men .3632 «0025:. 3:332:23qu 8.20m. 83on

 

@050on méoaamob $38.5 do“. «no
new “:5 90.55 mun. Co 9st use 2058.93 05 5 now: 35:3 395... ._. 63a...

30

For extraction from twigs, samples with fruiting bodies were observed under a
dissecting scope at 10X magnification and a length of twig, approximately 1 to 2cm in
length, was selected to be ground in a mortar and pestle using liquid nitrogen.
Approximately 200 to 300ul of ground twig material was added to a 1.5m1 microfuge
tube.

For fungal extraction, 2X CTAB extraction buffer was used, as compared to 1X
extraction buffer for woody material (Jasalavich et al. 2000). Extraction of DNA then
proceeded according to Jasalavich et a1. (2000). After alcohol precipitation, samples were
incubated at —20°C for 30 minutes to an hour, before centrifugation. DNA pellets were
brought up in 50p] of PCR-grade sterile H20. A single ten-fold dilution was made of all
DNA samples prior to PCR.

PCR reaction conditions

PCR was performed using 25 ul reactions containing various prepackaged buffers
and polymerases. Primers lTSl F (5' CTTGGTCATTTAGAGGAAGTAA 3') (Gardes and
Bruns 1993) and ITS4 (3' CGTATAGTTATTCGCCTCCT 5') (White et a1. 1990) were
used to amplify the internal transcribed spacer (ITSI, ITSII and the 5.88 ribosomal gene)
for use in sequencing. Approximately 10 to 50ng of DNA template were used per PCR
reaction with ITSIF and ITS4 primers. For amplification using species-specific primers,
lpl of the extracted genomic DNA solution was added to the reaction mixture, with both
undiluted and diluted extractions being tested for the presence of D. destructiva. PCR
reactions were conducted using a DNA thermal cycler (Perkin-Elmer, Norwalk, CT,
USA). Thermal cycling conditions consisted of an initial denaturation (94°C, 4 min), 35

cycles of denaturation (94°C, 1 min), annealing (54°C for ITS 1 F and ITS4, 57°C for

31

species-specific primers, 1 min), and extension (72°C, 1 min 30 s), followed by one final
cycle of extension (72°C, 10 min), then storage at 4°C. All extracted samples were tested
twice using species-species PCR primers, and the appropriate negative and positive
controls were included. PCR products were resolved on 1.5% agarose gels containing
100ng/ml of ethidium bromide and lKB plus molecular markers (Invitrogen, Carlsbad,
CA, USA), which were viewed under UV light.
Sequencing and primer design

The ITSI and ITSII region of D. destructiva isolate #23 was amplified using
ITSlF and ITS4, as noted above. Dye terminated capillary electrophoresis on an ABI
Prism 3700 DNA Analyzer (Applied Biosystems, Forest City, CA, USA) at the Michigan
State University Genomics Technology Support Facility was used to sequence PCR
products. Sequence results were confirmed as D. destructiva by alignment to other D.
destructiva ITS sequences published on GenBank using BLAST (Altschul et al. 1990).
From the sequence of D. destructiva isolate #23, species-specific forward primer DdF (5’
GGTGCTACCCAGAAACCCATTG 3’) and reverse primer DdR (3’
CTGATCACCCTGAAGAACGGCA 5’) were generated using the PrimerSelect 3.05
module of DNAStar (DNAStar, Madison, WI, USA). Primers DdF and DdR were then
used to amplify a region of D. destructiva DNA, which was then sequenced. This PCR
product was also rim through the BLAST search engine to confirm it was a match to
existing D. destructiva sequences.
Primer specificity

Primer specificity was tested against DNA extracted from other species of

Discula and an unidentified species of Phama that was routinely cultured from infected

32

plant material (Table 1). In addition, primers DdF and DdR were tested against D.
destructiva cultured from suspected infection sites in Michigan (Table 1). To ensure that
primers DdF and DdR were not selectively amplifying extracted plant DNA from
minipreps, woody extracts from both live and dead uninfected dogwood twigs were
tested using universal nuclear small primers NS7 (5'
GAGGCAATAACAGGTCTGTGATGC 3') and N88 (3'
AGGCATCCACTTGGACGCCT 5') (White et al. 1990, Gardes and Bruns 1993).
Primer sensitivity

Sensitivity of primers DdF and DdR to D. destructiva was evaluated by two
methods. The first test conducted examined the amount of acervuli growing on woody
twigs that could allow a positive PCR reaction. Fruiting bodies were miniprepped in
varying quantities from twigs collected from transect 5 in Allegan County, Michigan.
These samples were positively identified by microsc0pic means and cultures as D.
destructiva. Quantities of acervuli tested included: 110, 80, 60, 40, 30, 20, 15 and 10 per
miniprep.

The second test for primer sensitivity was performed by making a progressive 10-
fold dilution of DNA extracted from woody twigs. Dilution concentrations tested ranged
from 100ng down to lOOpg and were quantified with a Beckman DU 7400
spectrOphotometer (Beckman Coulter, Fullerton, CA, USA).

Testing Michigan forest and nursery samples with Discula primers

Twig samples from 13 forest sites in Michigan with fruiting bodies were

miniprepped as described above. A ten-fold dilution of the extracted DNA was made, and

both undiluted and diluted DNA was tested using primers DdF and DdR.

33

Nursery samples collected in 2001, which were examined microscopically were
also used in a comparative study utilizing the PCR primers DdF and DdR. As described
above, DNA was extracted from twigs of individual tree samples and a single 10-fold
dilution of the extracted DNA was performed. Both diluted and undiluted extraction
products were tested by PCR. In addition, rDNA universal primers NS7 and N88 were
tested to check for DNA extract viability (White et al. 1990, Gardes and Bruns 1993).

RESULTS
Development of forest health monitoring plots in Michigan

The responses gathered from the dogwood questionnaires enabled the scouting
and location of stands containing significant densities of flowering dogwood. A total of
36 counties in Michigan were identified as containing C. flarida based on our
observations and V083 (V053 1985) (Figure 1). The northern most stands of C. flarida
located in this survey were in Mason County, although individual plants may survive
further north. In addition, several dogwood stands that were believed to be infected were
identified. Health monitoring plots containing 100m GPS generated transects were set up
throughout the range of flowering dogwood in Michigan at sites meeting transect criteria.
A total of 52 GPS health monitoring transects, covering 26 counties, were established
(Table 2, Figure 1).

Inspection of Michigan nurseries for dogwood anthracnose

The inspection of imported C. flarida and C. kausa from five Michigan nurseries
in 2001 showed that trees were being imported from a total of ten different out of state
nurseries. Our method of sample collection produced a total of 17 sample sets, consisting

of 15 sets of various C. flarida cultivars and two sets of C. kausa. A total of 342 trees

34

were examined, with samples being collected from 209 of the trees. The other 133 trees
examined did not possess dead or diseased twigs for collection (Table 3).

Inspections in 1997 confirmed that dogwood anthracnose was spreading within
nurseries to unsold C. flarida and C. kausa trees. Four nurseries were found to have both
infected C. flarida and C. kausa trees showing leaf spot symptoms and acervuli of D.
destructiva.

Microscopic analysis of nursery samples

Microscopic analysis of twigs from all 209 trees containing samples yielded a
total of 35 trees, which were identified as infected with D. destructiva based on
morphology of conidia and conidiophores. This nursery stock gave a 10.23% rate of
infection for all 342 trees examined. Of the 35 trees diagnosed as infected, 33 were
various cultivars of C. flarida, and two were C. kausa. Both groups of C. kausa had trees
identified as infected. Fourteen of the 17 sample sets were found to have at least one tree
identified as having D. destructiva. All five states included in this study were exporting
infected flowering dogwood trees to Michigan. Eight of the 10 nurseries from the five
states were transporting infected dogwoods to Michigan (Table 3, Figure 2).
Species-specific PCR primer design and specificity

Amflification of the ITSI, 5.8S and ITSII region of D. destructiva isolate #23 for
initial sequencing gave a product 579 bp in length, which was confirmed as the correct
species through matches with existing ITS sequences of D. destructiva available on
GenBank. Primers DdF and DdR produced a 460 bp fragment consisting of part of the
ITSI, 5.83 and ITSII region of DNA amplified from D. destructiva Michigan isolate

Galesburg, which also was a 100% match against D. destructiva sequences on GenBank.

35

Table 2. Location of dogwood health monitoring plots in Michigan by county
and geographic coordinates developed using global positioning systems, and
the presence or absence of dogwood anthracnose.

 

 

Discula
Transect destructiva

Location (County) number Latitude Longitude positive 9%)
Berrien T1 42°00'23.4" 86°32'05. 2" -
Berrien T2 41 °54'58. 5" 86°27'15.0" -
Van Buren T3 42°21'21.6" 85°51'20.9" +
Van Buren T4 42°07'46.0" 85°56'42.0" -
Allegan T5 42°33'04.4" 85°59'41 .3" +
Allegan T6 42°39'18.7" 86°00'26. 1" -
Allegan T7 42°35'15.0" 86°1 2'44. 5" -
Kent T8 42°55' 1 6. 9" 85°42'30. 1" -
Ottawa T9 43°00'59.0" 86°05'35. 5" -
Kent T10 43°10'50.6" 85°30'42. 1" -
Ottawa T1 1 42°48'02.3" 86°07'36.4" -
Kent T12 43°13'07.6" 85°26'06.2" -
Ottawa T13 43°05'57. 1" 86°10'38.4" -
Muskegon T14 43°17'50.2" 86°07'34.7" -
Muskegon T15 43°27'38.6" 86°16'53.2" -
Eaton T16 42°31 '1 9.6" 84°39'49.7" -
Montcalm T17 43°10'20.7" 85°05'00.5" -
Montcalm T18 43°10'20.4" 85°04'59.9" -
Eaton T19 42°26'22.4" 84°58'46.7" -
Barty T20 42°37'31.4" 85°30'44.7“ -
Calhoun T21 42°26'27. 8" 85°1 0'52.5" -
Kalamazoo T22 42°22'01 .9" 85°31 '53. 1 " +
Kalamazoo T23 42°26'26.9" 85°34'1 5. 7" +
Kalamazoo T24 42°22'06.8" 85°21‘08.0" +
Jackson T25 42°12'19.7" 84°14'51 .5" -
St. Joseph T26 41 °56'27.8" 85°41 '02.5" +
Branch T27 41 °48'05. 0" 84°56'56. 7" -
Branch T28 41°51 '24.5" 85°03'30. 0" -
Branch T29 41 °51 '23. 5" 85°03'31.2" -
St. Joseph T30 41 °50'50. 9" 85°2 7'1 8. 5" -
lngham T31 42°42'50.6" 84°22'25. 8" -
lngham T32 42°33'23. 3" 84°22'02. 7" -
Jackson T33 42°1 9'28. 3" 84°07'41 .5" -
Washtenaw T34 42°17'56.1" 83°45'38.4" -
Jackson T35 42°06'41 .8" 84°35'36.9" -
Genesee T36 43°05'1 5. 3" 83°51 '36. 1" -
Oakland T37 42°31 '49.0" 83°39'57.6" -

36

Table 2 (can’t).

Livingston
Wayne
Wayne
Lapeer
Oakland
Lenawee
Van Buren
Cass
HiIIsdaIe
Kent
Muskegon
Kent
Mason
Kalamazoo
Clinton

T38
T39
T40
T41
T42
T43
T44
T45
T46
T47
T48
T49
T50
T51
T52

42°45'22.5"
42°10'22.6"
42°06'21 .2"
43°07'27s"
42°52'17.2"
42°02'04.5"
42°04'29.7"
41°57'4o.4"
41°54'21 .5"
42°56'33.5“
43°1 104.5"
43°02'21.9"
43°52'02.3"
42°18'00.4"
42°48'45.2"

37

83°43'00. 5"
83°2 5'24. 9"
83°1 8'42.6"
83°14'59.0"
83°26'30. 1"
84°06'49. 9"
86°08'59. 8"
86°03'52. 3"
84°34'1 9. 2"
85°34'39.6"
86°1 6'57.6"
85°2 9'30.6"
86°08'24.4"
85°25'04.8"
84°22'38. 7"

.95? 0. .0300 .0 :05 00.005 .0 5:928:00 m .m <20 8.00:: $:8.5 .0 08890 0.... 0.86:. 9.80. 02.89“. a

89.3an0 b953,: mun. .0 0.8. new <20 80.03: Eobmoam .0 00:80:. a... 0.8.0:. 8:80. 02.88 N

02.03800 .5 .0. 30.9.99: cam 08 0.0.5930 .0 $0600 0cm 802.020 9.39. 80.058 02:83. .

 

 

 

 

8.0 an. E 3.3 5208:. .x.

3 E. .. mm 8.4. New 0.20...
o w v o 9. 0-0 8005.... 09.9050 00:0... .0 2L. 9
o F o F ON 0.: .030 030.050 00.00.. .0
o N o m 0.. .90 .020 09.00050 00.00... .0 z... m
o N F F N N 0-0 902 $0.20.. .0 ..< m
N m r 0 or Do 6.0. £030 8:9. .0
o of N or B 3-0 9.8: $0.00.. .0
o m N m or .m.\. 923 $0.00.. .0 IO N
o m P m m .0... £2220 8:9. .0 IO 0
v v F v v 07¢ .95 09.9050 $000... .0
m m m E Z .2. m 0.5.0 00.20.. .0
F 2 m ON FN 0-0 052 00.20.. .0
m m N m m 50 0.85.60 09.0.9.0 00.000 .0 «.0 m
o N v N E .9-.. 05:. 00.00... .0
0 ON N N mm .38 a 0:0.0 0000.. .0 02 v
o 3 v 3 FN NT? 052 00.20.. .0 02 m
o 9 v N 0N .o Tm 800:5 09.0.05 00.00.. .0 02 N
o N. o N. or .N... .020 09.905 $0.00.. .0 02 F

”=5 05 «02 20056.0 00.0808 (20 005588 3.0 823:0 00.8mm 38m 0 E852

“:5 208.... 0:0 32 0.05:0 0.008225 >0 0:0 0302.00 000... .80...
5.; «0.. 3 5.; men. .3 .2538 8o: 8358
.”25:00.. 002... so>§non 000; 5.3 002...

 

00:00.00 .08.... mon. 08 5080.25.98: 0:...03800
$805 .0. .EON c. c0952.). 0. 00:09.0 800.. E093: 8:9. 8:260 08 $0.00: 8:500 .0 20:88:. .n 0.00...

38

 

1' Ill! I'llIl IIlll'lI

00.0.56 9.00: «0500 *

.IIIIII I I

 

05.0.. 00 5.52

 

 

. «525000 0.3005 5.? www.mi - 00.0900 00.2... .80... I

08005.0... 0.5302 0.:0
. o
I: I I. I I I. ~ IN I. I I I. In I.
.{r

I III .1, I I I ..... I oF

II I I I II I I I.
f I I II I III I- I I I I I I I III IIIII I II I III ON
I I I I I I I I. I II I I I I II I. I I . 3
I I I I I I II I I I I I I I I on
I I I I II I I I I I I I I I I I I . mm
III I I I I I II I I I I I I I I I I .I ov
IIIIIIIIIII III II I I I I I I I I II III II II: .Imv
II-IIIIIII-IIIIIII III I II III I II I I II I I II II I I .I. I- I. on

 

 

 

mm

 

0.00.00 0:: 3:03: .00 on? 0:: 30.00908 60:08:: 0.0 :0 0080 69:03:80 0300.5
£05 08000:. 8 bfiofioomobmfi 00:00:00. .83.. E 800 E83: :0.:0~\§500 00:09:. .00 80:5: 05 3.30% 3.33. .N «Bag

9001; go JaqumN

39

Specificity of PCR primers DdF and DdR were tested against extracts of hyphae
of D. destructiva isolated in Michigan, other species of Discula, and an unidentified
Phoma sp. that was ubiquitous in the twigs (Table 1). Primers DdF and DdR did not
amplify Discula campestris (Pass) Anx. and the unknown Phoma sp. (Figure 3).
Amplification of a 460 bp fragment of DNA did occur for all D. destructiva isolates
(Figure 3 and 4), in addition to both isolates of Disculafiaxinae (Peck) Redlin and Stack
(Figure 3).

Testing of PCR primer set NS7 and N88 against plant DNA from uninfected live
and dead twig extracts gave positive bands approximately 375 bp in length. No
amplification of plant DNA from the same uninfected live and dead twig extracts
occurred in the presence of primer set DdF and DdR (Figure 5).

PCR primer sensitivity

PCR amplification using DdF and DdR with DNA extracted from decreasing
quantities of acervuli produced bright bands visible down to 40 acervuli. Faint bands
showed detection of D. destructiva as low as 15 acervuli but no detection occurred at 10
acervuli (Figure 5).

A ten-fold serial dilution of DNA from a woody twig infected with dogwood
anthracnose, amplified with PCR primers DdF and DdR. produced visible bands.
Amplification of fungal DNA was visible at the 100ng and lOng concentration, however
no bands were detected at the lug and lOOpg level (Figure 5).

PCR analysis Michigan forest twig samples and nursery samples

Symptomatic twig samples from a total of 13 Michigan forest sites from 7

different counties were collected (Table 4. Figure 1). Only 11 of these sites contained

40

transects as the other two located at Fort Custer Training Center were only noted as
infected (Figure 1). Testing of total extracted DNA from symptomatic forest twigs using
primers DdF and DdR yielded visible bands for all samples (Figure 4). PCR amplification
bands were of the same size as those amplified from cultures obtained from the same
forest samples and D. destructiva isolate #23 (Figure 4).

DNA was extracted from twig samples of all 209 nursery trees analyzed
microscopically and was used in PCR-based detection using the species-specific primers
DdF and DdR. Positive PCR results, indicating the presence of D. destructiva, were
obtained for 15 trees of the 209 possible (Table 3, Figure 6). This yielded an overall
infection rate of 4.39% for all 342 nursery trees examined. PCR diagnosis showed that
two of the 10 nurseries, one each in Oregon and Ohio, were found to be exporting
infected dogwood trees to Michigan (Table 3). Of the 15 trees that tested positive for D.
destructiva, 13 were from C. florida received from nursery #5 in Oregon. Two C. kausa
from nursery #7 in Ohio also tested positive for D. destructiva (Table 3). Figure 7A
shows the PCR test results of C. florida cultivar Cloud 9, trees 1 to 8 from nursery #5,
Oregon. Trees 1, 2, 4 and 5 all tested positive for D. destructiva by primers DdF and
DdR. Figure 7B shows amplification of all minipreps of the same sample set using
universal primers NS7 and N88. Overall success of DNA extraction from twigs was
81.34% when checked with NS7 and N88 (Table 3).

Ten of the 15 trees yielding positive PCR results coincided with the diagnosis
from the microscopic study (Figure 8). The other five trees that expressed a positive PCR
reaction were not identified as having D. destructiva based on initial microsc0pic

examination of fruiting bodies.

41

M12 3 4 5 6 7 89101112131415161718M

Figure 3. PCR amplification of nuclear rDNA from total DNA isolated from pure
cultures of various Discula species and a unknown Phoma sp. Electrophoresis in
1.5% (wt/vol) agarose in lx TAE. The two outer lanes contain IKB plus molecular
markers (Invitrogen). Odd numbered lanes contain samples amplified using universal
primers lTSlF and lTS4, and even numbered lanes contain samples amplified using
primers DdF and DdR. Lanes 1 and 2, Discula destructiva #23; lanes 3 and 4, Discula
destructiva #100; lanes 5 and 6, Discula campesm's Williams Hollow; lanes 7 and 8,
Discula campestris Blessed Mountain; lanes 9 and 10, Discula fraxinae 9600l; lanes
ll and 12, Disculafraxinae 96003; lanes 13 and I4, Phoma sp. BlO-t; lanes 15 and
16, Phoma sp. GRl; lanes l7 and l8, no template DNA (i.e. negative controls).

42

 

bp

HH’I

4— L650

850
650
500
400
300

 

Figure 4. PCR amplification of rDNA from total DNA extracted from Michigan
cultures of Discula destructiva and woody twig extracts, both from symptomatic
forest samples. Electrophoresis in 1.5% (wt/vol) agarose in 1x TAE. The two outer
lanes contain lKB plus molecular markers (lnvitrogen). Lanes 1 to 18 contain
samples amplified using primers DdF and DdR. Lanes 1 to S, cultured Discula
destructiva; lanes 6 to 16, woody extracts. Lane 1, T3; lane 2, T26; lane 3, T5; lane 4,
T44; lane 5, T52; lane 6, T3; lane 7, T26; lane 8, T5; lane 9, T44; lane 10, T52; lane
ll, T24; lane 12, T48; lane 13, T47; lane 14, FCTC 25; lane l5, FCTC 40; lane 16,
T22. Lane 17, Discula destructiva #23 (i.e. positive control); lane 18, no DNA
template (i.e. negative control).

43

«1,650

850
650
500
400
300

ttttt

M l 2 3 4 5 8 9101112131415161718M

 

Figure 5. PCR products of rDNA, testing sensitivity of primers DdF and DdR against
DNA extraction of infected and uninfected woody material. Electrophoresis in 1.5%
(wt/vol) agarose in lx TAE. The two outer lanes contain lKB plus (lnvitrogen)
molecular marker. Lanes l to 3, samples amplified with universal primers NS7 and
NS8. Lane 1 and 2 contain uninfected live and dead twig samples, respectively; lane 3,
no DNA template (i.e. negative control). Lanes 4 to 18 contain samples amplified with
primers DdF and DdR. Lanes 4 and 5, live and dead uninfected twig samples,
respectively; lane 6 to 13, extraction of decreasing number of acervuli; lane 6, 110
acervuli; lane 7, 80 acervuli; lane 8, 6O acervuli; lane 9, 40 acervuli; lane 10, 30
acervuli; lane ll, 20 acervuli; lane 12, IS acervuli; lane l3, l0 acervuli. Lanes 14 to
17, progressive ten-fold dilutions (100ng, lOng, lng and lOOpg) of woody extract
containing Discula destructiva. Lane 18, no DNA template (i.e. negative control).

44

bp

  

Table 4. Michigan forest sites identified as infected with Discula destructiva from symptomatic

twigs, using microscopy and PCRJJrimers DdF and DdR.

1:21am
Allegan

C45

Area 51
Benner
FCTC 25
FCTC 40
Galesburg
Kal23

Grand Rapids
Muskegon

St. Joseph
Kal Haven
Van Buren

Location by County

Allegan
Cass
Kalamazoo
Kalamazoo
Kalamazoo
Kalamazoo
Kalamazoo
Kalamazoo
Kent
Muskegon
St. Joseph
Van Buren
Van Buren

45

Tranggcg Number Year Infggtgd

T5
T45
T24
T22
none

none
T52
T23
T47
T48
T26

T3

T44

2000
1997
1998
1993
1999
1999
1999
1995
1995
1995
1999
1997
1995

00.950 8:9. 0:500 .1 050.5000 0.I:.00.a :0: 05.000 000:... .m. 090an 000;. .80... I-

0080::0... 050002 020 :0005 0:20:00 5:02

 

 

 

 

I o

I m

Io_.

Imp

.om

Ina

.om

 

seer; )0 JeqwnN

mm

 

I ov

...I ...I-.I mv

 

-I-.I1I..,om

 

 

 

mm

0:0. 0:
.850 :0 :05 :800:w 8008:0050 0 :0 <75 30.00: 030:5 00 00:80:: 05 :0 0003 £03 0:: 009 808:: Mon. mam: 08:03:00.:

5:005 5.3 0808:. 00 000.300. . SON :. 000:: D080: :0.:0~\:.:E0U 08:00:: 00 00:3: 05 macaw 8.000% .0 0.5»...—

46

bp

M12 3 4 56 7 8 9101112131415161718M

f .

    

1,650

850
650
500
400
300

M123456789101112131415161718M bp

060 . 0&0 . .83 0

 

Figure 7. PCR amplification of dogwood woody extracts from nursery samples.
Nursery sample set: Cornusfloria’a var. cloud 9, from nursery #5, Oregon, individual
trees 1 to 8 tested. Electrophoresis in 1.5% (wt/vol) agarose in lx TAE. Amplification
using primer combinations DdF and DdR (A), and NS7 and NS8 (B). The two outer
lanes contain ”(8 plus (Invitrogen) molecular markers. For both gels, odd numbered
lanes contain undiluted samples and even numbered lanes contain 10 fold dilutions.
For both gels (A & B) lanes 1 and 2, tree 1; lanes 3 and 4, tree 2; lanes 5 and 6, tree 3;
lanes 7 and 8, tree 4; lanes 9 and 10, tree 5; lanes 11 and 12, tree 6; lanes l3 and 14,
tree 7; lanes 15 and 16, tree 8; lane 17, Discula destructiva #23 DNA template (i.e.
positive control); lane 18, no template DNA (i.e. negative control).

47

 

00.:E0m 00:8. 0:500 ¥ M. l; 02:00: mo: i i

00805.8. 050002 020

cc cc o

   

 

 

 

$00.11 go iaqumN

l -- l l -,- l llll‘ l l.l;ll ll -7 lllL: l.li|r 0

.88 5 000:: E005:
00::0~\§=:0U 08:09:: 80:: 63:05:00: 0300.5 :0 00:08:: 05 :0: 03:00: $300: M0: 0:: 2:080:28 :0 80:39:00 .a 0.5»:—

48

DISCUSSION

The use of species—specific PCR primers for diagnosing plant disease has been
well documented (Langrell 2002, Faggian et al. 1999, Kageyama 1997). In particular, this
method of detection is valuable when dealing with highly destructive pathogens, such as
Spongospora subterranea (powdery scab of potato) and Phytophthora ramorum (sudden
oak death) that are at risk of spreading further through their host range, either by natural
means or commercial movement (Bulman and Marshall 1998, Kox et a1. 2002).

With the eradication of flowering dogwood by D. destructiva occurring
throughout much of its native range, species-specific PCR primers may be useful in
aiding identification. In Michigan there has been only one county identified in 1993 as
infected (Daughtrey et al. 1996). The identification of new infection sites and
development of plans to protect the remaining stands of flowering dogwood could be
enhanced by the deveIOpment of these primers. One particular application would be
improved screening of imported dogwood nursery stock from infected states, which are
sold in Michigan for landscaping.

The internal transcribed spacer (ITS) region of the rDNA has been utilized
extensively by other researchers for developing species-specific PCR primers, due to it
often being conserved within species but not between species (White et al. 1990, Gardes
and Burns 1993). In this study the fungal specific forward primer ITS 1 F (Gardes and
Burns 1993) and the reverse primer ITS4 (White et al. 1990), were used to successfully
amplify the ITSI, 5.88 and ITSII regions of D. destructiva isolate #23 and Michigan
isolate Galesburg. The sequences of both isolates matched D. destructiva ITS sequences

from GenBank. The design of species—specific primers from these sequences yielded two

49

optimum primers, DdF and DdR. Primer DdF is downstream of the 188 rDNA and DdR
is upstream of the 28S rDNA, resulting in a PCR product 460 bp in length.

Testing against D. campestris and D. fiaxinae revealed that primers DdF and DdR
were not truly ‘species-specific,’ although they did not amplify DNA of D. campestris,
they did amplify a fragment of D. fraxinae DNA, producing a product similar in size to
that of D. destructiva. The absence of species—specificity within genera can be attributed
to a lack of heterogeneity of the ITS region being used to develop the primers. This has

been the case for pathogens such as the beech bark disease causing complex, where early

 

primers were unable to distinguish between Neonectria coccinea, N. coccinea var.
faginata and N. galligena (Brown et a1. 1993, Langrell 2002). In the case of dogwood
anthracnose where there is only one primary pathogen, the amplification of D. fraxinae
should not be cause for concern due to host specificity. Primer specificity against an
ubiquitous Phoma sp. that routinely was isolated from diseased dogwood samples in
Michigan was deemed more important. In addition, tests of DdF and DdR in conjunction
with extracted uninfected dogwood DNA confirmed that amplification of plant DNA did
not occur. This is an important consideration with any species-specific primer being used
in disease diagnosis from plant material in order to prevent false-positive reactions
(Zhang et al. 1997). The positive amplification of uninfected dogwood trees by targeting
the nuclear small (188) region of rDNA from uninfected dogwood extracts, using primers
NS7 and NS8, was a necessary check to ensure that DNA extractions from twigs yielded
quality DNA without contaminating substances that inhibit PCR amplification reactions
(Kim et al. 1999, Gardes and Bruns 1993). PCR-inhibiting substances were present in

DNA extracts of dogwood twigs when several standard extraction methods were used

50

(Taylor et al. 1993, Catal et al. 2001) and only the Jasalavich procedure (2000) was
effective in this study.

The ability to quantify primer sensitivity is another important aspect when using
species-specific primers for plant diagnostics. Researchers have gone about quantifying
primer sensitivity in various ways. One method that has been utilized is to simply make a
dilution series of template DNA from the targeted fungal pathogen (Glen et al. 2001,

F aggian et al. 1999). Another method utilized has been extracting various amounts of
infected plant material in order to determine the minimum amount needed to generate a
positive PCR reaction (Jasalavich et al. 2000, Faggian et al. 1999). Both of these methods
were employed for this study. DNA template extracted from heavily infected dogwood
twigs was quantified with a spectrophotometer, and then subjected to a progressive 10-
fold dilution series. This method allowed for detection in the range of 100ng to 10mg of
DNA. The number of acervuli needed to elicit a positive reaction was also determined.
Primers DdF and DdR produced bright bands from 110 acervuli to 40 acervuli, and
fainter bands down to 15 acervuli. Despite not knowing the exact amount of fungal
material being tested for, this is still seen as a valid test since fruiting bodies of D.
destructiva are being sought in nursery and forest inspections in the absence of trunk
cankers or leaf symptoms.

Dogwood anthracnose has moved largely unhindered through much of the eastern
states with devastating consequences (Daughtrey et al. 1996). For the states in the
western range of C. florida, patches of dogwood trees are isolated due to large expanses
of agricultural land and suburban development. In Michigan only one site in Kalamazoo

County was initially identified in 1993 as being infected with D. destructiva (Daughtrey

51

 

 

et al. 1996). No further studies were conducted to determine if the pathogen was
spreading to the various isolated stands of native dogwood around the state until 1997.
The importance of flowering dogwood in the ecosystem and for Michigan’s ornamental
nursery industry cannot be understated. A total of 52 permanent health-monitoring plots
were placed in 26 counties throughout the range of C. florida in Michigan utilizing GPS
technology, for the future monitoring and studies of disease impacts. For this study a total
of three forest transects were sought per county. However in many counties this was not
possible due to the lack of dogwood stands with densities sufficient for a minimum of ten
trees intersecting the 100m transect because of the fragmented nature of Michigan’s
forests. More extensive surveys and transects were established in Kalamazoo County due
to disease incidence and spread. This also allowed us to observe the initial infection site
(Transect 22) as an infection epicenter, in order to estimate the rate in which the pathogen
was spreading outward. The timeline as to when other 12 infections in the state occurred
is somewhat arbitrary as it was necessary to rely on property owner memories and
impressions, and our own estimate following inspections. Assuming that transect 22 in
Kalamazoo County was a single and isolated introduction of the pathogen, then disease
spread from the initial site to the Fort Custer Training Center could be estimated at
0.53km per year. Transects T47 and T48 are important to note as communication with the
homeowners at both sites strongly suggested that C. florida trees from nurseries planted
in close proximity to native dogwood stands initiated infections, as opposed to natural
spread of the pathogen.

Much of the spread of dogwood anthracnose in the eastern United States has been

through natural means, such as short distance rain splash (Daughtrey et al. 1988).

52

 

Dissemination of dogwood anthracnose into native stands by the movement of infected
nursery stock has been suggested (Daughtrey et al. 1996). Until this study no inspections
of newly arrived dogwood stock from infected states had been done to determine if the
nursery trade was escalating the spread of this disease through Michigan forests.

The inspection of five Michigan nurseries was the first attempting to quantify the
number of infected trees that were being imported from other states. For this study C.
florida and C. kausa were examined since they are the primary cultivars being sold for
landscaping. Hybrid cultivars of C. florida were not used for the study as some show
varying levels of resistance to D. destructiva, which could prove difficult to detect if the
morphology of the symptoms and signs changed as a result of the crossing (Ranney et al.
1995, Windham et al. 1998). However, it is important to keep in mind that while hybrids
and the Chinese flowering dogwood C. kousa may not be as susceptible to dogwood
anthracnose, they may act as carriers of the disease. C. kausa was examined despite this
as it possible that D. destructiva originated with it, and was heralded as a replacement
tree for C. florida in landscaping (Brown et al. 1996).

Another important aspect taken into account for this study was the timing of the
inspections. Examination of imported trees was done as soon as they arrived at the
nursery, and only newly imported trees were inspected. This was important since
excessive movement of the dogwood trees may have broken off the dead twigs that were
being collected. In addition, trees were sampled immediately incase the nursery contained
D. destructiva infected holdovers from the previous season that could have spread the

pathogen to the newly imported trees once they were placed in the same growing beds.

53

Many of the trees examined from the smaller size classes were devoid of twigs, either
alive or dead, which created sample sets with very few twigs for examination.

Microscopic diagnosis established that eight of the 10 out-of-state nurseries were
exporting infected dogwood trees to Michigan and that 10.23% of all trees examined
were infected. Discula-specific PCR primers established that at least 4.39% of trees were
infected. All five states included in the survey are infected with dogwood anthracnose but
no quarantine exists for controlling dogwood anthracnose and regular inspections are
limited (Daughtrey et al. 1996). The infection rates are relatively low, but it only takes
one infected tree to spread the pathogen to other healthy trees within the nursery.
Suburban sprawl continues to spread into forested areas, bringing with it the possibility of
planting an infected dogwood tree in close proximity of native stands. This type of action
has quite likely occurred in Michigan in at least three sites already (T22, T47 and T48),
where homeowners planted nursery flowering dogwood trees. Within a couple of years of
planting the imported dogwood, homeowners reported to us that they noticed a marked
decline in tree vigor, followed by a similar decline in the native dogwoods in the adjacent
woods.

Only 10 of the 15 trees that yielded positive results by PCR primer detection
coincided with the microscopic diagnosis. The differences between microscopic and PCR
diagnosis could have several explanations. From a micrOSCOpic perspective the five
samples that yielded a positive PCR reaction but were diagnosed as not being D.
destructiva could have been misidentified. Perhaps the wrong fruiting bodies were

selected for observation because many of the twigs collected had other fruiting bodies on

54

them from secondary pathogens. Also the D. destructiva acervuli could have already
exuded its spores, making accurate identification difficult.

There are several reasons from a molecular standpoint as to why the infection
percentages differed between microscopic and molecular diagnosis. The first possibility
is there may not have been enough fungal material (fruiting bodies) present on the
samples to elicit a positive reaction. With a minimum number of 15 acervuli needed for
detection, it is likely that some samples did not possess enough fungal material, yet could
allow for the spread of the pathogen. Additionally, the dead twigs may have been on the
tree for more than one year, allowing degradation of fungal DNA to occur.

One last possibility as to why primers DdF and DdR were not as sensitive as
microscopic methods could be due to genetic variation between isolates. While no studies
have been done on the ITS region of D. destructiva, there are studies that suggest some
genetic diversity overall is present in the population (Caetano-Anolles et al. 2001, Zhang
and Blackwell 2001, Zhang and Blackwell 2002). If genetic changes have occurred in the
ITS region of populations of D. destructiva over the past 30 years, it might have
prevented full detection by ITS primers DdF and DdR in this study.

While the sensitivity of primers DdF and DdR are somewhat low, PCR diagnosis
was valuable when used with suspect forest samples taken from transects during periods
of dry weather when other symptoms were difficult to spot. Other PCR protocols used in
disease diagnosis that may improve on this study include nested PCR (Hamelin 1996)
and magnetic capture hybridization (Jacobson 1995, Langrell and Barbara 2001).

The survey of Michigan’s flowering dogwood population shows that dogwood

anthracnose is a continued threat to the majority of the natural stands. Given enough time,

55

dogwood anthracnose will likely result in the high levels of mortality throughout the state
of Michigan, as observed in the eastern United States (Daughtrey et al. 1996, Hiers and
Evans 1997). Our study has shown that dogwood anthracnose has been spreading over
the last ten years in Michigan. How D. destructiva is spreading is difficult to determine.
On method could be from the movement of spores by insect vectors as seen with ladybird
beetles in a laboratory setting (Colby 1993, Colby et al. 1995, Colby et al. 1996, Hed et
al. 1999). However, since most stands of C. florida are highly fragmented in Michigan,
the likely scenario is the introduction of D. destructiva into new areas with infected
nursery plantings, followed by natural spread to contiguous stands. Spread of D.
destructiva within nurseries during the summer months to all C. florida and C. kausa

trees will allow for a continued increase in the number of infected trees that are being
planted throughout Michigan. Hopefully future studies utilizing the permanent GPS

forest health monitoring plots developed in this project will allow for continual

monitoring and research on the spread of dogwood anthracnose in Michigan in order to

conserve the flowering dogwood tree, C. florida.

56

LITERATURE CITED

Altschul, S. F., Gish, W., Miller, W., Myers, E. W., and Lipman, D. J. 1990. Basic local
alignment search tool. Journal of Molecular Biology 215: 403-410.

Anderson, J. B. and Stasovski, E. 1992. Molecular phylogeny of Northern Hemisphere
species of Armillaria. Mycologia 84:505-516.

Badenhop, B. B., Witte, W. T., and Glasgow, T. E. 1985. Production systems and costs
for producing balled and burlapped trees of dogwood cultivars. Tennessee, 1984. Univ. of
Term. Agric. Expt. Sta. Bul. 63 7.

Britton, K. O. 1989. Temperature, pH and free water effects on in vitro germination of
conidia of a Discula sp. isolated from dogwood anthracnose lesions. Phytopathology
79: 1203.

Britton, K. O. 1993. Anthracnose infection of dogwood seedlings exposed to natural
inoculum in western North Carolina. Plant Disease 77:34-37.

Britton, K. 0., Roncadori, R. W., and Hendrix, F. F. 1993. Isolation of Discula
destructiva and other fungi from seeds of dogwood trees. Plant Disease 77:1026-1028.

Brown, A. B., Windham, M. T., and Trigiano, R. N. 1996. Resistance to dogwood
anthracnose among Camus species. Journal of Aboriculture 22:83—85.

Brown, A. E., Muthumeenash, S., Sreenivasaprasad, 8., Mills, P. R., and Swinburne, T.
R. 1993. A PCR primer-specific to C ylindrocarpon heteronema for detection of the
pathogen in apple wood. FEMS Microbiology Letters 108: 117-120.

Bulman, S. R. and Marshall, J. W. 1998. Detection of Spongospora subterranea in potato
tuber lesions using the polymerase chain reaction (PCR). Plant Pathology 47: 759-766.

Byther, R. S. and Davidson, R. M., Jr. 1979. Dogwood anthracnose. Ornamental
Northwest Newletter 3(2):20-21.

Caetano-Anolles, G., Trigiano, R. N., and Windham, M. T. 2001. Patterns of evolution in
Discula fungi and the origin of dogwood anthracnose in North America, studied using
arbitrarily amplified and ribosomal DNA. Current Genetics 39:346-354.

Catal, M., Chastagner, G. A., and Adams, G. C. 2001. Detection, identification and
quantification of latent needlecast pathogens and endophytes in symptomless conifer
foliage by PCR and dot-blot assays. International Union of Forestry Research
Organizations, Helsinki, Finland. U.S.D.A. Forest Service General Technical Report
WO, vol. 61.

S7

Chen, W., Hoy, J. W., and Schneider, R. W. 1992. Species-specific polymorphisms in
transcribed ribosomal DNA of five Pythium species. Experimental Mycology 16:22—34.

Colby, D. M. 1993. Arthropod dissemination of Discula destructiva conidia on Camus
florida. M.S. thesis, University of Tennessee, Knoxville.

Colby, D. M., Windham, M. T., and Grant, J. F. 1995. Hippodamia convergens
(Coleoptera: Coccinellidae) dissemination of dogwood anthracnose fungus
(Melanconiales: Melanconiaceae). Environmental Ecology 24:1075-1079.

Colby, D. M., Windham, M. T., and Grant, J. F. 1996. Transportation and viability of
conidia of Discula destructiva on Hippodamia convergens. Plant Disease 80:804-805.

Dan, H., Ali-Khan, S. T., and Robb, J. 2001. Use of quantitative PCR diagnostics to
identify tolerance and resistance to Verticillium dahliae in potato. Plant Disease 8 5:700-
705.

Daughtrey, M. L. and Hibben, C. R. 1983. Lower branch dieback, a new disease of
northern dogwoods. (Abstr.) Phytopathology 73:365.

Daughtrey, M. L., Hibben, C. R., and Hudler, G. W. 1988. Cause and control of dogwood
anthracnose in northeastern United States. Journal of Aboriculture 14:159-164.

Daughtrey, M. L., Hibben, C. R., Britton, K. O., Windham, M. T., and Redlin, S. C.
1996. Dogwood anthracnose, understanding a disease new to North America. Plant
Disease 80:349-358.

Eyre, F. H., ed. 1980. Forest cover types of the United States and Canada. Society of
American Foresters, Washington, DC. p 148.

Faggian, R., Bulman, S. R., Lawrie, A. C., and Porter, 1. J. 1999. Specific polymerase
chain reaction primers for the detection of Plasmodiophora brassicae in soil and water.
Phytopathology 89:392-397.

Gardes, M. and Bruns, T. D. 1993. ITS primers with enhanced specificity for
basidiomycetes — application to the identification of mycorrhizae and rusts. Molecular
Ecology 2:113-118.

Glen, M., Tommerup, I. C., Bougher, N. L., and O’Brien, P. A. 2001. Specificity,
sensitivity and discrimination of primers for PCR-RF LP of larger basidiomycetes and
their applicability to identification of ectomycorrhizal fungi in Eucalyptus forests and
plantations. Mycological Research 105(2): 138-149.

Graveland, J. R., van der Wal, J. H., and Van Noordwijk, A. J. 1994. Poor reproduction

in forest passerines from decline of snail abundance on acidified soils. Nature 368:446-
448.

58

 

 

I

Harnelin, R. C., Berube, P., Gignac, M. and Bourassa, M. 1996. Identification of root rot
fungi in nusery seedlings by nested multiplex PCR. Applied and Environmental
Microbiology 62( 1 1): 4026-4031 .

Hartung, J. S., Daniel, J. F ., and Pruvost, O. P. 1993. Detection of Xanthomonas

campestris pv. am using the polymerase chain reaction. Applied Environmental
Microbiology 59:1143-1 148.

Hed, E. H., Windham, M. T., and Grant, J. F. 1999. Survival of conidia of Discula
destructiva in frass of the convergent lady beetle. Plant Disease 83:806-809.

Henson, J. M. and French, R. 1993. The polymerase chain reaction and plant disease
diagnosis. Annual Review of Phytopathology 31 :81-109.

Hibben, C. R. and Daughtrey, M. L. 1988. Dogwood anthracnose in northeastern United
States. Plant Disease 72:199-203.

Hiers, J. K. and Evans, J. P. 1997. Effects of anthracnose on dogwood mortality and
forest composition of the Cumberland Plateau (USA). Conservation Biology 11:1430-
1435.

Jacobsen, C. S. 1995. Microscale detection of specific bacterial DNA in soil with a
magnetic capture-hybridisation and PCR amplification assay. Applied and Environmental
Microbiology 61: 3347-3352.

Jasalavich, C. A., Ostrofsky, A., and Jellison, J. 2000. Detection and identification of
decay fungi in spruce wood by restriction fragment length polymorphism analysis of

amplified genes encoding rRNA. Applied and Environmental Microbiology 66:4725-
4734.

Kageyama, K., Ohyama, A., and Hyakumachi, M. 1997. Detection of Pythium ultimum
using polymerase chain reaction with species-specific primers. Plant Disease 81: 1 155-
I160.

Kim, S. H., Uzunovic, A. and Breuil, C. 1999. Rapid detection of Ophiostomapiceae and

0. quercus in stained wood by PCR. Applied and Environmental Microbiology 65:28 7-
290.

Knighten, J. L. and Anderson, R. L., eds. 1992. Results of the 1991 dogwood anthracnose
impact assessment and pilot test in the southeastern United States. USDA For. Serv.
South. Reg. Prot. Rep. R8-PR 23.

Knighten, J. L. and Anderson, R. L., eds. 1993. Results of the 1992 dogwood anthracnose

impact assessment and pilot test in the southeastern United States. USDA For. Serv.
South. Reg. Prot. Rep. R8-PR 24.

59

Kox, L., de Gruyter, H., Garbelotto, M., van Brouwershaven, I., Admiraa], J ., and
Baayen, R. 2002. Validation of a PCR method for detection and identification of
Phytaphthara ramarum. Sudden Oak Death Science Symposium: The State of Our
Knowledge. Dec. 15-18, 2002. Monterey, California.

Langrell, S. and Barbara, D. 2001. Magnetic capture hybridization for improved PCR
detection of Nectria galligena from lignified apple extracts. Plant Molecular Biology
Reporter 19: 5-11.

Langrell, S. 2002. Molecular detection of Neanecm'a galligena (syn. Nectria galligena).
Mycological Research 106(3): 280-292.

Lee, S. B. and Taylor, J. W. 1992. Phylogeny of five fungus-like protoctistan
Phytaphtara species, inferred from the internal transcribed spacers of ribosomal DNA.
Molecular Biology and Evolution 9:636-653.

McGee, C. E. 1986. Loss of Quercus spp. dominance in an undisturbed old-growth
forest. Journal of the Elisha Mitchell Scientific Society 102:10-15.

McLemore, B. F. 1990. Camus flarida L.: Flowering Dogwood. Pages 278-283, In:
Silvics of North America, Vol 2, Hardwoods. Burns, R. M. and Honkala, B. H., eds.
Forest Service, US. Dept. Agric., Washington, DC.

Mielke, M. E. and Langdon, K. 1986. Dogwood anthracnose threatens Catoctin Mountain
Park. Park Science (Winter):6-8.

Mitchell, W. A., Gibbs, P. A., and Martin, C. O. 1988. Flowering dogwood (Camus
flarida): section 7.5.9, US. Army Corps of Engineers Wildlife Resources Management
Manual, Tech Rept. EL-88-9, US. Army Engineer Waterways Experiment Station,
Vicksburg, MS. 25p.

Pirone, P. P. 1980. Parasitic fungus affects region’s dogwood. New York Times. Feb. 24,
pp.34,37.

Ranney, T. G., Grand, L. F ., and Knighten, J. L. 1995. Susceptibility of cultivars and
hybrids of kausa dogwood to dogwood anthracnose and powdery mildew. Journal of
Aboriculture 21(1): 11-16.

Redlin, S. C. 1991. Discula destructiva sp. nov., cause of dogwood anthracnose.
Mycologia 83:63 3-642.

Rossell, I. M., Rossell, C. R. Jr., Hining, K. J., and Anderson, R. L. 2001. Impacts of

dogwood anthracnose (Discula destructiva Redlin) on the fruits of flowering dogwood
(Camus flarida L.): Implications for wildlife. American Midland Naturalist 146:3 79-38 7.

60

 

Salogga, D. S. and Ammirati, J. F. 1983. Discula species associated with anthracnose of
dogwood in the Pacific Northwest. Plant Disease 67: 1290.

Schneeberger, N. F. and Jackson, W. 1989. Dogwood anthracnose at Catoctin Mountain
Park. USDA Forest Service NE. Area State Priv. F or. 16pp.

Servello, F. A., and Kirkpatrick, R. L. 1987. Regional variation in the nutritional ecology
of ruffed grouse. Journal of Wildlife Management 51:749-770.

Stiles, E. W. 1980. Patterns of fruit presentation and seed dispersal in bird-disseminated
woody plants in the eastern deciduous forests. The American Naturalist 1 16:670-688.

Taylor, B. H., Manhart, J. R., and Amasino, R. M. 1993. Isolation and characterization of
plant DNAs, p. 37-65. In B. R. Glick and J. E. Thompson (0d,), Methods in plant
molecular biology and biotechnology. CRC Press, Ann Arbor, Michigan.

Thomas, W. A. 1969. Accumulation and cycling of calcium by dogwood trees.
Ecological Monographs 39: 101-120.

Tuite, J. F. 1969. Plant pathological methods; fungi and bacteria. Burgess Pub. Co.,
Minneapolis, MN. 239pp.

Voss, E. G. 1985. Michigan flora: A guide to the identification and occurrence of the
native and natural seed-plants of the state. Cranbrook Institute of Science, Bloomfield
Hills, MI. v. II Dicots, pp. 675-681.

Wallace, L. L., and Dunn, E. L. 1980. Comparative photosynthesis of three gap phase
successional tree species. Oecologia 45:331-340.

White, T. J ., Bruns, T., Lee, S., and Taylor, J. 1990. Amplification and direct sequencing
of fungal ribosomal RNA genes for phylogenetics. In: Innis M.A., Gelfand D.H., Sninsky
J .J ., and White T.J. (eds), PCR Protocols: a Guide to Methods and Applications, pp. 315-
322. Academic Press, New York.

Wilson, K. 1987. Preparation of genomic DNA from bacteria, p. 2.4.1-2.4.5. In F. M.
Ausubel, R. Brent, R. E. Kingston, D. D. Moore, J. G. Seidmand, J. A. Smith, and K.

Struhl (ed.), Current protocols in molecular biology. John Wiley & Sons, Inc., New York,
N. Y.

Windham, M. T., Graham, E. T., Witte, W. T., Knighten, J. L., and Trigiano, R. N. 1998.
Carnusflarida ‘Appalachian Spring’: A white flowering dogwood resistant to dogwood
anthracnose. HortScience 33(7): 1265-1267.

Zhang, N. and Blackwell, M. 2002. Population structure of dogwood anthracnose fungus.
Phytopathology 92: 1276-1283.

61

 

Zhang, W. F., Wendel, J. F., Clark, L. G. 1997. Bamboozled again — inadvertent isolation
of fungal rDNA sequences from bamboos (Paaceae, Bambusaideae). Molecular
Phylogenetics and Evolution 8: 205-217.

62

 

 

 

93 02469 1507 I