454
Mycologia, 97(2), 2005, pp. 454–463.
q 2005 by The Mycological Society of America, Lawrence, KS 66044-8897
Genetic variation in the widespread lichenicolous fungus
Marchandiomyces corallinus
M. Carmen Molina1
Departamento de Matema´ticas y Fı´sica Aplicadas y
Ciencias de la Naturaleza, Escuela Superior de
Ciencias Experimentales y Tecnologı´a, Universidad Rey
Juan Carlos, 28933 Madrid, Espan˜a
Paula T. DePriest
Botany Section, United States National Herbarium,
National Museum of Natural History, Smithsonian
Institution, P.O. Box 37012, Washington, DC 20013-
7012
James D. Lawrey
Department of Environmental Science and Policy,
George Mason University, Fairfax, Virginia 22030-
4444
Abstract: The lichenicolous basidiomycete Mar-
chandiomyces corallinus is widely distributed in North
America and Europe, where it commonly is found on
a variety of lichens. Theoretically either of these char-
acteristics, a wide geographic range or generalized
host ecology, could provide opportunities for genetic
differentiation within this species. To determine how
genetic variation is partitioned in M. corallinus, 12
fungal isolates were obtained from locations in North
America and Europe; at two locations, in Washington
County, Maine, and on the Isle of Mull in Scotland,
fungi also were isolated from different lichen hosts.
Vegetative mycelial compatibility tests were used to
determine compatibility groupings from among the
isolates; in addition, several PCR amplification prod-
ucts (RAPD, nuITS rDNA) were obtained for each
isolate. A number of distinct compatibility groups
were recognizable based on geography, not host ecol-
ogy. In addition compatible isolates always were re-
stricted to either North America or Europe. However
RAPD markers indicated that compatible isolates are
not always genetically identical. The presence of se-
quence heterozygosity at specific positions indicated
that the isolates are heterokaryotic and a number of
distinct haplotypes could be identified based on ITS
variation at three separate locations. This type of ge-
netic variation in these fungi suggests that sexual re-
combination is possible and that genetic differentia-
tion has taken place recently as a result of geographic
isolation, not host switching.
Accepted for publication 21 Dec 2004.
1 Corresponding author. E-mail: cmolina@escet.urjc.es
Key words: basidiomycetes, host-parasite evolu-
tion, lichenicolous fungi, lichens, Marchandiomyces,
rDNA sequences
INTRODUCTION
Lichenicolous fungi form obligate associations with
lichens, either as saprotrophs that colonize dead li-
chen thalli or parasites that range in virulence from
relatively benign to aggressively pathogenic. The
most recent survey of these fungi by Lawrey and
Diederich (2003) lists more than 1500 species with
an estimate of over 3000. More than 95% of these
species are ascomycetous, although a number of ba-
sidiomycete groups have this habit as well (Sikaroodi
et al 2001). The host specificity of these fungi ap-
pears to be high (Diederich 2000), with as many as
95% thought to be associated with a single lichen
genus. Nevertheless some of the best known licheni-
colous fungi have wide geographic distributions and
broad host ecologies.
Several authors have discussed the evolution of the
lichenicolous habit and speculated on possible evo-
lutionary trends. Hawksworth (1978, 1982a, b, 1988a,
b) emphasized the reticulate nature of fungal habits
including the lichenicolous habit, an idea supported
by phylogenetic reconstructions of transitions among
nutritional modes in major fungal clades of ascomy-
cetous (i.e., Gargas et al 1995) and basidiomycetous
fungi (i.e., Hibbett et al 2000). Given the tools of
molecular biology, investigators now routinely ex-
plore microevolutionary patterns in parasite-host (es-
pecially plant-fungal) associations. At present lich-
enicolous fungi have not been studied genetically at
the intraspecific level, so the mechanisms responsible
for the origin and evolution of lichen-parasite inter-
actions have yet to be documented. Are there obvious
differences in the genetic variation of these fungi?
How different is the genetic variation of host-special-
ists and host-generalists? Is there evidence of genetic
differentiation among lichenicolous fungi that use
different lichen hosts? Is there evidence of geograph-
ic differentiation in widely distributed lichenicolous
fungi?
The lichenicolous basidiomycete M. corallinus
(Roberge) Diederich & D.Hawksw is collected com-
monly throughout eastern North America and Eu-
455MOLINA ET AL: GENETIC VARIATION IN MARCHANDIOMYCES CORALLINUS
rope from a variety of host lichens, including species
of Parmelia s.l., Physcia s.l., Lepraria, Pertusaria, Le-
canora and Lasallia. Infected lichens exhibit obvious
coral-colored bulbils clustered on the surface of the
thallus. A sexual stage has not been observed, and
the nuclear condition of the bulbilliferous stage is
not known. However basidiomycete teleomorph of
another species in the genus, M. aurantiacus (Lasch)
Diederich & Etayo, was described (Diederich et al
2003) as Marchandiobasidium auranticacum Dieder-
ich & Schultheis. The anamorph is common in pol-
luted habitats in Europe, and it also attacks a variety
of lichens, especially species of Physcia. In addition
to M. corallinus and M. aurantiacus the genus in-
cludes one other species, M. lignicola Lawrey & Died-
erich, that appears from molecular data to be most
closely related to M. corallinus but is lignicolous in
habit (DePriest et al 2005), indicating a remarkably
flexible nutritional ecology in the members of this
genus.
The presumed close relationships among Marchan-
diomyces and other mitosporic genera (Hobsonia, Il-
losporium), discussed at various times in the literature
(Lowen et al 1986), were shown by Sikaroodi et al
(2001) to be erroneous, as predicted by Etayo and
Diederich (1996). It should be noted that the most
recent edition of the Dictionary of the Fungi (Kirk
et al 2001) classifies Marchandiomyces as an anamor-
phic ascomycete, citing Sikaroodi et al (2001); this is
clearly a mistake because the latter study demon-
strates an unambiguous basidiomycetous position for
Marchandiomyces.
The phylogenetic reconstructions of DePriest et al
(2005) placed the three species of Marchandiomyces
in a clade made up of representatives of the types of
the genera Dendrocorticium, Duportella, Laeticorticium
and Vuilleminia. Various authors (Hibbett and Thorn
2001, Binder and Hibbett 2002, Hibbett and Binder
2002, Larsson et al 2004) have recognized the dis-
tinctiveness of this clade but have referred to it using
different names. Hibbett and Binder (2002) refer to
it as the Dendrocorticium clade, and Larsson et al
(2004) call it a corticioid clade; both studies make
clear that fungi traditionally considered to be corti-
cioid are widely distributed among the basidiomy-
cetes.
The wide geographic distribution and host ampli-
tude of M. corallinus would appear to present nu-
merous opportunities for genetic differentiation in
this fungus. During the past decade a small group of
samples has been obtained from various locations in
North America and Europe representing the known
range of this fungus, and in certain locations samples
were taken from different lichen hosts in the same
habitat. We since have used pairwise tests of mycelial
compatibility (Rayner 1991, Worrall 1997) and pres-
ence of various molecular markers (RAPD, nuITS
rDNA) to describe the isolates genetically and assess
the level of genetic differentiation among the popu-
lations they represent.
We expected results of these investigations to shed
light on a number of questions, among them: (i)
How genetically distinct are these bulbilliferous (pre-
sumably asexual) fungi? (ii) Is there evidence, direct
or indirect, for a heterokaryotic nuclear condition in
these fungi? (iii) How are vegetative compatibility
groups distributed geographically? (iv) Is genetic dif-
ferentiation correlated with geographic distance or
host ecology?
METHODS
Fungal specimens.—Isolates from 12 specimens of Marchan-
diomyces corallinus were used. Specimens were collected
from North America and Europe from a variety of locations
representing the known range of this fungus (TABLE I). In
two locations (Maine and Scotland) a sufficient amount of
material could be obtained from different lichen hosts
growing in the same habitat to test for genetic differentia-
tion possibly caused by host switching in the fungus.
Isolation of fungal cultures.—Fungal cultures were isolated
from freshly collected material. Infected thalli were washed
briefly in sterile water, and bulbils were removed using a
flamed needle. Bulbils were placed on either potato-dex-
trose agar (PDA) or Sabouraud’s medium (Difco Labora-
tories Inc., 10 g Bacto Neopeptone/L, 20 g Bacto Dextrose/
L), and mycelial outgrowths were subcultured monthly
(Lawrey 2002). Voucher cultures of new isolates were sent
to the American Type Culture Collection for reference.
These cultures were used for compatibility/incompatibility
tests and for DNA extraction.
Mycelial compatibility groupings.—Pairwise tests of mycelial
compatibility were done with 8-mm diam mycelial plugs tak-
en from the margin of a 14 d old colony of each isolate
and placed about 25 mm apart on 100 3 15 mm dishes
containing PDA. All possible pairwise combinations were
tested, including pairs from the same isolate. The dishes
were incubated 14 d and examined for the presence of an
aversion or barrage reaction in the zone of contact between
mycelial outgrowths. Isolates that formed a reaction were
assigned to different compatibility groups. All tests were
replicated.
DNA extraction.—Total DNA was extracted from each of the
cultures using the TES extraction protocol modified from
Grube et al (1995). Fungal tissue was ground in TES buffer
(100 mM Tris, 10 mM EDTA, 2% SDS, pH 8.0), with 1.4 M
NaCl and 10% CTAB, and extracted twice with 1 volume
chloroform : isoamyl (24:1). DNA was precipitated in 0.6
volume of isopropanol with sodium acetate and washed with
70–80% ethanol. The DNA pellet was suspended in 20–30
mL of deionized water, and the DNA was quantified by vi-
sualizing with ethidium bromide on a 1% agarose gel.
456 MYCOLOGIA
TABLE I. Specimens of Marchandiomyces corallinus used in the study with collector, location of collection, lichen substrate
and ATCC accession number
GMU isolate
No Collector Location Lichen substrate ATCC number
JL106-95
JL108-96
JL136-00
JL160-00
JL161-00
Lawrey 1619
Lawrey 1626
Cullen & Fox 216
Buck 37514
Buck 37404
Bear Is, Maryland (USA)
Augusta Co., Virginia (USA)
Isle of Mull, Scotland (Europe)
Howell Co., Missouri (USA)
Madison Co., Arkansas (USA)
Flavoparmelia baltimorensis
Lasallia papulosa
Pertusaria amara
Pertusaria neoscotia
Pertusaria plittiana
200796
200797
MYA-1118
MYA-1119
MYA-1120
JL167-00
JL171-00
JL172-00
JL173.00
JL213-01
Fox (no number)
Fox 322
Fox 322
Fox 322
Cole 9290
County Kerry, Ireland (Europe)
Isle of Mull, Scotland (Europe)
Isle of Mull, Scotland (Europe)
Isle of Mull, Scotland (Europe)
Washington Co., Maine (USA)
Tephromela atra
Ramalina subfarinacea
Parmelia sulcata
Hypogymnia physodes
Lasallia papulosa
MYA-1377
MYA-1378
MYA-1379
MYA-1380
MYA-2499
JL222-01
JL236-02
Diederich 14869
Hoffman (no number)
Moselle, France (Europe)
Washington Co., Maine (USA)
Porpidia cinereoatra
Xanthoparmelia sp.
MYA-2501
MYA-2827
FIG. 1. Position of primers used in this analysis. The nu-
SSU-1785-59 (MC1) primer is Marchandiomyces corallinus-
specific for the ITS region. The nu-ITS-146A-39 (MC9) and
nu-ITS-146G-39 (MC10) primers were used to establish the
heterokaryotic character of M. corallinus.
When the extractions were carried out using natural (not
isolated) material or when the amount of the culture was
small, DNA was extracted using the Dneasy Plant Mini Kit
(Qiagen) with minor modifications as described elsewhere
(Crespo et al 2001).
RAPD-PCR amplification.—The protocol for PCR amplifi-
cation is derived from Sikaroodi et al (2001) and modified
for RAPD-PCR amplification. Fragments were amplified
from the genomic DNA (;10 ng) with 1.25 units of Klentaq
1 (Ab peptides Inc.) in 100 mL PCR reactions in a reaction
buffer (10 mM Tris pH 8.3, 50 mM KCl and 2 mM MgCl2),
with 200 mM of each of the four dNTPs and 0.5 mM of each
primer. The reactions were carried out in a Perkin-Elmer
Cetus DNA Thermal Cycler, for 42 cycles with these condi-
tions for most reactions: template denaturation at 94 C for
1 min, primer annealing at 30 C for 1 min, and primer
extending at 72 C for 2 min (extended by 5 s in each cycle).
The primers used were P102 (59-GGTGGGGACT-39), P130
(59-GGTTATCCTC-39), P131 (59-GAAACAGCGT-39), P128
(59-GCATATTCCG-39) and P129 (59-GCGGTATAGT-39).
Amplification products were separated by electrophoresis,
stained with ethidium bromide and visualized under UV
light. All clearly visible bands were included in the analysis;
faint bands were considered unidentifiable. To assess the
reproducibility of our data RAPD amplifications were twice
repeated; all replicates produced identical band patterns.
nu-ITS PCR amplification.—DNA isolates were diluted (10–
300-fold) to get approximately 10 ng for amplifications. Am-
plification reactions were carried out as described in Sika-
roodi et al (2001). The nuITS1 region was amplified be-
tween primer nu-SSU-1766-59 (ITS5) and primer ITS2
(White et al 1990). We obtained one sequence from bulbils
of a field-collected specimen of M. corallinus ( JL236) grow-
ing on Xanthoparmelia sp, using a primer designed espe-
cially for this purpose, nu-SSU-1785-59 MC1 (FIG. 1) and
ITS4 (White et al 1990). Two new primers to amplify a frag-
ment of ITS1 also were designed to establish the presence
of one or two templates in the isolates: nu-ITS-146A-39 and
nu-ITS-146G-39 (FIG 1.). Both were paired with ITS5.
PCR products were purified of excess primers with either
of two protocols: (i) precipitation with 20% polyethylene
glycol and 2.5 mol/L NaCl or (ii) filtration through PCR
Wizards (Promega) following the manufacturer instruc-
tions. The concentration and size of the PCR amplification
products were estimated by comparing to nucleotide weight
and size markers after agarose gel electrophoresis and
stained with ethidium bromide and exposure to UV light.
DNA sequencing.—Double stranded PCR products were se-
quenced from each of the amplification primers and a num-
ber of internal sequencing primers (ITS2 and ITS3). Ap-
proximately 100 ng of cleaned products were sequenced
from 3.2 pM of primer with the PRISM Ready Reaction Dye
Deoxy Terminator Cycle Sequencing Kit (Applied Biosys-
tems). The reaction was carried out in a Perkin Elmer Cetus
DNA Thermal Cycler for 25 cycles under these conditions:
template denaturation was done at 96 C for 30 s, primer
annealing at 50 C for 15 s and primer extension at 60 C for
4 min. The cycle sequencing products were purified of ex-
cess dye with filtration through Sephedex G-50 Fine (Phar-
macia) columns and were run on a 4% polyacrylamide gel
in a 373A and 377 Automatic Sequencer (Applied Biosys-
tems).
457MOLINA ET AL: GENETIC VARIATION IN MARCHANDIOMYCES CORALLINUS
FIG. 2. Example of a gel showing amplification products
from DNA of Marchandiomyces corallinus isolates using
RAPD primer ‘‘102’’. Isolates 1 5 JL106(MD), 2 5
JL213(ME), 3 5 JL236(ME), 4 5 JL108(VA), 5 5
JL160(MO), 6 5 JL161(AR), 7 5 JL167(IRE), 8 5
JL136(SCO), 9 5 JL171(SCO), 10 5 JL172(SCO), 11 5
JL173(SCO), 12 5 JL222(FRA).
TABLE II. RAPD band patterns in each fungal isolate using five different primers. ? 5 unidentifiable, D9 5 D pattern except
in one band
RAPD
Primers
North America
JL106
MA
JL213
ME
JL236
ME
JL108
VA
JL160
MO
JL161
AR
Europe
JL167
IRE
JL136
SCO
JL171
SCO
JL172
SCO
JL173
SCO
JL222
FRA
‘‘102’’
‘‘128’’
‘‘129’’
‘‘130’’
‘‘131’’
A
A
A
A
A
A
B
B
B
A
A
B
B
C
A
A9
C
C
?
A
A9
C
B
B9
A
A9
C
B
B9
A
B
D
D
E
B
B
D
D
F
B
B
D9
D
G
C
B
D9
D
E
C9
B
D9
D
?
C9
B
F
D
?
F
Sequence compiling.—Base calling software (Sequencing
Analysis, ABI Prism, 2.1.1) was used to produce a prelimi-
nary nucleotide sequence. The nucleotide sequence frag-
ments were compiled with Sequence Navigator 1.0 (Applied
Biosystems). The sequences were confirmed by comparison
to sequences produced from the opposite strand and al-
tered by manual base calling where appropriate.
RESULTS
Mycelial compatibility.—Pairwise tests of mycelial com-
patibility were used to obtain a general idea of ge-
netic similarity. Genetically identical (or sufficiently
similar) mycelia will anastomose readily when grown
together on agar, but genetically incompatible my-
celia will form interaction zones. Several mycelial
compatibility (MC) groups were discovered for M.
corallinus (TABLE III), and they appear to be corre-
lated with geographic location because compatibility
was never observed between any North American
sample and any European sample. Samples from
North America form three compatibility groups:
MC1 is represented by a single culture JL106 (Mary-
land), MC2 is represented by two cultures from dif-
ferent lichens in the same location, JL213 and JL236
(Maine), and MC3 is represented by JL108 (Virgin-
ia), JL160 (Missouri) and JL161 (Arkansas). Samples
from Europe form four compatibility groups. Three
are represented by single isolates, MC4 from JL167
(Ireland), MC5 from JL136 (Scotland), MC7 from
JL222 (France), and a fourth (MC6) is represented
by three isolates from different lichen hosts in the
same habitat, JL171, JL172 and JL173 from Scotland.
In both North America and Europe samples from dif-
ferent lichen substrates in the same location are al-
ways from the same compatibility group (TABLE III).
European samples collected from Lochbuie Stone
Circle, Isle of Mull (Scotland), included those grow-
ing on Ramalina subfarinacea (Ramalinaceae), Par-
melia sulcata and Hypogymnia physodes (Parmeli-
aceae), and all are of the MC6 compatibility group.
North American samples collected at the Eagle Hill
Field Station in Maine from two lichen substrates,
Xanthoparmelia sp. and Lasallia papulosa, were of the
MC2 compatibility group.
RAPD analyses.—RAPD PCR band patterns were dis-
tinct for most isolates, even those from the same my-
celial compatibility group (FIG. 2, TABLE II); only
JL160 and JL161 had the same band pattern. RAPD
bands cannot distinguish heterokaryotic from hap-
loid nuclear conditions, nor can they distinguish ho-
mozygotes from heterozygotes in heterokaryotic in-
dividuals. Given the possibility that M. corallinus bul-
billiferous stages are heterokaryotic, we were hesitant
to assign much meaning to these data beyond the
observation that isolates are rarely identical, which
indicates a much higher level of genetic variation
among both North American and European popu-
lations than we expected for asexual fungi.
ITS1 results.—Sequences of certain of the cultures
(e.g., JL106, JL213, JL236) exhibited a double peak
458 MYCOLOGIA
FIG. 3. Electropherograms obtained from DNA using several primers. A. Sequence from culture JL106 (MD) that is
presumably heterozygous at the 146 position of ITS1 using the ITS2 primer (the same electropherogram was obtained using
ITS5), R 5 double peak A and G, Y 5 double peak C and T. B. Sequence of culture JL106 (MD) using M9 primer (nu-ITS-
146A-39, 59-CGGTTTACATTCAGAACGTA-39). C. Sequence of culture JL106 (MD) using M10 primer (nu-ITS-146G-39, 59-
CGGTTTACATTCAGAACGTG-39).
on positions 103 (A/G) and 146 (T/C) on the ITS1
segment (FIG. 3A) when they were obtained using
ITS5 or ITS2. When this fragment was sequenced us-
ing the primer MC9 (nu-ITS-146A-39, 59-CGGTTTA-
CATTCAGAACGTA-39), designed to match one of
the possible sequences for this region, a clear elec-
tropherogram was found with an ‘‘A’’ at the down-
stream 103 position (FIG. 3B). When this same frag-
ment was sequenced using the primer MC10 (nu-ITS-
146G-39, 59-CGGTTTACATTCAGAACGTG-39), de-
signed to match the other possible sequence for this
position, a uniform electropherogram without dou-
ble peaks was found with a ‘‘G’’ at the downstream
103 position (FIG. 3C). These molecular markers in-
dicated the presence of two distinct DNA haplotypes
in the extract obtained from these isolates, suggest-
ing a heterozygous state for these particular sequence
positions. One haplotype has ‘‘A’’ at position 103 and
‘‘T’’ at position 146, and the other has ‘‘G’’ at posi-
tion 103 and ‘‘C’’ and position 146. This is not true
of all isolates, however, because several cultures (e.g.,
JL136 from Scotland) exhibited a uniform electro-
pherogram when ITS2 (FIG. 4A) and ITS5 (FIG. 4B)
were used to sequence the ITS1 fragment. If these
are also heterokaryotic, this pattern suggests a ho-
mozygous state for this particular sequence position.
Certain isolates (e.g., JL171, JL172 and JL173 from
Scotland) also exhibited double peaks downstream
from the 141 ITS1 position when the electrophero-
gram was obtained with ITS2 (FIG. 4C) and down-
stream from the 132 ITS1 position when the ITS5 was
used (FIG. 4D). This alteration is also apparently a
consequence of overlap of two different templates.
One of them was constituted by the sequence 59-ACA-
CACACAC-39 (5AC repeated) and the other one by
59-ACACACACACAC-39 (6AC repeated). A uniform
electropherogram was obtained when both templates
had the same number of ‘‘ACs’’ at the position (FIG.
4A and B) and double peaks were detected when the
templates had a different number of ‘‘ACs’’ repeated
(FIG. 4C and D). Four haplotypes were obtained for
this microsatellite, two of which we are interpreting
as homozygous (5AC/5AC and 6AC/6AC) and two
heterozygous (5AC/4AC and 6AC/5AC). One group
of isolates (e.g., JL108, JL160, JL161) had double
peaks on position 103 (A/C) and on the position
459MOLINA ET AL: GENETIC VARIATION IN MARCHANDIOMYCES CORALLINUS
FIG. 4. Electropherograms obtained from DNA using two primers, ITS1 and ITS2. A. Sequence from culture JL136(SCO)
that is presumably homozygous for five ‘‘AC’’ repeats (molecular marker ‘‘132’’ in TABLE IV) in the ITS1, using the ITS2
primer. B. Sequence from culture JL136 (SCO) using the ITS5 primer. C. Sequence from culture JL173 (SCO) that is
presumably heterozygous for the number of ‘‘AC’’ repeats in the ITS1, using the ITS2 primer. D. Sequence from culture
JL173 (SCO) using the ITS5 primer. Asterisks (* *) show the positions where double peaks indicate the overlap of templates
with different numbers of ‘‘AC’’ repeats. The arrows show the downstream sequence direction.
region 133 to 138 (5AC/4AC). However given the
lack of variation on position 146 (T/T) we could not
fully resolve the DNA haplotypes.
Altogether our investigations identified three mo-
lecular markers from ITS1 at the 103, 132 and 146
positions. The inferred pattern of arrangement of
the various haplotypes in each of the sequences is
provided (TABLE IV). Patterns observed for these
markers generally match those obtained in the my-
celial compatibility tests (TABLE III). It is interesting,
however, that certain isolates (the North American
isolates JL106, JL213 and JL236 and the European
isolates JL136 and JL167) share one ITS haplotype
(Haplotype 1) but belong to different compatibility
groups. However RAPD markers for these isolates are
also clearly different, indicating that they are not
clonally derived.
DISCUSSION
It is becoming clear from recent molecular studies
that fungi assumed to be exclusively clonal actually
are capable of recombination in nature (Taylor et al
2000), and this appears to be the case with Marchan-
diomyces corallinus as well. The presence of larger-
than-expected ITS and RAPD variation in isolates of
460 MYCOLOGIA
TABLE III. Molecular markers at the 103, 132 and 146 ITS1 positions, mycelial compatibility group (MC), and RAPD band
patterns for each isolate
Fungal isolates
ITS1 Positions
‘‘103’’ ‘‘132’’ ‘‘146’’ MC RAPD patterns
JL106 MD
JL213 ME
JL236 ME
JL108 VA
JL160 MO
A/G
A/G
A/G
A/C
A/C
5AC/5AC
5AC/5AC
5AC/5AC
5AC/4AC
5AC/4AC
T/C
T/C
T/C
T/T
T/T
MC1
MC2
MC2
MC3
MC3
AAAAA
ABBBA
ABBCA
A9CC?A
A9CBB9A
JL161 AR
JL167 IRE
JL136 SCO
JL171 SCO
JL172 SCO
A/C
A/A
A/A
A/A
A/A
5AC/4AC
5AC/5AC
5AC/5AC
6AC/5AC
6AC/5AC
T/T
T/T
T/T
T/T
T/T
MC3
MC4
MC5
MC6
MC6
A9CBB9A
BDDEB
BDDFB
BD9DGC
BD9DEC9
JL173 SCO
JL222 FRA
A/A
A/A
6AC/5AC
6AC/6AC
T/T
T/T
MC6
MC7
BD9D?C9
BFD?F
TABLE IV. Inferred ITS haplotype combinations for each of the fungal isolates used in the study. Five distinct haplotypes
(1–5) were inferred from the 12 isolates. Haplotype sequences from cultures JL108 VA, JL160 MO, and JL161 AR could not
be fully resolved. Molecular markers at three ITS1 positions, as in Table III are shown in bold type. Complete haplotype 1
ITS sequences (ITS1, ITS2, and 5.8SS) for isolates JL106 MD and JL136 SCO were previously deposited as GenBank AY583324
and AY583325, respectively
Fungal isolates Inferred ITS haplotype combinations Haplotypes
‘‘103’’ ‘‘132’’ ‘‘146’’
JL106 MD
JL213 ME
JL236 ME
ACCCTTTCCGATCCGGTTCCGCGTCTTTTACACACACAC--TAT
GCCCTTTCCGATCCGGTTCCGCGTCTTTTACACACACAC--TAC
ACCCTTTCCGATCCGGTTCCGCGTCTTTTACACACACAC--TAT
GCCCTTTCCGATCCGGTTCCGCGTCTTTTACACACACAC--TAC
ACCCTTTCCGATCCGGTTCCGCGTCTTTTACACACACAC--TAT
GCCCTTTCCGATCCGGTTCCGCGTCTTTTACACACACAC--TAC
1
2
1
2
1
2
JL108 VA
JL160 MO
JL161 AR
MCC--TTCSGATYTGGTCCYACGTSTTTATCACACACAC--TAT
MCC--TTCSGATYTGGTCCYACGTSTTTATCACACAC----TAT
MCC--TTCSGATYTGGTCCYACGTSTTTATCACACACAC--TAT
MCC--TTCSGATYTGGTCCYACGTSTTTATCACACAC----TAT
MCC--TTCSGATYTGGTCCYACGTSTTTATCACACACAC--TAT
MCC--TTCSGATYTGGTCCYACGTSTTTATCACACAC----TAT
3
4
3
4
3
4
JL167 IRE
JL136 SCO
JL171 SCO
ACCCTTTCCGATCCGGTTCCGCGTCTTTTACACACACAC--TAT
ACCCTTTCCGATCCGGTTCCGCGTCTTTTACACACACAC--TAT
ACCCTTTCCGATCCGGTTCCGCGTCTTTTACACACACAC--TAT
ACCCTTTCCGATCCGGTTCCGCGTCTTTTACACACACAC--TAT
ACCCTTTCCGATCCGGTTCCGCGTCTTTTACACACACAC--TAT
ACCCTTTCCGATCCGGTTCCGCGTCTTTTACACACACACACTAT
1
1
1
1
1
5
JL172 SCO
JL173 SCO
JL222 FRA
ACCCTTTCCGATCCGGTTCCGCGTCTTTTACACACACAC--TAT
ACCCTTTCCGATCCGGTTCCGCGTCTTTTACACACACACACTAT
ACCCTTTCCGATCCGGTTCCGCGTCTTTTACACACACAC--TAT
ACCCTTTCCGATCCGGTTCCGCGTCTTTTACACACACACACTAT
ACCCTTTCCGATCCGGTTCCGCGTCTTTTACACACACACACTAT
ACCCTTTCCGATCCGGTTCCGCGTCTTTTACACACACACACTAT
1
5
1
5
5
5
M. corallinus and the presence of more than one dis-
tinct haplotype in some of these isolates suggests that
genetic recombination (or less likely hybridization)
is at least possible in this fungus and that geographic
differentiation has taken place. Even isolates ob-
tained from the same habitat have different RAPD
patterns, indicating that many populations of this
fungus are made up of more than one genet and that
few are derived clonally. The only isolates with iden-
tical RAPD patterns were from specimens collected
461MOLINA ET AL: GENETIC VARIATION IN MARCHANDIOMYCES CORALLINUS
in Missouri and Arkansas, indicating in this case the
existence of a single large, geographically widespread
clone. No sexual state has been described for M. cor-
allinus, but the finding that it is heterokaryotic in-
dicates that mating has taken place in the past, an
unexpected result given the absence of clamps in iso-
lated mycelia. Because most isolates are distinct ge-
netically and probably not clonally derived, we would
expect that there are a number of different mating
types in the species.
Mycelial compatibility groups in M. corallinus ex-
hibited some structure assignable to geography, with
distinctly North American and European groups and
no compatibility between the two continents. This
has been observed before for fungal plant pathogens.
For example a recent gene genealogical study
(O’Donnell et al 2000) of Fusarium graminearum, a
widespread virulent plant pathogen that causes Fu-
sarium head blight of wheat and barley, demonstrat-
ed that this species is not panmictic but comprises
seven phylogenetically distinct groups with limited
gene flow among them. The origin of this structure
appeared to be mainly geographic separation, but
host-mediated isolation also might have played a role.
In the case of M. corallinus, there is no indication
of a significant genetic separation caused by host
switching, as is sometimes observed in fungal plant
pathogens (e.g., Brem and Leuchtmann 2003, Har-
vey et al 2001). In our study, isolates from different
lichen hosts in the same habitat were always identical
in ITS sequence and intercompatible. However they
were never identical in RAPD pattern, suggesting that
different genotypes might exhibit minor differences
in host preference. A more detailed test of this will
compare within- and between-host genetic variation
in the same compatibility groups using RAPD mark-
ers. Given the intercompatibility of host groups, we
expect within- and between-group genetic differenc-
es to be no different.
Recent genetic studies of other widely distributed
fungal species demonstrate that genetic differentia-
tion may take place without either host switching or
geographic separation. For example, a study of cryp-
tic species in Stachybotrys chartarum (Cruse et al
2002), which has been implicated as a possible cause
of sick-building syndrome, indicated little differenti-
ation caused by geographic separation. Steenkamp et
al (2002) similarly found evidence for genetic differ-
entiation, even speciation, in various groups within
Fusarium subglutinans, but the cause of this differ-
entiation could not be attributed to either host-
switching or geographic distance.
It has been suggested that mycelial incompatibility
maintains the genetic identity of genotypes, although
genetic exchange between certain genotypes appears
to be permitted. In the case of M. corallinus, even
this small sample of isolates represents many differ-
ent compatibility groups and some are known only
from a single location at this point. A marked excep-
tion is the group formed by isolates from VA, AR and
MO, two of which (those from MO and AR) have
identical RAPD banding patterns. This particular
compatibility group is the most widely distributed
one that we studied and might represent a single
clone (MO and AR are apparently identical) that has
undergone subsequent genetic differentiation in
parts of its range.
Isolates from different lichen hosts in the same
habitat were always intercompatible and exhibited
the same ITS haplotypes, but the RAPD banding pat-
terns were different. Population genetic studies of
other bulbilliferous or sclerotial basidiomycetes gen-
erally show similar results. For example Punja and
Sun (2001) studied mycelial compatibility groups of
the widespread soilborne sclerotial basidiomycete
Sclerotium rolfsii (teleomorph Athelia rolfsii) that caus-
es diseases on a wide range of plant species. Isolates
were generally unique, single-member compatibility
groups structured geographically, but no clear rela-
tionship was found between compatibility and host
plant of origin. In addition RAPD patterns were usu-
ally distinct for each isolate, even those from the
same compatibility group, indicating few were clon-
ally derived. High levels of RAPD polymorphism also
are observed commonly both within and among com-
patibility groups in the sclerotial basidiomycete Rhi-
zoctonia solani (Duncan et al 1993, Bounou et al
1999).
Genetic polymorphism within compatibility groups
of asexual basidiomycetes can be detected using
markers other than RAPD markers. For example iso-
lates from the same compatibility group can have dif-
ferent ITS sequences in Sclerotium rolfsii (Harlton et
al 1995), Armillaria spp. (Guillaumin et al 1996) and
the Rhizoctonia solani species complex (Boysen et al
1996). In R. solani genetic differences among and
within compatibility groups also are detectable using
PCR amplification of SSU nrDNA (Liu et al 1995),
RFLPs (Vilgalys 1988, Jabaji-Hare et al 1990, Vilgalys
and Gonzales 1990) and isozymes (Laroche et al
1992), suggesting the existence of far greater genetic
polymorphism in these groups than we observed in
groups of M. corallinus.
It was not possible to determine reliably the phy-
logenetic structure within M. corallinus using the lim-
ited number of genetic markers we were able to iden-
tify in this study. North American isolates were dis-
tinctly different from European isolates in the ITS
sequence position 103, with all North American iso-
lates being A/G or A/C at this position and all Eu-
462 MYCOLOGIA
ropean isolates A/A. In addition North American
and European isolates always were incompatible.
However, at the ITS positions 132 and 146, no obvi-
ous continental geographic patterns can be seen and
there were only minor within-continent differences
among either North American or European isolates.
We reasoned that the RAPD data could not be used
to infer phylogenetic relationships inasmuch as they
may represent various combinations of heterokary-
otic banding patterns. There may be many of these
and no clear indication that they group isolates by
location or host type. Because these patterns might
have arisen as a consequence of mating events reg-
ulated by as yet unknown mating compatibility rela-
tionships, we hesitated to make much of them. It is
interesting that many of the ITS and RAPD markers
found in M. corallinus also are found in its close rel-
atives, M. aurantiacus, which like M. corallinus is lich-
enicolous, and M. lignicola, which is lignicolous. The
latter species is also undoubtedly heterokaryotic (it
has obvious clamps), but it has not been collected
from enough localities to know much about its range
(it has not yet been collected in Europe for exam-
ple). It might be possible in the future to identify
enough genetic markers in these species to use in a
phylogenetic analysis that addresses both the origin
of the nutritionally distinct species and the subse-
quent divergence of the geographically widespread
M. corallinus.
As Hawksworth and Rossman (1997) have noted,
lichenicolous fungi probably represent an important
source of new fungal species, especially because they
only recently have been collected extensively outside
Europe. Because nearly 95% of described lichenicol-
ous fungi are narrowly host specific the rare but ubiq-
uitous host-generalized species are exceptional and
interesting for this reason alone. In addition, how-
ever, they might represent heterogeneous assemblag-
es of cryptic species, as has been suggested (Lawrey
and Diederich 2003). In the case of M. corallinus,
among the most geographically widespread of lich-
enicolous fungi, cryptic speciation probably has not
taken place. However there appears to have been
some genetic differentiation within this species, gen-
erated by geographic, not host, separation.
ACKNOWLEDGMENTS
We thank Bill Buck, Mariette Cole, Nicholas Hoffmann,
Paul Diederich and Howard Fox for providing fresh mate-
rial for culture. We also thank an anonymous reviewer for
valuable comments.
LITERATURE CITED
Binder M, Hibbett DS. 2002. Higher level phylogenetic re-
lationships of homobasidiomycetes (mushroom-form-
ing fungi) inferred from four rDNA regions. Mol Phy-
logen and Evol 22:76–90.
Bounou S, Jabaji-Hare SH, Hogue R, Charest PM. 1999.
Polymerase chain reaction-based assay for specific de-
tection of Rhizoctonia solani AG-3 isolates. Mycol Res
103:1–8.
Boysen M, Borja M, Delmoral C, Salazar O, Rubio V. 1996.
Identification at strain level of Rhizoctonia solani AG4
isolates by direct sequence of asymmetric PCR products
of the ITS regions. Cur Gen 29:174–181.
Brem D, Leuchtmann A. 2003. Molecular evidence for host-
adapted races of the fungal endophyte Epichloe¨ bromi-
cola after presumed host shifts. Evolution 57:37–51.
Crespo A, Blanco O, Hawksworth DL. 2001. The potential
of mitochondrial DNA for establishing phylogeny and
stabilising generic concepts in the parmelioid lichens.
Taxon 50:807–819.
Cruse M, Telerant R, Gallagher T, Lee T, Taylor JW. 2002.
Cryptic species in Stachybotrys chartarum. Mycologia
94:814–822.
DePriest PT, Sikaroodi M, Lawrey JD, Diederich P. 2004.
Marchandiomyces lignicola sp. nov. shows recent and re-
peated transition between a lignicolous and a licheni-
colous habit. Mycological Research 108 (In press).
Diederich P. 2000. Host-specificity and co-evolution in lich-
enicolous fungi. In: The Fourth IAL Symposium, Pro-
gress and Problems in Lichenology at the Turn of the
Millennium, Book of Abstracts: 102. University of Bar-
celona, Barcelona.
, Schultheis B, Blackwell M. 2003. Marchandiobasi-
dium aurantiacum gen. et sp. nov., the teleomorph of
Marchandiomyces aurantiacus (Basidiomycota, Cerato-
basidiales). Mycol Res 107:523–527.
Duncan S, Barton JE, O’Brien PA. 1993. Analysis of varia-
tion in isolates of Rhizoctonia solani by random ampli-
fied polymorphic DNA assay. Mycol Res 97:1075–1082.
Etayo J, Diederich P. 1996. Lichenicolous fungi from the
western Pyrenees, France and Spain. II. More Deuter-
omycetes. Mycotaxon 60:415–428.
Gargas A, DePriest PT, Grube M, Tehler A. 1995. Multiple
origins of lichen symbioses in fungi suggested by SSU
rDNA phylogeny. Science 268:1492–1495.
Grube M, DePriest PT, Gargas A, Hafellner J. 1995. DNA
isolation from lichen ascomata. Mycological Research
99:1321–1324.
Guillaumin JJ, Anderson JB, Legrand P, Ghaharai S, Ber-
thelay S. 1996. A comparison of different methods for
the identification of genets of Armillaria spp. New Phy-
tologist 133:333–343.
Harlton CE, Le´vesque CA, Punja ZK. 1995. Genetic diversity
in Sclerotium (Athelia) rolfsii and related species. Phy-
topathology 85:1269–1281.
Harvey PR, Langridge P, Marshall DR. 2001. Genetic drift
and host-mediated selection cause genetic differentia-
tion among Gaeumannomyces graminis populations in-
fecting cereals in southern Australia. Mycol Res 105:
927–935.
Hawksworth DL. 1978. The taxonomy of lichen-forming
fungi, reflections on some fundamental problems. In:
463MOLINA ET AL: GENETIC VARIATION IN MARCHANDIOMYCES CORALLINUS
Street HE, ed. Essays in Plant Taxonomy. London: Ac-
ademic Press. p 211–243.
. 1982a. Secondary fungi in lichen symbioses: para-
sites, saprophytes and parasymbionts. J of Hattori Bot
Lab 52:357–366.
. 1982b. Co-evolution and the detection of ancestry
in lichens. J of Hattori Bot Lab 52:323–329.
. 1988a. The variety of fungal-algal symbioses, their
evolutionary significance, and the nature of lichens.
Bot J of Linnean Soc 96:3–20.
. 1988b. Coevolution of fungi with algae and cyano-
bacteria in lichen symbioses. In: Pirozynski KA, Hawks-
worth DL, eds. Coevolution of Fungi with Plants and
Animals. London: Academic Press. p 125–148.
, Rossman AY. 1997. Where are all the undescribed
fungi? Phytopathology 87:888–891.
Hibbett DS, Binder M. 2002. Evolution of complex fruiting-
body morphologies in homobasidiomycetes. Proc of R
Soc of London, B 269:1963–1969.
, Gilbert LB, Donoghue MJ. 2000. Evolutionary in-
stability of ectomycorrhizal symbioses in basidiomy-
cetes. Nature 407:506–508.
, Thorn RG. 2001. Basidiomycota: Homobasidiomy-
cetes. In: McLaughlin DJ, McLaughlin EG, Lemke PA,
eds. The Mycota, vol. VII, part B, Systematics and Evo-
lution. New York: Springer Verlag. p 121–168.
Jabaji-Hare SH, Meller Y, Gill S, Charest PM. 1990. Investi-
gation of genetic relatedness among anastomosis
groups of Rhizoctonia solani using cloned DNA probes.
Can J of Plant Path 12:393–404.
Kirk PM, Cannon PF, David JC, Stalpers JA. 2001. Ainsworth
and Bisby’s Dictionary of the Fungi. 9th ed. Egham:
CABI Bioscience.
Laroche JP, Jabaji-Hare SH, Charest PM. 1992. Differentia-
tion of two anastomosis groups of Rhizoctonia solani by
isozyme analysis. Phytopathology 82:1387–1393.
Larsson K-H, Larsson E, Ko˜ljalg U. 2004. High phylogenetic
diversity among corticioid homobasidiomycetes. Mycol
Res 108:983–1002.
Lawrey JD. 2002. Isolation and culture of lichenicolous fun-
gi. In: Kranner I, Beckett RP, Varma A, eds. Protocols
in lichenology-culturing, biochemistry, physiology and
use in biomonitoring. Berlin: Springer-Verlag. p 75–84.
, Diederich P. 2003. Lichenicolous fungi: interac-
tions, evolution and biodiversity. The Bryologist 106:
80–120.
Liu ZL, Domier LL, Sinclair JB. 1995. Polymorphism of
genes coding for nuclear 18S rRNA indicates genetic
distinctiveness of anastomosis group 10 from other
groups in the Rhizoctonia solani species complex. App
and Enviro Microb 61:2659–2664.
Lowen R, Brady BL, Hawksworth DL, Patterson RRM. 1986.
Two new lichenicolous species of Hobsonia. Mycologia
78:842–846.
O’Donnell K, Kistler HC, Tack BK, Casper HH. 2000. Gene
genealogies reveal global phylogeographic structure
and reproductive isolation among lineages of Fusarium
graminearum, the fungus causing wheat scab. Proc of
Nat Acad of Sci USA 97:7905–7910.
Punja ZK, Sun L-J. 2001. Genetic diversity among mycelial
compatibility groups of Sclerotium rolfsii (teleomorph
Athelia rolfsii) and S. delphinii. Mycol Res 105:537–546.
Rayner ADM. 1991. The challenge of the individualistic my-
celium. Mycologia 83:48–71.
Sikaroodi M, Lawrey JD, Hawksworth DL, DePriest PT. 2001.
The phylogenetic position of selected lichenicolous
fungi: Hobsonia, Illosporium and Marchandiomyces. My-
col Res 105:453–460.
Steenkamp ET, Wingfield BD, Desjardins AE, Marasas WFO,
Wingfield MJ. 2002. Cryptic speciation in Fusarium
subglutinans. Mycologia 94:1032–1043.
Taylor JW, Jacobson DJ, Kroken S, Kasuga T, Geiser DM,
Hibbett DS, Fisher MC. 2000. Phylogenetic species rec-
ognition and species concepts in fungi. Fung Gen and
Bio 31:21–32.
Vilgalys R. 1988. Genetic relatedness among anastomosis
groups in Rhizoctonia solani as measured by DNA-DNA
hybridization. Phytopathology 78:698–702.
, Gonzales D. 1990. Ribosomal DNA restriction frag-
ment length polymorphisms in Rhizoctonia solani. Phy-
topathology 78:151–158.
White TJ, Bruns TD, Lee SB, Taylor JW. 1990. Amplification
and direct sequencing of fungal ribosomal RNA genes
for phylogenetics. In: Innis MA, Gelfand DH, Sninsky
JJ, White TJ, eds. PCR Protocols: a guide to methods
and applications. San Diego: Academic Press. p 315–
322.
Worrall JJ. 1997. Somatic incompatibility in basidiomycetes.
Mycologia 89:24–36.