Content uploaded by Matthias Lutz
Author content
All content in this area was uploaded by Matthias Lutz
Content may be subject to copyright.
Systematics and Biodiversity 7(3): 297–306 Issued 1 September 2009
doi:10.1017/S1477200009990028 Printed in the United Kingdom C
The Natural History Museum
Martin Kemler1∗,
Matthias Lutz2,
Markus G¨
oker2,3,,
Franz Oberwinkler2&
Dominik Begerow1
1Lehrstuhl f¨ur Evolution und
Biodiversit¨at der Pflanzen, AG
Geobotanik, Ruhr-Universit¨at
Bochum, Universit¨atsstrasse
150, 44780 Bochum, Germany
2Lehrstuhl f¨ur Spezielle
Botanik und Mykologie,
Botanisches Institut,
Eberhard-Karls-Universit¨at,
Auf der Morgenstelle 1, 72076
T¨ubingen, Germany
3DSMZ – Deutsche Sammlung
von Mikroorganismen und
Zellkulturen GmbH,
Inhoffenstraße 7 B, 38124
Braunschweig, Germany
submitted August 2007
accepted March 2009
Hidden diversity in the
non-caryophyllaceous plant-parasitic
members of Microbotryum
(Pucciniomycotina: Microbotryales)
Abstract Members of the fungal genus Microbotryum are well-known parasites
on eudicotyledonous plant hosts. However, recent studies focused exclusively on
Microbotryum species being parasites in the anthers of Caryophyllaceae in which
strong host-specificity was confirmed by molecular analyses. Consequently, species
numbers have risen considerably as multi-host parasites were split up in so-called
cryptic species. We subjected three non-caryophyllaceous Microbotryum groups
to molecular phylogenetic analyses to see whether we would confirm multi-host
morphospecies or if host-specific cryptic species in these selected groups could be
revealed as well (i.e. a group of non-caryophyllaceous anther smuts, parasites on
different Fallopia species, and parasites on Polygonum bistorta and Polygonum vi-
viparum). We applied a multiple analysis strategy to correct for varying alignment
effects on a two-locus dataset (ITS and LSU rDNA). The results obtained by the differ-
ent approaches are uniform; high host-specificity exists in the non- caryophyllaceous
anther smuts, but overlapping host ranges occur in the parasites of Fallopia species.
Results for the parasites of Polygonum are similar, with Microbotryum bistortarum
being separated into three lineages and M. marginale forming a lineage on P. bis-
torta which apparently is conspecific with M. bistortarum p.p. Our study shows that
phylogenetic patterns within Microbotryum are much more complicated than de-
duced from morphological observations alone. Even though Microbotryum species
are highly host-specific, it is impossible to identify species based solely on host taxa
affiliation. Species status is reinstated for the anther smut on Salvia pratensis.
Key words cryptic species, host-specificity, Microbotryum salviae, molecular sys-
tematics, plant parasites
Introduction
Estimates of biodiversity extrapolate that over 50% of species
and much more than 50% of individuals on earth are parasitic
organisms (Begon et al., 2006). Thus, their role in ecosys-
tems can not be underestimated as they contribute highly to
the diversity of interactions between organisms. Some organ-
isms are described as parasitising multiple hosts, others as
occurring only on a single host species. Thereby, the range
of specificity in many ecological relationships, not only in
parasitic interactions, has been shown to be much more vari-
able over geographical space than was thought a few years
ago (i.e. geographic mosaic of coevolution; Thompson, 2005).
The spectrum of diversity of interactions ranges from differ-
∗Corresponding author. Email: martin.kemler@rub.de
ences between individual organisms over divergent popula-
tions of one species to the specificity of different closely related
species.
Quite recently, the advances in molecular systematics
have played an important role in revealing genetically dis-
tinct cryptic species within morphologically similar ones (e.g.
Fernandez et al., 2006; Robba et al., 2006; Smith et al., 2006).
This uncovering of an increasing number of cryptic species has
profound consequences for the understanding of speciation in
a group of organisms. However, in order to identify the pat-
terns of cryptic diversity and to elucidate potential underlying
mechanisms of speciation, thorough genus level based invest-
igations are needed (Trontelj & Fiser, 2009). The observation
of cryptic species has become especially evident in parasitic
organisms (Jousson et al., 2000; G¨
oker et al., 2007, 2009)
and one question raised by this discovery is how many of the
297
298 Martin Kemler et al.
multi-host parasites, and especially obligate ones, so far de-
scribed are really unspecific in their choice of hosts or whether
narrow host-specificity is the rule.
For plant–fungal relations, Zhou and Hyde (2001) defined
host-specificity as ‘a relationship in which a fungus derives its
nutrition from a live host plant during some phase of its life
cycle, and is restricted to a particular host or group of re-
lated species, but does not occur on other unrelated plants in
the same habitat’. In the obligate plant-parasitic basidiomycet-
ous fungal genus Microbotryum there are a number of species
parasitising several hosts. In the anther smuts of the Caryo-
phyllaceae, the best-known members of the genus, it has been
discussed for a very long time whether different host-specific
lineages exist and whether these should be treated as individual
species or rather as formae speciales (Liro, 1924; Lindeberg,
1959). Recent studies, based on molecular phylogenetic data,
clearly demonstrated that in the anther smuts of the Caryo-
phyllaceae much narrower species delimitations should be ap-
plied than was previously thought, in accordance with frequent
high host-specificity (Lutz et al., 2005, 2008; Kemler et al.,
2006). By sequencing the ITS region of the nuclear ribosomal
DNA and correlating these characters with morphological and
host attributes, new species (Microbotryum adenopetalae,M.
chloranthae-verrucosum,M. minuartiae,M. saponariae,M,
silenes-acaulis) were recognised within the caryophyllaceous
anther smuts (Lutz et al., 2005, 2008). High host-specificity
was also found by a multi-gene phylogenetic approach (Le Gac
et al., 2007) showing similar results as earlier studies based
solely on ITS sequence data (Lutz et al., 2005; Kemler et al.,
2006). Furthermore, population studies in the anther smuts
have shown that these parasites are often restricted to a few
host species or even a single species, and that there exists only
restricted or no gene flow between spatially adjacent parasites
on different hosts (Shykoff et al., 1999; Bucheli et al., 2000,
2001).
The goal of this study was to assess whether the applica-
tion of molecular methods would reveal that multi-host para-
sitic Microbotryum species on non-caryophyllaceous hosts
should be split up in highly host-specific parasites. There-
fore, we inferred phylogenies from concatenated ITS and LSU
sequence data for parasites of three selected groups of non-
caryophyllaceous hosts: a group of anther smuts including
Microbotryum betonicae on Stachys alopecuros and Salvia
pratensis,M. pinguiculae on Pinguicula alpina,andM. in-
termedium on Scabiosa spp.; M. anomalum on three different
Fallopia species; and a group of parasites on Polygonum bis-
torta and P. viviparum.
Materials and methods
The specimens investigated are listed in Table 1, ‘Supplement-
ary data’ on Cambridge Journals Online: http://www.journals.
cup.org/abstract_S1477200009990028
Nomenclature and assignment of specimens to species
was based on the current taxonomy in Microbotryum and fol-
lows Lutz et al. (2005) and V´
anky (1994, 1998).
DNA isolation, PCR, and sequencing
Ribosomal rDNA was used as a suitable genomic region
for reconstructing phylogenetic relationships in Microbotryum
(Lutz et al. 2005, 2008) before and has been used successfully
in other studies of smut fungal systematics (e.g. Begerow et al.,
2002; Stoll et al., 2005). In order to extract genomic DNA the
DNeasyTM Plant Mini Kit (Qiagen, Germany) was used.
The ITS region, localised between the 18S and 28S rRNA
genes, was amplified using the primer pairs ITS1 and ITS4
(White et al., 1990) or ITS1f and ITS4 (Gardes & Bruns, 1993),
respectively, to obtain an approximately 650 bp long DNA
fragment. The LSU region was amplified using the primer pair
LR0R and NL4 (O’Donnell, 1992, 1993; White et al., 1990)
to obtain an approximately 550 bp long DNA fragment. To
purify PCR products, the QIAquickTM Kit (Qiagen, Germany)
was used. Samples were sequenced with the BigDyeTM Ter-
minator Cycle Sequencing Kit V3.1 (Applied Biosys-
tems) on an automatic sequencer (ABI 3100 Genetic
Analyser). DNA sequences were deposited in GenBank;
accession numbers are given in Table 1, ‘Supplementary data’
on Cambridge Journals Online: http://www.journals.cup.org/
abstract_S1477200009990028
Alignment and phylogenetic analyses
Previous phylogenetic studies in Microbotryum have shown
that support values in phylogenetic analyses for different
groups depend to a large extend on the underlying DNA substi-
tution model and accordingly on the algorithm used for align-
ing (Kemler et al., 2006). This study also showed that these ef-
fects can be accounted for by use of a multiple analysis strategy
(Lee, 2001). Therefore, the obtained sequences were aligned
using three different alignment algorithms as implemented in
MAFFT 6.502 (Katoh et al., 2002, 2005), PCMA (Pei et al.,
2003), and POA (Lee et al., 2002), respectively, to account
for the variation in support values. MAFFT was run in default
mode with the maximum number of iterations set to 1000.
PCMA was also run in default mode. POA was run in pro-
gressive mode as invoked by a UNIX shell wrapper script that
also computes pair-wise sequence similarities with BLASTN;
the script is available at http://www.goeker.org/mg/aop/. As
suggested by Giribet and Wheeler (1999) and Gatesy et al.
(1993), to obtain reproducible results we avoided manipula-
tion of the alignment by hand as well as manual exclusion of
ambiguous sites. Only parts where approximately 30% or more
of the sequences had leading or trailing gaps were removed.
To obtain the appropriate substitution model for
Neighbour-Joining analyses, each alignment was analysed
with Modeltest 3.7 (Posada & Crandall, 1998) using the Akaike
information criterion. Under the respective model found,
NJ analyses were then performed with PAUP∗(Swofford,
2002) using the BioNJ algorithm (Gascuel, 1997). One thou-
sand rounds of BioNJ bootstrapping (Felsenstein, 1985) were
performed with PAUP∗under the same settings. Maximum
Likelihood (ML; Felsenstein, 1981) analyses were conduc-
ted with the RAxML 7.0.4 software (Stamatakis, 2006a) in-
voked with the –m GTRMIX option and the number of start-
ing trees for heuristic search set to 100. ML bootstrap analysis
Non-caryophyllaceous plant-parasites in Microbotryum 299
was performed with RAxML under the –m GTRCAT option
(Stamatakis, 2006b) and 1000 replicates. GTRCAT is an ef-
ficient approximation of the general time reversible (GTR;
Rodr´
ıguez et al., 1990) model of site substitution combined
with a gamma distribution; GTRMIX uses GTRCAT during
heuristic search but the full GTR+G model for the final likeli-
hood computation (Stamatakis, 2006a).
Maximum parsimony (MP; Fitch, 1971) analysis was
performed using a heuristic search in PAUP with 1000 replic-
ates of random sequence addition and subsequent TBR branch
swapping (multrees option in effect, steepest decent option in
effect). No more than 25 trees with a score greater than or
equal to 1 were saved per replicate. Bootstrap analysis under
the MP criterion was conducted in PAUP using 1000 replic-
ates. In each replicate, 25 rounds of random sequence addition
followed by TBR branch swapping were performed.
To test for compatibility of ITS and LSU based phylo-
genies the two DNA sets were analysed separately using the
MAFFT alignment and RAxML (−f a option and 100 rep-
licates). The obtained phylogenies were afterwards inspected
by eye to check for significantly supported incongruencies
between the topology.
As an outgroup two members of the Ustilentylomata-
ceae, which are supposed to be the sister family of the
Microbotryaceae (Bauer et al., 2006), were used. We also in-
cluded a comprehensive sample of other microbotryaceous
genera in our dataset, some of which had been shown to
be difficult to delineate from Microbotryum in earlier stud-
ies (Kemler et al., 2006). All alignments together with the
resulting trees inferred from them are included in the sup-
plementary material, which is available as ‘Supplementary
data’ on Cambridge Journals Online: http://www.journals.cup.
org/abstract_S1477200009990028
Results
DNA alignments
The alignments obtained with MAFFT had a total length of 851
bp for the ITS and 657 bp for the LSU data set. After truncation
of leading and trailing gaps, the alignments were concatenated
resulting in an alignment of 1159 bp including 333 parsimony-
informative sites. The respective values for PCMA were
838 bp, 656 bp, 1169 bp and 355 parsimony-informative sites,
and for POA 912 bp, 673 bp, 1182 bp and 298 parsimony-
informative sites. These differences are almost entirely caused
by the ITS locus. The LSU region was very homogeneous and
contained hardly any indels. Therefore alignments of the LSU
region differed only slightly or not at all between the different
alignment programs applied. The best log likelihood values
obtained and the final estimates for the alpha parameter were:
MAFFT, −7630.82, 0.23663; PCMA, −8045.84, 0.25161; and
POA, −7451.55, 0.22745. Numbers and lengths of the most
parsimonious trees found as well as their retention indices
(Farris, 1989) were: MAFFT, 9, 1235, 0.830; PCMA, 3, 1324,
0.847; and POA, 2, 1155, 0.823. The best substitution models
obtained for the NJ analyses using MAFFT were: TrN+I+G,
gamma distribution shape parameter 0.4135, proportion of in-
variable sites 0.3466. The respective values for PCMA are:
GTR+I+G, 0.4504 and 0.3300; and for POA are: GTR+I+G,
0.4990 and 0.4310.
Phylogenetic trees
General tree topology
No well-supported incongruencies could be found by the in-
spection of the phylogenies based on ITS and LSU (see supple-
mental material). As in a previous study (Kemler et al., 2006),
the topologies inferred from the different alignment algorithms
were similar but the phylogenies differed in bootstrap support
for certain groups. In Fig. 1 the ML tree with branch length
obtained from the MAFFT alignment is depicted. A majority-
rule consensus of the three most parsimonious trees obtained
from the PCMA alignment is shown in Fig. 2, together with the
host plants of the examined parasites. Branches that appeared
also in the strict consensus are drawn in bold. For three groups
of particular interest (non-caryophyllaceous anther smuts, Mi-
crobotryum anomalum, and parasites on Polygonum viviparum
and P. bistorta) the different support values under all condi-
tions are provided in Fig. 3.
Parasite taxa of special interest
A group of anther smuts, comprising Microbotryum betonicae,
M. pinguiculae and M. intermedium, formed a monophyletic
group supported by moderate to high bootstrap values. Within
these anther smuts, support values for two clusters, one con-
taining Microbotryum betonicae on Salvia pratensis and M.
intermedium and the other one M. betonicae on Stachys alo-
pecuros and M. pinguiculae, differed between the alignment
algorithms, and high bootstrap support for these two clusters
was only obtained for the MP analysis under the PCMA align-
ment. A sister-group relationship between the two Lamiaceae
parasites was only weakly supported by the ML analysis un-
der the MAFFT alignment (Fig. 1). All conditions produced
high bootstrap support for the groups that contained several
specimens of Microbotryum betonicae on Salvia pratensis,M.
betonicae on Stachys alopecuros,M. intermedium on Scabiosa
species, and M. pinguiculae on Pinguicula alpina (Fig. 3a).
The group comprising both M. anomalum and the caryo-
phyllaceous anther smuts was supported by PCMA for NJ and
MP and by POA only for ML. Microbotryum anomalum on
Fallopia species was supported as a monophyletic group with
a bootstrap value of 100 for all inference methods under all
alignments. A separation of Microbotryum anomalum in two
groups (mk067 and ml1163 vs. mk312, mk664 and ml1162)
was moderately supported by all methods except for NJ and
MP under the MAFFT algorithm (see Fig. 3b). Within the
caryophyllaceous anther smuts, all recognised species were
supported by high bootstrap values under all three alignments.
All algorithms supported the monophylum containing
both Microbotryum bistortarum on Polygonum bistorta and
P. viviparum as well as M. marginale on P. bistorta.Mi-
crobotryum bistortarum on Polygonum viviparum formed a
well-supported group under all conditions, except for the MP
under POA. The members of Microbotryum marginale on
300 Martin Kemler et al.
Figure 1 Best tree obtained in the Maximum Likelihood analysis of the concatenated ITS and LSU MAFFT alignments. The phylogeny was
inferred with RAxML under the GTRMIX model approximation. Branch lengths are scaled in terms of the expected numbers of
nucleotide substitutions per site. Numbers above branches are bootstrap values inferred with RAxML under GTRCAT. M. =
Microbotryum; Sphac. =Sphacelotheca. The tree was rooted with Ustilentyloma brefeldii and Ustilentyloma fluitans. Major
Microbotryum subgroups as referred to in the discussion are indicated.
Non-caryophyllaceous plant-parasites in Microbotryum 301
Figure 2 Majority-rule consensus of the three most parsimonious trees inferred from the concatenated PCMA-aligned ITS and LSU datasets.
Heuristic search for the best trees was conducted with PAUP∗. The tree was rooted with Ustilentyloma brefeldii and Ustilentyloma
fluitans. The symbols on the branches indicate the magnitude of parsimony bootstrap support for the different concatenated alignments
constructed with MAFFT (top), PCMA (bottom left) and POA (bottom right), respectively. Bold lines show groupings that appear in
the strict consensus tree. The names of the host plants are indicated on the right side. M. =Microbotryum; Sphac. =Sphacelotheca.
302 Martin Kemler et al.
Figure 3 For legend see facing page.
Non-caryophyllaceous plant-parasites in Microbotryum 303
Polygonum bistorta grouped together with specimen mk644 of
M. bistortarum on P. bistorta. As a sister group of the whole
clade, with moderate to high support, a specimen of Micro-
botryum bistortarum on P. bistorta was placed. Specimens of
M. pustulatum formed a moderately to highly supported mono-
phyletic group under all conditions, with the exception of MP
under MAFFT and NJ and MP under PCMA. Under most
conditions, Microbotryum pustulatum formed a monophylum
with M. bosniacum,Liroa emodensis,M. tuberculiforme and
Sphacelotheca polygoni-serrulati with moderate to high sup-
port (Fig. 3c).
Discussion
On the one hand the present study confirms several phylo-
genetic groupings that have been observed previously (Kemler
et al., 2006); on the other hand new results emerge from this
study. The main findings are described below.
Anther smuts on non-caryophyllaceous hosts
In a former study, Kemler et al. (2006) showed that anther
parasitism in Microbotryum may have evolved independently
at least twice. One group contains the caryophyllaceous an-
ther smuts and M. scabiosae on Knautia arvensis. The second
clade, which is also confirmed to be monophyletic by our cur-
rent study, comprises anther parasites of the three different
host families Dipsacaceae, Lamiaceae and Lentibulariaceae.
Microbotryum betonicae, considered in the literature as
parasitising both Salvia pratensis and Stachys species, shows
a clear separation between the parasites on Salvia pratensis
and on Stachys alopecuros. As the three alignment algorithms
resulted in different internal relationships of this group of an-
ther smuts, it could not be clarified in this study if these two
lineages form sister taxa or not. Whereas the DNA sequences
within the two clusters are identical, genetic distances between
the two groups are considerable, i.e. the proportion of sites dif-
fering between the groups being 2.64% for MAFFT, 2.58% for
PCMA and 2.51% for POA. Ciferri (1931) already suggested
two distinct species, based on spore size differences, and de-
scribed Ustilago salviae as the parasite of Salvia pratensis,
which Ferraris (1902) referred to as Ustilago violacea forma
salviae. Other studies (Kirchner, 1923; V´
anky, 1994), how-
ever, were not able to find differences in spore size between
these parasites and merged them into Microbotryum betonicae.
Nevertheless, it has been shown recently that spore size is not a
reliable character for defining monophyletic lineages in the ca-
ryophyllaceous anther smuts (Lutz et al., 2005). In that study,
spore sizes were variable among specimens of a single phylo-
genetic lineage, with intraspecific variation being sometimes
greater than interspecific variation. As our molecular study
demonstrates genetic separation between the two lineages that
correlates well with host preference, we reinstate the anther
parasite on Salvia pratensis as a species of its own.
Microbotryum salviae (Ferraris) Kemler & M. Lutz
comb. nov.
Mycobank No. MB 510807
Basionym Ustilago violacea (Pers.) Roussel f. salviae
Ferraris, Ann. R. Istit. Bot. Roma 9: 190, 1902.
Basionym Ustilago salviae (Ferraris) Cif., Ann. Mycol. 29:
5, 1931.
The splitting of the morphospecies Microbotryum betonicae
based on molecular characters extends the number of Micro-
botryum lineages in which cryptic species have been observed.
Microbotryum pinguiculae has been described as parasit-
ising several species of Pinguicula, but during our sampling
efforts we never encountered members of Pinguicula vulgaris
that were infected by M. pinguiculae even if they grew in close
proximity of infected Pinguicula alpina specimens (U. Bloss,
M. Kemler & M. Lutz, unpubl.). Other authors have found this
pattern in sympatric Pinguicula populations, too (Liro, 1924,
and references therein). Thus, the infections of Pinguicula vul-
garis observed by other workers may indicate a phylogenetic
lineage different from that on Pinguicula alpina.However,as
long as further studies are lacking, incipient host jumps cannot
be ruled out, as has been described for anther smuts in Ca-
ryophyllaceae (Antonovics et al., 2002; Lopez-Villavicencio
et al., 2005).
Our study reveals that within the non-caryophyllaceous
anther smuts jumps between very distantly related hosts most
likely have occurred. This seems to be the case even for
the parasites occurring on Stachys alopecuros and Salvia
pratensis, which belong to different subfamilies of Lamiaceae
(Wagstaff & Olmstead, 1997). Furthermore, if the relation-
ships inside this group of anther smuts evolved as indicated by
some of the phylogenetic trees obtained in our present study,
i.e. Microbotryum betonicae clustering with M. pinguiculae,
and M. salviae clustering with M. intermedium, at least two
host jumps to or from Lamiaceae must have occurred. This
picture of successive jumps to distantly related hosts might be
reconsidered if other hosts for Microbotryum betonicae (one
additionally described host), M. pinguiculae (three addition-
ally described hosts) and M. intermedium (four additionally
described hosts) will be added in future studies.
Figure 3 The phylogenetic relationships illustrated in Figs 3a–3c are compiled from subtrees of the ML topology shown in Fig. 1 and the MP
topology shown in Fig. 2. Nodes marked with a character indicate the bootstrap support for the corresponding node as printed in the
table below each figure. Nodes lacking bootstrap values above 50% in each analysis are not marked; characters from different
subtrees do not correspond to each other. The tables are sorted by the different alignment algorithms and inference methods. ML:
Maximum likelihood; MP: Maximum parsimony; NJ: Neighbour joining. (a) Phylogenetic relationships and support values between
the non-caryophyllaceous anther smuts. (b) Phylogenetic relationships and support values between specimens of Microbotryum
anomalum,aswellasbetweenM. anomalum and the caryophyllaceous anther smuts. (c) Phylogenetic relationships and support
values of Microbotryum bistortarum,M. marginale,andM. pustulatum on Polygonum bistorta and P. viviparum.
304 Martin Kemler et al.
Parasites on Fallopia
As it is hypothesised that the ancestor of all Microbotryum
species was a pathogen of Polygonaceae (Kemler et al.,
2006), evaluating host specificities within recent parasites
on Polygonaceae should be a major topic. One group of
Polygonaceae smuts in our study are the parasites on Fal -
lopia species. In our study one of the two specimens on Fal -
lopia convolvulus clusters together with one on F. dumetorum
and not with the other Microbotryum anomalum specimens,
including an additional parasite of F. convolvulus.How-
ever, there is no apparent geographical correlation between
samples that cluster together (Table 1, ‘Supplementary data’ on
Cambridge Journals Online: http://www.journals.cup.org/
abstract_S1477200009990028). Interestingly, it has been ob-
served by inoculation experiments that the smut of Fallopia
convolvulus was not able to infect specimens of F. dumetorum,
and vice versa (Liro, 1924). Based on this observation and the
different spore mass colours, Liro described two species, Ustil-
ago anomala on Fallopia dumetorum and U. carnea on F. con-
volvulus. Unfortunately, we were not able to either confirm or
reject these different colour morphs, regarding the herbarium
specimens we had access to. Therefore we were unable to cor-
relate this character to the two non-host-specific monophyletic
lineages within Microbotryum anomalum that were supported
in some of our phylogenies. Nevertheless, both Liro’s (1924)
as well as our observations indicate that several distinct lin-
eages of Microbotryum anomalum may exist. According to
our molecular data, these cryptic species, if confirmed by fu-
ture studies, would then have overlapping host ranges. On
the other hand, genetic distances within Microbotryum anom-
alum (ranging from 0.009% to 1.02% differing sites for the
MAFFT alignment) as it is currently described seem to be in
the same range as has been observed within other species of
Microbotryum, e.g. in M. lychnidis-dioicae the proportion of
differing sites for the MAFFT alignment is 0.86%. Therefore,
many more samples should be considered before taxonomic
consequences might be taken.
The third host in this group, Fallopia aubertii,isanin-
troduced plant in European gardens and can occasionally be
found naturalised in the wild (Sebald et al., 1993). It is an
important parent species for hybrids with the invasive Fal -
lopia japonica (Bailey & Stace, 1992; Tiebre et al., 2007).
Future studies in this group should assess whether infection
of Fallopia aubertii by Microbotryum can be linked to its re-
cent introduction in Europe and whether F. japonica also is
susceptible to pathogens of native European Fallopia species.
Parasites on Polygonum bistorta and Polygonum
viviparum
The other group of parasites on Polygonaceae occurs on Poly -
gonum bistorta and P. viviparum. According to the traditional
morphospecies concept, the separation between Microbotryum
bistortarum,M. pustulatum and M. marginale is solely based
on the location of the sori. Microbotryum bistortarum produces
its spores in flowers and bulbils, M. pustulatum produces pus-
tules in the centre of leaves, and M. marginale produces sori in
the margins of leaves. All three parasites can be found
in the same geographical site (e.g. Alp Flix, Gris-
ons, Switzerland; see Table 1, ‘Supplementary data’ on
Cambridge Journals Online: http://www.journals.cup.org/
abstract_S1477200009990028). The species status of these
Polygonum parasites and their phylogenetic relationship to
each other has been debated for quite some time (Liro, 1924).
Molecular data very clearly separate Microbotryum pustulatum
from a clade comprising both M. bistortarum and M. marginale
specimens (Figs 1 and 2). However, within the group contain-
ing Microbotryum bistortarum and M. marginale molecular
results point to a more complicated pattern of host-parasite
species relationships. In fact, tree topologies indicate that the
current taxonomic description of Microbotryum bistortarum
may contain more than one parasite species. The sequences
obtained from Microbotryum bistortarum parasitising on Pol y-
gonum viviparum form a well-supported group, which may be
regarded as a species of its own. In contrast, the sequences ob-
tained from Microbotryum bistortarum and M. marginale on
Polygonum bistorta form two distinct groups (Fig. 1) that are
not in accordance with current taxonomy. One of these groups
comprises all M. marginale specimens and one M. bistortarum
specimen, whereas the other lineage consists of one specimen
of M. bistortarum only. Whether these two lineages represent
two different species, or if the observed pattern is a result of
incomplete lineage sorting of alleles, remain to be clarified by
further studies. As long as this issue is not resolved, a conser-
vative taxonomic treatment should be applied; we suggest the
use of ‘Microbotryum bistortarum/marginale complex’ when
referring to these pathogens.
Our results demonstrate that sorus location alone is not
a sufficient diagnostic character for species delineation, but
needs further support from other sources. For instance, ar-
tificial inoculation could clarify whether sorus location is a
stable character or can vary on the same host. Such plasticity
of sorus location has been shown quite recently in the anther
smuts of Caryophyllaceae. In an inoculation experiment in-
fected flowers have been obtained that produced spores in the
capsules rather than in the anthers (Sloan et al., 2008). Also,
other parasites in the Microbotryaceae (e.g. Liroa emodensis)
have a rather plastic sorus location (Piepenbring, 2004).
Our study has shown a complicated phylogenetic pattern
in the parasites on these two Polygonum species. Therefore,
future research in this group should focus on both obtaining
and analysing additional collections as well as on conducting
cross-infection experiments. For instance, so far molecular
data are unavailable for Microbotryum pustulatum on Pol y -
gonum viviparum; such data might help to resolve the rela-
tionship between M. pustulatum and M. bistortarum on this
host.
Conclusions
Recent molecular studies in Microbotryum clearly show that
phylogenetic patterns within these plant parasites are complic-
ated and may differ from the degree of relationship inferred
from morphological and host data alone. For instance, the
work of Kemler et al. (2006) indicates that there may be two
Non-caryophyllaceous plant-parasites in Microbotryum 305
independent lineages of anther smuts and also that parasit-
ism on Dipsacaceae may have evolved twice. Parasites from
Polygonaceae may be considered as ancestors of Micro-
botryum as they exhibit a paraphyletic distribution. In cary-
ophyllaceous anther smuts, a considerable increase in spe-
cies numbers has resulted from the application of molecular
phylogenetic methods (Lutz et al., 2005, 2008; Kemler et al.,
2006). These studies also demonstrated that high host spe-
cificity can be observed in most lineages. Our present study
on non-caryophyllaceous anther smuts confirms this observa-
tion for other groups of Microbotryum. Thereby we show that
the number of Microbotryum species to be regarded as patho-
gens of single host species (as particularly evident in the case
of Polygonum bistorta) may also rise if a thorough molecu-
lar investigation is carried out. Thus, separation or merging
of Microbotryum species is not straightforward, but should be
based on a sufficient sampling effort as well as a careful cross-
comparison of molecular results, host data, and in many cases
also the outcome of inoculation experiments.
Acknowledgements
We are indebted to M. Weiß, M. Piatek and K. V ´
anky for help with the
new combination of M. salviae and to U. Simon and W. Maier for crit-
ically reading the text and improving the language. Financial support
by the Deutsche Forschungsgemeinschaft is gratefully acknowledged.
References
ANTONOVICS,A.,HOOD,M.E.&PARTAIN, J.L. 2002. The ecology
and genetics of a host shift: Microbotryum as a model system.
American Naturalist 160, S40–S53.
BAILEY,J.P.&STAC E , C.A. 1992. Chromosome number, morpho-
logy, pairing, and DNA values of species and hybrids in the genus
Fallopia (Polygonaceae). Plant Systematics and Evolution 180,
29–52.
BAUER ,R.,BEGEROW,D.,SAMPAIO,J.P., WEISS,M.&OBERWINKLER,
F. 2006. The simple-septate basidiomycetes: a synopsis. Mycolo-
gical Progress 5, 41–66.
BEGEROW,D.,LUTZ,M.&OBERWINKLER, F. 2002. Implications of
molecular characters for the phylogeny of the genus Entyloma.
Mycological Research 106, 1392–1399.
BEGON,M.,TOWNSEND,C.R.&HARPER, J.L. 2006. Ecology – From
Individuals to Ecosystems. Fourth Edition. Blackwell Publishing,
Malden.
BUCHELI,E.,GAUTS CHI ,B.&SHYKOFF, J.A. 2000. Host-specific
differentiation in the anther smut fungus Microbotryum violaceum
as revealed by microsatellites. Journal of Evolutionary Biology 13,
188–198.
BUCHELI,E.,GAUT SCH I,B.&SHYKOFF, J.A. 2001. Differences in
population structure of the anther smut fungus Microbotryum vi-
olaceum on two closely related host species, Silene latifolia and S.
dioica.Molecular Ecology 10, 285–294.
CIFERRI, R. 1931. Quinta contribuzione allo studio degli Ustilaginales.
Annals of Mycology 29, 1–74.
FARRIS, J.S. 1989. The retention index and the rescaled consistency
index. Cladistics 5, 417–419.
FELSENSTEIN, J. 1981. Evolutionary trees from DNA sequences: a
maximum likelihood approach. Journal of Molecular Evolution
17, 368–376.
FELSENSTEIN, J. 1985. Confidence limits on phylogenies: an approach
using the bootstrap. Evolution 39, 783–791.
FERRARIS, T. 1902. Primo elenco di Funghi del Piemonte. Annuario
Reale Istituto Botanico di Roma 9, 187–220.
FERNANDEZ, C.C., SHEVOCK, J.R., GLAZER,A.N.&THOMPSON,J.N.
2006. Cryptic species within the cosmopolitan desiccation-tolerant
moss Grimmia laevigata.Proceedings of the National Academy of
Sciences USA 103, 637–642.
FITCH, W.M. 1971. Towards defining the course of evolution: minimal
change for a specified tree topology. Systematic Zoology 20, 406–
416.
GASCUEL, O. 1997. BIONJ, an improved version of the NJ algorithm
based on a simple model of sequence data. Molecular Biology and
Evolution 14, 685–695.
GATESY,J.,DESALLE,R.&WHEELER, W. 1993. Alignment-
ambiguous nucleotide sites and the exclusion of systematic data.
Molecular Phylogenetics and Evolution 2, 152–157.
GARDES,M.&BRUNS, T.D. 1993. ITS primers with enhanced spe-
cificity for basidiomycetes – application to the identification of
mycorrhizae and rusts. Molecular Ecology 2, 113–118.
GIRIBET,G.&WHEELER, W.C. 1999. On gaps. Molecular Phylogen-
etics and Evolution 13, 132–143.
G¨
OKER,M.,VOGLMAYR,H.,RIETHM ¨
ULLER,A.&OBERWINKLER,F.
2007. How do obligate parasites evolve? A multi-gene phylogen-
etic analysis of downy mildews. Fungal Genetics and Biology 44,
105–122.
G¨
OKER,M.,VOGLMAYR,H.,GARC´
IA-BL´
AZQUEZ,G.&
OBERWINKLER, F. 2009. Species delimitation in downy mil-
dews: the case of Hyaloperonospora in the light of nuclear
ribosomal internal transcribed spacer and large subunit sequences.
Mycological Research 113, 308–325.
JOUSSON,O.,BARTOLI,P.&PAWL OW SK I, J. 2000. Cryptic speciation
among intestinal parasites (Trematoda: Digenea) infecting sym-
patric host fishes (Sparidae). Journal of Evolutionary Biology 13,
778–785.
KATO H ,K.,MISAWA,K.,KUMA,K.&MIYATA, T. 2002. MAFFT: a
novel method for rapid multiple sequence alignment based on fast
Fourier transform. Nucleic Acids Research 30, 3059–3066.
KATO H ,K.,KUMA,K.,TOH,H.&MIYATA, T. 2005. MAFFT version 5:
improvement in accuracy of multiple sequence alignment. Nucleic
Acids Research 33, 511–518.
KEMLER,M.,G
¨
OKER,M.,OBERWINKLER,F.&BEGEROW, D. 2006.
Implications of molecular characters for the phylogeny of the Mi-
crobotryaceae (Basidiomycota: Urediniomycetes). BMC Evolu-
tionary Biology 6, 35.
KIRCHNER, O. 1923. Der Antherenbrand von Salvia,Ustilago beton-
icae Beck. Zeitschrift f¨
ur Pflanzenkrankheiten 33, 97–104.
LEGAC,M.,HOOD, M.E., FOURNIER,E.&GIRAUD, T. 2007. Phylo-
genetic evidence of host-specific cryptic species in the anther smut
fungus. Evolution 61, 15–26.
LEE, M.S.Y. 2001. Unalignable sequences and molecular evolution.
Trends in Ecology and Evolution 16, 681–685.
LEE,C.,GRASSO,C.&SHARLOW, M.F. 2002. Multiple sequence
alignment using partial order graphs. Bioinformatics 18, 452–464.
LINDEBERG, B. 1959. Ustilaginales of Sweden. Symbolae Botanicae
Upsalienses 16, 1–175.
LIRO, J.I. 1924. Die Ustilagineen Finnlands I. Annales Academiae
Scientiarum Fennicae, Series A 17, 1–636.
LOPEZ-VILLAVICENCIO,M.,ENJALBERT,J.,HOOD, M.E., SHYKOFF,
J.A., RAQUI N,C.&GIRAUD, T. 2005. The anther smut disease
on Gypsophila repens: a case of parasite sub-optimal performance
following a recent host shift? Journal of Evolutionary Biology 18,
1293–1303.
LUTZ,M.,G
¨
OKER,M.,PIATEK,M.,KEMLER,M.,BEGEROW,D.&
OBERWINKLER, F. 2005. Anther smuts of Caryophyllaceae: Mo-
lecular characters indicate host-dependent species delimitation.
Mycological Progress 4, 225–238.
LUTZ,M.,PIATEK,M.,KEMLER,M.,CHLEBICKI,A.&OBERWINKLER,
F. 2008. Anther smuts of Caryophyllaceae: Molecular analyses
reveal further new species. Mycological Research 112, 1280–
1296.
O’DONNELL, K.L. 1992. Ribosomal DNA internal transcribed spacers
are highly divergent in the phytopathogenic ascomycete Fusarium
306 Martin Kemler et al.
sambucinum (Gibberella pulicaris). Current Genetics 22, 213–
220.
O’DONNELL, K.L. 1993. Fusarium and its near relatives. In:
REYNOLDS,D.R.&TAY LO R , J.W., Eds., The Fungal Holomorph:
Mitotic, Meiotic and Pleomorphic Speciation in Fungal Syste-
matics. CAB International, Wallingford, pp. 225–233.
PEI,J.,SADREYEV,R.&GRISHIN, N.V. 2003. PCMA: fast and ac-
curate multiple sequence alignment based on profile consistency.
Bioinformatics 19, 427–428.
PIEPENBRING, M. 2002. Morphology of Liroa emodensis (Micro-
botryales, Basidiomycota) on Polygonum chinense.Fungal Sci-
ence 17, 55–64.
POSADA,D.&CRANDALL, K.A. 1998. Modeltest: testing the model
of DNA substitution. Bioinformatics 14, 817–818.
ROBBA,L.,RUSSELL, S.J., BARKER,G.L.&BRODIE, J. 2006. As-
sessing the use of the mitochondrial cox1 marker for use in DNA
barcoding of red algae (Rhodophyta). American Journal of Botany
93, 1101–1108.
RODR´
IGUEZ,F.,OLIVER, J.L., MAR´
IN,A.&MEDINA, J.R. 1990. The
general stochastic model of nucleotide substitution. Journal of
Theoretical Biology 142, 485–501.
SEBALD,O.,SEYBOLD,S.&PHILIPPI, G. 1993. Die Farn- und
Bl¨
utenpflanzen Baden-W¨
urttembergs.Band1.EugenUlmer
GmbH & Co., Stuttgart.
SHYKOFF, J.A., MEYHOFER,A.&BUCHELI, E. 1999. Genetic isola-
tion among host races of the anther smut fungus Microbotryum
violaceum on three host plant species. International Journal of
Plant Sciences 160, 907–916.
SLOAN, D.B., GIRAUD,T.&HOOD, M.E. 2008. Maximized virulence
in a sterilizing pathogen: the anther-smut fungus and its co-evolved
hosts. Journal of Evolutionary Biology 21, 1544–1554.
SMITH, M.A., WOODLEY, N.E., JANZEN, D.H., HALLWACHS,W.&
HEBERT, P.D. 2006. DNA barcodes reveal cryptic host-specificity
within the presumed polyphagous members of a genus of parasitoid
flies (Diptera: Tachinidae). Proceedings of the National Academy
of Sciences USA 103, 3657–3662.
STAM ATAK I S, A. 2006a. RAxML-VI-HPC: maximum likelihood-
based phylogenetic analyses with thousands of taxa and mixed
models. Bioinformatics 22, 2688–2690.
STAM ATAK I S, A. 2006b. Phylogenetic models of rate heterogeneity:
A high performance computing perspective. In: Proceedings of
IPDP2006, Rhodos, Greece.
STOLL,M.,BEGEROW,D.&OBERWINKLER, F. 2005. Molecular phylo-
geny of Ustilago,Sporisorium, and related taxa based on com-
bined analyses of rDNA sequences. Mycological Research 109,
342–356.
SWOFFORD, D.L. 2002. PAU P∗. Phylogenetic Analysis using Parsi-
mony (∗and other methods). Version 4. Sinauer Associates,
Sunderland.
THOMPSON, J.N. 2005. The Geographic Mosaic of Coevolution.
University of Chicago Press, Chicago.
TIEBRE, M.S., VANDERHOEVEN,S.,SAAD,L.&MAHY, G. 2007. Hy-
bridization and sexual reproduction in the invasive alien Fallopia
(Polygonaceae) complex in Belgium. Annals of Botany 99, 193–
203.
TRONTELJ,P.&FISER, C. 2009. Cryptic species diversity should not
be trivialised. Systematics and Biodiversity 7, 1–3.
V´
ANKY, K. 1994. European Smut Fungi. Gustav Fischer Verlag,
Stuttgart.
V´
ANKY, K. 1998. The genus Microbotryum (smut fungi). Mycotaxon
67, 33–118.
WAGSTAFF ,S.J.&OLMSTEAD, R.G. 1997. Phylogeny of Labiatae and
Verbenaceae inferred from rbcL sequences. Systematic Botany 22,
165–179.
WHITE, T.J., BRUNS, T.D., LEE,S.&TAY LO R , J.W. 1990. Amplific-
ation and Direct Sequencing of Fungal Ribosomal RNA Genes
for Phylogenetics. In: INNIS, M.A., GELFAND, D.H., SNINSKY,J.J.
&W
HITE, T.J., Eds., PCR Protocols, a Guide to Methods and
Applications. Academic Press, San Diego, pp. 315–322.
ZHOU,D.&HYDE, K.D. 2001. Host-specificity, host-exclusivity,
and host-recurrence in saprobic fungi. Mycological Research 105,
1449–1457.
