This article appeared in a journal published by Elsevier. The attached
copy is furnished to the author for internal non-commercial research
and education use, including for instruction at the authors institution
and sharing with colleagues.
Other uses, including reproduction and distribution, or selling or
licensing copies, or posting to personal, institutional or third party
websites are prohibited.
In most cases authors are permitted to post their version of the
article (e.g. in Word or Tex form) to their personal website or
institutional repository. Authors requiring further information
regarding Elsevier’s archiving and manuscript policies are
encouraged to visit:
http://www.elsevier.com/copyright
Author's personal copy
Molecular Phylogenetics and Evolution 59 (2011) 587–602
Contents lists available at ScienceDirect
Molecular Phylogenetics and Evolution
journal homepage: www.elsevier.com/locate/ympev
Complex patterns of speciation in cosmopolitan ‘‘rock posy’’ lichens – Discovering
and delimiting cryptic fungal species in the lichen-forming Rhizoplaca
melanophthalma species-complex (Lecanoraceae, Ascomycota)
Steven D. Leavitt a,⇑, Johnathon D. Fankhauser b, Dean H. Leavitt c, Lyndon D. Porter d, Leigh A. Johnson a,
Larry L. St. Clair a
a
Department of Biology and M.L. Bean Life Science Museum, Brigham Young University, Provo, UT 84602, USA
Department of Plant Biology, University of Minnesota, 1445 Gortner Ave., St. Paul, MN 55108, USA
c
Department of Biology, San Diego State University, 5500 Campanile Drive, San Diego, CA 92182-4614, USA
d
USDA_ARS Vegetable and Forage Crops Research Unit, Prosser, WA 99350, USA
b
a r t i c l e
i n f o
Article history:
Received 6 August 2010
Revised 16 December 2010
Accepted 17 March 2011
Available online 1 April 2011
Keywords:
Coalescent theory
Species delimitation
Lichen species concepts
Rhizoplaca
Secondary metabolites
Species delimitation
Species tree phylogeny estimation
Vagrant lichens
a b s t r a c t
A growing body of evidence indicates that in some cases morphology-based species circumscription of
lichenized fungi misrepresents the number of existing species. The cosmopolitan ‘‘rock posy’’ lichen (Rhizoplaca melanophthalma) species-complex includes a number of morphologically distinct species that are
both geographically and ecologically widespread, providing a model system to evaluate speciation in
lichen-forming ascomycetes. In this study, we assembled multiple lines of evidence from nuclear DNA
sequence data, morphology, and biochemistry for species delimitation in the R. melanophthalma species-complex. We identify a total of ten candidate species in this study, four of which were previously
recognized as distinct taxa and six previously unrecognized lineages found within what has been thus
far considered a single species. Candidate species are supported using inferences from multiple empirical
operational criteria. Multiple instances of sympatry support the view that these lineages merit recognition as distinct taxa. Generally, we found little corroboration between morphological and chemical characters, and previously unidentified lineages were morphologically polymorphic. However, secondary
metabolite data supported one cryptic saxicolous lineage, characterized by orsellinic-derived gyrophoric
and lecanoric acids, which we consider to be taxonomically significant. Our study of the R. melanophthalma species-complex indicates that the genus Rhizoplaca, as presently circumscribed, is more diverse in
western North American than originally perceived, and we present our analyses as a working example of
species delimitation in morphologically cryptic and recently diverged lichenized fungi.
Ó 2011 Elsevier Inc. All rights reserved.
1. Introduction
Lichens are obligate symbiotic systems consisting of a filamentous fungus, a photosynthetic partner (eukaryotic alga and/or cyanobacterium), and, at least in some cases, non-photosynthetic
bacteria (Cardinale et al., 2008; Grube et al., 2009; Hodkinson
and Lutzoni, 2009; Selbmann et al., 2010). The lichenized condition
has been extremely successful for many fungal lineages, with an
estimated 40% of all ascomycetes forming lichens (Lutzoni et al.,
2001). Traditionally, morphology and the expression of signature
secondary metabolites have been used to infer taxonomic boundaries for lichen-forming fungi (Culberson, 1972; Hale, 1990;
Huneck, 1999; Huneck and Yoshimura, 1996). However, these
characters are often widely variable, and their homology has
⇑ Corresponding author. Present address: Department of Botany, Field Museum of
Natural History, 1400 S. Lake Shore Drive, Chicago, IL 60605-2496, USA.
E-mail address: sleavitt@fieldmuseum.org (S.D. Leavitt).
1055-7903/$ - see front matter Ó 2011 Elsevier Inc. All rights reserved.
doi:10.1016/j.ympev.2011.03.020
proven difficult to assess between and within taxonomic groups
(Blanco et al., 2004b; Crespo et al., 2007; LaGreca and Lumbsch,
2001; Lumbsch and Schmitt, 2001; Ott et al., 2004). A growing
body of evidence suggests that in many cases lichen species diversity has been underestimated, especially within morphologically
similar species with cosmopolitan distributions (Buschbom and
Mueller, 2006; Crespo and Pérez-Ortega, 2009; Kroken and Taylor,
2001; O’Brien et al., 2009; Printzen, 2009; Wedin et al., 2009; Wirtz
et al., 2008).
Species represent fundamental units of analysis in various subdisciplines of biology, so accurate species diagnoses are critical,
and reassessing current species delimitation is particularly relevant in lichenized fungi. Finding and applying the appropriate
character sets and analytical tools is one of the greatest challenges
with empirical species delimitation in lichen-forming fungi
(Crespo and Pérez-Ortega, 2009; Wirtz et al., 2008). In spite of
the complicated issues associated with attempts to empirically
circumscribe species, contemporary species concepts share the
Author's personal copy
588
S.D. Leavitt et al. / Molecular Phylogenetics and Evolution 59 (2011) 587–602
common view that species are segments of separately evolving
metapopulation lineages (de Queiroz, 1998, 1999, 2007; Mayden,
1999). This concept allows researchers to delimit species using different empirical properties and facilitates the development of new
methods to test hypotheses of lineage separation (de Queiroz,
2007). A rapidly growing interest in species delimitations has
resulted in novel approaches to investigate species boundaries
(Carstens and Dewey, 2010; Knowles and Carstens, 2007; O’Brien
et al., 2009; O’Meara, 2010; Sites and Marshall, 2004; Vieites
et al., 2009; Weisrock et al., 2010; Yang and Rannala, 2010), and
more properties supporting putative lineages are associated with
a higher degree of corroboration (de Queiroz, 2007). Although different data sets and operational criteria may give conflicting or
ambiguous results due to multiple evolutionary processes occurring within and between populations, the use of several independent suites of characters, such as morphology, geographic range,
host preference, chemistry, and cross-validation using inferences
from multiple empirical operational criteria have been shown to
establish robust species boundaries (Dayrat, 2005; Duminil et al.,
2006; Hey et al., 2003; O’Brien et al., 2009; Roe and Sperling,
2007; Ruiz-Sanchez and Sosa, 2010).
As traditional characters used to delimit lichen species may
misrepresent mycobiont diversity, we feel it is important to
address lichen species boundaries using multiple independent
datasets and operational criteria to effectively identify and delimit
lichen species. We selected the Rhizoplaca melanophthalma speciescomplex (Ascomycota, Lecanorales, Lecanoraceae) as a model system to assess species diversity for this study because of its broad
ecological and geographical distribution, morphological, chemical
and genetic diversity, and importance as a sensitive indicator of
environmental health (Arup and Grube, 2000; Aslan et al., 2004;
Dillman, 1996; Leuckert et al., 1977; Ugur et al., 2004; Zhou
et al., 2006). This group has previously been identified as a wellsupported monophyletic lineage and includes the placodiod crustose taxon, Lecanora novomexicana H. Magn., the umblicate taxon
R. melanophthalma (DC.) Leuckert & Poelt, and at least 4 vagrant,
or obligatory unattached, species (Arup and Grube, 2000).
R. melanophthalma sensu lato (s. l.) has a worldwide distribution, and in North America it ranges from the northern boreal zone
to Mexico along the Rocky Mountain corridor. It is commonly
found in the Intermountain Western United States growing in large
populations on rocky substrates. Specimens are generally
umbilicate (fixed to the substrate by a single point of attachment),
but often appear squamulose or pulvinate (polyphyllous), and considerable chemical variation is found within the species (McCune,
1987; Ryan, 2001). However, taxonomic status of lichens with distinct morphologies and chemotypes within R. melanophthalma s. l.
and closely related taxa within this group (including the placodiod
L. novomexicana and vagrant Rhizoplaca) remains uncertain (Arup
and Grube, 2000; McCune, 1987). The vagrant taxa are endemic
to the high plains and mountains of the central and northern Rocky
Mountains in western North America and are particularly susceptible to habitat fragmentation, altered fire dynamics, and agricultural conversion (McCune and Rosentreter, 2007; Rosentreter,
1993).
Speciation in lichenized fungi is, in general, understudied, and
we present our analyses of the R. melanophthalma species-complex
to represent the larger focus of this study, which is robust species
delimitation in morphologically cryptic lichenized fungi. In this
study we followed the general lineage concept (GLC; de Queiroz,
1998, 1999) as our non-operational species definition. We analyzed molecular data within a phylogenetic framework to identify
candidate species by examining monophyletic groups recovered in
the topology, and assessed the putative lineages across individual
gene trees to identify lineages that exhibited genealogical exclusivity, an expected pattern for divergent lineages (Avise and Ball,
1990; Baum and Shaw, 1995; Hudson and Coyne, 2002). Candidate
species were also evaluated within a population-level framework
to assess gene flow and genetic differentiation (O’Brien et al.,
2009), and we used multi-locus sequence data to identify genetic
clusters without a priori assignment of individuals (Groeneveld
et al., 2009; Weisrock et al., 2010). The phylogenetic relationships
among candidate species circumscribed in this study were also inferred within a coalescence framework (Heled and Drummond,
2010) and support of speciation events was assessed using a Bayesian species delimitation test (Yang and Rannala, 2010). Finally, we
investigated patterns in morphological and chemical variation and
geographical and ecological distributions for each candidate
species.
2. Materials and methods
2.1. Taxon sampling
We analyzed sequence data from 170 individual posy rock
lichens. The focal group was represented by the four know species
from the R. melanophthalma species-complex: L. novomexicana (6
from 4 localities); R. melanophthalma, including R. melanophthalma
subsp. melanophthalma (DC.) Leuckert & Poelt (127 vouchers from
37 localities), R. melanophthalma subsp. cerebriformis Rosentreter &
B. D. Ryan (1), R. melanophthalma subsp. crispa Rosentreter & B. D.
Ryan (1); Rhizoplaca haydenii, including R. haydenii ssp. arbuscula
Rosentreter (2 from a single locality) and R. haydenii spp. haydenii
(Tuck.) W. A. Weber (6 from 4 localities); Rhizoplaca idahoensis
Rosentreter & McCune (4 from 2 localities); and two species not
formally described (see McCune and Rosentreter, 2007), Rhizoplaca
cylindrica (1) and Rhizoplaca subidahoensis (1). Fig. 1 depicts the
high degree of morphological variation within the sampled R. melanophthalma species-complex in western North America. The present study emphasized umbilicate saxicolous forms; therefore
sampling of the lobate taxon L. novomexicana and vagrant taxa
was relatively limited. Collections of R. melanophthalma s. l. were
initially made in 1997 at ten, 9 15 m plots along an altitudinal
gradient (2200–3400 m) at Thousand Lakes Mountain (TLM),
Wayne County Utah, USA (Porter, 1998), and three additional
9 15 m plots (2200 m, 2800 m, and 3300 m) were collected on
the neighboring Boulder Mountain Plateau (BM), Wayne and Garfield Counties, Utah, in 2008. Seven individual thalli were randomly chosen from collections made at each plot to assess
ecological trends in distributions and reproductive isolation between candidate species identified in this study (see Section 3.3).
We also sampled 39 additional specimens from the R. melanophthalma species-complex, collected from 24 populations throughout
the Intermountain West, USA. Available internal transcribed spacer
sequences obtained from GenBank, representing 20 individuals,
were included to assess relationships within a broader taxonomic
and phylogeographic context. Rhizoplaca subdiscrepans (Nyl.) R.
Sant. and Rhizoplaca chrysoleuca (Sm.) Zopf were selected as outgroups (Arup and Grube, 2000; Cansaran et al., 2006; Zhou et al.,
2006). Collection information for all included specimens is summarized in Supplementary Table S1, and new voucher material generated for this study is housed at the Brigham Young University
Herbarium of Nonvascular Cryptogams (BRY), Provo, Utah, USA.
2.2. Molecular data and sequence alignment
Total genomic DNA was isolated using either the E.Z.N.A. Plant
DNA Kit (Omega Bio-Tek, Norcross, GA), following manufacturer’s
instructions, or the Prepease DNA Isolation Kit (USB, Cleveland,
OH), following the plant leaf extraction protocol. We generated
new sequence data via polymerase chain reaction (PCR) for three
Author's personal copy
589
S.D. Leavitt et al. / Molecular Phylogenetics and Evolution 59 (2011) 587–602
Fig. 1. Variation in morphology and habit within the Rhizoplaca melanophthalma species-complex (Lecanoraceae) in western North America: (A) the lobate, placodioid taxon
Lecanora novomexicana; (B) Rhizoplaca melanophthalma sensu lato (s.l.), with distinct light colored, pruinose apothecia discs; (C) Rhizoplaca melanophthalma sensu lato (s.l.),
umblicate form with distinct lobes and dark apothecia; (D) R. melanophthalma s.l., umblicate form lacking lobes with pruinose apothecia (E) R. melanophthalma s.l., erratic
form completely lacking umbilicus growing free on soil from western Idaho, with apothecia. Images F–I vagrant taxa endemic to the high plains and mountains of the
northern Rocky Mountains (F) R. melanophthalma ssp. crispa; (G) R. idahoensis; (H) R. haydenii; (I) R. haydenii ssp. arbuscula. Images A–C, F–I were provided by S. Sharnoff.
Table 1
Primers used for PCR amplification and sequencing of the nuclear ribosomal IGS, ITS, and group I intron markers and nuclear markers b-tubulin and MCM7.
Marker
Primer name
Forward primer sequence
Annealing temperature (°C)
Reference
IGS
IGS12
NS1R
ITS1F
ITS4
Bt3-LM
Bt10-LM
Bt_rhizo_F
Bt_rhizo_R
Mcm7–709for
Mcm7–1348rev
LecMCM7f
LecMCM7r
50 -AGTCTGTGGATTAGTGGCCG-30
50 -GAGACAAGCATATGACTAC-30
50 -CTTGGTCATTTAGAGGAAGTAA-30
50 - TCCTCCGCTTATTGATATGC-30
50 -GAACGTCTACTTCAACGAG-30
50 -TCGGAAGCAGCCATCATGTTCTT-30
50 -GCAACAAGTATGTTCCTCGTGC-30
50 -GTAAGAGGTGCGAAGCCAACC-30
50 -ACIMGIGTITCVGAYGTHAARCC-30
50 -GAYTTDGCIACICCIGGRTCWCCC AT-30
50 -TACCANTGTGATCGATGYGG-30
50 -GTCTCCRCGTATTCGCATNCC-30
66–56 (touchdown)
Carbone and Kohn (1999)
Carbone and Kohn (1999)
Gardes and Bruns (1993)
White et al. (1990)
Myllys et al. (2001)
Myllys et al. (2001)
Leavitt et al. (2011)
Leavitt et al. (2011)
Schmitt et al. 2009
Schmitt et al. 2009
This study
This study
ITS/group I intron
b-tubulin
MCM7
nuclear ribosomal loci and two low-copy protein-coding loci (Table
1). The nuRNA gene tandem repeat exists in large copy numbers
facilitating amplification of ribosomal loci from older specimens
(Thousand Lake Mountain collections, 1997). Although low levels
of intragenomic variation in fungal rDNA repeats suggest
convergent evolution in which homogenization is very rapid and
effectively maintains highly similar repeat arrays through recombination (Ganley and Kobayashi, 2007), previous studies have
confirmed the utility of the sampled ribosomal loci for speciesand population-level studies in lichenized ascomycetes (Blanco
et al., 2004a,b; Brunauer et al., 2007; Buschbom and Mueller,
2006; Gutiérrez et al., 2007; Kroken and Taylor, 2001; Lindblom
and Ekman, 2006; O’Brien et al., 2009; Thell, 1999; Wedin et al.,
2009; Wirtz et al., 2008). A gene duplication of b-tubulin has
55–60
55–60
66–56 (touchdown)
56
66–56 (touchdown)
occurred within Ascomycota, however the paralogs are easily distinguishable within the analyzed group, and the marker has been
successfully employed to investigate species-level relationships
in other lichenized ascomycetes (Buschbom and Mueller, 2006;
O’Brien et al., 2009; Wedin et al., 2009).
PCR cycling parameters used for amplifying the ITS, group I intron, and b-tubulin loci followed the methods of Blanco et al.
(2004b) and Wedin et al. (2009); cycling parameters for amplifying
the IGS followed a 66–56° touchdown reaction (Lindblom and
Ekman, 2006); and PCR cycling parameters amplifying the MCM7
fragment followed Schmitt et al. (2009). PCR products were
quantified on 1% agarose gel and stained with ethidium bromide.
In cases where no PCR product was visualized for the b-tubulin
and MCM7 loci, internally nested PCR reactions were performed
Author's personal copy
590
S.D. Leavitt et al. / Molecular Phylogenetics and Evolution 59 (2011) 587–602
using 0.3 ll of PCR product from the original reaction and newly
developed internal primers ‘BT-RhizoF’ and ‘BT-RhizoR’ for the tubulin fragment, and ‘LecMCM7f’ and ‘LecMCM7r’ for the MCM7
fragment. Nested PCR reactions followed the touchdown PCR cycling parameters described above used to amplify the IGS fragment. PCR fragments were cleaned using the PrepEase PCR
Purification Kit (USB, Cleveland, OH), and complementary strands
were sequenced using the same primers used for amplification.
Sequencing reactions were performed using the Big Dye3 (Applied
Biosystems, Foster City, CA), and products were run on an AB
3730xl automated sequencer at the DNA Sequencing Center, Brigham Young University Provo, Utah, USA.
Sequences were assembled and edited using Sequencher version 3.1.1 (Gene Codes Corporation, Ann Arbor, MI), and sequence
identity was confirmed with ‘megaBLAST’ search in GenBank
(Wheeler et al., 2006). Sequences were aligned in Muscle version
3.6 (Edgar, 2004), using default settings. Because recombination
within nuclear genes can lead to errors in estimated relationships,
we tested for intragenic recombination events in all sampled loci
using GENECONV (Padidam et al., 1999), Bootscan/Rescan (Martin
et al., 2005a), MaxChi (Smith, 1992), Chimaera (Posada and Crandall, 2001a), SiScan (Gibbs et al., 2000), LARD (Holmes et al., 1999),
Topal/DSS (McGuire and Wright, 2000), and 3Seq (Boni et al., 2007)
methods implemented in Recombination Detection Program v3.44
(Heath et al., 2006; Martin et al., 2005b). Additionally, we searched
alignments for evidence of recombination breakpoints using GARD
(Kosakovsky Pond et al., 2006), implemented in www.datamonkey.org web server.
2.3. Nucleotide polymorphism analyses and population differentiation
We used DnaSP 5.10 (Librado and Rozas, 2009) to calculate
basic nucleotide polymorphism statistics, including numbers of
haplotypes (H), total number of polymorphic sites (Npoly), average
pair-wise diversity per site, (p; Nei, 1987) for each candidate species (see Section 3.3). In addition, genetic differentiation between
candidate species was assessed by calculating FST values using
DnaSP and counting the number of fixed nucleotides for all pairwise comparisons (O’Brien et al., 2009). F-statistic calculations
were estimated from specimens with complete ITS, IGS, ß-tubulin,
and MCM7 dataset (the group I intron was missing in all specimens
assigned to a single candidate species, and this marker was therefore excluded from FST calculations). Aligned sequences were
scanned for fixed characters between each candidate species and
the remaining data matrix in DnaSP, and the total number of fixed
nucleotide positions was tabulated for each candidate species.
2.4. Phylogenetic analyses
Heterogeneity in phylogenetic signal among the sampled markers was assessed before combining the datasets. We performed
maximum likelihood (ML) analyses of the concatenated ribosomal
dataset (ITS, IGS, and group I intron), b-tubulin, and MCM7 markers
separately in RAxML version 7.0.4 (Stamatakis, 2006; Stamatakis
et al., 2008), using the ‘rapid bootstrapping’ option as implemented
on the CIPRES Web Portal. We used the GTRGAMMA model, which
includes a parameter (C) for rate heterogeneity among sites, and
chose not to include a parameter for estimating the proportion of
invariable sites because C mathematically account for this source
of rate heterogeneity by using 25 rate categories (Stamatakis,
2006). Support values for ribosomal, b-tubulin, and MCM7 phylogenies were examined for well-supported (P70%) conflicts between
data sets (Lutzoni et al., 2004).
Phylogenetic relationships were estimated from the combined
data set using mixed-model Bayesian inference (BI) in Mr.Bayes
version 3.1.2 (Huelsenbeck and Ronquist, 2001), using the default
priors, namely a flat Dirichlet prior for the relative nucleotide
frequencies and rate parameters, a discrete uniform prior for topologies, and an exponential distribution (mean 1.0) for the gammashape parameter and branch lengths. We used MrModeltest
version 2.3 (Nylander et al., 2004) to identify the best-fitting model
of evolution for each marker using the Akaike Information Criterion (AIC; Posada and Crandall, 2001b), and we treated each marker as a separate partition. Four independent replicate searches
were executed with eight chains; each run started with randomly
generated trees and consisted of sampling every 1000 generations
for 20,000,000 generations. To evaluate stationarity and convergence between runs, log-likelihood scores were plotted using TRACER version 1.5 (Rambaut and Drummond, 2007), ESS statistics,
and the average standard deviation in split frequencies were assessed (Hall, 2007). Trees generated prior to stationarity were discarded as burn-in (Huelsenbeck et al., 2001). The results were
summarized with a majority-rule consensus tree from the remaining trees from the four independent runs. Bayesian posterior probabilities (PP) were assessed at all nodes, and clades with PP P 0.95
were considered strongly supported (Huelsenbeck and Rannala,
2004).
We conducted a ML analysis using RAxML 7.0.4, allowing a
separate GTR model with unique parameter values for each locus
(Stamatakis, 2006; Stamatakis et al., 2008). We used the GTRGAMMA model and did not include a parameter for the proportion of
invariable sites (Stamatakis, 2006). A search combining 200 separate maximum likelihood searches (to find the optimal tree) and
1000 ‘‘fastbootstrap’’ replicates to evaluate nodal support was conducted on the complete dataset.
Exploratory phylogenetic reconstructions including GenBank
specimens represented solely by ITS sequences in the concatenated
dataset resulted in reduced nodal support across the topology and
important ambiguous relationships. Therefore we chose not to include GenBank accessions in the multilocus data matrix in order to
minimize the effect of missing data (Baurain et al., 2007). To assess
relationships within a broader geographic context we reconstructed the ITS gene tree using both BI and ML inference from
all available ingroup ITS sequences, including 20 sequences retrieved from the GenBank database, with Rhizoplaca chrysolueca selected as the outgroup (Arup and Grube, 2000; Zhou et al., 2006). BI
and ML reconstructions were performed for the complete ITS dataset as described above.
Establishing a preliminary set of species boundaries using concatenation provides a reasonable starting point for identifying
cryptic species and species tree inference (Le Gac et al., 2007;
Leache, 2009; Šlapeta et al., 2006). We used the topology from
the concatenated dataset to guide the identification of candidate
species for this study. We chose to circumscribe a total of 10 candidate species to represent four currently accepted taxa and six
phylogenetic lineages identified within the topology representing
R. melanophthalma s. l (see Results, Section 3.3). Following recommendations of Sites and Marshall (2004) and de Queiroz (2007), we
implemented multiple analytical approaches to assess species
boundaries for independent corroboration of the candidate species
identified in the current study.
2.5. Haplotype network reconstructions and genealogical concordance
Relationships of candidate species were evaluated between
individual haplotype networks to identify lineages that exhibited
genealogical exclusivity across multiple loci (Avise and Ball,
1990; Hudson and Coyne, 2002). The presence of the same clades
in the majority of single-locus genealogies is taken as evidence that
the clades represent reproductively isolated lineages (Dettman
et al., 2003; Pringle et al., 2005). We used statistical parsimony
to assess the genealogical relationship of every individual and
Author's personal copy
S.D. Leavitt et al. / Molecular Phylogenetics and Evolution 59 (2011) 587–602
compare relationships of candidate species between genes. Networks were constructed under a 95% parsimony probability criterion (Templeton et al., 1992) from concatenated ribosomal
sequences (ITS, IGS, intron), the b-tubulin, and the MCM7 fragments using the program TCS v1.21 (Clement et al., 2000). Gaps
were treated as missing data for the ribosomal network reconstruction to include voucher specimens missing one of the three
ribosomal loci. All protein-coding sequences were trimmed to the
length of the fragment resulting from nested PCR reactions and a
single sequence missing approximately half the fragment was removed from the -tubulin network analysis. All network uncertainties (i.e. closed loops) were treated following Templeton and Sing
(1993).
2.6. Bayesian population structure analysis
Individual-based approaches provide an alternative for identifying population structure and barriers to gene flow (Saisho and
Purugganan, 2007), as analyses based on predefined groups may
obscure patterns of differentiation (Latch et al., 2006; Rowe and
Beebee, 2007). We used a Bayesian population assignment test
implemented in STRUCTURE version 2.32 (Falush et al., 2003;
Pritchard et al., 2000) to infer population structure based on a combined genotypic matrix from SNPs data from all loci (ITS, IGS, group
I intron, b-tubulin, and MCM7), without using known geographic
location or putative species classification of the individual as priors. An admixture model was used with correlated allele frequencies. We implemented 15 replicate runs for each number of
assumed populations (K), with a range of K from 1 to12. Based
on preliminary runs, all analyses used 30,000 MCMC generations
to estimate the posterior distribution following a burn-in period
of 15,000 generations. In some cases, independent runs for K values
3–12 appeared to converge on different parameter space, and longer burn-in or MCMC runs did not improve convergence. Therefore,
we calculated the median log (ln) likelihood of each K value from
the four best-scoring runs. We calculated the modal value (DK)
based on the second order rate of change of the likelihood function
between successive K values (Evanno et al., 2005). In some scenarios the DK method may fail detect the true number of clusters (K),
and we examined subgroups created by the best individual assignments produced by STRUCTURE (Evanno et al., 2005; Groeneveld
et al., 2009; Saisho and Purugganan, 2007; Weisrock et al., 2010).
2.7. Species tree inference and speciation probabilities
We used the hierarchical Bayesian model implemented in
BEAST v. 1.6.1 (Heled and Drummond, 2010) to estimate the species trees for the ten candidates species. ⁄BEAST estimates the species tree directly from the sequence data, and incorporates the
coalescent process, uncertainty associated with gene trees, and
nucleotide substitution model parameters (Heled and Drummond,
2010). We selected 2–8 individuals from each candidate species,
representing the sampled genetic diversity. The group I intron
was missing from all individuals recovered in clade IVd (Fig. 2)
and was therefore excluded from the analysis. We specified the
best-fitting models of molecular evolution for each locus using
the same models used in the MrBayes v. 3.1.2 analysis under an
uncorrelated relaxed lognormal clock (Drummond et al., 2006).
We selected the lineage birth/death model and constant population size for species tree priors. Two independent MCMC analyses
were run for a total of 50 million generations (sampling every 1000
steps and excluding the 10 million generations of each run as burnin). We assessed convergence by examining the likelihood plots
through time using Tracer v. 1.5 (Rambaut and Drummond,
2007), and the effective sample sizes (ESS) of parameters of
⁄
591
interest were all above 200. Posterior probabilities of nodes were
computed from sampled trees after burn-in.
We estimated the marginal posterior probability of speciation
using the program BPP (Rannala and Yang, 2003; Yang and
Rannala, 2010). This method accommodates the species phylogeny
as well as lineage sorting due to ancestral polymorphism. We used
a gamma prior h G(2, 1000) on the population size parameters
with algorithm 0. The age of the root in the species tree was assigned the gamma prior s0 G(2, 1000), while the other divergence time parameters were assigned the Dirichlet prior (Yang
and Rannala, 2010). Because the prior distribution of h and s0 can
result in strong support for models containing more species
(Leaché and Fujita, 2010), we also explored a more conservative
combination of priors that should favor fewer species by assuming
large ancestral population sizes h G(1, 10) and relatively shallow
divergences among species s0 G(2, 2000). The species tree estimated in the ⁄BEAST analysis, representing the ten candidate species, was used as the fully resolved guide tree. Running the rjMCMC
analysis for 1000,000 generations with a burn-in of 100,000 produced consistent results across independent analyses initiated
with different starting seeds and species trees. Each analysis was
run at least twice to confirm consistency between runs.
2.8. Morphological and biochemical comparisons
Considering current studies (Arup and Grube, 2000; Cansaran
et al., 2006; Ryan, 2001; Zheng et al., 2007; Zhou et al., 2006), a total of 14 morphological characters were quantified from all specimens to identify potentially diagnostic characters for candidate
species identified in this study, including: point of attachment (distinctly umbilicate/squamulose), thallus form (polyphyllous/monophyllous), lobe morphology (distinct/intermediate/indistinct),
appearance of upper surface (dull/shiny), upper surface texture
(smooth/cracked), upper surface color (light to moderately greenish yellow/olive), lower surface texture (smooth/rough), lower surface edges (black near edges/not blackened edges), lower surface
color (tan/brown), apothecia morphology (sessile/basally constricted), apothecia pruinosity (heavily pruinose/moderately pruinose/not pruinose), thallus margin (entire/crenate), spore shapes
(ellipsoid/subglobose), spore size (continuous character).
Lichen compounds were extracted from 0.02 g liquid nitrogenground specimens overnight in acetone at 4 °C. The supernatant
was removed, dried, reconstituted in methanol, and analyzed using
high-performance liquid chromatography (HPLC). Retention index
values (RI) were calculated from benzoic acid and solorinic acid
controls (Feige et al., 1993; Lumbsch, 2002). For HPLC, we used
an Agilent Technologies 1200 series integrated system with a
Zorbax Eclipse XDB8-CB column (4.6 150 mm, 5 lm) regulated
at 30 °C, spectrometric detectors operating at 210, 254, 280,
310 nm, and a flow rate of 0.7 ml/min. Following established protocols (Feige et al., 1993; Lumbsch, 2002), two mobile phases, A
and B, were used: 1% aqueous orthophosphoric acid (A) and methanol (B). The run started with 30% B for 1 min and was raised to
70% B within 15 min of the start time, then to 100% B during an
additional 15 min, followed by isocratic elution in 100% B for the
final 20 min. Mobile phase B was decreased to 30% within 1 min
and the column was flushed with 30% B for 15 min following each
run. UV spectra of each peak were recorded and computermatched against a library of ultraviolet spectra from authentic
metabolites derived under identical conditions using Agilent
Chemstation software. The correlation of UV spectra with the standards in the library was greater than 99.9% for each substance
identified. When multiple library entries matched with this level
of identity, calculated R/I values were used to discriminate between compounds.
Author's personal copy
592
S.D. Leavitt et al. / Molecular Phylogenetics and Evolution 59 (2011) 587–602
Fig. 2. Relationships among sampled specimens collected from the Rhizoplaca melanophthalma group inferred from a maximum likelihood analysis of ribosomal and nuclear
DNA sequence data (2600 bp, ITS, IGS, group I intron, b-tubulin, and MCM7). Values at each node indicate non-parametric-bootstrap support/posterior probability. Only
support indices PBS 50/PP 0.50 are indicated. Clade numbers plotted to the right of the tree indicate candidate species. GenBank accessions represented solely by ITS
sequences were not included.
3. Results
For this study, 627 new sequences were generated, including
152 ITS, 129 IGS, 75 group 1 intron, 137 b-tubulin, and 134
MCM7 sequences. The combined data matrix consisted of 2639
aligned nucleotide position characters (Table 2; TreeBase Accession 11094). Missing data were generally limited to the outgroup
taxa R. chrysoleuca and R. subdiscrepans. However, we were unable
to generate group I intron sequences from all specimens recovered
in clade IVd and single individuals from R. haydenii subsp. arbuscula
(092f), R. idahoensis (093f) and clade II (693f) (defined below). All
representative haplotypes of the five gene fragments have been
deposited in GenBank under Accession Nos. HM576889–
HM577515 (Supplementary Table S2). No evidence for recombination within any of the nuclear markers was detected.
3.1. Polymorphism statistics and estimates of genetic differentiation
Polymorphism statistics are reported in Table 3. The greatest
nucleotide diversity for candidate species was generally recovered
Author's personal copy
593
S.D. Leavitt et al. / Molecular Phylogenetics and Evolution 59 (2011) 587–602
IV). Group I intron sequences were missing for all individuals assigned to clade IVd.
for ribosomal loci. High levels of genetic differentiation between all
pairs of candidate species were calculated from the combined data
set, as measured by FST (Table 4). Fixed differences between candidate species circumscribed in this study were identified from ribosomal markers for all pair-wise comparisons, and fixed differences
were identified in at least one of the protein-coding fragments for
40 of 45 pairwise comparisons (Table 4). The ribosomal data matrix showed the greatest number of fixed character differences between each candidate species compared to all remaining lineages;
while the protein-coding matrixes generally did not reveal fixed
character differences (Table 4). However, the b-tubulin fragment
revealed nine fixed nucleotide positions in clade I and 1 fixed locus
in clade IVb, and the MCM7 fragment revealed two fixed nucleotide
positions in clade I and five fixed characters in R. idahoensis (clade
3.2. Phylogenetic reconstructions
We detected limited discordance between the ribosomal,
b-tubulin, and MCM7 topologies restricted to clades with relatively
shallow evolutionary histories. Conflicting terminals are shown in
individual gene trees (Supplementary Figs. 1–3). Given the overall
congruence, the ribosomal, b-tubulin, and MCM7 gene regions
were combined to maximize the total number of characters for
phylogenetic analyses and branch length estimation (Rokas et al.,
2003; Wiens, 1998). Major relationships and branch support values
from a restricted dataset excluding the individuals showing incongruent positions when comparing the single gene phylogenies
were highly similar to the phylogeny resulting from the complete
dataset. The partitioned Bayesian analyses, summed from four
independent runs, yielded a consensus tree with a negative ln harmonic mean of 11,092.49. All parameters converged within the
first 25% of sampled generations, leaving a posterior distribution
estimated from 15,000 trees per run (60,000 total post-burn-in
sampled trees). The partitioned ML analysis yielded a single best
scoring tree lnL = 10,755.758. As the recovered trees were similar
across methods and the topologies did not show any strongly supported conflict; we present the results of the ML analysis with ML
bootstrap (BS) and Bayesian posterior probability (PP) values in
Fig. 2. The R. melanophthalma group is strongly supported as monophyletic and several other well-supported groups can be identified
in the tree.
Table 2
Genetic variability of sampled markers used in this study, including alignment length
(number of basepairs); variable and parsimony-informative (PI) sites for each
sampled locus; and locus-specific model of evolution identified using the Akaike
information criterion in MrModeltest. Numbers in parentheses indicate the number of
variable and parsimony-informative sites for the Rhizoplaca melanophthalma speciescomplex only.
Locus
Length
ITS
IGS
Group I intron
b-tubulin
MCM7
Total
# Variable sites
# PI sites
561
374
269
819
616
163 (91)
138 (84)
98 (44)
165 (90)
158 (123)
127 (57)
103 (54)
84(30)
132(55)
123 (42)
2639
722 (432)
569 (238)
Model selected
GTR + G
GTR + I
SYM + G
HKY + I + G
GTR + G
–
Table 3
Polymorphism statistics for candidate species within the R. melanophthalma species-complex. N, number of individuals sampled, Npoly, number of polymorphic sites; h, number of
unique haplotypes; p, estimate of 4 Nl per base pair using the average pairwise differences.
ITS
Clade
Clade
Clade
Clade
Clade
Clade
Clade
Clade
Clade
Clade
I (L. novomexicana)
II
III
IV (R. haydenii)
IV (R. h. spp. arbuscula)
IV (R. idahoensis)
IVa
IVb
IVc
IVd
Total
IGS
Intron
MCM7
b-tubulin
N/Npoly/h
p
N/Npoly/h
P
N/Npoly/h
P
N/Npoly/h
p
N/Npoly/h
p
3/0/1
24/35/17
13/5/5
5/6/4
2/1/2
3/3/2
3/3/3
14/9/7
5/1/2
55/11/10
0
0.00930
0.00188
0.00475
0.00182
0.00367
0.00427
0.00235
0.00088
0.00162
4/0/1
21/37/18
13/1/2
4/4/4
2/1/2
3/1/2
3/2/3
13/3/4
5/3/3
55/19/18
0
0.01776
0.0014
0.00318
0.00272
0.00272
0.00363
0.00265
0.00381
0.01191
3/0/1
23/19/17
13/0/1
5/4/3
1/0/1
2/0/1
3/0/1
14/3/4
5/0/1
0/na/na
0
0.1089
0
0.00723
na
0
0
0.00327
0
na
4/2/3
24/34/17
13/3/2
5/2/2
1/0/1
3/4/2
3/0/1
13/9/9
5/5/3
55/32/8
0.00146
0.01430
0.00067
0.00117
na
0.0039
0
0.00285
0.00439
0.00266
2/11/2
23/10/8
13/4/4
5/7/2
2/0/1
37316
3/0/1
13/19/6
5/2/2
55/5/6
0.02041
0.00278
0.00157
0.00779
0
0.00124
0
0.01308
0.00148
0.00040
127/91/52
0.02221
122/84/54
0.02494
69/43/27
0.03521
127/71/40
0.01309
126/112/33
0.01486
Table 4
Fixed differences and fixation indices (FST) for all pairwise comparisons of candidate species identified within R. melanophthalma species-complex. Numbers across the top row
correspond to candidate species numbers in the first column. Numbers of fixed differences (ribosomal/b-tubulin/MCM7 characters) are represented for all comparisons below the
diagonal and FST values are represented above the diagonal. The last column indicates total number of fixed nucleotides identified between each candidate species and the
remaining data matrix. Numbers within parentheses represent fixed ribosomal characters/fixed protein-coding characters. Accessions representing R. haydenii subspecies
arbuscula were not included in FST calculations because of the small sample sizes and pairwise comparisons are not represented.
Candidate species
1
2
3
4
5
6
7
8
9
10
Fixed characters
1. Clade I (L. novomexicana)
2. Clade II
3. Clade III
4. Clade IV (R. haydenii)
5. Clade IV (R. h. spp. arbuscula)
6. Clade IV (R. idahoensis)
7. Clade IVa
8. Clade IVb
9. Clade IVc
10. Clade IVd
–
49 (31/13/5)
77(55/18/4)
77(51/20/6)
82(54/19/9)
71(53/8/10)
65(38/20/7)
76(54/19/3)
76(51/18/7)
61(36/18/7)
0.77102
–
32(28/0/4)
32(26/0/6)
36(28/1/7)
33(28/0/5)
36(29/0/7)
31(29/2/0)
35(28/0/7)
22(16/0/6)
0.89534
0.75792
–
55(36/11/8)
56(39/9/8)
55(38/8/9)
54(36/10/8)
48(39/7/2)
55(39/8/8)
45(29/8/8)
0.86359
0.732
0.90524
–
7(2/4/1)
12(1/0/11)
27(21/5/1)
13(4/8/1)
18(14/3/1)
10(6/3/1)
na
na
na
na
–
15(2/0/13)
27(24/3/0)
11(5/6/0)
6(6/0/0)
8(8/0/0)
0.85763
0.69564
0.89291
0.58915
na
–
38(23/2/13)
15(5/5/5)
18(5/0/13)
14(6/0/12)
0.88863
0.76148
0.9382
0.82339
na
0.84298
–
30(23/7/0)
24(24/0/0)
14(14/0/0)
0.85574
0.72461
0.88716
0.67851
na
0.6808
0.82136
–
9(4/5/0)
13(8/5/0)
0.88172
0.75139
0.9273
0.66667
na
0.71146
0.83333
0.67031
–
7(7/0/0)
0.88085
0.7426
0.92874
0.71894
na
0.75427
0.80228
0.72953
0.66841
–
32(21/11)
3(3/0)
15(15/0)
1(0/1)
0 (0/0)
7(1/6)
7(7/0)
3(2/1)
1(1/0)
1(1/0)
Author's personal copy
594
S.D. Leavitt et al. / Molecular Phylogenetics and Evolution 59 (2011) 587–602
The ITS topology (Fig. 3) recovered most lineages identified in
the combined analyses. GenBank accessions representing individuals collected in Austria (AF159935), China (AY509791, EF095286,
and EF095297), and the United States (AF159929-Arizona and
AF159935-Arizona) were recovered in a well-supported clade
(91/1.0) corresponding to clade II identified in the combined analyses. Six accessions collected in China (EF095278, EF095280,
EF095283, EF095285, EF095287, and EF095290) were recovered
within a well-supported clade (81/0.98) corresponding to clade
IVb from the combined analyses, and two accession representing
R. cerebriformis (AF159942, Idaho, USA) and R. subidahoensis
(AF159944, Idaho, USA) were recovered within a well-supported
clade (90/1.0) corresponding to clade IVa from the combined analyses. A single accession representing R. cylindrical (AF159941, Idaho, USA) was recovered in a clade with high ML bootstrap support
(82) and weak PP support (0.79) corresponding to clade IVd in the
combined analyses. Two vagrant accessions representing R. idahoensis (AF159943-Idaho, USA) and R. haydenii subsp. haydenii
(AF159937-Idaho, USA) were recovered in a well-supported clade
(85/1.0) containing individuals all assigned to clades IVb, IVc, R.
haydenii, R. haydenii ssp. arbuscula, and R. idahoensis in the combined analyses. L. novomexicana was recovered as polyphyletic in
two well-supported lineages; one containing specimens collected
in northeastern Utah, and the second (clade V, Fig. 3) in two
Fig. 3. The maximum likelihood ITS topology obtained from all sampled specimens and available GenBank accessions collected from the Rhizoplaca melanophthalma speciescomplex. Values at each node indicate non-parametric-bootstrap support/posterior probability. Only support indices PBS 50/PP 0.50 are indicated. Clade numbers plotted to
the right of the tree indicate lineages corresponding to candidate species shown in Fig. 2.
Author's personal copy
S.D. Leavitt et al. / Molecular Phylogenetics and Evolution 59 (2011) 587–602
GenBank accessions, one from Arizona (AF159923) and the other
from New Mexico (AF159923). However, the relationship between
the L. novomexicana lineages lacked strong statistical support.
3.3. Candidate species
We circumscribed 10 candidate species based on the results
from our phylogenetic reconstructions and current taxonomic
boundaries for additional empirical testing of species boundaries.
However, individuals represented by GenBank accessions were
not included because they were represented by a single marker.
L. novomexicana (clade I, Fig. 2) was recovered as a well-supported
lineage (BS = 100/PP = 1.0), sister to the remaining R. melanophthalma taxa with weak nodal support. Clade II was recovered with
high nodal support (95/1.0), and corresponds to a genetically and
morphologically diverse assemblage of umbilicate saxicolous specimens collected throughout the intermountain western United
States, all containing usnic and psoromic acids. However, the relationship of clade II with other well-supported sister lineages lacks
strong nodal support (43/0.89). Clade III was also recovered with
strong support (100/1.0), and is represented by umbilicate saxicolous individuals with little morphological or genetic variation collected from two plots (BM-3 and TLM-9) on the Aquarius Plateau in
south central Utah, U.S.A. Clade III was recovered with strong nodal
support (94/0.98) as sister to a fourth well-supported clade (99/
1.0) containing a chemically diverse assemblage of umbilicate
595
and vagrant specimens (clade IV). Seven additional candidate species were identified within clade IV to accommodate currently described vagrant taxa and an exhaustive subdivision of the
remaining specimens.
All sampled vagrant taxa were recovered within a single
monophyletic clade with weak nodal support (BS and PP < 50/
0.50). R. idahoensis, R. haydenii subsp. haydenii, and R. haydenii
subsp. arbuscula were treated as independent lineages based on
current taxonomic circumspection. Both R. idahoensis and R. haydenii subsp. arbuscula were recovered as well-supported monophyletic lineages (94/1.0 and 81/1.0, respectively), while R.
haydenii subsp. haydenii was found in two well-supported clades.
A single saxicolous specimen with unique lobe morphology (715f)
was recovered within the R. haydenii subsp. haydenii clade. In
addition to the currently described vagrant taxa, four candidate
species are circumscribed to accommodate exhaustive subdivision
within the larger clade. Clade IVa (Fig. 2) was recovered with
strong nodal support (100/1.0) and contains three morphologically and geographically diverse individuals. All specimens containing lecanoric or orsellinic acids were recovered within clade
IVb with moderate to strong nodal support (BS = 83; PP = 0.93).
Clade IVc (Fig. 2) was also recovered with strong support (82/
1.0), and included five individuals; and clade IVd included the
remaining 55 individuals. Although this lineage was recovered
as monophyletic, it lacked strong support in the combined phylogenetic reconstructions.
Fig. 4. Geographical distributions of candidate Rhizoplaca species in the Intermountain western USA. Colors refer to different lineages, indicated in key. Insert shows
distributions of putative lineages along two altitudinal gradients in southern Utah, USA. A total of 7 individual were included from each plot and the proportion of candidate
species recovered at each plot is represented.
Author's personal copy
596
S.D. Leavitt et al. / Molecular Phylogenetics and Evolution 59 (2011) 587–602
Fig. 5. (A) Unrooted statistical parsimony haplotype networks at 95% probability of the ribosomal, MCM7, and b-tubulin loci representing relationship within the R.
melanophthalma species-complex. Each candidate species is designated by a different color. Size of circles is proportional to the number of individuals of a given haplotype,
and black dots represent inferred haplotypes not sampled. (B) Correspondence between candidate species identified from the combined maximum likelihood analysis and the
population clusters identified using STRUCTURE. Numbers at nodes represent maximum likelihood bootstrap values and posterior probabilities, and relationships within
candidate species are collapsed for ease of presentation (see Fig. 2 for detailed relationships). Candidate species are mapped to corresponding clusters in the STRUCTURE plot.
Each population cluster is represented by a different color, and vertical bars within each cluster represent individuals and the proportion of a bar assigned to a single color
represents the posterior probability that an individual is assigned to that cluster. The colors in the topology and STRUCTURE plot correspond to candidate species colors
shown in Fig. 5A. phylogenetic hypothesis of relationships in the Rhizoplaca melanophthalma species-complex in western North America.
Author's personal copy
597
S.D. Leavitt et al. / Molecular Phylogenetics and Evolution 59 (2011) 587–602
Geographic distributions of candidate species and the distribution of these species along altitudinal transects on Thousand Lakes
Mountain and Boulder Mountain, Utah are summarized in Fig. 4.
3.4. Haplotype networks
We recovered a total of five independent haplotype networks
for the combined ribosomal data set, and two networks for both
the -tubulin and MCM7 datasets (Fig. 5A). The ribosomal network
haplotypes separated by up to 15 mutational steps had greater
than 95% probability of being parsimoniously connected. In the
b-tubulin and MCM7 distinct networks were connected by up to
11 or 10 mutational steps, respectively. For all markers, clade I
(L. novomexicana) formed an independent network. In addition,
clades II, III, and IVa formed independent networks constructed
from the ribosomal dataset, while clades IVc, IVb, IVd, R. haydenii
3.5. Bayesian population structure
The median ML values of the Bayesian clustering analysis using
STRUCTURE with estimates of K = 1–12 are shown in Fig. 6A. These
analyses reveal a general pattern of a plateau with a decrease in
median maximum likelihood values above a K = 6 level. In contrast,
the DK method indicates that a K = 2 model best fits the data
(Fig. 6B; DK = 137.170 for K = 2; DK = < 25 for all other K values).
However, the plateau in likelihood values around K = 6 suggest a
-2000
B 140
-3000
120
-4000
∆K value
Likelihood (lnL)
A
subsp. arbuscula (clade IV), R. haydenii subsp. haydenii (clade IV),
and R. idahoensis (clade IV), were found on a single network. In
both the -tubulin and MCM7 datasets clades II, III, IVa, IVb, IVc,
IVd, R. haydenii subsp. arbuscula (clade IV), R. haydenii subsp. haydenii (clade IV), and R. idahoensis (clade IV) were found on a single
network.
-5000
-6000
-7000
100
80
60
-8000
40
-9000
20
0
-10000
1 2 3 4 5 6 7 8 9 10 11 12
2
3
4
5
6
7
8
9
10 11 12
Inferred clusters (K)
Inferred clusters (K)
Fig. 6. Plots of calculations for K values 1–12 in STRUCTURE analysis of the combined dataset. (A) The mean log probability of the data for K = 1–12, calculated from the four
best scoring runs for each K value. (B) DK values for K = 2–12.
R. chrysoleuca
clade I
1.0,1.0,1.0
clade III
clade II
0.99,1.0,1.0
clade IVb
0.57,1.0,1.0
clade IVa
1.0,1.0,1.0
clade IVc
0.63,1.0,0.98
0.54,1.0,0.97
clade IVd
0.75,1.0,0.98
R. h. subsp.
arbuscula
0.95,1.0,0.96
R. h. subsp.
haydenii
0.54,0.95,0.62
R. idahoensis
0.02
Fig. 7. The coalescent-based species tree and Bayesian species delimitation results for the Rhizoplaca melanophthalma species-complex. Posterior probabilities at nodes
indicate support from the BEAST analysis, speciation probabilities under prior means = 0.001, and prior means = 0.1.
Author's personal copy
598
S.D. Leavitt et al. / Molecular Phylogenetics and Evolution 59 (2011) 587–602
Table 5
Chemotypic variation by candidate species in the R. melanophthalma species-complex based on HPLC analysis. Superscript number following acid nominal indicate acid
occurrence: 1major or minor; 2major or not present; 3minor or not present; 4minor or trace; and 5trace or not present.
Acid
Clade I
Clade II
Clade III
R. haydenii
(clade IV)
R. h. ssp. arbuscula
(clade IV)
R. idahonesis
(clade IV)
Clade IVa
Clade IVb
Clade IVc
Clade IVd
Usnic1
Psoromic2
Lecanoric2
Orsellinic3
Gyrophoric5
Constipatic3
Dehydroconstipatic3
Dehydroprotoconstipatic3
Subpsoromic acid3
20 -O-demethylsubpsoromic4
20 -O-demethylpsoromic3
1
1
0
0
0
0
0.25
0.25
0.25
0.75
0.75
1
0.91
0
0
0
0.91
0.91
0.7
0.43
0.52
0.39
1
1
0
0
0
0.64
0.36
0.36
1
1
0.82
1
0
0
0
0
0
0
0
0
0
0
1
0
0
0
0
0
0
0
0
0
0
1
0
0
0
0
0
0
0
0
0
0
1
0.66
0
0
0
1
1
0.33
0
1
1
1
1
0.57
0.64
0.43
0.93
0.93
0.86
0.57
0.29
0.5
1
0.40
0
0
0
1
1
1
1
1
0.5
1
0.95
0
0
0
0.91
0.95
0.55
0.78
0.87
0.73
higher number of population clusters (Fig. 6A). A plot of individual
membership coefficients for K = 6 reveals a high number of population clusters with average individual membership coefficients
(i.e. posterior probabilities) greater than 0.9 (Fig. 5B). Population
clusters inferred for K > 6 did not yield additional clusters with
high membership coefficients. Therefore, we place our focus on
K = 6 as an uppermost level of population structure. The K = 6 model is generally consistent with the identified candidate species.
However, all vagrant species (R. haydenii subsp. haydenii, R. haydenii subsp. arbuscular, and R. idahoensis) were recovered within a
single population cluster, along with all individuals assigned to
clade IVc in the combined phylogenetic analysis. A total of three
saxicolous accessions (554f, 556f, and 715F) and three erratic, or
facultatively unattached, accessions (668f, 669f, 670f) were assigned to the cluster with vagrant taxa. Clades IVa and IVd were
also recovered as a single population cluster; however, membership coefficients for individuals with posterior probabilities were
<0.71 for clade IVa and P0.87 for clade IVd.
3.6. Species tree inference and speciation probabilities
The species tree resulting from the analysis of the ten candidate
species estimated with ⁄BEAST is shown in Fig. 7. Within the R.
melanophthalma species-complex, clade I, representing L. novomexicana, was recovered with strong support (posterior probability
>0.99) as the earliest diverging lineage. All candidate species recovered within clade IV in the concatenated analyses formed a wellsupported monophyletic clade and clade II was sister to this lineage, although without significant posterior support. Relationships
among candidate species recovered within clade IV lack strong statistical support, although the three vagrant species formed a wellsupported clade (PP > 0.95), sister to clades IVa, IVc, and IVd. The
Bayesian species delimitation results for the R. melanophthalma
species-complex are shown in Fig. 7. Bayesian species delimitation
supports the ten candidate species guide tree with speciation probabilities > 0.95 on all nodes. Exploratory analyses using different
prior distributions for h and s0 did not affect speciation
probabilities.
3.7. Morphology and chemistry
We adopted the approach of Wiens and Penkrot (2002), suggesting that in order for characters to diagnose a lineage they must
be invariant for alternative character states or show no overlap in
trait values. Both vegetative morphology and reproductive characters, spore size and shape, were highly variable within some candidate species, and overall we were unable to identify morphological
or reproductive characters corroborating candidate species following Wiens and Penkrot (2002).
Occurrence of the 11 most common compounds identified in
HPLC analyses within each lineage is summarized in Table 5. The
majority of specimens belonged to the usnic/psoromic acids chemotype (119 specimens, including all specimens of L. n ovomexicana),
having a broad geographical and ecological distribution; 9 specimens contained usnic, psoromic, and lecanoric acid; and 5 specimens contained usnic, psoromic, and orsellinic acid. All sampled
vagrant specimens expressed usnic acid only. In addition to the
previously reported psoromic acid, we found 2’-O-demethylsubpsoromic acid, 2’-O-demethylpsoromic acid, and the recently described b-orcinol depsidone, subpsoromic acid (Elix 2000). The
dibenzofuran-derivative, usnic acid, was present in all samples,
and some combination of the aliphatic acids – dehydroprotoconstipatic acid and constipatic acid – were present in all individuals,
except the sampled vagrant taxa. We found gyrophoric (triorsellininc) acid and also the monocyclic-depside precursor, orsellinic
acid, restricted to specimens assigned to clade IVb (defined in
3.3) in the combined molecular analyses, in addition to previous
reports for lecanoric (diorsellinic) acid (Arup and Grube, 2000;
McCune, 1987).
4. Discussion
In this study, we assembled multiple lines of evidence to identify and delimit candidate species within the R. melanophthalma
species-complex. Based on all of the available evidence, we circumscribe ten candidate species within this complex, and most fall
within a nominal taxon currently recognized as a single cosmopolitan species, R. melanophthalma. Genetic patterns, generated by
population-level processes operating within divergent lineages,
provide an informative perspective about the process of speciation
in the R. melanophthalma species-complex. Results of the empirical
tests delimiting species are summarized in Table 6.
Our results provide a compelling case of diversification within
the R. melanophthalma species-complex using molecular data and
multiple analytical tools, although most candidate species were
not supported unambiguously by independent datasets. Besides
the placodiod crustose taxon, L. novomexicana, we found that the
greatest morphological and chemical variation was restricted to
closely related lineages (sampled vagrant taxa and clades IVb and
IVc), while morphological and chemical characters supporting
more divergent groups were not identified. Ecological interactions
are expected to drive phenotypic divergence during the early
stages of lineage diversification when species richness is low and
available niches are ‘‘open’’ (Schluter, 2000). The ecological transition from a saxicolous attached form to morphologically distinct
vagrant forms appears to follow the ecological theory of adaptation
(Funk et al., 2006). The STRUCTURE analysis assigned all vagrant
forms to a single population cluster, suggesting a recent divergence
Author's personal copy
S.D. Leavitt et al. / Molecular Phylogenetics and Evolution 59 (2011) 587–602
599
Table 6
Summary of data supporting candidate species within the R. melanophthalma species-complex. Fixed characters, the total number of fixed nucleotide characters relative to the
remaining data matrix (combined ribosomal loci-b-tubulin-MCM7); genealogical, candidate species recovered as an exclusive lineage in gene haplotype networks, ‘⁄0 indicate
support from individual ribosomal, b-tubulin, and MCM7 network reconstructions; STRUCTURE, indicates if the candidate species was recovered as a unique population cluster in
the Bayesian clustering analysis; BPP, posterior support for speciation events calculated using a Bayesian modeling approach; independent characters support, support from
independent morphological or chemical data.
Candidate species
Fixed characters
Genealogical
exclusivity
STRUCTURE
BPP
(PP > 0.95)
Independent character support
Clade
Clade
Clade
Clade
Clade
Clade
Clade
Clade
Clade
Clade
Yes
Yes
Yes
Yes
No
Yes
Yes
Yes
Yes
Yes
Yes⁄⁄⁄
Yes⁄-⁄
Yes⁄⁄⁄
No
No
No
Yes⁄–
Yes⁄⁄Yes⁄–
Yes⁄-⁄
Yes
Yes
Yes
=vagrant taxa & clade
=vagrant taxa & clade
=vagrant taxa & clade
=clade IVa & IVd
Yes
=vagrant taxa & clade
=clade IVa & IVd
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Lobate, placodioid thallus morphology
Not identified
Not identified
Vagrant thallus morphology and usnic acid only
Vagrant thallus morphology and usnic acid only
Vagrant thallus morphology and usnic acid only
Not identified
Lecanoric/orsellinic acid are exclusive to this lineage
Not identified
Not identified
I (L. novomexicana)
II
III
IV (R. haydenii)
IV (R. h. ssp. arbuscula)
IV (R. idahonesis)
IVa
IVb
IVc
IVd
(21–9–2)
(3–0–0)
(15–0–0)
(0–0–1)
(1–0–5)
(7–0–0)
(2–1–0)
(1–0–0)
(1–0–0)
of morphologically diverse vagrant taxa. However, the inclusion of
saxicolous attached taxa within this cluster suggests a recent
divergence from saxicolous attached forms or an underlying genetic predisposition to vagrancy in at least some saxicolous lineages.
Previous research (Leavitt et al., submitted for publication) indentified multiple independent origins of vagrancy within the lichen
genus Xanthoparmelia (Parmeliaceae), but our data suggest that
that vagrancy in the R. melanophthalma species-complex is limited
to a single closely related lineage, even among morphologically
distinct vagrant forms. However, a broader sample of vagrant individuals is essential to adequately addressing this question, particularly R. haydenii subsp. haydenii recently described in China
(Zheng et al., 2007).
Previous studies have used thin-layer-chromatography (TLC) to
characterize lichen secondary metabolic products within Rhizoplaca. In this study HPLC provided a more sensitive approach to determine secondary metabolite diversity within the R. melanophthalma
group, as many newly reported compounds here would be masked
by other compounds, or likely found at levels undetectable by TLC.
While data have supported the taxonomic use of some secondary
metabolic characters for delimiting lichen taxa (Schmitt and
Lumbsch, 2004; Tehler and Källersjö, 2001), other studies found
no correlation between chemotypes and lineages identified using
molecular phylogenetic reconstructions (Articus et al., 2002;
Buschbom and Mueller, 2006; Nelsen and Gargas, 2009; Velmala
et al., 2009). We have identified chemical characters corroborating
some lineages identified within the R. melanophthalma group,
including: clade IVb containing a combination of orsellinic, lecanoric, and gyrophoric acids; and R. haydenii, R. haydenii, ssp. arbuscula, and R. idahoensis all lack aliphatic acids related to constipatic
acid. However, we were unable to identify secondary metabolic
characters supporting most identified putative lineages, including
the most genetically divergent clades.
McCune (1987) suggested three hypotheses to explain chemical
diversity in the genus Rhizoplaca: (1) chemotypes are sibling species that cannot or seldom hybridize assuming there are no reproductive barriers, (2) factors favoring polymorphism in chemistry
do not differ markedly between regions, or (3) the polymorphism
is neutral to natural selection. Although the present study was
not designed to explicitly test these hypotheses, our results indicate within the usnic/psoromic acid race multiple lineages co-occur. The usnic/psoromic/lecanoric acid race appears to be a
distinct lineage also containing specimens lacking lecanoric acid
but expressing the lecanoric acid precursor, orsellinic acid.
Additional studies will be needed to fully elucidate the relationship
between R. melanophthalma s.l. containing lecanoric or orsellinic
acids. Our sampling of the usnic acid chemical race in the
R. melanophthalma species-complex was limited to a single
IVc
IVc
IVc
IVc
saxicolous attached individual (715f) and all vagrant taxa. The
saxicolous R. melanophthalma chemical race containing placodiolic
acid was not sampled and its relationship to sampled taxa remains
in question.
Porter (1998) reported a correlation between some secondary
metabolites and elevation in R. melanophthalma populations along
an altitudinal gradient on Thousand Lakes Mountain, Utah. Besides
the strict correlation of lecanoric and orsellinic acid with clade IVb,
the present study did not identify any specific correlations between lineages identified from molecular data and expressed secondary metabolites on Thousand Lake Mountain, suggesting that
the production of most minor compounds may be environmentally
induced. A combination of species diversity in lichen-forming symbionts (alga and fungus) and ecological factors may explain secondary metabolite variation among the Thousand Lake Mountain
populations (Brunauer et al., 2007).
The current study was generally limited to the Intermountain
region of western North America, and robust data from a broader
geographic sampling will be essential to understand the general
geographic distribution of the candidate species identified in this
study. We anticipate that with improved sampling, additional lineages may be identified within the R. melanophthalma species-complex, particularly within L. novomexicana s. l. However, with the
exception of L. novomexicana, GenBank accessions were recovered
within the candidate species identified from our combined dataset
set from samples in western North America in the ITS topology,
suggesting our candidate species may have wide-spread distributions. The occurrence of cohesive cosmopolitan lineages found
sympatrically with closely related divergent populations poses
challenging questions about the processes that yield and maintain
cohesive lineages within widespread lichenized ascomycetes.
Clade-specific ecological or microhabitat differences considered
alone do not appear to offer a plausible explanation of how sympatric diversification may occur in the candidate species. Some lineages exhibit extensive microsympatry (i.e., divergent lineages
occurring within a single sampled plot), as well as the production
of abundant perennial apothecia (sexual fruiting bodies) without
detectable gene flow or hybridization between microsympatric
individuals. This pattern suggests that candidate species may have
achieved a significant level of reproductive isolation. However, the
role of spatio-temporal isolation in lichenized fungal reproduction
is relatively unexplored. It has been proposed that competition for
symbiotic partners may be a major driver of diversity in mutualistic relationships (Bruns, 1995; O’Brien et al., 2009) and investigating competition for symbionts may provide insights into
mechanisms that possibly drive sympatric speciation.
Within lichenized fungi, gene trees have often been used to infer species boundaries, and the over-reliance on a single locus has
Author's personal copy
600
S.D. Leavitt et al. / Molecular Phylogenetics and Evolution 59 (2011) 587–602
been problematic in delimiting species because gene duplication,
horizontal gene transfer, and deep coalescence may create conflict
between the sampled gene tree and the true species tree (de Queiroz and Donoghue, 1990; Maddison, 1997). Furthermore, simulations have shown that concatenation of independent genetic loci
can mislead phylogeny estimation (Edwards et al., 2007). Interestingly, major topological differences between our concatenated and
coalescent-based phylogeny were recovered (Figs. 2 and 7).
Although our study emphasized species delimitation, phylogenetic
relationships have crucial significance for understanding the role of
biogeography, character evolution, and accurately dating divergences. The coalescent-based species tree assembled from the gene
trees using ⁄BEAST provides less support for many of the relationships recovered, and these posteriors could more accurately reflect
the level of uncertainty in these nodes (Leaché and Rannala, 2010).
Our data suggest that a coalescent approach, explicitly incorporating the independent sorting of gene trees while reconstructing the
species trees, is likely crucial in order to recover the best estimate
of the R. melanophthalma complex species phylogeny. Improved
sampling of the R. melanophthalma species-complex with additional genetic loci will be necessary to assemble an unambiguous
species tree, stabilize the taxonomy of this group, and infer phylogeographical patterns.
In some cases, rapidly evolving molecular characters may reach
fixation in ephemerally isolated demes, with the potential to reticulate with other conspecific lineages at some point in the future
(O’Hara, 1993). Additionally, phylogenetic structure can extend below the level of the species, particularly within asexual and haploid
genomes (Birky et al., 1989, 1983; Davis, 1996; de Queiroz and
Donoghue, 1990) making species limits based on molecular data
within lichenized fungi particularly susceptible to excessive subdivision. Here we use a Bayesian modeling approach, incorporating
the coalescent process and gene tree uncertainties, to generate
the posterior probabilities of speciation (Yang and Rannala,
2010). Although the guide tree has been shown to play a critical
role in the outcome of the species delimitation model (Leaché
and Fujita, 2010), in this study significant posterior probabilities
of speciation were recovered between even the closely related vagrant Rhizoplaca species. Given the fairly strict criteria the BPP assumes to designate species (Yang and Rannala, 2010), we feel that
recognizing ten species provides a reasonable perspective on species diversity within the sampled R. melanophthalma species-complex. Our results indicate that in some cases morphological and
chemical data provide additional lines of evidence of supporting
distinct lineages. However, other candidate species are apparently
morphologically and chemically cryptic.
In spite of the limitations in delimiting taxa using molecular
data, most of the candidate species identified in this study, were
not supported by diagnostic morphological or chemical characters,
and the effective use of genetic data appears to be an essential approach to appropriately identify natural groups in many fungal lineages (Crespo and Pérez-Ortega, 2009). The authors plan a detailed
taxonomic revision for the R. melanophthalma species-complex in
the near future, including additional taxonomic and morphological
sampling to more fully characterize boundaries between candidate
species. Results from this study suggest that robust taxon and
molecular data sampling, using appropriate empirical operational
criteria to delimit species, may provide an improved perspective
on the diversification of lichenized fungi, compared to traditional
morphological and chemical characters. However, we are not advocating the use of genetic data to the exclusion of other evidence for
delimiting species; because corroboration of species boundaries
via independent lines of evidence is important to the establishment of robust hypotheses of species diversity.
Analysis of the R. melanophthalma species-complex comprises
the larger focus of this study, which is using robust species
delimitation in morphologically cryptic and recently diverged lichenized fungi. Rhizoplaca – as traditionally circumscribed – is a
small, morphologically diverse lichen genus represented by 9 species (Arup and Grube, 2000; Zhou et al., 2006). This study indicates
overall diversity within umbilicate Rhizoplaca species may be
vastly underestimated, as multiple previously unrecognized lineages were identified within the R. melanophthalma group. Previous studies have identified well-supported lineages within R.
chrysoleuca corresponding to two phenotypic groups (Zhou et al.,
2006), and divergent, well-supported clades within the outgroup
taxon R. chrysoleuca were also recovered in this study, which suggests an additional unrecognized diversity. Extending the present
sampling of the R. melanophthalma species-complex to include a
broader geographic context and robust sampling of underrepresented lineages will be critical to improve the understanding of
the mechanisms driving speciation in lichenized fungi. Furthermore, an extension of the present sampling to other closely related
cosmopolitan Rhizoplaca and Lecanora species-complexes will provide a potential opportunity for developing a comprehensive classification system for other closely related taxa. Additionally,
continued investigation of independent characters supporting candidate lineages will be essential for generating robust hypotheses
of species boundaries.
Acknowledgments
We thank Byron Adams (Provo), Eric Green (Provo), Roger
Rosentreter (Boise), Imke Schmitt (Minnesota), and Jack Sites (Provo) for valuable discussion and comments on early versions of this
manuscript; Christopher Jones and Peter Ririe for laboratory assistance; and LauraDawn Leavitt (Provo) and Gajendra Shrestha (Provo) for invaluable help in preparing figures. We would also like to
thank Jack Elix (Canberra) for providing a digital HPLC library and
Thorsten Lumbsch (Chicago) for a collection of authentic substances. This study was supported, in part, by funds from the University of Minnesota to Imke Schmitt (St. Paul), Brigham Young
University graduate mentoring and graduate research fellowship
awards to SDL, and a Walmart Foundation Internship Grant to
JDF. The funding sources had no role in study design, data collection and analysis, preparation or decision to publish this
manuscript.
Appendix A. Supplementary material
Supplementary data associated with this article can be found, in
the online version, at doi:10.1016/j.ympev.2011.03.020.
References
Articus, K., Mattsson, J.E., Tibell, L., Grube, M., Wedin, M., 2002. Ribosomal DNA and
beta-tubulin data do not support the separation of the lichens Usnea florida and
U. Subfloridana as distinct species. Mycol. Res. 106, 412–418.
Arup, U., Grube, M., 2000. Is Rhizoplaca (Lecanorales, lichenized Ascomycota) a
monophyletic genus? Can. J. Botany 78, 318–327.
Aslan, A., Budak, G., Karabulut, A., 2004. The amounts Fe, Ba, Sr, K, Ca and Ti in some
lichens growing in Erzurum province (Turkey). J. Quant. Spectrosc. Radiat.
Transf. 88, 423–431.
Avise, J.C., Ball, A.S., 1990. Principles of genealogical concordance in species
concepts and biological taxonomy. In: Futuyma, D., Antonovics (Eds.), Oxford
Surveys in Evolutionary Biology. Oxford University Press, Oxford.
Baum, D.A., Shaw, K.L., 1995. Genealogical perspectives on the species problem. In:
Hoch, P.C., Stephenson, A.G. (Eds.), Experimental and Molecular Approaches to
Plant Biosystematics. Missouri Botanical Garden, St. Louis, pp. 289–303.
Baurain, D., Brinkmann, H., Philippe, H., 2007. Lack of resolution in the animal
phylogeny: closely spaced cladogeneses or undetected systematic errors? Mol.
Biol. Evol. 24, 6–9.
Birky Jr., C.W., Fuerst, P., Maruyama, T., 1989. Organelle gene diversity under
migration, mutation, and drift: equilibrium expectations, approach to
equilibrium, effects of heteroplasmic cells, and comparison to nuclear genes.
Genetics 121, 613–627.
Author's personal copy
S.D. Leavitt et al. / Molecular Phylogenetics and Evolution 59 (2011) 587–602
Birky Jr., C.W., Maruyama, T., Fuerst, P., 1983. An approach to population and
evolutionary genetic theory for genes in mitochondria and chloroplasts, and
some results. Genetics 103, 513–527.
Blanco, O., Crespo, A., Divakar, P.K., Esslinger, T.L., Hawksworth, D.L., Thorsten
Lumbsch, H., 2004a. Melanelixia and Melanohalea, two new genera segregated
from Melanelia (Parmeliaceae) based on molecular and morphological data.
Mycol. Res. 108, 873–884.
Blanco, O., Crespo, A., Elix, J.A., Hawksworth, D.L., Lumbsch, H.T., 2004b. A molecular
phylogeny and a new classification of parmelioid lichens containing
Xanthoparmelia-type lichenan (Ascomycota: Lecanorales). Taxon 53, 959–975.
Boni, M.F., Posada, D., Feldman, M.W., 2007. An exact nonparametric method for
inferring mosaic structure in sequence triplets. Genetics 176, 1035–1047.
Brunauer, G., Hager, A., Grube, M., Türk, R., Stocker-Wörgötter, E., 2007. Alterations
in secondary metabolism of aposymbiotically grown mycobionts of Xanthoria
elegans and cultured resynthesis stages. Plant Physiol. Biochem. 45, 146–151.
Bruns, T.D., 1995. Thoughts on the processes that maintain local species diversity of
ectomycorrhizal fungi. Plant Soil 170, 63–73.
Buschbom, J., Mueller, G.M., 2006. Testing ‘‘species pair’’ hypotheses: evolutionary
processes in the lichen-forming species complex Porpidia flavocoerulescens and
Porpidia melinodes. Mol. Biol. Evol. 23, 574–586.
Cansaran, D., Aras, S., Kandemir, I., Halici, M.G., 2006. Phylogenetic relations of
Rhizoplaca Zopf. from anatolia inferred from ITS sequence data. Z. Naturforsch.
Sect. C: J. Biosci. 61, 405–412.
Carbone, I., Kohn, L.M., 1999. A method for designing primer sets for speciation
studies in filamentous Ascomycetes. Mycologia 91, 553–556.
Cardinale Jr., M., J.V.d.C., Müller, H., Berg, G., Grube, M., 2008. In situ analysis of the
bacterial community associated with the reindeer lichen Cladonia arbuscula
reveals predominance of Alphaproteobacteria. FEMS Microb. Ecol. 66, 63–71.
Carstens, B.C., Dewey, T.A., 2010. Species delimitation using a combined coalescent
and information-theoretic approach: an example from North American Myotis
bats. Syst. Biol. 54, 400–414.
Clement, M., Posada, D., Crandall, K.A., 2000. TCS: a computer program to estimate
gene genealogies. Mol. Ecol. 9, 1657–1659.
Crespo, A., Lumbsch, H.T., Mattsson, J.-E., Blanco, O., Divakar, P.K., Articus, K., Wiklund,
E., Bawingan, P.A., Wedin, M., 2007. Testing morphology-based hypotheses of
phylogenetic relationships in Parmeliaceae (Ascomycota) using three ribosomal
markers and the nuclear RPB1 gene. Mol. Phylogenet. Evol. 44, 812–824.
Crespo, A., Pérez-Ortega, S., 2009. Cryptic species and species pairs in lichens: a
discussion on the relationship between molecular phylogenies and
morphological characters. Anales del Jardin Botanico de Madrid 66 (S1), 71–81.
Culberson, C.F., 1972. Improved conditions and new data for identification of lichen
products by standardized thin-layer chromatographic method. J. Chromatogr. A
72, 113–125.
Davis, J.I., 1996. Phylogenetics, molecular variation, and species concepts.
Bioscience 46, 502–511.
Dayrat, B., 2005. Towards integrative taxonomy. Biol. J. Linn. Soc. 85, 407–415.
de Queiroz, K., 1998. The General Lineage Concept of Species, Species Criteria, and
the Process of Speciation: A Conceptual Unification and Terminological
Recommendations. Oxford University Press, Oxford.
de Queiroz, K., 1999. The general lineage concept of species and the defining
properties of the species category. Species, New Interdisciplinary Essays, 49–89.
de Queiroz, K., 2007. Species concepts and species delimitation. Syst. Biol. 56, 879–
886.
de Queiroz, K., Donoghue, M.J., 1990. Phylogenetic systematics and species
revisited. Cladistics 6, 83–90.
Dettman, J.R., Jacobson, D.J., Taylor, J.W., 2003. A multilocus genealogical approach
to phylogenetic species recognition in the model eukaryote Neurospora.
Evolution 57, 2703–2720.
Dillman, K.L., 1996. Use of the lichen Rhizoplaca melanophthalma as a biomonitor in
relation to phosphate refineries near Pocatello. Idaho Environ. Pollut. 92, 91–96.
Drummond, A., Ho, S., Phillips, M., Rambaut, A., 2006. Relaxed phylogenetics and
dating with confidence. PLoS Biol. 4, e88.
Duminil, J., Caron, H., Scotti, I., Cazal, S.-O., Petit, R., 2006. Blind population genetics
survey of tropical rainforest trees. Mol. Ecol. 15, 3505–3513.
Edgar, R.C., 2004. MUSCLE: a multiple sequence alignment method with reduced
time and space complexity. BMC Bioinformatics 5, 1–19.
Edwards, S.V., Liu, L., Pearl, D.K., 2007. High-resolution species trees without
concatenation. Proc. Natl. Acad. Sci. USA 104, 5936–5941.
Evanno, G., Regnaut, S., Goudet, J., 2005. Detecting the number of clusters of
individuals using the software STRUCTURE: a simulation study. Mol. Ecol. 14,
2611–2620.
Falush, D., Stephens, M., Pritchard, J.K., 2003. Inference of population structure
using multilocus genotype data: linked loci and correlated allele frequencies.
Genetics 164, 1567–1587.
Feige, G.B., Lumbsch, H.T., Huneck, S., Elix, J.A., 1993. Identification of lichen
substances by a standardized high-performance liquid chromatographic
method. J. Chromatogr. 646, 417–427.
Funk, D.J., Nosil, P., Etges, W.J., 2006. Ecological divergence exhibits consistently
positive associations with reproductive isolation across disparate taxa. Proc.
Natl. Acad. Sci. USA 103, 3209–3213.
Ganley, A.R.D., Kobayashi, T., 2007. Highly efficient concerted evolution in the
ribosomal DNA repeats: total rDNA repeat variation revealed by whole-genome
shotgun sequence data. Genome Res. 17, 184–191.
Gardes, M., Bruns, T.D., 1993. ITS primers with enhanced specificity for
basidiomycetes – application to the identification of mycorrhizae and rusts.
Mol. Ecol. Notes 2, 113–118.
601
Gibbs, M.J., Armstrong, J.S., Gibbs, A.J., 2000. Sister-scanning: a Monte Carlo
procedure for assessing signals in recombinant sequences. Bioinformatics 16,
573–582.
Groeneveld, L., Weisrock, D., Rasoloarison, R., Yoder, A., Kappeler, P., 2009. Species
delimitation in lemurs: multiple genetic loci reveal low levels of species
diversity in the genus Cheirogaleus. BMC Evol. Biol. 9, 30.
Grube, M., Cardinale, M., de Castro Jr., J.V., Muller, H., Berg, G., 2009. Species-specific
structural and functional diversity of bacterial communities in lichen
symbioses. ISME J. 3, 1105–1115.
Gutiérrez, G., Blanco, O., Divakar, P., Lumbsch, H., Crespo, A., 2007. Patterns of Group
I intron presence in nuclear SSU rDNA of the Lichen Family Parmeliaceae. J. Mol.
Evol. 64, 181–195.
Hale, M.E., 1990. A Synopsis of the Lichen Genus Xanthoparmelia (Vainio) Hale
(Ascomycotina, Parmeliaceae). Smithsonian Institution Press, Washington, DC.
Hall, B.G., 2007. Phylogenetic Trees Made Easy: A How-To Manual. Sinauer
Associates, Sunderland, MA.
Heath, L., van der Walt, E., Varsani, A., Martin, D.P., 2006. Recombination patterns in
Aphthoviruses mirror those found in other picornaviruses. J. Virol. 80, 11827–
11832.
Heled, J., Drummond, A.J., 2010. Bayesian inference of species trees from multilocus
data. Mol. Biol. Evol. 27, 570–580.
Hey, J., Waples, R.S., Arnold, M.L., Butlin, R.K., Harrison, R.G., 2003. Understanding
and confronting species uncertainty in biology and conservation. Trends Ecol.
Evol. 18, 597–603.
Hodkinson, B., Lutzoni, F., 2009. A microbiotic survey of lichen-associated bacteria
reveals a new lineage from the Rhizobiales. Symbiosis 49, 163–180.
Holmes, E.C., Worobey, M., Rambaut, A., 1999. Phylogenetic evidence for
recombination in dengue virus. Mol. Biol. Evol. 16, 405–409.
Hudson, R.R., Coyne, J.A., 2002. Mathematical consequences of the genealogical
species concept. Evolution 56, 1557–1565.
Huelsenbeck, J.P., Rannala, B., 2004. Frequentist properties of Bayesian posterior
probabilities of phylogenetic trees under simple and complex substitution
models. Syst. Biol. 53, 904–913.
Huelsenbeck, J.P., Ronquist, F., 2001. MrBayes: bayesian inference of phylogenetic
trees. Bioinformatics 17, 754–755.
Huelsenbeck, J.P., Ronquist, F., Nielsen, R., Bollback, J.P., 2001. Bayesian inference of
phylogeny and its impact on evolutionary biology. Science 294, 2310–2314.
Huneck, S., 1999. The significance of lichens and their metabolites.
Naturwissenschaften 86, 559–570.
Huneck, S., Yoshimura, I., 1996. Identification of Lichen Substances. Springer, New
York.
Knowles, L.L., Carstens, B.C., 2007. Delimiting species without monophyletic gene
trees. Syst. Biol. 56, 887–895.
Kosakovsky Pond, S.L., Posada, D., Gravenor, M.B., Woelk, C.H., Frost, S.D.W., 2006.
GARD: a genetic algorithm for recombination detection. Bioinformatics 22,
3096–3098.
Kroken, S., Taylor, J.W., 2001. A gene genealogical approach to recognize
phylogenetic species boundaries in the lichenized fungus Letharia. Mycologia
93, 38–53.
LaGreca, S., Lumbsch, H.T., 2001. Three species of Lecanora new to North America,
with notes on other poorly known Lecanoroid lichens. Bryologist 104, 204–211.
Latch, E., Dharmarajan, G., Glaubitz, J., Rhodes, O., 2006. Relative performance of
bayesian clustering software for inferring population substructure and
individual assignment at low levels of population differentiation. Conserv.
Genet. 7, 295–302.
Le Gac, M., Hood, M.E., Fournier, E., Giraud, T., 2007. Phylogenetic evidence of hostspecific cryptic species in the anther smut fungus. Evolution 61, 15–26.
Leache, A.D., 2009. Species tree discordance traces to phylogeographic clade
boundaries in North American fence lizards (Sceloporus). Syst. Biol. 58, 547–559.
Leaché, A.D., Fujita, M.K., 2010. Bayesian species delimitation in West African forest
geckos (Hemidactylus fasciatus). Proc. Roy. Soc. B: Biol. Sci. 277, 3071–3077.
Leaché, A.D., Rannala, B., 2010. The accuracy of species tree estimation under
simulation: a comparison of methods. Syst. Biol. doi:10.1093/sysbio/syq073.
Leavitt, S.D., Johnson, L.A., Goward, T., St. Clair, L.L., submitted for publication.
Species delimitation in taxonomically difficult lichen-forming fungi: an
example from morphologically and chemically diverse Xanthoparmelia
(Parmeliaceae) in North America.
Leavitt, S.D., Johnson, L.A., St. Clair, L.L., 2011. Species delimitation and evolution in
morphologically and chemically diverse communities of the lichen-forming
genus Xanthoparmelia (Parmeliaceae, Ascomycota) in western North America.
Am. J. Bot. 98, 1–14.
Leuckert, C., Poelt, J., Hahnel, G., 1977. Zur Chemotaxonomie der eurasischen Arten
der Flechtengattung rhizoplaca. Nova Hedwigia 28, 71–129.
Librado, P., Rozas, J., 2009. DnaSP v5: a software for comprehensive analysis of DNA
polymorphism data. Bioinformatics 25, 1451–1452.
Lindblom, L., Ekman, S., 2006. Genetic variation and population differentiation in
the lichen-forming ascomycete Xanthoria parietina on the island Storfosna,
central Norway. Mol. Ecol. 15, 1545–1559.
Lumbsch, H.T., 2002. Analysis of phenolic products in lichens. In: Kranner, I.,
Beckett, R.P., Varma, A. (Eds.), Protocols in Lichenology. Springer, Berlin, pp.
281–295.
Lumbsch, H.T., Schmitt, I., 2001. Molecular data suggest that the Lichen Genus
Pertusaria is not monophyletic. Lichenologist 33, 161–170.
Lutzoni, F., Kauff, F., Cox, C.J., McLaughlin, D., Celio, G., Dentinger, B., Padamsee, M.,
Hibbett, D., James, T.Y., Baloch, E., Grube, M., Reeb, V., Hofstetter, V., Schoch, C.,
Arnold, A.E., Miadlikowska, J., Spatafora, J., Johnson, D., Hambleton, S., Crockett,
Author's personal copy
602
S.D. Leavitt et al. / Molecular Phylogenetics and Evolution 59 (2011) 587–602
M., Shoemaker, R., Sung, G.-H., Lucking, R., Lumbsch, T., O’Donnell, K., Binder, M.,
Diederich, P., Ertz, D., Gueidan, C., Hansen, K., Harris, R.C., Hosaka, K., Lim, Y.-W.,
Matheny, B., Nishida, H., Pfister, D., Rogers, J., Rossman, A., Schmitt, I., Sipman,
H., Stone, J., Sugiyama, J., Yahr, R., Vilgalys, R., . Assembling the fungal tree of
life: progress, classification, and evolution of subcellular traits. Am. J. Bot. 91,
1446–1480.
Lutzoni, F., Pagel, M., Reeb, V., 2001. Major fungal lineages are derived from lichen
symbiotic ancestors. Nature 411, 937–940.
Maddison, W.P., 1997. Gene trees in species trees. Syst. Biol. 46, 523–536.
Martin, D.P., Posada, D., Crandall, K.A., Williamson, C., 2005a. A modified bootscan
algorithm for automated identification of recombinant sequences and
recombination breakpoints. Anglais 21, 98–102.
Martin, D.P., Williamson, C., Posada, D., 2005b. RDP2: recombination detection and
analysis from sequence alignments. Bioinformatics 21, 260–262.
Mayden, R.L., 1999. Consilience and a hierarchy of species concepts: advances
towards closure on the species puzzle. J. Nematol. 31, 95–116.
McCune, B., 1987. Distribution of chemotypes of Rhizoplaca in North America.
Bryologist 90, 6–14.
McCune, B., Rosentreter, R., 2007. Biotic Soil Crust Lichens of the Columbian Basin.
Monographs in North American Lichenology, Corvalis.
McGuire, G., Wright, F., 2000. Improved detection of mosaic sequences within
multiple sequence alignments. Bioinformatics 16, 130–134.
Myllys, L., Lohtander, K., Tehler, A., 2001. b-Tubulin, ITS and group I intron
sequences challenge the species pair concept in Physcia aipolia and P. caesia.
Mycologia 93, 335–343.
Nei, M., 1987. Molecular Evolutionary Genetics. Columbia University Press, New
York.
Nelsen, M.P., Gargas, A., 2009. Assessing clonality and chemotype monophyly in
Thamnolia (Icmadophilaceae). Bryologist 112, 42–53.
Nylander, J.A.A., Ronquist, F., Huelsenbeck, J.P., Nieves-Aldrey, J., 2004. Bayesian
phylogenetic analysis of combined data. Syst. Biol. 53, 47–67.
O’Brien, H.E., Miadlikowska, J., Lutzoni, F., 2009. Assessing reproductive isolation in
highly diverse communities of the lichen-forming fugnal genus Peltigera.
Evolution 63, 2076–2086.
O’Hara, R.J., 1993. Systematic generalization, historical fate, and the species
problem. Syst. Biol. 42, 231–246.
O’Meara, B.C., 2010. New heuristic methods for joint species delimitation and
species tree inference. Syst. Biol. 59, 59–73.
Ott, S., Brinkmann, M., Wirtz, N., Lumbsch, H.T., 2004. Mitochondrial and nuclear
ribosomal DNA data do not support the separation of the Antarctic lichens
Umbilicaria kappenii and Umbilicaria antarctica as distinct species. Lichenologist
36, 227–234.
Padidam, M., Sawyer, S., Fauquet, C.M., 1999. Possible emergence of new
geminiviruses by frequent recombination. Virology 265, 218–225.
Porter, L., 1998. Rhiz. Brigham Young University, Provo, Utah.
Posada, D., Crandall, K.A., 2001a. Evaluation of methods for detecting recombination
from DNA sequences: computer simulations. Proc. Natl. Acad. Sci. USA 98,
13757–13762.
Posada, D., Crandall, K.A., 2001b. Selecting the best-fit model of nucleotide
substitution. Syst. Biol. 50, 580–601.
Pringle, A., Baker, D.M., Platt, J.L., Wares, J.P., Latgé, J.P., Taylor, J.W., 2005. Cryptic
speciation in the cosmopolitan and clonal human pathogenic fungus Aspergillus
fumigatus. Evolution 59, 1886–1899.
Printzen, C., 2009. Lichen systematics: the role of morphological and molecular data
to reconstruct phylogenetic relationships. Progress in Botany 71. Springer Berlin
Heidelberg, pp. 233–275.
Pritchard, J.K., Stephens, M., Donnelly, P., 2000. Inference of population structure
using multilocus genotype data. Genetics 155, 945–959.
Rambaut, A., Drummond, A.J., 2007. Tracer v1.5. <http://beast.bio.ed.ac.uk/Tracer>.
Rannala, B., Yang, Z., 2003. Bayes estimation of species divergence times and
ancestral population sizes using DNA sequences from multiple loci. Genetics
164, 1645–1656.
Roe, A.D., Sperling, F.A.H., 2007. Population structure and species boundary
delimitation of cryptic Dioryctria moths: an integrative approach. Mol. Ecol.
16, 3617–3633.
Rokas, A., Williams, B.L., King, N., Carroll, S.B., 2003. Genome-scale approaches to
resolving incongruence in molecular phylogenies. Nature 425, 798–804.
Rosentreter, R., 1993. Vagrant lichens in North America. Bryologist 96, 333–338.
Rowe, G., Beebee, T.J.C., 2007. Defining population boundaries: use of three Bayesian
approaches with microsatellite data from British natterjack toads (Bufo
calamita). Mol. Ecol. 16, 785–796.
Ruiz-Sanchez, E., Sosa, V., 2010. Delimiting species boundaries within the
neotropical bamboo Otatea (Poaceae: Bambusoideae) using molecular,
morphological and ecological data. Mol. Phylogenet. Evol. 54, 344–356.
Ryan, B.D., 2001. Rhizoplaca. In: Nash, T.H., III., Ryan, B.D., Gries, C., Bungartz, F.
(Eds.), Lichen Flora of the Greater Sonoran Desert Region. Lichens Unlimited,
Tempe.
Saisho, D., Purugganan, M.D., 2007. Molecular phylogeography of domesticated
barley traces expansion of agriculture in the old world. Genetics 177, 1765–1776.
Schluter, D., 2000. The Ecology of Adaptive Radiations. Oxford University Press,
Oxford.
Schmitt, I., Crespo, A., Divakar, P.K., Fankhauser, J.D., Herman-Sackett, E., Kalb, K.,
Nelsen, M.P., Rivas-Plata, E., Shimp, A.D., Widhelm, T., Lumbsch, H.T., 2009. New
primers for promising single-copy genes in fungal phylogenies and systematics.
Persoonia 23, 35–40.
Schmitt, I., Lumbsch, H.T., 2004. Molecular phylogeny of the Pertusariaceae
supports secondary chemistry as an important systematic character set in
lichen-forming ascomycetes. Mol. Phylogenet. Evol. 33, 43–55.
Selbmann, L., Zucconi, L., Ruisi, S., Grube, M., Cardinale, M., Onofri, S., 2010.
Culturable bacteria associated with Antarctic lichens: affiliation and
psychrotolerance. Polar Biol. 33, 71–83.
Sites, J.W., Marshall, J.C., 2004. Operational criteria for delimiting species. Ann. Rev.
Ecol. Evol. Syst. 35, 199–227.
Šlapeta, J., López-García, P., Moreira, D., 2006. Global dispersal and ancient cryptic
species in the smallest Marine eukaryotes. Mol. Biol. Evol. 23, 23–29.
Smith, J.M., 1992. Analyzing the mosaic structure of genes. J. Mol. Evol. 34, 126–129.
Stamatakis, A., 2006. RAxML-VI-HPC: maximum likelihood-based phylogenetic
analyses with thousands of taxa and mixed models. Bioinformatics 22, 2688–
2690.
Stamatakis, A., Hoover, P., Rougemont, J., 2008. A rapid bootstrap algorithm for the
RAxML web servers. Syst. Biol. 57, 758–771.
Tehler, A., Källersjö, M., 2001. Parmeliopsis ambigua and P. hyperopta (Parmeliaceae):
species or chemotypes? Lichenologist 33, 403–408.
Templeton, A.R., Crandall, K.A., Sing, C.F., 1992. A cladistic analysis of phenotypic
associations with haplotypes inferred from restriction endonuclease mapping
and DNA sequence data. III. Cladogram estimation. Genetics 132, 619–633.
Templeton, A.R., Sing, C.F., 1993. A cladistic analysis of phenotypic associations with
haplotypes inferred from restriction endonuclease mapping. IV. Nested Analyses
with cladogram uncertainty and recombination. Genetics 134, 659–669.
Thell, A., 1999. Group I intron versus ITS sequences in phylogeny of cetrarioid
lichens. Lichenologist 31, 441–449.
Ugur, A., Özden, B., Saç, M., Yener, G., Altinbasß, Ü., Kurucu, Y., Bolca, M., 2004.
Lichens and mosses for correlation between trace elements and 210Po in the
areas near coal-fired power plant at Yatağan, Turkey. J. Radioanal. Nucl. Chem.
259, 87–92.
Velmala, S., Myllys, L., Halonen, P., Goward, T., Ahti, T., 2009. Molecular data show
that Bryoria fremontii and B. Tortuosa (Parmeliaceae) are conspecific.
Lichenologist 41, 231–242.
Vieites, D.R., Wollenberg, K.C., Andreone, F., Köhler, J., Glaw, F., Vences, M., 2009.
Vast underestimation of Madagascar’s biodiversity evidenced by an integrative
amphibian inventory. Proc. Natl. Acad. Sci. 106, 8267–8272.
Wedin, M., Westberg, M., Crewe, A.T., Tehler, A., Purvis, O.W., 2009. Species
delimitation and evolution of metal bioaccumulation in the lichenized
Acarospora smaragdula (Ascomycota, Fungi) complex. Cladistics 25, 161–172.
Weisrock, D.W., Rasoloarison, R.M., Fiorentino, I., Ralison, J.M., Goodman, S.M.,
Kappeler, P.M., Yoder, A.D., 2010. Delimiting species without nuclear
monophyly in Madagascar’s mouse lemurs. PLoS ONE 5, e9883.
Wheeler, D.L., Barrett, T., Benson, D.A., Bryant, S.H., Canese, K., Chetvernin, V.,
Church, D.M., DiCuccio, M., Edgar, R., Federhen, S., Geer, L.Y., Kapustin, Y.,
Khovayko, O., Landsman, D., Lipman, D.J., Madden, T.L., Maglott, D.R., Ostell, J.,
Miller, V., Pruitt, K.D., Schuler, G.D., Sequeira, E., Sherry, S.T., Sirotkin, K.,
Souvorov, A., Starchenko, G., Tatusov, R.L., Tatusova, T.A., Wagner, L., Yaschenko,
E., 2006. Database resources of the National Center for Biotechnology
Information. Nucl. Acids Res., gkl1031.
White, T.J., Bruns, T.D., Lee, S., Taylor, J. (Eds.), 1990. Amplification and Direct
Sequencing of Fungal Ribosomal RNA Genes for Phylogenetics. Academic Press,
San Diego.
Wiens, J., Penkrot, T., 2002. Delimiting species using DNA and morphological
variation and discordant species limits in spiny lizards (Sceloporus). Syst. Biol.
51, 69–91.
Wiens, J.J., 1998. Combining data sets with different phylogenetic histories. Syst.
Biol. 47, 568–581.
Wirtz, N., Printzen, C., Lumbsch, H.T., 2008. The delimitation of Antarctic and
bipolar species of neuropogonoid Usnea (Ascomycota, Lecanorales): a cohesion
approach of species recognition for the Usnea perpusilla complex. Mycol. Res.
112, 472–484.
Yang, Z., Rannala, B., 2010. Bayesian species delimitation using multilocus sequence
data. Proc. Natl. Acad. Sci. 107, 9264–9269.
Zheng, X.-L., Sheng, H.-M., An, L.-Z., 2007. Phylogenetic analysis of lichen-forming
fungi Rhizoplaca Zopf from China based on ITS data and morphology. Z.
Naturforsch. Sect. C: J. Biosci. 62c, 757–764.
Zhou, Q.-M., Guo, S.-Y., Huang, M.-R., Wei, J.-C., 2006. A study of the genetic
variability of Rhizoplaca chrysoleuca using DNA sequences and secondary
metabolic substances. Mycologia 98, 57–67.