New primers for promising single-copy genes in fungal phylogenetics and systematics.

Developing powerful phylogenetic markers is a key concern in fungal phylogenetics. Here we report degenerate primers that amplify the single-copy genes Mcm7 (MS456) and Tsr1 (MS277) across a wide range of Pezizomycotina (Ascomycota). Phylogenetic analyses of 59 taxa belonging to the Eurotiomycetes, Lecanoromycetes, Leotiomycetes, Lichinomycetes and Sordariomycetes, indicate the utility of these loci for fungal phylogenetics at taxonomic levels ranging from genus to class. We also tested the new primers in silico using sequences of Saccharomycotina, Taphrinomycotina and Basidiomycota to predict their potential of amplifying widely across the Fungi. The analyses suggest that the new primers will need no, or only minor sequence modifications to amplify Saccharomycotina, Taphrinomycotina and Basidiomycota.

INTRODUCTION

Molecular systematics has revolutionised our view of fungal evolution. Recent large scale sequencing efforts resulted in comprehensive multi-locus phylogenies, which have significantly improved our understanding of phylogenetic relationships within fungi (Binder & Hibbett 2002, Lumbsch et al. 2004, Lutzoni et al. 2004, James et al. 2006). These data led to the first phylogenetic classification of the Fungi (Hibbett et al. 2007). However, early events in fungal evolution still remain uncertain because of missing support and resolution at the backbone of the phylogeny. We lack information, for example, about the relationships of the different ascomycete classes to one another, or the evolution within major lineages, such as the lichenised Lecanoromycetes, or the basidiomycete clade Agaricomycetes. Robust and well-supported phylogenies are essential for a better understanding of fungal evolution, and a prerequisite for studies aiming at reconstructing the evolution of non-molecular characters on the background of a molecular phylogeny.

Commonly used molecular loci in fungal phylogenetics include nuclear and mitochondrial ribosomal rDNA (18S, 28S, ITS, IGS, mtSSU, mtLSU), as well as protein-coding genes, such as RNA polymerases (RPB1 and RPB2), β-tubulin, γ-actin, ATP synthase (ATP6), and elongation factor EF-1α (TEF1α). Some single-copy protein-coding genes such as RPB1 and RPB2 are promising for yielding well resolved and highly supported phylogenies (Liu & Hall 2004, Reeb et al. 2004, Crespo et al. 2007, Lumbsch et al. 2007). Other protein-coding genes, such as the tubulins, are present in the genome in multiple copies and thus have the potential of being phylogenetically misleading (Landvik et al. 2001). Generally, slow evolving loci are more suitable for reconstruction of deep phylogenetic relationships, while loci with high rates of evolution are better for the reconstruction of more recent evolutionary events. Ribosomal loci with high and heterogeneous rates of change, such as ITS, IGS and mtSSU rDNA, can be used to distinguish taxa at the genus and species level. However, the non-coding regions of these loci are prone to significant length variation, making alignment of distantly related taxa problematic. Fast evolving ribosomal genes are therefore less useful in large scale concatenated analyses involving higher-level phylogenetic relationships. Molecular systematists are constantly searching for loci that are conserved enough to produce reliable alignments, and at the same time have sufficient variability to yield well resolved and well supported phylogenies. Analysing phylogenetic relationships at lower and higher taxonomic levels simultaneously, while using only a few loci, is desirable, because sequencing entire genomes or even multiple loci is not feasible for many phylogenetically interesting taxa. Fungal material suitable for molecular study is often limited, and culturing of many species impossible.

In a recent study Aguileta et al. (2008) used a bioinformatics approach to assess the performance of single-copy protein-coding genes for fungal phylogenetics. Their analyses of 30 published fungal genomes revealed two loci, MS277 and MS456, which outperformed all other single-copy genes in phylogenetic utility. MS277 corresponds to the gene Tsr1, required for rRNA accumulation during biogenesis of the ribosome (Gelperin et al. 2001), while MS456 corresponds to the gene Mcm7, a DNA replication licensing factor required for DNA replication initiation and cell proliferation (Moir et al. 1982, Kearsey & Labib 1998). Alignments based on these two loci alone recovered phylogenies that had the same topology, resolution power, and branch support as phylogenies based on a concatenated analysis of all 135 orthologous single-copy genes identified from fungal genomes (Aguileta et al. 2008). Strikingly, the authors report that most protein-coding genes commonly used in fungal systematics, such as RPB1, RPB2, TEF1α, β-tubulin, and γ-actin are not found among the best performing genes.

In the current study we designed degenerate primers to amplify a 600–800 bp fragment of each, MS277 and MS456, over a wide range of Pezizomycotina. We tested variability and phylogenetic utility of these loci at taxonomic levels ranging from genus to class. Our analyses include in silico comparisons of the new primers to sequences of Saccharomycotina and Basidiomycota to predict primer utility in these phylogenetic groups.

MATERIALS AND METHODS

Material and GenBank sequences used in the current study are listed in Table 1. We designed new degenerate primers based on amino acid alignments of Mcm7 (MS456) and Tsr1 (MS277) of euascomycete sequences available in GenBank. These alignments included members of Dothideomycetes, Eurotiomycetes, Leotiomycetes and Sordariomycetes. Primer sequences and annealing conditions are reported in Table 2 and 3. The locations of the fragments amplified by the new primers are indicated in Fig. 1. We used Aspergillus nidulans mRNA sequences of Mcm7 and Tsr1 as reference sequences (GenBank accession numbers XM_658504 and XM_658778). Saccharomycotina, Taphrinomycotina and Basidiomycota used for in silico analysis of primer fit are listed in Table 4.

Table 1

Material and DNA sequences used in this study.

Species	Order	Class	Source	GenBank accession
				Mcm7 (MS456)	Tsr1 (MS277)
Ajellomyces capsulatus	Onygenales	Eurotiomycetes	–	XM_001538714	XM_001541629
Arctomia delicatula	incertae sedis	Lecanoromycetes	Sweden, 2002, Palice s.n. (F)	GQ272388	GQ272430
Arctomia teretiuscula	incertae sedis	Lecanoromycetes	China (GZU – holotype)	GQ272389	GQ272431
Aspergillus clavatus	Eurotiales	Eurotiomycetes	–	XM_001275314	XM_001275562
Aspergillus fumigatus	Eurotiales	Eurotiomycetes	–	XM_750254	XM_750526
Aspergillus nidulans	Eurotiales	Eurotiomycetes	–	XM_658504	XM_658778
Aspergillus nidulans	Eurotiales	Eurotiomycetes	–	XM_001213626	XM_001208611
Aspergillus niger	Eurotiales	Eurotiomycetes	–	XM_001397760	XM_001399262
Aspergillus oryzae	Eurotiales	Eurotiomycetes	–	XM_001826176	XM_001821764
Aspicilia caesiocinerea	Pertusariales	Lecanoromycetes	USA, Lumbsch 19277e (F)	GQ272390	GQ272432
Aspicilia cinerea	Pertusariales	Lecanoromycetes	USA, Lumbsch 19190c (F)	GQ272391	GQ272433
Botryotinia fuckeliana	Helotiales	Leotiomycetes	–	XM_001556412	XM_001554531
Bulbothrix apophysata	Lecanorales	Lecanoromycetes	Costa Rica, Lücking 16650btu (F)	GQ272392	GQ272434
Cetrariastrum andense	Lecanorales	Lecanoromycetes	Peru, Lumbsch 19334 (MAF)	GQ272429	GQ272471
Cetrariastrum dubitans	Lecanorales	Lecanoromycetes	Peru, Lumbsch 19366 (MAF)	GQ272427	GQ272470
Chaetomium globosum	Sordariales	Sordariomycetes	–	XM_001220296	XM_001225626
Coccidioides immitis	Onygenales	Eurotiomycetes	–	XM_001240385	XM_001245725
Dermatocarpon intestiniforme	Verrucariales	Eurotiomycetes	Turkey, 27.7.1997, John (F)	GQ272393	GQ272435
Dermatocarpon miniatum	Verrucariales	Eurotiomycetes	Germany, 17.10.2001, Zimmermann (F)	GQ272394	GQ272436
Everniastrum lipidiferum	Lecanorales	Lecanoromycetes	Peru, Lumbsch 19309b (MAF)	GQ272395	GQ272437
Everniopsis trulla	Lecanorales	Lecanoromycetes	Peru, Lumbsch 19309b (F)	GQ272396	GQ272438
Flavoparmelia marchantii	Lecanorales	Lecanoromycetes	Australia, (MAF-Lich 10492)	GQ272420	GQ272463
Gibberella zeae	Hypocreales	Sordariomycetes	–	XM_387281	XM_384579
Lecanora allophana	Lecanorales	Lecanoromycetes	Turkey, Lumbsch 19618d (F)	GQ272399	GQ272444
Lecanora carpinea	Lecanorales	Lecanoromycetes	Turkey, Lumbsch 19611m (F)	GQ272400	GQ272443
Lecanora chlarotera	Lecanorales	Lecanoromycetes	Turkey, Lumbsch 19622e (F)	GQ272398	GQ272440
Lecanora margarodes	Lecanorales	Lecanoromycetes	Australia, Lumbsch 19086b (F)	GQ272401	GQ272439
Lecanora pulicaris	Lecanorales	Lecanoromycetes	Turkey, Lumbsch 19627c (F)	GQ272419	GQ272441
Lecanora subcarpinea	Lecanorales	Lecanoromycetes	Turkey, Lumbsch 19622a (F)	GQ272428	GQ272442
Lobothallia radiosa	Pertusariales	Lecanoromycetes	Switzerland, 9.8.2004, Lumbsch (F)	GQ272397	GQ272445
Magnaporthe grisea	Sordariales	Sordariomycetes	–	XM_364455	XM_368157
Malcolmiella psychotrioides	Ostropales	Lecanoromycetes	Costa Rica, Lücking s.n. (F)	GQ272412	GQ272456
Malcolmiella sp. 1	Ostropales	Lecanoromycetes	Thailand, Kalb 37092 (hb. Kalb)	GQ272402	GQ272447
Malcolmiella sp. 2	Ostropales	Lecanoromycetes	Thailand, Kalb 36969 (hb. Kalb)	GQ272411	GQ272455
Malcolmiella sp. 3	Ostropales	Lecanoromycetes	Thailand, Kalb 37093 (hb. Kalb)	GQ272405	GQ272450
Malcolmiella sp. 4	Ostropales	Lecanoromycetes	Thailand, Kalb 36858 (hb. Kalb)	GQ272403	GQ272448
Malcolmiella sp. 5	Ostropales	Lecanoromycetes	Thailand, Kalb 37060 (hb. Kalb)	GQ272407	GQ272446
Malcolmiella sp. 6	Ostropales	Lecanoromycetes	Thailand, Kalb 37072 (hb. Kalb)	GQ272408	GQ272452
Malcolmiella sp. 7	Ostropales	Lecanoromycetes	Thailand, Kalb 36832 (hb. Kalb)	GQ272406	GQ272451
Malcolmiella sp. 8	Ostropales	Lecanoromycetes	Thailand, Kalb 37005 (hb. Kalb)	GQ272409	GQ272453
Malcolmiella sp. 9	Ostropales	Lecanoromycetes	Thailand, Kalb 36963 (hb. Kalb)	GQ272404	GQ272449
Malcolmiella sp. 10	Ostropales	Lecanoromycetes	Thailand, Kalb 37086 (hb. Kalb)	GQ272410	GQ272454
Neosartorya fischeri	Eurotiales	Eurotiomycetes	–	XM_001260497	XM_001260746
Neurospora crassa	Sordariales	Sordariomycetes	–	XM_958785	XM_951859
Ochrolechia parella	Pertusariales	Lecanoromycetes	Turkey, Lumbsch 19625g (MIN)	GQ272421	GQ272464
Ochrolechia subpallescens	Pertusariales	Lecanoromycetes	USA, Lumbsch 19900a & Schmitt (MIN)	GQ272422	GQ272465
Parmeliopsis hyperopta	Lecanorales	Lecanoromycetes	Spain (MAF-Lich 10181)	GQ272426	GQ272468
Peltula euploca	Lichinales	Lichinomycetes	USA, Lumbsch 19923b & Schmitt (MIN)	GQ272424	GQ272467
Penicillium marneffei	Eurotiales	Eurotiomycetes	–	XM_002146315	XM_002148793
Pertusaria amara	Pertusariales	Lecanoromycetes	USA, Lumbsch 19925a & Schmitt (MIN)	GQ272423	GQ272466
Pertusaria velata	Pertusariales	Lecanoromycetes	USA, Lumbsch 19913c & Schmitt (MIN)	GQ272425	GQ272469
Podospora anserina	Sordariales	Sordariomycetes	–	XM_001912857	XM_001909251
Psiloparmelia denotata	Lecanorales	Lecanoromycetes	Peru, Lumbsch 19302g (F)	GQ272413	GQ272457
Pyrenula subpraelucida	Pyrenulales	Eurotiomycetes	Costa Rica, Lücking 17550f (F)	GQ272414	GQ272459
Pyrgillus javanicus	Pyrenulales	Eurotiomycetes	Australia, Lumbsch 19115e (F)	GQ272415	GQ272458
Sclerotinia sclerotiorum	Helotiales	Leotiomycetes	–	XM_001586126	XM_001593622
Umbilicaria leprosa	incertae sedis	Lecanoromycetes	Peru, Lumbsch 19355a (F)	GQ272416	GQ272460
Usnea endochrysaea	Lecanorales	Lecanoromycetes	USA, Buck 51175 (hb. Lendemer)	GQ272417	GQ272461
Verrucaria muralis	Verrucariales	Eurotiomycetes	Czech Republic, Palice 6011 (hb. Palice)	GQ272418	GQ272462

Table 2

Primers developed in the current study.

Primer Name	Direction	Sequence (5′-3′)	Position in A. nidulans mRNA (XM_658504 and XM_658778)	Corresponding amino acid sequence in A. nidulans (AN5992 and AN6266)	Length	Degeneracy
Mcm7-709for	For	ACI MGI GTI TCV GAY GTH AAR CC	709	TRVSDVKP	23 bp	32
Mcm7-1348rev	Rev	GAY TTD GCI ACI CCI GGR TCW CCC AT	1348	MGDPGVAKS	26 bp	16
Mcm7-1447rev	Rev	C ATI ACI GCI GCI GTR AGR CC	1447	GLTAAVM	21 bp	4
Tsr1-1453for	For	GAR TTC CCI GAY GAR ATY GAR CT	1453	EFPDEIEL	23 bp	32
Tsr1-1459for	For	CCI GAY GAR ATY GAR CTI CAY CC	1459	PDEIELHP	23 bp	32
Tsr1-2308rev	Rev	CTT RAA RTA ICC RTG IGT ICC	2308	GTHGYFK	21 bp	8

Table 3

Annealing conditions and PCR success rates for primers used in this study.

Gene	Primer combination	Approximate fragment length	Annealing temp.	PCR success (% of attempts)
Mcm7 (MS456)	Mcm7-709for/Mcm7-1348rev	640 bp	56 °C	80 %
Mcm7 (MS456)	Mcm7-709for/Mcm7-1447rev	740 bp	56 °C	50 %
Tsr1 (MS277)	Tsr1-1459for/Tsr1-2308rev	750 bp	49 °C	40 %
Tsr1 (MS277)	Tsr1-1453for/Tsr1-2308rev	750 bp	49 °C	40 %

Fig. 1

Locations of the new primers for Mcm7 and Tsr1 using Aspergillus nidulans mRNA (XM_658504 and XM_658778) as reference sequence. Shaded areas in Tsr1 indicate regions of high sequence variability.

Table 4

Taxa used to test the fit of the new primers in silico.

Taxon	Mcm7	Tsr1
Saccharomycotina
Ashbya gossypii	NP_984137	NP_984911
Kluyveromyces lactis	XP_454998	XP_454177
Saccharomycetes cerevisiae	NP_009761	NP_010223
Yarrowia lipolytica	XP_501070	XP_500653
Taphrinomycotina
Schizosaccharomyces pombe	NP_596545	NP_593391
Basidiomycota
Coprinopsis cinerea	EAU88865	EAU91047
Cryptococcus neoformans	XP_571487	XP_570891
Ustilago maydis	EAK87259	EAK85759

Molecular procedures

We extracted total genomic DNA from our samples using the Qiagen Plant Mini Kit (Qiagen). PCR reactions (25 μL) contained PuReTaq Ready-To-Go PCR beads (GE Healthcare), 1.25 μL of each primer (10 mM), 19.5 μL H2O, and 3 μL DNA template. Alternatively we used 0.125 μL AmpliTaq Gold Taq (Applied Biosystems), 2.5 μL buffer, 2 μL dNTPs, 2.5–4 μL MgCl (20 mM), 0–5 μL BSA, 1.25 μL of each primer, and 3 μL DNA template. We found that increasing the amount of forward primer Tsr1-1459for to 2.5 μL, as well as adding 2 μL MgCl (20 mM) to PCR reactions involving PCR beads often improved PCR results. PCR cycling conditions for Mcm7-709for/Mcm7-1447rev and Mcm7-709for/Mcm7-1348rev (MS456) were: initial denaturation 94 °C for 10 min, followed by 38 cycles of 94 °C for 45 s, 56 °C for 50 s, 72 °C for 1 min, and final elongation 72 °C for 5 min. PCR cycling conditions for Tsr1-1459for/Tsr1-2308rev (MS277) were the same as above except with 49 °C annealing temperature. Amplification products were stained with EZ-Vision DNA dye (Amresco) and viewed on 1 % low melt agarose gels. We excised bands of the expected length from the gel and purified them using GELase (Epicentre). Alternatively, PCR products were cleaned using the Bioclean Columns kit (Biotools, Madrid) according to the manufacturer’s instructions. We sequenced the fragments using Big Dye v3.1 chemistry (Applied Biosystems) and the same primers as for PCR. Cycle sequencing was executed with the following program: initial denaturation for 1 min at 96 °C followed by 32 cycles of 96 °C for 15 s, 50 °C for 10 s, 60 °C for 4 min. Sequenced products were precipitated with 25 μL of 100 % EtOH mixed with 1 μL of 3 M NaOAC, and 1 μL of EDTA, before they were loaded on an ABI PRISMTM 3730 DNA Analyser (Applied Biosystems). We assembled partial sequences using SeqMan v4.03 (Lasergene) and edited conflicts manually. We aligned the sequences based on amino acid sequence using ClustalW as implemented in the program BioEdit v7.0.9 (Hall 1999) and subsequently translated them back to nucleotides.

Phylogenetic analyses

We assembled two alignments including the same 59 taxa each. For phylogenetic analysis we used a maximum parsimony (MP), maximum likelihood (ML) and a Bayesian approach (B/MCMC) (Larget & Simon 1999, Huelsenbeck et al. 2001). We performed all analyses on the single gene alignments as well as on a combined alignment. We tested for potential conflict between individual datasets by comparing the 75 % MP bootstrap consensus trees.

We used PAUP v4.0 (Swofford 2003), GARLI v0.96 (Zwickl 2006) and MrBayes v3.1.2. (Huelsenbeck & Ronquist 2001) to analyse the alignments. MP analyses included 100 replicates with random sequence additions and TBR branch swapping in effect. MP bootstrapping (Felsenstein 1985) was performed based on 2 000 replicates with the same settings as for the MP search. Likelihood analyses were run using the GTR+I+G model and default settings in GARLI. For Bayesian analyses we partitioned the dataset into three parts (each codon position) and each partition was allowed to have its own parameter values (Nylander et al. 2004). No molecular clock was assumed, and no interpartition rate heterogeneity was allowed. Heating of the chains was set to 0.2. A run with 3 000 000 generations starting with a random tree and employing 4 simultaneous chains was executed for the individual datasets. Every 100th tree was saved into a file. The first 300 000 generations (i.e. the first 3 000 trees) were deleted as the ‘burn in’ of the chain. For the combined alignment dataset we executed a run with 6 000 000 generations and deleted the initial 600 000 generations (i.e. the first 6 000 trees). We plotted the log-likelihood scores of sample points against generation time using TRACER v1.0 (http://tree.bio.ed.ac.uk/software/tracer/) to ensure that stationarity was achieved after the first 300 000 (600 000 for the combined alignment dataset) generations by checking whether the log-likelihood values of the sample points reached a stable equilibrium value (Huelsenbeck & Ronquist 2001). Additionally, we used AWTY (Nylander et al. 2008) to compare splits frequencies in the different runs and to plot cumulative split frequencies to ensure that stationarity was reached. We calculated a majority rule consensus tree with average branch lengths of the remaining 54 000 trees (27 000 from each of the parallel runs) using the sumt option of MrBayes. For the combined alignment dataset the majority rule consensus tree consisted of 108 000 (2 × 54 000) trees from the stationarity phase. Posterior probabilities were obtained for each clade. Clades with posterior probabilities ≥ 0.95 were considered as strongly supported. Phylogenetic trees were visualised using the program Treeview (Page 1996).

RESULTS

We report 84 new sequences of Mcm7 (MS456) and Tsr1 (MS277) for 42 lichenised ascomycetes belonging to the classes Eurotiomycetes, Lecanoromycetes and Lichinomycetes (Table 1). PCR success rates for our newly developed primers were highest for the primer combination Mcm7-709for/Mcm7-1348rev (± 80 %), while Mcm7-709for/Mcm7-1447rev worked in ± 50 % of the attempted PCRs, and the Tsr1 primers in ± 40 %. Multiple bands were sometimes present when we used the primer combinations Mcm7-709for/Mcm7-1447rev and Tsr1-1459for/Tsr1-2308rev. Tsr1-1453for is a modification of Tsr1-1459for that we used under the same annealing conditions. We used the Aspergillus nidulans mRNA sequences of Mcm7 (XM_658504) and Tsr1 (XM_658778) as references for the locations of our primers. The full length genomic DNA sequences of Aspergillus nidulans Mcm7 and Tsr1 contain 1–2 introns of ± 60 bp length, which, however, do not overlap with the sequence fragments amplified by primers developed in this study. We found introns (length: 189–272 bp) with characteristic GT-intron-AG splice sites near the reverse primer (Tsr1-2308rev) in Tsr1 in three Lecanora species. Two hypervariable regions containing many gaps (Tsr1: positions 198–221 and 518–628) were excluded from the phylogenetic analysis. The Mcm7 alignment contained no gaps and no ambiguously aligned regions. Properties of the sequences and alignments are summarized in Table 5. We performed parsimony bootstrap analyses on each individual dataset, and examined 75 % bootstrap consensus trees for conflict (Lutzoni et al. 2004). We used the program Modeltest v3.7 (Posada & Crandall 1998) to determine the nucleotide substitution model that best fit our data. For both datasets the program selected the GTR+I+G model.

Table 5

Mcm7 and Tsr1 sequence and alignment properties.

	Mcm7 (MS456)	Tsr1 (MS277)
Introns	None	some (length: 189–272 bp)
Total alignment length (bp)	573	827
Hypervariable (excluded) sites	None	198
Variable sites	357/573 (62.3 %)	489/629 (77.7 %)
Constant sites	216/573 (37.7 %)	140/629 (22.3 %)
Within-genus sequence variation (p-distances) excluding hypervariable sites:
Malcolmiella (11 OTUs)	0.0055–0.2227	0.0332–0.2193
Aspergillus (7 OTUs)	0.0230–0.2307	0.0357–0.3076
Lecanora (6 OTUs)	0.0377–0.2756	0.0226–0.4148

The tree topologies obtained from the single gene datasets resulting from MP, ML and Bayesian analyses did not show any strongly supported conflicts. Thus, we present only the B/MCMC tree of the combined analysis (Fig. 2). Statistical values and number of supported nodes obtained by MP, ML and Bayesian analyses of single and combined datasets are summarised in Table 6. The Sordariomycetes were used as outgroup. The classes Sordariomycetes, Leotiomycetes, Eurotiomycetes and Lecanoromycetes are monophyletic and highly supported (PP ≥ 95). Lichinomycetes is only represented by a single species, Peltula euploca. The phylogenetic estimate obtained from the combined analysis of Mcm7 and Tsr1 agrees with previously published phylogenies (Gargas et al. 1995, James et al. 2006). Lecanoromycetes form a supported sister group relationship with Eurotiomycetes. Basal to this are Lichinomycetes and Leotiomycetes. Within Lecanoromycetes, the subclasses Lecanoromycetidae and Ostropomycetidae form supported groups, while the genus Umbilicaria is in an unsupported position at the base of Lecanoromycetes. Within Eurotiomycetes, Eurotiomycetidae and Chaetothyriomycetidae form supported clades. We included multiple species/strains of the genera Aspergillus (7), Lecanora (6), and Malcolmiella (11) to assess within-genus variation of the analysed loci, as well as resolution power at low taxonomic levels. Genetic distances within Aspergillus, Lecanora and Malcolmiella are reported in Table 5. Each of these genera forms a supported monophyletic clade with high internal resolution and support (Fig. 2).

Fig. 2

Phylogeny of Pezizomycotina (Ascomycota) based on a combined alignment of Mcm7 (MS456) and Tsr1 (MS277) sequences. Total alignment length is 1203 bp. This is a 50 % majority rule consensus tree based on a sampling of 108 000 B/MCMC trees. Bold branches indicate posterior probabilities ≥ 0.95. Numbers above branches are maximum parsimony bootstrap support values ≥ 70 based on 2 000 random addition replicates.

Table 6

Comparison of phylogenetic analyses (MP, ML, B/MCMC) between single and combined datasets.

	Mcm7 (MS456)	Tsr1 (MS277)	Combined
MP tree length	3537	4606	8200
Number of MP trees	1	12	8
Consistency Index (CI) excluding uninformative sites	0.195	0.216	0.205
# of nodes supported by bootstrap ≥ 70 in MP analyses (based on 2 000 replicates)	23	30	37
ML score using GTR+I+G (GARLI)	−13732	−18424	−32262
# of nodes supported by PP ≥ 95 in B/MCMC analyses	36	38	44

We aligned selected members of Saccharomycotina, Taphrinomycotina and Basidiomycota (Table 4) with our datasets and compared the new primer sequences to the corresponding positions in these taxa. The low number of mismatches suggests that the new primers will need no adjustments or only minor modifications to also fit these phylogenetic groups (Fig. 3).

Fig. 3

Comparison of the new primers to homologous sequences in Saccharomycotina (Ashbya, Kluyveromyces, Saccharomyces, Yarrowia), Taphrinomycotina (Schizosaccharomyces) and Basidiomycota (Coprinopsis, Cryptococcus, Ustilago). 100 % matches between primer sequence and gene sequences studied are indicated by grey shading. High sequence similarities indicate that the new primers are likely to fit in members of the analysed groups. Some primer sequences may require slight modifications.

DISCUSSION

We developed new degenerate primers, which amplify fragments of the single-copy protein-coding genes Mcm7 and Tsr1 in Pezizomycotina. Our study confirms that Mcm7 and Tsr1 are suitable loci for the reconstruction of phylogenetic relationships among fungi (Aguileta et al. 2008). We were able to obtain sequences from representatives of 5 classes and 11 orders of euascomycetes, demonstrating the ability of the primers to amplify a wide range of unrelated taxa. Additionally we tested primer fit in silico using members of Saccharomycotina, Taphrinomycotina and Basidiomycota and found that the new primers can be used for these groups as well, possibly with slight sequence modifications.

Our analyses within Pezizomycotina show that Mcm7 and Tsr1 are able to resolve large scale as well as fine scale phylogenetic relationships. The sequences are alignable across a wide range of unrelated taxa and at the same time have sufficient variability to resolve within-genus relationships (Table 5). This property sets the new loci apart from commonly used ribosomal markers, such as ITS or mtSSU, which also have the power to resolve lower level phylogenetic relationships, but may yield ambiguous and saturated alignments, when used to compare distantly related taxa. We predict that Mcm7 and Tsr1 have an even higher potential to resolve phylogenetic relationships between fungi when analyzed in combination with other routinely used datasets, such as 18S, 28S, RPB1 and RPB2.

Mcm7 and Tsr1 are two relatively long (~ 2.5 kb) single-copy genes which can be aligned across major fungal lineages, such as Ascomycota and Basidiomycota (Aguileta et al. 2008). The fact that Homo sapiens sequences can be used as outgroups (Aguileta et al. 2008, www.systematicbiology.org, online Appendix 5) indicates that these loci might also be useful for phylogenetic studies involving fungi as well as non-fungal organisms.