Identification of species in fungi has always been challenging due to their cryptic nature. Furthermore, universally accepted DNA barcode markers are lacking, with the ribosomal internal transcribed spacer region (ITS) showing the highest probability of successful identification of the broadest range of fungi (Schoch et al., 2012). In orchid mycorrhizae, the ITS as well as the mitochondrial large subunit (LSU) are most commonly used. However, dependence on only two sequence loci for accurate species identification lacks discrimination, especially for fine-scale ecological and evolutionary interaction studies between orchids and their mycorrhizal symbionts.

Tulasnella J. Schröt, (family Tulasnellaceae, order Cantharellales) is an endophytic fungus occurring in the roots, stems, or protocorns of a range of orchids, trees, and liverworts. In Australia, sexually deceptive orchids within the tribe Drakaeinae such as Chiloglottis R. Br. and Drakaea Lindl. are known to form mycorrhizal associations with narrow groups of monophyletic Tulasnella lineages (Roche et al., 2010; Phillips et al., 2011). However, the actual number of Tulasnella species associated with the orchid genera has not been fully resolved. Furthermore, population-level studies are rare in orchid—mycorrhizal associations because suitable population-level markers often are unavailable. Consequently, to facilitate investigations into evolutionary interactions among Tulasnella species and their orchid hosts, markers are needed at both the species and population levels.

METHODS AND RESULTS

Fungal isolation and DNA extractions were conducted for Tulasnella mycorrhizal fungi as described in Roche et al. (2010) from a range of host species within the genera Arthrochilus F. Muell., Chiloglottis, Drakaea, and Paracaleana Blaxell (Appendix 1).

Phylogenetic loci development —Tulasnella isolates from Chiloglottis aff. jeanesii (an undescribed taxon most closely related to C. jeanesii D. L. Jones) and D. elastica Lindl. were grown in liquid culture and DNA extracted as described previously (Roche et al., 2010).

Sequences for each isolate were generated using a 3-kb pair-end sequence library (Roche, 454 Life Sciences, Branford, Connecticut, USA) on a GS FLX 454 platform using GS XL70 sequencing chemistry. The sequences for each Tulasnella isolate were separately assembled using CLC Genomics Workbench software (CLC bio, Aarhus, Denmark) with the software's standard assembly parameters.

To design phylogenetic markers, we performed a de novo assembly within CLC with both Tulasnella isolates. This provided a consensus sequence to target high-homology sequence regions shared by both Tulasnella isolates. Some 3300 reads were shared out of approximately 20 000 reads per species. To maximize sequence length only contigs of both species >200 bp (418 in total) were investigated further. To increase amplification success across the range of Tulasnella-orchid host species, GenBank BLAST searches ( http://www.ncbi.nlm.nih.gov/) were conducted on the 418 contigs to locate other related sequences. The top BLAST hits to annotated or predicted genes from Basidomycota and Ascomycota fungi were downloaded and included in alignments within the program Geneious version 5.5.6 (Drummond et al., 2012) for 83 consensus sequences. Of these, we selected 30 consensus sequences to design primers; selection was based on product length, ease of primer design, and gene identity. Primers were designed using Primer3 (Rozen and Skaletsky, 2000). Twenty-one (70%) of the 30 primer pairs amplified in both target species. Consistent and high quality cross-genus amplification occurred for seven primer sets in Tulasnella from Arthrochilus, Chiloglottis, Drakaea, and Paracaleana (Table 1).

TABLE 1.

Characteristics of phylogenetic primers for Tulasnella isolates in this study.

PCR reactions were performed in 30-µL reactions containing 1× PCR buffer (1.5 mM MgCl₂, 50 mM KCl, 10 mM Tris-HCl pH 8.3 final concentration; QIAGEN Gmbh, Hilden, Germany), 0.2 µM dNTPs, 5 µg bovine serum albumin (BSA) (New England Biolabs, Ipswich, Massachusetts, USA), 200 nM of each primer, 1 U of Taq polymerase (QIAGEN), and 20–100 ng of template DNA. A touchdown thermal profile was used for initial testing consisting of a 3-min denaturation at 94°C; followed by 12 touchdown cycles at 94°C (30 s), 66°C (40 s) (−3°C/second cycle), 72°C (1 min); 30 cycles at 94°C (30 s), 48°C (40 s), 72°C (1 min); and final extension at 72°C for 20 min. One locus performed better at a fixed annealing temperature of 55°C (Table 1). Products were sequenced bidirectionally with an ABI PRISM BigDye Terminator version 3.1 sequencing kit (Applied Biosystems, Carlsbad, California, USA) on an ABI 3100 automated sequencer. Sequences were edited using the program Sequencher version 4.7 (GeneCodes, Ann Arbor, Michigan, USA), aligned in Geneious, and estimates of variability were performed with MEGA5 (Tamura et al., 2011). The seven markers showed between 24% and 35% nucleotide diversity and 15% and 29% parsimony informative sites (Appendix 2). Amplification was also achieved for some loci in Tulasnella from other orchid genera including Cryptostylis R. Br., Corybas Salisb., and Diuris Sm., as well as for Sebacina Tul. & C. Tul. from Caladenia R. Br. and Glossodia R. Br., and Ceratobasidium D. P. Rogers from Rhizanthella R. S. Rogers (Appendix 2).

TABLE 2.

Characteristics of microsatellite primers for Tulasnella isolates in this study.

Microsatellite primer design —The 454 reads of Tulasnella from C. aff. jeanesii were screened for di-, tri-, tetra-, penta-, and hexanuclotide repeats using the online software MSATFINDER ( http://www.genomics.ceh.ac.uk/msatfinder/). A total of 800 contigs with simple sequence repeats (SSR) were detected. Using the criterion of at least eight repeat units in the sequence, we designed 24 primer pairs using Primer3 (Rozen and Skaletsky, 2000). Two additional SSR loci were obtained via a genomic library enriched for CAG repeats with target clones identified by PCR using published methods (Adcock et al., 2005).

Screening of the loci was performed on several Tulasnella individuals from eight Chiloglottis species sourced from one or two geographic locations per species (Appendix 3). Forward primers had universal M13 tails added as per the method of Schuelke (2000). PCRs were performed in 30-µL reactions containing 1× PCR buffer (1.5 mM MgCl₂, 50 mM KCl, 10 mM Tris-HCl pH 8.3 final concentration; QIAGEN), 0.2 µM dNTPs, 5 µg BSA (New England Biolabs), 200 nM of nonlabeled primer, 50 nM of M13-labeled primer, 100 nM of 21M13 primer (labeled with FAM, NED, or VIC; Applied Biosystems), and 20–100 ng of template DNA. PCR was performed using the touchdown thermal profile described earlier. Some loci required further optimization of annealing temperatures (Table 2). Fragment analysis was performed on an ABI 3100 sequencer with amplified products mixed with a 500 LIZ (Applied Biosystems) size ladder, and genotyping determined using GeneMapper version 3.7 software (Applied Biosystems). Eleven loci (two loci from the genomic library) (Table 2) amplified reliably and were polymorphic for Tulasnella in all eight Chiloglottis host species. Allelic diversity and genotype analyses were performed using GenAlEx version 6.5 (Peakall and Smouse, 2012). All loci resulted in two alleles per locus, consistent with a dikaryotic haploid nature of related genera. Twenty-four genotypes were found among 42 Tulasnella isolates assayed. Genotypes were not shared among Tulasnella isolates from eight Chiloglottis species, or between sites within a host species (Appendix 4).

CONCLUSIONS

We successfully designed polymorphic coding and noncoding markers for Tulasnella mycorrhizal fungi from numerous species within four genera of Australian orchids. Some loci are also useful at higher taxonomic levels because they amplify and provide useful sequences for Sebacina and/or Ceratobasidium. We found that the microsatellite markers are sufficiently polymorphic to investigate species and population-level diversity of Tulasnella from Chiloglottis hosts. These newly developed polymorphic markers will be useful to investigate diversity, phylogenetic relationships, and specificity of the mycorrhizal—orchid associations.