Draft genome of the fungus-growing termite pathogenic fungus Ophiocordyceps bispora (Ophiocordycipitaceae, Hypocreales, Ascomycota)

This article documents the public availability of genome sequence data and assembled contigs representing the partial draft genome of Ophiocordyceps bispora. As one of the few known pathogens of fungus-farming termites, a draft genome of O. bispora represents the opportunity to further the understanding of disease and resistance in these complex termite societies. With the ongoing attempts to resolve the taxonomy of the Hypocralaean family, more genetic data will also help to shed light on the phylogenetic relationship between sexual and asexual life stages. Next generation sequence data is available from the European Nucleotide Archive (ENA) under accession PRJEB13655; run numbers: ERR1368522, ERR1368523, and ERR1368524. Genome assembly available from ENA under accession numbers: FKNF01000001–FKNF01000302. Gene prediction available as protein fasta, nucleotide fasta and GFF file from Mendeley Data with accession doi:10.17632/r99fd6g3s4.2 (http://dx.doi.org/10.17632/r99fd6g3s4.2).


Value of the data
Ophiocordyceps bispora is the first pathogen of fungus-farming termites (Macrotermitinae) to have its genome sequenced.
O. bispora represents one of the few known parasites of fungus-farming social insects. The O. bispora draft genome assembly will be of value for future comparative and phylogenetic analyses.
Further study of O. bispora's relationship with Hirsutella thompsonii could potentially identify a teleomorph-anamorph connection bringing the two species in line with the 2011 reccommendations for The International Code of Nomenclature for algae, fungi and plants [2].

Data
We present a partial draft genome assembly with gene prediction of the fungus-farming termite pathogenic fungus Ophiocordyceps bispora (Ophiocordycipitaceae, Hypocreales). O. bispora infects the reproductive alate caste of several termite genera, including fungus-growing termites in the genus Macrotermes in Africa [3][4][5][6][7]. It is one of the few known pathogens of fungus-farming social insects and its Macrotermes host apparently has no other known diseases [8,9]. We also provide a two-gene phylogenetic analysis ( Fig. 1) showing that O. bispora is closely related to an asexual form of Ophiocordyceps: Hirsutella thompsonii. The stark difference between sexual and asexual life stages in the Hypocrales family imply the different life stages of many species are classified twice [10]. Since 2011, the International Code of Nomenclature for algae, fungi and plants has been changed so that one fungus can have only one name [2]. Efforts to reconcile the taxonomy of the Hypocreales with these changes are ongoing [7,11,12], and the draft genome of O. bispora presented here may prove valuable in this effort.

Library construction protocol
Total fungal DNA was extracted from the fruiting bodies protruding from the termites by crushing and dissolving the sample in extraction buffer (10 mM Tris, 10 mM sodium chloride (NaCl), 5 mM calcium chloride (CaCl), 2.5 mM ethylenediaminetetraacetic acid (EDTA), 1% sodiumdodecyl sulfate (SDS), 10% proteinase K and 6.17 mg/mL dithiothreitol (DTT)) overnight at 56°C, followed by organic extraction of the DNA using one volume of chloroform. The DNA-containing supernatant was purified by mixing with 10 Â volumes of modified PB buffer [13] and using a MinElute spin column (Qiagen) as demonstrated elsewhere [14]. The DNA was eluted in 60 μL EB buffer after a ten-minute incubation of the spin column with buffer at 37°C and subsequently fragmented using a Diagenode Bioruptor with a program of 30 s on, 90 s off for 6 cycles. Illumina compatible shotgun sequencing library was generated using reagents from New England Biolabs kit for the 454, E#6070 L as described elsewhere [15]. The finished library was indexed and amplified for sequencing using 10 μL template and a mastermix consisting of 1 Â AmpliTaq Gold buffer (Invitrogen), 2.5 mM MgCl 2 (Invitrogen), 0.8 μg/μL bovine serum albumin (BSA), 0.25 mM dNTP (Thermo Fischer Scientific, 25 mM stock), 0.2 μM forward and reverse indexed primer (10 μM stock), 0.2 U/μL AmpliTaq Gold enzyme (Invitrogen) and molecular grade water to a final reaction volume of 100 μL. The library was amplified in an Applied Biosystems 2720 Thermal Cycler, using the following conditions: 95°C for 1 min, followed by a number of cycles of 95°C for 30 sec, 60°C for 30 s and 72°C for 1 min for 5 cycles and finished by 7 min at 72°C. Finally, the library was purified using a QiaQuick spin column (Qiagen), following manufacturer's instructions. The library was pooled with other samples and sequenced on 10% of a lane on an Illumina Hiseq 2500 instrument at The Danish National High-Throughput DNA Sequencing Centre, Copenhagen, Denmark.

Processing
Pipeline: DNA of the insect pathogenic fungus O. bispora was shotgun sequenced using Illumina HiSeq. Raw sequencing reads are available as three fastq files with accession nos. ERR1368522, ERR1368523, and ERR1368524.
Run data file type: Fastq File Names (Runs: ERR1368522, ERR1368523, and ERR1368524, respectively): ○ TOG_QEHU_Term_fung_03_15_ACATAC_L004_R1_001.fastq.gz ○ TOG_QEHU_Term_fung_03_15_ACATAC_L004_R1_002.fastq.gz ○ TOG_QEHU_Term_fung_03_15_ACATAC_L004_R1_003.fastq.gz Table 1 The fungal strains included in the phylogeny including known host taxa and GenBank accession numbers for nrLSU and nrSSU sequences. The Geneious ver. 8.0.5. read-mapping and assembly tool, which also takes read quality into account when mapping, were used to assemble the draft O. bispora genome. The genome of the closely related Hirsutella thompsonii MTCC 6686 [1] was used as reference genome. The final assembled O. bispora draft genome consisted of 302 contigs and was 6,359,382 bp long. The maximum contig length, with a mean coverage of 3.1, was 158,451 bp with an N50 of 41,819 bp and a GC of 58.5%. Assembled contigs are available with accession nos. FKNF01000001-FKNF01000302. Protein coding sequences were predicted on both strands in the O. bispora draft genome using Augustus ver. 3.2.1 [16] using default parameters and found 3324 Open Reading Frames (ORFs).
The common fungal barcoding genes nuclear-ribosomal Large Sub-Unit (nrLSU) and nuclear ribosomal Small Sub-Unit (nrSSU) were located in the draft O. bispora genome assembly using blastn searches with Geneious 4.8.5. The identified O. bispora nrLSU and nrSSU sequences were combined with other Hypocrealean sequences from GenBank (Table 1), and a concatenated alignment of nrSSU and nrLSU was produced using MUSCLE [17]. A Bayesian phylogenetic analysis was performed using Topali v2 (Runs: 2; Generations: 1,500,000; Burn in: 50%) [18].