Two +ssRNA mycoviruses cohabiting the fungal cultivar of leafcutter ants

Leafcutter ants are dominant herbivores in the Neotropics and rely on a fungus (Leucoagaricus gongylophorus) to transform freshly gathered leaves into a source of nourishment rather than consuming the vegetation directly. Here we report two virus-like particles that were isolated from L. gongylophorus and observed using transmission electron microscopy. RNA sequencing identified two +ssRNA mycovirus strains, Leucoagaricus gongylophorus tymo-like virus 1 (LgTlV1) and Leucoagaricus gongylophorus magoulivirus 1 (LgMV1). Genome annotation of LgTlV1 (7401 nt) showed conserved domains for methyltransferase, endopeptidase, viral RNA helicase, and RNA-dependent RNA polymerase (RdRp). The smaller genome of LgMV1 (2636 nt) contains one open reading frame encoding an RdRp. While we hypothesize these mycoviruses function as symbionts in leafcutter farming systems, further study will be needed to test whether they are mutualists, commensals, or parasites. Supplementary Information The online version contains supplementary material available at 10.1186/s12985-024-02465-0.


Culture condition
Leafcutter ant colonies were initially collected in Gamboa, Panama, and kept in climatecontrolled rooms (25°C, 75% humidity, minimal daylight) at the University of Copenhagen.
Staphylae were isolated from fungus gardens using flame-sterilized acupuncture needles and grown on Petri dishes (90 mm) containing 20-mL potato dextrose agar.From these plates, potato dextrose broth was inoculated and kept on a rotating shaker for at least 21 days at 25°C, minimal daylight.Fungi from axenic in vitro culture were used for further analysis.

Virus particle extraction
Freeze-dried fungal tissue was ground in heat-sterilized sand and 10-mL binding buffer (20mM Tris-HCl pH 7.4, 1mM CaCl2, 1mM MnCl2, 0.5M NaCl) using a mortar and pestle.The mixture was centrifuged for ten min at 16.000 g and supernatant was then filtered through syringe filters connected in this size order: 5 µm, 2.7 µm, and 0.2 µm.Concanavalin A Sepharose TM 4B (Con A Sep; Cytiva) was prepared following manufacturer's protocol and incubated with 20 mL of sample filtrate in a tumbling shaker for one hour at room temperature.
After incubation, the tubes were placed vertically for ten min during which time the Con A Sep settled.The supernatant was then discarded, and the sample was mixed with 10 mL wash buffer (binding buffer, 20mM α-D-Mannopyranoside) for two min.Con A Sep was then left to settle, and supernatant was discarded.This sample was mixed with 4 mL elution buffer (20mM Tris-HCl pH 7.4, 0.5M NaCl, 0.8M α-D-Mannopyranoside) for 30 min in a tumbling shaker.The resulting eluate was collected by filtration through a syringe barrel using a 5 µm filter.Eluates were dialyzed in distilled water through D-TubeTM Dialyzer Maxi, MWCO 12-14 kDa (Novagen®) overnight and the water changed after the first two hours.

PCR primers and conditions
Although we attempted designing multiple primer pairs using NCBI's primer design tool (https://www.ncbi.nlm.nih.gov/tools/primer-blast/)covering several regions of the sequence, none of these worked for the Tymovirales sequence.
For the Botourmiaviridae, we first performed cDNA synthesis using iScript™ cDNA Synthesis kit (BIO-RAD) following the manufacturer's protocol.20 μl reactions were set up by adding 4 μl 5x iScript Reaction Mix, 1 μl iScript Reverse Transcriptase, 11 μl Nuclease-free water, and 4 μl RNA template (a total of ~15.5 ng RNA) and incubated for 5 min at 25°C, 20 min at 46°C, and 1 min at 95°C.Next, we amplified the selected region using a specific primer pair designed and optimized to amplify a 164-nt region between the nt positions 1844 and 2009 (forward 5'-CACTTGGCGTGTGTTGGAAG-3'; reverse 5'-AGAGGTCCCATTTTGCCTCC-3').To amplify the cDNA, we used a Taq DNA polymerase RED (Ampliqon) following the manufacturer instructions with an annealing temperature of 62°C.PCR products were cleaned with PureIT ExoZAP (Ampliqon, Denmark) following the manufacturer's instructions, and Sanger sequenced with the same primer pair at Eurofins (Germany).
The sequencing results validated the presence of the LgMv1 in the isolate Ac2012-1 of L. gongylophorus.

Identification of mycovirus sequences
A custom database was created from the NCBI viral genome database by retrieving RefSeq protein sequences from viral genomes with algae, plants, and fungal hosts.Assembled nucleotide and amino acid sequences from the Ac2012-1 virome were analyzed using blastx and blastp, respectively, with both NCBI Blast v.2.12.0+ [1] and Diamond v.2.0.13 [2] against the custom database.We used an e-value cutoff of 1e-3.From the blast results, the top five hits with the highest similarity to the query sequence were saved for examination of mycoviruses.
The longest assembled nucleotide isoform was then chosen as a representative for each of the retrieved blast results, and subjected to NCBI blastx algorithms against all organisms, to ensure the most significant alignment.The longest ORF was identified by subjecting the longest nucleotide isoform to NCBI ORF finder (ORFfinder Viewer -NCBI (nih.gov)).

Visualization of read depth in LgMv1 and LgTlv1
We used the Integrative Genome Viewer (IGV) v.2.15.4 to examine the BAM files containing reads mapped to the viral sequences (Fig. S1).This allowed us to visualize the coverage and distribution of the reads across the viral genomes.

Leucoagaricus gongylophorus tymo-like virus 1 comparison
In a recent study, Jo et al. [3] identified multiple mycoviruses sequences from previously RNAseq dataset publicly available in NCBI GenBank.Within these sequences, three tymolike virus sequences were identified in the transcriptome shotgun assembly (TSA) from a L. gongylophorus (isolate Ae322) assembled by De Fine Licht et al. [4].The RNA was sequenced using polyA enriched libraries and Illumina and 454 technologies.
During the transcriptomic assembly process, it is common to generate truncated or chimeric transcripts treated by the program as isoforms.This error could be responsible for the assembly of the three sequences in De Fine Licht et al. study.
When we used blastn to align the nucleotide sequences of our tymo-like virus sequence to the three "Leucoagaricus tymovirus" identified by Jo et al. (Fig. S2, Table S2, Table S3).We observed complete overlap of the three sequences to our tymo-like sequence.Additionally, the three sequences identified by Jo et al. [3] also overlap with each other in more than 1500 nucleotides (Fig. S1).We believe that the error-prone sequencing technologies, the assembly process, and virus polymorphism, might have generated several artifact sequences (referred by the assemblers as isoforms) leading to the identification of multiple similar virus sequences namely Leucoagaricus tymovirus A isolate Won, Leucoagaricus tymovirus B isolate Won, Leucoagaricus tymovirus B isolate Cho.When aligning the four sequences (Fig. S3), we found a stop codon in the sequence of isolate "Cho" which is not present in the other sequences.We believe this is a sequencing error, that is also supported by the presence of several N's in such sequence.This suggests a low-quality sequencing that might have further introduced erroneous nucleotides likely generating false mismatches, and subsequently multiple 'isoforms'.S3.Pairwise percentage identity comparison between the three "Leucoagaricus tymovirus" from Jo et al. [3] at nucleotide and amino acid level.

Figure S1 :
Figure S1: IGV view of read mapping coverage of the sequences generated in this study, showing LgMv1 (top) and LgTlv1 (bottom).The coverage tracks display the depth and distribution of reads mapped to each viral genome, illustrating the variation in read density across different regions of the genomes.Polymorphic sites are indicated by colored bars within the reads, highlighting the genetic variability present in the viral sequences.

Figure S3 :
Figure S3: AliView screenshot showing a segment of the tymo-like sequences aligned.The colors represent the amino acid encoded by each codon plus stop-codon (black).

Table S1 .
Detailed information about the colony used in this study.