One of the most important evolutionary transitions is the cooperation of vastly unrelated species (symbiosis). What stabilizes symbiotic associates or causes their collapse (dysbiosis) is a mystery.  Recent advances due to high throughput sequencing techniques indicate that most symbioses (and eukaryotic organisms) are in fact communities of interacting bacteria, fungi in addition to macroscopic hosts and symbionts. Fungus-gardening (‘attine’) ants have had a symbiotic relationship with specific fungal lineages for millions of years. These ants are excellent model systems for exploring the basis of specificity as the host and symbionts are macroscopic and can be easily experimentally manipulated. The most derived attine ants (the ‘higher attines’) typically grow two broad clades of fungi. Most ants in the genus Trachymyrmex tend to grow Clade–B fungi, which are a group of undescribed Leucoagaricus species, whereas the leaf–cutting ants in the genera Acromyrmex and Atta tend to grow Clade–A fungi (Leucoagaricus gongylophorus). However, there are some species of Trachymyrmex that are known to grow Clade A fungi and some leaf-cutting ants that grow Clade B fungi.  Previous work has shown switching the cultivar grown by Trachymyrmex septentrionalis ants, from Clade–B to Clade–A fungus, creates an unstable symbiosis between the ants and their grown cultivar, so that a sudden and catastrophic decrease in the size of their fungal garden invariably results. One hypothesis is that the stability of ant–fungal combinations is maintained by interactions among members of the microbiome of fungus–gardening ants and their fungus gardens. We explored whether changing fungal partners impacts the microbiomes of the host ants and their symbiotic fungus by performing cross fostering experiments that forced ants to grow novel fungi. Specifically, these experiments forced ants of two Trachymyrmex species (T. pomonae and T. septentrionalis) that normally grow Clade–B fungi to grow Clade–A fungi and compared these to a species that is known to growth both Clade-A and Clade-B (T. arizonensis). The experiments revealed that Trachymyrmex ants altered their novel Clade–A garden microbiomes and that these were similar to that of the ‘control or sham switched’ Clade–B fungus gardens. These results suggest that ants play a role in determining the structure of the microbiome of their fungus gardens. Since these combinations are not stable, it is possible that the ‘novel’ microbiome structured by the Trachymyrmex ants is a factor in driving symbiotic collapse. Such findings suggest each fungal clade may have specificity with assorted bacterial communities.

DNA extraction and sequencing of the 16S samples was performed at MR DNA in Shallowater, Texas (http://www.mrdnalab.com/). DNA extractions were conducted using Qiagen DNEasy Powersoil Pro Kits (Qiagen item no. 12888). The sequencing provider conducts regular negative controls. DNA sequences were amplified from whole ants and fungus using primers 27F 5’AGRGTTTGATCMTGGCTCAG and 519R 5’GTNTTACNGCGGCKGCTG that span the V1-V3 hypervariable regions of the 16S rRNA gene (Ishak et al., 2011b). Sequences were amplified using the HotStarTaq Plus Master Mix Kit (Qiagen) under the following conditions: 94°C for 3 minutes, followed by 28 cycles of 94°C for 30 seconds, 53°C for 40 seconds and 72°C for 1 minute, after which a final elongation step at 72°C for 5 minutes was performed. After the samples were amplified and checked for adequate genetic yields, the sub-samples were pooled back together and purified using calibrated Ampure XP beads. The purified and pooled PCR product was used to create a DNA library and sequenced using the Illumina MiSeq platform in PEx300 mode.

Microbiome sequences from MR DNA were processed using Qiime2–2020.6 (Qiime2) (Bolyen et al., 2019). To input the sequences into Qiime2, sequences were first processed utilizing the MR DNA Fastq Processor (http://www.mrdnalab.com/mrdnafreesoftware/fastq–processor.html) in order for the sequences to be converted into a format (Earth Microbiome Project) that Qiime2 accepts. The demux plugin (https://github.com/qiime2/q2–demux) was used to demultiplex the inputted sequences and the DADA2 plugin was used to denoise the data (Callahan et al., 2016). The demux package and DADA2 were utilized as they have been consistently used for ant and insect microbiome analyses (Deguenon et al., 2019; Flynn et al., 2021; Ramalho et al., 2020). When using the DADA2 plugin, sequences associated with T. arizonensis and T. pomonae were truncated down to 280 base pairs, while sequences associated with T. septentrionalis were truncated down to 260 base pairs, as the average quality score dipped below 20 beyond these points. Taxonomy classification with 99% operational taxonomic unit (OTU) similarity was performed utilizing the SILVA 138_QIIME database (Quast et al., 2013; Yilmaz et al., 2014). To do this, a taxonomic classifier was created using the “feature–classifier fit–classifier–naive–bayes” command and the SILVA database (Bokulich et al., 2018). This classifier was used to assign sequences a taxonomic classification using the “feature–classifier” plugin (Bokulich et al., 2018) with the “classify–sklearn” command. An OTU table for each species was created by inputting a tabulated taxonomic bar plot, created using the “taxa barplot” command, of the sequences into Qiime2 View (https://view.qiime2.org). The utilized Qiime2 processing pipeline has been deposited on GitHub (https://github.com/bsbringhurst/Consequences–of–Horizontal–Fungus–Exchange). Sequences that were classified as chloroplasts, mitochondria, or unassigned were manually removed from the OTU tables. Attached are these OTU tables, along with the sample associated metadata, in .csv format.