Nematophagous fungus Arthrobotrys oligospora mimics olfactory cues of sex and food to lure its nematode prey

Abstract

To study the molecular basis for predator-prey coevolution, we investigated how Caenorhabditis elegans responds to the predatory fungus Arthrobotrys oligospora. C. elegans and other nematodes were attracted to volatile compounds produced by A. oligospora. Gas-chromatographic mass-spectral analyses of A. oligospora-derived volatile metabolites identified several odors mimicking food cues attractive to nematodes. One compound, methyl 3-methyl-2-butenoate (MMB) additionally triggered strong sex- and stage-specific attraction in several Caenorhabditis species. Furthermore, when MMB is present, it interferes with nematode mating, suggesting that MMB might mimic sex pheromone in Caenorhabditis species. Forward genetic screening suggests that multiple receptors are involved in sensing MMB. Response to fungal odors involves the olfactory neuron AWCs. Single-cell RNA-seq revealed the GPCRs expressed in AWC. We propose that A. oligospora likely evolved the means to use olfactory mimicry to attract its nematode prey through the olfactory neurons in C. elegans and related species.

Keywords: AWC; Arthrobotrys oligospora; C. elegans; attraction; evolutionary biology; genomics; neuroscience; olfactory mimicry.

Figure 1.. A. oligospora attracts many nematode species, including C. elegans.

(A) Schematic representation of the 4-point chemotaxis assays and the formulation used to calculate the chemotaxis index (CI). (B) C. elegans adults are attracted to A. oligospora, but the dauers and the L1 larvae have much weaker attraction (Mean ± SD, n = 4–12 trials). (C) Speed of wild-type C. elegans on A. oligospora culture or medium only control. Locomotion of the animals was tracked by the commercial WormLab system (MBF bioscience). (D) Many Caenorhabditis nematodes and the more distant Panagrellus redivivus, Pristionchus pacificus, and Bursaphelenchus xylophilus were attracted to A. oligospora (Mean ± SD, n = 4–12 trials). (E) SEM image of C. elegans trapped in A. oligospora. Scale bar = 10 μm. DOI: http://dx.doi.org/10.7554/eLife.20023.002

Figure 2.. C. elegans are attracted to volatile compounds produced by A. oligospora.

(A) Schematic representation of the volatile chemotaxis assays in which A. oligospora were grown in compartmentalized Petri dishes that contained three sections. The CI was calculated based on the formula presented. (B) C. elegans adults preferentially migrate to the section that was inoculated with A. oligospora (Mean ± SD, n = 6–7 trials). (C) A. oligospora attraction is reduced in several mutants defective in AWC neuron function (Mean ± SD, n = 3–4 trials). (D) Laser ablation of the AWC neurons reduced attraction to A. oligospora. DOI: http://dx.doi.org/10.7554/eLife.20023.003

Figure 3.. GC x GC-TOFMS analyses identified A. oligospora-derived odorants that are attractive to C. elegans.

(A) Representative two-dimensional gas chromatogram obtained from SPME sampling of headspace above A. oligospora cultures. The total detector response is indicated by the intensity of red in the chromatogram. Black dots represent discrete elution peaks identified during data processing (shown for a signal-to-noise threshold of 100). The elution peaks labeled 1–5 represent compounds that were detected only in A. oligospora cultures, and not in the medium-only controls. (B) Chemotaxis assays of A. oligospora-derived odorants. Chemicals were tested at a 10⁻² dilution and truffle oil was tested undiluted (Mean ± SD, n = 6 trials). (C) N2 adults were strongly attracted to MMB across a wide-range of concentrations (Mean ± SD, n = 6 trials). (D) C. elegans prefers MMB over isoamyl alcohol (IAA; Mean ± SD, n = 3–7 trials). The MMB preference index was calculated according to the formula in Figure 1. (E) Chemotaxis plot showing C. elegans chemotactic response to different odorants after 2 hr adaptation to A. oligospora culture (red) or the medium-only control (blue; Mean ± SD, n = 4–10 trials). DOI: http://dx.doi.org/10.7554/eLife.20023.004

Figure 4.. AWC^on neurons respond to A. oligospora-derived odorants.

(A) GCaMP6 measurements of the AWC^on neurons in response to A. oligospora-derived odorants (10⁻³) and buffer control (n = 4–7 trials for each odorant tested). (B) GCaMP6 measurements of the AWC^on neurons in response to MMB in the unc-13 mutant background (n = 11 and 14 for MMB and buffer control respectively). (C) The AWC- mutant is defective in MMB attraction. Chemotaxis response to 10⁻⁴ MMB was measured for the wild-type (N2) or AWC- mutant (ceh-36). (D) The AWC- mutant is more likely to survive predation by A. oligospora. Y axis indicates % of nematodes that were still alive after 16 hr incubation with A. oligospora culture (Mean ± SD, n = 10–12 trials). DOI: http://dx.doi.org/10.7554/eLife.20023.005

Figure 5.. MMB triggers a sex- and lifestage-specific attraction in Caenorhabditis species and interferes with mating in C. afra.

(A) Responses of C. elegans hermaphrodites, males, L1 larvae and dauers to a 10⁻⁴ dilution of MMB (Mean ± SD, n = 4–10 trials). (B) Responses of Caenorhabditis sp. and P. pacificus to a 10⁻⁴ dilution of MMB (Mean ± SD, n = 4–10 trials). (C) Mating behavior of C. afra was recorded. Y axis indicates the time required for the nematodes to mate; if no mating was observed within the 30 min tracking period, data points were represented as 30 min. (D) Chemotaxis of C. afra male and female nematodes in response to 10⁻² diacetyl. DOI: http://dx.doi.org/10.7554/eLife.20023.006

Figure 6.. Forward genetic screens isolated two mutants with strong defects in MMB chemotaxis.

(A) Response of the two mutant lines to MMB, IAA, and Benzaldehyde (BA; n = 4–8 trials for each odorant tested). The mutants showed strong defects in MMB chemotaxis but remained attracted to two other odorants, IAA and BA. (B) SNP mapping with the Hawaiian polymorphic strain CB4856 revealed that the mutations in both mutants were located on chromosome X. Mutants #7–22 and #18–10 were crossed to CB4856 and three F2 progeny from each mutant line that had MMB chemotaxis defects were analyzed for eight SNP markers on each chromosome. (C) Mutants #7–22 and #18–10 failed to complement. Responses of the F1 progeny from a cross between the two mutants to 10⁻⁴ dilution of MMB. (Mean ± SD, n = 5–8 trials). DOI: http://dx.doi.org/10.7554/eLife.20023.007

Figure 6—figure supplement 1.. Gene structure of odr-7.

Asterisks indicate two nonsense mutations caused by EMS mutagenesis. DOI: http://dx.doi.org/10.7554/eLife.20023.008

Figure 6—figure supplement 2.. Fosmid clone WRM0639bF05 which contains odr-7 rescues the MMB attraction defect of odr-7 (sy837).

Responses of the mutant or the rescue line to 10⁻⁴ dilution of MMB. (Mean ± SD, n = 8–16 trials). DOI: http://dx.doi.org/10.7554/eLife.20023.009

Figure 7.. Candidate GPCRs expressed in AWC neurons.

Gene expression in AWC neurons. Expression data are shown for 5326 C. elegans genes (5294 protein-coding and 32 non-coding RNAs) that exhibited above-background expression in our pooled AWC RNA-seq (i.e., that had a minimum expression level, in a 99% credibility interval, of ≥0.1 TPM). The x-axis shows AWC-specificity, computed as the ratio of AWC gene expression (in TPM) to larval gene expression (in TPM) for each gene. The y-axis shows the absolute magnitude of AWC gene activity in TPM. Among most genes (labeled ‘AWC’ and shown as gray dots) are highlighted 32 genes encoding G-protein coupled receptors shown as green triangles and 984 genes encoding housekeeping functions shown as red dots. The housekeeping genes were identified in a previous single-cell RNA-seq analysis (Schwarz et al., 2012). Full identities, annotations, and expression data for all C. elegans genes are provided in Figure 7—source data 1. DOI: http://dx.doi.org/10.7554/eLife.20023.010

Figure 8.. A. oligospora produce volatiles that mimic food and sex cues to attract C. elegans via its olfactory neuron AWCs.

DOI: http://dx.doi.org/10.7554/eLife.20023.015