orco Mutagenesis Causes Loss of Antennal Lobe Glomeruli and Impaired Social Behavior in Ants

Abstract

Life inside ant colonies is orchestrated with diverse pheromones, but it is not clear how ants perceive these social signals. It has been proposed that pheromone perception in ants evolved via expansions in the numbers of odorant receptors (ORs) and antennal lobe glomeruli. Here, we generate the first mutant lines in the clonal raider ant, Ooceraea biroi, by disrupting orco, a gene required for the function of all ORs. We find that orco mutants exhibit severe deficiencies in social behavior and fitness, suggesting they are unable to perceive pheromones. Surprisingly, unlike in Drosophila melanogaster, orco mutant ants also lack most of the ∼500 antennal lobe glomeruli found in wild-type ants. These results illustrate that ORs are essential for ant social organization and raise the possibility that, similar to mammals, receptor function is required for the development and/or maintenance of the highly complex olfactory processing areas in the ant brain. VIDEO ABSTRACT.

Keywords: CRISPR; Formicidae; Ooceraea biroi; chemical communication; evo-devo; mutagenesis; odorant receptors; pheromones.

Figure 1. Number of OR Genes and orco Mutagenesis

(A) Phylogeny with numbers of ORs for ants (green) and other insects (black), showing ant OR expansion (Table S1). (B) Position of predicted CRISPR/Cas9 cut site in Orco protein model (red circle). Frameshift mutations at this position truncate the wild-type protein between the third and fourth transmembrane domains, and the resultant mutant protein is unlikely to form functional ion channels. (C) Proportion of Illumina sequencing reads of orco amplicons with insertions or deletions (indels) relative to gRNA sequence in G0s, showing mutation rates of at least 97% in some individuals. Red circle indicates predicted CRISPR/Cas9 cut site. Protospacer adjacent motif (PAM) in bold. (D) Wild-type orco sequence compared to sequences for the two orco^wt/− and the five orco^−/− mutant lines. Deletions are shown in red and insertions in green. orco^−/− ants have two frameshift alleles and are therefore expected to be complete loss-of-function orco mutants. Each of these lines arose independently; n indicates the number of ants of each line (G1s and subsequent generations) used across all experiments in this study (see Table S3 for a description of the lines used in each specific experiment). PAM in bold. See also Figure S1 and Tables S1–S2.

Figure 2. Loss of Orco Expression and OSNs in orco^−/− Ants

(A) Antennal section of wild-type O. biroi showing DAPI counterstain (grey), Orco immunostain (red), and the merged image (black). Wild type ants possess a dense region of Orco-positive OSNs in the center of the antenna. (B) Antennal section of orco^−/− ant. orco^−/− ants lack Orco staining, indicating that the full-length Orco protein is absent. orco^−/− ants also lack the dense region of cells in the center of the antenna, indicating that most or all of the OSNs that would express orco in wild-types are absent in orco^−/− ants. Scale bars are 20 μm.

Figure 3. Reduced Antennal Lobes in orco^−/− Ants

O. biroi: (A) Dorsal (n-ventral) 3D projections of orco^−/− and wild-type ants. Antennal lobes indicated by dashed lines. orco^−/− antennal lobe is highly reduced relative to wild-type. Two orco^−/− ants had 90 and 91 glomeruli relative to 493 and 509 glomeruli for two wild-type ants (one of the wild-type reconstructions has been published previously (McKenzie et al., 2016)); small differences between replicates within treatments might reflect reconstruction errors or actual biological variation. (B) Antennal lobe volumes for wild-type (n = 5), orco^wt/− (n = 5), and orco^−/− (n = 4) ants (orco^wt/− and orco^−/− were age-matched at approximately 4 months old). orco^−/− ants, but not orco^wt/− ants, have significantly smaller antennal lobes than wild-type. D. melanogaster: (C) Anterior (n-ventral) 3D projections for wild-type and orco^−/− brains from one month old flies. Antennal lobes indicated by dashed lines. orco^−/− antennal lobe is similar to wild-type. Two orco^−/− flies had 43 and 44 glomeruli, and two wild-type flies each had 46 glomeruli. These glomerulus numbers were higher than has been previously reported, which is likely due to differences in sample preparation and imaging techniques. Slight differences in glomerulus numbers between replicates may be due to reconstruction errors, or may reflect modest antennal lobe phenotypes in orco mutant flies (STAR Methods). (D) Antennal lobe volumes for wild-type (n = 5) and orco^−/− (n = 5) flies. Volumes of wild-type and orco^−/− antennal lobes are not significantly different (p = 0.20, t-test). Scale bars are 20 μm. NS: not significant. Genotypic classes marked by different letters are significantly different (p < 0.05) after ANOVA followed by Tukey’s test (B). See also Tables S1 and S3.

Figure 4. Deficient Olfactory and Social Behavior in orco^−/− Ants.

(A) Example trajectories of wild-type and orco^−/− ants in Sharpie assay. (B) Line crossing events for wild-type (n = 9), orco^wt/− (n = 10), and orco^−/− (n = 8) ants in Sharpie assays, with ants from Figure 4A highlighted. Wild-type and orco^wt/− ants cross Sharpie lines (red) less frequently than printed lines (grey), but orco^−/− ants cross both lines at approximately equal frequencies. (C) Repulsion indices for ants in Sharpie assays. Repulsion index is calculated as proportion of printed line crosses. orco^−/− ants, but not orco^wt/− ants, are significantly less repelled than wild-types. (D) Example colony used for trail pheromone analysis. The same colony, containing a mixture of wild-type, orco^wt/−, and orco^−/− ants, is shown twice, with a wild-type or orco^−/− focal ant highlighted. (E) Example trail pheromone analysis. Trajectories, during which ants were moving and edges were excluded, were used to create 2-D histograms, or density maps, for each ant in the colony. These density maps were compared to the actual and randomized density maps for all other ants in the colony. The wild-type density map is more strongly correlated with the actual colony density map than with the randomized colony density map, while the orco^−/− density map is poorly correlated with both colony density maps. (F) Pearson correlation coefficients for individual ant density maps with the actual or randomized density map of the other ants in the colony. Pearson correlation coefficients for wild-type ants (n = 28), but not for orco^−/− ants (n = 9), were significantly higher in actual than randomized density maps (sample sizes were too small to test orco^wt/−). (G) Example colony showing an individual outside of the nest. (H) Distances traveled in 24 hr videos by ants in experimental colonies. orco^−/− ants (n = 10), but not wild-type (n = 40) or orco^wt/− (n = 8) ants, exhibit a wandering phenotype. (I) Time without contacting other ants in 24 hr videos. orco^−/− ants spend more time without contact than wild-type or orco^wt/− ants. ***p < 0.001; NS: not significant. Genotypic classes marked by different letters are significantly different (p < 0.05) after ANOVA followed by Tukey’s test (C), or from log-likelihood ratio tests on generalized linear mixed models followed by Tukey’s tests with colony as a random variable and actual/randomized maps (F) or genotypic class (H,I) as fixed variables. See also Figure S3 and Table S3.

Figure 5. Reduced Fitness in orco^−/− Ants

(A) Eggs laid per day over a two week period by orco^wt/− (n = 12) and orco^−/− (n = 14) ants relative to wild-type average (dotted line). orco^−/− ants laid significantly fewer eggs than orco^wt/− ants. Both orco^wt/− and orco^−/− ants laid significantly fewer eggs than the wild-type average of 0.34 eggs per day. Wild-type data are given as an average, rather than individual values, because most ants in each colony were wild-type and it was therefore not possible to obtain individual egg-laying rates for wild-type ants (see STAR Methods). (B) Survival of identically-reared and age-matched wild-type (n = 42), orco^wt/− (n = 7), and orco^−/− (n = 13) ants over a 34 day period. Survival of orco^−/− ants was significantly lower than that of wild-type ants. Survival of orco^wt/− ants was not statistically tested due to small sample size, but no trend toward reduced survival was observed. *p < 0.05; ***p < 0.001; NS: not significant. p values from an unpaired two-way Wilcoxon test (comparison of orco^wt/− and orco^−/− egg-laying rates) and one-way Wilcoxon tests (comparisons of orco^wt/− and orco^−/− egg laying rates to wild-type) using the mean egg-laying rate of wild-type ants in this experiment (A) or from a Fisher exact test (B). See also Table S3.