Abstract
Signal diversity in communication systems plays an essential role in maintaining mating boundaries between closely related species. To preserve fitness, it has been hypothesized that signal-receptor coupling is maintained via strong purifying selection. However, because strong negative selection antagonizes diversity, how communication systems retain their potential for diversification is puzzling. We propose that one possible solution to this conundrum is receptor pleiotropy. Specifically, we demonstrate that Gr8a, a member of the gustatory receptor family in Drosophila, is a pleiotropic receptor that contributes to both the perception and production of inhibitory mating pheromones in the peripheral nervous system and pheromone producing oenocytes, respectively. Together, our data provide an elegant genetic solution to a long-standing evolutionary conundrum.
One Sentence Summary The Drosophila chemoreceptor Gr8a contributes to the maintenance of pheromonal signal-receptor coupling via its pleiotropic action in both the perception and production of mating pheromones.
Communication systems are essential for determining species mating boundaries via functionally coupled signal-receptor pairs (1–4). Because qualitative or quantitative changes in either signals or receptors could carry fitness costs, the coupling of signal-receptor pairs should be maintained via purifying selection (4–6). Yet, closely related species often utilize distinct communication signals (7–10). Therefore, it is puzzling how coupled pheromone production and perception can be maintained by purifying selection; yet, retain the potential for signal diversification (4, 5, 11). Here we demonstrate that the perception and production of mating pheromones in Drosophila, two independent biological processes that reside in different tissues, are genetically coupled via the pleiotropic action of pheromone receptors in the peripheral nervous system and pheromone-producing oenocytes.
In Drosophila, cuticular hydrocarbons (CHCs) act as mating pheromones, which are essential for the integrity of reproductive boundaries between closely related species (12–18). The perception of CHCs, which are produced in the fat body and abdominal oenocytes (11, 13, 19), is mediated by specialized gustatory-like receptor neurons (GRNs) in appendages and the proboscis (20–24). We chose members of the Gustatory receptor (Gr) gene family as candidates for testing our hypothesis because several family members have already been implicated in the detection of excitatory and inhibitory pheromones (25–29). Since the expression of most Gr’s in gustatory receptor neurons (GRNs) has already been established (30, 31), we reasoned that candidate pleiotropic pheromone receptors should be also expressed in abdominal oenocytes (13). An RT-PCR screen identified 24, out of the 59 Gr family members in the Drosophila genome, as chemoreceptors with abdominal-enriched expression (Table S1).
Next, we focused our analyses on Gr8a, which was previously shown to contribute to the detection of the non-proteinogenic amino acid L-Canavanine (32). We found that Gr8a is expressed in 14–16 GRNs in the proboscis (Fig. 1A), and two paired GRNs in the foreleg pretarsus (Fig. 1B) of males and females. It is also expressed in abdominal oenocyte-like cells in males but not females (Fig. 1C). The sexually dimorphic expression pattern of Gr8a was further supported by qRT-PCR (Fig. 1D). We also found that Gr8a is co-expressed with the oenocyte marker Desat1 (13), as well as Desat1-negative cells with fat body-like morphology (Fig. 1E to G). These data suggest that in addition to its chemosensory function in males and females, Gr8a also functions in pheromone production system in males.
Therefore, we next investigated whether Gr8a, and the GRNs that express it, are required for sensory functions associated with mating decisions in males and females. We found that blocking neuronal transmission in Gr8a-expressing GRNs with the tetanus toxin (TNT) in females resulted in shorter copulation latency when courted by wild-type males (Fig. 2A). Similarly, homozygous (Fig. 2B) and hemizygous (Fig. 2C) Gr8a mutant females exhibited shorter copulation latency relative to wild-type controls, which could be rescued by the transgenic expression of a Gr8a cDNA in all Gr8a-expressing cells (Fig. 2D). In contrast, Gr8a and the neurons that express do not seem to contribute to male courtship latency or index towards wild-type virgin females (fig. S1). Since mating decisions in flies involve both excitatory and inhibitory signals (13, 33), a simple interpretation of these data is that in females, Gr8a contributes to the perception of male-emitted inhibitory mating signals.
Because Gr8a expression is enriched in male oenocytes, and Gr8a mutant females seem to be unable to sense a copulation inhibitory signal emitted by males, we next tested the hypothesis that Gr8a mutant males are unable to produce or release the putative copulation inhibitory signal detected by virgin females. Indeed, we found that wild-type virgin females exhibited shorter copulation latency towards Gr8a mutant males, which suggest these males did not produce/release the inhibitory signal important for the copulation decision of virgin females (Fig. 2E).
As predicted by our behavioral data, the Gr8a mutation also has a significant effect on the overall CHC profile of males (Fig. 3A). Analyses of individual pheromonal components revealed a significant contribution of Gr8a to levels of specific components in males, including alkenes and methyl-branched CHCs (Fig. 3B and Table S2), which have been implicated in mating decisions in several Drosophila species (13, 14, 17). Together, behavioral and pheromonal data indicate that Gr8a action contributes to mating decisions in females by co-regulating the female perception and male production of an inhibitory pheromone, which is consistent with Gr8a pleiotropy.
Previous studies showed that male Drosophila increase their fitness by transferring inhibitory mating pheromones to females during copulation, which lowers their overall attractiveness (13, 34–36). Therefore, we hypothesized that Gr8a mutant males would have less ability to produce/transfer inhibitory pheromones during copulation, and would not be able to detect inhibitory signals in mated females. Accordingly, we found that wild-type males fail to recognize mated status of wild-type females that previously mated with Gr8a mutant males, and Gr8a mutant males are not able to recognize the mated status of wild-type females that previously mated with wild-type males (Fig 3F). Together, these data indicate that Gr8a is required in males for the production/transfer, and subsequent detection, of an inhibitory mating signal in females. Therefore, Gr8a contributes to the regulation of both pre-and post-mating decisions in males and females by regulating the perception and production/release/transfer of inhibitory chemical mating signals.
Here we demonstrated that a pleiotropic gene that encodes a putative pheromone receptor can simultaneously regulate the perception and production of pheromones important for mating decisions in Drosophila. Nevertheless, we still do not understand the exact mechanism by which Gr8a exerts its pleiotropic action. However, how a chemoreceptor like Gr8a contributes to CHC production in oenocytes is not obvious. We speculate that Gr8a could regulate the synthesis and/or secretion of specific CHCs by acting as an oenocyte-intrinsic receptor, which integrates feedback information to the complex genetic network that regulates CHC synthesis (Fig 4). We also do not know yet the chemical identity of the ligand of Gr8a. Previous studies indicated cVA and CH503 as inhibitory mating pheromones that are transferred from male to females during copulation. However, these chemicals are not likely to function as Gr8a ligands because the volatile cVA acts primarily via the olfactory receptor Or67d (34, 35, 37), and CH503 has been reported to signal via Gr68a-expressing neurons, which are anatomically distinct from the Gr8a GRNs we describe here (36, 38).
Although we do not know yet whether the pleiotropic action of Gr8a supported the rapid species diversification in Drosophila, phylogenetic analysis of Gr8a indicated that its protein sequence and sexually dimorphic expression pattern are conserved across Drosophila species (fig. S2A), and alignment of orthologous sequences revealed that at least one predicted extracellular region is hypervariable (fig. S2C and D). These data suggest that pleiotropic pheromone receptors may have played a role in maintaining the functional coupling of the production and perception of mating pheromones while still retaining the capacity for species diversification.
Whether genetic coupling serves as an important mechanism for signal-receptor co-evolution in mating systems remains an open question (39, 40). Here we provide experimental data, which indicate that pleiotropic receptors can maintain signal-receptor coupling in a mating communication systems. We do not know yet whether pleiotropic chemoreceptor genes also contribute to pheromone-receptor coupling in other species or to communication systems that depend on other sensory modalities. Nevertheless, population genetics studies in crickets suggest that pleiotropy might be playing a role in auditory signal-receptor coupling as well (40, 41). While specific identities of the pleiotropic genes in these systems are mostly unknown, these data suggest that the genetic coupling of signal-receptor pairs in communication systems might be more common than previously thought.
Acknowledgments
We thank members of the Ben-Shahar lab for valuable comments on earlier versions of the manuscript. We thank the Millar laboratory (UC Riverside) for assistance with CHC analyses. We thank Nabeel Chowdhury for assistance with the RT-PCR screen, and Paula Kiefel for technical help with the generation of transgenic flies. This work was supported by NSF grant 1545778 and NIH grant NS089834 awarded to YB-S.