A Single Gene Affects Both Ecological Divergence and Mate Choice in Drosophila

See allHide authors and affiliations

Science  07 Mar 2014:
Vol. 343, Issue 6175, pp. 1148-1151
DOI: 10.1126/science.1249998

Evolutionary Duality

When ecological divergence involves traits that also contribute to mating behavior, such divergence could lead to rapid reproductive isolation and speciation. However, there are very few experimental demonstrations of such “dual” traits. Chung et al. (p. 1148, published online 13 February) now demonstrate that specific cuticular hydrocarbons are a dual trait that affects both desiccation resistance and mate choice in the widely distributed Australian species Drosophila serrata. These compounds have largely been lost from its rainforest-restricted, desiccation-sensitive, closely related sibling D. birchii.


Evolutionary changes in traits involved in both ecological divergence and mate choice may produce reproductive isolation and speciation. However, there are few examples of such dual traits, and the genetic and molecular bases of their evolution have not been identified. We show that methyl-branched cuticular hydrocarbons (mbCHCs) are a dual trait that affects both desiccation resistance and mate choice in Drosophila serrata. We identify a fatty acid synthase mFAS (CG3524) responsible for mbCHC production in Drosophila and find that expression of mFAS is undetectable in oenocytes (cells that produce CHCs) of a closely related, desiccation-sensitive species, D. birchii, due in part to multiple changes in cis-regulatory sequences of mFAS. We suggest that ecologically influenced changes in the production of mbCHCs have contributed to reproductive isolation between the two species.

The evolution of traits with dual roles in ecological divergence and mate choice could cause populations to become reproductively isolated through local adaptation (1, 2). However, the direct contribution of ecological divergence to speciation is uncertain, in part because relatively few systems have been investigated experimentally, and the genes affecting dual traits have not been identified. Insect cuticular hydrocarbons (CHCs) are potential dual traits (3). CHCs seal the cuticle, protecting against desiccating environments (4), and can also act as pheromones that influence mate choice and mating success (5). Whereas specific CHC molecules act as pheromones (6, 7), no specific CHC or class of CHCs has been demonstrated to be involved directly in desiccation resistance.

One class of CHCs, methyl-branched CHCs (mbCHCs), which have melting points above ambient temperature, are hypothesized to help maintain a barrier against evaporative water loss (8). D. serrata, a habitat generalist found outside of and on the fringes of the rainforest on the east coast of Australia, is relatively desiccation resistant and produces relatively large amounts of mbCHCs (29% of all CHCs) (Fig. 1A). In contrast, its close relative D. birchii, a habitat specialist found exclusively in the humid rainforest, is extremely sensitive to desiccation and produces very low amounts of mbCHCs (911) (3% of all CHCs) (Fig. 1 and fig. S1A). D. serrata and D. birchii show prezygotic isolation, which appears to be mediated by chemical signals (12). Notably, mbCHC levels are one factor correlated with male mating success in D. serrata (13, 14). If mbCHCs contribute to both mating success and desiccation resistance, then mbCHCs would constitute a dual trait, and the loss of mbCHCs in D. birchii could have contributed to its reproductive isolation from D. serrata.

Fig. 1 Closely related Drosophila species differ in mbCHC levels, whose synthesis is controlled by the mFAS gene.

(A) D. serrata and D. birchii occupy different habitats, but their ranges overlap on the east coast of Australia. The habitat generalist D. serrata produces higher levels of mbCHCs (29%) than the rainforest specialist D. birchii (3%). (B) The biosynthetic pathways for CHC production in insects. Two different forms of FAS are responsible for the synthesis of methyl-branched and straight-chained CHCs in Drosophila (18). (C) RNA in situ hybridization of the three Drosophila FAS genes on adult D. melanogaster males. CG3523 transcript was detected in the fat body. CG17374 and CG3524 transcripts were detected in the oenocytes (arrowheads), the site of CHC synthesis. Insets show right side of second tergite (abdominal segment). For the CG3523 inset, the fat body was removed. Expression of all three FASs is sexually monomorphic. (D) Oenocyte-specific RNAi knockdown of CG3524 in D. melanogaster results in a large reduction of mbCHCs (results shown in males), suggesting that it is a methyl-branched specific FAS (mFAS).

To explore this scenario, we identified the gene(s) responsible for the differences in mbCHCs between D. serrata and D. birchii. Because the levels of both major mbCHCs (2Me-C26 and 2Me-C28) are lower in D. birchii than in D. serrata (10) (fig. S1), we reasoned that the underlying genetic differences between these two species must have occurred early in the fatty acid synthase (FAS) pathway responsible for mbCHC synthesis (Fig. 1B).

Biochemical studies have identified two forms of FASs in insects, one for producing unbranched CHCs such as alkanes, monoenes, and dienes, and the other for producing mbCHCs (15). The D. melanogaster genome contains three putative FASs: CG3523, CG3524, and CG17374 (16). In situ hybridization of RNA probes for all three FASs in adult abdomens revealed that CG3523 is expressed only in the adult fat body, whereas CG17374 and CG3524 are both expressed in adult D. melanogaster oenocytes (Fig. 1C), where CHCs are produced (17).

RNA interference (RNAi)–mediated knockdown of CG3524 and CG17374 in D. melanogaster by a ubiquitously expressed driver (tubulin-GAL4) was lethal. An oenocyte-specific driver (oenoGAL4) (17) inducing CG17374 RNAi was also lethal, whereas oenoGAL4-driven RNAi knockdown of CG3524 produced viable adult flies in which mbCHC production was strongly and specifically reduced (Fig. 1D). This suggests that CG3524 is a mbCHC-specific FAS (mFAS). RNAi-mediated knockdown of CG17374 in oenocytes after adult flies eclosed yielded viable adults with unchanged CHC profiles, whereas knockdown of CG3524 in adults reduced mbCHCs, thus confirming the specific role of CG3524 as mFAS.

We then isolated the orthologous gene in the D. serrata genome and drove D. serrata mFAS RNAi under the direct control of an oenocyte-specific enhancer (fig. S2), obtaining two independent lines of transgenic D. serrata flies carrying this construct (mFASRNAi-1 and mFASRNAi-2). mbCHC levels were reduced in males and females of both lines, indicating that the role of mFAS in producing mbCHCs is conserved among Drosophila (Fig. 2A and table S1).

Fig. 2 Loss of mbCHCs strongly reduces desiccation resistance in D. serrata.

(A) Oenocyte-specific RNAi knockdown of mFAS in D. serrata results in almost complete loss of mbCHCs (results shown in males). (B) Oenocyte-specific RNAi knockdown of mFAS in D. serrata results in a dramatic reduction of desiccation resistance in both male (t test; df = 17; t = 26.7, P = 6.4 × 10−12) and female (df = 18; t = 65.7; P = 1.8 × 10−17) mFASRNAi-1 flies. The mean time to mortality of 10 flies in each of n = 9 to 10 vials per genotype was compared with t tests (two-tailed, unpaired, unequal variance). Mean ± SEM is shown.

To assess whether mbCHCs contribute to desiccation resistance in D. serrata, we subjected D. serrata mFAS-RNAi and wild-type flies to a desiccation assay. We found that the D. serrata mFAS-RNAi flies exhibited reduced desiccation resistance (Fig. 2B). To investigate whether the loss of desiccation resistance was due to the specific loss of mbCHCs, we applied a mixture of purified synthetic mbCHCs to D. serrata mFAS-RNAi flies. The application of synthetic mbCHCs significantly increased resistance (fig. S3B), indicating that the low resistance of mFAS-RNAi flies was due to the specific loss of mbCHCs. We also applied synthetic mbCHCs to D. birchii flies and observed increased desiccation resistance (fig. S3B), suggesting that the comparatively low desiccation resistance of this species may be due at least in part to its low levels of mbCHCs.

The D. birchii mFAS gene (GenBank JX627578) contains a complete coding sequence, and cDNA could be isolated from adults, indicating that the gene is intact and functional. However, mFAS expression was not detectable in D. birchii oenocytes by in situ hybridization, suggesting that transcription in these cells is reduced or absent (Fig. 3). mFAS is expressed in the oenocytes of all other Drosophila species examined, including D. serrata (fig. S4). Thus, mFAS expression in oenocytes appears to have been lost in the D. birchii lineage.

Fig. 3 mFAS is not expressed in D. birchii oenocytes.

RNA in situ hybridization reveals that, whereas mFAS shows a sexually monomorphic expression pattern in D. serrata oenocytes (left), it is undetectable in either sex in D. birchii (right). To confirm that the lack of transcript detection was not due to any defect of the D. birchii RNA probe, we showed that this probe detected mFAS expression in D. serrata (center). Arrowheads point to oenocytes; open arrowheads indicate no visible expression.

To determine whether loss of mFAS expression is due to cis-regulatory changes, we made green fluorescent protein (GFP) reporter constructs with segments of the mFAS gene from both species and compared their activity when transformed into D. melanogaster. Both 5′ and intronic mFAS sequences from D. birchii exhibited lower activity than the homologous regions from D. serrata (fig. S5), indicating that the reduction in D. birchii mFAS expression is partly due to multiple cis-regulatory changes at the locus.

The ecological divergence of the rainforest-restricted D. birchii and widely distributed D. serrata involved reductions of desiccation resistance, mFAS expression, and mbCHCs. mbCHCs are one factor that has been correlated with male mating success in D. serrata (13, 14), and thus could have played a role in reproductive isolation between the species. To determine whether mbCHCs directly affect mating in D. serrata, we conducted a series of mate-choice experiments in which mbCHC levels were manipulated. When single wild-type males were presented with a choice between a wild-type female and a mFAS-RNAi female, we did not observe any mating preference (n = 105; χ2 = 0.77; P = 0.38). This suggests that D. serrata males do not discriminate on the basis of mbCHC levels in females. In contrast, when single wild-type females were given the choice between a wild-type male and either a mFASRNAi-1 or mFASRNAi-2 male, they preferred males with normal production of mbCHCs over males with reduced production (Table 1) [this effect was dependent on the age of the flies (see table S2)].

Table 1 mbCHCs affect mating success in D. serrata.

(A) Three-day-old wild-type D. serrata males had significantly higher mating success than mFAS-RNAi flies in female choice assays where single wild-type females were given a choice between the two genotypes. (B) Wild-type and mFASRNAi-2 males were perfumed with 2Me-C26 and 2Me-C28 individually. Females are given a choice between a perfumed male and a nonperfumed male with the same genotype. Perfuming with 2Me-C26 significantly increased mating success of both wild-type and mFAS-RNAi flies.

View this table:

To determine whether increasing the level of mbCHCs also influenced mating success, we perfumed wild-type males with either 2Me-C26 or 2Me-C28 (table S3). We found that males perfumed with 2Me-C26 exhibited increased mating success, whereas 2Me-C28 had no detectable effect (Table 1). Finally, we investigated whether perfuming mFAS-RNAi males with either 2Me-C26 or 2Me-C28 increased male mating success and found that 2Me-C26 again increased male mating success. Together, our results indicate that mbCHCs, whose production is controlled by the mFAS gene, affect mating success as well as desiccation resistance, and thus function as a dual trait in D. serrata.

To determine whether the difference in mbCHC levels might also contribute to premating isolation between D. serrata and D. birchii, we investigated whether manipulating the level of mbCHCs in D. birchii could break the interspecies mating barrier. We perfumed D. birchii males and females with synthetic mbCHCs before introducing them to D. serrata females and males, respectively. In neither case did we observe any copulations. Importantly, however, we also observed that no males of either species even attempted to court females of the other species, a necessary preamble before females exhibit a choice. These observations suggest that courtship by males is also governed by cues that differ between the species, such that modulation of mbCHCs alone is not sufficient to surmount the barrier to mating. This is not surprising because numerous factors contribute to courtship and mating among Drosophila species, and these traits may also be subject to rapid divergence (18, 19).

Although restoring mbCHCs to D. birchii was not sufficient to surmount its reproductive isolation from D. serrata, the demonstrated role of mbCHCs as a component of mate choice in D. serrata and the loss of mbCHCs in its rainforest-restricted sibling suggest a plausible scenario for mbCHCs in ecological speciation. That is, as populations of the common ancestor of these two species diverged ecologically, and if the rainforest-adapting population lost mbCHCs and mFAS expression before or during speciation (perhaps due to relaxed constraints on desiccation resistance), then such changes would likely result in increasing reproductive isolation between the populations. Subsequent changes at other genetic loci would then complete and/or reinforce speciation (fig. S6).

It is becoming increasingly apparent that adaptation to different environments may lead to reproductive isolation (1). Given the dual role of CHCs in insect water balance and sexual communication, and their great chemical diversity, changes in CHC production in ecologically diverging populations may be an important general contributor to insect speciation.

Supplementary Materials

Materials and Methods

Figs. S1 to S6

Tables S1 to S4

References (2029)

References and Notes

  1. Acknowledgments: We thank V. Kassner and J. Selegue for technical help and advice; T. Shirangi for initial discussions about this project; M. Giorgianni and the University of Wisconsin Genome Center for genome sequencing; T. Werner and S. Koshikawa for advice on transgenesis; and J-C. Billeter and J. Levine (University of Toronto) for the oenoGAL4 flies. D.W.L. is a Howard Hughes Medical Institute Fellow of the Life Sciences Research Foundation. S.B.C. is a Howard Hughes Medical Institute investigator. DNA sequences of D. birchii mFAS and D. serrata mFAS were deposited into GenBank under accession numbers JX627578 and JX627579. Partial cDNA sequence of D. birchii mFAS obtained from Sanger sequencing was deposited into GenBank under accession number KF791130. H.C. and S.B.C. designed and conceived the project. H.C. performed the experiments with help from D.W.L, H.D.D., and K.V. J.G.M. ran and analyzed the gas chromatography–mass spectrometry data and synthesized mbCHCs. H.C., D.W.L and S.B.C. wrote the manuscript with input from H.D.D. and J.G.M.
View Abstract

Stay Connected to Science

Navigate This Article