Regulatory changes in the fatty acid elongase eloF underlie the evolution of sex-specific pheromone profiles in Drosophila prolongata

Luo, Yige; Takau, Ayumi; Li, Jiaxun; Fan, Tiezheng; Hopkins, Ben R.; Le, Yvonne; Ramirez, Santiago R.; Matsuo, Takashi; Kopp, Artyom

doi:10.1186/s12915-025-02220-z

Research
Open access
Published: 30 April 2025

Regulatory changes in the fatty acid elongase eloF underlie the evolution of sex-specific pheromone profiles in Drosophila prolongata

BMC Biology volume 23, Article number: 117 (2025) Cite this article

516 Accesses
2 Altmetric
Metrics details

Abstract

Background

Pheromones play a key role in regulating sexual behavior throughout the animal kingdom. In Drosophila and other insects, many cuticular hydrocarbons (CHCs) are sexually dimorphic, and some are known to perform pheromonal functions. However, the genetic control of sex-specific CHC production is poorly understood outside of the model species D. melanogaster. A recent evolutionary change is found in D. prolongata, which, compared to its closest relatives, shows greatly increased sexual dimorphism in both CHCs and the chemosensory system responsible for their perception. A key transition involves a male-specific increase in the proportion of long-chain CHCs.

Results

Perfuming D. prolongata females with the male-biased long-chain CHCs reduces copulation success, suggesting that these compounds function as sex pheromones. The evolutionary change in CHC profiles correlates with a male-specific increase in the expression of multiple genes involved in CHC biosynthesis, including fatty acid elongases, reductases and other key enzymes. In particular, elongase F, which is responsible for producing female-specific pheromones in D. melanogaster, is strongly upregulated in D. prolongata males compared both to females and to males of the sibling species. Mutations in eloF reduce the amount of long-chain CHCs, resulting in a partial feminization of pheromone profiles in D. prolongata males. Transgenic experiments show that sex-biased expression of eloF is caused in part by a putative transposable element honghaier insertion in its regulatory region.

Conclusions

These results show that cis-regulatory changes in the eloF gene, along with other changes in the CHC synthesis pathway, contribute to the evolution of sexual communication.

Background

Communication, both between and within the sexes, plays a pivotal role in sexual selection and the evolution of sexual dimorphism [7, 26, 27, 121, 153]. However, our understanding of the genetic control of both signaling and signal perception remains limited outside traditional model systems. In insects, as well as other animals, pheromones are one of the key methods of communication [27, 134, 135, 159]. Among the most common insect pheromones are cuticular hydrocarbons (CHCs), which affect a wide range of social and non-social functions including maintaining water balance [34, 35, 150], resource acquisition [10], social aggregation [138], cohort recognition [143], mate choice [35, 87, 132], aggression [148, 161], and signaling fecundity [105] and immunocompetence [89].

Much of our understanding of pheromone communication comes from Drosophila, where several chemicals have been confirmed to have pheromonal effects [21, 49, 73, 159]. In D. melanogaster, the male-specific compounds cis-Vaccenyl Acetate (cVA) and 7-Tricosene (7 T) promote aggression when perceived by males and increase receptivity when perceived by females [42, 60, 82, 120, 149]. In contrast, the female-specific 7,11-heptacosadiene (7,11-HD) functions as an aphrodisiac [53]. 7,11-HD initiates a neural cascade that flows from peripheral chemoreceptors to the central nervous system to stimulate male courtship behavior, whereas the perception of 7 T inhibits the courtship circuitry in males and regulates reproductive functions in females [25, 124, 144].

The importance of CHCs in mating behavior can contribute to the evolution of reproductive barriers [38, 47]. The best-studied example is found in D. melanogaster and its sibling species D. simulans, where interspecific differences in the processing of the 7 T and 7,11-HD signals contribute to pre-mating isolation [38, 47]. Within D. melanogaster, a higher abundance of female-specific 5,9-heptacosadiene in African populations contributes to the partial isolation between African and non-African strains [48, 50, 61, 158]. Divergent pheromone profiles also contribute to reproductive isolation in other Drosophila species, including the 9-pentacosene between different populations of D. elegans [67], 2-methyl hexacosane between D. serrata and D. birchii [35, 66], and 10-heptadecen-2-yl acetate between different subspecies of D. mojavensis [72].

Understanding the genetic basis of pheromone evolution has been facilitated by a well-characterized pathway for CHC biosynthesis. In insects, key steps in this process, including fatty acid synthesis, desaturation, elongation, and decarboxylation, are highly conserved [19, 155]. These reactions take place mainly in adult oenocytes, a specialized cell type located beneath the abdominal epidermis [16, 52, 100]. Dietary lipids, such as palmitic and stearic acids, are CoA-activated by fatty acyl synthases, followed by the introduction of position-specific double bonds catalyzed by desaturases. Elongation proceeds with the incorporation of malonyl-CoA, adding two carbons at a time to the growing precursor chain. The synthesis of very-long-chain CHCs is catalyzed by fatty acid elongases (FAEs), with the additional involvement of three other categories of enzymes: 3-keto-acyl-CoA-reductase (KAR), 3-hydroxy-acyl-CoA dehydratase (HACD), and trans-enoyl-CoA-reductase (TER) [31, 155, 159]. Fatty acyl-CoA reductases (FARs) act on the very long chain fatty acyl-CoAs produced by the elongation process, reducing them to aldehydes. From these aldehydes, mature CHCs are produced by oxidative decarboxylation catalyzed by insect-specific cytochrome P450 [114]. The multi-stage chemistry that creates the final structure of CHCs offers multiple points at which the end products can be modified. Variation in CHC profiles has been attributed to genes controlling the positions of double bonds [32, 41], methyl branches [35], and chain length [31, 38, 110, 119].

Sexually dimorphic CHCs have been observed in most Drosophila species that have been examined (81/99) [73], but our understanding of how sex-specific pheromones are produced continues to be based on genetic studies in D. melanogaster. A complete feminization of pheromone profiles can be achieved by targeted expression of the female sex determiner, transformer (tra), in adult male oenocytes [52]. Downstream, at least two key enzymes are under the control of the sex differentiation pathway: elongase F (eloF) and desaturase F (desatF, also known as Fad2), which control carbon chain elongation and the production of alkadienes, respectively [31, 32]. Female-specific expression of these enzymes contributes to the production of 7,11-HD, the critical female pheromone in D. melanogaster, as well as to the higher abundance of very long chain CHCs in females. However, a comparative analysis has shown that the female-restricted expression of desatF has evolved relatively recently, in the common ancestor of D. melanogaster and D. erecta, and that more distantly related Drosophila species express desatF in a sexually monomorphic manner that correlates with sexually monomorphic diene abundance [128]. The evolution of sex-biased desatF expression in the D. melanogaster lineage was associated with the gain of binding sites for doublesex (dsx), the key transcription factor that acts downstream of tra to direct the sexual differentiation of somatic cells, in the oenocyte enhancer of desatF [128]. Outside of D. melanogaster and its closest relatives, the genetic basis of sex-specific pheromone production, and especially the synthesis of male-specific pheromones that are found in many Drosophila species [73], is largely unknown.

In this report, we examine the genetic basis and evolutionary origin of male-biased pheromones in D. prolongata. This species exhibits multiple derived sex-specific traits compared to its close relatives [131], making it an attractive model for investigating coevolution between signals and receptors that mediate sexual communication. Along with many species-specific features of mating behavior and male-male aggression [6, 104, 127], D. prolongata has strongly diverged from its relatives both in the chemical signals and in their receptors. On the sensory perception side, this species shows a dramatic, sex-specific increase in the number of gustatory organs on the front legs of males [96]. Leg gustatory organs have a well-characterized role in sex-specific pheromone perception in D. melanogaster [25, 124, 144], and D. prolongata males use their front legs extensively in both courtship and male-male aggression [6, 104, 126, 127, 160], suggesting that this morphological change may have important behavioral consequences. And on the signaling side, D. prolongata shows a recently evolved, strongly sex-biased CHC profile [98]. Specifically, the difference involves the relative amounts of three serial chemical homologs, 9-tricosene (9T), 9-pentacosene (9P), and 9-heptacosene (9H). These molecules differ only in the length of the carbon backbone, and are likely to share common biosynthetic origin. While its closest relatives such as D. rhopaloa and D. carrolli are sexually monomorphic in the abundance of these CHCs, D. prolongata males show a dramatic increase in the amounts of 9P and 9H, and a concomitant reduction in the amount of 9T, compared to females.

To identify the genetic changes responsible for the evolutionary transition from sexually monomorphic to sexually dimorphic CHC profiles, we compared gene expression in pheromone-producing tissues between D. prolongata and D. carrolli. We find that D. prolongata males show increased expression of many enzymes involved in CHC synthesis, including multiple fatty acyl elongases and reductases. We show that eloF, which is responsible for the female-biased abundance of long-chain CHCs in D. melanogaster, is expressed in a male-specific manner in D. prolongata, due in part to changes in its cis-regulatory sequences, and is partly responsible for the increased abundance of 9P and 9H in D. prolongata males. Finally, we confirm that these CHCs affect sexual behavior. Together, our results reveal one of the genetic mechanisms responsible for a recent evolutionary change in sexual communication.

Results

Perfuming with male-specific pheromones reduces female mating success

Male-biased chemical cues are often used in a reproductive context, for example as inhibitory signals against female remating that function as chemical mate-guarding strategy [53, 69, 86, 108]. We previously showed that D. prolongata exhibits a male-specific increase in the abundance of two long-chain CHCs, 9-pentacosene (9P) and 9-heptacosene (9H) [98]. To investigate the role of these hydrocarbons in mating behavior, we examined male–female interactions by pairing single virgin males with single virgin females perfumed with synthetic 9P or 9H. On average, each female received ~ 350 ng of extra 9P in the 9P treatment and ~ 90 ng of additional 9H in the 9H treatment, as shown by GC–MS (Fig. 1 A’, B’). Perfumed females, therefore, had a masculinized pheromone profile with an abundance of male-biased hydrocarbons intermediate between those observed in normal D. prolongata males and females (Fig. 1A, B).

In mating trials, nearly all males encountered their female partners at the mating arena (Additional file 1: Fig. S1 A). Males rarely showed threatening behavior toward the perfumed females, a stereotypical aggressive behavior displayed towards other males (Additional file 1: Fig. S1B), suggesting that males could still recognize the sex identities of females with modified CHC profiles by using other chemical or non-chemical cues. In the 9H treatment, we observed a non-significant decrease in the rate of courtship initialization (N = 32, logistic regression z-test, p = 0.25, Additional file 1: Fig. S1 C) and leg vibration (p = 0.066, Additional file 1: Fig. S1D), which may suggest reduced motivation in males. Despite the non-significant effects on the individual elements of courtship behavior, we found a strong decrease in copulation success when females were perfumed with either 9P (N = 32, p = 0.0475) or 9H (p = 0.023, Fig. 1 C) compared with the hexane control (59%, N = 32). The proportion of pairs that mated in the 9H treatment (28% mated) was reduced by half compared to the hexane control (59% mated). The success rate was reduced less in the 9P treatment (34% mated), even though a larger amount of the synthetic CHC was introduced. This disparity may suggest that 9H was perceived as a stronger masculinity cue than 9P, and therefore outweighed 9P in mate evaluation and decision-making during courtship. Simultaneous perfuming with both 9P and 9H did not result in further inhibition of courtship and copulation (Fig. 1; Additional file 1: Fig. S1). It is possible, however, that higher concentrations of 9P and/or 9H would induce stronger behavioral changes.

A key Drosophila pheromone, cis-vaccenyl-acetate (cVA), is transferred from males to females during mating and subsequently inhibits courtship by rival males [46, 53, 69, 108]. Male-biased CHCs are also transferred to females in many other Drosophila species [73]. We therefore tested whether D. prolongata males transferred 9P or 9H to females during mating. However, no transfer was observed (Additional file 2: Fig. S2), suggesting that while these CHCs reduce female attractiveness, they are unlikely to be involved in chemical mate guarding or male-male competition in a manner similar to cVA.

Gene expression shows stronger sexual dimorphism in D. prolongata than in D. carrolli

We previously showed that sexually dimorphic pheromone profiles, with an increased abundance of 9P and 9H in males, have evolved in D. prolongata from a sexually monomorphic ancestor [98]. To identify the genes responsible for this evolutionary transition, we performed RNA sequencing on dissected oenocyte-enriched tissues (abbreviated as oenocyte dissections) in sexually mature adults of both sexes of D. prolongata and D. carrolli, followed by differential gene expression analysis. We defined our candidate genes as those that show (1) differential expression between males and females in D. prolongata (Fig. 2A) and (2) differential expression between males of D. prolongata and D. carrolli (Fig. 2B). To also account for the possibility that both D. prolongata and D. carrolli are sexually dimorphic, but the extent or direction of sex bias differs between the two species, we also required that the differentially expressed genes show interaction effects between species and sex (Fig. 2C).

In the comparison between male and female D. prolongata, 526 genes were identified as differentially expressed (Fig. 2A). Differentially expressed genes (DEGs) are almost equally likely to be female-biased (262 genes) as male-biased (264 genes). We found many more genes (2812) that were differentially expressed between males of D. prolongata and D. carrolli (Fig. 2B). These genes were slightly more likely to be enriched in D. carrolli (Binomial test, p = 2.4e-3), with 46.7% (1325 genes) having higher expression in D. prolongata. 91 genes showed significant interaction between species and sex (Fig. 2C). Consistent with D. prolongata being more sexually dimorphic in various phenotypes, the latter DEGs are more likely to show stronger sexual dimorphism in D. prolongata than in D. carrolli (Binomial test, p = 2.6e-10), with only 17.6% (16 genes) showing stronger dimorphism in D. carrolli.

Sexually dimorphic and species-biased genes are enriched for lipid metabolism functions

The sexually dimorphic pheromone profile of D. prolongata is mainly attributable to a lower abundance of the shorter-chain 9T, and a higher abundance of the longer-chain 9P and 9H, in males [98]. These compounds differ only in the number of carbons, suggesting a simple chemical basis for their differences – namely, a higher carbon chain elongation activity in males compared to females. To identify the molecular pathways that may underlie male–female differences in CHC profiles, we performed Gene Ontology (GO) enrichment analysis of the genes that show sex-biased expression in D. prolongata and oenocyte expression in D. melanogaster. We found 26 significantly enriched GO terms, of which the top 6 categories are all associated with lipid metabolism (Table 1,Fig. 3; Additional file 3: Fig. S3).

Table 1 Significant GO terms in the comparison between D. prolongata males and females

Full size table

Long-chain fatty acyl CoA metabolic process (GO:0035336) shows particular enrichment in D. prolongata males compared to females (Table 1; Fig. 3; Additional file 3: Fig. S3). These genes include seven fatty acyl reductases (FARs): CG17560, CG17562, CG14893, CG4020, CG5065, CG8306, and CG30427 [33, 54] and six fatty acid elongases (FAEs), including elongase F (eloF), CG9458, CG16904, CG9459, CG33110, and bond. Some members of both FAR and FAE gene families have been shown to affect the production and relative ratios of long-chain and short-chain pheromones [31, 43, 110, 119]. We also found an enrichment of genes associated with transmembrane transport (26 genes, GO:0055085) (Table 1; Additional file 3: Fig. S3). This may reflect the need for CHCs to be transported from the oenocytes to the cuticle, which likely involves crossing the intervening epithelium and several layers of cell membrane [19]. As expected, we also found an enrichment of genes involved in somatic sex differentiation (GO:0007548), including Sex-lethal (Sxl), transformer (tra), doublesex (dsx), and yolk proteins yp1, yp2, and yp3, which are known molecular targets of Dsx [64, 77, 157].

In the GO enrichment analysis of genes that show differential expression between D. prolongata and D. carrolli males and oenocyte expression in D. melanogaster, we identified 32 overrepresented and 7 underrepresented GO terms (Additional file 4: Table S1; Additional file 5: Fig. S4). As in the sex bias analysis, the enriched GO terms include terminal lipid metabolism processes, such as fatty acid elongation (GO: 0030497, see child GO terms GO:0034625, GO:0034626 and GO:0019367 in Additional file 5: Fig. S4). A closely related process is the very-long-chain fatty acid biosynthetic process (GO:0042761), which contains the 3-hydroxy-acyl-CoA-dehydratase (HACD) Hacd2 and the trans-enoyl-CoA-reductase (TER) Sc2, which exhibits extremely D. carrolli-biased expression (> 12,000-fold change) and is sexually monomorphic in D. prolongata. Both HACDs and TERs are required for the elongation step during the synthesis of very-long-chain fatty acids, as is another enzyme class, 3-keto-acyl-CoA-reductases (KAR, [155]). Among predicted KARs, CG13284 showed differential expression between D. prolongata and D. carrolli. The long-chain fatty acyl CoA metabolic process (GO:0035336 also showed strong enrichment in this analysis (Additional file 4: Table S1; Additional file 5: Fig. S4. In addition to the 5 FARs identified in the male–female comparison, we detected significant differential expression of CG18031, a FAR that was previously shown to function in larval oenocytes [37].

Beyond the genes immediately related to the pheromone synthesis pathway, we also observed wider differences in lipid metabolism. The lipid catabolic process (GO:0016042), in addition to being male-biased in D. prolongata, also shows higher expression in D. prolongata compared to D. carrolli (Fig. 3, Additional file 5: Fig. S4). This GO term contains 74 species-biased genes, many of which have annotated or predicted function in lipid storage, mobilization, and transport. Representative examples include the medium-chain acyl-CoA dehydrogenase Mcad [39], hormone-sensitive lipase Hsl [15], juvenile hormone epoxide hydrolases Jheh1 [28] and Jheh2 [55], ABC-type fatty-acyl-CoA transporter ABCD [56], long-chain-3-hydroxyacyl-CoA dehydrogenase Mtpα [76], and predicted acetyl-CoA C-acetyltransferase yip2 [84].

In summary, GO enrichment analyses indicate that, consistent with the lipidic nature of CHCs, a disproportionately high number of genes involved in lipid metabolism are differentially expressed between males and females and between D. prolongata and D. carrolli. These genes, in particular fatty acid elongases and fatty acyl reductases, could underlie the evolution of sexually dimorphic pheromone profiles in D. prolongata.

Candidate genes show increased male bias in D. prolongata

By intersecting the three selection criteria (male vs. female D. prolongata (Fig. 2A), male D. prolongata vs. D. carrolli (Fig. 2B), and interaction effects of species and sex (Fig. 2C)), we reduced the number of top candidate genes to 53, most of which show their highest expression in D. prolongata males (Fig. 2D). To test for correlated expression among these genes, we performed hierarchical clustering of genes and samples. The samples of D. prolongata males showed the greatest differences from the other samples (Fig. 4, left). We identified four major clusters of genes with distinct expression patterns (Fig. 4, top). The largest two clusters (red and purple) consist of 39 genes that show higher expression in D. prolongata males compared both to D. carrolli and to conspecific females. Most of these genes do not show significant sex differences in D. carrolli. Compared to the monomorphic D. carrolli, genes in the red cluster (24 genes) are strongly upregulated in D. prolongata males, while those in the purple cluster (15 genes) are downregulated in D. prolongata females (Fig. 4).

In principle, the evolution of male-biased pheromone profiles in D. prolongata could be explained either by species-specific increase or by species-specific reduction in the expression of genes in the pheromone biosynthesis pathway. The former pattern appears to dominate. In the third-largest cluster (blue in Fig. 4, 12 genes), most genes have mildly dimorphic expression in D. carrolli and increased dimorphism in D. prolongata, while fewer have overall lower expression in D. prolongata. The last and smallest cluster (orange, 3 genes) shows generally higher expression in D. prolongata, especially in females (Fig. 4).

In summary, D. prolongata males show a distinctive gene expression profile due mainly to male-specific upregulation of multiple genes. While the genes showing monomorphic or female-biased expression in D. prolongata (orange and blue clusters) may be necessary for the synthesis of species-specific pheromones, the genes directly responsible for the male-biased pheromone profile are more likely to be part of the male-enriched red and orange clusters. These clusters contain a number of fatty acid elongases and other genes involved in fatty acid metabolism (Fig. 4), whose increased expression may account for the evolution of male-specific CHC profile in D. prolongata.

Functionally related genes are distributed in local genomic blocks

Genes involved in CHC metabolism are likely to be expressed in the oenocytes of other Drosophila species, including the well-studied model D. melanogaster. We intersected the 53 candidate genes identified above with the oenocyte-expressed genes from the D. melanogaster Fly Cell Atlas data [90]. For genes that do not have gene-level annotations, we inferred their functions from the associated GO terms. Most functionally related candidates, including genes from the very-long-chain fatty-acyl-CoA metabolic process (CG17560, CG17562, CG8306, CG5065) and fatty acid elongation (eloF, CG9458, CG9459, CG16904) show oenocyte expression in D. melanogaster (Fig. 4). We cannot rule out that some of the other genes are not detected in the Fly Cell Atlas due to the gene drop-out typical of single-nucleus data, and or that some genes are expressed in oenocytes in D. prolongata but not in D. melanogaster or vice versa.

We found that many candidate genes are spatially clustered in the genome (Fig. 4). In particular, the fatty acid elongases eloF, CG8534, CG9458, CG9459, and CG16904 are all located in a ~ 10 kb genomic neighborhood (Fig. 4). Except for CG16904, which shows reduced expression in D. prolongata females compared to D. carrolli, the other four genes in this cluster have increased expression in D. prolongata males. Between D. prolongata males and females, eloF is 217-fold enriched in males, CG8534 156-fold, CG16904 87-fold, CG9458 148-fold, and CG9459 54-fold (Fig. 5A). This combination of spatial clustering and common sex bias suggests that these genes may share some cis-regulatory elements.

The best-studied elongase gene is eloF, which encodes a bona fide fatty acid elongase. Oenocyte-specific knockdown of eloF in D. melanogaster leads to a reduction in the abundance of long-chain hydrocarbons, which are female-specific in that species [31]. On the other hand, oenocyte-specific knockdown of CG9458 is not sufficient to change the balance between long- and short-chain CHCs in D. melanogaster [43].

Another example of spatial clustering is found among 5 fatty acyl reductases. CG8303, CG8306, and CG5065 are tandemly arranged in the genome (Fig. 4). These FARs are likely to be involved in essential lipid metabolism as RNAi knockdown leads to lethality [54]. Two other FARs, CG17560 and CG17562, are located in a separate genomic cluster. Oenocyte-specific knockdown of CG17562 affects the production of short-chain mono alkenes and long-chain alkanes in D. melanogaster females [33]. Lastly, a local cluster is formed by three genes involved in ecdysteroid metabolism (CG9519, CG9522, and CG12539). Ecdysone regulates pheromone biosynthesis in D. melanogaster [9, 33] and houseflies [2, 17, 18, 20]. Interestingly, hormone receptor 4 (Hr4), which encodes a nuclear receptor responding to ecdysone [75], is also among the 53 candidate genes. All 4 genes (CG9519, CG9522, CG12539, and Hr4) show strongly sexually dimorphic expression in D. prolongata while being sexually monomorphic in D. carrolli (Fig. 4). In conclusion, it is possible that correlated changes in the expression of genes involved in CHC synthesis were facilitated in part by their clustered genomic arrangement.

Fatty-acid elongase eloF shows extremely male-biased expression in D. prolongata

Among the 53 candidate genes, we identified elongase F (eloF) as the top candidate underlying the observed sexual dimorphism of CHC profiles in D. prolongata (p = 7.29e-10, Fig. 2A-C, Fig. 4). Expression of this gene is strongly male-biased in D. prolongata, with a 217-fold difference between males and females based on RNA-seq data, but is not sexually dimorphic in D. carrolli (p = 0.80, Fig. 5). The only other gene with a comparable sex bias is roX1, a long non-coding RNA involved in X-chromosome dosage compensation. Moreover, eloF shows 79-fold higher expression in D. prolongata males compared to D. carrolli males.

To validate these results, we used quantitative PCR (qPCR) to amplify eloF transcripts from an independent set of oenocyte dissections. qPCR results support the strong male bias in D. prolongata (3413-fold enrichment, Fig. 6A) and higher expression level in males of D. prolongata over D. carrolli (82-fold enrichment, Fig. 6A). Contrary to the RNA-seq results, qPCR results suggest a modest (fourfold) but significant male-biased eloF expression in D. carrolli (Fig. 6A). This is consistent with previously described CHC phenotypes, where longer-chain hydrocarbons are slightly more enriched in males than females of D. carrolli [98]. Despite this discrepancy between RNA-seq and qPCR, it is clear that sex differences are far less pronounced in D. carrolli, suggesting that a transition toward a strongly sexually dimorphic expression of eloF has occurred in D. prolongata.

To test whether the evolution of sexual dimorphism in eloF expression was tissue-specific, we also included in our qPCR study the heads of the same flies from which oenocyte samples were collected. In the brain, eloF shows little, if any, expression in either sex [79]. We found consistently low but detectable levels of eloF transcripts in the heads, which were several orders of magnitude lower than in oenocytes (Fig. 6B) and could be due to the presence of fat body tissue in the head. We also found that eloF expression was significantly higher in D. prolongata male heads than in the other groups (p < 0.05), resulting in a sexually dimorphic pattern in D. prolongata (27-fold difference) and a sexually monomorphic pattern in D. carrolli. Therefore, male-biased eloF expression is not entirely limited to oenocytes, although the extent of sexual dimorphism is much greater in oenocytes than in the head.

Loss of eloF partially feminizes the pheromone profile of male D. prolongata

Increased expression of eloF correlates with the increased abundance of long-chain CHCs in male D. prolongata. The D. melanogaster eloF, which is a 1:1 ortholog of the D. prolongata gene, encodes a bona fide fatty acid elongase sufficient to elongate fatty acids in yeast heterologous expression assays, whereas its RNAi knockdown reduces the amount of female-biased long-chain hydrocarbons in D. melanogaster [31]. This suggests that evolutionary changes in eloF expression could be responsible for the male-specific increase in the abundance of the 9P and 9H pheromones in D. prolongata. To test this hypothesis, we generated two loss-of-function eloF mutants (eloF[-]) in D. prolongata using CRISPR/Cas9 mutagenesis: an early frameshift resulting in a likely null allele and a 45 bp in-frame deletion, which disrupts a predicted transmembrane domain of EloF that is conserved with multiple mammalian fatty elongases [31] and may affect protein localization (Fig. 7A). Gas chromatography and mass spectrometry (GC–MS) analysis of these mutants and their wild-type progenitors showed that, qualitatively, they contained the same CHCs that were previously reported in wild-type D. prolongata (Additional file 6: Table S2). The one exception is a minor alka-diene constituent, x,y-tricosadiene, which is shared between sexes and is not fully characterized.

We observed a strong feminization of male pheromone profiles in both eloF[-] mutants (Fig. 7B, C; Additional files 7–8: Fig. S5-6). In contrast, only a subtle effect is seen in females (Fig. 7C). Consistent with its molecular function, eloF[-] flies show decreased production of long-chain hydrocarbons, which is much more pronounced in males than in females (Fig. 7B; Additional files 7–8: Fig. S5-6). In males, we found a ~ 50% reduction in the amount of 9P and a near absence of 9H. Concurrently, eloF[-] males show increased abundance of 9T to a level comparable to wild-type females (Fig. 7B; Additional file 7: Fig. S5). Since 9T is an early terminal product derived from a common metabolic precursor with 9P and 9H during carbon chain elongation, the increased abundance of 9T may be a direct consequence of reduced 9P and 9H synthesis. Notably, the total abundance of all CHCs as well as that of 9-monoenes remained unchanged (Additional file 9: Fig. S7), indicating that disruption of eloF inhibits elongation of specific male-biased pheromone precursors without having a general inhibitory effect on CHC synthesis. While the degree of sexual dimorphism is reduced in eloF[-] mutants, CHC profiles remain dimorphic (Fig. 7C). This incomplete feminization suggests that other FAE or FAR genes, which also show strongly male-biased expression in D. prolongata (Fig. 4; Fig. 5A) may act in parallel with eloF in the production of long-chain CHCs.

eloF[-] mutations did not significantly affect courtship or copulation in single male–female pairs (Additional file 10: Table S3) despite reduced 9H and 9P abundance in males. Instead, they affected male-male interactions, although the effects were not consistent between the two eloF[-] mutants. Males with the 45 bp deletion in the transmembrane domain of eloF showed increased rate of boxing, a typical male-male aggressive behavior in D. prolongata (Additional file 10: Table S3), whereas decapitated males with the early stop codon elicited higher frequency of misdirected courtship from other males (Additional file 10: Table S3). These results suggest that other signals must be involved in male-male communication alongside 9H and 9P.

The elongase gene cluster including eloF is conserved

We compared the genomic neighborhood of the eloF locus between the rhopaloa species subgroup (D. prolongata, D. fuyamai, D. kurseongensis, D. rhopaloa, and D. carrolli), its nearest outgroup D. elegans, and D. melanogaster. In D. melanogaster, eloF is part of a ~ 10 kb cluster with four other fatty acid elongases, which likely evolved by tandem duplication [139]. We found the same five predicted elongases, in the same order and orientation and with the same exon/intron structure, in all species of our focal clade (Fig. 5B), indicating deep origin and strong conservation of the elongase cluster.

Despite the strong evidence of sex- and species-specific regulation of eloF (Fig. 2; Fig. 5A; Fig. 6A), it is possible that changes in EloF protein activity contribute to the derived male-specific CHC profile seen in D. prolongata. To test this hypothesis, we compared the coding sequences of eloF between D. prolongata and the other four species of the rhopaloa subgroup, which are sexually monomorphic in the abundance of 9P and 9H [98]. There is a high overall degree of protein sequence conservation (> 90%) in the coding region (Additional file 11: Fig. S8), and the protein identity between D. prolongata and D. carrolli is 96.5%. While we found 20 single nucleotide variants (SNVs) distinguishing the reference genomes of these two species, with 9 of them resulting in predicted amino acid substitutions, our RNA-seq data show that all these substitutions are polymorphic in one or both species and none are fixed between species (Additional file 11: Fig. S8, Additional file 12: Table S4). In the absence of fixed amino acid differences between D. prolongata and D. carrolli, coding sequence divergence in eloF is unlikely to contribute to the evolution of CHC profiles.

A species-specific transposable element insertion in eloF in D. prolongata

Our evidence points to changes in eloF transcription as the main cause of sex-specific pheromone profiles in D. prolongata. To identify the likely cis-regulatory elements of eloF, we examined the flanking intergenic and intronic regions of eloF in the rhopaloa subgroup. The most drastic difference between D. prolongata and all other species is a ~ 900 bp insertion in the otherwise conserved (~ 500 bp, > 72% sequence identity) downstream region of eloF (Fig. 5). The inserted sequence contains two predicted binding sites for the doublesex (dsx) transcription factor, the main regulator of somatic sexual differentiation in Drosophila and other insects [64, 77, 157], as well as several predicted binding sites for bric-à-brac 1 (bab1), a TF that regulates the development of abdominal segments [78, 118] (Additional file 13: Fig. S9). These sites, along with the rest of the insertion, are absent in the other species. Our inspection of the single-cell Fly Cell Atlas [90] shows that both TFs are expressed in adult oenocytes, and that the upstream regions of oenocyte-biased genes show a significant enrichment for bab1 binding motifs (p = 1.75e-60). These observations suggest that the insertion in the 3’ region of eloF may have contributed to the species- and sex-specific increase in eloF expression in D. prolongata.

We found hundreds of highly similar copies of this insertion throughout the genome of D. prolongata (Additional file 14: Table S5), suggesting that it may be a transposable elements (TE). We named this putative TE “honghaier” after the mythical Chinese character capable of self-duplication. honghaier is found in high copy numbers in all five species of the rhopaloa subgroup, but not in D. elegans or D. melanogaster (Additional file 14: Table S5), suggesting that it originated or invaded this lineage relatively recently. honghaier is AT-rich (~ 60%, Additional file 13: Fig. S9), a common feature of miniature inverted-repeat transposable elements (MITE), and has a 414 bp open reading frame (ORF), which is predicted to be transcribed in the direction opposite to eloF (Additional file 13: Fig. S9). The strongest sequence similarity between honghaier and known TEs is found with DNAREP1_DM (53%) [71] and wukong (52.5%), a mosquito MITE element [147]. However, honghaier is unlikely to be a true MITE, as these elements usually do not have coding potential [147]. honghaier also lacks typical hallmarks of TEs such as terminal inverted repeats (TIRs) and flanking short direct repeats stemming from target site duplication [125].

Although changes in gene expression caused by TE insertions are common [11, 40, 63, 137, 146], and the honghaier insertion in eloF correlates with its divergent expression profile in D. prolongata, we cannot rule out contributions from other cis-regulatory changes. There are multiple fixed SNVs between D. prolongata and D. carrolli in the upstream (~ 300 bp, Additional file 15: Fig. S10) and intronic (~ 70 bp, Additional file 16: Fig. S11) regions of eloF, despite overall high conservation (94.3% for upstream and 96.9% for intron). However, these substitutions do not affect any predicted binding motifs for dsx, bab1, or other TFs known to regulate abdominal development or sexual differentiation.

eloF downstream region drives gene expression in oenocytes

To test whether increased expression of eloF in D. prolongata is due to changes in the cis-regulatory regions of eloF, we generated transgenic GFP reporter strains where the eloF loci from D. prolongata and D. carrolli were transformed into D. melanogaster. First, we cloned the entire eloF region between the flanking genes CG16904 and CG8534 (Dpro eloF WT^(l) and Dcar eloF WT^(l)). In these constructs, the eloF transcript is in the same orientation as the GFP reporter, while the honghaier insertion is in the opposite orientation (Fig. 6C, Additional file 17: Fig. S12 A-B). To examine the effects of the honghaier insertion, we also made two TE-swap constructs, one with honghaier removed from D. prolongata (Dpro eloF WT^(l)-TE) and the other with honghaier added to D. carrolli (Dcar eloF WT^(l) + TE) (Fig. 6C, Additional file 17: Fig. S12B). We observed little, if any, GFP expression by qPCR (Additional file 17: Fig. S12E; Additional file 18: Table S6). In females, GFP transcripts were not detectable (Ct > 40), while in males they were present at very low levels (Ct > 35). We then cloned only the downstream regions of eloF, generating both wild-type reporters (Dpro eloF WT^(s) and Dcar eloF WT^(s)) and TE-swap constructs (Dpro eloF WT^(s)-TE and Dcar eloF WT^(s) + TE). These constructs were designed so that the downstream eloF sequences were upstream of the GFP reporter, and the honghaier insertion in the forward orientation relative to the promoter (Additional file 17: Fig. S12 C-D).

In the adult dorsal abdominal epidermis of transgenic flies carrying short eloF reporters, we observed GFP expression in both sexes in stripes of tissue in the posterior half of each segment (Additional file 19: Fig. S13). This region corresponds to the location of the pheromone-producing oenocytes [16], suggesting that the downstream region of eloF contains an oenocyte-specific enhancer.

eloF allele from D. prolongata drives sexually dimorphic expression in D. melanogaster

As both D. prolongata and D. carrolli express eloF in the abdomen, we expect the differences in enhancer activity to be more quantitative than qualitative. We therefore compared transgenic reporter activity by qPCR. In the long constructs, which contained the eloF coding sequence, we compared the eloF transcript levels. The D. carrolli allele was expressed at similar levels in males and females (Fig. 6D). The D. prolongata allele was expressed at a ~ 20-fold higher level than the D. carrolli allele and showed significant sexual dimorphism (Fig. 6D). Surprisingly, the D. prolongata allele was expressed at a higher level in females compared to males. While this direction is opposite to what is observed at the endogenous eloF locus in D. prolongata, it matches the phenotype of D. melanogaster, in which eloF expression and long-chain CHC abundance are higher in females than in males [31]. This indicates that while the D. prolongata eloF allele, in contrast to the D. carrolli allele, encodes sex-specific regulatory information, its effect on transcription depends on the trans-regulatory background, which appears to have diverged between D. prolongata and D. melanogaster. The removal of the honghaier insertion from the D. prolongata allele, or the addition of this insertion to the D. carrolli allele, eliminated the differences in reporter activity both between species and between males and females (Fig. 6E), suggesting that this insertion is necessary, but not sufficient, for driving sex-specific expression of eloF.

We then used the short reporter constructs to compare GFP transcript expression driven by the wild-type and TE-swapped alleles of the downstream eloF region that contains the honghaier insertion in D. prolongata. We observed a modest (twofold) but significant sexual dimorphism, also in the direction of females having higher expression (Additional file 17: Fig. S12 F). However, GFP expression was low in both sexes (~ 29 Ct), and there was no significant difference between the D. prolongata and D. carrolli alleles in either sex (Additional file 17: Fig. S12 F), suggesting that the downstream region and the honghaier insertion alone are not sufficient to confer species-specific transcriptional regulation, at least in the D. melanogaster genetic background. Alternatively, it is possible that eloF enhancers are sensitive to the sequence, position, and relative orientation of the interacting promoter.

Discussion

In this study, we show that sexually dimorphic pheromones affect mating behavior in D. prolongata and identify a key gene underlying the evolution of sex-specific pheromone profiles in this species (Fig. 8). A cis-regulatory change in the eloF gene is an important, though not the only, component of the genetic changes that distinguish D. prolongata from its close, sexually monomorphic relatives. Below, we discuss these findings in the context of our still limited but growing knowledge of the evolution and functional roles of Drosophila pheromones.

Male-specific hydrocarbons reduce female mating success

Sex-specific visual, acoustic and chemical cues play vital roles in mate recognition. The divergence of communication systems helps maintain species boundaries and can drive the evolution of reproductive isolation, as seen in the coevolution of nuptial colors and color vision in sticklebacks [22], wing color patterns and co-evolved mate preferences in Heliconius butterflies [70], matching conspecific mating duets sung by male and female lacewings [152], or the divergent pheromone blends between two sympatric races of the European corn borer [92]. In Drosophila, sexually dimorphic CHCs mediate mate recognition and allow males to differentiate potential mates from competitors [65]. For example, in D. melanogaster, the male-biased 7-tricosene (7 T) evokes male-male aggression, whereas the female-biased 7,11-heptacosadiene (7,11-HD) elicits courtship behavior even when applied to a dummy female [53, 68].

The male-biased 9P and 9H in D. prolongata may serve as one of the cues that facilitate mate recognition, though other signals including visual cues are clearly important [140]. Our perfuming studies show that 9H, and to a lesser extent 9P, reduce mating success when applied to females. This reduction is not due to a lack of courtship interest, but could instead be related to reduced leg vibration, a species-specific behavior performed by D. prolongata males that increases female receptivity [127]. This suggests that the lower relative amounts of 9P and 9H in females compared to males are important for the proper progression of male courtship toward females. Although this effect is relatively subtle in the no-choice assays that pair a single male with a single female, it could be more significant in nature, where a single male is choosing among multiple females and vice versa. The neurophysiological mechanisms by which 9P and 9H influence male behavior remain to be identified. An obvious hypothesis is that they are perceived by gustatory organs that are present in unusually large numbers on the front legs of D. prolongata males [96].

Identifying other functions of 9P and 9H is complicated both by the fact that eloF mutations do not fully block the synthesis of these compounds, and by the complex mix of visual, chemical, and auditory signals that mediate Drosophila mating behavior. Some pheromones, such as cis-vaccenyl-acetate (cVA), are transferred from males to females during mating, and function as a post-mating anti-aphrodisiac signal [46, 53, 69, 108]. However, we find no evidence that D. prolongata males transfer 9P or 9H to females, suggesting that the long-chain CHCs are unlikely to act by reducing female re-mating. This finding is not surprising, since many sex-specific CHCs are not transferred during mating [73]. The effect of long-chain CHCs on male-male interactions appears to be limited. While we observe an increase in male-male aggression and misdirected courtship toward males, these effects are inconsistent between the two mutant alleles of eloF although both alleles reduce the abundance of 9H and 9P and increase 9T levels.

Beyond intraspecific communication, 9P and 9H could contribute to sexual isolation between sibling species, similar to the roles of 7 T and 7,11-HD in the isolation between D. melanogaster and D. simulans [68, 124]. Pre-mating isolation between D. prolongata and its relatives is strong; we have never observed an interspecific mating. Female D. prolongata could potentially use a lack of 9P or 9H to reject mating attempts from males of other species, although, as in the intraspecific communication, this would likely be only one of several cues. Compared to all of its relatives, D. prolongata has highly derived male mating behavior and greatly exaggerated sexual dimorphism in multiple traits, including reversed sexual size dimorphism, pigmentation, and the organization of the chemosensory system [96, 97, 127]. In this context, deciphering the behavioral and ecological roles of 9P and 9H may elucidate why a strongly male-biased pheromone profile has evolved in D. prolongata but not in any of its close relatives.

eloF is a major gene controlling long-chain pheromone production in male D. prolongata

Our results show that the evolution of sexually dimorphic pheromone profiles in D. prolongata is due to a large extent to changes at the eloF locus (Fig. 8). eloF mutations have a particularly strong effect on the elongation of C25 to C27, and a somewhat milder effect on the elongation of C23 to C25. eloF is a well-characterized fatty acid elongase that has been shown to catalyze the conversion of long-chain fatty acyl CoA to very long-chain fatty acyl CoA in yeast assays [31]. In D. melanogaster, long-chain hydrocarbons are enriched in females [49, 51], and knocking down eloF expression in D. melanogaster elicits a female-specific reduction in their abundance. However, ectopic expression of eloF in D. melanogaster males does not increase the abundance of long-chain CHCs, indicating that eloF is necessary but not sufficient for their synthesis [31]. Unlike other genes whose disruption leads to an overall increase or decrease in CHC production [43, 114, 155], we show that eloF mutations in D. prolongata alter the relative abundance of short vs. long-chain monoenes without affecting total monoene amounts, or the amounts of CHC more generally.

In principle, increased EloF activity in D. prolongata could be due to either regulatory or coding sequence changes. Even small differences in protein sequence can have a major effect on enzyme function, with drastic phenotypic consequences [59, 107, 151]. However, we find no fixed coding sequence differences in eloF between D. prolongata and its sibling species D. carrolli, in which the amounts of 9P and 9H are nearly monomorphic. On the other hand, D. prolongata shows extreme sexual dimorphism in eloF transcript abundance as well as overall higher eloF expression relative to D. carrolli. In D. prolongata, eloF expression is > 3,000-fold higher in males than in females, while only a fourfold difference between the sexes is detected in D. carrolli. These observations indicate that increased 9P and 9H production in D. prolongata males is due to changes in eloF expression rather than EloF protein activity (Fig. 8).

Interaction of cis- and trans-regulatory factors in the control of sex-biased eloF expression

The eloF allele of D. prolongata, but not D. carrolli, drives sexually dimorphic gene expression in D. melanogaster, suggesting the presence of cis-regulatory elements that respond to the sexual differentiation pathway. This pathway, including the doublesex (dsx) transcription factor and the transformer (tra) RNA-binding protein that controls its sex-specific splicing, is the primary mediator of sex-specific cell differentiation in somatic tissues [64, 77, 157]. Across a number of Drosophila species, the binding of Dsx to the regulatory region of the desatF (Fad2) gene is responsible for female-specific expression of that gene in adult oenocytes, and thus for the female-specific production of the 7,11-HD pheromone [128]. The D. melanogaster eloF ortholog, which also contributes to the synthesis of female-specific pheromones, has also been shown to be under the control of tra [31]. Overall, however, the regulatory program that controls sex-specific differentiation of oenocytes remains to be characterized.

Perhaps the most surprising part of our results is that the direction of sex bias was reversed in reporter assays. Instead of recapitulating the male-biased expression of eloF seen in D. prolongata, the D. prolongata eloF gene is expressed in a female-biased fashion when placed into the D. melanogaster genome. That is, the direction of sex bias replicates the pattern seen in the D. melanogaster host [31] and not in the D. prolongata donor. This indicates that, while the eloF locus itself encodes the potential for sexually dimorphic expression, the realization of this potential depends on the trans-regulatory background, which has clearly diverged between D. prolongata and D. melanogaster. The mechanistic basis of this reversal is not clear. It could indicate either that the regulation of eloF by the sexual differentiation pathway is indirect, or that dsx interacts with other transcription factors whose expression differs between species.

The most conspicuous sequence change at the eloF locus is the presence of a species-specific insertion of the honghaier TE-like element in D. prolongata. TEs are a major source of cis-regulatory changes underlying the evolution of gene expression [5, 24, 36, 62, 99, 137, 146], so it is tempting to speculate that the honghaier insertion is at least partly responsible for the increased expression of eloF in D. prolongata males. Consistent with this idea, the honghaier element contains predicted binding sites for the dsx and bab1 transcription factors. However, the results of reporter assays defy a simple explanation, as the downstream eloF region that contains the honghaier insertion in D. prolongata is not sufficient to confer sex- and species-specific expression observed in the longer reporters. Moreover, swapping the honghaier insertion between the D. prolongata and D. carrolli alleles shows that this insertion is necessary but not sufficient for species- and sex-specific expression. This suggests that the downstream region may interact with other parts of the eloF locus or the wider elongase cluster. Although enhancers are generally modular [30], numerous exceptions are known where interactions among several regions within the locus are necessary for correct gene expression [96, 106]. Another, not mutually exclusive explanation is that eloF enhancers are promoter-specific – that is, their activity depends on the sequence and the relative position and orientation of the interacting promoter [14, 83, 109].

Finally, we note that eloF is part of a compact genomic cluster with four other elongases, and that all five genes show strongly male-biased expression in D. prolongata but not in D. carrolli. This raises the possibility that their expression is controlled in part by shared cis-regulatory elements, and that some of the enhancers that control eloF expression may be located outside of the immediate eloF locus. Co-regulation of clustered genes is not uncommon. It contributes, for example, to the co-regulation of Hox [57] and Iroquois (Irx/iro) [142] genes in vertebrates, while in Drosophila shared enhancers contribute to the concerted expression of bab, pdm, and other closely linked genes [23, 94].

Multiple genes likely contribute to sex-specific pheromone profiles in D. prolongata

Disruption of eloF causes a strong, but not complete feminization of the male CHC profile. In particular, while the longest-chain CHC, 9H, is almost absent in eloF mutants, the most abundant male-biased pheromone 9P shows only a ~ 50% reduction in mutant males, and remains the most abundant CHC component. Since one of the mutant alleles, a premature stop codon early in the first exon, is almost certainly a molecular null, this suggests that other genes must contribute to the synthesis of 9P. Consistent with this, eloF knockdown in D. melanogaster reduces the amount of very long chain CHCs, but does not fully eliminate them [31, 156].

RNA-seq shows that sex-specific expression of lipid metabolism genes in D. prolongata is not restricted to eloF. All four other elongases in the genomic cluster that contains eloF show strongly male-biased expression, as do many other enzymes that function in lipid metabolism. Unfortunately, most of these candidate genes have not been characterized nearly as well as eloF. In the elongase cluster, only CG9458 has been studied so far, and its disruption does not apparently impact the elongation of pheromone precursors in D. melanogaster [43]. One possibility is that there is a degree of functional redundancy among the five clustered elongases, so that a simultaneous elimination of several (or all) of them would be required to feminize the male CHC profile more completely.

Another class of genes that may contribute to the sex-specific pheromone profiles of D. prolongata are fatty acid reductases (FARs). Enzymes in this large (17 genes in D. melanogaster) but poorly characterized gene family control the reduction of fatty acyl CoA to aldehydes and alcohols before they are converted to hydrocarbons by decarbonylation [114, 154, 159]. In moths, natural variation in FAR genes is responsible for the divergence of pheromone blends between populations and species [85, 91], while in Drosophila serrata, FAR2-B, a recently duplicated ortholog of the D. melanogaster CG17560, explains sexually antagonistic variation in the relative amounts of short-chain and long-chain hydrocarbons [119]. In D. prolongata, the set of top candidate genes includes five FARs, including the ortholog of CG17560/FAR2-B (Fig. 4). Four of these genes are upregulated in D. prolongata males compared both to D. carrolli and to D. prolongata females, while the fifth, CG17562, is downregulated in D. prolongata males. The functional significance of these differences is unknown at this point. Since FAR-controlled decarboxylation competes with FAE-controlled elongation in determining whether precursors give rise to terminal products (pheromones) or to longer precursors with additional carbons (Fig. 8), one possible explanation is that the reduction of CG17562 in D. prolongata males facilitates elongation of 9T precursors to 9P/9H precursors instead of direct production of 9T. More generally, at least some FARs appear to be broad-spectrum enzymes: changes in, or disruption of, a single gene can alter the relative abundance of multiple long- and short-chain CHCs [43, 85, 91, 119]. At the same time, FARs, like elongases, show some level of substrate specificity [34]. It is possible that the FARs with male-biased and female-biased expression in D. prolongata have different substrate specificity (which may also vary among species), and that changes in their relative expression in males vs females contribute to the production of sex-specific pheromone blends. Genetic and biochemical evidence will be needed to test this hypothesis.

Conclusions

Our results show that cis-regulatory changes in the eloF gene, along with other changes in the CHC synthesis pathway, contribute to the evolution of sexual communication. The exchange of signals involved in mating behavior is exceedingly complex. eloF is only one of the genes responsible for sex-specific pheromone profiles, while sex-specific pheromones are only one of the sensory cues underlying male-male and male–female communication. Putting together the complete puzzle will require not only identifying the missing pieces but also understanding how they interconnect.

Methods

Fly rearing and dissection for gene expression analysis

Flies were raised on standard cornmeal media and kept at room temperature (20–22 °C) under natural light–dark cycle. For behavioral experiments and CRISPR gene editing, we used the reference genome strain of D. prolongata [95], which was derived by four generations of full-sib inbreeding from the SaPa strain collected by Dr. H. Takamori [98]. For RNA-seq and qPCR experiments, we used the BaVi strain of D. prolongata and the reference genome strain of D. carrolli [98]. Both D. prolongata strains show strongly sexually dimorphic pheromone profiles, with females consistently distinguishable from males (F-type, [98]). For RNA-seq analysis, four biological replicates were prepared for each sex of D. prolongata and D. carrolli, resulting in 16 libraries. For qPCR experiments, three biological replicates were prepared for each species and sex. To obtain tissue samples enriched for oenocytes, we dissected the dorsal abdominal body wall (“cuticle fillet”) as described [16]; these samples are referred to as oenocyte dissections hereafter. Each biological replicate contained ten cuticle fillets. For head dissections, each biological replicate contained ten heads coming from the same tissue donors as the oenocyte dissections. All flies used for gene expression analysis were isolated as virgins and aged for seven days (D. prolongata and D. carrolli) or five days (transgenic D. melanogaster). Unless noted otherwise, tissues were dissected in chilled Shields and Sang M3 Insect Medium (Sigma-Aldrich, St. Louis, MO).

RNA extraction

Total RNA was extracted from dissected fly tissues following the Trizol protocol (Ambion, Carlsbad, CA). Purified RNA was pelleted by isopropanol overnight at −20 °C, washed by freshly made, pre-chilled 70% Ethanol (EtOH) 2 times, and dissolved in 20 µl of DEPC-treated water (Ambion, Carlsbad, CA). To mitigate batch effects, flies were collected from the same food bottle, and flies collected on different dates were evenly distributed to each Trizol-containing tube. To ensure purity (A260/A280 > 1.9, A260/A230 > 1.5), isolated RNA was analyzed on a Nanodrop Spectrophotometer (ND-1000) using the software Nanodrop 1000 3.8.1. To ensure the integrity of RNA (2 sharp peaks of ribosomal RNA), gel electrophoresis was performed on an Agilent 2100 Bioanalyzer (Agilent) using RNA Nano Chips (Agilent). The concentration of RNA was determined using a Qubit 2.0 Fluorometer (Invitrogen) and Broad Range RNA Assay kit (Life Technologies). Finally, total RNA was DNase treated to remove carry-over genomic DNA following the rigorous DNA removal recommendations of the Turbo DNA-free kit (Invitrogen, Carlsbad, CA).

Library construction, sequencing, and read mapping

cDNA libraries for RNA-seq were made using TruSeq Stranded RNA kit (Illumina, San Diego, CA) following the low throughput (LT) procedures in the user manual. 500 ng of total RNA was used as starting material, and mRNA was selected by polyT enrichment. Reverse-transcribed cDNA was ligated with adapters uniquely barcoded for each library, followed by 11-cycle PCR amplification in an Applied Biosystems 2720 Thermal Cycler (Applied Biosystem, Waltham, MA). Thermocycling conditions were set as follows: 98 °C for 30 s, 11 cycles of 98 °C for 10 s, 60 °C for 30 s, 72 °C for 30 s, and a final 72 °C for 5 min. Amplified fragments were analyzed on an Agilent 2100 Bioanalyzer (Agilent) using High Sensitivity DNA Chips (Agilent). Unimodal fragment size distribution was consistently observed in all libraries, with a median fragment size of ~ 300 bp. The resulting cDNA libraries were initially quantified by Qubit 2.0 Fluorometer (Invitrogen) using the DNA High Sensitivity kit (Life Technologies) and further quantified by qPCR using the KAPA Library Quantification kit (Roche, Cape Town, South Africa). Barcoded cDNA libraries were subsequently pooled in equal molar ratios. To mitigate batch effects, all 16 libraries were prepared on two consecutive days, and within each day, two biological replicates of each group were processed.

Multiplexed libraries were sequenced on HiSeq4000 on PE-150 mode by Novogene (https://www.novogene.com). Reads were preprocessed (quality trimmed and deduplicated) by HTStream [111]. Cleaned reads were aligned to species-specific reference genomes [74, 95] using STAR (version 2.7.3a, [44]) with the following flags: –sjdbOverhang 149 –genomeSAindexNbases 13 –genomeChrBinNbits 18 –quantMode GeneCounts. Feature annotations were predicted by combining MAKER pipelines [29] and Liftoff [129] from existing genomic features of D. melanogaster (release 6.36) and D. elegans (Gnome annotation version 101). RNA-seq data and alignment statistics are summarized in (Additional files 20–21: Tables S7 and S8).

Differential gene expression analysis

Paired-end fragments data were extracted from concordant read pairs. R packages"limma"[115] and"edgeR"[116] were used to detect differentially expressed genes. To account for variations in sequence depth and RNA compositions between samples, sample-specific normalization factors were calculated using the Trimmed Mean of M-values (TMM) method [117]. For each gene, counts per million (cpm) were computed. To remove genes with low expression, those with less than 2 cpm across all samples were excluded. Genes that could not be identified in both D. prolongata and D. carrolli (~ 400 genes) were also excluded, resulting in a set of 9143 genes. To obtain normalized expression data, TMM-based normalization factors were applied, followed by calculating the log₂ cpm of these genes. To account for the mean–variance trend associated with each gene (e.g., genes with low mean expression tend to have larger variances), voom transformation was applied to estimate the weights of genes [88]. These weights were then used to fit a weighted linear model on normalized expression data (log₂ cpm), using groups (sex x species) as predictors.

Three one-way comparisons were defined to identify candidate genes for pheromone divergence: (1) contrasting D. prolongata males with D. prolongata females; (2) contrasting D. prolongata males with D. carrolli males and (3) comparing the magnitude of the male–female difference between D. prolongata and D. carrolli. This last test helps to identify changes that result either from differences in the magnitude of sexual dimorphism in each species (i.e., which one shows greater male–female difference?) or the direction of sexual dimorphism in each species (i.e., is the direction of sex-biasedness the same?). Linear contrasts were made for each one-way comparison. To account for variance that comes from random factors (not low expression), we used empirical Bayes smoothing [133]. To adjust for multiple testing, false discovery rate (FDR) corrections were applied to raw p values [12], and significant differentially expressed genes (DEGs) were reported with FDR < 0.05 (Additional files 22–27: Data Files 1–6).

To be considered candidates, genes had to be reported as significantly different in all three one-way comparisons (referred to as the three-way comparison hereafter). P values for the three-way comparison were constructed by taking the maximum of p values from the one-way comparisons, resulting in 53 candidate genes (p < 0.05, Fig. 2). To visualize the expression profiles of top candidates, hierarchical clustering was performed on their standardized expression levels. Euclidean distances were calculated, and UPGMA (i.e., average linkage) was used for the hierarchical clustering. The same metrics were used to cluster both samples and genes. Volcano plots (Fig. 2) were generated by the"ggplot2"package, and heatmaps (Fig. 3) by the"stats"package, both of which were subsequently polished by the Inkscape software (https://inkscape.org).

Gene Ontology (GO) enrichment analysis

GO enrichment analysis was performed for DEGs identified from each one-way comparison (i.e., male vs. female in D. prolongata, males of D. prolongata vs. D. carrolli, and the magnitude and direction of sexual dimorphism in D. prolongata vs. D. carrolli). To determine whether these DEGs were expressed in oenocytes ancestrally, we consulted the single-cell Fly Cell Atlas data from D. melanogaster [90]. ~ 3000 cells annotated as adult oenocytes (FBbt:00003185) were retrieved from the"10 × relaxed dataset"on SCope (https://scope.aertslab.org/#/FlyCellAtlas/FlyCellAtlas%2Fr_fca_biohub_oenocyte_10x.loom/gene). Oenocyte expressors were defined as genes that were detected in at least ten cells (i.e., at least one transcript in each of 10 cells) and had at least 50 cumulative read counts in either female or male samples. This produced a list of ~ 6200 oenocyte expressors (Additional file 28: Data File 7).

We annotated GO terms with the following criteria. For genes annotated with D. melanogaster CG/CR numbers, their GO annotations from the R Bioconductor package"org.Dm.eg.db"(version 3.14.0) were used. This provided GO annotation for 7741 genes. For genes annotated with D. melanogaster CG/CR numbers that did not have GO annotations in"org.Dm.eg.db", their GO annotations were retrieved from orthoDB (version 10.1, https://www.orthodb.org). This provided GO annotation for another 314 genes. For genes annotated with D. elegans LOC numbers, their GO annotations were retrieved from orthoDB (version 10.1). This provided GO annotation for additional 529 genes. In this way, a gene-GO map was built to cover 93.8% (8584) of the entire gene set used in the RNA-seq analysis.

R Bioconductor package"TopGO"(version 2.46.0, [3]) was used to perform enrichment analysis on Biological Processes GO terms. To control for potential artifacts due to small GO categories, those with < 10 associated genes were excluded, as recommended by the program. To account for the tree topology between GO terms, the modified elimination algorithm weight01 [4] was used. By doing this, parent nodes of significant child nodes are less likely to be annotated unless they contain substantially more significant genes not covered in their children. This also helps balance a low false positive rate and a high recall rate. P values were obtained from Fisher exact test [45] and the Kolmogorov–Smirnov test [1]. No p-value adjustment was performed as recommended by the program developer [3]. Instead, candidate GO terms were defined as those with p values < 0.05 for both Fisher and Kolmogorov–Smirnov tests.

Quantitative polymerase chain reaction (qPCR) analysis of gene expression

To quantify the expression of endogenous eloF in D. prolongata and D. carrolli, two-step qPCR was performed (i.e., cDNA synthesis followed by separate qPCR analysis). cDNA was synthesized from 1 µg of DNase-treated total RNA using Superscript III (Invitrogen) following kit recommendations. To prime the reverse transcription reaction, a volume ratio of 1:1 random hexamer (Invitrogen) and oligo dT (Invitrogen) was used. Reactions were performed in a thermal cycler (Applied Biosciences) with the following conditions: initial incubation at 25 °C for 5 min, reverse transcription at 50 °C for 50 min, and enzyme inactivation at 70 °C for 15 min. The resulting single-stranded cDNA was diluted by a factor of 100 and stored at −20 °C prior to qPCR.

Green Fluorescence Protein (GFP) expression in reporter assays (see “Design of reporter constructs” below) was quantified using one-step RT-qPCR (i.e., combining reverse transcription and PCR amplification in the same tube). This was done because the reporter GFP is a single-exon gene (Additional file 17: Fig. S12), so that amplification of GFP transcripts could be confounded by even trace amounts of GFP DNA. The entire experiment was conducted on a clean bench free of DNA contaminants, and PCR-grade water (IBI Scientific, Dubuque, IA) was used to assemble the reaction. As GFP expression was preliminarily found to be low, 300 ng DNase-treated RNA was used for each reaction. To prepare no-reverse-transcriptase (NRT) controls, total RNA samples from 3 biological replicates were pooled in equal mass ratios and received the same treatment. All NRT controls showed Ct > 35 (Additional file 18: Table S6), indicating sufficient removal of genomic DNA.

qPCR reactions were assembled using SsoAdvanced SYBR Green PCR Supermix kit (Bio-Rad, Hercules, CA) on Bio-Rad CFX96 Real-time PCR system. Amplification was performed in 10 µl total volumes with a 4 µl template (1:100 diluted cDNA or 300 ng DNase treated total RNA) and 100 nM of each primer in a 96-well optical plate (Bio-Rad). Melt-curve analysis was performed on the PCR products to assess the presence of unintended products. Thermocycling conditions are set as follows. For qPCR: initial denaturing at 95 °C for 1 min, followed by 40 cycles of 95 °C for 10 s and 60 °C for 10 s. For RT-qPCR: Reverse transcription at 50 °C for 10 min, initial denaturing at 95 °C for 1 min, followed by 40 cycles of 95 °C for 10 s and 60 °C for 10 s. For melt-curve analysis: from 65 °C to 95 °C at an increment of 0.5 °C, hold 5 s for each temperature step. For both one-step or two-step qPCR, three biological replicates were made for each group, and each reaction was technically replicated three times to obtain an average Ct value. Technical reproducibility was consistent with standard deviations within 0.5 Ct [123]. Ribosomal protein L32 (Rpl32) was chosen as a reference gene for its stable expression level [113]. Standard curves were built to determine primer amplification performance (e.g., primer efficiency) (Additional file 29: Fig. S14). Specifically, qPCR was performed on a diluted DNA template covering at least 6 log range. qPCR amplification metrics were determined for each gene with the slope of a linear regression model [112]. Relative efficiencies were calculated according to the equation: \(\text{E }= (\text{dilution factor }-1/\text{slope }- 1)\text{ x }100\text{\%}\). Primer sequences, design considerations, coefficient of determination (R²), and amplification efficiencies are summarized in Additional file 30: Table S9. As all primers had near-perfect amplification efficiency, the ΔΔCt method [93] was used for the relative quantification of genes of interest (eloF and GFP).

To model normalized expression levels, a two-way ANOVA with interaction effects between genotypes (species) and sex was used, similar to the section"Statistical analysis of mutant CHC profiles."Statistical significance for genotype, sex, and their interactions was tested by comparing the full model with a reduced model after dropping the term of interest. Type III variance partitioning was used, and Tukey's method was used to determine which construct has a significantly higher expression level.

Cuticular lipids extraction

Virgin D. prolongata with wild-type or mutant eloF were individually isolated within 12 h after eclosion. After aging for 7 days, individual flies were frozen at −20 °C, transferred to pure hexane (Sigma-Aldrich), soaked for 5 min at room temperature, and vortexed for 30 s. To ensure complete CHC extraction, 40 µl of pure hexane was used for females and 80 µl for males, due to the large size difference. Crude extracts were air-dried overnight and stored at 4 °C before GC–MS analysis. To quantify the absolute amount of each analyte, hexane containing 10 ng/µl n-heneicosane (nC26, Sigma-Aldrich) and 10 ng/µl n-triacotane (nC30, Sigma-Aldrich) as alkane standards was used to resolubilize crude extracts. 40 µl of this solvent was used for females and 80 µl for males.

Gas chromatography (GC) and mass spectrometry (MS) analyses.

GC–MS analysis was performed as in [98] with the following modifications. The oven temperature was programmed to first ramp from 160 °C to 280 °C at a rate of 8 C/min, hold at 280 °C for 1 min, and increase from 280 °C to 315 °C at a rate of 15 °C/min, followed by a final 1 min hold at 315 °C. The flow rate of carrier gas (helium) was optimized to 1 ml/min. Individual chromatographic peaks were first called using the built‐in ChemStation integrator of MSD ChemStation Enhanced Data Analysis Software vF.01.00 (Agilent Technologies, Santa Clara, CA), with initial peak width of 0.030 and an initial threshold of 16. Manual adjustments were made to include minor peaks and deconvolute overlapping peaks. Analytes were then identified (Additional file 6: Table S2) and quantified as described previously [98]. Briefly, all CHCs were normalized by alkane standards and scaled in units of nanograms per individual fly.

Female perfuming experiments

Synthetic (Z)−9-Pentacosene (9P) was purchased from Cayman Chemical (Ann Arbor, MI), and (Z)−9-Heptacosene (9H) was kindly provided by Dr. Jocelyn Miller (University of California, Riverside). To prepare perfuming vials, batches of hexane solutions containing 9P (9P treatment), 9H (9H treatment), or nothing (control) were added to and air-dried inside 2 mL glass vials (Agilent Technologies, #5182–0715, Santa Clara, CA). 50 µg 9P and 10 µg 9H were used to ensure consistent and biologically reasonable perfuming (Fig. 1). Flies were perfumed according to a modified protocol of [16], briefly summarized as follows. Groups of eight virgin, 7-day old female flies were placed inside clean 2 mL glass vials (Agilent Technologies) and vortexed on medium speed to capture the CHC profile before the perfuming study. To perfume with synthetic hydrocarbons, the same group of 8 flies was subsequently transferred to the perfuming vial prepared as described above and vortexed intermittently. 4 groups of flies were prepared per day, resulting in a total of 8 groups containing 64 individuals. Perfumed flies were allowed to recover for 3 h and divided randomly into two equal groups, with one group of four used immediately for assessing male–female interactions (assay group) and the other saved for confirming the transfer of desired CHCs (validation group). 200 µl of pure hexane was used to extract CHCs from the validation group. Both pre-and post-perfuming crude extracts were resolubilized with alkane standards as described above (see"cuticular lipids extraction"), except that 20 µl was used for pre-perfuming samples (N = 8) and 100 µl for post-perfumed samples (N = 8). To quantify changes in the CHCs of interest, all samples were analyzed by GC–MS as described above.

Behavioral assays

Cameras and the behavior arena were set up as previously described [145]. For male–female interaction experiments, a single virgin female was paired with a single virgin male inside a food podium. For male-male interaction experiments, two virgin males of the same genotype were placed together without any females being present. For misdirected courtship experiments, a pair consisting of one wild-type male and one wild-type female was combined with a single decapitated male, whose genotype was either wild-type or eloF[-]. The genotypes and numbers of individuals are reported separately for each experiment in the figure legends. The flies were videotaped for 1 h, and binary metrics of previously characterized behaviors, including"encounter,""threatening,""courtship,""leg vibration,"“wing vibration”,"copulation", and “boxing” were scored from the video recordings [80, 127]. Courtship was quantified as the proportion of males that continued to court the female after the initial encounter. Leg vibration, whereby the male vigorously shakes the female’s abdomen with his front legs, is a behavior specific to D. prolongata [126, 127].

To test whether male-biased hydrocarbons are transferred to females during mating, 6–8 day-old virgin males and females from the reference genome strain were placed together in single pairs (n = 16). Behavior was observed in the morning for 1 h to determine whether mating occurred. To test for quantitative changes in CHC profiles, whole-body pheromone extractions were performed on mated and unmated females on the same day after the observation concluded. Socially naive females and males were included as controls.

Statistical analysis of behavioral changes

A logistic regression model was used for each binary behavior (e.g., courtship) with the genotype as the only explanatory variable, and an ordinary linear regression model for each continuous behavior (e.g., copulation duration). Z-tests were performed on coefficients from logistic regression to determine the p-value for each comparison between eloF[-] mutant and wild-type alleles. t-tests were performed on coefficients from ordinary linear regression.

Statistical estimation of hydrocarbon transfer

Hydrocarbon transfer was estimated from the increase in the abundance of the analyte of interest (9P or 9H) after perfuming. For each group of 4 females used in behavioral tests, we created a parallel group of 4 females that were subjected to the same perfuming procedure but were not used in behavioral assays (see “female perfuming preparation” above). This replicate group was used to validate the transfer of desired CHCs, in conjunction with pheromones extracted from the same group before perfuming. Instead of simply taking the difference in 9P (or 9H) abundance before and after perfuming, we calibrated the post-perfuming abundance of 9P (or 9H) by a method analogous to standard curves to mitigate technical variation as follows. For each perfuming group of eight flies, a calibration curve was made by regressing the post-perfuming on the pre-perfuming abundances of all CHCs except those modified in treatment (e.g., leaving out 9P in 9P treatment). A general agreement was found between pre-post pairs of endogenous CHCs, with coefficients of determination (R²) ranging from 0.85 to 0.98 (Additional file 31: Fig. S15). Leveraging this property, expected post-perfuming abundance of 9P (or 9H) if no synthetic 9P (or 9H) were transferred (i.e., the"counterfactual"abundance) was then predicted based on the sample-specific standard curve. Likewise, 95% confidence intervals were constructed around the expected abundance. Finally, the hydrocarbon transfer was estimated as the difference between the observed and"counterfactual"abundance.

Gene editing by CRISPR/Cas9 mutagenesis

To create null mutants for elongase F (eloF) in D. prolongata, two guide RNAs were designed that target its first exon (Fig. 6). Guide RNA sequences were as follows: gRNA43: 5'-TCTGCTATTTGTCCTCAAGGTGG-3’ and gRNA84: 5'-AGAGTACCCAGAGCAACCCATGG-3'. Embryo injection and mutation screening were conducted as described [140]. Deletion of sequences between two guide RNAs was confirmed by Sanger sequencing. Two mutant strains were obtained: one with a frameshift mutation resulting in an early stop codon, and the other with a 45 bp in-frame deletion resulting in the loss of 15 amino acids (Fig. 6).

Statistical analysis on mutant CHC profiles

To determine the effects of eloF on pheromone production in D. prolongata, we examined the CHC profiles of both homozygous eloF mutant strains generated by CRISPR/Cas9 mutagenesis. The reference genome strain, in which these mutants were induced, was used as the control. Multivariate and univariate analyses were performed on the absolute quantity (on a logarithmic scale) of 18 consensus CHCs that are shared between sexes and collectively account for > 98% of total CHC abundance. Prior to principal component analysis (PCA), CHC abundances were centered to zero means but not standardized to unit variance, so PCA was conducted on the sample covariance matrix. In the PCA scatter plot, 95% confidence regions for each group (genotype x sex) were estimated assuming underlying bivariate t-distributions. To determine whether (1) pheromone profiles of wild-type and eloF mutant flies were significantly different and (2) whether mutation effects differed between sexes, we used two-way ANOVA models with interaction effects between sex and genotype, followed by Tukey's range test for all pairwise comparisons. The ANOVA model was specified as follows: Log(abundance) ~ sex + genotype + sex * genotype. Data management (R package suite"tidyverse") and statistical modeling (R packages"car,""multcomp,""lsmeans") were conducted by in-house R scripts (R Core Team 2022), with plots generated by the"ggplot2"package and subsequently polished by the Inkscape software (version 0.92.4, https://inkscape.org).

Comparative sequence analysis

Sequences surrounding the eloF locus (~ 2 kb) were extracted from reference genomes of each species [74, 95]. To study sequence evolution, multiple alignments of DNA sequences were conducted using Clustal Omega (version 1.2.2 [130]) using the default parameters. To examine sequence divergence of eloF orthologs, single nucleotide variants (SNVs) in the coding sequence (CDS) of eloF were called (Additional file 12: Table S4) by manual inspection of RNA-seq reads that mapped to a nearby region using the software IGV (version 2.4.11, Broad Institute). Open reading frames (ORFs) were predicted based on the standard genetic code and required a minimum of 400 base pairs. To identify the genetic nature of"honghaier,"a putative transposable element, and its associated ORF, the web application BLASTn (version 2.13.0 +) was used to search against all NCBI databases and the database of known transposable elements Dfam [136]. To visualize the phylogenetic distribution of honghaier and associated ORF (Additional file 14: Table S5), local standalone blastn databases were made from genome assemblies, and command-line-based BLASTn (version 2.2.31 +) was used.

To assess the sequence complexity of the honghaier insertion, a preliminary dot plot (not shown) was made using the EMBOSS (version 6.5.7) tool dotmatcher, with a word size of 10. De novo motif discovery was subsequently made to identify the repeating units using MEME-suite (https://meme-suite.org) software MEME (version 5.3.2, [8]). The following command-line flags were used: “-dna -mod anr -nmotifs 3 -revcomp”. The AT content was estimated by averaging the occurrence of adenosine (A) and thymine (T) in a window of 50 bp. Unless otherwise noted, sequence analysis was conducted in Geneious Prime (version 2021.0.3, Biomatters, www.geneious.com).

Transcription factor (TF) binding motif analysis

Since the exact motif sequences that activate gene expression in adult oenocytes are largely unknown, we used de novo prediction to identify TF-binding motifs that are enriched in adult oenocytes. A list of genes annotated as being differentially expressed in adult oenocytes over other tissues (referred to as oenocyte markers hereafter) in D. melanogaster was downloaded from single-cell Fly Cell Atlas ("10X relaxed dataset", [90]). Marker genes were stringently filtered using log fold change cutoff > 1 and a p-value cutoff of 1e-10. Using R Bioconductor packages"org.Dm.eg.db"(feature annotation database, version 3.14.0),"TxDb.Dmelanogaster.UCSC.dm6.ensGene"(transcript database, version 3.12.0), and"BSgenome.Dmelanogaster.UCSC.dm6"(genome database, version 1.4.1), genes were further filtered by the following criteria. Oenocyte marker genes must (1) have a matching FlyBase unique gene identifier and (2) map to chromosome X, 2, or 3. This resulted in a final set of 956 oenocyte-enriched markers (Additional file 32: Data File 8). This list included the previously reported oenocyte markers desaturase F (desatF, [32]) and elongase F (eloF, [31]). For each oenocyte marker, up to 1 kb of the upstream promoter region was extracted for motif enrichment analysis.

Motif enrichment analysis was performed on the retrieved upstream sequences (Additional file 33: Data File 9) by Meme Suite software AME (version 5.3.2, [102]) using the following command line flags: -control –shuffle– -scoring avg -method fisher. The iDMMPMM motif database downloaded from Meme Suite provided 39 known motifs with well-supported DNase-I footprint evidence [81]. We observed significant enrichment for binding motifs associated with the TFs bric-a-brac (bab1, p = 1.75e-60) and Mothers against dpp (Mad, p = 2.14e-21). Both these genes are expressed in the adult oenocytes of D. melanogaster (Additional file 28: Data File 7). Other candidates were not considered because their p values were several orders higher than the top 2 candidates.

Using bab1 and Mad as candidate motifs that may underlie oenocyte development, motif occurrence analysis was performed on non-protein-coding regions of eloF across 5 species in the rhopaloa subgroup using Meme Suite software FIMO (version 5.3.2, [58]). The following command line flags were used:"–parse-genomic-coord, –thresh 0.001". As no matches corresponding to Mad were found, only bab1 binding motifs were reported (Additional file 13: Fig. S9). In addition to tissue-specific motifs, individual motif occurrence analysis was performed on sex-related motifs by FIMO. The binding motifs of the doublesex (dsx) TF were retrieved from Shirangi et al. [128], FlyReg [13], Fly Factor Survey (https://mccb.umassmed.edu/ffs), and JASPAR (9 th release, https://jaspar.genereg.net). As no match was found for the motif reported by Shirangi et al., sex motifs included three targets: dsx from JASPAR and dsx-F and dsx-M from FlyReg (where both proteins have identical binding sequences).

Design of reporter constructs

We generated transgenic D. melanogaster strains that carried orthologous eloF sequences from D. prolongata and D. carrolli. “Long” constructs were designed to cover the entire eloF locus and its whole flanking region (between the flanking genes CG16904 on the left and CG8534 on the right):"Dpro eloF WT^(l)"with the allele from D. prolongata and"Dcar eloF WT^(l)"with the allele from D. carrolli (Additional file 17: Fig. S12). The downstream region of eloF contains a putative transposable element (TE) insertion, which we named"honghaier", in D. prolongata but not in D. carrolli or any other species at this location. Two additional constructs were therefore produced by a TE swap: one engineered allele had honghaier removed from the D. prolongata allele ("Dpro eloF WT^(l)—TE"), and the other had honghaier inserted into the D. carrolli allele ("Dcar eloF WT^(l) + TE,"Additional file 17: Fig. S12). In addition, “short” reporter constructs were designed with the DNA sequences of the downstream region of eloF (between CG16904 on the left and eloF on the right; note that the two genes are transcribed in head-to-head orientation). Similar to the “long” constructs, two of the short constructs were wild-type alleles of each species,"Dpro eloF WT^(s)"and"Dcar eloF WT^(s)", while the other two were produced by the honghaier swap:"Dpro eloF WT^(s)—TE"and"Dcar eloF WT^(s) + TE"(Additional file 17: Fig. S12).

To clone the reporter sequences, DNA fragments were amplified by SeqAmp (Takara Bio, San Jose, CA), a proofreading DNA polymerase (See Additional file 30: Table S9 for primers used in this cloning experiment). PCR-amplified DNA fragments were first Gibson-cloned into linearized pCR8 vectors (Invitrogen, Carlsbad, CA) using Gibson Assembly Master Mix (New England Biolabs, Ipswich, MA) according to kit recommendations. We then conducted a Gateway reaction to transfer the DNA inserts into the destination vector pGreenFriend (Additional file 17: Fig. S12, [103]) by Gateway recombination reaction using LR Clonase II Enzyme mix (Invitrogen). The pGreenFriend vector has a GFP reporter driven by the Drosophila Synthetic Core Promoter (Additional file 17: Fig. S12). Final constructs were bulk-purified using a QIAGEN midi-prep kit (QIAGEN, Redwood City, CA) and confirmed by Sanger sequencing (McLab Sequencing, San Francisco, CA). Chemically competent E. coli strain NEB5alpha H2987 (New England Biolabs) was used for transformation.

Transgenic strains

The pGreenFriend vector has a single attB site that allows it to integrate into attP anchor sites in the D. melanogaster genome (Additional file 17: Fig. S12). 30 µg of purified plasmids were sent to BestGene (https://www.thebestgene.com) for embryo injection. The genotype of injected flies was y¹ w^67c23; P{CaryP}attP40, with the attp40 landing site on the second chromosome [101]. Transformed G0 flies were crossed to yw flies, and the resulting G1 progeny were genotyped to verify successful integration. Heterozygous flies carrying attP insertion (attP40*) were selected based on orange eye color. Confirmed insertions were balanced and flies homozygous for the attP40* site with the reporter insertion were selected from these balanced strains and used for antibody staining and RT-qPCR.

Tissue dissection and antibody staining

Homozygous transgenic flies were isolated as virgins and the dorsal abdominal body wall was dissected in 1 × Tris-NaCl-Triton (TNT) buffer (100 mM Tris pH7.5, 300 mM NaCl, 0.5% Triton-X). Flies with the yw genotype were used as negative control to account for oenocyte autofluorescence. A standard fixation protocol was adopted [141]. Tissues were pooled and fixed in a fixation buffer (100 mM Tris pH7.5, 300 mL NaCl, 4% paraformaldehyde (Electron Microscopy Sciences, Hatfield, PA)) for 20 min on a rotation platform set to a gentle speed at a frequency of 0.33 Hz. Fixed tissue was then washed in 1 mL 1xTNT for 15 min three times. Post-wash tissues were stored in 1 mL fresh 1xTNT at 4 °C until antibody staining.

For GFP staining, fixed tissues were first transferred to 3 × 3 dissection plates and washed with 300 µl 1xPhosphate-Buffered Saline with Triton-X 100 (PBST buffer) for 15 min. To reduce non-specific antibody binding, washed tissues were blocked in 180 µl of 5% normal goat serum (Jackson Immunoresearch, West Grove, PA) for 30 min, followed by three times 1 × PBST wash (300 µl, 15 min each). Tissues were stained using 300 µl of primary antibody at 4 °C overnight and washed three times in 300 µl 1 × PBST on the following day. Immediately after primary staining, tissues were stained using 300 µl of secondary antibody at 4 °C for 1 h and washed three times in 300 µl 1 × PBST (15 min each). Stained tissues were stored in 1 mL 1xPBST at 4 °C and covered with aluminum. All antibody staining steps were performed in the dark and incubated on a nutator set to gentle speed. The ingredients of buffers used are summarized as follows. 1xPBST buffer was prepared by adding Triton-X to 1xPhosphate-Buffered Saline (1 × 1xPBS, Corning, Manassas, VA) to a final concentration of 0.4% (v2v). The blocking solution was freshly made by adding goat serum (Jackson Immunoresearch) in 1xPBST to a final concentration of 5% (v2v) and stored at 4 °C upon use. Staining solutions were freshly made by diluting primary (Chicken-anti-GFP, Invitrogen) or secondary antibodies (goat anti-chicken-AF488, Jackson Immunoresearch) in 1 × blocking solution to a final concentration of 1:200. At least four dissected cuticle filets were studied for each short construct.

To facilitate mounting, fully stained dorsal cuticles were flattened by trimming to an approximately rectangular shape. Under the dissecting microscope, the margins of the A1 segment and A6/A7 segments were first removed, and the lateral sides of the remaining segments were trimmed by fine scissors. After flattening, the dorsal cuticle was placed on a cover slide (22 × 22 mm, thickness 1.5, Corning) with the interior facing up. A 20 µl of antifade FluromountG reagent (Electron Microscopy Sciences) was added and spread evenly to reduce the formation of air bubbles. The mounting slide (3''× 1''x 1 mm, Fisher Scientific, Pittsburgh, PA) was placed on top of the cover slide to finish preparation. Mounted slides were stored in a slide binder at 4 °C before imaging.

Confocal microscopy

Mounted tissues were imaged using a Leica SP8 confocal microscope. The 488 nm laser was used to visualize GFP. Images were taken every 5 µm using a 20X objective and digital zoom. Images were further processed in Fiji ImageJ [122] to project everything on a z-stack with maximum intensity.

Data availability

The RNA-seq datasets generated during the current study are available in the Sequence Read Archive (SRA) under BioProject ID PRJNA1178650. Other data generated or analyzed during this study are included in this published article and supplementary information files

Abbreviations

CHC:: Cuticular hydrocarbon
cVA:: cis-Vaccenyl Acetate
7 T:: 7-Tricosene
7,11-HD:: 7,11-heptacosadiene
9 T:: (Z)-9-tricosene
9P:: (Z)-9-pentacosene
9H:: (Z)-9-heptacosene
nC26:: n-heneicosane
nC30:: n-triacotane
FAEs:: Fatty acid elongases
KAR:: Keto-acyl-CoA-reductase
HACD:: Hydroxy-acyl-CoA dehydratase
TER:: Trans-enoyl-CoA-reductase
FARs:: Fatty acyl-CoA reductases
dsx :: Doublesex
Sxl :: Sex-lethal
tra :: Transformer
eloF :: Elongase F
eloF[-]:: Loss-of-function eloF mutants
desatF :: also known as Fad2 Desaturase F
Hr4 :: Hormone receptor 4
GFP :: Green Fluorescent Protein
bab1 :: Bric-à-brac 1
Irx/iro :: Iroquois
Rpl32 :: Ribosomal protein L32
Mad :: Mothers against dpp
LT:: Low throughput
DAG:: Directed acyclic graph
TMM:: Trimmed mean of M-values
cpm:: Counts per million
FDR:: False discovery rate
NRT:: No-reverse-transcriptase control
R² :: Coefficient of determination
GC:: Gas chromatography
MS:: Mass spectrometry
PCA:: Principal component analysis
qPCR:: Quantitative PCR
DEGs:: Differentially expressed genes
GO:: Gene Ontology
SNVs:: Single-nucleotide variants
TE:: Transposable elements
TIRs:: Terminal inverted repeats
MITE:: miniature inverted-repeat transposable elements
ORF:: Open reading frame
CDS:: Coding DNA sequence
TF:: Transcription factor
TNT:: Tris-NaCl-Triton
PBST:: Phosphate-Buffered Saline with Triton-X 100

References

Ackermann M, Strimmer K. A general modular framework for gene set enrichment analysis. BMC Bioinformatics. 2009;10(1):47. https://doiorg.publicaciones.saludcastillayleon.es/10.1186/1471-2105-10-47.
Article CAS PubMed PubMed Central Google Scholar
Adams TS, Dillwith JW, Blomquist GJ. The role of 20-hydroxyecdysone in housefly sex pheromone biosynthesis. J Insect Physiol. 1984;30(4):287–94. https://doiorg.publicaciones.saludcastillayleon.es/10.1016/0022-1910(84)90129-X.
Article CAS Google Scholar
Alexa A, Rahnenfuhrer J. 2022. topGO: Enrichment Analysis for Gene Ontology. https://doiorg.publicaciones.saludcastillayleon.es/10.18129/B9.bioc.topGO. [accessed 2022 Nov 1]. https://bioconductor.org/packages/topGO/.
Alexa A, Rahnenführer J, Lengauer T. Improved scoring of functional groups from gene expression data by decorrelating GO graph structure. Bioinformatics. 2006;22(13):1600–7. https://doiorg.publicaciones.saludcastillayleon.es/10.1093/bioinformatics/btl140.
Article CAS PubMed Google Scholar
Almojil D, Bourgeois Y, Falis M, Hariyani I, Wilcox J, Boissinot S. The Structural, Functional and Evolutionary Impact of Transposable Elements in Eukaryotes. Genes. 2021;12(6):918. https://doiorg.publicaciones.saludcastillayleon.es/10.3390/genes12060918.
Article CAS PubMed PubMed Central Google Scholar
Amino K, Matsuo T. Intra- Versus Inter-Sexual Selection on Sexually Dimorphic Traits in Drosophila prolongata. Zoolog Sci. 2020;37(3):210–6. https://doiorg.publicaciones.saludcastillayleon.es/10.2108/zs200010.
Article PubMed Google Scholar
Andersson M. Sexual selection. Princeton, NJ: Princeton Univ. Press; 1994.
Book Google Scholar
Bailey TL, Elkan C. Fitting a mixture model by expectation maximization to discover motifs in biopolymers. Proc Int Conf Intell Syst Mol Biol. 1994;2:28–36.
CAS PubMed Google Scholar
Baron A, Denis B, Wicker-Thomas C. Control of pheromone production by ovaries in Drosophila. J Insect Physiol. 2018;109:138–43. https://doiorg.publicaciones.saludcastillayleon.es/10.1016/j.jinsphys.2018.07.003.
Article CAS PubMed Google Scholar
Bartelt RJ, Schaner AM, Jackson LL. cis-Vaccenyl acetate as an aggregation pheromone inDrosophila melanogaster. J Chem Ecol. 1985;11(12):1747–56. https://doiorg.publicaciones.saludcastillayleon.es/10.1007/BF01012124.
Article CAS PubMed Google Scholar
Barth NKH, Li L, Taher L. Independent Transposon Exaptation Is a Widespread Mechanism of Redundant Enhancer Evolution in the Mammalian Genome. Genome Biol Evol. 2020;12(3):1–17. https://doiorg.publicaciones.saludcastillayleon.es/10.1093/gbe/evaa004.
Article CAS PubMed PubMed Central Google Scholar
Benjamini Y, Hochberg Y. Controlling the False Discovery Rate: A Practical and Powerful Approach to Multiple Testing. J R Stat Soc Ser B Methodol. 1995;57(1):289–300.
Article Google Scholar
Bergman CM, Carlson JW, Celniker SE. Drosophila DNase I footprint database: a systematic genome annotation of transcription factor binding sites in the fruitfly. Drosophila melanogaster Bioinforma Oxf Engl. 2005;21(8):1747–9. https://doiorg.publicaciones.saludcastillayleon.es/10.1093/bioinformatics/bti173.
Article CAS Google Scholar
Bergman DT, Jones TR, Liu V, Ray J, Jagoda E, Siraj L, Kang HY, Nasser J, Kane M, Rios A, et al. Compatibility rules of human enhancer and promoter sequences. Nature. 2022;607(7917):176–84. https://doiorg.publicaciones.saludcastillayleon.es/10.1038/s41586-022-04877-w.
Article CAS PubMed PubMed Central Google Scholar
Bi J, Xiang Y, Chen H, Liu Z, Grönke S, Kühnlein RP, Huang X. Opposite and redundant roles of the two Drosophila perilipins in lipid mobilization. J Cell Sci. 2012;125(15):3568–77. https://doiorg.publicaciones.saludcastillayleon.es/10.1242/jcs.101329.
Article CAS PubMed Google Scholar
Billeter J-C, Atallah J, Krupp JJ, Millar JG, Levine JD. Specialized cells tag sexual and species identity in Drosophila melanogaster. Nature. 2009;461(7266):987–91. https://doiorg.publicaciones.saludcastillayleon.es/10.1038/nature08495.
Article CAS PubMed Google Scholar
Blomquist GJ, Adams TS, Dillwith JW. Induction of female sex pheromone production in male houseflies by ovary implants or 20-hydroxyecdysone. J Insect Physiol. 1984;30(4):295–302. https://doiorg.publicaciones.saludcastillayleon.es/10.1016/0022-1910(84)90130-6.
Article CAS Google Scholar
Blomquist GJ, Adams TS, Halarnkar PP, Gu P, Mackay ME, Brown LA. Ecdysteroid induction of sex pheromone biosynthesis in the housefly, Musca domestica—Are other factors involved? J Insect Physiol. 1992;38(4):309–18. https://doiorg.publicaciones.saludcastillayleon.es/10.1016/0022-1910(92)90131-V.
Article CAS Google Scholar
Blomquist GJ, Bagnères A-G. Insect hydrocarbons biology, biochemistry, and chemical ecology. Cambridge, UK: Cambridge Univ. Press; 2010.
Book Google Scholar
Blomquist GJ, Tillman JA, Reed JR, Gu P, Vanderwel D, Choi S, Reitz RC. Regulation of enzymatic activity involved in sex pheromone production in the housefly. Musca domestica Insect Biochem Mol Biol. 1995;25(6):751–7. https://doiorg.publicaciones.saludcastillayleon.es/10.1016/0965-1748(95)00015-N.
Article CAS PubMed Google Scholar
Bontonou G, Wicker-Thomas C. Sexual Communication in the Drosophila Genus. Insects. 2014;5(2):439–58. https://doiorg.publicaciones.saludcastillayleon.es/10.3390/insects5020439.
Article PubMed PubMed Central Google Scholar
Boughman JW, Rundle HD, Schluter D. Parallel evolution of sexual isolation in sticklebacks. Evolution. 2005;59(2):361–73. https://doiorg.publicaciones.saludcastillayleon.es/10.1111/j.0014-3820.2005.tb00995.x.
Article PubMed Google Scholar
Bourbon H-MG, Benetah MH, Guillou E, Mojica-Vazquez LH, Baanannou A, Bernat-Fabre S, Loubiere V, Bantignies F, Cavalli G, Boube M. A shared ancient enhancer element differentially regulates the bric-a-brac tandem gene duplicates in the developing Drosophila leg. PLOS Genet. 2022;18(3): e1010083. https://doiorg.publicaciones.saludcastillayleon.es/10.1371/journal.pgen.1010083.
Article CAS PubMed PubMed Central Google Scholar
Bourque G, Leong B, Vega VB, Chen X, Lee YL, Srinivasan KG, Chew J-L, Ruan Y, Wei C-L, Ng HH, et al. Evolution of the mammalian transcription factor binding repertoire via transposable elements. Genome Res. 2008;18(11):1752–62. https://doiorg.publicaciones.saludcastillayleon.es/10.1101/gr.080663.108.
Article CAS PubMed PubMed Central Google Scholar
Bray S, Amrein H. A Putative Drosophila Pheromone Receptor Expressed in Male-Specific Taste Neurons Is Required for Efficient Courtship. Neuron. 2003;39(6):1019–29. https://doiorg.publicaciones.saludcastillayleon.es/10.1016/S0896-6273(03)00542-7.
Article CAS PubMed Google Scholar
Broder ED, Elias DO, Rodríguez RL, Rosenthal GG, Seymoure BM, Tinghitella RM. Evolutionary novelty in communication between the sexes. Biol Lett. 2021;17(2):20200733. https://doiorg.publicaciones.saludcastillayleon.es/10.1098/rsbl.2020.0733.
Article PubMed PubMed Central Google Scholar
Buchinger TJ, Li W. Chemical communication and its role in sexual selection across Animalia. Commun Biol. 2023;6(1):1178. https://doiorg.publicaciones.saludcastillayleon.es/10.1038/s42003-023-05572-w.
Article PubMed PubMed Central Google Scholar
Campbell PM, Healy MJ, Oakeshott JG. Characterisation of juvenile hormone esterase in Drosophila melanogaster. Insect Biochem Mol Biol. 1992;22(7):665–77. https://doiorg.publicaciones.saludcastillayleon.es/10.1016/0965-1748(92)90045-G.
Article CAS Google Scholar
Cantarel BL, Korf I, Robb SMC, Parra G, Ross E, Moore B, Holt C, Alvarado AS, Yandell M. MAKER: An easy-to-use annotation pipeline designed for emerging model organism genomes. Genome Res. 2008;18(1):188–96. https://doiorg.publicaciones.saludcastillayleon.es/10.1101/gr.6743907.
Article CAS PubMed PubMed Central Google Scholar
Carroll SB. Evo-Devo and an Expanding Evolutionary Synthesis: A Genetic Theory of Morphological Evolution. Cell. 2008;134(1):25–36. https://doiorg.publicaciones.saludcastillayleon.es/10.1016/j.cell.2008.06.030.
Article CAS PubMed Google Scholar
Chertemps T, Duportets L, Labeur C, Ueda R, Takahashi K, Saigo K, Wicker-Thomas C. A female-biased expressed elongase involved in long-chain hydrocarbon biosynthesis and courtship behavior in Drosophila melanogaster. Proc Natl Acad Sci. 2007;104(11):4273–8. https://doiorg.publicaciones.saludcastillayleon.es/10.1073/pnas.0608142104.
Article CAS PubMed PubMed Central Google Scholar
Chertemps T, Duportets L, Labeur C, Ueyama M, Wicker-Thomas C. A female-specific desaturase gene responsible for diene hydrocarbon biosynthesis and courtship behaviour in Drosophila melanogaster. Insect Mol Biol. 2006;15(4):465–73. https://doiorg.publicaciones.saludcastillayleon.es/10.1111/j.1365-2583.2006.00658.x.
Article CAS PubMed Google Scholar
Chiang YN, Tan KJ, Chung H, Lavrynenko O, Shevchenko A, Yew JY. Steroid Hormone Signaling Is Essential for Pheromone Production and Oenocyte Survival. PLOS Genet. 2016;12(6): e1006126. https://doiorg.publicaciones.saludcastillayleon.es/10.1371/journal.pgen.1006126.
Article CAS PubMed PubMed Central Google Scholar
Chung H, Carroll SB. Wax, sex and the origin of species: Dual roles of insect cuticular hydrocarbons in adaptation and mating. BioEssays. 2015;37(7):822–30. https://doiorg.publicaciones.saludcastillayleon.es/10.1002/bies.201500014.
Article CAS PubMed PubMed Central Google Scholar
Chung H, Loehlin DW, Dufour HD, Vaccarro K, Millar JG, Carroll SB. A Single Gene Affects Both Ecological Divergence and Mate Choice in Drosophila. Science. 2014;343(6175):1148–51. https://doiorg.publicaciones.saludcastillayleon.es/10.1126/science.1249998.
Article CAS PubMed Google Scholar
Chuong EB, Elde NC, Feschotte C. Regulatory activities of transposable elements: from conflicts to benefits. Nat Rev Genet. 2017;18(2):71–86. https://doiorg.publicaciones.saludcastillayleon.es/10.1038/nrg.2016.139.
Article CAS PubMed Google Scholar
Cinnamon E, Makki R, Sawala A, Wickenberg LP, Blomquist GJ, Tittiger C, Paroush Z, Gould AP. Drosophila Spidey/Kar Regulates Oenocyte Growth via PI3-Kinase Signaling. PLoS Genet. 2016;12(8): e1006154. https://doiorg.publicaciones.saludcastillayleon.es/10.1371/journal.pgen.1006154.
Article CAS PubMed PubMed Central Google Scholar
Combs PA, Krupp JJ, Khosla NM, Bua D, Petrov DA, Levine JD, Fraser HB. Tissue-specific cis-regulatory divergence implicates eloF in inhibiting interspecies mating in Drosophila. Curr Biol. 2018;28(24):3969–75. https://doiorg.publicaciones.saludcastillayleon.es/10.1016/j.cub.2018.10.036.
Article CAS PubMed PubMed Central Google Scholar
Course MM, Scott AI, Schoor C, Hsieh C-H, Papakyrikos AM, Winter D, Cowan TM, Wang X. Phosphorylation of MCAD selectively rescues PINK1 deficiencies in behavior and metabolism. Mol Biol Cell. 2018;29(10):1219–27. https://doiorg.publicaciones.saludcastillayleon.es/10.1091/mbc.E18-03-0155.
Article PubMed PubMed Central Google Scholar
Cridland JM, Thornton KR, Long AD. Gene Expression Variation in Drosophila melanogaster Due to Rare Transposable Element Insertion Alleles of Large Effect. Genetics. 2015;199(1):85–93. https://doiorg.publicaciones.saludcastillayleon.es/10.1534/genetics.114.170837.
Article CAS PubMed Google Scholar
Dallerac R, Labeur C, Jallon J-M, Knipple DC, Roelofs WL, Wicker-Thomas C. A Δ9 desaturase gene with a different substrate specificity is responsible for the cuticular diene hydrocarbon polymorphism in Drosophila melanogaster. Proc Natl Acad Sci. 2000;97(17):9449–54. https://doiorg.publicaciones.saludcastillayleon.es/10.1073/pnas.150243997.
Article CAS PubMed PubMed Central Google Scholar
Datta SR, Vasconcelos ML, Ruta V, Luo S, Wong A, Demir E, Flores J, Balonze K, Dickson BJ, Axel R. The Drosophila pheromone cVA activates a sexually dimorphic neural circuit. Nature. 2008;452(7186):473–7. https://doiorg.publicaciones.saludcastillayleon.es/10.1038/nature06808.
Article CAS PubMed Google Scholar
Dembeck LM, Böröczky K, Huang W, Schal C, Anholt RR, Mackay TF. 2015. Genetic architecture of natural variation in cuticular hydrocarbon composition in Drosophila melanogaster. eLife. 4:e09861. https://doiorg.publicaciones.saludcastillayleon.es/10.7554/eLife.09861.001.
Dobin A, Davis CA, Schlesinger F, Drenkow J, Zaleski C, Jha S, Batut P, Chaisson M, Gingeras TR. STAR: ultrafast universal RNA-seq aligner. Bioinforma Oxf Engl. 2013;29(1):15–21. https://doiorg.publicaciones.saludcastillayleon.es/10.1093/bioinformatics/bts635.
Article CAS Google Scholar
Drăghici S, Sellamuthu S, Khatri P. Babel’s tower revisited: A universal resource for cross-referencing across annotation databases. Bioinforma Oxf Engl. 2006;22(23):2934–9. https://doiorg.publicaciones.saludcastillayleon.es/10.1093/bioinformatics/btl372.
Article CAS Google Scholar
Everaerts C, Farine J-P, Cobb M, Ferveur J-F. Drosophila Cuticular Hydrocarbons Revisited: Mating Status Alters Cuticular Profiles. PLoS ONE. 2010;5(3): e9607. https://doiorg.publicaciones.saludcastillayleon.es/10.1371/journal.pone.0009607.
Article CAS PubMed PubMed Central Google Scholar
Fan P, Manoli DS, Ahmed OM, Chen Y, Agarwal N, Kwong S, Cai AG, Neitz J, Renslo A, Baker BS, et al. Genetic and neural mechanisms that inhibit Drosophila from mating with other species. Cell. 2013;154(1):89–102. https://doiorg.publicaciones.saludcastillayleon.es/10.1016/j.cell.2013.06.008.
Article CAS PubMed Google Scholar
Fang S, Takahashi A, Wu C-I. A mutation in the promoter of desaturase 2 is correlated with sexual isolation between Drosophila behavioral races. Genetics. 2002;162(2):781–4. https://doiorg.publicaciones.saludcastillayleon.es/10.1093/genetics/162.2.781.
Article CAS PubMed PubMed Central Google Scholar
Ferveur J-F. Cuticular hydrocarbons: their evolution and roles in Drosophila pheromonal communication. Behav Genet. 2005;35(3):279–95. https://doiorg.publicaciones.saludcastillayleon.es/10.1007/s10519-005-3220-5.
Article PubMed Google Scholar
Ferveur J-F, Cobb M, Boukella H, Jallon J-M. World-wide variation in Drosophila melanogaster sex pheromone: behavioural effects, genetic bases and potential evolutionary consequences. Genetica. 1996;97(1):73–80. https://doiorg.publicaciones.saludcastillayleon.es/10.1007/BF00132583.
Article CAS PubMed Google Scholar
Ferveur J-F, Jallon J-M. Genetic control of male cuticular hydrocarbons in Drosophila melanogaster. Genet Res. 1996;67(3):211–8. https://doiorg.publicaciones.saludcastillayleon.es/10.1017/S0016672300033693.
Article CAS PubMed Google Scholar
Ferveur J-F, Savarit F, O’Kane CJ, Sureau G, Greenspan RJ, Jallon J-M. Genetic feminization of pheromones and its behavioral consequences in Drosophila males. Science. 1997;276(5318):1555–8. https://doiorg.publicaciones.saludcastillayleon.es/10.1126/science.276.5318.1555.
Article CAS PubMed Google Scholar
Ferveur J-F, Sureau G. Simultaneous influence on male courtship of stimulatory and inhibitory pheromones produced by live sex-mosaic Drosophila melanogaster. Proc R Soc Lond B Biol Sci. 1996;263(1373):967–73. https://doiorg.publicaciones.saludcastillayleon.es/10.1098/rspb.1996.0143.
Article CAS Google Scholar
Finet C, Slavik K, Pu J, Carroll SB, Chung H. Birth-and-Death Evolution of the Fatty Acyl-CoA Reductase (FAR) Gene Family and Diversification of Cuticular Hydrocarbon Synthesis in Drosophila. Genome Biol Evol. 2019;11(6):1541–51. https://doiorg.publicaciones.saludcastillayleon.es/10.1093/gbe/evz094.
Article CAS PubMed PubMed Central Google Scholar
González J, Macpherson JM, Petrov DA. A Recent Adaptive Transposable Element Insertion Near Highly Conserved Developmental Loci in Drosophila melanogaster. Mol Biol Evol. 2009;26(9):1949–61. https://doiorg.publicaciones.saludcastillayleon.es/10.1093/molbev/msp107.
Article CAS PubMed PubMed Central Google Scholar
Gordon HB, Valdez L, Letsou A. 2018. Etiology and treatment of adrenoleukodystrophy: new insights from Drosophila. Dis Model Mech. 11(6):dmm031286. https://doiorg.publicaciones.saludcastillayleon.es/10.1242/dmm.031286.
Gould A, Morrison A, Sproat G, White RA, Krumlauf R. Positive cross-regulation and enhancer sharing: two mechanisms for specifying overlapping Hox expression patterns. Genes Dev. 1997;11(7):900–13. https://doiorg.publicaciones.saludcastillayleon.es/10.1101/gad.11.7.900.
Article CAS PubMed Google Scholar
Grant CE, Bailey TL, Noble WS. FIMO: scanning for occurrences of a given motif. Bioinformatics. 2011;27(7):1017–8. https://doiorg.publicaciones.saludcastillayleon.es/10.1093/bioinformatics/btr064.
Article CAS PubMed PubMed Central Google Scholar
Gratten J, Beraldi D, Lowder B, v, McRae A f, Visscher P m, Pemberton J m, Slate J. Compelling evidence that a single nucleotide substitution in TYRP1 is responsible for coat-colour polymorphism in a free-living population of Soay sheep. Proc R Soc B Biol Sci. 2007;274(1610):619–26. https://doiorg.publicaciones.saludcastillayleon.es/10.1098/rspb.2006.3762.
Article CAS Google Scholar
Grillet M, Dartevelle L, Ferveur J-F. A Drosophila male pheromone affects female sexual receptivity. Proc R Soc Lond B Biol Sci. 2006;273(1584):315–23. https://doiorg.publicaciones.saludcastillayleon.es/10.1098/rspb.2005.3332.
Article CAS Google Scholar
Grillet M, Everaerts C, Houot B, Ritchie MG, Cobb M, Ferveur J-F. Incipient speciation in Drosophila melanogaster involves chemical signals. Sci Rep. 2012;2(1):224. https://doiorg.publicaciones.saludcastillayleon.es/10.1038/srep00224.
Article CAS PubMed PubMed Central Google Scholar
Herpin A, Schartl M, Depincé A, Guiguen Y, Bobe J, Hua-Van A, Hayman ES, Octavera A, Yoshizaki G, Nichols KM, et al. 2021. Allelic diversification after transposable element exaptation promoted gsdf as the master sex determining gene of sablefish. Genome Res. https://doiorg.publicaciones.saludcastillayleon.es/10.1101/gr.274266.120. [accessed 2021 Aug 9]. https://genome.cshlp.org/content/early/2021/07/23/gr.274266.120.
Hof AE, van’t, Campagne P, Rigden DJ, Yung CJ, Lingley J, Quail MA, Hall N, Darby AC, Saccheri IJ. The industrial melanism mutation in British peppered moths is a transposable element. Nature. 2016;534(7605):102–5. https://doiorg.publicaciones.saludcastillayleon.es/10.1038/nature17951.
Article CAS Google Scholar
Hopkins BR, Kopp A. Evolution of sexual development and sexual dimorphism in insects. Curr Opin Genet Dev. 2021;69:129–39. https://doiorg.publicaciones.saludcastillayleon.es/10.1016/j.gde.2021.02.011.
Article CAS PubMed PubMed Central Google Scholar
Howard RW, Blomquist GJ. Ecological, behavioral, and biochemical aspects of insect hydrocarbons. Annu Rev Entomol. 2005;50(1):371–93. https://doiorg.publicaciones.saludcastillayleon.es/10.1146/annurev.ento.50.071803.130359.
Article CAS PubMed Google Scholar
Howard RW, Jackson LL, Banse H, Blows MW. 2003. Cuticular hydrocarbons of Drosophila birchii and D. serrata: identification and role in mate choice in D. serrata. J Chem Ecol. 29(4):961–976. https://doiorg.publicaciones.saludcastillayleon.es/10.1023/A:1022992002239.
Ishii K, Hirai Y, Katagiri C, Kimura MT. Sexual isolation and cuticular hydrocarbons in Drosophila elegans. Heredity. 2001;87(4):392–9. https://doiorg.publicaciones.saludcastillayleon.es/10.1046/j.1365-2540.2001.00864.x.
Article CAS PubMed Google Scholar
Jallon J-M. A few chemical words exchanged by Drosophila during courtship and mating. Behav Genet. 1984;14(5):441–78. https://doiorg.publicaciones.saludcastillayleon.es/10.1007/BF01065444.
Article CAS PubMed Google Scholar
Jallon JM, Antony C, Benamar O. An anti-aphrodisiac produced by Drosophila-Melanogaster males and transferred to females during copulation. C R Acad Sci Ser III Sci. Vie. 1981;292:1147–9.
Google Scholar
Jiggins CD, Estrada C, Rodrigues A. Mimicry and the evolution of premating isolation in Heliconius melpomene Linnaeus. J Evol Biol. 2004;17(3):680–91. https://doiorg.publicaciones.saludcastillayleon.es/10.1111/j.1420-9101.2004.00675.x.
Article CAS PubMed Google Scholar
Kapitonov VV, Jurka J. Molecular paleontology of transposable elements in the Drosophila melanogaster genome. Proc Natl Acad Sci. 2003;100(11):6569–74. https://doiorg.publicaciones.saludcastillayleon.es/10.1073/pnas.0732024100.
Article CAS PubMed PubMed Central Google Scholar
Khallaf MA, Auer TO, Grabe V, Depetris-Chauvin A, Ammagarahalli B, Zhang D-D, Lavista-Llanos S, Kaftan F, Weißflog J, Matzkin LM, et al. 2020. Mate discrimination among subspecies through a conserved olfactory pathway. Sci Adv. 6(25):eaba5279. https://doiorg.publicaciones.saludcastillayleon.es/10.1126/sciadv.aba5279.
Khallaf MA, Cui R, Weißflog J, Erdogmus M, Svatoš A, Dweck HKM, Valenzano DR, Hansson BS, Knaden M. Large-scale characterization of sex pheromone communication systems in Drosophila. Nat Commun. 2021;12(1):4165. https://doiorg.publicaciones.saludcastillayleon.es/10.1038/s41467-021-24395-z.
Article CAS PubMed PubMed Central Google Scholar
Kim BY, Wang JR, Miller DE, Barmina O, Delaney E, Thompson A, Comeault AA, Peede D, D’Agostino ER, Pelaez J, et al. 2021. Highly contiguous assemblies of 101 drosophilid genomes. Coop G, Wittkopp PJ, Sackton TB, editors. eLife. 10:e66405. https://doiorg.publicaciones.saludcastillayleon.es/10.7554/eLife.66405.
King-Jones K, Charles J-P, Lam G, Thummel CS. The Ecdysone-Induced DHR4 Orphan Nuclear Receptor Coordinates Growth and Maturation in Drosophila. Cell. 2005;121(5):773–84. https://doiorg.publicaciones.saludcastillayleon.es/10.1016/j.cell.2005.03.030.
Article CAS PubMed Google Scholar
Kishita Y, Tsuda M, Aigaki T. Impaired fatty acid oxidation in a Drosophila model of mitochondrial trifunctional protein (MTP) deficiency. Biochem Biophys Res Commun. 2012;419(2):344–9. https://doiorg.publicaciones.saludcastillayleon.es/10.1016/j.bbrc.2012.02.026.
Article CAS PubMed Google Scholar
Kopp A. Dmrt genes in the development and evolution of sexual dimorphism. Trends Genet. 2012;28(4):175–84. https://doiorg.publicaciones.saludcastillayleon.es/10.1016/j.tig.2012.02.002.
Article CAS PubMed PubMed Central Google Scholar
Kopp A, Duncan I, Carroll SB. Genetic control and evolution of sexually dimorphic characters in Drosophila. Nature. 2000;408(6812):553–9. https://doiorg.publicaciones.saludcastillayleon.es/10.1038/35046017.
Article CAS PubMed Google Scholar
Kudo A, Shigenobu S, Kadota K, Nozawa M, Shibata TF, Ishikawa Y, Matsuo T. Comparative analysis of the brain transcriptome in a hyper-aggressive fruit fly. Drosophila prolongata Insect Biochem Mol Biol. 2017;82:11–20. https://doiorg.publicaciones.saludcastillayleon.es/10.1016/j.ibmb.2017.01.006.
Article CAS PubMed Google Scholar
Kudo A, Takamori H, Watabe H, Ishikawa Y, Matsuo T. Variation in morphological and behavioral traits among isofemale strains of Drosophila prolongata (Diptera: Drosophilidae). Entomol Sci. 2015;18(2):221–9. https://doiorg.publicaciones.saludcastillayleon.es/10.1111/ens.12116.
Article Google Scholar
Kulakovskiy IV, Makeev VJ. Discovery of DNA motifs recognized by transcription factors through integration of different experimental sources. Biophysics. 2009;54(6):667–74. https://doiorg.publicaciones.saludcastillayleon.es/10.1134/S0006350909060013.
Article Google Scholar
Kurtovic A, Widmer A, Dickson BJ. A single class of olfactory neurons mediates behavioural responses to a Drosophila sex pheromone. Nature. 2007;446(7135):542–6. https://doiorg.publicaciones.saludcastillayleon.es/10.1038/nature05672.
Article CAS PubMed Google Scholar
Kwon D, Mucci D, Langlais KK, Americo JL, DeVido SK, Cheng Y, Kassis JA. Enhancer-promoter communication at the Drosophila engrailedlocus. Development. 2009;136(18):3067–75. https://doiorg.publicaciones.saludcastillayleon.es/10.1242/dev.036426.
Article CAS PubMed PubMed Central Google Scholar
Larkin A, Marygold SJ, Antonazzo G, Attrill H, Dos Santos G, Garapati PV, Goodman JL, Gramates LS, Millburn G, Strelets VB, et al. FlyBase: updates to the Drosophila melanogaster knowledge base. Nucleic Acids Res. 2021;49(D1):D899–907. https://doiorg.publicaciones.saludcastillayleon.es/10.1093/nar/gkaa1026.
Article CAS PubMed Google Scholar
Lassance J-M, Groot AT, Liénard MA, Antony B, Borgwardt C, Andersson F, Hedenström E, Heckel DG, Löfstedt C. Allelic variation in a fatty-acyl reductase gene causes divergence in moth sex pheromones. Nature. 2010;466(7305):486–9. https://doiorg.publicaciones.saludcastillayleon.es/10.1038/nature09058.
Article CAS PubMed Google Scholar
Laturney M, Billeter J-C. Drosophila melanogaster females restore their attractiveness after mating by removing male anti-aphrodisiac pheromones. Nat Commun. 2016;7(1):12322. https://doiorg.publicaciones.saludcastillayleon.es/10.1038/ncomms12322.
Article CAS PubMed PubMed Central Google Scholar
Laturney M, Moehring AJ. The Genetic Basis of Female Mate Preference and Species Isolation in Drosophila. Int J Evol Biol. 2012;2012: e328392. https://doiorg.publicaciones.saludcastillayleon.es/10.1155/2012/328392.
Article Google Scholar
Law CW, Chen Y, Shi W, Smyth GK. voom: precision weights unlock linear model analysis tools for RNA-seq read counts. Genome Biol. 2014;15(2):R29. https://doiorg.publicaciones.saludcastillayleon.es/10.1186/gb-2014-15-2-r29.
Article CAS PubMed PubMed Central Google Scholar
Lawniczak MKN, Barnes AI, Linklater JR, Boone JM, Wigby S, Chapman T. Mating and immunity in invertebrates. Trends Ecol Evol. 2007;22(1):48–55. https://doiorg.publicaciones.saludcastillayleon.es/10.1016/j.tree.2006.09.012.
Article PubMed Google Scholar
Li H, Janssens J, Waegeneer MD, Kolluru SS, Davie K, Gardeux V, Saelens W, David F, Brbić M, Leskovec J, et al. 2021. Fly Cell Atlas: a single-cell transcriptomic atlas of the adult fruit fly. [accessed 2021 Sep 1]. https://www.biorxiv.org/content/https://doiorg.publicaciones.saludcastillayleon.es/10.1101/2021.07.04.451050v1.
Liénard MA, Hagström ÅK, Lassance J-M, Löfstedt C. Evolution of multicomponent pheromone signals in small ermine moths involves a single fatty-acyl reductase gene. Proc Natl Acad Sci. 2010;107(24):10955–60. https://doiorg.publicaciones.saludcastillayleon.es/10.1073/pnas.1000823107.
Article PubMed PubMed Central Google Scholar
Linn CE, Young MS, Gendle M, Glover TJ, Roelofs WL. Sex pheromone blend discrimination in two races and hybrids of the European corn borer moth. Ostrinia nubilalis Physiol Entomol. 1997;22(3):212–23. https://doiorg.publicaciones.saludcastillayleon.es/10.1111/j.1365-3032.1997.tb01161.x.
Article CAS Google Scholar
Livak KJ, Schmittgen TD. Analysis of relative gene expression data using real-time quantitative PCR and the 2(-Delta Delta C(T)) Method. Methods San Diego Calif. 2001;25(4):402–8. https://doiorg.publicaciones.saludcastillayleon.es/10.1006/meth.2001.1262.
Article CAS PubMed Google Scholar
Loker R, Mann RS. Divergent expression of paralogous genes by modification of shared enhancer activity through a promoter-proximal silencer. Curr Biol CB. 2022;32(16):3545-3555.e4. https://doiorg.publicaciones.saludcastillayleon.es/10.1016/j.cub.2022.06.069.
Article CAS PubMed Google Scholar
Luecke D, Luo Y, Krzystek H, Jones C, Kopp A. 2024. Highly contiguous genome assembly of Drosophila prolongata—a model for evolution of sexual dimorphism and male-specific innovations. G3 GenesGenomesGenetics.:jkae155. https://doiorg.publicaciones.saludcastillayleon.es/10.1093/g3journal/jkae155.
Luecke D, Rice G, Kopp A. Sex-specific evolution of a Drosophila sensory system via interacting cis- and trans-regulatory changes. Evol Dev. 2022;24(1–2):37–60. https://doiorg.publicaciones.saludcastillayleon.es/10.1111/ede.12398.
Article CAS PubMed PubMed Central Google Scholar
Luecke DM, Kopp A. Sex-specific evolution of relative leg size in Drosophila prolongata results from changes in the intersegmental coordination of tissue growth. Evolution. 2019;73(11):2281–94. https://doiorg.publicaciones.saludcastillayleon.es/10.1111/evo.13847.
Article PubMed PubMed Central Google Scholar
Luo Y, Zhang Y, Farine J-P, Ferveur J-F, Ramírez S, Kopp A. Evolution of sexually dimorphic pheromone profiles coincides with increased number of male-specific chemosensory organs in Drosophila prolongata. Ecol Evol. 2019;9(23):13608–18. https://doiorg.publicaciones.saludcastillayleon.es/10.1002/ece3.5819.
Article PubMed PubMed Central Google Scholar
Lynch VJ, Leclerc RD, May G, Wagner GP. Transposon-mediated rewiring of gene regulatory networks contributed to the evolution of pregnancy in mammals. Nat Genet. 2011;43(11):1154–9. https://doiorg.publicaciones.saludcastillayleon.es/10.1038/ng.917.
Article CAS PubMed Google Scholar
Makki R, Cinnamon E, Gould AP. The Development and Functions of Oenocytes. Annu Rev Entomol. 2014;59(1):405–25. https://doiorg.publicaciones.saludcastillayleon.es/10.1146/annurev-ento-011613-162056.
Article CAS PubMed PubMed Central Google Scholar
Markstein M, Pitsouli C, Villalta C, Celniker SE, Perrimon N. Exploiting position effects and the gypsy retrovirus insulator to engineer precisely expressed transgenes. Nat Genet. 2008;40(4):476–83. https://doiorg.publicaciones.saludcastillayleon.es/10.1038/ng.101.
Article CAS PubMed PubMed Central Google Scholar
McLeay RC, Bailey TL. Motif Enrichment Analysis: a unified framework and an evaluation on ChIP data. BMC Bioinformatics. 2010;11(1):165. https://doiorg.publicaciones.saludcastillayleon.es/10.1186/1471-2105-11-165.
Article CAS PubMed PubMed Central Google Scholar
Miller SW, Rebeiz M, Atanasov JE, Posakony JW. Neural precursor-specific expression of multiple Drosophila genes is driven by dual enhancer modules with overlapping function. Proc Natl Acad Sci U S A. 2014;111(48):17194–9. https://doiorg.publicaciones.saludcastillayleon.es/10.1073/pnas.1415308111.
Article CAS PubMed PubMed Central Google Scholar
Minekawa K, Amino K, Matsuo T. A courtship behavior that makes monandrous females polyandrous. Evolution. 2020;74(11):2483–93. https://doiorg.publicaciones.saludcastillayleon.es/10.1111/evo.14098.
Article PubMed Google Scholar
Monnin T. 2006. Chemical recognition of reproductive status in social insects. In: Annales Zoologici Fennici. JSTOR. p. 515–530. http://www.jstor.org/stable/23736759.
Museridze M, Ceolin S, Mühling B, Ramanathan S, Barmina O, Sekhar PS, Gompel N. 2024. Entangled and non-modular enhancer sequences producing independent spatial activities. :2024.07.08.602541. https://doiorg.publicaciones.saludcastillayleon.es/10.1101/2024.07.08.602541. [accessed 2024 Sep 13]. https://www.biorxiv.org/content/https://doiorg.publicaciones.saludcastillayleon.es/10.1101/2024.07.08.602541v1.
Nachman MW, Hoekstra HE, D’Agostino SL. The genetic basis of adaptive melanism in pocket mice. Proc Natl Acad Sci. 2003;100(9):5268–73. https://doiorg.publicaciones.saludcastillayleon.es/10.1073/pnas.0431157100.
Article CAS PubMed PubMed Central Google Scholar
Ng SH, Shankar S, Shikichi Y, Akasaka K, Mori K, Yew JY. Pheromone evolution and sexual behavior in Drosophila are shaped by male sensory exploitation of other males. Proc Natl Acad Sci. 2014;111(8):3056–61. https://doiorg.publicaciones.saludcastillayleon.es/10.1073/pnas.1313615111.
Article CAS PubMed PubMed Central Google Scholar
Ohtsuki S, Levine M, Cai HN. Different core promoters possess distinct regulatory activities in the Drosophila embryo. Genes Dev. 1998;12(4):547–56.
Article CAS PubMed PubMed Central Google Scholar
Pei X-J, Fan Y-L, Bai Y, Bai T-T, Schal C, Zhang Z-F, Chen N, Li S, Liu T-X. Modulation of fatty acid elongation in cockroaches sustains sexually dimorphic hydrocarbons and female attractiveness. PLOS Biol. 2021;19(7): e3001330. https://doiorg.publicaciones.saludcastillayleon.es/10.1371/journal.pbio.3001330.
Article CAS PubMed PubMed Central Google Scholar
Petersen KR, Streett DA, Gerritsen AT, Hunter SS, Settles ML. 2015. Super deduper, fast PCR duplicate detection in fastq files. In: Proceedings of the 6th ACM Conference on Bioinformatics, Computational Biology and Health Informatics. New York, NY, USA: Association for Computing Machinery. (BCB ’15). p. 491–492. [accessed 2022 Dec 6]. https://doiorg.publicaciones.saludcastillayleon.es/10.1145/2808719.2811568.
Pfaffl MW. A new mathematical model for relative quantification in real-time RT–PCR. Nucleic Acids Res. 2001;29(9): e45. https://doiorg.publicaciones.saludcastillayleon.es/10.1093/nar/29.9.e45.
Article CAS PubMed PubMed Central Google Scholar
Ponton F, Chapuis M-P, Pernice M, Sword GA, Simpson SJ. Evaluation of potential reference genes for reverse transcription-qPCR studies of physiological responses in Drosophila melanogaster. J Insect Physiol. 2011;57(6):840–50. https://doiorg.publicaciones.saludcastillayleon.es/10.1016/j.jinsphys.2011.03.014.
Article CAS PubMed Google Scholar
Qiu Y, Tittiger C, Wicker-Thomas C, Le Goff G, Young S, Wajnberg E, Fricaux T, Taquet N, Blomquist GJ, Feyereisen R. An insect-specific P450 oxidative decarbonylase for cuticular hydrocarbon biosynthesis. Proc Natl Acad Sci U S A. 2012;109(37):14858–63. https://doiorg.publicaciones.saludcastillayleon.es/10.1073/pnas.1208650109.
Article PubMed PubMed Central Google Scholar
Ritchie ME, Phipson B, Wu D, Hu Y, Law CW, Shi W, Smyth GK. limma powers differential expression analyses for RNA-sequencing and microarray studies. Nucleic Acids Res. 2015;43(7): e47. https://doiorg.publicaciones.saludcastillayleon.es/10.1093/nar/gkv007.
Article CAS PubMed PubMed Central Google Scholar
Robinson MD, McCarthy DJ, Smyth GK. edgeR: a Bioconductor package for differential expression analysis of digital gene expression data. Bioinformatics. 2010;26(1):139–40. https://doiorg.publicaciones.saludcastillayleon.es/10.1093/bioinformatics/btp616.
Article CAS PubMed Google Scholar
Robinson MD, Oshlack A. A scaling normalization method for differential expression analysis of RNA-seq data. Genome Biol. 2010;11(3):R25. https://doiorg.publicaciones.saludcastillayleon.es/10.1186/gb-2010-11-3-r25.
Article CAS PubMed PubMed Central Google Scholar
Rogers WA, Salomone JR, Tacy DJ, Camino EM, Davis KA, Rebeiz M, Williams TM. Recurrent Modification of a Conserved Cis-Regulatory Element Underlies Fruit Fly Pigmentation Diversity. PLOS Genet. 2013;9(8): e1003740. https://doiorg.publicaciones.saludcastillayleon.es/10.1371/journal.pgen.1003740.
Article CAS PubMed PubMed Central Google Scholar
Rusuwa BB, Chung H, Allen SL, Frentiu FD, Chenoweth SF. Natural variation at a single gene generates sexual antagonism across fitness components in Drosophila. Curr Biol. 2022;32(14):3161-3169.e7. https://doiorg.publicaciones.saludcastillayleon.es/10.1016/j.cub.2022.05.038.
Article CAS PubMed Google Scholar
Ruta V, Datta SR, Vasconcelos ML, Freeland J, Looger LL, Axel R. A dimorphic pheromone circuit in Drosophila from sensory input to descending output. Nature. 2010;468(7324):686–90. https://doiorg.publicaciones.saludcastillayleon.es/10.1038/nature09554.
Article CAS PubMed Google Scholar
Schaefer HM, Ruxton GD. 2015. Signal Diversity, Sexual Selection, and Speciation. Annu Rev Ecol Evol Syst. 46(Volume 46, 2015):573–592. https://doiorg.publicaciones.saludcastillayleon.es/10.1146/annurev-ecolsys-112414-054158.
Schindelin J, Arganda-Carreras I, Frise E, Kaynig V, Longair M, Pietzsch T, Preibisch S, Rueden C, Saalfeld S, Schmid B, et al. Fiji: an open-source platform for biological-image analysis. Nat Methods. 2012;9(7):676–82. https://doiorg.publicaciones.saludcastillayleon.es/10.1038/nmeth.2019.
Article CAS PubMed Google Scholar
Scott Adams P. 2007. Data analysis and reporting. Taylor & Francis. [accessed 2022 Dec 6]. https://www.taylorfrancis.com/chapters/edit/https://doiorg.publicaciones.saludcastillayleon.es/10.4324/9780203967317-11/data-analysis-reporting-pamela-scott-adams.
Seeholzer LF, Seppo M, Stern DL, Ruta V. Evolution of a central neural circuit underlies Drosophila mate preferences. Nature. 2018;559(7715):564–9. https://doiorg.publicaciones.saludcastillayleon.es/10.1038/s41586-018-0322-9.
Article CAS PubMed PubMed Central Google Scholar
Senft AD, Macfarlan TS. Transposable elements shape the evolution of mammalian development. Nat Rev Genet. 2021;22(11):691–711. https://doiorg.publicaciones.saludcastillayleon.es/10.1038/s41576-021-00385-1.
Article CAS PubMed Google Scholar
Setoguchi S, Kudo A, Takanashi T, Ishikawa Y, Matsuo T. Social context-dependent modification of courtship behaviour in Drosophila prolongata. Proc R Soc Lond B Biol Sci. 2015;282(1818):20151377. https://doiorg.publicaciones.saludcastillayleon.es/10.1098/rspb.2015.1377.
Article Google Scholar
Setoguchi S, Takamori H, Aotsuka T, Sese J, Ishikawa Y, Matsuo T. Sexual dimorphism and courtship behavior in Drosophila prolongata. J Ethol. 2014;32(2):91–102. https://doiorg.publicaciones.saludcastillayleon.es/10.1007/s10164-014-0399-z.
Article Google Scholar
Shirangi TR, Dufour HD, Williams TM, Carroll SB. Rapid evolution of sex pheromone-producing enzyme expression in Drosophila. PLoS Biol. 2009;7(8): e1000168. https://doiorg.publicaciones.saludcastillayleon.es/10.1371/journal.pbio.1000168.
Article CAS PubMed PubMed Central Google Scholar
Shumate A, Salzberg SL. Liftoff: accurate mapping of gene annotations. Bioinforma Oxf Engl. 2020;37(12):1639–43. https://doiorg.publicaciones.saludcastillayleon.es/10.1093/bioinformatics/btaa1016.
Article CAS Google Scholar
Sievers F, Wilm A, Dineen D, Gibson TJ, Karplus K, Li W, Lopez R, McWilliam H, Remmert M, Söding J, et al. Fast, scalable generation of high-quality protein multiple sequence alignments using Clustal Omega. Mol Syst Biol. 2011;7:539. https://doiorg.publicaciones.saludcastillayleon.es/10.1038/msb.2011.75.
Article PubMed PubMed Central Google Scholar
Singh BK, Gupta JP. Two new and two unrecorded species of the genus Drosophila Fallen (Diptera: Drosophilidae) from Shillong, Meghalaya. India Proc Zool Soc Calcutta. 1977;30(1–2):31–8.
Google Scholar
Smadja C, Butlin RK. On the scent of speciation: the chemosensory system and its role in premating isolation. Heredity. 2008;102(1):77–97. https://doiorg.publicaciones.saludcastillayleon.es/10.1038/hdy.2008.55.
Article CAS PubMed Google Scholar
Smyth GK. 2004. Linear Models and Empirical Bayes Methods for Assessing Differential Expression in Microarray Experiments. Stat Appl Genet Mol Biol. 3(1). https://doiorg.publicaciones.saludcastillayleon.es/10.2202/1544-6115.1027. [accessed 2022 Nov 14]. https://www.degruyter.com/document/doi/https://doiorg.publicaciones.saludcastillayleon.es/10.2202/1544-6115.1027/html.
Steiger S, Stökl J. The role of sexual selection in the evolution of chemical signals in insects. Insects. 2014;5(2):423–38. https://doiorg.publicaciones.saludcastillayleon.es/10.3390/insects5020423.
Article PubMed PubMed Central Google Scholar
Stökl J, Steiger S. Evolutionary origin of insect pheromones. Curr Opin Insect Sci. 2017;24:36–42. https://doiorg.publicaciones.saludcastillayleon.es/10.1016/j.cois.2017.09.004.
Article PubMed Google Scholar
Storer J, Hubley R, Rosen J, Wheeler TJ, Smit AF. The Dfam community resource of transposable element families, sequence models, and genome annotations. Mob DNA. 2021;12(1):2. https://doiorg.publicaciones.saludcastillayleon.es/10.1186/s13100-020-00230-y.
Article CAS PubMed PubMed Central Google Scholar
Sundaram V, Cheng Y, Ma Z, Li D, Xing X, Edge P, Snyder MP, Wang T. Widespread contribution of transposable elements to the innovation of gene regulatory networks. Genome Res. 2014;24(12):1963–76. https://doiorg.publicaciones.saludcastillayleon.es/10.1101/gr.168872.113.
Article CAS PubMed PubMed Central Google Scholar
Suzuki T. 1980. 4, 8-Dimethyldecanal: The aggregation pheromone of the flour beetles, Tribolium castaneum and T. confusum (Coleoptera: Tenebrionidae). Agric Biol Chem. 44(10):2519–2520.
Szafer-Glusman E, Giansanti MG, Nishihama R, Bolival B, Pringle J, Gatti M, Fuller MT. A Role for Very-Long-Chain Fatty Acids in Furrow Ingression during Cytokinesis in Drosophila Spermatocytes. Curr Biol. 2008;18(18):1426–31. https://doiorg.publicaciones.saludcastillayleon.es/10.1016/j.cub.2008.08.061.
Article CAS PubMed PubMed Central Google Scholar
Takau A, Matsuo T. Contribution of visual stimuli to mating and fighting behaviors of Drosophila prolongata. Entomol Sci. 2022;25(4): e12529. https://doiorg.publicaciones.saludcastillayleon.es/10.1111/ens.12529.
Article Google Scholar
Tanaka K, Barmina O, Kopp A. Distinct developmental mechanisms underlie the evolutionary diversification of Drosophila sex combs. Proc Natl Acad Sci. 2009;106(12):4764–9. https://doiorg.publicaciones.saludcastillayleon.es/10.1073/pnas.0807875106.
Article PubMed PubMed Central Google Scholar
Tena JJ, Alonso ME, de la Calle-Mustienes E, Splinter E, de Laat W, Manzanares M, Gómez-Skarmeta JL. An evolutionarily conserved three-dimensional structure in the vertebrate Irx clusters facilitates enhancer sharing and coregulation. Nat Commun. 2011;2(1):310. https://doiorg.publicaciones.saludcastillayleon.es/10.1038/ncomms1301.
Article CAS PubMed Google Scholar
Thome BL. Termite-termite interactions: workers as an agonistic caste. Psyche (Stuttg). 1982;89(1–2):133–50.
Google Scholar
Toda H, Zhao X, Dickson BJ. The Drosophila female aphrodisiac pheromone activates ppk23+ sensory neurons to elicit male courtship behavior. Cell Rep. 2012;1(6):599–607. https://doiorg.publicaciones.saludcastillayleon.es/10.1016/j.celrep.2012.05.007.
Article CAS PubMed Google Scholar
Toyoshima N, Matsuo T. Fight outcome influences male mating success in Drosophila prolongata. J Ethol. 2023;41(2):119–27. https://doiorg.publicaciones.saludcastillayleon.es/10.1007/s10164-023-00778-1.
Article Google Scholar
Trizzino M, Park Y, Holsbach-Beltrame M, Aracena K, Mika K, Caliskan M, Perry GH, Lynch VJ, Brown CD. Transposable elements are the primary source of novelty in primate gene regulation. Genome Res. 2017;27(10):1623–33. https://doiorg.publicaciones.saludcastillayleon.es/10.1101/gr.218149.116.
Article CAS PubMed PubMed Central Google Scholar
Tu Z. Three novel families of miniature inverted-repeat transposable elements are associated with genes of the yellow fever mosquito. Aedes aegypti Proc Natl Acad Sci. 1997;94(14):7475–80. https://doiorg.publicaciones.saludcastillayleon.es/10.1073/pnas.94.14.7475.
Article CAS PubMed Google Scholar
Wang L, Anderson DJ. Identification of an aggression-promoting pheromone and its receptor neurons in Drosophila. Nature. 2010;463(7278):227–31.
Article CAS PubMed Google Scholar
Wang L, Han X, Mehren J, Hiroi M, Billeter J-C, Miyamoto T, Amrein H, Levine JD, Anderson DJ. Hierarchical chemosensory regulation of male-male social interactions in Drosophila. Nat Neurosci. 2011;14(6):757–62. https://doiorg.publicaciones.saludcastillayleon.es/10.1038/nn.2800.
Article CAS PubMed PubMed Central Google Scholar
Wang Z, Receveur JP, Pu J, Cong H, Richards C, Liang M, Chung H. 2022. Desiccation resistance differences in Drosophila species can be largely explained by variations in cuticular hydrocarbons. M. Riddiford L, editor. eLife. 11:e80859. https://doiorg.publicaciones.saludcastillayleon.es/10.7554/eLife.80859.
Weill M, Lutfalla G, Mogensen K, Chandre F, Berthomieu A, Berticat C, Pasteur N, Philips A, Fort P, Raymond M. Insecticide resistance in mosquito vectors. Nature. 2003;423(6936):136–7. https://doiorg.publicaciones.saludcastillayleon.es/10.1038/423136b.
Article CAS PubMed Google Scholar
Wells MM, Henry CS. The role of courtship songs in reproductive isolation among populations of green lacewings of the genus Chrysoperla (Neuroptera: Chrysopidae). Evolution. 1992;46(1):31–42. https://doiorg.publicaciones.saludcastillayleon.es/10.1111/j.1558-5646.1992.tb01982.x.
Article PubMed Google Scholar
West-Eberhard MJ. Darwin’s forgotten idea: the social essence of sexual selection. Neurosci Biobehav Rev. 2014;46(Pt 4):501–8. https://doiorg.publicaciones.saludcastillayleon.es/10.1016/j.neubiorev.2014.06.015.
Article PubMed Google Scholar
Wicker-Thomas C, Chertemps T. 2010. Molecular biology and genetics of hydrocarbon production. Insect Hydrocarb Biol Biochem Chem Ecol.:53–74.
Wicker-Thomas C, Garrido D, Bontonou G, Napal L, Mazuras N, Denis B, Rubin T, Parvy J-P, Montagne J. Flexible origin of hydrocarbon/pheromone precursors in Drosophila melanogaster. J Lipid Res. 2015;56(11):2094–101. https://doiorg.publicaciones.saludcastillayleon.es/10.1194/jlr.M060368.
Article CAS PubMed PubMed Central Google Scholar
Wicker-Thomas C, Guenachi I, Keita YF. Contribution of oenocytes and pheromones to courtship behaviour in Drosophila. BMC Biochem. 2009;10:21.
Article PubMed PubMed Central Google Scholar
Williams TM, Carroll SB. Genetic and molecular insights into the development and evolution of sexual dimorphism. Nat Rev Genet. 2009;10(11):797–804. https://doiorg.publicaciones.saludcastillayleon.es/10.1038/nrg2687.
Article CAS PubMed Google Scholar
Wu CI, Hollocher H, Begun DJ, Aquadro CF, Xu Y, Wu ML. Sexual isolation in Drosophila melanogaster: a possible case of incipient speciation. Proc Natl Acad Sci. 1995;92(7):2519–23. https://doiorg.publicaciones.saludcastillayleon.es/10.1073/pnas.92.7.2519.
Article CAS PubMed PubMed Central Google Scholar
Yew JY, Chung H. Insect pheromones: An overview of function, form, and discovery. Prog Lipid Res. 2015;59:88–105. https://doiorg.publicaciones.saludcastillayleon.es/10.1016/j.plipres.2015.06.001.
Article CAS PubMed Google Scholar
Yoshimizu T, Akutsu J, Matsuo T. An Indirect Cost of Male-Male Aggression Arising from Female Response. Zoolog Sci. 2022;39(6):514–20. https://doiorg.publicaciones.saludcastillayleon.es/10.2108/zs210116.
Article PubMed Google Scholar
Zwarts L, Versteven M, Callaerts P. Genetics and neurobiology of aggression in Drosophila. Fly (Austin). 2012;6(1):35–48. https://doiorg.publicaciones.saludcastillayleon.es/10.4161/fly.19249.
Article CAS PubMed PubMed Central Google Scholar

Download references

Acknowledgements

We thank Dr. Jocelyn Miller (University of California, Riverside) for providing synthetic 9-heptacosene, Logan Blair and David Luecke for advice on experiments and comments on the manuscript, and Majken Horton, Madison Hypes, Yingxin Su, and Olga Barmina for technical assistance.

Funding

This work was supported by NIH grant R35 GM122592 to AK, JSPS grants No. 18H02507 and No. 24 K01767 to TM, funds from the UC Davis Center for Population Biology to YL, and David and Lucile Packard Foundation grant 2014-40378 to SRR. The funders had no role in the conceptualization, design, data collection, analysis, decision to publish, or preparation of the manuscript.

Author information

Yige Luo and Ayumi Takau contributed equally to this work.

Authors and Affiliations

Department of Evolution and Ecology, University of California, Davis, USA
Yige Luo, Jiaxun Li, Tiezheng Fan, Ben R. Hopkins, Yvonne Le, Santiago R. Ramirez & Artyom Kopp
Department of Agricultural and Environmental Biology, The University of Tokyo, Tokyo, Japan
Ayumi Takau & Takashi Matsuo
Georgia Institute of Technology, 225 North Avenue NW, Atlanta, GA, 30332, USA
Jiaxun Li
San Joaquin General Hospital, 500 W Hospital Road, French Camp, CA, 95231, USA
Yvonne Le

Authors

Yige Luo
View author publications
You can also search for this author inPubMed Google Scholar
Ayumi Takau
View author publications
You can also search for this author inPubMed Google Scholar
Jiaxun Li
View author publications
You can also search for this author inPubMed Google Scholar
Tiezheng Fan
View author publications
You can also search for this author inPubMed Google Scholar
Ben R. Hopkins
View author publications
You can also search for this author inPubMed Google Scholar
Yvonne Le
View author publications
You can also search for this author inPubMed Google Scholar
Santiago R. Ramirez
View author publications
You can also search for this author inPubMed Google Scholar
Takashi Matsuo
View author publications
You can also search for this author inPubMed Google Scholar
Artyom Kopp
View author publications
You can also search for this author inPubMed Google Scholar

Contributions

Y Luo designed the project, performed GC–MS, RNA-seq and other molecular experiments, developed bioinformatic workflows, performed statistical modeling and analysis of transcriptomic and genomic data, and wrote the manuscript AT generated eloF mutants and performed behavioral experiments JL generated a transgenic reporter construct TF assisted with oenocyte dissections BRH edited the manuscript Y Le assisted with Drosophila experiments SRR provided methodology and resources for the GC–MS analysis, helped with the troubleshooting and interpretation of GC–MS experiments, and commented on the manuscript TM designed genetic and behavioral experiments, performed and behavioral experiments, and wrote sections of the manuscript AK designed and directed the project and edited the manuscript.

Corresponding author

Correspondence to Artyom Kopp.

Ethics declarations

Ethics approval and consent to participate

Not applicable.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.

Additional information

Publisher’s Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Supplementary Information

12915_2025_2220_MOESM1_ESM.tif

Additional file 1: Figure S1. Variable effect of perfuming on male-female interactions. Stacked bar plots showing success rates of (A) Encounter, (B) Threatening, (C) Courtship (defined as the proportion of males that continued to court the female after the initial encounter), and (D) Leg vibration (where the male vigorously shakes the female’s abdomen with his front legs) across three perfuming conditions (N = 32 for each treatment). N.S., nonsignificant results based on comparison between treatment and control in a logistic regression model. P values are as follows: *** p < 0.001, ** p < 0.01, *, p < 0.05, p < 0.1.

12915_2025_2220_MOESM2_ESM.tif

Additional file 2: Figure S2. Male-biased long-chain CHCs are not transferred to females during mating. Boxplots showing the abundance of 9T, 9P and 9H, with cis-vaccenyl-acetate (cVA) as a positive control. Shown are control wild-type females, WT females that did not mate with a WT male, WT females that mated with a WT male, and control WT males. Pheromone abundance is measured in nanograms per fly and shown on log10 scale. Overlayed jitter points are samples of each sex * mating status combination, color-coded by genotype. Significance results of all pairwise comparisons (Tukey HSD test followed by significant omnibus ANOVA F-tests) are summarized in the format of compact letter display (using R packages"multicomp" and "lsmeans").

12915_2025_2220_MOESM3_ESM.tif

Additional file 3: Figure S3. Differential gene expression between males and females of D. prolongata. (A) Sex differentiation and substance transport. (B) Terminal processes of amino acid metabolism. Directed Acyclic Graph (DAG) of significant GO terms and their parent terms in biological processes. Significant (p < 0.05) and non-significant GO terms are color-coded and represented by ellipses and rectangular boxes, respectively. Significant GO terms can be underrepresented (blue) or overrepresented (red) based on Fisher’s exact test. Arrows indicate hierarchical relationships. GO terms at the same hierarchical level are placed at the same vertical position. Significant GO terms that are also enriched between males and females of D. carrolli have dashed borders.

Additional file 4: Table S1. Significant GO terms in the comparison between males of D. prolongata and D. carrolli.

12915_2025_2220_MOESM5_ESM.tif

Additional file 5: Figure S4. Differential gene expression between males of D. prolongata and D. carrolli. (A) Terminal lipid metabolism processes. (B) Substance transport, development, and reproduction. (C) Amino acid metabolism processes. (D) Signal transduction, cell-cell adhesion, and aggressive behavior. Directed Acyclic Graph (DAG) of significant GO terms and their parent terms in biological processes. Significant (p < 0.05) and non-significant GO terms are color-coded and represented by ellipses and rectangular boxes, respectively. Significant GO terms can be underrepresented (blue) or overrepresented (red) based on Fisher’s exact test. Arrows indicate hierarchical relationships. GO terms at the same level are positioned at the same vertical position. GO terms under the lipid metabolic process (GO:0006629) are connected by green arrows and have green borders in (A). Significant GO terms that are also enriched between females of D. prolongata and D. carrolli have dashed borders.

Additional file 6: Table S2. Cuticular lipid description.

12915_2025_2220_MOESM7_ESM.tif

Additional file 7: Figure S5. eloF is responsible for elongating the precursors of long-chain 9-monoenes. Boxplots showing the abundance of 9H (A), 9P (B), 9T (C), and the aggregate 9-Monoenes (D) across genotypes in each sex, with abundance in nanograms shown on log₁₀ scale. Overlayed jitter points are samples of each sex * genotype combination, with color-coded genotypes. Significance of all pairwise comparisons (Tukey HSD test followed by significant omnibus ANOVA F-tests) are summarized in the format of compact letter display (using R packages "multicomp" and "lsmeans"). Note the decrease in the abundance of 9P and 9H in both sexes, and an increase in the abundance of 9T in males, in eloF mutants, while the total abundance of 9-monoenes remains approximately constant.

12915_2025_2220_MOESM8_ESM.tif

Additional file 8: Figure S6. eloF is responsible for elongating the precursors of long-chain CHCs. Boxplots showing the log₂ ratio of 9-monoenes (A-B) and other CHCs (C-D) with adjacent odd-numbered carbons across genotypes in each sex.

12915_2025_2220_MOESM9_ESM.tif

Additional file 9: Figure S7. Little effect of eloF mutations on total CHC abundance. Boxplots showing the aggregate abundance of 7-Monoenes (A), 9-Monoenes (B), branched alkanes (C), straight-chain alkanes (D), and overall CHCs (E) across genotypes in each sex, with abundance in nanograms shown on log₁₀ scale.

Additional file 10: Table S3. eloF[-] mutant behavior.

12915_2025_2220_MOESM11_ESM.tif

Additional file 11: Figure S8. No fixed protein sequence differences between D. prolongata and D. carrolli eloF orthologs. Multiple alignment on translated amino acid sequences across five species in the rhopaloa species subgroup, with species phylogeny on the left and the consensus sequence at the bottom. Numbers above the consensus sequence are coordinates showing the consensus length (257 AA). For the alleles of each species, site-wise disagreement from the consensus is represented in gray shade. For D. carrolli and D. prolongata, single nucleotide polymorphisms (SNPs) that lead to changes in amino acids are highlighted in red. Polymorphic sites are represented in dashed rectangles. In D. prolongata, amino acid sequences deleted in one CRISPR mutant (eloF[-] Δ45) are in cyan shade. Feature annotations are displayed above the protein sequence, with dark gray boxes representing eloF exons. All features have their direction labeled as arrowheads.

Additional file 12: Table S4. Sites segregating in the coding region of eloF in D. prolongata and D. carrolli.

12915_2025_2220_MOESM13_ESM.tif

Additional file 13: Figure S9. D. prolongata-specific honghaier insertion in the downstream region of eloF. (A) The downstream region of eloF in D. prolongata, showing the insertion of the TE-like repetitive element honghaier. Feature annotations are displayed below DNA sequence, with the red box representing the honghaier insertion, the orange box representing its predicted ORF, and the purple box showing the BLAST hit to the DNAREP_DM1 transposable element (Dfam). The motif track shows putative binding sites for transcription factors including dsx (JASPAR, dark green), dsx (FlyReg, light green), dmrt99B (JASPAR, yellow-green), and bab1 (iDMMPMM, pink). The repeat track includes short TGTC repeats (cyan) and three de novo motifs: MEME-1 (brown), MEME-2 (pink), and MEME-3 (steel blue). (B) Alignment of the conserved downstream region of eloF (shaded region in A) across species, with species phylogeny on the left and consensus sequence at the bottom. Numbers above the DNA sequence are coordinates showing the length of the consensus (606 bp) and alignment (529 bp). For the alleles of each species, nucleotide-wise disagreement from the consensus is represented in a color-coded vertical line for nucleotide substitutions (A: red, C: blue, G: yellow, T: green), and a horizontal line for nucleotide deletions. The track of percent identity is color coded as follows: green for perfect (100%) agreement, yellow-green for intermediate (30-99%) agreement, and red for low (<30%) agreement. All features have their direction labeled as arrowheads when applicable.

Additional file 14: Table S5. High fidelity honghaier sequence occurrence.

12915_2025_2220_MOESM15_ESM.tif

Additional file 15: Figure S10. The upstream region of eloF is conserved in the rhopaloa species subgroup. Multiple alignment of the upstream region of eloF, with schematic gene structure displayed on top. The track of percent identity is color-coded as follows: green for perfect (100%) agreement, yellow-green for intermediate (30-99%) agreement, and red for low (<30%) agreement. Numbers above the percent identity track are coordinates showing the length of the consensus (310 bp) and alignment (309 bp). For alleles from each species, nucleotide-wise disagreement from the consensus is represented in a color-coded vertical line for nucleotide substitutions (A: red, C: blue, G: yellow, T: green), and a horizontal line for nucleotide deletions. Predicted transcription factor (TF) binding motifs are displayed below the DNA sequence as follows: dsx (JASPAR, dark green); dsx (FlyReg, light green); bab1 (iDMMPMM, pink). All features have their direction labeled as arrowheads when applicable.

12915_2025_2220_MOESM16_ESM.tif

Additional file 16: Figure S11. The intron of eloF in conserved in the rhopaloa species subgroup. Multiple alignment of the intronic region of eloF, with genomic context displayed on top. Numbers above the DNA sequence are coordinates showing the consensus length (68 bp). For the alleles of each species, site-wise disagreement from the consensus is represented in gray shade. No sex (dsx) or tissue (bab1) motifs were identified.

12915_2025_2220_MOESM17_ESM.tif

Additional file 17: Figure S12. Design and analysis of GFP reporter constructs containing eloF sequences. (A) Schematic illustration of the eloF locus and the two flanking genes. (B) “Long” constructs containing the entire eloF locus including flanking sequences. Dpro eloF WT^(l) and Dcar eloF WT^(l) carry wild-type eloF loci from D. prolongata and D. carrolli, respectively. The other two constructs were made by removing the honghaier TE insertion from the D. prolongata sequence (Dpro eloF WT^(l)-TE) or adding the D. prolongata honghaier insertion to the D. carrolli sequence (Dcar eloF WT^(l)+TE). The eloF locus is placed into the pGreenFriend vector in the forward orientation, so that eloF is transcribed in the same direction as GFP while the honghaier insertion is in the opposite direction. (C).“Short” constructs containing only the downstream eloF sequences. As in the “long” constructs, two constructs contain the wild-type alleles from D. prolongata and D. carrolli, while the other two were made by TE swap. Here, the downstream eloF sequences are placed into the pGreenFriend vector in the flipped orientation, so that the direction of the honghaier insertion is the same as GFP transcription. In (B) and (C), alignment coordinates are displayed on top. Black lines indicate disagreement between theD. prolongata and D. carrolli alleles, vertical for single nucleotide variants and horizontal for short indels. Feature annotations are displayed below DNA sequence, with green box representing genes, yellow box representing CDS, red box representing the honghaier insertion, and the orange box representing its predicted ORF. All features have their direction labeled by arrowheads when applicable. (D) Schematic illustration of pGreenFriend vector, where GFP is driven by the Drosophila synthetic core promoter (DSCP, yellow-green). (E) In the “long” constructs containing the entire eloF locus, GFP reporter expression is low and does not differ significantly between genotypes. (F) In the “short” constructs containing only the downstream eloF region, all eloF alleles have similar effects on GFP expression in transgenic D. melanogaster, with a slight (~2-fold) female bias (p <0.05). Y axis shows the relative expression of eloF with respect to the reference gene Rpl32 (measured in ΔCt). For each group, three biological replicates, each an average of three technical replicates, are represented by jitter points. Males are in filled symbols; females are in open symbols. Wild-type reporter alleles are represented with circles and TE-swapped alleles with triangles.

12915_2025_2220_MOESM18_ESM.docx

Additional file 18: Table S6. qPCR analysis of GFP transcript expression driven by eloF “long” constructs (complete eloF locus including flanking regions).

12915_2025_2220_MOESM19_ESM.tif

Additional file 19: Figure S13. eloF downstream sequences drive GFP expression in adult abdominal oenocytes. Confocal images of GFP protein stained with anti-GFP antibodies, showing dissected male and female dorsal abdominal body walls. Non-transgenic yw flies are used as a negative control. Transgenic flies carry the “short” constructs containing the eloF downstream region (see Additional file 17: Fig S12).

Additional file 20: Table S7. RNA-seq data summary.

Additional file 21: Table S8. RNA-seq read mapping statistics.

12915_2025_2220_MOESM22_ESM.csv

Additional file 22: Data File 1. List of genes differentially expressed between D. prolongata males and D. prolongata females.

12915_2025_2220_MOESM23_ESM.csv

Additional file 23: Data File 2. List of genes differentially expressed between D. prolongata males and D. carrolli males.

12915_2025_2220_MOESM24_ESM.csv

Additional file 24: Data File 3. List of genes differentially expressed between D. carrolli males and D. carrolli females.

12915_2025_2220_MOESM25_ESM.csv

Additional file 25: Data File 4. List of genes differentially expressed between D. prolongata females andD. carrolli females.

12915_2025_2220_MOESM26_ESM.csv

Additional file 26: Data File 5. Genes for which the direction or magnitude of male-female (MF) differences in expression levels differs between D. prolongata and D. carrolli.

12915_2025_2220_MOESM27_ESM.csv

Additional file 27: Data File 6. Three one-way comparisons used to identify candidate genes for sex-specific pheromone divergence, showing gene names ordered by union P values.

Additional file 28: Data File 7. Genes expressed in the adult oenocytes of D. melanogaster.

12915_2025_2220_MOESM29_ESM.tif

Additional file 29: Figure S14. All primers used for quantitative PCR have near-perfect amplification performance. Standard curves of Rpl32 are based on cDNA from mixed-sex whole-body RNA of D. prolongata (A) and D. carrolli (B). Standard curves of eloF are based on cDNA from mixed-sex whole-body RNA of D. prolongata (C) and D. carrolli (D). Standard curve of GFP is based on empty pGreenFriend vector (E). Dilution factors are 10-fold for (A), (B) and (E); 8-fold for (C), and 3-fold for (D). Points represent average values, with error bars showing standard deviations calculated from three technical replicates. Lines represent the best linear fit, showing the estimated equation and coefficient of determination (R²).

Additional file 30: Table S9. Primers used for cloning and qPCR.

12915_2025_2220_MOESM31_ESM.tif

Additional file 31: Figure S15. Calibration of candidate CHC transfer based on unperfumed CHCs. Panels of calibration standard curves for 9H treatment (A), 9P treatment (B), and hexane control (C). Each panel represents CHC profiles sampled from an independent group of 8 flies subjected to the same experimental procedures as those used in behavioral studies. Within each panel, each point represents individual CHC, with its abundance before perfuming procedure (pre-abundance) indicated as the x coordinate value, and abundance after perfuming procedure (post-abundance) indicated as the y coordinate value. CHCs other than the spiked-in compound (9P or 9H) were used to build the standard curve (dotted blue line), showing the estimated equation and coefficient of determination (R²). Counterfactual post-perfuming abundances of 9P and 9H are estimated from standard curves (light green point for 9P and orange point for 9H) as if no synthetic compounds were added, along with their 95% confidence interval (error bars). Abundance is measured in nanograms and standardized to abundance per individual fly.

12915_2025_2220_MOESM32_ESM.xlsx

Additional file 32: Data File 8. Oenocyte-specific marker genes in D. melanogaster, showing log fold changes (oenocytes vs. other tissues) and associated P values.

Additional file 33: Data File 9. 1kb upstream DNA sequences of the oenocyte marker genes listed in Data File 8.

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.

Reprints and permissions

About this article

Cite this article

Luo, Y., Takau, A., Li, J. et al. Regulatory changes in the fatty acid elongase eloF underlie the evolution of sex-specific pheromone profiles in Drosophila prolongata. BMC Biol 23, 117 (2025). https://doiorg.publicaciones.saludcastillayleon.es/10.1186/s12915-025-02220-z

Download citation

Received: 29 October 2024
Accepted: 17 April 2025
Published: 30 April 2025
DOI: https://doiorg.publicaciones.saludcastillayleon.es/10.1186/s12915-025-02220-z

Regulatory changes in the fatty acid elongase eloF underlie the evolution of sex-specific pheromone profiles in Drosophila prolongata

Abstract

Background

Results

Conclusions

Background

Results

Perfuming with male-specific pheromones reduces female mating success

Gene expression shows stronger sexual dimorphism in D. prolongata than in D. carrolli

Sexually dimorphic and species-biased genes are enriched for lipid metabolism functions

Candidate genes show increased male bias in D. prolongata

Functionally related genes are distributed in local genomic blocks

Fatty-acid elongase eloF shows extremely male-biased expression in D. prolongata

Loss of eloF partially feminizes the pheromone profile of male D. prolongata

The elongase gene cluster including eloF is conserved

A species-specific transposable element insertion in eloF in D. prolongata

eloF downstream region drives gene expression in oenocytes

eloF allele from D. prolongata drives sexually dimorphic expression in D. melanogaster

Discussion

Male-specific hydrocarbons reduce female mating success

eloF is a major gene controlling long-chain pheromone production in male D. prolongata

Interaction of cis- and trans-regulatory factors in the control of sex-biased eloF expression

Multiple genes likely contribute to sex-specific pheromone profiles in D. prolongata

Conclusions

Methods

Fly rearing and dissection for gene expression analysis

RNA extraction

Library construction, sequencing, and read mapping

Differential gene expression analysis

Gene Ontology (GO) enrichment analysis

Quantitative polymerase chain reaction (qPCR) analysis of gene expression

Cuticular lipids extraction

Female perfuming experiments

Behavioral assays

Statistical analysis of behavioral changes

Statistical estimation of hydrocarbon transfer

Gene editing by CRISPR/Cas9 mutagenesis

Statistical analysis on mutant CHC profiles

Comparative sequence analysis

Transcription factor (TF) binding motif analysis

Design of reporter constructs

Transgenic strains

Tissue dissection and antibody staining

Confocal microscopy

Data availability

Abbreviations

References

Acknowledgements

Funding

Author information

Authors and Affiliations

Contributions

Corresponding author

Ethics declarations

Ethics approval and consent to participate

Consent for publication

Competing interests

Additional information

Publisher’s Note

Supplementary Information

Rights and permissions

About this article

Cite this article

Share this article

Keywords

BMC Biology

Contact us