A comparative analysis of highly conserved sex-determining genes between Apis mellifera and Drosophila melanogaster

INTRODUCTION

Most metazoans express two sexual phenotypes, and the choice of sexual cell fate is a developmental process similar to the choice between becoming an epidermal cell or a neuron. One of the most completely characterized genetic regulatory hierarchies is that regulating sexual differentiation. Sex determination during development in Drosophila melanogaster (Diptera) is transmitted through a cascade of regulatory genes to the terminal differentiation genes, and their products are responsible for the sexually dimorphic characteristics of the adult (Baker and Ridge, 1980; Baker et al., 1987; Baker, 1989; Burtis, 1993). The initial signal at the top of the regulatory cascade shows surprising forms of solutions to switch the genetic program to produce two alternative developmental fates, a male or a female. In organisms with heteromorphic sex chromosomes, males and females can be either heterogametic (i.e., XY) or homogametic (i.e., XX), independent of the phylogenetic distance. Furthermore, some organisms have no distinct sex chromosomes; males and females are produced by a specific composition in the alleles of a single sex-determining locus or combination of loci (i.e., Hymenopteran haploid-diploid system) (White, 1973; Bull, 1983; Marín and Baker, 1998). Despite the considerable variation in the initial signal at the top of the regulatory hierarchy, sex determination shares some general features at the bottom of the pathway in all species studied elsewhere (Marín and Baker, 1998; Raymond et al., 1998; Graham et al., 2003).

The gene doublesex (dsx) at the bottom of sex-determination hierarchy of the insects D. melanogaster (Burtis and Baker, 1989), Bombyx mori (Ohbayashi et al., 2000), Megaselia scalaris (Kuhn et al., 2000), Musca domestica (Hediger et al., 2004), Bactrocera tryoni (Shearman and Frommer, 1998), and Ceratitis capitata (Graham et al., 2003) is alternatively spliced to encode a sex-specific transcription factor with a DM DNA-binding motif. Genes that encode proteins containing a DM domain are required in male development in the nematode Caenorhabditis elegans (male abnormal-3, mab-3) and humans (dmrt1), suggesting that at least some aspects of sexual regulation have a common evolutionary origin (Erdman and Burtis, 1993; Cline and Meyer, 1996; Raymond et al., 1998; Raymond et al., 2000). The regulation of yolk protein gene expression in D. melanogaster is the best-characterized DSX function at the molecular level. The binding of DSX-M (male-specific protein) to the fat body enhancer in the promoter region of yolk proteins (yp-1 and yp-2) represses the expression of these genes, whereas binding of DSX-F (female-specific protein) to the same sequences cooperates with other factors (BZIP-1) to activate transcription of yolk proteins in the fat body (An and Wensink, 1995a,b). Genitalia formation in flies also requires the integration of several pathways: 1) decapentaplegic (dpp), hedgehog (hh) and wingless (wg) to inform position in the genital disc, 2) the homeotic gene abdominal-B (abd-B) to define differences between genital disc segments, and 3) dsx to play the major role in the formation of genitalia and analia (Sanchez et al., 2001; Sánchez and Guerrero, 2001; Keisman and Baker, 2001; Keisman et al., 2001; Estrada et al., 2003; DeFalco et al., 2004). Other functional roles of dsx are related to the target genes responsible for sex-specific abdominal pigmentation (bric-a-brac, bab) and male courtship behavior (takeout and fruitless) (Kopp et al., 2000; Dauwalder et al., 2002; Kyriacou, 2005).

When compared to D. melanogaster, little is known about the molecular genetic basis of haplodiploid sex determination. About 20% of animal species are haplodiploid; in these species, unfertilized haploid eggs develop into males and fertilized diploid eggs into females. The honey bee Apis mellifera (Hymenoptera) is an emergent model organism, and recently a simplified model of sex determination in honey bees has been published to explain the initial signal that depends on one allele or two different alleles of a single gene, the complementary sex determiner (csd). Heterozygosity generates an active protein that initiates female development, while hemizygosity/homozygosity results in a non-active CSD protein and default male development. An orthologous dsx transcript has been reported to occur in honey bees, but this study has not been published yet (Beye et al., 2003; Beye, 2004).

Currently, the honey bee genome is in the fourth release, Amel_v3.0 (http://www.hgsc.bcm.tmc.edu/projects/honeybee), and it opens a new possibility for a comparative approach to functional genomics between different insect species. This can be fruitful in exploring conserved and variable genetic elements in a biological process among different species (i.e., D. melanogaster (Dm) and A. mellifera (Am)). We identified in the A. mellifera genome 13 genes homologous to D. melanogaster genes that participate in the sex determination and differentiation process. We focused preferentially on the dsx sequence because it is one of the most conserved genes involved in the sex determination mechanism in D. melanogaster and other insects. Amdsx encodes a putative protein with a DNA-binding motif named DM domain. Comparisons between DM-related gene family members in the fruitfly and honey bee indicated the most probable dsx ortholog in A. mellifera. Multiple sequence comparisons among available insect dsx genes pointed to the most probable Amdsx candidate. Finally, a male-specific partial Amdsx gene transcript was found in total RNA of the honey bee larva.

MATERIAL AND METHODS

Computational analysis

A list of annotated genes involved in sex determination (GO:0007530) and sex differentiation (GO:0007548) in D. melanogaster was retrieved from Gene Ontology (GO) database (http://www.geneontology.org/) without redundant UniProt and isoform sequences. Similar A. mellifera sequences were searched on the predicted protein official set (ftp://beeftp:analysis@ftp.beegenome.hgsc.bcm.tmc.edu/GenePredictions/) at the Human Genome Sequencing Center of Baylor College of Medicine (HGSC-BCM) using BLASTP (with expected value less than 1e-18 and soft masking options). Only the alignments with more than 40% identity are reported in this manuscript. Version 3.0 of the whole genome shotgun sequence assembly (ftp://ftp.hgsc.bcm.tmc.edu/pub/data/Amellifera) was used to infer the partial genomic architecture of the putative Amdsx gene, as well as the genomic location of other hypothetical conserved genes. Insect dsx gene sequences were extracted from GenBank (http://www.ncbi.nlm.nih.gov/): Bactrocera oleae (CAD67987), Anastrepha obliqua (AAY25167), B. mori (BAB13472), D. melanogaster (AAF54169), M. scalaris (AAK38832), M. domestica (AAR23813), Anopheles gambiae (AAX48939), and C. capitata (AAN63597).

A pipeline was designed using Python programming tools (http://www.python.org/) and Biopython modules (http://www.biopython.org/) to identify the most closely related sequences in different proteomes of D. melanogaster and A. mellifera. The method used for identifying a group of related (gene family) dsx sequences between these two species was based on reciprocal best hit of alignments (Chervitz et al., 1998); only matched sequences with a very conservative statistical value (e-value < 1e-18) were considered.

Multiple protein sequence comparisons were generated using CLUSTALW (Thompson et al., 1994) with default parameters (gap opening 10.0, gap extension penalty 0.1). Protein distance-matrix analysis was performed with PHYLIP, version 3.64 (Felsenstein, 1989). One thousand bootstrap replications were carried out using the program SEQBOOT. Distance matrix was created using PROTDIST with PAM matrix (Dayhoff, 1979) to construct neighbor-joining (Saitou and Nei, 1987) trees using the NEIGHBOR program. The consensus tree was calculated for each set of trees produced in the program NEIGHBOR.

The DSX protein-conserved domains were predicted with Pfam Hidden Markov Models (HMM) for “DM DNA-binding domain” (Pfam accession: PF00751) and for “DMRTA motif” (Pfam accession: PF03474) (http://www.sanger.ac.uk/Software/Pfam/) using the program HMMER (the current release is 2.3.2, see http://hmmer.wustl.edu/). Only the very conserved motifs were considered, with expected values less than 1e-10.

DNA isolation, cloning, PCR amplification, and DNA sequencing

First-strand cDNA was synthesized by RT-PCR (SuperScript II; Invitrogen) from 10 µg total RNA isolated from female and male embryos and larvae. Aliquots of first-strand cDNA products were employed in PCR reactions using PCR master mix (Promega). The thermal cycling program consisted of 1 min at 95°C, followed by 35 cycles of 1 min at 95°C, 1 min at 59°C, 1 min at 72°C, and a final extension step at 72°C for 10 min. Three different primers were used in different combinations for PCR amplification, where one set was used to amplify a non-sex-specific region of the transcript (P1: 5’-TGCGAGAAGTGTAAGATCAC-3’ and P2: 5’-GTGCTCCAATAGAATTTCCAC-3’) and another one was used to amplify a male-specific region of the transcript (P1: 5’-TGCGAGAAGTGTAAGATCAC-3’ and P3: 5’-GCACGACTAGGTTGGGACAT-3’). The amplification products were analyzed by electrophoresis on 1% agarose gels. The non-sex-specific fragments of about 443 bp and the male-specific fragments of about 560 bp corresponding to dsx gene partial sequence were purified and subcloned into the EcoRI site of pGEM-T easy plasmid (Promega). Insert-containing plasmids were subjected to sequencing reactions using the M13-reverse and M13-forward universal primers.

RESULTS

Computational localization of conserved honey bee genes involved in sex determination and sex differentiation processes

Table 1 shows a total of 13 protein sequences of 29 non-redundant (only Flybase sequences) initial sequences that were retrieved from GO database annotated to be involved in sex determination (GO:0007530) and sex differentiation (GO:0007548) of the fruit fly D. melanogaster. All 29 sequences were aligned against the predicted honey bee protein database (Official set of predicted protein sequences from Baylor College of Medicine), and only the very conserved genes were selected (at least 40% of identity and e-value < 1e-18). There are three gene products commonly participating in both biological processes in the fruifly, and they are highly conserved in the honey bee genome, namely sexlethal (fruitfly sxl FBgn0003659 is 68% identical to the honey bee GB13127-PA), intersex (fruitfly ix FBgn0001276 is 43% identical to honey bee GB19364-PA) and dsx (fruifly dsx FBgn0000504 is 51% identical to honey bee GB18426-PA). Among the others, 10 genes are important elements responsible for establishment of complete sexual dimorphism, but none one of them is as well studied and known as the dsx gene. In the following, we describe the main functional activity of the most conserved gene, dsx, already studied in other insects as regulators of sexual phenotype.

The dsx gene encodes a highly conserved zinc-finger transcription factor with a DM DNA domain. This domain is remarkable for a novel pattern of cysteines and histidines that bind to the DNA minor groove (Erdman and Burtis, 1993; Zhu et al., 2000). The DM domain was named on the basis of its occurence in DSX (D. melanogaster) and MAB-3 (C. elegans) (Raymond et al., 1998). The DM motif is conserved in metazoan sex determination. In humans, these genes are involved in testis differentiation, where dmrt1 and dmrt2 deletions are associated with XY sex reversal, even given the integrity of the male-determining gene, sex-determining region Y (sry) (Sinclair et al., 1990). In zebrafish and mammals, a DM gene, terra, appears to function in mesodermal patterning in both sexes (Meng et al., 1999; Volff et al., 2003). The dsx homolog may be the most ancient one at the bottom of the sex-determining cascade in metazoans, according to the “bottom-up” hypothesis (Wilkins, 1995).

The 13 genes reported in Table 1 are involved in three different levels of molecular regulation: alternative splicing, gene transcription and signal transduction. The six transcription factors are: dsx (Zn-finger with a DM-binding domain), ix (protein domain without DNA binding), fruitless (fru; Zn-finger with protein binding), deadpan (dpn; basic helix-loop-helix dimerization region, bHLH), dissatisfaction (dsf; Zn-finger with steroid hormone receptor activity), runt (run; RNA polymerase II activity), bab (helix-turn-helix), and sex combs reduced (scr; RNA polymerase II activity). Three genes participate in the splicing mechanism, they are: sexlethal (sxl), transformer2 (tra2), and female lethal (fl). Two genes participate in the signal transduction pathway, they are: hopscotch (hop) and protein kinase 61C (pk61C) (FlyBase - http://flybase.bio.indiana.edu/ and GO - http://www.geneontology.org/).

Except for the ix gene, all other genes turned out to be similar to more than one gene in the honey bee genome. It can be inferred that these homologous genes are members of gene families with at least one conserved shared domain. Four genes, fru, dsf, bab, and scr, have a remarkably large number of homologs, leading us to infer that they are members of transcription factor families that are very common in A. mellifera and D. melanogaster. Two different dsx- related genes were reported in this first analysis, and a comparison between all possible homologs in the genome of these two insects is necessary to predict the best candidate for the dsx ortholog.

Alignments of the DM-related family of proteins between D. melanogaster and A. mellifera

A list of eight DM-related protein sequences found in the D. melanogaster and A. mellifera databases was generated by reciprocal best hit of alignments, considering only matched sequences with a very conservative statistical value (e-value < 1e-18) (Chervitz et al., 1998). The phylogenetic relationship among the eight aligned products encoded by DM-related genes revealed four clusters of paralogs, in agreement with a previous study on the phylogeny of dsx-like genes and the evolution of sex determination (Ottolenghi et al., 2002) (Figure 1A). All related genes encode proteins with the DM domain (Pfam accession: PF00751; Figure 1B, black boxes) at the N-terminal region, but two groups of paralogs showed a conserved motif outside the DM domain at the C-terminal region termed DMRTA domain (Pfam accession: PF03474; Figure 1B, grey boxes). There are three forms of evidence that consider GB18426-PA and Dmdsx (AAF54169) as orthologs instead of GB15791-PA and Dmdsx. First, the amino acid sequences encoded by the orthologs GB18426-PA and Dmdsx (AAF54169) showed 51% identity, while the comparison between GB15791-PA and Dmdsx revealed 44.7% identity. Second, a second significant local alignment with 34% identity can be observed only between GB18426-PA and Dmdsx (not shown in Table 1 because only the best high score pair is shown). Third, only the DM domain was found in GB18426-PA and Dmdsx at the N-terminal region (e-value < 1e-10); two domains are present in GB15791-PA (DM and DMRTA), but the DMRTA domain is not present in the Dmdsx gene (Figure 1B). The most probable orthologs, GB18426-PA and Dmdsx, are located on chromosomes 5 and 3R, respectively. GB18260-PA and Dmdmrt93B (AAF55843) showed 44% identity and two domains (DM and DMRTA), and they are located on chromosomes 8 and 3R, respectively (Figure 1B). The most conserved orthologs, GB12040-PA and Dmdmrt11E (AAF48261), showed 78% identity and are located on chromosomes 5 and X, respectively (Figure 1B). Finally, the last clusters, GB15791-PA and Dmdmrt99B (AAF56919), are the least conserved orthologs with 39% identity. They share the DM and DMRTA domains and are located on chromosomes 1 and 3R, respectively. No synteny seems to exist between these DM genes in these insects.

Alignments with the DSX protein of insects

The orthologs of dsx genes in other insects were searched against the nr database at GenBank using the Dmdsx male isoform (AAF54169) as query. Nine insect dsx orthologs were aligned: B. oleae (CAD67987), A. obliqua (AAY25167), B. mori (BAB13472), D. melanogaster (AAF54169), M. scalaris (AAK38832), M. domestica (AAR23813), A. gambiae (AAX48939), C. capitata (AAN63597), and A. mellifera (GB18426-PA). Two highly conserved regions were evident in the alignments (Figure 2), and they have been experimentally described as functionally essential to the DSX proteins. The N-terminal conserved region includes the distinct class of an intertwined Zn-finger DNA-binding domain (DBD), that binds in the minor groove of DNA (Figure 2, DBD site I and site II) (Erdman and Burtis, 1993; Zhu et al., 2000) and a non-sex-specific oligomerization domain (Figure 2, OD1) that forms dimers that bind to regulatory sites of target genes. The C-terminal region corresponds to the oligomerization domain 2 (Figure 2, OD2).

Molecular cloning of A. mellifera dsx cDNA

A 443-bp cDNA fragment of Amdsx (GenBank accession: AY375535) was first isolated from female embryonic RNA using the combination of P1 and P2 primers designed on the two most highly conserved regions of the dsx gene, DBD/OD1 and OD2 (Figure 3A and D). This partial sequence of Amdsx represents the common region of the three first exons in both sexes. A male-specific cDNA fragment of 560 bp was obtained from male larval RNA using the combination of P1 and P3 primers (Figure 3A and D). P3 primer was designed based on an exon transcribed only in males (Figure 3A and D). The putative male-specific Amdsx sequence was aligned against the dsx family of the fruit fly, and the most similar sequence was Dmdsx (AAF54169). Furthermore, when comparing this partial sequence against the predicted honey bee official database set, the most similar sequence was GB18426-PA (Figure 3B) as we previously reported in the phylogenetic analysis (Figure 1A). The Amdsx is located in Group5.6 (chromosome 5, BCM-HGSP) and has at least three non-sex-specific exons. The first exon contains the DM motif (DBD/OD1 domain), and the third exon contains the non-sex-specific segment of the OD2 domain (Figure 3C). The fourth exon was predicted by computational methods and was experimentally inferred as male-specific. This study corroborates the sources of evidence of alternative splicing mechanisms in all insects studied so far. This sex-specific alternative splicing hypothesis is strengthened by the presence of an alanine (A) in the first position of the fourth exon at the male complement of the OD2 domain (Figure 2, black arrow) while glycine (G) is the most common residue at the first position in the female-specific exon (not shown in this paper). Beye et al. (2003) have mentioned that an ortholog dsx transcript is present in the honey bee and encodes a putative male- and female-specific protein by alternative splicing, but no experimental evidence had been shown to date.

DISCUSSION

We made a search of 29 nonredundant genes annotated in the GO database (Ashburner et al., 2000), which pointed out 13 highly conserved genes (at least 40% identity and e-value < 1e-18) involved in sex determination (GO:0007530) and sex differentiation (GO:0007548) in A. mellifera. Eight of 13 genes of the conserved genes encode a variety of transcription factors (dsx, ix, fru, dpn, dsf, run, bab, and scr, see Table 1). Three of 13 conserved genes, encoding proteins that participate in the initial signal (sxl, tra2 and fl) and are involved in the splicing mechanisms to direct the choice of sexual fate, were found to be conserved in the A. mellifera genome. Despite the conservation in structure, no evidence has been found concerning the functional conservation of these initial signal genes, even inside a single fly genus (Musca) (Meise et al., 1998). The last two conserved genes are related to cellular communication by means of signal transduction pathways (hop and pk61C). These signal transduction pathways are considered very important for the formation of genitalia and analia by means of dpp, hh and wg pathways (Estrada et al., 2003). All these regulators are important in sexual development in D. melanogaster, and some of them have been described as highly conserved in structure and they function even in distant species (Cline and Meyer, 1996; Marín and Baker, 1998; Graham et al., 2003).

The sex-determination pathways have evolved from the bottom up, as hypothesized by Wilkins (1995) and expanded by others (Marín and Baker, 1998; Schütt and Nöthiger, 2000). According to this hypothesis, the dsx and ix genes are highly conserved genes at the bottom of the sex determination hierarchy; we found that they are also conserved in the honey bee genome, at least in structure. Functional conservation for these two genes has been reported for metazoans (Raymond et al., 1998; Suzuki et al., 2003; Hediger et al., 2004; Siegal and Baker, 2005). The ix product can act together with DSX-F in a complex to achieve the same type of functionality as DSX-M alone, repressing or activating target genes in a female-specific manner (Siegal and Baker, 2005). DSX-F and IX form a complex that binds to the regulatory region of yp-1 and yp-2 to control the transcription of these female-specific genes (Coschigano and Wensink, 1993; Garrett-Engele et al., 2002).

In honey bees, the initial signal is not dependent on an X:A ratio or any sex chromosome-linked gene, but on the complementation of allelic products. In complementary sex determination, the single allelic (hemizygous or homozygous) proteins are non-functional and males develop by default, while the combination of different alleles (heterozygosity) results in an active protein, triggering the female sexual program (Whiting, 1943; Beye et al., 2003). The csd gene encodes an SR-protein that acts at the top of the sex-determination hierarchy to regulate the formation of sex-specific Amdsx transcripts in a way very similar to that of TRA in D. melanogaster and C. capitata (Beye et al., 2003; Beye, 2004).

The dsx gene encodes a transcription factor containing an intertwined CCHC and HCCC Zn-finger DNA-binding domain at the N-terminal region of the protein, termed DM domain (Erdman and Burtis, 1993; Raymond et al., 1998; Zhu et al., 2000). DM-related genes are involved in sexual development and in somite development in very distant metazoan phyla (Ottolenghi et al., 2002; Volff et al., 2003). In insects, dsx regulates somatic sexual differentiation. It encodes two functional products (DSX-F and DSX-M) produced by alternative splicing to yield a male or female pleiotropic factor acting on several independent target genes in a sex-specific manner (Burtis and Baker, 1989; Marín and Baker, 1998). The DSX proteins bind to a DNA palindromic sequence at the regulatory region of target genes as dimers. The DNA-binding affinities of female and male DSX are indistinguishable when considering only the DBD/OD1 domain, but the sex-specific OD2 domain makes crucial contributions to form dimeric structures necessary for the repression or activation of sex-specific target genes (Erdman et al., 1996; Cho and Wensink, 1998). The sex-specific sequence of OD2 domain is related to the sex specificity of DSX interaction with the transcriptional machinery or to DNA-binding cooperativity (An et al., 1996).

In our comparative approach, four homologs of dsx genes were found in the A. mellifera genome, and these are highly conserved when compared with the fruit fly (D. melanogaster). The clusters of paralogs indicated the best candidate for the putative Amdsx, GB18426-PA. Multiple alignment of dsx sequences of insects showed that two main conserved domains (DM or DBD/OD1 and OD2) are present in the predicted sequence of the honey bee dsx. A male-specific cDNA fragment (560 bp) and a non-sex-specific cDNA fragment (443 bp) were isolated from larval and embryonic transcripts, cloned and sequenced to confirm the presence of this dsx ortholog in the honey bee, but no functional analysis was conducted. Since mutant experiments in honey bees are rather impracticable, knockdown by double-strand RNA (dsRNA) or RNA interference (RNAi), and in situ hybridization can be a reasonable solution to test the functional role of the orthologous Amdsx (Amdam et al., 2003; Xavier-Neto and Behringer, 2005). Choosing a set of good phenotypic markers is a very important prerequisite, and the computational methods described in this study could be a simple and efficient solution to predict an initial set for good candidates for Amdsx target genes through an evolutionary approach.

Of particular interest to help understand the evolution of developmental pathways are the aspects of the regulatory network topology. Considering the high degree of conservation of the DNA-binding domain in the DSX transcription factor, conserved pleiotropic effects of this protein can act as a powerful force against evolutionary change, which implies some degree of conservation at the regulatory region of essential target genes responsible for sexual development. Currently, bioinformatics has turned out to be an efficient and powerful tool to make abstractions and to formalize theoretical concepts in biology, allowing more precise predictions based on mathematical evidence and not only on descriptive diagrams.

ACKNOWLEDGMENTS

We are very grateful to the Baylor College of Medicine Human Genome Sequencing Center for the free access to A. mellifera genomic sequences and the official set of predicted proteins. We also thank Klaus Hartfelder for critical reading of the manuscript. We thank MEC-CAPES and CNPq for financial support and the Open Source community, especially Python and Biopython projects, for their freely available programming tools.

REFERENCES

Amdam GV, Simoes ZL, Guidugli KR, Norberg K, et al. (2003). Disruption of vitellogenin gene function in adult honeybees by intra-abdominal injection of double-stranded RNA. BMC Biotechnol. 20: 1-3.

An W and Wensink PC (1995a). Integrating sex- and tissue-specific regulation within a single Drosophila enhancer. Genes Dev. 9: 256-266.

An W and Wensink PC (1995b). Three protein binding sites form an enhancer that regulates sex- and fat body-specific transcription of Drosophila yolk protein genes. EMBO J. 14: 1221-1230.

An W, Cho S, Ishii H and Wensink PC (1996). Sex-specific and non-sex-specific oligomerization domains in both of the doublesex transcription factors from Drosophila melanogaster. Mol. Cell. Biol. 16: 3106-3111.

Ashburner M, Ball CA, Blake JA, Botstein D, et al. (2000). Gene ontology: tool for the unification of biology. The gene ontology consortium. Nat. Genet. 25: 25-29.

Baker BS (1989). Sex in flies: the splice of life. Nature 340: 521-524.

Baker BS and Ridge KA (1980). Sex and the single cell. I. On the action of major loci affecting sex determination in Drosophila melanogaster. Genetics 94: 383-423.

Baker BS, Nagoshi RN and Burtis KC (1987). Molecular genetic aspects of sex determination in Drosophila. Bioessays 6: 66-70.

Beye M (2004). The dice of fate: the csd gene and how its allelic composition regulates sexual development in the honey bee, Apis mellifera. Bioessays 26: 1131-1139.

Beye M, Hasselmann M, Fondrk MK, Page RE, et al. (2003). The gene csd is the primary signal for sexual development in the honeybee and encodes an SR-type protein. Cell 114: 419-429.

Bull JJ (1983). Evolution of sex determining mechanisms. Benjamim Cummings, Menlo Park, CA, USA.

Burtis KC (1993). The regulation of sex determination and sexually dimorphic differentiation in Drosophila. Curr. Opin. Cell Biol. 5: 1006-1014.

Burtis KC and Baker BS (1989). Drosophila doublesex gene controls somatic sexual differentiation by producing alternatively spliced mRNAs encoding related sex-specific polypeptides. Cell 56: 997-1010.

Chervitz SA, Aravind L, Sherlock G, Ball CA, et al. (1998). Comparison of the complete protein sets of worm and yeast: orthology and divergence. Science 282: 2022-2028.

Cho S and Wensink PC (1998). Linkage between oligomerization and DNA binding in Drosophila doublesex proteins. Biochemistry 37: 11301-11308.

Cline TW and Meyer BJ (1996). Vive la difference: males vs females in flies vs worms. Annu. Rev. Genet. 30: 637-702.

Coschigano KT and Wensink PC (1993). Sex-specific transcriptional regulation by the male and female doublesex proteins of Drosophila. Genes Dev. 7: 42-54.

Dauwalder B, Tsujimoto S, Moss J and Mattox W (2002). The Drosophila takeout gene is regulated by the somatic sex-determination pathway and affects male courtship behavior. Genes Dev. 16: 2879-2892.

Dayhoff MO (1979). Atlas of protein sequence and structure. National Biomedical Research Foundation, Washington, DC, USA.

DeFalco T, Le BS and Van DM (2004). Abdominal-B is essential for proper sexually dimorphic development of the Drosophila gonad. Mech. Dev. 121: 1323-1333.

Erdman SE and Burtis KC (1993). The Drosophila doublesex proteins share a novel zinc finger related DNA binding domain. EMBO J. 12: 527-535.

Erdman SE, Chen HJ and Burtis KC (1996). Functional and genetic characterization of the oligomerization and DNA binding properties of the Drosophila doublesex proteins. Genetics 144: 1639-1652.

Estrada B, Casares F and Sanchez-Herrero E (2003). Development of the genitalia in Drosophila melanogaster. Differentiation 71: 299-310.

Felsenstein J (1989). PHYLIP: phylogeny inference package. Version 3.2. Cladistics 5: 164-166.

Garrett-Engele CM, Siegal ML, Manoli DS, Williams BC, et al. (2002). Intersex, a gene required for female sexual development in Drosophila, is expressed in both sexes and functions together with doublesex to regulate terminal differentiation. Development 129: 4661-4675.

Graham P, Penn JK and Schedl P (2003). Masters change, slaves remain. Bioessays 25: 1-4.

Hediger M, Burghardt G, Siegenthaler C, Buser N, et al. (2004). Sex determination in Drosophila melanogaster and Musca domestica converges at the level of the terminal regulator doublesex. Dev. Genes Evol. 214: 29-42.

Keisman EL and Baker BS (2001). The Drosophila sex determination hierarchy modulates wingless and decapentaplegic signaling to deploy dachshund sex-specifically in the genital imaginal disc. Development 128: 1643-1656.

Keisman EL, Christiansen AE and Baker BS (2001). The sex determination gene doublesex regulates the A/P organizer to direct sex-specific patterns of growth in the Drosophila genital imaginal disc. Dev. Cell 1: 215-225.

Kopp A, Duncan I, Godt D and Carroll SB (2000). Genetic control and evolution of sexually dimorphic characters in Drosophila. Nature 408: 553-559.

Kuhn S, Sievert V and Traut W (2000). The sex-determining gene doublesex in the fly Megaselia scalaris: conserved structure and sex-specific splicing. Genome 43: 1011-1020.

Kyriacou CP (2005). Behavioural genetics: sex in fruitflies is fruitless. Nature 436: 334-335.

Marin I and Baker BS (1998). The evolutionary dynamics of sex determination. Science 281: 1990-1994.

Meise M, Hilfiker-Kleiner D, Dubendorfer A, Brunner C, et al. (1998). Sex-lethal, the master sex-determining gene in Drosophila, is not sex-specifically regulated in Musca domestica. Development 125: 1487-1494.

Meng A, Moore B, Tang H, Yuan B, et al. (1999). A Drosophila doublesex-related gene, terra, is involved in somitogenesis in vertebrates. Development 126: 1259-1268.

Ohbayashi F, Suzuki MG, Mita K, Okano K, et al. (2000). A homologue of the Drosophila doublesex gene is transcribed into sex-specific mRNA isoforms in the silkworm, Bombyx mori. Comp. Biochem. Physiol. B 128: 145-158.

Ottolenghi C, Fellous M, Barbieri M and McElreavey K (2002). Novel paralogy relations among human chromosomes support a link between the phylogeny of doublesex-related genes and the evolution of sex determination. Genomics 79: 333-343.

Raymond CS, Shamu CE, Shen MM, Seifert KJ, et al. (1998). Evidence for evolutionary conservation of sex-determining genes. Nature 391: 691-695.

Raymond CS, Murphy MW, O’Sullivan MG, Bardwell VJ, et al. (2000). Dmrt1, a gene related to worm and fly sexual regulators, is required for mammalian testis differentiation. Genes Dev. 14: 2587-2595.

Saitou N and Nei M (1987). The neighbor-joining method: a new method for reconstructing phylogenetic trees. Mol. Biol. Evol. 4: 406-425.

Sanchez L and Guerrero I (2001). The development of the Drosophila genital disc. Bioessays 23: 698-707.

Sanchez L, Gorfinkiel N and Guerrero I (2001). Sex determination genes control the development of the Drosophila genital disc, modulating the response to Hedgehog, Wingless and Decapentaplegic signals. Development 128: 1033-1043.

Schütt C and Nöthiger R (2000). Structure, function and evolution of sex-determining systems in Dipteran insects. Development 127: 667-677.

Shearman DC and Frommer M (1998). The Bactrocera tryoni homologue of the Drosophila melanogaster sex-determination gene doublesex. Insect Mol. Biol. 7: 355-366.

Siegal ML and Baker BS (2005). Functional conservation and divergence of intersex, a gene required for female differentiation in Drosophila melanogaster. Dev. Genes Evol. 215: 1-12.

Sinclair AH, Berta P, Palmer MS, Hawkins JR, et al. (1990). A gene from the human sex-determining region encodes a protein with homology to a conserved DNA-binding motif. Nature 346: 240-244.

Suzuki MG, Funaguma S, Kanda T, Tamura T, et al. (2003). Analysis of the biological functions of a doublesex homologue in Bombyx mori. Dev. Genes Evol. 213: 345-354.

Thompson JD, Higgins DG and Gibson TJ (1994). CLUSTAL W: improving the sensitivity of progressive multiple sequence alignment through sequence weighting, position-specific gap penalties and weight matrix choice. Nucleic Acids Res. 22: 4673-4680.

Volff JN, Zarkower D, Bardwell VJ and Schartl M (2003). Evolutionary dynamics of the DM domain gene family in metazoans. J. Mol. Evol. 57: S241-S249.

White M (1973). Animal cytology and evolution. Cambridge University Press, Cambridge, UK.

Whiting PW (1943). Multiple alleles in complementary sex determination of habrobracon. Genetics 28: 365-382.

Wilkins AS (1995). Moving up the hierarchy: a hypothesis on the evolution of a genetic sex determination pathway. Bioessays 17: 71-77.

Xavier-Neto J and Behringer RR (2005). Developmental biology takes on a Latin American rhythm. Cell 122: 1-6.

Zhu L, Wilken J, Phillips NB, Narendra U, et al. (2000). Sexual dimorphism in diverse metazoans is regulated by a novel class of intertwined zinc fingers. Genes Dev. 14: 1750-1764.