RNA sequencing reveals sexually dimorphic gene expression before gonadal differentiation in chicken and allows comprehensive annotation of the W-chromosome
- Katie L Ayers†1, 2, 3,
- Nadia M Davidson†1,
- Diana Demiyah4,
- Kelly N Roeszler1,
- Frank Grützner5,
- Andrew H Sinclair1, 2, 6,
- Alicia Oshlack1Email author and
- Craig A Smith1, 2, 6Email author
© Ayers et al.; licensee Springer. 2013
Received: 23 November 2012
Accepted: 25 March 2013
Published: 25 March 2013
Birds have a ZZ male: ZW female sex chromosome system and while the Z-linked DMRT1 gene is necessary for testis development, the exact mechanism of sex determination in birds remains unsolved. This is partly due to the poor annotation of the W chromosome, which is speculated to carry a female determinant. Few genes have been mapped to the W and little is known of their expression.
We used RNA-seq to produce a comprehensive profile of gene expression in chicken blastoderms and embryonic gonads prior to sexual differentiation. We found robust sexually dimorphic gene expression in both tissues pre-dating gonadogenesis, including sex-linked and autosomal genes. This supports the hypothesis that sexual differentiation at the molecular level is at least partly cell autonomous in birds. Different sets of genes were sexually dimorphic in the two tissues, indicating that molecular sexual differentiation is tissue specific. Further analyses allowed the assembly of full-length transcripts for 26 W chromosome genes, providing a view of the W transcriptome in embryonic tissues. This is the first extensive analysis of W-linked genes and their expression profiles in early avian embryos.
Sexual differentiation at the molecular level is established in chicken early in embryogenesis, before gonadal sex differentiation. We find that the W chromosome is more transcriptionally active than previously thought, expand the number of known genes to 26 and present complete coding sequences for these W genes. This includes two novel W-linked sequences and three small RNAs reassigned to the W from the Un_Random chromosome.
KeywordsSex determination Embryonic chicken gonad W chromosome Avian sex, RNA-seq
Sex determination in mammals and birds occurs at fertilization, with the differential inheritance of sex chromosomes . Mammals have an XX/XY sex chromosome system, characterised by male heterogamety (XY), while birds have a ZZ/ZW system and female heterogamety (ZW). In eutherian mammals, almost all genes on the Y chromosome are associated with male development and fitness, namely the testis-determining SRY gene and genes required for spermatogenesis . The Z and W sex chromosomes of birds evolved independently of the eutherian X and Y, and hence birds lack the SRY gene . Like the mammalian X chromosome, the avian Z chromosome is large and gene-rich, while the W, like the mammalian Y, is typically small and largely heterochromatic. The 82 Mb chicken Z chromosome harbours over 1,000 genes and is highly conserved among avians [4, 5], while the W has far fewer genes and appears to have undergone independent degradation across the various avian groups [6, 7]. The exact roles of Z and W genes in avian sex determination remain to be fully resolved. Sex determination may involve a dominant-acting female determinant carried on the female-specific W chromosome, or a Z dosage-based mechanism could prevail . Consistent with the latter, the Z-linked DMRT1 and HEMOGEN genes are involved in testis development in the chicken embryo [9, 10]. However, a potential role for the W sex chromosome in avian sex determination cannot yet be excluded.
Attempts to address the potential role of the W chromosome in avian sex determination have been hampered by the poor understanding of this chromosome and its gene content. The chicken W chromosome has an estimated size of approximately 55 Mb, most of which (at least 70%) comprises repetitive elements of the XhoI, EcoR1 and Ssp1 classes [11, 12]. In addition, long arrays of interstitial telomeric sequences have been described on the W. The exact number of genes on the W is obscure. The chicken genome, from a female Red Jungle fowl, was sequenced in 2004  but assembly and annotation of the W has remained incomplete, and it is estimated that only 4% of the chicken W has been mapped . This amounts to approximately 1.2 Mb of assembled sequence on the W in the most recent release of the Chicken genome (Gallus_gallus-4.0; November 2011) . Until recently, around 12 bona fide genes had previously been verified as W-linked, including CHD1W, ATP5A1W, HINTW, UBAP2W, NIPBL, hnRNPW, KCMF, SMAD2, SPIN, MIER3, ZFR and ZNF532 [6, 12, 16–20] (chicken genome release 67, Ensembl) . In a study of W chromosome gene expression in different breeds of chickens, Moghadam and colleagues recently increased the number of putative W genes to 21, with several genes suggested to have multiple copies . However, expression data for these and other W genes at critical time points during embryonic sexual differentiation are lacking.
It has recently been proposed that avian sexual differentiation is at least partially cell autonomous. Cell autonomous sexual differentiation would involve purely intrinsic genetic factors, independent of any extrinsic signalling. In contrast, the long-held dogma is that vertebrate sexual differentiation is non-cell autonomous, that is, it does not rely purely on intrinsic factors in the cell and involves some degree of extrinsic signalling. (Sex hormones secreted from the gonads to induce female or male sexual dimorphisms, for example.) Zhao and colleagues (2010) used data from naturally occurring gynandomorphic chickens (birds that are half male and half female) and cross-sex embryonic tissue transplantations to show that sexual differentiation may be cell autonomous, at least partly independent of hormones . This study would predict molecular sexual differentiation (that is, sexually dimorphic gene expression) prior to morphological sexual differentiation (that is, development of testes or ovaries and other sex dimorphisms) (reviewed in ). However, large-scale data in support of this hypothesis in avians have not previously been reported in any detail.
This study uses RNA sequencing to examine the degree of cell autonomous sexual differentiation in birds and to characterise the W transcriptome in chicken embryos. Comprehensive analysis of gene expression in early embryos reveals significant sexually dimorphic gene expression well prior to gonadal sex differentiation, involving both the sex chromosomes and autosomes. These findings support the idea that avian sexual differentiation is at least partly cell autonomous at the molecular level . Different sets of genes showed sexual dimorphism in chicken blastoderms versus gonads, indicating that distinctive molecular pathways underlying sexual differentiation operate in different tissues. In addition, this analysis has allowed the full characterisation of gene expression from the enigmatic W (female) sex chromosome. Full characterisation of these W transcripts allows definitive comparisons with their Z homologues, shedding light on avian sex determination.
Expression profiling of blastoderms and embryonic day 4.5 gonads reveals at least partial cell autonomous molecular sexual differentiation in chicken
Deep transcriptome sequencing was used to profile gene expression at two developmental time points in males and females; 12-h blastoderms (Hamburger and Hamilton stage 1) and day 4.5 embryonic gonads (stage 26) . The rationale for using these times points was our focus on sex determination. Blastoderms represent the earliest accessible post-laying developmental stage, prior to primitive streak formation and gastrulation. This stage was chosen to specifically address the question of cell autonomous molecular sex differentiation pre-dating morphological differentiation. The second tissue, embryonic gonads at day 4.5, represents the time when the gonads are still morphologically identical in each sex ('bipotential').
Sequenced read-pairs were mapped to the chicken genome, (galGal3), using the TopHat 1.3.1 software . The overlap of read-pairs with Ensembl genes was then counted. Differential expression analysis was undertaken by testing the female counts against male counts at both time-points using edgeR , with a false discovery rate (FDR) <0.05. Genes known to be expressed sexually dimorphically in E4.5 gonads served as positive controls. For example, DMRT1 and AMH are known to be male upregulated by approximately two-fold in E4.5 gonads, and FOXL2 is expressed only in female gonads at E5.0. Meanwhile, both Aromatase and SOX9 are expressed after E4.5, and were expected to be non-dimorphic in our datasets . These patterns were confirmed in the RNA-seq (see Additional file 1, Figure S1), validating the sequencing results.
Some genes involved in sexual differentiation of the gonads begin to be dimorphically expressed between the sexes only in the gonads (for example, FOXL2). These genes therefore showed a different pattern of sexual dimorphism from blastoderm to gonad; that is, the female: male ratio was significantly different between tissues. To identify additional genes showing this difference in sexual dimorphism, we used a robust statistical test for the difference in female: male fold change between tissues (see Additional file 1 and 4). Of all of the 43 W-linked Ensembl sequences we detected, none of these genes showed a different pattern of sexual dimorphism (Figure 1C). For Z-genes, of which 262 were differentially expressed in at least one tissue, only three genes (1%) showed a significantly different pattern of sexual dimorphism between tissues (Figure 1C). This included two MHM locus genes that show a large increase in female expression in the gonads, and the Endothelial Tyrosine Kinase (TEK) gene that was more dimorphic in blastoderm (ENSGALG00000018479, ENSGALG00000023324, ENSGALG00000001840). However, 40 autosomal genes (25%) and 22 Un_random genes (45%) showed a different pattern of sexual dimorphism between tissues (Figure 1C). These data indicate that sex-specific molecular pathways that manifest in the blastoderms differ to those in gonads, primarily due to difference in autosomal gene expression. Given that sex chromosome genes did not generally deviate in their female: male ratio between the two tissues, it is interestingly to speculate how these genes can then activate different downstream genes in blastoderms and gonads. In the case of DMRT1, a known testis determinant that is differentially expressed with a similar ratio in both tissues, we found that the average expression level ((male+female)/2) of this gene dramatically increased from blastoderm to gonad. We therefore tested all genes for significant differential expression between tissues, regardless of sex (see Additional files 1 and 4). Indeed, a number of sexually dimorphically expressed W and Z genes also showed differential expression between the tissues, that is, W-genes - 23 (53%), Z-genes - 194 (74%) (Figure 1D and Additional file 4). Those genes upregulated in the gonads (Figure 1D and Additional file 4) are therefore interesting candidate gonadal sexual differentiation genes. Altogether, the data indicate that the relative expression levels of dimorphic sex-linked genes could explain their ability to regulate different downstream genes in different tissues.
To shed light on the molecular pathways that might underlie cell-autonomous sexual differentiation, we first screened the datasets for genes implicated in gonadal sex differentiation. A list was compiled of 117 genes previously linked tovertebrategonadogenesis (see Additional file 5). The set of genes differentially expressed between male and female blastoderms was significantly enriched with these gonadogenesis genes (P = 0.0098, Fisher's exact test), mostly due to non-compensated Z-linked genes (for example, 17βHSDB4 (ENSGAL00000002187), DMRT1 (ENSGAL00000010160) and CFC1B (ENSGAL00000012623)). Only one differentially expressed autosomal 'implicated gonadal gene' was found differentially expressed in this tissue (VNN1, ENSGALG00000013992) (Additional file 2). No genes previously proven to have a role in gonadal sex differentiation were found among the W-linked sequences. Taken together, these data indicate that sex chromosome genes do not activate known sexual differentiation pathways in blastoderms of either sex.
In contrast to the blastoderms, the gonads showed a different set of differentially expressed autosomal genes, many of which have a known link to gonadal sex differentiation, such as FOXL2, Anti-Müllerian Hormone (AMH), INHA (Inhibin-A) and HSP70 (ENSGALG00000011715). This indicates that the sex chromosomes have initiated developmental programs specifically associated with gonadal sex differentiation at embryonic day 4.5 (stage 25), 1 to 2 days prior to the onset of gonadal sexual differentiation (stages 29-30).
These findings show that different sexually dimorphic molecular pathways are engaged by the sex chromosomes in the two different tissues examined here. However, most of the differentially expressed genes detected in blastoderms and gonads had no previous connection to sexual differentiation. To characterise potential pathways activated in female versus male tissues, we assessed gene ontology of all genes that were exclusively differentially expressed in only blastoderms or gonads, using the DAVID programs . The top three clusters of GO terms for each group are shown in (Additional file 1, Figure S10). Given the low number of genes, very few GO terms showed significant enrichment when we corrected for multiple testing (Benjamini), however numerous genes involved in cell stress and DNA damage repair showed sex differential expression in the blastoderm. Various members of the hepatic fibrosis pathway were also differentially expressed (A2M and Col3A1 upregulated in females, and IL1R2 and EGF in males). Sex-specific gene expression in the liver has been previously described, involving >1,000 genes that affect a wide range of biological processes . For genes specifically differentially expressed in the gonads, the top GO terms included neuroactive ligand-receptor interaction (P value = 0.0081). These genes include the GABA receptor alpha 4 and two glutamate receptors. Among the list of autosomal genes expressed in the day 4.5 gonads in a sexually dimorphic fashion were several transcription factors not previously linked to sex per se (four upregulated in females and six in males) (Additional file 2). These data reveal the engagement of different molecular pathways at the two tissues assayed.
As mentioned previously, Z- and W-linked genes generally showed similar dimorphic expression in both blastoderms and gonads (Figure 1C), while many autosomal genes were only differentially expressed in one tissue (Figure 1C). In addition, a significant proportion of genes annotated to the Un_Random chromosome showed a stable dimorphic expression profile (Figure 1C). These unassigned sequences may in fact map to sex chromosomes, especially the poorly annotated W chromosome. Evidence for this conclusion was supported by their sex expression ratios, which were similar to those of the sex chromosome genes (Additional files 2 and 3). As the chicken W sex chromosome and its potential role in avian sex determination is currently poorly understood, we exploited the RNA-seq data to definitively annotate these sequences and to address W-linked gene expression.
Annotation of the W chromosome
Given the robust W-linked gene expression in both blastoderms to E4.5 gonads, we considered that this chromosome might play an important role in sex determination and cell autonomous molecular sex differentiation. However, the chicken W sex chromosome is currently incompletely annotated and its sequence is only partially assembled. In light of the challenges associated with sequencing the W chromosome, we investigated the W transcriptome and its potential role in avian sex determination in more detail. Two approaches were used to construct full-length open reading frames for the W transcriptome, genome-guided and genome independent (de novo) assembly.
Potential mis-annotated W genes were initially identified by their female specific expression (Ensembl) (see Methods and Materials). However, a significant fraction of reads also mapped outside annotated genes. Most notably, 62% of reads that mapped to the Un_random chromosome were not in annotated genes (Additional file 1, Figure S3). We hypothesised that some of these sequences were unannotated W-linked contigs. In order to identify unannotated genes, we extended our analysis of the RNA-seq data using a genome guided transcript assembly procedure, Cufflinks . The Cufflinks analysis was performed by mapping to the chicken genome (Galgal4), and a set of chicken transcripts was created by running Cufflinks 1.3.0 on the mapped reads (see Additional file 1). A significant proportion of expressed W genes identified through the Cufflinks analysis encoded retroviral elements with at least one open reading frame that showed homology to a retroviral protein (with an e-value <0.001, using BLAST). Together with pseudogenes, these sequences were filtered out of subsequent analysis (see Methods and Materials and Supplementary Methods in Additional file 1).
Annotation and expression of chicken W chromosome genes.
Expression (FPKM) Female
% ID with paralog
ATP synthase subunit alpha
Transcription factor BTF3-like
Uncharacterized protein C18ORF25
Chromodomain helicase DNA binding protein
Female Associated Factor
Golgi phosphoprotein 3-like
Histidine triad nucleotide binding protein
Heterogeneous nuclear ribonucleoprotein K
E3 ubiquitin-protein ligase KCMF1-like
Mesoderm induction early response 3
Neural precursor cell expressed developmentally downregulated 4-like
Nipped B/sister chromatid cohesion 2
Ras GTPase-activating protein 1-like
Ribosomal protein 17
Mothers against decapentaplegic homolog 2-like
TGF-beta signal pathway antagonist Smad7
Sia-alpha-2,3-Gal-beta-1,4-GlcNAc-R: alpha 2,8-sialyltransferase-like
Activated RNA polymerase II transcriptional coactivator p15-like
Ubiquitin associated protein 2
Ubiquitin-conjugating enzyme E2R 2
Valosin containing protein
Zinc finger RNA-binding protein
Zinc finger protein 532-like
Zinc finger SWIM domain-containing protein 6-like
W_random autosome ChrUn
MicroRNA 7b (in intron of HNRPK-W)
Small nucleolar RNA SNORD58 (in intron of RPL17-W)
Small nucleolar RNA SNORD58 (in intron of RPL17-W)
Small nucleolar RNA SNORD121A (in intron of UBAP2-W)
Small nucleolar RNA SNORD121A (in intron of UBAP2-W)
Using this approach, the open reading frames of W-linked mRNAs could be completely assembled, with the exception of one gene, NEDD4-like-W. This resulted in the characterisation of ten genes where previously only part of their ORF was known, due to the poor assembly of the W chromosome. Of these genes, we estimate that, on average, 60% of open reading frame sequence was previously missing from that available in either Ensembl or on Genbank (Additional file 6). Subsequent expression analysis of W genes was carried out by mapping reads to the newly assembled complete cDNA sequences. Final FPKM values (Fragments Per Kilobase of exon per Million fragments mapped) are presented in Table 1.
The complete W-linked transcripts identified in this RNA-seq screen showed gene ontologies with varying functions, none of which are patently associated with sexual differentiation (Table 1 and Table S3 - Additional file 6). However, the lists included genes specifying proteins associated with transcriptional regulation (ZSWIM6-W, ZNF532-W, MIER3-W, BTF3-W, SUB1-W), signalling (SMAD2-W, SMAD7-W, RASA1-W), the ubiquitination pathway (UBAP2-W, UBE2R2-W), the antioxidant thioredoxin (TXN-like-W), the ATP synthase, ATP5A1-W, and the aberrant nucleotide-binding protein, HINTW. As noted above, most W-linked genes did not show significant expression changes between the two tissues examined (Table 1). The most highly expressed genes across both tissues were HINT-W, RPL7-W, ATPA5A1-W, BTF3-W, VCP-like-W and hnRNKP.
Confirmation of female-restricted expression and W-linkage
Conservation and relative expression of W-linked genes and their Z gametologues
The complete W-linked transcripts assembled in this study were used to screen for homologues on the Z chromosome (gametologues). All W-linked genes with the exception of FAF (Female Associated Factor) were found to have gametologues on the Z. There was high sequence homology between almost all W- and Z-linked copies at both the DNA level (average of 88.4% identity, Table 1) and the protein level (average of 90.3% identity, Table 1). An exception was the gene HINTW, which showed 41% sequence and 48.5% amino acid homology with its Z gametologue (Table 1). Evolution of novel function among W-linked genes was examined by calculating the rates of synonymous and non-synonymous substitution for each Z-W gene pair  and is represented in Table 1, and as a sliding window across each gene pair in Figure S7 (Additional file 1). The dN/dS value for HINTW was the highest of all W genes (0.6) indicating that it has undergone the least amount of purifying selective pressure.
This study indicates that cells of the chicken embryo commence sexual differentiation at the molecular level early in development, through sexually dimorphic gene expression from both sex chromosomes and autosomes. This involves the enigmatic W sex chromosome, and it supports recent observations indicating that sexual differentiation may be at least partly cell autonomous in birds. This does not, however, exclude an important role for non-cell autonomous sexual differentiation, which involves hormonal signalling and is clearly very important in birds as it is in all animals. However, the data presented here support the notion that cell autonomous factors are also important, and there is very likely an interaction between both mechanisms to ensure complete and proper phenotypic sexual differentiation.
Previous studies have also reported sexually dimorphic gene expression in chicken embryos pre-dating gonadal development [32, 41, 42]. These studies utilised microarrays or subtractive hybridisation to uncover genes regulated in a sex-specific manner prior to gonadal differentiation in the chicken embryo. Our study utilised the earliest time point to date (12 h, stage 1 blastoderms), and has employed the most sensitive and comprehensive method available to assay gene expression. RNA-seq detects all expressed genes, including those that are unannotated, compared to microarray analysis, which only detects sequences arrayed on the microarray . The finding that sexually dimorphic gene expression can pre-date gonadal sex differentiation is not limited to birds, but is likely to be a feature of all animals. Sex differences in gene expression before gonadogenesis have been reported in mouse and bovine embryos [43, 44].
We found that W- and Z-linked genes showing sexually dimorphic expression in blastoderms largely maintained this differential expression profile in the gonads, pointing to a pervasive role for the sex chromosomes in the continuity of cell autonomous sexual development. The sexually dimorphic expression of autosomal genes indicates that female- and male-specific pathways are engaged as early as the blastoderm stage. However, a different set of autosomal genes showed sexually dimorphic expression in the undifferentiated gonads, most notably genes implicated in gonadal development. This indicates that the sex chromosomes activate different sex-specific pathways in different tissues.
The mechanism through which the Z and W sex chromosomes act to direct sexual differentiation pathways in early embryos is still unclear. One hypothesis is that the W chromosome carries a dominant-acting female determinant. This could confer 'femaleness' in a cell autonomous fashion early in development, but it may also initiate ovarian differentiation of the gonads at day 4.5. The RNA-seq data presented here has allowed us to draw some conclusions regarding the potential role of the W in avian sex determination. Contrary to previous assumptions, we found that the W is highly transcriptionally active in chicken embryos. Our novel analysis strategy allowed us to construct full-length transcripts for essentially the entire W transcriptome in two tissues. Using these we could correct and expand the annotation of the chicken W, providing complete transcript sequences and open reading frames for almost all 26 genes (Table 1 and Table S3 - Additional file 6). Most of these genes have previously been reported as W-linked in chicken [20, 22]. However, a thorough assessment of expression prior to or during gonadal development has been lacking, and limited sequence analysis has been carried out for these W-linked genes. In addition, we describe here two additional W linked genes and three W-linked small RNAs (Table 1). The assembly of full-length W transcripts described has allowed the accurate comparison of W and Z gametologues, providing key information on the degree of divergence between gene pairs and their relative expression levels. Most of the expressed W genes have open reading frames that are almost or completely homologous to their Z-linked gametologues (Additional file 1, Figure S7), which were also expressed in both blastoderms and gonads. This includes both the known and novel genes identified here. These W and Z genes are therefore likely to be acting as 'de facto autosomal' pairs, and hence almost all W-linked genes are unlikely female sex determinants. Indeed most of these genes encode proteins associated with general functions in the cell, such as ubiquitination, lipid and energy metabolism, and chromosome structure (Table 1). The two exceptions are HINTW, which is divergent from its Z gametologue, and FAF, which lacks a Z gametologue.
HINTW (Histidine Triad nucleotide binding, W-linked) is highly expressed in female chicken embryos at the blastoderm and gonadal stages (Table 1) . The high level of HINTW expression in females most likely reflects the multiple copies of HINTW that reside on the chicken W . While an endogenous HINTW protein has yet to be demonstrated, it is predicted to specify an aberrant nucleotide binding protein, lacking a key catalytic domain. The Z-linked copy, HINTZ, encodes a bona fide nucleotide binding protein, mediating the hydrolysis of nucleoside monophosphates  and it is hypothesised that HINTW may operate in a dominant negative fashion, blocking the function of HINTZ [8, 45]. While experimental over-expression of HINTW mRNA fails to alter male development in chicken embryos , the potential role of HINTW in avian sex determination requires closer examination. One W-linked transcript detected in our screen did not have a Z gametologue; FAF (Female Associated Factor) . The FAF gene product shows no clear homology to known proteins and its potential role in sexual differentiation is not clear at this stage.
The same set of W-linked genes expressed in female blastoderms was also expressed in E4.5 female gonads, yet a different set of autosomal genes was expressed. Therefore, if W-linked genes play a role in female development, they must engage different pathways in the different tissues. In the developing female gonad, a putative ovary determinant would directly or indirectly activate expression of the key FOXL2 gene, which is thought to then activate CYP19A1 and hence the oestrogen synthesis that is central to ovarian differentiation. Since we detected female enriched FOXL2 expression already in E4.5 gonads, the direct activator of this gene must also be present in our E4.5 female dataset (Additional file 5). This requires further analysis.
The alternative mechanism mediating avian sex determination involves Z-gene dosage. As there is no global Z chromosome dosage compensation system in birds, a sex determining mechanism may rely upon Z gene dosage inequality between the sexes. One or more non-dosage compensated Z-linked genes could direct sexual differentiation in individual cells. Our findings indicate that like the W, many Z-linked genes show sexually dimorphic expression that does not differ between the tissues assayed, with only 1% showing significantly different levels of dimorphic expression between the two tissues. This indicates that the sex chromosome influence sexual differentiation in cells, but this does not explain how sex-linked genes could then trigger sexually dimorphic expression of different autosomal genes in each tissue. Many Z- and W- linked genes showed different average expression levels between tissues (Figure 1D), and therefore while female: male ratios may not differ, changes in total expression levels may play a role in each tissue. For example, the Z-linked testis-associated genes, DMRT1 and HEMOGEN, are more highly expressed in the gonads compared to the blastoderms. Therefore it is possible that additional Z-linked genes may play a role in sexual differentiation in different tissues through changes in their relative expression. It is noteworthy in this regard that many Z-linked genes with sexually dimorphic expression in the gonads were sex-related (Additional file 5). Previous studies have reported a clear bias for genes associated with sex and reproduction located on the avian Z sex chromosome [47, 48], making sex determination via Z dosage and expression levels an attractive model.
It has recently been hypothesised that multiple parallel pathways leading to sexual differentiation probably exist in mammalian embryos . Similarly, in birds, molecular pathways controlling sexual differentiation in a cell autonomous fashion in the early embryo would be different to those operating specifically to induce ovarian versus testicular development at later stages. This idea is supported by our findings that different sets of autosomal genes show sexually dimorphic expression in the blastoderm compared to the E4.5 embryonic gonad. It is noted, however, that ultimately, proper sexual differentiation must involve the interaction of both cell autonomous and non-cell autonomous factors. Despite the advances brought by this research, the exact mechanism underlying avian sex determination is still unclear.
The data reported here show sexually dimorphic gene expression in chicken blastoderms and gonads prior to gonadal sex differentiation. This supports the notion that sexual differentiation begins at the molecular level in a cell autonomous manner in the chicken embryo. This involves both sex-linked and autosomal genes, with different sets of genes being expressed in the two tissues, demonstrating the engagement of different sexually dimorphic pathways. While sexually dimorphic gene expression from the Z and W sex chromosomes does not dramatically differ between developmental stages, we show that significant differences in average levels of expression of sex-linked genes may be responsible for activation of different signalling pathways in a tissue specific manner. Furthermore, we comprehensively define the W chromosome transcriptome for two early embryonic tissues, annotating W genes and providing complete open reading frames for most genes, and characterising two novel W-linked genes and several small RNAs.
Materials and methods
All experiments were carried out on a single line of Specific Pathogen Free (SPF) embryos (SPAFAS) from the SPF White Leghorn strain of chick (Lohman-LSL).
Blastoderms: Eggs were incubated for 12 h at 38°C and the blastoderms were dissected in cold DEPC-treated PBS. Hamburger and Hamilton (HH) stage 1 blastoderms were taken, and any showing developed primitive streaks were discarded. Only the area pellucida was dissected from the vitelline membrane and used, and any remaining yolk was removed. A small piece of blastoderm was taken for genetic sexing by PCR (see below), and the remainder of the blastoderm was stored at -80°C until RNA extraction was performed.
Gonads: SPF eggs were incubated until stage 26 (embryonic day 4.5). Paired gonads were removed and stored at -80°C. Handplate tissue was used for sexing. Embryos were sexed by PCR as described previously [50, 51].
RNA extraction and sequencing
Tissues were pooled according to sex. Twelve blastoderms or 16 paired gonads were pooled for each replicate (two male replicates and two female replicates were collected for each tissue). Total RNA was extracted using the RNeasy micro kit (QIAGEN) (which enriches for mRNAs, that is, RNAs >200 base pairs in length). On column DNAsing was performed to remove contaminating genomic DNA. The resulting RNA was poly A-selected, reverse transcribed, fragmented, bar-coded and sequenced using the Illumina HiSeq2000 at Australian Genome Research Facility (AGRF) in Melbourne. We sequenced 100 base pair, paired ends reads. Four lanes were used and all samples were bar-coded and run on each lane. The RNA was sequenced to a depth of approximately 20 million read-pairs per sample per lane, giving a total for each sample of 80 million read-pairs (160 million reads). These data are available on the NCBI Sequence Read Archive (accession number SRA055442).
Sequences were screened to remove poor quality bases, and subsequent analysis of the data followed the three methods shown in Additional file 1, Figure S8 and described in detail in the Supplementary Methods (Additional file 1). Briefly, we mapped the trimmed reads to the genome and counted reads which overlapped Ensembl genes, we performed genome-guided transcriptome discovery using Cufflinks  and we assembled the transcriptome independent of the chicken genome using ABySS . For the Ensembl differential expression analysis and W-linked gene identification we mapped the read-pairs to the chicken genome, galGal3, using the TopHat 1.3.1 software , with default settings. We counted reads which mapped to annotated genes giving a count for each gene in each sample and then used edgeR for differential expression (DE) analysis. edgeR uses a Generalized Linear Modeling (GLM) framework to model the gene counts as a negative binomial distribution. Variation estimates are stabilized by borrowing information between genes using an empirical Bayes approach  and hypothesis tests are performed using a likelihood ratio test. This method takes into account both biological and technical sources of variation even for small numbers of replicates. In this case we had 3 degrees of freedom to estimate biological variability . Further details are described in the Supplementary Methods (Additional file 1). Differential expression for the Cufflinks defined genes was performed in a similar manner to the Ensembl analysis, however we mapped to the updated chicken genome Galgal4 and the set of chicken transcripts was created by running Cufflinks 1.3.0 on the mapped reads (see Additional file 1 - Supplementary Methods).
De-novoW gene assembly
The de-novo transcriptome assembly was performed using Abyss version 1.3.2 on reads from all samples . To obtain the full sequence of genes split over non-contiguous genomic regions, we merged the Ensembl, Cufflinks and Abyss W transcripts using CAP3 . This was achieved using the novel approach of scaffolding the transcripts using the sequence of their Z gametologues, as detailed in the Supplementary Methods (Additional file 1). Our de-novo assembled W gene sequences and their Z gametologue sequences are provided as Additional files 7 and 8.
Preparation of chromosomes and fluorescent in-situhybridization (FISH) of BAC clones
Mitotic metaphase chromosomes and interphase preparations were generated from the established chicken embryonic fibroblast cell lines. Fluorescent in-situ hybridization was carried out as described in the Supplementary Methods (Additional file 1).
Quantitative reverse transcription-PCR (qRT-PCR)
Dissected stage 1-4 blastoderms were pooled according to sex, including three biological replicates, RNA was extracted and reverse transcribed as previously described . Quantitative real-time PCR was carried out using the Roche UPL Assay Design Center and relative expression was determined using the comparative CT method (ΔΔCT), with samples normalized against HPRT and expressed as fold change, as described in the Supplementary methods (Additional file 1).
false discovery rate
fluorescent in situ hybridization
fragments per kilobase of exon per million fragments mapped
generalised linear modelling framework
Hamburger and Hamilton
quantitative reverse transcription-PCR
next-generation sequencing of RNA
specific pathogen free.
We thank the Australian Genomic Research Institute (AGRF) in Melbourne. AO is supported by NHMRC Career Development Fellowship (1051481). KA is supported by the Poultry CRC for Sex Determination. CS is funded through an Australian Research Council (ARC) Future Fellowship, while KR is funded through an ARC Discovery Project Grant. AS is supported by an NHMRC Program grant (546517). FG is an ARC Research Fellow. DD is supported by a PhD scholarship from the Malaysian Government. This research was conducted within the Poultry CRC, established and supported under the Australian Government's Cooperative Research Centres Program. MCRI is supported by the Victorian Government's Operational Infrastructure Support Program.
- Ferguson-Smith M: The evolution of sex chromosomes and sex determination in vertebrates and the key role of DMRT1. Sex Dev. 2007, 1: 2-11. 10.1159/000096234.PubMedView ArticleGoogle Scholar
- Griffin DK: Is the Y chromosome disappearing?--both sides of the argument. Chromosome Res. 2012, 20: 35-45. 10.1007/s10577-011-9252-1.PubMedView ArticleGoogle Scholar
- McBride D, Sang H, Clinton M: Expression of Sry-related genes in the developing genital ridge/mesonephros of the chick embryo. J Reprod Fertil. 1997, 109: 59-63. 10.1530/jrf.0.1090059.PubMedView ArticleGoogle Scholar
- Nanda I, Benisch P, Fetting D, Haaf T, Schmid M: Synteny conservation of chicken macrochromosomes 1-10 in different avian lineages revealed by cross-species chromosome painting. Cytogenet Genome Res. 2008, 132: 165-181.View ArticleGoogle Scholar
- Bellott DW, Skaletsky H, Pyntikova T, Mardis ER, Graves T, Kremitzki C, Brown LG, Rozen S, Warren WC, Wilson RK, Page DC: Convergent evolution of chicken Z and human X chromosomes by expansion and gene acquisition. Nature. 2010, 466: 612-616. 10.1038/nature09172.PubMedPubMed CentralView ArticleGoogle Scholar
- Fridolfsson AK, Cheng H, Copeland NG, Jenkins NA, Liu HC, Raudsepp T, Woodage T, Chowdhary B, Halverson J, Ellegren H: Evolution of the avian sex chromosomes from an ancestral pair of autosomes. Proc Natl Acad Sci USA. 1998, 95: 8147-8152. 10.1073/pnas.95.14.8147.PubMedPubMed CentralView ArticleGoogle Scholar
- Ellegren H: Evolution of the avian sex chromosomes and their role in sex determination. Trends Ecol Evol. 2000, 15: 188-192. 10.1016/S0169-5347(00)01821-8.PubMedView ArticleGoogle Scholar
- Smith CA: Sex determination in birds: HINTs from the W sex chromosome?. Sex Dev. 2007, 1: 279-285. 10.1159/000108934.PubMedView ArticleGoogle Scholar
- Smith CA, Roeszler KN, Ohnesorg T, Cummins DM, Farlie PG, Doran TJ, Sinclair AH: The avian Z-linked gene DMRT1 is required for male sex determination in the chicken. Nature. 2009, 461: 267-271. 10.1038/nature08298.PubMedView ArticleGoogle Scholar
- Nakata T, Ishiguro M, Aduma N, Izumi H, Kuroiwa A: Chicken hemogen homolog is involved in the chicken-specific sex-determining mechanism. Proc Natl Acad Sci USA. 2013, doi: 10.1073/pnas.1218714110Google Scholar
- Saitoh Y, Mizuno S: Distribution of XhoI and EcoRI family repetitive DNA sequences into separate domains in the chicken W chromosome. Chromosoma. 1992, 101: 474-477. 10.1007/BF00352469.PubMedView ArticleGoogle Scholar
- Itoh Y, Mizuno S: Molecular and cytological characterization of SspI-family repetitive sequence on the chicken W chromosome. Chromosome Res. 2002, 10: 499-511. 10.1023/A:1020944414750.PubMedView ArticleGoogle Scholar
- Consortium ICG: Sequence and comparative analysis of the chicken genome provide unique perspectives on vertebrate evolution. Nature. 2004, 432: 695-716. 10.1038/nature03154.View ArticleGoogle Scholar
- Chen N, Bellott DW, Page DC, Clark AG: Identification of avian W-linked contigs by short-read sequencing. BMC Genomics. 2012, 13: 183-10.1186/1471-2164-13-183.PubMedPubMed CentralView ArticleGoogle Scholar
- Consortium ICG: Gallus gallus 4.0. 2011, 22/11/2011Google Scholar
- Ellegren H: First gene on the avian W chromosome (CHD) provides a tag for universal sexing of non-ratite birds. Proc Biol Sci. 1996, 263: 1635-1641. 10.1098/rspb.1996.0239.PubMedView ArticleGoogle Scholar
- Hori T, Asakawa S, Itoh Y, Shimizu N, Mizuno S: Wpkci, encoding an altered form of PKCI, is conserved widely on the avian W chromosome and expressed in early female embryos: implication of its role in female sex determination. Mol Biol Cell. 2000, 11: 3645-3660.PubMedPubMed CentralView ArticleGoogle Scholar
- Axelsson E, Smith NG, Sundstrom H, Berlin S, Ellegren H: Male-biased mutation rate and divergence in autosomal, z-linked and w-linked introns of chicken and Turkey. Mol Biol Evol. 2004, 21: 1538-1547. 10.1093/molbev/msh157.PubMedView ArticleGoogle Scholar
- Wahlberg P, Stromstedt L, Tordoir X, Foglio M, Heath S, Lechner D, Hellstrom AR, Tixier-Boichard M, Lathrop M, Gut IG, Andersson L: A high-resolution linkage map for the Z chromosome in chicken reveals hot spots for recombination. Cytogenet Genome Res. 2007, 117: 22-29. 10.1159/000103161.PubMedView ArticleGoogle Scholar
- Nam K, Ellegren H: The chicken (Gallus gallus) Z chromosome contains at least three nonlinear evolutionary strata. Genetics. 2008, 180: 1131-1136. 10.1534/genetics.108.090324.PubMedPubMed CentralView ArticleGoogle Scholar
- Ensembl: Chicken genome Release 67.
- Moghadam HK, Pointer MA, Wright AE, Berlin S, Mank JE: W chromosome expression responds to female-specific selection. Proc Natl Acad Sci USA. 2012, doi: 10.1073/pnas.1202721109Google Scholar
- Zhao D, McBride D, Nandi S, McQueen HA, McGrew MJ, Hocking PM, Lewis PD, Sang HM, Clinton M: Somatic sex identity is cell autonomous in the chicken. Nature. 2010, 464: 237-242. 10.1038/nature08852.PubMedPubMed CentralView ArticleGoogle Scholar
- Clinton M, Zhao D, Nandi S, McBride D: Evidence for avian cell autonomous sex identity (CASI) and implications for the sex-determination process?. Chromosome Res. 2012, 20: 177-190. 10.1007/s10577-011-9257-9.PubMedView ArticleGoogle Scholar
- Hamburger V, Hamilton HL: A series of normal stages in the development of the chick embryo. J Morph. 1951, 88: 49-92. 10.1002/jmor.1050880104.PubMedView ArticleGoogle Scholar
- Trapnell C, Pachter L, Salzberg SL: TopHat: discovering splice junctions with RNA-Seq. Bioinformatics. 2009, 25: 1105-1111. 10.1093/bioinformatics/btp120.PubMedPubMed CentralView ArticleGoogle Scholar
- Robinson MD, McCarthy DJ, Smyth GK: edgeR: a Bioconductor package for differential expression analysis of digital gene expression data. Bioinformatics. 2010, 26: 139-140. 10.1093/bioinformatics/btp616.PubMedPubMed CentralView ArticleGoogle Scholar
- Smith CA, McClive PJ, Hudson QJ, Sinclair AH: Male-specific cell migration into the developing gonad is conserved between mammals and birds and involves PDGF signalling. Developmental Biology. 2005, 284: 337-350. 10.1016/j.ydbio.2005.05.030.PubMedView ArticleGoogle Scholar
- Oshlack A, Robinson MD, Young MD: From RNA-seq reads to differential expression results. Genome Biol. 2010, 11: 220-10.1186/gb-2010-11-12-220.PubMedPubMed CentralView ArticleGoogle Scholar
- Itoh Y, Melamed E, Yang X, Kampf K, Wang S, Yehya N, Van Nas A, Replogle K, Band MR, Clayton DF, Schadt EE, Lusis AJ, Arnold AP: Dosage compensation is less effective in birds than in mammals. J Biol. 2007, 6: 2-10.1186/jbiol53.PubMedPubMed CentralView ArticleGoogle Scholar
- Arnold AP, Itoh Y, Melamed E: A bird's-eye view of sex chromosome dosage compensation. Annu Rev Genomics Hum Genet. 2008, 9: 109-127. 10.1146/annurev.genom.9.081307.164220.PubMedView ArticleGoogle Scholar
- Zhang SO, Mathur S, Hattem G, Tassy O, Pourquie O: Sex-dimorphic gene expression and ineffective dosage compensation of Z-linked genes in gastrulating chicken embryos. BMC Genomics. 2010, 11: 13-10.1186/1471-2164-11-13.PubMedPubMed CentralView ArticleGoogle Scholar
- Itoh Y, Kampf K, Arnold AP: Possible differences in the two Z chromosomes in male chickens and evolution of MHM sequences in Galliformes. Chromosoma. 2011, 120: 587-598. 10.1007/s00412-011-0333-x.PubMedView ArticleGoogle Scholar
- Teranishi M, Shimada Y, Hori T, Nakabayashi O, Kikuchi T, Macleod T, Pym R, Sheldon B, Solovei I, Macgregor H, Mizuno S: Transcripts of the MHM region on the chicken Z chromosome accumulate as non-coding RNA in the nucleus of female cells adjcent to the DMRT1 locus. Chromosome Res. 2001, 9: 147-165. 10.1023/A:1009235120741.PubMedView ArticleGoogle Scholar
- Dennis G, Sherman BT, Hosack DA, Yang J, Gao W, Lane HC, Lempicki RA: DAVID: Database for Annotation, Visualization, and Integrated Discovery. Genome Biol. 2003, 4: P3-10.1186/gb-2003-4-5-p3.PubMedView ArticleGoogle Scholar
- Waxman DJ, O'Connor C: Growth hormone regulation of sex-dependent liver gene expression. Mol Endocrinol. 2006, 20: 2613-2629. 10.1210/me.2006-0007.PubMedView ArticleGoogle Scholar
- Trapnell C, Williams BA, Pertea G, Mortazavi A, Kwan G, van Baren MJ, Salzberg SL, Wold BJ, Pachter L: Transcript assembly and quantification by RNA-Seq reveals unannotated transcripts and isoform switching during cell differentiation. Nat Biotechnol. 2010, 28: 511-515. 10.1038/nbt.1621.PubMedPubMed CentralView ArticleGoogle Scholar
- Simpson JT, Wong K, Jackman SD, Schein JE, Jones SJ, Birol I: ABySS: a parallel assembler for short read sequence data. Genome Res. 2009, 19: 1117-1123. 10.1101/gr.089532.108.PubMedPubMed CentralView ArticleGoogle Scholar
- Robertson G, Schein J, Chiu R, Corbett R, Field M, Jackman SD, Mungall K, Lee S, Okada HM, Qian JQ, Griffith M, Raymond A, Thiessen N, Cezard T, Butterfield YS, Newsome R, Chan SK, She R, Varhol R, Kamoh B, Prabhu AL, Tam A, Zhao Y, Moore RA, Hirst M, Marra MA, Jones SJ, Hoodless PA, Birol I: De novo assembly and analysis of RNA-seq data. Nat Methods. 2010, 7: 909-912. 10.1038/nmeth.1517.PubMedView ArticleGoogle Scholar
- Charif D, Thioulouse J, Lobry JR, Perriere G: Online synonymous codon usage analyses with the ade4 and seqinR packages. Bioinformatics. 2005, 21: 545-547. 10.1093/bioinformatics/bti037.PubMedView ArticleGoogle Scholar
- Yamada D, Koyama Y, Komatsubara M, Urabe M, Mori M, Hashimoto Y, Nii R, Kobayashi M, Nakamoto A, Ogihara J, Kato J, Mizuno S: Comprehensive search for chicken W chromosome-linked genes expressed in early female embryos from the female-minus-male subtracted cDNA macroarray. Chromosome Res. 2004, 12: 741-754.PubMedView ArticleGoogle Scholar
- Lin YP, Chen LR, Chen CF, Liou JF, Chen YL, Yang JR, Shiue YL: Identification of early transcripts related to male development in chicken embryos. Theriogenology. 2010, 74: 1161-1178. 10.1016/j.theriogenology.2010.05.017. e1161-1168PubMedView ArticleGoogle Scholar
- Nef S, Schaad O, Stallings NR, Cederroth CR, Pitetti JL, Schaer G, Malki S, Dubois-Dauphin M, Boizet-Bonhoure B, Descombes P, Parker KL, Vassalli JD: Gene expression during sex determination reveals a robust female genetic program at the onset of ovarian development. Dev Biol. 2005, 287: 361-377. 10.1016/j.ydbio.2005.09.008.PubMedView ArticleGoogle Scholar
- Bermejo-Alvarez P, Rizos D, Rath D, Lonergan P, Gutierrez-Adan A: Sex determines the expression level of one third of the actively expressed genes in bovine blastocysts. Proc Natl Acad Sci USA. 2010, 107: 3394-3399. 10.1073/pnas.0913843107.PubMedPubMed CentralView ArticleGoogle Scholar
- Moriyama S, Ogihara J, Kato J, Hori T, Mizuno S: PKCI-W forms a heterodimer with PKCI-Z and inhibits the biological activities of PKCI-Z in vitro, supporting the predicted role of PKCI-W in sex determination in birds. J Biochem. 2006, 139: 91-97. 10.1093/jb/mvj004.PubMedView ArticleGoogle Scholar
- Smith CA, Roeszler KN, Sinclair AH: Genetic evidence against a role for W-linked histidine triad nucleotide binding protein (HINTW) in avian sex determination. Int J Dev Biol. 2009, 53: 59-67. 10.1387/ijdb.082742cs.PubMedView ArticleGoogle Scholar
- Storchova R, Divina P: Nonrandom representation of sex-biased genes on chicken Z chromosome. J Mol Evol. 2006, 63: 676-681. 10.1007/s00239-006-0022-1.PubMedView ArticleGoogle Scholar
- Naurin S, Hasselquist D, Bensch S, Hansson B: Sex-biased gene expression on the avian z chromosome: highly expressed genes show higher male-biased expression. PLoS One. 2012, 7: e46854-10.1371/journal.pone.0046854.PubMedPubMed CentralView ArticleGoogle Scholar
- Arnold AP: The end of gonad-centric sex determination in mammals. Trends Genet. 2011, 28: 55-61.PubMedPubMed CentralView ArticleGoogle Scholar
- Kodama H, Saitoh H, Tone M, Kuhara S, Sakaki Y, Mizuno S: Nucleotide sequences and unusual electrophoretic behavior of the W chromosome-specific repeating DNA units of the domestic fowl, Gallus gallus domesticus. Chromosoma. 1987, 96: 18-25. 10.1007/BF00285878.PubMedView ArticleGoogle Scholar
- Clinton M, Haines L, Belloir B, McBride D: Sexing chick embryos: a rapid and simple protocol. Br Poult Sci. 2001, 42: 134-138. 10.1080/713655025.PubMedView ArticleGoogle Scholar
- Trapnell C, Roberts A, Goff L, Pertea G, Kim D, Kelley DR, Pimentel H, Salzberg SL, Rinn JL, Pachter L: Differential gene and transcript expression analysis of RNA-seq experiments with TopHat and Cufflinks. Nat Protoc. 2012, 7: 562-578.PubMedPubMed CentralView ArticleGoogle Scholar
- McCarthy DJ, Chen Y, Smyth GK: Differential expression analysis of multifactor RNA-Seq experiments with respect to biological variation. Nucleic Acids Res. 2012, 40: 4288-4297. 10.1093/nar/gks042.PubMedPubMed CentralView ArticleGoogle Scholar
- Huang X, Madan A: CAP3: A DNA sequence assembly program. Genome Res. 1999, 9: 868-877. 10.1101/gr.9.9.868.PubMedPubMed CentralView ArticleGoogle Scholar
- Smith CA, Shoemaker CM, Roeszler KN, Queen J, Crews D, Sinclair AH: Cloning and expression of R-Spondin1 in different vertebrates suggests a conserved role in ovarian development. BMC Dev Biol. 2008, 8: 72-10.1186/1471-213X-8-72.PubMedPubMed CentralView ArticleGoogle Scholar
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.