Sex-dimorphic gene expression and ineffective dosage compensation of Z-linked genes in gastrulating chicken embryos
© Zhang et al; licensee BioMed Central Ltd. 2010
Received: 20 July 2009
Accepted: 7 January 2010
Published: 7 January 2010
Considerable progress has been made in our understanding of sex determination and dosage compensation mechanisms in model organisms such as C. elegans, Drosophila and M. musculus. Strikingly, the mechanism involved in sex determination and dosage compensation are very different among these three model organisms. Birds present yet another situation where the heterogametic sex is the female. Sex determination is still poorly understood in birds and few key determinants have so far been identified. In contrast to most other species, dosage compensation of bird sex chromosomal genes appears rather ineffective.
By comparing microarrays from microdissected primitive streak from single chicken embryos, we identified a large number of genes differentially expressed between male and female embryos at a very early stage (Hamburger and Hamilton stage 4), long before any sexual differentiation occurs. Most of these genes are located on the Z chromosome, which indicates that dosage compensation is ineffective in early chicken embryos. Gene ontology analyses, using an enhanced annotation tool for Affymetrix probesets of the chicken genome developed in our laboratory (called Manteia), show that among these male-biased genes found on the Z chromosome, more than 20 genes play a role in sex differentiation.
These results corroborate previous studies demonstrating the rather inefficient dosage compensation for Z chromosome in birds and show that this sexual dimorphism in gene regulation is observed long before the onset of sexual differentiation. These data also suggest a potential role of non-compensated Z-linked genes in somatic sex differentiation in birds.
Many metazoan species have dimorphic sex chromosomes. The imbalanced or differential expression of sex determination genes on the sex chromosomes of a given species controls the genetic cascade that eventually leads to dimorphic development of reproductive organs and secondary sexual characteristics. Different mechanisms of sex determination have been uncovered in genetically tractable systems such as C. elegans, Drosophila and the mouse . The majority of genes located on the sex chromosomes, however, are not involved in sex determination. Their imbalanced expression in the two sexes can have deleterious consequences in species like mammals . Thus, various species have evolved different mechanisms to equalize the expression levels (dosage compensation) of these genes.
In mammals, the parity of female and male expression of X genes is achieved through inactivation of one whole X chromosome in the female and upregulation of the single X gene copy in both females and males to equalize the expression level with the autosomes . In the fly, both copies of X chromosomes are active in the female . An increase in the expression of the single X gene copy in the male brings the expression level closer to that of the two female copies and close to that of autosomes. In C. elegans, transcription from the two active X gene copies in hermaphrodites is first decreased by one-fold to equal that of a single X gene copy in the male . Then, transcription from the two decreased X gene copies in hermaphrodites and the single copy in the male are further increased to equal that of autosomal copies .
The avian species represents an interesting but poorly understood system in which the homogametic karyotype (ZZ) corresponds to the male, while the heterogametic karyotype (ZW) corresponds to the female [4, 5]. This is in contrast to mammals or Drosophila, in which the homogametic karyotype (XX) represents the female and the heterogametic (XY) represents the male. It is currently unclear whether the male sex is determined by the presence of the two Z chromosomes, or whether the female sex is defined because of the presence of the W chromosome [4, 5]. Recently, microarray-based, genome-wide, gene-profiling studies of male and female adult zebra finch and chicken embryos have demonstrated that a majority of genes located on the Z chromosome are not dosage-compensated, and thus are expressed at a much higher level in males than in females [6–12]. The lack of compensation of Z genes could lead to an increase in expression of male determinants in ZZ individuals, ultimately leading to male sex determination. Consistent with this idea, DMRT1, a Z-linked gene was recently shown to be required for male gonadal sex determination . Alternatively, it has been proposed that the PKCIW gene (also known as HINTW), which is located on the W chromosome, is involved in female sex determination [14, 15] and inhibits the activity of its homolog on the Z chromosome, PKCIZ. This hypothesis is closer to what is observed in mammalian sex determination in which the male sex is defined by specific male determinants, such as the sex determining region Y (Sry) gene located on the Y chromosome . The functional significance of PKCIW in female sex determination remains, however, unknown.
Here, we use chicken whole-genome Affymetrix microarrays to compare gene expression profiles of the primitive streak of individual male and female gastrulating chicken embryos at day 1 of incubation. This is a critical stage at which axis formation and neural, ectodermal, mesodermal and endodermal patterning begin. Dosage compensation is observed already at this stage for a small portion of Z-linked genes. Compared to dosage compensation of Z-linked genes as reported during the late stages , dosage compensation of Z-linked genes at this stage shares similarities, such as an enrichment at the male hypermethylated (MHM) region, but differs in terms of which genes are compensated. It indicates a temporal regulation of dosage compensation in the chicken. We also report that a large number of genes, which reside mainly on the Z and W chromosomes, are dimorphically expressed between male and female embryos. Surprisingly, functional annotation of the male-biased Z genes revealed significant enrichment in the genes involved in male sex determination/differentiation. We conclude that the lack of Z chromosome-wide dosage compensation may be part of the mechanism of sex differentiation at early stages when there is no obvious morphological difference between sexes.
Results and Discussion
Sex-biased gene expression in Hamburger and Hamilton Stage 4 chicken embryos
When computing the F:M ratios of mean expression levels of the 14548 probesets, 35 were detected as being expressed significantly higher in females with a FDR of 5% (Additional file 3), while 27 out of the 35 probesets differentially expressed in females exhibit a F:M ratio greater than 1.5. These 27 probesets are regarded as dimorphically female-biased (Additional file 4). Three probesets are on chr1, four on chrZ, nine on chrUn_random, three are of unknown location and eight are on chrW (Figure 1b). Since the W chromosome is only partially sequenced and annotated at present, it is likely that at least some of the nine probesets (aligned on chrUn_random) with F:M ratios higher than four are also on the W chromosome. Among these genes, interesting candidates include the transcription factor BTF3, which shows a F:M ratio of 130 and was identified as a transcriptional co-activator for the estrogen receptor . The STAT-related gene NP_001012932.1, which shows a F:M ratio of 7, is also of interest since STAT5 was implicated in sexual dimorphic growth downstream of growth hormones in mammals .
An obvious difference between the male- and female-biased/dimorphic probesets is the value of the M:F ratio (mean 1.75) and F:M ratio (mean 95.2). While the male-biased probesets are detected in males at a level ~one-fold higher than in females, the female-biased probesets exhibit a much higher increase. This difference may be explained as the result of a majority of male-biased genes being located on the Z chromosome, which has two gene copies in males and only one copy in females. Among the female-biased genes, those on chrUn_random may be not-yet-annotated genes on W chromosome. If so, the bulk of the female-biased genes would be on W, which has one copy in females but is absent in males. However, the expression of chr1 genes with F:M ratio greater than 2 cannot be attributed to the difference in gene copy number.
Absence of chromosome-wide compensation of Z-linked genes
To determine if this M:F ratio of Z genes can be explained by a global increased expression of Z chromosome genes compared to autosomes in males or by a decreased expression on the single Z chromosome in females, the mean expression levels of autosomal genes in comparison with Z genes in both females and males were plotted. In males, the expression level of Z genes is similar to that of autosomes (P-value 0.226, Wilcoxon Rank Sum Test). However, in females, there is a significant decrease in the expression of Z genes compared to autosomal genes (P-value 1.107e-07, Wilcoxon Rank Sum Test) (Figure 2b). This is in sharp contrast to all three different situations in mammals, flies and worms, and is consistent with the results reported in older embryos [6, 7, 9], indicating that the high M:F ratio of Z genes is achieved through the relatively low expression of genes from the single copy on Z in females. In other words, dosage compensation between females and males, and between Z chromosome and autosomes, is either nonexistent or ineffective at the chromosome-wide level.
Regional differences in the distribution of compensated gene expression along the Z chromosome
In chicken embryos, dosage compensation with a M:F ratio of expression of 1:1 occurs in a small portion of Z-linked genes that are mostly associated with the male hypermethylated MHM [8, 21]. To examine the distribution trend of M:F ratios across the Z chromosome position, the running averages of ratios were calculated as previously described . In agreement with these two studies, a major valley with compensated genes enriched at MHM region was observed (Figure 3c). This region corresponds to a short genomic region located on the Z chromosome, which is hypermethylated in males and, which in females, produces a large non-coding RNA that remains associated with the MHM region. The W chromosome is required for the transcription of the female-specific MHM transcript, supporting the possibility that in the chicken, sex determination requires the W chromosome . The MHM non-coding RNA was shown to be expressed in HH5 to HH6 female chicken embryos . However, there is no probeset in the Affymetrix GeneChip that detects MHM non-coding RNA. Thus, we do not know whether MHM non-coding RNA is already expressed in HH4 chicken embryos.
We compared the results of compensated and non-compensated gene lists to those of Itoh  and Melamed . Valley and peak genes from the Melamed et al. study were compared with compensated and non-compensated genes in this study, respectively. Thirteen out of 61 Melamed valley genes were shared with the compensated genes and six with non-compensated genes, and 32 of 121 Melamed et al. peak genes were shared by non-compensated genes and eight with compensated genes from our study. Many of the peak and valley genes of the Melamed study were not expressed in our dataset (data not shown). Genes with a M:F ratio > 1.5 were extracted from the brain study of Itoh et al and compared to the male-biased probesets (see additional data file 2 of Itoh et al, 2007). Of the 277 male-biased probesets from Itoh et al, 157 (56%) were represented in the male-biased set in our study (data not shown). Interestingly, the major Zq peak centering on position 6.0 × 107 bp observed in E14 embryonic brain, heart and liver  and E10, E15, E19 adult brain and gonad tissues  is not observed in our study (Figure 3c). Only 40 of the 121 genes of the peak identified as non-compensated  were identified in this analysis. In contrast, a major peak is observed at a different position around 4.6 × 107 bp on the chromosome arm Zq in our study (Figure 3c). Furthermore, genes distally located to the MHM on Zp are expressed at relatively higher M:F ratios, with multiple peaks in contrast to the levels reported in the brain, heart and liver tissues of E14 embryos (Figure 3c)  and E10, E15, E19 adult brain and gonad tissues .
Using brain and gonad tissues collected from E10 to adult stages, Mank & Ellegren  computed the "amplitude" of gene expression on Z chromosome. They reached the conclusion that Z chromosome dosage compensation does not have enrichment at the MHM region and is not subject to temporal regulation within the same tissue type. This different conclusion was a result of using "amplitude" instead of "M:F ratio" and a different view on whether genes with low (female-biased) M:F ratios should be considered as compensated [22, 23]. We computed the same "amplitude" profile of Z-linked genes in addition to the "M:F ratio" profile. The result showed that the valley at MHM region still existed (Additional file 8). This argues that at least in gastrulating chicken embryos, dosage compensation is enriched at the MHM region. These results support the existence of temporal differences in the regulation of the compensated and non-compensated genes during embryonic development. Similarly, dosage compensation in the mouse is also a temporal process. In undifferentiated female and male mouse embryonic stem cells and embryonic blastocysts, the F:M ratio of X-linked genes is 1.61. In differentiated cells, the F:M ratio drops to 1.10 . Although the major peak at 6.0 × 107 bp is observed regardless of tissues types in later stage embryos , comparing undifferentiated tissues at gastrulation stage with differentiated somatic and gonadal tissues at E10 and later stages may not be directly relevant. Thus, we cannot rule out that the difference in dosage compensation patterns between gastrulation stage embryos and much later stage embryos is also tissue specific.
To further investigate the behavior of compensated and non-compensated genes, the expression levels of compensated and non-compensated probesets along with probesets on the entire Z chromosome in females and males were plotted. The mean expression levels of compensated and non-compensated genes on the Z chromosome in females were similar. The mean expression level of compensated genes in the male is similar to that of the female. However, the mean expression level of non-compensated genes in males is twice to that in females (Figure 3d). These findings support the conclusion that within the Z chromosome, dosage compensation is achieved through relative downregulation of the expression of the two copies of Z genes in the male, but not through upregulation of a single copy of Z genes in the female.
Dimorphic expression of genes implicated in sex determination and physiology during early embryonic development
The determinants that control sex determination in birds are unknown . In mammals, the Sry gene on the Y chromosome initiates a gene cascade that induces differentiation of the gonads into testes . There is currently no evidence of the existence of an SRY ortholog in the chicken. In our male- and female-biased gene lists, we do not find homologs of other genes involved in the mammalian gonad differentiation pathway, such as: STF1, SOX3, SOX9 or WNT4 . In the chicken, these four genes are not located on the Z chromosome. In humans, loss of a copy of the sex determinant DMRT1 can lead to sex reversal in which a genetically male individual adopts a female phenotype . DMRT1 is Z-linked and sexually dimorphic in the chicken gonad [26, 27]. DMRT1 is required for male gonadal sex determination . This gene has homologs in mammals, flies and worms, which are also involved in sex determination . However, in our gene lists, DMRT1 is not significantly male-biased: Gga.260.2.S1_a_at, M:F ratio 1.3, q-value 54%.
The lack of Z chromosome-wide compensation in gastrulating chicken embryos could be explained because male-biased genes were selected to escape compensation in order to serve an active role in male sex determination. In a series of previous studies , no sex-biased gene was reported to be implicated in sexual determination or differentiation . To re-examine this possibility, a gene ontology (GO) analysis test was conducted on the differentially regulated genes between males and females. The quality of the chicken genome GO annotation is extremely poor compared to mouse and human. To circumvent this, we developed an annotation database called Manteia (see Methods section; manuscript in preparation), which links the chicken Affymetrix probesets to the corresponding NCBI or ENSEMBL gene models and merges the available information for both. In addition, Manteia offers the possibility to enhance this annotation by inferring to the chicken gene models the annotation (such as GO terms) of their corresponding human and mouse orthologs. The database provides convenient statistical tools to evaluate the enrichment of GO (or phenotype, see below) categories compared to the genome or to the transcriptome represented on the microarray using an approach originally developed for GoMiner .
GO analysis of male-biased genes.
steroid biosynthetic process
sterol biosynthetic process
Androgen Receptor Signaling Pathway
Male-biased genes implicated in steroid metabolism, steroid receptor activity and spermatid and testis development.
hydroxysteroid (17-beta) dehydrogenase 4 (HSD17B4)
zinc finger, CCHC domain containing 7 (ZCCHC7)
3-hydroxy-3-methylglutaryl-Coenzyme A reductase (HMGCR)
zinc finger, CCHC domain containing 9 (ZCCHC9)
C-8 sterol isomerase activity, Hypothetical protein, clone 7e2Gg-AFFY-23616
acetyl-Coenzyme A acyltransferase 2 (ACAA2)
protein kinase, AMP-activated, alpha 1 catalytic subunit (PRKAA1)
StAR-related lipid transfer protein 4 (StARD4)
Phosphatidate phosphohydrolase type 2a (PAP2a)
DnaJ (Hsp40) homolog, subfamily A, member 1RCJMB04
Bardet-Biedl syndrome 4 protein (BBS4)
Fanconi anemia, complementation group G (FANCG)
Fanconi anemia, complementation group C (FANCC)
nuclear receptor subfamily 2, group C, member 1 (NR2C1)
Hexosaminidase B (HEXB)
Phospholipase A-2-activating protein (PLAAP)
pelota homolog (Drosophila) (PELOTA)
prostaglandin receptor EP4 subtypeL (PTGER4)
alpha-methylacyl-CoA racemase (AMACR)
In Drosophila, PELOTA mutation causes a male-specific failure of spermatocyte meiosis . In the mouse, Spindlin (SPIN1 homolog) encodes a protein associated with the oocyte meiotic spindle , and is highly similar to Ssty, a gene on the Y chromosome that is exclusively expressed in spermatids, and which is important for their development . Female chickens also have a SPINW homolog (Gga.1845.1.S1_at) on the W chromosome, but it is not expressed at this stage (data not shown), suggesting that the identified Z-linked SPINZ homolog could, in fact, be related to Ssty. Thus, male-biased expression of Z-linked PELOTA and SPINZ may play a role in spermatocyte development in the chicken.
Strikingly, enzymes involved in steroid synthesis and genes mediating testosterone activity are enriched in the noncompensated, male-specific Z-linked genes. Testosterone, one of the major steroids, is synthesized from cholesterol. We identified several genes involved in cholesterol synthesis - such as ACAA2 (acetyl-Coenzyme A acyltransferase 2) and HMGCR (3-hydroxy-3-methylglutaryl-Coenzyme A reductase), the rate-limiting enzyme in cholesterol synthesis - that are enriched in male-biased genes. StARD4, which is implicated in transporting cholesterol for steroid synthesis; the C-8 sterol isomerase, the prostaglandin receptor EP4, subtype L (PTGER4) and the hydroxysteroid (17-β) dehydrogenase (HSD17B4) are also found in the Z-linked dimorphic genes.
The androgen receptor mediates testosterone or dihydrotestosterone activity and is dimorphically expressed in the adult mouse brain . Nuclear receptor subfamily 2, group C, member 1 (NR2C1), also called testicular receptor 2 and which is normally expressed in testis and prostate in mammals, is acting as a co-receptor of an androgen receptor . It is also listed as a non-compensated Z-linked gene. Additional genes that are involved in the control of reproductive functions (such as, hexosaminidase B [HEXB]), which is involved in the breakdown of gangliosides), are also found in the non-compensated Z-linked genes. In humans, mutations of HEXB cause Sandhoff disease, a progressive neurodegenerative disorder characterized by an accumulation of GM2 gangliosides in neurons. Additionally, HEXB has been shown to have much higher activity in male infants. This bias is correlated with higher testosterone levels in males . This supports the notion that chicken orthologs of these genes are involved in sexual differentiation.
Using the MGI (Mouse Genome Informatics) phenotype descriptions of mouse knockouts in Manteia, the enrichment in specific phenotypes in the mouse orthologs of the non-compensated chicken genes of the Z-chromosome was calculated. Significant enrichment was found in categories, such as seminiferous tubule degeneration (three genes, p = 0.002), abnormal seminiferous tubule morphology (six genes, p = 0.00091), abnormal testis morphology (eight genes, p = 0.00088), abnormal sex gland morphology (eight genes, p = 0.0049), abnormal male reproductive anatomy (eight genes, p = 0.002), small testis (five genes, p = 0.01), abnormal fertility/fecundity (12 genes, p = 0.002), reproductive system phenotype (13 genes, p = 0.07), and abnormal male reproductive physiology (eight genes, p = 0.004). The genes associated with these phenotypes largely overlap the ones identified by GO analysis (Additional file 9).
GO analysis of the 114 compensated Z genes revealed that the major categories enriched include: intracellular signaling cascade (22 genes) or regulation of transcription (21 genes), but do not include significant processes related to sex determination or differentiation (data not shown) [8–10]. Strikingly, the FEM1C gene, which acts downstream of the DMRT1 homolog Mab3 in the C. elegans sex determination cascade, belongs to this list of compensated genes . Using the MGI data, enrichment in specific phenotypes for the mouse orthologs of these chicken genes was also analyzed, revealing that 32 of these genes are associated with lethality-embryonic perinatal phenotypes (p = 0.0003791394). The most significant phenotypes observed were related to the nervous system and muscle development, but no enrichment for phenotypes related to sexual differentiation was observed (data not shown).
Early dimorphic expression in the developing embryo of genes implicated in steroid hormone activity and reproduction
In this report, we used a microarray approach to analyze genes exhibiting sexual dimorphism in the primitive streak of gastrulating male and female chicken embryos, several days before the appearance of gonadal primordia and sexual differentiation. Thus far, the earliest embryos analyzed by this approach were harvested at day 4 [6, 7, 9], around the beginning of sexual differentiation. Strikingly, we observed that even before the segregation of the three embryonic germ layers, a large number of genes present on the Z chromosome are expressed at a much higher level in males than in females. Since males exhibit two copies of the Z chromosome and females only one, these results are consistent with inefficient dosage compensation of the Z chromosome genes in the chicken [7, 8].
During late embryonic stages in mammals, the developing testes secrete two major hormones: anti-Mullerian and testosterone. The anti-Mullerian hormone inhibits the female development of the reproductive tract and external genitalia, while testosterone promotes masculine development . Testis-derived testosterone not only induces male configuration of the reproductive tract and external genitalia, but also circulates to induce permanent changes in discrete nuclei of hypothalamus and other regions of the brain, leading to structural differences and distinct mating behaviors . It is generally believed that testis is the main, if not the only, source of testosterone that acts on the embryonic and adult male brain. This concept is supported by the fact that the brain of castrated male mice is transformed into a female-like brain , which highlights the importance of testis and testosterone in the differentiation of dimorphic structures of the brain. In birds, steroid hormones also play important roles in somatic sex dimorphism, especially in the dimorphism of brain regions controlling singing behaviors of song birds, such as zebra finches . However, gonads seem to have little or no role in sex dimorphism in the adult zebra finch brain, since castrated males can still sing male-specific courtship songs . Existence of gynandromorphs in birds clearly supports the notion of sexual dimorphism being largely independent from circulating hormone levels. In some gynandromorph birds, each side of the animal exhibits secondary sexual characteristics, such as feather coloration and patterns that are specific to each sex . In a reported case, one half largely developed like a female, despite the presence of a testis .
Our data demonstrate enrichment for genes involved in male sexual differentiation among the non-compensated Z-linked genes. Dimorphic expression of these genes at the gastrulation stage in chicken embryos is striking and can hardly be related to gonadal development, suggesting that, rather, they act on somatic tissues. Interestingly, male zebra finch brain tissue has been shown to be capable of synthesizing estrogen locally. This autonomous synthesis of estrogen is necessary and sufficient for some aspects of male-specific neural development in vitro [41, 42]. Given the profound effect of testosterone on differentiation of both gonadal and somatic sex, it is tempting to suggest that the chicken may have evolved a mechanism different than mammals to synthesize high levels of testosterone for somatic sex differentiation in males independent of testis formation. Male-biased expression of Z-linked sex differentiation gene expression is likely a result of ineffective dosage compensation.
Sample preparation and microarrays
Fertilized White Leghorn chicken eggs were purchased from commercial sources. Embryos were incubated at 38°C for 23 h to reach HH4-4+ before harvesting. A 400 micron × 400 micron square tissue consisting of the three germ layers was excised at the 70% to 90% primitive streak level of an individual embryo. In this way, the primitive streak proper and epiblast lateral to it were sampled. Thus, this tissue encompasses the prospective somitic mesoderm, neuroectoderm and endoderm. Total RNA was extracted, amplified two cycles and biotin-labeled before being hybridized to Affymetrix Chicken GeneChips. Detailed procedures are essentially the same as previously described . In total, 18 samples were prepared from 18 individual embryos randomly chosen with regard to sex. Sexes were identified by virtue of the presence or absence of high expression of PKCIW genes in 9 of the 18 samples. To monitor the quality of RNA extraction and amplifications during the procedures, some embryos were electoporated with plasmids expressing various constructs, such as the fluorescent reporter Zsgreen. Since no statistical difference was observed in the sexually dimorphic genes between control and electroporated embryos of the same sex, all the samples were subsequently pooled in the analysis. Original microarray data was deposited into NCBI-GEO with access using the following link: http://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?token=xxoffuogsawoavo&acc=GSE16064
The original dataset consisted of 38535 probesets. Since the chicken genome is not very well annotated in terms of the genes corresponding to probesets, and hence to avoid losing information, we decided to treat each probeset as a potential gene. The dataset was normalized by Affymetrix Microarray Suite 5 (MAS5) preprocessing method. To remove the unexpressed genome, the dataset was filtered on the present calls calculated from MAS5. Probesets that were called "Present" in at least one half of the total samples were selected, resulting in 14548 probesets (37% genome). Significance Analysis of Microarrays (SAM) was used to identify differentially regulated genes at a cutoff of a 5% FDR. Applying different thresholds for filtering (Present in one third of the samples, 6 of 18) yielded similar results after SAM analysis. The significance of overlaps between two sets of genes was calculated by defining the appropriate sample space (filtered list of 14548 probesets for differentially expressed genes and Z-chromosome probesets for compensated and non-compensated overlaps) and then randomly selecting the specified number of genes in each group. This was simulated 10000 times to obtain a P-value. The comparisons with Melamed and Arnold studies  were done using DAVID  by extracting the Entrez gene identifiers from the ID conversion interface. The probeset annotation consisted of standard chicken genome annotation as well as from orthology to other organisms (computed by Manteia). Some of the probesets corresponded to 2 or more models of the same gene. This duplication of probeset-gene annotation can result in faulty conclusions in GO analysis. Manteia removes the multiplicity of annotation by assigning a single gene to a probeset by either removing the most poorly annotated genes or choosing one randomly when all are well annotated. In Table 2, when multiple probesets corresponded to a single gene, only the probesets with significant fold change were retained. All statistical analyses and graph generation were conducted using the Bioconductor 2.0 package on R-2.6.1 platform.
The 406 differentially regulated probesets between males and females corresponding to 246 Ensembl genes were chosen for further analysis. Manteia-GOstat interface http://research.stowers-institute.org:8000/Manteia was used to calculate the significantly enriched GO categories in the dataset. The P-values were corrected for multiple tests using FDR . Some of the relevant GO Biological Process (BP) categories along with their adjusted P-values are listed in Table 1. Some of the BP categories (such as, "steroid BP" and "Sterol BP") were significantly enriched; whereas, other categories (such as "spermatogenesis") have a P-value < 0.2, and thus were not included.
Cloning and in situ hybridization
Genes were cloned by PCR amplification using sequence-specific primers and cDNA prepared from a mixed pool of HH3 to HH6 chicken embryos. To make antisense in situ probes, T7 transcription sequence was added to the 3' reverse primer. Antisense in situ probes were made using T7 in vitro transcription reactions with the PCR products as templates. The primer sequences for the cognitive genes are listed as follows (The T7 promoter sequence is highlighted in bold.):
FANCG: forward (AACTACAGGCACCTTTGCACC); reverse (CTAATACGACTCACTATAGGG GGCTATTTCCACTCCTGATCTC).
HSD17B4: forward (GGAGCTTCAGTGGTTGTGAATG); reverse (CTAATACGACTCACTATAGGG CTGCTGCGGAGAAACAAAGGAG).
PELOTA: forward (TACCACACCATTGAGCTGGAGC); reverse (CTAATACGACTCACTATAGGG TCTGTTTCTGGTACTGCTGGAG).
PGTER4: forward (AAGTCCAGGAAGGAGCAGAAGG); reverse (CTAATACGACTCACTATAGGG TTCAATGTCTCAGTGGGGAAGG).
PKCIW: forward (CTGTGAGATACCCACCCTCAG); reverse(CTAATACGACTCACTATAGGG AAGCAGTGTCAAAACTCCGAG).
StARD4: forward (ATGCAGTGGTGTGGCGTAAACC); reverse (CTAATACGACTCACTATAGGG CCAGAGGAACATCTTCCATGTG).
TXNL1: forward (TGAGTAACAAGTACCCTCAGG); reverse (CTAATACGACTCACTATAGGG CACGTAATGGGGATTACAACAG).
To identify the sex of embryos before harvesting for in situ, we genotyped embryos using PKCIW 3' UTR-specific, primer-based PCR reactions on genomic DNA. Briefly, a piece of embryonic tissue was extracted and lysed in lysis buffer. 2 μl of crude lysate was added as a template to PCR reaction using the primer pair: forward (CTGTGAGATACCCACCCTCAG)/reverse (TGAGATACCCACCCTCAGTCC). Genomic DNA of female embryos gives rise of a product band of 500 bp, but does not in male embryos. Fluoresceine-labeled antisense riboprobes of genes of interest were synthesized in vitro. In situ probes were co-hybridized with Digoxigenin-labeled PKCIW probes to HH4-to-HH6 embryos. Alkaline phosphotase conjugated anti-fluoresceine Fab and horseradish peroxidase conjugated anti-Digoxigenin Fab (Roche) were diluted and added together at 1:1000. First, PKCIW in situ was revealed as red fluorescence with fluorescent TSA-Cy3 substrate (Perkin Elmer). Then, the sex of samples was identified by the presence or absence of PKCIW in situ signal. Samples were then regrouped as females or males. The in situ signal of the gene of interest was revealed by adding NBT/BCIP substrate for alkaline phosphotase reaction. Time of coloration was the same for the female group and male groups. Female and male samples of the same gene were photographed at the same illumination and exposure conditions so that in situ signals could be directly compared.
Hamburger and Hamilton
false discovery rate
male hypermethylated region
We thank all members of our laboratory for useful discussions and S. Esteban for artwork. This research was funded in part from NIH grant R02HD043158 to O.P., the Stowers Institute for Medical Research and Howard Hughes Medical Institute.
- Zarkower D: Establishing sexual dimorphism: conservation amidst diversity?. Nat Rev Genet. 2001, 2 (3): 175-185. 10.1038/35056032.PubMedView ArticleGoogle Scholar
- Heard E, Disteche CM: Dosage compensation in mammals: fine-tuning the expression of the X chromosome. Genes Dev. 2006, 20 (14): 1848-1867. 10.1101/gad.1422906.PubMedView ArticleGoogle Scholar
- Straub T, Becker PB: Dosage compensation: the beginning and end of generalization. Nat Rev Genet. 2007, 8 (1): 47-57. 10.1038/nrg2013.PubMedView ArticleGoogle Scholar
- Ellegren H: Hens, cocks and avian sex determination. A quest for genes on Z or W?. EMBO Rep. 2001, 2 (3): 192-196. 10.1093/embo-reports/kve050.PubMed CentralPubMedView ArticleGoogle Scholar
- Smith CA, Sinclair AH: Sex determination: insights from the chicken. Bioessays. 2004, 26 (2): 120-132. 10.1002/bies.10400.PubMedView ArticleGoogle Scholar
- Scholz B, Kultima K, Mattsson A, Axelsson J, Brunstrom B, Halldin K, Stigson M, Dencker L: Sex-dependent gene expression in early brain development of chicken embryos. BMC neuroscience. 2006, 7: 12-10.1186/1471-2202-7-12.PubMed CentralPubMedView ArticleGoogle Scholar
- Itoh Y, Melamed E, Yang X, Kampf K, Wang S, Yehya N, Van Nas A, Replogle K, Band MR, Clayton DF: Dosage compensation is less effective in birds than in mammals. J Biol. 2007, 6 (1): 2-10.1186/jbiol53.PubMed CentralPubMedView ArticleGoogle Scholar
- Melamed E, Arnold AP: Regional differences in dosage compensation on the chicken Z chromosome. Genome Biol. 2007, 8 (9): R202-10.1186/gb-2007-8-9-r202.PubMed CentralPubMedView ArticleGoogle Scholar
- Ellegren H, Hultin-Rosenberg L, Brunstrom B, Dencker L, Kultima K, Scholz B: Faced with inequality: chicken do not have a general dosage compensation of sex-linked genes. BMC biology. 2007, 5: 40-10.1186/1741-7007-5-40.PubMed CentralPubMedView ArticleGoogle Scholar
- Mank JE, Hultin-Rosenberg L, Webster MT, Ellegren H: The unique genomic properties of sex-biased genes: insights from avian microarray data. BMC Genomics. 2008, 9: 148-10.1186/1471-2164-9-148.PubMed CentralPubMedView ArticleGoogle Scholar
- Mank JE, Ellegren H: All dosage compensation is local: Gene-by-gene regulation of sex-biased expression on the chicken Z chromosome. Heredity. 2009, 102 (3): 312-320. 10.1038/hdy.2008.116.PubMedView ArticleGoogle Scholar
- Arnold AP, Itoh Y, Melamed E: A bird's-eye view of sex chromosome dosage compensation. Annual review of genomics and human genetics. 2008, 9: 109-127. 10.1146/annurev.genom.9.081307.164220.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 (7261): 267-271. 10.1038/nature08298.PubMedView 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. Journal of biochemistry. 2006, 139 (1): 91-97. 10.1093/jb/mvj004.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 (10): 3645-3660.PubMed CentralPubMedView ArticleGoogle Scholar
- Sekido R, Lovell-Badge R: Sex determination and SRY: down to a wink and a nudge?. Trends Genet. 2009, 25 (1): 19-29. 10.1016/j.tig.2008.10.008.PubMedView ArticleGoogle Scholar
- Hamburger V, Hamilton HL: A series of normal stages in the development of the chick embryo (1951). DevDyn. 1992, 195: 231-272.Google Scholar
- Tusher VG, Tibshirani R, Chu G: Significance analysis of microarrays applied to the ionizing radiation response. Proc Natl Acad Sci USA. 2001, 98 (9): 5116-5121. 10.1073/pnas.091062498.PubMed CentralPubMedView ArticleGoogle Scholar
- Green CD, Thompson PD, Johnston PG, El-Tanani MK: Interaction between transcription factor, basal transcription factor 3, and the NH2-terminal domain of human estrogen receptor alpha. Mol Cancer Res. 2007, 5 (11): 1191-1200. 10.1158/1541-7786.MCR-07-0123.PubMedView ArticleGoogle Scholar
- Lichanska AM, Waters MJ: How growth hormone controls growth, obesity and sexual dimorphism. Trends Genet. 2008, 24 (1): 41-47. 10.1016/j.tig.2007.10.006.PubMedView ArticleGoogle Scholar
- Teranishi M, Shimada Y, Hori T, Nakabayashi O, Kikuchi T, Macleod T, Pym R, Sheldon B, Solovei I, Macgregor H: Transcripts of the MHM region on the chicken Z chromosome accumulate as non-coding RNA in the nucleus of female cells adjacent to the DMRT1 locus. Chromosome Res. 2001, 9 (2): 147-165. 10.1023/A:1009235120741.PubMedView ArticleGoogle Scholar
- Melamed E, Elashoff D, Arnold AP: Evaluating dosage compensation on the chicken Z chromosome: should effective dosage compensation eliminate sexual bias?. Heredity. 2009, 103 (5): 357-359. 10.1038/hdy.2009.91.PubMedView ArticleGoogle Scholar
- Mank JE, Ellegren H: Sex bias in gene expression is not the same as dosage compensation. Heredity. 2009, 103 (5): 434-10.1038/hdy.2009.90.PubMedView ArticleGoogle Scholar
- Lin H, Gupta V, Vermilyea MD, Falciani F, Lee JT, O'Neill LP, Turner BM: Dosage compensation in the mouse balances up-regulation and silencing of X-linked genes. PLoS Biol. 2007, 5 (12): e326-10.1371/journal.pbio.0050326.PubMed CentralPubMedView ArticleGoogle Scholar
- Raymond CS, Shamu CE, Shen MM, Seifert KJ, Hirsch B, Hodgkin J, Zarkower D: Evidence for evolutionary conservation of sex-determining genes. Nature. 1998, 391 (6668): 691-695. 10.1038/35618.PubMedView ArticleGoogle Scholar
- Smith CA, McClive PJ, Western PS, Reed KJ, Sinclair AH: Conservation of a sex-determining gene. Nature. 1999, 402 (6762): 601-602. 10.1038/45130.PubMedView ArticleGoogle Scholar
- Raymond CS, Kettlewell JR, Hirsch B, Bardwell VJ, Zarkower D: Expression of Dmrt1 in the genital ridge of mouse and chicken embryos suggests a role in vertebrate sexual development. Dev Biol. 1999, 215 (2): 208-220. 10.1006/dbio.1999.9461.PubMedView ArticleGoogle Scholar
- Zeeberg BR, Feng W, Wang G, Wang MD, Fojo AT, Sunshine M, Narasimhan S, Kane DW, Reinhold WC, Lababidi S: GoMiner: a resource for biological interpretation of genomic and proteomic data. Genome Biol. 2003, 4 (4): R28-10.1186/gb-2003-4-4-r28.PubMed CentralPubMedView ArticleGoogle Scholar
- Eberhart CG, Wasserman SA: The pelota locus encodes a protein required for meiotic cell division: an analysis of G2/M arrest in Drosophila spermatogenesis. Development. 1995, 121 (10): 3477-3486.PubMedGoogle Scholar
- Oh B, Hwang SY, Solter D, Knowles BB: Spindlin, a major maternal transcript expressed in the mouse during the transition from oocyte to embryo. Development. 1997, 124 (2): 493-503.PubMedGoogle Scholar
- Burgoyne PS, Mahadevaiah SK, Sutcliffe MJ, Palmer SJ: Fertility in mice requires X-Y pairing and a Y-chromosomal "spermiogenesis" gene mapping to the long arm. Cell. 1992, 71 (3): 391-398. 10.1016/0092-8674(92)90509-B.PubMedView ArticleGoogle Scholar
- Shah NM, Pisapia DJ, Maniatis S, Mendelsohn MM, Nemes A, Axel R: Visualizing sexual dimorphism in the brain. Neuron. 2004, 43 (3): 313-319. 10.1016/j.neuron.2004.07.008.PubMedView ArticleGoogle Scholar
- Mu X, Chang C: TR2 orphan receptor functions as negative modulator for androgen receptor in prostate cancer cells PC-3. The Prostate. 2003, 57 (2): 129-133. 10.1002/pros.10282.PubMedView ArticleGoogle Scholar
- Oberkotter LV, Tenore A, Koldovsky O: Sex differences in serum levels of N-acetyl-beta-hexosaminidase in infancy: correlation of enzyme activity with testosterone levels. Early human development. 1980, 4 (3): 325-331. 10.1016/0378-3782(80)90035-3.PubMedView ArticleGoogle Scholar
- Wilhelm D, Palmer S, Koopman P: Sex determination and gonadal development in mammals. Physiol Rev. 2007, 87 (1): 1-28. 10.1152/physrev.00009.2006.PubMedView ArticleGoogle Scholar
- Morris JA, Jordan CL, Breedlove SM: Sexual differentiation of the vertebrate nervous system. Nat Neurosci. 2004, 7 (10): 1034-1039. 10.1038/nn1325.PubMedView ArticleGoogle Scholar
- Handa RJ, Corbier P, Shryne JE, Schoonmaker JN, Gorski RA: Differential effects of the perinatal steroid environment on three sexually dimorphic parameters of the rat brain. Biology of reproduction. 1985, 32 (4): 855-864. 10.1095/biolreprod32.4.855.PubMedView ArticleGoogle Scholar
- Gurney ME, Konishi M: Hormone-Induced Sexual Differentiation of Brain and Behavior in Zebra Finches. Science. 1980, 208 (4450): 1380-1383. 10.1126/science.208.4450.1380.PubMedView ArticleGoogle Scholar
- Arnold AP: The effects of castration on song development in zebra finches (Poephila guttata). J Exp Zool. 1975, 191 (2): 261-278. 10.1002/jez.1401910212.PubMedView ArticleGoogle Scholar
- Agate RJ, Grisham W, Wade J, Mann S, Wingfield J, Schanen C, Palotie A, Arnold AP: Neural, not gonadal, origin of brain sex differences in a gynandromorphic finch. Proc Natl Acad Sci USA. 2003, 100 (8): 4873-4878. 10.1073/pnas.0636925100.PubMed CentralPubMedView ArticleGoogle Scholar
- Holloway CC, Clayton DF: Estrogen synthesis in the male brain triggers development of the avian song control pathway in vitro. Nat Neurosci. 2001, 4 (2): 170-175. 10.1038/84001.PubMedView ArticleGoogle Scholar
- London SE, Monks DA, Wade J, Schlinger BA: Widespread capacity for steroid synthesis in the avian brain and song system. Endocrinology. 2006, 147 (12): 5975-5987. 10.1210/en.2006-0154.PubMed CentralPubMedView ArticleGoogle Scholar
- Dequeant ML, Glynn E, Gaudenz K, Wahl M, Chen J, Mushegian A, Pourquie O: A complex oscillating network of signaling genes underlies the mouse segmentation clock. Science. 2006, 314 (5805): 1595-1598. 10.1126/science.1133141.PubMedView ArticleGoogle Scholar
- Huang da W, Sherman BT, Lempicki RA: Systematic and integrative analysis of large gene lists using DAVID bioinformatics resources. Nature protocols. 2009, 4 (1): 44-57. 10.1038/nprot.2008.211.PubMedView ArticleGoogle Scholar
- Benjamini Y, Drai D, Elmer G, Kafkafi N, Golani I: Controlling the false discovery rate in behavior genetics research. Behavioural brain research. 2001, 125 (1-2): 279-284. 10.1016/S0166-4328(01)00297-2.PubMedView 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.