Differentially expressed genes from RNA-Seq and functional enrichment results are affected by the choice of single-end versus paired-end reads and stranded versus non-stranded protocols.

BACKGROUND: RNA-Seq is now widely used as a research tool. Choices must be made whether to use paired-end (PE) or single-end (SE) sequencing, and whether to use strand-specific or non-specific (NS) library preparation kits. To date there has been no analysis of the effect of these choices on identifying differentially expressed genes (DEGs) between controls and treated samples and on downstream functional analysis. RESULTS: We undertook four mammalian transcriptomics experiments to compare the effect of SE and PE protocols on read mapping, feature counting, identification of DEGs and functional analysis. For three of these experiments we also compared a non-stranded (NS) and a strand-specific approach to mapping the paired-end data. SE mapping resulted in a reduced number of reads mapped to features, in all four experiments, and lower read count per gene. Up to 4.3% of genes in the SE data and up to 12.3% of genes in the NS data had read counts which were significantly different compared to the PE data. Comparison of DEGs showed the presence of false positives (average 5%, using voom) and false negatives (average 5%, using voom) using the SE reads. These increased further, by one or two percentage points, with the NS data. Gene ontology functional enrichment (GO) of the DEGs arising from SE or NS approaches, revealed striking differences in the top 20 GO terms, with as little as 40% concordance with PE results. Caution is therefore advised in the interpretation of such results. By comparison, there was overall consistency in gene set enrichment analysis results. CONCLUSIONS: A strand-specific protocol should be used in library preparation to generate the most reliable and accurate profile of expression. Ideally PE reads are also recommended particularly for transcriptome assembly. Whilst SE reads produce a DEG list with around 5% of false positives and false negatives, this method can substantially reduce sequencing cost and this saving could be used to increase the number of biological replicates thereby increasing the power of the experiment. As SE reads, when used in association with gene set enrichment, can generate accurate biological results, this may be a desirable trade-off.

RNA-Seq analysis of Gtf2ird1 knockout epidermal tissue provides potential insights into molecular mechanisms underpinning Williams-Beuren syndrome.

BACKGROUND: Williams-Beuren Syndrome (WBS) is a genetic disorder associated with multisystemic abnormalities, including craniofacial dysmorphology and cognitive defects. It is caused by a hemizygous microdeletion involving up to 28 genes in chromosome 7q11.23. Genotype/phenotype analysis of atypical microdeletions implicates two evolutionary-related transcription factors, GTF2I and GTF2IRD1, as prime candidates for the cause of the facial dysmorphology. RESULTS: Using a targeted Gtf2ird1 knockout mouse, we employed massively-parallel sequencing of mRNA (RNA-Seq) to understand changes in the transcriptional landscape associated with inactivation of Gtf2ird1 in lip tissue. We found widespread dysregulation of genes including differential expression of 78 transcription factors or coactivators, several involved in organ development including Hey1, Myf6, Myog, Dlx2, Gli1, Gli2, Lhx2, Pou3f3, Sox2, Foxp3. We also found that the absence of GTF2IRD1 is associated with increased expression of genes involved in cellular proliferation, including growth factors consistent with the observed phenotype of extreme thickening of the epidermis. At the same time, there was a decrease in the expression of genes involved in other signalling mechanisms, including the Wnt pathway, indicating dysregulation in the complex networks necessary for epidermal differentiation and facial skin patterning. Several of the differentially expressed genes have known roles in both tissue development and neurological function, such as the transcription factor Lhx2 which regulates several genes involved in both skin and brain development. CONCLUSIONS: Gtf2ird1 inactivation results in widespread gene dysregulation, some of which may be due to the secondary consequences of gene regulatory network disruptions involving several transcription factors and signalling molecules. Genes involved in growth factor signalling and cell cycle progression were identified as particularly important for explaining the skin dysmorphology observed in this mouse model. We have noted that a number of the dysregulated genes have known roles in brain development as well as epidermal differentiation and maintenance. Therefore, this study provides clues as to the underlying mechanisms that may be involved in the broader profile of WBS.

Protein tyrosine kinase 2 beta (PTK2B), but not focal adhesion kinase (FAK), is expressed in a sexually dimorphic pattern in developing mouse gonads.

Sexual reproduction is essential for the propagation and the maintenance of fitness of our species, and is dependent on the correct development of the bipotential genital ridges into testes and ovaries. Although several transcription factors, secreted signaling molecules, and their receptors have been found to be important for testis determination and early gonad development, comparatively little research has been carried out into intracellular signal transduction pathways activated during these processes. Focal adhesion kinase (FAK) and protein tyrosine kinase 2 beta (PTK2B) form one group of cytosolic tyrosine kinases that are known to be important for processes such as cell proliferation, differentiation, and motility. Here, we describe the temporal and spatial expression patterns of Fak and Ptk2b mRNA and protein during sex determination and early gonadogenesis in mouse embryos. Ptk2b mRNA and PTK2B protein were expressed in testes from 11.5 days post coitum onward, predominantly in developing Sertoli cells, in a SOX9-dependent manner. Fak mRNA and FAK protein were expressed in gonads of both sexes at all stages examined. Our data suggest cell type- and stage-specific roles for PTK2B during early testis development.

Whole Mount In Situ Hybridization of Mid-Gestation Mouse Embryos.

In situ hybridization is a powerful technique that allows the visualization of specific RNA species in biological samples in exquisite detail. It has been particularly well explored in the field of developmental genetics. The spatial and temporal patterns of RNA expression provide us with critical information on likely gene function during embryonic development, and often inform the decision on whether to attempt further gene manipulation approaches. Furthermore, once a mouse strain with altered gene function has been created, in situ hybridization is a critical tool for revealing how the development of embryos with the mutation differs from that of wild-type embryos, and thus infer the function of the altered gene. Here, a well-tested protocol used to visualize RNA expression in whole-mount mid-gestation mouse embryos ranging from 8.5 to 14.5 days post-coitum (dpc) is described. © 2020 Wiley Periodicals LLC. Basic Protocol 1: RNA probe synthesis Alternate Protocol: Preparation of DNA template by PCR Basic Protocol 2: Embryo dissection Basic Protocol 3: Whole mount in situ hybridization Support Protocol: Generation of embryo powder.

Sox9-dependent expression of Gstm6 in Sertoli cells during testis development in mice.

Glutathione S-transferases (GSTs) are an important family of multifunctional enzymes that play a role in the protection of tissues by the detoxification of hazardous and carcinogenic compounds. We found previously that Gstm6 is upregulated in the somatic cells of male mouse fetal gonads relative to female gonads. In this study, we describe the spatial and temporal expression pattern of Gstm6 during mouse development. We show that Gstm6 is predominantly expressed in the reproductive system, at significantly higher levels in XY gonads compared with XX gonads from 11.5 dpc onwards, and remains expressed in the testes in adult mice. Its expression is associated with the Sertoli cell lineage, and is dependent on the expression of the male sex-determining gene Sox9. Our data suggest that Gstm6 plays a male-specific role in gonad development or function, possibly by modulating the exposure of somatic tissue and/or germ cells to endogenous or exogenous toxicants.

Aard is specifically up-regulated in Sertoli cells during mouse testis differentiation.

Aard (alanine and arginine rich domain) is a gene of unknown function, previously reported to show sexually dimorphic expression in fetal mouse gonads. Here we describe the spatio-temporal expression profile of Aard during gonad development. The period of elevated mRNA expression coincides with early differentiation of the testis and is limited to Sertoli cells of the developing testis cords. Although low levels of Aard transcripts were detected in other tissues by quantitative RT-PCR assays, high levels of Aard expression is specific to the testis in both embryonic and adult mice.

Epidermal YAP activity drives canonical WNT16/ß-catenin signaling to promote keratinocyte proliferation in vitro and in the murine skin.

The skin constantly self-renews throughout adult life. Wnt/ß-catenin signaling plays a key role in promoting keratinocyte proliferation in the hair follicles and in the interfollicular epidermis. A recent report demonstrated that epidermal YAP activity drives ß-catenin activation to promote keratinocyte proliferation in the murine skin. However, it remains unclear whether this is caused by paracrine activation of canonical Wnt signaling or through other YAP/ß-catenin regulatory interactions. In the present study, we found that XAV939-inhibition of canonical WNT signaling in skin of YAP2-5SA-ΔC mice resulted in diminished ß-catenin activation, reduced keratinocyte proliferation, and a mitigation of the hyperplastic abnormalities in the interfollicular epidermis, signifying a canonical WNT ligand-dependent mechanism. Our subsequent analyses determined that WNT16 is produced in response to YAP activity in keratinocytes both in vitro and in vivo, and that WNT16 drives HaCaT keratinocyte proliferation via canonical WNT16/ß-catenin signaling. We conclude that under normal physiological conditions WNT16 is the paracrine WNT ligand secreted in response to epidermal YAP activity that promotes cell proliferation in the interfollicular epidermis. This study delineates a fundamental YAP-driven mechanism that controls normal skin regeneration, and that may be perturbed in human regenerative disease displaying increased YAP and WNT signaling activity.

Plau and Tgfbr3 are YAP-regulated genes that promote keratinocyte proliferation.

Yes-associated protein (YAP) is a mechanosensor protein and a downstream effector of the Hippo kinase pathway, which controls organ growth, cell proliferation, survival, maintenance and regeneration. Unphosphorylated YAP translocates to the nucleus where it acts as a cofactor of primarily the TEAD transcription factors to activate target gene transcription and cell proliferation. Perturbed YAP activation results in tumorigenesis. The pathways downstream of activated YAP that drive cell proliferation remain relatively unexplored. In this study, we employed YAP2-5SA-∆C transgenic mice, which overexpress a mildly activated YAP mutant protein in basal keratinocytes leading to increased proliferation of the epidermal stem/progenitor cell populations. We performed massively-parallel sequencing of skin biopsy mRNA (RNA-Seq) and found dysregulation of 1491 genes in YAP2-5SA-∆C skin, including many with roles in cell activation and proliferation. Furthermore, we found that 150 of these dysregulated genes harbored YAP/TEAD binding motifs in the 3' UTR, suggesting that these may be direct YAP/TEAD target genes in the control of epidermal regeneration. Further validation and functional characterization assays identified Plau and Tgfbr3 as prime candidate genes that may be activated by epidermal YAP activity in the mouse skin in vivo to promote keratinocyte proliferation. This study provides novel insights into the mechanisms regulated by YAP that control tissue homeostasis, and in particular in conditions where YAP is aberrantly activated such as in neoplastic and regenerative skin disease.

Msx1 and Dlx5 act independently in development of craniofacial skeleton, but converge on the regulation of Bmp signaling in palate formation.

Msx and Dlx homeoproteins control the morphogenesis and organization of craniofacial skeletal structures, specifically those derived from the pharyngeal arches. In vitro Msx and Dlx proteins have opposing transcriptional properties and form heterodimeric complexes via their homeodomain with reciprocal functional repression. In this report we examine the skeletal phenotype of Msx1; Dlx5 double knock-out (DKO) mice in relationship with their expression territories during craniofacial development. Co-expression of Dlx5 and Msx1 is only observed in embryonic tissues in which these genes have independent functions, and thus direct protein interactions are unlikely to control morphogenesis of the cranium. The DKO craniofacial phenotypes indicate a complex interplay between these genes, acting independently (mandible and middle ear), synergistically (deposition of bone tissue) or converging on the same morphogenetic process (palate growth and closure). In the latter case, the absence of Dlx5 rescues in part the Msx1-dependent defects in palate growth and elevation. At the basis of this effect, our data implicate the Bmp (Bmp7, Bmp4)/Bmp antagonist (Follistatin) signal: in the Dlx5(-/-) palate changes in the expression level of Bmp7 and Follistatin counteract the reduced Bmp4 expression. These results highlight the importance of precise spatial and temporal regulation of the Bmp/Bmp antagonist system during palate closure.

Positive regulatory interactions between YAP and Hedgehog signalling in skin homeostasis and BCC development in mouse skin in vivo.

Skin is a highly plastic tissue that undergoes tissue turnover throughout life, but also in response to injury. YAP and Hedgehog signalling play a central role in the control of epidermal stem/progenitor cells in the skin during embryonic development, in postnatal tissue homeostasis and in skin carcinogenesis. However, the genetic contexts in which they act to control tissue homeostasis remain mostly unresolved. We provide compelling evidence that epidermal YAP and Hedgehog/GLI2 signalling undergo positive regulatory interactions in the control of normal epidermal homeostasis and in basal cell carcinoma (BCC) development, which in the large majority of cases is caused by aberrant Hedgehog signalling activity. We report increased nuclear YAP and GLI2 activity in the epidermis and BCCs of K14-CreER/Rosa-SmoM2 transgenic mouse skin, accompanied with increased ROCK signalling and ECM remodelling. Furthermore, we found that epidermal YAP activity drives GLI2 nuclear accumulation in the skin of YAP2-5SA-ΔC mice, which depends on epidermal ß-catenin activation. Lastly, we found prominent nuclear activity of GLI2, YAP and ß-catenin, concomitant with increased ROCK signalling and stromal fibrosis in human BCC. Our work provides novel insights into the molecular mechanisms underlying the interplay between cell signalling events and mechanical force in normal tissue homeostasis in vivo, that could potentially be perturbed in BCC development.

Epidermal YAP2-5SA-ΔC Drives ß-Catenin Activation to Promote Keratinocyte Proliferation in Mouse Skin In Vivo.

The epidermis is a highly regenerative tissue. YAP is a pivotal regulator of stem/progenitor cells in tissue regeneration, including in the epidermis. The molecular mechanisms downstream of YAP that activate epidermal cell proliferation remain largely unknown. We found that YAP and ß-catenin co-localize in the nuclei of keratinocytes in the regenerating epidermis in vivo and in proliferating HaCaT keratinocytes in vitro. Inactivation of YAP in HaCaT keratinocytes resulted in reduced activated ß-catenin and reduced keratinocyte numbers in vitro. In addition, we found that in the hyperplastic epidermis of YAP2-5SA-ΔC mice, the mutant YAP2-5SA-ΔC protein was predominantly localized in the keratinocyte nuclei and caused increased expression of activated nuclear ß-catenin. Accordingly, ß-catenin transcriptional activity was elevated in the skin of live YAP2-5SA-ΔC/TOPFLASH mice. Lastly, loss of ß-catenin in basal keratinocytes of YAP2-5SA-ΔC/K14-creERT/CtnnB1-/- mice resulted in reduced proliferation of basal keratinocytes and a striking rescue of the hyperplastic abnormalities. Taken together, our work shows that YAP2-5SA-ΔC drives ß-catenin activity to promote basal keratinocyte proliferation in the mouse skin in vivo. Our data shine new light on the etiology of regenerative dermatological disorders and other human diseases that display increased YAP and ß-catenin activity.

CUBIC Protocol Visualizes Protein Expression at Single Cell Resolution in Whole Mount Skin Preparations.

The skin is essential for our survival. The outer epidermal layer consists of the interfollicular epidermis, which is a stratified squamous epithelium covering most of our body, and epidermal appendages such as the hair follicles and sweat glands. The epidermis undergoes regeneration throughout life and in response to injury. This is enabled by K14-expressing basal epidermal stem/progenitor cell populations that are tightly regulated by multiple regulatory mechanisms active within the epidermis and between epidermis and dermis. This article describes a simple method to clarify full thickness mouse skin biopsies, and visualize K14 protein expression patterns, Ki67 labeled proliferating cells, Nile Red labeled sebocytes, and DAPI nuclear labeling at single cell resolution in 3D. This method enables accurate assessment and quantification of skin anatomy and pathology, and of abnormal epidermal phenotypes in genetically modified mouse lines. The CUBIC protocol is the best method available to date to investigate molecular and cellular interactions in full thickness skin biopsies at single cell resolution.

The expression pattern of EVA1C, a novel Slit receptor, is consistent with an axon guidance role in the mouse nervous system.

The Slit/Robo axon guidance families play a vital role in the formation of neural circuitry within select regions of the developing mouse nervous system. Typically Slits signal through the Robo receptors, however they also have Robo-independent functions. The novel Slit receptor Eva-1, recently discovered in C. elegans, and the human orthologue of which is located in the Down syndrome critical region on chromosome 21, could account for some of these Robo independent functions as well as provide selectivity to Robo-mediated axon responses to Slit. Here we investigate the expression of the mammalian orthologue EVA1C in regions of the developing mouse nervous system which have been shown to exhibit Robo-dependent and -independent responses to Slit. We report that EVA1C is expressed by axons contributing to commissures, tracts and nerve pathways of the developing spinal cord and forebrain. Furthermore it is expressed by axons that display both Robo-dependent and -independent functions of Slit, supporting a role for EVA1C in Slit/Robo mediated neural circuit formation in the developing nervous system.

Yap controls stem/progenitor cell proliferation in the mouse postnatal epidermis.

Tissue renewal is an ongoing process in the epithelium of the skin. We have begun to examine the genetic mechanisms that control stem/progenitor cell activation in the postnatal epidermis. The conserved Hippo pathway regulates stem cell turnover in arthropods through to vertebrates. Here we show that its downstream effector, yes-associated protein (YAP), is active in the stem/progenitor cells of the postnatal epidermis. Overexpression of a C-terminally truncated YAP mutant in the basal epidermis of transgenic mice caused marked expansion of epidermal stem/progenitor cell populations. Our data suggest that the C-terminus of YAP controls the balance between stem/progenitor cell proliferation and differentiation in the postnatal interfollicular epidermis. We conclude that YAP functions as a molecular switch of stem/progenitor cell activation in the epidermis. Moreover, our results highlight YAP as a possible therapeutic target for diseases such as skin cancer, psoriasis, and epidermolysis bullosa.

Sex-specific expression of a novel gene Tmem184a during mouse testis differentiation.

During mouse embryogenesis, the fate of the bipotential gonads is sealed around 10.5 days post coitum (dpc) when the Y-linked gene Sry specifies the differentiation of testes in males, whereas in females, absence of Sry results in ovary formation. Apart from the pivotal action of Sry, many other genes are known to be involved in sex determination and subsequent differentiation. Much is still unknown regarding the regulatory hierarchy governing these events and many more sex differentiation genes are yet to be discovered. In this study, we investigated the expression of Tmem184a, a novel gene encoding a protein of unknown function, but with predicted kinase activity, during mouse embryogenesis. We show that Tmem184a is expressed at high levels in the developing testis from 11.5 dpc, a time of active proliferation and differentiation. Tmem184a expression is further shown to be expressed exclusively within the Sertoli cells of the developing testis cords, suggesting that it may mediate sex-specific signaling events during Sertoli cell differentiation.

A high-resolution anatomical ontology of the developing murine genitourinary tract.

Cataloguing gene expression during development of the genitourinary tract will increase our understanding not only of this process but also of congenital defects and disease affecting this organ system. We have developed a high-resolution ontology with which to describe the subcompartments of the developing murine genitourinary tract. This ontology incorporates what can be defined histologically and begins to encompass other structures and cell types already identified at the molecular level. The ontology is being used to annotate in situ hybridisation data generated as part of the Genitourinary Development Molecular Anatomy Project (GUDMAP), a publicly available data resource on gene and protein expression during genitourinary development. The GUDMAP ontology encompasses Theiler stage (TS) 17-27 of development as well as the sexually mature adult. It has been written as a partonomic, text-based, hierarchical ontology that, for the embryological stages, has been developed as a high-resolution expansion of the existing Edinburgh Mouse Atlas Project (EMAP) ontology. It also includes group terms for well-characterised structural and/or functional units comprising several sub-structures, such as the nephron and juxtaglomerular complex. Each term has been assigned a unique identification number. Synonyms have been used to improve the success of query searching and maintain wherever possible existing EMAP terms relating to this organ system. We describe here the principles and structure of the ontology and provide representative diagrammatic, histological, and whole mount and section RNA in situ hybridisation images to clarify the terms used within the ontology. Visual examples of how terms appear in different specimen types are also provided.

Expression profiling of purified mouse gonadal somatic cells during the critical time window of sex determination reveals novel candidate genes for human sexual dysgenesis syndromes.

Despite the identification of SRY as the testis-determining gene in mammals, the genetic interactions controlling the earliest steps of male sex determination remain poorly understood. In particular, the molecular lesions underlying a high proportion of human XY gonadal dysgenesis, XX maleness and XX true hermaphroditism remain undiscovered. A number of screens have identified candidate genes whose expression is modulated during testis or ovary differentiation in mice, but these screens have used whole gonads, consisting of multiple cell types, or stages of gonadal development well beyond the time of sex determination. We describe here a novel reporter mouse line that expresses enhanced green fluorescent protein under the control of an Sf1 promoter fragment, marking Sertoli and granulosa cell precursors during the critical period of sex determination. These cells were purified from gonads of male and female transgenic embryos at 10.5 dpc (shortly after Sry transcription is activated) and 11.5 dpc (when Sox9 transcription begins), and their transcriptomes analysed using Affymetrix genome arrays. We identified 266 genes, including Dhh, Fgf9 and Ptgds, that were upregulated and 50 genes that were downregulated in 11.5 dpc male somatic gonad cells only, and 242 genes, including Fst, that were upregulated in 11.5 dpc female somatic gonad cells only. The majority of these genes are novel genes that lack identifiable homology, and several human orthologues were found to map to chromosomal loci implicated in disorders of sexual development. These genes represent an important resource with which to piece together the earliest steps of sex determination and gonad development, and provide new candidates for mutation searching in human sexual dysgenesis syndromes.

Genetics of shoulder girdle formation: roles of Tbx15 and aristaless-like genes.

The diverse cellular contributions to the skeletal elements of the vertebrate shoulder and pelvic girdles during embryonic development complicate the study of their patterning. Research in avian embryos has recently clarified part of the embryological basis of shoulder formation. Although dermomyotomal cells provide the progenitors of the scapular blade, local signals appear to have an essential guiding role in this process. These signals differ from those that are known to pattern the more distal appendicular skeleton. We have studied the impact of Tbx15, Gli3, Alx4 and related genes on formation of the skeletal elements of the mouse shoulder and pelvic girdles. We observed severe reduction of the scapula in double and triple mutants of these genes. Analyses of a range of complex genotypes revealed aspects of their genetic relationship, as well as functions that had been previously masked due to functional redundancy. Tbx15 and Gli3 appear to have synergistic functions in formation of the scapular blade. Scapular truncation in triple mutants of Tbx15, Alx4 and Cart1 indicates essential functions for Alx4 and Cart1 in the anterior part of the scapula, as opposed to Gli3 function being linked to the posterior part. Especially in Alx4/Cart1 mutants, the expression of markers such as Pax1, Pax3 and Scleraxis is altered prior to stages when anatomical aberrations are visible in the shoulder region. This suggests a disorganization of the proximal limb bud and adjacent flank mesoderm, and is likely to reflect the disruption of a mechanism providing positional cues to guide progenitor cells to their destination in the pectoral girdle.

Pisrt1, a gene implicated in XX sex reversal, is expressed in gonads of both sexes during mouse development.

XX sex reversal syndromes not involving Sry provide an opportunity to identify and study genes important for sexual development. The polled intersex syndrome (PIS) in goats, which shares some features with blepharophimosis, ptosis, epicanthus inversus syndrome (BPES) in humans, exemplifies such syndromes. BPES is caused by defects in the forkhead transcription factor gene FOXL2, while PIS is caused by a large deletion of goat chromosome 1q43 that affects transcription of the genes Pisrt1 and Foxl2. Pisrt1 is a non-translated gene that has a sexually dimorphic expression pattern in goats. Here, we describe the structure and expression of the mouse Pisrt1 locus, to investigate its likely role in ovarian development more broadly in mammals. This gene showed some sequence similarity, and was found in a similar genomic context, to its goat and human orthologues. Expression analyses indicated that Pisrt1 is transcribed, and its mRNA polyadenylated and exported to the cytoplasm, but no significant open reading frames were found in a 1.5kb mouse genomic region corresponding to goat Pisrt1. Pisrt1 transcripts were expressed very broadly among tissues of the developing mouse embryo, and at similar levels in male and female gonads at each stage examined, as determined by in situ hybridisation and RT-PCR. This profile of expression suggests that Pisrt1 is unlikely to contribute to sex-specific events during gonadal development in mice and that divergent pathways of ovarian development operate among different mammalian species.

Sertoli cell differentiation is induced both cell-autonomously and through prostaglandin signaling during mammalian sex determination.

We have raised an antibody specifically recognizing endogenous mouse SRY protein and used it to investigate the molecular and cellular mode of action of SRY in testis determination. We find that expression of SRY protein closely mirrors the expression of Sry mRNA in mouse genital ridges and is detectable for 6 to 8 h after the mRNA ceases to be detectable. The subset of somatic cells that expresses SRY begins to express SOX9 almost immediately. Since these SOX9-positive cells go on to develop as Sertoli cells, it appears that SRY expression marks the pre-Sertoli cell lineage and leads to up-regulation of Sox9 expression cell-autonomously. However, a small proportion of SOX9-positive cells did not appear to express SRY, possibly reflecting the additional involvement of paracrine signaling in activating Sox9 transcription in these cells. We confirmed by ex vivo cell mixing experiments that SRY is able to engage receptor-mediated signaling to up-regulate Sox9 expression. Finally, we showed by employing specific inhibitors that the causative signaling molecule is prostaglandin D2 (PGD2) and that PGD2 can induce Sox9 transcription in cultured XX gonads. Our data indicate a mechanism whereby Sry uses both a cell-autonomous mechanism and a PGD2-mediated signaling mechanism to stimulate expression of Sox9 and induce the differentiation of Sertoli cells in vivo.