RESEARCH ARTICLE Molecular Reproduction & Development (2014)
Direct and Indirect Consequences on Gene Expression of a Thyroid Hormone Receptor Alpha 1 Mutation Restricted to Sertoli Cells 3 FABRICE CHATONNET,1 GABRIEL LIVERA,2 BETTY FUMEL,3 SOPHIE FOUCHECOURT ,
AND
ERIC FRED FLAMANT1*
1
nomique Fonctionnelle de Lyon, Universite de Lyon, Universite Lyon 1, INRA, CNRS, Cedex, France Institut de Ge , Fontenay-Aux-Roses, France INSERM U967, CEA/DSV/iRCM/SCSR/LDG, Univ Paris Diderot, Sorbonne Paris Cite 3 de Tours, IFCE, Physiologie de la Reproduction et des Comportements, Nouzilly, France INRA UMR85, CNRS, Universite 2
SUMMARY Thyroid hormone is required for the timely transition of Sertoli cells from proliferative to differentiating and maturing. This transition takes place during a critical developmental period in mammals, which in mice is the first post-natal week. In order to identify the underlying molecular mechanisms of this differentiation process, we used Cre/loxP technology to selectively block the function of the thyroid hormone receptor TRa1 in Sertoli cells. We then used RNA-seq to analyze the changes in gene expression induced in the post-natal testis. This differential analysis provides genetic clues to the initial testicular defects resulting from disrupted thyroid hormone signaling, and suggests that Sertoli cells influence germ cells soon after their birth.
Corresponding author: nomique Fonctionnelle Institut de Ge de Lyon de Lyon, INRA, CNRS Universite rieure de Lyon Ecole Normale Supe e d’Italie 46 alle 69364, Lyon Cedex 07, France. E-mail: [email protected]
court and Fre de ric Flamant Sophie Fouche contributed equally to the work.
Mol. Reprod. Dev. 2014. ß 2014 Wiley Periodicals, Inc. Received 3 March 2014; Accepted 22 October 2014
INTRODUCTION Sertoli cells of the adult testis provide the physical environment and factors required for germ cells to differentiate into spermatozoa. The number of Sertoli cells is fixed in the adult as their proliferation is arrested, and they rarely die (Sharpe et al., 2003). The extent of Sertoli cell proliferation, which takes place mainly in the perinatal period in rodents, defines the number of germ cells that can be supported as well as the total gamete output after puberty. The timing of Sertoli cell post-natal differentiation is thus a critical parameter of male fertility, yet little is
ß 2014 WILEY PERIODICALS, INC.
Published online in Wiley Online Library (wileyonlinelibrary.com). DOI 10.1002/mrd.22437
known about the gene regulatory events underlying this process. Thyroid hormones including 3,30 ,5-triiodothyronine (T3) and thyroxine (T4), the low-activity precursor of T3 exert a pleiotropic influence on development and cell differentiation in all vertebrates. T3 acts mainly by binding to members of the
Abbreviations: Amh, anti-Müllerian hormone; P#, post-natal day #; T3, triiodo-thyronine; TR, thyroid hormone receptor.
Molecular Reproduction & Development
nuclear-receptor superfamily. Two mammalian genes, Thra and Thrb, encode the three receptors TRa1, TRb1, and TRb2 (collectively called thyroid receptors [TRs]). TRs are present on most if not all cell types, functioning as ligandactivated transcription factors that directly regulate the transcription of a large number of target genes (Yen et al., 2006). They do so by binding to DNA response elements, which are mainly related to the DR4 consensus element (50 AGGTCANNNNAGGTCA) that is recognized by a TR/RXR heterodimer and is found in regulatory regions of their target genes (Chatonnet et al., 2013). Although the physiological relevance of non-genomic pathways is still debated, reports indicate that TR-encoded products can also participate in signaling at the plasma membrane (Kalyanaraman et al., 2014; Martin et al., 2014). In rodents, a peak in circulating T3 levels occurs during the post-natal period. A transient T3 deficiency at this stage prolongs the Sertoli cell proliferation period and has irreversible consequences on testis development, including an increase in testis weight and daily sperm production in adults (Cooke and Meisami., 1991; Kirby et al., 1992). Reciprocally, excess T3 accelerates Sertoli cell differentiation, resulting in reduced testis weight (van Haaster et al., 1993). Yet, transient variations in T3 concentration have a broad effect on pituitary hormone production, which can also indirectly alter testis development. Therefore, experiments that globally modulate T3 levels do not clarify how Sertoli cells directly respond to T3 stimulation during testis development (Wagner et al., 2008). Genetic experiments using mice indicate that an early effect of T3 on testis development is to influence Sertoli cell differentiation. Like hypothyroid rats, mice carrying a knockout allele of Thra, but not Thrb, have more Sertoli cells and heavier testes (Holsberger et al., 2005). We recently used Cre/loxP recombination to eliminate TRa1 specifically in Sertoli cells (Fumel et al., 2012). This was achieved by crossing mice carrying the TRaAMI allele of Thra (Quignodon et al., 2007) with mice carrying the Amh:Cre transgene, which drives the expression of the Cre recombinase in Sertoli cells using the anti-Müllerian hormone (Amh) promoter (Lecureuil et al., 2002). These mice, called TRaAMI/SC, express the dominant-negative TRa1L400R only in Sertoli cells, and display a testicular phenotype that is very similar to the one reported in hypothyroid or Thra-knockout mice. These conditional mice unambiguously establish the cell-autonomous effects of T3 on Sertoli cells and indicate that the adult testicular phenotype resulting from early transient hypothyroidism is a long-term consequence of a defect in the T3-dependent Sertoli cell differentiation. Thus, while early T3 deficiency broadly influences many cell types along the entire gonadotrophic axis, the ability of T3 to specifically stimulate Sertoli cell differentiation is a key element of its impact on testis development. The fact that TRa1 directly influences cell-cycle exit and differentiation of Sertoli cells offers an opportunity to access the genetic programs that control these cellular processes. We performed a comparative RNA-seq analysis on whole testis of TRaAMI/SC mice at the post-natal stage to gain further insight into the repertoire of TRa1 target genes as
2
CHATONNET
ET AL.
well as the molecular mechanisms underlying the control of Sertoli cell differentiation by T3. This broad survey identified 203 differentially expressed genes, among which we were able to distinguish between TRa1 target genes in Sertoli cells and those genes whose expression was altered in the germ-cell lineage as a consequence of the Sertoli cell phenotype.
RESULTS A Mouse Model With Altered T3 Signaling in Sertoli Cells In the previously described transgenic mouse model TRaAMI/SC (Fumel et al., 2012), individual mice are heterozygous for the TRaAMI allele of Thra (AMI standing for Activation-function-2 Mutation, Inducible), in which transcription is interrupted by a cassette flanked by two loxP sequences, and Cre recombination is driven by the promoter of the Amh gene. Cre-mediated recombination in Sertoli cells eliminates the stop-codon cassette, triggering the expression of the TRa1L400R receptor, a dominant-negative variant, in heterozygous cells that efficiently prevents TRa1mediated response and generally produces a phenotype analogous to T3 deficiency (Quignodon et al., 2007). Circulating testosterone (Fumel et al., 2012) and luteinizing hormone levels (unpublished data) are not changed in TRaAMI/SC mice. Cell-cycle exit and differentiation of Sertoli cells, however, are delayed. At post-natal Day 3 (P3), no histological defect is visible and the number of Sertoli cells has not yet changed in TRaAMI/SC mice (Fig. S1).
Differential Gene Expression Analysis in the TraAMI/SC Testis We reasoned that differential gene expression analysis performed at P3 would identify primary defects in Sertoli cells and allow us to gain insight into the molecular mechanisms underlying the delayed Sertoli cells differentiation. We prepared polyadenylated RNA from whole testes from two mutant and two control littermates. These samples were submitted for RNA-seq to obtain a genome-wide view of gene expression. Using the DESeq2 software, we found 1,710 differentially expressed genes compared to 10,305, the number of genes whose expression level exceeds 10 reads per millions in all samples; false discovery rate
Direct and Indirect Consequences on Gene Expression of a Thyroid Hormone Receptor Alpha 1 Mutation Restricted to Sertoli Cells 3 FABRICE CHATONNET,1 GABRIEL LIVERA,2 BETTY FUMEL,3 SOPHIE FOUCHECOURT ,
AND
ERIC FRED FLAMANT1*
1
nomique Fonctionnelle de Lyon, Universite de Lyon, Universite Lyon 1, INRA, CNRS, Cedex, France Institut de Ge , Fontenay-Aux-Roses, France INSERM U967, CEA/DSV/iRCM/SCSR/LDG, Univ Paris Diderot, Sorbonne Paris Cite 3 de Tours, IFCE, Physiologie de la Reproduction et des Comportements, Nouzilly, France INRA UMR85, CNRS, Universite 2
SUMMARY Thyroid hormone is required for the timely transition of Sertoli cells from proliferative to differentiating and maturing. This transition takes place during a critical developmental period in mammals, which in mice is the first post-natal week. In order to identify the underlying molecular mechanisms of this differentiation process, we used Cre/loxP technology to selectively block the function of the thyroid hormone receptor TRa1 in Sertoli cells. We then used RNA-seq to analyze the changes in gene expression induced in the post-natal testis. This differential analysis provides genetic clues to the initial testicular defects resulting from disrupted thyroid hormone signaling, and suggests that Sertoli cells influence germ cells soon after their birth.
Corresponding author: nomique Fonctionnelle Institut de Ge de Lyon de Lyon, INRA, CNRS Universite rieure de Lyon Ecole Normale Supe e d’Italie 46 alle 69364, Lyon Cedex 07, France. E-mail: [email protected]
court and Fre de ric Flamant Sophie Fouche contributed equally to the work.
Mol. Reprod. Dev. 2014. ß 2014 Wiley Periodicals, Inc. Received 3 March 2014; Accepted 22 October 2014
INTRODUCTION Sertoli cells of the adult testis provide the physical environment and factors required for germ cells to differentiate into spermatozoa. The number of Sertoli cells is fixed in the adult as their proliferation is arrested, and they rarely die (Sharpe et al., 2003). The extent of Sertoli cell proliferation, which takes place mainly in the perinatal period in rodents, defines the number of germ cells that can be supported as well as the total gamete output after puberty. The timing of Sertoli cell post-natal differentiation is thus a critical parameter of male fertility, yet little is
ß 2014 WILEY PERIODICALS, INC.
Published online in Wiley Online Library (wileyonlinelibrary.com). DOI 10.1002/mrd.22437
known about the gene regulatory events underlying this process. Thyroid hormones including 3,30 ,5-triiodothyronine (T3) and thyroxine (T4), the low-activity precursor of T3 exert a pleiotropic influence on development and cell differentiation in all vertebrates. T3 acts mainly by binding to members of the
Abbreviations: Amh, anti-Müllerian hormone; P#, post-natal day #; T3, triiodo-thyronine; TR, thyroid hormone receptor.
Molecular Reproduction & Development
nuclear-receptor superfamily. Two mammalian genes, Thra and Thrb, encode the three receptors TRa1, TRb1, and TRb2 (collectively called thyroid receptors [TRs]). TRs are present on most if not all cell types, functioning as ligandactivated transcription factors that directly regulate the transcription of a large number of target genes (Yen et al., 2006). They do so by binding to DNA response elements, which are mainly related to the DR4 consensus element (50 AGGTCANNNNAGGTCA) that is recognized by a TR/RXR heterodimer and is found in regulatory regions of their target genes (Chatonnet et al., 2013). Although the physiological relevance of non-genomic pathways is still debated, reports indicate that TR-encoded products can also participate in signaling at the plasma membrane (Kalyanaraman et al., 2014; Martin et al., 2014). In rodents, a peak in circulating T3 levels occurs during the post-natal period. A transient T3 deficiency at this stage prolongs the Sertoli cell proliferation period and has irreversible consequences on testis development, including an increase in testis weight and daily sperm production in adults (Cooke and Meisami., 1991; Kirby et al., 1992). Reciprocally, excess T3 accelerates Sertoli cell differentiation, resulting in reduced testis weight (van Haaster et al., 1993). Yet, transient variations in T3 concentration have a broad effect on pituitary hormone production, which can also indirectly alter testis development. Therefore, experiments that globally modulate T3 levels do not clarify how Sertoli cells directly respond to T3 stimulation during testis development (Wagner et al., 2008). Genetic experiments using mice indicate that an early effect of T3 on testis development is to influence Sertoli cell differentiation. Like hypothyroid rats, mice carrying a knockout allele of Thra, but not Thrb, have more Sertoli cells and heavier testes (Holsberger et al., 2005). We recently used Cre/loxP recombination to eliminate TRa1 specifically in Sertoli cells (Fumel et al., 2012). This was achieved by crossing mice carrying the TRaAMI allele of Thra (Quignodon et al., 2007) with mice carrying the Amh:Cre transgene, which drives the expression of the Cre recombinase in Sertoli cells using the anti-Müllerian hormone (Amh) promoter (Lecureuil et al., 2002). These mice, called TRaAMI/SC, express the dominant-negative TRa1L400R only in Sertoli cells, and display a testicular phenotype that is very similar to the one reported in hypothyroid or Thra-knockout mice. These conditional mice unambiguously establish the cell-autonomous effects of T3 on Sertoli cells and indicate that the adult testicular phenotype resulting from early transient hypothyroidism is a long-term consequence of a defect in the T3-dependent Sertoli cell differentiation. Thus, while early T3 deficiency broadly influences many cell types along the entire gonadotrophic axis, the ability of T3 to specifically stimulate Sertoli cell differentiation is a key element of its impact on testis development. The fact that TRa1 directly influences cell-cycle exit and differentiation of Sertoli cells offers an opportunity to access the genetic programs that control these cellular processes. We performed a comparative RNA-seq analysis on whole testis of TRaAMI/SC mice at the post-natal stage to gain further insight into the repertoire of TRa1 target genes as
2
CHATONNET
ET AL.
well as the molecular mechanisms underlying the control of Sertoli cell differentiation by T3. This broad survey identified 203 differentially expressed genes, among which we were able to distinguish between TRa1 target genes in Sertoli cells and those genes whose expression was altered in the germ-cell lineage as a consequence of the Sertoli cell phenotype.
RESULTS A Mouse Model With Altered T3 Signaling in Sertoli Cells In the previously described transgenic mouse model TRaAMI/SC (Fumel et al., 2012), individual mice are heterozygous for the TRaAMI allele of Thra (AMI standing for Activation-function-2 Mutation, Inducible), in which transcription is interrupted by a cassette flanked by two loxP sequences, and Cre recombination is driven by the promoter of the Amh gene. Cre-mediated recombination in Sertoli cells eliminates the stop-codon cassette, triggering the expression of the TRa1L400R receptor, a dominant-negative variant, in heterozygous cells that efficiently prevents TRa1mediated response and generally produces a phenotype analogous to T3 deficiency (Quignodon et al., 2007). Circulating testosterone (Fumel et al., 2012) and luteinizing hormone levels (unpublished data) are not changed in TRaAMI/SC mice. Cell-cycle exit and differentiation of Sertoli cells, however, are delayed. At post-natal Day 3 (P3), no histological defect is visible and the number of Sertoli cells has not yet changed in TRaAMI/SC mice (Fig. S1).
Differential Gene Expression Analysis in the TraAMI/SC Testis We reasoned that differential gene expression analysis performed at P3 would identify primary defects in Sertoli cells and allow us to gain insight into the molecular mechanisms underlying the delayed Sertoli cells differentiation. We prepared polyadenylated RNA from whole testes from two mutant and two control littermates. These samples were submitted for RNA-seq to obtain a genome-wide view of gene expression. Using the DESeq2 software, we found 1,710 differentially expressed genes compared to 10,305, the number of genes whose expression level exceeds 10 reads per millions in all samples; false discovery rate
