- Email: [email protected]
Tissue-specific knockouts of steroidogenic factor 1
Molecular and Cellular Endocrinology 215 (2004) 89–94
Review
Tissue-specific knockouts of steroidogenic factor 1 Liping Zhao, Marit Bakke, Neil A. Hanley, Gregor Majdic, Nancy R. Stallings, Pancharatnam Jeyasuria, Keith L. Parker∗ Department of Internal Medicine, Division of Endocrinology, University of Texas Southwestern Medical Center, 5323 Harry Hines Blvd., Dallas, TX 75390-8857, USA
Abstract Targeted gene disruption has produced knockout (KO) mice globally deficient in the orphan nuclear receptor steroidogenic factor 1 (SF-1). These SF-1 KO mice lacked adrenal glands and gonads, and also had impaired expression of gonadotropins in pituitary gonadotropes and marked structural abnormalities of the ventromedial hypothalamic nucleus (VMH). To define the roles of SF-1 within discrete sites of the hypothalamic-pituitary-steroidogenic organ axis, we have sought to make tissue-specific SF-1 KO mice (as reviewed here). We first used adrenal transplants to restore adrenal function in global SF-1 KO mice, providing a physiological form of a “VMH-specific” KO to study the roles of SF-1 in weight regulation. These adrenal-transplanted SF-1 KO mice became obese due to decreased locomotor activity, providing a novel model of hypothalamic obesity. Mice with a pituitary-specific KO of SF-1 mediated by the Cre-loxP recombination strategy exhibited hypogonadotropic hypogonadism, revealing essential roles of SF-1 in pituitary function in vivo. Ongoing studies seek to inactivate SF-1 in the brain or specific gonadal cell types, thereby defining its roles in development and function at these sites. In addition, we review our use of bacterial artificial chromosome transgenesis to develop a fluorescent marker for cells that express SF-1. © 2003 Elsevier Ireland Ltd. All rights reserved. Keywords: Tissue-specific knockouts; Cre recombinase; Steroidogenesis; Green fluorescent protein
1. Introduction Steroidogenic factor 1 (SF-1) (officially designated NR5A1) was first identified and characterized as a key mediator of the expression of the cytochrome P450 steroid hydroxylases in transfection studies in steroidogenic cell lines. Following the cloning of a SF-1 cDNA (Lala et al., 1992; Honda et al., 1993), its sequence revealed that SF-1 is a member of the nuclear hormone receptor family of transcriptional regulators. Many laboratories subsequently showed that SF-1 acts at multiple levels of the hypothalamic-pituitary-steroidogenic organ axis to regulate the expression of genes that are important for regulated steroidogenesis. Studies with mice globally deficient in SF-1 due to homozygosity for a null allele of SF-1 (i.e., SF-1 KO mice) confirmed its essential roles in endocrine development (reviewed in Morohashi and Omura, 1996; Parker and Schimmer, 1997; Sadovsky and Crawford, 1998). Although ∗ Corresponding author. Tel.: +1-214-648-5027; fax: +1-214-648-5044. E-mail address: [email protected] (K.L. Parker).
initial stages of adrenogonadal development progressed in the absence of SF-1 such that the indifferent gonad could be visualized at E10.5 and persisted until ∼E11.5–E12, newborn SF-1 KO mice lacked adrenal glands and gonads (Fig. 1). Because their gonads regressed before hormones that mediate male sexual differentiation are produced at ∼E12.5, the internal and external urogenital tracts of SF-1 KO mice were female irrespective of genetic sex. SF-1 KO mice also had impaired gonadotrope expression of a number of genes that regulate gonadal steroidogenesis, including the common ␣ subunit of glycoprotein hormones, the beta subunit of luteinizing hormone (LH), the beta subunit of follicle-stimulating hormone (FSH), and the receptor for gonadotropin-releasing hormone. However, treatment of SF-1 KO mice with exogenous gonadotropins induced some expression of gonadotropins in the region of the pituitary where the gonadotropes normally reside, suggesting that gonadotropes or their precursors persist in SF-1 KO mice (Ikeda et al., 1995). Moreover, they also had marked structural abnormalities of the ventromedial hypothalamic nucleus (VMH), a region of the mediobasal hypothalamic linked to feeding and appetite regulation and female reproductive behavior (Ikeda et al., 1995). Finally, although
0303-7207/$ – see front matter © 2003 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.mce.2003.11.009
90
L. Zhao et al. / Molecular and Cellular Endocrinology 215 (2004) 89–94
Fig. 1. Adrenal and gonadal agenesis in SF-1 KO mice. Shown are the dissected genito-urinary tracts from wild-type and SF-1 KO mice. Reprinted with permission from Luo et al. (1994). (A) SF-1 KO female; (B) wild-type female; (C) SF-1 KO male; (D) wild-type male. (a) Adrenal; (k) kidney; (od) oviduct; (o) ovary; (t) testis; (e) epididymis.
the functional consequences remain undefined, the SF-1 KO mice had defects in their splenic parenchyma (Morohashi et al., 1999).
2. Results and discussion
weighing almost twice as much as WT mice (50 g SF-1 KO versus 28 g WT)—and had markedly increased adiposity. In contrast to many other genetic models of hypothalamic obesity, the mice did not eat in excess, but rather had decreased spontaneous locomotor activity. These findings implicate SF-1 and the VMH in metabolic processes distinct from appetite control.
2.1. Adrenal transplants to restore adrenal function show that SF-1 KO mice exhibit late-onset obesity
2.2. Tissue-specific KO of SF-1
Although previous studies have linked the VMH to weight regulation, the specific role of SF-1 in these processes remains undefined. To allow us to study appetite and weight regulation in global SF-1 KO mice without confounding effects of glucocorticoid therapy, we used adrenal transplants to restore their adrenal function (Majdic et al., 2002). These mice were viable and had serum corticosterone levels that were indistinguishable from wild-type (WT) mice. Beginning at approximately 10 weeks of age, they became significantly heavier than the WT littermates—eventually
Analyses of adrenal-transplanted SF-1 KO mice suggested that SF-1 plays important roles in weight regulation. However, the complex phenotype of SF-1 KO mice complicated efforts to distinguish primary versus secondary effects at various sites of the hypothalamic-pituitary-steroidogenic organ axis. To define specific roles of SF-1 at different sites, we have used the Cre-loxP recombination strategy to make tissue-specific SF-1 KO mice (Zhao et al., 2001). In the first step of this approach (Fig. 2), we modified the locus encoding SF-1 to place recognition sites for
L. Zhao et al. / Molecular and Cellular Endocrinology 215 (2004) 89–94
91
Fig. 2. Strategy for tissue-specific knockout of SF-1. The strategy for placing loxP sites around the last exon encoding the SF-1 AFA-2 motif is shown as modified from Zhao et al., 2001. Following tissue-specific Cre expression, the last exon is deleted to yield the recombined locus. When coupled with tissue-specific transgenes driving Cre expression, this approach generates tissue-specific knockouts of SF-1. N: NcoI restriction endonuclease site; K: KpnI; X: XhoI; H: HindIII. The solid line under the recombined allele shows a probe used in Southern blot hybridization to identify correctly targeted clones.
Cre recombinase (termed loxP sites for “locus of crossing over”) around the last exon that encodes the AF-2 transcriptional activation motif and the 3 -transcription termination-polyadenylation signals, generating an allele termed “floxed” for “flanked by loxP sites” (F). The tissue-specific Cre transgene contained the regulatory sequences of the ␣ subunit of glycoprotein hormones (␣GSU-Cre), and thus targeted Cre expression to pituitary gonadotropes and thyrotropes (Cushman et al., 2000). Genetic crosses were used to produce mice with one conditional floxed SF-1 allele and one allele that had already undergone germline recombination to generate a non-conditional, null allele (designated R). To minimize effects of ectopic Cre expression by the Cre transgene, crosses typically involved one parent with one R SF-1 allele and one WT SF-1 allele (i.e., SF-1 R/+) and the ␣GSU-Cre transgene, and a second parent homozygous for the F SF-1 allele (SF-1 F/F) but lacking the Cre transgene. As controls, we also examined mice that carried the F/R SF-1 alleles without the Cre transgene and mice that carried the Cre transgene with at least one wild-type SF-1 allele. To assess the effect of this targeted genetic modification on SF-1 expression in various tissues, we used an antiserum against full-length SF-1 in immunohistochemical analyses. Mice carrying the F/R SF-1 alleles and the ␣GSU-Cre transgene had normal SF-1 immunoreactivity in the adrenal cortex, testes, ovaries, and VMH. In contrast, SF-1 immunoreactivity was reduced to background levels in the anterior pituitary gland—the intended site of tissue-specific knockout. Both male and female F/R, ␣GSU-Cre mice produced by the crosses described above were sterile and
showed no signs of secondary sexual development. In addition, sex steroid-dependent tissues (e.g., seminal vesicles and prostate in males and uterus in females) were markedly hypoplastic. These findings suggest that their synthesis of sex steroids is severely impaired. To determine the basis for this reproductive phenotype, we examined the histology of SF-1-expressing tissues in pituitary-specific SF-1 KO mice. The adrenal glands, VMH, and anterior pituitary of pituitary-specific SF-1 KO mice appeared grossly intact (data not shown). In contrast, the gonads were markedly hypoplastic. The testes contained immature seminiferous tubules with Sertoli cells and immature spermatogonia, indicating that testis determination, germ cell migration, and sex cord formation occurred normally (Fig. 3A). However, germ cells were severely decreased in number and mature sperm were not seen. In the interstitial region, Leydig cells were drastically decreased in number and showed none of the histological features of steroidogenic cells. Although primary, secondary, and antral follicles were present in the ovaries, preovulatory follicles and corpora lutea were not seen (Fig. 3C). Coupled with the hypoplastic uteri, these findings suggest strongly that sex steroid production is severely impaired in pituitary-specific SF-1 KO mice. The expression of FSH and LH in the F/R, ␣GSU-Cre mice was diminished considerably, while levels of ACTH, TSH, growth hormone, and prolactin all were normal (data not shown). The normal expression of TSH is particularly noteworthy, because it demonstrates a specific defect in gonadotropes rather than a non-specific, toxic effect of Cre expression in both gonadotropes and thyrotropes.
92
L. Zhao et al. / Molecular and Cellular Endocrinology 215 (2004) 89–94
Fig. 3. Gonadal hypoplasia in pituitary-specific SF-1 KO mice. (A) Testis; (B) testis after treatment with PMSG; (C) ovary; (D) ovary after treatment with PMSG modified from (Zhao et al., 2001).
These results suggest that the impaired reproduction in the pituitary-specific SF-1 KO mice results from decreased transcription of multiple SF-1 target genes in the gonadotropes. However, because we did not show that the pituitary-specific SF-1 KO mice can respond to exogenous gonadotropins, as was done with the global SF-1 KO mice (Ikeda et al., 1995), we cannot state whether the gonadotropes or their precursors persist in the anterior pituitaries of pituitary-specific SF-1 KO mice.
To prove that the gonads were functional, we administered exogenous pregnant mares’ serum gonadotropins (PMSG) to the pituitary-specific SF-1 KO mice. In males, treatment for 5 days induced sperm maturation associated with luminal opening of the seminiferous tubules and hyperplasia of the Leydig cells (Fig. 3B). In females, PMSG induced both follicular maturation and the formation of corpora lutea (Fig. 3D). These studies establish that the gonads of pituitary-specific SF-1 KO mice are
Fig. 4. Effect of CNS-specific inactivation of SF-1 on VMH structure. Sections of the mediobasal hypothalami from WT mice and mice carrying the conditional SF-1 allele and the nestin-Cre transgene are shown. Note the absence of a discrete VMH structure in the CNS-specific KO section versus the well-defined structure in the WT section.
L. Zhao et al. / Molecular and Cellular Endocrinology 215 (2004) 89–94
functional if stimulated with gonadotropins (Zhao et al., 2001). 2.2.1. CNS-specific KO of SF-1 The studies with adrenal-transplanted SF-1 KO mice described above argue strongly that the VMH abnormality is associated with a metabolic syndrome, but provide a very cumbersome model for further studies. We therefore sought to generate VMH-specific SF-1 KO mice. We are unaware of any promoter that is expressed specifically within the VMH, but reasoned that pan-neuronal expression should suffice, since SF-1 in the CNS is expressed only in the VMH. In preliminary results, a nestin-Cre transgene coupled with the conditional SF-1 allele has resulted in mice with marked structural abnormalities of the VMH (Fig. 4, associated with virtually complete abrogation of SF-1 immunoreactivity in the hypothalamus and normal levels elsewhere (Zhao, unpublished observation). Moreover, preliminary studies suggest that these CNS-specific SF-1 KO mice have a propensity to increased adiposity and decreased spontaneous locomotor activity. These studies
93
strongly support the utility of these mice in exploring the roles of the VMH, and specifically of SF-1, in metabolic regulation. 2.3. Development of a transgenic marker for SF-1-expressing cells To facilitate analyses of SF-1-expressing cells in various sites, it would be of great value to have a non-invasive lineage marker. To this end, we used BAC transgenesis to place an enhanced green fluorescent protein reporter gene (eGFP) under the control of 50 kb of SF-1 regulatory sequences (Stallings et al., 2002). As shown in Fig. 5, the SF-1/eGFP transgene was expressed in the adrenal cortex, gonads, and VMH, showing that this region of the SF-1 locus contains sufficient information to target gene expression to these sites. Of note, expression was not seen in pituitary, suggesting that expression in gonadotropes requires additional elements not found in this construct. Analyses of mouse embryos indicated that the SF-1/eGFP transgene faithfully recapitulates the ontogeny of SF-1 expression in the gonads, adrenal cortex, and VMH. This transgene therefore should provide a valuable tool to follow the fate of SF-1-expressing cells during development, both in WT and SF-1 KO mice.
Acknowledgements Studies described here were supported by NIH grant DK54480 (KLP). NAH was supported by a Wellcome Trust Clinical Training Fellowship and Marit Bakke was supported by a fellowship from the Swedish Medical Research Council.
References
Fig. 5. The SF-1/eGFP transgene is expressed in the adrenal cortex, testes, ovary, and VMH (Stallings et al., 2002). Tissues from SF-1/eGFP transgenic mice were harvested and analyzed by fluorescence microscopy for eGFP expression. Shown are brightfield views of sections stained with hematoxylin-eosin (left panels) and fluorescent photomicrographs (right panels). (A) Adrenal; (B) testis; (C) ovary; (D) mediobasal hypothalamus.
Cushman, L.J., Burrows, H.L., Seasholtz, A.F., Lewandoski, M., Muzyczka, N., Camper, S.A., 2000. Cre-mediated recombination in the pituitary gland. Genesis 28, 167–174. Honda, S.-I., Morohashi, K.-I., Nomura, M., Takeya, H., Kitajima, M., Omura, T., 1993. Ad4BP regulating steroidogenic P-450 gene is a member of steroid hormone receptor superfamily. J. Biol. Chem. 268, 7494–7502. Ikeda, Y., Luo, X., Abbud, R., Nilson, J.H., Parker, K.L., 1995. The nuclear receptor steroidogenic factor 1 is essential for the formation of the ventromedial hypothalamic nucleus. Mol. Endocrinol. 9, 478– 486. Lala, D.S., Rice, D.A., Parker, K.L., 1992. Steroidogenic factor I, a key regulator of steroidogenic enzyme expression, is the mouse homolog of fushi tarazu-factor I. Mol. Endocrinol. 6, 1278–1287. Luo, X., Ikeda, Y., Parker, K.L., 1994. A cell-specific nuclear receptor is essential for adrenal and gonadal development and sexual differentiation. Cell 77, 481–490. Majdic, G., Young, M.A., Gomez-Sanchez, E., Anderson, P., Szczepaniak, L., Dobbins, R., McGarry, J.D., Parker, K.L., 2002. SF-1 knockout mice are a novel genetic model of hypothalamic obesity. Endocrinology 143, 607–614.
94
L. Zhao et al. / Molecular and Cellular Endocrinology 215 (2004) 89–94
Morohashi, K.-I., Omura, T., 1996. Ad4BP/SF-1, a transcription factor essential for the transcription of steroidogenic cytochrome P450 genes and for the establishment of the reproductive function. FASEB J. 10, 1569–1577. Morohashi, K., Tsuboi-Asai, H., Matsushita, S., Suda, M., Nakashima, M., Sasano, H., Hataba, Y., Li, C.L., Fukata, J., Irie, J., Watanabe, T., Nagura, H., Li, E., 1999. Structural and functional abnormalities in the spleen of an mFtz-F1 gene-disrupted mouse. Blood 93, 1586–1594. Parker, K.L., Schimmer, B.P., 1997. Steroidogenic factor 1: a key mediator of endocrine development and function. Endocr. Rev. 18, 361–377.
Sadovsky, Y., Crawford, P.A., 1998. Developmental and physiologic roles of the nuclear receptor steroidogenic factor-1 in the reproductive system. J. Soc. Gynecol. Invest. 5, 6–12. Stallings, N.R., Hanley, N.A., Bakke, M., Majdic, G., Zhao, L., Parker, K.L., 2002. Development of a green fluorescent protein lineage marker for steroidogenic factor 1. Mol. Endocrinol. 16, 2360– 2370. Zhao, L., Bakke, M., Cushman, L.J., Burrows, H.L., Parlow, A.F., Camper, S.A., Parker, K.L., 2001. Steroidogenic factor 1 is essential for pituitary gonadotrope function. Development 128, 147–154.
Review
Tissue-specific knockouts of steroidogenic factor 1 Liping Zhao, Marit Bakke, Neil A. Hanley, Gregor Majdic, Nancy R. Stallings, Pancharatnam Jeyasuria, Keith L. Parker∗ Department of Internal Medicine, Division of Endocrinology, University of Texas Southwestern Medical Center, 5323 Harry Hines Blvd., Dallas, TX 75390-8857, USA
Abstract Targeted gene disruption has produced knockout (KO) mice globally deficient in the orphan nuclear receptor steroidogenic factor 1 (SF-1). These SF-1 KO mice lacked adrenal glands and gonads, and also had impaired expression of gonadotropins in pituitary gonadotropes and marked structural abnormalities of the ventromedial hypothalamic nucleus (VMH). To define the roles of SF-1 within discrete sites of the hypothalamic-pituitary-steroidogenic organ axis, we have sought to make tissue-specific SF-1 KO mice (as reviewed here). We first used adrenal transplants to restore adrenal function in global SF-1 KO mice, providing a physiological form of a “VMH-specific” KO to study the roles of SF-1 in weight regulation. These adrenal-transplanted SF-1 KO mice became obese due to decreased locomotor activity, providing a novel model of hypothalamic obesity. Mice with a pituitary-specific KO of SF-1 mediated by the Cre-loxP recombination strategy exhibited hypogonadotropic hypogonadism, revealing essential roles of SF-1 in pituitary function in vivo. Ongoing studies seek to inactivate SF-1 in the brain or specific gonadal cell types, thereby defining its roles in development and function at these sites. In addition, we review our use of bacterial artificial chromosome transgenesis to develop a fluorescent marker for cells that express SF-1. © 2003 Elsevier Ireland Ltd. All rights reserved. Keywords: Tissue-specific knockouts; Cre recombinase; Steroidogenesis; Green fluorescent protein
1. Introduction Steroidogenic factor 1 (SF-1) (officially designated NR5A1) was first identified and characterized as a key mediator of the expression of the cytochrome P450 steroid hydroxylases in transfection studies in steroidogenic cell lines. Following the cloning of a SF-1 cDNA (Lala et al., 1992; Honda et al., 1993), its sequence revealed that SF-1 is a member of the nuclear hormone receptor family of transcriptional regulators. Many laboratories subsequently showed that SF-1 acts at multiple levels of the hypothalamic-pituitary-steroidogenic organ axis to regulate the expression of genes that are important for regulated steroidogenesis. Studies with mice globally deficient in SF-1 due to homozygosity for a null allele of SF-1 (i.e., SF-1 KO mice) confirmed its essential roles in endocrine development (reviewed in Morohashi and Omura, 1996; Parker and Schimmer, 1997; Sadovsky and Crawford, 1998). Although ∗ Corresponding author. Tel.: +1-214-648-5027; fax: +1-214-648-5044. E-mail address: [email protected] (K.L. Parker).
initial stages of adrenogonadal development progressed in the absence of SF-1 such that the indifferent gonad could be visualized at E10.5 and persisted until ∼E11.5–E12, newborn SF-1 KO mice lacked adrenal glands and gonads (Fig. 1). Because their gonads regressed before hormones that mediate male sexual differentiation are produced at ∼E12.5, the internal and external urogenital tracts of SF-1 KO mice were female irrespective of genetic sex. SF-1 KO mice also had impaired gonadotrope expression of a number of genes that regulate gonadal steroidogenesis, including the common ␣ subunit of glycoprotein hormones, the beta subunit of luteinizing hormone (LH), the beta subunit of follicle-stimulating hormone (FSH), and the receptor for gonadotropin-releasing hormone. However, treatment of SF-1 KO mice with exogenous gonadotropins induced some expression of gonadotropins in the region of the pituitary where the gonadotropes normally reside, suggesting that gonadotropes or their precursors persist in SF-1 KO mice (Ikeda et al., 1995). Moreover, they also had marked structural abnormalities of the ventromedial hypothalamic nucleus (VMH), a region of the mediobasal hypothalamic linked to feeding and appetite regulation and female reproductive behavior (Ikeda et al., 1995). Finally, although
0303-7207/$ – see front matter © 2003 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.mce.2003.11.009
90
L. Zhao et al. / Molecular and Cellular Endocrinology 215 (2004) 89–94
Fig. 1. Adrenal and gonadal agenesis in SF-1 KO mice. Shown are the dissected genito-urinary tracts from wild-type and SF-1 KO mice. Reprinted with permission from Luo et al. (1994). (A) SF-1 KO female; (B) wild-type female; (C) SF-1 KO male; (D) wild-type male. (a) Adrenal; (k) kidney; (od) oviduct; (o) ovary; (t) testis; (e) epididymis.
the functional consequences remain undefined, the SF-1 KO mice had defects in their splenic parenchyma (Morohashi et al., 1999).
2. Results and discussion
weighing almost twice as much as WT mice (50 g SF-1 KO versus 28 g WT)—and had markedly increased adiposity. In contrast to many other genetic models of hypothalamic obesity, the mice did not eat in excess, but rather had decreased spontaneous locomotor activity. These findings implicate SF-1 and the VMH in metabolic processes distinct from appetite control.
2.1. Adrenal transplants to restore adrenal function show that SF-1 KO mice exhibit late-onset obesity
2.2. Tissue-specific KO of SF-1
Although previous studies have linked the VMH to weight regulation, the specific role of SF-1 in these processes remains undefined. To allow us to study appetite and weight regulation in global SF-1 KO mice without confounding effects of glucocorticoid therapy, we used adrenal transplants to restore their adrenal function (Majdic et al., 2002). These mice were viable and had serum corticosterone levels that were indistinguishable from wild-type (WT) mice. Beginning at approximately 10 weeks of age, they became significantly heavier than the WT littermates—eventually
Analyses of adrenal-transplanted SF-1 KO mice suggested that SF-1 plays important roles in weight regulation. However, the complex phenotype of SF-1 KO mice complicated efforts to distinguish primary versus secondary effects at various sites of the hypothalamic-pituitary-steroidogenic organ axis. To define specific roles of SF-1 at different sites, we have used the Cre-loxP recombination strategy to make tissue-specific SF-1 KO mice (Zhao et al., 2001). In the first step of this approach (Fig. 2), we modified the locus encoding SF-1 to place recognition sites for
L. Zhao et al. / Molecular and Cellular Endocrinology 215 (2004) 89–94
91
Fig. 2. Strategy for tissue-specific knockout of SF-1. The strategy for placing loxP sites around the last exon encoding the SF-1 AFA-2 motif is shown as modified from Zhao et al., 2001. Following tissue-specific Cre expression, the last exon is deleted to yield the recombined locus. When coupled with tissue-specific transgenes driving Cre expression, this approach generates tissue-specific knockouts of SF-1. N: NcoI restriction endonuclease site; K: KpnI; X: XhoI; H: HindIII. The solid line under the recombined allele shows a probe used in Southern blot hybridization to identify correctly targeted clones.
Cre recombinase (termed loxP sites for “locus of crossing over”) around the last exon that encodes the AF-2 transcriptional activation motif and the 3 -transcription termination-polyadenylation signals, generating an allele termed “floxed” for “flanked by loxP sites” (F). The tissue-specific Cre transgene contained the regulatory sequences of the ␣ subunit of glycoprotein hormones (␣GSU-Cre), and thus targeted Cre expression to pituitary gonadotropes and thyrotropes (Cushman et al., 2000). Genetic crosses were used to produce mice with one conditional floxed SF-1 allele and one allele that had already undergone germline recombination to generate a non-conditional, null allele (designated R). To minimize effects of ectopic Cre expression by the Cre transgene, crosses typically involved one parent with one R SF-1 allele and one WT SF-1 allele (i.e., SF-1 R/+) and the ␣GSU-Cre transgene, and a second parent homozygous for the F SF-1 allele (SF-1 F/F) but lacking the Cre transgene. As controls, we also examined mice that carried the F/R SF-1 alleles without the Cre transgene and mice that carried the Cre transgene with at least one wild-type SF-1 allele. To assess the effect of this targeted genetic modification on SF-1 expression in various tissues, we used an antiserum against full-length SF-1 in immunohistochemical analyses. Mice carrying the F/R SF-1 alleles and the ␣GSU-Cre transgene had normal SF-1 immunoreactivity in the adrenal cortex, testes, ovaries, and VMH. In contrast, SF-1 immunoreactivity was reduced to background levels in the anterior pituitary gland—the intended site of tissue-specific knockout. Both male and female F/R, ␣GSU-Cre mice produced by the crosses described above were sterile and
showed no signs of secondary sexual development. In addition, sex steroid-dependent tissues (e.g., seminal vesicles and prostate in males and uterus in females) were markedly hypoplastic. These findings suggest that their synthesis of sex steroids is severely impaired. To determine the basis for this reproductive phenotype, we examined the histology of SF-1-expressing tissues in pituitary-specific SF-1 KO mice. The adrenal glands, VMH, and anterior pituitary of pituitary-specific SF-1 KO mice appeared grossly intact (data not shown). In contrast, the gonads were markedly hypoplastic. The testes contained immature seminiferous tubules with Sertoli cells and immature spermatogonia, indicating that testis determination, germ cell migration, and sex cord formation occurred normally (Fig. 3A). However, germ cells were severely decreased in number and mature sperm were not seen. In the interstitial region, Leydig cells were drastically decreased in number and showed none of the histological features of steroidogenic cells. Although primary, secondary, and antral follicles were present in the ovaries, preovulatory follicles and corpora lutea were not seen (Fig. 3C). Coupled with the hypoplastic uteri, these findings suggest strongly that sex steroid production is severely impaired in pituitary-specific SF-1 KO mice. The expression of FSH and LH in the F/R, ␣GSU-Cre mice was diminished considerably, while levels of ACTH, TSH, growth hormone, and prolactin all were normal (data not shown). The normal expression of TSH is particularly noteworthy, because it demonstrates a specific defect in gonadotropes rather than a non-specific, toxic effect of Cre expression in both gonadotropes and thyrotropes.
92
L. Zhao et al. / Molecular and Cellular Endocrinology 215 (2004) 89–94
Fig. 3. Gonadal hypoplasia in pituitary-specific SF-1 KO mice. (A) Testis; (B) testis after treatment with PMSG; (C) ovary; (D) ovary after treatment with PMSG modified from (Zhao et al., 2001).
These results suggest that the impaired reproduction in the pituitary-specific SF-1 KO mice results from decreased transcription of multiple SF-1 target genes in the gonadotropes. However, because we did not show that the pituitary-specific SF-1 KO mice can respond to exogenous gonadotropins, as was done with the global SF-1 KO mice (Ikeda et al., 1995), we cannot state whether the gonadotropes or their precursors persist in the anterior pituitaries of pituitary-specific SF-1 KO mice.
To prove that the gonads were functional, we administered exogenous pregnant mares’ serum gonadotropins (PMSG) to the pituitary-specific SF-1 KO mice. In males, treatment for 5 days induced sperm maturation associated with luminal opening of the seminiferous tubules and hyperplasia of the Leydig cells (Fig. 3B). In females, PMSG induced both follicular maturation and the formation of corpora lutea (Fig. 3D). These studies establish that the gonads of pituitary-specific SF-1 KO mice are
Fig. 4. Effect of CNS-specific inactivation of SF-1 on VMH structure. Sections of the mediobasal hypothalami from WT mice and mice carrying the conditional SF-1 allele and the nestin-Cre transgene are shown. Note the absence of a discrete VMH structure in the CNS-specific KO section versus the well-defined structure in the WT section.
L. Zhao et al. / Molecular and Cellular Endocrinology 215 (2004) 89–94
functional if stimulated with gonadotropins (Zhao et al., 2001). 2.2.1. CNS-specific KO of SF-1 The studies with adrenal-transplanted SF-1 KO mice described above argue strongly that the VMH abnormality is associated with a metabolic syndrome, but provide a very cumbersome model for further studies. We therefore sought to generate VMH-specific SF-1 KO mice. We are unaware of any promoter that is expressed specifically within the VMH, but reasoned that pan-neuronal expression should suffice, since SF-1 in the CNS is expressed only in the VMH. In preliminary results, a nestin-Cre transgene coupled with the conditional SF-1 allele has resulted in mice with marked structural abnormalities of the VMH (Fig. 4, associated with virtually complete abrogation of SF-1 immunoreactivity in the hypothalamus and normal levels elsewhere (Zhao, unpublished observation). Moreover, preliminary studies suggest that these CNS-specific SF-1 KO mice have a propensity to increased adiposity and decreased spontaneous locomotor activity. These studies
93
strongly support the utility of these mice in exploring the roles of the VMH, and specifically of SF-1, in metabolic regulation. 2.3. Development of a transgenic marker for SF-1-expressing cells To facilitate analyses of SF-1-expressing cells in various sites, it would be of great value to have a non-invasive lineage marker. To this end, we used BAC transgenesis to place an enhanced green fluorescent protein reporter gene (eGFP) under the control of 50 kb of SF-1 regulatory sequences (Stallings et al., 2002). As shown in Fig. 5, the SF-1/eGFP transgene was expressed in the adrenal cortex, gonads, and VMH, showing that this region of the SF-1 locus contains sufficient information to target gene expression to these sites. Of note, expression was not seen in pituitary, suggesting that expression in gonadotropes requires additional elements not found in this construct. Analyses of mouse embryos indicated that the SF-1/eGFP transgene faithfully recapitulates the ontogeny of SF-1 expression in the gonads, adrenal cortex, and VMH. This transgene therefore should provide a valuable tool to follow the fate of SF-1-expressing cells during development, both in WT and SF-1 KO mice.
Acknowledgements Studies described here were supported by NIH grant DK54480 (KLP). NAH was supported by a Wellcome Trust Clinical Training Fellowship and Marit Bakke was supported by a fellowship from the Swedish Medical Research Council.
References
Fig. 5. The SF-1/eGFP transgene is expressed in the adrenal cortex, testes, ovary, and VMH (Stallings et al., 2002). Tissues from SF-1/eGFP transgenic mice were harvested and analyzed by fluorescence microscopy for eGFP expression. Shown are brightfield views of sections stained with hematoxylin-eosin (left panels) and fluorescent photomicrographs (right panels). (A) Adrenal; (B) testis; (C) ovary; (D) mediobasal hypothalamus.
Cushman, L.J., Burrows, H.L., Seasholtz, A.F., Lewandoski, M., Muzyczka, N., Camper, S.A., 2000. Cre-mediated recombination in the pituitary gland. Genesis 28, 167–174. Honda, S.-I., Morohashi, K.-I., Nomura, M., Takeya, H., Kitajima, M., Omura, T., 1993. Ad4BP regulating steroidogenic P-450 gene is a member of steroid hormone receptor superfamily. J. Biol. Chem. 268, 7494–7502. Ikeda, Y., Luo, X., Abbud, R., Nilson, J.H., Parker, K.L., 1995. The nuclear receptor steroidogenic factor 1 is essential for the formation of the ventromedial hypothalamic nucleus. Mol. Endocrinol. 9, 478– 486. Lala, D.S., Rice, D.A., Parker, K.L., 1992. Steroidogenic factor I, a key regulator of steroidogenic enzyme expression, is the mouse homolog of fushi tarazu-factor I. Mol. Endocrinol. 6, 1278–1287. Luo, X., Ikeda, Y., Parker, K.L., 1994. A cell-specific nuclear receptor is essential for adrenal and gonadal development and sexual differentiation. Cell 77, 481–490. Majdic, G., Young, M.A., Gomez-Sanchez, E., Anderson, P., Szczepaniak, L., Dobbins, R., McGarry, J.D., Parker, K.L., 2002. SF-1 knockout mice are a novel genetic model of hypothalamic obesity. Endocrinology 143, 607–614.
94
L. Zhao et al. / Molecular and Cellular Endocrinology 215 (2004) 89–94
Morohashi, K.-I., Omura, T., 1996. Ad4BP/SF-1, a transcription factor essential for the transcription of steroidogenic cytochrome P450 genes and for the establishment of the reproductive function. FASEB J. 10, 1569–1577. Morohashi, K., Tsuboi-Asai, H., Matsushita, S., Suda, M., Nakashima, M., Sasano, H., Hataba, Y., Li, C.L., Fukata, J., Irie, J., Watanabe, T., Nagura, H., Li, E., 1999. Structural and functional abnormalities in the spleen of an mFtz-F1 gene-disrupted mouse. Blood 93, 1586–1594. Parker, K.L., Schimmer, B.P., 1997. Steroidogenic factor 1: a key mediator of endocrine development and function. Endocr. Rev. 18, 361–377.
Sadovsky, Y., Crawford, P.A., 1998. Developmental and physiologic roles of the nuclear receptor steroidogenic factor-1 in the reproductive system. J. Soc. Gynecol. Invest. 5, 6–12. Stallings, N.R., Hanley, N.A., Bakke, M., Majdic, G., Zhao, L., Parker, K.L., 2002. Development of a green fluorescent protein lineage marker for steroidogenic factor 1. Mol. Endocrinol. 16, 2360– 2370. Zhao, L., Bakke, M., Cushman, L.J., Burrows, H.L., Parlow, A.F., Camper, S.A., Parker, K.L., 2001. Steroidogenic factor 1 is essential for pituitary gonadotrope function. Development 128, 147–154.