Abstract
Steroidogenic acute regulatory protein (StAR) is essential for adrenal and gonadal steroidogenesis, stimulating the translocation of cholesterol to the inner mitochondrial membrane where steroidogenesis commences. StAR mutations in humans cause congenital lipoid adrenal hyperplasia (lipoid CAH), an autosomal recessive condition with severe deficiencies of all classes of steroid hormones. We previously described StAR knockout mice that mimic many features of lipoid CAH patients. By keeping StAR knockout mice alive with corticosteroid replacement, we now examine the temporal effects of StAR deficiency on the structure and function of steroidogenic tissues. The adrenal glands, affected most severely at birth, exhibited progressive increases in lipid deposits with aging. The testes of newborn StAR knockout mice contained scattered lipid deposits in the interstitial region, presumably in remnants of fetal Leydig cells. By 8 weeks of age, the interstitial lipid deposits worsened considerably and were associated with Leydig cell hyperplasia. Despite these changes, germ cells in the seminiferous tubules appeared intact histologically, suggesting that the StAR knockout mice retained some capacity for androgen biosynthesis. Sperm maturation was delayed, and the germ cells exhibited histological features of apoptosis, consistent with suboptimal androgen production. Immediately after birth, the ovaries of StAR knockout mice appeared normal. After the time of normal puberty, however, prominent lipid deposits accumulated in the interstitial region, accompanied by marked luteinization of stromal cells and incomplete follicular maturation that ultimately culminated in premature ovarian failure. These studies provide the first systematic evaluation of the developmental consequences of StAR deficiency in the various steroidogenic organs.
INTRODUCTION
A critical component of regulated steroidogenesis is the induction of steroid biosynthesis by pituitary trophic hormones. Temporally, this hormonal induction is divided into two phases: acute effects that reflect increased mobilization and delivery of cholesterol precursor to the inner mitochondrial membrane, and chronic effects that result from increased transcription of genes that encode key components of the steroidogenic complex (reviewed in Ref. 1). A mitochondrial phosphoprotein, designated the steroidogenic acute regulatory protein (StAR), was identified and proposed to play key roles in the acute induction of steroidogenesis (2). Compelling evidence for StAR’s essential role in steroidogenesis came from analyses of patients with congenital lipoid adrenal hyperplasia (lipoid CAH), an autosomal recessive disorder characterized by severely impaired adrenal and gonadal steroidogenesis coupled with characteristic lipid deposits in the steroidogenic tissues (3, 4).
To further our understanding of the pathogenesis of lipoid CAH, we generated StAR knockout mice (5). These mice established essential roles of StAR in regulated steroidogenesis in mice and demonstrated a spectrum of severity of lipid deposits in the primary steroidogenic cells. In newborn StAR knockout mice, the adrenal glands lacked their normal cellular architecture and had abundant lipid deposits, presumably reflecting the fact that the mouse adrenal cortex normally produces steroids in utero. In contrast, the testes contained only scattered lipid deposits, while the ovaries appeared completely normal. Based on a similar hierarchy in the impairment of steroid production in patients with lipoid CAH, Bose et al. (4) proposed a two-hit model of lipoid CAH. According to this model, lipoid CAH patients initially retain some capacity for StAR-independent steroidogenesis; thereafter, progressive lipid accumulation in steroidogenic cells, driven at least partly by trophic hormone stimulation, kills the cells and completely abrogates steroidogenic capacity.
In this report, we used corticosteroid replacement to keep StAR knockout mice alive for differing periods of time after birth, thereby allowing us to assess the temporal effects of StAR deficiency. Our results, which demonstrate progressive pathological changes in the gonads after the time of normal sexual maturation, strongly support the two-hit model of lipoid CAH.
RESULTS
Steroid Hormone Levels in StAR Knockout Mice
Previous samples for analyses of hormone levels in newborn StAR knockout mice potentially contained maternally derived steroids. By rescuing StAR knockout mice with corticosteroid replacement as described in Materials and Methods, we were able to examine the effect of StAR deficiency on circulating steroid hormones later in life. As shown in Table 1, plasma corticosterone, the major circulating glucocorticoid in mice, was markedly decreased in StAR knockout mice at 8 weeks of age, although levels still exceeded the lower limit of assay sensitivity. This finding suggests that the adrenal cortex retained some residual capacity for corticosteroid production, even at 8 weeks. Although markedly decreased, plasma testosterone in males also was detectable. In females, progesterone was decreased and ovarian morphology was markedly abnormal (see below), Surprisingly, estradiol levels in StAR knockout females did not differ significantly from wild-type values. Given the low levels of circulating estradiol, it is unclear whether this finding reflects unimpeded production of ovarian steroids or limitations in the assay that mask defects in ovarian steroidogenesis. As discussed below, the uterus and oviducts of StAR knockout females were markedly hypoplastic, suggesting that estrogen action on target tissues is impaired despite the “normal” circulating levels.
Steroid Hormone Levels in StAR Knockout Mice
| Estradiol (pg/ml) | Progesterone (pg/ml) | Testosterone (ng/ml) | Corticosterone (ng/ml) | |||||
|---|---|---|---|---|---|---|---|---|
| Female | ||||||||
| WT (n = 5) | 10.2 ± 1.9 | ⌝ | 497 ± 177 | ⌝ | ND | 305 ± 77 | ⌝ | |
| NS | a | a | ||||||
| KO (n = 5) | 16.2 ± 3.8 | ⌟ | 39 ± 8 | ⌟ | ND | 9 ± 2 | ⌟ | |
| Male | ||||||||
| WT (n = 5) | ND | 290 ± 70 | ⌝ | 8.1 ± 3.2 | ⌝ | 103 ± 32 | ⌝ | |
| a | a | a | ||||||
| KO (n = 5) | ND | 37 ± 6 | ⌟ | 0.8 ± 0.1 | ⌟ | 17 ± 2 | ⌟ | |
| Estradiol (pg/ml) | Progesterone (pg/ml) | Testosterone (ng/ml) | Corticosterone (ng/ml) | |||||
|---|---|---|---|---|---|---|---|---|
| Female | ||||||||
| WT (n = 5) | 10.2 ± 1.9 | ⌝ | 497 ± 177 | ⌝ | ND | 305 ± 77 | ⌝ | |
| NS | a | a | ||||||
| KO (n = 5) | 16.2 ± 3.8 | ⌟ | 39 ± 8 | ⌟ | ND | 9 ± 2 | ⌟ | |
| Male | ||||||||
| WT (n = 5) | ND | 290 ± 70 | ⌝ | 8.1 ± 3.2 | ⌝ | 103 ± 32 | ⌝ | |
| a | a | a | ||||||
| KO (n = 5) | ND | 37 ± 6 | ⌟ | 0.8 ± 0.1 | ⌟ | 17 ± 2 | ⌟ | |
NS, P = 0.2.
Values that were statistically different.
Steroid Hormone Levels in StAR Knockout Mice
| Estradiol (pg/ml) | Progesterone (pg/ml) | Testosterone (ng/ml) | Corticosterone (ng/ml) | |||||
|---|---|---|---|---|---|---|---|---|
| Female | ||||||||
| WT (n = 5) | 10.2 ± 1.9 | ⌝ | 497 ± 177 | ⌝ | ND | 305 ± 77 | ⌝ | |
| NS | a | a | ||||||
| KO (n = 5) | 16.2 ± 3.8 | ⌟ | 39 ± 8 | ⌟ | ND | 9 ± 2 | ⌟ | |
| Male | ||||||||
| WT (n = 5) | ND | 290 ± 70 | ⌝ | 8.1 ± 3.2 | ⌝ | 103 ± 32 | ⌝ | |
| a | a | a | ||||||
| KO (n = 5) | ND | 37 ± 6 | ⌟ | 0.8 ± 0.1 | ⌟ | 17 ± 2 | ⌟ | |
| Estradiol (pg/ml) | Progesterone (pg/ml) | Testosterone (ng/ml) | Corticosterone (ng/ml) | |||||
|---|---|---|---|---|---|---|---|---|
| Female | ||||||||
| WT (n = 5) | 10.2 ± 1.9 | ⌝ | 497 ± 177 | ⌝ | ND | 305 ± 77 | ⌝ | |
| NS | a | a | ||||||
| KO (n = 5) | 16.2 ± 3.8 | ⌟ | 39 ± 8 | ⌟ | ND | 9 ± 2 | ⌟ | |
| Male | ||||||||
| WT (n = 5) | ND | 290 ± 70 | ⌝ | 8.1 ± 3.2 | ⌝ | 103 ± 32 | ⌝ | |
| a | a | a | ||||||
| KO (n = 5) | ND | 37 ± 6 | ⌟ | 0.8 ± 0.1 | ⌟ | 17 ± 2 | ⌟ | |
NS, P = 0.2.
Values that were statistically different.
Developmental Effects of StAR Deficiency in the Adrenal Glands
The adrenal cortex is the best studied steroidogenic organ in human patients with lipoid CAH. In StAR knockout mice, the adrenal cortex, which exhibited the most marked abnormalities in newborn pups (5), underwent progressive increases in lipid deposition, as revealed by vacuolated regions in hematoxylin-eosin-stained sections (Fig. 1) and marked lipid deposits demonstrated by oil red O staining (Fig. 2). At 8 weeks, these changes were especially marked in the zona glomerulosa and zona fasciculata, while the zona reticularis was relatively preserved and the adrenal medulla lacked discernable pathological findings. Unlike human patients with lipoid CAH (6), cholesterol crystals were not present in the mouse adrenal sections, and there was no evidence of adrenocortical hyperplasia at 8 weeks (i.e. the cortex-medulla ratio was not increased in StAR knockout adrenals). The postnatal administration of corticosteroids to StAR knockout mice before weaning at 3 weeks may have partially suppressed ACTH release and ameliorated the progression of the adrenal lesions. By 16 weeks, adrenocortical hyperplasia was apparent, accompanied by markedly increased lipid deposits (data not shown).
Histology of Wild-Type and StAR Knockout Adrenal Glands from Mice at Different Ages As described in Materials and Methods, adrenal glands were isolated from wild-type (WT) and StAR knockout (KO) mice either immediately after birth (PO) or at 8 weeks of age (8w), and sections were prepared and stained with hematoxylin-eosin. Panel A, WT, P0. Panel B, WT, 8w. Panel C, KO, P0. Panel D, KO, 8w. c, Cortex; m, medulla.
Oil Red O Staining of Lipid Deposits in Steroidogenic Organs of Wild-Type and StAR Knockout Mice Steroidogenic organs were harvested shortly after birth (P0, adrenal and testis; P7, ovary) or at 8 weeks of age from wild-type or StAR knockout mice. Panel A, WT adrenal, P0. Panel B, WT adrenal, 8w. Panel C, WT testis, P0. Panel D, WT testis, 8w. Panel E, WT ovary, P7. Panel F, WT ovary, 8w. Panel G, KO adrenal, P0. Panel H, KO adrenal, 8w. Panel I, KO testis, P0. Panel J, KO testis, 8w. Panel K, KO ovary, P7. Panel L, KO ovary, 8w.
Developmental Effects of StAR Deficiency in the Testes
The testes of lipoid CAH patients have been studied in less detail (7, 8). Because concurrent treatment with sex steroids was not required for survival, we were able to study the full developmental effects of StAR deficiency in the gonads of StAR knockout mice. Anatomically, the testes in StAR knockout mice are found in the inguinal canal. A similar position is observed in testicular feminization mice, which have deficient androgen action due to a mutation in the androgen receptor (9). As shown in Figs. 2 and 3, the testes of newborn StAR knockout mice appeared relatively intact, with only scattered foci of lipid deposits in the interstitial region. These deposits presumably are in remnants of fetal Leydig cells, a distinct population of cells that produce androgens in utero and then regress postnatally (10, 11). As StAR knockout mice aged, lipid accumulated throughout the interstitial region, consistent with recruitment of the adult population of Leydig cells. At 8 weeks of age (Fig. 3), the cytoplasm of the Leydig cells appeared foamy, and oil red O staining (Fig. 2) revealed progression in the severity of their lipid deposits. Moreover, Leydig cell hyperplasia was noted in the StAR knockout testes (Fig. 3G); at 8 weeks of age, oil red O-positive areas accounted for 10.5% of the area of StAR knockout testes (vs. 3.18% of wild-type testes). These studies document a progressive worsening in the integrity of the steroidogenic cells, precisely as predicted by the two-stage model.
Histology of the Testes from Wild-Type and StAR Knockout Mice at Different Ages Testes were isolated from wild-type and StAR knockout mice at P0, 4w, or 8w. Higher power views of sections from the KO mice at 8w are shown in the right panels. Panel A, WT, P0. Panel B, WT, 4w. Panel C, WT, 8w. Panel D, KO, P0. Panel E, KO, 4w. Panel F, KO, 8w. Panel G, KO 8w, high power. Note Leydig cell hyperplasia in the interstitial region. Panel H, KO, 8w, high power. The solid arrow points to multinucleated germ cells within the seminiferous tubules, and the dashed arrow points to a germ cell with chromatin condensation.
Previous studies have suggested that spermatogenesis absolutely requires androgens (reviewed in Ref. 12). Despite marked lipid deposits in the Leydig cells of StAR knockout testes at 8 weeks, microscopic examination revealed residual capacity for spermatogenesis. The basal spermatogonia in the seminiferous tubules were intact (Fig. 3) and mature spermatids were detected in the epididymis (data not shown). In addition, as shown in Fig. 4, StAR knockout testes at 8 weeks had comparable immunohistochemical staining for CREM, a marker for postmeiotic germ cells (13), and cyclin A1, which is expressed during the first meiotic division (14). These results, in conjunction with the low-butdetectable plasma testosterone levels (Table 1), indicate that StAR knockout testes retain sufficient androgen biosynthesis to support germ cell maturation, a finding that persists even at 24 weeks of age (data not shown). Examination at earlier stages of development, however, showed that the initial wave of spermatogenesis in StAR knockout testes is delayed. Whereas wild-type seminiferous tubules at 4 weeks contained mature elongated spermatids (Fig. 3B), germ cell maturation in StAR knockout mice had not progressed beyond step 14 [stage II/III elongating spermatids (Fig. 3E)]. Moreover, the tubules contained multinucleated giant cells (Fig. 3H, solid arrow), and individual germ cells exhibited chromatin condensation (dashed arrow), features consistent with programmed cell death. TUNEL assays (Fig. 4) revealed considerably increased labeling in germ cells of some tubules, consistent with DNA fragmentation. Collectively, these findings suggest that androgen production in StAR knockout mice is inadequate to maintain optimal spermatogenesis.
Expression of Markers of Germ Cell Maturation and TUNEL Assay of Programmed Cell Death in Testis Sections Sections from WT or KO testes at 8w were analyzed as described in Materials and Methods by immunohistochemistry with antisera specific for CREM or cyclin A or TUNEL assay. Panel A, WT, anti-CREM antibody. Panel B, WT, anti-cyclin A antibody. Panel C, WT, TUNEL assay. Panel D, KO, anti-CREM antibody. Panel E, KO, anti-cyclin A antibody. Panel F, KO, TUNEL assay. Panel G, KO, TUNEL assay, high-power.
Developmental Effects of StAR Deficiency in the Ovaries
We next examined the ovaries of StAR knockout mice. At 1 week of age (P7), both wild-type and StAR knockout ovaries contained immature follicles at the one- and two-cell stages (Fig. 5), and no lipid deposits were seen in oil red O staining (Fig. 2). By 8 weeks of age, when sexual maturation normally has occurred, the StAR knockout ovaries exhibited impaired follicular maturation and contained abundant lipid deposits, primarily in stromal cells that exhibited a luteinized appearance (Fig. 2) and expressed high levels of the cholesterol side-chain cleavage enzyme (Fig. 5, N and P). Although some follicles contained luteinized theca cells, there were no corpora lutea (i.e. structures composed of well developed luteinized granulosa and theca interna cells with dissolution of the basement membrane) in any section examined (Fig. 5 and data not shown). The markedly decreased level of circulating progesterone further suggests that the decreased steroidogenesis impaired ovulation. In contrast, both granulosa cells and theca interna cells appeared relatively normal in the remaining follicles. Although their circulating estrogen levels were not clearly decreased, StAR knockout females had sexually immature uterus and oviducts (data not shown), suggesting impaired estrogen action.
Histology of Ovaries from Wild-Type and StAR Knockout Mice at Different Ages Ovaries were isolated from WT and StAR KO mice at the indicated ages and were processed as described in Materials and Methods. Top panels show hematoxylin-eosin staining of sections. Middle panels show photomicrographs of ovarian sections analyzed by in situ hybridization with an antisense probe for cholesterol side-chain cleavage enzyme. The bottom panels show oil red O staining. Panel A, WT, P7, low power. Panel B, WT 8w, low power. Panel C, WT, 24 w, low power. Panel D, WT, P7, high power. Panel E, WT, 8w, high power. Panel F, WT, 24w, high power. Panel G, KO, P7, low power. Panel H, KO, 8w, low power. Panel I, KO, 24 w, low power. Panel J, KO, P7, high power. Panel K, KO, 8w, high power. Panel L, KO, 24w, high power. Panel M, WT, 8w, brightfield, P450scc. Panel N, WT, 8w, darkfield, P450scc. Panel O, KO, 8w, brightfield, P450scc. Panel P, KO, 8w, darkfield, P450scc. Panel Q, WT, 24w, oil red O. Panel R, KO, 24w, oil red O. CL, Corpus luteum.
By 24 weeks, the StAR knockout ovaries were larger than the wild-type ovaries. They retained only a few scattered follicles and were composed largely of vacuolated stromal cells, consistent with the onset of premature ovarian failure (Fig. 5, I and L). These studies, which provide the first detailed analyses of ovarian histopathology in the absence of StAR, again are consistent with the two-hit model of lipoid CAH.
Developmental Effects of StAR Deficiency on Accessory Sex Organs
Limited analyses of 46, XY patients with lipoid CAH suggested that structures derived from the wolffian ducts were normal, while the external genitalia were completely feminized (15). In contrast, there were well documented examples of 46, XX patients who underwent menarche and breast development (4). The ability to maintain StAR knockout mice until the time of normal puberty allowed us to examine the effect of StAR deficiency on development of accessory sex organs. We noted previously that the epididymal structures of wild-type and StAR knockout mice were indistinguishable at birth (5); comparable histology of the epididymis and vas deferens also was seen in older StAR knockout mice (data not shown). In contrast, while the seminal vesicles appeared relatively normal at birth, they were clearly hypoplastic at 4 and 8 weeks of age (Fig. 6). Microscopically, the seminal vesicles contained only simple tubular structures with few convolutions (data not shown). Finally, the prostate was hypoplastic, both immediately after birth and at 8 weeks of age (Fig. 6), and on microscopic examination contained only primary branching tubules (data not shown). These findings suggest that residual capacity for androgen synthesis in the absence of StAR, although sufficient to support survival of wolffian structures in utero, is insufficient to virilize more distal accessory sex glands at the time of normal puberty.
Effect of StAR Knockout on Male Accessory Sex Organs The male accessory sex glands were harvested from WT and StAR KO mice at the indicated ages. For P0, sagittal sections were stained with hematoxylin-eosin and analyzed by photomicroscopy. For 4w and 8w, the structures were processed as described in Materials and Methods. Panel A, WT, P0. Panel B, KO, P0. Panel C, WT and KO seminal vesicles, 4w. Panel D, WT and KO seminal vesicles, 8w. Panel E, WT and KO prostate, 4w. Panel F, WT and KO prostate, 8w.
DISCUSSION
In this report, we extend our analyses of the StAR knockout mice, examining the temporal effects of StAR deficiency on steroidogenic organs in vivo and defining the effects of the StAR knockout on development of male accessory sex organs and sperm maturation. Collectively, these studies provide new insights into the pathogenesis of StAR deficiency, particularly with respect to developmental events in gonads that support the two-hit model for the pathogenesis of lipoid CAH (5). As predicted by this model, we observed a hierarchy of involvement of the different steroidogenic organs. Adrenal glands were affected severely at birth and subsequently showed progressive increases in lipid deposition. Although the Leydig cells contained scattered foci of lipid deposits shortly after birth, more generalized lipid deposits were not seen until the time of normal sexual maturation. Hyperplastic Leydig cells containing florid lipid deposits ultimately filled the interstitial region. As noted previously (5), the ovaries appeared normal at birth and lacked any lipid deposits. When kept alive until after the time of normal sexual maturation, the ovaries exhibited histological abnormalities, with decreased numbers of follicles, absent corpora lutea, and proliferation of vacuolated stromal cells that strongly expressed P450scc (Fig. 5). These findings strongly support the concept that persistent trophic hormone stimulation of StAR-deficient steroidogenic organs leads to progressively worsening lipid deposits and diminished function of steroidogenic cells.
Steroidogenic cells obtain cholesterol from multiple sources, including de novo biosynthesis, hydrolysis of cholesterol esters, and uptake from circulating lipoproteins [in mice, predominantly high density lipoprotein (HDL) taken up by the scavenger receptor-B1 pathway]. The StAR knockout mice provide a model system to address the relative contributions of these different pathways to the lipid deposits that accumulate in steroidogenic cells. In situ hybridization analyses did not reveal increased expression of SR-B1 in StAR knockout adrenals or testes (T. Hasegawa, unpublished observation). However, preliminary studies indicate that double knockout mice lacking both ApoA1 and StAR, although severely deficient in steroidogenesis, accumulate considerably less lipid in their steroidogenic cells (T. Hasegawa, unpublished observation). These findings implicate circulating HDL as the major source of the cholesterol that accumulates in StAR knockout mice.
Few studies have examined testicular structure, spermatogenesis, or development of male secondary sexual organs in 46,XY subjects with lipoid CAH. Although differing in age and location of testes (i.e. abdominal vs. inguinal), most lipoid CAH patients have had normal appearing Sertoli cells, Leydig cell hyperplasia, and decreased numbers of germ cell precursors; internal genitalia have included normal epididymis and vas deferens, with no description of the prostate. Similarly, StAR knockout mice had normal appearing epididymis and vas deferens and mature spermatids within the epididymis (data not shown), supporting some residual capacity for androgen biosynthesis. They did exhibit some signs of impaired spermatogenesis, with delayed germ cell maturation at 4 weeks of age and increased apoptosis in developing spermatocytes. Moreover, the seminal vesicles and prostate were markedly hypoplastic, and there was no virilization of external genital structures.
Developmentally, the seminal vesicles, epididymis, and ductus deferens all arise from the Wolffian ducts, whereas the prostate develops from the urogenital sinus. The apparent discrepancy in the development of organs derived from the wolffian ducts probably reflects differing thresholds for paracrine actions of androgens in the immediate vicinity of the testes (i.e. epididymis and ductus deferens) vs. endocrine actions at more distal sites (i.e. seminal vesicles). The necessity to convert testosterone to dihydrotestosterone for full virilization may further exacerbate the effects of the testosterone deficiency, as supported by the prostate hypoplasia in patients with genetic defects in the type 2 isozyme of steroid 5α-reductase (16). However, knockout mice lacking both type 1 and 2 isozymes of 5α-reductase undergo normal male sexual differentiation (M. Mahendroo and D. Russell, personal communication), suggesting that there may be species-dependent differences in these processes.
In part because their gonads are not removed to prevent malignant transformation, even less is known about the structure and function of ovaries in 46,XX patients with lipoid CAH. The spontaneous onset of breast development and menarche in 46,XX patients with lipoid CAH was one factor that prompted the two-hit model. The worsening ovarian histopathology in StAR knockout mice, beginning at the time of normal puberty, is entirely consistent with this model. Under persistent gonadotropin stimulation, the ovaries of StAR knockout mice developed progressive lipid deposition, particularly within the stromal cells. Although their serum estradiol levels did not differ from those in wild-type mice, the uterus and oviducts of StAR knockout mice were markedly hypoplastic, indicating that estrogen production was inadequate to stimulate normal development of these secondary sex organs. No corpora lutea were detected, and progesterone levels were markedly impaired, indicating that the ovaries of StAR knockout mice were anovulatory. In this regard, the relative roles of defects in estrogen vs. progesterone biosynthesis in the ovaries of StAR knockout mice warrant further study. Ovulation also is impaired in knockout mice lacking progesterone receptor (17), suggesting that abnormal production of progesterone at least partly explains the ovarian phenotype.
A recent study documented the expression of StAR transcripts in the rat brain, colocalizing with transcripts for cholesterol side chain cleavage enzyme and 3β-hydroxysteroid dehydrogenase in the hippocampus, dentate gyrus, and granular and Purkinje cells of the cerebellum (18). Given the proposed roles of endogenous steroids, designated neurosteroids, within the central nervous system (19), it is of interest to analyze neuronal function in StAR knockout mice. To date, the histology of different brain regions where StAR is expressed does not appear distinguishable from the same regions in wild-type mice (L. Zhao, unpublished observation). However, it is important to note that neurophysiological analyses of StAR knockout mice have not been performed, and this area needs further investigation. Ultimately, we may need to make tissue-specific knockouts that ablate StAR in the brain but retain its expression in the primary steroidogenic tissues to delineate specific roles of StAR in the central nervous system.
We noted previously that some StAR knockout mice exhibited signs of respiratory distress. However, analyses of the lungs of newborn StAR knockout mice did not reveal any consistent histological abnormalities, and in situ hybridization analyses showed comparable expression of genes encoding surfactant proteins A and C in wild-type and knockout mice (T. Hasegawa, unpublished observation). Although it is tempting to ascribe the apparent respiratory distress to impaired lung maturation secondary to glucocorticoid deficiency, results to date have not elucidated the molecular basis for this aspect of the StAR knockout phenotype.
In summary, our studies, which demonstrate progressive histopathological changes in the gonads after the time of normal puberty, strongly support the two-hit model of lipoid CAH. The StAR knockout mice, and immortalized cell lines derived from their steroidogenic organs, hold considerable promise as a system to expand our understanding of the mechanisms by which StAR facilitates cholesterol translocation to the inner mitochondrial membrane. Through studies such as these, we hope to increase our understanding of how StAR makes its essential contributions to adrenocortical and gonadal steroidogenesis and endocrine function.
MATERIALS AND METHODS
Reagents
Oligonucleotides and reagents for PCR were purchased from Life Technologies, Inc. (Gaithersburg, MD) and PE Applied Biosystems (Norwalk, CT), respectively. Radionuclides were purchased from Amersham Pharmacia Biotech (Arlington Heights, IL), and routine laboratory reagents were purchased from Sigma (St. Louis, MO). Anti-CREM antiserum was purchased from Santa Cruz Biotechnology, Inc. (Santa Cruz, CA), and anti-cyclin A antiserum was a generous gift from Dr. Debra Wolgemuth (Columbia University, New York, NY).
Generation of StAR Knockout Mice
All animal studies were approved by the Institutional Review Committee at University of Texas Southwestern Medical Center. The StAR knockout mice were generated as previously described (5) and maintained as +/− heterozygotes, which were crossed to produce StAR knockout pups. Mice were maintained in the University of Texas Southwestern Animal Resources Center on a 12-h light, 12-h dark cycle and were given food and water ad libitum. StAR genotypes were determined by PCR analyses of tail DNA using forward (5′-AAGAGCTCAACTGGAGAGCAC-3′) and reverse (5′-TACTTAGCACTTCGTCCCCGT-3′) primers. Genetic sex was determined by PCR analyses for Sry as previously described (20).
Corticosteroid Rescue of StAR Knockout Mice
StAR knockout mice were rescued with a modification of a previously described steroid replacement regimen that includes both glucocorticoids and mineralocorticoids (21). Stock solutions of dexamethasone 21-phosphate (4 mg/ml in H2O), fludrocortisone acetate (5 mg/ml in 95% ethanol), and hydrocortisone (4 mg/ml in 95% ethanol) were prepared and stored at 4 C. A corticosteroid cocktail was made by diluting the stock solutions in olive oil (1:10,000 for dexamethasone and fludrocortisone, 1:10 for hydrocortisone). All newborn pups were injected with 0.05 ml sc once daily until StAR knockout pups were identified by PCR analysis. Steroid injections then were continued in StAR knockout pups until weaning, when the mice were provided ad libitum with 0.9% sodium chloride as drinking solution and all steroid injections were stopped.
Histological Analyses
Histological analyses were carried out with 4% paraformaldehyde or Bouin’s fixed tissue specimens with hematoxylin and eosin staining, or frozen tissue sections for oil red O staining with hematoxylin counterstaining. Areas positive for oil red O staining were measured in 20 high-power fields (40×: 3.2 × 1–8 mm) in representative specimens using CAS 200 morphometrical analysis as described (22). To examine prostate morphology, glandular and stromal compartments were separated by a modification of a published method (23). Immunohistochemical analyses were carried out with primary antibodies including anti-CREM (1:100 dilution) and anti-cyclin A1 (1:500 dilution). Control reactions were performed with normal rabbit serum (DAKO Corp., Carpinteria, CA). The TUNEL assay for DNA fragmentation was performed with a kit purchased from Intergen (Purchase, NY) according to the manufacturer’s protocol. In situ hybridization was performed with an antisense probe specific for the cholesterol side-chain cleavage enzyme as previously described (24).
Hormone Assays
RIAs were performed by Dr. David Hess, Oregon Regional Primate Center using serum collected from +/+ and −/− mice at 8 weeks of age. Steroids analyzed included corticosterone [limit of detection, 2 ng/ml; intraassay % coefficient of variation (CV), 8.5%; % recovery, 88.3], progesterone (limit of detection, 30 pg/ml; intraassay % CV, 10.7%; % recovery, 87.8), testosterone (limit of detection, 0.1 ng/ml; intraassay % CV, 6.2%; % recovery, 69.1), and estradiol (limit of detection, 3.0 pg/ml; intraassay % CV, 9.5%; % recovery, 79.1). The means and ses of each group were calculated, and the statistical significance of differences was determined by the Mann-Whitney U test.
Acknowledgments
We thank Dr. Beverly Koller and Ann Latour for invaluable assistance in producing the StAR knockout mice, Drs. Martin Matzuk and William Rainey for helpful discussions, Dr. Deborah Wolgemuth for the anticyclin A1 antiserum, Dr. David Hess for measurements of circulating steroid hormones, and Dr. Yelena Krimkevich for excellent technical assistance.
This work was supported by grant support from the NIH (DK-54028 and DK-54480 to K.L.P.).
