Gonadotropin-Induced Adrenocortical Neoplasia in NU/J Nude Mice
http://www.100md.com
《内分泌学杂志》
Departments of Pediatrics (M.B., E.G., M.H., D.B.W.) and Molecular Biology and Pharmacology (I.B., D.B.W.), Washington University School of Medicine, St. Louis Children’s Hospital, St. Louis, Missouri 63110
Program for Developmental and Reproductive Biology (H.P., S.K., M.H.), Children’s Hospital, Biomedicum Helsinki, University of Helsinki, 00290 Helsinki, Finland
Department of Physiology (N.R.), Institute of Biomedicine, University of Turku, 20520 Turku, Finland
Department of Physiology (J.L.), University of Oulu, 90220 Oulu, Finland
Abstract
In response to prepubertal gonadectomy certain inbred mouse strains, including DBA/2J, develop sex steroid-producing adrenocortical neoplasms. This phenomenon has been attributed to a lack of gonadal hormones or a compensatory increase in gonadotropins. To assess the relative importance of these mechanisms, we created a new inbred model of adrenocortical neoplasia using female NU/J nude mice. These mice developed adrenocortical neoplasms in response to either gonadectomy or gonadotropin elevation from xenografts of human chorionic gonadotropin (hCG)-secreting Chinese hamster ovary cells. In each instance the adrenal tumors resembled the neoplasms found in gonadectomized DBA/2J mice and were composed of spindle-shaped A cells and lipid-laden B cells. Both cell populations were defined by ectopic expression of GATA-4 and an absence of the adrenocortical markers melanocortin-2-receptor and steroid 21-hydroxylase, but only B cells expressed the gonadal steroidogenic markers inhibin-, LH receptor, P450c17, and P450c19. Expression of sex steroidogenic markers was attenuated in the neoplastic adrenal cortex of hCG-treated vs. gonadectomized mice. Whereas neoplastic adrenals were an obvious source of estradiol in gonadectomized mice, ovaries appeared to be the major source of this hormone in hCG-treated mice. Gonadectomy and hCG treatment elicited comparable increases in serum estradiol, but testosterone levels increased significantly only in hCG-treated mice. We conclude that chronic gonadotropin elevation, caused by either gonadectomy or hCG administration, signals a population of cells in the adrenal subcapsular region of permissive mice to undergo differentiation along a gonadal rather than an adrenal lineage. Thus, NU/J nude mice can be used as a model to study both neoplasia and adrenogonadal lineage specification.
Introduction
SUBCAPSULAR ADRENOCORTICAL neoplasms may arise in the setting of gonadal failure and chronic gonadotropin elevation, as in postmenopausal women (1, 2, 3, 4) and men with acquired testicular atrophy (5). Some of these adrenal tumors resemble gonadal stroma and produce sex steroids in response to continuous gonadotropin stimulation. Incidental adrenocortical tumors are present in up to 6% of older individuals (6, 7), and a subset of these tumors is gonadotropin responsive (7, 8). Sex steroid-producing adrenocortical neoplasms also occur in neutered domestic animals, including pet ferrets (9, 10). A lack of gonadal hormones or a compensatory increase in gonadotropins is postulated to drive adrenocortical neoplasia, although the relative importance of these mechanisms remains controversial (10, 11, 12, 13).
Inbred laboratory mice make a good model to study this phenomenon. Certain mouse strains (e.g. DBA/2J, CE) develop sex steroid-producing adrenocortical neoplasms in response to prepubertal gonadectomy (11, 14). Organ transplantation experiments have established that the adrenal glands of susceptible strains exhibit an inherent predisposition to tumor formation in response to the hormonal changes that accompany gonadectomy (15). The genetic basis of strain susceptibility is unknown but seems to correlate with reduced fertility and a polymorphism in the coding region of steroidogenic factor-1 (SF-1), which may affect steroidogenesis both in adrenal cortex and gonads (16, 17). Neoplasms in the adrenals of these mice arise at multiple foci in the subcapsular region and invade the underlying zones of adrenal cortex in a fashion reminiscent of the centripetal pattern of adrenal renewal (18). The tumors are characterized by nests of steroidogenic cells, termed B cells, which appear amid ovoid- or spindle-shaped cells, known as A cells (11, 14).
Previously we characterized adrenocortical neoplasms in gonadectomized DBA/2J mice and ferrets and showed that the gene products GATA-4, LH receptor (LHR), and inhibin- are hallmarks of these tumors (19, 20). In the present study, we have developed a new inbred model of adrenocortical neoplasia to assess the relative contributions of gonadectomy vs. gonadotropin elevation to the induction and progression of adrenocortical neoplasia. Our new model employs NU/J mice, an inbred strain carrying nude (Foxn1nu), a recessive loss-of-function mutation in the gene encoding Foxn1, which is associated with athymia (21). Nude (Foxn1nu/Foxn1nu) mice are widely used in cell and organ transplantation experiments. Early studies with different strains of nude or thymectomized mice demonstrated deficiencies in reproductive function, including reduced fertility and decreased sex steroid production (22, 23). In light of these reproductive abnormalities, we postulated that athymic nude mice might be predisposed to gonadectomy-induced adrenocortical tumorigenesis. Here we show that NU/J nude mice, like strain DBA/2J, respond to gonadectomy by increasing serum gonadotropin levels and developing sex steroid-producing neoplasms in the adrenal cortex and that similar neoplastic changes occur in intact nude mice after chronic gonadotropin stimulation.
Materials and Methods
Experimental animals and surgery
All animal work was carried out in accordance with National Institutes of Health Guidelines for the Care and Use of Laboratory Animals and was approved by the institutional animal studies committee. Weanling mice homozygous for the Foxn1nu mutation on an NU/J background were purchased from The Jackson Laboratory (Bar Harbor, ME). The mice were specific pathogen free and housed five per cage in controlled conditions of light (12 h light, 12 h darkness) and temperature (21 C). They were fed commercial mouse chow (Purina) and tap water ad libitum. Female mice (3–4 wk of age) were anesthetized with ip administration of ketamine (75 mg/kg), xylazine (4 mg/kg), and acepromazine (0.75 mg/kg) and subjected to gonadectomy or sham surgery. At specified times the mice were killed by CO2 inhalation and blood samples were collected. Harvested tissue was fixed overnight in 4% paraformaldehyde, embedded in paraffin, and then sectioned (4 μm) for histochemical staining (hematoxylin-eosin or toluidine blue) or immunohistochemistry. For in situ hybridization isolated adrenals were frozen in OCT (Tissue-Tek, Torrance, CA). TRIzol reagent (Invitrogen, Carlsbad, CA) was used to isolate total RNA for RNase protection and RT-PCR assays.
Cell culture and inoculations
Chinese hamster ovary (CHO) cells stably expressing a biologically active single-chain human chorionic gonadotropin (hCG) variant (hCG-CHO cells) were prepared as described elsewhere (24, 25). hCG-CHO cells or control CHO cells were maintained in Ham’s F-12 medium [supplemented with penicillin (100 U/ml), streptomycin (100 μg/ml), and 2 mM glutamine] containing 5% fetal bovine serum and G418 (125 μg/ml) at 37 C in a humidified atmosphere of 5% CO2-95% air. To inoculate nude mice, hCG-CHO or control CHO cells were harvested with trypsin, washed in PBS, resuspended in Ham’s F-12 medium (without serum) at a concentration of 3 x 107 cells/ml. Immediately after gonadectomy or sham surgery, nude mice were injected sc in each flank with 0.2 ml of this cell suspension.
Hormone measurements
Serum LH concentrations were measured using an immunofluorometric assay (Wallac Oy, Turku, Finland), as described previously (26). Serum levels of hCG and testosterone were measured using commercial assay kits (Diagnostic Products Corp., Los Angeles, CA). Serum estradiol levels were determined by RIA (Spectria, Orion Diagnostica, Espoo, Finland). Corticosterone levels were measured by RIA (IDS, Boldon, UK). Data reported are expressed as the mean ± SEM. Statistical significance was determined using Student’s t test in Microsoft Excel data analysis software (Microsoft Corp., Seattle, WA).
Immunohistochemistry
Paraffin-embedded, paraformaldehyde-fixed tissue sections were deparaffinized and rehydrated (27). Endogenous peroxidase was quenched by 30 min incubation in 1% H2O2 in PBS. Slides were then treated with avidin/biotin blocking (SP-2001; Vector Laboratories, Burlingame, CA) for 20 min, rinsed in three changes of PBS, and then treated with protein block (X0909; Dako, Carpinteria, CA) for 10 min. Sections were incubated overnight at 4 C with primary antibodies (Table 1). After rinsing for 30 min in PBS, secondary antibody (Table 1) was applied for 30 min at 30 C. After rinsing in PBS, an avidin-biotin immunoperoxidase system (Vectastain Elite ABC kit; Vector Laboratories) and diaminobenzidine (Sigma Chemicals, St. Louis, MO) were used to visualize the bound antibody (27). Slides were counterstained with Gill’s hematoxylin.
In situ hybridization
Tissue cryosections (10 μm) were subjected to in situ hybridization as described (19). [33P]-labeled antisense riboprobes were prepared using the following linearized templates: 1) melanocortin-2 receptor (MC2R), BamHI-digest pCRII plasmid containing nucleotides 23–663 of the cDNA (12), T7 polymerase; 2) steroid 21-hydroxylase (21-OHase), EcoRI digest of a cDNA, T3 polymerase (28); 3) Müllerian inhibitory substance (MIS) receptor II, EcoRV-digested pCRII plasmid containing nucleotides 681-1080 of the cDNA plasmid, Sp6 polymerase (29); 4) P450c17, as described elsewhere (19).
RNase protection assays
These assays were performed with a commercially available kit, RPA III (Ambion, Austin, TX) and 5 μg of total RNA. The linearized templates used have been described previously (19).
RT-PCR assays
Reverse transcription was performed with 3 μg of total RNA and random hexamer primers (Promega, Madison, WI). After 10 min incubation at 65 C, the reaction mixture was supplemented with 5x avian myeloblastosis virus-reverse transcription buffer (Fisher Scientific, Pittsburgh, PA), MgCl2 to 25 mM, and deoxynucleotide triphosphates to a final concentration of 2 mM each, 80 U RNase inhibitor (Promega), and 20 U avian myeloblastosis virus reverse transcriptase (Stratagene, La Jolla, CA), and the reaction was run for 1 h at 42 C in the final volume of 40 μl. The quality of each cDNA was determined by the relative level of amplification of the mouse Hprt gene. The following cDNA primers were used: P450c19 (ovarian-specific alternative first exon), forward primer (+1 to +22) 5'-ACAGCATTGTGATTGTCCCTCT-3', reverse primer (+324 to +344) 5'-CATCTTGCGCTATTTGGCCTC-3' (30); hypoxanthine-guanine phosphoribosyl transferase (HPRT), forward primer (+531 to +539) 5'-GCTGGTGAAAAGGACCTCT-3', reverse primer (+750 to +769 5'-CACAGGACTAGAACACCTGC-3', GenBank accession no. BC083145.1. The PCR contained 2.5% of the cDNA product, Pfu buffer (Stratagene), 20 pmol of each primer, and 1 U of Pfu polymerase (Stratagene) in a final volume of 20 μl. The thermal cycler conditions were: 94 C (4 min), 1 cycle; 94 C (1 min), 57 C (1 min) for P450c19, and 56 C (1 min) for HPRT, 72 C (1 min), 35 cycles; and 72 C (10 min), one cycle. Agarose gel electrophoresis (1.2%) in the presence of ethidium bromide demonstrated a single band of the expected size for each of the PCR primer pairs. Authenticity of the P450c19 PCR product was confirmed by direct DNA sequencing.
Results
Morphological changes in the adrenal glands of gonadectomized NU/J nude mice
We hypothesized that nude mice, like other inbred strains with impaired sex steroid production, might be predisposed to gonadectomy-induced adrenocortical tumorigenesis. To test this possibility, weanling female NU/J nude mice were subjected to sham surgery or gonadectomy. Within weeks of gonadectomy, patches of neoplastic cells appeared beneath the adrenal capsule and extended into the cortex. By 2 months wedge-shaped neoplastic lesions containing spindle-shaped A cells and lipid-laden B cells were evident in the adrenals (Fig. 1, A vs. B), and by 4 months tumors occupied a large fraction of the adrenal cortex (Fig. 1, C vs. D). In addition, there was evidence of mast cell infiltration in the neoplastic tissue (Fig. 1, E vs. F), a phenomenon known to accompany gonadectomy-induced tumorigenesis in susceptible inbred mice (31). These morphological changes were indistinguishable from those seen in the adrenal glands of DBA/2J mice gonadectomized for comparable lengths of time (14, 19).
Ectopic expression of gonadal transcripts in the adrenals of gonadectomized NU/J nude mice
To characterize the gonadectomy-induced changes in NU/J mice on a molecular level, we performed in situ hybridization on adrenal glands. As markers of normal adrenocortical steroidogenesis, we used MC2R, which binds ACTH and promotes adrenocortical cell proliferation and differentiation, and steroid 21-OHase, an enzyme required for glucocorticoid production. At 3 wk after gonadectomy, neither MC2R nor 21-OHase mRNA was expressed in the neoplastic areas (Fig. 2, A and B), indicating that these cells were incapable of performing typical adrenocortical functions. Instead the cells in the adrenal marker-deficient areas contained transcripts for gonadal markers. mRNA for MIS receptor type II (MISIIR), which is found in gonadal somatic cells of males and females (32) and has been observed in some gonadal and extragonadal epithelial tumors (33), was detected in the tumorous adrenals (Fig. 2C). High-magnification bright-field images confirmed the presence of this transcript in small, densely packed ovoid- or spindle-shaped cells that lacked lipid droplets, presumed to represent a subset of A cells or a perhaps a transitional cell type (Fig. 2E). P450c17, which is required for synthesis of gonadal sex steroids and normally is not expressed in adrenal cortex beyond fetal development (34), was expressed by large, lipid-laden B cells in the neoplastic adrenocortical tissue (Fig. 2, D and F).
Ectopic expression of gonadal proteins in the adrenals of gonadectomized NU/J nude mice
As in other models of gonadectomy-induced adrenocortical neoplasia (19, 35), immunohistochemistry showed the presence of GATA-4 protein in the nuclei of both A and B cells (Fig. 3A), whereas cytoplasmic LHR (Fig. 3B) and inhibin- (Fig. 3C) were observed only in B cells. P450c19 (aromatase), an enzyme involved in physiological sex steroidogenesis in gonads, also was detected in neoplastic B cells (Fig. 3D), underscoring their capacity to synthesize estrogen. The presence of MISIIR mRNA in the subcapsular region of the neoplastic lesions (Fig. 2, C and E) prompted us to examine expression of its ligand, MIS. This factor is expressed by gonadal cells in males and females and can be used as a tumor marker for sex cord-derived neoplasms and germinal epithelium-derived carcinomas (36). In gonads, the Mis gene is a known target of transcriptional activation by GATA-4 (37, 38). MIS immunoreactivity was detected in B cells and a subset of A cells (Fig. 3E). This raises the possibility that MIS produced by B cells acts in a paracrine fashion on A cells or vice versa, perhaps influencing events involved in the phenotypic switch from adrenal- to gonadal-type steroidogenesis. SF-1, which cooperates with GATA-4 in the transcriptional activation of Mis (38) and inhibin- (39, 40), was observed in B cells and normal adrenocortical cells but not in A cells (Fig. 3F). Estrogen receptor (ER)-, a nuclear hormone receptor implicated in normal and neoplastic cell proliferation, was observed mainly in nuclei of B cells and probably endothelial cells (Fig. 3G). In addition, we examined expression of Smad3, an intracellular mediator of TGF signaling (41). Smad3 immunoreactivity was evident in B cells (Fig. 3H) and in a subset of normal adrenocortical cells beneath the capsule. However, Smad3 was not detected in A cells. Similarly, -catenin immunoreactivity was seen in the nuclei of B cells (Fig. 3I). The colocalization of SF-1, Smad3, and -catenin suggests that both the TGF and Wnt signaling pathways operate in B cells (42). To assess the proliferation of neoplastic cells, we performed immunostaining for proliferating cell nuclear antigen (PCNA). This antigen was detected in both neoplastic and nonneoplastic regions. In the neoplastic tissue, it was expressed more prominently in B cells than in A cells (Fig. 3J). Thus, B cells, even though differentiated and steroidogenic, continue to proliferate in these neoplastic regions.
Hormonal changes in gonadectomized NU/J nude mice
The transition from adrenal- to gonadal-type steroidogenesis was verified by hormone measurements in serum of mice 2 months after sham surgery or gonadectomy (Table 2). Some nude mouse strains have been reported to have reduced plasma levels of gonadotropins due to impaired pituitary secretion, which limits sex steroid synthesis in the ovary (22, 43). However, we found that the baseline and postgonadectomy levels of LH in NU/J nude mice were comparable with those seen in DBA/2J or FVB/N mice (19), increasing from 0.36 ±.02 ng/ml in sham-operated females to 3.6 ± 1.7 ng/ml after gonadectomy (P < 0.05). Levels of estradiol were also higher than expected for nude mice (22, 43), amounting to 27 ± 5.7 pg/ml in sham-operated vs. 183 ± 61 pg/ml in ovariectomized females (P < 0.05). Testosterone was at the lower limit of detection for the assay, and only a modest increase was observed after gonadectomy. Collectively, these measurements confirmed the capacity of neoplastic cells in the adrenal cortex to become a source of gonadal hormones after gonadectomy. In contrast to sex steroids, corticosterone levels were reduced from 86 ± 10 ng/ml in intact nude mice to 56 ± 5.5 ng/ml (P < 0.05) in gonadectomized animals, suggesting a diminished capacity to produce glucocorticoids, presumed to reflect replacement of a large portion of adrenal cortex by neoplastic cells without compensatory growth or increased activity in the remaining adrenal tissue. We assume that the presence of LHR on the neoplastic B cells allows these cells to respond to elevated LH levels and produce sex steroids. Conversely, the lack of LHR expression on normal adrenocortical cells together with a decrease in serum corticosterone levels indicates that LH does not supplant ACTH in the stimulation of glucocorticoid production.
Hormonal changes in hCG-treated NU/J nude mice
To assess the adrenal response to elevated levels of gonadotropins in the presence of ovaries, we took advantage of the ability of NU/J nude mice to accept xenografts. CHO cells stably expressing a single-chain variant of hCG known to activate LHR (24) or control CHO cells were injected sc into the flanks of intact or gonadectomized female NU/J nude mice. In preliminary studies we verified that conditioned media from hCG-CHO and control CHO cells did not contain detectable levels of estradiol or testosterone. Inoculated mice were killed after 3–7 wk, when flank tumors were approximately 1 cm in diameter. The mean levels of hCG in serum at a time of killing were 31 ± 14 IU/ml in the intact hCG-treated mice. LH levels reached 2.1 ± 0.4 ng/ml in these mice, which we attributed to cross-reactivity of the assay with highly elevated levels of hCG rather than to a true increase in circulating LH. Despite individual variability in serum hCG levels, all of the hCG-treated mice exhibited an increase in circulating sex steroids. Whereas the levels of estradiol (174 ± 42 pg/ml) were comparable with those seen in gonadectomized nude mice, the levels of testosterone (320 ± 60 ng/dl) were significantly higher in the hCG-treated intact mice. A similar increase in serum testosterone has been reported in bovine LH-CTP transgenic mice (13). In response to hCG, serum corticosterone levels in nongonadectomized mice reached 128 ± 3.5 ng/ml, which was higher than observed in gonadectomized nude mice.
In mice subjected to both gonadectomy and xenografting, hCG levels were comparable with those in intact hCG-treated mice (Table 2). Adrenocortical tumors that formed in these animals did not differ with regard to morphology or expression of GATA-4, compared with mice treated with gonadectomy alone (data not shown). However, serum testosterone levels (in this case reflecting extragonadal production) were 2.5 times higher in the mice treated with hCG and gonadectomy than in mice treated with gonadectomy alone (50 ± 7.5 ng/dl) (Table 2).
Neoplastic changes in adrenals from hCG-treated NU/J nude mice
To determine the consequences of chronic hCG elevation on the adrenal cortex, we harvested adrenal glands from NU/J nude mice 3 wk after inoculation with hCG-CHO cells. The adrenal glands of intact NU/J nude mice inoculated with control CHO cells showed rare subcapsular GATA-4-positive cells but no frank neoplasia (data not shown). In contrast, the adrenal glands of hCG-treated mice showed neoplastic changes that resembled those seen in ovariectomized animals. As in gonadectomized mice, GATA-4-expressing A and B cells were observed in neoplastic adrenocortical tissue of hCG-treated intact mice (Fig. 4, A and F). Inhibin- immunoreactivity was evident in B cells found in the adrenals of both gonadectomized (Fig. 4B) and hCG-treated (Fig. 4G) mice; however, in the latter case, the inhibin--producing cells appeared to have less cytoplasm and inhibin- had a more lacy distribution. This might have reflected a temporal delay in differentiation of the B cells or a reduction in inhibin- expression in the hCG-treated adrenals. Friend-of-GATA-2 (Fog-2), a cofactor known to interact with GATA-4 during the differentiation of sex steroidogenic cells (37, 44), was detected in the nuclei of A and B cells (Fig. 4, C and H). In addition to nuclear staining, we observed some cytoplasmic FOG-2 immunoreactivity of unclear significance. The overlapping expression patterns of GATA-4 and FOG-2 suggested that these two factors may interact to maintain the gonadal steroidogenic phenotype in the neoplastic tissue. We also examined expression of PCNA, which in both gonadectomized (Fig. 4D) and hCG-treated (Fig. 4I) adrenals was expressed in B cells and in a subset of normal adrenal cells, as was the case at 4 months after gonadectomy (Fig. 3J). There was little or no PCNA detected in A cells, except those directly under the adrenal capsule, suggesting that tumor expansion occurs primarily through growth and proliferation of B at this stage of tumor development. In the adrenals of gonadectomized mice, ER immunoreactivity was observed mainly in neoplastic areas (Fig. 4E), whereas in hCG-treated mice, ER was expressed more broadly, localizing to both B cells and normal adrenal cells (Fig. 4J).
hCG-induced changes in the ovaries of NU/J nude mice may limit sex steroid production by neoplastic adrenocortical tissue
The ovarian changes induced by inoculation of hCG-CHO cells mimicked those seen in transgenic mice that constitutively express - and -subunits of hCG (45) or bovine LH-CTP (46). The ovaries of hCG-CHO mice were enlarged, cystic and hemorrhagic with thecal layer expansion and stromal cell proliferation (Fig. 5B), whereas the ovaries from mice inoculated with wild-type CHO cells appeared normal (Fig. 5A). Abundant GATA-4 immunoreactivity was evident in the stimulated ovaries. Normally expression of GATA-4 in postmitotic cells of the corpus luteum is weak, whereas another member of the GATA family, GATA-6, is expressed in both follicular cells and corpus luteum (47). Thus, the histological changes in hCG-treated ovaries are more reminiscent of a luteoma than physiologic luteinization. RNase protection assays on ovarian RNA from individual nude mice showed no significant difference in the levels of GATA-6 mRNA (Fig. 5B, lanes 1 and 2 vs. 3 and 4) or P450c17 mRNA (Fig 6B, lanes 5 and 6 vs. 7 and 8) in response to hCG. However, the levels of LHR mRNA were much higher in hCG-treated ovaries, compared with those from mice treated with control CHO cells (Fig. 5B, lanes 5 and 6 vs. 7 and 8).
To assess the steroidogenic capacity of hCG-induced neoplastic cells in the adrenal gland, we measured the levels of P450c17 and LHR transcripts using RNase protection assays. Both transcripts were present in adrenals from gonadectomized NU/J nude mice (Fig. 6A, lanes 1 and 2). hCG treatment of gonadectomized mice caused a 2-fold decrease in the levels of mRNA for both factors (Fig. 6A, lanes 3 and 4). Thus, the effects of gonadectomy and hCG were not additive over the 3- to 7-wk period of treatment; rather, hCG attenuated the impact of gonadectomy on adrenal sex steroidogenic activity. Surprisingly, neither P450c17 nor LHR was detected in adrenals from intact mice treated with hCG-CHO (Fig. 6A, lanes 7 and 8) or control CHO (Fig. 6A, lanes 5 and 6) cells. To compare the relative expression of P450c19 mRNA in adrenals from mice subjected to gonadectomy and/or hCG-treatment, we performed RT-PCR analysis using the housekeeping gene Hprt as a control. Expression of P450c19 is tissue-specifically regulated by alternative first exons and promoters (30), and the ovarian-specific first exon product was detected in the adrenals of gonadectomized and to a lesser extent hCG-treated mice (Fig. 6B). Use of this tissue-specific first exon underscores the gonadal character of the neoplastic steroidogenic tissue. The limited steroidogenic capacity of B cells in neoplastic adrenals of hCG-treated mice may be a consequence of decreased of adrenal LHR expression in response to chronic gonadotropin stimulation. Chronic hCG exposure apparently did not affect normal adrenocortical function, as reflected by slightly increased levels of corticosterone in hCG-treated intact mice.
Discussion
The adrenocortical lesions that appear in gonadectomized inbred mice are thought to reflect metaplasia of competent cells in the subcapsular region, which transform into gonadal-like stroma in response to changes in the hormonal milieu. Like other classic examples of metaplasia (e.g. Barrett’s esophagus), gonadectomy-induced adrenocortical metaplasia arises in a self-renewing epithelium, is caused by chronic hormonal stimulation or tissue injury, and has the potential to evolve into frank neoplasia (48). It is unclear whether gonadectomy-induced adrenocortical metaplasia represents transdifferentiation, the direct conversion of one differentiated cell type into another, or instead is an example of stem cell metaplasia (48).
Our new NU/J nude mouse model affords insights into the molecular events underlying this form of tissue-type switching and sheds light on the developmental mechanisms that control normal steroidogenic cell differentiation. We used two approaches to induce metaplasia in the adrenal cortex of these inbred nude mice: gonadectomy and xenografts of hCG-secreting cells. By characterizing the adrenal glands from these animals, we have expanded the list of molecular markers that delineate the different cell populations within the neoplastic adrenocortical tissue. Based on their morphology, marker expression (GATA-4, MIS, and MISIIR) and low PCNA-labeling index, we propose that A cells resemble stromal cells of the postmenopausal ovary, which are known to metabolize cholesterol but have limited capacity for steroid hormone biosynthesis, particularly sex steroids (49). B cells, which have a relatively high PCNA-labeling index and express GATA-4, ER, LHR, P450c17, and P450c19, resemble follicular theca cells from patients with polycystic ovary syndrome, a condition marked by chronically elevated LH levels (50, 51). The interrelationship between A and B cells and their cell(s) of origin remain enigmatic. We (19) and others (11) have hypothesized that nonsteroidogenic A cells transform into B cells based on the circumstantial evidence that A cells precede B cells, that B cells always arise in the company of A cells, and that both cell types express GATA-4. However, direct evidence that B cells are descendants of A cells is lacking. Lineage tracing studies are needed to definitively establish the origin of these cell types.
NU/J nude mice develop adrenocortical neoplasms in response to either gonadectomy or xenografts of hCG-secreting cells, suggesting that chronic gonadotropin elevation is the principal signal responsible for this process. Although the neoplastic lesions in hCG-treated intact nude mice morphologically resemble those seen in their gonadectomized counterparts, they have reduced sex steroidogenic capacity, implying that gonadotropin elevation does not explain all aspects of the phenotypic switch. The mechanism responsible for this limited adrenal sex steroidogenic capacity is unknown but may involve decreased LH signaling through down-regulation of LHR in the setting of persistently elevated hCG (52) or hCG-induced degradation of LHR mRNA (53). hCG stimulates ovarian production of estradiol and testosterone, and high levels of these gonadal hormones might directly inhibit steroidogenesis in neoplastic adrenocortical cells. Sex steroids have been shown to act directly on gonadal steroidogenic cells to inhibit P450c17 (54, 55).
Our nude mouse model suggests that multiple signaling pathways influence the adrenocortical changes that accompany gonadectomy. Ectopic expression of both MIS and MISIIR was observed early in tumorigenesis in subcapsular areas devoid of normal adrenocortical markers, suggesting that this ligand-receptor pair may be involved in the switch to gonadal-type differentiation. Perhaps MIS, acting in a paracrine fashion, inhibits proliferation or steroidogenesis in A cells or facilitates their centripetal migration. MIS expression has been demonstrated in gonadal tumors, and genetic studies suggest that inhibin- and MIS exert a synergistic effect on tumorigenesis in gonads. In inhibin- null mice, the absence of MIS causes earlier onset and more aggressive development of testicular tumors (56). The detection of SF-1, -catenin, and Smad3 in B cells suggests potential involvement of TGF and Wnt signaling pathways in mechanisms for phenotypic switching, including down-regulation of ACTH signaling and up-regulation of gonadal steroidogenesis. Based on the selective expression of the above markers in A or B cells, we assume that gonadectomy-induced adrenocortical tumorigenesis involves multiple steps and that different sets of signals are required for A vs. B cell differentiation.
Not all inbred mouse strains respond to prepubertal gonadectomy by developing sex steroid-producing adrenocortical tumors. Inbred strain susceptibility to gonadectomy-induced adrenocortical tumorigenesis appears to correlate with the in vivo and in vitro capacity of normal gonadal tissue to produce steroid hormones in response to unopposed gonadotropin stimulation (19). In DBA/2J mice, this may be a consequence of a polymorphism in the SF-1 coding sequence (17), whereas in NU/J nude mice, it may be due to the effects of athymia on function of the hypothalamus-pituitary-gonadal/hypothalamus-pituitary-adrenal axes (22, 23). Gonadectomy-induced adrenocortical neoplasia in mice and ferrets may be the byproduct of an organism’s efforts to compensate for unfavorable conditions. In DBA/2J and NU/J nude mice, these neoplastic changes are benign. Inbred strains that develop malignant lesions after gonadectomy (i.e. CE/J) may have genetic changes that either exacerbate adrenocortical hyperplasia or promote malignant transformation.
As in mice and ferrets, adrenocortical tumor formation in humans appears to be influenced by both the hormonal milieu and the inherent susceptibility of the adrenal gland. Subcapsular adrenocortical neoplasms with histological features resembling luteinized ovarian stroma (thecal metaplasia) have been reported in postmenopausal women (1, 2, 3, 4) and men with acquired testicular atrophy (5), suggesting that the hormone changes associated with gonadal failure (e.g. elevated LH, decreased sex steroids, etc.) may contribute to adrenal tumor development in humans. Analysis of heritable and spontaneous types of human adrenocortical tumors has documented alterations in either cell surface receptors or their downstream effectors that make adrenocortical tissue susceptible to hormone-associated neoplastic transformation. In a majority of cases, these genetic alterations lead to adrenocortical neoplasms that produce excess glucocorticoids, but in a few cases, notably those involving young children, the tumors produce sex steroids and inhibin-, like the adrenocortical tumors that develop in prepubertally gonadectomized mice and ferrets.
The nude mouse model described in this paper should prove useful in characterizing the roles of gonadotropins, sex steroids, and other hormones in adrenocortical metaplasia and tumorigenesis.
Acknowledgments
We thank Keith Parker, Perrin White, and Jorma Toppari for providing plasmid constructs. We thank the Digestive Diseases Research Core Center staff for assistance with immunohistochemistry. We thank Susan Porter-Tinge for technical assistance.
Footnotes
This work was supported by National Institutes of Health Grants HL61006 and DK52574, March of Dimes FY02-203, Barnes-Jewish Foundation, Mallinckrodt Foundation, Finnish Pediatric Research Foundation, and Juselius Foundation.
Abbreviations: CHO, Chinese hamster ovary; ER, estrogen receptor; hCG, human chorionic gonadotropin; HPRT, hypoxanthine-guanine phosphoribosyl transferase; LHR, LH receptor; MC2R, melanocortin-2 receptor; MIS, Müllerian inhibitory substance; MISIIR, MIS receptor type II; 21-OHase, 21-hydroxylase; PCNA, proliferating cell nuclear antigen; SF-1, steroidogenic factor-1.
References
Fidler WJ 1977 Ovarian thecal metaplasia in adrenal glands. Am J Clin Pathol 67:318–323
Werk Jr EE, Sholiton LE, Kalejs L 1973 Testosterone-secreting adrenal adenoma under gonadotropin control. N Engl J Med 289:767–770
Givens JR, Andersen RN, Wiser WL, Coleman SA, Fish SA 1974 A gonadotropin-responsive adrenocortical adenoma. J Clin Endocrinol Metab 38:126–133
Carney JA 1987 Unusual tumefactive spindle-cell lesions in the adrenal glands. Hum Pathol 18:980–985
Romberger CF, Wong TW 1989 Thecal metaplasia in the adrenal gland of a man with acquired bilateral testicular atrophy. Arch Pathol Lab Med 113:1071–1075
Thompson GB, Young Jr WF 2003 Adrenal incidentaloma. Curr Opin Oncol 15:84–90
Bertherat J, Mosnier-Pudar H, Bertagna X 2002 Adrenal incidentalomas. Curr Opin Oncol 14:58–63
Lacroix A, Baldacchino V, Bourdeau I, Hamet P, Tremblay J 2004 Cushing’s syndrome variants secondary to aberrant hormone receptors. Trends Endocrinol Metab 15:375–382
Russfield AB 1975 Experimental endocrinopathies. Methods Achiev Exp Pathol 7:132–148
Schoemaker NJ, Teerds KJ, Mol JA, Lumeij JT, Thijssen JH, Rijnberk A 2002 The role of luteinizing hormone in the pathogenesis of hyperadrenocorticism in neutered ferrets. Mol Cell Endocrinol 197:117–125
Firth CH, Dunn TB 1994 Tumours of the adrenal gland. IARC Sci Publ 111:595–609
Beuschlein F, Looyenga BD, Bleasdale SE, Mutch C, Bavers DL, Parlow AF, Nilson JH, Hammer GD 2003 Activin induces x-zone apoptosis that inhibits luteinizing hormone-dependent adrenocortical tumor formation in inhibin-deficient mice. Mol Cell Biol 23:3951–3964
Mikola M, Kero J, Nilson JH, Keri RA, Poutanen M, Huhtaniemi I 2003 High levels of luteinizing hormone analog stimulate gonadal and adrenal tumorigenesis in mice transgenic for the mouse inhibin- subunit promoter/Simian virus 40 T-antigen fusion gene. Oncogene 22:3269–3278
Murthy AS, Brezak MA, Baez AG 1970 Postcastrational adrenal tumors in two strains of mice: morphologic, histochemical, and chromatographic studies. J Natl Cancer Inst 45:1211–1222
Huseby RA, Bittner JJ 1951 Differences in adrenal responsiveness to post-castrational alteration as evidenced by transplanted adrenal tissue. Cancer Res 11:954–961
Stalvey JR, Payne AH 1983 Luteinizing hormone receptors and testosterone production in whole testes and purified Leydig cells from the mouse: differences among inbred strains. Endocrinology 112:1696–1701
Frigeri C, Tsao J, Cordova M, Schimmer BP 2002 A polymorphic form of steroidogenic factor-1 is associated with adrenocorticotropin resistance in Y1 mouse adrenocortical tumor cell mutants. Endocrinology 143:4031–4037
Morley SD, Viard I, Chung BC, Ikeda Y, Parker KL, Mullins JJ 1996 Variegated expression of a mouse steroid 21-hydroxylase/-galactosidase transgene suggests centripetal migration of adrenocortical cells. Mol Endocrinol 10:585–598
Bielinska M, Parviainen H, Porter-Tinge SB, Kiiveri S, Genova E, Rahman N, Huhtaniemi IT, Muglia LJ, Heikinheimo M, Wilson DB 2003 Mouse strain susceptibility to gonadectomy-induced adrenocortical tumor formation correlates with the expression of GATA-4 and luteinizing hormone receptor. Endocrinology 144:4123–4133
Peterson RA, Kiupel M, Bielinska M, Kiiveri S, Heikinheimo M, Capen CC, Wilson DB 2004 Transcription factor GATA-4 is a marker of anaplasia in adrenocortical neoplasms of the domestic ferret (Mustela putorius furo). Vet Pathol 41:446–449
Balciunaite G, Keller MP, Balciunaite E, Piali L, Zuklys S, Mathieu YD, Gill J, Boyd R, Sussman DJ, Hollander GA 2002 Wnt glycoproteins regulate the expression of FoxN1, the gene defective in nude mice. Nat Immunol 3:1102–1108
Rebar RW, Morandini IC, Erickson GF, Petze JE 1981 The hormonal basis of reproductive defects in athymic mice: diminished gonadotropin concentrations in prepubertal females. Endocrinology 108:120–126
Rebar RW, Morandini IC, Petze JE, Erickson GF 1982 Hormonal basis of reproductive defects in athymic mice: reduced gonadotropins and testosterone in males. Biol Reprod 27:1267–1276
Sugahara T, Pixley MR, Minami S, Perlas E, Ben Menahem D, Hsueh AJ, Boime I 1995 Biosynthesis of a biologically active single peptide chain containing the human common alpha and chorionic gonadotropin beta subunits in tandem. Proc Natl Acad Sci USA 92:2041–2045
Ben Menahem D, Jablonka-Shariff A, Hyde RK, Pixley MR, Srivastava S, Berger P, Boime I 2001 The position of the and subunits in a single chain variant of human chorionic gonadotropin affects the heterodimeric interaction of the subunits and receptor-binding epitopes. J Biol Chem 276:29871–29879
Haavisto AM, Pettersson K, Bergendahl M, Perheentupa A, Roser JF, Huhtaniemi I 1993 A supersensitive immunofluorometric assay for rat luteinizing hormone. Endocrinology 132:1687–1691
Anttonen M, Ketola I, Parviainen H, Pusa AK, Heikinheimo M 2003 FOG-2 and GATA-4 are coexpressed in the mouse ovary and can modulate Müllerian-inhibiting substance expression. Biol Reprod 68:1333–1340
Milstone DS, Shaw SK, Parker KL, Szyf M, Seidman JG 1992 An element regulating adrenal-specific steroid 21-hydroxylase expression is located within the slp gene. J Biol Chem 267:21924–21927
Mishina Y, Rey R, Finegold MJ, Matzuk MM, Josso N, Cate RL, Behringer RR 1996 Genetic analysis of the Müllerian-inhibiting substance signal transduction pathway in mammalian sexual differentiation. Genes Dev 10:2577–2587
Honda S, Harada N, Takagi Y 1996 The alternative exons 1 of the mouse aromatase cytochrome P-450 gene. Biochim Biophys Acta 1305:145–150
Kim JS, Kubota H, Kiuchi Y, Doi K, Saegusa J 1997 Subcapsular cell hyperplasia and mast cell infiltration in the adrenal cortex of mice: comparative study in 7 inbred strains. Exp Anim 46:303–306
Jamin SP, Arango NA, Mishina Y, Hanks MC, Behringer RR 2002 Requirement of Bmpr1a for Müllerian duct regression during male sexual development. Nat Genet 32:408–410
Fynn-Thompson E, Cheng H, Teixeira J 2003 Inhibition of steroidogenesis in Leydig cells by Mullerian-inhibiting substance. Mol Cell Endocrinol 211:99–104
Keeney DS, Jenkins CM, Waterman MR 1995 Developmentally regulated expression of adrenal 17-hydroxylase cytochrome P450 in the mouse embryo. Endocrinology 136:4872–4879
Rahman NA, Kiiveri S, Rivero-Muller A, Levallet J, Vierre S, Kero J, Wilson DB, Heikinheimo M, Huhtaniemi I 2004 Adrenocortical tumorigenesis in transgenic mice expressing the inhibin -subunit promoter/SV40 virus T-antigen transgene: Relationship between ectopic expression of luteinizing hormone receptor and transcription factor GATA-4. Mol Endocrinol 18:2553–2569
Rey R, Sabourin JC, Venara M, Long WQ, Jaubert F, Zeller WP, Duvillard P, Chemes H, Bidart JM 2000 Anti-Müllerian hormone is a specific marker of Sertoli- and granulosa-cell origin in gonadal tumors. Hum Pathol 31:1202–1208
Tevosian SG, Albrecht KH, Crispino JD, Fujiwara Y, Eicher EM, Orkin SH 2002 Gonadal differentiation, sex determination and normal Sry expression in mice require direct interaction between transcription partners GATA4 and FOG2. Development 129:4627–4634
Tremblay JJ, Viger RS 1999 Transcription factor GATA-4 enhances Müllerian inhibiting substance gene transcription through a direct interaction with the nuclear receptor SF-1. Mol Endocrinol 13:1388–1401
Feng ZM, Wu AZ, Chen CL 1998 Testicular GATA-1 factor up-regulates the promoter activity of rat inhibin -subunit gene in MA-10 Leydig tumor cells. Mol Endocrinol 12:378–390
Ketola I, Rahman N, Toppari J, Bielinska M, Porter-Tinge SB, Tapanainen JS, Huhtaniemi IT, Wilson DB, Heikinheimo M 1999 Expression and regulation of transcription factors GATA-4 and GATA-6 in developing mouse testis. Endocrinology 140:1470–1480
Shi Y, Massague J 2003 Mechanisms of TGF- signaling from cell membrane to the nucleus. Cell 113:685–700
Gummow BM, Winnay JN, Hammer GD 2003 Convergence of Wnt signaling and steroidogenic factor-1 (SF-1) on transcription of the rat inhibin gene. J Biol Chem 278:26572–26579
Rebar RW, Morandini IC, Benirschke K, Petze JE 1980 Reduced gonadotropins in athymic mice: prevention by thymic transplantation. Endocrinology 107:2130–2132
Tremblay JJ, Viger RS 2003 Novel roles for GATA transcription factors in the regulation of steroidogenesis. J Steroid Biochem Mol Biol 85:291–298
Matzuk MM, DeMayo FJ, Hadsell LA, Kumar TR 2003 Overexpression of human chorionic gonadotropin causes multiple reproductive defects in transgenic mice. Biol Reprod 69:338–346
Mann RJ, Keri RA, Nilson JH 2003 Consequences of elevated luteinizing hormone on diverse physiological systems: use of the LHCTP transgenic mouse as a model of ovarian hyperstimulation-induced pathophysiology. Recent Prog Horm Res 58:343–375
Heikinheimo M, Ermolaeva M, Bielinska M, Rahman NA, Narita N, Huhtaniemi IT, Tapanainen JS, Wilson DB 1997 Expression and hormonal regulation of transcription factors GATA-4 and GATA-6 in the mouse ovary. Endocrinology 138:3505–3514
Tosh D, Slack JM 2002 How cells change their phenotype. Nat Rev Mol Cell Biol 3:187–194
Jabara S, Christenson LK, Wang CY, McAllister JM, Javitt NB, Dunaif A, Strauss III JF 2003 Stromal cells of the human postmenopausal ovary display a distinctive biochemical and molecular phenotype. J Clin Endocrinol Metab 88:484–492
Kaaijk EM, Sasano H, Suzuki T, Beek JF, van Der Veen F 2000 Distribution of steroidogenic enzymes involved in androgen synthesis in polycystic ovaries: an immunohistochemical study. Mol Hum Reprod 6:443–447
Katulski K, Bornstein S, Figiel M, Wand D, Warenik-Szymankiewicz A, Trzeciak WH 1998 Typical hormonal profiles are accompanied by increased immunoreactivity of theca folliculi steroid 17-hydroxylase P450 in polycystic ovary syndrome. J Endocrinol Invest 21:304–309
Ascoli M, Fanelli F, Segaloff DL 2002 The lutropin/choriogonadotropin receptor, a 2002 perspective. Endocr Rev 23:141–174
Wang L, Menon KM 2005 Regulation of luteinizing hormone/chorionic gonadotropin receptor messenger ribonucleic acid expression in the rat ovary: relationship to cholesterol metabolism. Endocrinology 146:423–431
Johnson DC, Crane LH 1995 Inhibitory and stimulatory effect of oestrogens upon ovarian 17-hydroxylase/C17,20-lyase in immature hypophysectomized rats treated with gonadotrophin. J Endocrinol 145:59–67
Magoffin DA, Erickson GF 1982 Direct inhibitory effect of estrogen on LH-stimulated androgen synthesis by ovarian cells cultured in defined medium. Mol Cell Endocrinol 28:81–89
Matzuk MM, Finegold MJ, Mishina Y, Bradley A, Behringer RR 1995 Synergistic effects of inhibins and müllerian-inhibiting substance on testicular tumorigenesis. Mol Endocrinol 9:1337–1345(Malgorzata Bielinska, Ele)
Program for Developmental and Reproductive Biology (H.P., S.K., M.H.), Children’s Hospital, Biomedicum Helsinki, University of Helsinki, 00290 Helsinki, Finland
Department of Physiology (N.R.), Institute of Biomedicine, University of Turku, 20520 Turku, Finland
Department of Physiology (J.L.), University of Oulu, 90220 Oulu, Finland
Abstract
In response to prepubertal gonadectomy certain inbred mouse strains, including DBA/2J, develop sex steroid-producing adrenocortical neoplasms. This phenomenon has been attributed to a lack of gonadal hormones or a compensatory increase in gonadotropins. To assess the relative importance of these mechanisms, we created a new inbred model of adrenocortical neoplasia using female NU/J nude mice. These mice developed adrenocortical neoplasms in response to either gonadectomy or gonadotropin elevation from xenografts of human chorionic gonadotropin (hCG)-secreting Chinese hamster ovary cells. In each instance the adrenal tumors resembled the neoplasms found in gonadectomized DBA/2J mice and were composed of spindle-shaped A cells and lipid-laden B cells. Both cell populations were defined by ectopic expression of GATA-4 and an absence of the adrenocortical markers melanocortin-2-receptor and steroid 21-hydroxylase, but only B cells expressed the gonadal steroidogenic markers inhibin-, LH receptor, P450c17, and P450c19. Expression of sex steroidogenic markers was attenuated in the neoplastic adrenal cortex of hCG-treated vs. gonadectomized mice. Whereas neoplastic adrenals were an obvious source of estradiol in gonadectomized mice, ovaries appeared to be the major source of this hormone in hCG-treated mice. Gonadectomy and hCG treatment elicited comparable increases in serum estradiol, but testosterone levels increased significantly only in hCG-treated mice. We conclude that chronic gonadotropin elevation, caused by either gonadectomy or hCG administration, signals a population of cells in the adrenal subcapsular region of permissive mice to undergo differentiation along a gonadal rather than an adrenal lineage. Thus, NU/J nude mice can be used as a model to study both neoplasia and adrenogonadal lineage specification.
Introduction
SUBCAPSULAR ADRENOCORTICAL neoplasms may arise in the setting of gonadal failure and chronic gonadotropin elevation, as in postmenopausal women (1, 2, 3, 4) and men with acquired testicular atrophy (5). Some of these adrenal tumors resemble gonadal stroma and produce sex steroids in response to continuous gonadotropin stimulation. Incidental adrenocortical tumors are present in up to 6% of older individuals (6, 7), and a subset of these tumors is gonadotropin responsive (7, 8). Sex steroid-producing adrenocortical neoplasms also occur in neutered domestic animals, including pet ferrets (9, 10). A lack of gonadal hormones or a compensatory increase in gonadotropins is postulated to drive adrenocortical neoplasia, although the relative importance of these mechanisms remains controversial (10, 11, 12, 13).
Inbred laboratory mice make a good model to study this phenomenon. Certain mouse strains (e.g. DBA/2J, CE) develop sex steroid-producing adrenocortical neoplasms in response to prepubertal gonadectomy (11, 14). Organ transplantation experiments have established that the adrenal glands of susceptible strains exhibit an inherent predisposition to tumor formation in response to the hormonal changes that accompany gonadectomy (15). The genetic basis of strain susceptibility is unknown but seems to correlate with reduced fertility and a polymorphism in the coding region of steroidogenic factor-1 (SF-1), which may affect steroidogenesis both in adrenal cortex and gonads (16, 17). Neoplasms in the adrenals of these mice arise at multiple foci in the subcapsular region and invade the underlying zones of adrenal cortex in a fashion reminiscent of the centripetal pattern of adrenal renewal (18). The tumors are characterized by nests of steroidogenic cells, termed B cells, which appear amid ovoid- or spindle-shaped cells, known as A cells (11, 14).
Previously we characterized adrenocortical neoplasms in gonadectomized DBA/2J mice and ferrets and showed that the gene products GATA-4, LH receptor (LHR), and inhibin- are hallmarks of these tumors (19, 20). In the present study, we have developed a new inbred model of adrenocortical neoplasia to assess the relative contributions of gonadectomy vs. gonadotropin elevation to the induction and progression of adrenocortical neoplasia. Our new model employs NU/J mice, an inbred strain carrying nude (Foxn1nu), a recessive loss-of-function mutation in the gene encoding Foxn1, which is associated with athymia (21). Nude (Foxn1nu/Foxn1nu) mice are widely used in cell and organ transplantation experiments. Early studies with different strains of nude or thymectomized mice demonstrated deficiencies in reproductive function, including reduced fertility and decreased sex steroid production (22, 23). In light of these reproductive abnormalities, we postulated that athymic nude mice might be predisposed to gonadectomy-induced adrenocortical tumorigenesis. Here we show that NU/J nude mice, like strain DBA/2J, respond to gonadectomy by increasing serum gonadotropin levels and developing sex steroid-producing neoplasms in the adrenal cortex and that similar neoplastic changes occur in intact nude mice after chronic gonadotropin stimulation.
Materials and Methods
Experimental animals and surgery
All animal work was carried out in accordance with National Institutes of Health Guidelines for the Care and Use of Laboratory Animals and was approved by the institutional animal studies committee. Weanling mice homozygous for the Foxn1nu mutation on an NU/J background were purchased from The Jackson Laboratory (Bar Harbor, ME). The mice were specific pathogen free and housed five per cage in controlled conditions of light (12 h light, 12 h darkness) and temperature (21 C). They were fed commercial mouse chow (Purina) and tap water ad libitum. Female mice (3–4 wk of age) were anesthetized with ip administration of ketamine (75 mg/kg), xylazine (4 mg/kg), and acepromazine (0.75 mg/kg) and subjected to gonadectomy or sham surgery. At specified times the mice were killed by CO2 inhalation and blood samples were collected. Harvested tissue was fixed overnight in 4% paraformaldehyde, embedded in paraffin, and then sectioned (4 μm) for histochemical staining (hematoxylin-eosin or toluidine blue) or immunohistochemistry. For in situ hybridization isolated adrenals were frozen in OCT (Tissue-Tek, Torrance, CA). TRIzol reagent (Invitrogen, Carlsbad, CA) was used to isolate total RNA for RNase protection and RT-PCR assays.
Cell culture and inoculations
Chinese hamster ovary (CHO) cells stably expressing a biologically active single-chain human chorionic gonadotropin (hCG) variant (hCG-CHO cells) were prepared as described elsewhere (24, 25). hCG-CHO cells or control CHO cells were maintained in Ham’s F-12 medium [supplemented with penicillin (100 U/ml), streptomycin (100 μg/ml), and 2 mM glutamine] containing 5% fetal bovine serum and G418 (125 μg/ml) at 37 C in a humidified atmosphere of 5% CO2-95% air. To inoculate nude mice, hCG-CHO or control CHO cells were harvested with trypsin, washed in PBS, resuspended in Ham’s F-12 medium (without serum) at a concentration of 3 x 107 cells/ml. Immediately after gonadectomy or sham surgery, nude mice were injected sc in each flank with 0.2 ml of this cell suspension.
Hormone measurements
Serum LH concentrations were measured using an immunofluorometric assay (Wallac Oy, Turku, Finland), as described previously (26). Serum levels of hCG and testosterone were measured using commercial assay kits (Diagnostic Products Corp., Los Angeles, CA). Serum estradiol levels were determined by RIA (Spectria, Orion Diagnostica, Espoo, Finland). Corticosterone levels were measured by RIA (IDS, Boldon, UK). Data reported are expressed as the mean ± SEM. Statistical significance was determined using Student’s t test in Microsoft Excel data analysis software (Microsoft Corp., Seattle, WA).
Immunohistochemistry
Paraffin-embedded, paraformaldehyde-fixed tissue sections were deparaffinized and rehydrated (27). Endogenous peroxidase was quenched by 30 min incubation in 1% H2O2 in PBS. Slides were then treated with avidin/biotin blocking (SP-2001; Vector Laboratories, Burlingame, CA) for 20 min, rinsed in three changes of PBS, and then treated with protein block (X0909; Dako, Carpinteria, CA) for 10 min. Sections were incubated overnight at 4 C with primary antibodies (Table 1). After rinsing for 30 min in PBS, secondary antibody (Table 1) was applied for 30 min at 30 C. After rinsing in PBS, an avidin-biotin immunoperoxidase system (Vectastain Elite ABC kit; Vector Laboratories) and diaminobenzidine (Sigma Chemicals, St. Louis, MO) were used to visualize the bound antibody (27). Slides were counterstained with Gill’s hematoxylin.
In situ hybridization
Tissue cryosections (10 μm) were subjected to in situ hybridization as described (19). [33P]-labeled antisense riboprobes were prepared using the following linearized templates: 1) melanocortin-2 receptor (MC2R), BamHI-digest pCRII plasmid containing nucleotides 23–663 of the cDNA (12), T7 polymerase; 2) steroid 21-hydroxylase (21-OHase), EcoRI digest of a cDNA, T3 polymerase (28); 3) Müllerian inhibitory substance (MIS) receptor II, EcoRV-digested pCRII plasmid containing nucleotides 681-1080 of the cDNA plasmid, Sp6 polymerase (29); 4) P450c17, as described elsewhere (19).
RNase protection assays
These assays were performed with a commercially available kit, RPA III (Ambion, Austin, TX) and 5 μg of total RNA. The linearized templates used have been described previously (19).
RT-PCR assays
Reverse transcription was performed with 3 μg of total RNA and random hexamer primers (Promega, Madison, WI). After 10 min incubation at 65 C, the reaction mixture was supplemented with 5x avian myeloblastosis virus-reverse transcription buffer (Fisher Scientific, Pittsburgh, PA), MgCl2 to 25 mM, and deoxynucleotide triphosphates to a final concentration of 2 mM each, 80 U RNase inhibitor (Promega), and 20 U avian myeloblastosis virus reverse transcriptase (Stratagene, La Jolla, CA), and the reaction was run for 1 h at 42 C in the final volume of 40 μl. The quality of each cDNA was determined by the relative level of amplification of the mouse Hprt gene. The following cDNA primers were used: P450c19 (ovarian-specific alternative first exon), forward primer (+1 to +22) 5'-ACAGCATTGTGATTGTCCCTCT-3', reverse primer (+324 to +344) 5'-CATCTTGCGCTATTTGGCCTC-3' (30); hypoxanthine-guanine phosphoribosyl transferase (HPRT), forward primer (+531 to +539) 5'-GCTGGTGAAAAGGACCTCT-3', reverse primer (+750 to +769 5'-CACAGGACTAGAACACCTGC-3', GenBank accession no. BC083145.1. The PCR contained 2.5% of the cDNA product, Pfu buffer (Stratagene), 20 pmol of each primer, and 1 U of Pfu polymerase (Stratagene) in a final volume of 20 μl. The thermal cycler conditions were: 94 C (4 min), 1 cycle; 94 C (1 min), 57 C (1 min) for P450c19, and 56 C (1 min) for HPRT, 72 C (1 min), 35 cycles; and 72 C (10 min), one cycle. Agarose gel electrophoresis (1.2%) in the presence of ethidium bromide demonstrated a single band of the expected size for each of the PCR primer pairs. Authenticity of the P450c19 PCR product was confirmed by direct DNA sequencing.
Results
Morphological changes in the adrenal glands of gonadectomized NU/J nude mice
We hypothesized that nude mice, like other inbred strains with impaired sex steroid production, might be predisposed to gonadectomy-induced adrenocortical tumorigenesis. To test this possibility, weanling female NU/J nude mice were subjected to sham surgery or gonadectomy. Within weeks of gonadectomy, patches of neoplastic cells appeared beneath the adrenal capsule and extended into the cortex. By 2 months wedge-shaped neoplastic lesions containing spindle-shaped A cells and lipid-laden B cells were evident in the adrenals (Fig. 1, A vs. B), and by 4 months tumors occupied a large fraction of the adrenal cortex (Fig. 1, C vs. D). In addition, there was evidence of mast cell infiltration in the neoplastic tissue (Fig. 1, E vs. F), a phenomenon known to accompany gonadectomy-induced tumorigenesis in susceptible inbred mice (31). These morphological changes were indistinguishable from those seen in the adrenal glands of DBA/2J mice gonadectomized for comparable lengths of time (14, 19).
Ectopic expression of gonadal transcripts in the adrenals of gonadectomized NU/J nude mice
To characterize the gonadectomy-induced changes in NU/J mice on a molecular level, we performed in situ hybridization on adrenal glands. As markers of normal adrenocortical steroidogenesis, we used MC2R, which binds ACTH and promotes adrenocortical cell proliferation and differentiation, and steroid 21-OHase, an enzyme required for glucocorticoid production. At 3 wk after gonadectomy, neither MC2R nor 21-OHase mRNA was expressed in the neoplastic areas (Fig. 2, A and B), indicating that these cells were incapable of performing typical adrenocortical functions. Instead the cells in the adrenal marker-deficient areas contained transcripts for gonadal markers. mRNA for MIS receptor type II (MISIIR), which is found in gonadal somatic cells of males and females (32) and has been observed in some gonadal and extragonadal epithelial tumors (33), was detected in the tumorous adrenals (Fig. 2C). High-magnification bright-field images confirmed the presence of this transcript in small, densely packed ovoid- or spindle-shaped cells that lacked lipid droplets, presumed to represent a subset of A cells or a perhaps a transitional cell type (Fig. 2E). P450c17, which is required for synthesis of gonadal sex steroids and normally is not expressed in adrenal cortex beyond fetal development (34), was expressed by large, lipid-laden B cells in the neoplastic adrenocortical tissue (Fig. 2, D and F).
Ectopic expression of gonadal proteins in the adrenals of gonadectomized NU/J nude mice
As in other models of gonadectomy-induced adrenocortical neoplasia (19, 35), immunohistochemistry showed the presence of GATA-4 protein in the nuclei of both A and B cells (Fig. 3A), whereas cytoplasmic LHR (Fig. 3B) and inhibin- (Fig. 3C) were observed only in B cells. P450c19 (aromatase), an enzyme involved in physiological sex steroidogenesis in gonads, also was detected in neoplastic B cells (Fig. 3D), underscoring their capacity to synthesize estrogen. The presence of MISIIR mRNA in the subcapsular region of the neoplastic lesions (Fig. 2, C and E) prompted us to examine expression of its ligand, MIS. This factor is expressed by gonadal cells in males and females and can be used as a tumor marker for sex cord-derived neoplasms and germinal epithelium-derived carcinomas (36). In gonads, the Mis gene is a known target of transcriptional activation by GATA-4 (37, 38). MIS immunoreactivity was detected in B cells and a subset of A cells (Fig. 3E). This raises the possibility that MIS produced by B cells acts in a paracrine fashion on A cells or vice versa, perhaps influencing events involved in the phenotypic switch from adrenal- to gonadal-type steroidogenesis. SF-1, which cooperates with GATA-4 in the transcriptional activation of Mis (38) and inhibin- (39, 40), was observed in B cells and normal adrenocortical cells but not in A cells (Fig. 3F). Estrogen receptor (ER)-, a nuclear hormone receptor implicated in normal and neoplastic cell proliferation, was observed mainly in nuclei of B cells and probably endothelial cells (Fig. 3G). In addition, we examined expression of Smad3, an intracellular mediator of TGF signaling (41). Smad3 immunoreactivity was evident in B cells (Fig. 3H) and in a subset of normal adrenocortical cells beneath the capsule. However, Smad3 was not detected in A cells. Similarly, -catenin immunoreactivity was seen in the nuclei of B cells (Fig. 3I). The colocalization of SF-1, Smad3, and -catenin suggests that both the TGF and Wnt signaling pathways operate in B cells (42). To assess the proliferation of neoplastic cells, we performed immunostaining for proliferating cell nuclear antigen (PCNA). This antigen was detected in both neoplastic and nonneoplastic regions. In the neoplastic tissue, it was expressed more prominently in B cells than in A cells (Fig. 3J). Thus, B cells, even though differentiated and steroidogenic, continue to proliferate in these neoplastic regions.
Hormonal changes in gonadectomized NU/J nude mice
The transition from adrenal- to gonadal-type steroidogenesis was verified by hormone measurements in serum of mice 2 months after sham surgery or gonadectomy (Table 2). Some nude mouse strains have been reported to have reduced plasma levels of gonadotropins due to impaired pituitary secretion, which limits sex steroid synthesis in the ovary (22, 43). However, we found that the baseline and postgonadectomy levels of LH in NU/J nude mice were comparable with those seen in DBA/2J or FVB/N mice (19), increasing from 0.36 ±.02 ng/ml in sham-operated females to 3.6 ± 1.7 ng/ml after gonadectomy (P < 0.05). Levels of estradiol were also higher than expected for nude mice (22, 43), amounting to 27 ± 5.7 pg/ml in sham-operated vs. 183 ± 61 pg/ml in ovariectomized females (P < 0.05). Testosterone was at the lower limit of detection for the assay, and only a modest increase was observed after gonadectomy. Collectively, these measurements confirmed the capacity of neoplastic cells in the adrenal cortex to become a source of gonadal hormones after gonadectomy. In contrast to sex steroids, corticosterone levels were reduced from 86 ± 10 ng/ml in intact nude mice to 56 ± 5.5 ng/ml (P < 0.05) in gonadectomized animals, suggesting a diminished capacity to produce glucocorticoids, presumed to reflect replacement of a large portion of adrenal cortex by neoplastic cells without compensatory growth or increased activity in the remaining adrenal tissue. We assume that the presence of LHR on the neoplastic B cells allows these cells to respond to elevated LH levels and produce sex steroids. Conversely, the lack of LHR expression on normal adrenocortical cells together with a decrease in serum corticosterone levels indicates that LH does not supplant ACTH in the stimulation of glucocorticoid production.
Hormonal changes in hCG-treated NU/J nude mice
To assess the adrenal response to elevated levels of gonadotropins in the presence of ovaries, we took advantage of the ability of NU/J nude mice to accept xenografts. CHO cells stably expressing a single-chain variant of hCG known to activate LHR (24) or control CHO cells were injected sc into the flanks of intact or gonadectomized female NU/J nude mice. In preliminary studies we verified that conditioned media from hCG-CHO and control CHO cells did not contain detectable levels of estradiol or testosterone. Inoculated mice were killed after 3–7 wk, when flank tumors were approximately 1 cm in diameter. The mean levels of hCG in serum at a time of killing were 31 ± 14 IU/ml in the intact hCG-treated mice. LH levels reached 2.1 ± 0.4 ng/ml in these mice, which we attributed to cross-reactivity of the assay with highly elevated levels of hCG rather than to a true increase in circulating LH. Despite individual variability in serum hCG levels, all of the hCG-treated mice exhibited an increase in circulating sex steroids. Whereas the levels of estradiol (174 ± 42 pg/ml) were comparable with those seen in gonadectomized nude mice, the levels of testosterone (320 ± 60 ng/dl) were significantly higher in the hCG-treated intact mice. A similar increase in serum testosterone has been reported in bovine LH-CTP transgenic mice (13). In response to hCG, serum corticosterone levels in nongonadectomized mice reached 128 ± 3.5 ng/ml, which was higher than observed in gonadectomized nude mice.
In mice subjected to both gonadectomy and xenografting, hCG levels were comparable with those in intact hCG-treated mice (Table 2). Adrenocortical tumors that formed in these animals did not differ with regard to morphology or expression of GATA-4, compared with mice treated with gonadectomy alone (data not shown). However, serum testosterone levels (in this case reflecting extragonadal production) were 2.5 times higher in the mice treated with hCG and gonadectomy than in mice treated with gonadectomy alone (50 ± 7.5 ng/dl) (Table 2).
Neoplastic changes in adrenals from hCG-treated NU/J nude mice
To determine the consequences of chronic hCG elevation on the adrenal cortex, we harvested adrenal glands from NU/J nude mice 3 wk after inoculation with hCG-CHO cells. The adrenal glands of intact NU/J nude mice inoculated with control CHO cells showed rare subcapsular GATA-4-positive cells but no frank neoplasia (data not shown). In contrast, the adrenal glands of hCG-treated mice showed neoplastic changes that resembled those seen in ovariectomized animals. As in gonadectomized mice, GATA-4-expressing A and B cells were observed in neoplastic adrenocortical tissue of hCG-treated intact mice (Fig. 4, A and F). Inhibin- immunoreactivity was evident in B cells found in the adrenals of both gonadectomized (Fig. 4B) and hCG-treated (Fig. 4G) mice; however, in the latter case, the inhibin--producing cells appeared to have less cytoplasm and inhibin- had a more lacy distribution. This might have reflected a temporal delay in differentiation of the B cells or a reduction in inhibin- expression in the hCG-treated adrenals. Friend-of-GATA-2 (Fog-2), a cofactor known to interact with GATA-4 during the differentiation of sex steroidogenic cells (37, 44), was detected in the nuclei of A and B cells (Fig. 4, C and H). In addition to nuclear staining, we observed some cytoplasmic FOG-2 immunoreactivity of unclear significance. The overlapping expression patterns of GATA-4 and FOG-2 suggested that these two factors may interact to maintain the gonadal steroidogenic phenotype in the neoplastic tissue. We also examined expression of PCNA, which in both gonadectomized (Fig. 4D) and hCG-treated (Fig. 4I) adrenals was expressed in B cells and in a subset of normal adrenal cells, as was the case at 4 months after gonadectomy (Fig. 3J). There was little or no PCNA detected in A cells, except those directly under the adrenal capsule, suggesting that tumor expansion occurs primarily through growth and proliferation of B at this stage of tumor development. In the adrenals of gonadectomized mice, ER immunoreactivity was observed mainly in neoplastic areas (Fig. 4E), whereas in hCG-treated mice, ER was expressed more broadly, localizing to both B cells and normal adrenal cells (Fig. 4J).
hCG-induced changes in the ovaries of NU/J nude mice may limit sex steroid production by neoplastic adrenocortical tissue
The ovarian changes induced by inoculation of hCG-CHO cells mimicked those seen in transgenic mice that constitutively express - and -subunits of hCG (45) or bovine LH-CTP (46). The ovaries of hCG-CHO mice were enlarged, cystic and hemorrhagic with thecal layer expansion and stromal cell proliferation (Fig. 5B), whereas the ovaries from mice inoculated with wild-type CHO cells appeared normal (Fig. 5A). Abundant GATA-4 immunoreactivity was evident in the stimulated ovaries. Normally expression of GATA-4 in postmitotic cells of the corpus luteum is weak, whereas another member of the GATA family, GATA-6, is expressed in both follicular cells and corpus luteum (47). Thus, the histological changes in hCG-treated ovaries are more reminiscent of a luteoma than physiologic luteinization. RNase protection assays on ovarian RNA from individual nude mice showed no significant difference in the levels of GATA-6 mRNA (Fig. 5B, lanes 1 and 2 vs. 3 and 4) or P450c17 mRNA (Fig 6B, lanes 5 and 6 vs. 7 and 8) in response to hCG. However, the levels of LHR mRNA were much higher in hCG-treated ovaries, compared with those from mice treated with control CHO cells (Fig. 5B, lanes 5 and 6 vs. 7 and 8).
To assess the steroidogenic capacity of hCG-induced neoplastic cells in the adrenal gland, we measured the levels of P450c17 and LHR transcripts using RNase protection assays. Both transcripts were present in adrenals from gonadectomized NU/J nude mice (Fig. 6A, lanes 1 and 2). hCG treatment of gonadectomized mice caused a 2-fold decrease in the levels of mRNA for both factors (Fig. 6A, lanes 3 and 4). Thus, the effects of gonadectomy and hCG were not additive over the 3- to 7-wk period of treatment; rather, hCG attenuated the impact of gonadectomy on adrenal sex steroidogenic activity. Surprisingly, neither P450c17 nor LHR was detected in adrenals from intact mice treated with hCG-CHO (Fig. 6A, lanes 7 and 8) or control CHO (Fig. 6A, lanes 5 and 6) cells. To compare the relative expression of P450c19 mRNA in adrenals from mice subjected to gonadectomy and/or hCG-treatment, we performed RT-PCR analysis using the housekeeping gene Hprt as a control. Expression of P450c19 is tissue-specifically regulated by alternative first exons and promoters (30), and the ovarian-specific first exon product was detected in the adrenals of gonadectomized and to a lesser extent hCG-treated mice (Fig. 6B). Use of this tissue-specific first exon underscores the gonadal character of the neoplastic steroidogenic tissue. The limited steroidogenic capacity of B cells in neoplastic adrenals of hCG-treated mice may be a consequence of decreased of adrenal LHR expression in response to chronic gonadotropin stimulation. Chronic hCG exposure apparently did not affect normal adrenocortical function, as reflected by slightly increased levels of corticosterone in hCG-treated intact mice.
Discussion
The adrenocortical lesions that appear in gonadectomized inbred mice are thought to reflect metaplasia of competent cells in the subcapsular region, which transform into gonadal-like stroma in response to changes in the hormonal milieu. Like other classic examples of metaplasia (e.g. Barrett’s esophagus), gonadectomy-induced adrenocortical metaplasia arises in a self-renewing epithelium, is caused by chronic hormonal stimulation or tissue injury, and has the potential to evolve into frank neoplasia (48). It is unclear whether gonadectomy-induced adrenocortical metaplasia represents transdifferentiation, the direct conversion of one differentiated cell type into another, or instead is an example of stem cell metaplasia (48).
Our new NU/J nude mouse model affords insights into the molecular events underlying this form of tissue-type switching and sheds light on the developmental mechanisms that control normal steroidogenic cell differentiation. We used two approaches to induce metaplasia in the adrenal cortex of these inbred nude mice: gonadectomy and xenografts of hCG-secreting cells. By characterizing the adrenal glands from these animals, we have expanded the list of molecular markers that delineate the different cell populations within the neoplastic adrenocortical tissue. Based on their morphology, marker expression (GATA-4, MIS, and MISIIR) and low PCNA-labeling index, we propose that A cells resemble stromal cells of the postmenopausal ovary, which are known to metabolize cholesterol but have limited capacity for steroid hormone biosynthesis, particularly sex steroids (49). B cells, which have a relatively high PCNA-labeling index and express GATA-4, ER, LHR, P450c17, and P450c19, resemble follicular theca cells from patients with polycystic ovary syndrome, a condition marked by chronically elevated LH levels (50, 51). The interrelationship between A and B cells and their cell(s) of origin remain enigmatic. We (19) and others (11) have hypothesized that nonsteroidogenic A cells transform into B cells based on the circumstantial evidence that A cells precede B cells, that B cells always arise in the company of A cells, and that both cell types express GATA-4. However, direct evidence that B cells are descendants of A cells is lacking. Lineage tracing studies are needed to definitively establish the origin of these cell types.
NU/J nude mice develop adrenocortical neoplasms in response to either gonadectomy or xenografts of hCG-secreting cells, suggesting that chronic gonadotropin elevation is the principal signal responsible for this process. Although the neoplastic lesions in hCG-treated intact nude mice morphologically resemble those seen in their gonadectomized counterparts, they have reduced sex steroidogenic capacity, implying that gonadotropin elevation does not explain all aspects of the phenotypic switch. The mechanism responsible for this limited adrenal sex steroidogenic capacity is unknown but may involve decreased LH signaling through down-regulation of LHR in the setting of persistently elevated hCG (52) or hCG-induced degradation of LHR mRNA (53). hCG stimulates ovarian production of estradiol and testosterone, and high levels of these gonadal hormones might directly inhibit steroidogenesis in neoplastic adrenocortical cells. Sex steroids have been shown to act directly on gonadal steroidogenic cells to inhibit P450c17 (54, 55).
Our nude mouse model suggests that multiple signaling pathways influence the adrenocortical changes that accompany gonadectomy. Ectopic expression of both MIS and MISIIR was observed early in tumorigenesis in subcapsular areas devoid of normal adrenocortical markers, suggesting that this ligand-receptor pair may be involved in the switch to gonadal-type differentiation. Perhaps MIS, acting in a paracrine fashion, inhibits proliferation or steroidogenesis in A cells or facilitates their centripetal migration. MIS expression has been demonstrated in gonadal tumors, and genetic studies suggest that inhibin- and MIS exert a synergistic effect on tumorigenesis in gonads. In inhibin- null mice, the absence of MIS causes earlier onset and more aggressive development of testicular tumors (56). The detection of SF-1, -catenin, and Smad3 in B cells suggests potential involvement of TGF and Wnt signaling pathways in mechanisms for phenotypic switching, including down-regulation of ACTH signaling and up-regulation of gonadal steroidogenesis. Based on the selective expression of the above markers in A or B cells, we assume that gonadectomy-induced adrenocortical tumorigenesis involves multiple steps and that different sets of signals are required for A vs. B cell differentiation.
Not all inbred mouse strains respond to prepubertal gonadectomy by developing sex steroid-producing adrenocortical tumors. Inbred strain susceptibility to gonadectomy-induced adrenocortical tumorigenesis appears to correlate with the in vivo and in vitro capacity of normal gonadal tissue to produce steroid hormones in response to unopposed gonadotropin stimulation (19). In DBA/2J mice, this may be a consequence of a polymorphism in the SF-1 coding sequence (17), whereas in NU/J nude mice, it may be due to the effects of athymia on function of the hypothalamus-pituitary-gonadal/hypothalamus-pituitary-adrenal axes (22, 23). Gonadectomy-induced adrenocortical neoplasia in mice and ferrets may be the byproduct of an organism’s efforts to compensate for unfavorable conditions. In DBA/2J and NU/J nude mice, these neoplastic changes are benign. Inbred strains that develop malignant lesions after gonadectomy (i.e. CE/J) may have genetic changes that either exacerbate adrenocortical hyperplasia or promote malignant transformation.
As in mice and ferrets, adrenocortical tumor formation in humans appears to be influenced by both the hormonal milieu and the inherent susceptibility of the adrenal gland. Subcapsular adrenocortical neoplasms with histological features resembling luteinized ovarian stroma (thecal metaplasia) have been reported in postmenopausal women (1, 2, 3, 4) and men with acquired testicular atrophy (5), suggesting that the hormone changes associated with gonadal failure (e.g. elevated LH, decreased sex steroids, etc.) may contribute to adrenal tumor development in humans. Analysis of heritable and spontaneous types of human adrenocortical tumors has documented alterations in either cell surface receptors or their downstream effectors that make adrenocortical tissue susceptible to hormone-associated neoplastic transformation. In a majority of cases, these genetic alterations lead to adrenocortical neoplasms that produce excess glucocorticoids, but in a few cases, notably those involving young children, the tumors produce sex steroids and inhibin-, like the adrenocortical tumors that develop in prepubertally gonadectomized mice and ferrets.
The nude mouse model described in this paper should prove useful in characterizing the roles of gonadotropins, sex steroids, and other hormones in adrenocortical metaplasia and tumorigenesis.
Acknowledgments
We thank Keith Parker, Perrin White, and Jorma Toppari for providing plasmid constructs. We thank the Digestive Diseases Research Core Center staff for assistance with immunohistochemistry. We thank Susan Porter-Tinge for technical assistance.
Footnotes
This work was supported by National Institutes of Health Grants HL61006 and DK52574, March of Dimes FY02-203, Barnes-Jewish Foundation, Mallinckrodt Foundation, Finnish Pediatric Research Foundation, and Juselius Foundation.
Abbreviations: CHO, Chinese hamster ovary; ER, estrogen receptor; hCG, human chorionic gonadotropin; HPRT, hypoxanthine-guanine phosphoribosyl transferase; LHR, LH receptor; MC2R, melanocortin-2 receptor; MIS, Müllerian inhibitory substance; MISIIR, MIS receptor type II; 21-OHase, 21-hydroxylase; PCNA, proliferating cell nuclear antigen; SF-1, steroidogenic factor-1.
References
Fidler WJ 1977 Ovarian thecal metaplasia in adrenal glands. Am J Clin Pathol 67:318–323
Werk Jr EE, Sholiton LE, Kalejs L 1973 Testosterone-secreting adrenal adenoma under gonadotropin control. N Engl J Med 289:767–770
Givens JR, Andersen RN, Wiser WL, Coleman SA, Fish SA 1974 A gonadotropin-responsive adrenocortical adenoma. J Clin Endocrinol Metab 38:126–133
Carney JA 1987 Unusual tumefactive spindle-cell lesions in the adrenal glands. Hum Pathol 18:980–985
Romberger CF, Wong TW 1989 Thecal metaplasia in the adrenal gland of a man with acquired bilateral testicular atrophy. Arch Pathol Lab Med 113:1071–1075
Thompson GB, Young Jr WF 2003 Adrenal incidentaloma. Curr Opin Oncol 15:84–90
Bertherat J, Mosnier-Pudar H, Bertagna X 2002 Adrenal incidentalomas. Curr Opin Oncol 14:58–63
Lacroix A, Baldacchino V, Bourdeau I, Hamet P, Tremblay J 2004 Cushing’s syndrome variants secondary to aberrant hormone receptors. Trends Endocrinol Metab 15:375–382
Russfield AB 1975 Experimental endocrinopathies. Methods Achiev Exp Pathol 7:132–148
Schoemaker NJ, Teerds KJ, Mol JA, Lumeij JT, Thijssen JH, Rijnberk A 2002 The role of luteinizing hormone in the pathogenesis of hyperadrenocorticism in neutered ferrets. Mol Cell Endocrinol 197:117–125
Firth CH, Dunn TB 1994 Tumours of the adrenal gland. IARC Sci Publ 111:595–609
Beuschlein F, Looyenga BD, Bleasdale SE, Mutch C, Bavers DL, Parlow AF, Nilson JH, Hammer GD 2003 Activin induces x-zone apoptosis that inhibits luteinizing hormone-dependent adrenocortical tumor formation in inhibin-deficient mice. Mol Cell Biol 23:3951–3964
Mikola M, Kero J, Nilson JH, Keri RA, Poutanen M, Huhtaniemi I 2003 High levels of luteinizing hormone analog stimulate gonadal and adrenal tumorigenesis in mice transgenic for the mouse inhibin- subunit promoter/Simian virus 40 T-antigen fusion gene. Oncogene 22:3269–3278
Murthy AS, Brezak MA, Baez AG 1970 Postcastrational adrenal tumors in two strains of mice: morphologic, histochemical, and chromatographic studies. J Natl Cancer Inst 45:1211–1222
Huseby RA, Bittner JJ 1951 Differences in adrenal responsiveness to post-castrational alteration as evidenced by transplanted adrenal tissue. Cancer Res 11:954–961
Stalvey JR, Payne AH 1983 Luteinizing hormone receptors and testosterone production in whole testes and purified Leydig cells from the mouse: differences among inbred strains. Endocrinology 112:1696–1701
Frigeri C, Tsao J, Cordova M, Schimmer BP 2002 A polymorphic form of steroidogenic factor-1 is associated with adrenocorticotropin resistance in Y1 mouse adrenocortical tumor cell mutants. Endocrinology 143:4031–4037
Morley SD, Viard I, Chung BC, Ikeda Y, Parker KL, Mullins JJ 1996 Variegated expression of a mouse steroid 21-hydroxylase/-galactosidase transgene suggests centripetal migration of adrenocortical cells. Mol Endocrinol 10:585–598
Bielinska M, Parviainen H, Porter-Tinge SB, Kiiveri S, Genova E, Rahman N, Huhtaniemi IT, Muglia LJ, Heikinheimo M, Wilson DB 2003 Mouse strain susceptibility to gonadectomy-induced adrenocortical tumor formation correlates with the expression of GATA-4 and luteinizing hormone receptor. Endocrinology 144:4123–4133
Peterson RA, Kiupel M, Bielinska M, Kiiveri S, Heikinheimo M, Capen CC, Wilson DB 2004 Transcription factor GATA-4 is a marker of anaplasia in adrenocortical neoplasms of the domestic ferret (Mustela putorius furo). Vet Pathol 41:446–449
Balciunaite G, Keller MP, Balciunaite E, Piali L, Zuklys S, Mathieu YD, Gill J, Boyd R, Sussman DJ, Hollander GA 2002 Wnt glycoproteins regulate the expression of FoxN1, the gene defective in nude mice. Nat Immunol 3:1102–1108
Rebar RW, Morandini IC, Erickson GF, Petze JE 1981 The hormonal basis of reproductive defects in athymic mice: diminished gonadotropin concentrations in prepubertal females. Endocrinology 108:120–126
Rebar RW, Morandini IC, Petze JE, Erickson GF 1982 Hormonal basis of reproductive defects in athymic mice: reduced gonadotropins and testosterone in males. Biol Reprod 27:1267–1276
Sugahara T, Pixley MR, Minami S, Perlas E, Ben Menahem D, Hsueh AJ, Boime I 1995 Biosynthesis of a biologically active single peptide chain containing the human common alpha and chorionic gonadotropin beta subunits in tandem. Proc Natl Acad Sci USA 92:2041–2045
Ben Menahem D, Jablonka-Shariff A, Hyde RK, Pixley MR, Srivastava S, Berger P, Boime I 2001 The position of the and subunits in a single chain variant of human chorionic gonadotropin affects the heterodimeric interaction of the subunits and receptor-binding epitopes. J Biol Chem 276:29871–29879
Haavisto AM, Pettersson K, Bergendahl M, Perheentupa A, Roser JF, Huhtaniemi I 1993 A supersensitive immunofluorometric assay for rat luteinizing hormone. Endocrinology 132:1687–1691
Anttonen M, Ketola I, Parviainen H, Pusa AK, Heikinheimo M 2003 FOG-2 and GATA-4 are coexpressed in the mouse ovary and can modulate Müllerian-inhibiting substance expression. Biol Reprod 68:1333–1340
Milstone DS, Shaw SK, Parker KL, Szyf M, Seidman JG 1992 An element regulating adrenal-specific steroid 21-hydroxylase expression is located within the slp gene. J Biol Chem 267:21924–21927
Mishina Y, Rey R, Finegold MJ, Matzuk MM, Josso N, Cate RL, Behringer RR 1996 Genetic analysis of the Müllerian-inhibiting substance signal transduction pathway in mammalian sexual differentiation. Genes Dev 10:2577–2587
Honda S, Harada N, Takagi Y 1996 The alternative exons 1 of the mouse aromatase cytochrome P-450 gene. Biochim Biophys Acta 1305:145–150
Kim JS, Kubota H, Kiuchi Y, Doi K, Saegusa J 1997 Subcapsular cell hyperplasia and mast cell infiltration in the adrenal cortex of mice: comparative study in 7 inbred strains. Exp Anim 46:303–306
Jamin SP, Arango NA, Mishina Y, Hanks MC, Behringer RR 2002 Requirement of Bmpr1a for Müllerian duct regression during male sexual development. Nat Genet 32:408–410
Fynn-Thompson E, Cheng H, Teixeira J 2003 Inhibition of steroidogenesis in Leydig cells by Mullerian-inhibiting substance. Mol Cell Endocrinol 211:99–104
Keeney DS, Jenkins CM, Waterman MR 1995 Developmentally regulated expression of adrenal 17-hydroxylase cytochrome P450 in the mouse embryo. Endocrinology 136:4872–4879
Rahman NA, Kiiveri S, Rivero-Muller A, Levallet J, Vierre S, Kero J, Wilson DB, Heikinheimo M, Huhtaniemi I 2004 Adrenocortical tumorigenesis in transgenic mice expressing the inhibin -subunit promoter/SV40 virus T-antigen transgene: Relationship between ectopic expression of luteinizing hormone receptor and transcription factor GATA-4. Mol Endocrinol 18:2553–2569
Rey R, Sabourin JC, Venara M, Long WQ, Jaubert F, Zeller WP, Duvillard P, Chemes H, Bidart JM 2000 Anti-Müllerian hormone is a specific marker of Sertoli- and granulosa-cell origin in gonadal tumors. Hum Pathol 31:1202–1208
Tevosian SG, Albrecht KH, Crispino JD, Fujiwara Y, Eicher EM, Orkin SH 2002 Gonadal differentiation, sex determination and normal Sry expression in mice require direct interaction between transcription partners GATA4 and FOG2. Development 129:4627–4634
Tremblay JJ, Viger RS 1999 Transcription factor GATA-4 enhances Müllerian inhibiting substance gene transcription through a direct interaction with the nuclear receptor SF-1. Mol Endocrinol 13:1388–1401
Feng ZM, Wu AZ, Chen CL 1998 Testicular GATA-1 factor up-regulates the promoter activity of rat inhibin -subunit gene in MA-10 Leydig tumor cells. Mol Endocrinol 12:378–390
Ketola I, Rahman N, Toppari J, Bielinska M, Porter-Tinge SB, Tapanainen JS, Huhtaniemi IT, Wilson DB, Heikinheimo M 1999 Expression and regulation of transcription factors GATA-4 and GATA-6 in developing mouse testis. Endocrinology 140:1470–1480
Shi Y, Massague J 2003 Mechanisms of TGF- signaling from cell membrane to the nucleus. Cell 113:685–700
Gummow BM, Winnay JN, Hammer GD 2003 Convergence of Wnt signaling and steroidogenic factor-1 (SF-1) on transcription of the rat inhibin gene. J Biol Chem 278:26572–26579
Rebar RW, Morandini IC, Benirschke K, Petze JE 1980 Reduced gonadotropins in athymic mice: prevention by thymic transplantation. Endocrinology 107:2130–2132
Tremblay JJ, Viger RS 2003 Novel roles for GATA transcription factors in the regulation of steroidogenesis. J Steroid Biochem Mol Biol 85:291–298
Matzuk MM, DeMayo FJ, Hadsell LA, Kumar TR 2003 Overexpression of human chorionic gonadotropin causes multiple reproductive defects in transgenic mice. Biol Reprod 69:338–346
Mann RJ, Keri RA, Nilson JH 2003 Consequences of elevated luteinizing hormone on diverse physiological systems: use of the LHCTP transgenic mouse as a model of ovarian hyperstimulation-induced pathophysiology. Recent Prog Horm Res 58:343–375
Heikinheimo M, Ermolaeva M, Bielinska M, Rahman NA, Narita N, Huhtaniemi IT, Tapanainen JS, Wilson DB 1997 Expression and hormonal regulation of transcription factors GATA-4 and GATA-6 in the mouse ovary. Endocrinology 138:3505–3514
Tosh D, Slack JM 2002 How cells change their phenotype. Nat Rev Mol Cell Biol 3:187–194
Jabara S, Christenson LK, Wang CY, McAllister JM, Javitt NB, Dunaif A, Strauss III JF 2003 Stromal cells of the human postmenopausal ovary display a distinctive biochemical and molecular phenotype. J Clin Endocrinol Metab 88:484–492
Kaaijk EM, Sasano H, Suzuki T, Beek JF, van Der Veen F 2000 Distribution of steroidogenic enzymes involved in androgen synthesis in polycystic ovaries: an immunohistochemical study. Mol Hum Reprod 6:443–447
Katulski K, Bornstein S, Figiel M, Wand D, Warenik-Szymankiewicz A, Trzeciak WH 1998 Typical hormonal profiles are accompanied by increased immunoreactivity of theca folliculi steroid 17-hydroxylase P450 in polycystic ovary syndrome. J Endocrinol Invest 21:304–309
Ascoli M, Fanelli F, Segaloff DL 2002 The lutropin/choriogonadotropin receptor, a 2002 perspective. Endocr Rev 23:141–174
Wang L, Menon KM 2005 Regulation of luteinizing hormone/chorionic gonadotropin receptor messenger ribonucleic acid expression in the rat ovary: relationship to cholesterol metabolism. Endocrinology 146:423–431
Johnson DC, Crane LH 1995 Inhibitory and stimulatory effect of oestrogens upon ovarian 17-hydroxylase/C17,20-lyase in immature hypophysectomized rats treated with gonadotrophin. J Endocrinol 145:59–67
Magoffin DA, Erickson GF 1982 Direct inhibitory effect of estrogen on LH-stimulated androgen synthesis by ovarian cells cultured in defined medium. Mol Cell Endocrinol 28:81–89
Matzuk MM, Finegold MJ, Mishina Y, Bradley A, Behringer RR 1995 Synergistic effects of inhibins and müllerian-inhibiting substance on testicular tumorigenesis. Mol Endocrinol 9:1337–1345(Malgorzata Bielinska, Ele)