Lactation-induced WAP-SV40 Tag transgene expression in C57BL/6J mice leads to mammary carcinoma.

Two transgenic lineages were generated by directing the expression of SV40 T antigen to the mammary gland of inbred C57BL/6J mice using the whey acidic protein (WAP) promoter. In one lineage, WAPTag 1, multiparous female mice developed mammary adenocarcinoma with an average latency period of 13 months. The histopathological phenotype was heterogeneous, tumours occurred in a stochastic fashion, normal tissue was located next to neoplastic tissue, the mammary tumours usually developed and were remarkably similar to that observed in human cases. In addition, male and virgin females developed a poorly differentiated SV40 T antigen-positive soft tissue sarcoma, also at 13 months of age. In the other lineage, WAPTag 3, some parous females developed mammary tumours, but most mice succumbed to osteosarcomas arising from the os petrosum at 5.5 to 6 months of age and on necropsy, renal adenocarcinomas were also found. Appearance of these unexpected tumour types demonstrates the non-specific expression of SV40 Tag under the control of the WAP promoter. The expression of SV40 Tag in mammary glands at different stages of development was also examined, and only actively lactating glands were positive. This suggests that the abundant cyclic synthesis of SV40 Tag associated with pregnancy is required for mammary tumorigenesis in these lineages.

Multigenic and imprinting control of ovarian granulosa cell tumorigenesis in mice.

Spontaneous juvenile ovarian granulosa cell (GC) tumors that occur in young girls are similar to GC carcinomas that develop in SWR-derived inbred mice. We analyzed female offspring from a series of matings among SWR and SJL inbred mice for chromosomal loci underlying tumor susceptibility. Intercross F2 female mice were produced by reciprocal matings of (SWR x SJL)F1 and (SJL x SWR)F1 parents. Tumorigenesis in these F2 mice as well as in SWXJ recombinant inbred and congenic strains of mice derived from SWR and SJL showed significant (P < 0.001) association with Gct1, a dominant susceptibility locus on chromosome (CHR) 4 and with Gct2 on CHR 12. Suggestive (P < 0.01) association was found with Gct3 on CHR 15. A fourth susceptibility locus, Gct4 on CHR X, was demonstrated with a strong parent-of-origin effect associated with the paternal genotype. Imprinting and complex interactions among these four loci combine to establish the probability for GC tumorigenesis in this mouse model.

Mechanisms of hormone-mediated carcinogenesis of the ovary in mice.

Experimental ovarian carcinogenesis has been investigated in inbred and hybrid strains of mice and induced by a diversity of mechanisms including X-irradiation, oocytotoxic xenobiotic chemicals, ovarian grafting to ectopic or orthotopic sites, neonatal thymectomy, mutant genes reducing germ cell populations, and aging. The mechanisms are briefly reviewed whereby disruptions in the function of graafian follicles results in a spectrum of ovarian proliferative lesions including tumors. The findings in mutant mice support the concept of a secondary (hormonally-mediated) mechanism of ovarian carcinogenesis in mice associated with sterility. Multiple pathogenetic factors that either destroy or diminish the numbers of graafian follicles in the ovary result in decreased sex hormone secretion (especially estradiol-17 beta) leading to a compensatory over-production of pituitary gonadotrophins (particularly luteinizing hormone), which places the mouse ovary at an increased risk to develop tumors. The intense proliferation of ovarian surface epithelium and stromal (interstitial) cells with the development of unique tubular adenomas in response to sterility does not appear to have a counterpart in the ovaries of women.

Granulosa cell tumorigenesis in genetically hypogonadal-immunodeficient mice grafted with ovaries from tumor-susceptible donors.

The SWR and SWXJ recombinant inbred strains of mice develop heritable, pubertal onset ovarian granulosa cell (GC) tumors with characteristics similar to those observed for human juvenile GC tumors. We utilized this murine model to determine: (a) whether spontaneous tumorigenesis is an intrinsic property of the susceptible ovary; (b) whether pubertal developmental stage affects tumorigenesis; and (c) whether tumorigenesis depends on extraovarian regulation provided by an immune system or a hypothalamic-pituitary gonadotropin system. To test these questions, ovaries from tumor-susceptible donors were grafted beneath the kidney capsules of hosts with differing immunological and hormonal capabilities. Hosts for these ovarian grafts were: (a) immunologically intact, syngeneic mice; (b) immune-deficient, allogeneic mice homozygous for the severe combined immune deficiency (scid/scid) mutation; and (c) scid/scid mice segregating for the hypogonadal (hpg) mutation, yielding gonadotropin-deficient hpg/hpg scid/scid and gonadotropin replete +/? (hpg/+ or +/+) scid/scid littermates. Donors and hosts of differing ages were used to address questions of developmental effects on tumorigenesis. Grafts were examined 6 to 10 wk after implantation for ovarian morphology and tumor incidence. Results showed that ovary grafts from susceptible female mice formed spontaneous GC tumors equally well in both syngeneic and immune-deficient scid/scid hosts. In each type of host, the incidence of grafts exhibiting spontaneous tumor development declined significantly with increasing age of both donor and host. In addition, prepubertal ovary grafts formed spontaneous tumors in hormonally normal +/? scid/scid but not in hormonally deficient hpg/hpg scid/scid hosts. Finally, treatment of hpg/hpg scid/scid host mice with the androgenic steroid hormone precursor, dehydroepiandrosterone, resulted in GC tumor formation in the tumor-susceptible ovary grafts. We conclude that pubertal onset, spontaneous tumorigenesis in the susceptible ovaries is: (a) independent of an intact immune system; (b) terminated by completion of ovarian maturation as a cyclic organ; (c) not dependent on extraovarian factors unique to the genetically susceptible host; and (d) potentially initiated by androgenic steroids in the absence of an intact hypothalamic-pituitary gonadotropin axis. We hypothesize that ovarian androgens synthesized in response to normal gonadotropin stimulation initiate spontaneous tumorigenesis in the genetically susceptible ovary.

Genetic susceptibility for C19 androgen induction of ovarian granulosa cell tumorigenesis in SWXJ strains of mice.

Susceptibility to pubertal onset, malignant granulosa cell (GC) tumors of the ovary is inherited in SWR/Bm and certain SWR-related SWXJ recombinant inbred strains of mice. In some SWXJ strains, GC tumors occur spontaneously (spontaneous strains), and in others GC tumors can only be induced by treatment with dehydroepiandrosterone (DHEA-dependent strains). A gene controlling susceptibility to both spontaneous and DHEA-induced GC tumorigenesis, Gct, has been assigned to Chromosome 4. Additional research on the role of steroids in GC tumorigenesis has revealed a second gene controlling response to C19 androgenic steroids. Spontaneous strains showed increased tumor frequency after treatment with testosterone (T), whereas DHEA-dependent strains showed no GC tumors following T treatment. Within treatment groups, serum steroid data from DHEA, T, and control treated mice showed no consistent differences between spontaneous and DHEA-dependent strains with respect to progesterone, DHEA, androstenedione, dihydrotestosterone, T, estrone, or estradiol. Thus, observed differences in GC tumor responsiveness to exogenous steroids were not due to different patterns of steroid metabolism among spontaneous and DHEA-dependent strains. Further studies on the range of effective C19 steroids were conducted using one spontaneous and one DHEA-dependent strain. The spontaneous strain showed increased GC tumor frequency in response to dihydrotestosterone and androsterone treatment, whereas the DHEA-dependent strain showed no response. This result suggests that spontaneous strains may be sensitive to a broad range of C19 steroids. To determine whether genetic differences in endogenous steroid levels have a role in spontaneous GC tumorigenesis, serum steroid levels were measured in SWR/Bm and SJL/Bm progenitor strains during the developmental period of risk between 22 and 38 days of age. With the exception of transiently increased DHEA at 22 days, there were no consistent differences in steroid levels analyzed. Thus, serum steroid profiles were not reliably prognostic for GC tumorigenesis. In conclusion, GC tumor induction in response to T treatment has co-segregated with susceptibility to spontaneous GC tumors in the SWXJ recombinant inbred strains. Thus, the second gene in our ovarian granulosa cell tumor model regulates responsiveness to T. We propose to name this gene spontaneous ovarian tumorigenesis (Sot), with alleles for susceptibility (s) carried by spontaneous strains and resistance (r) carried by DHEA-dependent strains.

Ovarian granulosa cell tumorigenesis in SWR-derived F1 hybrid mice: preneoplastic follicular abnormality and malignant disease progression.

A high incidence (27.5%; 174 of 633) of spontaneous, malignant ovarian granulosa cell tumors develop in (SWR x SWXJ-9)F1 hybrid females between 3 and 6 weeks. Granulosa cell tumors developed in predictable stages, starting as preneoplastic lesions appearing as hyperemic follicles on the ovarian surface. These follicles were characterized by hypertrophied theca, degenerating oocytes, and large fluid- or erythrocyte-filled antra lined by irregular masses of granulosa cells. Rapidly proliferating granulosa cells filled the antra and the theca/interstitial cells became more dysplastic as granulosa cell tumors developed. Thus the morphology of the preneoplastic lesion suggests that disturbed mechanisms for normal follicular development underlie granulosa cell tumor initiation. Estradiol treatment before but not after preneoplastic lesions appeared inhibited granulosa cell tumor formation. By 6 to 9 months 42% of these mice show metastases in major abdominal and thoracic organs. Thus this model can be experimentally analyzed both for mechanisms of granulosa cell tumor initiation and subsequent malignant progression.

Epidermal growth factor receptors in spontaneous ovarian granulosa cell tumors of SWR-derived mice.

Epidermal growth factor (EGF) receptor binding properties were examined in spontaneous ovarian granulosa cell (GC) tumors from SWR and SWR-derived strains of mice. EGF binding was measured at room temperature in tissue homogenates from GC tumors and normal ovaries from adult randomly cycling mice. GC tumor tissue displayed significantly increased EGF binding and 2 receptor populations (R1 and R2). Normal ovarian tissue appeared to have only one receptor population with a dissociation constant (KD) similar to the R1 (high-affinity) receptor in GC tumors. In subsequent experiments, GC tumor and normal granulosa cells from immature mice were analyzed in primary cultures for EGF binding, immunofluorescence microscopy for receptors, and cell proliferation. After 24 hr in culture, the GC tumors bound 10-fold more EGF/micrograms protein than did normal granulosa cells. GC tumor cells, but not normal granulosa cells, showed specific immunofluorescence when reacted with a polyclonal antibody to mouse EGFR. During 96 hr in culture, GC tumor cells, but not normal cells, showed a significant proliferative response to EGF. In conclusion, the EGF binding capacity is markedly increased in GC tumor cells and the proliferation data suggest that this growth factor supports tumor growth in the SWR model system.

Gene for ovarian granulosa cell tumor susceptibility, Gct, in SWXJ recombinant inbred strains of mice revealed by dehydroepiandrosterone.

Spontaneous, malignant ovarian granulosa cell (GC) tumors occur in pubertal SWR and specific SWXJ recombinant inbred strains of mice. Treatment of these mice with dehydroepiandrosterone (DHEA), an adrenal secretory steroid with anticancer actions against spontaneous and carcinogen-induced tumors of different tissues, gave unexpected results. Diet supplemented with 0.4% DHEA (a) induced significantly more GC tumors in spontaneous tumor-susceptible strains (SWR and SWXJ-1, -4, and -9), (b) induced the first GC tumors observed in five previously tumor-free strains (SWXJ-6, -7, -8, -10, and -12), and (c) failed to induce GC tumors in SJL and in the remaining six SWXJ strains (SWXJ-2, -3, -5, -11, -13, and -14). The strain distribution pattern of DHEA-induced GC tumor susceptibility versus resistance was compared with strain distribution patterns for 35 different loci known to distinguish SWR and SJL progenitor strains. A complete match of DHEA-induced GC tumors with pancreas-2 (Pan-2) on mouse chromosome 4 was found. We have named this new locus GC tumor susceptibility (Gct), with the Gcts (susceptible) allele found in SWR and the Gctr (resistant) allele found in SJL mice. The Gct locus is closely linked to pancreas-2, Pan-2, but the order of genes is not yet confirmed. In addition, data from F1 progeny of matings between SWR and selected inbred strains provide suggestive evidence for a second gene controlling GC tumor incidence that we hypothesize involves steroid metabolism. Differences in GC tumor incidence data from reciprocal F1 progeny of matings between SWR and SJL mice reveal a strong maternal effect that may represent yet a third gene. These data support a heritable basis for GC tumorigenesis in the SWR model involving a small number of genes.

Induction of ovarian granulosa cell tumors in SWXJ-9 mice with dehydroepiandrosterone.

Spontaneous ovarian granulosa cell (GC) tumors develop in SWXJ-9 inbred mice at approximately the time of puberty. The effect of dehydroepiandrosterone (DHEA), a steroid secreted by the adrenals and reported to have antitumor actions, was examined in this ovarian tumor model. In contrast with expectations, administration of diet supplemented with 0.4% DHEA or Silastic capsules containing 10 mg DHEA resulted in a significant multifold increase in GC tumor incidence. Similar studies with metabolites of DHEA, i.e., testosterone (TESTO), dihydrotestosterone (DHT), and 17 beta-estradiol (E2), revealed that TESTO was as effective as DHEA in increasing GC tumor incidence. DHT was without effect, and E2 suppressed GC tumor incidence. Serum steroid levels and steroid target tissue responses were assessed to determine if a correlation between a change in level or response to specific steroids and GC tumorigenesis existed. In both tumor-free and GC tumor host mice, dietary or capsular treatment with DHEA, TESTO, or DHT resulted in substantial alteration in one or more of serum steroids, DHEA, androstenedione, TESTO, and DHT, in addition to the administered steroid. No consistent correlation was observed between changes in a single steroid or pattern of steroids and GC tumorigenesis. Although significant increases in serum estrogens could be detected in GC tumor hosts treated with DHEA but not TESTO, estrogens did not induce these tumors. Treatment with E2 increased only serum E2 levels. In tumor-free mice, DHEA and E2 treatments were associated with vaginal cytological evidence of estrogen action, whereas the androgens induced a leukocytic pattern. Eighty-eight % of GC tumor host mice, regardless of steroid treatment, showed a vaginal cytology pattern that included cornified cells. The evidence presented in this report leads us to hypothesize that (a) spontaneous and steroid-induced GC tumorigenesis in these mice have the same mechanism, and (b) subtle increases in DHEA or a closely related metabolite during the peripubertal period may initiate GC tumors in these genetically susceptible mice. The mechanism whereby these steroids initiate GC tumorigenesis remains to be determined.

Gonadotropin uptake in genetic and irradiation models of ovarian tumorigenesis.

Genetic and irradiation models of ovarian tumorigenesis were investigated for evidence that elevated gonadotropins have a role in tumorigenesis. Wx/Wv mice lack oocytes at birth, develop complex mesothelial adenomas by 6 mo, and additional ovarian tumor types later. Uptake of iodinated human chorionic gonadotropin (125I-hCG) was measured in mice aged 1 to 30 mo, and uptake iodinated human follicle-stimulating hormone (125I-hFSH) was measured in mice aged 1 to 12 mo. Gonadotropin uptake by Wx/Wv ovaries in vivo declined quickly and was undetectable by 6 mo. Irradiated ovaries rapidly lost oocytes and follicular structures, formed mesothelial adenomas by 5 mo, and later formed additional types of ovarian tumors. In the irradiation model, 125I-hCG uptake also declined quickly and was undetectable by 3 mo of age. Neither the surface nor the tubular epithelium of the mesothelial adenoma were consistently labeled by 125I-hCG in autoradiography studies with either model. Although these data do not exclude an acute role for gonadotropins in initiation of preneoplastic events, they do indicate that ovarian cells do not require chronic gonadotropin stimulation during subsequent tumorigenesis. These findings are discussed in relation to additional models of ovarian tumorigenesis.

Ovarian tumors not induced by irradiation and gonadotropins in hypogonadal (hpg) mice.

Hypogonadal (hpg/hpg) mice deficient in gonadotropin-releasing hormone were used to study gonadotropin involvement in ovarian tumorigenesis following gamma irradiation. In the first experiment, 30-day-old hpg/hpg and normal (+/-) littermate mice were irradiated. The same mice were killed 10-15 mo later, and autopsies were performed. Ovaries of irradiated hpg/hpg mice were devoid of oocytes, but retained follicular structures. Neither mesothelial adenomas nor granulosa cell tumors were observed. In contrast, all irradiated +/- mice formed mesothelial adenomas or granulosa cell tumors, or both. Therefore, oocyte death in the absence of gonadotropins did not initiate ovarian tumorigenesis. In the second experiment, irradiated and nonirradiated hpg/hpg and +/- mice were injected 3 times weekly for 180 days with either low or high doses of pregnant mare serum gonadotropin (PMSG) and human chorionic gonadotropin (hCG) in combination. Irradiation reduced ovarian mass and markedly reduced ovarian weight increase in response to exogenous gonadotropins. Follicular dissolution and stromal cell hypertrophy were observed in saline-treated and gonadotropin-treated +/- mice that had been irradiated, and in hpg/hpg mice given the high gonadotropin dose. Mesothelial adenoma formation was observed in 100% of saline-treated, 14% of low dose-treated, and 11% of high dose-treated +/- mice. No mesothelial adenomas developed in any hpg/hpg or nonirradiated +/- mice, despite gonadotropin-induced stromal luteinization. These results indicate that, in the absence of gonadotropins, irradiation leads only to the loss of oocytes. The presence of gonadotropins was necessary to promote follicular dissolution and stromal luteinization, but was insufficient to stimulate mesothelial adenoma formation.