Growth hormone mRNA in mammary gland tumors of dogs and cats.

We have shown recently that in the dog progestin administration results in mammary production of immunoreactive growth hormone (GH). At present we demonstrate the expression of the gene encoding GH in the mammary gland of dogs and cats using reverse-transcriptase PCR. GH mRNA was found in the great majority of normal mammary tissues as well as benign and malignant mammary tumors of the dog and was associated with the presence of immunoreactive GH in cryostat sections. The mammary PCR product proved to be identical to that of the pituitary. The highest expression levels were found after prolonged treatment with progestins. In carcinomas GH mRNA was also found in progesterone receptor-negative tissue samples, indicating that after malignant transformation GH gene expression may become progestin independent. GH mRNA was also present in mammary tissues of cats with progestin-induced fibroadenomatous changes. It is concluded that GH gene expression occurs in normal, hyperplastic, and neoplastic mammary tissue of the dog. The expression in normal tissue is stimulated by progestins and might mediate the progestin-stimulated development of canine mammary tumors. The demonstration of progestin-stimulated GH expression in mammary tissue of cats indicates that the phenomenon is more generalized among mammals.

Progestin-induced growth hormone excess in the dog originates in the mammary gland.

In the dog endogenous progesterone and synthetic progestins may incite overproduction of GH, resulting in acromegaly and insulin resistance. This progrestin-induced excessive GH secretion is characterized by disappearance of the pulsatile secretion pattern and insensitivity to both stimulation with GHRH and inhibition with a somatostatin analog. This progestin-induced GH hypersecretion is not associated with neoplastic transformation at the pituitary level. These observations were the impetus for a search of a possible extrapituitary site of GH production. In four ovariohysterectomized dogs elevated plasma GH levels (46.5 +/- 7.7 micrograms/liter; mean +/- SEM) were induced by administration of synthetic progestins. In these dogs hypophysectomy did not led to a significant decrease in plasma GH levels. Analysis of the GH content of various tissue homogenates revealed that the highest GH immunoreactivity was found in extracts of the mammary gland. Ectopic production of GH in the mammary gland was confirmed by lowering of plasma GH concentration to values within the reference range within 2 h after complete mammectomy in two dogs with progestin-induced elevations of plasma GH levels. In one of these dogs the arterial and elevations of plasma GH levels. In one of these dogs the arterial and venous GH concentrations across the mammary gland were measured and an arterio-venous GH gradient was demonstrated. Displacement studies in the RIA and analysis by reversed-phase HPLC revealed that mammary-derived GH is highly similar to pituitary-derived GH. Immuno-histochemical staining revealed that GH immunoreactivity was localized in focal areas of hyperplastic ductular epithelium. In mammary tissue of healthy untreated female dogs no GH immunoreactivity was found. It is concluded that treatment of dogs with synthetic progestins can induce the overproduction of GH in the mammary gland. This GH is biologically active, highly similar to pituitary derived GH, and originates from foci of hyperplastic ductular epithelium of the mammary gland.

Progestin treatment in the dog. I. Effects on growth hormone, insulin-like growth factor I and glucose homeostasis.

The effects of two synthetic progestins, medroxyprogesterone acetate (MPA) and proligestone (PROL), on the release of growth hormone (GH) and glucose metabolism were studied in two groups of eight ovariohysterectomized dogs. Eight injections of long-acting progestins were administered at 3-week intervals. Recovery was studied in four dogs of each treatment group in the 6 months following cessation of progestin administration. Treatment with both MPA and PROL resulted in similar increases in plasma levels of GH and insulin-like growth factor I (IGF-I). The GH responses to both clonidine and growth hormone-releasing hormone became impaired. In neither treatment group did the elevated plasma GH levels decrease after administration of the synthetic somatostatin analogue SMS 201-995. The size and shape of the pituitary gland were not changed by progestin treatment. After cessation of progestin administration, basal plasma levels of GH and IGF-I did not return to pretreatment values. The GH response to growth hormone-releasing hormone remained impaired for at least 6 months after the last progestin administration. In both treatment groups, glucose homeostasis was sustained initially by increased insulin production. Prolonged treatment with MPA and PROL resulted in glucose intolerance. No amelioration was observed during the recovery period in either group. A small number of dogs developed diabetes mellitus. In more than 50% of the dogs in both treatment groups small mammary tumours developed. The recently discovered local production of GH probably played a role in mammary tumorigenesis.(ABSTRACT TRUNCATED AT 250 WORDS)

Progestin treatment in the dog. II. Effects on the hypothalamic-pituitary-adrenocortical axis.

The effects of two synthetic progestins, medroxyprogesterone acetate (MPA) and proligestone (PROL), on the hypothalamic-pituitary-adrenocortical (HPA) axis were studied in two groups of eight ovariohysterectomized dogs each. Eight injections of long-acting progestins were administered at 3-week intervals. Recovery of the HPA axis was studied in four dogs of each group in the following 6 months. Basal levels of adrenocorticotrophin (ACTH) and cortisol in plasma and the urinary corticoid/creatinine ratio were measured. The responsiveness of the HPA axis was investigated by stimulation with ovine corticotrophin-releasing hormone. Both MPA and PROL caused sawtooth patterns of suppression of basal ACTH and cortisol levels in plasma, synchronous with the time of administration. The suppression of the adrenocortical component of the HPA axis was most pronounced. Adrenocorticotrophin production also was affected but to a lesser extent and occurred especially in PROL-treated dogs. Soon after the cessation of progestin administration ACTH levels increased, sometimes with a rebound. In both groups basal cortisol levels and urinary corticoid/creatinine ratios did not return to pretreatment levels until 6 months after the last progestin injection. It is concluded that MPA and PROL act as glucocorticoid agonists and suppress the HPA axis. The suppression at the adrenocortical level may last for 6 months.

Responsiveness to corticotropin-releasing hormone and vasopressin in canine Cushing's syndrome.

Corticotropin-releasing hormone (CRH) and vasopressin are the most important hypothalamic factors regulating adrenocorticotropic hormone (ACTH) secretion. In this study we have investigated the responsiveness of the pituitary-adrenocortical axis to intravenous administration of CRH or lysine vasopressin (LVP) in 16 control dogs, 22 dogs with pituitary-dependent hyperadrenocorticism and five dogs with hyperadrenocorticism due to an adrenocortical tumor, using doses of CRH and LVP that caused equivalent ACTH responses in the control dogs. After CRH administration, the increment in plasma ACTH was significantly (p < 0.05) lower in dogs with pituitary-dependent hyperadrenocorticism (221 +/- 53 ng/l) than that in control dogs (279 +/- 41 ng/l). In the dogs with pituitary-dependent hyperadrenocorticism, the relative increases in ACTH after CRH were significantly (p < 0.05) lower than those after LVP. Despite the absence of an increase in ACTH following LVP administration in dogs with hyperadrenocorticism due to an adrenocortical tumor, there was a significant increase in plasma cortisol, the increment (790 +/- 238 nmol/l) being not statistically different from that in the control dogs (412 +/- 37 nmol/l). We conclude that in spite of the changes inherent to pituitary-dependent hyperadrenocorticism, i.e. neoplastic transformation of corticotropic cells and hypercortisolism, there is persistence of responsiveness to hypophysiotropic hormones. The ACTH secretion by corticotropic cells in pituitary-dependent hyperadrenocorticism was relatively less sensitive to stimulation with CRH than with LVP. Adrenocortical tumors develop an aberrant sensitivity to LVP.

Binding specificity of medroxyprogesterone acetate and proligestone for the progesterone and glucocorticoid receptor in the dog.

The use of the synthetic progestin medroxyprogesterone acetate (MPA) for estrus prevention in the dog can result in overproduction of growth hormone, suppression of plasma glucocorticoid levels, and the induction of mammary tumors. Proligestone (PROL) was claimed to be devoid of these unwanted side effects. In the present study, the binding characteristics of MPA and PROL for the canine progesterone receptor (PR) and glucocorticoid receptor (GR) were investigated. The apparent inhibition constants for the PR and GR of MPA and PROL were compared with those of progesterone, ORG 2058, and a number of corticosteroids. MPA and PROL had high affinities for both the PR and the GR. The rank order for displacement of the binding of the PR ligand [3H]ORG 2058 from the canine uterine receptor was: MPA approximately ORG 2058 > PROL > progesterone >> cortisol, dexamethasone, and spironolactone. The rank order for displacement of the specific binding of the GR ligand [3H]dexamethasone from the canine liver receptor was: dexamethasone > cortisol > MPA > PROL > progesterone >> aldosterone approximately spironolactone. The apparent inhibition constants of PROL for both the PR and the GR were approximately 10 times higher than those of MPA. The ratios of the inhibition constants for the GR and PR appeared to be equal for PROL and MPA. It is concluded that although MPA has higher affinities for the PR and GR than PROL, both progestins have a similar in vitro binding specificity, which is less than that of progesterone. These findings are consistent with suppression of the adrenal cortex and the induction of growth hormone secretion in the mammary gland after MPA and PROL treatment in dogs.

Progestin-induced growth hormone (GH) production in the treatment of dogs with congenital GH deficiency.

The recent demonstration of the ability of progestins to induce the expression of the growth hormone (GH) gene in the mammary gland of dogs and cats opens possibilities for the treatment of some forms of GH deficiency with progestins. Therefore, one male and one female German shepherd dog with congenital dwarfism because of a pituitary anomaly were treated with subcutaneous injections of medroxyprogesterone acetate (MPA) in doses of 2.5-5.0 mg per kg body weight, initially at 3-wk intervals and subsequently at 6-wk intervals. In both dogs, body sizes increased and a complete adult hair coat developed. Undesirable side-effects were recurrent periods of pruritic pyoderma in both dogs and cystic endometrial hyperplasia with mucometra in the female dog. Parallel with the physical improvements, plasma insulin-like growth factor I concentrations rose sharply. Plasma GH concentrations tended to rise, but never exceeded the upper limit of the reference range. Nevertheless, one of the dogs developed slight acromegalic features, possibly because mammary GH, unlike pituitary GH, is released evenly throughout the day. Even moderate increases in circulating GH concentration may, therefore, give rise to overexposure. It is concluded that long-term treatment with MPA can be used as an alternative for heterologous GH in the treatment of congenital GH deficiency in the dog.

Characterization of and radioimmunoassay for canine thyroglobulin.

Canine thyroglobulin (cTg) has been isolated and purified. It has similar electrophoretic patterns as Tg from other mammalian species. The main fraction had a MW of 660,000, whereas also fractions of a MW of approximately 1,300,000 (dimer) and 330,000 (subunit) were present. The iodine content was 0.8 to 1.0% (w/w). cTg did not cross-react with antibodies against human Tg to a degree that would allow the use of a radioimmunoassay for human Tg for the determination of cTg in serum or plasma. Therefore a polyclonal antiserum was raised against cTg and a homologous radioimmunoassay was developed, which was sensitive (0.4 micrograms/l) and specific (cross-reactivity in cTg assay of human Tg, goat Tg, T4, T3, and DIT less than 0.01%). Plasma Tg levels in normal dogs of both sexes and aged 3-15 years amounted to 192 +/- 73 micrograms/l (mean +/- SD, n = 30). There was no relation between plasma Tg and T4 levels.

Comparison of the histological changes in the dog after treatment with the progestins medroxyprogesterone acetate and proligestone.

Administration of progestins in the dog may result in overproduction of growth hormone, suppression of the hypothalamic-pituitary-adrenocortical axis, and insulin resistance. In this paper we present a comparison of the histological findings in control dogs and dogs treated with either medroxyprogesterone acetate (MPA) or proligestone (PROL). Depot preparations of MPA or PROL were administered (SC) at 3-week intervals in two groups of seven ovariohysterectomized beagle dogs, after which three dogs of each group were killed. After a 6-month period without hormone treatment during which recovery was studied, the remaining dogs received five additional injections at the same interval and were subsequently killed. Tissue samples of four intact female beagle dogs served as controls. Progestin treatment resulted in atrophy of the adrenal cortex. In both MPA- and PROL-treated dogs, the thickness of the combined zona fasciculata and reticularis was significantly smaller than in control animals. In the mammary glands of progestin-treated dogs there were well developed alveoli and normal ducts adjacent to foci of hyperplastic ductular epithelium. Five dogs in each treatment group had developed benign mammary tumours which varied from simple tubular and papillary adenomas to benign complex and mixed tumours, whereas no mammary tumours were observed in the control animals. In each treatment group, steroid-induced hepatopathy was observed in the liver of three dogs. Vacuolation of the cells of the islets of Langerhans and the epithelium of the intercalated ducts was present in two dogs of each treatment group and was only observed after the second series of progestin administrations. Incidental findings included chronic pyelonephritis, aspecific dermatitis, and mucinous dysplasia of the gall bladder. No abnormalities were found in sections of spleen, lung, brain, or pituitary gland. There were no significant differences in the frequencies of the various abnormalities between MPA- and PROL-treated dogs. Our findings correspond with the clinical and biochemical results after treatment of dogs with MPA and PROL. The high incidence of mammary tumours might be associated with our recent finding that in the dog progestins induce ectopic production of growth hormone in the mammary gland. The dog might be a good model for further studies on hormonally induced breast cancers.

Mammary growth hormone and tumorigenesis--lessons from the dog.

The discovery in the early 1990s that progestin-induced growth hormone (GH) excess in the dog originates in the mammary gland can be seen as a hallmark in the research on the pathogenesis of mammary cancer in the dog. The local biosynthesis and release of GH may provide a highly proliferative environment in the mammary gland, which contributes to the development and/or progression of mammary tumours. Before final goals such as prevention of tumour formation or inhibition of tumour promotion can be achieved it is of eminent importance to elucidate the mechanism of progesterone-induced mammary GH production and the mechanism of local autocrine/paracrine action of GH. These local GH effects may be achieved through direct growth stimulating effects of GH as well as by indirect effects mediated by the stimulation of the biosynthesis of insulin-like growth factor-I (IGF-I). The biological effects of the IGFs largely depend on the presence of IGF binding proteins (IGFBPs) which may both enhance or inhibit the activity of the IGFs. This review concentrates on recent advances in the understanding of the local mammary GH-IGF axis and the lessons which can be drawn from the dog for mammary cancer research in other species.

[A female dog who grew progressively more lethargic and hoarse]. / Een teef die progressief slomer en heser werd.

A female dog (Collie dog, eight years of age, non-spayed) was referred to the University Clinic for Companion Animals with signs and symptoms suggesting endogenous progesterone-induced acromegaly and cystic endometrial hyperplasia. The dog had glucose intolerance, but the growth hormone concentration in plasma was within the reference range. The latter was probably due to the decline of progesterone at the end of the luteal phase, resulting in an abrogation of the process of progesterone-induced growth hormone hypersecretion. After ovariohysterectomy the glucose-tolerance normalized.

Progestins and growth hormone excess in the dog.

Chronic overproduction of growth hormone in man and the cat is most often caused by a GH-producing tumour of the pituitary gland. In dogs the usual cause is quite different. In this species endogenous progestins (during metestrus) and exogenous progestins (used to prevent estrus) cause excessive GH secretion that is reversible. For a better understanding of the mechanism of progestin-induced GH synthesis and secretion, studies were carried out in healthy dogs and in dogs presented with GH excess. The effectiveness of stimulation of GH secretion by hGHRH, and clonidine and of the inhibition with the somatostatin analogue SMS 201-995 were evaluated in healthy male dogs. SMS 201-995 had no influence on basal plasma GH levels, but significantly inhibited the hGHRH and clonidine-evoked GH release. In healthy female dogs the basal concentrations of plasma GH were significantly higher during metestrus than during anestrus. The elevated basal plasma GH levels were associated with a diminished responsiveness to stimulation with clonidine. In female dogs with experimental or spontaneous progestin-induced chronic overproduction of GH (and IGF-I), plasma GH levels were completely unresponsive to stimulation with hGHRH and clonidine and to inhibition with SMS 201-995. It is concluded that in the female dog the progestin-induced GH excess has characteristics of autonomous secretion.

The role of progestins, insulin-like growth factor (IGF) and IGF-binding proteins in the normal and neoplastic mammary gland of the bitch: a review.

The growth hormone (GH) and insulin-like growth factor (IGF) system is an important regulatory system of mammalian epithelial cell proliferation and differentiation. The biological effects of the IGFs are modulated by six different binding proteins (IGFBPs). Progestins play an important role in the regulation of the dynamics of mammary gland development and involution through the modulation of these growth regulating factors. In dogs and cats, progestins stimulate the local production of GH in the mammary gland. In dogs, this results in high plasma concentrations of GH and a concomitant increase in plasma IGF-I and IGFBP-3 concentrations. The administration of progestins also induces high plasma concentrations of IGF-II, even before GH concentrations start to increase. In the mammary gland of the normal bitch, IGFBP-5 and IGFBP-2 are the main IGFBPs expressed. Progestin administration results in a decrease of mRNA encoding IGFBP-5, but does not alter the concentration of mRNA encoding IGFBP-2. This local mammary system of GH, IGFs and IGFBPs plays an important role in the regulation of mammogenesis, lactation and involution. However, the presence of a high proliferative environment may also enhance the risk of malignant transformation and promotion of tumour growth with an associated inhibition of programmed cell death.

Effects of progestin administration on the hypothalamic-pituitary-adrenal axis and glucose homeostasis in dogs.

The effects of medroxyprogesterone acetate (MPA) and proligestone (PROL) on the hypothalamic-pituitary-adrenocortical axis and glucose homeostasis were studied in two groups of eight ovariohysterectomized beagle bitches. In addition, the binding characteristics of MPA and PROL for the progesterone and glucocorticoid receptor were investigated. The administration of both progestins resulted in suppression of the hypothalamic-pituitary-adrenal axis. Whereas basal plasma concentrations of adrenocorticotrophic hormone (ACTH) were only moderately affected, the basal plasma concentrations of cortisol and the cortisol:creatinine ratio in urine were significantly decreased after the first administration of both progestins. In the group given MPA the increase of ACTH after stimulation with corticotrophin-releasing hormone (CRH) remained normal but it was suppressed in the group treated with PROL. In both treatment groups the increase of cortisol after stimulation with CRH was lower. After cessation of progestin administration both basal and stimulated plasma ACTH concentrations returned to pretreatment concentrations within a few weeks. In contrast, it took 6 month to restore the basal plasma cortisol concentrations and cortisol:creatinine ratios in urine. Paradoxically, the stimulated cortisol concentrations returned to normal shortly after the cessation. Histological examinations revealed a severe atrophy of the zona fasciculata and reticularis of the adrenal gland in all treated dogs and a steroid-induced hepatopathy in 50% of them. During the first half of the progestin treatment, glucose homeostasis was maintained by increased plasma concentrations of insulin in both groups. After prolonged treatment the response to a glucose load became impaired. None of these parameters improved during the 6 month recovery period. Histological changes in the pancreas, characteristics of diabetes mellitus, were found in two dogs of each group. Most probably, the glucocorticoid action of the progestins is not the sole explanation for the insulin resistance since progestin treatment resulted in a concomitant increase in plasma concentrations of growth hormone which has diabetogenic properties. Experiments in vitro confirmed the strong glucocorticoid component of MPA and PROL. The inhibition constants (Ki) of PROL for both the progesterone receptor (PR) and the glucocorticoid receptor (GR) were approximately then times higher than those of MPA. Nonetheless, the ratios of the Ki for the GR and PR indicated that the specificity of MPA and PROL was only slightly different but considerably smaller than that of progesterone. It is long-term treatment with high doses of these progestins may result in a iatrogenic Cushing's syndrome and diabetes mellitus.

Circulating thyroglobulin measurements by homologous radioimmunoassay in dogs with thyroid carcinoma.

Circulating thyroglobulin was measured in 20 dogs with thyroid cancer, using a homologous polyclonal radioimmunoassay. Plasma Tg levels exceeded the normal range in 14 (70%) dogs, and ranged from 6 to 2902 micrograms/l (median 608). Plasma Tg levels tended to decrease from follicular carcinomas to solid-follicular carcinomas, to solid carcinomas (p less than 0.05). Plasma Tg levels were also higher in scintigraphically hot tumours than in cold ones. Other relationships between circulating Tg and clinical, pathological, and functional parameters were not found, except a poor (R = 0.49) but significant (p = 0.04) correlation between Tg and T4 levels. Plasma Tg was measured before surgery and at least once during follow-up, in 9 dogs. After hemithyroidectomy, a decrease was observed in 8 dogs. In 7 of these 8 dogs, plasma Tg levels declined within the reference range at the first postoperative sample. In the ninth dog, where metastases were detected 14 months after surgical treatment, plasma Tg slightly increased, yet within normal range. It is concluded that measurement of plasma Tg levels might be useful for monitoring the postoperative course of the disease in individual dogs with thyroid cancer.