Hormones and endocrine glands of fishes. Studies of fish endocrinology reveal major physiologic and evolutionary problems.

The structures discussed above illustrate some of the evolutionary knowledge to be obtained from studies of the comparative biology of the endocrine system among the fishes (see also 17, 78). The interrenal (adrenocortical) gland is an example of an endocrine structure which shows great morphologic variation in the vertebrate series, and among the fishes themselves. The structural variability implies little selective value for any particular pattern, and, in fact, biochemically the interrenal tissue would seem to be involved in much the same kind of steroidogenesis in the vertebrates generally. However, the hormone aldosterone may be a tetrapod novelty. The caudal neurosecretory system is a good example of a ubiquitous endocrine apparatus, among at least teleostean and elasmo-branch fishes, for which a function has yet to be elucidated and which continues to challenge the comparative physiologist. This system, along with the Stannius corpuscles, is lacking in the tetrapods. The existence of these structures makes it clear that the endocrine biology of fishes cannot be tacitly summarized as being essentially similar to that of the tetrapods, only less well developed. The prolactin situation illustrates the existence of a gland-the pituitary-present among fishes as among tetrapods, which secretes a product only partly related to the tetrapod hormone and having a very different functional significance. In this case it is clear that "the hormone and the uses to which it is put" have undergone evolutionary change during vertebrate phylogeny.

Teleostean urophysis: urotensin II and ion transport across the isolated skin of a marine teleost.

The caudal neurosecretory peptide urotensin II rapidly inhibits by 30 percent the short-circuit current across the isolated skin of a marine teleost. The effect appears to be specific and cannot be attributed to actions of epinephrine, urotensin I, or arginine vasotocin. The results strongly suggest that urotensin II may act directly on ion-transporting cells involved in teleostean hypoosmoregulation.

Adenosis-like lesions and other cervicovaginal abnormalities in mice treated perinatally with estrogen.

Female mice from three inbred strains (BALB/cCrgl, C3H/Crgl, and C57BL/Crgl) and one noninbred stock [RU:NCS (RU)] were treated perinatally with estradiol benzoate (EB), diethylstilbestrol (DES), or sesame oil and were killed on postnatal days 30--36. A combined prenatal and neonatal regime of EB injections resulted in the abnormal presence of columnar epithelium in the vaginal fornices of some of the mice from each strain or stock. The same epithelial abnormalities were also present in the vaginal fornices of 30-day-old RU:NCS(RU) mice that had been treated only neonatally with EB or DES. The incidence of these lesions was 40--67% in the mice treated prenatally and neonatally with EB, 68% in the neonatal EB treatment group, and 100% in the neonatal DES treatment group. The columnar cells were arranged either as single layers in areas of the fornical lining epithelium or as glandlike or cystic structures in the subepithelial stroma. No cells of this type were detected in any of the samples from sesame oil-inoculated control mice. No comparable epithelial lesions were detected in the common cervical canal of the perinatally estrogen-treated animals, but this treatment consistently resulted in gross structural abnormalities at this site.

Growth of mouse mammary glands after neonatal sex hormone treatment.

Mammary gland growth was examined in female mice immediately and during 1 month after neonatal treatment with sex hormones. BALB/c mice were given daily injections of hormones for the first 5 days after birth. As a criterion of growth, the number of ductal branchings was counted in stained wholemounts of the fourth (inguinal) pair of glands. At day 6, the growth of the glands was significantly inhibited in animals treated with 17 beta-estradiol (E) or diethylstilbestrol (DES); after treatment with progesterone (P) or androgens, no immediate effect was evident. At day 33, neonatal treatment with E stimulated growth, as did neonatal treatment with testosterone (T) or 5 alpha-dihydrotestosterone (5 alpha-DHT); treatment with a high dose of DES inhibited growth. Treatment with P or 5 beta-dihydrotestosterone had no effect. In conclusion, neonatal treatment with sex hormones has two distinct effects on the growth of mouse mammary gland: 1) inhibition by estrogens, observed immediately after treatment, and 2) stimulation by E, T, and 5 alpha-DHT, observed 4 weeks after treatment.

Effect of immunosuppression on neonatally diethylstilbestrol-induced genital tract lesion and tumor development in female mice.

Neonatal treatment of female rodents with the synthetic estrogen diethylstilbestrol [(DES) CAS: 56-53-1; alpha, alpha'-diethyl-4,4'-dimethoxystilbene] results in immunodeficiency that persists into adulthood, along with progressive development of genital tract lesions. DES-induced immunosuppression was examined as a possible promoter of genital tract lesion and tumor development in BALB/cCrgl female mice. Secondary suppression of T-cell immunity after neonatal DES treatment failed to increase lesion incidence or to promote genital tract tumor development in 8-month-old mice. Although hyperplastic lesions were present in 70% of the genital tracts from DES-treated mice at 8 months of age, apparently few neoplastic cells are present at this time, since transplantation of genital tracts from these animals into syngeneic hosts did not yield tumors. To reduce the possibility that potential tumors were too immunogenic to survive in the syngeneic hosts, genital tract pieces from 12-month-old DES-treated female mice were transplanted into immunosuppressed hosts; only 1 tumor resulted. The tumor was a mixed adenocarcinoma, squamous cell carcinoma, and sarcoma and was immunogenic. These results suggest that immunosurveillance by T-cells is relatively unimportant in DES-induced genital tract lesion development.

Long-term responses of the mouse uterus to neonatal diethylstilbestrol treatment and to later sex hormone exposure.

The effects of ovariectomy at 1 month of age and continuous 17 beta-estradiol (E2) and/or progesterone (P) replacement on the uterus of BALB/cCrgl mice neonatally treated with diethylstilbestrol [(DES) CAS: 56-53-1; alpha-alpha'-diethyl-4,4'-stilbenediol] or sesame oil were recorded after 1, 4, 7, and 10 months of treatment. DES-exposed uteri were found to be hypoplastic, less responsive to the growth-promoting effects of E2, and more likely to develop smooth muscle abnormalities after continuous hormonal treatment than similarly treated control uteri. Neonatal DES treatment led to leukocytic infiltration, disruption in the organization of the inner circular smooth muscle layer, and development of a population of epithelial cells believed to respond to later E2 treatment by proliferation and stratification (squamous metaplasia). Qualitative and quantitative responses to continuous P treatment and the development of cystic glandular hyperplasia and adenomyosis were found to be unaltered by neonatal DES administration. The relevance of these results to the problems of uterine abnormalities observed in women exposed to DES in utero is discussed.

Long-term effects of neonatal steroid exposure on mammary gland development and tumorigenesis in mice.

Newborn female mice of three strains--BALB/cfC3H [mammary tumor virus (MuMTV)-infected], BALB/c, and C57BL (both virus-free)--were given injections of 17beta-estradiol or testosterone, alone or in combination with ovine prolactin, for the first 5 days of life. Half of each group of mice were ovariectomized at 40 days of age, and all mice were killed between 6 and 16 months of age. Mammary glands of BALB/cfC3H mice receiving steroid hormones were better developed than those of mice not receiving steroids. Androgen induced a higher incidence of grossly dilated ducts and secretion-filled alveoli. Mammary nodule and tumor incidences were higher in steroid-treated mice than in controls; androgen resulted in higher incidences than did estrogen. The age of onset of mammary tumors was also earlier after neonatal steroid treatment. In BALB/c mice, neonatal injections of estrogen induced some alveolar development of the mammary gland; neonatal injections of ovine prolactin had a greater effect. The mammary glands of C57BL mice did not show any evidence of stimulation by neonatal hormone treatment, which indicated the probability of strain differences. However, no nodules or tumors occurred in these MuMTV-free strains. Therefore, MuMTV was essential for neoplastic mammary responses to neonatal hormone treatment. Ovariectomy prevented alveolar development and abnormal changes in the mammary glands of all groups, thus indicating that ovary-independent alterations in the mammary gland were not induced by neonatal steroid treatment. We concluded that neonatal steroid exposure resulted in increased mammary tumor risk in mice, but only in the presence of both MuMTV and ovaries.

Cervicovaginal and mammary gland abnormalities in BALB/cCrgl mice treated neonatally with progesterone and estrogen, alone or in combination.

Neonatal female BALB/cCrgl mice (mammary tumor virus unexpressed) were given a daily injection of estradiol and/or progesterone for 5 days, beginning within 36 hr after birth. About one-half of each group was ovariectomized when 40 days old, and all mice were killed when between 18.5 and 26 months of age. Neonatal progesterone leads to ovary-dependent persistent vaginal cornification and hyperplasia. In addition, 16 of the 24 progesterone-treated mice had genital tract lesions, and 4 of these showed predominantly glandular features. No such lesions were observed in either oil-treated or untreated mice. Lesions were also observed in both intact and ovariectomized mice treated with estrogen-progesterone combinations, but most of the lesions were not as severe as those seen in mice treated neonatally with progesterone alone, and they were predominantly squamous in appearance. Although mammary tumors were not observed in either the control or the neonatally steroid-treated intact mice, many in the latter groups possessed hyperplastic alveolar-like mammary nodules and other abnormalities.

Long-term effects of neonatal treatment with progesterone, alone and in combination with estrogen, on the mammary gland and reproductive tract of female BALB/cfC3H mice.

Neonatal female mice of the BALB/cfC3H strain were given 5 daily injections of 17beta-estradiol and progesterone alone and in combination, beginning within 36 hr after birth. Half of the mice in each group were ovariectomized at 40 days of age, and all were killed at tumor age of at 12 months of age. Mice receiving progesteron (100 mug daily) alone showed ovary-dependent persistent vaginal cornification. When neonatal progesterone and estradiol were given concurrently, the occurrence of persistent vaginal cornification was significantly lower than in mice receiving neonatal estradiol treatment alone. Progesterone alone produced hyperplastic downgrowths and lesions of both vaginal and cervical epithelia, but to a lesser degree than occurred in mice treated neonatally with estrogen. When progesterone was given concurrently with 17beta-estradiol, the incidence of lesions was lower but their severity was greater. The low doses of 17beta-estradiol and progesterone each resulted in an earlier age of onset and a higher incidence of mammary tumors; this also occurred after both combined estrogen-progesterone treatments. In treated mice ovariectomized on Day 40, normal mammary development did not occur and mammary tumors failed to appear, regardless of neonatal treatment. The data indicate a clear effect of neonatal progesterone exposure on both the genital tract and the mammary apparatus of female mice.

Alteration of sodium transport in mouse mammary epithelium associated with neoplastic transformation.

Using electrophysiological techniques we have examined the apical membrane ionic permeabilities of primary cell cultures of the mouse mammary gland in the midpregnant, preneoplastic, and neoplastic states. Membrane Na+ permeability changed with tumorigenesis, whereas K+ and Cl- permeabilities were unaltered. With tracer flux techniques the unidirectional efflux rate constant of 22Na was found to be greater in tumor cells than it is in normal cells. This increase in 22Na efflux was eliminated by the addition of ouabain. The results are interpreted as an increase in Na+ permeability and in Na+-K+-ATPase activity with the neoplastic transformation. The presence or absence of the virus in midpregnant cells does not seem to affect Na+ permeability.

Induction of abnormal epithelial changes by estrogen in neonatal mouse vaginal transplants.

Adenosis occurred in transplanted C57BL and BALB/c mice Müllerian-derived reproductive tract regions, cervix, and/or fornix (FX), and middle vagina but never in the urogenital sinus-derived portion of the vagina, after a 1-mo exposure to endogenous ovarian hormones or exogenous estradiol (E2). Grafts in ovariectomized hosts did not exhibit adenosis, confirming its dependence on estrogen. C57BL FX and midvaginal transplants from 1-, 3-, and 5-day-old donors but not from 7- or 10-day-old donors developed adenosis, indicating a critical period before day 6. Prolonged E2 exposure (to 2 mo) decreased the adenosis incidence observed in the C57BL FX group but not in midvaginal transplants. Progesterone added during the second half of transplantation to continuing exogenous E2 prevented this reduction in the FX group; however, adenosis incidence in the similarly treated middle vagina group was less than that observed after 1 or 2 mo of E2 treatment alone. Progesterone present throughout the 2-mo transplantation period did not significantly affect adenosis incidence induced by 2-mo exposure of midvaginal or FX grafts to E2 alone. Changes suggestive of squamous cell carcinoma were found in a few BALB/c midvaginal grafts after E2 exposure for 1 mo and in some C57BL midvaginal and FX grafts after E2 and progesterone exposure for 2 mo.

Long-term alterations in histology and steroid receptor levels of the genital tract and mammary gland following neonatal exposure of female BALB/cCrgl mice to various doses of diethylstilbestrol.

The relation of the dosage of diethylstilbestrol (DES) administered neonatally to the incidence and severity of genital tract and mammary gland lesions and to the levels of sex hormone receptors was examined using a mouse model for human intrauterine DES exposure. Female BALB/cCrgl mice received various doses of DES (ranging from 5 X 10(-1)-10(-5) micrograms daily for the first 5 days of life) or the sesame oil vehicle alone. In the vagina, at all ages examined (1, 2, 6, and 12 months) cytosolic estrogen receptors are consistently decreased after high doses of neonatal DES (10(-1) and 1 microgram). In contrast, at the same ages, vaginal cytosolic progestin receptors increase after identical doses. In the uterus, the 1-microgram dose of neonatal DES also consistently decreases cytosolic estrogen receptors while increasing cytosolic progestin receptors at 1, 2, and 6 months of age. Histologically, neonatal doses of 5 X 10(-2) micrograms DES result in vaginal lesions at 2 months. With age, this threshold level decreases, implying interaction with an altered hormonal milieu. The uterus shows a sensitivity similar to that of the vagina in regard to the histopathological effects of neonatal DES. The ovary and mammary glands are 10- to 100-fold more sensitive to neonatal DES exposure.

Estrogen and progestin receptors in mouse vaginal epithelium and fibromuscular wall.

Both sodium molybdate and Percoll density gradient stabilize the hormone-binding capacities of the estrogen and progestin receptors and individually increase the recovery of these receptors in prepared cytosols of the separated mouse vaginal epithelium and fibromuscular wall. Their effects are additive. The concentrations of estrogen receptors are similar in the epithelial and fibromuscular compartments, whereas progestin receptor concentrations are higher in the epithelium.

Effects of somatostatin and urotensin II on tilapia pituitary prolactin release and interactions between somatostatin, osmotic pressure Ca++, and adenosine 3',5'-monophosphate in prolactin release in vitro.

Both somatostatin (SRIF) and urotensin II, a dodecapeptide from the teleost caudal neurosecretory system, inhibit PRL release from the organ-cultured rostral pars distalis of the tilapia, Sarotherodon mossambicus, in a dose-related manner. The inhibitory action of SRIF on PRL release was completely prevented by the presence of the calcium ionophore A23187. PRL release was also blocked when Ca++ was excluded from the incubation medium, even in the presence of the ionophore. Both dibutyryl cAMP (dbcAMP) and the phosphodiesterase inhibitor 3-isobutyl-1-methylxanthine, alone or in combination, stimulated PRL release during incubation in high osmotic pressure medium. The effect of dbcAMP appeared to be dose related. Together, dbcAMP and 3-isobutyl-1-methylxanthine were also effective in preventing the inhibition of PRL release by SRIF. These results are consistent with the notion that Ca++, and possibly cAMP, may be important mediators of PRL secretion, and it is likely that SRIF may inhibit PRL release by blocking a Ca++- or cAMP-mediated mechanism.

17 beta-estradiol increases plasma calcitonin levels in salmonid fish.

The functional role of calcitonin in teleost fish is in question. Data on the role of calcitonin in calcium regulation are inconsistent, and while a participation in some aspects of sexual maturation has been strongly indicated, the exact function is not known. To establish if there exists a functional relationship between 17 beta-estradiol (E2) and calcitonin in salmonid species, rainbow trout (Oncorhynchus mykiss), Atlantic salmon (Salmo salar), and coho salmon (Oncorhynchus kisutch) were injected ip with a single or repeated doses of E2. It is concluded that E2 treatment increases plasma calcitonin levels directly or indirectly, and that it is possible that E2 is responsible for the rise in calcitonin levels during late sexual maturation of female salmonids. In accord with earlier studies, no correlation was found between changes in calcitonin levels and free plasma calcium levels. It seems clear that changes in free plasma calcium levels are not the primary cause of the plasma calcitonin changes in teleost fish. It is possible that calcitonin is involved in mobilizing calcium or directing its mobilization by protection of certain calcium pools during vitellogenesis. However, the increase in calcitonin occurs so close to ovulation that a reproductive role other than a calcium regulatory one is likely. The possibility of transfer of calcitonin itself to the developing oocytes and a subsequent role in embryonic development must also be considered.

Experimental diabetes mellitus in a teleost fish. II. Roles of insulin, growth hormone (GH), insulin-like growth factor-I, and hepatic GH receptors in diabetic growth inhibition in the goby, Gillichthys mirabilis.

Insulin-dependent diabetes mellitus (IDDM), when untreated or poorly controlled in mammals, results in growth retardation. To determine whether the same relationship exists in an ectothermic vertebrate, IDDM-like symptoms were induced in a teleost fish, the goby Gillichthys mirabilis, by surgical removal of its pancreatic endocrine (islet) organ. Isletectomized (Ix) gobies lost body weight, their skeletal growth was retarded, as measured by changes in body length, and they exhibited a 50% reduction in cartilage 35SO4 incorporation in vitro, consistent with changes that occur in mammals with IDDM. Injections of bovine insulin into the Ix fish restored body growth parameters to control levels and stimulated cartilage 35SO4 incorporation in a dose-related manner. In contrast to mammals with IDDM, which are resistant to GH action, injection of teleost GH stimulated cartilage 35SO4 incorporation in the Ix fish. Furthermore, whereas cartilage from rats with IDDM is resistant to stimulation by insulin-like growth factor-I (IGF-I) in vitro, cartilage explants from the Ix fish were highly responsive to recombinant bovine IGF-I, exhibiting a dose-dependent stimulation of 35SO4 incorporation. As far as we are aware, these results represent the first demonstration of diabetic growth inhibition in an ectothermic vertebrate. This inhibition is similar to that which occurs in mammals with IDDM in some respects, but is different in others, as the diabetic fish did not develop resistance to growth stimulation by either GH or IGF-I. While these results support a role for insulin in maintaining the GH-IGF-I-growth axis in this ectothermic vertebrate, there may be important differences in the role of insulin in the promotion of anabolic processes.

Development-associated changes in thyroxine kinetics in juvenile salmon.

Smoltification is a transformation that occurs in some species of salmon, during which solitary fish in fresh water become schooling fish and migrate to the sea. This process is accompanied by large increases in plasma T4. T4 secretion rate and other parameters of T4 metabolism in juvenile coho salmon were estimated by applying kinetic analyses to measurements of the disappearance of injected T4 radiotracer from plasma. Studies were performed at the beginning (March) and end (May) of the increase in T4 concentration in fresh water and seawater. Early and intensive sampling permitted characterization of a very fast initial component of the T4 disappearance curve when analyses included a zero time datum derived from an independent estimate of plasma volume. The plasma volume, equal to 1.77% of body weight, was obtained by measuring the disappearance of radiolabeled albumin from the plasma in two other groups of animals in fresh water and seawater. There were 3- to 7-fold changes in T4 production, distribution, and metabolism between March and May, whereas environment (fresh water vs. seawater) had relatively minor effects on T4 kinetics. In fresh water, the T4 secretion rate was 4.48 ng/h in March and 1.50 ng/h in May. The total T4 pool size was 37.8 ng in March and 12.2 ng in May. Plasma-tissue T4 fluxes were 3- to 7-fold greater in May. Relatively less T4 was distributed in tissue in May (63% vs. 83%), and T4 spent much less time in tissue in May than in March during each pass through the tissue space (11 min vs. 3.1 h). We propose that the difference in secretion rate and a redistribution of T4 between blood and tissues contribute to both the rise and fall in the plasma T4 concentration between March and May. Changes in T4 kinetics during salmonid smoltification resemble those occurring during amphibian metamorphosis and mammalian gestation and neonatal life, and may reflect an increased requirement and an important role for thyroid hormones during periods of rapid development in vertebrates in general.

Growth responses of prostatic epithelial cells from male mice neonatally exposed to diethylstilbestrol in serum-free collagen gel culture.

Growth of anterior and ventral prostatic epithelial cells from mice neonatally exposed to diethylstilbestrol (DES) and from unexposed control mice was compared at different time points in serum-free collagen gel culture. A longer maintenance of the initial plating density (lag in growth) was observed in cultured DES-exposed ventral prostatic cells. Neonatal DES exposure resulted in two colony types: one similar to colonies arising from unexposed cells and one which appears to be non-growing. Keratinization was observed in some DES-exposed anterior prostatic cell colonies. Removal of epidermal growth factor from the serum-free medium significantly decreased growth in 3 of the 4 groups compared with their growth in the complete serum-free medium.

Growth of mammary epithelial cells from neonatally sex hormone-exposed mice in serum-free collagen gel culture.

Neonatal sex hormone treatments are known to cause an increase in mammary tumors in female mice with expressed mammary tumor virus (MTV). The growth of mammary epithelial cells from mice treated neonatally with sex hormones was studied in response to growth-stimulatory factors in a serum-free collagen gel culture system which sustains the growth of normal mammary epithelial cells. Animals were treated with hormones or oil-vehicle for the first 5 days after birth. Cells from control mice at 2 and 3 months of age showed a maximal growth response to insulin at 5-10 micrograms/ml and LiCl at 5-20 mM. Cells responded to epidermal growth factor at all concentrations used (1, 10 and 50 ng/ml). In contrast, mammary epithelial cells from mice treated neonatally with estrogen (estradiol and diethylstilbestrol (DES] showed a reduced growth response to the growth factors tested.

Vaginal abnormalities in ovariectomized BALB/cCrgl mice after neonatal exposure to different doses of diethylstilbestrol.

Newborn BALB/cCrgl female mice received five daily injections of various doses of diethylstilbestrol (DES), 0.0001-10 micrograms. Mice were killed at 6 days of age or at 4 months after ovariectomy at 40-42 days. Subepithelial nodules of polygonal cells in the upper (Mullerian) vagina during early postnatal life were associated with the later occurrence of ovary-independent persistent stratification with or without cornification in mice treated neonatally with 0.1-10 micrograms DES and thus are a possible predictor of this phenomenon. The thresholds for the induction of ovary-independent epithelial pegs, downgrowths and adenosis (glandular formations) were 0.1 microgram and 0.5 microgram DES/day, respectively.