0013-7227/91/1293-1221$03.00/0 Endocrinology Copyright (0 1991 by The Endocrine Society
Vol. 129, No. 3 Printed in U.S.A.
Fluctuations in the Proportions of Growth Hormone- and Prolactin-Secreting Cells during the Bovine Estrous Cycle* RHONDA D. KINEMAN, DONALD M. HENRICKS, WILLIAM J. FAUGHT, AND L. STEPHEN FRAWLEY Division of Molecular and Cellular Endocrinology, Department of Anatomy and Cell Biology, Medical University of South Carolina (R.D.K., W.J.F., L.S.F.), Charleston, South Carolina 29425; and Department of Animal, Dairy, and Veterinary Sciences, Clemson University (D.M.H.), Clemson, South Carolina 29631
ABSTRACT. Reverse hemolytic plaque assays were performed on monodispersed pituitary cells from cattle at various stages of the estrous cycle in an attempt to determine whether short term fluctuations in the gonadal steroid milieu influenced the proportions of pituitary cells that released GH and/or PRL. Phase of the estrous cycle was initially determined by gross ovarian morphology and later confirmed by determination of estradiol-17/3 and progesterone peripheral serum concentrations. Animals were subdivided into four groups according to phase of cycle: early luteal (EL; day 1-10), midluteal (ML; day 11-16), late luteal (LL; day 17-19), and follicular (F; day 20-21). Plaque assays demonstrated that the percentage of all pituitary cells that released PRL was greater in the EL phase than during the ML or F phases, whereas the relative abundance of GH-secreting
M
cells remained unchanged. A more critical analysis of hormonesecreting subtypes revealed that the increase in total PRL secretors could be attributed almost exclusively to an increase in the abundance of those cells that released both GH and PRL (mammosomatotropes). Accompanying this augmentation of dual hormone-secretors was a decrease in the proportion of cells that released GH alone without a change in the abundance of cells that secreted only PRL. These results strongly suggest that during the estrous cycle there is a bidirectional interconversion among cells that release only GH and mammosomatotropes. Moreover, the relationship between ratios of acidophilic subpopulations and stage of reproductive cycle indicates that ovarian steroids may regulate this phenomenon. (Endocrinology 129: 1221-1225, 1991)
OUNTING evidence suggests that pituitary acidophils (cells which secrete GH and/or PRL) are a dynamic cell population capable of functional interconversion when presented with the appropriate steroidal cue from the gonads. Such a process might enable an animal to satisfy changing requirements for GH or PRL without modifying appreciably the total number of acidophils within the pituitary gland. Indeed, recent studies demonstrate clearly that the proportions of pituitary cells which produce GH, PRL, or both hormones (mammosomatotropes) fluctuate considerably throughout the annual reproductive cycles of bats and musk shrews and during the 6-week transition from early pregnancy through the conclusion of lactation in rats (1-3). Yet to be resolved, however, is the issue of whether more dynamic oscillations in circulating steroids (such as occur during the estrous cycle) could influence the proportions
of acidophilic phenotypes. Preliminary attempts in our laboratory to detect fluctuations in acidophilic subpopulations during the rat estrous cycle have yielded equivocal results. This observation was not totally unexpected since the elevations of estradiol and progesterone during this 4-day cycle are extremely brief when compared with those of other species, and may be too short lived to alter the functional phenotypes of pituitary acidophils (4). Accordingly, we sought a species with a more extended estrous cycle for use as an experimental model. The cow seemed to satisfy our requirement since the estrous cycle is 21 days in mean length with a follicular phase lasting approximately 3 days and an extended luteal phase of 16-18 days (5, 6). Therefore, in the present study we utilized reverse hemolytic plaque assays to characterize functional subtypes of acidophils in pituitary cell cultures derived from cows in various stages of the estrous cycle.
Received April 2, 1991. Address all correspondence and requests for reprints to: Dr. L. Stephen Frawley, Department of Anatomy and Cell Biology, Medical University of South Carolina, 171 Ashley Avenue, Charleston, South Carolina 29425. * This work was supported by USDA Grant 9000720 (to L.S.F.) and National Research Service Award DK-08420 (to R.D.K.).
Materials and Methods Tissue source and reproductive status of the tissue donors Anterior pituitaries from randomly cycling females of various beef breeds were obtained from Clemson University (Clemson,
1221
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 13 November 2015. at 20:58 For personal use only. No other uses without permission. . All rights reserved.
1222
GH- AND PRL-SECRETING CELLS DURING BOVINE ESTROUS CYCLE
SC; n = 6) or a commercial slaughterhouse (Brown Packing Co., Inc., Gaffney, SC; n = 12). Studies were conducted between November 1990 and January 1991. The specific phase of the cycle was estimated by examination of ovarian structures at the time of slaughter (i.e. corpora luteal color and size, and follicular size) in conjunction with assessment of serum concentrations (jugular/carotid samples were taken during exsanguination or cardiac samples were obtained approximately 10 min postmortem) of estradiol-17/3 (E2) and progesterone (P4) by RIA (7, 8). Animals were chosen for this study only if a corpus albicans was present from the previous cycle. Data from only one animal were later excluded because serum steroid concentrations were inconsistent with ovarian structure thus precluding an accurate estimate of ovarian stage. This animal had an undetectable E2 level in the presence of follicles greater than 20 mm in diameter. Remaining animals were assigned to one of four groups: early luteal (EL; day 1-10, n = 5), midluteal (ML; day 11-16, n = 5), late luteal (LL; day 17-19, n = 4), and follicular (F; day 20-21, n = 4). The estimated day of ovulation was designated day 1. Preparation of monodispersed pituitary cells The enzymatic dispersion of bovine pituitaries has been previously described by Kineman et al. (9). In brief, anterior pituitaries were excised within 25 min of slaughter and placed in ice-cold Hank's balanced salt solution (calcium and magnesium free) containing 25 mM HEPES, 0.1% BSA, 100 U/ml penicillin, and 100 Mg/ml streptomycin (pH 7.35) for transport. Upon return to our laboratory, the tissue was diced into 1 mm3 pieces and digested using a solution of minimum essential medium containing 0.1% collagenase type I (138 U/mg), 0.1% hyaluronidase, 0.1% deoxyribonuclease I, 0.01% soybean trypsin inhibitor, 0.1% BSA, nonessential amino acids, antibiotics, and 25 mM HEPES. Liberated cells were washed free of enzymes with Dulbecco's modified Eagle's medium containing 0.1% BSA, nonessential amino acids, and antibiotics. The latter medium is henceforth referred to as DMEM. Monodispersed cells were then cultured for 18 h at 37 C in a water-saturated atmosphere of 95% air:5% CO2 in DMEM containing 10% heat-inactivated horse serum. All culture media and supplements were purchased from Grand Island Biological Co. (Grand Island, NY). Enzymes were obtained from Sigma (St. Louis, MO) with the exception of collagenase type I, which was purchased from Worthington Biochemicals (Freehold, NJ). Simultaneous plaque assay for determination of single and dual hormone secretors A detailed procedure of the simultaneous plaque assay has been described by Kineman et al. (10). In brief, monodispersed cells were combined with protein-A-coated ovine red blood cells (oRBC), and this mixture (1.75 X 104 cells/ml in a 12% oRBC suspension) was infused into an assay chamber where the cells attached as a monolayer after a 45-min incubation (37 C; 95% air:5% CO2). The chambers were then washed free of unattached cells and subsequently filled with DMEM containing one of the following combinations: 1) rabbit antibovine GH serum (1:80, R-l-1-3; Dr. T. Elsasser, USDA, Beltsville, MD) and rabbit antiovine PRL serum (1:160, DJB-9-026; Dr. D. J. Bolt, USDA Animal Hormone Program); 2) GH antiserum and
Endo • 1991 Vol 129 • No 3
1:160 normal rabbit serum; or 3) PRL antiserum and 1:80 normal rabbit serum. Incubation with antibodies was performed in the presence of bovine GRF (10~7 M; Sigma) and TRH (10~7 M, Peninsula Laboratories, Inc., Belmont, CA). Cells were incubated with antibody solution for 18 h and subsequently incubated with a 1:100 dilution of guinea pig serum as a source of complement (Grand Island Biological Co.). Plaque development occurred within 50 min after addition of complement, and the reaction was stopped by infusing fixative. Cell monolayers were stained with toluidine blue in fixative to facilitate visualization of pituitary cells in the oRBC mat. To assess the proportion of cells that released hormone, at least 200 cells were counted on each slide, and three slides were analyzed for each treatment from a single experiment. Quantification of the relative abundance of plaque-forming cells on slides incubated with both GH and PRL antibodies simultaneously provides information on the percentage of all acidophils (cells which secrete GH, PRL, or both hormones), whereas incubation with either GH or PRL antiserum alone allows for the quantification of cells which secrete GH or PRL, respectively. If a population of dual-hormone secretors is present, the sum of the proportion of GH- and PRL-secreting cells obtained from assays performed with each antibody separately will be greater than the percentage of acidophils as determined by simultaneous application of both antisera. Therefore, we used the following equations to estimate the relative abundance of single- and dual-hormone secretors: % acidophils - % GH antisera alone = % PRL only; % acidophils - % PRL antisera alone = % GH only; % acidophils - (% GH only + % PRL only) = % mammosomatotropes. A more detailed description of these calculations has been previously published (10). Statistical analysis Differences associated with reproductive status were determined by analysis of variance followed by Duncan's multiple range test.
Results The initial estimation of the day of the reproductive cycle was made by examination of ovarian structures, and this assessment was later confirmed by quantification of serum E2 and P 4 concentrations. The mean serum concentration of E2 and P 4 for animals in each of the four cycle phases (EL, ML, LL, and F) is presented in Fig. 1. In the EL phase, both P 4 and E2 concentrations were low. During the ML phase, E2 concentrations did not differ from those of EL whereas P 4 values were elevated (P < 0.05). Animals in the LL phase had P 4 concentrations comparable to those of the ML phase while E2 values were far greater (P < 0.05) than ML values. Last, in the F phase P 4 concentrations fell to EL levels while E2 values remained high. The proportion of anterior pituitary cells that secreted either GH or PRL is presented in Fig. 2. The percentages of PRL secretors from animals in the ML and F phase were significantly lower than those of EL phase animals
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 13 November 2015. at 20:58 For personal use only. No other uses without permission. . All rights reserved.
GH- AND PRL-SECRETING CELLS DURING BOVINE ESTROUS CYCLE
1223
25-,
T
w z o cr UJ
CO UJ
o o cc
CO
50-
cc UJ 2 cc
• 40 -
T
O UL UJ
o
3020-
a 10 -
Q.
0 -
EL EL
ML
LL
ML
LL
OVARIAN STATUS
OVARIAN STATUS FlG. 1. Serum concentrations of E2 and P4 for cows in various stages of the estrous cycle (EL, ML, LL, F). In this and subsequent figures, values represent the mean and SE and significantly different means (P < 0.05) are identified by different letters.
ALL GH
FIG. 3. Percentage of all pituitary cells that formed plaques when both GH and PRL antisera were used concurrently. This combined approach was used to estimate the size of the entire acidophil population. Pituitary cell cultures were derived from cattle in the EL, ML, LL, or F phase of the estrous cycle.
tained fewer mammosomatotropes when compared with those from EL and F phase animals, whereas the ovarian status of the donor did not influence the percentage of cells that released only PRL.
ALL PRL
Discussion 20 10-
ML
LL
OVARIAN STATUS
FIG. 2. Proportions of all pituitary cells that released PRL or GH. Donor cows were in the EL, ML, LL, or F phase of the estrous cycle. The proportions of hormone-secreting cells were determined after an 18-h incubation with GH antiserum (ALL GH) or PRL antiserum (ALL PRL).
(P < 0.05) but did not differ from those in the LL phase. In contrast, the relative abundance of GH secretors remained relatively constant throughout all four ovarian phases. Interestingly, the fluctuations in the percentages of PRL-secreting cells during the cycle were not accompanied by proportional changes in the percentages of total acidophils (Fig. 3). The relative abundance of anterior pituitary cells that secreted GH only, PRL alone, or both hormones was calculated from data contained in Figs. 2 and 3 and is presented in Fig. 4. All three acidophilic cell types were present in pituitary cultures derived from cows in each phase of the cycle, and their proportions varied with the reproductive stage of the donor animal. The percentage of all pituitary cells that secreted only GH was larger in pituitary cultures from ML animals than in those from the EL, LL, or F phase of the cycle (P < 0.05). Conversely, pituitary cell cultures from ML phase cows con-
These results are the first to demonstrate a definitive relationship between stage of estrous cycle and changes in functional phenotype within the acidophilic cell population of the adenohypophysis. Specifically, the overall abundance of cells that released PRL was reduced in the ML and F phases relative to EL phase values, whereas the proportion of GH secretors was maintained. Further subdivision of GH- and PRL-releasing cells into single and dual hormone secretors revealed dramatic shifts in the relative abundance of these subpopulations at various stages of the estrous cycle. Inasmuch as these shifts were not accompanied by concomitant changes in the total number of acidophils, it seems reasonable to propose that bovine acidophils, like those of other species (1-3), are capable of undergoing functional interconversions. This possibility is supported further by the apparent dynamics of acidophil shifts. For instance, the drop in mammosomatotropes that occurred during the transition from the EL to the ML phase was accompanied by a commensurate increase in GH-only cells, suggesting that a subpopulation of dual-hormone secretors lost the ability to release PRL while maintaining GH-secretory capacity. Bidirectionality of this interconversion is suggested by the resurgence of mammosomatotropes in the LL and F phases. The fact that the latter took place in the absence of a change in the proportion of PRL-only cells and total acidophils indicates that a subpopulation of GH-secreting cells could have reacquired the ability to release PRL. The dynamic nature of the acidophilic population has
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 13 November 2015. at 20:58 For personal use only. No other uses without permission. . All rights reserved.
GH- AND PRL-SECRETING CELLS DURING BOVINE ESTROUS CYCLE
1224
EL
ML
LL
F
20
c/>
c
MS
DC LLJ
T
BC
DC O LL LU
AB
A
T
T
EL
ML
LL
F
EL
ML
LL
10-
T
OVARIAN STATUS FIG. 4. Proportions of pituitary cells from cattle in the EL, ML, LL, and F phases that released only GH (GH), only PRL (PRL), or both GH and PRL (MS) as calculated from data contained in Figs. 2 and 3.
also been observed in a variety of other reproductive situations. For example, immunocytochemical techniques have been used to colocalize GH and PRL to the same pituitary cell (and in some instances to the same secretory granule) of the rat and musk shrew (1). In these studies it was found that mammosomatotropes were more prevalent in pituitaries of pregnant animals than in those of lactators or virgin females. In a companion
Endo• 1991 Voll29«No3
study (2) these same investigators noted that bihormonal cells were numerous during pregnancy in the bat and accounted for most, if not all, PRL cells during hibernation and the period of arousal prior to fertilization. Such shifts in acidophils have also been documented by reverse hemolytic plaque assay techniques. Porter et al. (3) observed that the relative abundance of PRL-secreting cells increased and that of GH secretors decreased during the transition from early pregnancy through lactation without changes in the proportion of total acidophils. These reciprocal shifts in the relative proportions of PRL and GH secretors appeared to be the result of an increment in PRL-only cells which was preceded by both a loss of mammosomatotropes and GH-only cells. Follow-up studies by the same group demonstrate further the flexibility of the acidophilic subpopulations because the dramatically altered ratios of single and dual hormone secretors found at the end of lactation returned to prepregnancy levels within 4 days of weaning (11). This rapid conversion of acidophil secretory capacity was also observed in the present study where over a relatively short period of approximately 7 days (from the EL to ML phase) there was a dramatic reduction of dual hormone secretors which was reestablished as the cycle advanced into the LL and F phases. Although this is the first report of short term, cyclic changes in acidophil cell function associated with ovarian secretory patterns, the idea that the steroid hormone milieu could modulate the interconversion of GH and PRL cells is not a novel one. This possibility was first suggested by Goluboff and Ezrin (12) who identified acidophils by tinctorial methods and reported that the abundance of PRL cells increased while that of GH cells decreased as pregnancy advanced in the human. The possibility that these reciprocal shifts might be induced by placental steroids (presumably estrogen) was suggested by the observation that exogenous estrogen therapy induced radiolabeled somatotropes to take on the morphological characteristics of PRL-producing cells in the rat (13). Direct evidence that estrogens might subserve such a modulatory role in shaping acidophilic phenotype was provided by Boockfor et al. (14) who cultured monodispersed pituitary cells from male rats in the presence or absence of E2 for 6 days and then performed reverse hemolytic plaque assays to assess the ability of individual cells to release GH, PRL, or both hormones. They found that E2 caused a marked increase in the proportion of mammosomatotropes and a commensurate decrease in the relative abundance of cells that released GH alone. An effect of gonadal steroids on the ratios of GH-only and dual hormone secretors also has been observed recently (10) in experiments with bulls and steers (castrates). In gonad-intact bulls approximately 9% of all pituitary cells released both GH and PRL, whereas
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 13 November 2015. at 20:58 For personal use only. No other uses without permission. . All rights reserved.
GH- AND PRL-SECRETING CELLS DURING BOVINE ESTROUS CYCLE
21% released only GH and 45% PRL alone. This contrasts markedly with the situation in steers where mammosomatotropes comprised 22% of the pituitary cell population and cells that released GH only or PRL only accounted for 8% and 46%, respectively. These results, coupled with those of the present study clearly indicate that gonadal status (presumably steroidogenic in nature) has a profound effect on the secretory phenotype of bovine acidophils. The shift toward more PRL secretors (owing primarily to an increase in mammosomatotropes) during the onset of the follicular phase is not easily reconciled with the pattern of circulating PRL during the bovine estrous cycle. One might logically expect an increment in PRL cell number to be accompanied by a commensurate increase in serum PRL concentrations. However, circulating PRL levels are highly variable throughout the estrous cycle in cows and do not appear to be directly influenced by fluctuations in the gonadal steroids (15). Although an increase in serum PRL occasionally occurs at estrus in cattle, this is believed to be a consequence of lordosis behavior since a rise in PRL concentrations is not observed if animals are prohibited from interacting physically (16, 17). In addition, ovariectomy has only a transient effect on serum PRL with levels falling slightly 6 h post ovariectomy but returning to preoperative levels by 10 h (18). Likewise, acute and chronic estrogen replacement does not profoundly affect circulating PRL in cows (18). Taken together, one might question whether the bovine pituitary is capable of responding to gonadal steroids. However, this does not appear to be the case since high affinity estrogen and androgen receptors have been identified in bovine pituitaries (19), and treatment of bovine pituitary cell cultures with estrogen augments PRL synthesis and release (20). In light of these discrepancies concerning regulation of PRL secretion by gonadal steroids, it seems premature to speculate further about their physiological relevance to our observations of cyclic shifts among acidophilic subtypes in cattle. Acknowledgments The authors wish to thank Dr. T. Elsasser (USDA) who generously provided us with the bovine GH antiserum and Dr. D. Bolt (USDA) for the gift of the ovine PRL antiserum. We also wish to thank Mrs. Mary Ackerman for help in the preparation of this manuscript.
References 1. Ishibashi T, Shiino M 1989 Co-localization pattern of growth hormone (GH) and prolactin (PRL) within the anterior pituitary cells in the female rat and female musk shrew. Anat Rec 223:185-
1225
193 2. Ishibashi T, Shiino M 1989 Subcellular localization of prolactin in the anterior pituitary cells of the female Japanese house bat, pipstrellus abramus. Endocrinology 124:1056-1063 3. Porter TE, Hill JB, Wiles CD, Frawley LS 1990 Is the mammosomatotrope a transitional cell for the functional interconversion of growth hormone- and prolactin-secreting cells? Suggestive evidence from virgin, gestating, and lactating rats. Endocrinology 127:2789-2794 4. Smith MS, Freeman ME, Neill JD 1975 The control of progesterone secretion during the estrous cycle and early pseudopregnancy in the rat: prolactin, gonadotropin and steroid levels associated with rescue of the corpus luteum of pseudopregnancy. Endocrinology 96:219-226 5. Donaldson LE, Bassett JM, Thorburn GD 1970 Peripheral plasma progesterone concentration of cows during puberty, oestrous cycles, pregnancy and lactation, and the effects of under-nutrition or exogenous oxytocin on progesterone concentration. J Endocrinol 48:599-614 6. Christensen DS, Hopwood ML, Wiltbank JN 1974 Levels of hormones in the serum of cycling beef cows. J Anim Sci 38:577-583 7. Breuel KF, Spitzer JC, Gimenez T, Henricks DM, Gray SL 1988 Effect of holding time and temperature of bovine whole blood on concentration of progesterone, estradiol-17j8 and estrone in plasma and serum samples. Theriogenology 30:613-627 8. Schrick FN, Spitzer TC, Jenkins JC, Henricks DM, Altheon TG 1990 Effect of dietary energy restriction on metabolic and endocrine responses during the estrous cycle of the suckled beef cow. J Anim Sci 68:3313-3321 9. Kineman RD, Faught WJ, Frawley LS 1990 Bovine pituitary cells exhibit a unique form of somatotrope secretory heterogeneity. Endocrinology 127:2229-2235 10. Kineman RD, Faught WJ, Frawley LS 1991 Mammosomatotropes are abundant in bovine pituitaries: influence of gonadal status. Endocrinology 128:2229-2233 11. Porter TE, Wiles CD, Frawley LS 1991 Evidence for bidirectional interconversion of mammotropes and somatotropes: rapid reversion of acidophilic cell types to pregestational proportions after weaning. Endocrinology 129:1215-1220 12. Goluboff LG, Ezrin C 1969 Effect of pregnancy on the somatotroph and the prolactin cell of the human adenohypophysis. J Clin Endocrinol Metab 29:1533-1538 13. Stratmann IE, Ezrin C, Sellers EA 1974 Estrogen-induced transformation of somatotrophs into mammotrophs in the rat. Cell Tissue Res 152:229-238 14. Boockfor FR, Hoeffler JP, Frawley LS 1986 Estradiol induces a shift in cultured cells that release prolactin or growth hormone. Am J Physiol 25O:E1O3-E1O5 15. Swanson LV, Hafs HD 1971 LH and prolactin in blood serum from estrus to ovulation in Holstein heifers. J Anim Sci 33:1038-1041 16. Koprowski JA, Tucker HA 1973 Serum prolactin during various physiological states and its relationship to milk production in the bovine. Endocrinology 92:1480-1487 17. Edgerton LA, Hafs HD 1973 Serum luteinizing hormone, prolactin, glucocorticoid and progestin in dairy cows from calving to gestation. J Dairy Sci 56:451-458 18. Beck TW, Smith VG, Seguin BE, Convey EM 1976 Bovine serum LH, GH and prolactin following chronic implantation of ovarian steroids and subsequent ovariectomy. J Anim Sci 42:461-468 19. Armstrong Jr EG, Villee CA 1977 Characterization and comparison of estrogen and androgen receptors of calf anterior pituitary. J Steroid Biochem 8:258-292 20. Padmanabhan V, Convey EM 1979 Estradiol-170 stimulates basal and thyrotropin releasing hormone induced prolactin secretion by bovine pituitary cells in primary culture. Mol Cell Endocrinol 14:103-112
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 13 November 2015. at 20:58 For personal use only. No other uses without permission. . All rights reserved.
Vol. 129, No. 3 Printed in U.S.A.
Fluctuations in the Proportions of Growth Hormone- and Prolactin-Secreting Cells during the Bovine Estrous Cycle* RHONDA D. KINEMAN, DONALD M. HENRICKS, WILLIAM J. FAUGHT, AND L. STEPHEN FRAWLEY Division of Molecular and Cellular Endocrinology, Department of Anatomy and Cell Biology, Medical University of South Carolina (R.D.K., W.J.F., L.S.F.), Charleston, South Carolina 29425; and Department of Animal, Dairy, and Veterinary Sciences, Clemson University (D.M.H.), Clemson, South Carolina 29631
ABSTRACT. Reverse hemolytic plaque assays were performed on monodispersed pituitary cells from cattle at various stages of the estrous cycle in an attempt to determine whether short term fluctuations in the gonadal steroid milieu influenced the proportions of pituitary cells that released GH and/or PRL. Phase of the estrous cycle was initially determined by gross ovarian morphology and later confirmed by determination of estradiol-17/3 and progesterone peripheral serum concentrations. Animals were subdivided into four groups according to phase of cycle: early luteal (EL; day 1-10), midluteal (ML; day 11-16), late luteal (LL; day 17-19), and follicular (F; day 20-21). Plaque assays demonstrated that the percentage of all pituitary cells that released PRL was greater in the EL phase than during the ML or F phases, whereas the relative abundance of GH-secreting
M
cells remained unchanged. A more critical analysis of hormonesecreting subtypes revealed that the increase in total PRL secretors could be attributed almost exclusively to an increase in the abundance of those cells that released both GH and PRL (mammosomatotropes). Accompanying this augmentation of dual hormone-secretors was a decrease in the proportion of cells that released GH alone without a change in the abundance of cells that secreted only PRL. These results strongly suggest that during the estrous cycle there is a bidirectional interconversion among cells that release only GH and mammosomatotropes. Moreover, the relationship between ratios of acidophilic subpopulations and stage of reproductive cycle indicates that ovarian steroids may regulate this phenomenon. (Endocrinology 129: 1221-1225, 1991)
OUNTING evidence suggests that pituitary acidophils (cells which secrete GH and/or PRL) are a dynamic cell population capable of functional interconversion when presented with the appropriate steroidal cue from the gonads. Such a process might enable an animal to satisfy changing requirements for GH or PRL without modifying appreciably the total number of acidophils within the pituitary gland. Indeed, recent studies demonstrate clearly that the proportions of pituitary cells which produce GH, PRL, or both hormones (mammosomatotropes) fluctuate considerably throughout the annual reproductive cycles of bats and musk shrews and during the 6-week transition from early pregnancy through the conclusion of lactation in rats (1-3). Yet to be resolved, however, is the issue of whether more dynamic oscillations in circulating steroids (such as occur during the estrous cycle) could influence the proportions
of acidophilic phenotypes. Preliminary attempts in our laboratory to detect fluctuations in acidophilic subpopulations during the rat estrous cycle have yielded equivocal results. This observation was not totally unexpected since the elevations of estradiol and progesterone during this 4-day cycle are extremely brief when compared with those of other species, and may be too short lived to alter the functional phenotypes of pituitary acidophils (4). Accordingly, we sought a species with a more extended estrous cycle for use as an experimental model. The cow seemed to satisfy our requirement since the estrous cycle is 21 days in mean length with a follicular phase lasting approximately 3 days and an extended luteal phase of 16-18 days (5, 6). Therefore, in the present study we utilized reverse hemolytic plaque assays to characterize functional subtypes of acidophils in pituitary cell cultures derived from cows in various stages of the estrous cycle.
Received April 2, 1991. Address all correspondence and requests for reprints to: Dr. L. Stephen Frawley, Department of Anatomy and Cell Biology, Medical University of South Carolina, 171 Ashley Avenue, Charleston, South Carolina 29425. * This work was supported by USDA Grant 9000720 (to L.S.F.) and National Research Service Award DK-08420 (to R.D.K.).
Materials and Methods Tissue source and reproductive status of the tissue donors Anterior pituitaries from randomly cycling females of various beef breeds were obtained from Clemson University (Clemson,
1221
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 13 November 2015. at 20:58 For personal use only. No other uses without permission. . All rights reserved.
1222
GH- AND PRL-SECRETING CELLS DURING BOVINE ESTROUS CYCLE
SC; n = 6) or a commercial slaughterhouse (Brown Packing Co., Inc., Gaffney, SC; n = 12). Studies were conducted between November 1990 and January 1991. The specific phase of the cycle was estimated by examination of ovarian structures at the time of slaughter (i.e. corpora luteal color and size, and follicular size) in conjunction with assessment of serum concentrations (jugular/carotid samples were taken during exsanguination or cardiac samples were obtained approximately 10 min postmortem) of estradiol-17/3 (E2) and progesterone (P4) by RIA (7, 8). Animals were chosen for this study only if a corpus albicans was present from the previous cycle. Data from only one animal were later excluded because serum steroid concentrations were inconsistent with ovarian structure thus precluding an accurate estimate of ovarian stage. This animal had an undetectable E2 level in the presence of follicles greater than 20 mm in diameter. Remaining animals were assigned to one of four groups: early luteal (EL; day 1-10, n = 5), midluteal (ML; day 11-16, n = 5), late luteal (LL; day 17-19, n = 4), and follicular (F; day 20-21, n = 4). The estimated day of ovulation was designated day 1. Preparation of monodispersed pituitary cells The enzymatic dispersion of bovine pituitaries has been previously described by Kineman et al. (9). In brief, anterior pituitaries were excised within 25 min of slaughter and placed in ice-cold Hank's balanced salt solution (calcium and magnesium free) containing 25 mM HEPES, 0.1% BSA, 100 U/ml penicillin, and 100 Mg/ml streptomycin (pH 7.35) for transport. Upon return to our laboratory, the tissue was diced into 1 mm3 pieces and digested using a solution of minimum essential medium containing 0.1% collagenase type I (138 U/mg), 0.1% hyaluronidase, 0.1% deoxyribonuclease I, 0.01% soybean trypsin inhibitor, 0.1% BSA, nonessential amino acids, antibiotics, and 25 mM HEPES. Liberated cells were washed free of enzymes with Dulbecco's modified Eagle's medium containing 0.1% BSA, nonessential amino acids, and antibiotics. The latter medium is henceforth referred to as DMEM. Monodispersed cells were then cultured for 18 h at 37 C in a water-saturated atmosphere of 95% air:5% CO2 in DMEM containing 10% heat-inactivated horse serum. All culture media and supplements were purchased from Grand Island Biological Co. (Grand Island, NY). Enzymes were obtained from Sigma (St. Louis, MO) with the exception of collagenase type I, which was purchased from Worthington Biochemicals (Freehold, NJ). Simultaneous plaque assay for determination of single and dual hormone secretors A detailed procedure of the simultaneous plaque assay has been described by Kineman et al. (10). In brief, monodispersed cells were combined with protein-A-coated ovine red blood cells (oRBC), and this mixture (1.75 X 104 cells/ml in a 12% oRBC suspension) was infused into an assay chamber where the cells attached as a monolayer after a 45-min incubation (37 C; 95% air:5% CO2). The chambers were then washed free of unattached cells and subsequently filled with DMEM containing one of the following combinations: 1) rabbit antibovine GH serum (1:80, R-l-1-3; Dr. T. Elsasser, USDA, Beltsville, MD) and rabbit antiovine PRL serum (1:160, DJB-9-026; Dr. D. J. Bolt, USDA Animal Hormone Program); 2) GH antiserum and
Endo • 1991 Vol 129 • No 3
1:160 normal rabbit serum; or 3) PRL antiserum and 1:80 normal rabbit serum. Incubation with antibodies was performed in the presence of bovine GRF (10~7 M; Sigma) and TRH (10~7 M, Peninsula Laboratories, Inc., Belmont, CA). Cells were incubated with antibody solution for 18 h and subsequently incubated with a 1:100 dilution of guinea pig serum as a source of complement (Grand Island Biological Co.). Plaque development occurred within 50 min after addition of complement, and the reaction was stopped by infusing fixative. Cell monolayers were stained with toluidine blue in fixative to facilitate visualization of pituitary cells in the oRBC mat. To assess the proportion of cells that released hormone, at least 200 cells were counted on each slide, and three slides were analyzed for each treatment from a single experiment. Quantification of the relative abundance of plaque-forming cells on slides incubated with both GH and PRL antibodies simultaneously provides information on the percentage of all acidophils (cells which secrete GH, PRL, or both hormones), whereas incubation with either GH or PRL antiserum alone allows for the quantification of cells which secrete GH or PRL, respectively. If a population of dual-hormone secretors is present, the sum of the proportion of GH- and PRL-secreting cells obtained from assays performed with each antibody separately will be greater than the percentage of acidophils as determined by simultaneous application of both antisera. Therefore, we used the following equations to estimate the relative abundance of single- and dual-hormone secretors: % acidophils - % GH antisera alone = % PRL only; % acidophils - % PRL antisera alone = % GH only; % acidophils - (% GH only + % PRL only) = % mammosomatotropes. A more detailed description of these calculations has been previously published (10). Statistical analysis Differences associated with reproductive status were determined by analysis of variance followed by Duncan's multiple range test.
Results The initial estimation of the day of the reproductive cycle was made by examination of ovarian structures, and this assessment was later confirmed by quantification of serum E2 and P 4 concentrations. The mean serum concentration of E2 and P 4 for animals in each of the four cycle phases (EL, ML, LL, and F) is presented in Fig. 1. In the EL phase, both P 4 and E2 concentrations were low. During the ML phase, E2 concentrations did not differ from those of EL whereas P 4 values were elevated (P < 0.05). Animals in the LL phase had P 4 concentrations comparable to those of the ML phase while E2 values were far greater (P < 0.05) than ML values. Last, in the F phase P 4 concentrations fell to EL levels while E2 values remained high. The proportion of anterior pituitary cells that secreted either GH or PRL is presented in Fig. 2. The percentages of PRL secretors from animals in the ML and F phase were significantly lower than those of EL phase animals
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 13 November 2015. at 20:58 For personal use only. No other uses without permission. . All rights reserved.
GH- AND PRL-SECRETING CELLS DURING BOVINE ESTROUS CYCLE
1223
25-,
T
w z o cr UJ
CO UJ
o o cc
CO
50-
cc UJ 2 cc
• 40 -
T
O UL UJ
o
3020-
a 10 -
Q.
0 -
EL EL
ML
LL
ML
LL
OVARIAN STATUS
OVARIAN STATUS FlG. 1. Serum concentrations of E2 and P4 for cows in various stages of the estrous cycle (EL, ML, LL, F). In this and subsequent figures, values represent the mean and SE and significantly different means (P < 0.05) are identified by different letters.
ALL GH
FIG. 3. Percentage of all pituitary cells that formed plaques when both GH and PRL antisera were used concurrently. This combined approach was used to estimate the size of the entire acidophil population. Pituitary cell cultures were derived from cattle in the EL, ML, LL, or F phase of the estrous cycle.
tained fewer mammosomatotropes when compared with those from EL and F phase animals, whereas the ovarian status of the donor did not influence the percentage of cells that released only PRL.
ALL PRL
Discussion 20 10-
ML
LL
OVARIAN STATUS
FIG. 2. Proportions of all pituitary cells that released PRL or GH. Donor cows were in the EL, ML, LL, or F phase of the estrous cycle. The proportions of hormone-secreting cells were determined after an 18-h incubation with GH antiserum (ALL GH) or PRL antiserum (ALL PRL).
(P < 0.05) but did not differ from those in the LL phase. In contrast, the relative abundance of GH secretors remained relatively constant throughout all four ovarian phases. Interestingly, the fluctuations in the percentages of PRL-secreting cells during the cycle were not accompanied by proportional changes in the percentages of total acidophils (Fig. 3). The relative abundance of anterior pituitary cells that secreted GH only, PRL alone, or both hormones was calculated from data contained in Figs. 2 and 3 and is presented in Fig. 4. All three acidophilic cell types were present in pituitary cultures derived from cows in each phase of the cycle, and their proportions varied with the reproductive stage of the donor animal. The percentage of all pituitary cells that secreted only GH was larger in pituitary cultures from ML animals than in those from the EL, LL, or F phase of the cycle (P < 0.05). Conversely, pituitary cell cultures from ML phase cows con-
These results are the first to demonstrate a definitive relationship between stage of estrous cycle and changes in functional phenotype within the acidophilic cell population of the adenohypophysis. Specifically, the overall abundance of cells that released PRL was reduced in the ML and F phases relative to EL phase values, whereas the proportion of GH secretors was maintained. Further subdivision of GH- and PRL-releasing cells into single and dual hormone secretors revealed dramatic shifts in the relative abundance of these subpopulations at various stages of the estrous cycle. Inasmuch as these shifts were not accompanied by concomitant changes in the total number of acidophils, it seems reasonable to propose that bovine acidophils, like those of other species (1-3), are capable of undergoing functional interconversions. This possibility is supported further by the apparent dynamics of acidophil shifts. For instance, the drop in mammosomatotropes that occurred during the transition from the EL to the ML phase was accompanied by a commensurate increase in GH-only cells, suggesting that a subpopulation of dual-hormone secretors lost the ability to release PRL while maintaining GH-secretory capacity. Bidirectionality of this interconversion is suggested by the resurgence of mammosomatotropes in the LL and F phases. The fact that the latter took place in the absence of a change in the proportion of PRL-only cells and total acidophils indicates that a subpopulation of GH-secreting cells could have reacquired the ability to release PRL. The dynamic nature of the acidophilic population has
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 13 November 2015. at 20:58 For personal use only. No other uses without permission. . All rights reserved.
GH- AND PRL-SECRETING CELLS DURING BOVINE ESTROUS CYCLE
1224
EL
ML
LL
F
20
c/>
c
MS
DC LLJ
T
BC
DC O LL LU
AB
A
T
T
EL
ML
LL
F
EL
ML
LL
10-
T
OVARIAN STATUS FIG. 4. Proportions of pituitary cells from cattle in the EL, ML, LL, and F phases that released only GH (GH), only PRL (PRL), or both GH and PRL (MS) as calculated from data contained in Figs. 2 and 3.
also been observed in a variety of other reproductive situations. For example, immunocytochemical techniques have been used to colocalize GH and PRL to the same pituitary cell (and in some instances to the same secretory granule) of the rat and musk shrew (1). In these studies it was found that mammosomatotropes were more prevalent in pituitaries of pregnant animals than in those of lactators or virgin females. In a companion
Endo• 1991 Voll29«No3
study (2) these same investigators noted that bihormonal cells were numerous during pregnancy in the bat and accounted for most, if not all, PRL cells during hibernation and the period of arousal prior to fertilization. Such shifts in acidophils have also been documented by reverse hemolytic plaque assay techniques. Porter et al. (3) observed that the relative abundance of PRL-secreting cells increased and that of GH secretors decreased during the transition from early pregnancy through lactation without changes in the proportion of total acidophils. These reciprocal shifts in the relative proportions of PRL and GH secretors appeared to be the result of an increment in PRL-only cells which was preceded by both a loss of mammosomatotropes and GH-only cells. Follow-up studies by the same group demonstrate further the flexibility of the acidophilic subpopulations because the dramatically altered ratios of single and dual hormone secretors found at the end of lactation returned to prepregnancy levels within 4 days of weaning (11). This rapid conversion of acidophil secretory capacity was also observed in the present study where over a relatively short period of approximately 7 days (from the EL to ML phase) there was a dramatic reduction of dual hormone secretors which was reestablished as the cycle advanced into the LL and F phases. Although this is the first report of short term, cyclic changes in acidophil cell function associated with ovarian secretory patterns, the idea that the steroid hormone milieu could modulate the interconversion of GH and PRL cells is not a novel one. This possibility was first suggested by Goluboff and Ezrin (12) who identified acidophils by tinctorial methods and reported that the abundance of PRL cells increased while that of GH cells decreased as pregnancy advanced in the human. The possibility that these reciprocal shifts might be induced by placental steroids (presumably estrogen) was suggested by the observation that exogenous estrogen therapy induced radiolabeled somatotropes to take on the morphological characteristics of PRL-producing cells in the rat (13). Direct evidence that estrogens might subserve such a modulatory role in shaping acidophilic phenotype was provided by Boockfor et al. (14) who cultured monodispersed pituitary cells from male rats in the presence or absence of E2 for 6 days and then performed reverse hemolytic plaque assays to assess the ability of individual cells to release GH, PRL, or both hormones. They found that E2 caused a marked increase in the proportion of mammosomatotropes and a commensurate decrease in the relative abundance of cells that released GH alone. An effect of gonadal steroids on the ratios of GH-only and dual hormone secretors also has been observed recently (10) in experiments with bulls and steers (castrates). In gonad-intact bulls approximately 9% of all pituitary cells released both GH and PRL, whereas
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 13 November 2015. at 20:58 For personal use only. No other uses without permission. . All rights reserved.
GH- AND PRL-SECRETING CELLS DURING BOVINE ESTROUS CYCLE
21% released only GH and 45% PRL alone. This contrasts markedly with the situation in steers where mammosomatotropes comprised 22% of the pituitary cell population and cells that released GH only or PRL only accounted for 8% and 46%, respectively. These results, coupled with those of the present study clearly indicate that gonadal status (presumably steroidogenic in nature) has a profound effect on the secretory phenotype of bovine acidophils. The shift toward more PRL secretors (owing primarily to an increase in mammosomatotropes) during the onset of the follicular phase is not easily reconciled with the pattern of circulating PRL during the bovine estrous cycle. One might logically expect an increment in PRL cell number to be accompanied by a commensurate increase in serum PRL concentrations. However, circulating PRL levels are highly variable throughout the estrous cycle in cows and do not appear to be directly influenced by fluctuations in the gonadal steroids (15). Although an increase in serum PRL occasionally occurs at estrus in cattle, this is believed to be a consequence of lordosis behavior since a rise in PRL concentrations is not observed if animals are prohibited from interacting physically (16, 17). In addition, ovariectomy has only a transient effect on serum PRL with levels falling slightly 6 h post ovariectomy but returning to preoperative levels by 10 h (18). Likewise, acute and chronic estrogen replacement does not profoundly affect circulating PRL in cows (18). Taken together, one might question whether the bovine pituitary is capable of responding to gonadal steroids. However, this does not appear to be the case since high affinity estrogen and androgen receptors have been identified in bovine pituitaries (19), and treatment of bovine pituitary cell cultures with estrogen augments PRL synthesis and release (20). In light of these discrepancies concerning regulation of PRL secretion by gonadal steroids, it seems premature to speculate further about their physiological relevance to our observations of cyclic shifts among acidophilic subtypes in cattle. Acknowledgments The authors wish to thank Dr. T. Elsasser (USDA) who generously provided us with the bovine GH antiserum and Dr. D. Bolt (USDA) for the gift of the ovine PRL antiserum. We also wish to thank Mrs. Mary Ackerman for help in the preparation of this manuscript.
References 1. Ishibashi T, Shiino M 1989 Co-localization pattern of growth hormone (GH) and prolactin (PRL) within the anterior pituitary cells in the female rat and female musk shrew. Anat Rec 223:185-
1225
193 2. Ishibashi T, Shiino M 1989 Subcellular localization of prolactin in the anterior pituitary cells of the female Japanese house bat, pipstrellus abramus. Endocrinology 124:1056-1063 3. Porter TE, Hill JB, Wiles CD, Frawley LS 1990 Is the mammosomatotrope a transitional cell for the functional interconversion of growth hormone- and prolactin-secreting cells? Suggestive evidence from virgin, gestating, and lactating rats. Endocrinology 127:2789-2794 4. Smith MS, Freeman ME, Neill JD 1975 The control of progesterone secretion during the estrous cycle and early pseudopregnancy in the rat: prolactin, gonadotropin and steroid levels associated with rescue of the corpus luteum of pseudopregnancy. Endocrinology 96:219-226 5. Donaldson LE, Bassett JM, Thorburn GD 1970 Peripheral plasma progesterone concentration of cows during puberty, oestrous cycles, pregnancy and lactation, and the effects of under-nutrition or exogenous oxytocin on progesterone concentration. J Endocrinol 48:599-614 6. Christensen DS, Hopwood ML, Wiltbank JN 1974 Levels of hormones in the serum of cycling beef cows. J Anim Sci 38:577-583 7. Breuel KF, Spitzer JC, Gimenez T, Henricks DM, Gray SL 1988 Effect of holding time and temperature of bovine whole blood on concentration of progesterone, estradiol-17j8 and estrone in plasma and serum samples. Theriogenology 30:613-627 8. Schrick FN, Spitzer TC, Jenkins JC, Henricks DM, Altheon TG 1990 Effect of dietary energy restriction on metabolic and endocrine responses during the estrous cycle of the suckled beef cow. J Anim Sci 68:3313-3321 9. Kineman RD, Faught WJ, Frawley LS 1990 Bovine pituitary cells exhibit a unique form of somatotrope secretory heterogeneity. Endocrinology 127:2229-2235 10. Kineman RD, Faught WJ, Frawley LS 1991 Mammosomatotropes are abundant in bovine pituitaries: influence of gonadal status. Endocrinology 128:2229-2233 11. Porter TE, Wiles CD, Frawley LS 1991 Evidence for bidirectional interconversion of mammotropes and somatotropes: rapid reversion of acidophilic cell types to pregestational proportions after weaning. Endocrinology 129:1215-1220 12. Goluboff LG, Ezrin C 1969 Effect of pregnancy on the somatotroph and the prolactin cell of the human adenohypophysis. J Clin Endocrinol Metab 29:1533-1538 13. Stratmann IE, Ezrin C, Sellers EA 1974 Estrogen-induced transformation of somatotrophs into mammotrophs in the rat. Cell Tissue Res 152:229-238 14. Boockfor FR, Hoeffler JP, Frawley LS 1986 Estradiol induces a shift in cultured cells that release prolactin or growth hormone. Am J Physiol 25O:E1O3-E1O5 15. Swanson LV, Hafs HD 1971 LH and prolactin in blood serum from estrus to ovulation in Holstein heifers. J Anim Sci 33:1038-1041 16. Koprowski JA, Tucker HA 1973 Serum prolactin during various physiological states and its relationship to milk production in the bovine. Endocrinology 92:1480-1487 17. Edgerton LA, Hafs HD 1973 Serum luteinizing hormone, prolactin, glucocorticoid and progestin in dairy cows from calving to gestation. J Dairy Sci 56:451-458 18. Beck TW, Smith VG, Seguin BE, Convey EM 1976 Bovine serum LH, GH and prolactin following chronic implantation of ovarian steroids and subsequent ovariectomy. J Anim Sci 42:461-468 19. Armstrong Jr EG, Villee CA 1977 Characterization and comparison of estrogen and androgen receptors of calf anterior pituitary. J Steroid Biochem 8:258-292 20. Padmanabhan V, Convey EM 1979 Estradiol-170 stimulates basal and thyrotropin releasing hormone induced prolactin secretion by bovine pituitary cells in primary culture. Mol Cell Endocrinol 14:103-112
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 13 November 2015. at 20:58 For personal use only. No other uses without permission. . All rights reserved.
