- Email: [email protected]
The effect of bromocriptine on the mammary glands of hyperprolactinemic rats. A histologic, ultrastructural and morphometric study
Exp. Path. 26, 131
1-1(J (H'K-f,
1) Department of Pathology, St. Michael's Hospital, University of Toronto, Ontario, Canada. 2) Department of Internal Medicine, Fniversity of Virginia, Charlottesville, VA, U.S.A.
The effect of bromocriptine on the mammary glands of hyperprolactinemic rats. A histologic, ultrastructural and morphometric study By 1. C. MURRAyl), D. J. MCCOMB l ), P. HELLMANN 2), K. KOVACS!) and M. O. THORNER2) With 7 figures (Received November 26, 1983) Address for correspondence: lAIN C. MURRAY (c/o Dr. KALMAN KOVACS), Department of Pathology St. Michael's Hospital, 30 Bond Streeet, Toronto, Ontario M5B 1W8, Canada
Key words: bromocriptine; hyperprolactinemia; mammary gland; morphometry; ultrastructure; rat
Summary Old female Long-Evans rats with elevated serum prolactin levels were used to study the effect of bromocriptine on the stimulated mammary glands by histology, electron microscopy, and morphometry. Untreated rats were divided into normoprolactinemic (below 30 ng prolactin/ml serum) and hyperprolactinemic control groups (above 30 ng/ml). Treated groups were injected subcutaneously with 1 mg/kg of bromocriptine once daily for 1 day, 30 days and 44 days. One group was treated with bromocriptine for 30 days, and then the drug was withdrawn for an additional 14 days. The light and electron microscopic features of breasts of hyperprolactinemic control, 1 day treatment and withdrawal groups resembled those of postpartum lactating breasts. A unique feature was the presence of large membrane-bound bodies, possibly lysosomes. The breasts of normoprolactinemic control, 30 and 44-day treatment groups were similar by histology and ultrastructure to involuted breasts after cessation of lactation. However, many large lysosome-like bodies and microvilli werr found in the alveolar cells which remained tightly apposed. Hyperprolactinemic control rats showed breast stimulation which could be correlated with serum prolactin levels. Serum prolactin concentrations fell within 1 day of bromocriptine treatment, whereas breast tissue responded only after a longer interval. Pituitary weights could be correlated with serum prolactin levels. Withdrawal of bromocriptine resulted in an elevation of serum prolaetin levels and breasts returned to the stimulated stage. It can be concluded that bromocriptine reversibly suppresses the stimulation of mammary glands in hyperprolactinemie rats.
Introduction In women, hyperprolactinemia is often due to a prolactin (PRL)-secreting pituitary adenoma or PRL cell hyperplasia (10). PRL excess stimulates the mammary glands and causes proliferation of the alveolar epithelium (2, 4, 9, 19, 24). Bromocriptine, a dopamine agonist, inhibits PRL secretion and is given to women to treat non-puerperal galactorrhea (16, 23). The drug is known to suppress hyperprolactinemia, restore ovulation, normalize the menstrual cycle, stop galactorrhea, abolish infertility and reduce the size of PRLsecreting pituitary adenomas (10, 16, 21, 22, 23). Bromocriptine has been shown to inhibit the growth of hyperplastie nodules in the mammary glands of mice (25). To our knowledge, the effect of bromocriptinr on ultrastructure of mammary glands has not been reported in hyperprolactinemie rats.
131
In this study, old female Long-Evans rats were used as a model of the hyperprolactinemic state. These rats have been reported to have PRL cell adenoma or hyperplasia leading to hypersecretion (11, 12, 15). We investigated the effect of bromocriptine and, consequently, the effrct of reduction in serum PRL levels on the mammary glands by light microscopy, electron mieroscopy, and morphometry.
Materials and J1ethods One hundred and twenty-seven female Long-Evans rats of approximately 20 months of age were divided into control and bromocriptine-treated groups. All groups were fed Purina laboratory rat chow ad libitum. The control animals were injected subcutaneously once daily with a 10 % ethanol vehicle and were later divided into a normoprolactinemic group with serum PRL levels under 30 ng/ml and a hyperprolactinemic group with serum PRL levels above 30 ng/ml. The treated animals were injected with bromocriptine subcutaneously at the dose of 1 mg/kg/day, and were grouped according to duration of bromocriptine therapy; there were 1 day, 30 day and 44 day treatment groups. Another group was treated with bromocriptine for 30 days and the drug was withdrawn for a further 14 days. The animals were killed by decapitation and blood was collected for measurements of serum PRL levels by radioimmunoassay. The pituitaries were removed and weighed before fixation. The histologic findings from the pituitaries of these rats will be described elsewhere (MCCOMB et al., in preparation). For light microscopy, breast tissue was removed and fixed in 10 % formalin and embedded in paraffin. Sections were stained with hematoxylin-eosin (HE). For ultrastructural studies, formalin-fixed breast tissue was transferred to Sorensen's phosphate buffer, postfixed in 2.5 % glutaraldehyde and subsequently in 1 % OS04' dehydrated in graded ethanol, processed through propylene oxide and embedded in epoxy resin. Semithin sections were stained with toluidine blue and examined by a light microscope. Ultrathin sections were stained with uranyl acetate and lead citrate and investigated with a Philips 300 electron microscope. Morphometry was performed using a Leitz A.S.M. digital analyzer. HE sections of breast were at 200 x magnification. Parenchymal tissue, including the acini and ducts within a 0.58 mm 2 grid was measured and the areas for each rat were determined from the 5-7 fields studied per animal (1). The toluidine blue-stained semithin sections were studied at a 1,000 X magnification. Cytoplasmic and nuclear areas were measured and the cytoplasmic to nuclear ratios were determined for each animal. The Wilcoxon signed rank test (20) was applied to detect significant differences between mean serum PRL levels, mean pituitary weights, and mean areas of parenchymal tissue for each group. The coefficients of correlation of the measured variables (serum PRL levels, pituitary weights, breast parenchymal areas, and acinar cell cytoplasmic to nuclear ratios) were calculated and their significance tested.
Results Light microscopic findi ngs The breast tissue of the normoprolactinemic control rats was composed of alveoli and ducts interspersed throughout adipose tissue (fig. 1 a). The histology was similar to that of unstimulated breasts as described in previous studies (4, 9, 20). The alveolar cells were flat, enclosed a small lumen and were surrounded by a basement membrane and myoepithelial cells. Some of the alveolar cells contained brown pigment granules in their cytoplasm. The hyperprolactinemic control animals (fig. 1 b) showed morphologic evidence of breast stimulation similar to that described by previous authors in postpartum lactating rats (4,9) and in rabbit mammary glands stimulated by exogenous PRL (24). Much more secretory tissue was apparent, and the alveoli appeared larger. The dilated alveolar lumina were filled with faintly staining, acidophilic material; the alveolar cells were columnar and contained many fat droplets and brown pigment granules. While the 1-day treatment group exhibited breast stimulation similar to that of the hyperprolactinemic control rats, the 30 (fig. 1 c) and 44-day treatment groups closely resembled the normoprolactinemic control group. The acini were smaller and less numerous. The alveolar <:ells were flat and contained a few lipid droplets and several brown pigment granules.
132
Fig. 1. Breast tissue of old female rats. Epoxy resin embedded sections stained with toluidine blue. x 400. a) Normoprolactinemie control rat. Alveolar cells show some lipid droplets. b) Hyperprolactinemic control rat. Proliferation of alveolar cells is evident. The cytoplasm of alveolar cells is well developed. The alveolar lumina are distended and contain secretory material. c) Rat treated with bromocriptine for 30 days. The breast resembles that of normoprolactinemic control rats; it has a few lipid droplets and the alveolar lumina contain less secretory material. d) Rat withdrawn from bromocriptine. Proliferation of alveoli is seen.
133
Fig. 2. Breast of hyperprola,etinemic eontrol rat, showing a large lipid droplet, electron dense granules, many mitochondria, a,bundant RER, and part of a megalysosome. The alveolar lumen is lined by microvilli. x ii,SOO. Fig. 3. Breast of a hyperprolactinemie eontrol rat, showing many lipid droplets, mitochondria, and an abundance of RER. x 10,200.
134
Fig. 4. Breast of rat treated with bromocriptine for 44 days. RER is inconspicuous; many lysosomes are visible. The alveolar lumen shows many mi(·rovilli. A tightly apposed alveolar cell border is visible. x 11,910. Fig. 5. Breast of rat treated with bromoeriptine for 44 days. Megalysosomes are shown in the cytoplasm of an alveolar cell lacking abundant RER and lipid droplets. x 7,950.
135
Fig. 6. Breast of rat withdrawn from bromocriptine. Many mitochondria, abundant RER, numerou lipid droplets and electron dense granules are noted in the cytoplasm of alveolar cells. The lumina surface shows many microvilli. The alveolar lumina are filled with many lipid droplets and protein granules. x 5,670. Fig. 7. Breast of rat withdrawn from bromocriptine. Mitochondria and RER are prominent in the cytoplasm of alveolar cells. The basement membrane and capillary are easily noticeable. x 5,670.
136
The withdrawal group (fig. 1 d) showed stimulation similar to that of the hyperprolactinemic group with hyperplasia flf alveolar eells, abundant secretory material in the dilated lumina, and numerous fat droplets and brown pigment granules within the alveolar cell cytoplasm. Ducts showed no morphologic differenees. Electron microseopic findings Compared with the breasts of normoprolactinemic control rats, the alveolar cells of hyperprolactinemie eontrol rats (figs. 2 and 3) contained more cytoplasmic lipid droplets and electron dense granules (4, 5, 13, 18, 24). The luminal surface of each alveolar cell was lined by microvilli. The alveolar cells were tightly apposed to one another and were connected by junctional complexes. The alveolar cell nuelei were large and round or were indented by large lipid droplets. The rough endoplasmic reticulum (RER) and Golgi apparatus were well developed and prominent. Many alveolar cells contained large irregular lysosome-like bodies, each surrounded by a limiting membrane. Some alveolar cells had large cytoplasmic vacuoles which contained many electron dense granules clumped together. The alveolar lumina were generally distended and contained lipid droplets, in some cases surrounded by an osmiophilic membrane and a small amount of cytoplasm, and other electron dense granules described by others (4, 9). With the exception of the large lysosome-like bodies, the ultrastructural findings were in accordance with those of postpartum lactating breasts (4, 5, 9, 18, 24). The I-day treatment group exhibited fine structural features similar to those of the hyperprolactinemie eontrol group. In the alveolar cells of the breasts of rats treated with bromoeriptine for 30 and 44 days (figs. 4 and 5), the RER, and Golgi apparatus appeared to be less eonspieuous. The lipid droplets and eleetron dense granules were reduced in size and number in both the alveolar eytoplasm and alveolar lumina. The number of small and large lysosomes seemed to have inereased. The alveolar eells remained tightly apposed, and the mierovilli showed no atrophy. Apart from the nomal microvilli, the abundanee of large lysosome-like bodies and close approximation, the alveolar eells were similar in ultrastrueture to involuting alveolar eells following lactation as described in other studies (4, 6, 7, 8, 13, 17). The withdrawal group (figs. 6 and 7) showed ultrastructural features similar to those of postpartum laetating breasts, except that they possessed several large lysosomelike bodies. Duets were not studied at the eleetron microscopic level. Morphometrie findings Table 1 shows the mean serum PRL levels, pituitary weights, and breast parenchymal tissue areas for each group. It can be seen that serum PRL concentrations were lower in the bromoeriptine-treated groups than in the untreated controls, and that the withdrawal group had a higher mean serum PRL level. The pituitaries weighed more in the hyperprolaetinemic control and withdrawal groups than in the other groups. The hyperprolactinemie Table 1. Means (± standard errors of means) or serum PRL levels, pituitary weights and breast parenchymal tissue areas Group
Normoprolactinemic control H yperprolactinemic control i-day treatment 30-day treatment 44-day treatment Withdrawal
# Rats Serum PRL Pituitary weight (ng/ml)
11
39 14 28 12 .23
17 203 10 21 37 133
± 3 ± 54 ± 3+ ± 5+ ± 8*+ ± 20*
(mg)
20 50 22 30 23 38
± ± ± ± ± ±
2 9 3+ 5 2 6*
Breast Parenchymal tissue area (mm2) 0.037 0.061 0.045 0.024 0.028 0.070
± 0.007 ± 0.008* ± 0.007 ± 0.003+ ± 0.008+ ± 0.013
* Significant difference from normoprolactinemic controls (p < 0.01) + Significant difference from hyperprolactinemic controls (p < 0.01) 10 Exp. Path. 26, H. 3
137
control, l-(la~' trpatllIPllt alld withdrawal g-roup~ all had more parenchymal tissue areas than the nOTmopTolaetinemie control, 30 and 44-day treatment groups. The mean serum PRL levels significantly differed between the i-day treatment group, the 30-day treatment and withdrawal groups, bptwepll the 30-day treatment group and the 44-day treatment and withdrawal groups, and brtween the 44-day treatment and withdrawal groups. The I-day treatment group showed a significant difference in parenehymal tissue area from the 30-day treatment group, and both the 30 and 44-day treatment groups from the withdrawal group. There was a signifieant eorrelation between serum PRL levels and breast parenchymal tissue areas for individual rats, and in the hyperprolactinemic control and 30-day treatment groups (p < 0.02). Serum PRL levpls eorrelated significantly with pituitary weights in the hyperprolactinemic ('ontrol, 30-day treatment and withdrawal groups (p < 0.(1). Significant correlation was also found between pituitary weight with parenchymal tissue areas, in the hyperprolactinemic control, 30-day treatment and withdrawal groups (p < 0.01) and the I-day treatment group (p < 0.(5). No statistical significance was obtained when the alveolar cytoplasmic to nuclear ratios were correlated with serum PRL levels or pituitary weight.
Discussion The mammary glands of hyperprolactinemic control rats show ultrastructural features similar to those of postpartum lactating rats (4, 5, 9) with the exception of the presence of large electron lucent inclusions in the cytoplasm of alveolar cells. These structures may represent lysosomes and because of their size they can be termed megalysosomes. They may playa role in the removal of lipid droplets and undischarged secretory material from the cytoplasm; however, further work is required to determine their origin and functional significance. It is also evident that the mean pituitary weights and mean breast parenchymal tissue areas are all higher in the hyperprolactinemic control group than in the normoprolactinemic control group. In the hyperprolactinemic control group, breast parenchymal tissue areas can be correlated with serum PRL levels (2, 4, 9, 19), the pituitary weights with breast parenchymal tissue areas, and the serum PRL levels with pituitary weights. The latter findings in the rat support reports that serum PRL levels correlate with pituitary adenoma size in human patients (14). The ultrastructural features of the secretory tissue in the breasts of rats treated with bromocriptine for 1 day were not different from those of the hyperprolactinemic control group. Since serum PRL levels are decreased within 1 day, it is apparent that serum PRL levels respond to bromocriptine faster than breast morphology. It is assumed that bromocripiine has an indirect effect on breast morphology by reducing serum PRL levels. However, further studies are necessary to exclude the possibility of a direct effect of bromocriptine on the breast. The 30 and 44-day treatment groups show a significant reduction in serum PRL levels and breast parenchymal tissue areas, indicating that the breast responds to bromocriptine within 30 days. In these rats, breast ultrastructure is similar to that of involuting breasts following cessation of lactation (4, 6, 7, 8, 9, 13, 17). HOLLMAN (9) noted that when the breast undergoes involution, the alveolar cells tend to spread apart and are held together only by junctional eomplexes. Such a distention was not apparent in the bromocriptine-treated groups. Normal involution involves an increase in luminal pressure, since the sucklings are no longer withdrawing milk from the lumina (4). Bromocriptine merely brings about an arrest in production of milk; hence, there is no sudden increase in luminal pressure and the cells remain tightly apposed. This could explain the fact that the microvilli do not disappear, as noticed b;' CI:'o1TI et al. (3). Tn the withdrawal g-roup. breast ultrastructure, serum PRL levels, and breast parenchymal tissue areas are similar to thosp of the hyperprolactinemic controls indicating that suppression by bromoeriptine is reyersiblr, and if treatment is withdrawn the above para-
138
meters return to a pretreatnHmt "tatl'. It is noteworthy that the withdrawal group shows correlation between serum PRL levels and pituitary weights, pituitary weights and breast parenchymal tissue areas, but not serum PRL levels and breast parenchymal tissue areas. These results can best be explained by the fact that serum PRL levels tend to fluctuate substantially during the day, whereas the weight of the pituitary is more constant. Thus, in this case, it seems that pituitary weight may be a more reliable estimate of the stimulus for breast growth than a singh' determination of serum PRL concentration. This experiment shows that hyperprolactinemic control rats possess stimulated breasts similar to those of postpartum lactating breasts, with the exception of the presence of megalysosomes. Treatment with bromoeriptine brings about an involution of the breasts differing from normal involution only in the presence of microvilli and tightly apposed cells. Withdrawal of bromocriptine treatment tends to return the breasts and serum PRL levels to their original hyperprolactinemic state, indicating reversibility of the effects of the drug
Acknowledgements This work was supported in part by St. Michael's Hospital Research Society and Sandoz Limited Basle, Switzerland. The advice of Dr. D. MURRAY, the technical assistance of Mrs. S. BRIGGS, Miss S. COHEN, Mrs. G. ILSE, Mrs. N. LOZINSKI and Mrs. A. PITTAWAY, and the secretarial work of Mrs. W. WLODARSKI are gratefully acknowledged.
Literature 1. HENSON, G. K., and S. J. FOLLEY, The effeet of oxytocin on mammary gland involution in the rat. J. Endocrino!. 16,189-201 (1957). 2. CHATTERTON, R. T., J. A. HARRIS, W. J. KING and R. M. WYNN, Ultrastructural alterations 3. 4.
5. 6. 7. 8.
in mammary glands of pregnant rats after ovariectomy and hysterectomy: Effect of adrenal steroids and prolactin. Am. J. Obstet. Gyneco!. 133, 694-702 (1979). CINTI, S., G. BALERCIA, R. ROSSI, and A. SBARBATI, Modificazioni ultrastrutturali della ghiandola mammalia di ratto in allattamento indotte dal trattamento con bromocriptina. Boll. Soc. Ital. BioI. Sper. 08, 1406-1412 (1982). COWIE, A. T., and J. S. TINDAL, The physiology of lactation. Arnold, London 1971, pp. 1-392. HELMINEN, H. J., and J. 1. E. ERICSSON, Studies on mammary gland involution. 1. On the ultrastructure of the lactating mammary gland. J. Ultrastruct. Res. 20, 193-213 (1968). - - Studies on mammary gland involution. II. Ultrastructural evidence for auto- and heterophagocytosis. J. Ultrastruct. Res. 20, 214-227 (1968). - - Studies on mammary gland involution. III. Alterations outside auto- and heterophagoeytosis pathways for cytoplasmic degradation. J. Ultrastruct. Res. 20, 228-239 (1968). - - and S. ORRENIUS, Sudies on mammary gland involution. IV. Histochemical and biochemical observations on alterations in lysosomes and lysosomal enzymes. J. Ultrastruct. Res.
20, 240-252 (1968).
9. HOLLMANN, K. H., Cytology and fine structure of the mammarp gland. In: Lactation: A Comprehensive Treatise: Mammary Gland/Development and Maintenance, eds. 1. B. 1. LARSON and V. R. SMITH. Academic Press, New York 1974, pp. 3-95. 10. KINCH, R. A., The use of bromocriptine in obstetrics and gynecology. Ferti!. Steril. 33, 463-470
(1980).
11. KOVACS, K., E. HORVATH, F. G. lLSE, C. EZRIN, and D. lLSE, Spontaneous pituitary adenomas in aging rats. A light microscopic, immunocytological and fine structural study. Beitr. Pathol. 161, 1-16 (1977). 12. - G. lLSE, N. RYAN, D. J. MCCOMB, E. HORVATH, H. J. CHEN, and P. G. WALFISH, Pituitary prolactin eell hyperplasia. Hormone Res. 312, 8'7-95 (1980). 13. LASCELLES, A. K., and C. S. LEE, Involution of the mammary gland. In: Lactation: A Comprehensive Treatise: The Mammary Gland/Human Lactation/Milk Synthesis IV., ed. B. C. LARSON. Academic Press, New York 1978, pp. 115-177. 14. MCCOMB, D ..J., K. KOYACS, E. HORVATH, W. SINGER, D. W. KILLINGER, H. S. SMYTH, C. EZRIN, and M. H. WEISS, Correlative ultrastructural morphometry of human prolactin-producing adenomas. Acta Neurochir. 03, 217--225 (1980). 15. - N. RYAN, E. HOIW.\TH, K. KOYACS, E. NAGY, I. BERCZI, 1. DOMOKOZ, and F. A. LASZLO, Five different adenomas derived from the rat adenohypophysis: immunocytochemical and ultrastructural study ..J. Nat!. Caneer Inst. 66, 1103-1111 (1981). 16. NILLIUS, S. J., T. BERGH, P. D. LllNIJBERG, J. STAHLE, and L. WIDE, Regression of a prolactinsecreting pituitary tumor during long-term treatment with bromocriptine. Ferti!. Steri!. 30,
710-712 (1978).
10*
139
17. RICH.\IW" R. C.. and G. K. BE",OX, l'ltrastrudural changes accompanying involution of the mammary gland in the albino rat . .T. Endocrinol.lil, 127-135 (1971). 18..- - Structural changes associated with inhibition of involution of the mammary gland in the albino rat. J. Endoerinol. 51, 137-148 (1971). 19. Suru, R. P. C., and H. G. FIUESEN, :Vlechanism of action of prolactin in the control of mammary gland function. Ann. Rev. Physiol. 42, 83-9G (1980). 20. STEEL, R. G. D., and J. H. TORBIE, Prineiples and Procedures of Statistics. MacGraw-Hill Book Co., Inc., New York 19GO, pp. 492. 21. TINDALL, G. T., K. KOVACS, E. HORVATH, and M. O. THORNER, Human prolactin-producing adenomas and bromocriptine. A histological, immunocytological, ultrastructural and morphometric study. J. Clin. Endocrinol. Metab. lili, 1178-1183 (1982). 22. THORNER, ~L 0., W. H. MARTIN, A. D. ROGOL, J. L. MORRIS, R. O. PERRYMAN, B. P. CONWAY, S. S. HOWARDS, M. G. WOLFMAN, and R M. MACLEOD, Rapid regression of pituitary prolactinomas during bromocriptine treatment. J. Clin. Endocrinol. Metab. iiI, 438-445 (1980). 23. - A. S. McNEILLY, C. HAGEN, and G. M. BESSER, Long-term treatment of galactorrhea and hypogonadism with bromocriptine. Br. Med. J. II, 419-422 (1974). 24. TOBIN, H., J. B. J OSIMOVICH, and H. SALAZAR, The ultrastructure of the mammary gland during prolactin-induced lactogenesis in the rabbit. Endocrinology 90, 15G9-1577 (1972). 25. YANAI, R, and H. NAGASAWA, Suppression of mammary hyperplastic nodule formation and pituitary prolactin secretion in mice induced by ergocornine or 2-bromo-iX-ergocriptine. J. Nat. Cancer Inst. 41i, 1105-1112 (1970).
140
1-1(J (H'K-f,
1) Department of Pathology, St. Michael's Hospital, University of Toronto, Ontario, Canada. 2) Department of Internal Medicine, Fniversity of Virginia, Charlottesville, VA, U.S.A.
The effect of bromocriptine on the mammary glands of hyperprolactinemic rats. A histologic, ultrastructural and morphometric study By 1. C. MURRAyl), D. J. MCCOMB l ), P. HELLMANN 2), K. KOVACS!) and M. O. THORNER2) With 7 figures (Received November 26, 1983) Address for correspondence: lAIN C. MURRAY (c/o Dr. KALMAN KOVACS), Department of Pathology St. Michael's Hospital, 30 Bond Streeet, Toronto, Ontario M5B 1W8, Canada
Key words: bromocriptine; hyperprolactinemia; mammary gland; morphometry; ultrastructure; rat
Summary Old female Long-Evans rats with elevated serum prolactin levels were used to study the effect of bromocriptine on the stimulated mammary glands by histology, electron microscopy, and morphometry. Untreated rats were divided into normoprolactinemic (below 30 ng prolactin/ml serum) and hyperprolactinemic control groups (above 30 ng/ml). Treated groups were injected subcutaneously with 1 mg/kg of bromocriptine once daily for 1 day, 30 days and 44 days. One group was treated with bromocriptine for 30 days, and then the drug was withdrawn for an additional 14 days. The light and electron microscopic features of breasts of hyperprolactinemic control, 1 day treatment and withdrawal groups resembled those of postpartum lactating breasts. A unique feature was the presence of large membrane-bound bodies, possibly lysosomes. The breasts of normoprolactinemic control, 30 and 44-day treatment groups were similar by histology and ultrastructure to involuted breasts after cessation of lactation. However, many large lysosome-like bodies and microvilli werr found in the alveolar cells which remained tightly apposed. Hyperprolactinemic control rats showed breast stimulation which could be correlated with serum prolactin levels. Serum prolactin concentrations fell within 1 day of bromocriptine treatment, whereas breast tissue responded only after a longer interval. Pituitary weights could be correlated with serum prolactin levels. Withdrawal of bromocriptine resulted in an elevation of serum prolaetin levels and breasts returned to the stimulated stage. It can be concluded that bromocriptine reversibly suppresses the stimulation of mammary glands in hyperprolactinemie rats.
Introduction In women, hyperprolactinemia is often due to a prolactin (PRL)-secreting pituitary adenoma or PRL cell hyperplasia (10). PRL excess stimulates the mammary glands and causes proliferation of the alveolar epithelium (2, 4, 9, 19, 24). Bromocriptine, a dopamine agonist, inhibits PRL secretion and is given to women to treat non-puerperal galactorrhea (16, 23). The drug is known to suppress hyperprolactinemia, restore ovulation, normalize the menstrual cycle, stop galactorrhea, abolish infertility and reduce the size of PRLsecreting pituitary adenomas (10, 16, 21, 22, 23). Bromocriptine has been shown to inhibit the growth of hyperplastie nodules in the mammary glands of mice (25). To our knowledge, the effect of bromocriptinr on ultrastructure of mammary glands has not been reported in hyperprolactinemie rats.
131
In this study, old female Long-Evans rats were used as a model of the hyperprolactinemic state. These rats have been reported to have PRL cell adenoma or hyperplasia leading to hypersecretion (11, 12, 15). We investigated the effect of bromocriptine and, consequently, the effrct of reduction in serum PRL levels on the mammary glands by light microscopy, electron mieroscopy, and morphometry.
Materials and J1ethods One hundred and twenty-seven female Long-Evans rats of approximately 20 months of age were divided into control and bromocriptine-treated groups. All groups were fed Purina laboratory rat chow ad libitum. The control animals were injected subcutaneously once daily with a 10 % ethanol vehicle and were later divided into a normoprolactinemic group with serum PRL levels under 30 ng/ml and a hyperprolactinemic group with serum PRL levels above 30 ng/ml. The treated animals were injected with bromocriptine subcutaneously at the dose of 1 mg/kg/day, and were grouped according to duration of bromocriptine therapy; there were 1 day, 30 day and 44 day treatment groups. Another group was treated with bromocriptine for 30 days and the drug was withdrawn for a further 14 days. The animals were killed by decapitation and blood was collected for measurements of serum PRL levels by radioimmunoassay. The pituitaries were removed and weighed before fixation. The histologic findings from the pituitaries of these rats will be described elsewhere (MCCOMB et al., in preparation). For light microscopy, breast tissue was removed and fixed in 10 % formalin and embedded in paraffin. Sections were stained with hematoxylin-eosin (HE). For ultrastructural studies, formalin-fixed breast tissue was transferred to Sorensen's phosphate buffer, postfixed in 2.5 % glutaraldehyde and subsequently in 1 % OS04' dehydrated in graded ethanol, processed through propylene oxide and embedded in epoxy resin. Semithin sections were stained with toluidine blue and examined by a light microscope. Ultrathin sections were stained with uranyl acetate and lead citrate and investigated with a Philips 300 electron microscope. Morphometry was performed using a Leitz A.S.M. digital analyzer. HE sections of breast were at 200 x magnification. Parenchymal tissue, including the acini and ducts within a 0.58 mm 2 grid was measured and the areas for each rat were determined from the 5-7 fields studied per animal (1). The toluidine blue-stained semithin sections were studied at a 1,000 X magnification. Cytoplasmic and nuclear areas were measured and the cytoplasmic to nuclear ratios were determined for each animal. The Wilcoxon signed rank test (20) was applied to detect significant differences between mean serum PRL levels, mean pituitary weights, and mean areas of parenchymal tissue for each group. The coefficients of correlation of the measured variables (serum PRL levels, pituitary weights, breast parenchymal areas, and acinar cell cytoplasmic to nuclear ratios) were calculated and their significance tested.
Results Light microscopic findi ngs The breast tissue of the normoprolactinemic control rats was composed of alveoli and ducts interspersed throughout adipose tissue (fig. 1 a). The histology was similar to that of unstimulated breasts as described in previous studies (4, 9, 20). The alveolar cells were flat, enclosed a small lumen and were surrounded by a basement membrane and myoepithelial cells. Some of the alveolar cells contained brown pigment granules in their cytoplasm. The hyperprolactinemic control animals (fig. 1 b) showed morphologic evidence of breast stimulation similar to that described by previous authors in postpartum lactating rats (4,9) and in rabbit mammary glands stimulated by exogenous PRL (24). Much more secretory tissue was apparent, and the alveoli appeared larger. The dilated alveolar lumina were filled with faintly staining, acidophilic material; the alveolar cells were columnar and contained many fat droplets and brown pigment granules. While the 1-day treatment group exhibited breast stimulation similar to that of the hyperprolactinemic control rats, the 30 (fig. 1 c) and 44-day treatment groups closely resembled the normoprolactinemic control group. The acini were smaller and less numerous. The alveolar <:ells were flat and contained a few lipid droplets and several brown pigment granules.
132
Fig. 1. Breast tissue of old female rats. Epoxy resin embedded sections stained with toluidine blue. x 400. a) Normoprolactinemie control rat. Alveolar cells show some lipid droplets. b) Hyperprolactinemic control rat. Proliferation of alveolar cells is evident. The cytoplasm of alveolar cells is well developed. The alveolar lumina are distended and contain secretory material. c) Rat treated with bromocriptine for 30 days. The breast resembles that of normoprolactinemic control rats; it has a few lipid droplets and the alveolar lumina contain less secretory material. d) Rat withdrawn from bromocriptine. Proliferation of alveoli is seen.
133
Fig. 2. Breast of hyperprola,etinemic eontrol rat, showing a large lipid droplet, electron dense granules, many mitochondria, a,bundant RER, and part of a megalysosome. The alveolar lumen is lined by microvilli. x ii,SOO. Fig. 3. Breast of a hyperprolactinemie eontrol rat, showing many lipid droplets, mitochondria, and an abundance of RER. x 10,200.
134
Fig. 4. Breast of rat treated with bromocriptine for 44 days. RER is inconspicuous; many lysosomes are visible. The alveolar lumen shows many mi(·rovilli. A tightly apposed alveolar cell border is visible. x 11,910. Fig. 5. Breast of rat treated with bromoeriptine for 44 days. Megalysosomes are shown in the cytoplasm of an alveolar cell lacking abundant RER and lipid droplets. x 7,950.
135
Fig. 6. Breast of rat withdrawn from bromocriptine. Many mitochondria, abundant RER, numerou lipid droplets and electron dense granules are noted in the cytoplasm of alveolar cells. The lumina surface shows many microvilli. The alveolar lumina are filled with many lipid droplets and protein granules. x 5,670. Fig. 7. Breast of rat withdrawn from bromocriptine. Mitochondria and RER are prominent in the cytoplasm of alveolar cells. The basement membrane and capillary are easily noticeable. x 5,670.
136
The withdrawal group (fig. 1 d) showed stimulation similar to that of the hyperprolactinemic group with hyperplasia flf alveolar eells, abundant secretory material in the dilated lumina, and numerous fat droplets and brown pigment granules within the alveolar cell cytoplasm. Ducts showed no morphologic differenees. Electron microseopic findings Compared with the breasts of normoprolactinemic control rats, the alveolar cells of hyperprolactinemie eontrol rats (figs. 2 and 3) contained more cytoplasmic lipid droplets and electron dense granules (4, 5, 13, 18, 24). The luminal surface of each alveolar cell was lined by microvilli. The alveolar cells were tightly apposed to one another and were connected by junctional complexes. The alveolar cell nuelei were large and round or were indented by large lipid droplets. The rough endoplasmic reticulum (RER) and Golgi apparatus were well developed and prominent. Many alveolar cells contained large irregular lysosome-like bodies, each surrounded by a limiting membrane. Some alveolar cells had large cytoplasmic vacuoles which contained many electron dense granules clumped together. The alveolar lumina were generally distended and contained lipid droplets, in some cases surrounded by an osmiophilic membrane and a small amount of cytoplasm, and other electron dense granules described by others (4, 9). With the exception of the large lysosome-like bodies, the ultrastructural findings were in accordance with those of postpartum lactating breasts (4, 5, 9, 18, 24). The I-day treatment group exhibited fine structural features similar to those of the hyperprolactinemie eontrol group. In the alveolar cells of the breasts of rats treated with bromoeriptine for 30 and 44 days (figs. 4 and 5), the RER, and Golgi apparatus appeared to be less eonspieuous. The lipid droplets and eleetron dense granules were reduced in size and number in both the alveolar eytoplasm and alveolar lumina. The number of small and large lysosomes seemed to have inereased. The alveolar eells remained tightly apposed, and the mierovilli showed no atrophy. Apart from the nomal microvilli, the abundanee of large lysosome-like bodies and close approximation, the alveolar eells were similar in ultrastrueture to involuting alveolar eells following lactation as described in other studies (4, 6, 7, 8, 13, 17). The withdrawal group (figs. 6 and 7) showed ultrastructural features similar to those of postpartum laetating breasts, except that they possessed several large lysosomelike bodies. Duets were not studied at the eleetron microscopic level. Morphometrie findings Table 1 shows the mean serum PRL levels, pituitary weights, and breast parenchymal tissue areas for each group. It can be seen that serum PRL concentrations were lower in the bromoeriptine-treated groups than in the untreated controls, and that the withdrawal group had a higher mean serum PRL level. The pituitaries weighed more in the hyperprolaetinemic control and withdrawal groups than in the other groups. The hyperprolactinemie Table 1. Means (± standard errors of means) or serum PRL levels, pituitary weights and breast parenchymal tissue areas Group
Normoprolactinemic control H yperprolactinemic control i-day treatment 30-day treatment 44-day treatment Withdrawal
# Rats Serum PRL Pituitary weight (ng/ml)
11
39 14 28 12 .23
17 203 10 21 37 133
± 3 ± 54 ± 3+ ± 5+ ± 8*+ ± 20*
(mg)
20 50 22 30 23 38
± ± ± ± ± ±
2 9 3+ 5 2 6*
Breast Parenchymal tissue area (mm2) 0.037 0.061 0.045 0.024 0.028 0.070
± 0.007 ± 0.008* ± 0.007 ± 0.003+ ± 0.008+ ± 0.013
* Significant difference from normoprolactinemic controls (p < 0.01) + Significant difference from hyperprolactinemic controls (p < 0.01) 10 Exp. Path. 26, H. 3
137
control, l-(la~' trpatllIPllt alld withdrawal g-roup~ all had more parenchymal tissue areas than the nOTmopTolaetinemie control, 30 and 44-day treatment groups. The mean serum PRL levels significantly differed between the i-day treatment group, the 30-day treatment and withdrawal groups, bptwepll the 30-day treatment group and the 44-day treatment and withdrawal groups, and brtween the 44-day treatment and withdrawal groups. The I-day treatment group showed a significant difference in parenehymal tissue area from the 30-day treatment group, and both the 30 and 44-day treatment groups from the withdrawal group. There was a signifieant eorrelation between serum PRL levels and breast parenchymal tissue areas for individual rats, and in the hyperprolactinemic control and 30-day treatment groups (p < 0.02). Serum PRL levpls eorrelated significantly with pituitary weights in the hyperprolactinemic ('ontrol, 30-day treatment and withdrawal groups (p < 0.(1). Significant correlation was also found between pituitary weight with parenchymal tissue areas, in the hyperprolactinemic control, 30-day treatment and withdrawal groups (p < 0.01) and the I-day treatment group (p < 0.(5). No statistical significance was obtained when the alveolar cytoplasmic to nuclear ratios were correlated with serum PRL levels or pituitary weight.
Discussion The mammary glands of hyperprolactinemic control rats show ultrastructural features similar to those of postpartum lactating rats (4, 5, 9) with the exception of the presence of large electron lucent inclusions in the cytoplasm of alveolar cells. These structures may represent lysosomes and because of their size they can be termed megalysosomes. They may playa role in the removal of lipid droplets and undischarged secretory material from the cytoplasm; however, further work is required to determine their origin and functional significance. It is also evident that the mean pituitary weights and mean breast parenchymal tissue areas are all higher in the hyperprolactinemic control group than in the normoprolactinemic control group. In the hyperprolactinemic control group, breast parenchymal tissue areas can be correlated with serum PRL levels (2, 4, 9, 19), the pituitary weights with breast parenchymal tissue areas, and the serum PRL levels with pituitary weights. The latter findings in the rat support reports that serum PRL levels correlate with pituitary adenoma size in human patients (14). The ultrastructural features of the secretory tissue in the breasts of rats treated with bromocriptine for 1 day were not different from those of the hyperprolactinemic control group. Since serum PRL levels are decreased within 1 day, it is apparent that serum PRL levels respond to bromocriptine faster than breast morphology. It is assumed that bromocripiine has an indirect effect on breast morphology by reducing serum PRL levels. However, further studies are necessary to exclude the possibility of a direct effect of bromocriptine on the breast. The 30 and 44-day treatment groups show a significant reduction in serum PRL levels and breast parenchymal tissue areas, indicating that the breast responds to bromocriptine within 30 days. In these rats, breast ultrastructure is similar to that of involuting breasts following cessation of lactation (4, 6, 7, 8, 9, 13, 17). HOLLMAN (9) noted that when the breast undergoes involution, the alveolar cells tend to spread apart and are held together only by junctional eomplexes. Such a distention was not apparent in the bromocriptine-treated groups. Normal involution involves an increase in luminal pressure, since the sucklings are no longer withdrawing milk from the lumina (4). Bromocriptine merely brings about an arrest in production of milk; hence, there is no sudden increase in luminal pressure and the cells remain tightly apposed. This could explain the fact that the microvilli do not disappear, as noticed b;' CI:'o1TI et al. (3). Tn the withdrawal g-roup. breast ultrastructure, serum PRL levels, and breast parenchymal tissue areas are similar to thosp of the hyperprolactinemic controls indicating that suppression by bromoeriptine is reyersiblr, and if treatment is withdrawn the above para-
138
meters return to a pretreatnHmt "tatl'. It is noteworthy that the withdrawal group shows correlation between serum PRL levels and pituitary weights, pituitary weights and breast parenchymal tissue areas, but not serum PRL levels and breast parenchymal tissue areas. These results can best be explained by the fact that serum PRL levels tend to fluctuate substantially during the day, whereas the weight of the pituitary is more constant. Thus, in this case, it seems that pituitary weight may be a more reliable estimate of the stimulus for breast growth than a singh' determination of serum PRL concentration. This experiment shows that hyperprolactinemic control rats possess stimulated breasts similar to those of postpartum lactating breasts, with the exception of the presence of megalysosomes. Treatment with bromoeriptine brings about an involution of the breasts differing from normal involution only in the presence of microvilli and tightly apposed cells. Withdrawal of bromocriptine treatment tends to return the breasts and serum PRL levels to their original hyperprolactinemic state, indicating reversibility of the effects of the drug
Acknowledgements This work was supported in part by St. Michael's Hospital Research Society and Sandoz Limited Basle, Switzerland. The advice of Dr. D. MURRAY, the technical assistance of Mrs. S. BRIGGS, Miss S. COHEN, Mrs. G. ILSE, Mrs. N. LOZINSKI and Mrs. A. PITTAWAY, and the secretarial work of Mrs. W. WLODARSKI are gratefully acknowledged.
Literature 1. HENSON, G. K., and S. J. FOLLEY, The effeet of oxytocin on mammary gland involution in the rat. J. Endocrino!. 16,189-201 (1957). 2. CHATTERTON, R. T., J. A. HARRIS, W. J. KING and R. M. WYNN, Ultrastructural alterations 3. 4.
5. 6. 7. 8.
in mammary glands of pregnant rats after ovariectomy and hysterectomy: Effect of adrenal steroids and prolactin. Am. J. Obstet. Gyneco!. 133, 694-702 (1979). CINTI, S., G. BALERCIA, R. ROSSI, and A. SBARBATI, Modificazioni ultrastrutturali della ghiandola mammalia di ratto in allattamento indotte dal trattamento con bromocriptina. Boll. Soc. Ital. BioI. Sper. 08, 1406-1412 (1982). COWIE, A. T., and J. S. TINDAL, The physiology of lactation. Arnold, London 1971, pp. 1-392. HELMINEN, H. J., and J. 1. E. ERICSSON, Studies on mammary gland involution. 1. On the ultrastructure of the lactating mammary gland. J. Ultrastruct. Res. 20, 193-213 (1968). - - Studies on mammary gland involution. II. Ultrastructural evidence for auto- and heterophagocytosis. J. Ultrastruct. Res. 20, 214-227 (1968). - - Studies on mammary gland involution. III. Alterations outside auto- and heterophagoeytosis pathways for cytoplasmic degradation. J. Ultrastruct. Res. 20, 228-239 (1968). - - and S. ORRENIUS, Sudies on mammary gland involution. IV. Histochemical and biochemical observations on alterations in lysosomes and lysosomal enzymes. J. Ultrastruct. Res.
20, 240-252 (1968).
9. HOLLMANN, K. H., Cytology and fine structure of the mammarp gland. In: Lactation: A Comprehensive Treatise: Mammary Gland/Development and Maintenance, eds. 1. B. 1. LARSON and V. R. SMITH. Academic Press, New York 1974, pp. 3-95. 10. KINCH, R. A., The use of bromocriptine in obstetrics and gynecology. Ferti!. Steril. 33, 463-470
(1980).
11. KOVACS, K., E. HORVATH, F. G. lLSE, C. EZRIN, and D. lLSE, Spontaneous pituitary adenomas in aging rats. A light microscopic, immunocytological and fine structural study. Beitr. Pathol. 161, 1-16 (1977). 12. - G. lLSE, N. RYAN, D. J. MCCOMB, E. HORVATH, H. J. CHEN, and P. G. WALFISH, Pituitary prolactin eell hyperplasia. Hormone Res. 312, 8'7-95 (1980). 13. LASCELLES, A. K., and C. S. LEE, Involution of the mammary gland. In: Lactation: A Comprehensive Treatise: The Mammary Gland/Human Lactation/Milk Synthesis IV., ed. B. C. LARSON. Academic Press, New York 1978, pp. 115-177. 14. MCCOMB, D ..J., K. KOYACS, E. HORVATH, W. SINGER, D. W. KILLINGER, H. S. SMYTH, C. EZRIN, and M. H. WEISS, Correlative ultrastructural morphometry of human prolactin-producing adenomas. Acta Neurochir. 03, 217--225 (1980). 15. - N. RYAN, E. HOIW.\TH, K. KOYACS, E. NAGY, I. BERCZI, 1. DOMOKOZ, and F. A. LASZLO, Five different adenomas derived from the rat adenohypophysis: immunocytochemical and ultrastructural study ..J. Nat!. Caneer Inst. 66, 1103-1111 (1981). 16. NILLIUS, S. J., T. BERGH, P. D. LllNIJBERG, J. STAHLE, and L. WIDE, Regression of a prolactinsecreting pituitary tumor during long-term treatment with bromocriptine. Ferti!. Steri!. 30,
710-712 (1978).
10*
139
17. RICH.\IW" R. C.. and G. K. BE",OX, l'ltrastrudural changes accompanying involution of the mammary gland in the albino rat . .T. Endocrinol.lil, 127-135 (1971). 18..- - Structural changes associated with inhibition of involution of the mammary gland in the albino rat. J. Endoerinol. 51, 137-148 (1971). 19. Suru, R. P. C., and H. G. FIUESEN, :Vlechanism of action of prolactin in the control of mammary gland function. Ann. Rev. Physiol. 42, 83-9G (1980). 20. STEEL, R. G. D., and J. H. TORBIE, Prineiples and Procedures of Statistics. MacGraw-Hill Book Co., Inc., New York 19GO, pp. 492. 21. TINDALL, G. T., K. KOVACS, E. HORVATH, and M. O. THORNER, Human prolactin-producing adenomas and bromocriptine. A histological, immunocytological, ultrastructural and morphometric study. J. Clin. Endocrinol. Metab. lili, 1178-1183 (1982). 22. THORNER, ~L 0., W. H. MARTIN, A. D. ROGOL, J. L. MORRIS, R. O. PERRYMAN, B. P. CONWAY, S. S. HOWARDS, M. G. WOLFMAN, and R M. MACLEOD, Rapid regression of pituitary prolactinomas during bromocriptine treatment. J. Clin. Endocrinol. Metab. iiI, 438-445 (1980). 23. - A. S. McNEILLY, C. HAGEN, and G. M. BESSER, Long-term treatment of galactorrhea and hypogonadism with bromocriptine. Br. Med. J. II, 419-422 (1974). 24. TOBIN, H., J. B. J OSIMOVICH, and H. SALAZAR, The ultrastructure of the mammary gland during prolactin-induced lactogenesis in the rabbit. Endocrinology 90, 15G9-1577 (1972). 25. YANAI, R, and H. NAGASAWA, Suppression of mammary hyperplastic nodule formation and pituitary prolactin secretion in mice induced by ergocornine or 2-bromo-iX-ergocriptine. J. Nat. Cancer Inst. 41i, 1105-1112 (1970).
140