THE ANATOMICAL RECORD 223:252-256 (1989) Progesterone Secretion and Mitochondria1Size of Aging Porcine Corpora Lutea VALERIE ADAIR, MARVIN H. STROMER, AND LLOYD L. ANDERSON Department of Animal Science, Iowa State University, Ames, Iowa 50011 ABSTRACT A functional dependency between the nongravid uterus and the ovaries is essential to luteolysis and the return to estrus in the pig. After mating of gilts, the corpora lutea develop, and they are required for the maintenance of pregnancy t o a normal duration of about 114 days. Hysterectomy of luteal phase (day 6) nongravid gilts results in persistence of the corpora lutea to 150 days. We report that these corpora secrete greater quantities ( P < 0.025) of progesterone than during the later half of gestation (days 54-108). Although aging corpora lutea remain functional for at least an additional 35 days, an abrupt reduction by half in progesterone secretion (16 ng/ml) occurs about day 114 in hysterectomized gilts that coincides with the prepartum decrease to basal serum levels (<0.5 ng/ml) at parturition (day 114) and during lactation. Aging corpora lutea remain large (averaging >450 mg) on days 124 and 136 in hysterectomized gilts, whereas they regress (averaging < 75 mg) in the lactating dams. Mitochondria continue to increase in size in aging corpora lutea of hysterectomized gilts until day 136; in contrast, they decrease during the postpartum period in lactating dams. A precisely timed signal, possibly of ovarian origin or from the CNS and pituitary gland, entrains in hysterectomized and pregnant pigs at day 113 that results in marked shifts in relaxin and progesterone secretion. Progesterone secretion and mitochondria1 features suggest that porcine corpora lutea seem genetically controlled and are preprogrammed at estrus for the duration of pregnancy, regardless of the presence of conceptuses or absence of the uterus. Progesterone and relaxin are produced by porcine corpora lutea during pregnancy and after hysterectomy (Belt et al., 1971; Anderson et al., 1973, 1983; Musah et al., 1984). Although the normal duration of the estrous cycle is about 21 days in the pig, the corpora lutea secrete progesterone and small amounts of relaxin (Masuda et al., 1967; Anderson et al., 1973; Sherwood and Rutherford, 1981). Pregnancy lasts about 114 days in this species, and soon after mating progesterone secretion peaks early and remains elevated from days 8-108, when it decreases before parturition. The corpora lutea are required for the maintenance of pregnancy; ovariectomy any time results in abortion within 36 hr (du Mesnil du Buisson and Dauzier, 1957). Although these corpora in hysterectomized gilts are capable of remaining large ( > 450 mg) at least 35 days beyond the time of normal regression at parturition, abrupt changes occur in their capacity to secrete both progesterone and relaxin (Felder et al., 1986). The ephemeral nature of mammalian corpora lutea can be characterized by their autonomy of progesterone secretion, responsiveness to the luteolytic effects of prostaglandins, and their ability t o make prostaglandins (Anderson et al., 1969; Rothchild, 1981).We describe sequential profiles of progesterone secretion into peripheral blood during periods exceeding the normal life span of the corpora lutea in 0 1989 ALAN R. LISS, INC. hysterectomized gilts, as compared with pregnant and lactating animals, as well as the fine structure of mitochondria in aging IuteaI cells. MATERIALS AND METHODS Purebred Yorkshire gilts, approximately 125 kg body weight, that had exhibited at least one estrous cycle (21 f 2 days; mean f S.E.)were either mated at estrus (day 0) or remained unmated and were hysterectomized at day 6 by surgical procedures described previously (Anderson et al., 1961).Anesthesia was induced by intravenous injection of thiamylal sodium (0.8-1.0 gm Surital, Parke-Davis, Morris Plains, NJ) and maintained by a closed-circuit system of halothane (4-9%; Ayerst Laboratories, New York, NY) and 0 2 (500-1,000 cc/min).During hysterectomy, the corpora lutea on each ovary were marked with a loop of silk suture for later identification. Animals were laparotomized on days 100, 112, 124, and 136 to obtain luteal tissue for electron microscopy. During these laparotomies, samples of luteal tissue from ~ Received November 2, 1987; accepted August 22, 1988. 253 PROGESTERONE AND MITOCHONDRIA IN CORPORA LUTEA EXPERIMENTAL DESIGN MATED LAPAROTOMY FOR COLLECTION OF LUTEAL TISSUE ON DAYS 0 lm.112.124.136 1 PREGNANT YORKSHIRE GILTS (6) I DAY I I I Y I t 171I I’?I 1 I I t 171I tl?t BLOOD COLLECTED EVERY SIXTH DAV FROM ANTERIOR VENA CAVA HYSTERECTOMY 1 1 I I I I ITII I”? a PARTURITION MPAROTOMV FOR COLLECTION OF LUTEAL TISSUE ON DAYS 0 loo.112,124.136 VORKSHIRE DHTS HYSTERECTOMIZED ON DAY 6 (9) 1 I DAV D I I 24 40 I 96 1 1 1 1 1 I cells representing different regions of the corpora lutea were examined for mitochondrial size in each animal. The number of mitochondria measured per photomicrograph ranged from 18 to 160. The statistical analysis of data on mitochondrial size was based on animal as the experimental unit, not on luteal cell or mitochondria within cell. For statistical analysis the experimental units in this study were the individual gilts, and they were assigned to treatments a t random. Data were analyzed as a splitplot by using the statistical analysis for the general linear model and by Student’s t-test for comparisons among treatment groups (Snedecor and Cochran, 1980). 120 144 160 I. l. t. t. l. l. l, t. l, l. l , l . t . l . l . l . l . l . l 1 1 1 1 1 l t 1 72 BLOOD COLLECTED EVERV SIXTH DAY FROM ANTERIOR VENA CAVA Fig. 1. Description of experimental groups indicating days for sequential collections of peripheral venous blood for radioimmunoassay of progesterone and for sequential laparotomies to obtain luteal tissue for electron microscopic examination. The number of gilts in each group is indicated in parentheses. both ovaries were removed for electron microscopy. Blood samples were allowed to clot at 4°C and were then centrifuged (2,OOOg; 4”C), and serum was stored at -20°C for radioimmunoassay of progesterone. The description of experimental groups and number of gilts in each group are indicated in parentheses in Figure 1. Progesterone was extracted from duplicate aliquots of 200 p1 serum with benzene:hexane (1:2 vol/vol) by modification of a procedure described previously. A third replicate served as a recovery for determining procedural losses by addition of 5,000 cpm (1,2,6,7-[N]-3H)progesterone (97.0 Ci/mmol; New England Nuclear Corp., Boston, MA). Serum extracts were assayed for progesterone as described by Anderson et al. (1979) with the fully characterized antibody GDN 337 (Niswender, 1973; Gibori et al., 1977). The sensitivity of the assay was defined as the amount of progesterone that yielded 95% of cpm in buffer control tubes; this ranged from 50 to 80 pg. Intraassay and interassay coefficients of variance (CV) were 2.0 and 5.0%, respectively. Mean blanks from ovariectomized-hysterectomized gilts were < 0.05 ngiml (N = 4). The precision and accuracy were evaluated by adding 0.05, 0.10, 0.25,0.50, 1.00,2.50, and 5.00 ng/ml progesterone to serum from ovariectomized-hysterectomized gilts, and the recoveries of six replicates were (mean f S.E.) 0.07 f 0.01, 0.15 f 0.01, 0.39 f 0.01, 0.73 f 0.009, 1.41 0.01, 3.25 5 0.07, and 5.5 k 0.10 ng/ml, respectively. Fixation and dehydration of corpora lutea were done a t 2°C. Luteal tissue was cut into cubes <1 mm and fixed 3 h r in 2.5% glutaraldehyde in Millonig’s phosphate buffer, pH 7.25, rinsed in Millonig’s buffer, and transferred to 1% osmium tetroxide for 2 hr. During the last step of the acetone dehydrations, the tissue was allowed to warm (24°C) for infiltration and embedding in a n Epon-Araldite resin. Silver sections (60-80 pm) were cut on glass knives with a n LKB Ultrotome I11 microtome. Thin sections were double-stained (with 2% uranyl acetate in methanol and then lead citrate) and examined in a n RCA EMU-4 electron microscope. At least four randomly selected photomicrographs of luteal RESULTS The experimental design and details for the collection of sequential samples of blood (10 ml), hysterectomy, and the sequential laparotomies for collection of luteal tissue are presented in Figure 1. The concentrations of progesterone in peripheral blood serum were determined by radioimmunoassay as described in Figure 2. In nongravid gilts hysterectomized a t day 6, progesterone serum levels remained significantly greater ( P < 0.025) on days 54 through 108 than during pregnancy (Fig. 2). Parturition occurred a n average of 114 days after mating in the six pregnant gilts. Progesterone blood levels decreased abruptly in hysterectomized gilts at the time equivalent to that of parturition. By using a split-plot analysis, circulating levels of progesterone during days 114 to 168 were lower (P < 0.001) than those during days 54 to 108 in both hysterectomized and pregnant/ lactating gilts (Fig. 2). Hysterectomized group means remained significantly greater (P < 0.0001) than those of pregnantAactating animals during days 114 to 168. By day 120, progesterone concentrations were < 1ngiml and remained consistently low throughout lactation. A postpartum estrous cycle occurred in two of the six lac- HYSTERECTOMY PARTURITION c. Q DAYS AFTER ESTRUS Fig. 2. Progesterone concentrations in peripheral blood serum of six Yorkshire gilts during pregnancy and early lactation ( 0 )compared with those in nine unmated gilts hysterectomized (A)on day 6 (day 0 = estrus). During lactation a postpartum estrous cycle occurred in Y2084 (0) and Y2113 (U), whereas the other four animals remained acyclic as indicated by basal blood levels of progesterone from days 120-168. 254 V. ADAIR ET AL. Fig. 3.Mitochondria were prominent in luteal cells of hysterectomized and pregnant gilts, but they were significantly greater in diameter in the hysterectomized compared with pregnant gilts at day 100 (a, b). The mitochondria of luteal cells in both groups of gilts contained fenestrated larnelliform cristae at this time (a,b). From days 100 to 136, the corpora lutea in hysterectomized gilts remained large, but they regressed in the pregnant and lactating dams. By day 136, most luteal cells in hysterectomized gilts contained numerous large mitochondria and little evidence of regression (c), whereas a few of them contained mitochondria of different diameters and intracellular as well as intercellular collagen fibers. In lactating dams at day 136, regressed luteal tissue contained cells with collagen and few mitochondria (d). Bar = 1 pm. tating gilts, as indicated by increased progesterone levels after day 138 (Fig. 2). On days 100 and 112, large corpora lutea (>450 mg) Reproductive No. of Perimeter (pm)of mitochondria’ typify hysterectomized and pregnant gilts. The corpora status gilts Day 100 Day 112 Day 124 Day 136 lutea persisted and remained larger (averaging >450 pregnancyand 1.58 1.86 1.g7 1.44 mg) on days 124 and 136 in the hysterectomized gilts, whereas they regressed (averaging < 75 mg) in the laclactation After 1.80* 2.13* 2.43** 2,32** tating dams. Mitochondria were prominent in luteal cells and contained fenestrated lamelliform cristae in hysterectomy corpora lutea from both groups of hysterectomized and ‘Values are the group means for animals within reproductive status pregnantgilts (Fig. 3). At days 100 and 112, the mitoand day. Micrographs were randomly chosen from each sample. A sheet of clear acetate with parallel lines 1 cm apart was randomly were larger (‘<O.O1) in hysterectomized as placed on each micrograph. Mitochondria intercepted by the lines compared with those of pregnant gilts (S.E. difference = were measured with a Bioquant I1 Image Analysis System (R & M 0.067; Table 1). Although they increased (P<0.0005) in Biornetrics, Nashville, TN). size in both groups from days 100 to 112 (S.E. difference “ P < 0.01 compared with corresponding day of pregnancy and - 0.044), the mitochondria remained significantly larger lactation. * + P < 0.001 compared with corresponding day of pregnancy and in the hysterectomized gilts to day 136 (Fig. 3, Table 11, lactation. whereas they decreased in size in the lactating animals TABLE 1. Size of mitochondria in aging porcine corpora lutea PROGESTERONE AND MITOCHONDRIA IN CORPORA LUTEA by that time. Sequential changes in the fine structure of aging luteal cells in hysterectomized gilts and mated animals indicated marked shifts in populations of electron-dense granules, as has been observed in previous studies (Belt et al., 1971; Anderson et al., 1983; Fields and Fields, 1985; Anderson, 1987). DISCUSSION The results presented here indicate that circulating levels of progesterone in hysterectomized gilts are consistently greater than corresponding stages throughout normal pregnancy. Although the corpora lutea remain large to day 150, there is a n abrupt decrease in progesterone secretion in these hysterectomized gilts, which coincides with that found in normal animals at parturition. Estrone (El) and 176-estradiol (176-E~)are primarily of fetal-placental origin in this species, and circulating levels of them peak just preceding parturition (Fevre et al., 1968; Anderson et al., 1983). After hysterectomy, peripheral blood concentrations of E l and 170-Ez remain consistently low from days 6 to 168 (Anderson et al., 1983). Although exogenous estrogens are luteotropic in this species, the mechanisms of their action are undefined (Gardner et al., 1963; Anderson et al., 1973). In hysterectomized gilts given intramuscular injections of 17fl-E~to mimic blood levels of endogenous estrogen during normal pregnancy, progesterone secretion is suppressed to levels similar to those of normal pregnancy (Musah et al., 1984).Nevertheless, progesterone secretion decreases abruptly in hysterectomized gilts on days 113-114 in the absence or presence of exogenous estrogens. Furthermore, relaxin blood concentrations peak at day 113 in hysterectomized gilts, the same day as found during late pregnancy (Felder et al., 1986). Thus, extrinsic (uterine) prostaglandins may switch on intrinsic (luteal) prostaglandins to cause luteolysis a t the termination of a n estrous cycle or pregnancy, whereas after hysterectomy, the rate of luteal regression is limited to the intrinsic process. Whether luteal demise during late pregnancy results primarily from prostaglandins of uterine and luteal cell origin and during late hysterectomy from luteal cells is not known. Prolactin (PRL) secretion is tonically inhibited by the porcine hypothalamus, as indicated by elevated PRL blood levels after hypophysial stalk transection (Anderson et al., 1982). In mated gilts hypophysial-stalk transected at days 30 and 50, the pregnancies fail, whereas in those stalk-transected a t days 70 and 90, the pregnancies are maintained to normal term (du Mesnil du Buisson and Denamur, 1969). Exogenous PRL maintains pregnancy and progesterone secretion a t least 10 days in gilts hypophysectomized at day 70, whereas luteinizing hormone (LH) is ineffective (du Mesnil du Buisson, 1973). In hypophysectomized-hysterectomized gilts, LH or human chorionic gonadotropin (hCG) maintains luteal function and progesterone secretion to day 30, whereas the corpora lutea regress in those similarly treated but with the uterus intact (Anderson et al., 1965, 1967). Daily injections of purified porcine PRL from days 110120 in pregnant and hysterectomized gilts increased circulating levels of progesterone and relaxin (Felder et al., 1988). These results provide clear evidence that porcine PRL prolongs secretory function by aging corpora lutea with sustained progesterone as well as relaxin 255 secretion in pregnant and hysterectomized gilts. Furthermore, PRL maintains aging corpora lutea and progesterone secretion from days 110-120 in hypophysectomized-hysterectomized gilts (Li et al., 1987). The data in Table 1 indicate that, a t estrus, a n unknown process begins that results in increased luteal mitochondria size. Mitochondria reached their maximum size in pregnant gilts at parturition and are maintained a t that size for about 10 days after parturition. Because luteal mitochondrial size in hysterectomized gilts peaked approximately 10 days later and regressed only slightly by day 136, a signal from the uterus and/ or a hormonal shift associated with lactation may be required for mitochondrial regression. The similar amount of decline in serum progesterone levels in both groups of animals suggests that mitochondrial size is not directly related to progesterone synthesis. Larger and more mitochondria may, however, be required to meet the metabolic needs of the larger corpora lutea in the hysterectomized gilts. The evidence presented here indicates that the presence of the conceptuses or the absence of the uterus, fetuses, and their associated placental estrogens are required for prolonged progesterone secretion in the pig. As the corpus luteum grows older, the aging luteal cells may be preprogrammed to have a n inherent life span of approximately 113 days. Mechanisms initiating a decrease in progesterone secretion during pregnancy and after hysterectomy may be related to aging of porcine luteal cells. These abrupt and autonomous shifts in progesterone secretion on about days 113-114 in hysterectomized as well as pregnant pigs may be genetically controlled. Progesterone secretion continues at halfmaximal values in hysterectomized gilts while the mitochondria continue to increase in size. Although exogenous PRL can maintain elevated progesterone secretion by aging corpora lutea in hysterectomized gilts, a role for endogenous PRL secretion in the abrupt shifts in progesterone secretion is unknown. ACKNOWLEDGMENTS We thank Professor D.F. Cox for statistical analyses and M.E. Shell, C.R. Bohnker, A. Alcivar, R. Perezgrovas, J.R. Molina, and R. Poyner for assistance. This work was supported by U.S. Department of Agriculture, A R S , CSRS, and OGPS Competitive grant 85-CRCR-11862. Journal Paper 5-12077 of the Iowa Agriculture and Home Economics Experiment Station, Ames, IA (Projects 2443, 2444, and 2754). LITERATURE CITED Anderson, L.L. 1987 Regulation of relaxin secretion and its role in pregnancy. Adv. Exp. Med. Biol., 219:421-463. Anderson, L.L., V. Adair, M.H. Stromer, and W.G. McDonald 1983 Relaxin production and release after hysterectomy in the pig. Endocrinology, 113:677-686. Anderson, L.L., J.G. Berardinelli, P.V. Malven, and J.J. Ford 1982 Prolactin secretion after hypophysial stalk transection in pigs. Endocrinology, IIlt380-384. Anderson, L.L., K.P. Bland, and R.M. Melampy 1969 Comparative aspects of uterine-luteal relationships. Recent Progr. Horm. Res., 25:57-104. Anderson, L.L., R.L. Butcher, and R.M. Melampy 1961 Subtotal hysterectomy and ovarian function in gilts. Endocrinology, 69:571580. Anderson, L.L., G.W. Dyck, H. Mori, D.M. Henricks, and R.M. Melampy 1967 Ovarian function in pigs following hypophysial stalk 256 V. ADAIR ET AL. transection or hypophysectomy. Am. J. Physiol., 212:1188-1194. Anderson, L.L., J.J. Ford, R.M. Melampy, and D.F. Cox 1973 Relaxin i n porcine corpora lutea during pregnancy and after hysterectomy. Am, J. Physiol., 225:1215-1219. Anderson, L.L., D.1,. Hard, and L.P. Kertiles 1979 Progesterone secretion and fetal development during prolonged starvation i n the pig. Am, J. Physiol., 256:E335-IXl41. Anderson, L.L., P.C. LBgIise, F. du Mesnil du Buisson, and P. Rombauts 1965 Interaction dcs hormones gonadotropes et de l’ut&us dans le maintien du tissu luteal ovarien chez la truie. C.R. Acad. Sci. Paris, 261:3675-3678. Belt, W.D., L.L. Anderson, L.F. Cavazos, and R.M. hIelamp1- 1971 Cytoplasmic granules and relaxin levels in porcine corpora lutea. Endocrinology, 89: 1-10. du Mesnil du Buisson, F. 1973 Pacteurs luteotropes chez l a truie. In: Le Corm Jaune. R. Denamur and A. Netter. eds. Masson e t Cie. Paris: pp. 225-237. du Mesnil du Buisson, F., and L. Dauzier 1967 lnlluence de I’ovariectomie chez la truie pendant la gestation. C E l . Soc. Biol., 151t311313. du Mesnil du Buisson, F., and R. Ilenamur 1969 Mecanismes du contrBle de la fonction luteale chcz la truie, la brebis et la vache. In: IIIrd International Congress on Endocrinology. C. Gaul, ed. Excerpta Medica. Amsterdam, pp. 927-934 Felder, K.J., J. Klindt, D.J. Bolt, and L.L. Anderson 1988 Relaxin and progesterone secretion as affected by luteinizing hormone and prolactin after hysterectomy in t h e pig. Endocrinology, 122:1751-1760 Felder, K..J., J.R. Molina, A.M. lknoit, and L.L. Anderson 1986 Precise timing for peak relaxin and (lccrcasrd pi‘ugestcrnnc. secretion after hysterectomy in the pig. Endocrinology, 119r1502-1509. FBvre, J., P.C. Leglise, and 1’. Rombauts 1968 Du role de l’hypophyse e t des ovain.s dnn la biosynthese des oestrogcnes a u cours de l a gestation chez la truie. Ann. Biol h i m . Hiochirn. Biophys., 8:225233. Fields, P.A., and M.J. Fields 1985 Ultrastructural localization of relaxin in t h e corpus l u t c w r r i ()[ lhe nonpregnant, pseudopregnant, and pregnant pig. Uiol. Itqirod., :32:1169-1179. Gardner, M.L., N.L. Firsl, and L.E. Casida 1963 Effect of exogenous estrogens on corpus Iriteurn maintenance in gilts. J. Anim. Sci., 22:132-134. Gibori, G., E. Pintezak, a n d I. Rothchild 1977 The role of estrogen in the regulation of lutcxl progcslerone secretion in t h e r a t after day 12 of pregnancy. Kndocrinology, 100:1483-1495. Li, Y., J.R. Molina, J. Klindt, D.J. Bolt, and L.L. Anderson 1987 Prolactin maintains progesterone secretion by aging corpora lutea in hypophysectomizcd pigs. J. Anim. Sci. [Suppl. 11, 65:366 (Abstract 410). Musah, A.I., J.J. Ford. and L.L. Anderson 1984 Progesterone secretion a s affected by 17fi-estradiol after hysterectomy in the pig. Endocrinology, 115~1876-1882. Masuda, H., L.L. Anderson, T1.M. Henricks, and R.M. Melamp?. 1967 Progesterone in ovarian venous plasma and corpora lutea of the pig. EndocrinoloLT, 80:240-246. Niswender, G.D. 1973 Influence of the site of conjugation on the specificity of antibodies to progesterone. Steroids, 22:413-423. Rothchild, I. 1981 The regulation of the mammalian corpus luteum. Recent Prog. Horm. Res., 37:183 298. Sherwood, O.D., and J.E. Rutherfnrd 1981 Relaxin imrnunoactivity levels i n ovarian extracts obtained frum r a t s during variocs reproductive states and from adult cycling pigs. Endocrinology, 108.11711177. Snedecor, G.W., and W.G. Ctxhran (1980) Statistical Methods, ed. 6. Iowa State University Press, Ames, p. 258.