0013-7227/90/1272-0637$02.00/0 Endocrinology Copyright© 1990 by The Endocrine Society
Vol. 127, No. 2 Printed in U.S.A.
Reorientation of Prostaglandin F Secretion by Calcium Ionophore, Estradiol, and Prolactin in Perifused Porcine Endometrium* TIMOTHY S. GROSSt, MARK A. MIRANDO, KATHLEEN H. YOUNG*, SASKIA BEERS, FULLER W. BAZER, AND WILLIAM W. THATCHER Dairy Science Department (T.S.G., W. W.T.) and Animal Science Department (M.A.M., K.H. Y., S.B., F.W.B.), University of Florida, Gainesville, Florida 32611
3), pregnant (n = 4), and EV-induced pseudopregnant (n = 4) gilts was collected on day 14 after estrus and perifused with BSA or PRL to both surfaces. Orientation of PGF secretion was initially endocrine in Exps 1 and 2 (P < 0.01). In Exp 1, calcium ionophore shifted the orientation of PG secretion from endocrine to exocrine (P < 0.01). In Exp 2, neither EV in vivo nor PRL in vitro alone altered the orientation of PGF secretion. However, EV and PRL interacted to reorient secretion of PGF from endocrine to exocrine at 6 h (P < 0.01) and 12 h (P < 0.01) after EV. In Exp 3, the orientation of PGF secretion was endocrine in cyclic gilts and exocrine in pregnant and pseudopregnant gilts (status x side; P < 0.05). PRL did not alter the orientation of PGF secretion regardless of reproductive status. These results suggest that the reorientation of endometrial PG secretion in pigs during the establishment of pregnancy involves interactive effects of estrogens and PRL, possibly through increased calcium cycling across the uterine epithelium. (Endocrinology 127: 637642, 1990)
ABSTRACT. Establishment of pregnancy in pigs requires a shift in endometrial prostaglandin (PG) F secretion from an endocrine (toward the myometrium and uterine vasculature) to an exocrine (toward the uterine lumen) orientation. Three experiments utilized bilateral endometrial perifusion devices for separate perifusion of myometrial and luminal surfaces to determine whether promoting calcium cycling across the luminal epithelial surface, which may be induced by interactive effects of estradiol and PRL, is involved in the reorientation of PGF secretion during establishment of pregnancy in pigs. In Exp 1, endometrium from cyclic gilts (n = 7) on day 14 after estrus was perifused with either saline (control) or calcium ionophore A23187 added to luminal surface perifusion buffer. In Exp 2, cyclic gilts (n = 7) at day 11 after estrus received an im injection of estradiol valerate (EV) after unilateral hysterectomy and the remaining EV-stimulated uterine horn removed at 6 h (n = 3) or 12 h (n = 4) after EV. Both surfaces of these endometrial samples were perifused with buffer containing either 250 ng/ml BSA or porcine PRL. In Exp 3, endometrium from cyclic (n =
E
STABLISHMENT of pregnancy involves maintenance of luteal function due to suppression or alteration of uterine luteolytic mechanisms (1). In cattle (2) and sheep (3), the antiluteolytic mechanism during early pregnancy involves an attenuation of endometrial responsiveness to stimulators of prostaglandin (PG) biosynthesis {e.g. oxytocin). However, porcine endometrium is capable of responding to oxytocin with increased PG secretion regardless of reproductive status (4). Bazer and
Received March 31,1990. Address requests for reprints to: Dr. F. W. Bazer, Animal Science Department, Building 459, Shealy Drive, University of Florida, Gainesville, Florida 32611. * Part of these studies was presented at the Annual Conference of the Society for the Study of Fertility, Edinburgh, Scotland, 1988 (Abstract 106). This work was supported in part by United States Department of Agriculture Grants 86-CRCR-1-2106 and 85-CRCR-l1871 and is published as Journal series no. 9675 of the Florida Agricultural Experimental Station. t Current address: Henry Doorly Zoo, Omaha, Nebraska 68107. X Current address: Department of Biochemistry, Molecular and Cellular Biology, Northwestern University, Evanston, Illinois 60208.
Thatcher (5) proposed that a reorientation of endometrial PGF release into the uterine lumen (exocrine mode) occurs at the onset of pregnancy in pigs. Using a perifusion system developed by Lacroix and Kann (6), Gross et al. (4) demonstrated that in vitro secretion of PGF was in an endocrine direction, toward the myometrium and the utero-ovarian vasculature, during the estrous cycle. However, during early pregnancy, secretion of PGF was exocrine, toward the uterine lumen. Such a reorientation in PG secretion could prevent release of PGF into the uterine vasculature and allow corpus luteum maintenance during early pregnancy in the pig as proposed by Bazer and Thatcher (5). This reorientation in endometrial secretion of PGF occurs between days 10 and 12 (4) of gestation when estrogen secretion is initiated by pig blastocysts (7-9). Pseudopregnancy can be induced in pigs by administration of estrogens during days 11 through 15 of the estrous cycle (10). Conceptus products, possibly estrogens, alter endometrial calcium dissipation and possibly calcium
637
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 15 November 2015. at 22:41 For personal use only. No other uses without permission. . All rights reserved.
638
REORIENTATION OF ENDOMETRIAL PG SECRETION
flux (11). Associated events might allow PGF to be sequestered in the uterine lumen so that it is less available to exert a luteolytic effect (5, 11). In addition, initiation of secretion of estrogens by pig blastocysts is associated with an acute release of calcium into the uterine lumen on days 11 and 12 of the estrous cycle, and then its reuptake by endometrial and/or conceptus tissues (11, 12). PG secretion by endometrial explants from guinea pigs (13) is dependent upon the presence of extracellular calcium. Agents such as calcium ionophore (Cal) A23187 increase calcium cycling across cell membranes (14) and are capable of altering PG biosynthesis (15, 16). Therefore, Cal might increase calcium flux or calcium uptake by endometrium and thereby alter endometrial secretion of PGF. PRL receptors have been detected in pig endometrial membranes, and a decrease in electrolyte release into the uterine lumen of hypoprolactinemic pigs has been reported (17). Estradiol induces endometrial PRL receptors in the pig (18, 19) within 6 h after treatment. Therefore, the possibility of an interaction between estrogen and PRL for the induction of calcium cycling across the endometrial epithelium exists. Increased calcium cycling might then alter the orientation of endometrial secretion of PGF similar to that observed during early pregnancy in pigs. The present study determined whether Cal would alter the orientation of secretion of PGF by perifused endometrium from cyclic gilts. In addition, perifused endometrium from cyclic gilts was used to determine interactions between estrogen and PRL which could change the direction of endometrial secretion of PGF from an endocrine to an exocrine direction.
Materials and Methods Materials Radioisotope [5,6)8)ll)12,14,15-3H]PGF2a (specific activity: 180 Ci/mmol) was purchased from Amersham Corporation (Arlington Heights, IL). Antiserum to PGF2a was provided courtesy of Dr. T. G. Kennedy (University of Western Ontario, Ontario, Canada). Calcium ionophore A23187 was purchased from Sigma Chemical Company (St. Louis, MO). PRL (USDABl) was supplied courtesy of Dr. Doug Bolt (United States Department of Agriculture, Agriculture Research Service, Beltsville, MD). Animals Sexually mature cross-bred (Duroc x Yorkshire x Hampshire) gilts were observed for estrous behavior in the presence of a boar at 0730 h each day. Day of onset of estrus was designated day 0 of the cycle. All gilts were assigned randomly to cyclic (n = 17; not bred), pregnant (n = 4, bred to boar at 0, 10, and 24 h after estrus), or pseudopregnant (n = 4) reproductive statuses. Pseudopregnancy was induced by daily sc injec-
Endo• 1990 Vol 127-No 2
tions of estradiol valerate (5 mg in 1 ml corn oil) on days 11 through 13 of the estrous cycle. Pregnant and cyclic gilts for Exp 3 were treated similarly with corn oil. Exp 1: effect of Cal on the orientation of secretion of PGF by perifused endometrium from cyclic gilts Gilts (cyclic; n = 7) were hysterectomized and endometrial tissue obtained on day 14 of the estrous cycle. Endometrium was dissected from myometrial tissue and placed into sterile Krebs-Ringer Bicarbonate solution (KRB) which was gassed with O2-CO2 (95:5), washed twice with fresh KRB and dissected into two pieces of approximately 4 cm2 each as described previously (4). Each endometrial sample (two per gilt) was placed into a bilateral perifusion device which allowed separation of myometrial and luminal surfaces so that each was perifused in a 0.5 ml hemichamber (4, 6). Tissue was stretched and anchored into each device with the aid of four needles.
Each chamber, submerged in a 39 C water bath, was attached to dual 60 ml syringes containing freshly gassed KRB and perifused at a rate of 3 ml/10 min for 2.5 h using a Harvard Infusion Pump System (Harvard Apparatus Co., Dover, MA). Fractions were collected every 10 min (3 ml) using a fraction collector (Gilson Electronics Inc., Middleton, WI) for a total of 15 fractions. Treatments were administered to the luminal surface during fractions 4-12. Treatments were: 1) control; and 2) Cal A23187 (40 /IM) to stimulate calcium flux. Fractions were stored at —20 C until analyzed for concentrations of PGF. Exp 2: interactive effect of estrogen and PRL on the orientation of PGF secretion by perifused endometrium from cyclic gilts Three gilts (cyclic) were unilaterally hysterectomized on day 11 after estrus and, immediately thereafter, received an im injection of estradiol valerate (5 mg in 1 ml corn oil). The remaining uterine horn was removed at 6 h after the estradiol valerate injection. Endometrium from each horn was collected and placed into two perifusion chambers as described in Exp 1. Endometrium was perifused with KRB for 2.5 h and fractions collected every 10 min. Treatment with either BSA or PRL (250 ng/ml) was administered in perifusate to both surfaces (two chambers per uterine horn per animal) during fractions 4-12. Four additional gilts (cyclic), at day 11 after estrus, received an im injection of estradiol valerate (5 mg in 1 ml corn oil) immediately after unilateral hysterectomy and the remaining uterine horn was removed at 12 h after estradiol valerate injection. Endometrium from each horn was collected and placed into two perifusion chambers as described in Exp 1. Endometrium was perifused with KRB for 3 h and fractions collected every 10 min. Treatment with either BSA or PRL (250 ng/ml) was administered in perifusate to both surfaces (two chambers per uterine horn per animal) during fractions 4-15. Fractions were stored at —20 C until analyzed for concentration of PGF. Exp 3: effect of PRL on the orientation of PGF secretion by perifused endometrium from cyclic, pregnant, and estrogeninduced pseudopregnant gilts Cyclic (n = 3, not bred), pregnant (n = 4, bred to boar at 0, 10, and 24 h after estrus), and pseudopregnant (n = 4) gilts
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 15 November 2015. at 22:41 For personal use only. No other uses without permission. . All rights reserved.
REORIENTATION OF ENDOMETRIAL PG SECRETION were hysterectomized on day 14 and endometrium placed into two perifusion chambers for each gilt as described in Exp 1. Endometrium was perifused with KRB for 2.5 h, fractions collected every 10 min, and treatments of either BSA or PRL (250 ng/ml) administered to both surfaces during fractions 412. Fractions were stored at —20 C until analyzed for concentrations of PGF.
RIAofPGF Fractions were analyzed for PGF using a direct RIA procedure (4) with an antibody characterized by Kennedy (20) and [3H]PGF2a (described above). Standard curves were prepared in KRB with known amounts of radioinert PGF2a (0, 10, 25, 50, 100, 250, 500, 1000, 2500, and 5000 pg). An antiserum dilution of 1:5000 was used, and the minimum concentration per tube which was distinguishable from zero was 10 pg. Crossreactivities of the PGF antiserum with other PGs were 94% for PGFia, 2.4% for PGE2, and less than 0.1% for 13,14-dihydro15-keto-PGF2a, PGE1( and arachidonic acid. Unextracted KRBperifusate samples were assayed for PGF in duplicate using 0.3 ml aliquots. A pooled KRB-perifusate sample (~10 ng PGF/ ml) was assayed serially in 0.01, 0.025, 0.05, 0.1, 0.2, and 0.3 ml volumes (final volume of 0.3 ml with blank KRB). This inhibition curve was parallel to the standard curve, with the test for homogeneity of regression indicating that the curves did not differ. Further characterization of the assay involved measurement of known amounts (10, 25, 50,100, 250, 500,1000, and 2500 pg) of PGF in KRB-perifusate [Y = -5.7 + 1.03X; Y = amount of PGF measured (picograms/0.3 ml); X = amount of PGF added (picograms/0.3 ml); R2 = 0.947]. Interassay and intraassay coefficients of variation were 12.0% and 13.1%, respectively. Statistical analyses Data were analyzed statistically by least squares analysis of variance using the General Linear Models procedure of the Statistical Analysis System (21). PG secretion rates (PGF) were analyzed for a split-plot design with sample periods (time) as the subplot effect. For Exp 1, data were analyzed using the model components of gilt, treatment, side, time, and all interactions. For Exp 2, data were analyzed using the model components of gilt, PRL, estrogen, side, time, and all interactions. For Exp 3, data were analyzed using the model components of reproductive status, gilt (nested within status), PRL, side, time, and all interactions. Effects of treatment, side, and reproductive status were examined over time for the entire perifusion period in each experiment. Tests of hypotheses were made using appropriate error terms according to expectations of the mean squares (22) with gilt considered to be random and all other effects considered to be fixed.
Results Expl This experiment determined whether endometrial secretion of PGF from the myometrial and/or luminal surfaces was affected by treatment with Cal. Endome-
639
trium from cyclic gilts at day 14 after estrus secreted PGF in primarily an endocrine orientation (Table 1); secretion of PGF was 80% greater (P < 0.01) from the myometrial than luminal surface during the pretreatment period. Interactions of treatment X side (P < 0.01) and treatment X side X time (P < 0.01) indicated that Cal A23187 (stimulates calcium flux) reoriented secretion of PGF more toward an exocrine direction within 1 h of treatment (Table 1) such that only 45% of the total PGF was secreted from the myometrial surface during the final 30 min of treatment (fraction 10-12) compared with 65% from the myometrial surface of control endometrium. Cal primarily increased (56%) secretion of PGF from the luminal surface, but also decreased (13%) secretion from the myometrial surface. However, when treatment with Cal was terminated, rates of secretion of PGF returned to pretreatment levels and primarily an endocrine orientation (Table 1). Cal did not significantly alter overall rate of PGF secretion. Exp 2 This experiment determined whether an interaction between estradiol valerate and PRL affects secretion of PGF from perifused endometrium. Endometrium from Day 11 of the estrous cycle secreted PGF in primarily an endocrine orientation (P < 0.01; Tables 2 and 3) which was 81% (Table 3) to 118% (Table 2) greater from the myometrial than luminal surface during the pretreatment period (treatment with PRL). Interactions of estradiol x PRL (P < 0.01) and estradiol x PRL x side (P < 0.05) were detected for tissue obtained at 6 h after estradiol treatment (Table 2) and estradiol X PRL X side X time (P < 0.01) for tissue obtained 12 h after estradiol treatment (Table 3). Treatment with PRL did not alter secretion of PGF by perifused endometrium obtained before estradiol valerate treatment (i.e. at 0 h; Tables 2 and 3). However, for endometrium taken from gilts at 6 h (Table 2) and 12 h (Table 3) after estradiol valerate treatment and removal of the first uterine horn, PRL increased secretion of PGF from the luminal surface by 27% (Table 3) to 46% (Table 2) while decreasing secretion from the myometrial surface by 17% (Table 3) to 18% (Table 2). These changes in PGF secretion, in response to an interaction between estradiol valerate and PRL, reoriented PGF secretion from an endocrine to an exocrine direction within 1 h of in vitro treatment such that only 46% of the total PGF was secreted from the myometrial surface during the final 30 min of treatment with PRL (Table 2, fractions 10-12 and Table 3, fractions 13-15), compared with 69% from the myometrial surface of BSA-treated endometrium. Treatment with estradiol in vivo or PRL in vitro did not significantly
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 15 November 2015. at 22:41 For personal use only. No other uses without permission. . All rights reserved.
640
REORIENTATION OF ENDOMETRIAL PG SECRETION
Endo-ioao Voll27«No2
TABLE 1. Effect of Cal on endometrial secretion rates of PGF (pg/10 min) by luminal and myometrial surfaces at day 14 after estrus for Exp 1 Fraction no. Treatment period
Surface
Treatment
Luminal Myometrial Luminal Myometrial
Control Cal, A23187
1-3
4-6
7-9
10-12
13-15
399 671 361 700
378 790 504 664
396 815 547 620
382 714 643 546
405 797 474 673
Endometrium was perifused with KRB (3 ml/10 min) for 2.5 h, and fractions were collected every 10 min (15 fractions total). Treatments (control and Cal A23187) were administered to the luminal surface during fractions 4-12. There were effects of side (P < 0.01), treatment x side (P < 0.01), and treatment x side x time (P < 0.01). Pooled SEM = 35.2, and each value represents the mean of three samples obtained from each of seven gilts. TABLE 2. Endometrial secretion rates of PGF (pg/10 min) by luminal and myometrial surfaces on day 11 after estrus for Exp 2 at 0 and 6 h after estrogen treatment in vivo Hours after estrogen
Treatment
0
BSA
0
PRL
6
BSA
6
PRL
Fraction no. Treatment period
Surface
Luminal Myometrial Luminal Myometrial Luminal Myometrial Luminal Myometrial
1-3
4-6
7-9
10-12
13-15
185 455 228 445 176 403 177 386
180 473 232 396 204
199 529 182 400 183 423 246 341
190 427 410 186 483 327
185 431 200 498 187 565 308
274
403
428 200 331
174
Endometrium was perifused with KRB (3 ml/10 min) for 2.5 h, and fractions were collected every 10 min (15 fractions total). Treatments (BSA and PRL) were administered to luminal and myometrial surfaces during fractions 4-12. There were effects of side (P < 0.01), estrogen x PRL (P < 0.01), and estrogen x PRL x side (P < 0.05). Pooled SEM = 18.6, and each value represents the mean of three samples obtained from each of three gilts. TABLE 3. Endometrial secretion rates of PGF (pg/10 min) by luminal and myometrial surfaces on day 11 after estrus for Exp 2 at 0 and 12 h after estrogen treatment in vivo
Hours after estrogen
Treatment
0
BSA
0
PRL
12
BSA
12
PRL
Fraction no. Treatment period
Surface Luminal Myometrial Luminal Myometrial Luminal Myometrial Luminal Myometrial
1-3
4-6
7-9
10-12
13-15
16-18
297 646 371 563 323 622 332 704
329 637 357 625 339 583 329 729
341 632 345 554 296 586 381 632
309 617 349 585 282 530 439 525
327 600 377 582 301 609 535 463
319 636 341 503 306 528 427 645
Endometrium was perifused with KRB (3 ml/10 min) for 3 h, and fractions were collected every 10 min (18 fractions total). Treatments (BSA and PRL) were administered to luminal and myometrial surfaces during fractions 4-15. There were effects of side (P < 0.01) and estrogen x PRL x side x time (P < 0.01). Pooled SEM = 51.5, and each value represents the mean of three samples obtained from each of four gilts.
alter overall rate of PGF secretion.
This experiment determined whether PRL could enhance the endocrine to exocrine reorientation of PGF secretion that occurs during early pregnancy and estro-
gen-induced pseudopregnancy. Endometrium from cyclic gilts at day 14 after estrus secreted PGF primarily in an endocrine orientation, whereas endometrium from pregnant and pseudopregnant gilts at day 14 after estrus secreted PGF primarily in an exocrine direction (Table 4; status X side interaction, P < 0.05). Endometrium from cyclic gilts secreted 63% of the PGF from the myometrial surface and 37% from the luminal surface
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 15 November 2015. at 22:41 For personal use only. No other uses without permission. . All rights reserved.
REORIENTATION OF ENDOMETRIAL PG SECRETION TABLE 4. Endometrial secretion rates of PGF (pg/10 min) by luminal and myometrial surfaces for cyclic, pregnant, and estrogen-induced pseudopregnant gilts at day 14 after estrus for Exp 3
Status Cyclic
Treatment
Surface
BSA
Luminal Myometrial Luminal Myometrial Luminal Myometrial Luminal Myometrial Luminal Myometrial Luminal Myometrial
PRL Pregnant
BSA PRL
Pseudopregnant
BSA PRL
Fraction no. Treatment period 1-3
4-12
1041 1913 1108 1740
606
813
1720 1061 1825 1460
2232 1275 2261 1245
936 640
1866 1400 732 448 709 506
13-15
904
816
2513 1754 1106
2216 1656 1379
673
557
1165
1209
632
913
Endometrium was perifused with KRB (3 ml/10 min) for 2.5 h, and fractions were collected every 10 min (15 fractions total). Treatments (BSA and PRL) were administered to luminal and myometrial surfaces during fractions 4-12. There was an effect of status X side (P < 0.05) but no treatment effect regardless of reproductive status. Pooled SEM = 243.4, and each value represents the mean of three, nine, and three samples (for fraction numbers 1-3, 4-12, and 13-15, respectively) obtained from each of three cyclic, four pregnant, and four pseudopregnant gilts.
during the pretreatment period, whereas endometrium from pregnant and pseudopregnant gilts secreted only 42 and 40% (respectively) of the PGF the myometrial surface and 58 and 60% (respectively) from the luminal surface during the pretreatment period. PRL did not significantly alter PGF secretion from endometrium regardless of reproductive status, nor did PRL enhance the endocrine to exocrine reorientation of PGF secretion of pregnant and pseudopregnant gilts. Mean PGF secretion also did not differ significantly among reproductive statuses. Discussion Previous results (4) demonstrated that secretion of PGF during the estrous cycle in gilts is primarily endocrine in direction (toward the myometrium and uterine vasculature), whereas secretion of PGF during early pregnancy is primarily exocrine in direction (toward the uterine lumen). Results from the present study confirm this orientation in direction of endometrial PGF secretion in cyclic and pregnant gilts at day 14 after estrus. In addition, these data indicate that estrogens, possibly of conceptus origin, are involved in the establishment of pregnancy in pigs by reorienting endometrial PGF secretion into the uterine lumen. Estrogens appear to initiate a calcium surge (12) which alters endometrial function. Associated events allow PGF to be sequestered in the uterine lumen so that it is
641
unavailable to exert a luteolytic effect. In addition, initiation of secretion of estrogens by pig blastocysts is associated with an acute release of calcium into the uterine lumen on days 11 and 12 of the estrous cycle, and then its reuptake by endometrial and/or conceptus tissues during early pregnancy (11,12). Secretion by PGF by endometrial explants from guinea pigs (13) and cows (Danet-Desnoyers, G., T. S. Gross, and W. W. Thatcher, unpublished results) is dependent upon the presence of extra-cellular calcium. Agents such as Cal A23187 increase calcium cycling across cell membranes (14) and, therefore, are capable of altering PG biosynthesis (15, 16). The present results demonstrate a reorientation of secretion of PGF by perifused endometrium from cyclic gilts when treated with Cal and indicate a possible role of calcium cycling across the endometrial epithelium in this reorientation of endometrial secretion of PGF during early pregnancy. Cal has been shown to primarily increase intracellular calcium (14); however, Cal has also been shown to directly affect cell membranes and subsequent PG synthesis in the absence of calcium (15). Therefore, the effects of Cal on PG secretion observed in the current study might be direct effects rather than mediated through increased calcium flux. Additional studies with other factors that alter calcium uptake/flux are necessary to demonstrate the mechanism by which Cal alters PG secretion. Preliminary studies (Gross, T. S., and F. W. Bazer, unpublished results) do not indicate any direct alteration in PG secretion by perifused endometrium from gilts in response to Verapamil (blocks calcium uptake), Amiloride (inhibits sodium transport), Trimethoxybenzoic acid (calcium antagonist), or KC1. Additional determinations in conjunction with Cal might indicate the possible mechanism by which Cal alters endometrial PG secretion. Estrogens also induce endometrial PRL receptors in the pig (18, 19). PRL receptors have been demonstrated for endometrial membranes from pigs, and a decrease in movement of electrolytes into the uterine lumen of hypoprolactinemic pigs has been detected (17). Therefore, an interaction between estrogen and PRL may influence induction of calcium cycling across the endometrial epithelium. Indeed, PRL increased secretion of PGF from the luminal surface while decreasing secretion from the myometrial surface of endometrium collected at 6 or 12 h after estradiol valerate treatment and removal of the first uterine horn on day 11 of the estrous cycle. An effect on uterine function of the remaining horn 6 to 12 h after unilateral hysterectomy has not been detected (12) and, therefore, the responses observed after PRL treatment in the present study are believed to be due to prior exposure to estradiol valerate rather than to the previous surgical procedure. These results indicate an
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 15 November 2015. at 22:41 For personal use only. No other uses without permission. . All rights reserved.
642
REORIENTATION OF ENDOMETRIAL PG SECRETION
interactive effect of estrogen and PRL, which results in a partial reorientation in the direction of secretion of PGF. Such an interaction between estrogen and PRL might be involved in the reorientation of endometrial secretion of PGF that normally occurs during early pregnancy in pigs. In contrast, the orientation and magnitude of PGF secretion by endometrium collected on day 14 after estrus was not altered by PRL treatment regardless of reproductive status (cyclic, pregnant, or estrogen-induced pseudopregnant). These results indicate that endometrium is no longer capable of responding to PRL once PGF secretion has reoriented from an endocrine to an exocrine direction which occurs between days 10 and 12 of pregnancy and pseudopregnancy. In addition, these results indicate that effects of PRL may be limited to before day 14 after estrus if involved in the establishment of pregnancy. Indeed, endometrial PGF secretion has been shown to begin to reorient from an endocrine to exocrine direction by day 12 of pregnancy (4). Therefore, an ability to respond to PRL at day 11 after estrus (Exp 2), but not at day 14 (Exp 3) might indicate an involvement of PRL along with estrogen in the establishment of pregnancy. In conclusion, estrogens (from the conceptuses or other sources) may act on the endometrium to induce PRL receptors which then allow PRL to bind to its receptor and stimulate an increase in calcium cycling across the endometrial epithelium. This calcium cycling may be associated with the reorientation of direction of secretion of PGF from an endocrine (toward the uterine vasculature) to an exocrine (toward the uterine lumen) direction. Since the reorientation of secretion of PGF from an endocrine to an exocrine direction occurs in gilts induced into pseudopregnancy by exogenous estradiol valerate (4), there is no basis to assume that a secretory product of the conceptus other than estrogen is required to induce this phenomenon.
3. 4.
5.
6. 7. 8.
9. 10.
11.
12.
13. 14. 15. 16. 17. 18.
Acknowledgment The technical assistance of Ms. M. F. V. Fliss is greatly appreciated. 19.
References 20. 1. Thatcher WW, Hansen PJ, Gross TS, Helmer SD, Plante C, Bazer FW 1989 Antiluteolytic effects of bovine trophoblast protein-one. J Reprod Fertil [Suppl]37:91-99 2. Gross TS, Thatcher WW, Lacroix MC, Hansen PJ 1988 Prostaglandin secretion by perifused bovine endometrium: secretion to-
21. 22.
Endo • 1990 Voll27-No2
wards the myometrial and luminal sides at day 17 after estrus as altered by pregnancy. Prostaglandins 35:343-357 Bazer FW, Vallet JL, Harney JP, Gross TS, Thatcher WW 1989 Comparative aspects of maternal recognition of pregnancy between sheep and pigs. J Reprod Fertil [Suppl] 37:85-89 Gross TS, Lacroix MC, Bazer FW, Thatcher WW, Harney JP 1988 Prostaglandin secretion by perifused porcine endometrium: further evidence for an endocrine versus exocrine secretion of prostaglandins. Prostaglandins 35:327-342 Bazer FW, Thatcher WW 1977 Theory of maternal recognition of pregnancy in swine based on oestrogen controlled endocrine versus exocrine secretion of prostaglandin F2a by the uterine endometrium. Prostaglandins 14:397-406 Lacroix MC, Kann G 1983 Discriminating analysis of in vitro prostaglandin release by myometrial and luminal sides of the ewe endometrium. Prostaglandins 25:853-862 Perry JS, Heap RB, Burton RD, Gadsby JE 1976 Endocrinology of the blastocyst and its role in the establishment of pregnancy. J Reprod Fertil 22 [Suppl l]:85-93 Bazer FW, Geisert RD, Thatcher WW, Roberts RM 1982 The establishment and maintenance of pregnancy. In: Cole DJA, Foxcroft GR (eds) Control of Pig Reproduction. Butterworth Co., London, pp 227-252 Fischer HE, Bazer FW, Fields MJ 1985 Steroid metabolism by endometrial and conceptus tissues during early pregnancy and pseudopregnancy in gilts. J Reprod Fertil 75:69-78 Frank M, Bazer FW, Thatcher WW, Wilcox JC 1977 A study of prostaglandin F2o as the luteolysin in swine. III. Effects of estradiol valerate on prostaglandin F, progestins, estrone and estradiol concentrations in the utero-ovarian vein of nonpregnant gilts. Prostaglandins 14:1183-1191 Bazer FW, Mejia AM, Clark WR, Vallet JL Protein, electrolytes and glucose in uterine flushings of pregnant and non-pregnant gilts. Proceedings of the 17th Annual Meeting of The Society for the Study of Reproduction, Laramie, WY, 1984, p 176 (Abstract) Geisert RD, Thatcher WW, Roberts RM, Bazer FW 1982 Establishment of pregnancy in the pig. III. Endometrial secretory response to estradiol valerate administered on day 11 of the estrous cycle. Biol Reprod 27:957-965 Riley SC, Poyser NL 1987 Prostaglandin production by the guineapig endometrium: is calcium necessary? J Endocrinol 113:463-471 Pressman BC 1976 Biological applications of ionophores. Annu Rev Biochem 451:501-530 Gemsa D, Seitz M, Kramer W, Grimm W, Till G, Resch K 1979 Ionophore A23187 raises cyclic AMP levels in macrophages by stimulating prostaglandin E formation. Exp Cell Res 188:55-62 Knapp HR, Oelz O, Roberts LJ, Sweetman BJ, Oates JA, Reed PW 1977 Ionophores stimulate prostaglandin and thromboxane biosynthesis. Proc Natl Acad Sci USA 74:4251-4255 Young KH, Kraeling RR, Bazer FW 1989 Effects of prolactin on conceptus survival and uterine secretory activity in pigs. J Reprod Fertil 86:713-722 Dehoff KH, Bazer FW, Collier RJ Ontogeny of prolactin receptors in porcine uterine endometrium during. Proceedings from the 4th International Prolactin Congress, Quebec, Canada, 1984, p 95 (Abstract) Young KH, Kraeling RR, Bazer FW 1990 Effect of pregnancy and exogenous ovarian steroids on endometrial prolactin receptor ontogeny and uterine secretory response in pigs. Biol Reprod, in press Kennedy TG 1985 Evidence for the involvement of prostaglandins throughout the decidual cell reaction in the rat. Biol Reprod 33:140-152 SAS User's Guide, Statistics 1982 SAS Institute Inc., Cary, NC Snedecor GW, Cochran WG 1980 Statistical Methods. Iowa State University Press, Ames, IA, pp 324-325
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 15 November 2015. at 22:41 For personal use only. No other uses without permission. . All rights reserved.
Vol. 127, No. 2 Printed in U.S.A.
Reorientation of Prostaglandin F Secretion by Calcium Ionophore, Estradiol, and Prolactin in Perifused Porcine Endometrium* TIMOTHY S. GROSSt, MARK A. MIRANDO, KATHLEEN H. YOUNG*, SASKIA BEERS, FULLER W. BAZER, AND WILLIAM W. THATCHER Dairy Science Department (T.S.G., W. W.T.) and Animal Science Department (M.A.M., K.H. Y., S.B., F.W.B.), University of Florida, Gainesville, Florida 32611
3), pregnant (n = 4), and EV-induced pseudopregnant (n = 4) gilts was collected on day 14 after estrus and perifused with BSA or PRL to both surfaces. Orientation of PGF secretion was initially endocrine in Exps 1 and 2 (P < 0.01). In Exp 1, calcium ionophore shifted the orientation of PG secretion from endocrine to exocrine (P < 0.01). In Exp 2, neither EV in vivo nor PRL in vitro alone altered the orientation of PGF secretion. However, EV and PRL interacted to reorient secretion of PGF from endocrine to exocrine at 6 h (P < 0.01) and 12 h (P < 0.01) after EV. In Exp 3, the orientation of PGF secretion was endocrine in cyclic gilts and exocrine in pregnant and pseudopregnant gilts (status x side; P < 0.05). PRL did not alter the orientation of PGF secretion regardless of reproductive status. These results suggest that the reorientation of endometrial PG secretion in pigs during the establishment of pregnancy involves interactive effects of estrogens and PRL, possibly through increased calcium cycling across the uterine epithelium. (Endocrinology 127: 637642, 1990)
ABSTRACT. Establishment of pregnancy in pigs requires a shift in endometrial prostaglandin (PG) F secretion from an endocrine (toward the myometrium and uterine vasculature) to an exocrine (toward the uterine lumen) orientation. Three experiments utilized bilateral endometrial perifusion devices for separate perifusion of myometrial and luminal surfaces to determine whether promoting calcium cycling across the luminal epithelial surface, which may be induced by interactive effects of estradiol and PRL, is involved in the reorientation of PGF secretion during establishment of pregnancy in pigs. In Exp 1, endometrium from cyclic gilts (n = 7) on day 14 after estrus was perifused with either saline (control) or calcium ionophore A23187 added to luminal surface perifusion buffer. In Exp 2, cyclic gilts (n = 7) at day 11 after estrus received an im injection of estradiol valerate (EV) after unilateral hysterectomy and the remaining EV-stimulated uterine horn removed at 6 h (n = 3) or 12 h (n = 4) after EV. Both surfaces of these endometrial samples were perifused with buffer containing either 250 ng/ml BSA or porcine PRL. In Exp 3, endometrium from cyclic (n =
E
STABLISHMENT of pregnancy involves maintenance of luteal function due to suppression or alteration of uterine luteolytic mechanisms (1). In cattle (2) and sheep (3), the antiluteolytic mechanism during early pregnancy involves an attenuation of endometrial responsiveness to stimulators of prostaglandin (PG) biosynthesis {e.g. oxytocin). However, porcine endometrium is capable of responding to oxytocin with increased PG secretion regardless of reproductive status (4). Bazer and
Received March 31,1990. Address requests for reprints to: Dr. F. W. Bazer, Animal Science Department, Building 459, Shealy Drive, University of Florida, Gainesville, Florida 32611. * Part of these studies was presented at the Annual Conference of the Society for the Study of Fertility, Edinburgh, Scotland, 1988 (Abstract 106). This work was supported in part by United States Department of Agriculture Grants 86-CRCR-1-2106 and 85-CRCR-l1871 and is published as Journal series no. 9675 of the Florida Agricultural Experimental Station. t Current address: Henry Doorly Zoo, Omaha, Nebraska 68107. X Current address: Department of Biochemistry, Molecular and Cellular Biology, Northwestern University, Evanston, Illinois 60208.
Thatcher (5) proposed that a reorientation of endometrial PGF release into the uterine lumen (exocrine mode) occurs at the onset of pregnancy in pigs. Using a perifusion system developed by Lacroix and Kann (6), Gross et al. (4) demonstrated that in vitro secretion of PGF was in an endocrine direction, toward the myometrium and the utero-ovarian vasculature, during the estrous cycle. However, during early pregnancy, secretion of PGF was exocrine, toward the uterine lumen. Such a reorientation in PG secretion could prevent release of PGF into the uterine vasculature and allow corpus luteum maintenance during early pregnancy in the pig as proposed by Bazer and Thatcher (5). This reorientation in endometrial secretion of PGF occurs between days 10 and 12 (4) of gestation when estrogen secretion is initiated by pig blastocysts (7-9). Pseudopregnancy can be induced in pigs by administration of estrogens during days 11 through 15 of the estrous cycle (10). Conceptus products, possibly estrogens, alter endometrial calcium dissipation and possibly calcium
637
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 15 November 2015. at 22:41 For personal use only. No other uses without permission. . All rights reserved.
638
REORIENTATION OF ENDOMETRIAL PG SECRETION
flux (11). Associated events might allow PGF to be sequestered in the uterine lumen so that it is less available to exert a luteolytic effect (5, 11). In addition, initiation of secretion of estrogens by pig blastocysts is associated with an acute release of calcium into the uterine lumen on days 11 and 12 of the estrous cycle, and then its reuptake by endometrial and/or conceptus tissues (11, 12). PG secretion by endometrial explants from guinea pigs (13) is dependent upon the presence of extracellular calcium. Agents such as calcium ionophore (Cal) A23187 increase calcium cycling across cell membranes (14) and are capable of altering PG biosynthesis (15, 16). Therefore, Cal might increase calcium flux or calcium uptake by endometrium and thereby alter endometrial secretion of PGF. PRL receptors have been detected in pig endometrial membranes, and a decrease in electrolyte release into the uterine lumen of hypoprolactinemic pigs has been reported (17). Estradiol induces endometrial PRL receptors in the pig (18, 19) within 6 h after treatment. Therefore, the possibility of an interaction between estrogen and PRL for the induction of calcium cycling across the endometrial epithelium exists. Increased calcium cycling might then alter the orientation of endometrial secretion of PGF similar to that observed during early pregnancy in pigs. The present study determined whether Cal would alter the orientation of secretion of PGF by perifused endometrium from cyclic gilts. In addition, perifused endometrium from cyclic gilts was used to determine interactions between estrogen and PRL which could change the direction of endometrial secretion of PGF from an endocrine to an exocrine direction.
Materials and Methods Materials Radioisotope [5,6)8)ll)12,14,15-3H]PGF2a (specific activity: 180 Ci/mmol) was purchased from Amersham Corporation (Arlington Heights, IL). Antiserum to PGF2a was provided courtesy of Dr. T. G. Kennedy (University of Western Ontario, Ontario, Canada). Calcium ionophore A23187 was purchased from Sigma Chemical Company (St. Louis, MO). PRL (USDABl) was supplied courtesy of Dr. Doug Bolt (United States Department of Agriculture, Agriculture Research Service, Beltsville, MD). Animals Sexually mature cross-bred (Duroc x Yorkshire x Hampshire) gilts were observed for estrous behavior in the presence of a boar at 0730 h each day. Day of onset of estrus was designated day 0 of the cycle. All gilts were assigned randomly to cyclic (n = 17; not bred), pregnant (n = 4, bred to boar at 0, 10, and 24 h after estrus), or pseudopregnant (n = 4) reproductive statuses. Pseudopregnancy was induced by daily sc injec-
Endo• 1990 Vol 127-No 2
tions of estradiol valerate (5 mg in 1 ml corn oil) on days 11 through 13 of the estrous cycle. Pregnant and cyclic gilts for Exp 3 were treated similarly with corn oil. Exp 1: effect of Cal on the orientation of secretion of PGF by perifused endometrium from cyclic gilts Gilts (cyclic; n = 7) were hysterectomized and endometrial tissue obtained on day 14 of the estrous cycle. Endometrium was dissected from myometrial tissue and placed into sterile Krebs-Ringer Bicarbonate solution (KRB) which was gassed with O2-CO2 (95:5), washed twice with fresh KRB and dissected into two pieces of approximately 4 cm2 each as described previously (4). Each endometrial sample (two per gilt) was placed into a bilateral perifusion device which allowed separation of myometrial and luminal surfaces so that each was perifused in a 0.5 ml hemichamber (4, 6). Tissue was stretched and anchored into each device with the aid of four needles.
Each chamber, submerged in a 39 C water bath, was attached to dual 60 ml syringes containing freshly gassed KRB and perifused at a rate of 3 ml/10 min for 2.5 h using a Harvard Infusion Pump System (Harvard Apparatus Co., Dover, MA). Fractions were collected every 10 min (3 ml) using a fraction collector (Gilson Electronics Inc., Middleton, WI) for a total of 15 fractions. Treatments were administered to the luminal surface during fractions 4-12. Treatments were: 1) control; and 2) Cal A23187 (40 /IM) to stimulate calcium flux. Fractions were stored at —20 C until analyzed for concentrations of PGF. Exp 2: interactive effect of estrogen and PRL on the orientation of PGF secretion by perifused endometrium from cyclic gilts Three gilts (cyclic) were unilaterally hysterectomized on day 11 after estrus and, immediately thereafter, received an im injection of estradiol valerate (5 mg in 1 ml corn oil). The remaining uterine horn was removed at 6 h after the estradiol valerate injection. Endometrium from each horn was collected and placed into two perifusion chambers as described in Exp 1. Endometrium was perifused with KRB for 2.5 h and fractions collected every 10 min. Treatment with either BSA or PRL (250 ng/ml) was administered in perifusate to both surfaces (two chambers per uterine horn per animal) during fractions 4-12. Four additional gilts (cyclic), at day 11 after estrus, received an im injection of estradiol valerate (5 mg in 1 ml corn oil) immediately after unilateral hysterectomy and the remaining uterine horn was removed at 12 h after estradiol valerate injection. Endometrium from each horn was collected and placed into two perifusion chambers as described in Exp 1. Endometrium was perifused with KRB for 3 h and fractions collected every 10 min. Treatment with either BSA or PRL (250 ng/ml) was administered in perifusate to both surfaces (two chambers per uterine horn per animal) during fractions 4-15. Fractions were stored at —20 C until analyzed for concentration of PGF. Exp 3: effect of PRL on the orientation of PGF secretion by perifused endometrium from cyclic, pregnant, and estrogeninduced pseudopregnant gilts Cyclic (n = 3, not bred), pregnant (n = 4, bred to boar at 0, 10, and 24 h after estrus), and pseudopregnant (n = 4) gilts
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 15 November 2015. at 22:41 For personal use only. No other uses without permission. . All rights reserved.
REORIENTATION OF ENDOMETRIAL PG SECRETION were hysterectomized on day 14 and endometrium placed into two perifusion chambers for each gilt as described in Exp 1. Endometrium was perifused with KRB for 2.5 h, fractions collected every 10 min, and treatments of either BSA or PRL (250 ng/ml) administered to both surfaces during fractions 412. Fractions were stored at —20 C until analyzed for concentrations of PGF.
RIAofPGF Fractions were analyzed for PGF using a direct RIA procedure (4) with an antibody characterized by Kennedy (20) and [3H]PGF2a (described above). Standard curves were prepared in KRB with known amounts of radioinert PGF2a (0, 10, 25, 50, 100, 250, 500, 1000, 2500, and 5000 pg). An antiserum dilution of 1:5000 was used, and the minimum concentration per tube which was distinguishable from zero was 10 pg. Crossreactivities of the PGF antiserum with other PGs were 94% for PGFia, 2.4% for PGE2, and less than 0.1% for 13,14-dihydro15-keto-PGF2a, PGE1( and arachidonic acid. Unextracted KRBperifusate samples were assayed for PGF in duplicate using 0.3 ml aliquots. A pooled KRB-perifusate sample (~10 ng PGF/ ml) was assayed serially in 0.01, 0.025, 0.05, 0.1, 0.2, and 0.3 ml volumes (final volume of 0.3 ml with blank KRB). This inhibition curve was parallel to the standard curve, with the test for homogeneity of regression indicating that the curves did not differ. Further characterization of the assay involved measurement of known amounts (10, 25, 50,100, 250, 500,1000, and 2500 pg) of PGF in KRB-perifusate [Y = -5.7 + 1.03X; Y = amount of PGF measured (picograms/0.3 ml); X = amount of PGF added (picograms/0.3 ml); R2 = 0.947]. Interassay and intraassay coefficients of variation were 12.0% and 13.1%, respectively. Statistical analyses Data were analyzed statistically by least squares analysis of variance using the General Linear Models procedure of the Statistical Analysis System (21). PG secretion rates (PGF) were analyzed for a split-plot design with sample periods (time) as the subplot effect. For Exp 1, data were analyzed using the model components of gilt, treatment, side, time, and all interactions. For Exp 2, data were analyzed using the model components of gilt, PRL, estrogen, side, time, and all interactions. For Exp 3, data were analyzed using the model components of reproductive status, gilt (nested within status), PRL, side, time, and all interactions. Effects of treatment, side, and reproductive status were examined over time for the entire perifusion period in each experiment. Tests of hypotheses were made using appropriate error terms according to expectations of the mean squares (22) with gilt considered to be random and all other effects considered to be fixed.
Results Expl This experiment determined whether endometrial secretion of PGF from the myometrial and/or luminal surfaces was affected by treatment with Cal. Endome-
639
trium from cyclic gilts at day 14 after estrus secreted PGF in primarily an endocrine orientation (Table 1); secretion of PGF was 80% greater (P < 0.01) from the myometrial than luminal surface during the pretreatment period. Interactions of treatment X side (P < 0.01) and treatment X side X time (P < 0.01) indicated that Cal A23187 (stimulates calcium flux) reoriented secretion of PGF more toward an exocrine direction within 1 h of treatment (Table 1) such that only 45% of the total PGF was secreted from the myometrial surface during the final 30 min of treatment (fraction 10-12) compared with 65% from the myometrial surface of control endometrium. Cal primarily increased (56%) secretion of PGF from the luminal surface, but also decreased (13%) secretion from the myometrial surface. However, when treatment with Cal was terminated, rates of secretion of PGF returned to pretreatment levels and primarily an endocrine orientation (Table 1). Cal did not significantly alter overall rate of PGF secretion. Exp 2 This experiment determined whether an interaction between estradiol valerate and PRL affects secretion of PGF from perifused endometrium. Endometrium from Day 11 of the estrous cycle secreted PGF in primarily an endocrine orientation (P < 0.01; Tables 2 and 3) which was 81% (Table 3) to 118% (Table 2) greater from the myometrial than luminal surface during the pretreatment period (treatment with PRL). Interactions of estradiol x PRL (P < 0.01) and estradiol x PRL x side (P < 0.05) were detected for tissue obtained at 6 h after estradiol treatment (Table 2) and estradiol X PRL X side X time (P < 0.01) for tissue obtained 12 h after estradiol treatment (Table 3). Treatment with PRL did not alter secretion of PGF by perifused endometrium obtained before estradiol valerate treatment (i.e. at 0 h; Tables 2 and 3). However, for endometrium taken from gilts at 6 h (Table 2) and 12 h (Table 3) after estradiol valerate treatment and removal of the first uterine horn, PRL increased secretion of PGF from the luminal surface by 27% (Table 3) to 46% (Table 2) while decreasing secretion from the myometrial surface by 17% (Table 3) to 18% (Table 2). These changes in PGF secretion, in response to an interaction between estradiol valerate and PRL, reoriented PGF secretion from an endocrine to an exocrine direction within 1 h of in vitro treatment such that only 46% of the total PGF was secreted from the myometrial surface during the final 30 min of treatment with PRL (Table 2, fractions 10-12 and Table 3, fractions 13-15), compared with 69% from the myometrial surface of BSA-treated endometrium. Treatment with estradiol in vivo or PRL in vitro did not significantly
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 15 November 2015. at 22:41 For personal use only. No other uses without permission. . All rights reserved.
640
REORIENTATION OF ENDOMETRIAL PG SECRETION
Endo-ioao Voll27«No2
TABLE 1. Effect of Cal on endometrial secretion rates of PGF (pg/10 min) by luminal and myometrial surfaces at day 14 after estrus for Exp 1 Fraction no. Treatment period
Surface
Treatment
Luminal Myometrial Luminal Myometrial
Control Cal, A23187
1-3
4-6
7-9
10-12
13-15
399 671 361 700
378 790 504 664
396 815 547 620
382 714 643 546
405 797 474 673
Endometrium was perifused with KRB (3 ml/10 min) for 2.5 h, and fractions were collected every 10 min (15 fractions total). Treatments (control and Cal A23187) were administered to the luminal surface during fractions 4-12. There were effects of side (P < 0.01), treatment x side (P < 0.01), and treatment x side x time (P < 0.01). Pooled SEM = 35.2, and each value represents the mean of three samples obtained from each of seven gilts. TABLE 2. Endometrial secretion rates of PGF (pg/10 min) by luminal and myometrial surfaces on day 11 after estrus for Exp 2 at 0 and 6 h after estrogen treatment in vivo Hours after estrogen
Treatment
0
BSA
0
PRL
6
BSA
6
PRL
Fraction no. Treatment period
Surface
Luminal Myometrial Luminal Myometrial Luminal Myometrial Luminal Myometrial
1-3
4-6
7-9
10-12
13-15
185 455 228 445 176 403 177 386
180 473 232 396 204
199 529 182 400 183 423 246 341
190 427 410 186 483 327
185 431 200 498 187 565 308
274
403
428 200 331
174
Endometrium was perifused with KRB (3 ml/10 min) for 2.5 h, and fractions were collected every 10 min (15 fractions total). Treatments (BSA and PRL) were administered to luminal and myometrial surfaces during fractions 4-12. There were effects of side (P < 0.01), estrogen x PRL (P < 0.01), and estrogen x PRL x side (P < 0.05). Pooled SEM = 18.6, and each value represents the mean of three samples obtained from each of three gilts. TABLE 3. Endometrial secretion rates of PGF (pg/10 min) by luminal and myometrial surfaces on day 11 after estrus for Exp 2 at 0 and 12 h after estrogen treatment in vivo
Hours after estrogen
Treatment
0
BSA
0
PRL
12
BSA
12
PRL
Fraction no. Treatment period
Surface Luminal Myometrial Luminal Myometrial Luminal Myometrial Luminal Myometrial
1-3
4-6
7-9
10-12
13-15
16-18
297 646 371 563 323 622 332 704
329 637 357 625 339 583 329 729
341 632 345 554 296 586 381 632
309 617 349 585 282 530 439 525
327 600 377 582 301 609 535 463
319 636 341 503 306 528 427 645
Endometrium was perifused with KRB (3 ml/10 min) for 3 h, and fractions were collected every 10 min (18 fractions total). Treatments (BSA and PRL) were administered to luminal and myometrial surfaces during fractions 4-15. There were effects of side (P < 0.01) and estrogen x PRL x side x time (P < 0.01). Pooled SEM = 51.5, and each value represents the mean of three samples obtained from each of four gilts.
alter overall rate of PGF secretion.
This experiment determined whether PRL could enhance the endocrine to exocrine reorientation of PGF secretion that occurs during early pregnancy and estro-
gen-induced pseudopregnancy. Endometrium from cyclic gilts at day 14 after estrus secreted PGF primarily in an endocrine orientation, whereas endometrium from pregnant and pseudopregnant gilts at day 14 after estrus secreted PGF primarily in an exocrine direction (Table 4; status X side interaction, P < 0.05). Endometrium from cyclic gilts secreted 63% of the PGF from the myometrial surface and 37% from the luminal surface
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 15 November 2015. at 22:41 For personal use only. No other uses without permission. . All rights reserved.
REORIENTATION OF ENDOMETRIAL PG SECRETION TABLE 4. Endometrial secretion rates of PGF (pg/10 min) by luminal and myometrial surfaces for cyclic, pregnant, and estrogen-induced pseudopregnant gilts at day 14 after estrus for Exp 3
Status Cyclic
Treatment
Surface
BSA
Luminal Myometrial Luminal Myometrial Luminal Myometrial Luminal Myometrial Luminal Myometrial Luminal Myometrial
PRL Pregnant
BSA PRL
Pseudopregnant
BSA PRL
Fraction no. Treatment period 1-3
4-12
1041 1913 1108 1740
606
813
1720 1061 1825 1460
2232 1275 2261 1245
936 640
1866 1400 732 448 709 506
13-15
904
816
2513 1754 1106
2216 1656 1379
673
557
1165
1209
632
913
Endometrium was perifused with KRB (3 ml/10 min) for 2.5 h, and fractions were collected every 10 min (15 fractions total). Treatments (BSA and PRL) were administered to luminal and myometrial surfaces during fractions 4-12. There was an effect of status X side (P < 0.05) but no treatment effect regardless of reproductive status. Pooled SEM = 243.4, and each value represents the mean of three, nine, and three samples (for fraction numbers 1-3, 4-12, and 13-15, respectively) obtained from each of three cyclic, four pregnant, and four pseudopregnant gilts.
during the pretreatment period, whereas endometrium from pregnant and pseudopregnant gilts secreted only 42 and 40% (respectively) of the PGF the myometrial surface and 58 and 60% (respectively) from the luminal surface during the pretreatment period. PRL did not significantly alter PGF secretion from endometrium regardless of reproductive status, nor did PRL enhance the endocrine to exocrine reorientation of PGF secretion of pregnant and pseudopregnant gilts. Mean PGF secretion also did not differ significantly among reproductive statuses. Discussion Previous results (4) demonstrated that secretion of PGF during the estrous cycle in gilts is primarily endocrine in direction (toward the myometrium and uterine vasculature), whereas secretion of PGF during early pregnancy is primarily exocrine in direction (toward the uterine lumen). Results from the present study confirm this orientation in direction of endometrial PGF secretion in cyclic and pregnant gilts at day 14 after estrus. In addition, these data indicate that estrogens, possibly of conceptus origin, are involved in the establishment of pregnancy in pigs by reorienting endometrial PGF secretion into the uterine lumen. Estrogens appear to initiate a calcium surge (12) which alters endometrial function. Associated events allow PGF to be sequestered in the uterine lumen so that it is
641
unavailable to exert a luteolytic effect. In addition, initiation of secretion of estrogens by pig blastocysts is associated with an acute release of calcium into the uterine lumen on days 11 and 12 of the estrous cycle, and then its reuptake by endometrial and/or conceptus tissues during early pregnancy (11,12). Secretion by PGF by endometrial explants from guinea pigs (13) and cows (Danet-Desnoyers, G., T. S. Gross, and W. W. Thatcher, unpublished results) is dependent upon the presence of extra-cellular calcium. Agents such as Cal A23187 increase calcium cycling across cell membranes (14) and, therefore, are capable of altering PG biosynthesis (15, 16). The present results demonstrate a reorientation of secretion of PGF by perifused endometrium from cyclic gilts when treated with Cal and indicate a possible role of calcium cycling across the endometrial epithelium in this reorientation of endometrial secretion of PGF during early pregnancy. Cal has been shown to primarily increase intracellular calcium (14); however, Cal has also been shown to directly affect cell membranes and subsequent PG synthesis in the absence of calcium (15). Therefore, the effects of Cal on PG secretion observed in the current study might be direct effects rather than mediated through increased calcium flux. Additional studies with other factors that alter calcium uptake/flux are necessary to demonstrate the mechanism by which Cal alters PG secretion. Preliminary studies (Gross, T. S., and F. W. Bazer, unpublished results) do not indicate any direct alteration in PG secretion by perifused endometrium from gilts in response to Verapamil (blocks calcium uptake), Amiloride (inhibits sodium transport), Trimethoxybenzoic acid (calcium antagonist), or KC1. Additional determinations in conjunction with Cal might indicate the possible mechanism by which Cal alters endometrial PG secretion. Estrogens also induce endometrial PRL receptors in the pig (18, 19). PRL receptors have been demonstrated for endometrial membranes from pigs, and a decrease in movement of electrolytes into the uterine lumen of hypoprolactinemic pigs has been detected (17). Therefore, an interaction between estrogen and PRL may influence induction of calcium cycling across the endometrial epithelium. Indeed, PRL increased secretion of PGF from the luminal surface while decreasing secretion from the myometrial surface of endometrium collected at 6 or 12 h after estradiol valerate treatment and removal of the first uterine horn on day 11 of the estrous cycle. An effect on uterine function of the remaining horn 6 to 12 h after unilateral hysterectomy has not been detected (12) and, therefore, the responses observed after PRL treatment in the present study are believed to be due to prior exposure to estradiol valerate rather than to the previous surgical procedure. These results indicate an
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 15 November 2015. at 22:41 For personal use only. No other uses without permission. . All rights reserved.
642
REORIENTATION OF ENDOMETRIAL PG SECRETION
interactive effect of estrogen and PRL, which results in a partial reorientation in the direction of secretion of PGF. Such an interaction between estrogen and PRL might be involved in the reorientation of endometrial secretion of PGF that normally occurs during early pregnancy in pigs. In contrast, the orientation and magnitude of PGF secretion by endometrium collected on day 14 after estrus was not altered by PRL treatment regardless of reproductive status (cyclic, pregnant, or estrogen-induced pseudopregnant). These results indicate that endometrium is no longer capable of responding to PRL once PGF secretion has reoriented from an endocrine to an exocrine direction which occurs between days 10 and 12 of pregnancy and pseudopregnancy. In addition, these results indicate that effects of PRL may be limited to before day 14 after estrus if involved in the establishment of pregnancy. Indeed, endometrial PGF secretion has been shown to begin to reorient from an endocrine to exocrine direction by day 12 of pregnancy (4). Therefore, an ability to respond to PRL at day 11 after estrus (Exp 2), but not at day 14 (Exp 3) might indicate an involvement of PRL along with estrogen in the establishment of pregnancy. In conclusion, estrogens (from the conceptuses or other sources) may act on the endometrium to induce PRL receptors which then allow PRL to bind to its receptor and stimulate an increase in calcium cycling across the endometrial epithelium. This calcium cycling may be associated with the reorientation of direction of secretion of PGF from an endocrine (toward the uterine vasculature) to an exocrine (toward the uterine lumen) direction. Since the reorientation of secretion of PGF from an endocrine to an exocrine direction occurs in gilts induced into pseudopregnancy by exogenous estradiol valerate (4), there is no basis to assume that a secretory product of the conceptus other than estrogen is required to induce this phenomenon.
3. 4.
5.
6. 7. 8.
9. 10.
11.
12.
13. 14. 15. 16. 17. 18.
Acknowledgment The technical assistance of Ms. M. F. V. Fliss is greatly appreciated. 19.
References 20. 1. Thatcher WW, Hansen PJ, Gross TS, Helmer SD, Plante C, Bazer FW 1989 Antiluteolytic effects of bovine trophoblast protein-one. J Reprod Fertil [Suppl]37:91-99 2. Gross TS, Thatcher WW, Lacroix MC, Hansen PJ 1988 Prostaglandin secretion by perifused bovine endometrium: secretion to-
21. 22.
Endo • 1990 Voll27-No2
wards the myometrial and luminal sides at day 17 after estrus as altered by pregnancy. Prostaglandins 35:343-357 Bazer FW, Vallet JL, Harney JP, Gross TS, Thatcher WW 1989 Comparative aspects of maternal recognition of pregnancy between sheep and pigs. J Reprod Fertil [Suppl] 37:85-89 Gross TS, Lacroix MC, Bazer FW, Thatcher WW, Harney JP 1988 Prostaglandin secretion by perifused porcine endometrium: further evidence for an endocrine versus exocrine secretion of prostaglandins. Prostaglandins 35:327-342 Bazer FW, Thatcher WW 1977 Theory of maternal recognition of pregnancy in swine based on oestrogen controlled endocrine versus exocrine secretion of prostaglandin F2a by the uterine endometrium. Prostaglandins 14:397-406 Lacroix MC, Kann G 1983 Discriminating analysis of in vitro prostaglandin release by myometrial and luminal sides of the ewe endometrium. Prostaglandins 25:853-862 Perry JS, Heap RB, Burton RD, Gadsby JE 1976 Endocrinology of the blastocyst and its role in the establishment of pregnancy. J Reprod Fertil 22 [Suppl l]:85-93 Bazer FW, Geisert RD, Thatcher WW, Roberts RM 1982 The establishment and maintenance of pregnancy. In: Cole DJA, Foxcroft GR (eds) Control of Pig Reproduction. Butterworth Co., London, pp 227-252 Fischer HE, Bazer FW, Fields MJ 1985 Steroid metabolism by endometrial and conceptus tissues during early pregnancy and pseudopregnancy in gilts. J Reprod Fertil 75:69-78 Frank M, Bazer FW, Thatcher WW, Wilcox JC 1977 A study of prostaglandin F2o as the luteolysin in swine. III. Effects of estradiol valerate on prostaglandin F, progestins, estrone and estradiol concentrations in the utero-ovarian vein of nonpregnant gilts. Prostaglandins 14:1183-1191 Bazer FW, Mejia AM, Clark WR, Vallet JL Protein, electrolytes and glucose in uterine flushings of pregnant and non-pregnant gilts. Proceedings of the 17th Annual Meeting of The Society for the Study of Reproduction, Laramie, WY, 1984, p 176 (Abstract) Geisert RD, Thatcher WW, Roberts RM, Bazer FW 1982 Establishment of pregnancy in the pig. III. Endometrial secretory response to estradiol valerate administered on day 11 of the estrous cycle. Biol Reprod 27:957-965 Riley SC, Poyser NL 1987 Prostaglandin production by the guineapig endometrium: is calcium necessary? J Endocrinol 113:463-471 Pressman BC 1976 Biological applications of ionophores. Annu Rev Biochem 451:501-530 Gemsa D, Seitz M, Kramer W, Grimm W, Till G, Resch K 1979 Ionophore A23187 raises cyclic AMP levels in macrophages by stimulating prostaglandin E formation. Exp Cell Res 188:55-62 Knapp HR, Oelz O, Roberts LJ, Sweetman BJ, Oates JA, Reed PW 1977 Ionophores stimulate prostaglandin and thromboxane biosynthesis. Proc Natl Acad Sci USA 74:4251-4255 Young KH, Kraeling RR, Bazer FW 1989 Effects of prolactin on conceptus survival and uterine secretory activity in pigs. J Reprod Fertil 86:713-722 Dehoff KH, Bazer FW, Collier RJ Ontogeny of prolactin receptors in porcine uterine endometrium during. Proceedings from the 4th International Prolactin Congress, Quebec, Canada, 1984, p 95 (Abstract) Young KH, Kraeling RR, Bazer FW 1990 Effect of pregnancy and exogenous ovarian steroids on endometrial prolactin receptor ontogeny and uterine secretory response in pigs. Biol Reprod, in press Kennedy TG 1985 Evidence for the involvement of prostaglandins throughout the decidual cell reaction in the rat. Biol Reprod 33:140-152 SAS User's Guide, Statistics 1982 SAS Institute Inc., Cary, NC Snedecor GW, Cochran WG 1980 Statistical Methods. Iowa State University Press, Ames, IA, pp 324-325
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 15 November 2015. at 22:41 For personal use only. No other uses without permission. . All rights reserved.
