Uterine secretory alterations coincident with embryonic mortality in the gilt after exogenous estrogen administration

Sixteen crossbred gilts were assigned randomly to receive either an i.m. injection of sesame oil (control) or estrogen (E), 5 mg of estradiol valerate, on d 9 and 10 of pregnancy. Gilts were unilaterally hysterectomized on either d 12 and 14 or 16 and 18. Uterine horns were flushed with 20 ml of .9% sterile NaCl solution to recover conceptus tissue. Conceptuses and endometrial explants were cultured for 24 h with 100 microCi [3H] leucine in 15 ml of minimum essential media. After dialysis, culture media were submitted to 2D-polyacrylamide gel electrophoresis and incorporated proteins were analyzed by fluorography. Normal, intact conceptus tissue was recovered from control gilts. Estrogen-treated gilts flushed on d 12 and 14 contained intact conceptuses; however, uteri from two gilts on d 16 and three on d 18 contained degenerating conceptus tissue. Comparison of endometrial polypeptides synthesized in vitro indicated an alteration in E-treated gilts on d 12 through 18. Although similar polypeptides were present, a band of polypeptides with a Mr of approximately 30,000 and pI from 7.9 to 8.9 and a larger, acidic polypeptide (Mr = 100,000, pI 3.5 to 5.0) were faint or absent in E-treated gilts. Conceptuses elongated normally in the altered uterine environment, but failed to survive past d 14 in E-treated gilts. Although loss of specific polypeptides in E-treated gilts coincides with conceptus death, their function in conceptus development or attachment is unknown.     

Interrelationships among conceptus size, uterine protein secretion, fetal erythropoiesis, and uterine capacity

The interrelationships among d-11 conceptus size, d-105 placental weight, placental efficiency (the ability of the placenta to support fetal growth and development), fetal erythropoiesis, and uterine capacity were examined in 1/2 Meishan, 1/2 White crossbred gilts that were unilaterally ovariohysterectomized at 90 to 100 d of age. In Exp. 1, gilts were mated after at least one normal estrous cycle and then slaughtered at 105 d of gestation and number of fetuses and CL, placental weights, fetal weights, hematocrits, fetal plasma iron, and fetal plasma folate were measured. In Exp. 2, gilts were mated and plasma progesterone was measured on d 2 and 3 of gestation. On d 11, the length of the remaining uterine horn was recorded and the uterine horn was flushed with minimal essential medium. Number of CL, conceptus number, conceptus diameters, and total uterine flush retinol-binding protein (tRBP), acid phosphatase (tAP), and folate-binding protein (tFBP) were measured. Gilts were mated again and slaughtered at 105 d of pregnancy and the same traits measured in Group 1 were recorded. Plasma progesterone concentrations on d 2 and 3 were correlated with average conceptus diameter on d 11 (r = 0.60, P < 0.01, for each day). In contrast, tRBP (r = 0.49, P < 0.01), tAP (r = 0.53, P < 0.01), and tFBP (r = 0.51, P < 0.01) in uterine flushings on d 11 were only correlated with d-3 plasma progesterone concentrations. No correlations between d-11 average conceptus diameter or d-11 uterine length with d-105 uterine capacity were observed. Uterine capacity was negatively correlated with placental weight, fetal weight and fetal hematocrit (r = -0.36, P < 0.01; r = -0.44, P < 0.01; r = -0.32, P < 0.01; respectively). Hematocrits were correlated with fetal plasma iron (r = 0.50, P < 0.01) and folates (r = 0.44, P < 0.01). Hematocrit, plasma iron, and plasma folate were each correlated with residual fetal weights after adjusting for placental weight (a measure of placental efficiency), and accounted for 11% of the variation in this trait. These data suggest that conceptus diameter and uterine protein secretion on d 11 may be influenced by the onset of progesterone secretion by the CL, but do not support an influence of conceptus growth during early pregnancy on uterine capacity. These results also suggest that reducing placental and fetal weights will likely result in increased uterine capacity.     

Dynamics of ovarian follicular development in cattle following hysterectomy and during early pregnancy

Three experiments were conducted to evaluate ovarian follicular dynamics and functional activity during pregnancy in cattle. In 11 pregnant Charolais cows of Experiment I, size of largest follicle, number of follicles and accumulated follicle size were reduced by day 27 of pregnancy on the ovary bearing the corpus luteum (CL) but not on the non-CL bearing ovary. In experiment II, local attenuation of ovarian follicular development on the CL bearing ovary of seven pregnant heifers was evident compared to the contralateral ovary without the CL. However, in four hysterectomized heifers, follicular development was sustained on both the CL- and non-CL bearing ovaries when CL maintenance was achieved without presence of the uterus or conceptus. In Experiment III, steroidogenic characteristics of the largest and second largest follicles at 17 d postestrus were evaluated for seven pregnant and six cyclic cattle. Follicle by physiological status interactions were detected for both aromatase activity of the follicle and follicular fluid concentrations of estradiol and progesterone. In cyclic cows, the largest follicle had appreciably more aromatase activity than did the second largest follicle; whereas, aromatase activity of the largest follicle from pregnant cows was less than that of cyclic cows. However, in pregnant cows the second largest follicle became the estrogen-active follicle, and this follicle occurred with a higher frequency on the ovary contralateral to the CL-bearing ovary. These changes in aromatase activity were reflected by parallel changes in estrogen concentrations of follicular fluid. The higher progesterone concentration in follicular fluid of the largest follicle in pregnant cows provided further confirmation of their atretic status. In conclusion, during early pregnancy the conceptus and/or uterus ipsilateral to the conceptus appear to secrete compounds which alter local follicular steroidogenic activity and attenuate subsequent follicular growth between 17 to 34 d of pregnancy on the CL-bearing ovary. This local mechanism acting within the ovary may contribute to the antiluteolytic effects of early pregnancy in cattle.     

Regression of induced corpora lutea by human chorionic gonadotropin in prepuberal gilts

The effect of daily injections of human chorionic gonadotropin (HCG) on luteal maintenance in hysterectomized prepuberal gilts induced to ovulate and in hysterectomized mature gilts was studied. Twenty-four pre-puberal gilts, 120 to 130 d of age, were induced to ovulate with 1,000 IU pregnant mare serum gonadotropin followed 72 h later with 500 IU HCG. Nine of the 24 prepuberal gilts (bred controls) were artificially inseminated on d 0 (d 0 = d after HCG). Mature gilts that had displayed one or more estrous cycles of 17 to 22 d were used (d 0 = onset of estrus). All gilts, except the bred controls, were totally hysterectomized on d 6 to 9 and their corpora lutea (CL) marked with charcoal. From d 10 through 29, eight prepuberal and 10 mature hysterectomized gilts received daily injections of 500 IU HCG in saline while seven prepuberal and eight mature hysterectomized gilts received daily injections of saline vehicle. Jugular blood samples were quantitated by radioimmunoassay for estrogen and 13,14-dihydro-15-keto prostaglandin F2 alpha (PGFM), a metabolite of prostaglandin F2 alpha. One bred control gilt was pregnant on d 30, indicating that the prepuberal gilts used in the experiment were prepuberal. All mature gilts and six of seven prepuberal gilts that received saline had maintained CL to d 30. Eight of 10 mature gilts that received HCG had maintained CL to d 30, while only two of eight (P less than .05) prepuberal gilts that received HCG maintained CL to d 30. All gilts receiving HCG had numerous follicles and accessory luteal structures.(ABSTRACT TRUNCATED AT 250 WORDS)     

Effect of number of conceptuses on maternal mammary development during pregnancy in the pig

An experiment was conducted to determine if manipulation of conceptus number could alter estrogen, progesterone or prolactin concentrations in gilts and thereby influence mammary development. Estradiol-valerate injections, oviduct ligation or no treatment were utilized to form three groups of gilts with zero (n=5), 4 to 7 (n=4), or 8 to 11 (n=4) conceptuses, respectively. Blood samples were taken throughout pregnancy, and at day 110 of gestation all gilts were slaughtered and total mastectomies were performed. Mammary glands were analyzed for wet weights, dry weights, RNA and DNA contents. Results indicated conceptuses stimulated mammary development since mammary glands from pseudopregnant gilts had only 22% as much DNA as mammae from pregnant gilts. However, mammary DNA was not different between gilts with 4 to 7 or 8 to 11 conceptuses. The relationship between conceptus number and mammary gland DNA in gilts was best described by the quadratic equation y = 577 + 583x − 38x², where y = total mammary DNA (mg) and x = conceptus number. Significant positive among-animal and within-group correlations between estrogen-sulfate or estrone concentrations in blood and maternal mammary DNA content suggest that estrogens may be the primary conceptus-derived mammogenic hormone in gilts.     

Adequacy of main uterine vein and the ovarian artery in the local venoarterial pathway for uterine-induced luteolysis in ewes.

The local uteroovarian pathway for uterine-induced luteolysis was studied in 34 ewes. Bilaterally ovulating ewes were used to study the role of the main uterine vein (uterine branch of uteroovarian vein) and unilaterally ovulating ewes were used to study the role of the ovarianartery. Surgical operation was done on day 7, 8, or 9 of diestrus and ewes were necropsied on day 20. In 4 bilaterally ovulating control ewes (unilaterally hysterectomized), mean weight of corpus luteum (CL) was less (P smaller than 0.05) on the side of the intact uterine horn (55 mg) than on the side of unilateral hysterectomy (548 mg). In 4 treated ewes (unilaterally hysterectomized and with surgical anastomosis of uterine vein from the intact side to the hysterectomized side), mean weight of CL on the hysterectomized side (153 mg) was less (P smaller than 0.05) than on the uterine-intact side (391 mg). The mean weight of CL on the hysterectomized side in the treated group was not significantly different from that of the uterine-intact side in the control group. Unilaterally ovulating ewes were randomized into 5 groups: (1) controls which were uterine intact, (2) controls in which a unilateral hysterectomy was done ipsilateral to CL, (3) unilateral hysterectomy done ipsilateral to CL and anastomosis of the ovarian branch of the ovarian artery on the uterine-intact side to the ovarian branch of the ovarian artery on the hysterectomized side, (4) controls in which unilateral hysterectomy was done contralateral to CL and the ovarian vascular pedicle on the uterine-intact side was isolated, and (5) unilateral hysterectomy done contralateral to CL, isolation of ovarian vascular pedicle on the uterine-intact side, and anastomosis of the ovarian branch of the ovarian artery on the unilaterally hysterectomized side to the ovarian branch of the ovarian artery on the uterine-intact side. Mean weight of CL was less (P smaller than 0.05) in groups 1 (56 mg), 3 (40 mg), and 4 (120 mg) than in groups 2 (408 mg) or 5 (473 mg). The mean weight of CL was not significantly different among groups 1, 3, and 4 or between groups 2 and 5. Results demonstrate that the local luteolytic effect of the uterus in ewes involves a venoarterial pathway and indicate that the main uterine vein is an adequate uterine outlet and the ovarian artery is an adequate ovarian inlet for the venoarterial pathway without the necessity of other possible concomitant local routes.     

Sialomucin Complex (Muc4) Expression in Porcine Endometrium During the Oestrous Cycle and Early Pregnancy

The non‐invasive type of implantation in the pig is characterized by the maintenance of a thick glycocalyx coating on the uterine epithelial surface microvilli. Present study investigated the alteration in the sialomucin complex (Muc4) expression during the oestrous cycle and early pregnancy in the pig. Endometrial tissue samples were immunostained with the primary antibody to the Muc4 transmembrane subunit ASGP‐2. Muc4 immunostaining increased in the surface and glandular epithelia between days 5 and 10 of oestrous cycle. Immunostaining continued to increase on day 12 with the greatest intensity of uterine Muc4 immunostaining detected on day 15 of the oestrous cycle and early pregnancy. Endometrial Muc4 expression in cyclic gilts decreased dramatically during early proestrous but continued to remain abundant in the surface and glandular epithelium of pregnant gilts during the period of conceptus attachment to the uterine surface.     

滋养层细胞囊泡与奶牛低品质胚胎共移植对受胎率的影响

将奶牛低品质胚胎与囊胚滋养层细胞囊泡（trophoblastic vesicles,TRV）共移植,TRV来源于体外受精培养 7 d、体内培养7 d囊胚的囊胚滋养层细胞。胚胎移植受体牛超声波妊娠诊断共进行2次,即发情的第26～43天和第38～73天。第1次妊娠诊断时试验组（与TRV共移植组）妊娠率（66.7%,16/24）显著高于对照组（34.5%,10/29）（P＜0.05）。在第1次和第2次妊娠诊断期间试验组有3头牛的胚胎死亡。试验组和对照组的分娩率分别为41.7%（10/24）和27.6%（8/29）,两者间无显著差异。两组间受体牛的妊娠期和后代出生重相似。     

Relationship between Prenatal Survival Rate at 70 days of Gestation and Morphometric Parameters of Vagina, Uterus and Placenta in Gilts

Swine uterine capacity affects litter size, and it could be used as a selection parameter of reproductive performance. Although there are some controversial results, evidences show that the catheter penetration length is positively correlated with litter size, and it could be used as a tool for predicting selection methods. The aim of this study was to determine whether there is any association between the prenatal survival rate and placental size at 70 days of gestation, the vaginal length [catheter penetration length during artificial insemination (AI)] and the uterine capacity in a homogeneous group of gilts. Sixty-six commercial-line gilts in pre-pubertal phase had their oestrus induced by hormonal treatment [600 UI of Equine Chorionic Gonadtrophin (eCG) i.m. and after a 72-h period 5 mg of luteinizing hormone (LH) i.m.], but only 40 gilts showed cyclicity after induction. The AI catheter penetration length was tested on these 40 gilts at the moment of AI using a calibrated AI catheter. Four gilts returned to oestrus and the other 36 were killed at around day 69 of pregnancy. The uterine length and weight showed a significant and positive correlation with the prenatal survival rate (p <0.05). The catheter penetration length was unable to predict the conceptus survival rate on 70 days of gestation; however, the uterine size influenced the survival rate positively. The mean placental area was positively correlated with the mean placental weight (p <0.0001), and both with the mean foetal weight (p <0.0001 and p <0.001, respectively). The analysis of the results obtained showed that neither did the catheter penetration length measurement during AI, nor the prenatal survival rate on day 70 of pregnancy predict the uterine capacity, but the uterine and placental size had a significant influence on the prenatal survival and foetus weight, respectively.     

Administration of Thyroxine Affects the Morphometric Parameters and VEGF Expression in the Uterus and Placenta and the Uterine Vascularization but does Not Affect Reproductive Parameters in Gilts During Early Gestation

The aim of this study was to evaluate the effects of thyroxine administration on morphometric parameters, expression of vascular endothelial growth factor (VEGF) and vascularization in the uterus and placenta and reproductive parameters in gilts at 70 days of gestation. At 150 days of age, i.e., before first heat, 20 gilts were randomly divided into two experimental groups: treated (n = 10) and control (n = 10). The treated group received a daily dose of 400 μg of l ‐thyroxine (T₄) in their diet until slaughter and the control group received only typical meals. Before artificial insemination, blood was collected to determine plasma total T₄. The gilts were inseminated in the second oestrus and slaughtered at 70 days of gestation. The foetal thyroid follicular epithelium height, number, size and weight of foetuses; foetal myogenesis, corpora lutea number, embryonic mortality rate, uterine weight, placental weight and placental fluid volume were measured. Histomorphometric and immunohistochemical analysis of uterus and placenta were determined. The averages of all variables were compared by the Student’s t‐test. The gilts treated with thyroxine showed significant increase of plasma total T₄. At 70 days of gestation, the heights of the trophoblastic epithelium, endometrial epithelium and endometrial gland epithelium were significantly higher in the group treated with T₄. The expression of cytoplasmatic and nuclear VEGF in trophoblastic cells and the number of blood vessels per field in endometrial stroma were significantly higher in the gilts treated with T₄. No other significant differences between groups were obtained with respect to other parameters (p > 0.05). We conclude that oral administration of T₄ up to 70 days of pregnancy in gilts affects the morphometric parameters, the expression of placental VEGF and the uterine vascularization but does not affect reproductive parameters in gilts during early gestation.     

Effect of number of conceptuses on maternal hormone concentrations in the pig

This study examined the effect of number of conceptuses on maternal concentrations and profiles of estrogen sulfate, estrone, estradiol-17 beta, progesterone and prolactin in gilts. Estradiol-valerate injections were used to induce pseudopregnancy (O conceptuses; n = 5) and oviduct ligation or no treatment were utilized to obtain pregnant gilts with 4 to 7 (n = 4), or 8 to 11 (n = 4) conceptuses, respectively. Blood samples were collected every 10 d from d 10 through 110 of pregnancy or pseudopregnancy. At 110 d after onset of estrus, all gilts were slaughtered and numbers and(or) weights of fetuses, corpora lutea, placentae and the empty uterus were determined. Concentrations of estrogen sulfate and estrone, but not progesterone or prolactin, were associated with fetal number, total fetal weight, total placental weight or empty uterine weight. In contrast, only progesterone was highly correlated with number of corpora lutea. Results suggest that most conjugated estrogen, estrone and estradiol were of fetal-placental origin, whereas little, if any, placental production of progesterone or prolactin occurred. Increases in estrogen sulfate and estrone concentrations were observed at gestation d 30 and from d 70 to 100. The latter increase coincides with previously established increases in the rate of maternal mammary development and fetal growth.     

Characterization of porcine uterine estrogen sulfotransferase

A quantitative trait locus (QTL) for uterine capacity is located on chromosome 8. Comparison of porcine and human genetic maps suggested that the estrogen sulfotransferase (STE) gene may be located near this region. The objectives of this study were to clone the full coding region for STE, compare endometrial STE gene expression between Meishan and White composite pigs during early pregnancy, and map the STE gene. We obtained a clone (1886 bp) containing the full coding region of STE by iterative screening of an expressed sequence tag library. Endometrial STE mRNA expression in White composite gilts was determined by Northern blotting on days 10, 13, and 15 of the estrous cycle; and on days 10, 13, 15, 20, 30, and 40 of pregnancy. STE mRNA expression was elevated (P<0.01) on days 20 and 30 of pregnancy compared to other days of the cycle or pregnancy. Endometrial STE mRNA expression during early pregnancy, determined using real-time RT-PCR, was elevated (P<0.01) on day 20 compared to day 15, decreased (P=0.02) between days 20 and 30, and decreased further (P<0.01) between days 30 and 40 in both Meishan and White composite pigs. Expression of STE mRNA was greater (P=0.01) in White composite pigs compared to Meishan pigs. Using a microsatellite from an STE containing BAC genomic clone, the STE gene was mapped to 65 centimorgans on chromosome 8. Because STE mRNA expression differs between Meishan and White composite pigs, the STE gene may be a candidate for the uterine capacity QTL.     

Reducing endogenous estrogen during development alters hormone production by porcine Leydig cells and seminiferous tubules

High levels of estrogen produced by boar testes and the presence of estrogen receptors in both interstitial and tubular compartments are consistent with a direct role for estrogen in regulation of testicular cell function. This study investigated the importance of estrogen on hormone production by Leydig cells and seminiferous tubules in the developing boar. Thirty-six 1-week-old littermate pairs of boars were treated weekly with vehicle or 0.1 mg/kg BW Letrozole, an aromatase inhibitor, until castration at 2, 3, 4, 5, 6, 7, or 8 months. Tissue was collected and Leydig cells and seminiferous tubules were isolated. In a separate study, five untreated boars (ages 1.5-4 months) were castrated and Letrozole was added in vitro to Leydig cell and seminiferous tubule cultures. Leydig cells were cultured for 24h with and without porcine LH. Media were assayed for estradiol (E(2)) and testosterone (T) concentrations by RIA. Seminiferous tubules were cultured for 4h with and without porcine FSH; media were assayed for E(2) and immunoreactive inhibin (INH). In vivo aromatase inhibition decreased basal E(2) and increased basal T production by cultured Leydig cells. Basal seminiferous tubule production of E(2) but not INH was reduced. Decreasing estrogen synthesis in vivo did not alter LH-induced Leydig cell E(2) production or FSH-induced seminiferous tubule INH production. INH production decreased with advancing age regardless of treatment. In conclusion, in vivo aromatase inhibition altered baseline steroid production by cultured Leydig cells and seminiferous tubules but had little effect on response to gonadotropins.     

The effect of breed and intrauterine crowding on fetal erythropoiesis on day 35 of gestation in swine

In a previous report, it was suggested that intrauterine crowding impaired fetal erythropoiesis and that fetal erythropoiesis was accelerated in Meishan pigs during early pregnancy. Because these conclusions were based on limited numbers of observations, the present experiment was undertaken to provide a more extensive investigation of these phenomena. Intact white crossbred gilts, unilaterally hysterectomized-ovariectomized (UHO) white crossbred gilts, and intact Meishan gilts (n = 13 to 16 per group) were mated after at least one estrous cycle of normal duration (17 to 23 d). Gilts were laparotomized at d 35 of pregnancy, the uterine horns were exteriorized and opened near each fetus, and a blood sample was collected from each fetus. The uterine horn was then surgically removed, and each fetus and placenta was weighed. All fetal blood samples were measured for hematocrit, red blood cell number, and hemoglobin. Erythropoietin and the percentages of nucleated cells and reticulocytes were also measured in blood samples from the largest and smallest living fetus in each litter. Fetal hematocrits were not affected by treatment. Blood cell counts were greater (P < 0.01) in fetuses of Meishan gilts than in White crossbred intact or UHO gilts. Hemoglobin was less (P < 0.01) in fetuses of Meishan gilts than in fetuses of White crossbred intact or UHO gilts. The percentages of nucleated (immature) cells and reticulocytes were both less (P < 0.01) in fetuses of Meishan intact gilts. Erythropoietin was also lower (P < 0.01) in fetuses of Meishan gilts. As observed previously, fetal weight was correlated (r = 0.38; P < 0.01) with blood hemoglobin concentration. These results confirm that fetal erythropoiesis in Meishan gilts is accelerated compared with White crossbred gilts. These results are consistent with the hypothesis that faster blood cell development could be beneficial to fetal survival in swine.     

Effect of total or partial uterus extirpation on uterus-projecting neurons in porcine inferior mesenteric ganglion. C. Changes in expression of apoptosis-associated (Bcl-2 and Bax) and regeneration-associated (GAP-43) proteins

The expression of Bcl-2 and Bax proteins was studied with immunohistochemistry, immuoblotting and RT-PCR in the uterine horn- and uterine cervix-projecting neurons of the inferior mesenteric ganglion (IMG) of the sexually immature gilts after partial or total hysterectomy. Additionally, the expression of regeneration-associated protein GAP-43 was studied in these neurons with immunohistochemistry. The uterus-projecting neurons were identified with retrograde fluorescent tracer Fast Blue (FB). The weak immunoreactivity to Bcl-2 and GAP-43 and moderately intense immunoreactivity to Bax was revealed in all FB+ (FB+) neurons of control and hysterectomized pigs. No difference in the intensity of immunostaining for Bcl-2, Bax and GAP-43 was found between control and hysterectomized gilts. Immunoblotting revealed the presence of Bcl-2 and Bax proteins in IMGs of control and hysterectomized animals and no difference in the band intensities between control and experimental groups was detected. RT-PCR detected weak induction of bcl-2 and bax only in the ganglia of animals which had undergone total hysterectomy. It was found that the axotomy of the uterus-projecting neurons located in the porcine IMG did not change the expression of the studied substances (Bcl-2, Bax and GAP-43) at protein level and only the induction of bcl-2 and bax at the level of RNA was visible.     

Serum prolactin response to thyrotropin releasing hormone in estrogen treated hypophysial stalk-transected gilts

Twelve crossbred gilts, 169 ± 3 days of age and 72.8 ± 3.4 kg body weight, were hypophysial stalk-transected (HST)1 or sham hypophysial stalk-transected (S-HST). Gilts were ovariectomized 6 days later and assigned to four treatments of 3 gilts each in a 2 × 2 factorial arrangement. One-half of the HST and S-HST gilts received 5 mg estradiolbenzoate (EB) or corn oil vehicle im at 0800 hr daily for 5 days beginning 64 ± 3 days after HST or S-HST. Blood was collected by jugular vein cannula at 0830 and 0900 hr the day after the last injection of EB or oil. Immediately after the 0900 hr sample, 200 μg thyrotropin releasing hormone (TRH) were injected (iv). Mean basal serum prolactin (PRL) concentration was similar for HST (10.3 ± 1.0 ng/ml) and S-HST (12.3 ± 1.7 ng/ml) gilts, however mean basal serum PRL concentration was greater (P<.05) for EB-treated gilts (13.7 ± 1.3 ng/ml) than for oil-treated gilts (8.8 ± .5 ng/ml). Mean serum PRL concentration of all gilts increased within 10 min and returned to approximately 20 ng/ml by 150 min after TRH. Maximum serum PRL concentrations at 10 min after TRH were greater (P<.01) for S-HST (255.9 ± 29.6 ng/ml) than HST gilts (83.4 ± 18.8 ng/ml), but were not different for EB (198.0 ± 50.6 ng/ml) and oil-treated gilts (141.4 ± 36.3 ng/ml). Area under the serum PRL response curve after TRH was greater (P<.005) for S-HST than HST gilts and for EB than oil-treated gilts (P<.05). These results do not eliminate the possible influence of estrogen on PRL secretion at the hypothalamus, but do indicate that estrogen directly stimulated the anterior pituitary gland to secrete PRL.