EFFECT OF ESTRADIOL ON HYPOTHALAMIC GnRH AND PITUITARY AND SERUM LH AND FSH IN OVARIECTOMIZED PIGS 1

Transcription

1 EFFECT OF ESTRADIOL ON HYPOTHALAMIC GnRH AND PITUITARY AND SERUM LH AND FSH IN OVARIECTOMIZED PIGS 1 N. M. Cox and J. H. Britt 2 North Carolina State University, Raleigh Summary Two experiments were conducted to measure pituitary gonadotropins, hypothalamic-gonadotropin releasing hormone (GnRH) and pituitary response to GnRH during periods when serum luteinizing hormone (LH) was suppressed by estradiol-17/~ (E2) in ovariectomized pigs. In the first experiment, 10 ovariectomized gilts were assigned to two groups of five each according to time of slaughter (24 or 36 h after injection). Within each group, gilts were given corn oil (n = 2) or 400 #g E2 (n = 3). Neither serum nor anterior pituitary (AP) concentrations of follicle-stimulating hormone (FSH) were affected by E2. Serum Lit was suppressed from 12 to 26 h after E2. Concentrations of LH in AP were unchanged at 24 h, but increased at 36 h after E2 injection. Concentrations of GnRH in medial basal hypothalamus (MBH), stalkmedian eminence (SME) and hypophyseal portal area (HPA) were lower at 24 h after E2 than in oil-treated gilts. At 36 h after E2, suppressive effects of E2 on LH in serum had subsided and concentrations of LH in AP and GnRH in MBH and SME were greater than in oil-treated controls. The observation that E2 suppressed LH in serum without a detectable suppression of LH in AP led to the hypothesis that E2 had caused the suppression of serum LH by suppression of GnRH release. In a second experiment, 12 ovariectomized gilts were assigned to receive corn oil (n = 4), 400/Jg E: (n = 4) or 400/~g E2 plus GnRH (1.5 #g/h; n = 4). Patterns of LH in sera of E2-treated ~nimals were similar to those in the first experiment, with serum LH in E2-treated gilts suppressed from 4 to 32 h after treatment. However, in gilts receiving GnRH in addition to Eu, serum LH concentrations during 20 to 32 h after treatment were intermediate between gilts receiving E2 alone and controls. Thus the pituitary of the pig is capable of responding to GnRH when LH is normally suppressed by Ez. These experiments provide two lines of evidence that suppression of serum LH by E2 is due at least in part to suppression of GnRH. These experiments also establish the hypothalamus as a site for negative feedback of E2 in the female pig. (Key Words: Luteinizing Hormone, Follicle Stimulating Hormone, Gonadotropin Releasing Hormone, Hypothalamus, Anterior Pituitary, Gilts. ) I ntroduction Several studies have implicated estrogen as 1 Paper no of the Journal Series of the North the primary ovarian hormone responsible for Carolina Agr. Res. Service, Raleigh. The use of trade names in this publication does not imply endorsement control of the preovulatory secretion of gonaby the North Carolina Agr. Res. Service of the pro- dotropins in swine. An increase in serum ducts named, nor criticism of similar ones not men- estrogens preceded the preovulatory surge of tioned. We thank Dr. G. D. Niswender for providing luteinizing hormone (LH) by about 48 h, both the porcine LH antiserum; Dr. V. L. Estergreen for during the estrous cycle (Henricks et al., 1972) donating the estrogen antiserum; Drs. R. J. Ryan and R. J. Whitley for supplying the porcine FSH antiserum and at the postweaning estrus (Stevenson et al., and purified porcine FSH (IA3-c2); Dr. L. E. Reichert, 1981). Administration of estrogen to intact pigs Jr., for providing purified porcine LH (LER 786-3) (Elsaesser and Foxcroft, 1978) or ovariectoand Drs. Ray Rippel, Abbott Laboratories, and M. D. mized pigs (Lantz and Zimmerman, 1972; Brown, Ceva Laboratories, for providing the GnRH. We acknowledge the excellent technical assistance of Stevenson et al., 1981)caused a rapid depression J. K. Huff. of serum LH followed by an increase between 2 Dept. of Anita. Sci. 48 and 60 h later. Serum follicle stimulating 901 JOURNAL OF ANIMAL SCIENCE, Vol. 55, No. 4, 1982

2 902 COX AND BRITT hormone (FSH) levels also increased in ovariectomized sows given estradiol after weaning (Stevenson et al., 1981). Amount of LH and FSH in the pituitary changed during the estrous cycle (Parlow et al., 1964) and around the postweaning estrus (Crighton and Lamming, 1969), but changes in pituitary concentrations of gonadotropins have not been related to serum levels of gonadotropins or steroids in the same pig. Little information is available on mechanisms by which estrogens influence gonadotropins in pigs. In other species, effects of estrogen on synthesis and secretion of gonadotropins have been attributed to direct effects on the anterior pituitary, to alterations in pituitary sensitivity to gonadotropin-releasing hormone (GnRH) and to modulation of hypothalamic GnRH. Estradiol caused an increase in intracellular content and release of LH in ovine (Huang and Miller, 1980) and bovine (Padmanabhan et al., 1978) pituitary cells in vitro. Estrogens also increased LH release in response to GnRH in vivo (Reeves et al., 1971; Zolman et al., 1974; Beck and Convey, 1977) and in vitro (Drouin et al., 1976; Padmanabhan et al., 1978; Huang and Miller, 1980) in several species other than the pig. Estradiol inhibited release of FSH by ovine and porcine pituitary cell cultures but caused synthesis and secretion of FSH in rat pituitary cell cultures (Miller and Wu, 1981). After ovariectomy in rats, decreases in hypothalamie GnRH and increases in serum LH were partially or completely reversed by administration of estrogen (Baram and Koch, 1977; Chert et al., 1977; Gross, 1980). Concentrations of GnRH in hypophyseal portal plasma increased during estrogen-induced LH surges in ovariectomized rats (Sarkar and Fink, 1979; Sherwood and Fink, 1980) and monkeys (Neill et al., 1977). To our knowledge there have been no reports relating measurements of endogenous GnRH to the steroid or gonadotropin environment in the female pig. Therefore, the present experiments were conducted to (1) determine if effects of exogenous estrogen on LH and FSH were accompanied by changes in endogenous GnRH and (2) investigate pituitary responsiveness to GnRH during a period when LH was suppressed by estradiol. Experimental Procedure Exp. 1. Ten sexually mature Duroc gilts weighing kg were bilaterally ovariecto- mized 40 d before the experiment. Beginning 1 wk before the study, gilts were maintained in individual pens (1.5 x 3.7 m) and fed a cornsoybean meal (IFN 4-O2-931 and ) diet supplemented with vitamins and minerals according to NRC (1979) guidelines. Gilts were exposed to natural daylight through windows and room lights were on throughout the experiment. A cannula was inserted, without anesthesia, into the anterior vena cava of each gilt about 4 h before the beginning of blood sampiing. Gilts were divided into two groups according to the time of slaughter: either 24 or 36 h after treatment. Three gilts in each group received (im) 400 #g estradiol-17/3 (E2) in 2 ml corn oil and two gilts in each group received (ira) 2 ml corn oil alone. Blood samples were taken immediately prior to injection and every 6 h until 12 h before slaughter, when the interval between samples was reduced to 2 h. The last sample was taken 4 h before slaughter (e.g., 20 or 32 h after treatment). Sampling was terminated at these times so gilts could be transported to the slaughter facility. Time of treatment in the 24-h group was delayed 12 h from that of the 36-h group so that all animals could be slaughtered at the same time. Within 30 min of slaughter, the pituitary gland and an area of brain encompassing the hypothalamus were excised and placed on solid CO2 before transfer to storage at -20 C for 2 wk. Each hypothalamic area was divided into the medial basal hypothalamus (MBH) and stalk-median eminence (SME). The MBH was defined as a block of tissue bounded rostrally by the optic chiasm, caudally by the mammilary body, laterally by the cerebral peduncles and dorsally by a cut 7 mm deep. The SME was easily detached (without cutting) from the hypthalamus. The anterior pituitary was divided into two parts, the hypophyseal portal area (HPA) and the remainder of the anterior pituitary (AP). The HPA, represented by a clearly discernible vascular area, was dissected from the AP. The MBH, SME and HPA were weighed and homogenized in 3 (SME and HPA) or 6 (MBH) ml 2 N acetic acid, according to the procedure of Estes et al. (1977). After centrifugation at 10,000 x g for 30 rain, 1 ml 5 N NaOH was added to the supernatants that were then frozen, lyophylized, reconstituted with double distilled H20 and refrozen until assayed for GnRH. Each AP was weighed and homogenized in gel-phosphate-buffered saline (PBS;.14 M

3 LH, FSH AND GnRH AFTER EsTRADIOL IN OVEX PIGS 903 sodium chloride,.01 M phosphate,.1% gelatin, ph 7.3) at a concentration of approximately 200 mg tissue/ml. The samples were centrifuged for 30 min at 3,000 g, and the supernatants were frozen and stored at -20 C before hormone assays. All pituitary and hypothalamic hormone concentrations were expressed for wet tissue. Serum samples were allowed to clot at 5 C for 24 h before serum was obtained by centrifugation and stored at -20 C before hormone assay. Serum and tissue LH, FSH and total uncow jugated estrogens were quantified by radioimmunoassay procedures described by Stevenson et at. (1981). Pituitary extracts were diluted 1:10,000 for LH and FSH assays. When extracts prepared in this fashion were assayed at volumes ranging from 50 to 200/al, curves plotted were parallel to standard curves. Radioimmunoassays for GnRH were conducted on undiluted tissue extracts using anti-gnrh 3 at a dilution of 1:80,000. Characteristics of this antiserum have been reported previously (Koch et al., 1973). Synthetic GnRI-I 4 was used for standards and was iodinated according to the procedure of Nett et al. (1973). Sensitivity of the assay, defined as the least amount that could be distinguished from the zero tube, was 25 pg GnRH. Inhibition curves prepared with each of the tissue extracts (MBH, SME and HPA) were parallel to standard curves. Radioimmunoassay of extracts of cerebral cortex and posterior pituitary gave values below the sensitivity of the assay. Recovery of GnRH added to tissue blocks before extraction or to extracts prior to assay were 99.2% and i04.1%, respectively. Recovery of 12SI-GnRH added to tissue blocks before extraction was 97.9%. Intra- and interassay coefficients of variation for the GnRH assay were 10 and 27%, respectively. Exp. 2. In the first experiment, concentrations,of LH in serum as well as concentrations of GnRH in MBH, SME and HPA were suppressed at 24 h after E2 injection. Therefore, a second experiment was designed to determine whether injections of GnRH would increase serum LH during the period of E2-induced suppression of LH. Because FSH concentrations were not affected by E2 in Exp. 1, FSH was amiles-yeda Laboratory, Rehovot, Israel. 4 Abbott Laboratories, Chicago. not measured in this experiment. Three Duroc and nine crossbred gilts ovariectomized 40 d previously and weighing kg were hensed and fed similarly to gilts in Exp. 1. One Duroc and three crossbred gilts were assigned to each of three treatments, so that average body weights were similar among treatments. Treatments were: 2 ml corn oil (oil), 400#g E2 in 2 ml corn oil (E2) and 400 /ag E2 with 1.5 /ag GnRH in physiological saline every hour for 72 h (E2-GnRH). Estradiol and oil were admin- istered as single im injections, and GnRH injections and blood sampling were via anterior vena cava cannulas inserted approximately 6 h before beginning the experiment. Serum samples were obtained immediately before beginning injections (0 h) and at 2-h intervals for 72 h. Statistical Analysis. For analysis of the effect of E2 on serum hormone levels in Exp. 1, data were grouped relative to the time of treatment with corn oil or E2 regardless of the time of slaughter. Data obtained at O, 6, 12 and 18 h after injection (the only sampling periods both 24 and 36 h groups had in common) were used in split-plot analysis of variance for repeated measurements within animals (Gill and Hafs, 1971). Independent variables in the statistical model were treatment (E2 or oil)~ gilt within treatment, hour and the treatment x hour interaction. Treatment differences were tested by the split error term, gilts within treatment. Similar models were used in Exp. 2 to analyze treatment effects. In some analyses, data from oibtreated gilts were deleted so the E2 and E2-GnRH treatments could be compared. Comparison of treatment" means for a particular sampling period was by t-test when variances were homogeneous and by approximate t-test when variance were heterogeneous (Sokal and Rohlf, 1969). One~way analysis of variance was used to determine effects of treatment on hormone concentrations in tissues taken at slaughter in Exp. 1 and on time to first increase in LH in Exp. 2 (Sokal and Rohlf, 1969). In Exp. 2 the time to first increase in LH was defined as the hour when LH concentration first reached a value >50% above the preceding nadir. Comparisons among treatment means were made by Duncan's multiple range test (Snedecor and Coehran, 1967). Results In the first experiment, total estrogens in

4 904 COX AND BRITT serum increased from basal values of less than 40 pg/ml to peak levels of pg/ml at 6 h after E2 treatment and were significantly greater than levels in oil-treated gilts (table 1). Estrogens were still elevated in E2-treated gilts 12 h after treatment. NO differences in total estrogens between E2- and oil-treated gilts were noted beyond 12 h after treatment. Serum FSH concentrations fluctuated between 2 and 14 ng/ml and no differences were detected between oil- and E2-treated gilts during any of the sampling periods (table 1). Analysis of variance indicated that the effect of E2 on serum FSH was not significant, but variation due to gilt within treatment was significant. In Exp. 1, mean serum LH concentrations during 12 through 26 h after treatment were lower (P<.05) in E2- than oil-treated gilts (table 1). Concentrations of LH in individual E2- treated gilts were greater than 1 ng/ml in only one instance from 12 to 28 h after treatment. Least-squares mean concentrations of LH and FSH in AP are depicted in figure 1. Data from oil-treated gilts, two each killed 24 or 36 h after corn oil, were combined. These control data were compared with data from gilts killed d -..ad200 ~ 100 im O. H LL k,-,oo ~ 100 rl 1 M M CONTROL HOURS AFTER E 2 Figure 1. Least-squares means for pituitary LH (LER 786-3) and FSH (IA3-c2) concentrations in gilts receiving corn oil (controls; n = 4) and gilts killed 24 h (n = 3) or 36 h (n = 3) after E 2 treatment (Exp. 1). Vertical lines represent SE. Pituitary LH at 36 h after E 2 was greater than pituitary LH both at 24 h after E 2 treatment and in control animals (P<.05). TABLE 1. SERUM CONCENTRATIONS OF TOTAL ESTROGENS, FSH AND LH IN OVARIECTOMIZED GILTS GIVEN ESTRADIOL (400 ~ug) OR CORN OIL (EXP. 1) Hours after Total estrogens FSH LH injection Oil E 2 Oil E 2 Oil E pg/ml ng/ml ab c c c c c.5.04 c 14 22, 50 d , , 3.2 c c 16 18, ~7, , 4.1 c.4.1 c c.3.1 c 20 43, , , 2.1 c c 24 47, , , 2.6 c.6.1 c 26 53, , , 2.2 c.5.1 c 28 46, , , , , , 1.9! , , , ames+n SE. bat O, 6, 12, 18 h n = 4 for oil- and n = 6 for E2-treated gilts. At all other periods n = 2 for oil and n = 3 for E 2 -treated gilts. CValues with common superscripts differ (t-test; P~.05). dwhere n = 2, individual hormone concentrations are given.

5 LH, FSH AND GnRH AFTER ESTRADIOL IN OVEX PIGS or 36 h after E2 treatment. Mean pituitary FSH concentration~ ranged from to 117 t 30/~g/g and were not affected by treatment with E2. Pituitary LH concentration at 36 h after E2 ( #g/g)was significantly higher than in both the oil-treated group ( /ag/g) and the group killed 24 h after E2 treatment ( /~g/g). Least-squares mean concentrations of GnRH in MBH, SME and HPA for the three treatment groups are illustrated in figure 2. Mean concentrations of GnRH in MBH were 10.3 t 1.6, and ng/g for control gilts and gilts killed 24 or 36 h after E2 treatment, respectively. Concentration of GnRH in MBH of gilts killed 36 h after E2 treatment was greater than in oil-treated gilts (P<.I) or gilts killed 24 h after E2 treatment (P<.03). Concen ~ trations of GnRH in SME paralleled those in MBH; gilts killed 36 h after E2 treatment had a greater mean concentration than the control gilts (P<.I) or the 24-h group (P<.03). Concentrations of GnRH in SME were , and 1, ng/g in the control, 24- and 36-h groups, respectively. In contrast to the other tissues, GnRH in HPA was lower (P<.05) at 36 h after E2 treatment than in oil-treated gilts. Similarly, GnRH in HPA at 24 h after E2 was lower (P<.05) than in controls. Concentrations of GnRH in HPA were , and ng/g in control, 24- and 36-h groups, respectively. In Exp. 2, the interaction between treatment and time after treatment for LH was significant (P<.O1, figure 3). From 4 to 32 h after treatment, LH was suppressed in gilts that had received E2 or E2-GnRH compared with oil-treated controls (P<.01). After 32 h, there were no differences in LH concentration among the three treatments. After 32 h, LH concentrations averaged (oil treatment), (E2 treatment) and ng/ml (E2-GnRH treatment). When oil-treated animals were omitted from analyses, there were treatment differences in pattern of LH response in time (hour X treatment interaction, P<.05) between gilts given E2 or E2-GnRH. Time to first increase in LH after initial suppression was less for E2-GnRH ( h) than for E2 ( h) treatment (P<.05). From 20 to 32 h after treatment, LH concentrations for gilts given E2-GnRH averaged ng/ml, approximately 40% of values for oil-treated animals. In contrast, LH concentration in gilts given only "I" 1000 X "l- MEDIAL BASAL J. 12 HYPOTHALAMUS STALKoM EDIAN EMINENCE 60O 400 ~'] 200 ~'~ 50 HYPOPHYSEAL PORTAL AREA CONTROL HOURS AFTER E 2 Figure 2. Least-squares means for GnRH concentrations in MBH, SME and HPA in gilts receiving corn oil (controls; n = 4) and gilts killed 24 (n = 3) or 36 h (n = 3) after E 2 treatment (Exp. 1). Vertical lines represent SE. In MBH and SME, GnRH concentrations at 36 h after E 2 treatment were greater than in control (P<.I) or 24-h groups (P<.03). In HPA, GnRH concentrations were greater in controls than in 24- or 36-h groups (P<.05). E2 averaged ng/ml from 20 to 32 h after treatment. Average concentrations of LH for each sampling period during 20 to 32 h after treatment for gilts given E2 and E2-GnRH are in table 2. Discussion No changes in pituitary or serum concentrations of FSH were detected following the pulse of E2 in this study. Previously we observed that serum FSH increased after sequential injections of E2 in ovariectomized sows (Stevenson et al., 1981), but in that study, FSH did not increase until after a preovulatory-like surge of LH had occurred. In the present study, no such surge in LH occurred. Moreover, in the present experiment the surge of total free estrogens in serum after E2 was of shorter duration (12 h compared with approximately 24 h) and of severalfold greater concentration (greater than

6 906 COX AND BRITT OIL 2.4 ~... E2 2.2 ~! : :E2-GnRH i/,.o i i ; o 1.8 a :~ ' ; *. ~ = +',!',o ~ ~, + o't.: ~.o-o. p.o.v, o /* /I-~o!l o I 1.6 I,.. 9 "", ',0, 9 ~ #'o ;;.~,.,,. -"* t,., 1.4 9, j3 # ~" ; - *1 ~ I :. 6 :." '+ ~ * I *.. r - " 9 9 ~., i,l[, -, ~' %'l''t*l *'to ~,ll~: II ~*, /o [o \~.'~ F, r 1.2 ; ~, wit IV~ i ~'~.. ;- '~,,; J o o ~ ~'~ ~ t it,; ~ :/ + i + "4 ~ ' ::[ HOURS AFTER TREATMENT Figure 3. Mean LH (LER 786-3) concentrations in serum of gilts sampled every 2 h for 72 h after single injections of oil, 400 ~ag E 2 or 400 ~ag E 2 followed by 1.5 ~g GnRH every hour (Exp. 2). Standard errors of the means ranged from.3 to.4 for the oil treatment and from.05 to.4 for E~ and E2-GnRH treatments. 300 compared with 60 to 80 pg/ml) than is normally observed before ovulation (Henricks et al., 1972; Stevenson et al., 1981). Thus, the duration of estrogen treatment may have been insufficient to affect FSH. Concentrations of LH in sera and pituitaries were altered after treatment with E2. Serum LH was suppressed through 26 (Exp. 1) to 32 (Exp. 2) h after E2 treatment and then rose to levels similar to oil-treated controls. The time course of suppression of LH in the present study was similar to results of other experiments TABLE 2. CONCENTRATIONS OF LH IN OVARIECTOMIZED GILTS GIVEN 400 #G E 2 OR 400/zG E 2 PLUS GnRH (EXP. 2) Hours from Treatment E 2 injection E a Ea-GnRHa b,7 +~ atreatment differences exist at all periods (P<.05). bmean SE~ in which estrogens were administered to female pigs (Lantz and Zimmerman, 1972; Elsaesser and Foxcroft, 1978; Elsaesser and Parvizi, 1979). The previous reports, however, described preovulatory-like LH surges 48 to 60 h after estrogens administered at greater doses than used in the present study. It appears that the level of E2 administerd in the present experiments was appropriate to suppress LH without eliciting a preovulatory-like surge. The low GnRH levels in SME and HPA at 24 h after E2 could indicate either that release of GnRH had occurred or that accumulation of GnRH in these tissues was inhibited. The data support the latter concept because no increases in serum or pituitary LH, which would have indicated appreciable GnRH release, were detected during 24 h after E2 treatment. Thus, the fact that GnRH was suppressed in SME and HPA at 24 h after E2, yet LH in AP was not suppressed, suggests a link between an apparent inhibition of GnRH release and suppression of serum LH after E2 treatment. Lowered GnRH in HPA appears to indicate that less GnRH reached the pituitary in E2-treated gilts than in control gilts. Results from Exp. 2 demonstrate that the pituitary is capable of responding to GnRH with increased LH secretion during a period when serum LH is suppressed by E2. This result is consistent with the observation in Exp. 1 that

7 LH, FSH AND GnRH AFTER ESTRADIOL IN OVEX PIGS 907 E2 lowered endogenous GnRH release. Kesner et al. (1981) also observed that the pituitary responded to exogenous GnRH during a time when LH was suppressed by E2 in steers and ovariectomized heifers, and they theorized that endogenous GnRH release in cattle was suppressed by E 2. The reason why the response to GnRH injections in Eu-treated gilts was only 40% of control levels during 20 to 32 h after E2 treatment cannot be determined from the present study. One explanation for this failure to restore LH to control levels is that administration of GnRH hourly did not simulate adequately the endogenous secretory patterns of GnRH, or the hourly dose of GnRH may have been too low. However, GnRH administered in a similar manner induced estrus and ovulation in anestrous lactating sows (Cox and Britt, 1982a). Perhaps simulation of GnRH release in ovariectomized gilts requires a different level or pattern of GnRH administration than that used in the present study. Another mechanism by which E2 lowered serum LH may have been by direct suppression of LH release by the anterior pituitary, and the GnRH injections may have overridden this suppression. Direct evidence for E2-induced suppression of LH production by the pig pituitary is lacking. However, results obtained with cultured piuitary cells of species other than pigs support the concept of an estrogeninduced enhancement of LH secretion rather than a suppression (Drouin et al., 1976; Padmanabhan et al., 1978; Huang and Miller, 1980). Given these considerations, it is more likely that suppression of LH in serum during 24 h after E2 was mediated primarily by inhibition of GnRH secretion. At 36 h after E2, LH in AP and GnRH in MBH and SME were at greater concentrations than in controls. It is impossible to attribute these increased hormone concentrations soley to either direct stimulation of LH and GnRH synthesis by E2 or simply a subsiding of E2- induced suppression of these hormones. These increased levels of LH and GnRH may have been responsible for the increased serum levels of LH by 36 h in E2-treated gilts. However, the fact that GnRH in HPA was still suppressed 36 h after E2 treatment prevents the unqualified conclusion that increased LH in AP and serum was due to increased GnRH secretion per se. In weaned sows, increased GnRH concentration in HPA occurred later than increased GnRH in MBH and SME (Cox and Britt, 1982b). Killing animals at 36 h after E 2 may have been too early for concentrations of GnRH in HPA to reflect increases in MBH and SME. It is possible that E2 caused increased LH in AP without a concomitant increase in HPA concentration of GnRH simply by increasing the number of GnRH receptors in the AP (Clayton et al., 1980). The effects of GnRH on synthesis and secretion of LH are enhanced by estrogens in vivo (Reeves et al., 1971; Zolman et al., 1974; Beck and Convey, 1977) and in vitro (Drouin et al., 1976; Padmanabham et al., 1978; Huang and Miller, 1980). Additionally, in ovariectomized rats, estrogen increased the amount of plasma obtained from hypophyseal portal vessels per unit of time (Sarkar and Fink, 1979; Sherwood and Fink, 1980). Hence, increased portal vessel flow rates after E2 treatment could have enabled more GnRH to reach the AP even though concentration of GnRH in HPA was unchanged. In the present study, the time-dependent alerations in GnRH in MBH, SME and HPA and LH in AP suggest that E2 initiated a series of processes that together affected serum LH. These processes led first to a suppression of serum LH and a decline in hypothalamic GnRH and then to a rebound in these hormones. The relative importance of GnRH-mediated alterations of pituitary LH, direct effects of E2 on pituitary LH and interactions between GnRH and E2 cannot be determined from this study. However, two liines of evidence indicate involvement of endogenous GnRH in suppression of LH by E:. First, GnRH concentrations in SME and HPA were suppressed at 24 h after E2 and second, administration of GnRH caused LH secretion during a period when LH was suppressed by E2. Collectively, these results point to the hypothalamus as a site for negative feedback action of estrogen in female swine. Literature Cited Baram, T. and Y. Koch Evidence for the dependence of serum luteinizing hormone surge on a transient, enhanced secretion of gonadotropinreleasing hormone from the hypothalamus. Neuroendocrinology 23:151. Beck, T. W. and E. M. Convey Estradlol control of serum luteinizing hormone concentrations in the bovine. J. Anim. Sci. 45:1096. Chen, H. T., J. Geneau and J. Meites Effects of castration, steroid replacement, and hypophysectomy on hypothalamic LHRH and serum LH. Proc. Soc. Exp. Biol. Med. 156:127. Clayton, R. N., A. R. Solano, A. Garcia-Vela, M. L.

8 908 COX AND BRITT Dufau and K. J. Catt Regulation of pitu!tary receptors for gonadotropin-releasing hormone during the rat estrous cycle. Endocrinology 107:699. Cox, N. M. and J. H. Britt. 1982a. Induction of fertile estrus in lactating sows with hourly injections of gonadotropin-releasing hormone. J. Anita. Sci. 55(Suppl. 1):49. Cox, N. M. and J. H. Britt. 1982b. Relationships between endogenous gonadotropin-releasing hormone, gonadotropins and follicular development after weaning in sows. Biol. Reprod. (In press). Crighton, D. B. and G. E. Lamming The iactational anoestrus of the sow: The status of the anterior pituitary-ov'-trian system during lactation and after weaning J. Endoerinol. 43:507. Drouin, J., L. Lagace and F. Labrie Estradiolinduced increase of the LH responsiveness to LH releasing hormone (LHRH) in rat anterior pituitary cells in culture. Endocrinology 99:1477. Elsaesser, F. and G. R. Foxcroft Maturational changes in the characteristics of oestrogen-induced surges of luteinizing hormone in immature domestic gilts. J. Endocrinol. 78:455. Elsaesser, F. and N. Parvizi Estrogen feedback in the pig: Sexual differentiation and the effect of prenatal testosterone treatment. Biol. Reprod. 20:1187. Esres, K. S., V. Padmanabhan and E. M. Convey Localiztion of gonadotropin releasing hormone (GnRH) within the bovine hypothalamus. Biol. Reprod. 17:706. Gill, J. L. and H. D. Hafs Analysis of repeated measurements of artimals. J. A'nim. Sci. 33:331. Gross, D. S Effect of castration and steroid replacement on immunoreactive gonadotropinreleasing hormone in the hypothalamus and preoptic area. Endocrinology 106:1442. tlenricks, D. M., H. D. Guthrie and D. L. Handlin Plasma estrogen, progesterone and luteinizing hormone levels during the estrous cycle in pigs. Biol. Repro& 6:210. Huang, E. S. and W. L. Miller Effects of estradiol-17~ on basal and luteinizing hormone releasing hormone-induced secretion of luteinizing hormone and follicle stimulating hormone by ovine pituitary cell cultures. Biol. Reprod. 23:124. Kesner, J. S., E. M. Convey and C. R. Anderson Evidence that estradiol induces the LH surge in cattle by increasing pituitary sensitivity to LHRH and then increasing the LHRH release. Endocrinology 108:1386. Koch, Y., M. Wilchek, M. Fridkin, P. Chobsieng, U. Zor and H. R. Lindner Production and characterization of an antiserum to synthetic gonadotropin-releasing hormone. Biochem. Bio- phys. Res. Comm. 55:616. Lantz, W. B. and D. R. Zimmerman Plasma LH levels in estradiol-treated ovariectomized gilts. J. Anlm. Sci. 35:1120 (Abstr.). Miller, W. L. and J. Wu Estrogen regulation of follicle stimulating hormone production in vitro: Species variation. Endocrinology 108:673. Neill, J. D., J. M. Patton, R. A. Dailey, R. C. Tsou and G. T. Tindall Luteinizing hormone releasing hormone (LHRH) in pituitary stalk blood of rhesus monkeys: Relationship to level of LH release. Endocrinology 101:430. Nett, T. M., A. M. Akbar, G. D. Niswender, M. T. Hedlund and W. F. White A radioimmunoassay for gonadotropin-releasing hormone (Gn- RH) in serum. J. Clin. Endocrinoi. Metab. 36:880. NRC Nutrient Requirements of Domestic Animals, No. 2. Nutrient Requirements of Swine. Eighth Revised Ed. National Academy of Sciences-National Research Council, Washington, IX:. Padmanabhan, V., J. S. Kesner and E. M. Convey Effect of estradiol on basal and luteinizing hormone releasing hormone (LHRH)-induced release of luteinizing hormone (LH) from bovine pituitary cells in culture. Biol. Reprod. 18:608. Parlow, A. F., L. L. Anderson and R. M. Melampy Pituitary follicle stimulating hormone and luteinizing horr~one concentrations in relation to reproductive stages of the pig. Endocrinology 75:365. Reeves, J. J., A. Arimura and A. V. Schally Pituitary responsiveness to purified luteinizing hormone-releasing hormone (LHRH) at various stages of the estrous cycle in sheep. J. Anita. Sci. 32:123. Sarkar, D. K. and G. Fink Effects of gonadal steroids on output of luteinizing hormone releasing factor into pituitary stalk blood in the female rat. J. Endoctinol. 80:303. Sherwood, N. M. and G. Fink Effect of ovariectomy and adrenalectomy on luteinizing hormonereleasing hormone in pituitary stalk blood from female rats. Endocrinology 106:363. Snedeeor, G. W. and W. G. Cochran Statistical Methods (6th Ed.). Iowa State Univ. Press, Ames. Sokal, R. R. and F. J. Rohlf Biometry. The Principles and Practices of Statistics in Biological Research. W. H. Freeman and Co., San Francisco. Stevenson, J. S., N. M. Cox and J. H. Britt Role of the ovary in controlling LH, FSH, and prolactin secretion during and after lactation in pigs. Biol. Reprod. 24:241. Zolman, J., E. M. Convey and J. H. Britt Relationships between the luteinizing hormone response to gonadotropin releasing hormone and endogenous steroids. J. Anim. Sci. 39:355.