MOLECULAR REPRODUCTION AND DEVELOPMENT 28:35-39 (1991)
Immunoreactive Inhibin in Follicular Fluid Is Related to Meiotic Stage of the Oocyte During Final Maturation of the Porcine Follicle K.F. MILLER,' S. XIE? AND W.F. POPE' Departments of 'Obstetrics and Gynecology and 'Animal Science, The Ohio State University, Columbus, Ohio
ABSTRACT The concentration and content of inhibin was determined in individual porcine follicles from gilts ovariectomized at various times after the onset of estrus. In one experiment, gilts (n = 5) were ovariectomized at 0, 10, or 20 hr after the onset of estrus and the follicular fluids from all large follicles individually aspirated. In a second experiment, gilts (n = 6) were ovariectomized at 21, 24, 27, 30, or 34 hr after the onset of estrus; follicular fluids were aspirated; and each oocyte was stained and evaluated for cytogenetic stage of meiotic maturation. lnhibin was determined in diluted follicular fluids with a radioimmunoassay based on a synthetic peptide replica of port of the cx subunit of porcine inhibin. lnhibin values were expressed in terms of thousands of units (kU) of a World Health Organization inhibin standard (86/ 690). Concentration of inhibin did not vary among hours (overall mean 248 kU/ml). Total follicular content of inhibin also was not different among hours (overall mean 57 kU/follicle). When follicles were classified on the basis of the maturation of the oocyte, significant differences were found. Concentration of inhibin in follicles with a germinal vesicle-stage oocyte was 138 kU/ml, whereas follicles with more mature oocytes had concentrations of between 204 and 254 kU/ml. Follicular content of inhibin showed a similar pattern with 34.9 kU/follicle at germinal vesicle stage, increasing to 42.556.1 kU/follicle at later stages. Quantities of inhibin were also negatively skewed and were positively correlated to follicular content of estradiol and dermatan sulfate. Key Words: Oogenesis, Ovary, Estradiol, Atresia, Folliculostatin
were found to peak a t day 20 and then decline precipitously during estrus (Eiler and Nalbandov, 1977). A subsequent study used prepubertal gilts treated with pregnant mare serum gonadotropin (PMSG) and human chorionic gonadotropin (hCG) in a n attempt to achieve more precise control over the timing of follicular growth and ovulation (Ainsworth et al., 1980). Concentrations of estrogens and progesterone in pooled follicular fluid were found to increase during the 72 h r after PMSG. Following hCG treatment at 72 hr, the concentration of estrogens declined rapidly and remained low until ovulation, and the concentration of progesterone remained constant until a final increase occurred shortly before ovulation. Pooled follicular fluid concentrations of androstenedione, testosterone, and dihydrotestosterone did not change significantly during treatment. A recent study examined the content of steroids in individual follicles isolated from gilts that were ovariectomized a t various times after the onset of estrus (Xie e t al., 1990). The progesterone content of follicles increased just prior to ovulation and the follicular content of estradiol decreased sharply in agreement with other studies. When the same follicular fluid data were classified based on the maturation of the oocyte within each follicle, a more precise characterization of follicular endocrine changes was produced (Xie et al., 1990). The follicular content of all steroids decreased coincident with germinal vesicle breakdown. As oocytes approached metaphase 11, the content of progesterone within follicles increased, while the content of estradiol further decreased. In a limited study of inhibin concentrations measured by bioassay in pooled porcine follicular fluid, decreasing concentrations were found with a n increase in follicle size between days 10 and 18 (Van de Wiel e t al., 1983). Furthermore, concentrations of inhibin found in large follicles (6-12 mm) during estrus were 35% less
INTRODUCTION The hormonal composition of follicular fluid has been demonstrated to change markedly during preovulatory growth and development of the porcine follicle. Follicular fluid concentrations of estrogen, progesterone, and androgen in gilts that had several follicles surgically sampled on known days of the estrous cycle
0 1991 WILEY-LISS, INC.
Received May 21, 1990; accepted July 26, 1990. Address reprint requests to Dr. Kurt F. Miller, Department of Obstetrics and Gynecology, The Ohio State University, 1654 Upham Drive, Columbus, OH 43210.
36
K.F. MILLER ET AL.
than concentrations in large follicles on day 18. Another report, utilizing bioassay measurement of pooled follicular fluid from the eight to ten largest follicles, found no difference in concentration of inhibin among days 7,11, and 15 but approximately a 40% decrease in concentration of inhibin on day 19 (Suter et al., 1986). The present report will describe the concentration and content of inhibin in individual follicles collected from gilts ovariectomized a t 0, 10, or 20 h r after the onset of estrus. We will also report the concentration and content of inhibin in a set of follicles collected at 21, 24, 27, 30, or 34 h r after the onset of estrus and classified by stage of oocyte maturation. Other endocrine data for this set of follicles has previously been reported (Xie et al., 1990).
MATERIALS AND METHODS Experiment 1 Crossbred gilts were penned adjacent to boars and observed for estrous behavior every 4 hr. At 0,10, or 20 h r after the onset of estrus, gilts (n = 5) were surgically ovariectomized, each follicle of 6 mm diameter or greater was aspirated into a syringe containing 400 p1 of phosphate-buffered saline (PBS) with 0.1% (wiv) gelatin. The follicular fluid volume was determined and the diluted follicular fluid was then centrifuged at 1,800g for 10 min. The resulting supernatant was aliquoted into small vials and stored frozen for later hormone assay. Experiment 2 Crossbred gilts were penned adjacent to boars and observed for estrous behavior every 4 hr. At 21, 24, 27, 30, or 34 h r after the onset on estrus, gilts (n = 6) were surgically ovariectomized, each follicle of 6 mm diameter or greater was aspirated into a syringe containing 500 pl of PBS with 0.1% (w/v) gelatin. The follicular fluid volume was determined, the oocyte was recovered in less than 10 p1 of fluid, and the remaining diluted follicular fluid was centrifuged a t 1,800g for 10 minutes. The resulting supernatant was aliquoted (50 pl) into small vials and stored frozen for later analysis. The follicular fluid content of estradiol, progesterone, androstenedione, testosterone, and dermatan sulfate a s well as the follicular wall content of prostaglandins FZcx and E, have been previously reported for these gilts (Xie et al., 1990). As reported previously, oocytes were prepared individually and classified into one of five meiotic maturation stages: germinal vesicle (GV), germinal vesicle breakdown (GVBD), metaphase I (Met I), anaphase I and telophase I (Ana I1Telo I), and metaphase I1 (Met 11). Radioimmunoassay for Inhibin The radioimmunoassay for porcine inhibin was developed using a synthetic peptide replica (Peninsula Labs, Belmont, CA) of the 1-32 amino terminal sequence of the 01 chain of 32 kd porcine inhibin (Mason
et al., 1985). This peptide was conjugated to bovine serum albumin and used to raise antisera in rabbits. A modification of this peptide (Tyro)was used for radioiodination by methods previously described (Miller et al., 1986). The assay was conducted in 0.14 M NaC1, 0.02 M NaPO,, 0.05 M NaHCO,, 0.01% (wiv) thimerosal, 0.001% ( w h ) phenol red, pH 7.4, with 0.05% (w1v) polyoxyethylenesorbitan monolaurate added to reduce nonspecific binding. Standards, unknowns, and radiolabel were diluted in assay buffer containing 0.05% (w/v) bovine serum albumin (BSA; fraction V) prior to assay. After incubation for 48 h r a t room temperature, bound and free antigen were separated with a sheep antirabbit gammaglobulin antibody in the presence of 2% polyethylene glycol. The working standard was a pool of porcine follicular fluid (pFF-2) collected from approximately 500 prepubertal gilts slaughtered a t a local slaughterhouse. This preparation was in t u r n standardized against the World Health Organization (WHO) Porcine Inhibin Research Standard (861690). All values are reported in terms of WHO units (U) or kilounits (kU). Both the WHO standard and a n NIH folliculostatin preparation (FS RP 119-6-4) were parallel to the pFF-2 working standard. Similarly, serial dilutions of unknown follicular fluids collected in these experiments were also parallel to standards. Intra- and interassay coefficients of variation were 8.3% and 10.3%,respectively. The least detectable amount of inhibin, defined a s a 5% decrease in binding, was 2.2 0.3 U per assay tube. As has been reported for other inhibin radioimmunoassays, serum or serum proteins interfered with binding to the antibody (Hasegawa et al., 1988; Robertson et al., 1988; Schanbacher, 1988). This possible source of variation was controlled for in the following manner. First, because the follicular fluid samples were diluted a n average of 614-fold (range 314-1,450fold) the quantity of interfering protein in the assay was quite small (less than 10 pg, based on a 5.5% protein concentration of porcine follicular fluid; Suter et al., 1986). This is much less than the amount of porcine serum albumin needed to interfere with binding (200 pg). Second, a constant amount of BSA (100 pg) was present in all assay tubes because the radiolabel, the standards, and the unknowns were all diluted in assay buffer containing 0.05% ( w h ) BSA. Thus the follicular fluid sample contribution of protein is small in both absolute terms (less than 10 pg) and relative terms (less than 10% of total protein).
*
Statistical Analysis All analyses were conducted with release 6.03 of SAS for the personal computer (SAS, 1988). The general linear model procedure was used to conduct a split plot analysis of variance to examine differences among hours in both the follicular concentration of inhibin and total follicular content of inhibin. Within each ex-
PORCINE FOLLICULAR INHIBIN DURING MEIOSIS TABLE 1. Number of Follicles per Gilt (mean f SD), Concentration of Inhibin in Follicular Fluid, and Total Follicular Inhibin Content in Ovaries Collected From Gilts at 0, 10, or 20 hr After the Onset of Estrus (Experiment 1; n = 5)* Variable Number of folliclesigilt Concentration of inhibin (kU1ml) Total inhibin content (kUifollicle)
0
Hours after onset of estrus 10 20
17.6 t 2.1 225
?
80
68.7 ? 29.1
19.2 ? 1.9
20.4 i- 3.6
325
2
118
306
?
107
77.1
i-
26.8
64.2
?
19.4
*Quantities of inhibin are expressed in thousands of units (kU) of the WHO inhibin research standard (861690).
periment, the effect of hour was tested using animal within hour a s the error term. Means were compared with Duncan’s multiple range test. In experiment 2, data were further cross-classified by meiotic maturation stage. The effect of stage was tested using animal by stage within hour a s the error term, and the interaction of stage with hour was tested by the residual. Additionally, both concentration and content data for each gilt were tested for normality of distribution with the univariate procedure of SAS. The number of animals with significantly skewed data and the mean coefficients of skewness were calculated. In experiment 2, the relationships between follicular content of inhibin and the content of estradiol, progesterone, androstenedione, testosterone, and dermatan sulfate previously determined (Xie et al., 1990) were examined by calculating the partial correlation coefficients using the correlation procedure of SAS.
RESULTS Experiment 1 The mean number of follicles 6 mm in diameter or greater did not differ among gilts ovariectomized at 0, 10, or 20 h r after the onset of estrus (Table 1). The concentration and total content of inhibin varied greatly among gilts. Mean concentration of all follicles within a n animal ranged from 168 to 444 kU/ml; similarly, mean follicular content within a n animal ranged from 46.2 to 103.0 kUifollicle. There were no differences in either concentration or total follicular content of inhibin among hours (Table 1). Four of fifteen gilts had significantly skewed data for inhibin concentration, whereas two of fifteen had skewed data for inhibin content. Mean coefficients of skewness were -0.64 and -0.38 for concentration and content, respectively. Experiment 2 The mean number of follicles 6 mm in diameter or greater did not differ among hours (Table 2). As re-
37
ported previously, some of these follicles were classified as atretic based on their steroidal content. The mean inhibin concentration and content of the follicles classified as atretic were 85 2 75 kU/ml and 16.0 18.1 kUifollicle, respectively. However, evaluation of all of the follicles or of only those follicles that were not classified as atretic resulted in the same pattern of changes and similar statistical differences. The data from all of the follicles are presented here. The concentration and total content of inhibin within a n animal ranged from 71 to 422 kUiml and from 12.3 to 106.1 kUifollicle, respectively. There were no differences among hours for either concentration or total follicular content of inhibin (Table 2). Data from all follicles collected over the five times and classified by the stage of maturation of the oocyte are presented in Table 3. Concentrations of inhibin were higher (P < 0.05) in follicles containing GVBD oocytes than in those follicles which contained GV oocytes (Table 3). A further though nonsignificant increase in the concentration of inhibin was found in follicles containing Met I, Ana IiTelo I, or Met I1 oocytes. Changes in total follicular content of inhibin were similar with follicles a t later stages containing more inhibin than follicles at GVBD, which in t u r n had greater content than follicles with GV-stage oocytes. Five of thirty gilts had significantly skewed data for inhibin concentration whereas 13 of 30 had significantly skewed data for inhibin content. The mean coefficients of skewness were -0.40 and -0.42 for concentration and content, respectively. The partial correlation of total inhibin content (after statistically removing the effects of hour and stage) with previously determined follicular contents indicated a significant positive correlation with estradiol ( + 0.189, P < 0.001) and dermatan sulfate ( 0.296, P < 0.001) but no significant correlation with progesterone, androstenedione, or testosterone.
*
+
DISCUSSION There are several reports of bioassay measurement of inhibin in pooled porcine follicular fluids (Van de Wiel et al., 1983; Suter et al., 1986). This is the first report of inhibin values for individual porcine follicles determined by radioimmunoassay. Comparison of the absolute levels of immunoreactive inhibin in follicular fluid that we report to levels reported previously is difficult because there was no international reference standard until recently (Burger and Igarashi, 1988). As a point of reference, we found that mean concentrations of inhibin in follicular fluid at all times sampled (0-34 h r after onset of estrus) were higher than the concentrations of inhibin in pools of follicular fluid collected from slaughtered prepubertal gilts (171 t 24 kUiml; n = 9 pools of 50 ovaries). Previous reports indicate a decrease in pooled follicular fluid concentrations of inhibin during the follicular phase (40% less than a t midcycle; Van de Wiel et al., 1983; Suter et al., 1986).
38
K.F. MILLER ET AL.
TABLE 2. Number of Follicles per Gilt (mean SD), Concentration of Inhibin in Follicular Fluid, and Total Follicular Inhibin Content in Ovaries Collected From Gilts at 21, 24, 27, 30, and 34 hr After the Onset of Estrus (ExDeriment2: n = 6)" Variable Number of folliclesigilt Concentration of inhibin (kUiml) Total inhibin content (kUifollicle)
21 17.2 ? 2.7 215 138 54.9 40.6
* *
Hours after onset of estrus 27 30 16.0 t 2.6 16.2 2 3.5 15.7 t 1.6 240 t 136 207 2 129 222 t 116 50.8 t 29.1 49.9 -c 35.5 43.7 26.4 24
*
34 15.5 2.5 242 112 46.2 2 22.2
* *
*Quantities of inhibin are expressed in thousands of units (kU) of the WHO inhibin research standard (861690).
TABLE 3. Mean (kSD) Concentration of Inhibin in Follicular Fluid and Total Follicular Inhibin Content in Ovarian Follicles Collected from 30 Gilts Between 21 and 34 hr After the Onset of Estrus and Classified According to the Meiotic Maturation of the Oocyte Within Each Follicle (Experiment 2)* Variable Number of follicles Concentration of inhibin (kUiml) Total inhibin content (kUifollicle)
GV 69 13tla t 83 34.ga 2 25.5
Meiotic staee of oocvte GVBD Met I Ana IiTelo I 122 58 21 254b t 154 242b 106 204b t 149 52.gb.' -+ 22.0 56.1' 40.7 42.kjazb t 34.9
*
Met I1 178 236b 115 49.6b%c t 27.3
*
"Quantities of inhibin are expressed in thousands of units (kU) of the WHO inhibin research standard (861690). a-'Within a row, means with common superscripts were not different.
These reported changes are consistent with a report of peripheral blood concentrations of inhibin during the estrous cycle (Hasegawa et al., 1988). With once daily sampling, plasma concentrations of inhibin were found to decrease approximately 70% during the 3 days prior to ovulation. Because all the sampling times in the present report were within the 48 h r prior to ovulation no significant decreases could be detected among times. It was noted that there were large variations in mean inhibin content and concentration among animals. This variation may be responsible for the difference between experiments in the quantities of inhibin found at 20 or 21 hr. The decrease in inhibin concentrations reported to occur during estrus in the pig may be caused by the surge in luteinizing hormone (LH) that occurs a t about the time of onset of estrus (Van de Wiel et al., 1981). Studies in the rat indicated that the decrease in inhibin gene expression that occurs in preovulatory follicles is dependent on the gonadotropin surge (Woodruff e t al., 1989). The lack of a significant decrease in follicular fluid inhibin among the hours examined in the present experiments may have been because the concentration of inhibin had already declined or because a change in gene expression was not reflected in a change in follicular fluid content. An alternative possibility is that the timing andlor regulation of inhibin in the porcine ovary is different than that of the rat. In this regard, a study in cattle treated with prostaglandin to induce luteolysis found a somewhat different pattern of inhibin concentrations in follicular fluid measured by bioassay (Padmanabhan et al., 1984). Concentrations of inhibin increased for 48 h r during the induced follicular phase, concentrations of inhibin then decreased
for approximately 24 h r prior to the LH surge and then increased subsequent to the surge. Consistent with a previous report of the follicular endocrinology of some of these same follicles (Xie et al., 1990), the follicular fluid content and concentration of inhibin were significantly skewed in some of the gilts. The mean coefficients of skewness were -0.38 to -0.64 indicating that a majority of follicles had more inhibin than did a minority. The degree of skewness was not a s great as those reported for estradiol and progesterone but is in agreement with the model of a majority of more advanced follicles (Xie et al., 1990). A significant positive partial correlation was found between the follicular content of inhibin and the follicular content of estradiol. Thus, removing the effects of hour and stage, the follicular content of estradiol and inhibin within each follicle are positively correlated. A positive relationship has been previously reported between ovarian vein estradiol and inhibin-like activity in unilaterally ovariectomized gilts (Redmer et al., 1986). In follicular fluid collected from women treated with exogenous gonadotropins, a negative correlation between estradiol and bioactive inhibin was found (Pampfer et al., 1989). In contrast, peripheral serum concentrations of estradiol and inhibin in gonadotropin-treated women have been reported to be highly positively correlated (McLachlan et al., 1986). Concentrations of inhibin in follicular fluid have been reported to be higher in nonatretic follicles than in atretic follicles of the cow (Padmanabhan et al., 1984) but equivalent in the ewe (Tsonis et al., 1983). In experiment 2 of the present study, size-matched atretic follicles were found to have only one-third of the inhibin content of nonatretic follicles. The association of
PORCINE FOLLICULAR INHIBIN DURING MEIOSIS increased concentrations of inhibin with increased stages of oocyte maturation may be a reflection of this with the healthiest follicles undergoing maturation sooner than the nonatretic but less healthy ones. Consequently, increased concentrations of inhibin are found in follicles that have more mature oocytes. It has recently been reported that bovine follicular fluid contains two a-subunit precursor proteins (Robertson et al., 1989). Assuming that such proteins exist in porcine follicular fluid they would be expected to cross react to some degree in the radioimmunoassay. Estimates of the relative abundance of these proteins in bovine follicular fluid indicate that they are present a t about 15% the level of inhibin. The biological function, interconvertability, and temporal changes of these proteins are not known, but in this study their existence may inflate the inhibin values by as much as 15%. In summary, porcine inhibin in individual follicles was not found to vary among hours during the interval of 0 to 34 h r after onset of estrus. However, a significant relationship was found between quantities of inhibin in individual follicles and the cytogenetic stage of oocyte maturation during the interval between 21 and 34 h r after onset of estrus.
ACKNOWLEDGMENTS This research was supported in part by federal and state funds appropriated to the Ohio Agricultural Research and Development Center, The Ohio State University (manuscript No. 138-90) and by U.S.D.A. competitive grant 85-CRCR-1-1875 and Hatch Project 814.
REFERENCES Ainsworth L, Tsang BK, Downey BR, Marcus GJ, Armstrong DT (19801: Interrelationships between follicular fluid steroid levels, gonadotropic stimuli, and oocyte maturation during preovulatory development of porcine follicles. Biol Reprod 23:621-627. Burger HG, Igarashi M (1988): Inhibin: definition and nomenclature, including related substances. Endocrinology 122:1701-1702. Eiler H, Nalbandov AV (19771: Sex steroids in follicular fluid and blood plasma during the estrous cycle of pigs. Endocrinology 100: 331-338. Hasegawa Y, Miyamoto K, Iwamura S, Igarashi M (1988):Changes in serum concentrations of inhibin in cyclic pigs. J Endocrinol 118: 211-219. Mason AJ, Hayflick JS, Ling N, Esch F, Ueno N , Ying SY, Guillemin
39
R, Niall H, Seeburg, PH (1985):Complementary DNA sequences of ovarian follicular fluid inhibin show precursor structure and homology with transforming growth factor-beta. Nature 318: 659-663. McLachlan RI, Robertson DM, Healy DL, deKretser DM, Burger HG (19861: Plasma inhibin levels during gonadotropin-induced ovarian hyperstimulation for IVF: a new index of follicular function? Lancet 1:1233-1234. Miller KF, Bolt DJ, Goldsby RA (1986):A rapid technique using radiolabeled soluble antigen to screen hybridoma culture supernatants. Methods Enzymol 121:433-437. Padmanabhan V, Convey EM, Roche J F , Ireland JJ (19841: Changes in inhibin-like bioactivity in ovulatory and atretic folicles and utero-ovarian venous blood after prostaglandin-induced luteolysis in heifers. Endocrinology 115:1332-1340. Pampfer S, Vankrieken L, Loumaye E, De Hertogh R, Thomas K (19891: Inhibin and renin in follicular fluids of patients with one or two ovaries stimulated with a GnRH agonist and gonadotrophins. Hum Reprod 4:396-402. Redmer DA, Christenson RK, Ford J J , Day BN, Goodman AL (1986): Inhibin-like activity in ovarian venous serum after unilateral ovariectomy in prepubertal gilts. Biol Reprod 34:357-362. Robertson DM, Giacometti M, Foulds LM, Lahnstein J , Goss NH, Hearn MTW, deKretser DM (1989): Isolation of inhibin alpha subunit precursor proteins from bovine follicular fluid. Endocrinology 125:2141-2149. Robertson DM, Hayward S, Irby D, Jacobsen J , Clarke L, McLachlan RI, deKretser DM (1988): Radioimmunoassay of rat serum inhibin: Changes after PMSG stimulation and gonadectomy. Mol Cell Endocrinol 58:l-8. SAS Institute Inc. (19881: “SAS User’s Guide, Release 6.03 Edition.” Cary, NC: SAS Institute. Schanbacher B (1988): Radioimmunoassay of inhibin: Serum responses to unilateral and bilateral orchidectomy. Endocrinology 123:2323-2330. Suter DE, Bahr JM, Dziuk PJ, Schwartz NB (1986):Effect of stage of the porcine estrous cycle and method of collection on folliculostatin in porcine follicular fluid. Anim Reprod Sci 11:43-49. Tsonis CG, Quigg H, Lee VWK, Leversha L, Trounson AO, Findlay J K (1983): Inhibin in individual ovine follicles in relation to diameter and atresia. J Reprod Fertil 67:83-90. Van de Wiel DFM, Bar-Ami S, Tsafriri A, de Jong FH (1983):Oocyte maturation inhibitor, inhibin and steroid concentrations in porcine follicular fluid a t various stages of the oestrous cycle. J Reprod Fertil 68:247-252. Van de Wiel DFM, Erkens J, Koops W, Vos E, Van Landeghem AAJ (19811: Periestrous and midluteal time courses of circulating LH, FSH, prolactin, estradiol-17 beta and progesterone in the domestic pig. Biol Reprod 24:223-233. Woodruff TK, D’Agostino J , Schwartz NB, Mayo KE (1989): Decreased inhibin gene expression in preovulatory follicles requires primary gonadotropin surges. Endocrinology 124:2193-2199. Xie S, Broermann DM, Nephew KP, Ottobre JS, Day ML, Pope WF (1990):Changes in follicular endocrinology during final maturation of porcine oocytes. Domestic Anim Endocrinol 7:75-82.
Immunoreactive Inhibin in Follicular Fluid Is Related to Meiotic Stage of the Oocyte During Final Maturation of the Porcine Follicle K.F. MILLER,' S. XIE? AND W.F. POPE' Departments of 'Obstetrics and Gynecology and 'Animal Science, The Ohio State University, Columbus, Ohio
ABSTRACT The concentration and content of inhibin was determined in individual porcine follicles from gilts ovariectomized at various times after the onset of estrus. In one experiment, gilts (n = 5) were ovariectomized at 0, 10, or 20 hr after the onset of estrus and the follicular fluids from all large follicles individually aspirated. In a second experiment, gilts (n = 6) were ovariectomized at 21, 24, 27, 30, or 34 hr after the onset of estrus; follicular fluids were aspirated; and each oocyte was stained and evaluated for cytogenetic stage of meiotic maturation. lnhibin was determined in diluted follicular fluids with a radioimmunoassay based on a synthetic peptide replica of port of the cx subunit of porcine inhibin. lnhibin values were expressed in terms of thousands of units (kU) of a World Health Organization inhibin standard (86/ 690). Concentration of inhibin did not vary among hours (overall mean 248 kU/ml). Total follicular content of inhibin also was not different among hours (overall mean 57 kU/follicle). When follicles were classified on the basis of the maturation of the oocyte, significant differences were found. Concentration of inhibin in follicles with a germinal vesicle-stage oocyte was 138 kU/ml, whereas follicles with more mature oocytes had concentrations of between 204 and 254 kU/ml. Follicular content of inhibin showed a similar pattern with 34.9 kU/follicle at germinal vesicle stage, increasing to 42.556.1 kU/follicle at later stages. Quantities of inhibin were also negatively skewed and were positively correlated to follicular content of estradiol and dermatan sulfate. Key Words: Oogenesis, Ovary, Estradiol, Atresia, Folliculostatin
were found to peak a t day 20 and then decline precipitously during estrus (Eiler and Nalbandov, 1977). A subsequent study used prepubertal gilts treated with pregnant mare serum gonadotropin (PMSG) and human chorionic gonadotropin (hCG) in a n attempt to achieve more precise control over the timing of follicular growth and ovulation (Ainsworth et al., 1980). Concentrations of estrogens and progesterone in pooled follicular fluid were found to increase during the 72 h r after PMSG. Following hCG treatment at 72 hr, the concentration of estrogens declined rapidly and remained low until ovulation, and the concentration of progesterone remained constant until a final increase occurred shortly before ovulation. Pooled follicular fluid concentrations of androstenedione, testosterone, and dihydrotestosterone did not change significantly during treatment. A recent study examined the content of steroids in individual follicles isolated from gilts that were ovariectomized a t various times after the onset of estrus (Xie e t al., 1990). The progesterone content of follicles increased just prior to ovulation and the follicular content of estradiol decreased sharply in agreement with other studies. When the same follicular fluid data were classified based on the maturation of the oocyte within each follicle, a more precise characterization of follicular endocrine changes was produced (Xie et al., 1990). The follicular content of all steroids decreased coincident with germinal vesicle breakdown. As oocytes approached metaphase 11, the content of progesterone within follicles increased, while the content of estradiol further decreased. In a limited study of inhibin concentrations measured by bioassay in pooled porcine follicular fluid, decreasing concentrations were found with a n increase in follicle size between days 10 and 18 (Van de Wiel e t al., 1983). Furthermore, concentrations of inhibin found in large follicles (6-12 mm) during estrus were 35% less
INTRODUCTION The hormonal composition of follicular fluid has been demonstrated to change markedly during preovulatory growth and development of the porcine follicle. Follicular fluid concentrations of estrogen, progesterone, and androgen in gilts that had several follicles surgically sampled on known days of the estrous cycle
0 1991 WILEY-LISS, INC.
Received May 21, 1990; accepted July 26, 1990. Address reprint requests to Dr. Kurt F. Miller, Department of Obstetrics and Gynecology, The Ohio State University, 1654 Upham Drive, Columbus, OH 43210.
36
K.F. MILLER ET AL.
than concentrations in large follicles on day 18. Another report, utilizing bioassay measurement of pooled follicular fluid from the eight to ten largest follicles, found no difference in concentration of inhibin among days 7,11, and 15 but approximately a 40% decrease in concentration of inhibin on day 19 (Suter et al., 1986). The present report will describe the concentration and content of inhibin in individual follicles collected from gilts ovariectomized a t 0, 10, or 20 h r after the onset of estrus. We will also report the concentration and content of inhibin in a set of follicles collected at 21, 24, 27, 30, or 34 h r after the onset of estrus and classified by stage of oocyte maturation. Other endocrine data for this set of follicles has previously been reported (Xie et al., 1990).
MATERIALS AND METHODS Experiment 1 Crossbred gilts were penned adjacent to boars and observed for estrous behavior every 4 hr. At 0,10, or 20 h r after the onset of estrus, gilts (n = 5) were surgically ovariectomized, each follicle of 6 mm diameter or greater was aspirated into a syringe containing 400 p1 of phosphate-buffered saline (PBS) with 0.1% (wiv) gelatin. The follicular fluid volume was determined and the diluted follicular fluid was then centrifuged at 1,800g for 10 min. The resulting supernatant was aliquoted into small vials and stored frozen for later hormone assay. Experiment 2 Crossbred gilts were penned adjacent to boars and observed for estrous behavior every 4 hr. At 21, 24, 27, 30, or 34 h r after the onset on estrus, gilts (n = 6) were surgically ovariectomized, each follicle of 6 mm diameter or greater was aspirated into a syringe containing 500 pl of PBS with 0.1% (w/v) gelatin. The follicular fluid volume was determined, the oocyte was recovered in less than 10 p1 of fluid, and the remaining diluted follicular fluid was centrifuged a t 1,800g for 10 minutes. The resulting supernatant was aliquoted (50 pl) into small vials and stored frozen for later analysis. The follicular fluid content of estradiol, progesterone, androstenedione, testosterone, and dermatan sulfate a s well as the follicular wall content of prostaglandins FZcx and E, have been previously reported for these gilts (Xie et al., 1990). As reported previously, oocytes were prepared individually and classified into one of five meiotic maturation stages: germinal vesicle (GV), germinal vesicle breakdown (GVBD), metaphase I (Met I), anaphase I and telophase I (Ana I1Telo I), and metaphase I1 (Met 11). Radioimmunoassay for Inhibin The radioimmunoassay for porcine inhibin was developed using a synthetic peptide replica (Peninsula Labs, Belmont, CA) of the 1-32 amino terminal sequence of the 01 chain of 32 kd porcine inhibin (Mason
et al., 1985). This peptide was conjugated to bovine serum albumin and used to raise antisera in rabbits. A modification of this peptide (Tyro)was used for radioiodination by methods previously described (Miller et al., 1986). The assay was conducted in 0.14 M NaC1, 0.02 M NaPO,, 0.05 M NaHCO,, 0.01% (wiv) thimerosal, 0.001% ( w h ) phenol red, pH 7.4, with 0.05% (w1v) polyoxyethylenesorbitan monolaurate added to reduce nonspecific binding. Standards, unknowns, and radiolabel were diluted in assay buffer containing 0.05% (w/v) bovine serum albumin (BSA; fraction V) prior to assay. After incubation for 48 h r a t room temperature, bound and free antigen were separated with a sheep antirabbit gammaglobulin antibody in the presence of 2% polyethylene glycol. The working standard was a pool of porcine follicular fluid (pFF-2) collected from approximately 500 prepubertal gilts slaughtered a t a local slaughterhouse. This preparation was in t u r n standardized against the World Health Organization (WHO) Porcine Inhibin Research Standard (861690). All values are reported in terms of WHO units (U) or kilounits (kU). Both the WHO standard and a n NIH folliculostatin preparation (FS RP 119-6-4) were parallel to the pFF-2 working standard. Similarly, serial dilutions of unknown follicular fluids collected in these experiments were also parallel to standards. Intra- and interassay coefficients of variation were 8.3% and 10.3%,respectively. The least detectable amount of inhibin, defined a s a 5% decrease in binding, was 2.2 0.3 U per assay tube. As has been reported for other inhibin radioimmunoassays, serum or serum proteins interfered with binding to the antibody (Hasegawa et al., 1988; Robertson et al., 1988; Schanbacher, 1988). This possible source of variation was controlled for in the following manner. First, because the follicular fluid samples were diluted a n average of 614-fold (range 314-1,450fold) the quantity of interfering protein in the assay was quite small (less than 10 pg, based on a 5.5% protein concentration of porcine follicular fluid; Suter et al., 1986). This is much less than the amount of porcine serum albumin needed to interfere with binding (200 pg). Second, a constant amount of BSA (100 pg) was present in all assay tubes because the radiolabel, the standards, and the unknowns were all diluted in assay buffer containing 0.05% ( w h ) BSA. Thus the follicular fluid sample contribution of protein is small in both absolute terms (less than 10 pg) and relative terms (less than 10% of total protein).
*
Statistical Analysis All analyses were conducted with release 6.03 of SAS for the personal computer (SAS, 1988). The general linear model procedure was used to conduct a split plot analysis of variance to examine differences among hours in both the follicular concentration of inhibin and total follicular content of inhibin. Within each ex-
PORCINE FOLLICULAR INHIBIN DURING MEIOSIS TABLE 1. Number of Follicles per Gilt (mean f SD), Concentration of Inhibin in Follicular Fluid, and Total Follicular Inhibin Content in Ovaries Collected From Gilts at 0, 10, or 20 hr After the Onset of Estrus (Experiment 1; n = 5)* Variable Number of folliclesigilt Concentration of inhibin (kU1ml) Total inhibin content (kUifollicle)
0
Hours after onset of estrus 10 20
17.6 t 2.1 225
?
80
68.7 ? 29.1
19.2 ? 1.9
20.4 i- 3.6
325
2
118
306
?
107
77.1
i-
26.8
64.2
?
19.4
*Quantities of inhibin are expressed in thousands of units (kU) of the WHO inhibin research standard (861690).
periment, the effect of hour was tested using animal within hour a s the error term. Means were compared with Duncan’s multiple range test. In experiment 2, data were further cross-classified by meiotic maturation stage. The effect of stage was tested using animal by stage within hour a s the error term, and the interaction of stage with hour was tested by the residual. Additionally, both concentration and content data for each gilt were tested for normality of distribution with the univariate procedure of SAS. The number of animals with significantly skewed data and the mean coefficients of skewness were calculated. In experiment 2, the relationships between follicular content of inhibin and the content of estradiol, progesterone, androstenedione, testosterone, and dermatan sulfate previously determined (Xie et al., 1990) were examined by calculating the partial correlation coefficients using the correlation procedure of SAS.
RESULTS Experiment 1 The mean number of follicles 6 mm in diameter or greater did not differ among gilts ovariectomized at 0, 10, or 20 h r after the onset of estrus (Table 1). The concentration and total content of inhibin varied greatly among gilts. Mean concentration of all follicles within a n animal ranged from 168 to 444 kU/ml; similarly, mean follicular content within a n animal ranged from 46.2 to 103.0 kUifollicle. There were no differences in either concentration or total follicular content of inhibin among hours (Table 1). Four of fifteen gilts had significantly skewed data for inhibin concentration, whereas two of fifteen had skewed data for inhibin content. Mean coefficients of skewness were -0.64 and -0.38 for concentration and content, respectively. Experiment 2 The mean number of follicles 6 mm in diameter or greater did not differ among hours (Table 2). As re-
37
ported previously, some of these follicles were classified as atretic based on their steroidal content. The mean inhibin concentration and content of the follicles classified as atretic were 85 2 75 kU/ml and 16.0 18.1 kUifollicle, respectively. However, evaluation of all of the follicles or of only those follicles that were not classified as atretic resulted in the same pattern of changes and similar statistical differences. The data from all of the follicles are presented here. The concentration and total content of inhibin within a n animal ranged from 71 to 422 kUiml and from 12.3 to 106.1 kUifollicle, respectively. There were no differences among hours for either concentration or total follicular content of inhibin (Table 2). Data from all follicles collected over the five times and classified by the stage of maturation of the oocyte are presented in Table 3. Concentrations of inhibin were higher (P < 0.05) in follicles containing GVBD oocytes than in those follicles which contained GV oocytes (Table 3). A further though nonsignificant increase in the concentration of inhibin was found in follicles containing Met I, Ana IiTelo I, or Met I1 oocytes. Changes in total follicular content of inhibin were similar with follicles a t later stages containing more inhibin than follicles at GVBD, which in t u r n had greater content than follicles with GV-stage oocytes. Five of thirty gilts had significantly skewed data for inhibin concentration whereas 13 of 30 had significantly skewed data for inhibin content. The mean coefficients of skewness were -0.40 and -0.42 for concentration and content, respectively. The partial correlation of total inhibin content (after statistically removing the effects of hour and stage) with previously determined follicular contents indicated a significant positive correlation with estradiol ( + 0.189, P < 0.001) and dermatan sulfate ( 0.296, P < 0.001) but no significant correlation with progesterone, androstenedione, or testosterone.
*
+
DISCUSSION There are several reports of bioassay measurement of inhibin in pooled porcine follicular fluids (Van de Wiel et al., 1983; Suter et al., 1986). This is the first report of inhibin values for individual porcine follicles determined by radioimmunoassay. Comparison of the absolute levels of immunoreactive inhibin in follicular fluid that we report to levels reported previously is difficult because there was no international reference standard until recently (Burger and Igarashi, 1988). As a point of reference, we found that mean concentrations of inhibin in follicular fluid at all times sampled (0-34 h r after onset of estrus) were higher than the concentrations of inhibin in pools of follicular fluid collected from slaughtered prepubertal gilts (171 t 24 kUiml; n = 9 pools of 50 ovaries). Previous reports indicate a decrease in pooled follicular fluid concentrations of inhibin during the follicular phase (40% less than a t midcycle; Van de Wiel et al., 1983; Suter et al., 1986).
38
K.F. MILLER ET AL.
TABLE 2. Number of Follicles per Gilt (mean SD), Concentration of Inhibin in Follicular Fluid, and Total Follicular Inhibin Content in Ovaries Collected From Gilts at 21, 24, 27, 30, and 34 hr After the Onset of Estrus (ExDeriment2: n = 6)" Variable Number of folliclesigilt Concentration of inhibin (kUiml) Total inhibin content (kUifollicle)
21 17.2 ? 2.7 215 138 54.9 40.6
* *
Hours after onset of estrus 27 30 16.0 t 2.6 16.2 2 3.5 15.7 t 1.6 240 t 136 207 2 129 222 t 116 50.8 t 29.1 49.9 -c 35.5 43.7 26.4 24
*
34 15.5 2.5 242 112 46.2 2 22.2
* *
*Quantities of inhibin are expressed in thousands of units (kU) of the WHO inhibin research standard (861690).
TABLE 3. Mean (kSD) Concentration of Inhibin in Follicular Fluid and Total Follicular Inhibin Content in Ovarian Follicles Collected from 30 Gilts Between 21 and 34 hr After the Onset of Estrus and Classified According to the Meiotic Maturation of the Oocyte Within Each Follicle (Experiment 2)* Variable Number of follicles Concentration of inhibin (kUiml) Total inhibin content (kUifollicle)
GV 69 13tla t 83 34.ga 2 25.5
Meiotic staee of oocvte GVBD Met I Ana IiTelo I 122 58 21 254b t 154 242b 106 204b t 149 52.gb.' -+ 22.0 56.1' 40.7 42.kjazb t 34.9
*
Met I1 178 236b 115 49.6b%c t 27.3
*
"Quantities of inhibin are expressed in thousands of units (kU) of the WHO inhibin research standard (861690). a-'Within a row, means with common superscripts were not different.
These reported changes are consistent with a report of peripheral blood concentrations of inhibin during the estrous cycle (Hasegawa et al., 1988). With once daily sampling, plasma concentrations of inhibin were found to decrease approximately 70% during the 3 days prior to ovulation. Because all the sampling times in the present report were within the 48 h r prior to ovulation no significant decreases could be detected among times. It was noted that there were large variations in mean inhibin content and concentration among animals. This variation may be responsible for the difference between experiments in the quantities of inhibin found at 20 or 21 hr. The decrease in inhibin concentrations reported to occur during estrus in the pig may be caused by the surge in luteinizing hormone (LH) that occurs a t about the time of onset of estrus (Van de Wiel et al., 1981). Studies in the rat indicated that the decrease in inhibin gene expression that occurs in preovulatory follicles is dependent on the gonadotropin surge (Woodruff e t al., 1989). The lack of a significant decrease in follicular fluid inhibin among the hours examined in the present experiments may have been because the concentration of inhibin had already declined or because a change in gene expression was not reflected in a change in follicular fluid content. An alternative possibility is that the timing andlor regulation of inhibin in the porcine ovary is different than that of the rat. In this regard, a study in cattle treated with prostaglandin to induce luteolysis found a somewhat different pattern of inhibin concentrations in follicular fluid measured by bioassay (Padmanabhan et al., 1984). Concentrations of inhibin increased for 48 h r during the induced follicular phase, concentrations of inhibin then decreased
for approximately 24 h r prior to the LH surge and then increased subsequent to the surge. Consistent with a previous report of the follicular endocrinology of some of these same follicles (Xie et al., 1990), the follicular fluid content and concentration of inhibin were significantly skewed in some of the gilts. The mean coefficients of skewness were -0.38 to -0.64 indicating that a majority of follicles had more inhibin than did a minority. The degree of skewness was not a s great as those reported for estradiol and progesterone but is in agreement with the model of a majority of more advanced follicles (Xie et al., 1990). A significant positive partial correlation was found between the follicular content of inhibin and the follicular content of estradiol. Thus, removing the effects of hour and stage, the follicular content of estradiol and inhibin within each follicle are positively correlated. A positive relationship has been previously reported between ovarian vein estradiol and inhibin-like activity in unilaterally ovariectomized gilts (Redmer et al., 1986). In follicular fluid collected from women treated with exogenous gonadotropins, a negative correlation between estradiol and bioactive inhibin was found (Pampfer et al., 1989). In contrast, peripheral serum concentrations of estradiol and inhibin in gonadotropin-treated women have been reported to be highly positively correlated (McLachlan et al., 1986). Concentrations of inhibin in follicular fluid have been reported to be higher in nonatretic follicles than in atretic follicles of the cow (Padmanabhan et al., 1984) but equivalent in the ewe (Tsonis et al., 1983). In experiment 2 of the present study, size-matched atretic follicles were found to have only one-third of the inhibin content of nonatretic follicles. The association of
PORCINE FOLLICULAR INHIBIN DURING MEIOSIS increased concentrations of inhibin with increased stages of oocyte maturation may be a reflection of this with the healthiest follicles undergoing maturation sooner than the nonatretic but less healthy ones. Consequently, increased concentrations of inhibin are found in follicles that have more mature oocytes. It has recently been reported that bovine follicular fluid contains two a-subunit precursor proteins (Robertson et al., 1989). Assuming that such proteins exist in porcine follicular fluid they would be expected to cross react to some degree in the radioimmunoassay. Estimates of the relative abundance of these proteins in bovine follicular fluid indicate that they are present a t about 15% the level of inhibin. The biological function, interconvertability, and temporal changes of these proteins are not known, but in this study their existence may inflate the inhibin values by as much as 15%. In summary, porcine inhibin in individual follicles was not found to vary among hours during the interval of 0 to 34 h r after onset of estrus. However, a significant relationship was found between quantities of inhibin in individual follicles and the cytogenetic stage of oocyte maturation during the interval between 21 and 34 h r after onset of estrus.
ACKNOWLEDGMENTS This research was supported in part by federal and state funds appropriated to the Ohio Agricultural Research and Development Center, The Ohio State University (manuscript No. 138-90) and by U.S.D.A. competitive grant 85-CRCR-1-1875 and Hatch Project 814.
REFERENCES Ainsworth L, Tsang BK, Downey BR, Marcus GJ, Armstrong DT (19801: Interrelationships between follicular fluid steroid levels, gonadotropic stimuli, and oocyte maturation during preovulatory development of porcine follicles. Biol Reprod 23:621-627. Burger HG, Igarashi M (1988): Inhibin: definition and nomenclature, including related substances. Endocrinology 122:1701-1702. Eiler H, Nalbandov AV (19771: Sex steroids in follicular fluid and blood plasma during the estrous cycle of pigs. Endocrinology 100: 331-338. Hasegawa Y, Miyamoto K, Iwamura S, Igarashi M (1988):Changes in serum concentrations of inhibin in cyclic pigs. J Endocrinol 118: 211-219. Mason AJ, Hayflick JS, Ling N, Esch F, Ueno N , Ying SY, Guillemin
39
R, Niall H, Seeburg, PH (1985):Complementary DNA sequences of ovarian follicular fluid inhibin show precursor structure and homology with transforming growth factor-beta. Nature 318: 659-663. McLachlan RI, Robertson DM, Healy DL, deKretser DM, Burger HG (19861: Plasma inhibin levels during gonadotropin-induced ovarian hyperstimulation for IVF: a new index of follicular function? Lancet 1:1233-1234. Miller KF, Bolt DJ, Goldsby RA (1986):A rapid technique using radiolabeled soluble antigen to screen hybridoma culture supernatants. Methods Enzymol 121:433-437. Padmanabhan V, Convey EM, Roche J F , Ireland JJ (19841: Changes in inhibin-like bioactivity in ovulatory and atretic folicles and utero-ovarian venous blood after prostaglandin-induced luteolysis in heifers. Endocrinology 115:1332-1340. Pampfer S, Vankrieken L, Loumaye E, De Hertogh R, Thomas K (19891: Inhibin and renin in follicular fluids of patients with one or two ovaries stimulated with a GnRH agonist and gonadotrophins. Hum Reprod 4:396-402. Redmer DA, Christenson RK, Ford J J , Day BN, Goodman AL (1986): Inhibin-like activity in ovarian venous serum after unilateral ovariectomy in prepubertal gilts. Biol Reprod 34:357-362. Robertson DM, Giacometti M, Foulds LM, Lahnstein J , Goss NH, Hearn MTW, deKretser DM (1989): Isolation of inhibin alpha subunit precursor proteins from bovine follicular fluid. Endocrinology 125:2141-2149. Robertson DM, Hayward S, Irby D, Jacobsen J , Clarke L, McLachlan RI, deKretser DM (1988): Radioimmunoassay of rat serum inhibin: Changes after PMSG stimulation and gonadectomy. Mol Cell Endocrinol 58:l-8. SAS Institute Inc. (19881: “SAS User’s Guide, Release 6.03 Edition.” Cary, NC: SAS Institute. Schanbacher B (1988): Radioimmunoassay of inhibin: Serum responses to unilateral and bilateral orchidectomy. Endocrinology 123:2323-2330. Suter DE, Bahr JM, Dziuk PJ, Schwartz NB (1986):Effect of stage of the porcine estrous cycle and method of collection on folliculostatin in porcine follicular fluid. Anim Reprod Sci 11:43-49. Tsonis CG, Quigg H, Lee VWK, Leversha L, Trounson AO, Findlay J K (1983): Inhibin in individual ovine follicles in relation to diameter and atresia. J Reprod Fertil 67:83-90. Van de Wiel DFM, Bar-Ami S, Tsafriri A, de Jong FH (1983):Oocyte maturation inhibitor, inhibin and steroid concentrations in porcine follicular fluid a t various stages of the oestrous cycle. J Reprod Fertil 68:247-252. Van de Wiel DFM, Erkens J, Koops W, Vos E, Van Landeghem AAJ (19811: Periestrous and midluteal time courses of circulating LH, FSH, prolactin, estradiol-17 beta and progesterone in the domestic pig. Biol Reprod 24:223-233. Woodruff TK, D’Agostino J , Schwartz NB, Mayo KE (1989): Decreased inhibin gene expression in preovulatory follicles requires primary gonadotropin surges. Endocrinology 124:2193-2199. Xie S, Broermann DM, Nephew KP, Ottobre JS, Day ML, Pope WF (1990):Changes in follicular endocrinology during final maturation of porcine oocytes. Domestic Anim Endocrinol 7:75-82.
