Measurement of dimeric inhibin in porcine serum: Evidence for low concentrations and existence of binding proteins

DOMESTIC ANIMAL ENDOCRINOLOGY Vol. 13(6):503-510, 1996

ELSEVIER

MEASUREMENT OF DIMERIC INHIBIN IN PORCINE SERUM: EVIDENCE FOR LOW CONCENTRATIONS AND EXISTENCE OF BINDING PROTEINS’y2 R.L. Meyer and J.E. Wheaton Department

of Animal Science, University St. Paul, MN 55108

of Minnesota

Received January 23, 1996

ABSTRACT An immunoradiometric assay and serum extraction procedure were developed to measure dimeric inhibin in porcine serum with minimal interference by putative inhibin-binding proteins. Assay sensitivity was 50 pg/tube, and it incorporated antibodies against the N-terminal region of inhibin’s o-subunit, ol-(l-25)-Ab, and against the C-terminal region of inhibin’s PA-subunit. To determine whether inhibin-binding proteins were present in porcine serum, serum was incubated with [‘*‘II-recombinant human (rh)-inhibin and then chromatographed by gel filtration. Radioiodinated rh-inhibin was associated with protein(s) ~-600 kDa. Radioiodinated rh-inhibin also was incubated with a,-macroglobulin, an inhibin-binding protein in human and rat serum. Elution profiles were similar for serum and cY,-macroglobulin. Serum- and o,-macroglobulin-[‘251]rhinhibin complexes dissociated upon exposure to 8 M urea. Porcine serum was treated with urea, after which inhibin was isolated and concentrated. The recovery of rh-inhibin added to starting serum was 28%. Concentrations of endogenous dimeric inhibin were ~28 pg/ml in serum collected from sows at random stages of the estrous cycle and were <21 pglml in serum collected from sows 2 d postweaning. Results demonstrate that 1) concentrations of dimeric inhibin are low in porcine serum, and 2) an inhibin-binding protein(s), consistent with a,-macroglobulin, is present in porcine serum. 0 Elsevier Science Inc. 1996

INTRODUCTION Serum inhibin concentrations have been measured in pigs by the use of radioimmunoassays (RIA) that incorporate antibodies (Ab) against the N-terminal region of inhibin’s a-subunit or against purified inhibin (1,2). Inhibin values derived from these assays ranged from 0.1 to 4 rig/ml. It is uncertain whether these values reflect dimeric inhibin and(or) other immunoreactive inhibin-like molecules. The specificity of inhibin RIA has been questioned because of the recognition of inhibin-like o-subunit monomers and because of interference by serum albumins (3,4). More recently, two-site immunoassays have been developed to detect dimeric inhibin (5,6). These assays use Ab against inhibin’s OL-and P-subunits. Results have revealed that concentrations of dimeric inhibin in the rat, monkey, and human are substantially lower than previous RIA estimates (7-10). Furthermore, the development of two-site immunoassays led to the discovery of inhibin-binding proteins (11,12). Tnhibin-binding proteins may harbor dimeric inhibin and thus contribute to the low concentrations detected. Inhibin was initially purified from porcine follicular fluid (pFF), and inhibin-like activity increases in ovarian venous serum after unilateral ovariectomy in prepubertal gifts (13,14). These observations suggest that ovarian inhibin plays an endocrine role in the regulation of follicle-stimulating hormone in the pig. The objective of this study was to determine concentrations of dimeric inhibin in porcine serum. Methodology was devel0 Elsevier Science Inc. 1996 655 Avenue of the Americas, New York, NY 10010

0739-7240/96/$7 5.00 PII SO739-7240(96)00086-O

504

oped to measure proteins.

MEYER AND WHEATON

dimeric

inhibin

with minimal

MATERIALS

interference

by putative

inhibin-binding

AND METHODS

Laboratory procedures were conducted at 25” C. Means + SE are presented in text and figures. Materials. Chemicals were from Sigma Chemical Co. (St. Louis, MO), unless noted otherwise. Solution constituents were as follows: phosphate-buffered saline (PBS; 0.01 M NaPO,, 0.13 M NaCI, 0.01% thimerosal, [pH 7.01); gel-PBS (0.1% gelatin and 0.05% Tween-20 in PBS); gel-NaPO, (0.1% gelatin in 0.2 M NaPO, [pH 6.01, containing 0.05% Tween-20 and 0.02% NaN,); HEPES solution (50 mM HEPES with 150 mM NaCl and 0.02% NaN, [pH 7.51); Tris buffer (50 mM Tris-HCl [pH 7.51, 0.2 M NaCl, 0.05% CHAPS, and 0.001% polymethylsulfonyl fluoride). Radioiodinations. Recombinant human (rh) inhibin-A (2.5 kg/40 pl of 0.1 M NaPO, [pH 7.11) was transferred to a reaction vial coated with 0.25 pg of Iodo-Gen (Pierce Chem. Co., Rockford, IL), and 0.5 mCi of Na 12’1 (NEN, Boston, MA) was added. After a 4-min reaction period, the mixture was fractionated through Sephadex G-25- 150 eluted with gel-NaPO,. Void volume fractions were pooled, diluted to 20 ml with gel-NaPO,, and added to a 0.8 x 5 cm column of reactive red 120-agarose. The agarose bed was washed with 10 ml of gel-NaPO,, and then bound material was eluted with 10 ml of gel-NaPO, containing 1 M KC1 and 4 M urea. Fractions containing [1251]rh-inhibin were pooled and desalted over Sephadex G-25- 150. The specific activity of [‘251]rh-inhibin was -35 p,Ci/p,g. Before use, [‘251]rh-inhibin was purified over Sephadex G-200-120 (I x 50 cm column) to remove any higher molecular weight aggregates. Mouse monoclonal Ab against the synthetic peptide B,-(82-114) (E4; 15) was radioiodinated as described above, except that reaction products were eluted with gel-PBS containing 0.05 M EDTA. The specific activity of [ lz51]PA-(82- 114)-Ab was -30 pCi/pg. Immunoradiometric Assay. Capture Ab was immunoaffinity-purified ovine polyclonal Ab raised against a synthetic fragment of ovine inhibin’s o-subunit, o-( l-25), that had been conjugated to human o-globulin (16). For immunoaffinity purification, o-( l25)-peptide was coupled to Affi-Gel 10 (Bio-Rad, Hercules, CA) by use of the method of Vaughan et al. (17). Affi-Gel 10-o-(1-25) was mixed overnight with antiserum, then collected in a column, and washed first with PBS, then with 1 M sodium thiocyanate [pH 5.81 and again with PBS. Antibody was eluted with 0.1 N glycine-HCl [pH 21 and neutralized immediately with 1 M Tris [pH 8.01 containing 150 mM NaCl. Eluate was dialyzed against HEPES solution, concentrated to 1.5 mg of protein/ml and coupled to Affi-Gel 10 at a ratio of 3 mg of protein/ml of Affi-Gel 10. Test solutions were added to microcentrifuge tubes and diluted with gel-PBS to a final volume of 1.4 ml. Hydrogen peroxide was added and mixed to yield a final concentration of 1%. After a I-hr stand, 10 ~1 of Affi-Gel-o-( l-25)-Ab was added and tubes were placed on a shaking platform for 24 hr. After this, Affi-Gel-a-( l-25)-Ab was washed three times with 1 ml of gel-PBS and then [‘251]B,-(82-114)-Ab (2 x lo5 cpm) was added for the detection of the B,-subunit. Tubes were incubated another 24 hr and then washed. Radioactivity retained on Affi-Gel-c-w-( I-25))Ab was determined, and that specifically bound was calculated by subtracting radioactivity in tubes without serum extract. Serum. Pooled serum samples for the binding protein study and the initial inhibin isolation study were collected from two multiparous Landrace x Yorkshire sows without regard to stage of the estrous cycle. Pooled serum samples for the second inhibin isolation study were collected from seven Landrace x Yorkshire sows at 2 d postweaning. Inhibin

DIMERIC INHIBIN AND BINDING

PROTEINS

IN PORCINE

SERUM

505

levels had been reported to be elevated at this time (2). Blood was allowed to clot for 2 hr at room temperature and then stored overnight at 4°C. Serum was stored at -20°C. Binding Proteins. Serum (2 ml) and gel-NaPO, (2 ml) were incubated with 0.1 pmol of [‘251]rh-inhibin for 24 hr. Human a,-macroglobulin (1 mg/2 ml of gel-NaPO,) and 2 ml of gel-NaPO, were incubated with 0.025 pmol of [iz51]rh-inhibin for 24 hr. A l-ml aliquot of each solution was loaded on a Sephadex G-200-120 column (1 x 50 cm) and eluted with gel-NaPO,. To the remaining solutions, urea and Tris buffer constituents were added to produce a final concentration of 8 M urea in Tris buffer. Urea solutions were mixed for 24 hr, and then 1-ml aliquots were chromatographed on a Superdex 200 column (1 x 50 cm; Pharmacia, Pitscataway, NJ) in 8 M urea-Tris buffer. Blue dextran was used to determine void volumes. Isolation of Inhibin. One 20-ml serum aliquot was spiked with 0.05 pmol of [1251]rhinhibin, a second was spiked with 100 ng of nonradiolabeled rh-inhibin, and a third served as control. After mixing for 24 hr, urea and Tris buffer constituents were added to make each aliquot an 8 M urea-Tris buffer solution. After 24 hr, urea-Tris buffer solutions were chromatographed over Superdex 200 (5 x 50 cm column) in 8 M urea-Tris buffer. On the basis of the elution profile of [‘251]rh-inhibin in the serum aliquot that had been spiked with [i2’I]rh-inhibin, corresponding fractions were pooled from the elution of the other two serum aliquots. Pooled fractions were exchanged into gel-NaPO, and concentrated with a diafiltration unit (Amicon, Beverly, MA). Immunoreactive inhibin-like material was isolated by mixing solutions with 2 ml of Affi-Gel-o-( 1-25)-Ab for 24 hr. Slurries were transferred to a column in which Affi-Gel-ol-(1-25)-Ab was washed with 20 ml of PBS. Immunoreactive material was eluted with 20 ml of 0.1 M glycine-HCl and immediately neutralized with 1 M Tris-150 mM NaCl [pH 8.01. Eluate was exchanged into Dulbecco’s PBS and concentrated 4.6-fold over starting serum volume. The recovery of inhibin was monitored by use of the immunoradiometric assay (IRMA) to measure rhinhibin in spiked serum before and after various procedural steps. The procedure was repeated with three 20-ml aliquots of control serum. Isolated immunoreactive inhibin-like material from the three serum aliquots was pooled, exchanged, and concentrated 6.1-fold over starting serum volume. RESULTS IRMA. Parallel dose-response curves were obtained with rh-inhibin and a preparation of pFF (Figure 1). Cross-reaction with rh-activin (l-20 ng/tube) and rh-follistatin (l-20 ngltube) was nondetectable. Sensitivity was 50 pgltube, and the recovery of rh-inhibin (0.02-10 ng) added to 1.4 ml of porcine serum was 73 + 4%. Intra-assay and interassay coefficients of variation were 5.5 and 7.0%, respectively. Binding Proteins. The incubation of [125I]rh-inhibin with serum and with 02macroglobulin resulted in two chromatographic peaks of radioactivity (Figure 2). The first eluted in the void volume (>600 kDa) and comprised 20-25% of the recovered radioactivity. The second eluted in a volume similar to that in which [1251]rh-inhibin in buffer eluted. The treatment of [‘?]rh-inhibin-spiked serum and a,-macroglobulin with 8 M urea resulted in a single chromatographic peak of radioactivity. The peak eluted in a volume similar to that of [‘251]rh-inhibin in urea (Figure 2). Isolation of Inhibin. Radiolabeled rh-inhibin eluted with the latter portion of serum proteins (Figure 3). The recovery of [‘25Ilrh-inhibin after chromatography was 48%. Fractions from the elution of the serum aliquot that had been spiked with nonradiolabeled rh-inhibin were pooled beginning midway through the serum protein peak and continuing 75 ml beyond, as shown in Figure 3. Fractions from the elution of the control serum aliquot were pooled likewise. The recovery of nonradiolabeled rh-inhibin was 46% after

506

MEYER AND WHEATON

100,

c v

E 8>r IO,

z >

‘Z tu

.-0 :

tY

l-

0---L

I

0.01

’ “““‘I

’ “““‘I

0.10 rh-lnhibin

’ “““‘I

’ “““‘I

1.oo 10.0 100 (ng) or inhibin preparation

’ “““‘I

’

1000 (~1)

Figure 1. Inhibin-binding curves in the IRMA. Curves presented are for rh-inhibin in buffer (0), rh-inhibin in unextracted porcine serum (O), a reactive red agarose partially purified preparation of pFF (*), rh-inhibin incubated with porcine serum and subjected to inhibin isolation procedures (A), and porcine serum subjected to inhibin isolation procedures (A).

chromatography, exchange into gel-NaPO,, and concentration. The immunoaffinity adsorption step had a recovery factor of 69%, and the exchange into Dulbecco’s PBS and further concentration had a recovery factor of 87%. The cumulative recovery of rh-inhibin was 28%. Dimeric inhibin immunoreactivities isolated from the rh-inhibin-spiked and control serum aliquots are presented in Figure 1; that in the latter was nondetectable. Thus, factoring assay sensitivity (50 pg/tube), recovery (28%) and serum equivalents assayed (6.44 ml = 1.4-ml sample volume assayed x 4.6 serum concentration factor), concentrations of dimeric inhibin in sows during random stages of the estrous cycle were ~28 pg/ml. The concentration of dimeric inhibin isolated from serum collected 2 d postweaning also was nondetectable. The threshold concentration for detection was 21 pg/ml. DISCUSSION Dimeric inhibin in porcine serum was not detected. The IRMA did detect rh-inhibin that had been added to porcine serum. It also recognized dimeric forms of porcine inhibin based on the similar dose-response curves produced by rh-inhibin and a partially purified preparation of pFF. Concentrations of dimeric inhibin in porcine serum were evidently below the 21 pg/ml detection limit of the IRMA. Such concentrations would be well below the RIA inhibin values reported previously for sows (1,2). An analogous disparity between RIA and two-site immunoassay estimates of inhibin concentrations in human and rat sera has been attributed to extensive RIA cross-reaction with or-inhibin monomers (8-10). o-Inhibin monomers have been identified in human serum and bovine plasma (18,19). The sensitivity of the IRMA was similar to that of other two-site immunoassays (7,20). Groome and O’Brien (21) have recently reported the development of an ultrasensitive two-site immunoassay capable of detecting 2 pg/ml rh-inhibin. To date, that assay has been validated only for human serum (personal communication). It appears that an equally sensitive assay also will be needed to determine concentrations of circulating inhibin during the estrous cycle of the sow.

DIMERIC INHIBIN AND BINDING

10

15

PROTEINSIN

20

PORCINE

25

30

507

SERUM

35

40

Milliliters Figure 2. Elution profiles of [rZ51]rh-inhibin. The upper panel depicts profiles of [ “sI]rh-inhibin after incubation with buffer (0). porcine serum (0). and u,-macroglobulin (*). Preparations were chromatographed with a 1 x 50 cm column of Sephadex G-200-120 eluted with gel-NaPO,. The flow rate was 0.06 mUmin, and the void volume was 13-17 ml. The lower panel shows profiles of the same incubates in 8 M urea-Tris buffer solution. Chromatographic conditions differed from above in that Superdex 200 and 8 M urea-Tris buffer were used for medium and elution, respectively. These conditions reduced the volume in which rh-inhibin eluted. This was attributed to differences in gel filtration media and to an apparent increase in the molecular mass of rh-inhibin in 8 M urea.

The lack of detection of dimeric inhibin in porcine serum may reflect the presence of a nondetectable form rather than low concentrations. Multiple molecular weight forms ranging from 28 to 120 kDa have recently been isolated from human plasma (22). Higher molecular weight forms may have been excluded during the inhibin isolation procedure. Another possibility is that inhibin-B may be the major circulating form of inhibin. The extent to which the IRMA recognizes inhibin-B is unknown. This possibility seems unlikely, however, because concentrations of inhibin PA-subunit mRNA have been detected in porcine follicles and have been shown to change during the estrous cycle (23). Inhibin-B may be the predominant form of inhibin in males (24). Groome et al. (25) reported that men’s plasma had an inhibin-B concentration of 80 pg/ml, whereas concentrations of inhibin-A were nondetectable. Inhibin-binding proteins are present in human and rat sera and, on the basis of the results presented here, in porcine serum as well. A portion of [‘251]rh-inhibin added to

508

MEYER AND WHEATON

IOOr4

80-

6040-

20-

400

500

600 700 Milliliters

800

900

Figure 3. Elution patterns of serum proteins and [ ‘Z51]rh-inhibin in 8 M urea-Tris buffer. Porcine serum that had been spiked with [‘*‘I]rh-inhibin was treated with urea and then chromatographed with a 5 x 50 cm column of Superdex 200 eluted with 8 M urea-Tris buffer. The flow rate was 1.1 ml/min. Absorbance was monitored at 280 nm (solid line), and fractions were counted for radioactivity (dashed line). Fractions that were subsequently pooled are indicated by the bar.

porcine serum complexed with high-molecular-weight species. In human and rat sera, the primary inhibin-binding protein is cu,-macroglobulin. cu,-Macroglobulin is a 7 18-kDa tetrameric protein that binds inhibin, activin, and growth factors (11,26). Human and porcine a,-macroglobulin are similar structurally and circulate at concentrations of 2-4 mg/ml (27,28). The incubation of [ ‘?]rh-inhibin with c-u,-macroglobulin produced an elution profile much like that after the incubation of [ ‘251]rh-inhibin with porcine serum. These results are consistent with ar,-macroglobulin serving as an inhibin-binding protein in porcine serum. The recovery of rh-inhibin added to porcine serum was reduced 27% compared with that added to buffer. The loss matched the proportion of [“‘I]rh-inhibin that associated with serum-binding proteins and a,-macroglobulin. Muttukrishna et al. (29) reported that a,-macroglobulin resulted in a 14% underestimation of inhibin levels. Complexes of [‘251]rh-inhibin-binding protein were dissociated in 8 M urea. Urea was used in an original inhibin purification protocol to disrupt noncovalent protein-protein interactions (30). In summary, serum concentrations of mature dimeric inhibin-A in sows are less than 21 pg/ml, lower than previously reported RIA estimates. An inhibin-binding protein is present in porcine serum that is consistent with a,-macroglobulin. ACKNOWLEDGMENTS/FOOTNOTES ’This work was supported by MGAR funds, Project 0302.48 16-75, and by NIH Grant 1RO 1-HD28426-01 A 1. Published as Contribution No. 22,347 of the scientific journal series of the Minnesota Agric. Exp. Sta. ’ Correspondence: J.E. Wheaton, Department of Animal Science, 495 Animal Science/Veterinary Medicine Building, 1988 Fitch Ave., University of Minnesota, St. Paul, MN 55108; FAX: (612) 625-2743. Appreciation is expressed to the USDA Animal Hormone Program, the NIDDK and the National Hormone and Pituitary program for RIA preparations and rh-follistatin. The authors also thank J.P. Mather, Genentech, Inc., South San Francisco, for rh-inhibin and Y. Eto (Kawasaki, Japan) for rh-activin.

DIMERIC INHIBIN AND BINDING

PROTEINS

IN PORCINE

SERUM

509

REFERENCES 1. Hasegawa Y, Miyamoto K, Iwamura S, Igarashi M. Changes in serum concentrations of inhibin in cyclic pigs. J Endocr 118:211-219, 1988. 2. Trout WE, Killen JH, Christenson RK, Schanbacher BD, Ford JJ. Effects of weaning on concentrations of inhibin in follicular fluid and plasma of sows. J Reprod Fert 94: 107-l 14, 1992. 3. Robertson DM. The measurement of inhibin. Reprod Fert Dev 2:101-105, 1990. ewes and serum albumin: 4. Bolt DJ, Caldwell DW. Inhibin-like activity in plasma from ovariectomized Pitfalls for radioimmunoassay. Domest Anim Endocrinol 957-69, 1992. immunosorbent assay for inhibin. Biol Reprod 45: 5. Betteridge A, Craven RP. A two-site enzyme-linked 748-754, 1991. 6. Groome, N. Ultrasensitive two-site assays for inhibin-A and activin-A using monoclonal antibodies raised to synthetic peptides. J Immunol Methods 145:65-69, 1991. 7. Baly DL, Allison DE, Kmmmen LA, Woodruff TK, Soules MR. Chen SA, Fendly BM, Bald LN, Mather JP, Lucas C. Development of a specific and sensitive 2-site enzyme-linked immunosorbent assay for measurement of inhibin-A in serum. Endocrinology 132:2099-2108, 1993. 8. Fahy PA, Wilson CA, Beard AJ, Groome NP, Knight PG. Changes in inhibin-A (o-PA dimer) and total cr-inhibin in the peripheral circulation and ovaries of rats after gonadotrophin-induced follicular development and during the normal oestrous cycle. J Endocr 147:271-283, 1995. 9. Knight PG. Muttukrishna S, Groome N, Webley GE. Evidence that most of the radioimmunoassayable inhibin secreted by the corpus luteum of the common Marmoset monkey is of a nondimeric form. Biol Reprod 47:554-560, 1992. 10. Groome NP, Illingworth PI, Obrien M, Cooke I, Ganesan TS, Baird DT, McNeilly AS. Detection of dimeric inhibin throughout the human menstrual cycle by two-site enzyme immunoassay. Clin Endocrinol 40:717723, 1994. 11. Vaughan JM, Vale, WW. cY,-Macroglobulin is a binding protein of inhibin and activin. Endocrinology 132:2038-2050, 1993. 12. Krummen LA, Woodruff TK, Deguzman G, Cox ET, Baly DL, Mann E, Garg S, Wong WL, Cossum P, Mather JP. Identification and characterization of binding proteins for inhibin and activin in human serum and follicular fluids. Endocrinology 132:477479, 1993. 13. Rivier J, Spiess J, McClintock R, Vaughan J, Vale W. Purification and partial characterization of inhibin from porcine follicular fluid. Biochem Biophys Res Commun 133:12&127, 1985. 14. Redmer DA, Christenson RK, Ford JJ, Day BN, Goodman AL. Inhibin-like activity in ovarian venous serum after unilateral ovariectomy in prepubertal gilts. Biol Reprod 34:357-362, 1986. 15. Groome N, Lawrence M. Preparation of monoclonal antibodies to the beta A subunit of ovarian inhibin using a synthetic peptide immunogen. Hybridoma 10:309-316, 1991. 16. Meyer RL, Carlson KM, Rivier J, Wheaton, JE. Antiserum to an inhibin alpha-chain peptide neutralizes inhibin bioactivity and increases ovulation rate in sheep. J Anim Sci 69:747-754, 1991. 17. Vaughan JM, Rivier J, Conigan AZ, McClintock R, Campen CA, Jolley D, Voglmayr JK, Bardin CW, Rivier C, Vale W. Detection and purification of inhibin using antisera generated against synthetic peptide fragments. In: Methods in Enzymology, Vol. 168, Conn PM (ed.). Academic Press, Orlando, p. 588-617, 1989. 18. Schneyer AL, Mason AJ, Burton LE, Ziegner JR, Crowley WF Jr. Immunoreactive inhibin a-subunit in human serum: implications for radioimmunoassay. J Clin Endocrinol Metab 70:1208-1212, 1990. 19. Knight PG, Beard AJ, Wrathall JHM, Castillo RJ. Evidence that the bovine ovary secretes large amounts of monomeric inhibin cx subunit and its isolation from bovine follicular fluid. J Mol Endocrinol 2:189-200, 1989. 20. Knight PG, Muttukrishna S. Measurement of dimeric inhibin using a modified two-site immunoradiometric assay specific for oxidized (Met 0) inhibin. J Endocr 141:417425, 1994. 21. Groome N, O’Brien M. Immunoassays for inhibin and its subunits; further applications of the synthetic peptide approach. J Immunol Methods 165:167-176, 1993. 22. Robertson DM, Sullivan J, Watson M, Cahir N. Inhibin forms in human plasma. J Endocr 144:261-269, 1995. 23. Guthrie HD, Rohan RM, Rexrod CE Jr, Cooper BS. Changes in concentrations of follicular inhibin (Yand BA subunit messenger ribonucleic acids and inhibin immunoreactivity during preovulatory maturation in the pig. Biol Reprod 47:1018-1025, 1992. 24. de Jong FH, de Winter JP, KIaij IA, Themmen APN, Wesseling JG. Inhibin, activin and activin receptors in the testis: Localization, effects and the activin paradox. In: Inhibin and Inhibin-Related Proteins, Burger HG (ed.). Ares-Serono Symposia Publications, Rome, p. 233-244, 1994.

510

MEYER AND WHEATON

2.5. Groome NP, O’Brien M, Illingworth P, Pai R, Mather J, Priddle J, McNeilly AS. Inhibin-B is a major circulating form of inhibin in men and women. 77th Annual Meeting of The Endocrine Society, Washington, D.C. p. 103 (Abstract OR42-l), 1995. 26. Niemuller CA, Randall KJ, Webb DJ, Gonias SL, Lamarre J. n,-Macroglobulin conformation determines binding affinity for activin A and plasma clearance of activin A rqmacroglobulin complex. Endocrinology 1365343-5349, 1995. 27. Tsuru D, Tomimatsu M, Fujiwara K, Kawahara K. A neutral subtilopeptidase inhibitor from porcine serum: some evidence for a,-macroglobulin. J Biochem (Tokyo) 77: 1305-l 3 12, 1975. 28. Starkey PM. The evolution of human alpha 2-macroglobulin: Further evidence for a common ancestry with the complement proteins. Ann NY Acad Sci 421:112-l 18, 1983. 29. Muttukrishna S, Fowler PA, Groome NP, Mitchell GG, Robertson WR, Knight PG. Serum concentrations of dimeric inhibin during the spontaneous human menstrual cycle and after treatment with exogenous gonadotrophin. Hum Reprod 9:1634-1642, 1994. 30. Miyamoto K, Hasegawa Y, Fukuda M, Nomura M, Igarashi M, Kangawa K, Matsuo H. Isolation of porcine follicular fluid inhibin of 32 K daltons. Biochem Biophys Res Commun 129:39@lO3, 1985.