different roles of prepubertal and postpubertal …...sections (4 mm) were mounted on superfrost...

Different Roles of Prepubertal and Postpubertal GermCells and Sertoli Cells in the Regulation of SerumInhibin B Levels*

ANNA-MARIA ANDERSSON, JØRN MULLER, AND NIELS E. SKAKKEBÆK

Department of Growth and Reproduction, University of Copenhagen, DK-2100 Copenhagen, Denmark

ABSTRACTTo elucidate the role of germ cells in the regulation of inhibin B

secretion, serum inhibin B levels in prepubertal boys and adult menwhom had a concurrent testicular biopsy showing either normal orimpaired testicular function were compared. In addition, by immu-nohistochemistry the cellular localization of the two subunits of in-hibin B (a and bB) were examined in adult testicular tissue withnormal spermatogenesis, spermatogenic arrest, or Sertoli cell onlytubules (SCO) as well as in normal testicular tissue from an infant anda prepubertal boy. Adult men with testicular biopsy showing normalspermatogenesis (n 5 8) or spermatogenic arrest (n 5 5) had medianinhibin B levels of 148 pg/mL (range, 37–463 pg/mL) and 68 pg/mL(range, 29–186 pg/mL), respectively, corresponding to normal or near-normal levels of our reference population (165 and 31–443 pg/mL; n 5358). Men with SCO (n 5 9) had undetectable or barely detectable (n 51) serum levels of inhibin B. In contrast to adults, prepubertal boyswith SCO (n 5 12) all had measurable serum inhibin B levels thatcorresponded to our previously determined normal range in healthyprepubertal boys (n 5 114). However, in postpubertal samples from

the same SCO boys, inhibin B levels were undetectable as in the adultSCO men. Intense inhibin a-subunit immunostaining was evident inSertoli cells in both prepubertal and adult testes. In the prepubertaltestis, positive immunostaining for the bB-subunit was observed inSertoli cells. In the adult testis, intense immunostaining for the bB-subunit was evident in germ cells from the pachytene spermatocyteto early spermatid stages and to a lesser degree in Leydig cells, butnot in Sertoli cells or other stages of germ cells. Thus, surprisingly,in adult men the two subunits constituting inhibin B were expressedby different cell types. We speculate that during puberty Sertoli cellmaturation induces a change in inhibin subunit expression. Thus,immature Sertoli cells express both a and bB inhibin subunits,whereas fully differentiated Sertoli cells only express the a-subunit.The correlation in adult men between serum inhibin B levels andspermatogenesis may be due to the fact that inhibin B in adult menis possibly a joint product of Sertoli cells and germ cells, including thestages from pachytene spermatocytes to early spermatids. (J ClinEndocrinol Metab 83: 4451–4458, 1998)

THE GONADAL peptide hormone inhibin, which exertsnegative feedback regulation of pituitary production

and secretion of FSH, exists in two principal bioactive forms:the inhibin A dimer, consisting of an a-subunit and a bA-subunit, and the inhibin B dimer, consisting of an a-subunitand a bB-subunit (1). The subunits are covalently linked bya disulfide bond. By the implementation of a new immuno-assay format that made it possible to measure these twobioactive inhibin forms specifically, it has been shown thatinhibin B is the principal circulating bioactive inhibin formin men (2). Activins are structurally related dimers that arecomposed of two b-subunits and that can stimulate FSHrelease. Inhibin B is presumed to be produced primarily bythe Sertoli cells of the testis based on in vitro studies (3–6) andthe localization of inhibin subunit peptide and messengerribonucleic acid in testicular tissue from rats (7–9), primates(10), and humans (9) by immunohistochemistry and in situhybridization. However, the regulation of inhibin B secretionis still largely unknown. Studies of the role of gonadotropinsor androgens in inhibin regulation are equivocal. Several

studies indicate that the presence of specific germ cell typesmay stimulate the production and secretion of inhibin asshown in the rat (11–14) and indicated in men with impairedspermatogenesis, who generally have very low or undetect-able serum inhibin B levels (15). On the other hand, we haverecently shown that prepubertal boys, who do not have sper-matogenesis and therefore lack the relevant germ cell types,secrete inhibin B in readily measurable levels (16, 17). To furtherelucidate the role of germinal cells in the regulation of inhibinB secretion we measured serum inhibin B levels in men witheither normal ongoing spermatogenesis, spermatogenic arrest,or Sertoli cell only syndrome (SCO) based on a testicular biopsy.In addition, we measured serum inhibin B levels in boys andadolescents previously treated for acute lymphoblastic leuke-mia (ALL), some of whom became SCO as a result of thetreatment. The expression and cellular localization of inhibin a-and bB-subunits were investigated using immunohistochem-istry in testicular biopsies from adult men with normal or ab-normal spermatogenesis and from prepubertal boys.

Based on our observations, we hypothesize that inhibin Bin adult men may be produced by Sertoli cells and sper-matocytes in collaboration.

Subjects and MethodsSubjects

Prepubertal boys. Simultaneous testicular biopsies and blood sampleswere available from 34 boys previously treated for ALL, who had tes-ticular biopsy performed for screening for lymphoblastic infiltration. Of

Received June 23, 1998. Revision received August 27, 1998. AcceptedSeptember 17, 1998.

Address all correspondence and requests for reprints to: Anna-MariaAndersson, M.Sc, Ph.D., Department of Growth and Reproduction, sec-tion GR 5064, Copenhagen University Hospital, Rigshospitalet, Bleg-damsvej 9, DK-2100 Copenhagen Ø, Denmark. E-mail: [email protected].

* This work was supported by the Danish Research Councils (Grant9700833).

0021-972X/98/$03.00/0 Vol. 83, No. 12Journal of Clinical Endocrinology and Metabolism Printed in U.S.A.Copyright © 1998 by The Endocrine Society

4451

these boys, 12 acquired a SCO pattern as a consequence of their treat-ment, 13 boys had testes with normal numbers of germinal cells, and 9boys had testes with reduced numbers of germinal cells after terminationof treatment for ALL. From 17 boys, longitudinal blood samples andtesticular biopsies were available, and from 6 of these boys postpubertalblood samples were also available. For immunocytochemistry, testiculartissue was obtained at autopsy from 2 boys, aged 4.5 months and 7 yr,who had died in sudden accidents.

Adult men. Concurrent testicular biopsies and blood samples were ob-tained from men referred to our clinic for infertility and from testicularcancer patients who had biopsy performed as screening for carcinomain situ in the contralateral testis. Based on the testicular biopsy and semenparameters, men were grouped according to the presence of normalspermatogenesis (n 5 8), spermatogenic arrest (n 5 5), or SCO syndrome(n 5 9). The men with normal spermatogenesis included five men withobstruction of the efferent ducts, two testicular cancer patients with oneremaining normal testis after unilateral orchidectomy, and one patientwith normal spermatogenesis in one testis and spermatogenic arrest ofthe other. Men with spermatogenic arrest were defined as having bi-lateral spermatogenic arrest and azoospermia. The SCO men were de-fined as having SCO and azoospermia and included four men withidiopathic bilateral SCO and five unilaterally orchidectomized men withSCO in the remaining testis due to irradiation for carcinoma in situ.

Serum samples and testicular biopsies

Serum was separated after clotting and was stored at 220 C untilhormone measurements. Adult serum samples were measured withina few weeks from the time of sampling, whereas the time of storage ofserum samples from ALL treated boys varied from 15 yr to a few months.Serum inhibin B was measured in an enzyme immunometric assay,previously described (18). This assay is specific for the bioactive inhibinB dimer (a-bB). The sensitivity of the inhibin B assay was 18 pg/mL, andthe intra- and interassay coefficients of variation were less than 12% andless than 17%, respectively.

Testicular biopsies of approximately 2 3 2 mm (boys) or 3 3 3 mm(adult men) were obtained by open surgery and fixed in Stieve’s fixative.For evaluation of spermatogenic status, slides were stained with hema-toxylin-eosin before evaluation by light microscopy.

Testicular biopsy and blood sampling in boys treated for ALL asscreening for testicular relapse was approved by the local ethical board.In the adult subjects, testicular biopsies and blood samples were per-formed as part of a clinical routine evaluation. The patients had giveninformed consent to the procedures.

Immunohistochemistry

Sections (4 mm) were mounted on SuperFrost Plus slides (Menzel,Braunschweig, Germany) and dried overnight at 50 C. Before incubationwith primary antibody, sections were dewaxed, rehydrated in gradedethanol, and washed in water and 0.05 mol/L Tris-HCl (pH 7.4) and 0.15mol/L NaCl (TBS). Sections were subjected to antigen retrieval by mi-crowaving in 0.01 mol/L citrate buffer (pH 5.5) on full power (750 watts)for 2 min and on 40% power for an additional 18 min. Sections wereallowed to cool at room temperature for 20 min before being washed for

5 min in TBS and subsequently incubated in 1% H2O2 in TBS for 30 minto block endogenous peroxidase. Purified monoclonal antibodies di-rected against a-subunit (19) (code R1) were used at a concentration of2 mg/mL, and two different antibodies against the bB-subunit (18) (codeC5 and 12/13) were used at concentrations of 25 and 5 mg/mL, respec-tively. The C5 and 12/13 antibodies directed against the bB-subunit had0.5–1% cross-reactivity with the bA-subunit (18). Primary antibodieswere diluted in TBS and incubated on sections in a humid chamberovernight at 4 C. The following day sections were washed three timesin TBS (3 min/wash), incubated with biotinylated goat antimouse Igs(Zymed histostain kit, Zymed Laboratories, Inc., San Francisco, CA) for30 min, and then washed again three times in TBS. For detection ofbound antibodies, sections were subsequently incubated with peroxi-dase-conjugated streptavidin-biotin complex (Zymed Histostain) for 30min, followed by washing three times in TBS. Color was developed inan aminoethyl carbazole substrate solution (Zymed Histostain kit). After6–10 min of color development, sections were washed in tap water,counterstained with hematoxylin, and mounted in glycerol vinyl alcohol(Zymed Laboratories, Inc.). On sections from prepubertal testes an al-ternative immunostaining procedure was also used. In the alternativeprocedure the second layer of antibody was exchanged with unlabeledrabbit antimouse antibody (Dako Corp., Glostrup, Denmark) diluted1:50 in TBS, and the third layer was exchanged with alkaline phospha-tase-antialkaline phosphatase complex (Dako Corp.) diluted 1:25 in TBS.After a final wash in 0.05 mol/L Tris-HCl (pH 8.7), color was developedin a new fuchsin substrate solution (Sigma, St. Louis, MO). The speci-ficity of the antibodies was controlled by using normal mouse seruminstead of primary antibodies and by preabsorbing the antibodies withthe corresponding peptide.

Statistics

Differences in inhibin B levels among men with normal spermato-genesis, spermatogenic arrest, and SCO were tested by the Mann-Whit-ney U test. The influence of storage on serum inhibin B levels was testedby linear regression on data that were square root transformed to obtaina good approximation to the normal distribution.

ResultsSerum inhibin B levels in boys and adult men

The median and range of serum inhibin B levels within thedifferent groups are summarized in Table 1. For comparison,serum inhibin B levels in a normal population of 114 pre-pubertal boys and 358 adult men, which have been describedpreviously (17), are also included in the table.

Serum inhibin B levels in samples from boys treated forALL with either normal testicular pattern or SCO are plottedaccording to age in relation to normal ranges in Fig. 1. Serafrom boys treated for ALL had been stored at 220 C from afew weeks up to 15.3 yr (median, 4.8 yr) before being ana-lyzed for inhibin B. Because of the large variation in length

TABLE 1. Serum inhibin B levels (median and range) in prepubertal boys and adult men with normal or impaired testicular function

Prepubertal boys treatedfor ALL with normal

testicular pattern

Prepubertal boys treatedfor ALL with SCO

pattern

Normal population ofprepubertal boysa

Serum inhibin B (pg/mL) 68 (53–131) 80 (34–215) 64 (,18–258)Age (yr) 7.3 (2.0–11.1) 10.0 (2.8–13.2) (5–10)n 16 12 114

Adult men with normalspermatogenesis

Adult men withspermatogenic arrest

Adult men withSCO pattern

Normal population ofadult mena

Serum inhibin B (pg/mL) 148 (37–463) 68 (29–186) ,18 (,18–22) 165 (31–443)Age (yrs) 36 (25–47) 32 (28–39) 33 (23–64) (25–50)n 8 5 9 358a Normal reference ranges were previously published (17).

4452 ANDERSSON, MULLER, AND SKAKKEBÆK JCE & M • 1998Vol 83 • No 12

of storage, the long term storage stability of inhibin B had tobe considered. There was no time trend in prepubertal in-hibin B levels when plotted according to storage time (r2 50.005; P 5 0.638), indicating that inhibin B was relativelystable during long term storage. Furthermore, serum inhibinB levels in prepubertal boys treated for ALL with normaltestis pattern were all within the normal range for prepu-bertal boys, as were the levels in prepubertal boys with SCO.In all postpubertal samples from boys who acquired a SCOpattern, serum inhibin B levels were undetectable (Fig. 1).

Serum inhibin B levels in the group of men with testicularbiopsies showing normal spermatogenesis, including thethree men who only had one testis with normal spermato-genesis, were all within the normal range for adult men.Serum inhibin B levels in the group of men with spermat-ogenic arrest were in the low normal range. However, therewas no statistically significant difference between the levelsin men with normal spermatogenesis and those in men withspermatogenic arrest (by Mann-Whitney U test). In contrast,all men with SCO, except one, had unmeasurable serumlevels of inhibin B.

Expression of inhibin B subunits in human prepubertal andadult testes

In prepubertal testes, intense positive immunostaining forinhibin a-subunit was observed that was localized to Sertoliand interstitial cells (Fig. 2A). Positive immunostaining forinhibin bB-subunit was also observed in Sertoli cells of pre-pubertal testes (Fig. 2B). In adult testes with ongoing sper-matogenesis, positive immunostaining for inhibin a-subunitwas observed in both Sertoli cells and Leydig cells (Fig. 3, Aand B). A similar a-subunit immunostaining pattern wasobserved in testes with spermatogenic arrest (Fig. 4A) and in

testes with SCO (Fig. 4B). In adult testes with normal sper-matogenesis and with spermatogenic arrest, intense immu-nostaining for the inhibin bB-subunit was located in germcells from pachytene spermatocytes to round spermatids(Figs. 3, C and D, and 4C). An identical staining pattern wasobserved using two different monoclonal anti-bB antibodies.In contrast, no staining for the inhibin bB-subunit was ob-served in Sertoli cells; spermatogonia; preleptotene, lepto-tene, or zygotene spermatocytes; or late spermatids (fromtypes Sb–Sd). However, Leydig cells were also positive forbB-subunit immunostaining. In adult testes with SCO, nointratubular immunostaining for the bB-subunit was ob-served (Fig. 4D). Occasionally, faint bB-subunit immuno-staining was observed in a few Leydig cells, but, in general,Leydig cells were also negative for bB-subunit staining inSCO testes.

Discussion

In our study in which serum inhibin B levels could bedirectly correlated to the testicular status based on a con-current testicular biopsy, we have demonstrated that theregulation of inhibin B secretion changes during puberty.Whereas serum inhibin B levels in prepubertal boys wereindependent of the presence of germinal cells, serum inhibinB levels in postpubertal men were closely related to thepresence of germinal cells from pachytene spermatocytes toearly spermatids. In the rat, inhibin production has also beencorrelated to the presence of certain germ cell types (11–14),although in this species inhibin production, as measured inan a-subunit assay, was highest in the presence of latespermatids.

Human adult serum inhibin B levels, however, were notdependent on complete spermatogenesis, as serum inhibin B

FIG. 1. Serum inhibin B levels in prepubertal boys treated for ALL plotted against age. Actual testicular status at time of sampling is indicatedby the marks as defined in the legend. The gray area (5th and 95th percentiles) indicates reference levels in normal healthy boys (16). The linesconnect longitudinal samples in the same individual.

ROLE OF GERM CELLS IN REGULATION OF INHIBIN B 4453

levels were measurable in all five men with bilateral sper-matogenic arrest at the spermatocyte level, although the in-hibin B levels tended to be lower than those observed in anormal male population, including men with documentedcomplete spermatogenesis. Furthermore, levels of serum in-hibin B in men with only one normal functioning testis werewell within the normal range, indicating that the remainingtestis after unilateral orchidectomy may be able to compen-sate for the missing testis with respect to inhibin B secretion.

The regulation of inhibin B secretion in prepubertal boysis not clear. The localization of both a- and bB-subunits tointerstitial and Sertoli cells in the prepubertal human testisobserved in this study is in agreement with previous findingsin the rat (8, 9) and human (9). We recently showed thatserum inhibin B levels are temporarily highly elevated post-natally, presumably as a consequence of the temporary ac-tivation of the pituitary shortly after birth (17), as also shownpreviously (20), indicating that inhibin B secretion in thenewborn is under the influence of gonadotropins. Later inchildhood, gonadotropin levels are low, but perhaps suffi-cient to sustain the readily measurable serum levels of in-hibin B observed throughout childhood (16). Thus, seruminhibin B levels in hypogonadotropic boys are in the lowrange of healthy boys (21), although this may be due to areduced number of Sertoli cells rather than to decreasedinhibin B production caused by low FSH levels. Comparedto the staining intensity observed for the bB-subunit in theadult testis, the staining in prepubertal testes was relativelyweak. This was also true in the testis from a boy aged 4.5months, although this is an age when serum inhibin B levelsis known to be increased above adult levels (17).

Several studies showing localization of the inhibin a-sub-unit in human testis exist that, in general, agree on the lo-calization of this subunit to the Sertoli cells and interstitial/Leydig cells in adult testes (22–25). Inhibin bB-subunitexpression has previously been localized to Sertoli cells in theadult rat (8, 9) and monkey (10). Recently, a study on theimmunolocalization of the inhibin/activin bB-subunit in theadult human testis using the same antibodies as in ourpresent study was published that showed colocalization ofthe a- and bB-subunits in the Sertoli cells (26). In contrast tothese findings, we found that the bB-subunit was localizedto pachytene spermatocytes and round spermatids as well asto Leydig cells, whereas the a-subunit was localized to Sertolicells and Leydig cells in the adult human testis. We do notknow the reason for this discrepancy, but it may be due todifferences in the fixation and processing of the tissue. Thefacts that we were able to eliminate our bB-subunit stainingpattern by preabsorbing the antibody with a bB peptide, thatthe staining pattern was highly reproducible in several dif-ferent tissue section from different men, and that our bB-subunit staining was restricted to selected germ cell typeswith no or negligible background staining of other cell typesstrongly indicate that our bB-subunit staining was specific.However, as both subunits are needed to produce inhibin B,our findings raises the question of where circulating inhibinB is produced in the adult human testis. As both subunits arepresent in Leydig cells, it cannot be excluded that these cellscontribute to the circulating levels of inhibin B. However,studies in rats have shown that the major route via which

FIG. 2. Cellular localization of inhibin a- and inhibin/activin bB-subunits in 4.5-month-old postnatal human testis. Strong immuno-staining for inhibin a-subunit was observed in both interstitial andSertoli cells (A). A weak, but positive, immunostaining for the inhibin/activin bB-subunit could be observed in Sertoli cells (B) when themonoclonal anti-bB antibody code C5 was used. Immunostaining withthe monoclonal anti-bB antibody code 12/13 resulted in an identicalstaining pattern (not shown). Immunostaining with preimmunemouse serum (C) or preabsorbed antibodies (not shown) did not resultin any positive staining. Magnification, 3250. The staining patternfor the a- and bB-subunits in the 4.5-month-old testis was similar tothe pattern observed in the 7-yr-old testis (not shown).


inhibin reaches the bloodstream is through secretion into theseminiferous tubular fluid, thus supporting the idea of anintratubular production of inhibin (27). Alternatively, it may

be speculated that the Sertoli cells express the inhibin bB-subunit at levels below the sensitivity of our immunohisto-chemistry.

FIG. 3. Cellular localization of inhibin a- and inhibin/activin bB-subunits in adult human testis with normal spermatogenesis. Strong im-munostaining for inhibin a-subunit was observed in both Leydig and Sertoli cells (A, 3250; B, 3630). Immunostaining for inhibin/activinbB-subunit with the anti-bB antibody code C5 was localized to pachytene spermatocytes, round spermatides, and Leydig cells (C, 3250; D, 3630).Immunostaining with the monoclonal anti-bB antibody code 12/13 resulted in an identical staining pattern (not shown). Immunostaining withpreimmune mouse serum (E, 3250) or with preabsorbed antibodies (not shown) did not result in any positive staining.


FIG. 4. Cellular localization of inhibin a- and inhibin/activin bB-subunits in adult human testis with spermatogenic arrest and SCO. Strongimmunostaining for inhibin a-subunit was observed in Leydig and Sertoli cells in both testes with spermatogenic arrest (A) and those with SCO(B). Immunostaining for inhibin/activin bB-subunit in testes with spermatogenic arrest (C) was similar to the staining observed in testis withcomplete spermatogenesis, with localization of the staining to pachytene spermatocytes, round spermatides, and Leydig cells. SCO testes weregenerally negative for staining for the bB-subunit (D), although occasionally faint bB-subunit staining could be observed in a few Leydig cells.Immunostaining with preimmune mouse serum [spermatogenic arrest (E) and SCO (F)] or with preabsorbed antibodies (not shown) did notresult in any positive staining. Magnification, 3250.


The site of dimerization of inhibin subunits is not known.By electron microscopy, evidence of transfer of inhibina-subunit from Sertoli cells to spermatocytes has been re-ported in the human testis (24). However, although inhibinsubunit dimerization may take place intracellularly in thespermatocytes, it could also theoretically take place extra-cellularly. Creation of a disulfide bond from the sulfhydrylgroups of two cystein residues is a passive reaction that doesnot require energy and that takes place in a nonreducingenvironment, such as in the endoplasmic reticulum, Golgi,coated vesicles, or even extracellular fluid. Manson et al. haveshown that the propeptide sequences present in the b-sub-unit precursors are necessary for dimerization, presumablyby providing the proper configuration for the involved cys-tein residues to be exposed (28, 29). However, several studieshave shown that a high proportion of the inhibin forms inplasma consist of unprocessed or partly processed precursordimers (30, 31), indicating that the structural elements nec-essary for dimerization of subunits are still present aftersecretion. Further processing of inhibin/activin precursorsinto the bioactive hormone by cleavage of the proproteinregion can also take place extracellularly (32).

We speculate that in the adult human testis, inhibin B maybe a joint product of Sertoli cells and spermatocytes, andhence the change in localization of the bB subunit duringmaturation may explain why serum inhibin B levels are germcell dependent in the mature testis and not in the immaturetestis in man. Furthermore, our hypothesis that inhibin sub-unit dimerization and processing may take place postsecre-tionally may explain the abundance of non- or partly pro-cessed as well as monomeric forms of inhibin in serum. Thefive men with spermatogenic arrest who were included inthis study all had spermatogenic progress that included thelevel of pachytene spermatocytes and all had measurableserum inhibin B levels. The somewhat lower serum inhibinB levels in these men could be explained by the fact that, ingeneral, they have fewer spermatocytes beyond the pachy-tene stage and no spermatids. If, as we believe, bB-subunitexpression by pachytene spermatocytes and round sperma-tids correlates with serum inhibin B levels, we would expectmen with spermatogenic arrest at a level before thepachytene stage to have very low or unmeasurable seruminhibin B levels.

bB-subunits may also bind together as homodimers tomake activin B. Thus, the expression of the bB-subunit bycertain stages of spermatocytes may not have anything to dowith inhibin B, but, rather, may imply expression of activinB by these types of spermatocytes. As no activin B dimer-specific antibody is available to our knowledge, it is notpossible by immunohistochemistry or in situ hybridization todistinguish whether the observed bB peptide expression inspermatocytes is present as free bB-subunits or activin Bdimers. To our knowledge no specific activin B immunoas-says are presently available, but development of such anassay would facilitate the specific quantitative measurementof activin B in enriched spermatocyte preparations and thusaid in solving this question.

In conclusion, during pubertal maturation of the humanmale, testicular inhibin B production becomes germ cell de-pendent. This change in inhibin B regulation may be ex-

plained by a maturational change in the localization of ex-pression of the inhibin bB-subunit from the Sertoli cells tospecific germ cell stages in the spermatogenic process,namely to cells from the pachytene spermatocyte to roundspermatid stages. Our suggestions that inhibin B is a jointproduct, and that the subunits of which may originate fromdifferent cell types propose to our knowledge a new conceptin the regulation of peptide hormone and growth factorexpression. This maturational change in the production andregulation of inhibin B levels thus provides a close link be-tween serum inhibin B levels (and thus FSH feedback reg-ulation) and spermatogenesis.

Acknowledgments

We thank Stine Ehlern Jessen and Majbrit Kvist for skilled technicalassistance, and we are grateful to Dr. Richard Sharpe for his constructivecomments during the preparation of this manuscript.

References

1. De Kretser DM, McFarlane JR. 1996 Inhibin in the male. J Androl. 17:179–82.2. Illingworth PJ, Groome NP, Bryd W, et al. 1996 Inhibin-B: a likely candidate

for the physiologically important form of inhibin in man. J Clin EndocrinolMetab. 81:1321–1325.

3. Steinberger A, Steinberger E. 1976 Secretion of an FSH-inhibiting factor bycultured Sertoli cells. Endocrinology. 99:918–921.

4. Verhoeven G, Franchimont P. 1983 Regulation of inhibin secretion by Sertolicell-enriched cultures. Acta Endocrinol (Copenh). 102:136–143.

5. Skinner MK, McLachlan RI, Bremner WJ. 1989 Stimulation of sertoli cellinhibin secretion by the testicular paracrine factor PModS. Mol Cell Endocri-nol. 66:239–249.

6. Grootenhuis AJ, Timmerman MA, Hordijk PL, de Jong FH. 1990 Inhibin inimmature rat Sertoli cell conditioned medium: a 32 kDa alfabeta-B dimer. MolCell Endocrinol. 70:109–116.

7. Merchenthaler I, Culler MD, Petrusz P, Negro-Vilar A. 1987 Immunocyto-chemical localization of inhibin in rat and human reproductive tissues. MolCell Endocrinol. 54:239–243.

8. Roberts V, Meunier H, Sawchenko PE, Vale W. 1989 Differential productionand regulation of inhibin subunits in rat testicular cell types. Endocrinology.125:2350–2359.

9. Majdic G, McNeilly AS, Sharpe RM, Evans LR, Groome NP, Saunders PTK.1997 Testicular expression of inhibin and activin subunits and follistatin in therat and human fetus and neonate and during postnatal development in the rat.Endocrinology. 138:2136–2147.

10. Zhang T, Zhou H-M, Liu Y-X. 1997 Expression of plasminogen activator andinhibitor, urokinase receptor and inhibin subunits in rhesus monkey testes.Mol Hum Reprod. 3:223–231.

11. Bhasin S, Krummen L, Morelos BS, et al. 1989 Stage-dependent expressionof inhibin a and bB subunits during the cycle of the rat seminiferous epithe-lium. Endocrinology. 124:987–991.

12. Pineau C, Sharpe RM, Saunders PT, Gerard N, Jegou B. 1990 Regulation ofSertoli cell inhibin production and of inhibin a-subunit mRNA levels byspecific germ cell types. Mol Cell Endocrinol. 72:13–22.

13. Allenby G, Foster PMD, Sharpe RM. 1991 Evidence that secretion of immu-noactive inhibin by seminiferous tubules from the adult rat testis is regulatedby specific germ cell types: correlation between in vivo and in vitro studies.Endocrinology. 128:467–476.

14. Klaij IA, van Pelt AMM, Timmerman MA, Blok LJ, de Rooij DG, de JongFH. 1994 Expression of inhibin subunit mRNAs and inhibin levels in thetestes of rats with stage-synchronized spermatogenesis. J Endocrinol.141:131–141.

15. Anawalt BD, Bebb RA, Matsumoto AM, et al. 1996 Serum inhibin B levelsreflect sertoli cell function in normal men and men with testicular dysfunction.J Clin Endocrinol Metab. 81:3341–3345.

16. Andersson A-M, Juul A, Petersen JH, Muller J, Groome NP, Skakkebæk NE.1997 Serum inhibin B in healthy pubertal and adolescent boys: relation to age,stage of puberty, and follicle-stimulating hormone, luteinizing hormone tes-tosterone and estradiol levels. J Clin Endocrinol Metab. 82:3976–3982.

17. Andersson A-M, Toppari J, Haavisto A-M, et al. 1998 Longitudinal repro-ductive hormone profiles in infants: peak of inhibin B levels in infant boysexceeds levels in adult men. J Clin Endocrinol Metab. 83:675–681.

18. Groome NP, Illingworth PJ, O’Brien M, et al. 1996 Measurement of dimericinhibin B throughout the human menstrual cycle. J Clin Endocrinol Metab.81:1401–1405.


19. Groome NP, Illingworth PJ, O’Brien M, et al. 1994 Detection of dimericinhibin throughout the human menstrual cycle by two-site enzyme immuno-assay. Clin Endocrinol (Oxf). 40:717–723.

20. Burger HG, Yamada Y, Bangah ML, McCloud PI, Warne GL. 1991 Serumgonadotropin, sex steroid and immunoreactive inhibin levels in the first twoyears of life. J Clin Endocrinol Metab. 72:682–686.

21. Raivio T, Toppari J, Perheentupa A, McNeilly AS, Dunkel L. 1997 Treatmentof prepubertal gonadotrophin-deficient boys with recombinant human follicle-stimulating hormone. Lancet. 350:263–264.

22. Bergh A, Cajander S. 1990 Immunohistochemical localization of inhibin-alphain the testes of normal men and in men with testicular disorders. Int J Androl.13:463–469.

23. Vliegen MK, Schlatt S, Weinbauer GF, Bergmann M, Groome NP, NieschlagE. 1993 Localization of inhibin/activin subunits in the testis of adult nonhumanprimates and men. Cell Tissue Res. 273:261–268.

24. Forti G, Vannelli GB, Barni T, Balboni GC, Orlando C, Serio M. 1992Sertoli-germ cells interactions in the human testis. J Steroid Biochem Mol Biol.43:419–422.

25. Vannelli GB, Barni T, Forti G, et al. 1992 Immunolocalization of inhibin

a-subunit in the human testis. A light- and electron-microscopy study. CellTissue Res. 269:221–227.

26. Anderson RA, Irvine DS, Balfour C, Groome NP, Riley SC. 1998 Inhibin Bin seminal plasma: testicular origin and relationship to spermatogenesis. HumReprod. 13:920–926.

27. Maddocks S, Sharpe RM. 1989 The route of secretion of inhibin from the rattestis. J Endocrinol 120:R5–R8.

28. Gray AM, Mason AJ. 1990 Requirement for activin A and transforming growthfactor-b1 pro-regions in homodimer assembly. Science. 247:1328–1330.

29. Mason AJ. 1994 Functional analysis of the cysteine residues of activin A. MolEndocrinol. 8:325–332.

30. Robertson DM, Sullivan J, Watson M, Cahir N. 1995 Inhibin forms in humanplasma. J Endocrinol. 144:261–269.

31. Robertson DM, Cahir N, Findlay JK, Burger HG, Groome N. 1997 Thebiological and immunological characterization of inhibin A and B forms inhuman follicular fluid and plasma. J Clin Endocrinol Metab. 82:889–896.

32. Mason AJ, Farnworth PG, Sullivan J. 1996 Characterization and determina-tion of the biological activites of noncleavable high molecular weight forms ofinhibin A and activin A. Mol Endocrinol. 10:1055–1065.


different roles of prepubertal and postpubertal …...sections (4 mm) were mounted on superfrost...

Documents