- Email: [email protected]
The effects of inhibin purified from bovine follicular fluid in several in vitro pituitary cell culture systems
Molecular
and Cell&r
Elsevier Scientific
Endocrinology,
Publishers
Ireland,
46 (1986)
29
29-36
Ltd.
MCE 01478
The effects of inhibin purified from bovine follicular in several in vitro pituitary cell culture systems D.M. Robertson, Department
of Anatomy,
(Received
Key word.7: follicostatin;
FSH inhibitory
M.S. Giacometti
Monash
material:
University,
1 November
inhibin
Wellington
1985; accepted
fluid
and D.M. de Kretser Road, Clayion,
21 February
Vicrorra 3168 (Austraha)
1986)
in vitro assay
Summary
The effects of inhibin purified from bovine follicular fluid (bFF) and a charcoal-treated bFF preparation were investigated in several inhibin in vitro systems based on the use of pituitary cells in culture. FSH, LH, TSH and PRL were determined in the medium and cell extracts before and after a 4 h LHRH stimulation test. Both pure inhibin and bFF preparations markedly inhibited the basal release, cell content and LHRH-stimulated release of FSH in a parallel dose-dependent manner with minor or negligible effects on LH, TSH and PRL. Using parallel line bioassay statistics the inhibin activities of the purified inhibin preparations in the various in vitro systems were calculated with the charcoal-treated bFF as standard. Significantly higher inhibin values were obtained using the basal release procedure than with the cell content or LHRH-stimulated release procedures. This difference was influenced by the length of time the inhibin preparations were in culture. The highly purified preparations showed no signs of cytotoxicity in culture as assessed by a “Cr release test. It is concluded that purified bFF inhibin is specific in suppressing pituitary FSH and not LH, TSH or PRL. The observation that purified bFF inhibin is more active in the basal release procedure is attributed to a loss of activity of the bFF standard in this system in contrast to that observed in either the cell content or LHRH-stimulated release procedure,
Introduction In a recent study (Robertson et al., 1985) a procedure for isolating pure inhibin from bovine follicular fluid was detailed. Inhibin was shown to be a protein with a molecular weight of 56000, which consists of 2 subunits of molecular weight 43000 and 15000. The specific activity, as determined by an in vitro bioassay method (Scott et al., 1980) was 200000 U/mg with a purification factor of approximately 3500-fold. The present study provides data concerning the bioactivity of the 56000 Da inhibin in 3 in vitro 0303-7207/86/$03.50
0 1986 Elsevier Scientific
Publishers
Ireland.
systems utilizing pituitary cells in culture based on basal FSH release (De Jong et al., 1979) FSH cell content (Scott et al., 1980) and LHRH-stimulated FSH release (Eddie et al., 1979; Scott and Burger, 1981). It represents the first detailed characterisation of the bioactivity of a number of pure inhibin preparations and assesses the activity of these preparations. Materials and methods Na,“Cr, was obtained in a sterilized chloride solution (100-400 mCi/mg) Ltd.
sodium from
30
Amersham Australia. Sydney. Synthetic was supplied by Bachem, Torrance, CA.
LHRH
Inhihin preparations Bovine ovaries were obtained from local abattoirs and bFF was aspirated into a chilled vessel containing aprotinin (10 U/ml) and phenylmethylsulphonyl fluoride (PMSF, 24 pg/ml). The bFF was stored frozen at - 20°C and subsequently charcoal treated for 30 min at 4°C with an equal volume of 1% Norit A charcoal in 10 mM phosphate buffer, pH 7.4, containing 0.9% NaCl and 0.1%’ bovine serum albumin (BSA). The supernatant was diluted and portioned in 250 ~1 aliquots, snap-frozen in solid COz/ethanol and stored at - 20°C. This charcoal-treated bFF preparation had an inhibin activity of 520 U/ml in terms of an earlier described ovine testicular lymph (OTL) standard, which had an arbitrary unitage of 1 U/mg (Scott et al., 1980). No significant change in inhibin potency of this bFF preparation was observed over a 3-month period when assayed against the OTL standard. The bFF preparation did not suppress LH cell content nor was it cytotoxic to the cultured pituitary cells as assessed by a “Cr release procedure (Robertson et al.. 1982). Five highly purified inhibin preparations isolated from separate collections of bFF were isolated according to the method of Robertson et al. (1985). The purity of these preparations was assessed by silver staining on SDS-polyacrylamide gel electrophoresis. Under non-reducing conditions a single band was observed with a molecular weight of 56000. In 2 of the 5 preparations a minor contamination was observed in the 60 kDa region; however, this material has been shown to be biologically inactive (Robertson et al., 1985). Under reducing conditions 2 sub-units with molecular weights of 42 000 and 15 500 were observed. These molecular weight values are similar to previously published values (Robertson et al., 1985). Owing to the low yields of purified inhibin per batch (- 10 pg), protein measurements by amino acid analysis were only determined in 2 batches resulting in a mean specific activity of 200 000 U/mg protein (Robertson et al., 1985). Subsequent batches were checked for inhibin bioactivity and purity on SDS-PAGE and the mass of inhibin recovered was calculated from the above
Non-red.
Red.
67 43 29 20 14
Fig. 1. Analytical SDS-PAGE of a purified inhibin preparation used in this study. The inhibin preparation (I) was electrophoresed under reducing and non-reducing conditions on a 12.5% polyacrylamide gel slab and silver stained. Horizontal bands present in the 60-70 kDa region of the reduced gels are due to 2-mercaptoethanol. Molecular weight standards(S) BSA 67000, ovalbumin 43000, carbonic anhydraae 29000. goose egg lysozyme 20 300. chick egg lysozyme 14 300.
specific activity. The SDS-PAGE pattern of one of the preparations used in this study is presented in Fig. 1. Inhibin activity was determined by the in vitro bioassay of Scott et al. (1980) based on the suppression of FSH cell content in pituitary cell cultures using the charcoal-treated bFF preparation described above as standard. The purified preparations were diluted to a working concentration of approximately 10 U/ml or 1 U/culture well for investigation in the proposed studies. In vitro culture procedures The pituitary cell culture conditions employed in the in vitro assay of inhibin (Scott et al., 1980; Au et al., 1983) were used. In brief: pituitaries from adult male rats of the Sprague-Dawley strain were enzymatically dispersed and the cells (125 000 cells/well) were cultured in 16 mm culture plates (3524; Costar, Cambridge, MA) in Dulbecco’s modified eagles medium containing 10% fetal calf
31
serum (DMEM, 500 ~1) for 48 h in a 5% CO, atmosphere in air. Charcoal-treated bFF (100 ~1) and purified bFF inhibin (100 ~1) diluted in DMEM + 0.1% BSA were added in quadruplicate at multiple dilutions and incubated for 72 h. The cell media were removed and saved, the cells were washed once with DMEM, and lysed with DMEM + 0.1% Triton X-100 (500 ~1). In plates in which an LHRH stimulation test was performed, the cells were incubated in the presence of DMEM + 0.1% BSA (500 ~1) containing 10 nM LHRH. After 4 h the media were removed and saved and the remaining cells lysed with DMEM + 0.1% Triton X-100. The various in vitro procedures were compared in 4 experiments each consisting of 4 plates, each of 24 wells. Two plates contained DMEM and 5 doses of charcoal-treated bFF (0.125-2 U inhibin/well) and the other 2 plates contained DMEM and 5 doses of purified inhibin with doses up to l-2 U inhibin/well. Media removed on the fifth day were designated ‘the hormone basal recharcoallease fraction’. Two plates containing treated bFF and purified inhibin were lysed with Triton X-100 and the lysate was designated as ‘the hormone cell content fraction’. The remaining 2 plates were incubated with LHRH, the media designated as ‘ the LHRH-stimulated hormone release fraction’ and the remaining cells after lysis, designated as ‘the hormone cell content fraction after LHRH stimulation’. With this design the hormone levels in the basal release fraction and either the cell content or LHRH-stimulated release fraction were directly compared in the same plate as a means of reducing experiment error attributed to between-plate variability (see Table 2). The 5’Cr cytotoxicity test was performed published (Robertson et al., 1982).
as
Radioimmunoassay FSH was assayed using an anti-hFSH serum ‘251-rFSH (NIH-rFSH-I,) as tracer and (20,‘26), NIH-rFSH-RP-1 as standard. A 2-day discontinuous assay employing a polyethylene glycol-facilitated second antibody step was used. The sensitivity of the assay was < 0.1 pg/ml, the within-assay variation l-2% and between-assay variation 6-9%. LH was assayed using an anti-rat LH serum
‘251-rLH (NIH-rLH-I,) as tracer and (NIH-S6), NIH-rLH-I, as standard. A one-day equilibrium assay was employed using a second antibody step as for FSH. The sensitivity of the assay was 0.4 ng/ml. the within-assay variation 2-3% and between-assay variation 9-10%. TSH was assayed using an anti-rat TSH serum as tracer and (NIH-SS), ‘251-rTSH (NIH-rTSH-I,) NIH-rTSH-RP-1 as standard. Assay conditions were similar to LH. The sensitivity of the assay was 125 ng/ml, the within-assay variation 2-3% and between-assay variation 7-12%. PRL was assayed using an anti-rat PRL serum (NIH-S8), ‘251-rPRL (NIH-rPRL-I,) as tracer and NIH-rPRL-RP-3 as standard. Assay conditions were similar to LH. The sensitivity of the assay was < 1.5 ng/ml, the within-assay variation 2-4% and between-assay variation 9-12%. Statistical analysis Inhibin potency estimates with the charcoaltreated bFF preparation as standard as well as slope values of the dose-response lines and indices of precision (X) were determined using standard parallel line bioassay statistics. Significance of differences between control and test cultures was assessed using Dunnett’s (1955) ranking test or by paired r-test. Results
Both charcoal-treated bFF and purified inhibin significantly suppressed FSH levels in a parallel dose-dependent manner in the various in vitro systems examined (Fig. 2). In contrast, with a high dose (1 U) of purified inhibin the effects on LH, TSH and PRL were not significant using Dunnett’s test although by paired t-test a small significant decrease was observed with the basal release of TSH (Table 1). A comparison by paired r-test of hormone levels in control cultures and cultures containing 2 U charcoal-treated bFF showed a significant decrease in FSH in the same groups as that presented in Table 1 for purified inhibin and significant although minor decreases in LH cell content (17%, P < 0.05) and PRL basal release (12%, P < 0.05). Using the dose-dependent suppression of FSH by inhibin as the assay end point in the various in
FSH
&yypj
CELL
PRL
CONTENT
100
80
’
E $
80
BASAL
RELEASE
8 40
AFTER
0.06
0.25
LHRH-STIMULATED
RELEASE
1
INHIBIN DILUTION w
8: z 8.
6 6
&&kk%Mtm_%
‘P-l-l-l 00.13
0.5
CFF
RELEASE
2 O.O~zx INHIBIN DILUTION
Ftg. 2. The relative levels of FSH, LH, TSH and PRL in pituitary cell media and extracts following incubation with charcoal-treated bFF and highly purified preparations of bFF inhibin. Results presented are the mean f SD of 4 experiments with 4 different purified inhibin preparations for FSH and 3 experiments with 3 purified preparations for LH, TSH and PRL. The initial dilution of purified inhibin contained approximately 1 U inhibin/culture well for each experiment. Basal release refers to hormone released into the medium in the absence of LHRH. The effect of charcoal-treated bFF and 3 highly purified preparations of bFF inhibin on “Cr release, (0) mean +-SD of 4 replicates, (a) mean* SD of mean values from 3 purified inhibin preparations. (a) DMEM.
33
tl +I -+I +I +I
-*0,x5 --WOO jdddd
34
TABLE
2
THE RATIO OF INHIBIN LEVELS OBTAINED NUMBER OF PURE INHIBIN PREPARATIONS DIFFERENT IN VITRO CULTURE CONDITIONS
FOR A USING
Statistical analysis is based on a paired f-test of values tained by the respective assays. Mean i SD.
ob-
In vitro cultures
n
Ratio
Basal release/cell content Basal release/LHRH-stimulated release Cell content/LHRH-stimulated release Cell content after LHRH stimulation/ LHRH-stimulated release
4 4 4
1.39kO.17 * 1.62 kO.22 * 1.11 iO.19 NS
3
1.62kO.17
*
NS, not significant. * P i 0.05 TABLE
3
COMPARISON OF INHIBIN LEVELS OF A HIGHLY PURIFIED PREPARATION AT DIFFERENT DAYS OF CULTURE WHEN ASSAYED BY THE BASAL RELEASE (BR) AND CELL CONTENT (CC) ASSAY PROCEDURES Mean and standard.
95% confidence
limits.
charcoal-treated
bFF
Days of culture
Basal release procedure
Cell content procedure
Ratio BR/CC
1 2 3 4
NP 483 (398 : 585) 526 (444 : 622) * 594 (536 : 609) *
487 (307 : 680) a 454 (380 : 540) 432 (390 : 470) 469(395:556)
1.06 1.21 1.27
as
a Units of inhibin in original stock solution. * Significant difference (P < 0.05) observed with corresponding value obtained in the cell content procedure. as based on the failure of its 95% confidence limits to overlap the mean value obtained by the other procedure. NP, not parallel. TABLE
4
CHARACTERISTICS
OF THE VARIOUS
Mean + SD of results from 4 separate
* (U/ml)
ED,, (U/ml) Slope of FSH inhibition dose-response lines ** Index of precision (X)
IN VITRO
PROCEDURES
IN THEIR
RESPONSES
TO INHIBIN
experiments.
FSH cell content Sensitivity
vitro systems studied, the inhibin activity in the purified preparations was calculated employing parallel line bioassay statistics using the charcoaltreated bFF as standard. Parallelism was observed between the 2 preparations in the linear region of the respective dose-response lines in each experiment with each procedure. The indices of precision were considered satisfactory (Table 4). Significantly higher values were obtained with the basal release and the hormone cell content following LHRH stimulation in vitro systems than with the other systems (Table 2). To assess if the differences in potency obtained between the various in vitro systems relate to differences in relative stability of the purified and charcoal-treated bFF preparations in culture, a purified inhibin preparation was incubated for 1, 2, 3 and 4 days in culture and its potency, in terms of charcoal-treated bFF, determined. The ratio of activities comparing the basal release with the cell content procedure increased with time in culture (Table 3). An examination of the EDd0 values of the dose-response lines for the 2 preparations in both procedures on days 2-4 showed a 30-40% decrease, while in contrast no change was observed with the bFF preparation in the basal release procedure. The purified inhibin preparations were not cytotoxic in culture as based on the failure to observe any significant change in “Cr release with increasing doses of inhibin (Fig. 2). A number of characteristics of the various in vitro procedures are presented in Table 4. Largely
0.14 + 0.03 0.85 i 0.23 18.0 f5.2 0.13
FSH basal release
0.12&0.02 0.48 * 0.20 24.6
i3.7 0.09
* Units of inhibin required to suppress FSH to 90% of controls. bFF standard curve. ** Expressed as B change in FSH/ln dose. See Fig. 1. a ED,,.
LHRH-stimulated FSH release 0.13 *0.04 0.65 * 0.43 21.7
is.4 0.12
All values are based on extrapolation
FSH cell content following LHRH stimulation 0.09 * 0.02 0.57 f 0.21 .+ 16.1
k6.3 0.18
of the linear region of the
35
similar values for the slope of the FSH inhibition dose-response lines, index of precision and sensitivity (ED,,, and ED,, or ED,, values) were observed for the various in vitro systems. Discussion Results of this study indicate that pure inhibin isolated from bFF suppresses FSH without affect‘ing LH. TSH and PRL to any marked extent in the various in vitro systems compared. A failure to observe a marked effect on LH in both cell content and basal release procedures is consistent with that seen with bFF and ovine testicular lymph preparations (De Jong et al., 1979; Scott et al., 1980; Baker et al.. 1981; Franchimont et al., 1981: Scott and Burger, 1981). However. the failure to see an effect of both pure inhibin and charcoaltreated bFF on LH in the LHRH-stimulated procedure is at variance with that seen with bFF (20-30% suppression, De Jong et al., 1979) ovine testicular lymph (25550%, Baker et al., 1981; Franchimont et al., 1981; Scott and Burger, 1981) and granulosa cell and Sertoli cell extracts (Massicotte et al., 1984a. b). While there may be an intrinsic difference related to the inhibin source and sample pretreatment, the basis of the differences may lie in the LHRH incubation conditions employed, since in this study a 4 h incubation with 10 nM LHRH was used while De Jong et al. (1978) used a 6 h incubation period with 50 nM LHRH. Clearly these variables will have to be examined in future studies. The observation that TSH and PRL are either unaffected or largely unaffected is consistent with the expected effect of inhibin on pituitary cells. Similar results have been observed with other inhibin sources, rete testis fluid extracts and seminal plasma fractions (Franchimont et al., 1978). A comparison of potency estimations for the purified material in terms of the charcoal-treated bFF standard shows that higher values were obtained with the basal release procedure and the procedure based on the cell content following LHRH stimulation than with either the cell content and LHRH stimulation release procedure and that the magnitude of the difference was proportional to the length of the culture period. If the differences were attributed to a relative loss of
activity in culture by one of the preparations this loss would be reflected in changes in potency with time. In the cell content procedure no change in potency was observed on days 2, 3 and 4 of culture suggesting a similar stability for the 2 preparations. However, in the basal release procedure an increase in potency was observed over the corresponding time period. An examination of the assay sensitivity (e.g., ED,,,) for the bFF standard and purified material dose-response lines indicated that the ED,,, values for bFF standard for days 2-4 of culture remained unchanged while the purified material decreased by 40%‘. These results suggest that as far as FSH basal release is concerned, and in contrast to its effects on FSH cell content. bFF inhibin is less stable than purified inhibin. A decrease in ED,,, of similar magnitude was noted with both preparations in the cell content procedure. Even though the basis for these observations is not understood, the fact that the purified material behaves differently to unfractionated bFF suggests that either some property of the inhibin molecule has undergone modification during the purification process or that other factors are present in crude preparations which interact with the action of inhibin. Differences between these types of assays have been observed previously although the standard and test preparations were either from different sources and species (De Jong et al., 1983) or unstated (Baker et al., 1981). When inhibin unitage is expressed in terms of mass, based on the specific activity of purified preparations of bFF inhibin (200000 U/mg), the sensitivity (ED,,) of the assay ranged from 0.5 to 1.0 ng/ml. Previous attempts to measure inhibin in plasma by in vitro bioassays have indicated levels (Scott et al., 1980) below detection, which for a maximum plasma volume of 100 ~1 would correspond to < 5-10 ng/ml. Based on the measurement of other protein hormones of the same molecular size, these levels are still sufficiently high to be measured by conventional radioimmunoassay techniques once the appropriate reagents become available. When ED,, values for inhibin are expressed in molar terms (4-8 X 10-l’ M) and compared to the ED,,, of LHRH on stimulating FSH release in pituitary cells in vitro (1 nM LHRH, unpublished
36
data), it is evident that under the conditions employed in this study the pituitary cells are more sensitive to inhibin than LHRH. Acknowledgements We wish to thank Ms. C. O’Reilly and Ms. S. Ellis for their technical assistance and Mr. T. Gill for collecting the bFF. The RIA reagents were gifts of the NIADDK. This study was supported by a grant from the National Health and Medical Research Council of Australia. References Au, C.L., Robertson, D.M. and de Kretser, D.M. (1983) Endocrinology 112, 2399244. Baker, H.W.G., Eddie, L.W., Higginson, R.E., Hudson, B.. Keogh, E.J. and NiaII. H.D. (1981) In: Intragonadal Regulation of Reproduction, Eds.: P. Franchimont and C.P. Channing (Academic Press, New York) pp. 193-228. De Jong, F.H., Smith, SD. and Van Der Molen, H.J. (1979) J. Endocrinol. 80. 91-102. De Jong, F.H., Jansen, E.H.J.M.. Steenbergen, J.. Van Dijk, S. and Van Der Molen. H.J. (1983) In: Role of Peptides and
Proteins in Control of Reproduction, Eds.: SM. McCann and D.S. Dhindsa (Elsevier Science Publishers, Amsterdam) pp. 257-273. Dunnett. C.W. (1955) J. Am. Stat. Assoc. 50, 1096-1121. Eddie, L.W., Baker. H.W.G., Higginson, R.E. and Hudson, B. (1979)J. Endocrinol. 81, 49960. Franchimont, P.. Demoulin. A., Verstraelen-Proyard, J., Hazee-Hagelstein, M.T., Walton, J.S. and Waites, G.M.H. (1978) Int. J. Androl. 2. 69-79. Franchimont, P., Henderson, K.. Verhoeven, G., Hazee-HageIstein. M.-T.. Charlet-Renard. C.. Demoulin. A., Bourguignon J.-P. and Lecomte-Yerna, M.-J. (1981) In: Intragonadal Regulation of Reproduction, Eds.: P. Franchimont and C.P. Channing (Academic Press, New York) pp. 167-191. Massicotte, J., Legace, L., Godbout. M. and Labrie, F. (1984a) J. Endocrinol. 100, 133-140. Massicotte. J.. Legace, L., Labrie. F. and Dorrington, J.H. (1984b) Am. J. Physiol. 247, E4955E504. Robertson, D.M.. Au. C.L. and de Kretser, D.M. (1982) Mol. Cell. Endocrinol. 26. 119-127. Robertson. D.M.. Foulds, L.M., Leversha, L., Morgan, F.J., Hearn. M.T.W., Burger, H.G., Wettenhall, R.E.H. and de Kretser. D.M. (1985) Biochem. Biophys. Res. Commun. 126, 220-226. Scott. R.S. and Burger, H.G. (1981) Biol. Reprod. 24, 541-550. Scott, R.S.. Burger, H.G. and Quigg, H. (1980) Endocrinology 107. 1536-1542.
and Cell&r
Elsevier Scientific
Endocrinology,
Publishers
Ireland,
46 (1986)
29
29-36
Ltd.
MCE 01478
The effects of inhibin purified from bovine follicular in several in vitro pituitary cell culture systems D.M. Robertson, Department
of Anatomy,
(Received
Key word.7: follicostatin;
FSH inhibitory
M.S. Giacometti
Monash
material:
University,
1 November
inhibin
Wellington
1985; accepted
fluid
and D.M. de Kretser Road, Clayion,
21 February
Vicrorra 3168 (Austraha)
1986)
in vitro assay
Summary
The effects of inhibin purified from bovine follicular fluid (bFF) and a charcoal-treated bFF preparation were investigated in several inhibin in vitro systems based on the use of pituitary cells in culture. FSH, LH, TSH and PRL were determined in the medium and cell extracts before and after a 4 h LHRH stimulation test. Both pure inhibin and bFF preparations markedly inhibited the basal release, cell content and LHRH-stimulated release of FSH in a parallel dose-dependent manner with minor or negligible effects on LH, TSH and PRL. Using parallel line bioassay statistics the inhibin activities of the purified inhibin preparations in the various in vitro systems were calculated with the charcoal-treated bFF as standard. Significantly higher inhibin values were obtained using the basal release procedure than with the cell content or LHRH-stimulated release procedures. This difference was influenced by the length of time the inhibin preparations were in culture. The highly purified preparations showed no signs of cytotoxicity in culture as assessed by a “Cr release test. It is concluded that purified bFF inhibin is specific in suppressing pituitary FSH and not LH, TSH or PRL. The observation that purified bFF inhibin is more active in the basal release procedure is attributed to a loss of activity of the bFF standard in this system in contrast to that observed in either the cell content or LHRH-stimulated release procedure,
Introduction In a recent study (Robertson et al., 1985) a procedure for isolating pure inhibin from bovine follicular fluid was detailed. Inhibin was shown to be a protein with a molecular weight of 56000, which consists of 2 subunits of molecular weight 43000 and 15000. The specific activity, as determined by an in vitro bioassay method (Scott et al., 1980) was 200000 U/mg with a purification factor of approximately 3500-fold. The present study provides data concerning the bioactivity of the 56000 Da inhibin in 3 in vitro 0303-7207/86/$03.50
0 1986 Elsevier Scientific
Publishers
Ireland.
systems utilizing pituitary cells in culture based on basal FSH release (De Jong et al., 1979) FSH cell content (Scott et al., 1980) and LHRH-stimulated FSH release (Eddie et al., 1979; Scott and Burger, 1981). It represents the first detailed characterisation of the bioactivity of a number of pure inhibin preparations and assesses the activity of these preparations. Materials and methods Na,“Cr, was obtained in a sterilized chloride solution (100-400 mCi/mg) Ltd.
sodium from
30
Amersham Australia. Sydney. Synthetic was supplied by Bachem, Torrance, CA.
LHRH
Inhihin preparations Bovine ovaries were obtained from local abattoirs and bFF was aspirated into a chilled vessel containing aprotinin (10 U/ml) and phenylmethylsulphonyl fluoride (PMSF, 24 pg/ml). The bFF was stored frozen at - 20°C and subsequently charcoal treated for 30 min at 4°C with an equal volume of 1% Norit A charcoal in 10 mM phosphate buffer, pH 7.4, containing 0.9% NaCl and 0.1%’ bovine serum albumin (BSA). The supernatant was diluted and portioned in 250 ~1 aliquots, snap-frozen in solid COz/ethanol and stored at - 20°C. This charcoal-treated bFF preparation had an inhibin activity of 520 U/ml in terms of an earlier described ovine testicular lymph (OTL) standard, which had an arbitrary unitage of 1 U/mg (Scott et al., 1980). No significant change in inhibin potency of this bFF preparation was observed over a 3-month period when assayed against the OTL standard. The bFF preparation did not suppress LH cell content nor was it cytotoxic to the cultured pituitary cells as assessed by a “Cr release procedure (Robertson et al.. 1982). Five highly purified inhibin preparations isolated from separate collections of bFF were isolated according to the method of Robertson et al. (1985). The purity of these preparations was assessed by silver staining on SDS-polyacrylamide gel electrophoresis. Under non-reducing conditions a single band was observed with a molecular weight of 56000. In 2 of the 5 preparations a minor contamination was observed in the 60 kDa region; however, this material has been shown to be biologically inactive (Robertson et al., 1985). Under reducing conditions 2 sub-units with molecular weights of 42 000 and 15 500 were observed. These molecular weight values are similar to previously published values (Robertson et al., 1985). Owing to the low yields of purified inhibin per batch (- 10 pg), protein measurements by amino acid analysis were only determined in 2 batches resulting in a mean specific activity of 200 000 U/mg protein (Robertson et al., 1985). Subsequent batches were checked for inhibin bioactivity and purity on SDS-PAGE and the mass of inhibin recovered was calculated from the above
Non-red.
Red.
67 43 29 20 14
Fig. 1. Analytical SDS-PAGE of a purified inhibin preparation used in this study. The inhibin preparation (I) was electrophoresed under reducing and non-reducing conditions on a 12.5% polyacrylamide gel slab and silver stained. Horizontal bands present in the 60-70 kDa region of the reduced gels are due to 2-mercaptoethanol. Molecular weight standards(S) BSA 67000, ovalbumin 43000, carbonic anhydraae 29000. goose egg lysozyme 20 300. chick egg lysozyme 14 300.
specific activity. The SDS-PAGE pattern of one of the preparations used in this study is presented in Fig. 1. Inhibin activity was determined by the in vitro bioassay of Scott et al. (1980) based on the suppression of FSH cell content in pituitary cell cultures using the charcoal-treated bFF preparation described above as standard. The purified preparations were diluted to a working concentration of approximately 10 U/ml or 1 U/culture well for investigation in the proposed studies. In vitro culture procedures The pituitary cell culture conditions employed in the in vitro assay of inhibin (Scott et al., 1980; Au et al., 1983) were used. In brief: pituitaries from adult male rats of the Sprague-Dawley strain were enzymatically dispersed and the cells (125 000 cells/well) were cultured in 16 mm culture plates (3524; Costar, Cambridge, MA) in Dulbecco’s modified eagles medium containing 10% fetal calf
31
serum (DMEM, 500 ~1) for 48 h in a 5% CO, atmosphere in air. Charcoal-treated bFF (100 ~1) and purified bFF inhibin (100 ~1) diluted in DMEM + 0.1% BSA were added in quadruplicate at multiple dilutions and incubated for 72 h. The cell media were removed and saved, the cells were washed once with DMEM, and lysed with DMEM + 0.1% Triton X-100 (500 ~1). In plates in which an LHRH stimulation test was performed, the cells were incubated in the presence of DMEM + 0.1% BSA (500 ~1) containing 10 nM LHRH. After 4 h the media were removed and saved and the remaining cells lysed with DMEM + 0.1% Triton X-100. The various in vitro procedures were compared in 4 experiments each consisting of 4 plates, each of 24 wells. Two plates contained DMEM and 5 doses of charcoal-treated bFF (0.125-2 U inhibin/well) and the other 2 plates contained DMEM and 5 doses of purified inhibin with doses up to l-2 U inhibin/well. Media removed on the fifth day were designated ‘the hormone basal recharcoallease fraction’. Two plates containing treated bFF and purified inhibin were lysed with Triton X-100 and the lysate was designated as ‘the hormone cell content fraction’. The remaining 2 plates were incubated with LHRH, the media designated as ‘ the LHRH-stimulated hormone release fraction’ and the remaining cells after lysis, designated as ‘the hormone cell content fraction after LHRH stimulation’. With this design the hormone levels in the basal release fraction and either the cell content or LHRH-stimulated release fraction were directly compared in the same plate as a means of reducing experiment error attributed to between-plate variability (see Table 2). The 5’Cr cytotoxicity test was performed published (Robertson et al., 1982).
as
Radioimmunoassay FSH was assayed using an anti-hFSH serum ‘251-rFSH (NIH-rFSH-I,) as tracer and (20,‘26), NIH-rFSH-RP-1 as standard. A 2-day discontinuous assay employing a polyethylene glycol-facilitated second antibody step was used. The sensitivity of the assay was < 0.1 pg/ml, the within-assay variation l-2% and between-assay variation 6-9%. LH was assayed using an anti-rat LH serum
‘251-rLH (NIH-rLH-I,) as tracer and (NIH-S6), NIH-rLH-I, as standard. A one-day equilibrium assay was employed using a second antibody step as for FSH. The sensitivity of the assay was 0.4 ng/ml. the within-assay variation 2-3% and between-assay variation 9-10%. TSH was assayed using an anti-rat TSH serum as tracer and (NIH-SS), ‘251-rTSH (NIH-rTSH-I,) NIH-rTSH-RP-1 as standard. Assay conditions were similar to LH. The sensitivity of the assay was 125 ng/ml, the within-assay variation 2-3% and between-assay variation 7-12%. PRL was assayed using an anti-rat PRL serum (NIH-S8), ‘251-rPRL (NIH-rPRL-I,) as tracer and NIH-rPRL-RP-3 as standard. Assay conditions were similar to LH. The sensitivity of the assay was < 1.5 ng/ml, the within-assay variation 2-4% and between-assay variation 9-12%. Statistical analysis Inhibin potency estimates with the charcoaltreated bFF preparation as standard as well as slope values of the dose-response lines and indices of precision (X) were determined using standard parallel line bioassay statistics. Significance of differences between control and test cultures was assessed using Dunnett’s (1955) ranking test or by paired r-test. Results
Both charcoal-treated bFF and purified inhibin significantly suppressed FSH levels in a parallel dose-dependent manner in the various in vitro systems examined (Fig. 2). In contrast, with a high dose (1 U) of purified inhibin the effects on LH, TSH and PRL were not significant using Dunnett’s test although by paired t-test a small significant decrease was observed with the basal release of TSH (Table 1). A comparison by paired r-test of hormone levels in control cultures and cultures containing 2 U charcoal-treated bFF showed a significant decrease in FSH in the same groups as that presented in Table 1 for purified inhibin and significant although minor decreases in LH cell content (17%, P < 0.05) and PRL basal release (12%, P < 0.05). Using the dose-dependent suppression of FSH by inhibin as the assay end point in the various in
FSH
&yypj
CELL
PRL
CONTENT
100
80
’
E $
80
BASAL
RELEASE
8 40
AFTER
0.06
0.25
LHRH-STIMULATED
RELEASE
1
INHIBIN DILUTION w
8: z 8.
6 6
&&kk%Mtm_%
‘P-l-l-l 00.13
0.5
CFF
RELEASE
2 O.O~zx INHIBIN DILUTION
Ftg. 2. The relative levels of FSH, LH, TSH and PRL in pituitary cell media and extracts following incubation with charcoal-treated bFF and highly purified preparations of bFF inhibin. Results presented are the mean f SD of 4 experiments with 4 different purified inhibin preparations for FSH and 3 experiments with 3 purified preparations for LH, TSH and PRL. The initial dilution of purified inhibin contained approximately 1 U inhibin/culture well for each experiment. Basal release refers to hormone released into the medium in the absence of LHRH. The effect of charcoal-treated bFF and 3 highly purified preparations of bFF inhibin on “Cr release, (0) mean +-SD of 4 replicates, (a) mean* SD of mean values from 3 purified inhibin preparations. (a) DMEM.
33
tl +I -+I +I +I
-*0,x5 --WOO jdddd
34
TABLE
2
THE RATIO OF INHIBIN LEVELS OBTAINED NUMBER OF PURE INHIBIN PREPARATIONS DIFFERENT IN VITRO CULTURE CONDITIONS
FOR A USING
Statistical analysis is based on a paired f-test of values tained by the respective assays. Mean i SD.
ob-
In vitro cultures
n
Ratio
Basal release/cell content Basal release/LHRH-stimulated release Cell content/LHRH-stimulated release Cell content after LHRH stimulation/ LHRH-stimulated release
4 4 4
1.39kO.17 * 1.62 kO.22 * 1.11 iO.19 NS
3
1.62kO.17
*
NS, not significant. * P i 0.05 TABLE
3
COMPARISON OF INHIBIN LEVELS OF A HIGHLY PURIFIED PREPARATION AT DIFFERENT DAYS OF CULTURE WHEN ASSAYED BY THE BASAL RELEASE (BR) AND CELL CONTENT (CC) ASSAY PROCEDURES Mean and standard.
95% confidence
limits.
charcoal-treated
bFF
Days of culture
Basal release procedure
Cell content procedure
Ratio BR/CC
1 2 3 4
NP 483 (398 : 585) 526 (444 : 622) * 594 (536 : 609) *
487 (307 : 680) a 454 (380 : 540) 432 (390 : 470) 469(395:556)
1.06 1.21 1.27
as
a Units of inhibin in original stock solution. * Significant difference (P < 0.05) observed with corresponding value obtained in the cell content procedure. as based on the failure of its 95% confidence limits to overlap the mean value obtained by the other procedure. NP, not parallel. TABLE
4
CHARACTERISTICS
OF THE VARIOUS
Mean + SD of results from 4 separate
* (U/ml)
ED,, (U/ml) Slope of FSH inhibition dose-response lines ** Index of precision (X)
IN VITRO
PROCEDURES
IN THEIR
RESPONSES
TO INHIBIN
experiments.
FSH cell content Sensitivity
vitro systems studied, the inhibin activity in the purified preparations was calculated employing parallel line bioassay statistics using the charcoaltreated bFF as standard. Parallelism was observed between the 2 preparations in the linear region of the respective dose-response lines in each experiment with each procedure. The indices of precision were considered satisfactory (Table 4). Significantly higher values were obtained with the basal release and the hormone cell content following LHRH stimulation in vitro systems than with the other systems (Table 2). To assess if the differences in potency obtained between the various in vitro systems relate to differences in relative stability of the purified and charcoal-treated bFF preparations in culture, a purified inhibin preparation was incubated for 1, 2, 3 and 4 days in culture and its potency, in terms of charcoal-treated bFF, determined. The ratio of activities comparing the basal release with the cell content procedure increased with time in culture (Table 3). An examination of the EDd0 values of the dose-response lines for the 2 preparations in both procedures on days 2-4 showed a 30-40% decrease, while in contrast no change was observed with the bFF preparation in the basal release procedure. The purified inhibin preparations were not cytotoxic in culture as based on the failure to observe any significant change in “Cr release with increasing doses of inhibin (Fig. 2). A number of characteristics of the various in vitro procedures are presented in Table 4. Largely
0.14 + 0.03 0.85 i 0.23 18.0 f5.2 0.13
FSH basal release
0.12&0.02 0.48 * 0.20 24.6
i3.7 0.09
* Units of inhibin required to suppress FSH to 90% of controls. bFF standard curve. ** Expressed as B change in FSH/ln dose. See Fig. 1. a ED,,.
LHRH-stimulated FSH release 0.13 *0.04 0.65 * 0.43 21.7
is.4 0.12
All values are based on extrapolation
FSH cell content following LHRH stimulation 0.09 * 0.02 0.57 f 0.21 .+ 16.1
k6.3 0.18
of the linear region of the
35
similar values for the slope of the FSH inhibition dose-response lines, index of precision and sensitivity (ED,,, and ED,, or ED,, values) were observed for the various in vitro systems. Discussion Results of this study indicate that pure inhibin isolated from bFF suppresses FSH without affect‘ing LH. TSH and PRL to any marked extent in the various in vitro systems compared. A failure to observe a marked effect on LH in both cell content and basal release procedures is consistent with that seen with bFF and ovine testicular lymph preparations (De Jong et al., 1979; Scott et al., 1980; Baker et al.. 1981; Franchimont et al., 1981: Scott and Burger, 1981). However. the failure to see an effect of both pure inhibin and charcoaltreated bFF on LH in the LHRH-stimulated procedure is at variance with that seen with bFF (20-30% suppression, De Jong et al., 1979) ovine testicular lymph (25550%, Baker et al., 1981; Franchimont et al., 1981; Scott and Burger, 1981) and granulosa cell and Sertoli cell extracts (Massicotte et al., 1984a. b). While there may be an intrinsic difference related to the inhibin source and sample pretreatment, the basis of the differences may lie in the LHRH incubation conditions employed, since in this study a 4 h incubation with 10 nM LHRH was used while De Jong et al. (1978) used a 6 h incubation period with 50 nM LHRH. Clearly these variables will have to be examined in future studies. The observation that TSH and PRL are either unaffected or largely unaffected is consistent with the expected effect of inhibin on pituitary cells. Similar results have been observed with other inhibin sources, rete testis fluid extracts and seminal plasma fractions (Franchimont et al., 1978). A comparison of potency estimations for the purified material in terms of the charcoal-treated bFF standard shows that higher values were obtained with the basal release procedure and the procedure based on the cell content following LHRH stimulation than with either the cell content and LHRH stimulation release procedure and that the magnitude of the difference was proportional to the length of the culture period. If the differences were attributed to a relative loss of
activity in culture by one of the preparations this loss would be reflected in changes in potency with time. In the cell content procedure no change in potency was observed on days 2, 3 and 4 of culture suggesting a similar stability for the 2 preparations. However, in the basal release procedure an increase in potency was observed over the corresponding time period. An examination of the assay sensitivity (e.g., ED,,,) for the bFF standard and purified material dose-response lines indicated that the ED,,, values for bFF standard for days 2-4 of culture remained unchanged while the purified material decreased by 40%‘. These results suggest that as far as FSH basal release is concerned, and in contrast to its effects on FSH cell content. bFF inhibin is less stable than purified inhibin. A decrease in ED,,, of similar magnitude was noted with both preparations in the cell content procedure. Even though the basis for these observations is not understood, the fact that the purified material behaves differently to unfractionated bFF suggests that either some property of the inhibin molecule has undergone modification during the purification process or that other factors are present in crude preparations which interact with the action of inhibin. Differences between these types of assays have been observed previously although the standard and test preparations were either from different sources and species (De Jong et al., 1983) or unstated (Baker et al., 1981). When inhibin unitage is expressed in terms of mass, based on the specific activity of purified preparations of bFF inhibin (200000 U/mg), the sensitivity (ED,,) of the assay ranged from 0.5 to 1.0 ng/ml. Previous attempts to measure inhibin in plasma by in vitro bioassays have indicated levels (Scott et al., 1980) below detection, which for a maximum plasma volume of 100 ~1 would correspond to < 5-10 ng/ml. Based on the measurement of other protein hormones of the same molecular size, these levels are still sufficiently high to be measured by conventional radioimmunoassay techniques once the appropriate reagents become available. When ED,, values for inhibin are expressed in molar terms (4-8 X 10-l’ M) and compared to the ED,,, of LHRH on stimulating FSH release in pituitary cells in vitro (1 nM LHRH, unpublished
36
data), it is evident that under the conditions employed in this study the pituitary cells are more sensitive to inhibin than LHRH. Acknowledgements We wish to thank Ms. C. O’Reilly and Ms. S. Ellis for their technical assistance and Mr. T. Gill for collecting the bFF. The RIA reagents were gifts of the NIADDK. This study was supported by a grant from the National Health and Medical Research Council of Australia. References Au, C.L., Robertson, D.M. and de Kretser, D.M. (1983) Endocrinology 112, 2399244. Baker, H.W.G., Eddie, L.W., Higginson, R.E., Hudson, B.. Keogh, E.J. and NiaII. H.D. (1981) In: Intragonadal Regulation of Reproduction, Eds.: P. Franchimont and C.P. Channing (Academic Press, New York) pp. 193-228. De Jong, F.H., Smith, SD. and Van Der Molen, H.J. (1979) J. Endocrinol. 80. 91-102. De Jong, F.H., Jansen, E.H.J.M.. Steenbergen, J.. Van Dijk, S. and Van Der Molen. H.J. (1983) In: Role of Peptides and
Proteins in Control of Reproduction, Eds.: SM. McCann and D.S. Dhindsa (Elsevier Science Publishers, Amsterdam) pp. 257-273. Dunnett. C.W. (1955) J. Am. Stat. Assoc. 50, 1096-1121. Eddie, L.W., Baker. H.W.G., Higginson, R.E. and Hudson, B. (1979)J. Endocrinol. 81, 49960. Franchimont, P.. Demoulin. A., Verstraelen-Proyard, J., Hazee-Hagelstein, M.T., Walton, J.S. and Waites, G.M.H. (1978) Int. J. Androl. 2. 69-79. Franchimont, P., Henderson, K.. Verhoeven, G., Hazee-HageIstein. M.-T.. Charlet-Renard. C.. Demoulin. A., Bourguignon J.-P. and Lecomte-Yerna, M.-J. (1981) In: Intragonadal Regulation of Reproduction, Eds.: P. Franchimont and C.P. Channing (Academic Press, New York) pp. 167-191. Massicotte, J., Legace, L., Godbout. M. and Labrie, F. (1984a) J. Endocrinol. 100, 133-140. Massicotte. J.. Legace, L., Labrie. F. and Dorrington, J.H. (1984b) Am. J. Physiol. 247, E4955E504. Robertson, D.M.. Au. C.L. and de Kretser, D.M. (1982) Mol. Cell. Endocrinol. 26. 119-127. Robertson. D.M.. Foulds, L.M., Leversha, L., Morgan, F.J., Hearn. M.T.W., Burger, H.G., Wettenhall, R.E.H. and de Kretser. D.M. (1985) Biochem. Biophys. Res. Commun. 126, 220-226. Scott. R.S. and Burger, H.G. (1981) Biol. Reprod. 24, 541-550. Scott, R.S.. Burger, H.G. and Quigg, H. (1980) Endocrinology 107. 1536-1542.