- Email: [email protected]
Assessment of the contribution of Leydig cells to the secretion of inhibin by the rat testis
Molecular and Cellular Endocrinology, 67 (1989) .~13-118 Elsevier Scientific Publishers Ireland, Ltd.
113
MOLCEL 02194
Rapid Paper Assessment of the contribution of Leydig cells to the secretion of inhibin by the rat testis S. M a d d o c k s a n d R.M. Sharpe MRC Reproductive Biology Unit, Centre for Reproductive Biology; Edinburgh EH3 9EW', ScotlamL U.K. (Received 8 September 1989; accepted 22 September 1989)
Key words: inhibit; E~|io,ne dimethane sulphonate; Testosterone treatment; Testicular venous blood; Spermatic venous blood; Leydi8 cell culture; Human chot~onic 8onadotrophin
Summary Cultured Leydig cells secreted 1.3-4.3 ng t'26a-itdlibin/106 cells/24 h, and although this was unaffected by human chorionic gonado~rophin (hCG), these cells could contribute to the intratesticular and blood levels of inhibin. The present study evaluated this contribution in rats in which the Leydig cells were destroyed by injection of ethane dimethane sulphonate (EDS). In these animals, inhibin levels increased in testicular interstitial fluid (IF), and in testicular (TV) and spermatic (SV) venous blood. In EDS-treated rats supplemented for 21 days with I or 25 mg testosterone esters to maintain full spermatogenesis and/or suppress the elevated follicle-stimulating hormone (FSH) levels and prevent Leydig cell regeneration, significant changes occurred in the levels of inhibin in IF, in TV and SV plasma and in the route of secretion of inhibin from the testis (i.e. via IF or seminiferous tubule fluid). However, none of these changes was related to the presence or absence of Leydig cells. It is concluded that Leydig cells make little contribution to the intratesticular and blood levels of inhibin in the adult rat.
lntroduetlon While it has long been recognised that Sertoli cells are the source of inhibin in the male, the role(s) of this hormone has become confused (McLachlan et al., 1988). Recent studies have shown that the levels of immunoactive inhibin in rats and men increase after administration of high doses of human chorionic gonadotrophin (hCG) (McLachlan et al., 1988; Sharpe et al., 1988a; Drummond et al., 1989), an effect that is independent of
Address for correspondence: R.M. Sharpe, MRC Reproductive Biology Unit, Centre for Reproductive Biology, 37 Chalmers Street, Edinburgh EH3 9EW, Scotland, U.K.
steroid synthesis (Sharpe et al., 1988a) but is abolished when the Leydig cells are destroyed with ethane dimethane suiphonate (EDS; Drummond et al., 1989). This suggests that LH-regulated, non-steroidogenic Leydis cell products are involved in the paracrine regulation of inhibin sec~'etion by the Sertoli cells. However, a recent study has shown that rat Leydig cells produce immuno- and bioactive inhibin in culture (Risbridger et al., 1989), and this may make a significant contribution to the levels measured in vivo. In the present study, in addition to assessing the in vitro production of inhibin by adult rat Leydig cells, we have also examined the contribution that such secretion makes in vivo by assessing the route of secretion of inhibin from the testis, and
0303-7207/89/$03.50 © 1989 Elsevier Scientific Pubfishers Ireland, Ltd.
114
therefore its likely source within the testis, in normal and Le,,dig cc!!deple~'d adult rats. Materials and methods
Animal:, and treatments. Sprague-Dawley rats aged 90-110 days from our own colony were given a single i.p. injectton of 75 mg/kg EDS in 1 : 3 (v/v) dimethylsulphoxide: water (DMSO). This treatment destroys all Leydig cells within 2 days, but they regenerate from 12-14 days onwards (Kerr et al., 1985; Bartlett et al., 1986). In addition to EDS, some rats were also injected s.c. every 3 days with 25 or 1 mg testosterone esters (TE) (Sustanon, Organon Labs.) in 0.1 ml arachis oil, beginning at the time of EDS administration. Normal spermatogenesis and testis weight are only maintained with 25 nag TE, whilst both 25 and 1 mg doses suppress plasma levels of LH and FSH to subnormal levels and prevent Leydig cell regeneration (Sharpe et al., 1988b, c). Control rats were injected with DMSO alone or together with 1 or 25 mg TE. Groups of control and EDS-treated rats were sampled at 6 days after injection, whilst at 21 days, groups of control and EDS-treated rats, with or without TE supplementation were sampled, as described below. Collection of samples. Each rat was anaesthetized with ether, heparinized by i.e. injection of 2500 U heparin (Pularin; Duncan Flockart & Co.) and samples of testicular (TV), spermatic (SV), and peripheral (PV) venous blood collected (Maddocks and Sharpe, 1989a, b), and the plasma separated by centrifugation. Testicular interstitial fluid (IF) from each animal was collected overnight at 4°C (Sharpe and Cooper, 1983). Assessment of the source of intratesticular inhibin. In the normal adult rat, inhibin from the Sertoli cell is secreted mainly into seminiferous tubule fluid (STF; 'apical' secretion) and is then resorbed into blood from the rete testis (Maddocks and Sharpe, 1989a). Inhibin is also secreted via the base of the Sertoli cell ('basal' secretion) into testicular IF. Inhibin secreted by the Leydig cells will pass into IF, but not into STF. In addition to directly measurhlg levels in IF, the relative contribution that inhibin in IF and STF make to the overall levels leaving the testis in blood can be assessed by the collection of TV, SV
and PV blood samples (Maddocks and Sharpe, 1989a, 1990). The transfer of inhibin from IF into TV blood is reflected in the difference between inhibin levels in TV blood and those in PV blood (i.e. entering the testis). As TV blood passes through the mediastinal venous plexus overlying the rete testis it 'picks up' inhibin secreted into STF, and this is reflected in the difference between inhibin levels in TV and SV blood, the latter being collected above the mediastinal venous plexus. Although this 'pick-up' is clearly evident in control rats (Maddocks and Sharpe, 1989a) it is underestimated clue to the dilution of venous blood in the spermatic cord with incoming arterial blood such that testosterone levels in SV blood are only 40-60% of those in TV blood (Maddocks and Sharpe, 1989). By assessing this dilution of testosterone, the level of inhibin measured in SV samples can be 'corrected' to determine the total inhibin leaving the testis in blood (Maddocks and Sharpe, 1990) and this was done in the present study. In rats in which the dilution of SV blood could not be assessed directly due to non-detectable testosterone (EDS + 6 days) or a non-testicular source of testosterone (exogenous TE administration), the dilution factor determined in vehicletreated control rats was applied, since the dilution of SV blood by incoming arterial blood remains constant irrespective of hormonal and spermatogenie status (Maddocks and Sharpe, 1989b). Leydig cell isolation and culture. Leydig cells were isolated from groups of four rats by collagenase digestion followed by purification on discontinuous gradients of Percoll (Sharpe and Fraser, 1983). Aliquots of 0.05 × 106 cells, of which 73-84~ were histochemically identifiable as Leydig cells, were then incubated for 24 h at 320C in 0.25 ml culture medium, in the presence or absence of 5 nM hCG (Chorulon, Intervet). Hormone measurement. Inhibin was measured by radioimmunoassay (RIA) using an antibody ($55) raised in sheep to the 1-26 sequence of the a-subunit of porcine inhibin (Sharpe et al., 1988c). By reference to an in vitro bioassay for inhibin using ovine pituitary cells (Tsonis et al., 1986), the RIA has been validated for the assay of inhibin in samples of rat plasma, interstitial fluid and Sertoli cell-conditioned medium (Sharpe et al., 1988c; Maddocks and Sharpe, 1989a, 1990; unpublished
115
data). In virtually every instance the level of inhibin determined by RIA has been substantiated by bioassay. Therefore, the potential contribution of free a-subunit forms of inhibin to the levels me.~sured by RIA appears to be negligible; but this remains a possibility. The pattern of change in IF levels of immunoactive inhibin following EDS treatment was confirmed by the assay of pools of IF in the in vitro bioassay, after treatment of the IF with 1 mg/ml dextran-coated charcoal (Norit A; Sigma) to lower steroid levels. Peripheral plasma levels of LH and FSH were measured by RIA (Fraser and Sandow, 1977), using NIADDK kits, whilst the plasma and IF levels of testosterone were measured by RIA as described previously (Sharpe and Cooper, 1983). Statistical analysis. Data were analysed by analysis of variance and the paired or Student's t-test. Results
Inhibin secretion by isolated Leydig cells in culture. Leydig cell cultures secreted variable levels of inhibin after 24 h of incubation, but addition of hCG had no effect on this secretion (Table 1).
TABLE 1 IN VITRO PRODUCTION OF IMMUNOACTIVE INHIBIN IN A 24 h PERIOD BY THREE SEPARATE PREPARATIONS OF ISOLATED RAT LEYDIG CELLS, CULTURED IN THE PRESENCE OR ABSENCE OF 5 nM hCG (MEAN +. SEM, n = 4) Cell preparation number
pg equiv, i-26a inhibin/lO 6 Leydig cells Basal
+ hCG (5 nM)
1 2 3
4 313 + 417 2132 +. 138 1280 + 390
3157 +. 438 2181 +. 540 924+. 91
Effect of Leydig cell destruction and regeneration on inhibin secretion in vivo. Six days after EDS treatment, all Leydig cells are destroyed and testosterone concentrations in IF were undetectable ( < 10 ng/ml; controls 182 + 50 ng/ml; mean + SEM; n = 6). However, in the absence of Leydig cells, inhibin levels in IF increased significantly (Table 2). By 21 days after EDS treatment, Leydig cells begin to regenerate, and IF testosterone became detectable (35 + 12 ng/ml), but remained lower than in controls (213 + 84 ng/ml). Inhibin levels in IF at this time were still significantly greater than in controls (Table 2) and this was
TABLE 2 EFFECT OF TREATMENT WITH EDS A N D / O R SUPPLEMENTATION WITH 1 OR 25 mg TE ON TESTICULAR WEIGHT, THE IF LEVELS OF INHIBIN AND THE ROUTE OF SECRETION OF INHIBIN FROM THE TESTIS (MEANS+ S~M~ n -- 5 OR 6) Testicular weights are included to indicate the degree of impairment (or otherwise) of spermatogenesis. Treatment
Control+6days EDS+6days~
Testis
IF
weight (mg)
lnhibin (ng/ml) Basal b
Percent
Apical c
Percent
1681+79 1498+49
8.2+0.7 23.3+.3.6"*
98+ 63 487+. 9 6 " *
4.3+2.1 16.1+.3.8
2210+. 66 2827+.534
95.7+.2.1 2308+694 83.9+.3.8 * 3313+.507
13.5+.0.7 13.7+1.8 14.04-0.9
128+ 65 268±166 388+.130
6.0+2.8 10.7+.6.6 18.4+6.0
1694 +. 199 2 535 +. 430 1983+.408
94.0 +. 2.8 89.3 +. 6.6 81.6+.6.0
1822 +. 253 2 803 +. 343 2371 +.367
460+130 * 393 +. 120 234+.101
9.8+3.2 9.4 4- 5.4 18.4+8.1
4229-1-474"** 1878+'454 1001+.113 *
90.2+3.2 90.6+.5.4 81.6+.8.1
46894"483 *** 2114+.496 1237+.101 ***
Control+21 days 1607+44 +25 mgTE 1569+85 +1 mgTE 1242+63 EDS+21 days + 25 tug TE ¢ t + l mgTE
965-1-40*** 20.1+3.6 * 1555 + 53 17.5 +. 3.8 906+~5 ** 19.6+2.2 *
Inhibin secretion by Sertoli cells ' (P8 equivalents !-2ea.inhibin/ml ) Total
Note that rats in treatment groups marked with ¢r had no Leydig cells. a See Materials and Methods for details of the methods u , ~ to determine the route of inhibin secretion. b Although basal secretion refers to secretion via the base of the Sertoli cells, inhibin from the Leydig cells (if present) could have contributed to these values. © Apical secretion refers to inhibin secreted into STF and resorbed from the fete testis. Ley~g cells could not have contributed to inhibin secreted via this route. * P < 0.05, * * P < 0.01, * * * P < 0.001 in comparison with values in the respective control group.
116
I 7000"
I-I
6000
I sv
5000
IPV
4000
"IV
Controls
I
Few Leydig cells
T
~
U'i
D
FSH
400
m
300
200
Leydig cells
30o0 c
~5 2000 om .c: .~! 1000 (D
100
0""
v-
I
¢3" G)
v ,.o e,.
e,m ..g_ J= ._¢
4000:
"13 m m u)
I
=!
Leydig cells
m
200 8
3000
o
,ooo:
100
1000-'
a.
3000
0
ND.
I
4000
= ~ o
0
S
"
I maTE
Leydig cells
25 mg TE
0
~ 3
I
No Leydig cells
1
200
2000
100
1000 .
Control
Ei~
~
llr
Control
d" f l
N.D.
.
0
EI~
Treatment Fig. 1. Levels of immunoactive inhibin in testicular (TV), spermatic (SV), and peripheral (PV) venous blood (left panel), and of FSH and LH in PV blood (right panel), in rats 21 days after administration of vehicle or EDS, alone or tog=ther with 1 or 25 mg testosterone esters (TE) every 3 days. Values are means:/: SEM (n = 5). N.D. = non-detectable. * P < 0.05, ** P < 0.01, *** P < 0.001 in comparison with value in the respective control group. Note that any contribution to inhibin levels by Leydig cells should he reflected in a change in inhibin levels in TV plasma; however, where significant changes in inhibin did occur (e.g. EDS + 1 mg 'i~:) they were a reflection of altered SV levels of inhibin and these are determined largely by apical secretion of inhibin by the Sertoli cells (see Table 2).
confirmed by inhibin bioassay (controls ffi 6.1 (95~ confidence limits 5.0-7.6); EDS = 14.8 (12.1-18.5) U / m l equiv, o R T F standard). Plasma L H and F S H levels were increased significantly at both 6 and 21 days after E D S treatment (e.g. Fig. 1). The secretion of inhibin into S T F and its uptake from the fete testis leads to a dramatic increase in inhibin levels in SV blood in controls
(Fig. 1). This pattern of secretion was largely unchanged in rats treated 6 (no Leydig cells) or 21 (regenerating Leydig cells) days previously with EDS, although inhibin levels were increased compared to controls, especially at 21 days (Fig. 1, Table 2). While inhibin levels secreted via the base of the Sertoli cells were increased significantly after E D S treatment, this increase was c o m p a r a b l e
117
at 6 (no Leydig cells) and 21 (regenerating Leydig cells) days (Table 2), suggesting that Leydig ~ n s make tittle contribution to IF levels of inhibin in the adult rat testis. However, as this treatment also induces supranormal FSH levels and impairment of spermatogenesis due to androgen withdrawal, the importance of these other changes was assessed by treating rats with EDS and supplementing them with 1 or 25 mg TE for 21 days. Effect of Leydig cell destruction with testosterone replacement on the route of secretion of inhibin from the testis. Supplementation of control and EDStreated rats with 1 or 25 mg TE suppressed LH and FSH levels and prevented Leydig ceil regeneration (Fig. 1); however, only 25 mg TE mainrained normal testis weight (Table 2). In controls, TE treatment had no significant effect on overall inhibin secretion whereas TE administration to EDS-treated rats lowered inhibin towards control levels (Fig. 1). The route of secretion of inhibin from the testis was not altered drastically by these treatments, and although in both control and EDS-treated rats supplemented with 1 mg TE there was an increase in the proportion of inhibin secreted via the base of the Sertoli cell into IF, this was independent of the presence or absence of Leydig cells (Table 2).
Discussion This study demonstrates that rat Leydig cells secrete inhibin in culture, confirming the findings of Risbridger et al. (1989). However, we found no effect of hCG on inhibin secretion by these cells, which contrasts with the 3-fold increase in immunoactive (but not bioactive) inhibin levels induced by rLH in the study of Risbridger et al. (1988). This discrepancy could be due to the different RIA used in our study, but using the same RIA as that used by Risbridger et al. (1989) we have stiff found no effect of hCG on Leydig cell inhibin secretion in vitro (unpublished data). Assuming an average secretion of ~ ng/106 Leydig cells/24 h (Table 1) and 30 × 106 Leydig cells/testis (Mori and Christensen, 1980), Leydig cells could produce 60 ng inhibin/ testis/ 24 h. The IF volume of the rat testis is about 200 gl (Sharpe and Cooper, 1983) giving a level of 300 ng inhibin/ml/24 h. This 'concentration' will obvi-
ously be affected by IF turnover and the rate of transfer into TV blood, but could still account for a substantial proportion of IF inhibin in controls (8-13 ng/ml; Table 2). However, the destruction of all Leydig ceils using EDS did not cause any reduction in IF levels of inhibin, in fact quite the reverse, with significant increases occurring in both the IF levels of inhibin and in the total inhibin leaving the testis. It can be argued that this increase reflects the abnormal situation created by Leydig cell destruction, namely the supranormal FSH levels and/or the disruption of spermatogenesis. However, when either one or both of these changes were prevented by the administration of 1 or 25 mg TE respectively to EDS-treated rats, there was n o dramatic change in either the IF levels of inhibin or in the route of its secretion into blood (Table 2). Indeed, where an overall decrease in inhibin secretion did occur in rats lacking Leydig cells (e.g. EDS + 1 mg TE), this change could be accounted for by reduced secretion of inh~bin from the apex of the Sertoti cell (Table 2). In both control and EDS-treated rats supplemented with 1 mg TE, there was a signific~a~ i~crease in the ~basal' secretion of inhibin but this change was not related to the presence or absence of Leydig cells (Table 2, Fig. 1). The 'basal' secretion of inhibin varied in direct relationship to the IF levels of inhibin (Table 2), as would be expected, but judged on the present findings, the contribution by the Leydig cells to the IF levels of inhibin, and thus to the levels in IN' and PV blood, is probably small. The present finding of no effect of hCO on inhibin secretion by Leydig cells in vitro suggests that the increase in IF and blood levels of inhibin in vivo following treatment of rats with hCO (Sharpe et al., 1988a; Drummond et al., 1989) is not due to increased secretion of inhibin by the Leydig cells. This conclusion is supported by the demonstration that a substantial proportion of the hCG-induced increase in inhibin levels leaving the testis in blood is due to increased secretion into STF (Maddocks and Sharpe, 1990), a route to which the Leydig cells could not contribute. Our in vivo findings suggest that Leydig cells do not contribute in a major way to the intratesticular or blood levels of inhibin in the adult rat. Further studies will be necessary to determine the
118
physiological significance of inhibin production by the Leydig cells. Acknowledgements We thank Irene Cooper and Rose Leask for sklh'ed help, Dr. R.W. Kelly for synthesizing EDS, Dr. J. Rivier for synthetic porcine 1-26a-inhibin, and the NIADDK (U.S.A.) for LH and FSH RIA kits. S.M. is a Sir Robert Menzies Memorial Scholar in Medicine.
References Bartlett, J.M.S., Kerr, J.B. and Sharpe, R.M. (1986) J. Androl. 7, 240-253. Drummond, A.E., Risbridger, G.P. and de Kretser, D.M. (1989) Endocrinology 125, 510-515. Fraser, H.M. and Sandow, J. (1977) J. Endocrinol. 74, 291-296. Kerr, J.B., Donachie, K. and Rommerts, F.F.G. (1985) Cell Tissue Res. 242, 145-156.
Maddocks, S. and Sharpe, R.M. (1989a) J. Endocrinol. 120, RS-RS. Maddocks, S. and Sharpe, R.M. (1989b) J. Endocrinol. 122, 323-329. Maddocks, S. and Sharpe, R.M. (1990) Endocrinology (submitted). McLachlan, R.I., Robertson, D.M., de Kretser, D.M. and Burger, H.G. (1988) Clin. Endocrinol. 29, 77-112. Mori, H, and Christensen, A.K. (1980) J. Cell Biol. 84, 340-352. Risbridger, G.P., Clements, J., Robertson, D.M., Drummond, A.E., Muir, J., Burger, H.G. and de Kretser, D.M. (1989) Mol. Cell. Endocrinol. 66, 119-122. Sharpe, R.M. and Cooper, I. (1983) J. Reprod. Fertil. 69, 125-135. Sharpe, R.M. and Fraser, H.M. (1983) Mol. Cell. Endocrinol. 33, 131-146. Sharpe, R.M., Kerr, J.B. and Maddocks, S. (1988a) Mol. Cell. Endocrinol. 60, 243-247. Sharpe, R.M., Fraser, H.M. and Ratnasooriya, W.D. (1988b) Int. J. Androl. 11, 507-523. Sharpe, R.M., Swanston, I., Cooper, I., Tsouis, C.G. and McNeilly, A.S. (1988c) J. Endocrinol. 119, 315-326. Tsonis, C.G., McNeilly, A.S. and Baird, D.T. (1986) J. Endo¢rinol. 110, 341-352.
113
MOLCEL 02194
Rapid Paper Assessment of the contribution of Leydig cells to the secretion of inhibin by the rat testis S. M a d d o c k s a n d R.M. Sharpe MRC Reproductive Biology Unit, Centre for Reproductive Biology; Edinburgh EH3 9EW', ScotlamL U.K. (Received 8 September 1989; accepted 22 September 1989)
Key words: inhibit; E~|io,ne dimethane sulphonate; Testosterone treatment; Testicular venous blood; Spermatic venous blood; Leydi8 cell culture; Human chot~onic 8onadotrophin
Summary Cultured Leydig cells secreted 1.3-4.3 ng t'26a-itdlibin/106 cells/24 h, and although this was unaffected by human chorionic gonado~rophin (hCG), these cells could contribute to the intratesticular and blood levels of inhibin. The present study evaluated this contribution in rats in which the Leydig cells were destroyed by injection of ethane dimethane sulphonate (EDS). In these animals, inhibin levels increased in testicular interstitial fluid (IF), and in testicular (TV) and spermatic (SV) venous blood. In EDS-treated rats supplemented for 21 days with I or 25 mg testosterone esters to maintain full spermatogenesis and/or suppress the elevated follicle-stimulating hormone (FSH) levels and prevent Leydig cell regeneration, significant changes occurred in the levels of inhibin in IF, in TV and SV plasma and in the route of secretion of inhibin from the testis (i.e. via IF or seminiferous tubule fluid). However, none of these changes was related to the presence or absence of Leydig cells. It is concluded that Leydig cells make little contribution to the intratesticular and blood levels of inhibin in the adult rat.
lntroduetlon While it has long been recognised that Sertoli cells are the source of inhibin in the male, the role(s) of this hormone has become confused (McLachlan et al., 1988). Recent studies have shown that the levels of immunoactive inhibin in rats and men increase after administration of high doses of human chorionic gonadotrophin (hCG) (McLachlan et al., 1988; Sharpe et al., 1988a; Drummond et al., 1989), an effect that is independent of
Address for correspondence: R.M. Sharpe, MRC Reproductive Biology Unit, Centre for Reproductive Biology, 37 Chalmers Street, Edinburgh EH3 9EW, Scotland, U.K.
steroid synthesis (Sharpe et al., 1988a) but is abolished when the Leydig cells are destroyed with ethane dimethane suiphonate (EDS; Drummond et al., 1989). This suggests that LH-regulated, non-steroidogenic Leydis cell products are involved in the paracrine regulation of inhibin sec~'etion by the Sertoli cells. However, a recent study has shown that rat Leydig cells produce immuno- and bioactive inhibin in culture (Risbridger et al., 1989), and this may make a significant contribution to the levels measured in vivo. In the present study, in addition to assessing the in vitro production of inhibin by adult rat Leydig cells, we have also examined the contribution that such secretion makes in vivo by assessing the route of secretion of inhibin from the testis, and
0303-7207/89/$03.50 © 1989 Elsevier Scientific Pubfishers Ireland, Ltd.
114
therefore its likely source within the testis, in normal and Le,,dig cc!!deple~'d adult rats. Materials and methods
Animal:, and treatments. Sprague-Dawley rats aged 90-110 days from our own colony were given a single i.p. injectton of 75 mg/kg EDS in 1 : 3 (v/v) dimethylsulphoxide: water (DMSO). This treatment destroys all Leydig cells within 2 days, but they regenerate from 12-14 days onwards (Kerr et al., 1985; Bartlett et al., 1986). In addition to EDS, some rats were also injected s.c. every 3 days with 25 or 1 mg testosterone esters (TE) (Sustanon, Organon Labs.) in 0.1 ml arachis oil, beginning at the time of EDS administration. Normal spermatogenesis and testis weight are only maintained with 25 nag TE, whilst both 25 and 1 mg doses suppress plasma levels of LH and FSH to subnormal levels and prevent Leydig cell regeneration (Sharpe et al., 1988b, c). Control rats were injected with DMSO alone or together with 1 or 25 mg TE. Groups of control and EDS-treated rats were sampled at 6 days after injection, whilst at 21 days, groups of control and EDS-treated rats, with or without TE supplementation were sampled, as described below. Collection of samples. Each rat was anaesthetized with ether, heparinized by i.e. injection of 2500 U heparin (Pularin; Duncan Flockart & Co.) and samples of testicular (TV), spermatic (SV), and peripheral (PV) venous blood collected (Maddocks and Sharpe, 1989a, b), and the plasma separated by centrifugation. Testicular interstitial fluid (IF) from each animal was collected overnight at 4°C (Sharpe and Cooper, 1983). Assessment of the source of intratesticular inhibin. In the normal adult rat, inhibin from the Sertoli cell is secreted mainly into seminiferous tubule fluid (STF; 'apical' secretion) and is then resorbed into blood from the rete testis (Maddocks and Sharpe, 1989a). Inhibin is also secreted via the base of the Sertoli cell ('basal' secretion) into testicular IF. Inhibin secreted by the Leydig cells will pass into IF, but not into STF. In addition to directly measurhlg levels in IF, the relative contribution that inhibin in IF and STF make to the overall levels leaving the testis in blood can be assessed by the collection of TV, SV
and PV blood samples (Maddocks and Sharpe, 1989a, 1990). The transfer of inhibin from IF into TV blood is reflected in the difference between inhibin levels in TV blood and those in PV blood (i.e. entering the testis). As TV blood passes through the mediastinal venous plexus overlying the rete testis it 'picks up' inhibin secreted into STF, and this is reflected in the difference between inhibin levels in TV and SV blood, the latter being collected above the mediastinal venous plexus. Although this 'pick-up' is clearly evident in control rats (Maddocks and Sharpe, 1989a) it is underestimated clue to the dilution of venous blood in the spermatic cord with incoming arterial blood such that testosterone levels in SV blood are only 40-60% of those in TV blood (Maddocks and Sharpe, 1989). By assessing this dilution of testosterone, the level of inhibin measured in SV samples can be 'corrected' to determine the total inhibin leaving the testis in blood (Maddocks and Sharpe, 1990) and this was done in the present study. In rats in which the dilution of SV blood could not be assessed directly due to non-detectable testosterone (EDS + 6 days) or a non-testicular source of testosterone (exogenous TE administration), the dilution factor determined in vehicletreated control rats was applied, since the dilution of SV blood by incoming arterial blood remains constant irrespective of hormonal and spermatogenie status (Maddocks and Sharpe, 1989b). Leydig cell isolation and culture. Leydig cells were isolated from groups of four rats by collagenase digestion followed by purification on discontinuous gradients of Percoll (Sharpe and Fraser, 1983). Aliquots of 0.05 × 106 cells, of which 73-84~ were histochemically identifiable as Leydig cells, were then incubated for 24 h at 320C in 0.25 ml culture medium, in the presence or absence of 5 nM hCG (Chorulon, Intervet). Hormone measurement. Inhibin was measured by radioimmunoassay (RIA) using an antibody ($55) raised in sheep to the 1-26 sequence of the a-subunit of porcine inhibin (Sharpe et al., 1988c). By reference to an in vitro bioassay for inhibin using ovine pituitary cells (Tsonis et al., 1986), the RIA has been validated for the assay of inhibin in samples of rat plasma, interstitial fluid and Sertoli cell-conditioned medium (Sharpe et al., 1988c; Maddocks and Sharpe, 1989a, 1990; unpublished
115
data). In virtually every instance the level of inhibin determined by RIA has been substantiated by bioassay. Therefore, the potential contribution of free a-subunit forms of inhibin to the levels me.~sured by RIA appears to be negligible; but this remains a possibility. The pattern of change in IF levels of immunoactive inhibin following EDS treatment was confirmed by the assay of pools of IF in the in vitro bioassay, after treatment of the IF with 1 mg/ml dextran-coated charcoal (Norit A; Sigma) to lower steroid levels. Peripheral plasma levels of LH and FSH were measured by RIA (Fraser and Sandow, 1977), using NIADDK kits, whilst the plasma and IF levels of testosterone were measured by RIA as described previously (Sharpe and Cooper, 1983). Statistical analysis. Data were analysed by analysis of variance and the paired or Student's t-test. Results
Inhibin secretion by isolated Leydig cells in culture. Leydig cell cultures secreted variable levels of inhibin after 24 h of incubation, but addition of hCG had no effect on this secretion (Table 1).
TABLE 1 IN VITRO PRODUCTION OF IMMUNOACTIVE INHIBIN IN A 24 h PERIOD BY THREE SEPARATE PREPARATIONS OF ISOLATED RAT LEYDIG CELLS, CULTURED IN THE PRESENCE OR ABSENCE OF 5 nM hCG (MEAN +. SEM, n = 4) Cell preparation number
pg equiv, i-26a inhibin/lO 6 Leydig cells Basal
+ hCG (5 nM)
1 2 3
4 313 + 417 2132 +. 138 1280 + 390
3157 +. 438 2181 +. 540 924+. 91
Effect of Leydig cell destruction and regeneration on inhibin secretion in vivo. Six days after EDS treatment, all Leydig cells are destroyed and testosterone concentrations in IF were undetectable ( < 10 ng/ml; controls 182 + 50 ng/ml; mean + SEM; n = 6). However, in the absence of Leydig cells, inhibin levels in IF increased significantly (Table 2). By 21 days after EDS treatment, Leydig cells begin to regenerate, and IF testosterone became detectable (35 + 12 ng/ml), but remained lower than in controls (213 + 84 ng/ml). Inhibin levels in IF at this time were still significantly greater than in controls (Table 2) and this was
TABLE 2 EFFECT OF TREATMENT WITH EDS A N D / O R SUPPLEMENTATION WITH 1 OR 25 mg TE ON TESTICULAR WEIGHT, THE IF LEVELS OF INHIBIN AND THE ROUTE OF SECRETION OF INHIBIN FROM THE TESTIS (MEANS+ S~M~ n -- 5 OR 6) Testicular weights are included to indicate the degree of impairment (or otherwise) of spermatogenesis. Treatment
Control+6days EDS+6days~
Testis
IF
weight (mg)
lnhibin (ng/ml) Basal b
Percent
Apical c
Percent
1681+79 1498+49
8.2+0.7 23.3+.3.6"*
98+ 63 487+. 9 6 " *
4.3+2.1 16.1+.3.8
2210+. 66 2827+.534
95.7+.2.1 2308+694 83.9+.3.8 * 3313+.507
13.5+.0.7 13.7+1.8 14.04-0.9
128+ 65 268±166 388+.130
6.0+2.8 10.7+.6.6 18.4+6.0
1694 +. 199 2 535 +. 430 1983+.408
94.0 +. 2.8 89.3 +. 6.6 81.6+.6.0
1822 +. 253 2 803 +. 343 2371 +.367
460+130 * 393 +. 120 234+.101
9.8+3.2 9.4 4- 5.4 18.4+8.1
4229-1-474"** 1878+'454 1001+.113 *
90.2+3.2 90.6+.5.4 81.6+.8.1
46894"483 *** 2114+.496 1237+.101 ***
Control+21 days 1607+44 +25 mgTE 1569+85 +1 mgTE 1242+63 EDS+21 days + 25 tug TE ¢ t + l mgTE
965-1-40*** 20.1+3.6 * 1555 + 53 17.5 +. 3.8 906+~5 ** 19.6+2.2 *
Inhibin secretion by Sertoli cells ' (P8 equivalents !-2ea.inhibin/ml ) Total
Note that rats in treatment groups marked with ¢r had no Leydig cells. a See Materials and Methods for details of the methods u , ~ to determine the route of inhibin secretion. b Although basal secretion refers to secretion via the base of the Sertoli cells, inhibin from the Leydig cells (if present) could have contributed to these values. © Apical secretion refers to inhibin secreted into STF and resorbed from the fete testis. Ley~g cells could not have contributed to inhibin secreted via this route. * P < 0.05, * * P < 0.01, * * * P < 0.001 in comparison with values in the respective control group.
116
I 7000"
I-I
6000
I sv
5000
IPV
4000
"IV
Controls
I
Few Leydig cells
T
~
U'i
D
FSH
400
m
300
200
Leydig cells
30o0 c
~5 2000 om .c: .~! 1000 (D
100
0""
v-
I
¢3" G)
v ,.o e,.
e,m ..g_ J= ._¢
4000:
"13 m m u)
I
=!
Leydig cells
m
200 8
3000
o
,ooo:
100
1000-'
a.
3000
0
ND.
I
4000
= ~ o
0
S
"
I maTE
Leydig cells
25 mg TE
0
~ 3
I
No Leydig cells
1
200
2000
100
1000 .
Control
Ei~
~
llr
Control
d" f l
N.D.
.
0
EI~
Treatment Fig. 1. Levels of immunoactive inhibin in testicular (TV), spermatic (SV), and peripheral (PV) venous blood (left panel), and of FSH and LH in PV blood (right panel), in rats 21 days after administration of vehicle or EDS, alone or tog=ther with 1 or 25 mg testosterone esters (TE) every 3 days. Values are means:/: SEM (n = 5). N.D. = non-detectable. * P < 0.05, ** P < 0.01, *** P < 0.001 in comparison with value in the respective control group. Note that any contribution to inhibin levels by Leydig cells should he reflected in a change in inhibin levels in TV plasma; however, where significant changes in inhibin did occur (e.g. EDS + 1 mg 'i~:) they were a reflection of altered SV levels of inhibin and these are determined largely by apical secretion of inhibin by the Sertoli cells (see Table 2).
confirmed by inhibin bioassay (controls ffi 6.1 (95~ confidence limits 5.0-7.6); EDS = 14.8 (12.1-18.5) U / m l equiv, o R T F standard). Plasma L H and F S H levels were increased significantly at both 6 and 21 days after E D S treatment (e.g. Fig. 1). The secretion of inhibin into S T F and its uptake from the fete testis leads to a dramatic increase in inhibin levels in SV blood in controls
(Fig. 1). This pattern of secretion was largely unchanged in rats treated 6 (no Leydig cells) or 21 (regenerating Leydig cells) days previously with EDS, although inhibin levels were increased compared to controls, especially at 21 days (Fig. 1, Table 2). While inhibin levels secreted via the base of the Sertoli cells were increased significantly after E D S treatment, this increase was c o m p a r a b l e
117
at 6 (no Leydig cells) and 21 (regenerating Leydig cells) days (Table 2), suggesting that Leydig ~ n s make tittle contribution to IF levels of inhibin in the adult rat testis. However, as this treatment also induces supranormal FSH levels and impairment of spermatogenesis due to androgen withdrawal, the importance of these other changes was assessed by treating rats with EDS and supplementing them with 1 or 25 mg TE for 21 days. Effect of Leydig cell destruction with testosterone replacement on the route of secretion of inhibin from the testis. Supplementation of control and EDStreated rats with 1 or 25 mg TE suppressed LH and FSH levels and prevented Leydig ceil regeneration (Fig. 1); however, only 25 mg TE mainrained normal testis weight (Table 2). In controls, TE treatment had no significant effect on overall inhibin secretion whereas TE administration to EDS-treated rats lowered inhibin towards control levels (Fig. 1). The route of secretion of inhibin from the testis was not altered drastically by these treatments, and although in both control and EDS-treated rats supplemented with 1 mg TE there was an increase in the proportion of inhibin secreted via the base of the Sertoli cell into IF, this was independent of the presence or absence of Leydig cells (Table 2).
Discussion This study demonstrates that rat Leydig cells secrete inhibin in culture, confirming the findings of Risbridger et al. (1989). However, we found no effect of hCG on inhibin secretion by these cells, which contrasts with the 3-fold increase in immunoactive (but not bioactive) inhibin levels induced by rLH in the study of Risbridger et al. (1988). This discrepancy could be due to the different RIA used in our study, but using the same RIA as that used by Risbridger et al. (1989) we have stiff found no effect of hCG on Leydig cell inhibin secretion in vitro (unpublished data). Assuming an average secretion of ~ ng/106 Leydig cells/24 h (Table 1) and 30 × 106 Leydig cells/testis (Mori and Christensen, 1980), Leydig cells could produce 60 ng inhibin/ testis/ 24 h. The IF volume of the rat testis is about 200 gl (Sharpe and Cooper, 1983) giving a level of 300 ng inhibin/ml/24 h. This 'concentration' will obvi-
ously be affected by IF turnover and the rate of transfer into TV blood, but could still account for a substantial proportion of IF inhibin in controls (8-13 ng/ml; Table 2). However, the destruction of all Leydig ceils using EDS did not cause any reduction in IF levels of inhibin, in fact quite the reverse, with significant increases occurring in both the IF levels of inhibin and in the total inhibin leaving the testis. It can be argued that this increase reflects the abnormal situation created by Leydig cell destruction, namely the supranormal FSH levels and/or the disruption of spermatogenesis. However, when either one or both of these changes were prevented by the administration of 1 or 25 mg TE respectively to EDS-treated rats, there was n o dramatic change in either the IF levels of inhibin or in the route of its secretion into blood (Table 2). Indeed, where an overall decrease in inhibin secretion did occur in rats lacking Leydig cells (e.g. EDS + 1 mg TE), this change could be accounted for by reduced secretion of inh~bin from the apex of the Sertoti cell (Table 2). In both control and EDS-treated rats supplemented with 1 mg TE, there was a signific~a~ i~crease in the ~basal' secretion of inhibin but this change was not related to the presence or absence of Leydig cells (Table 2, Fig. 1). The 'basal' secretion of inhibin varied in direct relationship to the IF levels of inhibin (Table 2), as would be expected, but judged on the present findings, the contribution by the Leydig cells to the IF levels of inhibin, and thus to the levels in IN' and PV blood, is probably small. The present finding of no effect of hCO on inhibin secretion by Leydig cells in vitro suggests that the increase in IF and blood levels of inhibin in vivo following treatment of rats with hCO (Sharpe et al., 1988a; Drummond et al., 1989) is not due to increased secretion of inhibin by the Leydig cells. This conclusion is supported by the demonstration that a substantial proportion of the hCG-induced increase in inhibin levels leaving the testis in blood is due to increased secretion into STF (Maddocks and Sharpe, 1990), a route to which the Leydig cells could not contribute. Our in vivo findings suggest that Leydig cells do not contribute in a major way to the intratesticular or blood levels of inhibin in the adult rat. Further studies will be necessary to determine the
118
physiological significance of inhibin production by the Leydig cells. Acknowledgements We thank Irene Cooper and Rose Leask for sklh'ed help, Dr. R.W. Kelly for synthesizing EDS, Dr. J. Rivier for synthetic porcine 1-26a-inhibin, and the NIADDK (U.S.A.) for LH and FSH RIA kits. S.M. is a Sir Robert Menzies Memorial Scholar in Medicine.
References Bartlett, J.M.S., Kerr, J.B. and Sharpe, R.M. (1986) J. Androl. 7, 240-253. Drummond, A.E., Risbridger, G.P. and de Kretser, D.M. (1989) Endocrinology 125, 510-515. Fraser, H.M. and Sandow, J. (1977) J. Endocrinol. 74, 291-296. Kerr, J.B., Donachie, K. and Rommerts, F.F.G. (1985) Cell Tissue Res. 242, 145-156.
Maddocks, S. and Sharpe, R.M. (1989a) J. Endocrinol. 120, RS-RS. Maddocks, S. and Sharpe, R.M. (1989b) J. Endocrinol. 122, 323-329. Maddocks, S. and Sharpe, R.M. (1990) Endocrinology (submitted). McLachlan, R.I., Robertson, D.M., de Kretser, D.M. and Burger, H.G. (1988) Clin. Endocrinol. 29, 77-112. Mori, H, and Christensen, A.K. (1980) J. Cell Biol. 84, 340-352. Risbridger, G.P., Clements, J., Robertson, D.M., Drummond, A.E., Muir, J., Burger, H.G. and de Kretser, D.M. (1989) Mol. Cell. Endocrinol. 66, 119-122. Sharpe, R.M. and Cooper, I. (1983) J. Reprod. Fertil. 69, 125-135. Sharpe, R.M. and Fraser, H.M. (1983) Mol. Cell. Endocrinol. 33, 131-146. Sharpe, R.M., Kerr, J.B. and Maddocks, S. (1988a) Mol. Cell. Endocrinol. 60, 243-247. Sharpe, R.M., Fraser, H.M. and Ratnasooriya, W.D. (1988b) Int. J. Androl. 11, 507-523. Sharpe, R.M., Swanston, I., Cooper, I., Tsouis, C.G. and McNeilly, A.S. (1988c) J. Endocrinol. 119, 315-326. Tsonis, C.G., McNeilly, A.S. and Baird, D.T. (1986) J. Endo¢rinol. 110, 341-352.