Journal of Reproductive Immunology, 22 (1992) 117-126
117
Elsevier Scientific Publishers Ireland Ltd.
JRI 00777
Immunosuppressive activity in the rat seminiferous tubules P. P611/inen a, M. v o n E u l e r b'c, S. Sainio-P611/inen a, K. J a h n u k a i n e n a, H. H a k o v i r t a a, O. S 6 d e r b a n d M. P a r v i n e n a aDepartment of Anatomy, Institute of Biomedicine, University of Turku, SF-20520 Turku (Finland) and bpediatric Endocrinology Unit and CDepartment of Oncology, Karolinska Institute, S-10401 Stockholm (Sweden) (Accepted for publication 28 February 1992)
Summary Rat seminiferous tubule segments in defined stages of the epithelial cycle were isolated by transillumination-assisted microdissection. The segments were cultured together with ConA-stimulated peripheral blood lymphocytes (PBL) and incorporation of 3H-labelled thymidine was measured. Tubule segments in stages II-VIII of the seminiferous epithelial cycle inhibited PBL proliferation significantly more than stages IX-I. Inhibition was lowest in stages IX-XII and increased progressively to reach a maximum in stages II-VIII. In a more detailed analysis, tubules in stages V and VI inhibited PBL proliferation significantly less than stage II tubules. No significant difference was observed between stages 1I and VII. The immunosuppressive activity had molecular weights of - 25 kDa and - 65 kDa in stage II-VIII seminiferous tubules. In stage II-VI seminiferous tubules activity was present also at - 1 0 kDa. The results suggest that the seminiferous tubules produce highmolecular weight immunosuppressive activity in a stage-dependent way. In addition to its contribution to the immunologically privileged milieu of the testis this activity may also be involved in the physiological regulation of DNA synthesis in the seminiferous epithelium. Key words: rat testis," seminiferous tubules; spermatogenesis; irnmunosuppression; growth factors," protectin
Correspondence to: Dr. Pasi P. P611~inen, Department of Anatomy, Institute of Biomedicine, University of Turku, Kiinamyllynkatu 10, 20520 Turku, Finland. 0165-0378/92/$05.00 © 1992 Elsevier Scientific Publishers Ireland Ltd. Printed and Published in Ireland
118
Introduction
The rodent testicular interstitium is proposed to be an immunologically privileged site, where activation of lymphocytes is inhibited by local products, leading to prolonged survival of intra-testicular allografts in the absence of immunosuppressive drugs (see Maddocks and Setchell, 1990). The nature of the factors involved in the formation and regulation of the immunologically privileged status of the testicular interstitial tissue has been largely unknown. Accumulating evidence indicates a major contribution of testicular soluble factors locally inhibiting lymphocyte activation in a paracrine manner (see P611~nen et al., 1990). Recent data indicate that lymphocytes bear receptors for at least 15 different testicular peptides (see P611/inen et al., 1990) and for estrogens (Cohen et al., 1983), but not for androgens (Grossman et al., 1979; Raveche et al., 1980; Sasson and Mayer, 1981; Cohen et al., 1983). However, in gel filtration chromatography of extracellular fluid from the testicular interstitial tissue (P611~nen et al., 1988, 1989b) and testicular extracts (Sainio-P611~inen et al., 1991) only four peaks of lymphocyte proliferation inhibiting activity with molecular ratios (Mr) of 25, 65, 200 and 400 kDa were observed. One of them, protectin D, with a Mr of - 2 5 kDa (for definition of protectin, see P611~nen et al., 1990), is especially prominent in cryptorchidism and hypogonadotrophic hypogonadism (Sainio-P611finen et al., 1991), in which the seminiferous epithelium undergoes significant changes. This suggests that the 25 kDa testicular immunosuppressive factor may have a physiological function in the seminiferous tubules. The present study was performed to determine whether immunosuppressive activity is produced by the seminiferous tubules and if the production of this activity is dependent on the stage of the seminiferous epithelial cycle. Materials and Methods
Animals Adult Wistar (Turku) or Lewis (ALAB, Stockholm, Sweden) rats were used as donors of lymphocytes and testis tissue. Isolation of seminiferous tubule segments Seminiferous tubule segments in defined stages of the seminiferous epithelial cycle (Leblond and Clermont, 1952) were isolated for co-cultures with lymphocytes using transillumination-assisted microdissection as previously described (Parvinen and Ruokonen, 1982).
119
Seminiferous tubule extracts A total length of 1 m of 2 mm tubule segments in stages II-VI, VII-VIII, I X - X I I and X I I I - I of the seminiferous epithelial cycle were homogenized in a volume of 500 ~1 of saline by vortexing vigorously in an Eppendorf tube for 5 min. The homogenates were centrifuged in an Eppendorf centrifuge with full speed for 15 min. The supernatants were collected and centrifuged again. The second supernatants were used for chromatography. Isolation of peripheral blood mononuclear cells (PBL) Blood was collected with a syringe containing heparin from the right ventricle of the heart whilst saline was infused into the left ventricle. Mononuclear cells were isolated by Ficoll-Paque (5.7% Ficoll 400 solution, Pharmacia Fine Chemicals, Uppsala, Sweden) gradient centrifugation as originally described by B6yum (1968). The isolated cells (average 3 × 107 cells from 50 ml of blood obtained by perfusion) were washed twice with RPMI 1640 medium before culture. The cells were mainly lymphocytes when examined by light microscopy. A few monocytes and some erythrocytes were also present. Co-cultures of seminiferous tubules and lymphocytes Seminiferous tubule segments of 2 mm length were cultured with 2 x 105 peripheral blood lymphocytes (PBL) in 200 ~1 of RPMI 1640 containing 5 /zg/ml Concanavalin A and 5% fetal calf serum (FCS) in an atmosphere of 5% CO2 in air at 37°C for 72 h. After 48 h of culture, 7.4 kBq of [6-3H]thymidine (3H-TdR, specific activity 185 GBq/mmol) in 20 t~l RPMI 1640 was added to each well. The cultures were harvested 16 h later onto glass fiber filter discs using a multiple cell harvester. The filter discs were put into scintillation vials and covered by Optiphase 'hisafe' 3 scintillation fluid (Pharmacia-LKB, Uppsala, Sweden). Radioactivity on the filter discs was measured in a/3-counter. Protectin bioassay The washed peripheral blood mononuclear cells were counted in 0.1% Trypan blue solution and diluted to 4 × 106/ml in RPMI 1640 containing 10% FCS. An aliquot of 50 #1 of the cell suspension was pipetted to wells of a standard 96-well titer plate with U-shaped wells (2 × 105cells/well). The cells were stimulated by adding 50/~1 of 20 t~g/ml ConA solution (final concentration 5/zg/ml) in RPMI 1640 containing 10% FCS. Aliquots of 50 #1 of the seminiferous tubule extract gel filtration fractions were added to the wells. Finally, 50 /zl of serum-free RPMI 1640 was added to each well to reach the final volume of 200 /~1. Each culture was made in triplicate. After 48 h of culture at 37°C in an atmosphere of 5% CO2 in air, 7.4 kBq of
120
6-3H-thymidine in 20 ~1 RPMI 1640 was added to each well. The ceils were harvested 16 h later onto glass fiber filter discs using a multiple cell harvester. Radioactivity on the filter discs was measured using liquid scintillation counting.
HPLC size exclusion chromatography An aliquot of 100 t~l of the seminiferous tubule extract was applied to a TSK G3000SW (7.5 × 600 mm, Pharmacia-LKB, Uppsala, Sweden) gel filtration column. The proteins were eluted with phosphate-buffered saline (PBS) using a flow rate of 1.0 ml/min. Fractions of 0.5 ml were collected. The absorbance of the eluent was monitored at 280 nm. Molecular size standards were thyroglobulin (669 kDa), ferritin (440 kDa), catalase (232 kDa), aldolase (158 kDa), albumin (67 kDa), ovalbumin (43 kDa), chymotrypsinogen A (25 kDa) and ribonuclease A (13.7 kDa) Kay values were calculated using the equation Kav = V e - V o / V t - V o , where Vt is the total bed volume (26.51 ml) and Vo the column void volume (11.76 ml) obtained by measuring elution volume for blue dextran 2000. lie is the elution volume of each standard molecule. The Kay values (on the linear scale) were plotted against the corresponding molecular weights (on a logarithmic scale). Kay values were calculated for the eluted peaks of immunosuppressive activity and their corresponding Mr values estimated from the standard curve. Data expression and statistical analysis Inhibitions were expressed as normalized inhibition indices: (E cpm (control): no) cpm (test) Normalized inhibition index =
~
× 100% (E cpm (control): no) -
-
t/ "
n
cpm (test)
where n~ is the number of control cultures and n t the number of test cultures. Mean counts/min from control cultures were divided by individual counts/min values from test cultures to obtain inhibition indices for each culture. Mean inhibition indices were calculated for each set of cultures performed using the same test lymphocyte population (i.e., lymphocytes obtained from the same individual rat). The inhibition indices obtained in each individual culture were divided by the mean inhibition index and the obtained values expressed as percentages to derive values (normalized inhibition indices) which are comparable between different experiments of the same type despite using different test lymphocyte populations. These percentages were
121
expressed as means and standard errors of means. Differences between groups were analyzed with Student's t-test. Results
Irnmunosuppressive activity in the seminiferous tubules When segments of seminiferous tubules (n = 9) in defined stages of the seminiferous epithelial cycle were cultured with PBL, it was observed that tubules in stages II-VI and VII-VIII inhibited the ConA-stimulated PBL proliferation to the greatest extent (Fig. la). The inhibition was minimal by tubules in stages I X - X I I and increased progressively to reach maximum in stages II-VIII. Inhibitions obtained by tubules in stages II-VI and VII-VIII
~00
tooi
175
175 -I
150
1
125
1
IO0
1
a)
)
75
50
25
IX-XII "from
XlII-I
ll-Vl
IX-XlI, p < 0 , 0 0 5
@(rein Xllld. p < 0.05 Stage
Vii-VIII
=
"from
----
>
>
>
~
>
"0
>
'~
>
>
II, p < 0 0 5
Stage
Fig. 1. (a) Effect of 2 m m seminiferous tubule segments (n = 9) in various stages of the seminiferous epithelial cycle on 3H-TdR incorporation to ConA-stimulated peripheral blood lymphocytes in coculture. The columns and bars represent mean and S.E.M. of normalized inhibition indices. (b) More exact analysis of the stages II-VII, where the production of the immunosuppressive activity was highest. Tubules in stages I I - V I I (n = 17) were cut sequentially into 2-mm segments containing only one type of cell association. The inhibition obtained by tubules in stages V and VI was significantly (P < 0.05) lower than that obtained by the stage II tubules. No significant differences in inhibition were observed between stage II and stage VII tubules. The columns and bars represent mean and S.E.M. of normalized inhibition indices.
122
were significantly higher than those obtained by stages IX-XII (P < 0.005). No significant difference between tubules in stages II-VI and VII-VIII could be observed. Only tubules in stages II-VI inhibited PBL proliferation significantly more than those in stages XIII-I (P < 0.05). In a more detailed analysis of stages II-VIII, seminiferous tubules (n = 17) in these stages were cut sequentially into segments containing only one type b)
a} 200
200
0
o
100
100
#
0
0 Fraction No:
Fraction No:
Fig. 2. (a) Inhibition of lymphocyte proliferation by gel filtration fractions of stage II-VI seminiferous tubule extracts and the absorbance of the eluate at 280 nm, (b) inhibition of lymphocyte proliferation by stage VII-VIII seminiferous tubule extracts and the absorbance of the eluate at 280 nm. The figures are 'means of normalized inhibition indices of 2 experiments (2 cultures in each experiment). Ovalbumin (OVA, 43 kDa), chymotrypsinogen A (CTG, 25 kDa) and ribonuclease A (RIB, 13.7 kDa) are marker proteins used for calibration.
123
of cell association. It was observed that inhibitions obtained by tubules in stages V and VI were significantly lower than those obtained by stage II tubules (Fig. lb). The highest inhibitory activity was present in stage II. A smaller peak of inhibitory activity was localized to stage VIIb, but this peak did not reach statistically significant difference from stages V and VI. No significant difference was observed between stages II and VIIa-d.
Molecular size distribution profile of seminiferous tubule immunosuppressive activity Gel filtration chromatography of extracts of seminiferous tubule segments in stages I I - V I and V I I - V I I I showed one peak of immunosuppressive activity with an apparent Mr of - 2 5 kDa (elution volume: 19.5-22.0 ml, Fig. 2) and another peak with an apparent Mr of - 6 5 kDa (elution volume: 18.0-18.5 ml). In gel filtration of extracts of seminiferous tubule segments in stages II-VI, activity could be observed also at an apparent Mr of - 5 - 1 0 kDa (elution volume: 22.5-24.5 ml). In stages I X - I , only very little activity could be observed, although some activity was present at - 6 5 kDa in stages X I I I - I (data not shown). The absorbance of the eluate at 280 nm showed a peak at 60-70 kDa and a smaller peak at - 10 kDa. Discussion
The present results show that seminiferous tubules in stages II-VIII produce significantly more high molecular weight immunosuppressive activity than tubules in stages I X - I (P < 0.05). This suggests that the activity, in addition to its observed effect on lymphocytes may also have a physiological function in the seminiferous tubules. Such a suggestion is compatible with the previous observation that the levels of high-molecular weight immunosuppressive activity increase in cryptorchidism and hypogonadotrophic hypogonadism (Sainio-P611/inen et al., 1991), in which the seminiferous epithelium undergoes remarkable changes. Also the - 2 5 kDa immunosuppressive factor present in the seminiferous tubule extracts corresponds in size to the - 2 5 kDa immunosuppressive factor observed previously in the abdominal testis and the testis of hypophysectomized rats (Sainio-P611/inen et al., 1991). This suggests that the - 2 5 kDa immunosuppresive factor present in the seminiferous tubules is similar to, or identical with the - 25 kDa activity in the extracts of abdominal testis and the testis of hypophysectomized animals (P611/inen et al., 1990). Because the 25 kDa testicular immunosuppressive factor is especially prominent in experimental cryptorchidism and hypogonadotrophic hypogonadism (Sainio-P611finen et al., 1991), where there are only few germ cells in the seminiferous epithelium, it is possible that
124
the Sertoli cells secrete protectin D. This suggestion is supported by the recent observation that Sertoli cells in vitro secrete an immunosuppressive agent with a Mr of 10-25 kDa (Wyatt et al., 1989). The physiological function of the immunosuppressive activity in the seminiferous tubules is presently unknown. It is possible that it is related to the regulation of germ cell DNA synthesis, since the DNA synthesis in sequentially cut staged seminiferous tubule segments as shown recently by Parvinen et al. (1991) correlates inversely with the levels of immunosuppressive activity in the staged seminiferous tubule segments. Especially in stages II-VII, it can be clearly seen that the immunosuppressive activity drops from the high levels in stage II with low DNA synthesis to significantly decreased levels in stages V and VI, where there is high DNA synthesis in the type B spermatogonia and then rises again in substages VIIa-c, where the DNA synthesis is minimal. In seminiferous tubule segments isolated using the transillumination assisted microdissection technique there is high DNA synthesis in stages I X - X I I (Parvinen et al., 1991). This coincides with the lowest levels of immunosuppressive activity observed in the present experiments. The rise in immunosuppressive activity between stages I X - X I I and X I I l - I is accompanied by a decrease in DNA synthesis. The role of the testicular immunosuppressive activity in regulation of germ cell DNA synthesis remains to be studied, but it seems that the 25-kDa and the 65-kDa testicular immunosuppressive factors, previously named as protectins D and C, respectively (P611finen et al., 1990), present in the seminiferous tubules, are not identical with the previously reported spermatogonial chalone (Clermont and Mauger, 1973), because the spermatogonial chalone has a Mr less than l0 kDa as judged from its ability to pass through Amicon filters (BustosObregon, 1986). The peak of immunosuppressive activity at Mr - 5 - 1 0 kDa present in the stage I I - V I seminiferous tubules may for its part represent spermatogonial chalone not only because of its size but also because the spermatogonial chalone is suggested to act in stage IIl (see de Rooij et al., 1989). It is interesting that the levels of testicular interleukin-l-like (tlL-l) activity are minimal in stage VII (S6der et al., 1991), where there is a peak of testicular immunosuppressive activity. This suggests that the testicular immunosuppressive activity produced by stage VII seminiferous tubules may counteract the effects of tIL-1 on thymocytes in the IL-1 bioassay. However, there is no drop in the levels of testicular interkeukin-l-like activity in stage II, where the levels of testicular immunosuppressive activity are the highest. This may suggest that different factors are responsible for the high immunosuppressive activity in stage II and stage VII or that there is virtually no production of tIL-1 in stage VII, The latter alternative is supported by the present observations that extracts of stage VII-VIII seminiferous tubule segments contain only the 25-kDa and the 65-kDa peaks of immunosup-
125
pressive activity and that the same peaks are present also in stages II-VI seminiferous tubule extracts. Thus, the present results strengthen the previous evidence that tlL-1 may be a growth factor for both spermatogonia and pre-leptotene spermatocytes (P611/inen et al., 1989a; S6der et al., 1991; Parvinen et al., 1991). It may be of physiological significance that there is a small peak of immunosuppressive activity in stage VIIb, because the spermatogenic autoantigens appear on pre-leptotene spermatocytes and spermatogonia in stage VII ( (Yule et al., 1990), thus creating a need for a local immunosuppressive mechanism. In conclusion, the present observations show that the stage II-VIII seminiferous tubules produce high-molecular weight immuonosuppressive activity to a significantly greater extent than stage I X - I seminiferous tubules. This activity is localized to molecular ratios of - 2 5 kDa and - 6 5 kDa in stages II-VIII, but in stages II-VI activity is present also at - 5 - 1 0 kDa. The inverse correlation of the immuonosuppressive activity with DNA synthesis in the seminiferous tubules suggests that this activity may be involved in the control of spermatogonial growth.
Acknowledgments This study was supported by the Nordic Minister Council and the Academy of Finland (proj. no: 1071091).
References B6yum, A. (1968) Isolation of mononuclear cells and granulocytes from human peripheral blood. Scand. J. Clin. Lab. Invest. 21 (Suppl. 97), 77-89. Bustos-Obregon, E. (1986) Spermatogonial proliferation: role of the chalones. In: Andrology: Male Fertility and Sterility (Holstein, ed.), pp. 293-311. Academic Press, New York. Clermont, Y. and Mauger, A. (1973) Existence of a spermatogonial chalone in the rat testis. Cell Tissue Kinet. 7, 165-172. Cohen, J.H.M., Danel, L., Cordier, G., Saez, S. and Revillard, J.-P. (1983) Sex steroid receptors in peripheral T cells: absence of androgen receptors and restriction of estrogen receptors to OKT8-positive cells. J. ImmunoL 131, 2767-2771. De Rooij, D.G., Van Dissel-Emiliani, F.M. and Van Pelt, A.M. (1989) Regulation of spermatogonial proliferation. Ann. N.Y. Acad. Sci. 564, 140-153. Grossman, C.J., Sholiton, L.J. and Nathan, P. (1979) Rat thymic estrogen receptor. I. Preparation, location and physicochemical properties. J. Steroid Biochem. 11, 1233-1240. Leblond, C.P. and Clermont, Y. (1952) Definition of the stages of the cycle of the seminiferous epithelium in the rat. Ann. N.Y. Acad. Sci. 55, 548-573. Maddocks, S. and Setchell, B.P. (1990) Recent evidence for the immune privilege in the testis. J. Reprod. Immunol. 18, 9-18. Parvinen, M. and Ruokonen, A. (1982) Endogenous steroid in rat seminiferous tubules. Comparison of the stages of the epithelial cycle isolated by transillumination-assisted microdissection. J. Androl. 3, 211-220,
126 Parvinen, M., S6der, O., Mali, P., Fr6ysa, B. and Ritz6n, E.M. (1991) In vitro stimulation of stagespecific deoxyribonucleic acid synthesis in rat seminiferous tubule segments by interleukin-lc~. Endocrinology, in press. P611/inen, P., S6der, O. and Uksila, J. (1988) Testicular immunosuppressive protein. J. Reprod. lmmunol. 14, 125-138. P611~inen, P., S6der, O. and Parvinen, M. (1989a) Interleukin-la stimulation of spermatogonial proliferation in vivo. Reprod. Fertil. Dev. 1, 85-87. P611/inen, P., S6der, O., Uksila, J., Nikula, H., Kaipia, A., Kangasniemi, M., Punnonen, J., Huhtaniemi, I. and Parvinen. M. (1989b) Testicular immunosuppressive peptide. Transplantation Proc. 21, 1144-1146.
P611~inen, P., von Euler, M. and S6der, O. (1990) Testicular immunoregulatory factors. J. Reprod. Immunol. 18, 51-76. Raveche, E.S., Vigorsky, R.A., Rice, M.K. and Steinberg, A.D. (1980) Murine thymic androgen receptors. J. Immunopharmacol. 2, 425-435. Sainio-P611/inen, S., P611/inen, P. and Setchell, B.P. (1991 ) Testicular immunosuppressive activity in experimental hypogonadotrophic hypogonadism and unilateral cryptorchidism. J. Reprod. Immunol. 20, 59-72. Sasson, S. and Mayer, M. (1981) Effect of androgenic steroids on rat thymus and thymocytes in suspension. J. Steroid Biochem. 14, 509-517. S6der, O., Syed, V., Callard, G.V., Toppari, J., P611/inen, P., Parvinen, M., Fr6ysa, B. and Ritz~n, E.M. (1991) Production and secretion of an interleukin-I-like factor is stage-dependent and correlates with spermatogonial DNA synthesis in the rat seminiferous epithelium. Int. J. Androl. 14, 223-231. Wyatt, C.R., Law, L., Magnuson, J.A., Griswold, M.D. and Magnuson, N.S. (1988) Suppression of lymphocyte proliferation by protein secreted by cultured Sertoli cells. J. Reprod. Immunol. 14, 27-40. Yule, T.D., Mahi-Brown, C.A. and Tung, K.S.K. (1990) Role of testicular autoantigens and influence of lymphokines in testicular autoimmune disease. J. Reprod. Immunol. 18, 89-103.
117
Elsevier Scientific Publishers Ireland Ltd.
JRI 00777
Immunosuppressive activity in the rat seminiferous tubules P. P611/inen a, M. v o n E u l e r b'c, S. Sainio-P611/inen a, K. J a h n u k a i n e n a, H. H a k o v i r t a a, O. S 6 d e r b a n d M. P a r v i n e n a aDepartment of Anatomy, Institute of Biomedicine, University of Turku, SF-20520 Turku (Finland) and bpediatric Endocrinology Unit and CDepartment of Oncology, Karolinska Institute, S-10401 Stockholm (Sweden) (Accepted for publication 28 February 1992)
Summary Rat seminiferous tubule segments in defined stages of the epithelial cycle were isolated by transillumination-assisted microdissection. The segments were cultured together with ConA-stimulated peripheral blood lymphocytes (PBL) and incorporation of 3H-labelled thymidine was measured. Tubule segments in stages II-VIII of the seminiferous epithelial cycle inhibited PBL proliferation significantly more than stages IX-I. Inhibition was lowest in stages IX-XII and increased progressively to reach a maximum in stages II-VIII. In a more detailed analysis, tubules in stages V and VI inhibited PBL proliferation significantly less than stage II tubules. No significant difference was observed between stages 1I and VII. The immunosuppressive activity had molecular weights of - 25 kDa and - 65 kDa in stage II-VIII seminiferous tubules. In stage II-VI seminiferous tubules activity was present also at - 1 0 kDa. The results suggest that the seminiferous tubules produce highmolecular weight immunosuppressive activity in a stage-dependent way. In addition to its contribution to the immunologically privileged milieu of the testis this activity may also be involved in the physiological regulation of DNA synthesis in the seminiferous epithelium. Key words: rat testis," seminiferous tubules; spermatogenesis; irnmunosuppression; growth factors," protectin
Correspondence to: Dr. Pasi P. P611~inen, Department of Anatomy, Institute of Biomedicine, University of Turku, Kiinamyllynkatu 10, 20520 Turku, Finland. 0165-0378/92/$05.00 © 1992 Elsevier Scientific Publishers Ireland Ltd. Printed and Published in Ireland
118
Introduction
The rodent testicular interstitium is proposed to be an immunologically privileged site, where activation of lymphocytes is inhibited by local products, leading to prolonged survival of intra-testicular allografts in the absence of immunosuppressive drugs (see Maddocks and Setchell, 1990). The nature of the factors involved in the formation and regulation of the immunologically privileged status of the testicular interstitial tissue has been largely unknown. Accumulating evidence indicates a major contribution of testicular soluble factors locally inhibiting lymphocyte activation in a paracrine manner (see P611~nen et al., 1990). Recent data indicate that lymphocytes bear receptors for at least 15 different testicular peptides (see P611/inen et al., 1990) and for estrogens (Cohen et al., 1983), but not for androgens (Grossman et al., 1979; Raveche et al., 1980; Sasson and Mayer, 1981; Cohen et al., 1983). However, in gel filtration chromatography of extracellular fluid from the testicular interstitial tissue (P611~nen et al., 1988, 1989b) and testicular extracts (Sainio-P611~inen et al., 1991) only four peaks of lymphocyte proliferation inhibiting activity with molecular ratios (Mr) of 25, 65, 200 and 400 kDa were observed. One of them, protectin D, with a Mr of - 2 5 kDa (for definition of protectin, see P611~nen et al., 1990), is especially prominent in cryptorchidism and hypogonadotrophic hypogonadism (Sainio-P611finen et al., 1991), in which the seminiferous epithelium undergoes significant changes. This suggests that the 25 kDa testicular immunosuppressive factor may have a physiological function in the seminiferous tubules. The present study was performed to determine whether immunosuppressive activity is produced by the seminiferous tubules and if the production of this activity is dependent on the stage of the seminiferous epithelial cycle. Materials and Methods
Animals Adult Wistar (Turku) or Lewis (ALAB, Stockholm, Sweden) rats were used as donors of lymphocytes and testis tissue. Isolation of seminiferous tubule segments Seminiferous tubule segments in defined stages of the seminiferous epithelial cycle (Leblond and Clermont, 1952) were isolated for co-cultures with lymphocytes using transillumination-assisted microdissection as previously described (Parvinen and Ruokonen, 1982).
119
Seminiferous tubule extracts A total length of 1 m of 2 mm tubule segments in stages II-VI, VII-VIII, I X - X I I and X I I I - I of the seminiferous epithelial cycle were homogenized in a volume of 500 ~1 of saline by vortexing vigorously in an Eppendorf tube for 5 min. The homogenates were centrifuged in an Eppendorf centrifuge with full speed for 15 min. The supernatants were collected and centrifuged again. The second supernatants were used for chromatography. Isolation of peripheral blood mononuclear cells (PBL) Blood was collected with a syringe containing heparin from the right ventricle of the heart whilst saline was infused into the left ventricle. Mononuclear cells were isolated by Ficoll-Paque (5.7% Ficoll 400 solution, Pharmacia Fine Chemicals, Uppsala, Sweden) gradient centrifugation as originally described by B6yum (1968). The isolated cells (average 3 × 107 cells from 50 ml of blood obtained by perfusion) were washed twice with RPMI 1640 medium before culture. The cells were mainly lymphocytes when examined by light microscopy. A few monocytes and some erythrocytes were also present. Co-cultures of seminiferous tubules and lymphocytes Seminiferous tubule segments of 2 mm length were cultured with 2 x 105 peripheral blood lymphocytes (PBL) in 200 ~1 of RPMI 1640 containing 5 /zg/ml Concanavalin A and 5% fetal calf serum (FCS) in an atmosphere of 5% CO2 in air at 37°C for 72 h. After 48 h of culture, 7.4 kBq of [6-3H]thymidine (3H-TdR, specific activity 185 GBq/mmol) in 20 t~l RPMI 1640 was added to each well. The cultures were harvested 16 h later onto glass fiber filter discs using a multiple cell harvester. The filter discs were put into scintillation vials and covered by Optiphase 'hisafe' 3 scintillation fluid (Pharmacia-LKB, Uppsala, Sweden). Radioactivity on the filter discs was measured in a/3-counter. Protectin bioassay The washed peripheral blood mononuclear cells were counted in 0.1% Trypan blue solution and diluted to 4 × 106/ml in RPMI 1640 containing 10% FCS. An aliquot of 50 #1 of the cell suspension was pipetted to wells of a standard 96-well titer plate with U-shaped wells (2 × 105cells/well). The cells were stimulated by adding 50/~1 of 20 t~g/ml ConA solution (final concentration 5/zg/ml) in RPMI 1640 containing 10% FCS. Aliquots of 50 #1 of the seminiferous tubule extract gel filtration fractions were added to the wells. Finally, 50 /zl of serum-free RPMI 1640 was added to each well to reach the final volume of 200 /~1. Each culture was made in triplicate. After 48 h of culture at 37°C in an atmosphere of 5% CO2 in air, 7.4 kBq of
120
6-3H-thymidine in 20 ~1 RPMI 1640 was added to each well. The ceils were harvested 16 h later onto glass fiber filter discs using a multiple cell harvester. Radioactivity on the filter discs was measured using liquid scintillation counting.
HPLC size exclusion chromatography An aliquot of 100 t~l of the seminiferous tubule extract was applied to a TSK G3000SW (7.5 × 600 mm, Pharmacia-LKB, Uppsala, Sweden) gel filtration column. The proteins were eluted with phosphate-buffered saline (PBS) using a flow rate of 1.0 ml/min. Fractions of 0.5 ml were collected. The absorbance of the eluent was monitored at 280 nm. Molecular size standards were thyroglobulin (669 kDa), ferritin (440 kDa), catalase (232 kDa), aldolase (158 kDa), albumin (67 kDa), ovalbumin (43 kDa), chymotrypsinogen A (25 kDa) and ribonuclease A (13.7 kDa) Kay values were calculated using the equation Kav = V e - V o / V t - V o , where Vt is the total bed volume (26.51 ml) and Vo the column void volume (11.76 ml) obtained by measuring elution volume for blue dextran 2000. lie is the elution volume of each standard molecule. The Kay values (on the linear scale) were plotted against the corresponding molecular weights (on a logarithmic scale). Kay values were calculated for the eluted peaks of immunosuppressive activity and their corresponding Mr values estimated from the standard curve. Data expression and statistical analysis Inhibitions were expressed as normalized inhibition indices: (E cpm (control): no) cpm (test) Normalized inhibition index =
~
× 100% (E cpm (control): no) -
-
t/ "
n
cpm (test)
where n~ is the number of control cultures and n t the number of test cultures. Mean counts/min from control cultures were divided by individual counts/min values from test cultures to obtain inhibition indices for each culture. Mean inhibition indices were calculated for each set of cultures performed using the same test lymphocyte population (i.e., lymphocytes obtained from the same individual rat). The inhibition indices obtained in each individual culture were divided by the mean inhibition index and the obtained values expressed as percentages to derive values (normalized inhibition indices) which are comparable between different experiments of the same type despite using different test lymphocyte populations. These percentages were
121
expressed as means and standard errors of means. Differences between groups were analyzed with Student's t-test. Results
Irnmunosuppressive activity in the seminiferous tubules When segments of seminiferous tubules (n = 9) in defined stages of the seminiferous epithelial cycle were cultured with PBL, it was observed that tubules in stages II-VI and VII-VIII inhibited the ConA-stimulated PBL proliferation to the greatest extent (Fig. la). The inhibition was minimal by tubules in stages I X - X I I and increased progressively to reach maximum in stages II-VIII. Inhibitions obtained by tubules in stages II-VI and VII-VIII
~00
tooi
175
175 -I
150
1
125
1
IO0
1
a)
)
75
50
25
IX-XII "from
XlII-I
ll-Vl
IX-XlI, p < 0 , 0 0 5
@(rein Xllld. p < 0.05 Stage
Vii-VIII
=
"from
----
>
>
>
~
>
"0
>
'~
>
>
II, p < 0 0 5
Stage
Fig. 1. (a) Effect of 2 m m seminiferous tubule segments (n = 9) in various stages of the seminiferous epithelial cycle on 3H-TdR incorporation to ConA-stimulated peripheral blood lymphocytes in coculture. The columns and bars represent mean and S.E.M. of normalized inhibition indices. (b) More exact analysis of the stages II-VII, where the production of the immunosuppressive activity was highest. Tubules in stages I I - V I I (n = 17) were cut sequentially into 2-mm segments containing only one type of cell association. The inhibition obtained by tubules in stages V and VI was significantly (P < 0.05) lower than that obtained by the stage II tubules. No significant differences in inhibition were observed between stage II and stage VII tubules. The columns and bars represent mean and S.E.M. of normalized inhibition indices.
122
were significantly higher than those obtained by stages IX-XII (P < 0.005). No significant difference between tubules in stages II-VI and VII-VIII could be observed. Only tubules in stages II-VI inhibited PBL proliferation significantly more than those in stages XIII-I (P < 0.05). In a more detailed analysis of stages II-VIII, seminiferous tubules (n = 17) in these stages were cut sequentially into segments containing only one type b)
a} 200
200
0
o
100
100
#
0
0 Fraction No:
Fraction No:
Fig. 2. (a) Inhibition of lymphocyte proliferation by gel filtration fractions of stage II-VI seminiferous tubule extracts and the absorbance of the eluate at 280 nm, (b) inhibition of lymphocyte proliferation by stage VII-VIII seminiferous tubule extracts and the absorbance of the eluate at 280 nm. The figures are 'means of normalized inhibition indices of 2 experiments (2 cultures in each experiment). Ovalbumin (OVA, 43 kDa), chymotrypsinogen A (CTG, 25 kDa) and ribonuclease A (RIB, 13.7 kDa) are marker proteins used for calibration.
123
of cell association. It was observed that inhibitions obtained by tubules in stages V and VI were significantly lower than those obtained by stage II tubules (Fig. lb). The highest inhibitory activity was present in stage II. A smaller peak of inhibitory activity was localized to stage VIIb, but this peak did not reach statistically significant difference from stages V and VI. No significant difference was observed between stages II and VIIa-d.
Molecular size distribution profile of seminiferous tubule immunosuppressive activity Gel filtration chromatography of extracts of seminiferous tubule segments in stages I I - V I and V I I - V I I I showed one peak of immunosuppressive activity with an apparent Mr of - 2 5 kDa (elution volume: 19.5-22.0 ml, Fig. 2) and another peak with an apparent Mr of - 6 5 kDa (elution volume: 18.0-18.5 ml). In gel filtration of extracts of seminiferous tubule segments in stages II-VI, activity could be observed also at an apparent Mr of - 5 - 1 0 kDa (elution volume: 22.5-24.5 ml). In stages I X - I , only very little activity could be observed, although some activity was present at - 6 5 kDa in stages X I I I - I (data not shown). The absorbance of the eluate at 280 nm showed a peak at 60-70 kDa and a smaller peak at - 10 kDa. Discussion
The present results show that seminiferous tubules in stages II-VIII produce significantly more high molecular weight immunosuppressive activity than tubules in stages I X - I (P < 0.05). This suggests that the activity, in addition to its observed effect on lymphocytes may also have a physiological function in the seminiferous tubules. Such a suggestion is compatible with the previous observation that the levels of high-molecular weight immunosuppressive activity increase in cryptorchidism and hypogonadotrophic hypogonadism (Sainio-P611/inen et al., 1991), in which the seminiferous epithelium undergoes remarkable changes. Also the - 2 5 kDa immunosuppressive factor present in the seminiferous tubule extracts corresponds in size to the - 2 5 kDa immunosuppressive factor observed previously in the abdominal testis and the testis of hypophysectomized rats (Sainio-P611/inen et al., 1991). This suggests that the - 2 5 kDa immunosuppresive factor present in the seminiferous tubules is similar to, or identical with the - 25 kDa activity in the extracts of abdominal testis and the testis of hypophysectomized animals (P611/inen et al., 1990). Because the 25 kDa testicular immunosuppressive factor is especially prominent in experimental cryptorchidism and hypogonadotrophic hypogonadism (Sainio-P611finen et al., 1991), where there are only few germ cells in the seminiferous epithelium, it is possible that
124
the Sertoli cells secrete protectin D. This suggestion is supported by the recent observation that Sertoli cells in vitro secrete an immunosuppressive agent with a Mr of 10-25 kDa (Wyatt et al., 1989). The physiological function of the immunosuppressive activity in the seminiferous tubules is presently unknown. It is possible that it is related to the regulation of germ cell DNA synthesis, since the DNA synthesis in sequentially cut staged seminiferous tubule segments as shown recently by Parvinen et al. (1991) correlates inversely with the levels of immunosuppressive activity in the staged seminiferous tubule segments. Especially in stages II-VII, it can be clearly seen that the immunosuppressive activity drops from the high levels in stage II with low DNA synthesis to significantly decreased levels in stages V and VI, where there is high DNA synthesis in the type B spermatogonia and then rises again in substages VIIa-c, where the DNA synthesis is minimal. In seminiferous tubule segments isolated using the transillumination assisted microdissection technique there is high DNA synthesis in stages I X - X I I (Parvinen et al., 1991). This coincides with the lowest levels of immunosuppressive activity observed in the present experiments. The rise in immunosuppressive activity between stages I X - X I I and X I I l - I is accompanied by a decrease in DNA synthesis. The role of the testicular immunosuppressive activity in regulation of germ cell DNA synthesis remains to be studied, but it seems that the 25-kDa and the 65-kDa testicular immunosuppressive factors, previously named as protectins D and C, respectively (P611finen et al., 1990), present in the seminiferous tubules, are not identical with the previously reported spermatogonial chalone (Clermont and Mauger, 1973), because the spermatogonial chalone has a Mr less than l0 kDa as judged from its ability to pass through Amicon filters (BustosObregon, 1986). The peak of immunosuppressive activity at Mr - 5 - 1 0 kDa present in the stage I I - V I seminiferous tubules may for its part represent spermatogonial chalone not only because of its size but also because the spermatogonial chalone is suggested to act in stage IIl (see de Rooij et al., 1989). It is interesting that the levels of testicular interleukin-l-like (tlL-l) activity are minimal in stage VII (S6der et al., 1991), where there is a peak of testicular immunosuppressive activity. This suggests that the testicular immunosuppressive activity produced by stage VII seminiferous tubules may counteract the effects of tIL-1 on thymocytes in the IL-1 bioassay. However, there is no drop in the levels of testicular interkeukin-l-like activity in stage II, where the levels of testicular immunosuppressive activity are the highest. This may suggest that different factors are responsible for the high immunosuppressive activity in stage II and stage VII or that there is virtually no production of tIL-1 in stage VII, The latter alternative is supported by the present observations that extracts of stage VII-VIII seminiferous tubule segments contain only the 25-kDa and the 65-kDa peaks of immunosup-
125
pressive activity and that the same peaks are present also in stages II-VI seminiferous tubule extracts. Thus, the present results strengthen the previous evidence that tlL-1 may be a growth factor for both spermatogonia and pre-leptotene spermatocytes (P611/inen et al., 1989a; S6der et al., 1991; Parvinen et al., 1991). It may be of physiological significance that there is a small peak of immunosuppressive activity in stage VIIb, because the spermatogenic autoantigens appear on pre-leptotene spermatocytes and spermatogonia in stage VII ( (Yule et al., 1990), thus creating a need for a local immunosuppressive mechanism. In conclusion, the present observations show that the stage II-VIII seminiferous tubules produce high-molecular weight immuonosuppressive activity to a significantly greater extent than stage I X - I seminiferous tubules. This activity is localized to molecular ratios of - 2 5 kDa and - 6 5 kDa in stages II-VIII, but in stages II-VI activity is present also at - 5 - 1 0 kDa. The inverse correlation of the immuonosuppressive activity with DNA synthesis in the seminiferous tubules suggests that this activity may be involved in the control of spermatogonial growth.
Acknowledgments This study was supported by the Nordic Minister Council and the Academy of Finland (proj. no: 1071091).
References B6yum, A. (1968) Isolation of mononuclear cells and granulocytes from human peripheral blood. Scand. J. Clin. Lab. Invest. 21 (Suppl. 97), 77-89. Bustos-Obregon, E. (1986) Spermatogonial proliferation: role of the chalones. In: Andrology: Male Fertility and Sterility (Holstein, ed.), pp. 293-311. Academic Press, New York. Clermont, Y. and Mauger, A. (1973) Existence of a spermatogonial chalone in the rat testis. Cell Tissue Kinet. 7, 165-172. Cohen, J.H.M., Danel, L., Cordier, G., Saez, S. and Revillard, J.-P. (1983) Sex steroid receptors in peripheral T cells: absence of androgen receptors and restriction of estrogen receptors to OKT8-positive cells. J. ImmunoL 131, 2767-2771. De Rooij, D.G., Van Dissel-Emiliani, F.M. and Van Pelt, A.M. (1989) Regulation of spermatogonial proliferation. Ann. N.Y. Acad. Sci. 564, 140-153. Grossman, C.J., Sholiton, L.J. and Nathan, P. (1979) Rat thymic estrogen receptor. I. Preparation, location and physicochemical properties. J. Steroid Biochem. 11, 1233-1240. Leblond, C.P. and Clermont, Y. (1952) Definition of the stages of the cycle of the seminiferous epithelium in the rat. Ann. N.Y. Acad. Sci. 55, 548-573. Maddocks, S. and Setchell, B.P. (1990) Recent evidence for the immune privilege in the testis. J. Reprod. Immunol. 18, 9-18. Parvinen, M. and Ruokonen, A. (1982) Endogenous steroid in rat seminiferous tubules. Comparison of the stages of the epithelial cycle isolated by transillumination-assisted microdissection. J. Androl. 3, 211-220,
126 Parvinen, M., S6der, O., Mali, P., Fr6ysa, B. and Ritz6n, E.M. (1991) In vitro stimulation of stagespecific deoxyribonucleic acid synthesis in rat seminiferous tubule segments by interleukin-lc~. Endocrinology, in press. P611/inen, P., S6der, O. and Uksila, J. (1988) Testicular immunosuppressive protein. J. Reprod. lmmunol. 14, 125-138. P611~inen, P., S6der, O. and Parvinen, M. (1989a) Interleukin-la stimulation of spermatogonial proliferation in vivo. Reprod. Fertil. Dev. 1, 85-87. P611/inen, P., S6der, O., Uksila, J., Nikula, H., Kaipia, A., Kangasniemi, M., Punnonen, J., Huhtaniemi, I. and Parvinen. M. (1989b) Testicular immunosuppressive peptide. Transplantation Proc. 21, 1144-1146.
P611~inen, P., von Euler, M. and S6der, O. (1990) Testicular immunoregulatory factors. J. Reprod. Immunol. 18, 51-76. Raveche, E.S., Vigorsky, R.A., Rice, M.K. and Steinberg, A.D. (1980) Murine thymic androgen receptors. J. Immunopharmacol. 2, 425-435. Sainio-P611/inen, S., P611/inen, P. and Setchell, B.P. (1991 ) Testicular immunosuppressive activity in experimental hypogonadotrophic hypogonadism and unilateral cryptorchidism. J. Reprod. Immunol. 20, 59-72. Sasson, S. and Mayer, M. (1981) Effect of androgenic steroids on rat thymus and thymocytes in suspension. J. Steroid Biochem. 14, 509-517. S6der, O., Syed, V., Callard, G.V., Toppari, J., P611/inen, P., Parvinen, M., Fr6ysa, B. and Ritz~n, E.M. (1991) Production and secretion of an interleukin-I-like factor is stage-dependent and correlates with spermatogonial DNA synthesis in the rat seminiferous epithelium. Int. J. Androl. 14, 223-231. Wyatt, C.R., Law, L., Magnuson, J.A., Griswold, M.D. and Magnuson, N.S. (1988) Suppression of lymphocyte proliferation by protein secreted by cultured Sertoli cells. J. Reprod. Immunol. 14, 27-40. Yule, T.D., Mahi-Brown, C.A. and Tung, K.S.K. (1990) Role of testicular autoantigens and influence of lymphokines in testicular autoimmune disease. J. Reprod. Immunol. 18, 89-103.
