Journal of Reproductive Immunology, 17 (1990) 1U-126 Elsevier Scientific Publishers Ireland Ltd.
115
JR1 00645
Pregnancy zone protein inhibits production of interleukin-2 but does not affect interleukin-2 receptor expression on T cell activation Shigeru Saito”, Hiratsugu Hashimotoa, Kunio Yonemasub and Motohiko Ichijo” ‘Department of Obstetrics and Gynecology, Nara Medical University and bDepartment of Bacteriology, Nara Medical University, Shijo-cho, Kashihara, 634 (Japan) (Accepted for publication 10 November 1989)
Summary The effect of pregnancy zone protein (PZP), which exhibits increased levels in the blood during pregnancy, on T cells was examined. PZP was found to suppress DNA synthesis following stimulation with phytohemagglutinin (PHA), concanavalin A (ConA) or CD3 antigen or in the mixed lymphocyte reaction (MLR). This effect of PZP was mediated by a reduction in interleukin-2 (IL-2) production, and was abolished by exogenous recombinant IL-2 administration. PZP did not affect the proliferation of T cells following stimulation with the calcium ionophore A23187 and phorbol 12-myristate 13-acetate (TPA). These results suggest that PZP acts on the T cell surface and reduces IL-2 production, but not IL-2R expression, and does not directly affect Ca2+influx or protein kinase C. Key words: pregnancy zone protein; interleukin-2; interleukin-2 receptor; T cell activation.
Introduction In the course of pregnancy, a variety of pregnancy-associated proteins and pregnancy-specific proteins appear in the serum. Of these proteins, the following are thought to suppress the blastogenesis of T cells and to be involved in the maintenance of pregnancy: crude fraction of human chorionic gonadotrophin (hCG) (Kaye and Jones, 1971) human placental lactogen (hPL) (Cerni et al., 1977), pregnancy-specific /3-l glycoprotein (SPl) (Cerni et al., 1977), pregnancy-associated plasma protein-A (PAPP-A) (Bischef et al., 1982), pregnancy zone protein (PZP, a,-PAG, SP3, a,-AP glyco0165-0378/90/$03.50 0 1990 Elsevier Scientific Publishers Ireland Ltd. Published and Printed in Ireland
116
gen, PAM) (Von Shoultz et al., 1973; Stimson, 1976), sex steroid hormones (Wyle and Kent, 1977), blocking antibody (Rocklin et al., 1982) and antiidiotypic antibody (Singal et al., 1984). However, the mechanism by which the blastogenesis of T cells is suppressed by these proteins remains obscure. The present study considers the mechanism by which proliferation of T cells is suppressed by PZP, which appears to play an important role in the maintenance of pregnancy as it is found on the surface of the chorionic villi, T cells and monocytes (Chemnitz et al., 1982; Stimson, 1977) and also prolongs the survival of embryonal mouse heart allografts (Svendsen et al., 1978). Materials and Methods Purification of PZP The detailed procedures for purification of PZP have been reported else1989). Briefly, pregnancy serum was applied to a where (Hashimoto, Zn-chelating Sepharose 6B column and the fraction passing through the column was adsorbed onto a CL-Sepharose 4B column coupled with anti-PZP rabbit IgG. PZP was then eluted with a solution of 3 M MgCl, and 0.6 M NaCl (pH 4.5) and purified by removal of IgG with protein A Sepharose 4B and subsequent high-performance liquid chromatography using a Superrose 6. Cells Perip.reral blood mononuclear cells (PBMC) from healthy individuals were collected by Ficoll-Hypaque density centrifugation and maintained in RPM1 1640 medium supplemented with 10% fetal calf serum (FCS) at a concentration of 1 x 106/ml. T-cell activation 0KT3 monoclonal antibody (Ortho Diagnostics, U.S.A.) was coupled to cyanogen bromide-activated Sepharose 4B (Pharmacia Fine Chemicals, Sweden) at a monoclonal antibody to gel ratio of 100 pg: 1 ml. The PBMC were dispensed into a 96-well microplate (200 pi/well) and stimulated for 3 days with 5 pg/ml of PHA (Difco Lab., USA), 5 pg/ml of ConA (Difco, Lab., USA) or l/100 volume of 0KT3 antibody-coupled Sepharose 4B, followed by DNA determination by assessing [3H]thymidine uptake. A one-way MLR was carried out as follows. PBMC from a third party were incubated at 37OC for 60 min in the presence of mitomycin C (50 pg/ ml). The cells were then washed three times with PBS and poured into a 96well microplate (1 x lw/lOO ~1 per well) to serve as a stimulator. Responder cells were added to each well (2 x 105/100 ~1 per well) and incubated for 5 days.
117
T-cell rich PMBC (T-cell content over 95%) were prepared by the E rosette method. To these mononuclear cells A23187 (0.1 @I) and TPA (5 rg/ml) were added. Four days later, [3H]thymidine uptake in the mononuclear cells treated with PZP (200, 100, 50, 25, 12.5 pg/ml) was compared with that in PZP untreated mononuclear cells. IL-2 determination Supernatants were collected from the cultures of mononuclear cells stimulated with PHA, ConA or 0KT3 antibody-coupled Sepharose 4B after 2 days, and from the MLR reaction after 3 days. The IL-2 content of the supematants was bioassayed with CTLL-2 cells as reported elsewhere (Saito et al., 1988) and expressed as units of rIL-2 (Shionogi Pharmacology, Osaka, Japan).
Alteration in suppression of DNA synthesis by PZP following exogenous rIL-2 administration PBMC were stimulated with PHA, ConA, 0KT3 antibody and a one-way MLR in the presence of PZP (100 pg/ml). rIL-2 was added in concentrations of 1 and 100 units/ml to these cells, which were incubated for 3 days following PHA, ConA and OKT3 antibody stimulation and for 5 days in the MLR at 37OC and 5% CO,. After this, 0.8 FCi of [‘Hlthymidine was added and pulses were applied for 4 h. In addition to these, PBMC were stimulated with ConA with or without PZP. Seventy-two hours later, they were washed three times with PBS and suspended in an RPM1 1640 medium (enriched with FCS) to a concentration of 1 x 106/ml. After rIL-2 was added at concentrations of 1, 5, 10, 20, 50 and 100 units/ml, they were incubated for 44 h and 0.8 FCi [3H]thymidine was added and pulses were applied for 4 h. Results
Identification of PZP with SDS-polyacrylamide gel electrophoresis (SDSPAGE) The purified PZP had a single band in SDS-PAGE, showing no contamination with a-,-macroglobulin. Its mol. wt. was 310,000 before reduction and 180,000 after reduction with mercaptoethanol (Fig. 1).
Effect of PZP on blastogenesis of T cells Following stimulation of the PBMC with PHA, ConA, OKT3 antibodycoupled Sepharose 4B, or the MLR reaction, PZP suppressed DNA synthesis in a concentration-dependent fashion. In each reaction system,
118
310KD +
80KD
M(-1
MW
Fig. 1. Silver staining of SDS-PAGE of the purified PZP preparation in the presence of mercaptoethanol [M( + )] or without mercaptoethanol [M( - )]. A single band of 310 kDa was recognized under nonreducing conditions. This band change to 180 kDa under reducing conditions. No other protein was detected in the purified PZP.
DNA synthesis was depressed by about l/2 following treatment with 100 pg/ ml of PZP (Fig. 2). Effect of PZP on IL-2 and IL-2R If PZP inhibits IL-2 or disturbs IL-2R function, PZP would be expected to affect DNA synthesis in IL-2 dependent CTLL-2 cells. In the present study, however, DNA synthesis by CTLL-2 cells was not suppressed by PZP at any of the IL-2 concentrations examined (1 and 100 units/ml). This result suggests that PZP neither acts as an IL-2 inhibitor nor disturbs IL-2R function (Fig. 3). IL-2 production following stimulation of PBMC IL-2 production following stimulation with PHA, ConA, OKT3-coupled sepharose 4B or MLR was markedly reduced by treatment with 100 pg/ml of PZP (Table 1).
119
0:
PHA
0:
ConA
a:
OKTBAb
0:
MLR
m: TPA+A23107
b
Ii.5
2’5
concentration
sb of PZP
Ii0
2i)o
(/rg/ml)
Fig. 2. Mononuclear cell response to PHA, ConA, OKT3 antibody, MLR and T cell response to TPA and A23187 in the presence of PZP. [‘H]TdR of mononuclear cells or T cells with each stimulation without PZP was determined to be 100% [“H]TdR. In each stimulation mononuclear cells or T cells treated with PZP (200, 100, 50, 25, 12.5 pg/ml) is presented as % of [“H]TdR without PZP. 0 - 0 : stimulated by PHA; O-O : stimulated by ConA; A-A : stimulated by 0KT3 antibody; Cl-0 : one way MLR; O-Qstimulated by TPA and A23187. Vertical bars show mean f S.D. in three cases.
Alteration in suppression of DNA synthesis by PZP following exogenous rIL-2 administration The PZP-induced suppression of DNA synthesis following stimulation with PHA, ConA, OKT3-coupled Sepharose 4B or MLR was slightly abrogated by treatment with 1 units/ml of rIL-2. The effect of PZP on mononuclear cells was completely abolished by treatment with 100 units/ ml of IL-2 (Fig. 4). In addition, ConA stimulated cells in the presence of PZP exhibited a capacity to proliferate equal to that in the absence of PZP even with a low IL-2 concentration (1 m 20 units). This indicates that ConA blasts in the presence of PZP possess the same levels of IL-2R and IL-2R function (Fig. 5). Effect of PZP on stimulation with A23187and TPA DNA synthesis in peripheral T cells following stimulation with A23 187 (a calcium ionophore) and TPA (a protein kinase C activator) was not affected at all by PZP (Pig. 2).
120
concentration
of
PZP
-
IL-2
1
IL-2 IU/ml
IOOU/ml
(~g/ml)
Fig. 3. Effect of PZP on proliferation of the IL-2 dependent cell line, CTLL-2. [rH]TdR of CTLL-2 cells at 1 and 100 units/ml r-IL-2 concentration was examined. DNA synthesis by CTLL-2 cells at 1 and 100 units/ml r-IL-2 was not affected by 12.5,25,50, 100 and 200 pg/ml PZP. 0 - 0 : [rH]TdR of CTLL-2 cells at 100 units/ml r-IL-2; A-A : [‘H]TdR of CTLL-2 cells at 1 unit/ml r-IL-2.
Discussion Since Kasakura (1973) reported the finding of non-specific immunosuppressive factors in the serum of pregnant women these have been clarified one by one. Of these factors, PZP is known to suppress the division of T cells stimulated with PHA, ConA or tuberculin (Von Schoultz et al., 1973; Stimson, 1976). However, the mechanism for this effect of PZP remains obscure. Because an extensive amino acid sequence homology exists between PZP and a,-macroglobulin (Sand et al., 1985), it seems likely that most of the results from previous studies on PZP do not reflect the features of PZP but rather those of PZP modified by the coexisting a,-macroglobulin. Some recent investigators have in fact corrected the results of their studies (Povlsen et al,, 1987). Therefore, it now seems necessary for us to review the conventional clinical studies of PZP. In the present investigations we were able to confirm that a preparation of PZP which is not contaminated by a,-macroglobulin suppresses lymphocyte
PZP
C-1 4.91 2.38 0.13
(+Y
1.18 < 0.04 0.04
PZP
t-r
1.87 0.16 0.18
t-1 3.84 0.18 6.22
0.89 0.89 < 0.04
PZP
OKT3Ab
(+I
‘IL-2 production following stimulation without PZP. YL-2 production following stimulation in the presence of 100 &ml of PZP. ‘IL-2 production indicated in units/ml. The detection limit is 0.04 units/ml of IL-2.
Case 2 (Units/ml) Case 3 (Units/ml)
Case 1 (Units/ml~
Con A
PHA
IL-2 production by mononuclear cells stimulated with PHA, ConA, OKT3Ab and MLR with or without PZP.
TABLE I
0.78 0.04 0.04
(+I
2.09 7.20 1.6
(-)
PZP
MLR
< 0.04 0.53 < 0.04
(+I
122
PHA
OKT3Ab
PHA+PZP
OKTBAb+PZP
PHA+PZP+IL-2
OKTSAb+PZP(;l,L;:
W/ml) PHA+PZP+IL-2 (lOOU/ml)
OKT3Ab+PZP+ILm-2 (10OU/ml)
ConA
MLR
ConA+PZP
MLR+PZP
ConA+PZP+IL-2
MLR+PZP+IL-2
W/ml) ConA+PZP+IL-2 (IOOUlml)
MLR+PZP+IL-2
W/ml) (IOOU/ml)
I
0
2 ‘l-l-TdR
4
6
(cpmX
4
L
6
0
10-9
I 2
4 “H-TdR
6 (cpm
X
14
1 O-‘)
Fig. 4. Alteration in suppression of DNA synthesis by PZP following exogenous r-IL-2 administration. DNA synthesis of PHA, ConA, 0KT3 antibody and one-way MLR stimulated mononuclear cells was suppressed by 100 pg/ml of PZP. 1 or 100 units/ml r-IL-2 was added to the cultures and their [)H]TdR was measured. The PZP-induced suppression of DNA synthesis under all stimulations was slightly overcome by treatment with 1 unit/ml of r-IL-2 and completely abolished by treatment with 100 units/ml of r-IL-2.
1 I-* 01
5
IO Concentration
20 of rlL-2
50
100
(U/ml)
Fig. 5. Proliferation response of ConA-blasts in the presence or absence of PZP. DNA synthesis of ConA-stimulated mononuclear cells by 1, 5, 10,20,50 and 100 units IL-2 was examined by [?H]thymidine uptake (see Materials and Methods). 0 - 0 : [)H]thymidine uptake in ConA-blasts in the absence of PZP. n -m : [‘Hlthymidine uptake in Con&blasts in the presence of PZP.
123
blastogenesis following stimulation with PI-IA or ConA and in an MLR reaction. Furthermore, this study revealed that PZP suppresses T cell activation through the CD3 molecule which mediates the intracellular signal transduction when T cells are stimulated by antigen via the T cell receptors (Alcover et al., 1987). In these experiments, DNA synthesis was suppressed by about l/2 by 100 pg/ml of PZP. In the third trimester of pregnancy the average concentration of PZP is 1 mg/ml (Soultz, 1974), hence PZP can suppress T cell proliferation at physiological concentrations. The signals that induce proliferation of T cells are conducted via CD2 and CD3 molecules (Alcover et al., 1987), and the stimulus with PHA, ConA and anti-CD3 antibody is conducted to the cells via either of these two molecules. In the presence of calcium ionophore and protein kinase C activator, T cells begin proliferation without the requirement of signals with CD2 or CD3 molecules (Isakof et al., 1987). The ability of T cells to proliferate following stimulation with Ca ionophore and protein kinase C activator was not suppressed by PZP in the present study. These results lead us to suggest that PZP does not act on the intracellular signal transduction system, but that it suppresses the stimulus conduction system on the cell membrane. Regarding the mechanism for suppression of T cell division by PZP, it was found that PZP does not act as an IL-2 inhibitor or directly act on IL2R, but that it suppresses IL-2 production due to T cell stimulation. As 100 units of extrinsic rIL-2 abolished the ability of PZP to suppress T cell proliferation (Fig. 4), and ConA-stimulated cells treated with PZP exhibited DNA synthesis equal to those in the absence of PZP, even with a low IL-2 concentration (1 m 20 units) (Fig. 5), it seems likely that there is a sufficient amount of high affinity IL-2R expressed even in the presence of PZP. Nicholas et al. (1984) have reported the suppression of MLR-stimulated IL-2 production by retroplacental sera and Saito et al. (1987) have demonstrated the potent ability of pregnancy sera to suppress PHA-stimulated IL-2 production. Our study documents that PZP is one of the factors involved in the suppression of IL-2 production caused by retroplacental sera and sera from pregnant women. When IL-2 production is suppressed, not only T cell division but also all of the following are suppressed: augmentation of NK activity by IL-2 (Kuribayashi et al., 1981), induction of macrophage killer activity (Markousky et al., 1987), induction of cytotoxic T cells (Gills et al., 1979), and induction of lymphokine activated killer cells (Grim et al., 1983). Considering this and the finding that PZP is present also on the trophoblast villus surface (Cemitz et al., 1982), it seems possible that PZP may play an important role in preventing the rejection of the allogeneic fetus by maternal immunocompetent cells. Other factors which suppress T cell proliferation have been reported.
124
Clark et al. (1985, 1988) identified soluble suppressor factors from decidua of mice that inhibit the proliferation of IL-2 dependent cells and downregulate IL-2R expression on T cells. Matsuzaki et al. (1989) reported an immunosuppressive factor, which they called trophoblast-derived immunoregulatory factor, that was shown to suppress the proliferation of T cells co-stimulated with TPA and A23187. This mechanism of immunosuppression is clearly different from that of PZP. Many factors may operate and in different ways, to prevent the maternal rejection of the fetus. A possible mechanism for the ability of PZP to suppress IL-2 production is that the conduction of the signals inducing proliferation of T cells is obstructed by the binding of PZP to the T cell receptors or stimulus-mediating molecules (CD2 or CD3) on the T cell membrane or to the area in the vicinity of these receptors or molecules. In fact, some reports suggest that PZP binds to the CD2 molecule and the DR antigen (Horne et al., 1978; Thomson et al., 1979). Activation of T cells with either protein kinase C or calcium influx stimulates expression of IL-2R. In contrast, neither of the two activates the IL-2 gene, but together they induce levels of IL-2 production (Isakov and Altman, 1985; Isakov et al., 1986). As a result of the binding of PZP on the T cell membrane either of the routes from signal transduction molecules to protein kinase C or calcium influx may be disturbed. This may lead to PZP mediating a reduction of IL-2 production but not IL-2R expression. From now on, we need to study which molecules on the T cell membrane are capable of binding PZP and result in immunosuppression. References Atcover, A., Ramarli, D., Richardson, N.E., Chang, H.C. and Reinherz, E.L. (1987) Functional and molecular aspects of human T lymphocyte activation via T3-Ti and Tll pathways. lmmunol. Rev. 95, S-36. Bischof, P., Lanber, K., de Wurstemberger, B. and Girard (1982) Inhibition of lymphocyte transformation by pregnancy associated plasma protein-A (PAPP-A). J. Clin. Lab. Immunol. 7,61-65. Cerni, C., Tatra, G. and Bohn, H. (1977) Immunosuppression by human placental lactogen (hPL) and the pregnancy specific glycoprotein (SPI). Arch. Gynak. 223, l-7. Chemnitz, J., Han, J., Svendsen, P., Folkersen, J., Westergaard, J-G. and Christensen, B.C. (1982) Immunohistochemical demonstration of human and murine pregancy-associated serum proteins in maternal and placenta1 tissue. Biblo. Anat. 22, 87-92. Clark, D.A., Chaput, A., Walker, C. and Rosenthal, K.L. (1985) Active suppression of host-vs-graft reaction in pregnant mice. VI. Soluble suppressor activity obtained from decidua of allopregnant mice blocks the response to IL-2. J. Immunol. 134, 1659-1664. Clark, D.A., Falbo, M., Rowley, R.B., Banwatt, D. and Clark, J.S. (1988) Active suppression of hostvs-graft reaction in pregnant mice. IX. Soluble suppressor activity obtained from allopregnant mouse decidua that blocks the cytolytic effector response to IL-2 is related to transforming growth factor-p,. J. Immunol. 141,3833-3840. Gills, S., Baker, D.E., Union, N.A. and Smith, K.A. (1979) The in vitro generation and sustained culture of nude cytotoxic T-lymphocytes. J. Exp. Med. 149, 1460-1476. Grim, E.A., Ramsey, K.M., Mazumder, A., Wilson, D.J. and Rosenberg, S.A (1983) Lymphokine acti-
125 vated kiier &l phenomenon. Il. Precursor phenotype is serologically distinct from peripheral blood T lymphocytes, memory cytotoxic thymus derived lymphocytes and natural killer cells. J. Exp. Med. 157,884-887. Hashimoto, H. (1989) Alteration, purification and biological effects of pregnancy zone protein (PZP, pregnancy-associated u-globulin) J. Nara Med. Assoc. 40,29-38. Home, C.H.W., Thomson, A.W., Towler, C.M., MacMillan, F.K. and Gibb, L.M. Relationship of pregnancy-associated alpha 2-glycoprotein (aiPAG) to peripheral blood leukocytes. &and. J. lmmunol. 8,75-80. Isakov, N. and Altman, A. (1985) Tumor promoters in conjunction with calcium ionophores mimic antigenie stimulation by reactivation of alloantigen-primed T lymphocytes. J. lmmunol. 135, 36743680. lsakov, N., Scholz, W. and Altman, A. (1986) Signal transduction and intracellular events in T lymphocyte activation. lmmunol. Today 7,271. lsakov, N., Mally, M.I., Scholz, W. and Altman, A. (1987) T-lymphocyte activation: The role of protein kinase C and the bifurcating inositol phospholipid Aignal transduction pathway. lmmunol. Rev. 95, 89-111. Kasakura, S. (1973) Is cortisol responsible for inhibition of MLC reaction by pregnancy plasma? Nature 246,496-497. Kaye, M.D. and Jones, W.R. (1971) Effects of human chorionic gonadotropin on in vitro lymphocyte transformation. Am J. Obstet. Gynecol. 109, 1029-1031. Kuribayashi, K., Gills, S., Kern, D.E. and Henney, C.S. (1981) Murine NK cell cultures: Effects of interleukin-2 and interferon on cell growth and cytotoxic reactivity. J. lmmunol. 126; 2321-2327. Malkousky, M., Loveland, B., North, M., Asherson, G.L., Gao, L., Ward, P. and Fiers, W. (1987) Recombinant interleukin-2 directly augments the cytotoxity of human monocytes. Nature 325, 262265. Matsuzaki, N., Okada, T., Kameda, T., Negoro, I., Saji, T. and Tanizawa, 0. (1989) Trophoblastderived immunoregulatory factor: Demonstration of the biological function and physicochemical characteristics of the factor derived from choriocarcinoma cell lines. Am. J. Reprod. Immunol. 19, 121-127. Nicholas, N.S., Panayi, G.S. and Nouri, A.M.Z. (1984) Human pregnancy serum inhibits interleukin-2 production. Clin. Exp. lmmunol. 58,587-595. Povlsen, J.V., lngerslev, J. and Petersen, C.M. (1987) Application of an enzyme-linked immunoassay for the measurement of pregnancy zone protein (PZP) in cell culture supernatants and sera. Stand. J. Clin. Lab. Invest 47,207-213. Rocklin, R.E., Kitzmiller, J.L. and Kaye, M.D. Immunology of the maternal-fetal relationship. (1979) Ann. Rev. Med. 30: 375-404. Saito, S., Saito, M., lbaragi, T. and lchijo, M. (1987) Lymphokine activated killer (LAK) activity in normal pregnancy. Jpn. J. Clin. lmmunol. 10,286-292. Saito, S., Saito, M., Kato, Y., Moriyama, I and Ichijo, M. (1988) Interleukin-2 production by human fetal lymphocytes. J. Reprod. Immunol. 14,247-255. Sand, O., Folkersen, J., Westergaard, J.G. and Scottrup-Jensen, L. (1985) Characterization of human pregnancy zone protein: comparison with human a,-macroglobulin. J. Biol. Chem. 260, 1572315735. Singal, D.P., Butler, L., Liao, S.-K. and Joseph, S. (9184). The fetus as an allograft: evidence for antiidiotypic antibodies induced by pregnancy. Am. J. Reprod. Immunol., 6, 145-15 1. Soultz, B.V. (1974) A quantitative study of the pregnancy zone protein in the sera of pregnant and puerperal women. Am. J. Obstet. Gynecol. 119.792-797. Stimson, W.H. (1976) Studies on the immunosuppressive properties of a pregnancy-associated amacroglobulin. Clin. Exp. lmmunol. 25, 159-206. Stimson, W.H. (1977) Identification of pregnancy-associated a-macroglobulin on the surface of peripheral blood leukocyte populations. Clin. Exp. Immunol. 28,4.45-452. Svendsen, P., Stigbrand, T., Teisner, B., Folkersen, J., Damber, M.G., von Echoultz, B., Kemp, E. and Svehag, S.E. (1978) Immunosuppressive effect of human pregnancy zone protein on H-2 incomparible mouse heart allografts. Acta. Pathol. Microbial. Stand [C) 86, 199-201.
126 Thomson, A.W., Hanter, C.B.J., Cruickchank, N. and Horne, C.H.W. (1979) Study of pregnancyassociated alpha 2-glycoprotein in relation to populations of human blood leukocytes. Int. Archs. Allergy Appl Immunol. 58,25 l-259. von Schoultz, B., Stigbrind, T. and Tarnvik, A. (1973) Inhibition of PHA-induced lymphocyte stimulation by the pregnancy zone protein. FEBS Lett. 38,23-26. Wyle, F.A. and Kent, J.R. (19770 Immuno-suppression by sex steroid hormones. Clin. Exp. Immunol. 27,407-415.
115
JR1 00645
Pregnancy zone protein inhibits production of interleukin-2 but does not affect interleukin-2 receptor expression on T cell activation Shigeru Saito”, Hiratsugu Hashimotoa, Kunio Yonemasub and Motohiko Ichijo” ‘Department of Obstetrics and Gynecology, Nara Medical University and bDepartment of Bacteriology, Nara Medical University, Shijo-cho, Kashihara, 634 (Japan) (Accepted for publication 10 November 1989)
Summary The effect of pregnancy zone protein (PZP), which exhibits increased levels in the blood during pregnancy, on T cells was examined. PZP was found to suppress DNA synthesis following stimulation with phytohemagglutinin (PHA), concanavalin A (ConA) or CD3 antigen or in the mixed lymphocyte reaction (MLR). This effect of PZP was mediated by a reduction in interleukin-2 (IL-2) production, and was abolished by exogenous recombinant IL-2 administration. PZP did not affect the proliferation of T cells following stimulation with the calcium ionophore A23187 and phorbol 12-myristate 13-acetate (TPA). These results suggest that PZP acts on the T cell surface and reduces IL-2 production, but not IL-2R expression, and does not directly affect Ca2+influx or protein kinase C. Key words: pregnancy zone protein; interleukin-2; interleukin-2 receptor; T cell activation.
Introduction In the course of pregnancy, a variety of pregnancy-associated proteins and pregnancy-specific proteins appear in the serum. Of these proteins, the following are thought to suppress the blastogenesis of T cells and to be involved in the maintenance of pregnancy: crude fraction of human chorionic gonadotrophin (hCG) (Kaye and Jones, 1971) human placental lactogen (hPL) (Cerni et al., 1977), pregnancy-specific /3-l glycoprotein (SPl) (Cerni et al., 1977), pregnancy-associated plasma protein-A (PAPP-A) (Bischef et al., 1982), pregnancy zone protein (PZP, a,-PAG, SP3, a,-AP glyco0165-0378/90/$03.50 0 1990 Elsevier Scientific Publishers Ireland Ltd. Published and Printed in Ireland
116
gen, PAM) (Von Shoultz et al., 1973; Stimson, 1976), sex steroid hormones (Wyle and Kent, 1977), blocking antibody (Rocklin et al., 1982) and antiidiotypic antibody (Singal et al., 1984). However, the mechanism by which the blastogenesis of T cells is suppressed by these proteins remains obscure. The present study considers the mechanism by which proliferation of T cells is suppressed by PZP, which appears to play an important role in the maintenance of pregnancy as it is found on the surface of the chorionic villi, T cells and monocytes (Chemnitz et al., 1982; Stimson, 1977) and also prolongs the survival of embryonal mouse heart allografts (Svendsen et al., 1978). Materials and Methods Purification of PZP The detailed procedures for purification of PZP have been reported else1989). Briefly, pregnancy serum was applied to a where (Hashimoto, Zn-chelating Sepharose 6B column and the fraction passing through the column was adsorbed onto a CL-Sepharose 4B column coupled with anti-PZP rabbit IgG. PZP was then eluted with a solution of 3 M MgCl, and 0.6 M NaCl (pH 4.5) and purified by removal of IgG with protein A Sepharose 4B and subsequent high-performance liquid chromatography using a Superrose 6. Cells Perip.reral blood mononuclear cells (PBMC) from healthy individuals were collected by Ficoll-Hypaque density centrifugation and maintained in RPM1 1640 medium supplemented with 10% fetal calf serum (FCS) at a concentration of 1 x 106/ml. T-cell activation 0KT3 monoclonal antibody (Ortho Diagnostics, U.S.A.) was coupled to cyanogen bromide-activated Sepharose 4B (Pharmacia Fine Chemicals, Sweden) at a monoclonal antibody to gel ratio of 100 pg: 1 ml. The PBMC were dispensed into a 96-well microplate (200 pi/well) and stimulated for 3 days with 5 pg/ml of PHA (Difco Lab., USA), 5 pg/ml of ConA (Difco, Lab., USA) or l/100 volume of 0KT3 antibody-coupled Sepharose 4B, followed by DNA determination by assessing [3H]thymidine uptake. A one-way MLR was carried out as follows. PBMC from a third party were incubated at 37OC for 60 min in the presence of mitomycin C (50 pg/ ml). The cells were then washed three times with PBS and poured into a 96well microplate (1 x lw/lOO ~1 per well) to serve as a stimulator. Responder cells were added to each well (2 x 105/100 ~1 per well) and incubated for 5 days.
117
T-cell rich PMBC (T-cell content over 95%) were prepared by the E rosette method. To these mononuclear cells A23187 (0.1 @I) and TPA (5 rg/ml) were added. Four days later, [3H]thymidine uptake in the mononuclear cells treated with PZP (200, 100, 50, 25, 12.5 pg/ml) was compared with that in PZP untreated mononuclear cells. IL-2 determination Supernatants were collected from the cultures of mononuclear cells stimulated with PHA, ConA or 0KT3 antibody-coupled Sepharose 4B after 2 days, and from the MLR reaction after 3 days. The IL-2 content of the supematants was bioassayed with CTLL-2 cells as reported elsewhere (Saito et al., 1988) and expressed as units of rIL-2 (Shionogi Pharmacology, Osaka, Japan).
Alteration in suppression of DNA synthesis by PZP following exogenous rIL-2 administration PBMC were stimulated with PHA, ConA, 0KT3 antibody and a one-way MLR in the presence of PZP (100 pg/ml). rIL-2 was added in concentrations of 1 and 100 units/ml to these cells, which were incubated for 3 days following PHA, ConA and OKT3 antibody stimulation and for 5 days in the MLR at 37OC and 5% CO,. After this, 0.8 FCi of [‘Hlthymidine was added and pulses were applied for 4 h. In addition to these, PBMC were stimulated with ConA with or without PZP. Seventy-two hours later, they were washed three times with PBS and suspended in an RPM1 1640 medium (enriched with FCS) to a concentration of 1 x 106/ml. After rIL-2 was added at concentrations of 1, 5, 10, 20, 50 and 100 units/ml, they were incubated for 44 h and 0.8 FCi [3H]thymidine was added and pulses were applied for 4 h. Results
Identification of PZP with SDS-polyacrylamide gel electrophoresis (SDSPAGE) The purified PZP had a single band in SDS-PAGE, showing no contamination with a-,-macroglobulin. Its mol. wt. was 310,000 before reduction and 180,000 after reduction with mercaptoethanol (Fig. 1).
Effect of PZP on blastogenesis of T cells Following stimulation of the PBMC with PHA, ConA, OKT3 antibodycoupled Sepharose 4B, or the MLR reaction, PZP suppressed DNA synthesis in a concentration-dependent fashion. In each reaction system,
118
310KD +
80KD
M(-1
MW
Fig. 1. Silver staining of SDS-PAGE of the purified PZP preparation in the presence of mercaptoethanol [M( + )] or without mercaptoethanol [M( - )]. A single band of 310 kDa was recognized under nonreducing conditions. This band change to 180 kDa under reducing conditions. No other protein was detected in the purified PZP.
DNA synthesis was depressed by about l/2 following treatment with 100 pg/ ml of PZP (Fig. 2). Effect of PZP on IL-2 and IL-2R If PZP inhibits IL-2 or disturbs IL-2R function, PZP would be expected to affect DNA synthesis in IL-2 dependent CTLL-2 cells. In the present study, however, DNA synthesis by CTLL-2 cells was not suppressed by PZP at any of the IL-2 concentrations examined (1 and 100 units/ml). This result suggests that PZP neither acts as an IL-2 inhibitor nor disturbs IL-2R function (Fig. 3). IL-2 production following stimulation of PBMC IL-2 production following stimulation with PHA, ConA, OKT3-coupled sepharose 4B or MLR was markedly reduced by treatment with 100 pg/ml of PZP (Table 1).
119
0:
PHA
0:
ConA
a:
OKTBAb
0:
MLR
m: TPA+A23107
b
Ii.5
2’5
concentration
sb of PZP
Ii0
2i)o
(/rg/ml)
Fig. 2. Mononuclear cell response to PHA, ConA, OKT3 antibody, MLR and T cell response to TPA and A23187 in the presence of PZP. [‘H]TdR of mononuclear cells or T cells with each stimulation without PZP was determined to be 100% [“H]TdR. In each stimulation mononuclear cells or T cells treated with PZP (200, 100, 50, 25, 12.5 pg/ml) is presented as % of [“H]TdR without PZP. 0 - 0 : stimulated by PHA; O-O : stimulated by ConA; A-A : stimulated by 0KT3 antibody; Cl-0 : one way MLR; O-Qstimulated by TPA and A23187. Vertical bars show mean f S.D. in three cases.
Alteration in suppression of DNA synthesis by PZP following exogenous rIL-2 administration The PZP-induced suppression of DNA synthesis following stimulation with PHA, ConA, OKT3-coupled Sepharose 4B or MLR was slightly abrogated by treatment with 1 units/ml of rIL-2. The effect of PZP on mononuclear cells was completely abolished by treatment with 100 units/ ml of IL-2 (Fig. 4). In addition, ConA stimulated cells in the presence of PZP exhibited a capacity to proliferate equal to that in the absence of PZP even with a low IL-2 concentration (1 m 20 units). This indicates that ConA blasts in the presence of PZP possess the same levels of IL-2R and IL-2R function (Fig. 5). Effect of PZP on stimulation with A23187and TPA DNA synthesis in peripheral T cells following stimulation with A23 187 (a calcium ionophore) and TPA (a protein kinase C activator) was not affected at all by PZP (Pig. 2).
120
concentration
of
PZP
-
IL-2
1
IL-2 IU/ml
IOOU/ml
(~g/ml)
Fig. 3. Effect of PZP on proliferation of the IL-2 dependent cell line, CTLL-2. [rH]TdR of CTLL-2 cells at 1 and 100 units/ml r-IL-2 concentration was examined. DNA synthesis by CTLL-2 cells at 1 and 100 units/ml r-IL-2 was not affected by 12.5,25,50, 100 and 200 pg/ml PZP. 0 - 0 : [rH]TdR of CTLL-2 cells at 100 units/ml r-IL-2; A-A : [‘H]TdR of CTLL-2 cells at 1 unit/ml r-IL-2.
Discussion Since Kasakura (1973) reported the finding of non-specific immunosuppressive factors in the serum of pregnant women these have been clarified one by one. Of these factors, PZP is known to suppress the division of T cells stimulated with PHA, ConA or tuberculin (Von Schoultz et al., 1973; Stimson, 1976). However, the mechanism for this effect of PZP remains obscure. Because an extensive amino acid sequence homology exists between PZP and a,-macroglobulin (Sand et al., 1985), it seems likely that most of the results from previous studies on PZP do not reflect the features of PZP but rather those of PZP modified by the coexisting a,-macroglobulin. Some recent investigators have in fact corrected the results of their studies (Povlsen et al,, 1987). Therefore, it now seems necessary for us to review the conventional clinical studies of PZP. In the present investigations we were able to confirm that a preparation of PZP which is not contaminated by a,-macroglobulin suppresses lymphocyte
PZP
C-1 4.91 2.38 0.13
(+Y
1.18 < 0.04 0.04
PZP
t-r
1.87 0.16 0.18
t-1 3.84 0.18 6.22
0.89 0.89 < 0.04
PZP
OKT3Ab
(+I
‘IL-2 production following stimulation without PZP. YL-2 production following stimulation in the presence of 100 &ml of PZP. ‘IL-2 production indicated in units/ml. The detection limit is 0.04 units/ml of IL-2.
Case 2 (Units/ml) Case 3 (Units/ml)
Case 1 (Units/ml~
Con A
PHA
IL-2 production by mononuclear cells stimulated with PHA, ConA, OKT3Ab and MLR with or without PZP.
TABLE I
0.78 0.04 0.04
(+I
2.09 7.20 1.6
(-)
PZP
MLR
< 0.04 0.53 < 0.04
(+I
122
PHA
OKT3Ab
PHA+PZP
OKTBAb+PZP
PHA+PZP+IL-2
OKTSAb+PZP(;l,L;:
W/ml) PHA+PZP+IL-2 (lOOU/ml)
OKT3Ab+PZP+ILm-2 (10OU/ml)
ConA
MLR
ConA+PZP
MLR+PZP
ConA+PZP+IL-2
MLR+PZP+IL-2
W/ml) ConA+PZP+IL-2 (IOOUlml)
MLR+PZP+IL-2
W/ml) (IOOU/ml)
I
0
2 ‘l-l-TdR
4
6
(cpmX
4
L
6
0
10-9
I 2
4 “H-TdR
6 (cpm
X
14
1 O-‘)
Fig. 4. Alteration in suppression of DNA synthesis by PZP following exogenous r-IL-2 administration. DNA synthesis of PHA, ConA, 0KT3 antibody and one-way MLR stimulated mononuclear cells was suppressed by 100 pg/ml of PZP. 1 or 100 units/ml r-IL-2 was added to the cultures and their [)H]TdR was measured. The PZP-induced suppression of DNA synthesis under all stimulations was slightly overcome by treatment with 1 unit/ml of r-IL-2 and completely abolished by treatment with 100 units/ml of r-IL-2.
1 I-* 01
5
IO Concentration
20 of rlL-2
50
100
(U/ml)
Fig. 5. Proliferation response of ConA-blasts in the presence or absence of PZP. DNA synthesis of ConA-stimulated mononuclear cells by 1, 5, 10,20,50 and 100 units IL-2 was examined by [?H]thymidine uptake (see Materials and Methods). 0 - 0 : [)H]thymidine uptake in ConA-blasts in the absence of PZP. n -m : [‘Hlthymidine uptake in Con&blasts in the presence of PZP.
123
blastogenesis following stimulation with PI-IA or ConA and in an MLR reaction. Furthermore, this study revealed that PZP suppresses T cell activation through the CD3 molecule which mediates the intracellular signal transduction when T cells are stimulated by antigen via the T cell receptors (Alcover et al., 1987). In these experiments, DNA synthesis was suppressed by about l/2 by 100 pg/ml of PZP. In the third trimester of pregnancy the average concentration of PZP is 1 mg/ml (Soultz, 1974), hence PZP can suppress T cell proliferation at physiological concentrations. The signals that induce proliferation of T cells are conducted via CD2 and CD3 molecules (Alcover et al., 1987), and the stimulus with PHA, ConA and anti-CD3 antibody is conducted to the cells via either of these two molecules. In the presence of calcium ionophore and protein kinase C activator, T cells begin proliferation without the requirement of signals with CD2 or CD3 molecules (Isakof et al., 1987). The ability of T cells to proliferate following stimulation with Ca ionophore and protein kinase C activator was not suppressed by PZP in the present study. These results lead us to suggest that PZP does not act on the intracellular signal transduction system, but that it suppresses the stimulus conduction system on the cell membrane. Regarding the mechanism for suppression of T cell division by PZP, it was found that PZP does not act as an IL-2 inhibitor or directly act on IL2R, but that it suppresses IL-2 production due to T cell stimulation. As 100 units of extrinsic rIL-2 abolished the ability of PZP to suppress T cell proliferation (Fig. 4), and ConA-stimulated cells treated with PZP exhibited DNA synthesis equal to those in the absence of PZP, even with a low IL-2 concentration (1 m 20 units) (Fig. 5), it seems likely that there is a sufficient amount of high affinity IL-2R expressed even in the presence of PZP. Nicholas et al. (1984) have reported the suppression of MLR-stimulated IL-2 production by retroplacental sera and Saito et al. (1987) have demonstrated the potent ability of pregnancy sera to suppress PHA-stimulated IL-2 production. Our study documents that PZP is one of the factors involved in the suppression of IL-2 production caused by retroplacental sera and sera from pregnant women. When IL-2 production is suppressed, not only T cell division but also all of the following are suppressed: augmentation of NK activity by IL-2 (Kuribayashi et al., 1981), induction of macrophage killer activity (Markousky et al., 1987), induction of cytotoxic T cells (Gills et al., 1979), and induction of lymphokine activated killer cells (Grim et al., 1983). Considering this and the finding that PZP is present also on the trophoblast villus surface (Cemitz et al., 1982), it seems possible that PZP may play an important role in preventing the rejection of the allogeneic fetus by maternal immunocompetent cells. Other factors which suppress T cell proliferation have been reported.
124
Clark et al. (1985, 1988) identified soluble suppressor factors from decidua of mice that inhibit the proliferation of IL-2 dependent cells and downregulate IL-2R expression on T cells. Matsuzaki et al. (1989) reported an immunosuppressive factor, which they called trophoblast-derived immunoregulatory factor, that was shown to suppress the proliferation of T cells co-stimulated with TPA and A23187. This mechanism of immunosuppression is clearly different from that of PZP. Many factors may operate and in different ways, to prevent the maternal rejection of the fetus. A possible mechanism for the ability of PZP to suppress IL-2 production is that the conduction of the signals inducing proliferation of T cells is obstructed by the binding of PZP to the T cell receptors or stimulus-mediating molecules (CD2 or CD3) on the T cell membrane or to the area in the vicinity of these receptors or molecules. In fact, some reports suggest that PZP binds to the CD2 molecule and the DR antigen (Horne et al., 1978; Thomson et al., 1979). Activation of T cells with either protein kinase C or calcium influx stimulates expression of IL-2R. In contrast, neither of the two activates the IL-2 gene, but together they induce levels of IL-2 production (Isakov and Altman, 1985; Isakov et al., 1986). As a result of the binding of PZP on the T cell membrane either of the routes from signal transduction molecules to protein kinase C or calcium influx may be disturbed. This may lead to PZP mediating a reduction of IL-2 production but not IL-2R expression. From now on, we need to study which molecules on the T cell membrane are capable of binding PZP and result in immunosuppression. References Atcover, A., Ramarli, D., Richardson, N.E., Chang, H.C. and Reinherz, E.L. (1987) Functional and molecular aspects of human T lymphocyte activation via T3-Ti and Tll pathways. lmmunol. Rev. 95, S-36. Bischof, P., Lanber, K., de Wurstemberger, B. and Girard (1982) Inhibition of lymphocyte transformation by pregnancy associated plasma protein-A (PAPP-A). J. Clin. Lab. Immunol. 7,61-65. Cerni, C., Tatra, G. and Bohn, H. (1977) Immunosuppression by human placental lactogen (hPL) and the pregnancy specific glycoprotein (SPI). Arch. Gynak. 223, l-7. Chemnitz, J., Han, J., Svendsen, P., Folkersen, J., Westergaard, J-G. and Christensen, B.C. (1982) Immunohistochemical demonstration of human and murine pregancy-associated serum proteins in maternal and placenta1 tissue. Biblo. Anat. 22, 87-92. Clark, D.A., Chaput, A., Walker, C. and Rosenthal, K.L. (1985) Active suppression of host-vs-graft reaction in pregnant mice. VI. Soluble suppressor activity obtained from decidua of allopregnant mice blocks the response to IL-2. J. Immunol. 134, 1659-1664. Clark, D.A., Falbo, M., Rowley, R.B., Banwatt, D. and Clark, J.S. (1988) Active suppression of hostvs-graft reaction in pregnant mice. IX. Soluble suppressor activity obtained from allopregnant mouse decidua that blocks the cytolytic effector response to IL-2 is related to transforming growth factor-p,. J. Immunol. 141,3833-3840. Gills, S., Baker, D.E., Union, N.A. and Smith, K.A. (1979) The in vitro generation and sustained culture of nude cytotoxic T-lymphocytes. J. Exp. Med. 149, 1460-1476. Grim, E.A., Ramsey, K.M., Mazumder, A., Wilson, D.J. and Rosenberg, S.A (1983) Lymphokine acti-
125 vated kiier &l phenomenon. Il. Precursor phenotype is serologically distinct from peripheral blood T lymphocytes, memory cytotoxic thymus derived lymphocytes and natural killer cells. J. Exp. Med. 157,884-887. Hashimoto, H. (1989) Alteration, purification and biological effects of pregnancy zone protein (PZP, pregnancy-associated u-globulin) J. Nara Med. Assoc. 40,29-38. Home, C.H.W., Thomson, A.W., Towler, C.M., MacMillan, F.K. and Gibb, L.M. Relationship of pregnancy-associated alpha 2-glycoprotein (aiPAG) to peripheral blood leukocytes. &and. J. lmmunol. 8,75-80. Isakov, N. and Altman, A. (1985) Tumor promoters in conjunction with calcium ionophores mimic antigenie stimulation by reactivation of alloantigen-primed T lymphocytes. J. lmmunol. 135, 36743680. lsakov, N., Scholz, W. and Altman, A. (1986) Signal transduction and intracellular events in T lymphocyte activation. lmmunol. Today 7,271. lsakov, N., Mally, M.I., Scholz, W. and Altman, A. (1987) T-lymphocyte activation: The role of protein kinase C and the bifurcating inositol phospholipid Aignal transduction pathway. lmmunol. Rev. 95, 89-111. Kasakura, S. (1973) Is cortisol responsible for inhibition of MLC reaction by pregnancy plasma? Nature 246,496-497. Kaye, M.D. and Jones, W.R. (1971) Effects of human chorionic gonadotropin on in vitro lymphocyte transformation. Am J. Obstet. Gynecol. 109, 1029-1031. Kuribayashi, K., Gills, S., Kern, D.E. and Henney, C.S. (1981) Murine NK cell cultures: Effects of interleukin-2 and interferon on cell growth and cytotoxic reactivity. J. lmmunol. 126; 2321-2327. Malkousky, M., Loveland, B., North, M., Asherson, G.L., Gao, L., Ward, P. and Fiers, W. (1987) Recombinant interleukin-2 directly augments the cytotoxity of human monocytes. Nature 325, 262265. Matsuzaki, N., Okada, T., Kameda, T., Negoro, I., Saji, T. and Tanizawa, 0. (1989) Trophoblastderived immunoregulatory factor: Demonstration of the biological function and physicochemical characteristics of the factor derived from choriocarcinoma cell lines. Am. J. Reprod. Immunol. 19, 121-127. Nicholas, N.S., Panayi, G.S. and Nouri, A.M.Z. (1984) Human pregnancy serum inhibits interleukin-2 production. Clin. Exp. lmmunol. 58,587-595. Povlsen, J.V., lngerslev, J. and Petersen, C.M. (1987) Application of an enzyme-linked immunoassay for the measurement of pregnancy zone protein (PZP) in cell culture supernatants and sera. Stand. J. Clin. Lab. Invest 47,207-213. Rocklin, R.E., Kitzmiller, J.L. and Kaye, M.D. Immunology of the maternal-fetal relationship. (1979) Ann. Rev. Med. 30: 375-404. Saito, S., Saito, M., lbaragi, T. and lchijo, M. (1987) Lymphokine activated killer (LAK) activity in normal pregnancy. Jpn. J. Clin. lmmunol. 10,286-292. Saito, S., Saito, M., Kato, Y., Moriyama, I and Ichijo, M. (1988) Interleukin-2 production by human fetal lymphocytes. J. Reprod. Immunol. 14,247-255. Sand, O., Folkersen, J., Westergaard, J.G. and Scottrup-Jensen, L. (1985) Characterization of human pregnancy zone protein: comparison with human a,-macroglobulin. J. Biol. Chem. 260, 1572315735. Singal, D.P., Butler, L., Liao, S.-K. and Joseph, S. (9184). The fetus as an allograft: evidence for antiidiotypic antibodies induced by pregnancy. Am. J. Reprod. Immunol., 6, 145-15 1. Soultz, B.V. (1974) A quantitative study of the pregnancy zone protein in the sera of pregnant and puerperal women. Am. J. Obstet. Gynecol. 119.792-797. Stimson, W.H. (1976) Studies on the immunosuppressive properties of a pregnancy-associated amacroglobulin. Clin. Exp. lmmunol. 25, 159-206. Stimson, W.H. (1977) Identification of pregnancy-associated a-macroglobulin on the surface of peripheral blood leukocyte populations. Clin. Exp. Immunol. 28,4.45-452. Svendsen, P., Stigbrand, T., Teisner, B., Folkersen, J., Damber, M.G., von Echoultz, B., Kemp, E. and Svehag, S.E. (1978) Immunosuppressive effect of human pregnancy zone protein on H-2 incomparible mouse heart allografts. Acta. Pathol. Microbial. Stand [C) 86, 199-201.
126 Thomson, A.W., Hanter, C.B.J., Cruickchank, N. and Horne, C.H.W. (1979) Study of pregnancyassociated alpha 2-glycoprotein in relation to populations of human blood leukocytes. Int. Archs. Allergy Appl Immunol. 58,25 l-259. von Schoultz, B., Stigbrind, T. and Tarnvik, A. (1973) Inhibition of PHA-induced lymphocyte stimulation by the pregnancy zone protein. FEBS Lett. 38,23-26. Wyle, F.A. and Kent, J.R. (19770 Immuno-suppression by sex steroid hormones. Clin. Exp. Immunol. 27,407-415.
