Developmental and Comparative Immunology,Vol. 16, pp. 425-430, 1992 Printed in the USA. All rights reserved.

0145-305X/92 $5.00 + .00 Copyright © 1992 Pergamon Press Ltd.

IMMUNE SIGNALLING AT THE MATERNAL-FETAL INTERFACE AND TROPHOBLAST DIFFERENTIATION T h o m a s G. W e g m a n n and Larry J. G u i l b e r t Department of Immunology, University of Alberta, 8-65 Medical Sciences Building, Edmonton, Alberta T6G 2H7 Canada (Submitted August 1991; Accepted September 1991)

DAbstract--In this review the afferent and efferent signals involved in immune signalling at the maternal-fetal interface are highlighted in the light of recent information. MHC antigen expression is reviewed. Immunizing mothers against class I and II MHC antigens can prevent spontaneous fetal resorption in mice. In addition, the CSF family of cytokines is not only secreted in placental tissues, but also play a role in trophoblast proliferation and differentiation. GM-CSF in particular appears to promote trophoblast syncytialization and the synthesis of human chorionic gonadotropin and human placental lactogen. Recent knock-out experiments indicate that CSF-1 is essential for complete fertility. Finally, the fact that the CSF cytokines promote HIV release from macrophages indicates that the knowledge gained in this area could lead to a better understanding of the transmission of the HIV virus from the mother to the fetus through the trophoblast. []Keywords--MHC; Fetal resorption; Trophoblast; GM-CSF; CSF-1; Fertility; Cytokines; HIV. Introduction In considering the m a t e r n a l - f e t a l immune interaction two reasonably wellgrounded facts must be kept in mind. The first fact is that it seems to be unnecessary to have an intact maternal immune system to reproduce. A number of Address correspondence to Dr. Thomas G. Wegmann.

observations have been made that lead to this conclusion, including the observation that doubly mutant SCID-beige mice, lacking B- and T-cell function and virtually all N K function, can nevertheless successfully reproduce (1). In addition, mice that are homozygous for the [3-2 microglobulin k n o c k - o u t mutation are also capable of reproducing in the absence of virtually all M H C class I expression at the cell surface (2,3). The second fact is that there is nonetheless a useful role for maternal immunity in reproduction. By now, it is a b u n d a n t l y clear in mice that immunization against paternal and third party alloantigens can prevent recurrent spontaneous fetal resorption, while leading to increased fetal and placental size (4,5). The same has been claimed to be true in humans, but since proper clinical trials are difficult to organize, the claim remains controversial (6). Recent work indicates that reimmunizing primed human females at certain time w i n d o w s during p r e g n a n c y with paternal leukocytes can dramatically influence live birth weights in a positive fashion (7). In order to better comprehend the mechanisms of maternal immune interaction in the placenta leading to this type of effect, a brief overview of recent developments involving M H C signalling, u n u s u a l e f f e c t o r cells involved in abortion, and the role of cytokines in reproduction will be examined here.

425

426

Afferent Signalling: The Role of MHC in Reproduction In human population studies there have been a number of studies which show that females who share MHC class I and/or class II antigens with their husbands have increased spontaneous abortion. Other studies, however, have failed to reproduce these findings, although when all the studies are considered together MHC mismatch appears to have a positive effect in most studies with a good deal of heterogeneity in the populations studied (8). One of the clearest studies is that of Ober and her colleagues, who have shown that Hutterite women living in a relatively closed colony show significantly more early abortion, as measured by HCG surge, if they share HLC antigens with their husbands [(9,10), and personal communication, 1991]. Recent studies have clarified the nature of the class I MHC antigens expressed on human extravillous trophoblast, which is the type of trophoblast that erodes through maternal basement membrane, replaces vascular endothelium in the uterine spiral arteries, and eventually metastasizes to the lungs during normal pregnancy. The antigen expressed is H L A - G and is apparently unique to the trophoblast (ll). How it relates to the above observations is not clear because it is a monomorphic antigen, but a number of investigators are now actively pursuing what role it may play in immune signalling at the maternal-fetal interface. For example, it is not known whether adults are tolerant to it. In addition, there is a 37 kDa acidic isoform of HLA-G which appears to be secreted and not expressed in the cell membrane, because it can be immunoprecipitated from tissue culture and not surface labelled by lactoperoxidase (12). It will be most interesting to see whether HLA-G influences the unusual effector

T.G. Wegmann and L. J. Guilbert

cells present at the maternal-fetal interface that interacts with trophoblast, as described in the next section. Beyond the expression of MHC antigens in p l a c e n t a is the quest i on of whether they bear any relevance to reproductive function. This question has been answered in the affirmative, for the first time, in a series of experiments using an induced abortion model in mice to be elaborated on below (13). In brief, the double-stranded RNA molecule poly IC will induce fetal resorption if given on day 3 after conception. We have recently shown that one can reverse this abortion, which mimics natural spontaneous fetal resorption through the stimulation of activated N K cells, by immunizing mice with cells which differ mutationally at single class I genes, employing the socalled bm series of mutant C57B1/6 mice. In addition, L cells transfected with class I genes also reverse the fetal resorption when c o m p a r e d with n o n t r a n s f e c t e d control L cells. Finally, we have confirmed these results using peptides prepared by Dr. B. Singh at the University of Alberta that represent immunodominant regions of classical MHC class I genes. An interesting observation is that the class I antigens do not have to be paternal in type. Thus, immune reactions directed at a third party class I product can serve to generate the type of immune response in the mother favorable to fetal survival. This is perhaps not surprising, in view of experiments from V. Toder's lab in which complete Freund's adjuvant can reverse the high spontaneous fetal resorption rate seen in CBA x DBA/2J mice (14). Thus, for the first time, antiMHC reactivity of the mother has been clearly shown to be involved in enhancing reproductive performance, although the CFA and third party class I experiments suggest that the response can be systemic in origin and not necessarily initiated by stimulation at the f e t a l placental unit. Possible mechanisms of

Trophoblast differentiation

how this might be working will be considered later in this review.

Efferent Signalling The Role of Natural Killer Cells in Spontaneous Abortion Recent evidence indicates that there may be a role for a type of natural killer cell in spontaneous fetal resorption in mice. Initial interest in these cells was stimulated when Baines and his colleagues showed that their local presence in the placenta correlated with increased fetal resorption seen in CBA x DBA-2 matings. The fetal resorption could be prevented by injecting the pregnant mice with anti-asialo-GM-1 antibody, which destroys the NK cells (18). In addition, one could enhance fetal resorption using double-stranded RNA, which stimulates an increase in the activity of these cells, no doubt acting through the elevation of ~/interferon levels. Furthermore, the resorption can be adoptively transferred by using spleen cells from poly ICtreated mice, and this activity disappears upon treatment of the spleen cells with anti-asialo-GM-1 antibody (19). Thus, there is a reasonably direct correlation b e t w e e n the natural killer-like cells found at the maternal-fetal interface and the increased spontaneous resorption seen in CBA x DBA/2 mice. As pointed out above, one can reverse this type of fetal resorption by injecting class I MHC alloantigens into the mother prior to pregnancy. The exact mechanisms by which this happens have yet to be clarified, but there is no doubt that an understanding of the cytokines involved in these interactions will play a role in elucidating the mechanisms involved. Analogous cells have been found in the human decidua as well. These so-called large granular lymphocytes are of bone marrow origin and in the human have

427

been shown to kill first trimester trophoblast cells following IL-2 stimulation in culture (16). This has yet to be demonstrated in mice, although the murine IL2-activated cells bearing N K markers (termed granulated metrial gland cells) can kill YAC but not p815 target cells (16,17).

The Role of Cytokines in Maternal-Fetal Immune Communication A number of observations have led various authors to postulate a role for lymphohematopoietic cytokines in the reproductive process. For example, a number of years ago we reported that IL3, GM-CSF, and CSF-I stimulated the proliferation of murine placental cells and enhanced their phagocytic capability (20). More recently we have shown that the responding cells are most likely trophoblast, because their cytoskeleton is c y t o k e r a t i n - p o s i t i v e and vimentinnegative (Mogil, R., manuscript submitted). Pollard and his colleagues have described experiments in which CSF-1 is produced in fairly large quantities in uterine epithelium, and this production is under hormonal influence, rising rapidly under the combined influence of estrogen and progesterone seen during early stages of pregnancy (21). In addition, cells of the trophoblast bear CSF receptors (22,23). Our recent experiments indicate that GM-CSF and CSF-1 both play a differentiative role in human trophoblast ( G a r c i a - L l o r e t , M., et al., manuscript submitted). Cytotrophoblast cells eventually join together to form syncytiotrophoblast, which forms the principal layer between the maternal and fetal circulations. Trophoblast syncytialization is enhanced by the presence of GM-CSF and CSF-1. Antibodies to GMCSF inhibit this process in vitro. In addition, both of these cytokines enhance

428

production of human placental lactogen and human chorionic gonadotropin in placental cell preparations in vitro. In addition, the human choriocarcinoma cell lines J E G , JAR, and B E W O , thought to represent trophoblast derivatives, spontaneously produce GM-CSF and CSF-1, and their proliferation can be inhibited by antibodies that react with GM-CSF and also antibodies that react with the receptor for CSF-I. Thus, there is developing evidence to suggest that these cytokines may be autocrine for at least some forms of trophoblast and that they may be serving a differentiative role in the trophoblast. Recently we have found by in situ hybridization that GM-CSF is expressed in the spongiotrophoblast layer in the murine placenta (24). GM-CSF may also play a role in the earlier phases of reproductive performance. GM-CSF enhances the implantation of mouse embryos into uterine glandular epithelium in vitro, suggesting that it may be membrane active (25). In addition, GM-CSF when injected in vivo can prevent fragmentation of mouse embryos, and in particular the trophectoderm surrounding the embryo in strains where it is prone to occur in vitro (26). We have shown that GM-CSF is coded for by a series of unusually high molecular weight messenger RNAs in the placenta (27). A similar phenomenon with respect to CSF-I expression in the mouse placenta is clearly caused by differential exon splicing, changing the protein so that it becomes membrane bound (Pollard, J. W. personal communication). These authors have further shown that females bearing a spontaneous mutation that blocks CSF-1 expression in the mouse show fertility, but it is reduced (28,29). If both the male and the female are homozygous for this nonproductive mutation, fertility is completely eliminated. No doubt, knock-out experiments involving other cytokines will provide a

T.G. Wegmann and L. J. Guitbert

clearer picture of the role of all of the cytokines that have been found in the placenta to date in enhancing reproductive performance. But it is probably not premature at this stage to speculate that some variety of MCH or other antigen stimulation at the maternal-fetal interface during normal pregnancy leads to the paracrine production of cytokines through maternal immune interactions at the maternal-fetal interface. These cytokines could then enhance the growth and function of the fetal trophoblast, and ultimately of the embryo itself, perhaps through the release of placental lactogens. Although such paracrine interactions are not necessary for successful reproduction, they no doubt enhance it. The fact that GM-CSF and CSF-1 can also be produced by fetal placental cells, including trophoblasts, indicates that the cytokines themselves may be centrally important to pregnancy but that they could originate either from fetal (nonimmune) or maternal (immune) sources.

These observations find relevance when considering the growing problem of perinatal AIDS. GM-CSF and CSF-I are potent stimulators of H1V release from infected macrophages. We have recently shown that syncytiotrophoblast displays CD4 molecules on its surface, as well as Fc-,/RIII receptor molecules. Gpl60 can bind to spongiotrophoblast cells and this binding can be inhibited with anti-CD4 antibodies (30). We have recently developed glass bead columns that are capable of isolating virtually 100% pure trophoblast, and now are in the process of seeing whether HIV infectivity is influenced by the lymphohematopoietic cytokines listed above. Preliminary experiments in collaboration with L. Guilbert and M. Meltzer indicate that GM-CSF and CSF-I do play a role here, but more experiments are necessary in order to clarify that role. It may be ironic that the very cytokines that

Trophoblast differentiation

429

drive differentiation of the trophoblastic interface between the mother and the fetus may also enhance the infectability of the fetus by HIV virus coming from the maternal side. An understanding of these events will hopefully lead to more intelligently designed therapies to prevent maternal to fetal transmission of the HIV

virus, as well as to novel means for controlling reproductive performance.

Acknowledgements--Some of the work described in this paper was supported by the Medical Research Council of Canada and Health and Welfare Canada.

References 1. Croy, B. A.; Chapeau, C. Evaluation of pregnancy immunotrophism hypothesis by assessment of the reproductive performance of young adult mice of genotype scid/sic.bg/bg. J. Reprod. Fertil. 88:231-239; 1990. 2. Zijlstra, M. ; Bix, M. ; Simister, N. E.; Loring, J. M.; Raulet, D. H.; Jaenisch, R. 132microglobulin deficient mice lack CD4-8 + cytolitic T cells. Nature 344:742-746; 1990. 3. Koller, B. H.; Marrack, P.; Kappler, J. W.; Smithies, O. Normal development of mice deficient in 132M, MHC class I proteins, and CD8 plus T cells. Science 248:1227-1230; 1990. 4. Chaouat, G. ; Kiger, N.; Wegmann, T. G. Vaccination against spontaneous abortion in mice. J. Reprod. Immunol. 5:389-392; 1983. 5. Chaouat, G.; Kolb, J. P.; Chaffaux, S.; Riviere, M.; Lankar, D.; Athanassakis, 1.; Green, D.; Wegmann, T. G. The placenta and the survival of the fetal allograft. In: Gill, T. J. III; Wegmann, T. G., eds. Immunoregulation of fetal survival. New York: Oxford University Press; 1987:239-251. 6. Section IX: Round Table--Immune abortion and immunotherapy in human: The controversy. In: Chaouat, G.; Mowbray, J., eds. Cellular and molecular biology of the maternofetal relationship. Colloque Inserm vol. 212. Montrouge, France: John Libbey Eurotext Ltd.; 1991:261-306. 7. Mowbray, J. E; Underwood, J. L. Effect of paternal lymphocyte immunisation on birthweight and pregnancy outcome. In: Chaout, G.; Mowbray, J., eds. Cellular and molecular biology of the materno-fetal relationship. Colloque Inserm vol. 212. Montrouge, France: John Libbey Eurotext Ltd.; 1991:295-302. 8. Gill, T. J. Immunogenetics of recurrent abortions in experimental animals and in humans. In: Chaout, G., ed. The immunology of the fetus. Boca Raton, FL: CRC Press, Inc.; 1990: 293-306. 9. Ober, C. L.; Martin, A. O.; Simpson, J. L.; Hauck, W. W.; Amos, D. B.; Kostyu, D. D.; Fotino, M.; Allen, E H., Jr. Shared HLA antigens and reproductive performance among Hutterites, Am. J. Hum. Genet. 35:994-1004; 1983. 10. Ober, C. L.; Hauck, W. W.; Kostyu, D. D.; O'Brien, B.; Elias, S.; Simpson, J. L.; Martin,

I 1.

12.

13. 14. 15.

16.

17.

18.

19.

20.

21.

22.

A. O. Adverse effects of human leukocyte antigen-DR sharing on fertility: A cohort study in a human isolate. Fertil. Steril. 44:227-232; 1985. Ellis, S. A.; Palmer, M. S.; McMichael, A. J. Human trophoblast and the choriocarcinoma cell line BeWo express a truncated HLA class I molecule. J. Immunol. 144:731-735; 1990. Kovats, S.; Main, E. L.; Librach, C.; Stubblebine, M.; Fisher, S. J.; DeMars, R. A class I antigen, HLA-G, expressed in human trophoblasts. Science 248:220-223; 1990 Kinsky, R.; Capovic, M.; Menu, E. ; et al. (submitted). Toder, V.; Strassburger, D. Non-specific immunopotentiation and pregnancy loss. Res. Immunol. 141:181-184; 1990. King, A.; Loke, Y. W. Human trophoblast and JEG choriocarcinoma cells are sensitive to lysis by IL-2 stimulated decidual NK cells. Cell. Immunol. 129:435-448; 1990. Croy, B. A.; Reed, N.; Malashenko, B-A.; Kim, K.; Kwon, B. S. Demonstration of YAC target cell lysis by murine granulated metrial gland cells. Cell. Immunol. 133:116-126; 1991. Linnemeyer, P. A.; Pollack, S. B. Murine granulated metrial gland cells at uterine implantation sites are matural killer lineage cells. J. Immunol. 147:2530-2535; 1991. Gendron, R. L.; Farookhi, R.; Baines, M. G. Resorption of CBA/J x DBA/2 mouse conceptuses in CBA/J uteri correlates with failure of the feto-placental unit to suppress natural killer cell activity. J. Reprod. Fetil. 89:277284; 1990. Kinsky, R.; Delage, G.; Rosin, N.; Thang, M. N.; Hoffmann, M.; Chaouat, G. A murine model of NK cell mediated resorption. Am. J. Reprod. Immunol. 23:73-77; 1990. Athanassakis, I.; Bleackley, R. C.; Paetkau, V.; Guilbert, L. ; Barr, P. J.; Wegmann, T. G. The immunostimulatory effect of T cells and T cell lymphokines on murine fetally-derived placental cells. J. Immunol. 138:37-44; 1987. Bartocci, A.; Pollard, J. W.; Stanley, E. R. Regulation of colony stimulating factor 1 during pregnancy. J. Exp. Med. 164:956-961; 1986. Arceci, R. J.; Shanahan, F.; Stanley, E. R.; Pollard, J. W. Temporal expression and location of colony-stimulating factor 1 (CSF-1) and

430

23.

24.

25.

26.

T.G. Wegmann and L. J. Guilbert

its receptor in the female reproductive tract are consistent with CSF-1 regulated placental development. Proc. Natl. Acad. Sci. USA 86: 8818-8822; 1989. Regenstreif, L. J.; Rossant, J. Expression of the c-fms proto-oncogene and of the cytokine, CSF-1, during mouse embryogenesis. Dev. Biol. 133:284-294; 1989. Kanzaki, H.; Crainie, M.; Lin, H.; Yui, J.; Guilbert, L. J.; Wegmann, T. G. The in situ expression of granulocyte-macrophage colony stimulating factor (GM-CSF) mRNA at the maternal-fetal interface. Growth Factors 5:69-74; 1991. Robertson, S.; Lavranos, T. S.; Seamark, R. E In vitro models of the maternal-fetal interface. In: Wegmann, T. G.; Nisbett-Brown, E.; Gill, T. J., I11, eds. Molecular and cellular immunology of the maternal-fetal interface. New York: Oxford University Press; 1991: 191-206. Tartakovsky, B.; Ben-Yair, E. Abnormal development of CBA/J-female derived embryos: Restoration by GM-CSE Dev. Biol. (in press).

27. Crainie, M.; Guilbert, L.; Wegmann, T. G. Expression of novel cytokine transcripts in the murine placenta. Biol. Reprod. 43:999-1005: 1990. 28. Wiktor-Jedrzejczak, W.; Bartocci, A.; Ferrante, A. W.; A h m e d - A n s a r i , A.; Snell, K. W.; Pollard, J. W.; Stanley, E. R. Total absence of colony stimulating factor-I in the macrophage-deficient esteopetrotic (op/op) mouse. Proc. Natl. Acad. Sci USA 87:48284832; 1990. 29. Wiktor-Jedrzejczak, W.: Urbanowska, E.; Aukerman, S. L.; Pollard, J. W.; Stanley, E. R.; Ralph, P.; Ansari, A. A.; Sell, K. W.; Szperl, M. Correction by CSF-I of defects in osteopetrotic op/op mouse suggests local, developmental and humoral requirements for this growth factor. Exp. Hematol. (in press). 30. David, F. J. E.; Autran, B.; Tran, H. C.; Menu, E. ; Raphael, M. ; Debre, P.; Hsi, B. L. ; Wegmann, T. G.; Barre-Sinousi, F.; Chaouat, G. CD4-positive human trophoblast cells may explain transmission of HIV-1 infection from mother to child. Clin. Exp. lmmunol. (in press).

Immune signalling at the maternal-fetal interface and trophoblast differentiation.

In this review the afferent and efferent signals involved in immune signalling at the maternal-fetal interface are highlighted in the light of recent ...
422KB Sizes 0 Downloads 0 Views