Hypothesis to unify an embryonic defence system and HIV infection.

rejected? Why don’t all females make anti-sperm antibodies S (equivalent to glycodelin A but in seminal plasma) Clark when inseminated with spermatozoa? These fundamental quesprovides evidence that both forms of glycodelin contain tions, which present difficult intellectual and experimental immunosuppressive moieties that are mimicked by pathogenic challenges to the scientist, are yet to be resolved. Recently, organisms, e.g. schistosomes express N-linked terminal fucoClark et al. (1996, 1997) have presented their novel feto– sylated LacdiNac sequences (the same as found in glycodelin embryonic defence system (Hu-FEDS) hypothesis which proA), and terminal LeX/LeY active sequences (the same as found poses mechanisms whereby the fetus can survive in the uterus on glycodelin S). Interestingly, LeY glycans have a greater by evading immune destruction (Clark et al., 1996). This expression on HIV infected cells and, although the precise hypothesis, and its extension (Clark et al., 1997) which role played by these glycans on the immune system is unknown, considers the evasion of the immune response by pathogens it is likely that they induce aberrant immune responses or [e.g. human immunodeficiency virus (HIV)] with its parallels mediate signals that may protect cells from an adverse immune to the reproductive system, is thought-provoking, challenging response. and well presented. The arguments must now stimulate scientIn contrast to the intensity of research investigating the ists to critically evaluate, and subsequently test by experimentarelationship between the immune system and survival of the tion, many of the predictions suggested in order that significant fetus, there is a paucity of knowledge (and ideas) about survival progress can be made in this critical area of reproductive and transport of spermatozoa in the female tract. Spermatozoa biology. enter a very hostile environment. Within minutes of entering The first presentation of the Hu-FEDS hypothesis suggested cervical mucus, spermatozoa (not seminal plasma) induce (with evidence) that specific carbohydrate moieties could block a massive influx of leukocytes (Pandya and Cohen, 1985; initial sperm egg binding, selectin-mediated adhesions and Thompson et al., 1992). This appears to be a poor strategy for suppress lymphocyte-mediated responses and thus, using such survival but preliminary data suggest that in contrast to the mechanisms, the fetus would be able to nullify the immune bovine, these leukocytes do not phagocytose the spermatozoa response (Clark et al., 1996). The hypothesis primarily emphas(Barratt et al., 1990). Many questions remain about this ized the role of glycodelin A which, in addition to inhibiting phenomenon, e.g. the molecular stimulus for this infiltration. sperm–zona interaction, was demonstrated in vitro to have What proportion of sperm are phagocytosed? Is phagocytosis immunosuppressive activities on both lymphocytes and natural selective, i.e. are the good or bad spermatozoa phagocytosed? killer (NK) cells. It was suggested that ‘glycodelin A inhibits How do the phagocytes recognize the spermatozoa etc? Whatsuch diverse immune responses because its unusual glycans ever the dynamics of the system, enough spermatozoa clearly are similar in structure to those attached to immune signalling survive to achieve acceptable pregnancy rates. As neutrophils or adhesion proteins’. The temporal and spatial expression of are poor antigen-presenting cells, perhaps this is one mechanglycodelin A supported its role in providing an immunosuppresism to reduce the antigen load. In addition, even if phagocytosed sive environment for the embryo. by antigen-presenting cells, the spermatozoa (following the In an extension of the Hu-FEDS hypothesis, Clark et al. ideas of Clark et al., 1987) could subvert the immune recogni(1997) suggest that pathogens including HIV and schistosomes tion system, for example by blocking class I presentation thus use the same (or very similar) carbohydrate dependent adheevading a T cell response. Such a mechanism is known to sion/signalling systems to protect themselves from the immune occur in viral infections, e.g. cytomegalovirus (CMV) (Gilbert system as suggested for human gametes and the developing et al., 1996). In addition to these systems, many other varied fetus. Therefore, a detailed understanding of this system is mechanisms could be used to avoid an immune response, e.g. absolutely necessary, not only for reproductive biologists, but limited expression of histocompatibility antigens (Guillaudeux also for other scientists wishing to dissect the mechanisms et al., 1996), absorption of seminal plasma products which are used by pathogens to mediate their effectiveness. In extending known to be immunosuppressive, e.g. glycodelin S, prostathe Hu-FEDS hypothesis to HIV, Clark shows that gametes and glandins (Kelly, 1995) etc. However, to determine the physioHIV are potential carriers of immunosuppressive carbohydrate logical significance of these mechanisms, experiments must sequences. For example, the biantennary N-linked glycan be related to the situation in vivo. For example, does glycodelin carrying the bisecting GlcNAc sequence suppresses NK cell