Adrian Bot, Stefan Antohi and Constantin Bona

The Department of Microbiology, Mount Sinai Medical School, New York, USA

Received 4/8/97; Accepted 4/15/97; On-line 5/1/97

2. INTRODUCTION

Early studies demonstrated that the immune responsiveness varies with the stage of the development. Countless observations showed that during the early stages of postnatal development or during the fetal period, exposure to antigens was followed by unresponsiveness. Billingham et al. (1) noted that newborns injected with allogeneic hematopoietic cells failed to reject subsequent skin grafts. Later, Owen (2) showed that genetically non-identical bovine twins which shared placenta and were exposed to each other's blood, failed to reject allogeneic cells as adults. Based on these data and theoretical considerations, Burnet (3) proposed that the ability to distinguish self from non-self results from the deletion of self reactive clones during the early stages of the immune system development. This concept was supported by experiments that showed an impaired ability to produce specific antibodies following neonatal injection of large amounts of foreign antigens (4,5) or antibodies specific for idiotypic, allotypic or isotypic determinants of Ig receptors (6-8). Furthermore, Etlinger and Chiller (9) showed that aggregated human gamma-globulin, a highly immunogenic T-dependent antigen in euthymic mice, was able to induce immune specific unresponsiveness when injected into newborn mice. More recent data in transgenic animals also showed that thymic exposure to soluble or cellular antigens (10) as well as exposure of immature B cells to membrane-bound antigens (11) caused deletion of the antigen-specific clones. Thus, it was indirectly assumed that neonatal susceptibility to tolerance was a consequence of the predominance of immature lymphocytes during this developmental stage.

Based on these assumptions regarding the neonatal susceptibility to tolerance, several groups succeeded in preventing or delaying the occurrence of spontaneous or experimental autoimmune diseases by injection of self-antigens into newborns. Prevention of experimental allergic encephalitis (EAE) by treatment of neonates with myelin basic protein (MBP) (12) or an MBP immunodominant peptide (13) provided a good example of anergizing the autoreactive T cell clones during the neonatal stage of life. In this particular case, the autoreactive T cell clones are not deleted during the thymic stage of development because of the lack of MBP self antigen. It is known that compacted myelin is produced in substantial amounts only after the first postnatal week and it is essentially sequestred in the nervous system. Thus, neonatal exposure to exogenous MBP prevented the subsequent responsiveness of the MBP-specific T cells. In the same manner, neonatal injection with glutamic acid decarboxylase (GAD) was also capable of delaying the spontaneous occurrence of diabetes in NOD female mice (14). Noteworthy, while the animals injected with MBP or GAD antigens as neonates developed specific T cell unresponsiveness, they were able to produce anti-MBP or anti-GAD antibodies (12,14). The phenomenon of 'split tolerance' was further investigated by Singh et al. (15) who demonstrated that neonatal inoculation of the 106-116 hen egg lysozyme (HEL) peptide in tolerogenic doses induced a mixed Th1/Th2 response that evolved into a strongly biased Th2 response after subsequent antigen exposure. Furthermore, following in vitro treatment with anti-IL-4 or anti-IL-10 monoclonal antibodies (mAbs), the Th2 response was reversed to a Th1 response associated with increased T cell proliferation (15). Mice were able to mount, at all stages, HEL 106-116 specific antibodies, the isotype pattern being largely determined by the Th profile (15). In some cases, neonatal injection of self peptides may rather facilitate or trigger autoimmune diseases. For example, treatment of neonatal NZB/NZW F1 mice with a self-autoantigenic peptide from an anti-DNA mAb that was followed by Th2 response and decreased T cell proliferation, led to the induction of pathogenic IgG anti-dsDNA auto-antibodies associated with signs of glomerulonephritis (15). Thus, it appears that injection of neonates with antigens may have various outcomes regarding the susceptibility for autoimmune diseases, depending on the particular pathogenic mechanism that leads to the disease.

In a distinct experimental system, neonatal B6AF1 female mice injected with a self ZP3 ovarian-specific peptide, developed autoimmune ovarian disease (AOD) associated with a significant T cell proliferative ability and Th1/Th2 mixed profile (16). In contrast, males that were identically treated developed strong Th2 responses associated with low T cell proliferation and resistence to induction of AOD (16). This study suggested that Th1 pathogenic cells induced by injection of neonates with ZP3 peptide, were rescued by continuous exposure to the endogenous self-antigen in the female mice. In another study, it was shown that the continuous exposure to low doses of antigen following injection of neonates with plasmid, led to the priming of specific CTLs rather than unresponsiveness (17). Furthermore, Th1 and CTL induction were noted subsequent to neonatal injection of antigens at low doses or in complete Freund's adjuvant (CFA) (18,19). CTLs specific for H-Y antigens were induced by transfer of dendritic cells from adult males to newborn females, indicating that 'professional' presentation may lead to effective cellular responses in neonates (20). Thus, whereas low doses of antigens presented by professional APC (pAPC) are associated with Th1 and CTL induction upon immunization of neonates, moderate or high-doses of antigens lead to generation of Th2 cells that have a poor proliferative potential. Consequently, neonates exposed to antigens display the tendency to develop an immunodeviated response rather than a true immunologic tolerance.

During the past years, studies conducted in our laboratory regarding the effect of DNA immunization of neonates, showed that sustained exposure to low doses of antigens during the early stages of postnatal development was followed by significant priming of protective humoral and cellular responses rather than induction of tolerance. Thus, this review is aimed at analyzing information, sometimes contradictory, regarding the immune responsiveness of neonatal B and T cells.