UNDERSTANDING REJECTION OF XENOGRAFTS

Specific Depletion of Preformed IgM Natural Antibodies by Administration of Anti-μ Monoclonal Antibody Suppresses Hyperacute Rejection of Pig to Baboon Renal Xenografts. Transplantation 2000; 70: 935.Dehoux JP, Hori S, Talpe S, Bazin H, Latinne D, Soares MP, Gianello P. More than a decade of investigations devoted to understanding rejection of xenografts has led us to a detailed description of the events that may participate in the rejection process. The report by Dehoux et al. in this issue contributes to answering the question of to what extent the many biological mechanisms potentially involved actually play a role in rejection in vivo. They demonstrate that XNA of the IgM isotype are the true mediators of hyperacute rejection (HAR). HAR of pig vascularized xenografts in primates is now understood as the deposition of xenoreactive natural antibodies (XNA) that in turn activate complement. Complement activation alters the endothelial monolayer and perturbs several anti-coagulant and anti-oxidant molecules normally expressed on EC. In addition, the possibility exists that complement (through iC3b deposition) and xenoreactive IgG opsonize EC that become targets for CD11b/CD18+ polynuclear cells and for CD16+ cells. Activation of C1q by XNA may also link complement with clotting through its platelet activating activity (1). However, evidence has accumulated in vitro that XNA of the IgM isotype, or cross-linked IgG, can signal and activate, at least partially, endothelial cells independent of complement activation, and lead to the expression of genes that are implicated in rejection (2). Many pig-to-primate xenotransplantations have used transgenic pig lines expressing human regulators of complement activity such as CD55, acting from C3 downward in the complement cascade. Therefore, these experiments have addressed the role of down stream complement components in xenograft rejection, and actually not that of XNA and early complement components. Such experiments show that controlling C3 downward in the complement cascade prevents HAR but allows for an acute vascular rejection to occur, which is correlated with increased xenoantibody titers and with endothelial cell activation. This leads to the interpretation that behind HAR, XNA, and induced antibodies, mainly against Galα1,3Gal epitopes, participate in xenograft rejection independent of complement activation. The reasons why the anti-Galα1,3Gal XNA response is so strong in xenotransplantation has recently been reviewed by Tanemura et al. (3). These include: 1) anti-Gal membrane IgM and IgD on pre-existing anti-Gal B cells capture galactosylated xenoglycoproteins and give B cells the first activation signal; 2) as B cells are efficient antigen presenting cells in vivo (4) they process peptides derived from these xenoglycoproteins and associate them with class II antigens, and subsequently receive help from the many T cells able to recognize these xenopeptides. This help completes their activation (signal 2) process by inducing their proliferation, isotype switch, and somatic mutations. We have recently shown in the hamster to rat model that the Vβ repertoire of T cells implicated in xenograft rejection is polyclonal thus enabling recognition of many xenopeptides, whereas the Vβ repertoire of T cells implicated in allograft rejection is restricted to the use of dominant reactivities (5). Eliminating Galα1,3Gal epitopes from pig tissues represents therefore a major issue. The recent success in cloning pigs by nuclear transfer brings the possibility of inactivation of the α1,3-galactosyltransferase gene that controls the synthesis of this epitope. Organs and tissues from Gal−/− pigs would therefore be hidden from for the human immune system, as far as anti-Gal antibodies are concerned. Another approach to eliminate an early attack by anti-graft immunoglobulins in addition to protecting from complement attack and modifying the antigenicity of the graft, is the elimination of recipient XNA. Very specific immunoadsorption using αGal affinity columns (6) or anti-Ig columns has been demonstrated to be useful in removing anti-Gal antibodies in baboons. Although such immunoadsorbtion reduces XNA by more than 80% and prevent HAR of grafted pig kidneys and hearts, it does not prevent a rapid rebound of Ig after grafting and this cannot be abrogated by the use of potent immunosuppressors, including cyclophosphamide (7). Targeting IgM molecules with anti-mu antibodies as presented by Dehoux et al. in this issue of Transplantation, has so far been the most efficient depletion performed in vivo. It acts first through eliminating serum IgM by complexation. Second, it has been known from the early 1970s that anti-IgM antibodies block terminal B cell differentiation and prevent IgM secretion. In rodents treated from birth with anti-μ or anti-μ and anti-δ antibodies, B cell differentiation is suppressed, leading to both B cell depletion in secondary lymphoid organs and depletion of circulating antibodies of all isotypes (8). In adults, however, it results in the modulation of B cell differentiation and leads to the down-regulation of IgM production and in the up-regulation of IgG2C in rats, or IgG1 and IgG3 in mice. The question of whether anti-μ administration acts also at the B cell level in primates has not been resolved but in vitro experiments presented here, showing that anti-μ impairs IgM secretion by human and baboon B cells, suggests that it does. In vivo, sustained anti-μ administration in baboons resulted in a complete but transient depletion of circulating IgM, the rebound of circulating IgM being correlated with the generation of elicited antibodies directed against the infused monoclonal antibody (of rat origin) and its elimination from the circulation. Interestingly, these anti-antibodies included immunoglobulins of the IgM and IgG isotypes reacting against Gal epitopes, with Galα1,3-Gal carbohydrates being present on antibodies from mice and rats. One might have expected that an antibody response against an infused heterologous antibody is lower if the infused antibody itself induces an immunosuppression, such as expected with an anti-μ antibody. The reason for the vigorous IgG response directed against the LO-BM2 may lie in the phenomenon described by Denis et al. (4), where anti-μ are processed in vivo by B cells that in turn generate an IgG anti-donor response (IgG2C response in rats and an IgG1/IgG3 in mice). In the first pig-to-baboon xenotransplantations, XNA of the IgM isotype were pointed out as the major Ig subclass playing a role in rejection. Both IgM and IgG XNA were deposited on the grafted tissue but, whereas IgM was deposited on the vasculature, IgG infiltrated the tissue and was largely codeposited with albumin, indicating a passive adsorption on damaged tissues. However, anti-gal antibodies of the IgG isotype were abundant and bound to porcine endothelial cells in vitro, such that the biological role of IgG in xenograft rejection remained controversial. Despite its short-term effect, anti-IgM administration in vivo brings about our comprehension of xenograft rejection: it shows that specific inhibition of IgM molecules is sufficient to circumvent HAR, demonstrating clearly that preformed XNA of the IgG isotype cannot induce HAR, and that accessory mechanisms that may theoretically participate in HAR, such as opsonization, antibody dependent cellular cytotoxicity, or intravascular coagulation due to molecular incompatibilities, are actually not instrumental in HAR. What remains to be clarified is whether a temporary elimination of xenoantibodies, or anti-Galα1,3Gal antibodies, will allow porcine xenografts to be accommodated. This concept was developed in the hamster-to-rat xenotransplantation model where hamster hearts can survive under cyclosporin A treatment after return of xenoantibodies and complement, and is associated with the up-regulation of antiapoptotic and antioxydant genes. What also remains to be clarified is whether the rebounding anti-pig antibodies, including IgG and IgM, observed in transplanted baboons, are identical to the original anti-pig XNA, or, alternatively, if pig xenoantigens elicit a class-switched response with higher intensity and affinity that generates a delayed form of xenograft rejection involving the active participation of mechanisms also depending on IgG. Targeting B cells with bioreagents, including anti-μ plus anti-δ treatment, a strategy recently shown by the same group to suppress B cell responses in mice and rats (9), might well be a valuable approach to complement the immunomodulatory drugs currently being evaluated in xenotransplantation.

Récupéré en direct depuis OpenAlex et désinversé. Les résumés ne sont pas conservés dans cette base de données : les index inversés représentent 8,6 Go des 9,3 Go de texte de la base, et le serveur dispose de 13 Go libres.