ARE PIGS TRANSGENIC FOR HUMAN COMPLEMENT REGULATORY PROTEINS NECESSARY FOR XENOTRANSPLANTATION?

Role and Regulation of Pig CD59 and Membrane Cofactor Protein/CD46 Expressed on Pig Aortic Endothelial Cells. Transplantation 2000; 70: 667.van den Berg CW, Rix C, Hanna SM, Perez de la Lastra JM, and Morgan BP. There are remarkably few functional comparisons of complement regulatory proteins in the literature, which is perhaps surprising in the light of the common coexpression of membrane cofactor protein (MCP)/CD46, decay-accelerating factor (DAF)/CD55, and CD59 on many cells in humans (and pigs). The report by van den Berg et al. on pages 667–673 of this issue presents a fresh view on the roles of complement regulatory proteins in protecting cells against damage or lysis mediated by complement. They studied two porcine proteins, membrane cofactor protein (MCP or CD46 homolog) controlling C3 and C4 activation, and porcine CD59 controlling assembly of the terminal complement complex. Using human serum as the source of complement, they focus attention on the relative efficiencies of these proteins and their potential roles in resisting the hyperacute rejection of pig xenografts by humans. Their data led them to favor porcine CD59 as the more efficient protein preventing cell lysis and to encourage analyses of the promotor regions of the respective genes to determine whether hyperexpression of the porcine proteins could be induced in xenografts. This article and previous work from this group raise a number of major questions about strategies for xenotransplantation and the assays commonly used to assess the efficacy of complement regulators. Their model target cells are primary cultures of porcine aortic endothelial cells, which, as they point out, are the immediate target cells of vascularized xenografts. Endothelial cells are recognized as being highly sensitive to the effects of ischemia during transplantation and, once activated, are responsible for catalyzing some of the mechanisms that result in their destruction. By using a primary cell culture, van den Berg et al. have also attempted to avoid changes that occur in culture-adapted cell lines. Some of the most interesting and unexpected results they obtained illustrate how rapidly cells can alter phenotype. Within several passages, the cells bound more human IgM and IgG; unfortunately, no mechanisms for this change are identified. Neither were any cells grown in autologous pig serum to test whether the fetal bovine serum and normal calf serum in the culture medium were contributing to augmentation of the antigenicity of the endothelial cells. A twofold increase in MCP but an approximately threefold reduction in CD59 within five passages was also observed. If such changes occur in xenografts in vivo, many current experimental strategies to achieve successful xenotransplantation will require review. The data provided on the function of porcine MCP require careful interpretation. Although increased expression of MCP after phorbol 12-myristate 13-acetate treatment of cells correlated with increased resistance to complement-mediated lysis, they show that increased MCP expression with cell passage number occurred with increased susceptibility to lysis, suggesting that the concurrent reduction in CD59 played a greater role in the outcome. Their previous work, using cofactor assays, indicated that porcine MCP was a most effective regulator of human complement (1). Their present implication that the MCP/CD46 protein is of lesser interest for xenotransplantation may be underestimating its effects. There is a view being expressed that CD46 only regulates the alternative pathway of complement activation, though the data in the literature are not so clear cut. The classical pathway is driven much faster, in part by the C3 amplification feedback loop, which increases the quantities of C3b generated, of C3b deposited, and of the classical pathway C5-convertase. C3b deposition indicates complement activation but not necessarily inevitable cell damage or lysis. Transfected and transgenic CD46 certainly protects cells against both alternative and classical pathway complement activation and cell lysis (2, 3). Current strategies to reduce the antigenicity of porcine tissues (see 4) might create a scenario where regulation of alternative pathway activation would be of critical importance for long-term xenograft function, where cell lysis is not the most important indicator of complement function. For example, control of C3 and C4 activation limits production of the anaphylatoxins C3a, C4a, and C5a, which are major chemokines inducing inflammatory responses in vivo. Thus, porcine or human MCP/CD46 may in some circumstances be the choice regulator. The data presented in the article add weight to previous work by this group counteracting the dogma of homologous restriction and indicate that pig tissues exposed to human natural anti-Galα(1,3)Gal antibodies and the components of the human complement cascade are not extraordinarily susceptible to complement-mediated lysis simply because the porcine regulators fail to regulate human complement. Thus, it is reasonable to ask whether up-regulated or hyperexpressed porcine complement regulators, such as DAF, CD59, and MCP, would be equally efficient as the transgenic human regulators that are currently being studied (4), and raise the important question of whether human promoters should be used in transgenes or might porcine promoters be equally or even more responsive? Thefinal issue that is worthy of reflection is the inhibition of complement regulator function by blocking antibodies. van den Berg et al. noted uniform CD59 inhibition by the antibodies they tested, whereas MCP inhibition was not readily detected except in the in vitro cofactor assays. (Particular monoclonal antibodies do inhibit CD46 protection against lysis, as long as Fab fragments are used (2)). Assuming that successful xenotransplantation of pig organs in humans will eventually be achieved but without the induction of complete immune tolerance, anti-pig antibodies will most likely continue to be generated despite immunosuppressive therapy. It is conceivable that some antibodies will be generated against endogenous porcine complement regulators and may inhibit their function. Indeed, there is precedence in an experimental model of active Heymann nephritis in rats where membranous nephropathy could be induced by natural anti-Crry antibodies (5), which react with a homolog of CD46 and CD55. Thus, increasing expression of proteins that in themselves will not be targets of antigraft immunity may be advantageous. In summary, the authors have drawn attention to the fact that porcine regulators of complement activity function efficiently against human complement and provide significant protection against cell lysis. Nevertheless, if xenotransplantation is to be fully successful, pursuit of transgenic human regulators, in combination with other strategies, may remain the path to follow.