BACKGROUND: The Gal alpha(1,3)Gal epitope is of interest as, in pig-to-primate xenotransplantation, it is the major target of naturally occurring human IgM and IgG antibodies, leading to hyperacute rejection. Human and Old World monkeys make anti-Gal alpha(1,3)Gal antibodies as they lack a functional gene and do not express Gal alpha(1,3)Gal. Interestingly, the cultured fibroblasts of some other species, such as chickens, have been reported also not to express Gal alpha(1,3)Gal–if this is true for other tissues, and chickens do not express Gal alpha(1,3)Gal antigen, then they would have anti-Gal antibodies–which could have diagnostic and therapeutic value, particularly as chicken antibodies do not fix mammalian complement. METHODS: Standard serological methods were used to characterize the antibodies. Several baboons received pig kidney xenografts that had been perfused with hyperimmune chicken anti-Gal antibodies. RESULTS AND CONCLUSIONS: We now demonstrate that chickens do not express Gal alpha(1,3)Gal on their red cells, leukocytes, or tissues, and that their serum contains large amounts of anti-Gal alpha(1,3)Gal antibodies. In addition, chickens could be immunized to produce high-titer, high-avidity antibodies (9.5x10(9) M(-1))–an avidity considerably greater than that of the Gal alpha(1,3)Gal binding lectin IB4 (2.9x10(8) M(-1)) or Gal antibodies in human serum (2.2x10(5) M(-1)). Chicken antibodies, obtained from both normal and immunized chickens, could block the in vitro cytolysis of pig endothelial cells or lymphocytes by human or baboon antibodies. However, such antibodies tested in vivo in pig-to-baboon xenotransplantation failed to block hyperacute rejection and, indeed, may have accelerated this.