Few major histocompatibility complex (MHC)-based mate choice studies include all MHC genes at the inter-individual, sperm-egg, and mother-fetus recognition levels. We tested three hypotheses of female mate choice in a 17-year study of the giant panda (Ailuropoda melanoleuca) while using ten functional MHC loci (four MHC class I loci: Aime-C, Aime-F, Aime-I, and Aime-L; six MHC class II loci: Aime-DRA, Aime-DRB3, Aime-DQA1, Aime-DQA2, Aime-DQB1, and Aime-DQB2); five super haplotypes (SuHa, SuHaI, SuHaII, DQ, and DR); and, seven microsatellites. We found female choice for heterozygosity at Aime-C, Aime-I, and DQ and for disassortative mate choice at Aime-C, DQ, and DR at the inter-individual recognition level. High mating success occurred in MHC-dissimilar mating pairs. No significant results were found based on any microsatellite parameters, suggesting that MHCs were the mate choice target and there were no signs of inbreeding avoidance. Our results indicate Aime-DQA1- and Aime-DQA2-associated disassortative selection at the sperm-egg recognition level and a possible Aime-C- and Aime-I-associated assortative maternal immune tolerance mechanism. The MHC genes were of differential importance at the different recognition levels, so all of the functional MHC genes should be included when studying MHC-dependent reproductive mechanisms.
Keywords: MHC-I- and MHC-II-dependent inter-individual recognition; MHC-I-based mother-fetus recognition; MHC-II-associated sperm-egg recognition; giant panda; long-fragment super haplotype.