The genus Citrus established in 1753 by Carl Linnaeus belongs to the subfamily Aurantioideae of the family Rutaceae. A series of study from morphology to molecular markers on the taxonomy and evolution of the genus Citrus has been carried out both in China and abroad in the past more than 250 years. However, the current views about the subgenus subdivision and the number of species in Citrus are still divergent. The phylogenetic framework of Citrus (the number of wild primitive species, their origins and interspecific relationships) has been unestablished. Swingle (1943) classified the genus Citrus and its five close relatives, Poncirus, Fortunella, Microcitrus, Eremocitrus and Clymenia into the true citrus fruits group based mainly on the number of stamen and structure of pulp-vesicles of the genera. Because of bud mutation, long cultivation history and hybrid compatibility, the taxonomic and evolutionary relationships between Citrus and its close relatives have remain unresolved. Citrus is one of the world's most important fruit crops, a better understanding of the phylogenetic relationship between Citrus and its closely related genera is needed for futrue variety breeding, germplasm conservation and utilization.Amplified fragment length polymorphism (AFLP) is a efficient, simple and stable molecular marker. In the present study, fifty-nine genotypes representing six genera of the true citrus fruit trees group (Aurantioideae, Rutaceae) were analyzed using AFLP technique to study their taxonomic and phylogenetic relationships. Fifteen primer combinations (out of 64 screened) were selected based on the polymorphism and quality of the bands produced by the primer and used in the present study. 312 bands were obtained, of which 305 (97.8%) were polymorphic. The UPGMA tree of all genotypes was constructed based on the AFLP data using PAUP* beta version 4.0b8 software. Our AFLP molecular tree clearly confirms that the true citrus fruit trees group is monophyletic and supports the division of the group into genera mainly based on morphological characters, except that the genus Fortunella was nested within the genus Citrus cluster (as a monophyletic sub-branch). The subdivision of the genus Citrus into subgenus Papeda and subgenus Eucitrus as suggested by W. T. Swingle based on morphological characters and the point of view that C. medica, C. grandis and C. reticulata are the three basic species of the subgenus Eucitrus were also supported by our molecular data. In addition, contrary to the expectation based on morphological data, our molecular data demonstrated that C. ichangensis was more closely related to C. junos of the subgenus Eucitrus than to the C. hystrix of the subgenus Papeda. Furthermore, our study provided the first evidence that C. mangshanensis was basal to all the loose-skinned citrus types. Finally, the taxonomic and phylogenetic relationships among the six genera and the important genotypes of the genus Citrus were discussed in details. |