Background: APOBEC3 (A3) proteins deaminate DNA cytosines and block the replication of retroviruses and retrotransposons. Each A3 gene encodes a protein with one or two conserved zinc-coordinating motifs (Z1, Z2 or Z3). The presence of one A3 gene in mice (Z2-Z3) and seven in humans, A3A-H (Z1a, Z2a-Z1b, Z2b, Z2c-Z2d, Z2e-Z2f, Z2g-Z1c, Z3), suggests extraordinary evolutionary flexibility. To gain insights into the mechanism and timing of A3 gene expansion and into the functional modularity of these genes, we analyzed the genomic sequences, expressed cDNAs and activities of the full A3 repertoire of three artiodactyl lineages: sheep, cattle and pigs. Results: Sheep and cattle have three A3 genes, A3Z1, A3Z2 and A3Z3, whereas pigs only have two, A3Z2 and A3Z3. A comparison between domestic and wild pigs indicated that A3Z1 was deleted in the pig lineage. In all three species, read-through transcription and alternative splicing also produced a catalytically active double domain A3Z2-Z3 protein that had a distinct cytoplasmic localization. Thus, the three A3 genes of sheep and cattle encode four conserved and active proteins. These data, together with phylogenetic analyses, indicated that a similar, functionally modular A3 repertoire existed in the common ancestor of artiodactyls and primates (i.e., the ancestor of placental mammals). This mammalian ancestor therefore possessed the minimal A3 gene set, Z1-Z2-Z3, required to evolve through a remarkable series of eight recombination events into the present day eleven Z domain human repertoire. Conclusion: The dynamic recombination-filled history of the mammalian A3 genes is consistent with the modular nature of the locus and a model in which most of these events (especially the expansions) were selected by ancient pathogenic retrovirus infections.
Bibliographical noteFunding Information:
We thank L. Beach, H. Malik, M. Murtaugh and M. Stenglein for valuable feedback. We thank K. Tennill and R. Godtel for assistance with BAC DNA sequencing, D. Shiroma and S. Fahrenkrug for help identifying pig BAC clones, M. Stenglein for several expression plasmids, M. Titus for use of her microscope, L. Hartman for pig and sheep samples, C. Knutson for cow samples, J. Zimmerman and R. Molina for peccary blood, P. Krauseman for peccary brain tissue and M. Ruen and O. Holland for opossum samples. R. LaRue is a member of the University of Minnesota CMB Graduate Program. S. Jónsson was the 2004–2005 Val Bjornson Icelandic Exchange Scholarship recipient. M. Lajoie was supported by a Canadian Institutes of Health Research studentship. D. Bertrand and N. El-Mabrouk were supported by grants from the Fonds Québécois de la Recherche sur la Nature et les Technologies and the Natural Sciences and Engineering Research Council of Canada. R. Harris was supported in part by a Searle Scholarship and a University of Minnesota McKnight Land Grant Assistant Professorship. This work was also supported by NIH grant AI064046. The University of Minnesota Advanced Genetic Analysis Facility assisted with DNA sequencing.