Human immunodeficiency virus type 1 (HIV-1) exclusively selects tRNA3Lys as the primer for the initiation of reverse transcription, even though both tRNA3Lys and tRNA1,2Lys are found in HIV-1 virions. Alteration of the HIV-1 primer-binding site (PBS) to be complementary to alternate tRNAs results in the use of those tRNAs for replication, indicating that primer complementarity with the PBS is an important determinant of primer selection. In previous studies, we have exploited this fact to develop a system in which yeast (Saccharomyces cerevisiae) tRNAPhe is provided in trans to complement the replication of HIV-1 with a PBS complementary to yeast tRNAPhe. Recent studies have demonstrated that the presence of lysyl-tRNA synthetase in HIV-1 virions might account for the preference for the selection of tRNA3Lys in HIV-1 replication. To establish a complementation system more reflective of HIV-1 primer selection, we have altered the HIV-1 PBS to be complementary to the Escherichia coli tRNA3Lys, which shares near identity with mammalian tRNA3Lys except in the 3'-terminal 18-nucleotide sequence that binds to the PBS. E. coli tRNA3Lys expressed from a plasmid was aminoacylated in mammalian cells. Cotransfection of cells with a plasmid that encodes E. coli tRNA3Lys and a plasmid encoding an HIV-1 provirus with a PBS complementary to E. coli tRNA3Lys resulted in the production of infectious virus. A comparison of the two complementation systems revealed that higher levels of intracellular E. coli tRNA3Lys than of yeast tRNAPhe were needed to achieve equal levels of infectious virus, indicating that there was no preferential selection of E. coli tRNA3Lys. To examine the specificity of tRNALys selection, E. coli tRNA3Lys was modified to tRNA1,2Lys. This tRNA was also aminoacylated when expressed in mammalian cells and complemented the infectivity of HIV-1 at levels similar to those seen for E. coli tRNA3Lys. Additional mutations in the anticodon of E. coli tRNA3Lys were constructed; these mutations did not significantly correlate with the capacity of the tRNA primer to complement infectivity of HIV-1, even though they had a drastic effect on the aminoacylation of the tRNAs. The results of these studies demonstrate that E. coli tRNA3Lys provided in trans can complement HIV-1 genomes with the PBS altered to E. coli tRNA3Lys. However, the capacity of tRNA3Lys to interact with lysyl-tRNA synthetase does not entirely explain the enhanced preference for selection of tRNA3Lys for the replication of HIV-1.