K. Verhoef et B. Berkhout, A second-site mutation that restores replication of a Tat-defective human immunodeficiency virus, J VIROLOGY, 73(4), 1999, pp. 2781-2789
We previously constructed a large set of mutants of the human immunodeficie
ncy virus type 1 (HIV-1) regulatory protein Tat with conservative amino aci
d substitutions in the activation domain. These Tat variants were analyzed
in the context of the infectious virus, and several mutants were found to b
e defective for replication. In an attempt to obtain second-site suppressor
mutations that could provide information on the Tat protein structure, som
e of the replication-impaired viruses were used as a parent for tbe isolati
on of revertant viruses with improved replication capacity. Sequence analys
is of revertant viruses frequently revealed changes within the tat gene, mo
st often first-site reversions either to the wild-type amino acid or to rel
ated amino acids that restore, at least partially, the Tat function and vir
us replication. Of 30 revertant cultures, we identified only one second-sit
e suppressor mutation. The inactive Y26A mutant yielded the second-site sup
pressor mutation Y47N that partially restored trans-activation activity and
virus replication. Surprisingly, when the suppressor mutation was introduc
ed in the wild-type Tat background, it also improved the transactivation fu
nction of this protein about twofold. We conclude that the gain of function
measured for the Y47N change is not specific for the Y26A mutant, arguing
against a direct interaction of Tat amino acids 26 and 47 in the three-dime
nsional fold of this protein. Other revertant viruses did not contain any a
dditional Tat changes, and some viruses revealed putative second-site Tat m
utations that did not significantly improve Tat function and virus replicat
ion. We reason that these mutations were introduced by chance through found
er effects or by linkage to suppressor mutations elsewhere in the virus gen
ome. In conclusion, the forced evolution of mutant HIV-1 genomes, which is
an efficient approach for the analysis of RNA regulatory motifs, seems less
suited for the analysis of the structure of this small transcription facto
r, although protein variants with interesting properties can be generated.