A system has been designed to study the in vivo forward rate of mutati
on of human immunodeficiency virus type 1 (HIV-1) during one round of
replication. A HIV-1 shuttle vector was used that contained the lacZ a
lpha peptide gene as a reporter for mutations. The forward mutation ra
te of HIV-1 was found to be 3 x 10(-5) mutations per target base pair
per cycle, or about 20-fold lower than the error rates reported for pu
rified HIV-1 reverse transcriptase with sense-strand RNA and DNA templ
ates of the lacZ alpha peptide gene in a cell-free system. To test the
hypothesis that the vpr gene product might, at least in part, account
for the lower mutation rate observed in vivo, a HIV-1 vector was repl
icated to determine if the mutation rate was higher in the absence of
the wild-type vpr gene product. A vpr(-) shuttle vector had an overall
mutation rate as much as 4-fold higher than that of the parental vect
or. A shuttle vector with an amino acid substitution in vpr that preve
nts efficient incorporation of Vpr into virus particles was found to h
ave a mutation frequency similar to that of the vpr(-) vector, and was
interpreted to indicate a requirement for Vpr incorporation into the
virus particle in order to observe the influence of vpr on the mutatio
n rate. Replication of a vpr(-) shuttle vector in the presence of a wi
ld-type vpr expression plasmid led to a mutation frequency similar to
that of the parental vector, suggesting that the vpr mutation could be
complemented in trans. Immunoprecipitation analysis indicated that Vp
r virion incorporation coincided with the influence of vpr on the muta
tion rate. (C) 1996 Academic Press, Inc.