CTL resistance to Nef is associated with early killing of HIV-1–infected cells. The timing of CTL killing of 1CC4.14 cells after acute infection with VSV-G Env-pseudotyped NL4-3-ΔEnv was assessed by serial ⁵¹Cr measurements over 24 hours. (A) Specific lysis plotted over time by epitope. Results with CTL clones are plotted with symbols and the average across clones is plotted with a broad gray line. (B) Estimates of time to reach 10% specific lysis (K₁₀) are indicated. The estimates were obtained by fitting average lysis curves for each epitope with logarithmic regression. Note that CTLs targeting the B*15-restricted Nef TY11 and the B*57-restricted Pol KW10 epitopes were unable to recognize virus-infected cells within the first 24 hours of infection, and therefore each epitope was assigned a conservative K₁₀ value of 24 hours. The inability of these CTL clones to recognize virus-infected targets was not because of inactivity of the clones, because they efficiently killed cells infected with HIV-1 containing the Nef M20A mutation (data not shown). (C) K₁₀ values of Gag-specific versus non-Gag–specific CTLs are compared. Each point represents the K₁₀ value for an epitope; the horizontal bar represents the mean across epitopes. Statistical significance was evaluated with a 2-tailed Student t test. (D) K₁₀ values are plotted against Nef-effect ratios for all epitopes. Statistical significance was tested with a Pearson test. Note that these results were obtained using HIV-1 containing wild-type Nef; nearly identical values were obtained using virus with Nef-M20A, although the efficiencies of infected cell killing were slightly higher (data not shown). Asterisk indicates results after removing an outlier, the B*57-restricted Rev RY10 epitope.