An adenoviral system that expresses Nef at levels similar to that found during HIV infection. (A) Map of HXB-EP. The marker gene PLAP has been cloned into the envelope open reading frame, and Nef is expressed under its normal regulatory elements. The hatched boxes indicate nonfunctional genes. The arrow indicates the location of the frameshift mutation in Nef that disrupts its expression. (B) Primary T cells transduced with HXB-EP have decreased surface expression of HLA-A2. Primary T cells transduced with HXB-EP that did or did not express Nef were stained for HLA-A2 and PLAP. Cells falling to the right of the vertical line are PLAP positive (infected), and cells falling below the horizontal line are negative for HLA-A2. The percentage of infected cells is indicated, as well as the percentage of Nef-expressing cells that have low (9%) or high (3%) HLA-A2 expression levels. (C) Western blot analysis of Nef and Gag expression from the indicated genome. Lysates were normalized for total protein and percentage of cells positive for intracellular Gag. The immunoblot was probed with an antibody against Nef and then stripped and reprobed with an antibody against Gag. The low level of Gag expression by the molecular clone 94UG114 has been previously reported and is thought to be due to low Rev activity encoded by this genome (3). (D) Determination of the adenoviral MOI (based on 293-cell infectivity) required for efficient CEM T-cell transduction. CEM T cells were transduced with the indicated MOI of adenovirus control (filled curve) or Nef-expressing adenovirus (solid line) and stained for HLA-A2 and CD4. The reduction (n-fold) in cell surface expression based on mean fluorescence intensity is indicated. (E) Adenoviral Nef expression levels are comparable to HXB-EP. CEM T cells were transduced with HXB-EP or the indicated MOI of Nef-expressing adenovirus and harvested 72 h later. Lysates normalized for total protein levels and transduction rates were separated by SDS-polyacrylamide gel electrophoresis and then analyzed by Western blotting using an antibody against Nef (AG11, 1:1,000; James Hoxie, University of Pennsylvania) (14, 15).

Nef mutants that are unable to affect HLA-A2 expression do not inhibit the transport of HLA-A2 to the cell surface. (A) CEM T cells transduced with the indicated adenovirus as described in the legend to Fig. 2 were harvested after 3 days, pulse-labeled for 15 min, and chased for 1 h in the presence of a cell-impermeable biotinylation reagent to label cell surface proteins. The cells were lysed and immunoprecipitated with an antibody against HLA-A2. The immunoprecipitates were then eluted by boiling in 10% SDS, and one-third was analyzed directly (total). The remaining two-thirds was reprecipitated with avidin agarose to selectively precipitate the HLA-A2 on the cell surface (surface) before separation by SDS-polyacrylamide gel electrophoresis. Control cells (ctrl) were the parental CEM T cells that did not express HLA-A2. A reduced yield of HLA-A2 was obtained in this Nef⁻ sample, but the fraction that reached the cell surface (33%) was typical of three experiments. (B) The decrease (n-fold) in cell surface transport was determined using phosphorimager analysis. The percentage of cell surface HLA-A2 in the Nef⁻ sample was divided by that in the Nef-expressing samples. The means ± standard deviations of results from two separate experiments are shown. (C) Nef does not alter HLA-A2 cell surface stability in HIV-infected primary T cells. The cell surface stability of HLA-A2 in HIV-infected primary T cells was measured by flow cytometry using PLAP as a marker of infection as previously described (16).

Analysis of Nef mutants. (A) A schematic diagram of Nef functional domains important for Nef to affect MHC-I trafficking (1, 6, 13, 21, 32). (B) Western blot of NL4-3 Nef expression levels. CEM-A2 T cells were transduced with the indicated adenovirus as previously described (16), with MOIs adjusted so that equal expression was obtained (wild-type Nef, 50:1; V₁₀EΔ17-26, 400:1; M₂₀A, 400:1; E_62-65Q, 600:1; P_72/75A, 1,000:1; P_75/78A, 1,000:1; or D₁₂₃G, 400:1). Lysates from transduced cells were immunoblotted and probed with an antibody directed against Nef (2949, 1:5,000; obtained from National Institutes of Health AIDS Repository, contributed by Ronald Swanstrom) (29) and goat anti-rabbit-horseradish peroxidase (Zymed, 1:50,000). (C) The effect of Nef mutants on MHC-I and CD4 cell surface expression. CEM-A2 T cells were transduced as described for panel B, stained 72 h later for either HLA-A2 (BB7.2 [22], 1:100) or CD4 (Serotech, 1:50), and analyzed by flow cytometry. The filled curve represents cells transduced with the control adenovirus, and the solid line represents cells transduced with an adenovirus expressing the indicated Nef construct. (D) The inhibition of cell surface expression (n-fold) based on mean fluorescence intensity. The means ± standard deviations of results from two experiments are shown.

Nef mutants defective at inhibiting MHC-I cell surface expression do not coprecipitate with HLA-A2. (A, B, and C) CEM T cells were transduced with the indicated adenovirus for 3 days using the MOIs indicated in the legend to Fig. 2 or a 100:1 MOI for HXB L_164/165A Nef. The ability of each mutant to coprecipitate with HLA-A2 was assessed by immunoprecipitation and Western blot analysis (top panel) using an antibody directed against Nef (2949, 1:5,000; obtained from National Institutes of Health AIDS Repository, contributed by Ronald Swanstrom) (29) and goat anti-rabbit-horseradish peroxidase (Zymed, 1:50,000). Input controls were incubated in 150 mM dithiothreitol at 37°C for 30 min and analyzed by Western blotting as described above, except a higher dilution of secondary antibody was used (1:200,000). In panel B, two MOIs of wild-type Nef (low [25:1] and high [50:1]) were used because of low expression of some mutants. Control cells (ctrl) were CEM T cells that do not express HLA-A2.