Functional analysis of the modified frameshift sites in a luciferase reporter assay. (A) The nucleotide sequence representing the original NL4-3-derived frameshift site (fs-NL), the inactivated frameshift site (fs-inact), and the artificial frameshift site (fs-AL) in provirus AL were cloned 5′ of a firefly luciferase reporter gene to allow for frameshift-dependent expression of luciferase from the −1 frame. The luciferase reading frame is depicted in gray, and nucleotide positions altered with respect to the original frameshift site are underlined. An additional construct with a modified slippery sequence driving frameshift-independent reporter synthesis from the −1 frame was used as positive control (fs-luc). (B) 293T cells were transiently transfected with the indicated reporter constructs or an empty vector (mock), and cell-associated luciferase activity was quantified after 48 h. Luciferase activity (relative light units [RLU]) determined for fs-luc was set to 100%, and all other values were related accordingly. Results from one representative blot comprising mean values and standard deviations from three independent transfection experiments are shown.

Influence of p6*-specific mutations in AL on frameshift efficiency. (A) On the basis of provirus AL, further modifications were introduced into the p6* sequence, resulting in virus mutants ALn (substitution of the amino-terminal residues in boldface), ALΔ (deletion of the central p6* sequence indicated by dots), and ALΔ-myc (carrying a 10-amino-acid myc tag instead of the central p6* region [indicated by a box] and flanking substitutions in p6* [boldface]). Those amino acids overlapping the RNA region involved in frameshifting are highlighted by a gray box, and p6* residues contributing to carboxyl-terminal p6*-PR cleavage are underlined. Amino-terminal (N), internal (I), and carboxyl-terminal (C) cleavage sites in p6* are indicated by arrowheads. (B) Modifications in the stem-loop regions of ALn and ALΔ compared to the AL-specific structure are indicated by black arrows. Stem-loop structures were predicted as described by Zuker and coworkers (37, 40), with the calculated free energy values (ΔG) shown below. (C) Functional analysis of frameshift efficiency in a luciferase reporter assay. The frameshift regions of the indicated AL derivatives were cloned 5′of a firefly luciferase reporter gene to allow for frameshift-dependent expression of luciferase from the −1 frame as illustrated in Fig. 2A. The corresponding reporter constructs fs-AL, fs-ALn, and fs-ALΔ; the control vector fs-luc; or an empty vector (mock) was used to transfect 293T cells, and cell-associated luciferase activity was quantified after 48 h. Luciferase activity (relative light units [RLU]) determined for fs-luc was set to 100%, and all other values were related accordingly. Results for one representative blot comprising mean values and standard deviations from three independent transfection experiments are shown.

Packaging of heterologous sequences expressed from the p6* reading frame. (A) A schematic representation of the AL Gag and Gag-Pol precursors is shown, with the p6* central region highlighted by a shaded box. A myc tag and a GFP reporter gene were inserted into the p6* reading frame, yielding ALΔ-myc and ALΔ-GFP provirus constructs. (B) 293T cells were transfected with AL-, ALΔ-myc, or an empty vector (mock) in the presence of the PR inhibitor Saquinavir, and myc-specific products (a to d) were detected in particles released into culture supernatants at 48 h posttransfection by Western blotting. (C) 293T cells were transfected the indicated constructs, and particles were analyzed with a GFP-specific antibody. GFP-specific products are indicated by arrows at the right.

Gag and Gag-Pol polyproteins encoded by NL4-3 (NL) and AL proviruses. (A) Schematic representation of the Gag and Gag-Pol polyprotein precursors. The p6* domain within Gag-Pol is highlighted in black, and the p1-p6^gag region inserted in the Gag-Pol precursor of AL is shaded gray; the five residues appended to p6^gag in Gag are indicated by a black vertical line. MA, matrix protein; NC, nucleocapsid protein; IN, integrase. (B) The Gag (gray) and Pol (black) products translated in the presence of the original frameshift site in NL are indicated on top, and the corresponding passage synthesized in AL after destruction of the slippery site is shown below. The active (NL) and inactive (AL) slippery sites are underlined, and the 3′ nucleotides contributing to the stem-loop structure are in parentheses. (C) An artificial frameshift site was introduced at the 3′ end of the gag gene in AL, resulting in the synthesis of a Gag-Pol polyprotein with p6* directly fused to p6^gag. The gag reading frame in AL is terminated by an artificial stop codon (boxed) and is extended by the residues ANFLG. The E₄→V₄ substitution in p6* is marked by a circle, and numbers at the left indicate amino acid positions within Gag and Gag-Pol with respect to the Gag start codon.

Influence of frameshift displacement on virus maturation, infectivity, and replication. (A) 293T cells were transiently transfected with the indicated viral plasmids. For analysis of virus composition, particles released into cell supernatants were harvested at 72 h posttransfection and sedimented through a 20% sucrose cushion. Viral proteins were separated by SDS-PAGE and analyzed by Western blotting using a monoclonal CA-specific antibody. Positions of molecular mass markers are shown at the left, and arrows at the right indicate Gag-specific protein bands. (B) To evaluate viral infectivity, virus-containing supernatants from 293T cells were harvested at 72 h posttransfection, normalized for CA content by ELISA, and used to infect CD4-positive TZM-bl indicator cells. At 48 hours postinfection, blue cells were counted and the values obtained were related to NL infections (100%). Error bars indicate the standard deviations from triplicate infections. (C and D) To monitor viral replication kinetics, PBMCs (C) and MT4 cells (D) were infected in duplicate with the virus-containing supernatants, and spreading infections were monitored by quantification of CA amounts in culture supernatants by ELISA. Results from one representative experiment are shown, with error bars indicating the standard deviations for duplicate infections.

Characterization of particle composition, infectivity, and replication of the AL-derived p6* mutants. (A) 293T cells were transiently transfected with the indicated proviral plasmids, and particles released into cell supernatants were harvested at 72 h posttransfection and sedimented through a 20% sucrose cushion. Particle-associated proteins were separated by SDS-PAGE and analyzed by Western blotting using CA-, Nef-, and RT-specific antisera. Positions of molecular mass markers are shown at the left, and arrows at the right indicate specific protein bands. (B) To evaluate viral infectivity, virus-containing supernatants from 293T cells were harvested at 72 h posttransfection, normalized for CA content by ELISA, and used to infect CD4-positive TZM-bl indicator cells. At 48 hours postinfection, blue cells were counted and the values obtained were related to AL infections (100%). Error bars indicate the standard deviations from triplicate infections. To analyze viral replication kinetics, PBMCs (C), CEM cells (D), and MT-4 cells (E) were infected in duplicate with the virus-containing supernatants, and spreading infections were monitored by quantification of CA amounts in culture supernatants by ELISA. Results from one representative experiment are shown, with error bars indicating the standard deviations from duplicate infections.

Influence of p6* insertions on viral maturation and infectivity. 293T cells were transfected with the indicated provirus constructs or an empty vector (mock), and virus-containing supernatants (VP) (A, C, and D) or cells (B and E) were harvested at 48 h posttransfection and analyzed by Western blotting using PR-, CA-, or RT-specific antisera. (B) Cells were transfected in the presence of the PR inhibitor Saquinavir (PI) and were harvested at 24 h posttransfection. Positions of molecular mass markers are shown at the left, and arrows at the right indicate specific protein bands. (F) Virus particles released into cell supernatants were harvested at 72 h posttransfection and sedimented through a 20% sucrose cushion. To determine precursor-associated CA amounts, virus preparations were quantified for CA content by ELISA prior to (black bars) or after (gray bars) processing with an excess of HIV-1 rPR. (G) To measure viral infectivity, virus-containing supernatants from 293T cells were harvested at 72 h posttransfection, normalized for CA content by ELISA, and used to infect CD4-positive TZM-bl indicator cells. At 48 hours postinfection, blue cells were counted, and the values obtained were related to AL infections (100%). Error bars indicate the standard deviations from triplicate infections.