Transduction of human immunodeficiency virus type 1 vectors lacking encapsidation and dimerization signals

Abstract

The encapsidation signal (Psi) and the nested dimerization initiation site are important for efficient packaging of human immunodeficiency virus type 1 (HIV-1) genomic RNA dimers. Consequently, these signals are included in all HIV-1 vectors. Here, we provide evidence demonstrating that these elements in such vectors are not absolutely required for vector transduction. In single-cycle infection assays, vectors with Psi deleted (DeltaPsi) were transduced with only a two- to fivefold reduction compared to the wild type. The transduction of DeltaPsi showed typical products of reverse transcription and vector integration; however, in vitro and in vivo dimerization assays demonstrated the lack of normal dimerization of the DeltaPsi vector. The reduction in transduction reflected a similar reduction in packaging. Nevertheless, a relatively high specificity of packaging was retained, as the DeltaPsi vector was encapsidated at a level 4 orders of magnitude higher than that for overexpressed, nonretroviral cellular mRNA and 15 orders of magnitude higher than that for a murine leukemia virus (MLV)-based vector, all containing the same reporter gene, suggesting a Psi-independent mechanism of packaging. The fact that HIV-1 and MLV vectors were encapsidated with a much higher level of efficiency than the cellular RNA suggests that the genomic RNAs of different retroviruses share common features and/or pathways that target them to encapsidation. Overall, these results formally demonstrate that packaging and dimerization signals are not required for the early stages of infection and can be deleted without risking a total loss of vector transduction. Deletion of these signals should enhance the safety of these vectors.

FIG. 1.

Schematic diagrams of HIV-1 vectors used in this study. (A) Schematic diagram of the pHR′ vector and expanded view of its core encapsidation signal. Shown are the 5′ and 3′LTRs (gray boxes with R sequence highlighted in black), the truncated gag open reading frame (boxed with a wavy line), the Rev-responsive elements (RRE), and the CMV internal promoter driving the reporter gene (in this study, gfp or lacZ). The drawing is not to scale. The expanded view of the encapsidation signal shows the four stem-loops (SL1 to SL4). The palindromic sequence of the DIS, which also forms a BssHII restriction site, is boxed. The SD site is represented by an arrowhead. Shaded letters represent the start codon of the gag open reading frame. The sequence between the solid arrows was deleted to make the ΔΨ mutation. (B) Representation of the sequence replacing SL1 to SL4. In the ΔΨ mutant, a PmeI restriction site was introduced together with the SD site.

FIG. 2.

Expression and transduction of an HIV-1-based vector with an encapsidation signal deletion. (A) 293T cells were transfected with helper plasmids pCMVΔR8.2 and pMD.G together with pHR′CMV-GFP (wt) or pHR′CMV-GFPΔΨ (ΔΨ) vector DNA. Two days posttransfection, culture media were collected and equal volumes (diluted 1:3 in the presence of 8 μg/ml polybrene) were used to infect naïve 293T cells. Infected cells (solid lines) were analyzed for GFP expression 2 days postinfection by FACS and were compared to mock-infected cells (dashed lines). (B) Graph summarizing total GFP expression (see Materials and Methods for calculation) in transfected (gray bars) or infected (black bars) 293T cells with either wt or ΔΨ vectors. The data are represented as the means ± the standard error of the means (n = 6) of the total GFP fluorescence intensities and are shown on a logarithmic scale with numerical annotation of the means.

FIG. 3.

Expression of the ΔΨ vector in infected cells represents genuine transduction. (A) FACS analysis of GFP expression following transfection (top panel) and infection (bottom panel) of 293T cells with the ΔΨ vector (solid black line). This vector was cotransfected with helper plasmids (pCMVΔR8.2 or pMD.G) as indicated. Cells were analyzed 2 days posttransfection, and their media (diluted 1:3 and supplemented with 8 μg/ml polybrene) were used to infect naïve 293T cells. Control cells (thin dashed line) were either mock-transfected or infected 293T cells. (B) FACS analysis of GFP expression in 293T cells infected with either wt or ΔΨ virions (solid lines) in the absence or presence of nevirapine (Nev; 12 μM). Cells were screened 5 days following infection. Dashed lines indicate control cells, as above. (C) Circle junction analysis of both the wt and ΔΨ vectors. Low-molecular-weight DNA was extracted, using the Hirt method, from 293T cells infected with media from cells expressing vector and pCMVΔR8.2 (1) or with vector and both pCMVΔR8.2 and pMD.G (2). The LTR-LTR junctions were amplified from the low-molecular-weight DNA samples, using PCR and primers derived from the U3 and R sequences in the LTRs (2LTRcircle-F and 2LTRcircle-R). The negative control (-) for PCR included all PCR components but no template DNA. (D) Maintenance of the wt packaging signal or the ΔΨ mutation in a transduced vector. The genomic DNA was extracted from the infected cells described for panel C. The 5′ portions of the vector sequences were detected by PCR, using primers (T7HIV5′R and Gag9) that flank the packaging signal or the deletion site. The identities of the amplified fragments were further confirmed by digestion with the indicated restriction enzymes. (E) Alu PCR analysis to detect integrated forms of wt and ΔΨ vectors. The genomic DNA extracted for panel D was used as a template in a nested PCR using primers derived from Alu and vector sequences. The amplified smear (left panel) was analyzed by Southern blotting (right panel) using a vector-specific probe. Cells infected with media lacking virions are shown in the lanes labeled “mock”.

FIG. 4.

Dimerization assays showing the absence of normal dimers in the ΔΨ vector. (A) In vitro dimerization assay. RNA fragments of wt and ΔΨ vectors encompassing sequences starting from the 5′ R and ending 200 nucleotides downstream of the gag initiation codon were transcribed in vitro. The RNA (25 μg/ml) was heat denatured at 95°C in dimerization buffer, then placed on ice immediately. A portion of the RNA was then incubated at 37°C to allow for dimer formation. RNA forms (250 ng/sample) were resolved in a 1% agarose gel containing ethidium bromide. (B) In vivo dimerization assay. Virions were purified from media of 293T cells cotransfected with either wt or ΔΨ and the helper plasmids through 25% sucrose cushions and ultracentrifugation. Virions were lysed with lysis buffer containing SDS, proteinase K, and tRNA. RNA was then purified with phenol-chloroform and ethanol precipitation. Equal-sized samples of wt and ΔΨ RNA were heated to the indicated temperatures for 10 min and separated on nondenaturing agarose gel. The RNA forms were detected by Northern blot analyses using a labeled antisense riboprobe (U3-R-gfp) harboring the vector-specific U3, R, and gfp sequences.

FIG. 5.

Specificity of vector RNA packaging. (A) Semiquantitative PCR was performed to detect packaged RNA in VLPs from producer cells that were transfected with helper plasmids and either pHR′CMV-GFP (wt), pHR′CMV-GFPΔΨ (ΔΨ), pEGFP-N3 (pGFP) or pQCXIP-EGFP (Mvec). The lanes labeled “mock” show control cells transfected with no DNA. RNA was extracted from the producer cells and VLP-containing media, treated with DNase, and then reverse transcribed. The cDNA (marked as RT-PCR) was serially diluted at a ratio of 1:10, as were equal-sized samples of negative controls consisting of matched RNA (marked as PCR). All these samples were then PCR amplified using primers specific to GFP. Real-time qPCR was performed on the same samples, and the values shown are standardized to the amount of wt RNA for cells and virions. (B) Western blot analysis of HIV capsid protein. Virions from the samples transfected for panel A were purified through 25% sucrose cushions and ultracentrifugation. Virion pellets were analyzed by Western blot analysis with anti-CA antibodies. The intensities of the bands were quantified by densitometry, and the values were standardized to that for the wt (shown below the panel). (C) Semiquantitative and quantitative PCRs were performed specifically to detect viral genomic RNA in the samples described for panel A. ND, not detected. The star indicates an empty lane with a spill-over from the adjacent lane.