Primate and feline lentivirus vector RNA packaging and propagation by heterologous lentivirus virions

Abstract

Development of safe and effective gene transfer systems is critical to the success of gene therapy protocols for human diseases. Currently, several primate lentivirus-based gene transfer systems, such as those based on human and simian immunodeficiency viruses (HIV/SIV), are being tested; however, their use in humans raises safety concerns, such as the generation of replication-competent viruses through recombination with related endogenous retroviruses or retrovirus-like elements. Due to the greater phylogenetic distance from primate lentiviruses, feline immunodeficiency virus (FIV) is becoming the lentivirus of choice for human gene transfer systems. However, the safety of FIV-based vector systems has not been tested experimentally. Since lentiviruses such as HIV-1 and SIV have been shown to cross-package their RNA genomes, we tested the ability of FIV RNA to get cross-packaged into primate lentivirus particles such as HIV-1 and SIV, as well as a nonlentiviral retrovirus such as Mason-Pfizer monkey virus (MPMV), and vice versa. Our results reveal that FIV RNA can be cross-packaged by primate lentivirus particles such as HIV-1 and SIV and vice versa; however, a nonlentivirus particle such as MPMV is unable to package FIV RNA. Interestingly, FIV particles can package MPMV RNA but cannot propagate the vector RNA further for other steps of the retrovirus life cycle. These findings reveal that diverse retroviruses are functionally more similar than originally thought and suggest that upon coinfection of the same host, cross- or copackaging may allow distinct retroviruses to generate chimeric variants with unknown pathogenic potential.

FIG. 1

Schematic representation of HIV-1, SIV, and MPMV packaging constructs and transfer vectors used in the cross-packaging studies. (A) HIV-based packaging construct and transfer vector. (B) SIV-based packaging construct and transfer vector. (C) MPMV-based packaging construct and transfer vector. (D) Control vector without any packaging signal but containing the SV-Hyg^r cassette to measure any signs of retrofection. CMV, cytomegalovirus promoter/enhancer; SV, simian virus 40 early promoter; hyg^r, hygromycin B phosphotransferase gene; SD, splice donor; CTE, constitutive transport element from MPMV.

FIG. 2

Schematic representation of the FIV packaging construct and transfer vectors used in the cross-packaging analysis of FIV RNA into HIV-1, SIV, and MPMV particles and vice versa. (A) FIV genome with open reading frames (ORFs) and FIV-based packaging construct. The U3 region of the FIV 5′ LTR was replaced with the CMV promoter to allow expression in human cells. (B) FIV-based transfer vectors. MB74, -72, -87, and -89 are chimeric derivatives of TR394 that contain portions of either HIV-1 or SIV sequences. MB89 contains a deletion in the Tat ORF, eliminating Tat expression. CMV, human cytomegalovirus promoter/enhancer; SV, simian virus 40 early promoter; hyg^r, hygromycin B phosphotransferase gene; SD, splice donor; CTE, constitutive transport element from MPMV.

FIG. 3

Packaging of retroviral RNAs into HIV-1 (A), SIV (B), FIV (C), and MPMV (D) particles. The transfer vector used in each trans- complementation assay is indicated to the left of panel I or at the top of panel IV. Panels I to III are slot blot analyses of cellular and viral RNAs hybridized with ³²P-labeled probes that anneal either to a housekeeping gene, the β-actin (panel I), or the hygromycin gene in the transfer vector (panels II and III). Whole-cell or viral RNAs were isolated from transiently transfected Cos cells in the three-plasmid trans-complementation assay and were DNase treated, diluted, and subjected to slot blot analysis. Panel I represents a 1:2 dilution of 1 μg of cellular RNA (1, 0.5, and 0.25 μg), while panel II represents a 1:5 dilution of 5 μg of cellular RNA (5, 1, and 0.2 μg) blotted. Panel III contains a 1:2 dilution of purified virion RNA harvested from a 3/4 portion of 9 ml of virus-containing medium from the transfected cultures. Panel IV shows the Western blot analyses of the remaining 1/4 portion of the purified virus particles harvested from each transfection in the corresponding trans-complementation assays using appropriate antisera (see Materials and Methods for details). TR174 (C), control transfer vector that does not contain the 5′ LTR, PBS, or the packaging signal; Mock, cells transfected without DNA.

FIG. 4

The hygromycin-resistant colonies obtained from the successful transduction of the hygromycin gene by HIV-1 particles contain only the proviral FIV sequences and not FIV transfer vector sequences. (A) Schematic representation of the PCR amplification strategy used to distinguish between proviral FIV DNA from FIV transfer vector DNA. Upon reverse transcription and integration, the resulting FIV provirus should contain a wild-type 5′ LTR rather than the artificially introduced chimeric CMV/LTR. Two rounds of PCR were conducted to amplify the specific products; the first round utilized the outer primers, while the second round utilized the inner primers shown. Small black arrows, primers used in the PCR amplification procedure; CMV, cytomegalovirus promoter/enhancer; SV, simian virus 40 early promoter; hyg^r, hygromycin B phosphotransferase gene; SD, splice donor; CTE, constitutive transport element from MPMV. (B) Ethidium bromide-stained images of agarose gels containing the amplified nested PCR products. pTR394, transfer vector plasmid containing the chimeric CMV/FIV LTR; p34TF10, FIV molecular clone containing the wild-type FIV LTR. The sizes of the expected products from each round of PCR are shown below the agarose gels. Sizes of the molecular size markers are indicated to the left of the image.

FIG. 5

Comparison of retroviral cis-acting sequences important during reverse transcription and integration, including the following: (A) PBS, primer binding site; (B) PPT, polypurine tract; (C) U3 att, 3′ attachment site in the U3 region of the 3′ LTR; and (D) U5 att, 5′ attachment site in the U5 region of the 5′ LTR. The canonical TG dinucleotide in the U3 att sequences and CA dinucleotide in the U5 att sequences are boxed. All sequences are shown in comparison to HIV-1, where “–” indicates homology with HIV-1, while differences are shown by the actual nucleotide. Columns to the right indicate the percent homology of each viral sequence to either HIV-1 or FIV with homology to self indicated as 100%.