The HIV-1 (NL4.3) core is restricted, but the NL4.3 envelope rescues the restriction in HeLa/CD4 cells. The restriction refers to the infectious titer on U87/CD4/CXCR4 cells compared to that on HeLa/CD4 cells. Error bars were calculated from the results from at least two independent experiments (see Materials and Methods). MCRΔenv or HIV-1 NL4.3Δenv was pseudotyped with NL4.3, MCR, sMCN, or VSV-G envelopes called MCR (NL4.3), NL4.3 (MCR), NL4.3 (sMCN), and NL4.3 (VSV-G). *, the HIV-1 NL4.3 core-based vectors encoding GFP were pseudotyped with MCR, sMCN, and VSV-G in a similar way (Materials and Methods). The titers of these viruses were determined by flow cytometry, and the restriction (n-fold) was plotted as the reduction of the HeLa/CD4 cell titer compared to that of U87/CD4/CXCR4 cells, as described before.

(A) A schematic diagram of MCR, MCN, and chimeras shows their phenotypes on HeLa/CD4 cells. (B) A working model for the results of the tropism studies are presented here. Restrictive cells have at least two routes of entry. One, which is mediated by a restrictive MCR envelope, can result in a route where Lv2 is operative, if the virus also has a restrictive core (a). This entry mechanism will result in a dead end to infection. If the virus has an unrestrictive core, such as MCN, then the Lv2 is bypassed (b) or the virus is rerouted to a permissive pathway (c and d). Nonrestrictive envelopes deliver the core to a permissive pathway (d) and bypass Lv2.

The restriction to infection of HeLa/CD4 and GHOST/CD4/CXCR4 cells is distinct from that of Fv1/Ref1/Lv1. (a) The Ref1-restricted MLV-N (VSV-G) and unrestricted MLV-B (VSV-G) were titrated on U87/CD4/CXCR4 and HeLa/CD4 cells and on the Fv1-null murine cell line MDTF. (b) Abrogation of Ref1 in HeLa/CD4 cells by MLV-N. Cells (10⁵) were pretreated either with control medium or with an MOI of 10 of MLV-Npuro (VSV-G) particles. Two hours later, the cells were exposed to an MOI of 0.1 of various VSV pseudotyped retroviral and lentiviral vectors encoding GFP. Inhibition of infection was plotted as the increase (n-fold) in the proportion of GFP-positive cells compared to the untreated cells and was measured by flow cytometry. (c) Pretreatment of cells with an MOI of 16 of MLV-N (VSV-G) or MLV-B (VSV-G) did not abrogate the restriction to infection of MCR on HeLa/CD4 or GHOST/CXCR4 cells.

Schematic genome representation of MCN and MCR and recombinants thereof. A representation of the HIV-2 genome is shown. The nucleic acid numbering system is in accordance with the convention of the first base of the 5′ R region being number 1. The terminal cloning sites of the molecular clones are indicated (NotI and NdeI). Additional unique sites (XhoI and PmlI) used for cloning are also shown (BstEII is not shown). The BsaBI site allows the transfer of the gp120 ectodomain of env between the molecular clones. The spliced donor/acceptor of tat and rev were adopted from the ROD molecular clone (accession no. M15390). A premature stop codon within the envelope of MCR is indicated with an arrow (position 8369). Also, there is a premature stop codon in the nef reading frame at position 8764 in MCR. The full-length molecular clone and swapped fragments of MCN (molecular clone nonrestricted) and MCR (molecular clone restricted) are represented by grey and solid fills, respectively. The white cross within MCR represents the premature stop codon in the CPT of the env gene. Gene swaps are indicated in italics preceded by the clone of origin in lowercase letters. For example, the MCN clone containing the envelope gene for MCR is MCNmcrenv.

Mapping the viral molecular determinants responsible for overcoming the restriction to infection in HeLa/CD4 cells. Molecular clones and recombinant viruses described in the legend to Fig. 2 were titrated on U87/CD4/CXCR4 and HeLa/CD4 cells, and the number of FFU per milliliter was calculated. Restriction refers to the infectious titer on U87/CD4/CXCR4 cells compared to that on HeLa/CD4 cells with error bars calculated from the results from at least two independent experiments (for calculation, see Materials and Methods). (a) Reciprocal swaps of env gene. (b) Restriction (n-fold) when the gag or nef gene is reciprocally swapped between MCR and MCN. (c and d) Reciprocal swaps of both gag and env. Results are representative of at least three experiments. (e) A single point mutation in gag rescues the gag restriction. Site-directed mutagenesis of MCR was used to introduce amino acid changes in Gag at positions 65 (MA), 209 (CA1), and 314 (CA2) or combinations of these. Viruses were titrated, and the restriction was calculated and compared to those of MCR and MCN. The results are representative of at least three experiments. Molecular clones MCR-MA (containing the MA mutation at position 65) and MCR-MA/CA2 (containing mutations at positions 65 and 314) showed no rescue. MCR-CA1 and the triple mutant MCR-MA/CA1/CA2 rescued the restriction to levels seen for MCRmcngag.

Production and characterization of molecular clones. (a) The strategy used to produce molecular clones of the HIV-2 isolates, prCBL-23 and CBL-23, is shown. Two fragments were PCR amplified with common primers. Primers AF and AR amplified a 5′ fragment of 2.2 kb, and primers BF and BR amplified a 3′ 8.5-kb fragment. A schematic representation of the two overlapping PCR products is shown. The overlapping fragments contain a unique XhoI site used to join both fragments. The restriction sites NotI and NdeI were used to insert the full-length HIV-2 genome into the pGEM-TEasy vector to produce replication-competent viruses in culture. (b) Tropism of the molecular clones for human cell lines. MCN and MCR were tested for tropism. Viral supernatants were titrated on the four cell lines indicated. Input was standardized on U87/CD4/CXCR4 cells previously determined to be nonrestrictive (29). After 3 days, infected cells were fixed and immunostained, and the number of FFU per milliliter was calculated. MCR is between 1.5 and 2.0 logs less infectious on HeLa/CD4 and GHOST/CXCR4 cells. NP2/CD4/CXCR4 cells are just as susceptible to infection as the unrestricted U87/CD4/CXCR4 cells. (c) The restriction to infection of HeLa/CD4 and GHOST/CXCR4 cells was previously shown to be overcome by providing the restricted prCBL-23 with a VSV-G envelope. The figure shows that when MCR is pseudotyped with a VSV-G envelope, MCR (VSV-G), the restriction in these cells is completely overcome. (d) To confirm that the restriction to MCR is postentry, as previously described for prCBL23 (29), we designed a Q-PCR assay to determine the levels of strong-stop reverse transcripts (see Materials and Methods). Briefly, strong-stop PCR products were measured from cell lines infected with equal doses of MCN or MCR (500 FFU on 10⁵ U87/CD4/CXCR4 cells). Cells were harvested 1, 2, 6, and 18 h postinfection. Q-PCR was performed on isolated total DNA. Results are expressed as the ratio of copies in unrestricted U87/CD4/CXCR4 cells to those in HeLa/CD4 cells. Figure 1d shows that equivalent levels of reverse transcripts were produced in the restrictive HeLa/CD4 cells by MCR and MCN. Although a fourfold difference in the ratio of reverse transcripts for MCR compared to MCN was noted at the 2-h time point, equivalent levels of transcripts were observed at the 6- and 18-h time points. Thus, MCR and MCN represent the phenotypes of prCBL23 and CBL23 previously described (29).