Zinc-finger nuclease editing of human cxcr4 promotes HIV-1 CD4(+) T cell resistance and enrichment

Abstract

HIV-1-infected individuals can harbor viral isolates that can use CCR5, as well as CXCR4, for viral entry. To genetically engineer HIV-1 resistance in CD4(+) T cells, we assessed whether transient, adenovirus delivered zinc-finger nuclease (ZFN) disruption of genomic cxcr4 or stable lentiviral expression of short hairpin RNAs (shRNAs) targeting CXCR4 mRNAs provides durable resistance to HIV-1 challenge. ZFN-modification of cxcr4 in CD4(+) T cells was found to be superior to cell integrated lentivirus-expressing CXCR4 targeting shRNAs when CD4(+) T cells were challenged with HIV-1s that utilizes CXCR4 for entry. Cxcr4 disruption in CD4(+) T cells was found to be stable, conferred resistance, and provided for continued cell enrichment during HIV-1 infection in tissue culture and, in vivo, in peripheral blood mononuclear cell transplanted NSG mice. Moreover, HIV-1-infected mice with engrafted cxcr4 ZFN-modified CD4(+) T cells demonstrated lower viral levels in contrast to mice engrafted with unmodified CD4(+) T cells. These findings provide evidence that ZFN-mediated disruption of cxcr4 provides a selective advantage to CD4(+) T cells during HIV-1 infection.

Figure 1

Short hairpin RNA (shRNA) and cxcr4 zinc-finger nuclease (ZFN)-mediated disruptions of surface CXCR4 affords variable protection from HIV-1 challenge. (a) CXCR4 expression on SupT1 T cells 7 days after transduction with lentiviral vectors containing CXCR4 shRNAs, siX4-1 or siX4-2, or an Ad5/F35 vector encoding cxcr4-ZFN, X4-ZFN. Treatment is noted on the top of each panel. Nontransduced SupT1 T cells (isotype control antibody), filled histograms, CXCR4 expression on the nontransduced SupT1 T cells, solid line histograms, and CXCR4 expression on siX4-1, siX4-2, or X4-ZFN transduced SupT1 T cells, dashed line histograms. For X4-ZFN transduced SupT1 T cells, note the distinct CXCR4 positive and negative populations. The values shown are the percentages of CXCR4 positive cells as defined by fluorescence intensity greater than the isotype control values of the nontransduced SupT1 T cells. Flow cytometry analyses and cell sorting are discussed in the Material and Methods section. (b) CXCR4 expression was low on siX4-1, siX4-2, or X4-ZFN transduced SupT1 T cells isolated for the absence of CXCR4. Cell lines generated in (a), above, were sorted for the absence of cell surface CXCR4. Each treatment is noted on the top of each panel. Nontransduced SupT1 T cells (isotype control antibody), filled histograms, CXCR4 expression on the nontransduced SupT1 T cells, solid line histograms, and CXCR4 expression on siX4-1, siX4-2, or X4-ZFN transduced SupT1 T cells, dashed line histograms. The values shown are the percentages of CXCR4 positive cells as defined by fluorescence intensity greater than the isotype control values of the nontransduced SupT1 T cells. (c) Only cxcr4 ZFN-modified SupT1 T cells were resistant to HIV-1 infection. Sorted, siX4-1, siX4-2, or X4-ZFN-modified SupT1 T cells, described in (b), were exposed to HIV-1_NL4-3 at an multiplicity of infection (MOI) of 0.1 and then HIV-1 infection was followed for 42 days. The percentage of HIV-1-infected SupT1 T cells was quantified by determining intracellular viral p24-positive cells with flow cytometry at the indicated times postinfection. Given the presence of significant viral p24-positive cells at day 12 (>60%) in nontransduced cells (NTD), siX4-1, and siX4-2 SupT1 T cell cultures, and ultimately viral-mediated depletion of these cells within 3–6 days, they were not evaluated further. Values represent the mean of duplicate infections and the difference between individual values composing each mean did not vary >10%. The kinetics of HIV-1 infection of SupT1 T cells transduced with empty lentiviral or Ad5/F35 vectors, which did not contain CXCR4 shRNAs or cxcr4-ZFN, is shown in Supplementary Figure S2.

Figure 2

Cxcr4 zinc-finger nuclease (ZFN)-modified SupT1 T cells were enriched after CXCR4-tropic HIV-1 challenge. (a) Cxcr4 ZFN-modified SupT1 T cells were protected from HIV-1_NL4-3 infection. SupT1 T cells were transduced with Ad5/F35 vectors expressing cxcr4-ZFN (unsorted), in Figure 1a, were determined to be 45% CXCR4 negative and 55% CXCR4 positive (data not shown). CXCR4 negative SupT1 T cells (sorted), prepared as described in Figure 1b, were used as a CXCR4 negative control. nontransduced cells (NTD), X4-ZFN unsorted, and X4-ZFN sorted SupT1 T cells were exposed to HIV-1_NL4-3 at an multiplicity of infection (MOI) of 0.1 and HIV-1 infection was followed for 35 days. The number of HIV-1 infected SupT1 T cells was quantified by flow cytometry at the indicated times postinfection, as discussed in Figure 1c. Given the presence of significant viral p24-positive cells at day 10 (>50%) in NTD SupT1 T cell cultures, and ultimately viral-mediated depletion 3–6 days later, they were not evaluated further. Values represent the mean of duplicate infections and the differences between values composing each mean did not exceed 15%. (b) Levels of cellular HIV-1_NL4-3 DNA decrease over time in cxcr4 ZFN-modified SupT1 T cells. Cellular HIV-1 DNA levels in NTD, X4-ZFN unsorted, and X4-ZFN sorted SupT1 T cells from (a) were determined at the indicated times post HIV-1 infection using quantitative PCR (qPCR). Given the presence of significant viral p24-positive cells at day 10 (>50%) in NTD SupT1 T cell cultures, and ultimately viral-mediated depletion 3–6 days later, they were not evaluated further. HIV-1 cellular DNA levels were expressed relative to hydroxymethylbilane synthase (HMBS) DNA levels, which serve as a DNA extraction control. Values represent the mean of duplicate qPCRs from two infections, where the individual qPCR values of each mean did not differ by >10%. (c) Surface CXCR4 negative SupT1 T cells enrich over NTD after HIV-1_NL4-3 infection. Surface CXCR4 expression levels on NTD, X4-ZFN unsorted, and X4-ZFN sorted SupT1 T cells were followed by flow cytometry at the indicated times after HIV-1 infection. CXCR4 percentages represent the mean of duplicate infections with individual values of each mean not differing by >15%. (d) Cxcr4 ZFN-modified cells increased over time after HIV-1 infection of X4-ZFN unsorted SupT1 T cells. Cxcr4 genomic disruption in X4-ZFN Unsorted SupT1 T cells from (b) was determined over time by utilizing the Surveyor nuclease assay (see Materials and Methods section). The lower migrating bands (indicated by arrows) in each lane indicates disrupted cxcr4 alleles. Note that by day 17 the ability to accurately estimate the level of genomic disruption was beyond the linear range of the assay and was designated as >80%. The percentage of genomic disruption was determined as described in the Materials and Methods section.

Figure 3

Zinc-finger nuclease (ZFN)-modification of cxcr4 of uninfected SupT1 T cells present in a T cell population undergoing spreading HIV-1 infection confers viral resistance and enrichment. (a) An established HIV-1_NL4-3 infection in SupT1 T cells was eliminated upon cxcr4-ZFN transduction. SupT1 T cells were first infected with HIV-1_NL4-3 at an multiplicity of infection (MOI) of 0.02 for 6 days before transduction with lentiviral vectors delivering CXCR4 short hairpin RNAs (shRNAs), siX4-1 or siX4-2 or Ad5/F35 delivering cxcr4-ZFN (X4-ZFN). Viral progression and protection of siX4-1, siX4-2, or X4-ZFN SupT1 T cells, as compared to nontransduced cells (NTD), was followed by flow cytometry at the indicated times post HIV-1 infection, as discussed in Figure 1c. HIV-1_NL4-3 with an arrow, day 0, indicates the time of viral infection and X4-ZFN with an arrow, at day 6, indicates the time point at which the infected cell population was treated with siX4-1, siX4-2, or cxcr4-ZFN. Given the presence of significant viral p24-positive cells at day 17 (60–80%) in NTD, siX4-1, and siX4-2 SupT1 T cell cultures, and ultimately viral-mediated depletion of the infected transduced cell populations 3 days later, they were not evaluated further. Values represent the mean of duplicate infections and differences between each of the two values used to determine each mean did not exceed 14%. (b) Levels of HIV-1 DNA decrease over time in ZFN-modified SupT1 T cells. HIV-1 DNA levels of SupT1 T cells from (a) were determined at the indicated times post HIV-1 infection using quantitative PCR (qPCR). Given the presence of significant viral p24-positive cells at day 17 (60–80%), and ultimately viral-mediated depletion of both NTD and siX4s transduced cell populations, they were not evaluated for cellular HIV-1 after day 6. The day 6 NTD culture served as a viral-infected cell control. HIV-1 DNA levels in the X4-ZFN and the day 6 NTD cells are expressed relative to hydroxymethylbilane synthase (HMBS) DNA levels, which serve as a DNA extraction control. For reference, HIV-1_NL4-3 with an arrow, day 0, indicates the time of viral infection and X4-ZFN with an arrow, day 6, indicates the time point at which the infected cell population was treated with vectors delivering siX4-1, siX4-2, or X4-ZFN. Values represent the mean of duplicate qPCRs from two infections, where the individual qPCR values of each mean did not differ by >10%. (c) X4-ZFN SupT1 T cells were highly enriched over NTD during concurrent infection. CXCR4 surface expression on SupT1 T cells from (a) was determined at indicated times after HIV-1 infection by flow cytometry to evaluate the changes in CXCR4 over time. Note that the NTD cell population contained few CXCR4 negative cells and by day 17 all NTD cells were virally depleted.

Figure 4

Cxcr4 zinc-finger nuclease (ZFN)-modified primary CD4⁺ T cells have a survival advantage during HIV-1 infection. (a) ZFN-modification reduces CXCR4-tropic HIV-1 infection in primary CD4⁺ T cells. CD4⁺ T cells were stimulated with anti-CD3/CD28 beads for 1 day and then nontransduced cells (NTD) or transduced with the Ad5/F35 cxcr4-ZFN vector (X4-ZFN) at an multiplicity of infection (MOI) of 500. Day 7 after transduction, the cells were infected with HIV-1_LAI at an MOI of 1.0. Infection of CD4⁺ T cells were quantified by flow cytometry, as discussed in Figure 1c, at the indicated times post HIV-1 infection. Values represent the mean of duplicate infections, with the individual values of each mean not differing by >10%. (b) Levels of cellular HIV-1 DNA decrease over time in cxcr4 ZFN-modified (X4-ZFN) CD4⁺ T cells, but not nontransduced (NTD) CD4⁺ T cells. Cellular HIV-1 DNA levels in CD4⁺ T cells from (a) were determined at the indicated times post HIV-1 infection using quantitative PCR (qPCR). HIV-1 DNA levels in the cxcr4 ZFN-modified and NTD were analyzed as discussed in Figure 2b. Values represent the mean of duplicate qPCRs from two infections, where the individual qPCR values of each mean did not differ by >10%. (c) Cxcr4 ZFN-modified CD4⁺ T cells were susceptible to CCR5-tropic HIV-1 infection. CD4⁺ T cells were stimulated and nontransduced (NTD) or Ad5/F35 cxcr4-ZFN vector transduced (X4-ZFN) as described in (a). Day 7 after transduction the cells were infected with HIV-1_JR-FL at an MOI of 1.0. Infection of CD4⁺ T cells was followed over time using flow cytometry as discussed in Figure 1c. Values represent the mean of duplicate infections, with the individual values of each mean not differing by >10%. (d) Cxcr4 ZFN-modified CD4⁺ T cells enrich over time after CXCR4-tropic HIV-1 challenge. CD4⁺ T cells were stimulated and nontransduced (NTD) or Ad5/F35 cxcr4-ZFN vector (X4-ZFN) transduced, as discussed in (a). The Surveyor assay was used to analyze cxcr4 genomic disruption over time in ZFN-modified CD4⁺ T cells exposed to CXCR4-trophic HIV-1_LAI and HIV-1_NL4-3, the CCR5-trophic HIV-1_JR-FL, or of uninfected CD4⁺ T cells. The percentage of cxcr4 genomic disruption was determined as described in Figure 2d and findings presented as fold-change over time relative to initial genomic disruption.

Figure 5

Selective advantage of cxcr4 zinc-finger nuclease (ZFN)-modified primary CD4⁺ T cells during HIV-1 infection in vivo in NSG mice. (a,b) Cxcr4-modified CD4⁺ T cells were present in the peripheral blood of NSG mice post HIV-1 infection. CD4⁺ T cells, either cxcr4-ZFN-modified or without modification, were adoptively transferred into NSG mice together with autologous peripheral blood mononuclear cells (PBMCs) that were not infected (a) or HIV-1 infected with CXCR4-tropic HIV-1_NL4-3 (b), as discussed in the Materials and Methods. The ZFN-modified CD4⁺ T cells used for these studies showed a 16% cxcr4 allelic disruption before transfer to NSG mice. A schematic of the experimental design is show in Supplementary Figure S6 and all methodology is presented in Materials and Methods section. Peripheral bleeds from mice were analyzed every other week for human CD3, CD4, and CD8 T cells by flow cytometry for up to 52 days after adoptive transfer of PBMCs to mice. The CD4⁺ T cell levels from individual mice with cxcr4 ZFN- (solid symbols) or unmodified (open symbols) CD4⁺ T cells are plotted. Numbers of mice/group: No HIV-1, unmodified: n = 7 and cxcr4 ZFN-modified: n = 8 (1 mouse died after the second bleed). HIV-1 infected, unmodified: n = 7 and cxcr4 ZFN-modified: n = 8 (note that n = 4 samples for day 52 analysis of CD4⁺ T cells, which is the result of the loss of 4 of 8 blood samples during collection). 1 mouse did not demonstrate detectable human CD4⁺ T cells in the peripheral blood, but did from spleen. Median of the values is shown as a solid line. *0.01 < P < 0.05, ** 0.001 < P < 0.01, *** P < 0.001 (unpaired two-tailed t-test at the 95% confidence level). The 2-day difference in the length of the study period of the NO HIV and HIV-1_NL4-3 mice results from a staggered sacrifice procedure to collect tissue. (c,d) CD4⁺ to CD3⁺ ratio remains high post HIV-1 infection in mice receiving cxcr4 ZFN-modified CD4⁺ T cells. The human CD4⁺/CD3⁺ ratio, determined by flow cytometry analyses of peripheral blood, from mice provided with cxcr4 ZFN- (solid symbols) and unmodified (open symbols) CD4⁺ T cells are plotted from uninfected mice (c) and mice infected with CXCR4-tropic HIV-1_NL4-3 (d). Median of the values is shown as a solid line. *0.01 < P < 0.05, **0.001 < P < 0.01, ***P < 0.001 (unpaired two-tailed t-test at confidence 95%). (e,f) Total splenic CD4⁺ T cells were abundant and the CD4⁺ to CD3⁺ ratio remains high post HIV-1 infection in mice receiving cxcr4 ZFN-modified CD4⁺ T cells at the conclusion of the study. Left panel (e), depicts total splenic CD4⁺ T cell numbers, at day 50, from uninfected mice (NO HIV-1) or at day 52 for HIV-1 infected (HIV-1_NL4-3) mice, which were engrafted with cxcr4 ZFN- (solid symbols) and unmodified (open symbols) CD4⁺ T cells and peripheral blood mononuclear cell (PBMC). The right panel (f), depicts splenic CD4⁺/CD3⁺ T cell ratios, at day 50, from uninfected mice (NO HIV-1) or HIV-1 at day 52 for infected mice (HIV-1_NL4-3) which were engrafted with cxcr4 ZFN- (solid symbols) and unmodified (open symbols) CD4⁺ T cells and PBMC. All groups of mice from (a,b) were necropsied 50 or 52 days after CD4⁺ T cell and PBMC transfer, CD4⁺ T cells were isolated from spleens of mouse groups not infected with HIV-1 and provided with unmodified CD4⁺ T cells, n = 7, or cxcr4 ZFN-modified CD4⁺ T cells, n = 6 mice (note: 8 mice were enrolled in the study, 1 mouse died after the second bleed and 1 sample was lost during the final CD4⁺ T cell collection). The HIV-1-infected mouse group which did not receive cxcr4 ZFN-modified CD4⁺ T cells had n = 7 mice, and mice that received cxcr4 ZFN-modified CD4⁺ T cells, had n = 8 mice. Median of the values is shown as a solid line. *0.01 < P < 0.05, **0.001 < P < 0.01, ***P < 0.001 (unpaired two-tailed t-test at confidence 95%). (g) Reduction in cellular HIV-1 DNA levels in splenic CD4⁺ T cells from mice receiving cxcr4 ZFN-modified CD4⁺ T cells. Shown are relative amounts of cellular HIV-1 DNA determined by qPCR from human CD4⁺ T cells isolated from spleens of mice, which were provided with non- (open symbols) or cxcr4 ZFN-modified (solid symbols) CD4⁺ T cells, which were not infected with HIV-1 (NO HIV-1) or infected with HIV-1 (HIV-1). For the HIV-1-infected group which did not receive cxcr4 ZFN-modified CD4⁺ T cells, all CD4⁺ T cells were pooled since too few CD4⁺ T cells from each mouse was obtained due to HIV-1-mediated cell loss to undertake individual assessment. CD4⁺ T cells from individual mice of the cxcr4 ZFN-modified CD4⁺ T cell group were assessed given adequate numbers of CD4⁺ T cells. HIV-1 DNA levels in all groups are expressed relative to hydroxymethylbilane synthase (HMBS) DNA levels, which serve as a DNA extraction control. Median of the values is shown as a solid line. Values represent the mean of duplicate qPCRs with a SE <10%. (h) NSG mice receiving cxcr4 ZFN-modified CD4⁺ T cells demonstrated enrichment for splenic CD4⁺ T cells with cxcr4 genomic disruption post HIV-1 infection. A plot of the percent of cxcr4 genomic disruption is shown for isolated splenic CD4⁺ T cells from mice that were provided with cxcr4 ZFN-modified CD4⁺ T and not infected with HIV-1 (NO HIV-1, n = 6 mice) and infected with HIV-1 (HIV-1, n = 8 mice). Information on NSG mouse groups is discussed in (c). Total DNA of human CD4⁺ T cells isolated from the mouse spleens was subjected to the Surveyor nuclease assay to assess allelic cxcr4 disruption. The Surveyor nuclease assay gel information is presented in Supplementary Figure S8. Median of the values is shown as a solid line. *0.01 < P < 0.05, **0.001 < P < 0.01, ***P < 0.001 (unpaired two-tailed t-test at confidence 95%).