Improved transduction of human sheep repopulating cells by retrovirus vectors pseudotyped with feline leukemia virus type C or RD114 envelopes

Abstract

Gene therapy for hematopoietic diseases has been hampered by the low frequency of transduction of human hematopoietic stem cells (HSCs) with retroviral vectors pseudotyped with amphotropic envelopes. We hypothesized that transduction could be increased by the use of retroviral vectors pseudotyped with envelopes that recognize more abundant cellular receptors. The levels of mRNA encoding the receptors of the feline retroviruses, RD114 and feline leukemia virus type C (FeLV-C), were significantly higher than the level of gibbon ape leukemia virus (GaLV) receptor mRNA in cells enriched for human HSCs (Lin- CD34+ CD38-). We cotransduced human peripheral blood CD34+ cells with equivalent numbers of FeLV-C and GALV or RD114 and GALV-pseudotyped retroviruses for injection into fetal sheep. Analysis of DNA from peripheral blood and bone marrow from recipient sheep demonstrated that FeLV-C- or RD114-pseudotyped vectors were present at significantly higher levels than GALV-pseudotyped vectors. Analysis of individual myeloid colonies demonstrated that retrovirus vectors with FeLV-C and RD114 pseudotypes were present at 1.5 to 1.6 copies per cell and were preferentially integrated near known genes We conclude that the more efficient transduction of human HSCs with either FeLV-C- or RD114-pseudotyped retroviral particles may improve gene transfer in human clinical trials.

Figure 1.

Northern blot analysis of FLVCR, RDR, and GALVR mRNA levels in RNA from HeLa, K562, and human bone marrow (HuBM) cells. Total RNA extracted from the indicated cells was separated on agarose gels and transferred to a nylon filter. The filter was sequentially probed with a DNA fragment containing FLVCR and GALVR sequences (A), a DNA fragment containing RDR and GALVR sequences (B), and a DNA fragment containing actin sequences (C). The approximate sizes of the mRNAs are indicated at the right. Blots in panels A and B were exposed for approximately 48 hours; panel C, approximately 10 hours.

Figure 2.

RT-PCR analysis of GALVR, RDR, and FLVCR mRNA levels in human bone marrow and hematopoietic progenitor and stem cells. RNA from FACS-sorted bone marrow fractions was analyzed by RT-PCR. The data for unfractionated (Unfr.) bone marrow was normalized to the data from Northern blot analysis of the same RNA samples as in Figure 1. The level of GALVR mRNA in unfractionated bone marrow was set as 1 and compared with the relative amount of RDR and FLVCR mRNA in unfractionated bone marrow, Lin^– CD34⁺ CD38⁺, and Lin^– CD34⁺ CD38^– cells. *Significantly higher than the level of GALVR mRNA (P < .001). **Significantly higher than the level of GALVR mRNA (P < .001); significantly lower than the level of FLVCR mRNA (P < .001). Error bars indicate standard deviation.

Figure 3.

Transduction of human CFU-Cs with GALV, RD114, or FeLV-C pseudotyped retroviral vectors. Slot blot analysis of 1.0 mL virus-containing medium was used to compare the titer of the different virus preparations (top panel). Human CD34⁺ peripheral blood CD34⁺ cells were transduced with GALV, RD114, or FeLV-C pseudotyped retrovirus particles. Transduced CFU-Cs were identified by PCR analysis of DNA extracted from each colony. Data shown was pooled from 3 experiments. Error bars indicate standard deviation.

Figure 4.

Cotransduction of K562 and human peripheral blood CD34⁺ cells with FeLV-C and GALV or RD114 and GALV-pseudotyped retrovirus particles. Slot blot analysis of 1.0 mL and 0.1 mL virus-containing medium was used to compare the titer of the different virus preparations (A). These titers were used to adjust the volumes of supernatants to ensure that equivalent numbers of each pseudotype were used. Human CD34⁺ peripheral blood CD34⁺ cells were cotransduced with FeLV-C and GALV or RD114 and GALV-pseudotyped virus preparations over a 4-day period. Parallel cultures of K562 cells were cotransduced with the same virus preparations each day (d1, d2, d3, d4). One culture was cotransduced over the 4-day period (d1-4). DNA was extracted from the K562 cultures and a portion of the CD34⁺ cells for analysis of the integration of the FeLV-C or RD114 and GALV-pseudotyped viruses by PCR using primers that span the nls in the MFG-nlsLacZ vector. (B) The analysis of FeLV-C/GALV- and RD114/GALV-transduced K562 cells is shown in the left and center panels. Sequences from the human β-globin were amplified in the same samples as a control. The analysis of FeLV-C/GALV (G/F)– and RD114/GALV (G/R)–transduced CD34⁺ cells used for transplantation into fetal sheep is shown in the right panel.

Figure 5.

Cotransduction of human S-RCs with FeLV-C and GALV or RD114 and GALV-pseudotyped retrovirus particles. Human CD34⁺ peripheral blood CD34⁺ cells were cotransduced with FeLV-C and GALV or RD114 and GALV-pseudotyped virus preparations over a 4-day period and transplanted into fetal sheep. DNA was extracted from sheep peripheral blood cells 6 months after transplantation and analyzed for the presence of human cells and the integration of the FeLV-C or RD114 and GALV-pseudotyped viruses by PCR analysis of 0.4 μg DNA. (A) Sequences from the human β-globin gene were amplified to detect the presence of human cells. F1 and F2 indicate analysis of DNA from 2 lambs infused with human CD34⁺ cells exposed to FeLV-C (LacZ) and GaLV (nls-LacZ) pseudotyped particles; R1a, R1b, and R2, analysis of DNA from 3 lambs infused with human CD34⁺ cells exposed to RD114 and GALV-pseudotyped particles; R1a and R1b, DNA from twin lambs infused with the same population of cells; –C, DNA extracted from control sheep; and +C, DNA extracted from K562 cells. (B) The relative amounts of integrated FeLV-C, RD114, and GALV-pseudotyped vectors were measured using primers that span the nls in the MFGs-nlsLacZ vector. 1.0 nls-LacZ indicates 3T3-cell DNA containing a single copy of the MFGs-nlsLacZ vector; 1.0 LacZ, 3T3-cell DNA containing a single copy of the MFGs-LacZ vector; 0.5 LacZ, 3T3-cell DNA containing a 1:1 dilution of this 1.0 LacZ DNA with untransduced 3T3-cell DNA; and –C, DNA extracted from control sheep. For quantitation, only the β-globin band and the 289-bp MFGs-LacZ and 310-bp MFGs-nlsLacZ bands were analyzed.

Figure 6.

Transduction of lymphoid and myeloid cells. Human lymphocytes (L), monocytes (M), and neutrophils (N) were enriched from the peripheral blood of sheep transplanted with transduced cells by light scattering and cell-surface antigen expression. DNA from these cells was analyzed for the presence of the human β-globin gene and the MFGs-LacZ and MFGs-nlsLacZ proviruses as in Figure 5.