Introduction

Hematopoietic stem cell gene therapy for HIV-1 infection may provide an approach to treat HIV-1 infection that could be complementary to drug-based approaches. Gene therapy for HIV-1 involves the transduction of CD34⁺ hematopoietic stem/progenitor cells to achieve expression of the transgene in the resultant CD4⁺ myeloid and lymphoid lineage cells in the periphery.

The HIV-1 REV protein functions to allow transport into the cytoplasm of incompletely spliced HIV-1 transcripts that encode the virion proteins GAG, POL, and ENV.^{1, 2, 3} Molecular clones of HIV that were deleted for the rev gene were shown to be unable to replicate.⁴ Although this would make REV an attractive drug target, no drugs directed against this essential viral protein have been developed so far.

The dominant-negative RevM10 mutant of the viral REV protein has been identified as being highly effective at suppressing viral replication.^{5, 6, 7, 8} RevM10 prevents the full-length and incompletely spliced HIV-1 transcripts from being exported from the nucleus to the cytoplasm because it has lost its ability to bind to the nuclear export factor Crm-1, despite retaining its ability to bind to the rev responsive element (RRE) of HIV-1 transcripts and to other REV protein molecules.^{9, 10, 11} Furthermore, RevM10 is a dominant-negative inhibitor of REV function, because one molecule can bind and inactivate several wild-type REV molecules.¹² In contrast to the problem of emergence of resistant HIV-1 to the standard drugs, viral escape mutations to revM10¹³ and mutations of the rev gene, or the RRE,¹⁴ appear to result in less fit virus. Gene transfer to hematopoietic stem cells can deliver the RevM10 dominant-negative gene into the target cells for HIV-1 infection.

Previous clinical studies have utilized murine retroviral vectors to deliver the RevM10 gene into CD4⁺ T cells^{15, 16} and CD34⁺ human hematopoietic progenitor cells.¹⁷ The results of these studies were modest but encouraging and identified the need to increase the transduction efficiency of CD34⁺ cells. Lentiviral vectors promise to be an improved delivery vehicle because of their superior ability to transduce non-dividing cells,¹⁸ such as primitive pluripotent human hematopoietic stem cells.

This study addresses the feasibility of using a lentiviral vector to deliver the RevM10 gene into human hematopoietic progenitor cells for clinical application. Because the codon usage of the rev transcript is suboptimal for translation in human cells, a codon optimized version of RevM10, called humanized revM10 (huM10), had been developed for production of higher levels of the dominant-negative protein.

The production of high-titer lentiviral vector carrying huM10 may be problematic, because the huM10 transgene of the vector will be expressed during the packaging process and may reduce cytoplasmic levels of the REV-dependant transcripts from both the packaging plasmid and the vector plasmid. Another concern using HIV-1-based lentiviral vectors in HIV-1-infected patients is the possibility of mobilization of vector sequences from the transduced cells to other cells within a recipient or to another person, by either recombination between vector sequences and those of wild-type HIV-1 or by pseudotyping of vector transcripts by wild-type HIV-1.

Here, we describe the identification of a self-inactivating (SIN) lentiviral vector, CCL-MND-huM10, and packaging constructs that produce high titers of a vector carrying the huM10 gene. We show the efficacy of the huM10 gene transduced by the lentiviral vector to inhibit replication of the HIV-1 strain JR-FL in the monocytic progeny of human CD34⁺ cells. And, we demonstrate that the SIN lentiviral vector carrying the huM10 gene did not undergo detectable mobilization by wild-type HIV-1.

Results

Identification of packaging plasmids that can produce high-titer lentiviral vectors containing the huM10 gene

One potential problem using a lentiviral vector to carry the huM10 gene is that expression of this dominant-negative rev gene during the packaging process may inhibit expression of the REV-dependent gag and pol transcripts from the packaging plasmid. To identify a packaging plasmid that can tolerate the inhibitory activity of huM10, we compared five different packaging constructs: 8.2, 8.9, pCD-NL/BH^*V 3, pMDLg/pRRE, and SYNGP. The second generation packaging construct 8.2¹⁹ retains the basic HIV-1 genome structure, but it has a large deletion of the envelope (env) with an additional translational stop codon introduced at the end of vpu. In addition to TAT and REV, 8.2 expresses all of the HIV-1 accessory proteins, namely VIF, VPU, VPR, and NEF. In contrast, 8.9,²⁰ which is based on 8.2, contains additional deletions that prevent expression of the accessory proteins VIF, VPU, VPR, and NEF. The pCD-NL/BH^*V 3 contains truncated coding regions for ENV, VIF, VPR, and VPU.²¹ Additionally, it has a 38 bp deletion between the 5' splice donor site and the ATG of gag, which may affect the strength of the splice donor site located in the leader region. A less effective 5' splice donor site may render this packaging construct less REV-dependent.²² The third-generation or split-packaging construct pMDLg/pRRE²³ is more like an expression plasmid than a modified HIV-1 genome structure, in that it is truncated immediately after the polymerase (pol) gene. It contains a 374 bp RRE fragment immediately downstream of pol because expression of the gag and pol genes is expected to remain dependent on REV. pMDLg/pRRE is further modified to contain an optimized translation initiation codon. The packaging plasmid SYNGP²⁴ is a synthetic, codon-optimized HIV-1-packaging plasmid that expresses GAG and POL proteins. pSYNGP does not contain any env sequences including rev sequences. Theoretically, it should be REV-independent because the codon optimization may eliminate the cis-repressive sequences of the native gag/pol transcripts.

Each of these packaging constructs was used to package the HIV-1 inhibitory vector CCL-MhuMIE, containing the huM10 gene linked to enhanced green fluorescent protein (eGFP) via an internal ribosome entry site, as well as the control vector CCL-ME, containing the eGFP gene only (Figure 1, Table 1). All vectors were pseudotyped with VSV-G using the pMD.G plasmid.²⁵

Figure 1.

Schematic diagram of lentiviral vectors. All of the vectors are SIN vectors because of deletions in the U3 region of the long terminal repeat (U3) and contain the packaging signal (), the complete RRE to facilitate RNA export and the central polypurine tract (cPPT) to facilitate nuclear entry of the pre-integration complex. Transcription of the transgene is controlled internally by the U3 region of the MND retroviral vector (MND). The control vector CCL-ME contains the eGFP transgene, the vector CCL-MhuM contains the huM10 transgene, and the vector CCL-MhuMIE contains both the huM10 and the eGFP transgene linked via an internal ribosome entry site.

Full figure and legend (81K)

Table 1 - Lentiviral vector constructs.

Full table

The CCL-ME vector (expressing only eGFP) was produced with an equally high titer of 2.6–5.5 10⁷ TU/ml (transducing units per ml) by all packaging constructs, with the exception of the pSYNGP construct (Figure 2a). The titer of CCL-ME produced by the pSYNGP construct was 99% lower (approximately 130-fold lower) than the CCL-ME titer produced by either MDLg/pRRE and CD-NL/BH^*V 3 (P0.001). Similar reduction of the titer of CCL-ME produced by the pSYNGP were measured when compared to 8.2 and 8.9 (P0.001).

Figure 2.

Titers of lentiviral vectors using different packaging constructs. (a) Comparison of the titers of the CCL-ME and CCL-MhuMIE vectors using different packaging constructs. Vector supernatants were analyzed for the number of GFP-transducing virions on HEK293 cells by serial dilution. The number of transduced HEK cells was determined by FACS and used to calculate TU/ml. The figure compares the mean GFP titers produced by CCL-ME (gray bars) to the mean GFP titers produced by CCL-MhuMIE (white bars) for each packaging plasmid. The SDs are indicated by the error bars (n=5). (b) Comparison of the level of GAG protein produced by the different packaging plasmid in the presence or absence of CCL-MhuMIE. Western blots using an anti-p24 GAG primary antibody were analyzed by densitometry. The top panel shows the amount of p24 GAG protein detected, with the blot re-developed with antibody to extracellular signal-regulated kinase2 as a loading control. The bar graph in the bottom panel compares the amount p24 GAG protein produced from the different packaging plasmid in the presence (white bars) or absence (gray bars) of the huM10 transgene. The amount of p24 GAG is graphed as the ratio of p24 GAG to the loading control protein extracellular signal-regulated kinase2. CEM cells infected with HIV-1 IIIB served as a positive control.

Full figure and legend (101K)

The CCL-MhuMIE vector (expressing huM10 and eGFP) was produced at reduced titers by the majority of the packaging constructs, when compared to the titers the different packaging constructs produced with CCL-ME. The 8.2 or 8.9-packaging constructs produced titers of CCL-MhuMIE that were reduced by 85% (6-fold lower), compared to the titers they produced with CCL-ME. In contrast, the MDLg/pRRE and the CD-NL/BH^*Vx3-packaging construct produced titers of CCL-MhuMIE that were less reduced, namely 69 and 67% (3.3- and 3-fold lower), respectively, compared to the titers they produced with CCL-ME. No reduction of the titer of the CCL-MhuMIE vector occurred, when it was packaged with the SYNGP-packaging construct, compared to CCL-ME; however, similarly to the low titer for CCL-ME with SYNGP, the absolute titer of CCL-MhuMIE packaged by SYNGP was only 0.6 10⁶ TU/ml (Figure 2a).

Significant differences were seen when the absolute titers of CCL-MhuMIE produced by the different packaging constructs were compared (Figure 2a). The absolute titer of CCL-MhuMIE produced by CD-NL/BH^*V 3 was 1.6 10⁷0.6 10⁷ TU/ml and thus significantly higher than those produced by either SYNGP, 8.2, or 8.9 (P0.001). Similar, the absolute titers of CCL-MhuMIE produced by MDLg/pRRE was 1.3 10⁷0.7 10⁷ TU/ml and thus also significantly higher than those produced by either SYNGP, 8.2, or 8.9 (P0.001). The titers of CCL-MhuMIE produced by either SYNGP, 8.2, or 8.9 were not statistically different (P0.156).

In summary, the titers of CCL-MhuMIE produced by either MDLg/pRRE or CD-NL/BH^*V 3 were significantly higher than those produced by either 8.2, 8.9, or SYNGP and were at absolute titers of 1.3–1.6 10⁷/ml, which would be sufficient to support large-scale production for clinical application.

Characterization of packaging plasmids that can produce high-titer lentiviral vectors containing the huM10 gene

The observation that both the pMDLg/pRRE and the pCD-NL/BH^*V 3-packaging constructs were less severely affected by the presence of huM10 in the vector (Figure 2a), raises the question of whether these packaging constructs are less sensitive to inhibition by huM10 or whether they simply produce more viral proteins. To address this question, the levels of HIV-1 p24 GAG protein were measured by Western blotting and densitometric analysis in cells transfected with the packaging plasmids plus either pCCL-ME or pCCL-MhuMIE (Figure 2b).

In the absence of huM10 (cells co-transfected with control CCL-ME vector), the pMDLg/pRRE, pCD-NL/BH^*Vx3, and pSYNGP plasmids produced the highest levels of HIV-1 p24 GAG protein, with 1.5- or 3.6-fold lower amounts seen from either 8.2 or 8.9.

In the presence of huM10 (cells co-transfected with CCL-MhuMIE), the relative reduction of HIV-1 p24 GAG protein was most pronounced for the pMDLg/pRRE and 8.9-packaging constructs (2.1- and 2.6-fold), with no reduction measured for pSYNGP. However, the level of HIV-1 p24 GAG protein produced by the pMDLg/pRRE, pCD-NL/BH^*V 3, and pSYNGP plasmids in the presence of huM10 remained higher than the levels of HIV-1 p24 GAG produced by either 8.2 or 8.9 in the absence of huM10. With the exception of pSYNGP, the HIV-1 p24 GAG level produced by each packaging plasmid correlated well with the measured titer for each vector. This observation indicates that the levels of HIV-1 p24 GAG protein produced from the packaging plasmids are a key determinant of vector titer. The major exception to this generality was the pSYNGP plasmid, which produced high levels of HIV-1 p24 GAG protein, even in the presence of huM10, yet produced low titers of vectors. The mechanism responsible for this observation is unknown.

Both the pMDLg/pRRE and the pCD-NL/BH^*Vx3-packaging plasmids produced the CCL-MhuMIE vector to high titers. The pMDLg/pRRE plasmid is a so-called "third-generation" lentiviral-packaging system developed for potentially increased safety for clinical use, with only the HIV-1 gag and pol genes present and the tat and rev genes removed. In contrast, the pCD-NL/BH^*Vx3 is a "second-generation" packaging plasmid that retains the general HIV-1 genome structure and contains the tat and rev genes, placing it at theoretically higher risk for generating replication-competent lentivirus by recombination with the vector plasmid during co-transfection. Therefore, subsequent studies were carried out using the pMDLg/pRRE-packaging plasmid.

Transduction efficiency of CD34⁺ cells and lack of ex vivo toxicity in CD34⁺ cells by high-titer lentiviral vectors containing the huM10 gene

The lentiviral vectors were evaluated for their ability to transduce efficiently and safely human CD34⁺ cells from healthy donor bone marrow or umbilical cord blood. In addition to the vectors CCL-MhuMIE (expressing huM10 and eGFP) and CCL-ME (expressing only eGFP) described above, the vector CCL-MhuM (expressing only huM10) was added to our comparison study (Figure 1). The CCL-MhuM vector, lacking the eGFP gene, is a prototype for a vector that would be used for clinical trials.

When transduction efficiencies were measured by eGFP expression in the bulk cultured cells, 53% (6.3) of the cells transduced with CCL-ME were positive and 41.4% (3.3) of the cells transduced with CCL-MhuMIE were positive (Figure 3a). When transduction efficiencies were measured by the number of DNA-positive colony forming units (CFU), 66% (CCL-ME), 50% (CCL-MhuMIE), and 55% (CCL-MhuM) of the CFU were positive (Figure 3a). The concordance of measured transduction efficiency by the two different methods was good for the CCL-ME and CCL-MhuMIE vectors that could be measured by both fluorescence-activated cell sorting (FACS) for eGFP expression and polymerase chain reaction (PCR) for vector-containing CFU.

Figure 3.

Analysis of Transduction of human CD34⁺ cells by lentiviral vectors. Human CD34⁺ cells were transduced with either the CCL-ME, CCL-MhuMIE, or CCL-MhuM lentiviral vectors and analyzed for the gene transfer efficiency, hematopoietic progenitor growth, and for CD4-expression. (a) Gene transfer efficiencies determined by FACS analysis of eGFP expression in cells after 7–10 days of bulk culture and by quantitative-PCR of CFU grown in methylcellulose for 14 days. The % of cells expressing eGFP (% eGFP⁺ cells) is shown with white bars. The percentage CFU positive for vector sequences (eGFP and/or huM10) were calculated from analysis of 50 colonies in each experiment. The mean values of all experiments were graphed (hatched bars). (b, c) Numbers and types of hematopoietic colonies grown from transduced CD34⁺ cells from (b) bone marrow and (c) cord blood. The SDs are indicated by the error bars (n=4). The white bars represent the number of granulocytic/monocytic colonies (CFU-GM), the hatched bars represent the number of burst forming colonies of the erythroid lineage (BFU-E), and the stippled bars represent the number of colonies that produced cells of the granulocytic, monocytic, erythroid, and megakaryocytic lineages (CFU-GEMM). (d) Transduced CD34⁺ cells were grown in bulk culture for 10 days and then analyzed by FACS for the total percentage with expression of eGFP (% eGFP⁺), for the total percentage expressing CD4 (% CD4⁺), and the percentage of the transduced cells that expressed CD4 (CD4⁺/eGFP⁺). CD34⁺ cells were mock transduced (black bar), transduced with CCL-ME (gray bar), or CCL-MhuMIE (white bar).

Full figure and legend (120K)

Potential cytotoxicity of the transgene huM10 was measured by examining the functionality and viability of the cells transduced with huM10 compared to cells that were mock transduced or transduced with the eGFP control gene. To validate the evaluation for cytotoxicity, long-term expression of the huM10 transgene in cells derived from transduced CD34⁺ cells was confirmed by Northern blot analysis. We chose Northern blot analysis to determine transgene expression at the RNA level, because we have not yet identified an antibody that can reliable detect the expression of M10 protein from either the CCL-MhuM or the CCL-MhuMIE vectors, or wild-type REV protein in HIV-1-infected cells. Northern blot analysis revealed high levels of huM10 transcripts produced in CCL-MhuM transduced CD34⁺ cells 15 days post-transduction. In two independent Northern blot analyses, there were smears of smaller-sized fragments that also hybridized to the huM10 probe, suggesting some instability of the transcript containing huM10 (that was not seen for the -actin transcript from the same sample) (Supplementary Figure S1). In contrast, the amount and stability of the CCL-MhuMIE transcript is similar to the amount and stability of the -actin transcript. These analyses demonstrate that there was expression of the huM10 gene in the samples analyzed that did not show evidence of cytotoxicity.

Functionality was evaluated by measuring the differentiation function of CD34⁺ using the CFU assay and by monitoring the emergence of CD4⁺ myelomonocytic cells in bulk culture. Analyses of CFU grown from transduced CD34⁺ cells derived from human bone marrow (Figure 3b) and umbilical cord blood (Figure 3c) revealed no difference in the number and type of colonies produced by mock-transduced CD34⁺ cells or by CD34⁺ cells transduced with either the vectors containing huM10 (CCL-MhuMIE and CCL-MhuM) or the vector with just eGFP (CCL-ME). Analyses of the level of CD4 expression in the bulk culture revealed that the total numbers of cells expressing CD4 were similar in bulk cultures derived from umbilical cord blood CD34⁺ that were either mock transduced or transduced with CCL-ME or CCL-MhuMIE (Figure 3d). Although the absolute numbers of transduced cells differed between the latter two cultures, of the transduced cells the number of cells expressing CD4 was similar, namely 53.1% for CCL-ME and 50.4% for CCL-MhuMIE cultures

Viability of the cultures was monitored by counting the cells using trypan blue exclusion and by measuring their metabolic activity. Metabolic activity was monitored by periodically measuring the level of ATP, which would only be present in metabolically active cells. No differences among the different cultures were detected (data not shown). Both the functionality and viability data indicate that the presence of the huM10 transgene did not impair the ability of CD34⁺ cells to proliferate and differentiate ex vivo.

In summary, lentiviral vectors carrying the huM10 gene were able to efficiently transduce CD34⁺ cells and did not interfere with the ability of the progenitor cells to differentiate ex vivo.

Anti-HIV activity of huM10 transduced CD34⁺ progeny cells

CD34⁺ cells were transduced with the CCL-MhuM vector or with the control CCL-ME vector and cultured under conditions that promote differentiation of the cells into CD4-expressing myelomonocytic cells.²⁶ The resultant cell pool is a heterogeneous mix of non-transduced and transduced cells of both the granulocytic and monocytic lineages. Such differentiated cells were challenged by infection with the monocytotropic strain of HIV-1, JR-FL using an multiplicity of infection (MOI) of 0.005, to determine the potential inhibitory effects of the huM10 gene on viral replication.

In a first experiment (Figure 4a) carried out with CD34⁺ cells isolated from bone marrow, HIV-1 replication peaked on day 15 post-infection. At that time point, viral growth was suppressed in duplicate cultures by 91.6 and 95.4%. In these cultures, the copy number of the huM10 gene was 1 and 0.97 copies per cell, respectively. Suppression of viral growth measured over the entire course of the experiment was between 91 and 99.2%.

Figure 4.

HIV-1 challenge of cultures derived from transduced human CD34⁺ cells. (a) HIV-1 viral replication was measured in cultures derived from transduced human bone marrow CD34⁺ cells a, (c), or (b) cord blood cells. The amount of HIV-1 p24 GAG protein produced by the culture transduced with the control vector CCL-ME (filled circle, solid line) was compared to cultures transduced with the vector CCL-MhuM (open symbols, hatched line).

Full figure and legend (38K)

In a second experiment (Figure 4b) carried out with CD34⁺ cells isolated from umbilical cord blood, HIV-1 replication peaked on day 19 post-infection. At that time point, viral growth was suppressed by 95.4 and 99.5%. In these cultures, the copy number of the huM10 gene was 2.3 and 2.5 copies per cell, respectively. Suppression of viral growth measured over the entire course of the experiment was between 89 and 99.8%.

A third HIV-1 challenge study was performed using 2- and 4-fold higher MOI of HIV-1 (0.01 and 0.02) to infect duplicate cultures derived from transduced bone marrow CD34⁺ cells (Figure 4c). HIV-1 replication peaked earlier in this experiment than in the ones that used a lower MOI of HIV-1 (day 10 vs days 15–19). Culture A showed 92.8–94.5% inhibition of HIV-1, and culture B showed 99.99% inhibition of HIV-1. These cultures contained an average of 0.2 and 0.3 vector copies/cell, respectively.

Thus at the range of MOI of HIV-1 JR-FL used, the CCL-MhuM vector potently inhibited HIV-1 replication in primary human cells derived from transduced CD34⁺ cells, without any steps taken to select for transduced cells.

Effects of huM10 on the mobilization of lentiviral vectors by HIV-1 co-infection

To measure the potential occurrence of mobilization of lentiviral vectors by co-infection with wild-type HIV-1, an experiment was designed to favor the event (Figure 5). Three pairs of vectors were analyzed, one with intact LTR (pHR') and two with "SIN" LTR (CCL and SMPU), with each pair consisting of one vector containing both eGFP and huM10 and one vector containing only eGFP (Table 1). It would be predicted that non-SIN vectors would be mobilized at a greater frequency than the SIN vectors, owing to their greater capacity to produce full-length (Psi-containing) transcripts from the 5' LTR, compared to SIN vectors with deletions of the LTR enhancers and promoters.²⁷

Figure 5.

Schematic flow chart of assay to detect lentiviral vector mobilization. 

Full figure and legend (47K)

In the primary cells, infection with HIV-1 IIIB at the high MOI of 0.1 resulted in vigorous viral replication, with 4–10 g/ml p24 produced in the non-transduced control cells as well as in the cells transduced with the SIN vectors lacking huM10 (CCL-MXIE and SMPU-MXIE) (Figure 6a). In contrast, on day 14 post-infection, cells transduced with the SIN vectors containing huM10 (CCL-MhuMIE and SMPU-MhuMIE) produced 189- and 108-fold less HIV-1, respectively.

Figure 6.

Assay to detect lentiviral vector mobilization. HIV-1 IIIB infection of (a) primary and (b) secondary CEM cells. Primary CEM cells were transduced with SIN vectors, CCL (circles), and SMPU (squares), and with a non-SIN vector, HR' (triangle). Each type of vector either carried the eGFP transgene alone (solid symbol, solid line) or the eGFP and huM10 transgene together (open symbol, hatched line). Transduced and naïve CEM (filled diamond, solid line) were infected with HIV-1 IIIB and analyzed for the numbers of viral particles produced in each culture by measuring the amount of extracellular p24. HIV-1 GAG protein (20 ng) produced after 22 days by the primary cells were used to infect the naïve secondary cells and serially collected sample of cell culture supernatant were analyzed for the amount of extracellular p24 GAG protein. Quantitative-PCR analysis of the secondary indicator CEM cells infected with 20 ng of (c) HIV-1 p24 GAG or with 5 ml of the (d) primary culture medium. Cell pellets of the secondary indicator CEM cells were harvested on day 22 of the experiment and analyzed with by quantitative-PCR for the eGFP and the huM10 sequences. The numbers of vector copies per cell (eGFP: gray bar, huM10: white bar) were calculated by using a standard consisting of a CEM clone that contains one copy of SMPU-MhuMIE per cell.

Full figure and legend (120K)

Interestingly, the non-SIN vector expressing only the eGFP gene (HR'-MXIE) was as active in suppressing viral replication as the non-SIN vector expressing the huM10 gene (HR'-MhuMIE). Suppression of HIV-1 replication by non-SIN lentiviral vectors lacking specific anti-HIV-1 genes has been reported,^{28, 29, 30} possibly, due to the TAR and RRE sequences contained in the pHR' backbone serving as decoys to bind TAT and REV, limiting their availability for the wild-type HIV-1. Another possibility that would account for the inhibition of HIV-1 replication in cells transduced by non-SIN vectors would be pseudotyping of vector transcripts resulting in HIV-1 virions lacking two copies of the HIV-1 genomic RNA.

The second step of the experiment was designed to detect and measure any of the vector sequences that may be mobilized by pseudotyping by the wild-type HIV-1. This was accomplished by infecting naïve CEM cells, serving as secondary indicator cells, with the cell-free viral supernatant harvested from the primary vector-transduced CEM cells that were infected with wild-type HIV-1.

In the first experiment, volumes of cell-free culture medium that contained an equivalent amount of HIV-1 p24 GAG (20 ng) were used to infect the secondary indicator cells, to account for the differences in growth of HIV-1 discussed above. The outgrowth of HIV-1 in the secondary cells was essentially identical in all cases (Figure 6b). However, because the cells transduced with the SIN vectors CCL and SMPU lacking huM10 supported viral replication so vigorously, the volume used to achieve 20 ng of HIV-1 p24 GAG would have contained a much smaller percentage of the total number of virions produced compared to the samples of the cells transduced with the non-SIN vectors. Therefore, to increase the sensitivity of the assay for the SIN vectors, in a second experiment, equal volumes (5 ml) of cell-free culture medium from the transduced and infected primary CEM cells were added to the secondary indicator cells.

In both experiments, huM10 sequences from either the CCL- or SMPU-based SIN vector were not detected in the secondary indicator cells (Figure 6c and d). In contrast, eGFP sequences from both the CCL or SMPU-based SIN vector were detectable, but only in the second experiment using 5 ml of viral supernatant, which in the case of the control SIN vectors CCL-MXIE and SMPU-MXIE increased the sensitivity of the assay 250-fold. Specifically, the frequency of eGFP sequences detected in this experiment was 5.8 10^-5 for the SMPU-MXIE vector and 5 10^-5 for the CCL-MhuMIE and 3 10^-4 for the SMPU-MhuMIE vector, respectively. As huM10 sequences were not detectable in either the CCL-MhuMIE or the SMPU-MhuMIE samples, it is possible that a rare vector was mobilized that had deleted the huM10 gene. However, as the values are near the limit of detection of the assay, it is also possible that the measured sequences represent contamination of eGFP sequences.

In contrast to the SIN vectors, significant mobilization was detected form both the non-SIN vectors, HR'-MXIE and HR'-MhuMIE (Figure 6c and d). In the experiment using 20 ng of input HIV-1 p24 GAG, the HR'-MXIE sample contained 2.7 10^-4 eGFP sequences, whereas in the experiment using 5 ml or 250-fold larger amounts of input virus, the HR'MXIE sample contained 6.2 10^-2 eGFP sequences. The measured increase in eGFP sequence frequency between the two experiments is 228-fold, corresponding to the 250-fold difference in sensitivity of the two assays. In contrast, the HR'-MhuMIE sample contained 6.3 10^-3 eGFP sequences and 9 10^-3 huM10 sequences, constituting a near 10-fold reduced frequency of mobilization when the huM10 gene was present (Figure 6d). In the experiment using the low 20 ng of input virus, the HR'-MhuMIE sample contained 3.8 10^-5 eGFP sequences in the absence of any detectable huM10 sequences. Again, we cannot distinguish between the possibility of a recombination event leading to a deletion of the huM10 gene or low level of eGFP DNA contamination.

These data indicate that non-SIN lentiviral vectors can be readily mobilized by co-infection with HIV-1, although the presence of huM10 in the vector lowers the extent of mobilization by at least 10-fold. SIN vectors containing only eGFP were 1,000-fold less likely to be mobilized than the comparable non-SIN vector. Mobilization of the SIN vectors containing the huM10 gene was only detected in the second experiment, where a larger amount of primary supernatant was evaluated and then only eGFP sequences but not huM10 sequences were detected (Figure 6c and d).

Discussion

As predicted, the presence of the huM10 gene in lentiviral vectors did lower the vector titers that could be produced. We found that huM10 decreased the production of HIV-1 GAG proteins from the packaging plasmid, presumably by inhibiting the action of REV needed for nuclear export of the HIV-1 transcripts. Examination of several HIV-1-packaging plasmids identified two that produced high levels of HIV-1 GAG proteins; although the levels of GAG were decreased by co-transfection with plasmids for the vectors carrying huM10, they still remained sufficiently high to produce adequate vector titers for large-scale production.

The CCL-huM10 vector produced by the optimal-packaging plasmid was capable of efficient transduction of human CD34⁺ cells. Under the conditions that were used, the percentages of cells that were transduced ranged from 41 to 57%, with a range of 0.2–2 copies per cell, without any efforts to select or enrich the transduced cells. These levels of transduction are typical of those achievable using lentiviral vectors for human CD34⁺ cells, which may translate to higher levels of gene-containing cells in vivo after transplantation than with retroviral vectors.³¹ Lentiviral vectors have been shown to be capable of achieving effective gene transfer to human CD34⁺ cells within 1–2 days of ex vivo culture and stimulation of the CD34⁺ cells, which may better preserve their stem cell function than the prolonged ex vivo culture and stimulation of 3–5 days needed for effective transduction by retroviral vectors.³²

Gratifyingly, we observed that expression of the huM10 gene in human CD34⁺ cells and their progeny was not toxic to the cells and did not impair their differentiation. Similar numbers of myeloid–erythroid colonies and mature CD4⁺ myeloid cells were produced from CD34⁺ cells expressing huM10 as from control-transduced cells. We have previously reported that CD34⁺ cells transduced with huM10 using a retroviral vector and transplanted into HIV-1-infected subjects led to production of myeloid and T cells with the huM10 gene.¹⁷ This lack of adverse effects on the stem/progenitor cells is important for the effective generation of mature T cells and other cells expressing huM10 after transplantation of transduced CD34⁺ cells.

The huM10 gene was shown to potently inhibit HIV-1 replication in vitro, as described previously with retroviral vectors carrying huM10.^{5, 6, 7, 8} This inhibition could be seen even in primary cell populations that had not been selected for expression of huM10 and that contained relatively low levels of integrated vectors per cell (e.g., 0.2–2 proviral copies/cell). The potency of any intracellular immunization approach in vivo depends on three factors: the intrinsic activity of the antiviral transgene or sequences, immune responses that may eliminate cells expressing the antiviral transgene and the emergence of viral resistance. Our data would suggest that the intrinsic activity of huM10 might be sufficient for inhibition of HIV-1 replication in vivo, although the relative potency of huM10 has not been compared to other anti-HIV-1 genes.

We developed an in vitro assay to measure mobilization of lentiviral vector transcripts by wild-type HIV-1, essentially a marker rescue assay. As has been previously described, vector mobilization of a lentiviral vector with intact LTR and carrying a reporter gene is readily detectable^{30, 33} and is reduced in the presence of revM10.³⁰ To validate our mobilization assay we included an intact HIV vector pair carrying either the eGFP reporter gene only or both the eGFP gene and the huM10 gene. We measured significant mobilization of the vector carrying eGFP only, which was reduced 10-fold when the same intact vector carried in addition the HIV-1 inhibitory gene huM10. In contrast to a previously described study,²⁹ we also detected low levels of mobilization for lentiviral vectors with a SIN LTR that had been deleted of the enhancers and promoter, when they carried the reporter gene eGFP, but not when they carried the huM10 gene. Vectors with the SIN LTR that carried the huM10 gene did not undergo mobilization, within the limits of detection of this in vitro assay. Presumably, the expression of huM10 inhibits HIV-1 replication to further limit spread of vector. This absence of mobilization of the lentiviral vector with huM10 adds to safety for clinical applications in HIV-1-infected subjects, by minimizing risks from vector spread to additional cells or beyond the treated subject.

It is difficult to extrapolate from the in vitro gene transfer levels and the extent of inhibition of HIV challenge shown by these studies to predict what may occur in the clinical gene therapy setting. The effects on HIV-1 infection would be functions of the percentage of transduced CD34⁺ cells infused, the extent of their engraftment and contribution to immune cell pools, the level of expression of huM10 in HIV target cells, and the relative MOI of HIV-1 faced by cells in compartments such as lymph nodes. It is not likely that sufficiently high percentages of blood and immune cells will be obtained with the suppressive gene to decrease production of HIV-1 (e.g., 90–99%). But, expression of huM10 may protect cells from HIV-induced cytopathicity and allow persistence of functional immune cells when other cells are being eliminated by HIV-1, and could augment other anti-retroviral approaches. Based on these findings, we plan to perform a clinical trial using the CCL-MND-huM10 to transduce CD34⁺ cells from HIV-1-infected subjects.

Materials and Methods

Lentiviral vectors and packaging plasmids. Lentiviral vectors were produced based on the pCCL²³ and SMPU³⁴ SIN vectors and the pHR' vector with intact LTR²⁵ to express either eGFP, huM10, or both under transcriptional control of the MND LTR U3 region³⁵ (Table 1). (See Supplementary Materials and Methods for details of vector construction).

The HIV-1-packaging plasmids p8.2¹⁹ and p8.9²⁰ were kind gifts of Inder Verma. pMDLg/pRRE²³ and pRSV-rev were kind gifts of Luigi Naldini (CellGenesys, Foster City CA). A modified version of the CD-NL/BH-packaging plasmid²¹ called pCD-NL/BH^*V 3 was provided by Jacob Reiser. The pSYNGP-packaging plasmid²⁴ was a kind gift of L Mithrophanous (University of Oxford, Oxford, UK). The pMD.G²³ plasmid expressing the VSV-G glycoprotein was used for all vector preparations. Viral vector supernatant was produced as described previously.³⁶

Cell lines. 293HEK (ATCC CRL-1573) and 293T (ATCC CRL-1268) were cultured in D10: Dulbecco's modified Eagle's medium (Mediatech, Herndon, VA) containing 10% fetal bovine serum (Omega Scientific, Tarzana, CA), 50 U/ml penicillin, 50 g/ml streptomycin, and 2 mM L-glutamine (Gemini Bioproducts, Woodland, CA). CEM cells (ATCC CCL-119) were cultured in R10: RPMI (Irvine Scientific, Santa Ana, CA) containing 10% fetal bovine serum 50 U/ml penicillin, 50 g/ml streptomycin, and 2mM L-glutamine.

CD34⁺ cell isolation and lentiviral transduction and monocytic differentiation. CD34⁺ cells were purified from normal human cord blood or from bone marrow specimens, as described.²⁶ Collection and use of the cord blood and bone marrow specimens were approved by the IRB at Childrens Hospital Los Angeles.

For the transduction, 1 10⁵ purified CD34⁺ cells were cultured in 48-well tissue culture plates (Becton Dickinson Labware, Franklin Lakes, NJ) coated with recombinant fibronectin fragment CH-296 (TaKara Shuzo Co., Kyoto, Japan) in transduction medium: X-Vivo 15 (Cambrex Bio Science Walkersville, Walkersville, MD) containing 50 ng/ml c-kit ligand (R&D Systems, Minneapolis, MN), 50 ng/ml flt-3 ligand (R&D Systems) and 50 ng/ml thrombopoietin (R&D Systems). The cells were prestimulated overnight in transduction medium followed by transduction for a 24 h period with lentiviral vector at a concentration of 1 10⁸ TU/ml. After transduction, CD34⁺ were cultured in recombinant fibronectin fragment CH-296 coated 12-well plates (Becton Dickinson Labware) and allowed to differentiate into myelomonocytic cells in BBMM: IMDM (Cambrex Bio Science Walkersville) containing 20% fetal bovine serum, 0.5% bovine serum albumin (Sigma, St Louis, MO), 10^-4 M -mercaptoethanol (Sigma), 10^-6 M hydrocortisone (Sigma), 50 U/ml penicillin, 50 g/ml streptomycin, 2 mM L-glutamine, 50 ng/ml IL3 (R&D Systems), 50 ng/ml IL6 (R&D systems), and 50 ng/ml c-kit ligand. Phosphate-buffered saline was used for all cell-washing procedures and all cells were cultured at 37°C in a 5% CO₂ humidified incubator.

Western blot. HEK 293T cells were transfected with 10 g of vector plasmid, 10 g of packaging plasmid, and 2 g of VSV-G envelope plasmid pMD-G using the CaPO₄ method (Invitrogen, Carlsbad, CA). Seventy-two hours after transfection the cells were pelleted and lysates were assayed for HIV-1 p24 GAG protein by Western blot analysis (See Supplementary Materials and Methods for details of Western blot analysis).

Colony forming assay. Between 7–10 days post-transduction of the CD34⁺ cells, the cells were counted and analyzed in the methylcellulose-supported colony-forming assay. Briefly, 5,000 and 10,000 cells, each in duplicate, were plated into 1 ml of Methocult (Stem Cell Technologies, Vancouver BC, Canada) supplemented with 20 U/ml erythropoietin (Amgen, Thousand Oaks CA). After 14 days, colonies containing at least 50 cells were counted under an inverted microscope.

CD4 staining. Aliquots of 1 10⁶ cells were removed from the bulk culture, pelleted and incubated with 10 l IgG1 pure, and 10 l CD4-PE, or 10 l IgG1-PE (Becton-Dickinson Immunocytometry Systems, San Jose, CA) in a total volume of 100 l phosphate-buffered saline for 30 min on ice. After a wash step, the cells were re-suspended in 300 l phosphate-buffered saline and analyzed by FACS (FACSvantage, Becton-Dickinson Immunocytometry Systems).

Viability assay. Aliquots (100 l) of cultured cells were removed from the bulk culture and analyzed on white 96-well plates (Corning, New York, NY) for the amount of ATP present using the CellTiter-Glo^® Luminescent Cell Viability Assay (Promega, Madison, WI) according to the manufacturer's instructions.

HIV-1 JR-FL challenge assay of myelomonocytic cells derived from transduced CD34⁺ cells. The viral stocks of HIV-1 JR-FL were produced by propagation on primary human peripheral blood mononuclear cells. The TCID₅₀ titers of the viral stock were determined on peripheral blood mononuclear cells. Between 7 and 9 days post-transduction, the cultures derived from the CD34⁺ cells were infected by the monocytotropic strain of HIV-1 JR-FL at an MOI of 0.005. The MOI was calculated based on the total number of cells exposed to virus. Typically, 1 10⁶ cells were incubated with 50 l of a 1 10⁵ TCID₅₀/ml HIV stock in a final volume of 200 l BBMM. The cells were cultured for a period of 3–4 weeks in BBMM and re-fed every 2–4 days. Aliquots of cell culture supernatant, collected at each medium change, were stored at -80°C and analyzed for the amount of HIV-1 p24 by enzyme-linked immunosorbent assay (HIV-1 p24 antigen assay, Coulter, Miami, FL) according to the manufacturer's instruction.

Transgene copy number analyses. At 3–4 weeks post-transduction, aliquots of CD34⁺ derived cells were collected for vector copy number analysis using quantitative PCR. DNA was extracted using the Quiagen Dneasy Tissue Kit (Quiagen, Valencia, CA) and assayed for the level of huM10 vector sequences using real-time PCR as described.¹⁷

Transgene expression analysis by Northern blot. Aliquots of cells derived from CD34⁺ cells were analyzed for the presence of huM10 transcripts 15 days post-transduction. (See Supplementary Materials and Methods for details of Northern blot analysis).

Mobilization assay. Primary cells were CEM human T cells transduced with the different lentiviral vectors and sorted by FACS for eGFP-expression to obtain populations of cells that contained integrated lentiviral vector proviruses. 1 10⁶ CEM cells were transduced with lentiviral vectors at an MOI of 20 using spinoculation (3,000 g 2 h) to achieve high levels of vector provirus per cell. Vectors used include the SIN vectors CCL-MXIE, CCL-MhuMIE, SMPU-MXIE, and SMPU-MhuMIE and the non-SIN vectors HR'-MXIE and HR'-MhuMIE. At 10 days post-transduction, the eGFP-positive cells were obtained by FACS on a FACSVantage flow cytometer (Becton-Dickinson Immunocytometry Systems) and using LysysII software (Becton-Dickinson Immunocytometry Systems).

Transduced primary CEM cells (1 10⁶) were incubated with HIV-IIIB at an MOI of 0.1 for 2 h. After the incubation, the cells were washed with R10 and cultured with serial passage at a cell density of between 1 10⁵ and 1 10⁶ cells/ml for a period of three weeks. During this period the cells were split twice a week and aliquots of cell culture supernatant were collected each time and stored at -80°C for later HIV-1 p24 measurement. Cell-free virus-containing cell culture supernatant from the last time point of culture of the transduced primary cells that would contain HIV-1 and any mobilized vector was applied to non-transduced secondary CEM indicator cells, normalized by either using the volume of supernatant that contained 20 ng of HIV-1 GAG p24 or by using 5 ml volume of each of the supernatants. Infection of these secondary non-transduced CEM cells was carried out by the identical method to that described for the primary transduced CEM cells.

Statistical analysis. The primary method of statistical analysis was multivariate linear regression analysis. Indicator variables were included for the day of the assay in order to account for a possible assay day effect. The analysis was conducted both for the original data as well as with the log transformation. As the analysis with the non-transformed data fit better that analysis was decided upon, although the results were similar in both analysis.

The initial analysis included as fixed factors all five groups (group1=pMDgagpol, group2=pCDNL, group3=pSYNgagpol, group4=del8.2, and group5=del8.9) and two types (type1=pCCLME and type 2=pCCLMhuMIE), as well as day effect indicator variables and possible interactions. Interaction terms were added to a complete main effects model as necessary. Subsequently nonsignificant terms were deleted in order to obtain a more parsimonious model. The resulting final model had an R² of 91.8%.

References

1. Malim, MH, Hauber, J, Lee, SY, Maizel, JV and Cullen, BR (1989). The HIV-1 rev trans-activator acts through structured target sequence to activate nuclear export of unspliced viral mRNA. Nature 338: 254–257. | Article | PubMed | ISI | ChemPort |
2. Chang, DD and Sharp, PA (1990). Messenger RNA transport and HIV rev regulation. Science 249: 614–515.
3. Hope, TJ (1999). The ins and outs of HIV rev. Arch Biochem Biophys 365: 186–191. | Article | PubMed | ChemPort |
4. Cullen, BR (1991). Regulation of HIV-1 gene expression. FASEB J 5: 2361–2368. | PubMed | ISI | ChemPort |
5. Bevec, D, Dobrovnik, M, Hauber, J and Böhnlein, E (1992). Inhibition of human immunodeficiency virus type 1 replication in human T cells by retroviral-mediated gene transfer of a dominant-negative Rev transactivator. Proc Natl Acad Sci 89: 9870–9874. | Article | PubMed | ChemPort |
6. Malim, MH et al. (1992). Stable expression of transdominant rev protein in human T cells inhibits human immunodeficiency virus replication. J Exp Med 176: 1197–1201. | Article | PubMed | ISI | ChemPort |
7. Bahner, I, Zhou, C, Yu, X-J, Hao, Q-L, Guatelli, JC and Kohn, DB (1993). Comparison of trans -dominant inhibitory mutant human immunodeficiency virus type 1 genes expressed by retroviral vectors in human t lymphocytes. J Virol 67: 3199–3207. | PubMed | ChemPort |
8. Bauer, G et al. (1997). Inhibition of HIV-1 replication after transduction of G-CSF mobilized CD34(+) cells from HIV-1-infected donors using retroviral vectors containing anti-HIV-1 genes. Blood 89: 2259–2267. | PubMed | ISI | ChemPort |
9. Fischer, U, Huber, J, Boelens, WC, Mattai, IW and Lührmann, R (1995). The HIV-1 Rev activation domain is a nuclear export signal that accesses an export pathway used by specific cellular RNAs. Cell 82: 475–483. | Article | PubMed | ISI | ChemPort |
10. Fritz, CC, Zapp, ML and Green, MR (1995). A human nucleoporin-like protein that specifically interacts with HIV Rev. Nature 376: 530–533. | Article | PubMed | ISI | ChemPort |
11. Yi, R, Bogerd, HP and Cullen, BR (2002). Recruitment of the Crm1 nuclear export factor is sufficient to induce cytoplasmic expression of incompletely spliced human immunodeficiency virus mRNA. J Virol 76: 2036–2042. | Article | PubMed | ChemPort |
12. Malim, MH and Cullen, BR (1991). HIV-1 structural gene expression requires the binding of multiple Rev monomers to the viral RRE-implications for HIV-1 latency. Cell 65: 241–248. | Article | PubMed | ISI | ChemPort |
13. Hamm, TE, Rekosh, D and Hammarskjöld, M-L (1999). Selection and characterization of human immunodeficiency virus type 1 mutants that are resistant to inhibition by the transdominant negative RevM10 protein. J Virol 73: 5741–5747. | PubMed | ChemPort |
14. Saurya, S, Lichtenstein, Z and Karpas, A (2005). Defective rev response element (RRE) and rev gene in HAART treated AIDS patients with discordance between viral load and CD4(+) T-cell counts. J Clin Virol 33: 324–327. | Article | PubMed | ChemPort |
15. Woffendin, C, Rang, U, Yang, Z-Y, Xu, L and Nabel, G (1996). Expression of a protective gene prolongs survival of T cells in human immunodeficiency virus-infected patients. Proc Natl Acad Sci USA 93: 2889–2894. | Article | PubMed | ChemPort |
16. Ranga, U et al. (1998). Enhanced T cell engraftment after retroviral delivery of an antiviral gene in HIV-infected individuals. Proc Natl Acad Sci USA 95: 1201–1206. | Article | PubMed | ChemPort |
17. Podsakoff, GM et al. (2005). Selective survival of peripheral blood lymphocytes in children with HIV-1 following delivery of an anti-HIV gene to bone marrow CD34(+) cells. Mol Ther 12: 77–86. | Article | PubMed | ISI | ChemPort |
18. Uchida, N et al. (1998). HIV, but not murine leukemia virus, vectors mediate high efficiency gene transfer into freshly isolated GO/G1 human hematopoietic stem cells. Proc Natl Acad Sci USA 95: 11939–11944. | Article | PubMed | ChemPort |
19. Naldini, L, Blömer, U, Gage, FH, Trono, D and Verma, IM (1996b). Efficient transfer, integration and sustained long-term expression of the transgene in adult rat brains injected with a lentiviral vector. Proc Natl Acad Sci USA 93: 11382–11388. | Article | PubMed | ChemPort |
20. Zufferey, R, Nagy, D, Mandel, RJ, Naldini, L and Trono, D (1997). Multiply attenuated lentiviral vector achieves efficient gene delivery in vivo. Nat Biotechnol 15: 871–875. | Article | PubMed | ISI | ChemPort |
21. Mochizuki, H, Scwartz, JP, Tanaka, K, Brady, RO and Reiser, J (1998). High-titer human immunodeficiency virus type-1-based vector systems for gene delivery into nondividing cells. J Virol 72: 8873–8883. | PubMed | ISI | ChemPort |
22. Cui, Y, Iwakuma, T and Chang, LJ (1999). Contributions of viral splice site and cis-regulatory elements to lentivirus vector function. J Virol 73: 6171–6176. | PubMed | ISI | ChemPort |
23. Dull, T et al. (1998). A third generation lentivirus vector with a conditional packaging system. J Virol 72: 8463–8471. | PubMed | ISI | ChemPort |
24. Kotsopoulou, E, Kim, VN, Kingsman, AJ, Kingsman, SM and Mitrophanous, KA (2000). A Rev-independent human immunodeficiency virus type 1 (HIV-1)-based vector that exploits a codon-optimized hiv-1 gag-pol gene. J Virol 74: 4839–4852. | Article | PubMed | ISI | ChemPort |
25. Naldini, L et al. (1996). In vivo gene delivery and stable transduction of non-dividing cells by a lentiviral vector. Science 272: 263–267. | Article | PubMed | ISI | ChemPort |
26. Bahner, I, Kearns, K, Hao, Q-L, Smogorzewska, EM and Kohn, DB (1996). Transduction of human CD34+ hematopoietic progenitor cells by a retroviral vector expressing an RRE decoy inhibits human myelomonocytic cells produced in long-term culture. J Virol 70: 4352–4360. | PubMed | ChemPort |
27. Logan, AC, Haas, DL, Kafr, T and Kohn, DB (2004). Integrated self-inactivating lentiviral vectors produce full-length genomic transcripts competent for encapsidation and integration. J Virol 78: 8421–8436. | Article | PubMed | ChemPort |
28. Corbeau, P and Wong-Staal, F (1998). Anti-HIV effects of HIV vectors. Virol 243: 268–274. | ChemPort |
29. Bukovsky, AA, Song, J-P and Naldini, L (1999). Interaction of human immunodeficiency virus-derived vectors with wild-type virus in transduced cells. J Virol 73: 7087–7092. | PubMed | ISI | ChemPort |
30. Klimatcheva, E, Planelles, V, Day, SL, Fulreader, F, Renda, M and Rosenblatt, J (2001). Defective lentiviral vectors are efficiently trafficked by HIV-1 and inhibit viral replication. Mol Ther 3: 928–939. | Article | PubMed | ChemPort |
31. Case, SS et al. (1999). Stable transduction of quiescent CD34+CD38-human hematopoietic cells by HIV-1 based lentiviral vectors. Proc Natl Acad Sci 96: 2988–2993. | Article | PubMed | ChemPort |
32. Mazurier, F, Gan, OI, McKenzie, JL, Doedens, M and Dick, JE (2004). Lentivector-mediated clonal tracking reveals intrinsic heterogeneity in the human hematopoietic stem cell compartment and culture-induced stem cell impairment. Blood 103: 545–552. | Article | PubMed | ISI | ChemPort |
33. Evans, JT and Garcia, V (2000). Lentivirus vector mobilization and spread by human immunodeficiency virus. Hu Gene Ther 11: 2331–2339. | ChemPort |
34. Wang, X et al. (2003). Dynamic tracking of human hematopoietic stem cell engraftment using in vivo bioluminescense imaging. Blood 102: 3478–3482. | Article | PubMed | ChemPort |
35. Challita, PM, Skelton, D, El-Khoueiry, A, Yu, XJ, Weinberg, KI and Kohn, DB (1995). Multiple modifications in cis-elements of the LTR of retroviral vectors lead to increased expression and decreased DNA methylation in embryonic carcinoma (EC) cells. J Virol 69: 748–755. | PubMed | ChemPort |
36. Logan, AC, Nightingale, SJ, Haas, DL, Cho, GJ, Pepper, KA and Kohn, DB (200). Factors influencing the titer and infectivity of lentiviral vectors. Hum Gene Ther 15: 976–988.

Acknowledgements

These studies were supported by a research Grant from the National Institutes of Health (1 RO1 AI52798-01) and by a Distinguished Clinical Scientists Award from the Doris Duke Charitable Foundation to DBK We thank the obstetric nurses of the Delivery Rooms at Kaiser-Permanente Hospital, Sunset Blvd., Los Angeles and at Queen of Angeles Hollywood Presbyterian Hospital, Vermont Ave., Los Angeles for the diligent collection of umbilical cord blood samples without which these studies would not have been possible.

Supplementary Material

Figure S1. Presence of huM10 transcripts in cells derived from transduced CD34⁺ cells.

Materials and Methods.