Introduction

Low levels of genetically modified cells can prevail in vivo if the modified cells have a selective advantage, as shown in the case of gene therapy treatment of severe combined immunodeficiency^1,2. For disorders in which a naturally occurring selective advantage does not exist, enforced in vivo selection is one approach to increase the fraction of genetically modified hematopoietic cells. One strategy is to use a drug-resistant gene to enrich the transduced cells. Examples include variants of dihydrofolate reductase (DHFR)^3,4, multiple drug resistance protein 1 (MDR1)⁵, and variants of DNA alkyltransferase such as methylguanine methyltransferase (MGMT)^6,7,8,9,10. However, adult cancer patients receiving MDR1-transduced autologous CD34⁺ cells followed by etoposide treatment showed no enrichment for the transduced cells¹¹. DHFR variants confer resistance to methotrexate and trimetrexate (TMTX), but these drugs do not exert selective pressure in the stem cell compartment. Although this limitation can be overcome by combined treatment with TMTX and the nucleoside transport inhibitor nitrobenzylmercaptopurine riboside 5'-monophosphate³, in vivo selection in a large animal model achieved only transient selection¹². MGMT protects cells by removing the DNA adducts induced by nitrosourea, such as bis-chloroethyl nitrosourea (BCNU)¹³. O⁶-Benzylguanine (BG) is frequently used in combination with BCNU since BG acts by functionally inactivating endogenous MGMT. To facilitate the selection strategy, MGMT variants that contain specific amino acid substitutions conferring BG resistance were used¹⁴. In murine serial transplantation studies, drug treatment of animals transplanted with DHFR- or MGMT-transduced cells demonstrated that selection occurred at the level of hematopoietic progenitor cells (HPCs)¹³. However, selection with either system was also accompanied by a significant reduction in HPC numbers¹⁵. This raises safety concerns as it could lead to hematopoietic failure. Due to the shortcomings of these approaches, alternative selection strategies need to be developed.

We have evaluated a potential lymphocyte selection scheme based on the expression of a variant of inosine monophosphate dehydrogenase 2 (IMPDH2), the rate-limiting enzyme in the de novo purine biosynthesis pathway¹⁶. IMPDH catalyzes the conversion of inosine monophosphate (IMP) to xanthosine monophosphate¹⁶. Humans have two IMPDH isoenzymes, termed type 1 and type 2, that are 84% identical at the amino acid level and possess indistinguishable catalytic activities. Since proliferating lymphocytes depend more than other cell types on de novo synthesis of purines, IMPDH inhibitors such as mycophenolic acid (MPA) and its prodrug mycophenolate mofetil (MMF) can selectively inhibit lymphocyte proliferation and confer immunosuppression¹⁶. In addition to the extensive clinical experience with MMF in organ and bone marrow transplantation, MMF has been used to supplement anti-HIV-1 chemotherapy because it has a synergistic interaction with nucleoside reverse transcriptase inhibitors^17,18,19. The IMPDH2 variant we use contains two mutations in the MPA binding site that lead to a significantly increased resistance to MPA. In this study, we used HIV vectors to deliver a small interfering RNA (siRNA) gene targeted to HIV-1 together with this IMPDH2 variant into T lymphocytes, peripheral blood mononuclear cells (PBMCs), and CD34⁺ cells; optimized the conditions for MPA selection; and characterized the phenotypes of the transduced cells, including their ability to secret cytokines upon stimulation and the ability of the selected cells to inhibit HIV-1 replication. Our results suggest that this system may be applied as a selection strategy to enrich transduced T cells in vivo.

Results

IMPDH2 expression conferred MPA resistance to T lymphocytes

To facilitate detection of the transduced T cells under MPA selection, we fused the green fluorescent protein (GFP) gene with the IMPDH2 variant to generate IMPDH2-GFP (I-GFP). The gene encoding I-GFP was controlled by the promoter derived from spleen focus-forming virus (SFFV), and we cloned it into a HIV vector backbone to generate pHIV7/I-GFP (Fig. 1A). We also constructed a second vector, pHIV7/p24/I-GFP, containing an H1 promoter-driven short hairpin (sh) RNA gene specific for the HIV gag transcript. We generated these vectors from 293T cells and used them to transduce human T lymphocytes in this study. To serve as a control, we generated and used the previously described HIV7/SF-GFP²⁰ containing the GFP gene in this study. To determine if MPA resistance could be demonstrated, we transduced both cell types with the vectors and cultured them in MPA at varying concentrations. After an 8-day selection in 3 M MPA, the GFP⁺ fraction of HIV7/I-GFP-transduced H9 cells increased from 81 to 99% and the GFP⁺ fraction of HIV7/p24/I-GFP-transduced cells increased from 45 to 94% (Fig. 1B). Serial passaging of MPA-selected cells for up to 2 months retained the same percentage of GFP⁺ cells in each population (data not shown), suggesting that stable expression of the IMPDH2–GFP fusion protein did not affect cell proliferation. Similarly, when PBMC-derived T cells were analyzed by fluorescence-activated cell sorting (FACS), we noted that selection in 0.4 M MPA led to an increase in the GFP⁺ cell fraction from 14 to 51% in HIV7/I-GFP-transduced cells and from 26 to 60% in HIV7/p24/I-GFP-transduced cells (Fig. 1C).

Figure 1.

MPA selection of T cells transduced with the HIV vector expressing the IMPDH2 variant. (A) The structures of the HIV vectors containing the I-GFP fusion gene (hatched box) and the p24 shRNA gene (solid box) with the direction of transcription indicated by the arrows. (B) Enrichment of the transduced H9 cells by MPA selection. H9 cells were transduced with either vector shown in (A), followed by selection in 3 M MPA for 8 days. The fraction of GFP⁺ cells was determined by FACS analysis. (C) Enrichment of the transduced PBMCs by MPA selection. Similar to H9 cells, transduced PBMCs were selected in 0.4 M MPA for 8 days and the fraction of GFP⁺ cells was determined by FACS.

Full figure and legend (140K)

Escalation of the MPA concentration from 3.1 to 24.8 M had only minimal effect on the viability of HIV7/I-GFP-transduced and HIV7/p24/I-GFP-transduced H9 cells (Fig. 2A). In contrast, mock-transduced or HIV7/SF-GFP-transduced H9 cells were eliminated by MPA at a concentration of 6.2 M. PBMCs were more sensitive than H9 cells to the selection, and 0.4 M MPA was sufficient to eliminate mock-transduced or control vector-transduced cells (Fig. 2B). The HIV7/I-GFP-transduced and HIV7/p24/I-GFP-transduced PBMCs were resistant to an MPA concentration as high as 3.1 M (Fig. 2B). MPA-resistant PBMCs expanded 251-fold during a period of 10 days in selection, and more than 98% of these cells expressed CD3 (data not shown), indicating that T cells were selected by this procedure. These studies demonstrated that IMPDH2 could serve as an effective drug selection marker to enrich for transduced T cells in culture. In addition, a wide range of MPA concentrations could be used for such a selection process without significantly affecting cell proliferation or viability (data not shown).

Figure 2.

Dose–response study with MPA. (A) Selection of transduced H9 cells with increasing mycophenolic acid (MPA) concentrations. (B) Selection of transduced PBMCs with increasing MPA concentrations. Percentage viable cells is shown as determined by trypan blue exclusion after 10-day selection in the MPA concentration for each vector-transduced cell, as indicated.

Full figure and legend (99K)

Cytokine production from MPA-selected T lymphocytes

To ensure that the MPA selection did not interfere with normal T cell activities, we measured cytokine production from MPA-selected PBMCs upon stimulation with antibodies against CD3 and CD28. For six cytokines tested, mock- and vector-transduced cell populations responded to stimulation with increased cytokine production (Fig. 3). However, the cytokine responses were generally lower in cells transduced with HIV7/p24/I-GFP. This could be attributed to potential "off-target" effects of stable siRNA expression in PBMCs²¹. However, given the variability of the assay in healthy subjects²², this quantitative difference could be due to the small sample size (four patients) we used in this assay. Since PBMCs used for this study were transduced and expanded in culture for at least 1 month in the continuous presence of MPA selection, these results demonstrated that cytokine production from these PBMCs was not inhibited by the selection procedure or by the stable expression of either the I-GFP fusion protein or the p24 siRNA.

Figure 3.

Cytokine production profile from MPA-selected PBMCs. The cytokine levels were measured by cytometric bead array (BD Biosciences, San Jose, CA, USA). MPA-selected PBMCs were washed and stimulated with antibodies to CD3 and CD28. The medium was collected prestimulation (white bars) and 24 h poststimulation (black bars).

Full figure and legend (109K)

I-GFP and shRNA expression did not interfere with the ability of CD34⁺ cells to differentiate in culture

HPCs represent the major target for the delivery of anti-HIV RNAi genes. To ensure that p24 shRNA or I-GFP fusion protein expression does not interfere with HPC differentiation, we transduced cord blood CD34⁺ cells with the three vectors shown in Fig. 1A and plated them in methylcellulose. We scored colony formation and GFP expression after 2 weeks. As shown in Table 1, colonies derived from BFU-E, CFU-GM, and CFU-GEMM were readily detectable. The fractions of GFP⁺ colonies in HIV7/I-GFP- and HIV7/p24/I-GFP-transduced cells were 25 and 43%, respectively. As the fraction of GFP⁺ colonies in HIV7/p24/I-GFP-transduced cells was similar to that in HIV7/SF-GFP-transduced cells, we concluded that I-GFP and shRNA expression did not interfere with hematopoietic progenitor cell differentiation in this culture system.

Table 1 - Methylcellulose colony formation from transduced CD34⁺ cells.

Full table

Transduced CD34 progenitor cells can be differentiated into myelomonocytic cell lineages in liquid culture. As shown in Fig. 4A, we detected a significant fraction of CD14⁺/GFP⁺ cells in liquid culture after a 3-week incubation of the transduced CD34⁺ cells in the presence of interleukin-3 (IL-3), IL-6, and stem cell factor (SCF). HIV7/SF-GFP-transduced cells had the highest GFP mean fluorescence intensity. This is consistent with our observation that HIV7/SF-GFP gave the highest mean fluorescence intensity in the HT1080 fibroblasts we used for vector titering (data not shown). The lower GFP mean fluorescence intensity derived from I-GFP could be due to protein fusion that reduced GFP fluorescence. Myelomonocytic differentiation of the transduced cells as evaluated for CD14 expression was not inhibited (41% for HIV7/SF-GFP, 61% for HIV7/I-GFP, 53% for HIV7/p24/I-GFP). In comparison, 36% of mock-transduced cells expressed CD14 (Fig. 4A). We also pooled myeloid colonies from the methylcellulose and placed them in cytokine-containing medium to differentiate these cells further into mature macrophages. As shown in Fig. 4B, both nontransduced (nonshaded) and vector-transduced (shaded) cells expressed similar levels of CD14, HLA-DR, CD4, CCR5, and CXCR4 on their surfaces. Thus, normal macrophage differentiation could occur in the presence of I-GFP or shRNA expression.

Figure 4.

Differentiation of transduced CD34⁺ cells in culture. CD34⁺ cells were transduced with the indicated vector and allowed to differentiate into myelomonocytic or erythroid cells in liquid culture for 3 weeks. The differentiated cells were then analyzed by GFP and lineage marker expression. (A) CD14 staining for myelomonocytic cell differentiation. (B) Staining of macrophages derived from transduced CD34⁺ cells for cell surface markers including CD14, HLA-DR, CD4, CCR5, and CXCR4. The shaded area denotes transduced cells and the nonshaded area denotes nontransduced cells. (C) Glycophorin A staining for erythroid cell differentiation.

Full figure and legend (234K)

As a measure of erythroid differentiation, we held the transduced CD34⁺ cells in liquid culture for 3 weeks in the presence of SCF and erythropoietin, followed by FACS analysis for glycophorin A expression. As shown in Fig. 4C, we observed a significant fraction of glycophorin A⁺/GFP⁺ erythroid cells. Similar to myelomonocytic cells, the fractions of the transduced erythroid cells with different vectors were similar (75% for HIV7/SF-GFP, 79% for HIV7/I-GFP, 62% for HIV7/p24/I-GFP, 75% for mock control). In addition, we examined Wright-stained slide preparations microscopically and observed the presence of reticulocyte and hemoglobinized erythrocytes (data not shown). These results demonstrate that expression of I-GFP and the p24 shRNA did not interfere with the ability of cord blood CD34⁺ cells to differentiate in culture.

Up-regulation of B7.1 expression and phagocytosis of macrophages derived from transduced CD34⁺ cells

Macrophages normally express low levels of B7 costimulatory molecules that present antigen to memory and effector T cells. However, upon activation with various stimuli, B7 in macrophages is up-regulated on the cell surface²³. To determine if the macrophages derived from the transduced CD34⁺ cells were functionally normal in up-regulating the B7 costimulatory molecule, we stimulated the cells with lipopolysaccharide (LPS) and analyzed them for B7.1 expression by FACS. As shown in Fig. 5, irrespective of the vector used, B7.1 expression was up-regulated in the transduced cells to levels similar to those in untransduced cells.

Figure 5.

B7.1 up-regulation as a measure of stimulatory response. Mock-transduced and vector-transduced macrophages derived from CD34⁺ cells were stimulated by adding 5 g/ml lipopolysaccharides (LPS) to the culture. After 24 h, the cells were stained with PE–Cy5-conjugated anti-B7.1 and analyzed by FACS.

Full figure and legend (153K)

Integral to antigen presentation is the ability of macrophages to phagocytose foreign material²⁴. To determine whether macrophages derived from the transduced CD34⁺ cells are capable of normal phagocytosis, we mixed tetramethylrhodamine-conjugated Escherichia coli with the transduced cells and analyzed the ingested bacteria by FACS. As shown in Fig. 6, phagocytosis by the transduced macrophages was normal compared with the macrophages derived from mock-transduced CD34⁺ cells. HeLa-derived Magi-CXCR4 cells²⁵ served as nonphagocytic controls. These results suggest that expression of the p24 shRNA and I-GFP did not interfere with HPC differentiation and normal functions of the differentiated progenies in culture.

Figure 6.

The phagocytosis assay for differentiated macrophages. Both LPS and tetramethylrhodamine-conjugated E. coli (K12) Bioparticles (Invitrogen, Carlsbad, CA, USA) were added to the cell culture after transduction and differentiation as described in the text. Two hours after stimulation, the CD14⁺ cells were analyzed by FACS for GFP-induced fluorescence or for tetramethylrhodamine-induced fluorescence, detectable in the FL2 channel as phycoerythrin (PE). Negative control cells are Magi-CXCR4 cells transduced with HIV7/SF-GFP, which serve as nonphagocytic cells treated similarly, and mock-transduced macrophages without E. coli Bioparticle stimulation.

Full figure and legend (109K)

Inhibition of HIV-1 replication by p24 shRNA in MPA-selected T cells

To determine whether p24 shRNA expression in MPA-selected cells can confer resistance to HIV-1 replication, we challenged MPA-resistant H9 cells and PBMCs transduced with either HIV7/I-GFP or HIV7/p24/I-GFP with HIV-1_NL4–3. As shown in Fig. 7A, HIV7/p24/I-GFP transduced H9 cells inhibited HIV-1 replication by at least 5 orders of magnitude compared with HIV7/I-GFP or mock-transduced cells. With the same challenge dose as H9 cells, MPA-selected PBMCs transduced with HIV7/p24/I-GFP also demonstrated a similar ability to inhibit HIV-1 replication compared with mock- or HIV7/I-GFP-transduced cells (Fig. 7B). The emergence of the resistant HIV-1 4 weeks after challenge was due to the selection of HIV-1 mutants with nucleotide changes in the siRNA target site (data not shown). Together, these results suggested that the MPA selection process did not interfere with the ability of T cells either to inhibit or to support HIV-1 replication, depending on whether an anti-HIV-1 shRNA was expressed.

Figure 7.

Inhibition of HIV-1 replication in MPA-selected T cells expressing p24 shRNA. (A) Transduced H9 cells were challenged with HIV-1_NL4–3 at an m.o.i. of 0.02. The production of HIV-1 p24 in the culture supernatant was monitored by p24 ELISA. HIV7/SF-GFP-transduced cells were grown in the absence of MPA before the challenge experiment, whereas HIV7/I-GFP- and HIV7/p24/I-GFP-transduced cells were selected in 3 M MPA for 2 weeks. The MPA was removed from the culture medium 5 days before the challenge experiment. (B) Transduced PBMCs were challenged with the same virus as described in (A). The conditions for challenge were similar to those of (A) except that the MPA concentration used for cell selection was 0.4 M.

Full figure and legend (86K)

Discussion

One challenge of gene therapy is how to transduce a sufficient number of the target cells to produce a therapeutic effect. In vivo drug selection addresses this problem and potentially allows a low fraction of the transduced cells to be expanded. This is most clearly demonstrated by the delivery of mutant MGMT cDNA into mouse and canine models^6,26. In both models, cytotoxic drug treatment produced a significant and sustained multilineage increase in genetically modified HPCs. However, the drug toxicities in bone marrow, liver, and lung and a depletion of overall HPC number induced by the MGMT selection scheme suggest that alternative approaches to enrich the transduced cells in vivo should be explored.

In the current study, we have evaluated the use of IMPDH2 as a way to enrich directly for the transduced T lymphocytes. The selection scheme based on this enzyme used to treat HIV infection provides at least two advantages: First, it uses MPA and its prodrug MMF, a clinically approved immunosuppressive drug, for the enrichment of the transduced cells. This drug specifically suppresses lymphocyte proliferation but has less effect on other hematopoietic cell lineages. We confirmed this property by showing that MPA had little effect on the survival of methylcellulose colonies consisting mostly of erythroid, granulocytic, and monocytic cell lineages (data not shown). Second, MMF is an antiviral agent and has a synergistic effect when combined with other nucleoside analog inhibitors of reverse transcriptase to suppress HIV replication. Our study showed that IMPDH2-expressing T cells survived a wide range of MPA selection with little effect on cell proliferation. MPA-resistant cells continued to respond to T cell stimulating signals and produce both TH1 and TH2 types of cytokines. These studies showed that HIV vector-mediated transfer of the mutant IMPDH2 cDNA could protect T lymphocytes in culture from the toxic effects of MPA and thus provide a basis to use this system for direct cell expansion in vivo.

The two amino acid changes that we introduced into IMPDH2 were based on a study of selected mouse neuroblastoma cells having increased resistance to MPA²⁷. The murine IMPDH2 enzyme isolated from one such resistant cell line containing an alteration of threonine-333 to isoleucine and serine-351 to tyrosine exhibited a 2400-fold increased k_i for MPA. As the crystal structures of IMPDH2 in complex with IMP and MPA were solved²⁸, it showed that the hydroxyl group of Thr-333 formed hydrogen bonds with the C1 carbonyl oxygen in MPA, whereas Ser-351 did not contact MPA. A single point mutation in Thr-333 in IMPDH2 led to a 300-fold increase in k_i for MPA, while the role that Ser-351 plays in MPA resistance remains unclear. To design a mutant that has the least affinity for MPA, and that also allows for more efficient selection, we chose to mutate the amino acid at both positions. Importantly, the T333I IMPDH2 mutant still retained at least 30% of the catalytic activity²⁸. Since Ser-351 is not involved in the interaction with the IMPDH2 enzymatic active site²⁸, the mutation is not expected to alter the enzymatic activity significantly. The ability of the mutated IMPDH2 to induce MPA resistance in the transduced lymphocytes is consistent with this prediction. The fact that the transduced cells survived a wide range of MPA also reflected the extent of the reduced affinity of this mutant enzyme toward MPA.

It has been shown before that primary lymphocytes could be stimulated to proliferate with anti-CD3 and anti-CD28 antibodies coimmobilized on beads²⁹; it remained unclear whether stable IMPDH2/shRNA expression or MPA selective pressure could alter the ability of these cells to proliferate. A previous study had demonstrated that transfection of an siRNA directed to HIV gag effectively inhibited HIV-1 replication in cell culture³⁰. We used an shRNA form of this same RNAi since the hairpin structure appears to have improved expression (data not shown). Of note, even in the presence of an otherwise lymphotoxic agent and with expression of the double-stranded RNA transgene, the MPA-resistant lymphocytes retained many of their physiological functions: the selected cells could expand exponentially in culture, the expanded T cells were able to mount a normal cytokine response when stimulated, and they could support or inhibit HIV replication depending on the absence or presence of the shRNA gene. The fusion protein, I-GFP, when expressed in transduced cells, produces lower mean fluorescence intensity of GFP than does the SF-GFP gene. Except for this, experimental data in our study indicate no deleterious effect of the IMPDH2 expression on the cellular systems evaluated here. Although we have not tested whether MPA selection affects polyclonal cell growth, the study by Levine et al. demonstrated that prolonged culture of CD4 cells with the same cell expansion scheme resulted in a diverse T cell repertoire²⁹. Thus, it is likely that MPA selection plus cell expansion may allow the production of polyclonal populations of T cells with TH1 and TH2 phenotypes.

Our studies in CD34⁺ cells demonstrated that I-GFP and p24 shRNA expression did not interfere with ex vivo HPC differentiation and the differentiated myelomonocytic cells exhibited normal macrophage functions such as B7.1 up-regulation and phagocytosis upon stimulation. Since no system that allows ex vivo CD34⁺ cell differentiation into T cells is currently available, the issue of whether HPC-derived T cells can be selected by MPA needs to be addressed with in vivo studies. Direct in vivo selection of the transduced lymphocytes with the IMPDH2 mutant represents a new approach to enrich the transduced cell fraction. Unlike the selection of HPCs with MGMT, IMPDH2 selection largely expands the lymphocyte fraction. Lineage-specific cell expansion could be further modulated by using T-cell-specific cis-regulatory sequences, thereby increasing the safety of using such a strategy for gene therapy against HIV infection. As T cells have a definite life span in vivo, the persistence of the in vivo MMF-selected lymphocytes remains to be determined. The evaluation of this MMF application in vivo as a strategy to achieve efficient selection of genetically transduced cells remains to be addressed. Ample information on the dosing of this drug in both animal models and human patients exists, and we are currently exploring these issues in animal models.

Materials and methods

Plasmids

To express anti-HIV shRNAs, the following oligonucleotides were synthesized: p24, 5'-GATCCCCGATTGTACTGAGAGACAGGCTTTCAAGAGAAGCCTGTCTCTCAGTACAA
TCTTTTTGGAAA-3' and 5'-AGCTTTTCCAAAAAGATTGTACTGAGAGACAGGCTTCTCTTGAAAGCCTGTCTCT
CAGTACAATCGGG-3'. Complementary oligonucleotides were annealed and cloned into pBP containing the 233-bp H1 promoter fragment in a pBluescript SK(-) backbone. Successful cloning was confirmed by DNA sequence analysis. The 2.3-kb I-GFP gene containing the GFP gene fused to the 5' end of the IMPDH2 cDNA was generated by PCR. In brief, a 1.5-kb cDNA encoding IMPDH2 was PCR amplified from primary human T lymphocytes using a pair of primers with the following sequences: 5'-GCTATCTGCAGGCCGCCACCATGGCCGACTACCTGATTAG-3' and 5'-CTAGCTCTAGATCAGAAAAGCCGCTTCTCATAC-3'. The fragment was cloned into pBluescript SK(-) and subjected to site-directed in vitro mutagenesis to generate T333I and S351Y mutations with the QuickChange II site-directed mutagenesis kit (Stratagene, La Jolla, CA, USA). The mutated IMPDH2 cDNA was further amplified using the following two primers with the upstream primer containing the 3' end of the GFP coding sequence: 5'-GTATGAGAAGCGGCTTTTCGGAGGCGGTGGAATGAGCAAGGGCGAGGAA CTG-3' and 5'-ATCGAAGCTTGCTAGCTCAGAAAAGCCGCTTCTC-3'. The GFP gene was amplified using the following two primers with the downstream primer containing the 5' end of the mutated IMPDH coding sequence: 5'-TACGGTACCCACCGGCGGCCGCCACCATGAGCAAGGGCGAGGAACTGTTC-3'. and 5'-CAGTTCCTCGCCCTTGCTCATTCCACCGCCTCCGAAAAGCCGCTTCTCATC-3'. The PCR products of the IMPDH2 and GFP genes were annealed and further amplified with the downstream primer of the IMPDH2 cDNA and the upstream primer of the GFP gene. The PCR product was digested with KpnI and HindIII and cloned into pBluescript SK(-). The I-GFP gene was placed under the control of the 0.6-kb SFFV promoter and cloned into pHIV7 to generate pHIV7/I-GFP. A 0.3-kb fragment containing the p24 shRNA gene linked to the H1 promoter was inserted immediately upstream of the SFFV promoter in pHIV7/I-GFP and generated pHIV7/p24/I-GFP. Construction of pHIV7/SF-GFP has been described previously²⁰.

Cell lines and virus

293T and HT1080 cells were maintained in high-glucose (4.5 g/liter) Dulbecco's modified Eagles' medium (DMEM) supplemented with 10% fetal bovine serum (FBS), 2 mM glutamine, and 100 mg/liter gentamicin. H9 cells were grown in RPMI 1640, 10% FBS, 2 mM glutamine, and 100 mg/liter gentamicin. HIV vectors were produced by transient transfection of 293T cells with each vector construct, pCMV-HIV-1 and pCMV-G. Vectors were harvested 40 h after transfection and titered on HT1080 cells for GFP expression by FACS analysis. HIV-1_NL4–3 was generated by transfection of pNL4-3 into 293T cells. The virus was harvested 40 h after transfection and titered on HeLa-CD4-LTR--gal cells as described²⁵. For HIV-1 challenge, cells were infected in duplicate with HIV-1_NL4–3 at an m.o.i. ranging between 0.0002 and 0.02. The infected cells were continually passaged twice weekly and the supernatants were collected for p24 detection using an HIV-1 p24 antigen assay kit (Beckman Coulter, Fullerton, CA, USA).

Transduction and MPA selection of H9 cells

H9 cells were transduced with the indicated vector for 18 h at an m.o.i. of 1 to 4. The optimal dose for MPA selection was predetermined by a dose–response curve (data not shown). Selection was initiated 6 days after transduction with the indicated doses. Cell enrichment was determined by FACS of GFP⁺ cells and cell viability was determined by trypan blue exclusion analysis.

Transduction and MPA selection of PBMCs

PBMCs from healthy donors were isolated by Ficoll–Hypaque (Amersham Biosciences, Uppsala, Sweden). To stimulate cell proliferation, 10⁶ PBMCs were seeded onto a plate coated with 50 g/ml retronectin (Biowhittaker, Walkersville, MD, USA). Dynabeads (Dynal Biotech LLC, Brown Deer, WI, USA) coated with the antibodies against CD3 and CD28 were added in the ratio of 3 beads/cell. The cells were stimulated for 18 h in RPMI 1640 supplemented with 10% FBS and 2 mM L-gutamine. This was followed by vector transduction for 16 h at an m.o.i. of 5. The transduced culture was maintained in RPMI medium supplemented with 50 units/ml IL-2 (R & D Systems, Minneapolis, MN, USA). The transduction efficiency was measured by FACS after day 5. Selection with 0.41 M MPA began 6 days after transduction and was maintained for 10 days.

Cytokine production upon PBMC stimulation

To assay for cytokine production, the cells were washed and restimulated for 24 h with Dynabeads coated with CD3/CD28 antibodies. Culture supernatants were collected immediately before restimulation and 24 h after. Cytokine production was measured by cytometric bead array assay (CBA; Becton–Dickinson, San Jose, CA, USA) as described previously^31,32.

CD34 transduction

CD34⁺ cells were isolated from umbilical cord blood samples with the approval of the Institutional Review Board and the Institutional Biosafety Committee of the City of Hope National Medical Center. About 5 10⁴ to 1 10⁵ CD34⁺ cells were transduced with the indicated vector at an m.o.i. of 20 in 200 l of Iscove's modified Dulbecco's medium (IMDM) containing 15% BIT9500 (Stem Cell Technology, Inc., Vancouver, BC, Canada) supplemented with Flt 3, SCF, and thrombopoietin at 50/50/10 ng/ml (R & D)²⁰. The conditions used for CD34 cell differentiation into myelomonocytic and erythroid lineages were described previously²⁰.

Methylcellulose colony formation from the transduced CD34⁺ cells

For methylcellulose colony formation, the transduced CD34⁺ cells were plated in 0.9% methylcellulose medium according to the manufacturer's instructions (Stem Cell Technology) and incubated at 37°C for 2 weeks. The colonies were scored by morphology and GFP expression by fluorescence microscopy.

Differentiation of CD34⁺ cells derived from fetal liver into macrophages

CD34⁺ hematopoietic progenitor cells were purified from human fetal liver using monoclonal antibody-conjugated immunomagnetic beads (Miltenyi Biotech, Auburn, CA, USA)³³. Cells were cultured in IMDM containing 10% FBS and 25 ng/ml of each of the cytokines IL-3, IL-6, and SCF. Cells were transduced twice with the indicated HIV vectors. To drive transduced CD34⁺ cells toward the myeloid lineage, cells were differentiated in Methocult medium (Stem Cell Technology) for 12 days. To derive monocytes/macrophages, the myeloid colonies were then pooled and cultured in DMEM containing 10% FBS and 25 ng/ml each of the cytokines GM-CSF and M-CSF for 5 days. Subsequent experiments on phenotypic analysis, B7.1 up-regulation, and phagocytosis were performed on these in vitro-derived pure populations of mature macrophages as described below. Monoclonal antibodies used for cell surface markers were purchased from BD Biosciences (San Jose, CA, USA).

Measurement of macrophage functions by B7 up-regulation and by phagocytosis

To determine whether the transduced macrophages produced were able to mount a normal stimulatory response, LPS (5 g/ml) was added to the macrophage cultures. Twenty-four hours after stimulation, the cells were stained with phycoerythrin (PE)–Cy5-conjugated anti-B7.1 and analyzed for B7.1 up-regulation by FACS³⁴. For phagocytosis assay, both LPS and tetramethylrhodamine-conjugated E. coli (K12) Bioparticles (Invitrogen) at 5 g/ml were added to the cell culture. The cells were analyzed for GFP and tetramethylrhodamine, which is detectable in the FL2 channel as PE in a flow cytometer (Beckman Coulter Epics XL)³⁴.

Challenge by wild-type HIV-1

Transduced T cells were seeded at 5 10⁵ cells per well onto 24-well plates and infected with increasing amounts of HIV-1_NL4–3 from m.o.i. 0.0002 to m.o.i. 0.02 for 24 h. The infected cells were continually passaged, and supernatant was collected twice weekly. The levels of p24 in the supernatant were determined by the p24 antigen assay kit (Beckman Coulter, Fullerton, CA).

References

1. Allay, J. A., et al. (1998). In vivo selection of retrovirally transduced hematopoietic stem cells. Nat. Med. 4: 1136–1143. | Article | PubMed | ISI | ChemPort |
2. Cavazzana-Calvo, M., et al. (2000). Gene therapy of human severe combined immunodeficiency (SCID)-X1 disease. Science 288: 669–672. | Article | PubMed | ISI | ChemPort |
3. Spencer, H. T., Sleep, S. E., Rehg, J. E., Blakley, R. L. and Sorrentino, B. P. (1996). A gene transfer strategy for making bone marrow cells resistant to trimetrexate. Blood 87: 2579–2587. | PubMed | ISI | ChemPort |
4. Allay, J. A., Galipeau, J., Blakley, R. L. and Sorrentino, B. P. (1998). Retroviral vectors containing a variant dihydrofolate reductase gene for drug protection and in vivo selection of hematopoietic cells. Stem Cells 16 (Suppl. 1): 223–233. | PubMed |
5. Carpinteiro, A., et al. (2002). Genetic protection of repopulating hematopoietic cells with an improved MDR1-retrovirus allows administration of intensified chemotherapy following stem cell transplantation in mice. Int. J. Cancer 98: 785–792. | Article | PubMed | ChemPort |
6. Davis, B. M., Roth, J. C., Liu, L., Xu-Welliver, M., Pegg, A. E. and Gerson, S. L. (1999). Characterization of the P140K, PVP(138–140)MLK, and G156A O₆-methylguanine-DNA methyltransferase mutants: implications for drug resistance gene therapy. Hum. Gene Ther. 10: 2769–2778. | Article | PubMed | ISI | ChemPort |
7. Davis, B. M., Koc, O. N. and Gerson, S. L. (2000). Limiting numbers of G156A O(6)-methylguanine-DNA methyltransferase-transduced marrow progenitors repopulate nonmyeloablated mice after drug selection. Blood 95: 3078–3084. | PubMed | ISI | ChemPort |
8. Gerson, S. L. (2000). Drug resistance gene transfer: stem cell protection and therapeutic efficacy. Exp. Hematol. 28: 1315–1324. | Article | PubMed | ISI | ChemPort |
9. Ragg, S., et al. (2000). Direct reversal of DNA damage by mutant methyltransferase protein protects mice against dose-intensified chemotherapy and leads to in vivo selection of hematopoietic stem cells. Cancer Res. 60: 5187–5195. | PubMed | ISI | ChemPort |
10. Reese, J. S., Qin, X., Ballas, C. B., Sekiguchi, M. and Gerson, S. L. (2001). MGMT expression in murine bone marrow is a major determinant of animal survival after alkylating agent exposure. J. Hematother. Stem Cell Res. 10: 115–123. | Article | PubMed | ISI | ChemPort |
11. Abonour, R., et al. (2000). Efficient retrovirus-mediated transfer of the multidrug resistance 1 gene into autologous human long-term repopulating hematopoietic stem cells. Nat. Med. 6: 652–658. | Article | PubMed | ISI | ChemPort |
12. Persons, D. A., et al. (2004). Transient in vivo selection of transduced peripheral blood cells using antifolate drug selection in rhesus macaques that received transplants with hematopoietic stem cells expressing dihydrofolate reductase vectors. Blood 103: 796–803. | Article | PubMed | ISI | ChemPort |
13. Sawai, N., Zhou, S., Vanin, E. F., Houghton, P., Brent, T. P. and Sorrentino, B. P. (2001). Protection and in vivo selection of hematopoietic stem cells using temozolomide, O⁶-benzylguanine, and an alkyltransferase-expressing retroviral vector. Mol. Ther. 3: 78–87. | Article | PubMed | ISI | ChemPort |
14. Gerson, S. L., Allay, E., Vitantonio, K. and Dumenco, L. L. (1995). Determinants of O₆-alkylguanine-DNA alkyltransferase activity in human colon cancer. Clin. Cancer Res. 1: 519–525. | PubMed | ISI | ChemPort |
15. Sawai, N., Persons, D. A., Zhou, S., Lu, T. and Sorrentino, B. P. (2003). Reduction in hematopoietic stem cell numbers with in vivo drug selection can be partially abrogated by HOXB4 gene expression. Mol. Ther. 8: 376–384. | Article | PubMed | ISI | ChemPort |
16. Allison, A. C. and Eugui, E. M. (2000). Mycophenolate mofetil and its mechanisms of action. Immunopharmacology 47: 85–118. | Article | PubMed | ISI | ChemPort |
17. Coull, J. J., Turner, D., Melby, T., Betts, M. R., Lanier, R. and Margolis, D. M. (2001). A pilot study of the use of mycophenolate mofetil as a component of therapy for multidrug-resistant HIV-1 infection. J. Acquired Immune Defic. Syndr. 26: 423–434. | ChemPort |
18. Hossain, M. M., Coull, J. J., Drusano, G. L. and Margolis, D. M. (2002). Dose proportional inhibition of HIV-1 replication by mycophenolic acid and synergistic inhibition in combination with abacavir, didanosine, and tenofovir. Antiviral Res. 55: 41–52. | Article | PubMed | ChemPort |
19. Margolis, D. M., et al. (2002). The addition of mycophenolate mofetil to antiretroviral therapy including abacavir is associated with depletion of intracellular deoxyguanosine triphosphate and a decrease in plasma HIV-1 RNA. J. Acquired Immune Defic. Syndr. 31: 45–49. | ChemPort |
20. Yam, P. Y., Li, S., Wu, J., Hu, J., Zaia, J. A. and Yee, J. K. (2002). Design of HIV vectors for efficient gene delivery into human hematopoietic cells. Mol. Ther. 5: 479–484. | Article | PubMed | ISI | ChemPort |
21. Jackson, A. L., et al. (2003). Expression profiling reveals off-target gene regulation by RNAi. Nat. Biotechnol. 21: 635–637. | Article | PubMed | ISI | ChemPort |
22. Villacres, M. C., et al. (2005). Reduced type 1 and type 2 cytokines in antiviral memory T helper function among women coinfected with HIV and HCV. J. Clin. Immunol. 25: 134–141. | Article | PubMed | ChemPort |
23. Hart, D. N., Starling, G. C., Calder, V. L. and Fernando, N. S. (1993). B7/BB-1 is a leucocyte differentiation antigen on human dendritic cells induced by activation. Immunology 79: 616–620. | PubMed | ChemPort |
24. Devitt, A., Moffatt, O. D., Raykundalia, C., Capra, J. D., Simmons, D. L. and Gregory, C. D. (1998). Human CD14 mediates recognition and phagocytosis of apoptotic cells. Nature 392: 505–509. | Article | PubMed | ISI | ChemPort |
25. Kimpton, J. and Emerman, M. (1992). Detection of replication-competent and pseudotyped human immunodeficiency virus with a sensitive cell line on the basis of activation of an integrated beta-galactosidase gene. J. Virol. 66: 2232–2239. | PubMed | ISI | ChemPort |
26. Neff, T., et al. (2002). Pharmacologically regulated in vivo selection in a large animal. Blood 100: 2026–2031. | Article | PubMed | ChemPort |
27. Serrano, L. (2001). A novel drug selection system for isolating T lymphocytes utilizing a minimally manipulated human type 11 inosine monophosphate dehydrogenase. Mol. Ther. 3: S152.
28. Sintchak, M. D., et al. (1996). Structure and mechanism of inosine monophosphate dehydrogenase in complex with the immunosuppressant mycophenolic acid. Cell 85: 921–930. | Article | PubMed | ISI |
29. Levine, B. L., et al. (1997). Effects of CD28 costimulation on long-term proliferation of CD4⁺ T cells in the absence of exogenous feeder cells. J. Immunol. 159: 5921–5930. | PubMed | ISI | ChemPort |
30. Novina, C. D., et al. (2002). siRNA-directed inhibition of HIV-1 infection. Nat. Med. 8: 681–686. | Article | PubMed | ISI | ChemPort |
31. Cook, E. B., et al. (2001). Simultaneous measurement of six cytokines in a single sample of human tears using microparticle-based flow cytometry: allergics vs. non-allergics. J. Immunol. Methods 254: 109–118. | Article | PubMed | ISI | ChemPort |
32. Oliver, K. G., Kettman, J. R. and Fulton, R. J. (1998). Multiplexed analysis of human cytokines by use of the FlowMetrix system. Clin. Chem. 44: 2057–2060. | PubMed | ISI | ChemPort |
33. Bai, J., Gorantla, S., Banda, N., Cagnon, L., Rossi, J. and Akkina, R. (2000). Characterization of anti-CCR5 ribozyme-transduced CD34⁺ hematopoietic progenitor cells in vitro and in a SCID-hu mouse model in vivo. Mol. Ther. 1: 244–254. | Article | PubMed | ChemPort |
34. Anderson, J. A. and Ramesh (2005). CXCR4 and CCR5 shRNA transgenic CD34⁺ cell derived macrophages are functionally normal and resist HIV-1 infection. Retrovirology 2: 53–63. | Article | PubMed | ChemPort |

Acknowledgements

We thank Ying Yu for making the lentiviral constructs and Wen Chang for making the IMPDH2–GFP fusion gene. We are deeply grateful to StemCyte, Inc. (Arcadia, CA, USA) for their generous gift of cord blood used in this study. This work is supported by grant POI AI061839.