The Efficiency of Simian Foamy Virus Vector Type-1 (SFV-1) in Nondividing Cells and in Human PBLs

Virology 280, 243–252 (2001) doi:10.1006/viro.2000.0773, available online at http://www.idealibrary.com on

The Efficiency of Simian Foamy Virus Vector Type-1 (SFV-1) in Nondividing Cells and in Human PBLs Ayalew Mergia,* ,1 Soumya Chari,* Dennis L. Kolson,† Maureen M. Goodenow,‡ and Tina Ciccarone* *Department of Pathobiology, College of Veterinary Medicine, University of Florida, Gainesville, Florida 32611; †Department Pathology, Immunology and Laboratory Medicine, University of Florida College of Medicine Health Science Center, Gainesville, Florida 32610; ‡Department of Neurology, University of Pennsylvania School of Medicine, Philadelphia, Pennsylvania 19104-6140 Received September 5, 2000; returned to author for revision November 21, 2000; accepted November 30, 2000; published online January 30, 2001 Current retroviral vectors based on murine leukemia virus (MuLV) are unable to efficiently transduce nondividing cells. Lentiviruses, such as the human immunodeficiency virus 1 (HIV-1) are efficient at transducing nondividing, growth-arrested, and post-mitotic cells, but due to complex safety considerations, they may have limited potential for human clinical gene transfer. For this reason, alternatives to MuLV and HIV-1 vectors need to be explored. In this paper, we have found that simian foamy virus vector (SFV-1) containing a CMV-LacZ expression cassette is able to efficiently transduce multiple cell types of various species that include epithelial, lymphoid, and hematopoietic-derived human cell lines and fibroblast cell lines of several species. Previously it was reported that foamy virus replication is cell cycle dependent (P. D. Bieniasz, R. A. Weiss, and M. O. McClure, 1995. J. Virol. 69, 7295–7299). However, others studies demonstrated nuclear import of viral DNA in arrested cells (A. Saibi, F. Puvion-Dutilleul, M. Schmid, J. Peries, and H. d. The 1997. J. Virol. 71, 1155–1161). Here, we show efficient LacZ transduction by SFV-1 vectors in several chemically arrested cell lines and terminally differentiated human neurons. SFV-1 vector can transduce cell lines arrested in G1/S phase of the cell cycle by aphidicolin treatment with similar efficiencies to that of dividing cells. The terminally differentiated human neural cell line, NT2N, was transduced with 30–50% efficiency, corroborating our data obtained with the arrested cell lines. To further examine whether the SFV-1 vector can efficiently deliver a gene into clinically important cells for gene therapy, we transduced primary human peripheral blood cells (PBLs) in the presence and absence of phytohemagglutanin (PHA) stimulation. We observed 81% transduction efficiency in non-stimulated PBLs and 87% in PHA-stimulated PBLs with vector infection carried out twice in 8 hours intervals at a multiplicity of infection of 1. Together, these data indicate that SFV-1 based retroviral vectors may provide a safe, efficient alternative to current onco- and lentiviral vectors for gene transfer in cells from a broad spectrum of lineages across species boundaries. © 2001 Academic Press

phages, mucosal dendritic cells, and quiescent T lymphocytes are non-proliferating targets of critical importance for transmission, propagation, and spread of the human immunodeficiency virus (HIV) in the infected host (Fauci, 1996; Levy, 1993). This feature allows vectors based on HIV to transduce nondividing cells and integrate their proviral DNA (Lewis et al., 1992; Lewis and Emerman, 1994). However, HIV vectors for human gene therapy are limited by safety concerns for patients and remain a public health issue. Other lentivirus vectors have a restricted host tropism and it is unclear whether they infect human cells. This stumbling block for developing vectors based on animal lentiviruses has recently been rectified by psuedotyping the system with vesicular stomatitis virus envelope glycoprotein G (VSV-G) (Poeschla et al., 1998b). Nevertheless, the comparative safety and efficacy of vectors based on non-primate lentiviruses remain to be determined. As an alternative to lentivirus vectors, foamy viruses may provide a safe and efficient means of targeting nondividing cells, satisfying current concerns with existing retrovirus vectors. Foamy viruses are found in many

INTRODUCTION Retroviruses provide an efficient means for stably introducing foreign DNA into the cell genome of a broad spectrum of target cells (Mulligan, 1993). The life cycle of retroviruses involves integration of the genome into the host DNA, replication, and transmission to the progeny of infected cells (Coffin et al., 1997). For this reason, retrovirus-mediated gene therapy is a promising approach for the treatment of human disease. Most retroviral vector systems use some components of murine leukemia virus (MuLV) (Miller et al., 1993; Vile and Rusell, 1995). However, MuLV replication requires at least one round of cell division for proviral integration into target cells (Miller et al., 1990; Roe et al., 1993; Springett et al., 1989). Thus, these vectors are inadequate to stably transduce nondividing and/or terminally differentiated cells, which severely restricts their potential utility for clinical gene transfer. In contrast, terminally differentiated macro-

1 To whom reprint requests should be addressed. Fax: (352)-3929704. E-mail: [email protected].

243

0042-6822/01 $35.00 Copyright © 2001 by Academic Press All rights of reproduction in any form reserved.

244

MERGIA ET AL.

mammalian species and appear to be ubiquitous. These viruses can be propagated efficiently in various cell types of several species (Mergia et al., 1996). Cultured epithelial and fibroblast cells as well as lymphoid and neural-originated cells support the growth of foamy viruses. Each foamy virus isolate can infect several mammalian species, and virus has been recovered from many organs in these animals (Hooks and Detrick-Hooks, 1981). To date, no disease has been attributed to foamy virus in naturally or experimentally infected animals (Flugel, 1991; Mergia and Luciw, 1991; Weiss, 1988). Furthermore, animal caretakers accidentally infected with SFV remain healthy (Heneine et al., 1998; Schweizer et al., 1997). The facts that foamy viruses have a broad host range, do not cause disease, and persist in infected humans (Callahan et al., 1999; Heneine et al., 1998; Schweizer et al., 1997) make them an ideal vector system for human disease gene therapy. Recently, SFV-1- and human foamy virus (HFV)-based vectors capable of gene transfer were developed (Mergia and Wu, 1998; Russell and Miller, 1996; Schmidt and Rethwilm, 1995; Wu et al., 1998). Similar to other retroviruses, the sequences within the 5⬘ untranslated leader region and the 5⬘ end of the gag gene are important cis-acting elements for foamy virus vector construction (Wu et al., 1998). However, foamy virus replication is unique and requires additional features for vector construction. For example, a second cis-acting element that is critical for foamy virus vectors is located at the 3⬘ end of the pol gene (Erlwein et al., 1998; Heinkelein et al., 1998; Wu et al., 1998). The absolute requirement of this sequence in the pol gene for foamy virus vectors is novel among retroviruses, although its function in foamy virus replication remains to be determined. The ability of foamy virus vectors to transduce stationary cells remains controversial. Russell and Miller (1996) demonstrated transduction of stationary phase human fibroblasts by HFV vectors, albeit less efficiently than dividing fibroblasts. In these studies, efficiency of gene transduction by HFV was higher than that of MuLV. Gene transfer with foamy virus vector in primate hematopoietic cells also showed better transduction than that of MuLV (Hirata et al., 1996). In support of this, Saib et al. (1997) were able to demonstrate that foamy virus DNA can enter the nucleus of G1/S-phase-arrested cells, although no viral gene expression was observed. In contrast to these studies, others have reported that productive foamy virus infection is cell cycle dependent (Bieniasz et al., 1995). To determine whether a foamy virus vector could be developed for expression of transduced genes in sationary as well as proliferating cells, we evaluated an SFV-1 vector that we previously developed (Wu and Mergia, 1999) for transduction of G1/S phase-arrested cell lines and post-mitotic human neurons. In addition, we evaluated the efficiency of SFV-1 vector in phytohe-

FIG. 1. Schematic representations of SFV-1 vector (pCV7-9) and packaging construct (pCGPETH) used to generate SFV-1 vector socks. The constructions of these plasmids were described previously (Wu and Mergia, 1999). The provirus genome of SFV-1 is shown in the top panel. The numbers represent the nucleotide positions in the proviral genome. CMV is the promoter of the human cytomegalovirus immediate early gene.

magglutanin (PHA) stimulated and unstimulated human peripheral blood cells (PBLs). RESULTS SFV-1 vector system The construction of the SFV-1 vector plasmid pVC7-9 is described elsewhere (Wu and Mergia, 1999). In pVC7-9, the minimal human CMV promoter was fused to the transcription start site located in the R region of the 5⬘LTR to direct high-level vector transcription from ⫹1 of the SFV-1 genome (Fig. 1). This technique has been shown to produce rapid, high-titer MuLV vector constructs by transient transfection (Naviaux et al., 1996). The vector contains SFV-1 sequences from the beginning of the R to the end of the pol gene, the PPT and the 3⬘ LTR. To monitor for the efficiency of transduction, a LacZ gene under the control of CMV promoter is placed between the pol gene and PPT. Packaging plasmid (pCGPETH) for SFV-1 vector is constructed by placing the structural genes gag, pol, and env, and tas downstream from the CMV promoter (Wu and Mergia, 1999). Splice donor of the 5⬘ SFV-1 is provided between CMV promoter and the gag gene in pCGPETH. The SV40 poly A signal sequence is supplied at the 3⬘ end. SFV-1 vector was produced by transient co-transfection of each of vector plasmid pVC7-9 and packaging plasmid pCGPETH with LipofectAMINE transfection reagent (Gibco-BRL). Efficiency of SFV-1 vector in multiple cell lineages Foamy viruses have a broad cell tropism and infect a wide variety of cell types from different species. As a result, foamy virus-based vectors are thought to be ideal

SAFE ALTERNATIVE FOR GENE TRANSFER IN CELLS

245

TABLE 1 Transduction of Multiple Cell Types by SFV-1 Vector, pCV7-9 Adherent cell lines

Cell line

Cell type

Titer ⫾ SE (⫻10 ⫺6 B-gal positive cells/ml) a

293T 293 HeLa HOS.CD4/CXCR4 HOS.CD4/CCR5 COS7 CRFK Cf2Th

Human kidney fibroblast Human kidney fibroblast Human epithelial cell Human osteosarcoma fibroblast Human osteosarcoma fibroblast African green monkey kidney fibroblast Crandell feline kidney fibroblast Canine thymus fibroblast

4.1 ⫾ 0.296 1.8 ⫾ 0.059 0.5 ⫾ 0.032 2.2 ⫾ 0.190 3.1 ⫾ 0.207 3.1 ⫾ 0.470 2.5 ⫾ 0.205 0.9 ⫾ 0.008

Relative transduction efficiency b 1.0 0.42 0.12 0.54 0.76 0.96 0.59 0.22

Suspension cell lines Cell line HuT78 K562

Cell type

Transduction efficiency c

Human T cell Human hematopoietic

69.3 ⫾ 6.1% 56.4 ⫾ 4.3%

a Titers given as number of ␤-Gal-positive cells/ml of supernatant, and calculated based on three independent titration experiments performed in duplicate ⫾ standard error of the means (SE). b Relative transduction efficiency was calculated as SFV-1 titer on each cell divided by the titer obtained on the 293T cell line, where transduction efficiency of 293T cells ⫽ 1.00 (100%). c Data represent percent ␤-galactosidase-positive cells of total cell population ⫾ standard error of the means from three independent experiments performed in triplicate.

for gene delivery into cells of different lineages. To determine the efficiency of SFV-1 vector transduction and to verify that the replication-defective derivative vector construct retained the cell tropism of the parental wild-type virus, SFV-1 vector particles were tested for transduction of fibroblasts, epithelial-, lymphoid-, neuronal-, and hematopoietic-derived cell lines, from human and non-human sources. Virus stocks were prepared by transfecting the SFV-1 vector and packaging plasmid into 293T cells. These cells gave the highest titer vector production (ⱕ5 ⫻ 10 6/ml, data not shown) compared to all cell lines listed in Table 1. We have previously shown, as with other retroviral vector systems, transient high-titer vector can be achieved in the SFV-1 system by replacing the U3 region with the CMV promoter and transfecting 293 cells (Wu and Mergia, 1999). The 293T cells, derived from 293 parental cells, consistently gave enhanced vector production compared to the parental cell line. SFV-1 vector stock prepared in 293T cells was then used to infect cell lines listed in Table 1 to determine the efficiency of SFV-1 vector transduction over broad host ranges. Target adherent cells showed titers ranging from 5 ⫻ 10 5 (HeLa cells) to 4.1 ⫻ 10 6 (293T), while 69% of the non-adherent human T cell line (Hut-78) and 56% of the hematopoietic line (K562) were transduced. These observations revealed that our SFV-1 vector can efficiently transduce cells of distinct cell lineages across different species and suggest a potentially broad clinical application for foamy virus vector-based gene therapy.

SFV-1 vector in growth-arrested cells Previous reports indicated that wild-type HFV, SFV-1, and SFV-6 fail to establish productive infection in G1/Sarrested cells, suggesting that foamy virus vector cannot be utilized to deliver genes in nondividing cells (Bieniasz et al., 1995). However, a subsequent report (Russell and Miller, 1996) demonstrated HFV infection and replication in stationary phase human fibroblasts. To determine whether a foamy virus vector could indeed infect and transduce nondividing cells, we tested our SFV-1 vector in aphidicolin arrested cells (G1/S phase of the cell cycle). Cells were transduced with MuLV or SFV-1 vector after 24 h with or without aphidicolin exposure. Medium was replenished daily with or without aphidicolin for an additional 48 h prior to ␤-Gal assay. Transductions were performed in multiple cell types to verify that cellular factors alone did not contribute to transduction efficiency in arrested cells. Efficiency of transduction was calculated as a percentage of ␤-Gal positive cells in arrested cells versus dividing cells. Consistent with previous reports (Lewis and Emerman, 1994; Olsen, 1998; Poeschla et al., 1998; Springett et al., 1989) the MuLV vector was inefficient at transducing growth-arrested cells with ⬍1% of the arrested cell population staining positive for ␤-Gal in comparison to the transduction of dividing cells (Table 2). For the SFV-1 vector, transduction efficiencies in the G1/S phase-arrested 293T, CRFK, and HeLa cells were 98.5 ⫾ 26.8, 90.4 ⫾ 7.2, and 108 ⫾ 19.9% that of dividing

246

MERGIA ET AL. TABLE 2 Transduction Efficiency of SFV-1 and MuLV Vectors in Multiple Cells in the Presence and Absence of Aphidicolin Percent transduction efficiency (T.E.)

Cell line

Vector

293T

MuLV SFV-1

HeLa

MuLV SFV-1

CRFK

MuLV SFV-1

Titer (⫻10 ⫺6) a

Dividing b

Arrested b

Relative T.E. c

0.02 1.0 0.01 1.0 0.02 1.0 0.01 1.0 0.02 1.0 0.01 1.0

3.4 ⫾ 0.2 90.8 ⫾ 0.8 2.5 ⫾ 0.1 90.0 ⫾ 3.7 8.8 ⫾ 3.9 48.6 ⫾ 4.4 0.3 ⫾ 0.1 11.2 ⫾ 3.2 8.8 ⫾ 3.9 80.0 ⫾ 12.5 2.5 ⫾ 0.6 73.3 ⫾ 15.0

0.0 ⫾ 0.0 0.1 ⫾ 0.1 1.5 ⫾ 0.1 88.7 ⫾ 8.8 0.0 ⫾ 0.0 0.7 ⫾ 0.3 0.1 ⫾ 0.0 12.1 ⫾ 4.0 0.0 ⫾ 0.0 0.7 ⫾ 0.3 1.5 ⫾ 0.5 66.3 ⫾ 14.8

0.2 ⫾ 0.2 0.2 ⫾ 0.1 59.6 ⫾ 10.4 98.5 ⫾ 26.8 0.1 ⫾ 0.1 0.3 ⫾ 0.3 11.8 ⫾ 5.8 108.8 ⫾ 20.0 0.0 ⫾ 0.0 0.9 ⫾ 0.4 14.8 ⫾ 2.8 90.4 ⫾ 7.2

a

Titers determined on 293T cells and represent number of transducing particles/ml as determined by LacZ expression. Transduction efficiency in dividing and arrested cells was calculated as total number of LacZ-positive cells divided by total cell population at time of transduction. c Relative transduction efficiency was calculated as the percent of arrested population stained LacZ-positive divided by the percent of proliferating population-stained positive. Results are from three experiments performed in triplicate. By ANOVA, transduction efficiencies between low-titer SFV-1 and MuLV vectors was statistically significant (␣ ⫽ 0.05). b

cells, respectively. The reason for the higher transduction efficiency in comparison to the previous study with HFV vector (Russell and Miller, 1996) is due to our high titer SFV-1 vector. Transduction with a low-titer SFV-1 vector resulted in low-level gene transfer into growtharrested cells (Table 2). This result is comparable to what is reported with the gene transduction using the HFV vector in stationary phase cell culture. By increasing vector titers 100-fold, we were able to enhance transduction in the arrested populations to similar levels as dividing cells. To verify that growth-arrested cells were stably transduced, the aphidicolin-treated cells were induced to cycle and passaged for several weeks. As expected, SFV-1 vector transduced cells were maintained for 9–12 passages, and the number of ␤-Gal positive cells increased with each passage, though the percentage did not increase. In contrast, no ␤-Gal-positive cells were observed after the first passage of cells transduced with MuLV vector. These results demonstrated similar transduction efficiencies for SFV-1 in both dividing and arrested cells regardless of cell type, verifying that simian foamy virus can infect and deliver a heterologous gene with persistent expression in growtharrested cells. Transduction of postmitotic human neuron cells To further investigate the utility of SFV-1 vectors in nondividing cells, transduction efficiency was examined in terminally differentiated human neuron cells (NT2.N). NT2.N neurons are functional, post-mitotic human neurons derived from NT2 precursor cells by retinoic acid

treatment and differential adhesion. These cells were cultured as described under Materials and Methods with a final plating on Matrigel basement membrane matrix in the presence of a mitotic inhibitor. Cells remain terminally differentiated after removal from mitotic inhibition. Cells plated at a density of 1.8 ⫻ 10 5/cm 2 were infected with either SFV-1 or MuLV vector particles at an m.o.i. of 2.5. Assaying for ␤-Gal activity revealed that 30–50% of the neuronal cells were positive indicating efficient transduction (Fig. 2). Infection of the neurons with MuLV vector showed 0–⬍1% transduction efficiency. The SFV-1 transduced neurons showed ␤-Gal activity both in cell bodies and axonal processes (Fig. 2). These results further established evidence that the SFV-1 vector can efficiently deliver and express heterologous genes into nondividing cells independent of the method of cell-cycle arrest. Transduction of human peripheral blood lymphocytes (PBLs) Development of a vector that can effectively deliver genes into PBLs is important for gene therapy based treatment of immune disorders. Simian foamy viruses have been readily isolated from PBLs of naturally and experimentally infected animals as well as infected animal caretakers (Falcone et al., 1999; Heneine et al., 1998; Hooks and Detrick-Hooks, 1981; Neuman-Haefelin et al., 1993; von Laer et al., 1996). To test the efficiency of SFV-1 vector in human PBLs, Ficoll gradient purified cells obtained from five different donors were independently PHA stimulated and infected twice with SFV-1 vector at a

SAFE ALTERNATIVE FOR GENE TRANSFER IN CELLS

247

FIG. 2. Transduction of SFV-1 vector in postmitotic human neuron (NT2) cells. Both MuLV and SFV-1 vectors were titered by infection of 293T cells and equal amounts were used to infect NT2 cells. (A) The transduction efficiencies of SFV-1 and MuLV vectors are presented. (B) Uninfected (left) and SFV-1 vector infected (right) terminally differentiated NT2 cells. Each panel represents a pure population of NT2 neurons, which have aggregated to form a large cluster composed of several hundred cells. Individual neurons are visible around the edge of the large central cluster and are densely stained blue within the cluster and surrounding it. Large neuritic fasicles extend from the cluster and represent multiple individual neurites that have coalesced to form fasicles.

m.o.i. of 1 in 8-h intervals. Gene transduction was determined by ␤-Gal assay 72 h p.i. The percentage of transduced PBLs was 87% (Table 3), which is significantly higher than that of other retroviral vectors (Douglas et al., 1999; Vandendriessche et al., 1995; Wu et al., 1999; Yang et al., 1999). Transduction efficiency of human PBLs with the MuLV vector was ⬍1%, again consistent with previous reports of infection of PBLs with equivalent moi (Culver et al., 1990, 1991; Morgan et al., 1994). In an attempt to reflect anticipated SFV-1 vector-cell interaction in vivo, we chose to use an experimental culture system of PBLs and vector infection initiated in an unstimulated state. Cells cultured in complete medium in the presence of 30 U/ml IL-2 without PHA stimulation were infected with SFV-1 or MuLV vector at the same day TABLE 3 Transduction of Human Primary Human Peripheral Blood Lymphocytes by SFV-1 Vector

Cell type Unstimulated PBLs (30 U IL-2 only) 48 hours stimulated PBLs (5 ␮g/ml PHA-P ⫹ 30 U/ml IL-2)

Uninfected control

MuLVampho a

SFV-1 vector b

0

⬍1%

0

⬍1%

81 ⫾ 6.3% (77–90%) 87 ⫾ 6.0% (79–94%)

Note. Data are the percent transduction ⫾ standard error of the mean for each experiment performed in triplicate. Results shown are from independent experiments performed in triplicate with samples from three different donors ( a) for MuLV and 5 different donors ( b) for SFV-1. Numbers in parentheses indicate range of transduction efficiency among donors. Both MuLV and SFV-1 vectors were titered by infection of 293T cells and equal amounts were used to infect stimulated and unstimulated PBLs.

of cell harvest. FACS analysis of these cells at Day 0 (at the time vector infection) revealed that 96% were at G0/G1, 0.39% at S, and 3.61% at G2/M phases of the cell cycle (Fig. 3). At 72 h, when cells were harvested for ␤-Gal assay, the vast majority of the cells were at G0/G1 (94%) of the cell cycle. The percentage of SFV-1 vector transduction in these unstimulated PBLs was 81% (Table 3). These results indicate that SFV-1 vector can readily deliver and express transduced genes into both unstimulated and stimulated PBLs, reflecting the usage of the foamy virus vector for transduction of PBL population naturally occurring in vivo. DISCUSSION In this report, we demonstrate the ability of a SFV-1based vector to mediate efficient transduction of various arrested and nondividing cell types, including lymphocytes and neurons. The SFV-1 vector is capable of gene transduction in multiple cell types of different species, reflecting the wide cell tropism associated with foamy virus infection in the animal and propagation in cell culture. This is a major advantage of foamy virus vectors. Furthermore, we have shown that lymphocytes and neurons that are critical for gene transfer in vivo can be infected with SFV-1 vector. These experiments ascertain the potential use of SFV vectors in critical cellular targets in vivo. Not only have we demonstrated that an SFV-1 vector can transduce fibroblast cell lines from different species, and human cell lines of epithelial, lymphoid, and hematopoietic origins, but we also have shown that transduction in all cells is similar. This suggests that cell receptors for primate foamy viruses are expressed at compa-

248

MERGIA ET AL.

FIG. 3. Cell-cycle analysis of PHA-stimulated and -unstimulated human PBLs. The populations of cells at different phases of the cell cycle are given as percent of total population. FACS analyses were performed at the time SFV-1 vector infection and when cell were harvested to determine for LacZ transduction by ␤-Gal assay. Vector infection was carried out at Day 0 (immediately following isolation) for unstimulated cells and at 48 h (at the time of cell activation) for stimulated cells. Cells were stained for ␤-Gal 72 h p.i. On all days, the numbers of cells of the unstimulated cells at G0/G1 remained ⬎94%, while cells progressing through the S to G2/M was ⬍1%. In contrasts, stimulated cells showed a dramatic increase in the number of cells progressing through the S phase from 0.4% at day of isolation to 23% after 48 h of PHA-stimulation.

rable levels on several cell types from different species. Whether the receptor is a protein or another class of cell-surface molecules is not known. Other retroviral vectors require pseudotyping the envelope protein with other viruses in order to broaden cell tropism. This notion may make these vectors cumbersome to utilize as a vehicle to deliver genes in vivo. Since foamy viruses demonstrate naturally broad tissue tropism (Hooks and Detrick-Hooks, 1981; Neuman-Haefelin et al., 1993), SFV-1 vector may be an ideal gene-delivery system. Previously, it was reported that foamy viruses failed to replicate in cells that were arrested in the G1/S phase of the cell cycle by aphidicolin treatment (Bieniasz et al., 1995), although others have shown HFV replication in stationary phase cells (Bieniasz et al., 1995). Our results showed that SFV-1 vector particles can indeed transduce and express transgenes in growth-arrested and postmitotic cells. The efficiency of transduction in these arrested cells is equivalent to that of lentivirus vectors (Klimatcheva et al., 1999; Naldini et al., 1996; Poeschla et al., 1998a,b). In addition, transduction efficiency (30–50%) of postmitotic

human neurons exceeds that of HIV-1 basal vectors (Klimatcheva et al., 1999; Naldini et al., 1996; Poeschla et al., 1998a,b), emphasizing the potential utility of SFV-1 vector in nondividing cells in vivo. In foamy virus infection of G1/S arrested cells, nuclear import of the preintegration complex allows translocation of the genome into the nucleus (Saib et al., 1997). This observation is in support of our finding that noncycling cells can be transduced with the SFV-1 vector. Previous studies showed that, despite nuclear transport of the genome, no 5⬘ foamy virus LTR-driven gene expression was observed in growth-arrested cells (Saib et al., 1997). However, our transgene expression is supported by a CMV promoter, which is active in many growth-arrested cells (Klimatcheva et al., 1999; Naldini et al., 1996; Poeschla et al., 1998a,b). It is possible that the foamy virus promoters may not be effective in growth-arrested cells, thereby limiting gene expression. Furthermore, by increasing vector titers 100-fold, we were able to enhance transduction in the arrested populations to similar levels as dividing cells. Therefore, the reason for the higher transduction efficiency

SAFE ALTERNATIVE FOR GENE TRANSFER IN CELLS

in comparison to previous studies with HFV vector (Russell and Miller, 1996) is in part due to our high-titer SFV-1 vector. Human PBLs are primary targets for gene therapy of inherited and acquired disorders of the immune system. Development of a gene-transfer system that warrants efficient transduction of human PBLs is critical for various applications in gene therapy. The high efficiency of PBL transduction by SFV-1 vector is a major advantage. In contrast, MuLV-mediated gene transfer into primary lymphocytes is generally low (⬍10%) when a low m.o.i. of vector is used (Culver et al., 1990, 1991; Morgan et al., 1994). Consistent with these previous reports, we have also observed ⬍1% transduction when human PBLs were infected with MuLV vector particles at an m.o.i. of 1. This low-level infection is attributed to MuLV’s inefficiency at transducing nondividing cells and restricted replication in human PBLs despite entry (Naldini et al., 1996; Woffendin et al., 1994). A higher transduction efficiency ranging from 20 to 50% in human primary lymphocytes has been accomplished with MuLV vectors only after enhancing vector titers by pseudotyping with vesicular stomtitis virus envelope glycoprotein G (VSV-G) and improving the method of infections with complex manipulations of the system (Burns et al., 1993; Pollok et al., 1998; Sharma et al., 1996; Vandendriessche et al., 1995; Wu et al., 1999; Yang et al., 1999). HIV-based vectors also have some significant limitations. Since HIV infects CD4-positive cells and nondividing cells, HIV-based vectors were thought to be potentially efficient in primary lymphoid cells (Fauci, 1996; Lewis et al., 1992; Spina et al., 1995; von Schwedler et al., 1994; Weinberg et al., 1991). However, low transduction efficiency is also associated with HIV vectors, although transduction efficiency of HIV vectors can be increased several-fold by psuedotyping with VSV-G and subsequent concentration procedures (Akkina et al., 1996; Naldini et al., 1996). Recently, transduction efficiency ⱕ60% of activated human PBLs was observed with VSV-G pseudotyped HIV vector (Douglas et al., 1999). However, we have demonstrated gene transfer with an SFV-1 vector in both PHA-stimulated and -unstimulated PBLs ⬎80% with infection at a significantly lower m.o.i. and without any manipulation of the vector system. This level of vector transduction by SFV-1 is higher than that reported for any other retroviral vector. The reason for the high efficiency of SFV-1 vector may have to do with the distinct mechanisms of virus replication. Reverse transcription occurs late in the foamy virus replication before the virion is released from the infected cells, and ⬃20% of virion contain fully reverse transcribed DNA (Moebes et al., 1997; Yu et al., 1996; Yu et al., 1999). In cases where the barrier for gene transfer is at the point of reverse transcription, foamy virus has an advantage since 20% of the virus particles contain fully reverse transcribed DNA genome that is infectious. A low pool of nucleotides

249

available for reverse transcription in unstimulated lymphocytes following lentivirus vector infection could affect the efficiency of gene transfer (Korin and Zack, 1998; Unutmaz et al., 1999; Zack et al., 1990). Thus, an SFV-1 based-vector will be able to provide a highly efficient method for clinical gene transfer into primary human PBLs. Furthermore, since no disease is associated with foamy viruses, the SFV-1 vector system may prove to be a safe delivery vehicle for human gene therapy than currently utilized vectors. The constructions of SFV-1 vectors and packaging DNAs with minimal genome sequences are in progress to improve the potential utility of the foamy virus vector system in clinical application. MATERIALS AND METHODS Cell culture Cell lines 293 and 293T (human kidney fibroblasts), HeLa (human epithelial cell), HuT 78 (human T cell), K562 (human hematopoietic cell), NT2 (NTera2C1.D1, human neurons), Cos-7 (African green monkey kidney fibroblasts), CRFK (Crandell feline kidney fibroblasts), and murine L929 (murine fibroblasts) were obtained from American Type Culture Collection (Rockville, MD). HOS.CD4/CCR5 and HOS.CD4/CXCR4 cell lines (human osteosarcoma) were made available by the AIDS Research and Reference Reagent Program, Division of AIDS, NAIAD, NIH. All fibroblast (except CRFK cells), NT2 cells, and HeLa cell lines were maintained in Dulbecco’s modified Eagle’s medium (DMEM) containing 10% fetal calf serum (FCS). CRFK cells were cultured in minimal essential media (MEM) supplemented with 1% nonessential amino acids and 10% fetal horse serum (FHS). HuT78 were maintained in RPMI medium containing 10% FCS. K562 were propagated in Iscove’s modified DMEM supplemented with 4 mM L-glutamine and 10% FCS. HOS.CD4/CCR5 and HOS.CD4/CXCR4 cell lines were cultured in DMEM, 2.5 mM HEPES, 10% FBS, and 1 ␮g/ml puromycin. Plasmids Construction of the recombinant SFV-1 packaging plasmid (pCGPETH) and vector plasmid (pCV7-9) were described previously (Wu and Mergia, 1999). The pCGPETH plasmid contains the SFV-1 gag, pol, env, and tas genes under the transcriptional control of the human cytomegalovirus immediate early (CMV-IE) gene promoter, and simian virus 40 (SV40) polyA additional signal sequence (Fig. 1). The pCV7-9 plasmid contains SFV-1 sequences extending from the R region of the 5⬘ LTR to the 3⬘ end of pol (nucleotide positions 1170–7080) and sequences extending from the orf-2 region upstream from the polypurine tract at the 3⬘ end of the genome to the 3⬘LTR (nucleotide positions 10,749–12,972). CMV-IE promoter was fused to the R region to initiate transcrip-

250

MERGIA ET AL.

tion from the beginning of R (⫹1 of the vector genome). pCV7-9 also contains the lacZ marker gene under the transcriptional control of the human CMV-IE promoter inserted between the end of pol gene and the 3⬘ SFV-1 segment. The MuLV vector system plasmids, pCLampho (packaging plasmid) and pCLMFG-lacZ (MuLV vector containing the Lac-Z gene), were gifts from Inder M. Verma of the Salk Institute, San Diego, CA. Transfections and production of packaged vectors Transfections of 293T cells were carried out in six-well plates by the liposome-mediated method with the reagent LipofectAMINE according to manufacturer’s recommendations (Life Technologies, Inc., Gaithersburg, MD). Cells were seeded at a density of 4.5 ⫻ 10 5 in six-well plates and co-transfected with 1–2 ␮g of each pCV7-9 and pCGPETH plasmids or pCLampho and pCLMFG-lacZ. To enhance foamy virus vector production, cells were stimulated the next day with 10 mM sodium butyrate (Sigma) for 12 h and supernatants were harvested at Day 3 or 4 post-transfection. The amounts of SFV-1 vectors produced were titered on fresh 293T cells plated at a density of 2.5 ⫻ 10 4 in 24-well plates. Forty-eight hours after infection, cells were stained for ␤-galactosidase (␤-Gal) activity. Briefly, cells were washed twice in phosphate-buffered saline (PBS), then fixed with 0.05% glutaraldehyde in PBS. Cells were washed in PBS an additional three times and incubated in 5-bromo-4-chloro-3-indolyl-␤-D-galactopyransoside (Xgal) for a minimum of 4 h. Cultures were monitored under phase contrast microscopy for appearance of blue staining, and positive cells were scored. Cell growth arrest and flow cytometry analysis Cells were treated with aphidicolin to arrest them at G1/S phase of the cell cycle. Hela, 293T, and CRFK cell lines, were incubated with varying amounts of aphidicolin and pulsed with 3H-thymidine for 18 h to initially determine the concentration of aphidicolin to prevent DNA replication. Aphidicolin concentrations resulting in ⬍5% 3H-uptake in comparison to normally growing cultures were used for further cell-cycle-arrest studies. HeLa, 293T, and CRFK were treated with 20 ␮g/ml aphidicolin, as described previously (Lewis and Emerman, 1994; Olsen, 1998; Poeschla et al., 1998b; Springett et al., 1989), prior to infection with SFV-1 vectors and replenished daily until staining for ␤-Gal activity. Cellcycle arrest with aphidicolin was verified at 24 h and 72 h by flow cytometry for DNA content by propidium iodide staining as described previously (Vindolv and Christensen, 1990). Postmitotic human neurons NT2N human neurons were prepared as previously described (Pleasure et al., 1992). Briefly, NTera 2 cells

were differentiated to neurons (NT2N) by culturing (3.2 ⫻ 10 3 cells/cm 2) in the presence of 10 ␮M all-trans retinoic acid for 5 weeks followed by replating at 1:6 density without retinoic acid in the presence of mitotic inhibitors (10 ␮M 5-fluoro-2⬘-deoxyuridine, 10 ␮M uridine, 1 ␮M cytosine-␣-D-arabinose furanoside). One week later, a final replating was done using culture vessels coated with Matrigel (Collaborative Biomedical Products, Bedford, MA) without mitotic inhibitors. Postmitotic NT2N neurons were cultured at a density of 1.8 ⫻ 10 5 cells/cm 2 and were inoculated with SFV-1 or MuLV vector at a m.o.i. of 2.5 in 10% FBS/DMEM. After 72 h, cultures were washed (PBS), fixed with 1% formaldehyde and 0.2% glutaraldehyde (in PBS) for 5 min at room temperature, washed, and stained for ␤-Gal activity. Primary peripheral blood lymphocytes Human peripheral blood lymphocytes (PBLs) were isolated by Ficoll-Hypaque density gradient centrifugation. Adherent cells were removed by plating gradient-purified samples and harvesting suspension cells. PBLs were maintained in RPMI 1640 with 10% FBS and 30 U/ml interleukin 2 (IL-2). Cells were infected with MuLV or SFV-1 vector at the same day of isolation or stimulated with 5 ␮g/ml phytohemagglutanin (PHA-P, Difco) for 48 h prior to vector infection. Cells were seeded in 24-well plates (5 ⫻ 10 5 per well), centrifuged at room temperature for 10 min at 2500 rpm, and resuspended in 300 ␮l (moi of 1) of retroviral supernatants. The resuspended cells were then spun at 1500 rpm for 1 h as described previously (Bahner et al., 1996; Bunnell et al., 1995). The infection procedure was repeated 8 h later, and cells were stained for ␤-Gal activity 72 h after infection. ACKNOWLEDGMENTS This research is supported by a grant from the National Institutes of Health to A. Mergia (AI-39126). S. Chari is in part supported by the Cancer Training Grant from the National Cancer Institute and the National Institutes of Health (CA09126-22).

REFERENCES Akkina, R. K., Walton, R. M., Chen, M. L., Li, Q. X., Planelles, V., and Chen, I. S. (1996). High-efficiency gene transfer into CD34⫹ cells with a human immunodeficiency virus type 1-based retroviral vector pseudotyped with vesicular stomatitis virus envelope glycoprotein G. J. Virol. 70, 2581–2585. Bahner, I., Kearns, K., Hao, Q. L., Smogorzewska, E. M., and Kohn, D. B. (1996). Transduction of human CD34⫹ hematopoietic progenitor cells by a retroviral vector expressing an RRE decoy inhibits human immunodeficiency virus type 1 replication in myelomonocytic cells produced in long-term culture. J. Virol. 70, 4352–4360. Bieniasz, P. D., Weiss, R. A., and McClure, M. O. (1995). Cell cycle dependence of foamy retrovirus infection. J. Virol. 69, 7295–7299. Bunnell, B. A., Muul, L. M., Donahue, R. E., Blaese, R. M., and Morgan, R. A. (1995). High-efficiency retroviral-mediated gene transfer into human and nonhuman primate peripheral blood lymphocytes. Proc. Natl. Acad. Sci. USA 92, 7739–7743. Burns, J. C., Friedmann, T., Driever, W., Burrascano, M., and Yee, J. K.

SAFE ALTERNATIVE FOR GENE TRANSFER IN CELLS (1993). Vesicular stomatitis virus G glycoprotein pseudotyped retroviral vectors: Concentration to very high titer and efficient gene transfer into mammalian and nonmammalian cells. Proc. Natl. Acad. Sci. USA 90, 8033–8037. Callahan, M. E., Switzer, W. M., Matthews, A. L., Roberts, B. D., Heneine, W., Folks, T. M., and Sandstrom, P. A. (1999). Persistent zoonotic infection of a human with simian foamy virus in the absence of an intact orf-2 accessory gene. J. Virol. 73, 9619–9624. Coffin, J. M., Hughes, S. H., and Varmus, H. E. (Eds.) (1997). Retroviruses. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory Press. Culver, K., Cornetta, K., Morgan, R., and et al. (1991). Lymphocytes as cellular vehicles for gene therapy in mouse and man. Proc. Natl. Acad. Sci. USA 88, 3155–3159. Culver, K. W., Morgan, R. A., Osborne, W. R., Lee, R. T., Lenschow, D., Able, C., Cornetta, K., Anderson, W. F., and Blaese, R. M. (1990). In vivo expression and survival of gene-modified T lymphocytes in rhesus monkeys. Hum. Gene Ther. 1, 399–410. Douglas, J., Kelly, P., Evans, J. T., and Garcia, J. V. (1999). Efficient transduction of human lymphocytes and CD34⫹ cells via human immunodeficiency virus-based gene transfer vectors. Hum. Gene Ther. 10, 935–945. Erlwein, O., Bieniasz, P. D., and McClure, M. O. (1998). Sequences in pol are required for transfer of human foamy virus-based vectors. J. Virol. 72, 5510–5516. Falcone, V., Leupold, J., Clotten, J., Urbanyi, E., Herchenroder, O., Spatz, W., Volk, B., Bohm, N. A., Toniolo, A. M., and Neumann-Haefelin, D. S. (1999). Sites of simian foamy virus persistence in naturally infected African green monkeys: Latent provirus is ubiquitous, whereas viral replication is restricted to the oral mucosa. Virology 257, 7–14. Fauci, A. S. (1996). Host factors and the pathogenesis of HIV-induced disease. Nature 384, 529–534. Flugel, R. M. (1991). Spumaviruses: A group of complex retroviruses. J. AIDS 4, 739–750. Heinkelein, M., Schmidt, M., Fischer, N., Moebes, A., Lindemann, D., Enssle, J., and Rethwilm, A. (1998). Characterization of a cis-acting sequence in the Pol region required to transfer human foamy virus vectors. J. Virol. 72, 6307–6314. Heneine, W., Switzer, W. M., Sandstrom, P., Brown, J., Vedapuri, S., Schable, C. A., Khan, A. S., Lerche, N. W., Schweizer, M., NeumannHaefelin, D., Chapman, L. E., and Folks, T. M. (1998). Identification of a human population infected with simian foamy viruses. Nat. Med. 4, 403–407. Hirata, R. K., Miller, A. D., Andrews, R. G., and Russell, D. W. (1996). Transduction of hematopoietic cells by foamy virus vectors. Blood 88, 3654–3661. Hooks, J. J., and Detrick-Hooks, B. (1981). Spumavirinae: Foamy virus group infections. Comparative aspects and diagnosis. In “Comparative Diagnosis of Viral Disease” (E. Kurstak, and C. Kurstak, Eds.), Vol. 4, pp. 599–618. Academic Press, Inc., New York. Klimatcheva, E., Resenblatt, J. D., and Planelles, V. (1999). Lentiviral vectors and gene therapy. Front Biosci. 4, D482–496. Korin, Y. D., and Zack, J. A. (1998). Progression to the G1b phase of the cell cycle is required for completion of human immunodeficiency virus type 1 reverse transcription in T cells. J. Virol. 72, 3161–3168. Levy, J. (1993). Pathogenesis of human immunodeficiency virus infection. Microbiol. Rev. 57, 183–289. Lewis, P. F., and Emerman, M. (1994). Passage through mitosis is required for oncoretroviruses but not for the human immunodeficiency virus. J. Virol. 68, 510–516. Lewis, P., Hensel, M., and Emerman, M. (1992). Human immunodeficiency virus infection of cells arrested in the cell cycle. EMBO J. 11, 3053–3058. Mergia, A., Leung, N. J., and Blackwell, J. (1996). Cell tropism of the simian foamy virus type 1 (SFV-1). J. Med. Primatol. 25, 2–7. Mergia, A., and Luciw, P. A. (1991). Replication and regulation of primate foamy viruses. Virology 184, 475–482.

251

Mergia, A., and Wu, M. (1998). Characterization of provirus clones of simian foamy virus type 1 (SFV-1). J. Virol. 72, 817–822. Miller, A. D., Miller, D. G., Garcia, J. V., and Lynch, C. M. (1993). Use of retroviral vectors for gene transfer and expression. Methods Enzymol. 217, 581–599. Miller, D. G., Adam, M. A., and Miller, A. D. (1990). Gene transfer by retroviral vectors occurs only in cells that are actively replicating at the time of infection. Mol. Cell Biol. 10, 4239–4242. Moebes, A., Enssle, J., Bieniasz, P. D., Heinkelein, M., Lindemann, D., Bock, M., McClure, M. O., and Rethwilm, A. (1997). Human foamy virus reverse transcription that occurs late in the viral replication cycle. J. Virol. 71, 7305–7311. Morgan, R. A., Baler-Bitterlich, G., Ragheb, J. A., Wong-Staal, F., Gallo, R. C., and Anderson, W. F. (1994). Further evaluation of soluble CD4 as an anti-HIV type 1 gene therapy: Demonstration of protection of primary human peripheral blood lymphocytes from infection by HIV type 1. AIDS Res. Hum. Retroviruses 10, 1507–1515. Mulligan, R. C. (1993). The basic science of gene therapy. Science 260, 926–932. Naldini, L., Blomer, U., Gallay, P., Ory, D., Mulligan, R., Gage, F. H., Verma, I. M., and Trono, D. (1996). In vivo gene delivery and stable transduction of nondividing cells by a lentiviral vector. Science 272, 263–267. Naviaux, R. K., Costanzi, E., Haas, M., and Verma, I. M. (1996). The pCL vector system: Rapid production of helper-free, high-titer, recombinant retroviruses. J. Virol. 70, 5701–5705. Neuman-Haefelin, D., Fleps, U., Renne, R., and Scweizer, M. (1993). Foamy viruses. Intervirology 35, 196–207. Olsen, J. C. (1998). Gene transfer vectors derived from equine infectious anemia virus. Gene. Ther. 5, 1481–1487. Pleasure, S. J., Page, C., and Lee, V. M.-Y. (1992). Pure, postmitotic, polarized human neurons derived from NTera 2 cells provide a system for expressing exogenous proteins in terminally differentiated neurons. J. Neurosci. 12, 1802–1815. Poeschla, E., Gilbert, J., Li, X., Huang, S., Ho, A., and Wong-Staal, F. (1998a). Identification of a human immunodeficiency virus type 2 (HIV-2) encapsidation determinant and transduction of nondividing human cells by HIV-2-based lentivirus vectors. J. Virol. 72, 6527– 6536. Poeschla, E. M., Wong-Staal, F., and Looney, D. J. (1998b). Efficient transduction of nondividing human cells by feline immunodeficiency virus lentiviral vectors. Nat. Med. 4, 354–357. Pollok, K. E., Hanenberg, H., Noblitt, T. W., Schroeder, W. L., Kato, I., Emanuel, D., and Williams, D. A. (1998). High-efficiency gene transfer into normal and adenosine deaminase-deficient T lymphocytes is mediated by transduction on recombinant fibronectin fragments. J. Virol. 72, 4882–4892. Roe, T., Reynolds, T. C., Yu, G., and Brown, P. O. (1993). Integration of murine leukemia virus DNA depends on mitosis. EMBO J. 12, 2099– 2108. Russell, D. W., and Miller, A. D. (1996). Foamy virus vectors. J. Virol. 70, 217–222. Saib, A., Puvion-Dutilleul, F., Schmid, M., Peries, J., and The, H. d. (1997). Nuclear targeting of incoming human foamy virus Gag proteins involves a centriolar step. J. Virol. 71, 1155–1161. Schmidt, M., and Rethwilm, A. (1995). Replicating foamy virus-based vectors directing high level expression of foreign genes. Virology 210, 167–178. Schweizer, M., Falcone, V., Gange, J., Turek, R., and Neumann-Haefelin, D. (1997). Simian foamy virus isolated from an accidentally infected human individual. J. Virol. 71, 4821–4824. Sharma, S., Cantwell, M., Kipps, T. J., and Friedmann, T. (1996). Efficient infection of a human T-cell line and of human primary peripheral blood leukocytes with a pseudotyped retrovirus vector. Proc. Natl. Acad. Sci. USA 93, 11842–11847. Spina, C. A., Guatelli, J. C., and Richman, D. D. (1995). Establishment of

252

MERGIA ET AL.

a stable, inducible form of human immunodeficiency virus type 1 DNA in quiescent CD4 lymphocytes in vitro. J. Virol. 69, 2977–2988. Springett, G. M., Moen, R. C., Anderson, S., Blaese, R. M., and Anderson, W. F. (1989). Infection efficiency of T lymphcytes with amphotropic retroviral vectors is cell cycle dependent. J. Virol. 63, 3865– 3869. Unutmaz, D., Kewal-Ramani, V., Marmon, S., and Littman, D. (1999). Cytokine signals are sufficient for HIV-1 infection of resting human T lymphocytes. J. Exp. Med. 189, 1735–1746. Vandendriessche, T., Chuah, M. K., Chiang, L., Chang, H. K., Ensoli, B., and Morgan, R. A. (1995). Inhibition of clinical human immunodeficiency virus (HIV) type 1 isolates in primary CD4⫹ T lymphocytes by retroviral vectors expressing anti-HIV genes. J. Virol. 69, 4045–4052. Vile, R. G., and Rusell, S. J. (1995). Retroviruses as vectors. Br. Med. Bull. 51, 12–30. Vindolv, L. L., and Christensen, I. J. (1990). A review of techniqies and results obtained in one laboratory by an integrated system of methods designed for routine clinical flow cytometric analysis. Cytometry 11, 753–770. von Laer, D., Neumann-Haefelin, D., Heeney, J. L., and Schweizer, M. (1996). Lymphocytes are the major reservoir for foamy viruses in peripheral blood. Virology 221, 240–244. von Schwedler, U., Kornbluth, R. S., and Trono, D. (1994). The nuclear localization signal of the matrix protein of human immunodeficiency virus type 1 allows the establishment of infection in macrophages and quiescent T lymphocytes. Proc. Natl. Acad. Sci. USA 91, 6992– 6996. Weinberg, J. B., Matthews, T. J., Cullen, B. R., and Malim, M. H. (1991).

Productive human immunodeficiency virus type 1 (HIV-1) infection of nonproliferating human monocytes. J. Exp. Med. 174, 1477–1482. Weiss, R. A. (1988). Foamy retroviruses. A virus in search of a disease. Nature 333, 497–498. Woffendin, C., Yang, Z. Y., Udaykumar, X. L., Yang, N. S., Sheehy, M. J., and Nabel, G. J. (1994). Nonviral and viral delivery of a human immunodeficiency virus protective gene into primary human T cells. Proc. Natl. Acad. Sci. USA 91, 11581–11585. Wu, A. I., Liu, X., Mazumder, A., Bellanti, J. A., and Meehan, K. R. (1999). Improvement of gene transduction efficiency in T lymphocytes using retroviral vectors. Hum. Gene. Ther. 10, 977–982. Wu, M., Chari, S., Yanchis, T., and Mergia, A. (1998). Cis-acting sequences required for simian foamy virus type 1 (SFV-1) vectors. J. Virol. 72, 3451–3454. Wu, M., and Mergia, A. (1999). Packaging cell lines for simian foam virus type 1 (SFV-1) vectors. J. Virol. 73, 4498–4501. Yang, S., Delgado, R., King, S. R., Woffendin, C., Barker, C. S., Yang, Z. Y., Xu, L., Nolan, G. P., and GJ, G. J. N. (1999). Generation of retroviral vector for clinical studies using transient transfection. Hum. Gene Ther. 10, 123–132. Yu, S. F., Baldwin, D. N., Gwynn, S. R., Yendapalli, S., and Linial, M. L. (1996). Human foamy virus replication: A pathway distinct from that of retroviruses and hepadnaviruses. Science 271, 1579–1582. Yu, S. F., Sullivan, M. D., and Linial, M. L. (1999). Evidence that the human foamy virus genome is DNA. J. Virol. 73, 1565–1572. Zack, J., Arrigo, S., Weitsman, S., Go, A., Maisip, A., and Chen, I. (1990). HIV-1 entry into quiescent lymphocytes: Molecular analysis reveals a labile, latent viral structure. Cell 61, 213–222.