- Email: [email protected]
Kinetics of Efficient Recombinant Adeno-Associated Virus Transduction in Retinal Pigment Epithelial Cells
Experimental Cell Research 267, 184 –192 (2001) doi:10.1006/excr.2001.5236, available online at http://www.idealibrary.com on
Kinetics of Efficient Recombinant Adeno-Associated Virus Transduction in Retinal Pigment Epithelial Cells Yvonne K. Y. Lai,* Fabienne Rolling,† Elizabeth Baker,‡ and Piroska E. Rakoczy 1,* *Centre for Ophthalmology and Visual Science, University of Western Australia, Nedlands, Western Australia, Australia, †Lions Eye Institute, Nedlands, Western Australia; and ‡Molecular Cytogenetics, Women’s and Children’s Hospital, Adelaide, South Australia
The aim of this study was to investigate the premise that retinal pigment epithelial (RPE) cells are more permissive to recombinant adeno-associated virus (rAAV) transduction than other cells. We investigated the kinetics and mechanisms of rAAV transduction in RPE cells and found that the transduction efficiencies of cultured RPE cells HRPE51 and ARPE19 were significantly higher than those of 293 (P < 0.008) and HeLa (P < 0.025) cells. In addition, RPE cells reached maximum transduction efficiency at a much lower m.o.i. (m.o.i. 10) than 293 cells (m.o.i. 25). Competition experiments using 1 g/ml heparin inhibited the high level of transduction in RPE cells by 30%, but additional heparin failed to reduce rAAV transduction further. Southern hybridization of low-molecular-weight DNA from transduced RPE cells indicated that 42% of single-stranded rAAV DNA was translocated into the nucleus by 2 h postinfection. By 6 h postinfection, double-stranded rAAV DNA was observed, which coincided with the onset of transgene expression. Southern and fluorescence in situ hybridization of total genomic DNA indicated that long-term transgene expression in RPE cells was maintained by the integration of rAAV into the cellular chromosome. Together, these results suggest that the high permissiveness of RPE cells is not related to the presence of heparan sulfate receptors or nuclear trafficking but may be due to an enhanced rate of second-strand synthesis and that integration in RPE cells is responsible for longterm transgene expression. © 2001 Academic Press Key Words: gene transfer; adeno-associated virus; retinal pigment epithelium; second-strand synthesis; virus integration.
INTRODUCTION
Adeno-associated virus (AAV)-mediated transduction and transgene expression have been reported in a 1
To whom correspondence and reprint requests should be addressed at Centre for Ophthalmology and Visual Science, University of Western Australia, 2 Verdun Street, Nedlands, WA 6009, Australia, Fax: ⫹61 8 9381 0700. E-mail: [email protected]. 0014-4827/01 $35.00 Copyright © 2001 by Academic Press All rights of reproduction in any form reserved.
wide range of cells and tissues, both in vitro and in vivo. These include hematopoietic cells, muscle, liver, lung, central nervous system, and retina [1– 6]. However, the level of transduction varies considerably among different cell types. For example, several studies have shown extensive transgene expression at the injection site following intramuscular injection of recombinant AAV (rAAV) vectors [2, 7]. In contrast, efficiency of rAAV gene transfer to the airway epithelia is much lower [8]. Furthermore, even though rAAV transduction in the brain is highly efficient overall, the degree of transgene expression is not uniform between the different cell types and regions of the brain [3, 9]. Similarly, wide variations in transduction efficiency of rAAV have been observed within the retina. Ganglion cells, inner retinal cells, photoreceptors, and retinal pigment epithelial (RPE) cells have been reported as expressing transgenic protein following rAAV delivery in vivo [5, 10 –12]. In our hands, transduction of rAAVCMV-gfp in the rat eye was detected mostly in the RPE, which showed intense and persistent green fluorescent protein (GFP) expression in up to 90% of the subretinally injected area [6, 13]. The ability to target and transfer genes into RPE cells is important, as this cell layer is believed to play an important role in ocular diseases such as age-related macular degeneration and several types of retinal degenerations [14 –17]. This preferential targeting of the RPE in vivo by rAAV has led us to question whether RPE cells are more permissive to rAAV transduction than other cell types and to explore some possible mechanisms that might facilitate these high transduction levels. Additionally, we examined the longevity of transgene expression and the molecular state of the rAAV vector persisting in RPE cells in vitro. METHODS Cells and viruses. Three types of human RPE cell cultures were used: two established cell lines, ARPE19 and D407 [18, 19], and a low-passage primary culture, HRPE51, established in our laboratory [20]. Two other cell lines were used as controls, Ad5-transformed kidney 293 and human cervical carcinoma HeLa. All cells were
184
185
rAAV GENE TRANSFER INTO RPE CELLS cultured in Dulbecco’s modified essential medium (DMEM; GIBCO BRL, Grand Island, NY) supplemented with 10% fetal calf serum (FCS; GIBCO), 100 units/ml penicillin (GIBCO), and 100 g/ml streptomycin (GIBCO). The cells were maintained at all times at 37°C and 5% CO 2. The identity of RPE cultures was confirmed by positive cytokeratin AE1/AE3 immunochemistry (DAKO, Carpinteria, CA). Vector rAAVCMV-gfp, based on AAV type 2, carries the enhanced GFP and neomycin resistance (neo r) genes, which are driven by the cytomegalovirus (CMV) and SV40 promoters, respectively [6]. In all of the following experiments except the heparin competition assay, CsCl gradient-purified rAAVCMV-gfp was used. The virus was prepared and titer determined in 293 cells as previously described [6]. For the heparin competition assay, rAAVCMV-gfp was purified on an iodixanol (Optiprep; Nycomed, Oslo, Norway) density step gradient, followed by heparin affinity chromatography (HiTrap Heparin; Amersham Pharmacia, Uppsala, Sweden) [21]. Purified rAAVCMV-gfp was heat-inactivated at 56°C for 1 h. Absence of helper Ad contamination was confirmed by the lack of cytopathic effect when 293 cells were exposed to the same amounts of rAAVCMV-gfp as those used in this study. Ad5dl309, used in the transduction experiments, was produced in 293 cells and the titer determined by limiting dilution. Transduction assays. Virus infections were done in DMEM supplemented with 2% FCS. Cells (10 5) were infected with rAAVCMVgfp at a multiplicity of infection (m.o.i.; transducing units per cell) of 0.1, 1, 10, or 25 and in the presence or absence of Ad5dl309 (m.o.i. of 120 pfu/cell). After a 24-h infection, cells were harvested, washed with phosphate-buffered saline (PBS), and fixed in 2% paraformaldehyde. The extent of rAAVCMV-gfp transduction was then measured using flow cytometry on a FACSCalibur machine (Becton Dickinson, Franklin Lakes, NJ). Heparin inhibition. The assay was performed as previously described [22] with minor modifications. Briefly, heparin column-purified rAAVCMV-gfp at a m.o.i. of 2 was incubated with 1, 15, and 30 g/ml soluble heparin (Sigma Chemical Co., St. Louis, MO) in DMEM for 1 h at 37°C. The heparin–rAAVCMV-gfp mixture was then allowed to adsorb to 10 5 ARPE19 cells (grown in 6-well plates) for 1 h at 37°C. The cells were washed with PBS, overlaid with DMEM supplemented with 2% FCS, harvested after 40 h, and analyzed by flow cytometry. Cellular fractionation. HRPE51, 293, and HeLa cells grown in 6-well plates were infected with rAAVCMV-gfp at a m.o.i. of 1. After a 2-h incubation, the cells were trypsinized, mixed with FCS to inactivate the trypsin, washed with PBS, and pelleted. Nuclear and cytoplasmic fractions were obtained using a previously described protocol [23]. Briefly, the cell pellet was resuspended in 200 l homogenization buffer (10 mM Hepes, pH 7.9, 10 mM KCl, 0.1 mM EDTA, 0.1 mM EGTA, 1 mM DTT, 0.5 mM PMSF) and vortexed. After 15 min incubation on ice, 12.5 l 10% Nonidet P-40 was added and vortexed vigorously for 10 s. Test of cell lysis by trypan blue exclusion [24] typically gave a yield of around 95% stained nuclei. The nuclei were then pelleted at 10,000g for 30 s. The cytoplasmic fraction contained within the supernatant was collected and the nuclear pellet remaining was washed with homogenization buffer. Low-molecular-weight or Hirt DNA was isolated from both cytoplasmic and nuclear fractions using previously described protocols [25]. Hirt DNA was resolved by agarose gel electrophoresis and analyzed by Southern hybridization. The relative densities of the autoradiographic signals were approximated using the Scion Image program (Scion Corp, Frederick, MD). Analysis of second-strand synthesis. HRPE51 and 293 cells seeded in 6-well plates were infected with rAAVCMV-gfp at a m.o.i. of 1. The 293 cells were also coinfected with Ad at a m.o.i. of 120. At 6, 24, and 48 h postinfection, the cells were trypsinized, washed with PBS, and pelleted. Hirt DNA was isolated and the final DNA pellet resuspended in 20 l Tris–EDTA buffer, pH 8.0. Hirt DNA was
resolved by agarose gel electrophoresis and then subjected to Southern hybridization. Generation and selection of transduced clonal populations. RPE D407 cells were grown to 80% confluence in six-well plates. Each well either was infected with 10 8 rAAVCMV-gfp particles (0.166 fmol) or was transfected with 0.1 g pEGFP-N1 (32 fmol; Clontech, Heidelberg, Germany). pEGFP-N1 contains the same GFP–neo r construct as rAAVCMV-gfp but without any AAV sequences. Plasmid transfection was achieved using Fugene 6 reagent (Roche, Indianapolis, IN). Twenty-four hours after infection or transfection, RPE cells from each well were passaged into a 15-cm-diameter plate. G418 (GIBCO), a neomycin analogue, was added the following day to a final concentration of 0.7 mg/ml. The numbers of neo r and GFP-expressing clones were determined by counting ten randomly selected fields under fluorescence microscopy. After 2 weeks of selection, resistant RPE clones generated by rAAVCMV-gfp were isolated using cloning rings and cultured as described above. Analysis of stable vector DNA. Hirt DNA and total genomic DNA were isolated from individual rAAV-transduced RPE clones [24, 25]. Approximately 20 g total genomic DNA was digested with restriction endonucleases, and the molecular structure of vector DNA present was then analyzed by agarose gel electrophoresis and Southern hybridization. Southern hybridization. Hirt DNA and digested genomic DNA were resolved on 0.8% agarose gels and transferred to nylon membranes (Gene Screen Plus; NEN Life Science, Boston, Massachusetts) by Southern blotting. The membranes were hybridized with a [ 32P]-dCTP-labeled GFP–neo r probe and exposed to radiographic films. Fluorescence in situ hybridization. Metaphase chromosomes were harvested from cells in culture, hypotonically treated with 0.075 M KCl, fixed in methanol–acetic acid (3:1), spread onto cleaned glass slides, and left to dry overnight. The probe specific for the GFP–neo r genes was nick-translated with biotin-14 – dATP and hybridized in situ to the metaphases at a final concentration of 15 ng/l. The fluorescence in situ hybridization (FISH) method was as described in Callen et al. [26], except that chromosomes were stained before analysis with both propidium iodide (as a counterstain) and diaminophenylindole (DAPI) (for chromosome identification). Images of metaphase preparations were visualized on an AX70 microscope (Olympus, Australia) fitted with a cooled CCD camera, filter wheel, and the ChromoScan image collection and enhancement system (Applied Imaging Corp., Newcastle Upon Tyne, UK). FISH signals and the DAPI banding pattern were merged for figure preparation.
RESULTS
Kinetics of RPE Cell Transduction in Vitro The ability of rAAV to transduce RPE cells was determined by infecting two human RPE cell lines and a low-passage RPE primary culture with a rAAV vector carrying the GFP gene, rAAVCMV-gfp. The number of fluorescent cells, as measured by FACS analysis, showed that the level of transduction by rAAVCMV-gfp in RPE cells was high at a m.o.i. of 1 (Fig. 1A). All three RPE cultures showed efficiencies greater than 31% and up to 46% for HRPE51. When compared with 293 cells, the transduction efficiencies of all three RPE cultures were significantly higher than that of the 293 cells (P ⬍ 0.05). This higher efficiency was especially evident with HRPE51 and ARPE19, which had at least 22.6% more GFP-expressing cells than 293 (P ⬍ 0.008
186
LAI ET AL.
Effect of Heparin Competition on rAAV Transduction in RPE Cells Preincubation of rAAV with heparin has been shown to completely block rAAV transduction in HeLa cells, suggesting that heparan sulfate proteoglycans (HSPG) are primary cellular receptors for AAV [22]. To determine whether the presence of HSPG receptors could be responsible for the high level of rAAV transduction in RPE cells, ARPE19 cells were infected with rAAVCMV-gfp that had been preincubated with increasing concentrations of heparin. Use of heparin column-purified rAAVCMV-gfp ensured that all viral particles were able to bind to the added heparin. In the presence of 1 g/ml heparin, rAAVCMV-gfp transduction was inhibited by 30% (Fig. 2A). However, higher concentrations of heparin could not block transduction any further, indicating that saturation of rAAV particles with heparin failed to completely inhibit rAAV binding and entry into RPE cells. Subcellular Distribution of Viral DNA in RPE Cells
FIG. 1. The efficiency of rAAVCMV-gfp transduction in RPE cells in vitro. (A) HRPE51, ARPE19, D407, control HeLa, and 293 cells were infected with rAAVCMV-gfp at a m.o.i. of 1. (B) HRPE51, ARPE19, and 293 cells infected with rAAVCMV-gfp at increasing m.o.i.s (0.1, 1, 10, 25). The transduction efficiency was determined as the percentage of GFP-expressing cells at 24 h postinfection (A and B, n ⫽ 2, mean value ⫾ SD).
and P ⬍ 0.007, respectively). The transduction efficiencies of HRPE51 and ARPE19 were also significantly higher than that of HeLa cells at this low m.o.i. (P ⬍ 0.02 and P ⬍ 0.03, respectively). Further transduction of HRPE51, ARPE19, and 293 cells with increasing m.o.i.s showed that the profiles of transgene expression were markedly different between RPE and 293 cells (Fig. 1B). At a high m.o.i. of 25, the upper transduction limit was reached by all three cultures with efficiencies around 80%; however, the rate at which the cultures attained this maximum differed significantly. At a m.o.i. of 10, RPE cells were already maximally transduced, but only 60% of 293 cells expressed GFP. A slower rate of increase was also obtained with HeLa cells (data not shown).
Impaired intracellular trafficking of viral DNA to the nucleus was recently identified as a barrier to efficient rAAV transduction [27]. To investigate the rate of rAAV DNA entry into the RPE cell nucleus, low-molecular-weight or Hirt DNA was isolated from the cytoplasmic and nuclear fractions of HRPE51, 293, and HeLa cells transduced by rAAVCMV-gfp at 2 h postinfection. The Hirt DNA was then analyzed by Southern hybridization using a GFP–neo r-specific probe, and the relative amounts of input viral DNA present were approximated using Scion Image software. By the 2 h time point, approximately 42% of single-stranded rAAV DNA were detected in the nuclei of HRPE51 cells (Fig. 2B, lanes 1 and 2). Analysis of 293 and HeLa cells showed that 40% and 47%, respectively, of the signal were found in the nuclei of these cells (lanes 3– 6). Comparison of these relative densities indicated that similar amounts of viral DNA were present in the nuclear fractions of the three cultures, suggesting that rAAV DNA successfully entered into the nuclei of HRPE51 cells and at rates similar to those of HeLa and 293 cells. Second-Strand Synthesis in RPE Cells Second-strand synthesis has been proposed as the major obstacle to efficient rAAV transduction and may be enhanced by coinfection with helper Ad [28, 29]. As mentioned, in the absence of Ad RPE cells were more readily transduced by rAAVCMV-gfp than 293 cells (Figs. 1A and 1B). However, the addition of Ad to these cells did not influence rAAV transduction in the ARPE19 cell line or low-passage primary HRPE51 (Fig. 3A). On the other hand, the number of 293 cells
rAAV GENE TRANSFER INTO RPE CELLS
187
formation of double-stranded vector genomes, Hirt DNA was isolated from HRPE51, and 293 plus Ad cells at different time points after infection. As expected, Ad coinfection of 293 enhanced conversion of singlestranded rAAVCMV-gfp to double-stranded DNA forms, with little input viral DNA detected after 24 h (Fig. 3B, lanes 5 and 6). Surprisingly, a similar pattern was observed for HRPE51 without Ad, in which most of the single-stranded vector had been converted to duplex replication forms (monomer and dimer; Rfm and Rfd) at 24 h, and by 48 h very little input singlestranded DNA was present (lanes 2 and 3). In contrast, a significant amount of single-stranded viral DNA remained after 24 h in HeLa cells (lane 7). Furthermore, as early as 6 h postinfection, double-stranded forms of the vector genome were already clearly visible in HRPE51 (lane 1), which coincided with the onset of GFP expression in some cells (data not shown). A band at around 2 kb was also obtained in all samples. The identity of this fragment is unclear but may represent circular intermediates. Nevertheless, in HRPE51 and 293 cells, this fragment was significantly lower in intensity than the replicative DNA forms. From this result, it appears that second-strand synthesis is enhanced in HRPE51 cells, similar to that in Ad-infected 293 cells, suggesting that this faster conversion to transcriptionally active forms may account for the higher level of transduction seen in RPE cells. Comparison of rAAVCMV-gfp Transduction and Plasmid Transfection in RPE Cells
FIG. 2. (A) Infection of ARPE19 with rAAVCMV-gfp preincubated with the indicated concentrations of soluble heparin. Each point represents the average percentage GFP-expressing cells relative to the maximum level obtained without addition of heparin (n ⫽ 2, mean value ⫾ SD). (B) Southern blot analysis of Hirt DNA from nuclear (Nuc.) and cytoplasmic (Cyt.) fractions of rAAVCMV-gfptransduced HRPE51, 293, and HeLa cells at 2 h postinfection. The position of single-stranded viral DNA (ssDNA) is indicated.
expressing GFP was significantly increased in the presence of Ad, especially at lower rAAV m.o.i.s of 0.1 to 10 (P ⬍ 0.03). Coinfection with Ad increased the transduction efficiency of 293 cells to levels similar to those obtained in ARPE19 and HRPE51, thereby indicating that the low efficiency in 293 observed previously in the absence of Ad was due to this rate-limiting step of second-strand synthesis. To determine whether there was a correlation between the level of expression in RPE cells and the
After establishing that RPE cells are readily transduced, the stability of rAAV expression in RPE cells was compared with plasmid transfections. The RPE D407 cell line grown in 6-well plates was either transduced with rAAVCMV-gfp or transfected with plasmid pEGFP-N1, which contains the same GFP–neo r construct as rAAVCMV-gfp. After several days of G418 treatment, many single cell-derived clones were evident in both samples. However, after 10 days under continual G418 selection, the majority of pEGFP-N1transfected clones had started to die. In contrast, most of the rAAVCMV-gfp-transduced clones remained resistant to G418 and continued to express GFP. In total, approximately 6% GFP-expressing clonal populations were obtained with rAAVCMV-gfp infection. This was seven times more than that generated from pEGFP-N1 transfection, even though a 200-fold excess molar quantity of plasmid was transfected per well. Of the rAAVCMV-gfp clones generated, seven were isolated, expanded, and maintained without G418 selection. After 50 passages, they were still stably expressing GFP (data not shown).
188
LAI ET AL.
FIG. 3. (A) Infection of HRPE51, ARPE19, and 293 cells with rAAVCMV-gfp at increasing m.o.i.s (0.1, 1, 10) in the absence (⫺) or presence (⫹) of Ad5dl309 (m.o.i. 120) (n ⫽ 2, mean value ⫾ SD). (B) Southern blot analysis of Hirt DNA from rAAVCMV-gfp-transduced HRPE51 minus Ad (lanes 1–3) and 293 plus Ad (m.o.i. 120) cells (lanes 4 – 6) at 6, 24, and 48 h postinfection. Hirt DNA from transduced HeLa cells at 24 h postinfection is also shown (lane 7). Bands corresponding to ssDNA and duplex Rfm and Rfd forms are indicated.
Molecular State of rAAVCMV-gfp in RPE Cells in Vitro To determine whether the greater stability of rAAVCMV-gfp in RPE cells could be attributed to a
particular molecular state of the vector (episomal or integrated), both Hirt and total genomic DNA were isolated from the seven rAAVCMV-gfp-transduced clones and analyzed for vector presence.
rAAV GENE TRANSFER INTO RPE CELLS
189
FIG. 4. Integration of rAAVCMV-gfp in transduced RPE D407 clones. (A) Southern blot analysis of total genomic DNA extracted from seven GFP ⫹–neo r clones and a parent untransduced D407 cell line. DNA was digested with EcoRV (I) and HindIII (II). Each lane corresponds to the clone number. (B) FISH analysis of integrated vector sequence in four representative GFP ⫹–neo r clones: clone 2 (I), clone 4 (II), clone 5 (III), and clone 6 (IV). A FISH signal was detected in each clone (white dot, arrow), which indicated integration into chromosomes 7, 17, 6, and 1, respectively.
Southern blot analysis of Hirt DNA from the D407 clones at early passage 6 did not reveal any free monomers or dimers of the rAAVCMV-gfp genome (data not shown). Instead, hybridizing bands were detected in total genomic DNA, suggesting vector integration into the RPE chromosome. Digestion with EcoRV (no restriction site within the vector; Fig. 4AI) generated single bands greater than 12 kb, suggesting integration at only one site per clone. Restriction with other enzymes indicated that each integration event involved only a single vector copy (data not shown). Additional analysis with HindIII-digested DNA (single cut within vector; Fig. 4AII) revealed different-sized
bands pointing to distinct viral-cellular junctions between clones. This suggested that these integration events might have occurred randomly at various chromosomal sites. FISH analysis was used to further investigate the sites of vector integration. Hybridization of metaphase chromosomes with biotinylated GFP–neo r probes revealed the presence of vector-specific FITC signal at a single chromosomal site in each of the seven clones. Analysis of these chromosomes revealed that rAAVCMV-gfp had integrated into different sites for each clone. Figure 4B shows FISH results from four clones with vector integration into chromosomes 7, 17,
190
LAI ET AL.
6, and 1 for clones 2, 4, 5, and 6, respectively, strongly supporting random vector integration. DISCUSSION
Previous studies have shown that rAAV can mediate efficient gene transfer into the retina [5, 10 –12]. The retina is composed of several different cell layers, which vary in permissiveness to rAAV transduction. These differences in cell permissiveness in vivo are consistent with previous in vitro studies that demonstrated varying rAAV transduction efficiencies between different cell types [30, 31]. These studies reported more efficient transduction in 293 and HeLa cells than in many other cell lines, such as ovarian CHO-K1, bronchial epithelial IB3, and carcinoma KB and SW480 cells. In the present study, we have shown that RPE cells not only were more permissive to rAAV transduction, but also reached maximum transduction at a much lower m.o.i. than 293 or HeLa cells. This result suggests that rAAV-mediated gene transfer and expression occur much faster in RPE cells than in any of the above cell lines and supports our in vivo observations [6, 13] of highly efficient rAAV uptake in RPE cells. This high permissiveness of RPE cells might be due to various factors, some of which include the presence and abundance of cellular receptors, second-strand synthesis, the dephosphorylated state of singlestranded D-sequence-binding protein (ssD-BP), and nuclear trafficking [22, 27–29, 32, 33]. Summerford and Samulski [22] reported total inhibition of rAAV-lacZ transduction in HeLa cells by heparin competition and thus proposed that HSPGs acted as primary receptors for AAV. Likewise, our results suggested that HSPGs were present on RPE cells and were partly responsible for rAAV binding. However, there is currently some disagreement in the literature regarding the completeness of heparin competition in HeLa cells [22, 33, 34]. We found that saturation of rAAVCMV-gfp particles with heparin could not completely inhibit transduction in RPE cells. This suggests that HSPGs are not responsible for the high permissiveness of RPE cells and supports the findings of Qiu et al. [33], who proposed the existence of another highaffinity receptor for AAV in addition to low-affinity heparan sulfate binding. Nuclear trafficking of viral DNA has also been identified as an obstacle to efficient rAAV transduction [27]. Translocation of rAAV DNA from the cytoplasm to the nucleus was shown to be efficient in 293 and HeLa cells [27]. Based on Southern analysis of Hirt DNA, the amount of single-stranded rAAV DNA detected in the nuclei of RPE cells at 2 h postinfection was similar to those in 293 and HeLa cells. This suggests that intracellular trafficking is neither impaired nor enhanced in
RPE cells and therefore cannot account for the high transduction efficiencies in these cells. It has been well established that coinfection of 293 cells with Ad increases the transduction efficiency of rAAV [28, 29]. However, no increase was observed in RPE cells in the presence of Ad. This lack of an Adinduced increase in transduction efficiency cannot be related to impairment of Ad infection, as RPE cells are known to be highly permissive to Ad infection [20, 35]. It was also interesting to note that the transduction efficiency of RPE cells without Ad was comparable to that of 293 plus Ad. Previous studies have established that this augmentation of rAAV transduction by Ad is due to enhanced second-strand synthesis [28, 29]. To determine if high rAAV transduction in RPE cells was related to enhanced second-strand synthesis, we compared Hirt DNA from transduced HRPE51 and 293⫹Ad cells and found that formation of a doublestranded viral genome in these RPE cells was significantly enhanced to a level similar to that in Ad-infected 293 cells. These data suggest that enhanced viral genome conversion might contribute to the greater permissiveness of RPE cells. Certain elements or cellular factors that promote this conversion may be present in HRPE51 cells. Ali et al. [5] showed that ␥-irradiation improved rAAV-lacZ transduction in the retina. The mechanism involved has not been established but possibly resembles that of UV irradiation, which stimulates formation of viral circular intermediates [36]. However, the negligible amount of circular intermediates observed in this study suggests that the cellular factors intrinsically present in RPE cells may act differently from genotoxic stresses such as ␥- and UV irradiation and the adenoviral E2a gene product [25, 36, 37]. A possible factor contributing to this high permissiveness could be increased levels of dephosphorylated ssD-BP, which has been found to enhance second-strand synthesis [32]. Further studies are needed to see if RPE cells in vivo also behave similarly and to fully elucidate the mechanisms involved, including any other factors that may contribute to this high permissiveness. Consistent with other studies [38], in vitro transduction of RPE cells with rAAV generated more stable clones than transient plasmid pEGFP-N1 transfection. We further established that the stability of expression in rAAV-transduced clones was due to vector integration. However, we noted that the number of clones obtained was low compared with the number of RPE cells initially transduced by rAAV. Similar observations were made in vivo in which rAAV-mediated expression in the rat RPE peaked at 2 months but dropped sharply between 2 to 3 months, with an overall decrease of 50% of GFP-positive cells over the 2- to 12-month period [13]. The number of transduced RPE cells remained relatively stable after 12 months [W. Y.
rAAV GENE TRANSFER INTO RPE CELLS
Shen, personal communication]. Promoter shutdown might be involved in this drop in GFP expression, but the low rates of integration in vitro suggest that this gradual decrease in vivo may be associated with the loss of episomal vector genomes, while the lower expression remaining after 12 months represents integrated vector genomes. Malik et al. [39] has recently demonstrated that high, short-term transgene expression in human primary myotubes did not correlate with the level of integrated forms of rAAV but was related to the appearance of double-stranded episomal genomes. Nevertheless, there is still an apparent discrepancy between the effectiveness of rAAV integration in vivo and in vitro, with more RPE cells stably transduced in vivo than in culture. The level of integration observed in vitro here might have been influenced by RPE cell proliferation and the use of selective pressure, resulting in the rapid and premature elimination of cells that still contain silent, single-stranded rAAV genomes. Consequently, this in vitro experiment may not fully reflect the situation in vivo. To establish the molecular state of rAAV in nonproliferating RPE cells in vivo and to investigate any involvement of promoter shutdown in signal loss, further studies using techniques capable of separating and analyzing single cells are required. In this study, we have shown that RPE cells were more efficiently transduced by rAAV than other cells in the absence of external biological, chemical, or genotoxic stresses, and that this higher permissiveness might be due to the presence of intrinsic cellular factors that enhances rAAV genome conversion. Understanding the underlying mechanisms responsible for the high transduction rate in RPE cells in vitro may lend a greater insight into the processes occurring in RPE cells in vivo. This knowledge will help to optimize the utility of rAAV vectors in RPE-specific gene therapy and, more importantly, may help the development of new strategies to enhance rAAV delivery into other less permissive cell types.
gene transfer by an adeno-associated virus (AAV) vector. Brain Res. 713, 99 –107. 4.
Flotte, T. R., Afione, S. A., Conrad, C., McGrath, S. A., Solow, R., Oka, H., Zeitlin, P. L., Guggino, W. B., and Carter, B. J. (1993). Stable in vivo expression of the cystic fibrosis transmembrane conductance regulator with an adeno-associated virus vector. Proc. Natl. Acad. Sci. USA 90, 10613–10617.
5.
Ali, R. R., Reichel, M. B., De Alwis, M., Kanuga, N., Kinnon, C., Levinsky, R. J., Hunt, D. M., Bhattacharya, S. S., and Thrasher, A. J. (1998). Adeno-associated virus gene transfer to mouse retina. Hum. Gene Ther. 9, 81– 86.
6.
Rolling, F., Shen, W. Y., Tabarias, H., Constable, I., Kanagasingam, Y., Barry, C., and Rakoczy, P. (1999). Evaluation of adeno-associated virus-mediated gene transfer into the rat retina by clinical fluorescence photography. Hum. Gene Ther. 10, 641– 648.
7.
Snyder, R. O., Spratt, S. K., Lagarde, C., Bohl, D., Kaspar, B., Sloan, B., Cohen, L. K., and Danos, O. (1997). Efficient and stable adeno-associated virus-mediated transduction in the skeletal muscle of adult immunocompetent mice. Hum. Gene Ther. 8, 1891–1900.
8.
Halbert, C. L., Standaert, T. A., Aitken, M. L., Alexander, L. E., Russell, D. W., and Miller, A. D. (1997). Transduction by adenoassociated virus vectors in the rabbit airway: Efficiency, persistence and readministration. J. Virol. 71, 5932–5941.
9.
Bartlett, J. S., Samulski, R. J., and McCown, T. J. (1998). Selective and rapid uptake of adeno-associated virus type 2 in brain. Hum. Gene Ther. 9, 1181–1186.
10.
Bennett, J., Maguire, A. M., Cideciyan, A. V., Schnell, M., Glover, E., Anand, V., Aleman, T. S., Chirmule, N., Gupta, A. R., Huang, Y., Gao, G. P., Nyberg, W. C., Tazelaar, J., Hughes, J., Wilson, J. M., and Jacobson, S. G. (1999). Stable transgene expression in rod photoreceptors after recombinant adeno-associated virus-mediated gene transfer to monkey retina. Proc. Natl. Acad. Sci. USA 96,9920 –9925.
11.
Bennett, J., Duan, D., Engelhardt, J. F., and Maguire, A. (1997). Real-time, noninvasive in vivo assessment of adenoassociated virus-mediated retinal transduction. Invest. Ophthalmol. Vis. Sci. 38, 2857–2963.
12.
Flannery, J. G., Zolotukhin, S., Vaquero, M. I., LaVail, M. M., Muzyczka, N., and Hauswirth, W. W. (1997). Efficient photoreceptor-targeted gene expression in vivo by recombinant adenoassociated virus. Proc. Natl. Acad. Sci. USA 94, 6916 – 6921.
13.
Rolling, F., Shen, W. Y., Barnett, N. L., Tabarias, H., Kanagasingam, Y., Constable, I., and Rakoczy, P. E. (2000). Long-term real-time monitoring of adeno-associated virus-mediated gene expression in the rat retina. Clin. Exp. Ophthalmol. 28, 382– 386.
14.
Rakoczy, P. E., Sarks, S. H., Daw, N., and Constable, I. J. (1999). Distribution of cathepsin D in human eyes with or without age-related maculopathy. Exp. Eye Res. 69, 367–374.
15.
Spilsbury, K., Garrett, K. L., Shen, W. Y., Constable, I. J., and Rakoczy, P. E. (2000). Overexpression of vascular endothelial growth factor (VEGF) in the retinal pigment epithelium leads to the development of choroidal neovascularization. Am. J. Pathol. 157, 135–144.
16.
Gu, S., Thompson, D., Srikumari, C., Lorenz, B., Finckh, U., Nicoletti, A., Murthy, K. R., Rathmann, M., Kumaramanickavel, G., Denton, M. J., Gal, A. (1997). Mutations in RPE65 cause autosomal recessive childhood-onset severe retinal dystrophy. Nat. Genet. 17, 194 –197.
17.
Morimura, H., Fishman, G. A., Grover, S. A., Fulton, A. B., Berson, E. L., and Dryja, T. P. (1998). Mutations in the RPE65 gene in patients with autosomal recessive retinitis pigmentosa
This study was partially supported by Meditech Research Limited and the Wellcome Trust (E.B.).
REFERENCES 1.
Ponnazhagan, S., Mukherjee, P., Wang, X. S., Qing, K., Kube, D. M., Mah, C., Kurpad, C., Yoder, M. C., Srour, E. F., and Srivastava, A. (1997). Adeno-associated virus type 2-mediated transduction in primary human bone marrow-derived CD34 ⫹ hematopoietic progenitor cells: Donor variation and correlation of transgene expression with cellular differentiation. J. Virol. 71, 8262– 8267.
2.
Xiao, X., Li, J., and Samulski, R. J. (1996). Efficient long-term gene transfer into muscle tissue of immunocompetent mice by adeno-associated virus vector. J. Virol. 70, 8098 – 8108.
3.
McCown, T. J., Xiao, X., Li, J., Breese, G. R., and Samulski, R. J. (1996). Differential and persistent expression patterns of CNS
191
192
18.
19.
20.
21.
22.
23.
24.
25.
26.
27.
28.
29.
LAI ET AL. or leber congenital amaurosis. Proc. Natl. Acad. Sci. USA 95, 3088 –3093. Dunn, K. C., Aotaki-Keen, A. E., Putkey, F. R., and Hjelmeland, L. M. (1996). ARPE-19, a human retinal pigment epithelial cell line with differentiated properties. Exp. Eye Res. 62, 155–169. Davis, A. A., Bernstein, P. S., Bok, D., Turner, J., Nachtigal, M., and Hunt, R. C. (1995). A human retinal pigment epithelial cell line that retains epithelial characteristics after prolonged culture. Invest. Ophthalmol. Vis. Sci. 36, 955–964. da Cruz, L., Rakoczy, P., Perricaudet, M., and Constable, I. J. (1996). Dynamics of gene transfer to retinal pigment epithelium. Invest. Ophthalmol. Vis. Sci. 37, 2447–2454. Zolotukhin, S., Byrne, B. J., Mason, E., Zolotukhin, I., Potter, M., Chestnut, K., Summerford, C., Samulski, R. J., and Muzyczka, N. (1999). Recombinant adeno-associated virus purification using novel methods improves infectious titer and yield. Gene Ther. 6, 973–985. Summerford, C., and Samulski, R. J. (1998). Membrane-associated heparan sulfate proteoglycan is a receptor for adenoassociated virus type 2 virions. J. Virol. 72, 1438 –1445. Sperinde, V., and Nugent, M. A. (1998). Heparan sulfate proteoglycans control intracellular processing of bFGF in vascular smooth muscle cells. Biochemistry 37, 13153–13164. Ausubel, F., Brent, R., Kingston, R., Moore, D., Seidman, J., Smith, J., and Struhl, K. (1995). “Current Protocols in Molecular Biology,” Wiley, Cambridge. Sanlioglu, S., Duan, D., and Engelhardt, J. F. (1999). Two independent molecular pathways for recombinant adeno-associated virus genome conversion occur after UV-C and E4orf6 augmentation of transduction. Hum. Gene Ther. 10, 591– 602. Callen, D. F., Baker, E., Eyre, H. J., Chernos, J. E., Bell, J. A., and Sutherland, G. R. (1990). Reassessment of two apparent deletions of chromosome 16p to an ins(11;16) and a t(1;16) by chromosome painting. Ann. Genet. 33, 219 –221. Hansen, J., Qing, K., Kwon, H. J., Mah, C., and Srivastava, A. (2000). Impaired intracellular trafficking of adeno-associated virus type 2 vectors limits efficient transduction of murine fibroblasts. J. Virol. 74, 992–996. Fisher, K. J., Gao, G. P., Weitzman, M. D., DeMatteo, R., Burda, J. F., and Wilson, J. M. (1996). Transduction with recombinant adeno-associated virus for gene therapy is limited by leadingstrand synthesis. J. Virol. 70, 520 –532. Ferrari, F. K., Samulski, T., Shenk, T., and Samulski, R. J. (1996). Second-strand synthesis is a rate-limiting step for efficient transduction by recombinant adeno-associated virus vectors. J. Virol. 70, 3227–3234.
Received October 18, 2000 Revised version received February 13, 2001 Published online June 8, 2001
30.
Rutledge, E. A., Halbert, C. L., and Russell, D. W. (1998). Infectious clones and vectors derived from adeno-associated virus (AAV) serotypes other than AAV type 2. J. Virol. 72, 309 –319.
31.
Chiorini, J. A., Kim, F., Yang, L., and Kotin, R. M. (1999). Cloning and characterization of adeno-associated virus type 5. J. Virol. 73, 1309 –1319.
32.
Qing, K., Khuntirat, B., Mah, C., Kube, D. M., Wang, X. S., Ponnazhagan, S., Zhou, S., Dwarki, V. J., Yoder, M. C., and Srivastava, A. (1998). Adeno-associated virus type 2-mediated gene transfer: Correlation of tyrosine phosphorylation of the cellular single-stranded D sequence-binding protein with transgene expression in human cells in vitro and murine tissues in vivo. J. Virol. 72, 1593–1599.
33.
Qiu, J., Handa, A., Kirby, M., and Brown, K. E. (2000). The interaction of heparin sulfate and adeno-associated virus 2. Virology 269, 137–147.
34.
Sanlioglu, S., Benson, P. K., Yang, J., Atkinson, E. M., Reynolds, T., and Engelhardt, J. F. (2000) Endocytosis and nuclear trafficking of adeno-associated virus type 2 are controlled by Rac1 and phosphatidylinositol-3 kinase activation. J. Virol. 74, 9184 –9196.
35.
Bennett, J., Wilson, J., Sun, D., Forbes, B., and Maguire, A. (1994). Adenovirus vector-mediated in vivo gene transfer into adult murine retina. Invest. Ophthalmol. Vis. Sci. 35, 2535– 2542.
36.
Sanlioglu, S., Benson, P., and Engelhardt, J. F. (2000). Loss of ATM function enhances recombinant adeno-associated virus transduction and integration through pathways similar to UV irradiation. Virology 268, 68 –78.
37.
Duan, D., Sharma, P., Dudus, L., Zhang, Y., Sanlioglu, S., Yan, Z., Yue, Y., Ye, Y., Lester, R., Yang, J., Fisher, K. J., and Engelhardt, J. F. (1999). Formation of adeno-associated virus circular genomes is differentially regulated by adenovirus E4 ORF6 and E2a gene expression. J. Virol. 73, 161–169.
38.
Philip, R., Brunette, E., Kilinski, L., Murugesh, D., McNally, M. A., Ucar, K., Rosenblatt, J., Okarma, T. B., and Lebkowski, J. S. (1994). Efficient and sustained gene expression in primary T lymphocytes and primary and cultured tumor cells mediated by adeno-associated virus plasmid DNA complexed to cationic liposomes. Mol. Cell. Biol. 14, 2411–2418.
39.
Malik, A. K., Monahan, P. E., Allen, D. L., Chen, B. G., Samulski, R. J., and Kurachi, K. (2000). Kinetics of recombinant adeno-associated virus-mediated gene transfer. J. Virol. 74, 3555–3565.
Kinetics of Efficient Recombinant Adeno-Associated Virus Transduction in Retinal Pigment Epithelial Cells Yvonne K. Y. Lai,* Fabienne Rolling,† Elizabeth Baker,‡ and Piroska E. Rakoczy 1,* *Centre for Ophthalmology and Visual Science, University of Western Australia, Nedlands, Western Australia, Australia, †Lions Eye Institute, Nedlands, Western Australia; and ‡Molecular Cytogenetics, Women’s and Children’s Hospital, Adelaide, South Australia
The aim of this study was to investigate the premise that retinal pigment epithelial (RPE) cells are more permissive to recombinant adeno-associated virus (rAAV) transduction than other cells. We investigated the kinetics and mechanisms of rAAV transduction in RPE cells and found that the transduction efficiencies of cultured RPE cells HRPE51 and ARPE19 were significantly higher than those of 293 (P < 0.008) and HeLa (P < 0.025) cells. In addition, RPE cells reached maximum transduction efficiency at a much lower m.o.i. (m.o.i. 10) than 293 cells (m.o.i. 25). Competition experiments using 1 g/ml heparin inhibited the high level of transduction in RPE cells by 30%, but additional heparin failed to reduce rAAV transduction further. Southern hybridization of low-molecular-weight DNA from transduced RPE cells indicated that 42% of single-stranded rAAV DNA was translocated into the nucleus by 2 h postinfection. By 6 h postinfection, double-stranded rAAV DNA was observed, which coincided with the onset of transgene expression. Southern and fluorescence in situ hybridization of total genomic DNA indicated that long-term transgene expression in RPE cells was maintained by the integration of rAAV into the cellular chromosome. Together, these results suggest that the high permissiveness of RPE cells is not related to the presence of heparan sulfate receptors or nuclear trafficking but may be due to an enhanced rate of second-strand synthesis and that integration in RPE cells is responsible for longterm transgene expression. © 2001 Academic Press Key Words: gene transfer; adeno-associated virus; retinal pigment epithelium; second-strand synthesis; virus integration.
INTRODUCTION
Adeno-associated virus (AAV)-mediated transduction and transgene expression have been reported in a 1
To whom correspondence and reprint requests should be addressed at Centre for Ophthalmology and Visual Science, University of Western Australia, 2 Verdun Street, Nedlands, WA 6009, Australia, Fax: ⫹61 8 9381 0700. E-mail: [email protected]. 0014-4827/01 $35.00 Copyright © 2001 by Academic Press All rights of reproduction in any form reserved.
wide range of cells and tissues, both in vitro and in vivo. These include hematopoietic cells, muscle, liver, lung, central nervous system, and retina [1– 6]. However, the level of transduction varies considerably among different cell types. For example, several studies have shown extensive transgene expression at the injection site following intramuscular injection of recombinant AAV (rAAV) vectors [2, 7]. In contrast, efficiency of rAAV gene transfer to the airway epithelia is much lower [8]. Furthermore, even though rAAV transduction in the brain is highly efficient overall, the degree of transgene expression is not uniform between the different cell types and regions of the brain [3, 9]. Similarly, wide variations in transduction efficiency of rAAV have been observed within the retina. Ganglion cells, inner retinal cells, photoreceptors, and retinal pigment epithelial (RPE) cells have been reported as expressing transgenic protein following rAAV delivery in vivo [5, 10 –12]. In our hands, transduction of rAAVCMV-gfp in the rat eye was detected mostly in the RPE, which showed intense and persistent green fluorescent protein (GFP) expression in up to 90% of the subretinally injected area [6, 13]. The ability to target and transfer genes into RPE cells is important, as this cell layer is believed to play an important role in ocular diseases such as age-related macular degeneration and several types of retinal degenerations [14 –17]. This preferential targeting of the RPE in vivo by rAAV has led us to question whether RPE cells are more permissive to rAAV transduction than other cell types and to explore some possible mechanisms that might facilitate these high transduction levels. Additionally, we examined the longevity of transgene expression and the molecular state of the rAAV vector persisting in RPE cells in vitro. METHODS Cells and viruses. Three types of human RPE cell cultures were used: two established cell lines, ARPE19 and D407 [18, 19], and a low-passage primary culture, HRPE51, established in our laboratory [20]. Two other cell lines were used as controls, Ad5-transformed kidney 293 and human cervical carcinoma HeLa. All cells were
184
185
rAAV GENE TRANSFER INTO RPE CELLS cultured in Dulbecco’s modified essential medium (DMEM; GIBCO BRL, Grand Island, NY) supplemented with 10% fetal calf serum (FCS; GIBCO), 100 units/ml penicillin (GIBCO), and 100 g/ml streptomycin (GIBCO). The cells were maintained at all times at 37°C and 5% CO 2. The identity of RPE cultures was confirmed by positive cytokeratin AE1/AE3 immunochemistry (DAKO, Carpinteria, CA). Vector rAAVCMV-gfp, based on AAV type 2, carries the enhanced GFP and neomycin resistance (neo r) genes, which are driven by the cytomegalovirus (CMV) and SV40 promoters, respectively [6]. In all of the following experiments except the heparin competition assay, CsCl gradient-purified rAAVCMV-gfp was used. The virus was prepared and titer determined in 293 cells as previously described [6]. For the heparin competition assay, rAAVCMV-gfp was purified on an iodixanol (Optiprep; Nycomed, Oslo, Norway) density step gradient, followed by heparin affinity chromatography (HiTrap Heparin; Amersham Pharmacia, Uppsala, Sweden) [21]. Purified rAAVCMV-gfp was heat-inactivated at 56°C for 1 h. Absence of helper Ad contamination was confirmed by the lack of cytopathic effect when 293 cells were exposed to the same amounts of rAAVCMV-gfp as those used in this study. Ad5dl309, used in the transduction experiments, was produced in 293 cells and the titer determined by limiting dilution. Transduction assays. Virus infections were done in DMEM supplemented with 2% FCS. Cells (10 5) were infected with rAAVCMVgfp at a multiplicity of infection (m.o.i.; transducing units per cell) of 0.1, 1, 10, or 25 and in the presence or absence of Ad5dl309 (m.o.i. of 120 pfu/cell). After a 24-h infection, cells were harvested, washed with phosphate-buffered saline (PBS), and fixed in 2% paraformaldehyde. The extent of rAAVCMV-gfp transduction was then measured using flow cytometry on a FACSCalibur machine (Becton Dickinson, Franklin Lakes, NJ). Heparin inhibition. The assay was performed as previously described [22] with minor modifications. Briefly, heparin column-purified rAAVCMV-gfp at a m.o.i. of 2 was incubated with 1, 15, and 30 g/ml soluble heparin (Sigma Chemical Co., St. Louis, MO) in DMEM for 1 h at 37°C. The heparin–rAAVCMV-gfp mixture was then allowed to adsorb to 10 5 ARPE19 cells (grown in 6-well plates) for 1 h at 37°C. The cells were washed with PBS, overlaid with DMEM supplemented with 2% FCS, harvested after 40 h, and analyzed by flow cytometry. Cellular fractionation. HRPE51, 293, and HeLa cells grown in 6-well plates were infected with rAAVCMV-gfp at a m.o.i. of 1. After a 2-h incubation, the cells were trypsinized, mixed with FCS to inactivate the trypsin, washed with PBS, and pelleted. Nuclear and cytoplasmic fractions were obtained using a previously described protocol [23]. Briefly, the cell pellet was resuspended in 200 l homogenization buffer (10 mM Hepes, pH 7.9, 10 mM KCl, 0.1 mM EDTA, 0.1 mM EGTA, 1 mM DTT, 0.5 mM PMSF) and vortexed. After 15 min incubation on ice, 12.5 l 10% Nonidet P-40 was added and vortexed vigorously for 10 s. Test of cell lysis by trypan blue exclusion [24] typically gave a yield of around 95% stained nuclei. The nuclei were then pelleted at 10,000g for 30 s. The cytoplasmic fraction contained within the supernatant was collected and the nuclear pellet remaining was washed with homogenization buffer. Low-molecular-weight or Hirt DNA was isolated from both cytoplasmic and nuclear fractions using previously described protocols [25]. Hirt DNA was resolved by agarose gel electrophoresis and analyzed by Southern hybridization. The relative densities of the autoradiographic signals were approximated using the Scion Image program (Scion Corp, Frederick, MD). Analysis of second-strand synthesis. HRPE51 and 293 cells seeded in 6-well plates were infected with rAAVCMV-gfp at a m.o.i. of 1. The 293 cells were also coinfected with Ad at a m.o.i. of 120. At 6, 24, and 48 h postinfection, the cells were trypsinized, washed with PBS, and pelleted. Hirt DNA was isolated and the final DNA pellet resuspended in 20 l Tris–EDTA buffer, pH 8.0. Hirt DNA was
resolved by agarose gel electrophoresis and then subjected to Southern hybridization. Generation and selection of transduced clonal populations. RPE D407 cells were grown to 80% confluence in six-well plates. Each well either was infected with 10 8 rAAVCMV-gfp particles (0.166 fmol) or was transfected with 0.1 g pEGFP-N1 (32 fmol; Clontech, Heidelberg, Germany). pEGFP-N1 contains the same GFP–neo r construct as rAAVCMV-gfp but without any AAV sequences. Plasmid transfection was achieved using Fugene 6 reagent (Roche, Indianapolis, IN). Twenty-four hours after infection or transfection, RPE cells from each well were passaged into a 15-cm-diameter plate. G418 (GIBCO), a neomycin analogue, was added the following day to a final concentration of 0.7 mg/ml. The numbers of neo r and GFP-expressing clones were determined by counting ten randomly selected fields under fluorescence microscopy. After 2 weeks of selection, resistant RPE clones generated by rAAVCMV-gfp were isolated using cloning rings and cultured as described above. Analysis of stable vector DNA. Hirt DNA and total genomic DNA were isolated from individual rAAV-transduced RPE clones [24, 25]. Approximately 20 g total genomic DNA was digested with restriction endonucleases, and the molecular structure of vector DNA present was then analyzed by agarose gel electrophoresis and Southern hybridization. Southern hybridization. Hirt DNA and digested genomic DNA were resolved on 0.8% agarose gels and transferred to nylon membranes (Gene Screen Plus; NEN Life Science, Boston, Massachusetts) by Southern blotting. The membranes were hybridized with a [ 32P]-dCTP-labeled GFP–neo r probe and exposed to radiographic films. Fluorescence in situ hybridization. Metaphase chromosomes were harvested from cells in culture, hypotonically treated with 0.075 M KCl, fixed in methanol–acetic acid (3:1), spread onto cleaned glass slides, and left to dry overnight. The probe specific for the GFP–neo r genes was nick-translated with biotin-14 – dATP and hybridized in situ to the metaphases at a final concentration of 15 ng/l. The fluorescence in situ hybridization (FISH) method was as described in Callen et al. [26], except that chromosomes were stained before analysis with both propidium iodide (as a counterstain) and diaminophenylindole (DAPI) (for chromosome identification). Images of metaphase preparations were visualized on an AX70 microscope (Olympus, Australia) fitted with a cooled CCD camera, filter wheel, and the ChromoScan image collection and enhancement system (Applied Imaging Corp., Newcastle Upon Tyne, UK). FISH signals and the DAPI banding pattern were merged for figure preparation.
RESULTS
Kinetics of RPE Cell Transduction in Vitro The ability of rAAV to transduce RPE cells was determined by infecting two human RPE cell lines and a low-passage RPE primary culture with a rAAV vector carrying the GFP gene, rAAVCMV-gfp. The number of fluorescent cells, as measured by FACS analysis, showed that the level of transduction by rAAVCMV-gfp in RPE cells was high at a m.o.i. of 1 (Fig. 1A). All three RPE cultures showed efficiencies greater than 31% and up to 46% for HRPE51. When compared with 293 cells, the transduction efficiencies of all three RPE cultures were significantly higher than that of the 293 cells (P ⬍ 0.05). This higher efficiency was especially evident with HRPE51 and ARPE19, which had at least 22.6% more GFP-expressing cells than 293 (P ⬍ 0.008
186
LAI ET AL.
Effect of Heparin Competition on rAAV Transduction in RPE Cells Preincubation of rAAV with heparin has been shown to completely block rAAV transduction in HeLa cells, suggesting that heparan sulfate proteoglycans (HSPG) are primary cellular receptors for AAV [22]. To determine whether the presence of HSPG receptors could be responsible for the high level of rAAV transduction in RPE cells, ARPE19 cells were infected with rAAVCMV-gfp that had been preincubated with increasing concentrations of heparin. Use of heparin column-purified rAAVCMV-gfp ensured that all viral particles were able to bind to the added heparin. In the presence of 1 g/ml heparin, rAAVCMV-gfp transduction was inhibited by 30% (Fig. 2A). However, higher concentrations of heparin could not block transduction any further, indicating that saturation of rAAV particles with heparin failed to completely inhibit rAAV binding and entry into RPE cells. Subcellular Distribution of Viral DNA in RPE Cells
FIG. 1. The efficiency of rAAVCMV-gfp transduction in RPE cells in vitro. (A) HRPE51, ARPE19, D407, control HeLa, and 293 cells were infected with rAAVCMV-gfp at a m.o.i. of 1. (B) HRPE51, ARPE19, and 293 cells infected with rAAVCMV-gfp at increasing m.o.i.s (0.1, 1, 10, 25). The transduction efficiency was determined as the percentage of GFP-expressing cells at 24 h postinfection (A and B, n ⫽ 2, mean value ⫾ SD).
and P ⬍ 0.007, respectively). The transduction efficiencies of HRPE51 and ARPE19 were also significantly higher than that of HeLa cells at this low m.o.i. (P ⬍ 0.02 and P ⬍ 0.03, respectively). Further transduction of HRPE51, ARPE19, and 293 cells with increasing m.o.i.s showed that the profiles of transgene expression were markedly different between RPE and 293 cells (Fig. 1B). At a high m.o.i. of 25, the upper transduction limit was reached by all three cultures with efficiencies around 80%; however, the rate at which the cultures attained this maximum differed significantly. At a m.o.i. of 10, RPE cells were already maximally transduced, but only 60% of 293 cells expressed GFP. A slower rate of increase was also obtained with HeLa cells (data not shown).
Impaired intracellular trafficking of viral DNA to the nucleus was recently identified as a barrier to efficient rAAV transduction [27]. To investigate the rate of rAAV DNA entry into the RPE cell nucleus, low-molecular-weight or Hirt DNA was isolated from the cytoplasmic and nuclear fractions of HRPE51, 293, and HeLa cells transduced by rAAVCMV-gfp at 2 h postinfection. The Hirt DNA was then analyzed by Southern hybridization using a GFP–neo r-specific probe, and the relative amounts of input viral DNA present were approximated using Scion Image software. By the 2 h time point, approximately 42% of single-stranded rAAV DNA were detected in the nuclei of HRPE51 cells (Fig. 2B, lanes 1 and 2). Analysis of 293 and HeLa cells showed that 40% and 47%, respectively, of the signal were found in the nuclei of these cells (lanes 3– 6). Comparison of these relative densities indicated that similar amounts of viral DNA were present in the nuclear fractions of the three cultures, suggesting that rAAV DNA successfully entered into the nuclei of HRPE51 cells and at rates similar to those of HeLa and 293 cells. Second-Strand Synthesis in RPE Cells Second-strand synthesis has been proposed as the major obstacle to efficient rAAV transduction and may be enhanced by coinfection with helper Ad [28, 29]. As mentioned, in the absence of Ad RPE cells were more readily transduced by rAAVCMV-gfp than 293 cells (Figs. 1A and 1B). However, the addition of Ad to these cells did not influence rAAV transduction in the ARPE19 cell line or low-passage primary HRPE51 (Fig. 3A). On the other hand, the number of 293 cells
rAAV GENE TRANSFER INTO RPE CELLS
187
formation of double-stranded vector genomes, Hirt DNA was isolated from HRPE51, and 293 plus Ad cells at different time points after infection. As expected, Ad coinfection of 293 enhanced conversion of singlestranded rAAVCMV-gfp to double-stranded DNA forms, with little input viral DNA detected after 24 h (Fig. 3B, lanes 5 and 6). Surprisingly, a similar pattern was observed for HRPE51 without Ad, in which most of the single-stranded vector had been converted to duplex replication forms (monomer and dimer; Rfm and Rfd) at 24 h, and by 48 h very little input singlestranded DNA was present (lanes 2 and 3). In contrast, a significant amount of single-stranded viral DNA remained after 24 h in HeLa cells (lane 7). Furthermore, as early as 6 h postinfection, double-stranded forms of the vector genome were already clearly visible in HRPE51 (lane 1), which coincided with the onset of GFP expression in some cells (data not shown). A band at around 2 kb was also obtained in all samples. The identity of this fragment is unclear but may represent circular intermediates. Nevertheless, in HRPE51 and 293 cells, this fragment was significantly lower in intensity than the replicative DNA forms. From this result, it appears that second-strand synthesis is enhanced in HRPE51 cells, similar to that in Ad-infected 293 cells, suggesting that this faster conversion to transcriptionally active forms may account for the higher level of transduction seen in RPE cells. Comparison of rAAVCMV-gfp Transduction and Plasmid Transfection in RPE Cells
FIG. 2. (A) Infection of ARPE19 with rAAVCMV-gfp preincubated with the indicated concentrations of soluble heparin. Each point represents the average percentage GFP-expressing cells relative to the maximum level obtained without addition of heparin (n ⫽ 2, mean value ⫾ SD). (B) Southern blot analysis of Hirt DNA from nuclear (Nuc.) and cytoplasmic (Cyt.) fractions of rAAVCMV-gfptransduced HRPE51, 293, and HeLa cells at 2 h postinfection. The position of single-stranded viral DNA (ssDNA) is indicated.
expressing GFP was significantly increased in the presence of Ad, especially at lower rAAV m.o.i.s of 0.1 to 10 (P ⬍ 0.03). Coinfection with Ad increased the transduction efficiency of 293 cells to levels similar to those obtained in ARPE19 and HRPE51, thereby indicating that the low efficiency in 293 observed previously in the absence of Ad was due to this rate-limiting step of second-strand synthesis. To determine whether there was a correlation between the level of expression in RPE cells and the
After establishing that RPE cells are readily transduced, the stability of rAAV expression in RPE cells was compared with plasmid transfections. The RPE D407 cell line grown in 6-well plates was either transduced with rAAVCMV-gfp or transfected with plasmid pEGFP-N1, which contains the same GFP–neo r construct as rAAVCMV-gfp. After several days of G418 treatment, many single cell-derived clones were evident in both samples. However, after 10 days under continual G418 selection, the majority of pEGFP-N1transfected clones had started to die. In contrast, most of the rAAVCMV-gfp-transduced clones remained resistant to G418 and continued to express GFP. In total, approximately 6% GFP-expressing clonal populations were obtained with rAAVCMV-gfp infection. This was seven times more than that generated from pEGFP-N1 transfection, even though a 200-fold excess molar quantity of plasmid was transfected per well. Of the rAAVCMV-gfp clones generated, seven were isolated, expanded, and maintained without G418 selection. After 50 passages, they were still stably expressing GFP (data not shown).
188
LAI ET AL.
FIG. 3. (A) Infection of HRPE51, ARPE19, and 293 cells with rAAVCMV-gfp at increasing m.o.i.s (0.1, 1, 10) in the absence (⫺) or presence (⫹) of Ad5dl309 (m.o.i. 120) (n ⫽ 2, mean value ⫾ SD). (B) Southern blot analysis of Hirt DNA from rAAVCMV-gfp-transduced HRPE51 minus Ad (lanes 1–3) and 293 plus Ad (m.o.i. 120) cells (lanes 4 – 6) at 6, 24, and 48 h postinfection. Hirt DNA from transduced HeLa cells at 24 h postinfection is also shown (lane 7). Bands corresponding to ssDNA and duplex Rfm and Rfd forms are indicated.
Molecular State of rAAVCMV-gfp in RPE Cells in Vitro To determine whether the greater stability of rAAVCMV-gfp in RPE cells could be attributed to a
particular molecular state of the vector (episomal or integrated), both Hirt and total genomic DNA were isolated from the seven rAAVCMV-gfp-transduced clones and analyzed for vector presence.
rAAV GENE TRANSFER INTO RPE CELLS
189
FIG. 4. Integration of rAAVCMV-gfp in transduced RPE D407 clones. (A) Southern blot analysis of total genomic DNA extracted from seven GFP ⫹–neo r clones and a parent untransduced D407 cell line. DNA was digested with EcoRV (I) and HindIII (II). Each lane corresponds to the clone number. (B) FISH analysis of integrated vector sequence in four representative GFP ⫹–neo r clones: clone 2 (I), clone 4 (II), clone 5 (III), and clone 6 (IV). A FISH signal was detected in each clone (white dot, arrow), which indicated integration into chromosomes 7, 17, 6, and 1, respectively.
Southern blot analysis of Hirt DNA from the D407 clones at early passage 6 did not reveal any free monomers or dimers of the rAAVCMV-gfp genome (data not shown). Instead, hybridizing bands were detected in total genomic DNA, suggesting vector integration into the RPE chromosome. Digestion with EcoRV (no restriction site within the vector; Fig. 4AI) generated single bands greater than 12 kb, suggesting integration at only one site per clone. Restriction with other enzymes indicated that each integration event involved only a single vector copy (data not shown). Additional analysis with HindIII-digested DNA (single cut within vector; Fig. 4AII) revealed different-sized
bands pointing to distinct viral-cellular junctions between clones. This suggested that these integration events might have occurred randomly at various chromosomal sites. FISH analysis was used to further investigate the sites of vector integration. Hybridization of metaphase chromosomes with biotinylated GFP–neo r probes revealed the presence of vector-specific FITC signal at a single chromosomal site in each of the seven clones. Analysis of these chromosomes revealed that rAAVCMV-gfp had integrated into different sites for each clone. Figure 4B shows FISH results from four clones with vector integration into chromosomes 7, 17,
190
LAI ET AL.
6, and 1 for clones 2, 4, 5, and 6, respectively, strongly supporting random vector integration. DISCUSSION
Previous studies have shown that rAAV can mediate efficient gene transfer into the retina [5, 10 –12]. The retina is composed of several different cell layers, which vary in permissiveness to rAAV transduction. These differences in cell permissiveness in vivo are consistent with previous in vitro studies that demonstrated varying rAAV transduction efficiencies between different cell types [30, 31]. These studies reported more efficient transduction in 293 and HeLa cells than in many other cell lines, such as ovarian CHO-K1, bronchial epithelial IB3, and carcinoma KB and SW480 cells. In the present study, we have shown that RPE cells not only were more permissive to rAAV transduction, but also reached maximum transduction at a much lower m.o.i. than 293 or HeLa cells. This result suggests that rAAV-mediated gene transfer and expression occur much faster in RPE cells than in any of the above cell lines and supports our in vivo observations [6, 13] of highly efficient rAAV uptake in RPE cells. This high permissiveness of RPE cells might be due to various factors, some of which include the presence and abundance of cellular receptors, second-strand synthesis, the dephosphorylated state of singlestranded D-sequence-binding protein (ssD-BP), and nuclear trafficking [22, 27–29, 32, 33]. Summerford and Samulski [22] reported total inhibition of rAAV-lacZ transduction in HeLa cells by heparin competition and thus proposed that HSPGs acted as primary receptors for AAV. Likewise, our results suggested that HSPGs were present on RPE cells and were partly responsible for rAAV binding. However, there is currently some disagreement in the literature regarding the completeness of heparin competition in HeLa cells [22, 33, 34]. We found that saturation of rAAVCMV-gfp particles with heparin could not completely inhibit transduction in RPE cells. This suggests that HSPGs are not responsible for the high permissiveness of RPE cells and supports the findings of Qiu et al. [33], who proposed the existence of another highaffinity receptor for AAV in addition to low-affinity heparan sulfate binding. Nuclear trafficking of viral DNA has also been identified as an obstacle to efficient rAAV transduction [27]. Translocation of rAAV DNA from the cytoplasm to the nucleus was shown to be efficient in 293 and HeLa cells [27]. Based on Southern analysis of Hirt DNA, the amount of single-stranded rAAV DNA detected in the nuclei of RPE cells at 2 h postinfection was similar to those in 293 and HeLa cells. This suggests that intracellular trafficking is neither impaired nor enhanced in
RPE cells and therefore cannot account for the high transduction efficiencies in these cells. It has been well established that coinfection of 293 cells with Ad increases the transduction efficiency of rAAV [28, 29]. However, no increase was observed in RPE cells in the presence of Ad. This lack of an Adinduced increase in transduction efficiency cannot be related to impairment of Ad infection, as RPE cells are known to be highly permissive to Ad infection [20, 35]. It was also interesting to note that the transduction efficiency of RPE cells without Ad was comparable to that of 293 plus Ad. Previous studies have established that this augmentation of rAAV transduction by Ad is due to enhanced second-strand synthesis [28, 29]. To determine if high rAAV transduction in RPE cells was related to enhanced second-strand synthesis, we compared Hirt DNA from transduced HRPE51 and 293⫹Ad cells and found that formation of a doublestranded viral genome in these RPE cells was significantly enhanced to a level similar to that in Ad-infected 293 cells. These data suggest that enhanced viral genome conversion might contribute to the greater permissiveness of RPE cells. Certain elements or cellular factors that promote this conversion may be present in HRPE51 cells. Ali et al. [5] showed that ␥-irradiation improved rAAV-lacZ transduction in the retina. The mechanism involved has not been established but possibly resembles that of UV irradiation, which stimulates formation of viral circular intermediates [36]. However, the negligible amount of circular intermediates observed in this study suggests that the cellular factors intrinsically present in RPE cells may act differently from genotoxic stresses such as ␥- and UV irradiation and the adenoviral E2a gene product [25, 36, 37]. A possible factor contributing to this high permissiveness could be increased levels of dephosphorylated ssD-BP, which has been found to enhance second-strand synthesis [32]. Further studies are needed to see if RPE cells in vivo also behave similarly and to fully elucidate the mechanisms involved, including any other factors that may contribute to this high permissiveness. Consistent with other studies [38], in vitro transduction of RPE cells with rAAV generated more stable clones than transient plasmid pEGFP-N1 transfection. We further established that the stability of expression in rAAV-transduced clones was due to vector integration. However, we noted that the number of clones obtained was low compared with the number of RPE cells initially transduced by rAAV. Similar observations were made in vivo in which rAAV-mediated expression in the rat RPE peaked at 2 months but dropped sharply between 2 to 3 months, with an overall decrease of 50% of GFP-positive cells over the 2- to 12-month period [13]. The number of transduced RPE cells remained relatively stable after 12 months [W. Y.
rAAV GENE TRANSFER INTO RPE CELLS
Shen, personal communication]. Promoter shutdown might be involved in this drop in GFP expression, but the low rates of integration in vitro suggest that this gradual decrease in vivo may be associated with the loss of episomal vector genomes, while the lower expression remaining after 12 months represents integrated vector genomes. Malik et al. [39] has recently demonstrated that high, short-term transgene expression in human primary myotubes did not correlate with the level of integrated forms of rAAV but was related to the appearance of double-stranded episomal genomes. Nevertheless, there is still an apparent discrepancy between the effectiveness of rAAV integration in vivo and in vitro, with more RPE cells stably transduced in vivo than in culture. The level of integration observed in vitro here might have been influenced by RPE cell proliferation and the use of selective pressure, resulting in the rapid and premature elimination of cells that still contain silent, single-stranded rAAV genomes. Consequently, this in vitro experiment may not fully reflect the situation in vivo. To establish the molecular state of rAAV in nonproliferating RPE cells in vivo and to investigate any involvement of promoter shutdown in signal loss, further studies using techniques capable of separating and analyzing single cells are required. In this study, we have shown that RPE cells were more efficiently transduced by rAAV than other cells in the absence of external biological, chemical, or genotoxic stresses, and that this higher permissiveness might be due to the presence of intrinsic cellular factors that enhances rAAV genome conversion. Understanding the underlying mechanisms responsible for the high transduction rate in RPE cells in vitro may lend a greater insight into the processes occurring in RPE cells in vivo. This knowledge will help to optimize the utility of rAAV vectors in RPE-specific gene therapy and, more importantly, may help the development of new strategies to enhance rAAV delivery into other less permissive cell types.
gene transfer by an adeno-associated virus (AAV) vector. Brain Res. 713, 99 –107. 4.
Flotte, T. R., Afione, S. A., Conrad, C., McGrath, S. A., Solow, R., Oka, H., Zeitlin, P. L., Guggino, W. B., and Carter, B. J. (1993). Stable in vivo expression of the cystic fibrosis transmembrane conductance regulator with an adeno-associated virus vector. Proc. Natl. Acad. Sci. USA 90, 10613–10617.
5.
Ali, R. R., Reichel, M. B., De Alwis, M., Kanuga, N., Kinnon, C., Levinsky, R. J., Hunt, D. M., Bhattacharya, S. S., and Thrasher, A. J. (1998). Adeno-associated virus gene transfer to mouse retina. Hum. Gene Ther. 9, 81– 86.
6.
Rolling, F., Shen, W. Y., Tabarias, H., Constable, I., Kanagasingam, Y., Barry, C., and Rakoczy, P. (1999). Evaluation of adeno-associated virus-mediated gene transfer into the rat retina by clinical fluorescence photography. Hum. Gene Ther. 10, 641– 648.
7.
Snyder, R. O., Spratt, S. K., Lagarde, C., Bohl, D., Kaspar, B., Sloan, B., Cohen, L. K., and Danos, O. (1997). Efficient and stable adeno-associated virus-mediated transduction in the skeletal muscle of adult immunocompetent mice. Hum. Gene Ther. 8, 1891–1900.
8.
Halbert, C. L., Standaert, T. A., Aitken, M. L., Alexander, L. E., Russell, D. W., and Miller, A. D. (1997). Transduction by adenoassociated virus vectors in the rabbit airway: Efficiency, persistence and readministration. J. Virol. 71, 5932–5941.
9.
Bartlett, J. S., Samulski, R. J., and McCown, T. J. (1998). Selective and rapid uptake of adeno-associated virus type 2 in brain. Hum. Gene Ther. 9, 1181–1186.
10.
Bennett, J., Maguire, A. M., Cideciyan, A. V., Schnell, M., Glover, E., Anand, V., Aleman, T. S., Chirmule, N., Gupta, A. R., Huang, Y., Gao, G. P., Nyberg, W. C., Tazelaar, J., Hughes, J., Wilson, J. M., and Jacobson, S. G. (1999). Stable transgene expression in rod photoreceptors after recombinant adeno-associated virus-mediated gene transfer to monkey retina. Proc. Natl. Acad. Sci. USA 96,9920 –9925.
11.
Bennett, J., Duan, D., Engelhardt, J. F., and Maguire, A. (1997). Real-time, noninvasive in vivo assessment of adenoassociated virus-mediated retinal transduction. Invest. Ophthalmol. Vis. Sci. 38, 2857–2963.
12.
Flannery, J. G., Zolotukhin, S., Vaquero, M. I., LaVail, M. M., Muzyczka, N., and Hauswirth, W. W. (1997). Efficient photoreceptor-targeted gene expression in vivo by recombinant adenoassociated virus. Proc. Natl. Acad. Sci. USA 94, 6916 – 6921.
13.
Rolling, F., Shen, W. Y., Barnett, N. L., Tabarias, H., Kanagasingam, Y., Constable, I., and Rakoczy, P. E. (2000). Long-term real-time monitoring of adeno-associated virus-mediated gene expression in the rat retina. Clin. Exp. Ophthalmol. 28, 382– 386.
14.
Rakoczy, P. E., Sarks, S. H., Daw, N., and Constable, I. J. (1999). Distribution of cathepsin D in human eyes with or without age-related maculopathy. Exp. Eye Res. 69, 367–374.
15.
Spilsbury, K., Garrett, K. L., Shen, W. Y., Constable, I. J., and Rakoczy, P. E. (2000). Overexpression of vascular endothelial growth factor (VEGF) in the retinal pigment epithelium leads to the development of choroidal neovascularization. Am. J. Pathol. 157, 135–144.
16.
Gu, S., Thompson, D., Srikumari, C., Lorenz, B., Finckh, U., Nicoletti, A., Murthy, K. R., Rathmann, M., Kumaramanickavel, G., Denton, M. J., Gal, A. (1997). Mutations in RPE65 cause autosomal recessive childhood-onset severe retinal dystrophy. Nat. Genet. 17, 194 –197.
17.
Morimura, H., Fishman, G. A., Grover, S. A., Fulton, A. B., Berson, E. L., and Dryja, T. P. (1998). Mutations in the RPE65 gene in patients with autosomal recessive retinitis pigmentosa
This study was partially supported by Meditech Research Limited and the Wellcome Trust (E.B.).
REFERENCES 1.
Ponnazhagan, S., Mukherjee, P., Wang, X. S., Qing, K., Kube, D. M., Mah, C., Kurpad, C., Yoder, M. C., Srour, E. F., and Srivastava, A. (1997). Adeno-associated virus type 2-mediated transduction in primary human bone marrow-derived CD34 ⫹ hematopoietic progenitor cells: Donor variation and correlation of transgene expression with cellular differentiation. J. Virol. 71, 8262– 8267.
2.
Xiao, X., Li, J., and Samulski, R. J. (1996). Efficient long-term gene transfer into muscle tissue of immunocompetent mice by adeno-associated virus vector. J. Virol. 70, 8098 – 8108.
3.
McCown, T. J., Xiao, X., Li, J., Breese, G. R., and Samulski, R. J. (1996). Differential and persistent expression patterns of CNS
191
192
18.
19.
20.
21.
22.
23.
24.
25.
26.
27.
28.
29.
LAI ET AL. or leber congenital amaurosis. Proc. Natl. Acad. Sci. USA 95, 3088 –3093. Dunn, K. C., Aotaki-Keen, A. E., Putkey, F. R., and Hjelmeland, L. M. (1996). ARPE-19, a human retinal pigment epithelial cell line with differentiated properties. Exp. Eye Res. 62, 155–169. Davis, A. A., Bernstein, P. S., Bok, D., Turner, J., Nachtigal, M., and Hunt, R. C. (1995). A human retinal pigment epithelial cell line that retains epithelial characteristics after prolonged culture. Invest. Ophthalmol. Vis. Sci. 36, 955–964. da Cruz, L., Rakoczy, P., Perricaudet, M., and Constable, I. J. (1996). Dynamics of gene transfer to retinal pigment epithelium. Invest. Ophthalmol. Vis. Sci. 37, 2447–2454. Zolotukhin, S., Byrne, B. J., Mason, E., Zolotukhin, I., Potter, M., Chestnut, K., Summerford, C., Samulski, R. J., and Muzyczka, N. (1999). Recombinant adeno-associated virus purification using novel methods improves infectious titer and yield. Gene Ther. 6, 973–985. Summerford, C., and Samulski, R. J. (1998). Membrane-associated heparan sulfate proteoglycan is a receptor for adenoassociated virus type 2 virions. J. Virol. 72, 1438 –1445. Sperinde, V., and Nugent, M. A. (1998). Heparan sulfate proteoglycans control intracellular processing of bFGF in vascular smooth muscle cells. Biochemistry 37, 13153–13164. Ausubel, F., Brent, R., Kingston, R., Moore, D., Seidman, J., Smith, J., and Struhl, K. (1995). “Current Protocols in Molecular Biology,” Wiley, Cambridge. Sanlioglu, S., Duan, D., and Engelhardt, J. F. (1999). Two independent molecular pathways for recombinant adeno-associated virus genome conversion occur after UV-C and E4orf6 augmentation of transduction. Hum. Gene Ther. 10, 591– 602. Callen, D. F., Baker, E., Eyre, H. J., Chernos, J. E., Bell, J. A., and Sutherland, G. R. (1990). Reassessment of two apparent deletions of chromosome 16p to an ins(11;16) and a t(1;16) by chromosome painting. Ann. Genet. 33, 219 –221. Hansen, J., Qing, K., Kwon, H. J., Mah, C., and Srivastava, A. (2000). Impaired intracellular trafficking of adeno-associated virus type 2 vectors limits efficient transduction of murine fibroblasts. J. Virol. 74, 992–996. Fisher, K. J., Gao, G. P., Weitzman, M. D., DeMatteo, R., Burda, J. F., and Wilson, J. M. (1996). Transduction with recombinant adeno-associated virus for gene therapy is limited by leadingstrand synthesis. J. Virol. 70, 520 –532. Ferrari, F. K., Samulski, T., Shenk, T., and Samulski, R. J. (1996). Second-strand synthesis is a rate-limiting step for efficient transduction by recombinant adeno-associated virus vectors. J. Virol. 70, 3227–3234.
Received October 18, 2000 Revised version received February 13, 2001 Published online June 8, 2001
30.
Rutledge, E. A., Halbert, C. L., and Russell, D. W. (1998). Infectious clones and vectors derived from adeno-associated virus (AAV) serotypes other than AAV type 2. J. Virol. 72, 309 –319.
31.
Chiorini, J. A., Kim, F., Yang, L., and Kotin, R. M. (1999). Cloning and characterization of adeno-associated virus type 5. J. Virol. 73, 1309 –1319.
32.
Qing, K., Khuntirat, B., Mah, C., Kube, D. M., Wang, X. S., Ponnazhagan, S., Zhou, S., Dwarki, V. J., Yoder, M. C., and Srivastava, A. (1998). Adeno-associated virus type 2-mediated gene transfer: Correlation of tyrosine phosphorylation of the cellular single-stranded D sequence-binding protein with transgene expression in human cells in vitro and murine tissues in vivo. J. Virol. 72, 1593–1599.
33.
Qiu, J., Handa, A., Kirby, M., and Brown, K. E. (2000). The interaction of heparin sulfate and adeno-associated virus 2. Virology 269, 137–147.
34.
Sanlioglu, S., Benson, P. K., Yang, J., Atkinson, E. M., Reynolds, T., and Engelhardt, J. F. (2000) Endocytosis and nuclear trafficking of adeno-associated virus type 2 are controlled by Rac1 and phosphatidylinositol-3 kinase activation. J. Virol. 74, 9184 –9196.
35.
Bennett, J., Wilson, J., Sun, D., Forbes, B., and Maguire, A. (1994). Adenovirus vector-mediated in vivo gene transfer into adult murine retina. Invest. Ophthalmol. Vis. Sci. 35, 2535– 2542.
36.
Sanlioglu, S., Benson, P., and Engelhardt, J. F. (2000). Loss of ATM function enhances recombinant adeno-associated virus transduction and integration through pathways similar to UV irradiation. Virology 268, 68 –78.
37.
Duan, D., Sharma, P., Dudus, L., Zhang, Y., Sanlioglu, S., Yan, Z., Yue, Y., Ye, Y., Lester, R., Yang, J., Fisher, K. J., and Engelhardt, J. F. (1999). Formation of adeno-associated virus circular genomes is differentially regulated by adenovirus E4 ORF6 and E2a gene expression. J. Virol. 73, 161–169.
38.
Philip, R., Brunette, E., Kilinski, L., Murugesh, D., McNally, M. A., Ucar, K., Rosenblatt, J., Okarma, T. B., and Lebkowski, J. S. (1994). Efficient and sustained gene expression in primary T lymphocytes and primary and cultured tumor cells mediated by adeno-associated virus plasmid DNA complexed to cationic liposomes. Mol. Cell. Biol. 14, 2411–2418.
39.
Malik, A. K., Monahan, P. E., Allen, D. L., Chen, B. G., Samulski, R. J., and Kurachi, K. (2000). Kinetics of recombinant adeno-associated virus-mediated gene transfer. J. Virol. 74, 3555–3565.