Efficient transduction of murine B lymphocytes and B ... - Nature

Gene Therapy (2001) 8, 938–945  2001 Nature Publishing Group All rights reserved 0969-7128/01 $15.00 www.nature.com/gt

RESEARCH ARTICLE

Efficient transduction of murine B lymphocytes and B lymphoma lines by modified adenoviral vectors: enhancement via targeting to FcR and heparancontaining proteins L Li1, TJ Wickham2, and AD Keegan1 1

Department of Immunology, Holland Laboratory, American Red Cross, Rockville, MD; and 2GenVec, Inc. Gaithersburg, MD, USA

Murine lymphocytes are relatively refractory to efficient transfection or retroviral gene transduction. Adenovirus has been used as a vector to transduce a wide variety of cell types. Several advantages of adenoviruses are their ability to transduce non-cycling cells and to transduce the majority of cells in a population. Unfortunately, lymphocytes are not susceptible to infection with conventional adenovirus. Therefore, to express genes efficiently in murine B cells, we tested the ability of genetically modified adenovirus to transduce the ␤-galactosidase gene. We found that adenovirus containing polylysine in the fiber knob was able to efficiently transduce lipopolysaccharide (LPS)-activated splenic B cells and the B lymphoma line M12.4.1; greater than 80% of the cells expressed ␤-galactosidase activity. However, small resting B cells did not express activity unless treated with LPS after infection. This transduction was mediated by inter-

action with charged molecules since heparan-sulfate, and to a lesser degree chondroitan sulfate, inhibited the transduction. In addition, adenovirus containing a FLAG epitope in the fiber protein was used to target the FcR expressed on B cells using an anti-FLAG antibody. In the presence of antiFLAG, the modified adenovirus was able to efficiently transduce LPS-activated B cells and several B cell lymphoma lines. Interestingly, in the absence of anti-FLAG, there was low level transduction in the LPS-blasts and in M12.4.1 that was not inhibited by soluble adenovirus fiber protein or agents that block RGD-integrin interactions. These results demonstrate that modified adenovirus efficiently transduce B lymphoctyes which will be critical for targeting genes to normal or malignant B cells. Gene Therapy (2001) 8, 938– 945.

Keywords: B lymphocytes; adenovirus; transduction

Introduction Normal lymphocytes and primary tumors of the lymphocyte lineage are historically difficult to transfect efficiently. Development of a system to achieve efficient gene expression in normal cells and cancer cells is an important goal with broad research and therapeutic utility in human and animal models. Previous attempts to transduce B and T lymphocytes using retrovirus, DEAE dextran, or liposomes have had limited success. Recent attempts to express genes in activated murine T cells using retrovirus-based approaches1 or gene gun technology2 have resulted in gene expression with an efficiency reported to be between only 10 and 30%.1,2 This relatively low efficiency hinders the ability to express genes in normal lymphocyte populations to test their contribution to normal biological responses without further manipulation. Recent results suggest that the adenovirus may be a suitable vector to express genes in primary lymphoCorrespondence: AD Keegan, Immunology Department, Jerome Holland Laboratory, American Red Cross, 15601 Crabbs Branch Way, Rockville, MD 20855, USA Received 21 November 2000; accepted 6 January 2001

cytes.3–7 Adenovirus infects a host cell by interaction of at least two coat proteins with two cell surface receptors. The fiber protein interacts with the coxsackievirus-adenovirus receptor3 (CAR).8,9 In addition, it has been reported that the fiber protein can interact with the ␣2domain of human major histocompatibility complex (MHC) class I.10,11 After the fiber-mediated attachment of adenovirus to the cell surface, the penton base interacts with ␣v␤3 and ␣v␤5 integrins which in some cell types can enhance/mediate virus internalization through RGD motifs.12–14 Previous hindrances to the use of adenovirus as a vector for targeting genes are the limited pattern of expression and low surface levels of CAR. For example, macrophages, T cells, and B cells express little to no CAR.13–15 Some success in transducing human B-CLL has been obtained if the cells were activated to elevate the ␣v integrins and transduced at a very high multiplicity of infection.3 Over the last several years, numerous approaches have been tried to expand the tropism of adenovirus, including the development of mice expressing CAR as a transgene under the control of a T cell-specific promoter5 and the targeting of adenovirus binding to other receptors.15–24 Adenovirus targeting has been achieved by using bispecific antibodies to redirect virus to alternative cell surface

Adenoviral transduction of murine B cells L Li et al

939

Figure 1 Schematic of Adz, Adz.pK7, Adz.F(FLAG) and Adz.PB(FLAG) virions. The Adz.pK7 schematic depicts the positively charged lysines on the C-terminus of each trimeric fiber coat protein. The Adz.F(FLAG) and Adz.PB(FLAG) schematics depict the expression of FLAG epitopes on the knob of the fiber coat protein and on the penton base protein, respectively.

Figure 2 Transduction of primary B cells or B cell lines by Adz or Adz.pK7. The indicated cells (1–10 × 105 per well) were transduced with increasing concentration of adenovirus (䊐, Adz wild-type, 䊏, Adz.pK7) as described in Materials and methods. ␤-Galactosidase activity was measured by using a chemiluminescence assay. Enzyme activity is reported in RLU. Reported RLU represent the average of duplicate measurements. Similar results were obtained in four independent experiments.

receptors17,24 and to generate recombinant adenovirus that contain defined receptor-binding motifs in the fiber protein or penton base.6,20 For example, a modified adenovirus containing a polylysine stretch at the N-terminus of the fiber protein (Adz.pK7) has been shown to interact with heparan-containing receptors and increase tropism to endothelial cells, smooth muscle cells, fibroblasts, and macrophages.20–23 In addition, this form of adenovirus has recently been shown to transduce human acute myeloid leukemia cells and fresh human myeloma cells.4 The expression of genes in primary murine B lymphocyte populations for basic mechanistic and pre-clinical analyses is an important goal. Therefore, in this study we have tested the efficacy of two different recombinant adenoviruses to transduce the ␤-galactosidase gene into primary murine B cells and murine B cell lymphoma lines. One virus contains a polylysine heparan-binding motif while the other contains the FLAG epitope. We found that the modified adenovirus can efficiently transduce lipopolysaccharide (LPS)-activated B cells and B lymphoma lines, but not resting B cells. Importantly, these modified viruses transduce the majority of the target population, making them desirable vectors to deliver genes of interest.

Results In order to achieve efficient transduction of genes into normal and malignant B cells, we tested the ability of several different versions of modified adenovirus to transduce murine B cells (Figure 1) with cDNA encoding ␤-galactosidase driven by the CMV promoter. Various

concentrations of wild-type adenovirus (Adz) or Adz.pK7 containing polylysine in the fiber protein were used to transduce small resting B cells, B cells activated with LPS, and the B lymphoma cell line M12.4.1 (Figure 2). The epithelial HeLa cell line was used as a positive control for efficient fiber-mediated adenovirus transduction. By analysis of ␤-galactosidase activity in whole cell lysates (Figure 2), the Adz and the Adz.pK7 were equally efficient at transducing HeLa cells reaching 苲106 RLU at 3000 particles per cell, indicating that the polylysine residues on the Adz.pK7 do not increase or decrease the fiber-CAR interactions. However, neither preparation was able to transduce ␤-galactosidase activity in resting B cells over background seen in the mock-transduced sample. Stimulation of resting B cells with PMA to activate the CMV promoter did not alter the level of ␤-galactosidase activity measured. However, treatment of the resting B cells with LPS for 16 h after infection slightly increased the ␤-galactosidase activity from background levels of 6800 relative light units (RLU) to 26 000 RLU using 3000 particles per cell suggesting that some virus was gaining entry. Wild-type Adz virus possessed a modest transduction efficiency for LPS blasts (104 RLU at 3000 particles per cell) and the M12.4.1 lymphoma (105 RLU at 3000 particles per cell). Interestingly, the Adz.pK7 was 10-fold more efficient at transducing the LPS blasts and M12.4.1 than the Adz (105 and 106 RLU at 3000 particles per cell respectively). The efficiency of transduction of M12.4.1 by Adz.pK7 approached that of HeLa cells. Analysis of ␤galactosidase activity at the single cell level in normal B cell populations revealed more detail on the adenoviral Gene Therapy


940

Figure 3 Uniform transduction of LPS-activated B cells. Purified resting B cells or B cells activated with LPS as described in Materials and methods were transduced with media alone (dashed line), Adz.F WT (dotted line), or Adz.pK7 (solid grey line) at 2000 particles per cell. ␤-Galactosidase activity in the cells was detected by the FluoReporter LacZ flow cytometry kit. Untransduced and unstained cells are shown in the solid histogram. Similar results were obtained in three independent experiments.

transduction (Figure 3). In the resting B cell population, there was only a minor subpopulation of cells expressing enzyme activity over background, consistent with the lack of activity in the whole cell lysate assay. Strikingly, in the LPS-activated B cells, the Adz.pK7 transduced more than 85% of the cells while the wild-type virus showed little activity over the mock-transduced control. Greater than 90% of HeLa cells were found to express ␤-

galactosidase activity after transduction with either WT Adz or Adz.pK7 (data not shown). These results indicate that the polylysine-modified adenovirus is able to transduce murine B cell blasts and B cell lymphomas efficiently and is a good candidate as a vector to target specific genes to these populations. To determine whether the enhanced transduction efficiency of Adz.pK7 was due to its ability to interact with negatively charged molecules on the cell surface, blocking studies were performed with heparan sulfate and the less charged molecule chondroitan sulfate (Figure 4). Neither of these proteoglycans had a major effect on the level of ␤-galactosidase activity transduced by Adz.pK7 in HeLa cells. However, heparan sulfate profoundly inhibited transduction of enzyme activity in LPSB cells (70% decrease at 5 ␮g/ml) and the M12.4.1 lymphoma cells (90% decrease). Chondroitan sulfate was also able to inhibit transduction, however, higher concentrations were necessary and the maximal degree of inhibition was less than that seen using heparan sulfate. These results indicate that the Adz.pK7 enhances gene transduction of murine B cells by interaction with negatively charged molecules on the cell surface such as receptors that contain heparan sulfate. B lymphocytes possess large numbers of cell surface receptors for the Fc portion of IgG,25 in particular, they express the Fc␥RII-b1 isoform. While this form of Fc␥R does not participate in receptor-mediated endocytosis via clatherin-coated pits,26 it has been shown to enhance the uptake of antibody coated Toxoplasma gondii.27 We reasoned it may be a good candidate for targeting viral entry into polyclonal B cell populations. Therefore, we next tested whether targeting adenovirus to the B cell Fc receptor would enhance transduction efficiency using Adz.F(FLAG). This engineered virus contains the FLAG epitope in the fiber protein while the Adz.PB(FLAG) contains the FLAG epitope in the penton base (Figure 1). We used a monoclonal anti-FLAG antibody of the IgG1 isotype to target the virus to the Fc receptor. These modified virus were incubated with resting and activated primary B cells, several murine B lymphoma lines (CH31, A20, and M12.4.1) and HeLa cells in the presence or absence of the anti-FLAG antibody (Figure 5). Both the Adz.F(FLAG) and the Adz.PB(FLAG) virus were able to transduce efficiently HeLa cells in the presence or absence of anti-FLAG (106 RLU at 3000 particles per cell).

Figure 4 Specificity of transduction by Adz.pK7. The indicated cells were incubated with Adz.Pk7 (2000 particles per cell) in the presence or absence of the indicated concentrations (␮g/ml) of heparan sulfate (black bars) or chondroitan sulfate (hatched bars) for 1 h at 37°C. The enzyme activity of whole cell lysate was analyzed 24 h after transduction. Gene Therapy


941

Figure 5 Transduction of primary B cells and B lymphoma cell lines by Adz.F(FLAG) or Adz.PB(FLAG). The indicated cells were pre-coated with (䊉) or without (䊊) ␣-FLAG antibody (5 ␮g/ml) for 30 min on ice, and after washing cells, Adz.F(FLAG) or Adz.PB(FLAG) were added and incubated for 1 h at 37°C. The enzyme activity in whole cell lysates was measured 24 h after transduction.

These results indicate that the addition of the FLAG epitope to the fiber or to the penton base neither inhibited nor enhanced the fiber–CAR interaction on HeLa cells. Neither virus was able to transduce resting B cells in the presence or absence of anti-FLAG. However, the ability of these virus to transduce activated B cells and B cell lymphoma lines showed interesting heterogeneity. The Adz.PB(FLAG) virus was not able to transduce these B cell populations in the presence or absence of anti-FLAG even though it was clearly able to transduce HeLa. The Adz.F(FLAG) virus did not transduce the immature B cell lymphoma cell line CH31 (sIgMhi, IgD−) in the presence or absence of anti-FLAG, however, it was able to transduce moderate levels of ␤-galactosidase activity (104 RLU at 3000 particles per cell) in the presence of anti-FLAG to the mature, isotype switched (sIgG+) B cell lymphoma A20. Both the LPS-activated B cells and the M12.4.1 mature phenotype B cell lymphoma line (sIgMlow, sIgD+) showed moderate transduction by the Adz.F(FLAG) virus alone (5 × 104 RLU at 3000 particles per cell). In the presence of the anti-FLAG antibody, transduction of M12.4.1 was increased 10-fold and approached transduction levels seen in HeLa. Addition of anti-FLAG slightly enhanced the transduction efficiency in LPS-activated B cells at lower multiplicity of infection (300–2000 particles per cell). By the single cell assay (Figure 6), 43% of the cells expressed ␤-galactosidase activity with a mean fluorescence intensity of 12.2 in the absence of anti-FLAG when transduced with Adz.F(FLAG), while in the presence of anti-FLAG, 80% of the cells expressed ␤-galactosidase activity with a mean fluorescence intensity of 19.3. Consistant with the results using whole cell lysates (Figure 5), the Adz.PB(FLAG) virus was not able to transduce LPS-activated B cells. All of these B cell populations express similar levels of Fc␥RII based on FACS staining with anti-FcR antibody (data not shown). To determine whether the anti-FLAG-mediated enhancement in transduction was due to Fc␥R on the B cell populations, monoclonal anti-Fc␥RII/III (2.4G2) or normal rabbit serum was used to saturate cell surface Fc receptors before addition of anti-FLAG and adenovirus (Figure 7). Both agents were able to inhibit the anti-FLAG

Figure 6 Effect of anti-FLAG antibody on transduction efficiency of LPSactivated B cells. LPS-B cells were pre-incubated with or without antiFLAG as indicated (5 ␮g/ml). Subsequently, the cells were transduced with 2000 particles per cell of Adz.F(FLAG) or Adz.PB(FLAG). Enzyme activity was measured in single cells by FACS.

enhanced level of ␤-galactosidase activity detected in whole cell lysates down to the levels seen in the absence of anti-FLAG antibody for all three cell types. This is seen most clearly in the A20 cells where the Adz.F(FLAG) virus has a very low efficiency of transduction in the absence of anti-FLAG antibody. These results indicate that targeting to Fc␥R on the surface of B cells can enhance transduction by adenovirus. Both the M12.4.1 B cell lymphoma line and LPS-activated primary B cells showed some sensitivity to transduction by a WT Ad.z and the Adz.F(FLAG) in the absence of anti-FLAG antibody. These results suggest that these cell types possess a cell surface receptor molecule for adenovirus. While lymphocytes are generally considered to be CAR negative,13–15 these populations Gene Therapy


942

Figure 7 Contribution of Fc␥R to anti-FLAG-mediated transduction with Adz.F(FLAG). The indicated cells were pre-incubated with ␣-FcR antibody 2.4G2 (hatched bars) (2.5–20 ␮g/ml), rabbit serum (speckled bars) (2.5–20%) for 30 min on ice. Cells were then cultured in the presence (+) or absence (−) of ␣-FLAG antibody (5 ␮g/ml) for 30 min on ice. After washing, the cells were transduced with Adz.F(FLAG) or Adz.PB(FLAG) with 2000 particles per cell. The ␤-galactosidase activity of whole cell lysate was measured 24 h after transduction.

may express low levels of CAR. In addition, the adenoviral transduction could be mediated by interaction of ␣v integrins with the RGD motifs found in the penton base. Indeed, in addition to expressing Fc␥ receptors, it is reported that normal murine B cells express ␣v integrins28 which can act as adenovirus coreceptors.12 To determine the level of expression of ␣v integrin on the surface of resting murine B cells and LPS blasts we used the anti-␣v integrin antibody, anti-CD51 (Figure 8). These results show that resting B cells express ␣v integrins and that LPS activation increases expression by four- to five-fold. To examine the role of fiber–CAR and RGD–␣v integrin interactions in murine B cell transduction, we tested whether soluble fiber protein or RGD peptides could inhibit adenovirus transduction of LPS blasts and M12.4.1 cells as compared with HeLa cells (Figure 9). Transduction of HeLa cells by Adz.F(FLAG) was dramatically inhibited by soluble fiber protein, but not by the RGE or RGD peptides indicating that transduction of HeLa cells is largely determined by fiber–CAR interaction. However, neither soluble fiber nor RGD peptide inhibited the transduction of LPS-B cells or M12.4.1 cells even though these cells express ␣v integrins based on positive staining with anti-CD51 antibody (Figure 8 and data not shown). Similar results were obtained when a blocking anti-CD51 antibody was included in the transduction assay (data not shown). These results suggest that neither fiber-CAR nor RGD–␣v integrin interaction contribute substantially to the transduction of the LPS-activated B cells or M12.4.1 cells.

Discussion In this report, we have shown that modified adenovirus efficiently transduces normal murine B cell blasts and several murine B lymphoma lines. One lymphoma line M12.4.1 was transduced by the polylysine-modified virus as efficiently as the adenovirus permissive Hela cell line. The finding that the vast majority of cells are transduced make the modified adenoviral vector a promising approach to studying the importance of signaling molecules in the functions of normal lympocytes. The high level of transduction efficiency (苲85%) in the LPS-activated B cell population will allow for signaling studies to be performed on transduced populations of cells without further manipulation such as cell sorting or cloning. Transduction with the Adz.FpK7 virus achieved a higher level of ␤-galactosidase activity compared to the Adz.F(FLAG) virus; due to this higher level of gene delivery and ease of use, the polylysine-modified adenovirus appears to be the better vector for delivery to normal murine B cell blasts. Resting B cells were not transduced by Adz.FpK7 or Adz.F(FLAG). While the use of PMA to induce expression of ␤-galactosidase from the CMV promoter did not enhance activity, LPS modestly (3.8-fold) enhanced ␤-galactosidase activity. These results indicate that adenovirus can gain entry to resting B cells, but it is not yet clear whether the low level of transduction is due to inefficient adenoviral entry or the lack of expression of critical genes to process the viral DNA and allow transcription from the CMV promoter. Interestingly, a recent

Figure 8 Expression of ␣v integrins on murine B cells. Resting B cells or B cells activated with LPS for 48 h were stained with anti-␣v (anti-CD51)phycoerythrin in the presence of Fc block. The cells were analyzed by FACS. Gene Therapy


943

Figure 9 Contribution of CAR and integrins to adenovirus transduction. LPS-activated B cells, M12.4.1 cells, and Hela cells (1 × 105 per well) were pre-incubated with RGE inactive peptide (1.0 mg/ml), RGD peptide (1.0 mg/ml), soluble fiber protein (3 ␮g/ml) and RGD plus fiber protein for 30 min on ice as indicated. Transduction with Adz.F(FLAG) was performed with 2000 particles per cell. ␤-Galactosidase activity of whole cell lysates was measured and the RLU are presented as an average of duplicate measurements.

study found that resting B cells were much more resistant to adenoviral transduction than T cells.29 Wan et al29 found that resting T lymphocytes from CAR transgenic mice were transduced by adenoviral vectors using either the CMV promoter or the UbC promoter. However, resting B cells from the CAR transgenic mice were not transduced by either, even though they expressed CAR on the cell surface. Even after activation with mitogen, only 28% of the CAR-transgenic B cells showed evidence of transduction. Therefore, it is likely that the low level of transduction we observed in resting cells cannot be simply explained by our use of the CMV promoter, but rather reflects the inherent difficulty in transducing B cells with adenovirus. Even with the limitations of the CMV promoter in lymphocytes, the polylysine modification examined here allowed for high level of gene expression in ⬎85% of the cells. Both the pK7 virus and the FLAG virus transduced B cells with specificity for the targeted receptor. Transduction with Adz.FpK7 was dependent on negatively charged molecules expressed on B cells since heparan sulfate efficiently blocked transduction. In addition, treatment of a number of cell types with heparinase reduced the ability of pK7 to mediate transduction indicating that the polylysine epitope on Adz.FpK7 targets to heparancontaining proteins on the cell surface.4,15 Mature, naive B cells lack expression of the heparan sulfate proteoglycan (HSPG) syndecan-1.29 However, after activation and differentiation B cells express high levels of syndecan-1 (CD138).30–32 Elevated expression of HSPG such as syndecan-1 could act as a targets for the pK7 virus in LPSstimulated B cells.4 Interestingly, targeting adenovirus to the Fc␥II receptor on B cells using monoclonal IgG1 specific for the FLAG epitope resulted in enhanced transduction of the B lymphomas A20 and M12.4.1 and LPS blasts at low multiplicity of infection. B cells express the b1 isoform of Fc␥RII, which contains 47 more amino acids than the b2 macrophage-specific isoform.25 These additional amino acids have been shown to suppress the clatherin-coated pit-mediated endocytosis of immune complexes containing protein antigen via the Fc receptor.26 This additional sequence contains the immunoreceptor tyrosine-based inhibitory motif (ITIM) that recruits the SH2-domain containing 5′-inositol phosphatase (SHIP).33 Recruitment of SHIP to the cytoplasmic region of Fc␥RII cross-linked to a neighboring antigen receptor, results in inhibition of antigen receptor signaling. Therefore, the Fc␥RII-b1 iso-

form is not considered endocytic, although it is able to patch and cap.26 Nevertheless, we found that targeting adenovirus to this receptor using the anti-FLAG antibody resulted in enhancement of transduction. While the Fc␥RII-b1 isoform does not endocytose immune complexes containing simple antigens,26 it has been shown to mediate internalization of heat-killed Toxoplasma gondii coated with antibody.27 It was proposed that other cell surface molecules might participate in the internalization of this infectious agent. It is possible that the Fc␥RII allows binding of adenovirus to the B cell surface and that some other molecule mediates the internalization. Since the macrophage-specific b2 isoform is more active in its ability to internalize immune complexes,26 this strategy for targeting adenovirus to FcR may be more efficacious in monocyte/macrophage populations. In addition, while we were not able to transduce resting B cells with this approach, another approach using a bispecific antibody recognizing both anti-IgM and anti-FLAG might prove more efficient. Surprisingly, we found that the Adz.PB(FLAG) virus was not able to transduce B cells in the presence or absence of anti-FLAG. The inability of the Adz.PB(FLAG) virus to transduce B cells in the presence or absence of anti-FLAG could be due to the location of the FLAG epitope. Its position in the penton base is near the RGD sequences17 and its insertion deletes several of them. This initially suggested to us that the RGD sequence was important for mediating viral transduction of murine B cells as has been shown for human B cells.3,7,14 However, inhibition studies (shown in Figure 9) did not reveal a major role for RGD–integrin interactions in the transduction of LPS-activated B cells. Other possibilities include (1) the fiber knob of the Adz.PB(FLAG) may cause steric hindrance of the anti-FLAG antibody from interacting effectively with both the FLAG epitope and the Fc receptor17 or (2) the sequence of the penton base near the RGD motifs is important for interactions with unknown cell surface proteins. The LPS blasts were transduced at low levels by WT adenovirus and the Adz.F(FLAG) in the absence of antiFLAG antibody by an unknown mechanism. While these cells express ␣v integrins, the low level of transduction was not substantially inhibited by soluble adenovirus fiber or RGD peptide. However, it is now becoming clear that there may be more players in adenovirus transduction than CAR and ␣v integrins.6 Several studies have suggested that human MHC molecules may also act as Gene Therapy


944

receptors for adenovirus in human cells, and it is possible that murine MHC molecules may be able to mediate lowlevel transduction. In addition to ␣v integrins, other integrins have been shown to interact with the RGD motif found in the penton base including ␣5␤1, ␣m␤2, and ␣6␤1.6 It is also possible that these cells may express a novel receptor for adenovirus that mediates modest transduction levels. Based on these studies and others,3,4 the use of modified adenoviral vectors appears a promising approach for targeting genes into murine B cells. Interestingly, we found that different B lymphoma lines show heterogeneity in their sensitivity to transduction with the Adz.FpK7 and the Adz.F(FLAG) Fc-targeted approach. This differential sensitivity could be potentially important in designing and selecting adenoviral vectors to target specific B lymphomas in humans for use in the gene therapy of cancer and will need further exploration.

Materials and methods Reagents The anti-Fc␥RII and III antibody, 2.4G2,34 and anti-␣v integrin28 CD51 coupled to phycoerythrin were purchased from Pharmingen (San Diego, CA, USA). AntiFLAG antibody, M2, heparan sulfate, chondroitan sulfate, and LPS were obtained from Sigma Chemical (St Louis, MO, USA). RGD and RGE peptides were purchased from Life Technologies (Gaithersburg, MD, USA). Soluble adenovirus fiber protein was produced as described.12 Viruses The E1- and E3-deleted wild-type adenovirus type 5 (Adz.WT) contains the ␤-galactosidase gene under a cytomegalovirus (CMV) promoter; it binds to host cells through the coxsackievirus-adenovirus receptor (CAR). The modified adenovirus Adz.pK7 was derived as described15 from the E4− vector, Adz.11A, to incorporate additional lysine residues on the C-termini of the fiber protein that targets the virus to broadly expressed heparan-containing cellular receptors. Adz.F(FLAG) and Adz.PB(FLAG) were modified to express the FLAG epitope on the fiber coat protein20 or penton base protein,17 respectively, as previously described. These viruses were purified and titered as described.15,17,20 Cell lines and primary murine B cells The murine B lymphoma cell lines, A20, CH31, and M12.4.1 were maintained in RPMI complete medium (CM: RPMI supplemented with 5% FBS, 100 U/ml penicillin-streptomycin, 2 mm glutamine, and 1 × 10−7 m 2mercaptoethanol). HeLa cells (obtained from ATCC, Manassas, VA, USA) were maintained in DMEM-CM without 2-ME and were used as a positive control for transduction. Mouse splenic B cells were purified from C57Bl/6 mice by careful mincing of spleens followed by incubation in anti-Thy 1 serum and complement (Cedarlane Laboratories, Hornby, ON, Canada) to remove T cells. The small, resting B cells were purified by Percoll (Pharmacia Biotech, Uppsala, Sweden) density gradient fractionation. Activated B cells were produced by culturing resting B cells with LPS (50 ␮g/ml) for 48 h in RPMI-CM. Gene Therapy

Adenovirus transduction For adenovirus transduction, 2–10 × 106 cells were suspended in a 96-well U-bottom plate in 200 ␮l of serumfree medium. Adenoviral preparations were added to these cells at varying concentrations (physical particles per cell) and incubated for 1 h at 37°C with plate rocking at 15 min intervals. The cells were washed once with PBS, resuspended in fresh CM, and incubated at 37°C for 24 h. Cells were tested for viability and subsequently analyzed for expression of ␤-galactosidase activity (see below). None of the adenoviral preparations demonstrated toxicity to B cells at the concentrations used in these studies. To neutralize adenovirus binding and transduction, cells were pre-treated with various concentrations of several neutralizing reagents for 30 min at 4°C before adding the virus. To block Fc-mediated uptake of the FLAGtagged virus, antibodies against the FcR (2.4G2) or normal rabbit serum as a source of polyclonal IgG were used. To inhibit uptake due to charge interaction of the pK7 virus, heparan sulfate or chondroitan sulfate were used. A soluble form of the adenoviral fiber coat protein and the peptides RGD and RGE were used to establish the contributions of CAR and integrins in the virus uptake. ␤-Galactosidase assay ␤-Galactosidase activity in whole cell lysates was measured by using a detection kit from TROPIX (Bedford, MA, USA) according to the manufacturer’s protocol. Briefly, the cells were washed in the 96-well plate with PBS before addition of lysis solution. The plates were incubated for 10 min at room temperature, followed by the addition of 100 ␮l of reaction solution containing the substrate 1,2-dioxetane. The plates were incubated for 60–90 min. The RLU were measured by a Microlite luminometer (Dynatech Laboratories, Chantilly, VA, USA). For detection of ␤-galactosidase in single cells, the FluoReporter LacZ Flow Cytometry Kit (Molecular Probes, Eugene, OR, USA) was used according to the suggested protocol. Briefly, single cell suspensions (5– 10 × 105 in 100 ␮l) were prewarmed in a 37°C water bath for 20 min. A pre-warmed solution of fluorescein di-␤-dgalactopyranoside (FDG) was added to each sample. The cells were then incubated for 1 min in a 37°C water bath. Subsequently, ice-cold staining medium containing 1 ␮g/ml propidium iodide and 300 ␮m choloroquine was added to stop FDG loading. The cells were analyzed for levels of green fluoresence by FACScan (Becton Dickinson, San Jose, CA, USA).

Acknowledgements We acknowledge Ms Helen Wang and Ms Xiulan Qi for excellent technical assistance and Dr Robert Hawley for critical reading of the manuscript. This work was supported by United States Public Health Service Grant CA77415 to ADK, the American Red Cross, and GenVec Inc.

References 1 Ouyang W et al. Inhibition of Th1 development mediated by GATA-3 through an IL-4-independent mechanism. Immunity 1998; 9: 745–755.


2 Huang, H, Panettier C, Hu-Li J, Paul WE. Transient transfection of primary helper T cells by particle-mediated gene transfer. J Immunol Meth 1998; 215: 173–177. 3 Cantwell MJ, Sharma S, Friedman T, Kipps TJ. Adenovirus vector infection of chronic lympocytic leukemia B cells. Blood 1996; 88: 4676–4683. 4 Gonzalez RR et al. Transduction of bone marrow cells by the AdZ.F(pK7) modified adenovirus demonstrates preferential gene transfer in myeloma cells. Hum Gene Ther 1999; 10: 2709– 2717. 5 Leon RP et al. Adenoviral-mediated gene transfer in lymphocytes. Proc Natl Acad Sci USA 1998; 95: 13159–13164. 6 Wickham TJ. Targeting adenovirus. Gene Therapy 2000; 7: 110– 114. 7 Huang S et al. Growth arrest of Epstein–Barr virus immortalized B lymphocytes by adenovirus-delivered ribozymes. Proc Natl Acad Sci USA 1997; 94: 8156–8161. 8 Bergelson JM et al. Isolation of a common receptor for coxsakie B viruses and adenoviruses 2 and 5. Science 1997; 275: 1320–1323. 9 Bergelson JM et al. The murine CAR homolog is a receptor for coxsackie B viruses and adenoviruses. J Virol 1988; 72: 415–419. 10 Hong SS. Adenovirus type 5 fiber knob binds to MHC class I ␣2 domain at the surface of human epithelial and B lymphoblastoid cells. EMBO J 1997; 16: 2294–2306. 11 Davidson E, Kirby I, Elliott T, Santis G. The human HLA-A* 0201 allele, expressed in hamster cells, is not a high-affinity receptor for adenovirus type 5 fiber. J Virol 1999; 73: 4513–4517. 12 Wickham TJ, Mathias P, Cheresh DA, Nemerow GR. Integrins, ␣v␤3 and ␣v␤5 promote adenovirus internalization but not virus attachment. Cell 1993; 73: 309–319. 13 Huang S, Endo RI, Nemerow GR. Upregulation of integrins ␣v␤3 and ␣v␤5 on human monocytes and T cells facilitates adenovirus-mediated gene delivery. J Virol 1995; 69: 2257–2263. 14 Nemerow GR, Stewart PL. Role of ␣v integrins in adenovirus cell entry and gene delivery. Microbiol Molec Biol Rev 1999; 63: 725–734. 15 Wickham, TJ, Roelvink PW, Brough DE, Kovesdi I. Adenovirus targeted to heparan-containing receptors increases its gene delivery efficiency to multiple cell types. Nat Biotech 1996; 14: 1570–1573. 16 Wickham TJ, Carrion ME, Kovesdi I. Targeting of adenovirus penton base to new receptors through replacement of its RGD motif with other receptor-specific peptide motifs. Gene Therapy 1995; 10: 750–760. 17 Wickham TJ et al. Targeted adenovirus gene transfer to endothelial and smooth muscle cells by using bispecific antibodies. J Virol 1996; 70: 6831–6838.

18 Douglas JT et al. Targeted gene delivery by tropism-modified adenoviral vectors. Nat Biotech 1996; 14: 1574–1578. 19 Hong SS et al. Enhancement of adenovirus-mediated gene delivery by use of an oligopeptide with dual binding specificity. Hum Gene Ther 1999; 10: 2709–2717. 20 Wickham TJ et al. Increased in vitro and in vivo gene transfer by adenovirus vectors containing chimeric fiber proteins. J Virol 71: 8221–8229. 21 Hidaka C et al. CAR-dependent and CAR-independent pathways of adenovirus vector-mediated gene transfer and expression in human fibroblasts. J Clin Invest 1999; 103: 579–587. 22 Frey BM et al. High efficiency gene transfer into ex vivo expanded human hematopoietic progenitors and precursor cells by adenovirus vectors. Blood 1998; 91: 2781–2792. 23 Bouri K et al. Polylysine modification of adenoviral fiber protein enhances muscle cell transduction. Hum Gene Ther 1999; 10: 1633–1640. 24 Wickham TJ et al. Targeted adenovirus-mediated gene delivery to T-cells via CD3. J Virol 1997; 71: 7663–7669. 25 Ravetch JV, Kinet J-P. Fc receptors. Annu Rev Immunol 1991; 9: 457–492. 26 Amigorena S et al. Cytoplasmic domain heterogeneity and functions of IgG Fc receptors in B lymphocytes. Science 1992; 256: 1808–1812. 27 Joiner KA et al. Toxoplasma gondii: fusion competence of parasitophorous vacuoles in Fc receptor-transfected fibroblasts. Science 1990; 249: 641–646. 28 Gerber DJ et al. Expression of ␣v and ␤3 integrin chains on murine lymphocytes. Proc Natl Acad Sci USA 1996; 93: 14698– 14703. 29 Wan YY et al. Transgenic expression of the coxsackie/ adenovirus receptor enables adenoviral-mediated gene delivery in naive T cells. Proc Natl Acad Sci USA 2000; 97: 13784–13789. 30 Costes V et al. The Mil5 monoclonal antibody (anti-syndecan1) is a reliable marker for quantifying plasma cells in paraffinembedded bone marrow biopsy specimens. Hum Pathol 30: 1405–1411. 31 Sanderson RD, Lalor P, Bernfield M. B lymphocytes express and lose syndecan at specific stages of differentiation. Cell Regul 1989; 1: 27–32. 32 Yamashita Y et al. Syndecan-4 is expressed by B lineage lymphocytes and can transmit a signal for formation of dendritic processes. J Immunol 1999; 162: 5940–5948. 33 Ravetch JV. Fc receptors. Curr Opinion Immunol 1997; 9: 121–125. 34 Unkeless JC. Characterization of a monoclonal antibody directed against mouse macrophage and lymphocyte Fc receptors. J Exp Med 150: 580–596.

945

Gene Therapy