J Biol Chem, Vol. 274, Issue 27, 19087-19094, July 2, 1999

Mannose Polyethylenimine Conjugates for Targeted DNA Delivery into Dendritic Cells^*

Sandra S. Diebold, Margaretha Kursa^§, Ernst Wagner^§, Matt Cotten^¶, and Martin Zenke

From the  Max-Delbrück-Center for Molecular Medicine, Robert-Rössle-Str. 10, D-13092 Berlin, Germany, ^§ Boehringer Ingelheim Austria, Dr. Boehringer-Gasse 5-11, A-1121 Vienna, Austria, and the ^¶ Institute for Molecular Pathology, Dr. Bohr-Gasse 7, A-1030 Vienna, Austria

	ABSTRACT

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Cell surface-bound receptors represent suitable entry sites for gene delivery into cells by receptor-mediated endocytosis. Here we have taken advantage of the mannose receptor that is highly expressed on antigen-presenting dendritic cells for targeted gene transfer by employing mannosylpolyethylenimine (ManPEI) conjugates. Several ManPEI conjugates were synthesized and used for formation of ManPEI/DNA transfection complexes. Conjugates differed in the linker between mannose and polyethylenimine (PEI) and in the size of the PEI moiety. We demonstrate that ManPEI transfection is effective in delivering DNA into mannose receptor-expressing cells. Uptake of ManPEI/DNA complexes is receptor-specific, since DNA delivery can be competed with mannosylated albumin. Additionally, incorporation of adenovirus particles into transfection complexes effectively enhances transgene expression. This is particularly important for primary immunocompetent dendritic cells. It is demonstrated here that dendritic cells transfected with ManPEI/DNA complexes containing adenovirus particles are effective in activating T cells of T cell receptor transgenic mice in an antigen-specific fashion.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Targeted gene delivery capitalizes on the presence of specific cell surface receptors for DNA uptake into cells by receptor-mediated endocytosis (1-3). Therefore, receptor binding ligands are coupled to polycationic compounds like polylysine (pL)¹ that bind and condense DNA. Following this concept, transferrin polylysine (TfpL)-based gene transfer systems were developed to target transferrin receptor for DNA delivery into cells (1, 4-7). Binding of TfpL/DNA complexes to transferrin receptor causes internalization and DNA uptake into the endosomal compartment (8). To facilitate DNA release from this compartment, endosomolytic agents (such as inactivated adenoviruses) were included in transfection complexes and were demonstrated to effectively enhance gene transfer efficiency (8-10). More recently, transferrin polyethylenimine (TfPEI) conjugates have been synthesized and used for DNA delivery, thereby combining the high intrinsic transfection efficacy of polyethylenimine (PEI) with receptor-targeted gene transfer (11, 12). PEI possesses DNA binding and condensing activity together with a high pH buffering capacity that is believed to protect DNA from degradation and to enhance exit from the endosomal compartment. Accordingly, PEI is effective in gene delivery into a variety of cell types even without the addition of cell binding ligands or endosomolytic agents (13, 14). Here we investigated whether the mannose receptor that is abundantly expressed on dendritic cells (DC) represents a suitable entry site for targeted gene delivery into DC using mannosylated PEI (ManPEI).

DC are professional antigen-presenting cells that occur in peripheral organs like skin, where these cells are exposed to antigens, which they capture and process (15-18). Upon inflammatory stimuli, DC migrate to lymphoid tissue and present processed antigens on major histocompatibility complex (MHC) class I and II molecules to T cells, to elicit an antigen-specific T cell response. Because of their central role in the initiation of primary immune responses, there is high interest in employing DC for immunotherapy of diseases, such as cancer (19-21). Following such approaches, gene-modified DC offer several potential advantages over peptide/protein-pulsed DC. For example, gene-modified DC can be expected to induce T cell responses against multiple and/or undefined epitopes of tumor antigens, possibly in the context of both MHC class I and II, and with any MHC allele. Furthermore, the expression of chemokines and cytokines in DC simultaneously with tumor-specific and/or associated antigens would additionally allow modulation of the immune response. DC and T cell functions are effectively regulated by a variety of cytokines, and local cytokine production by DC might represent an important adjunct for T cell activation in medical therapy, for example in cancer patients who are often immunosuppressed. However, so far the generation of gene-modified immunocompetent DC has remained difficult mainly due to limitations in DNA delivery techniques (12, 21, 22).

DC express high levels of mannose receptor and mannose receptor-related receptor that are used for endocytosis and phagocytosis of a variety of antigens that expose mannose and fucose residues (16, 23-26). DC also express transferrin receptor, albeit at lower levels.² Following ligand binding, internalization, and release of cargo, both mannose and transferrin receptor are recycled and transported back to the cell surface, where they allow repeated internalization of new ligand molecules (1, 4, 27). For this reason, mannose receptor might be similar to transferrin receptor and particularly suited for targeted delivery of DNA into cells by employing synthetic mannose polycation conjugates and using a strategy that was successfully applied before for transferrin receptor. Here we describe the synthesis of ManPEI conjugates and have investigated uptake of ManPEI/DNA transfection complexes and their efficacy in DNA delivery. Based on previous successful enhancement of receptor-mediated gene delivery by endosomolytic agents, ManPEI/DNA complexes containing endosome-disrupting adenovirus particles were also generated and investigated.

	EXPERIMENTAL PROCEDURES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Cells and Cell Culture-- Human DC were generated from peripheral blood monocytes by treatment with granulocyte-macrophage colony-stimulating factor (GM-CSF) and interleukin-4 (IL-4) as described (28, 29). Peripheral blood mononuclear cells were isolated from buffy coat preparations of healthy donors by Ficoll Hypaque centrifugation (density of 1.077 g/cm³; Eurobio, Paris, France) followed by T cell depletion with aminoethylthiouronium bromide (Sigma)-treated sheep red blood cells. The T cell-depleted cell fraction was then depleted of B cells and residual T cells using anti-CD19 and anti-CD2 ferromagnetic beads (Dynabeads M-450 Pan-B (CD19) and Pan-T (CD2); Dynal). The cells obtained were cultured in RPMI 1640 medium (Life Technologies, Inc.) containing 10% inactivated fetal calf serum (Life Technologies, Inc.), 2 mM glutamine, 100 units/ml penicillin and streptomycin in the presence of GM-CSF (500 units/ml; kindly provided by Novartis) and IL-4 (500 units/ml; kindly provided by Schering-Plough) at 37 °C in 5% CO₂ atmosphere for 6-10 days.

Mouse DC were prepared from bone marrow (30). Briefly, bone marrow cells were obtained from hind legs of female C57BL/6 mice and seeded on tissue culture dishes to remove strongly adherent cells. After 1 h, nonadherent cells were recovered and transferred into a new dish. Cultures were grown in RPMI 1640 medium containing 10% inactivated fetal calf serum, 50 µM -mercaptoethanol, 100 units/ml penicillin and streptomycin, and 300 units/ml mouse GM-CSF. At days 1 and 3, nonadherent cells were removed, and adherent cells were further cultured. At day 7, nonadherent and loosely adherent DC were harvested and used for transfection experiments.

Human macrophages were prepared from peripheral blood monocytes essentially as described for human DC (see above) but incubated in culture medium without cytokines for 2-10 days. B cells were obtained by immunomagnetic bead purification (see above) and cultured for 2 days to facilitate the release of beads.

BM2 chicken myeloblasts (31) were grown in Dulbecco's modified Eagle's medium (Life Technologies, Inc.) supplemented with 8% fetal calf serum (Life Technologies, Inc.), 2% chicken serum (Sigma), 10 mM HEPES, and 100 units/ml penicillin and streptomycin.

Plasmid Vectors and Adenovirus-- The following reporter constructs were used: pCluc plasmid (32) encoding the Photinus pyralis luciferase gene and pEGFP-C1 (CLONTECH Laboratories) bearing a mutated variant of green fluorescent protein (GFP); pcDNA3--gal containing -galactosidase cDNA in pcDNA-3 vector (Invitrogen); and pcDNA3-OVA encoding the chicken ovalbumin gene prepared by cloning the EcoRI/XbaI fragment of pGEM-OVA (33) into pcDNA-3. In all constructs, expression of the respective reporter gene is under control of the cytomegalovirus immediate early enhancer/promoter. Plasmid DNA was prepared by alkaline lysis followed by Triton-X114 purification to remove lipopolysaccharide (34).

Active wild type adenovirus type 5 (wtAd) and active or psoralen-inactivated E4-defective Ad5 strain dl1014 (E4Ad; Ref. 35) was used (36, 37). Virus growth, purification, biotinylation of E4Ad, and inactivation with psoralen were performed as described previously (38). The viral stocks were quantified by protein content (1 mg/ml protein = 3.4 × 10¹² viral particles/ml; Ref. 39).

Synthesis of ManPEI Conjugates-- ManPEI conjugates were synthesized similarly as described for mannose polylysine conjugates (40). Man-itc-PEI conjugate had mannose linked to PEI via a phenylisothiocyanate bridge using mannopyranosylphenyl isothiocyanate (Sigma) as coupling reagent. PEI was purified by gel filtration in 250 mM NaCl and subsequent dialysis against water. Coupling was performed by reacting 25 mg of PEI in 0.33 ml of water with 25 mg of mannopyranosylphenyl isothiocyanate in 0.2 ml of dimethyl sulfoxide for at least 1 day, followed by dilution with 4 ml of water and adjustment to 0.5 M sodium chloride, cation exchange chromatography (Bio-Rad Macroprep High S, salt gradient from 0.5 to 3 M sodium chloride) and dialysis against 150 mM sodium chloride. Conjugates were analyzed for content of PEI and mannose, by using ninhydrin assay (41) and a resorcinol sulfuric acid method (42), respectively. Man-itc-PEI²⁵ and Man-itc-PEI⁸⁰⁰ consisted of 25-kDa low molecular mass PEI or 800-kDa high molecular mass PEI (Aldrich) containing mannopyranosylphenyl isothiocyanate/PEI at a 1:1.4 and 1:1.3 weight ratio, respectively. This represents an average modification of every tenth (25 kDa) or ninth (800 kDa) PEI nitrogen with mannose. Man-bio-PEI conjugate was obtained by reductive amination with mannobiose in an analogous fashion as for the lactosylation described in Ref. 32. Reaction of 25 mg of PEI (800 kDa) in 0.5 ml of 250 mM aqueous sodium chloride with 25 mg of mannobiose (Sigma) and reduction with three portions of sodium cyanoborohydride (2 mg, at 1-h intervals), followed by dilution, cationic exchange chromatography, and dialysis (as described above), resulted in a conjugate that contained 800-kDa high molecular mass PEI at a mannobiose/PEI weight ratio of 1:2.

ManPEI Transfection-- ManPEI transfection complexes were generated by mixing 4 µg of plasmid DNA in 300 µl of HEPES-buffered saline (150 mM NaCl, 20 mM HEPES, pH 7.4) with various amounts of ManPEI conjugate in 300 µl of HEPES-buffered saline followed by incubation at room temperature (20 min). For formation of Ad/ManPEI/DNA complexes, an increasing amount of adenovirus particles (see above) was added, and samples were incubated for an additional 20 min. 300 µl of this transfection solution were given to 5 × 10⁵ cells in 500 µl of serum-free culture medium in 24-well plates; all transfections were done in duplicates. Transfection medium was replaced by complete culture medium after 4 h. Reporter gene assays were performed at day 1-2 after transfection.

In ManPEI transfection/blocking experiments, cells were preincubated with various amounts of mannosylated bovine serum albumin (ManBSA, Sigma) in 500 µl of serum-free culture medium for 30 min. Then 300 µl of ManPEI transfection complex was added, and transfection was performed as described above. As control, cells were preincubated with an equivalent amount of unconjugated bovine serum albumin (BSA; Sigma), transfected, and processed accordingly.

Reporter Gene Assays-- For luciferase assays, cells were washed once with phosphate-buffered saline and lysed by three cycles of "freeze and thaw" in 0.25 M Tris buffer, pH 7.5 (43). Luciferase activity of lysate was measured in a Lumat LB9501 (Berthold, Wildbad, Germany) and normalized for protein content. All values represent means of duplicates or of multiple measurements with the S.D. values indicated. GFP expression was detected by fluorescence microscopy (Axiophot, Zeiss) or by flow cytometry using a FACScalibur devise (Beckton Dickinson) and employing propidium iodide (PI) staining for gating on viable cells (see below).

Flow Cytometry-- Surface antigen expression of DC was analyzed by flow cytometry. To block unspecific binding, cells were incubated first in staining buffer (phosphate-buffered saline plus 1% BSA, fraction V, Sigma; 30 min, 4 °C) containing 1% human IgG for human DC (Beriglobin; Behringwerke, Marburg, Germany) and then reacted with unconjugated or fluorescein isothiocyanate (FITC)-labeled mouse monoclonal antibodies (1 h, 4 °C). Samples containing unlabeled antibodies were stained with FITC-conjugated goat anti-mouse antibody (Sigma; 45 min, 4 °C). Cells were washed three times, resuspended in staining buffer and PI (2 µg/ml, Sigma) for gating on viable cells, and analyzed by flow cytometry using a FACScalibur device with CELLQuest software (Becton Dickinson). The antibodies used were as follows: MHC class I (HLA-A, -B, -C, clone G46-2.6; PharMingen), MHC class II (HLA-DQ, clone SPVL3; Immunotech; and HLA-DR, clone CR3/43; DAKO), CD80 (B7/BB1, clone MAB104; Immunotech), CD86 (B70/B7-2, clone 2331; PharMingen), and mannose receptor (clone 19.2; PharMingen) for human DC and MHC class I (H-2D^b, clone KH95; PharMingen), MHC class II (I-A^b,d,q/I-E^d,k; ATCC no. TIB-120), CD80 (B7-1, clone 1G10; PharMingen), and DEC-205 (NLDC-145; ATCC no. HB-290) for analysis of murine cells.

Mannose receptor expression was determined by incubating cells with 1 mg/ml FITC-labeled mannosylated BSA (FITC-ManBSA; Sigma) for 1 h at 37 °C in the presence of 10 mM sodium fluoride to prevent receptor internalization (26). Control cells were processed similarly but incubated at 4 °C; under these conditions, no binding of ManBSA to receptor is found. Cells were then washed twice and resuspended in staining buffer followed by flow cytometry as above. PI (2 µg/ml) was used for gating on viable cells. To block binding of FITC-ManBSA to mannose receptor, cells were preincubated for 20 min with various amounts of unlabeled ManBSA or as a control with unconjugated BSA; FITC-ManBSA was added, and samples were processed as described above.

Cell viability was determined by PI staining of nonviable cells and flow cytometry (see above). The proportion of viable cells was determined with viability of untreated cells set at 100%.

Ovalbumin (OVA)-specific T Cell Activation-- Splenocytes of OT-I mice (44, 45) were prepared, and CD8⁺ T cells were obtained by immunomagnetic bead purification using MACS anti-CD8 Microbeads (Miltenyi Biotec). These T cells express a transgenic T cell receptor that recognizes OVA-(257-264) peptide on H-2K^b and were cocultured with irradiated mouse DC (5000 rads) in 96-well microtiter plates. Transfected DC were used at day 1 after transfection; untreated DC and DC pulsed with 0.5 µM OVA-(257-264) peptide (SIINFEKL) were employed as controls. Further controls were T cells stimulated by phorbol 12-myristate 13-acetate (25 ng/ml; Sigma) and ionomycin (1 µg/ml; Sigma). After 1 day, culture supernatant was harvested and tested by enzyme-linked immunosorbent assay for interleukin-2 (IL-2) production (R&D Systems). At day 5 of coculture, cells were labeled [³H]thymidine (Amersham Pharmacia Biotech; 1 µCi/well) and harvested 6 h later, and [³H]thymidine incorporation was measured in a Microbeta counter (Wallac, Turku, Finland). All values of [³H]thymidine incorporation represent means of triplicates.

	RESULTS

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Gene Delivery by Receptor-mediated Endocytosis via Mannose Receptor-- For generation of human DC, peripheral blood mononuclear cells were cultured in the presence of GM-CSF and IL-4 following standard procedures (28, 29). The cell populations obtained routinely contained 96-98% DC as determined by cell surface marker expression and flow cytometry and were used for transfection at day 7-9 of culture. Mouse DC were obtained from bone marrow cells cultured with GM-CSF for 7 days (30) and used for transfection. By that time, both human and mouse DC exhibited the typical morphology of DC and expressed high levels of MHC class I and class II and of the costimulatory molecules B7.1 and B7.2 (data not shown). Cells were highly active in stimulating T cell proliferation in allogenic mixed leukocyte reactions. As expected, both human DC and macrophages expressed mannose receptor from day 2 to 10 in culture as determined by staining with anti-mannose receptor-specific antibody (Fig. 1A), while the starting cell population was negative (data not shown). In addition, mannose receptor expression on DC was consistently found to be higher than on macrophages. Mannose receptor levels were also detected by binding of FITC-ManBSA, yielding the same result (Fig. 1B). Human B cells were negative. Interestingly, mouse bone marrow-derived DC showed considerably lower mannose receptor levels than human DC yet were competent in taking up transfection complexes via mannose receptor (see below).

View larger version (30K): [in this window] [in a new window]   Fig. 1.   Mannose receptor expression on mouse and human DC and schematic representation of ManPEI conjugates. A, human DC and macrophages (M) were analyzed for mannose receptor expression on day 6 of culture by staining with mannose receptor-specific antibody and analysis by flow cytometry (dark areas). Open areas represent staining with control antibody. B, human DC and macrophages (M; days 7 and 2 of culture, respectively), human B cells, and mouse DC (day 7 of culture) were analyzed for mannose receptor expression by incubation with FITC-labeled mannosylated BSA (FITC-ManBSA) at 37 °C and analysis by flow cytometry (dark areas). Control, incubation with FITC-ManBSA at 4 °C (open areas). C, schematic representation of Man-bio-PEI and Man-itc-PEI conjugates and of Ad/ManPEI complexes containing Man-itc-PEI conjugate. The mannopyranosyl and phenylisothiocyanate linkers present in Man-bio-PEI and Man-itc-PEI, respectively, are indicated. Ad/ManPEI complexes are formed by charged interactions of PEI with adenovirus capsid proteins.

To investigate the efficacy of gene delivery into DC via mannose receptor, several ManPEI conjugates were synthesized. Man-itc-PEI conjugate has a phenylisothiocyanate bridge for linking the mannose moiety to PEI, while Man-bio-PEI consists of the disaccharide mannobiose linked to PEI (Fig. 1C). Conjugates contained either low or high molecular weight PEI (25- and 800-kDa PEI, in the following referred to as PEI²⁵ and PEI⁸⁰⁰, respectively). To generate ManPEI/DNA transfection complexes harboring a luciferase reporter gene, ManPEI conjugate and plasmid DNA were mixed, and complexes were transfected into DC. Cells were analyzed for luciferase activity 1-2 days later. To determine whether incorporation of adenovirus in transfection complexes would increase gene delivery, Ad/ManPEI/DNA transfection complexes were also generated (Fig. 1C). Briefly, ManPEI conjugate and luciferase plasmid DNA were mixed to form ManPEI/DNA complexes, followed by the addition of adenovirus particles that bind to PEI by charged interactions with negative domains on the viral hexon.

For initial studies, we used the BM2 myeloblast cell line that expresses moderate mannose receptor levels (see below) and that can be grown as homogenous cell population to large cell numbers. In these experiments, a conjugate/DNA ratio of 1:1 was found to be optimal, yielding 0.6-1.6 × 10⁸ light units/mg of protein for both Man-itc-PEI and Man-bio-PEI conjugate (Fig. 2A). Unconjugated PEI was about 500-1000-fold less efficient than either ManPEI conjugate.

View larger version (28K): [in this window] [in a new window]   Fig. 2.   ManPEI transfection of Man-bio-PEI and Man-itc-PEI conjugates. A, ManPEI/DNA and PEI/DNA transfection complexes containing 800-kDa PEI and the luciferase reporter gene were transfected into BM2 cells. To determine optimal transfection efficiencies, various conjugate/DNA ratios were analyzed as indicated (1:10, 1:2, 1:1, 2:1, and 10:1 (w/w)). For PEI/DNA, the optimal polycation/DNA ratio of 1:1 (w/w) is shown. Luciferase activity was determined at day 2 after transfection. The activity of Man-itc-PEI and Man-bio-PEI conjugates containing a phenylisothiocyanate and mannopyranosyl linker, respectively, is depicted. B, BM2 cells were transfected as in A and analyzed by flow cytometry for PI staining of nonviable cells on day 1 after transfection. The proportion of viable cells (percentage of total cell number) is shown with viability of untreated cells set at 100%. One representative experiment of three is shown. Gray squares, Man-itc-PEI; black squares, Man-bio-PEI; open circle, PEI. PEI in conjugates and unmodified PEI were 800 kDa. C, human DC were transfected and analyzed for cell viability on day 1 as in B. Essentially the same result was obtained on day 2 after transfection. Gray squares, Man-itc-PEI; black squares, Man-bio-PEI; open circle, PEI.

To investigate a potential cytotoxic effect of the ManPEI conjugates used, the proportion of viable cells following transfection was determined by PI staining and flow cytometry. For BM2 cells, viability decreased with increasing amounts of ManPEI in transfection complexes in a dose-dependent fashion and was at the optimal 1:1 conjugate/DNA ratio, 60.5, 46.7, and 44.5% for Man-itc-PEI, Man-bio-PEI, and unmodified PEI, respectively (Fig. 2B). Interestingly, ManPEI transfection affected the viability of human DC to a lesser extent (86.7, 60.1, and 75.8% for Man-itc-PEI, Man-bio-PEI, and unmodified PEI, respectively). Thus, Man-itc-PEI and Man-bio-PEI behaved very similarly, and Man-itc-PEI conjugate was chosen for further transfection experiments (in the following referred to as ManPEI), since it only marginally affected cell viability.

Next, the influence of low and high molecular mass PEI (25- and 800-kDa PEI, respectively) on transfection efficiency of ManPEI conjugates was investigated. Again, a conjugate/DNA ratio of 1:1 was found to be optimal, resulting in BM2 cells in about 10⁸ light units/mg of protein for both ManPEI²⁵ and ManPEI⁸⁰⁰ (Fig. 3A and data not shown). As expected, unconjugated PEI was less effective than ManPEI. To further extend these results, ManPEI transfection was applied to human DC that were obtained from peripheral blood mononuclear cells by in vitro differentiation in the presence of GM-CSF and IL-4. Fig. 3B shows that ManPEI²⁵ and ManPEI⁸⁰⁰ conjugates were equally efficient in delivery of a luciferase reporter gene into DC, with ManPEI being more potent than unconjugated PEI. Essentially the same result was obtained with bone marrow-derived mouse DC (data not shown). In addition, as expected for transfection of primary cells, the luciferase values obtained were lower than those measured for the BM2 cell line.

View larger version (35K): [in this window] [in a new window]   Fig. 3.   ManPEI transfection of BM2 cells and human DC. A, ManPEI/DNA and PEI/DNA transfection complexes containing low or high molecular weight PEI (25 or 800 kDa, respectively) and the luciferase reporter gene were generated and transfected into BM2 cells. The conjugate/DNA ratio was 1:1 (w/w). ManPEI25 and ManPEI800 containing a phenylisothiocyanate linker (Man-itc-PEI) were used, and cells were analyzed for luciferase activity at day 2 after transfection. B, human DC were transfected with ManPEI/DNA and PEI/DNA complexes as in A and analyzed for luciferase activity at day 1 after transfection. S.D. values of three independent experiments are shown.

Uptake of ManPEI/DNA transfection complexes was clearly receptor-specific. BM2 cells express moderate levels of mannose receptor as demonstrated by binding of FITC-labeled ManBSA and analysis by flow cytometry (Fig. 4A). Furthermore, the addition of unlabeled ManBSA effectively reduced binding of FITC-labeled ManBSA to BM2 cells, while unconjugated BSA did not. Most importantly, ManBSA severely reduced luciferase expression in ManPEI transfection experiments in a dose-dependent fashion (Fig. 4B), indicating that ManBSA blocks uptake of ManPEI/DNA transfection complexes by binding to the limited number of mannose receptor molecules present on the cell surface.

View larger version (21K): [in this window] [in a new window]   Fig. 4.   ManPEI transfection is receptor-specific. A, mannose receptor expression in BM2 cells was determined by incubation with FITC-labeled ManBSA without competitor at 37 °C and analysis by flow cytometry (untreated, dark area). To block receptor-specific binding, cells were preincubated with unlabeled ManBSA (ManBSA) or BSA (BSA) and incubated with FITC-ManBSA at 37 °C followed by flow cytometry (dark areas). As experimental controls, cells were incubated with FITC-ManBSA at 4 °C (open areas). B, ManPEI/DNA transfection complexes (Man-itc-PEI, conjugate/DNA ratio of 1:1 (w/w)) containing 800-kDa PEI and a luciferase reporter gene were transfected into BM2 cells without preincubation (untreated) or after preincubation with 0.1 and 1 mg of ManBSA (ManBSA) to block mannose receptor-mediated DNA uptake. As control, cells were preincubated with unconjugated BSA (BSA). Luciferase activity at day 2 after transfection is shown.

Adenovirus Particles Enhance Gene Expression by ManPEI Transfection-- Previous studies demonstrated that adenovirus can effectively enhance receptor-mediated gene delivery in several systems due to its potent endosomolytic activity (4, 8-10). Therefore, adenoviral particles were incorporated in ManPEI transfection complexes to determine if this would increase the efficacy of ManPEI transfection. Ad/ManPEI/DNA complexes (Fig. 1C) were generated using wtAd and E4Ad and analyzed in transfection experiments. To test for contribution of viral gene expression to transgene expression, active and psoralen-inactivated E4Ad were employed. We demonstrate that in primary mouse and human DC, the addition of wtAd dramatically enhances luciferase expression by 1000- and 10,000-fold, respectively, while in BM2 cells the effect was only moderate (10-fold; Fig. 5). E4Ad was only slightly less efficient than wtAd in both DC and BM2 cells. However, psoralen inactivation of E4Ad clearly compromised adenovirus-augmented luciferase expression in DC, while the decline with psoralen-inactivated virus was only modest for BM2 cells. This might indicate that in DC reporter gene activity is influenced by viral gene expression, presumably via the cytomegalovirus promoter.

View larger version (22K): [in this window] [in a new window]   Fig. 5.   Adenovirus particles effectively enhance the efficacy of ManPEI transfection. A, Ad/ManPEI/DNA transfection complexes containing a luciferase reporter gene and 300 adenovirus particles/cell were generated and transfected into BM2 cells. Psoralen-inactivated or active E4 adenovirus (E4Ad* and E4Ad, respectively) or wtAd was used. Transfection of ManPEI/DNA complexes without adenovirus is shown as control. Man-itc-PEI containing 800-kDa PEI was employed; conjugate/DNA ratio was 1:1 (w/w). Luciferase activity at day 2 after transfection is shown. Human (B) and mouse (C) DC were transfected with Ad/ManPEI/DNA (3000 adenovirus particles/cell) complexes and ManPEI/DNA (control) containing the luciferase reporter gene as in A and analyzed for luciferase activity at day 1 after transfection.

To determine whether adenovirus-augmented ManPEI transfection was receptor-specific, blocking experiments with ManBSA were performed. An increasing number of adenovirus particles in Ad/ManPEI/DNA transfection complexes enhanced luciferase expression in human DC, while this effect was less pronounced in BM2 cells (Fig. 6, A and B). ManBSA effectively competed with Ad/ManPEI/DNA complex binding to mannose receptor and, at low adenovirus particle numbers, reduced luciferase activity by 3- and 100-fold in DC and BM2 cells, respectively. Interestingly, Ad/ManPEI/DNA transfection complexes containing higher numbers of adenovirus particles showed a less severe reduction or no reduction in transgene expression in response to ManBSA. This finding indicates that under these conditions uptake of Ad/ManPEI/DNA complexes predominantly occurs by mechanisms independent of mannose receptor, possible via the adenovirus internalization route. Unconjugated BSA, used as control, left luciferase activity unaffected irrespective of the number of adenovirus particles per transfection complex (data not shown).

View larger version (23K): [in this window] [in a new window]   Fig. 6.   Competition of Ad/ManPEI/DNA transfection by ManBSA. A, Ad/ManPEI/DNA complexes (Man-itc-PEI with 800-kDa PEI; conjugate/DNA ratio 1:1 (w/w); 300, 1000, and 3000 E4 adenovirus particles/cell) were transfected into BM2 cells without preincubation (gray bars) or after preincubation with 1 mg of ManBSA (black bars) to block mannose receptor-mediated DNA uptake. Luciferase activity was determined at day 2 after transfection. B, Ad/ManPEI transfection of luciferase gene into human DC as in A. Luciferase activity was determined at day 1 after transfection.

Finally, to determine the proportion of transgene-expressing cells, ManPEI/DNA transfection complexes containing GFP expression plasmid were generated. BM2 cells were transfected by employing the same transfection conditions as for luciferase vector and then analyzed for GFP expression by flow cytometry. ManPEI²⁵ and ManPEI⁸⁰⁰ conjugate yielded 2.8 and 1.3% GFP-positive cells, respectively, and the proportion of GFP-positive cells was further increased by the presence of adenovirus particles in transfection complexes (12.8 and 4%, respectively; Table I). ManPEI⁸⁰⁰ conjugate containing mannobiose behaved in a very similar manner to the respective ManPEI conjugate containing a isothiocyanate linker (Table I). Increasing the number of adenovirus particles per complex was associated with elevated cytotoxicity and therefore did not augment transgene expression (data not shown).

View this table: [in this window] [in a new window]   Table I GFP expression following ManPEI transfection ManPEI/DNA transfection complexes containing Man-itc-PEI or Man-bio-PEI and 25- or 800-kDa PEI, as indicated, and GFP expression plasmid were transfected into BM2 cells. Transfection complexes contained 0 or 1000 psoralen-inactivated E4 adenovirus particles/cell ( Ad and + Ad, respectively). 2 days after transfection, cells were analyzed for GFP expression by flow cytometry. The proportion of GFP-positive cells is given in percentage of total cell number with S.D. as indicated. As a control, PEI/DNA complexes with 25- or 800-kDa PEI with the optimal polycation/DNA ratio of 1:1 (w/w) are shown.

OVA-specific T Cell Activation Induced by Ad/ManPEI-transfected Mouse DC-- We next determined whether ManPEI-transfected DC are competent in inducing antigen-specific T cell responses. Therefore, mouse DC were transfected with an OVA-encoding expression plasmid using ManPEI or Ad/ManPEI transfection complexes. At day 1 after transfection, DC were irradiated and cocultured with CD8⁺ T cells of OT-I mice, which express a transgenic T cell receptor specific for OVA-(257-264) peptide presented by MHC class I H-2K^b. IL-2 production was determined at day 1 of coculture, and T cell proliferation was measured at day 5. It was found that Ad/ManPEI/OVA-transfected DC initiate an antigen-specific T cell response as evidenced by stimulation of IL-2 production and T cell proliferation (Fig. 7, A and B). ManPEI transfection of OVA cDNA without adenovirus particles did not induce T cell activation, which is probably due to low antigen expression level (see above). As expected, -galactosidase-transfected DC were inactive, as were untreated DC, while peptide-pulsed DC were highly active.

View larger version (23K): [in this window] [in a new window]   Fig. 7.   Ad/ManPEI-transfected mouse DC induce antigen-specific T cell activation. A, IL-2 production of splenic CD8+ OT-I T cells at day 1 of coculture with ManPEI/OVA and Ad/ManPEI/OVA (3000 E4Ad particles/cell) transfected mouse DC (lane 3 and 5, respectively) was determined by enzyme-linked immunosorbent assay. -Galactosidase-transfected DC served as experimental control (lanes 2 and 4). Lane 1, untreated DC; lane 6, T cells only; lane 7, SIINFEKL peptide-pulsed DC; lane 8, phorbol 12-myristate 13-acetate plus ionomycin-treated T cells. B, [3H]thymidine (TdR) incorporation of CD8+ OT-I T cells at day 5 of coculture with ManPEI/OVA- and Ad/ManPEI/OVA-transfected mouse DC as in A. Untreated DC and -galactosidase-transfected DC, T cells only, peptide-pulsed DC, and phorbol 12-myristate 13-acetate plus ionomycin-treated T cells were used as controls as in A. Means of triplicate values of a responder/stimulator ratio of 1:3 are shown.

	DISCUSSION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Here we describe the synthesis of ManPEI conjugates and their analysis for receptor-targeted gene delivery. ManPEI transfection represents a fully synthetic delivery system that capitalizes on gene transfer by receptor-mediated endocytosis via surface-bound mannose receptor that is highly expressed on antigen-presenting DC. By employing ManPEI rather than mannosylated polylysine conjugates, we took advantage of the higher transfection potential of conjugated PEI as observed for transferrin conjugates (TfPEI versus TfpL) in various cell types (11) and also in initial TfPEI transfection studies in DC (12). Mannosylated polylysine conjugates were employed in related studies for targeting mannose receptor in macrophages (40, 46).

Several ManPEI conjugates were synthesized by reductive amination with mannobiose or by coupling with mannosylphenylisothiocyanate to generate Man-bio-PEI and Man-itc-PEI conjugates, respectively. Both conjugates were found to exhibit similar physical properties and transfection potential when tested in mannose receptor-positive BM2 myeloblasts. The influence of low and high molecular mass PEI (25- and 800-kDa PEI, respectively) on ManPEI/DNA transfection was also studied. Again, ManPEI²⁵ and ManPEI⁸⁰⁰ conjugates showed similar transfection efficiencies. Additionally, blocking experiments demonstrated that the uptake of ManPEI/DNA transfection complexes was mannose receptor-specific.

While these studies demonstrated that ManPEI conjugates are effective in gene delivery via mannose receptor, their transfection potential for primary human and mouse DC was found to be rather low. This appears not to be due to differences in mannose receptor levels, which were the same for BM2 cells and mouse DC and even higher in human DC. Importantly, incorporation of adenovirus particles in ManPEI transfection complex dramatically increased transgene expression as observed for both wild type Ad particles as well as replication-deficient E4 Ad particles. This might be due to the fact that, following uptake of such Ad/ManPEI/DNA complexes via mannose receptor, the adenovirus component facilitates DNA release from the endosomal compartment, similar to its action in adenovirus-augmented transferrinfection (4, 8-10). Alternatively adenovirus itself might contribute to uptake of Ad/ManPEI/DNA complexes via the adenovirus infection route. Such an idea would be in line with the finding that Ad/PEI/DNA transfection complexes containing plasmid DNA bound to adenovirus carrier via PEI (36, 37, 47) are effective in delivering genes into DC (48). Furthermore, Ad/ManPEI/DNA transfection complexes containing a high number of adenovirus particles were found to be less affected by blocking the mannose receptor internalization route with ManBSA than complexes not containing adenovirus particles. Thus, Ad/ManPEI/DNA complexes apparently bind to cells and deliver DNA at least in part via the adenovirus moiety of the complex.

It appears therefore that in DC uptake of DNA via the adenovirus internalization route is more effective than via mannose receptor, while for BM2 cells both routes are equally efficient. In DC, the difference between both uptake pathways might be the extent of endosomal degradation that could be particularly high for mannose receptor-targeted complexes. We have attempted to address this question by applying agents that increase the endosomal pH and thereby inhibit lysosomal degradation like chloroquin and monensin (6, 7, 49). However, so far these studies have met with only limited success due to the high unspecific cytotoxicity of the compounds used.

Finally, our study demonstrates that ManPEI/DNA complexes containing adenovirus particles are effective in activating T cells from T cell receptor transgenic mice in an antigen-specific manner. DC transfected with ManPEI/DNA complexes without adenovirus particles were deficient in inducing such a T cell response, which is presumably related to low transgene expression, and higher expression levels might be required to overcome this limitation. Current experiments aim at incorporating synthetic endosome-disruptive influenza peptides (50) in ManPEI transfection complexes to address this question. Such a system would have the advantage of being fully synthetic.

The present study opens the possibility to use ManPEI transfection for gene delivery into DC to study DC function and to develop DC-based approaches of immunotherapy, e.g. of cancer or viral or infectious diseases. Other transfection techniques so far tested for DC were mostly inefficient and associated with high unspecific cytotoxicity and low transgene expression (12, 22). Additionally, DC from peripheral blood monocytes, as used in this study, are largely postmitotic and difficult to infect with recombinant retroviruses that rely on proliferating cells (51, 52). More recently recombinant adenovirus vectors were applied for transduction of DC (12, 22, 53, 54). The ManPEI transfection system described in this paper is particularly versatile and offers several advantages over viral vectors. For example, very large DNA constructs (more than 100 kilobase pairs) can be transfected to ensure long lasting transgene expression (36). More importantly, several plasmid DNAs can be transfected simultaneously to induce and/or modulate immune responses. Current experiments address the question of whether such a modulation of the immune response with gene-modified DC can be achieved by coexpression of the antigen with specific cytokines and chemokines that activate or attract T cells.

	ACKNOWLEDGEMENTS

We thank Novartis (Vienna, Austria) and Schering-Plough, (Kenilworth) for recombinant human GM-CSF and IL-4, respectively. We are most grateful to R. Holzhauser for conjugate synthesis; F. R. Carbone, M. Lutz, and G. Schuler for OT-I mice; T. Pezzutto and J. Westermann for plasmid DNA; and S. M. Kurz for recombinant mouse GM-CSF. We thank T. Blankenstein for helpful discussions, T. Schüler for advice in T cell preparation, C. Esslinger for careful reading of the manuscript, and I. Gallagher for expert secretarial assistance.

	FOOTNOTES

^* This work was supported in part by Deutsche Forschungsgemeinschaft Grant SFB 506 (to M. Z.) and by a grant of the Max-Delbrück-Center Gene Therapy Program (to S. S. D.).The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

To whom correspondence should be addressed: Max-Delbrück-Center for Molecular Medicine, Robert-Rössle-Str. 10, D-13092 Berlin, Germany. Tel.: 49-30-9406-3343; Fax: 49-30-9406-3329; E-mail: zenke{at}mdc-berlin.de.

² S. S. Diebold and M. Zenke, unpublished observations.

	ABBREVIATIONS

The abbreviations used are: pL, polylysine; TfpL, transferrin polylysine; PEI, polyethylenimine; TfPEI, transferrin polyethylenimine; DC, dendritic cell(s); ManPEI, mannosylated polyethylenimine; MHC, major histocompatibility complex; GM-CSF, granulocyte macrophage-colony stimulating factor; IL, interleukin; GFP, green fluorescent protein; OVA, ovalbumin; Ad, adenovirus; wtAd, wild type Ad; ManBSA, mannosylated bovine serum albumin; BSA, bovine serum albumin; PI, propidium iodide; FITC, fluorescein isothiocyanate.

	REFERENCES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

1. Cotten, M., and Wagner, E. (1993) Curr. Opin. Biotechnol. 4, 705-710[CrossRef][Medline] [Order article via Infotrieve]
2. Perales, J. C., Ferkol, T., Molas, M., and Hanson, R. W. (1994) Eur. J. Biochem. 226, 255-266[Medline] [Order article via Infotrieve]
3. Wu, G. Y., and Wu, C. H. (1987) J. Biol. Chem. 262, 4429-4432[Abstract/Free Full Text]
4. Cotten, M., Wagner, E., and Birnstiel, M. L. (1993) Methods Enzymol. 217, 618-644[Medline] [Order article via Infotrieve]
5. Wagner, E., Zenke, M., Cotten, M., Beug, H., and Birnstiel, M. L. (1990) Proc. Natl. Acad. Sci. U. S. A. 87, 3410-3414[Abstract/Free Full Text]
6. Zenke, M., Steinlein, P., Wagner, E., Cotten, M., Beug, H., and Birnstiel, M. L. (1990) Proc. Natl. Acad. Sci. U. S. A. 87, 3655-3659[Abstract/Free Full Text]
7. Cotten, M., Langle-Rouault, F., Kirlappos, H., Wagner, E., Mechtler, K., Zenke, M., Beug, H., and Birnstiel, M. L. (1990) Proc. Natl. Acad. Sci. U. S. A. 87, 4033-4037[Abstract/Free Full Text]
8. Wagner, E., Zatloukal, K., Cotten, M., Kirlappos, H., Mechtler, K., Curiel, D. T., and Birnstiel, M. L. (1992) Proc. Natl. Acad. Sci. U. S. A. 89, 6099-6103[Abstract/Free Full Text]
9. Curiel, D. T., Agarwal, S., Wagner, E., and Cotten, M. (1991) Proc. Natl. Acad. Sci. U. S. A. 88, 8850-8854[Abstract/Free Full Text]
10. Cotten, M., Wagner, E., Zatloukal, K., and Birnstiel, M. L. (1993) J. Virol. 67, 3777-3785[Abstract/Free Full Text]
11. Kircheis, R., Kichler, A., Wallner, G., Kursa, M., Ogris, M., Felzmann, T., Buchberger, M., and Wagner, E. (1997) Gene Ther. 4, 409-418[CrossRef][Medline] [Order article via Infotrieve]
12. Diebold, S. S., Cotten, M., Wagner, E., and Zenke, M. (1998) Adv. Exp. Med. Biol. 451, 449-455[Medline] [Order article via Infotrieve]
13. Boussif, O., Lezoualch, F., Zanta, M. A., Mergny, M. D., Scherman, D., Demeneix, B., and Behr, J. P. (1995) Proc. Natl. Acad. Sci. U. S. A. 92, 7297-7301[Abstract/Free Full Text]
14. Abdallah, B., Hassan, A., Benoist, C., Goula, D., Behr, J. P., and Demeneix, B. A. (1996) Hum. Gene Ther. 7, 1947-1954[Medline] [Order article via Infotrieve]
15. Peters, J. H., Gieseler, R., Thiele, B., and Steinbach, F. (1996) Immunol. Today 17, 273-278[CrossRef][Medline] [Order article via Infotrieve]
16. Cella, M., Engering, A., Pinet, V., Pieters, J., and Lanzavecchia, A. (1997) Nature 388, 782-787[CrossRef][Medline] [Order article via Infotrieve]
17. Austyn, J. M. (1998) Curr. Opin. Hematol. 5, 3-15[Medline] [Order article via Infotrieve]
18. Banchereau, J., and Steinman, R. M. (1998) Nature 392, 245-252[CrossRef][Medline] [Order article via Infotrieve]
19. Girolomoni, G., and Ricciardi-Castagnoli, P. (1997) Immunol. Today 18, 102-104[CrossRef][Medline] [Order article via Infotrieve]
20. Tuting, T., Storkus, W. J., and Lotze, M. T. (1997) J. Mol. Med. 75, 478-491[CrossRef][Medline] [Order article via Infotrieve]
21. Schuler, G., and Steinman, R. M. (1997) J. Exp. Med. 186, 1183-1187[Free Full Text]
22. Arthur, J. F., Butterfield, L. H., Roth, M. D., Bui, L. A., Kiertscher, S. M., Lau, R., Dubinett, S., Glaspy, J., McBride, W. H., and Economou, J. S. (1997) Cancer Gene Ther. 4, 17-25[Medline] [Order article via Infotrieve]
23. Jiang, W., Swiggard, W. J., Heufler, C., Peng, M., Mirza, A., Steinman, R. M., and Nussenzweig, M. C. (1995) Nature 375, 151-155[CrossRef][Medline] [Order article via Infotrieve]
24. Steinman, R. M., and Swanson, J. (1995) J. Exp. Med. 182, 283-288[Free Full Text]
25. Sallusto, F., Cella, M., Danieli, C., and Lanzavecchia, A. (1995) J. Exp. Med. 182, 389-400[Abstract/Free Full Text]
26. Avrameas, A., McIlroy, D., Hosmalin, A., Autran, B., Debre, P., Monsigny, M., Roche, A. C., and Midoux, P. (1996) Eur. J. Immunol. 26, 394-400[Medline] [Order article via Infotrieve]
27. Stahl, P. D. (1992) Curr. Opin. Immunol. 4, 49-52[CrossRef][Medline] [Order article via Infotrieve]
28. Romani, N., Gruner, S., Brang, D., Kampgen, E., Lenz, A., Trockenbacher, B., Konwalinka, G., Fritsch, P. O., Steinman, R. M., and Schuler, G. (1994) J. Exp. Med. 180, 83-93[Abstract/Free Full Text]
29. Sallusto, F., and Lanzavecchia, A. (1994) J. Exp. Med. 179, 1109-1118[Abstract/Free Full Text]
30. Inaba, K., Inaba, M., Romani, N., Aya, H., Deguchi, M., Ikehara, S., Muramatsu, S., and Steinman, R. M. (1992) J. Exp. Med. 176, 1693-1702[Abstract/Free Full Text]
31. Moscovici, C. (1975) Curr. Top. Microbiol. Immunol. 71, 79-101[Medline] [Order article via Infotrieve]
32. Plank, C., Zatloukal, K., Cotten, M., Mechtler, K., and Wagner, E. (1992) Bioconjugate Chem. 3, 533-539[CrossRef][Medline] [Order article via Infotrieve]
33. Boczkowski, D., Nair, S. K., Snyder, D., and Gilboa, E. (1996) J. Exp. Med. 184, 465-472[Abstract/Free Full Text]
34. Cotten, M., Baker, A., Saltik, M., Wagner, E., and Buschle, M. (1994) Gene Ther. 1, 239-246[Medline] [Order article via Infotrieve]
35. Bridge, E., and Ketner, G. (1989) J. Virol. 63, 631-638[Abstract/Free Full Text]
36. Baker, A., Saltik, M., Lehrmann, H., Killisch, I., Mautner, V., Lamm, G., Christofori, G., and Cotten, M. (1997) Gene Ther. 4, 773-782[CrossRef][Medline] [Order article via Infotrieve]
37. Baker, A., and Cotten, M. (1997) Nucleic Acids Res. 25, 1950-1956[Abstract/Free Full Text]
38. Cotten, M., Saltik, M., Kursa, M., Wagner, E., Maass, G., and Birnstiel, M. L. (1994) Virology 205, 254-261[CrossRef][Medline] [Order article via Infotrieve]
39. Lemay, P., Boudin, M. L., Milleville, M., and Boulanger, P. (1980) Virology 101, 131-143[CrossRef][Medline] [Order article via Infotrieve]
40. Erbacher, P., Bousser, M. T., Raimond, J., Monsigny, M., Midoux, P., and Roche, A. C. (1996) Hum. Gene Ther. 7, 721-729[Medline] [Order article via Infotrieve]
41. Cotten, M., Baker, A., Birnstiel, M. L., Zatloukal, K., and Wagner, E. (1996) Current Protocols in Human Genetics , pp. 12.3.1-12.3.33, John Wiley and Sons, Inc., New York
42. Monsigny, M., Petit, C., and Roche, A.-C. (1988) Anal. Biochem. 175, 525-530[CrossRef][Medline] [Order article via Infotrieve]
43. Disela, C., Glineur, C., Bugge, T., Sap, J., Stengl, G., Dodgson, J., Stunnenberg, H., Beug, H., and Zenke, M. (1991) Genes Dev. 5, 2033-2047[Abstract/Free Full Text]
44. Hoquist, K. A., Jameson, S. C., Heath, W. R., Howard, J. L., Bevan, M. J., and Carbone, F. R. (1994) Cell 76, 17-27[CrossRef][Medline] [Order article via Infotrieve]
45. Kurts, C., Heath, W. R., Carbone, F. R., Allison, J., Miller, J. F., and Kosaka, H. (1996) J. Exp. Med. 184, 923-930[Abstract/Free Full Text]
46. Ferkol, T., Perales, J. C., Mularo, F., and Hanson, R. W. (1996) Proc. Natl. Acad. Sci. U. S. A. 93, 101-105[Abstract/Free Full Text]
47. Fasbender, A., Zabner, J., Chillon, M., Moninger, T. O., Puga, A. P., Davidson, B. L., and Welsh, M. J. (1997) J. Biol. Chem. 272, 6479-6489[Abstract/Free Full Text]
48. Diebold, S. S., Lehrmann, H., Kursa, M., Wagner, E., Cotten, M., and Zenke, M. (1999) Hum. Gene Ther. 10, 775-786[CrossRef][Medline] [Order article via Infotrieve]
49. Dean, R. T., Jessup, W., and Roberts, C. R. (1984) Biochem. J. 217, 27-40[Medline] [Order article via Infotrieve]
50. Plank, C., Oberhauser, B., Mechtler, K., Hoch, C., and Wagner, E. (1994) J. Biol. Chem. 269, 12918-12924[Abstract/Free Full Text]
51. Aicher, A., Westermann, J., Cayeux, S., Willimsky, G., Daemen, K., Blankenstein, T., Uckert, W., Dorken, B., and Pezzutto, A. (1997) Exp. Hematol. 25, 39-44[Medline] [Order article via Infotrieve]
52. Westermann, J., Aicher, A., Quin, Z., Cayeux, S., Daemen, K., Blankenstein, T., Dörken, B., and Pezzutto, A. (1998) Gene Ther. 5, 264-271[CrossRef][Medline] [Order article via Infotrieve]
53. Ribas, A., Butterfield, L. H., McBride, W. H., Jilani, S. M., Bui, L. A., Vollmer, C. M., Lau, R., Dissette, V. B., Hu, B., Chen, A. Y., Glaspy, J. A., and Economou, J. S. (1997) Cancer Res. 57, 2865-2869[Abstract/Free Full Text]
54. Song, W., Kong, H. L., Carpenter, H., Torii, H., Granstein, R., Rafii, S., Moore, M. A. S., and Crystal, R. G. (1997) J. Exp. Med. 186, 1247-1256[Abstract/Free Full Text]