Thoracic and Cardiovascular Surgery, University of Texas M.D. Anderson Cancer Center, Houston, Texas

Transcription

1 2000 Nature America, Inc /00/$15.00/ 0 Gene delivery by an epidermal growth factor/dna polyplex to small cell lung cancer cell lines expressing low levels of epidermal growth factor receptor Klaus Stensgaard Frederiksen, 1 Niels Abrahamsen, 1 Richard J. Cristiano, 2 Lars Damstrup, 1 and Hans Skovgaard Poulsen 1 1 Section of Radiation Biology, National University Hospital, Copenhagen, Denmark; and 2 Department of Thoracic and Cardiovascular Surgery, University of Texas M.D. Anderson Cancer Center, Houston, Texas In the present study, we wanted to determine whether efficient gene delivery using an epidermal growth factor (EGF)/DNA polyplex could be accomplished in small cell lung cancer (SCLC) cell lines expressing low EGF receptor (EGFR) levels. EGFR expression levels and transduction efficiencies with polyplexes were examined in five SCLC cell lines and two controls. EGFR expression was examined by binding assays and demonstrated low EGFR levels ranging from 3.6 to 87.4 fmol/mg protein. The SCLC cell lines exhibited high sensitivity to adenovirus infection, which was an important determinant for transduction efficiency when adenovirus was used as an endosomolytic agent. The transduction efficiencies with EGF/DNA polyplexes ranged from 41% 3.5% to 73% 4.6% in the EGFR-positive SCLC cell lines. In the controls lacking EGFRs, only 5% 1.0% and 8% 1.8% of the cells were transduced. Furthermore, the transduction efficiency could be reduced from 50% 4.9% to 18% 1.1% when excess EGF was added to compete with the EGF/DNA polyplexes. In the present study, receptor-targeted gene delivery to SCLC cell lines has been demonstrated for the first time. Our results indicate that even low receptor expression levels in the target cells are sufficient for efficient and specific in vitro gene delivery with EGF/DNA polyplexes. Cancer Gene Therapy (2000) 7, Key words: Epidermal growth factor; DNA polyplex; small cell lung cancer; targeted gene delivery; nonviral vectors. Received October 9, 1998; accepted March 6, Address correspondence and reprint requests to Dr. Hans Skovgaard Poulsen, Finsen Center, 6321, National University Hospital, 9, Blegdamsvej, DK-2100 Copenhagen Denmark. address: skovgaard@rh.dk Small cell lung cancer (SCLC) comprises 25% of all lung cancer cases and is generally treated with different combinations of chemotherapy alone or in combination with radiotherapy. The prognosis for SCLC patients is very poor, with a 5-year survival rate that is 5%. 1 Because of this poor prognosis, there is an urgent requirement for research into the development of new therapies, such as gene therapy for SCLC. From the first clinical trials of gene therapy for cancer, the crucial importance of suitable gene delivery systems has become evident. When the malignancy is disseminated, a delivery system that can specifically target tumor cells throughout the body may be decisive for the successful application of gene therapy for advanced cancer. One way to accomplish targeted gene delivery is by the coupling of receptor-specific ligands to a plasmid containing the therapeutic gene. Gene delivery to specific cell types by receptor-mediated endocytosis of molecular conjugate vectors has been successful with a variety of different ligands, such as transferrin, 2 folate, 3 asialoorosomucoid, 4 and epidermal growth factor (EGF). 5 EGF/ DNA polyplexes consist of molecules of EGF coupled to a plasmid vector via poly-lysine bridges, which form ionic bonds to the negatively charged DNA (Fig 1). The EGF receptor (EGFR) is overexpressed in non-small cell lung cancer (NSCLC) tumors, and efficient gene delivery using EGF/DNA polyplexes has been reported previously in NSCLC cell lines. 5 In the present study, we wanted to determine whether specific and efficient gene delivery with EGF/DNA polyplexes could be accomplished in small cell lung cancer (SCLC) cell lines exhibiting low levels of EGFR expression. The EGF binding capacity was determined in different SCLC cell lines, and cell lines exhibiting different EGFR expression levels were included in EGF/ DNA polyplex transfections to determine the influence of EGFR concentration on gene delivery by EGFRmediated endocytosis. In addition, we examined the susceptibility of the SCLC cell lines to adenovirus infection. Variations in the sensitivity to adenovirus infection among the different cell lines could influence the transfection efficiency with EGF/DNA polyplexes, because replication-deficient adenovirus was used as an agent for endosomal lysis. Using EGF/DNA polyplexes, we were able to obtain highly efficient gene delivery to SCLC cell lines expressing very low levels of EGFR. This is the first 262 : pp

2 FREDERIKSEN, ABRAHAMSEN, CRISTIANO, ET AL: GENE DELIVERY BY AN EGF/DNA POLYPLEX TO SCLC CELL LINES 263 X-100) and incubated for 30 minutes at room temperature. An aliquot of sample was applied from each tube for counting in a Wallac -counter (Turku, Finland). Protein concentration was determined for each sample using the bicinchoninic acid protein kit (Pierce Europe, BA Oud Beijerland, The Netherlands). Nonspecific binding of EGF was determined in the presence of 30,000 pm of unlabeled EGF, and the proportion of nonspecific binding was calculated for each sample. The specific binding was calculated by subtracting nonspecific binding from total binding. The free EGF in the samples was calculated as the total EGF in the sample minus the specific binding. Calculation of the dissociation constant (K d ) and the maximal EGF binding capacity (B max ) was done as described by Scatchard. 13 Figure 1. Representation of an EGF/DNA polyplex. Biotin (B)- coupled EGF and PLL (K) are initially bound to streptavidin. The positively charged poly-lysine (K) bridges form ionic bonds with the negatively charged DNA, whereby the EGF molecules are linked to the DNA. demonstration of receptor-targeted gene delivery to SCLC cell lines. Our findings may have important implications for the proper selection of target receptors for gene delivery by receptor-mediated endocytosis. MATERIALS AND METHODS Cell culture The cell lines used in the experiments were A431, 6 an epidermoid carcinoma; CPH54A, 7 GLC2, 8 DMS114, 9 DMS153, 9 and DMS456, 10,11 small cell lung carcinomas; and SW620, 12 a colon adenocarcinoma. All cell lines, except SW620, were cultured in 175-cm 2 flasks (Nalge Nunc International, Naperville, Ill) at 37 C, 5% CO 2, and 80% humidity. SW620 was cultured in 175-cm 2 flasks at 37 C in atmospheric air. The cell lines were cultured in appropriate medium (Life Technologies, Roskilde, Denmark) containing 10% fetal calf sera (FCS) (Life Technologies) without antibiotics. EGFR binding studies To obtain a quantitative analysis of the EGFR expression levels in the included cell lines, receptor binding studies using 125 I-labeled EGF were performed. Cells were harvested and washed twice in chilled binding buffer with bovine serum albumin (BSA) (128 mm NaCl, 5 mm KCl, 5 mm MgSO 4, 1.2 mm CaCl 2,50mMN-2-hydroxyethylpiperazine-N -2-ethanesulfonic acid (ph 7.2), and 2% BSA) for 30 minutes. After washing, the cells were incubated with a fixed concentration (25 or 100 pm) of 125 I-EGF (New England Nuclear, Boston, Mass) and increasing concentrations (0 30,000 pm) of nonlabeled EGF (Calbiochem, Bad Soden, Germany) in a final volume of 1 ml of binding buffer with BSA. Incubations were carried out at room temperature for 2 hours with overhead turning of tubes. After incubation, cells were washed twice with ice-cold binding buffer without BSA. After the final wash, cells were resuspended in 1 ml of solubilization buffer (128 mm NaCl, 0.25 mm ethylenediaminetetraacetic acid, 0.5 mm tris(hydroxymethyl)aminomethane (ph 7.2), and 1% Triton EGF/DNA polyplex formation The coupling of EGF to DNA was carried out in a stepwise procedure. The first steps involved the coupling of biotinylated EGF to biotinylated poly-l-lysine (PLL). EGF (Biosource International, Camarillo, Calif) and PLL (average molecular weight of 22,500; Sigma, St. Louis, Mo) were biotinylated with biotin hydroxysuccinimide ester (Pierce, Rockford, Ill) as described previously. 5 The biotinylated EGF (1.2 biotin/egf, biotin M) was added dropwise to streptavidin (0.17 M) (Pierce) with vortexing in a sterile 1.5-mL tube. After a 5-minute incubation at room temperature, the biotin-pll (4.5 biotin/pll, NH 2 30 M, biotin M) was then added dropwise to the reaction mixture with vortexing. The molar ratios between EGF and streptavidin were 1:1, and the ratios between EGF and PLL were 4.5:1. After a 5-minute incubation, the resulting EGF/PLL conjugates were mixed with the plasmid and incubated at room temperature for 30 minutes to allow for polyplex formation. In the polyplex formation, the positively charged lysines will interact with the negatively charged plasmid to form electrostatic bonds between the EGF/PLL and the plasmid (Fig 1). The polyplex solution had a final DNA amount of 10 g in 1 ml of 5 mm N-2- hydroxyethylpiperazine-n -2-ethanesulfonic acid buffer (ph 7.0). Physical characteristics such as the size, charge, and stability of the EGF/DNA polyplex have been described previously by Xu et al. 14 The EGF ligand used was modified to contain only one biotin per EGF molecule. This modification has been shown previously not to change the binding affinity of EGF. 15 Adenovirus isolation The replication-defective adenoviruses dl312 (a gift of Dr. Thomas Shenk, Princeton University, Princeton, NJ) and Ad5CMV -gal were propagated on 293 cells and purified and stored using techniques reported previously. 16,17 Viral titer was determined by ultraviolet spectrophotometric analysis. 16,17 Transfection and transduction experiments The examined cell lines were cultured in 12-well dishes (Costar, Cambridge, Mass). At the time of transduction or transfection, the cells were subconfluent. Before the addition of polyplexes to the cells, replication-defective adenovirus (dl312) was mixed with the polyplexes. The amount of added adenovirus was adjusted to equal a multiplicity of infection (MOI) of To each well, 333 L (3 g of DNA) of polyplex reaction was added as well as 500 L of cell culture medium containing 2% FCS. In transduction experiments with the adenovirus vector Ad5CMV -gal, cells were incubated

3 264 FREDERIKSEN, ABRAHAMSEN, CRISTIANO, ET AL: GENE DELIVERY BY AN EGF/DNA POLYPLEX TO SCLC CELL LINES Figure 2. EGF binding to the SCLC cell line GLC2. A Scatchard plot of the binding data (insert) is shown. In the Scatchard plot, the maximal specific binding of EGF (B max ) is given by the x-interception (y 0). The dissociation constant K d is calculated from the slope of the curve (y ) as K d (y ) 1. with 500 L of cell culture medium containing 2% FCS and 1000 virus particles per cell. Cells were incubated for 2 hours at 37 C. After incubation, the medium was gently aspirated and 1 ml of cell culture medium containing 10% FCS was added per well. Cells were incubated at 37 C overnight. For the visualization of -galactosidase ( -gal) activity, cells were fixed and stained with 5-bromo-4-chloro-3-indolyl b-dgalactoside (Promega, Madison, Wis) staining solution. The cells were fixed for 5 minutes in cold 0.5% glutaraldehyde (Sigma). After aspiration of the fixative, cells were incubated with 1 ml of 5-bromo-4-chloro-3-indolyl b-d-galactoside staining solution per well overnight. The percentage of blue-stained cells was determined by counting the cells lying within the grid of a Leitz Wetzlar Periplan 10 ocular (Leica, Wetzlar, Germany) mounted on a reverse microscope. Cells were photographed using a Kodak digital camera (Rochester, NY). RESULTS EGFR binding studies To quantify the number of EGFRs in the SCLC cell lines and controls, binding studies using 125 I-labeled EGF were performed. Cell suspensions were incubated with a fixed concentration of 125 I-EGF and increasing concentrations of unlabeled EGF. An example of EGF binding data and a Scatchard analysis from an SCLC cell line (GLC2) are shown in Figure 2. From the Scatchard analysis of the binding data, it was determined that GLC2 expressed a mean of 87.4 fmol of EGFR per milligram of protein. The binding analysis demonstrated one high-affinity EGF binding site in all the SCLC cell lines tested. The cell lines bound between 3.6 and 87.4 fmol of EGF per milligram of protein, with a K d ranging from * M. The positive control cell line A431 bound 7829 fmol of EGF per mg of protein and had a K d of 3.57 * 10 9 M. The K d of EGF binding in A431 has been determined previously to be 5 *10 9 M. 18 The negative control SW620 showed no evidence of receptorspecific binding of EGF in two independent experiments. In our laboratory, the SCLC cell line DMS456 has been shown previously not to contain any specific EGF binding sites. 19 All the binding data are presented in Table 1. Transfection experiments with EGF/DNA polyplexes To study in vitro gene delivery by EGFR-mediated endocytosis in SCLC cells, transfection experiments using EGF/DNA polyplexes were performed in the SCLC cell lines DMS114, DMS153, DMS456, CPH54A, and GLC2. For visualization of transfection, a plasmid vector containing the Escherichia coli lacz gene under the control of the cytomegalovirus promotor/enhancer 20 was used in the transfection experiments. To avoid lysosomal degradation of internalized EGF/DNA polyplexes, a replication-deficient serotype 5 adenovirus mutant (dl312), 21 was coincubated with the EGF/DNA polyplexes. Table 1. EGF Receptor Binding in SCLC Cell Lines and Controls Cell line Mean values (B max ) Mean values (K d ) DMS fmol/mg protein 1.92 * M DMS fmol/mg protein 3.90 * M CPH54A 31.1 fmol/mg protein 2.46 * M GLC fmol/mg protein 6.78 * M DMS456 No specific binding detected A fmol/mg protein 3.57 * 10 9 M SW620 No specific binding detected

4 FREDERIKSEN, ABRAHAMSEN, CRISTIANO, ET AL: GENE DELIVERY BY AN EGF/DNA POLYPLEX TO SCLC CELL LINES 265 In all SCLC cell lines containing EGFRs, highly efficient gene delivery was accomplished with this vector system (Fig 3, A C and E). The percentage of cells staining positive for -gal activity ranged from 41% 3.5% in GLC2 to 73% 4.6% in CPH54A (Table 2). Initially the optimum polyplex amount was found by incubating the cell lines with increasing concentrations of EGF/DNA polyplexes, and the transfection experiments were performed with the given optimum polyplex concentrations. A polyplex amount of 10 g of DNA per three wells was used in all cell lines, except for DMS114, for which the optimum polyplex concentration was found to be 8 g ofdnaper three wells. The cell lines tested were incubated with an MOI of 3 *10 3. This saturating titer has been found previously to result in efficient gene delivery in NSCLC cell lines using EGF/DNA polyplexes. 5 To study the specificity of the EGF/DNA polyplex uptake, the vector system was tested in the cell lines DMS456 and SW620, which contained no functional EGFRs. The observed transfection efficiency in SW620 was only 5% 1.0%; in DMS456 (Fig 3D), the efficiency was 10% 2.3%. Furthermore, a competition analysis was performed in DMS114 and SW620 by adding an excess of free EGF (30,000 pm) along with the EGF/ DNA complexes. In DMS114, the competition with free EGF reduced the percentage of blue cells from 50% 4.9% to 18% 1.1% (Fig 3, E F). To determine whether this reduction was due to inhibition of polyplex binding to the EGFR, competition with a nonspecific inhibitor was performed. When adding an excess of insulin (60 M), no reduction in transfection efficiency was observed. In CPH54A, the transfection efficiency could only be reduced from 73% 5.0% to 48% 12.2% in competition studies with excess EGF. The low transfection efficiencies in the EGFR-negative controls SW620 and DMS456 were not affected by excess EGF. When EGF/DNA polyplex transfections were performed without functional adenovirus, no blue-staining cells could be observed (data not shown). Transduction experiments with Ad5CMV -gal To examine the susceptibility of the included cell lines to adenovirus infection, transduction experiments with the adenovirus vector Ad5CMV -gal were prepared. Because adenovirus was used as an endosomolytic agent, we wanted to determine whether variations in cell line susceptibility to adenovirus infection could influence the transfection efficiency with EGF/DNA polyplexes. Ad5CMV -gal contains the bacterial -gal gene under the transcriptional control of the cytomegalovirus promoter. The cell lines were infected with Ad5CMV -gal at an MOI of In all of the included SCLC cell lines, a high percentage of the cells could be infected by the adenovirus vector (Table 2). The transduction efficiencies ranged from 42% 1.7% in DMS456 to 99% 0.5% in DMS153. Somewhat lower transduction efficiencies were observed in the control cell lines SW620 and especially A431 (Table 2). In SW620, 27% 9.0% of the cells stained blue, but in A431 the transduction efficiency was only 3% 0.8%. DISCUSSION The evolution of vector systems that specifically target disseminated tumor cells throughout the body has evolved to be one of the major issues concerning the development of gene therapy for advanced cancer. In the present study, important aspects of targeted gene delivery to SCLC through ligand-receptor interactions have been examined. In the first part of the study, we wanted to examine the influence of cellular EGFR expression levels on the transfection efficiency using EGF/DNA polyplexes. Initially the EGFR expression was examined in the included SCLC and control cell lines by binding assays using 125 I-EGF. The SCLC cell lines DMS114, DMS153, CPH54A, and GLC2 all exhibited receptor-specific binding of EGF. These cell lines represented four different levels of EGFR expression, with mean B max values ranging from 3.6 fmol/mg protein in DMS114 to 87.4 fmol per milligram of protein in GLC2 (Table 1). The B max and K d values for EGF binding in the SCLC cell lines were similar to previously published findings of EGFR expression in SCLC cell lines. 19,22 The determined levels of EGFR expression are low compared with NSCLC cell lines. In a study of the EGFR expression in nine NSCLC cell lines, the mean B max for EGF binding was 505 fmol per milligram of protein. 22 The transfection experiments with EGF/DNA polyplexes showed that the observed low levels of EGFR expression were sufficient for highly efficient in vitro gene delivery to SCLC cells (Table 2). The transfection efficiency of 52% 2.1% in DMS114 only expressing 3.6 fmol of EGFR per milligram of protein emphasized this conclusion especially. This finding, together with the observed transfection efficiency of 35% 12.1% in the positive control A431, expressing 7829 fmol of EGFR per milligram of protein, strongly indicates that overexpression of a target receptor is not a prerequisite for efficient gene delivery with complexes of ligand/dna. Furthermore, it stresses the importance of a restricted expression of the target receptor to the target tissue, because even low receptor expression in nontarget tissues may be sufficient for undesirable gene delivery to these tissues. These conclusions may have important implications for the proper selection of suitable target receptors for gene delivery by receptor-mediated internalization. To test that the EGF/DNA polyplex uptake occurred specifically through the EGFR, transfection experiments were performed in the cell lines SW620 and DMS456, which has no specific EGF binding. Only 5% 1.0% and 10% 2.3% of the cells in SW620 and DMS456, respectively, stained positive for -gal activity after transfection with EGF/DNA polyplexes. In comparison with the transfection efficiencies accomplished in the

5 266 FREDERIKSEN, ABRAHAMSEN, CRISTIANO, ET AL: GENE DELIVERY BY AN EGF/DNA POLYPLEX TO SCLC CELL LINES Figure 3. EGF/DNA polyplex-transduced in vitro-cultured cell lines (A F) stained for -gal activity. CPH54A (A), DMS153 (B), GLC2 (C), DMS456 (D), DMS114 (E), all without competition, and DMS114 with 30 M of EGF competition (F). Also see Table 2. EGFR-positive SCLC cell lines, these results indicate that the uptake of EGF/DNA polyplexes is specifically mediated through the EGFR. Nonspecific interactions between the adenovirus and the EGF/DNA polyplex, which could form conjugates available for internalization through the Coxsackie adenovirus receptor (CAR), may explain the low levels of nonspecific transfection in SW620 and DMS456. To further investigate the uptake of EGF/DNA polyplexes, competition studies adding excess free EGF were performed. In SW620 and DMS456, the competition with free EGF did not influence the transfection effi-

6 FREDERIKSEN, ABRAHAMSEN, CRISTIANO, ET AL: GENE DELIVERY BY AN EGF/DNA POLYPLEX TO SCLC CELL LINES 267 Table 2. Transfection Efficiencies in Tested Cell Lines Cell line EGF/DNA polyplexes (percentage of blue cells) Ad5CMV -gal virus (percentage of blue cells) CPH54A 73% 4.6% 94% 2.1% DMS114 52% 2.1% 88% 1.0% DMS153 48% 1.5% 99% 0.5% DMS456 10% 2.3% 42% 1.7% GLC2 41% 3.5% 44% 3.8% A431 35% 12.1% 3% 0.8% SW620 5% 1.0% 27% 9.0% ciency, which confirms that the observed transfections in these cell lines were nonspecific. In the SCLC cell lines DMS114 and CPH54A, the percentage of positive staining cells were reduced in competition studies with excess EGF. This indicates that the gene delivery occurs through the EGFR. The differences between the observed transfection efficiencies in the SCLC cell lines seems to be due to factors other than the different B max and K d values for EGF binding. Recalling that the reciprocal of K d expresses the affinity of the receptor binding, it could be postulated that the high K d value of A431, at least in part, could account for the relatively low transfection efficiency found in this cell line. A431 has been shown previously to contain both high- and low-affinity EGF binding sites. 18 Approximately % of the EGFRs in A431 are high-affinity receptors with a K d of 7 * M. Nevertheless, other factors such as differences in the sensitivity to the agent for endosomal lysis (i.e., adenovirus dl312) could also contribute to the observed disparities in the transfection efficiencies. It is generally assumed that the presence of the adenovirus particle in the endosome compartment containing an internalized ligand/dna polyplex is dependent upon the binding of the adenovirus to its own receptor (i.e., the CAR). 23 When cells are incubated with polyplexes of ligand/dna and adenovirus, both polyplexes and virus particles will initially bind to their receptors on the cell surface. A fraction of the receptorbound adenovirus and ligand/dna is internalized in the same endosome, and only DNA from these endosomes will escape lysosomal degradation. 23 Therefore, differences in the expression levels of CAR could be a very important determinant for the efficiency of gene delivery with EGF/DNA polyplexes in the present study. Furthermore, differences in the levels of integrin v 5 may also influence the adenovirus-mediated endosome disruption. 24 During the binding and internalization of the adenovirus particle by receptor-mediated endocytosis, the base of the viral penton fibers binds to the integrin v 5, which promotes the adenovirus penetration into the cytoplasm. It has been reported recently that 15 human lung cancer cell lines exhibited pronounced differences in their levels of integrin v To determine whether differences in the susceptibility to adenovirus infection could account for the observed differences in the EGF/DNA polyplex transfection efficiencies, transduction experiments using saturating amounts of an adenovirus vector containing the -gal gene were performed in the cell lines included in the EGF/DNA polyplex transfections (Table 2). The percentage of positive staining cells in this experiment directly reflects the susceptibility to adenovirus infection in each cell line. As described above, it is crucial for the intracellular release of internalized EGF/DNA polyplexes that adenovirus particles are cointernalized to mediate endosomal lysis. In our experiments, adenovirus internalization occurs through the native infection pathway independently of EGF/DNA uptake. Therefore the transduction efficiencies with Ad5CMV -gal will reflect how well adenovirus can act as an endosomolytic agent in the EGF/DNA transfections. The three cell lines CPH54A, DMS114, and DMS153 were the most susceptible to adenovirus infection, with transduction efficiencies of 94% 2.1%, 88% 1.0%, and 99% 0.5%, respectively. These cell lines were also the most efficiently transduced with polyplexes of EGF/DNA (Table 2). Furthermore, the very low susceptibility of A431 to adenovirus infection may explain why this cell line, which greatly overexpresses the EGFR, was the least susceptible to transfection with EGF/DNA polyplexes. These results suggest that sensitivity to adenovirusmediated endosomal lysis was an important determinant for the transfection efficiencies in the EGFR mediated transfections. Future vector systems for receptor-targeted gene delivery could benefit from having an intrinsic endosomolytic activity replacing the adenovirus used in the present study. The development of vector systems that can be administered in vitro and specifically hit target cells throughout the body may be decisive for the successful clinical application of gene therapy for advanced cancer. In the present study, we have demonstrated for the first time receptor-targeted gene delivery to SCLC cell lines using ligand/dna polyplexes. Furthermore, we have shown that even low EGFR expression levels are sufficient for highly efficient and specific gene delivery to SCLC cell lines through EGFR-mediated endocytosis. The importance of restricted expression of a target receptor to the target tissue has been stressed, because even low receptor expression in nontarget tissues may be sufficient for undesirable gene delivery to these tissues. Also, from the present in vitro study of EGF/DNA polyplex gene delivery, overexpression of a target receptor does not seem to be necessary for efficient gene delivery through receptor-mediated internalization. When replication-deficient adenovirus is employed as an endosomolytic agent, the sensitivity of the target cell to adenovirus infection seems to be an important determinant for the efficiency of receptor-mediated gene delivery. The EGFR is expressed in 50% of tested SCLC cell lines. 19 Complementary ligands may therefore be useful in the development of vector systems for receptortargeted gene delivery to SCLC cells. We are currently testing ligand/dna polyplexes for new candidate receptors in SCLC cell lines. Future experiments will elucidate whether these vector systems specifically can target SCLC cells in vitro and in xenotransplanted nude mice.

7 268 FREDERIKSEN, ABRAHAMSEN, CRISTIANO, ET AL: GENE DELIVERY BY AN EGF/DNA POLYPLEX TO SCLC CELL LINES ACKNOWLEDGMENTS We thank Tine V. Jensen for excellent technical assistance and Thomas Carlslund for expert help with the photographic layout. This work was supported by fabrikant Einar Willumsens mindelegat, Bristol Myers Squibb, and Danish Cancer Society Grants and REFERENCES 1. Lassen U, Østerlind K, Hansen M, et al. Long-term survival in small-cell lung cancer: posttreatment characteristics in patients surviving 5 to 18 years an analysis of 1,714 consecutive patients. J Clin Oncol. 1995;13: Wagner E, Zenke M, Cotten M, et al. Transferrin-polycation conjugates as carriers for DNA uptake into cells. Proc Natl Acad Sci USA. 1990;87: Gottschalk S, Cristiano RJ, Smith LC, et al. Folate receptor-mediated DNA delivery into tumor cells: potosomal disruption results in enhanced gene expression. Gene Ther. 1994;1: Wu CH, Wilson JM, Wu GY. Targeting genes: delivery and persistent expression of a foreign gene driven by mammalian regulatory elements in vivo. J Biol Chem. 1989;264: Cristiano RJ, Roth JA. Epidermal growth factor-mediated DNA delivery into lung cancer cells via the epidermal growth factor receptor. Cancer Gene Ther. 1996;3: Giard DJ, Aaronson SA, Todaro GJ, et al. In vitro cultivation of human tumors: establishment of cell lines derived from a series of solid tumors. J Natl Cancer Inst. 1973;51: Engelholm SA, Spang-Thomsen M, Vindeløv LL, et al. Comparison of characteristics of human small cell carcinoma of the lung in patients, in vitro, and transplanted into nude mice. Acta Pathol Microbiol Immunol Scand Sect A. 1986;94: de Leij L, Postmus PE, Buys CHCM, et al. Characterization of three new variant type cell lines derived from small cell carcinoma of the lung. Cancer Res. 1985;45: Pettengill OS, Sorenson GD, Wurster-Hill DH, et al. Isolation and growth characteristics of continuous cell lines from small-cell carcinoma of the lung. Cancer. 1980; 45: Pettengill OS, Cate CC, Flint CF, et al. In vitro isolation of malignant cells from small cell carcinomas. In: Pretlow TG, Pretlow TP, eds. Cell Separation: Methods and Selected Applications. New York: Academic Press; 1984: Sorenson GD, Pettengill OS, Del Prete SA, et al. Biomarkers in small cell carcinoma of the lung. In: Aisner J, ed. Lung Cancer. New York: Churchill Livingstone; 1984: Leibovitz A, Stinson JC, McCombs WB, et al. Classification of human colorectal adenocarcinoma cell lines. Cancer Res. 1976;36: Scatchard G. The attractions of proteins for small molecules and ions. Ann N Y Acad Sci. 1949;51: Xu B, Wiehle S, Roth JA, Cristiano RJ. The contribution of poly-l-lysine, epidermal growth factor, and streptavidin to EGF/PLL/DNA polyplex formation. Gene Ther. 1998;5: Spitzer E, de Los Angeles M, Perez R, Grosse R. Binding properties of biotinylated epidermal growth factor to its receptor on cultured cells and tissue sections. J Cell Biochem. 1989;41: Nguyen DM, Wiehle SA, Roth JA, Cristiano RJ. Gene delivery into malignant cells in vivo by a conjugated adenovirus/dna complex. Cancer Gene Ther. 1997;4: Nguyen DM, Wiehle SA, Roth JA, Cristiano RJ. Delivery of the p53 tumor suppressor gene into lung cancer cells by an adenovirus/dna complex. Cancer Gene Ther. 1997;4: Kawamoto T, Sato JD, Le A, et al. Growth stimulation of A431 cells by epidermal growth factor: identification of high-affinity receptors for epidermal growth factor by an anti-receptor monoclonal antibody. Proc Natl Acad Sci USA. 1983;80: Damstrup L, Rygaard K, Spang-Thomsen M, et al. Expression of the epidermal growth factor receptor in human small cell lung cancer cell lines. Cancer Res. 1992;52: MacGregor GR, Caskey CT. Construction of plasmids that express E. coli -galactosidase in mammalian cells. Nucleic Acids Res. 1989;25: Jones N, Shenk T. An adenovirus type 5 early gene function regulates expression of other early viral genes. Proc Natl Acad Sci USA. 1979;76: Haeder M, Rotsch M, Bepler G, et al. Epidermal growth factor receptor expression in human lung cancer cell lines. Cancer Res. 1988;48: Cristiano RJ, Smith LC, Kay MA, et al. Hepatic gene therapy: efficient gene delivery and expression in primary hepatocytes utilizing a conjugated adenovirus-dna complex. Proc Natl Acad Sci USA. 1993;90: Wickham TJ, Filardo EJ, Cheresh DA, et al. Integrin v 5 selectively promotes adenovirus-mediated cell membrane permeabilization. J Cell Biol. 1994;127: Takayama K, Ueno H, Pei X, et al. The levels of integrin v 5 may predict the susceptibility to adenovirus-mediated gene transfer in human lung cancer cells. Gene Ther. 1998;5: