- Email: [email protected]
Adenoviral Serotypes in Gene Therapy for Esophageal Carcinoma
Journal of Surgical Research 140, 50 –54 (2007) doi:10.1016/j.jss.2006.12.006
Adenoviral Serotypes in Gene Therapy for Esophageal Carcinoma Willem A. Marsman, M.D., Ph.D.,1 John G. Wesseling, Ph.D., Abelkarim el Bouch, M.D., Piter J. Bosma, Ph.D., and J. Jan B. van Lanschot, M.D., Ph.D. Department of Gastroenterology, Academic Medical Center, Amsterdam, The Netherlands Submitted for publication July 26, 2006
INTRODUCTION Purpose. Adenoviral gene therapy could potentially play a role in the treatment of esophageal cancer and Barrett’s esophagus. The adenoviruses can be categorized in different serotypes. The goal of the present study was to investigate the transduction efficacy of different adenoviral serotypes in different models of esophageal cancer and Barrett’s esophagus. Methods. Chimeras of the adenoviral serotype 5 backbone and fibers of serotypes 5, 16, 35, 40, and 50 were constructed with PCR technology. For esophageal cancer, cell lines were used originating from with adenocarcinoma and squamous cell carcinoma, respectively. Differentiating Caco-2 cells were used as an in vitro model for Barrett’s esophagus. GFP was used as a reporter gene and transduction efficacy was assessed by flow cytometry. Results. Overall transduction was rather efficient in the cancer cell lines. Especially serotype 16 and 50 exhibited an improved transduction compared with the other serotypes. In the Caco-2 cell lines, we observed a decreased transduction upon differentiation of the cells. All serotypes had a very limited transduction and no serotype had an additional value in this setting. Conclusions. Some serotypes could have an additional value in the development of gene therapy for esophageal cancer. Especially serotype 16 and 50 exhibited an improved transduction in esophageal cancer cells compared with the native serotype 5. In the setting of Barrett’s esophagus, none of the serotypes had an improved potency as in differentiated intestinal cells all serotypes had a very limited transduction. © 2007
Esophageal cancer is an aggressive disease with a poor prognosis. In general, curative treatment consists of a surgical resection, but even then there is high chance of recurrent disease leading to a 5-y survival of only 25%. Adjuvant modalities, such as chemo- and radiotherapy, are being developed to improve the prognosis of patients with esophageal cancer. Gene therapy could also play a role as an adjuvant treatment modality for esophageal cancer. Barrett’s esophagus is a wellknown precursor lesion for the esophageal adenocarcinoma. When high grade dysplasia has developed in a Barrett’s esophagus, these patients are candidates for a surgical resection. In this setting, gene therapy could be used to treat the dysplasia endoscopically, may be combined with an endoscopic mucosal resection, to avoid a surgical resection. Adenovirus serotype 5 (Ad5) is the most commonly used gene-delivering vector in cancer gene therapy studies [1]. Especially in vitro, Ad5 has an excellent transduction efficacy and is able to induce cell death in different cancer cell lines using different suicide genes. In vivo, however, tumor transduction efficacy is rather limited [2]. To improve transduction, adenoviruses are targeted toward tumor-specific antigens. This strategy has shown significant enhancement of in vitro transduction of tumor cells [3]. Enhanced targeting can also be achieved by exploiting the specific tropisms of different adenoviral serotypes. Until now 51 different adenoviral serotypes have been identified [4]. Differences in tropism of each adenoviral serotype lead to different clinical symptoms in humans. Ad5, for instance, has a natural tropism for pulmonary epithelium, while Ad40 has a more specific tropism for the gastrointestinal tract. The tropism of an adenovirus is mainly determined by its fiber and fiber knob [5]. Modification of these proteins has shown to alter the specific tropism of adenoviral vectors [6, 7]. Recently, the group of Havenga
Elsevier Inc. All rights reserved.
Key Words: gene therapy; adenovirus; Barrett’s esophagus; serotypes; esophageal cancer.
1
To whom correspondence and reprint requests should be addressed at Department of Gastroenterology, Academic Medical Center, Meibergdreef 9, 1115 CC Amsterdam, The Netherlands. E-mail: [email protected].
0022-4804/07 $32.00 © 2007 Elsevier Inc. All rights reserved.
50
MARSMAN ET AL.: ADENOVIRAL SEROTYPES IN GENE THERAPY
et al. constructed a library of adenoviral chimeras consisting of an E1-deleted Ad5 backbone and capsid fibers derived from different adenoviral serotypes [8]. A library of Ad5-based vectors carrying the fibers of alternative serotypes was then created. In the present study, we tested the transduction efficacy of several different chimeric adenoviral serotypes in different models of esophageal carcinoma and Barrett’s esophagus. Cell lines were used as an in vitro model for esophageal cancer. Differentiated Caco-2 cells, which have an intestinal phenotype, were used as an in vitro model for intestinal metaplasia as it is present in Barrett’s esophagus. MATERIALS AND METHODS Adenoviral Vectors The fiber-chimeric adenoviruses were constructed and kindly provided by Dr M. Havenga (Crucell Holland BV, Leiden, The Netherlands). The chimeras were constructed as follows: different fiber genes were PCR amplified with “subgroup-specific” oligonucleotides and cloned in a plasmid containing the Ad5 viral genome sequence [9]. The Ad5 fiber sequence was deleted by PCR, while simultaneously unique restriction sites were introduced to allow for insertion of the fiber molecules derived from other serotypes. The generation and purification of the fiber-chimeric adenoviral vectors were performed on PER.C6 cells [10]. Purification was performed with a two-step cesium chloride purification protocol and all virus aliquots were stored at ⫺80°C. Viral titers were determined by HPLC and plaque titration assays and expressed as viral particles (vp) and plaque forming units (pfu) respectively [11, 12]. The ratio of vp versus pfu did not exceeded 100. In the present study, we used the fiber-chimeric adenoviral vectors with the fibers from adenoviral serotypes 5, 16, 35, 40, and 50 (Fig. 1).
In Vitro Transduction Protocols The human esophageal tumor cell lines TE1, TE2, OE19, and OE33 were used as an in vitro model for esophageal carcinoma. TE1 and TE2 were originally derived from esophageal squamous cell carcinoma, while OE19 and OE33 were originally derived from esophageal adenocarcinoma [13, 14]. The cell lines were propagated every 3 d and cultured in Dulbecco’s modified Eagle’s medium (DMEM) supplemented with L-glutamine (300 g/mL), 10% heatinactivated fetal calf serum, penicillin (100 U/mL) and streptomycin
51
(100 g/mL). To establish differentiated cells, Caco-2 cells underwent a long-term culture of 2 wk where medium was refreshed every 2 d. Under these circumstances, the Caco-2 cells undergo a spontaneous differentiation, which was checked by villin expression on Western blotting. To assess adenoviral transduction in the various cell lines, 10 5 cells/well were plated in a 24-well plate and incubated overnight to allow adherence. Before transduction, the cells were washed with phosphate buffered saline (PBS) and incubated with an adenoviral vector for 1 h with a multiplicity of infection of 100 vp per cell. Fresh culture medium was added and 48 h thereafter, transduction efficacy was assessed by measuring green fluorescent protein (GFP) expression using flow cytometry. The cells were prepared for flow cytometry as follows: the cells were trypsinized and fixated with 4% buffered formalin. Then they were pelleted at 1200 rpm for 5 min and resuspended in PBS with 1% bovine serum albumin. Resuspended cells were used for flow cytometry (Calibrite; Becton Dickinson Immunocytometry Systems; Franklin Lakes, NJ). At least 10,000 cells were counted and data on the number of transduced cells were obtained by setting a 1% quadrant marker for nonspecific staining in the nontransduced cells.
Western Blotting To check for differentiation of the Caco-2 cells, villin expression was determined by Western blotting. Caco-2 confluent layers from day 0 (after reaching confluence), day 7, 14, and 21 were lysed by adding 0.40 mL of ice-cold homogenizing buffer (20% glycerol, 0.1 M Tris-HCl, and 10 mM EDTA, pH 7.4) containing 1% protease inhibitor cocktail (Sigma, St. Louis, MO). A sample of mouse ileum was used as a positive control. The cells were frozen and thawed in ⫺80°C three times and sonicated for 10 s. Homogenates were dissolved in a sample buffer (2% sodium dodecyl sulfate (SDS), 5% sucrose, 5% 2-mercaptoethanol, and 50 mM Tris-HCl, pH 6.8) to a final concentration of 25 g/40 L and subsequently separated on a 6% SDSpolyacrylamide gel. Villin was detected using a mouse monoclonal antibody (Chemicon, Temecula, CA), and horseradish peroxidase anti-mouse (Transduction Laboratories, Lexington, KY) was used as secondary antibody. The immunocomplex was detected with the Lumilight-Plus chemiluminescence detection system and the Lumi Imager (Roche Diagnostics, Mannheim, Germany). Amido black staining of the Western blot membranes was used to ensure equal loading of proteins.
Statistics All experiments were performed in triplicate and expressed in mean values and standard deviation. The SPSS 12.0 software package (SPSS Inc., Chicago, IL) was used for the statistical calculations.
FIG. 1. Construction of the adenoviral chimeras. The adenoviral type 5 backbone is combined with the fibers of different adenoviral serotypes, which all have a specific length.
52
JOURNAL OF SURGICAL RESEARCH: VOL. 140, NO. 1, JUNE 1, 2007
FIG. 2. Transduction of the esophageal squamous cell carcinoma cell lines TE1 and TE2 by adenoviral (100 vp/cell) chimeras representing different serotypes. *P ⫽ ⬍0.05 for transduction efficacy of chimera 16 and 50 versus 5. Student’s t-test was used to test whether differences in transduction were significant. All values were two-sided and a P value ⬍ 0.05 was considered significant.
RESULTS
Two days after seeding of Caco-2 cells, there was no expression of villin, while after 7 d there was villin expression, which had further increased 14 and 21 d after seeding (Fig. 4). As the macroscopic quality of the cell deck was poor at 21 d, compared with 14 d, it was decided to use 14 d cultured Caco-2 cells for transduction experiments. Caco-2 cells were used as a differentiating model to study intestinal gene transfer of the adenoviral chimeras in vitro. The native chimera Ad5 exhibited a decreased transduction upon differentiation from 11.7% (SD; 0.7) in undifferentiated cells to 2.3% (SD; 0.5) in differentiated Caco-2 cells (Fig. 5). Chimeras 16, 35, and 50 exhibited a relatively high transduction in undifferentiated Caco-2 cells at a confluence of 50%, with an efficacy up to 33.8% (SD; 4.5) for chimera 50. This transduction efficacy, however, decreased markedly upon reaching 100% confluence and even further upon full differentiation of the Caco-2 cells (P ⬍ 0.05). In fully differentiated Caco-2 cells, none of the chimeras had a sufficient transduction with chimera Ad35 exhibiting the highest transduction of 4.0% (SD; 0.6) only.
Transduction in Esophageal Cancer Cell Lines
The native Ad5 revealed a higher transduction efficacy for the squamous cell carcinoma cell line TE1 compared with the cell line TE2, with a transduction efficacy of 34.6% (SD;4.2) and 9.1% (SD;2.4), respectively, (Fig. 2). This difference was statistically significant (Student’s t-test: P ⬍ 0.05). The adenoviral chimeras 16 and 50 exhibited an improved transduction in both cell lines. Chimera 16 had a transduction efficacy of 87.3% (SD; 6.3) and 52.2% (SD; 1.8) for TE1 and TE2, respectively, which was significantly higher compared with the native Ad5. Also chimera 50 had a significantly improved transduction with an efficacy of 75.8% (SD; 2.2) and 28.0% (SD; 1.0) in TE1 and TE2, respectively. Overall, chimeras 16 and 50 exhibited a two to threefold increased transduction compared to the native Ad5. The adenoviral chimeras 35 and 40 did not exhibit an improved transduction compared with chimera 5 in both cell lines. Ad5 had a transduction efficacy of 24.5% (SD; 6.0) and 21.8% (SD; 2.4) in the OE19 and OE33 cell lines, respectively (Fig. 3). Chimera 16 exhibited an improved transduction of 58.2% (SD; 4.7) and 71.8% (SD; 3.0), and also chimera 50 exhibited an improved transduction of 41.6% (SD; 4.1) and 51.9 (SD; 13.5), for OE19 and OE33, respectively. These transduction efficacies were significantly better compared with the Ad5 transduction efficacy. Also in these cells lines, chimeras 35 and 40 exhibited a similar transduction compared with the native Ad5. Transduction in Differentiating Cells
Differentiation of the Caco-2 cells was assessed by the expression of villin, as shown by Western blotting.
DISCUSSION
Adenovirus serotype 5 is the most commonly used gene-delivering vector for cancer gene therapy studies because it is easy to modify, can contain a large therapeutic gene, and can be produced in large quantities. The limitations of this vector, however, are the limited transduction in vivo and the immunogenicity of the vector. Therefore, the use of alternative genedelivering vectors needs to be explored. One possibility for this is to exploit the specific tropism of the different adenoviral serotypes to enhance the transduction efficiency of tumor cells. Traditionally, these serotypes might be less immunogenic. In the present study, we investigated the potential use of adenoviral serotypes in the setting of Barrett’s esophagus or esophageal cancer. Barrett’s esophagus is
FIG. 3. Transduction of the esophageal adenocarcinoma cell lines OE19 and OE33 by adenoviral chimeras (100 vp/cell) representing different serotypes. *P ⫽ ⬍0.05 for transduction efficacy of chimera 16 and 50 versus 5.
MARSMAN ET AL.: ADENOVIRAL SEROTYPES IN GENE THERAPY
53
FIG. 4. Western blot of villin expression in differentiating Caco-2 cells, at day 2, 7, 14, and 21 after reaching confluence. Villin expression increased upon long-term culture. Mouse ileum served as a positive control. (Color version of figure is available online.)
a premalignant condition for the development of esophageal adenocarcinoma. When high grade dysplasia has developed, a surgical resection used to be the treatment of choice, which is associated with a substantial morbidity and mortality. In esophageal cancer cell lines, the chimeras 16 and 50 exhibited promising results compared with the chimeras 5, 35, and 40. These results are comparable with previous results of transduction with a genetic retargeted adenovirus in pancreatic cancer cells [15]. Differentiated Caco-2 cells were used as a model of Barrett’s esophagus, and none of the chimeras showed a significant transduction in this model, not even chimera 40, which has a natural tropism for the intestine. Adenoviruses can be categorized in several serotypes, which have their own specific tropism. Until now, a total of 51 adenoviral serotypes have been identified. These serotypes can be divided into six species (A-F) based on several parameters, including hemagglutination pattern, virion polypeptide profile, and DNA restriction endonuclease pattern [16, 17]. Species B, C, D, and E form the R cluster as they are commonly associated with respiratory infections [18]. This is in contrast to subgenus A and F that are generally associated with gastrointestinal infections and form the G cluster. Members of species C, including serotype 5, are the most frequently isolated adenovirus reported to the WHO and predominantly
FIG. 5. Transduction of different adenoviral serotypes (100 vp/ cell) in undifferentiated and differentiated Caco-2 cells; 50% represents undifferentiated cells with 50% confluence, and 100% represents undifferentiated cells with 100% confluency. Differentiated (diff) Caco-2 cells had undergone a 2 wk long-term culture. For all different chimeras, transduction was significantly less in differentiated cells compared with undifferentiated cells.
result in infections of the adenoids and respiratory tract [4]. In this study, we decided to use chimera 16, 35, and 50 as these serotypes have previously been shown to effectively transduce cancer cells [8]. Ad16 belongs to cluster B:1 and is associated with respiratory infections while Ad35 belongs to cluster B:2 and is associated with infections of the urinary tract. Ad40 belongs to species F and is associated with gastrointestinal infections [19]. Therefore, in theory, serotype 40 could be a suitable gene-delivering vector in the intestine, Barrett’s esophagus, or differentiated Caco-2 cells. Ad50, a recently discovered adenovirus, was isolated from fecal specimens in patients with AIDS, and also belongs to subgroup B [20]. Gene therapy for esophageal cancer can potentially be applied in three settings. One setting is gene therapy for patients with an established esophageal carcinoma. In this setting, a therapeutic vector can be injected in the primary tumor or in a metastatic lesion to downstage the tumor. This could be useful as a palliative measure or as a curative measure when the downstaged tumor can be resected after induction therapy. Another setting is for the patient with high-grade dysplasia in a Barrett’s esophagus. These patients are at high risk for the development of carcinoma and therefore a surgical esophagectomy is generally advised [21]. As a surgical esophagectomy is associated with a high morbidity and even mortality, alternative, mainly endoscopic, strategies are currently being developed. Techniques such as endoscopic mucosal resection or photodynamic therapy could be combined with local application of gene-delivering vectors. The third strategy might consist of systemic targeted therapy for minimal residual disease or potentially present micrometastases after surgical esophagectomy. For the first setting, we investigated transduction efficacy of the adenoviral chimeras in several esophageal cancer cell lines. In that setting, chimeras 16 and 50 exhibited the most promising results. Serotype 16 and 50 both belong to subgroup B and, recently, CD46 has been identified as the receptor for these serotypes [22]. In addition, it has been shown that CD46 is up-regulated in cancer cells [23]. This could explain improved transduction of these serotypes. For the second strategy, we studied differentiating Caco-2 cells. In that setting
54
JOURNAL OF SURGICAL RESEARCH: VOL. 140, NO. 1, JUNE 1, 2007 9.
Rea D, Havenga MJ, van Den AM, et al. Highly efficient transduction of human monocyte-derived dendritic cells with subgroup B fiber-modified adenovirus vectors enhances transgeneencoded antigen presentation to cytotoxic T cells. J Immunol 2001;166:5236.
10.
Havenga MJ, Lemckert AA, Grimbergen JM, et al. Improved adenovirus vectors for infection of cardiovascular tissues. J Virol 2001;75:3335.
11.
Hehir KM, Armentano D, Cardoza LM, et al. Molecular characterization of replication-competent variants of adenovirus vectors and genome modifications to prevent their occurrence. J Virol 1996;70:8459.
12.
Shabram PW, Giroux DD, Goudreau AM, et al. Analytical anion-exchange HPLC of recombinant type-5 adenoviral particles. Hum Gene Ther 1997;8:453.
13.
Nishihira T, Kasai M, Mori S, et al. Characteristics of two cell lines (TE-1 and TE-2) derived from human squamous cell carcinoma of the esophagus. Gann 1979;70:575.
14.
Rockett JC, Larkin K, Darnton SJ, et al. Five newly established esophageal carcinoma cell lines: Phenotypic and immunological characterization. Br J Cancer 1997;75:258.
15.
The authors acknowledge Menzo Havenga from Crucell B. V., Leiden, The Netherlands, for providing the different adenoviral chimeras.
Wesseling JG, Bosma PJ, Krasnykh V, et al. Improved gene transfer efficiency to primary and established human pancreatic carcinoma target cells via epidermal growth factor receptor and integrin-targeted adenoviral vectors. Gene Ther 2001;8:969.
16.
Hierholzer JC. Further subgrouping of the human adenoviruses by differential hemagglutination. J Infect Dis 1973;128: 541.
REFERENCES
17.
Wadell G. Classification of human adenoviruses by SDSpolyacrylamide gel electrophoresis of structural polypeptides. Intervirology 1979;11:47.
18.
Bailey A, Mautner V. Phylogenetic relationships among adenovirus serotypes. Virology 1994;205:438.
19.
Uhnoo I, Wadell G, Svensson L, et al. Importance of enteric adenoviruses 40 and 41 in acute gastroenteritis in infants and young children. J Clin Microbiol 1984;20:365.
20.
De Jong JC, Wermenbol AG, Verweij-Uijterwaal MW, et al. Adenoviruses from human immunodeficiency virus-infected individuals, including two strains that represent new candidate serotypes Ad50 and Ad51 of species B1 and D, respectively. J Clin Microbiol 1999;37:3940.
21.
Rice TW, Falk GW, Achkar E, et al. Surgical management of high-grade dysplasia in Barrett’s esophagus. Am J Gastroenterol 1993;88:1832.
22.
Gaggar A, Shayakhmetov DM, Lieber A. CD46 is a cellular receptor for group B adenoviruses. Nat Med 2003;9:1408.
23.
Ni S, Gaggar A, Di Paolo N, et al. Evaluation of adenovirus vectors containing serotype 35 fibers for tumor targeting. Cancer Gene Ther 2006.
none of the serotypes showed a preferential transduction in differentiated cells. The decreased transduction in that setting could implicate that cancer cells are preferentially transduced compared with more differentiated precursor cells. In conclusion, we explored the use of different adenoviral chimeras for the potential use in esophageal carcinoma or its precursor lesion, i.e., the Barrett’s esophagus. For esophageal carcinoma adenoviral chimera 16 and 50 might be more favorable as they exhibit an increased transduction in esophageal cancer cell lines, compared with the native adenovirus. For potential treatment of Barrett’s esophagus, the described chimeras are not very promising as their transduction in differentiated Caco-2 cells was very limited. More research on targeting and in vivo application of genedelivering vectors is needed for further development of gene therapy in patients with esophageal carcinoma or Barrett’s esophagus. ACKNOWLEDGMENTS
1. 2. 3.
4. 5. 6.
7.
8.
Gomez-Navarro J, Curiel DT, Douglas JT. Gene therapy for cancer. Eur J Cancer 1999;35:2039. Kanerva A, Hemminki A. Modified adenoviruses for cancer gene therapy. Int J Cancer 2004;110:475. Rots MG, Curiel DT, Gerritsen WR, et al. Targeted cancer gene therapy: The flexibility of adenoviral gene therapy vectors. J Control Release 2003;87:159. Wadell G. Molecular epidemiology of human adenoviruses. Curr Top Microbiol Immunol 1984;110:191. Bergelson JM. Receptors mediating adenovirus attachment and internalization. Biochem Pharmacol 1999;57:975. Goossens PH, Vogels R, Pieterman E, et al. The influence of synovial fluid on adenovirus-mediated gene transfer to the synovial tissue. Arthritis Rheum 2001;44:48. Yotnda P, Onishi H, Heslop HE, et al. Efficient infection of primitive hematopoietic stem cells by modified adenovirus. Gene Ther 2001;8:930. Havenga MJ, Lemckert AA, Ophorst OJ, et al. Exploiting the natural diversity in adenovirus tropism for therapy and prevention of disease. J Virol 2002;76:4612.
Adenoviral Serotypes in Gene Therapy for Esophageal Carcinoma Willem A. Marsman, M.D., Ph.D.,1 John G. Wesseling, Ph.D., Abelkarim el Bouch, M.D., Piter J. Bosma, Ph.D., and J. Jan B. van Lanschot, M.D., Ph.D. Department of Gastroenterology, Academic Medical Center, Amsterdam, The Netherlands Submitted for publication July 26, 2006
INTRODUCTION Purpose. Adenoviral gene therapy could potentially play a role in the treatment of esophageal cancer and Barrett’s esophagus. The adenoviruses can be categorized in different serotypes. The goal of the present study was to investigate the transduction efficacy of different adenoviral serotypes in different models of esophageal cancer and Barrett’s esophagus. Methods. Chimeras of the adenoviral serotype 5 backbone and fibers of serotypes 5, 16, 35, 40, and 50 were constructed with PCR technology. For esophageal cancer, cell lines were used originating from with adenocarcinoma and squamous cell carcinoma, respectively. Differentiating Caco-2 cells were used as an in vitro model for Barrett’s esophagus. GFP was used as a reporter gene and transduction efficacy was assessed by flow cytometry. Results. Overall transduction was rather efficient in the cancer cell lines. Especially serotype 16 and 50 exhibited an improved transduction compared with the other serotypes. In the Caco-2 cell lines, we observed a decreased transduction upon differentiation of the cells. All serotypes had a very limited transduction and no serotype had an additional value in this setting. Conclusions. Some serotypes could have an additional value in the development of gene therapy for esophageal cancer. Especially serotype 16 and 50 exhibited an improved transduction in esophageal cancer cells compared with the native serotype 5. In the setting of Barrett’s esophagus, none of the serotypes had an improved potency as in differentiated intestinal cells all serotypes had a very limited transduction. © 2007
Esophageal cancer is an aggressive disease with a poor prognosis. In general, curative treatment consists of a surgical resection, but even then there is high chance of recurrent disease leading to a 5-y survival of only 25%. Adjuvant modalities, such as chemo- and radiotherapy, are being developed to improve the prognosis of patients with esophageal cancer. Gene therapy could also play a role as an adjuvant treatment modality for esophageal cancer. Barrett’s esophagus is a wellknown precursor lesion for the esophageal adenocarcinoma. When high grade dysplasia has developed in a Barrett’s esophagus, these patients are candidates for a surgical resection. In this setting, gene therapy could be used to treat the dysplasia endoscopically, may be combined with an endoscopic mucosal resection, to avoid a surgical resection. Adenovirus serotype 5 (Ad5) is the most commonly used gene-delivering vector in cancer gene therapy studies [1]. Especially in vitro, Ad5 has an excellent transduction efficacy and is able to induce cell death in different cancer cell lines using different suicide genes. In vivo, however, tumor transduction efficacy is rather limited [2]. To improve transduction, adenoviruses are targeted toward tumor-specific antigens. This strategy has shown significant enhancement of in vitro transduction of tumor cells [3]. Enhanced targeting can also be achieved by exploiting the specific tropisms of different adenoviral serotypes. Until now 51 different adenoviral serotypes have been identified [4]. Differences in tropism of each adenoviral serotype lead to different clinical symptoms in humans. Ad5, for instance, has a natural tropism for pulmonary epithelium, while Ad40 has a more specific tropism for the gastrointestinal tract. The tropism of an adenovirus is mainly determined by its fiber and fiber knob [5]. Modification of these proteins has shown to alter the specific tropism of adenoviral vectors [6, 7]. Recently, the group of Havenga
Elsevier Inc. All rights reserved.
Key Words: gene therapy; adenovirus; Barrett’s esophagus; serotypes; esophageal cancer.
1
To whom correspondence and reprint requests should be addressed at Department of Gastroenterology, Academic Medical Center, Meibergdreef 9, 1115 CC Amsterdam, The Netherlands. E-mail: [email protected].
0022-4804/07 $32.00 © 2007 Elsevier Inc. All rights reserved.
50
MARSMAN ET AL.: ADENOVIRAL SEROTYPES IN GENE THERAPY
et al. constructed a library of adenoviral chimeras consisting of an E1-deleted Ad5 backbone and capsid fibers derived from different adenoviral serotypes [8]. A library of Ad5-based vectors carrying the fibers of alternative serotypes was then created. In the present study, we tested the transduction efficacy of several different chimeric adenoviral serotypes in different models of esophageal carcinoma and Barrett’s esophagus. Cell lines were used as an in vitro model for esophageal cancer. Differentiated Caco-2 cells, which have an intestinal phenotype, were used as an in vitro model for intestinal metaplasia as it is present in Barrett’s esophagus. MATERIALS AND METHODS Adenoviral Vectors The fiber-chimeric adenoviruses were constructed and kindly provided by Dr M. Havenga (Crucell Holland BV, Leiden, The Netherlands). The chimeras were constructed as follows: different fiber genes were PCR amplified with “subgroup-specific” oligonucleotides and cloned in a plasmid containing the Ad5 viral genome sequence [9]. The Ad5 fiber sequence was deleted by PCR, while simultaneously unique restriction sites were introduced to allow for insertion of the fiber molecules derived from other serotypes. The generation and purification of the fiber-chimeric adenoviral vectors were performed on PER.C6 cells [10]. Purification was performed with a two-step cesium chloride purification protocol and all virus aliquots were stored at ⫺80°C. Viral titers were determined by HPLC and plaque titration assays and expressed as viral particles (vp) and plaque forming units (pfu) respectively [11, 12]. The ratio of vp versus pfu did not exceeded 100. In the present study, we used the fiber-chimeric adenoviral vectors with the fibers from adenoviral serotypes 5, 16, 35, 40, and 50 (Fig. 1).
In Vitro Transduction Protocols The human esophageal tumor cell lines TE1, TE2, OE19, and OE33 were used as an in vitro model for esophageal carcinoma. TE1 and TE2 were originally derived from esophageal squamous cell carcinoma, while OE19 and OE33 were originally derived from esophageal adenocarcinoma [13, 14]. The cell lines were propagated every 3 d and cultured in Dulbecco’s modified Eagle’s medium (DMEM) supplemented with L-glutamine (300 g/mL), 10% heatinactivated fetal calf serum, penicillin (100 U/mL) and streptomycin
51
(100 g/mL). To establish differentiated cells, Caco-2 cells underwent a long-term culture of 2 wk where medium was refreshed every 2 d. Under these circumstances, the Caco-2 cells undergo a spontaneous differentiation, which was checked by villin expression on Western blotting. To assess adenoviral transduction in the various cell lines, 10 5 cells/well were plated in a 24-well plate and incubated overnight to allow adherence. Before transduction, the cells were washed with phosphate buffered saline (PBS) and incubated with an adenoviral vector for 1 h with a multiplicity of infection of 100 vp per cell. Fresh culture medium was added and 48 h thereafter, transduction efficacy was assessed by measuring green fluorescent protein (GFP) expression using flow cytometry. The cells were prepared for flow cytometry as follows: the cells were trypsinized and fixated with 4% buffered formalin. Then they were pelleted at 1200 rpm for 5 min and resuspended in PBS with 1% bovine serum albumin. Resuspended cells were used for flow cytometry (Calibrite; Becton Dickinson Immunocytometry Systems; Franklin Lakes, NJ). At least 10,000 cells were counted and data on the number of transduced cells were obtained by setting a 1% quadrant marker for nonspecific staining in the nontransduced cells.
Western Blotting To check for differentiation of the Caco-2 cells, villin expression was determined by Western blotting. Caco-2 confluent layers from day 0 (after reaching confluence), day 7, 14, and 21 were lysed by adding 0.40 mL of ice-cold homogenizing buffer (20% glycerol, 0.1 M Tris-HCl, and 10 mM EDTA, pH 7.4) containing 1% protease inhibitor cocktail (Sigma, St. Louis, MO). A sample of mouse ileum was used as a positive control. The cells were frozen and thawed in ⫺80°C three times and sonicated for 10 s. Homogenates were dissolved in a sample buffer (2% sodium dodecyl sulfate (SDS), 5% sucrose, 5% 2-mercaptoethanol, and 50 mM Tris-HCl, pH 6.8) to a final concentration of 25 g/40 L and subsequently separated on a 6% SDSpolyacrylamide gel. Villin was detected using a mouse monoclonal antibody (Chemicon, Temecula, CA), and horseradish peroxidase anti-mouse (Transduction Laboratories, Lexington, KY) was used as secondary antibody. The immunocomplex was detected with the Lumilight-Plus chemiluminescence detection system and the Lumi Imager (Roche Diagnostics, Mannheim, Germany). Amido black staining of the Western blot membranes was used to ensure equal loading of proteins.
Statistics All experiments were performed in triplicate and expressed in mean values and standard deviation. The SPSS 12.0 software package (SPSS Inc., Chicago, IL) was used for the statistical calculations.
FIG. 1. Construction of the adenoviral chimeras. The adenoviral type 5 backbone is combined with the fibers of different adenoviral serotypes, which all have a specific length.
52
JOURNAL OF SURGICAL RESEARCH: VOL. 140, NO. 1, JUNE 1, 2007
FIG. 2. Transduction of the esophageal squamous cell carcinoma cell lines TE1 and TE2 by adenoviral (100 vp/cell) chimeras representing different serotypes. *P ⫽ ⬍0.05 for transduction efficacy of chimera 16 and 50 versus 5. Student’s t-test was used to test whether differences in transduction were significant. All values were two-sided and a P value ⬍ 0.05 was considered significant.
RESULTS
Two days after seeding of Caco-2 cells, there was no expression of villin, while after 7 d there was villin expression, which had further increased 14 and 21 d after seeding (Fig. 4). As the macroscopic quality of the cell deck was poor at 21 d, compared with 14 d, it was decided to use 14 d cultured Caco-2 cells for transduction experiments. Caco-2 cells were used as a differentiating model to study intestinal gene transfer of the adenoviral chimeras in vitro. The native chimera Ad5 exhibited a decreased transduction upon differentiation from 11.7% (SD; 0.7) in undifferentiated cells to 2.3% (SD; 0.5) in differentiated Caco-2 cells (Fig. 5). Chimeras 16, 35, and 50 exhibited a relatively high transduction in undifferentiated Caco-2 cells at a confluence of 50%, with an efficacy up to 33.8% (SD; 4.5) for chimera 50. This transduction efficacy, however, decreased markedly upon reaching 100% confluence and even further upon full differentiation of the Caco-2 cells (P ⬍ 0.05). In fully differentiated Caco-2 cells, none of the chimeras had a sufficient transduction with chimera Ad35 exhibiting the highest transduction of 4.0% (SD; 0.6) only.
Transduction in Esophageal Cancer Cell Lines
The native Ad5 revealed a higher transduction efficacy for the squamous cell carcinoma cell line TE1 compared with the cell line TE2, with a transduction efficacy of 34.6% (SD;4.2) and 9.1% (SD;2.4), respectively, (Fig. 2). This difference was statistically significant (Student’s t-test: P ⬍ 0.05). The adenoviral chimeras 16 and 50 exhibited an improved transduction in both cell lines. Chimera 16 had a transduction efficacy of 87.3% (SD; 6.3) and 52.2% (SD; 1.8) for TE1 and TE2, respectively, which was significantly higher compared with the native Ad5. Also chimera 50 had a significantly improved transduction with an efficacy of 75.8% (SD; 2.2) and 28.0% (SD; 1.0) in TE1 and TE2, respectively. Overall, chimeras 16 and 50 exhibited a two to threefold increased transduction compared to the native Ad5. The adenoviral chimeras 35 and 40 did not exhibit an improved transduction compared with chimera 5 in both cell lines. Ad5 had a transduction efficacy of 24.5% (SD; 6.0) and 21.8% (SD; 2.4) in the OE19 and OE33 cell lines, respectively (Fig. 3). Chimera 16 exhibited an improved transduction of 58.2% (SD; 4.7) and 71.8% (SD; 3.0), and also chimera 50 exhibited an improved transduction of 41.6% (SD; 4.1) and 51.9 (SD; 13.5), for OE19 and OE33, respectively. These transduction efficacies were significantly better compared with the Ad5 transduction efficacy. Also in these cells lines, chimeras 35 and 40 exhibited a similar transduction compared with the native Ad5. Transduction in Differentiating Cells
Differentiation of the Caco-2 cells was assessed by the expression of villin, as shown by Western blotting.
DISCUSSION
Adenovirus serotype 5 is the most commonly used gene-delivering vector for cancer gene therapy studies because it is easy to modify, can contain a large therapeutic gene, and can be produced in large quantities. The limitations of this vector, however, are the limited transduction in vivo and the immunogenicity of the vector. Therefore, the use of alternative genedelivering vectors needs to be explored. One possibility for this is to exploit the specific tropism of the different adenoviral serotypes to enhance the transduction efficiency of tumor cells. Traditionally, these serotypes might be less immunogenic. In the present study, we investigated the potential use of adenoviral serotypes in the setting of Barrett’s esophagus or esophageal cancer. Barrett’s esophagus is
FIG. 3. Transduction of the esophageal adenocarcinoma cell lines OE19 and OE33 by adenoviral chimeras (100 vp/cell) representing different serotypes. *P ⫽ ⬍0.05 for transduction efficacy of chimera 16 and 50 versus 5.
MARSMAN ET AL.: ADENOVIRAL SEROTYPES IN GENE THERAPY
53
FIG. 4. Western blot of villin expression in differentiating Caco-2 cells, at day 2, 7, 14, and 21 after reaching confluence. Villin expression increased upon long-term culture. Mouse ileum served as a positive control. (Color version of figure is available online.)
a premalignant condition for the development of esophageal adenocarcinoma. When high grade dysplasia has developed, a surgical resection used to be the treatment of choice, which is associated with a substantial morbidity and mortality. In esophageal cancer cell lines, the chimeras 16 and 50 exhibited promising results compared with the chimeras 5, 35, and 40. These results are comparable with previous results of transduction with a genetic retargeted adenovirus in pancreatic cancer cells [15]. Differentiated Caco-2 cells were used as a model of Barrett’s esophagus, and none of the chimeras showed a significant transduction in this model, not even chimera 40, which has a natural tropism for the intestine. Adenoviruses can be categorized in several serotypes, which have their own specific tropism. Until now, a total of 51 adenoviral serotypes have been identified. These serotypes can be divided into six species (A-F) based on several parameters, including hemagglutination pattern, virion polypeptide profile, and DNA restriction endonuclease pattern [16, 17]. Species B, C, D, and E form the R cluster as they are commonly associated with respiratory infections [18]. This is in contrast to subgenus A and F that are generally associated with gastrointestinal infections and form the G cluster. Members of species C, including serotype 5, are the most frequently isolated adenovirus reported to the WHO and predominantly
FIG. 5. Transduction of different adenoviral serotypes (100 vp/ cell) in undifferentiated and differentiated Caco-2 cells; 50% represents undifferentiated cells with 50% confluence, and 100% represents undifferentiated cells with 100% confluency. Differentiated (diff) Caco-2 cells had undergone a 2 wk long-term culture. For all different chimeras, transduction was significantly less in differentiated cells compared with undifferentiated cells.
result in infections of the adenoids and respiratory tract [4]. In this study, we decided to use chimera 16, 35, and 50 as these serotypes have previously been shown to effectively transduce cancer cells [8]. Ad16 belongs to cluster B:1 and is associated with respiratory infections while Ad35 belongs to cluster B:2 and is associated with infections of the urinary tract. Ad40 belongs to species F and is associated with gastrointestinal infections [19]. Therefore, in theory, serotype 40 could be a suitable gene-delivering vector in the intestine, Barrett’s esophagus, or differentiated Caco-2 cells. Ad50, a recently discovered adenovirus, was isolated from fecal specimens in patients with AIDS, and also belongs to subgroup B [20]. Gene therapy for esophageal cancer can potentially be applied in three settings. One setting is gene therapy for patients with an established esophageal carcinoma. In this setting, a therapeutic vector can be injected in the primary tumor or in a metastatic lesion to downstage the tumor. This could be useful as a palliative measure or as a curative measure when the downstaged tumor can be resected after induction therapy. Another setting is for the patient with high-grade dysplasia in a Barrett’s esophagus. These patients are at high risk for the development of carcinoma and therefore a surgical esophagectomy is generally advised [21]. As a surgical esophagectomy is associated with a high morbidity and even mortality, alternative, mainly endoscopic, strategies are currently being developed. Techniques such as endoscopic mucosal resection or photodynamic therapy could be combined with local application of gene-delivering vectors. The third strategy might consist of systemic targeted therapy for minimal residual disease or potentially present micrometastases after surgical esophagectomy. For the first setting, we investigated transduction efficacy of the adenoviral chimeras in several esophageal cancer cell lines. In that setting, chimeras 16 and 50 exhibited the most promising results. Serotype 16 and 50 both belong to subgroup B and, recently, CD46 has been identified as the receptor for these serotypes [22]. In addition, it has been shown that CD46 is up-regulated in cancer cells [23]. This could explain improved transduction of these serotypes. For the second strategy, we studied differentiating Caco-2 cells. In that setting
54
JOURNAL OF SURGICAL RESEARCH: VOL. 140, NO. 1, JUNE 1, 2007 9.
Rea D, Havenga MJ, van Den AM, et al. Highly efficient transduction of human monocyte-derived dendritic cells with subgroup B fiber-modified adenovirus vectors enhances transgeneencoded antigen presentation to cytotoxic T cells. J Immunol 2001;166:5236.
10.
Havenga MJ, Lemckert AA, Grimbergen JM, et al. Improved adenovirus vectors for infection of cardiovascular tissues. J Virol 2001;75:3335.
11.
Hehir KM, Armentano D, Cardoza LM, et al. Molecular characterization of replication-competent variants of adenovirus vectors and genome modifications to prevent their occurrence. J Virol 1996;70:8459.
12.
Shabram PW, Giroux DD, Goudreau AM, et al. Analytical anion-exchange HPLC of recombinant type-5 adenoviral particles. Hum Gene Ther 1997;8:453.
13.
Nishihira T, Kasai M, Mori S, et al. Characteristics of two cell lines (TE-1 and TE-2) derived from human squamous cell carcinoma of the esophagus. Gann 1979;70:575.
14.
Rockett JC, Larkin K, Darnton SJ, et al. Five newly established esophageal carcinoma cell lines: Phenotypic and immunological characterization. Br J Cancer 1997;75:258.
15.
The authors acknowledge Menzo Havenga from Crucell B. V., Leiden, The Netherlands, for providing the different adenoviral chimeras.
Wesseling JG, Bosma PJ, Krasnykh V, et al. Improved gene transfer efficiency to primary and established human pancreatic carcinoma target cells via epidermal growth factor receptor and integrin-targeted adenoviral vectors. Gene Ther 2001;8:969.
16.
Hierholzer JC. Further subgrouping of the human adenoviruses by differential hemagglutination. J Infect Dis 1973;128: 541.
REFERENCES
17.
Wadell G. Classification of human adenoviruses by SDSpolyacrylamide gel electrophoresis of structural polypeptides. Intervirology 1979;11:47.
18.
Bailey A, Mautner V. Phylogenetic relationships among adenovirus serotypes. Virology 1994;205:438.
19.
Uhnoo I, Wadell G, Svensson L, et al. Importance of enteric adenoviruses 40 and 41 in acute gastroenteritis in infants and young children. J Clin Microbiol 1984;20:365.
20.
De Jong JC, Wermenbol AG, Verweij-Uijterwaal MW, et al. Adenoviruses from human immunodeficiency virus-infected individuals, including two strains that represent new candidate serotypes Ad50 and Ad51 of species B1 and D, respectively. J Clin Microbiol 1999;37:3940.
21.
Rice TW, Falk GW, Achkar E, et al. Surgical management of high-grade dysplasia in Barrett’s esophagus. Am J Gastroenterol 1993;88:1832.
22.
Gaggar A, Shayakhmetov DM, Lieber A. CD46 is a cellular receptor for group B adenoviruses. Nat Med 2003;9:1408.
23.
Ni S, Gaggar A, Di Paolo N, et al. Evaluation of adenovirus vectors containing serotype 35 fibers for tumor targeting. Cancer Gene Ther 2006.
none of the serotypes showed a preferential transduction in differentiated cells. The decreased transduction in that setting could implicate that cancer cells are preferentially transduced compared with more differentiated precursor cells. In conclusion, we explored the use of different adenoviral chimeras for the potential use in esophageal carcinoma or its precursor lesion, i.e., the Barrett’s esophagus. For esophageal carcinoma adenoviral chimera 16 and 50 might be more favorable as they exhibit an increased transduction in esophageal cancer cell lines, compared with the native adenovirus. For potential treatment of Barrett’s esophagus, the described chimeras are not very promising as their transduction in differentiated Caco-2 cells was very limited. More research on targeting and in vivo application of genedelivering vectors is needed for further development of gene therapy in patients with esophageal carcinoma or Barrett’s esophagus. ACKNOWLEDGMENTS
1. 2. 3.
4. 5. 6.
7.
8.
Gomez-Navarro J, Curiel DT, Douglas JT. Gene therapy for cancer. Eur J Cancer 1999;35:2039. Kanerva A, Hemminki A. Modified adenoviruses for cancer gene therapy. Int J Cancer 2004;110:475. Rots MG, Curiel DT, Gerritsen WR, et al. Targeted cancer gene therapy: The flexibility of adenoviral gene therapy vectors. J Control Release 2003;87:159. Wadell G. Molecular epidemiology of human adenoviruses. Curr Top Microbiol Immunol 1984;110:191. Bergelson JM. Receptors mediating adenovirus attachment and internalization. Biochem Pharmacol 1999;57:975. Goossens PH, Vogels R, Pieterman E, et al. The influence of synovial fluid on adenovirus-mediated gene transfer to the synovial tissue. Arthritis Rheum 2001;44:48. Yotnda P, Onishi H, Heslop HE, et al. Efficient infection of primitive hematopoietic stem cells by modified adenovirus. Gene Ther 2001;8:930. Havenga MJ, Lemckert AA, Ophorst OJ, et al. Exploiting the natural diversity in adenovirus tropism for therapy and prevention of disease. J Virol 2002;76:4612.