Adenovirus-mediated Gene Therapy for Human Head and Neck Squamous Cell Cancer’

Inactivation of the tumor suppressor gene j6Th (4 is the most common genetic alteration in human head and neck squamous cell cancer (HNSCC), making it an ideal target for gene replacement. We constructed a replicationdefective, recombinant adenovirus capable of directing a high level of J6INK4A protein expression (AdS-p16) to investigate its benefit in treating HNSCC. Initial in vitro experiments in four human HNSCC cell lines demonstrated that AdS-p16 treatment significantly inhibits cell growth with up to 96% efficiency. Flow cytometric analysis showed that AdS-p16 induced a maximum G1-S cell cycle arrest of 90%. Subsequent studies in a nude mouse model demonstrated that AdS-p16 treatment significantly reduced (cell line 011) or stabilized (ceilline 012) established tumors when compared with control treatments (P < 0.008). These results demonstrate for the first time a significant antitumor effect of Ad5-p16 against human HNSCC in vivo and support the potential application of AdS-p16 to treat locally advanced, unresectable, or metastatic head and neck cancer, as well as microscopic residual disease after surgical resection. INTRODUCTION Worldwide, there are approximately 500,000 new cases of HNSCC3 each year, with over 50,000 of these cases occurring here in the United States (1). Standard therapies for advanced HNSCC include radical surgical procedures with adjuvant radiotherapy and the possible use of chemotherapy in selected Received 12/15/97; revised 4/15/98; accepted 4/17/98. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. ‘ This work was supported in part by NIH/National Institute of Dental Research Grant R29 DEl 1 722-02 (to B. W. 0.). 2 To whom requests for reprints should be addressed, at Gene and Molecular Therapy Division, Department of Otolaryngology, Head and Neck Surgery, P. 0. Box 41402, Baltimore, MD 21203-6402. Phone: (410) 955-8409; Fax: (410) 955-0035. 3 The abbreviations used are: HNSCC, head and neck squamous cell carcinoma; CDK, cyclin-dependent kinase; CMV, cytomegalovirus; Rb, retinoblastoma; PFU, plaque-forming unit(s); X-Gal, 5-bromo-4-chloro3-indolyl3-D-galactopyranoside. cases. Surgical resection frequently results in significant cosmetic deformity, along with functional deficits in speech, swallowing, and upper extremity strength. Radiation and chemotherapy are not innocuous and also bring substantial morbidity including: mandible and laryngeal cartilage radionecrosis, tissue fibrosis, mucosal atrophy, and xerostomia. as well as gastrointestinal, bone marrow, renal, and ototoxicity (2). Unfortunately, despite the widespread use of these aggressive therapeutic modalities, the overall rate of survival for patients with HNSCC has failed to significantly improve over the last 30 years. Most studies still report overall survival rates of -45%, with 2-year survival rates of 30% or less for patients with advanced stage III or IV disease (3). As a consequence of these poor outcomes in advanced HNSCC, new treatment strategies based on the therapeutic application of genes are being investigated. Because most treatment failures in HNSCC patients tend to be local-regional, with less than 10% of patients succumbing to distant metastasis alone (4), these patients may be ideal candidates for direct gene transfer to persistent or recurrent disease. Promising gene therapy strategies to date have been founded on the use of highly efficient adenovirus vectors to deliver therapeutic genes to advanced HNSCC. The herpes virus thymidine kinase gene (suicide gene therapy) and the tumor suppressor gene p53 are the two most widely studied therapies in preclinical trials (5-9). The tumor suppressor gene 16 4A is also a strong candidate HNSCC gene therapy for several reasons: (a) mactivation of J6INK4A the most frequent genetic alteration in human HNSCC, with over 80% of primary HNSCCs inactivating J6INK4A through deletion, mutation, or promoter methylation (10); (b) the transfection of plasmid vectors containing p16 4’ ’ into numerous tumor cell lines (1 1), including HNSCC cell lines (12), has led to both growth suppression and G,-S cell cycle arrest; (c) recombinant j6 (4A adenoviruses have been shown to inhibit in vitro and in vivo tumor cell proliferation in cell lines derived from human lung (13) and prostate carcinomas (14), as well as suppressing glioma (15) invasion in vitro. The J6D4K4A gene encodes for a 156-amino acid protein that was initially identified as a specific inhibitor of the CDKs, CDK4 and CDK6 (16). CDK4 and CDK6 are the major catalytic partners of cyclins Dl, D2, and D3 and regulate progression through the G,-S transition of the cell cycle by assisting in the cyclin D-dependent phosphorylation of the retinoblastoma susceptibility gene product, pRb (17). Phosphorylation of pRb is equivalent to functional inactivation and results in the release of transcription factors essential for S-phase progression (18, 19). Thus, by binding to and inhibiting CDK4 and CDK6, J6IN (4A prevents both pRb phosphorylation and subsequent progression into the S phase of the cell cycle (20). In a previous study, we transfected plasmids containing full-length J6INK4A eDNA under the control of a CMV promoter into three well-characterized HNSCC cell lines and demResearch. on November 1, 2017. © 1998 American Association for Cancer clincancerres.aacrjournals.org Downloaded from 1698 plOI 4 4A Adenovirus-mediated Gene Therapy for Human HNSCC Table 1 Characterizati on of HNSCC cell lines Cell line Tissue origin p16 p53 Rb 011 012 022 029 Larynx Oral cavity Larynx Larynx WT’ Meth” WI’ HDC WT WI WI WI WT WI WI WI a Sequence is wild-type (WI) with low but detectable endogenous protein by Western blotting. b Hypcrmethylated promoter region that fails to make detectable endogenous pI6INK4A protein by Western blotting. C HD. homozygous deleted. onstrated marked growth inhibition in vitro through a G,-S cell cycle arrest (12). However, to fully explore the potential of j6I (4A as a therapeutic agent against HNSCC, we needed to overcome the low transfection efficiencies inherent in plasmidbased transfection techniques. We subsequently constructed a replication-defective recombinant adenovirus, Ad5-p16, to take advantage of the high transduction efficiency of the adenovirus system. Major advantages of the recombinant adenovirus system are: its well-documented ability to infect multiple cell types; its capability to infect nonproliferating cells; and the ease in which a recombinant adenovirus can be produced at the high titers needed for in vivo use. To directly examine the therapeutic potential of ]6INK4A against human HNSCC, we treated four well-characterized human HNSCC cell lines with Ad5-p16. AdS-p16 was able to significantly arrest tumor cell growth both in vitro and in vivo, regardless of the status of the endogenous pl&N Z4 gene, demonstrating that effective in vivo treatment of human HNSCC in nude mice can be achieved using adenovirusmediated transfer of the human J6IN (4A gene. MATERIALS AND METHODS Cell Lines. Primary tumor explants were derived from squamous cell carcinomas of the head and neck from patients at the Johns Hopkins Hospital Department of Otolaryngology (cell lines 01 1, 012, 022, and 029) and have been well characterized by both double-stranded DNA sequencing and Western blot analysis (Table 1 ). All four human HNSCC cell lines were maintained in RPMI 1640 with 10% fetal bovine serum and 1% penicilbin/streptomycin. Cell line 293 (American Type Culture Collection) was maintained in DMEM (Life Technologies, Inc.) with 10% fetal bovine serum and 1% penicillin/streptornycin. All cell lines were maintained at 37#{176}C, 5% CO2 with humidification, and passaged just prior to 100% confluence. J6Ir 4 cDNA Subcboning and the Construction of pAdl.CMVp16. Human pj6hr 4A eDNA in pBluescript (pBS-p16) from a HeLa cell eDNA library was confirmed by double-stranded DNA sequencing prior to being subcloned into the adenovirus vector pAdl.CMV (gift of Savio Woo, Mount Sinai Hospital, New York, NY). pBS-p16 was than sequentially digested with restriction enzymes PvuII and BamHI to generate a I . 1-kb fragment containing the human J6IN (4A eDNA, which was then directionally subcloned into vector pAd 1.CMV at the BamHI and Echo R5 restriction sites of the polylinker. The final product, pAdl.CMVpI6, contained an expression cassette that placed the J6LN (4A eDNA under the control of the strong CMV immediate-early gene promoter (21), along with adenoviral sequences required for in vivo recombination with the helper plasmid pJMl7. Generation and Confirmation of Recombinant Adenovirus AdS-p16. To generate recombinant adenovirus, plasmids pAdl.CMVpI6 and adenoviral helper plasmid pJM17 (Microbix) were cotransfected into 85-90% subconfluent 293 cells using a Lipofectamine (Life Technologies, Inc.)-based protocol. Transfected 293 cell monolayers were overlaid with 0.5% agarose in DMEM and maintained as described until the appearance of plaques (22). To obtain both a clonal recombinant virus and increase viral titers by eliminating defective virions, recombinant plaques were purified a minimum of three times prior to amplification using established methods (22). Prior to amplification, recombinant plaques were screened by PCR analysis using a single set of primers: CMV-5 (GGTCTATATAAGCAGAGC), an 18-mer, 5’ 3’ primer specific to the CMV promoter; and pl6-EX1-3 (GTCAGCCGAAGGCTCCAT), an 18-mer, 3’ -* 5’ primer specific to exon 1 of J6IN (4A, to confirm the presence of j61N (4 eDNA within the selected plaques. A second round of PCR analysis was used to screen the viral lysate from Ad5-p16 plaques for wild-type adenovirus contamination (23) prior to large-scale viral purification. Amplified individual clones of Ad5-p16, free from wild-type adenovirus contamination, were purified by cesium chloride ultracentrifugation using established methods (22) and stored at -80#{176}Cin single-use aliquots. The viral titer (PFU/ml) of purifled Ad5-p16 stocks were determined by plaque assays on 293 cells using standard methods (22). The recombinant adenovirus AdS-lacZ (gift from Dr. Frank Graham, McMaster University, Hamilton, Ontario, Canada) carries the lacZ gene from Escherichia coli under the control of the strong CMV immediate-early gene promoter and is similar to Ad5-p16. 3-Galactosidase Transduction Assay. To assess the ability of our adenoviral vectors to transfer genes to all four of our HNSCC cell lines, monolayers of all four cell lines were treated with Ad5-lacZ and treated 48 h later with X-Gal. Increasing doses of Ad5-lacZ, from 0 to 50 PFU/cell, were used to determine a multiplicity of infection to maximize gene transfer (>85% stain positive) and minimize viral toxicity. The percentages of positive-staining cells were determined by scoring 500 cells on triplicate dishes. Western Blotting. Confirmation 0fpJ6IN 4A protein cxpression after AdS-p16 infection was performed in all four HNSCC cell lines. Duplicate samples of log phase cells were seeded at a density of 5 X l0 in 35-mm2 culture dishes 24 h before treatment. Cells were either mock-treated with PBS or treated with the viruses Ad5-lacZ or Ad5-p16 as indicated. Cell monolayers were incubated with 0.5 ml of media containing virus at a concentration of 5 PFU/cell for 1 h at 37#{176}C before adding an additional 2.5 ml of media and continuing the incubation for an additional 23 h. The virus-containing medium was removed by aspiration, and cells were incubated in 2.0 ml of fresh media and maintained at 37#{176}C. Forty-eight h after treatment, the medium was removed, and the cell monolayers were washed with PBS at room temperature. Cell monolayers were than lysed by adding 0.2 ml of ice-cold RIPA buffer containing 100 ig/m1 of phenylmethylsulfonyl fluoride, followed by sonication for 3 5 to reduce viscosity. The resulting cell lysate was than incubated on ice for 1 h before centrifuging at 15,000 X g Research. on November 1, 2017. © 1998 American Association for Cancer clincancerres.aacrjournals.org Downloaded from 011 012 022 029 Clinical Cancer Research 1699