An oncolytic adenovirus that expresses the HAb18 and interleukin 24 genes exhibits enhanced antitumor activity in hepatocellular carcinoma cells

Hepatocellular carcinoma (HCC) is characterized by alterations in multiple genes. High expression of CD147 on the surface of HCC cells promotes proliferation. The monoclonal antibody HAb18 recognizes CD147. We constructed an oncolytic adenoviral vector to express HAb18 (ZD55-HAb18) in HCC cells. Interleukin 24 (IL24) was co-expressed through the use of an F2A linker. ZD55-HAb18-IL24 decreased HCC cell viability to a greater degree than either ZD55-HAb18 or ZD55-IL24 alone. ZD55-HAb18-IL24 also induced apoptosis and autophagy in PLC/PRF/5 HCC cells. Intratumoral injection of ZD55-HAb18-IL24 repressed tumor growth in a PLC/PRF/5 xenograft model. Our results suggest that antibody-antitumor gene conjugation elicited a stronger antitumor effect than the antibody alone, and that this strategy could broaden the applications of antibody-based therapies in HCC.


INTRODUCTION
Hepatocellular carcinoma (HCC) is a multi-factorial disease that involves cross-talk between several pathways [1]. Combination approaches for HCC therapy based on multi-targeted conjugates have significant advantages [2]. Combination endostatin/sFlt-1 antiangiogenic gene therapy was shown to be highly effective in a rat model of HCC [3]. Complete eradication of HCC was achieved by combined vasostatin gene therapy and B7H3-mediated immunotherapy [4]. Drozdzik et al. demonstrated that a suicide gene in combination with interleukin-12 was more efficient than therapy with one gene alone in a murine model of HCC [5].
CD147 is a transmembrane glycoprotein that belongs to the immunoglobulin superfamily and is highly enriched on the surface of human HCC cells [6]. CD147 plays a critical role in tumor progression and metastasis [7]. Recent studies have revealed that knockdown of CD147 reduced cell proliferation and improved chemo-sensitivity in many cancer cells [8,9]. HAb18 is a singlechain monoclonal antibody fragment (scFv). The variable regions of the heavy chain and light chains are linked by a (GGGS) 3 peptide. HAb18 is secreted by hybridoma cells in BALB/c mice immunized with human HCC tissue extracts. The antigen recognized by HAb18 is CD147, which was identified in a screen of an HCC cDNA library [10][11][12]. Xu et al. reported that blocking CD147 with HAb18 mAb inhibited HCC growth and metastasis in vivo [13].
Interleukin 24 (IL24) is a cytokine that belongs to the IL-10 family of cytokines [14]. Preclinical studies have shown that ectopic expression of IL24 induces apoptosis in cancer cells with no significant cytotoxicity to normal cells [15,16]. IL24 has synergistic effects in various human cancers when combined with other agents. For example, Ad-IL24 combined with a selective inhibitor of EGFR (gefitnib) induces apoptotic cell death in nonsmall cell lung cancer [17]. In another study, treatment of Her-2/neu-overexpressing breast cancer cells with Ad-IL24 in combination with a monoclonal antibody targeting

Research Paper
the Her-2/neu receptor inhibited cell growth [18]. These studies indicate therapeutic antibodies combined with IL24 might enhance antitumor efficacy.
Oncolytic adenoviral vectors are promising cancer therapies [19][20][21]. The ONYX-015 vector was engineered to lack expression of the E1B55KD viral protein. Reduced replication of ONYX-015 in normal cells results from defective export of late viral RNA. This is because the E1B55KD protein facilitates preferential transport of viral RNA during the late stages of oncolytic adenoviral infection [22,23]. We constructed the oncolytic adenovirus ZD55, in which the E1B55KD gene was deleted and a cloning site added in order to insert foreign antitumor genes [24]. Previous studies have shown that ZD55 carrying the IL24 gene could selectively replicate in tumor cells and inhibit cell growth more effectively than ONYX-015 and a replication-defective adenovirus carrying the IL24 gene [25]. In another study, a combination of ZD55 carrying the TRAIL gene and ZD55 carrying the Smac gene suppressed the growth of HCC tumors in mice [26].
Our results indicate oncolytic adenoviruses carrying a combination of a therapeutic monoclonal antibody and a cytokine may be effective HCC therapies.

CD147 is highly expressed in HCC and promotes cell proliferation and chemoresistance
CD147 expression in HCC tissue (T) and adjacent non-cancerous tissue (N) was evaluated by quantitative real-time PCR (qRT-PCR) ( Figure 1A). CD147 was found to be overexpressed in tumor compared to normal tissue. Examination of CD147 expression in several HCC cell lines revealed higher CD147 mRNA and protein expression in cancer cells compared to normal human hepatocytes (QSG-7701 cells) ( Figure 1B and 1C).
To investigate the effects of CD147 on HCC cell proliferation, we overexpressed CD147 in PLC/PRF/5, Hep3B, and HepG2 cells using a lentiviral infection strategy. Cells that stably overexpressed CD147 were selected with puromycin ( Figure 1D and 1E). Overexpression of CD147 promoted cell proliferation ( Figure 1F) and conferred resistance to cytotoxic chemotherapy ( Figure 1G). We next knocked down CD147 using lentivirus-mediated shRNA ( Figure 1H and 1I). Knockdown of CD147 inhibited cell growth ( Figure  1J) and sensitized the cells to chemotherapy ( Figure  1K). These results demonstrated that CD147 enhanced cell proliferation and protected cells from the effects of chemotherapy, suggesting that it might be an attractive therapeutic target in HCC.

ZD55-HAb18-IL24 exhibits enhanced cytotoxicity in vitro
The effects of CD147 prompted us to assess whether inhibition of CD147 with the monoclonal antibody HAb18 could have therapeutic efficacy. We also simultaneously introduced IL24 into the oncolytic adenoviral vector since multi-target agents have been widely investigated for cancer treatment. We hypothesized that a combination of HAb18 and IL24 could improve the therapeutic efficacy. To test this hypothesis, we constructed an oncolytic adenovirus that expressed the HAb18 (ZD55-HAb18) and IL24 (ZD55-IL24) genes. We linked the two genes with an F2A sequence and cloned the product into an oncolytic adenoviral vector to generate ZD55-HAb18-IL24 ( Figure  2A). PCR analysis of the E1B-55KD gene confirmed that all the oncolytic adenoviruses were free of wild-type adenovirus contamination ( Figure 2B).
The expression of exogenous HAb18, IL24, and GFP in PLC/PRF/5 cells was detected by western blot ( Figure 2C). PLC/PRF/5 cells infected with ZD55-HAb18-IL24 expressed HAb18 and IL24 individually with bands corresponding to the correct molecular weights. The expression of both proteins in ZD55-HAb18-IL24-treated cells was similar to expression in cells infected with either ZD55-HAb18 or ZD55-IL24 individually, indicating that the F2A peptides mediated efficient generation of the individual HAb18 and IL24 proteins. We observed decreases in MMP-2 and MMP-9 expression in ZD55-HAb18-and ZD55-HAb18-IL24-treated PLC/PRF/5 cells, which suggested that oncolytic adenoviral expression of mAb HAb18 inhibited CD147 function ( Figure 2D).
To compare the abilities of the oncolytic adenoviruses to replicate in tumor and normal cell lines, PLC/PRF/5 and QSG-7701 cells were infected with oncolytic adenoviruses at an MOI of 10. The relative replication efficacy of the oncolytic adenoviruses was measured by qRT-PCR of adenovirus gene E1A cDNA. The oncolytic adenoviruses replicated effectively in tumor cells ( Figure 2E). In contrast, the replication ability was significantly reduced in normal cells. These results demonstrated that oncolytic adenoviruses selectively replicated in tumor cells.

ZD55-HAb18-IL24 induces apoptosis and autophagy
We assessed whether ZD55-HAb18-IL24 could induce apoptosis in PLC/PRF/5 cells 48 hours after oncolytic adenovirus infection using Annexin V/propidium iodide (PI) staining. The percentage of apoptotic cells (Annexin V + /PI + ) in the ZD55-HAb18-IL24-treated group was higher than in the other groups ( Figure 4A). We also detected several key members of caspase-dependent apoptotic signaling cascades by western blot. Enhanced  poly ADP-ribose polymerase (PARP) cleavage, and decreased expression of procaspase-3 and procaspase-9 were observed in ZD55-HAb18-IL24-treated PLC/ PRF/5 cells ( Figure 4B). Because cleaved caspase-9 is an essential initiator of the mitochondrial apoptosis signaling pathway, we analyzed changes in the mitochondrial membrane potential (Δψm) using JC-1 staining. An increase in the number of cells with mitochondrial membrane potential loss was observed in ZD55-HAb18-IL24-treated PLC/PRF/5 cells ( Figure 4C). These data suggested that ZD55-HAb18-IL24 induced apoptosis in PLC/PRF/5 cells through the mitochondrial pathway.

ZD55-HAb18-IL24 suppresses tumor growth in vivo
To evaluate the therapeutic potential of oncolytic adenoviruses in vivo, oncolytic adenoviruses and corresponding volumes of PBS were intratumorally injected in PLC/PRF/5 xenografts. Treatment with ZD55-HAb18-IL24 delayed tumor growth compared to treatment with the other oncolytic adenoviruses ( Figure  5A). The mice were sacrificed 33 days after injection. The tumor volume and weight in the ZD55-HAb18-IL24-treated group were 297.8 ± 67.5 mm 3 and 47 ± 0.05 g, respectively. These tumors were significantly smaller than the tumors in the other groups ( Figure 5B and 5C), indicating that ZD55-HAb18-IL24 was able to efficiently inhibit PLC/PRF/5 xenograft growth. HAb18 and IL24 expression in tumors was assessed by qRT-PCR, and the serum HAb18 level examined by enzymelinked immunosorbent assays (ELISA) at the end of the experiment. ZD55-HAb18-IL24 expressed high levels of HAb18 and IL24 in tumors ( Figure 5D). The expression of HAb18 in serum remained relatively high for > 1 month ( Figure 5E). We analyzed the kinetics of HAb18 expression in nude mice bearing PLC/PRF/5 tumors. These results demonstrated that ZD55-HAb18 and ZD55-HAb18-IL24 efficiently expressed HAb18 (Supplementary Figure S2). Additionally, no significant increases in the expression of aspartate aminotransferase (AST), alanine aminotransferase (ALT), blood urea nitrogen (BUN), or creatinine were observed after treatment (Supplementary Figure S3).
The apoptotic and autophagic effects of ZD55-HAb18-IL24 in PLC/PRF/5 xenografts were analyzed using terminal deoxynucleotidyl transferase dUTP nick end labeling (TUNEL) assays. A higher percentage of positive cells was observed in the ZD55-HAb18-IL24treated tumors than in the other groups ( Figure 5F and 5G). Tumors in the ZD55-HAb18-IL24-treated groups displayed more autophagic structures than tumors in the other groups, and no autophagic structures were observed in PBS-treated tumors by electron microscopy ( Figure 5F). Because angiogenesis plays a critical role in tumor progression, we examined the expression of VEGFA in adenovirus-treated cells in vitro and in vivo. A significant decrease in VEGFA expression was observed after adenovirus infection (Supplementary Figure S4). Immunohistochemical staining of CD31, a vascular endothelial marker, revealed reduced microvessel density in ZD55-HAb18-IL24-treated tumors ( Figure 5F and 5H).

DISCUSSION
HCC can result from both genetic and epigenetic alterations [29]. Many factors involved in HCC development are potential therapeutic targets. CD147 is an HCC-associated antigen and plays important roles in HCC progression [30]. Our data indicate that CD147 promotes HCC cell proliferation and chemoresistance (Figure 1). Overexpression of CD147 was also shown to promote proliferation and chemoresistance in bladder cancer T24 cells (Supplementary Figure S5). These findings suggest that CD147 might be a broad-spectrum target for cancer therapy.
Monoclonal antibodies are advantageous for the treatment of cancer [31]. In order to achieve clinical therapeutic effects, antibody levels in serum must be high [32]. Adeno-associated virus vectors that transferred the anti-HIV mAb gene yielded low levels of antibody [33]. Recombinant oncolytic Newcastle Disease Virus carrying the full-length mouse-human chimeric HAb18 gene (cHAb18) induced HCC cell necrosis and prolonged survival in mice through inhibition of local tumor metastasis [34]. Here, we used the oncolytic adenoviral vector ZD55 to mediate HAb18 mAb expression in combination with the therapeutic cytokine IL24, which were linked by an F2A sequence. High expression of HAb18 and IL24 could still be observed 33 days after injection of ZD55-HAb18-IL24 into nude mice bearing PLC/PRF/5 xenograft tumors ( Figure 5D and 5E). However, when the vector was injected into to C57/BL6 mice bearing syngeneic colon tumors, low serum HAb18 expression was observed. This could have resulted from defective replication of the oncolytic adenovirus in murine cells (Supplementary Figure S6). www.impactjournals.com/oncotarget Oncolytic adenoviral vectors can be generated using two strategies. One strategy is to use a tumorspecific promoter to drive an essential viral replication gene. Another strategy is to delete a viral element that is required for virus replication in normal cells but dispensable in tumor cells. The adenovirus E1B55KD enhances cyclin E expression and is required for efficient viral DNA replication [35]. E1B55KD deletion might attenuate oncolytic adenovirus replication in tumor cells. Luo et al. constructed the AdCN205-IL24 vector in which the hTERT promoter was used to control the expression of E1A in which the CR2 region was deleted. The expression of IL24 was controlled by the adenoviral E3 promoter. AdCN205-IL24 could express IL24 and suppress tumor growth [36].
Blocking HAb18G/CD147 using mAb HAb18 was shown to inhibit HCC growth in vivo [37]. Licartin ( 131 I-labled HAb18) was approved for clinical use by the Food and Drug Administration of China in April 2005 and is an effective and safe treatment for HCC [38]. IL24 has been widely investigated because it has selective antitumor effects in HCC [36,39,40]. Ad-IL24 was effective in phase I/II clinical trials in patients with advanced cancers [41]. We hypothesized that the combination of the HAb18 and IL24 could result in enhanced antitumor effects. We demonstrated that oncolytic adenoviruses carrying both genes more effectively inhibited HCC cell growth than either gene alone both in vitro and in vivo. Furthermore, increases in apoptosis and autophagy were observed in ZD55-HAb18-IL24-treated cells (Figure 4 and 5). Collectively, our data indicate that ZD55-HAb18-IL24 has anti-tumor effects and is a potential therapeutic strategy for HCC.

Patient samples
Adjacent, non-cancerous tissue samples were anonymously obtained from HCC patients treated at the Center of Clinical Oncology at the Affiliated Hospital of Xuzhou Medical College. All human samples were obtained with informed consent, and approval for usage was obtained from the Ethics Committee of the Affiliated Hospital of Xuzhou Medical College. Studies using these samples were approved by the Institutional Review Board of the Institute of Biochemistry and Cell Biology, Shanghai Institutes for Biological Sciences, Chinese Academy of Sciences.

Cell culture
The human HCC cell lines (PLC/PRF/5, HepG2, BEL-7404, and Hep3B), human bladder cancer cell line (T24), human hepatocytes (QSG-7701), HEK-293T cells, and the murine colon cancer cell line (MC38) were purchased from the Cell Bank of the Type Culture Collection of the Chinese Academy of Sciences (Shanghai, China). The HEK-293 cell line was obtained from Microbix Biosystems Inc. (Toronto, Canada). All cells were cultured according to the manufacturer's instructions.

Adenovirus construction
The GFP, IL24, HAb18, and HAb18-F2A-IL24 expression cassettes were inserted into pShuttle-E1B (Δ55) with E1B55K deletion. Oncolytic adenovirus plasmids were generated by homologous recombination of the shuttle vector and the adenoviral backbone plasmid in E. coli BJ5183 cells. Oncolytic adenoviruses were packaged and amplified in HEK-293 cells, and then purified by gradient CsCl centrifugation. Virus titer was measured using the Quick Titer Adenovirus Titer Immunoassay Kit (Cell Biolabs, San Diego, CA, USA), or plaque assays (HEK293 cells). Viral genomic DNA was extracted using the Blood Genome Extract Kit and the manufacturer's protocol (Generay, Shanghai, China) for identification. The absence of wild-type contamination was demonstrated by PCR with corresponding primers (Supplementary Table  S1). The F2A sequence was generated from the FMDV 2A sequence through the addition of a furin cleavage site sequence (RAKR) to the N-terminus of the FMDV 2A sequence.

Quantitative RT-PCR
Total RNA was isolated using TRIzol (CWBIO, Beijing, China). Single-strand cDNA was synthesized using the ReverTra Ace qPCR RT Kit (Toyobo, Osaka, Japan). RNA expression was analyzed using SuperReal Premix Plus (TIANGEN, Beijing, China) according to the manufacturer's protocols. The sequences of all primers are shown in Supplementary Table S1.

Proliferation assays
Cell proliferation was measured using the CCK-8 kit (Dojindo, Kumamoto, Japan). Following treatment, the cells were incubated with 10 μL CCK-8 at 37°C for 2 h. Absorbance was measured at 450 nm and 630 nm using a Biotek Eon Microplate Reader.

Migration assays
Cell migration assays were performed using transwell inserts (8 μm, BD Biosciences, San Jose, CA, USA) placed in 24-well plates. Cells were cultured in the upper chambers of the transwell inserts in 200 μL serum-free medium. The medium in the bottom chamber contained 10% fetal bovine serum. Following a 12 h incubation, the inserts were stained with 0.5% crystal violet and imaged.

Flow cytometry
Apoptosis was analyzed using the Annexin V-FITC Apoptosis Detection kit (Beyotime Biotechnology, Shanghai, China) according to the manufacturer's instructions. The mitochondrial membrane potential was evaluated by staining the cells with the JC-1 fluorescent probe (BD Biosciences) and then subjecting them to flow cytometry using a FACS Calibur flow cytometer (BD Biosciences).

Immunofluorescence microscopy
Cells were fixed with 4% (w/v) paraformaldehyde, permeabilized using 0.1% (w/v) Triton X-100 and blocked with 1% bovine serum albumin. The cells were stained with an anti-LC3 antibody (Sigma) overnight and then incubated with a Cy3-labeled secondary antibody in PBS with 1% FBS for 1 hour at room temperature. Images were captured using a laser scanning confocal microscope (Olympus, Tokyo, Japan).

Animal experiments
All in vivo experiments were performed according to protocols approved by the U.S. Public Health Service Policy on Humane Care and Use of Laboratory Animals. Four-week-old female BALB/c nude mice and C57/BL6 mice were purchased from the Animal Core Facility (Shanghai, China). We pre-mixed 5 × 10 6 PLC/PRF/5 cells with Matrigel (BD Biosciences) at a 1:1 ratio and subcutaneously injected the cells into the right flank of each mouse. A total of 1 × 10 6 MC38 cells were injected into C57/BL6 mice. When the tumors reached 80-120 mm 3 , the mice were randomly divided into five groups (six mice per group). Oncolytic adenovirus (2.5 × 10 8 PFU per mouse) or PBS was intratumorally injected every other day (four injections total). The tumor volume was measured using a Vernier caliper every 4 days and calculated using the following equation: (length × width 2 ) / 2. At the end of the experiment, the tumors were resected from the sacrificed mice for immunohistochemical analysis. The levels of AST, ALT, creatinine, and BUN in serum were measured.

ELISA
Serum HAb18 expression was quantified by ELISA using a human IgG ELISA detection kit (Mlbio, Shanghai, China). All experiments were performed according to the manufacturer's protocols.

Immunohistochemistry and TUNEL assays
Tumor tissue was fixed in 4% formaldehyde overnight, embedded in paraffin, and sectioned (5 μm thickness) for immunohistochemical analysis. Sections were stained with an anti-CD31 antibody (Cell Signaling Technology, Danvers, MA, USA). TUNEL assays were performed according to the manufacturer's protocol (Roche, Basel, Switzerland). Hematoxylin was used as a counterstain.

Transmission electron microscopy
Tumor tissue was fixed with 2.5% glutaraldehyde overnight followed by 1% osmium tetroxide for 1.5 h. The tissue samples were then dehydrated using a graded series of ethanol. Samples were then rinsed with acetone and permeated overnight with embedding buffer. Sections of 70 nm thickness were dual-stained with 2% uranyl acetate and lead citrate. Autophagosomes were examined by transmission electron microscopy (FEI, Hillsboro, OR, USA).

Statistical analysis
Data are presented as the mean ± standard deviation (S.D). Comparisons between two groups were performed with Student's t-tests and the GraphPad Prism 6.0 software (GraphPad Software, USA).

ACKNOWLEDGMENTS
We thank Dr. Zhi-Nan Chen of the Fourth Military Medical University for providing the mAb HAb18 www.impactjournals.com/oncotarget construct. We thank Lan-Ying Sun for assistance with cell culture and the staff at the Cell Analysis Center and Animal Core Facility (Institute of Biochemistry and Cell Biology, Shanghai Institutes for Biological Sciences, Chinese Academy of Sciences) for providing technical assistance.