A vesicular stomatitis virus glycoprotein epitope-incorporated oncolytic adenovirus overcomes CAR-dependency and shows markedly enhanced cancer cell killing and suppression of tumor growth

Abstract

Utility of traditional oncolytic adenovirus (Ad) has been limited due to low expression of coxsackie and adenovirus receptor (CAR) in cancer cells which results in poor infectivity of Ads. Here with an aim of improving the efficiency of Ad's entry to the cell, we generated a novel tropism-expanded oncolytic Ad which contains the epitope of vesicular stomatitis virus glycoprotein (VSVG) at the HI-loop of Ad fiber. We generated 9 variants of oncolytic Ads with varying linkers and partial deletion to the fiber. Only one VSVG epitope-incorporated variant, RdB-1L-VSVG, which contains 1 linker and no deletion to fiber, was produced efficiently. Production of 3-dimensionaly stable fiber in RdB-1L-VSVG was confirmed by immunoblot analysis. RdB-1L-VSVG shows a remarkable improvement in cytotoxicity and total viral yield in cancer cells. RdB-1L-VSVG demonstrates enhanced cytotoxicity in cancer cells with subdued CAR-expression as it can be internalized by an alternate pathway. Competition assays with a CAR-specific antibody (Ab) or VSVG receptor, phosphatidyl serine (PS), reveals that cell internalization of RdB-1L-VSVG is mediated by both CAR and PS. Furthermore, treatment with RdB-1L-VSVG significantly enhanced anti-tumor effect in vivo. These studies demonstrate that the strategy to expand oncolytic Ad tropism may significantly improve therapeutic profile for cancer treatment.

Keywords: CAR dependency; VSVG; fiber; oncolytic adenovirus; therapeutic efficacy.

Figure 1. Construction of VSVG epitope-incorporated fiber-modified oncolytic Ads

A. To construct VSVG-incorporated oncolytic Ad (RdB-VSVG), 9 variants of fiber shuttle vectors were constructed and utilized for homologous recombination with viral total oncolytic Ad vector (RdB). B. Polymerase chain reaction (PCR) analysis of a fiber-modified Ad (RdB-1L-VSVG). The fiber genotype was confirmed by PCR amplification with primers specific for the fiber. The 713 or 800 bp fiber genes from RdB (lane 1) or RdB-1L-VSVG (lane 2) were amplified respectively. Left lane is a DNA marker with 1-kb DNA ladder. C. Western blot analysis. A549 cells were infected with RdB or RdB-1L-VSVG at MOI of 10. Fiber monomer and trimer were observed under either denaturing or non-denaturing condition, respectively. Cell lysates were probed with antibodies against Ad fiber knob.

Figure 2. Cancer cell killing effect of RdB-1L-VSVG

A. MTT assay in CAR-positive cancer. CAR-positive various cancer cells (A549, U343, U87MG, Hep3B, C33A, and Hela) were treated with dE1, RdB, or RdB-1L-VSVG. At 2–4 days post infection, MTT assay was performed. B. MTT assay in CAR-negative cancer. CAR-negative cancer cells (MCF7 and MDA-MB-435) were treated with PBS, dE1, RdB, or RdB-1L-VSVG. At 4 days post infection, MTT assay was performed. C. MTT assay in normal fibroblast cells. Normal fibroblast cells (HDF and BJ) were treated with PBS, dE1, RdB, or RdB-1L-VSVG. At 4 days post infection, MTT assay was performed. Each cell line was tested at least three times and data shown are representative experiments. **P < 0.01, ***P < 0.001.

Figure 3. Mechanism of cellular uptake of RdB-1L-VSVG in various cancer cell lines

A. Competition analysis with coxsackie virus and adenovirus receptor (CAR)-specific Ab. U343, A549, and MCF7 cells were pre-incubated at 4°C using CAR-specific Ab (RmcB, 1 μg/ml). RdB or RdB-1L-VSVG was then added at an MOI of 20 (for U343 and A549 cells) or 100 (for MCF7 cells), respectively. Two days later, the cell viability was determined by MTT assay. All data are presented as means ± standard deviation (SD). ***P < 0.001. B. Competition analysis with phosphatidyl-serine (PS). U343, A549, and MCF7 cells were pre-incubated at 4°C using phosphatidyl-serine (PS) (16 μg/ml). RdB or RdB-1L-VSVG was then added at an MOI of 20 (for U343 and A549 cells) or 100 (for MCF7 cells), respectively. Two days later, the cell viability was determined by MTT assay. All data are presented as means ± standard deviation (SD). ***P < 0.001. Each cell line was tested at least three times and data shown are representative experiments.

Figure 4. Viral production

Monolayers of A549 (5 MOI), U343 (5 MOI), and MCF7 (200 MOI) cells were infected with RdB or RdB-1L-VSVG. The total virus present in cell and supernatant was extracted at 48 and 72 hours post-infection and titers were determined by TCID₅₀. Viral production assay was conducted at least three times, and data shown are representative experiments.

Figure 5. Anti-tumor effect and survival rate of oncolytic Ads in tumor xenograft model

Therapeutic efficacy of RdB or RdB-1L-VSVG against A549 and MCF7 tumors established in male athymic nude mice. Tumors were established by subcutaneous implantation of 1 × 10⁷ cells and allowed to grow to an average size of 100 mm³. Animals with established tumors were randomized into three treatment groups of 7–8 animals each and treatment was initiated (day 1). Each group received intratumoral injection of PBS, RdB, or RdB-1L-VSVG (5 × 10⁸ PFU for A549 and 1 × 10⁹ PFU for MCF7) on days 1, 3, and 5 (vertical arrows). A. Evaluation of anti-tumor effect of VSVG-incorporating oncolytic Ads. Tumor growth was monitored on a 2 day interval by measuring the short and long length of the tumor. Tumor volume was estimated on the basis of the following formula: volume = 0.523 LW². **P < 0.01, ***P < 0.001. B. Survival curve analysis of oncolytic Ads in tumor xenograft model. The percentage of surviving mice was determined by monitoring the tumor growth related events (A549; tumor size < 500 mm³, MCF7; tumor size < 1000 mm³) over a period.

Figure 6. Histological analysis of tumor tissue treated with oncolytic Ads

A549 and MCF7 tumors established in nude mice were injected with PBS, RdB, or RdB-1L-VSVG on day 1, 3, 5 and the tumors were harvested on day 7 for histological and immunohischemical analysis. A. H & E staining B. immunohistochemical staining of Ad E1A and C. TUNEL. Staining was performed in the sections of tumors treated with PBS, RdB, or RdB-1L-VSVG. D. Semi-quantitative analysis of E1A- or TUNNEL-stained sections using MetaMorph image analysis software. All results are shown as means ± STDEV. ***P < 0.001 compared with RdB.

Figure 6. Histological analysis of tumor tissue treated with oncolytic Ads

A549 and MCF7 tumors established in nude mice were injected with PBS, RdB, or RdB-1L-VSVG on day 1, 3, 5 and the tumors were harvested on day 7 for histological and immunohischemical analysis. A. H & E staining B. immunohistochemical staining of Ad E1A and C. TUNEL. Staining was performed in the sections of tumors treated with PBS, RdB, or RdB-1L-VSVG. D. Semi-quantitative analysis of E1A- or TUNNEL-stained sections using MetaMorph image analysis software. All results are shown as means ± STDEV. ***P < 0.001 compared with RdB.