Oncosuppressive suicide gene virotherapy "PVH1-yCD/5-FC" for pancreatic peritoneal carcinomatosis treatment: NFκB and Akt/PI3K involvement

Abstract

Peritoneal carcinomatosis is common in advanced pancreatic cancer. Despite current standard treatment, patients with this disease until recently were considered incurable. Cancer gene therapy using oncolytic viruses have generated much interest over the past few years. Here, we investigated a new gene directed enzyme prodrug therapy (GDEPT) approach for an oncosuppressive virotherapy strategy using parvovirus H1 (PV-H1) which preferentially replicates and kills malignant cells. Although, PV-H1 is not potent enough to destroy tumors, it represents an attractive vector for cancer gene therapy. We therefore sought to determine whether the suicide gene/prodrug system, yCD/5-FC could be rationally combined to PV-H1 augmenting its intrinsic oncolytic activity for pancreatic cancer prevention and treatment. We showed that the engineered recombinant parvovirus rPVH1-yCD with 5-FC treatment increased significantly the intrinsic cytotoxic effect and resulted in potent induction of apoptosis and tumor growth inhibition in chemosensitive and chemoresistant cells. Additionally, the suicide gene-expressing PV-H1 infection reduced significantly the constitutive activities of NFκB and Akt/PI3K. Combination of their pharmacological inhibitors (MG132 and LY294002) with rPVH1-yCD/5-FC resulted in substantial increase of antitumor activity. In vivo, high and sustained expression of NS1 and yCD was observed in the disseminated tumor nodules and absent in normal tissues. Treatment of mice bearing intraperitoneal pancreatic carcinomatosis with rPVH1-yCD/5-FC resulted in a drastic inhibition of tumor cell spreading and subsequent increase in long-term survival. Together, the presented data show the improved oncolytic activity of wPV-H1 by yCD/5-FC and thus provides valuable effective and promising virotherapy strategy for prevention of tumor recurrence and treatment. In the light of this study, the suicide gene parvovirotherapy approach represents a new weapon in the war against pancreatic cancer. Moreover, these preliminary accomplishments are opening new field for future development of new combined targeted therapies to have a meaningful impact on advanced cancer.

Figure 1. Schematic representation of recombinant parvoviral vector constructions.

The upper diagram depicts the empty vector DNA clone PVH1-Δ800 which has 800-bp deletion in the VP coding region and carries a multiple cloning sequence (MluI/SmaI polylinker) at the VP2 translation initiation site. DNA inserts encoding for GFP or catalytic yCD were introduced in polylinker using XhoI and BamH1 restriction enzymes. The resulting rPVH1-GFP and rPVH1-yCD plasmids were used for the production of corresponding non-replicative recombinant parvoviruses.

Figure 2. Virus infection and transduction efficiency.

A) Cytopathic effect of wtPV-H1. The tumor cell lines BxPc3, Panc1 and AsPc1 were infected at MOIs ranging from 1 to 100 and after 72 hours, cell viability was measured by MTT assay. Indicated data represent the mean of three independent experiments realized in triplicate. B) For pancreatic tumor cell infection study, the cell cultures were blocked in G0/G1 phase by serum starvation for 72 hours and thereafter, tumor cells were infected with wtPV-H1or rPVH1-yCD at MOI 10 for 12 hours, and subsequently transferred on nitrocellulose filters. The detection of NS1 viral gene was realized according to the protocol of PV-H1 titration mentioned above. The detected radioactive spots reflect the number of NS1-hybridized probes. Asterisks indicate significant difference (*** p<0.001) observed between permissive cells (Panc1 and AsPc1) and non-permissive (or low) BxPc3 cells. C) Representative microscopy fluorescence images of AsPc1, Panc1 and BxPc3 cell cultures infected with rPVH1-GFP. Cells were infected at MOI 5 and 10 during 48 hours. D) Fluorescence intensity was quantified by flow cytometry. Results represent the percentage of fluorescent cells reported to total cells. E) For transduction efficiency. Kinetics of NS1 and yCD expression were carried out on AsPc1, Panc1 and BxPc3 cells. Tumor cells were infected with rPVH1-yCD (MOI 10) and total RNA extractions were performed 12, 24, 48 and 72 h post-infection and subjected to real time RT-PCR measurement. Relative quantitation of gene expression was achieved by normalization against the endogenous GAPDH housekeeping gene expression. Error bars represent the standard error of three independent experiments and run in quadruplets.

Figure 3. rPVH1 expressing yCD exhibits significant oncolytic effects in human pancreatic tumor cells.

A) Cytotoxic effect of rPVH1-yCD/5-FC treatment. Pancreatic tumor cells were infected with rPVH1-yCD at MOIs (1 to 100) and after 48 hours, cells were either left untreated or treated for three days with different concentrations of 5-FC. Cell survival was assessed by MTT assays and the percentage of cell viability was calculated by comparison to control cell cultures (untreated cells). B) Clonogenic survival assay. Cells were non-infected or infected with wtPV-H1, rPVH1-GFP or rPVH1-yCD (MOI 10). Cells infected with rH1-yCD were further treated with 5-FC (250 µg/ml) for 48 hours. Then, cells were trypsinized and replated at low density as described in Materials and methods. After 14 days, colony formation was visualized using Giemsa staining and the number of colony was calculated. The number of colony-forming units from non-treated cells was defined as 100% of the surviving fraction. Data represent the mean of three experiments realized in triplicate. Asterisks indicate significant difference observed in PVH1-treated cells compared to untreated cells (mock) (*p<0.05, ***p<0.001). C) Apoptosis induction. Tumor cells were infected with wtPV-H1 or rPVH1-yCD or rPVH1-GFP (MOI 10). After 48 hours of 5-FC treatment, apoptotic cells were assessed using AnnexinV/PI staining. Results represent mean percentage of apoptotic cells ± s.e.m. of three independent experiments. Asterisks indicate significant difference (***p<0.001) observed in rPVH1-yCD/5-FC treated cells compared to wtPVH1-treated cells. D) in parallel, tumor cells were harvested and subjected to caspase3/7 assays. Results represent the average ± s.e.m of three independent experiments and expressed as fold induction compared with untreated control (*p<0.05, ***p<0.001). E) Western blot of apoptogene expression after AsPC1 tumor cell infection with wtPV-H1 or PVH1-yCD plus 5-FC treatment. Data indicate expression levels of PARP, Bax and TRAIL compared to the constitutive β-Actin protein. Similar expression profile was observed for BxPc3 and Panc1.

Figure 4. Effect of rPVH1-yCD/5FC treatment on NFκB and Akt/PI3K activity.

We first investigated the NFκB constitutive activity in AsPc1, Panc1 and BxPc3 tumor cell lines and compared it to wtPV-H1 (A) and rPVH1yCD/5FC (B) -infected tumor cells. Thus, tumor cells were transfected with plasmids expressing the luciferase reporter gene, pNFκB-Luc and 24 hours later, cells were infected with wtPV-H1 or rPVH1-yCD. One set of rPVH1yCD-infected cells was further treated with 5-FC. After 48 hours, cells were lysed and luciferase activity was measured using the same amount of protein. Reported data are the mean of three experiments performed in triplicate. The high levels of constitutive NFκB activity observed in AsPc1 and Panc1 were reduced slightly after wtPV-H1 infection (*p<0.05). The rPVH1-yCD/5FC treatment resulted in a significant reduction of constitutive NFκB in Panc1 (**p<0.01) and AsPc1 (***p<0.001). The rPVH1-yCD/5FC data were further confirmed by EMSA experiments. Whole cell extracts were prepared as described in Materials and methods. The specificity of the band corresponding to NFκB-DNA complex was determined using non-labeled NFκB consensus oligonucleotide as a specific competitor. C) wtPV-H1 and rPVH1-yCD/5-FC decrease Akt/PI3K constitutive activity. After 48 h of infection and 5-FC treatment, cell lysates were prepared and used to analyze the protein expression of Akt, PI3K and their phosphorylation forms by Western blotting. Compared to the control (1) indicating the high constitutive activity, the pAkt (Ser473) and pPI3K (p85) were strongly inhibited by wtPV-H1 (2) and rPVH1-yCD/5-FC (3). D) MG132 and LY94002 reduce significantly the transcriptional activity of NFκB. Data represent the fold induction compared to untreated control cells, *p<0.05, **p<0.01 and ***p<0.001. E) NFκB and Akt/PI3K inhibitors (MG132, LY294002) combination enhances the cytopathic effects induced by wtPV-H1 and rPVH1-yCD/5-FC. The indicated mean percentages of cell death are representative of three experiments realized in triplicate. The (†) symbol represents a significant difference between wtPV-H1 or rPVH1-yCD/5-FC treatment alone and their combination with NFκB or PI3K/Akt inhibitors; ††p<0.01, †††p<0.001. The (*) symbol indicates a significant difference between wtPV-H1 or rPVH1-yCD/5-FC treatment and MG132 or LY294002 inhibitors; *p<0.05, **p<0.01.

Figure 5. In vivo antitumor activity of wild type and GDEPT oncosuppressive parvoviruses on pancreatic peritoneal carcinomatosis AsPc1 tumor model.

Mice with peritoneal carcinomatosis (n = 12) received i.p injection of wtPV-H1, rPVH1-GFP or rPVH1-yCD (1.10⁸ particles). Mice treated with rPVH1-yCD also received a daily 5-FC (250 mg/kg/day) i.p administration for 14 days. A and B) Real time quantitative RT-PCR measurement and Western blot analysis of NS1, GFP and yCD expression in tumor nodules and normal tissues. For wtPV-H1 infection, two mice per group were sacrificed and relative expression of NS1 was assessed in different organs from tumor-free and tumor-bearing animals. As well, the expression of NS1, GFP and yCD was evaluated on different biopsies arising from tumor-bearing animals injected with either rPVH1-GFP or rPVH1-yCD. GAPDH expression was used as internal control to generate a standard curve. The columns are the mean of three independent experiments; bars, s.e.m. (C,D) CEA tumor marker levels and animal survival data of treatment initiated 2 weeks and (E,F) 2 days post-tumor cell inoculation. For tumor growth and invasion, studies were performed by tumor maker measurement. Blood samples were drawn every 5 days, and serum CEA levels were measured by ELISA assays (C,E). For Kaplan-Meier survival curve determination, mice bearing AsPc1 peritoneal carcinomatosis were monitored for survival for 120 days (D,F). Asterisks indicate significant differences observed after PVH1-treated mice compared to non-treated mice (Mock) or rPVH1-yCD/5-FC compared to wtPV-H1; (*p<0.05, **p<0.01 and ***p<0.001).

Figure 6. PVH1-yCD/5-FC did not induce significant systemic toxicity in mice.

As indicated in Panel A, compared to the untreated group (mock), the body weight of mice receiving wtPVH1 or rPVH1-yCD/5-FC was not affected during the first 21 days period of treatment. Panel B) the concentrations (µg/ml) of 5-FC and 5-FU were determined using HPLC measurement 48 hours after prodrug i.p injections. In the control group (mock), the concentration of 5-FC in the plasma samples (P) was approximately 14 µg/ml. Its level in normal and tumor tissues (NT, TT) was comparable (6 to 6.25 µg/ml). In the PVH1-yCD/5-FC -treated mice, the levels of 5-FC were inferior (4 µg/ml in plasma samples and 2–2.65 µg/ml in normal and tumor tissue). As indicated in panel B- bottom, in the control group as well as in PVH1-yCD/5FC –treated group, the 5-FU was not detected neither in plasma samples nor in normal tissue. 5-FU (5 µg/ml) was only detected in tumor tissue extracts from the PVH1-yCD/5FC–treated mice. Panel C- For viral presence, DNA was extracted and quantitative real-time PCR was done to assess viral copy number per total genomic cellular DNA. The viral DNA was only detected at low level in normal tissue at day 2 after i.p injection of replicative wtPV-H1 or PVH1-yCD. However, in tumor nodule tissues, viral DNA was detected in all the samples (1,2,3,4) at day 2 and 10 post-infection. Columns are the mean of triplicate assays; bars, s.e.m.