A NOTCH-sensitive uPAR-regulated oncolytic adenovirus effectively suppresses pancreatic tumor growth and triggers synergistic anticancer effects with gemcitabine and nab-paclitaxel

Abstract

Notch signaling pathway is an embryonic program that becomes reactivated in pancreatic cancer and contributes to cancer stem cell (CSC) maintenance. We explored the concept of oncolytic adenoviral activity in response to Notch activation signaling, in the context of a chimeric promoter with uPAR regulatory sequences, as a strategy to drive its activity in neoplastic and CSC. We explored the advantages of a chemo-virotherapy approach based on synergistic combinations. Regulatory sequences recognized by the transcriptional factor CSL upstream a minimal uPAR promoter were engineered in adenoviral vectors and in the oncolytic adenovirus AdNuPARmE1A. Viral response to Notch signaling, and viral potency in cell lines and pancreatic cancer stem cells (PCSC) was tested. Preclinical toxicity and antitumor efficacy in xenografts and Patient-derived xenografts (PDX) mouse models was evaluated, as unimodal or in combination with gemcitabine+nab-paclitaxel. Mechanistic studies were conducted to explore the synergism of combined therapies.We demonstrate that CSL-binding site optimized-engineered sequences respond to Notch activation in AdNuPARmLuc and AdNuPARmE1A. AdNuPARmE1A showed strong lytic effects in pancreatic cancer cell lines and PCSC. AdNuPARmE1A displayed attenuated activity in normal tissues, but robust antitumor effects in xenograft and PDX models, leading to a reduced capacity of treated tumors to form tumorspheres. Chemo-virotherapy treatment enlarged therapeutic response in both tumor models. Synergistic effects of the combination resulted from viral sensitization of apoptotic cell death triggered by chemotherapy.In summary we present a novel effective oncolytic adenovirus, AdNuPARmE1A that reduces PCSC and presents synergistic effects with gemcitabine and nab-paclitaxel, supporting further clinical development.

Keywords: cancer stem cells; gemcitabine; nab-paclitaxel; oncolytic adenovirus; pancreatic cancer.

Figure 1. Notch-dependent transcriptional activity from the chimeric promoter controlled by the uPAR regulatory sequences and Notch responsive elements

(A) Reporter plasmids that express the luciferase gene under the control of uPAR, 1xSPSuPAR or 3xSPSuPARm promoters were tested for luciferase activity in NIH/3T3 cells under the presence, or not, of NICD. Hes1LUC plasmid was used as positive control of Notch response. (B) Reporter adenovirus was designed using 3xSPSuPARm promoter (upper panel). MIAPaCa-2 cells were transduced with 10 MOI of Ad3xSPSuPARmLuc and then transfected with increasing amounts of pNICD encoding plasmid. Luciferase activity was analyzed 48 h after transfection. (C) Luciferase activity of MIAPaCa-2 cells transduced with AduPARLUC or AdNuPARmLUC (10 MOI) and transfected with 1 μg pNICD. (D) Luciferase activity in BxPC-3 cells transduced with AdNuPARmLUC (5 MOI) and treated with 30 μM of DAPT. Statistical differences were analyzed by Student's t-test (E) AduPARLUC and AdNuPARmLUC activity in several pancreatic cancer models: BxPC3 and PANC-1 cell lines (left panel), BxPC3 and PANC-1 spheres (middle panel) and CP15 and CP13 tumorspheres derived from PDX (right panel). All cells were transduced at 5 MOI and luciferase expression was analyzed at 48 h after transduction and normalized by the % of infection (% of GFP expressing-cells). Results are expressed as a mean +/− SEM of at least three independent experiments (*p < 0.05; **p < 0.01; ***p < 0.001).

Figure 2. AdNuPARmE1A oncolytic adenovirus is regulated by the Notch pathway and displays a high oncolytic activity in pancreatic cancer models in vitro

(A) Schematic representation of the oncolytic adenovirus AdNuPARmE1A. Expression of the E1A adenoviral gene is controlled by the 3xSPSuPARm promoter. SINEB2 insulator sequence was cloned upstream the promoter sequence. (B) Western blot of E1A gene expression in BxPC-3 cells infected with AduPARE1A or AdNuPARmE1A (50 MOIs) and treated, or not, with 30 μM of DAPT for 48 h. Bar graph shows quantification of E1A expression normalized to GAPDH levels (right panel). Results are expressed as a mean +/− SEM of n = 7 independent experiments (*p < 0.05; **p < 0.01). (C) In vitro oncolytic activity of AdNuPAREmE1A compared to Adwt in PANC-1, BxPC-3, MIAPaca-2, CP15-Luc HPDE and HK-2 cell lines. Cells were seeded in triplicate and treated with a dose range of adenoviruses (vp/cell). Cell viability was measured 72 h post-infection by MTT assay and normalized to mock treated cultures. IC50 mean values +/− SEM of at least four independent experiments are represented at the table (*p < 0.05; **p < 0.01). (D) In vitro oncolytic activity of AdNuPAREmE1A compared to Adwt in CP15 tumorspheres. Cell viability was measured 72 h post-infection by WST-1 assay and normalized to mock treated cultures (n = 3 independent experiments). (E) Quantification of viral production in CP15 tumorspheres infected with AdNuPARmE1A or Adwt upon several passages by qPCR. Results are expressed as the mean of relative viral genomes +/− SEM of three independent experiments.

Figure 3. Toxicity profile of AdNuPARmE1A after systemic administration in immunocompetent mice

(A) Percentage of body weight variation in immunocompetent mice after intravenous administration of Adwt or AdNuPARmE1A (5·10¹⁰ vp/mouse), or saline in the control group. Bar graph shows the percentage of body weight variation at day 3 post-administration (right panel). (B) Assesment of hepatotoxicity by the determination of AST, ALT and total bilirubin in the serum of treated mice. (C) Viral genomes production in livers of treated mice. (D) qPCR of E1A mRNA expression in livers of treated mice, relative to GDX expression. Results are expressed as the mean +/− SEM of n ≥ 8 animals/group, or n = 5 in the saline control group. (*p < 0.05; **p < 0.01).

Figure 4. In vivo antitumoral activity of AdNuPARmE1A as a single agent or in combination with gemcitabine and nab-paclitaxel in MIAPaCa-2 tumors

(A) Mice bearing subcutaneous MIAPaCa-2 tumors were treated with gemcitabine (100 mg/kg) and nab-paclitaxel (30mg/kg) twice a week (T1), with a single dose of AdNuPARmE1A (5·10¹⁰vp) (T2), with the combined treatment (T3) or with saline in the control group, (upper panel). Follow-up of tumor growth was monitored every other day for 25 days (lower panel). (B) Percentage of tumor volume (mm³) of saline and treated tumors at the end of the experiment. (C) qPCR of E1A mRNA expression in MIAPaCa-2 treated tumors normalized to β-actin levels. Results are expressed as the mean +/− SEM of at least n = 8 tumors/group (*p < 0.05; **p < 0.01; ***p < 0.001).

Figure 5. In vivo antitumoral activity of AdNuPARmE1A as a single agent or in combination with gemcitabine and nab-paclitaxel in CP15 PDX tumors

(A) Mice bearing subcutaneous tumor fragments of CP15 PDX tumors were treated with gemcitabine plus nab-paclitaxel (T1), with a single dose of AdNuPARmE1A (5·10¹⁰vp) (T2), with the combined treatment (T3) or with saline in the control group, arrow indicates the day when the chemotherapy was stopped (upper panel). Follow-up of CP15 PDX tumor volume (mm³) represented as percentage of growth (lower panel), (n ≥ 12 tumors/treatment group). (B) Image of three representative CP15 tumors for each treatment group. (C) Percentage of tumor volume (mm³) of CP15 treated tumors at the end of the experiment. (D) qPCR of E1A mRNA expression in CP15 treated tumors normalized to β-actin levels (E) Tumorspheres grown as anchorage-independent colonies from CP15 treated tumors, representative images (left panel), quantification of the number of tumorspheres (right panel). Results are expressed as the mean +/− SEM of n = 4 tumors/treatment group at 22 days after plating (*p < 0.05; **p < 0.01; ***p < 0.001).

Figure 6. Cytotoxic and synergistic effects of the combination of AdNuPARmE1A with gemcitabine plus nab-paclitaxel in cellular models

(A) Dose –response curves of nab-paclitaxel, AdNuPARmE1A or combined treatment in BxPC-3 cell line. Cells were seeded in triplicate and treated with a dose range of nab-paclitaxel (nM) and/or AdNuPARmE1A (vp/cell). Cell viability was measured 72 h later by MTT assay and is expressed as cell viability normalized to mock treated cultures. IC50 values for monotherapy or combination therapy for each treatment are represented in bar graphs. (B) Combination index values (CI) for the interaction of nab-paclitaxel and AdNuPARmE1A are calculated as a function of inhibitory fractions. Results are expressed as a mean +/− SEM of five independent experiments. (C) Cell viability in BxPC-3 cells treated with gemcitabine plus nab-paclitaxel, AdNuPARmE1A or combined treatment. Cells were seeded in triplicate and treated with several doses of nab-paclitaxel (nM) plus gemcitabine (ng/ml) and/or AdNuPARmE1A (vp/cell). Cell viabilitiy was measured 72 h later by MTT assay. Cell viability percentages relative to mock treated cells are represented in bar graphs as the mean +/− SEM of four independent experiments. Statistical differences were analyzed by Student's t-test. (D) Loss of mitochondrial transmembrane potential (ΨΔ_m) analysis in BxPC-3 cultured cells treated with AdNuPARmE1A (4000 vp/cell), with 50 ng/ml of gemcitabine plus 10 nM of nab-paclitaxel or with the combined treatment for 48 h, and staining with 15 nm 3,3′-diexyloxacarbocyanine iodide (DiOC₆(3)). (E) Cell viability in BxPC-3 and CP15-Luc spheres treated with gemcitabine plus nab-paclitaxel, AdNuPARmE1A or combined treatment. Single cell suspensions of BXPC-3 and CP15-Luc cells were seeded at a density 2.0.10⁴ cells/mL with appropriate medium. Forty-eight hours later, BxPC-3 and CP15-Luc spheres were treated with nP 25 nM or 50 nM, gemcitabine 5 ng/mL or 2.5 ng/ml, AdNuPARmE1A 5000 vp/cell or 1000 vp/cell respectively alone or in combination and 72 h later, the percentage of viable spheres (1st generation) was determined by MTT assay. Then, 1st generation spheres were dissociated to single cell suspensions, re-plated at a density of 2.0.10⁴ cells/mL and 72 h later, CP15-Luc 2nd generation spheres images (10× magnification) were captured and the percentage of viable cells was determined by MTT assay. Statistical differences were analyzed by Student's t-test. (F) Luciferase activity in BxPC-3 and MIAPaCa-2 cell lines transduced with reporter adenovirus AdNuPARmLuc (10 MOIs) and treated with gemcitabine and/or nab-paclitaxel for 24 h. Results are expressed as the mean +/− SEM of at least four independent experiments. (*p < 0.05; **p < 0.01; ***p < 0.001).