Utilizing tumor hypoxia to enhance oncolytic viral therapy in colorectal metastases

Abstract

Objective: To determine the effects of hypoxia-induced ribonucleotide reductase (RR) production on herpes oncolytic viral therapy.

Summary background data: Hypoxia is a common tumor condition correlated with therapeutic resistance and metastases. Attenuated viruses offer a unique cancer treatment by specifically infecting and lysing tumor cells. G207 is an oncolytic herpes virus deficient in RR, a rate-limiting enzyme for viral replication.

Methods: A multimerized hypoxia-responsive enhancer was constructed (10xHRE) and functionally tested by luciferase assay. 10xHRE was cloned upstream of UL39, the gene encoding the large subunit of RR (10xHRE-UL39). CT26 murine colorectal cancer cells were transfected with 10xHRE-UL39, incubated in hypoxia (1% O2) or normoxia (21% O2), and infected with G207 for cytotoxicity assays. CT26 liver metastases, with or without 10xHRE-UL39, were created in syngeneic Balb/C mice (n = 40). Livers were treated with G207 or saline. Tumors were assessed and stained immunohistochemically for G207.

Results: 10xHRE increased luciferase expression 33-fold in hypoxia versus controls (P < 0.001). In normoxia, 10xHRE-UL39 transfection did not improve G207 cytotoxicity. In hypoxia, G207 cytotoxicity increased 87% with 10xHRE-UL39 transfection versus nontransfected cells (P < 0.001). CT26 were resistant to G207 alone. Combining 10xHRE-UL39 with G207 resulted in a 66% decrease in tumor weights (P < 0.0001) and a 65% reduction in tumor nodules (P < 0.0001) versus G207 monotherapy. 10xHRE-UL39-transfected tumors demonstrated greater viral staining.

Conclusions: Hypoxia-driven RR production significantly enhances viral cytotoxicity in vitro and reduces tumor burden in vivo. G207 combined with RR under hypoxic control is a promising treatment for colorectal cancer, which would otherwise be resistant to oncolytic herpes virus alone.

FIGURE 1. Constructing a vector with an intact UL39 gene under hypoxic control. A, The pGL3 promoter vector (Promega, Madison, WI) with its native luciferase reporter gene. B, Cloning the hypoxia responsive enhancer (10xHRE) into the XhoI restriction site of pGL3. C, The luciferase gene was removed with HindIII and XbaI (New England Biolabs, Beverly, MA) and replaced with the UL39 gene to form 10xHRE-UL39.

FIGURE 2. Expression of hypoxia-inducible factor 1α (HIF-1α) by CT26 murine colorectal carcinoma cells in normoxia and hypoxia. CT26 cells were incubated in either 21% O₂ or 1% O₂ for 24 hours. HIF-1α enzyme-linked immunosorbent assay was performed (Active Motif, Carlsbad, CA) on 20 μg of nuclear protein from each sample. There was a 6.6-fold increase in HIF-1α expression in 1% O₂ compared with 21% O₂ (P < 0.01).

FIGURE 3. 10xHRE enhancer function assessed by luciferase activity. CT26 murine colorectal cancer cells were transfected with 1.6 μg of our 10xHRE-pGL3 construct or the native pGL3 reporter vector (Promega, Madison, WI). Cells were subsequently incubated in either 21% O₂ or 1% O₂ for 18 hours, and luciferase assay was performed. Data are expressed in terms of fold increase in luciferase activity compared with pGL3 plasmid transfection alone. 10xHRE-pGL3 transfection resulted in a 33-fold increase in luciferase activity in 1% O₂ (P < 0.001). Expression in 21% O₂ increased 1.6-fold (P < 0.001).

FIGURE 4. G207 cytotoxicity in CT26 murine colorectal cancer cells. A, CT26 cells demonstrated resistance to G207 in normoxia (21% O₂) at a multiplicity of infection (MOI: viral plaque-forming unit to tumor cell ratio) of 1. Our 10xHRE-UL39 construct failed to improve cell kill in normoxia. B, CT26 cells were resistant to G207 at an MOI of 1 in hypoxia (1% O₂). Transfecting our 10xHRE-UL39 plasmid significantly increased G207 cytotoxicity by day 5, with nearly complete cell death by day 10 (P < 0.001).

FIGURE 5. Treatment of murine liver metastases with G207 and 10xHRE-UL39 (n = 10 per group). Balb/C mice underwent splenic injection with 6 × 10⁴ CT26 murine colorectal cancer cells transfected with our 10xHRE-UL39 construct or no DNA. Mice were reinjected 24 hours later with either 5 × 10⁴ plaque-forming units of G207 or phosphate-buffered saline (PBS). Liver tumor burdens were assessed 14 days later. A, There was a 66% decrease in tumor weight (P < 0.0001) with the combination of 10xHRE-UL39 and G207. There was no significant difference among tumor weights when given monotherapy with PBS, 10xHRE-UL39, or G207. B, There was a 65% reduction in tumor nodules (P < 0.0001) when both 10xHRE-UL39 and G207 were given. There was no significant difference among nodule counts when mice received monotherapy with PBS, 10xHRE-UL39, or G207.

FIGURE 6. Detection of herpes simplex virus (HSV) by immunohistochemistry of Balb/C livers with metastatic CT26 murine colorectal cancer. Fixed tissues were sectioned, stained with a rabbit anti-HSV-1 polyclonal antibody, and examined under light microscopy. The presence of brown represents positive staining for HSV envelope proteins. There was appreciably stronger staining for HSV in 10xHRE-UL39-transfected tumors treated with G207 versus mock-transfected tumors receiving G207. Viral staining was confined to the tumorigenic regions of livers. A, Untreated liver metastases (original magnification ×10). B, G207-treated, mock-transfected liver metastases (original magnification ×10). C, G207-treated, 10xHRE-UL39-transfected liver metastases (original magnification ×10). D, G207-treated, 10xHRE-UL39-transfected liver metastases (original magnification ×60).