Production of bioactive soluble interleukin-15 in complex with interleukin-15 receptor alpha from a conditionally-replicating oncolytic HSV-1.

Oncolytic type-1 herpes simplex viruses (oHSVs) lacking the γ134.5 neurovirulence gene are being evaluated for treatment of a variety of malignancies. oHSVs replicate within and directly kill permissive cancer cells. To augment their anti-tumor activity, oHSVs have been engineered to express immunostimulatory molecules, including cytokines, to elicit tumor-specific immune responses. Interleukin-15 (IL-15) holds potential as an immunotherapeutic cytokine because it has been demonstrated to promote both natural killer (NK) cell-mediated and CD8(+) T cell-mediated cytotoxicity against cancer cells. The purpose of these studies was to engineer an oHSV producing bioactive IL-15. Two oHSVs were constructed encoding murine (m)IL-15 alone (J100) or with the mIL-15 receptor α (mIL-15Rα, J100D) to determine whether co-expression of these proteins is required for production of bioactive mIL-15 from oHSV. The following were demonstrated: i) both oHSVs retain replication competence and cytotoxicity in permissive tumor cell lines. ii) Enhanced production of mIL-15 was detected in cell lysates of neuro-2a cells following J100D infection as compared to J100 infection, suggesting that mIL-15Rα improved mIL-15 production. iii) Soluble mIL-15 in complex with mIL-15Rα was detected in supernates from J100D-infected, but not J100-infected, neuro-2a, GL261, and CT-2A cells. These cell lines vary in permissiveness to oHSV replication and cytotoxicity, demonstrating soluble mIL-15/IL-15Rα complex production from J100D was independent of direct oHSV effects. iv) The soluble mIL-15/IL-15Rα complex produced by J100D was bioactive, stimulating NK cells to proliferate and reduce the viability of syngeneic GL261 and CT-2A cells. v) J100 and J100D were aneurovirulent inasmuch as no neuropathologic effects were documented following direct inoculation into brains of CBA/J mice at up to 1x10(7) plaque forming units. The production of mIL-15/mIL-15Rα from multiple tumor lines, as well as the lack of neurovirulence, renders J100D suitable for investigating the combined effects of oHSV and mIL-15/IL-15Rα in various cancer models.