Experimental gene therapy for an in vitro model of proliferative vitreoretinopathy

BACKGROUND: Proliferative vitreoretinopathy (PVR) is the leading cause of failure of retinal reattachment surgery. Since a key component of PVR is cell proliferation, we performed a study to examine whether the ribonucleotide-reductase-deficient herpes simplex virus type I (HSV-I) mutant hrR3 can be effective at destroying proliferating retinal pigment epithelial (RPE) cells and thus prevent epiretinal membrane formation and PVR, while sparing nondividing cells, such as neurons. METHODS: Primary cultures of rat RPE cells and rat cortical neurons were infected with 300 microL of hrR3 HSV-I to achieve a multiplicity of infection of 1.0. After 1 hour at 37 degrees C, 700 microL of growth medium was added to raise the total volume of medium to 1 mL. At 0, 12, 24 and 36 hours the cultures were observed, and the ratio of dead cells to live cells was determined. HSV infection and protein expression were confirmed by a beta-galactosidase histochemical assay or an antihuman HSV-I immunoassay, or both. RESULTS: At 24 hours more than 95% of the RPE cells and neurons stained positively for HSV infection, although beta-galactosidase was expressed predominantly in RPE cells. At 36 hours 72% (standard deviation 2.1%) of the RPE cells were dead. There was no noticeable cell death in the neuronal or mock-infected control cultures. INTERPRETATION: The results suggest that the hrR3 mutant strain of HSV-I can be used to infect and selectively kill actively proliferating rat RPE cells while sparing normal, nonreplicating cells. This model may be used to explore potential therapies for PVR in humans.