To express Hsp72 exogenously in RGC-5 cells that are not easily transfected by conventional methods, the cells were transduced with a recombinant adenovirus expressing human Hsp72. Western blot analysis showed that protein bands of 72 kDa corresponding to Hsp72 were clearly detected in lysates of cells transduced with a recombinant adenovirus expressing human Hsp72. The intensities of the bands were found to be in proportion to the degree of viral transduction. In untransduced control cells, a faint band equivalent to endogenous Hsp72 expression was barely detectable (Fig. 1A). Immunocytochemistry yielded consistent results showing that cytoplasmic staining for characteristic Hsp72 expression could be detected exclusively in transduced cells but not in untransduced cells, confirming the adenovirus-mediated Hsp72 gene expression (Fig. 1B).

To examine the effect of Hsp72 on the death of RGC-5 cells caused by apoptosis, the cells were treated with etoposide immediately after viral transduction. Compared with control cells (Fig. 2A), early cell death was seen in a small proportion of cells in the culture without serum (Fig. 2B), and the degree of cell death was accelerated by the treatment of etoposide (Fig. 2C). However, the etoposide-induced cell death appeared to be blocked by Hsp72 expression, as observed by the enhanced viability of cells that were transduced immediately before etoposide treatment (Fig. 2D). To confirm the effect of Hsp72 on the survival of RGC-5 cells, a cell proliferation assay was performed on cells treated with etoposide following viral transduction (Fig. 3). The rate of viable cells was reduced by 44.8 ± 6.7% and 80.8 ± 9.2% relative to control cells at 5 and 20 µM etoposide, respectively. Viral transduction itself might be toxic to cells because cell viability was slightly decreased when Hsp72 was expressed in control cells. The Hsp72 expression, however, seemed to be protective to cells undergoing apoptosis in that the rate of cell death in the cultures exposed to etoposide was progressively reduced as the degree of viral transduction increased. At the highest level of Hsp72 expression, the number of viable cells recovered was increased by 24.0 ± 3.1% and 94.1 ± 10.3% over that in the control cultures at 5 and 20 µM etoposide, respectively.

In this study, we addressed whether Hsp72 can protect RGC-5 cells from apoptotic cell death. RGC-5 cells are the only immortalized RGC cell line established to date and are used widely in glaucoma research, although it is uncertain whether the cell line is of mouse or rat origin [10]. To investigate directly the effect of Hsp72 on the death of RGC-5 cells, it is necessary to express the protein exogenously in the cells. Like other cell lines, however, RGC-5 cells show very low transfection efficiency. To overcome this obstacle, we constructed a recombinant adenovirus in which a human Hsp72 cDNA is under the control of the human cytomegalovirus promoter for its constitutive expression and demonstrated that the viral vector was successful in both transducing RGC-5 cells and expressing its transgene inside the cells. We showed that the degree of cell death by etoposide was inversely proportional to the degree of viral transduction in the cells. Etoposide, an inhibitor of topoisomerase II, is a well-known apoptosis-inducing reagent [11]. Therefore, our results present direct evidence for the cytoprotective ability of Hsp72 against apoptotic cell death.

The protective role of heat shock was first described in light-damaged rat retina and Hsp72 protein was identified as the entity primarily responsible for the response [12,13]. A growing body of evidence has accumulated that induction of Hsp72 is essential for neuronal tolerance to a wide variety of stressful conditions including ischemic or excitotoxic insults, seizure, oxidative stress, and apoptotic stimuli [5,8,9]. The present study demonstrates that Hsp72 expression itself is sufficient to protect RGC-5 cells from apoptotic cell death. Apoptosis, or programmed cell death, of RGCs is one of the earlier signs of the pathogenesis of glaucoma. Because inhibition of RGC death or a halt of its progression is the key to glaucoma treatments, the ability of Hsp72 to intervene in the process of apoptosis is attractive as a novel therapeutic intervention in glaucoma. Gene therapy, which is a practical alternative to current therapies that are limited to reduction of elevated intraocular pressure, is an approach by which Hsp72 could be useful in a clinical context. Recent studies reported that Hsp72 induction by zinc or geranylgeranylacetone administration increased RGC survival in a rat glaucoma model, suggesting that Hsp72 could be therapeutic for glaucoma [14,15].

The eye may perhaps be the first organ for which gene therapy is used routinely in a clinical setting because it has advantages such as well-defined anatomy with small size, translucent media allowing excellent visual localization, and easy accessibility by which therapeutic agents can be delivered to the vicinity of cells with a particular defect [16]. The replication-deficient adenovirus generated in this study is a sophisticated gene delivery tool and such viral-based vectors are expected to have enormous potential as a mammalian expression system for gene therapy.

Recently, it has been reported that Hsp72 can be identified in the extracellular compartments and the extracellular Hsp72 acts as a cytokine that induces inflammation and modulates the innate immune response [17-20]. The dual role of Hsp72 as a molecular chaperone and cytokine suggests that fine tuning its expression may be a potential route for a novel therapeutic approach to glaucoma and other neurodegenerative diseases.