advertisement
Historically, the molecular analysis of retinal and optic nerve pathology in glaucoma was limited to specialized and expensive studies using non-human primates with experimental glaucoma, or to limited short-term experiments, in vitro, using primary cultures of ganglion cells isolated from rat retinas harvested in the first week of life. Over the last 10-15 years, three important new tools have emerged that have resulted in an explosion of new findings in the research community. Two of these have been the development of rat models of ocular hypertension and the use of mice, particularly the characterization of the disease phenotype exhibited by the DBA/2J inbred line. Cell culture studies benefited from a third new tool, that being a cell line derived from primary rat retinal ganglion cells immortalized by transformation with the ψ2 E1A virus (Krishnamoorthy et al. Mol Brain Res 2001; 86: 1-12). These cells exhibited remarkable molecular and phenotypic similarities to ganglion cells, including the expression of well-characterized ganglion cell marker genes and a dependence on the growth factors BDNF and NT-4 for survival. In addition, they could be treated with drugs like succinyl concanavalin A (sConA), staurosporine, or histone deacetylase inhibitors, which would cause them to differentiate into neuron-like cells, and exhibit a sensitivity to glutamate. The development of this cell line enabled several research programs to use them as a starting point to evaluate and study molecular changes occurring during stress conditions in ganglion cells. Since their initial description, over 80 different publications studying them have appeared in the peer-reviewed literature. Over this time, however, cells from this line appeared to change, losing, for example, their sensitivity to glutamate after differentiation. To address this apparent change form the original source, Van Bergen et al. (1034) set out to re-characterize RGC-5 cells they had received from the founding laboratory on two different occasions spaced 5 months apart. These investigators sequenced two regions of genomic and mitochondrial DNA, respectively, and found that the cells originated from mouse, and not rat. This observation was also independently verified by a different laboratory using a third sample of cells.
RGC-5 cells currently in circulation are mischaracterizedAdditionally, the cells examined in the Van Bergen report failed to express both neuronal and ganglion cell marker genes, either before or after differentiation with sConA. What does this mean for the future of studies using RGC-5 cells? First, it should be noted that there were some limitations of the study presented here. Cell marker expression studies were restricted to immunoreactivity assays comparing rat retina and optic nerve samples against extracts of RGC-5 cells. Since the antibodies used were principally monoclonals generated in mice, it is not surprising that they exhibited limited reactivity to mouse derived RGC-5 samples. Thus, speculation that these cells do not express neuronal or ganglion cell markers is not an easy conclusion to draw from these results. Clearly, in this case, follow-up studies characterizing gene expression profiles at the transcript level are necessary to fully characterize these cells. Additionally, the authors limited the scope of their study by only differentiating cells with sConA. Since other compounds appear to be more potent and reliable at differentiating RGC-5 cells to a neuronal phenotype, it would be appropriate to conduct more comprehensive studies before discarding them as non-neuronal, or molecularly dissimilar to ganglion cells.
Stocks of cells need to be extensively re-evaluated to re-legitimize them as an in-vitro model for glaucoma related research
Ultimately, however, the DNA analysis work done by this group shows beyond question that the cells currently in circulation are mischaracterized. It appears that the stocks of cells need to be extensively reevaluated to re-legitimize them as an in vitro-model for glaucomarelated research, although ultimately, they may still be an excellent tool for broader neuroscience applications.