Like most tissues, the eye as a structure contains a great diversity of cell types. These range from the cells of the lens and the iris, to the melanin-containing retinal pigment epithelium, the cells of the choroid and retinal vasculature, and of course the numerous photoreceptors, neurons and supportive Muller cells of the retina. Such cellular heterogeneity will restrict the detection of changes in gene expression, as if a transcript is expressed in a rare cell type, even a massive change in expression may be undetectable as it reflects only a small fraction of the RNA population measured. Alternatively, changes in gene expression in one cell type may be masked by compensatory changes in other cell types, for example rhythms of gene expression in anti-phase. Even when changes in expression can be detected in heterogeneous tissues, determining the source of the change may be impossible. Overall changes could be due to a change in all cells within the tissue, a change in only a fraction of the cells, a change in the cellular makeup of the tissue, e.g., neuronal loss, reactive gliosis or invasion of inflammatory cell types (Dougherty and Geschwind, 2005).
The counterpoint to these problems is that dissection or dissociation of specific compartments or cell types may affect the dynamic expression profile within the tissue. Reduction in starting tissue volume will also result in a decreased RNA yield and may even necessitate pooling individual samples, resulting in a decreased sample size or, in the worst case scenario, a single pooled sample. Moreover, the dissociation of cell types within heterogeneous tissues may lead to cell damage and release of endogenous RNases that will compromise RNA quality. Finally, the tissue yield obtained even from similar samples may be more variable following dissection or dissociation. This is certainly the case with retinal dissection, as the whole eye represents a discrete organ, and even given very careful dissection, retinal yields are considerably more variable.
As such, the use of heterogeneous tissues presents a trade-off between the problems of studying mixed cell populations and the problems of introduced variance and potential RNA loss associated with producing cell-specific populations.
Was this article helpful?