Optimizing gene expression analysis in archival brain tissue

Analysis of gene expression in the brain is a valuable tool to study the function of the brain under normal and pathological conditions. Although there are many techniques used to measure gene expression the validity of any such experiment is directly related to the quality of the RNA in the samples. The most readily available source of human brain tissue is post-mortem and while frozen tissue is sometimes available, most archived tissue is fixed and paraffin-embedded. The use of fixed tissue for expression analysis introduces variables, which must be considered in the experimental design. In addition, factors associated with clinical variability of the patient and with tissue procurement can affect RNA transcript levels. In order to illustrate the effects of two common tissue fixatives, formalin and ethanol, on the quality of RNA for expression analysis we compare RNA extracted from these fixed tissues to the gold standard, flash-frozen tissue. We describe RNA extraction from fixed tissue and ways to assess the quality or intactness of the RNA using reverse transcription combined with polymerase chain reaction amplification. An advantage of using archived tissue is the ease with which single cells or subpopulations of cells can be obtained by laser microdissection. The successful isolation of RNA from microdissected cells is also presented. From our results and a review of the literature we conclude that RNA from fixed tissues is a viable source of RNA for expression analysis which should enable new experimental approaches and discoveries as long as attention is given to variables that can affect RNA at all levels of analysis.