Transcriptional profiling of degraded RNA in cryopreserved and fixed tissue samples obtained at autopsy

Abstract

Background: Traditional multiplexed gene expression methods require well preserved, intact RNA. Such specimens are difficult to acquire in clinical practice where formalin fixation is the standard procedure for processing tissue. Even when special handling methods are used to obtain frozen tissue, there may be RNA degradation; for example autopsy samples where degradation occurs both pre-mortem and during the interval between death and cryopreservation. Although specimens with partially degraded RNA can be analyzed by qRT-PCR, these analyses can only be done individually or at low levels of multiplexing and are laborious and expensive to run for large numbers of RNA targets.

Methods: We evaluated the ability of the cDNA-mediated Annealing, Selection, extension, and Ligation (DASL) assay to provide highly multiplexed analyses of cryopreserved and formalin fixed, paraffin embedded (FFPE) tissues obtained at autopsy. Each assay provides data on 1536 targets, and can be performed on specimens with RNA fragments as small as 60 bp.

Results: The DASL performed accurately and consistently with cryopreserved RNA obtained at autopsy as well as with RNA extracted from formalin-fixed paraffin embedded tissue that had a cryopreserved mirror image specimen with high quality RNA. In FFPE tissue where the cryopreserved mirror image specimen was of low quality the assay performed reproducibly on some but not all specimens.

Conclusion: The DASL assay provides reproducible results from cryopreserved specimens and many FFPE specimens obtained at autopsy. Gene expression analyses of these specimens may be especially valuable for the study of non-cancer endpoints, where surgical specimens are rarely available.

Figure 1

DASL RNA Profiling Technique (From [5]).

Figure 2

Comparison of Intact and Degraded Cryopreserved RNAs. A. Electropherogram constructed from Bioanalyzer microchip electrophoresis of RNA extracted from cryopreserved tissue graded "Passed" by Asterand quality assessment method. Prominent bands can be seen as sharp peaks at 18S and 28S indicating highly intact RNA. C. Sample that did not pass Asterand quality measures. Electropherogram shows an 18S peak larger than the 28S peak and an increased amount of noise throughout the electropherogram. B. & D. The panels on the right show scattergrams of the signal intensities of individual genes for replicate analyses of each sample, and the associated Pearson correlation coefficient squared (R²). Both samples provided reproducible results.

Figure 3

Comparison of Intact and Degraded FFPE Extracted RNAs. A. Electropherogram of RNA graded "Passed" by Asterand quality measures from cryopreserved tissue of a mirror image preparation of the FFPE sample being studied. One can see clear peaks for both 18S and 28S ribosomal RNA with low amount of noise between the peaks. C. Cryopreserved mirror sample graded "Failed" by Asterand quality measure. A lack of distinguishable peaks and also a large amount of fragments in the low molecular weight range indicates a large amount of degradation. B. & D. The right shows the reproducibility of the assay based on scattergrams of signal intensities from extracted FFPE RNA on replicate DASL assays and the obtained Pearson R² correlation coefficient squared (R²). Both scattergrams demonstrate highly reproducible results.