The isolation of nucleic acids from fixed, paraffin-embedded tissues-which methods are useful when?

Abstract

Museums and pathology collections around the world represent an archive of genetic material to study populations and diseases. For preservation purposes, a large portion of these collections has been fixed in formalin-containing solutions, a treatment that results in cross-linking of biomolecules. Cross-linking not only complicates isolation of nucleic acid but also introduces polymerase "blocks" during PCR. A wide variety of methods exists for the recovery of DNA and RNA from archival tissues, and although a number of previous studies have qualitatively compared the relative merits of the different techniques, very few have undertaken wide scale quantitative comparisons. To help address this issue, we have undertaken a study that investigates the quality of nucleic acids recovered from a test panel of fixed specimens that have been manipulated following a number of the published protocols. These include methods of pre-treating the samples prior to extraction, extraction and nucleic acid purification methods themselves, and a post-extraction enzymatic repair technique. We find that although many of the published methods have distinct positive effects on some characteristics of the nucleic acids, the benefits often come at a cost. In addition, a number of the previously published techniques appear to have no effect at all. Our findings recommend that the extraction methodology adopted should be chosen carefully. Here we provide a quick reference table that can be used to determine appropriate protocols for particular aims.

Figure 1. The association of levels of PCR amplifiable human nuDNA and RNA with digestion time.

Note that in conventional quantitative real-time PCR assays, the measure of PCR amplifiable DNA or RNA (the Ct value) is inversely related to the starting concentration of template. Therefore, lower Ct values indicate higher original PCR amplifiable DNA or RNA yields.

Figure 2. DNA and RNA yields (ng/µg original tissue) in six different nucleic acid extracts from sample PM85L.

Each extract derives from a different digestion technique (a–f). For full details of the digestion protocols refer to main text. The data clearly demonstrates that total (but not necessarily amplifiable) DNA and RNA yields are greatly reduced in extracts performed using hot alkali techniques , .

Figure 3. The relative PCR-amplifiable nuDNA yields resulting from different extraction techniques (a–f) on 3 different formalin fixed samples.

Only sample PM85 was extracted with techniques (d–f). Techniques (a–c) are based on the hot-alkali methods of Shi et al. , , while techniques (d–f) are conventional proteinase k digestion techniques with extended digestion times. Note that in conventional quantitative real-time PCR assays, the measure of PCR amplifiable DNA (the Ct value) is inversely related to the starting concentration of DNA. Therefore, lower Ct values indicate higher original PCR amplifiable DNA yields. Although too small a sample to provide statistically supported findings, the data suggest that while there is little difference between the efficacies of the different hot alkali techniques themselves, they provide marginally higher levels of PCR amplifiable nuDNA.