Extraction of Nucleic Acids from FFPE Samples

The collection of tissue samples for subsequent microscopic examination usually entails the use of formaldehyde as a fixative. In tissue, the initial reaction of the formaldehyde is thought to occur primarily with the ε-amine-N group of lysine residues of proteins, resulting in the formation of aminomethanol moieties. These are thought to react, in turn, with less reactive amide- and guanidyl-N groups [1] as well as the modified ring structures of Tyr, Trp, and His residues [2]. Although the reaction of nucleic acids (NA) with formaldehyde is known to involve the exocyclic amino groups of A, C, and G residues [3], the complexity of the reaction in the milieu of biological systems is unknown and probably involves many additive reactions that form a tight meshwork of crosslinks between proteins, nucleic acids, and other biomolecules at unknown sites. Although this is perfect for creating a stable specimen that maintains the tissue’s ultrastructure and withstands processing and sectioning, it creates a problem for retrieving any of these “fixed” molecules. Although some reversal of the crosslinks by chemical methods is possible [4], the final products of the fixation reactions normally possess a stability that requires conditions severe enough to degrade RNA.

After extensive research, we have found the best methods to free and purify RNA involve performing enzymatic proteolysis under optimized conditions, followed by solid-phase extraction of the nucleic acid on glass-fiber or other solid supports. DNA can be isolated in a similar fashion, but as it is inherently more stable, conditions are often more severe (longer incubation times, higher temperatures, etc.)—an unexpected by-product of which is the chemical removal of more crosslinks and adducts.

Despite the advantages of these modified extraction methods, neither DNA nor RNA escapes undamaged from the treatment and fixation. However, both can still be used successfully for some downstream applications (e.g., TaqMan Assays). The formaldehyde treatment and embedding step inevitably cause some fragmentation, which can be seen when examining the NA by electrophoretic methods. In addition, other lesions can be present that aren’t readily apparent but do affect the ability of the NA to serve as a template. This becomes apparent when the researcher tries to amplify different fragment sizes from the same gene. Short amplicons of 60–70 bp will give a lower Cq than amplicons in the 200–300 bp range, although this observation does not directly correlate to the average size of the fragmented population. Surprisingly, the chemical damage appears to increase with the age of the FFPE samples, even though paraffin embedding should displace all water and air from the tissue matrix.

Read more: FFPE RNA isolation

We offer two popular products for the preparation of both RNA and DNA from FFPE tissue: the RecoverAll Total Nucleic Acid Isolation Kit for FFPE and the MagMAX FFPE Total Nucleic Acid Isolation Kit, both offered under the Ambion brand. In addition, the DNA purification module from the latter can be purchased separately as the Invitrogen MagMAX FFPE DNA Isolation Kit. Both procedures perform proteolytic digestions under defined temperatures, followed by purification with solid-phase extraction. We also offer a kit for RNA purification from FFPE. 

The two total nucleic acid procedures differ in two main respects: the method for removing paraffin wax and the solid phase chosen for final extraction of the nucleic acid. The RecoverAll kit achieves replacement of the wax by water through a series of soaks in xylene (or limonene) and various dilutions of ethanol in water. Paraffin wax removal using the MagMAX kit occurs by the direct incubation of a slice of the embedded tissue in the proteolytic solution (after the addition of a novel ingredient that helps to penetrate and lift away the wax). Therefore, the temperature regimens for the two kits are different (Table 1). The second major difference is that the RecoverAll kit uses a glass-fiber filter in a spin-column format for its solid phase recovery step, while the MagMAX FFPE kit uses MagMAX magnetic beads. This allows for higher-throughput processing either by manual extraction or transfer of the protocol to an automated platform, such as a MagMAX Express Magnetic Particle Processor.

The performance of the two kits is comparable. Some of the observed trends are due more to variability between different slices used for the replicates (note the error bars) than any true difference arising from the methods. Overall, both kits provide a fast and easy solution for nucleic acid extraction from FFPE samples in a range of formats that yield material suitable for downstream applications.

Table 1. Overview of the RecoverAll Total Nucleic Acid Isolation Kit and the MagMAX FFPE Total Nucleic Acid Isolation Kit procedures

Figure 1. Comparison of RNA results using different FFPE kits.

Panel A: RNA yield

After extraction, RNA concentration was measured using the Nanodrop instrument. Fair amount of RNA was calculated using the concentration and elution volume.

Panel B. mRNA qRT-PCR results for ACTB target

Equal mass amounts of total RNA (based on Nanodrop measurements) were run through qRT-PCR reactions. cDNA was created using Applied Biosystems' High Capacity cDNA Reverse Transcription Kit. Once complete, 2 µL of each cDNA reaction was run through triplicate qPCR reactions using Applied Biosystems Assays and Universal PCR master mix 11 and standard reaction conditions. Raw C_T results were averaged for all replicates and graphed.

Panel C. miRNA qRT-PCR results for Let7a target

Equal mass amounts of total RNA (based on Nanodrop measurements) were run through qRT-PCR reactions. cDNA was created using Applied Biosystems' TaqMan MicroRNA Reverse Transcription Kit and gene specific primers. Once complete, 2 µL of each cDNA reaction was run through triplicate qPCR reactions using Applied Biosystems Assays and Universal PCR master mix 11 and standard reaction conditions. Raw C_T results were averaged for all replicates and graphed.

Figure 2. Yield and qRT-PCR results from nucleic acid purified with FFPE kits. (A) DNA concentration was measured using the NanoDrop instrument. Yield was calculated based on the measured concentration and the elution volume. (B) qPCR was carried out for TAF9 (rat) and RNase P (human) targets. Equal mass amounts of TAF9 (rat) and RNAse P (human) DNA (purified using specified FFPE kits) were assayed in triplicate in qRT-PCR reactions using the Applied Biosystems assays and Universal PCR Master Mix 11 and standard reaction conditions. Raw C_T results were averaged for all replicates and graphed.

1. Fraenkel-Conrat H, Olcott HS (1948) The reaction of formaldehyde with proteins; cross-linking between amino and primary amide or guanidyl groups. J Am Chem Soc 70:2673–2684. 
2. Fraenkel-Conrat H, Olcott HS (1948) Reaction of formaldehyde with proteins; cross-linking of amino groups with phenol, imidazole, or indole groups. J Biol Chem 174:827–843. 
3. Chaw YF, Crane LE, Lange P et al. (1980) Isolation and identification of cross-links from formaldehyde-treated nucleic acids. Biochemistry 19:5525–5531. 
4. Masuda N, Ohnishi T, Kawamoto S et al. (1999) Analysis of chemical modification of RNA from formalin-fixed samples and optimization of molecular biology applications for such samples. Nucleic Acids Res 27:4436–4443.