RNA extraction from ten year old formalin-fixed paraffin-embedded breast cancer samples: a comparison of column purification and magnetic bead-based technologies

Abstract

Background: The development of protocols for RNA extraction from paraffin-embedded samples facilitates gene expression studies on archival samples with known clinical outcome. Older samples are particularly valuable because they are associated with longer clinical follow up. RNA extracted from formalin-fixed paraffin-embedded (FFPE) tissue is problematic due to chemical modifications and continued degradation over time. We compared quantity and quality of RNA extracted by four different protocols from 14 ten year old and 14 recently archived (three to ten months old) FFPE breast cancer tissues. Using three spin column purification-based protocols and one magnetic bead-based protocol, total RNA was extracted in triplicate, generating 336 RNA extraction experiments. RNA fragment size was assayed by reverse transcription-polymerase chain reaction (RT-PCR) for the housekeeping gene glucose-6-phosphate dehydrogenase (G6PD), testing primer sets designed to target RNA fragment sizes of 67 bp, 151 bp, and 242 bp.

Results: Biologically useful RNA (minimum RNA integrity number, RIN, 1.4) was extracted in at least one of three attempts of each protocol in 86-100% of older and 100% of recently archived ("months old") samples. Short RNA fragments up to 151 bp were assayable by RT-PCR for G6PD in all ten year old and months old tissues tested, but none of the ten year old and only 43% of months old samples showed amplification if the targeted fragment was 242 bp.

Conclusion: All protocols extracted RNA from ten year old FFPE samples with a minimum RIN of 1.4. Gene expression of G6PD could be measured in all samples, old and recent, using RT-PCR primers designed for RNA fragments up to 151 bp. RNA quality from ten year old FFPE samples was similar to that extracted from months old samples, but quantity and success rate were generally higher for the months old group. We preferred the magnetic bead-based protocol because of its speed and higher quantity of extracted RNA, although it produced similar quality RNA to other protocols. If a chosen protocol fails to extract biologically useful RNA from a given sample in a first attempt, another attempt and then another protocol should be tried before excluding the case from molecular analysis.

Figure 1

Amplification of G6PD by multiplex RT-PCR and analyzed by electrophoresis on 2% agarose gels. A. Representative examples of amplification products from RNA extracted from ten year old FFPE, samples 6 and 12; B. representative examples of amplification products from RNA extracted from recently archived months old FFPE, samples 15, 23, 22, 24. RNA fragments extracted from all 28 of the ten year old and months old samples were of sufficient size to generate shorter amplicons of G6PD 67 bp and 151 bp, visible on the gel; the 242 bp amplicon was visualized in none of the 10 year old samples and only six of 14 (43%) of the months old samples. All non-visualizing amplification products in the multiplex experiments were confirmed by PCR with single primer sets. M = marker (DNA ladder).