Applied Biosystems scientists examined the effect of section thickness (5–40 µm) on microRNA (miRNA) isolation from formaldehyde- or paraformaldehyde-fixed, paraffin-embedded (FFPE) tissues. Their experiments show how thicker sections, which contain more intact cells, produce more reliable results for miRNA analysis.


A concern when working with FFPE tissues is that nucleic acids are trapped and modified through protein-protein and protein-nucleic acid crosslinks. Because miRNAs are so small (~21 nucleotides) and adducts tend to occur about every hundred nucleotides, it is reasonable to assume that the majority of miRNAs in the section are unmodified. However, if these miRNAs dissolve, even to a slight degree, in xylene or ethanol, they can be lost during the deparaffinization steps unless a barrier is imposed. The primary barrier to miRNA loss is the cell membrane, which is often breached with single-cell-layer sections (usually ~7 µm). In this experiment, we tested the hypothesis that the use of thin sections would negatively affect miRNA recovery and expression analysis from FFPE tissues.

Experimental Design

Duplicate sections were cut with thicknesses of 5, 10, 20, and 40 µm from human breast tumor and lung tumor FFPE tissue samples. Deparaffinization, digestion, and RNA isolation were performed according to the RecoverAll™ Kit protocol with a single elution of 60 µL at 50ºC. miRNA expression was measured using 10 ng of total RNA from each sample with TaqMan® MicroRNA Assays for six targets (miR-16, Let-7a, miR-106a, miR-320, miR-134, and miR-224) and the U6 endogenous control. Raw CT values were collected and averaged.


The relative expression patterns of the miRNA targets remained unaltered for all samples (Figure 1), but the 5-µm sections showed decreased miRNA expression: CT values were 1–2 cycles higher than that of the 40-µm sections. The 10-µm sections showed a slight decrease in miRNA expression, but the decreased expression is clearly more pronounced in the 5-µm samples. The U6 control, which detects a longer and thus more cross-linked and protected RNA, remained unaltered for all samples in this study.

Figure 1. miRNA Expression Analysis From FFPE Sections of Different Thicknesses. The graphs show the average raw CT values for sections of different thickness for formaldehyde-fixed, paraffin-embedded (FFPE) human breast tumor (A) and lung tumor (B) tissues. Total RNA (10 ng) from each sample underwent a single round of reverse transcription (RT) using the TaqMan® MicroRNA Reverse Transcription Kit. Each RT reaction (1.33 µL) was run in triplicate PCRs using TaqMan MicroRNA Assays for 6 targets and the U6 endogenous control on the Applied Biosystems 7900HT Fast Real-Time PCR System. CT values from 6 reactions (triplicates from duplicate isolations) were averaged, and the standard deviation was calculated.


Our data indicate that for FFPE tissue, a section thickness of 10 µm or higher yields more sensitive data for miRNA analysis. Thinner sections can provide adequate sample for expression analysis and produce reliable results, but they may prove problematic for less-abundant targets as the CT values could move beyond the accepted range of detection.

Scientific Contributors
Emily Zeringer, Rick Conrad • Applied Biosystems, Austin, TX

For Research Use Only. Not for use in diagnostic procedures.