Overcoming Challenges in Gene Expression Analysis of FFPE Samples

With RT2 PreAmp and RT2 Profiler PCR Array technology
  • Main Image Navi
Yexun Wang, Emi Arikawa, Min You, Shankar Sellappan, and Li Shen
QIAGEN, Fredrick, MD, USA

Gene expression profiling of formalin-fixed, paraffin-embedded (FFPE) samples can be challenging due to the process of tissue fixation, which causes chemical modifications that decrease the quality and availability of recovered RNA. This study demonstrates a novel preamplification step using gene-specific primers matched to a specific RT2 Profiler PCR Array, which results in an increase in the sensitivity of gene expression analysis of at least 100-fold.
Introduction
Results and discussion
Conclusion
Introduction
FFPE tissue samples represent an invaluable source of material for biomedical research and enable scientists to discover important links between molecular and clinical information. However, working with FFPE samples can be challenging due to nucleic acid crosslinking, fragmentation, and limited sample size. The ability to perform reliable gene expression analysis of FFPE tissue samples depends on the success of several steps in the workflow, including RNA extraction, reverse transcription (RT), qPCR primer design, and preamplification.

In this study, the results from a Human Cancer PathwayFinder RT2 Profiler PCR Array show that introducing a preamplification step using gene specific primers that are matched to the RT2 Profiler PCR Array increases the sensitivity at least 100-fold in real-time PCR gene analysis. With this RT2 PreAMP technology, FFPE RNA can serve as a powerful tool for both retrospective and prospective gene profiling studies for biomarker identification and characterization.

Results and discussion
Considerations for successful gene expression analysis from FFPE samples
Careful consideration of every stage of the FFPE gene expression analysis workflow is necessary to achieve reliable results (see figure Gene expression analysis of FFPE samples). RNA extraction from FFPE samples must ensure effective paraffin removal, de-crosslinking of RNA, RNA hydrolysis, and recovery of RNA fragments (for effective purification from of RNA from FFPE samples, we suggest the RNeasy FFPE Kit; click here for additional solutions for working with FFPE samples). Additional considerations include use of random or gene-specific primers for the RT reaction, designing qPCR primers for long or short amplicons, and strategies to increase signal sensitivity without altering the original gene expression profiles.
Choice of primers for efficient reverse transcription of FFPE samples
When working with FFPE RNA, fewer primers will generate cDNAs containing a target amplicon due to the high degree of RNA fragmentation that occurs during the FFPE fixation process (see figure Reverse transcription of RNA from FFPE samples). The use of gene-specific primers for the RT reaction increases the probability that cDNA corresponding to the amplicon of interest will be produced (see figure Gene-specific primers for RT from FFPE samples).
Considerations when designing primers for real-time PCR
PCR primers are often designed to generate amplicons that range in size from 130 to 200 bp. Although this is suitable for intact RNA, the high degree of FFPE RNA fragmentation means that smaller PCR amplicons are necessary to ensure amplification will occur (see figure Effective primer design for qRT-PCR from FFPE samples).
Strategies for amplifying FFPE RNA without altering the gene expression profile
Preamplification of RNA samples can make it possible to perform more analyses on limited FFPE samples. However, it is essential that the method used amplifies all regions of the RNA equally to ensure that the gene expression profile of preamplified samples is equal to the gene expression profile of unamplified samples. RT2 PreAmp Primer Mixes, which are designed specifically for RT2 Profiler PCR Arrays, effectively amplify RNA without altering the gene expression profile (see figure Highly comparable CT values from PreAMP and unamplified RNA isolated from an FFPE sample).
Gene expression analysis of RNA from FFPE samples using the Human Cancer PathwayFinder RT2 Profiler PCR Array
Due to small sample amounts, it is not always possible to obtain complete gene expression profiles using RNA from FFPE samples. Amplification of RNA using the PreAMP process enables quantification of genes that could not be detected without preamplification (see figure The PreAMP process makes more genes quantifiable from FFPE samples).

When comparing the gene expression profiles of 2 different types of FFPE tissue (liver and intestine), comparable results were obtained whether the RT2 PreAMP system or unamplified RNA was used (see figure Comparable fold changes).

Conclusion
The RT2 PreAMP system solves major challenges in real-time PCR-based gene expression analysis using FFPE samples. It greatly improves the lower limit of quantitation for FFPE RNA and also enables easy integration with the RT2 Profiler PCR Array platform to analyze pathway-focused gene panels. With the improvement in both reverse transcription and qPCR primer design coupled with the introduction of a novel workflow tailored to the unique nature of these samples, better utilization of valuable FFPE sample resources is expected. These results demonstrate that PreAMP technology and the RT2 Profiler PCR Array system is a valuable tool available for biomarker discovery or validation utilizing FFPE samples.

For up-to-date licensing information and product-specific disclaimers, see the respective QIAGEN kit handbook or user manual. QIAGEN kit handbooks and user manuals are available at www.qiagen.com or can be requested from QIAGEN Technical Services or your local distributor.

Imagens
Gene expression analysis from FFPE samples.
Gene expression analysis from FFPE samples.
The protocol for performing PCR array-based, pathway-focused gene expression analysis on FFPE or other chemically fixed samples.
Reverse transcription of RNA from FFPE samples.
Reverse transcription of RNA from FFPE samples.
[A] For intact RNA from fresh samples, any one of the primers can generate cDNA that contains the target amplicon. [B] For fragmented RNA from FFPE samples, only primer 4 can generate cDNA that contains the target amplicon.
BC_0436_RT2ProfilerPCRArray
Gene-specific (GS) primers improve the efficiency of reverse transcription of FFPE RNA.
RNA (500 ng) extracted from one 20 μm section of a five year old human spleen FFPE block was reversed transcribed using either random hexamers or a pool of 89 gene-specific primers positioned downstream of the PCR amplicons. CT values were lower when the GS primers were used, indicating that the GS primers provided more usable cDNA for real-time PCR analysis.
Effective primer design for qRT-PCR from FFPE samples.
Effective primer design for qRT-PCR from FFPE samples.
[A] RNA extracted from one 20 μm section of a five year old human spleen block was reversed transcribed and analyzed using two sets of assays from different primer designs: regular amplicons (132 to 191 bp) and short amplicons (53 to 81 bp). [B] cDNA transcribed from 1 μg high quality human universal RNA was reversed transcribed and analyzed using two sets of 45 gene assays from different primer designs: regular amplicons (132 to 191 bp) and short amplicons (53 to 81 bp). Primers designed for shorter PCR amplicons generally yield lower CT values for fragmented RNA from FFPE sources, but not intact RNA, as they provide more usable cDNA for real-time PCR analysis.
Highly comparable CT values from PreAMP and unamplified RNA isolated from an FFPE sample.
Highly comparable CT values from PreAMP and unamplified RNA isolated from an FFPE sample.
RNA extracted from a human spleen FFPE block was reverse transcribed and prepared for qPCR, with and without the PreAMP process. Both cDNAs were run on the Human Cancer PathwayFinder RT2 Profiler PCR Array (PAHS-033). [A] Every assay CT value is improved by approximately the same number of cycles after the PreAMP process. Only genes which have CT values lower than 31.5 in the reaction without PreAMP are shown here. [B] The scatter plot of raw CT values. [C] Inter-assay consistency of the PreAMP process using the mouse Toll-like Receptor Signaling Pathway PCR Array (PAMM-018). Two independent PreAMP reactions were set up using the same amount of mouse cDNA. ΔCT (CTGOI — CTHKG) values for 89 genes from two separate runs shows a high correlation, r >0.99.
BC_0437_RT2ProfilerPCRArray
The PreAMP process makes more genes quantifiable from FFPE samples.
RNA extracted from a five year old human intestine FFPE block was reverse transcribed with or without the PreAMP process. Both cDNAs were run on Human Cancer PathwayFinder RT2 Profiler PCR Array (PAHS-033). Shown here are 18 genes which were called "absent" using standard RT-PCR array procedures but produce quantitative CT values after PreAMP.
Comparable fold changes.
Comparable fold changes.
RNA samples extracted from human spleen and intestine FFPE blocks were reverse transcribed with or without the PreAMP process. All four cDNA samples were run on Human Cancer PathwayFinder RT2 Profiler PCR Array. Shown here are the relative gene expression differences between the two tissues. All the genes showing significantly changed expression in original FFPE samples were also detected as significantly changed in the same direction after the PreAMP process (in white region [A] and [B]). Although some genes showed fold-change values (as identified by the arrow) in the opposite directions, the values were not significant in either preamplified or unamplified samples. Only genes that have CT values lower than 31.5 in both unamplified spleen and intestine samples are shown here.