Introduction

Formalin-fixed, paraffin-embedded (FFPE) tissue offers a vast source of biopsy specimens for which the clinical outcome is well documented and thus optimal for retrospective studies (1). Over the past decade, FFPE tissue sections have been increasingly used as a source of RNA for gene expression profiling (2,3,4,5). However, RNA extracted from FFPE samples has proven to be problematic. First, RNA is significantly degraded in the tissue before, during, and after formalin fixation (6,7). Extended storage also results in more severe RNA deterioration, leading to lower extracted RNA yields. Secondly, the fixed RNA becomes more resistant to extraction due to cross-linking with proteins (2,8). Thirdly, formalin fixation causes mono-methylol addition on all four bases, resulting in a lower efficiency of reverse transcription (8,9). As a result, the current methods of gene expression analysis, such as real-time reverse transcription PCR (RT-PCR), which rely on RNA extraction and reverse transcription, have been consistently less sensitive using archived FFPE tissues than with fresh and frozen tissues (5)(10,11,12).

The branch DNA technology is a sandwich nucleic acid hybridization assay that provides a unique approach for messenger RNA (mRNA) detection and quantification by amplifying the reporter signal rather than target sequences (13,14,15,16). By measuring mRNA directly from crude cell lysates and tissue homogenates, the assay avoids variations or errors inherent to extraction and amplification of target sequences. To explore its potential application in archived clinical specimens, we have developed a modified branch DNA assay to quantitatively measure mRNA directly in FFPE tissue homogenates. In this procedure, FFPE sections are treated to remove paraffin, break the cross-linking bonds between RNA and proteins, and lyse the cells to release mRNA. An aliquot of the soluble tissue homogenate is then applied directly to the branch DNA assay. Here we compare the impact of formalin fixation and RNA degradation on the performance of the branch DNA and real-time RT-PCR assays as well as the detection sensitivity measuring mRNA expression in FFPE lung tumor specimens.

Materials and Methods

Tissue Homogenate Preparation and Total RNA Isolation from FFPE Tissue Sections

FFPE specimens from 3- and 14-year-old human lung adenocarcinoma and normal adjacent tissues were obtained from Analytical Biological Services, Inc. (Wilmington, DE, USA). Twenty sections of FFPE specimens from the same tissue block were scraped off from glass slides using a scalpel and transferred to a microcentrifuge tube. One milliliter of xylene-containing EZ-DeWax™ (BioGenex, San Ramon, CA, USA) was added to each sample. After briefly vortex mixing and incubating at room temperature for 5 min, the tissue samples were centrifuged in a microcentrifuge at 16,000× g for 2 min, and the supernatants were removed. One milliliter 70% ethanol was added to the samples, and the samples were vortex mixed and centrifuged in a microcentrifuge at 16,000× g for 2 min. The samples were dewaxed again with the EZ-DeWax and washed with 70% ethanol for two more times before they were divided equally for tissue homogenate preparation or total RNA isolation. To prepare tissue homogenate for the branch DNA assay, 300 µL Homogenizing Solution (Genospectra, Fremont, CA, USA) containing 100 µg proteinase K were added to each dewaxed sample, and the samples were incubated at 65 °C overnight. Total RNA was isolated from the dewaxed samples using Optimum™ FFPE RNA Isolation kit (Ambion, Austin, TX, USA). The total RNA was quantified with a SpectraMax® 384 plus (Molecular Device, Sunnyvale, CA, USA).

Formalin Fixation of HeLa Cells

HeLa cells (ATCC, Manassas, VA, USA) were cultured in Dulbecco's modified Eagle's medium (DMEM; Invitrogen, Carlsbad, CA, USA) at an approximate density of 1–2 × 10⁶ cells/mL. Ten million HeLa cells were collected, incubated with 10 mL RNAlater® (Ambion) for 4 h to stabilize the RNA, and washed twice with 50 mL ice-cold 1× phosphate-buffered saline (PBS). The cells were then incubated with or without 10 mL 1% formalin at 4°C for 1 or 16 h. After being washed twice with 1× PBS, the cells were either lysed with branch DNA Lysis Mixture (Genospectra) or used for RNA isolation with the Optimum FFPE RNA Isolation kit.

RNA Degradation

Fifty micrograms human lung total RNA (Ambion) were treated with 0.1 M NaOH at room temperature. At predetermined time points, an equal volume of 0.1 M HCl was added to the sample to neutralize NaOH and stop RNA degradation. The resulting degraded RNA was used for branch DNA or real-time RT-PCR assays as described below.