RNA from Air-Dried Frozen Sections for RTPCR and Differential Display

Evaluation of in vivo gene expression in specific types of cells is difficult because animal tissues are complex mixtures of many cell types, and RNA extracted from homogenized fresh whole tissue is derived from all cell types present. In an attempt to improve specificity, RNA has been isolated from specific cells microdissected from fixed, paraffin-embedded tissue sections (2,6), but the yield and quality of RNA have generally been suboptimal. Here we report an alternative method for recovering high-quality RNA from specific cells microdissected from airdried frozen histological sections of unfixed tissue and its analysis by reversetranscription polymerase chain reaction (RT-PCR) and by the relatively new technique of differential display RTPCR (DDRT-PCR) (1,3,4). To prepare air-dried frozen sections, unfixed tissue is quickly snap-frozen or removed from the freezer and adhered to a cryostat chuck. It is critical to avoid thawing in order to inhibit RNase activity. Sections are cut at 20 μm and thaw-mounted on glass slides. The slides are dried in a 37°C incubator for 5 min and either microdissected immediately or stored airtight with a desiccant at < -70°C. Typical equipment for microdissection includes an inverted microscope (with 4× and 10× objectives) and an attached mechanical micromanipulator for holding and manipulating the cutting tool (e.g., a 30gauge hypodermic needle superglued into a glass microcapillary tube). Specific cells can be visualized without staining the tissue if the microscope contrast is high (e.g., by unfocusing the condenser or using phase-contrast rings) and if an adjacent hematoxylinand eosin-stained section is used as a guiding template. Alternatively, staining briefly (30 s) with aqueous hematoxylin and quickly re-drying the slide allow for better direct visualization and only slightly decrease the yield of RNA. Accurately separating different types of cells requires familiarity with the histopathological features of the tissue. Approximately 1 × 104 cells are needed to obtain sufficient RNA to run one primer pair in DDRT-PCR or RTPCR. Thus, dissecting one to several slides may be required depending on the target cell distribution and density. Obtaining sufficient cells can be tedious and time-consuming if the target cells are rare and/or scattered in small groups. Harvesting enough cells can be greatly facilitated by pre-selecting specimens containing relatively large areas of apposed target cells. It is possible to routinely prepare samples enriched to >95% target cellularity, although 100% purity is nearly impossible due to intermingled capillary endothelium, fibroblasts, lymphocytes, etc. Once the tissue is harvested, RNA can be extracted from air-dried frozen histological sections of breast tissue using methods previously described (6). Following isopropanol-precipitation, sample RNA pellets are washed with 200 μL 80% ethanol, dried under a vacuum for 5 min, or alternatively, the pellets are swabbed with a cotton-tipped appplicator, quickly air-dried and then resuspended in 10 μL of 10× RT buffer (100 mM Tris-HCl, pH 8.3, 500 mM KCl and 30 mM MgCl2). RT of sample RNA was carried out in 0.5-μL thin-wall microcentrifuge tubes (PGC Scientific, Gaithersburg, MD, USA) in a total volume of 96 μL RT buffer (10 mM Tris-HCl, pH 8.3, 50 mM KCl and 3 mM MgCl2) along with 1 mM dNTPs, 0.5 nmol of random hexamer primers and 5 μL of sample RNA (the amount of RNA utilized in the reaction should be empirically determined for different tissues sources). Samples were denatured at 94°C for 2 min, cooled to 42°C for 2 min and then incubated at 42°C for 45 min following the addition of 2.0 U of avian myeloblastosis virus (AMV) reverse transcriptase (Life Sciences, St. Petersburg, FL, USA). Following denaturation of the reverse transcriptase at 94°C for 3 min, 0.1 nmol each of sense and antisense primers designed to amplify a small region of the human estrogen receptor (ER) and 2.5 U of AmpliTaq DNA Polymerase (Perkin-Elmer, Norwalk, CT, USA) were added. The samples were covered with 100 μL of mineral oil, and the reaction tubes were heated at 94°C for 1 min. PCR was then carried out in a Perkin-Elmer cycler using 40 cycles of 94°C for 1 min, 55°C for 2 min and 72°C for 3 min. These parameters have proven to be optimum when amplifying this fragment of the ER gene. However, cycling parameters should be optimized for each gene-specific primer pair utilized. These cycling steps were then followed by a single extension cycle of 94°C for 1 min, 55°C for 2 min and 72°C for 7 min. The Taq polymerase was then denatured at 98°C for 10 min, and the samples were cooled to 27°C. Ten microliters of each sample were loaded onto a 5% minipolyacrylamide gel (Bio-Rad, Hercules, CA, USA), separated by electrophoresis and then stained with a 1% solution of ethidium bromide. As shown in Figure 1, PCR products of the expected size of 170 bp were amplified from four breast tissue samples using oligonucleotide primers specific for separate exons of the ER. It is advisable to choose primers across