Scripps Research Logo

DNA Array Core Facility

RNA Isolation

While we do not normally offer RNA extraction as a service in the DNA Array Core Facility, preparation of RNA for microarray analysis is a critical step and one of the most important sources of variability in microarray data. RNA extraction is required whenever microarray studies are to be performed. Successful RNA extraction will yield total RNA with minimal degradation and free of any contaminating RNAses. We advise using our RNA Extraction Protocol when preparing your RNA for submittal to the Core Facility.

RNA Extraction:

Our lab has expended a significant effort towards development and optimization of RNA extraction protocols for use in subsequent microarray hybridization experiments. Important issues related to RNA extraction include variability in RNA yields from various tissues and states of cell activation, processing different quantities of tissue, and purity of extracted RNA. The end product of a successful RNA isolation procedure will yield minimally degraded total RNA free from contaminating agents, such as RNAses and other proteins, and genomic DNA.

Total RNA yields (in ug/mg tissue) from various mouse tissues

Figure 1. (A) Total RNA yields (in ug/mg tissue) from various mouse tissues. All tissues were flash frozen in liquid nitrogen immediately after harvesting and stored at -80°C prior to RNA extraction. Frozen tissue samples were ground using a mortar and pestle (Fisher) before immersion in Trizol reagent. Total RNA was extracted from 700-1500mgs of tissue per sample type using Invitrogen's recommended protocol for the Trizol reagent.

Bargraph of Total RNA yields from mouse splenocytes nonactivated and 24 or 48 hour activated

Figure 1. (B) Total RNA yields from mouse splenocytes nonactivated and 24 or 48 hour activated. Splenocytes were obtained from mice by dissecting out the spleen and straining through a wire mesh. The cell suspensions were washed with PBS and layered on histopaque. After a 30 minute spin at 1800rpm the layer containing the splenocytes was removed, washed 3 times in PBS and counted. For the unactivated splenocytes, the cells were resuspended in 10ml PBS in 15ml tubes. The cells were pelleted and the pellets were immediately frozen in liquid nitrogen and stored at -80°C. A portion of the cells were activated with ionomycin and PMA (500ng/ml and 10ng/ml, respectively) and harvested at 24hr and 48hr time points. Total RNA was extracted from these cells as described above.

RNA can be extracted from a variety of starting materials, including whole or partial tissue samples, cultured cells, or purified cell populations (for example a lymphocyte prep from whole blood or cells selected by FACS or laser capture microdissection techniques). A survey of RNA yields from various mouse tissues is shown in Figure 1A. These results indicate an approximately 10-fold range of RNA yields/mg of tissue across different tissue types. Figure 1B shows RNA yields/cell for mouse splenocytes (nonactivated as well as 24 and 48 hours after activation with ionomycin/PHA). RNA yields per cell after 48 hours of activation increase by five-fold relative to nonactivated cells.

These data are indicative of the variability in cellular RNA yields across different tissue and cell types as well as between different physiologic states of similar cell populations. Our data for RNA yields per cell (Figure 1B) are somewhat less than other reported values (20-40pg/cell, Roozemnond, 1976; Uemura, 1980; 3-6 pg/cell, Baugh 2001; Alberts, 1994) and may reflect either an underestimation of cell numbers used to extract the RNA and/or our conservative RNA isolation procedures that include either Trizol (Invitrogen) or QiaShredder/RLT (Qiagen) extraction, a DNAse treatment to remove genomic DNA contamination and a final RNeasy column (Qiagen) purification. In our experience column purification steps may result in 10-50% loss of RNA depending on how much material is applied to the columns.

RNA Preservation:

It is critical that messenger RNA be preserved in its last functioning, physiological state and be prevented from degrading. This is achieved by one of two methods for tissues: snap freezing in liquid nitrogen or emersion in RNA Later. The liquid nitrogen process is particularly suitable for larger tissue samples ( >50mg). Tissues of this size are easily visualized and ground by a liquid nitrogen cooled mortar and pestle (Fisher) with negligible loss during processing. Smaller tissue samples, such as needle core biopsies or murine lymph nodes, are more suitable to preservation in RNA Later solution, a product distributed by both Ambion and Qiagen. This product is designed to penetrate cell membranes and inactivate cellular and other contaminating RNAses. Tissue pieces must be no greater than 5mm in thickness in order to allow adequate infiltration by the RNA Later solution and effectively prevent degradation. Small tissue samples preserved in RNA Later can be homogenized in tissue grinders (Fisher).

We have identified a difference in performance between RNA Later from Ambion vs. Qiagen at subzero temperatures. The Qiagen product does not freeze to a solid, but becomes progressively thicker as the temperature drops. At -20°C the Ambion variety freezes outright, forming crystals which are hard to redissolve at higher temperatures. This crystalline precipitate makes it difficult to separate out small pieces of tissue preserved in this solution. We generally recommend the Qiagen product for preservation of harvested tissues.

When working with cell preparations such as purified lymphocytes, pelleting the cells and resuspending the pellet in a chaotropic agent most effectively accomplishes both the preservation and recovery of cellular RNA. (Trizol or RLT buffers are commonly used for this purpose.) After suspension in the chaotropic agent, the cellular material can be homogenized using a QiaShredder column and the RNA extracted using an RNeasy column.

Regardless of the method used to extract the total cellular RNA, we have observed a significant improvement in yields of cDNA synthesized from RNA extracted using protocols that include a final column purification step (e.g. RNeasy). We believe this final "washing" step may remove residual components from the RNA that may inhibit one or more of the enzymes used in cDNA synthesis. In addition, a DNAse digestion step may be performed on the extracted RNA if gel analysis reveals contaminating genomic DNA. Contaminating DNA can interfere with accurate quantitation of RNA. We normally use Ambions's DNA Free DNAse digestion kit for this purpose.

RNA Quantitation:

At the conclusion of the extraction procedure, quantitation is done by taking the spectrophotomic absorbance at 260nm and multiplying the reading by 40ug/ml and the appropriate dilution factor. Assessment of RNA quality is done in two ways: by calculating a spectrophotomic A260/A280 ratio and by electrophoretic analysis. The A260/A280 ratio should fall in the range of 1.8 - 2.2. Electrophoresis can be performed on an agarose gel for samples in which 1ug can be spared for the analysis. For smaller samples, we use the Agilent Bioanalyzer (Figure 2) because this microelectrophoresis instrument can analyze as little as 50ng of total RNA. Electrophoresis results should show two strong, distinct bands representing ribosomal RNA and a light smear behind the ribosomal bands representing the messenger RNA. In addition, significant high molecular weight product indicative of contaminating genomic DNA should not be observed.

Bargraphs of electropherograms generated by the Agilent Bioanalyzer

Figure 2. Electropherograms generated by the Agilent Bioanalyzer showing (A) high quality, (B) degraded and (C) partially degraded total RNA samples. Each trace represents 50ng of total RNA and data are scaled to match peak heights between traces. (The two distinctive peaks in figure A represent the ribosomal RNA)

References:

  1. Alberts, B. Molecular Biology of the Cell (Garland, New York and London; 1994).
  2. Baugh LR, Hill AA, Brown EL, Hunter CP (2001). Quantitative analysis of mRNA amplification by in vitro transcription. Nucleic Acids Res Mar1; 29(5): E29
  3. Roozemond RC (1976). Ultramicrochemical determination of nucleic acids in individual cells using the Zeiss UMSP-I microspectrophotometer. Application to isolated rat hepatocytes of different ploidy classes. Histochem J Nov; 8(6)A: 625-38.
  4. Uemura E (1980). Age-related changes in neuronal RNA content in rhesus monkeys (Macaca mulatta). Brain Res Bull Mar-Apr; 5(2): 117-9.

Affymetrix Technical Note:

An Analysis of Blood Processing Methods to Prepare Samples for GeneChip Expression Profiling, Affymetrix, (pdf, 30 pages)