For studying differential gene expression by reverse-transcriptase PCR (RT-PCR) applications, eliminating genomic DNA contamination from the cDNA is essential prior to amplification of the specific gene sequence. Residual genomic DNA can disturb the amplification of the target gene from cDNA and lead to false-positive results. DNase I treatment is generally used to remove genomic DNA from RNA samples. However, RNA can degrade totally or partially during the DNase treatment. It has also been shown that some contamination of genomic DNA can remain after the DNase treatment, even after an overnight incubation (1). Certain tissue types can be a problem because some contain elevated levels of RNases (2,3) or a limited amount of RNA (4). Fruit tissues are regarded as particularly problematic because of high amounts of secondary metabolites, polysaccharides, and elevated levels of RNases (5). In our experiments with RNA from different plant tissues, especially bilberry (Vaccinium myrtillus) fruit and leaf, we found that RNA degraded during the DNase treatment. We also noticed that despite the DNase treatment, some genomic DNA was always left in the samples when RT-PCR gene expression analysis was performed on Scots pine (Pinus sylvestris) needles. To overcome these problems, we developed a method based on electrophoresis and gel-extraction filter tubes.
Bilberry fruits and leaves and Scots pine needles growing in a forest in Oulu, Finland, were snap frozen in liquid nitrogen and stored at °80°C until use. Total RNA was isolated from the samples with the method described for pine trees (6) that had been modified by Jaakola et al. (7). The quality of the isolated RNA was verified on a 1% (w/v) ethidium bromide-stained agarose gel and from the absorbance measurements at wavelengths 230, 260, and 280 nm.
From each sample, 5 µg were taken for DNase treatment with TURBO™ DNase (Ambion, Austin, TX, USA), according to the manufacturer's instructions (1 U/1 µg DNA) using SUPER-ase-In™ RNase inhibitor (Ambion) in a reaction volume of 50 µL at 37°C for 30 min. (Figure 1)A shows the isolated RNA before and after the DNase treatment. As seen in the figure, RNA from bilberry fruits (lane 6) is totally degraded, and RNA from leaf sample (lane 5) is mostly degraded after the DNase treatment. The RNA from pine needles is still intact, but the amount has been reduced to 1/5 of the original.Figure 1.
In addition, 5 µg of each sample of non-DNase-treated RNA were reverse transcribed using Moloney murine leukemia virus (M-MLV) reverse transcriptase (Invitrogen, Carlsbad, CA, USA) with anchored oligo(dT) primers using standard methods in a reaction volume of 20 µL. The samples were separated for 15 min at 40 volts (V) on a 1% ethidium bromide-stained agarose gel prepared in Tris-acetate-EDTA (TAE) buffer. The cDNA fragments were trimmed out of the gel according to the size information (<10 kb), leaving the genomic DNA on the gel ((Figure 1)B). The fragments were extracted from agarose gel slices with the Montage® DNA Gel Extraction Kit (Millipore, Bedford, MA, USA), which is designed to extract DNA fragments that are 100 to 10,000 bp in size in one 10-min centrifugation at 5000× g. After the centrifugation, the solution containing the cDNA was precipitated by adding two volumes of ice-cold absolute ethanol, incubating at -70°C for 30 min, and centrifuging at 18,000× g for 15 min at 4°C. The pellet was suspended in 10 µL sterilized water.
For verifying the purity of the cDNA from genomic DNA, PCR was performed using the gel-purified cDNA or genomic DNA as template. For the latter, DNA was isolated from pine needles and bilberry leaves with the method described by Pirttilä et al. (8). DyNA-zyme™ DNA polymerase (Finnzymes, Espoo, Finland) was used for the PCR amplification in a final volume of 25 µL. Conditions for PCR for pine and bilberry cDNA and DNA were 30 cycles of 94°C for 60 s, 54°–63°C for 2 min, and 72°C for 2 min. (Table 1) presents the annealing temperature for each primer. The PCR products were separated on a 1% (w/v) agarose gel stained with ethidium bromide. Precise sizes for expected amplification products are given in (Table 1); in all cases, the primer pairs encompass a region containing at least one intron, thus allowing cDNA and genomic DNA amplification products to be distinguished. As a first analysis, primer pairs against glyceraldehyde-3-phosphate dehydrogenase (GAPD) were used to amplify products from pine and bilberry cDNA and DNA samples ((Figure 1)C). No bands indicative of genomic DNA contamination were seen in the gel-purified cDNA samples. To confirm the importance of the purification step, additional primer pairs against actin (ACT), chalcone synthase (CHS), and pinosylvin synthase (BBS) were used for amplifying fragments from pine cDNA samples with and without gel separation and from the genomic DNA as a control ((Figure 1)D). In samples without gel separation, using CHS and BBS primers, the amplification occurred primarily from genomic DNA instead of cDNA. With ACT primers, both cDNA and genomic DNA were equally amplified from the untreated samples ((Figure 1)D). These results clearly show how the presence of genomic DNA disturbs the amplification of a target gene from cDNA. In quantitative RT-PCR, the results would thus be greatly distorted.Table 1. Sequences of the Primers and Information of the Reaction Conditions and Amplification Products
Ta, annealing temperature; gDNA, genomic DNA.
The elimination of genomic DNA from RNA or cDNA is required to study gene expression with RT-PCR or micro-arrays. A common strategy for removing genomic DNA is DNase treatment; however, typical problems include the degradation of the RNA and the presence of residual contaminating genomic DNA. Our results show that the method described here is suitable for obtaining intact and functional cDNA that is free of genomic DNA. We have found this method useful with all RNA samples, but especially with samples that degrade easily during the DNase treatment step. The method is efficient because no time-consuming extraction steps with hazardous chemicals are needed. Flohr et al. (9) reported the possibility of performing the DNase treatment on cDNA instead of RNA, which is also an alternative for the RNA samples that degrade easily during the DNase treatment. However, because of the extraction steps, some loss of the sample always occurs during the DNase treatment. Although some reduction in sample amount is also encountered during the gel-extraction step, we have found this method more reliable because residual genomic DNA was frequently detected when DNase treatment was used.
This work was supported by Academy of Finland, SUNARE project (Sustainable Use of Natural Resources) grant no. 52741, and by the Emil Aaltonen Foundation.
The authors declare no competing interests.