DNA-free RNA isolation protocols for Arabidopsis thaliana, including seeds and siliques
© Oñate-Sánchez and Vicente-Carbajosa; licensee BioMed Central Ltd. 2008
Received: 05 August 2008
Accepted: 20 October 2008
Published: 20 October 2008
High throughput applications of the reverse transcriptase quantitative PCR (RT-qPCR) for quantification of gene expression demand straightforward procedures to isolate and analyze a considerable number of DNA-free RNA samples. Published protocols are labour intensive, use toxic organic chemicals and need a DNase digestion once pure RNAs have been isolated. In addition, for some tissues, the amount of starting material may be limiting. The convenience of commercial kits is often prohibitive when handling large number of samples.
We have established protocols to isolate DNA-free RNA from Arabidopsis thaliana tissues ready for RT-qPCR applications. Simple non-toxic buffers were used for RNA isolation from Arabidopsis tissues with the exception of seeds and siliques, which required the use of organic extractions. The protocols were designed to minimize the number of steps, labour time and the amount of starting tissue to as little as 10–20 mg without affecting RNA quality. In both protocols genomic DNA (gDNA) can be efficiently removed from RNA samples before the final alcohol precipitation step, saving extra purification steps before cDNA synthesis. The expression kinetics of previously characterized genes confirmed the robustness of the procedures.
Here, we present two protocols to isolate DNA-free RNA from Arabidopsis tissues ready for RT-qPCR applications that significantly improve existing ones by reducing labour time and the use of organic extractions. Accessibility to these protocols is ensured by its simplicity and the low cost of the materials used.
Reverse transcriptase quantitative PCR (RT-qPCR) is a very useful and sensitive technique to quantitate changes in gene expression. Increasing number of labs routinely use the agarose-based or/and real-time versions of the technique. A significant number of replicates are required to ensure significance of the results, which adds a high-throughput component to the RT-qPCR. These requirements impose the need to have simplified protocols to reduce labour and costs without compromising RNA integrity or yield.
Since Arabidopsis thaliana is a model organism for plant biologists, a number of RNA extraction protocols have been developed. Although there are several commercial kits available, the cost can be prohibitive when processing large numbers of samples. Published protocols use organic extractions with toxic chemicals, which involve the use of fume hoods to diminish health risks and lengthen the procedure [1–5]. Moreover, none of these protocols reduce genomic DNA (gDNA) contamination enough below PCR-detection levels. Digestion with DNase removes traces of DNA and is compulsory if the RNA samples are devoted to RT-qPCR. DNase digestion after the final RNA precipitation step involves adding extra salts and protein to the sample and, since they can affect the efficiency of the cDNA synthesis, additional purification steps are required. The amount of starting tissue is another factor to consider, especially when enough sample quantities are difficult to obtain and to speed up work by fitting volumes into standard microfuge tubes (SMTs). Polysaccharides and other compounds frequently contaminate RNA samples from seeds and siliques  and protocols suitable for RNA isolation from other tissues have to be adapted. Although high-salt extraction buffers have been used to get rid of those contaminants [1–3, 5] large RNAs may become insoluble in these buffers, therefore decreasing the complexity and information contained in the samples.
Here we report improved RNA isolation protocols for the model plant Arabidopsis thaliana that address the problems mentioned above. Both protocols start with small amounts of tissue and buffer volumes to fit the SMT format, and incorporate a DNase digestion before the final purification steps.
RNA extraction from Arabidopsis tissues other than seeds and siliques
To demonstrate the quality of the samples and sensitivity of the protocol, RT-qPCR was performed on RNA samples isolated from Arabidopsis leaves of the Columbia-0 ecotype (wild type) and leaves of transgenic plants overexpressing AtDOF31 (35S:AtDOF31). AtDOF31 is a DOF transcription factor (TF) with very low relative transcript abundance, whose expression cannot be detected by in situ hybridizations or by promoter::GUS fusions (Oñate-Sánchez et al., unpublished). Gene-specific oligonucleotides were chosen to amplify regions near the 5' end to ensure RNA integrity and cDNA synthesis efficiency and a ubiquitin gene was amplified for sample normalization. cDNA was synthesized according to , diluted 10 times and 2 μl were used as a template for real time PCR together with forward and reverse oligonucleotides (0.5 μM each) in 1× Power SYBR Green Master mix (Applied Biosystems). Cycling conditions (Applied Biosystems 7300 instrument) were as follows: 10 min at 95°C, 50 cycles of 15 s at 95°C and 60 s at 60°C, linked to a default dissociation stage program to detect non-specific amplification. As shown in Figure 2B, basal levels of AtDOF31 expression were detected in the wild type sample and several transgenic lines with different levels of AtDOF31 overexpression relative to the control (Col-0) were identified.
RNA extraction from Arabidopsis seeds and siliques
Arabidopsis seeds and siliques are tissues with a high content of polysaccharides and other compounds, which can appear as contaminants of RNA samples  and attempts to isolate clean RNA from these tissues using the protocol described above were unsuccessful. Although there are published protocols that use organic extractions [1–5], we have improved them by using simple buffers and including a step to precipitate carbohydrates, which bypass the need to use high salt concentrations in the initial steps of purification (Figure 1 – Protocol 2). We have also shortened the protocol and slightly increased yield, despite starting from smaller amounts of tissue. Due to the small size of Arabidopsis seeds, the availability of sufficient quantities of starting material is a limiting factor, especially when working with conditions or genotypes producing low seed set. Published protocols on seed RNA isolation use 50 mg or higher amounts of tissue and/or non-SMTs at some point of the procedure [2, 5]. Higher RNA yields (1,584 ± 304 μg/g) than published protocols were obtained using only 20 mg of seeds. We observed that using phenol directly over the extraction buffer produced brownish RNA pellets, probably due to the presence of phenolic compounds. We decided to use chloroform prior to phenol to remove the phenolic compounds [ and references therein]. To shorten the protocol we mixed the extraction buffer with chloroform, thus avoiding RNA degradation and removing the phenolic compounds at the same time. The aqueous phase was then extracted with acid phenol to remove proteins and as much DNA as possible, and chloroform was added in the same step to facilitate the separation of organic and aqueous phases and to eliminate any possible contamination with phenol. Genomic DNA contamination was removed after RNA was precipitated with ClLi (data not shown). Protein and salts from the DNase digestion were removed during the final RNA purification step, which was divided into two parts to precipitate firstly, carbohydrates with low concentrations of sodium acetate and ethanol and secondly, RNA by increasing those amounts up to standard concentrations. RNA absorbance (A260/A230 and A260/A280) values (always above 2.0) and their electropherograms indicated the absence of undesirable contaminants (Figure 2C).
We have described two simplified, inexpensive and robust protocols to isolate DNA-free RNA from Arabidopsis tissues. As far as we know, all non-commercial protocols for RNA isolation from plant tissues use organic solvents during the first steps. We present here a protocol that uses two simple buffers without organic solvents that includes a 30 min DNase digestion and requires only 90 min to obtain 24 DNA-free RNA samples. Although only data from leaves is presented here, RNA from most Arabidopsis tissues can be isolated since we have successfully used it with stems, roots and flowers (data not shown) and also from different vegetative tissues of the legume model plant Medicago truncatula (Gao, personal communication). The only tissues not amenable for the previous protocol were seeds and siliques and we streamlined and improved existing protocols that use organic extractions for RNA purification from those recalcitrant tissues. This protocol has also been successfully used in the isolation of RNAs from developing barley seeds (Oñate-Sánchez and Vicente-Carbajosa, unpublished). Altogether, our results indicate that the use of these protocols can be extended to other plant species.
We determined that the RNA obtained at an intermediate step of the isolation protocol was of sufficient quality to allow DNase digestion before the final purification steps. This modification shortens the protocol and maximizes yield bypassing the need to double precipitate the RNA. RNA integrity and purity are key factors influencing cDNA synthesis and ultimately detection and accurate quantification of RNAs, especially low abundance RNAs such as those derived from many TF coding genes. AtDOF31 and AtBME3 are expressed at low levels and AtBME3 mRNA has a length (1700 bp) above the average for a TF. After cDNA synthesis using oligo dT and qPCR using specific oligonucleotides annealing as far as possible from the polyA, we were able to detect and quantify the expression levels of both genes in all samples analyzed, confirming sensitivity and specificity.
Accessibility to these protocols is ensured by its simplicity and low cost of the materials used. With the increase of genomic approaches involving the manipulation of large numbers of samples, we feel these protocols will be extremely useful for plant scientists.
We thank Inmaculada Gude and Elaine Smith for excellent technical assistance. We also thank Prof. Pilar Carbonero and Drs. Louise Thatcher and Hannetz Roschzttardtz for critical reading of the manuscript. Financial support from Spanish Agency for International Cooperation (A/012927/07), Ministerio de Educación y Ciencia (Spain; MEC BMC2006-07258), Universidad Politécnica de Madrid (UPM-M 0700204101), Fundación Genoma España trilateral TRI21/REGULATORS, GEN2003-20859-C03-01 and a Marie Curie International Reintegration Grant (36524) to L.O., are gratefully acknowledged. L.O. was supported by the Spanish Ministry for Education and Science with a Ramón y Cajal contract.
- Carpenter CD, Simon AE: Preparation of RNA. Methods in Molecular Biology, Arabidopsis Protocols. Edited by: Martínez-Zapater JM, Salinas J. 1998, Totowa: Humana Press, 82: 85-89.View ArticleGoogle Scholar
- Vicient CM, Delseny M: Isolation of total RNA from Arabidopsis thaliana seeds. Anal Biochem. 1999, 268: 412-413. 10.1006/abio.1998.3045.View ArticlePubMedGoogle Scholar
- Sharma AD, Gill PK, Singh P: RNA isolation from plant tissues rich in polysaccharides. Anal Biochem. 2003, 314: 319-321. 10.1016/S0003-2697(02)00689-9.View ArticlePubMedGoogle Scholar
- Weigel D, Glazebrook J: Arabidopsis: A laboratory manual. 2002, New York: Cold Spring Harbor, Cold Spring Harbor Laboratory PressGoogle Scholar
- Suzuki Y, Kawazu T, Koyama H: RNA isolation from siliques, dry seeds, and other tissues of Arabidopsis thaliana. BioTechniques. 2004, 37: 542-544.PubMedGoogle Scholar
- Heath EM: Low pH RNA isolation reagents, method, and kit. United States Patent 5973137. 1999Google Scholar
- Oñate-Sánchez L, Singh KB: Identification of Arabidopsis ethylene-responsive element binding factors with distinct induction kinetics after pathogen infection. Plant Physiology. 2002, 128: 1313-1322. 10.1104/pp.010862.PubMed CentralView ArticlePubMedGoogle Scholar
- Yamaguchi S, Kamiya Y, Sun T: Distinct cell-specific expression patterns of early and late gibberellin biosynthetic genes during Arabidopsis seed germination. Plant J. 2001, 28: 443-453. 10.1046/j.1365-313X.2001.01168.x.View ArticlePubMedGoogle Scholar
- Liu PP, Koizuka N, Martin RC, Nonogaki H: The BME3 (Blue Micropylar End 3) GATA zinc finger transcription factor is a positive regulator of Arabidopsis seed germination. Plant J. 2005, 44: 960-971. 10.1111/j.1365-313X.2005.02588.x.View ArticlePubMedGoogle Scholar
- Wobus U, Weber H: Seed maturation: genetic programmes and control signals. Curr Opin Plant Biol. 1999, 2: 33-38. 10.1016/S1369-5266(99)80007-7.View ArticlePubMedGoogle Scholar
- Hoekstra FA, Golovina EA, Buitink J: Mechanisms of plant desiccation tolerance. Trends Plant Sci. 2001, 6: 431-438. 10.1016/S1360-1385(01)02052-0.View ArticlePubMedGoogle Scholar