Reference genes for accurate gene expression analyses across different tissues, developmental stages and genotypes in rice for drought tolerance
- Isaiah M. Pabuayon†1,
- Naoki Yamamoto†1Email author,
- Jennylyn L. Trinidad1,
- Toshisangba Longkumer1,
- Manish L. Raorane1 and
- Ajay Kohli1Email author
© Pabuayon et al. 2016
Received: 3 March 2016
Accepted: 3 July 2016
Published: 18 July 2016
Quantitative reverse transcription PCR (qRT-PCR) has been routinely used to quantify gene expression level. This technique determines the expression of a target gene by comparison to an internal control gene uniformly expressed among the samples analyzed. The reproducibility and reliability of the results depend heavily on the reference genes used. To achieve successful gene expression analyses for drought tolerance studies in rice, reference gene selection should be based on consistency in expression across variables. We aimed to provide reference genes that would be consistent across different tissues, developmental stages and genotypes of rice and hence improve the quality of data in qRT-PCR analysis.
Ten candidate reference genes were screened from four ubiquitously expressed gene families by analyzing public microarray data sets that included profiles of multiple organs, developmental stages, and water availability status in rice. These genes were evaluated through qRT-PCR experiments with a rigorous statistical analysis to determine the best reference genes. A ubiquitin isogene showed the best gene expression stability as a single reference gene, while a 3-gene combination of another ubiquitin and two cyclophilin isogenes was the best reference gene combination. Comparison between the qRT-PCR and in-house microarray data on roots demonstrated reliability of the identified reference genes to monitor the differential expression of drought-related candidate genes.
Specific isogenes from among the regularly used gene families were identified for use in qRT-PCR-based analyses for gene expression in studies on drought tolerance in rice. These were stable across variables of treatment, genotype, tissue and growth stage. A single gene and/or a three gene set analysis is recommended, based on the resources available.
KeywordsDrought Gene expression qRT-PCR Reference gene Rice Transcriptome
Real-time quantitative reverse transcription PCR (qRT-PCR) is widely used to assess the status of gene expression under variable developmental and environmental conditions (Gachon et al. 2004). Variables for transgene expression (Kohli et al. 2006) also necessitate its analysis by qRT-PCR (Trijatmiko et al. 2016). For the results to be dependable, selection of reference genes that express consistently across the tissue, genotype and developmental stage variables is critical (Wong and Medrano 2005; Guénin et al. 2009). Unless a stably expressed gene is used as a reference, expression level in samples are overestimated or underestimated, and identification of differential expressions is prone to errors. In general, ubiquitously expressed genes that would maintain consistent expression are used. However, these have not been thoroughly tested to be so under different variables. Also, the reference genes generally belong to a gene family and isogenes can be rather variably expressed in different tissues of different genotypes.
In rice, several housekeeping genes have been examined on various samples under different conditions but mostly at seedling stage (Kim et al. 2003; Jain et al. 2006; Narsai et al. 2010; Moraes et al. 2015). Consistent reference genes for rice under drought stress at both, the seedling and reproductive stage are not known. Earlier studies with genes such as the actin, ubiquitin, ubiquitin-conjugating enzyme E2 and eukaryotic elongation factor 1α (Moumeni et al. 2011; Sharoni et al. 2012; Minh-Thu et al. 2013; Campo et al. 2014) suffer from lack of information on isogene analysis.
An accurate gene expression analysis is central to obtaining insights on adaptation mechanisms to drought stress. Plants regulate the expression of many genes to adapt to water deficit conditions (Shinozaki and Yamaguchi-Shinozaki 2007). A number of drought stress responsive genes were analyzed (Deyholos 2010; Alter et al. 2015), and genes contributing to drought tolerance were suggested (Bhatnagar-Mathur et al. 2008; Osakabe et al. 2014). Drought transcriptome studies in various plant species revealed a largely common response involving similar pathways and genes but again these were largely confined to studies on leaves at seedling stage drought in genotypes known as models for drought tolerance (Lenka et al. 2011; Minh-Thu et al. 2013).
We have redressed the situation by identifying reference genes for relative quantification of transcripts in rice under drought studies, taking into consideration tissues, genotypes and growth stages. Candidate reference genes were pre-screened from public transcriptome microarray datasets by a novel systematic strategy, and then experimental evaluation of the candidates was conducted using qRT-PCR and a rigorous statistical analysis. The best reference genes identified were validated comparatively between qRT-PCR and microarray data.
In principle, reference genes should be independent of any biological response to the treatment under study e.g. drought stress in our case. Good reference genes can be screened from large-scale transcriptome data but the effectiveness of this approach would depend on the depth and breadth of the transcriptome experiments, which almost never simultaneously addressed the variables of treatment, genotype, growth stage and tissues. Thus, in the publicly available data sets, each of which lack addressing one or more of these variables, we focused on the four gene families i.e. glyceraldehyde-3-phosphate dehydrogenase (GAPDH), actin, ubiquitin and cyclophilin, that have been recurrently used as reference genes in various organisms. Any additional potential candidate reference genes or gene families would have to be validated in various organisms. Hence, there was value in our conservative approach of considering isogene expression differences within and among the tried and tested four gene families, under the variables of treatment, genotype, tissue and growth stage. Moreover, since reference gene combinations have been recommended (Vandesompele et al. 2002), our effort was to come up with a combination of genes expressed stably under different variables to provide a strict and dependable reference parameter.
Expression stability of candidate reference genes
Mean SD of ∆Ct
Rank on expression stability
A ubiquitin isogene (LOC_Os03g13170) represented the highest stability as a single gene reference. Among the 2-gene combinations, the pair of a GAPDH (LOC_Os08g03290) and another ubiquitin isogene (LOC_Os08g19830) had the best expression stability. Among the 3-gene combinations, two cyclophilin genes (LOC_Os07g29390, LOC_Os06g49480) and yet another ubiquitin isogene (LOC_Os02g06640) showed the best expression stability. Finally, among the 4-gene combinations, the set of a cyclophilin (LOC_Os02g02890) the GAPDH (LOC_Os08g03290) and two ubiquitin isogenes (LOC_Os05g06770 and LOC_Os08g19830) was the most stable reference (Table 1). The reason why different ubiquitin isogenes were extracted in the above listed best reference genes could be explained by the idea that most ubiquitously expressed genes showed divergent expressions and a combination of multiple genes minimized such divergence.
A number of publications exist about particular genes, and various statistical treatments, which can lead to dependable qRT-PCR results (Kozera and Rapacz 2013). Statistical veracity in assessing gene expression by qRT-PCR is an important component of the technique (Vandesompele et al. 2002; Silver et al. 2006). These methods suggested the superiority of multiple reference genes-based analysis to improve qRT-PCR accuracy, and recently Campo et al. (2014) applied a combination of reference genes to monitor dependable expression of their target genes. Utilization of multiple reference genes is a way to relieve the effect of expression variability of single reference genes (Vandesompele et al. 2002). A single reference gene tends to show higher expression variability due to a number of biological factors, but additional reference genes, which are differentially dependent on those factors, could cancel such effects. In this study, we used different combinations of reference genes in order to determine the most reliable gene combination. Based on the mean SD of ∆Ct, the expression stability of the reference increased as more reference genes were used (Table 1). Notably, only the 2-gene and 4-gene combinations have common genes between the different numbers of reference genes used. On the other hand, some single reference genes showed low expression stability. These results indicated the necessity of appropriate reference gene selection for more accurate analysis of gene expression.
Several statistical algorithms have been applied to validate reference genes in plants (Gue´nin et al. 2009; Saha and Blumwald 2014). The algorithms can be classified into two categories: one evaluates absolute Ct values among reference genes such as CV and one-way ANOVA; the other measures pair wise variations of Ct values among genes such as geNorm and the ΔCt approach. The first strategy is robust when mRNA content per total RNA is identical among samples compared, but fails when it differs. The second strategy overcomes this shortcoming, although the estimated gene expression stabilities may include some errors (Guenin et al. 2009). Application of the ∆Ct approach pointed to LOC_Os03g13170 as the most stable reference gene, in accordance with the results from geNorm (data not shown). The reliability of the best reference gene was validated by comparison between the qRT-PCR and the microarray result (Fig. 4).
Jain et al. (2006) documented a ubiquitin (UBQ5) and eEF-1α as the best reference genes to be used. Moraes et al. (2015) reported another ubiquitin (UBQ10) as the best reference. In the present study, another ubiquitin (LOC_Os03g13170) was observed to be the best single reference. However, multiple reference genes showed higher expression stabilities and higher reliability. The best reference gene combinations found in this study, to our knowledge, are not reported yet. These combinations were discovered by exhausting all possible combinations of the candidate reference genes. The best single reference gene was not included in the most stable gene combinations (Table 1). As such, it should not be assumed that combining other genes to a relatively stable reference will automatically yield a better reference.
In conclusion, we identified reference genes for relative quantification of transcript across rice samples under drought conditions, according to their broad expression stability under different variables. Our study indicated that the 3-reference gene combination was the most reliable. Most importantly, some isogenes we identified as broadly stable were different from the ones commonly used as reference genes. This study provides a basis for quantifying gene expression with high accuracy, leading to identification of drought regulated genes, and further leading to better understanding of drought tolerance mechanisms in plants. The identified reference genes are useful to gain insights into regulation of candidate genes against drought stress in rice. In addition, we present an efficient method for screening reference genes and their combinations.
Ct, threshold cycle; CV, coefficient of variation; GAPDH, glyceraldehyde 3-phosphate dehydrogenase; mRNA, messenger ribonucleic acid; qRT-PCR, quantitative reverse transcription polymerase chain reaction; RNA, ribonucleic acid; RV, residual variance; SD, standard deviation; WRS, weighted rank score
We thank Mr. Francisco V. Gulay and Ms. Maria Elena Escosura for plant material preparation. We also thank Ms. Blesilda Enriquez for total RNA preparation. This study was partially supported by the ‘Rapid Mobilization of Alleles for Rice Cultivar Improvement in Sub-Saharan Africa’ project of the Bill and Melinda Gates Foundation and Generation Challenge Program (Project: GCP.3007.05) and from the German Federal Ministry for Economic Cooperation and Development (GIZ Project 10.7860.9-001.00).
IMP and JLT carried out the experimental procedures. NY designed the study. IMP and NY did the data analysis and prepared the manuscript. TL and MLR reviewed the manuscript and added inputs for improvement. AK conceived the idea of the study, organized inputs from all the authors and finalized the manuscript. All authors read and approved the final manuscript.
The authors declare that they have no competing interests.
Open AccessThis article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made.
- Alter S, Bader KC, Spannagi M, Wang Y, Bauer E, Schön CC, Mayer KF (2015) DroughtDB: an expert-curated compilation of plant drought stress genes and their homologs in nine species. Database. doi:10.1093/database/bav046Google Scholar
- Bhatnagar-Mathur P, Vadez V, Sharma KK (2008) Transgenic approaches for abiotic stress tolerance in plants: retrospect and prospects. Plant Cell Rep 27:411–424View ArticlePubMedGoogle Scholar
- Campo S, Baldrich P, Messeguer J, Lalanne E, Coca M, San Segundo B (2014) Overexpression of a calcium-dependent protein kinase confers salt and drought tolerance in rice by preventing membrane lipid peroxidation. Plant Physiol 165:688–704View ArticlePubMedPubMed CentralGoogle Scholar
- Deyholos MK (2010) Making the most of drought and salinity transcriptomics. Plant Cell Environ 33:648–654View ArticlePubMedGoogle Scholar
- Dixit S, Biswal AK, Min A, Henry A, Oane RH, Raorane ML, Longkumer T, Pabuayon IM, Mutte SK, Vardarajan AR, Miro B, Govindan G, Albano-Enriquez B, Pueffeld M, Sreenivasulu N, Slamet-Loedin I, Sundarvelpandian K, Tsai YC, Raghuvanshi S, Hsing YC, Kumar A, Kohli A (2015) Action of multiple intra-QTL genes concerted around a co-localized transcription factor underpins a large effect QTL. Sci Rep 5:15183View ArticlePubMedPubMed CentralGoogle Scholar
- Gachon C, Mingam A, Charrier B (2004) Real-time PCR: what relevance to plant studies? J Ex Bot 55:1445–1454View ArticleGoogle Scholar
- Guénin S, Mauriat M, Pelloux J, Van Wuytswinkel O, Bellini C, Gutierrez L (2009) Normalization of qRT-PCR data: the necessity of adopting a systematic, experimental conditions-specific, validation of references. J Exp Bot 60:487–493View ArticlePubMedGoogle Scholar
- Jain M, Nijhawan A, Tyagi AK, Khurana JP (2006) Validation of housekeeping genes as internal control for studying gene expression in rice by quantitative real-time PCR. Biochem Biophys Res Comm 345:646–651View ArticlePubMedGoogle Scholar
- Kim BR, Nam HY, Kim SU, Kim SI, Chang YJ (2003) Normalization of reverse transcription quantitative-PCR with housekeeping genes in rice. Biotechnol Lett 25:1869–1872View ArticlePubMedGoogle Scholar
- Kohli A, González-Melendi P, Abranches R, Capell T, Stoger E, Christou P (2006) The quest to understand the basis and mechanisms that control expression of introduced transgenes in crop plants. Plant Signaling and Behavior 1:185–195View ArticlePubMedPubMed CentralGoogle Scholar
- Kozera B, Rapacz M (2013) Reference genes in real-time PCR. J Applied Genet 54:391–406View ArticleGoogle Scholar
- Lenka SK, Katiyar A, Chinnusamy V, Bansal KC (2011) Comparative analysis of drought-responsive transcriptome in Indica rice genotypes with contrasting drought tolerance. Plant Biotechnol J 9:315–327View ArticlePubMedGoogle Scholar
- Livak KJ, Schmittgen TD (2001) Analysis of relative gene expression data using real-time quantitative PCR and the 2−ΔΔC T method. Methods 25:402–408View ArticlePubMedGoogle Scholar
- Minh-Thu PT, Hwang DJ, Jeon JS, Nahm BH, Kim YK (2013) Transcriptome analysis of leaf and root of rice seedling to acute dehydration. Rice 6:38View ArticlePubMedPubMed CentralGoogle Scholar
- Moraes GP, Benitez LC, do Amaral MN, Vighi IL, Auler PA, da Maia LC, Bianchi VJ, Braga EJB (2015) Evaluation of reference genes for RT-qPCR studiesin the leaves of rice seedlings under salt stress. Genet Mol Res 14:2384–2398View ArticlePubMedGoogle Scholar
- Moumeni A, Satoh K, Kondoh H, Asano T, Hosaka A, Venuprasad R, Serraj R, Kumar A, Leung H, Kikuchi S (2011) Comparative analysis of root transcriptome profiles of two pairs of drought-tolerant and susceptible rice near-isogenic lines under different drought stress. BMC Plant Biol 11:174View ArticlePubMedPubMed CentralGoogle Scholar
- Narsai R, Ivanova A, Ng S, Whelan J (2010) Defining reference genes un Oryza sativa using organ, development, biotic and abiotic transcriptome datasets. BMC Plant Biol 10:56View ArticlePubMedPubMed CentralGoogle Scholar
- Osakabe Y, Osakabe K, Shinozaki K, Tran LS (2014) Response of plants to water stress. Front Plant Sci 5:86View ArticlePubMedPubMed CentralGoogle Scholar
- Saha P, Blumwald E (2014) Assessing reference genes for accurate transcript normalization using quantitative real-time PCR in pearl millet [Pennisetum glaucum (L.) R. Br]. PLoS One 9:e106308View ArticlePubMedPubMed CentralGoogle Scholar
- Sharoni AM, Nuruzzaman M, Satoh K, Moumeni A, Attia K, Venuprasad R, Serraj R, Kumar A, Leung H, Islam AK, Kikuchi S (2012) Comparative transcriptome analysis of AP2/EREBP gene family under normal and hormone treatments, and under two drought stresses in NILs setup by Aday Selection and IR64. Mol Genet Genomics 287:1–19View ArticlePubMedGoogle Scholar
- Shinozaki K, Yamaguchi-Shinozaki K (2007) Gene networks involved in drought stress response and tolerance. J Exp Bot 58:221–227View ArticlePubMedGoogle Scholar
- Silver N, Best S, Jiang J, Thein SL (2006) Selection of housekeeping genes for gene expression studies in human reticulocytes using real-time PCR. BMC Mol Biol 7:33View ArticlePubMedPubMed CentralGoogle Scholar
- Trijatmiko KR, Arines FM, Oliva N, Slamet-Loedin IH, Kohli A (2016) Molecular Analyses of Transgenic Plants. Meth Mol Biol. 1385:201–222.View ArticleGoogle Scholar
- Vandesompele J, De Preter K, Pattyn F, Poppe B, Van Roy N, De Paepe A, Speleman F (2002) Accurate normalization of real-time quantitative RT-PCR data by geometric averaging of multiple internal control genes. Genome Biol 3:research0034View ArticlePubMedPubMed CentralGoogle Scholar
- Wong ML, Medrano JF (2005) Real-time PCR for mRNA quantitation. Biotechniques 39:75View ArticlePubMedGoogle Scholar