- Email: [email protected]
Validation of reference genes for quantitative real-time PCR during latex regeneration in rubber tree
GENE-40356; No. of pages: 6; 4C: Gene xxx (2015) xxx–xxx
Contents lists available at ScienceDirect
Gene journal homepage: www.elsevier.com/locate/gene
Validation of reference genes for quantitative real-time PCR during latex regeneration in rubber tree Xiangyu Long a,1, Bin He a,b,1, Xinsheng Gao a, Yunxia Qin a, Jianghua Yang a, Yongjun Fang a, Jiyan Qi a, Chaorong Tang a,⁎ a Key Laboratory of Biology and Genetic Resources of Rubber Tree, Ministry of Agriculture, Rubber Research Institute, Chinese Academy of Tropical Agricultural Sciences, Danzhou, Hainan 571737, PR China b College of Agronomy, Hainan University, Haikou, Hainan 570228, PR China
a r t i c l e
i n f o
Article history: Received 12 December 2014 Received in revised form 26 February 2015 Accepted 13 March 2015 Available online xxxx Keywords: Hevea brasiliensis Reference genes Latex regeneration Algorithms Quantitative real-time PCR
a b s t r a c t In rubber tree, latex regeneration is one of the decisive factors influencing the rubber yield, although its molecular regulation is not well known. Quantitative real-time PCR (qPCR) is a popular and powerful tool used to understand the molecular mechanisms of latex regeneration. However, the suitable reference genes required for qPCR are not available to investigate the expressions of target genes during latex regeneration. In this study, 20 candidate reference genes were selected and evaluated for their expression stability across the samples during the process of latex regeneration. All reference genes showed a relatively wide range of the threshold cycle values, and their stability was validated by four different algorithms (comparative delta Ct method, Bestkeeper, NormFinder and GeNorm). Three softwares (comparative delta Ct method, NormFinder and GeNorm) exported similar results that identify UBC4, ADF, UBC2a, eIF2 and ADF4 as the top five suitable references, and 18S as the least suitable one. The application of the screened references would improve accuracy and reliability of gene expression analysis in latex regeneration experiments. © 2015 Elsevier B.V. All rights reserved.
1. Introduction In biological research, the technology of gene expression is broadly applied to understand the biological roles and interrelation of genes in molecular pathways. At transcriptional level, several methods provide high sensitivity and accuracy in the quantification of gene expression, such as transcriptome sequencing, cDNA microarray, Northern blotting and quantitative real-time PCR (qPCR) technology (Kubista et al., 2006; Josefsen and Nielsen, 2011; Lang et al., 2014; Le et al., 2014). Owing to technical ease, low reagent cost, less hand-on time and high throughput, qPCR is increasingly and widely used to measure the expression of target genes across different samples (Kubista et al., 2006). Abbreviations: qRT-PCR, Quantitative real-time PCR; Ct, threshold cycle (previously); Cq, quantification cycle; HKGs, Housekeeping genes; SD, standard deviation; CV, coefficient of variation; 18S, 18s ribosomal RNA; ACT7a, Actin (ACTIN7); ACT7b, Actin (ACTIN7); ADF, Actin depolymerizing factor; ADF4, Actin depolymerizing factor 4; eIF1Aa, Eukaryotic translation initiation factor 1A; eIF1Ab, Eukaryotic translation initiation factor 1A; eIF2, Eukaryotic translation initiation factor; eIF3, Eukaryotic translation initiation factor; FP, F-box family protein; PTP, Trosine phosphatase; RH2a, DEAD box RNA helicase,RH2; RH2b, DEAD box RNA helicase,RH2; ROC3, Cytosolic cyclophilin (ROC3); UBC1, Ubiquitin-protein ligase; UBC2a, Ubiquitin-protein ligase (ATUBC2); UBC2b, Ubiquitin-protein ligase (ATUBC2); UBC3, Ubiquitin-protein ligase; UBC4, Ubiquitin-protein ligase; YLS8, Mitosis protein YLS8. ⁎ Corresponding author. E-mail address: [email protected] (C. Tang). 1 These authors contributed equally to this work.
Unfortunately, there are several factors which can affect the quantitative measurement of gene expression by qPCR, including initial sample amount, RNA recovery, RNA integrity and efficiency of cDNA synthesis. To achieve accurate and stable results, it is essential that one or several reference genes should be used as internal control to normalize variations (Vandesompele et al., 2002; Andersen et al., 2004; Huggett et al., 2005). Theoretically, an ideal reference gene is stably expressed in various samples across different experimental conditions or treatments. Housekeeping genes (HKGs) have been historically used as reference genes for normalization, relating to basal cell activities and cellular structure components (Thellin et al., 1999; Daud and Scott, 2008; Frericks and Esser, 2008). Several housekeeping genes, including 18S or 26S ribosomal RNA (18S or 26S rRNA), glyceraldehyde-3-phosphate dehydrogenase (GAPDH), α or β-actin (ACT), β or γ-tubulin (TUB), ubiquitin C (UBC) and elongation factor-1 alpha (EF-1α) have been validated as suitable reference genes for qPCR analysis in the past (Bustin, 2000; Goidin et al., 2001; Kim and Kim, 2003; Lossos et al., 2003; Mitter et al., 2009). However, no genes are universally stable across different plant species and differing experimental conditions. More than chance, the usage of the reported suitable reference genes in a different species or an altered condition leads to misleading results. Recently, more candidate reference genes are isolated and identified using gene expression profile data in many plants, including Saccharum (Ling et al., 2014), Elaeis guineensis (Chan et al., 2014), Tectona grandis L.f. (Galeano et al.,
http://dx.doi.org/10.1016/j.gene.2015.03.026 0378-1119/© 2015 Elsevier B.V. All rights reserved.
Please cite this article as: Long, X., et al., Validation of reference genes for quantitative real-time PCR during latex regeneration in rubber tree, Gene (2015), http://dx.doi.org/10.1016/j.gene.2015.03.026
2
X. Long et al. / Gene xxx (2015) xxx–xxx
2014), Iris lactea var. chinensis (Gu et al., 2014), Triticum aestivum L. (Long et al., 2010). To improve the results accuracy, before performing qPCR analyses in a new experimental system, it is vital to evaluate multiple reference genes and utilize the most suitable one(s) to quantify gene expression. Four systematic and statistical algorithms, comparative delta Ct method (Silver et al., 2006), Bestkeeper (Pfaffl et al., 2004), NormFinder (Andersen et al., 2004) and GeNorm (Vandesompele et al., 2002) have been developed to assess and identify the most suitable reference genes for qPCR data normalization. Natural rubber, cis-polyisoprene, is an essential industrial substance and strategic material, and is mainly acquired from Hevea brasiliensis, the Pará rubber tree. Natural rubber biosynthesis occurs in the cytoplasm (latex) of laticifers, which are a strong sink and highly specialized cells in the phloem (Metcalfe, 1967; Lewinsohn, 1991). In production, huge amount of latex (30–40% of its dry matter as natural rubber) is harvested by tapping for every two or three days. So, latex regeneration between two consecutive tappings is one of the major limiting factors determining rubber yield (Kongsawadworakul et al., 2009). And the earlier studies found that physiologically, latex regeneration mainly depends on the availability and metabolism of sugar and nitrogen compounds (Tupy, 1973; Pujade-Renaud et al., 1994). At present, the development of molecular biology makes the studies of gene expression and regulation related to natural rubber biosynthesis become a hotspot in undressing the mechanisms of latex regeneration. For instance, several researchers have focused on the genes involved in the supply and utility of sucrose in the latex, using gene expression profile especially qPCR analysis (Dusotoit-Coucaud et al., 2010; Tang et al., 2010; Liu et al., 2015). To reveal the underlying molecular mechanism and regulatory network of latex regeneration, it is necessary and crucial to screen the suitable reference genes for normalization in relevant gene expression analysis. Although we previously characterized a total of 22 candidate genes for their suitability as reference genes in several experimental conditions in rubber tree (Li et al., 2011), the screening of the suitable references for studying latex regeneration has not been touched. In this study, 20 common reference genes were tested for their expression stability across the samples during the process of latex regeneration. Of the four softwares used for evaluating the suitability of reference genes, three (comparative delta Ct method, NormFinder and GeNorm) exported similar results showing UBC4, ADF, UBC2a, eIF2 and ADF4 as the suitable reference genes for normalization in latex regeneration samples. Evaluation and application of them would improve accuracy and reliability of gene expression analysis in latex regeneration experiments. 2. Materials and methods 2.1. Plant materials In this study, two clones of rubber tree, RRIM600 and CATAS628 were applied as experimental materials with different types of metabolism. Two types of tree were growing for 13 years at the experimental plantation of the Rubber Research Institute of the Chinese Academy of Tropical Agricultural Sciences (CATAS) (Danzhou, Hainan, China), and which had been regularly tapped for 6 years in a half spiral pattern, every three days (S/2, d/3). For each rubber clone, 30 trees with similar girth, growth vigor and rubber yield were selected and divided into 6 groups. All groups were tapped simultaneously at the first tapping, and then at 6 h, 12 h, 24 h, 48 h, 72 h and 96 h, respectively, one group for each time was tapped again. After every tapping, latex was collected for RNA extraction according to previous description by Tang (Tang et al., 2007, 2010).
integrity of the RNA samples was checked by agarose gel electrophoresis, and the concentration and quality were examined by NanoDrop 2000 (Thermo, USA) at 230 nm, 260 nm and 280 nm. Synthesis of cDNA was performed using the RevertAidTM First Strand cDNA Synthesis Kit (Fermentas, Canada) following the manufacturer's protocol. 2.3. qPCR A total of 20 candidate reference genes, 18S, ACT7a, ACT7b, ADF, ADF4, eIF2, eIF3, eIF1Aa, eIF1Ab, FP, PTP, RH2a, RH2b, ROC3, UBC1, UBC2a, UBC2b, UBC3, UBC4 and YLS8 were selected from our previous study, and their detail information refers to our previous study including gene annotation, primer sequence, and so on (Table 1S). The realtime PCR was carried out using the SYBR® Premix Ex Taq™ II (Perfect Real Time) (Takara, Dalian, China) and the ABI 7500 Fast Real Time PCR System (Applied Biosystems, Foster City, CA, USA). The PCR reaction system and procedures were described previously (Tang et al., 2007, 2010; Li et al., 2011). The ABI 7500 Software v2.0.6 was used for visualizing and analyzing the data, including the quantification cycle values, PCR efficiency and correlation coefficients. 2.4. Data analysis After collecting and converting the quantification cycle data (Cq), Cq average values were calculated to statistical analysis by SPSS 13 (http:// www.spss.com/). To obtain reliable results, comparative delta Ct method, Bestkeeper, NormFinder and GeNorm were used to analyze expression stability of reference genes, according to their instructions (RefFinder, http://www.leonxie.com/referencegene.php?type=reference). Pearson correlation coefficients were detected for ranking results from four different algorithms, using Minitab 15 software (http://www.minitab. com/). 3. Results 3.1. Expression profiles of candidate reference genes A total of 20 candidate reference genes were selected for determining the most stable one during the process of latex regeneration in the rubber varieties of RRIM600 and CATAS628. Amplification of each reference gene in 24 samples (three replicates per sample) produced 72 Cq values, and samples with missing Cq values or inconsistencies between replicates (Cq differences N0.5 cycle) were removed from the analysis. Based on the standard curves using a serial dilution of cDNA samples, the amplification efficiencies of those primers ranged from 94.18% to 102.28%, and the regression coefficient R2 for all primers varied between 0.997 and 1.000. Over all samples, the 20 candidate reference genes had a wide range of the Cq values, and the mean Cq values of those gene ranged from 15.60 to 23.07 across all the samples. Among these candidate reference genes, 18S was the most abundantly expressed gene (mean Cq ± SD = 15.60 ± 1.07) followed by eIF3 (mean Cq ± SD = 16.38 ± 0.72), whereas RH2a was the least abundantly expressed gene (mean Cq ± SD = 23.07 ± 1.04). There were small standard deviation and coefficient of variation in all samples. The eIF3 standard deviation (SD) was the lowest (0.72) while PTP presented the largest variation between Cq values (SD = 1.14). The coefficient of variation of UBC4 was the smallest (4.18%), while that of 18S was the largest (6.88%) (Table S1). Additionally, using individual value plot to evaluate and compare all samples, the results showed that those genes had similar distribution or trend except 18S and FP (Fig. 1).
2.2. RNA isolation and cDNA synthesis 3.2. Expression stability of the twenty candidate reference genes Total RNA was extracted by the protocol as described previously by Tang (Tang et al., 2007). RNA samples were treated with DNase I (TaKaRa) to eliminate the trace contaminants of genomic DNA. The
In order to make a more detailed expression analysis of the candidate reference genes, the 24 samples were divided into three
Please cite this article as: Long, X., et al., Validation of reference genes for quantitative real-time PCR during latex regeneration in rubber tree, Gene (2015), http://dx.doi.org/10.1016/j.gene.2015.03.026
X. Long et al. / Gene xxx (2015) xxx–xxx
3
(0.134) in the third set. Four candidate reference genes, ROC3, PTP, FP and 18S had the worst stability in the first, second and third sets. The ranks of candidate reference genes from the least to most stable expression were very similar between the first two sets, and the correlation coefficient value was 0.680 (p = 0.001). However, there were two genes that showed great fluctuation of expression stability, including YLS8 (5th in the first set, 19th in the second set) and UBC1 (14th in the first set, 5th in the second set). 3.4. NormFinder
Fig. 1. Expression levels of candidate reference genes tested using the qPCR cycle threshold values (Cq). The red-dot represents reference gene expression level (Cq) in each samples and the black-dot represents the mean expression level of reference genes (arithmetical mean of Cq). (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)
experiment sets. The first set consisted of 12 samples from RRIM600. The second set was comprised of 12 samples from CATAS628. In the third set, both the first and second sample sets were included. The four softwares using different algorithms, including comparative delta Ct method, Bestkeeper, NormFinder and GeNorm, were used to analyze and evaluate the stability for candidate reference genes in three experiment sets.
NormFinder program is used to calculate an arbitrary stability value and standard error, considering the intragroup and intergroup variations of each reference gene in all groups (Andersen et al., 2004). According to NormFinder results, the rank of stability in the first set was somewhat similar to that in the second set, and the correlation coefficient value was 0.439 (p = 0.053) (Table 2). Four reference genes, YLS8, UBC3, ADF4 and ADF, were rated as the most stable reference genes in the first set, with stability values of 0.057–0.083. However, both in the second and third sets, UBC4, ADF, eIF2 and UBC2b were ranked as the four most stable reference genes although ordering differed between the two sets. In any of the three sets, 18S was always determined as the least stable reference gene with stability values of 1.074–2.211, which were above the default limit of 1.0 in the NormFinder program. Additionally, YLS8 showed large fluctuations of expression stability in the three sets, ranking 1st in the first set, 18th in the second set and 15th in the third set. 3.5. Delta Ct method
The program GeNorm was utilized to calculate the expression stability M based on the average pair-wise variation between all studied genes in the three experimental sets (the first, second, and third sets) (Table 1). All of the genes showed high expression stability with M values of less than 0.600 (0.593 in the first set, 0.419 in the second set and 0.548 in the third set), which were below the default limit of 1.5 in the GeNorm program. From the results of GeNorm, the top ranked genes were ADF4 = UBC2b (0.099) N UBC3 (0.117) N UBC4 (0.126) in the first set, ADF4 = UBC2a (0.107) N UBC3 (0.120) N UBC4 (0.132) in the second set, and UBC2a = UBC4 (0.124) N ADF4 (0.128) N UBC3
The comparative delta-Ct method calculates the standard deviation of Cq differences within each sample for each pairwise comparison with the other genes to define the stable reference genes (Silver et al., 2006), and which is similar to the ΔΔCt method for real-time quantitative PCR (Livak and Schmittgen, 2001). A correlation value of ranking between the first and second sets was 0.549 (p = 0.012), indicating a similar result obtained for the 20 reference genes evaluated (Table 3). The comparative delta-Ct results showed that ADF4 was identified as the most stable reference gene with stability value of 0.376 in the first set, UBC4 with stability value of 0.297 in the second set and UBC4 with stability value of 0.373 in the third set. Interestingly, consistent results were obtained in all three sets that 18S was the least stable gene, with stability value of 2.244 in the first set, 1.102 in the second set and
Table 1 Expression stability of 20 candidate reference genes as calculated by GeNorm.
Table 2 Expression stability of 20 candidate reference genes as calculated by NormFinder.
3.3. GeNorm
Rank
1 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
The first set
The second set
The third set
Rank
Gene
M values
Gene
M values
Gene
M values
ADF4 | UBC2b UBC3 UBC4 YLS8 ADF UBC2a eIF2 eIF1Aa RH2b RH2a ACT7b eIF1Ab UBC1 ACT7a eIF3 ROC3 PTP FP 18S
0.099 0.117 0.126 0.131 0.136 0.140 0.145 0.152 0.158 0.166 0.173 0.181 0.189 0.197 0.224 0.251 0.277 0.410 0.593
ADF4 | UBC2a UBC3 UBC4 UBC1 ADF eIF1Aa eIF2 UBC2b RH2b eIF3 ACT7a eIF1Ab RH2a ACT7b PTP FP ROC3 YLS8 18S
0.107 0.120 0.132 0.148 0.158 0.173 0.181 0.188 0.195 0.204 0.216 0.226 0.243 0.258 0.274 0.292 0.318 0.343 0.419
UBC2a | UBC4 ADF4 UBC3 ADF eIF2 UBC2b RH2b UBC1 eIF1Aa ACT7a RH2a ACT7b eIF1Ab eIF3 YLS8 PTP ROC3 FP 18S
0.124 0.128 0.134 0.154 0.167 0.175 0.183 0.193 0.203 0.214 0.224 0.237 0.252 0.267 0.284 0.303 0.333 0.416 0.548
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
The first set
The second set
The third set
Gene
Stability
Gene
Stability
Gene
Stability
YLS8 UBC3 ADF4 ADF UBC4 UBC2a eIF1Aa UBC2b eIF2 UBC1 eIF1Ab ACT7a RH2b RH2a ACT7b eIF3 ROC3 PTP FP 18S
0.057 0.062 0.068 0.083 0.086 0.124 0.124 0.125 0.128 0.142 0.165 0.199 0.203 0.229 0.264 0.312 0.407 0.528 1.523 2.211
UBC4 ADF eIF2 UBC2b eIF1Aa UBC2a ADF4 RH2b UBC3 UBC1 eIF3 ACT7a eIF1Ab RH2a ACT7b FP PTP YLS8 ROC3 18S
0.065 0.067 0.107 0.117 0.125 0.133 0.162 0.162 0.169 0.177 0.181 0.215 0.226 0.325 0.359 0.370 0.416 0.483 0.487 1.074
ADF UBC4 eIF2 UBC2b UBC3 ADF4 UBC1 UBC2a RH2b eIF1Aa ACT7a eIF3 eIF1Ab RH2a YLS8 ACT7b PTP ROC3 FP 18S
0.073 0.081 0.113 0.122 0.129 0.143 0.155 0.160 0.179 0.179 0.205 0.264 0.270 0.279 0.330 0.342 0.464 0.526 1.086 1.710
Please cite this article as: Long, X., et al., Validation of reference genes for quantitative real-time PCR during latex regeneration in rubber tree, Gene (2015), http://dx.doi.org/10.1016/j.gene.2015.03.026
4
X. Long et al. / Gene xxx (2015) xxx–xxx
Table 3 Expression stability of 20 candidate reference genes as calculated by comparative delta Ct method. Rank
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
The first set
The second set
The third set
Gene
SD
Gene
SD
Gene
SD
ADF4 UBC2b UBC4 ADF YLS8 eIF1Aa UBC3 RH2b UBC2a RH2a eIF2 eIF1Ab UBC1 ACT7b ACT7a eIF3 ROC3 PTP FP 18S
0.376 0.390 0.391 0.391 0.391 0.397 0.400 0.410 0.414 0.424 0.425 0.431 0.439 0.443 0.456 0.582 0.616 0.651 1.598 2.244
UBC4 ADF UBC2a eIF1Aa UBC2b ADF4 eIF2 RH2b UBC3 UBC1 eIF3 ACT7a eIF1Ab RH2a ACT7b FP PTP ROC3 YLS8 18S
0.297 0.300 0.315 0.320 0.322 0.324 0.325 0.333 0.335 0.340 0.359 0.369 0.374 0.415 0.439 0.476 0.485 0.565 0.579 1.102
UBC4 ADF ADF4 UBC2b UBC3 RH2b UBC2a eIF2 UBC1 eIF1Aa ACT7a RH2a eIF1Ab ACT7b eIF3 YLS8 PTP ROC3 FP 18S
0.373 0.376 0.384 0.386 0.394 0.399 0.401 0.402 0.411 0.416 0.434 0.445 0.474 0.490 0.501 0.528 0.581 0.667 1.163 1.743
values exceed 1.0, reference genes are considered as unstable ones and should be avoided for gene expression normalization. Nine genes in the first set, two genes in the second set and five genes in the third set had SD values of larger than 1.0 (1.004–1.402) and CV values of 4.37–8.14. 3.7. Comprehensive analysis of expression stability in the third set
1.743 in the third set. Again, YLS8 exhibited a big change for stable ranking as 1st in the first set but 18th in the second set, consistent with the outputs from GeNorm and NormFinder.
For a comprehensive judgment of the suitable reference genes in the third set, the Pearson correlations were calculated using the ranks from the most stable to the least stable among the four methods (comparative delta Ct method, Bestkeeper, NormFinder and GeNorm) used in this study (Fig. 2). The Pearson correlations for the four stability tests showed a significant or extremely significant positive correlation. Three high Pearson correlations were observed between the outcomes from delta Ct method and GeNorm (r = 0.950, p = 0.000), delta Ct method and NormFinder (r = 0.949, p = 0.000), and NormFinder and GeNorm (r = 0.908, p = 0.000), indicating that ranking results from the above three methods (delta Ct, GeNorm and NormFinder) were nearly identical in the third set. However, Bestkeeper had a relatively lower correlation with the other three methods, showing a correlation efficient of 0.516 with delta Ct method, 0.559 with GeNorm and 0.675 with NormFinder. According to the four methods in the third set, five reference genes (UBC4, ADF, UBC2a, eIF2 and ADF4) were rank as the most stable expression ones, whereas five reference genes, 18S, FP, PTP, ROC3 and ACT7b were considered as the least stable expression ones in the third set (Tables 1, 2 and 3).
3.6. Bestkeeper
4. Discussion
BestKeeper ranks the reference genes according to the standard deviation (SD) of their Cqs, and the output includes more information, for example the coefficient of variation (CV) and correlation (r) (Pfaffl et al., 2004). There was a large difference in reference gene stability between the first and second sets, and the coefficient of correlation was only 0.018 (p = 0.940) for ranking results (Table 4). The stable reference genes are those with low coefficient of variance and standard deviation, and high coefficient of correlation. eIF3 was ranked as the most stable expression genes with the CV ± SD value of 4.34 ± 0.71 and correlation value of 0.979 in the first set, and the CV ± SD value of 4.39 ± 0.720 and correlation value of 0.983 in the third set. In the second set, eIF1Ab had low CV ± SD value of 3.38 ± 0.706 and high correlation value of 0.985, showing the most stable expression pattern. Additionally, if the SD
In rubber tree, as one of the decisive factors that influence the rubber yield, latex regeneration is an extremely complex metabolic process including biosynthesis of nature rubber, supply and utility of sucrose, remodeling of organelles and recovering of other compounds (Tupy, 1973; Kongsawadworakul et al., 2009). Presently, gene expression and regulation have been widely used to understand the molecular mechanisms of latex regeneration from one or another aspect (Dusotoit-Coucaud et al., 2010; Tang et al., 2010). The qPCR has routinely become the main technique to conduct gene expression analysis. To get reliable gene expression results for the examined genes during latex regeneration, it is important that more ideal reference genes are screened for normalization of gene expression. In previous research, 18S and ACTIN genes have been the most widely used reference genes for expression normalization in rubber tree (Dusotoit-Coucaud et al., 2009; Duan et al., 2010; Li et al., 2010; Tang et al., 2010). However, extensive analysis is not conducted to validate their suitability as reference genes in those studies. In our previous study, in order to screen more suitable reference genes, 22 candidate genes were evaluated for expression stability in various tissues, different hormone treatments, tappings, individual trees and genotypes. The results showed that those candidate reference genes display distinct expression stability in different experimental conditions, and UBC2a and UBC4 are considered as the suitable reference genes in all conditions taken together (Li et al., 2011). In this study, we analyzed and compared 20 candidate genes for expression stability across 24 samples during the process of latex regeneration in rubber varieties of RRIM600 and CATAS628. Five genes, UBC4, ADF, UBC2a, eIF2 and ADF4 were considered as the top five suitable reference genes to normalize all the samples, especially UBC4 showing the most stable expression when the results of the four methods are taken together (Tables 1, 2 and 3). Ubiquitin-protein ligase genes (UBC4 and UBC2a) are frequently used as appropriate reference gene in other plants, including Arabidopsis thaliana, Brachypodium distachyon and Brachiaria brizantha (Czechowski et al., 2005; Hong et al., 2008; Silveira et al., 2009). Another gene family, actin-depolymerizing factor 4 (ADF and ADF4), which plays an important role in remodeling the actin cytoskeleton (Xu
Table 4 Expression stability of 20 candidate reference genes as calculated by Bestkeeper. Rank
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
The first set
The second set
The third set
Gene
SD
Gene
SD
Gene
SD
eIF3 eIF2 UBC2a ACT7a YLS8 UBC4 UBC3 UBC2b ADF4 ADF ROC3 eIF1Aa RH2b eIF1Ab UBC1 RH2a ACT7b PTP 18S FP
0.712 0.872 0.897 0.919 0.939 0.953 0.954 0.990 0.993 0.996 0.998 1.011 1.012 1.028 1.033 1.064 1.073 1.272 1.279 1.402
eIF1Ab ROC3 eIF3 eIF1Aa FP eIF2 ADF UBC4 ACT7a 18S UBC2b UBC3 RH2b UBC1 UBC2a YLS8 ADF4 ACT7b PTP RH2a
0.706 0.712 0.728 0.752 0.756 0.782 0.793 0.813 0.824 0.830 0.858 0.861 0.868 0.869 0.871 0.883 0.906 0.974 1.004 1.010
eIF3 eIF2 ACT7a UBC4 UBC2a ADF eIF1Aa UBC3 YLS8 UBC2b eIF1Ab RH2b ADF4 UBC1 ROC3 ACT7b RH2a 18S FP PTP
0.720 0.827 0.876 0.883 0.886 0.901 0.902 0.907 0.913 0.932 0.943 0.947 0.949 0.951 0.963 1.029 1.037 1.073 1.104 1.138
Please cite this article as: Long, X., et al., Validation of reference genes for quantitative real-time PCR during latex regeneration in rubber tree, Gene (2015), http://dx.doi.org/10.1016/j.gene.2015.03.026
X. Long et al. / Gene xxx (2015) xxx–xxx
5
Fig. 2. Comparison of the ranking results from comparative delta Ct method, Bestkeeper, NormFinder and GeNorm. The correlation was evaluated for the ranking results of 20 candidate reference genes in all samples, by comparative delta Ct method, Bestkeeper, NormFinder and GeNorm. Correlation coefficient (r) values are shown (*p b 0.05, **p b 0.01).
et al., 2008), also has relatively stable expression in latex generation. Similar to the previous findings in other plants, 18S had very poor expression stability across latex regeneration samples in rubber tree. Some researchers reported that the 18S and 26S were not good choice as reference genes (Sturzenbaum and Kille, 2001; Zhu and Dong, 2006), because the rRNA is much higher expressed compared with the mRNA and easily fluctuates with various biological and drug factors. Additionally, the rankings for 20 candidate reference genes differed between the varieties of RRIM600 and CATAS628, especially for several ones. For example, the mitosis protein gene, YLS8 performed quite stable expression in RRIM600, which ranked first as analyzed in two methods (Tables 2 and 3) and fifth in the other two methods (Tables 1 and 4). In contrast, the same gene was among the five least stable reference genes as analyzed in CATAS628 with any of the four methods (Tables 1–4). However, YLS8 was a novel suitable reference in A. thaliana (Remans et al., 2008), and considered as the most suitable reference gene for hormone and tapping treatments in rubber tree (Remans et al., 2008; Li et al., 2011). The four common methods, comparative delta Ct method, Bestkeeper, NormFinder and GeNorm are often used to check the stability of reference genes. In the past research, the results from comparative delta Ct method, NormFinder and GeNorm have been reported to be more consistent among them than those of Bestkeeper (Zhang et al., 2012; Robledo et al., 2014). Similar results were obtained in this study, which showed a high correlation among comparative delta Ct method, NormFinder and GeNorm, especially between comparative delta Ct method and GeNorm (Fig. 2). Of course, each type of softwares has its own risks and benefits according to different algorithms. Two softwares, GeNorm and comparative delta Ct method take a similar algorithm approach to rank reference genes from the most to least stable expression, using pairwise comparison for standard deviation between one and all other ones (Vandesompele et al., 2002; Silver et al., 2006). However, co-regulation of reference genes that have similar function or basic roles in biological processes could influence the validity of the two methods (Long et al., 2010). NormFinder software could be used to control the influence of co-regulation, taking into account intra-group and inter-group variations to evaluate the expression stability (Andersen et al., 2004). So, NormFinder and GeNorm are considered as the useful and common softwares for selecting the suitable reference genes when combined together. The Bestkeeper directly calculates the standard deviation to measure the
variation, outputting the stability of reference genes. However, there is obvious problem that the gene with low standard deviation may not be considered as suitable reference gene, because the cDNA samples could not be controlled at completely consentient level produced by RNA extraction and reverse transcription steps (Pfaffl et al., 2004). Our results here again reinforce the idea that no reference genes are absolutely stable, because their expressions may be irregular and unsteady in some experiments. So, to obtain reliable expression results, it is absolutely necessary that the expression stability should be evaluated to select the suitable reference gene using different softwares. It is better to chose more than one reference gene for normalization, and these genes should be involved in different biological functions and pathways. In brief, we conclude that UBC4, ADF, UBC2a, eIF2 and ADF4 are the first set of reference genes selected to normalize gene expressions in different latex regeneration samples, by applying comparative delta Ct method, NormFinder and GeNorm together. Our results will be of great benefit to the gene expression-based studies for the investigation of the molecular regulation of latex regeneration in rubber tree. Acknowledgments This research was supported by the 863 program (2013AA102605), the National Natural Science Foundation of China (31300570) and the Natural Science Foundation of Hainan province (312029). Appendix A. Supplementary data Supplementary data to this article can be found online at http://dx. doi.org/10.1016/j.gene.2015.03.026. References Andersen, C.L., Jensen, J.L., Orntoft, T.F., 2004. Normalization of real-time quantitative reverse transcription-PCR data: a model-based variance estimation approach to identify genes suited for normalization, applied to bladder and colon cancer data sets. Cancer Res. 64, 5245–5250. Bustin, S.A., 2000. Absolute quantification of mRNA using real-time reverse transcription polymerase chain reaction assays. J. Mol. Endocrinol. 25, 169–193. Chan, P.L., Rose, R.J., Abdul, M.A., Zainal, Z., Low, E.T., Ooi, L.C., Ooi, S.E., Yahya, S., Singh, R., 2014. Evaluation of reference genes for quantitative real-time PCR in oil palm elite planting materials propagated by tissue culture. PLoS One 9, e99774.
Please cite this article as: Long, X., et al., Validation of reference genes for quantitative real-time PCR during latex regeneration in rubber tree, Gene (2015), http://dx.doi.org/10.1016/j.gene.2015.03.026
6
X. Long et al. / Gene xxx (2015) xxx–xxx
Czechowski, T., Stitt, M., Altmann, T., Udvardi, M.K., Scheible, W.R., 2005. Genome-wide identification and testing of superior reference genes for transcript normalization in Arabidopsis. Plant Physiol. 139, 5–17. Daud, I.I., Scott, M.E., 2008. Validation of reference genes in cervical cell samples from human papillomavirus-infected and -uninfected women for quantitative reverse transcription-PCR assays. Clin. Vaccine Immunol. 15, 1369–1373. Duan, C., Rio, M., Leclercq, J., Bonnot, F., Oliver, G., Montoro, P., 2010. Gene expression pattern in response to wounding, methyl jasmonate and ethylene in the bark of Hevea brasiliensis. Tree Physiol. 30, 1349–1359. Dusotoit-Coucaud, A., Brunel, N., Kongsawadworakul, P., Viboonjun, U., Lacointe, A., Julien, J.L., Chrestin, H., Sakr, S., 2009. Sucrose importation into laticifers of Hevea brasiliensis, in relation to ethylene stimulation of latex production. Ann. Bot. 104, 635–647. Dusotoit-Coucaud, A., Kongsawadworakul, P., Maurousset, L., Viboonjun, U., Brunel, N., Pujade-Renaud, V., Chrestin, H., Sakr, S., 2010. Ethylene stimulation of latex yield depends on the expression of a sucrose transporter (HbSUT1B) in rubber tree (Hevea brasiliensis). Tree Physiol. 30, 1586–1598. Frericks, M., Esser, C., 2008. A toolbox of novel murine house-keeping genes identified by meta-analysis of large scale gene expression profiles. Biochim. Biophys. Acta 1779, 830–837. Galeano, E., Vasconcelos, T.S., Ramiro, D.A., De Martin, V.F., Carrer, H., 2014. Identification and validation of quantitative real-time reverse transcription PCR reference genes for gene expression analysis in teak (Tectona grandis L.f.). BMC Res. Notes 7, 464. Goidin, D., Mamessier, A., Staquet, M.J., Schmitt, D., Berthier-Vergnes, O., 2001. Ribosomal 18S RNA prevails over glyceraldehyde-3-phosphate dehydrogenase and beta-actin genes as internal standard for quantitative comparison of mRNA levels in invasive and noninvasive human melanoma cell subpopulations. Anal. Biochem. 295, 17–21. Gu, C.S., Liu, L.Q., Xu, C., Zhao, Y.H., Zhu, X.D., Huang, S.Z., 2014. Reference gene selection for quantitative real-time RT-PCR normalization in Iris lactea var. chinensis roots under cadmium, lead, and salt stress conditions. ScientificWorldJournal 2014, 532713. Hong, S.Y., Seo, P.J., Yang, M.S., Xiang, F., Park, C.M., 2008. Exploring valid reference genes for gene expression studies in Brachypodium distachyon by real-time PCR. BMC Plant Biol. 8, 112. Huggett, J., Dheda, K., Bustin, S., Zumla, A., 2005. Real-time RT-PCR normalisation; strategies and considerations. Genes Immun. 6, 279–284. Josefsen, K., Nielsen, H., 2011. Northern blotting analysis. Methods Mol. Biol. 703, 87–105. Kim, S., Kim, T., 2003. Selection of optimal internal controls for gene expression profiling of liver disease. Biotechniques 35, 456–458. Kongsawadworakul, P., Viboonjun, U., Romruensukharom, P., Chantuma, P., Ruderman, S., Chrestin, H., 2009. The leaf, inner bark and latex cyanide potential of Hevea brasiliensis: evidence for involvement of cyanogenic glucosides in rubber yield. Phytochemistry 70, 730–739. Kubista, M., Andrade, J.M., Bengtsson, M., Forootan, A., Jonak, J., Lind, K., Sindelka, R., Sjoback, R., Sjogreen, B., Strombom, L., Stahlberg, A., Zoric, N., 2006. The real-time polymerase chain reaction. Mol. Aspects Med. 27, 95–125. Lang, V., Usadel, B., Obermeyer, G., 2014. De novo sequencing and analysis of the lily pollen transcriptome: an open access data source for an orphan plant species. Plant Mol. Biol. 87, 69–80. Le, M.Q., Pagter, M., Hincha, D.K., 2014. Global changes in gene expression, assayed by microarray hybridization and quantitative RT-PCR, during acclimation of three Arabidopsis thaliana accessions to sub-zero temperatures after cold acclimation. Plant Mol. Biol. 87, 1–15. Lewinsohn, T., 1991. The geographical distribution of plant latex. Chemoecology 2, 64–68. Li, D., Deng, Z., Chen, C., Xia, Z., Wu, M., He, P., Chen, S., 2010. Identification and characterization of genes associated with tapping panel dryness from Hevea brasiliensis latex using suppression subtractive hybridization. BMC Plant Biol. 10, 140. Li, H., Qin, Y., Xiao, X., Tang, C., 2011. Screening of valid reference genes for real-time RTPCR data normalization in Hevea brasiliensis and expression validation of a sucrose transporter gene HbSUT3. Plant Sci. 181, 132–139. Ling, H., Wu, Q., Guo, J., Xu, L., Que, Y., 2014. Comprehensive selection of reference genes for gene expression normalization in sugarcane by real time quantitative RT-PCR. PLoS One 9, e97469.
Liu, S.J., Lan, J.X., Zhou, B.H., Qin, Y.X., Zhou, Y.H., Xiao, X.H., Yang, J.H., Gou, J.Q., Qi, J.Y., Huang, Y.C., Tang, C.R., 2015. HbNIN2, a cytosolic alkaline/neutral-invertase, is responsible for sucrose catabolism in rubber-producing laticifers of Hevea brasiliensis (para rubber tree). New Phytol. 206 (In press). Livak, K.J., Schmittgen, T.D., 2001. Analysis of relative gene expression data using real-time quantitative PCR and the 2(-delta delta C(T)). Method 402–408. Long, X.Y., Wang, J.R., Ouellet, T., Rocheleau, H., Wei, Y.M., Pu, Z.E., Jiang, Q.T., Lan, X.J., Zheng, Y.L., 2010. Genome-wide identification and evaluation of novel internal control genes for Q-PCR based transcript normalization in wheat. Plant Mol. Biol. 74, 307–311. Lossos, I.S., Czerwinski, D.K., Wechser, M.A., Levy, R., 2003. Optimization of quantitative real-time RT-PCR parameters for the study of lymphoid malignancies. Leukemia 17, 789–795. Metcalfe, C.R., 1967. Distribution of latex in the plant kingdom. Econ. Bot. 21, 115–127. Mitter, K., Kotoulas, G., Magoulas, A., Mulero, V., Sepulcre, P., Figueras, A., Novoa, B., Sarropoulou, E., 2009. Evaluation of candidate reference genes for QPCR during ontogenesis and of immune-relevant tissues of European seabass (Dicentrarchus labrax). Comp. Biochem. Physiol. B Biochem. Mol. Biol. 153, 340–347. Pfaffl, M.W., Tichopad, A., Prgomet, C., Neuvians, T.P., 2004. Determination of stable housekeeping genes, differentially regulated target genes and sample integrity: BestKeeper–Excel-based tool using pair-wise correlations. Biotechnol. Lett. 26, 509–515. Pujade-Renaud, V., Clement, A., Perrot-Rechenmann, C., Prevot, J.C., Chrestin, H., Jacob, J.L., Guern, J., 1994. Ethylene-induced increase in glutamine synthetase activity and mRNA levels in Hevea brasiliensis latex cells. Plant Physiol. 105, 127–132. Remans, T., Smeets, K., Opdenakker, K., Mathijsen, D., Vangronsveld, J., Cuypers, A., 2008. Normalisation of real-time RT-PCR gene expression measurements in Arabidopsis thaliana exposed to increased metal concentrations. Planta 227, 1343–1349. Robledo, D., Hernandez-Urcera, J., Cal, R.M., Pardo, B.G., Sanchez, L., Martinez, P., Vinas, A., 2014. Analysis of qPCR reference gene stability determination methods and a practical approach for efficiency calculation on a turbot (Scophthalmus maximus) gonad dataset. BMC Genomics 15, 648. Silveira, E.D., Alves-Ferreira, M., Guimaraes, L.A., Da, S.F., Carneiro, V.T., 2009. Selection of reference genes for quantitative real-time PCR expression studies in the apomictic and sexual grass Brachiaria brizantha. BMC Plant Biol. 9, 84. Silver, N., Best, S., Jiang, J., Thein, S.L., 2006. Selection of housekeeping genes for gene expression studies in human reticulocytes using real-time PCR. BMC Mol. Biol. 7, 33. Sturzenbaum, S.R., Kille, P., 2001. Control genes in quantitative molecular biological techniques: the variability of invariance. Comp. Biochem. Physiol. B Biochem. Mol. Biol. 130, 281–289. Tang, C., Qi, J., Li, H., Zhang, C., Wang, Y., 2007. A convenient and efficient protocol for isolating high-quality RNA from latex of Hevea brasiliensis (para rubber tree). J. Biochem. Biophys. Methods 70, 749–754. Tang, C., Huang, D., Yang, J., Liu, S., Sakr, S., Li, H., Zhou, Y., Qin, Y., 2010. The sucrose transporter HbSUT3 plays an active role in sucrose loading to laticifer and rubber productivity in exploited trees of Hevea brasiliensis (para rubber tree). Plant Cell Environ. 33, 1708–1720. Thellin, O., Zorzi, W., Lakaye, B., De Borman, B., Coumans, B., Hennen, G., Grisar, T., Igout, A., Heinen, E., 1999. Housekeeping genes as internal standards: use and limits. J. Biotechnol. 75, 291–295. Tupy, J., 1973. The level and distribution pattern of latex sucrose along the trunk of Hevea brasiliensis Mull. arg. as affected by the sink region induced by latex tapping. Physiologie Vaegaetale, pp. 1–11. 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, H34. Xu, J.H., Qin, Z.H., Liao, Y.S., Xie, M.Q., Li, A.X., Cai, J.P., 2008. Characterization and expression of an actin-depolymerizing factor from Eimeria tenella. Parasitol. Res. 103, 263–270. Zhang, Y., Chen, D., Smith, M.A., Zhang, B., Pan, X., 2012. Selection of reliable reference genes in Caenorhabditis elegans for analysis of nanotoxicity. PLoS One 7, e31849. Zhu, Z.W., Dong, C.S., 2006. The Stability Comparison of Housekeeping Genes as Internal Standards. pp. 807–809.
Please cite this article as: Long, X., et al., Validation of reference genes for quantitative real-time PCR during latex regeneration in rubber tree, Gene (2015), http://dx.doi.org/10.1016/j.gene.2015.03.026
Contents lists available at ScienceDirect
Gene journal homepage: www.elsevier.com/locate/gene
Validation of reference genes for quantitative real-time PCR during latex regeneration in rubber tree Xiangyu Long a,1, Bin He a,b,1, Xinsheng Gao a, Yunxia Qin a, Jianghua Yang a, Yongjun Fang a, Jiyan Qi a, Chaorong Tang a,⁎ a Key Laboratory of Biology and Genetic Resources of Rubber Tree, Ministry of Agriculture, Rubber Research Institute, Chinese Academy of Tropical Agricultural Sciences, Danzhou, Hainan 571737, PR China b College of Agronomy, Hainan University, Haikou, Hainan 570228, PR China
a r t i c l e
i n f o
Article history: Received 12 December 2014 Received in revised form 26 February 2015 Accepted 13 March 2015 Available online xxxx Keywords: Hevea brasiliensis Reference genes Latex regeneration Algorithms Quantitative real-time PCR
a b s t r a c t In rubber tree, latex regeneration is one of the decisive factors influencing the rubber yield, although its molecular regulation is not well known. Quantitative real-time PCR (qPCR) is a popular and powerful tool used to understand the molecular mechanisms of latex regeneration. However, the suitable reference genes required for qPCR are not available to investigate the expressions of target genes during latex regeneration. In this study, 20 candidate reference genes were selected and evaluated for their expression stability across the samples during the process of latex regeneration. All reference genes showed a relatively wide range of the threshold cycle values, and their stability was validated by four different algorithms (comparative delta Ct method, Bestkeeper, NormFinder and GeNorm). Three softwares (comparative delta Ct method, NormFinder and GeNorm) exported similar results that identify UBC4, ADF, UBC2a, eIF2 and ADF4 as the top five suitable references, and 18S as the least suitable one. The application of the screened references would improve accuracy and reliability of gene expression analysis in latex regeneration experiments. © 2015 Elsevier B.V. All rights reserved.
1. Introduction In biological research, the technology of gene expression is broadly applied to understand the biological roles and interrelation of genes in molecular pathways. At transcriptional level, several methods provide high sensitivity and accuracy in the quantification of gene expression, such as transcriptome sequencing, cDNA microarray, Northern blotting and quantitative real-time PCR (qPCR) technology (Kubista et al., 2006; Josefsen and Nielsen, 2011; Lang et al., 2014; Le et al., 2014). Owing to technical ease, low reagent cost, less hand-on time and high throughput, qPCR is increasingly and widely used to measure the expression of target genes across different samples (Kubista et al., 2006). Abbreviations: qRT-PCR, Quantitative real-time PCR; Ct, threshold cycle (previously); Cq, quantification cycle; HKGs, Housekeeping genes; SD, standard deviation; CV, coefficient of variation; 18S, 18s ribosomal RNA; ACT7a, Actin (ACTIN7); ACT7b, Actin (ACTIN7); ADF, Actin depolymerizing factor; ADF4, Actin depolymerizing factor 4; eIF1Aa, Eukaryotic translation initiation factor 1A; eIF1Ab, Eukaryotic translation initiation factor 1A; eIF2, Eukaryotic translation initiation factor; eIF3, Eukaryotic translation initiation factor; FP, F-box family protein; PTP, Trosine phosphatase; RH2a, DEAD box RNA helicase,RH2; RH2b, DEAD box RNA helicase,RH2; ROC3, Cytosolic cyclophilin (ROC3); UBC1, Ubiquitin-protein ligase; UBC2a, Ubiquitin-protein ligase (ATUBC2); UBC2b, Ubiquitin-protein ligase (ATUBC2); UBC3, Ubiquitin-protein ligase; UBC4, Ubiquitin-protein ligase; YLS8, Mitosis protein YLS8. ⁎ Corresponding author. E-mail address: [email protected] (C. Tang). 1 These authors contributed equally to this work.
Unfortunately, there are several factors which can affect the quantitative measurement of gene expression by qPCR, including initial sample amount, RNA recovery, RNA integrity and efficiency of cDNA synthesis. To achieve accurate and stable results, it is essential that one or several reference genes should be used as internal control to normalize variations (Vandesompele et al., 2002; Andersen et al., 2004; Huggett et al., 2005). Theoretically, an ideal reference gene is stably expressed in various samples across different experimental conditions or treatments. Housekeeping genes (HKGs) have been historically used as reference genes for normalization, relating to basal cell activities and cellular structure components (Thellin et al., 1999; Daud and Scott, 2008; Frericks and Esser, 2008). Several housekeeping genes, including 18S or 26S ribosomal RNA (18S or 26S rRNA), glyceraldehyde-3-phosphate dehydrogenase (GAPDH), α or β-actin (ACT), β or γ-tubulin (TUB), ubiquitin C (UBC) and elongation factor-1 alpha (EF-1α) have been validated as suitable reference genes for qPCR analysis in the past (Bustin, 2000; Goidin et al., 2001; Kim and Kim, 2003; Lossos et al., 2003; Mitter et al., 2009). However, no genes are universally stable across different plant species and differing experimental conditions. More than chance, the usage of the reported suitable reference genes in a different species or an altered condition leads to misleading results. Recently, more candidate reference genes are isolated and identified using gene expression profile data in many plants, including Saccharum (Ling et al., 2014), Elaeis guineensis (Chan et al., 2014), Tectona grandis L.f. (Galeano et al.,
http://dx.doi.org/10.1016/j.gene.2015.03.026 0378-1119/© 2015 Elsevier B.V. All rights reserved.
Please cite this article as: Long, X., et al., Validation of reference genes for quantitative real-time PCR during latex regeneration in rubber tree, Gene (2015), http://dx.doi.org/10.1016/j.gene.2015.03.026
2
X. Long et al. / Gene xxx (2015) xxx–xxx
2014), Iris lactea var. chinensis (Gu et al., 2014), Triticum aestivum L. (Long et al., 2010). To improve the results accuracy, before performing qPCR analyses in a new experimental system, it is vital to evaluate multiple reference genes and utilize the most suitable one(s) to quantify gene expression. Four systematic and statistical algorithms, comparative delta Ct method (Silver et al., 2006), Bestkeeper (Pfaffl et al., 2004), NormFinder (Andersen et al., 2004) and GeNorm (Vandesompele et al., 2002) have been developed to assess and identify the most suitable reference genes for qPCR data normalization. Natural rubber, cis-polyisoprene, is an essential industrial substance and strategic material, and is mainly acquired from Hevea brasiliensis, the Pará rubber tree. Natural rubber biosynthesis occurs in the cytoplasm (latex) of laticifers, which are a strong sink and highly specialized cells in the phloem (Metcalfe, 1967; Lewinsohn, 1991). In production, huge amount of latex (30–40% of its dry matter as natural rubber) is harvested by tapping for every two or three days. So, latex regeneration between two consecutive tappings is one of the major limiting factors determining rubber yield (Kongsawadworakul et al., 2009). And the earlier studies found that physiologically, latex regeneration mainly depends on the availability and metabolism of sugar and nitrogen compounds (Tupy, 1973; Pujade-Renaud et al., 1994). At present, the development of molecular biology makes the studies of gene expression and regulation related to natural rubber biosynthesis become a hotspot in undressing the mechanisms of latex regeneration. For instance, several researchers have focused on the genes involved in the supply and utility of sucrose in the latex, using gene expression profile especially qPCR analysis (Dusotoit-Coucaud et al., 2010; Tang et al., 2010; Liu et al., 2015). To reveal the underlying molecular mechanism and regulatory network of latex regeneration, it is necessary and crucial to screen the suitable reference genes for normalization in relevant gene expression analysis. Although we previously characterized a total of 22 candidate genes for their suitability as reference genes in several experimental conditions in rubber tree (Li et al., 2011), the screening of the suitable references for studying latex regeneration has not been touched. In this study, 20 common reference genes were tested for their expression stability across the samples during the process of latex regeneration. Of the four softwares used for evaluating the suitability of reference genes, three (comparative delta Ct method, NormFinder and GeNorm) exported similar results showing UBC4, ADF, UBC2a, eIF2 and ADF4 as the suitable reference genes for normalization in latex regeneration samples. Evaluation and application of them would improve accuracy and reliability of gene expression analysis in latex regeneration experiments. 2. Materials and methods 2.1. Plant materials In this study, two clones of rubber tree, RRIM600 and CATAS628 were applied as experimental materials with different types of metabolism. Two types of tree were growing for 13 years at the experimental plantation of the Rubber Research Institute of the Chinese Academy of Tropical Agricultural Sciences (CATAS) (Danzhou, Hainan, China), and which had been regularly tapped for 6 years in a half spiral pattern, every three days (S/2, d/3). For each rubber clone, 30 trees with similar girth, growth vigor and rubber yield were selected and divided into 6 groups. All groups were tapped simultaneously at the first tapping, and then at 6 h, 12 h, 24 h, 48 h, 72 h and 96 h, respectively, one group for each time was tapped again. After every tapping, latex was collected for RNA extraction according to previous description by Tang (Tang et al., 2007, 2010).
integrity of the RNA samples was checked by agarose gel electrophoresis, and the concentration and quality were examined by NanoDrop 2000 (Thermo, USA) at 230 nm, 260 nm and 280 nm. Synthesis of cDNA was performed using the RevertAidTM First Strand cDNA Synthesis Kit (Fermentas, Canada) following the manufacturer's protocol. 2.3. qPCR A total of 20 candidate reference genes, 18S, ACT7a, ACT7b, ADF, ADF4, eIF2, eIF3, eIF1Aa, eIF1Ab, FP, PTP, RH2a, RH2b, ROC3, UBC1, UBC2a, UBC2b, UBC3, UBC4 and YLS8 were selected from our previous study, and their detail information refers to our previous study including gene annotation, primer sequence, and so on (Table 1S). The realtime PCR was carried out using the SYBR® Premix Ex Taq™ II (Perfect Real Time) (Takara, Dalian, China) and the ABI 7500 Fast Real Time PCR System (Applied Biosystems, Foster City, CA, USA). The PCR reaction system and procedures were described previously (Tang et al., 2007, 2010; Li et al., 2011). The ABI 7500 Software v2.0.6 was used for visualizing and analyzing the data, including the quantification cycle values, PCR efficiency and correlation coefficients. 2.4. Data analysis After collecting and converting the quantification cycle data (Cq), Cq average values were calculated to statistical analysis by SPSS 13 (http:// www.spss.com/). To obtain reliable results, comparative delta Ct method, Bestkeeper, NormFinder and GeNorm were used to analyze expression stability of reference genes, according to their instructions (RefFinder, http://www.leonxie.com/referencegene.php?type=reference). Pearson correlation coefficients were detected for ranking results from four different algorithms, using Minitab 15 software (http://www.minitab. com/). 3. Results 3.1. Expression profiles of candidate reference genes A total of 20 candidate reference genes were selected for determining the most stable one during the process of latex regeneration in the rubber varieties of RRIM600 and CATAS628. Amplification of each reference gene in 24 samples (three replicates per sample) produced 72 Cq values, and samples with missing Cq values or inconsistencies between replicates (Cq differences N0.5 cycle) were removed from the analysis. Based on the standard curves using a serial dilution of cDNA samples, the amplification efficiencies of those primers ranged from 94.18% to 102.28%, and the regression coefficient R2 for all primers varied between 0.997 and 1.000. Over all samples, the 20 candidate reference genes had a wide range of the Cq values, and the mean Cq values of those gene ranged from 15.60 to 23.07 across all the samples. Among these candidate reference genes, 18S was the most abundantly expressed gene (mean Cq ± SD = 15.60 ± 1.07) followed by eIF3 (mean Cq ± SD = 16.38 ± 0.72), whereas RH2a was the least abundantly expressed gene (mean Cq ± SD = 23.07 ± 1.04). There were small standard deviation and coefficient of variation in all samples. The eIF3 standard deviation (SD) was the lowest (0.72) while PTP presented the largest variation between Cq values (SD = 1.14). The coefficient of variation of UBC4 was the smallest (4.18%), while that of 18S was the largest (6.88%) (Table S1). Additionally, using individual value plot to evaluate and compare all samples, the results showed that those genes had similar distribution or trend except 18S and FP (Fig. 1).
2.2. RNA isolation and cDNA synthesis 3.2. Expression stability of the twenty candidate reference genes Total RNA was extracted by the protocol as described previously by Tang (Tang et al., 2007). RNA samples were treated with DNase I (TaKaRa) to eliminate the trace contaminants of genomic DNA. The
In order to make a more detailed expression analysis of the candidate reference genes, the 24 samples were divided into three
Please cite this article as: Long, X., et al., Validation of reference genes for quantitative real-time PCR during latex regeneration in rubber tree, Gene (2015), http://dx.doi.org/10.1016/j.gene.2015.03.026
X. Long et al. / Gene xxx (2015) xxx–xxx
3
(0.134) in the third set. Four candidate reference genes, ROC3, PTP, FP and 18S had the worst stability in the first, second and third sets. The ranks of candidate reference genes from the least to most stable expression were very similar between the first two sets, and the correlation coefficient value was 0.680 (p = 0.001). However, there were two genes that showed great fluctuation of expression stability, including YLS8 (5th in the first set, 19th in the second set) and UBC1 (14th in the first set, 5th in the second set). 3.4. NormFinder
Fig. 1. Expression levels of candidate reference genes tested using the qPCR cycle threshold values (Cq). The red-dot represents reference gene expression level (Cq) in each samples and the black-dot represents the mean expression level of reference genes (arithmetical mean of Cq). (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)
experiment sets. The first set consisted of 12 samples from RRIM600. The second set was comprised of 12 samples from CATAS628. In the third set, both the first and second sample sets were included. The four softwares using different algorithms, including comparative delta Ct method, Bestkeeper, NormFinder and GeNorm, were used to analyze and evaluate the stability for candidate reference genes in three experiment sets.
NormFinder program is used to calculate an arbitrary stability value and standard error, considering the intragroup and intergroup variations of each reference gene in all groups (Andersen et al., 2004). According to NormFinder results, the rank of stability in the first set was somewhat similar to that in the second set, and the correlation coefficient value was 0.439 (p = 0.053) (Table 2). Four reference genes, YLS8, UBC3, ADF4 and ADF, were rated as the most stable reference genes in the first set, with stability values of 0.057–0.083. However, both in the second and third sets, UBC4, ADF, eIF2 and UBC2b were ranked as the four most stable reference genes although ordering differed between the two sets. In any of the three sets, 18S was always determined as the least stable reference gene with stability values of 1.074–2.211, which were above the default limit of 1.0 in the NormFinder program. Additionally, YLS8 showed large fluctuations of expression stability in the three sets, ranking 1st in the first set, 18th in the second set and 15th in the third set. 3.5. Delta Ct method
The program GeNorm was utilized to calculate the expression stability M based on the average pair-wise variation between all studied genes in the three experimental sets (the first, second, and third sets) (Table 1). All of the genes showed high expression stability with M values of less than 0.600 (0.593 in the first set, 0.419 in the second set and 0.548 in the third set), which were below the default limit of 1.5 in the GeNorm program. From the results of GeNorm, the top ranked genes were ADF4 = UBC2b (0.099) N UBC3 (0.117) N UBC4 (0.126) in the first set, ADF4 = UBC2a (0.107) N UBC3 (0.120) N UBC4 (0.132) in the second set, and UBC2a = UBC4 (0.124) N ADF4 (0.128) N UBC3
The comparative delta-Ct method calculates the standard deviation of Cq differences within each sample for each pairwise comparison with the other genes to define the stable reference genes (Silver et al., 2006), and which is similar to the ΔΔCt method for real-time quantitative PCR (Livak and Schmittgen, 2001). A correlation value of ranking between the first and second sets was 0.549 (p = 0.012), indicating a similar result obtained for the 20 reference genes evaluated (Table 3). The comparative delta-Ct results showed that ADF4 was identified as the most stable reference gene with stability value of 0.376 in the first set, UBC4 with stability value of 0.297 in the second set and UBC4 with stability value of 0.373 in the third set. Interestingly, consistent results were obtained in all three sets that 18S was the least stable gene, with stability value of 2.244 in the first set, 1.102 in the second set and
Table 1 Expression stability of 20 candidate reference genes as calculated by GeNorm.
Table 2 Expression stability of 20 candidate reference genes as calculated by NormFinder.
3.3. GeNorm
Rank
1 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
The first set
The second set
The third set
Rank
Gene
M values
Gene
M values
Gene
M values
ADF4 | UBC2b UBC3 UBC4 YLS8 ADF UBC2a eIF2 eIF1Aa RH2b RH2a ACT7b eIF1Ab UBC1 ACT7a eIF3 ROC3 PTP FP 18S
0.099 0.117 0.126 0.131 0.136 0.140 0.145 0.152 0.158 0.166 0.173 0.181 0.189 0.197 0.224 0.251 0.277 0.410 0.593
ADF4 | UBC2a UBC3 UBC4 UBC1 ADF eIF1Aa eIF2 UBC2b RH2b eIF3 ACT7a eIF1Ab RH2a ACT7b PTP FP ROC3 YLS8 18S
0.107 0.120 0.132 0.148 0.158 0.173 0.181 0.188 0.195 0.204 0.216 0.226 0.243 0.258 0.274 0.292 0.318 0.343 0.419
UBC2a | UBC4 ADF4 UBC3 ADF eIF2 UBC2b RH2b UBC1 eIF1Aa ACT7a RH2a ACT7b eIF1Ab eIF3 YLS8 PTP ROC3 FP 18S
0.124 0.128 0.134 0.154 0.167 0.175 0.183 0.193 0.203 0.214 0.224 0.237 0.252 0.267 0.284 0.303 0.333 0.416 0.548
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
The first set
The second set
The third set
Gene
Stability
Gene
Stability
Gene
Stability
YLS8 UBC3 ADF4 ADF UBC4 UBC2a eIF1Aa UBC2b eIF2 UBC1 eIF1Ab ACT7a RH2b RH2a ACT7b eIF3 ROC3 PTP FP 18S
0.057 0.062 0.068 0.083 0.086 0.124 0.124 0.125 0.128 0.142 0.165 0.199 0.203 0.229 0.264 0.312 0.407 0.528 1.523 2.211
UBC4 ADF eIF2 UBC2b eIF1Aa UBC2a ADF4 RH2b UBC3 UBC1 eIF3 ACT7a eIF1Ab RH2a ACT7b FP PTP YLS8 ROC3 18S
0.065 0.067 0.107 0.117 0.125 0.133 0.162 0.162 0.169 0.177 0.181 0.215 0.226 0.325 0.359 0.370 0.416 0.483 0.487 1.074
ADF UBC4 eIF2 UBC2b UBC3 ADF4 UBC1 UBC2a RH2b eIF1Aa ACT7a eIF3 eIF1Ab RH2a YLS8 ACT7b PTP ROC3 FP 18S
0.073 0.081 0.113 0.122 0.129 0.143 0.155 0.160 0.179 0.179 0.205 0.264 0.270 0.279 0.330 0.342 0.464 0.526 1.086 1.710
Please cite this article as: Long, X., et al., Validation of reference genes for quantitative real-time PCR during latex regeneration in rubber tree, Gene (2015), http://dx.doi.org/10.1016/j.gene.2015.03.026
4
X. Long et al. / Gene xxx (2015) xxx–xxx
Table 3 Expression stability of 20 candidate reference genes as calculated by comparative delta Ct method. Rank
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
The first set
The second set
The third set
Gene
SD
Gene
SD
Gene
SD
ADF4 UBC2b UBC4 ADF YLS8 eIF1Aa UBC3 RH2b UBC2a RH2a eIF2 eIF1Ab UBC1 ACT7b ACT7a eIF3 ROC3 PTP FP 18S
0.376 0.390 0.391 0.391 0.391 0.397 0.400 0.410 0.414 0.424 0.425 0.431 0.439 0.443 0.456 0.582 0.616 0.651 1.598 2.244
UBC4 ADF UBC2a eIF1Aa UBC2b ADF4 eIF2 RH2b UBC3 UBC1 eIF3 ACT7a eIF1Ab RH2a ACT7b FP PTP ROC3 YLS8 18S
0.297 0.300 0.315 0.320 0.322 0.324 0.325 0.333 0.335 0.340 0.359 0.369 0.374 0.415 0.439 0.476 0.485 0.565 0.579 1.102
UBC4 ADF ADF4 UBC2b UBC3 RH2b UBC2a eIF2 UBC1 eIF1Aa ACT7a RH2a eIF1Ab ACT7b eIF3 YLS8 PTP ROC3 FP 18S
0.373 0.376 0.384 0.386 0.394 0.399 0.401 0.402 0.411 0.416 0.434 0.445 0.474 0.490 0.501 0.528 0.581 0.667 1.163 1.743
values exceed 1.0, reference genes are considered as unstable ones and should be avoided for gene expression normalization. Nine genes in the first set, two genes in the second set and five genes in the third set had SD values of larger than 1.0 (1.004–1.402) and CV values of 4.37–8.14. 3.7. Comprehensive analysis of expression stability in the third set
1.743 in the third set. Again, YLS8 exhibited a big change for stable ranking as 1st in the first set but 18th in the second set, consistent with the outputs from GeNorm and NormFinder.
For a comprehensive judgment of the suitable reference genes in the third set, the Pearson correlations were calculated using the ranks from the most stable to the least stable among the four methods (comparative delta Ct method, Bestkeeper, NormFinder and GeNorm) used in this study (Fig. 2). The Pearson correlations for the four stability tests showed a significant or extremely significant positive correlation. Three high Pearson correlations were observed between the outcomes from delta Ct method and GeNorm (r = 0.950, p = 0.000), delta Ct method and NormFinder (r = 0.949, p = 0.000), and NormFinder and GeNorm (r = 0.908, p = 0.000), indicating that ranking results from the above three methods (delta Ct, GeNorm and NormFinder) were nearly identical in the third set. However, Bestkeeper had a relatively lower correlation with the other three methods, showing a correlation efficient of 0.516 with delta Ct method, 0.559 with GeNorm and 0.675 with NormFinder. According to the four methods in the third set, five reference genes (UBC4, ADF, UBC2a, eIF2 and ADF4) were rank as the most stable expression ones, whereas five reference genes, 18S, FP, PTP, ROC3 and ACT7b were considered as the least stable expression ones in the third set (Tables 1, 2 and 3).
3.6. Bestkeeper
4. Discussion
BestKeeper ranks the reference genes according to the standard deviation (SD) of their Cqs, and the output includes more information, for example the coefficient of variation (CV) and correlation (r) (Pfaffl et al., 2004). There was a large difference in reference gene stability between the first and second sets, and the coefficient of correlation was only 0.018 (p = 0.940) for ranking results (Table 4). The stable reference genes are those with low coefficient of variance and standard deviation, and high coefficient of correlation. eIF3 was ranked as the most stable expression genes with the CV ± SD value of 4.34 ± 0.71 and correlation value of 0.979 in the first set, and the CV ± SD value of 4.39 ± 0.720 and correlation value of 0.983 in the third set. In the second set, eIF1Ab had low CV ± SD value of 3.38 ± 0.706 and high correlation value of 0.985, showing the most stable expression pattern. Additionally, if the SD
In rubber tree, as one of the decisive factors that influence the rubber yield, latex regeneration is an extremely complex metabolic process including biosynthesis of nature rubber, supply and utility of sucrose, remodeling of organelles and recovering of other compounds (Tupy, 1973; Kongsawadworakul et al., 2009). Presently, gene expression and regulation have been widely used to understand the molecular mechanisms of latex regeneration from one or another aspect (Dusotoit-Coucaud et al., 2010; Tang et al., 2010). The qPCR has routinely become the main technique to conduct gene expression analysis. To get reliable gene expression results for the examined genes during latex regeneration, it is important that more ideal reference genes are screened for normalization of gene expression. In previous research, 18S and ACTIN genes have been the most widely used reference genes for expression normalization in rubber tree (Dusotoit-Coucaud et al., 2009; Duan et al., 2010; Li et al., 2010; Tang et al., 2010). However, extensive analysis is not conducted to validate their suitability as reference genes in those studies. In our previous study, in order to screen more suitable reference genes, 22 candidate genes were evaluated for expression stability in various tissues, different hormone treatments, tappings, individual trees and genotypes. The results showed that those candidate reference genes display distinct expression stability in different experimental conditions, and UBC2a and UBC4 are considered as the suitable reference genes in all conditions taken together (Li et al., 2011). In this study, we analyzed and compared 20 candidate genes for expression stability across 24 samples during the process of latex regeneration in rubber varieties of RRIM600 and CATAS628. Five genes, UBC4, ADF, UBC2a, eIF2 and ADF4 were considered as the top five suitable reference genes to normalize all the samples, especially UBC4 showing the most stable expression when the results of the four methods are taken together (Tables 1, 2 and 3). Ubiquitin-protein ligase genes (UBC4 and UBC2a) are frequently used as appropriate reference gene in other plants, including Arabidopsis thaliana, Brachypodium distachyon and Brachiaria brizantha (Czechowski et al., 2005; Hong et al., 2008; Silveira et al., 2009). Another gene family, actin-depolymerizing factor 4 (ADF and ADF4), which plays an important role in remodeling the actin cytoskeleton (Xu
Table 4 Expression stability of 20 candidate reference genes as calculated by Bestkeeper. Rank
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
The first set
The second set
The third set
Gene
SD
Gene
SD
Gene
SD
eIF3 eIF2 UBC2a ACT7a YLS8 UBC4 UBC3 UBC2b ADF4 ADF ROC3 eIF1Aa RH2b eIF1Ab UBC1 RH2a ACT7b PTP 18S FP
0.712 0.872 0.897 0.919 0.939 0.953 0.954 0.990 0.993 0.996 0.998 1.011 1.012 1.028 1.033 1.064 1.073 1.272 1.279 1.402
eIF1Ab ROC3 eIF3 eIF1Aa FP eIF2 ADF UBC4 ACT7a 18S UBC2b UBC3 RH2b UBC1 UBC2a YLS8 ADF4 ACT7b PTP RH2a
0.706 0.712 0.728 0.752 0.756 0.782 0.793 0.813 0.824 0.830 0.858 0.861 0.868 0.869 0.871 0.883 0.906 0.974 1.004 1.010
eIF3 eIF2 ACT7a UBC4 UBC2a ADF eIF1Aa UBC3 YLS8 UBC2b eIF1Ab RH2b ADF4 UBC1 ROC3 ACT7b RH2a 18S FP PTP
0.720 0.827 0.876 0.883 0.886 0.901 0.902 0.907 0.913 0.932 0.943 0.947 0.949 0.951 0.963 1.029 1.037 1.073 1.104 1.138
Please cite this article as: Long, X., et al., Validation of reference genes for quantitative real-time PCR during latex regeneration in rubber tree, Gene (2015), http://dx.doi.org/10.1016/j.gene.2015.03.026
X. Long et al. / Gene xxx (2015) xxx–xxx
5
Fig. 2. Comparison of the ranking results from comparative delta Ct method, Bestkeeper, NormFinder and GeNorm. The correlation was evaluated for the ranking results of 20 candidate reference genes in all samples, by comparative delta Ct method, Bestkeeper, NormFinder and GeNorm. Correlation coefficient (r) values are shown (*p b 0.05, **p b 0.01).
et al., 2008), also has relatively stable expression in latex generation. Similar to the previous findings in other plants, 18S had very poor expression stability across latex regeneration samples in rubber tree. Some researchers reported that the 18S and 26S were not good choice as reference genes (Sturzenbaum and Kille, 2001; Zhu and Dong, 2006), because the rRNA is much higher expressed compared with the mRNA and easily fluctuates with various biological and drug factors. Additionally, the rankings for 20 candidate reference genes differed between the varieties of RRIM600 and CATAS628, especially for several ones. For example, the mitosis protein gene, YLS8 performed quite stable expression in RRIM600, which ranked first as analyzed in two methods (Tables 2 and 3) and fifth in the other two methods (Tables 1 and 4). In contrast, the same gene was among the five least stable reference genes as analyzed in CATAS628 with any of the four methods (Tables 1–4). However, YLS8 was a novel suitable reference in A. thaliana (Remans et al., 2008), and considered as the most suitable reference gene for hormone and tapping treatments in rubber tree (Remans et al., 2008; Li et al., 2011). The four common methods, comparative delta Ct method, Bestkeeper, NormFinder and GeNorm are often used to check the stability of reference genes. In the past research, the results from comparative delta Ct method, NormFinder and GeNorm have been reported to be more consistent among them than those of Bestkeeper (Zhang et al., 2012; Robledo et al., 2014). Similar results were obtained in this study, which showed a high correlation among comparative delta Ct method, NormFinder and GeNorm, especially between comparative delta Ct method and GeNorm (Fig. 2). Of course, each type of softwares has its own risks and benefits according to different algorithms. Two softwares, GeNorm and comparative delta Ct method take a similar algorithm approach to rank reference genes from the most to least stable expression, using pairwise comparison for standard deviation between one and all other ones (Vandesompele et al., 2002; Silver et al., 2006). However, co-regulation of reference genes that have similar function or basic roles in biological processes could influence the validity of the two methods (Long et al., 2010). NormFinder software could be used to control the influence of co-regulation, taking into account intra-group and inter-group variations to evaluate the expression stability (Andersen et al., 2004). So, NormFinder and GeNorm are considered as the useful and common softwares for selecting the suitable reference genes when combined together. The Bestkeeper directly calculates the standard deviation to measure the
variation, outputting the stability of reference genes. However, there is obvious problem that the gene with low standard deviation may not be considered as suitable reference gene, because the cDNA samples could not be controlled at completely consentient level produced by RNA extraction and reverse transcription steps (Pfaffl et al., 2004). Our results here again reinforce the idea that no reference genes are absolutely stable, because their expressions may be irregular and unsteady in some experiments. So, to obtain reliable expression results, it is absolutely necessary that the expression stability should be evaluated to select the suitable reference gene using different softwares. It is better to chose more than one reference gene for normalization, and these genes should be involved in different biological functions and pathways. In brief, we conclude that UBC4, ADF, UBC2a, eIF2 and ADF4 are the first set of reference genes selected to normalize gene expressions in different latex regeneration samples, by applying comparative delta Ct method, NormFinder and GeNorm together. Our results will be of great benefit to the gene expression-based studies for the investigation of the molecular regulation of latex regeneration in rubber tree. Acknowledgments This research was supported by the 863 program (2013AA102605), the National Natural Science Foundation of China (31300570) and the Natural Science Foundation of Hainan province (312029). Appendix A. Supplementary data Supplementary data to this article can be found online at http://dx. doi.org/10.1016/j.gene.2015.03.026. References Andersen, C.L., Jensen, J.L., Orntoft, T.F., 2004. Normalization of real-time quantitative reverse transcription-PCR data: a model-based variance estimation approach to identify genes suited for normalization, applied to bladder and colon cancer data sets. Cancer Res. 64, 5245–5250. Bustin, S.A., 2000. Absolute quantification of mRNA using real-time reverse transcription polymerase chain reaction assays. J. Mol. Endocrinol. 25, 169–193. Chan, P.L., Rose, R.J., Abdul, M.A., Zainal, Z., Low, E.T., Ooi, L.C., Ooi, S.E., Yahya, S., Singh, R., 2014. Evaluation of reference genes for quantitative real-time PCR in oil palm elite planting materials propagated by tissue culture. PLoS One 9, e99774.
Please cite this article as: Long, X., et al., Validation of reference genes for quantitative real-time PCR during latex regeneration in rubber tree, Gene (2015), http://dx.doi.org/10.1016/j.gene.2015.03.026
6
X. Long et al. / Gene xxx (2015) xxx–xxx
Czechowski, T., Stitt, M., Altmann, T., Udvardi, M.K., Scheible, W.R., 2005. Genome-wide identification and testing of superior reference genes for transcript normalization in Arabidopsis. Plant Physiol. 139, 5–17. Daud, I.I., Scott, M.E., 2008. Validation of reference genes in cervical cell samples from human papillomavirus-infected and -uninfected women for quantitative reverse transcription-PCR assays. Clin. Vaccine Immunol. 15, 1369–1373. Duan, C., Rio, M., Leclercq, J., Bonnot, F., Oliver, G., Montoro, P., 2010. Gene expression pattern in response to wounding, methyl jasmonate and ethylene in the bark of Hevea brasiliensis. Tree Physiol. 30, 1349–1359. Dusotoit-Coucaud, A., Brunel, N., Kongsawadworakul, P., Viboonjun, U., Lacointe, A., Julien, J.L., Chrestin, H., Sakr, S., 2009. Sucrose importation into laticifers of Hevea brasiliensis, in relation to ethylene stimulation of latex production. Ann. Bot. 104, 635–647. Dusotoit-Coucaud, A., Kongsawadworakul, P., Maurousset, L., Viboonjun, U., Brunel, N., Pujade-Renaud, V., Chrestin, H., Sakr, S., 2010. Ethylene stimulation of latex yield depends on the expression of a sucrose transporter (HbSUT1B) in rubber tree (Hevea brasiliensis). Tree Physiol. 30, 1586–1598. Frericks, M., Esser, C., 2008. A toolbox of novel murine house-keeping genes identified by meta-analysis of large scale gene expression profiles. Biochim. Biophys. Acta 1779, 830–837. Galeano, E., Vasconcelos, T.S., Ramiro, D.A., De Martin, V.F., Carrer, H., 2014. Identification and validation of quantitative real-time reverse transcription PCR reference genes for gene expression analysis in teak (Tectona grandis L.f.). BMC Res. Notes 7, 464. Goidin, D., Mamessier, A., Staquet, M.J., Schmitt, D., Berthier-Vergnes, O., 2001. Ribosomal 18S RNA prevails over glyceraldehyde-3-phosphate dehydrogenase and beta-actin genes as internal standard for quantitative comparison of mRNA levels in invasive and noninvasive human melanoma cell subpopulations. Anal. Biochem. 295, 17–21. Gu, C.S., Liu, L.Q., Xu, C., Zhao, Y.H., Zhu, X.D., Huang, S.Z., 2014. Reference gene selection for quantitative real-time RT-PCR normalization in Iris lactea var. chinensis roots under cadmium, lead, and salt stress conditions. ScientificWorldJournal 2014, 532713. Hong, S.Y., Seo, P.J., Yang, M.S., Xiang, F., Park, C.M., 2008. Exploring valid reference genes for gene expression studies in Brachypodium distachyon by real-time PCR. BMC Plant Biol. 8, 112. Huggett, J., Dheda, K., Bustin, S., Zumla, A., 2005. Real-time RT-PCR normalisation; strategies and considerations. Genes Immun. 6, 279–284. Josefsen, K., Nielsen, H., 2011. Northern blotting analysis. Methods Mol. Biol. 703, 87–105. Kim, S., Kim, T., 2003. Selection of optimal internal controls for gene expression profiling of liver disease. Biotechniques 35, 456–458. Kongsawadworakul, P., Viboonjun, U., Romruensukharom, P., Chantuma, P., Ruderman, S., Chrestin, H., 2009. The leaf, inner bark and latex cyanide potential of Hevea brasiliensis: evidence for involvement of cyanogenic glucosides in rubber yield. Phytochemistry 70, 730–739. Kubista, M., Andrade, J.M., Bengtsson, M., Forootan, A., Jonak, J., Lind, K., Sindelka, R., Sjoback, R., Sjogreen, B., Strombom, L., Stahlberg, A., Zoric, N., 2006. The real-time polymerase chain reaction. Mol. Aspects Med. 27, 95–125. Lang, V., Usadel, B., Obermeyer, G., 2014. De novo sequencing and analysis of the lily pollen transcriptome: an open access data source for an orphan plant species. Plant Mol. Biol. 87, 69–80. Le, M.Q., Pagter, M., Hincha, D.K., 2014. Global changes in gene expression, assayed by microarray hybridization and quantitative RT-PCR, during acclimation of three Arabidopsis thaliana accessions to sub-zero temperatures after cold acclimation. Plant Mol. Biol. 87, 1–15. Lewinsohn, T., 1991. The geographical distribution of plant latex. Chemoecology 2, 64–68. Li, D., Deng, Z., Chen, C., Xia, Z., Wu, M., He, P., Chen, S., 2010. Identification and characterization of genes associated with tapping panel dryness from Hevea brasiliensis latex using suppression subtractive hybridization. BMC Plant Biol. 10, 140. Li, H., Qin, Y., Xiao, X., Tang, C., 2011. Screening of valid reference genes for real-time RTPCR data normalization in Hevea brasiliensis and expression validation of a sucrose transporter gene HbSUT3. Plant Sci. 181, 132–139. Ling, H., Wu, Q., Guo, J., Xu, L., Que, Y., 2014. Comprehensive selection of reference genes for gene expression normalization in sugarcane by real time quantitative RT-PCR. PLoS One 9, e97469.
Liu, S.J., Lan, J.X., Zhou, B.H., Qin, Y.X., Zhou, Y.H., Xiao, X.H., Yang, J.H., Gou, J.Q., Qi, J.Y., Huang, Y.C., Tang, C.R., 2015. HbNIN2, a cytosolic alkaline/neutral-invertase, is responsible for sucrose catabolism in rubber-producing laticifers of Hevea brasiliensis (para rubber tree). New Phytol. 206 (In press). Livak, K.J., Schmittgen, T.D., 2001. Analysis of relative gene expression data using real-time quantitative PCR and the 2(-delta delta C(T)). Method 402–408. Long, X.Y., Wang, J.R., Ouellet, T., Rocheleau, H., Wei, Y.M., Pu, Z.E., Jiang, Q.T., Lan, X.J., Zheng, Y.L., 2010. Genome-wide identification and evaluation of novel internal control genes for Q-PCR based transcript normalization in wheat. Plant Mol. Biol. 74, 307–311. Lossos, I.S., Czerwinski, D.K., Wechser, M.A., Levy, R., 2003. Optimization of quantitative real-time RT-PCR parameters for the study of lymphoid malignancies. Leukemia 17, 789–795. Metcalfe, C.R., 1967. Distribution of latex in the plant kingdom. Econ. Bot. 21, 115–127. Mitter, K., Kotoulas, G., Magoulas, A., Mulero, V., Sepulcre, P., Figueras, A., Novoa, B., Sarropoulou, E., 2009. Evaluation of candidate reference genes for QPCR during ontogenesis and of immune-relevant tissues of European seabass (Dicentrarchus labrax). Comp. Biochem. Physiol. B Biochem. Mol. Biol. 153, 340–347. Pfaffl, M.W., Tichopad, A., Prgomet, C., Neuvians, T.P., 2004. Determination of stable housekeeping genes, differentially regulated target genes and sample integrity: BestKeeper–Excel-based tool using pair-wise correlations. Biotechnol. Lett. 26, 509–515. Pujade-Renaud, V., Clement, A., Perrot-Rechenmann, C., Prevot, J.C., Chrestin, H., Jacob, J.L., Guern, J., 1994. Ethylene-induced increase in glutamine synthetase activity and mRNA levels in Hevea brasiliensis latex cells. Plant Physiol. 105, 127–132. Remans, T., Smeets, K., Opdenakker, K., Mathijsen, D., Vangronsveld, J., Cuypers, A., 2008. Normalisation of real-time RT-PCR gene expression measurements in Arabidopsis thaliana exposed to increased metal concentrations. Planta 227, 1343–1349. Robledo, D., Hernandez-Urcera, J., Cal, R.M., Pardo, B.G., Sanchez, L., Martinez, P., Vinas, A., 2014. Analysis of qPCR reference gene stability determination methods and a practical approach for efficiency calculation on a turbot (Scophthalmus maximus) gonad dataset. BMC Genomics 15, 648. Silveira, E.D., Alves-Ferreira, M., Guimaraes, L.A., Da, S.F., Carneiro, V.T., 2009. Selection of reference genes for quantitative real-time PCR expression studies in the apomictic and sexual grass Brachiaria brizantha. BMC Plant Biol. 9, 84. Silver, N., Best, S., Jiang, J., Thein, S.L., 2006. Selection of housekeeping genes for gene expression studies in human reticulocytes using real-time PCR. BMC Mol. Biol. 7, 33. Sturzenbaum, S.R., Kille, P., 2001. Control genes in quantitative molecular biological techniques: the variability of invariance. Comp. Biochem. Physiol. B Biochem. Mol. Biol. 130, 281–289. Tang, C., Qi, J., Li, H., Zhang, C., Wang, Y., 2007. A convenient and efficient protocol for isolating high-quality RNA from latex of Hevea brasiliensis (para rubber tree). J. Biochem. Biophys. Methods 70, 749–754. Tang, C., Huang, D., Yang, J., Liu, S., Sakr, S., Li, H., Zhou, Y., Qin, Y., 2010. The sucrose transporter HbSUT3 plays an active role in sucrose loading to laticifer and rubber productivity in exploited trees of Hevea brasiliensis (para rubber tree). Plant Cell Environ. 33, 1708–1720. Thellin, O., Zorzi, W., Lakaye, B., De Borman, B., Coumans, B., Hennen, G., Grisar, T., Igout, A., Heinen, E., 1999. Housekeeping genes as internal standards: use and limits. J. Biotechnol. 75, 291–295. Tupy, J., 1973. The level and distribution pattern of latex sucrose along the trunk of Hevea brasiliensis Mull. arg. as affected by the sink region induced by latex tapping. Physiologie Vaegaetale, pp. 1–11. 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, H34. Xu, J.H., Qin, Z.H., Liao, Y.S., Xie, M.Q., Li, A.X., Cai, J.P., 2008. Characterization and expression of an actin-depolymerizing factor from Eimeria tenella. Parasitol. Res. 103, 263–270. Zhang, Y., Chen, D., Smith, M.A., Zhang, B., Pan, X., 2012. Selection of reliable reference genes in Caenorhabditis elegans for analysis of nanotoxicity. PLoS One 7, e31849. Zhu, Z.W., Dong, C.S., 2006. The Stability Comparison of Housekeeping Genes as Internal Standards. pp. 807–809.
Please cite this article as: Long, X., et al., Validation of reference genes for quantitative real-time PCR during latex regeneration in rubber tree, Gene (2015), http://dx.doi.org/10.1016/j.gene.2015.03.026