Advertisement
Browse Subject Areas
?

Click through the PLOS taxonomy to find articles in your field.

For more information about PLOS Subject Areas, click here.

  • Loading metrics

Selection and Validation of Reference Genes for Real-Time Quantitative PCR in Hyperaccumulating Ecotype of Sedum alfredii under Different Heavy Metals Stresses

  • Jian Sang ,

    Contributed equally to this work with: Jian Sang, Xiaojiao Han

    Affiliations State Key Laboratory of Tree Genetics and Breeding, Beijing, China, Key Laboratory of Tree Genomics, The Research Institute of Subtropical Forestry, Chinese Academy of Forestry, Fuyang, Zhejiang, China

  • Xiaojiao Han ,

    Contributed equally to this work with: Jian Sang, Xiaojiao Han

    Affiliations State Key Laboratory of Tree Genetics and Breeding, Beijing, China, Key Laboratory of Tree Genomics, The Research Institute of Subtropical Forestry, Chinese Academy of Forestry, Fuyang, Zhejiang, China

  • Mingying Liu,

    Affiliations State Key Laboratory of Tree Genetics and Breeding, Beijing, China, Key Laboratory of Tree Genomics, The Research Institute of Subtropical Forestry, Chinese Academy of Forestry, Fuyang, Zhejiang, China

  • Guirong Qiao,

    Affiliations State Key Laboratory of Tree Genetics and Breeding, Beijing, China, Key Laboratory of Tree Genomics, The Research Institute of Subtropical Forestry, Chinese Academy of Forestry, Fuyang, Zhejiang, China

  • Jing Jiang,

    Affiliations State Key Laboratory of Tree Genetics and Breeding, Beijing, China, Key Laboratory of Tree Genomics, The Research Institute of Subtropical Forestry, Chinese Academy of Forestry, Fuyang, Zhejiang, China

  • Renying Zhuo

    zhuory@gmail.com

    Affiliations State Key Laboratory of Tree Genetics and Breeding, Beijing, China, Key Laboratory of Tree Genomics, The Research Institute of Subtropical Forestry, Chinese Academy of Forestry, Fuyang, Zhejiang, China

Selection and Validation of Reference Genes for Real-Time Quantitative PCR in Hyperaccumulating Ecotype of Sedum alfredii under Different Heavy Metals Stresses

  • Jian Sang, 
  • Xiaojiao Han, 
  • Mingying Liu, 
  • Guirong Qiao, 
  • Jing Jiang, 
  • Renying Zhuo
PLOS
x

Abstract

Real-time Quantitative PCR (RT-qPCR) has become an effective method for accurate analysis of gene expression in several biological systems as well as under different experimental conditions. Although with high sensitivity, specificity and broad dynamic range, this method requires suitable reference genes for transcript normalization in order to guarantee reproducible and meaningful results. In the present study, we evaluated five traditional housekeeping genes and five novel reference genes in Hyperaccumulating ecotype of Sedum alfredii, a well known hyperaccumulator for heavy metals phytoremediation, under Cd, Pb, Zn and Cu stresses of seven different durations. The expression stability of these ten candidates were determined with three programs - geNorm, NormFinder and BestKeeper. The results showed that all the selected reference genes except for SAND could be used for RT-qPCR normalization. Among them UBC9 and TUB were ranked as the most stable candidates across all samples by three programs together. For the least stable reference genes, however, BestKeeper produced different results compared with geNorm and NormFinder. Meanwhile, the expression profiles of PCS under Cd, Pb, Zn and Cu stresses were assessed using UBC9 and TUB respectively, and similar trends were obtained from the results of the two groups. The distinct expression patterns of PCS indicated that various strategies could be taken by plants in adaption to different heavy metals stresses. This study will provide appropriate reference genes for further gene expression quantification using RT-qPCR in Hyperaccumulator S. alfredii.

Introduction

The contamination of soil and underground water by heavy metals is not only posing an unacceptable hazard to human and animal health but also placing an overwhelming burden on eco-environment [1,2]. In the past decade, exploiting natural hyperaccumulator plants to decontaminate heavy metal polluted soil, namely phytoremediation, has been regarded as an promising and eco-friendly strategy with low cost [3-5]. More than 400 plant species were already identified as natural heavy metal hyperaccumulators (account for about 0.2% of all angiosperms) [6], among them Hyperaccumulating ecotype of S. alfredii is a Cd/Zn/Pb co-hyperaccumulating species native to China which can also confer strong tolerance to heavy metals stresses [7]. To date, several critical physiological indices, such as accumulation capacity and translocation rate of different metal ions from roots to shoots and their regularities of distribution, have been characterized in this plant [8-11]. However, minimal knowledge is available on the molecular mechanism underlying hyper-accumulation and hyper-tolerance to heavy metals in S. alfredii. In order to explore the molecular mechanism and make full use of its gene resources for further phytoremediation breeding, it is necessary to dig out the candidate genes responsible for metal hyper-accumulation and analyze their expression patterns under heavy metals stresses [12,13].

Gene expression analysis by real-time quantitative PCR (RT-qPCR) is extensively used in many fields of plant biological research, including responses to abiotic and biotic stresses. Compared with conventional methods, RT-qPCR is more practical in gene expression analysis for its high sensitivity, specificity and a broad dynamic range even with limited amounts of RNA sample [14-16]. Regardless of its numerous advantages, this method requires appropriate normalization methods to ensure reliable and meaningful gene expression measurement. Among several strategies reported, it is the most popular to make use of suitable reference genes for gene expression normalization. The ideal reference genes should be expressed at a relatively constant level among samples from different tissues and developmental stages, and also should be unaffected by different experimental treatments [17-19]. During the previous gene expression analysis in S. alfredii, ACT was often used for RT-qPCR normalization [20-22], however, without any systematic study validating reference genes under abiotic stress reported for S. alfredii. Recent studies revealed that some traditional housekeeping genes such as ACT, TUB and EF-1α were not constantly stably expressed under different experimental conditions [23-25]. Therefore, it is necessary to perform a selection and validation of reference genes in a preliminary experiment to appraise their practicability for RT-qPCR normalization in S. alfredii.

In the present study, we compared the stability of ten candidate genes in Hyperaccumulating ecotype of S. alfredii exposed to four kinds of heavy metals: CdCl2, ZnSO4, Pb(NO3)2 and CuSO4 for seven different durations, respectively. Among them, five candidate genes (ACT2, TUB, GADPH, UBC9, EF-1α) are used as traditional housekeeping genes, while the others (SAND, TIP41, AP-2, TBP, APT) are relatively novel reference genes. Transcriptome data of Hyperaccumulating ecotype of S. alfredii was used for the source of potential reference genes. Our results will provide suitable RT-qPCR normalization for accurate gene expression analysis in Hyperaccumulating ecotype of S. alfredii.

Results

Identification of reference gene candidates and primer specificity and efficiency

A total of ten reference genes were selected as potential candidates for RT-qPCR normalization through a search for the orthologs of Arabidopsis reference genes in Hyperaccumulating ecotype of S. alfredii transcriptome data. These genes are the orthologs of AT3G18780 (Actin-2, ACT2), AT1G20010 (beta-Tubulin, TUB), AT4G27960 (Ubiquitin conjugating enzyme 9, UBC9), AT4G34270 (TIP41-like family protein, TIP41), AT2G28390 (SAND family protein, SAND), AT5G46630 (AP-2 complex subunit mu-like, AP-2), AT1G55520 (TATA binding protein, TBP), AT1G27450 (Adenine Phosphoribosyltransferase, APT), AT5G60390 (Elongation factor 1-alpha, EF-1α), AT1G13440 (glyceraldehyde 3-phosphate dehydrogenase, GAPDH). To verify the accuracy of the selected gene sequences, RT-qPCR products were sequenced respectively, and their sequences were listed in the File S1. The estimated PCR amplification efficiency of these candidates varied from 1.963 for GAPDH to 2.082 for TIP41, and the correlation coefficients (R2) ranged between 0.9964 for AP-2 and 0.9997 for APT, respectively (Table 1). Melting curves analysis with single distinctive peak revealed that no primer dimers were generated from non-specific amplification (Figure 1a). Furthermore, the amplification specificity was confirmed by 2% agarose gel electrophoresis with single fragment of the expected size (Figure 1b).

Gene SymbolGeneBank Accession nmuberPrimer sequence(forward/reverse)Amplicon Length (bp)Amplification efficiency R2
ACT2KF652103TTCCGGTGATGGTGTCAGTCA/ACAATTTCCCGCTCAGCAGTG 1652.0330.9980
TUBKF652104TTATGGCGATTCCGAGCTTCA/ATTATTTCCAGCGCCGGATTG1932.0570.9981
EF-1αKF652105CATTATGAACCACCCCGGTCA/CCGCATCTCCATTCTTCAGGA1642.0710.9979
UBC9KF652106TGGCGTCGAAAAGGATTCTGA/CCTTCGGTGGCTTGAATGGATA1982.0830.9969
GAPDHKF652107AAGAGCCGCCTCGTTCAACAT/TGATTGCCGCCTTGATCTCAT1911.9750.9965
SANDKF652108TGGAGCGAAACCCTAGCTTCA/TCATCCACGCGGATTGAACTT 2112.0730.9980
TIP41KF652109ATGCACTTGCTGGATGGAAGC/TCCTTGCAGTCCTCCCATTGA2231.9630.9996
AP-2KF652110ACTTTGGTGGGGCTTTTGACG/GCAAGGTTGCATTGGGAACAG1952.0270.9964
TBPKF652111ATGGCTGAGCAAAGCATGGAG/CACAGCAGCAAAACGCTTAGGA1802.0240.9967
APTKF652112GCGCTAGCAATTGGTGCAAAA/GCAGCACATAAAGTGCCACCA1912.0590.9997
PCSKF652113TGTTGTGGCTCCAACCCTTG/TTGACGCAAGTGCATCACCTC 1921.9150.9990

Table 1. Candidate reference genes, primers and different parameters derived from RT-qPCR analysis.

CSV
Download CSV
thumbnail
Figure 1. Confirmation of primer specificity and amplicon size.

(a) Melting curves of ten candidate reference genes and one target gene showing single peaks. (b) Agarose gel (2%) showing specific RT-qPCR products of the expected size for each genes. M1 and M2 represent 2000bp and 500bp DNA marker respectively.

https://doi.org/10.1371/journal.pone.0082927.g001

Expression levels of the selected candidate reference genes

The average cycle threshold (Ct) values were calculated to reveal the differences in transcript levels among the reference genes with different heavy metals treatments. The ten candidates showed a relatively wide dispersal of transcript levels with average Ct values ranged from 19.29 for EF-1α to 26.26 for TIP41 (Figure 2), and most of them were moderately abundant (lying between 20 and 25). Among them EF-1α was the most abundant reference gene and TIP41 was the least one. The coefficient of variance (CV) values were also computed to reveal the expression stability of these reference genes. TBP had the least variation with a CV value of 2.32%, while GAPDH was the most variable one with a CV value of 6.53%.

thumbnail
Figure 2. RT-qPCR Ct values for reference genes.

Ct values for each reference gene in all Hyperaccumulating ecotype of S. alfredii samples. A line across the box depicts the median. The box indicates the 25% and 75% percentiles, whiskers caps represent the maximum and minimum values, dots represent outliers.

https://doi.org/10.1371/journal.pone.0082927.g002

Expression stability of candidate reference genes

Three different programs were applied to evaluate the expression stability of the selected reference gene candidates: geNorm [26], NormFinder [27] and BestKeeper [28]. To select stably expressed genes, we collected Ct values data across all samples. These Ct data were used directly for stability calculations for BestKeeper analysis, and then transformed to relative quantities using the delta-Ct method for geNorm and NormFinder analysis.

a): geNorm analysis.

The expression stability measure M for all selected reference genes was calculated by geNorm program based on the average pairwise variation between all the tested reference genes. The most stable reference gene has the lowest M value, whereas the least stable one has the highest M value [26]. In our analysis, all reference genes under different heavy metals stresses had an M value less than the geNorm threshold of 1.5 that could be recognized as stable (Figure 3). For treatments with Zn and Pb, UBC9 and TUB were the most stably expressed reference genes, which was consistent with the results from different plant organs. ACT2 and UBC9 were ranked as the most stable under Cd stress, while ACT2 and APT were ranked as the most stable under Cu stress. On the whole, UBC9 and TUB had the lowest M value with the most stability and TIP41 had the least stability across all samples. The discrepancy existed in stability ranking of reference genes under different heavy metals stresses revealed that virous strategies were taken by Hyperaccumulating ecotype of S. alfredii in response to specific heavy metal.

thumbnail
Figure 3. Expression stability (M) of ten reference genes across all 31 samples calculated by geNorm.

(a) all 31 samples, (b) different tissue/organs, (c) Cd2+ treatment, (d) Pb2+ treatment, (e) Zn2+ treatment, (f) Cu2+ treatment.

https://doi.org/10.1371/journal.pone.0082927.g003

To determine the optimal number of reference genes required for reliable normalization, geNorm was used to calculate their pairwise variation (Vn/Vn+1) between two sequential normalization factors NFn and NFn+1 (Figure 4). Analysis of the pairwise variation for all samples showed that the V2/3 value (0.209) and V3/4 (0.171) were higher than the recommended threshold of 0.15, which was universally accepted as the cutoff value [25]. When the fourth reference gene was included, V4/5 value (0.130) dropped below 0.15, indicating that four reference genes would be sufficient for proper gene expression normalization in all samples of Hyperaccumulating ecotype of S. alfredii.

thumbnail
Figure 4. Pairwise variation (V) calculated by geNorm to determine the optimal number of reference genes.

The average pairwise variations (Vn/Vn+1) were analyzed to measure the effect of adding additional reference gene on the RT-qPCR normalization for all samples.

https://doi.org/10.1371/journal.pone.0082927.g004

b): NormFinder analysis.

Another adopted program NormFinder is used to calculate the stability value of reference genes based on their intra- and inter-expression variation of expression [27]. Those exhibited lower average expression stability values are regarded as more stably expressed reference genes. The stability values for all the reference genes by NormFinder calculation were shown in Table 2. Although the stability ranking of the ten candidates with this method showed somewhat different from geNorm under specific heavy metals stresses, when considering all samples together, the five most stable reference genes: UBC9 (0.304), TUB (0.319), ACT2 (0.322), TBP (0.395) and AP-2 (0.459) were ranked consistently with geNorm. SAND and TIP41 were identified as the least stable candidates by both geNorm and NormFinder. Similar to geNorm program, NormFinder also displayed a difference in the stability ranking of reference genes with different heavy metal treatments.

RankTotalTissue/organsCd PbZn Cu
1UBC9 (0.304)TBP (0.094)UBC9 (0.143)UBC9 (0.093)TUB (0.110)TUB (0.207)
2TUB (0.319)ACT2 (0.304)TUB (0.178)ACT2 (0.161)APT (0.220)APT (0.268)
3ACT2 (0.322)TUB (0.438)ACT2 (0.415)TUB (0.188)UBC9 (0.297)ACT2 (0.274)
4TBP (0.395)AP-2 (0.438)TBP (0.514)TBP (0.338)TBP (0.37)AP-2 (0.315)
5AP-2 (0.459)UBC9 (0.517)EF-1α (0.551)APT (0.350)AP-2 (0.438)UBC9 (0.327)
6EF-1α (0.518)APT (0.563)APT (0.619)AP-2 (0.356)ACT2 (0.481)TBP (0.432)
7APT (0.559)EF-1α (0.570) AP-2 (0.782)TIP41 (0.602)GAPDH (0.523)EF-1α (0.694)
8GAPDH (0.659)GAPDH (0.605)TIP41 (0.793)GAPDH (0.621)EF-1α (0.549)GAPDH (0.742)
9TIP41 (0.759)SAND (0.679)SAND (0.897)SAND (0.748)SAND (0.711)TIP41 (1.002)
10SAND (0.761)TIP41 (0.734)GAPDH (0.914)EF-1α (0.806)TIP41 (0.806)SAND (1.054)

Table 2. Ranking of candidate reference genes in order of their expression stability as calculated by NormFinder.

CSV
Download CSV

c): BestKeeper analysis.

BestKeeper program is used to evaluate the expression stability of reference genes based on their coefficient of correlation (r) to the BestKeeper Index (BI) by the calculation of standard deviation (SD) and percentage covariance (CV) [28]. In the present study, the BestKeeper analysis revealed that the best correlations were obtained for TUB (0.945) and UBC (0.878) with P value of 0.001 across all samples, while SAND (0.247) with a P value as high as 0.181 was excluded (Table 3). However, the two unstable reference genes GAPDH and TIP41 identified by geNorm and NormFinder got relatively high correlation coefficient value of 0.878 and 0.866, which might result from the distinct adopted statistical algorithms. Therefore UBC9 and TUB were ranked as the most stable reference genes and SAND was ranked as the least one by three programs together in Hyperaccumulating ecotype of S. alfredii under different heavy metals stresses.

RankGene name rP-valueCV(%)SD
1TUB 0.950.0013.951.8
2UBC90.880.0013.291.6
3GAPDH0.870.0016.532.52
4TIP41 0.830.0014.492.33
5ACT2 0.800.0013.031.57
6EF-1α 0.790.0015.272.07
7AP-2 0.79 0.0013.771.92
8TBP0.720.0012.321.46
9APT0.670.0013.301.71
10SAND 0.250.1813.271.75

Table 3. Statistics results by BestKeeper program for ten selected genes based on Ct values.

CSV
Download CSV

Validation of the selected reference gene candidates

In order to ensure the reliability of the selected reference gene candidates, we analyzed the expression patterns of phytochelatin synthase (PCS) across all samples under different heavy metals stresses. The two most appropriate reference genes UBC9 and TUB selected by geNorm, NormFinder and BestKeeper were used for the RT-qPCR normalization of PCS, respectively. Phytochelatins, produced by the enzyme phytochelatin synthase (PCS), have been reported to be a class of cellular chelating agents which can play an important role in heavy metal detoxification, and it can be activated by various heavy metal ions [5]. The normalization of RT-qPCR using UBC9 and TUB generated similar expression patterns of PCS (Figure 5). An increased transcript level of PCS induced by heavy metals could be detected, while the specific expression patterns of PCS under Cd, Pb, Zn and Cu stresses were different from each other. The relative expression of PCS in different organs without any heavy metals treatments was also calculated. Its transcript was expressed relatively higher in roots and stems than leaves, and no significant difference was found when using the two different reference genes for normalization, respectively.

thumbnail
Figure 5. The expression quantification of PCS using UBC9 and TUB as internal controls under different heavy metals stresses for seven durations and in different tissue/organs.

(a) Cd2+ treatment, (b) Pb2+ treatment, (d) Zn2+ treatment, (d) Cu2+ treatment, (e) different tissue/organs without any heavy metals treatments.

https://doi.org/10.1371/journal.pone.0082927.g005

Discussion

Nowadays, RT-qPCR has become an effective method for gene expression patterns analysis in plant biological studies. Since the relative expression of target genes can be influenced significantly by reference genes, it is necessary to select ideal reference genes for the reliable and reproducible gene expression measurement. Some commonly used traditional housekeeping genes were doubted in recent studies for their variable expressions when tested with different experimental treatments [23]. The application of novel genes then became more and more popular for gene expression analysis with RT-qPCR [33-36]. However, our results showed that traditional housekeeping genes were expressed more stable than novel genes for RT-qPCR normalization under different heavy metals stresses in S. alfredii. Thus, the selection of reference genes across all samples under different experimental conditions is indeed an essential job before further gene expression analysis.

As a Cd/Zn/Pb co-hyperaccumulator, S. alfredii is well known for its potential utilization in the detoxification of contaminated soils by heavy metals [7-11]. Previous studies of the mechanism involved in heavy metals hyper-accumulation and hyper-tolerance for this plant were generally focused on physiological view, which was unable to reveal its intrinsic characteristic essentially. So far, no large-scale related molecular biological research has been carried out in S. alfredii due to limited available sequence information. Taking advantage of transcriptome data, a precious sequence resource for non-model plant systems, many heavy metals stresses-responsive genes will be identified with comprehensive functional analysis in the near future. In this study, we performed a selection and validation of reference genes for further accurate gene expression analysis using RT-qPCR in S. alfredii.

Three Excel-based programs geNorm, NormFinder and BestKeeper were used in the present study to evaluate the expression stability of ten selected reference gene candidates in the total of 31 samples. Those samples were extracted from Hyperaccumulator S. alfredii with four kinds of heavy metal treatments (CdCl2, Pb(NO3)2, CuSO4 and ZnSO4) for seven process durations: 0 h, 0.5 h, 6 h, 12 h, 24 h, 48 h and 72 h, respectively. Although with different statistical algorithms and analytical methods, same results were produced by three programs that UBC9 and TUB were the most stably expressed reference genes and SAND the least one across all samples. However, as the unstable candidates selected by both geNorm and NormFinder, GAPDH and TIP41 were ranked relatively stable by BestKeeper. This inconsistent result indicated that using geNorm and NormFinder in some cases might not be enough for the studies on reference genes selection. Although some previous studies only used the two programs to select reference genes successfully [29-32], we suggested that taking into consideration the stability ranking of BestKeeper could make the results more convincing. Besides the different stability ranking resulted from distinct algorithms, another discrepancy was observed in the stability ranking with four kinds of heavy metals treatments using same software, which might imply that diverse reactions appeared in Hyperaccumulator S. alfredii in response to different heavy metals stresses. In previous studies, ACT was often used as internal control for expression quantification of heavy metals stresses responsive genes in S. alfredii [20-22]. In our study, ACT was also regarded as stable reference genes, however, UBC9 and TUB were suggested to be better for RT-qPCR normalization in S. alfredii, which might provide more accurate expression measurement of target genes.

To validate the reference gene candidates selected by three programs, the transcript level of phytochelatin synthetase (PCS) under different heavy metals stresses were quantified using the two most stable reference genes UBC9 and TUB for RT-qPCR normalization, respectively. There was no obvious difference existed in the expressional trends of PCS generated by normalization of UBC9 compared with that by normalization of TUB for all samples, further indicating that both TUB and UBC9 were appropraite reference genes. A heated debate occurred several years ago on whether phytochelatin synthetase could be induced by heavy metals in hyperaccumulator S. alfredii. As reported by Sun et al., phytochelatin (PC) could not be detected using the physiological methods - HPLC under Cd, Pb and Zn stresses [36]. While Zhang et al. reported that PC formation could be induced by Cd and Pb, but not Zn stress with the same HPLC method [37]. Our expression analysis of PCS using RT-qPCR indicated that PCS could be induced by Cd, Pb, Zn and Cu treatments at transcript levels with different expression patterns in hyperaccumulator S. alfredii. This discrepancy might be resulted from distinct expression regulations of PCS between transcription and translation levels under different heavy metals stresses, which is worth for further exploration.

Conclusion

To sum up, five traditional housekeeping genes and five novel reference genes from Hyperaccumulator S. alfredii were evaluated under four kinds of different heavy metals stresses and in different organs using geNorm, NormFinder and BestKeeper programs. The results showed that UBC9 and TUB were the most appropriate reference genes for the RT-qPCR normalization across all samples. Furthermore, we validated the two optimal reference genes in all samples by quantifying the expression level of phytochelatin synthetase. This study constitutes the first systematic exploration in S. alfredii to identify optimal reference genes for RT-qPCR normalization under different heavy metals stresses.

Materials and Methods

Ethics Statement

These field studies did not involve any protected species. No specific permits were required for the collection of samples in the study location.

Plant materials and treatments

Hyperaccumulating ecotype of S. alfredii were collected from an old Pb/Zn mined area in Quzhou city, Zhejiang Province, PR China. After 14 d water-cultivation in artificial climate chest at 25~28°C under long day (16 hours light/8 hours dark) conditions, the plants were treated with 400 μM CdCl2, 400 μM Pb(NO3)2, 500 μM CuSO4 and 500 μM ZnSO4, respectively. Then the roots of the treated plants were sampled at 0 h, 0.5 h, 6 h, 12 h, 24 h, 48 h and 72 h. Three different tissue/organs including roots, stems and leaves of the plants without any heavy metals treatments were also sampled. Each treatment was replicated three times. All the samples were frozen in liquid nitrogen immediately after collection and then stored at - 80°C for further study.

Total RNA extraction and first strand cDNA synthesis

Total RNA was extracted using the Total RNA Purification Kit (NORGEN, Thorold, Canada) and then treated with DNase I (TaKaRa Bio, Dalian, China) to remove genomic DNA contamination according to the manufacturer’s recommendations. The integrity of total RNA was checked with denaturing 1.0% (p/v) agarose gel electrophoresis. A NanoDrop2000 spectrophotometer (Thermo, Wilmington, USA) was then used to determine the purity and concentration of the RNA samples. Only those with appropriate absorbance ratio of A260/A280 and A260/A230 were used for cDNA synthesis. After that, 1 μg of RNA sample was used for the first strand cDNA synthesis in a total volume of 20 μL with the superscript Ш first strand synthesis system followed by the RNase H step (Invitrogen, Carlsbad, USA). The cDNA products were diluted 30-fold prior to use in RT-qPCR.

PCR primer design

The sequences of ten reference gene candidates and one target gene were obtained from Hyperaccumulating ecotype of S. alfredii transcriptome data by BLAST search using the sequences of reported Arabidopsis reference genes with score value higher than 250 and identity value higher than 95%. Primer3 program (http://frodo.wi.mit. edu/primer3/) was used to design the primers pairs for all those genes with the optimal Tm at 58-60°C, product size ranging from 150 to 200 bp, primer length between 19 and 21 nucleotides and a GC content of 45 to 55%. The products of real-time PCR with a 2% agarose gel electrophoresis and melt curve analysis were conducted to check the specificity of all primers.

RT-qPCR

RT-qPCR amplification was performed in 96-well plates using a 7300 Real Time PCR System (Applied Biosystems, CA, USA) with SYBR® Premix Ex TaqTM Kit (TaKaRa Bio, Dalian, China) according to the manufacturer’s protocol. Each PCR reaction was performed in 20 μl volumes containing: 2 μl cDNA, 10 μl SYBR® Premix Ex TaqTM, 0.4 μl of both 10 μM forward primer and reverse primer, 6.8 μl PCR-grade water and 0.4 μl ROX reference dye. The standard thermal profile used in RT-qPCR was as follows: 95°C for 10 s,40 cycles of 95°C for 5 s and 60°C for 31 s. After that, a dissociation curve was generated (from 60°C to 95°C) to check the specificity of amplicons. Threshold cycles (Ct) were automatically determined and each PCR reaction was performed in triplicate.

Analysis of gene stability

Standard curves were used to calculate the gene-specific PCR efficiency from 5-fold series dilution of the mixed cDNA template for each primer pair. The correlation coefficients (R2) and slope values were acquired from the standard curve, and PCR amplification efficiencies (E) were calculated according to the equation E = (5-1/slope - 1) ×100%.

Three Excel-based softwares geNorm, NormFinder and BestKeeper were used to evaluate the expression stability of all the candidate genes. The geNorm software is used to calculate the expression stability value (M) for reference genes based on the mean pairwise variation between all the tested genes (http://medgen.ugent.be/~jvdesomp/genorm/), and NormFinder is used to rank the expression stability of reference genes according to their intra- and interexpression variation of expression (http://www.mdl.dk/publicationsnormfinder.htm). While BestKeeper uses an index for the evaluation of expression stability which is calculated by standard deviation (SD) and percentage covariance (CV) (http://www.gene-quantification.de/bestkeeper.html).

Supporting Information

File S1.

Sequence of the RT-qPCR products for the eleven selected genes (Their primers sequences were highlighted).

https://doi.org/10.1371/journal.pone.0082927.s001

(DOC)

Author Contributions

Conceived and designed the experiments: JS XH RZ. Performed the experiments: JS XH GQ. Analyzed the data: JS ML XH. Contributed reagents/materials/analysis tools: ML RZ JJ. Wrote the manuscript: JS XH.

References

  1. 1. Diels L, Van der Lelie N, Bastiaens L (2002) New developments in treatment of heavy metal contaminated soils. Rev Environ Sci Biotechnol 1: 75-82. doi:https://doi.org/10.1023/A:1015188708612.
  2. 2. Volesky B, Holan Z (1995) Biosorption of heavy metals. Biotechnol Progress 11: 235-250. doi:https://doi.org/10.1021/bp00033a001.
  3. 3. Ali H, Khan E, Sajad MA (2013) Phytoremediation of heavy metals-Concepts and applications. Chemosphere.
  4. 4. McGrath S, Zhao J, Lombi E (2002) Phytoremediation of metals, metalloids, and radionuclides. Adv Agron 75: 1-56.
  5. 5. McGrath SP, Zhao FJ (2003) Phytoextraction of metals and metalloids from contaminated soils. Curr Opin Biotechnol 14: 277-282. doi:https://doi.org/10.1016/S0958-1669(03)00060-0. PubMed: 12849780.
  6. 6. Visioli G, Marmiroli N (2013) The proteomics of heavy metal hyperaccumulation by plants. J Proteomics 79: 133-145. doi:https://doi.org/10.1016/j.jprot.2012.12.006. PubMed: 23268120.
  7. 7. Yang X, Long X, Ni W, Fu C (2002) Sedum alfredii H: a new Zn hyperaccumulating plant first found in China. Chinese Sci Bull, 47(19): 1634-1637. doi:https://doi.org/10.1360/02tb9359.
  8. 8. Lu L, Tian S, Zhang M, Zhang J, Yang X et al. (2010) The role of Ca pathway in Cd uptake and translocation by the hyperaccumulator Sedum alfredii. J Hazard Mater 183: 22-28. doi:https://doi.org/10.1016/j.jhazmat.2010.06.036. PubMed: 20674155.
  9. 9. Lu LL, Tian SK, Yang XE, Wang XC, Brown P et al. (2008) Enhanced root-to-shoot translocation of cadmium in the hyperaccumulating ecotype of Sedum alfredii. J Exp Bot 59: 3203-3213. doi:https://doi.org/10.1093/jxb/ern174. PubMed: 18603654.
  10. 10. Tian S, Lu L, Labavitch J, Yang X, He Z et al. (2011) Cellular sequestration of cadmium in the hyperaccumulator plant species Sedum alfredii. Plant Physiol 157: 1914-1925. doi:https://doi.org/10.1104/pp.111.183947. PubMed: 22025609.
  11. 11. Yang X, Long X, Ye H, He Z, Calvert D et al. (2004) Cadmium tolerance and hyperaccumulation in a new Zn-hyperaccumulating plant species (Sedum alfredii Hance). Plant Soil 259: 181-189. doi:https://doi.org/10.1023/B:PLSO.0000020956.24027.f2.
  12. 12. Krämer U (2010) Metal hyperaccumulation in plants. Annu Rev Plant Biol 61: 517-534. doi:https://doi.org/10.1146/annurev-arplant-042809-112156. PubMed: 20192749.
  13. 13. Gao J, Sun L, Yang X et al. (2013) Transcriptomic Analysis of Cadmium Stress Response in the Heavy Metal Hyperaccumulator Sedum alfredii Hance. PLOS ONE, 8(6): e64643. doi:https://doi.org/10.1371/journal.pone.0064643. PubMed: 23755133.
  14. 14. Bustin SA (2002) Quantification of mRNA using real-time reverse transcription PCR (RT-PCR): trends and problems. J Mol Endocrinol 29: 23-39. doi:https://doi.org/10.1677/jme.0.0290023. PubMed: 12200227.
  15. 15. Bustin SA, Benes V, Nolan T, Pfaffl MW (2005) Quantitative real-time RT-PCR–a perspective. J Mol Endocrinol 34: 597-601. doi:https://doi.org/10.1677/jme.1.01755. PubMed: 15956331.
  16. 16. Gachon C, Mingam A, Charrier B (2004) Real-time PCR: what relevance to plant studies? J Exp Bot 55: 1445-1454. doi:https://doi.org/10.1093/jxb/erh181. PubMed: 15208338.
  17. 17. Huggett J, Dheda K, Bustin S, Zumla A (2005) Real-time RT-PCR normalisation; strategies and considerations. Genes Immun 6: 279-284. doi:https://doi.org/10.1038/sj.gene.6364190. PubMed: 15815687.
  18. 18. Udvardi MK, Czechowski T, Scheible WR (2008) Eleven golden rules of quantitative RT-PCR. Plant Cell 20: 1736-1737. doi:https://doi.org/10.1105/tpc.108.061143. PubMed: 18664613.
  19. 19. Brunner AM, Yakovlev IA, Strauss SH (2004) Validating internal controls for quantitative plant gene expression studies. BMC Plant Biol 4: 14. doi:https://doi.org/10.1186/1471-2229-4-14. PubMed: 15317655.
  20. 20. Chao Y, Zhang M, Feng Y, Yang X, Islam E (2010) cDNA-AFLP analysis of inducible gene expression in zinc hyperaccumulator Sedum alfredii Hance under zinc induction. Environ Exp Bot 68: 107-112. doi:https://doi.org/10.1016/j.envexpbot.2009.11.013.
  21. 21. Zhang M, Senoura T, Yang X, Chao Y, Nishizawa NK (2011) Lhcb2 gene expression analysis in two ecotypes of Sedum alfredii subjected to Zn/Cd treatments with functional analysis of SaLhcb2 isolated from a Zn/Cd hyperaccumulator. Biotechnol Lett 33: 1865-1871. doi:https://doi.org/10.1007/s10529-011-0622-8. PubMed: 21516315.
  22. 22. Zhang M, Senoura T, Yang X, Nishizawa NK (2011) Functional analysis of metal tolerance proteins isolated from Zn/Cd hyperaccumulating ecotype and non-hyperaccumulating ecotype of Sedum alfredii Hance. FEBS Lett 585: 2604-2609. doi:https://doi.org/10.1016/j.febslet.2011.07.013. PubMed: 21781966.
  23. 23. Gutierrez L, Mauriat M, Guénin S, Pelloux J, Lefebvre JF et al. (2008) The lack of a systematic validation of reference genes: a serious pitfall undervalued in reverse transcription-polymerase chain reaction (RT‐PCR) analysis in plants. Plant Biotechnol J 6: 609-618. doi:https://doi.org/10.1111/j.1467-7652.2008.00346.x. PubMed: 18433420.
  24. 24. Hong SY, Seo PJ, Yang MS, Xiang F, Park CM (2008) Exploring valid reference genes for gene expression studies in Brachypodium distachyon by real-time PCR. BMC Plant Biol 8: 112. doi:https://doi.org/10.1186/1471-2229-8-112. PubMed: 18992143.
  25. 25. Rubie C, Kempf K, Hans J, Su T, Tilton B et al. (2005) Housekeeping gene variability in normal and cancerous colorectal, pancreatic, esophageal, gastric and hepatic tissues. Mol Cell Probes 19: 101-109. doi:https://doi.org/10.1016/j.mcp.2004.10.001. PubMed: 15680211.
  26. 26. Vandesompele J, De Preter K, Pattyn F, Poppe B, Van Roy et al (2002) Accurate normalization of real-time quantitative RT-PCR data by geometric averaging of multiple internal control genes. Genome Biol, 3(7): research0034. PubMed: 12184808.
  27. 27. Andersen CL, Jensen JL, Ørntoft TF (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(15): 5245-5250. doi:https://doi.org/10.1158/0008-5472.CAN-04-0496. PubMed: 15289330.
  28. 28. Pfaffl MW, Tichopad A, Prgomet C, Neuvians TP (2004) Determination of stable housekeeping genes, differentially regulated target genes and sample integrity: BestKeeper–Excel-based tool using pair-wise correlations. Biotechnol Lett, 26(6): 509-515. doi:https://doi.org/10.1023/B:BILE.0000019559.84305.47. PubMed: 15127793.
  29. 29. Mafra V, Kubo KS, Alves-Ferreira M, Ribeiro-Alves M, Stuart RM et al. (2012) Reference genes for accurate transcript normalization in citrus genotypes under different experimental conditions. PLOS ONE, 7(2): e31263. doi:https://doi.org/10.1371/journal.pone.0031263. PubMed: 22347455.
  30. 30. Li R, Xie W, Wang S, Wu Q, Yang N et al. (2013) Reference gene selection for qRT-PCR analysis in the sweetpotato whitefly, Bemisia tabaci (Hemiptera: Aleyrodidae). PLOS ONE, 8(1): e53006. doi:https://doi.org/10.1371/journal.pone.0053006. PubMed: 23308130.
  31. 31. Hu R, Fan C, Li H, Zhang Q, Fu YF (2009) Evaluation of putative reference genes for gene expression normalization in soybean by quantitative real-time RT-PCR. BMC Mol Biol, 10(1): 93. doi:https://doi.org/10.1186/1471-2199-10-93. PubMed: 19785741.
  32. 32. Expósito-Rodríguez M, Borges AA, Borges-Pérez A, Pérez JA (2008) Selection of internal control genes for quantitative real-time RT-PCR studies during tomato development process. BMC Plant Biol, 8(1): 131. doi:https://doi.org/10.1186/1471-2229-8-131. PubMed: 19102748.
  33. 33. Kumar V, Sharma R, Trivedi PC, Vyas GK, Khandelwal V (2011) Traditional and novel references towards systematic normalization of qRT-PCR data in plants. Aust J Crop Sci, 5: 1455-1468.
  34. 34. Mafra V, Kubo KS, Alves-Ferreira M, Ribeiro-Alves M, Stuart RM (2012) Reference genes for accurate transcript normalization in citrus genotypes under different experimental conditions. PLOS ONE, 7(2): e31263. doi:https://doi.org/10.1371/journal.pone.0031263. PubMed: 22347455.
  35. 35. Han X, Lu M, Chen Y, Zhan Z, Cui Q et al. (2012) Selection of reliable reference genes for gene expression studies using real-time PCR in tung tree during seed development. PLOS ONE, 7(8): e43084. doi:https://doi.org/10.1371/journal.pone.0043084. PubMed: 22912794.
  36. 36. Sun Q, Ye ZH, Wang XR, Wong MH (2007) Cadmium hyperaccumulation leads to an increase of glutathione rather than phytochelatins in the cadmium hyperaccumulator Sedum alfredii. J Plant Physiol, 164(11): 1489-1498. doi:https://doi.org/10.1016/j.jplph.2006.10.001. PubMed: 17207552.
  37. 37. Zhang Z, Gao X, Qiu B (2008) Detection of phytochelatins in the hyperaccumulator Sedum alfredi exposed to cadmium and lead. Phytochemistry, 69(4): 911-918. doi:https://doi.org/10.1016/j.phytochem.2007.10.012. PubMed: 18023461.