Stable Reference Gene Selection for RT-qPCR Analysis in<i> Synechococcus elongatus</i> PCC 7942 under Abiotic Stresses

Luo, Xiao; Li, Jiaxing; Chang, Tianliang; He, Hongyan; Zhao, Yi; Yang, Xue; Zhao, Yuwei; Xu, Yao

doi:https://doi.org/10.1155/2019/7630601

BioMed Research International

On this page

Abstract Introduction Materials and Methods Results Discussion Conclusion Abbreviations Data Availability Conflicts of Interest Authors’ Contributions Acknowledgments Supplementary Materials References Copyright Related Articles

Research Article | Open Access

Volume 2019 | Article ID 7630601 | https://doi.org/10.1155/2019/7630601

Stable Reference Gene Selection for RT-qPCR Analysis in Synechococcus elongatus PCC 7942 under Abiotic Stresses

Xiao Luo,^1,2,3Jiaxing Li,^1,2,3Tianliang Chang,^1,2,3Hongyan He,^1,2,3Yi Zhao,^1,2,3Xue Yang,^1,2Yuwei Zhao ,^1,2,3and Yao Xu^1,2,3

Academic Editor: Mansour El-Matbouli

Received10 Jan 2019

Revised15 Mar 2019

Accepted07 Apr 2019

Published21 Apr 2019

Abstract

Synechococcus elongatus PCC 7942 (S. elongatus PCC 7942) is a model cyanobacteria species for circadian clock mechanism studies. It has also been widely used as a bioreactor to produce biofuels and other metabolic products. Quantitative real-time PCR (qPCR) technology is the most commonly used method for studying the expression of specific genes, in which the relative expression level of target genes is calibrated by stably expressed internal reference genes. In this work, we examined the expression of nine candidate reference genes in time-course samples of S. elongatus PCC 7942 under no treatment (control), NaCl-stress conditions, H₂O₂-stress conditions, and high light-stress conditions. Based on the qPCR amplification parameters, the stability ranking of these candidate reference genes was established by three statistical software programs, geNorm, NormFinder, and BestKeeper. Considering all the stress conditions or high light stress alone, the results showed that the combination of prs and secA was the best choice for the double reference gene calibration method by qPCR. The combination of secA and ppc, rimM and rnpA, rnpA, and ilvD was most stable under no treatment, NaCl-stress conditions, and H₂O₂-stress conditions, respectively. rimM was stable under only special conditions and should be carefully chosen. 16S and rnpB were not suitable as internal reference genes for S. elongatus PCC 7942 qPCR experiments under all experimental conditions. To validate the above results, a cyanobacterial core clock gene, kaiC, was used to evaluate the actual performance of the optimized reference genes by qPCR, as well as the worst reference genes under different stress conditions. The results indicated that the best reference gene yielded more accurate calibration results for qPCR experiments carried out in S. elongatus PCC 7942 time-course samples.

1. Introduction

Synechococcus elongatus PCC 7942 (S. elongatus PCC 7942) is a freshwater single cell photosynthetic autotrophic prokaryote [1, 2]. Its 2.74 Mb genome constitutes 2,715 genes and encodes approximately 2,612 proteins [3]. S. elongatus PCC 7942 is also the simplest model organism for studying circadian rhythms [4, 5], with approximately 30% of its genes showing circadian rhythmic expression profiles that peak at dawn or dusk [6, 7]. The cyanobacterial clock is controlled by genetic, physiological, and metabolic processes, such as cell division, nitrogen fixation, photosynthesis, amino acid uptake, carbohydrate synthesis, and respiration [8, 9]. The central oscillator of S. elongatus PCC 7942 is composed by three clock proteins, KaiA, KaiB, and KaiC [10]. Hexameric KaiC proteins, the key components of the S. elongatus PCC 7942 clock, have autokinase, phosphotransferase, and ATPase activities and display a circadian rhythmic oscillation of phosphorylation and dephosphorylation every day [11]. Interestingly, circadian rhythmic oscillations due to KaiC phosphorylation can be reconstituted in vitro when purified KaiA, KaiB, KaiC proteins and an appropriate amount of ATP are incubated together [12]. During the growth of cyanobacteria, the most common stresses come from various environmental abiotic factors, such as salinity [13], light stress [14], and accumulated peroxide [15]. To understand the effects of different types of stress on cyanobacteria, it is necessary to evaluate the expression patterns of cyanobacteria exposed to different stress conditions.

Quantitative real-time PCR (qPCR) technology can be used to monitor the production and accumulation of amplification products in each cycle of the PCR reaction by adding a fluorescent dye or a fluorescent probe to the PCR reaction. This method is characterized by its simplicity, sensitivity, high specificity and high throughput features. qPCR has become a basic and standard technique for detecting or comparing the level of a defined gene in a given experimental condition [16, 17]. However, many parameters in a qPCR experiment can affect the authenticity of the experimental results, such as the quality of the sample RNA, the efficiency of the reverse transcription reaction, and the stability of the internal reference gene. Standardized qPCR requires the elimination of systematic errors between different sample groups as much as possible, so that the final experimental results reveal the closest biological changes [18]. This makes it particularly important to reduce the variability of the experimental parameters and to enhance the stability of the experimental setup when using large-scale sample populations for analyzing specific gene expression levels, such as using the qPCR technique, to perform rhythmic expression detection of defined gene populations in time-course samples. The internal reference gene used to calibrate the expression of the target gene is one of the key technical means for achieving qPCR standardization and increasing the stability of the experimental system. Although commonly used internal reference genes are usually the so-called housekeeping genes necessary to maintain cell survival [19], the results of earlier studies show that the expression levels of housekeeping genes under different experimental treatments and environmental conditions significantly fluctuate [20]. When calibrating the expression level of a target gene, the use of relatively unstable housekeeping genes may lead to misinterpretation of the real results [21]. Therefore, when qPCR technology is used, not only the appropriate biological repeats should be set, the quality of RNA samples and reverse transcription synthesis cDNA templates should be strictly controlled, and the optimal internal reference genes under experimental conditions should be carefully screened [22, 23].

The prescreening of housekeeping genes, which will be used as internal reference genes in qPCR, is mainly carried out in studies of eukaryotes. Except for a few reports in the cyanobacteria Synechocystis sp. PCC 6803 [24] and Synechococcus sp. PCC 7002 [25], the screening of qPCR internal reference gene in other prokaryotes, including S. elongatus PCC 7942, has not been reported. When using qPCR for transcriptional studies in cyanobacteria, the most commonly used internal reference gene encoding is 16s RNA (16S ribosomal RNA) because it is highly conserved during evolution [26–29]. In addition, rnpA [24, 30], rnpB [24, 31–33], ppc [34], ilvD [24], secA [24, 25], petB [24, 25, 30], rim [25], and prs [24] are also commonly used internal reference genes in cyanobacterial transcriptome research. In recent years, various algorithms for analyzing the stability of the expression levels of housekeeping genes have been developed. Currently, there are various free software packages that are available online, such as geNorm [35], NormFinder [36], BestKeeper [37], ΔCt approach [38], and RefFinder [39]. Based on the Ct values of the qPCR reaction from individual candidate reference genes, the corresponding stability evaluation is given according to a similar algorithm, though different software packages designate different weight values to the core parameters, such as the Ct value, amplification efficiency, and intergroup difference of the qPCR reaction. The final ranking results of the candidate reference genes could also be inconsistent between different software programs. Therefore, it is necessary to use two or more algorithms simultaneously to identify the most stably expressed reference gene [39]. Our current study used RT-PCR to detect the expression patterns of nine candidate reference genes (shown in Table 1) under NaCl, high light, and H₂O₂ conditions in the cyanobacteria S. elongatus PCC 7942. Three kinds of software, namely, geNorm, NormFinder, and BestKeeper, were used to evaluate the stability of these candidate internal reference genes for qPCR experiments under various stress conditions. Finally, we used a core circadian clock gene kaiC [40, 41] to verify the optimal reference genes screened by the software programs. The relevant research results provide an experimental basis for further study of gene expression analysis in S. elongatus PCC 7942 under stress conditions.

2. Materials and Methods

2.1. Strains, Culture Conditions, and Sample Collection

Wild-type S. elongatus PCC 7942 (R² strain) was purchased from the Wuhan Institute of Aquatic Biology, Chinese Academy of Sciences. The S. elongatus PCC 7942 cells were cultured in 150 ml fresh BG11 (Haibo, Qingdao, China) liquid medium with gentle shaking (100 rpm) at 30°C under 50 μE•m⁻²•s⁻¹ continuous light provided by cool-white fluorescent bulbs. When the OD₇₅₀ of cyanobacteria cultures reached approximately 0.3, the cultures were divided into four groups, namely, no treatment control (NTC), NaCl-treated group (NaCl), H₂O₂-treated group (HO), and high light intensity-treated group (HL). Three biological replicates were set up in each group. After being entrained with two 12-h light/12-h dark cycles at 30°C, the cyanobacteria cultures were treated with 1% NaCl (w/v) or 1 mM H₂O₂ in fresh liquid BG11 medium for 48 h. For the HL group, the entrained cyanobacteria cells were cultured under 300 μE•m⁻²•s⁻¹ for 48 h after entrainment. Subsequently, 10 ml of the cyanobacteria cell suspension from each group was centrifuged at 4,500 rpm for 15 min and stored in a −80°C freezer for the following experiments. Time-course samples from NTC, NaCl, HO, and HL groups were continually collected every 4 h for 48 h.

2.2. Total RNA Extraction and cDNA Synthesis

The frozen time-course cyanobacteria samples were resuspended and incubated in 100 μL of lysozyme solution (RNase-free, Sangon, Shanghai, China) for 10 min at RT, according to the manufacturer’s instructions. Total RNA from the digested samples was isolated using a Bacterial Total RNA Extraction and Purification Kit (Sangon). To detect the quality total RNA samples, all RNA samples were detected by 1.5% agarose gel electrophoresis. As shown in Figure S1, the total RNA samples obtained by the method reported in these work were qualified to be the templates for following reverse transcription and qPCR reactions. The A₂₆₀/A₂₈₀ ratios of total RNA samples varied from 2.0 to 2.1, while their A₂₆₀/A₂₃₀ ratios were around 1.8 (as shown in Figure S2). All these data indicated that the quality of total RNAs was guaranteed and could completely meet the requirement of the following qPCR experiments. The total RNA samples were simultaneously treated with 1 μL/mL Dnase I (Rnase-free, Takara, Dalian, China) and RNase Inhibitor (Takara). The total RNA samples were diluted to a final concentration of 1.0 μg/μL and used as templates for cDNA synthesis using the PrimeScript™ II First Strand cDNA Synthesis Kit (TaKaRa), according to the manufacturer’s instructions. A universal DNA Purification Kit (TIANGEN, China) was used to purify cDNA products.

2.3. Primer Design and Amplification Characteristics of Candidate Reference Genes

Nine candidate reference genes from various species of cyanobacteria were selected from previous reports. The sequences of the homologous internal reference genes were downloaded from the database CYORF http://cyano.genome.jp/. The detailed information for these candidate reference genes is shown in Table 1. Software Primer Premier (version 5.0) was used to design the primers for qPCR amplification. All primers were designed to meet the following standards: a melting temperature (Tm) between 50 and 65°C, a primer length between 19 and 25 bp, a product length between 100 and 300 bp, and a GC content of 50–60%. The specificity of each primer pair was tested via semiquantitative RT-PCR using cDNA from NTC samples as the templates, whereas the Premix Taq™ Kit (TaKaRa) was used to catalyze the amplification reactions. The semiquantitative RT-PCR amplified product for each candidate gene was separated on 1% agarose gel and visualized using a GelDoc XR System (Bio-Rad, Hercules, CA, USA).

2.4. The Performance of qPCR

qPCR was carried out on a Bio-Rad CFX96 Real-time PCR System (Bio-Rad), according to the manufacturer’s instructions. The 10 μL reaction was composed of 5 μL of 2×SYBR Premix Ex TaqII (TaKaRa), 0.25 μL (20 μM) of each upstream primer and downstream primer, 3.5 μL of ddH₂O, and 1 μL of appropriately diluted cDNA template. A two-step RT-qPCR was performed using the following protocol: predenaturation at 95°C for 3 minutes, followed by denaturation at 95°C for 10 seconds and annealing at the appropriate temperature for 30 seconds (the individual optimized annealing temperature for the different candidate reference genes is listed in Table 1), which was repeated for 40 cycles. To confirm the specificity of the primers, melting curve analysis was carried out during the heating phase from the individual optimized annealing temperatures to 95°C at the end of each cycle. Three replicates were set for the cDNA of each biological sample, and the final Ct value of each individual experiment was counted by the average of three independent biological replicates from different batches of experiment, whereas each biological replicate was represented by an average of three parallel technical replicates from the same cDNA templates. The Ct values from all samples were uploaded to GraphPad Prism v6.04 Software to generate box-and-whisker plots, which could indicate the Ct value within the 25th and 75th percentiles (Figure 1). Whiskers of each sample covered the Ct values in the 10th and 90th percentiles, whereas a horizontal solid line in the box represented the median.

The E value corresponded to the efficiency of a given qPCR amplification [22]. To determine the E value of a qPCR reaction, we prepared a 1:10 serial dilution of the original cDNA template (diluted from 10⁻² to 10⁻⁸). Subsequently, each candidate gene was amplified using the diluted cDNA template using standard qPCR. The E value and the regression coefficient (R²) of qPCRs were calculated according to the dilution gradients and the Ct values obtained from the individual qPCRs. The E value was calculated by the following equation: (Table 1). Uniformly spaced amplification curves produced a linear standard curve with a reaction efficiency of 90–110% for qPCR [22].

2.5. Analysis of Reference Gene Expression Stability

The reference gene stability analysis software programs geNorm (version 3.5), NormFinder (version 0.953), and BestKeeper (version 1.0) were used to analyze the performance of the candidate housekeeping genes under different stress conditions. The geNorm software determined the two best internal reference genes by determining their stability values, represented as M values. The M value of each candidate reference gene was calculated by comparing the average standard deviation between the log-transformed expression ratios of the target genes to that of another candidate gene [42]. The reference gene with the lowest M value was rejected for further analysis. This comparison process between the candidates was repeated until the two most stable reference genes were identified [22]. An M value of 1.5 was used as a criterion for evaluating the stability of a tested reference gene. When the M value of a candidate was lower than 1.5, it was considered an internal reference gene. When the M value was higher than 1.5, the candidate gene was excluded. In addition, the pairwise variation comparisons between two consecutive normalization factors () determined the number of optimal internal reference genes that were needed in the qPCR, with 0.15 set as a threshold. When the value of was lower than 0.15, it indicates that less than n additional reference genes were necessary for the following target gene calibration processes [35]. NormFinder was used to assess the stability of the reference gene expression based on the intra- and intergroup variations among the candidates. With a similar criteria to geNorm but calculated by different strategies, NormFinder chose the gene with lowest standard deviation as the most stable reference gene [43], that is, the gene with the lowest M value was considered as the most stable one. BestKeeper is a Microsoft Excel based tool for identifying the most suitable reference gene by pairwise correlation analysis among the candidates [37]. Descriptive statistical results of Ct values of all reference genes under different experimental conditions, including standard deviation (SD) and variance coefficient (CV), were calculated. The Pearson correlation coefficient, designated as r, for each candidate internal reference gene pair was calculated. A highly correlated candidate reference gene with a low standard deviation (SD < 1) was combined into the index value (i.e., a normalization factor) using a geometric mean of its Ct values, and each candidate reference gene and index value were determined by BestKeeper software (the correlation coefficient r value was defined as the BestKeeper Index), and the probability (P) value. The gene having the highest r value and a p-value below 0.05 was considered to be the most stable.

3. Results

3.1. The Candidate Reference Genes for qPCR

Nine common reference genes in cyanobacteria, 16S, petB, ppc, ilvD, prs, rnpA, rnpB, secA, and rimM, were selected to be evaluated as candidate references for the analysis of gene expression under stress conditions (NaCl, H₂O₂, and high light stress). The applicability of the primers designed for the qPCR amplification of these nine candidate genes were verified by RT-PCR and the sequencing of the amplified products (Figure S3). Subsequently, the specificity of the qualified primers was tested by unimodal melting curve analysis (Figure S4). The information of all nine candidates, such as their gene IDs, gene definitions, PCR primer sequences, size of the PCR products, Tm values, amplification efficiencies, correlation coefficients (R²), and median Ct values, is listed in Table 1. The data for KaiC, the target gene for verifying of candidate references, is listed in Table S1. In general, the sizes of the qPCR products of all the tested genes ranged from 102 bp to 296 bp, with efficiency E values ranging from 90.4% to 105.1%, and the R² distributing from 0.996 to 0.999. As shown in the box-and-whisker plot in Figure 1, the distribution of the Ct values from the nine reference genes under different conditions varied from 6.31 to 30.91. The 16S RNA gene encoded the most abundant transcripts in cyanobacteria S. elongatus PCC 7942, and the 16S RNA had the lowest median Ct value (Ct<10, Table 1) compared to other candidate reference genes during qPCR amplification when the same amounts of total RNA were used as templates for reverse transcription reactions.

3.2. Analysis Based on geNorm Software

Processed by geNorm software, the M values of all the candidate reference genes were used to determine the rank of their expressing stabilities. The threshold of the M value was set at 1.5. A smaller M value was indicative of more stable gene expression. However, M values higher than 1.5 caused the rejection of the candidate to be a qualified internal reference gene. According to result of the geNorm statistical analysis, the M values of all the candidate internal reference genes were lower than 1.5 under all the tested experimental conditions (Figure 2). When the data from the four tested conditions were counted, secA and prs had the lowest M value of 0.505 and were the most stably expressed candidate genes. Furthermore, the M values of 0.300 and 0.265 under NTC and HL, the lowest score among all the candidates, rendered secA and prs as the most suitable reference genes under these situations. When NaCl and HO were the stress factors, prs and ppc (M under NaCl=0.237 and M under HO=0.321) were the most stable reference genes by geNorm software. Therefore, either under individual experimental conditions or all four conditions, prs was the most stably expressed candidate reference gene suggested by geNorm software. rnpB was the most unstable candidate under NaCl or HL, and rimM was the most unstable candidate under NTC, whereas 16S was the most unstable under HO.

(a)

(b)

(c)

(d)

(e)

Additionally, geNorm software can also estimate the minimal numbers of the necessary reference genes in a reliable qPCR reaction by a pairwise variation comparison between the reference genes (). If the was less than 0.15, it indicated that the using of n+1 reference genes to calibrate the relative expression of target genes was unnecessary. As shown in Figure 3, under NTC, NaCl, HO, and HL conditions, the V_2/3 values were all below 0.15 (0.1390, 0.1353, 0.1002, and 0.1154, respectively), indicating that the combined use of two reference genes was necessary for qPCR experiments in cyanobacteria under all tested stress conditions. In this study, the best reference gene pairs under NTC and HL conditions given by geNorm were secA and prs. Under NaCl and HO stress, the best gene pairs are prs and ppc.

3.3. Analysis Based on NormFinder Software

When the expression stability of candidate reference genes was evaluated by NormFinder analysis, a lower stability value was indicative of a more stably expressed reference gene. The ranking results of the candidate genes suggested by the NormFinder are shown in Table 2. When all data from the four experimental groups were evaluated comprehensively, prs and secA were the most stable genes, which was consistent with the results obtained by geNorm software. When the NTC condition was considered, secA and ppc were the most stable reference genes. By contrast, the most stably expressed reference genes were rimM and ilvD under NaCl and rnpA and ilvD under HO stress. In regard to HL stress, rnpA and prs were the most stable candidates. rnpB displayed the most unstable expression in either NaCl or HL groups, though 16S is the most unstable candidate under HO stress. In the NTC group, rimM was the most unstable gene under NTC conditions, but at the same time this gene was the most stable under NaCl conditions. This contradiction indicated that the stable expression of a so-called housekeeping gene under normal growth conditions, such as rimM, could be severely changed by environmental stresses such as salt in this work.

Under the three stress conditions, rimM, ilvD, and rnpA shared much higher ranks arranged by NormFinder than that obtained via geNorm analysis. A possible reason for these differences between these two sets of software can be ascribed to their different build-in algorithms and weighting methods to the same row data based on the Ct value from the qPCR amplifications. In summary, similar ranking results of the candidate genes were obtained by NormFinder and geNorm software.

3.4. Analysis Based on BestKeeper Software

BestKeeper is another popular software, which has been used to optimize the reference genes in qPCR. Unlike the aforementioned software programs, BestKeeper uses a unique indicator, designated as the BestKeeper Index, to evaluate the stability of all candidate reference genes [37]. In this work, the BestKeeper Index performances of the candidate reference genes suggested by the BestKeeper software analysis are ranked and listed in Table 3. From the combined results of all the experimental groups, the top five candidate genes in the list, namely, secA, prs, 16S, petB, and rnpA showed strong correlation (r>0.92), and the P-values of these candidates were 0.001. When only the NTC condition was concerned, 16S and secA showed strong correlations with BestKeeper Index (r>0.90). Under NaCl stress, only petB had r>0.90. However, under HO and HL stress, the top three genes had a very strong correlation with the BestKeeper Index (r>0.95). A high r value indicated that these candidate genes exhibited highly similar stable expression patterns. In contrast, the r value of rnpB in each group (0.477<r<0.745) was much lower. In addition, the p-value (0.003<p<0.099) was higher than that for the other candidate genes. Thus, rnpB was considered to be the least reliable reference gene under the preset conditions of the present study in S. elongatus PCC 7942. The performance of rimM varied dramatically under the different conditions, which was verified to be one of the most unstable candidate genes under NTC stress (r=0.331, p=0.271) and HO stress (r = 0.506, p = 0.077), but also was ranked into the top three confident references under NaCl stress (r = 0.882, p = 0.001).

3.5. Comprehensive Stability Analysis of Candidate Reference Genes

Due to the fact that the results from the three software programs were inconsistent, we conducted a comprehensive assessment of all the analyzed data by rearranging the order of the nine candidate reference genes based on the basic results of geNorm, NormFinder, and BestKeeper evaluation. Firstly, the optimal number of essential reference genes in a qPCR reaction was determined to be two, according to the geNorm analysis shown in Tables 4 and 5. In regard to the data integrated from all the four tested conditions or the data under HL stress alone, prs and secA were the most stable expressed reference genes. Under NTC conditions, secA and ppc were the most stable genes. In terms of the NaCl group, rimM and rnpA were the best reference genes, whereas under HO conditions, rnpA and ilvD were considered to be the best. On the other hand, rimM was the most unstable gene under NTC and HO conditions, whereas rnpB was found to be the most unstable expressed candidate gene under NaCl and HL.

3.6. Validation Test of the Chosen Reference Genes

To verify the validation and applicability of the optimized reference genes suggested by the three software programs, we used the well-studied cyanobacterial clock gene kaiC to evaluate the three sets of reference genes, the best reference gene combination (as recommended by geNorm software), and the use of the best or the worst reference gene. Due to the fact that rimM was the best reference gene under NaCl stressed condition and was unstable under the other tested experimental conditions (Table 4), we chose the second best gene, rnpA, as the optimal reference gene. As shown in Table 4, under all experimental conditions, the circadian expression of kaiC peaked near the subjective dusk (approximately 9-12h), regardless of whether the optimal reference gene combination or a single optimal reference gene had been used to calibrate the Ct value of the qPCR amplification. This was consistent with a previous report on the circadian expression profile of kaiC under constant lighting conditions [10]. Interestingly, the single use of rnpA to calibrate the expression of a target gene under HO showed better performance in qPCR than the use of rnpA and ilvD in combination as geNorm software had recommended.

When calibrated by rimM, the rejected candidate reference gene under NTC stress, the diurnal peak of kaiC was significantly deviated from its reported trend. When the rejected candidates were used as internal references in the qPCR experiments, the peak time of the target gene not only was seriously affected, and even caused a reversed phase to the results similar to rimM under HO stress, but also resulted in some misleading findings, as the usage of rnpB displayed an incorrect diurnal double-peaks phenotype of kaiC expression under NaCl stress. Similar results were also found under HO conditions. When the rejected reference gene, rnpB, was used to calibrate the relative expression of kaiC by qPCR, the amplitude of the target gene diurnal expression profiles was significant dampened, and the SD values between samples were enhanced than those when the suggested best reference gene or the best reference combination was used to calibrate the same qPCR data. Therefore, the results of the rhythmic expression of kaiC by different candidate reference genes have indicated that the best reference gene or best reference combination recommended by geNorm, NormFinder, and BestKeeper software can be applied convincingly in qPCR reactions under NaCl, HO, or HL stress conditions in cyanobacterium S. elongatus PCC 7942.

4. Discussion

When the expression of the target gene is analyzed at the level of RNA transcription, there are inevitable operational errors due to the rapid extraction of large-scale biological samples and the subsequent preparation and quantification of the RNA samples. In addition, the absolute expression level of the same target gene can display significant errors between different biological groups or technical repetitions among different sample groups, which always leads to inaccurate biometric analysis. To address this issue, researchers often select one or more appropriate stably expressed genes, the so-called “housekeeping” genes as internal references. First, the absolute expression level of the target gene is calibrated based on the expression level of the internal reference gene. Thereafter, the normalized relative expression level of the target gene is used as an indicator to objectively describe the differential expression of the target gene under different experimental conditions or at different developmental stages [44]. qPCR is a common technique used for differential expression analysis under various experimental conditions for limited-scale closed gene populations at the RNA level. It has been reported that the selection of different reference genes significantly affects the acquisition of convincing qPCR results [45]. Using inappropriate reference genes for raw data calibration results in significant bias in the final results [46]. In this study, nine housekeeping genes in cyanobacteria S. elongatus PCC 7942 were selected as candidate reference genes for studies that evaluated the expression stability in cyanobacteria time-course RNA samples by qPCR that were held under different experimental conditions. To avoid bias in the experimental results caused by the use of a single evaluation software and a single algorithm, we selected three mainstream gene expression stability evaluation software programs based on different algorithms, namely, geNorm, NormFinder, and BestKeeper. The expression stability of selected nine candidate reference genes was detected and evaluated using the qPCR method. Based on the analysis of the three software programs, the expression stability of these candidate reference genes was comprehensively arranged, and the best reference gene of S. elongatus PCC 7942 under different experimental conditions was selected.

Pinto et al. reported that secA is the most suitable reference gene in the expression analysis of cyanobacteria Lyngbya aestuarii CCY 9616 and can be widely used in the calibration of the cyanobacteria qPCR amplification data [24]. In addition, Szekeres et al. showed in cyanobacteria Synechococcus sp. PCC 7002 that secA is stable under the given experimental conditions and is the best reference for RT-qPCR experiments [25]. These findings are consistent with the combination of the best reference genes we screened under NTC stress. Our results indicated that the secA gene could be stably expressed in different species of cyanobacteria under nonstress conditions. It is the optimal choice as the universal reference gene of cyanobacteria in qPCR experiments in the absence of environmental stress. In previous studies, the rnpA gene encoding the protein subunit of ribonuclease P (RNase P) was identified in various species of cyanobacteria, such as Synechocystis sp. PCC 6803, Lyngbya aestuarii CCY 9616, Nostoc sp. PCC 7120, and Synechococcus sp. PCC 7002, and it is the best internal reference gene [24, 25, 47]. Likewise, our results showed that in S. elongatus PCC 7942, rnpA was the most stable gene under HO stress. The ilvD gene ranks second in stability under HO stress, which is similar to a previous study that reported this gene obtained from Nostoc sp. PCC 7120 could serve as an optimal reference gene under specific experimental conditions [24].

Surprising results were found in the stability analysis of transcripts amount of the rimM gene that encodes the 16S rRNA processing protein. Although the rimM gene ranked 8th in the comprehensive ranking of the nine candidate reference genes under the four experimental conditions, it was considered to be the worst qPCR reference gene under NTC and HO conditions. However, the expression stability of this gene ranked first under NaCl stress. RNA microarray analysis carried out in Synechocystis sp. PCC 6803 showed that salt stress from the environment strongly induced the expression of genes encoding ribosomal proteins [48]. Therefore, we hypothesized that NaCl treatment may have affected the processing and assembling of the ribosome in S. elongatus PCC 7942, so that the expression of rimM gene may have been involved in the regulation of cells, where it was stably expressed. Before Szekeres’s report in 2014, rimM has not been reported as a reference gene for cyanobacterial gene expression studies. Based on the performance of the rimM gene in that study, Szekeres et al. suggested that rimM should be carefully selected as internal reference gene in studies involving cyanobacterial gene expression by qPCR. This is because it is necessary to verify the expression stability of multiple candidate reference genes under given experimental conditions and then choose the most stable reference to calibrate the relative expression of specific target genes [25]. Based on our results on the calibration of the target gene kaiC (Figure 4(b)), we confirmed that rimM was a suitable reference gene in S. elongatus PCC 7942 under NaCl stress. However, a double reference calibration by combining rimM with other suitable reference genes would further ensure the authenticity of the qPCR experimental results. In regard to the stability evaluation results of the nine candidate reference genes under the four simultaneous conditions, we found that the prs gene can be stably expressed under various stress conditions. Therefore, prs is an optimal reference gene for qPCR experiments in S. elongatus PCC 7942 under various environmental conditions such as NTC, HO, and HL (Table 4). The prs gene is one of the important housekeepers in cyanobacteria, and it encodes ribose phosphate pyrophosphate kinase, a key enzyme that is involved in the pentose phosphate pathway. These results also form an important physiological basis for the stable expression of this gene under various environmental stress conditions, as reported in this study.

(a)

(b)

(c)

(d)

Figure 4

The relative expression level of the cyanobacterial core clock gene kaiC under different experimental conditions was normalized to the most stable paired reference genes, the most stable single reference gene, and the most unstable reference gene. Templates were collected for 0–24 h under each treatment condition. (a) NTC. (b) NaCl. (c) HO. (d) HL. Bars represent the standard error from three biological replicates. The white bars under the X-axis indicate , the subjective daytime, and the gray bars under the X-axis indicate , the subjective night time.

In phylogenetic studies of prokaryotes, 16S ribosomal RNA is often selected as the target sequence for species-specific targeted amplification because of its conservative evolution and stable high expression [49, 50]. Interestingly, the 16S gene is also one of the most commonly used reference genes in qPCR experiments involving cyanobacteria [24, 29, 51–54]. The results of Pinto et al. showed that 16S effectively calibrates the expression of target genes in Nostoc sp. PCC 7120 and Synecocystis sp. PCC 6803 under various experimental conditions. However, in different species of cyanobacteria, the use of 16S as a reference gene for qPCR is controversial because the interpretation of results can be biased [25]. The results of present work also showed that the basal expression level of the 16S gene was too high, and its median Ct was 9.13 in qPCR (see Table 1). When 16S gene was used to calibrate specific low-abundance target genes, the very small median Ct of the target gene was very likely to artificially mask the differences in the relative expression of target genes between different sample groups. In addition, the comprehensive stability of the 16S gene under NaCl, HO, and HL stress was ranked 7th and 8th, respectively. These results showed that the expression of the 16S gene in cyanobacteria S. elongatus PCC 7942 under stress conditions was very unstable and was not the best choice as the reference gene in the qPCR reaction. This is due to the significant imbalance in the abundance between rRNA and mRNA molecules [55]. The different intracellular RNA degradation mechanisms also significantly affected the different half-lives of rRNA and mRNA [56]. On the other hand, the qPCR experiment itself usually requires the selection of a stably expressed gene having a similar Ct value as the reference gene [57]. Although the mRNA abundance of rnpB was also relatively high, the comprehensive ranking of the rnpB gene under all tested experimental conditions was always in the last two digits of the ranking, indicating that the expression of rnpB was unstable under almost all tested environmental conditions. This is consistent with the results of Pinto et al. Thus, rnpB is also not suitable for use as an internal reference gene in S. elongatus PCC 7942.

In summary, our results indicated that, in cyanobacteria S. elongatus PCC 7942, secA was the optimal reference gene in qPCR experiments under nonstress conditions, and rnpA and ilvD served as optimal reference genes under NaCl or HO conditions, respectively. prs could be used in a variety of experimental conditions for S. elongatus PCC 7942. However, reliable reference genes in qPCR experiments are always highly specific for individual environmental conditions. When performing qPCR in specific physiological or pathological conditions, it is important to first carefully evaluate and screen the reference genes that are most suitable for the preestablished experimental conditions as a necessary prerequisite for ensuring the credibility and reproducibility of the experiment [58]. The results of the current study showed that the expression stability of homologous housekeeping genes is not identical in different species of cyanobacteria or different sampling conditions. Previous evidence has shown that no housekeeping gene expression is stable under all environmental conditions [57]. In addition, if the reference gene stability evaluation results for a specific sample show that it is necessary to calibrate the Ct data of the target gene using two or more reference gene combinations together, the use of the two highest-ranking reference genes can simultaneously provide sufficient credibility and reproducibility for the qPCR experiment.

5. Conclusion

Based on geNorm, NormFinder, and BestKeeper gene expression stability analysis software programs, we studied the expression stability of nine candidate reference genes under four different experimental conditions in S. elongatus PCC 7942 and comprehensively evaluated their expression. The results of statistical analysis showed that the candidate reference genes prs and secA were the most stable in all samples tested, secA and ppc were most suitable for NTC stress, rimM and rnpA were most suitable for NaCl stress, rnpA and ilvD were most suitable for HO stress, and prs and secA were most suitable for HL stress. rimM was stably expressed only under specific conditions and should be carefully selected in future studies. 16S and rnpB were not suitable as reference genes. We anticipate that our results will provide a reliable experimental basis for the expression analysis of target genes under various stress conditions in S. elongatus PCC 7942 by the qPCR method.

Abbreviations

S. elongatus PCC 7942:	Synechococcus elongatus PCC 7942
qPCR:	Quantitative real-time polymerase chain reaction
16S:	16S ribosomal RNA
ALL:	All samples
NTC:	No treatment control
NaCl:	NaCl-treated group
HO:	H₂O₂-treated group
HL:	High light intensity-treated group.

Data Availability

The dataset supporting the conclusions of this study is included within the article and supplementary information files.

Conflicts of Interest

The authors declare that there are no conflicts of interest regarding the publication of this paper.

Authors’ Contributions

Xiao Luo and Yuwei Zhao conceived and designed the experiments. Xiao Luo, Jiaxing Li, and Tianliang Chang performed the experiments, analyzed the data, and prepared figures and/or tables. Hongyan He, Yi Zhao, and Xue Yang contributed to the sample analysis. Xiao Luo wrote drafts of the paper. Yuwei Zhao and Yao Xu approved the final paper.

Acknowledgments

This work was supported by grants from the National Natural Science Foundation of China (Grant no. 31470329), the Research Project of Provincial Key Laboratory of Shaanxi (Grant no. 17JS127), and the Research Project of Key Laboratory of Resource Biology and Biotechnology in Western China (Grant no. ZSK2018005).

Supplementary Materials

Supplementary Figure S1: 1.5% agarose gel electrophoresis pattern of total RNA from time-course NTC cyanobacteria samples (no treatment control). Figure S2: the quantity and quality detection of total RNA from time-course NTC cyanobacteria samples (no treatment control) by a BioSpec-nano spectrophotometer. The ratio of A₂₆₀/A₂₈₀ and A₂₆₀/A₂₃₀ from each total RNA sample was also shown in graphs. (a) NTC-CT₀. (b) NTC-CT₄. (c) NTC-CT₈. (d) NTC-CT₁₂. (e) NTC-CT₁₆. (f) NTC-CT₂₀. (g) NTC-CT₂₄. Figure S3: after amplification of each candidate reference gene PCR, the size of the expected fragment was visualized using agarose gel (1.5%) electrophoresis. M represents the 2000 bp DNA marker. Figure S4: melting curves of the nine candidate reference genes showed single peaks in qRT-PCR. Table S1: qPCR parameters for the amplification of kaiC under different experimental conditions. (Supplementary Materials)

References

A. R. Grossman, M. R. Schaefer, G. G. Chiang, and J. L. Collier, “The phycobilisome, a light-harvesting complex responsive to environmental conditions,” Microbiological Reviews, vol. 57, no. 3, pp. 725–749, 1993.
View at: Google Scholar
K. Hirokawa, M. Nagaya, H. Aiba, and T. Mizuno, “A cyanobacterial gene that interferes with the phosphotransfer signal transduction involved in the osmoregulatory expression of ompC and ompF in Escherichia coli,” Journal of the Agricultural Chemical Society of Japan, vol. 60, no. 2, pp. 277–283, 1996.
View at: Google Scholar
H. Cui, Y. Wang, Y. Wang, and S. Qin, “Genome-wide analysis of putative peroxiredoxin in unicellular and filamentous cyanobacteria,” BMC Evolutionary Biology, vol. 12, no. 1, article no. 220, 2012.
View at: Google Scholar
S. E. Cohen and S. S. Golden, “Circadian rhythms in cyanobacteria,” Microbiology Molecular Biology Reviews Mmbr, vol. 79, no. 4, pp. 373–385, 2015.
View at: Publisher Site | Google Scholar
C. H. Johnson, T. Mori, and Y. Xu, “A cyanobacterial circadian clockwork,” Current Biology, vol. 18, no. 17, pp. R816–R825, 2008.
View at: Publisher Site | Google Scholar
M. A. Woelfle and C. H. Johnson, “No promoter left behind: global circadian gene expression in cyanobacteria,” Journal of Biological Rhythms, vol. 21, no. 6, pp. 419–431, 2006.
View at: Publisher Site | Google Scholar
H. Ito, M. Mutsuda, Y. Murayama et al., “Cyanobacterial daily life with Kai-based circadian and diurnal genome-wide transcriptional control in Synechococcus elongatus,” Proceedings of the National Acadamy of Sciences of the United States of America, vol. 106, no. 33, pp. 14168–14173, 2009.
View at: Publisher Site | Google Scholar
Y. Liu, N. F. Tsinoremas, C. H. Johnson et al., “Circadian orchestration of gene expression in cyanobacteria,” Genes & Development, vol. 9, no. 12, pp. 1469–1478, 1995.
View at: Publisher Site | Google Scholar
S. S. Golden, M. Ishiura, C. H. Johnson, and T. Kondo, “Cyanobacterial circadian rhythms,” Annual Review of Plant Biology, vol. 48, no. 1, pp. 327–354, 1997.
View at: Publisher Site | Google Scholar
M. Ishiura, S. Kutsuna, S. Aoki et al., “Expression of a gene cluster kaiABC as a circadian feedback process in cyanobacteria,” Science, vol. 281, no. 5382, pp. 1519–1523, 1998.
View at: Publisher Site | Google Scholar
Y. Xu, T. Mori, R. Pattanayek, S. Pattanayek, M. Egli, and C. H. Johnson, “Identification of key phosphorylation sites in the circadian clock protein KaiC by crystallographic and mutagenetic analyses,” Proceedings of the National Acadamy of Sciences of the United States of America, vol. 101, no. 38, pp. 13933–13938, 2004.
View at: Publisher Site | Google Scholar
M. Nakajima, K. Imai, H. Ito et al., “Reconstitution of circadian oscillation of cyanobacterial KaiC phosphorylation in vitro,” Science, vol. 308, no. 5720, pp. 414-415, 2005.
View at: Publisher Site | Google Scholar
R. Waditee-Sirisattha, H. Kageyama, Y. Tanaka, M. Fukaya, and T. Takabe, “Overexpression of halophilic serine hydroxymethyltransferase in fresh water cyanobacterium Synechococcus elongatus PCC7942 results in increased enzyme activities of serine biosynthetic pathways and enhanced salinity tolerance,” Archives of Microbiology, vol. 199, no. 1, pp. 29–35, 2017.
View at: Publisher Site | Google Scholar
T. D. B. Mackenzie, R. A. Burns, and D. A. Campbell, “Carbon status constrains light acclimation in the cyanobacterium Synechococcus elongatus,” Plant Physiology, vol. 136, no. 2, pp. 3301–3312, 2004.
View at: Publisher Site | Google Scholar
A. Perelman, A. Uzan, D. Hacohen, and R. Schwarz, “Oxidative stress in Synechococcus sp. strain PCC 7942: Various mechanisms for H2O2 detoxification with different physiological roles,” Journal of Bacteriology, vol. 185, no. 12, pp. 3654–3660, 2003.
View at: Publisher Site | Google Scholar
O. Thellin, B. ElMoualij, E. Heinen, and W. Zorzi, “A decade of improvements in quantification of gene expression and internal standard selection,” Biotechnology Advances, vol. 27, no. 4, pp. 323–333, 2009.
View at: Publisher Site | Google Scholar
S. A. Bustin, V. Benes, T. Nolan, and M. W. Pfaffl, “Quantitative real-time RT-PCR - A perspective,” Molecular Endocrinology, vol. 34, no. 3, pp. 597–601, 2005.
View at: Publisher Site | Google Scholar
S. Derveaux, J. Vandesompele, and J. Hellemans, “How to do successful gene expression analysis using real-time PCR,” Methods, vol. 50, no. 4, pp. 227–230, 2010.
View at: Publisher Site | Google Scholar
O. Thellin, W. Zorzi, B. Lakaye et al., “Housekeeping genes as internal standards: use and limits,” Journal of Biotechnology, vol. 75, no. 2-3, pp. 291–295, 1999.
View at: Publisher Site | Google Scholar
K. Dheda, J. F. Huggett, S. A. Bustin, M. A. Johnson, G. Rook, and A. Zumla, “Validation of housekeeping genes for normalizing RNA expression in real-time PCR,” BioTechniques, vol. 37, no. 1, pp. 112–119, 2004.
View at: Publisher Site | Google Scholar
T. Remans, K. Smeets, K. Opdenakker, D. Mathijsen, J. Vangronsveld, and A. Cuypers, “Normalisation of real-time RT-PCR gene expression measurements in Arabidopsis thaliana exposed to increased metal concentrations,” Planta, vol. 227, no. 6, pp. 1343–1349, 2008.
View at: Publisher Site | Google Scholar
S. Taylor, M. Wakem, G. Dijkman, M. Alsarraj, and M. Nguyen, “A practical approach to RT-qPCR-Publishing data that conform to the MIQE guidelines,” Methods, vol. 50, no. 4, pp. S1–S5, 2010.
View at: Publisher Site | Google Scholar
S. C. Taylor and E. M. Mrkusich, “The state of RT-quantitative PCR: Firsthand observations of implementation of minimum information for the publication of quantitative real-time PCR experiments (MIQE),” Journal of Molecular Microbiology and Biotechnology, vol. 24, no. 1, pp. 46–52, 2014.
View at: Publisher Site | Google Scholar
F. Pinto, C. C. Pacheco, D. Ferreira, P. Moradas-Ferreira, and P. Tamagnini, “Selection of suitable reference genes for RT-qPCR analyses in cyanobacteria,” PLoS ONE, vol. 7, no. 4, Article ID e34983, 2012.
View at: Google Scholar
E. Szekeres, C. Sicora, N. Dragoş, and B. Drugă, “Selection of proper reference genes for the cyanobacterium Synechococcus PCC 7002 using real-time quantitative PCR,” FEMS Microbiology Letters, vol. 359, no. 1, pp. 102–109, 2014.
View at: Publisher Site | Google Scholar
J. M. Rinta-Kanto, A. J. A. Ouellette, G. L. Boyer, M. R. Twiss, T. B. Bridgeman, and S. W. Wilhelm, “Quantification of toxic Microcystis spp. during the 2003 and 2004 blooms in western Lake Erie using quantitative real-time PCR,” Environmental Science & Technology, vol. 39, no. 11, pp. 4198–4205, 2005.
View at: Publisher Site | Google Scholar
E. Urbach, D. L. Robertson, and S. W. Chisholm, “Multiple evolutionary origins of prochlorophytes within the cyanobacterial radiation,” Nature, vol. 355, no. 6357, pp. 267–270, 1992.
View at: Publisher Site | Google Scholar
S. Imamura, K. Tanaka, M. Shirai, and M. Asayama, “Growth phase-dependent activation of nitrogen-related genes by a control network of group 1 and group 2 σ factors in a cyanobacterium,” The Journal of Biological Chemistry, vol. 281, no. 5, pp. 2668–2675, 2006.
View at: Publisher Site | Google Scholar
L. Schäfer, M. Sandmann, S. Woitsch, and G. Sandmann, “Coordinate up-regulation of carotenoid biosynthesis as a response to light stress in Synechococcus PCC7942,” Plant, Cell & Environment, vol. 29, no. 7, pp. 1349–1356, 2006.
View at: Publisher Site | Google Scholar
G. D. Price, F. J. Woodger, M. R. Badger, S. M. Howitt, and L. Tucker, “Identification of a SulP-type bicarbonate transporter in marine cyanobacteria,” Proceedings of the National Acadamy of Sciences of the United States of America, vol. 101, no. 52, pp. 18228–18233, 2004.
View at: Publisher Site | Google Scholar
I. Mary, C.-J. Tu, A. Grossman, and D. Vaulot, “Effects of high light on transcripts of stress-associated genes for the cyanobacteria Synechocystis sp. PCC 6803 and Prochlorococcus MEDA and MIT9313,” Microbiology, vol. 150, no. 5, pp. 1271–1281, 2004.
View at: Publisher Site | Google Scholar
T. Fujimori, M. Higuchi, H. Sato et al., “The mutant of sll1961, which encodes a putative transcriptional regulator, has a defect in regulation of photosystem stoichiometry in the cyanobacterium Synechocystis sp. PCC 6803,” Plant Physiology, vol. 139, no. 1, pp. 408–416, 2005.
View at: Publisher Site | Google Scholar
C.-J. Tu, J. Shrager, R. L. Burnap, B. L. Postier, and A. R. Grossman, “Consequences of a deletion in dspA on transcript accumulation in Synechocystis sp. strain PCC6803,” Journal of Bacteriology, vol. 186, no. 12, pp. 3889–3902, 2004.
View at: Publisher Site | Google Scholar
F. J. Woodger, M. R. Badger, and G. D. Price, “Inorganic carbon limitation induces transcripts encoding components of the CO2-concentrating mechanism in Synechococcus sp. PCC7942 through a redox-independent pathway,” Plant Physiology, vol. 133, no. 4, pp. 2069–2080, 2003.
View at: Publisher Site | Google Scholar
J. Vandesompele, K. De Preter, F. Pattyn et al., “Accurate normalization of real-time quantitative RT-PCR data by geometric averaging of multiple internal control genes,” Genome Biology, vol. 3, no. 7, Article ID 0034.11, 2002.
View at: Google Scholar
C. L. Andersen, J. L. Jensen, and T. F. Ørntoft, “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 Research, vol. 64, no. 15, pp. 5245–5250, 2004.
View at: Publisher Site | Google Scholar
M. W. Pfaffl, A. Tichopad, C. Prgomet, and T. P. Neuvians, “Determination of stable housekeeping genes, differentially regulated target genes and sample integrity: bestKeeper—excel-based tool using pair-wise correlations,” Biotechnology Letters, vol. 26, no. 6, pp. 509–515, 2004.
View at: Publisher Site | Google Scholar
N. Silver, S. Best, J. Jiang, and S. L. Thein, “Selection of housekeeping genes for gene expression studies in human reticulocytes using real-time PCR,” BMC Molecular Biology, vol. 7, no. 33, 2006.
View at: Publisher Site | Google Scholar
F. Jacob, R. Guertler, S. Naim et al., “Careful selection of reference genes is required for reliable performance of RT- qPCR in human normal and cancer cell lines,” PLoS ONE, vol. 8, no. 3, Article ID e59180, 2013.
View at: Publisher Site | Google Scholar
K. Imai, Y. Kitayama, and T. Kondo, “Elucidation of the role of clp protease components in circadian rhythm by genetic deletion and overexpression in cyanobacteria,” Journal of Bacteriology, vol. 195, no. 19, pp. 4517–4526, 2013.
View at: Publisher Site | Google Scholar
T. Wishiwaki, Y. Satomi, M. Nakajima et al., “Role of KaiC phosphorylation in the circadian clock system of Synechococcus elongatus PCC 7942,” Proceedings of the National Acadamy of Sciences of the United States of America, vol. 101, no. 38, pp. 13927–13932, 2004.
View at: Publisher Site | Google Scholar
Q. Zhang, M. Zhang, J. Li et al., “Selection and validation of reference genes for RT-PCR expression analysis of candidate genes involved in morphine-induced conditioned place preference mice,” Journal of Molecular Neuroscience, vol. 66, no. 4, pp. 587–594, 2018.
View at: Publisher Site | Google Scholar
S. Wen, X. Chen, F. Xu, and H. Sun, “Validation of reference genes for real-time quantitative PCR (qPCR) Analysis of avibacterium paragallinarum,” PLoS ONE, vol. 11, no. 12, Article ID e0167736, 2016.
View at: Google Scholar
D. A. Sandercock, J. E. Coe, P. Di Giminiani, and S. A. Edwards, “Determination of stable reference genes for RT-qPCR expression data in mechanistic pain studies on pig dorsal root ganglia and spinal cord,” Research in Veterinary Science, vol. 114, pp. 493–501, 2017.
View at: Publisher Site | Google Scholar
X. Yang, H. Pan, L. Yuan, and X. Zhou, “Reference gene selection for RT-qPCR analysis in Harmonia axyridis, a global invasive lady beetle,” Scientific Reports, vol. 8, no. 1, article no. 2689, 2018.
View at: Publisher Site | Google Scholar
R. Shivhare and C. Lata, “Selection of suitable reference genes for assessing gene expression in pearl millet under different abiotic stresses and their combinations,” Scientific Reports, vol. 6, article no. 23036, 2016.
View at: Google Scholar
A. Pascual and A. Vioque, “Cloning, purification and characterization of the protein subunit of ribonuclease P from the cyanobacterium Synechocystic sp. PCC 6803,” European Journal of Biochemistry, vol. 241, no. 1, pp. 17–24, 1996.
View at: Publisher Site | Google Scholar
Y. Kanesaki, I. Suzuki, S. I. Allakhverdiev, K. Mikami, and N. Murata, “Salt stress and hyperosmotic stress regulate the expression of different sets of genes in Synechocystis sp. PCC 6803,” Biochemical and Biophysical Research Communications, vol. 290, no. 1, pp. 339–348, 2002.
View at: Publisher Site | Google Scholar
M. Lopez-Fernandez, A. Cherkouk, R. Vilchez-Vargas et al., “Bacterial diversity in bentonites, engineered barrier for deep geological disposal of radioactive wastes,” Microbial Ecology, vol. 70, no. 4, pp. 922–935, 2015.
View at: Publisher Site | Google Scholar
A. Ballot, L. Cerasino, V. Hostyeva, and S. Cirés, “Variability in the sxt gene clusters of PSP toxin producing Aphanizomenon gracile strains from Norway, Spain, Germany and North America,” PLoS ONE, vol. 11, no. 12, Article ID e0167552, 2016.
View at: Google Scholar
T. Sakamoto, S. Higashi, H. Wada, N. Murata, and D. A. Bryant, “Low-temperature-induced desaturation of fatty acids and expression of desaturase genes in the cyanobacterium Synechococcus sp. PCC 7002,” FEMS Microbiology Letters, vol. 152, no. 2, pp. 313–320, 1997.
View at: Publisher Site | Google Scholar
F. Engelbrecht, K. Marin, and M. Hagemann, “Expression of the ggpS gene, involved in osmolyte synthesis in the marine cyanobacterium Synechococcus sp. strain PCC 7002, revealed regulatory differences between this strain and the freshwater strain Synechocystis sp. strain PCC 6803,” Applied and Environmental Microbiology, vol. 65, no. 11, pp. 4822–4829, 1999.
View at: Google Scholar
Y. Hihara, A. Kamei, M. Kanehisa, A. Kaplan, and M. Ikeuchi, “DNA microarray analysis of cyanobacterial gene expression during acclimation to high light,” The Plant Cell, vol. 13, no. 4, pp. 793–806, 2001.
View at: Publisher Site | Google Scholar
H. Song, J. Xu, M. Lavoie et al., “Biological and chemical factors driving the temporal distribution of cyanobacteria and heterotrophic bacteria in a eutrophic lake (West Lake, China),” Applied Microbiology and Biotechnology, vol. 101, no. 4, pp. 1685–1696, 2017.
View at: Publisher Site | Google Scholar
M. Solanas, R. Moral, and E. Escrich, “Unsuitability of using ribosomal RNA as loading control for Northern blot analyses related to the imbalance between messenger and ribosomal RNA content in rat mammary tumors,” Analytical Biochemistry, vol. 288, no. 1, pp. 99–102, 2001.
View at: Publisher Site | Google Scholar
B. S. Ferguson, H. Nam, R. G. Hopkins, and R. F. Morrison, “Impact of reference gene selection for target gene normalization on experimental outcome using real- time qRT-PCR in adipocytes,” PLoS ONE, vol. 5, no. 12, Article ID e15208, 2010.
View at: Google Scholar
T. Hruz, M. Wyss, M. Docquier et al., “RefGenes: identification of reliable and condition specific reference genes for RT-qPCR data normalization,” BMC Genomics, vol. 12, no. 1, pp. 1–14, 2011.
View at: Publisher Site | Google Scholar
Z. Tong, Z. Gao, F. Wang, J. Zhou, and Z. Zhang, “Selection of reliable reference genes for gene expression studies in peach using real-time PCR,” BMC Molecular Biology, vol. 10, article 71, 2009.
View at: Publisher Site | Google Scholar

Copyright

Copyright © 2019 Xiao Luo et al. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

PDF Download Citation

Download other formats

Order printed copies

Views

2766

Downloads

1628

Citations