Figures
Abstract
Quantitative real-time PCR (qRT-PCR) is a reliable and reproducible technique for measuring mRNA expression. To facilitate gene expression studies and obtain more accurate qRT-PCR analysis, normalization relative to stable housekeeping genes is mandatory. In this study, ten housekeeping genes, including beta-actin (Actin) , elongation factor 1 α (EF1A) , glyceralde hyde-3-phosphate dehydrogenase (GAPDH) , ribosomal protein L13 (RPL13) , ribosomal protein 49 (RP49) , α-tubulin (Tubulin) , vacuolar-type H+-ATPase (v-ATPase) , succinate dehydrogenase subunit A (SDHA) , 28S ribosomal RNA (28S) , and 18S ribosomal RNA (18S) from the two-spotted spider mite, Tetranychus urticae, were selected as the candidate reference genes. Four algorithms, geNorm, Normfinder, BestKeeper, and the ΔCt method, were used to evaluate the performance of these candidates as endogenous controls across different developmental stages. In addition, RefFinder, which integrates the above-mentioned software tools, provided the overall ranking of the stability/suitability of these candidate reference genes. Among them, PRL13 and v-ATPase were the two most stable housekeeping genes across different developmental stages. This work is the first step toward establishing a standardized qRT-PCR analysis in T. urticae following the MIQE guideline. With the recent release of the T. urticae genome, results from this study provide a critical piece for the subsequent genomics and functional genomics research in this emerging model system.
Citation: Yang C, Pan H, Liu Y, Zhou X (2015) Stably Expressed Housekeeping Genes across Developmental Stages in the Two-Spotted Spider Mite, Tetranychus urticae. PLoS ONE 10(3): e0120833. https://doi.org/10.1371/journal.pone.0120833
Academic Editor: Christian Schönbach, Nazarbayev University, KAZAKHSTAN
Received: June 9, 2014; Accepted: February 7, 2015; Published: March 30, 2015
Copyright: © 2015 Yang et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited
Data Availability: All relevant data are within the paper and its Supporting Information files.
Funding: This research was supported by a start-up fund from the University of Kentucky to XGZ, a grant from USDA BRAG grant (Award Agreement No.: 3048108827) to XGZ, and a Special Fund for Agroscience Research in the Public Interest (Award Agreement No.: 201303028) to YL. These agencies had no role in study design, data collection/analysis, manuscript preparation, or the decision to publish.
Competing interests: The authors have declared that no competing interests exist.
Introduction
Quantitative real-time PCR (qRT-PCR) is a rapid and reliable method for the detection and quantification of gene expression during different biological processes [1]. Although qRT-PCR analysis has been a primary tool in the molecular biology research, limitations still exist, including variation in RNA extraction, reverse transcription and normalization, and PCR efficiency [2, 3]. To ensure the accuracy, a critical component in qRT-PCR analysis is to normalize data by measuring in parallel the expression of a reference gene from the same samples. Housekeeping genes, involved in basic cellular functions, are typically maintaining stable and constitutive expression in all cells and regardless of physiological conditions. Consequently, housekeeping genes have been widely adopted in many molecular and genomics studies as a reference for calibration purposes, including qRT-PCR analysis [1,4].
The two-spotted spider mite, Tetranychus urticae Koch, is one of the most important pest species worldwide. Considered as the most polyphagous species within the family of Tetranychidae, T. urticae can infest nearly 800 plant species, including peppers, tomatoes, potatoes, beans, maize, strawberries, and ornamental plants such as roses [5]. T. urticae lays its egg on the leaf, damages host plants by sucking cell contents from the leaf, and leaves tiny pale spots or scars where the green epidermal cells have been destroyed [5]. Although the individual lesions are small in size, attacks by hundreds and thousands of spider mites can significantly reduce the photosynthetic capability of plants [5]. The control of T. urticae has traditionally relied on synthetic insecticides and acaricides. As a result, T. urticae has developed resistance to almost every major active compound [6]. With the advent of T. urticae genome [7], the first complete genome being sequenced from a chelicerate species, there is an unprecedented opportunity to investigate the genetic basis of pesticide resistance and other biological phenomena in T. urticae [6, 8, 9]. To take advantage of these genomic resources, establishing a standardized qRT-PCR procedure in T. urticae following the MIQE (Minimum Information for publication of Quantitative real-time PCR Experiments) guidelines [10] will be instrumental for the subsequent genomics and functional genomics studies in this emerging pest model.
The objective of this study was to address an important aspect of gene expression studies in T. urticae: the selection and validation of appropriate reference genes with stable expression under different developmental stages. Here, ten housekeeping genes, including beta-actin (Actin), elongation factor 1 α (EF1A), glyceralde hyde-3-phosphate dehydrogenase (GAPDH), ribosomal protein L13 (RPL13), ribosomal protein 49 (RP49), α-tubulin (Tubulin), vacuolar-type H+-ATPase (v-ATPase), succinate dehydrogenase subunit A (SDHA), 28S ribosomal RNA (28S), and 18S ribosomal RNA (18S) from the T. urticae genome were selected as the candidate reference genes [7]. The stability of these reference genes was investigated across different developmental stages. To validate the recommendations, the effectiveness of these candidates was further examined by qRT-PCR analysis using CYP392D8, a P450 gene associated with the abamectin resistance [9].
Materials and Methods
Ethics Statement
The Tetranychus urticae colony was collected from honeyvine milkweed, Cynanchum laeve (syn. Ampelamus albidus) in a greenhouse (20–28°C) at the University of Kentucky. No specific permit was required for the described collections. T. urticae is a common mite species with agricultural importance in the USA.
Sample preparation
The T. urticae was maintained in a climate chamber at 23°C with a photoperiod of 14: 10 (L: D) and 50% relative humidity. Fifteen adult females were allowed to oviposit for 12 h on a milkweed leaf resting on wet filter paper in a petri dish (9 cm diameter). When the adults were removed, there were approximately 100 eggs in each petri dish. One week after the egg eclosion, approximately 60 nymphs (a mix of protonymphs and deutonymphs) were collected. Adults were collected directly from the milkweed, and a total of 40 mixed-sex individuals were used in one biological replicate. All samples were preserved in 1.5 ml microcentrifuge tubes and stored at −80°C after being snap frozen in liquid nitrogen. For egg, nymph, and adult samples, each sample had three technical replications, and each developmental stage was independently collected three times.
Total RNA extraction and cDNA synthesis
Tetranychus urticae samples (100 eggs, 60 nymphs, or 40 adults) were homogenized in a 1.5 ml RNase-free microcentrifuge tube containing 400 μl TRIzol reagent (Invitrogen, Carlsbad, CA). The whole body homogenates were centrifuged at 12000 g for 15 min and then supernatant was transferred to a new 1.5 ml microcentrifuge tube. A volume of 100 μl chloroform was added to the supernatant and then the mixture was incubated at room temperature for 10 min and then centrifuged at 4°C, 12000 g for 15 min. After that, the supernatant was transferred to a new 1.5 ml microcentrifuge tube, 200 μl Isopropyl alcohol was added to it, and the mixture precipitated at room temperature for 10 min. Then, the supernatant was discarded after the mixture was centrifuged at 4°C, 12000 g for 8 min, and then 500 μl 75% alcohol was added and centrifuged at 4°C, 7500 g for 5 min to wash the pellet. Finally, the pellet was air dried for 5 min and then dissolved in 30 μl ddH2O. DNAse treated total RNA was denatured at 75°C for 5 min and immediately chilled on ice. The concentration of RNA from the egg, nymph, and adult was quantified with a NanoDrop 2000c spectrophotometer with the result for egg (83.2 ± 17.2 ng/μl), nymph (104.0 ± 10.3 ng/μl), and adult (113.7 ± 23.1 ng/μl). First strand cDNA was synthesized from 0.1 μg of total RNA with M-MLV reverse transcription kit (Invitrogen, USA) using a random N primer according the manufacturer’s recommendations. The cDNA was diluted five-fold for the subsequent qPCR studies.
Reference gene selection and primer design
Table 1 lists the candidate reference genes and primer sets associated with these candidates. PCR amplifications were performed in 50 μl reactions, containing 10 μl 5×PCR Buffer (Mg2+ Plus), 1 μl dNTP mix (10 mM of each nucleotide), 5 μl of each primer (10μM each), and 0.25 μl of Go Taq (5 u/μl) (Promega). The PCR parameters were as follows: one cycle of 94°C for 3 min; 35 cycles of 94°C for 30 s, 59°C for 45 s and 72°C for 1 min; and a final cycle of 72°C for 10 min. Amplicons of the expected size were purified and cloned into the pCR4-TOPO vector (Invitrogen, Carlsbad, CA) for the sequencing confirmation.
Quantitative real-time PCR
Gene-specific primers (Table 1) were used in qRT-PCR reactions (20 μl), containing 7 μl of ddH2O, 10 μl of 2×SYBR Green MasterMix (Bio-Rad), 1 μl of each specific primer (10 μM), and 1 μl of first-strand cDNA template. The qPCR program included an initial denaturation for 3 min at 95°C followed by 40 cycles of denaturation at 95°C for 10 s, annealing for 30 s at 55°C, and extension for 30 s at 72°C. For melting curve analysis, a dissociation step cycle (55°C for 10 s, and then 0.5°C for 10 s until 95°C) was added. The reactions were set up in 96-well format Microseal PCR plates (Bio-Rad) in triplicates.
Reactions were performed in a MyiQ single Color Real-Time PCR Detection System (BioRad). The presence of a single peak in the melting curve analysis was used to confirm gene-specific amplification and to rule out the non-specific amplification and the generation of primer-dimer. qRT-PCR efficiency was determined for each gene using slope analysis with a linear regression model. Relative standard curves for the transcripts were generated with serial dilutions of cDNA (1/5, 1/25, 1/125, 1/625, and 1/3125). The corresponding qRT-PCR efficiencies (E) were calculated according to the equation: E = (10[-1/slope] -1)×100.
Stability of gene expression
All biological replicates were used to calculate the average Ct value. The stability of the ten housekeeping genes was evaluated by algorithms geNorm [1], NormFinder [11], BestKeeper [12], and the comparative ΔCt method [13]. Finally, we compared and ranked the tested candidates based on a web-based analysis tool RefFinder (http://www.leonxie.com/referencegene.php).
Validation of reference gene selection
CYP392D8, a cytochrome P450 associated with abamectin resistance in T. urticae [9], was used as a target gene to evaluate the performance of candidate reference genes. CYP392D8 expression was profiled in three life stages: egg, nymph, and adult. Two different normalization factors (NFs) were calculated based on (1) the geometric mean of the genes with the lowest Geomean value (as determined by RefFinder), and (2) a single reference gene with the lowest or highest Geomean value. Relative quantification of CYP392D8 expression was calculated using the 2-ΔΔCt method [14].
Results
Validation and transcriptional profiling of candidate reference genes
Initially, ten candidate reference genes were investigated by reverse transcription polymerase chain reaction (RT-PCR). All genes examined in T. urticae were visualized as a single amplicon with expected size on a 1.5% agarose gel (S1 Fig.). All amplicons were sequenced and displayed 100% identity with their corresponding transcripts. Furthermore, gene-specific amplification of these genes was confirmed by a single peak in real-time melting-curve analysis (S2 Fig.). A standard curve was generated for each gene, using five-fold serial dilution of the pooled cDNAs. The correlation coefficient and PCR efficiency for each standard curve are shown in Table 1.
The mean and standard deviation (SD) of the Ct values for egg, nymph, and adult were summarized in S1 Table. With high SD values, the expression of RP49, Actin, and EF1A varied substantially. In contrast, 18S, GAPDH, and RPL13 had the least variable expressions reflected in their low SD values. Additionally, SDHA (Ctavg = 31.39), and RPL13 (Ctavg = 30.39) had the highest Ct values, suggesting that they are the least expressed gene candidates in egg, nymph, and adult. 18S (Ctavg = 15.89) and 28S (Ctavg = 14.94) had the lowest Ct values, indicating that they are the most expressed gene candidates in eggs, nymphs, and adults (Fig. 1, S1 Table).
The expression of candidate reference genes was documented in Ct-value. The median is represented by the line in the box. The interquartile rang is bordered by the upper and lower edges, which indicate the 75th and 25th percentiles, respectively.
Optimal number of reference genes for normalization
To decide the minimal number of genes mandatory for normalization, the V-value was computed by geNorm. Beginning with the two most stable genes, the software automatically adds another gene following a step-wise format to recalculate the normalization factor ratio. If the newly constructed V-value is lower than the 0.15 cut-off value, the starting number of gene pairings will be sufficient for the consistent normalization. Otherwise, additional genes need to be incorporated until the V-value reach the threshold, i.e., no significant changes in the normalization factor. The first V-value<0.15 was V2/3 (Fig. 2A), suggesting that the pairing of the two most stable reference genes was reliable enough for the normalization throughout the developmental stages.
A) Pairwise variation (V) analysis. geNorm first calculates an expression stability value (M) for each gene and then compares the pair-wise variation (V) of this gene with the others. The pair-wise variation (Vn/Vn+1) was analyzed between the normalization factors NFn and NFn+1 by the geNorm software to determine the optimal number of references genes required for normalization. A threshold of V<0.15 was suggested for the valid normalization [1]. B) Ranking of the 11 housekeeping genes based on the stability value (M).
Determining the best candidate reference genes based on the four statistical algorithms
GeNorm bases its ranking on the geometric mean of the SD of each transformed gene set of pair combinations (M-value). The lower the M-value is, the higher the ranking. Recommended M values for geNorm are M<0.5 for homogeneous samples and M<1 for heterogenous samples [15].
RPL13 and v-ATPase were co-ranked as the most stable reference genes (M = 0.35). The overall ranking of candidate reference genes from the most stable to the least stable was: RPL13, GAPDH, 18S, v-ATPase, Tubulin, EF1A, 28S, SDHA, Actin, and RP49 (Fig. 2B).
ΔCt method ranks the stability of candidate reference genes based on the pair-wise comparisons. Using raw Ct value, the average SD of each gene set is inversely proportional to its stability. As shown in S2 Table and Table 2, RPL13 (0.76) was the most stable reference gene. The overall ranking of candidate reference genes from the most stable to the least stable was: RPL13, v-ATPase, 18S, Tubulin, GAPDH, 28S, EF1A, SDHA, Actin, and RP49 (Table 2).
Comprehensive ranking by RefFinder
RefFinder is a web-based platform integrating all four above-mentioned algorithms to evaluate and select reference genes from extensive experimental datasets. In this study, all algorithms except geNorm recommended RPL13 as the most stable reference gene (Table 2), which is consistent with the ranking from RefFinder. The comprehensive ranking of candidate reference genes from the most stable to the least stable was: RPL13, GAPDH, v-ATPase, 18S, Tubulin, 28S, EF1A, SDHA, Actin, and RP49. Among them, GM values of Actin and RP49 were both higher than 9.0 (Table 2), indicating the lowest stability and the least likelihood of serving as the reference gene for qRT-PCR analysis in T. urticae across different developmental stages.
Validation of the recommended reference genes
To examine the validity of the selected reference genes (genes with the low Geomean value are considered stable), the expression profile of a target gene, CYP392D8, was investigated under different developmental stages. Similar expression profiles of CYP392D8 were observed using either one [RPL13; NF1], two [RPL13 and GAPDH; NF (1–2)], or three best reference genes [RPL13, GAPDH, and v-ATPase; NF (1–3)] for the normalization (Fig. 3). The expression of CYP392D8 was numerically lower at the nymph stage using a single reference gene with the highest Geomean value (RP49; NF10) (Fig. 3).
Expression profiles of CYP392D8 were investigated using different normalization factors. The expression of CYP392D8 was normalized using the best reference gene (NF1), the top two [NF (1–2)], the top three reference genes [NF (1–3)], and the worst reference gene (NF10). Bars represent the means and standard error of three biological replications.
Discussion
qRT-PCR quantification demands a comprehensive normalization by reference genes to offset confounding variations among the extensive experimental datasets. Most gene expression studies in the literature use a single endogenous control; this will profoundly influence the statistical outcome and may lead to inaccurate data interpretation [16]. Most recently, there are a influx of reference gene selection studies in insects, including whitefly, diamondback moth, brown planthopper, beet armyworm, oriental leafworm moth, Colorado potato beetle, oriental fruit fly, Russian wheat aphid, and pea aphid, etc [17–25]. In comparison, reference gene selections have largely been ignored in Tetranychidae species. Attribute to their economical importance, carmine spider mite, T. cinnabarinus (Boisduval) [26], T. urticae [27], and citrus red mite, Panonychus citri (McGregor) [28] are the only three examples in Tetranychidae. In Yue et al. 2013 [27], the mRNA expression profiles of eight housekeeping genes were investigated in fenpropathrin-susceptible and -resistant strains of T. urticae. Our major concern is that there was not sufficient replication in the experimental design of Yue et al. 2013 [27]; the entire dataset was generated from approximately 400 fenpropathrin-susceptible and 400 fenpropathrin-resistant adult females, respectively. Additionally, our experiments have contrasting designs in that Yue et al. 2013 focused on the selection of reference genes under the impact of insecticide resistance [27], while we investigated the expression profiles of ten housekeeping genes across different developmental stages of T. urticae.
In our study, except for 18S and 28S, the Ct values of these selected reference genes are rather high (S1 Table); however, the Ct values for reference genes are mostly around 15–25 in most of the insects [17–23]. As for the high Ct values, only 0.1 μg of total RNA was used for the cDNA synthesized in this study, whereas, 1.0 μg of total RNA with M-MLV reverse transcription kit was used in previous studies [17–23]. This could be the main factor contributing to the high Ct values. In addition, the expression of reference genes differs in different organisms. For example, the Actin gene was highly expressed in the citrus red mite, Panonychus citri with the Ct values around 23 across various experimental conditions [28], while, the Actin gene was less expressed in the T. urticae strain in our study with the Ct value around 28. The Actin gene was also less expressed in another T. urticae strain with the Ct value around 27 (Sunny Yoon, University of Kentucky, personal communication), although different primer pairs of the same gene were used between our studies.
Although housekeeping genes are constitutively expressed to maintain cellular function, they do not necessarily meet the prerequisites for a good reference gene that can be ‘expressed at constant levels across various conditions [16–18]. In fact, there are no "universal" reference genes that are stably expressed and suitable for the entire cell and tissue, and various experimental conditions [17–23]. Therefore, customized reference gene selection under specific experimental conditions is highly recommended [29].
There has been ongoing discussion about the optimal number of reference genes required for qRT-PCR analysis. To avoid biased normalization, multiple reference genes have been adopted to analyze gene expression under various experimental conditions. A single reference gene is usually insufficient to normalize the expression of target genes [30]. Results from our validation study with a targeted P450 gene are consistent with the multi-gene normalizer concept, suggesting that the use of two reference genes is sufficient to normalize the expression data and provides more conservative estimation of target gene expression in T. urticae (Figs. 2B and 3). As a result, we recommend that two internal references are necessary for investigating gene expressions in T. urticae across different developmental stages. The combined results not only provide a standardized procedure for quantification of gene expression in T. urticae, but also lay a solid foundation for the genomics and functional genomics research in this emerging pest species.
Supporting Information
S1 Fig. The agrose gel profile of the ten candidate reference genes.
M, EZ Load 100 bp Molecular Ruler; Templates in the PCR reactions were as follows: 1) EF1A; 2) SDHA; 3) GAPDH; 4) Tubulin; 5) RPL13; 6) RP49; 7) 18S; 8) 28S; 9) v-ATPase; and 10) Actin.
https://doi.org/10.1371/journal.pone.0120833.s001
(TIFF)
S2 Fig. Melting curve of the ten candidate reference genes.
https://doi.org/10.1371/journal.pone.0120833.s002
(TIF)
S1 Table. The mean and standard deviation (SD) of the Ct values of the ten candidate reference gene.
https://doi.org/10.1371/journal.pone.0120833.s003
(DOCX)
S2 Table. Pairwise comparison of candidate reference genes.
https://doi.org/10.1371/journal.pone.0120833.s004
(DOCX)
S3 Table. Ranking of the candidate reference genes based on their crossing point (CP) values by BestKeeper.
https://doi.org/10.1371/journal.pone.0120833.s005
(DOCX)
Acknowledgments
The authors are grateful to anonymous reviewers and the editor for their constructive criticisms. Special thanks go to Dr. Jamin Dreyer for his comments on an earlier draft, Dr. Joe Deas for his help to revise the grammar, and Dr. Rumei Li (Shandong Academy of Agricultural Sciences) for her assistance with the data analysis. The information reported in this paper (No.14-08-066) is part of a project of the Kentucky Agricultural Experiment Station and is published with the approval of the Director.
Author Contributions
Conceived and designed the experiments: HPP YL XGZ. Performed the experiments: HPP CXY. Analyzed the data: HPP. Contributed reagents/materials/analysis tools: XGZ. Wrote the paper: HPP CXY XGZ.
References
- 1. Vandesompele J, De Preter K, Pattyn F, Poppe B, Van Roy N, De Paepe A, et al. (2002) Accurate normalization of real-time quantitative RT-PCR data by geometric averaging of multiple internal control genes. Genome Biol 3: research0034.
- 2. Strube C, Buschbaum S, Wolken S, Schnieder T (2008) Evaluation of reference genes for quantitative real-time PCR to investigate protein disulfide isomerase transcription pattern in the bovine lungworm Dictyocaulus viviparus. Gene 425: 36–43. pmid:18761062
- 3. Bustin SA, Benes V, Nolan T, Pfaffl MW (2005) Quantitative real-time RT-PCR-a perspective. J Mol Endocrinol 34: 597–601. pmid:15956331
- 4. Eisenberg E, Levanon EY (2013) Human housekeeping genes, revisited. Trends Genet 29: 569–574. pmid:23810203
- 5.
Fasulo TR, Denmark HA (2009) Two spotted spider mite. Featured Creatures. University of Florida / Institute of Food and Agricultural Sciences.
- 6. Van Leeuwen T, Vontas J, Tsagkarakou A, Dermauw W, Tirry L (2010) Acaricide resistance mechanisms in the two-spotted spider mite Tetranychus urticae and other important Acari: A review. Insect Biochem Molec 40: 563–572. pmid:20685616
- 7. Grbić M, Leeuwen TV, Clark RM, Rombauts S, Rouzé P, Grbić V, et al. (2011) The genome of Tetranychus urticae reveals herbivorous pest adaptations. Nature 479: 487–492. pmid:22113690
- 8. Kwon DH, Park JH, Lee SH (2013) Screening of lethal genes for feeding RNAi by leaf disc-mediated systematic delivery of dsRNA in Tetranychus urticae. Pestic Biochem Phys 105: 69–75. pmid:24238293
- 9. Riga M, Tsakireli D, Ilias A, Morou E, Myridakis A, Stephanou EG, et al. (2014) Abamectin is metabolized by CYP392A16, a cytochrome P450 associated with high levels of acaricide resistance in Tetranychus urticae. Insect Biochem Molec 46: 43–53. pmid:24463358
- 10. Bustin SA, Benes V, Garson JA, Hellemans J, Huggett J, Kubista M, et al. (2009) The MIQE guidelines: minimum information for publication of quantitative real-time PCR experiments. Clin Chem 55: 611–622. pmid:19246619
- 11. 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: 5245–5250. pmid:15289330
- 12. 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: 509–515. pmid:15127793
- 13. Silver N, Best S, Jiang J, Thein SL (2006) Selection of housekeeping genes for gene expression studies in human reticulocytes using real-time PCR. BMC Mol Biol 7: 33. pmid:17026756
- 14. Livak KJ, Schmittgen TD (2001) Analysis of relative gene expression data using real-time quantitative PCR and the 2-ΔΔCt method. Methods 25, 402–408. pmid:11846609
- 15. Hellemans J, Mortier G, De Paepe A, Speleman F, Vandesompele J (2007) qBase relative quantification framework and software for management and automated analysis of real-time quantitative PCR data. Genome Biol 8: R19. pmid:17291332
- 16. Ferguson BS, Nam H, Hopkins RG, Morrison RF (2010) Impact of reference gene selection for target gene normalization on experimental outcome using real-time qRT-PCR in adipocytes. PLOS ONE 5: e15208. pmid:21179435
- 17. Li RM, Xie W, Wang SL, Wu QJ, Yang NN, Yang X, et al. (2013) Reference gene selection for qRT-PCR analysis in the sweetpotato whitefly, Bemisia tabaci (Hemiptera: Aleyrodidae). PLOS ONE 8: e53006. pmid:23308130
- 18. Fu W, Xie W, Zhang Z, Wang SL, Wu QJ, Liu Y, et al. (2014) Exploring valid reference genes for quantitative real-time PCR analysis in Plutella xylostella (Lepidoptera: Plutellidae). Int J Biol Sci 9: 792–802.
- 19. Zhu X, Yuan M, Shakeel M, Zhang YJ, Wang SL, Wang X, et al. (2014) Selection and evaluation of reference genes for expression analysis using qRT-PCR in the beet armyworm Spodoptera exigua (Hübner)(Lepidoptera: Noctuidae). PLOS ONE 9: e84730. pmid:24454743
- 20. Shi XQ, Guo WC, Wan PJ, Zhou LT, Ren XL, Ahmat T, et al. (2013) Validation of reference genes for expression analysis by quantitative real-time PCR in Leptinotarsa decemlineata (Say). BMC Res Notes 6: 93. pmid:23497596
- 21. Yuan M, Lu YH, Zhu X, Wan H, Shakeel M, Zhan S, et al. (2014) Selection and evaluation of potential reference genes for gene expression analysis in the Brown Planthopper, Nilaparvata lugens (Hemiptera: Delphacidae) using reverse-transcription quantitative PCR. PLOS ONE 9: e86503. pmid:24466124
- 22. Lu YH, Yuan M, Gao XW, Kang TH, Zhan S, Wan H, et al. (2013) Identification and validation of reference genes for gene expression analysis using quantitative PCR in Spodoptera litura (Lepidoptera: Noctuidae). PLOS ONE 8: e68059. pmid:23874494
- 23. Yang CX, Pan HP, Liu Y, Zhou XG (2014) Selection of reference genes for expression analysis using quantitative real-time PCR in the pea aphid, Acyrthosiphon pisum (Harris) (Hemiptera, Aphidiae). PLOS ONE 9: e110454. pmid:25423476
- 24. Shen GM, Jiang HB, Wang XN, Wang JJ (2010) Evaluation of endogenous references for gene expression profiling in different tissues of the oriental fruit fly Bactrocera dorsalis (Diptera: Tephritidae). BMC Mol Biol 11: 76. pmid:20923571
- 25. Sinha DK, Smith M (2014) Selection of reference genes for expression analysis in Diuraphis noxia (Hemiptera: Aphididae) fed on resistant and susceptible wheat plants. Sci Rep 4:
- 26. Sun W, Jin Y, He L, Lu WC, Li M (2010) Suitable reference gene selection for different strains and developmental stages of the carmine spider mite, Tetranychus cinnabarinus, using quantitative real-time PCR. J Insect Sci 10: 208. pmid:21265619
- 27. Yue XL, Gao XJ, Wang JJ, Lv JJ, Shen HM (2013) Selection of reference genes and study of the expression of detoxifying enzymes of Tetranychus urticae (Acari: Tetranychidae). Scientia Agricultura Sinica 46: 4542–4549.
- 28. Niu JZ, Dou W, Ding TB, Yang LH, Shen GM, Wang JJ (2012). Evaluation of suitable reference genes for quantitative RT-PCR during development and abiotic stress in Panonychus citri (McGregor) (Acari: Tetranychidae). Mol Biol Rep 39: 5841–5849. pmid:22203483
- 29. Liang P, Guo YJ, Zhou XG, Gao XW (2014) Expression Profiling in Bemisia tabaci under insecticide treatment: indicating the necessity for custom reference gene selection. PLOS ONE 9: e87514. pmid:24498122
- 30. Veazey KJ, Golding MC (2011) Golding selection of stable reference genes for quantitative RT-PCR comparisons of mouse embryonic and extra-embryonic stem cells. PLOS ONE 6: 27592.