Evaluation of candidate reference genes for gene expression analysis in the brassica leaf beetle, Phaedon brassicae (Coleoptera: Chrysomelidae)

Abstract

To accurately determine of target gene expression and eliminate the technical variation among the tested samples, one or several reference genes with stable expression are required as internal controls to normalize the data and to make accurate comparisons among experimental conditions. According to the reference genes documented in entomological research, the most commonly used genes are actin (ACT), tubulin (TUB), TATA-Box binding protein (TBP), glyceraldehyde-3-phosphate dehydrogenase (GAPDH), elongation factor 1-alpha (EF-1α), ribosomal proteins (RPs), and 18S ribosomal RNA (18S) [11–13]. [...]optimal sets of reference genes were recommended for the respective experimental conditions. Experimental treatment and sample collection For the gene expression analysis, the expression of candidate reference genes was tested in different developmental stages, tissues, and treatments to evaluate the stability of the candidate genes.

About the Authors:

Long Ma

Roles Methodology, Writing – original draft, Writing – review & editing

Affiliation: College of Life Sciences, Jiangxi Science & Technology Normal University, Nanchang, China

Ting Jiang

Roles Conceptualization, Methodology

Affiliation: Institute of Entomology, Jiangxi Agricultural University, Nanchang, China

Xiangya Liu

Roles Conceptualization, Methodology

Affiliation: Institute of Entomology, Jiangxi Agricultural University, Nanchang, China

Haijun Xiao

Roles Funding acquisition, Resources

Affiliation: Institute of Entomology, Jiangxi Agricultural University, Nanchang, China

Yingchuan Peng

Roles Validation

Affiliation: Institute of Entomology, Jiangxi Agricultural University, Nanchang, China

Wanna Zhang

Roles Formal analysis, Methodology

* E-mail: [email protected]

Affiliation: Institute of Entomology, Jiangxi Agricultural University, Nanchang, China

ORCID logo https://orcid.org/0000-0002-5646-6446

Introduction

The brassica leaf beetle, Phaedon brassicae Baly, is a notorious defoliator of crucifers and is widely distributed in East and South Asia with high fecundity [1, 2]. This beetle escapes from predators by dropping from host plants, and interestingly, the larvae exhibit less frequent dropping behavior than adults in response to attacks [3]. In the Yangtze River Valley, there are two distinct infestation peaks in the field: the single spring generation and the two generations in autumn, which undergo aestivating and hibernating imaginal diapause in soil, separately [1, 4, 5]. This beetle is a typical short-day species in which low temperature enhances the induction of its winter diapause, while high temperature suppresses the incidence of its summer diapause [1, 5]. In the last decades, this beetle has become a secondary chewing pest of brassicaceous vegetables in China, and recently, it has occurred frequently and caused potential threats to vegetable products. However, the application of insecticides is not always effective against P. brassicae due to its complex life history and high fecundity [4], and previous studies have explored a cadherin-based peptide as an enhancer for Cry3Aa-based products in controlling P. brassicae [2, 6]. To identify novel target genes for controlling P. brassicae, the accurate quantification of gene expression under different conditions is indispensable.

Gene transcription patterns in different tissues and developmental stages provide deep insights into their biological functions [7]. Taking advantage of high-throughput transcriptome sequencing, transcriptome analysis and data mining have become efficient in screening differentially expressed genes and quantifying the expression abundance of their transcripts. Real-time quantitative PCR (RT-qPCR) has become a powerful tool for validating gene expression profiles owing to its accuracy, specificity, sensitivity, dynamic range, and reproducibility [8, 9]. However, several factors, such as RNA purity and integrity, reverse transcription and PCR efficiency, and pipetting errors, can affect the accuracy of RT-qPCR [10]. In RT-qPCR, a common practice to calibrate target expression is to measure the expression of an internal control, namely reference gene, synchronously in the same sample [8]. Generally, reference genes are housekeeping genes that are constitutively expressed to maintain basic cellular function, and the selection of suitable reference genes has become a necessary step prior to RT-qPCR. To accurately determine of target gene expression and eliminate the technical variation among the tested samples, one or several reference genes with stable expression are required as internal controls to normalize the data and to make accurate comparisons among experimental conditions. According to the reference genes documented in entomological research, the most commonly used genes are actin (ACT), tubulin (TUB), TATA-Box binding protein (TBP), glyceraldehyde-3-phosphate dehydrogenase (GAPDH), elongation factor 1-alpha (EF-1α), ribosomal proteins (RPs), and 18S ribosomal RNA (18S) [11–13]. Ideally, reference genes should exhibit constant expression with respect to different developmental stages, tissues, or treatment conditions and do not co-regulate with the target gene. However, the transcription levels of such reference genes are not always stable under various biotic conditions (e.g. different tissues or developmental stages) or abiotic conditions (e.g. pesticide exposure or thermal stress), possibly leading to inconsistent results [14–16].

To date, a large body of research has suggested that there is no ‘universal’ reference gene applicable for various experimental conditions and all tissue types, even within the same insect species [10, 15, 17]. For instance, in the predatory lady beetle Hippodamia convergens, 28S, EF1A, and CypA were the best reference genes across different developmental stages, while GAPDH, CypA, and 28S were the most stable in different tissues; GAPDH and CypA were the most stable under photoperiod conditions [15]. Similarly, in Aphidius gifuensis, Colaphellus bowringi, Mythimna separata and Harmonia axyridis, the optimal set of reference genes significantly varies with developmental stage, gender, and diet [12, 18–20]. Taken together, these studies demonstrated that the selection of reliable reference genes under specific experimental conditions was pivotal before normalizing target gene expression.

The objective of this study was to discern the solidly expressed reference genes in P. brassicae across different developmental stages, in various tissues and in response to abiotic perturbations (thermal stress and pesticide exposure) for RT-qPCR analysis. For this purpose, four different analytical tools (geNorm, NormFinder, BestKeeper, and the comparative ΔCt algorithm) were used to assess the stability of candidate reference genes. These results were conclusively integrated in RefFinder to provide an overall ranking of the candidate reference genes. A target gene was selected to verify our findings. As a result, optimal sets of reference genes were recommended for the respective experimental conditions. To our knowledge, this is the first report of a comprehensive evaluation of reference genes in P. brassicae, and our results would facilitate the future research on functional studies of P. brassicae.

Materials and methods

Insect rearing

Individuals of P. brassicae were seized from a natural population in the experimental radish plantation of Jiangxi Agricultural University (Nanchang, Jiangxi, China). This strain has been reared for six generations without exposure to chemical insecticides. Laboratory rearing was conducted on radish (Raphanus sativus var. longipinnatus) leaves in transparent plastic containers (7.5 cm wide at base and 15.0 cm deep) under conditions of 25 ± 1°C, with relative humidity at 70 ± 10%, and a 12:12 h light:dark photoperiod. Fresh leaves were provided daily.

Experimental treatment and sample collection

For the gene expression analysis, the expression of candidate reference genes was tested in different developmental stages, tissues, and treatments to evaluate the stability of the candidate genes. To investigate the developmental expressions, specimens were sampled from eggs, larvae (first, second, third and fourth instar), pupae, and adults. For tissue expression analysis, tissues (including head, cuticula, fat body and gut) were dissected from the third-instar larvae in ice-cold phosphate-buffered saline (PBS). For the insecticide exposure experiment, the second-instar larvae were treated with sublethal doses of acetamiprid (2 μg/ml), dinotefuran (5 μg/ml), and abamectin (0.18 μg/ml) for 48 h, respectively, while an equal dose of PBS was applied to individuals set as controls. Briefly, the leaf discs of radish (2 cm diameter) were dipped in the solution with insecticide for 10 s and then left to air-dry at room temperature. Each piece of dipped leaf discs was placed in a Petri dish, into which four larvae were transferred. To examine the influence of temperature, the third-instar larvae were subjected to cold shock (10°C), heat shock (40°C), and control condition (25°C) for 2 h and 4 h, respectively. Each experiment contained at least 24 individuals, and was performed in three biological replicates. All samples were frozen immediately in liquid nitrogen and stored at -80°C until RNA isolation.

RNA isolation and cDNA synthesis

Total RNA was isolated using the Trizol reagent (Thermo Scientific, China) following the manufacturer’s protocol. The RNA quality was examined by 1% agarose gel electrophoresis, and its quantity was evaluated with an Agilent 2100 Bioanalyzer (Agilent Technologies). RNA samples with a value of OD 260/280 ratio between 1.8 and 2.0 were further applied into cDNA synthesis. After the removal of residual genomic DNA with DNase I (Promega, Madison, USA), 1 μg of purified total RNA was applied to the first-strand cDNA synthesis using a Fast Quant RT kit (Tiangen, Beijing, China).

Identification of candidate reference genes

To identify the stably expressed genes, eight candidate reference genes, namely Actin1, Actin2, EF-1α, GAPDH, α-Tub, RPL19, RPL32, and TBP, were selected from the P. brassicae RNA-seq transcriptome database (S1 Table). Prior to RT-qPCR detection, the open reading frames of these genes were confirmed by PCR amplification using specific primers. Other primer pairs used for RT-qPCR were designed using Primer Express software 3.0.1 (Applied Biosystems). Furthermore, primer specificity was screened by 1.5% agarose electrophoresis after PCR amplification.

Quantitative real-time PCR analysis

RT-qPCR was performed on a CFX96 Touch Real-time PCR detection system (Bio-Rad) using SYBR Green SuperReal PreMix Plus (Tiangen, Beijing, China). Amplifications were carried out under the following conditions, initial denaturation at 95°C for 10 min followed by 40 cycles of 5 s at 95°C and for 30 s at 60°C, followed by a melting curve stage (60 to 95°C) to confirm gene-specific amplification. After 5-fold dilution of cDNA template, 2 μl cDNA sample was incorporated in RT-qPCR reaction for a total volume of 25 μl. Each reaction was carried out in triplicate and the average cycle threshold (Ct) values of triplicates were calculated. Meanwhile, negative control without template was performed. A standard curve for each primer pair was constructed with serial dilutions of cDNA samples, and the corresponding amplification efficiency for each candidate gene was calculated following the equation: E = (10[−1/slope] −1)×100.

Stability analysis of reference gene expression

After RT-qPCR measurement, four algorithms, namely geNorm, NormFinder, BestKeeper, and the comparative ΔCt method, were employed to evaluate the stability of each candidate reference gene. In brief, the geNorm algorithm provides its ranking based on the mean pairwise variation (V-value) between all the tested genes to calculate the expression stability value (M) of genes, and a lower M-value indicates a higher stability [8]. In addition, geNorm was utilized to define the optimal number of reference genes credible for normalization, wherein a V-value within a threshold of 0.15 was set to determine whether additional reference genes were necessary. NormFinder, an Excel-based applet, ranks the gene expression stability based on the evaluation of their intra- and inter-group variation and a separate analysis of the sample subgroups in expression [21]. BestKeeper evaluates expression stability based on an index obtained by calculating the Ct set standard deviation (SD) and coefficient of variance (CV) [22]. Finally, a comprehensive analysis tool RefFinder (https://www.heartcure.com.au/reffinder/#) was used to integrate the results of the four different analytical methods and assess the rank of the reference genes based on their geometric mean [23].

Validation of the recommended reference genes

In insects, small heat shock proteins (sHSPs) function as molecular chaperones to protect cells from harsh conditions, prevent irreversible protein aggregation and become active in response to thermal stress [24]. To examine the reliability of the selected reference genes, the expression levels of sHSP20.0 (accession number MW538931), a target gene, were examined during exposure of P. brassicae to different thermal conditions (10, 25 and 40°C) for 2 h. After RT-qPCR, the optimal combination of reference genes (RPL32 and TBP), the optimal reference gene (RPL32), and the least stable gene (GAPDH) were selected for RT-qPCR normalization of the target gene, respectively. It is noteworthy that the normalization against two reference genes was performed using the geometric mean of the normalization factors. The relative expression of sHSP20.0 in each sample was calculated using the 2−ΔΔCt method [25].

Results

Determination of primer specificity and efficiency

For each candidate reference gene, a single amplicon was detected by agarose gel electrophoresis, ranging from 125 to 207 bp (S1 Fig). Consistent with this result, a single peak was observed in the melting curve analysis (S2 Fig). Furthermore, the amplification efficiency of eight candidate reference genes ranged from 96.16 to 106.75%, with correlation coefficients (R2) of more than 99% (Table 1; S3 Fig).

[Figure omitted. See PDF.]

Table 1. Primers for candidate reference genes used in RT-qPCR analyses.

https://doi.org/10.1371/journal.pone.0251920.t001

Ct values of candidate reference genes under different experimental conditions

The mean Ct values for the eight candidate reference genes ranged from 18 to 27 under the given experimental conditions (Fig 1). For different developmental stages, EF-1α exhibited the highest expression, followed by RPL32, RPL19 and α-TUB. Based on the tissue expression profiles, the most expressed reference gene was EF-1α, followed by RPL19 and α-TUB. According to their integrative performance in different treatments, RPL19 exhibited the lowest variation (below 2 cycles) in expression, and EF-1α and Actin2 showed the highest and lowest transcription levels, respectively.

[Figure omitted. See PDF.]

Fig 1. The cycle threshold (Ct) values of eight reference genes under different treatment conditions.

The vertical bars represent the standard deviation.

https://doi.org/10.1371/journal.pone.0251920.g001

Stability analysis of reference genes under different experimental conditions

In regards to the development expression analysis, RPL19 and TBP were ranked as the top two most stably expressed genes by NormFinder and geNorm, and the top candidates were EF-1α and TBP according to Bestkeeper and ΔCt method, respectively (Table 2). For tissue expression analysis based on Bestkeeper and ΔCt method, the top three candidates were RPL19, RPL32, and EF-1α. Besides, both NormFinder and geNorm recommended EF-1α as the most stably expressed gene. Under different temperature conditions, NormFinder and geNorm ranked RPL32, TBP, and α-TUB as the top three reference genes. Besides, both Bestkeeper and ΔCt method identified RPL19 as the most stable candidate. In the pesticide exposure experiment, the overall order based on NormFinder from the most stable to the least stable reference gene was: RPL32, α-TUB, RPL19, Actin2, TBP, EF-1α, GAPDH, Actin1. In addition, Bestkeeper identified TBP as the most stable one, and RPL19 showed the optimal stability based on the ΔCt method.

[Figure omitted. See PDF.]

Table 2. Stability of reference gene expression under different experimental conditions calculated by four different analytical tools.

https://doi.org/10.1371/journal.pone.0251920.t002

Integrating the evaluation of four programs, a comprehensive ranking of candidate reference genes was determined by RefFinder (Fig 2; Table 2). The results indicated that RPL19 and TBP were considered as the best reference genes across different developmental stages, RPL32 and EF-1α were recommended for tissue expression analysis, RPL32 and TBP were identified as the optimal candidates for the temperature experiment, and RPL32 and RPL19 were selected for the insecticide treatment.

[Figure omitted. See PDF.]

Fig 2. The stability of candidate reference gene expression under different conditions based on RefFinder.

The expression stabilities of candidate reference genes were evaluated in diverse conditions including developmental stage (A), tissue (B), temperature stress (C), and insecticide exposure (D).

https://doi.org/10.1371/journal.pone.0251920.g002

Quantitative analysis of candidate reference genes based on geNorm

The geNorm algorithm provided the optimal number of reference genes for credible normalization under a given experimental condition. For all tested treatments, our results indicated that all pairwise variations were under a predetermined threshold of 0.15 (Fig 3), suggesting that two reference genes were sufficient for the normalization of the target gene. Taking the developmental expression as an example, the first V-value < 0.15 was observed at V2/3, addressing that two reference genes were sufficient for reliable normalization.

[Figure omitted. See PDF.]

Fig 3. Optimal number of reference genes used for normalization of gene expression by geNorm program.

A value of pairwise variation (Vn/Vn+1) below 0.15 suggested that no extra gene was required for normalization of gene expression in this condition.

https://doi.org/10.1371/journal.pone.0251920.g003

Validation of selected reference genes

To confirm the reliability of the selected reference genes, the expression of sHSP20.0, a target gene, was detected under different thermal treatments (Fig 4). In the temperature treatment (from 25 to 40 °C), when the optimal reference set (RPL32 and TBP) or the optimal reference (RPL32) alone was utilized to normalize the target, the expressions of sHSP20.0 were up-regulated by 3.4 and 2.6 times, respectively, and significant difference was detected (P<0.01). In comparison, when the least stable reference gene GAPDH served as a normalizer, only a 0.8-fold increase was observed in sHSP20.0 expression (P<0.05), showing a decreased percentage of increase compared to those using the former two normalizers.

[Figure omitted. See PDF.]

Fig 4. The relative expression of target sHSP20.0 normalized by the recommended set of references (RPL32 and TBP), the optimal reference (RPL32) and the least stable gene (GAPDH), respectively.

Columns represent the expressions of sHSP20.0 in Phaedon brassicae when subjected to cold shock (10°C), heat shock (40°C) and the control (25°C) for 2 h. Error bars indicate SE; *, P < 0.05; **, P < 0.01 (Student’s t-test by SAS 9.20).

https://doi.org/10.1371/journal.pone.0251920.g004

Discussion

Currently, RT-qPCR is extensively used to quantify gene expression [9]. Several concerns have been raised regarding the variations in RT-qPCR method, and normalization of RT-qPCR data with reference genes is a commonly used strategy to calibrate experimental errors introduced by RT-qPCR [10]. It is noteworthy that the selection of reference genes with low expression variation is a prerequisite to ensure valid normalization and avoid inaccurate quantification of gene expression [14, 19]. Previous studies have documented that many frequently used reference genes differed significantly in their expression profiles under different treatments [11, 19, 26]. Therefore, before RT-qPCR operation, each candidate reference gene should be evaluated under specific experimental conditions to ensure a stable expression level, and this step has become a routine practice before using them to normalize target gene expression.

Our results revealed that the stability of a reference gene could be changed under different experimental conditions. However, many previous studies utilized a single endogenous control for different treatments and life stages to quantify gene expression, which can significantly affect statistical analyses and may result in false data interpretation [27]. Therefore, it is imperative to identify the optimal reference genes for specific conditions in a given species. Furthermore, our study revealed that the stability ranking of these reference genes was variable in certain circumstances due to the different algorithms used in the four analytical tools. For instance, when P. brassicae was subjected to pesticide exposure, RPL32 was listed as the optimal by NormFinder, and TBP was the top choice by Bestkeeper, whereas RPL19 was recommended by geNorm. To address this challenge, an integrated analysis to evaluate the dataset becomes necessary, and adopting the multiple instead of a single normalizer for RT-qPCR analysis is in demand. For these purposes, RefFinder integrates the results of these computational programs and calculates a comprehensive ranking value for candidate genes [15, 19, 20], and meanwhile, a suite of reference genes is specifically recommended instead of a single normalizer. Herein, the optimal number of reference genes was determined by geNorm, and our results proved that two references were sufficient for reliable normalization in each given condition.

Ribosomal proteins are known to play an essential role in ribosome assembly, and they, in conjunction with four ribosomal RNA (rRNA), make up the ribosomal subunits responsible for cellular protein translation [28]. Our results demonstrated that the ribosomal protein RPL19 was expressed stably across life stages and under pesticide application, and RPL32 showed high stability under thermal and pesticide treatment. Consistent with our results, many ribosomal proteins have been documented as the optimal reference for many insect species. Likewise, in the cabbage beetle C. bowringi, RPL19 was identified as the optimal reference for different sexes and under photoperiod treatments [20]. In other coleopterans, RP18 and RP4 were regarded as the most stable house-keeping genes in Leptinotarsa decemlineata (Say) [29], RpS9 showed a steady expression under a range of temperatures in Diabrotica undecimpunctata howardi [30], and RPL22e was selected as the suitable reference in different sexes of Mylabris cichorii [31]. Besides, RpS3 and RpS13A showed the highest stability under ultraviolet irradiation in Tribolium castaneum [32]. Moreover, in the aphid parasitoid wasp A. gifuensis, RPL13 was recommended as the optimal under diverse conditions including different developmental stages, sexes, and diverse diets [18]. In a tetranychid mite Tetranychus urticae, Rp49 was suitable not only for host plant shift studies but also for the investigations of acaricide susceptible and resistant populations [33]. EF-1α is a ubiquitous and conserved cytosolic protein among eukaryotic organisms and is responsible for catalyzing the binding of aminoacyl-transfer RNAs to the ribosome [34, 35]. Our study indicated that EF-1α showed the best performance in diverse developmental stages and tissues. Studies of other coleopterans (i.e. H. convergens and D. undecimpunctata howardi) also showed that EF-1α acted as the best reference gene across life stages [15, 30]. Similar results were documented in several lepidopteran species (i.e. M. separata, Danaus plexippus, and Diaphania caesalis), where EF-1α was identified as the most stable reference across life stages and in tissues [12, 36, 37].

One surprising finding was that the traditional reference gene GAPDH was listed as the least reliable reference gene in most experimental conditions. Likewise, the instability of GAPDH expression has been documented in different developmental stages and tissues of C. bowringi [20] and D. caesalis [36], in H. convergens under thermal stress [15], and in M. separata after pesticide exposure [12]. Previous literature has documented that GAPDH functions as a glycolytic enzyme involved in glycolysis [38], and it was presumed that any perturbation toward energy metabolism would have a potential impact on GAPDH expression. Recent evidence suggests that GAPDH is associated with cell proliferation under adverse conditions where its catalytic activity is impaired [38]. Considering these issues, it is inappropriate to adopt GAPDH as a reference under several abiotic stress conditions, such as starvation, pesticides, and thermal stress.

To further validate the reliability of the optimal reference genes in P. brassicae, the expression of a target gene sHSP20.0 was investigated under different thermal stress. As molecular chaperones, sHSPs assist in the correct folding of nascent proteins and combat protein aggregation induced by stresses, especially under thermal stress [24, 39]. In T. castaneum, hsp18.3 was dramatically up-regulated in response to enhanced heat stress but not to cold stress [24]. Similarly, sHSP19.1, from the oak silkworm, was strongly induced after heat shock [40]. Our results showed that sHSP20.0 expression was inconsistent when normalized to the least stable reference compared with that when normalized to the optimal reference set or the optimal reference alone. These findings revealed that the arbitrary selection of reference genes would lead to inaccurate or contradictory results for target genes [15, 41], and our results demonstrated that the combined use of optimal reference genes ensures greater accuracy in gene expression analysis.

In conclusion, our results demonstrated that unstable reference genes might result in incorrect interpretation of RT-qPCR results, and the optimal reference gene recommendations could avoid such bias in normalization. To date, this is the first study to investigate candidate reference genes for gene expression analysis in P. brassicae, and our findings would lay a foundation for functional research in P. brassicae.

Supporting information

S1 Fig. The agarose gel electrophoresis of eight candidate reference genes.

M, marker. Templates in the PCR reactions were as follows: 1) Actin2, 2) GAPDH, 3) RPL32, 4) α-TUB, 5) Actin1, 6) Ef-1α, 7) TBP, and 8) RPL19.

https://doi.org/10.1371/journal.pone.0251920.s001

(TIF)

S2 Fig. Melting curve analysis of eight candidate reference genes.

The gene-specific amplification was confirmed by a single peak in melting-curve analysis.

https://doi.org/10.1371/journal.pone.0251920.s002

(TIF)

S3 Fig. Standard curves of eight candidate reference genes.

https://doi.org/10.1371/journal.pone.0251920.s003

(TIF)

S1 Table. Sequences of candidate reference genes.

https://doi.org/10.1371/journal.pone.0251920.s004

(DOCX)

S1 Raw images. The raw image of S1 Fig.

https://doi.org/10.1371/journal.pone.0251920.s005

(PDF)

Citation: Ma L, Jiang T, Liu X, Xiao H, Peng Y, Zhang W (2021) Evaluation of candidate reference genes for gene expression analysis in the brassica leaf beetle, Phaedon brassicae (Coleoptera: Chrysomelidae). PLoS ONE 16(6): e0251920. https://doi.org/10.1371/journal.pone.0251920

References

1. Wang X, Zhou X, Lei C. Development, survival and reproduction of the Brassica leaf beetle, Phaedon brassicae Baly (Coleoptera: Chrysomelidae) under different thermal conditions. Pan-Pacific Entomologist. 2007; 83: 143–151.

2. Kim E, Jeoung S, Park Y, Kim K, Kim Y. A novel formulation of Bacillus thuringiensis for the control of Brassica leaf beetle, Phaedon brassicae (Coleoptera: Chrysomelidae). Journal of Economic Entomology. 2015; 108: 2556–2565. pmid:26470390

3. Matsubara S, Sugiura S. Host plant architecture afects the costs of dropping behaviour in Phaedon brassicae (Coleoptera: Chrysomelidae). Applied Entomology and Zoology. 2018; 53: 501–508.

4. Wang X, Xue F, Tan Y, Lei C. The role of temperature and photoperiod in diapause induction in the brassica leaf beetle, Phaedon brassicae (Coleoptera: Chrysomelidae). European Journal of Entomology. 2007; 104(4): 693–697.

5. Wang X, Zhou X, Lei C. Adaptation to aging and absence of host-plant in the brassica leaf beetle, Phaedon brassicae. Chinese Bulletin of Entomology. 2009; 46: 403–407.

6. Gao Y, Jurat-Fuentes JL, Oppert B, Fabrick JA, Liu C, Gao J, et al. Increased toxicity of Bacillus thuringiensis Cry3Aa against Crioceris quatuordecimpunctata, Phaedon brassicae and Colaphellus bowringi by a Tenebrio molitor cadherin fragment. Pest Management Science. 2011; 67(9): 1076–1081. pmid:21495115

7. Kozera B, Rapacz M. Reference genes in real-time PCR. Journal of Applied Genetics. 2013; 54: 391–406. pmid:24078518

8. Vandesompele J, De Preter K, Pattyn F, Poppe B, Van Roy N, De Paepe A, et al. Accurate normalization of real-time quantitative RT-PCR data by geometric averaging of multiple internal control genes. Genome Biology. 2002; 3: research0034. pmid:12184808

9. Costa-Silva J, Domingues D, Lopes FM. RNA-Seq differential expression analysis: An extended review and a software tool. PloS ONE. 2017; 12: e0190152. pmid:29267363

10. Bustin SA, Benes V, Nolan T, Pfaffl MW. Quantitative real-time RT-PCR—a perspective. Journal of Molecular Endocrinology. 2005; 34: 597–601. pmid:15956331

11. Shakeel M, Rodriguez A, Tahir UB, Jin F. Gene expression studies of reference genes for quantitative real-time PCR: an overview in insects. Biotechnology Letters. 2018; 40: 227–236. pmid:29124515

12. Li H, Dai C, Zhang C, He Y, Ran H, Chen S. Screening potential reference genes for quantitative real-time PCR analysis in the oriental armyworm, Mythimna separata. PLoS ONE. 2018; 13: e0195096. pmid:29617430

13. Nagy NA, Németh Z, Juhász E, Póliska S, Rácz R, Kosztolányi A, et al. Evaluation of potential reference genes for real-time qPCR analysis in a biparental beetle, Lethrus apterus (Coleoptera: Geotrupidae). PeerJ. 2017; 5: e4047. pmid:29201562

14. Tellin O, Zorzi W, Lakaye B, De Borman B, Coumans B, Hennen G, et al. Housekeeping genes as internal standards: use and limits. Journal of Biotechnology. 1999; 75: 291–295. pmid:10617337

15. Pan H, Yang X, Siegfried BD, Zhou X. A comprehensive selection of reference genes for RT-qPCR analysis in a predatory lady beetle, Hippodamia convergens (Coleoptera: Coccinellidae). PLoS ONE. 2015; 10: e0125868. pmid:25915640

16. Zhang S, An S, Li Z, Wu F, Yang Q, Liu Y, et al. Identification and validation of reference genes for normalization of gene expression analysis using qRT-PCR in Helicoverpa armigera (Lepidoptera: Noctuidae). Gene. 2015; 555: 393–402. pmid:25447918

17. Silver N, Best S, Jiang J, Thein SL. Selection of housekeeping genes for gene expression studies in human reticulocytes using real-time PCR. BMC Molecular Biology. 2006; 7: 33. pmid:17026756

18. Gao X, Zhang S, Luo J, Wang C, Lü L, Zhang L, et al. Identification and validation of reference genes for gene expression analysis in Aphidius gifuensis (Hymenoptera: Aphidiidae). PLoS ONE. 2017; 12: e0188477. pmid:29190301

19. Yang X, Pan H, Yuan L, Zhou X. Reference gene selection for RT-qPCR analysis in Harmonia axyridis, a global invasive lady beetle. Scientific Reports. 2018; 8: 2689. pmid:29426915

20. Tan Q, Zhu L, Li Y, Liu W, Ma W, Lei C, et al. A de novo transcriptome and valid reference genes for quantitative Real-Time PCR in Colaphellus bowringi. PLoS ONE. 2015; 10: e0118693. pmid:25692689

21. Andersen CL, Jensen JL, Ørntoft TF. 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. 2004; 64: 5245–5250. pmid:15289330

22. Pfaffl MW, Tichopad A, Prgomet C, Neuvians TP. Determination of stable housekeeping genes, differentially regulated target genes and sample integrity: BestKeeper-Excel-based tool using pair-wise correlations. Biotechnology Letters. 2004; 26: 509–515. pmid:15127793

23. Xie F, Xiao P, Chen D, Xu L, Zhang B. miRDeepFinder: a miRNA analysis tool for deep sequencing of plant small RNAs. Plant Molecular Biology. 2012; 80(1): 75–84. pmid:22290409

24. Xie J, Hu X, Zhai M, Yu X, Song X, Gao S. Characterization and functional analysis of hsp18.3 gene in the red flour beetle, Tribolium castaneum. Insect Science. 2019; 26: 263–273. pmid:28980406

25. Livak KJ, Schmittgen TD. Analysis of relative gene expression data using real-time quantitative PCR and the 2−ΔΔCt Method. Methods. 2001; 25: 402–408. pmid:11846609

26. Lü J, Chen S, Guo M, Ye C, Qiu B, Wu J, et al. Selection and Validation of Reference Genes for RT-qPCR Analysis of the Ladybird Beetle Henosepilachna vigintioctomaculata. Frontiers in Physiology. 2018; 9: 1614. pmid:30483159

27. Ferguson BS, Nam H, Hopkins RG, Morrison RF. Impact of reference gene selection for target gene normalization on experimental outcome using real-time qRT-PCR in adipocytes. PloS One. 2010; 5: e15208. pmid:21179435

28. Zhou X, Liao W, Liao J, Liao P, Lu H. Ribosomal proteins: functions beyond the ribosome. Journal of Molecular Cell Biology. 2015; 7: 92–104. pmid:25735597

29. Shi X, Guo W, Wan P, Zhou L, Ren X, Ahmat T, et al. Validation of reference genes for expression analysis by quantitative real-time PCR in Leptinotarsa decemlineata (Say). BMC Research Notes. 2013; 6: 93. pmid:23497596

30. Basu S, Pereira AE, Pinheiro DH, Wang H, Valencia-Jiménez A, Siegfried BD, et al. Evaluation of reference genes for real-time quantitative PCR analysis in southern corn rootworm, Diabrotica undecimpunctata howardi (Barber). Scientific Reports. 2019; 9(1): 10703. pmid:31341190

31. Wang Y, Wang Z, Huang Y, Liao Y, Yin Y. Identification of suitable reference genes for gene expression studies by qRT-PCR in the blister beetle Mylabris cichorii. Journal of Insect Science. 2014; 14: 94. pmid:25368050

32. Sang W, He L, Wang X, Zhu-Salzman K, Lei C. Evaluation of reference genes for RT-qPCR in Tribolium castaneum (Coleoptera: Tenebrionidae) under UVB Stress. Environmental Entomology. 2015; 44: 418–425. pmid:26313197

33. Morales MA, Mendoza BM, Lavine LC, Lavine MD, Walsh DB, Zhu F. Selection of reference genes for expression studies of xenobiotic adaptation in Tetranychus urticae. International Journal of Biological Sciences. 2016; 12(9): 1129–1139. pmid:27570487

34. Kristensen R, Torp M, Kosiak B, Holst-Jensen A. Phylogeny and toxigenic potential is correlated in Fusarium species as revealed by partial translation elongation factor 1 alpha gene sequences. Mycological Research. 2005; 109: 173–186. pmid:15839101

35. Tatsuka M, Mitsui H, Wada M, Nagata A, Nojima H, Okayama H. Elongation factor-1 alpha gene determines susceptibility to transformation. Nature. 1992; 359(6393): 333–336. pmid:1383827

36. Wang Z, Meng Q, Zhu X, Sun S, Liu A, Gao S, et al. Identification and evaluation of reference genes for normalization of gene expression in developmental stages, sexes, and tissues of Diaphania caesalis (Lepidoptera, Pyralidae). Journal of Insect Science. 2020; 20: 1–9.

37. Pan H, Yang X, Bidne K, Hellmich RL, Siegfried BD, Zhou X. Selection of Reference Genes for RT-qPCR Analysis in the Monarch Butterfly, Danaus plexippus (L.), a Migrating Bio-Indicator. PLoS One. 2015; 10(6): e0129482. pmid:26030778

38. Nicholls C, Li H, Liu J. GAPDH: a common enzyme with uncommon functions. Clinical and Experimental Pharmacology and Physiology. 2012; 39: 674–679. pmid:21895736

39. Chen X, Tan A, Palli SR. Identification and functional analysis of promoters of heat-shock genes from the fall armyworm, Spodoptera frugiperda. Scientific Reports. 2020; 10: 2363. pmid:32047182

40. Zhang J, Li Q, Sun Y, Tian J, Hu Z, Zhu B, et al. Molecular cloning and functional analysis of small heat shock protein 19.1 gene from the Chinese oak silkworm, Antheraea pernyi. Archives of Insect Biochemistry and Physiology. 2018; 99: e21516. pmid:30387888

41. Singh S, Pandher S, Gupta M, Kaur G, Rathore P. Reference Gene Selection in Phenacoccus solenopsis Tinsley (Hemiptera: Pseudococcidae) and Their Normalization Impact on Gene Expression in RNAi Studies. Journal of Economic Entomology. 2019; 112(1): 371–381. pmid:30329069