Identification of reference genes for studies of quantitative gene expression in male and female quail tissues

Abstract In industrial poultry, quail production has gained increasing prominence over the years. It is known that the intensification of genetic studies has contributed greatly to this growth, through techniques, such as analysis of gene expression by PCR, for example. This study aimed to evaluate stability and recommend reference genes for quantitative real-time PCR in different tissues from male and female broiler quails. The stability of 10 housekeeping genes (GAPDH, RPL5, MRPS27, MRPS30, TFRC, HMBS, EEF1, LDHA, B2M, and UBC) by means Bestkeeper, NormFinder, GeNorm softwares with ΔCq method. The tissues analyzed were: heart, thigh muscle, brain, and spleen, considering that they are tissues commonly used in nutrigenomic, immunological, and poultry performance research. As expected, the reference genes tested showed varying stability depending on the tissue evaluated. According to the present study, the most stable housekeeping genes were MRPS30, TFRC, and HMBS in heart; MRPS30, EEF1, and HMBS in thigh muscle; B2M, GAPDH, and UBC in brain; and EEF1, LDHA, and HMBS in spleen. Therefore, it is recommended to be used as reference genes for gene expression studies of male and female quails.


Introduction
For industrial poultry farming, quails rearing is an activity that has been emerging in the world.European quails (Coturnix coturnix coturnix) also known for meat production, began to develop commercially and gained space at expense of laying quails due to the first is considered an ease adaptation bird to breeding conditions, precocious sexual maturity, rapid growth, small feed consumption, great resistance to diseases, 1 and also used as an animal model in the laboratory.However, it is established that success in productivity depends not only because optimization of the production system considering only on animal performance; other factors are also extremely important; like genetic, for example.][4] In the genetic context, gene expression analysis has been strengthened in the scientific community, with constant improvements in methods promoting better access to various platforms.Moreover, the epigenome comprising different mechanisms, e.g., DNA methylation, remodeling, histone tail modifications, chromatin microRNAs, and long non-coding RNAs, interact with environmental factors like nutrition, pathogens, and climate to influence the expression profile of genes and the emergence of specific pheno-types. 5,6Multilevel interactions between the genome, epigenome, and environmental factors might occur. 7Furthermore, numerous lines of evidence suggest the influence of epigenome variation on health and production. 3,5,8he expression of eukaryotic genes is temporarily and multidimensionally controlled. 4Only a relatively small set of the entire genome is expressed in each type of tissue, and the expression of genes depends on the stage of development. 9Therefore, gene expression in eukaryotes is specific to each tissue. 2 Also, the number of gene products that are made in the same tissue as well as in other tissues that make up that product, regulates the expression of that gene. 10One of the basic activities in domestic animals is the study of genes and proteins related to economic traits and their study at the cellular or chromosomal level. 11mong the methods of genetic analysis, quantitative real-time polymerase chain reaction (RT-qPCR) is recurrent and one of the most used in gene expression studies.Through this technique, the genome sequence of interest is amplified billions of times, for ease of analysis, and in conjunction with a fluorophore, the quantification of these copies is performed in realtime. 13The results obtained from RT-qPCR require normalization and this can be accomplished with internal control for, besides other functions, the correction of the results that may be distorted due to different initial amounts of nucleic acid. 6,8,9he internal control analyses correspond to the study of a single or small set of genes, called reference genes, that have demonstrated significant expression on tissue and must remain invariable in every experimental condition, being their choice very influential and decisive in the results. 10,11The method commonly used to identify reliable reference genes is based on algorithms such geNorm, NormFinder, BestKeeper e Delta Ct (DCt), assessing genes stability based on quantification cycle (Cq) variance values in each physiological or experimental condition. 14Thus, the choice of reference genes must be made carefully, based on previous tests that prove their effectiveness, since various physiological and experimental conditions affect the levels of expression.
6][17][18][19] Some studies seek answers regarding animal genetic behavior in both sexes.For this purpose, is necessary to use the same set of reference genes to normalize the expression data from males and females samples, eliminating the sex effect. 12However, information about genes used as reference is still scarce for quails, reinforcing the importance of experiments that investigate their action, since these are considered based on the technique and, therefore, will influence subsequent genetic studies related to the current species.
Hence, the objective of this work was to assess 10reference genes stability and, according to the results, recommend them for RT-qPCR in different tissues of male and female broiler quails.

Animals
All procedures applied in this experiment were approved (process 09/2015) by the Animal Protection, Research, and Ethics Committee of the Federal University of Sergipe, Sergipe, Brazil.The tissues used in this study were: thigh muscle, brain, heart, and spleen, since they are commonly used in studies about animal productive performance and metabolic analyzes of an investigative nature.The tissues were collected from six broiler quails (Coturnix coturnix coturnix) at the age of 35 days, three males and three females (n ¼ 6).These animals were obtained from a commercial production poultry farm in São Crist ovão-SE city, raised under standard management conditions with nutrition conducted in accordance with each growing phase requirements.At 35 days of age, the animals were slaughtered by cervical dislocation; and the organs and tissues were collected and stored in sterile tubes containing sterile RNAlater solution (Ambion, RNA Carlsbad, CA, USA).The samples were stored at 4 C for 12 h and then at À20 C until total RNA extraction.

RNA extraction and cDNA synthesis
Total RNA was extracted using the Trizol reagent (Invitrogen, Carlsbad, CA, USA) according to the manufacturer's instructions, from 30 mg of each tissue.RNA concentration was determined by AstraGene UV/Vis spectrophotometer (AstraNet Inc., Bath, UK) using 2.5 mL of the extracted RNA.To eliminate possible sample contamination, they were subjected to treatment with DNase I Amplification Grade (Invitrogen, Carlsbad, CA, USA), then RNA integrity was evaluated using a 1% agarose gel stained with the gel red TM , at 100 V for 30 min.After this verification, the cDNA was built from 1 mg of total RNA from each tissue using the GoScript Reverse Transcription System kit (Promega, Madison, WI, USA), following the manufacturer's instructions.The resulting cDNA samples were stored at À20 C until PCR analysis was performed.

Selection of reference genes and RT-qPCR
Ten reference genes were chosen to assess stability during gene expression study in male and female broiler quails.These genes were selected based on literature, 20 for Gallus gallus species.The primers were designed using Primer Quest software provided by DNA Technologies, Inc (IDT, Coralville, IA, USA), and hereafter, the sequences were analyzed in Primer BLAST (http://blast.ncbi.nlm.nih.gov) to verify specificity; then were synthesized by Invitrogen Life Technologies.Further information and primer sequences used in this study are shown in Table 1.
All molecular reactions were performed in a CFX96 Real-Time PCR Detection System thermocycler, using SYBR green detection method with GoTaq qPCR Master Mix (Promega, Madison, WI, USA).
To asses each primer's efficiency, a composite sample (pool) was obtained containing a mixed sample from the different RNA profiles studied.This pool was then diluted to build standard curves for PCR optimization and efficiency calculations.Thus, four cDNA concentrations (5, 15, 45, and 135 ng) and four primer concentrations (200, 400, 800, and 1000 nM) were tested.The experimental conditions of PCR were as follows: a cycle of 95 C for 10 min, 40 cycles of 95 C for 10 and 60 s at 60-62 C.An additional step with a gradual increase in temperature was added, with a temperature from 60-62 to 95 C, to obtain the dissociation curve.
The amplification reaction was developed in different wells and in duplicate, and each primer pair had its specificity checked through the amplicon size, in electrophoresis at 1.5% agarose gel.The amplification efficiency was estimated for each reference gene according to PFAFFL: 21 After efficiency analysis, the most suitable annealing temperature, cDNA, and primer concentration were used to perform PCR reactions.The reactions occurred with volume of 10 mL, and Cq values were excluded from further analysis for control wells, once they have been exported to the electronic sheet, as well as values higher than 35 or undetected.

Expression stability of reference genes
The expression stability of the reference genes was analyzed using four methods: Bestkeeper, 22 GeNorm, 23 NormFinder, 24 and DCt method, 25 through RefFinder software, available at http://fulxie.0fees.us/?type=reference, which integrates the four algorithms to compare and classify genes stability.For GeNorm and NormFinder, the values used for calculation are those related to relative gene expression estimated from Cq values (min Cq-sample Cq).For Bestkeeper and DCq method, the values used for stability calculation were the Cq values introduced in the RefFinder.This tool analyzes each algorithm used separately and, also, provides a general classification of the best normalizer for experimental conditions tested, obtained by combining the results of the four algorithms.The results provided by RefFinder, considering gene expression stability for males and females broiler quails, can be displayed together or separately.
The BestKeeper algorithm provides values of standard deviation and coefficient of variation for the tested genes, where genes with a value of standard deviation higher than 1 are considered unstable and should be excluded from the analysis. 22Hence, a gene with the lowest standard deviation value is the first for stability classification.In GeNorm, classification is based on the value of average expression stability (M), which can reach the limit of 1.5 to be considered stable. 23hus, the lower M value indicates a higher gene expression stability.NormFinder, on the other hand, shows results based on the variance between samples, where the lower value indicates less variation and, consequently, greater stability. 24The DCq method can be considered the most rigid testing each gene stability by comparing its expression with all other genes, pair by pair.After obtaining a standard deviation for each comparison, this method obtains the average deviation for each gene.So, if DCq value between the pairs of genes remains constant for all tested samples, the smaller the deviation will be and, consequently, the greater the stability of its expression. 25o obtain the final classification, RefFinder findings are based on rankings results of the more or less stable genes obtained by each method, designating an appropriate weight for an individual gene, and calculating the geometric mean of the weights.Consequently, average values of stability provided by each software can be obtained, favoring to analyze and perceive genes stability for different studied tissues.

Amplification efficiency and reaction specificity
Considering regression coefficients (R 2 ) of the standard curve for the 10 reference genes, all were >0.98 and the amplification efficiencies were between 94 and 106%, revealing good linear correlation and relatively good specificity of the primers (Table 2).
Of eight genes evaluated on thigh muscle, three genes (GAPDH, TFRC, and EEF1) were highly expressed, with the lowest Cq values.The UBC gene was the least expressed (Cq !28) among all others.LDHA, B2M, and UBC genes presented variation coefficients up to 2.5%, evidenced by low dispersion in broiler quails' brain.Thus, considering standard deviation, only RPL5 (SD ¼ 1.28) and HMBS (SD ¼ 1.01) genes were shown to be unstable, and in this case can be observed that almost all evaluated genes showed moderate expression, with Cq values from 21 to 28.However, can be observed high expression variability in tissues.In this case of the brain, variability was quite heterogeneous where the gene with the least dispersion was the RPL5 (CV ¼ 5.69%) with a high standard deviation (SD ¼ 1.18); and in heart the EEF1 gene (CV ¼ 18.91%) also having high standard deviation (SD ¼ 3.87) (Table 3).The possibility of technical error for the EEF1 gene can be ruled out since the Cq values of the replicates were homogeneous within male/female tissues (Appendix).
In the spleen, seven genes were evaluated, and all of them showed moderate expression, except for GAPDH, which obtained Cq ¼ 18.03, considered highly expressive.However, despite the expression, this gene was quite dispersed and unstable (CV ¼ 9.06% and SD ¼ 1.64) in the tissue (Table 3).Also, like the EEF1 gene evaluated in heart tissue, Cq values of replicates for the GAPDH gene were homogeneous within male/female, discarding any technical errors.

Expression stability by RefFinder
The general stability classification, assessed by tissue, obtained in the present research is shown in Table 4.According to this, the most stable genes in thigh muscle were MRPS30 and EEF1, followed by the HMBS gene; in brain, B2M, UBC, and GAPDH; in heart, MRPS30, TFRC, and HMBS, in that order.For spleen, the most stable genes identified were EEF1, LDHA, and HMBS.

Overall classification and sex effect
Sex effect was estimated through the F test (p < 0.05), from the variability results generated in RefFinder for each reference gene on different tissues, and in male and female quails separately.Figures 1-4 show the overall classification of each gene expression variability, found in four tissues (thigh, brain, heart, and spleen) from male and female broiler quails, considering sex effect.Based on this classification, MRPS30, EEF1, and HMBS genes showed the most stable expression in thigh muscle; B2M, UBC, and GAPDH on brain; MRPS30 and TFRC followed by the HMBS on heart; and EEF1, LDHA, and HMBS on spleen of broiler quails.

Discussion
Can be perceived that the HMBS gene was one of the most stable in 75% of the evaluated tissues (thigh, heart, and spleen).Some studies evaluating HMBS gene stability, found this gene to be one of the most stable when considering several tissues in the same molecular analysis.Zhang, 26 for example, evaluating the stability of eight reference genes in ten Boer goat tissues, found that the HMBS gene was the third most stable and, therefore, recommended to calibrate gene expression analyzes in goat tissues, deriving from this one, through RT-qPCR.In muscle tissue, HMBS gene was still found to be the most stable by Nascimento, 20 in chickens' Pectoralis major muscle.This is illustrated by the fact that the HMBS gene encodes hydroxymethylbilane synthase enzyme production.This enzyme is involved, also, in the production of a molecule called heme, considered vital for all organs of the body in both males and females.The use of the HMBS gene as an endogenous control for RT-qPCR analysis in various tissues of avian species has been reported and has shown positive results, due to its high expression and stability. 27RPS30 gene is a mitochondrial ribosomal protein encoded by nuclear genes and, with high activity in muscle tissues; hence, this protein synthesis is aided by this gene within the mitochondria.In addition, ribosomes where this gene is part of, consist of 75% of proteins for the composition of rRNA and, therefore, is abundant in cells and their expression can be considered continuous.For this reason, it is believed that such genes were shown to be stable in this study, as they are continually required in cells as part of processes essential to their maintenance.EEF1 gene encodes a protein responsible for the alpha-1 elongation factor, responsible for the enzymatic delivery of aminoacyl tRNAs to the ribosome during protein synthesis.This is also an FES-1 alpha subunit complex isoform, GTPase, and is involved with protein actin, in proteolysis. 28,29B2M gene, on the other hand, encodes a whey protein found in association with the largest heavy chain class I histocompatibility complex (MHC) on the cell surface of all nucleated cells.In brain tissue, ubiquitin (UBC) also acts, which is necessary for the proper creation, maintenance, and disassembly of protein subdomains within the neuron, being its function predominant throughout the life of this cell. 30Thus, all genes indicated as more stable in the current four tissues studied executing, on large scale, on cell survival and metabolism, with relatively little participation in only one of the sexes' exclusive routes; for this reason, they  can be recommended for genic expression normalization in both male and female quails.
In the present study, the genes MRPS27 and LDHA in thigh muscle; RPL5 in brain, EEF1 in heart, and MRPS30 in spleen were considered unstable, and therefore, not recommended as a reference, when intended to analyze broiler quails gene expression in both sexes.However, due to their high coefficients of variation and deviation, the sex effect could be seen, enabling them to be recommended as reference genes when only one sex is being analyzed.In thigh muscle, MRPS27 protein, as the MRPS30, is encoded by nuclear genes with high activity in muscle tissues.However, besides its high expression, be stable and indispensable to mitochondrial cells; a difference in its production levels can be noted, due to differential protein metabolism between males and females, especially, in thigh muscle.This may explain to recommend this gene as an internal control for RT-qPCR analysis when intended to study muscle tissues of both male and female quails, which were analyzed separately.
For brain tissue, the present research indicates the GAPDH gene as a candidate for reference gene to standardize the expression of only from males, from females, or from males and females samples together.This gene has become one of the most used as a reference to standardize the expression of several samples in various animal species, 31 due to the fact that Glyceraldehyde 3-phosphate dehydrogenase (GAPDH) is directly involved in metabolic and non-metabolic processes in the cell.This enzyme catalyzes the sixth step of glycolysis, such as glucose breakdown, activation of transcription, initiation of apoptosis, transport of ER vesicles to Golgi, and axonal transport rapid or axoplasmic; 32 however, its expression varies greatly between individuals and in different stages of development. 33In avian species brain tissue, for example, GAPDH was used as endogenous control along with other genes, their expression in this tissue was stable and effective. 34In fact, in molecular studies seeking to evaluate the genic expression of sexually dimorphic genes related to gonadal differentiation in chickens' brain tissue, the GAPDH gene is the one that showed the best results as a reference gene, 35 according to the results obtained in the present study.
In the present research, the TFRC gene was the most stable and expressed in male quails heart.This gene encodes transferrin receptor Protein 1 responsible for cellular iron absorption.In healthy animals, male or female, both iron uptake and TFRC expression in heart tissue are stable, since this gene participates in cells' oxidative phosphorylation.The deletion of this gene in cardiomyocytes is the origin of coronary problems and could motivate lethal cardiomyopathy. 36n male and female quail spleen tissue, the EEF1 gene was the most stable when compared to the others.Genes expressed invariably in this tissue are related to immune response, independent of animal sex.In this case, when evaluating its genic expression in both sexes with normal health conditions, as accomplished in the current work, it is possible to recommend the use of the EEF1 gene as endogenous control of spleen analyzes involving, for example, T cells.The present study was capable to identify stable genes than can be used in further genic expression studies, being necessary to evaluate both sexes and different quail tissues.The differences found between tissues showed variability, as an important factor to be analyzed, and must be considered in any experimental process.For some genes, the sex effect can be noticed suggesting, therefore, that every analysis can be performed to clarify differential expression found for contrasting tissues and sexes.Studies evaluating genetic behavior in animals without distinction of sex are important and should be subject to constant improvements.Due to research scarcity related to molecular genetics in quails, the present study serves as an informational basis to continue research on gene expression in this species.In addition, also can be a basis for investigations of the same family animals, such as chickens and partridges for example, and consequently, promotes improvements in the entire production chain.

Conclusion
Based on algorithms BestKeeper, GeNorm, NormFinder, and DCq; the genes MRPS30, EEF1, and HMBS in thigh muscle; B2M, UBC, and GAPDH in brain; MRPS30, TFRC, and HMBS in heart; and, EEF1, LDHA, and HMBS in spleen tissues; are the most stable and, therefore, can be recommended to be used as an endogenous control in further gene expression studies related to male and female broiler quails.

Figure 1 .
Figure 1.Expression variability of reference genes evaluated in thigh muscle from broiler quails, within and between male and female individuals.a,b Different lowercase letters were significantly different (p < 0.05).

Figure 2 .
Figure 2. Expression variability of reference genes evaluated in brain from broiler quails, within and between male and female individuals.a,b Different lowercase letters were significantly different (p < 0.05).

Figure 3 .
Figure 3. Expression variability of reference genes evaluated in heart from broiler quails, within and between male and female individuals.a,b Different lowercase letters were significantly different (p < 0.05).

Figure 4 .
Figure 4. Expression variability of reference genes evaluated in spleen from broiler quails, within and between male and female individuals.a,b Different lowercase letters were significantly different (p < 0.05).

Table 1 .
Reference genes, specific primers, and parameters derived from quantitative real-time PCR analysis.

Table 2 .
Characteristics of PCR amplification reactions for current studied genes.

Table 3 .
Descriptive statistics and expression levels of reference genes in broiler quails at 35 days tested in four tissues (thigh, brain, heart, and spleen), obtained from the bestkeeper (n ¼ 6) software.

Table 4 .
Stability values, algorithmic, and general classification for reference genes in four tissues, evaluated in male and female broiler quails.