INTRODUCTION
Quantitative analysis of gene expression is essential in the field of biology and veterinary research to understand the gene regulatory network. Because quantitative reverse transcription polymerase chain reaction (qPCR) is able to simultaneously compare gene expression in various samples and provides high convenience, sensitivity, reproducibility, accuracy, and reliability, it is considered as the standard method for quantification of gene transcripts. However, the results of qPCR can be critically affected by several factors, including the quality of nucleic acids, amount of starting material, method of RNA preparation, tissue degradation, sampling method, specificity of PCR products, and DNA dye [
1,
2]. The most common technique used to adjust for these variations is normalization of the expression level of a target gene against constitutively expressed reference genes (RGs) [
3,
4].
The RGs play a critical role in basic cellular functions and cell survival, including energy generation, substance synthesis, and cell defense/death, and are believed to be stably expressed regardless of environmental and experimental conditions [
5]. Unfortunately, there is no single universal RG that is consistently expressed in all experimental situations; accumulated evidence has shown that expression of RGs is variable according to experimental conditions [
4,
5]. For instance, expression of commonly used RGs such as beta actin (
ACTB), glyceraldehyde-3-phosphate dehydrogenase (
GAPDH), 18S ribosomal RNA (
18S), and hypoxanthine phosphor-ribosyl transferase (
HPRT) is dependent on tissue type, developmental stage, and other factors [
6–
8]. In fact, the application of inappropriate RGs for normalization of the target gene could result in large fluctuations in expression among the tested samples, false conclusions, and misleading interpretations of gene expression [
2,
9]. Therefore, validation of the proper RGs for each experimental condition is considered as a prerequisite for reliable results by qPCR and can improve the accuracy and reproducibility of the study [
5].
Because pigs (
Sus scrofa) are one of the most economically important types of livestock, understanding how reproduction is regulated inherently or affected by external factors can offer important insights to aid in the development of husbandry strategies in the animal farm industry and understand the reproductive physiology of animals [
10]. Normal reproductive function is dependent on the hypothalamic–pituitary–gonadal (HPG) axis; the HPG axis for reproduction is centrally controlled by a complex regulatory network of excitatory and inhibitory signals [
11]. In detail, the hypothalamus secretes pulsatile gonadotropin-releasing hormone (GnRH) via the hypophyseal portal system to induce gonadotropin pulsatile secretion, including follicle-stimulating hormone and luteinizing hormone (LH), and the gonadotropins then stimulate the growth of ovarian follicles to preovulatory follicles, where estrogen is highly secreted. The estrogen activates the
GnRH surge, which is followed by the LH surge for ovulation and corpus luteum (CL) formation. This inherent feedback system of the HPG axis can be altered by several stimuli, including disease, weight, nutrients, age, season, and stress, that can lead to loss of the normal preovulatory LH surge, estrus cyclicity, and fertility [
11,
12]. Therefore, gene expression study in HPG axis-related female tissues in response to internal or external stimuli has been widely conducted using qPCR [
13,
14].
Several studies have evaluated various RGs to clarify the most stable gene under each experimental condition. Similar efforts have also been conducted in pig to find the most stable RGs in various cell types, embryos, several tissues, and infection [
2,
4,
5,
7,
15]. However, a clear list of suitable RGs in the HPG axis-related tissues in sows is still lacking. In addition, since the expression of genes in the reproductive system fluctuates depending on the stage of estrus cycle or pregnancy in animals, it is important to find the most stable RGs regardless of different stages of estrus cycle for the future experiment; application of other sets of RGs at each different stage of estrus cycle in a study is inconclusive and impractical [
16]. Therefore, the aim of the present study was to evaluate stability within the pool of nine commonly used RGs in the HPG axis-related tissues of sows regardless of the stages of estrus cycle as transcript levels of RGs may vary among different types of tissues or between different estrus cycles. The cycle threshold (Ct) values determined by qPCR, which were obtained from candidate RGs in HPG axis-related tissues, were assessed for their stability by means of the geNorm and Normfinder programs. The RGs evaluated in the study will be helpful for investigating the molecular mechanisms involved in the HPG axis in female pigs.
DISCUSSION
An understanding of the HPG axis in the domestic animals is important in order to improve the output of livestock products as well as to understand animal reproductive disorders such as anestrus, silent heat, and ovarian cysts. In addition, its application can improve the efficiency of reproductive performance, including normal ovarian cyclicity, increased wean-to-estrus intervals, and higher pregnancy ratio and litter size [
18]. Because gene expression studies using qPCR in HPG axis-related female tissues have been widely conducted to understand reproductive physiology, selection of adequate and stable RGs is regarded as a prerequisite step for deducing the reliable results for an exact comparison of mRNA transcription in different samples or tissues [
8]. Unfortunately, there is no single universal RG that is constantly expressed in all types of tissues and is not regulated by internal and external stimuli [
4]. And the expression of RGs in the reproductive tissues can be changed by inherent body conditions such as the stage of the estrus cycle or pregnancy in female pigs [
16]. In addition, several previous articles and
Figure 5 in the present study demonstrate that the application of stable or unstable RGs can change the outcome and conclusions of a study; for instance, whereas normalization with stable RGs could generate significant difference in target gene expression between groups, unstable ones showed no significance and possibly led to false conclusions [
2,
9]. Therefore, we mainly focused on uncovering the most stable RGs in the HPG axis-related tissues of sows regardless of the different stages of estrus cycle from the pool of nine commonly used RGs by means of stable RG-finding programs (geNorm and Normfinder). Since there has been no universal standard consensus method for the validation of stability of RGs, we mainly used geNorm for gene stability analysis, followed by reconfirmation by Normfinder to avoid tool-dependent results; both programs have been frequently used for finding stable RGs, and these comparisons by different programs for RG selection may allow a better evaluation of the most reliable controls [
2,
3,
5,
15,
16,
19,
20]. We found that results of the three most stable RGs were highly consistent between the two programs; a slight difference in stability rankings between programs could be explained by their different algorithms [
2,
19]. Comprehensively, both programs concluded that
PPIA,
HPRT1, and
YWHAZ were the most stable RGs in the HPG axis in sows regardless of the stages of estrus cycle and that traditional RGs, including
18S and
ACTB, were less stable (
Figures 3,
4). To the authors’ knowledge, the present study is the first to report on stable RGs in the HPG axis of pigs.
The present results explain the stability of RGs in HPG axis in sows; mainly,
PPIA,
HPRT1,
YWHAZ,
18S, and
ACTB. Even though they are widely used as RGs due to their consistent roles in cell survival, their expression is affected by several stimuli.
PPIA is a prototypical cyclophilin family member, an enzyme that catalyzes the reversible cis/trans interconversion of the imide bond in proline residues and is known as cyclosporin binding protein and inhibitor of serinethreonine phosphatase. Therefore, changes in expression of
PPIA are highly related to inflammatory disorders and cancers [
21].
HPRT1 is a transferase enzyme that plays a pivotal role in the cell cycle by generating purine nucleotides via the purine salvage pathway, and its elevation is now considered as a marker with clinical significance in human disease, including several tumors, because demand for
HPRT1 in cell cycling is increased for nucleotide synthesis [
22].
YWHAZ acts on cell growth, apoptosis, migration, and invasion. Furthermore, upregulation of
YWHAZ is also highly related to tumor progression [
23]. Because
ACTB encodes a structural protein of cytoskeleton as an indispensable component of the cytoskeleton in the cell for cell migration, cell division, and regulation of gene expression, change of
ACTB expression is associated with cancer and changes in response to external stimuli [
9].
18S is a component of the ribosomal RNA and plays a role in the biogenesis and function of ribosomes in the cell; its expression is variable in cultured goat follicles and ovarian tissue derived from healthy or diseased humans [
9]. Therefore, the fact that RGs could be affected by several influences makes the validation of RGs under each experimental condition important and suggests that random selection of commonly used RGs is no longer acceptable. In particular, several published articles have demonstrated differences in the transcript level (Ct value) of RGs. Ten of twelve RGs exhibited different Ct values in porcine mesenchymal stem cells (MSCs) before and after differentiation [
5]. In addition, the Ct values of seven of nine RGs differed according to the type of porcine blastocysts [
2]. Similarly, significant differences in Ct values among experimental groups have been routinely found [
4,
7,
19–
21]. In agreement with these articles, the present study found that the Ct values of seven of nine candidate RGs showed significant differences (p<0.05) in the HPG axis-related tissues of sows, indicating differences in tissue transcript levels even in RGs and suggesting the necessity of further validation of RGs for qPCR assay (
Figure 2).
In porcine specimens, similar efforts to discover the most stable RGs in each experimental condition have been conducted. Expression of
YWHAZ is the most stable RG in porcine MSCs regardless of differentiation induction [
5], intact alveolar macrophages (AMs) [
4], peripheral blood mononuclear cells (PBMCs) with lipopolysaccharide (LPS) and lipoteichoic acid (LTA) stimulation [
19], tissue (PBMCs, lymph nodes, intestinal mucosa, stomach, liver, spleen, thymus, lung, kidney, heart, and skin) at different ages including newborn, young, and adult pigs [
8], and adipose with muscle tissue [
24].
HPRT1 was also determined to be one of the best RGs in porcine samples for the pregnant ovary across physiological time points (heat and 15, 30, 45, and 60 days of pregnancy) [
16], different parts of the gastrointestinal (GI) tract of piglets during the weaning process [
22], and
Actinobacillus pleuropneumoniae-infected tissues including white blood cells, liver, and lymph nodes [
15].
PPIA is stably expressed in intact PBMCs, polyinosinic: polycytidylic acid-stimulated PBMCs [
20], and several tissues at different ages [
8]. In addition, in most of the aforementioned cases, traditional RGs such as
GAPDH,
ACTB, and/or
18S were determined to be the least stable in porcine samples in porcine MSCs [
5], pregnant ovary [
16], intact AMs [
4], LPS and LTA-stimulated PBMCs [
19], the GI tract of piglets [
22], heat-stressed blood [
25],
Actinobacillus pleuropneumoniae-infected tissues [
15], and several tissues at different ages [
8]. Furthermore, expression of these traditional RGs is unstable in several types of porcine blastocysts produced by
in vivo, parthenogenetic activation,
in vitro fertilization, and somatic cell nuclear transfer [
2] and in various pig tissues including the diaphragm, heart, kidney, liver, lungs, muscle, spleen, and stomach [
3,
6,
15].
Although there is currently no published data on stable RGs in HPG axis-related tissues of sows, similar studies have been conducted in other species. In agreement with the findings of the present study,
PPIA,
HPRT1, and/or
YWHAZ have been validated as the most stable RGs in HPG-axis tissues including the brain, pituitary, ovary, and testis in songbirds [
26], testosterone-influenced hypothalamus and kidney of rats [
27], normal and adenoma tissues of pituitary gland from dogs and mice [
28], and human polycystic ovarian syndrome [
29]. The stability of traditional RGs such as
GAPDH,
ACTB, and/or
18S was moderate to low in testosterone-influenced hypothalamus and kidney of rats [
27], hypothalamus of chicken under different feeding status [
30], HPG axis in songbird [
26], pituitary gland from normal/adenoma dogs and mice [
28], and human polycystic ovarian syndrome [
29].