Expression characteristics and regulatory mechanism of Apela gene in liver of chicken (Gallus gallus)

Apela, a novel endogenous peptide ligand for the G-protein-coupled apelin receptor, was first discovered and identified in human embryonic stem cells in 2013. Apela has showed some biological functions in promoting angiogenesis and inducing vasodilatation of mammals by binding apelin receptor, but little is known about its expression characteristics and regulatory mechanism in chicken. In the present study, the coding sequences of Apela in chicken was cloned. The evolution history and potential function of Apela were analyzed. Subsequently, the spatiotemporal expression characteristics of chicken Apela were investigated. Furthermore, the regulatory mechanism of Apela mRNA responsing to estrogen was explored by in vitro and in vivo experiments. The results showed that the length of the CDs of Apela mRNA was 165 bp and encoded a protein consisting of 54 amino acids residues with a transmembrane domain in chicken. The Apela was derived from the same ancestor of Apelin, and abundantly expressed in liver, kidney and pancreas tissues. The expression levels of Apela in the liver of hens were significantly higher at the peak-laying stage than that at the pre-laying stage (p ≤ 0.05). The Apela mRNA levels were significantly up-regulated in primary hepatocytes treated with 17β-estradiol (p ≤ 0.05), and could be effectively inhibited by estrogen receptor antagonists MPP, ICI 182780 and tamoxifen. It indicated that chicken Apela expression was regulated by estrogen via estrogen receptor α (ERα). In individual levels, both the contents of TG, TC and VLDL-c in serum, and the expression of ApoVLDLII and Apela in liver markedly up-regulated by 17β-estradiol induction at 1mg/kg and 2mg/kg concentrations (p ≤ 0.05). This study lays a foundation for further research on Apela involving in hepatic lipid metabolism.


Introduction
The endogenous peptide apelin is identified as a novel adipokine to regulate fat metabolism in mammals, exerting its biological function by binding the G-protein-coupled apelin receptor regulatory mechanism remains unclear. In this study, evolutionary history and dynamic expression patterns of chicken Apela were investigated by phylogenetic tree and qRT-PCR analyses, respectively. Furthermore, the expression regulation of Apela was investigated by in vivo and vitro experiments. This study will lay a foundation for further research on Apela involving in hepatic lipid metabolism.

Animals and sample preparation
The birds used in the study are HY-LINE hens. They were raised under the same condition with normal feeding procedure. The liver tissues of hens at the age of 1 day, 1, 10, 15, 20, 30 and 35 weeks (6 birds for each point) were collected, respectively. Meanwhile, ten other tissues including ovary, lung, heart, spleen, kidney, duodenum, glandular stomach, pancreas and pectorales of hens were also collected at the age of 10 and 30 weeks, and stored at -80˚C until use. Forty-eight 10-week-old HY-LINE hens with similar body weights were randomly divided into four groups (12 birds for each group), including three estrogen-treated groups at different concentrations (0.5, 1.0, and 2.0 mg/kg body weight of 17β-estradiol) and one control group (olive oil only). 17β-estradiol (Sigma-Aldrich, St. Louis, MO, USA) was dissolved in olive oil. The mode of administration was intramuscular injections. After injection for 12 h and 24h, the blood samples of each group was collected from wing veins (6 birds at each observing time) and was left at room temperature for about 1.5 hours to separate the serum. Then, the birds were euthanized, and the liver tissues were isolated. The serum and tissue samples were stored as described above.
The animal use and care protocols for these experiments were approved by the Institutional Animal Care and Use Committee (IACUC) of Henan Agricultural University Zhengzhou, P.R. China (Permit Number: 11-0085).

Blood plasma biochemical index detection
The blood plasma biochemical index of associated with lipid synthesis and metabolism mainly include triglyceride (TG), low-density lipoprotein cholesterol (LDL-c), high-density lipoprotein cholesterol (HDL-c), total cholesterol (TC) and very low-density lipoprotein cholesterol (VLDL-c) [25,28]. Among them, the cotent of first four index in serum samples were detected using a blood analyzer (Hitachi 7100), while the cotent of VLDL-c was indirectly calculated by the Friedewald formula: VLDL-c = TC-HDL-c-LDL-c [26,27,29,30].

Isolation and culture of chicken primary hepatocytes
Fertilized eggs without specific pathogen for the experiment were purchased from Beijing Meiliyaweitong Experimental Animal Technology Co. Ltd. (Beijing, China). Fertilized eggs were incubated to 18 embryo age in thermostatic wet incubator. The embryonic hepatocytes were isolated according to the method as previously described [28,31]. The seed plate and culture method of hepatocytes were following by our previous studies [29,32].

Stimulation of chicken primary hepatocytes
To further confirm estrogen regulating Apela expression in vitro, 17β-estradiol, dissolved in ethyl alcohol, were used to stimulate primary hepatocytes. The hepatocytes were starved for 6 h when the confluence reached 80%, then treated with different concentrations of 17β-estradiol (0, 25, 50, 100 nM) for 12 h (six repeats for each treatment). The control groups (0 nM 17β-estradiol) were treated with solvent ethyl alcohol only.
To determine which estrogen receptor subtypes response to estrogen regulating Apela expression, 17β-estradiol and estrogen receptor antagonists (ICI 182780, tamoxifen and MPP) were used for stimulating primary hepatocytes. ICI 182780 and tamoxifen are antagonists of ERα and ERβ, tamoxifen also induces GPR30, and MPP (methyl-piperidino-pyrazole) is a highly selective antagonist of ERα. Before treatment, estrogen receptor antagonists were dissolved in DMSO. Similarly, after 6 h of starvation, the primary hepatocytes were co-treated with 17β-estradiol and different estrogen receptor antagonists. In details, the experimental treatment was divided into 5 groups with 6 biological replications of each treatment. The treatment group of estrogen receptor antagonists (group 1, 2, 3) were firstly treated with a final concentration of 1 μM MPP, 1 μM tamoxifen, 1 μM ICI 182,780 for 4 h, respectively, and followed by a treatment of 100 nM 17β-estradiol for 12 h. Estrogen treatment group (group 4) was firstly treated with 1 μM solvent DMSO for 4 h and then a same dose of 17β-estradiol for 12 h. The control group (group 5) was treated with the two corresponding solvents mentioned above. The cells were collected and stored at -80˚C for further use.

RNA extraction and cDNA synthesis
RNAiso Plus reagent (Takara, Dalian, China) were used to extract the total RNA of tissues and cells. The RNA integrity and quality were tested through gel electrophoresis. The total RNA concentrations were detected by a Nanodrop 2000 spectrophotometer (Thermo Fisher Scientific, Waltham, MA, USA). cDNA was synthesized using a PrimeScript TM RT Reagent kit with gDNA Eraser (Takara) according to the manufacturer's protocols, and stored at -20˚C.

Cloning of chicken Apela gene
To clone the coding sequence (CDs) of Apela, the PCR primers (Table 1) were designed with Primer 5.0 software, according to the predicted Gallus gallus Apela sequence (GenBank: XM_015285307). PCR reaction volume was 20 μL, including 10 μL of 2× pfu MasterMix (Tian-genBiotech Co. Ltd., Beijing, China), 0.8 μL forward / reverse primers (10 μM), 6.4 μL of deionized water and 2 μL of first-strand cDNA. The PCR procedure was set as follows: 95˚C for 4 min; 30 cycles at 95˚C for 30 s, 60˚C for 30 s, and 72˚C for 30 s; followed by 72˚C for 10 min. The PCR products with correct target fragments were sequenced Sangon Biotech, Co., Ltd. (Shanghai, China) with three times independently.

Phylogenetic analysis
The amino acid sequence alignments of Apela and Apelin were performed using Clustal W. And the phylogenetic tree was constructed by using the neighbor-joining method in MEGA7.0 [30,33]. The structure prediction for amino acid sequence of chicken apela was implemented using the SMART software (http://smart.embl-heidelberg.de/).

Statistical analysis
The mRNA level of Apela and ApoVLDLII relative to β-actin was calculated by the 2 −ΔΔCT method. The variance analysis among groups were carried out using One-way ANOVA on SPSS version 20.0. p � 0.05 was considered statistically significant. All values are presented as mean ± SEM.

Sequence analysis of chicken Apela gene
The cDNA synthesized from the RNA of liver tissue was used as the template to clone the Apela gene ( Fig 1A). The sequencing results of PCR product showed that the length of the CDs of Apela mRNA was 165 bp, which encoded a protein consisting of 54 amino acide residues ( Fig 1B), identical to the nucleotide sequence and amino acid sequence of chicken Apela (KX017223 and APU52335.1, respectively) reported before. The functional domain analysis showed that Apela was composed of two domains including the low-complexity region (LCR) domain and the transmembrane region domain ( Fig 1C). The transmembrane region domain was located at 7-26 aa. The LCR was located at 2-18 aa. The amino acid sequences alignment of Apela among chickens, humans and rats indicated that the sequences of Apela shared 59% and 57% identify to human and rat, respectively ( Fig 1D).

Phylogenetic analysis of Apela and Apelin
Phylogenetic analysis was carried out based on the Apela and Apelin amino acid sequences alignment among different species, including mammals (Homo sapiens, Pan troglodytes, Microcebus murinus, Bos taurus, Capra hircus, and Mus musculus), avians (Gallus gallus, Coturnix japonica, Parus major, and Taeniopygia guttata), reptiles (Gekko japonicus), amphibians (Xenopus tropicalis and Gavialis gangeticus) and fish (Danio rerio). The GenBank ID of Apela and Apelin in these species were listed in Table 2. The phylogenetic tree revealed that both Apelin and Apela genes from different species were classified into the same cluster except Apelin from Pan troglodytes was classified into Apela cluster, and Apelin and Apela are paralogous genes which may originate from the same ancestor (Fig 2).

Spatiotemporal expression of chicken Apela gene
To understand the expression distribution of Apela, Apelin and Apelin receptor in chicken, the relative mRNA expression levels of the gene in different tissues of 10-week-old and 30-weekold chickens were analyzed by qRT-PCR. The results showed that Apela gene was extensively expressed in all tissues tested, and the expression levels were relatively higher in liver, pancrea and kidney compared to the other tissues at the two stages (Fig 3). Interestingly, Apela was sharply up-regulated only in liver of 30-week-old hens, compared with 10-week-old hens.  Furthermore, the relative expression of Apela gene in hen livers at different development phases were investigated. It was revealed that, compared to the sexually immaturity (1d, and 1w-20w), the Apela mRNA was significantly up-regulated at peak-laying period (30w and 35w) (p � 0.05), while there is no significant change among 1d, 1w, 10w,15w and 20w (p > 0.05), and between 30w and 35w (p > 0.05) (Fig 4). However, the expression of Apelin and Apelin receptor were significantly decreased at the peak-laying stage in comparison with that in pre-laying stage (S1 Fig).  Table 2. https://doi.org/10.1371/journal.pone.0238784.g002

Expression regulation of Apela gene by estrogen in vitro
Consider that there are large increase in lipid synthesis of liver and estrogen levels in laying period, the experiment explore the effect of 17β-estradiol on hepatic Apela expression in vitro.
the Apela mRNA levels were investigated in chicken primary hepatocytes treated with different concentrations of 17β-estradiol for 12 h. The results showed that ApoVLDLII, a positive control gene of estrogen response, was significantly up-regulated expression in a dose-dependent manner (p � 0.05) (Fig 5A), indicating that exogenous 17β-estradiol exerted the biological function of estrogen in the cell level. Apela in chicken primary hepatocytes also exhibited a significantly up-regulated expression when 17β-estradiol concentrations achieved at 50 nM and 100 nM (p � 0.05) (Fig 5B). Combining the expression changes of ApoVLDLII and Apela, 100nM 17β-estradiol concentrations was the optimal concentration that induced Apela expression. In contrary, the expression of Apelin and Apelin receptor exhibited down-regulated trend after 17β-estradiol treated primary hepatocytes (S2 Fig).
To further identify which estrogen receptor subtype responded to estrogen regulation of Apela expression, firstly, we checked the expression levels of ER α and ER β in the liver of prelaying hens (20 weeks old) and peak-laying hens (30 weeks old). The results suggested that the expression levels of ER α and ER β increased with sexual maturation (S3A and S3B Fig). Then, we treated the primary embryonic hepatocytes of chicken with 17β-estradiol or co-treated with 17β-estradiol and different ER antagonists (Fig 6A and 6B). The results showed that compared to the control group, the mRNA levels of ApoVLDLII (the positive control gene) increased significantly in 100 nM 17β-estradiol treatment group (p � 0.05), and then decreased significantly to the level of control group in the co-treatment group of 17β-estradiol and ER antagonist MPP, tamoxifen or ICI 182780 (p � 0.05). Similarly, the expression of Apela was significantly up-regulated accompanied by 17β-estradiol treatment, whereas recovered significantly to the level of control group in the co-treatment group of 17β-estradiol and ER antagonist MPP, tamoxifen or ICI 182780 (p � 0.05). For the estrogen receptor subtype antagonists, ICI 182780 and tamoxifen are antagonists of ERα and ERβ, tamoxifen also induces GPR30, and MPP is a highly selective antagonist of ERα. Therefore, the results implied that the expression of Apela in chicken liver was regulated by estrogen through ERα.

The effect of estrogen on Apela expression in individuals
To further validate the effect of estrogen on chicken Apela expression in individuals, 10-weekold hens were treated with different doses of 17β-estradiol for 12 h and 24 h. The serum biochemical index showed estrogen could notably induce the content increase of TG when treating for 12 h (p � 0.05), while the contents of TC and VLDL-c was no significant change (p � 0.05) (Fig 7A). When treating for 24 h, the contents of TG, TC and VLDL-c dramatically increased at 1.0 and 2.0 mg/kg of 17β-estradiol treatment concentrations (p � 0.05) (Fig 7A). Therefore, 17β-estradiol-treated liver tissues for 24 hours were selected for gene expression analysis. The gene expression results showed ApoVLDLII was significantly up-regulated expression in a dose-dependent manner accompanied by the increase of 17β-estradiol concentrations (p � 0.05) (Fig 7B). However, when 17β-estradiol treatment concentrations reached 1.0 mg/kg and 2.0 mg/kg levels, Apela were induced to be significantly up-regulated in chicken livers (p � 0.05) (Fig 7C).

Discussion
Previous reports showed that the apelin peptide deficiency in mice didn't cause impact to birth rate or embryonic development [6], while the loss of APJ in mice resulted in a drop of birth rate in the expected Mendelian ratio [7], breaks the hypothesis that apelin was the sole  endogenous ligand of APJ. The discovery and identification of Apela in human embryonic stem cells and the function of its binding APJ confirmed Apela was the second endogenous ligand of APJ [8]. Since then, it brings a blossom on Apela-APJ research. Recently, Apela was detected in the adult heart as a novel endogenous ligand of APJ [33,36] and was also abundant in human placentas to enhance placental development to prevent preeclampsia [34,37]. At present, researches on the Apela/APJ pathway and function have mainly focused on human and mouse, no systematic expression and regulation characteristics of Apela has been reported in chicken.
Here we cloned the CDs of Apela mRNA, analyzed the amino acid sequences and implemented functional domain analysis. The amino acid sequences alignment of Apela among chickens, humans and rats showed that Apela possessed a pair of conserved cysteines, and the last 13 residues were nearly invariant, which was consistent with the analysis of other report [8,27]. For Apelin, there were various of active protein subunits, such as, Apelin-55, Apelin-17, Apelin-13 and Pyr-Apelin-13 [35,38]. They all contained the 13 amino acids in the C-precursor protein were thought to be the active units of each subunit [36,39]. Similar to Apelin, The control groups (0 mg/kg) were treated with olive oil only, and the treatment groups were treated with 0.5, 1.0 and 2.0 mg/kg of 17β-estradiol. Each data point represents the mean ± SEM of 6 chicken. Different letters mean a significant difference between groups (p � 0.05), and the same letter means no significant difference between groups (p > 0.05).
https://doi.org/10.1371/journal.pone.0238784.g007 the Apela amino acids had high similarity in the 13 amino acids [37,40], so we speculated that the Apela and Apelin were likely to retain similar binding mechanism. However, in this study, the phylogenetic analysis showed that Apelin and Apela were classed into two clusters, which suggested some modest differences of biological functions between these two orthologous genes.
Expression patterns of apela are various in different species, which may associate with its biological functions. In Zebrafish, apela is expressed predominantly in the testes and weakly in other tissues including the intestine, brain and heart [41]. In human beings, expression of apela is associated with pre-implantation human development, and restricted to a few tissues, including cardiovascular tissues and two endocrine organs, the kidneys and placenta in the adult [42]. In rodents, expression of apela can be detected in the stage of zygotic transcription, peaked at the blastocyst stage, and mainly expressed in kidney of adult rat to regulate fluid homeostasis by directly binding APJ to activate the Gi pathway [43]. As for chicken, both our study and other previous study [41] showed that Apela is mainly expressed in liver and other two endocrine organs kidney and pancreas. Apela mRNA expression exhibited a down-regulation in the peak-laying stage in comparison to that in pre-laying stage in chicken. How does the expression changes of Apela in kidney and pancreas tissues correlate to the physiological changes are not clear yet, and need to be clarified by more experiments. However, Apela mRNA expression levels in chicken livers at the laying stages are significantly higher than that of pre-laying stages. It was suggested that Apela might play an important role in avian liver lipid metabolism. As we know, the most of de novo synthesis of fatty acid occurs in chicken liver [20,39,44,45]. During the laying stages in chicken, proteins and hydrophobic lipids including triacylglycerols, cholesteryl, cholestery esters, phospholipids and free fatty acids are actively synthesized in liver. Further studies focusing on the function of the chicken Apela gene in the lipid metabolism of the kidney are needed.
It has been demonstrated that estrogen is a key regulator controlling the lipid synthesis and transport in liver of laying hen [40,46]. We conducted the estrogen treatment experiments in vitro and in vivo by using 17β-estradiol. The expression of Apela mRNA was increased in a dose-dependent manner both in embryonic primary hepatocytes and liver tissue after 17βestradiol treatment. This finding indicated that 17β-estradiol could induce the expression of chicken Apela gene. It is well known that estrogen stimulates gene expression by binding to its receptors, including nuclear receptors (ERα and ERβ) [41,42,47,48] and a membrane receptor (GPR30) [43,49]. To further investigate regulatory mechanism of estrogen function on Apela expression, the expression levels of ER α and ER β in the liver of pre-and peak-laying hens were examined first, and found that the expression levels of ER α and ER β increased with sexual maturation. It should be the true because, in general, the estrogen level of hens rises significantly during the egg-laying period, and estrogen acts on many of genes involved in hepatic metabolisms, e.g. lipid and fatty acid metabolism, by interaction with either ER α or ER β, and certainly other pathways. Then the effects of 17β-estradiol on promoting Apela mRNA expression were checked by using the different combinations of 17β-estradiol and ER antagonists, and found that Apela expression could be partially inhibited by either the ER α antagonist MPP, or the ER α and ER β antagonists TAM and ICI 182,780. Because MPP is a highly selective antagonist of ERα, and ICI 182, 780 and tamoxifen are antagonist of ERα and ERβ, meanwhile tamoxifen also induces GPR30, a membrane receptor. We speculated that the expression of Apela was activated by estrogen in chicken embryo hepatocytes mediated by ERα. Of course, it requires further experimental verification.
During the laying stage, TG and cholesterol are synthesized in large quantities in hen's liver and are assembled by ApoVLDLⅡ and apolipoprotein B into VLDLy, which is a yolk-targeted VLDL, induced by estrogen [41]. The concurrent inductions of avian hepatic lipogenesis and plasma lipids by estrogen exogenous injection have been examined in immature avian [24]. Our results that the contents of TG, TC, and VLDL-c in serum of 10-week hens stimulated by 17β-estradiol increased significantly, and the liver ApoVLDLII expression of the corresponding treatment were also remarkable increase, which is consistent with previous reports. Therefore, the significantly upregulated expression of chicken liver Apela induced by 17β-estradiol suggested that Apela might involve in hepatic lipid metabolism of laying chicken.
As we usually believed, Apela plays its roles via binding APJ. However, previous study demonstrated that Apela mRNA has regulatory roles on gene expression depending on species. A previous study reported a non-coding role of Apela in promoting p53-mediated DNA damage-induced apoptosis DIA via interaction between heterogeneous nuclear ribonucleoprotein L (hnRNPL) and 3' UTR of Apela in embryonic stem cells of mouse origin (mESCs) [19]. However, in human pluripotent embryonic stem cells (hESCs), the mature ELA functions as an endogenous hormonal peptide secreted by hESCs in a paracrine manner and signals through an unknown receptor to the PI3K/AKT pathway to sustain survival and self-renewal of ESCs. We analyzed the the features of 3'UTR sequences of Apela between mouse and chicken [42]. The 3'UTRs are of very different lengths and show a low homology between these two orthologous genes, which implied the 3'UTR of Apela gene in chicken and mouse may play different roles, but need to be explored fuether.
In conclusion, the CDs of Apela mRNA was cloned. Amino acid sequence alignment analysis showed that the last 13 AA residues of Apela were nearly invariant in chicken, human and rat, implicating similarity of Apela binding Apelin receptor between chicken and mammals. Evolutionary analysis showed that Apela gene had a distant evolutionary pattern between chickens and mammals, implicating specificity of Apela function in chicken. Furthermore, the Apela gene was highly expressed in liver, kidney and pancreas in chicken, and its expression reached a peak in the livers of hens at the peak-laying stages. The expression of the Apela gene was induced by estrogen via ERα.