The role of cyadox and recombinant growth hormone (rGH) on the promotion of growth through epigenetics

Background: Cyadox is an effective growth-promoting antibiotic, which is similar to the role of recombinant growth hormone (rGH). Current studies have shown that cyadox can promote animal growth through altering intestinal microora, improving protein utilization and increasing protein synthesis. Increasing evidence suggests that epigenetics are also closely related to growth. However, the potential role of epigenetics in the cyadox for growth has not been explored. Results: Here, we used recombinant growth hormone (rGH) and cyadox to study the relationship between growth and changes in epigenetics including DNA methylation, histone modication and chromatin structure. Bisulte DNA sequencing (BSP) assay suggested that cyadox and rGH treatments increased IGF-1 expression partially by hypomethylation at CpG sites within the promoter region of IGF-1, which was regulated by DNA methyltransferases (DNMTs). We also observed an enrichment of H3K4me3 and H3K27ac at the promoter regions of IGF-1 by ChIP-qPCR assay, which contributed to an increase in IGF-1 transcription. In addition, immunouorometric assay displayed cellular accessible chromatin structure following the treatment of cyadox and rGH, facilitating the combination of transcription factors and DNA and thus enhancing gene transcription. Conclusions: Taken together, our ndings indicated that cyadox and rGH promote cell growth partially through epigenetic changes, providing a prospect for the development of animal growth-promoting drugs in the future. binding protein; DTT, DL-Dithiothreitol; modied eagle medium; DMSO, dimethylsulfoxide; DNMTs; DNA methyltransferases; DNA methyltransferase DNMT DNA methyltransferase DNMT DNA ECL, chemiluminescence; EGF, epidermal growth factor; fetal bovine guanine; GH, growth hormone; GHR, growth hormone Acetyl-Histone histone deacetylase IGF-1, insulin-like IGF-1R, insulin-like growth factor IUGR, intrauterine retardation; Kdm lysine Kmt lysine lysine

DNA methylation which is dynamically regulated by DNA methyltransferases (DNMTs) [18], plays a key role in maintaining normal cell differentiation and regulating gene expression [19]. For example, DNA methylation is closely associated with programmed inhibition of tissue differentiation and activation of genes involved in tissue-speci c processes [19]. DNA methylation can in uence the growth and development of the embryo [20], fetus and infant [21].
Additionally, different types of histone modi cation also have a signi cant impact on growth and development [22,23]. Tri-Methyl-Histone H3 (Lys4) (H3K4me3) and Acetyl-Histone H3 (Lys27) (H3K27ac) are generally linked to gene activation and growth and development of the body [24]. H3K27ac and H3K4me3 modi cations mainly occur around transcriptional start sites (TSSs) [25,26]. At present, the research of histone modi cation mainly focuses on the growth, maturation and activation of germ cells such as oocytes [23] and sperm [27], and the occurrence and development of diseases [28].
DNA methylation and histone modi cation in the nucleus can also affect the three-dimensional structure of chromatin, thus in uencing on gene expression related to growth and development [29,30]. Chromatin remodeling can lead to changes in the position and structure of nucleosomes, leading to changes in chromatin [31]. Although, there are many studies on growth and epigenetics, but the role of DNA methylation, histone modi cation and chromatin remodeling in antimicrobial growth promotion remains unknown.
It is well known that growth hormone (GH) can enhance somatic growth and tissue repair [32]. Similar to the function of GH, insulin-like growth factor 1 (IGF-1) is also essential for the body's growth [33]. In present study, we used the rat liver cell line BRL because IGF-1 is mainly secreted by the liver which is the main organ of drug metabolism. DNA methylation and histone modi cation can control the expression of IGF-1 [34,35]. However, the relationship between DNA methylation, histone modi cation of IGF-1gene and cyadox-inducing cell growth remains unclear. Thus, in present study, we used most sophisticated molecular biology techniques including quantitative real-time polymerase chain reaction (RT-PCR), western blot, bisul te DNA sequencing (BSP), and immuno uorescence assay to illustrate the potential mechanisms.  Cell viability assay Cells treated with cyadox at different concentrations (0, 1, 2, 4, 6, 8, and 10 µM) were seeded onto 96-well plates with 1 × 10 5 /mL cells per well and then incubated at 37℃ for 0.5, 1, 2, 4, 8, 12, and 24 h, respectively. After incubation, the cells were treated with 0.5 mg/mL solution of MTT (20 µL/well) at 37℃ for 4 h. Next, after discarding the supernatants of each well, 150 µL of dimethylsulfoxide (DMSO) was added to dissolve the purple formazan crystals. After 10 min, the optical density (OD) was read on a microplate reader (Bio-Tek Instruments, Inc., Winooski, VT, USA). The results were calculated as: the relative viability of cell = (OD of the sample well -OD of the blank well) / (OD of control well -OD of the blank well) × 100%. Sample well: pore with cell, MTT solution and drug solution; blank well: a cell-free pore with a medium and MTT solution; control well: pore with cell, MTT solution but without drug solution. From these data, we established the optimal culture time and cyadox concentration. Then, We tested nine concentrations of rGH (1.59 × 10 − 4 , 3.18 × 10 − 4 , 6.37 × 10 − 4 , 1.27 × 10 − 3 , 2.55 × 10 − 3 , 5.09 × 10 − 3 , 1.02 × 10 − 2 , 2.04 × 10 − 2 and 4.08 × 10 − 2 IU/mL) using the same method and conditions as above.

Reagents and chemicals
RNA extraction and RT-PCR assay BRL cells grown to near con uence in 12-well plates were subjected to serum-free medium for 12 h of starvation. The serum-free DMEM medium was replaced with serum-free DMEM medium containing cyadox (1 µM) and rGH (6.37 × 10 − 4 IU/mL). After 0.5, 1, 2 and 4 h, total RNA was extracted from cells using RNAiso plus (TaKaRa, Japan) according to the manufacturer's instructions. Reverse transcription was performed on 1 µg of total RNA by using a PrimeScript RT reagent kit. CDNAs obtained were diluted and used for RT-PCR.
Western blot assay BRL cells (1 × 10 5 /mL) were seeded in 6-well plates and incubated respectively with 1 µM cyadox and 6.37 × 10 − 4 IU/mL rGH for 0.5, 1, 2 and 4 h. Total protein was isolated in a preparation of radio immunoprecipitation assay (RIPA) lysis buffer, then a sonic oscillator was used to break the cells or organelles. The concentration of proteins was quanti ed by bicinchoninic acid (BCA) assay (Beyotime, PR, China). Equal amounts of protein were separated by SDS-PAGE electrophoresis (12.5%) and transferred to nitrocellulose membranes (Millipore, Bedford, MA, USA). The blots were blocked and then labeled with primary antibodies to Acetyl-Histone-H3 (Lys) (D5E4) XP Rabbit monoclonal antibody (mAb) (CST, USA), Tri-Methyl-Histone H3 (Lys4) (C42D8) Rabbit mAb (CST, USA) and β-actin overnight at 4℃. After washing, blots were incubated for 1 h with the corresponding secondary antibody at a 1:5000 dilution. Immunoreactive bands were detected using the Luminata Classico Western HRP Substrate Kit (Millipore, Bedford, MA, USA), and signal intensity and images were captured with an LAS-4000 luminescent image analyser (Fuji lm, Tokyo, Japan).

ChIP-qPCR assay
ChIP-qPCR was used to found the amount of H3K4me3 and H3K27ac in the promoter of IGF-1 gene in BRL cells treated with either 1 µM cyadox or 6.37 × 10 − 4 IU/mL rGH for 4 h. The ChIP assay was completed using the SimpleChIP® Enzymatic Chromatin IP Kit (Magnetic Beads) (CST, USA). The qPCR primers used for the ChIP analysis were shown in Table 3. in the dark. All of the above-cited antibodies were diluted in 1% PBS. Then, the cells were incubated in DAPI (5 µM, Invitrogen) staining solution for 5 min and washed in PBS to monitor cellular uorescence with an UltraVIEW VOX Confocal system (PerkinElmer, Co., Norwalk, Conn., USA) using a 60×, 1.4 NA oil immersion objective lens.

Statistical analysis
Data were analyzed by SPSS 16.0 for Windows. All results were presented as means ± standard deviation (SD). Compared with control group, signi cant differences were indicated by *p < 0.05.

Analysis of MTT
As shown in Fig. 1A, the cell viability of cyadox (1 µM) treatment was the highest in 4 h. When cyadox (1 µM) treated cells for more than 4 hours, the cell viability decreased, which may be due to the increased toxicity of drugs to cells as the time of drug treatment increased; on the other hand, high concentration cyadox (higher than 1 µM) is toxic to cells. Therefore, it was observed the viability was the highest after cyadox (1 µM) treatment for 4 h. And one study revealed that cyadox (1 µM for 4 h) could enhance growth of porcine primary hepatocytes [6]. Thus, the subsequent experiments were performed using 1 µM cyadox for 4 h.
BRL cells viability was measured by MTT assay after treatment with a range of rGH concentrations between 1.59 × 10 − 4 and 4.08 × 10 − 2 IU/mL for 4 h (Fig. 1B). Cell viability was higher when cells were exposed to 6.37 × 10 − 4 IU/mL rGH for 4 h than for the other tested concentrations. Based on these results, we concluded that incubation with 6.37 × 10 − 4 IU/mL rGH for 4 h was more conducive to cell growth.
Promoter hypomethylation mediates upregulation of IGF-1 under rGH and cyadox Furthermore, we also tested whether rGH and cyadox affected genomic DNA methylation level in BRL cells. We found that cyadox and rGH increased 5-mC levels compared to control group. This effect could be reversed by the addition of DNA methylation inhibitors, DAC ( Fig. 2A). The expression of DNMTs regulated the level of 5-mC level was a dynamic change. As shown in Fig. 2F ~ 2K, gene expression of DNMT1 signi cantly increased than control group after BRL cells were exposed to 1 µM cyadox or 6.37 × 10 − 4 IU/mL rGH for 4 h, respectively, which agreed well with 5-mC levels. DNMT 3A and DNMT 3B expression also increased to varying degrees after cells were treated with cyadox and rGH for different times.
To further investigate DNA methylation level in the promoter regions of IGF-1 in uenced by rGH and cyadox, the BSP was used to detect 10 CpG sites (Fig. 2B-E). Among them, methylation levels of CpG sites (-1314, -1284, -1229, -1227, -2) of IGF-1 promoter regions signi cantly decreased under rGH and cyadox. Additionally, the methylation level of CpG site (-128) was lower than control group after rGH treatment in BRL cells. These results indicated that rGH and cyadox reduced the expression of IGF-1 by partially DNA hypomethylation.
Histone modi cation participates in high expression of IGF-1 under rGH and cyadox To investigate the relationship between expression of IGF-1 and histone modi cation, H3K27ac and H3K4me3 levels were detected by western blot after cells were exposed to cyadox and rGH, respectively (Fig. 3A). The total gray value related to H3K27ac (Fig. 3B) and H3K4me3 (Fig. 3C) was analyzed by software of Image J. We found that the level of β-actin was similar in all treatments. The levels of H3K27ac and H3K4me3 were also similar at different times in control group. The levels of H3K27ac and H3K4me3 decreased following cyadox treatment (1 µM) or rGH treatment (6.37 × 10 − 4 IU/mL) for 4 h. By ChIP-qPCR, we found that positions around the promoter 1 and promoter 2 regions of IGF-1 were enriched with H3K27ac and H3K4me3 following cyadox and rGH treatments (Fig. 3D). The same trend was observed for H3K4me3 (Fig. 3E). H3K27ac and H3K4me3 are related to gene activation. Our results suggested that the level of H3K27ac was increased signi cantly at the sites − 1036, -541, + 10, +1291, + 2232 and + 2850 when BRL cells were treated with cyadox or rGH. The level of H3K4me3 was increased signi cantly at the sites − 1036, -541, + 1291, +2232 and + 2850 when BRL cells were treated with cyadox, and at the sites − 1036, -541, + 10, +1291, and + 2232 when BRL cells were treated with rGH. The high level of H3K27ac and H3K4me3 on promoter 1 and promoter 2 of IGF-1 can relax the chromatin and increase transcription levels. In addition, H3K4me3 and H3K27ac can be regulated by Kdm 5a, Kdm 5b, Kmt 2a, and Crebbp, Kat 2a, Kat 2b, HDAC1, respectively. As shown in Fig. 4, these genes mRNA levels presented a dynamic expression in the period of 0.5 ~ 4 h after BRL cells were exposed to 1 µM cyadox or 6.37 × 10 − 4 IU/mL rGH, indicating drugs may regulate histone modi cation level of IGF-1 gene by disrupting gene expression related to histone modi cation.

Cyadox and rGH alter the chromatin status of BRL cells
Histone modi cation is mainly enriched on chromatin. To con rm the effect of cyadox and rGH on the state of chromatin loosening in BRL cells, we carried out immuno uorescence analysis using H3K27ac and H3K4me3 antibodies. The results about H3K27ac suggested that the gaps among nucleosomes were wider and chromatin was lost when cells were exposed to cyadox or rGH. However, when DAC was added to the culture medium together with rGH or cyadox, it showed contrary results (Fig. 5C4, 5C5). In addition, the main metabolite of cyadox, 1, 4-Bidesoxycyadox could also not change gaps among nucleosomes (Fig. 5C6) and couldn't make chromatin loose.
Results of H3K4me3 immuno uorescence displayed that the gaps among nucleosomes were wider, and chromatin was looser when cells were exposed to rGH. Cyadox did not make the chromatin loose compared with the control group. Moreover, when DAC was added to the culture medium together with rGH or cyadox, the visible gaps formed in the nucleus were larger, and the space between the whole chromosomes in the nucleus was smaller (Fig. 6C4, 6C5). In addition, 1, 4-Bidesoxycyadox and couldn't make chromatin loose (Fig. 6C6). Based on the above results, we speculated chromatin remodeling was involved in the growth-promoting mechanisms of both drugs.

Discussion
Our results demonstrated that rGH and cyadox affected DNA methylation and histone modi cation of IGF-1 gene promoter. In addition, these two drugs can also affect the state of chromatin loosening in BRL cells. These ndings shed new light on the relationships between epigenetics and body's growth.
Currently, quite many studies have suggested that cyadox can improve the growth performance of animals including pigs and poultry [6,7,[37][38][39][40]. Antimicrobials including cyadox might in uence growth via an endocrine axis such as IGF-1, GH and epidermal growth factor (EGF) [41]. One study has found that pigs given 50 mg/kg cyadox diet showed the greater average daily gain (ADG) and the better food . Insulin also stimulates hepatic IGF-1 release [46,47]. As known, the increase of IGF-1 expression can promote cell proliferation and differentiation in cells [48,49]. In present study, cyadox also increased cell viability and enhanced the expression of IGF-1 in BRL cells, which suggests that cyadox may promote cell proliferation partially through high expression of IGF-1 gene.
DNA methylation is an important growth-promoting mechanism of cyadox and rGH. Although, there are many studies suggested that cyadox as growth promoting agent, however, those were focused on the changes of intestinal microorganisms and apparent digestibility of nutrients [50]. In recent years, the role of DNA methylation in animal growth has gained extensive attention. DNA methylation mainly takes place in CpG dinucleotides, adding a methyl group to the cytosine to form 5-mC [51]. DNA methylation is closely related to increased gene expression, whereas hypermethylation usually leads to inhibition of genes expression [52]. DNA methylation can regulate several biological events, including X-chromosome inactivation, embryonic development, genomic imprinting, transcriptional regulation, and chromatin modi cation [53]. However, the role of DNA methylation in cyadox-induced growth is unknown. The present study demonstrated that cyadox and rGH induced genome-wide DNA hypermethylation, which is regulated by DNMT1. Hypermethylation is usually not harmful, whereas hypomethylation typically leads to growth arrest or apoptosis in mammalian cells [54]. Thus, the proper hypermethylation in cells is thought to be bene cial for growth and development. To our knowledge, this is the rst study to demonstrate the effects of rGH and Cyadox on whole genome 5-mC level. Importantly, we found that DNA methylation level of IGF-1 promoter was reduced in different extent at ten CpG sites after rGH and cyadox treatments, which facilitates the transcription of IGF-1, increasing IGF-1 expression. These ndings indicated that IGF-1 expression would be increased due to local hypomethylation under rGH and cyadox, which is in line with the ndings of a report [55].
DNA methylation is regulated by DNMTs mainly DNMT1, DNMT 3A, DNMT 3B. DNMTs use Sadenosylmethionine (SAM) as a methyl donor to catalyze a methyl group to add to the cytosine ring to form methylcytosine [56]. DNMT1 is responsible for the restoration of hemimethylated to full methylation (maintenance methylation) and is the predominant mammalian DNA methylating enzyme. DNMT 3A and DNMT 3B create new DNA methylation sites (de novo methylation) [57]. In present study, the gene expression of DNMTs (DNMT1, DNMT 3A) decreased rst and then increased in the time effect of cyadox and rGH. This may be responsible for maintaining cell homeostasis through demethylation during the compensatory process; and then with the continuous stimulation of drugs, drugs may downregulate certain genes through hypermethylation, and thus play its own pharmacological role. Although the mRNA lever of DNMT1 and DNMT 3A genes signi cantly increased than control group after BRL cells were exposed to 1 µM cyadox for 4 h, it only re ects genome-wide methylation level in cells. After the drug acts on the cells, some genes may be hypermethylated and others may be hypomethylated or unchanged.
H3K4me3 and H3K27ac can induce gene activation and play an important role in the growth [24].
However, little known about how rGH and cyadox affect H3K27ac and H3K4me3 levels. Present study found that the H3K27ac and H3K4me3 levels are reduced by cyadox and rGH treatments. Modi ed histones can interact with other proteins to affect gene expression [58]. Histone acetylation can activate gene transcription and mainly occurs in lysines [59]. Histone methylation can regulate the chromatin structure and growth of the cells, and mainly occurs in lysine and arginine residues [60]. The decreased level of total H3K27ac and H3K4me3 in BRL cells exposed to cyadox and rGH may be the result of the combination of different genomic protein modi cations. Cyadox and rGH can increase the levels of H3K4me3 and H3K27ac around promoter 1 and promoter 2 of IGF-1, inducing IGF-1 expression. Some active enhancers are usually marked by H3K27ac, which is around transcriptional start sites (TSSs) [25,26,61,62]. Some active enhancers are also associated with H3K4me3, which is related to RNA polymerase II activity [63]. Thus, modi cation of H3K4me3 might enable the activity of RNA polymerase II, thus increasing IGF-1 transcription initiation. In addition, the increased level of H3K27ac was rst to be found in promoter 1 and promoter 2 of IGF-1 when the BRL cells were treated with rGH and cyadox, respectively.
Histone methylation can be regulated Kdm 5a, Kdm 5b [64] and Kmt 2a [65]. Kdm 5a and Kdm 5b can inhibit H3K4me3 [66], however, Kmt 2a, a methyltransferase, can activate H3K4me3 [67]. We found that cyadox and rGH increased the level of Kdm 5a and Kdm 5b expression. The high expression of Kdm 5a and Kdm 5b and low expression of Kmt 2a caused the low level of H3K4me3 in BRL cells under cyadox and rGH. Additionally, Histone acetylation can be regulated by genes of Crebbp [68], Kat 2a [69], Kat 2b [70] and HDAC1 [71]. Crebbp, Kat 2a, and Kat 2b are histone acetyltransferase, and HDAC1 is a histone deacetylase. Our present results suggested that cyadox and rGH downregulated the level of Kat 2a and Crebbp, and increased HDAC1 expression, which could lower the H3K27ac level in cells.
Chromatin structure is closely related to gene activation and inhibition in cells [72]. The current results displayed that the visible space in the nucleus was smaller and gaps among nucleosomes were wider after BRL cells were treated with rGH and cyadox, which may be caused by DNA methylation and histone modi cation changes [73]. These results indicated that rGH and cyadox loosens the chromatin of BRL cells, facilitating transcriptional activation of growth-related genes.

Conclusion
In summary, our research ( Fig. 7) revealed that cyadox and rGH elevated IGF-1 expression via hypomethylation and the enrichment of H3K4me3, H3K27ac in the promoter region of IGF-1. Additionally, accessible chromatin structure of BRL cells induced by cyadox and rGH is conducive to genes transcription and promotes cell growth. Taken together, the current data provide a novel insight for drugsinduced growth of animals from the perspective of epigenetic changes, contributing to future research in the eld.  following treatment with rGH (6.37×10-4 IU/mL). All the results were shown as mean ± SD (n = 3), which were three separate experiments performed in triplicate. Signi cant differences were indicated by *p < 0.05, **p < 0.01, ***p < 0.0002, ****p < 0.0001, versus control.