Chinese Herbs Antagonize The Effects of Bisphenol A On Kiss1 Expression Through Epigenetic Regulation In The Promoter Region

Background: Bisphenol A (BPA), an environmental endocrine disruptor, is involved in precocious puberty. Chinese Herbs for Nourishing Yin and Purging Fire (NYPF) have been effective in delaying early onset of puberty. However, the underline mechanism is unknown. As epigenetics plays an important role in advanced puberty caused by extraneous factors. We intend to investigate the epigenetic impact of BPA and NYPF herbs on Kiss1 gene, a vital gene driving puberty onset, in hypothalamic neuron cells. Methods: GT1-7 cells derived from GnRH neurons were used in the study. A series of concentration of BPA and serum containing NYPF herbs were prepared in advance. GT1-7 cells were administrated with BPA or vehicle rst, followed by NYPF herbs or normal saline (NS) for 24 hours. Afterward, quantication of Kiss1 expression was performed through Realtime-PCR and Weston-blot methods. In addition, epigenetic modications at Kiss1 promoter were examined through Bisulphite Sequencing (BSP) and Chromatin immunoprecipitation combined real-time PCR(cid:0)Chip-qPCR(cid:0)assays. Results: Compared to control group, GT1-7 cells with BPA 10mg/l administration presented lower methylation status (61.25% Vs. 76.56%) concomitantly with enhanced abundance of mixed-lineage leukemia 1(MLL1) (1.05%±0.15 Vs. 0.27%±0.01) and Tri-Methyl-Histone 3 at lysine 4 (H3K4me3) (0.77% ±0.51 Vs. 0.02%±0.04) at Kiss1 promoter, leading to elevated expression of Kiss1. On the contrary, TCM for NYPF promoted increased DNA methylation (83.75% Vs. 72.5%), accompanied with less deposition of MLL1/H3K4me3 (MLL1: 0.14%±0.02 Vs. 0.97%±0.09; H3K4me3: 0.01%±0.00 Vs. 0.57%±0.30) at Kiss1 promoter than BPA 10mg/l group, which resulted in markedly inhibition of Kiss1 expression. Conclusions: BPA may induce elevated expression of Kiss1 through inuencing DNA methylation and histone modication at Kiss1 promoter. And NYPF herbs reverse the adverse effects of BPA on Kiss1 gene through epigenetic modication as well.


Introduction
Precocious puberty is de ned as early emergence of secondary sexual characteristics in girls (before 8 years old) and boys (before 9 years old). Global trends with advanced puberty, as well as early menarche for girls, within the recent hundred years have been reported. Recent researches implicates that both environmental factors and nutritional status are involved in the process of puberty onset 1 .
Environmental endocrine-disrupting chemicals (EDCS), including Bisphenol A (BPA) and its analogs and phthalates, have attracted attention due to their role in induction of precocious puberty 2 . BPA is mainly used to synthesize polycarbonate or epoxy resin and widely exists in the environment. It is usually detected in children's blood or urine samples, even in maternal amniotic uid 3,4 . Studies have shown that BPA is an estrogenic compound and induces precocious puberty in female pups 5,6 . Epidemiological researchers also nd that the content of BPA in serum or urine increases in precocious puberty girls 7,8 , further supporting the theory that BPA disrupts the regular process of puberty in both children and animal pups.
Traditional Chinese Medicine (TCM) is an effective treatment for precocious puberty in China. According to TCM theory, precocious puberty is generally caused by changes of Yin De ciency and Fire Excess, or Pathogenic Fire generated by Liver Depression, and Phlegm-Dampness Stagnation. Therefore, the therapeutic principle is nourishing Yin and purging Fire. We have explored for 40 years, in TCM clinical practice, a Nourishing Yin and Purging Fire formula capable of effectively delaying advanced onset of puberty. Preclinical experiments have shown that this NYPF herbs delay puberty process by regulation of Kiss1 9 , mTOR 10 and lin28/let7 11 expressions.
GnRH neurons in the hypothalamus, dynamically synthesizing gonadotropin releasing hormone (GnRH), are considered as puberty control center. GnRH, coded by GnRH1, is regulated by activatory and inhibitory neurotransmitter inputs. Before puberty initiation, it is mainly controlled by inhibitory neurotransmitters and only a small amount of GnRH is synthesized. As the age of puberty is approaching, activatory synaptic inputs become gradually dominant, leading to GnRH expression increased 12 . Kisspeptin, coded by Kiss1 gene, has been con rmed as one potent molecular to stimulate GnRH1 transcription. Additionally, Kiss1 neurons integrate nutrition and environmental signals and project onto GnRH neurons.
Hence, Kiss1 gene plays an important role in environmental-mediated control of sexual development 13 .
Epigenetic mechanisms, including DNA methylation, histone modi cation and miRNA and lincRNA regulation, have been demonstrated to play crucial roles in early onset of puberty induced by environmental factors 12 . Previous studies suggests that BPA exposure during early developmental period of animal pups alters expressions of DNA methyltransferase and histone post-translational regulatory enzymes 14 . Studies on TCM also found that herbal monomers or compounds are also involved in regulation of epigenetic related enzymes 15,16 . However, the epigenetic mechanisms underlying BPA and TCM's involvement in adolescence remain unknown. GT1-7 cells are derived from GnRH neurons and express both Kiss1 and GnRH1 genes. In this study, the cells were chosen to explore the relationship between BPA/ NYPF herbs and Kiss1/GnRH1 expressions, and furthermore the underline epigenetic mechanisms.

BPA Preparation
First, a series of concentration of BPA solutions (500ng/l, 50ug/l, 500ug/l, 5mg/l and 10mg/l) were prepared by dissolving different dose of BPA in an equal volume of phosphate buffered saline (PBS). Then, BPA-containing PBS diluted in cell medium, depending on the group designation, was used for an additional culture period of 24 hours.

Animals
Pregnant SD rats were purchased from Zhejiang Vital River Laboratory Animal Technology Co., Ltd

Pharmaceutic Serum
The preparation methods of Pharmaceutic Serum were based on previous research 18 with a little modi cation. The brief protocol was as follows. Female rats at postnatal 21 (P21) were randomly divided into TCM group or normal saline (NS) group (N=10 each). They were continuously given TCM for NYPF or equal volume of NS as designated by gavage twice a day, respectively, from P21 to P25. On the morning of P25, one hour after the last administration, rats were sacri ced and blood samples were drawn from abdominal aorta. The serum was then separated by centrifugation and inactivated through heating at 56℃ for 30 min followed by aseptic ltration. Finally, the samples were stored at -80℃ for later use. GT1-7 cells were intervened rst by BPA, followed by serum containing TCM or NS at the nal concentration of 10% (vol/vol). Then the cells were proceeded to culture for an additional 24 hours before harvesting.

Real-Time PCR (RT-PCR) Analysis
The mRNA levels of Kiss1 and GnRH1 were detected by RT-PCR. Total RNA was extracted using RNAiso Plus reagent (9109, TAKARA, Japan) and quanti ed by ultraviolet spectrophotometry (Nanodrop2000, Thermo-Scienti c). Then 1ug of RNA per sample was transcribed into cDNA using reverse transcriptase kit (RR036A, Takara, Japan). Each tube for RT-PCR contained 2ul of cDNA, 1.6ul of primers (0.8ul each), and TB Green® Premix Ex Taq™ II reagent (RR820B, TAKARA, Japan) in a nal volume of 20ul. And PCR ampli ed conditions were as follows: a cycle of 30s at 95°C for predenaturation and followed by 40 cycles for denaturation at 95°C for 15s and annealing at 60°C for 60s.
Glyceraldehyde-3-phosphate dehydrogenase (GAPDH) was used as an internal reference, using the 2 −ΔΔCt method to calculate the relative mRNA levels. All the primers used in the assay (shown in Table 1) were designed by Primer3 software and synthesized by Shanghai Sangon Biotech Inc. (Shanghai, China).

Chromatin Immunoprecipitation Assay
To assess the content of MLL1 and H3K4me3 at the promoter region of Kiss1, we performed Chip assays using chromatin extracted from each group according to the instructions of Magna ChIP® A/G Chromatin Immunoprecipitation Kit (17-10085, Millipore, USA). The brief protocols were as follows: cell samples counting 2*10 6 each group were washed once in ice-cold phosphate-buffered saline (PBS) containing a protease inhibitor cocktail II. Thereafter, cells were crosslinked by exposing to 1% formaldehyde for 10 min at room temperature. After two additional washing steps the samples were lysed with cell and nuclear lysis buffer, and sonicated for 30 s by 10 times to yield chromatin fragments of 200~500 base pairs (bp) using the Ultrasonic Processor (Bioruptor Pico, Diagenode, Belgium) at 4℃.
Size fragmentation was con rmed by agarose gel electrophoresis. The sonicated chromatin was clari ed by centrifugation at 12,000g for 10 min at 4℃, then take 50ul of supernatant into a clean EP tube, brought up to 500ul in chip dilution buffer for each reaction. Input sample with 50ul was removed from each tube of chromatin in advance and the remains were incubated with 5ug of MLL1(05-765, Millpore, USA) or H3K4me3(9727, CST, USA) antibodies and with 25μl of protein A or G beads solution (Dynabeads) for 4 hours with rotation at 4°C. Four hours later the beads were washed rst with 0.5 ml low-salt wash buffer, followed by high-salt wash buffer, LiCl buffer and nally with TE buffer. Thereafter, the complexes were eluted with 100μl of chip elution buffer containing proteinase K at 62°C for 2 hours with rocking. To reverse the crosslinking reaction the samples were incubated at 95°C for 10 min. Then DNA for qPCR analysis was recovered using ChIP DNA Clean & Concentrator columns. All the reagents mentioned above were included in the Magna ChIP Kit.

Real-time PCR Detection of Chromatin Immunoprecipitated DNA
The 5'-anking regions of Kiss1 gene between nt -2000 and nt +1, using transcriptional start site (TSS) as reference point, were supposed as promoter region. The putative promoter fragments were predicted online using the website (https://www.fruit y.org/cgi-bin/seq_tools/promoter.pl) and quali ed by RT-PCR using Chromatin Immunoprecipitated (IP) samples. Each reaction system was comprised of 2uL of IP or input sample, 2ul of primers (1ul each) and TB Green® Premix Ex Taq™ II reagent (Tli RNaseH Plus) (RR820B, Takara, Japan) in a nal volume 20 ul. Data are expressed as % of IP signal/input signal.

Statistical analysis
All the data were presented as mean ± standard error (SEM) and statistical analyses were performed using SPSS 17.0 (IBM Corporation, Armonk, NY, USA). One-way ANOVA was used to make comparisons among multiple groups. The data were rst subjected to a normality and an equal variance test. If the variance was homogeneous, the LSD tests were used. If not, the Dunnett T3 tests were conducted. In addition, comparison of proportion analysis was evaluated by Chi-square test. Values were considered to be signi cantly different as P < 0.05.

Effects of BPA on Kiss1 and GnRH1 expressions
Compared to control group, GT1-7 cells with BPA 10mg/l administration presented increased mRNA levels of Kiss1 and GnRH1 (P <0.05), as well as protein expressions (Fig1). In addition, GnRH1 mRNA levels also elevated in BPA 500ug/l group than control group.
Effects of TCM on Kiss1 and GnRH1 expressions TCM for NYPF inhibited the overexpression of Kiss1 and GnRH1 induced by BPA exposure. As shown in Fig2, NYPF herbs presented inhibiting effects on both mRNA and protein levels of these two genes.

Effects of BPA/TCM on DNA methylation at Kiss1 promoter
The fragments between nucleotide (nt) -2000 and nt +1 in the 5'-anking regions of Kiss1 gene were sequenced. Methylation differences were detected in the fragment between nt -547 and nt -155 among the four groups. As shown in Fig 3, compared to control group, the levels of DNA methylation decreased with BPA 10mg/l administration. Conversely, the methylation levels elevated in TCM mediated group than NS group. Furthermore, there were signi cant differences in methylation incidence at sequences ranged from nt -436 to nt -377, containing four methylation sites of CpG2/3/4/5, between BPA and control group (37.5% Vs. 62.5% P<0.05). The methylation rate of CpG3 (nt -427) was signi cantly different between TCM and NS group (100% Vs. 50% P<0.05).
Effects of BPA/TCM on Histone modi cation at Kiss1 promoter Chip-qPCR analysis were targeted at the DNA fraction between nucleotide (nt)-2068 and +1 using IP samples incubated with H3K4me3 or MLL1 antibodies. As shown in Fig 4, the content of MLL1 and H3K4me3 elevated simultaneously at the fragments between nt -657 and nt -540. Once administration of Chinese herbs, both H3K4me3 and MLL1 deposition decreased signi cantly at the same region. In addition, similar changes of H3K4me3 deposition were found in the following fragments (nt -2068 to -1948, nt -865 to -736). However, there was not alteration of MLL1 content at the two fragments.

Discussion
There is not dose-effect relationship between BPA exposure and precocious puberty. Invitro trials performed on embryonic hypothalamus cells from mice have shown that a BPA dosage of 20nM-20uM doesn't have a signi cant effect on GnRH1 mRNA expression. However, an increase to 200uM can signi cantly inhibit its expression 19 . Additionally, studies on the rhesus monkey hypothalamus establish that 10nM BPA inhibits the expressions of both Kiss1 and GnRH1 20 . Another study on hard-bone sh shows that 10ug/l of BPA can signi cantly increase the expression of Kiss1 and GnRH1 genes 21 . Our study performed on GT1-7 cells have veri ed that 10mg/l of BPA, equivalent to 440uM, signi cantly enhances expressions of Kiss1 and GnRH1.
Epigenetic modi cation is regarded as an important way of environmental factors, drugs and food affecting human growth and development. DNA methylation is a common manner of epigenetic modi cation. It has been demonstrated that DNA hyper-methylation is associated with gene silencing, while hypo-methylation is associated with gene activation. This is concurrent with prior ndings that hyper-methylation of Kiss1's CpG island near the TSS results in decrease in mRNA and protein levels of the Kiss1 gene in human tumor cell lines 22 . The expressions of Kiss1 are signi cantly increased when Azacytidine, one demethylated drug, is administrated, suggesting that DNA methylation affects the expression of Kiss1 23 . However, the CpG island is not present in the rodent Kiss1 gene, hinting at racial differences in its regulation 23 . In our study, although CpG island doesn't exist near the TSS of Kiss1 gene in mouse GT1-7 cells, DNA methylation incidence at the fragments between nt -547 to nt -155 is veri ed to associate with Kiss1 expression. 10mg/l of BPA promotes DNA hypo-methylation, contributing to expression of gene. However, the NYPF reverses DNA hypo-methylation induced by BPA, which subsequently leads to repression of Kiss1 expression. Therefore, our study suggests that the DNA methylation between nt -547 to nt -155 near TSS, especially at the range from nt -436 to nt -377 (the fragment of CpG2/3/4/5), may be a target of BPA and TCM for Kiss1 regulation in GT1-7 cells.
Chromatin histone modi cation is another major manner of epigenetic modi cation. Previous research identi es MLL1 and MLL3 as central components of an activating epigenetic machinery that dynamically regulate Kiss1 gene in rat KNDy neurons 24 . Preceding puberty, MLL1 changes the chromatin con guration at the promoters of Kiss1 from repressive to permissive by H3K4me2/3 deposits 24 . In our study, we also have found that high levels of H3K4me3 and MLL1 deposition at chromatin region from nt -657 to nt -540 near TSS is accompanied by enhanced Kiss1 gene expression. As promoters of active genes display high levels of active histone maker H3K4me3 and MLL1 24 , suggesting the fragment behaves as a promoter domain. It is reported that BPA involves in ovarian developmental process through modulation of DNA methyltransferase, TET enzyme, and histone modi cation enzymes 25,26 . Our study also manifests that BPA enhances gene expression through modi cation of MLL1 and H3K4me3.
This experiment indicates both DNA methylation and histone modi cation are two epigenetics mechanisms involved in regulation of Kiss1 expressions by TCM and BPA. It might not only represent the epigenetic mechanisms through which BPA induces precocious puberty but also be a viable indicator of the bene cial therapeutic role of TCM.
This experiment was carried out in-vitro, and further studies in-vivo are needed to clarify the relationship between BPA / TCM and precocious puberty.

Conclusions
In summary, our study shows that BPA promotes  Figure 1 The  and GnRh1. c Relative protein levels of Kiss1 and GnRH1.The data are presented as mean ± SEM (n = 3 in each group). **P < 0.01, *P < 0.05.

Figure 3
The effects of BPA and TCM of NYPF on methylation incidence at Kiss1 promoter. a DNA methylation status at Kiss1 promoter in vehicle group. b DNA methylation status at Kiss1 promoter with 10mg/l BPA administration. c DNA methylation status at Kiss1 promoter in BPA combined NS intervention group. d DNA methylation status at Kiss1 promoter in BPA combined TCM mediated group. e Total methylation levels at the region between nt -547 and nt -155 near TSS in each group. f The methylation levels at each CpG site in four groups. All the data was presented as percentages.