Epigenetic Mechanisms Of Selective Expression Of CYP3A Isoforms During Human Liver Development

Background and objectiveThe human cytochromes P450 3A(CYP3A) comprises the four CYP genes 3A4,3A5, 3A7 and 3A43. The CYP3 A subfamily enzymes are the most abundant CYPs in liver and intestinal enterocytes and play a major role in the metabolism of 60% of clinically used drugs from almost all therapeutic categories. CYP3A7 is the most abundant CYP3 A enzyme in fetal liver while CYP3A4 exclusively expresses in adult liver. For adults, the expression of the CYP3A4 may attribute to the genetype and induction status while development for children may have a profound effect on the changes in the drug metabolizing enzyme expression. Recent studies suggest that the developmental pattern of the CYP3 A activities is not liner with the age of growth from the prenatal period to adulthood. The greatest changes occur during development and impact drug ef?cacy in the neonate and young child.The profound changes between CYP3A7 and CYP3A4 in enzyme expression and activities will lead to a difference in drug ef?cacy during hepatic development. A pediatric patient does not deal with miniature men and women and reduced doses cannot be simply extrapolated from body weight adjustments in adults. Otherwise, a risk of adverse events will be possible. Hence, understanding the activity, developmental changes and specific epigenetic regulation mechanisms of individual CYP3 A forms is crucial for improving drug dosing and leading to optimal safety and ef?cacy in children.More-recent data suggest that the polymorphic nature of the CYP3A5 expression could cause clearance differences while polymorphisms in CYP3A4/3A7 genes seldom affect CYP3 A phenotyping. The interindividual variations of CYP3 A ontogeny can not be explained by genetic factors; whereas, the underlying molecule mechanisms regulating CYP3 A ontogeny remain poorly understood. How the CYP3A7 gene expression is suppressed while how hepatic the CYP3A4 gene is activated in the ?rst few weeks or months after birth ? Recent studies show epigenetic regulations may play an important role in hepatic organogenesis and cell differentiation. In our previous experiment we found that histone H3K4 dimethylation(H3K4me2) and histone H3K27 trimethylation(H3K27me3) were associated with the developmental and expression patterns of CYP3A4/3A7 genes. Hence, epigenetic mechanisms may de?nitely regulate CYP3 A ontogeny.How do the epigenetic mechanisms regulate the developmental expression of CYP3A? What determines the status of folding or unfolding chromatin? Which nuclear receptors regulate the developmental expression of CYP3A? How are histone modi?cation molecules such as H3K4me2 involved in regulating developmental expression patterns? In conclusion, we developed the hypothesis that nuclear receptors bind to CYP3A4/3A7 genes response elements at multiple genomic locations and recruit the epigentic modifers enzymes involved in methylation/ demethylation and change histone modi?cation status and thus facilitate the transcriptional regulation.Based on the above hypothesis, our objectives were :(1) to characterize the distinct developmental expression patterns of CYP3 A ontogeny and nuclear receptors ontogeny;(2) to elucidate the relationships between histone modification and the dynamic developmental expressions of CYP3A4/3A7 and nuclear receptors; and(3) to investigate the critical and complicated mechanisms needed to regulate the human CYP3A7-CYP3A4 switch during liver development, and thus to provide knowledge for the safe use of drugs in the neonate and young child.MethodsVivo study1. Subjects All human tissues were collected with informed consent following ethical and institutional guidelines of the First Affiliated Hospital of Zhengzhou University. A total of 110 liver tissues from an estimated gestational age(EGA) of 13 weeks to a postnatal age(PNA) 72 years were used in this study. This total includes 53 fetal samples and 97 postnatal samples. The 97 postnatal samples were derived from patients undergoing scheduled liver resections. Human fetal livers were obtained from spontaneous or elective abortions. The estimated gestational age was determined by ultrasonic measurement. The patients enrolled in this study did not take any other drugs(CYP3A4 inhibitors and inducers) and patients with impaired hepatic or renal function were excluded from participation. Even patients suffering from severe infectious diseases(hepatitis B and C, or HIV) were excluded. We used fetal samples at 13 weeks(EGA) as the control for each gene.2. Isolation of RNAs and realtime-q PCR analysis Total RNAs were extracted from human fetal, postnatal liver samples. RNAs were reverse-transcribed into c DNA using a Primer Script RT reagent kit with g DNA Eraser according to the manufacturer’s instructions. Real-time quantitative PCR(QPCR) for expression levels of CYP3A4, CYP3A7, and nuclear receptors was performed by the fluorescent dye SYBR Green methodology using the SYBR Premix Ex Taq TM and the ABI 7500 fast. The relative quantification of the steady state m RNA levels was calculated after normalization of glyceraldehyde-3-phosphate dehydrogenase(GAPDH). Relative quantification analysis was performed using the comparative CT(2–ΔΔCT) method.3. Measurement of CYP3A4/3A7 activity Liver samples were homogenated and microsomes were extracted by differential centrifugation. Protein concentrations were determined by using a BCA protein assay kit.Activities of midazolam 1-hydroxylation(MZ1H) were measured to estimate CYP3A4 activity and the velocity of decrease of DHEA was used to confirm CYP3A7 dependent metabolite in fetus and postnatal liver microsomes. High-performance liquid chromatograph(HPLC) was performed using Waters e2695. Km and Vmax values were calculated by linear regression and plotted with the Eadie-Hofstee equation.4. Preparation protein samples and western blotting analysis Proteins were extracted from liver samples and protein concentrations were determined using a BCA protein assay kit. Proteins were separated by 10% sodium dodecyl sulfate-polyacrylamide gel electrophoresis(SDS-PAGE) and transferred to PVDF membrane. CYP3A4 and CYP3A7 were detected with CYP3A4 and CYP3A7 rabbit polyclonal antibodies with a dilution of 1:20000 and a dilution of 1:5000respectively followed by detection with a chemiluminescent agent after incubating with the corresponding secondary antibodies with a dilution 1:2000 coupled to horseradish peroxidase(HRP). Samples were normalized to GAPDH, which was detected with mouse monoclonal antibody with a dilution of 1:10000. Signals were visualized and scanned using a chemiluminescence imaging system. The relative intensity of each band was determined by computer assisted densitometry.5. Qualitative analysis of CYP3 A protein by liquid chromatography-tandem mass spectrometry( LC-MS/MS)Fetal liver microsomal samples were extracted and the SDS-PAGE gel electrophoresis was performed and followed by coomassie brilliant blue staining. The gel articles containing 58 k Da were treated with 0.01 mg/ml trypsin. LC-MS/MS analyses were performed using nanospray way and the positive ions were tested mainly. The mass charge ratios of peptide and peptide fragments were collected by full scan(MS2 scan).MS/MS spectra were analyzed by the Mascot server(version 2.2). To match the raw file,the corresponding database was searched using Mascot 2.2 software and finally the protein identification results were obtained.6. Chromatin immunoprecipitation(Ch IP) sequence assay Chromatin immunoprecipitation(Ch IP) was performed using a Chromatin Immunoprecipitation Assay Kit according to the manufacturer’s instructions.Subsequently, the samples were purified and the DNA was used for CHIP-sequence.Pre-sequencing control assays were performed. Sequencing on Illumina Hi Seq 2000 incorporated the major stages in DNA sequencing, consisting of image analysis, base calling and alignment by BGI company. Reads were aligned to a human genome(UCSC HG19) using BOWTIE software(V2.1.0). The mapped reads were used for peak detection by MACS v2.0.9 software. Data Visualization, a brief introduction of visualizing the Ch IP-Seq profiles of each sample was given in the UCSC Genome Browser(http://www.genome.ucsc.edu). The overall fold changes of each epigenetic modification in a gene were represented by average peak values of all active regions within margins of the associated genes. Statistically significant Ch IP-enriched regions(peaks) were identified by a comparison of two group samples with a P-value.Vitro study1. Effects of methylation of histone on CYP3A4/3A7 expressions Human primary fetal liver cells were isolated and cultured and then treated with 0.6,1.2, 2.4 m M pargyline for 48 hours. Hep G2 cells were cultured and then treated with 0.03,0.3, 3 m M pargyline for 48 hours. After treatment, total RNAs were prepared from the cells to observe the expression of CYP3 A m RNA in primary fetal cells and Hep G2 cells.2. Effects of h HNF4 A si RNA on the expression of CYP3 A Hep G2 cells were cultured and transfected of human hepatocyte nuclear factor4A(h HNF4A) si RNA to investigate the effects of the expression of CYP3 A for 48 hours.Hep G2 cells were cultured and h HNF4 A si RNA was transfected to investigate the effects of the expression of CYP3 A for 48 hours. Then all total RNAs were extracted and reverse-transcribed into c DNA. Relative quantification analysis was performed using the comparative CT(2–ΔΔCT) method.3. Effects of histone modi?cation status in enhancers and promoters regions on expression and activities of CYP3A4/3A7 gene Hep G2 cells were cultured and then treated with 3 m M pargyline for 48 hours. The treated cells and control cells were crosslinked and the chromatins were sheared by sonication respectively. The Ch IP assay was performed with H3K4me2, and Ig G antibodies respectively. DNA and proteins complexes were washed, eluted and reverse crosslinked. Subsequently, the samples were purified by a DNA Purification Kit. The precipitated DNA was resuspended and was used for QPCR. QPCR analysis was carried out using the ABI PRISM 7900 Fast Real-Time PCR System. Primers were used to detect different regions of CYP3A4 promoter(-362 to +53) and enhancer(-7836 to-6093) and CYP3A7 promoter(-163 to +103) and(-4054 to-3421)(-6265 to-6247). Occupancy of H3K4me2 was normalized to input and expressed as the fold difference relative to control cells.Statistical Analysis SPSS version 19.0 was used for statistical analysis. Results are reported as mean ±S.E. Comparison of two groups was made with an unpaired, two-tailed Student’s t test.Pearson correlation analyses between m RNA levels and protein were performed. A P value of less than 0.05 was considered to be statistically significant.Results1. CYP3A4/3A7 and nuclear receptor expressions CYP3A7 m RNA expression was detected at 13 wks(EGA) and reached maximal level at 2.31(normalized to GAPDH) at 25 wks(EGA) and slightly decreased before birth and then showed substantial decreases by 0.05(normalized to GAPDH) during the ?rst 1year after birth. During the postnatal age(PNA)(1-18 yr) and adulthood(> 18 yr),CYP3A7 m RNA expression levels significantly decreased compared to the levels in the fetus. The expression levels of CYP3A4 m RNA were low at 2.43(normalized to GAPDH)before birth and increased slightly within the ?rst year after birth at 4.08(normalized to GAPDH). Then, the expression levels of CYP3A4 m RNA in PNA(1-18 yr) and PNA(>18 yr) were significantly higher at 21.81 and 10.81(normalized to GAPDH) respectively compared to the levels in the fetus(P<0.001,P < 0.01). CYP3A4 and CYP3A7 m RNA were expressed at highly variable levels(256, 628 fold) in adult and fetal livers respectively.The ontogenic expression of pregnane X receptor(PXR), constitutive androstane receptor(CAR), retinoid X receptor(RXR), CCAAT/enhancer-binding protein B(CEBPB), peroxisome proliferator-activated receptor A(PPARA), hepatocyte nuclear factor1A(HNF1A) and HNF4 A m RNA in postnatal livers signi?cantly increased compared with those in fetal livers, whereas the ontogenic expression of forkhead box protein A3(FOXA3) m RNA in postnatal livers(0-18yr) signi?cantly decreased compared with that in fetal livers. The correlations between PXR, HNF4 A, HNF1 A and CYP3A4 were significant in adult livers and most prominent for HNF4A(R = 0.52, P<0.01) and PXR(R = 0.47, P < 0.01). Correlations were also found between FOXA3,glucocorticoid receptor(GR) and CYP3A7 in fetal livers and most prominent for GR(R= 0.48, P < 0.001).2. Hepatic CYP3A4 and CYP3A7 activity CYP3A7 showed a signi?cantly higher activity with Vmax(5.90 μmol/min/mg protein) in fetal liver microsomes(EGA)(13-38 wks) than that of postnatal liver microsomes(Vmax =0.75 μmol/min/mg protein). CYP3A4 showed a signi?cantly higher activity in adult liver microsomes(18-60 yr) with Vmax(0.26 nmol/min/mg protein) than that of fetal liver microsomes(Vmax = 0.04 nmol/min/mg protein)(EGA)(13-38 wks).Similar trends with CLint were observed. In contrast, the apparent Km values for CYP3A4-catalyzed midazolam and CYP3A7-catalyzed DHEA 16a- hydroxylation did not signi?cantly differ between fetal liver microsomes and postnatal liver microsomes,respectively. Overall, fetal and postnatal liver samples were characterized by large interindividual variability with levels ranging from 0 to 36.4 μmol 3A7/mg protein(highest levels measured for a sample with EGA of 24 wks).3. The switch effect of histone methylations on the CYP3A4/3A7 genes ontogenic expression Genome-wide maps and comparison of H3K4me2, H3K27me3 modifications in human fetal and adult liver were performed. Speci?cally, on chromosome 7, the CYP3A4 gene was marked with strong enrichment of H3K4me2(fold enrichment: 4.17) around its TSS(± 5kb) in adult livers versus fetal livers(fold enrichment: 0). In contrast, the CYP3A7 gene exhibited strong enrichment of H3K4me2, around its TSS(± 5 kb) in fetal livers(fold enrichment: 4.82), compared with adult livers(fold enrichment: 0)respectively. There were enrichments of H3K4me2 around the PXR gene locus(chr3:11950032 5-119502876) in adult livers(fold enrichment: 6.81) and in fetal livers(fold enrichment: 3.75), around the HNF4 A gene locus(chr20: 42983566-42986259) in adult livers(fold enrichment: 7.5) and in fetal livers(fold enrichment: 3.73), around the HNF1 A gene locus(chr12: 121421565-121422822) in adult livers(fold enrichment:5.02) and in fetal livers(fold enrichment: 4.38) and around the RXR gene locus(chr9:137287725-137288917) adult livers(fold enrichment: 8.74) and in fetal livers(fold enrichment: 4.52), respectively.For the H3K27me3 profiles along the CYP3A4/7 and nuclear receptors locus during liver development:there were H3K27me3 fold changes along the CYP3 A on chromosome7 using CHIP-Seq on the UCSC between CYP3A4 and CYP3A7 gene locus 3 kbp downstream of 3’domains in fetal liver(fold enrichment 8.66).Our data as well as bioinformation analysis emphasize that H3K4me2 has been shown to be enriched in highly expressed genes, which is in line with the data in our first part of the study and have shown that the levels of H3K4me2 modifications at a promoter proximal region are well correlated to the expression of genes.4. Effects of h HNF4A-si RNA on CYP3 A expressions in Hep G2 cells To evaluate the efficacy of human HNF4A-small interfering on CYP3A4/A7 m RNA expression levels, Hep G2 cells were infected with HNF4A-small interfering. The expression level of HNF4 A m RNA was clearly attenuated in an h HNF4A-si RNA concentration-dependent manner(20-80 n M), with an approximately 78% decrease compared with control cells at 80 n M. Significant decrease in CYP3A4 and PXR m RNA levels were observed, by 75% and 76% of control cells, respectively, at 80 n M.5. Effects of histone modi?cation status in enhancers and promoters regions on expression and activities of CYP3 A genes Pargyline has an effect on primary fetal liver cells and Hep G2 cells proliferation.The expression levels of CYP3A7, ALB and AFP were markedly enhanced by treatment with 1.2, 2.4m M of pargyline compared with a control in primary fetal liver cells. The levels of CYP3A4 and CYP3A7 was markedly enhanced by treatment with 0.3, 3m M of pargyline compared with control in Hep G2 cells.We wondered whether interactions occurred between HNF4 A and LSD1, GR and LSD1 on the promoter of CYP3A4/3A7. To explore such a possibility, we first demonstrated in Hep G2 cells, using Ch IP assay, occupancy of H3K4me2 on human CYP3A4 promoter(-362 to +53) and enhancer segment(-7836 to-6093) harbored the overlapping HNF4 A binding site compared with a negative control. occupancy of H3K4me2 on human CYP3A7 promoter(-163 to +103) and enhancer segment(-4054 to-3421)(-6265 to-6247) harbored the overlapping GR binding site compared with negative control. The inhibitor of LSD1 causes an increase of H3K4me2 levels at CYP3A4 and CYP3A7 promoters and an upregulation of their expression. The dynamic H3K4me2 marker is important for CYP3A4/A7 gene expressions and that LSD1-mediated demethylation is a potentially important epigenetic mechanism underlying the CYP3 A ontogeny process.Conclusions1. The developmental expression of human hepatic CYP3 A forms has been characterized in the Chinese Han population. CYP3A4 and CYP3A7 are generally served as the major adult and fetal liver forms, respectively and exhibited a developmental switch with a gradual shift towards CYP3A4 expression throughout childhood; and,the associated CYP3A7 change occurs in the ?rst year after birth during liver maturation.2. The epigenetic patterns of H3K4me2 and H3K27me3 modifications are associated with the patterns of ontogenic expression of the CYP3 A genes during liver maturation. HNF4 A is likely a dominant regulator of basal CYP3A4 gene expression in the adult liver, whereas GR seems to be a major regulator of the CYP3A7 gene in the fetal liver. The enriched H3K4me2 in enhancers and promoters regions is important for CYP3A4/3A7 gene expressions; and, HNF4 A and GR, possibly through interaction with H3K4me2 coordinately activate their target genes finally.