Contribution of DNA methylation to the expression of FCGRT in human liver and myocardium

FcRn mediates recycling and transcytosis of IgG and albumin in various cell types. The MHC-class-I-like protein of the FcRn heterodimer is encoded by FCGRT. Few determinants of variable FCGRT expression in humans have been identified so far. In this study, we investigated the presence of DNA methylation in regulatory regions of FCGRT in samples of human liver and myocardium tissue, and we examined the impact of FCGRT methylation on FcRn expression in model cell lines. Quantitative DNA methylation analysis of the FCGRT locus revealed differentially methylated regions in DNA from liver and myocardium. Methylation status in individual CpG sites correlated with FCGRT mRNA expression. Data from model cell lines suggest that differential methylation in the −1058 to −587 bp regulatory region of FCGRT contributes to FcRn expression. Chromatin immunoprecipitation assays indicate that CpG site methylation impacts the binding of the methylation sensitive transcription factors Zbtb7a and Sp1. This study provides a foundation to further define the contribution of epigenetic factors during the control of FcRn expression and IgG traffic in human tissues.


Results
FCGRT methylation in human liver and myocardium. Bioinformatics analysis of a ≈2.0 Kb region encompassing up to 1.4 Kb upstream and 0.6 Kb downstream the start A +1 TG codon of FCGRT revealed three distinct CpG islands (CGI #1 : −1241 to −1086 bp relative to A +1 TG, CGI #2 : −155 to −9 bp, and CGI #3 : +101 to +218 bp) (Fig. 1). Quantitative DNA methylation analysis was performed to examine the extent of methylation in CpG sites located within the "distal" and "proximal" 5′UTR of FCGRT, as well as exon 1 and intron 1, in DNA samples from human liver and myocardium (Supplementary Table S1). The analysis encompassed approximately 82% (53 sites) of the possible CpG sites located within 1.4 Kb upstream the A +1 TG codon (Fig. 1). In samples from both tissues, DNA methylation was most pronounced at CpG sites located within the distal portion of the 5′UTR, while DNA methylation was largely absent within the proximal region (Fig. 1). On average, the extent of DNA methylation in the distal 5′UTR of FCGRT was 58.7 ± 12.9% in liver and 41.2 ± 9.1% in myocardium (Fig. 1). The extent of DNA methylation within the proximal 5′UTR was 4.6 ± 3.9% in liver and 4.5 ± 2.8% in myocardium (Fig. 1). DNA methylation analysis also included a total of 26 CpG sites within a 610 bp stretch downstream the A +1 TG codon (exon 1 and intron 1) (Fig. 1). Within these genic regions the overall extent of CpG methylation was low and similar between both tissues (liver: 8.6 ± 8.3%, myocardium: 8.4 ± 10.5%) (Fig. 1). A total of 11 CpG sites located near or within the 3′UTR (−90 to + 291 bp relative to the stop codon, T +1 GA) were analyzed and showed high levels of methylation (liver: 93.4 ± 4.4%, myocardium: 94.1 ± 5.3%) (Fig. 1).
FCGRT mRNA expression in liver and myocardium. The expression of FCGRT mRNA (transcript NM_004107.4) in samples from liver and myocardium varied by approximately 7-and 14-fold, respectively (Fig. 2a). Correlation analyses between methylation levels and FCGRT mRNA expression were performed across all CpG sites located within the 5′UTR. In liver tissue, there was a significant negative correlation between methylation levels at CpG site −707 bp and FCGRT mRNA expression (Fig. 2b). In myocardium, there were negative correlations between the extent of methylation at CpG sites −1017 bp, −903 bp, and −842 bp and FCGRT mRNA expression (Fig. 2b). The analysis was expanded by using DNA methylation and FGCRT mRNA expression data from 16 liver-derived cell lines available at the Cancer Cell Line Encyclopedia (Supplementary Table S1) (CCLE, https://portals.broadinstitute.org/ccle) 23 . In liver-derived cell lines, there was a negative correlation between the extent of DNA methylation in the −1021 to −628 bp region of FCGRT and FCGRT mRNA expression (R 2 = 0.4201, P = 0.0066) (Fig. 2c).
Binding of transcription factors to the 5′UtR of FCGRT. Bioinformatics analysis identified multiple potential binding sites for transcription factors to the 5′UTR of FCGRT (Fig. 6a). The region with differentially methylated CpG sites (i.e., −1058 to −587 bp) contained potential binding sites for Sp1 transcription factor (Sp1), GA-binding protein alpha chain (GABPα), and zinc finger and BTB domain-containing protein 7A www.nature.com/scientificreports www.nature.com/scientificreports/ (Zbtb7a) (Fig. 6a). These transcription factors were selected for further analysis based on the following criteria: 1) ubiquitous expression according to the human protein atlas database (http://www.proteinatlas.org), 2) potential binding to the FCGRT target region according to MeDReaders (http://medreader.org), and 3) previous evidence of interactions with the FCGRT promoter region [24][25][26][27] . AC16 cardiomyocytes were selected to analyze the role of FCGRT methylation on the binding of candidate transcription factors because the cells: 1) display intermediate level of FCGRT methylation, and 2) are sensitive to Aza treatment. In AC16 cells, Sp1 and Zbtb7a bind to the 5′UTR of FCGRT as suggested by chromatin immunoprecipitation experiments (ChIP), while GABPα showed no apparent binding to the same region (Fig. 6b,c). ChIP assays with primers spanning differentially methylated segments evidenced interactions between Sp1 and the −923 to −825 bp region (30 ± 17-fold enrichment, primer set 2), and Zbtb7a and the −1045 to −825 bp region of FCGRT (23 ± 3-fold enrichment, primer set 1, and 34 ± 22-fold enrichment, primer set 2) (Fig. 6c). No relevant interactions between any of the three transcription factors and the −729 to −592 bp region (primer set 3) were detected (Fig. 6b). No interactions between GABPα, Sp1, or Zbtb7a, and the −1045 to −825 bp region were evident after treating AC16 cells with Aza (5 µM for 72 h) (Fig. 6b,c). Confirmatory dot-blot assays showed that GABPα, Sp1, and Zbtb7a were expressed in AC16 cells under basal conditions and after incubations with Aza ( Supplementary Fig. S2).

Discussion
In this study, we investigated the role of DNA methylation as potential epigenetic regulator of human FCGRT expression. Quantitative DNA methylation analysis of the FCGRT locus revealed differentially methylated regions in DNA samples from liver and myocardium (Fig. 1). Seminal work by Mikulska and Simister showed that the −1529 to −570 bp region of FCGRT is essential for gene promoter activity 26 . This region exhibits variable levels of CpG methylation, and the extent of methylation in certain CpG sites (i.e., CpG sites: −1017, −903, −842, and −707 bp) correlates with FCGRT mRNA expression in liver and myocardial tissues (Fig. 2a,b). In line, global methylation levels in the −1529 to −570 bp region correlates with FCGRT mRNA expression in cell lines derived from hepatocellular carcinomas (Fig. 2c).
In general, increased DNA methylation in gene regulatory regions decreases gene expression 28,29 . HepG2 cells exhibited low levels of FCGRT locus methylation and relatively higher levels of basal FCGRT mRNA and FcRn protein expression in comparison to AC16 and SNU-475 cells (Fig. 3). Treatments with the de-methylating drug Aza decreased the levels of FCGRT locus methylation in AC16 and SNU-475 cells which was paralleled by increases in FCGRT mRNA and FcRn expression (Figs 4 and 5). Thus, data from these model cell lines further suggest that differential methylation in the −1058 to −587 bp regulatory region of FCGRT contributes to FcRn expression in liver and myocardium. Liver and myocardium are tissues composed of different cell types that exhibit variable levels of FcRn expression. For example, Kupffer cells from human liver express relatively www.nature.com/scientificreports www.nature.com/scientificreports/ high levels of FcRn in comparison to sinusoidal endothelial cells 30 . Thus, future studies should aim to elucidate the contribution of differential FCGRT methylation to the dynamics of FcRn expression in individual cell types from relevant tissues. Furthermore, epigenetically active substances such as 5-azacytidine and 5-aza-2′ deoxycytidine are used in the clinic for the therapy of myelodysplastic syndromes (MDS) and acute myeloid leukemia (AML) 31,32 . Since Aza increases the expression of FcRn in model cell lines, it will be of interest to evaluate whether Aza treatment modifies a) the pharmacological behavior (e.g., pharmacokinetics) of Fc-containing monoclonal antibody drugs, and b) albumin tumor consumption in the context of therapy for AML/MDS 33,34 .
Relatively few regulatory elements within the putative promoter region of human FCGRT have been characterized to date. Sachs et al. demonstrated that a polymorphic VNTR in the FCGRT promoter influences FcRn expression 17 . Mikulska showed that Sp1, Sp2, Sp3, c-Fos, c-Jun, YY1, C/EBPβ and C/EBPΔ interact with distinct regions in the FCGRT promoter 26,35 . Our results in AC16 cells suggest that Zbtb7a and Sp1 bind to overlapping methylated regions in the FCGRT promoter (Fig. 6). Zbtb7a and Sp1 interactions with FCGRT were abolished with Aza treatment (Fig. 6). Zbtb7a recognizes methylated CpGs and represses transcription by mechanisms that include prevention of transcription factor binding to DNA and recruitment of corepressors and histone deacetylase (HDAC) complexes [36][37][38] . Sp1 can recognize methylated CpG sites, and Zbtb7a is capable of interacting www.nature.com/scientificreports www.nature.com/scientificreports/ with Sp1 which results in transcriptional repression due to interference of Sp1's DNA binding activity 37,39,40 . Thus, ChIP data may reflect the presence of a Sp1-Zbtb7a complex bound to the −923 to −825 bp region of FCGRT. This region is relatively more sensitive to demethylation by Aza in cells. Interindividual levels of methylation in liver and myocardium are also more variable in this region and this variability may be relevant for the epigenetic control of FCGRT expression. Additional studies are necessary in order to elucidate the role of FCGRT methylation on the binding of Zbtb7a, Sp1, and other transcription factors in various cell types and tissues at different diseases states.
Our results suggest that differential DNA methylation in specific regions of the FCGRT gene promoter contributes to regulate the expression of FcRn. This study provides a foundation to further define the contribution of epigenetic factors during the control of FcRn expression and IgG traffic in human tissues.

Human tissue samples. The Institutional Review Board of the State University of New York at Buffalo
(UB-IRB) approved this research. UB-IRB determined that this research is not research with human subjects. This research meets exempt criteria, 45 CRF 46,101(b)(4). Human liver (n = 10) and myocardial (n = 10) tissue samples were provided by the National Disease Research Interchange (NDRI), the Cooperative Human Tissue Network (CHTN), and the Liver Tissue Procurement and Distribution System (LTPDS, National Institutes of Health Contract N01-DK-9-2310). As per Federal and State regulations, CHTN, NDRI, and LTPDS require every procurement site to obtain informed consent in compliance with all regulations governing that process in writing from any donor of human tissue (or the next of kin thereof) for the use of tissue for research. Each signed consent form is kept on file at the tissue acquisition site and information is never released to any third party. CHTN, NDRI, and LTPDS assign computer generated codes to each donor and do not maintain any information that could be used to identify the donor. CHTN, NDRI, and LTPDS provide anonymous samples coded with unique sample identification numbers. Procurement protocols for this project were reviewed and approved by NDRI, CHTN, and LTPDS.
DNA methylation analysis. Genomic DNA from tissues was extracted using the E.Z.N.A. Tissue DNA kit (Omega Bio-tek). Quantitative DNA methylation analysis of the FCGRT locus was performed with the EpiTYPER Mass Array System (Agena Bioscience) at the Genomics Shared Resource at Roswell Park Comprehensive Cancer Center (Buffalo, New York). Specific primer sets were used for amplification (Table S1). For each amplicon, standards consisting of 0%, 50%, and 100% methylated DNA were used as calibrators. The relative location of individual CpG sites was determined by using the translation start site A +1 TG of exon 1 as the origin. The relative location of CpG sites in the 3′UTR of FCGRT was determined by using the stop codon (T +1 GA) as the origin. www.nature.com/scientificreports www.nature.com/scientificreports/ Quantitative real-time polymerase chain reaction. Total RNA was isolated from cells using Trizol reagent following the manufacturer's instructions (Thermo Fisher). FCGRT mRNA expression was analyzed with specific primers (Table S2). Total RNA (12.5 ng) was reverse transcribed and amplified with the iTaq Universal SYBR Green One-Step Kit (Bio-Rad). FCGRT and reference genes (ACTB and B2M) were amplified in parallel in a CFX96 Touch Real-Time PCR Detection System (Bio-Rad) with the following cycling parameters: 50 °C for 10 min (reverse transcription), 95 °C for 1 min, followed by 44 cycles of 95 °C for 10 s, 60.5 °C for 20 s. Calibration curves were prepared to analyze linearity and PCR efficiency. qRT-PCR data were analyzed using the ΔΔCt method with CFX manager Software (Bio-Rad). The ΔCt method was utilized for determining the relative abundance of FCGRT mRNA.
Fluorescence microscopy. Cells were grown on glass coverslips, fixed in 4% paraformaldehyde in phosphate-buffered saline (PBS) for 20 min at 4 °C and permeabilized with 0.1% Triton X-100 and 200 mM glycine in PBS for 2 min at 4 °C, and washed with PBS. Samples were blocked in 3% bovine serum albumin (BSA)-PBS for 60 min at room temperature and incubated with rabbit anti-FcRn antibody (1/100, ab193148, Abcam) 2 h at room temperature followed by Alexa 546-conjugated anti-rabbit IgG (1/1,000; Invitrogen) for 1 h. Negative controls were performed by replacing the anti-FcRn antibody with rabbit IgG (ab171870, Abcam) at the same concentration during immunostaining. Controls for secondary antibody immunostaining specificity without primary antibody were also included. Nuclei were stained with DAPI. Samples were mounted onto glass slides using FluorSave (Calbiochem). Images (12 bits) were obtained with a monochrome digital camera (Carl Zeiss, Axiocam MRm) attached to a fluorescence microscope (Carl Zeiss, Axiovert 200 M) using PlanApoN 60 × 1.40 NA oil immersion objective. Identical microscope configuration and camera settings were maintained during image acquisition for conditions from the same experiment. Images from multiple fields (~10 fields/condition) were taken.
For FcRn quantification, images were processed using identical parameters with the Fiji software 41 . Images corresponding to different conditions were grouped in image stacks for further processing and comparisons. Background subtraction was identical for each stack, mean fluorescence in the 546 channel (corresponding to FcRn or negative controls) was measured by thresholding with the Huang method in Fiji, and signals from negative controls were then subtracted. Mean DAPI fluorescence signal was measured in the same way.
Bisulfite DNA sequencing. DNA from cells was extracted using the E.Z.N.A. Tissue DNA kit and converted with bisulfite using the MethylEdge Bisulfite Conversion System (Promega). Amplicons corresponding to the FCGRT 5′UTR were amplified using ZymoTaq PreMix (Zymo Research) and specific primers (Table S2). Amplicons were cloned into the pCR4-TOPO vector and transformed into TOP10 chemically competent E. Coli following the manufacturer's instructions (Thermo Fisher Sci.). Individual clones (n = 4-8) for each amplicon were selected, and grown in LB broth supplemented with ampicillin (Sigma-Aldrich). Plasmid DNA was extracted from each clone using the PureYield Plasmid Miniprep System (Promega) and directly sequenced at the Genomics Shared Resource at the Roswell Park Comprehensive Cancer Center (Buffalo, New York).
Chromatin immunoprecipitation. ChIP assays were performed using the ChIP-IT High Sensitivity kit (53040, Active Motif) according to the manufacturer's instructions. Briefly, AC16 cells were cross-linked with 1% formaldehyde at room temperature for 5 min, repeatedly washed with ice-cold PBS, and lysed using a Dounce homogenizer followed by centrifugation. Chromatin was fragmented using enzymatic shearing. A fraction of the mixture of protein-DNA complex was used as "input DNA". Sheared chromatin (1 μg) was then incubated overnight at 4 °C with 5 μg of rabbit anti-GABPα antibody (PA5-27735, Invitrogene), rabbit anti-Sp1 antibody (sc-14027-X, Santa Cruz Biotechnology), rabbit anti-Zbtb7a antibody (A300-548A, Bethyl Laboratories) or normal rabbit IgG (ab171870, Abcam). Immuno-precipitated DNA was eluted using protein G agarose beads, then the cross-linking was reversed and the DNA was purified.
Recovered DNA samples and inputs were analyzed by PCR and qPCR. Three fragments (123, 99, and 138 bp) corresponding to the FCGRT promoter region were amplified with primers listed in Table S2. PCR products were resolved on a 2% agarose gel stained with SYBR Safe (Thermo Fisher Scientific), and visualized under UV light. For qPCR, DNA samples were amplified with the iTaq Universal SYBR Green One-Step Kit (Bio-Rad) in a CFX96 Touch Real-Time PCR Detection System (Bio-Rad) with the following cycling parameters: 95 °C for 1 min, followed by 35 cycles of 95 °C for 30 s, 58 °C for 30 s. The amount of DNA in each sample was extrapolated from standard curves for each PCR reaction, and fold enrichment values for target DNA sequences relative to IgG controls samples were determined.
Densitometric analysis was performed with the Fiji software 41 . A region of interest (ROI) corresponding to the spotted region was manually selected, and the mean gray value was measured. Negative control signal was subtracted from all conditions and normalized to β-actin signal.
Bioinformatics. FCGRT reference sequence NM_004107.4 from the National Center for Biotechnology Information (Gene ID: 2217) was used for analysis. Predictions of CpG islands were performed with MethPrimer