Skip to main content
Download PDF

Altered chromatin landscape and 3D interactions associated with primary constitutional MLH1 epimutations

Clinical Epigenetics volume 16, Article number: 193 (2024) Cite this article

Abstract

Background

Lynch syndrome (LS), characterised by an increased risk for cancer, is mainly caused by germline pathogenic variants affecting a mismatch repair gene (MLH1, MSH2, MSH6, PMS2). Occasionally, LS may be caused by constitutional MLH1 epimutation (CME) characterised by soma-wide methylation of one allele of the MLH1 promoter. Most of these are “primary” epimutations, arising de novo without any apparent underlying cis-genetic cause, and are reversible between generations. We aimed to characterise genetic and gene regulatory changes associated with primary CME to elucidate possible underlying molecular mechanisms.

Methods

Four carriers of a primary CME and three non-methylated relatives carrying the same genetic haplotype were included. Genetic alterations were sought using linked-read WGS in blood DNA. Transcriptome (RNA-seq), chromatin landscape (ATAC-seq, H3K27ac CUT&Tag) and 3D chromatin interactions (UMI-4C) were studied in lymphoblastoid cell lines. The MLH1 promoter SNP (c.-93G > A, rs1800734) was used as a reporter in heterozygotes to assess allele-specific chromatin conformation states.

Results

MLH1 epimutant alleles presented a closed chromatin conformation and decreased levels of H3K27ac, as compared to the unmethylated allele. Moreover, the epimutant MLH1 promoter exhibited differential 3D chromatin contacts, including lost and gained interactions with distal regulatory elements. Of note, rare genetic alterations potentially affecting transcription factor binding sites were found in the promoter-contacting region of CME carriers.

Conclusions

Primary CMEs present allele-specific differential interaction patterns with neighbouring genes and regulatory elements. The role of the identified cis-regulatory regions in the molecular mechanism underlying the origin and maintenance of CME requires further investigation.

Introduction

Lynch syndrome (LS) is characterised by an increased risk of developing several types of cancers, mainly colon and endometrial tumours [1]. LS is mainly caused by germline genetic pathogenic variants in the mismatch repair (MMR) genes MLH1, MSH2, MSH6, and PMS2 [2]. Rarely, constitutional epigenetic defects (constitutional epimutations) in MLH1 and MSH2 may be causative of LS [2].

Constitutional epimutations are associated with soma-wide monoallelic methylation of a promoter CpG island throughout normal tissues [3,4,5,6,7]. Constitutional epimutations are classified as secondary when linked to a genetic alteration in cis, or as primary when there is no apparent genetic cause [3]. Moreover, primary epimutations usually show de novo occurrence and/or erasure in the next generation [8, 9]. While only secondary constitutional epimutations have been reported for MSH2 gene, all caused by germline deletions in the adjacent EPCAM gene [4], both primary and secondary epimutations of MLH1 have been described.

Constitutional MLH1 epimutations (CME) constitute a focal event that specifically and exclusively affects a region of 1.6 Kb (chr3:36,992,300–36,993,908) encompassing the CpG island spanning the bidirectional MLH1 and EPM2AIP1 promoter [10]. Secondary CMEs have been linked to rare point variants occurring within the promoter, exon 1 and intron 1 of MLH1, Alu insertions within the first MLH1 exon, and structural variants affecting the entire MLH1 gene, each resulting in varying levels of constitutional methylation [11,12,13,14,15,16,17]. However, most of the reported CME cases have not been associated with any cis-genetic variant located within the MLH1 differentially methylated region (DMR) and have been classified as putative “primary” CME. Very rarely, CMEs have been associated with non-Mendelian transmission to offspring [8, 9].

Somatic MLH1 promoter hypermethylation is one of the main causes of MMR deficiency in sporadic colorectal cancer (CRC) and endometrial tumours that show loss of MLH1 expression [18,19,20]. In MLH1-methylated CRC cell lines, Deng and colleagues defined a proximal region inside the MLH1 promoter (referred to as C-D) in which methylation correlated with loss of MLH1 transcription. Later studies showed that nucleosome occupancy at this region correlated with MLH1 transcriptional silencing and preceded DNA methylation in RKO CRC cell lines [21].

Spreading of methylation from Alu elements within MLH1 intron 1 towards the promoter has also been suggested as a potential underlying mechanism for MLH1 methylation in CRC cell lines and tumours [22]. In CRC, hypermethylation of MLH1 coexists with expression of the BRAF p.V600E variant [23], which is involved in the onset of the CpG island methylator phenotype (CIMP) through the action of the transcription factor (TF) MAFG [24,25,26]. On the other hand, the MLH1 c.-93G > A promoter variant (rs1800734) has been associated with an increased risk of MLH1-methylated CRC [25, 27, 28] and MLH1-methylated endometrial cancer [29, 30]. It has been proposed that binding of TFAP4 to the wildtype c.-93G allele precludes the binding of MAFG and prevents the recruitment of the DNA methyltransferase DNMT3B to the MLH1 promoter [24, 25].

The DMR in CME carriers is more confined compared to sporadic colorectal cancer [10]. While methylation is restricted to the shared MLH1-EMP2AIP1 CpG island in carriers of a CME, multiple genes flanking MLH1 in a 50 kb region are concomitantly hypermethylated in sporadic MSI, MLH1-hypermethylated CRC, leading to regional transcriptional silencing in these tumours [10, 31]. These differences in the observed methylation patterns suggest distinct mechanisms underlie the establishment of MLH1 methylation in sporadic tumours and in CME carriers. In the present work, we delve into the mechanism underlying primary CME and their epigenetic consequences by analysing the chromatin landscape and 3D interactions of the epimutant and wildtype MLH1 alleles along with genomic and transcriptome characterisation of de novo epimutation cases.

Material and methods

Patients and samples

Four CME carriers previously characterised by our group [10], showing hemiallelic methylation at the MLH1 promoter in blood DNA samples, were selected for this study. Families 1 and 2 showed transmission of the genetic haplotype associated with methylation, but in an unmethylated state in the offspring, indicating intergenerational erasure of the epimutation (Supplementary Fig. 1). These cases were therefore classified as “primary” epimutations. In Family 4, the methylated haplotype was not transmitted, and in Family 3, haplotypes were not studied (Supplementary Fig. 1). Family 3 and 4 were consequently classified as putative primary epimutations (no evidence of methylation-associated haplotype inheritance) (Table 1 and Supplementary Fig. 1).

Table 1 Variants identified within the MLH1 DMR and MLH1 gene
Full size table

All included CME carriers were heterozygous for the MLH1 promoter variant c.-93G > A (population allele frequency 0.232, gnomAD v3.1.2) (Table 1), which was used herein as a reporter to distinguish epimutant from unmethylated alleles in sequence analyses. As previously reported [10], the methylation-associated allele (MAA) was linked to the c.-93A allele in 3 out of 4 CME carriers (CME1-3) and to the c.-93G allele in the remaining one (CME4); also, the CME3 patient carried a small deletion (c.-234_-236del) in trans to the MAA (Table 1). One adult child of CME1 (R1) and two adult children of CME2 (R2.1 and R2.2) who harboured the MAA in a non-methylated state were included as control relatives (Table 1 and Supplementary Fig. 1).

Lymphoblastoid cell lines

Lymphoblastoid cell lines (LCLs) were used as the source material for the RNA-seq, ATAC-seq, CUT&Tag, and UMI-4C experiments. LCLs were established using Epstein Barr Virus as previously described [10]. Immortalised lymphoblastoid cells were grown at 37ºC in RPMI (Gibco #61870–010) supplemented with 10% FBS (Gibco #10270106), 1% Pen/Strep (Gibco #15140122) and 0.25 μg/mL Fungizone (Gibco #15290018).

Linked read library preparation, sequencing, and analyses

Chromium 10X linked-read whole genome sequencing (10X Genomics, Pleasanton, CA, USA) [32] was performed using blood DNA samples from the four CME carriers and R1 and R2.1 control relatives. Sequencing libraries were prepared using the Chromium Library Kit (10 × Genomics, Pleasanton, CA, USA) following manufacturer’s protocol. Library was sequenced on an Illumina NovaSeq 6000 system with 150-by-150-bp paired-end reads. The resulting BCL files were demultiplexed and converted to FASTQ files using Long Ranger (v2.2.2) 'mkfastq'.

Alignment and SNVs/indels inference

Raw sequencing FASTQ output was aligned with BWA-MEM (v0.7.17) to the NCBI Human Reference Genome Build hg38 (hg38). Duplicates were marked using Samblaster (v0.1.24), and BAMs were sorted and indexed using Samtools (v1.9) [33]. Germline variants were called using: (1) GATK Haplotypecaller (v4.1.8.1), genotyped by GenotypeGVCFs and filtered using VariantRecalibrator and ApplyVQSR [34], and (2) Strelka2 (v2.9.10) [35] using the germline configuration. Only “PASS” variants identified by both callers’ algorithms were retained. The intersection between the two datasets was performed using BCFtools (v1.3.1) [36]. Variants were annotated using ANNOVAR (v20191024) (refGene, gnomad30 genome, avsnp150) [37, 38]. Only variants identified with a read depth greater than 5, a mapping quality exceeding 40, and an allele frequency higher than 30% (putative germline) were considered.

Further filtering was performed to screen variants mapping to the DMR (chr3:36,992,300–36,993,908), MLH1 (chr3:36,993,226–37,050,896), LRRFIP2 (chr3:37,052,626–37,183,689) and promoter-contacting region (chr3:36,596,059–37,430,058), which spans 396 kb upstream and 436 kb downstream of the DMR. Variants with a population frequency < 1% in gnomAD v3.0 genome database were considered as rare variants.

Phasing of SNVs and indels

To obtain phase information we used a well-established pipeline. Briefly, Long Ranger (v2.2.2) ‘wgs’ was run to align the reads to the hg38 reference genome. Variants called by GATK (v3.5) [39] using the-vcmode parameter were annotated with phasing information (phased block ID and phased genotype) in the resulting VCF files. All filters-passed variants near the MLH1 gene (chr3:36,000,000–38,000,000) were collected and intersected with our previous final variant list.

Structural variants inference

Structural variants (SVs) were called using Delly (v1.1.6) [40], GRIDSS (v2.11.1–1) [41], Manta (v1.6.0) [42], and Smoove (v0.2.8) [43]. Only PASS variants were included, and ENCODE DAC blacklist was used to remove regions with anomalous, unstructured, and high signal/read counts. SVs were annotated as deletion, duplication, inversion, insertion, and breakends.

To control for population SVs, 2,504 low-coverage BAMs (hg38) and the PED file were downloaded from the 1,000 genomes AWS S3 bucket [44]. We extracted SV alignment evidence (discordant reads and split reads) from BAM control population using excord (v0.2.4) [45] with—discordant distance set to 500. Giggle (v0.6.3) and STIX (v1.0) were used to create an index and a database as described elsewhere [46]. The same methodology was applied to our patient cohort, and the inferred SVs were queried in both databases. SVs with evidence of > 9 counts in 1,000 genomes were defined as population variants and were removed. Coverage plots were generated for the remaining SVs using Samplot (v1.3.0) [47] and manually inspected.

Splicing prediction of identified variants

The potential effects of intronic variants on splicing were evaluated with the in silico tool SpliceAI [48] using max distance 2,000 bp for MLH1 and 10,000 bp for LRRFIP2 variants.

Insertion of transposable elements

MELT programme (v2.2.2) was employed to identify, annotate, and genotype non-reference Mobile Elements Insertions (MEIs), specifically Alu, LINE-1, HERVK and SINE-VNTR-Alu (SVA) elements. Default hg38 transposon files and parameters were used for the analysis. Only Alu insertions successfully passed all filters and were subsequently taken into consideration in the resulting VCF file. Heterozygous Alu insertions in CME carriers that were either absent or homozygous in control samples were considered as possible insertions and rearrangements.

Motif analysis

The package TFBSTools v1.36.0 [49] and JASPAR2022 (v.0.99.7) were used to predict binding sites for transcription factors with a minimum overlap score of 80% to the MLH1 promoter sequence (chr3:36,992,079–36,994,100).

The package motifbreakR v.2.12.3 [50] was used to identify the SNVs and indels that disrupt TF binding based on position probability matrices (PPM). Default settings were applied, using a P value threshold of 1 × 10–4. We selected MotifDb (v.1.40.0) as the chosen TF motifs database [51]. Output motifs from FlyFactor Survey, ScerTF, stamlab, and versions older than HOCOMOCOv10 and JASPAR2022 were not considered. Only variants classified in the output file as “strong” based on the setup parameters were considered.

To identify the motifs potentially bound by transcriptional repressors and activators or insulator proteins, we considered TFs included in the Gene Ontology categories “DNA-binding transcription repressor activity”, “DNA-binding transcription activator activity”, and “chromatin insulator sequence binding” obtained from AmiGO [52, 53]. In this analysis, we did consider binding motifs of TFs with potential repressor activity for which the alternative allele score was higher than the reference allele, as well as TFs with activator activity for which the alternative allele score was smaller than the reference allele.

RNA-seq library preparation and RNA isoform analysis

Total RNA from LCLs was isolated using Trizol (Ambion) and the RNeasy Mini Kit (Qiagen). RNA-seq libraries and 75-bp or 100-bp paired-end sequencing were undertaken at the Genomic Units from Centre Nacional d’Anàlisi Genòmica (CNAG) and Centre for Genomic Regulation (CRG) (Barcelona, Spain).

RNA-seq reads from FASTQ files were mapped to hg38 using Hisat2 (v2.1.0) [54], with the option --dta for downstream transcriptome assembly. Sorted and indexed BAM files were generated with samtools (v1.17) [33]. De-novo isoform discovery and quantification were performed using the StringTie (v2.2.1) [55] transcript assembler. First, StringTie was used to assemble the read alignments obtained in the previous step. A non-redundant set of transcripts from all RNA-seq samples was generated using the --merge option. Then, transcript abundances (TPMs) and read coverage tables were obtained for each of the input transcripts and annotated in a GTF file. SUPPA (v2.3) [56] was used to obtain the alternative splicing events from the GTF file and calculate the percent-splice-in (PSI) values using the transcript abundances per sample. The differential transcript usage between epimutant and control cells for MLH1 and adjacent genes (EPM2AIP2 and LRRFIP2) was calculated with SUPPA diffSPlice by applying a multiple testing correction test (FDR).

ATAC-seq library preparation and data processing

ATAC-seq libraries were prepared as previously described [57]. For each experiment, 50,000 cells were collected and incubated in 300 µl cold lysis buffer (10 mM Tris–HCl pH 7.4, 10 mM NaCl, 3 mM MgCl2, 0.1% Igepal CA-630) for 25 min on ice. Nuclei were centrifuged for 15 min at 500 g at 4°C with low acceleration and brake settings. Then, the pellet was resuspended in 100 µL of the lysis buffer and centrifuged again. After centrifugation, nuclei were resuspended in 25 µL of reaction buffer containing 2 µL of Tn5 transposase, 12.5 µL of TD buffer (Nextera DNA Library Prep Kit, 15,028,212, Illumina), and 10.5 µL of water. Samples were incubated at 37ºC for 1 h. After incubation, reaction was inactivated by adding 5 µL of clean-up buffer (900 mM NaCl, 300 mM EDTA), 2 µL of 5% SDS, 2 µL of Proteinase K (ThermoScientific #EO0491) and incubated for 30 min at 40 °C. DNA was isolated using 2X SPRI beads clean-up (AgencourtAMPureXP, Beckman-Coulter, #A63880) and eluted in 21 µL 10 mM Tris–HCl pH8.

For library preparation, purified DNA was amplified performing two sequential PCRs with 9 cycles each. PCR mix was prepared as: 21 μL DNA, 25 μL NEBNext HiFi 2 × PCR Master mix, 2 µL of universal Ad1 primer and 2 µL of uniquely barcoded Ad2 primer [58]. Library was amplified in a thermocycler using the following programme: 72°C for 5 min, 98ºC for 30 s, 9 cycles of 98°C for 10 s and 63°C for 30 s, and a final extension at 72°C for 1 min and at 4°C hold. First amplification was purified for small fragment selection using 0.6X SPRI beads clean-up, following the manufacturer’s instructions. DNA was eluted in 21 µL 10 mM Tris–HCl pH8. The purified PCR product was amplified again using the same conditions and purified using 1.8X SPRI beads clean-up. Library sequencing was undertaken at the Genomic Units from CRG (Barcelona, Spain).

Reads were mapped to version hg38 of the human genome using Bowtie2 [59](v2.4.1) with default parameters. Next, duplicates and reads mapping to non-canonical chromosomes were removed using Samtools [33] (v1.10). Offset correction was performed using the ATACseqQC R package [60] (v1.22.0). Peak calling was performed using MACS2 [61] (v2.2.7.1) with arguments “--q 0.05 –nomodel --shift -100 --extsize 200”.

CUT&Tag library preparation and data processing

CUT&Tag was performed as previously described [62] with minor modifications. Briefly, cells were harvested, counted, and centrifuged for 3 min at 600 g at room temperature. Cells were washed in Wash Buffer (20 mM HEPES pH 7.5, 150 mM NaCl, 0.5 mM Spermidine, 10 mM Sodium butyrate, 1X Protease inhibitor cocktail) and resuspended to 500,000 cells/mL. Concanavalin A coated (Bangs Laboratories, #BP531) magnetic beads were mixed with 10 volumes of binding buffer (20 mM HEPES pH 7.9, 10 mM KCl, 1 mM CaCl2, 1 mM MnCl2, 10 mM Sodium butyrate) and washed with 1.5 mL of binding buffer. Beads were resuspended in 1 volume of binding buffer and added to the cells. In each experiment, 100,000 cells and 10 μL of beads were used. The unbound supernatant was removed, and bead-bound cells were resuspended in 50 μL ice-cold Antibody buffer (20 mM HEPES pH 7.5, 150 mM NaCl, 0.5 mM Spermidine, 10 mM Sodium butyrate, 0.4 mM EDTA, 0.02% BSA, 0.05% Digitonin, 1X Protease inhibitor cocktail) and transferred to a LoBind tube. Primary antibody against H3K27ac (Abcam, #ab4729) was added (1:100) and incubated overnight on a rotating wheel at 4ºC. Next day, tubes were placed on the magnet stand to clear and the liquid drawn off. Secondary antibody (Antibodies Online, #ABIN101961) was diluted 1:100 in Dig-wash buffer (20 mM HEPES pH 7.5, 150 mM NaCl, 0.5 mM Spermidine, 10 mM Sodium butyrate, 0.05% Digitonin, 1X Protease inhibitor cocktail), added to the beads, and incubated on a rotating wheel at room temperature for 1 h. Tubes were placed on the magnet stand to clear and withdraw the liquid, and beads were washed three times with 1 mL of Dig-wash buffer. A 1:20 dilution of pA-Tn5 adapter complex (Cutana, 15–1017) was prepared in Dig-300 buffer (20 mM HEPES pH 7.5, 300 mM NaCl, 0.5 mM Spermidine, 10 mM Sodium butyrate, 0.01% Digitonin, 1X Protease inhibitor cocktail) and added to the beads. The tubes were mixed by gentle vortexing and incubated on a rotating wheel at room temperature for 1 h. After incubation, beads were washed three times in 1 mL of Dig-300 buffer. Tubes were placed on the magnet stand to draw off the liquid, and beads were resuspended in 300 μL of Tagmentation buffer (Dig-300 buffer, 10 mM MgCl2) and incubated at 37ºC for 1 h. To stop tagmentation and reverse cross-links, 10 µL 0.5 M EDTA, 3 µL 10% SDS and 2.5 µL 20 mg/mL Proteinase K were added to each sample and incubated for 1 h at 55 °C. DNA was purified by phenol–chloroform extraction and dissolved in 28 µL of TE (1 mM Tris–HCl pH 8, 0.1 mM EDTA).

For library amplification, 21 µL of DNA was mixed with 2 µL of a universal i5 and 2 μL of a unique barcoded i7 primer for each sample. A volume of 25 μLNEBNext HiFi 2X PCR Master mix was added to the mix. The following cycling conditions were used for library amplification: 72 °C for 5 min, 98 °C for 30 s, 13 cycles of 98 °C for 10 s and 63 °C for 10 s, and a final extension at 72 °C for 1 min and hold at 8 °C. Post-PCR clean-up was performed by adding 1.3X of Ampure XP beads (AgencourtAMPureXP, Beckman-Coulter, #A63880). The mix was incubated at ambient temperature for 5 min and washed twice gently with 80% ethanol. Samples were eluted in 25 µL 10 mM Tris–HCl pH 8. Library sequencing was undertaken at the Genomic Units from CRG (Barcelona, Spain).

Reads were mapped to hg38 using Bowtie2 [59] (v2.4.1) with parameters “–very-sensitive –no-mixed –no-discordant –phred33 -I 10—× 100”. Next, duplicates and reads mapping to non-canonical chromosomes were removed using Samtools [33] (v1.10). Peak calling was performed using MACS2 [61] (v2.2.7.1) with arguments “–broad –broadcutoff 0.1 –nomodel”.

ATAC-seq, CUT&Tag and RNA-seq differential analysis

For the ATAC-seq and CUT&Tag data, the R package DiffBind [63] (v3.8.3) was used to load peaks called with -log10 p-value > 2 and create consensus peaksets, which were obtained by selecting regions present in more than 2 samples within the same condition (epimutation carrier or control). Then, condition-specific consensus peaks were merged.

RNA-seq reads were aligned to GENCODE version 38 using Salmon [64] (v1.3.0). Results were loaded into R using tximport [65] (v1.26.0), transcript information was summarised into genes, and protein-coding genes were retained for downstream analyses.

Differential analyses of ATAC-seq, CUT&Tag, and RNA-seq data were performed using de DESeq2 R package [66] (v1.38.1) with the design “ ~ group + condition”, where group refers to the family and condition to whether the sample is from an epimutation carrier or a control. Thresholds for defining statistically significant changes were set at adjusted p value < 0.1.

UMI-4C

Library preparation

For each experiment, 4 million cells were collected and crosslinked with 1% paraformaldehyde in PBS for 10 min at room temperature and with gentle mixing. Glycine was added to a final concentration of 125 mM to quench the reaction. After 5 min, cells were washed twice with PBS and frozen. Cell pellet was incubated in 1 mL of cold lysis buffer (50 mM Tris–HCl pH 8, 150 mM NaCl, 5 mM EDTA, 1% Triton X-100, 0.5% NP-40, 1X protease inhibitor cocktail) for 30 min on ice with gentle mixing every 10 min. Following steps were performed as previously described [57] using the following upstream and downstream bait primers 5′-AGTGCCTTCAGCCAATCACC-3′ and 5′- TCAGTGCCTCGTGCTCA-3′. Libraries were sequenced on a NextSeq 500 using 2 × 75 bp reads.

UMI-4C processing and differential analysis

The UMI-4C reads were split based on the genotype of the reporter SNP c.-93G > A which was sequenced from the downstream bait primer. The allele-specific UMI-4C demultiplexed reads were then used as an input to UMI4Cats R package [67] to infer and quantify significant genomic interactions with the viewpoint. MLH1 promoter interaction profiles were thus generated separately for the epimutant (methylated) and the wild-type (unmethylated) alleles using the contacts UMI4C() function. Next, contacts were mapped within a 1 Mb region (chr3:36,493,518–37,493,518) centred on the MLH1 promoter using the makeUMI4C function with smoothing parameter min_win_factor = 0.03. Smoothed viewpoint-specific interaction profile plots were generated using the plotUMI4C function. Allele-specific differential contacts were inferred by Fisher’s exact test using the fisherUMI4C function with filter_low = 30 (FDR adjusted p-value 0.05).

Results

Absence of potentially causal genetic variants at the MLH1 and LRRFIP2 loci

We focus the study on four primary CME carriers (referred to as CME1 to CME4), for whom the presence of rare point variants in cis to the epimutation within the DMR spanning the MLH1 CpG island had been previously ruled out by Sanger sequencing [10] (Table 1). Here, we extended the search for genetic variants across the entire MLH1 gene by using linked-read whole genome sequencing (lrWGS). Three rare heterozygous intronic variants were identified within the MLH1 gene in CME2 and CME3. Two of these variants (c.1409 + 441A in intron 12, c.545 + 841G in intron 6) are in phase with the methylation-associated allele (MAA) (Table 1). None of the identified variants were predicted to alter splicing (Supplementary Table 1). The variant in phase with the MAA in CME2 was shared with its relative R2.1 (Table 1), who inherited the same haplotype as CME2 but in an unmethylated state (Supplementary Fig. 1). Phasing of variants in R2.1 was not possible due to the lack of heterozygous variants within the MLH1 promoter. Two other rare heterozygous variants were identified in the two control relatives but not in the CME carriers (Table 1 and Supplementary Table 1).

MLH1 belongs to a trio of reverse-forward-reverse genes together with its neighbouring genes EPM2AIP1 and LRRFIP2. Variants in LRRFIP2 could potentially influence MLH1 silencing through antisense transcription, as demonstrated for PRDX1 variants and MMACHC methylation [68]. Sixteen rare intronic variants were detected in LRRFIP2, none of them predicted to have an impact on splicing (Supplementary Table 2). In agreement with the absence of candidate splicing variants, LRRFIP2 or MLH1 aberrant transcripts were not identified by RNA-seq (data not shown).

Insertion or rearrangements of Alu sequences were not detected within the DMR nor MLH1 gene in any of the four CME carriers. Beyond these specified regions, we identified 11 heterozygous Alu insertions in CME carriers (Supplementary Table 3). Additionally, we detected 9 candidate structural variants (SVs), none of them within the DMR or MLH1 gene body in the two relatives and in three CME carriers (Supplementary Table 4). Therefore, in agreement with previous reports on primary CMEs, our results suggest that neither SNVs, mobile elements nor SVs were implicated in the mechanisms underlying CME in the four patients included in this study.

The epimutant allele features an inactive chromatin conformation at the MLH1 promoter

In the absence of genetic alterations clearly linked to the CMEs, we next sought to explore whether the epimutant allele features altered regulatory functions related to the allele-specific MLH1 loss of function. Thus, we first assayed chromatin accessibility by ATAC-seq in LCLs from the four CME carriers and the non-carrier relatives R1 and R2.2 (Fig. 1A). Compared to control cells, we observed that epimutant cells presented a reduced number of ATAC-seq normalised read counts mapping to the MLH1 promoter suggesting decreased chromatin accessibility (Fig. 1B). We took advantage of a heterozygous reporter genetic variant (c.-93G > A) located within the MLH1 promoter to determine whether the reduction in accessibility was predominantly associated with one of the two alleles. While in control cells ATAC-seq reads were mapped in similar proportions to the two alleles, in cells from CME carriers 98–100% of the chromatin accessibility signal came from the non-methylation-associated allele (non-MAA) (Fig. 1C). These data imply that in the epimutant cells the MAA promoter is in a closed conformation structure. Consistently, in CME3, the allele-specific accessibility was confirmed with an additional genetic variant (c.-234_-236del) in trans with the MAA (Supplementary Fig. 2A).

Fig. 1
figure 1

Characterisation of MLH1 promoter and functional validation of the epimutations. A UCSC screenshot at chr3:36,991,969–36,994,439 showing the ATAC and H3K27ac enrichment signal at the MLH1 promoter in each sample. Green and purple tracks represent control and epimutant (CME) cells, respectively. Below the peak profiles, the consensus peak region is indicated in grey. Normalised counts at the MLH1 promoter for ATAC (B) and H3K27ac (D) peaks. Wilcoxon p value = 0.13. Proportion of ATAC (C) and H3K27ac (E) reads harbouring each allele at MLH1 c.-93 (chr3:36,993,455). F Allelic expression of two common MLH1 (left) and EPM2AIP1 (right) SNPs in the heterozygous samples

Full size image

Next, we used CUT&Tag to map the deposition of H3K27ac, a histone modification typically associated with transcriptionally active chromatin. Consistent with ATAC-seq, we observed a reduction in the H3K27ac reads mapping to the MLH1 promoter in epimutant cells as compared to controls (Fig. 1D). Again, after splitting the reads based on the heterozygous reporter genetic variant at c.-93, we observed discordant proportions of allele-specific H3K27ac enrichments in the cells of controls versus CME carriers and on the non-MAAs (Fig. 1E, Supplementary Fig. 2B). Overall, these data indicate that the MAA MLH1 promoter lacks both accessibility and active chromatin marks exclusively in epimutant cells. Importantly, in the cells of the non-methylated controls, the same allele exhibited an active chromatin configuration.

We applied RNA-seq to gauge the impact of DNA methylation and inactive chromatin conformation at the MLH1 promoter on gene expression activity in LCLs. We assessed the allele-specific expression of two common MLH1 and EPM2AIP1 SNPs, c.655G > A (rs1799977) and c.*2570 T > G (rs9311149), respectively. Most samples were heterozygous for at least one of these SNPs; however, CME3 and R1 were homozygous for both SNPs hence excluded from this analysis. As expected, while the control cells (R2.2) expressed both alleles at similar levels, the epimutant cells exhibited monoallelic expression (Fig. 1F). Our findings were consistent with the monoallelic expression previously observed in CME1 and CME2 primary lymphocytes [10].

Differences in chromatin accessibility and activity between control and CME cells were also analysed at the genome-wide level. Our principal component analyses showed an overall high similarity between the chromatin landscapes of the different cell lines (Supplementary Fig. 3A, B), suggesting that cancer predisposition in CME is linked to focal events rather than genome-wide chromatin remodelling. Of note, significant differences were observed in chromatin accessibility for three regions (Supplementary Fig. 3C, Supplementary Table 5) and in H3K27ac occupancy for four regions (Supplementary Fig. 3D, Supplementary Table 6), between the CMEs and controls, although none of the associated genes was found to be differentially expressed (Supplementary Table 7). We found only 52 differentially expressed genes in epimutant cells compared to control cells (Supplementary Fig. 3F, Supplementary Table 7). Among them, we identified 3 TFs that were upregulated (BHLHA15, NR2F2, PMEPA1) and 3 downregulated (EFNA5, RSC1A1, SGMS2) in epimutant cells compared to the controls. We scanned the MLH1 promoter and found that 2 of the upregulated TFs had binding sites within the MLH1 promoter, namely NR2F2 and BHLHA15, both with the potential to act as repressor TFs [69, 70].

Wild type and epimutant alleles showed allele-specific promoter interactions

We next explored whether the differences in chromatin accessibility and activity between alleles were linked to allele-specific differences in the 3D chromatin interactions. To address this question, we developed an allele-specific UMI-4C assay taking advantage of the c.-93 reporter variant. Importantly, we chose to use the UMI-4C technique for this analysis to achieve a quantitative comparative assessment of the chromatin contacts, which is not guaranteed with other chromatin capture techniques such as regular 4C that are prone to PCR amplification biases. By careful comparison and strict statistical analyses of allele-specific UMI-4C data, we did not observe significant differences in the interactions profile of the two (unmethylated) c.-93 alleles in control cells (Supplementary Fig. 4). Strikingly, we observed a different 3D chromatin contact profile of the MAA when comparing control versus epimutant cells. In contrast, the non-MAA showed non-statistically significant differences when comparing epimutant and control cells (Fig. 2). The eight regions displaying statistically significant changes in MAA included gained and lost interactions (Fig. 2, right). The differential contacts and the nearby regions were marked by the presence of ATAC and H3K27ac peaks (Supplementary Fig. 5A, B). In fact, a permutation test showed that the differential contacts overlapped with accessible regions more than expected by chance (Supplementary Fig. 5A), indicating that the differential contacts likely harbour distal regulatory elements implicated in the regulation of MLH1 gene expression.

Fig. 2
figure 2

Contacts of the MLH1 promoter. Profiles of MLH1 promoter contacts in control (CTRL, green) and epimutant (CME, purple) cell lines for the non-MAA and the MAA. The viewpoint is indicated by the black triangle. Protein coding gene annotation and transcriptional direction are shown at the top. Below, the UMI-4C contacts normalised trends and its corresponding windows, with fill representing the log10 odds ratio of the contact differences in epimutant versus control cells. Asterisks on top of the windows indicate significant differences in the 3D contacts (FDR adjusted p-value < 0.05). The domainogram at the bottom shows the mean contact intensity log2 fold changes with increasing resolution between the two groups. Colour scale is the same in all graphs: green and purple indicate lost and gained contacts, respectively, in epimutant cells compared to controls

Full size image

Genetic characterisation of MLH1 promoter contacts identified variants predicted to alter transcription factor binding sites

Regulatory elements harbour binding sites for TFs. Genetic variants mapping to these regions can thus interfere with TF binding, resulting in modulation of gene expression [71], and can influence the local epigenetic state. We hypothesised that some regions might be more prone to accumulate variants than others. To explore whether genetic variants could potentially influence MLH1 silencing, we searched for rare variants mapping to regulatory regions identified by UMI-4C to be in physical contact with the MLH1 promoter. By using a sliding window approach, we uncovered an upstream differential contact exhibiting a high frequency of rare variant accumulation in CME carriers (Supplementary Fig. 5B). Specifically, a total of 99 distinct variants were found in the CME carriers, of which 34 were in phase with the MAA (Supplementary Fig. 5C). Interestingly, 31 of these 34 in-phase variants were predicted to alter transcription factor binding sites (Supplementary Fig. 5C, Supplementary Table 8). Eleven of these variants (11/31, 35.5%) came from a specific sample (CME3), which overall accounts for a higher number of rare variants in the promoter-contacting regions compared to the other carriers (Supplementary Fig. 5D). We did not find variants with a strong effect on insulator protein binding sites in any of the CME carriers. In contrast, we identified 8 rare variants whose alternative genotype was predicted to facilitate the binding of repressor TFs by enhancing similarity to their binding motifs, and 14 rare variants predicted to decrease binding affinity of an activator TF, although none of them were located within the promoter differential contacts (Supplementary Fig. 5C, Supplementary Table 8).

None of the SVs and Alu insertions found in our previous analyses were located within the MLH1 promoter-contacting region.

Overall, in cis genetic alterations potentially affecting TF binding sites at MLH1 promoter contacts that may have a role in predisposition to MLH1 silencing were identified in the four CME carriers included in this study.

Discussion

Here, we have characterised the chromatin landscape associated with the CME in four selected patients who carried a “primary” CME. For the first time, we have shown that, compared to the MLH1 promoter unmethylated allele, the constitutional MAA is less accessible and active. These observations are consistent with monoallelic loss of expression of MLH1 and EPM2AIP1, both genes regulated by the same promoter. Moreover, we have demonstrated that the promoter MAA exhibits differential 3D contacts compared to the non-MAA, including loss and gained interactions with distal regions. Finally, rare in cis genetic variants affecting the binding of repressors and activators were identified within the promoter-contacting region in the four CME carriers, which may have a role on predisposing the allele to MLH1 methylation (Fig. 3).

Fig. 3
figure 3

Schematic representation of the changes in 3D contacts and expression associated with MLH1 promoter methylation. In control cells (derived from CME’s relatives), MLH1 promoter is unmethylated and both alleles show the same pattern of interactions regardless of the c.-93 genotype, with concomitant biallelic expression of MLH1. In cells carrying the epimutation (epimutant cells) the MAA allele is methylated independently of the c.-93 genotype. MLH1 methylation and loss of transcription are coupled to changes in the 3D promoter interactions. Regulatory elements in physical contact with the MLH1 promoter can harbour genetic variants in cis with the MAA that may alter binding of TFs and could predispose to MLH1 epimutation

Full size image

Typically, CME carriers have been screened for in cis variants within the MLH1 promoter region. The four CME carriers studied here were previously screened for the presence of variants within the DMR [10, 72]. In the present work, we excluded the presence of shared rare point genetic variants in the entire MLH1 and LRRFIP2 genes. Instead, a limited number of rare intronic variants not predicted to impact splicing were identified in each case. Consistent with this, RNA-seq analysis also did not identify any aberrant transcripts. Similarly, insertions or rearrangements of Alu sequences or SVs, previously found in index cases with a secondary CME [4, 14, 73, 74], were also ruled out. Of note, structural rearrangements can modify chromatin topology and are associated with aberrant methylation of CpG islands [75,76,77,78]. Neither mobile element insertions or rearrangements nor SVs were detected within the DMR, across the MLH1 gene, or the distal regions in physical contact with the MLH1 promoter in the four CME carriers. Overall, the lack of any findings of singular genetic abnormalities in cis the epimutation in these four CME carriers is in agreement with the classification of “primary” CME.

We have profiled changes in chromatin landscape at the genome-wide level and identified only a few regions exhibiting differential accessibility or H3K27ac enrichment in CMEs as compared to control cells. This observation is consistent with the minimal transcriptional differences found in the cells from the CME and control groups and the focal nature of the CME. Of note, the MLH1 promoter contains binding motifs for two TFs, NR2F2 and BHLHA15, that we identified as differentially expressed between the CME and control cells. NR2F2 can either act as a transcriptional activator or repressor in a gene-specific manner to regulate developmental processes, such as angiogenesis, adipogenesis and neural differentiation [69]. On the other hand, BHLHA15 encodes for a transcription factor whose mouse orthologue Mist1 is expressed in post-implantation E10.5 embryos [70]. Despite the potential for both TFs to act as transcriptional repressors, there is no reported evidence of association with DNA methylation.

The overall small number of transcriptional and chromatin changes was expected, as the LCLs used in this study were derived from healthy lymphocytes obtained from CME carriers whose methylation pattern only differs from their control counterpart at the MLH1 locus [10]. The small differences observed between our CME and control cell lines may be caused by cell diversity and clonal evolution of LCLs through culturing, as previously reported [79, 80]. Nevertheless, differential analysis of chromatin accessibility, H3K27ac enrichment or gene expression performed on distinct cell groups may fail to capture allele-specific features. Allele-specific differences were evident when we could take advantage of informative heterozygous variants within the MLH1 promoter (c.-93G > A and c.-234_-236del) that uncovered differential chromatin accessibility in the ATAC and H3K27ac reads, and monoallelic expression of exonic MLH1 (c.655) and EPM2AIP1 (c.*2570) SNPs in the RNA-seq data.

Using our own UMI-4C approach, we could perform a quantitative allele-specific analysis and identify changes in the 3D contacts between the MAA and non-MAA in CME LCLs. These changes are mainly linked to the promoter methylation state, as both alleles in the control cell lines exhibited the same pattern of 3D contacts. A previous study suggested that the c.-93A variant may induce increased contacts between the MLH1 promoter and the DCLK3 gene resulting in its enhanced expression in CRC cell lines [81]. Based on our findings in LCLs, the observed changes in 3D contacts cannot be attributed to the c.-93A variant itself, suggesting the MLH1 promoter-DCLK3 contact may be restricted to CRC cell lines due to the binding of colon-specific factors not present in patient-derived LCLs.

The differential contacts identified when the two alleles (MAA and non-MAA) were compared likely contain regulatory elements that modulate MLH1 expression. Of note, two of the eight differential contacts identified in our study include previously reported putative regulatory regions that correlate with MLH1 expression across distinct cell types: a positive correlation in one and a negative correlation in the other [82]. Interestingly, the region that negatively correlates with MLH1 expression was found to have gained contacts with the MAA in epimutant cells compared to controls. Conversely, the region that positively correlated with MLH1 expression showed a loss of physical contacts on the MAA compared to the unmethylated allele.

Variants in distal regulatory elements influence chromatin contacts and gene expression [71, 83], and correlations between genetic variants and DNA methylation states exist, with 10–45% of the methylome being influenced by nearby genetic variants [84]. These variants tend to accumulate in non-genic regions and enhancers, while being depleted in CpG islands, 5’UTRs and regions upstream of transcription start site. Considering this, we hypothesise that variants located within the differential contacts might predispose to MLH1 methylation (Fig. 3). Although we did not identify any in cis genetic variants predicted to alter binding of activators or to enhance binding of repressor factors inside the differential contacts, we did detect in cis variants affecting motifs in all CME carriers across the promoter-contacting region. It is important to note that we could not determine the phase for all the detected variants within the promoter-contacting region, particularly in CME2, for which the phase block including the c.-93G > A reporter SNP did not cover the entire promoter-contacting region. Additionally, we only used four primary CMEs and two control relatives to determine the differential contacts, which limits the statistical power of the analysis. Our study represents the first comprehensive genetic and epigenetic (ATAC-seq, CUT&Tag, UMI-4C) characterisation of primary CME carriers using peripheral blood lymphocytes and derived LCLs. Furthermore, we have performed allele-specific analyses to investigate the alterations directly coupled with CME. The fact that the MLH1 promoter is the only differentially methylated region in “primary” CME cases indicates a focal defect, for which one hypothesis is that the existence of cis elements makes it prone to become methylated under certain environmental conditions.

The main limitation of our study lies in the use of LCLs. Primary epimutations likely arise at the preimplantation phase during embryo development or at the establishment of the DNA methylation pattern phase during germ cell maturation [8], stages in which the cis-regulatory landscape at the MLH1 locus may differ from that observed in LCLs. Future studies should consider using induced pluripotent stem cells that have undergone generalised DNA demethylation to capture the contribution of regulatory genetic variants to the establishment of a transcriptional silencing conformation of the MLH1 promoter. Regarding technical limitations, we used short-read sequencing data to infer SVs and transposable element insertions. However, long-read sequencing methods have shown greater potential to identify these sorts of variants [89, 90]. Applying long-read sequencing could result in the identification of additional genetic alterations affecting in-cis MLH1 regulatory regions.

In conclusion, by using a very comprehensive (epi)genetic approach, we have identified putative regulatory regions that might influence the epigenetic state of the MLH1 promoter. Also, we have explored genetic variants within the promoter-contacting regions which might confer susceptibility to the epimutation. As a result, our study provides valuable cis-regulatory maps that will facilitate the discovery of genetic variants implicated in MLH1 silencing. Further mechanistic studies, such as in pluripotent stem cells, are needed to elucidate the functional impact of the identified cis-regulatory regions in the origin and/or maintenance of CME.

Availability of data and materials

ATAC-seq, H3K27ac CUT&Tag, UMI-4C, RNA-seq and WGS data generated in this publication are deposited at the European Genome-phenome Archive at the European Bioinformatics Institute with the IDs EGAD50000000714, EGAD50000000711, EGAD50000000713, EGAD50000000712, and EGAD50000000710, respectively.

References

  1. Idos G, Valle L. Lynch syndrome. Seattle: University of Washington; 1993.

    Google Scholar 

  2. Lynch HT, Snyder CL, Shaw TG, Heinen CD, Hitchins MP. Milestones of Lynch syndrome: 1895–2015. Nat Rev Cancer. 2015;15:181–94.

    Article  PubMed  Google Scholar 

  3. Hitchins MP. The role of epigenetics in Lynch syndrome. Fam Cancer. 2013;12:189–205.

    Article  PubMed  Google Scholar 

  4. Ligtenberg MJL, Kuiper RP, Chan TL, Goossens M, Hebeda KM, Voorendt M, et al. Heritable somatic methylation and inactivation of MSH2 in families with Lynch syndrome due to deletion of the 3’ exons of TACSTD1. Nat Genet. 2009;41:112–7.

    Article  PubMed  Google Scholar 

  5. Kuiper RP, Vissers LELM, Venkatachalam R, Bodmer D, Hoenselaar E, Goossens M, et al. Recurrence and variability of germline EPCAM deletions in Lynch syndrome. Hum Mutat. 2011;32:407–14.

    Article  PubMed  Google Scholar 

  6. Hesson LB, Hitchins MP, Ward RL. Epimutations and cancer predisposition: importance and mechanisms. Curr Opin Genet Dev. 2010;20:290–8.

    Article  PubMed  Google Scholar 

  7. Hitchins MP, Ward RL. Constitutional (germline) MLH1 epimutation as an aetiological mechanism for hereditary non-polyposis colorectal cancer. J Med Genet. 2009;46:793–802.

    Article  PubMed  Google Scholar 

  8. Hitchins MP. Constitutional epimutation as a mechanism for cancer causality and heritability? Nat Rev Cancer. 2015;15:625–34.

    Article  PubMed  Google Scholar 

  9. Hitchins MP. Finding the needle in a haystack: identification of cases of Lynch syndrome with MLH1 epimutation. Fam Cancer. 2016;15:413–22.

    Article  PubMed  Google Scholar 

  10. Dámaso E, Castillejo A, del Arias M, Canet-Hermida J, Navarro M, Del Valle J, et al. Primary constitutional MLH1 epimutations: a focal epigenetic event. Br J Cancer. 2018;119:978–87.

    Article  PubMed  PubMed Central  Google Scholar 

  11. Cini G, Carnevali I, Quaia M, Chiaravalli AM, Sala P, Giacomini E, et al. Concomitant mutation and epimutation of the MLH1 gene in a Lynch syndrome family. Carcinogenesis. 2015;36:452–8.

    Article  PubMed  Google Scholar 

  12. Gylling A, Ridanpää M, Vierimaa O, Aittomäki K, Avela K, Kääriäinen H, et al. Large genomic rearrangements and germline epimutations in Lynch syndrome. Int J Cancer. 2009;124:2333–40.

    Article  PubMed  Google Scholar 

  13. Kwok CT, Ward RL, Hawkins NJ, Hitchins MP. Detection of allelic imbalance in MLH1 expression by pyrosequencing serves as a tool for the identification of germline defects in Lynch syndrome. Fam Cancer. 2010;9:345–56.

    Article  PubMed  Google Scholar 

  14. Leclerc J, Flament C, Lovecchio T, Delattre L, Ait Yahya E, Baert-Desurmont S, et al. Diversity of genetic events associated with MLH1 promoter methylation in Lynch syndrome families with heritable constitutional epimutation. Genet Med. 2018;20:1589–99.

    Article  PubMed  Google Scholar 

  15. Morak M, Schackert HK, Rahner N, Betz B, Ebert M, Walldorf C, et al. Further evidence for heritability of an epimutation in one of 12 cases with MLH1 promoter methylation in blood cells clinically displaying HNPCC. Eur J Hum Genet. 2008;16:804–11.

    Article  PubMed  Google Scholar 

  16. Pinto D, Pinto C, Guerra J, Pinheiro M, Santos R, Vedeld HM, et al. Contribution of MLH1 constitutional methylation for Lynch syndrome diagnosis in patients with tumor MLH1 downregulation. Cancer Med. 2018;7:433–44.

    Article  PubMed  PubMed Central  Google Scholar 

  17. Ward RL, Dobbins T, Lindor NM, Rapkins RW, Hitchins MP. Identification of constitutional MLH1 epimutations and promoter variants in colorectal cancer patients from the colon cancer family registry. Genet Med. 2013;15:25–35.

    Article  PubMed  Google Scholar 

  18. Goodfellow PJ, Buttin BM, Herzog TJ, Rader JS, Gibb RK, Swisher E, et al. Prevalence of defective DNA mismatch repair and MSH6 mutation in an unselected series of endometrial cancers. Proc Natl Acad Sci U S A. 2003;100:5908–13.

    Article  PubMed  PubMed Central  Google Scholar 

  19. Inamura K. Colorectal cancers: an update on their molecular pathology. Cancers (Basel). 2018;10:26.

    Article  PubMed  Google Scholar 

  20. Pasanen A, Loukovaara M, Bützow R. Clinicopathological significance of deficient DNA mismatch repair and MLH1 promoter methylation in endometrioid endometrial carcinoma. Mod Pathol. 2020;33:1443–52.

    Article  PubMed  Google Scholar 

  21. Hesson LB, Patil V, Sloane MA, Nunez AC, Liu J, Pimanda JE, et al. Reassembly of nucleosomes at the MLH1 promoter initiates resilencing following decitabine exposure. PLoS Genet. 2013;9:e1003636.

    Article  PubMed  PubMed Central  Google Scholar 

  22. Wang X, Fan J, Liu D, Fu S, Ingvarsson S, Chen H. Spreading of Alu methylation to the promoter of the MLH1 gene in gastrointestinal cancer. PLoS ONE. 2011;6:e25913.

    Article  PubMed  PubMed Central  Google Scholar 

  23. Deng G, Bell I, Crawley S, Gum J, Terdiman JP, Allen BA, et al. BRAF mutation is frequently present in sporadic colorectal cancer with methylated hMLH1, but not in hereditary nonpolyposis colorectal cancer. Clin Cancer Res. 2004;10(1 Pt 1):191–5.

    Article  PubMed  Google Scholar 

  24. Fang M, Ou J, Hutchinson L, Green MR. The BRAF oncoprotein functions through the transcriptional repressor MAFG to mediate the CpG island methylator phenotype. Mol Cell. 2014;55:904–15.

    Article  PubMed  PubMed Central  Google Scholar 

  25. Thomas R, Trapani D, Goodyer-Sait L, Tomkova M, Fernandez-Rozadilla C, Sahnane N, et al. The polymorphic variant rs1800734 influences methylation acquisition and allele-specific TFAP4 binding in the MLH1 promoter leading to differential mRNA expression. Sci Rep. 2019;9:13463.

    Article  PubMed  PubMed Central  Google Scholar 

  26. Weisenberger DJ, Siegmund KD, Campan M, Young J, Long TI, Faasse MA, et al. CpG island methylator phenotype underlies sporadic microsatellite instability and is tightly associated with BRAF mutation in colorectal cancer. Nat Genet. 2006;38:787–93.

    Article  PubMed  Google Scholar 

  27. Miyakura Y, Tahara M, Lefor AT, Yasuda Y, Sugano K. Haplotype defined by the MLH1-93G/A polymorphism is associated with MLH1 promoter hypermethylation in sporadic colorectal cancers. BMC Res Notes. 2014;7:835.

    Article  PubMed  PubMed Central  Google Scholar 

  28. Mrkonjic M, Roslin NM, Greenwood CM, Raptis S, Pollett A, Laird PW, et al. Specific variants in the MLH1 gene region may drive DNA methylation, loss of protein expression, and MSI-H colorectal cancer. PLoS ONE. 2010;5:1–10.

    Article  Google Scholar 

  29. Chen H, Taylor NP, Sotamaa KM, Mutch DG, Powell MA, Schmidt AP, et al. Evidence for heritable predisposition to epigenetic silencing of MLH1. Int J Cancer. 2007;120:1684–8.

    Article  PubMed  Google Scholar 

  30. Beiner ME, Rosen B, Fyles A, Harley I, Pal T, Siminovitch K, et al. Endometrial cancer risk is associated with variants of the mismatch repair genes MLH1 and MSH2. Cancer Epidemiol Biomarkers Prev. 2006;15:1636–40.

    Article  PubMed  Google Scholar 

  31. Hitchins MP, Lin VA, Buckle A, Cheong K, Halani N, Ku S, et al. Epigenetic inactivation of a cluster of genes flanking MLH1 in microsatellite-unstable colorectal cancer. Cancer Res. 2007;67:9107–16.

    Article  PubMed  Google Scholar 

  32. Zheng GXY, Lau BT, Schnall-Levin M, Jarosz M, Bell JM, Hindson CM, et al. Haplotyping germline and cancer genomes with high-throughput linked-read sequencing. Nat Biotechnol. 2016;34:303–11.

    Article  PubMed  PubMed Central  Google Scholar 

  33. Li H, Handsaker B, Wysoker A, Fennell T, Ruan J, Homer N, et al. The sequence alignment/map format and SAMtools. Bioinformatics. 2009;25:2078–9.

    Article  PubMed  PubMed Central  Google Scholar 

  34. McKenna A, Hanna M, Banks E, Sivachenko A, Cibulskis K, Kernytsky A, et al. The genome analysis toolkit: a mapreduce framework for analyzing next-generation DNA sequencing data. Genome Res. 2010;20:1297–303.

    Article  PubMed  PubMed Central  Google Scholar 

  35. Kim S, Scheffler K, Halpern AL, Bekritsky MA, Noh E, Källberg M, et al. Strelka2: fast and accurate calling of germline and somatic variants. Nat Methods. 2018;15:591–4.

    Article  PubMed  Google Scholar 

  36. Danecek P, Bonfield JK, Liddle J, Marshall J, Ohan V, Pollard MO, et al. Twelve years of SAMtools and BCFtools. Gigascience. 2021;10:1–4.

    Article  Google Scholar 

  37. Wang K, Li M, Hakonarson H. ANNOVAR: functional annotation of genetic variants from high-throughput sequencing data. Nucl Acids Res. 2010;38:e164.

    Article  PubMed  PubMed Central  Google Scholar 

  38. Yang H, Wang K. Genomic variant annotation and prioritization with ANNOVAR and wANNOVAR. Nat Protoc. 2015;10:1556–66.

    Article  PubMed  PubMed Central  Google Scholar 

  39. Van der Auwera GA, Carneiro MO, Hartl C, Poplin R, del Angel G, Levy-Moonshine A, et al. From FastQ data to high-confidence variant calls: the genome analysis toolkit best practices pipeline. Curr Protoc Bioinf. 2013;43:11.10.1-11.10.33.

    Google Scholar 

  40. Rausch T, Zichner T, Schlattl A, Stütz AM, Benes V, Korbel JO. DELLY: structural variant discovery by integrated paired-end and split-read analysis. Bioinformatics. 2012;28:333–9.

    Article  Google Scholar 

  41. Cameron DL, Baber J, Shale C, Valle-Inclan JE, Besselink N, van Hoeck A, et al. GRIDSS2: comprehensive characterisation of somatic structural variation using single breakend variants and structural variant phasing. Genome Biol. 2021;22:1–25.

    Article  Google Scholar 

  42. Chen X, Schulz-Trieglaff O, Shaw R, Barnes B, Schlesinger F, Källberg M, et al. Manta: rapid detection of structural variants and indels for germline and cancer sequencing applications. Bioinformatics. 2016;32:1220–2.

    Article  PubMed  Google Scholar 

  43. Pedersen B, Layer R, Quinlan A. smoove: structural-variant calling and genotyping with existing tools (version 0.2.8). [Computer software]. 2020. https://github.com/brentp/smoove.

  44. 1000 Genomes Phase 3 Reanalysis with DRAGEN 3.5 - Data Lakehouse Ready was accessed on April 2023. s3://1000genomes/phase3/data/.

  45. Layer R. excord version 0.2.4. [Computer software]. https://github.com/brentp/excord?tab=readme-ov-file.

  46. Chowdhury M, Pedersen BS, Sedlazeck FJ, Quinlan AR, Layer RM. Searching thousands of genomes to classify somatic and novel structural variants using STIX. Nat Methods. 2022;19:445–8.

    Article  PubMed  PubMed Central  Google Scholar 

  47. Belyeu JR, Chowdhury M, Brown J, Pedersen BS, Cormier MJ, Quinlan AR, et al. Samplot: a platform for structural variant visual validation and automated filtering. Genome Biol. 2021;22:1–13.

    Article  Google Scholar 

  48. Jaganathan K, Kyriazopoulou Panagiotopoulou S, McRae JF, Darbandi SF, Knowles D, Li YI, et al. Predicting splicing from primary sequence with deep learning. Cell. 2019;176:535-548.e24.

    Article  PubMed  Google Scholar 

  49. Tan G, Lenhard B. TFBSTools: An R/bioconductor package for transcription factor binding site analysis. Bioinformatics. 2016;32:1555–6.

    Article  PubMed  PubMed Central  Google Scholar 

  50. Coetzee SG, Coetzee GA, Hazelett DJ. MotifbreakR: an R/Bioconductor package for predicting variant effects at transcription factor binding sites. Bioinformatics. 2015;31:3847–9.

    Article  PubMed  PubMed Central  Google Scholar 

  51. Shannon P, Richards M. MotifDb: An Annotated Collection of Protein-DNA Binding Sequence Motifs. [R package version 1.40.0]. 2022.

  52. Ashburner M, Ball CA, Blake JA, Botstein D, Butler H, Cherry JM, et al. Gene ontology: tool for the unification of biology. Nat Genetics. 2000;25:25–9.

    Article  PubMed  Google Scholar 

  53. Carbon S, Douglass E, Good BM, Unni DR, Harris NL, Mungall CJ, et al. The gene ontology resource: enriching a GOld mine. Nucl Acids Res. 2021;49:D325–34.

    Article  Google Scholar 

  54. Kim D, Langmead B, Salzberg SL. HISAT: a fast spliced aligner with low memory requirements. Nat Methods. 2015;12:357–60.

    Article  PubMed  PubMed Central  Google Scholar 

  55. Pertea M, Kim D, Pertea GM, Leek JT, Salzberg SL. Transcript-level expression analysis of RNA-seq experiments with HISAT StringTie and Ballgown. Nat Protoc. 2016;11:1650–67.

    Article  PubMed  PubMed Central  Google Scholar 

  56. Trincado JL, Entizne JC, Hysenaj G, Singh B, Skalic M, Elliott DJ, et al. SUPPA2: Fast, accurate, and uncertainty-aware differential splicing analysis across multiple conditions. Genome Biol. 2018;19:1–11.

    Article  Google Scholar 

  57. Ramos-Rodríguez M, Raurell-Vila H, Colli ML, Alvelos MI, Subirana-Granés M, Juan-Mateu J, et al. The impact of proinflammatory cytokines on the β-cell regulatory landscape provides insights into the genetics of type 1 diabetes. Nat Genet. 2019;51:1588–95.

    Article  PubMed  PubMed Central  Google Scholar 

  58. Buenrostro JD, Giresi PG, Zaba LC, Chang HY, Greenleaf WJ. Transposition of native chromatin for fast and sensitive epigenomic profiling of open chromatin, DNA-binding proteins and nucleosome position. Nat Methods. 2013;10:1213–8.

    Article  PubMed  PubMed Central  Google Scholar 

  59. Langmead B, Salzberg SL. Fast gapped-read alignment with Bowtie 2. Nat Methods. 2012;9:357–9.

    Article  PubMed  PubMed Central  Google Scholar 

  60. Ou J, Liu H, Yu J, Kelliher MA, Castilla LH, Lawson ND, et al. ATACseqQC: a Bioconductor package for post-alignment quality assessment of ATAC-seq data. BMC Genomics. 2018;19:169.

    Article  PubMed  PubMed Central  Google Scholar 

  61. Zhang Y, Liu T, Meyer CA, Eeckhoute J, Johnson DS, Bernstein BE, et al. Model-based analysis of ChIP-Seq (MACS). Genome Biol. 2008;9:R137.

    Article  PubMed  PubMed Central  Google Scholar 

  62. Kaya-Okur HS, Wu SJ, Codomo CA, Pledger ES, Bryson TD, Henikoff JG, et al. CUT&Tag for efficient epigenomic profiling of small samples and single cells. Nat Commun. 2019;10:1–10.

    Article  Google Scholar 

  63. Ross-Innes CS, Stark R, Teschendorff AE, Holmes KA, Ali HR, Dunning MJ, et al. Differential oestrogen receptor binding is associated with clinical outcome in breast cancer. Nature. 2012;481:389–93.

    Article  PubMed  PubMed Central  Google Scholar 

  64. Patro R, Duggal G, Love MI, Irizarry RA, Kingsford C. Salmon provides fast and bias-aware quantification of transcript expression. Nat Methods. 2017;14:417–9.

    Article  PubMed  PubMed Central  Google Scholar 

  65. Soneson C, Love MI, Robinson MD. Differential analyses for RNA-seq: transcript-level estimates improve gene-level inferences. F1000Res. 2016;4:1521.

    Article  PubMed Central  Google Scholar 

  66. Love MI, Huber W, Anders S. Moderated estimation of fold change and dispersion for RNA-seq data with DESeq2. Genome Biol. 2014;15:550.

    Article  PubMed  PubMed Central  Google Scholar 

  67. Ramos-Rodríguez M, Subirana-Granés M, Pasquali L. UMI4Cats: an R package to analyze chromatin contact profiles obtained by UMI-4C. Bioinformatics. 2021;37:4240–2.

    Article  PubMed  PubMed Central  Google Scholar 

  68. Guéant J-L, Chéry C, Oussalah A, Nadaf J, Coelho D, Josse T, et al. A PRDX1 mutant allele causes a MMACHC secondary epimutation in cblC patients. Nat Commun. 2018;9:67.

    Article  PubMed  PubMed Central  Google Scholar 

  69. Polvani S, Pepe S, Milani S, Galli A. COUP-TFII in health and disease. Cells. 2020;9:101.

    Article  Google Scholar 

  70. Lemercier C, To RQ, Swanson BJ, Lyons GE, Konieczny SF. Mist1: a novel basic helix-loop-helix transcription factor exhibits a developmentally regulated expression pattern. Dev Biol. 1997;182:101–13.

    Article  PubMed  Google Scholar 

  71. Schwartz M. Can abnormal chromatin folding cause high-penetrance cancer predisposition? Physiol Genomics. 2022;54:380–8.

    Article  PubMed  Google Scholar 

  72. Pineda M, Mur P, Iniesta MD, Borràs E, Campos O, Vargas G, et al. MLH1 methylation screening is effective in identifying epimutation carriers. Eur J Hum Genet. 2012;20:1256–64.

    Article  PubMed  PubMed Central  Google Scholar 

  73. Morak M, Koehler U, Schackert HK, Steinke V, Royer-Pokora B, Schulmann K, et al. Biallelic MLH1 SNP cDNA expression or constitutional promoter methylation can hide genomic rearrangements causing Lynch syndrome. J Med Genet. 2011;48:513–9.

    Article  PubMed  Google Scholar 

  74. Morak M, Steinke-Lange V, Massdorf T, Benet-Pages A, Locher M, Laner A, et al. Prevalence of CNV-neutral structural genomic rearrangements in MLH1, MSH2, and PMS2 not detectable in routine NGS diagnostics. Fam Cancer. 2020;19:161–7.

    Article  PubMed  Google Scholar 

  75. Peifer M, Hertwig F, Roels F, Dreidax D, Gartlgruber M, Menon R, et al. Telomerase activation by genomic rearrangements in high-risk neuroblastoma. Nature. 2015;526:700–4.

    Article  PubMed  PubMed Central  Google Scholar 

  76. Zhang Y, Yang L, Kucherlapati M, Chen F, Hadjipanayis A, Pantazi A, et al. A pan-cancer compendium of genes deregulated by somatic genomic rearrangement across more than 1,400 cases. Cell Rep. 2018;24:515–27.

    Article  PubMed  PubMed Central  Google Scholar 

  77. Zhang Y, Chen F, Fonseca NA, He Y, Fujita M, Nakagawa H, et al. High-coverage whole-genome analysis of 1220 cancers reveals hundreds of genes deregulated by rearrangement-mediated cis-regulatory alterations. Nat Commun. 2020;11:736.

    Article  PubMed  PubMed Central  Google Scholar 

  78. Zhang Y, Yang L, Kucherlapati M, Hadjipanayis A, Pantazi A, Bristow CA, et al. Global impact of somatic structural variation on the DNA methylome of human cancers. Genome Biol. 2019;20:1–24.

    Article  Google Scholar 

  79. Ryan JL, Kaufmann WK, Raab-Traub N, Oglesbee SE, Carey LA, Gulley ML. Clonal evolution of lymphoblastoid cell lines. Lab Invest. 2006;86:1193–200.

    Article  PubMed  Google Scholar 

  80. SoRelle ED, Dai J, Bonglack EN, Heckenberg EM, Zhou JY, Giamberardino SN, et al. Single-cell RNA-seq reveals transcriptomic heterogeneity mediated by host–pathogen dynamics in lymphoblastoid cell lines. Elife. 2021;10:1–24.

    Google Scholar 

  81. Liu NQ, Ter Huurne M, Nguyen LN, Peng T, Wang SY, Studd JB, et al. The non-coding variant rs1800734 enhances DCLK3 expression through long-range interaction and promotes colorectal cancer progression. Nat Commun. 2017;8:1–10.

    Google Scholar 

  82. Sheffield NC, Thurman RE, Song L, Safi A, Stamatoyannopoulos JA, Lenhard B, et al. Patterns of regulatory activity across diverse human cell types predict tissue identity, transcription factor binding, and long-range interactions. Genome Res. 2013;23:777–88.

    Article  PubMed  PubMed Central  Google Scholar 

  83. Degtyareva AO, Antontseva E V., Merkulova TI. Regulatory snps: Altered transcription factor binding sites implicated in complex traits and diseases. Int J Mol Sci. 2021;22.

  84. Villicaña S, Bell JT. Genetic impacts on DNA methylation: research findings and future perspectives. Genome Biol. 2021;22:127.

    Article  PubMed  PubMed Central  Google Scholar 

  85. Filion GJP, Zhenilo S, Salozhin S, Yamada D, Prokhortchouk E, Defossez P-A. A family of human zinc finger proteins that bind methylated DNA and repress transcription. Mol Cell Biol. 2006;26:169–81.

    Article  PubMed  PubMed Central  Google Scholar 

  86. Alappat S, Zhang ZY, Chen YP. Msx homeobox gene family and craniofacial development. Cell Res. 2003;13:429–42.

    Article  PubMed  Google Scholar 

  87. Bendall AJ, Abate-Shen C. Roles for Msx and Dlx homeoproteins in vertebrate development. Gene. 2000;247:17–31.

    Article  PubMed  Google Scholar 

  88. Pla P, Monsoro-Burq AH. The neural border: Induction, specification and maturation of the territory that generates neural crest cells. Dev Biol. 2018;444(Suppl May):S36-46.

    Article  PubMed  Google Scholar 

  89. Kucuk E, van der Sanden BPGH, O’Gorman L, Kwint M, Derks R, Wenger AM, et al. Comprehensive de novo mutation discovery with HiFi long-read sequencing. Genome Med. 2023;15:34.

    Article  PubMed  PubMed Central  Google Scholar 

  90. Zhao X, Collins RL, Lee WP, Weber AM, Jun Y, Zhu Q, et al. Expectations and blind spots for structural variation detection from long-read assemblies and short-read genome sequencing technologies. Am J Hum Genet. 2021;108:919–28.

    Article  PubMed  PubMed Central  Google Scholar 

Download references

Acknowledgements

The authors thank the individuals who participated in this study and their families.

Funding

This study was supported by the National Institute of Health/National Cancer Institute (R01CA218342), the Spanish Ministry of Science and Innovation, which is part of Agencia Estatal de Investigación (AEI), through the Retos Investigación grants (SAF2015-68016-R, PID2019-111254RB-I00, PID2020-117099RB-I00, PID2023-151585OB-I00, and SAF2017-86242-R), CIBERONC (CB16/12/00234), and the Secretariat for Universities and Research of the Department of Business and Knowledge of the Generalitat de Catalunya grant to support the activities of research groups (2021 SGR 01112). We also thank the CERCA Program/Generalitat de Catalunya for institutional support. M.R.-R. is supported by the IMPULSO Talento Joven grant from DiabetesCERO and the EFSD/Lilly Young Investigator Award.

Author information

Authors and Affiliations

  1. Hereditary Cancer Group, ONCOBELL Program, Institut d’Investigació Biomèdica de Bellvitge (IDIBELL), L’Hospitalet de Llobregat, Spain

    Paula Climent-Cantó, Estela Dámaso, Fátima Marín, Covadonga Vara, Elisabet Munté, Gabriel Capellá & Marta Pineda

  2. Department of Medicine and Life Sciences, Universitat Pompeu Fabra, 08003, Barcelona, Spain

    Marc Subirana-Granés, Mireia Ramos-Rodríguez, Beatriz Pérez-González, Helena Raurell & Lorenzo Pasquali

  3. Molecular Genetics Laboratory, Foundation for the Promotion of Health and Biomedical Research of Valencia Region (FISABIO), University Hospital of Elche, 03203, Elche, Alicante, Spain

    Estela Dámaso & José Luis Soto

  4. Genomics Medicine Unit, Navarrabiomed, Hospital Universitario de Navarra (HUN), Universidad Pública de Navarra (UPNA), IdiSNA, 31008, Pamplona, Spain

    Ángel Alonso

  5. Department of Medicine (Oncology), Stanford Cancer Institute, Stanford University, Stanford, CA, 94305, USA

    GiWon Shin & Hanlee Ji

  6. Department of Biomedical Sciences, Cedars-Sinai Medical Center, Los Angeles, CA, 90048, USA

    Megan Hitchins

  7. Ciber Oncología (CIBERONC), Instituto Salud Carlos III, Madrid, Spain

    Fátima Marín, Gabriel Capellá & Marta Pineda

  8. Hereditary Cancer Program, Institut Català d’Oncologia (ICO), L’Hospitalet de Llobregat, Spain

    Paula Climent-Cantó, Estela Dámaso, Fátima Marín, Covadonga Vara, Elisabet Munté, Gabriel Capellá & Marta Pineda

Authors
  1. Paula Climent-Cantó
    View author publications

    You can also search for this author inPubMed Google Scholar

  2. Marc Subirana-Granés
    View author publications

    You can also search for this author inPubMed Google Scholar

  3. Mireia Ramos-Rodríguez
    View author publications

    You can also search for this author inPubMed Google Scholar

  4. Estela Dámaso
    View author publications

    You can also search for this author inPubMed Google Scholar

  5. Fátima Marín
    View author publications

    You can also search for this author inPubMed Google Scholar

  6. Covadonga Vara
    View author publications

    You can also search for this author inPubMed Google Scholar

  7. Beatriz Pérez-González
    View author publications

    You can also search for this author inPubMed Google Scholar

  8. Helena Raurell
    View author publications

    You can also search for this author inPubMed Google Scholar

  9. Elisabet Munté
    View author publications

    You can also search for this author inPubMed Google Scholar

  10. José Luis Soto
    View author publications

    You can also search for this author inPubMed Google Scholar

  11. Ángel Alonso
    View author publications

    You can also search for this author inPubMed Google Scholar

  12. GiWon Shin
    View author publications

    You can also search for this author inPubMed Google Scholar

  13. Hanlee Ji
    View author publications

    You can also search for this author inPubMed Google Scholar

  14. Megan Hitchins
    View author publications

    You can also search for this author inPubMed Google Scholar

  15. Gabriel Capellá
    View author publications

    You can also search for this author inPubMed Google Scholar

  16. Lorenzo Pasquali
    View author publications

    You can also search for this author inPubMed Google Scholar

  17. Marta Pineda
    View author publications

    You can also search for this author inPubMed Google Scholar

Contributions

G.C., L.P., and M.P. conceived the presented project. P.C.-C., E.D., H.R., M.P. and L.P. designed the experiments. P.C.-C., E.D., C.V., B.P.-G. and H.R. performed the experiments. J.L.S., A.A., and G.C. facilitated patient recruitment. P.C.-C., M.S.-G., M.R.-R., F.M., E.M., G.S., M.H., L.P., G.C. and M.P. analysed and interpreted the data. M.H., G.C., and M.P. obtained funding. P.C.-.C wrote the original manuscript. All authors reviewed the manuscript.

Corresponding authors

Correspondence to Gabriel Capellá, Lorenzo Pasquali or Marta Pineda.

Ethics declarations

Ethics approval and consent to participate

Written informed consent was obtained from all individuals enrolled. This study was approved by the Ethics Committee of the Institut d’Investigació Biomèdica de Bellvitge (code PR225/11).

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.

Additional information

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Gabriel Capellá, Lorenzo Pasquali and Marta Pineda have shared senior authorship.

Supplementary Information

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Climent-Cantó, P., Subirana-Granés, M., Ramos-Rodríguez, M. et al. Altered chromatin landscape and 3D interactions associated with primary constitutional MLH1 epimutations. Clin Epigenet 16, 193 (2024). https://doiorg.publicaciones.saludcastillayleon.es/10.1186/s13148-024-01770-3

Download citation

  • Received:

  • Accepted:

  • Published:

  • DOI: https://doiorg.publicaciones.saludcastillayleon.es/10.1186/s13148-024-01770-3

Keywords

Clinical Epigenetics

ISSN: 1868-7083

Contact us