Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2019 Jan;68(1):49-61.
doi: 10.1136/gutjnl-2017-314817. Epub 2017 Nov 15.

DNA methylation defines regional identity of human intestinal epithelial organoids and undergoes dynamic changes during development

Affiliations

DNA methylation defines regional identity of human intestinal epithelial organoids and undergoes dynamic changes during development

Judith Kraiczy et al. Gut. 2019 Jan.

Abstract

Objective: Human intestinal epithelial organoids (IEOs) are increasingly being recognised as a highly promising translational research tool. However, our understanding of their epigenetic molecular characteristics and behaviour in culture remains limited.

Design: We performed genome-wide DNA methylation and transcriptomic profiling of human IEOs derived from paediatric/adult and fetal small and large bowel as well as matching purified human gut epithelium. Furthermore, organoids were subjected to in vitro differentiation and genome editing using CRISPR/Cas9 technology.

Results: We discovered stable epigenetic signatures which define regional differences in gut epithelial function, including induction of segment-specific genes during cellular differentiation. Established DNA methylation profiles were independent of cellular environment since organoids retained their regional DNA methylation over prolonged culture periods. In contrast to paediatric and adult organoids, fetal gut-derived organoids showed distinct dynamic changes of DNA methylation and gene expression in culture, indicative of an in vitro maturation. By applying CRISPR/Cas9 genome editing to fetal organoids, we demonstrate that this process is partly regulated by TET1, an enzyme involved in the DNA demethylation process. Lastly, generating IEOs from a child diagnosed with gastric heterotopia revealed persistent and distinct disease-associated DNA methylation differences, highlighting the use of organoids as disease-specific research models.

Conclusions: Our study demonstrates striking similarities of epigenetic signatures in mucosa-derived IEOs with matching primary epithelium. Moreover, these results suggest that intestinal stem cell-intrinsic DNA methylation patterns establish and maintain regional gut specification and are involved in early epithelial development and disease.

Keywords: intestinal cell lines; intestinal development; intestinal epithelium; intestinal gene regulation; intestinal stem cell.

PubMed Disclaimer

Conflict of interest statement

Competing interests: None declared.

Figures

Figure 1
Figure 1
DNA methylation profiling of human intestinal epithelial organoids (IEOs) and primary intestinal epithelial cells (IECs). (A) Summary of experimental setup. Mucosal biopsies were obtained from distal small bowel (terminal ileum=TI) and distal large bowel (sigmoid colon=SC) and used for IEC purification or generation of IEOs. Brightfield images showing established IEOs in culture. (B) Multidimensional scaling plot of genome-wide DNA methylation profiles of purified IEC (labelled as ‘x’), mucosa-derived IEOs (labelled as ‘o’), whole biopsies from SC (labelled as ‘b’) and non-epithelial cell fraction of mucosal biopsies (labelled as ‘n’). (C) Venn diagram indicating the number of significant differentially methylated positions (DMPs) as well as overlap between TI and SC purified primary cells and respective organoids (adjusted p<0.01, n=16 IEC, n=10 IEO for each gut location.) Organoids were passages 1–11. (D) Scatterplots of average DNA methylation beta values of overlapping gut-segment specific DMPs (see figure 1C) in matched IECs and IEOs of the respective gut segment. Pearson correlation r=0.948 (TI) and r=0.955 (SC), p<2.2e-16. (E) Examples of differentially methylated regions displaying gut segment-specific methylation levels in both organoids and primary epithelium. Left panel region upstream of interleukin-6 receptor (IL6R), right panel special AT-rich sequence-binding protein 2-antisense 1 (SATB2-AS1). Chromosomal location, CpGs and DNA methylation (expressed as average beta value (ranging from 0=unmethylated to 1=fully methylated) of sample groups) are displayed. (F) Validation of DNA methylation profiles by pyrosequencing of bisulfite-converted DNA, showing per cent of CpG methylation in genomic regions of IL6R and SATB2-AS1. Data are presented as mean+SD of n=4 per group and representative for two independent experiments. (G) Heatmap of top 1000 DMPs, identified by comparing purified TI versus SC epithelium, in the respective organoids profiled at various passages (P). See also online supplementary figures S1 and S2. SCo, SC organoids; SCp, SC purified; TIo, TI organoids; TIp, TI purified.
Figure 2
Figure 2
Transcriptomic profiling of human intestinal epithelial organoids (IEOs) and primary intestinal epithelial cells (IECs). (A) Hierarchical clustering and sample heatmap of transcriptomes by RNA-sequencing of IECs and organoids from terminal ileum (TI) and sigmoid colon (SC). (B) Heatmap displaying gene expression (ie, rlog-transformed RNA-seq counts) of selected epithelial cell subset markers, genes involved in intestinal epithelial innate immune function and regulation of DNA methylation. (C) Venn diagram displaying number as well as overlap of differentially expressed genes (DEGs) comparing TI versus SC of primary IEC and organoids. Significance cut-off adj. p<0.01 and log2Fold Change>±1.5, n=11 (IEC) and 5 (IEO) for each gut segment. Organoids were passages 1–6. (D) RNA-seq read counts of selected marker genes found to retain gut segment-specific expression patterns in organoid culture. (E) Validation of differentially expressed marker genes by quantitative PCR on a validation sample set, represented as mean+SD, n=3–5 per group, *p<0.05. ALPI, intestinal alkaline phosphatase; CFTR, cystic fibrosis-transmembrane conductance regulator; CHGA, chromogranin A; CLDN15, claudin 15; DEFA, defensin alpha; DEFB1, defensin beta 1; DNMT, DNA methyltransferase; LGR5, leucin-rich repeat containing G protein-coupled receptor 5; LYZ, lysozyme; MUC2, mucin 2; OLFM4, olfactomedin-4; PIGR, polymeric immunoglobulin receptor; SATB2, special AT-rich sequence-binding protein 2; SLC5A1, solute carrier family 5 member 1; TET, ten-eleven translocation; TFF3, trefoil factor 3; TLR, Toll- like receptor.
Figure 3
Figure 3
DNA methylation-dependent, gut segment-specific induction of gene expression in intestinal epithelial organoids (IEOs) during in vitro differentiation. (A) (I) Brightfield images of terminal Ileum (TI) and sigmoid colon (SC)-derived IEOs in maintenance medium (TIo and SCo), and differentiation medium (dTIo and dSCo), showing change of gross phenotype and (II) change in intestinal stem cell and epithelial cell subset marker expression measured by real-time PCR. (B) Histogram of mean difference in DNA methylation (expressed as beta values) of all ~450K tested CpGs comparing differentiated with matched undifferentiated intestinal organoids, n=8 derived from four patients (C) DNA methylation level of selected CpGs showing stable gut segment-specific differences in undifferentiated and differentiated IEOs, boxplot of n=4 derived from four patients for each group. Array cg-identifier is listed on top. (D) Real-time PCR data showing induction of gene expression during in vitro differentiation, n=5 per group. (E) Immunofluorescent staining of undifferentiated and differentiated IEOs forFABP6 (red, upper panel) and MUC5B (red, lower panel) protein expression. Blue=DAPI, cell nuclei. (F) CpG methylation quantified by pyrosequencing located within ileal and colonic marker genes in IEOs at baseline (Ctrl) and after co-culture with DNA methyltransferase inhibitor (Aza-deoxycytidine (AdC)) over 48 hours followed by a 4-day recovery period. Data shown as mean+SD of n=4 per group. (G) Gene expression of the markers in (F) shown as absolute values normalised to glyceraldehyde-3-phosphate dehydrogenase (GAPDH). Mean+SD, n=3–4 per group. KI67, marker of proliferation Ki-67; LRIG1, leucine-rich repeats and immunoglobulin-like domains 1; SI, sucrase isomaltase; SOX9, SRY Box 9; VIL1, villin 1.
Figure 4
Figure 4
DNA methylation dynamics of human fetal intestinal epithelial organoids (IEOs). (A) Schematic illustration of fetal sample processing and brightfield microscopy images of fetal IEOs at time points 1, 3 and 5 days after seeding and under long-term maintenance conditions. (B) Multidimensional scaling plot of genome-wide DNA methylation profiles of fetal primary intestinal epithelial cells (IECs) and organoids derived from fetal proximal gut (FPG) and fetal distal gut (FDG). Data are displayed in the context of paediatric samples (see figure 1). ‘+’ indicates purified IEC, ‘number’ represents IEO sample and indicates passage. Arrows indicate longitudinal samples of the same IEO line. Fetal organoids: n=10 from eight individuals, paediatric organoids: n=10 from seven individuals (ie, three longitudinal samples). (C) Venn diagrams displaying significant differentially methylated positions (DMPs) (adj. p<0.01) and their overlaps comparing fetal IECs versus fetal IEOs (n=5–10 per group) and fetal IECs versus paediatric IECs (n=5–6 and 16). (D) Correlation plot of fold changes (log2FC) of the overlapping DMPs from (C) (‘maturation-associated DMPs’), indicating that nearly all DMPs show the same direction (ie, higher or lower methylation in fetal IEC with almost 100% concordance for small intestine, 99.7% concordance for large intestine. Correlation: Pearson’s r=0.946 (proximal) and 0.956 (distal), both adj p<0.0001). (E) Heatmap of the top 1000 overlapping DMPs between fetal versus paediatric IECs and fetal IECs versus IEOs (see C) with fetal organoids ordered according to passaging times (P1–P23). As a reference, fetal IECs and paediatric sample values are shown as group average (n=5–16). (F) DNA methylation assessed by pyrosequencing of CpGs located in the promoter region of Toll-like receptor 4 (TLR4), left, and paired-like homeodomain 1 (PITX1), right, during culture of fetal organoids. Sample groups are fetal IEC, early fetal organoids (passage ≤3) and late fetal organoids (passage ≥10). Mean+SD, n=4–5 per group, *p<0.05 and **p<0.01 versus IEC. See also online supplementary figure S4. FDGo, FDG organoids; FDGp, FDG purified epithelium; FPGo, FPG organoids; FPGp, FPG purified epithelium; SC sigmoid colon; TI, terminal ileum.
Figure 5
Figure 5
Transcriptional profiling of human fetal intestinal epithelial organoids (IEOs) and correlation with DNA methylation. (A) Multidimensional scaling plot of transcriptome data from fetal intestinal epithelial cells (IECs) and fetal organoids in the context of paediatric primary IECs. (B) Overlap of differentially expressed genes (DEGs) between fetal IEC versus fetal IEOs, and fetal IEC versus paediatric IEC (n=3–5 per fetal group, n=11 per paediatric group), separated for small intestine (left) and large intestine (right). DEG cut-off adj p<0.01 and log2FC>±1.5. (C) Scatterplots of log2FC values for each overlapping DEG transcript from the comparisons in (B). Pearson’s correlation r=0.94 (proximal) and r=0.96 (distal), both p<0.001. (D) Heatmap of top 1000 DEGs previously identified between fetal and paediatric IEC, shown in fetal organoids. Passage number indicated as ‘P’. (E) Venn diagram of overlapping genes that are both differentially methylated and expressed. (F) Five most significant gene ontologies (biological process) for genes both differentially methylated and differentially expressed in vivo and in vitro (see figures 4C and 5B), analysed individually for proximal and distal gut. See also online supplementary figure S5. DMPS, differentially methylated positions; FDG, fetal distal gut; FPG, fetal proximal gut; SC, sigmoid colon; TI, terminal ileum.
Figure 6
Figure 6
Involvement of ten-eleven translocation 1 (TET1) and DNA methyltransferases (DNMTs) in observed in vitro maturation of human fetal intestinal epithelial organoids (IEOs). (A) Gene targeting of TET1 in fetal IEOs: (I) Schematic display of targeting strategy using CRISPR-Cas9 and targeting vector, (II) confirmation of homology-directed repair (HDR) by PCR. Knockout (KO) line 1 (#1) was used for subsequent analysis. (III) Sequencing results of TET1 allele in WT and KO#1 samples aligned to reference. Underline indicates start codon; blue colour indicates gRNA target sequence, purple indicates PAM site. (B) Brightfield images of unaltered fetal proximal gut organoids (FPGo), that is, wildtype (WT) and TET1- KO fetal small intestinal IEOs derived from the same individual. (C) Quantification of DNA methylation change between two culturing time points in WT and TET1- KO. Plotted are the number of CpG positions displaying either loss (left) or gain (right) of DNA methylation in KO compared with WT organoids during culture. Time points analysed were WT P12 and P21, KO P8 and P18. p=0.02 (loss) and p=0.03 (gain), paired Wilcoxon test. (D) Heatmap displaying DNA methylation of top 1000 maturation-associated differentially methylated positions (see figure 4C) that lose DNA methylation, shown in the respective organoid. (E) Example of differentially methylated regions (DMRs) in the genes for the transcription factors homeobox A3 (HOXA3, left) and GATA binding protein 4 (GATA4, right) between WT and KO at two different time points in culture. KO early=P8, KO late=P18, WT early=P12, WT late=P21. (F) Beta values of CpG methylation in CLDN15 at different passages shown in the context of other sample groups. Each symbol represents a sample, bar indicates the average. (G) Effect on dynamic DNA methylation changes in response to treatment with DNMT inhibitors. Heatmap displaying DNA methylation of top 1000 maturation-associated DMPs that gain methylation between fetal and paediatric small intestinal IEC, shown in FPGo at different passages after treatment with Aza-deoxycytidine (AdC). FPG purified epithelium (FPGp) shown as average of n=6, terminal ileum purified (TIp) shown as average of n=16, FPGo shown as average of n=2 per condition. See also online supplementary figures S6 and S7.
Figure 7
Figure 7
Altered DNA methylation in organoids derived from gastric heterotopia. (A) Endoscopic image of rectal mucosa. Numbers indicate gastric heterotopic region (=1), mucosal ulcer (=2) and healthy, unaffected rectal mucosa (=3). (B) (I) Sections of healthy gastric tissue (STO), gastric heterotopic tissue in the rectum (GHR) and adjacent healthy rectum (REC). Shown are H&E staining as well as immunohistochemistry staining for hindgut marker CDX2 and gastric marker KRT7. (II) Intestinal epithelial organoids (IEOs) derived from stomach, gastric heterotopic and healthy rectal tissue. (C) Multidimensional scaling plot of genome-wide DNA methylation of different IEOs from the same donor. Samples were passages 3 and 5, indicated by superscript number next to the sample. STO-organoids were grown in gastric medium, REC-organoids in intestinal medium and GHR-organoids in both gastric (P3 and P5) and intestinal medium (P5). (D) Methylation levels of CpGs located in the genes of CDX2 (left) and KRT7 (right). Box plot of n=2–3 IEO per group. (E) Pyrosequencing data showing percentage of methylation in the genes of SATB2-AS1 and GATA2 in heterotopic and rectal IEOs. Data shown as mean+SD of n=2 per group. See also online supplementary figure S8. G, gastric heterotopic organoids; R, rectal organoids; S, stomach-derived organoids.

Comment in

References

    1. Jung P, Sato T, Merlos-Suárez A, et al. . Isolation and in vitro expansion of human colonic stem cells. Nat Med 2011;17:1225–7. 10.1038/nm.2470 - DOI - PubMed
    1. Sato T, Stange DE, Ferrante M, et al. . Long-term expansion of epithelial organoids from human colon, adenoma, adenocarcinoma, and Barrett’s epithelium. Gastroenterology 2011;141:1762–72. 10.1053/j.gastro.2011.07.050 - DOI - PubMed
    1. Fordham RP, Yui S, Hannan NR, et al. . Transplantation of expanded fetal intestinal progenitors contributes to colon regeneration after injury. Cell Stem Cell 2013;13:734–44. 10.1016/j.stem.2013.09.015 - DOI - PMC - PubMed
    1. Yu DH, Gadkari M, Zhou Q, et al. . Postnatal epigenetic regulation of intestinal stem cells requires DNA methylation and is guided by the microbiome. Genome Biol 2015;16:211 10.1186/s13059-015-0763-5 - DOI - PMC - PubMed
    1. Elliott EN, Sheaffer KL, Schug J, et al. . Dnmt1 is essential to maintain progenitors in the perinatal intestinal epithelium. Development 2015;142:2163–72. 10.1242/dev.117341 - DOI - PMC - PubMed

Publication types