Selective hypermethylation is evident in small intestine samples from infants with necrotizing enterocolitis

Abstract

Objective: Necrotizing enterocolitis (NEC) is the most common and lethal gastrointestinal disease affecting preterm infants. NEC develops suddenly and is characterized by gut barrier destruction, an inflammatory response, intestinal necrosis and multi-system organ failure. There is currently no method for early NEC detection, and the pathogenesis of NEC remains unclear.

Design: To further understand the molecular mechanisms that support NEC, we used solution phase hybridization and next-generation DNA sequencing of bisulfite converted DNA to perform targeted genome-wide analysis of DNA methylation at high read depth.

Results: We found that ileal samples from surgical NEC infants (n = 5) exist in a broadly hypermethylated state relative to their non-NEC counterparts (n = 9). These trends were not uniform, with hypermethylation being most consistently observed outside CpG islands and promoters. We further identified several biologically interesting gene promoters that displayed differential methylation in NEC and a number of biological pathways that appear dysregulated in NEC. We also found that DNA methylation patterns identified in ileal NEC tissue were correlated with those found and published previously in stool samples from NEC-affected infants.

Conclusion: We confirmed that surgical NEC is associated with broad DNA hypermethylation in the ileum, and this may be detectable in stool samples of affected individuals. Thus, an epigenomic liquid biopsy of stool may have significant potential as a biomarker with respect to the diagnostic/predictive detection of NEC. Our findings, along with recent similar observations in colon, suggest that epigenomic dysregulation is a significant feature of surgical NEC. These findings motivate future studies which will involve the longitudinal screening of samples obtained prior to the onset of NEC. Our long-term goal is the development of novel screening, diagnostic and phenotyping methods for NEC.

Keywords: DNA methylation; Epigenetics; Ileum; Intestine; Necrotizing enterocolitis; Neonatal.

Fig. 1

Distribution of methylation of non-NEC (blue) versus NEC (red) ileum by genomic element in A all CpG sites, B promoters without CGI, C promoter CGIs, D introns/exons without CGI, E intergenic CGI, F CGIs in introns/exons, G CGI shores, H intergenic regions without CGI. Note that only CpG sites read to a depth of 10 × or more were included in these analyses

Fig. 2

Comparison methylation category distribution (Hypo ≤ 20%, IM > 20% and < 80%, Hyper ≥ 80%) of non-NEC and NEC ileum by genomic element

Fig. 3

Non-NEC (blue) and NEC (red) methylation across Chromosome 1 broken down by genomic element

Fig. 4

A, B Comparison of percent methylation between NEC (red) and non-NEC (blue) between ileum tissue and previously published data [8] from WGBS of LCM enterocytes. A All sites shared. B Shared sites with p < 0.05 in both methods. C, D Comparison of methylation difference (NEC minus non-NEC) between ileum tissue and WGBS of LCM enterocytes. C All sites shared. D Shared sites with p < 0.05 in both methods. In each case, the dashed line represents a correlation value of 1 between the two data sets

Fig. 5

DNA methylation levels in the Oncostatin M (OSM) locus in ileum samples from NEC (red) and control (black) neonates. Differential methylation levels are shown across multiple CpG sites that cluster around regulatory elements in and around the flanking regions of the OSM gene locus

Fig. 6

A Comparison of percent methylation between NEC (red) and non-NEC/control (blue) between ileum tissue and previously published colon tissue . Shared sites with p < 0.05 in both tissues. B Comparison of methylation difference (NEC minus non-NEC/control) between ileum tissue and colon. Shared sites with a p < 0.05 in both tissues. In each case, the dashed line represents a correlation value of 1 between the two data sets

Fig. 7

A, B Comparison of percent methylation between NEC (red) and non-NEC/control (blue) between ileum tissue and stool. A All sites shared. B Shared sites with a minimum absolute methylation difference of 5% or more and p < 0.05 in both methods. C, D Comparison of methylation difference (NEC minus non-NEC/control) between ileum tissue and stool. C All sites shared. D Shared sites with a minimum absolute methylation difference of 5% or more and p < 0.05 in both methods. In each case, the dashed line represents a correlation value of 1 between the two data sets