P017 Small bowel transcriptomics define similarities and divergent pathways between Crohn, environmental enteric dysfunction (EED), and celiac diseases, implicating role for the microbiome in the pathogenesis

Haberman Ziv, Y.(1);Iqbal, N.T.(2);Braun, T.(1);Dexheimer, P.J.(3);Hadar, R.(1);Rahman, N.(2);Syed, S.(4);Moskaluk, C.(4);Moore, S.R.(4);Ali, S.A.(2);Denson, L.A.(3);

(1)Tel-HaShomer Sheba Medical Center, Department of Pediatric Gastroenterology, Ramat Gan, Israel;(2)Aga Khan University, Department of Pediatrics, Karachi, Pakistan;(3)Cincinnati Children’s Hospital Medical Center, Department of Pediatrics, Cincinnati, United States;(4)University of Virginia- Charlottesville, Department of Pediatrics, Charlottesville, United States; The Study of Environmental Enteropathy and Malnutrition (SEEM) Pakistan and CCF-supported RISK study


Crohn disease (CD), environmental enteric dysfunction (EED) and celiac disease are prevalent enteropathies linked to chronic gut inflammation. However, unlike celiac and Crohn that are prevalent in westernized countries, EED is seen in only in low- and middle-income countries of the developing world. Recent work has defined specific EED duodenal microbiota associated with stunting, and transmission of EED histologic changes and weight loss by microbial transfer to germ free mice.


We performed high throughput transcriptomics of 94 duodenal samples from the SEEM cohort:52 undernourished EED participants, 17 celiac and 25 healthy North American controls. We compared those to our previously reported transcriptomics of 248 newly diagnosed ileal CD from RISK. Comparison were performed between cases and controls within each cohort.


Above 90% of the expressed protein coding genes are shared between duodenum samples within SEEM, and Crohn ileal within RISK enabling the anticipated comparison. A substantial number of the down-regulated genes in undernourished EED participants overlapped with celiac (212/481, 44%) and with Crohn Disease (249/481, 52%).  Shared reduced signatures and functions included decrease in epithelial lipid metabolism, vitamins, and brush border transport signatures.  Relative to celiac disease, suppression of antioxidant and detoxification genes were EED- and Crohn-specific, which include the reduction of metallothioneins (MT family) that buffer against toxic metals, and aldo-keto NADPH- genes (AKR1C family) involved in detoxification of environmental compounds. A smaller proportion of the EED up-regulated signal overlapped with celiac (85/781), and we noted a more significant overlap with Crohn disease (248/781, Chi squares p<0.05). Relative to celiac disease, up-regulation of an innate anti-microbial DUOX2 and LCN2 gene, and of myeloid response signals, several CCL chemokines, and extracellular matrix signature were EED- and Crohn-specific. Consistent with this, greater level of DUOX2 protein was detected in duodenal biopsies from EED subjects vs. celiac and controls and in ileal CD vs. ileal controls. As opposed to the robust dysbiosis previously characterized in Crohn disease, we were unable to identify significant microbial signal in celiac patients in comparison to controls, while recent studies identified microbial patterns linked with EED.


We highlight important biologic signals widely overlapping between Crohn Disease (CD) and EED transcriptome in the small intestine, suggesting mechanisms involving altered gut microbiota.  Inhibiting innate immune activation or reverting Crohn dysbiosis may be beneficial future therapy for Crohn Disease and EED.