Concomitant COX-1 and COX-2 suppression is not sufficient to induce enteropathy associated with chronic NSAID use
Kayla Barekat, Soumita Ghosh, Christin Herrmann, Karl Keat, Charles-Antoine Assenmacher, Ceylan Tanes, Naomi Wilson, Ronan Lordan, Antonijo Mrčela, Lubica Rauova, Arjun Sengupta, Ujjalkumar Subhash Das, Robin Joshi, Elliot Friedman, Marylyn D. Ritchie, Kyle Bittinger, Aalim Weljie, Ken Cadwell, Frederic D. Bushman, Gary D. Wu, Garret A. FitzGerald, Emanuela Ricciotti
Kayla Barekat, Soumita Ghosh, Christin Herrmann, Karl Keat, Charles-Antoine Assenmacher, Ceylan Tanes, Naomi Wilson, Ronan Lordan, Antonijo Mrčela, Lubica Rauova, Arjun Sengupta, Ujjalkumar Subhash Das, Robin Joshi, Elliot Friedman, Marylyn D. Ritchie, Kyle Bittinger, Aalim Weljie, Ken Cadwell, Frederic D. Bushman, Gary D. Wu, Garret A. FitzGerald, Emanuela Ricciotti
View: Text | PDF
Research Article Gastroenterology Inflammation

Concomitant COX-1 and COX-2 suppression is not sufficient to induce enteropathy associated with chronic NSAID use

Abstract

NSAIDs are the most widely used medications for the management of chronic pain; however, they are associated with numerous gastrointestinal (GI) adverse events. Although many mechanisms have been suggested, NSAID-induced enteropathy is thought to be primarily due to inhibition of both COX-1 and -2, which results in suppression of prostaglandin synthesis. Yet surprisingly, we found that concomitant postnatal deletion of Cox-1 and -2 over 10 months failed to cause intestinal injury in mice unless they were treated with naproxen or its structural analog, phenylpropionic acid, which is not a COX inhibitor. Cox double-knockout mice exhibited a distinct gut microbiome composition, and cohousing them with controls rescued their dysbiosis and delayed the onset of NSAID-induced GI bleeding. In both the UK Biobank and All of Us human cohorts, coadministration of antibiotics with NSAIDs was associated with an increased frequency of GI bleeding. These results showed that prostaglandin suppression played a trivial role in NSAID-induced enteropathy. However, Cox deletion caused dysbiosis of the gut microbiome, which amplified the enteropathic response to NSAIDs.

Authors

Kayla Barekat, Soumita Ghosh, Christin Herrmann, Karl Keat, Charles-Antoine Assenmacher, Ceylan Tanes, Naomi Wilson, Ronan Lordan, Antonijo Mrčela, Lubica Rauova, Arjun Sengupta, Ujjalkumar Subhash Das, Robin Joshi, Elliot Friedman, Marylyn D. Ritchie, Kyle Bittinger, Aalim Weljie, Ken Cadwell, Frederic D. Bushman, Gary D. Wu, Garret A. FitzGerald, Emanuela Ricciotti

×

Figure 1

Postnatal deletion of Cox-1 and -2 reveals that chronic PG depletion alone is not sufficient to induce spontaneous GI injury in mice.

Options: View larger image (or click on image) Download as PowerPoint
Postnatal deletion of Cox-1 and -2 reveals that chronic PG depletion alo...
(A) Longitudinal tracking of universal inducible Cox-DKO mice was conducted over 10 months after tamoxifen exposure (administered at 10–12 weeks of age). Monthly urinary PGI2 (PGIM), PGD2 (PGDM), PGE2 (PGDM), and thromboxane (TxM) metabolite levels measured by LC-MS/MS, analyzed separately for each sex. n = 14 Cox-DKO mice and 6 Cre–/– controls. (B) Attempted induction of PG synthesis in Cox-DKO mice via LPS challenge. Mice received 1 mg/kg BW LPS by i.p. injection, and urine was collected overnight. n = 7–9 mice per group. Urinary PG metabolites measured by LC-MS/MS for male and female Cox-DKO mice and Cre–/– controls at baseline versus post-LPS exposure. *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001 by 1-way ANOVA. (C) Total ulcer count throughout the entire GI tract (including stomach, small intestine, and large intestine), as counted by a blinded pathologist, paired with qualitative/semiquantitative pathology scores throughout the entire GI tract for longitudinal tracking experiment. Each data point represents a single mouse. (D) Weekly hemoccult test results plotted as a Kaplan-Meier curve for percentage of each group that tested negative for blood in the stool. Any individual that tested positive was marked positive for that first week and all subsequent weeks, resembling a survival curve.