The impact of cyclooxygenase-2 selective and non-isoform selective NSAIDs on the gut microbiota

Nonsteroidal anti‐inflammatory drugs (NSAIDs) exert their analgesic and anti‐inflammatory effects by inhibiting the cyclooxygenase (COX)‐1 and/or COX‐2, enzymes involved in biotransformation of arachidonic acid into prostanoids. It was previously reported that some NSAIDs alter populations of bacteria in the gut and modify the way that NSAIDs work in the body. Here we directly compared the impact of NSAIDs, selective (celecoxib) and non‐selective (naproxen) for COX‐2 inhibition, on the composition of the gut microbiota in mice. We conducted a longitudinal study in adult male C57BL/6 mice that were fed ad libitum with a regular chow or with a diet containing celecoxib or naproxen for three weeks. Naproxen caused significant COX‐1 inhibition as reflected by a ~70% inhibition of urinary thromboxane metabolite (TxM), an in vivo index of COX‐1 activity, and a 60% inhibition of a COX‐1/COX‐2 urinary prostaglandin (PG)E 2 metabolite (PGEM), as measured by LC‐MS/MS. The treatment with celecoxib caused a significant 40% reduction of PGEM, but had no significant effect on TxM. Fecal samples were collected on day zero and day 21, DNA purified, 16S rRNA gene segments amplified using V1V2 primers, and amplicons sequenced. Community structure was compared using UniFrac, which generates distances between all pairs of samples as shared distances on a common phylogenetic tree, allowing assessment of clustering and gradients in community composition. For the unweighted analysis, the control group did not show a significant change over time (p=0.34; Permutational Multivariate Analysis of Variance Using Distance Matrices (‘Adonis’ function in R)), while a strong change was seen for naproxen (p=0.007; Adonis), and a lesser change was seen for celecoxib (p=0.023; Adonis). For the weighted analysis, the control group again did not change (p=0.31; Adonis), naproxen showed a trend (p= 0.089; Adonis), and no change was seen with celecoxib (p= 0.23; Adonis). Analysis of specific lineages showed strong longitudinal changes in some sequence clusters (operational taxonomic units), including Bacteroidetes S24‐7 and Clostridiales both p values < 10 −4 ). Moreover, we observed that naproxen caused gastroenteropathy in mice. The stomach of naproxen‐treated mice showed pyloric mucosal erosion and ulcers while the small intestine showed ulceration, inflammation in the lamina propria and epithelium, and thickening and blunting of the villi. Significantly fewer and less severe lesions were observed in mice treated with celecoxib both in the stomach and in the intestine. Thus, both NSAIDs have an impact on the composition of the gut microbiota, with the stronger effects seen for naproxen versus celecoxib. Effects were most evident in the unweighted analysis, indicating a greater effect on community membership than on their relative proportions. These data are consistent with the hypothesis that microbial dysbiosis is involved in the NSAID‐induced enteropathy that is more common with non‐isoform selective than COX‐2 selective NSAIDs. Support or Funding Information This work was supported by NIH grant U54 HL117798 to Garret A. FitzGerald. This abstract is from the Experimental Biology 2019 Meeting. There is no full text article associated with this abstract published in The FASEB Journal .