POLYSORBATE-80 EXACERBATES <i>PROTEUS MIRABILIS</i>-INDUCED INTESTINAL INFLAMMATION VIA STIMULATION OF ENZYME X SECRETION. THE ENIGMA STUDY.

Background Mounting evidence suggests that food additives, especially emulsifier polysorbate-80 (P80), promote colitis via gut microbiota. Here, we explored the underlying mechanism of P80 and specific species Proteus mirabilis (P. mirabilis), a pathobiont in gut, in pathogenesis and aggravation of intestinal inflammation.

Methods This study includes two independent cohorts comprised of Crohn’s disease (CD) and healthy controls, respectively Hong Kong (92 CD, 21 controls) and Australian (98 CD, 21 controls). Previous and current consumption of P80 was quantified using a validated 26-item food additive questionnaire. Fecal microbiota composition was assessed by shotgun metagenomic sequencing. The interaction between P80 and P. mirabilis was investigated in both in vitro and in vivo models. The effect of P80 on P. mirabilis was determined by bacterial transcriptome sequencing. The interested protein was identified by silver straining followed by mass spectrometry.

Results P. mirabilis was identified as the enriched microbiota in the CD patients with high food additive consumption when compared to controls in both Hong Kong and Australia cohorts. Co-cultivation of P. mirabilis and P80 induced more cell death on human epithelial cells lines, INT 407 and NCM 460, compared to P. mirabilis alone or Escherichia coli control groups (Figure 1A). P80 exposure increased expression of virulence genes in P. mirabilis by RNA-seq. The co-cultured conditional medium of P. mirabilis and P80 (Plus.CM) triggered cell death, which was attenuated by Proteinase K, implying that the secreted protein is the functional participant in this process (Figure 1B). Compared to other fractions, the proteins with molecular weight over 100kDa exhibited the most notable function. By proteomics identification, P. mirabilis-derived enzyme X was uncovered as the potential deleterious protein in arising and aggravating inflammation (Figure 1C). In mouse models, the mice orally challenged with P. mirabilis and fed with P80 diet had shortened colon length (Figure 2A) and higher Lipocalin-2 (LCN2) levels (Figure 2B) compared to P. mirabilis solely and control groups.

Conclusion We have explored that with the addition of P80, P. mirabilis exacerbated the inflammation both in vitro and in vivo. The enzyme X, derived from P. mirabilis, was identified being the possible deleterious participant in this process and may be targeted for future precision therapy.

This work is supported by the Croucher Senior Research Fellowship and The Leona M. and Harry B. Helmsley Charitable Trust.