Translational impact of the food-derived xeno-miRNA miR-168 in gastrointestinal cancers and prenoplastic conditions

Abstract

Introduction: Diet is one of the most important factors contributing to the multistep process of carcinogenesis. The functional role of exogenous xe-no-microRNAs (miRNAs) in this context is poorly understood. Recently, we reported the detectability of food-derived xeno miRNAs in various foods and in the gastrointestinal and colonic mucosa. However, little is known about the translational role of xeno-miRNA in the GI mucosa and specifically in GI diseases. Aims & Methods: In this work, we aimed to evaluate the potential clinical relevance of the xeno-miRNA miR-168 in the gastric mucosa along the preneoplastic conditions and gastric carcinogenesis. To this end, we performed quantitative analysis of miR-168 in different settings: 1. Samples from patients with normal mucosa (N), chronic non-atrophic (CNAG) and atrophic gastritis (CAG) and intestinal metaplasia (IM) (n=72); 2. Matched non-tumorous (NT-) and tumorous (T-) gastric cancer (GC) tissues (n=82) and 3) matched NT- and T-colorectal cancer (CRC) tissues (n=40). Survival analysis was performed using Kaplan-Meyer analysis. Results: MiR-168 was reproducibly detectable in the samples examined, with higher levels in stomach compared to colon tissue. In the stomach, a significantly higher level of miR-168 was observed in NT-GC compared to N, CNAG, AG/IM samples (p<0.01 each), but there was no difference related to H. pylori positivity or inflammation grade. Interestingly, miR-168 was higher in patients with moderate or severe AG/IM or OLGIM 3/4. Despite a significant correlation of miR-168 expression, paired sample analysis revealed a higher level of miR-168 in NT-GC compared to T-GC, with the highest level observed in the cardia and the lowest in the gastric mucosa of the antrum. Survival analysis showed only a small trend towards worse overall survival for patients with highest to lowest miR-168 levels, mostly during the first 3 years, but no long-term difference regardless of Lauren’s classification. MiR-168 levels in normal mucosa were not associated with overall survival. In correlation with matched GC samples, NT-CRC showed higher miR-168 levels compared to T-CRC, although overall there was a positive correlation between matched tissues. No survival difference was observed for higher or lower miR-168 levels in CRC. Conclusion: Food-derived xeno-miRNA are present in the gastric and colonic mucosa, with the highest levels in the stomach compared to the colon and in non-tumour compared to tumour tissues. While the potential impact of miR-168 on overall survival was weak, the higher level of miR-168 in patients with moderate and severe IM deserves further attention and further functional analyses are needed to better understand the functional role of xeno-miRNA in the GI tract.