MicroRNA Expression Profiling in Porcine Liver, Jejunum and Serum upon Dietary DON Exposure Reveals Candidate Toxicity Biomarkers

Abstract

Deoxynivalenol (DON), a frequent mycotoxin worldwide, impairs human and animal health. The response of microRNAs, small non-coding RNAs, to DON has been scarcely investigated, but holds remarkable potential for biomarker applications. Hence, we aimed to investigate DON-induced changes in the microRNA expression in porcine liver, jejunum and serum by combining targeted and untargeted analyses. Piglets received uncontaminated feed or feed containing 900 µg/kg and 2500 µg/kg DON for four weeks, followed by a wash-out period. In tissue, only slight changes in microRNA expression were detected, with ssc-miR-10b being downregulated in liver of DON-exposed piglets. In serum, several microRNAs were differentially expressed upon DON exposure, four of which were validated by qPCR (ssc-miR-16, ssc-miR-128, ssc-miR-451, ssc-miR-205). The serum microRNA response to DON increased over time and declined after removal of contaminated diets. Receiver operating curve analyses for individual microRNAs were significant, and a combination of the four microRNAs increased the predictive capacity for DON exposure. Predicted microRNA target genes showed enrichment of several pathways including PIK3-AKT, Wnt/β-catenin, and adherens junctions. This study gives, for the first time, a comprehensive view of the porcine microRNA response to DON, providing a basis for future research on microRNAs as biomarkers for mycotoxins.

Keywords: biomarker; deoxynivalenol; microRNA; mycotoxin; next generation sequencing.

Figure 1

Sample collection scheme. Sample types, days of sampling and analyses performed are shown (accl.: acclimatization phase; mtx: mycotoxin).

Figure 2

MicroRNA expression profiles of Control and DON exposed pigs in liver and jejunum. Heatmap of microRNA expression in all samples from the Control and DON exposed pigs.

Figure 3

MicroRNA abundance in serum samples based on sequencing data. (a) Principal component analysis based on microRNA abundance in the serum samples; (b) volcano plots of differentially expressed genes showing the fold changes and FDR (adjusted p-values) of microRNAs in the DON_LOW and DON_HIGH samples compared to Controls. Red dots: microRNAs with FDR < 10% and > 2-fold change; blue dots: microRNAs with FDR <10%; (c) normalized counts in liver and serum for all microRNAs in all analyzed samples; (d) normalized counts in jejunum and serum for all microRNAs in all samples. The eight selected microRNAs found in liver as well as in jejunum are highlighted. Each dot represents the abundance of a microRNA in a specific sample.

Figure 4

Comparison of microRNA expression in the serum based on qPCR and sequencing data. Log2 fold changes of in the DON_LOW (left) and DON_HIGH (right) compared to the Control group assessed via small RNA sequencing (blue) and qPCR (red).

Figure 5

MicroRNA expression in the serum based on qPCR. Comparison of microRNA expression in different treatment groups, normalized by the baseline expression at d0. Significant values for the treatment effect are shown above each plot based on 2-way ANOVA followed by Tukey’s multiple comparison test. Significant differences are marked as follows: ■ p < 0.1; * p < 0.01; ** p < 0.001.

Figure 6

ROC analysis based on microRNA expression measured by qPCR for serum samples at day 26. For this analysis, DON_LOW and DON_HIGH samples were pooled together as the treated group and compared to the Control. The Combined ROC curve shows the analysis based on the logistic regression model using a combination of these four microRNAs.