Social status regulates the hepatic miRNAome in rainbow trout: Implications for posttranscriptional regulation of metabolic pathways

Abstract

Juvenile rainbow trout develop social hierarchies when held in dyads, and the development of socially subordinate (SS) and social dominance (SD) phenotypes in this context has been linked to specific changes in the hepatic energy metabolism of all major macronutrients. Following our recently reported finding that transcript abundance of drosha, a key component of the microRNA (miRNA) biogenesis pathway, is increased in paired juvenile rainbow trout irrespective of social status compared to socially isolated (SI) controls, we here determined global changes of the hepatic miRNA pathway genes in detail at the transcript and protein level. Both SD and SS rainbow trout exhibited increased Ago2 protein abundance compared to SI rainbow trout, suggesting that hepatic miRNA function is increased in rainbow trout maintained in dyads. Given the well-described differences in hepatic intermediary metabolism between SD and SS rainbow trout, and the important role of miRNAs in the posttranscriptional regulation of metabolic pathways, we also identified changes in hepatic miRNA abundance between SS and SD rainbow trout using small RNA next generation sequencing. We identified a total of 24 differentially regulated miRNAs, with 15 miRNAs that exhibited increased expression, and 9 miRNAs that exhibited decreased expression in the liver of SS trout compared to SD trout. To identify potential miRNA-dependent posttranscriptional regulatory pathways important for social status-dependent regulation of hepatic metabolism in rainbow trout, we used an in silico miRNA target prediction and pathway enrichment approach. We identified enrichment for pathways related to metabolism of carbohydrates, lipids and proteins in addition to organelle-specific processes involved in energy metabolism, especially mitochondrial fusion and fission. Select predicted miRNA-mRNA target pairs within these categories were quantitatively analyzed by real-time RT-PCR to validate candidates for future studies that will probe the functional metabolic roles of specific hepatic miRNAs in the development of SD and SS metabolic phenotypes.

Fig 1. Schematic representation of the experimental design used to investigate the role of hepatic miRNAs in social status-dependent intermediary metabolism in juvenile rainbow trout, Oncorhynchus mykiss.

bw, body weight; NGS, next generation sequencing.

Fig 2

Relative steady-state mRNA abundance (+S.E.M.) of hepatic canonical miRNA biogenesis components exportin 5a (A), and argonaute 2a and argonaute 2b (B-C), as well as representative bands of argonaute protein abundance (D) in SI, SD and SS rainbow trout, Oncorhynchus mykiss. The mRNA abundance data were initially normalized using the Normagene approach, and protein data were initially normalized to total protein. Both mRNA abundance and protein data were then normalized to the SI group values to visualize relative fold-changes in SS and SD trout. Filled circles representing individual datapoints are additionally plotted for each group. A one-way ANOVAs followed by Tukey’s test was used for analysis. A p-value of p < 0.05 was used as cut-off for significant effects.

Fig 3

Number of raw reads, removed reads and mappable reads used in the small RNA next generation sequencing analysis (A), length distribution of reads (B), and Phred Score (C), an indicator of base call accuracy. A Phred score > 30 indicates 99.9% accuracy, or a 1:1000 probability of a false base call.

Fig 4. Heatmap showing hierarchical clustering of differentially regulated hepatic miRNAs between SS (SUB1-3) and SD (DOM1-3) rainbow trout, analyzed by t-tests.

See text for explanation.

Fig 5. Relative steady-state abundance (+S.E.M.) of the most abundant differentially regulated hepatic miRNA, miRNA-21-5p, as well as miRNA-722, miRNA-26a-5p, miRNA-let-7a and miRNA-152 in liver of SD and SS rainbow trout, Oncorhynchus mykiss.

A one-tailed Welch’s t-test was used for analysis, and a p-value of p<0.05 was used as cut-off for significant effects.

Fig 6. A gene network schematic depicting key predicted targets for miRNAs for gene ontology terms related to the theme of glucose regulation.

Pathways were constructed in Pathway Studio. Each target in the pathway is predicted to be regulated by a hepatic miRNA that is differentially regulated between SS and SD rainbow trout.

Fig 7

Steady state mRNA abundance (+S.E.M.) of genes involved in hepatic glucose metabolism, including gluconeogenic enzyme isoforms cytoplasmic phosphoenolpyruvate carboxykinase pck1 (A), mitochondrial phosphoenolpyruvate carboxykinase pck2 (B), gluconeogenic enzyme paralogues of glucose-6-phosphatase (C-G), and liver (H) and brain (I) isoforms of the glycogenolytic enzyme, glycogen phosphorylase. Data for SI, SD and SS rainbow trout (Oncorhynchus muykiss) were normalized using the Normagene algorithm, and then expressed relative to values for SI fish. A one-way ANOVAs followed by Tukey’s post-hoc was used for analysis. A p-value of p<0.05 was used as cut-off for significant effects.

Fig 8

Steady state mRNA abundance (+S.E.M.) of genes involved in hepatic lipid metabolism, hormone sensitive lipase, hsl (A), and those involved in the regulation of mitochondrial dynamics including mitofusin 1, mfn1 (B) mitofusin2, mfn2 (C) and mitochondrial fission protein, fis1 (D). Data for SI, SD and SS rainbow trout (Oncorhynchus mykiss) were normalized using the Normagene algorithm, and then expressed relative to values for SI fish. A one-way ANOVAs followed by Tukey’s post-hoc was used for analysis. A p-value of p<0.05 was used as cut-off for significant effects.

Fig 9. Circulating glucose concentrations in SD and SS rainbow trout (Oncorhynchus mykiss).

Data were analyzed using Welch’s t-test. A p<0.05 was used as cut-off for significant effect.