ABSTRACT: The current study was designed to investigate the circulating miRNA expression signature of dairy cows during several key time points around calving, and under different fatty acids (FA) treatments to get insights into different aspects of metabolic adaptations and their interaction with administrated FA. In this trial, 32 transition Holstein dairy cows were randomly assigned to one of the following treatments: 1-CTRL (n = 8; coconut oil, Bio-Kokosöl #665, Kräuterhaus Sanct Bernhard, KG, Bad Ditzenbach, Germany; 38 and 76 g/d in antepartum (AP) and postpartum (PP)), 2- Essential fatty acids (EFA, n=8, a combination of linseed oil (DERBY® Leinöl #4026921003087, DERBY Spezialfutter GmbH, Münster, Germany; 39 and 78 g/d in AP and PP) and safflower oil (GEFRO Distelöl, GEFRO Reformversand Frommlet KG, Memmingen, Germany; 2 and 4 g/d in AP and PP), 3- conjugated linoleic acid (CLA, n = 8, Lutalin® ; cis-9, trans-11, 10 g/d trans- 10, cis-12 CLA, BASF SE, Lampertheim, Germany; 19 and 38 g/d in AP and PP), and 4- EFA+CLA, a combination of linseed oil, safflower oil and Lutalin® from day 63 AP to day 63 PP. From each cow, four blood samples were collected individually on days -21, +1, +28, and +63 relative to calving, and the plasma fraction was used for RNA extraction. Total RNA was extracted from 128 plasma samples using the QIAGEN miRNeasy serum/plasma kit (Qiagen GmbH, Hilden, Germany), and checked for yield and purity using the small RNA kit on a 2100 Bioanalyzer (Agilent Technologies, Santa Clara, CA, USA). We used the QIAseq miRNA library kit (QIAGEN, Hilden, Germany) for library preparation and Caliper (PerkinElmer, Waltham, MA) for quality check. NGS was performed on the libraries using the NovaSeq 6000 (Illumina, San Diego, CA) setting on single-end 100 bp mode on the respective platform of IGA Technology Services, Udine, Italy (https://igatechnology.com/). Raw-sequencing data was checked for quality (format conversion to FASTQ, demultiplexing, Adapter Trimming, and UMI Remova) using the Illumina Bcl2fastq2 Conversion Software (version 2.20). Detected sequences were analyzed with the miRDeep2 (miRDeep2.pl) software package (version 2.0.0.5) to detect known miRNA and predict putative novel miRNA (using human, ovine, and goat miRNA datasets). MiRDeep2 was fed with the Bos taurus miRNA collection available at miRBase (www.mirbase.org). Read counts for each known and novel miRNA were compiled using the HTSeq-count script. The quantified counts were normalized using the TMM method via the edgeR (version 3.12.0) package in R software (Version 4.0). Only miRNA with at least one count per million over at least 2 samples were taken into consideration for the analysis. Differentially expressed miRNA (DEM) during the time points were identified by performing generalized linear model (GLM) likelihood ratio tests (glmLRT) using the GLM approach in edgeR. The DEMs have been interpreted separately I-during different time points, and also II-between treatments, to gain insight into metabolic and immune adaptation, as well as the role of fatty acids at each time point. A time-affected DEM identified key signaling pathways that are involved in initiating and regulating metabolic shifts during the transition from gestation (non-lactation) to lactation. Although, fatty acid treatments had only minor effects on miRNA expression (insufficient DEM to perform pathway analysis).
