Exploring the Regulatory Potential of Long Non-Coding RNA in Feed Efficiency of Indicine Cattle

Abstract

Long non-coding RNA (lncRNA) can regulate several aspects of gene expression, being associated with complex phenotypes in humans and livestock species. In taurine beef cattle, recent evidence points to the involvement of lncRNA in feed efficiency (FE), a proxy for increased productivity and sustainability. Here, we hypothesized specific regulatory roles of lncRNA in FE of indicine cattle. Using RNA-Seq data from the liver, muscle, hypothalamus, pituitary gland and adrenal gland from Nellore bulls with divergent FE, we submitted new transcripts to a series of filters to confidently predict lncRNA. Then, we identified lncRNA that were differentially expressed (DE) and/or key regulators of FE. Finally, we explored lncRNA genomic location and interactions with miRNA and mRNA to infer potential function. We were able to identify 126 relevant lncRNA for FE in Bos indicus, some with high homology to previously identified lncRNA in Bos taurus and some possible specific regulators of FE in indicine cattle. Moreover, lncRNA identified here were linked to previously described mechanisms related to FE in hypothalamus-pituitary-adrenal axis and are expected to help elucidate this complex phenotype. This study contributes to expanding the catalogue of lncRNA, particularly in indicine cattle, and identifies candidates for further studies in animal selection and management.

Keywords: Bos indicus; RNA-Seq; co-expression network; residual feed intake.

Figure 1

Pipeline used to identify and characterize new Long non-coding RNA (lncRNA) and identify relevant lncRNA for feed efficiency in Nellore cattle. DE—Differential expression; RIF—Regulatory Impact Factor; PCIT—Partial Correlation and Information Theory.

Figure 2

Classification of lncRNA. Class codes in relation to known genes (A) and known genes + algorithm-predicted genes (B); number of exons (C); and length in base pairs (D). Classification of lncRNA was according to Trapnell (2017), class codes are: = “-“ — Complete match of intron chain; “c”—Contained; “j”—Potentially novel isoform (fragment): At least one splice junction is shared with a reference transcript; “i”—A transfrag falling entirely within a reference intron; “o”—Generic exonic overlap with a reference transcript; “u”—Unknown, intergenic transcript; “x”—Exonic overlap with reference on the opposite strand; “s”—An intron of the transfrag overlaps a reference intron on the opposite strand (likely due to read mapping errors).

Figure 3

Number of expressed lncRNA identified per tissue type.

Figure 4

Minimum free energy structures encoding base-pair probabilities for lncRNA, color-coded from 0 (purple) to 1 (red). (A) TCONS_00051404—differentially expressed (DE) in muscle; (B) TCONS_00040537—DE and key regulator in adrenal gland; (C) TCONS_00111349—key regulator in liver; (D) TCONS_00106745—DE in liver and key regulator in adrenal gland; (E) TCONS_00222966—DE in hypothalamus and liver; (F) TCONS_00062811—DE in pituitary gland.

Figure 5

Number of DE (differentially expressed lncRNA) and Key (key regulator lncRNA) that are also quantitative trait loci (QTL) (overlap QTL for traits related to feed efficiency), miRNA (potential to be a miRNA precursor) and NONCODE (high similarity with previously described lncRNA).

Figure 6

Liver subnetwork comprised of the first neighbors of TCONS_00188391, TCONS_00111349, TCONS_00190687 and TCONS_00106745. The light color indicates mRNA and the dark color lncRNA; differentially expressed (DE) mRNA and lncRNA are indicated by the black border; key lncRNA and mRNA are indicated as a triangle. DE and key mRNA correspond to results from Alexandre et al. (2019 25).