Recently, lncRNAs have received much consideration and a growing number of studies have shown that lncRNAs play key roles in various physiological and pathological processes (Kern et al., 2018; Alexandre et al., 2020). Whereas many of the lncRNAs and their functions are known in different species, such as humans and mouse, lncRNA research is in its infancy in domestic animals, especially in chickens (Zhang et al., 2017). In addition, the role of lncRNAs in the liver of chicken in regulating feed efficiency-related traits is unclear. The most common way of identifying relevant lncRNAs is by differential expression between contrasting conditions (Bakhtiarizadeh and Salami, 2019). Hence, several potential known and novel lncRNAs were identified in the present study by comparing the gene expression profile of the liver of two extremely different chicken breeds (Iranian native chicken vs Ross breed). Comparing the features of the identified novel lncRNAs to protein coding genes showed consistency between our results and the previous studies, as lncRNAs had a lower expression, shorter transcript length (Zhu et al., 2017), fewer exons (Wang et al., 2018) and lower GC content relative to mRNAs (Bakhtiarizadeh et al., 2016; Bakhtiarizadeh and Salami, 2019). These findings indicate that the used bioinformatic pipeline is reliable.
Moreover, in agreement with our previous study, synteny and conservation analysis of the novel lncRNAs emphasized that synteny of lncRNAs are more conserved than their cross-species sequence conservation (Bakhtiarizadeh and Salami, 2019). Therefore, syntenic analysis can be considered as a useful approach for improving the prediction of novel lncRNAs in the genomes that are incompletely annotated. In brief, 1,110 annotated lncRNAs, 525 known lincRNAs, 68 known ilncRNAs, 454 novel lincRNAs and 133 novel ilncRNAs were found. Of these, 38, 1, and 14, known lincRNAs, known ilncRNAs and novel ilncRNAs, respectively, were identified as DEGs. Although differentially expressed genes can be considered as potential candidates related to feed efficiency, however further investigations are needed to be ensured about their potential roles in these biological processes. Hence, different functional analysis were applied to further understand the functions of these genes in regulating feed efficiency in chicken.
Recent studies suggested that the function of lncRNAs can be deduced by analyzing their co-expressed mRNAs or neighboring protein-coding genes in genome wide (Jandura and Krause, 2017). Accordingly, in the cis mode, the identified target genes of the novel lincRNAs were mainly involved in lipid, carbohydrate and growth metabolism, which are related to feed efficiency and make these novel genes as ideal candidates to investigate the regulatory mechanism of feed efficiency in chicken. The relationship between lipid metabolism and feed efficiency has been reported in previous studies, as animals with lower feed efficiency have higher fat deposition and cholesterol levels (Nafikov and Beitz, 2007; Karisa et al., 2014). One of these target genes was ethyl malonyl‑CoA decarboxylase (ECHDC1, targeted by lincRNA.103491.1), which is a new metabolite proof reading enzyme. ECHDC1 can eliminate ethyl malonyl‑CoA by converting it to butyryl‑CoA (Linster et al., 2011). Interestingly, ECHDC1 was a DEG in our study that was down-regulated in Ross breed and showed an opposite expression pattern with its neighboring lincRNA.103491.1. Moreover, these genes were located in the QTL regions associated with feed efficiency, which reinforce their potential function in this subject. These findings make these genes (lincRNA and its target gene) as interesting candidates for a follow-on experiment to assess their impact on feed efficiency in chicken. The other important target gene was VPS13C that was predicted to be targeted by a lincRNA.2454.1. VPS13C, a member of the VPS13 family of proteins (VPS13A, B, C, and D), regulates galectin-12 stability in adipocytes (Yang et al., 2016). In this regard, the important roles of galectin-12 in adipocyte differentiation and lipolysis have been reported (Yang et al., 2016). Also, this gene is suggested to play an important role in glucose homeostasis for high milk production in dairy cow (Lemley et al., 2008). In the present study, higher expression of this gene and the relevant lincRNA in the commercial breed can be considered as their importance in regulating fat and glucose homeostasis, which might be the cause of higher feed efficiency in commercial chicken than the native breed. This finding also emphasized the functional diversity of lncRNAs, which may contribute to widespread regulatory roles in the liver tissue of chicken. The membrane-bound PRLR, as a member of the cytokine receptor family, is closely related to the growth hormone receptor. Over 300 separate biological activities have been attributed to PRL including endocrine signaling, metabolism, control of water and electrolyte balance, growth and development (Bu et al., 2013). PRLR is predicted to be a target gene of lincRNA.15581.1 and its higher expression in chicken can increase feed efficiency by increasing growth rate and decreasing storage energy (Hou et al., 2020).
The genes that were identified as target of differentially expressed known lincRNAs and also were associated with feed efficiency were included B3GALT5, ACAA2, CDCA7L, CHST7, LRRC8D, LRRC8C, LMAN1, CTDSPL, EBP, and AGPAT3. Accordingly, ACAA2 that was predicted to be targeted by lincRNA.15325 is a gene involved in mitochondrial fatty acid oxidation. ACAA2 encodes an enzyme that catalyze the cleavage of 3-ketoacyl CoA to yield acetyl-CoA and acyl-CoA, the final step of the mitochondrial fatty acid beta-oxidation spiral (Abasht et al., 2019). In this study, lincRNA.15325 and its related target gene were up-regulated in the commercial breed. Taking into account its importance in fatty acid oxidation, it might be the cause of higher feed efficiency in commercial chicken than the native breed. It is reported that an increase in the expression of EBP, a key gene functioning in cholesterol biosynthesis, exhibited greater rates of gain and feed efficiency (Connor et al., 2010). This is supported by our result that lincRNA.2988 and its neighboring gene (EBP) showed higher expression in Ross breed and may impact feed digestion. In this context, B3GALT5 (β-1, 3-Galactosyltransferase) was predicted as a cis target gene of lincRNA.1424. Zeng et al. (2017) conducted a study to identify duodenum genes and pathways through transcriptional profiling in two extreme RFI phenotypes of the duck population. B3GALT was DEG between the two duck breeds. B3GALT5 is involved in glycosylation that is a metabolic pathway consisting of the enzymatic modification of proteins and lipids through the stepwise addition of sugars (Trinchera et al., 2014). Therefore, higher expression of lincRNA.1424 and its target gene in higher efficient chicken can increase feed efficiency with enzymatic modification of proteins and lipids. Also, lincRNA.14534, lincRNA.313, lincRNA.15325, lincRNA.5912, lincRNA.6167, lincRNA.1460 and lincRNA.2104 were up-regulated in the liver of high efficient breed (Ross) and their expression were positively correlated with LRRC8D, LRRC8C (Brunes et al., 2021), LMAN1 (Reyer et al., 2017), CTDSPL (Wolc et al., 2013), CDCA7L (Ramayo-Caldas et al., 2019), CHST7 (Dawson et al., 2006) and AGPAT3 (Zarek et al., 2017). Previous studies reported these target genes to be located within the most significant SNPs associated with RFI or feed efficiency.
Sixteen lincRNAs were found to be co-expressed significantly with the MCHR1 gene. This gene is related to chemical synaptic transmission and regulates energy homeostasis and body weight. The function of this gene in reducing feed intake, body weight and body fat have been reported in mice (Zhang et al., 2014). Interestingly, MCHR1 was up-regulated in the commercial breed, which is in agreement with its function to reduce the feed intake and increase feed efficiency. It is worth to note that animals with higher feed efficiency require less energy for metabolism (McKenna et al., 2019; Guinguina et al., 2020). Moreover, 15 lincRNAs were predicted as potential regulators of ADRA2A, in trans mode. Previous studies have shown that activation of ADRA2A, through inhibition of hormone-sensitive lipase, leads to inactivation of the adipocytes lipolysis receptor and prevents adipocyte accumulation (Sawczuk et al., 2013). The other known lincRNA (lincRNA.14916) was found to target GAD2, which is associated with stimulating food intake. LincRNA.14916 and its co-expression gene were up-regulated in the commercial breed. Accordingly, it appears to increased feed efficiency in commercial poultry by reducing maintenance energy (Boutin et al., 2003). As mentioned above, the predicted target genes for the known lincRNAs were similar to the known ilncRNAs, which indicate a synergistic effect between lincRNAs and ilncRNAs for regulating a common biological process. LincRNA.13441 (as a novel lincRNA) was predicted as a potential regulator of 133 genes. Its target genes were enriched in various biological categories including “calcium ion transport” and “muscle contraction regulation”. It is well documented that calcium signaling is an important modulator of lipid metabolism (Xue et al., 2001). Hence, the function of these lincRNAs (such as 537 ADRA2A and MCHR1) could be closely related to lipid metabolism as well as feed efficiency development due to their co-expressed targeted mRNAs in the commercial breed affect lipid metabolism To better understand how lncRNAs cooperate with their target genes, integrated networks (separately for each of lincRNAs and ilncRNAs) were constructed and one significant module was found in each of the networks. Interestingly, member genes of both modules were significantly enriched in functional categories related to feed efficiency including “lipid phosphorylation”, “glucose homeostasis”, “carbohydrate homeostasis”, “insulin receptor signaling pathway” and “calcium ion transport”. Energy metabolism is an important factor affecting the feed efficiency of livestock and poultry (Kaewpila et al., 2018). ADAR2A encodes adrenoceptor alpha 2A and is a regulator of catecholamines, which have been introduced to be associated with energy metabolism and fat metabolism. Catecholamine-stimulated whole body lipolysis and lipolysis in subcutaneous adipocytes are blunted in obesity (Blaak et al., 1994), thereby limiting lipid mobilization and favoring fat accumulation. Notably, ADAR2A was predicted to be target of 24 lncRNAs (lincRNA and ilncRNA) that might be regarded as key regulators. On the other hand, the other members of these module have been reported to be involved in calcium signaling pathways including MCHR1 (Pissios et al., 2003), P2RY1 (Choi et al., 2001) and F2R (Marchesi et al., 2019) (Fig. 6–7). According to a previous study calcium participated in the expression of genes related to lipid metabolism and prohibition of fat deposition (Cao et al., 2017).
QTL analysis revealed that 11 novel ilncRNAs and 47 novel lincRNAs as putative effective lncRNAs in feed efficiency related processes, as they were overlapped with the potential regions associated with RFI in genome wide. Of these, the predicted target genes of the four lincRNAs including lincRNA.10349 (target ECHDC1) (Linster et al., 2011), lincRNA.10336 (target SMAD3) (Yadav et al., 2011), lincRNA.2986 (target EBP) (Connor et al., 2010) and lincRNA.10372 (target TPD52L1) (Kamili et al., 2015) were related to lipid metabolism. For example, SMAD3 is a multifaceted regulator in adipose physiology, pathogenesis of obesity and type 2 diabetes (Yadav et al., 2011). As discussed above, animals with lower feed efficiency have higher fat deposition and cholesterol levels (Nafikov and Beitz, 2007; Karisa et al., 2014). In addition, six lincRNAs (linRNA.1964, linRNA.2201, linRNA.2215, linRNA.5649, linRNA.2920, linRNA.2178) were located in more than one QTL related feed efficiency. CTSC, a predicted cis target gene of lincRNA.2201, encodes a lysosomal cysteine proteinase and play a central role in bacterial killing and immune regulation in T lymphocytes. The effects of this gene on the observed difference between the pigs with low and high RFI have been reported (Gondret et al., 2017). The closest protein coding gene to lincRNA.1964 was DLEU7, which encodes a protein containing 221 amino acids. Fibroblast growth factor (FGF) regulates the expression of DLEU7 during early embryogenesis (Zhu et al., 2012). Moreover, association of this gene with human height has been reported (Weedon et al., 2008; Sovio et al., 2009; Kang et al., 2010), which can be suggested to play an important role in chicken growth and feed efficiency. The other gene related to growth rate was RAB30 (cis target of lincRNA.2215). Claire D’Andre et al. (2013) conducted a study on the identification and characterization of genes that control fat deposition in chicken. RAB30 was appeared to be down-regulated in slow-growing Xinghua chickens. lincRNAs that were located in QTL regions related to feed efficiency compared with other lincRNAs are more likely to be truly related to feed efficiency. Cis target genes of these lincRNAs that were involved in feed efficiency and RFI or has a known role in the growth rate and lipid metabolism, were interesting functional candidate genes responsible for feed efficiency in chicken. Our findings supported this hypothesis that lncRNAs may affect feed efficiency mainly through regulating lipid metabolism, glucose homeostasis, growth rate, immune system, modification of proteins and energy homeostasis. However further experiments still require to validate the suggested functions of these lncRNAs.