Due to rising feed costs, efficiency of feed nutrient use for better growth performance continues to be of significant interest (Holmgren and Feuz, 2015). Residual feed intake (RFI), a measure of feed efficiency in beef cattle, is known to be moderately heritable (Koch et al., 1963; Herd et al., 2004) and has been improved over the years via genetic selection (Arthur et al., 2001). However, factors other than genetic factors, including differences in host metabolism and gut microbiome contribute to variation in RFI (Herd and Arthur, 2009; Myer et al., 2017). Consequently, several studies have focused on understanding the physiological mechanisms that cause differences in RFI, the difference between an animal's actual intake vs. its predicted feed intake based on average daily gain (ADG) and metabolic body weight (Koch et al., 1963; Herd and Arthur, 2009). Animals with low (negative) RFI values consume less feed than expected and are feed efficient whereas animals with high (positive) RFI values consume more feed than expected and are feed inefficient (Koch et al., 1963).

In recent years, the advent of metabolomics has provided an opportunity to comprehensively analyze multiple metabolites in biological samples. Indeed, several studies have applied metabolomics to provide insight into the metabolic status of animals with varying RFI with a twin-goal of identifying blood metabolic signatures that could be used as predictive biomarkers due to the high cost associated with the direct measurement of RFI in animals (Goldansaz et al., 2020). Studies that have attempted to identify candidate biomarkers of RFI mostly applied analytical tools, such as nuclear magnetic resonance and liquid chromatography–mass spectrometry (LC-MS) (Connolly et al., 2019; Goldansaz et al., 2020). However, due to limited sensitivity and/or quantitative accuracy of these methods, only a small number of high-abundance metabolites can be analyzed (Pan and Raftery, 2007; Imperlini et al., 2016). The chemical isotope labeling (CIL) LC-MS is a metabolomics technique that provides a new opportunity to perform chemical-group-based metabolome profiling (Zhao and Li, 2020). This method can detect thousands of metabolites based on their chemical groups (such as amine/phenol, carbonyl, and carboxylic acid chemical groups) in biological samples thereby allowing a holistic view of the metabolome with highly accurate metabolite quantification (Zhao et al., 2019). Metabolites containing amine/phenol, carbonyl, and carboxylic acid chemical groups are common intermediates and/or end products of metabolisms of amino acids, carbohydrate (such as glucose), and fatty acids, respectively. Due to the functional roles of amino acids, carbohydrate, and fatty acids and their associated metabolic pathways on animal health and productivity, we hypothesized that metabolites related to their metabolisms could serve as candidate biomarkers of RFI. Therefore, the objective of this study was to analyze the plasma amine/phenol-, carbonyl-, and carboxylic acid-metabolome of crossbred beef steers divergent for high and low RFI to identify blood metabolic signatures that could serve as candidate biomarkers for divergent RFI in beef cattle.

The results of the performance of LFE and HFE steers are shown in Table 1. The average RFI values of HFE and LFE steers were −1.93 and 2.01 kg/d, respectively. The initial BW, final BW, and ADG were similar between the two groups (P > 0.05); however, LFE steers had greater (P = 0.01) DMI (16.0 kg/d) than HFE steers (11.9 kg/d).

Metabolites containing amine/phenol chemical group are common intermediate and/or end products of amino acid metabolism (Zhao et al., 2019). Amino acid metabolism contributes largely to the productivity of farm animals due to its functional roles in various biochemical and metabolic processes in the cells of animals including growth, production, and reproduction. Altered plasma amine/phenol-metabolome of the beef steers suggests the significance of amino acid metabolism to productivity and feed efficiency of those animals. In this study, four amine/phenol-containing metabolites were identified as candidate biomarkers to classify beef steers into high and low-RFI groups: methionine, 5-aminopentanoic acid, 6-aminohexanoic acid, 4-chlorolysine, and 7-cyano-7-carbaguanine. Relative concentration of plasma methionine was lower in HFE, relative to LFE steers. Methionine is known to be the first limiting amino acid in growing beef cattle when microbial protein is the only source of amino acids (Richardson and Hatfield, 1978), and its deficiency in diet has been reported to be associated with poor growth performance in growing beef cattle (Ragland-Gray et al., 1997). In addition to the role of methionine in tissue protein synthesis, methionine can serve as a precursor for synthesis of other amino acids such as taurine, cysteine, apolipoprotein, and can donate its methyl groups, via S-adenosyl methionine, for synthesis of choline, carnitine, creatine, and phospholipids which are all essential for improved skeletal muscle and hepatic lipid metabolism for energy supply especially during reduced supply of glucose. In this study, lower plasma level of methionine in HFE steers was accompanied with higher plasma levels of taurine and creatine, which are known to regulate lipid metabolism in humans and farm animals (da Silva et al., 2014; Karisa et al., 2014; Ibrahim et al., 2019; Li et al., 2021). In a similar study, Karisa et al. (2014) reported that plasma concentration of creatine was associated with RFI. In the same study, creatine was reported to interact with AMP activated protein kinase which is known to stimulate hepatic and skeletal muscle fatty acid oxidation. When there is an insufficient glucose supply due to low DMI, as observed in HFE steers, to meet energy demands for growth and other physiological processes, there is normally an increased hepatic lipid catabolism to generate acetyl-CoA, which can enter the citric acid cycle to generate energy in the form of ATP and/or be converted to oxidative fuels including ketones (Rui, 2014). In support of our results, several studies have reported differences in tissue mitochondrial function, and amino acid and energy metabolisms in animals divergent for low and high RFI (Bottje and Carstens, 2009; Baldassini et al., 2018; Mukiibi et al., 2018). In a previous study, Mukiibi et al. (2018) reported upregulation of hepatic genes responsible for lipid secretion, transport and efux in more-efficient beef cattle compared to less-efficient ones. In dairy cattle, Salleh et al. (2018) reported that RFI status is associated with regulation of energy via hepatic lipid metabolism. Considering these facts, we speculate that reduced plasma concentration of methionine in HFE steers was probably due to its increased uptake by the hepatic cells to synthesize other metabolites that aid lipid beta-oxidation to compensate for reduced energy supply due to low DMI.

4-chlorolysine, a derivative of lysine, was identified as a candidate biomarker of RFI in this study. Lysine plays a significant role in tissue protein synthesis and energy metabolism (Tome and Bos, 2007). Plasma lysine concentration has been reported to be associated with RFI in two previous studies in beef heifer and steer (Karisa et al., 2014; Jorge-Smeding et al., 2019), although no AUC values were provided in both studies. In our study, higher level of 4-chlorolysine in HFE steers also corresponded to higher levels of 5-aminopentanoic acid and 2-aminohexanedioic acid, which were also identified as candidate biomarkers in this study. 5-aminopentanoic acid and 2-aminohexanedioic acid are intermediate products of lysine degradation (Guidetti and Schwarcz, 2003). In fact, in a recent study, 2-aminohexanedioic acid, also known as aminoadipic acid, has been previously identified as a candidate serum metabolite biomarker of RFI in sheep (Goldansaz et al., 2020). A study in rat revealed induced cell death and reduced tissue protein synthesis with in vitro supplementation of aminoadipic acid (Nishimura et al., 2000). Thus, a high level of 4-chlorolysine and low levels of 2-aminohexanedioic acid and 5-aminopentanoic acid would be expected to result in increased tissue protein synthesis in HFE steers.

Two amino acid metabolism-related pathways, histidine metabolism and arginine biosynthesis, were enriched in HFE steers, relative to LFE. Histidine metabolism results in production of several metabolites including glutamate, histamine, and carnosine, all of which were increased in HFE steers. Glutamate promotes neural functioning, cell proliferation, and the production of other amino acids (Wünschiers et al., 2012; Holeček, 2020). Histamine can function as a homeostatic neurotransmitter while carnosine exhibits anti-inflammatory and cytoprotective effects by scavenging free radicals and reducing protein glycation mostly in skeletal muscle (Mendelson, 2008; Wünschiers et al., 2012; Holeček, 2020). Arginine biosynthesis pathway synthesizes arginine and several intermediate products including citrulline and ornithine, all of which were increased in HFE steers. Arginine helps in the control of normal cell division, wound healing, and removal of ammonia via the urea cycle (Rhoads and Wu, 2009; Wünschiers et al., 2012). Enrichment of these aforementioned pathways and their associated metabolites in HFE steers is evidence of better health and immune status, relative to LFE steers. Our results concur with prior investigations that have attempted to associate blood metabolites with RFI in cattle; several of these studies have reported an association of a good number of amino acid metabolism-related metabolites including creatine, tyrosine, glycine, glutamine, ornithine, aspartate, lysine, and valine with RFI in beef cattle, although no AUC values for these metabolites were reported (Karisa et al., 2014; Clemmons et al., 2017; Jorge-Smeding et al., 2019).

Several studies that analyzed the hepatic transcriptome of beef cattle with divergent RFI reported altered expressions of genes related to lipid and carbohydrate metabolisms (Mukiibi et al., 2018; Higgins et al., 2019), indicating possibility of differences in their hepatic metabolisms which would be expected to lead to alterations in blood concentrations of their metabolites. In our study, none of the plasma metabolites related to carbohydrate and fatty acid metabolisms (carbonyl- and carboxylic acid-metabolome), except methionine sulfoxide, which is an oxidized form of methionine, qualified as candidate biomarkers in this study. In fact, very few related metabolites were differentially abundant, and no associated metabolic pathways were different between HFE and LFE steers. These results were unexpected given the significant roles of carbohydrate and fatty acid metabolisms to the health and productivity of ruminants. In fact, blood glucose, an important carbonyl-containing metabolite and non-esterified fatty acids (NEFA), which are carboxylic acid-containing metabolites, are often used as markers of health and energy status in ruminants (Gleghorn et al., 2004; Adewuyi et al., 2005). A possible explanation for the lack of difference may be because glucose is in continuous supply via gluconeogenesis in ruminants (Young, 1977). Increased concentrations of blood NEFA reflect extensive fat mobilization from body reserves due to negative energy balance and fatty acid release from adipocytes most especially during lactation period in high-yielding dairy cows (Bowden, 1971). Unlike in dairy cows, there is a little need for an extensive body fat mobilization which is not expected to cause a significant change in blood NEFA (Clemmons et al., 2017). In agreement with our results, two previous studies observed no differences in blood glucose and NEFA in beef cattle divergent for low and high RFI (Bourgon et al., 2017; Clemmons et al., 2017).

The findings of the present study demonstrate differences in the plasma amine/phenol-metabolome of beef steers with divergent high and low RFI values indicating an association between blood amino acid metabolic signatures and RFI divergence in beef steers. Two amino acid metabolic pathways, histidine metabolism and arginine biosynthesis, were found to be associated with RFI. Five candidate metabolite biomarkers of divergent RFI related to amino acid metabolism pathway (methionine, methionine sulfoxide, 5-aminopentanoic acid, 2-aminohexanedioic acid, and 4-chlorolysine) were identified in this study. Validation studies using a larger cohort of beef cattle of different genetic pedigree are needed to confirm the robustness of the candidate plasma biomarkers identified in this study.

The original contributions presented in the study are included in the article/Supplementary Material, further inquiries can be directed to the corresponding author.

The animal study was reviewed and approved by Institutional Animal Care and Use Committees of West Virginia University.

IO and MW designed the experiment. GT, IO, SC, TS, and MI conducted the experiment and analyzed the data. IO, GT, and AP-C drafted the manuscript. IO reviewed the final manuscript together with GT and MI. All authors contributed to the article and approved the submitted version.

This work was funded by West Virginia University Experimental Station (scientific article number 3419) in support of U.S. Department of Agriculture hatch multi-state regional project W-3010.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.