Non-castrated animals fed RH tend to have lower albumin concentration than immunocastrated animals fed RH. Lower serum albumin concentration has been associated with increased animal feed efficiency (Paula et al., 2013). In this study, albumin was negatively correlated with slaughter, hot carcass, forequarter, brisket, short ribs, and flank weights, which are the main indicators of muscle deposition. Although albumin has a negative correlation with CK, which favors muscle deposition (Kaneko et al., 2008), the albumin difference found in this study may not be interpreted as a difference in muscle deposition because the main markers of the total amount of muscle mass (CK and creatinine) did not differ. Serum albumin also regulates the colloidal osmotic pressure of blood and binds and transports fatty acids, cholesterol, and some ions (copper, zinc, and calcium) during blood circulation (Evans, 2002). Thus, in this study, the increase of serum albumin in immunocastrated animals may be associated with the higher lipid metabolism in these animals than in non-castrated animals, regardless of the diet, as observed by the positive correlation between serum albumin and HDL, triglycerides, and cholesterol.

Regarding the effects of sexual condition and β-AA supplementation on lipid metabolism, non-castrated animals fed control and ZH tend to have lower serum cholesterol concentrations than immunocastrated animals fed control and ZH, respectively, which may be related to the testosterone effect (Dunshea et al., 2005; Kaneko et al., 2008), because there was no effect of β-AA supplementation on cholesterol concentration within each sexual condition. Cai et al. (2015) observed that castration significantly increased the serum levels of total cholesterol, low-density lipoprotein cholesterol, and triglycerides of pigs fed a high-fat and high-cholesterol diet, which suggests that testosterone deficiency is associated with increased serum cholesterol levels and with cholesterol metabolism.

Some studies have reported an increase in the serum NEFA concentration in response to β-AA supplementation (Eisemann et al., 1988; Eisemann and Bristol, 1998), which may indicate the mobilization of fat stores (greater lipolysis) to provide energy support for the physiological functions of other tissues (Bryant et al., 2010). However, in this study, immunocastrated animals fed ZH had lower serum NEFA concentrations than immunocastrated animals fed control and RH. Moreover, immunocastrated animals fed control and RH had higher serum NEFA concentrations than non-castrated animals fed control and RH diets. These results suggest that muscle tissue may be using NEFA as an energy source in response to the ZH- and testosterone-promoted greater muscle growth. Cônsolo et al. (2015) also reported a lower serum NEFA concentration in nonimplanted Nellore heifers fed ZH than in those fed control due to an increase in the energy demands for muscle growth in ZH-supplemented animals. Thus, although the NEFA concentrations of all treatments were below 0.4 mmol/L, which would indicate healthy animals with a positive energy balance, the likely use of NEFA as an energy source may suggest that the concentration was above 0.4 mmol/L, which would indicate adipose tissue lipolysis (Miksa et al., 2004).

Despite both non-castrated and ZH-fed animals having a known impact on protein and lipid metabolism due to an increase in the energy demands for muscle growth, combining immunocastration with β-AA supplementation did not have enough impact on animal metabolism to promote significant changes in animal performance and carcass traits, as reported by Antonelo et al. (2017) and Brigida et al. (2018). Therefore, the effects of immunocastration with β-AA supplementation on animal metabolism were also evaluated separately.

In the present study, immunocastration produced no changes in the blood electrolyte balance, which could be due to the lack of difference in water intake behavior (data not shown) between treatments. In addition, these results may be supported by the absence of differences in the blood concentrations of HCO₃ and PCO₂ or their ratio, which is a major mechanism for maintaining the blood pH (Kaneko et al., 2008). However, hemogasometry analysis revealed that blood chloride concentration was changed by sexual condition, but this change had no representative effect on acid-basic balance, since sodium and potassium remain unchanged (Kaneko et al., 2008). Furthermore, the values were within the accepted normal range for beef cattle (97-111 mEq/L; Kaneko et al., 2008).

Several researchers have reported that immunocastration increased fat deposition in the carcass of beef cattle (Dunshea et al., 2005; Amatayakul-Chantler et al., 2012, 2013; Gómez et al., 2017). These results were also observed in this study, in which immunocastrated animals had greater serum HDL and triglycerides concentrations than non-castrated animals. The transport of triglycerides to target tissues occurs via chylomicrons that, upon reaching the tissue, are broken down by the lipoprotein lipase enzyme into glycerol and fatty acids to be absorbed and stored by the cells (adipose tissue) or oxidized (muscles; Nelson and Cox, 2012). The increase in serum triglycerides in immunocastrated animals may indicate a high activity of the lipoprotein lipase enzyme, thereby enhancing lipogenesis and accumulation of adipose tissue, as suggested in this study, by the positive correlation between triglycerides and total fat trim.

Immunocastration has been also related to the lower muscle mass deposition (Dunshea et al., 2005; Walker et al., 2010; Amatayakul-Chantler et al., 2013; Gómez et al., 2017). The results observed in this study are in a good agreement with this statement, since immunocastrated animals presented a lower serum creatinine and CK concentration than non-castrated animals. Creatinine is derived from the catabolism of creatine and phosphocreatine, molecules mainly contained in muscle, and is proposed as a marker for muscle mass in a steady state (Rennie and Millward, 1983; Virgili et al., 1994). Blood creatinine concentration has been positively associated with lean muscle in sheep (Cameron, 1992; Clarke et al., 1996) and greater muscle mass in bulls (Morgan et al., 1993). Moreover, the increase of serum CK concentrations indicate a muscle damage generated from an increased long-term physical stress due to an increase in muscle deposition. According to Kaneko et al. (2008), when energy muscle demand is not enough and more energy is required for muscle contraction, CK catalyzes the transfer of the adenosine diphosphate to form adenosine triphosphate, thereby providing energy for muscle. Furthermore, CK phosphorylates the creatinine in creatine-phosphate, which is an energy storage molecule present mainly in the skeletal muscles (Kaneko et al., 2008). The positive correlations found in this study between creatinine and CK with some indicators of muscle deposition, such as Longissimus muscle area, hot carcass weight, dressing percentage, and weight of total retail cuts are in a good agreement with those statements.

Despite the tendency of animals fed ZH diet to have a lower sodium concentration than animals fed the control diet, the blood electrolyte balance was maintained and the blood pH was not changed. According to Freitas et al. (2010), the knowledge of blood electrolyte balance and its physiology and regulation is relevant because the health of the animal depends directly on the normal composition of fluid in body compartments. Therefore, the hemogasometry analysis revealed that β-AA supplementation did not affect the acid-base balance of feedlot Nellore cattle. Similarly, Abney et al. (2007) and Frese et al. (2016) also reported no difference in the hemogasometry of feedlot cattle fed β-AA diets.

β-adrenergic agonists are natural analogs of catecholamines; thus, they have glycogenolytic activity (Dunshea et al., 2005). In this study, animals fed ZH had lower serum glucose concentrations than animals fed the control diet, which may be associated with a higher energy demand proportioned by a greater muscle deposition in ZH animals. Cônsolo et al. (2015) also reported lower serum glucose concentrations in ZH-fed animals than in control animals. In this study, glucose concentration was negatively correlated with feed efficiency, which is in a good agreement with the fact that animals fed ZH have been associated with a higher performance in feedlot than non-supplemented animals (Arp et al., 2014; Cônsolo et al., 2015; Antonelo et al., 2017). Parr et al. (2011) also reported a higher performance in ZH-fed animals, which was related to an increase in insulin-like growth factor-1 and NEFA concentration in ZH-fed animals than in control animals (Parr et al., 2014). In this study, glucose was positively correlated with NEFA, which was positively correlated with BHBA, a ketone body produced from the metabolism of NEFA by the liver that is usually elevated in states of negative energy balance (Ospina et al., 2010). However, in this study, RH-fed animals had higher BHBA concentrations than control animals without differences in glucose concentration. Thus, in spite of the lack of effect of β-AA supplementation on lipolysis, these results indicated that animals fed RH were not able to metabolize BHBA at sufficient rates to avoid accumulation in the blood, likely due to a poor adaptation to a negative energy balance (Herdt, 2000). Eisemann and Bristol (1998) and Van Bibber-Krueger et al. (2015) observed no changes in BHBA concentrations when clembuterol or ZH were administered to finishing steers, respectively.

It has been reported that the use of β-AA in cattle diet promotes greater muscle hypertrophy (Dunshea et al., 2005; Avendaño-Reyes et al., 2006). The results of the present study are in a good agreement with this statement, because animals fed ZH diet had a higher serum creatinine and CK concentration, which are good markers for muscle deposition (Virgili et al., 1994; Kaneko et al., 2008), than animals fed the control diet over 30 d of β-AA supplementation. Moreover, animals fed ZH diet had a lower serum urea and total protein concentration than animals fed RH and control diets. Those results were also reported by Parr et al. (2014) and Van Bibber-Krueger et al. (2015), which may reflect a decrease in protein catabolism in skeletal muscle (Van Bibber-Krueger et al., 2015) or an increase in tissue nitrogen deposition (Brake et al., 2011). These differences in blood metabolites related to muscle deposition between RH and ZH may be caused by differences in the β receptor, which is coupled to provide a cellular β-AA effect. The RH binds to a β1 receptor, and ZH binds to a β2 receptor (Mersmann, 1998). It has already been reported that bovine animals have greater amounts of β2 cellular receptor than β1 (Mersmann, 1998); thus, ZH can provide greater effect on muscle hypertrophy compared with RH in beef cattle supplementation. Similar to the present study, Frese et al. (2016) reported higher CK concentrations in animals fed ZH than in those fed RH and control diets and stated that it could be associated with many factors, including myocardial infarction, damage of skeletal muscle, and increase in lean body mass. The same authors proved that diets containing β-AA have no effect on arrhythmia rates, which suggests that an increase in CK concentration represents an increase in lean body mass. In the present study, creatinine and CK concentrations were positively correlated with slaughter weight, Longissimus muscle area, hot carcass weight, dressing, and total weight of retail cuts, which also may suggest an increase in muscle deposition.