Effect of supplemental dietary slow-release urea on growth performance and physiological status of dairy heifers

2.1 Study site, animals and study design

The experiment was conducted in the Dingxi Livestock Cooperative (Dingxi city, Gansu province, China), a temperate semi-arid area. The average annual temperature ranges between 5.7 and 7.7°C, number of frost-free days ranges between 122 and 160, and the annual rainfall ranges between 350 and 600 mm. Sixteen Holstein heifers (322 ± 10 kg; 10 months of age) were penned individually under a roofed, three-sided shelter and were offered a diet consisting of 70% roughage and 30% concentrate, and with free access to water. The heifers were assigned randomly to one of four treatments (four heifers per treatment) that differed in the level of slow-release urea (Hebei Xingmu Agricultural Technology Co., Ltd) supplementation (0% [U₀], 1% [U₁], 1.5% [U_1.5], 2% [U₂] DM) (Table 1). The total mixed rations were formulated to provide equal intakes of metabolizable energy (ME) and N according to National Research Council (NRC, 2000). The diets were prepared daily and fed ad libitum twice each day, at 07:00 and 19:00 hours. Orts were weighed before morning feeding each day to calculate daily intake. Twenty days were allowed for adaptation to the conditions and 75 days for measurements.

2.2 Sample collection and analysis

The heifers were weighed every 2 weeks before feeding in the morning. FCR was calculated as DMI divided by weight gain (WG). A 10 ml caudal blood sample was collected in a vacuum tube from each heifer after each weighing. The blood sample was centrifuged at 3000 × g for 15 min and the serum was stored at −80°C for analysis. Aspartate transaminase (AST), glutamic-pyruvic transaminase (ALT), blood urea nitrogen (BUN), glucose (GLU), free fatty acid (FFA) and β-hydroxybutyrate (β-HB) were measured colorimetrically by an automatic biochemistry analyzer (Hitachi 7160), while insulin (INS) and growth hormone (GH) were measured by automatic radioimmunoassay (r-911, University of Science and Technology of China Industry Co., Lanzhou, China). All blood samples were analyzed by Bejing Sinoouk Institute of Biological Technology, Beijing, China. Rumen fluid samples from each heifer were collected monthly by using a 2 m long (diameter 1.2 cm) esophageal tube, which had a cylindrical hollow metal filter head (diameter 1.2 cm, length 8 cm) before morning feeding. The rumen fluid was strained through four layers of cheesecloth, and stored in 10 ml tubes at −80°C for determination of volatile fatty acids (VFA) and NH₃-N. Rumen fluid pH was measured by a pH meter (M90; Corning Inc., Corning, NY, USA). Feed samples were collected every 2 weeks, dried at 105°C for 48 hr in a forced-air oven and then ground through a 1 mm screen to analyze for dry matter (DM), crude protein (CP), ash, neutral detergent fiber (NDF) including sodium sulfite but without heat-stable amylase and acid detergent fiber (ADF) (AOAC, 1990; Van Soest, Robertson, & Lewis, 1991). VFA concentrations of rumen fluid were determined using high performance liquid chromatograph (1200; Agilent Technologies Co, Inc., Santa Clara, CA, USA), and the NH₃-N concentration was determined using a spectrophotometer (U-2910; Hitachi Construction Co., Ltd. Tokyo, Japan).

2.3 Statistical analysis

Data (means ± SEM) from the blood and rumen samples were averaged over the experimental period for each animal. The statistical analysis system package (SPSS 20.0, SPSS Inc., Chicago, IL, USA) was used to compare treatments by one way analysis of variance and least significant difference. Correlation analysis of variables was done using bivariate correlation analysis (SPSS correlation analysis). Statistical significance was accepted at p < .05.

3.1 DMI and ADG

DMI for U₀ and U₁ heifers were higher (p < .05) than for U_1.5 and U₂ heifers (Table 2). There was no difference (p > .05) in initial body weights (BW) of the heifers among the four treatments. Final body weights and ADG of U₁ and U₂ did not differ (p > .05) and were higher (p = .05,) than those in U₀ and U_1.5 (Table 2). The FCR of U_1.5 and U₂ was lower than in U₀ (p < .05) (Table 2).

3.2 Rumen characteristics and serum metabolites

The pH and VFA concentrations of rumen fluid did not differ (p > .05) among treatments (Table 3). The NH₃-N level increased with an increase of slow-release urea in the diet and was higher (p < .05) in U₂ than in the other treatments (Table 3). Urea level supplementation was correlated positively with ruminal NH₃-N concentration (p < .05).

There was no difference among groups (p > .05) in serum AST, ALT, GLU, β-HB and GH concentrations (Table 4). AST and ALT were positively correlated (p < .05). BUN concentration in U₀ was higher than in U₁ and U_1.5 and of FFA concentration in U₂ was higher (p < .05) than in the other three treatments. The concentration of serum INS in U₁ was higher (p < .05) than that in U₀ and U₂, but did not differ (p > .05) from U_1.5 (Table 4).

4.1 Effect of urea supplementation on DMI, ADG and FCR

Supplemental dietary urea can improve ADG (Gleghorn et al., 2004; Lazzarini et al., 2009; Wittayakun, Chainetr, Innaree, & Pranamornkith, 2016) but can also reduce DMI due to the poor palatability of urea and an imbalance between energy and nitrogen in the rumen (Redman, Kellaway, & Leibholz, 1980). In the present study, the addition of 1.5% and 2% DM of slow-release urea resulted in a reduced DMI. However, the U_1.5, along with the U₁ heifers, had the highest ADG. Consequently, we reasoned that the optimal level of slow-release urea supplement lay between 1% and 1.5% DM of the total mixed ration. Burque, Abdullah, Babar, Javed, and Hawaz (2008) reported that in calves supplemented with urea, DMI decreased significantly when the urea level exceeded 2.0% DM, but that weight gain and feed efficiency increased significantly with a urea level of up to 1.0% DM. In the present study, up to 17% flax meal was included in the diet in order to obtain isocaloric and isonitrogenous rations in the four treatments. Flax meal could hamper nutrient utilization because of anti-nutritional components (Bhatty, 1993; Madhusudhan, Ramesh, Ogawua, Sasaoka, & Singh, 1986). However, the U₁, U_1.5 heifers received up to 8% flax meal in their diets and Gilberry, Lardy, Hagberg, and Baur (2010) reported that 8% flax meal in the diet did not alter the ruminal or total tract digestion of organic matter, crude protein, NDF and ADF in growing beef cattle. The increased proportion of slow-release urea in the ration resulted in higher levels of rumen degradable protein, which promoted higher rumen microbial activity and proliferation, and increased DM digestibility and intake (Westwood, Lean, Garvin, & Wynn, 2000). Previous studies showed higher DMI in calves with 70% rumen degradable protein diet (Sultan, Javaid, Nadeem, Akhtar, & Mustafa, 2009), which is similar to the present rations of U₁ and U_1.5.

4.2 Effect of urea supplementation on rumen characteristics

No difference in ruminal pH among treatments was observed in the present study, which could be due to the high ratio of forage in the diet (70% DM), resulting in high buffering of the rumen fluid (Ding et al., 2015; Emmanuel et al., 2015; Preston, 1972; Taylor-Edwards et al., 2009). Rumen microbes can use rumen degradable protein in the form of NPN or true protein to synthesize microbial protein but may also deaminate amino acids, resulting in an increased NH₃-N concentration in the rumen fluid (Bach, Calsamiglia, & Stem, 2005). The results in the present study showed an increase of NH₃-N concentration in the rumen fluid with an increase of slow-release urea level in the diet, as was demonstrated in other studies (Gabler & Heinrichs, 2003; Griswold, Apgar, Bouton, & Firkins, 2003). However, concentrations of NH₃-N in all treatments of the present study were higher than 5 mg/dl, which is the minimum concentration recommended for rumen microbial growth (Satter & Slyter, 1974). The ruminal concentrations of each individual VFA (acetic acid, propionic acid, butyric acid) and of the total volatile acids were all higher, albeit non-significantly, in U₁ and U_1.5 than in U₀ and U₂. As VFA provide approximately 70% of the energy requirements of ruminants (Bergman, 1990), this would suggest that more energy was available for the U₁ and U_1.5 groups than for the U₀ and U₂ groups. However, conclusions must be taken with caution as Taylor-Edwards et al. (2009) reported no effect on VFA concentrations with urea supplementation.

Urea level supplementation was correlated positively with ruminal NH₃-N concentration. The quantity of NH₃-N absorbed across the ruminal wall is determined mainly by diet but also by ruminal factors, with the most important factors being rumen degradable protein, contributions of endogenous sources of nitrogen (e. g., urea) to the ruminal NH₃-N pool, and dietary ruminally available energy (Reynolds & Kristensen, 2008). After absorption across the ruminal wall into the portal blood, NH₃-N reaching the liver is detoxified primarily to urea in the ornithine cycle (Khajehdizaj, Taghizadeh, & Nobari, 2014).

4.3 Effect of urea on serum metabolites

The serum metabolite concentrations of AST, ALT, GLU, BUN, FFA, β-HB and INS are commonly used to assess the nutritional status of livestock (Khajehdizaj et al., 2014). With liver damage, AST and ALT are released into the blood and consequently, these metabolites are often used as indicators of the health of the liver. Ammonium nitrogen is detoxified in the liver (Khajehdizaj et al., 2014), and the high level of dietary urea may affect the liver. However, the differences of serum AST and ALT did not differ among the four treatments in the present study which meant that the slow-release urea was not detrimental to the heifers. These two serum metabolites respond similarly to liver damage and, consequently, there was a correlation between them.

The concentration of BUN reflects the efficiency of protein synthesis and the urea recycling rate and is commonly used to assess the nutritional status of livestock. The urea recycling rate is correlated positively with the degradable N intake (Lapierre & Lobley, 2001). An increase in BUN level can indicate an increase of protein catabolism and urea recirculation in the liver, and a decrease of nitrogen retention (Hall et al., 1995; Stanley et al., 2002). The decrease of serum BUN in the U₁ and U_1.5 heifers compared with U₀ heifers indicated an increase of protein synthesis. However, the increase of BUN concentration in the diet with 2% urea inclusion was due mainly to the low DMI. Although 1.5% urea inclusion also reduced DMI, the level of BUN was not affected, which resulted in a high rate of protein catabolism and urea recycling, as was suggested by Hoover and Stokes (1991).

Blood GH plays a role in growth and fat decomposition in ruminants by inhibiting glucose utilization by muscle and adipose tissue, and by promoting liver gluconeogenesis and decomposition of glycogen (Richard & Stephens, 2011). Therefore, an increase in blood GH is generally concomitant with an increase in blood GLU and a decrease in blood FFA. Similarly, insulin also plays a role in promoting lipogenesis (McGillicuddy et al., 2009). No significant difference in blood GLU among the four treatments emerged in the current study. The U₂ heifers had higher levels of FFAs and tended to have lower levels of INS than the other treatments, which was due to the low feed energy intake in this group. Studies have shown that GLU is insensitive to change in body energy status because of its homeostatic regulation (Grünwaldt et al., 2005; Herdt, 2000). Serum FFAs are related to cow energy status and depot fat mobilization as a consequence of energy mobilization (Block et al., 2001). The heifers with the 2% urea supplement increased their live weight, but their higher blood FFA concentrations than the other groups indicated a lower energy status of these dairy heifers.