Preliminary study of the effects of condensed barley distillers soluble on rumen fermentation and plasma metabolites in Japanese Black cows

Animals and Treatments

The animals used in this study were treated according to the Guidelines for Animal Experiments in the Faculty of Agriculture of Kyushu University (Fukuoka, Japan) and to the laws of the Japanese Government (Law No. 105 with notification No. 6).

Japanese Black cows (n = 9, non-pregnant cows, mean age 13.7 ± 2.5 years, initial mean body weight 546 ± 66 kg) were used in a replicated 3 × 3 Latin square design experiment over a period of 35 days. The cows were housed in individual stalls with free access to water and offered feed twice daily (at 09.00 and 16.00 hours). The experimental feeds were prepared based on the chemical composition and nutrients of CBDS (Table 1). One of the three treatments was feeding soybean meal to provide the same intake of crude protein as CBDS. Overall, three treatments were set up: 100% hay (Hay-t), 20% CBDS and 80% hay (CBDS-t), and 17.2% soybean meal and 82.8% hay (Soybean-t). The Japanese Feeding Standard for Beef Cattle (Agriculture, Forestry and Fisheries Research Council Secretariat (AFFRCS) 2008) recommends a protein content of 12% in feed for Japanese Black cattle, but considering the practical loss of protein, it also recommends adding more protein to compensate. Therefore, we provided a protein content of 14.4% in both the CBDS-t and Soybean-t feeds. The quantity of feeds was determined by the dry matter (DM) requirements for maintenance of the Japanese Feeding Standard for Beef cattle (Agriculture, Forestry and Fisheries Research Council Secretariat (AFFRCS) 2008). The daily quantity of feed was set at 1.22–1.35% of body weight. CBDS was fed after mixing with hay. The soybean meal was fed by top-dressing the hay. The CBDS was produced at the Haita Green Bio Office of the Sanwa Shurui Corporation (Usa, Japan). The hay was prepared mainly from orchard grass. The feed offered and refused was measured and recorded daily during the last 14 days of each period to calculate DM intake.

Sample Collection

Ruminal fluids were collected via the mouth using a rumen catheter (Sanshin Industrial, Yokohama, Japan) on day 35 of each period just before the morning feeding and 3 h later according to data from Okamoto (1979). The pH value of the ruminal fluid was measured using a glass electrode pH meter (HM-40 V; DKK-TOA Corp., Tokyo, Japan) immediately after sample collection. The ruminal fluids were filtered to remove feed particles using four layers of gauze and then centrifuged at 1200 × g for 10 min at 4°C. To remove the protein, perchloric acid solution was added to the supernatant at a proportion of 5%. The resultant fluid samples were then stored at −30°C until later analysis according to a previously reported procedure (Hosoda et al. 2012). Blood samples were collected from the jugular vein using heparinized tubes immediately and 3 h after the morning feeding on day 35 of each period. The samples were immediately centrifuged at 1500 × g for 10 min at 4°C, and the plasma was stored at −30°C until later analysis.

Chemical Analysis

The feed samples were dried in an oven at 60°C and ground to pass through a 1-mm screen. The air-dried samples were dried at 135°C for 2 h to determine DM. The crude protein (CP), ether extract (EE), crude fiber (CF) and crude ash (CA) of the samples were analyzed according to the methods of the Association of Official Analytical Chemists (AOAC) (2000). The neutral detergent fiber (NDF) content was determined using a heat-stable amylase, sodium sulfite and expressed exclusive of residual ash (Van Soest et al. 1991). The degraded intake crude protein (DIP) and soluble intake crude protein (SIP) contents were determined using the methods of Roe et al. (1991) and Krishnamoorthy et al. (1982), respectively. The non-fibrous carbohydrate (NFC) content was calculated using the formula: NFC = 100 – CP – EE – CA –NDF. The organic acid contents were determined using high-pressure liquid chromatography (Wahyono et al. 2015). The samples were stored at −30°C until later analysis. Before analysis for the organic acid, the CBDS was diluted with deionized water and deproteinized. The soybean meal was also extracted with deionized water and deproteinized. The supernatants were passed through a 0.45-µm filter under pressure. Using the same samples, the glycerol content was determined using a high-pressure liquid chromatograph (Takano et al. 2014) equipped with a Shodex Asahipak NH2P-50G 4A guard column (Showa Denko, Tokyo, Japan), a Shodex Asahipak NH2P-50 4E (4.6 mm × 250 mm) separation column and a refractive index detector. For separation, the column oven temperature was set at 50°C with isocratic elution being achieved using a mixture of acetonitrile and water (75:25 v/v) at a flow rate of 1.0 mL/min. The samples of deproteinized ruminal fluid were neutralized with potassium hydroxide solution and centrifuged at 1200 × g for 10 min. The ammonia concentrations were determined by the steam distillation method. The neutralized ruminal fluid was shaken with a cation exchange resin (Amberlite, IR 120B H AG; Organo Corporation, Tokyo, Japan) then centrifuged at 7700 × g for 3 min. The supernatants were passed through a 0.45 µm filter under pressure and the filtrates were used to determine the content of volatile fatty acids (VFA) using high-pressure liquid chromatography (Hosoda et al. 2005). The β-hydroxybutyric acid (BHBA) content of the plasma was determined by colorimetric analysis (DTX-880; Beckman Coulter, Tokyo, Japan) using the β-hydroxybutyrate Ketone Body Assay Kit (Cayman Chemical Co., Ann Arbor, MI, USA). The blood urea nitrogen (BUN), NH₃, total protein, total cholesterol and glucose contents in the plasma were determined using an automatic biochemical analyzer (Dri-Chem; Fujifilm, Tokyo, Japan).

Statistical Analysis

The data on all variables were analyzed using R version 3.1.2 (The R Foundation for Statistical Computing, Vienna, Austria). The data were analyzed by analysis of variance with treatment, experimental period and cow as factors. The differences in treatment means were tested using Tukey's multiple range tests. The differences in sampling times of the same treatment were tested using the paired t-test. The significance level for differences was set at P < 0.05. Data are shown as means and standard error of the means.

Nutrient Characteristics of CBDS as a Feed

Soybean meal is the most popular protein additive in cattle feed. The protein content of CBDS, 42.5% (DM basis), was generally high (Table 1). However, compared with soybean meal, 51.2% (DM basis), CBDS contains a relatively low level of crude protein (Table 1). CBDS exhibits higher values of DIP, SIP contained in CP, ether extract and NFC than soybean meal (Table 1). The total amounts of organic acids (citric, malic, succinic, lactic, formic and acetic acids) in the CBDS and soybean meal were 18.39% and 3.43% (DM basis), respectively (Table 1). The glycerol content of CBDS, 7.27% (DM basis), was higher than that of soybean meal, 0.05% (DM basis) (Table 1). This means that CBDS offers a higher potential and availability of protein and energy compared with soybean meal. The intake of hay was significantly lower in the CBDS-t and Soybean-t than in the Hay groups (P < 0.05) (Table 2). The intake of CP was significantly higher in the CBDS-t and Soybean-t than in the Hay-t groups (P < 0.05) (Table 2). The intake of NDF was significantly lower in the CBDS-t and Soybean-t than in the Hay-t groups (P < 0.05) (Table 2). The highest intake of NFC was observed in the CBDS-t group, followed by the Soybean-t group, with the Hay-t group having the lowest intake (P < 0.05) (Table 2). The total DM intake was not significantly different between treatments (Table 2). The cattle body weight was not significantly different between treatments (Table 2). This means that feed with 20% CBDS did not affect DM intake under the restricted feeding condition.

Relationship Between Protein Degradability, Ruminal Ammonia-N, and Plasma Metabolites and Ruminal pH

Okamoto (1979) reported that when sheep were fed hay and barley, the concentration of total VFA in the rumen was reached a maximum 3 h after feeding and it decreased slowly until 12 h after feeding. In other studies, in ruminant fed a forage-based diet, the total VFA reached a maximum value after 1.5–4 h (Sutoh et al. 1991; Carro et al. 2009; Zhang et al. 2012). The BHBA, which is converted from butyrate in the rumen epithelium, was higher between 2 and 3 h after feeding, or between 2 and 4 h in multiparous dairy cows (Blum et al. 2000; Nikkhah et al. 2008). Therefore, in the present study, we collected ruminal fluids and plasma just before the morning feeding and 3 h later.

Ruminal pH was not affected by the dietary treatments (Table 3). The highest initial concentration (just before the morning feeding) of ruminal ammonia-N was observed in the Soybean-t group, while the Hay-t group exhibited the lowest concentration (P < 0.05) (Table 3). The highest concentration of ruminal ammonia-N after 3 h was observed in the CBDS-t group, whereas the Hay-t group exhibited the lowest concentration (Table 3). The concentration of ammonia-N in the rumen after 3 h in the CBDS-t group was significantly higher than initially (P < 0.05) (Table 3). In the Soybean-t and Hay-t groups, the changes in the concentrations of ammonia-N in the rumen from initially to after 3 h were not significantly different (Table 3).

In the present study, the intake of CP in the CBDS-t group was similar to that in the Soybean-t group, but the intake of the degraded intake CP in the CBDS-t group was significantly higher than that in the Soybean-t group (P < 0.05) (Table 2). The CP of CBDS consists of 92.7% SIP and is highly soluble in the rumen (Table 1). Therefore, the CP of CBDS would be rapidly degraded after feeding. In the CBDS-t group, the concentration of ammonia-N in the rumen after 3 h rose by about 2.5-fold of that initially (Table 3). Regarding the in vitro ruminal degradation parameters of feed protein, for soybean meal and hay, respectively, the rapidly soluble fractions (a fraction) were 16 and 23%, the slowly degradable fractions (b fraction) were 82 and 58%, and the rate constant of disappearance for the b fractions (c fraction) were 9 and 5%/h (Agriculture, Forestry and Fisheries Research Council Secretariat (AFFRCS) 2008). Therefore, because a fraction values of soybean meal and hay were less and the ruminal degradation of the protein was slow, the peak concentrations of ammonia-N in the rumen in the Soybean-t and the Hay-t groups might also be low.

The initial BUN concentrations in the plasma and after 3 h were significantly higher in the CBDS-t and Soybean-t groups than in the Hay-t group (P < 0.05) (Table 4). There was no significant difference in the initial BUN concentration in the plasma and after 3 h between the CBDS-t and Soybean-t groups (Table 4). The BUN concentrations in the plasma after 3 h were higher than initially for all treatments (P < 0.05) (Table 4). There was no significant difference in NH₃ concentration in the plasma between treatments (Table 4).

The concentrations of BUN were highly correlated with the concentration of ammonia-N produced in the rumen (Wanapat et al. 2008). Furthermore, the BUN concentration could reflect nitrogen losses from rumen fermentation (Xin et al. 2010). This suggests that although the concentration of ammonia-N in the rumen after 3 h was significantly higher in the CBDS-t group (P < 0.05) (Table 3), the nitrogen loss in the CBDS-t group was probably not different from the Soybean-t group. Although the concentrations of BUN after 3 h were significantly higher than initially for all treatments (P < 0.05) (Table 4), the concentrations of ammonia-N in the rumen were not significantly different between the initial reading and after 3 h in the Soybean-t and Hay-t groups (Table 3). The peak concentration of BUN occurred about 1–2 h later compared with those of ammonia-N in the rumen (Sutoh et al. 1991; Cherdthong et al. 2011). Therefore, the peak concentration of ammonia-N in the rumen might have occurred within 3 h after feeding. This suggests that although the differences in the degradability of protein in the rumen between CBDS and soybean meal affected the production of ammonia-N in the rumen, it did not affect BUN and NH₃ in the plasma.

Ruminal VFA Concentration

The initial total VFA concentration was significantly lower in the CBDS-t than in the Soybean-t groups (P < 0.05), but the total VFA concentration after 3 h was not significantly different (Table 3). The initial acetate concentration was significantly lower in the CBDS-t than in the Soybean-t groups (P < 0.05), but the concentration after 3 h was not significantly different (Table 3). The Hay-t group showed no significant differences from the CBDS-t and Soybean-t groups in the concentrations of total VFA and acetate (Table 3). In contrast, although the initial concentrations of propionate and butyrate were significantly lower in the CBDS-t than in the Soybean-t groups (P < 0.05), after 3 h they were significantly higher in the CBDS-t than in the Soybean-t groups (P < 0.05) (Table 3). This suggests that the CBDS degraded rapidly in the rumen after feeding and the concentrations of propionate and butyrate increased, but that the VFA products did not last long compared with the diet containing soybean meal. Although the initial compositions of acetate, propionate and butyrate (% mol) in the rumen were not significantly different between the CBDS-t and the Soybean-t group, the composition of acetate after 3 h was lower in the CBDS-t than in the Soybean-t group and the compositions of propionate and butyrate after 3 h were significantly higher in the CBDS-t than in the Soybean-t group (P < 0.05) (Table 3). This suggests that the feeding CBDS alters the production ratios of VFA in the rumen.

Relationship Between NFC Configuration, Protein Degradability and Ruminal VFA Concentration

The largest intake of NFC was observed in the CBDS-t group, followed by the Soybean-t group, with the Hay-t group showing the smallest intake (Table 2). NFC consists of starch, sugars, water-soluble fibers and organic acids. When ruminants have been fed a large amount of concentrated feeds or feeds containing a large amount of starch, the concentration of total VFA in the rumen has been observed to be higher (Ørskov 1986; Marshall et al. 1992). Hattori et al. (2010) reported that the mono- and oligosaccharide content of CBDS was 9.0% (DM basis), composed of arabinose (7.4% DM basis), xylose (1.2%), sucrose (0.2%) and glucose (0.2%). Arabinose and xylose are the components of the cytoplasm. During the manufacturing process of shochu, Aspergillus is used for the glycation of starch. Therefore, the starch content of CBDS would be lower. The starch content of soybean meal was also less (Sniffen et al. 1992). In soybean meal, the oligosaccharides are sucrose (6.9–8.7%), stachyose (5.3–5.9%) and raffinose (0.6–1.2%) (Van Kempen et al. 2006). Therefore, although the CBDS was rich in organic acids, especially lactic acid and glycerol, the sugars of CBDS might contain less than soybean meal.

Although the initial concentrations of propionate and butyrate were not significantly different between the CBDS-t and Hay-t groups, after 3 h they were significantly higher in the CBDS-t than in the Hay-t group (P < 0.05) (Table 3). However, although the intake of NFC was larger in the Soybean-t than in the Hay-t group, the initial concentrations of propionate and butyrate and after 3 h were not significantly different (Table 3). The initial compositions (% mol) of propionate and butyrate were also not significantly different between treatments, but after 3 h they were significantly higher in the CBDS-t than in the Soybean-t and Hay-t groups (P < 0.05) (Table 3). The concentrations of lactic and citric acids in CBDS were higher and lower, respectively, than those in soybean meal (Table 1). The infusion of lactic acid in the rumen has been shown to increase the concentrations of propionate and butyrate but decrease the concentration of acetate (Terashima et al. 1976). Glycerol supplementation also increased the total VFA, propionate and butyrate in the rumen (Wang et al. 2009b). This suggests that the larger amounts of lactic acid and glycerol in CBDS increased its ability to produce propionate and butyrate in the rumen compared with soybean meal. Although the concentration of acetic acid in the rumen increased linearly with increasing citric acid supplementation, the concentrations of propionate and butyrate in the rumen did not increase (Wang et al. 2009a). In the present study, the intake of citric acid from soybean was lower than the lowest amount of citric acid supplementation in the study of Wang et al. (2009a). Therefore, citric acid in soybean meal might not have increased the concentration of acetic acid in the Soybean-t group.

The total VFA concentration in the Soybean-t group was not significantly different between the initial values and those after 3 h (Table 3). Because the NDF content in the hay was high (71.8%, DM basis), the digestibility of the hay would be low. Nitrogen supplementation has been shown to increase the digestibility coefficient of DM and NDF in cattle fed low-quality tropical forage (Souza et al. 2010). The stimulatory effect of nitrogen supplementation on NDF degradation is because of the improved rumen conditions increasing the availability of nitrogenous compounds for the growth of fibrolytic microorganisms (Detmann et al. 2009). The initial concentration of ammonia-N in the rumen in the Soybean-t group was higher than in the CBDS-t and Hay-t groups (Table 3). Therefore, because the microbial activity before feeding in the morning might be higher in the Soybean-t than in the CBDS-t and Hay-t groups, the total VFA concentration might also be significantly higher in the Soybean-t than in the CBDS-t and Hay-t groups (Table 3). On the other hand, because after 3 h the concentration of ammonia-N in the rumen in the CBDS-t was significantly higher than initially, NDF degradation in the CBDS-t might be higher and the acetate concentration after 3 h in the rumen in the CBDS-t might be significantly higher than initially (Table 3). This suggests that not with only lactic acid and glycerol, the difference in the degradability of protein in the rumen might cause a difference in the ruminal VFA production between CBDS and soybean meal.

Relationship Between Rumen VFA and Plasma BHBA and Glucose

The initial BHBA concentration in plasma was not significantly different between treatments (Table 4). After 3 h, it was significantly higher in the CBDS-t than in the Soybean-t and Hay-t groups (P < 0.05) (Table 4). Increasing the rate of intraruminal production of VFA has been shown to promote the proliferation of epithelial cells in the rumen. When butyrate production increases rapidly, the proliferation of rumen epithelial cells accelerates (Sakata & Tamate 1978). Papillary growth in the rumen is also stimulated by the administration of propionate and butyrate (Tamate et al. 1962). Plasma BHBA is a measure of rumen epithelial metabolic activity and is produced from butyrate as it passes through the rumen wall (Lane et al. 2000). Weigand et al. (1975) have also reported that 26–33% of butyrate absorbed by the rumen papillae was converted to BHBA. This suggests that the butyrate level increased rapidly in the rumen through feeding CBDS; the BHBA concentration also increased, after which the metabolite in the rumen villus would increase. Weigand et al. (1975) also reported that 18–24% of the valerate absorbed by the rumen papillae was converted to BHBA. In the present study, although the initial concentration of valerate was not significantly different between treatments, after 3 h it was significantly higher in the CBDS-t than in the Soybean-t and Hay-t groups (P < 0.05) (Table 3). This suggests that the high concentration of valerate in the rumen through feeding CBDS also led to the high concentration of BHBA in the plasma.

The concentrations of total protein and total cholesterol in the plasma were not significantly different between treatments (Table 4). The initial glucose concentrations in the plasma and after 3 h were not significantly different between treatments (Table 4). Sutoh et al. (1991) have reported that the concentration of propionic acid in the rumen fluid or in plasma was not directly related to the plasma glucose level, a similar result to the present study. Butyrate has been shown to significantly inhibit hepatic propionate use and conversion to glucose in sheep, goat and calf hepatocytes (Aiello et al. 1989). Although butyrate inhibits propionate utilization in the liver, greater than physiological quantities of butyrate may be necessary to cause that inhibition (Krehbiel et al. 1992). Although increasing the level of dietary lactose increased the ruminal butyrate and plasma BHBA levels and decreased the concentrations of glucose in plasma, the changes in plasma glucose and BHBA were not as great (Defrain et al. 2004). In the present study, the production of ruminal butyrate might not be great enough to affect glucose use between treatments as suggested by Krehbiel et al. (1992) and Defrain et al. (2004). However, the concentration of glucose in the plasma after 3 h in the CBDS-t group was lower than the initial concentration (P < 0.05) (Table 4). Because the BHBA was utilized as energy, a rapid increase of butyrate in the rumen by feeding CBDS might inhibit propionate use and convert to glucose in the liver.

Conclusion

Feeding CBDS to cows, which contains high levels of NFC, especially lactic acid and glycerol, can produce more propionate and butyrate in the VFA produced in the rumen. Consequently, feeding CBDS leads to a higher BHBA concentration in the plasma than by feeding soybean meal. Taken together, the present study suggests that feeding CBDS promotes rumen fermentation and butyrate metabolism.