Corn and soybean meal were used as a carbohydrate and a protein source, respectively. All feedstuffs were dried at 60°C and ground in a Wiley mill (Arthur H. Thomas, Philadelphia, PA, USA) through a 1 mm screen prior to enzyme and formaldehyde treatment. Enzymes and formaldehyde were purchased from Sigma Co. Ltd. (Sigma & Aldrich, St. Louis, USA). Corn was consecutively treated with alcalase (0.2%/wt) and α-amylase (1%/wt) and incubated for 4 h at 55°C to produce enzyme-treated corn (EC). Soybean meal was also treated with alcalase (0.2%/wt) and formaldehyde (0.5%/wt) to make enzyme-treated meal (ES) and formaldehyde-treated meal (FS). The incubation time and temperature of the S treatment was same as in the C process. In total, five treated experimental diets were prepared.

An in vivo trial was conducted to examine the degradation kinetics of each experimental diet. Ruminal dry matter (DM), organic matter (OM) and crude protein (CP) disappearance rates of each treatment feed were measured using the nylon bag technique as described by Ørskov et al. (1980). Three rumen-cannulated Holstein steers, weighing 350±18.77 kg, were used for the in situ experiment. Experimental animals were offered Timothy hay and a commercial concentrate feed (Cargill Agribrand Purina Korea Co. Ltd.) at a ratio of 6:4 twice daily (at 8 am and pm). The total amount of feed offered to animals was 2% of their body weight. The chemical composition of the commercial concentrate was 87.7% DM, 14.8% CP, 4.13% ether extract (EE), 5.22% ash and 71.5% total digestible nutrient DM basis. Fresh water and a mineral mixture were available to each animal at all times.

A total of 15 g of each experimental diet (C, EC, S, ES, FS) was weighed and allotted to nylon bags with a 53±10 μm (Bar Diamond, Idaho, USA) pore size. The bags were placed into the ventral sac of the rumen of each Holstein steer through the rumen cannula. The incubation was started at 8:30 am after feeding. All feed samples were incubated for 0, 3, 6, 9, 12, 24 and 48 h. After incubation, the bags were removed from the rumen and washed by hand under cold tap water to remove the feed particle attached to the surface of the bags until the rinse water ran clear. Control bags which were not incubated in the rumen (0 h) were washed in the same way. After washing, all nylon bags were dried at 65°C for 48 h and weighed to estimate the DM residues. The contents of the bags were stored at 4°C until chemical analysis.

The disappearance rates of DM, OM and CP were calculated as the difference between the weight of the bag contents before and after incubation. The degradability data obtained for each treatment were fitted using PROC NLIN (SAS, 1996) to the following equation (Ørskov and Mcdonald, 1979):

where P is the degradation rate at time ‘t’; a is the rapidly soluble fraction; b is the insoluble but potentially degradable fraction; c is the rate of degradation of fraction b; t is the incubation time.

The effective degradabilities of DM, OM and CP were calculated using the above parameters (a, b, c) and a ruminal outflow rate of 0.02, 0.05 and 0.08%/h by:

Where P is the effective degradability of DM, OM and CP, and k is the estimated rumen outflow rate (%/h).

Two carbohydrate samples (C, EC) and three protein samples (S, ES and FS) were prepared as described previously. Carbohydrate and protein samples were mixed in a ratio of 0.85:0.15 for a total of six treatment diets (CS, CES, CFS, ECS, ECES and ECFS) for the in vitro experiment to get chemical composition which is similar with commercial concentrated feed. The SI of the diets in the in vitro study was calculated from the degradation kinetics of OM and CP of the diet components from the in situ experiments. The SI of N to OM was calculated as described by Sinclair et al. (1993).

For the in vitro experiment, rumen fluid was collected and mixed two to three hours after morning feeding from three Holstein steers that were fed with the same feeds as used in in situ experiment. The strained rumen fluid was filtered through eight layers of cheesecloth and transported to the laboratory in a pre-warmed thermo-flask. It was then mixed with artificial saliva (McDougall, 1948) at a ratio of 1:2 under a CO₂ stream. Twenty-five milliliters of rumen fluid and buffer mixture was dispensed anaerobically into 50 ml serum bottles, each containing 1 g of experimental diet. The bottles were then filled with CO₂ gas and capped with a rubber stopper. The bottles were incubated for 0, 3, 6, 12 and 24 h in a 39°C incubator.

After fermentation, bottle contents were filtered through Whatman No. 541 paper (Fisher Scientific Company, Pittsburgh, PA, USA), and the residues were dried at 65°C for 48 h to measure DM digestibility. The filtrate was centrifuged at 10,000 g for 20 min at 4°C, and the supernatant was collected to determine NH₃-N and VFA concentration. The pH of the filtrate was measured immediately after centrifugation using a pH meter. After filtering the supernatant through a Millipore filter (0.22 μm pore size), VFA concentration was measured by gas chromatography using a Hewlett Packard 5880A gas chromatograph (Hewlett Packard, Palo Alto, Ca, USA) employing a method described by Erwin (1961). The concentration of NH₃-N was determined using a modified colorimetric method (Chaney and Marbach, 1962). The in vitro digesta samples were analyzed for purine content (Zinn and Owens, 1986) using HPLC (HP1100, Hewlett Packard, Palo Alto, Ca, USA) according to the modified method (Makkar and Becker, 1999; George et al., 2006) to estimate MPS. All feed components and in situ and in vitro residues were ground in a Wiley mill (Arther H. Thomas, Philadelphia, PA, USA) and analyzed for dry matter (DM), ash, fiber, and nitrogen contents using the Kjeldahl procedure with a Kjeltec (KjeltecTM 2200, Foss Tecator, Sweden) and for EE using diethyl ether with a Soxhlet as described in AOAC (1990). Neutral detergent fiber (NDF) and acid detergent fiber (ADF) were analyzed by the method described by Van Soest et al. (1991) with an Ankom 200 fiber analyzer (Ankom Technology, NY, USA).

The data for the in situ disappearance characteristics of the experimental diets were analyzed using PROC GLM (SAS, 1996). Differences between treatments were considered significant at p<0.05. The data for DM digestibility, pH variation, NH₃-N, VFA, and purine concentrations obtained from in vitro experiments were analyzed using a 2×3 factorial design using the MIXED procedure (SAS, 1996) with the following model:

Where:

The effects of C treatment, S treatment and their interaction were considered fixed. Significant differences (p<0.05) in treatment least square means were reported only if the Tukey-test (SAS, 1996) was also significant (p<0.05).

The chemical compositions of the test feeds are listed in Table 1. Although EC was treated with α-amylase and alcalase, the chemical composition of this feed was not influenced by the enzyme treatment. Similarly, treatment of soybean meal with enzyme or formaldehyde did not alter its chemical composition. However, the ruminal degradation characteristics of C and S were greatly influenced by enzyme and formaldehyde treatment (Table 2). ES showed higher (p<0.05) effective degradability of DM, CP, and OM than S, but FS had the lowest (p<0.05) effective degradability of DM, OM, and CP. When C and S were treated with enzyme, their ‘a’ fraction of DM, OM and CP increased, while the treatment of S with formaldehyde resulted in a lower (p<0.05) immediately soluble fraction ‘a’ and a lower degradation rate of the ‘b’ fraction compared with untreated S (Table 2). The potentially degradable P fraction (a+b) varied between 66.94% for S and 100% for FS. In a number of previously described experiments, the protein source was treated with formaldehyde to protect dietary protein from proteolysis by rumen microorganisms in an effort to increase N utilization efficiency in the rumen. For instance, Witt et al. (1999b) reported that formaldehyde treatment of S reduced not only the soluble N content but also the rate of degradation of the ‘b’ fraction. In our experiment, formaldehyde treatment increased the portion of ‘b’ fraction but the degradation rate of ‘b’ was decreased by chemical treatment. Therefore the ED value of DM, CP and OM in FS were lower than those of S significantly (p<0.05). Eghbali et al. (2011) examined the effects of different treatment methods for canola meal on rumen degradability of CP and DM in in situ studies. The protein fraction was measured based on Cornell Net Carbohydrate and Protein system. Treating with formaldehyde also reduced the protein degradation rate in the rumen in this study. Our results are also consistent with those of Eghbali et al. (2011), who suggested that aldehydes could crosslink with free amino acid groups in feeds so that protein degradability in the rumen might be reduced, but these complexes could be hydrolyzed in the abomasum and small intestine thus enabling protein digestion and absorption in the hindgut.

Exchange of feed ingredients (Herrera-Saldana et al., 1990; Shabi et al., 1998; Rotger et al., 2006) and adjustment of ingredient proportions (Sinclair et al., 1993; Witt et al., 1999a;b; Witt et al., 2000; Richardson et al., 2003; Ichinohe and Fujihara, 2008; Seo et al., 2010) are methods to manipulate the degree of synchronization of carbohydrate and nitrogen release in the rumen. In present experiment, we used the same carbohydrate and protein source treated with enzymes or formaldehyde and mixed at the same ratio to exclude the influence of the characteristics of different feeds on synchronization effects. As mentioned above, 0.85 g of carbohydrate source (C, EC) and 0.15 g of protein source (S, ES, FS) were mixed in order to vary synchronicity in the rumen and were used in in vitro experiment. In total, six kinds of experimental feeds were prepared, and the results of their proximate analysis and their fiber contents are presented in Table 3. All six diets had similar OM, CP, EE, NDF and ADF contents; the average proportion of each was 98.17%, 17.55%, 3.59%, 13.34% and 5.77% of DM, respectively. However, the SI of each experimental diet varied between 0.71 in the CFS group and 0.95 in the CES group. Feeds mixed with enzyme-treated carbohydrate or protein source (ECS, CES) had the highest SI, while the groups including formaldehyde treatment group (CFS, ECFS) showed lower SI, which may have been due insufficient N release of FS in the rumen.

All six experimental diets were used in the in vitro experiment to evaluate the effects of synchronicity on rumen fermentation characteristics and MPS. The in vitro DM digestibility, pH, and ammonia-N concentration as influenced by different SI are presented in Table 4. The DM digestibility of all treatments increased as incubation time progressed from 0 h to 24 h. A significant interaction between C and S treatment was detected at 24 h incubation for DM digestibility (p<0.01). CFS and ECFS with lower SI resulted in lower (p<0.01) digestibility than those of high and intermediate SI groups at all incubation times, and highly synchronous diets had the greatest (p<0.01) DM digestibility within the C groups. However, the degree of synchronicity did not influence digestibility within the EC groups. The results of the present study suggest that the rapidly fermentable fraction of the diet in the rumen is more important for ruminal digestion than the degree of synchronicity. Herrera-Saldana et al. (1990) also observed that barley-containing diets, which are more rapidly degraded than corn, had higher DM, OM, CP and starch digestibility than corn-based diets, regardless of the synchronicity of the carbohydrate and protein degradation rates. Since we used highly fermentable carbohydrate and protein sources in our experimental diets, the pH was rapidly dropped after 6 h of incubation in all treatments. Both treatments of C and S influenced the medium pH at most incubation times, with a more noticeable effect after 6 h, but the interaction between the treatments was observed only at 24 h of incubation. EC-containing diets had a lower (p<0.05) ruminal pH than those with untreated C as the carbohydrate source at 6 and 12 h incubation, indicating that rumen microbes degrade EC more efficiently than C. FS containing diets showed higher (p<0.01) ruminal pH irrespective of carbohydrate source after 6 h incubation.

The ammonia-N concentrations of CS, CES, ECS, and ECES diets increased as incubation progressed. Ammonia-N concentration has been reported to increase immediately after feeding for 2 to 3 h and decrease until the next feeding in other in vivo studies (Chumpawadee et al., 2006; Seo et al., 2010). Our data showed the opposite result, with the above-mentioned in vivo ammonia-N concentration increasing consistently, which may be an indication of poor utilization of ammonia for MPS or may be due to decreased absorption of ammonia from the rumen as no absorption into the body is possible under in vitro conditions. We used a batch culture in vitro system in which the end product of fermentation could not flow out, and thus ammonia-N was overproduced and accumulated continuously. FS-containing diets had lower (p<0.01) ammonia-N content than S- and ES-containing diets at all incubation times. This suggests that FS was not degraded to a significant extent by ruminal microbes. At 12 h and 24 h of incubation, EC diets showed a lower (p<0.05) ammonia-N concentration than C-containing diets, regardless the degree of synchronicity, indicating that efficient utilization of ammonia-N for MPS was achieved by ruminal microorganisms only when EC was offered as a carbohydrate source. The purine contents of all diets at 12 h of incubation were also confirmed the relationship between ammonia-N utilization and MPS, since the measurement of purine base production is an indirect method for evaluating MPS in the rumen (Makkar and Becker, 1999) (Table 5). Our data support the theory that lower ruminal NH₃-N concentration correlates with higher utilization of NH₃-N for MPS. There were no dietary treatment effects on purine base contents at 0 h of incubation. However, treatment effects of C and S and their interaction were significant (p<0.01) at 12 h of incubation (Table 5). When C was used as an energy source, purine base concentration increased as the diets were more synchronized; this effect was not observed when EC was an energy source. ECES had the highest (p<0.01) purine base content of all the diets, although it had only intermediate synchronicity of energy and protein supply in the rumen. These data imply that the supply of a highly fermentable energy source, rather than the synchronicity of energy and protein supply, was a more influential factor for stimulating MPS from ammonia-N (Hoover and Stokes, 1991). Rotger et al. (2006) have also concluded that rumen fermentation may be more limited by carbohydrate or protein availability than by the degree of synchronicity. A sufficient energy supply from a rapidly fermentable carbohydrate source may allow for the efficient incorporation of peptides, free amino acids, and/or ammonia-N into microbial cells, thereby reducing ammonia-N concentration in the rumen and increasing MP. Hristov et al. (1997) confirmed the effects of different levels of carbohydrate and ammonia availability on the utilization of ammonia and alpha-amino N by ruminal mixed microorganisms in vitro. They observed an increased uptake and incorporation of ammonia into microbial cells with increasing carbohydrate level regardless of the type of N compounds. Theoretically, synchronization of energy and N release in the rumen can increase MPS, N utilization, and rumen fermentation efficiency. However, rumen microorganisms can respond to a deficiency of nutrients by synthesizing intracellular storage polysaccharides when energy sources are abundant, or by utilizing cellular storage and recycled urea N when carbohydrate and N sources are scarce (Valkeners et al., 2004; Ichinohe and Fujihara, 2008).

The VFA concentrations for each treatment after 12 h of incubation are shown in Table 5. There were significant C and S treatment effects, and the interaction of effects on VFA concentrations was also significant. Similar to purine concentrations, total VFA production and individual VFA concentrations of the groups using EC as an energy source was higher (p<0.01) than in other groups (CS, CES and CFS). These parameters were not affected by the degree of synchronicity, as was seen with the purine base concentrations. All mixtures containing ECES had the highest values than any other treatments, in spite of having an intermediate SI value.

In summary, synchronization had positive effects on rumen fermentation, DM digestibility, and MPS when these were compared between low synchronous (CFS, ECFS) versus intermediate (CS, ECES) or high synchronous feeds (CES, ECS). However, there were no consistent results when we compared rumen fermentation characteristics between intermediate and high synchronous feeds. Meanwhile, rapidly fermentable energy and protein sources showed higher DM digestibility, VFA production, ammonia-N utilization, and purine base contents, indicating that the availability of energy or the protein source should be also considered as the other important factor for improving rumen fermentation and MPS.