The effects of dipeptidase inhibitor on peptide breakdown and VFA concentrations in rumen of sheep*

The objectives of this study were to examine the effect of dipeptidase inhibitor (DI) on the concentration of NH3-N, VFA and peptide N in rumen. Four Small-Tailed Han rams fi tted with rumen fi stula were fed with pelleted diets containing DI of different concentrations at four ten-days’ experimental periods, respectively. Rumen fl uids were taken through the fi stula at last two days of each period. The results indicated that DI inhibited the release of NH3-N severely (P<0.05). In addition, the peptide N concentration increased (P<0.05) with the increase of DI concentration. However, the VFA concentrations, except propionic acid, were not affected.


INTRODUCTION
Excessive breakdown of protein to ammonia in the rumen leads to ineffi cient utilization of dietary protein by ruminants, because much of the ammonia released is lost from the fermentation by diffusion across the rumen wall (Leng and Nolan, 1984). It would be benefi cial to the nutrition of the animal by inhibited the degradation process at any of the degradation steps, such as proteolysis, peptide hydrolysis or amino acid breakdown (Witt et al., 1998). Peptides are intermediates in the catabolic process and therefore it might be possible to regulate ammonia production through inhibiting the breakdown of peptides (Wallace, 1996). The breakdown of peptides in the rumen appears to be a two-stage process, whereby oligopeptides are broken down mainly by dipeptidyl peptidases, and the dipeptides released are hydrolysed by separate dipeptidases (Wallace et al., 1993). Many different rumen microorganisms possess dipeptidase activity, Prevotella ruminicola is one of the most active bacteria in the breakdown of proteins and peptides in the rumen (McKain et al., 1992). It is also one of the numerous species, which plays a prominent role in each of the steps in the generation of ammonia from protein, sometimes comprising >60% of the total bacterial population (Van Gylswyk, 1990). The dipeptidase activity of Prevotella ruminicola was found to be extremely sensitive to the metalloprotease inhibitor, 1,10-phenanthroline (Wallace et al., 1995). The aim of the present study was to examine the effects 1,10-phenanthroline, a dipeptidase inhibitor (DI), on peptide breakdown and VFA concentrations in the rumen fl uids of sheep.

Animals and feeding
Four castrated Small-Tailed Han rams with a mean BW of 36.08±1.75 kg were used. All the sheep were fi tted with permanent rumen fi stula (i.d., 55 mm, wall thickness, 2.5 mm), which were made of silicon rubber. The sheep were allowed to recover for two weeks prior to commencement of the study and then were placed in the metabolism crates. Animals were fed 1.1 kg/d pelletted diet with auto-feeder at 1 h intervals, and allowed free access to water.

Experimental diet
The diet was formulated based on the feeding standard of meat-producing sheep and goats (NY/T 816-2004). Diet composition and the nutrient level were showed in the Table 1.

Experimental design
The dipeptase inhibitor (DI) was added into pellets with the concentration of 0, 0.8, 1.0 and 1.2 g/kg, respectively. The experiment was composed of four ten-day's periods. The adaptation of each period was 8 days, then the last 2 days were sampling period in which 50 ml rumen fl uid samples were collected at 4 h intervals. All the samples were stocked under -20°C.

Analysis of samples
Samples of rumen fl uid were processed and analysed for volatile fatty acid (VFA), ammonia and the peptide concentrations. Ammonia nitrogen (NH 3 -N) was analysed by colorimetry as described by Chaney and Marbach (1962). Samples for VFA analysis were prepared as described by Jouany (1982) and analysed by gas chromatography. The peptide nitrogen of rumen fl uids was estimated by the method described by Chen et al. (1987).

Statistical analysis
Data were analysed by the one-way analysis of variance procedure of SAS (1999). The means compared with the method of Duncan's multiple-range test, and the signifi cance level was 0.05.

RESULTS
The DI had signifi cant effect on the production of NH 3 -N in rumen fl uid ( Table  2). The NH 3 -N concentrations of all experimental groups were lower than that of the control group. Especially, the NH 3 -N of group containing 1.0 g/kg DI was signifi cantly decreased 50.27% (P<0.05) compared with that of the control group, but no signifi cance were found with the group of containing 1.2 g/kg DI. The results indicated that DI could inhibit the release of NH 3 -N at a certain concentration. DI had signifi cant effect on the peptide N concentration of rumen fl uid ( Table  2). The peptide N concentration increased signifi cantly as the increasing of DI concentration compared with the control group. The concentration of peptide N of all experimental groups increased by 87.88, 106.06 (P<0.05) and 415.1% (P<0.05), respectively.
The concentration of VFA except propionic acid had no signifi cant changes between different groups (Table 3). The propionic acid concentrations of groups containing 0.8 and 1.2 g/kg DI were signifi cantly declined compared with the control group (P<0.05). The molar proportion of acetic acid of group containing 1.2 g/kg DI was signifi cantly higher (P<0.05), but the propionic acid was lower than in the control group. The rate of C 2 /C 3 of all experimental groups were higher than that of the control group, and the group of adding 1.2 g/kg DI had the highest rate (P<0.05). 3.78 ± 0.30 c 5.13 ± 1.00 ab 4.54 ± 1.23 bc 6.14 ± 1.00 a abc means with different superscripts within lines are signifi cantly different (P<0.05)

DISCUSSION
The activities of the main rumen bacterial species with high dipeptidase activity, Prevotella ruminicola, was inhibited 95.17% by 1,10-phenanthriline (Wallace et al., 1996). When trypticase, a pancreatic casein hydrolysate containing a mixture of oligopeptides, dipeptides and amino acids, was incubated with rumen fl uid, the production of ammonia and free amino groups was inhibited about 71% by 1,10-phenanthroline . Zhang et al. (2003) estimated the infl uence of 1,10-phenanthroline concentrations on in vitro fermentation of mixed ruminal microorganisms, the result indicated that 1,10-phenanthroline inhibited gas production and NH 3 -N concentration severely. In the present experiment, NH 3 -N, VFA and peptide nitrogen concentration in the rumen were measured when different amount of DI were supplemented into diets. The concentration of ammonia, which were products of amino acid breakdown, was lower than the control group, the results of present experiment also indicated that DI inhibited the release of ammonia nitrogen, and the 1.0 g/kg DI group had the lowest NH 3 -N concentration. There were possibilities that the lowered rumen NH 3 -N by DI supplementation may have adverse impact on rumen bacteria growth as some previous studies reveled that NH 3 -N was necessary for effi cient rumen bacteria growth. Schaefer et al. (1980) reported that the optimal rumen NH 3 -N concentration for rumen microbial growth was 1.4 mg/100 ml, and the result of Satter and Slyter (1974) was 5.0 mg/100 ml. However, Mehrez et al. (1977) reported that the optimal rumen NH 3 -N concentration to reach the optimal degradation of the DM of barley by dacron bags method was 19.4 mg/100 ml. The lowest rumen NH 3 -N concentration (16.74 mg/100 ml) was observed when 1.0 g/kg DI was supplemented in the present study. It was slightly lower than the recommendation of Mehrez et al. (1977), but higher than the results of Satter and Slyter (1974) and Schaefer et al. (1980), so DI supplementation may behave no signifi cant infl uence on rumen bacteria growth in the present study. Correspondingly, the peptide N showed signifi cant increase, which indicated apparently that DI inhibited the breakdown of peptide. Therefore, DI inhibited the breakdown of peptides and the dipeptidase actives of rumen microorganisms, and could be used as a means of controlling ammonia production from peptides in the rumen.
In present experiment, with the adding of DI, the propionic acid concentration decreased, it maybe related to the decrease of ammonia that was the substrate of some kinds of starch digestive bacteria, such us Ruminobacter amylophilus and Prevotella ruminicola, and these bacteria was mainly responsible for the produce of propionic acid in rumen. Moreover, the further research was needed to obtain the exact mechanism.
CONCLUSIONS It was concluded that dipeptidase inhibitor (DI) inhibited the release of ammonia from peptide and improved the production of peptides in rumen with the increase of DI. When DI concentration was higher than 1.0 g/kg in diet, the effect of DI in controlling the breakdown of peptide was effective. The VFA concentrations, except the propionic acid that was decreased, were not affected.