TEXT

Top Abstract Text References

Ruminant animals are exceptionally inefficient in their retention of nitrogen under conditions where the diet contains protein which can be fermented rapidly in the rumen (8). Protein is converted via peptides and amino acids to NH₃, which can be incorporated by ruminal microorganisms, but which, when present in excess, is absorbed rapidly across the ruminal wall (9). Proteolysis is carried out by many ruminal species (10, 19), and large variation between individual animals occurs (5). Peptide catabolism is for the most part a two-step process, whereby oligopeptides are first cleaved to dipeptides and then broken down by dipeptidases to amino acids (5, 15). The released amino acids are then deaminated by ciliate protozoa (16) and two categories of ruminal bacteria: the first category is comprised of many of the most numerous species of ruminal bacteria, which possess a low deaminative activity, and the second category is comprised of a much smaller population of so-called hyper-ammonia-producing (HAP) bacteria (1), which have much higher deaminative activity and which are believed to rely on amino acids rather than sugars for energy generation (1, 11-13). Inhibition of any of these catabolic steps would benefit nitrogen retention in ruminal fermentation. The aims of this study were to determine the site of action of 1-[(E)-2-(2-methyl-4-nitrophenyl)diaz-1-enyl]pyrrolidine-2-carboxylic acid (LY29) which, when added to steer rations, appeared to enhance nitrogen flow from the rumen (21), and to compare its antimicrobial effects with those of a known deaminase inhibitor, diphenyliodonium chloride (DIC) (3, 4).

LY29 was synthesized by the reaction of diazotized 2-methyl-4-nitroaniline with L-proline in alkaline solution. Thirty grams of 2-methyl-4-nitroaniline was added to an ice-cold solution of 75 ml of concentrated HCl in 400 ml of distilled water. Sodium nitrite was added as 100 ml of an ice-cold 30% aqueous solution, followed by 1 g of diatomaceous earth. The solution of diazotized 2-methyl-4-nitroaniline was filtered through Whatman no. 1 filter paper and kept on ice. One hundred and twenty-five grams of NaCO₃ and 75 g of NaCl were added to 450 ml of ice-cold water, followed by 42.6 g of L-proline, and then the solution of diazotized 2-methyl-4-nitroaniline was added dropwise, with constant mixing, over a period of 30 min. The solution was then taken to pH 4.5 by adding HCl, and the precipitate was collected by filtration and freeze-dried. Its structure was confirmed by matrix-assisted laser desorption ionization-time-of-flight mass spectrometry. The experiments described here were done with a single batch of LY29 provided by Elanco Animal Health, Indianapolis, Ind. DIC was from Aldrich.

Four ruminally cannulated adult sheep received a mixed diet (500, 299.5, 100, 91, and 9.5 g/kg of dry matter of hay, barley, molasses, fish meal, and minerals and vitamins, respectively) fed in equal meals of 500 g daily at 8 a.m. and 4 p.m. Samples of rumen fluid were removed 3 h after the morning feeding. The rumen fluid was kept warm and was strained through linen cloth before use. Incubations were carried out under O₂-free CO₂ in Hungate tubes at 39°C. Aliquots (7.5 ml) of ruminal fluid were added to the tubes, which already contained 7.5 µl of ethanol or 7.5 µl of ethanol containing 2 or 10 mM LY29 or DIC, in combination with either no added nitrogen source or 0.15 g of Trypticase (BBL, Becton Dickinson, Cockeysville, Md.), a pancreatic hydrolysate of casein containing mainly peptides, or 0.15 g of casein acid hydrolysate, containing mainly free amino acids (Oxoid, Basingstoke, Hampshire, United Kingdom). One-milliliter samples were removed at 2-h intervals up to 8 h. The samples were added to 0.25 ml of 25% trichloroacetic acid and cooled to 4°C. Pellets from trichloroacetic acid-treated samples were obtained by centrifuging at 28,000 × g for 15 min. Ammonia was measured by the phenol-hypochlorite method adapted from Whitehead et al. (20). Ammonia release was linear over the sampling period, so rates of ammonia release were calculated by linear regression.

Proteinase and peptidase activities were measured by using mixed ruminal bacteria from the same sheep, prepared by straining and centrifuging at 200 × g for 15 min in order to remove ciliate protozoa. Proteinase activity used ¹⁴C-labelled casein as previously described with whole bacteria (14), while peptidase assays used p-nitroanilide (pNA) substrates and sonicated bacteria, with protein being determined as described previously (17). The bacteriolytic activity of ruminal ciliate protozoa was measured by the rate of breakdown of [¹⁴C]leucine-labelled Selenomonas ruminantium in strained ruminal fluid (18).

Twenty different species of ruminal bacteria were examined for their sensitivities to 10 µM LY29 and 10 µM DIC. The source of most of the strains was given previously (2, 17), providing Anaerovibrio lipolytica 5S, Butyrivibrio fibrisolvens SH13, Eubacterium ruminantium 2388, Fibrobacter succinogenes S85, Lachnospira multipara D15d, Megasphaera elsdenii J1, Mitsuokella multiacidus 46/5, Prevotella albensis M384, Prevotella bryantii B₁4, Prevotella brevis GA33, Prevotella ruminicola 23, Ruminobacter amylophilus WP225, Ruminococcus albus SY3, Ruminococcus flavefaciens FD1, S. ruminantium Z108, Streptococcus bovis ES1, and Veillonella parvula L59. Clostridium aminophilum (ATCC 49906), Clostridium sticklandii (ATCC 12662), and Peptostreptococcus anaerobius (ATCC 27337) were obtained from culture collections. The anaerobic growth medium was the liquid form of Hobson's ruminal fluid-containing medium 2 (7), except for the growth medium for C. aminophilum, C. sticklandii, and P. anaerobius, which was basal M2 medium with no added sugars or lactate, with Trypticase added to a final concentration of 15 g/liter, as used in the cultivation of HAP bacteria (13). Cultivation was carried out in anaerobic Hungate culture tubes, and growth was measured by optical density at 650 nm. Growth experiments were done in triplicate with 5% inoculum volumes.

The rate of NH₃ production from Trypticase was 28% greater than the rate of NH₃ production from casein acid hydrolysate when strained ruminal fluid was incubated with Trypticase or casein acid hydrolysate in vitro (Table 1). LY29 and DIC both inhibited NH₃ production from Trypticase, by 17 and 28%, respectively, at inhibitor concentrations of 10 µM. This concentration was calculated to approximate likely ruminal fluid concentrations arising from effective feed inclusions of 50 ppm of DIC in 213-kg steers consuming 3.7 kg of dry matter per day (4) and 1 mg of LY29/kg of body weight in calves (21). NH₃ production from casein acid hydrolysate was also inhibited, by 17 and 37% by LY29 and DIC, respectively. Proteinase activity was unaffected by LY29 or DIC, and the main peptidase activities were unaffected, except for a small (<10%) inhibition of GlyPro-pNA breakdown (dipeptidyl peptidase IV) (Table 2). LY29 had no effect on the bacteriolytic activity of protozoa, whereas DIC tended to decrease protozoal activity slightly (Table 2). It was concluded, therefore, that effects on nitrogen flow in vivo (4, 21) and NH₃ production in vitro (Table 1) must have resulted mainly from inhibition of the final step in the breakdown pathway, namely the deamination of amino acids.

DIC did not inhibit the growth of any of the pure cultures of ruminal bacteria in 24-h incubations. The growth rate of C. sticklandii was decreased slightly, however, at 10 µM DIC (0.27 ± 0.03 h¹ versus 0.39 ± 0.02 h¹ in controls). LY29 at 10 µM prevented growth of C. aminophilum, C. sticklandii, and R. flavefaciens at 24 h and suppressed the yield at 24 h of some other species, notably A. lipolytica. Thus, the mode of action of LY29 appears to be different from the mode of action of DIC and may involve suppression of certain HAP bacteria. The structure of LY29 as a modified amino acid suggests that it may be a substrate analogue for amino acid deamination. DIC is thought to interfere with NADH metabolism (3).

The present results therefore identify the likely mode of action of a novel feed additive, LY29. LY29 principally affects deamination of amino acids by suppressing HAP bacteria. Its effect on NH₃ production in these sheep is relatively small (as illustrated in Table 1), but ruminants on different diets might have larger HAP populations and hence give greater responses to LY29. However, LY29 is selectively antibacterial. Given recent concerns about the spread of antibiotic resistance, and the finding that one of the bacteria inhibited by LY29 is an important cellulolytic species, it may be preferable to develop DIC rather than LY29 as a feed additive for improving nitrogen retention in ruminants.