Proteolytic activity of the bovine rumen microflora was studied with azocasein as the substrate. Approximately 25% of the proteolytic activity of rumen contents was recovered in the strained rumen fluid fraction, and the balance of the activity was associated with the particulate fraction. The proportion of proteinase activity associated with particulate material decreased when the quantity of particulate material in rumen contents was reduced. The specific activity of the proteinase from the bacterial fraction was 6 to 10 times higher than that from the protozoal fraction. Proteinase inhibitors of synthetic, plant, and microbial origin were tested on proteolytic activity of the separated bacteria. Synthetic proteinase inhibitors that caused significant inhibition of proteolysis included phenylmethylsulfonyl fluoride, N-tosyl-1-lysine chloromethyl ketone, N-tosylphenylalanine chloromethyl ketone, EDTA, cysteine, dithiothreitol, iodoacetate, and Merthiolate. Plant proteinase inhibitors that had an inhibitory effect included soybean trypsin inhibitors types IS and II-S and the lima bean trypsin inhibitor. Proteinase inhibitors of microbial origin that showed an inhibitory effect included antipain, leupeptin, and chymostatin; phosphoramidon and pepstatin had little effect. We tentatively concluded that rumen bacteria possess, primarily, serine, cysteine, and metalloproteinases. Ruminal proteolysis can result in a loss of high-quality dietary protein that would otherwise be directly digested and absorbed in the small intestine of the ruminant animal (7). The first step in protein degradation in the rumen is hydrolysis of proteins by proteinases to peptides and amino acids, which are either utilized directly by the microflora or degraded further by peptidases and deaminating enzymes to shortchain fatty acids and ammonia (12, 33, 34). Proteolysis has been suggested to be the ratelimiting step in the degradation of fraction 1 protein from alfalfa (31), although with other proteins, the utilization of amino acids was considered to be the limiting step (11, 26). Methods successful in improving protein utilization in ruminants by decreasing the apparent degradation of protein in the rumen have included chemical treatment of feed materials (11), defaunation (6), and inclusion of feed additives such as monensin and diaryliodonium compounds (13, 32). The most direct and perhaps the most effective means of decreasing the degradation of protein within the rumen is through the selective

Intact, metabolically active rumen protozoa prepared by gravity sedimentation and washing in a mineral solution at 10 to 15 degrees C had comparatively low proteolytic activity on azocasein and low endogenous proteolytic activity. Protozoa washed in 0.1 M potassium phosphate buffer (pH 6.8) at 4 degrees C and stored on ice autolysed when they were warmed to 39 degrees C. They also exhibited low proteolytic activity on azocasein, but they had a high endogenous proteolytic activity with a pH optimum of 5.8. The endogenous proteolytic activity was inhibited by cysteine proteinase inhibitors, for example, iodoacetate (63.1%) and the aspartic proteinase inhibitor, pepstatin (43.9%). Inhibitors specific for serine proteinases and metalloproteinases were without effect. The serine and cysteine proteinase inhibitors of microbial origin, including antipain, chymostatin, and leupeptin, caused up to 67% inhibition of endogenous proteolysis. Hydrolysis of casein by protozoa autolysates was also inhibited by cysteine proteinase inhibitors. Some of the inhibitors decreased endogenous deamination, in particular, phosphoramidon, which had little inhibitory effect on proteolysis. Protozoal and bacterial preparations exhibited low hydrolytic activities on synthetic proteinase and carboxypeptidase substrates, although the protozoa had 10 to 78 times greater hydrolytic activity (per milligram of protein) than bacteria on the synthetic aminopeptidase substrates L-leucine-p-nitroanilide, L-leucine-beta-naphthylamide, and L-leucinamide. The aminopeptidase activity was partially inhibited by bestatin. It was concluded that cysteine proteinases and, to a lesser extent, aspartic proteinases are primarily responsible for proteolysis in autolysates of rumen protozoa. The protozoal autolysates had high aminopeptidase activity; low deaminase activity was observed on endogenous amino acids.

Alkaline phosphatase (APase) activity of Megasphaera elsdenii was enhanced by PO4 2- limitation in batch culture; however, six other species of rumen bacteria tested showed no increase in APase activity under these conditions. Alkaline phosphatase was produced by the mixed rumen microflora even though the inorganic phosphorus concentration was as high as 10mM. The APase activity of the bacterial fraction from rumen fluid was not increased during incubation in a phosphorus-free culture medium. Since bacteria may account for greater than 80% of the APase activity in the rumen, this would suggest that the bulk of the APase activity in the rumen is synthesized constitutively. The bacterium responsible for most of the APase activity probably is Bacteroides ruminicola.

Experiments were initiated to select a sterilization method(s) that minimizes alterations in the digestive properties of cereal grains and, thus, would be suitable for the study of cereal grain digestion by pure cultures of ruminal bacteria. The following five treatments were examined: unsterilized (U), autoclaving with buffer (AB), autoclaving without buffer (AD), ethylene oxide (E), and gamma irradiation (I). Solubility of DM, starch, and CP was determined by soaking grain in buffer for 1 h followed by filtration through Whatman #54 filter paper. Ground corn and wheat from each treatment were placed in vials with a 1:1 mixture of Bryant's medium and ruminal inoculum. Vials were incubated for 4, 8, 12, 24, and 48 h and analyzed for starch content. Bacterial growth was not evident in sterilized, uninoculated samples. The AD treatment decreased the disappearance of CP in wheat and corn, whereas AB caused an increase in the disappearance of DM, CP, and starch in wheat (P less than .001) compared with U. Rates of microbial starch digestion for corn were 1.3, 1.5, 3.3, 14.7, and 3.5%/h and for wheat were 1.3, 3.4, 4.6, 17.1, and 4.6%/h for AD, E, I, AB, and U, respectively. Contrasts indicated that AD and AB differed (P less than .001) from U for both corn and wheat. It is likely that gelatinization of cereal starch enhanced microbial starch digestion in AB and the formation of Maillard products reduced starch digestion in AD. Corn and wheat sterilized with E or I had digestive properties that closely resembled those of U grain, and either sterilization method was suitable for studying cereal grain digestion.

Thirty-nine xylanolytic bacteria tentatively identified as Clostridia were isolated from a selection of agricultural and forest soil samples, and water–sediment–fibre samples from acidic springs in the Yellowstone National Park. Screening for xylan hydrolysis was performed using an enriched agar medium at pH 5.5 with Remazol–xylan as the indicator, with or without an initial enrichment using xylan as the major carbon source. From 13 of the most highly xylanolytic strains, 9 were tentatively identified as Clostridium acetobutylicum, 2 as Clostridium butyricum, and 2 as Clostridium beijerinckii. The C. acetobutylicum isolate 3BYR utilized 80% of oat spelt xylan as a carbon source during growth. The bacterium exhibited very high extracellular xylanase and xylosidase activities and, as well, α-L-arabinofuranosidase and α-glucuronidase activities. Glucuronidase activity was documented by the release of 4-O-methyl-α-D-glucuronic acid from birchwood xylan. The results of this work indicate the ubiquity of xylanolytic Clostridia, and that the previously unreported activity, α-glucuronidase, has been demonstrated in C. acetobutylicum. Key words: α-glucuronidase, xylanase, Clostridium acetobutylicum, xylan.