Nutrient digestion and nitrogen metabolism of dried fermentation biomass and various fractions of rumen microbes.

Abstract

An experiment was performed to investigate if fermentation biomass (FB), a dried bacterial co-product derived from lysine production (Ajinomoto Heartland, Inc.) can be used as a protein source in ruminant diets. Eight dual-flow continuous culture fermenters were inoculated with rumen fluid and used during one experimental period consisting of a 7-d adaptation period followed by 3 sampling days. Substrate for the microbes was provided by one of two isonitrogenous diets, CON or DFB. In CON, soybean meal (SBM) provided 57% of total CP, and in DFB, SBM and FB provided 12 and 45% of total CP, respectively. CON contained 3% molasses, 16% ground corn, 13% grass hay, 48% corn silage, and 20% SBM on a DM basis. DFB contained 3% molasses, 18.4% ground corn, 13% grass hay, 50% corn silage, 8.5% SBM, and 6.7% FB. Fermenters were supplied with 75 g/d of DM divided into 8 equal portions. Anaerobic conditions were maintained by infusion of N2; pH was maintained between 5.8 and 6.8; and temperature was set at 39&#61616;C. On sampling days, liquid and solid effluent were collected, combined, and homogenized to be used for chemical analysis and in vitro estimation of intestinal digestibility (ID). Treatment did not affect (P > 0.1) average, maximum, or minimum fermenter pH. There was no effect (P > 0.1) on apparent or true OM, NDF, or ADF digestibility (%) or total VFA (mM), although branched-chain VFA (mM) was higher (P = 0.01) in CON because isobutyrate (mol/100 mol) tended (P = 0.06) to increase with CON treatment. Source of N had no effect (P > 0.1) on total, dietary, or bacterial-N flows. Addition of FB decreased (P < 0.05) NH3-N flow from 0.4 to 0.2 g/d and tended to decrease (P =0.06) effluent NH3-N concentration from 17.1 to 9.7 mg/100 mL. His and Met flows increased (P < 0.05) from 0.48 to 0.53 and 0.18 to 0.20 g/d, respectively, when FB partially replaced SBM in the diet, but there were no effects (P > 0.1) on other AA or total AA flows. There was a trend (P = 0.08) in percent non-essential AA input (CON = 73.6% vs. DFB = 82.2%) in effluent; however, there was no effect (P > 0.1) on percent of essential AA or total AA input in effluent. Effluent from the DFB treatment was higher (P < 0.05) in ID than CON (CON = 70.4% vs. DFB = 79.6%). These results indicate that FB elicited a similar response in N metabolism and AA flows to SBM but had a higher estimated digestibility in the intestine, and has potential use as a protein source in ruminant diets. In a second experiment, rumen fluid was collected from 4 steers fed the same diet, and 3 populations of rumen microorganisms (liquid-associated bacteria (LAB), solid-associated bacteria (SAB), and liquid-associated protozoa (LAP)) were isolated by differential centrifugation. Intestinal digestibility of N (ID) was determined using the in vitro (pepsin and pancreatin) steps of the 3-step procedure of Calsamiglia and Stern (1995). Total-N content of each fraction was different (P < 0.05) with LAB having the highest N content (8.1 vs. 6.1 vs. 5.6% of DM for LAB, SAB, and LAP, respectively). Purine concentration was greatest (P < 0.05) in LAB compared with SAB and LAP (3.1 vs. 1.0 vs. 0.9 mg/g bacterial DM, respectively). Similarly, LAB had a higher (P < 0.05) purine:N ratio compared with SAB and LAP (37.4 vs. 12.3 vs. 11.1 mg/g of bacterial N, respectively). Total amino acids (AA) (wt/wt) varied (P < 0.05) with each treatment (35.9 vs. 29.8 vs. 27.72% for LAB, SAB, and LAP, respectively). Most individual AA (g/100 g of total AA and wt/wt) exhibited differences (P < 0.05) between the 3 populations. Essential AA content (g/100 g of total AA) was highest (P < 0.05) in LAB and lowest in LAP (47.15 vs. 46.02, respectively). Non-essential AA content was greatest (P < 0.05) in LAP followed by SAB and LAB (53.3 vs. 52.3 vs. 51.8 g/100 g total AA, respectively). LAB was highest (P < 0.05) in ID followed by LAP and SAB (71.2% vs. 68.2% vs. 57.5% for LAB, LAP, and SAB, respectively). These results demonstrate that microbial fractions have differing digestibilities in the small intestine and suggest that ID of mixed rumen microbes can be influenced by the relative amount of each microbial population.