1788
Nutrient supply estimations errors when using free ruminal bacteria as reference sample
The chemical composition of bacteria associated with solid (SAB) and liquid (LAB) rumen-digesta phases was studied to confirm errors when LAB is used as reference sample to estimate nutrient supply from SAB (González et al., 2012). Three rumen and duodenum cannulated wethers were fed three isoproteic diets for three successive periods. Diets included protein concentrates (sunflower meal and spring pea) untreated or treated with malic or ortophosphoric acid and heat supplied in this order. Diets were formulated with 45% oat (avena sativa) hay and 55% concentrate (fresh matter basis; 30% corn grain, 30% barley grain, 15% sunflower meal, 22% spring pea, 3% minerals and vitamins) and offered daily at 45 g/kg BW0.75 in six identical meals. Bacterial samples (SAB and LAB) were isolated after 12 d of continuous intraruminal infusion of (15NH4)2SO4 (25 mg 15N/d). Results were compared by variance analysis considering wethers (as blocks) and diets in the model. Compared with SAB, LAB showed consistently lower contents (g/kg DM) of OM (780 vs. 693; P < 0.001), starch- glucose (33.4 vs. 19.9; P = 0.013), and total lipids (221 vs. 113; P < 0.001), but greater CP content (402 vs. 429; P = 0.007) and 15N enrichment (atoms %: 0.0585 vs. 0.0817; P = 0.005). Present data fitted well with the relationship predicting SAB 15N enrichment from the same LAB value reported in the previously cited study (variation coefficient of the mean prediction error = 13.0 %). The mean ratio 15N-SAB/15N-LAB allows establishing CP supply from synthesized SAB is under evaluated by 25.6% when LAB is used as reference. This under evaluation was even greater for starch-glucose (44.2%) and lipids (61.9%) supply. These under evaluations were close to those previously reported by González et al., 2012. Taking into consideration that SAB is the main microbial source of nutrient supply, these errors should be factored in when attempting to improve ruminant nutrition. Predictions using 15N as marker are useful for correcting errors associated with the traditional use of LAB as reference sample, and therefore to obtain a more accurate estimate of microbial nutrient supply to the ruminant.
Keywords: rumen bacteria, chemical composition, 15N enrichment