Effect of form and of quality of feed on the concentrations of purine derivatives in urinary spot samples, daily microbial N supply and predictability of intake

This study examined the influence of dietary variation on the usefulness of urinary spot purine derivatives CPD) collected at three periods (8:00-9:30, 14:00-15:30 and 20:00-21:30 hours) as predictors of intake by cattle. Diets comprised teff (Eragrotis tef) straw, fresh and dried napier grass and fresh napier grass supplemented with 1.5 las fed) of dry alfalfa hay. Variation in dry matter intake (DMI) was further amplified by using calves ranging between 65 and 207 kg live weight (W). DMI and digestible organic matter (OM) ranged from 1.36 to 7.34 and from 0.67 to 3.88 (kg per day), respectively. The only differences tin intake, digestibility, microbial PD and N supply) among the diets which were significant at least at P=0.05 were due to supplementation with alfalfa. The excretion of allantoin and total PD were within the ranges 3.33-47.37 and 3.60-47.37 (mmol per day), respectively. The sampling period had no influence on the concentration of PD or creatinine (CR) nor on the PD:CR ratio in urine. Diets influenced (P<0.01) PD concentration such that the poorest quality diet (teff straw) exhibited the highest concentration, but diets had no effect on the PD:CR ratio. The PD:CR ratio or the PD concentrations had poor correlations with the intake and had no predictive power on their own. Further discussion was structured to develop ideas used to model intake based on PD concentrations and predicted urine output (UOP (g per day)- =exp(2.480+3.41 NDS)x W-0.942, where NDS (g/g DM) is neutral detergent solubles). The equation of intake on PD and UOP achieved 89% accuracy (OMI (g per day)= -0.15+0.556 UOP+0.038 PD8. where PD8 is the uncorrected PD concentration within 8:00-9:30 hours period). The slope and the intercept of the relationship between the predicted and observed intake were equal to 1 and 0, respectively, as in the equation OMI = 0.001(S.E. = 0.2099) + 1.00(S.E. = 0.072)OMIp (RSD = 0.422; Adj - R-2 = 0.89; n = 24; P < 0.0001); where the subscript 'P' denotes the predicted value. It is, thus, suggested that information on urinary spot PD concentration should be integrated with that on rumen osmolality modulators when developing intake models of free ranging livestock. The problem, however, is that though the approach is simple and may have general application, the ensuing models are likely to be region-specific because the pattern of UOP might be influenced by climatic factors and the animal's adaptation strategy to its environment. Further development should integrate the knowledge of the digestion kinetics and dilution rate in the rumen to determine an appropriate lag period that should intervene between the time feed is consumed and the time of spot-sampling such that the latter coincides with the peak concentration of PD in the urine. (C) 2000 Elsevier Science B.V. All rights reserved.