Dehydrated sugar beet pulp sample, animals and feeding

A sample without addition of urea or molasses, provided by an extraction plant, was tested on three wethers fitted with rumen cannulae (internal diameter 60 mm) and T-type duodenal cannulae (internal diameter 12 mm). Wethers were fed with a 2 : 1 (fresh weight basis) chopped oat hay-to-concentrate diet containing 136 and 402 g/kg dry matter (DM) of CP and neutral detergent fibre (NDF), respectively. The diet was started 10 days before the experimental period, at 40 g DM/kg BW^0·75 in six equal meals (every 4 h), starting at 09.00 h. Wethers were housed in individual pens and handled according to the animal care principles as published in the Spanish Royal Decree 1201/2005 (BOE 2007). The DBP sample was ground to pass a 2-mm screen for in situ studies and to pass a 1-mm screen for chemical composition, including AA contents.

Experimental procedures

The experimental schedule to evaluate DBP included: (i) its incubation in nylon bags in the rumen, labelling micro-organisms with ¹⁵N to correct the microbial contamination of these residues, (ii) a transit study of the dietary concentrate particles, (iii) the isolation from rumen contents of solid adherent bacteria (SAB) to be used as a reference to correct microbial contamination and (iv) a study of the intestinal digestion of rumen undegraded feed.

A solution of ¹⁵N enriched ammonium sulphate (98 atoms%, i.e. 98 of 100 atoms correspond to the ¹⁵N isotope) was continuously infused into the rumen (25 mg N¹⁵/wether/day) to label micro-organisms, beginning 5 days before the start of the ruminal incubations and continuing till the end. Before suppressing this infusion, representative samples of rumen content were obtained for SAB isolation as described by Rodríguez et al. (Reference Rodríguez, González, Alvir, Repetto, Centeno and Lamrani2000). These isolates were lyophilized and analysed for DM, organic matter (OM), CP, ¹⁵N/N and AA.

Ruminal incubations were performed using nylon bags (internal dimensions 110×70 mm) with a pore size of 46 μm (Saatilon, SAATI Serigrafia Iberica S.A., Almazora, Castellón, Spain) filled with c. 3 g (air-dry basis) of DBP. A total of 12 bags were incubated in the rumen of each animal in two incubation series for 2, 4, 8, 16, 24 and 48 h. All bags were placed in the rumen simultaneously in each series, just before the wethers were offered their first meal in the morning (i.e. at 09.00 h). Bags were washed with tap water and stored at −20 °C after collection from the rumen. After thawing, bags were washed three times for 5 min each in a turbine washing machine (Jata 580; JATA, Zubibitarte, 8 - 48220 Abadiano, Bizkaia, Spain). The same procedure was applied to two series of two bags to obtain the water insoluble fraction (0 h value). Afterwards, the bags were again stored at −20 °C, freeze-dried and immediately weighed to establish the degradation kinetics of DM. As these kinetics had a sigmoid behaviour in all wethers, DM disappearance (d) with rumen residence time (t, h) was fitted by the model of Van Milgen & Baumont (Reference Van Milgen and Baumont1995):

$$d = a + b \times \displaystyle{{(1 - {\rm e}^{k_\infty t} )} \over {[1 + ({{(k_\infty - k_0 )} / {k_0}} ) \times {\rm e}^{ - k_\infty t} ]}}$$

In this model, the constants a and b represent, respectively, the proportions of the soluble fraction and the non-soluble but degradable component, which disappears at a fractional rate per unit time which varies from k ₀ to k _∞. The undegradable fraction (r) was estimated as 1−(a+b). The RUDM was determined as the complementary value of that proposed by Gómez (Reference Gómez1997) to determine the effective degradability when ruminal degradation kinetics showed a sigmoid behaviour. Values of RUDM were estimated when only k _p was used (RUDM_p) and when k _c and k _p were used jointly (RUDM_c,p) as:

$${\rm RUDM}_{\rm p} = 1 - \left[ {a + bk_0 k_\infty ^2 \int_0^\infty {\displaystyle{{{\rm e}^{( - k_{\rm p} - k_\infty )t}} \over {(k_0 (k_\infty - k_0 ){\rm e}^{ - k_\infty t} )^2}} {\rm d}t}} \right]$$

$$\eqalign{{\rm RUDM}_{{\rm c,p}}} =1 - \left[{a + bk_0 k_\infty ^2 \int_0^\infty {\displaystyle{{(k_{\rm c} {\rm e}^{ - k_p t} - k_{\rm p} {\rm e}^{ - k_c t} ){\rm e}^{ - k_\infty t}} \over {(k_{\rm c} - k_{\rm p} )(k_0 (k_\infty - k_0 ){\rm e}^{ - k_\infty t} )^2}} {\rm d}t}} \right]$$

The primitive functions of these integrals were not found and therefore RUDM values were determined by mathematical approximation using the mathematical calculation software Derive 2 (Soft Warehouse Inc., Honolulu, USA).

Values of k _p and k _c were derived from the study of particle transit in the rumen-reticulum. This was performed between both incubation series by pulse dosing each animal with 50 g of the dietary concentrate labelled with ytterbium (Yb) before the first morning meal (i.e. at 09.00 h), which was normally consumed in 20 min. This concentrate had previously been washed with a commercial detergent in an automatic washing machine to eliminate its soluble components (Udén et al. Reference Udén, Colucci and Van Soest1980) and labelled by immersion for 24 h in a solution at pH 2·2 of YbCl₃ (10 mg Yb/g of feed) and citric acid (0·7 mol/mol of YbCl₃) to avoid marker fixation in sites with low affinity. Finally, the residue was carefully washed with water and oven-dried at 65 °C for 48 h. A total of 21 samples were obtained from the duodenal cannulae, the first sample before supplying the marker and the remainder between 1 and 96 h afterwards. These samples were oven-dried at 60 °C, milled (through 1 mm screen) and analysed for Yb. The pattern of Yb concentrations in duodenal digesta over time was described for each animal using the model of Dhanoa et al. (Reference Dhanoa, Siddons, France and Gale1985). In agreement with Ellis et al. (Reference Ellis, Matis and Lascano1979) and González et al. (Reference González, Ouarti, Rodríguez and Alvir2006), primary and secondary rate constants of this model were assumed as k _p and k _c values, respectively.

Values of RU of DM, CP, OM and AA non-corrected or corrected by microbial contamination were determined considering both k _p and k _c transit rates with the method proposed by Arroyo & González (Reference Arroyo and González2013). This method is based on generating composite samples representative of the chemical composition of the RUDM. For this purpose, the residues obtained at each incubation time were pooled in equal quantities for each animal. The resultant residues for 0, 2, 4, 8, 16, 24 and 48 h of incubation were considered representative of the composition of the rumen outflow of undegraded feed in the intervals 0–1, 1–3, 3–6, 6–12, 12–20, 20–36 and 36–72 h, respectively. Then the rumen outflow of undegraded feed in these intervals was established by the function proposed by Arroyo & González (Reference Arroyo and González2013). The proportions in which the different residues were mixed were calculated by the quotient of the flow in each interval and the total flow. The flow function cited above was established by considering single exponential kinetics. For this reason, single exponential kinetics equivalent to previous sigmoid kinetics were established using the previous values of a, b and r, and a constant degradation rate (0·058±0·0079 (standard deviation (s.d.)) providing the same RUDM_c,p values. The composite samples were analysed for CP, OM, AA and ¹⁵N/total N. The respective RU values were determined from the concentrations of the tested fraction in the composite samples (Y) and in the whole feed (X) and the RUDM_c,p values, as follows:

$${\rm RU} = Y \times {\rm RUDM}_{{\rm c,p}} /X$$

After a 10-day resting period to eliminate the ¹⁵N enrichment from the digesta, the IED of DM, CP and AA were determined. For this purpose, eight sub-samples (c. 200 mg) from the composite samples of each wether were weighed into mobile nylon bags with an approximately round shape (diameter c. 30 mm). These bags were introduced through the duodenal cannula and recovered in the faeces. Bags were inserted randomly at a rate of one bag every 15 min (8 bags/wether/day). The bags were subsequently processed by same methods as for rumen incubations, dried at 70 °C for 48 h and weighed. The IED of DM was calculated as the disappearance of DM from the bag during the intestinal incubation. Finally, the undigested residues obtained in each wether were pooled prior to CP, AA and ¹⁵N/total N analysis. The IED of each fraction was determined from its concentration in the composite samples (Y) and in the intestinal incubated residues (Z), and the values of IED of DM:

$${\rm IED} = {\rm 1}-[Z \times ({\rm 1}-{\rm IED}\,{\rm of}\,{\rm DM})/Y]$$

The microbial proportions of N and DM in the composite samples and in the intestinal incubated residues were determined using SAB as a reference as indicated by Rodríguez & González (Reference Rodríguez and González2006). The microbial proportions of OM were established as microbial DM×(OM in SAB/OM in the tested residue). The microbial contribution to a specific AA was determined as the microbial DM content of these samples×AA concentration in SAB expressed on DM.

Chemical analyses

Analyses of DBP, incubated residues or SAB were performed by the methods of AOAC (2000) for DM (procedure 934·01), ash (procedure 967·05), ether extract (procedure 920·39) and CP (N×6·25; procedure 968·06). Analyses of NDF (Van Soest et al. Reference Van Soest, Robertson and Lewis1991), acid detergent fibre (ADF) and acid detergent lignin (ADL) (Robertson & Van Soest Reference Robertson, Van Soest, James and Theander1981) were also performed in DBP. The insoluble nitrogen in neutral detergent (NDIN) and in acid detergent (ADIN) solutions was determined by N analysis of the respective residues. Ytterbium in duodenal contents was analysed in nitric acid extracts (Rodríguez et al. Reference Rodríguez, González, Alvir, Repetto, Centeno and Lamrani2000) by inductively coupled plasma emission spectroscopy (Perkin Elmer, ICP 5000). Nitrogen isotopic proportions were analysed by mass spectrometry (Barrie et al. Reference Barrie, Brookes, Prosser and Debney1995). Amino acids were determined after acid hydrolysis using alpha-aminobutyric acid as the internal standard. Therefore, tryptophan was not measured in the current study. This hydrolysis was preceded by an oxidation with performic acid to obtain methionine and cysteine values. These analyses were performed by using the HPLC system, after derivatizing AA with AQC (6-aminoquinolyl-N-hydroxysuccinimidyl carbamate) in accordance with Liu et al. (Reference Liu, Chang, Yan, Yu and Liu1995).

Statistical methods

The kinetics associated with the models used were fitted using non-linear regression models. The effects of the transit model (t: to use k _pv. to use k _c, k _p) on the value of RUDM as well as those of microbial contamination correction (c) in RU, IED and intestinal digested fractions (IDF) of DM, OM, CP and AA were studied by variance analyses by taking into account this primary factor (t or c) as well as the wethers (w) in the model (y _ij = μ+t _i (or c _i)+w _j+ε_ij). Considering AA as the primary factor, this same design was also used to compare microbial corrected results of RU, IED and IDF of individual AA v. total analysed AA (TAA) by protected least square differences. Differences between CP and TAA values for these last cited parameters were studied with a split-plot arrangement of treatments considering microbial correction (c) as the whole-plot and the nitrogenous fraction (f) as the sub-plot (y _ijk = μ+c _i+f _j+w _k+c _i × f _j+c _i × w _k+ε_ijk). All statistical analyses were performed using the GLM procedure of SAS (SAS Institute 1990).