The effect of the source of nitrogen supplementation on nitrogen balance, rates of plasma leucine turnover, protein synthesis and degradation in sheep

Combined experiments of the isotope dilution method of [1-¹³C]leucine, open-circuit calorimetry and nitrogen (N) balance test were used to determine the effect of the source of N supplementation on N balance, whole body protein synthesis (WBPS) and degradation (WBPD) in sheep. The experiment was performed in a replicated 3 × 3 Latin square design. The control diet consisted of timothy hay, ground maize and soybean meal. The urea diet was the control diet supplemented with 1.5% urea. The SBM diet contained the same N and metabolisable energy as the urea diet, which was reached by changing ground maize and soybean meal weights of the control diet. Nitrogen retention was greater (p < 0.05) for the urea diet than the control and SBM diets. Plasma urea concentrations were highest for the SBM diet, followed by the urea diet, and the control diet was lowest. The WBPS and WBPD did not differ between diets, but were numerically lower for the urea and SBM diets. These results suggest that in sheep, urea supplementation influenced N retention without clear changes in WBPS and WBPD.

Keywords: urea; feed supplements; soybean oil meal; amino acid metabolism; protein synthesis; protein degradation; sheep

In ruminants, dietary protein utilisation is more complex than in monogastric animals because microorganisms in the rumen play an important role for the digestion of diets. The rumen microorganisms synthesise microbial protein from both dietary protein and non-protein nitrogen (N) compounds. However, little information is currently available focusing on the effect of the source of N supplementation on amino acid and protein metabolism in ruminants. In sheep, increased dietary crude protein (CP) intake did not influence whole body protein synthesis (WBPS) but decreased whole body protein degradation (WBPD), resulting in increased N retention (Sano et al. [23]).

Recently, soybean production has not been sufficient for both human food and animal diets, which has become a serious problem. Moreover, soybean meal, one of the most important dietary protein sources, is costly, largely degraded in the rumen and less absorbed than other protein sources in lactating dairy cattle (Santos et al. [25]). Urea, a non-protein N, is available at a low cost (Mahouachi et al. [19]). Urea supplementation is widely applied and has benefits for the production of cows (Tedeschi et al. [28]; Zinn et al. [33]). It was hypothesised that dietary N supplementation as urea and soybean meal might influence amino acid metabolism and protein retention in ruminants by increasing microbial protein synthesis in the rumen (Archimède et al. [4]). Therefore, this experiment was designed to measure plasma leucine (Leu) turnover rate (LeuTR), as measured by the isotope dilution method using a stable isotope, and to calculate WBPS and WBPD combined with determination of Leu oxidation (LeuOX) and N balance in sheep fed a basal diet supplemented with urea and soybean meal.

The surgery and experimental procedures were reviewed and approved by the Animal Care Committee of Iwate University. All experiments were carried out without noticeable stress to the animals.

Six crossbred (Corriedale × Suffolk) shorn sheep (3 ewes and 3 rams), aged 1 to 2 years and 45±2 kg BW were used. The sheep were surgically prepared under anaesthesia with a skin loop enclosing the left carotid artery at least three months before the start of the experiment. The sheep were assigned to three dietary treatments. The basal diet (control diet) consisted of timothy hay (metabolisable energy [ME] 9.2 kJ/g air DM, 6.1% CP, 57.7% neutral detergent fibre [NDF]), ground maize (ME 13.8 kJ/g air DM, 5.4% CP) and soybean meal (ME 13.3 kJ/g air DM, 36.2% CP). For the second diet (urea diet) the control diet was supplemented with urea (0.86 g · kg BW−0.75 · d−1), and N intake was designed to be 150% of maintenance. The third diet (SBM diet) was mixed with the same N content as the urea diet by changing ground maize and soybean meal weights of the control diet. Metabolisable energy and metabolisable protein (MP) intakes assumed from the Agricultural and Food Research Council (AFRC [1]) were similar for all dietary treatments and were slightly above the maintenance levels, as listed in Table 1. The experiment utilised a replicated 3 × 3 Latin square design with three periods of 29 d. Animals were housed in individual pens in an animal room during the preliminary period and the first three weeks of the experiment. The sheep were then moved to metabolic cages in a controlled environment chamber at an air temperature of 23±1°C, with lighting present from 06:00 to 22:00 h. The animals received feed once daily at 14:00 h, which was commonly eaten within 1 h, but the manger was not removed until the next morning. Water was available ad libitum. The sheep were weighed on day 1, 15 and 29 of each dietary treatment.

Table 1. Diet formulation and intakes of crude protein (CP), metabolisable protein (MP) and metabolisable energy (ME) of the dietary treatments.

Nitrogen balance was determined over five successive days of the last week of each 29 day treatment. Faeces were collected for each 24 h, dried in a forced air oven (60°C, 48 h) and weighed at 5 d after placement at room temperature. An aliquot was ground and stored until analysis. Urine was collected for each 24 h in a bottle containing 100 ml of 6 N H2SO4 and an aliquot was stored (−30°C). On day 27 of each treatment, rumen fluid (50 ml) was collected through a stomach tube at 2 h after feeding. Immediately after determination of pH of rumen fluid by a pH meter (HM-10P, Toa Electronics Ltd., Japan), the rumen fluid was centrifuged at 1000 g for 10 min at 4°C (RS-18IV, Tomy, Japan). An aliquot (2 ml) acidified for ammonia determination and residuals of the rumen fluid were stored at −30°C until further analysis.

A catheter for infusion was inserted into a jugular vein on day 27 and a catheter for blood sampling was inserted into the skin loop of the carotid artery in the morning on day 28. Catheters were filled with a sterile solution of 3.8% trisodium citrate. The combined experiment of an isotope dilution method of [1-13C]Leu and open-circuit calorimetry was conducted to determine plasma LeuTR, LeuOX, WBPS and WBPD as reported by Sano et al. ([23]). At 11:00 h, the sheep were fitted with a clear head chamber (approximately 0.2 m3) for collecting gaseous samples throughout the samplings of blood and exhaled gas. After collecting blood and gaseous samples at the pre-infusion period, 10 μmol/kg BW0.75 of [1-13C]Leu (L-leucine-1-13C, 99 atom % excess 13C; Isotec Inc., A Matheson, USA Co., USA) and 3.5 μmol/kg BW0.75 of NaH13CO3 (sodium bicarbonate-13C, 99.2 atom % excess 13C; Isotec Inc., A Matheson, USA Co., USA) dissolved in saline solution (0.9% sodium chloride) were injected into the catheter for infusion as a priming dose. Then, [1-13C]Leu (4 mmol/l in saline) was continuously infused by a multichannel peristaltic pump (AC-2120, Atto, Japan) at a rate of 10 μmol · kg BW–0.75 · h−1 through the same catheter for 8 h. Blood samples were taken from the catheter for blood sampling immediately before (10 ml) and at 30-min intervals (5 ml) during the last 90 min of [1-13C]Leu infusion. Samples were transferred into centrifuge tubes containing heparin sodium and were chilled until centrifugation. Blood samples were centrifuged at 10,000 g for 10 min at 4°C, and the plasma was stored at −30°C until further analyses.

Open-circuit calorimetry (Metabolic Monitor, Coast Electronics, UK) was used to determine carbon dioxide (CO2) production and exhaled 13CO2 enrichments throughout the isotope dilution method. The CO2 production rate was continuously determined for 15 min at 5.5 h of [1-13C]Leu infusion and immediately after the end of [1-13C]Leu infusion. An aliquot of exhaled CO2 was collected in 4 ml of 1 N NaOH for 30 min immediately before and four times half hourly during the last 2 h of [1-13C]Leu infusion to determine the isotope enrichments of exhaled 13CO2 and the NaOH solution was stored at −30°C. The catheters were removed after collections of all blood and gas samples.

Nitrogen contents in diets, faeces, and urine were determined by a colorimetric method for ammonia N (Weatherburn [31]) after Kjeldahl digestion. The concentrations of plasma Leu and α-ketoisocaproic acid (α-KIC) and enrichments of plasma [1-13C]Leu and α-[1-13C]KIC were determined by gas chromatography mass spectrometry (QP-2010, Shimadzu, Japan) by the procedures of Rocchiccioli et al. ([22]) and Calder and Smith ([8]) as described previously (Sano et al. [23]). The isotopic abundance of exhaled 13CO2 was determined with a gas chromatography-combustion-isotope ratio mass spectrometric system (DELTAplus, Thermo Electron, USA). Concentrations of plasma free amino acids and urea at the pre-infusion period were determined with an automated amino acid analyser (JLC-500/V, JEOL, Japan). Plasma insulin concentration at the pre-infusion period was determined by a radioimmunoassay kit (IRI 'Eiken', Eiken Chemical, Japan). The intra- and interassay coefficients of variation were 6 and 9%, respectively. Concentrations of ammonia in the rumen fluid and plasma ammonia were determined using a diagnostic kit (Ammonia-test, Wako, Japan). Concentrations of total volatile fatty acids (VFA) in the rumen fluid (5 ml) were determined by titration using 0.1 mol/l NaOH during steam distillation. The molar ratio of individual VFA was then analysed by gas chromatography (HP-5890, Hewlett Packard, USA) equipped with a 30 m × 0.25 mm DB-FFAP capillary column (J & W Scientific, USA). The conditions included the injector and detector temperature of 170°C. Oven temperature was raised from 100°C to 165°C by 8°C/min gradient with carrier gas (He) flow of 1.0 ml/min.

Mean values with standard errors of the mean are given. Plasma LeuTR and LeuOX were calculated using the equations by Krishnamurti and Janssens ([15]):

Graph

where I is the infusion rate of [1-13C]Leu [μmol · kg BW−0.75 · h−1] and EKIC is the isotope enrichment of plasma α-[1-13C]KIC during the steady states (ratio of [1-13C]KIC to [1-12C]KIC), and:

Graph

where ECO2 is the isotope enrichment of exhaled 13CO2 (ratio of 13CO2 to 12CO2) and VCO2 [mmol · kg BW–0.75· d–1] is the CO2 production rate. The recovery fraction of the exhaled CO2 production in the animal body was estimated to be 0.81 as described by Allsop et al. ([2]) and Wolfe et al. ([32]). Whole body protein synthesis (WBPS) and WBPD were calculated from the following equations as described by Krishnamurti and Janssens ([15]):

Graph

Graph

Leucine concentration in carcass protein (6.6%) was used as described by Harris et al. ([12]).

All data were analysed with the MIXED procedure of SAS ([26]). The fixed effect in the model was period, diet, and period × diet interaction and the random effect was sheep. The sex effect was also tested. Results were considered significant at the p < 0.05 level. If the effect of diet was significant, the Tukey adjustment was used to compare diets and was considered significant at p < 0.05. Repeated statement was used for the time course of parameters during the isotope dilution method and the difference in least square means with the Tukey adjustment was used (p < 0.05).

The sheep consumed all the diets supplied; therefore, ME and MP intakes were estimated to be similar among the dietary treatments. Nitrogen intake and N excretion in urine differed (p < 0.001 and p < 0.01, respectively) between diets (Table 2). Nitrogen intake was highest (p < 0.05) for the urea diet, but urinary N excretion was highest (p < 0.05) for the SBM diet. Nitrogen excretion in faeces did not differ between diets. Nitrogen retention differed (p < 0.01) between diets, and was higher (p < 0.05) for the urea diet than the control and SBM diets. Nitrogen digestibility was higher (p < 0.05) for the urea and SBM diets than the control diet. Body weight change did not differ between diets.

Table 2. Effects of the source of nitrogen (N) supplementation on N balance, N digestibility and body weight change in sheep (n = 6).

At 2 h after feeding pH values in the rumen fluid differed (p < 0.001) between diets, and were higher (p < 0.05) for the urea and SBM diets than the control diet (Table 3). Ammonia concentrations in the rumen fluid differed (p < 0.01) between diets, and were higher (p < 0.05) for the urea diet and lower (p < 0.05) for the control diet than other diets. Concentrations of total VFA, acetate, propionate, n-butyrate, isovalerate and n-valerate in the rumen fluid did not differ between diets. Concentrations of isobutyrate differed (p < 0.01) between diets, and were higher (p < 0.05) for the SBM diet than other diets. The period effect was significant in isobutyrate (p < 0.01) and n-butyrate (p < 0.05).

Table 3. Effects of the source of nitrogen supplementation on ruminal characteristics in sheep (n = 6).

Plasma amino acids, ammonia, urea and insulin concentrations determined at the pre-infusion period of the isotope dilution method were listed in Table 4. Of plasma amino acids, glutamine concentrations differed (p < 0.001) between diets and were lower (p < 0.05) for the urea and SBM diets than for the control diet. Other amino acid concentrations did not differ between diets. Plasma urea concentrations differed (p < 0.01) between diets, and were higher (p < 0.05) for the SBM diet and lower (p < 0.05) for the control diet than other diets. Plasma ammonia and insulin concentrations did not differ between diets. Concentrations of plasma isoleucine, Leu, phenylalanine, lysine, serine, arginine, tyrosine and proline were significantly higher for the rams compared with the ewes.

Table 4. Effects of the source of nitrogen supplementation on plasma amino acid, urea, ammonia and insulin concentrations in sheep (n = 6).

Concentrations of plasma Leu and α-KIC and enrichments of plasma [1-13C]Leu and α-[1-13C]KIC and exhaled 13CO2 were almost constant during the latter periods of the isotope dilution method for each treatment (data not shown). The mean coefficients of variances of isotope enrichments during the corresponding period were 6.4, 4.9 and 19.6% for plasma [1-13C]Leu and α-[1-13C]KIC and exhaled 13CO2, respectively. Plasma Leu and α-KIC concentrations for the urea diet did not differ from those for the control diet, but plasma Leu concentrations for the SBM diet were lower (p < 0.05) than the other two diets (Table 5). Plasma LeuTR, LeuOX, WBPS and WBPD did not differ between diets, but the values of LeuTR, WBPS and WBPD were numerically lower for the urea and SBM diets compared with the control diet. Plasma LeuTR was greater (p < 0.01) for the rams compared with the ewes, but no other gender difference was detected in parameters of protein metabolism including the WBPS.

Table 5. Effects of the source of nitrogen supplementation on plasma leucine (Leu) and α-ketoisocaproic acid (α-KIC) concentrations, rates of Leu turnover and Leu oxidation, whole body protein synthesis and degradation in sheep (n = 6).

The present experiment demonstrated that in mature sheep dietary N source influenced N retention without clear changes in WBPS and WBPD. Urea supplementation benefits ruminant production and improves the function of microorganisms in the rumen (Tedeschi et al. [28]; Zinn et al. [33]; Wallace et al. [30]). Tedeschi et al. ([28]) reported that in growing and finishing cattle, supplemental urea (0.4–1.2% of diet DM) improved average daily gain and feed conversion. Zinn et al. ([33]) reported that in steers, average daily gain increased linearly with increasing urea level (0, 0.4, 0.8 and 1.2% DM basis), whereas feed efficiency remained unchanged. Wallace et al. ([30]) reported that microbial proteolytic activities of rumen fluid in sheep fed diets supplemented with urea were higher than with casein and was similar to that with egg albumin. The level of urea supplementation (1.5% of diet) in the present experiment was comparable to those described above. Plasma urea concentrations were higher for N supplemented diets, but plasma ammonia concentrations were comparable between the dietary treatments. When urea was supplemented to the basal diet, the postprandial plasma ammonia concentrations increased markedly because a large part of the ammonia produced from supplemental urea in the rumen was directly absorbed from the rumen (Sinclair et al. [27]) and ammonia-N concentrations did not differ between with and without urea supplementation by 4 h after feeding (Mahouachi et al. [19]). Protein degradability of soybean meal in the rumen was greater and availability of amino acids for absorption was less than corn gluten meal, wet brewers' grains and distillers' dried grains with solubles in lactating cattle (Santos et al. [25]). Therefore, degradability of urea and protein from the urea and SBM diets in the rumen was faster than the control diet, but urea utilisation would still be activated even at 22 h after feeding without the effect on plasma ammonia concentrations.

In the present experiment, all dietary treatments showed positive N balance; therefore, CP intake may be above the maintenance requirements. Although body weight change did not differ between the urea and SBM diets, N retention was greater for the urea diet compared with the SBM diet due to the differences in urinary N excretion. The urea recycling rate into the gut tended to increase urea infusion into the rumen of sheep (Obara and Dellow [20]). Therefore, the urea recycling rate would be increased for the urea diet and result in lower plasma urea concentrations than the SBM diet. The higher N digestibility for the urea and SBM diets agreed with the observation by Ferrell et al. ([9]). They suggested that the apparent N digestibilities for urea, soybean meal and ruminally undegraded protein supplementation to diets should be interpreted with caution, because most of the faecal N loss was attributed to metabolic faecal N. Moreover, the higher N retention for the urea diet accorded with the results obtained in sheep (Kempton et al. [14]; Archimède et al. [4]). Kempton et al. ([14]) reported that in growing lambs fed a low-protein cellulose diet, urea supplementation improved digestibility and retention of N. Archimède et al. ([4]) also reported that in rams fed a low quality of roughage diet, supplemental urea and sugar (2.3 and 5.0% of hay, respectively) improved N balance and stimulated the microbial proliferation and activity. On the other hand, Bernard et al. ([6]) reported that in Jersey cows fed whole cottonseed coated with combinations of corn starch (2.5 and 5%) and urea (0.25 and 0.5%) duodenal flow rates of total and bacterial N were not influenced by urea treatment. The inconsistency of the effect could partly be related to the level of supplemental urea and species used. Net hepatic uptake and portal release of α-amino N increased with both urea and soybean meal supplementation in sheep (Ferrell et al. [9]). Therefore, increased N digestibility for the urea and SBM diets and N retention for the urea diet in the present experiment may be partly related to increased microbial protein synthesis from dietary N.

Zinn et al. ([33]) reported that in steers fed twice daily, urea supplementation did not influence total VFA concentrations and molar percentages of acetate and propionate in the rumen fluid. Hsu et al. ([13]) observed the similar results of ruminal VFA in defaunated sheep. In the present experiment concentrations of acetate and propionate, the major VFA, in the rumen fluid were not influenced by N supplementation. Therefore, it seems that fermentation of carbohydrates is not modified by N supplementation, when ME intake is constant. Ruminal pH determined at 2 h after feeding, the only spot sample from a stomach tube, was within the normal range for all the dietary treatments and was higher for the urea and SBM diets than the control diet. Higher ruminal pH would be reflected by both unchanged VFA production and the alkalising effect of supplemental urea in the rumen (Zinn et al. [33]).

In the present experiment, enrichments of plasma α-[1-13C]KIC and exhaled 13CO2 from plasma [1-13C]Leu were used for calculation of whole-body protein metabolism instead of those of plasma [1-13C]Leu and exhaled 13CO2, because enrichments of plasma α-[1-13C]KIC represent the true kinetics (Magni et al. [18]). The enrichment ratio of plasma α-[1-13C]KIC to plasma [1-13C]Leu was 70±1%, agreeing with those obtained previously (Sano et al. [23]; Al-Mamun et al. [3]). Whole body protein synthesis was close to those reported in sheep using the same experimental methods (Krishnamurti and Janssens [15]; Sano et al. [23]), and was also similar to those in goats using the [2H5]phenylalanine model (Fujita et al. [11]). However, WBPS in sheep was considerably lower than growing steers (Lapierre et al. [16]), even when the data was compared based on the metabolic body size. Sano et al. ([23]) reported that in sheep WBPS and WBPD were changed toward reduction with increased CP intake using both enrichments of plasma [1-13C]Leu and α-[1-13C]KIC. The present experiment observed a similar trend, but plasma LeuTR, WBPS and WBPD did not differ significantly between diets. Whole body protein synthesis would be more influenced by dietary energy and MP intakes than N intake (Lapierre et al. [17]; Fujita et al. [10]).

Plasma amino acid concentrations were not significantly influenced by N supplementation, except that plasma glutamine concentrations were lower than the control diet. Ferrell et al. ([9]) reported that in sheep plasma α-amino N concentrations remained unchanged with both urea and soybean meal supplementation. It is possible that decreased plasma glutamine concentrations were attributable to reduced proteolysis in the body protein. In the present experiment, the gender effect was significant in eight plasma free amino acid concentrations and LeuTR. Similar results were obtained in healthy humans as Caballero et al. ([7]) reported postabsorptive plasma concentrations of plasma valine, leucine and isoleucine were significantly lower in young women than in young men. The difference would partly be reflected by secretion of gonadotropins (Zurek et al. [34]).

Unchanged plasma insulin concentrations suggested that insulin release was not modified by N supplementation, although plasma insulin concentrations were only determined at the pre-infusion period. Concerning insulin action on protein metabolism, intravenous insulin infusion reduced WBPS, WBPD and endogenous Leu appearance rate with the isotope dilution method of [1-14C]Leu in fed and fasted lambs, and lactating and dry goats (Oddy et al. [21]; Tesseraud et al. [29]). Bequette et al. ([5]) also reported that in lactating goats insulin infusion reduced whole-body and hind-leg protein metabolism with the isotope dilution method of [15N, 1-14C]Leu and a hyperinsulinemic-euglycemic clamp. Insulin sensitivity was influenced by CP intake in sheep (Sano and Terashima [24]). The impact on increased N balance for the urea diet may partly be related to enhanced action of insulin which reduces WBPS and WBPD.

In conclusion, N supplementation as urea and soybean meal did not influence protein synthesis and degradation in sheep clearly. The difference in N balance between N supplementation would be related to the urea recycling and insulin action.

The authors are grateful to Kim Taylor, University of Guelph, Canada, for his kind comments on the manuscript.