QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text (PDF) Purchase Article View Shopping Cart Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by van Milgen, J. Articles by Dubois, S. Search for Related Content PubMed PubMed Citation Articles by van Milgen, J. Articles by Dubois, S.

---

Articles

Energetic Efficiency of Starch, Protein and Lipid Utilization in Growing Pigs^{1 ,2}

INRA, Unité Mixte de Recherches sur le Veau et le Porc, 35590 Saint-Gilles, France

³To whom correspondence should be addressed at INRA-UMRVP, Domaine de la Prise, 35590 Saint Gilles, France. E-mail: jaap{at}st-gilles.rennes.inra.fr

	ABSTRACT

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
Mathematical models are increasingly used to predict the response of an animal to a changing nutrient supply. The objective of this experiment was to provide data that can be used in model development or evaluation and concerns the energetic efficiency with which nutrients are used for protein and lipid deposition. A basal diet (D1), limiting in lysine supply, was fed at 1.7 MJ metabolizable energy (ME)/(kg BW^0.60 · d¹) to growing pigs that weighed 60 kg. Four additional diets were formulated: the basal diet and a dietary supplement that consisted of starch (D2), starch and corn gluten meal (D3), starch and casein (D4) or starch and lipid (D5). The latter four diets were fed at 2.55 MJ ME/(kg BW^0.60 · d¹) and ensured the same intake of the basal diet across treatments; the difference was supplied by the supplement. Metabolic utilization of the basal diet and supplements was determined using nitrogen and energy balances (indirect calorimetry). The N retention was similar in pigs fed diets D1, D2, D3 and D5 but considerably higher in those fed D4. A data analysis model was developed to account for differences in ME utilization between nutrients. The ME not deposited as protein entered a common pool of energy, which was used for adenosine triphosphate synthesis or lipid deposition. The energetic efficiencies of ME utilization were 0.842, 0.520 and 0.883 for starch, protein and lipid, respectively. Due to the energy cost of protein deposition (or protein turnover), the energetic efficiencies of depositing dietary protein as protein or lipid were similar.

KEY WORDS: • pigs • energy efficiency • nutritional models • nutrient utilization

	INTRODUCTION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
In the past 25 y, increasing interest has developed in predicting growth performance in swine with the use of mathematical models. There is general consensus that energy intake often is the first limiting factor for protein deposition in young, growing pigs. Protein synthesis (and thus protein deposition) is an energetically expensive process, requiring (at least) 5 adenosine triphosphate (ATP)⁴ molecules per peptide bond synthesized. Additional dietary energy (i.e., energy supplied in excess of maintenance, activity, thermoregulation and protein synthesis) can be used for lipid deposition. The efficiencies with which nutrients provide energy for ATP or lipid synthesis are not necessarily the same. There is increasing awareness that excess dietary protein is an inefficient energy source. This, combined with the environmental burden of nitrogenous excretions, emphasizes the importance of optimizing the nitrogen (N) and energy supplies in domestic animals. The objective of the current experiment was therefore to determine the energetic cost of protein and lipid deposition with different nutritional sources.

	MATERIALS AND METHODS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
Experimental diets and feeding.

A basal diet containing wheat, corn and soybean meal was formulated to be first limiting in lysine supply [digestible lysine/digestible energy (DE) <0.5 g/MJ; Table 1 ]. This basal diet (D1) was fed at 1.70 MJ metabolizable energy (ME)/(kg BW^0.60 · d¹) to six Piétrain x (Landrace x Large White) barrows weighing 60 kg. Four other diets were formulated that consisted of a combination of the basal diet and a dietary supplement that provided ME from starch (D2), 50% starch plus 50% unbalanced protein (D3), 50% starch plus 50% balanced protein (D4) or 50% starch plus 50% lipid (D5). Diets D2 through D5 were offered to four groups of five or six pigs at 2.55 MJ ME/(kg BW^0.60 · d¹). Diet formulation and feeding level ensured that approximately two thirds of ME intake was provided by the basal diet and approximately one third was provided by the dietary supplement. Intake of the basal diet was similar for all five diets; the difference in intake between diets D1 and D2–D5 was due to the dietary supplement. Corn gluten meal was used as the unbalanced protein source, whereas casein was used as a balanced protein source. Because corn gluten meal contains some lysine (1.7% lysine in crude protein according to the National Research Council (1) , no synthetic lysine was added to D3. Low levels of tryptophan relative to the other neutral amino acids may depress feed intake; therefore, tryptophan was supplemented to ensure that the ratio between tryptophan and other neutral amino acids exceeded 0.04 (2) . Diets were pelleted and offered in four (approximately) equal meals at 0900, 1300, 1700 and 2100 h. Pigs consumed water ad libitum.

The (apparent) ileal digestibility of the five diets was determined in a 5 x 5 Latin Square using five Piétrain x (Landrace x Large White) barrows. The pigs were surgically prepared with an ileorectal anastomosis 3 wk before the experiment (3) . The experimental diets were fed during 7 d at 2.1 MJ ME/(kg BW^0.60 · d¹). The first 4 d were used as an adaptation to the experimental diets. Feces were collected during the last 3 d.

Fecal digestibility and energy and N balances.

A total of 30 Piétrain x (Landrace x Large White) barrows from 15 litters were used for the determination of energy and N balances. The energy and part of the N balance experiments were carried out in two 12-m³ open-circuit respiration chambers. Therefore, only two N and energy balances could be carried out at a time. Littermates received a different diet according to an alternating design (i.e., D1/D2, D3/D4, D5/D1, etc.). Temperature in the chambers was maintained at 24°C, and a 13-h lighting scheme (0830 to 2130 h) was adopted. The metabolism cages were mounted on force sensors (Kistler Instrumente AG, Winterthur, Switzerland), which produced a signal proportional to the physical activity of the pigs. In addition, the cages were equipped with infrared beams to detect standing (or sitting) of the pigs. The use of these two devices allowed us to distinguish between standing physical activity and resting physical activity.

The pigs were moved to the experiment building 10 d before the beginning of the experiment. This 10-d period were used to adapt the pigs to the diets and metabolism cages. The N balance (and fecal digestibility) was performed on d 11–18. On d 14, the pigs, in their cages, were moved to a respiration chamber for the measurement of energy balance. Measurement of the N and energy balance in the fed state terminated the morning of d 19, followed by weighing of the pigs. The pigs then reentered the respiration chamber for measurement of the fasting heat production (i.e., no food was offered to the pigs on d 19). This measurement was terminated the morning of d 20. The care and use of pigs met the requirements of French law 87.848 (October 19, 1987). An authorization to perform an experiment on living animals was given by the French Ministry of Agriculture and Fishery (certificates 7704 and 4739 for J. van Milgen and J. Noblet, respectively).

Chemical analyses.

The five diets were analyzed for dry matter (DM), organic matter, N, fat, crude fiber (4) , starch (Ewers’ polarimetric method) and gross energy (GE; adiabatic bomb calorimeter). The amino acid composition of diets was determined after 24-h acid hydrolysis (48 h for leucine, isoleucine and valine). Methionine and cysteine were hydrolyzed after preoxidation with formic acid. Amino acid contents were determined by HPLC (Alliance System; Waters France, Saint-Quentin-en-Yvelines, France) after precolumn derivatization with 6-aminoquinolyl-N-succinimidyl carbamate (AccQ.Fluor Reagent, Waters). Dietary supplements (i.e., corn starch, corn gluten meal, casein and vegetable oil) were analyzed only for criteria judged useful to verify diet composition. Fecal samples from individual pigs were analyzed for DM, organic matter, N and GE. A fecal sample pooled per treatment was analyzed for crude fiber and fat. The amino acid composition was determined in pooled ileal digesta. The N and GE contents were determined in individual urine samples.

The energy balance consists of measuring gas exchanges between the animal and its environment (see later), from which the heat production can be calculated (5) . The consumption or production of these gases (O₂, CO₂ and CH₄) was measured continuously. An aliquot of gaseous NH₃ was taken during measurement of the energy balance and analyzed separately. The O₂ was measured with a paramagnetic differential analyzer (Oxygor 6N; Maihak AG, Hamburg, Germany), whereas CO₂ and CH₄ were measured with infrared analyzers (Unor 6N; Maihak AG).

Data analysis.

Although respiration chambers are traditionally used to determine the (daily) energy balance of animals, modeling techniques can be used to determine the components of heat production (6 ,7) . In short, the method relates observed changes in gas concentration to eating behavior and physical activity. Three main components of heat production are distinguished (Fig. 1 ): the fasting heat production (FHP), the thermic effect of feeding (TEF) and the heat production due to physical activity (HP_activity). The FHP is estimated as the asymptotic nocturnal heat production after a period of food deprivation of 24 h. The heat production due to physical activity is estimated as the product of the signal detected by the force sensors and the estimated heat production per force signal unit. The TEF is then the difference between total heat production and the sum of FHP and HP_activity. In addition, a short-term and a long-term component of TEF may be distinguished. The short-term component (TEF_ST) relates to a dynamic component of heat production that is directly associated with the consumption of a meal (e.g., due to ingestion and digestion). In contrast, no distinguishable kinetics of heat production are observed for the long-term TEF (TEF_LT). Metabolic processes such as nutrient metabolism or fermentation are considered to be the origin of TEF_LT. Although feed consumption is a relatively rapid process, the associated TEF_ST has a prolonged duration. The time to dissipate half of TEF_ST after the ingestion of a meal (T_TEF) is one of the model parameters. The FHP, TEF_LT, TEF_ST, T_TEF and HP_activity were estimated statistically by regressing the observed O₂ and CO₂ concentrations on the independent variables (time, quantity and time of feed intake and signal of force sensors) with the model described previously (6 ,7) using ACSL/Optimize (8) .

View larger version (30K): [in this window] [in a new window]   Figure 1. Example of the components of heat production in a growing pig (60 kg) fed at 2.4 MJ metabolizable energy/(kg BW0.60 · d1) offered in four separate meals at 0900, 1300, 1700 and 2100 h. HPactivity, heat production due to physical activity; TEF, thermic effect of feeding.

To calculate energy values and energetic efficiencies of nutrients, we used a nested, multivariate regression model. The procedure is conceptually similar to a univariate multiple regression procedure but also accounts for the relations that exist between the dependent variables. Intakes of GE, DE and ME; protein deposition (PD); and lipid deposition (LD) were simultaneously regressed on the intakes of the basal diet, starch, gluten, casein and lipid. The corn gluten meal used in D3 contained some starch and lipid (177 and 25 g/kg, respectively). The fraction further referred to as "gluten" is the added corn gluten meal minus its starch and lipid contents. The latter two fractions were accounted for in the dietary components. The relations that exist among GE, DE and ME are relatively straightforward. The GE intake for a pig fed one of the five diets is based on the intake of each dietary component and the GE value of that component:

or

(1)

where GE_i is the GE intake (kJ/d) for a pig fed diet i (i = D1–D5), DMI is the DM intake for an individual pig (g/d), F_j is the fraction of each dietary component in the diet (j = basal diet, starch, gluten, casein or lipid) and GEc_j is the GE content of that component (kJ/g). The DE intake therefore is

(2)

where DE_i is the DE intake (kJ/d) for a pig fed diet i, and dc_j is the energy digestibility coefficient of the dietary components. Similarly, the ME intake is

(3)

where ME_i is the ME intake (kJ/d) for a pig fed diet i, and mc_j is the energy metabolizability coefficient of DE (i.e., ME/DE).

Diets were formulated so that the digestible lysine supply of the basal diet was the first limiting factor for protein deposition in D1, D2, D3 and D5. For D4, other factors will limit PD (e.g., energy intake or intrinsic animal factors). Nevertheless, it is possible that the noncasein supplements provoke a slight increase in PD. To account for this, PD is given as the sum of five components:

or

(4)

where PD_i is the protein deposition (kJ/d) for a pig fed diet i, and p_j is the fraction of ME_j that results in additional PD. If results were as anticipated, p_starch, p_gluten and p_lipid would not be different from zero.

It is assumed that all ME not used for PD enters a common pool that supplies energy for both lipid deposition and ATP synthesis (Fig. 2 ). Functions that require ATP include maintenance functions, physical activity and protein synthesis. The efficiencies with which different nutrients enter the common pool are not necessarily the same. This pool of "PD-free net energy (NE)" therefore is

where k_j is the efficiency of using ME supplied by a dietary supplement for the common pool of energy. The efficiency of using excess casein for the PD-free NE pool was assumed to be the same as that for gluten (k_protein). It was also assumed that the ATP requirement (for maintenance, activity and PD) has a greater priority for PD-free NE than LD. Consequently, PD-free NE not used for ATP synthesis is deposited as lipid. The ATP requirement was based on the measured HP_activity, the measured FHP and the estimated PD (from ^{eq. 4} ). Differences in the measured HP_activity may be due to differences in ATP requirements for activity per se but also to differences with which nutrients supply energy for this ATP requirement. To convert HP_activity to its equivalent of PD-free NE, it was multiplied by the weighted average of energetic efficiencies (k_diet):

The measured FHP was supposed to be indicative of maintenance energy functions. During fasting, the animal uses its body reserves (BR) (i.e., glycogen, protein and fat) for maintenance needs. A fraction of the heat production measured as FHP is due to the conversion of BR to ATP, and the remainder is due to heat loss from actual ATP utilization. If it is assumed that the diet has no effect on the relative utilization of BR for ATP synthesis, the PD-free NE equivalent of FHP equals k_BR x FHP, where k_BR is the energetic efficiency of using BR for NE. When pigs are fed, the same quantity of PD-free NE would be required for maintenance, but this energy is then supplied by the diet and not by BR. Finally, the energy requirement for PD is given by NE_PD x PD, where NE_PD is the PD-free NE requirement to deposit 1 kJ of PD. In summary, the PD-free NE not used for ATP synthesis is deposited as lipid:

(5)

where LD_i is the lipid deposition (kJ/d) for a pig fed diet i.

View larger version (23K): [in this window] [in a new window]   Figure 2. Schematic of different nutrient sources for adenosine triphosphate (ATP) synthesis and lipid deposition in growing pigs. FHP, fasting heat production; kBR, efficiency of using body reserves during fasting to supply energy for maintenance functions; kbasal, kstarch, kprotein and klipid, efficiency of using metabolizable energy from the basal diet, starch, protein (gluten or excess casein) or lipid, respectively, for lipid deposition of ATP synthesis; PD-free net energy, net energy not deposited as protein and available for lipid deposition of ATP synthesis.

	RESULTS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
General observations.

In general, the pigs consumed their daily ration and appeared to be in good health. For one pig (fed D2), a problem occurred with the automatic distribution of the meal. Another pig (fed D4) had an exceptionally low N deposition compared with other pigs in the same group (130 versus 190 g PD/d) and had diarrhea. Results for both of these pigs were not included in the statistical analysis. Pigs fed D1, D2 and D5 diets consumed their meals as intended (four meals per day), whereas those fed the protein supplements (D3 and D4 diets) tended to consume the distributed feed in more numerous but smaller meals.

Ileal and fecal digestibility.

Ileal and fecal digestibilities of the five diets are given in Table 2 . The average weight and DM feed intake of the five anastomosed pigs during the ileal digestibility study were 48.6 kg and 1.32 kg/d, respectively. Fecal digestibilities were determined as part of the N and energy balance studies. Except for fat, fecal digestibilities were higher than ileal digestibilities. Diet always affected the ileal digestibility, whereas animal and period affected the digestibility for five of the six variables tested. This period effect was due to a significantly lower digestibility in the first period (2.5 points lower for DM in period 1). This suggests that the pigs required a rather long period to adapt to the experimental conditions (e.g., cage, anastomosis). The DM, organic matter and energy digestibilities of diets D2–D5 were higher than those of diet D1, indicating that the dietary supplement was more digestible than the basal diet. This was confirmed by multiple regression analysis. Ileal DM digestibilities of both starch and casein were close to 100%, whereas those for gluten and lipid were somewhat lower (90 and 91%, respectively). The apparent ileal N digestibility of corn gluten meal (91%) was slightly lower than that of casein (96%). Amino acid digestibilities were similar for D1, D2 and D5 (all amino acids in these diets were supplied by the basal diet). Amino acid digestibilities of the protein supplements appeared higher than those of the basal diet.

Before the adaptation period, body weights of pigs designated to each diet were similar (50 kg). Consequently, differences in body weight among diets reported in Tables 3 and 4 are due to diet composition and feed intake level, which are cumulated during the adaptation and experimental period. The DM feed intake was lower for D5 than for D2–D4 due to its higher energy density (Table 3) . As anticipated, ingested and absorbed N was highest in D3 and D4. The N retention was by far the highest in D4. It did not differ among diets D2, D3 and D5, suggesting that amino acids supplied by corn gluten meal were deaminated and used for energetic purposes.

The TEF is calculated as the difference between the basal heat production (when fed) and the FHP. Considering only the supplemented diets, TEF was highest for D3 and D4 and lowest for D5. The high TEF for D4 was in part due to a numerically higher TEF_LT (P = 0.09), which is considered to reflect long-term metabolic processes such as protein and lipid deposition. The TEF_ST (representing the cost of intake and digestion) was of similar magnitude in D2–D4 but considerably lower for D5.

There was no difference in the heat production due to activity among the five diets. Heat production due to physical activity represented 13–20% of HP. Surprisingly, pigs fed D1 had the numerically highest HP_activity (P = 0.16), implying that a large proportion of ME was expended on activity and not on growth (13% of ME intake for D1, 7–8% of ME intake for diets D2–D5). In this group, close to 48% of HP_activity was spent standing without eating (24% in the other groups). Physical activity during eating (1 h/d) represented 16–20% of HP_activity. The residual standard deviation for HP_activity was of similar magnitude to that of total HP, suggesting important variation between individual pigs.

In fed pigs, the respiratory quotient was lowest for D1 and highest for D2. There was no difference (P = 0.68) between diets in the respiratory quotient during fasting. The ME/DE ratio was lowest for D3, slightly higher for D4 (it in part deposits, rather than fully deaminates, the supplement) and highest for D2 and D5. The ratio between TEF and ME was lowest for D5, slightly higher for D2 and highest for D3 and D4. This confirms the hypothesis that unbalanced proteins (D3) are deaminated and that the resulting energy is used less efficiently than energy from carbohydrates or lipids.

The residual standard deviations for DE, ME, PD and LD from the nested, multiple regression analysis were 252, 245, 300 and 391 kJ/d, respectively (there is no residual standard deviation for GE because it results from five equations with five unknowns). The GE contents of dietary supplements (Table 5 ) are similar to those found in most nutrition textbooks. Digestibilities of both starch and casein approached unity, whereas those for gluten and lipid were somewhat lower. All digestible starch and lipid were available for the animal without a noticeable energy loss in urine or as methane. In contrast, 12–16% of digestible protein energy was recovered in urine and thus not available for productive processes. A considerable proportion of ME from the basal diet and casein was used for PD (p_j; 13 and 42%, respectively). For both gluten and lipid, the proportion of ME used for PD did not differ from zero (P = 0.18). However, 4% of ME from starch was used for PD, which differed from zero (P < 0.01). Because all dietary supplements contained starch, this implies that between 5% (D4) and 15% (D2) of PD may be attributed to starch; the remainder is attributed to the basal diet alone for D2, D3 or D5 or to the basal diet (55%) and casein (40%) in D4. The efficiency of using PD-free ME for LD or ATP synthesis increased in the order protein, basal diet, starch and lipid (k_j in Table 5 ). The calculated energetic efficiencies of using PD-free ME for PD-free NE (k_diet) were 0.707, 0.755, 0.702, 0.715 and 0.763 for diets D1–D5, respectively. The use of BR for ATP synthesis (k_BR) was closer to the efficiency of protein than to those of the other supplements. For each 1 kJ of PD, 0.484 kJ of NE is required as support costs (i.e., ATP).

	DISCUSSION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

The diets were formulated so that lysine supply from the basal diet would be the first limiting factor for protein deposition. Consequently, it was anticipated that the supply of starch, corn gluten meal or lipid would affect only lipid deposition. The slightly higher N retention in D2 and D3 compared with D1 and D5 (Tables 3 and 4) was in part due to a greater apparently digestible lysine intake, on the one hand, and a relatively low N retention in one pig fed D1, on the other hand. The ratio between retained lysine (calculated as retained N x 6.25 x 0.07) and the apparently digestible lysine intake was 0.49 for D4; 0.80 for D1, D3 and D5; and 0.86 for D2. Because this ratio does not account for the maintenance lysine requirement and because the postabsorptive efficiency of lysine utilization is generally thought to be 0.70–0.80, lysine appeared to be the first limiting factor for PD in diets other than D4. Nevertheless, there was a small but significant PD response to the supply of starch (Table 5 ; p_starch 0). Although the idea of a single, first-limiting factor is appealing from a modeling point of view, pigs may not necessarily respond this way. The existence of multiple factors colimiting PD has been suggested for amino acids (12) .

Fasting and maintenance.

The technique of determining the energetic efficiency by adding nutrients to a basal diet and measuring the response in terms of energy retention dates back to more than a century ago. Kellner [cited by Blaxter (13) ] used this technique to determine the efficiency of nutrients relative to a maintenance ration fed to mature animals. The response that occurs appears to differ below and above maintenance energy requirements. Different models have been proposed that represent the relation between energy retention and ME intake [see Blaxter (13) and Emmans (14) for reviews]. Figure 3 represents the more or less classic view on the relation between ME intake and energy retention (i.e., the sum of PD and LD), on which many energy systems are based. The slope relating FHP to MEm is interpreted as the energetic efficiency for maintenance and reflects the feed efficiency relative to using energy from BR (i.e., FHP = k_m x MEm = k_diet/k_BR x MEm). In mature pigs, values for k_m are typically less than unity (15 ,16) , suggesting that dietary nutrients are used less efficiently than BR for maintenance purposes. In the present experiment, calculated k_m values range from 1.14 (D3) to 1.24 (D5) or from 0.84 (protein) to 1.43 (lipid). There probably are two reasons for the difference in k_m values between mature and growing animals. First, nutritional history (feed intake level) has been shown to affect FHP in growing pigs (17) , and data given in Table 4 at least do not contradict this. Because the relative feed intake will be lower in mature pigs, FHP will be more of an animal characteristic per se in mature pigs than in growing pigs. Consequently, in growing pigs, part of the diet effect is measured as FHP rather than considered as the inefficiency of nutrient utilization. A second explanation for differences in k_m may be due to differences in body composition (i.e., the nutrient source during fasting) between mature and growing pigs. Apart from having less body lipid, metabolically active organs (e.g., gastrointestinal tract and liver) will rapidly diminish in size in growing pigs during fasting. Constituents of these organs, such as protein, will then be available for ATP synthesis. The observation that k_BR was higher than k_protein but lower than k_basal, k_starch and k_lipid is consistent with this idea.

View larger version (9K): [in this window] [in a new window]   Figure 3. Traditional response curve of energy retention as a function of metabolizable energy (ME) supply in pigs. FHP, fasting heat production; kBR, efficiency of using BR during fasting to supply energy for maintenance functions; kdiet, marginal efficiency of using dietary ME for energy retention (above maintenance); km, efficiency of using dietary ME relative to that from BR; MEm, ME requirement for maintenance.

An important source of variation in energy expenditure is the physical activity. The coefficient of variation for physical activity was 2.5 times that of FHP, illustrating the variability among pigs. Physical activity was measured here in individually housed pigs with limited possibility for locomotion. It is therefore possible that under normal circumstances, the contribution of physical activity to heat production is even greater. This emphasizes the importance of accounting for physical activity in energy balance studies, especially when factors such as feeding level may affect the behavior of the pigs.

Relation with energy systems.

Energy systems are used to predict the energy value of feeds relative to energy requirements for maintenance and production. In contrast to systems based on ME, NE systems account for differences in utilization of dietary components as well as for differences in the composition of production. The current study is essentially an extension of the NE system (19 ,20) combined with a more recent approach concerning the dynamic components of heat production (6) .

NE is typically defined as the sum of retained energy and the FHP. In the current study, NE supply is defined as PD-free NE plus PD. The PD-free NE covers the total ATP requirement (i.e., maintenance, activity and protein deposition; k_BR x FHP + k_diet x HP_activity + NE_PD x PD), whereas the remainder is deposited as lipid. Consequently, the NE supply equals the sum of PD, LD and the ATP requirement. This ATP requirement, of an average of 746 kJ/(kg BW^0.60 · d¹), happens to be almost identical to the FHP used by Noblet et al. (19) to calculate the NE value of feeds [750 kJ/(kg BW^0.60 · d¹)]. Consequently, the observed efficiencies are close to those used by Noblet et al. (19) : 0.84, 0.52 and 0.88 versus 0.82, 0.58 and 0.90 for starch, protein and lipid, respectively. Similarly, apart from D4, calculated NE values were very similar to those predicted by Noblet et al. (19) (Table 4) . The difference for D4 is due to the fact that part of the casein is deposited as protein rather than deaminated and used for other energetic purposes, which implies a loss of energy (k_diet). The ATP cost of the additional PD induces an increased energy requirement (at the expense of LD) but without changing the supply of PD-free NE. This results in the ambiguous situation that the NE value of protein is not constant but instead depends on whether protein is deposited as protein or is used for other energetic purposes. This is illustrated in Table 5 , where the calculated NE value of casein is 50% greater than that of corn gluten meal (42% of the ME of casein was deposited as protein, whereas for corn gluten meal this was not different from 0%). With this reasoning, the NE value of protein theoretically ranges between than of gluten (all protein deaminated) and the DE value of protein (all protein deposited as protein). Casein is a high-quality protein but was fed in large excess of requirements (23% crude protein in D4), thereby considerably reducing its NE value. The NE value of proteins used in typical diets may be even >14 kJ/g [assuming 90% digestibility, 50% of protein DE deposited as PD, 15% energy loss in the urine and k_protein = 0.52; hence 23.7 x 0.90 x (0.50 + 0.50 x 0.85 x 0.52) = 15.4 kJ/g]. However, once protein requirements are met, the NE value of the additional protein supply would be close to 10 kJ/g. The results given in Table 5 are consistent with other work from our laboratory. Le Bellego et al. (21) reported that the supply of excess protein (compared with starch) increased urinary energy loss by 3.5 kJ/g and heat production by 7 kJ/g protein. Using the data from the current study (i.e., corn gluten meal compared with starch), these values are 3.4 and 6.0 kJ/g, respectively.

The parameter estimates given in Table 5 allow the calculation of several traditional indicators of energy metabolism, such as k_p and MEm. The k_p (i.e., ratio of PD to the corresponding ME input) can be calculated from NE_PD and the energy source for ATP synthesis. For example, if excess protein is used to provide ATP, k_p would equal 1/(0.484/0.520 + 1) = 0.52, whereas if starch or lipid was used, k_p would equal 0.63 and 0.65, respectively. It is interesting to note that the efficiencies of using dietary protein for PD (including the associated ATP cost) or LD are identical. The MEm (including physical activity) can be obtained from FHP x k_BR/k_diet + HP_activity and averaged as 849 kJ/(kg BW^0.60 · d^-1) across diets. This value is similar to that reported previously for growing pigs fed at different feed intake levels (22 ,23) . However, there are indications that MEm is higher [1 MJ/(kg BW^0.60 · d¹)] in animals consuming food ad libitum (24 ,25) .

Theoretical efficiency of nutrient utilization.

The efficiency with which nutrients can be utilized for lipid deposition or ATP synthesis depends, on the one hand, on the biochemical transformation of the nutrient and, on the other hand, on biophysical and physiological processes (e.g., transport, synthesis of enzymes). Baldwin (26) estimated the biochemical efficiency of synthesizing tripalmitin from glucose at 83.8%, which is similar to the value in the current experiment. Consequently, the current value seems rather high, because the costs of digestion and intermediary metabolism, such as glycogen storage, are not accounted for.

The biochemical efficiency of depositing dietary lipid as LD should be close to unity. The main energy cost involved seems to be the activation of fatty acids for reesterification of triacylglycerides (e.g., after hydrolysis of dietary lipid in the intestine or lipoproteins in adipose tissue). However, with 2-fold hydrolysis and reesterification, the loss of energy amounts to only 3% [i.e., 6 ATP molecules per hydrolysis/esterification, 0.074 MJ/mol ATP and 31.8 MJ/mol tripalmitin according to Baldwin (26) ; hence, 2 x 6 x 0.074/31.8]. However, dietary lipid not only can be deposited but also can be used for ATP synthesis. According to Armstrong (27) , the energy cost per ATP molecule synthesized is even slightly greater for lipids than for glucose (75.7 versus 74.1 kJ/ATP, respectively). The measured energetic efficiency therefore is a combination of using dietary lipid for both LD and ATP synthesis. There are conflicting reports on the extent to which dietary lipids contribute to maintenance requirements and, consequently, are catabolized in growing pigs. Based on a slaughter study, Flanzy et al. (28) recovered only half of the absorbed linoleic acid in the body. In contrast, Chwalibog et al. (29) used calorimetry data in combination with respiratory quotients of nutrient utilization to conclude that all dietary fat is retained in the body.

Most information concerning the biochemical efficiency of dietary amino acids is based on their utilization for ATP synthesis. However, because glucose and amino acids share a common intermediate for ATP and fatty acid synthesis (acetyl coenzyme A), it can be anticipated that both modes of expression (i.e., percent of energy input retained as lipid and kJ/ATP) reflect a similar phenomenon. As seen earlier, this contrasts with lipids where acetyl coenzyme A is not a common intermediate, resulting in different biochemical efficiencies. Data from Krebs (30) combined with the heat of combustion for amino acids and urea (31) indicate that the cost of ATP synthesis is higher for amino acids than that for glucose, ranging from 77.4 kJ/ATP for glutamate to 119.7 kJ/ATP for cysteine. For casein, this amounts to 89.5 kJ/ATP (27) , hence being 20% less efficient than glucose. However, the observed efficiency is 38% lower for excess protein than that for starch (Table 4) , suggesting that biochemistry accounts for a considerable part, but not all, of the difference.

An additional explanation for differences in energetic efficiency between protein and carbohydrate is the increased body protein turnover associated with high protein diets (32 ,33) . The repeated synthesis and breakdown of protein probably involve considerable energy expenditure. Synthesis of a peptide bond requires (at least) 5 ATP molecules, and also peptide bond hydrolysis has been associated with an ATP requirement. As indicated, the energetic efficiencies of using dietary protein for PD and LD are very similar. Nevertheless, at least one additional cycle of protein synthesis is required for amino acids to be deposited as PD compared with those deposited as LD. On the other hand, deposition of amino acids as LD implies an additional energy cost of 2 ATP/N for urea synthesis. A simple calculation using the ATP yield for glutamate (77.4 kJ/ATP) and assuming a requirement of 5 ATP molecules per peptide bond shows that the most efficient peptide formation (i.e., no breakdown) would equal 2570/(2570 + 5 x 77.4) = 0.87. If no ATP requirement is assumed for peptide hydrolysis, three turnover cycles before deposition (i.e., four times synthesis and three times breakdown) would result in an efficiency of 0.63.

The current experiment clearly showed that different nutrients are used with different energetic efficiencies. The use of a relatively simple modeling technique allowed extensive exploitation of the data (within the limits of the underlying model assumptions). Biochemical transformations explain to a large extent the observed efficiency of carbohydrate and lipid utilization. Although a considerable part of the (in)efficiency of protein utilization can be explained by biochemistry, other phenomena are implied. The observation that dietary protein is as efficiently deposited as lipid or as protein suggests that diet-induced protein turnover may play a role in this.

	ACKNOWLEDGMENTS

 
Technical support provided by R. Vilboux, A. Roger, P. Bodinier and S. Hillion and discussions on experimental design and data interpretation with Agribrands International and C.F.M. de Lange and S. Birkett (University of Guelph) are appreciated.

	FOOTNOTES

 
¹ Presented in part at the 15th Symposium on Energy Metabolism in Animals, September 11–16, 2000, Snekkersken, Denmark [van Milgen, J., Noblet, J. & Dubois, S. (2001) Energetic efficiency of nutrient utilization in growing pigs. In: Energy Metabolism in Animals (Chwalibog, A. & Jakobsen, K, eds.). Wagenigen Pers, Wageningen, the Netherlands].

² Supported by Agribrands International.

⁴ Abbreviations used: ATP, adenosine triphosphate; BR, body reserves; BW, body weight; dc, digestibility coefficient; D1, basal diet limiting in lysine supply; D2, diet containing the basal diet plus starch; D3, diet containing the basal diet plus starch and unbalanced protein; D4, diet containing the basal diet plus starch and balanced protein; D5, diet containing the basal diet plus starch and lipid; DE, digestible energy; DM, dry matter; DMI, dry matter intake; F, ingredient fraction in the diet; FHP, fasting heat production; HP_activity, heat production due to physical activity; k, efficiency of using metabolizable energy for net energy; LD, lipid deposition; mc, metabolizability coefficient; ME, metabolizable energy, MEm, metabolizable energy requirement for maintenance; N, nitrogen; NE, net energy; NE_PD, net energy requirement for protein deposition; p, fraction of metabolizable energy deposited as protein; PD, protein deposition; TEF, thermic effect of feeding; TEF_ST, short-term thermic effect of feeding; TEF_LT, long-term thermic effect of feeding.

Manuscript received September 6, 2000. Initial review completed October 28, 2000. Revision accepted December 19, 2000.

	REFERENCES

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 

1. National Research Council Nutrient Requirements of Swine 10th ed. 1998 National Academy Press Washington, D.C.

2. Sève B. Alimentation du porc en croissance: intégration des concepts de protéine idéale, de digestibilité digestive des acides aminés et d’énergie nette. INRA Prod. Anim. 1994;7:275-291

3. Hess V., Sève B. Effects of body weight and feed intake level on basal ileal endogenous losses in growing pigs. J. Anim. Sci. 1999;:3281-3288

4. Association of Official Analytical Chemists Official Methods of Analysis 15th ed. 1990 Association of Official Analytical Chemists Arlington, VA.

5. Brouwer E. Report of sub-committee on constants and factors. Energy Metabolism 1965 Proceedings of the 3rd Symposium Held at Troon Scotland (Blaxter, K. L. ed.), pp. 441–443. Academic Press, London, U.K.

6. van Milgen J., Noblet J., Dubois S., Bernier J. F. Dynamic aspects of oxygen consumption and carbon dioxide production in swine. Br. J. Nutr. 1997;78:397-410[Medline]

7. van Milgen J., Noblet J. Modelling energy expenditure in pigs. McNamara J. P. France J. Beever D. E. eds. Modelling Nutrient Utilization in Farm Animals 2000:103-114 CAB International Oxon, U.K.

8. AEgis Simulation Inc ACSL Optimize, version 2.4 1999 AEgis Simulation, Inc Huntsville, AL.

9. SAS Institute SAS/STAT User’s Guide, version 6 4th ed. 1989 SAS Institute Inc Cary, NC.

10. Neter J., Kutner M. H., Nachtsheim C. J., Wasserman W. Applied Linear Statistical Models 4th ed. 1996 Irwin Burr Ridge, IL.

11. Ratkowsky D. A. Nonlinear Regression Modeling: A Unified Practical Approach 1983 Marcel Dekker, Inc New York, NY.

12. Susenbeth A. Factors affecting lysine utilization in growing pigs: an analysis of literature data. Livest. Prod. Sci. 1995;43:193-204

13. Blaxter K. L. The Energy Metabolism of Ruminants 1962 Hutchinson Scienctific and Technical London, U.K.

14. Emmans G. C. Models of the relationship between animal performance and dietary energy. Moughan P. J. Verstegen M.W.A. Visser-Reyneveld M. I. eds. Modelling Growth in the Pig 1995:47-58 Wageningen Pers Wageningen, the Netherlands.

15. Blaxter K. L., Boyne A. W. Fasting and maintenance metabolism of sheep. J. Agric. Sci. (Camb.) 1982;99:611-620

16. Noblet J., Shi X. S., Dubois S. Metabolic utilization of dietary energy and nutrients for maintenance energy requirements in sows: basis for a net energy system. Br. J. Nutr. 1993;70:407-419[Medline]

17. Koong L. J., Nienaber J. A., Pekas J. C., Yen J. T. Effects of plane of nutrition on organ size and fasting heat production in pigs. J. Nutr. 1982;112:1638-1642

18. Knap, P. W. (2000) Variation in Maintenance Requirements of Growing Pigs in Relation to Body Composition. PH.D. thesis, Wageningen Agricultural University, Wageningen, the Netherlands.

19. Noblet J., Fortune H., Shi X. S., Dubois S. Prediction of net energy value of feeds for growing pigs. J. Anim. Sci. 1994;72:344-354[Abstract]

20. Noblet J., Shi X. S., Dubois S. Effect of body weight on net energy value of feeds for growing pigs. J. Anim. Sci. 1994;72:648-657[Abstract]

21. Le Bellego, L., van Milgen, J., Dubois, S. & Noblet, J. (2001) Energy utilization of low protein diets in growing pigs. J. Anim. Sci. (in press).

22. Quiniou N., Dourmad J. Y., Noblet J. Effect of energy intake on the performance of different types of pig from 45 to 100 kg body weight. 1. Protein and lipid deposition. Anim. Sci. 1996;63:277-288

23. van Milgen J., Quiniou N., Noblet J. Modelling the relation between energy intake and protein and lipid deposition in growing pigs. Anim. Sci. 2000;71:119-130

24. Noblet J., Karege C., Dubois S., van Milgen J. Metabolic utilization of energy and maintenance requirements in growing pigs: effects of sex and genotype. J. Anim. Sci. 1999;77:1208-1216[Abstract/Free Full Text]

25. van Milgen J., Noblet J. Energy partitioning in growing pigs: the use of a multivariate model as an alternative for the factorial analysis. J. Anim. Sci. 1999;77:2154-2162[Abstract/Free Full Text]

26. Baldwin R. L. Modeling Ruminant Digestion and Metabolism 1995 Chapman & Hall London, U.K.

27. Armstrong D. G. Cell bioenergetics and energy metabolism. Lenkeit W. Breirem K. Crasemann E. eds. Handbuch Der Tierernährung 1969:385-414 Parey Hamburg, Germany.

28. Flanzy J., François A. C., Rérat A. Utilisation métabolique des acides gras chez le porc. Ann. Biol. Anim. Biochem. Biophys. 1970;10:603-620

29. Chwalibog A., Jakobsen K., Henckel S., Thorbek G. Estimation of quantitative oxidation and fat retention from carbohydrate, protein and fat in growing pigs. J. Anim. Physiol. Anim. Nutr. 1992;68:123-135

30. Krebs H. A. The metabolic fate of amino acids. Munro H. N. Allison J. B. eds. Mammalian Protein Metabolism 1964:125-176 Academic Press New York, NY.

31. Livesey G. The energy equivalents of ATP and the energy values of food proteins and fats. Br. J. Nutr. 1984;51:15-28[Medline]

32. Reeds P. J., Fuller M. F., Cadenhead A., Lobley G. E., McDonald J. D. Effects of changes in the intakes of protein and non-protein energy on whole-body protein turnover in growing pigs. Br. J. Nutr. 1981;45:539-546[Medline]

33. Roth F. X., Gotterbarm G. G., Windisch W., Kirchgessner M. Influence of dietary level of dispensable amino acids on nitrogen balance and whole-body protein turnover in growing pigs. J. Anim. Physiol. Anim. Nutr. 1999;81:232-238

This article has been cited by other articles:

				E. Labussiere, S. Dubois, J. van Milgen, G. Bertrand, and J. Noblet Effects of Dietary Crude Protein on Protein and Fat Deposition in Milk-Fed Veal Calves J Dairy Sci, December 1, 2008; 91(12): 4741 - 4754. [Abstract] [Full Text] [PDF]

				G. Lopez and S. Leeson Assessment of the Nitrogen Correction Factor in Evaluating Metabolizable Energy of Corn and Soybean Meal in Diets for Broilers Poult. Sci., February 1, 2008; 87(2): 298 - 306. [Abstract] [Full Text] [PDF]

				C. S Johnston, S. L Tjonn, P. D Swan, A. White, H. Hutchins, and B. Sears Ketogenic low-carbohydrate diets have no metabolic advantage over nonketogenic low-carbohydrate diets Am. J. Clinical Nutrition, May 1, 2006; 83(5): 1055 - 1061. [Abstract] [Full Text] [PDF]

				J. Noblet and J. van Milgen Energy value of pig feeds: Effect of pig body weight and energy evaluation system J Anim Sci, January 1, 2004; 82(13_suppl): E229 - 238. [Abstract] [Full Text] [PDF]

				J. van Milgen and J. Noblet Partitioning of energy intake to heat, protein, and fat in growing pigs J Anim Sci, February 1, 2003; 81(14_suppl_2): E86 - 93. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) Purchase Article View Shopping Cart Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by van Milgen, J. Articles by Dubois, S. Search for Related Content PubMed PubMed Citation Articles by van Milgen, J. Articles by Dubois, S.