Introduction
The effects of age on lipid levels in poultry muscle depend on the species. In chickens, lipid levels in breast and thigh muscles remain stable or decrease with age, at least until 100 days (Touraille et al., Reference Touraille, Ricard, Kopp, Valin and Leclercq1981; Hamm and Ang, Reference Hamm and Ang1984; Rabot, Reference Rabot1998) and then, according to the few reports available for older chickens, lipid levels possibly increase (Marion and Miller, Reference Marion and Miller1968; Grey et al., Reference Grey, Robinson, Jones, Stock and Thomas1983). Wangen et al. (Reference Wangen, Marion and Hotchkiss1972) reported that lipid levels in the breast muscle remained unchanged in turkeys killed at intervals of 4 weeks from 4 to 28 weeks, whereas they increased in thigh muscle after 16 weeks of age. Komprda et al. (Reference Komprda, Sarmanova, Zelenka, Bakaj, Fialova and Fajmonova2002) found significant increases in lipid levels in breast and thigh muscles of turkeys slaughtered at 1-week intervals between 10 and 25 weeks of age. In ducks, only Setiawan et al. (Reference Setiawan, Babilé, Auvergne, Belveze and Latil1994) and Baéza et al. (Reference Baéza, Salichon, Marché, Wacrenier, Dominguez and Culioli2000) and (Reference Baéza, Dessay, Wacrenier, Marché and Listrat2002) showed an increase in lipid content of breast muscle of mule and Muscovy ducks between 6–8 and 12–13 weeks of age. Lipid content of muscle can also be manipulated by feeding levels, with feed restriction often decreasing intramuscular fat content and overfeeding inducing an increase. In particular, Auvergne (Reference Auvergne1992) showed that lipid levels in the breast meat of overfed ducks were 2.0-fold higher than in ducks fed ad libitum. In practice, the overfeeding period often lasts for 2 weeks and occurs after a period of feed restriction followed by a pre-overfeeding period. The aim of the feed restriction is to induce later a catch-up growth that may maximise the effects of overfeeding. During the pre-overfeeding period birds are trained to ingest increasing amounts of feed. The mule duck is the main species used to produce fatty liver.
It was therefore very interesting to study the effects of age on intramuscular lipid deposition in mule ducks usually reared for overfeeding, since this type of production also enabled us to study the effects of feeding levels. Both lipids and glucose represent energy sources for muscle metabolism, and their levels depend on circulating blood levels, and on the uptake and oxidation ability of muscles (Hocquette et al., Reference Hocquette, Ortigues-Marty, Pethick, Herpin and Fernandez1998a). It is generally accepted that muscle oxidative metabolism is prevalent at the beginning of postnatal development and is progressively replaced by glycolytic metabolism, as described by Bacou and Vigneron (Reference Bacou and Vigneron1976) in chicks, Briand et al. (Reference Briand, Boissonet, Laplace-Marieze and Briand1993) in rabbits and Baéza et al. (Reference Baéza, Salichon, Marché and Juin1998) in Muscovy ducks. These changes can be modified by feeding levels. For example, in one study, overfed ducks had higher β-hydroxyacyl CoA dehydrogenase (HAD) activity in breast muscles than ad libitum-fed ducks (Baéza et al., Reference Baéza, Rideau, Chartrin, Davail, Hoo-Paris, Mourot, Guy, Bernadet, Juin, Meteau and Hermier2005). The aim of this study was therefore to determine the effects of age (from 1 day post hatching to 98 days of age) and feeding levels (feed restriction followed by overfeeding v. ad libitum feeding) on lipid deposition in breast muscle (quantity and quality, localisation) of mule ducks compared with muscle energy metabolism (glycolytic and oxidative), plasma levels of lipids, glucose and insulin, and muscle capacity for lipid uptake (characterised by lipoprotein lipase activity).
Material and methods
Animals and experimental design
We used 140 male mule ducks provided by Grimaud (Roussay, France). The birds were distributed into two groups and reared under the usual conditions of light and temperature at the Experimental Station for Waterfowl Breeding (INRA Artiguères, France). From hatching to 42 days of age, all ducks were fed ad libitum. From 42 to 76 days of age, the ducks of the restricted then overfed group (ROF) were fed restricted (260 g of feed per duck per day between 43 and 68 days of age and 230 g of feed per duck per day between 69 and 76 days). Feed amounts were then progressively increased to reach 350 g at 83 days of age. At 84 days of age, ducks were overfed for 14 days with maize and maize meal (329 g at 84 days to reach 1532 g at 97 days of age). From 42 to 97 days of age, ducks of the ad libitum-fed group (ALF) were fed ad libitum with the growing diet. The main characteristics of the diets (starting, growing and overfeeding) are shown in Table 1. The details of diet composition have already been presented by Chartrin et al. (Reference Chartrin, Bernadet, Guy, Mourot, Duclos and Baéza2006). Feed consumption was recorded for each period per group.
† ME = metabolisable energy; CP = crude protein; SFA, MUFA, PUFA = saturated, mono-unsaturated and poly-unsatured fatty acids, respectively.
At 1, 14, 42, 75, 84, 91 and 98 days of age, 10 ducks per group, deprived of feed overnight, were randomly selected and weighed. Blood samples were drawn by puncture of the occipital venous sinus, collected on EDTA (1.2 g/l) in a vacuum tube and kept at 2- to 4°C during the subsequent procedure. Individual plasma samples were separated by centrifugation at 2000 g for 10 min at 4°C. Plasma samples were then frozen at − 20°C until further analysis. Ducks were then sacrificed by cervical dislocation. Immediately after bleeding, the liver and breast muscle (pectoralis major) were excised and weighed. One sample of p. major was quickly frozen in isopentane, cooled with liquid nitrogen and stored at − 80°C until histochemical analysis was performed, as previously described by Baéza et al. (Reference Baéza, Marche and Wacrenier1999). A sample of the p. major muscle was quickly frozen in liquid nitrogen and stored at − 80°C prior to determination of enzymatic activity. A sample of the p. major muscle was also frozen and stored at − 20°C for chemical analysis.
The present study was carried out in accordance with the French legislation on animal experimentation and with the authorisation of the French Ministry of Agriculture (Animal Health and Protection Directorate).
Histological analysis
The levels of triglycerides (stained with red oil = RO), and total lipid (stained with Sudan black = SB), were determined in muscle fibres according to type as previously described by Chartrin et al. (Reference Chartrin, Bernadet, Guy, Mourot, Duclos and Baéza2005): fast-twitch fibres with glycolytic energy metabolism (αW) or oxydo-glycolytic energy metabolism (αR). Staining intensities were expressed as luminance (Lc RO and Lc SB), with luminance being negatively correlated with lipid and triglyceride contents. We also evaluated the relative surfaces occupied by adipocytes on cross-sectional areas of each muscle sample as described by Chartrin et al. (Reference Chartrin, Bernadet, Guy, Mourot, Duclos and Baéza2005) and the results were expressed as the relative surface occupied by adipocyte cluster per observed field (%).
Blood analysis
Plasma glucose levels were determined enzymatically using a Beckman autoanalyser and the corresponding kit (Kadish et al., Reference Kadish, Little and Sternberg1968).
Plasma triglyceride (TG), total cholesterol (CT) and phospholipid (PL) levels were quantified by colorimetric enzymatic methods according to Fossati and Prencipe (Reference Fossati and Prencipe1982), Richmond (Reference Richmond1973) and Takayama et al. (Reference Takayama, Itoh, Nagasaki and Tanimizu1977), respectively, and using kits provided by Bio-Mérieux (Marcy-l'Etoile, France).
Plasma insulin levels were determined by radioimmunoassay as previously described by Simon et al. (Reference Simon, Freychet and Rosselin1974). The lowest limit of sensitivity was 2 μU/ml, and interassay variation was less than 10%.
Lipid analysis
Total lipids were extracted from muscle, by homogenising samples of minced tissue in chloroform/methanol 2/1 (v/v) and collecting gravimetrically (Folch et al., Reference Folch, Lees and Sloane Stanley1957). Lipid classes were determined using an Iatroscan (Iatron, Tokyo, Japan) with 10 silica-gel thin layer chromatography rods and a flame ionisation detector system (TLC-FID) according to Mares et al. (Reference Mares, Ranny, Sedlacek and Skorepa1983). The hydrogen flow rate was 160 ml/min, the air flow rate 2 l/min and scanning speed 0.3 cm/s. The software used (Boreal, JMBS Development, Grenoble, France) recorded chromatograms and integrated peaks with reference to an external standard (Sigma, St Quentin Fallavier, France). Fatty acid composition was determined after transmethylation of lipids (Morrisson and Smith, Reference Morrisson and Smith1964) by gas chromatography (Perkin Elmer Autosystem, St Quentin en Yvelines, France). Injector and detector (FID) temperatures were 250°C, and the carrier gas was nitrogen with a head column pressure of 16.5 p.s.i. using a capillary column (25 m × 0.22 mm, BPX70, SGE, Villeneuve St Georges, France). Methyl esters were identified and quantified by comparison with standards (Sigma, St Quentin Fallavier, France). Because of the small amounts of muscles at 1 and 14 days of age, analysis of lipids was performed with one pool of 20 muscles and two pools of 10 muscles, respectively.
Determination of enzyme activity
Lipoprotein lipase (LPL) activity was determined in the p. major muscle according to Hocquette et al. (Reference Hocquette, Graulet and Olivecrona1998b). The activity of the enzymes for glycolytic (lactate dehydrogenase, LDH) and oxidative (β-hydroxyacyl CoA dehydrogenase, HAD and citrate synthase, CS) muscle energy metabolism was assayed in the p. major muscle (Bass et al., Reference Bass, Brdiczka, Eyer, Hoper and Pette1969).
Statistical analysis
Data were analysed by analysis of variance using the general linear model procedure of Statistical Analysis Systems Institute (1989). Age effects were only tested in the ALF group between 1 and 98 days of age, and the effects of feeding levels were tested at each slaughter age between 75 and 98 days. By taking into account the values for all ages and feeding levels, we calculated Pearson correlations between LPL activity and lipid levels in the breast muscle, or phospholipid, triglyceride and cholesterol levels in plasma.
Results
Growth performance
Feed consumption was similar in both groups between 1 and 42 days of age (Table 2). Feed restriction corresponded to 79% of ad libitum ingestion between 43 and 75 days of age. Ducks from the ROF group ingested 0.3 kg less than ducks from the ALF group during the pre-overfeeding period (Table 2). At 83 days of age, ad libitum-fed ducks had ingested about 2.5 kg more feed than restricted ducks, and between 84 and 97 days of age overfed ducks had ingested three times the amount of feed ingested by ad libitum-fed ducks (Table 2).
The greatest increases in body weight and liver weight occurred between 1 and 14 days of age (10- and 15-fold, respectively), whereas the greatest increases in weight of the p. major muscle occurred between 14 and 42 days of age (27-fold, Table 3). Body weight remained stable in ALF ducks after 84 days of age. Feed restriction exerted no significant effect on body weight or on weight of the p. major muscle but it increased liver weight (+24% and +14% at 75 and 84 days of age, respectively) and the relative weight of liver to body weight (1.66 v. 1.40% and 1.74 v. 1.48% at 75 and 84 days of age, respectively). Overfeeding induced a marked increase in body weight (+12% and +41% after 1 and 2 weeks, respectively) and liver weight (2.8-fold and 6.7-fold heavier after 1 and 2 weeks, respectively) while the weight of the p. major muscle decreased during the 1st week of this period ( − 12%). At the end of the overfeeding period ducks of both groups exhibited similar muscle weights (Table 3).
a-f Significant effect of age for ad libitum-fed group (P < 0.05).
* Significant effect of feeding levels for a given age (between 75 and 98 days) (P < 0.05).
Blood parameters
Circulating glucose concentration increased between 1 and 14 days of age, then remained stable in the ALF group (Table 4). Neither feed restriction nor overfeeding significantly affected glycaemia levels (Table 4).
a-d Significant effect of age for ad libitum-fed group between 1 and 98 days of age (P < 0.05).
* Significant effect of feeding levels for a given age between 75 and 98 days (P < 0.05).
Plasma insulin levels remained stable until 75 days, and then significantly increased at 91 and 98 days in the ALF group (Table 4). Feed restriction resulted in a significant decrease ( − 25% at 75 days of age) and 2 weeks of overfeeding resulted in a significant increase in plasma insulin level (+67% at 98 days of age, Table 4).
Plasma triglyceride levels first increased between 1 and 14 days then remained stable until 98 days, except for the lowest value measured at 91 days in the ALF group (Table 4). Feed restriction had no significant effect, while overfeeding resulted in considerable increases in plasma triglyceride levels (3.7-fold and 5.1-fold at 91 and 98 days of age, respectively, Table 4).
Plasma phospholipid levels were high at 1 and 14 days of age (Table 4), then decreased and remained stable at around 3.2 mmol/l until the end of the experiment in the ALF group. Feed restriction had no significant effect, while overfeeding resulted in an increase in plasma phospholipid levels (1.3-fold and 1.6-fold at 91 and 98 days of age, respectively, Table 4).
Plasma cholesterol levels were high at 1 and 14 days of age (Table 4), then decreased and remained stable at around 4.34 mmol/l until the end of the experiment in the ALF group. Feed restriction had no significant effect, while overfeeding resulted in an increase in plasma cholesterol levels (1.4-fold and 1.6-fold at 91 and 98 days of age, respectively, Table 4).
Lipid composition of breast muscle
In the ALF group, muscle lipid content first decreased from 1 to 42 days of age ( − 57%, Table 5), and then increased significantly between 42 and 75 days of age (2.2-fold) and between 91 and 98 days of age (1.3-fold). Feed restriction significantly decreased muscle lipid content ( − 16%). Pre-overfeeding resulted in an increase in muscle lipid content, reaching similar levels to those observed in ad libitum-fed ducks at 84 days of age. Overfeeding resulted in an increase in muscle lipid content, particularly during the 2nd week (+12% and +40% at 91 and 98 days of age, respectively, Table 5).
a-d Significant effect of age for ad libitum-fed group between 42 and 98 days (P < 0.05); nd = not detected.
* Significant effect of feeding levels for a given age between 75 and 98 days (P < 0.05).
†,‡ Analysis performed with a pool of 20 muscles or two pools of 10 muscles, respectively.
Muscle triglyceride content increased in the ALF group between 1 and 14 days of age (+85%) then markedly decreased (value divided by 6.6 between 14 and 42 days of age, Table 5), but from 75 to 98 days of age it increased steadily (+98%). Feed restriction decreased ( − 31%) and overfeeding increased muscle triglyceride content, particularly during the 2nd week (+22% and +60% at 91 and 98 days of age, respectively, Table 5).
Muscle phospholipid content decreased in the ALF group between 1 and 14 days of age ( − 39%, Table 5), then increased regularly until 98 days of age (1.7-fold). Feed restriction had no significant effect while overfeeding slightly increased muscle phospholipid content at 98 days of age (+14%, Table 5).
The levels of esterified cholesterol in breast muscle, which was maximal at 1 day of age, dropped at 14 days of age (initial value divided by 36). A further decrease occurred until 84 days when it became undetectable (Table 5). At 1 day of age, the free cholesterol levels in breast muscle were high (Table 5), then decreased until 42 days of age (divided by 2.8). We recorded higher levels at 75 days of age, followed by a decrease at 84 days. The levels then remained stable at 84 and 98 days of age. Neither feed restriction nor overfeeding had significant effects on esterified or free cholesterol levels in breast muscle.
The main fatty acids in breast muscle were palmitic (C16:0), stearic (C18:0), oleic (C18:1n-9), linoleic (C18:2n-6) and arachidonic (C20:4n-6) acids (Table 6). Saturated, mono-unsaturated and poly-unsaturated fatty acid (SFA, MUFA and PUFA, respectively) levels decreased between 1 and 42 days of age in the ALF group, values being divided by 2.0, 3.0 and 1.7, respectively. Levels then increased regularly until 98 days of age (2.9-fold, 3.6-fold and 2.6-fold for SFA, MUFA and PUFA, respectively). Feed restriction decreased SFA, MUFA and PUFA levels in breast muscle ( − 12%, − 20% and − 11%, respectively, Table 6) and overfeeding increased only SFA and MUFA levels (+49% and +56%, respectively, at 98 days of age).
a-d Significant effect of age for ad libitum-fed group between 42 and 98 days (P < 0.05); nd = not detected.
* Significant effect of feeding levels for a given age between 75 and 98 days, (P < 0.05).
† SFA, UFA, PUFA = saturated, unsaturated and poly-unsatured fatty acids, respectively.
‡,§ Analysis performed with a pool of 20 muscles or two pools of 10 muscles, respectively.
Histological characteristics of breast muscle
The surface occupied by adipocytes on cross-sections of breast muscle was maximal at 1 and 14 days of age then decreased until 75 days of age (value divided by 12.3 between 1 and 75 days, Table 7). It then increased regularly until 98 days of age (6-fold). At 1 and 14 days of age the adipocytes were distributed around the fibre bundles (in the connective tissue) and also inside (between the muscle fibres) while at 98 days of age the adipocytes were mostly distributed around the fibre bundles (in the connective tissue and mainly around the vessels, Figure 1).
a-c Significant effect of age for ad libitum-fed group between 1 and 98 days (P < 0.05).
* Significant effect of feeding levels for a given age between 75 and 98 days (P < 0.05).
† αR and αW fibres: oxydo-glycolytic and glycolytic fast-twitch fibres, respectively. Luminance was negatively correlated with lipid levels.
In the P. major muscle, the average percentages of type αR and αW fibres were 81 and 19%, respectively. The cross-sectional area (CSA) of muscle fibres increased with age (96-fold between 1 and 98 days of age in the ALF group, Table 7). The greatest increase in fibre CSA occurred between 14 and 42 days of age (15-fold and 18-fold in αR and αW fibres, respectively). At 98 days of age, the CSA of αW fibres was 2.8-fold greater than that of αR fibres.
The total lipid content of αR fibres increased between 42 and 75 days of age in the ALF group (Table 7). It then remained stable until 98 days of age. The total lipid content of αW fibres was not affected by age. The triglyceride content of αR fibres increased (+51%) between 1 and 75 days of age in the ALF group. It then remained stable until 91 days of age before decreasing ( − 36%) significantly between 91 and 98 days of age. The triglyceride content of αW fibres increased (+57%) between 1 and 84 days of age. It then remained stable until 98 days of age.
Neither feed restriction nor overfeeding had a significant effect on the surface occupied by adipocytes, fibre typology (data not shown), fibre CSA or total lipid or triglyceride content in muscle fibres (Table 7).
Enzyme activity
LDH activity increased steadily in the ALF group between 1 and 91 days of age (8-fold, Table 7). The greatest increase in LDH activity occurred between 14 and 42 days of age (3.4-fold), but between 91 and 98 days of age, it decreased significantly ( − 23%). CS activity increased regularly with age in the ALF group (1.9-fold between 1 and 98 days of age, Table 8). HAD activity remained at low levels between 1 and 42 days of age. It then increased strongly (6.7-fold between 42 and 75 days of age) to remain stable until 98 days of age. In the ALF group, LPL activity was high at 1 and 14 days of age. It decreased considerably at 42 days of age (value divided by 6 between 14 and 42 days), increased significantly at 75 days of age (2.5-fold) to remain stable until 98 days of age (Table 8).
a-f Significant effect of age for ad libitum-fed group between 1 and 98 days (P < 0.05).
* Significant effect of feeding levels for a given age between 75 and 98 days (P < 0.05).
Feed restriction had no significant effect on LDH, CS, HAD or LPL activity. Overfeeding decreased LDH activity, particularly during the 2nd week ( − 22% at 98 days of age), and increased LPL activity, particularly during the 1st week (2.3-fold and 1.8-fold at 91 and 98 days of age, respectively, Table 8). It had no significant effect on CS or HAD activity.
Correlations between LPL activity and lipid levels in breast muscle, or phospholipid, triglyceride and cholesterol levels in plasma were the following: +0.33 (P < 0.01),+0.45 (P < 0.01),+0.13 (P>0.05) and +0.51 (P < 0.001), respectively.
Discussion
Age effects
The kinetics of fat deposition in the breast muscle might be divided into two periods, i.e. before and after 42 days of age, the greatest effects of age occurring between 14 and 42 days when the growth of the breast muscle was maximal (weight increased 27-fold, CSA increased 15 and 18-fold in αR and αW fibre types, respectively). The energy metabolism of birds is based on the use of lipids stored in egg yolk until hatching (Noble and Cocchi, Reference Noble and Cocchi1990; Klasing, Reference Klasing and Klasing1998). Total lipid content in the breast muscle of mule ducks was therefore high at 1 day of age (2.9 g per 100 g of muscle) and mainly composed of phospholipids (50% of total lipids) and cholesterol (29%), reflecting the high phospholipid and cholesterol plasma levels. They are transferred from the egg yolk to the liver during embryonic development (Noble and Cocchi, Reference Noble and Cocchi1990; Klasing, Reference Klasing and Klasing1998). Egg yolk is rich in oleic acid (C18:1n-9), which represents 33% of the total fatty acids (Noble and Cocchi, Reference Noble and Cocchi1990; Lin and Yang, Reference Lin and Yang2000). This may explain the high oleic acid levels measured in the breast muscle of mule ducks at 1 day of age (41% of total fatty acids). LPL activity was high at 1 and 14 days of age, allowing breast muscle to import blood lipids. Lipids were mainly stored in adipocytes in the breast muscle of 1-day-old ducklings which could be a specificity of birds as in most mammals, where triglycerides are mainly stored in the fibre cytoplasm at birth and adipocytes develop later (Hauser et al., Reference Hauser, Mourot, De Clercq, Genart and Remacle1997; Gondret et al., Reference Gondret, Mourot and Bonneau1998). By contrast, total lipid and triglyceride levels in muscle fibres were lower at hatching than at older ages. Breast muscle development occurred mainly after 14 days of age. The levels of enzyme activity reflecting energy metabolism (CS, HAD, LDH) were therefore low at hatching as previously reported in mammals.
We observed a decrease in total lipid content in the breast muscle of mule ducks induced mainly by a decrease in phospholipids and cholesterol between 1 and 42 days of age. In terms of fatty acid composition, the greatest decrease was in oleic acid. Lin and Yang (Reference Lin and Yang2000) also observed a decrease in oleic acid in the hearts of mule ducks between 1 and 21 days of age. The total lipid content in muscles is unchanged or decreased in turkeys and chickens, respectively, during the first few weeks of post-hatch development (Touraille et al., Reference Touraille, Ricard, Kopp, Valin and Leclercq1981; Hamm and Ang, Reference Hamm and Ang1984; Rabot, Reference Rabot1998). Triglyceride content in fibres increased in breast muscle of mule ducks at day 14 whereas the surface occupied by adipocytes decreased at day 42. Since CS and HAD activity increased, it can be hypothesised that muscle fibres use the lipids stored in adipocytes to supply their energy requirements for growth. The LPL activity of breast muscle decreased, in agreement with the decrease in LPL activity in adipose tissues of Aylesbury ducks (Anas platyrhynchos) reported by Evans (Reference Evans1972). It seems that LPL activity in breast muscle depends on plasma levels of lipids.
We observed an increase in total lipid content in the breast muscle of mule ducks, induced by an increase mainly in triglycerides, but also in phospholipids, between 42 and 98 days of age. At 98 days of age, triglycerides represented 58% of total lipids. Setiawan et al. (Reference Setiawan, Babilé, Auvergne, Belveze and Latil1994) and Baéza et al. (Reference Baéza, Salichon, Marché, Wacrenier, Dominguez and Culioli2000) and (Reference Baéza, Dessay, Wacrenier, Marché and Listrat2002) also reported an increase in total lipid content in the breast muscle of mule ducks between 42 or 56 and 91 days of age and in the breast muscle of Muscovy ducks between 56 and 84 days of age, respectively. In the present study, analysis of fatty acid composition showed that the main increase involved MUFA (particularly oleic acid) and SFA (particularly palmitic acid) provided by lipogenesis in the liver (Klasing, Reference Klasing and Klasing1998) or feed. During this second period, lipids were first stored in muscle fibres: the triglyceride content increased until 75 and 84 days of age in αR and αW fibre types, respectively. Later, lipids were stored in adipocytes which occupied an increasing surface from 75 days of age. This was consistent with a study involving rabbit muscles: in this species, the age-related increase in total muscle lipid and triglyceride contents was closely associated with the increase in numbers and size of clustered adipocytes (Gondret et al., Reference Gondret, Mourot and Bonneau1998). LPL activity increased in the breast muscle of mule ducks between 42 and 75 days of age allowing an increased uptake of circulating lipids, then remained stable. HAD, LDH and CS activity increased until 75, 91 and 98 days of age, respectively. LDH activity was much higher than CS and HAD activity. These findings were in agreement with previous reports in Muscovy ducks by Baéza et al. (Reference Baéza, Salichon, Marché and Juin1998) and confirm that oxidative metabolism is prevalent in breast muscle at the beginning of postnatal development whereas glycolytic metabolism takes over in later development, as reported by Bacou and Vigneron (Reference Bacou and Vigneron1976) in chicks and Briand et al. (Reference Briand, Boissonet, Laplace-Marieze and Briand1993) in rabbits. Plasma levels of glucose, insulin and triglycerides, measured following 12 h fasting, remained almost stable during the whole experimental period. Vasilatos-Younken (Reference Vasilatos-Younken1986) and McMurtry et al. (Reference McMurtry, Plavnik, Rosebrough, Steele and Proudman1988) reported decreases in plasma insulin with age in chickens from 28 to 84 days and from 6 to 54 days, respectively. In these studies, blood samples were drawn 6 or 2 h after a meal, respectively, which might explain the significant age effect on plasma insulin level.
Feed restriction effects between 75 and 84 days of age
Feed restriction corresponded to 79% of ad libitum ingestion between 43 and 75 days of age, but it did not alter body weight or breast muscle weight. It improved feed efficiency (feed conversion ratios were 6.44 and 9.55 for ROF and ALF groups, respectively), probably by reducing carcass fattiness and muscle lipid levels. McMurtry et al. (Reference McMurtry, Plavnik, Rosebrough, Steele and Proudman1988) found no effect of feed restriction on plasma insulin levels in chickens but in this study feed restriction was shorter (6 days) and occurred earlier (from 6 to 12 days of age). In our study, despite the decrease in plasma insulin levels induced by feed restriction, glycaemia, plasma triglyceride, phospholipid and total cholesterol levels and LPL activity in the breast muscle were not affected. The reduced plasma insulin levels probably reduced liver lipogenic activity and the amounts of neo-synthesised lipids exported to peripheral tissues such as muscle and adipose tissues, as previously reported by Tan et al. (Reference Tan, Ohtani and Tanaka1999) and Tan and Ohtani (Reference Tan and Ohtani2000). Since feed restriction was moderate and applied after maximal growth of the breast muscle, the decrease in total lipid content of the breast muscle was therefore moderate ( − 16%) and mainly affected triglyceride ( − 31%) and MUFA ( − 20%) levels. It had no significant influence on the energy metabolism of the breast muscle, the surface occupied by adipocytes, fibre type, CSA and total lipid and triglyceride contents. Restricted feed intake during fattening in rabbits has also been reported to reduce intramuscular lipid deposition without modifying muscle fibre characteristics (Gondret et al., Reference Gondret, Lebas and Bonneau2000).
Overfeeding effects
Kinetic studies during the overfeeding period have only reported findings on lipid deposition in duck livers (Baudonnet-Lenfant, Reference Baudonnet-Lenfant1993; Auvergne et al., Reference Auvergne, Rémignon, Babilé and Baudonnet-Lenfant1995; Robin et al., Reference Robin, Larroudé, Peyhorgue and Castaing2002 and Reference Robin, Peyhorgue and Bénard2004). Studies concerning lipid deposition in muscles have only compared pre-overfed v. overfed ducks slaughtered at different ages (Auvergne, Reference Auvergne1992; Hermier et al., Reference Hermier, Guy, Guillaumin, Davail, André and Hoo-Paris2003) and ad libitum-fed v. overfed ducks slaughtered at the same age (Zanusso et al., Reference Zanusso, Rémignon, Guy, Manse and Babilé2003; Chartrin et al., Reference Chartrin, Bernadet, Guy, Mourot, Duclos and Baéza2005 and Reference Chartrin, Bernadet, Guy, Mourot, Duclos and Baéza2006). In the present study, we compared ad libitum-fed and overfed ducks at three different stages, i.e. the beginning, middle and end of the overfeeding period. The main effects of overfeeding occurred during the 2nd week of this period. Overfeeding resulted in a considerable increase in body weight. Previous studies have shown that it resulted from a dramatic increase in the synthesis of lipids in the liver leading to fatty liver formation and to lipid export to peripheral tissues such as adipose tissues and muscles (Guy et al., Reference Guy, Hermier, Davail, Bely, André and Hoo-Paris1999; Hermier et al., Reference Hermier, Guy, Guillaumin, Davail, André and Hoo-Paris2003). Plasma triglyceride, phospholipid and total cholesterol levels therefore increased. Glycaemia measured 12 h after a meal was not affected by overfeeding, suggesting that the relative hyperinsulinaemia observed in overfed ducks was sufficient for normalisation of glycaemia and that overfed ducks do not become insulin resistant. LPL activity in the breast muscle was increased mainly during the first week and it could be hypothesised that this results from hyperinsulinaemia. The increase was less during the 2nd week. It is generally accepted that overfeeding leading to obesity in mammals increases LPL activity (Granneman and Wade, Reference Granneman and Wade1983; Dugail et al., Reference Dugail, Quignard-Boulange and Dupuy1986; Cleary and Phillips, Reference Cleary and Phillips1996). Davail et al. (Reference Davail, Rideau, Guy, André, Hermier and Hoo-Paris2003) measured plasma LPL activity in mule ducks 90 min after a meal, on the 1st and 10th days of overfeeding. They reported a considerable decrease in plasma LPL activity at the 10th day of overfeeding (1.17 v. 4.12 μmol palmitate per h per ml). This down regulation of LPL activity during the 2nd week of overfeeding needs further investigation.
As previously reported by Zanusso et al. (Reference Zanusso, Rémignon, Guy, Manse and Babilé2003) in Muscovy ducks, and Chartrin et al. (Reference Chartrin, Bernadet, Guy, Mourot, Duclos and Baéza2006) in Muscovy and Pekin ducks and their hybrids, hinny and mule ducks, total lipid levels in the breast muscle increased. This was mainly due to increases in triglyceride and MUFA levels, and to a lesser extent to increases in phospholipid and SFA levels. Overfeeding did not affect the fibre type or CSA of breast muscle fibres. The decrease in the luminance values for red oil and Sudan black staining, indicating increased triglyceride and total lipid levels in muscle fibres and the increase in the surface occupied by adipocytes, were not significant. We did not confirm the results of a previous study in which Chartrin et al. (Reference Chartrin, Bernadet, Guy, Mourot, Duclos and Baéza2005) reported that overfeeding significantly increased the triglyceride content of αR fibres and the surface occupied by adipocytes in breast muscle compared with ad libitum-fed ducks. In the latter study, the number of repetitions per treatment was 32 and the increase in total lipid content of breast muscle from mule ducks was also greater than in the present experiment (1.7-fold v. 1.4-fold). Nevertheless, the glycolytic metabolism of breast muscle in the present study was affected by overfeeding, since LDH activity was significantly decreased at 98 days of age. Zanusso et al. (Reference Zanusso, Rémignon, Guy, Manse and Babilé2003) found no significant effect of overfeeding on LDH activity in Muscovy ducks, but there was an increase in CS and HAD activity. It seems more relevant to consider the CS/LDH or HAD/LDH ratios. When they increase, the use of fatty acids for energy metabolism is enhanced.
Breast muscle weight decreased during the 1st week of overfeeding. It can be hypothesised that muscle proteolysis increased to provide amino acids to the liver for protein synthesis (hyperplasia and hypertrophy of hepatocytes, lipopoprotein synthesis). At the end of the overfeeding period, the muscle had recovered its initial weight.
Conclusion
The original contribution of this study was to describe the kinetics of lipid deposition in relation to its localisation (adipocytes and muscle fibres) in the breast muscle of ducks during growth post-hatch. This has not been reported for any domesticated bird to date. Two periods can be defined for age effects on intramuscular lipids in breast muscle: − 1 to 42 days of age when lipids stored in the adipocytes during embryonic life are used by the muscle fibres for growth and energy requirements, and − 42 to 98 days of age when the muscle again stores lipids, first in fibres then in adipocytes.
Feed restriction, corresponding to 79% of ad libitum ingestion, applied between 42 and 75 days of age only resulted in a slight decrease in total lipid content of the breast muscle, mainly involving triglyceride and MUFA levels. Overfeeding considerably increased total lipid levels in the breast muscle, mainly due to triglyceride and MUFA deposition occurring particularly during the 2nd week of this period. As found by Gondret et al. (Reference Gondret, Lebas and Bonneau2000) in rabbits, we observed that nutritional status regulates intramuscular lipid deposition in ducks without changing fibre-type composition.
During the first weeks of post-hatch development, the increase in HAD, CS and LDH activity depends on muscle development. After 91 days of age, LDH activity can be decreased by age or overfeeding leading to an increase in muscle lipid levels.
The LPL activity seems to be regulated by the plasma lipid levels. It is high when lipids circulate at high concentrations in plasma leading to increases in muscle lipid levels as observed during the first 2 weeks of post-hatch development and during overfeeding.
Acknowledgements
We thank the technical staff of INRA Artiguères for rearing and overfeeding ducks, Thierry Bordeau for his expert technical assistance in animal dissection and sample preparation and D. Chadeyron and N. Guivier of INRA Theix for their expertise in measurement of enzyme activities. We are grateful to Daniel Tanguy for improving the use of Racine software by modifying the Unix environment to the Windows environment.