Compensatory growth response in pigs: Effects on growth performance, composition of weight gain at carcass and muscle levels, and meat quality 1

Transcription

1 Compensatory growth response in pigs: Effects on growth performance, composition of weight gain at carcass and muscle levels, and meat quality 1 A. Heyer and B. Lebret 2 INRA, UMR 1079, Systèmes d Elevage et Nutrition Animale et Humaine, F Saint-Gilles, France ABSTRACT: A restriction/realimentation feeding strategy was applied to pigs to increase the age at market weight and final ADG, modify protein and lipid deposition rates at carcass and muscle levels, and thereby improve eating quality of the pork. A total of 126 Duroc (Large White Landrace) pigs (females and castrated males) were used. At the average BW of 30 kg, within litter and sex, pairs of littermates (blocked by BW) were randomly assigned to ad libitum (AL) feeding during growing (30 to 70 kg of BW) and finishing (70 to 110 kg of BW) periods (AL, n = 56), or restricted feeding at 65% of the ADFI of the AL pigs, on a BW basis, during the growing period and AL feeding during finishing (compensatory growth, CG; n = 56). In each feeding regimen, 15 pigs were slaughtered at 70 kg of BW, and 41 pigs were slaughtered at 110 kg of BW. Additionally, 14 pigs were slaughtered at 30 kg of BW to calculate tissue deposition rates. The CG pigs showed decreased ADG ( 35%, P = 0.001) during growing but increased ADG (+13%, P = 0.001) during finishing (i.e., compensatory growth) due to greater (P = 0.001) ADFI and G:F. Hence, CG pigs were 19 d older at 110 kg of BW than AL pigs. The CG pigs were leaner at 70 kg of BW than AL (e.g., 11.7 vs mm of average backfat thickness for CG and AL pigs, respectively, P = 0.023), whereas the differences were reduced at 110 kg of BW (20.6 vs mm of average backfat thickness for CG and AL pigs, respectively, P = 0.536). At 70 kg of BW, intramuscular fat (IMF) content of LM did not differ between CG and AL pigs (1.25 vs. 1.49%, respectively, P = 0.118), whereas CG pigs had less IMF in LM at 110 kg of BW (2.19 vs. 2.53% for CG and AL pigs, respectively, P = 0.034). Feeding regimen influenced the composition of weight gain. From 30 to 70 kg of BW, feed restriction reduced (P = 0.001) lean and adipose tissue deposition at the carcass level and protein and lipid deposition at the muscle level. From 70 to 110 kg of BW, the CG feeding strategy increased (P = 0.016) deposition of adipose but not of lean tissue at the carcass level. However, lipid and protein deposition at the muscle level were not affected. Thus, realimentation promoted deposition of subcutaneous fat over IMF. Feeding regimen hardly affected technological meat quality at 110 kg of BW. The CG feeding strategy decreased (P = 0.014) the meat juiciness score in relation to the decreased IMF but did not influence other sensory traits. Elevated IMF content and improved pork quality might be achieved by modifying the onset or duration of the restriction and realimentation periods. Key words: pig, compensatory growth, carcass, tissue deposition rate, intramuscular fat content, meat quality 2007 American Society of Animal Science. All rights reserved. J. Anim. Sci : doi: /jas INTRODUCTION Compensatory growth in growing/finishing pigs is a well known physiological phenomenon of accelerated final growth rate induced by a restricted food supply during the growing period followed by ad libitum (AL) 1 This project was financially supported by the French Région Bretagne, the Department PHASE - INRA and the program INRA- INAO. The authors acknowledge J. Noblet (INRA-SENAH) for calculations of diet composition. The skillful assistance of R. Delaunay, B. Duteil, M. Alix, J. Liger, J. F. Rouaud and N. Bonhomme at the experimental farm, slaughterhouse and laboratory (INRA-SENAH), and of K. Méteau at the sensory laboratory (INRA Le Magneraud, F Surgères, France) is gratefully acknowledged. feeding during finishing (McMeekan, 1940). Later studies revealed that compensatory growth influences the deposition rates of lean and adipose tissues and might thereby modify carcass composition at market weight (Campbell et al., 1983; Donker et al., 1986; Hornick et al., 2000). However, composition of weight gain at the muscle level during refeeding was not well described in pigs. Intramuscular fat (IMF) level is often reported to have beneficial effects on pork eating quality (Bejer- 2 Corresponding author: Benedicte.Lebret@rennes.inra.fr Received March 21, Accepted September 20,

2 770 Heyer and Lebret holm and Barton-Gade, 1986; Fernandez et al., 1999; Wood et al., 2004), although some authors showed only weak (Eikelenboom et al., 1996) or even no influence (Göransson et al., 1992). Hence, an increased IMF through a strategic feeding regimen would be a desirable tool in pork production. Because more than 80% of muscle lipids are stored in i.m. adipocytes (Essén- Gustavsson et al., 1994), raising their number could increase IMF. Gondret and Lebret (2002) showed an increase in the number of i.m. adipocytes with age independently of IMF, whereas the size of i.m. adipocytes and therefore IMF increases with energy intake. Thus, IMF might increase simultaneously with number and size of i.m. adipocytes because of greater final age and ADFI. Moreover, greater slaughter age is associated with improved pork quality for French consumers (MAP, 2005). Besides, it was demonstrated that muscle protein turnover at slaughter increased with final growth rate, which might improve meat tenderness (Therkildsen et al., 2002, 2004). Our study aimed to increase age at slaughter of pigs by an initial strong feed restriction until 70 kg of BW, followed by AL feeding up to 110 kg of BW to 1) increase age at slaughter of pigs and the potential for IMF deposition; 2) evaluate the composition of BW gain during restriction and realimentation; and 3) increase final ADFI and likely increase IMF, and thereby improve pork eating quality. Animals MATERIALS AND METHODS The experiment was conducted following French guidelines for animal care and use ( infolabos/reglementation/expearim.htm). All people involved in the experiment have an agreement for conducting experimental procedures on animals, delivered by the veterinary services. The experiment was conducted in the experimental farm of INRA [UMR SENAH (F Saint-Gilles, France)], on a total of 126 indoor-reared pigs [crossbred Duroc (Large White Landrace)] from 16 litters originating from 3 sires. The piglets were weaned after 28 d and received a conventional postweaning diet thereafter. At the average BW of 30 kg (SD = 3.7) and average age of 75 d (SD = 0.7), castrated males and females in each litter were chosen on the basis of their BW at birth and growth rate up to 30 kg of BW. Within litter and sex, pairs of littermates (blocked by BW) were randomly assigned to one of the following feeding regimens: the control group (n = 56) was fed AL with a growing diet (Table 1) during the growing period (from 30 to 70 kg of BW), and with a finishing diet during the finishing period (from 70 to 110 kg of BW); and the group for which a compensatory growth (CG; n = 56) was expected during the finishing period, was restricted-fed with the growing diet at 65% of the Table 1. Composition (ingredients and chemical composition) and nutrient content of the experimental diets, asfed basis Growing Finishing Ingredient, % diet diet Wheat Corn Barley Soybean meal Rapeseed oil Wheat bran Molasses L-Lysine HCl DL-Methionine L-Threonine L-Thryptophane 0.01 Bicalcium phosphate Calcium carbonate Marine salt Trace minerals and vitamin mixture Chemical composition, % DM CP Crude fat Crude fiber Starch Ash Lysine Nutritional value, 1 MJ of NE/kg Estimated from the INRA-AFZ (2002) tables. AL feed intake of the AL group (on a BW basis) during the growing period (from 30 to 70 kg of BW), and fed AL with the finishing diet from 70 to 110 kg of BW. Pigs were reared in individual pens (2.25 m 2 /pig, concrete floor without straw), in 2 compartments of the same building. Pairs of littermates of both sexes were assigned within treatments to 1 of the 2 compartments on the basis of their BW [average BW of 27.8 kg (SD = 2.3) for the lighter pigs and of 32.5 kg (SD = 3.0) for the heavier pigs, respectively] for a better calculation of the restricted diet of the CG pigs in each compartment. Feed intake and weight was registered weekly for each pig on experiment. Pigs were weighed without fasting, except at the beginning of the experiment and at the end of the growing (70 kg) and the finishing (110 kg) periods. One CG female pig was discarded from the experiment due to severe growing retardation. The feeding schemes of the CG pigs were calculated weekly within compartment and sex, on the basis of BW and ADFI of the AL pigs. During the restriction period, feed of the CG pigs was distributed every morning in 1 meal. All pigs had free access to water at all times. Apart from the 112 pigs distributed in the AL and CG feeding regimen, 14 piglets (7 castrated males and 7 females), originating from 12 litters and representative of their experimental littermates according to their BW at birth and their growth rate up to 30 kg of BW, were chosen at the average age of 75 d [average BW of 29.9 kg (SD = 2.5)] for immediate slaughter as controls

3 Compensatory growth response in pigs 771 at beginning of the experiment to enable further calculations of the composition of the weight gain. Pigs were slaughtered at BW (live weight) of 30 kg (LW30, n = 14), 70 kg (LW70, n = 30, equally distributed within feeding regimen, sexes, and compartments), or 110 kg (LW110, n = 82). The first aim of this study was to evaluate the responses at 110 kg of BW to the restriction-realimentation feeding strategy compared with AL feeding for various traits, including meat eating quality, thus explaining the lower number of pigs harvested at LW70 compared with LW110. Harvesting pigs at 30 and 70 kg of BW aimed to allow subsequent calculations of tissue deposition rates. The LW30 and LW70 pigs were representative of the animals due to their pedigree, BW at beginning of the growing period, and for LW70, ADG during the growing period. Pigs were fasted overnight, transported individually, and kept in individual pens for 2 h in lairage, where they had free access to water. They were showered with a small water jet for 1 min, at 5 min before being taken to the stunning location, and then were slaughtered by electrical stunning and exsanguination, in compliance with the current national regulations applied in slaughterhouses. Carcass Traits Weights of hot carcass, liver, and entire, depleted gut were recorded at slaughter. Mean backfat thickness [mean of measurements taken between the third and fourth lumbar vertebra (8 cm from the midline) and third and fourth from the last rib (6 cm from the midline) levels] and muscle depth (between third and fourth from the last rib, 6 cm from the midline) were measured using a Fat-O-Meter (SFK, Herlev, Denmark). Lean meat content (LMC FoM ) was estimated from carcass linear measurements for LW110 pigs, as described by Daumas et al. (1998). After 24 h at 4 C, the weights of the cold carcasses (the entire pig, without the viscera and head) and of wholesale cuts of the left side (ham, shoulder, belly, loin, and backfat) were recorded. For LW70 and LW110 pigs, lean meat content was calculated from detailed cutting values using the equation LMCcalc = %(ham) %(loin) 1.059%(backfat); Métayer and Daumas, On all LW30, LW70, and on 31 LW110 pigs (equally distributed between sex and compartments, and representative in terms of growth performance), the left ham was partially dissected into muscles (including intermuscular fat and intermuscular connective tissues), subcutaneous fat without skin, and bones. This partial dissection of the ham was used as an indicator of carcass composition (Desmoulin et al., 1988). Weights of each tissue and of dissected biceps femoris (BF) muscle were recorded. On the same pigs, the whole LM was dissected from the left loin and weighed. Muscle Composition and Meat Quality Traits Within 20 min postmortem, approximately 60-g samples of LM (at the third lumbar vertebra level) and BF muscles were excised on all LW110 pigs, immediately frozen in liquid nitrogen, and stored at 80 C until determinations of the initial ph (ph 1 ) and the glycolytic potential (GP) could be performed. The ph 1 was determined after homogenization of 2 g of muscle in 18 ml of 5 mm sodium iodoacetate, ph 7.0 (Ingold Xerolyte electrode, Knick ph-meter, Berlin, Germany). The GP was determined according to the method of Monin and Sellier (1985), as GP = 2([glycogen] + [glucose] + [glucose-6-phosphate]) + [lactate], in which the brackets indicate the concentration in micromoles per gram. After homogenization of 1 g of muscle in 10 ml of perchloric acid (0.55 M), glucose and glucose-6-phosphate were determined together using an enzymatic method (glucose HK, ABX Diagnostics kit, Montpellier, France). Lactate was also determined using an enzymatic method (Biomerieux kits, Marcy l Etoile, France). These analyses were performed on an automatic spectrophotometric analyzer (Cobas Mira Roche, Basel, Switzerland). Muscle glycogen content was determined from the glucose determination (see above) after hydrolysis by amiloglucosidase, as described by Talmant et al. (1989). Lactate, free glucose and glucose- 6-phosphate, and glucose issued from glycogen hydrolysis were expressed as micromoles per gram of wet tissue. The GP was expressed as micromole equivalent of lactate per gram of wet tissue. The following day, for all pigs in the experiment, 1 slice of LM (second lumbar vertebra level, approximately 2-cm thick) and BF (approximately 2-cm thick) were collected, trimmed of external fat, minced, and homogenized. A subsample was used for analyze DM of fresh LM and BF, which was calculated as the weight difference before and after drying at 105 C for 16 to 18 h. The remaining samples were freeze-dried, pulverized, and kept at 20 C under vacuum before chemical analyses. The IMF content was determined from the diethyl ether and petroleum ether extracts (Soxtec Avanti 2050, Foss, Höganäs, Sweden). Crude protein content was determined from the N concentration (Dumas method; AOAC, 1990) using a multiplication factor of The same day, on LW110 pigs, transverse sections of LM (last rib level) and BF muscles were collected on all pigs for direct determination of ultimate ph using the same apparatus as described above. After 1 h of blooming at 4 C under artificial light, the color was evaluated on these muscle samples through the determination of CIE L* (lightness), a* (redness), and b* (yellowness) coordinates (average of triplicate determinations, at 3 locations per sample) using a chromameter Minolta CR 300 (Osaka, Japan), with a D65 illuminant and a 1-cm-diam. aperture. The same day, another slice of approximately 100 g of LM (cranial to the last rib cut) was collected, trimmed of external fat and perimysium,

4 772 Heyer and Lebret weighed, and kept at 4 C in a plastic bag for determination of drip loss after 4 d postmortem (Honikel, 1998). Sensory Analyses The day after slaughter, on all LW110 pigs, a sample of the right loin (dorsal vertebra region) of each carcass was trimmed of external fat, kept at 4 C for 3 d, deboned but with muscles and adipose tissue adjacent to the LM kept in place, put under vacuum, frozen, and stored at 20 C before sensory analyses at the INRA-Le Magneraud (Surgères, France). Additionally, on 4 AL and 4 CG pigs, the cranial portion of the remaining loin was aged, prepared, and stored following the same method, for training sessions of the sensory panelists. Roasts (600 g) were thawed at 4 C during 48 h, trimmed of external fat, and cooked in an oven by dry heat for 10 min at 250 C and then by humid heat at 100 C up to a core temperature of 80 ± 2 C; i.e., a total cooking time of approximately 55 min. Then, the middle portion of 1-cm thick slices of cooked meat was cut into rectangular pieces and presented to the 12 trained panelists. After 2 training sessions with the training roasts, a descriptive test was performed by a trained sensory panel of 12 members. Because of the high variability in IMF content of the LM within the LW110 pigs (0.75 to 5.25%), roasts for sensory analyses were ranked according to IMF content of LM within feeding regimen and sex to avoid the confounding effects of feeding regimen and IMF content in sensory analyses. Four roasts, one from each feeding regimen and sex, and presenting a similar IMF level, were then assessed at each of the 20 sessions and scored by the panelists for tenderness, juiciness, fibrousness (perception of muscle fibers during mastication), flouriness (flour sensation after mastication), flavor, and pork flavor (specific flavor of pork) on a continuous scale from 0 (absent of the character) to 10 (very high intensity of the character). The average of individual panelist scores from each sample was used for the statistical analysis. Statistical Analyses Statistical analyses were performed with the GLM procedure (SAS Inst. Inc., Cary, NC). The model included the fixed effects of feeding regimen (AL or CG), sex (castrated male or female), initial BW (light or heavy), and their interactions (first level), and the random effect of sow within sire. The fixed effect of sensory session was added in the model for the analysis of sensory data. RESULTS AND DISCUSSION Performance and Carcass Quality Feed restriction at 65% of the AL level for CG pigs from 30 to 70 kg of BW resulted in a decrease (P = 0.001) by 35% in ADG (Table 2), whereas effect on G:F did not reach significance. Consequently, these pigs needed 22 more days to reach the 70 kg of BW mark i.e., average age at start of realimentation of CG pigs was 140 d compared with 118 d for the AL pigs (P = 0.001). Realimentation increased (P = 0.001) ADFI with 255% during finishing compared with growing phase for the CG pigs. The ADFI of CG pigs during finishing corresponded to 108% of those of AL pigs in the same period (P = 0.001), thus demonstrating a compensatory feed intake for the CG pigs during realimentation. However, total feed consumption from 30 to 110 kg of BW was not different between the feeding regimen (229.7 vs kg, P = 0.794). The CG pigs had additionally a greater (P = 0.001) G:F during realimentation. This led to a 13% greater (P = 0.001) ADG for CG pigs during finishing compared with AL pigs. Hence, CG pigs expressed a compensatory growth. Because of the realized compensatory growth in connection with a 19 d greater age at slaughter at 110 kg for the CG pigs, the objective of this study to produce older pigs with same market weight through a strategic feeding was reached. During growing and finishing periods, castrated males had greater (P = 0.001) ADFI than females (1.99 vs kg/d during growing, and 3.92 vs kg/d during finishing for castrated males and females, respectively) and greater (P 0.035) ADG (833 vs. 780 g/ d for the growing period and 1,211 vs. 1,160 g/d during the finishing period). The G:F did not differ between sexes during the growing period, whereas castrated males had greater (P = 0.001) G:F than females during finishing (0.32 vs. 0.30). Altogether, this led to lower (P = 0.001) age for castrated males than females at both LW70 (128 vs. 131 d) and LW110 (166 vs. 162 d). It should be noticed that castrated males and females had similar responses to feeding regimen in terms of ADFI and ADG during growing and finishing periods, and G:F during the growing period. An interaction (P = 0.004) between feeding regimen and sex was found for G:F during the realimentation period, with a significant effect of the feeding regimen in females (0.34 vs for CG and AL pigs, respectively) but not in castrated males (0.31 vs for CG and AL pigs, respectively). Initial BW influenced ADFI during growing, with lower (P = 0.002) ADFI for the light compared with the heavy pigs (1.89 vs kg/d), whereas there were no differences in ADFI during realimentation. Light pigs at start of the experiment had greater (P = 0.001) G:F during growing (0.43 vs kg/kg) but lower (P = 0.004) G:F during finishing (0.31 vs kg/kg) than heavy pigs. Initial BW did not affect ADG during growing and finishing periods. However, due to their differences in BW at start of the experiment, light pigs were older (P = 0.001) than heavy pigs at BW 70 (+4 d) and LW110 (+ 8 d). Light pigs were 1.5 kg heavier (P = 0.005) at LW110 than heavy pigs. Initial BW did not influence the effects of feeding strategy on growth performance traits.

5 Compensatory growth response in pigs 773 Table 2. Growth performance of pigs between 30 and 70 kg of BW and between 70 and 110 kg of BW and allocated to different feeding regimens 1 30 to 70 kg of BW P-value 2 70 to 110 kg of BW P-value 2 Item AL CG RSD F S W AL CG RSD F S W No. of animals Initial live weight, kg Initial age, d ADFI, kg/d ADG, g/d G:F, 3 kg/kg Weight at LW70/LW110, 4 kg Age at LW70/LW110, d AL = Ad libitum during growing and finishing periods; CG = restricted (65%) during the growing period (30 to 70 kg of BW) and AL during the finishing period (70 to 110 kg of BW). Values are least squares means; RSD = residual SD. 2 P-values for feeding regimen (F), sex (S), and initial BW (W). 3 Interaction between feeding regimen and sex was found for the 70 to 110 kg of BW period (P < 0.05). 4 LW70 = Live weight of 70 kg; LW110 = live weight of 110 kg. The occurrence of greater ADFI after a feed restriction period was discussed, with some controversy, in the literature. Our study corroborates findings of greater ADFI from Donker et al. (1986), Mersmann et al. (1987), and Therkildsen et al. (2004), whereas Prince et al. (1983), Valaja et al. (1992), and Oksbjerg et al. (2002) reported an unchanged level of ADFI during realimentation. Raj et al. (2004) suggested that the onset of the realimentation period plays an important role for the increase in ADFI. When realimentation occurred at 50 kg of BW, the authors reported greater ADFI from 50 to 80 kg of BW but not afterwards. In contrast, starting the realimentation phase at 80 kg of BW had no significant effect on ADFI compared with the AL-fed control. The beneficial effect of compensatory feed intake or greater G:F or both during realimentation on growth rate was earlier reported in several studies (e.g., Donker et al., 1986; Kristensen et al., 2002; Therkildsen et al., 2004). However, in another study under similar conditions as the present one, compensatory feed intake during realimentation from 80 to 110 kg of BW did not lead to a compensatory growth response of pigs because of a lower G:F (Lebret et al., 2004). Indeed, Kristensen et al. (2002) and Therkildsen et al. (2002) found that a realimentation period of 42 to 75 d was necessary to obtain full advantages of compensatory growth on performance. Thus, a strategic feeding might influence ADG depending on the onset, intensity, and duration of restriction and the onset and duration of realimentation (Campbell et al., 1983; Therkildsen et al., 2002; Mason et al., 2005). It has to be mentioned that in the current study, onset of realimentation and slaughtering occurred when pigs reached 70 or 110 kg of BW mark and not after predetermined time-scheduled growing and finishing periods, as practiced in most of the studies. Hence, a compensatory growth index could not be calculated. The study demonstrated that retarded growth performance during the restriction period increased (P = 0.004) dressing percent and decreased (P = 0.009) weights of depleted gut and liver, without modifying carcass weight. The feed restriction tended (P = 0.068) to increase calculated lean meat content (+1.7 percentage point) due to reduced (P 0.023) carcass fatness (backfat thickness and percentage) and a tendency (P = 0.060) for greater ham proportion in the carcass for CG pigs at LW70 (Table 3). Similar results were found at the ham level, with lower (P = 0.047) subcutaneous fat for CG than AL pigs, whereas difference between CG and AL pigs for total muscle or BF percentage did not reach significance. These effects of feed restriction on carcass fatness were similar to those reported in several earlier studies (Campbell et al., 1983; Donker et al., 1986; Therkildsen et al., 2002). After realimentation at slaughter at 110 kg of BW, the CG pigs had lower (P = 0.001) carcass dressing and greater (P = 0.001) weights of depleted gut and liver. The LMC FoM and muscle depth were similar between CG and AL pigs, and backfat thickness was no longer different between the feeding regimens. However, CG pigs had increased LMC calc (+ 1.3 percentage points, P = 0.024), in connection with slightly greater (P = 0.034) proportion of ham and lower (P = 0.001) proportion of backfat in the carcass. Differences in ham composition between CG and AL pigs did not reach significance. Except for greater (P = 0.014) weight of liver for castrated males than females (1.39 vs kg), there were no differences between sexes for carcass traits at LW70. At LW110, castrated males had lower (P = 0.001) LMC FoM and LMC calc in connection with greater (P = 0.001) backfat thickness and percentage (e.g., 57.7 vs. 60.0% LMC FoM, and 22.3 vs mm backfat thickness for castrated males and females, respectively). Proportions of shoulder were increased (P = 0.010) and proportions of ham and loin were decreased (P 0.014) in castrated males compared with females (data not shown). Similar effects were found at the ham level with lower (P = 0.047) muscle percentage for castrated males. Interactions (P 0.033) between feeding regimen

6 774 Heyer and Lebret Table 3. Carcass quality traits of pigs slaughtered at 70 or 110 kg of BW and allocated to different feeding regimens 1 30 to 70 kg of BW P-value 2 70 to 110 kg of BW P-value 2 AL CG RSD F S W AL CG RSD F S W No. of animals Carcass traits HCW, kg Dressing, 3 % Depleted gut, kg Liver, 3 kg Lean meat content FoM, % Lean meat content calc, % Muscle depth last rib, mm Backfat thickness average, mm Carcass composition, % Ham Shoulder Belly Loin Backfat No. of animals Partial ham dissection, % Muscle BF Subcutaneous fat Bones AL = Ad libitum during growing and finishing periods; CG = restricted (65%) during the growing period (30 to 70 kg of BW) and AL during the finishing period (70 to 110 kg of BW). Values are least squares means; RSD = residual SD. 2 P-values for feeding regimen (F), sex (S), and initial BW (W). 3 Interaction between feeding regimen and sex was found for the 30 to 70 kg of BW period (P < 0.05). and sex were found for dressing percent and weight of liver at LW70, with significant effects of the feeding regimen in females (79.9 vs. 77.8% for dressing and 1.21 vs kg for liver weight, for CG and AL pigs, respectively), but not in castrated males (78.6 vs. 78.1% for dressing and 1.39 vs kg for liver weight, for CG and AL pigs, respectively). At LW110, effects of feeding regimen on carcass traits did not differ between castrated males and females. Initial BW did not influence carcass traits at LW70 and LW110 and did not modify the effects of feeding strategy on carcass traits. Increased weight of gut in CG pigs may have been explained by an amplified development of the gastrointestinal tract due to greater ADFI during finishing. The greater liver weight in CG pigs is in accordance with Mersmann et al. (1987) and Bikker et al. (1996b), who stated that compensatory growth occurs principally in internal organs. Higher weights of internal organs explain the lower dressing percent of CG compared with AL pigs, but similar carcass weights. The similar carcass weights eased further comparison of carcass and quality traits between the 2 feeding regimens. Results on the effects of feeding regimen on carcass composition indicate that compensatory growth modifies the composition of weight gain with greater lipid than protein deposition at the carcass level, in agreement with Skiba et al. (2004). In accordance with the current study, Therkildsen et al. (2004) found no differences in LMC FoM of pigs showing compensatory growth (50 or 59 d of realimentation), whereas in an earlier study refed pigs (27 d of realimentation) had greater LMC FoM compared with AL-fed controls (Therkildsen et al., 2002). These results strengthen the finding that duration of realimentation period affects carcass composition as previously demonstrated by other studies (Campbell et al., 1983; Lebret et al., 2004; Mason et al., 2005). Chemical Composition of Muscle The chemical composition of LM and BF muscles differed between feeding regimens at LW70 (Table 4). The DM (LM, BF) contents were similar between feeding regimens, whereas protein contents were lower (P 0.049) in LM and BF, and IMF content was lower (P = 0.005) in the BF for the CG pigs. Effect of feeding regimen on IMF content in the LM did not reach significance. These results showed a decrease in fat deposition up to 20% in muscle due to feed restriction, in accordance with Ellis et al. (1996) and Wood et al. (1996). Sex and initial weight had no effect of muscle composition at 70 kg of BW. Interactions (P 0.029) were found between feeding regimen and initial BW for protein content in LM and BF, with a significant effect of feed restriction in pigs with low initial BW (21.9 vs. 22.5% protein in the LM and 21.0 vs. 22.2% protein in the BF for CG and AL pigs, respectively) but not in pigs with high initial BW (22.4 vs. 22.3% protein in the LM and 21.5 vs. 21.8% protein in the BF for CG and AL pigs, respectively). At 110 kg of BW, in LM and BF, DM and IMF contents were significantly lower for CG compared with AL pigs,

7 Compensatory growth response in pigs 775 Table 4. Chemical composition of LM and M. biceps femoris of pigs slaughtered at 70 or 110 kg of BW and allocated to different feeding regimen 1 30 to 70 kg of BW P-value 2 70 to 110 kg of BW P-value 2 Item AL CG RSD F S W AL CG RSD F S W No. of animals LM DM, % CP, 3 % IMF, % M. biceps femoris DM, % CP, 3 % IMF, % AL = Ad libitum during growing and finishing periods; CG = restricted (65%) during the growing period (30 to 70 kg of BW) and AL during the finishing period (70 to 110 kg of BW). Values are least squares means; RSD = residual SD. 2 P-values for feeding regimen (F), sex (S), and initial BW (W). 3 Interaction between feeding regimen and initial BW was found for the 30 to 70 kg of BW period (P < 0.05). but on the average level of DM and IMF increased during the finishing period. Thus, the objective to increase IMF through a greater age at slaughter and a greater final feed intake was not reached with the applied feeding regimen. Similarly, Therkildsen et al. (2002) and Mason et al. (2005) did not find an increase in IMF content using a restriction-realimentation feeding strategy. Protein contents were no longer different between CG and AL pigs in LM and BF muscles, in agreement with Kristensen et al. (2002) and Therkildsen et al. (2002). Castrated males had greater (P 0.009) IMF contents than females in LM and BF muscles (2.64 vs. 2.09% in the LM and 2.59 vs. 2.26% in the BF for castrated males and females, respectively), and greater (P = 0.003) DM content in the LM (26.9 vs. 26.4% for castrated males and females, respectively). Composition of LM and BF muscles did not depend on initial BW of pigs. Due to the weight measurements and chemical analysis of ham, LM and BF on pigs slaughtered at 30, 70, and 110 kg of BW, muscle, protein, and fat deposition rates of AL and CG pigs could be calculated (Table 5). Deposition rate calculations were made on pigs from CG and AL feeding regimens representative in terms of growth rate and carcass traits (n = 13 pigs per feeding regimen and per slaughter weight). Compared with AL feeding, restriction from 30 to 70 kg of BW decreased (P = 0.001) muscle tissue deposition and protein deposition in the LM and BF by around 30%. Similarly, lipid deposition in ham and in muscles was decreased (P 0.003) by 40 to 55%. Deposition rates were similar between castrated males and fe- Table 5. Muscle, protein and fat deposition in ham, LM and M. biceps femoris of pigs slaughtered at 70 or 110 kg of BW and allocated to different feeding regimens 1 30 to 70 kg of BW P-value 2 70 to 110 kg of BW P-value 2 AL CG RSD F S W AL CG RSD F S W No. of animals Deposition rates in ham Muscle, g/d Subcutaneous fat, g/d Bone, g/d Weight and deposition rates in LM Weight, kg Muscle, g/d Protein, g/d IMF, g/d Weight and deposition rates in M. biceps femoris Weight, kg Muscle, g/d Protein, g/d IMF, g/d AL = Ad libitum during growing and finishing periods; CG = restricted (65%) during the growing period (30 to 70 kg of BW) and AL during the finishing period (70 to 110 kg of BW). Values are least squares means. RSD = residual SD; IMF = intramuscular fat. 2 P-values for feeding regimen (F), sex (S), and initial BW (W).

8 776 Heyer and Lebret Table 6. Technological meat quality traits of LM and M. biceps femoris of pigs slaughtered at 110 kg of BW and allocated to different feeding regimens 1 LW110 P-value 3 Item 2 AL CG RSD F S W No. of animals LM ph ph u Drip loss 4days, % Glycolytic potential, mol/g Glucose (glyc.), 4 mol/g Glucose+G6P, mol/g Lactate, mol/g Minolta value L* a* b* M. biceps femoris ph ph u Glycolytic potential, mol/g Glucose (glyc.), 4 mol/g Glucose+G6P, mol/g Lactate, mol/g Minolta value L* a* b* AL = Ad libitum during growing and finishing periods; CG = restricted (65%) during the growing period and AL during the finishing period. Least squares means are indicated. 2 ph 1 = initial ph (20 min postmortem); ph u = ultimate ph. L* (lightness), a* (redness), and b* (yellowness). 3 P-values for feeding regimen (F), sex (S), and initial live weight (W); RSD = residual SD. 4 Glucose issued from glycogen hydrolysis. males. Deposition rates of ham muscles, LM, and BF, and protein in LM and BF were lower (P 0.039) in pigs having a low compared with a high initial BW. During realimentation, deposition rate of ham subcutaneous fat was increased for the CG pigs (+26%, P = 0.016), whereas IMF muscle and protein deposition rates were not significantly affected. Weight, muscle, and protein deposition rates in LM were decreased (P 0.049) in castrated males compared with females (2.4 vs. 2.6 kg for muscle weight, 20.1 vs g/d for muscle deposition rate, and 4.8 vs. 6.1 g/d for LM protein deposition rate, for castrated males and females, respectively). Low initial BW led to lower (P = 0.031) subcutaneous fat deposition and tended (P 0.084) to decrease LM weight and deposition rate, compared with high initial BW. Calculations of tissue deposition rates indicated that feed restriction influenced more the adipose than muscular tissue deposition at the carcass level, and the protein than IMF deposition at the muscle level, compared with AL feeding. In contrast, realimentation modified the composition of weight gain at the carcass level with greater deposition rate of adipose than muscular tissue, but realimentation did not modify the composition of weight gain at the muscle level. The improved deposition of subcutaneous fat over IMF during realimentation was an interesting finding. To our knowledge, effects of compensatory growth in pigs on composition of weight gain at the muscle level, i.e., protein and IMF deposition rates were not studied in previous experiments. Higher ADG of CG pigs during realimentation would be due to greater growth rate of adipose tissue and internal organs, but not of lean tissue growth rate, in alliance with Bikker et al. (1996a) and Skiba et al. Table 7. Sensory meat quality traits of cooked LM of pigs slaughtered at 110 kg of BW and allocated to different feeding regimens 1 LW110 P-value 2 Item AL CG RSD F S W No. of animals Tenderness Juiciness Fibrousness Flouriness Flavor Pork flavor AL = Ad libitum during growing and finishing periods; CG = restricted (65%) during the growing period (30 to 70 kg of BW) and AL during the finishing period (70 to 110 kg of BW). Values are least squares means; RSD = residual SD. 2 P-values for feeding regimen (F), sex (S), and initial BW (W).

9 Compensatory growth response in pigs 777 (2004). Altogether, this suggests that compensatory growth occurs mostly in the tissues mainly discriminated by feed restriction, i.e., internal organs and backfat, but not IMF. Our objective to increase IMF with the CG feeding regimen, based on the hypothesis of greater number and lipid deposition rate in intramuscular adipocytes by increasing age at slaughter and final ADFI (Gondret and Lebret, 2002), was not reached. An elevated IMF content in refed pigs might be achieved through forwarding the onset or modifying the duration of restriction and of realimentation periods. Meat Quality In the current study, initial and ultimate ph in LM and BF, and drip loss after 4 d postmortem did not differ between CG and AL pigs (Table 6). The GP was lower (P 0.023) for the CG compared with AL pigs in LM and BF, due to lower (P 0.074) glycogen content in both muscles, whereas free glucose, glucose-6-phosphate, and lactate levels remained uninfluenced by the feeding regimen. Despite the lower GP of the CG pigs, the ultimate ph remained unaffected. Color measurements showed a lower (P = 0.001) b* value in the loin of the CG pigs but no effect in the BF. Effects of sex were found, with lower (P 0.036) drip and lightness score (L*) in the LM of castrated males than females, in agreement with their lower (P 0.036) levels of lactate (Schäfer et al., 2002). Pigs with low initial BW had lower (P 0.030) GP in LM and BF muscles, due to less glycogen content in particular in the LM (P = 0.003). On the whole, feeding regimen hardly affected technological meat quality, in accordance with other studies (Kristensen et al., 2002; Therkildsen et al., 2002; Mason et al., 2005). For sensory meat quality, the CG pigs exhibited a lower juiciness score than the AL pigs (P = 0.013), whereas other sensory meat quality characters remained unaffected (Table 7). However, differences were rather small and might not be noticeable by consumers. The positive relation between IMF content and sensory pork quality is widely accepted in literature (Bejerholm and Barton-Gade, 1986; Fernandez et al., 1999; Wood et al., 2004). Indeed, we found a high positive correlation in the current study between IMF content and juiciness sore of loin (r = 0.62, P = 0.001) in agreement with results of Hodgson et al. (1991) and Fernandez et al. (1999), indicating that the lower meat juiciness score of CG pigs could be related to their lower IMF content. Meat from castrated males had a lower (P = 0.014) flavor intensity score than meat from females (4.7 vs. 4.9), but no difference between sexes was found for pork flavor score. We reported a lower IMF content in females than castrated males. Indeed, decreased IMF has been shown to be associated with increased PUFA concentration (Cameron and Enser, 1991; Wood et al., 2004), which is highly sensitive to peroxidation and development of off-flavors. Thus, the greater flavor intensity score of meat from females might have derived from an increased level of off-flavors due to increased oxidation of PUFA, compared with castrated males. Therkildsen et al. (2002) did not report any significant effect of feeding strategy on meat juiciness, in agreement with the current study. Inconsistent results were found concerning effects of restriction-realimentation feeding strategy on pork tenderness. In accordance with present results, Therkildsen et al. (2002) described similar tenderness scores in refed and control pigs. However, Kristensen et al. (2004) reported improved meat tenderness from refed females compared with controls, whereas meat from castrated males was not affected. IMPLICATIONS The objective to produce older pigs at commercial market weight through feed restriction during growing and realimentation during finishing periods was reached. Refed pigs expressed a compensatory growth during the finishing period compared with ad libitumfed pigs. During realimentation, composition of weight gain was modified (greater adipose than lean tissue deposition) at the carcass level. Hence, carcass quality of compensatory gain pigs was comparable to those of ad libitum-fed pigs at live weight of 110 kilograms. However, at muscle level, composition of weight gain was not modified. Consequently, the initial hypothesis of increased intramuscular fat deposition through increased average daily gain and age at live weight of 110 kilograms could not be verified. The restrictionrealimentation feeding strategy did not lead to improved meat eating quality. It can be hypothesized that an elevated intramuscular fat content in refed pigs, which would improve pork quality, might be achieved through forwarding the onset or modifying the duration of restriction and realimentation periods. LITERATURE CITED AOAC Official Methods of Analysis. 15th ed. Assoc. Off. Anal. Chem., Arlington, VA. Bejerholm, C., and P. Barton-Gade Effect of intramuscular fat level on eating quality of pig meat. Page in Proc. 32nd European Meeting. Meat Res. Workers, Ghent, Belgium. Bikker, P., M. W. A. Verstegen, and R. G. Campbell. 1996a. Performance and body composition of finishing gilts (45 to 84 kg) as affected by energy intake and nutrition in earlier life: II. Protein and lipid accretion in body components. J. Anim. Sci. 74: Bikker, P., M. W. A. Verstegen, B. Kemp, and M. W. Bosch. 1996b. Performance and body composition of finishing gilts (45 to 85 kg) as affected by energy intake and nutrition in earlier life: I. Growth of the body and body components. J. Anim. Sci. 74: Cameron, N. D., and M. B. Enser Fatty acid composition of lipid in longissimus dorsi muscle of Duroc and British Landrace pigs and its relationships with eating quality. Meat Sci. 29: Campbell, R., M. Taverner, and D. Curic Effects of feeding level from 20 to 45 kg on the performance and carcass composition of pigs grown to 90 kg live weight. Livest. Prod. Sci. 10:

10 778 Heyer and Lebret Daumas, G., D. Causeur, T. Dhorne, and E. Schollhammer Pig carcass grading methods authorized in France in Journées Recherche Porcine en France 30:1 6. Desmoulin, B., P. Ecolan, and M. Bonneau Estimation de la composition tissulaire des carcasses de porc: Récapitulatif de diverses méthodes utilisables en expérimentation. Prod. Anim. 1: Donker, R., L. Hartog, E. Brascamp, J. Merks, G. Noordewier, and G. Buiting Restriction of feed intake to optimize the overall performance and composition of pigs. Livest. Prod. Sci. 15: Eikelenboom, G., A. H. Hoving-Bolink, and P. G. van der Wal The eating quality of pork. 2. The influence of intramuscular fat. Fleischwirtschaft 76: Ellis, M., A. J. Webb, P. J. Avery, and I. Brown The influence of terminal sire genotype, sex, slaughter weight, feeding regime and slaughter-house on growth performance and carcass and meat quality in pigs and on the organoleptic properties of fresh pork. Anim. Sci. 62: Essén-Gustavsson, B., A. Karlsson, K. Lundström, and A. C. Enfält Intramuscular fat and muscle fibre lipid contents in halothane-gene-free pigs fed high or low protein diets and its relation to meat quality. Meat Sci. 38: Fernandez, X., G. Monin, A. Talmant, J. Mourot, and B. Lebret Influence of intramuscular fat content on the quality of pig meat - 1. Composition of the lipid fraction and sensory characteristics of m. longissimus lumborum. Meat Sci. 53: Gondret, F., and B. Lebret Feeding intensity and dietary protein level affect adipocyte cellularity and lipogenic capacity of muscle homogenates in growing pigs, without modification of the expression of sterol regulatory element binding protein. J. Anim. Sci. 80: Göransson, A., G. von Seth, and E. Tornberg The influence of intramuscular fat content on the eating quality of pork. Pages in Proc. 38th Int. Congr. Meat Sci. Technol., Clermont- Ferrand, France. Hodgson, R. R., G. W. Davis, G. C. Smith, J. W. Savell, and H. R. Cross Relationship between pork loin palatability traits and physical characteristics of cooked chops. J. Anim. Sci. 69: Honikel, K. O Reference methods for the assessment of physical characteristics of meat. Meat Sci. 49: Hornick, J. L., C. Van Eenaeme, O. Gerard, I. Dufrasne, and L. Istasse Mechanisms of reduced and compensatory growth. Domest. Anim. Endocrinol. 19: INRA-AFZ Tables de composition et de valeur nutritive des matières premières destinées aux animaux d élevage. D. Sauvant, J. M. Pérez, and G. Tran, ed. INRA, Paris, France. Kristensen, L., M. Therkildsen, M. D. Aaslyng, N. Oksbjerg, and P. Ertbjerg Compensatory growth improves meat tenderness in gilts but not in barrows. J. Anim. Sci. 82: Kristensen, L., M. Therkildsen, B. Riis, M. T. Sorensen, N. Oksbjerg, P. P. Purslow, and P. Ertbjerg Dietary-induced changes of muscle growth rate in pigs: Effects on in vivo and post-mortem muscle proteolysis and meat quality. J. Anim. Sci. 80: Lebret, B., I. Louveau, and F. Gondret Dietary induced changes in pig growth rate: Consequences on carcass, muscle and meat quality traits at slaughter at 110 kg BW. Page in Proc. Sustainable Pork Production: Welfare, Nutrition and Consumer Attitudes. A. H. Karlsson and H. J. Andersen, ed. Copenhagen, Denmark. MAP Notice technique définissant les critères minimaux à remplir pour l obtention d un label Viande de porc vendue à l état frais ou surgelé et préparations dérivées. Direction de la Politique Economique et Internationale, Ministère de l Agriculture et de la Pêche. pdf/nt8porchom2005.pdf Accessed Mar. 01, Mason, L. M., S. A. Hogan, A. Lynch, K. O Sullivan, P. G. Lawlor, and J. P. Kerry Effects of restricted feeding and antioxidant supplementation on pig performance and quality characteristics of longissimus dorsi muscle from Landrace and Duroc pigs. Meat Sci. 70: McMeekan, C. P Growth and development in the pig, with special references to carcass quality characters. III. Effects of plane of nutrition on the form and composition of the bacon pig. J. Agric. Sci. 30: Mersmann, H. J., M. D. MacNeil, S. C. Seideman, and W. G. Pond Compensator growth in finishing pigs after feed restriction. J. Anim. Sci. 64: Métayer, A., and G. Daumas Estimation of the lean meat content of pig carcasses using the measurement of different cuts. Journées Rech. Porcine en France 30:7 11. Monin, G., and P. Sellier Pork of low technological quality with a normal rate of muscle ph fall in the immediate postmortem period: The case of the Hampshire breed. Meat Sci. 13: Oksbjerg, N., M. T. Sorensen, and M. Vestergaard Compensatory growth and its effect on muscularity and technological meat quality in growing pigs. Acta Agric. Scand. Sect. A-Anim. Sci. 52: Prince, T., S. Jungst, and D. Kuhlers Compensatory responses to short-term feed restriction during the growing period in swine. J. Anim. Sci. 56: Raj, S., G. Skiba, H. Fandrejewski, and D. Weremko Feed intake of pigs during restriction and subsequent compensatory growth. Page in Proc. Sustainable Pork Production: Welfare, Nutrition and Consumer Attitudes. A. H. Karlsson and H. J. Andersen, ed., Copenhagen, Denmark. Schäfer, A., K. Rosenvold, P. P. Purslow, H. J. Andersen, and P. Henckel Physiological and structural events postmortem of importance for drip loss in pork. Meat Sci. 61: Skiba, G., H. Fandrejewski, S. Raj, and D. Weremko Compensatory response of pigs kept in different feeding strategies. Page in Proc. Sustainable Pork Production: Welfare, Nutrition and Consumer Attitudes. A. H. Karlsson and H. J. Andersen, ed. Copenhagen, Denmark. Talmant, A., X. Fernandez, P. Sellier, and G. Monin Glycolytic potential in longissimus muscle of Large White pigs as measured after in vivo sampling. Page 1129 in Proc. 35th Int. Congr. Meat Sci. Technol., Copenhagen, Denmark. Therkildsen, M., B. Riis, A. Karlsson, L. Kristensen, P. Ertbjerg, P. P. Purslow, M. D. Aaslyng, and N. Oksbjerg Compensatory growth response in pigs, muscle protein turnover and meat texture: Effects of restriction/realimentation period. Anim. Sci. 75: Therkildsen, M., M. Vestergaard, H. Busk, M. Jensen, B. Riis, A. Karlsson, L. Kristensen, P. Ertbjerg, and N. Oksbjerg Compensatory growth in slaughter pigs In vitro muscle protein turnover at slaughter, circulating IGF-I, performance and carcass quality. Livest. Prod. Sci. 88: Valaja, J., T. Alaviuhkola, K. Suomi, and I. Immonen Compensatory growth after feed restriction during the rearing period in pigs. Agri. Sci. Finland 1: Wood, J. D., S. N. Brown, G. R. Nute, F. M. Whittington, A. M. Perry, S. P. Johnson, and M. Enser Effects of breed, feed level and conditioning time on the tenderness of pork. Meat Sci. 44: Wood, J. D., G. R. Nute, R. I. Richardson, F. M. Whittington, O. Southwood, G. Plastow, R. Mansbridge, N. da Costa, and K. C. Chang Effects of breed, diet and muscle on fat deposition and eating quality in pigs. Meat Sci. 67: