Effect of the feeding level during the fattening phase on the productive parameters , carcass characteristics and quality of fat in heavy pigs *

The objective of this experiment was to study the effect of feeding restriction carried out at the beginning or at the end of the fi nishing period on the productive results, carcass characteristics and fat quality of Iberian pigs fattened with feed in confi nement. During the fi nishing period (77 days, from 100 to 150 kg of liveweight), the Iberian pigs were given the following feeding treatments: 1. R+AL (restricted feeding from the beginning of fi nishing period to 43 days later and ad libitum feeding from 44 days to the slaughter); 2. AL+R (ad libitum feeding from the beginning of the fi nishing period to 43 days later and restricted feeding from 44 days to slaughter); 3. C: constant feeding. The productive results (weight at slaughter, average daily gain and carcass weight) were higher (P<0.001) in C pigs than in R+AL and AL+R. The treatment had not signifi cant infl uence on carcass characteristics except for the variables dorsal fat thickness and ham weight with respect to carcass weight. The concentration of C18:1n-9 and total MUFA proportions observed in subcutaneous backfat were higher in AL+R and C groups than in R+AL (P<0.05), while the C18:0 and total saturated fatty acids proportions were lower in AL+R and C pigs than in R+AL (P<0.01). The dietary treatment had not signifi cant infl uence on the fatty acids proportions detected in the intramuscular and in liver fat. The results obtained in this experiment suggest the future development of other experimental designs, to improve the productive results and quality of Iberian pigs fed with feed in confi nement by means of feeding restriction strategies.


INTRODUCTION
Feed restriction as a farm animal production strategy can prevent an excessive lipid deposition during the fi nishing period and induce economical benefi ts.Restriction followed by ad libitum feeding produces a phenomenon of compensatory growth (McMeekan, 1940) that infl uences carcass composition (Donker et al., 1986).Hence, it has been reported in pigs of improved genotypes that feed restriction during the growing period (from 30-70 kg liveweight) reduces lean and adipose tissue at the carcass level, while from 70 to 110 kg liveweight the compensatory growth feeding strategy increases adipose but not lean tissue (Heyer and Lebret, 2007).The effect of feed restriction on total fatness could be of higher magnitude in heavy pigs (up to 160 kg liveweight) produced to obtain quality meat products.One example of heavy pig is the Iberian pig, fed mixed diets in confi nement.Nowadays, the range of quality among the different types of products from the Iberian pig is determined by the fatty acid profi le of fat (Rey et al., 2006).However, there is not information on the possible effects of the feed ingestion level during the fattening period on the quality characteristics (fatty acid profi le of the subcutaneous, intramuscular and liver fat) in Iberian pigs fed in confi nement.
Iberian pigs fed mixed diets during the fattening phase (from 100-150 kg) usually receive -a daily amount of feed that approaches 80-85% of the ad libitum ingestion with the purpose of avoiding overload of the digestive system and an excessive fattening that would penalize the carcass yield at slaughter (at least ten months of age according to the Spanish Quality Rule).The results obtained in previous experiments carried out on compensatory growth in pigs of improved genotypes selected for the production of lean meat (Donker et al., 1986;Critser et al., 1995;Daza et al., 2007a), suggested to consider the possibility that an initial restriction of the energy ingestion in the fattening phase of heavy pigs could lead to an improvement of the productive results, carcass characteristics and fat composition.On the other hand, an energy restriction at the end of the fi nishing period could reduce the fattening and improve carcass yield without a signifi cant alteration in the fatty acid profi le.
Hence, the objectives of the present experiment were to study the infl uence of an energy restriction carried out either at the beginning or at the end of the fi nishing phase on the productive parameters, carcass characteristics and fat composition (fatty acid profi le of subcutaneous, intramuscular and liver fat) of heavy pigs such as the Iberian pig fed in confi nement.DUNKER A. ET AL.

Animals and diets
Twenty four castrated male Torbiscal pigs (El Deheson del Encinar, Junta de Comunidades de Castilla-La Mancha, Oropesa, Toledo, Spain) were selected at 52.2 kg SEM±1.3 kg liveweight and approximately 6 months old.From 52.2 to approximately 100 kg (152 days from June to November), pigs received a commercial diet (1.7 kg/day).In the last fattening phase (from 97.7±2.6 kg) pigs were randomly distributed in three groups.Each pig was located in an individual pen (8 m 2 ) to control the feed ingestion.One group (R+AL) was fed a restriction feeding of 2.5 kg feed/day during the fi rst 43 days of the fattening phase, followed 20.16 C18:3 n-3 0.34 1 calculated values by ad libitum feeding until slaughter (34 days).A second group (AL+R) was fed ad libitum during the fi rst 43 days of the fattening phase, followed by a restriction feeding (2.5 kg feed/day) during the last 34 days until slaughter.The third group (C) received 4 kg feed /day during the whole fattening phase (77 days) (this feeding program is the most commonly used in the fattening phase for the Iberian pig reared indoors).All pigs were weighed at the beginning of the experiment, at the initial of the fattening phase, after 43 days of the initial fattening phase, and 18 h before slaughter.Water was provided ad libitum.The chemical and major fatty acids composition of the formulated experimental diets, are shown in Table 1.
Determination of the compositional analysis of feeds (in triplicate) was carried out according to AOAC (1996).Fatty acids of diets were extracted and quantifi ed by the one-step procedure of Sukhija and Palmquist (1988) from lyophilized samples.Fatty acid methyl esters were analysed by gas chromatography using a Hewlett Packard HP-5890 (Avondale, PA, USA) gas chromatograph equipped with a fl ame ionization detector and a capillary column (HP-Innowax, 30 m length × 0.32 mm internal diameter and 0.25 µm polyethylene glycol-fi lm thickness) (Lopez-Bote et al., 2002).

In vivo sample collection
Backfat biopsy samples were taken at the beginning of the fattening phase and 43 days after, at the level of the tail using a metal cylinder (diameter 0.25 cm) with a sharpened edge.All necessary precautions were taken to prevent animal discomfort during and after the in vivo sampling processes.This included tranquillization with 40 mg of azaperon (Stressnill, Labopica, Madrid) 1 h before biopsy and local anaesthesia with 2% lidocaine-HCl, immediately prior to sample collection.Afterwards, animals received a 2 ml penicillin intramuscular injection (300 000 IU ml-1, Labopica, Madrid).Backfat samples were separated into outer and inner layers which were independently analysed for fatty acid composition.

Slaughter, sample collection and chemical analysis
Animals were stunned and slaughtered at a local slaughter house (Talavera de la Reina, Toledo, Spain) at a liveweight of approximately 153.8±2.0 kg.The following measurements were taken: carcass weight, internal length of the carcass, left ham length, left ham perimeter, left ham weight, left shoulder weight, ham %, shoulder %, and subcutaneous fat thickness at the level of the last rib.
Samples of liver were taken at the slaughter time.Samples of Longissimus dorsi muscle at the level of the last rib were taken at cutting (24 h after slaughter), weighed, vacuum-packed in low-oxygen permeable fi lm and kept frozen at -20ºC until analysis.Backfat samples were also taken after slaughter at the level of the last rib and separated into outer and inner layers which were independently analysed for fatty acid composition.Analyses were carried out within three weeks of slaughter.
Subcutaneous fat (inner and outer layers) was extracted using the method of Bligh and Dyer (1959).Intramuscular and hepatic neutral and polar lipids were extracted by consecutive solvent elution with dichloromethane and dichloromethane/methanol (90/10 v/v) according to Marmer and Maxwell (1981).Fat extracts were methylated in the presence of sulphuric acid and analysed as described for feed fatty acids.

Statistical analysis
The data were analysed as a completely randomised design using the statistical program Statgraphics-Plus (version 5.1, 2001).The individual pig was the experimental unit for analysis of all data.The results were analysed by a covariate analysis in which treatment was the fi xed effect and initial weight the covariate.If covariate was not statistically signifi cant (P>0.05) it was removed from the model.The comparative analysis between means was conducted using the Duncan t-test.
Data are presented as the mean of each group and the mean standard error (SEM) together with the signifi cance levels.

Productive parameters
During the previous period to fattening (152 days from June-November) the average daily gain of the pigs was 298.9±16.8g and signifi cant differences were not detected for the three groups of pigs (R+AL:301.3;AL+R:294.1 and C:301.3).The average daily consumption of feed per pig was 1.72 g during that period.
The productive results obtained during the fi nishing phase according to the applied feeding pattern are given in Table 2.The average daily gain (ADG) was signifi cantly higher in the pigs that received a constant amount of feed (C) when compared with those that were restricted at the beginning (R+AL) or at the end of the fi nishing period (AL+R).Moreover, the group of pigs restricted at the end of the fattening phase showed a higher ADG than those restricted to the beginning.The restricted pigs consumed less feed that those that received a constant feeding (C) and the lowest average daily feed intake (ADFI) was for the pigs fed R+AL.
However, the feeding pattern did not have a signifi cant statistically infl uence on the total transformation index of the feed (IT).As expected, the feed restriction had a negative effect on the transformation index of the feed and it was especially marked in the pigs restricted at the end of the fattening that had higher liveweight (due to the previous ad libitum feeding) and consequently dedicated an important fraction of the feed to satisfy its maintenance necessities.
The group that was restricted in the beginning (R+AL) showed from 43-77 days a higher feed consumption (ADFI 2 ) and average daily gain (ADG 2 ).These results are in agreement with those reported by Critser et al. (1995) and Daza et al. (2007a).However, R+AL group did not reach the ADG of the pigs that received a constant feeding (C).Not completed compensatory growth have also been evidenced in other experiments carried out with pigs of improved genotypes for the production of lean meat (Prince et al.,1983;Donker et al., 1986) and in Iberian pigs fattened in free-range (Daza et al., 2005a) and with feed in confi nement (Daza et al., 2007a).On the other hand, the restriction applied at the end of the fi nishing period (AL+R) was too severe and produced the lowest ADG 2 and consequently, a lower total ADG than those pigs that received a constant feeding.024 SEM -standard error of the mean, ADG 1 -average daily gain during the fattening phase (from 0 to 43 days), ADG 2 -average daily gain during the fattening phase (from 43 to 77 days), ADG -average daily gain during whole period of fattening (0-77 days), ADFI 1 -average daily feed ingestion during the fattening phase (0-43 days), ADFI 2 -average daily feed ingestion during the fattening phase (from 43 to 77 days), ADFI -average daily feed ingestion during the fattening phase (0-77 days) IT 1 -transformation index of feed from the beginning of the fattening phase until 43 days, IT 2 -transformation index of feed from 43 days to fi nal period of fattening (77 days), IT -transformation index during the fattening period (from 0 to 77 days) values with different superscript are signifi cantly different at P<0.05

Carcass yield and characteristics
The feeding pattern did not have statistically signifi cant infl uence on the carcass characteristics except for the percentage of ham and fat thickness at the level of the last rib (Table 3).The ham percentage was signifi cantly higher in the pigs that were restricted at the beginning of the fattening period than in the other two groups.Moreover, the correlation coeffi cient among the ham percentage and the carcass weight was r=-0.56 (P<0.004).The reduction of the ham percentage according to the increase of the carcass weight was also observed by Menaya et al. (1998).The fat thickness was signifi cantly higher in the pigs that received a constant feeding (C) that in those restricted during the fi rst 43 days (R+AL) and not signifi cant differences were detected with the group restricted at the end of the period (AL+R).The correlation between the ham percentage of the ham and shoulder and the fat thickness was negative (r=-0.46;P<0.023 and r=-0.37;P<0.05, respectively) which is in agreement with the results found by Aparicio (1987) in Iberian pigs.
Moreover, in Iberian pigs the carcass yield increased with the slaughter weight (Menaya et al., 1998).However, in the present experiment, the range of the slaughter weight was small and consequently, signifi cant variations of the carcass yield were not detected.Similar results were found previously (Daza et al., 2007a).
Hence, the initial hypothesis that a restriction at the end of the fi nishing period would lead to a reduction of the fattening and so to an increase of the carcass yield has not been verifi ed in this experiment.

Backfat, muscle and liver fatty acid profi le
The fatty acid composition of the subcutaneous fat, outer and inner layers, at 43 days are presented in Table 4.The pigs that received a restricted feeding during this phase (R+AL) had, at the end of the same, lower C18:1 n-9 and MUFA and higher C18:2 n-6, C18:3 n-3, PUFA and n-6 than the other two groups of pigs (AL+R and C).The increase of the energy ingestion generates an increment of the proportions of saturated fatty acids and MUFA due in part to an increase of the lipogenic (Bee et al., 1999(Bee et al., , 2002) ) and of the steatoryl-CoA-desaturase enzymes (Enser, 1975;Daza et al., 2007a) at the same time that the proportions of the fatty acid C18:2 n-6, C18:3 n-3 and total PUFA decrease.In this experiment we estimated the activity of the enzyme steatoryl-CoA-desaturase (ECD), using the calculation of insaturation indexes (C16:1n-7 /C16:0 and C18:1n-9/C18:0).The activity of the enzyme ECD, from the beginning of the experiment up to 43 days later, was not signifi cantly affected by the treatment.However, Daza et al. (2007a) observed that a lower feeding level followed by a higher one does not mean an immediate increase of the lipogenic activity of the fatty acid synthetase enzyme (FAS).The higher pro-portion of C18:1 n-9 and MUFA in the AL+R group and C would be explained by the highest ingestion of feed and therefore, a higher ingestion of those fatty acids.In the R+AL group an inhibition of the ECD enzyme could also have occurred due to the high proportion of the PUFA which is in agreement with the data reported by Enser (1973) and Bee et al. (2002).The saturated fatty acids and those derived from the de novo synthesis are the main source for the ECD (Enser and Roberts, 1982).
Moreover, signifi cant interactions of the feeding treatment and the fat layer were also found on the proportion of C18:3 n-3, total n-3 and PUFA.Hence, the external layers were more susceptible to changes in the fatty acid proportions as affected by the feeding than the inner layer.
In Table 5 can be observed that at slaughter pigs receiving the feeding pattern AL+R and C had signifi cantly higher proportions of C18:1 n-9 and MUFA and lower C18:0 and SAT than the group R+AL.However, no signifi cant differences were detected for the C16:0, C18:2 n-6, C18:3 n-3, n-6, n-3 and PUFA.In this case, the activity of the ECD enzyme estimated as C18:1 n-9 /C18:0 was signifi cantly higher for the groups AL+R and C than for R+AL group.The higher proportions in C18:0, C18:1 n-9 and MUFA and the lower proportions in PUFA of the group AL+R and C led to an increase of the activity of the ECD and so higher proportions of C18:1 n-9 and MUFA in the subcutaneous fat.Kouba et al. (1997) found that the high activity of the enzyme ECD was related with high proportions of oleic acid in the tissues.The restriction (2.5 kg/day) applied to the pigs AL+R during the last 34 days of the experimental period was not suffi ciently severe to produce a reduction of the proportions of the SAT and MUFA and to increase PUFA.However, it should be observed that in the last phase (from 43 days to slaughter) the proportion of C18:1 n-9 increased more in the pigs R+AL than in AL+R and C (2.91, 1.64 and 2.60%, respectively; SEM=0.009;P<0.01), what could mean that if during this phase the pigs R+AL had more ingestion capacity probably then the proportion of oleic acid in the subcutaneous fat would have been higher than that for the pigs AL+R and C.
In addition, the subcutaneous fat of the outer layer had higher proportions of C18:1 n-9 and MUFA and lower C16:0, C18:0 and SAT than the inner layer.These results are in agreement with those observed by Migdal et al. (2001), López-Bote et al. (2002), Rey et al. (2005) and previously by Dean and Hildritch (1933) who reported that the degree of unsaturation from the outer to inner backfat layer was inversely related to the environmental temperature.The proportions of C18:2 n-6, C18:3 n-3, n-6 and PUFA were higher in the inner layer than in the outer 43 days after the beginning of the experiment.At slaughter, the inner layer had higher concentrations of C18:2 n-6, C18:3 n-3 and n-6 than the outer.However, the outer layer had a higher proportion in n-3 fatty acids (C20:5 n-3 and C22:6 n-3) (0.11 vs 0.09% and 0.51 vs 0.028%, respectively; P<0.05).The PUFA proportion according to the subcutaneous layer was very variable.In this way, Daza et al. (2005b) in crossed Iberian x Duroc pigs fed with feed in confi nement and Rey et al. (2005) in Iberian pigs found that the outer layer was more unsaturated than the inner, while Daza et al. (2007a) found the opposite results in Iberian pigs fattened with feed in confi nement.It seems that when feed is rich in PUFA the differences among the accumulated proportions of such fatty acids in the external and internal layers can be diminished or it can be even higher in the inner (Bee et al., 2002;López-Bote et al., 2002).In a previous study, Mac Grath et al. (1968) reported that when pigs were fed with a tallow supplemented diet there was a temperature gradient from the outer to inner backfat layer inversely related to total fat unsaturation, while no clear relationship was detected when pigs were fed with a maize oil-supplemented diet.Consequently these authors concluded that the relationship between depot fat unsaturation and environmental temperature is infl uenced by the polyunsaturated fatty acids received by the feeding.
The fatty acid profi le of the intramuscular fat according to the feeding pattern is shown in Table 6.The feeding level neither affected the composition of the intramuscular fat in the Longissimus dorsi muscle neither the ECD activity.DUNKER A. ET AL.Kuhn et al. (1995Kuhn et al. ( , 1997) ) found that a reduction of the feed ingestion did not affect the fatty acid composition of the intramuscular fat.So, to fi nd differences in the proportions of saturated fatty acids and MUFA, changes in the feed ingestion should be high (close to 30%).Kondracki (2000) and Rey et al. (2005) did not detect differences among the proportions of saturated, MUFA and PUFA in the neutral and polar lipids of the intramuscular fat in Iberian pigs fed acorn, or acorn and grass in confi nement.Qualitative important changes are needed in the feeding type during the period of fi nishing (for example free-range feeding vs a mixed diet in confi nement) to detect changes in the intramuscular fat (Tejeda et al., 2002).As it was expected, the proportions of saturated fatty acids and MUFA were higher in the neutral than in the polar lipids, while PUFA were higher in the polar than in the neutral lipids.These results are in agreement with those reported by Warnants et al. (1999), Bee et al. (2002) and Rey et al. (2005).
In Table 7 is shown the fatty acid profi le in the liver fat according to the feeding level.The pigs that were restricted at the end of the period (AL+R) had higher proportions of C22:6 n-3 (P<0.05) and total n-3 fatty acids (P<0.05) in the neutral lipids than the other two groups of pigs.However, differences were not statistically different in the remaining fatty acids.In accordance with Ruiz et al. (1998) the profi le of the fatty acids in the liver fat is a faithful refl ection of the fatty acid composition of the plasma and therefore, of the recent feeding received by the pigs.Hence, it should be expected that differences among the different experimental treatments in the liver fat were not detected.Do- cosahexanoic acid (C22:6 n-3) comes from the linolenic acid (C18:3 n-3) by the enzyme activity which also produces the transformation of the linoleic acid (C18:2 n-6) in arachidonic acid (C20:4 n-6).In the present experiment the increase in the proportion of C22:6 n-3 of the AL+R group was probably due to a higher enzyme activity and was responsible of the higher content of the total n-3 fatty acids because the other n-3 (C18:4 n-3, C20:5 n-3 and C22:5 n-3) did not change by the feeding, the ratio n-6/n-3 being more favourable in the group AL+R than in the remaining groups.The polar lipids of the liver fat were more saturated and less monounsaturated that the neutral lipids and the proportions of total PUFA were similar in both classes of lipids.These results are in agreement with those observed by Daza et al. (2007b) in Iberian pigs fattened in confi nement.

CONCLUSIONS
It is concluded, that the restriction of the energy ingestion at the beginning or at the end of the period of fi nishing Iberian pigs fed in confi nement does not produce an improvement of the productive results, quality of the carcass and quality of the fat when it is compared with a model of conventional constant feeding (average daily ingestion of 83% of the amount of feed received ad libitum during the fattening phase).So, the restriction of feed could be interesting from the economic point of view for producers.

Table 1 .
Ingredient and chemical composition of the experimental diets

Table 2 .
Effect of the feeding level on productive parameters of Iberian pigs fed in confi nement

Table 3 .
Effect of the feeding level on carcass parameters of Iberian pigs fed in confi nement

Table 4 .
Effect of the feeding level on fatty acid profi le (%) of subcutaneous fat at 43 days of fattening in Iberian pigs fed in confi nement

Table 5 .
Effect of the feeding level on fatty acid profi le (%) of subcutaneous fat at slaughter in Iberian pigs fed in confi nement

Table 6 .
Effect of the feeding level on fatty acid profi le (%) of Longissimus dorsi muscle in Iberian

Table 7 .
Effect of the feeding level on fatty acid profi le (%) of liver in Iberian pigs fed in confi nement