Corns with Different Nutritional Profiles on Growing and Finishing Pigs Feeding (30 to 90 kg)*

: Three experiments were carried out to determine the nutritional value and to verify the pig performance in growing and finishing phases (30 to 90 kg) fed on diets containing common corn (CC), high-lysine corn (HLC) and high-oil corn (HOC). In the total digestibility trial (Exp. I) 12 barrows were used. Values of digestible energy (DE) and metabolizable energy (ME) as-fed basis for CC, HLC and HOC, were: 3,396 and 3,275 kcal/kg; 3,248 and 3,139 kcal/kg; 3,445 and 3,308 kcal/kg, respectively. In order to determine the apparent and true ileal digestibility coefficients of amino acids, as well as the values of true digestible amino acids of the CC, HLC and HOC, an ileal digestibility trial was done (Exp. II) with T-cannulated barrows (“T” simple). The treatments consisted of three diets, with one of them as the sole source of protein (CC, HLC and HOC). In the performance experiment (Exp. III), 36 crossbred pigs, allotted in a completely randomized design with three treatments and 12 replications were used. Treatments consisted of three diets: 1 - CC; 2 - HLC and 3 - HOC. It was observed no difference for performance and carcass variables among the corns with different nutritional profiles. Results of the three experiments highlighted the importance of segregating corns in their real chemical and energetic composition as well as the values of true digestible amino acids for formulating diets for growing and finishing pigs. ( Words : Bio Economic Studies, Carcass,


INTRODUCTION
Corn is an important ingredient used in diets for pigs in Brazil. About 80% of the Brazilian production of corn is used for the preparation of animal feed. It is characterized as energetic food source (3,930 kcal of GE/kg), rich in starch (63.00%) and contains around 8.00% of CP and 3.60% of crude fat (Rostagno et al., 2005).
Currently, there are in the market new varieties and hybrids of corn, developed with the genetic improvement and manipulation by the introduction of technologies of molecular biology, which have superior nutritional profile when compared to the traditional genotypes, thus they are more suitable for animal feed.
Corn with high levels of lysine has 0.35% lysine (the common corn has 0.24%), arousing the interest of its use in feed for pigs because it gathers all the good agronomic characteristics when compared to the common corn, besides having higher content of tryptophan (Saldivar and Rooney, 1994). Corn with a high-oil content has 6.40% of crude fat (Rostagno et al., 2005) which increases the energy content allowing to formulate diets with higher energy density which can improve the performance of animals.
In many cases pig diets are formulated with values reported in tables of feed composition, although the average nutritional profile of the raw materials may differ. Thus, it has been constantly sought to formulate diets with values closer to the real chemical composition and energy value of the different types of corns available in the market (Vieira et al., 2007).
Beside the energy content it is also necessary to know the amount and the digestibility of amino acids present in the corn (Tanksley and Knab, 1993). However, values for the ileal digestibility of amino acids must be preferred due to the possible interference of the bacterial flora in the metabolism of amino acids in the large intestine (Fan and Sauer, 1995).
Formulation of diets based on standardized ileal digestible amino acids has been used by nutritionists, mainly because of the necessity to optimize the use of raw materials of high cost and to ensure a sufficient supply of digestible amino acids by the correction of deficiencies with supplementation of synthetic amino acids (Sakomura and Rostagno, 2007).
This study was carried out aiming to evaluate the use of corns with different nutritional profile in practical growing and finishing pig diets, and their effects on performance, carcass quality and economic return.

MATERIAL AND METHODS
The experiments were carried out at the Maringá State University Pig Farm in the State of Paraná (23° 21'S, 52° 04'W, at a height of 564 meters).
Three different corns were used: common corn (CC) -BR1030 hybrid, which was provided by Embrapa CNPMS; high-lysine corn (HLC) -BR473 variety, which is under development by Embrapa CNPMS and high-oil corn (HOC) -DAS766 hybrid, from the seed company DowAgroScience.
Three corns were ground in a hammermill equipped with a 2 mm screen. Chemical compositions and energy of the CC, of the HLC, of the HOC, of the diets and feces were obtained in the Laboratory of Food Analysis and Animal Nutrition at Maringá State University. Analysis of dry matter (DM), crude protein (CP), ashes, calcium (Ca), total phosphorus (Pt) and crude fat (CF) were carried out according to the methods described by Silva and Queiroz (2002). The determination of starch in food and in the feed and feces were carried out according to the enzymatic method proposed by Poor et al. (1989), adapted by Pereira and Rossi (1995). Values of gross energy (GE) were determined by means of adiabatic calorimeter (Parr Instrument Co. AC720), according to the procedures described by Silva and Queiroz (2002). Contents of amino acids in the corns were determined by HPLC in the CBO Laboratory Analysis in Campinas, Brazil.

Experiment I: Total digestibility trial
Initially, the three corns (CC, HLC and HOC) were analyzed for DM, GE, CP, organic matter (OM), CF, starch, ashes, Ca and Pt and total amino acids (Tables 3 and 5).
A total digestibility trial was carried out, using 12 crossbred barrows, from commercial line (TOPIGS), with initial body weight of 44.2±0.8 kg. The animals were individually allotted in metabolism cages type "PEKAS" in a controlled environment room and average room temperature with minimum of 18°C and maximum of 24°C.
CC, HLC and HOC were tested, replacing 30% of the reference diet, resulting in three test diets. Reference diet, based on corn and soybean meal was calculated to meet the requirements indicated by NRC (1998).
Four diets were studied: 1 = Reference Diet (RD); 2 = RD (70%)+CC (30%); 3 = RD (70%)+HLC (30%); 4 = RD (70%)+HOC (30%). Collections were done in two periods: the first one consisted of seven days of adaptation to the diets and cages follow by five days of collection of feces and urine; the second one consisted of 3-day interval and a five-day collection. From the first to the second period, the treatments were reallotted.
Other procedures for diets, feces and urine collection were carried out according to those described by Sakomura and Rostagno (2007), as below.
Pigs were fed twice daily (07:00 a.m. and at 01:30 p.m.,) in the proportions of 60 and 40% of the total quantity, respectively. Daily total quantity was defined according to the feed intake in the adaptation phase, based on the metabolic weight (kg 0.75 ) of each experimental unit. Diets were humidified with 15% of water. After each meal, water (3.0 ml/g of diet) was offered in the same feeders. Fe 2 O 3 (2% of inclusion) was used as a marker.
Feces were daily collected in the morning and packaged in freezer (-10°C). At the end of the collecting period, they were dried at 55°C for 72 hours for pre-drying. Later on, the samples were exposed to air in order to balance with room temperature and humidity. Then, they were weighed, ground and mixed and finally samples were collected for analysis.
Urine was collected once a day in plastic buckets with 20 ml of HCl 1:1, aiming not to have nitrogenous loss and the bacterial proliferation. A fabric known as "filó" was put on the collector funnel in order to hold possible fecal waste. Collection was done at 07:30 a.m. in which the liquid into the bucket was filled for fixed volumes to make the removal of the 20% aliquot easier. The urine was kept in flasks into the freezer for laboratory analysis, which were done in the Animal Nutrition Laboratory of Animal Science Department of Maringá State University (LANA/DZO/ DEM) by Silva and Queiroz (2002).
Digestibility coefficients of dry matter (DCDM), of crude protein (DCCP), of organic matter (DCOM), of gross energy (DCGE) and the metabolization coefficient of gross energy (MCGE), of the corns with different nutritional profiles were calculated as described by Matterson et al. (1965), thereby obtaining the values of digestible dry matter (DDM), digestible protein (DP), digestible organic matter (DOM), digestible energy (DE) and metabolized energy (ME).
Experimental design was completely randomized, with three treatments, six replicates and the experimental unit was the single pig. In order to evaluate differences between the digestibility coefficients of the CC, HLC and HOC, the data was submitted to ANOVA and average test (Newman-Keuls test, p<0.05), using the statistical software SAEG (UFV, 2000), in accordance with the following statistical model: Y ij = μ+T i +e ij where: Y ij = digestibility coefficients of corn i, j of repetition; μ = constant associated with all observations; T i = effect of type of corn i, where i = 1, 2, 3, (1 = CC, 2 = HLC, 3 = HOC), and ij = random error associated with each observation.

Experiment II: Ileal digestibility trial
CC, HLC and HOC were analyzed for amino acid composition (Table 5).
Average minimum external temperatures were 13.9± 3.0°C and 19.1±2.4°C and average maximum temperatures were 25.9±3.8°C and 30.0±4.0°C, respectively. The mean air relative humidity in the growing and finishing phases in the morning were 71.5±14.9% and 73.6±13.5% and in the afternoon were 45.9±13.5% and 46.6±16.9%, respectively.
Three crossbred pigs from commercial line (TOPIGS) were used, with 46.3±2.1 kg of BW, which were submitted to a surgery for implantation of a T-shaped cannula, as described by Bellaver (1989). After the surgery, the animals were transferred to individual concrete pens (3.80 m 2 /each), with nipple drinker at the back and semi-automatic feeders at the front, which provided free access to feed and water. These animals were under recovery for 20 days.
Treatments consisted of three diets, with a single source of protein food (CC, HLC and HOC) and oneprotein-free diet (OFD) for estimating endogenous amino acid losses.
Animals showed an average weight of 57.0±0.7 kg and were submitted to a five-day adaptation in order to regulate the intestinal flow and to a one-day collection of ileal digesta of the OFD. The supply procedures for the diets and the collection and processing of the digesta were carried out according to Sakomura and Rostagno (2007).
Ileal digesta of the three test diets was collected (CC, HLC and HOC). Animals were allotted to a double 3×3 Latin square design. First period lasted six days with five days of adaptation and one day of collection of ileal digesta. In the following periods there was a three-day adaptation to the diet and a one-day collection of ileal digesta.
Analyzed digesta of each feed were compounded of the six samples of digesta performed during the six periods.
Experimental diets had 0.5% of chromic oxide (Cr 2 O 3 ) as a marker in the digestibility determination. Daily quantity of diet was provided to each animal twice a day (at 7 a.m. and 7 p.m.). It was calculated according to the metabolic weight (kg 0.75 ).
Digesta samples were collected in polyethylene bags, attached to the cannula and, later on, put into plastic pots which were identified and put into the freezer (-5°C) until the end of the collecting period. At the end of this period, the samples were defrosted, weighed, mixed and lyophilized in order to avoid the amino acid degradation. Dried samples were ground, packaged in glass flasks with lid for future analysis.
Levels of chromium oxide and DM in the digests were determined according to the methods described by Silva and Queiroz (2002). The contents of amino acids in the diets and digesta samples were determined by HPLC in the CBO Laboratory Analysis in Campinas, Brazil.
Determination of ileal digestibility of amino acids was calculated based on the levels of chromium (Cr) in the diets and digests of pigs by means of calculating the indigestibility factor (IF) using the formulas described by Sakomura and Rostagno (2007).
To determine the coefficients of SID amino acid digestibility of the corns with different nutritional profiles, the values of endogenous amino acid losses were used which were obtained from the animals that received OFD.

Experiment III: Performance experiment using corns with different nutritional profiles
Once obtained the chemical composition, the energy values and the SID digestible amino acid values of CC, HLC and HOC, they were used in the formulations of the experimental diets in the growing and finishing phases (from 30 to 60 kg of BW and from 60 to 90 kg BW).
Thirty-six crossbred pigs (TOPIGS) were used from a commercial line, with an initial BW of 31.1±4.1 and a final body weight of 60.7±6.2 in the growing phase and a final body weight of 91.1±5.5 kg in the finishing phase.
Growing and finishing experiments were conducted during autumn season. Average minimum external temperatures were 19.1±1.3°C and 12.5±4.3°C and average maximum temperatures were 29.3±2.2°C and 21.5±3.5°C, respectively. The mean air relative humidity during the growing and finishing phase was 82.4±11.6% and 87.4±7.6% in the morning and 58.3±12.6% and 68.2±12.9% in the afternoon, respectively.
Animals were allotted in shed, covered with fibercement tiles, with 36 pens (1.90 m 2 /each) with nipple drinker at the back and semi-automatic feeders at the front. Diets and water were given ad libitum throughout the experimental period.
Treatments consisted of three diets (Tables 1 and 2), where: 1 = basal CC, 2 = basal HLC and 3 = basal HOC. Diets had the same nutritional levels and they met the requirements indicated by NRC (1998).
Once determined the chemical composition, the energy and amino acid values of the corns with different nutritional profiles (Tables 3, 4, 5, 6 and 7), such data were used for the formulations of the diets (Tables 1 and 2). For the other ingredients, the chemical composition and energy value determined by Rostagno et al. (2005) were used. L-lysine, DL-methionine and L-threonine were added, complying   Rostagno et al. (2005). with the pattern of the ideal protein in terms of digestible amino acids, as indicated by NRC (1998).
Pigs were distributed in a completely randomized design with three treatments and 12 replications, both in the growing and finishing phases. Experimental unit was the single pig. At the end of the growing phase, the animals were redistributed at random to the treatments for the finishing phase.
Pig and feeder weights were measured at the beginning and at the end of the experiment to calculate the daily feed intake (DFI), daily weight gain (DWG) and feed:gain rate (FGR) of each experimental unit. At the end of the growing and finishing phases the backfat thickness (BT) and loin depth (LD) at the P2 position were measured using a Sono-Grader equipment (Renco ® ).
At the end of the finishing phase, all the pigs were slaughtered in the slaughterhouse of the Maringá State University Pig Farm. Carcasses were refrigerated (1-2°C) for 24 hours to be subsequently submitted to quantitative evaluation, according to the Brazilian Method of Carcass Classification (ABCS, 1973). For qualitative evaluation of the carcass, samples were taken from M. Longissimus dorsi at the 8th and 10th vertebrae for subsequent measurement of intramuscular fat, that is, of marbling and the drip loss, as NPPC (1991). Areas of M. Longissimus dorsi and of fat were determined using digitizing tablet and the software SPRING (Câmara et al., 1996).
Economic analysis was calculated by the following expression adapted from Guidoni et al. (1997): where: MPC = maximum price of the corn (HLC and HOC) so that the diet which will be used has the same economic efficiency as the reference diet (CC); PRP = price per kilo of pig; Gain i = average weight gain of pigs of the treatment with corn (HLC or HOC); Gain 0 = average weight gain of pigs of the reference treatment (CC); P j = price of the other ingredients in each diet; C ji = percentage of the ingredient j in the diet i; FI i = average total feed intake per animal inherent to diet i; C j0 = percentage of the ingredient j in the reference diet; FI 0 = average total feed intake per animal on the reference diet; C li = percentage of the corn (HLC or HOC) in the diet i.
The following prices were used as inputs (region of Maringá, Paraná, Brazil; current prices of February 2008): common corn R$ 0.41, soybean meal R$ 0.76, limestone R$ 0.16, dicalcium phosphate R$ 2.10, sodium chloride R$ 0.34, vitamin and mineral mix for pigs in the growing phase R$ 5.00, vitamin and mineral mix for pigs in the finishing phase R$ 6.40, L-lysine R$ 7.08, DL-methionine R$ 24.00, L-threonine R$ 8.25 and growth promoter R$ 135.00. Price of the pig was R$ 3.50.
Results of the different variables were submitted to ANOVA and to the average test (Newman-Keuls test, p<0.05). Observations were analyzed with the following statistical model: Y ijk = μ+C j +e ijk, where Y ijk = observed value of variables for each individual j, receiving the corn i; μ = general constant; C j = effect of type of corn j (j = 1, 2 and 3); and e ij = random error associated with each observation. Statistical analyses were carried out using the statistical software SAEG (UFV, 2000).

RESULTS AND DISCUSSION
CP, lysine and threonine levels of the HLC were increased by 29.68; 57.14 and 31.25% respectively, compared to the CC, as expected. While the crude fat level of the HOC was increased by 5.47% compared to the CC (Tables 3 and 5). CP, lysine and threonine levels of the HLC were  Rostagno et al. (2005) when converted to the same basis of DM (Tables 3 and 5). Similarly, the crude fat level of the HOC was increased by 20.98% if compared to the corn (Rostagno et al., 2005). However, it was decreased by 31.76% if compared to the high-oil corn (Rostagno et al., 2005) and decreased by 9.74% if compared to DAS766 corn (Silva et al., 2006). In general, the results obtained from chemical composition are between the minimum and maximum values shown by van Mileen and Noblet (2004). These possible variations in the levels of the nutrients among the corns may be attributed to several other factors, such as the genetic potential of the seeds for this attribute, the level of fertilization used (especially N), the soil fertility and the climatic conditions (Noblet and van Mileen, 2004).

Experiment I: Total digestibility trial
There were no differences (p>0.05) in digestibility coefficients of DM, CP and of OM and in the metabolization coefficient of gross energy among the three corns. However, digestibility coefficient of GE was superior (p<0.05) for HOC compared to HLC (Table 4).
Comparing the DCGE and MCGE of the CC and HOC to those shown in the guides of Embrapa (1991) and Rostagno et al. (2005)  However, when comparing the three studied corns (CC, HLC and HOC) among themselves, the HOC had the highest level of CF (Table 3) and thus showed the best digestible energy and the best values of DE and ME. The same was observed by Adeola and Bajjalieh (1997) when examined different hybrids of corns with high level of oil and found variation from 3,400 to 3,570 kcal DE/kg compared to 3,290 kcal DE/kg of common corn.
HLC presented (Table 4) lower DCGE and MCGE, when compared to HOC, as well as the values in the guide by Rostagno et al. (2005), for high-lysine corn (89.76 for DCGE and 87.25% for MCGE). These results are similar to those published by Burgoon et al. (1992) who studied common corn, QPM corn (quality protein maize) and two types of corn with high protein level. They reported a DCGE of 89% for all the corns studied. ME:DE ratios of the three studied corns give a mean value of 0.96 which is in accordance with that one reported by NRC (1998), van Mileen andNoblet (2004) and Pozza et al. (2005).

Experiment II: Ileal digestibility trial
Amino acids values of the CC and HOC were similar to each other. However, the values of some essential amino acids (lysine, threonine, arginine, histidine and valine) of the HLC were numerically superior compared to the other corns (Table 5).
The amino acid composition of the three corns, the apparent digestibility coefficients (ADC), the true digestibility coefficients (TDC) and the true digestible amino acids (Tables 6 and 7) had similar values when compared to those presented by Rostagno et al. (2005) for corn, high-lysine corn and high-oil corn, as well as when compared to those listed by Fontes et al. (2007) for common corn and high-protein corn.         It is important to consider the variation in the digestibility of amino acids within each type of corn. Sauer and Ozimek (1986) reported that the digestibility of lysine ranges from 71% to 82% in corn common. A large part of this variation may be related to the relatively low content of amino acids in the cereal grains (Fontes et al., 2007). Small changes in the levels of endogenous amino acids could explain large variations in the values of apparent digestibility of amino acids, since this is expressed as percentages, especially for those amino acids which occur in small levels in grains such as lysine and tryptophan and to those which are found in relatively high concentrations in the endogenous secretions such as threonine, glycine and proline.

Experiment III: Performance test using corns with different nutritional profiles
In growing and finishing phases the variables DFI, DWG, FGR, BT and LD were similar between the treatments (Table 8). Similarly, in the finishing phase, the carcass variables (Table 9) were similar (p>0.05). This was expected as the fed diets were isonutrient (isocaloric, isophosphoric, isocalcium, isoaminoacidic for lysine, methionine+cystine and threonine). For the formulation of the diets the exact chemical composition and energy values as well as the SID digestible amino acids of the different corns were used. This has permitted the formulation of a diet which complied properly with the nutritional requirements proposed by NRC (1998). Thus, the pigs in the growing and finishing phases showed quite similar performance and carcass characteristics.
These results are consistent with those found by Spurlock et al. (1997), Lima et al. (2003) and Silva et al. (2006) who found no differences in the DFI and DWG of pigs in growing and finishing phases fed with diets containing total replacement of common corn by higher oil content corn.
Most studies evaluating high-oil level corn have been in animals for growing and finishing phases because of its higher energy content (Spurlock et al., 1997;Snow et al., 1998). Normally, pigs over 50 kg of body weight have got a good capacity to use supplemental dietary fat (Adeola and Bajjalieh, 1997;O'Quinn et al., 2000). Besides the age factor, Adeola and Bajjalieh (1997) highlighted the suitability of other dietary nutrients and the nutrient:calorie ratio in order to justify the different responses in performance with the addition of dietary fat.