Apparent and standardized ileal digestibility of protein and amino acids of several fi eld bean and pea varieties in growing pigs

The apparent ileal digestibility of protein and amino acids of fi eld beans (cv. Nadwiślański, Kodam, Martin and Optimal) and of peas (cv. Piast, Rola, Idol and Wiato) was determined by the indicator method using Cr2O3. The experiment was conducted in a double Latin square design (5 × 5 and 4 × 4) on nine barrows with an initial body weight of about 25 kg equipped with simple T-cannuale inserted into the terminal ileum. The only source of protein in the experimental diets was either fi eld beans or peas, whereas in the control diet soyabean oilmeal. The digestibility of protein and of most amino acids was signifi cantly lower in fi eld beans and peas (P<0.05) than in soyabean oilmeal. No signifi cant differences were found (P>0.05) in the digestibility of protein or amino acids among the tested varieties of fi eld beans. Signifi cant differences did, however, appear in the digestibility of protein and amino acids between whiteand coloured-fl owered pea varieties (P<0.05). Standardized protein and amino acid digestibility of the studied varieties of fi eld bean and peas was calculated. The coeffi cients of standardized digestibility were somewhat higher than those of apparent ileal digestibility.


INTRODUCTION
The systems of feed formulation currently in use for pigs recommend taking into account the amino acids digested in the small intestine (NRC, 1998;Grala et al., 1999). When determining apparent ileal digestibility of protein and amino acids, endogenous nitrogen compounds are not factored in, although they constitute a signifi cant pool of compounds excreted with faeces and have a relatively constant chemical composition (Jondreville et al., 1995). Losses of endogenous nitrogen are taken into account in modern standardized digestibility coeffi cients, which are a better indicator of the availability of amino acids in the gastrointestinal tract of pigs than apparent digestibility coeffi cients. The equations given by Degussa (2001) are used to calculate standardized digestibility coeffi cients.
Over the past years breeding has led to the registration of legume varieties having reduced levels of antinutrients, including tannins, ADF, and NDF (Gdala et al., 1992;Abrahamsson et al., 1993;Grala et al., 1993;Mariscal-Landin et al., 2002). Tannins form complexes with proteins, metal ions and other macromolecules, lowering their availability to animals (Abrahamsson et al., 1993). The tannins in fi eld beans and peas lower protein and energy digestibility, reduce body weight gain and feed utilization by pigs (Selgado et al., 2002a). As the tannin content of the diet increases, the activity of digestive enzymes has been found to decrease, accompanied by development of disorders in liver and digestive tract functions (Jansman, 1993).
It may be presumed that the seeds of the new varieties of legumes, with reduced levels of antinutrients, may be an alternative to soyabean oilmeal in pig nutrition.
The objective of this study was to determine the apparent and standardized digestibility of the protein and amino acids of selected varieties of fi eld beans and peas.

Protein sources
The study was conducted using the fi eld bean variety Nadwiślański (a traditional variety registered in 1955), and the recently registered varieties Kodam (conventional), Martin and Optimal (self-determinate), as well as pea varieties Piast and Rola (white-fl owers) and Idol and Wiato (coloured-fl owers). When choosing the varieties, their chemical composition, yield stability, protein yield per hectare (Wiatr, 1999) and antinutrient content were taken into account.

Animals, diets and feeding
The experiment was conducted on 9 Złotnicka White barrows with an average initial body weight of about 25 kg. The animals were fi tted with simple T-cannulae in the terminal ileum according to the method of van Leeuwen et al. (1991).
The animals were fed in a double Latin square design (5 × 5 and 4 × 4) with semisynthetic mixtures in which the only source of protein was legume seeds or soyabean oilmeal (SBM) (control group). Digestibility was determined by the indicator method using Cr 2 O 3 . The feed was given dry in two rations at 7.00 and 19.00.
The chemical composition of the protein feeds is given in Table 1, their amino acid composition in Table 2 and feed composition in Table 4.

Experimental procedure
The study period was composed of a nine-day adaptation period and a two-day experimental period. In the experimental period, samples of ileal contents (about 40 g) were collected every 4 h, frozen immediately and stored until chemical analysis. Fresh samples were used for determination of dry matter; the remaining components were assayed in freeze dried samples.

Chemical analyses
The basic chemical composition of feeds and ileal contents was determined according to AOAC (1996) procedures. Total protein was assayed by the Kjeldahl method using a Kjel Foss apparatus, crude fi bre, ADF and NDF, by the van Soest procedure on a Tecator apparatus. The amino acid content was determined using a type AAA-339 Mikrotechnia amino acid analyser following hydrolysis of samples in 6N HCL at 105 o C for 23 h. The tannin content was determined according to Kuhla and Ebmeier (1981), Cr 2 O 3 by the method of Saha and Gilbreath (1991) using a Carl Zeiss Jena AAS-3 spectrophotometer after wet mineralization of the samples.
Standardized ileal protein and amino acid digestibility coeffi cients were calculated on the basis of the equation and endogenous protein and amino acid losses given by Degussa (2001).

Statistical analysis
One-way analysis of variance was performed, the signifi cance of differences between varieties was calculated using Duncan's test. The statistical analysis was performed using Statgraphics Plus ver. 5.1 software.

RESULTS
The apparent ileal digestibility of protein and most amino acids was significantly (P<0.05) lower than that of soyabean oilmeal (SBM) ( Table 5) and did not signifi cantly differ (P>0.05) among varieties. The average digestibility of lysine, threonine, and methionine was 11, 13.6 and 10 percentage points, respectively,  lower than in SBM. No signifi cant differences were found, however, between SBM and: fi eld bean cv. Nadwiślański in the digestibility of arginine, histidine, and tyrosine; cv. Kodam and Martin, in the digestibility of arginine and proline; cv. Optimal in the digestibility of alanine, glycine, and proline. The digestibility coeffi cients of proline and methionine were lowest. The ileal digestibility of protein and some amino acids of peas were signifi cantly lower (P<0.05) than in SBM (Table 6). The ileal digestibility coeffi cients of pea protein ranged from 75.2 to 80.5%, with signifi cant differences (P<0.05) between the white-fl owered variety, Rola, and the coloured-fl owered Idol. Differences in the digestibility of lysine of the cultivars Piast, Rola, Wiato, of arginine, glycine and cystine (Piast and Rola), histidine (Rola and Wiato), threonine, phenylalanine, alanine, aspartic acid, glutamic acid (Rola) did not differ signifi cantly (P>0.05) from the digestibility of these amino acids in soyabean meal.

DISCUSSION
The ileal digestibility of fi eld beans and of peas was in agreement with the results reported by Grala et al. (1993), Degussa (2001), Mariscal-Landin et al. (2002) and Selgado et al. (2002a,b), among others, whereas the digestibility of lysine, methionine and threonine in our experiment was several percentage points lower. The differences in the digestibility of amino acids may have resulted from the use of different varieties of legumes (different chemical composition) and the age of the experimental animals.
The ileal digestibility of fi eld bean and pea protein was signifi cantly lower than in soyabean meal, which may have been an effect of the antinutrients contained in them (Abrahamsson et al., 1993;Grala et al., 1993;Jansman, 1993;Selgado et al., 2002a). The tannin level (Tabel 1) was higher in the coloured-fl owered varieties of peas than in the white-fl owered ones. Despite this, only the digestibility of cv. Idol peas, which have coloured fl owers, was signifi cantly lower than that of the while-fl owered varieties. The tannin level of fi eld beans did not differ among varieties and was relatively high. The low digestibility of proline found also in peas may have been related to the considerable excretion of a proline-rich protein in saliva. Jansman (1993) points to the negative infl uence of tannins on the availability of proline in rats fed high-tannin diets. This is explained by the ease with which the hydroxyl groups of phenolic compounds bind with groups in the relaxed structure of proline, resulting in a reduction of the antinutritive effect of tannins (Grala et al., 1993).
Fibre may also reduce digestibility as the result of adsorption of amino acids and binding of digestive enzymes (Sauer and Ozimek, 1986;Mosenthin et al., 1993;Selgado et al., 2002a). This component also has an exceptionally strong water-binding property, which may lead to accelerated digesta passage through the small intestine and reduction of feed digestibility (Grala et al., 1999). Some authors have found a negative correlation between both the fi bre content of feeds and the amount of protein bound with fi bre and the intestinal digestibility of protein and amino acids in pigs (Gdala et al., 1992;Fan et al., 2001). It should be stressed that the fi bre content of the control feed was lower than in the experimental feeds (Table 4). A reason for the lower digestibility of peas may have also been the presence of considerable amounts of non-starch polysaccharides (NSP) having a low digestibility (hemicellulose, cellulose, pectins). Jondreville et al. (1992) and Le Guen et al. (1995a,b) point out that starch in peas is resistant to amylolytic enzymes.
The low digestibility of methionine found both in this study and by other authors (Mosenthin et al., 1993;Frankiewicz, 1999), may be related to the relatively low content of this amino acid in fi eld beans and peas, and consequently, in the experimental feeds.
No differences were also found between soyabean meal and the pea and fi eld bean cultivars in the digestibility of those amino acids whose concentration in the experimental feeds (or seeds) was similar or higher than in the control feed. No statistically signifi cant differences were found among varieties of fi eld beans in the digestibility of protein and amino acids, which results from the similar chemical composition and lack of differences in antinutrient contents.
Despite the differences in the tannin contents among the white-and colouredfl owered pea varieties, signifi cant differences in apparent protein and lysine digestibility occurred only between one white-fl owered (Rola) and coloured-fl owered (Idol) variety of peas.
The determined ileal digestibility coeffi cients are coeffi cients of apparent digestibility since they do not factor in endogenous nitrogenous compounds. Many factors infl uence nutrient digestibility of various feeds and the amount of endogenous nitrogenous compounds. These include the protein and fi bre levels of the ration, its antinutrient content, and the method of determination (Grala et al., 1998). Sauer and Ozimek (1986) consider however, that the nutritional value of feeds determined on the basis of apparent digestibility coeffi cients is suffi cient in pig feeding practice.
The calculated standardized digestibility coeffi cients of the feeds studied by us are similar to those given by Degussa (2001) and differentiate to a smaller degree the legume varieties in terms of protein and amino acid digestibility than do coeffi cients of apparent digestibility.

CONCLUSIONS
The apparent ileal digestibility of protein and amino acids of fi eld beans and of peas are lower that those of soyabean oilmeal and do not differ among fi eld bean cultivars. Differences in digestibility of pea varieties are not unequivocally dependent on their tannin contents.