Variation in seed protein digestion of different pea (Pisum sativum L.) genotypes by cecectomized broiler chickens: 2. Relation between in vivo protein digestibility and pea seed characteristics, and identification of resistant pea polypeptides

Available online at www.sciencedirect.com

Livestock Science 113 (2008) 262 – 273 www.elsevier.com/locate/livsci

Variation in seed protein digestion of different pea (Pisum sativum L.) genotypes by cecectomized broiler chickens: 2. Relation between in vivo protein digestibility and pea seed characteristics, and identification of resistant pea polypeptides I. Gabriel a,⁎, L. Quillien c,e , F. Cassecuelle c , P. Marget c , H. Juin b , M. Lessire a , B. Sève d , G. Duc c , J. Burstin c a

INRA, UR83 Recherches Avicoles, F-37380 Nouzilly, France INRA, UE1206 Elevage Alternatif et Santé des Monogastriques, Le Magneraud, BP52, F-17700 Saint-Pierre-d’Amilly, Surgères, France c INRA, Unité de recherche Génétique et Ecophysiologie des Légumineuses, Domaine d’Époisses, F-21110 Bretenières, France d INRA, UMR1079, Système d’Elevage Nutrition Animale et Humaine, F-35590 Saint Gilles, France INRA, Unité de recherche sur les Biopolymères, Interactions, Assemblages, Rue de la Géraudière, BP71627, F-44316 Nantes cedex 3, France b

e

Received 12 October 2006; received in revised form 2 April 2007; accepted 4 April 2007

Abstract Eight pea genotypes characterized for their major protein fractions were used to investigate the effect of seed protein composition variability on protein digestibility in poultry. These genotypes of various pea types, were also variable in other seed components. They showed variations in their carbohydrate (insoluble fibre compounds, soluble fibre, soluble carbohydrates) and trypsin inhibitor (TI) contents. To exclude the effect of tannins and of particle size, the seeds were dehulled and micro-ground. They were incorporated as the only protein source in isoproteinaceous diets with similar metabolisable energy content and fed to cecectomized chickens. The average amino acid digestibility (apparent and true) and endogenous amino acid excretion were related with pea diet characteristics (protein composition, carbohydrate composition and TI activity). This allowed to precise which of the diet characteristics affect protein digestibility and endogenous excretion. Average apparent digestibility of amino acids was negatively correlated with insoluble fibre components (R = −0.71 to −0.72; p b 0.05) and TI activity (R = − 0.93; p b 0.001). Average endogenous losses of amino acids were positively correlated with soluble carbohydrate content (R = 0.77; p b 0.05) and TI activity (R = 0.84; p b 0.01). Average true digestibility of amino acids was positively correlated with the PA2 albumin level (R = 0.71; p b 0.05), and negatively with the legumin level (R = −0.72; p b 0.05). Resistant peptides extracted from chicken excreta were analysed through electrophoresis and identified by immunodetection. Intensity of detected resistant peptides showed variation among genotypes. However, for the 8 pea genotypes, the pea proteins, which persisted at the end of the digestive tract, were mainly albumin PA1b and lectin. Other minor peptides were also detected: vicilin, albumin PA2 and legumin peptides which migrated at the same level as β-subunits. © 2007 Elsevier B.V. All rights reserved. Keywords: Broiler chickens; Pisum sativum; Pea seed characteristics; Trypsin inhibitor; Protein digestibility; Resistant pea proteins

⁎ Corresponding author. Tel.: +33 2 47 42 76 47; fax: +33 2 47 42 77 78. E-mail address: [email protected] (I. Gabriel). 1871-1413/$ - see front matter © 2007 Elsevier B.V. All rights reserved. doi:10.1016/j.livsci.2007.04.005

I. Gabriel et al. / Livestock Science 113 (2008) 262–273

1. Introduction For feed manufacturers, it is important that raw material have high and reliable protein nutritional values. Variability in seed protein digestibility contributes to the limitation of inclusion of meal obtained from seeds in feeds. In peas, variations in protein digestibility have been observed both in poultry and in pigs (Conan and Carré, 1989; Igbasan et al., 1997; Grosjean et al., 1998, 1999; Gabriel et al., in press). There are several causes for this variability. Most seeds used in feeds are variable in composition. Some factors are known to have a negative effect on protein digestibility such as tannins (Grosjean et al., 1999), carbohydrates (Longstaff and McNab, 1991; Gdala et al., 1997), trypsin inhibitors (TI) (Huisman and Jansman, 1991) or particle size (Crévieu et al., 1997a). Variations in protein composition and structure are also involved. The composition of pea seed protein presents variability due to both genotype and environment (Burstin and Duc, 2006). Pea seed proteins have been classified in three main groups according to their solubility: water soluble albumins represent around 20–25% of seed proteins, salt-soluble globulins represent 55–65% and insoluble proteins represent around 15–20% (Guéguen, 1991). The albumin fraction contains very diverse proteins: the major ones, 11 kDa PA1 and 48–53 kDa PA2, and also lipoxygenase, glycosidases, TI and lectins. The globulin fraction includes two major storage protein groups encoded by multigene families and differing in their sedimentation coefficients: vicilins and convicilins (7S) and legumins (11S). Vicilins and convicilins are trimeres of 150 to 180 and 210 to 280 kDa respectively, composed of heterogeneous and differently matured polypeptides. Legumins are compact hexameres of 350 to 400 kDa associating acidic α-polypeptides and basic β-polypeptides. These proteins show various resistance to hydrolysis (Spencer et al., 1988; Crévieu et al., 1997b; Le Gall et al., 2005) as well as different amino acid composition, particularly with regard to sulphur-rich amino-acids which are rare in vicilins and high in albumins PA1 and PA2 (Gwiazda et al., 1980). In a previous study, we assessed the variability of endogenous amino acid excretion as well as true amino acid digestibility using the 15N dilution method, with seed meals from eight pea genotypes fed to cecectomized growing chickens (Gabriel et al., in press). Pea genotypes were chosen for their difference in protein composition. Thus they were of various types of peas (feed peas, garden peas and fodder peas) and therefore they also differed in their carbohydrate content and TI

263

activity. To limit variation due to known factors such as tannins and particle size, seeds were dehulled and micro-ground. The objective of the present study was to identify factors involved in the variability of amino acid digestibility and endogenous losses. We analysed the chemical composition, in particular protein composition, of each pea seed meal, in order to relate this composition to in vivo digestibility and endogenous losses. Moreover, resistant pea proteins extracted from chicken excreta were separated through sodium-dodecyl-sulfate polyacrylamide gel electrophoresis (SDSPAGE) and identified by immunodetection. 2. Materials and methods 2.1. Characterization of seed composition for the different pea genotypes Eight pea lines (cv ‘Ballet’, cv ‘Caméor’, China, E344, cv ‘Finette’, cv ‘Préclamex’, cv ‘Sommette’, VavD265) were field grown at INRA Dijon and harvested in the summer 2002. These genotypes have previously been described by Baranger et al. (2004). As two genotypes (VavD265 and E344) contained tannins, seeds from the 8 pea genotypes were dehulled (CREOL, Pessac, France). For each genotype, a seed sample was ground through a 0.5-mm screen and assayed for total nitrogen by the Kjeldahl procedure (ISO, 1997), and a factor of 6.25 was used to calculate the crude protein content. The pea samples were also assayed for amino acids (AFNOR, 1998a,b), starch (European Directive, 1999), soluble and insoluble fibre (AOAC, 1995), insoluble cell walls (AFNOR, 1998c), cellulose (AFNOR, 1993), soluble carbohydrates (European Directive, 1971), ash (AFNOR, 1977), tannins (INZO, 1999), TI (method of Kakade et al. (1974) modified by Valdebouze et al. (1980)) and dry matter (AFNOR, 1982). The protein composition was evaluated by fast protein liquid chromatography (FPLC) as described by Baniel et al. (1998). The relative quantity of each protein fraction was estimated by the ratio of the area below its corresponding peak on the chromatograms to the total area below the chromatogram curve. 2.2. Experimental design Peas were used as the only protein source in experimental diets. The diets were formulated to be isoproteinaceous (19.5%) and to contain similar metabolisable energy (2950 to 3030 kcal/kg). Carbohydrates and antinutritional factors content in the diets were

264

I. Gabriel et al. / Livestock Science 113 (2008) 262–273

calculated from their content measured in the pea samples. The animal experimental design was described in Gabriel et al. (in press). In summary, 48 cecectomized broiler chickens housed in individual cages (6 blocks × 8 birds) were fed the 8 experimental pea diets. Each chicken was fed two different diets during two successive periods (period 1: 18 to 25 days and period 2: 25 to 32 days). Thus for each diet, 12 birds were used, 6 birds in period 1 and 6 birds in period 2. Each experimental period consisted of 4 days adaptation followed by 4 days of measurements consisting of a digestion balance followed by a test meal of the labelled diet containing 15N-labelled experimental peas and chromic oxide. After the birds were fed ad libitum for 1 h (25 and 32 days) with these test meals, and feed withdrawal, excreta were collected hourly for 9 h, stored at − 20 °C, freeze-dried, weighed and micro-ground. The green coloured excreta samples (containing chromic oxide) were taken hourly and were analysed for total N and 15N enrichment. For each bird the samples were pooled, excluding those from early or late collections with the lowest enrichment values, and assayed for total N, 10 individual amino acids, their 15N enrichments, chromic oxide and uric acid. Samples from these excreta pools were used for SDS-PAGE and immunoblotting.

electroblotted on nitrocellulose membrane (pore diameter 0.2 μm) according to the method of Towbin et al. (1979). Sera diluted in PBS buffer (10 mM Na2HPO4, 150 mM NaCl, pH 7.2) were used to reveal bands corresponding to specific pea proteins (legumin, vicilin, PA2, lectin, PA1a, PA1b). The final revelation was obtained after incubation with goat-anti rabbit IgG conjugated to horseradish peroxidase (1/3000) and addition of 4-chloro-1-naphtol and hydrogen peroxide in methanol:PBS. 2.4. Statistical analysis Statistical analyses were performed using the Statview® software programme version 5 (Abacus Concepts, Berkeley, CA, USA). Correlation coefficients were calculated to determine the relationship among pea characteristics, on one hand, and between pea diet characteristics and average apparent amino acid digestibilities measured over the balance period, amino acid endogenous losses and true digestibility measured after the 15N-labelled test meal, on the other hand. For average amino acid endogenous losses and true digestibility, 9 among the 10 assayed individual amino acids were used since tyrosine enrichment could not be determined in all samples. 3. Results

2.3. SDS-PAGE and immunoblotting 3.1. Pea seed characteristics Proteins were extracted from micro-ground seeds of a control pea genotype (cv. ‘Frilène’) and from individual excreta pools, in a Tris–HCl (pH 8.5) buffer containing 1% SDS, at 100 °C for 20 min. Extracts were centrifuged at 12 000×g for 10 min and filtrated through a 0.2 μm filter (Millex-GV, Millipore, St Quentin en Yvelines, France). Extracts were then diluted in loading buffer (63 mM Tris–HCl, pH 6.8, SDS 2%, β-mercaptoethanol 2%, glycerol 25%, bromophenol blue 0.01%) to 0.82 μg protein per μl according to their dietary nitrogen content, as determined with the 15 N enrichment. Extracts were stored at − 20 °C until further analysis. For each sample, 10 μl were loaded on 10–20% polyacrylamide gradient gels (161–1460, Biorad, Marnes la Coquette, France) and SDS-PAGE was performed at 35 mA during 4 to 5 h, using a vertical gel system (Biorad). Proteins were fixed and stained with Coomassie Blue G (Biosafe Coomassie G, Biorad). All excreta pools were analysed and one representative pool, corresponding to one bird, was selected per period and per pea diet according to its protein pattern, for immunoblotting. After SDS-PAGE, proteins were

The seed protein content of the 8 genotypes studied ranged from 24.0% to 32.4% of DM in dehulled seed meal (Table 1). Starch content varied between 45.5% and 54.2% of DM, and was negatively correlated with protein content (R = − 0.87; p b 0.01). The insoluble fibre components measured by insoluble cell wall, insoluble fibre and cellulose contents varied between 7.1% and 9.5%, 8.9% and 11.5%, and 2.3% and 4.1% respectively. These insoluble fibre components were highly correlated. Thus positive correlations were observed between insoluble cell walls and insoluble fibre (R = 0.83; p = 0.01), between insoluble cell walls and cellulose (R = 0.85; p b 0.01), and between insoluble fibre and cellulose (R = 0.86; p b 0.01). Soluble fibre was not detected in half of the genotypes, and was present in very small amounts, from 0.4% to 0.6%, in the other genotypes. Soluble carbohydrates (sucrose and αgalactosides) varied between 4.3% and 6.8%. Ash varied between 3.2% and 4.1%. Traces of tannins were found in the two coloured pea genotypes

I. Gabriel et al. / Livestock Science 113 (2008) 262–273

265

Table 1 Chemical composition of dehulled seeds in the 8 pea genotypes (% of dry matter) Genotype

China Finette Sommette Caméor VavD265 Préclamex E 344 Ballet

N × 6.25 %

32.0 29.2 24.0 29.2 32.4 29.3 30.4 25.3

Starch %

Fibre

45.5 50.4 54.2 50.4 45.8 49.4 45.5 50.2

Soluble %

Insoluble %

Total %

Insoluble cell wall %

0.57 0.46 0.00 0.00 0.46 0.00 0.00 0.35

11.1 10.0 8.9 9.8 11.3 9.4 10.6 11.5

11.7 10.4 8.9 9.8 11.7 9.4 10.6 11.9

8.5 8.0 7.2 7.9 8.8 7.1 9.5 9.0

VavD265 and E344. TI activity varied between 1.9 and 6.8 TIU/mg. As peas were incorporated in different amounts in the experimental diets, contents of carbohydrate compounds, tannins and TI, were calculated for each diet (Table 2). Insoluble fibre, insoluble cell wall and cellulose contents varied between 6.1% and 8.8%, 4.6% and 6.9% and 1.5% and 3.1% respectively. Soluble fibre was present in very small amounts, 0.0% to 0.3%. Soluble carbohydrates amounts varied between 2.6% and 5.1%. TI content in pea diets varied between 1.3 and 4.9 TIU/mg. Seed protein composition was analysed by anion exchange FPLC. Chromatograms obtained for the 8 genotypes differed both in terms of peak height and peak position (Fig. 1), suggesting quantitative as well as qualitative variations in seed protein composition among the 8 pea genotypes. Relative quantities ranged from 10% to 14% for PA1, from 22% to 29% for PA2, from 15% to 20% for vicilin and 23% to 36% for legumin (Table 3). A negative correlation was observed between relative quantity of PA2 and legumin (R = − 0.94; p b 0.01). It should be noted that these figures

Cellulose %

3.1 3.3 2.6 2.6 3.6 2.3 3.4 4.1

Soluble carbohydrate %

Ash %

5.8 5.0 4.9 4.6 4.3 5.0 5.1 6.8

3.7 3.3 3.2 3.7 3.6 3.5 4.1 3.2

Tannin %

Trypsin inhibitor TIU/mg

0 0 0 0 0.16 0 0.02 0

6.8 2.3 3.2 1.9 3.2 3.2 5.8 6.3

only allow to compare the different genotypes within each protein fraction, but not to compare among fractions. 3.2. Pea seed characteristics and digestibility parameters An effect of pea genotype was observed in the apparent digestibility for all the assayed amino acids except for methionine, and in endogenous amino acid losses as well as true digestibility for the 10 assayed amino acids (Gabriel et al., in press). The mean apparent digestibility varied between 79.5% and 86.3% and was negatively correlated with insoluble fibre components (R = − 0.71 to − 0.72; p b 0.05) and TI contents (R = − 0.93; p b 0.001) (Table 4). The mean endogenous excretion of the 9 amino acids, varied between 3.6% and 5.4% of the ingested amino acids, and was positively correlated with soluble carbohydrate (R = 0.77; p b 0.05) and TI content (R = 0.84; p b 0.01) (Table 4). For the individual amino acids, high soluble carbohydrate content, for 6 amino acids out of the 10, and TI activity, for 7 amino acids out of the 10, were significantly

Table 2 Carbohydrate and antinutritional factors in the experimental pea diets (% of crude matter) a Genotype

China Finette Sommette Caméor VavD265 Préclamex E344 Ballet a

Fibre Soluble %

Insoluble %

Total %

Insoluble cell wall %

0.34 0.31 0.00 0.00 0.28 0.00 0.00 0.27

6.63 6.57 7.13 6.49 6.69 6.14 6.66 8.75

6.97 6.87 7.13 6.49 6.97 6.14 6.66 9.02

5.09 5.28 5.76 5.24 5.20 4.63 6.02 6.86

Calcutated from Table 1 values.

Cellulose %

1.84 2.14 2.07 1.74 2.16 1.49 2.17 3.10

Soluble carbohydrate %

Tannin %

3.49 3.28 3.89 3.06 2.55 3.26 3.19 5.13

0.00 0.00 0.00 0.00 0.10 0.00 0.01 0.00

Trypsin inhibitor TIU/mg 4.13 1.56 2.59 1.28 1.93 2.14 3.73 4.91

266

I. Gabriel et al. / Livestock Science 113 (2008) 262–273

Fig. 1. Anion exchange fast protein liquid chromatograms of the 8 pea genotypes. The different peaks correspond to the different seed protein fraction: albumin PA2 and PA1, vicilin, and legumin.

associated with higher endogenous excretion. Endogenous excretion of leucine was positively correlated with one fibre component measurement, insoluble fibre (R = 0.75; p b 0.05). Moreover, endogenous excretion of lysine was negatively correlated with the PA2 level (R = − 0.73; p b 0.05). The average true digestibility varied between genotypes from 84.4% and 90.2%, and was correlated with two major protein fractions (Table 5). It was positively correlated with the PA2 level (R = 0.71; p b 0.05), and negatively with the legumin level (R = − 0.72; p b 0.05). The true digestibility of 2 individual amino acids out of 10 was also positively correlated with the PA2 level (R = 0.71 to 0.72; p b 0.05), and high legumin levels were associated with lower true digestibility of 3 individual amino acids out of 10 (R = − 0.71 to −0.76; p b 0.05). True digestibility of aspartic acid was negatively correlated with TI activity (R = − 0.76; p b 0.05) (Table 5). 3.3. Resistant pea polypeptides Analyses were performed on excreta samples collected after the labelled diet. Firstly, one representative excreta sample was selected for each pea diet and each period, by electrophoresis protein pattern analysis. All selected samples were run on the same SDS polyacrylamide gradient gel (Fig. 2). Secondly, pea proteins were identified by immunoblotting with specific antibodies for legumin, vicilin, albumin PA2, PA1a, PA1b, and lectin. Results obtained for period 1 are presented in Fig. 3. Different degrees of degradation were observed for the different polypeptides, and for the different pea genotypes. The SDS-PAGE protein profiles obtained for excreta were very different from those obtained for proteins from the meal. The intensity of high molecular weight

bands was lower in the excreta than in the meal, while the intensity of low molecular weight bands was higher, due to protein degradation during the digestion process in birds. Using the anti-legumin antibody, acidic αpolypeptides of legumin were not detectable in the excreta (Fig. 3A). A slight band migrating at the same level as the basic β-polypeptides of legumin, for which the antibody presented a very low affinity, was detected in six of the excreta samples. The lack of detection in two of the samples may be due to the low affinity of the antibody used for these polypeptides. On Coomassie Blue stained gels, the intensity of the band corresponding to these polypeptides was low in the excreta as well as in the pea meal (Fig. 2). Using specific antibodies against vicilin, no band was observed in the excreta at the level of convicilin (Fig. 3B). Corresponding to a low intensity band in the excreta of the Coomassie Blue stained gel (Fig. 2), an intense band was observed on the immunoblot at 50 kDa in the excreta as well as in the pea meal (Fig. 3B) due to the high affinity of the antibody for this polypeptide. Several other peptides, of molecular weight around 35, 30 and 22 kDa, were detected on the immunoblot with a low intensity in the excreta (Fig. 3B). However, on the Coomassie Blue gel, the intensity of the band containing the 22 kDa peptide was higher in the excreta than in the meal, and showed quantitative variations between genotypes (Fig. 2). Using antibodies against albumin PA2, a faint band (26 kDa) was detected in half of the excreta samples (Fig. 3C). On Coomassie Blue gel, its intensity was similar in the excreta and in the meal (Fig. 2). A lower polypeptide was also detected (17 kDa) on the PA2 immunoblot in some samples (Fig. 3C). The antibody raised against albumin PA1a (6 kDa) could not detect any related peptide in the excreta (data not shown). The antibody raised against albumin PA1b (4 kDa) (Louis et al., 2004) was not able to detect this albumin in the meal Table 3 Seed protein composition in the 8 pea genotypes (%) a Genotype

PA1

PA2

Vicilin

Legumin

China Finette Sommette Caméor VavD265 Préclamex E 344 Ballet

10 11 14 11 13 12 14 14

22 27 28 24 29 27 23 27

17 19 16 15 17 20 16 17

36 24 23 33 24 24 34 24

a

Evaluation of protein composition by the ratio of the surface below the peak of each protein fraction obtained by FPLC chromatography on the total surface below the chromatogram curve (Baniel et al., 1998).

I. Gabriel et al. / Livestock Science 113 (2008) 262–273

267

Table 4 Correlations (R) between apparent amino acid digestibilities or endogenous losses and diet characteristics

PA1 PA2 Vicilin Legumin Insoluble fibre Insoluble cell wall Cellulose Soluble carbohydrate Trypsin inhibitor

Apparent digestibility

Endogen

Aa

Aa a

Ala

Asp

Ile

Leu

Lys

Phe

Pro

Thr

Tyr

Val

ns ns ns ns − 0.72⁎ − 0.71⁎ − 0.72⁎ ns − 0.93⁎⁎⁎

ns ns ns ns ns ns ns 0.77⁎ 0.84⁎⁎

ns ns ns ns ns ns ns 0.75⁎ 0.73⁎

ns ns ns ns ns ns ns 0.73⁎ ns

ns ns ns ns ns ns ns ns 0.87⁎⁎

ns ns ns ns 0.75⁎ ns ns 0.80⁎ 0.96⁎⁎⁎

ns − 0.73⁎ ns ns ns ns ns ns 0.76⁎

ns ns ns ns ns ns ns ns 0.75⁎

ns ns ns ns ns ns ns 0.75⁎ ns

ns ns ns ns ns ns ns 0.80⁎ ns

ns ns ns ns ns ns ns ns 0.73⁎

ns ns ns ns ns ns ns 0.72⁎ 0.92⁎⁎

ns: non significant, ⁎p b 0.05, ⁎⁎p b 0.01, ⁎⁎⁎p b 0.001. a Average of all amino acids, except tyrosine.

due to its low content (data not shown). However, this antibody detected a band of high intensity in all the excreta samples, with variation among genotypes (Fig. 3D). Since the variation of intensity between genotypes with antibody detection did not correspond to the variation of intensity between genotypes with Coomassie Blue G (Fig. 2), this band did probably not contain only the PA1b peptide. Due to their low content in peas, lectins were only faintly detectable in the meal by the anti-lectin (17 kDa) antibody, but they were highly detectable in the excreta (Fig. 3E). The similarity in variation of intensity between detection with specific antibody (Fig. 3E) and Coomassie Blue G (Fig. 2), suggested that this band was composed only of lectin. Several peptides (one band around 115 kDa, a band at 55 kDa, several bands in the range 25 to 35 kDa and under 20 kDa), were not detected in the pea meal, but were present in the excreta. They were not detected by any of the antibodies used in this study (Fig. 3). These peptides might be from endogenous proteins or hydro-

lysis products of pea proteins not detected by specific antibodies produced against the native protein (Le Gall et al., 2005). 4. Discussion The aim of the present study was to investigate the effect of seed protein composition variability on protein digestibility in poultry. For this, we used 8 pea genotypes that had previously been shown to differ in their seed protein composition. These genotypes including feed peas, garden peas and fodder peas, were also variable in other seed characteristics. 4.1. Pea characteristics We minimized when possible the effect of factors already identified as affecting protein digestibility such as tannins (Grosjean et al., 1999), particle size (Crévieu et al., 1997a) and TI (Huisman and Jansman,

Table 5 Correlations (R) between true amino acid digestibilities and diet characteristics True digestibility Aa PA1 PA2 Vicilin Legumin Insoluble fibre Insoluble cell wall Cellulose Soluble carbohydrate Trypsin inhibitor

a

ns 0.71⁎ ns − 0.72⁎ ns ns ns ns ns

Ala

Asp

Ile

Leu

Lys

Phe

Pro

Thr

Tyr

Val

ns ns ns ns ns ns ns ns ns

ns ns ns ns ns ns ns ns − 0.76⁎

ns 0.72⁎ ns −0.72⁎ ns ns ns ns ns

ns 0.71⁎ ns −0.71⁎ ns ns ns ns ns

ns ns ns − 0.76⁎ ns ns ns ns ns

ns ns ns ns ns ns ns ns ns

ns ns ns ns ns ns ns ns ns

ns ns ns ns ns ns ns ns ns

ns ns ns ns ns ns ns ns ns

ns ns ns ns ns ns ns ns ns

ns: non significant, ⁎p b 0.05, ⁎⁎p b 0.01, ⁎⁎⁎p b 0.001. a Average of all amino acids, except tyrosine.

268

I. Gabriel et al. / Livestock Science 113 (2008) 262–273

Fig. 2. SDS-polyacrylamide gel electrophoresis in reducing conditions (gradient 10–20%) of proteins extracted from excreta from chickens fed 8 different diets containing pea from the 8 pea genotypes after the test meal of the 15N-labelled diet during the two successive periods (Period 1: 25 days; Period 2: 32 days). Mr: molecular weight markers; 1–8: excreta of birds fed on diets obtained from the eight genotypes (1: China; 2: cv ‘Finette’; 3: cv ‘Sommette’; 4: cv ‘Caméor’; 5: VavD265; 6: cv ‘Préclamex’; 7: E344; 8: cv ‘Ballet’) for the periods 1 and 2; T: control pea (cv ‘Frilène’). ConVic: Convicilin; Vic: vicilin; PA2: albumin PA2; Legα: α-subunit of legumin; Legβ: β-subunit of legumin; Lec: lectin; PA1b: albumin PA1b.

1991). The tannin content after dehulling the seeds was very low, from 0% to 0.16%. Traces found in coloured pea genotypes were probably due to incomplete dehulling. The seeds were micro-ground to break the cotyledonary cell walls, which may act as a barrier between proteins and digestive enzymes (Crévieu et al., 1997a). The TI activity in the seeds ranged from low to intermediate values (1.9 to 6.8 TIU/mg), when compared to TI activity measured in a survey with 54 pea genotypes (1 to 14.6 TIU/mg, Bastianelli et al., 1998). Protein content and composition showed large variations. Protein content varied between 24.0% and 32.4% of DM in dehulled seeds, and peak surfaces corresponding to the major protein fractions varied between 10% and 14% of the total FPLC profile surface for PA1, between 22% and 29% for PA2, between 15% and 20% for vicilins and between 23% and 36% for legumins. Bastianelli et al. (1998) also reported, for smooth-seeded genotypes, a significant variability in the pea seed protein composition with protein content varying between 21.9% and 34.4%, and relative peak surfaces ranging from 29.0% to 56.9% for albumins, from 22.6% to 45.0% for vicilins and from 17.4% to 39.0% for legumins. We observed a negative correlation between PA2 and legumin contents, which is in accordance with the negative correlation between albumin and legumin content reported by Guéguen and Barbot (1988).

The carbohydrate composition (insoluble fibre compounds, soluble carbohydrates) was also analysed. Insoluble fibre components as measured by insoluble cell wall, insoluble fibre and cellulose contents varied between the genotypes. Insoluble cell walls which contain water insoluble non-starch polysaccharides, lignin and cell wall proteins (Carré and Brillouet, 1989) varied between 7.1% and 9.5%, which represented a similar range of variation than previously reported in pea (6.5% to 10.4%, Bastianelli et al., 1998). Insoluble fibre, which also includes water insoluble non-starch polysaccharides and lignin, varied between 8.9% and 11.5%. Although they do not include cell wall proteins, these values were higher than those found for insoluble cell walls, due to a lower efficiency of the insoluble fibre procedure in removing starch from pea samples (Carré and Brillouet, 1989). Cellulose, a component of insoluble fibre, ranged from 2.3% to 4.1%. The very low level of soluble fibre is characteristic for peas (Carré et al., 1984). The amount of soluble carbohydrates, which include mainly α-galactosides (53% to 83% of the soluble carbohydrates, VidalValverde et al., 2003) and sucrose, ranged from 4.3% to 6.8%, which is in the same range, from 3.9% to 8.2%, as reported by Bastianelli et al. (1998) for about 180 genotypes. These variations were maintained for all these compounds in the experimental pea isoproteinaceous diets, which contained different levels of pea seeds according to their protein content.

I. Gabriel et al. / Livestock Science 113 (2008) 262–273

269

Fig. 3. Identification of resistant peptides in excreta from chickens fed 8 different diets prepared with pea seeds from the 8 pea genotypes, by immunoblotting of SDS-PAGE. Proteins were extracted from a control pea meal and from excreta from chickens fed diets containing the 8 pea genotypes after the test meal of the 15N-labelled diet during the first period (25 days). Mr: molecular weight markers; 1–8: excreta of birds fed on diets obtained from the eight genotypes (1: China; 2: cv ‘Finette’; 3: cv ‘Sommette’; 4: cv ‘Caméor’; 5: VavD265; 6: cv ‘Préclamex’; 7: E344; 8: cv ‘Ballet’) for the period 1; T: control pea meal (cv ‘Frilène’). A: Legumin antibody: Legαβ (80): pro-protein of legumin; Legα(40): α-subunit of legumin; Legβ(20): β-subunit of legumin. B: Vicilin antibody: ConVic70: convicilin; Vic50, Vic35, Vic 33, Vic 22: vicilin constitutive polypeptides. C: Albumin PA2 antibody. D: Albumin PA1b antibody. E: Lectin antibody.

270

I. Gabriel et al. / Livestock Science 113 (2008) 262–273

4.2. Relation between pea diet characteristics and in vivo digestibility parameters, and resistant pea polypeptides Feeding different pea genotypes to the chicken (Conan and Carré, 1989; Igbasan et al., 1997; Grosjean et al., 1999; Gabriel et al., in press) or the pig (Grosjean et al., 1998; Hess et al., 1998) led to variations in protein digestibility and endogenous losses. In our study, we observed that whereas apparent amino acid digestibility and endogenous losses were mainly related to insoluble fibre, soluble carbohydrate contents or TI activity, true amino acid digestibility was mainly related to major protein fractions. The negative effect of insoluble fibre compounds on apparent protein digestibility was in accordance with previous results in the chicken (Carré and Leclercq, 1985; Longstaff and McNab, 1991) and in the pig (Sauer et al., 1977). It was associated with a positive effect of insoluble fibre on leucine endogenous loss. Whereas some fibre sources such as wood cellulose have no effect on endogenous excretion neither in poultry (Sibbald, 1980; Green, 1988) nor in pigs (Furuya and Kaji, 1992; Leterme et al., 1992), other insoluble fibre compounds have been reported to increase the endogenous losses in the chicken (Parsons et al., 1983; Raharjo and Farell, 1984) and in the pig (Green et al., 1987; Mariscal-Landin et al., 1995). For peas, a cotyledon fibre isolate has been shown to stimulate ileal endogenous protein losses in pigs (Leterme et al., 1996). Soluble carbohydrates had also a positive effect on endogenous losses of 6 among the 10 assayed amino acids. Although previous studies reported conflicting results from the effect of these compounds on protein digestibility (Irish et al., 1995; Gdala et al., 1997; Daveby et al., 1998), their effect on endogenous protein losses may be due to an increase of the bacterial biomass with fermentation of α-galactosides, as endogenous αgalactosidase is lacking in the chicken intestine (Carré et al., 1994). The negative effect of TI observed on apparent amino acid digestibility seemed to be due to the increase in endogenous losses. Although it has been suggested that pea TI are less effective inhibitors of trypsin than soyabean TI (Al Wesali et al., 1995), a study in chickens with near-isogenic lines of peas differing in TI activity (1.5–1.8 and 7.4–8.7 TIU/mg MS) showed a negative effect of these antinutritional factors on apparent digestibility (Wiseman et al., 2003). Previous studies showed no significant correlation between TI and protein digestibility in the chicken as well as in the

pig (Grosjean et al., 1998, 1999) using different pea genotypes. This was probably due to the interference of other seed components or particle size with greater effects, or due to estimation of protein digestibility in excreta of conventional animals and using nitrogen and not amino acids. The higher endogenous losses associated with the higher TI genotypes might be explained, in part, by the stimulation of pancreatic secretion due to formation of TI-proteolytic enzyme complexes in the intestinal lumen (Huisman and Jansman, 1991). The negative effect of TI on true aspartic acid digestibility may be due to the negative effect on protein hydrolysis, even though it was surprising that this effect was detected for only one amino acid. True amino acid digestibility was correlated with major protein fractions. A positive correlation was observed with the PA2 content and a negative correlation with the legumin content. The negative correlation between legumin and true digestibility was in agreement with the resistance of the β-subunits of legumin as shown by their immunochemical detection at the end of the intestine of the chicken, although the corresponding band was rather weak on the immunoblots, due to the low affinity of the antibody for these peptides. The resistance of these subunits of legumin was previously observed in the chicken (Crévieu et al., 1997b) and in the rumen (Spencer et al., 1988). Their resistance might be due to their highly ordered structure (Subirade et al., 1994) and high hydrophobicity (Lycett et al., 1984). The positive correlation of the PA2 fraction content with true digestibility and the faint detection of this protein on immunoblots suggested an efficient hydrolysis of this protein, even though its high cysteine content associated with the presence of disulfide bonds leading to a tight and globular structure (Gruen et al., 1987) would suggest a limited accessibility to digestive enzymes. In previous studies, this protein was detected at the terminal end of intestine of the chicken (Crévieu et al., 1997b) and the pig (Le Gall et al., 2005). However, it was not a major resistant polypeptide in the pig (Le Gall et al., 2005). No correlation was observed between vicilin content and endogenous losses or true digestibility, even though we identified some undigested vicilin peptides on the immunoblots. Vicilin has been shown to be highly susceptible to hydrolysis, with relatively low amounts of these peptides detected in the terminal ileum of the chicken (Crévieu et al., 1997b), the pig (Le Gall et al., 2005) and the calf (Lalles et al., 1998). PA1 had no effect on endogenous losses or true digestibility. Yet, PA1b was found to be highly resistant to hydrolysis in our study (Fig. 3) as previously reported in the pig (Le Gall et al., 2005). Albumin PA1b with 6

I. Gabriel et al. / Livestock Science 113 (2008) 262–273

cysteines per molecule forming 3 disulfide bonds (Higgins et al., 1986) may be difficult to hydrolyse. For these two proteins, vicilin and PA1, the lack of correlation with true digestibility may be due to the different resistance of the polypeptides quantified together in the corresponding peaks of FPLC (two subunits of albumin PA1, PA1a and PA1b, different vicilin peptides). Grouping together these different polypeptides may lead to the disappearance of their specific effects on digestibility. Moreover, other peptides which may act on digestibility, were not quantified in our study. Among the peptides which were shown to persist until the end of the digestive tract, β-subunit of lectin was among the most important. This peptide was also shown to be resistant in the intestine of the rat (Aubry and Boucrot, 1986) and the pig (Le Gall et al., 2005). Its resistance might be due to its compact structure and high content in β-sheet regions (Goldstein and Poretz, 1986), and to the binding of this protein to ligands (Pusztai et al., 1991). 5. Conclusion The results of this study showed that different factors may affect amino acid endogenous losses (TI, soluble carbohydrate) and true digestibility (major protein fraction) in young cecectomized broilers fed on dehulled and micro-ground pea. This emphasizes the need for a characterization of these two components of apparent digestibility, for a better understanding of the source of digestibility variation, and a more effective selection of improved genotypes. A finer characterization of pea protein composition might also help to determine more precisely the relationship between protein composition and true digestibility. Indeed, peptides belonging to the same protein class as measured by FPLC may have different susceptibility to hydrolysis, as α and β subunits of legumin, or PA1a and PA1b. This may explain the lack of correlation with vicilin and PA1 content. Moreover, other polypeptides not taken into account in our characterization of protein composition can have an effect on protein digestion. Acknowledgments We would like to thank R. Roy and her technical staff for the animal care (INRA, Surgères France), R. Maillard and G. Courteau assistant (Adisseo, France) for cecectomy, H. Houtin and C. Rond for the seed production (INRA, Bretenières, France), P. Carré (CREOL, France) for the pea dehulling, P. Colace (TECALIMAN, France) for the pea micro-grinding, J.

271

Gueguen (INRA, Nantes, France) and B. Desprez (Florimond-Desprez, France) for the helpful discussions during experiments, M. Legall for the scientific reading of the paper (INRA, Saint Gilles, France) and J. Williams (INRA, Nouzilly, France) for checking the English in the paper. The experiments described herein were funded by the French national programs Genoplante GOP-PeaC and GOP-PeaC2. References AFNOR, 1977. Aliment des animaux — Dosage des cendres brutes. AFNOR NF V 18–101. AFNOR, 1982. Aliment des animaux — Détermination de la teneur en eau. AFNOR ENR V 18–109. AFNOR, 1993. Produits agricoles et alimentaires — Détermination de la cellulose brute. AFNOR NF V 03–040. AFNOR, 1998a. Aliment des animaux — Dosage des acides aminés. AFNOR XP V 18–113. AFNOR, 1998b. Aliment des animaux — Dosage du tryptophane. AFNOR XP V 18–114. AFNOR, 1998c. Aliment des animaux — Détermination de la teneur en parois végétales insolubles dans l'eau. AFNOR XP V 18–111. Al Wesali, M., Lambert, N., Welham, T., Domoney, C., 1995. The influence of pea seed trypsin inhibitors on the in vitro digestibility of casein. J. Sci. Food Agric. 68, 431–437. AOAC, 1995. Total, soluble, and insoluble dietary fiber in foods. Enzymatic–gravimetric method, MES-TRIS buffer. AOAC Off. Method 991.43. Aubry, M., Boucrot, P., 1986. Etude comparée de la digestion des viciline, légumine et lectine radiomarquées de Pisum sativum chez le rat. Ann. Nutr. Metab. 30, 175–182. Baniel, A., Bertrand, D., Lelion, A., Guéguen, J., 1998. Variability in protein composition in pea seeds studied by FPLC and multidimensional analysis. Crop Sci. 38, 1568–1575. Baranger, A., Aubert, G., Arnau, G., Lainé, A.L., Deniot, G., Potier, J., Weinachter, C., Lejeune-Hénaut, I., Lallemand, J., Burstin, J., 2004. Genetic diversity within Pisum sativum using protein- and PCR-based markers. Theor. Appl. Genet. 108, 1309–1321. Bastianelli, D., Grosjean, F., Peyronnet, C., Duparque, M., Régnier, J. M., 1998. Feeding value of pea (Pisum sativum L.) 1. Chemical composition of different categories of pea. Anim. Sci. 67, 609–619. Burstin, J., Duc, G., 2006. The relationship between protein content and protein composition of pea seeds. Grain Legumes 44, 16–17. Carré, B., Leclercq, B., 1985. Digestion of polysaccharides, protein and lipids by adult cockerels fed on diets containing a pectic cellwall material from white lupin (Lupinus albus L.) cotyledon. Br. J. Nutr. 54, 669–680. Carré, B., Brillouet, J.M., 1989. Determination of water-insoluble cell walls in feeds: interlaboratory study. J. Assoc. Off. Anal. Chem. 72, 463–467. Carré, B., Plouzeau, M., Leclercq, B., 1984. Les glucides des principales matières premières utilisées en aviculture. Rev. Aliment. Anim. 381, 46–51. Carré, B., Bree, A., Gomez, J., 1994. α-Galactosides are poorly digested by germ-free chickens. Reprod. Nutr. Dev. 34, 617. Conan, L., Carré, B., 1989. Effect of autoclaving on metabolizable energy value of smooth pea seed (Pisum sativum) in growing chicks. Anim. Feed Sci. Technol. 26, 337–345.

272

I. Gabriel et al. / Livestock Science 113 (2008) 262–273

Crévieu, I., Carré, B., Chagneau, A.M., Guéguen, J., Melcion, J.P., 1997a. Effect of particle size of pea (Pisum sativum L.) flours on the digestion of their proteins in the digestive tract of broilers. J. Sci. Food Agric. 75, 217–226. Crévieu, I., Carré, B., Chagneau, A.M., Quillien, L., Guéguen, J., Bérot, S., 1997b. Identification of resistant pea (Pisum stivum L.) proteins in the digestive tract of chickens. J. Agric. Food Chem. 45, 1295–1300. Daveby, Y.D., Razdan, A., Aman, P., 1998. Effect of particle size and enzyme supplementation of diets based on dehulled peas on the nutritive value for broiler chickens. Anim. Feed Sci. Technol. 74, 229–239. European Directive, 1971. Première Directive de la Commission du 15 juin 1971 portant fixation de méthodes d'analyse communautaire pour le contrôle officiel des aliments des animaux (71/250/CEE). JO no L 155 du 12/7/1971, 13 pages. European Directive, 1999. Directive 1999/79/CE de la Commission du 27 juillet 1999 modifiant la troisième directive 72/199/CEE portant fixation de méthodes d'analyse communautaires pour le contrôle officiel des aliments des animaux. Journal officiel no L 209 du 07/ 08/1999, p. 0023–0027. Furuya, S., Kaji, Y., 1992. The effects of feed intake and purified cellulose on the endogenous ileal amino acid flow in growing pigs. Br. J. Nutr. 68, 463–672. Gabriel, I., Lessire, M., Juin, H., Burstin, J., Duc, G., Quillien, L., Thibault, J.N., Leconte, M., Hallouis, J.M., Ganier, P., Mézière, N., Sève, B., in press. Variation in seed protein digestion of different pea (Pisum sativum L.) genotypes by cecectomized broiler chickens: 1. Endogenous amino acid losses, true digestibility and in vitro hydrolysis of proteins. Livest. Sci. doi:10.1016/j.livsci.2007.04.002. Gdala, J., Jansman, A.J.M., Buraczewska, L., Huisman, J., Van Leeuwen, P., 1997. The influence of alpha-galactosidase supplementation on the ileal digestibility of lupin seed carbohydrates and dietary protein in young pigs. Anim. Feed Sci. Technol. 67, 115–125. Goldstein, I.J., Poretz, R.D., 1986. Isolation, physicochemical characterization, and carbohydrate-binding specificity of lectins. In: Liener, I.E., Sharon, N., Goldstein, I.J. (Eds.), The Lectins: Properties, Functions, and Applications in Biology and Medicine. Academic Press, Orlando, pp. 33–247. Green, S., 1988. Effect of dietary fibre and caecectomy on the excretion of endogenous amino acids from adult cockerels. Br. Poult. Sci. 29, 419–429. Green, S., Bertrand, S.L., Duron, M.J.C., Maillard, R.A., 1987. Digestibility of amino acids in maize, wheat and barley meals, measured in pigs with ileo-rectal anastomosis and isolation of the large intestine. J. Sci. Food Agric. 41, 29–43. Grosjean, F., Bastianelli, D., Bourdillon, A., Cerneau, P., Jondreville, C., Peyronnet, C., 1998. Feeding value of pea (Pisum sativum L.) Nutritional value in pig. Anim. Sci. 67, 621–625. Grosjean, F., Barrier-Guillot, B., Bastianelli, D., Rudeaux, F., Bourdillon, A., Peyronnet, C., 1999. Feeding value of three categories of pea (Pisum sativum L.) for poultry. Anim. Sci. 69, 591–599. Gruen, L.C., Guthrie, E., Blagrove, R.J., 1987. Structure of a major pea seed albumin: implication of a free sulphydryl group. J. Sci. Food Agric. 41, 167–178. Guéguen, J., 1991. Pea and fababean proteins. In: Hudson, B.J.F. (Ed.), Developments in Food Proteins. Elsevier, pp. 35–78. Guéguen, J., Barbot, J., 1988. Quantitative and qualitative variability of pea (Pisum sativum L.) protein composition. J. Sci. Food Agric. 42, 209–224.

Gwiazda, S., Schwenke, K.D., Rutkowski, A., 1980. Isolation and partial characterization of proteins from pea (Pisum sativum L.). Nahrung 24, 939–950. Hess, V., Thibault, J.N., Duc, G., Melcion, J.P., Van Eys, J., Sève, B., 1998. Influence de la variété et du microbroyage sur la digestibilité iléale de l'azote et des acides aminés du pois. Digestibilité réelle de l'azote et pertes endogènes spécifiques. Journ. Rech. Porc. Fr. 30, 223–229. Higgins, T.J.V., Chandler, P.M., Randall, P.J., Spencer, D., Beach, L. R., Blagrove, R.J., Kortt, A.A., Inglis, A.S., 1986. Gene structure, protein structure, and regulation of the synthesis of sulphur-rich protein in pea seeds. J. Biol. Chem. 261, 11124–11130. Huisman, J., Jansman, J.M., 1991. Dietary effects and some analytical aspects of antinutritional factors in peas (Pisum sativum), common beans (Phaseolus vulgaris) and soyabeans (Glycine max. L.) in monogastric farm animals. A literature reviews. Nutr. Abst. Rev., B 61, 901–921. Igbasan, F.A., Guenter, W., Slominski, B.A., 1997. Field peas: chemical composition and energy and amino acid availabilities for poultry. Can. J. Anim. Sci. 77, 293–300. INZO, 1999. Méthode interne par colorimétrie (ref N0841/01/99.1). Irish, G.G., Barbour, G.W., Classen, H.L., Tyler, R.T., Bedford, M.R., 1995. Removal of the α-galactosides of sucrose from soybean meal using either ethanol extraction or exogenous α-galactosidase and broiler performance. Poult. Sci. 74, 1484–1494. ISO, 1997. Aliment des animaux — Détermination de la teneur en azote et calcul de la teneur en protéines brutes — Méthode Kjeldahl. ISO 5983. Kakade, M.L., Rackis, J.J., Meghee, J.E., Puski, G., 1974. Determination of trypsin inhibitors activity of soy products: a collaborative analysis of an improved procedure. Cereal Chem. 51, 376–382. Lalles, J.P., Quillien, L., Toullec, R., 1998. Immunochemical identification of pea protein fragments escaping small intestinal digestion in the preruminant calf. In: Guéguen, J., Popineau, Y. (Eds.), Conference on Plant Proteins from European Crops. Food and Non-Food Applications. Springer-Verlag, Berlin, pp. 193–197. Le Gall, M., Quillien, L., Guéguen, J., Rogniaux, H., Sève, B., 2005. Identification of dietary and endogenous ileal protein losses in pigs by immunoblotting and mass spectrometry. J. Nutr. 135, 1215–1222. Leterme, P., Pirard, L., Thewis, A., 1992. A note on the effect of wood cellulose level in protein-free diets on the recovery and amino acid composition of endogenous protein collected from the ileum in pigs. Anim. Prod. 54, 163–165. Leterme, P., Thewis, A., van Leeuwen, P., Monmart, T., Huisman, J., 1996. Chemical composition of pea fibre isolates and their effect on the endogenous amino acid flow at the terminal ileum of the pig. J. Sci. Food Agric. 72, 127–134. Longstaff, M., McNab, J.M., 1991. The inhibitory effects of hull polysaccharides and tannins of field beans (Vicia faba L.) on the digestion of amino acids, starch and lipid and on digestive enzyme activities in young chicks. Br. J. Nutr. 65, 199–216. Louis, S., Delobel, B., Gressent, F., Rahioui, I., Quillien, L., Vallier, A., Rahbé, Y., 2004. Molecular and biological screening for insecttoxic seed albumins from four legume species. Plant Sci. 167, 705–714. Lycett, G.W., Croy, R.R.D., Shirsat, A.H., Boulter, D., 1984. The complete nucleotide sequence of a legumin gene from pea (Pisum sativum L.). Nucleic Acids Res. 12, 4493–4506. Mariscal-Landin, G., Sève, B., Colléaux, Y., Lebreton, Y., 1995. Endogenous amino nitrogen collected from pigs with end to end

I. Gabriel et al. / Livestock Science 113 (2008) 262–273 ileorectal anastomosis is affected by the method of estimation and altered by dietary fiber. J. Nutr. 125, 136–146. Parsons, C.M., Potter, L.M., Brown, R.D., 1983. Effects of dietary carbohydrate and of intestinal microflora on excretion of endogenous amino acids by poultry. Poult. Sci. 62, 483–489. Pusztai, A., Begbie, R., Grant, G., Ewen, S.W.B., Bardocz, S., 1991. Indirect effects of food antinutrients on protein digestibility and nutritional value of diets. In: Fuller, M.F. (Ed.), In vitro Digestion for Pigs and Poultry. CAB International, Wallingford, pp. 45–61. Raharjo, Y., Farell, D.J., 1984. A new biological method for determination amino acid digestibility in poultry feedstuffs using a simple cannula, and the influence of dietary fibre on endogenous amino acid output. Anim. Feed Sci. Technol. 12, 29–45. Sauer, W.C., Stothers, S.C., Parker, R.J., 1977. Apparent and true availabilities of amino acids in wheat and milling by-products for growing pigs. Can. J. Anim. Sci. 57, 775–784. Sibbald, I.R., 1980. The effects of dietary cellulose and sand on the combined metabolic plus endogenous energy and amino acid outputs of adult cockerels. Poult. Sci. 59, 836–844. Spencer, D., Higgins, T.J.V., Preer, M., Dove, H., Coombe, J.B., 1988. Monitoring the fate of dietary proteins in rumen fluid using gel electrophoresis. Br. J. Nutr. 60, 241–247.

273

Subirade, M., Guéguen, J., Pézolet, M., 1994. Conformational changes upon dissociation of a globular protein from pea: a Fourier transform infrared spectroscopy study. Biochim. Biophys. Acta 1205, 239–247. Towbin, H., Staehelin, T., Gordon, J., 1979. Electrophoretic transfer of proteins from polyacrylamide gels to nitrocellulose sheets. Proc. Natl. Acad. Sci. U. S. A. 76, 4350–4354. Valdebouze, P., Bergeron, P., Gaborit, T., Delort-Laval, J., 1980. Content and distribution of trypsin inhibitors and hemagglutinins in some legume seeds. Can. J. Plant Sci. 60, 695–701. Vidal-Valverde, C., Frias, J., Hernandez, A., Martin-Alvarez, P., Sierra, I., Rodriguez, C., Blasquez, I., Vicente, G., 2003. Assessment of nutritional compounds and antinutritional factors in pea (Pisum sativum) seeds. J. Sci. Food Agric. 83, 298–306. Wiseman, J., Al-Mazooqi, W., Welham, T., Domoney, C., 2003. The apparent ileal digestibility, determined with young broilers, of amino acids in near-isogenic lines of peas (Pisum sativum L.) differing in trypsin inhibitor activity. J. Sci. Food Agric. 83, 644–651.