Get 20M+ Full-Text Papers For Less Than $1.50/day. Subscribe now for You or Your Team.

Learn More →

Effects of Toasting Time on Digestive Hydrolysis of Soluble and Insoluble 00-Rapeseed Meal Proteins

Effects of Toasting Time on Digestive Hydrolysis of Soluble and Insoluble 00-Rapeseed Meal Proteins J Am Oil Chem Soc (2017) 94:619–630 DOI 10.1007/s11746-017-2960-8 ORIGINAL PAPER Effects of Toasting Time on Digestive Hydrolysis of Soluble and Insoluble 00‑Rapeseed Meal Proteins 1,2 2,3 4 Sergio Salazar‑Villanea · Erik M. A. M. Bruininx · Harry Gruppen · 5 6 2 Patrick Carré · Alain Quinsac · Antonius F. B. van der Poel Received: 7 October 2016 / Revised: 20 December 2016 / Accepted: 31 January 2017 / Published online: 11 February 2017 © The Author(s) 2017. This article is published with open access at Springerlink.com Abstract Thermal damage to proteins can reduce their in the hydrolysis rate of this fraction. Overall, longer toast- nutritional value. The effects of toasting time on the kinet- ing times increased the size of the peptides resulting after ics of hydrolysis, the resulting molecular weight distribu- hydrolysis of the RSM and the insoluble protein fraction. tion of 00-rapeseed meal (RSM) and the soluble and insolu- The hydrolysis kinetics of the soluble and insoluble protein ble protein fractions separated from the RSM were studied. fractions and the proportion of soluble:insoluble proteins in Hydrolysis was performed with pancreatic proteases to rep- the RSM explain the reduction in the rate of protein hydrol- resent in vitro protein digestibility. Increasing the toasting ysis observed in the RSM with increasing toasting time. time of RSM linearly decreased the rate of protein hydroly- sis of RSM and the insoluble protein fractions. The extent Keywords Hydrolysis rate · Maillard · Protein solubility · of hydrolysis was, on average, 44% higher for the insolu- Rapeseed meal · Toasting ble compared with the soluble protein fraction. In contrast, the rate of protein hydrolysis of the soluble protein fraction Abbreviations was 3–9-fold higher than that of the insoluble protein frac- DH Maximum degree of hydrolysis max tion. The rate of hydrolysis of the insoluble protein frac-DTT Dithiothreitol tion linearly decreased by more than 60% when compar- k Rate of h ydrolysis ing the untoasted to the 120 min toasted RSM. Increasing RSM Rapeseed meal the toasting time elicited the formation of Maillard reac- ε ε tion products (furosine, N -carboxymethyl-lysine and N - carboxyethyl-lysine) and disulfide bonds in the insoluble Introduction protein fraction, which is proposed to explain the reduction The production of defatted 00-rapeseed meal (RSM) involves toasting for the removal of remnant solvent after * Sergio Salazar-Villanea oil extraction, the inactivation of myrosinase and the degra- sergio.salazarvillanea@ucr.ac.cr dation of glucosinolates [1–3], the main antinutritional fac- Wageningen Livestock Research, Wageningen, The tors of RSM for monogastric animals. Increasing the toast- Netherlands ing time decreases protein solubility [1, 3] and the contents Animal Nutrition Group, Wageningen University of lysine [1, 3, 4] and reactive lysine [2, 3] of the resulting and Research, P.O. Box 338, 6700 AH Wageningen, The RSM. Reactive lysine can be considered as the protein- Netherlands bound lysine with a free ε-amino group [5]. Nutritionally, Agrifirm Innovation Center BV, Royal Dutch Agrifirm the changes in protein solubility, lysine and reactive lysine Group, Apeldoorn, The Netherlands contents have been noticed as reduction of in vitro crude Laboratory of Food Chemistry, Wageningen University protein digestibility [2], apparent ileal protein digestibility and Research, Wageningen, The Netherlands in broilers [6], standardized ileal protein digestibility in CREOL/OLEAD, Pessac, France pigs [2] and apparent total tract crude protein digestibility Terres Inovia, Paris, France in rats [1] and pigs [7]. 1 3 620 J Am Oil Chem Soc (2017) 94:619–630 Chemical and physical modifications can impair the porcine intestinal mucosa (50–100 units/g solid, EC 232- accessibility of proteins for enzymatic hydrolysis [8]. The 875-1) were obtained from Sigma-Aldrich (St. Louis, MO, proteases in the pancreatic secretions at the small intestine USA). The furosine, lysinoalanine and N -carboxymethyl- are highly specific. Trypsin cleaves peptide bonds involving lysine standards were obtained from PolyPeptide Labora- the carboxyl groups of either lysine or arginine [9], which tories (Strasbourg, France), whereas the rest of the stand- 13 15 are also the most susceptible amino acids to heat damage. ards ( C , N -lysine, lysine, lanthionine) were obtained 6 2 Lysine and arginine residues that are modified via Mail- from Sigma-Aldrich (Steinheim, Germany). lard reactions could, therefore, reduce enzyme accessibil- ity for proteolysis, finally reducing their standardized ileal digestibility [5]. Also, physical modifications, such as pro- Rapeseed Meals Preparation tein aggregation, might have a similar effect. Protein aggre- gates can be formed after hydrothermal processing and are An untoasted RSM was prepared by cold-pressing of the noticed as a reduction in protein solubility [10]. The forma- 00-rapeseeds, solvent-extraction and desolventization tion of aggregates can reduce protein accessibility for enzy- using indirect heat. The 00-rapeseeds were cold-pressed matic hydrolysis [11]. (La Mecanique Moderne MBU 75 type, Arras, France) In proteins that are accessible for enzymatic hydroly- at 250 kg/h with temperatures not exceeding 80 °C. sis, the gastric and intestinal enzymes cleave proteins into Solvent extraction was performed at temperatures not peptides, followed by cleavage into free small peptides higher than 55 °C on a belt extractor (B-1930, Desmet- and free amino acids by the brush border enzymes in the Ballestra, Zaventem, Belgium) at 230 l/h flow of hexane gut’s epithelium. The size of the peptides produced during and 160 kg/h flow of the rapeseed cake. Desolventization hydrolysis depends on the accessibility of the protein for with indirect heat (without direct steam) was performed the enzymes. Even at similar degrees of protein hydrolysis, in a desolventizer toaster (Schumacher type, Desmet- molecular size distribution of the resulting peptides can be Ballestra) for 60 min at temperatures of 90 ± 3 °C. different [12]. This might be the result of the selectivity of A batch of 150 kg of the untoasted RSM was toasted the enzymes or the accessibility of the proteins for cleavage with the use of direct steam (30 kg/h) for 120 min, with [12]. spot samples of 5 kg taken every 20 min (Fig. 1). Toast- The aim of the present study was to determine the effects ing of a separate batch of 150 kg of RSM was performed of toasting time during the production process of RSM on during the next day under the same conditions for replica- the kinetics of hydrolysis of proteins present in the com- tion. Temperatures during toasting on the first day ranged plete material, and in its soluble and insoluble fractions, from 107 to 112 °C and between 109 and 112 °C on the and on the resulting molecular size distribution of the pep- tides after hydrolysis. Previous results from our research group indicate that there is a highly significant positive cor - relation between nitrogen solubility and the rate of protein hydrolysis [3]. Therefore, we hypothesize that the rate of hydrolysis will be higher for the soluble protein fraction compared with the insoluble protein fraction, but that the rates at different toasting times will not vary within the soluble and the insoluble protein fractions. We also hypoth- esize that damage to the proteins due to prolonged toast- ing times will change the molecular size distribution of the peptides obtained after hydrolysis towards a larger size. Materials and Methods Materials Rapeseed meals were prepared from 00-rapeseed (Bras- sica napus) at the pilot plant of CREOL/OLEAD (Pes- sac, France). Trypsin (type IX-S, 13,000–20,000 BAEE units/mg protein, EC 232-650-8), chymotrypsin (type II, ≥40 units/mg protein, EC 232-671-2) and peptidase from Fig. 1 Schematic view of the design of the experiment 1 3 J Am Oil Chem Soc (2017) 94:619–630 621 second day. In total, 13 samples of RSM (1 untoasted RSM substrate to enzyme ratio was used for hydrolysis of RSM, and 12 toasted RSM) were obtained. Untoasted and toasted soluble and insoluble fractions. All hydrolyses were per- RSM were ground with a centrifugal mill (ZM200, Retsch, formed in duplicate. Haan, Germany) at 8000 rpm to pass a 1-mm sieve. The volume of alkali added was used for the calculation of the degree of hydrolysis according to Eq. 1, Fractionation of Proteins into Soluble and Insoluble Vb × Nb Fractions DH(%) = ×100 (1) α × mp × htot The water-soluble and water-insoluble fractions of the meals in which Vb is the volume added (ml), Nb is the normality were separated by suspending 25 g of the RSM in 250 ml of of the titration solution, α is the degree of dissociation of water. The pH of the suspension was adjusted to 8.0 with the α-NH group (in this case, 0.794 at 37 °C and pH 8.0), NaOH and magnetic stirring was applied for 20 min at room mp is the mass of protein (g) and htot is the total number of temperature. The soluble fraction was separated by centrifu- peptide bonds per gram of protein (7.8 meq/g) [14]. gation (11,900×g, 20 min, room temperature). Transparent Modelling of the degree of hydrolysis curves was based solutions were obtained, which were dialyzed extensively on second-order reaction kinetics, which is described in against a 0.01 M NaCl solution. After dialysis, the pH was Eq. 2, re-adjusted to 8.0 with NaOH. The soluble fractions were DH max kept at 4 °C and hydrolyzed within 24 h. The insoluble frac- DH(%) = DH − max (2) 1 + k × t × DH max tion was filtered through a nylon cloth and washed three times with 250 ml of water in order to remove soluble pro- in which DH is the maximum degree of hydrolysis (%), max teins and was freeze-dried prior to hydrolysis. k is the hydrolysis rate constant (/M×s) and t is the hydrol- ysis time. Fitting of this model was performed using the Analytical Methods MODEL procedure in SAS [15]. Nitrogen contents of the complete meals and the insoluble Nitrogen Solubility fractions were determined by combustion (AOAC 968.06, Thermo Quest NA 2100 Nitrogen and Protein Analyzer, Five hundred mg of RSM were suspended in 10 ml of Breda, The Netherlands). The N concentration of the sol- water and the suspension was adjusted to pH 8.0 using uble protein fractions was measured after an aliquot of 2 M NaOH. The suspension was stirred for 20 min at room 0.3 ml was oven-dried at 60 °C overnight. temperature and centrifuged (16,100×g, 15 min, room tem- perature). An aliquot (0.3 ml) was oven-dried overnight at Protein Hydrolysis 60 °C and analyzed for N content. Nitrogen solubility of the insoluble protein fractions were determined in water, In vitro protein hydrolysis was performed for 120 min 100 mM sodium phosphate buffer pH 7.5, and the same using a modification of the method described previ- buffer containing either 2% (w/v) SDS, 10 mM dithiothrei- ously [13]. Briefly, 10 ml of aqueous suspension contain- tol (DTT) or both 2% (w/v) SDS and 10 mM DTT. Briefly, ing 1 mg N/ml was adjusted to pH 8.0 in a titration unit 1.5 ml of these solutions were added to 75 mg of the insol- (719 S Titrino, Metrohm, Herisau, Switzerland) at 39 °C uble protein fractions. The suspensions were vortexed for using 0.1 M NaOH. At this point, 1 ml of enzyme solu- 20 s and mixed in a head-over-tail rotator for 20 min at tion containing 1.61 mg trypsin, 3.96 mg chymotrypsin 20 rpm. Following centrifugation (16,100×g, 15 min, room and 1.18 mg porcine intestinal peptidase was added and the temperature), 0.3 ml of the supernatant was oven-dried 120 min titration was initiated. This procedure was used for overnight at 60 °C and analyzed for N content. the complete RSM and the insoluble fractions. Concerning the soluble fraction, different concentrations Size Exclusion Chromatography (SEC) of N in the soluble fractions were obtained upon solubiliz- ing the RSM in water. Demineralized water was used for After hydrolysis, a sample of 1.5 ml was taken from the diluting these solutions to the concentration of the low- supernatant of the hydrolysate, heated at 99 °C for 15 min est one of the solutions (0.30 mg N/ml), to a volume of and centrifuged (16,100×g, 10 min, room temperature). In 10 ml. The solutions were adjusted to pH 8.0 in the titra- addition, samples from the water-soluble fraction of RSM tion unit using 0.05 M NaOH. At this point, 1 ml of an were analyzed after centrifugation (16,100×g, 15 min, enzyme solution containing 0.48 mg of trypsin, 1.19 mg of room temperature). The samples were analyzed in the chymotrypsin and 0.35 mg of porcine intestinal peptidase ÄKTA micro system (GE Healthcare, Uppsala, Sweden) was added and the 120 min titration initiated. The same using a Superdex 75 column (GE Healthcare) at a flow rate 1 3 622 J Am Oil Chem Soc (2017) 94:619–630 Table 1 Selected reaction monitoring conditions of 100 μl/min with UV detection at 220 nm. The eluent used was 10 mM sodium phosphate buffer of pH 7.0 con- a Compound Parent mass (Da) Fragment (m/z) taining 150 mM NaCl and 2% (w/v) SDS. The volume of Lysine 146 130 injection was 50 μl. A calibration curve of the elution vol- 13 15 C N -Lysine 154 137 umes in the column was obtained using threonine (119 Da), 6 2 N -Carboxymethyl-lysine 204 84, 130 proline-glycine-glycine (229 Da), vitamin B12 (1355 Da), Lanthionine 208 120 lysozyme (14,307 Da), β-lactoglobulin (18,400 Da), and N -Carboxyethyl-lysine 218 84, 130 ovalbumin (42,700 Da). Areas under the curve were inte- Lysinoalanine 233 128, 145 grated manually and the proportions of peptides based on Furosine 255 84, 130 AU response in each region (>10, 10–1.5 and <1.5 kDa) were calculated relative to the total area under the curve. Parent mass is defined as the molecular mass of the compounds before ionization Maillard Reaction Products, Crosslinked Compounds and Lysine N -carboxyethyl-lysine with concentrations of 0.01, 0.1, 1, The contents of furosine, N -[carboxymethyl]-lysine, lysino- 2.5, 5 and 10 mg/l of each standard was used to calculate alanine, lanthionine, N -[carboxyethyl]-lysine and lysine in the content of each compound. Compounds were quantified the RSM and the insoluble protein fractions were quanti- using the external standard calibration curve by plotting the fied by UHPLC-MS. The samples (10 mg) were hydrolyzed MS peak area divided by the MS peak area of the labelled with 1 ml of 6 M HCl during 24 h at 110 °C. The tubes were Lys, used as internal standard. Data were acquired and ana- dried under N flow and the dried material was re-suspended lyzed using XCalibur 2.2 software (Thermo Scientific). in 1 ml of UPLC-grade Milli-Q water, sonicated and centri- fuged (16,100×g, 3.5 min, room temperature). The superna- Statistical Analysis tant was diluted 50 times in eluent A that contained 1 mg/l 13 15 (w/v) C N -lysine (Sigma-Aldrich, Steinheim, Ger- 6 2 Linear and quadratic regressions were fitted using toast- many) as internal standard. Eluent A was UPLC-grade Mil- lipore water containing 0.1% (v/v) formic acid and eluent ing time as fixed effect in the model. Linear and quadratic effects were considered to be significant if the P value B was acetonitrile containing 0.1% (v/v) formic acid. The samples were analyzed using an Accela RP-UHPLC system was lower than 0.05 and as trends when the P value was between 0.05 and 0.10. Correlations between hydrolysis (Thermo Scientific, San Jose, CA, USA) with an Acquity BEH Amide Vanguard precolumn (2.1 × 50 mm, 1.7 μm parameters and molecular size distribution after hydrolysis were performed using the CORR procedure of SAS [15]. particle size) and an Acquity UPLC BEH 300 Amide col- umn (2.1 × 150 mm, 1.7 μm particle size). The column was maintained at 35 °C and the injection volume was 1 μl. The elution profile was as follows: 0–2 min isocratic on 80% Results and Discussion B, 2–3 min linear gradient from 80% B to 65% B, 3–5 min isocratic on 65% B, 5–7 min linear gradient from 65% B There were linear (P < 0.001) and quadratic (P = 0.001) effects of toasting time on N solubility of the RSM in water at to 40% B, 7–10 min isocratic on 40% B, 10–12 min linear gradient from 40% B to 80% B and 12–28 min isocratic pH 8.0 (Table 2). Solubility seems to decrease faster at shorter toasting times compared with longer ones. The decrease in on 80% B. The flow rate was 350 μ l/min. Mass spectro- metric data were obtained using a LTQ-VelosPro (Thermo N solubility can be caused by physical aggregation of pro- teins following protein unfolding [10]. In addition, chemical Scientific) equipped with a heated electrospray ionization (ESI) probe, coupled to the UHPLC system. The capillary modifications to proteins (e.g. formation of intermolecular disulfide bonds, Maillard reactions) might also be involved in voltage was set to 3 kV. The sheath gas flow rate was set at 20 and the auxiliary gas flow rate at 5 (arbitrary units). the solubility decrease [10]. Both of these phenomena (physi- cal aggregation and chemical modifications) might decrease A selected reaction monitoring (SRM) method (Table 1) was used for fragments analysis in negative ion mode for the accessibility of proteins for enzymatic hydrolysis. lysinoalanine and in positive ion mode for the other com- pounds. The normalized collision energy was set at 30 for Hydrolysis Kinetics furosine, lysine and lysinoalanine and at 35 for the other compounds, and the m/z width on the fragment was set to 1. Increasing the toasting time affected the hydrolysis profile of RSM (Fig. 2a). The hydrolysis rate constant An external standard calibration curve for furosine, lysine, lysinoalanine, lanthionine, N -carboxymethyl-lysine and (k) of the RSM decreased linearly (P < 0.001) with 1 3 J Am Oil Chem Soc (2017) 94:619–630 623 Table 2 N solubility at pH Toasting time N solubility RSM RSM Soluble protein frac- Insoluble protein 8.0 of the RSM and kinetic tion fraction parameters for the hydrolysis curves of RSM toasted for (% of total N) DH (%) k (/M×s) DH (%) k (/M×s) DH (%) k (/M×s) max max max different times and the soluble 0 min 31.3 20.0 6.8E−05 14.7 1.2E−04 19.5 4.0E−05 and insoluble fractions separated from these RSM 20 min 21.1 18.3 8.0E−05 14.8 1.4E−04 22.8 4.0E−05 40 min 17.7 19.2 7.0E−05 15.3 1.3E−04 21.1 4.3E−05 60 min 15.3 19.0 6.7E−05 14.3 1.5E−04 21.4 2.4E−05 80 min 12.3 20.4 5.0E−05 15.8 1.6E−04 21.9 2.2E−05 100 min 11.4 20.2 4.8E−05 14.5 1.7E−04 21.7 2.4E−05 120 min 9.9 21.7 3.4E−05 15.3 1.4E−04 22.3 1.5E−05 SEM 0.2 0.3 4.5E−06 0.2 6.7E−06 0.3 3.0E−06 P value b c d Linear <0.001 0.006 <0.001 0.47 0.18 0.34 <0.001 Quadratic 0.001 0.06 0.07 0.96 0.34 0.86 0.87 DH maximum degree of hydrolysis, k rate of protein hydrolysis, RSM rapeseed meal, SEM standard max error of the mean a 2 2 N solubility = 28.84 − 0.33 × time + 0.0015 × time (R = 0.95) b 2 RSM DH = 18.33 + 0.023 × time (R = 0.52) max c 2 RSM k = 8.30E−05 − 3.71E−07 × time (R = 0.80) d 2 Insolubles k = 4.46E−05 − 2.40E−07 × time (R = 0.78) increasing toasting time (Table 2). The k after toast- The k of the soluble protein fraction was 3–9-fold ing for 120 min was approximately 2-fold lower com- greater compared with that of the insoluble protein frac- pared with the k of the untoasted RSM (6.8E−05 vs. tion (Table 2), which matches our hypothesis. Higher rates 3.4E−05 /M×s, respectively). The decrease in k was of hydrolysis were reported previously [18] for soluble probably related to the restricted enzyme accessibility sodium caseinate compared with the insoluble form of the for proteolysis due to protein aggregation or chemi- same ingredient (casein), although these protein sources cal protein modifications [8]. Increasing the toasting were not heat-processed before hydrolysis. Both native and time caused a linear (P = 0.006) increase in the DH denatured proteins can remain in solution and this depends max of RSM (Table 2). The DH decreased 9% after the on their concentration and on their extent of aggregation max initial 20 min of toasting and subsequently gradually [19, 20]. Due to their high flexibility, native proteins in increased with increasing toasting time, up to a level in solution are in dynamic equilibrium with their distorted the 120 min-toasted RSM similar to the untoasted RSM forms and these distorted forms could be considered as (Fig. 2a). denatured [19], which could make them more accessible Toasting time had no effect on the k or the DH of for enzymatic cleavage. Furthermore, the relatively small max the soluble protein fraction (Table 2), as can be seen from size of the protein aggregates present in the soluble frac- the similar shapes of the degree of hydrolysis curves tion allows them to stay in solution [10]. Therefore, the (Fig. 2b). The soluble proteins could either be native or lower degree of aggregation of the proteins in the solu- aggregated proteins with a molecular size that allows ble fraction as compared with the proteins in the insolu- them to stay in solution [10, 16]. Aggregation of proteins ble fraction probably facilitated enzymatic hydrolysis, in solution into spherical particles occurs when heating thus increasing k. This would mean that a decrease in the is performed at a pH close to the isoelectric point of the proportion of soluble to insoluble proteins after thermal protein [17]. With longer heating times, there is second- processing leads to a decrease in the k for the complete ary aggregation of these spherical particles [17] into pro- material, as observed for the RSM in this study. Another tein condensates [16], which results in the reduction of possibility for the differences in k between the soluble and protein solubility. Protein aggregation mechanisms for insoluble fractions is that the insoluble fiber matrix struc- proteins in solution have been clearly identified before tures present in the latter, and not in the soluble fraction, [16, 17]. We assume that proteins thermally treated under might limit and decrease the rate of protein hydrolysis. semi-dry conditions (e.g. toasting or autoclaving) follow The k of the insoluble protein fraction decreased similar aggregation mechanisms as reported for proteins linearly (P < 0.001) by 62% when comparing in solution. the untoasted RSM to the 120 min-toasted RSM 1 3 624 J Am Oil Chem Soc (2017) 94:619–630 Fig. 2 Degree of hydrolysis during 120 min hydrolysis of a rapeseed meals toasted for different times and b soluble and c insoluble protein fractions separated from these rapeseed meals (Table 2), whereas no effects were noticed on the Nitrogen Solubility of the Insoluble Fractions DH of these fractions. As 70–90% of the proteins max in RSM correspond to insoluble proteins, the pat- The N solubilities of the insoluble protein fraction in the tern of the hydrolysis curves of the insoluble protein sodium phosphate buffer pH 7.5, buffer with 2% (w/v) fractions (Fig. 2c) were, as expected, similar to those SDS, buffer with 10 mM DTT and buffer with 2% (w/v) from the RSM. It is possible that there is an increase SDS and 10 mM DTT decreased with increasing toasting in the size of protein aggregates formed at increasing time (Fig. 3a). All solvents show a reduction in their solu- toasting times, which can hamper the penetration of bilizing power with increasing toasting time, which might the hydrolytic enzymes. be related to the increase in chemical modifications of the In vivo, enzymatic protein digestion starts at the stom- residues (e.g. Maillard reactions or crosslinking). Non- ach with pepsin and is followed by secretion of trypsin covalent electrostatic interactions can be cleaved by salt and chymotrypsin at the small intestine, where most of solutions [22], such as the phosphate buffer used in this the digestion occurs. The length of the small intestine experiment. Furthermore, SDS can cleave hydrogen bonds and the transit time, however, are limited and the extent and hydrophobic interactions, whereas DTT can cleave of protein digestion probably also depends on their rate disulfide bonds [22]. Non-covalent bonds may be impor - of digestion. These factors might be even more important tant for the stability of the insoluble aggregates. The SDS for poultry than for pigs, due to the short digestive tract solution solubilized twice the amount of N than the phos- of the former [21]. phate buffer alone, while the increase of N solubility with 1 3 J Am Oil Chem Soc (2017) 94:619–630 625 Fig. 3 a Nitrogen solubility of the insoluble protein fractions in of the insoluble protein fractions relative to the phosphate buffer of 100 mM phosphate buffer and the buffer containing either 2% SDS, solutions containing either 2% SDS, 10 mM DTT, or 2% SDS and 10 mM DTT or 2% SDS and 10 mM DTT. b Increase in N solubility 10 mM DTT additional DTT was minimal (Fig. 3a). The increase in N decreased the lysine content by 23%, which is similar to the solubility relative to the solubility by the phosphate buffer reduction in lysine content observed in the present study is reported in Fig. 3b. The relative importance of non-cova- after 100 min of toasting, regardless of the initial contents. lent bonds (solubilized by SDS solution) for the stability of Conversion of fructoselysine during 6 M HCl hydrolysis the aggregates does not change with the increasing toast- yields furosine (32%), pyridosine (16%) and regenerated ing time (Fig. 3b). With increasing toasting time, there is an lysine (56%) [24]. The N -carboxymethyl-lysine (Fig. 4a), increase in the relative amount of protein solubilized by the furosine (Fig. 4c) and N -carboxyethyl-lysine (Fig. 4d) DTT-containing buffer (Fig. 3b). This indicates that there contents in the RSM linearly (P < 0.001) increased with is formation of disulfide bonds in the insoluble aggregates increasing toasting time. The formation of these com- with increasing toasting time. The SDS–DTT solution sol- pounds with increasing toasting time does not completely ubilized more protein than SDS or DTT containing solu- account for the reduction in the lysine content, which tions separately. Synergy of SDS and DTT for N solubility also decreases linearly (P < 0.001; Fig. 4e). Whereas the has been reported before [23] for extruded soy proteins. In lysine content decreases 15.7 μg/mg CP from the 0 min to addition, the relative amount of protein solubilized by this the 120 min-toasted RSM, the sum of fructoselysine (cal- ε ε solution increased with increasing toasting time (Fig. 3b). culated from furosine), N -carboxymethyl-lysine and N - We suggest that cleavage of non-covalent bonds by SDS carboxyethyl-lysine only increases 11.1 μg/mg CP in the exposes extra disulfide bonds that can subsequently be same toasting time range. Other Maillard-derived com- cleaved by the DTT. pounds (e.g. [5-hydroxymethyl]-2-furfural) that were not determined in this experiment were probably also formed Maillard Reaction Products during toasting, which could account for this difference. The content of lysinoalanine (Fig. 4b) decreased during Formation of early (fructoselysine) and advanced Mail- the initial 20 min of toasting, but remained constant with ε ε lard reaction products (N -carboxymethyl-lysine and N - increasing toasting times. Formation of lysinoalanine is carboxyethyl-lysine) in the RSM was noticed with increas- favored with an alkaline pH [25], which is unlikely to have ing toasting time, along with a decrease in the lysine been applied during toasting of the RSM. No lanthionine content (Fig. 4). The lysine contents in the present study could be detected in any of the samples. (<5.3 g/100 g CP) were lower than those reported previ- Most of the chemically modified compounds formed ously [4] for untoasted (6.0 g/100 g CP) and commercially were present in the insoluble protein fraction. The con- toasted (5.6 g/100 g CP) canola meals. The decrease in tents of furosine (Fig. 4c) and N -carboxyethyl-lysine lysine content (7%) reported in that study when comparing (Fig. 4d) increased linearly (P < 0.001) with increasing the untoasted to the toasted canola meals corresponds in toasting times, whereas there were linear (P = 0.006) the present study to a RSM toasted for 20 min. In a previ- and quadratic (P = 0.009) effects of toasting time on the ous study [2], increasing the toasting time from 0 to 93 min N -carboxymethyl-lysine (Fig. 4a) content. The content 1 3 626 J Am Oil Chem Soc (2017) 94:619–630 Fig. 4 a N -[carboxymethyl]-lysine (CML), b lysinoalanine (LAL), c for different times and the insoluble protein fraction (gray) separated furosine (Fur), d N -[carboxyethyl]-lysine (CEL) and e lysine (Lys) from these rapeseed meals contents (μg/mg crude protein) in the rapeseed meals (black) toasted of lysine (Fig. 4e) decreased linearly (P < 0.001) with after 2 ml consist of unspecified protein/peptides that bind increasing toasting time. Contents (μg/mg CP) of furo- to the column. ε ε sine, N -carboxymethyl-lysine, N -carboxyethyl-lysine The largest fraction of soluble proteins from the intact and lysine were higher in the insoluble protein fractions RSM elute at a molecular mass >10 kDa (Fig. 5a). Intact compared with the complete RSM. The removal of the soluble proteins and soluble protein aggregates can be soluble protein fraction probably concentrates the aggre- located in this region [27]. The proportion of material gated and chemically modified insoluble fraction. The >10 kDa decreased (linear P < 0.001, quadratic P = 0.04) presence of these Maillard-derived compounds in the with increasing toasting time (Table 3). This was also insoluble fraction could delay the enzymatic cleavage of reflected by the linear increase (P < 0.001) of the propor - the available peptide bonds due to steric hindrance [26]. tion of proteins/peptides <10 kDa with increasing toasting In addition to protein aggregation, as described above, time. This increase was probably the result of a higher rela- the presence of these chemically modified compounds tive representation of this highly soluble fraction compared in the insoluble fraction could also explain the observed with the decreasing contents of soluble intact proteins at reduction of the k of the insoluble protein fractions. longer toasting times. There were no indications of other chemical crosslinks Hydrolysis of the RSM and their soluble and insolu- occurring: lysinoalanine (Fig. 4b) contents in the insolu- ble fractions changed their elution profiles compared with ble fraction do not change with increasing toasting time, the soluble intact RSM, as higher proportions of material whereas no lanthionine could be detected in any of the can be determined at lower molecular masses (<10 kDa; samples. Table 3). In the hydrolysates of RSM, the proportion of material <10 kDa corresponded to 80–86% of the quan- Molecular Size Distribution tified area (i.e. excluding the a-specific binding, reten- tion volume higher than 2 ml). The proportion of peptides Toasting time influenced the molecular weight distribution >10 kDa in the hydrolysates of the RSM decreased after the of the proteins in solution before hydrolysis (Fig. 5a) and initial 20 min of toasting and was not largely affected by the peptides obtained after hydrolysis of the RSM (Fig. 5b), a further increase in toasting time. Increasing the toasting the soluble (Fig. 5c) and the insoluble (Fig. 5d) protein time increased the overall molecular weight of these hydro- fractions. The fractions in the chromatograms that elute lysates, as an increasing (linear, P < 0.001) proportion of 1 3 J Am Oil Chem Soc (2017) 94:619–630 627 Fig. 5 Size exclusion chromatograms of the a soluble fractions of fractions of the rapeseed meals toasted for different times. Vertical rapeseed meals toasted for different times before hydrolysis and the dashed lines represent the cut-off points of 10 and 1.5 kDa hydrolysates of b rapeseed meals, c soluble and d insoluble protein peptides 1.5–10 kDa and a decreasing (linear, P < 0.001) However, increasing the toasting time decreased (linear proportion of peptides <1.5 kDa were determined at higher P < 0.001, quadratic P = 0.005) the proportion of mate- toasting times compared with lower ones. There was a neg- rial >10 kDa (Table 3). The nitrogen concentration before ative correlation (r = −0.73, P < 0.01) between the propor- hydrolysis was similar for all the hydrolyzed soluble frac- tion of peptides 1.5–10 kDa and the k of hydrolysis of the tions. Therefore, the decrease in the proportion of mate- RSM, whereas a positive correlation (r = 0.84, P < 0.001) rial >10 kDa is not expected to be due to the decrease in was determined for the proportion of peptides <1.5 kDa solubility with increasing toasting times (as explained (Table 4). The proportion of peptides 1.5–10 kDa in the previously for the soluble intact RSM), but to the facili- RSM hydrolysates was also positively correlated (r = 0.56, tated hydrolysis of soluble intact proteins. This was also P < 0.05) to the DH , whereas a negative correlation reflected by the overall decrease in the molecular weight of max (r = −0.74, P < 0.01) was determined between the propor- the hydrolysates, as the proportion of peptides 1.5–10 kDa tion of peptides <1.5 kDa and the DH . (P = 0.002) and <1.5 kDa (P < 0.001) linearly increased max The hydrolysates of the soluble protein fraction con- with increasing toasting times. The increase in the propor- tained a larger proportion of material >10 kDa com- tion of peptides <1.5 kDa with increasing toasting time pared with the RSM hydrolysates (Table 3). This can be obtained after hydrolysis of the soluble protein fraction explained by the presence of intact soluble proteins, which can be explained by an increased denaturation of the sol- were detected by SDS–PAGE in the hydrolysates of the uble proteins, exposing cleavage sites that were initially RSM and the soluble protein fraction (results not shown). not accessible for the enzymes [28]. A positive correlation 1 3 628 J Am Oil Chem Soc (2017) 94:619–630 1 3 Table 3 Relative molecular weight distribution (%) of intact RSM toasted for different times and the hydrolysates of these RSM, and soluble and insoluble protein fractions separated from these RSM Toasting time Intact RSM Hydrolysates RSM Soluble protein fraction Insoluble protein fraction >10 kDa 10–1.5 kDa <1.5 kDa >10 kDa 10–1.5 kDa <1.5 kDa >10 kDa 10–1.5 kDa <1.5 kDa >10 kDa 10–1.5 kDa <1.5 kDa 0 min 52.3 12.8 34.9 19.2 21.4 59.5 33.6 26.4 40.0 11.7 32.0 56.3 20 min 50.9 9.0 40.1 13.1 24.3 62.6 27.8 27.8 44.4 7.0 35.8 57.2 40 min 49.5 10.1 40.5 12.7 25.7 61.6 24.9 28.1 47.0 7.5 36.6 56.0 60 min 48.6 9.8 41.6 13.3 28.6 58.1 22.8 30.4 46.8 9.1 37.3 53.6 80 min 45.2 10.3 44.5 12.6 28.7 58.7 20.3 30.6 49.1 11.2 38.3 50.5 100 min 42.4 11.3 46.3 13.7 30.7 55.6 19.2 31.3 49.5 13.1 39.3 47.6 120 min 38.3 12.4 49.3 14.0 30.5 55.5 18.8 30.7 50.5 15.6 39.6 44.7 SEM 1.4 0.4 1.2 0.5 0.9 0.8 1.3 0.5 0.9 0.9 0.6 1.3 P value c e f h i Linear <0.001 0.15 <0.001 0.34 <0.001 <0.001 <0.001 0.002 <0.001 0.003 <0.001 <0.001 a b d g j k l Quadratic 0.04 0.006 0.76 0.006 0.13 0.32 0.005 0.18 0.07 <0.001 0.02 0.002 Abbreviations: RSM, rapeseed meal; SEM, standard error of the mean a 2 2 Intact >10 kDa = 51.84 − 0.017 × time − 0.00,079 × time (R = 0.91) b 2 2 intact 10–1.5 kDa = 11.45 − 0.07 × time + 0.00066 × time (R = 0.63) c 2 Intact <1.5 kDa = 36.38 + 0.10 × time (R = 0.85) d 2 2 RSM >10 kDa = 16.97 − 0.13 × time + 0.00093 × time (R = 0.58) e 2 RSM 10–1.5 kDa = 22.83 + 0.073 × time (R = 0.76) f 2 RSM <1 kDa = 62.63 − 0.060 × time (R = 0.67) g 2 2 Soluble >10 kDa = 32.81 − 0.23 × time + 0.00099 × time (R = 0.93) h 2 Soluble 10–1.5 kDa = 27.09 + 0.038 × time (R = 0.60) i 2 Soluble <1 kDa = 42.68 + 0.071 × time (R = 0.77) j 2 2 Insoluble >10 kDa = 10.10 − 0.093 × time + 0.0012 × time (R = 0.91) k 2 2 Insoluble 10–1.5 kDa = 32.74 + 0.11 × time − 0.00041 × time (R = 0.92) l 2 2 Insoluble <1 kDa = 57.16 − 0.014 × time − 0.00078 × time (R = 0.97) J Am Oil Chem Soc (2017) 94:619–630 629 Table 4 Pearson correlation Proportion of peptides after hydrolysis RSM Soluble protein Insoluble protein coefficients between the fraction fraction proportion of peptides after hydrolysis and the maximum DH k DH k DH k max max max degree of hydrolysis (DH ) max >10 kDa 0.16 −0.01 −0.25 −0.44 0.02 −0.77** and rate of hydrolysis (k) of the RSM and the soluble and 1.5–10 kDa 0.56* −0.73** 0.08 0.63* 0.44 −0.74** insoluble protein fractions <1.5 kDa −0.74** 0.84*** 0.32 0.25 −0.23 0.87*** separated from these RSM DH maximum degree of hydrolysis, k rate of protein hydrolysis, RSM rapeseed meal max Significance level: * P < 0.05, ** P < 0.01, *** P < 0.001 (r = 0.63, P < 0.05) was determined between the propor- contain peptides with larger sizes, in addition to intact pro- tion of peptides 1.5–10 kDa and the k of hydrolysis of the teins, can probably be considered to be less digestible com- soluble protein fraction (Table 4). pared with those with smaller sizes. In contrast to what was observed in the hydrolysates of the soluble protein fraction, increasing the toasting time increased the overall molecular weight of the peptides pre- Conclusion sent in the hydrolysates of the insoluble protein fraction (Fig. 5d). The proportion of material >10 kDa decreased The rate of protein hydrolysis of the soluble protein frac- after the initial 20 min of toasting, but steadily increased tion was 3–9-fold greater than that of the insoluble pro- with increasing toasting time (Table 3). This is not expected tein fraction. The decrease in the rate of hydrolysis of the to be due to the presence of intact proteins, as no clear RSM observed with increasing toasting time results from bands were detected by SDS–PAGE in these hydrolysates a combination of (1) the reduction in the proportion of fast (results not shown). The proportion of material >10 kDa hydrolysable soluble proteins to slowly digestible insoluble was negatively correlated to the k of hydrolysis (r = −0.77, proteins and (2) the decrease in the rate of hydrolysis of P < 0.01) of the insoluble protein fraction (Table 4). Fur- the insoluble proteins with increasing toasting time due to thermore, increasing the toasting time increased (linear the formation of disulfide bonds and/or chemically modi- P < 0.001, quadratic P = 0.02) the proportion of peptides fied amino acid residues. In addition, increasing the toast- 1.5–10 kDa and decreased (linear P < 0.001, quadratic ing time results in an overall increase of the size of the pep- P = 0.002) the proportion of peptides <1.5 kDa. There was tides after hydrolysis. Positive correlations were obtained a negative correlation (r = −0.74, P < 0.01) between the between the rates of protein hydrolysis of the RSM and the proportion of peptides 1.5–10 kDa and the k of hydrolysis insoluble protein fraction with the proportion of small pep- of the insoluble protein fraction, whereas a positive correla- tides (<1.5 kDa) after hydrolysis. tion (r = 0.87, P < 0.001) was determined with the propor- Acknowledgements The authors gratefully acknowledge the finan- tion of peptides <1.5 kDa (Table 4). cial support from the Wageningen UR “IPOP Customized Nutrition” The correlations between k and the size distribution of programme financed by Wageningen UR, the Dutch Ministry of Eco- the peptides in the hydrolysates of RSM and the insoluble nomic Affairs, WIAS, Agrifirm Innovation Center, ORFFA Addi- protein fraction could be explained by a shift of the hydro- tives BV, Ajinomoto Eurolysine s.a.s and Stichting VICTAM BV. We appreciate the support of Claire Butré with the UHPLC-MS measure- lytic mechanism from a more one-by-one type to a more ments. SSV acknowledges the support of the Universidad de Costa zipper-type-dominated system with increasing toasting Rica. time [19, 29]. Hydrolysis of most proteins shows an inter- mediate behavior between these two types of hydrolytic Compliance with ethical standards mechanisms [19]. In the one-by-one type of hydrolysis, the cleavage of peptide bonds in one protein is followed by a Conflict of interest The authors declare that they have no conflicts of interest. fast cleavage into smaller peptides. Overall, a higher pro- portion of intermediate peptides (Table 3) was determined in this study at short toasting times compared with long Open Access This article is distributed under the terms of the Crea- tive Commons Attribution 4.0 International License (http://crea- toasting times. In contrast, in the zipper-type hydrolysis, tivecommons.org/licenses/by/4.0/), which permits unrestricted use, several peptide bonds are cleaved simultaneously, which distribution, and reproduction in any medium, provided you give is followed by a slow conversion of large peptides into appropriate credit to the original author(s) and the source, provide a smaller peptides. The shift towards a zipper-type enzymatic link to the Creative Commons license, and indicate if changes were made. cleavage renders peptides of larger sizes, compared with the more one-by-one type of hydrolysis. Hydrolysates that 1 3 630 J Am Oil Chem Soc (2017) 94:619–630 14. Chabanon G, Chevalot I, Framboisier X, Chenu S, Marc I (2007) References Hydrolysis of rapeseed protein isolates: kinetics, characteriza- tion and functional properties of hydrolysates. Process Biochem 1. Jensen SK, Liu YG, Eggum BO (1995) The effect of heat treat- 42:1419–1428 ment on glucosinolates and nutritional value of rapeseed meal in 15. SAS Institute Inc (2011) The SAS system for Windows. SAS rats. Anim Feed Sci Technol 53:17–28 Institute Inc., Cary, NC, USA 2. Eklund M, Sauer N, Schöne F, Messerschmidt U, Rosenfelder P, 16. Amin S, Barnett GV, Pathak JA, Roberts CJ, Sarangapani PS Htoo JK, Mosenthin R (2015) Effect of processing of rapeseed (2014) Protein aggregation, particle formation, characterization under defined conditions in a pilot plant on chemical composi- and rheology. Curr Opin Colloid Interface Sci 19:438–449 tion and standardized ileal amino acid digestibility in rapeseed 17. Nicolai T, Durand D (2013) Controlled food protein aggregation meal for pigs. J Anim Sci 93:2813–2825 for new functionality. Curr Opin Colloid Interface Sci 18:249–256 3. Salazar-Villanea S, Bruininx EMAM, Gruppen H, Hendriks 18. Tonheim SK, Nordgreen A, Høgøy I, Hamre K, Rønnestad I WH, Carré P, Quinsac A, Van Der Poel AFB (2016) Physical and (2007) In vitro digestibility of water-soluble and water-insoluble chemical changes of rapeseed meal proteins during toasting and protein fractions of some common fish larval feeds and feed their effects on in vitro digestibility. J Anim Sci Biotechnol 7:62 ingredients. Aquaculture 262:426–435 4. Newkirk RW, Classen HL, Scott TA, Edney MJ (2003) The 19. Adler-Nissen J (1976) Enzymatic hydrolysis of proteins for digestibility and content of amino acids in toasted and non- increased solubility. J Agric Food Chem 24:1090–1093 toasted canola meals. Can J Anim Sci 83:131–139 20. Shivu B, Seshadri S, Li J, Oberg KA, Uversky VN, Fink AL 5. Hulshof TG, Bikker P, van der Poel AFB, Hendriks WH (2016) (2013) Distinct β-sheet structure in protein aggregates deter- Assessment of protein quality of soybean meal and 00-rapeseed mined by ATR–FTIR spectroscopy. Biochemistry 52:5176–5183 meal toasted in the presence of lignosulfonate by amino acid 21. Thorpe J, Beal JD (2000) Vegetable protein meals and the effects digestibility in growing pigs and Maillard reaction products. J of enzymes. In: Bedford MR, Partridge GG (eds) Enzymes in farm Anim Sci 94:1020–1030 animal nutrition. CABI Publishing, Oxfordshire, pp 125–143 6. Newkirk RW, Classen HL, Edney MJ (2003) Effects of prepress- 22. Liu KS, Hsieh FH (2008) Protein-protein interactions during solvent extraction on the nutritional value of canola meal for high-moisture extrusion for fibrous meat analogues and compari- broiler chickens. Anim Feed Sci Technol 104:111–119 son of protein solubility methods using different solvent systems. 7. Thacker PA, Newkirk RW (2005) Performance of growing- J Agric Food Chem 56:2681–2687 finishing pigs fed barley-based diets containing toasted or non- 23. Liu KS, Hsieh FH (2007) Protein-protein interactions in high toasted canola meal. Can J Anim Sci 85:53–59 moisture-extruded meat analogs and heat-induced soy protein 8. Gerrard JA, Lasse M, Cottam J, Healy JP, Fayle SE, Rasiah I, gels. J Am Oil Chem Soc 84:741–748 Brown PK, BinYasir SM, Sutton KH, Larsen NG (2012) Aspects 24. Krause R, Knoll K, Henle T (2003) Studies on the formation of of physical and chemical alterations to proteins during food pro- furosine and pyridosine during acid hydrolysis of different Ama- cessing—some implications for nutrition. Br J Nutr 108(Suppl dori products of lysine. Eur Food Res Technol 216:277–283 2):S288–S297 25. Schwarzenbolz U, Henle T (2010) Non-enzymatic modifications 9. Olsen JV, Ong SE, Mann M (2004) Trypsin cleaves exclusively of proteins under high-pressure treatment. High Pressure Res C-terminal to arginine and lysine residues. Mol Cell Proteomics 30:458–465 3:608–614 26. Puigserver AJ, Sen LC, Gonzales-Flores E, Feeney RE, Whi- 10. Wang W, Nema S, Teagarden D (2010) Protein aggregation— taker JR (1979) Covalent attachment of amino acids to casein. 1. pathways and influencing factors. Int J Pharm 390:89–99 Chemical modification and rates of in vitro enzymic hydrolysis 11. Carbonaro M, Bonomi F, Iametti S, Cappelloni M, Carnovale of derivatives. J Agric Food Chem 27:1098–1104 E (1998) Aggregation of proteins in whey from raw and heat- 27. He R, He H-Y, Chao D, Ju X, Aluko R (2014) Effects of high processed milk: formation of soluble macroaggregates and nutri- pressure and heat treatments on physicochemical and gelation tional consequences. LWT Food Sci Technol 31:522–529 properties of rapeseed protein isolate. Food Bioprocess Technol 12. Butré CI (2014) Introducing enzyme selectivity as a quantitative 7:1344–1353 parameter to describe the effects of substrate concentration on 28. O’Loughlin IB, Murray BA, Kelly PM, FitzGerald RJ, Brodkorb protein hydrolysis. Laboratory of Food Chemistry, Wageningen A (2012) Enzymatic hydrolysis of heat-induced aggregates of University, The Netherlands, p 208 whey protein isolate. J Agric Food Chem 60:4895–4904 13. Pedersen B, Eggum BO (1983) Prediction of protein digestibility 29. Linderstrom-Lang K (1953) The initial phases of the enzymatic by an in vitro enzymatic pH-stat procedure. Z Tierphysiol Tier- degradation of proteins. Bull Soc Chim Biol 35:100–116 ernahr Futtermittelkd 49:265–277 1 3 http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Journal of the American Oil Chemists' Society Pubmed Central

Effects of Toasting Time on Digestive Hydrolysis of Soluble and Insoluble 00-Rapeseed Meal Proteins

Journal of the American Oil Chemists' Society , Volume 94 (4) – Feb 11, 2017

Loading next page...
 
/lp/pubmed-central/effects-of-toasting-time-on-digestive-hydrolysis-of-soluble-and-C4Q3i17JDA

References (31)

Publisher
Pubmed Central
Copyright
© The Author(s) 2017
ISSN
0003-021X
eISSN
1558-9331
DOI
10.1007/s11746-017-2960-8
Publisher site
See Article on Publisher Site

Abstract

J Am Oil Chem Soc (2017) 94:619–630 DOI 10.1007/s11746-017-2960-8 ORIGINAL PAPER Effects of Toasting Time on Digestive Hydrolysis of Soluble and Insoluble 00‑Rapeseed Meal Proteins 1,2 2,3 4 Sergio Salazar‑Villanea · Erik M. A. M. Bruininx · Harry Gruppen · 5 6 2 Patrick Carré · Alain Quinsac · Antonius F. B. van der Poel Received: 7 October 2016 / Revised: 20 December 2016 / Accepted: 31 January 2017 / Published online: 11 February 2017 © The Author(s) 2017. This article is published with open access at Springerlink.com Abstract Thermal damage to proteins can reduce their in the hydrolysis rate of this fraction. Overall, longer toast- nutritional value. The effects of toasting time on the kinet- ing times increased the size of the peptides resulting after ics of hydrolysis, the resulting molecular weight distribu- hydrolysis of the RSM and the insoluble protein fraction. tion of 00-rapeseed meal (RSM) and the soluble and insolu- The hydrolysis kinetics of the soluble and insoluble protein ble protein fractions separated from the RSM were studied. fractions and the proportion of soluble:insoluble proteins in Hydrolysis was performed with pancreatic proteases to rep- the RSM explain the reduction in the rate of protein hydrol- resent in vitro protein digestibility. Increasing the toasting ysis observed in the RSM with increasing toasting time. time of RSM linearly decreased the rate of protein hydroly- sis of RSM and the insoluble protein fractions. The extent Keywords Hydrolysis rate · Maillard · Protein solubility · of hydrolysis was, on average, 44% higher for the insolu- Rapeseed meal · Toasting ble compared with the soluble protein fraction. In contrast, the rate of protein hydrolysis of the soluble protein fraction Abbreviations was 3–9-fold higher than that of the insoluble protein frac- DH Maximum degree of hydrolysis max tion. The rate of hydrolysis of the insoluble protein frac-DTT Dithiothreitol tion linearly decreased by more than 60% when compar- k Rate of h ydrolysis ing the untoasted to the 120 min toasted RSM. Increasing RSM Rapeseed meal the toasting time elicited the formation of Maillard reac- ε ε tion products (furosine, N -carboxymethyl-lysine and N - carboxyethyl-lysine) and disulfide bonds in the insoluble Introduction protein fraction, which is proposed to explain the reduction The production of defatted 00-rapeseed meal (RSM) involves toasting for the removal of remnant solvent after * Sergio Salazar-Villanea oil extraction, the inactivation of myrosinase and the degra- sergio.salazarvillanea@ucr.ac.cr dation of glucosinolates [1–3], the main antinutritional fac- Wageningen Livestock Research, Wageningen, The tors of RSM for monogastric animals. Increasing the toast- Netherlands ing time decreases protein solubility [1, 3] and the contents Animal Nutrition Group, Wageningen University of lysine [1, 3, 4] and reactive lysine [2, 3] of the resulting and Research, P.O. Box 338, 6700 AH Wageningen, The RSM. Reactive lysine can be considered as the protein- Netherlands bound lysine with a free ε-amino group [5]. Nutritionally, Agrifirm Innovation Center BV, Royal Dutch Agrifirm the changes in protein solubility, lysine and reactive lysine Group, Apeldoorn, The Netherlands contents have been noticed as reduction of in vitro crude Laboratory of Food Chemistry, Wageningen University protein digestibility [2], apparent ileal protein digestibility and Research, Wageningen, The Netherlands in broilers [6], standardized ileal protein digestibility in CREOL/OLEAD, Pessac, France pigs [2] and apparent total tract crude protein digestibility Terres Inovia, Paris, France in rats [1] and pigs [7]. 1 3 620 J Am Oil Chem Soc (2017) 94:619–630 Chemical and physical modifications can impair the porcine intestinal mucosa (50–100 units/g solid, EC 232- accessibility of proteins for enzymatic hydrolysis [8]. The 875-1) were obtained from Sigma-Aldrich (St. Louis, MO, proteases in the pancreatic secretions at the small intestine USA). The furosine, lysinoalanine and N -carboxymethyl- are highly specific. Trypsin cleaves peptide bonds involving lysine standards were obtained from PolyPeptide Labora- the carboxyl groups of either lysine or arginine [9], which tories (Strasbourg, France), whereas the rest of the stand- 13 15 are also the most susceptible amino acids to heat damage. ards ( C , N -lysine, lysine, lanthionine) were obtained 6 2 Lysine and arginine residues that are modified via Mail- from Sigma-Aldrich (Steinheim, Germany). lard reactions could, therefore, reduce enzyme accessibil- ity for proteolysis, finally reducing their standardized ileal digestibility [5]. Also, physical modifications, such as pro- Rapeseed Meals Preparation tein aggregation, might have a similar effect. Protein aggre- gates can be formed after hydrothermal processing and are An untoasted RSM was prepared by cold-pressing of the noticed as a reduction in protein solubility [10]. The forma- 00-rapeseeds, solvent-extraction and desolventization tion of aggregates can reduce protein accessibility for enzy- using indirect heat. The 00-rapeseeds were cold-pressed matic hydrolysis [11]. (La Mecanique Moderne MBU 75 type, Arras, France) In proteins that are accessible for enzymatic hydroly- at 250 kg/h with temperatures not exceeding 80 °C. sis, the gastric and intestinal enzymes cleave proteins into Solvent extraction was performed at temperatures not peptides, followed by cleavage into free small peptides higher than 55 °C on a belt extractor (B-1930, Desmet- and free amino acids by the brush border enzymes in the Ballestra, Zaventem, Belgium) at 230 l/h flow of hexane gut’s epithelium. The size of the peptides produced during and 160 kg/h flow of the rapeseed cake. Desolventization hydrolysis depends on the accessibility of the protein for with indirect heat (without direct steam) was performed the enzymes. Even at similar degrees of protein hydrolysis, in a desolventizer toaster (Schumacher type, Desmet- molecular size distribution of the resulting peptides can be Ballestra) for 60 min at temperatures of 90 ± 3 °C. different [12]. This might be the result of the selectivity of A batch of 150 kg of the untoasted RSM was toasted the enzymes or the accessibility of the proteins for cleavage with the use of direct steam (30 kg/h) for 120 min, with [12]. spot samples of 5 kg taken every 20 min (Fig. 1). Toast- The aim of the present study was to determine the effects ing of a separate batch of 150 kg of RSM was performed of toasting time during the production process of RSM on during the next day under the same conditions for replica- the kinetics of hydrolysis of proteins present in the com- tion. Temperatures during toasting on the first day ranged plete material, and in its soluble and insoluble fractions, from 107 to 112 °C and between 109 and 112 °C on the and on the resulting molecular size distribution of the pep- tides after hydrolysis. Previous results from our research group indicate that there is a highly significant positive cor - relation between nitrogen solubility and the rate of protein hydrolysis [3]. Therefore, we hypothesize that the rate of hydrolysis will be higher for the soluble protein fraction compared with the insoluble protein fraction, but that the rates at different toasting times will not vary within the soluble and the insoluble protein fractions. We also hypoth- esize that damage to the proteins due to prolonged toast- ing times will change the molecular size distribution of the peptides obtained after hydrolysis towards a larger size. Materials and Methods Materials Rapeseed meals were prepared from 00-rapeseed (Bras- sica napus) at the pilot plant of CREOL/OLEAD (Pes- sac, France). Trypsin (type IX-S, 13,000–20,000 BAEE units/mg protein, EC 232-650-8), chymotrypsin (type II, ≥40 units/mg protein, EC 232-671-2) and peptidase from Fig. 1 Schematic view of the design of the experiment 1 3 J Am Oil Chem Soc (2017) 94:619–630 621 second day. In total, 13 samples of RSM (1 untoasted RSM substrate to enzyme ratio was used for hydrolysis of RSM, and 12 toasted RSM) were obtained. Untoasted and toasted soluble and insoluble fractions. All hydrolyses were per- RSM were ground with a centrifugal mill (ZM200, Retsch, formed in duplicate. Haan, Germany) at 8000 rpm to pass a 1-mm sieve. The volume of alkali added was used for the calculation of the degree of hydrolysis according to Eq. 1, Fractionation of Proteins into Soluble and Insoluble Vb × Nb Fractions DH(%) = ×100 (1) α × mp × htot The water-soluble and water-insoluble fractions of the meals in which Vb is the volume added (ml), Nb is the normality were separated by suspending 25 g of the RSM in 250 ml of of the titration solution, α is the degree of dissociation of water. The pH of the suspension was adjusted to 8.0 with the α-NH group (in this case, 0.794 at 37 °C and pH 8.0), NaOH and magnetic stirring was applied for 20 min at room mp is the mass of protein (g) and htot is the total number of temperature. The soluble fraction was separated by centrifu- peptide bonds per gram of protein (7.8 meq/g) [14]. gation (11,900×g, 20 min, room temperature). Transparent Modelling of the degree of hydrolysis curves was based solutions were obtained, which were dialyzed extensively on second-order reaction kinetics, which is described in against a 0.01 M NaCl solution. After dialysis, the pH was Eq. 2, re-adjusted to 8.0 with NaOH. The soluble fractions were DH max kept at 4 °C and hydrolyzed within 24 h. The insoluble frac- DH(%) = DH − max (2) 1 + k × t × DH max tion was filtered through a nylon cloth and washed three times with 250 ml of water in order to remove soluble pro- in which DH is the maximum degree of hydrolysis (%), max teins and was freeze-dried prior to hydrolysis. k is the hydrolysis rate constant (/M×s) and t is the hydrol- ysis time. Fitting of this model was performed using the Analytical Methods MODEL procedure in SAS [15]. Nitrogen contents of the complete meals and the insoluble Nitrogen Solubility fractions were determined by combustion (AOAC 968.06, Thermo Quest NA 2100 Nitrogen and Protein Analyzer, Five hundred mg of RSM were suspended in 10 ml of Breda, The Netherlands). The N concentration of the sol- water and the suspension was adjusted to pH 8.0 using uble protein fractions was measured after an aliquot of 2 M NaOH. The suspension was stirred for 20 min at room 0.3 ml was oven-dried at 60 °C overnight. temperature and centrifuged (16,100×g, 15 min, room tem- perature). An aliquot (0.3 ml) was oven-dried overnight at Protein Hydrolysis 60 °C and analyzed for N content. Nitrogen solubility of the insoluble protein fractions were determined in water, In vitro protein hydrolysis was performed for 120 min 100 mM sodium phosphate buffer pH 7.5, and the same using a modification of the method described previ- buffer containing either 2% (w/v) SDS, 10 mM dithiothrei- ously [13]. Briefly, 10 ml of aqueous suspension contain- tol (DTT) or both 2% (w/v) SDS and 10 mM DTT. Briefly, ing 1 mg N/ml was adjusted to pH 8.0 in a titration unit 1.5 ml of these solutions were added to 75 mg of the insol- (719 S Titrino, Metrohm, Herisau, Switzerland) at 39 °C uble protein fractions. The suspensions were vortexed for using 0.1 M NaOH. At this point, 1 ml of enzyme solu- 20 s and mixed in a head-over-tail rotator for 20 min at tion containing 1.61 mg trypsin, 3.96 mg chymotrypsin 20 rpm. Following centrifugation (16,100×g, 15 min, room and 1.18 mg porcine intestinal peptidase was added and the temperature), 0.3 ml of the supernatant was oven-dried 120 min titration was initiated. This procedure was used for overnight at 60 °C and analyzed for N content. the complete RSM and the insoluble fractions. Concerning the soluble fraction, different concentrations Size Exclusion Chromatography (SEC) of N in the soluble fractions were obtained upon solubiliz- ing the RSM in water. Demineralized water was used for After hydrolysis, a sample of 1.5 ml was taken from the diluting these solutions to the concentration of the low- supernatant of the hydrolysate, heated at 99 °C for 15 min est one of the solutions (0.30 mg N/ml), to a volume of and centrifuged (16,100×g, 10 min, room temperature). In 10 ml. The solutions were adjusted to pH 8.0 in the titra- addition, samples from the water-soluble fraction of RSM tion unit using 0.05 M NaOH. At this point, 1 ml of an were analyzed after centrifugation (16,100×g, 15 min, enzyme solution containing 0.48 mg of trypsin, 1.19 mg of room temperature). The samples were analyzed in the chymotrypsin and 0.35 mg of porcine intestinal peptidase ÄKTA micro system (GE Healthcare, Uppsala, Sweden) was added and the 120 min titration initiated. The same using a Superdex 75 column (GE Healthcare) at a flow rate 1 3 622 J Am Oil Chem Soc (2017) 94:619–630 Table 1 Selected reaction monitoring conditions of 100 μl/min with UV detection at 220 nm. The eluent used was 10 mM sodium phosphate buffer of pH 7.0 con- a Compound Parent mass (Da) Fragment (m/z) taining 150 mM NaCl and 2% (w/v) SDS. The volume of Lysine 146 130 injection was 50 μl. A calibration curve of the elution vol- 13 15 C N -Lysine 154 137 umes in the column was obtained using threonine (119 Da), 6 2 N -Carboxymethyl-lysine 204 84, 130 proline-glycine-glycine (229 Da), vitamin B12 (1355 Da), Lanthionine 208 120 lysozyme (14,307 Da), β-lactoglobulin (18,400 Da), and N -Carboxyethyl-lysine 218 84, 130 ovalbumin (42,700 Da). Areas under the curve were inte- Lysinoalanine 233 128, 145 grated manually and the proportions of peptides based on Furosine 255 84, 130 AU response in each region (>10, 10–1.5 and <1.5 kDa) were calculated relative to the total area under the curve. Parent mass is defined as the molecular mass of the compounds before ionization Maillard Reaction Products, Crosslinked Compounds and Lysine N -carboxyethyl-lysine with concentrations of 0.01, 0.1, 1, The contents of furosine, N -[carboxymethyl]-lysine, lysino- 2.5, 5 and 10 mg/l of each standard was used to calculate alanine, lanthionine, N -[carboxyethyl]-lysine and lysine in the content of each compound. Compounds were quantified the RSM and the insoluble protein fractions were quanti- using the external standard calibration curve by plotting the fied by UHPLC-MS. The samples (10 mg) were hydrolyzed MS peak area divided by the MS peak area of the labelled with 1 ml of 6 M HCl during 24 h at 110 °C. The tubes were Lys, used as internal standard. Data were acquired and ana- dried under N flow and the dried material was re-suspended lyzed using XCalibur 2.2 software (Thermo Scientific). in 1 ml of UPLC-grade Milli-Q water, sonicated and centri- fuged (16,100×g, 3.5 min, room temperature). The superna- Statistical Analysis tant was diluted 50 times in eluent A that contained 1 mg/l 13 15 (w/v) C N -lysine (Sigma-Aldrich, Steinheim, Ger- 6 2 Linear and quadratic regressions were fitted using toast- many) as internal standard. Eluent A was UPLC-grade Mil- lipore water containing 0.1% (v/v) formic acid and eluent ing time as fixed effect in the model. Linear and quadratic effects were considered to be significant if the P value B was acetonitrile containing 0.1% (v/v) formic acid. The samples were analyzed using an Accela RP-UHPLC system was lower than 0.05 and as trends when the P value was between 0.05 and 0.10. Correlations between hydrolysis (Thermo Scientific, San Jose, CA, USA) with an Acquity BEH Amide Vanguard precolumn (2.1 × 50 mm, 1.7 μm parameters and molecular size distribution after hydrolysis were performed using the CORR procedure of SAS [15]. particle size) and an Acquity UPLC BEH 300 Amide col- umn (2.1 × 150 mm, 1.7 μm particle size). The column was maintained at 35 °C and the injection volume was 1 μl. The elution profile was as follows: 0–2 min isocratic on 80% Results and Discussion B, 2–3 min linear gradient from 80% B to 65% B, 3–5 min isocratic on 65% B, 5–7 min linear gradient from 65% B There were linear (P < 0.001) and quadratic (P = 0.001) effects of toasting time on N solubility of the RSM in water at to 40% B, 7–10 min isocratic on 40% B, 10–12 min linear gradient from 40% B to 80% B and 12–28 min isocratic pH 8.0 (Table 2). Solubility seems to decrease faster at shorter toasting times compared with longer ones. The decrease in on 80% B. The flow rate was 350 μ l/min. Mass spectro- metric data were obtained using a LTQ-VelosPro (Thermo N solubility can be caused by physical aggregation of pro- teins following protein unfolding [10]. In addition, chemical Scientific) equipped with a heated electrospray ionization (ESI) probe, coupled to the UHPLC system. The capillary modifications to proteins (e.g. formation of intermolecular disulfide bonds, Maillard reactions) might also be involved in voltage was set to 3 kV. The sheath gas flow rate was set at 20 and the auxiliary gas flow rate at 5 (arbitrary units). the solubility decrease [10]. Both of these phenomena (physi- cal aggregation and chemical modifications) might decrease A selected reaction monitoring (SRM) method (Table 1) was used for fragments analysis in negative ion mode for the accessibility of proteins for enzymatic hydrolysis. lysinoalanine and in positive ion mode for the other com- pounds. The normalized collision energy was set at 30 for Hydrolysis Kinetics furosine, lysine and lysinoalanine and at 35 for the other compounds, and the m/z width on the fragment was set to 1. Increasing the toasting time affected the hydrolysis profile of RSM (Fig. 2a). The hydrolysis rate constant An external standard calibration curve for furosine, lysine, lysinoalanine, lanthionine, N -carboxymethyl-lysine and (k) of the RSM decreased linearly (P < 0.001) with 1 3 J Am Oil Chem Soc (2017) 94:619–630 623 Table 2 N solubility at pH Toasting time N solubility RSM RSM Soluble protein frac- Insoluble protein 8.0 of the RSM and kinetic tion fraction parameters for the hydrolysis curves of RSM toasted for (% of total N) DH (%) k (/M×s) DH (%) k (/M×s) DH (%) k (/M×s) max max max different times and the soluble 0 min 31.3 20.0 6.8E−05 14.7 1.2E−04 19.5 4.0E−05 and insoluble fractions separated from these RSM 20 min 21.1 18.3 8.0E−05 14.8 1.4E−04 22.8 4.0E−05 40 min 17.7 19.2 7.0E−05 15.3 1.3E−04 21.1 4.3E−05 60 min 15.3 19.0 6.7E−05 14.3 1.5E−04 21.4 2.4E−05 80 min 12.3 20.4 5.0E−05 15.8 1.6E−04 21.9 2.2E−05 100 min 11.4 20.2 4.8E−05 14.5 1.7E−04 21.7 2.4E−05 120 min 9.9 21.7 3.4E−05 15.3 1.4E−04 22.3 1.5E−05 SEM 0.2 0.3 4.5E−06 0.2 6.7E−06 0.3 3.0E−06 P value b c d Linear <0.001 0.006 <0.001 0.47 0.18 0.34 <0.001 Quadratic 0.001 0.06 0.07 0.96 0.34 0.86 0.87 DH maximum degree of hydrolysis, k rate of protein hydrolysis, RSM rapeseed meal, SEM standard max error of the mean a 2 2 N solubility = 28.84 − 0.33 × time + 0.0015 × time (R = 0.95) b 2 RSM DH = 18.33 + 0.023 × time (R = 0.52) max c 2 RSM k = 8.30E−05 − 3.71E−07 × time (R = 0.80) d 2 Insolubles k = 4.46E−05 − 2.40E−07 × time (R = 0.78) increasing toasting time (Table 2). The k after toast- The k of the soluble protein fraction was 3–9-fold ing for 120 min was approximately 2-fold lower com- greater compared with that of the insoluble protein frac- pared with the k of the untoasted RSM (6.8E−05 vs. tion (Table 2), which matches our hypothesis. Higher rates 3.4E−05 /M×s, respectively). The decrease in k was of hydrolysis were reported previously [18] for soluble probably related to the restricted enzyme accessibility sodium caseinate compared with the insoluble form of the for proteolysis due to protein aggregation or chemi- same ingredient (casein), although these protein sources cal protein modifications [8]. Increasing the toasting were not heat-processed before hydrolysis. Both native and time caused a linear (P = 0.006) increase in the DH denatured proteins can remain in solution and this depends max of RSM (Table 2). The DH decreased 9% after the on their concentration and on their extent of aggregation max initial 20 min of toasting and subsequently gradually [19, 20]. Due to their high flexibility, native proteins in increased with increasing toasting time, up to a level in solution are in dynamic equilibrium with their distorted the 120 min-toasted RSM similar to the untoasted RSM forms and these distorted forms could be considered as (Fig. 2a). denatured [19], which could make them more accessible Toasting time had no effect on the k or the DH of for enzymatic cleavage. Furthermore, the relatively small max the soluble protein fraction (Table 2), as can be seen from size of the protein aggregates present in the soluble frac- the similar shapes of the degree of hydrolysis curves tion allows them to stay in solution [10]. Therefore, the (Fig. 2b). The soluble proteins could either be native or lower degree of aggregation of the proteins in the solu- aggregated proteins with a molecular size that allows ble fraction as compared with the proteins in the insolu- them to stay in solution [10, 16]. Aggregation of proteins ble fraction probably facilitated enzymatic hydrolysis, in solution into spherical particles occurs when heating thus increasing k. This would mean that a decrease in the is performed at a pH close to the isoelectric point of the proportion of soluble to insoluble proteins after thermal protein [17]. With longer heating times, there is second- processing leads to a decrease in the k for the complete ary aggregation of these spherical particles [17] into pro- material, as observed for the RSM in this study. Another tein condensates [16], which results in the reduction of possibility for the differences in k between the soluble and protein solubility. Protein aggregation mechanisms for insoluble fractions is that the insoluble fiber matrix struc- proteins in solution have been clearly identified before tures present in the latter, and not in the soluble fraction, [16, 17]. We assume that proteins thermally treated under might limit and decrease the rate of protein hydrolysis. semi-dry conditions (e.g. toasting or autoclaving) follow The k of the insoluble protein fraction decreased similar aggregation mechanisms as reported for proteins linearly (P < 0.001) by 62% when comparing in solution. the untoasted RSM to the 120 min-toasted RSM 1 3 624 J Am Oil Chem Soc (2017) 94:619–630 Fig. 2 Degree of hydrolysis during 120 min hydrolysis of a rapeseed meals toasted for different times and b soluble and c insoluble protein fractions separated from these rapeseed meals (Table 2), whereas no effects were noticed on the Nitrogen Solubility of the Insoluble Fractions DH of these fractions. As 70–90% of the proteins max in RSM correspond to insoluble proteins, the pat- The N solubilities of the insoluble protein fraction in the tern of the hydrolysis curves of the insoluble protein sodium phosphate buffer pH 7.5, buffer with 2% (w/v) fractions (Fig. 2c) were, as expected, similar to those SDS, buffer with 10 mM DTT and buffer with 2% (w/v) from the RSM. It is possible that there is an increase SDS and 10 mM DTT decreased with increasing toasting in the size of protein aggregates formed at increasing time (Fig. 3a). All solvents show a reduction in their solu- toasting times, which can hamper the penetration of bilizing power with increasing toasting time, which might the hydrolytic enzymes. be related to the increase in chemical modifications of the In vivo, enzymatic protein digestion starts at the stom- residues (e.g. Maillard reactions or crosslinking). Non- ach with pepsin and is followed by secretion of trypsin covalent electrostatic interactions can be cleaved by salt and chymotrypsin at the small intestine, where most of solutions [22], such as the phosphate buffer used in this the digestion occurs. The length of the small intestine experiment. Furthermore, SDS can cleave hydrogen bonds and the transit time, however, are limited and the extent and hydrophobic interactions, whereas DTT can cleave of protein digestion probably also depends on their rate disulfide bonds [22]. Non-covalent bonds may be impor - of digestion. These factors might be even more important tant for the stability of the insoluble aggregates. The SDS for poultry than for pigs, due to the short digestive tract solution solubilized twice the amount of N than the phos- of the former [21]. phate buffer alone, while the increase of N solubility with 1 3 J Am Oil Chem Soc (2017) 94:619–630 625 Fig. 3 a Nitrogen solubility of the insoluble protein fractions in of the insoluble protein fractions relative to the phosphate buffer of 100 mM phosphate buffer and the buffer containing either 2% SDS, solutions containing either 2% SDS, 10 mM DTT, or 2% SDS and 10 mM DTT or 2% SDS and 10 mM DTT. b Increase in N solubility 10 mM DTT additional DTT was minimal (Fig. 3a). The increase in N decreased the lysine content by 23%, which is similar to the solubility relative to the solubility by the phosphate buffer reduction in lysine content observed in the present study is reported in Fig. 3b. The relative importance of non-cova- after 100 min of toasting, regardless of the initial contents. lent bonds (solubilized by SDS solution) for the stability of Conversion of fructoselysine during 6 M HCl hydrolysis the aggregates does not change with the increasing toast- yields furosine (32%), pyridosine (16%) and regenerated ing time (Fig. 3b). With increasing toasting time, there is an lysine (56%) [24]. The N -carboxymethyl-lysine (Fig. 4a), increase in the relative amount of protein solubilized by the furosine (Fig. 4c) and N -carboxyethyl-lysine (Fig. 4d) DTT-containing buffer (Fig. 3b). This indicates that there contents in the RSM linearly (P < 0.001) increased with is formation of disulfide bonds in the insoluble aggregates increasing toasting time. The formation of these com- with increasing toasting time. The SDS–DTT solution sol- pounds with increasing toasting time does not completely ubilized more protein than SDS or DTT containing solu- account for the reduction in the lysine content, which tions separately. Synergy of SDS and DTT for N solubility also decreases linearly (P < 0.001; Fig. 4e). Whereas the has been reported before [23] for extruded soy proteins. In lysine content decreases 15.7 μg/mg CP from the 0 min to addition, the relative amount of protein solubilized by this the 120 min-toasted RSM, the sum of fructoselysine (cal- ε ε solution increased with increasing toasting time (Fig. 3b). culated from furosine), N -carboxymethyl-lysine and N - We suggest that cleavage of non-covalent bonds by SDS carboxyethyl-lysine only increases 11.1 μg/mg CP in the exposes extra disulfide bonds that can subsequently be same toasting time range. Other Maillard-derived com- cleaved by the DTT. pounds (e.g. [5-hydroxymethyl]-2-furfural) that were not determined in this experiment were probably also formed Maillard Reaction Products during toasting, which could account for this difference. The content of lysinoalanine (Fig. 4b) decreased during Formation of early (fructoselysine) and advanced Mail- the initial 20 min of toasting, but remained constant with ε ε lard reaction products (N -carboxymethyl-lysine and N - increasing toasting times. Formation of lysinoalanine is carboxyethyl-lysine) in the RSM was noticed with increas- favored with an alkaline pH [25], which is unlikely to have ing toasting time, along with a decrease in the lysine been applied during toasting of the RSM. No lanthionine content (Fig. 4). The lysine contents in the present study could be detected in any of the samples. (<5.3 g/100 g CP) were lower than those reported previ- Most of the chemically modified compounds formed ously [4] for untoasted (6.0 g/100 g CP) and commercially were present in the insoluble protein fraction. The con- toasted (5.6 g/100 g CP) canola meals. The decrease in tents of furosine (Fig. 4c) and N -carboxyethyl-lysine lysine content (7%) reported in that study when comparing (Fig. 4d) increased linearly (P < 0.001) with increasing the untoasted to the toasted canola meals corresponds in toasting times, whereas there were linear (P = 0.006) the present study to a RSM toasted for 20 min. In a previ- and quadratic (P = 0.009) effects of toasting time on the ous study [2], increasing the toasting time from 0 to 93 min N -carboxymethyl-lysine (Fig. 4a) content. The content 1 3 626 J Am Oil Chem Soc (2017) 94:619–630 Fig. 4 a N -[carboxymethyl]-lysine (CML), b lysinoalanine (LAL), c for different times and the insoluble protein fraction (gray) separated furosine (Fur), d N -[carboxyethyl]-lysine (CEL) and e lysine (Lys) from these rapeseed meals contents (μg/mg crude protein) in the rapeseed meals (black) toasted of lysine (Fig. 4e) decreased linearly (P < 0.001) with after 2 ml consist of unspecified protein/peptides that bind increasing toasting time. Contents (μg/mg CP) of furo- to the column. ε ε sine, N -carboxymethyl-lysine, N -carboxyethyl-lysine The largest fraction of soluble proteins from the intact and lysine were higher in the insoluble protein fractions RSM elute at a molecular mass >10 kDa (Fig. 5a). Intact compared with the complete RSM. The removal of the soluble proteins and soluble protein aggregates can be soluble protein fraction probably concentrates the aggre- located in this region [27]. The proportion of material gated and chemically modified insoluble fraction. The >10 kDa decreased (linear P < 0.001, quadratic P = 0.04) presence of these Maillard-derived compounds in the with increasing toasting time (Table 3). This was also insoluble fraction could delay the enzymatic cleavage of reflected by the linear increase (P < 0.001) of the propor - the available peptide bonds due to steric hindrance [26]. tion of proteins/peptides <10 kDa with increasing toasting In addition to protein aggregation, as described above, time. This increase was probably the result of a higher rela- the presence of these chemically modified compounds tive representation of this highly soluble fraction compared in the insoluble fraction could also explain the observed with the decreasing contents of soluble intact proteins at reduction of the k of the insoluble protein fractions. longer toasting times. There were no indications of other chemical crosslinks Hydrolysis of the RSM and their soluble and insolu- occurring: lysinoalanine (Fig. 4b) contents in the insolu- ble fractions changed their elution profiles compared with ble fraction do not change with increasing toasting time, the soluble intact RSM, as higher proportions of material whereas no lanthionine could be detected in any of the can be determined at lower molecular masses (<10 kDa; samples. Table 3). In the hydrolysates of RSM, the proportion of material <10 kDa corresponded to 80–86% of the quan- Molecular Size Distribution tified area (i.e. excluding the a-specific binding, reten- tion volume higher than 2 ml). The proportion of peptides Toasting time influenced the molecular weight distribution >10 kDa in the hydrolysates of the RSM decreased after the of the proteins in solution before hydrolysis (Fig. 5a) and initial 20 min of toasting and was not largely affected by the peptides obtained after hydrolysis of the RSM (Fig. 5b), a further increase in toasting time. Increasing the toasting the soluble (Fig. 5c) and the insoluble (Fig. 5d) protein time increased the overall molecular weight of these hydro- fractions. The fractions in the chromatograms that elute lysates, as an increasing (linear, P < 0.001) proportion of 1 3 J Am Oil Chem Soc (2017) 94:619–630 627 Fig. 5 Size exclusion chromatograms of the a soluble fractions of fractions of the rapeseed meals toasted for different times. Vertical rapeseed meals toasted for different times before hydrolysis and the dashed lines represent the cut-off points of 10 and 1.5 kDa hydrolysates of b rapeseed meals, c soluble and d insoluble protein peptides 1.5–10 kDa and a decreasing (linear, P < 0.001) However, increasing the toasting time decreased (linear proportion of peptides <1.5 kDa were determined at higher P < 0.001, quadratic P = 0.005) the proportion of mate- toasting times compared with lower ones. There was a neg- rial >10 kDa (Table 3). The nitrogen concentration before ative correlation (r = −0.73, P < 0.01) between the propor- hydrolysis was similar for all the hydrolyzed soluble frac- tion of peptides 1.5–10 kDa and the k of hydrolysis of the tions. Therefore, the decrease in the proportion of mate- RSM, whereas a positive correlation (r = 0.84, P < 0.001) rial >10 kDa is not expected to be due to the decrease in was determined for the proportion of peptides <1.5 kDa solubility with increasing toasting times (as explained (Table 4). The proportion of peptides 1.5–10 kDa in the previously for the soluble intact RSM), but to the facili- RSM hydrolysates was also positively correlated (r = 0.56, tated hydrolysis of soluble intact proteins. This was also P < 0.05) to the DH , whereas a negative correlation reflected by the overall decrease in the molecular weight of max (r = −0.74, P < 0.01) was determined between the propor- the hydrolysates, as the proportion of peptides 1.5–10 kDa tion of peptides <1.5 kDa and the DH . (P = 0.002) and <1.5 kDa (P < 0.001) linearly increased max The hydrolysates of the soluble protein fraction con- with increasing toasting times. The increase in the propor- tained a larger proportion of material >10 kDa com- tion of peptides <1.5 kDa with increasing toasting time pared with the RSM hydrolysates (Table 3). This can be obtained after hydrolysis of the soluble protein fraction explained by the presence of intact soluble proteins, which can be explained by an increased denaturation of the sol- were detected by SDS–PAGE in the hydrolysates of the uble proteins, exposing cleavage sites that were initially RSM and the soluble protein fraction (results not shown). not accessible for the enzymes [28]. A positive correlation 1 3 628 J Am Oil Chem Soc (2017) 94:619–630 1 3 Table 3 Relative molecular weight distribution (%) of intact RSM toasted for different times and the hydrolysates of these RSM, and soluble and insoluble protein fractions separated from these RSM Toasting time Intact RSM Hydrolysates RSM Soluble protein fraction Insoluble protein fraction >10 kDa 10–1.5 kDa <1.5 kDa >10 kDa 10–1.5 kDa <1.5 kDa >10 kDa 10–1.5 kDa <1.5 kDa >10 kDa 10–1.5 kDa <1.5 kDa 0 min 52.3 12.8 34.9 19.2 21.4 59.5 33.6 26.4 40.0 11.7 32.0 56.3 20 min 50.9 9.0 40.1 13.1 24.3 62.6 27.8 27.8 44.4 7.0 35.8 57.2 40 min 49.5 10.1 40.5 12.7 25.7 61.6 24.9 28.1 47.0 7.5 36.6 56.0 60 min 48.6 9.8 41.6 13.3 28.6 58.1 22.8 30.4 46.8 9.1 37.3 53.6 80 min 45.2 10.3 44.5 12.6 28.7 58.7 20.3 30.6 49.1 11.2 38.3 50.5 100 min 42.4 11.3 46.3 13.7 30.7 55.6 19.2 31.3 49.5 13.1 39.3 47.6 120 min 38.3 12.4 49.3 14.0 30.5 55.5 18.8 30.7 50.5 15.6 39.6 44.7 SEM 1.4 0.4 1.2 0.5 0.9 0.8 1.3 0.5 0.9 0.9 0.6 1.3 P value c e f h i Linear <0.001 0.15 <0.001 0.34 <0.001 <0.001 <0.001 0.002 <0.001 0.003 <0.001 <0.001 a b d g j k l Quadratic 0.04 0.006 0.76 0.006 0.13 0.32 0.005 0.18 0.07 <0.001 0.02 0.002 Abbreviations: RSM, rapeseed meal; SEM, standard error of the mean a 2 2 Intact >10 kDa = 51.84 − 0.017 × time − 0.00,079 × time (R = 0.91) b 2 2 intact 10–1.5 kDa = 11.45 − 0.07 × time + 0.00066 × time (R = 0.63) c 2 Intact <1.5 kDa = 36.38 + 0.10 × time (R = 0.85) d 2 2 RSM >10 kDa = 16.97 − 0.13 × time + 0.00093 × time (R = 0.58) e 2 RSM 10–1.5 kDa = 22.83 + 0.073 × time (R = 0.76) f 2 RSM <1 kDa = 62.63 − 0.060 × time (R = 0.67) g 2 2 Soluble >10 kDa = 32.81 − 0.23 × time + 0.00099 × time (R = 0.93) h 2 Soluble 10–1.5 kDa = 27.09 + 0.038 × time (R = 0.60) i 2 Soluble <1 kDa = 42.68 + 0.071 × time (R = 0.77) j 2 2 Insoluble >10 kDa = 10.10 − 0.093 × time + 0.0012 × time (R = 0.91) k 2 2 Insoluble 10–1.5 kDa = 32.74 + 0.11 × time − 0.00041 × time (R = 0.92) l 2 2 Insoluble <1 kDa = 57.16 − 0.014 × time − 0.00078 × time (R = 0.97) J Am Oil Chem Soc (2017) 94:619–630 629 Table 4 Pearson correlation Proportion of peptides after hydrolysis RSM Soluble protein Insoluble protein coefficients between the fraction fraction proportion of peptides after hydrolysis and the maximum DH k DH k DH k max max max degree of hydrolysis (DH ) max >10 kDa 0.16 −0.01 −0.25 −0.44 0.02 −0.77** and rate of hydrolysis (k) of the RSM and the soluble and 1.5–10 kDa 0.56* −0.73** 0.08 0.63* 0.44 −0.74** insoluble protein fractions <1.5 kDa −0.74** 0.84*** 0.32 0.25 −0.23 0.87*** separated from these RSM DH maximum degree of hydrolysis, k rate of protein hydrolysis, RSM rapeseed meal max Significance level: * P < 0.05, ** P < 0.01, *** P < 0.001 (r = 0.63, P < 0.05) was determined between the propor- contain peptides with larger sizes, in addition to intact pro- tion of peptides 1.5–10 kDa and the k of hydrolysis of the teins, can probably be considered to be less digestible com- soluble protein fraction (Table 4). pared with those with smaller sizes. In contrast to what was observed in the hydrolysates of the soluble protein fraction, increasing the toasting time increased the overall molecular weight of the peptides pre- Conclusion sent in the hydrolysates of the insoluble protein fraction (Fig. 5d). The proportion of material >10 kDa decreased The rate of protein hydrolysis of the soluble protein frac- after the initial 20 min of toasting, but steadily increased tion was 3–9-fold greater than that of the insoluble pro- with increasing toasting time (Table 3). This is not expected tein fraction. The decrease in the rate of hydrolysis of the to be due to the presence of intact proteins, as no clear RSM observed with increasing toasting time results from bands were detected by SDS–PAGE in these hydrolysates a combination of (1) the reduction in the proportion of fast (results not shown). The proportion of material >10 kDa hydrolysable soluble proteins to slowly digestible insoluble was negatively correlated to the k of hydrolysis (r = −0.77, proteins and (2) the decrease in the rate of hydrolysis of P < 0.01) of the insoluble protein fraction (Table 4). Fur- the insoluble proteins with increasing toasting time due to thermore, increasing the toasting time increased (linear the formation of disulfide bonds and/or chemically modi- P < 0.001, quadratic P = 0.02) the proportion of peptides fied amino acid residues. In addition, increasing the toast- 1.5–10 kDa and decreased (linear P < 0.001, quadratic ing time results in an overall increase of the size of the pep- P = 0.002) the proportion of peptides <1.5 kDa. There was tides after hydrolysis. Positive correlations were obtained a negative correlation (r = −0.74, P < 0.01) between the between the rates of protein hydrolysis of the RSM and the proportion of peptides 1.5–10 kDa and the k of hydrolysis insoluble protein fraction with the proportion of small pep- of the insoluble protein fraction, whereas a positive correla- tides (<1.5 kDa) after hydrolysis. tion (r = 0.87, P < 0.001) was determined with the propor- Acknowledgements The authors gratefully acknowledge the finan- tion of peptides <1.5 kDa (Table 4). cial support from the Wageningen UR “IPOP Customized Nutrition” The correlations between k and the size distribution of programme financed by Wageningen UR, the Dutch Ministry of Eco- the peptides in the hydrolysates of RSM and the insoluble nomic Affairs, WIAS, Agrifirm Innovation Center, ORFFA Addi- protein fraction could be explained by a shift of the hydro- tives BV, Ajinomoto Eurolysine s.a.s and Stichting VICTAM BV. We appreciate the support of Claire Butré with the UHPLC-MS measure- lytic mechanism from a more one-by-one type to a more ments. SSV acknowledges the support of the Universidad de Costa zipper-type-dominated system with increasing toasting Rica. time [19, 29]. Hydrolysis of most proteins shows an inter- mediate behavior between these two types of hydrolytic Compliance with ethical standards mechanisms [19]. In the one-by-one type of hydrolysis, the cleavage of peptide bonds in one protein is followed by a Conflict of interest The authors declare that they have no conflicts of interest. fast cleavage into smaller peptides. Overall, a higher pro- portion of intermediate peptides (Table 3) was determined in this study at short toasting times compared with long Open Access This article is distributed under the terms of the Crea- tive Commons Attribution 4.0 International License (http://crea- toasting times. In contrast, in the zipper-type hydrolysis, tivecommons.org/licenses/by/4.0/), which permits unrestricted use, several peptide bonds are cleaved simultaneously, which distribution, and reproduction in any medium, provided you give is followed by a slow conversion of large peptides into appropriate credit to the original author(s) and the source, provide a smaller peptides. The shift towards a zipper-type enzymatic link to the Creative Commons license, and indicate if changes were made. cleavage renders peptides of larger sizes, compared with the more one-by-one type of hydrolysis. Hydrolysates that 1 3 630 J Am Oil Chem Soc (2017) 94:619–630 14. Chabanon G, Chevalot I, Framboisier X, Chenu S, Marc I (2007) References Hydrolysis of rapeseed protein isolates: kinetics, characteriza- tion and functional properties of hydrolysates. Process Biochem 1. Jensen SK, Liu YG, Eggum BO (1995) The effect of heat treat- 42:1419–1428 ment on glucosinolates and nutritional value of rapeseed meal in 15. SAS Institute Inc (2011) The SAS system for Windows. SAS rats. Anim Feed Sci Technol 53:17–28 Institute Inc., Cary, NC, USA 2. Eklund M, Sauer N, Schöne F, Messerschmidt U, Rosenfelder P, 16. Amin S, Barnett GV, Pathak JA, Roberts CJ, Sarangapani PS Htoo JK, Mosenthin R (2015) Effect of processing of rapeseed (2014) Protein aggregation, particle formation, characterization under defined conditions in a pilot plant on chemical composi- and rheology. Curr Opin Colloid Interface Sci 19:438–449 tion and standardized ileal amino acid digestibility in rapeseed 17. Nicolai T, Durand D (2013) Controlled food protein aggregation meal for pigs. J Anim Sci 93:2813–2825 for new functionality. Curr Opin Colloid Interface Sci 18:249–256 3. Salazar-Villanea S, Bruininx EMAM, Gruppen H, Hendriks 18. Tonheim SK, Nordgreen A, Høgøy I, Hamre K, Rønnestad I WH, Carré P, Quinsac A, Van Der Poel AFB (2016) Physical and (2007) In vitro digestibility of water-soluble and water-insoluble chemical changes of rapeseed meal proteins during toasting and protein fractions of some common fish larval feeds and feed their effects on in vitro digestibility. J Anim Sci Biotechnol 7:62 ingredients. Aquaculture 262:426–435 4. Newkirk RW, Classen HL, Scott TA, Edney MJ (2003) The 19. Adler-Nissen J (1976) Enzymatic hydrolysis of proteins for digestibility and content of amino acids in toasted and non- increased solubility. J Agric Food Chem 24:1090–1093 toasted canola meals. Can J Anim Sci 83:131–139 20. Shivu B, Seshadri S, Li J, Oberg KA, Uversky VN, Fink AL 5. Hulshof TG, Bikker P, van der Poel AFB, Hendriks WH (2016) (2013) Distinct β-sheet structure in protein aggregates deter- Assessment of protein quality of soybean meal and 00-rapeseed mined by ATR–FTIR spectroscopy. Biochemistry 52:5176–5183 meal toasted in the presence of lignosulfonate by amino acid 21. Thorpe J, Beal JD (2000) Vegetable protein meals and the effects digestibility in growing pigs and Maillard reaction products. J of enzymes. In: Bedford MR, Partridge GG (eds) Enzymes in farm Anim Sci 94:1020–1030 animal nutrition. CABI Publishing, Oxfordshire, pp 125–143 6. Newkirk RW, Classen HL, Edney MJ (2003) Effects of prepress- 22. Liu KS, Hsieh FH (2008) Protein-protein interactions during solvent extraction on the nutritional value of canola meal for high-moisture extrusion for fibrous meat analogues and compari- broiler chickens. Anim Feed Sci Technol 104:111–119 son of protein solubility methods using different solvent systems. 7. Thacker PA, Newkirk RW (2005) Performance of growing- J Agric Food Chem 56:2681–2687 finishing pigs fed barley-based diets containing toasted or non- 23. Liu KS, Hsieh FH (2007) Protein-protein interactions in high toasted canola meal. Can J Anim Sci 85:53–59 moisture-extruded meat analogs and heat-induced soy protein 8. Gerrard JA, Lasse M, Cottam J, Healy JP, Fayle SE, Rasiah I, gels. J Am Oil Chem Soc 84:741–748 Brown PK, BinYasir SM, Sutton KH, Larsen NG (2012) Aspects 24. Krause R, Knoll K, Henle T (2003) Studies on the formation of of physical and chemical alterations to proteins during food pro- furosine and pyridosine during acid hydrolysis of different Ama- cessing—some implications for nutrition. Br J Nutr 108(Suppl dori products of lysine. Eur Food Res Technol 216:277–283 2):S288–S297 25. Schwarzenbolz U, Henle T (2010) Non-enzymatic modifications 9. Olsen JV, Ong SE, Mann M (2004) Trypsin cleaves exclusively of proteins under high-pressure treatment. High Pressure Res C-terminal to arginine and lysine residues. Mol Cell Proteomics 30:458–465 3:608–614 26. Puigserver AJ, Sen LC, Gonzales-Flores E, Feeney RE, Whi- 10. Wang W, Nema S, Teagarden D (2010) Protein aggregation— taker JR (1979) Covalent attachment of amino acids to casein. 1. pathways and influencing factors. Int J Pharm 390:89–99 Chemical modification and rates of in vitro enzymic hydrolysis 11. Carbonaro M, Bonomi F, Iametti S, Cappelloni M, Carnovale of derivatives. J Agric Food Chem 27:1098–1104 E (1998) Aggregation of proteins in whey from raw and heat- 27. He R, He H-Y, Chao D, Ju X, Aluko R (2014) Effects of high processed milk: formation of soluble macroaggregates and nutri- pressure and heat treatments on physicochemical and gelation tional consequences. LWT Food Sci Technol 31:522–529 properties of rapeseed protein isolate. Food Bioprocess Technol 12. Butré CI (2014) Introducing enzyme selectivity as a quantitative 7:1344–1353 parameter to describe the effects of substrate concentration on 28. O’Loughlin IB, Murray BA, Kelly PM, FitzGerald RJ, Brodkorb protein hydrolysis. Laboratory of Food Chemistry, Wageningen A (2012) Enzymatic hydrolysis of heat-induced aggregates of University, The Netherlands, p 208 whey protein isolate. J Agric Food Chem 60:4895–4904 13. Pedersen B, Eggum BO (1983) Prediction of protein digestibility 29. Linderstrom-Lang K (1953) The initial phases of the enzymatic by an in vitro enzymatic pH-stat procedure. Z Tierphysiol Tier- degradation of proteins. Bull Soc Chim Biol 35:100–116 ernahr Futtermittelkd 49:265–277 1 3

Journal

Journal of the American Oil Chemists' SocietyPubmed Central

Published: Feb 11, 2017

There are no references for this article.