In Vitro Hydrolysis by Pancreatic Elastases / and II Reduces

In vitro hydrolysis by pancreatic elastases / and II reduces β-lactoglobulin antigenicity M Gestin, C Desbois, Le Huërou-Luron, V Romé, Gwenola Le Drean, T Lengagne, L Roger, F Mendy, P Guilloteau

To cite this version:

M Gestin, C Desbois, Le Huërou-Luron, V Romé, Gwenola Le Drean, et al.. In vitro hydrolysis by pancreatic elastases / and II reduces β-lactoglobulin antigenicity. Le Lait, INRA Editions, 1997, 77 (3), pp.399-409. hal-00929534

HAL Id: hal-00929534 https://hal.archives-ouvertes.fr/hal-00929534 Submitted on 1 Jan 1997

HAL is a multi-disciplinary open access L’archive ouverte pluridisciplinaire HAL, est archive for the deposit and dissemination of sci- destinée au dépôt et à la diffusion de documents entific research documents, whether they are pub- scientifiques de niveau recherche, publiés ou non, lished or not. The documents may come from émanant des établissements d’enseignement et de teaching and research institutions in France or recherche français ou étrangers, des laboratoires abroad, or from public or private research centers. publics ou privés. Lait (1997) 77, 399-409 399 © Eisevier/INRA

Original article

ln vitro hydrolysis by pancreatic elastases 1 and II reduces p-Iactoglobulin antigenicity

M Gestin1, C Desbois ', 1Le Huërou-Luron!" , V Romé 1, G Le Dréan ', T Lengagnc-, L Roger/, F Mendy", P Guilloteau '

1 Laboratoire du Jeune Ruminant, INRA, 65, rue de Saint-Brieuc, 35042 Rennes Cedex; 2 Nutrinov, 85, rue de Saint-Brieuc, 35042 Rennes Cedex; 3 l, Place de Béarn, Saint-Cloud, France

(Received 15 November 1996; accepted 19 December 1996)

Summary - Bovine whey proteins such as œ-lactalbumin and ~-Iactoglobulin are with bovine caseins the most commonly used proteins in infant formulas owing to their high nutritional value. How- ever, these cow milk components are not always weil tolerated and can induce allergies in infants. The purpose of this study was therefore to investigate the gastric (pepsin) and pancreatic (trypsin, chymotrypsin and elastases 1and II)enzymatic hydrolysis of ~-Iactoglobulin with the aim of finding a means to reduce its antigenicity. Elastases 1and IIwere first purified from porcine pancreatic acetone powder. After differential precipitation steps, elastases 1and IIwere separated by cation-exchange chromatography. The conditions regarding ~-Iactoglobulin hydrolysis by gastric and/or pancreatic enzymes were similar to those used for hypoallergenic milk preparations. Elastase II and to a lesser extent elastase I, were effective in enhancing ~-Iactoglobulin hydrolysis via a mix of pepsin, trypsin and chymotrypsin and in reducing the residual antigenicity ofhydrolytic products. The sarne hydrolytic percentage was observed when elastase 1or IIwere added, while the residual antigenicity was lower in the presence of elastase IIthan in the presence of elastase 1. The introduction of elastases in the pancreatic IIÙX can therefore be proposed to enhance the hydrolysis of cow milk components in hypoallergenic milk preparations.

~-Iactoglobulin hydrolysis / pancreatic elastases 1and II / residual antigenicity / hypoallergenie milk

Résumé - L'hydrolyse in vitro de la ~-lactogIobuline par les élastases 1 et II pancréatiques réduit son antigénicité. Les protéines sériques du lait de vache telles que l'a-lactalbumine et la ~-Iacto- globuline, de haute qualité nutritionnelle, sont avec les caséines bovines les plus couramment utili- sées dans les laits de remplacement. Toutefois, ces protéines laitières ne sont pas toujours tolérées et peuvent induire des réactions de nature allergique chez les nourrissons. Afin de réduire l'antigénicité

* Correspondence and reprints 400 M Gestin et al de la ~-lactoglobuline,des hydrolysespar des enzymes gastrique (pepsine)et pancréatiques (trypsine, chymotrypsine et élastases 1et II) sont réalisées. Une purification des élastases est effectuée à partir de poudre acétonique de pancréas de porc. Après les étapes de précipitation fractionnée, les élastases 1et II sont séparées par chromatographie échangeuse de cations. Les conditions d'hydrolyse de la ~- lactoglobuline par les enzymes gastrique et/ou pancréatiques sont celles utilisées dans l'industrie pour la fabrication des laits hypoallergéniques. L'élastase II et dans une moindre mesure, l'élastase 1augmententl'efficacité de l'hydrolyse de la ~-lactoglobulinepar le mélangepepsine, trypsine et chymotrypsine et réduisent l'antigénicité résiduelle de ses produits d'hydrolyse. Le même pourcentage d'hydrolyse est obtenu en additionnantl'élastase 1ou l'élastase II. Cependant,l'antigénicité résiduelle est plus faible en présence d'élastase II. Ainsi, l'adjonction d'élastases dans un mélange pancréatique pourrait améliorer l'hydrolyse des protéines dans la fabrication des laits hypoallergéniques. hydrolyse de la ~-lactoglobuline / élastases 1and II pancréatiques / antigénicité résiduelle / lait hypoallergénique

INTRODUCTION enzymes to reduce in vitro allergenicity of œ-lactalbumin and ~-lactoglobulin (Pahud et Bovine whey proteins such as œ-Iactalbu- al, 1985; Asselin et al, 1988, 1989). Heat min and ~-lactoglobulin are with bovine treatment of whey proteins has been shown caseins the most commonly used proteins to reduce antigenicity, but has been associ- in infant formulas owing to their high nutri- ated with a loss of available lysine (Hep- tional value. They differ qualitatively and/or pell et al, 1984; Jost et al, 1987). Accord- quantitatively from human milk proteins. ing to Jost et al (1987), combining selective In addition to a high casein level (80 vs 35% hydrolysis by specifie proteases with pro- of total protein in human milk), cow milk cessing or subsequent heat treatments contains 50% ~-lactoglobulin in whey pro- appeared to be a promising approach in teins. This globular protein is lacking in developing a hypoallergenic infant formula. human milk. These cow milk components However, very few studies have investi- are not always weIl tolerated, since about gated the action of pancreatic proteases other 3% of children under 2 years of age present than trypsin and chymotrypsin on milk pro- allergie reactions. The allergie reactions are tein. Elastase II, which specifically cleaves promoted by rapid absorption of incom- globular protein, could improve milk pro- pletely digested milk products. This partial tein hydrolysis. The hydrolysis rate of digestion may be explained by a low gas- bovine casein, œ-Iactalbumin and ~-lac- tric acidity during infancy along with the toglobulin by human cationic elastase (elas- high buffering capacity of milk (Mason, tase II-like) exceeds the hydrolysis rate by 1962) and possibly a deficient proteolytic human anionic and cationic trypsins or response, particularly as regards elastase II, anionic elastase (elastase I-like) (Jakobsson in the digestive tract of infants suffering et al, 1983). Elastase 1 hydrolyses both from cow milk allergy (Jakobsson et al, œ-Iactalbumin and ~-lactoglobulin at a mod- 1983). erate hydrolytic rate, producing large and very small peptides, respectively (Schmidt ln order to pre vent atopic disease, pro- and PolI, 1991). tein hydrolysates are now given as of birth for several months to an increasing number The purpose of this study was therefore to of infants (Rigo et al, 1994). Pep sin asso- investigate gastric (pepsin) and pancreatic ciated with trypsin or chymotrypsin hydrol- (trypsin, chymotrypsin and elastases 1and II) ysis constitutes an efficient combination of enzymatic hydrolysis of ~-lactoglobulin as ~-Lactoglobulin hydrolysis by elastases 401 a means of reducing p-lactoglobulin anti- Enzyme activity assays genicity. The purified elastase II is not available commercially; we therefore had to Protein content was measured as described by purify pancreatic elastase II simultaneously Lowry et al (1951). Elastase 1 (EC 3.4.21.36) with pancreatic elastase 1 in order to carry and II (EC 3.4.21.71) activity assays were carried out at 25 "C in 0.2 mol/L Tris-HCl buffer out our study. (pH 8.0) with 0.01 mol/L succinyl-L-alanyl- L- alany 1-L-alanine-p-nitroanilide (Suc- Ala3-pN A; L1385, Bachem AG, Budendorf, Switzerland) (Bieth et al, 1974) and 0.01 mol/L succinyl-L- MATERIALS AND METHODS alanyl-L-alanyl-L-prolyl-L-Ieucine-p-nitroanilide (Suc-Alaz-Pro-Leu-pNA; Ll390, Bachem AG, Purification of porcine elastases 1 and II Budendorf, Switzerland) (DelMar et al, 1980) as substrates, respectively. Elastase II substrate is known to be substantially hydrolyzed by both Porcine elastases 1 and II were purified as previ- chymotrypsin and elastase 1 (especially by the ously described for the isolation of human elas- latter) (DelMar et al, 1980; Largman, 1983). tases 1 and II (Largman et al, 1976) with slight Therefore, chymotrypsin (EC 3.4.21.2) assays modifications. Briefly, 4 g acetone powder, rep- were performed at 25 "C in 0.05 mollL Tris-HCI resenting 25 g pancreatie tissue, were suspended (pH 8.0) containing 0.02 mol/L CaClz using suc- in 0.01 mollL CaClz,1 mollL Tris-base (pH 8.0), ciny 1-L-alany 1-L-alany 1-L-proly 1-L-pheny lala- activated at 4 "C for 30 min with 1: JO (w:w) nine-p-nitroanilide (Suc-Alaz- Pro- Phe-pN A; trypsin (L-I tosylamide 2-phenylethyl S7388, Sigma Chemicals, St Louis, MO, USA) as chloromethyl ketone; T8642, Sigma Chemicals, substrate (DelMar et al, 1980). The resulting St Louis, MO, USA) and centrifuged for 15 min enzymatic units were expressed as u mol of at 20 000 g. The supematant was precipitated at p-nitroaniline released per min (Il.l). pH 5.1 with 6% acetie acid and centrifuged for 10 min at 33 000 g. The latter supematant was then subjected to ammonium sulfate precipitation up to 55% saturation. After centrifugation Sodium dodecyl sulfate (SDS) at 27 000 g for 10 min, the precipitate was resus- gel electrophoresis pended in 0.01 mollL sodium phosphate (pH 6.5) and dialyzed (retention threshold: 6000-8000 Da) Cation-exchange chromatographie samples were for 4 h against water and maintained ovemight analyzed by SDS-polyacrylamide gel elec- against the 0.05 mmol/L sodium phosphate trophoresis (SDS-PAGE) using a 14% gel under (pH 6.5) buffer. At the different purification steps, denaturing conditions and stained with the retained supematants and pellets were tested Coomassie blue according to Laemmli (1970). for protein content and enzymatie activities, as The molecular mass marker proteins (Pharma- indicated below. cia Biotech, calibration kit) used were: œ-lactalbumin (14 400 Da), soybean trypsin inhibitor The activated material obtained was filtered (20 100 Da), carbonie anhydrase (30 000 Da), through 0.45-flm filters and separated on a FPLC ovalbumin (43 000 Da), bovine serum albumin apparatus equipped with a cation-exchange col- (67000 Da) and phosphorylase b (94 000 Da). urnn (Mono-S, HR5/5, Pharmacia, St Quentin-en- Yvelines, France) equilibrated in 0.01 mol/L sodium phosphate (pH 6.5) buffer. Proteins were eluted with a linear gradient of 0.01 mollL up to Hydrolysis of p-Iactoglobulin 0.2 mollL sodium phosphate (pH 6.5) buffer at a flow rate of 1 mL/min for 50 min. The eluate The hydrolysis of bovine ~-Iactoglobulin (L3908; was monitored at 280 nm and the protein peaks Sigma Chemicals, St Louis, MO, USA) was measured for their protein content and potential assayed as indicated by Kahn et al (1990). enzymatic activities. The entire procedure was Hydrolytic conditions for whey protein from cow repeated several times and fractions containing milk were those used for hypoallergenie milk either elastase 1 or elastase II were pooled, dia- preparations. ~-Iactoglobulin was hydrolysed by Iyzed ovemight against water, freeze-dried, resus- gastric, pancreatie or gastric plus pancreatic pended in water and stored at -50 oc. enzymes. ~-Iactoglobulin was incubated with 402 M Gestin et al

pepsin (Bovipep, Sanofi, Beaune, France) at toglobulin. Two to three determinations of resid- 42 -c for 1 h at pH 2.6. The enzyme/substrate ual antigenicity were carried out for each (E/S) ratio was 1:5000 (w/w). Incubation of the hydrolysate. ~-Iactoglobulin solution (pH 8.0) was performed at 37 "C for 2 h 30 when pancreatic enzymes were added. This yie1ded an E/S ratio of 1:285 for the commercial pancreatic trypsin and chy- RESULTS AND DISCUSSION motrypsin mix (PEM 2500S, Novo-Nordisk, Bagsvaert, Denmark) and of 1:2500 for the puri- Purification of porcine pancreatic fied elastase 1 and/or elastase II. Samples were elastases 1 and II quickly frozen by immersion in liquid nitrogen to stop the reaction and were then stored at -50 "C until chromatographie analysis. For each type of The different purification steps for both elas- hydrolysis, a specifie control was obtained under tases 1 and II are summarized in table 1. A the same conditions of hydrolysis without l2.5-fold purification of elastase 1 and an enzyme addition. 8.3-fold purification of elastase II were finally achieved with a purification yield of 2.5 and 2.1 %, respectively. Similar degrees Reverse-phase chromatography of purification and yields were obtained by Largman et al (1976) for the purification of Hydrolysates were applied to a C 18 column human pancreatic elastases 1and II. The loss (Vydac TP54, IOOA, 4.6 mm, L 25 cm) equili- of elastase II activity during the purifica- brated in 0.1 % trifluoroacetic acid (TFA) solution procedure, particularly after ion- tion. Elution was performed with a linear gradi- exchange chromatography, could be par- ent of 0-80% acetonitrile in 0.1 % TFA at a flow rate of 1 mL /min for 1 h. The percentage of ~- tially due to the separation of elastase II lactoglobulin hydrolysis was determined by cal- from elastase 1 and chymotrypsin. In fact, culating the ratio of areas of the chromatographie the apparent elastase II activity of pancreatic peaks corresponding to non-hydrolysed ~-Iac- extracts during the early stages of purifica- toglobulin after and before hydrolysis. Proteins tion represented elastase II activity poten- and peptides were monitored at 210 nm. For each tiated to sorne degree by elastase 1and chy- enzymatic hydrolysis, three to six assays were carried out and analysed by chromatography. motrypsin (DelMar et al, 1980; Largman, 1983). Separation of porcine elastases 1 and II Radial immunodiffusion analysis was easily accomplished by cation-exchange chromatography, since the two enzymes had The antigenic property of hydro1ysates was mea- been eluted at different salt concentrations sured with the radial immunodiffusion technique from Mono-S. Porcine elastase 1 is more of Mancini et al (1965), using rabbit antiserum cationic than porcine elastase II; their pHi raised against bovine ~-Iactoglobulin (Levieux, calculated from amino-acid sequences are 1980). Non-hydrolysed ~-Iactoglobulin was used 8.45 and 7.43, respectively (Shotton and as standard. The lower limit of this method should lie at - 75 ug antigen. The variation Hartley, 1970; Kawashima et al, 1987). The between assays amounted to 2% when expressed elution profile from the cation-exchange as the standard deviation in per cent of the mean. chromatography is shown in figure 1. Five For each sample, the residual antigenicity of ~- major peaks can be distinguished (table II). lactoglobulin hydrolysate was calculated by We can consider that elastase 1 constituted removing the antigenicity due to non-hydrolysed the main form in peak 5 and elastase II in ~-Iactoglobulin. The percentage of non-hydrolysed ~-Iactoglobulin was determined as indi- peak 1. In peak 5, elastase 1activity is likely cated above. The residual antigenicity was to be due exclusi vely to the elastase 1 expressed as a percentage of the antigenicity of enzyme, which can be responsible for the the same amount of non-hydrolysed ~-Iac- measured elastase II activity (DelMar et al, I3-Lactoglobulin hydrolysis by elastases 403

o o ,....,o 00 1:-

0\ a1:-

00 'Cia

25 ,....,o

o ""l o 00 ,...., 00

c:o

o 0\ r'l

1:- o 00 a,....,

o o 00 'Ci 'Ci 'Ci 404 M Gestin et al

Fig 1. Elution profile of purified elastases 1 and M II from activated porcine pancreatic acetone pow- N _4 der on Mono-S column. Each peak was individ- ~I ually collected for activity determination. Pro- teins were eluted with a linear gradient of ;1 0.0 1 mollL up to 0.2 mollL sodium phosphate M (pH 6.5) buffer at a constant flow rate of ::i [,1 - 1 mL/min for 50 min. G indicates start of elu- 2 i~ tion gradient. 1 M l:e "~ i Profil d'élution des élastases 1 et II purifiées sur 1 !5 colonne Mono-S à partir d'une poudre acéto- 1 Il nique activée de pancréas de porc. Chaque pic a 1 1 ~ été collecté individuellement pour la mesure des activités enzymatiques. Les protéines sont éluées Il l' avec un gradient linéaire de 0,01 mol/L à Il 0,2 mol/L de tampon phosphate de sodium, pH i 6,5 à flux constant (l mUmin) pendant 50 min. G indique le début du gradient d'élution. ~

t G

Retention lime (min)

Table II. Protein content and specifie activities of the main peaks from Mono-S chromatography. Teneur en protéines et activités spécifiques des principaux pics issus de la chromatographie sur colonne Mono-S.

Peak no Prote in Specifie activities (lU/mg) a (mg/mL) Substrates

Suc-Ala3-pNA Suc-Ala ç-Pro-Leu-iitl A Suc-Ala2-Pro-Phe-pNA

1 8.5 75 10364 4207 2 6.8 47 1637 219 3 8.6 47 622 464 4 7.7 678 6554 1612 5 4.8 4316 8758 360

"Conditions of enzyme activity analysis have been detailed in Materials and methods. "Les conditions d'analyse des activités enzymatiques sont détaillées dans la partie Materials and methods.

1980; Largman, 1983). In peak l. elastase 1 tamination by chymotrypsin. Under the pre- activity was very low, indicating that it was sent conditions of chymotrypsin activity not eluted in this peak. The measured activ- analysis, commercial chymotrypsin demon- ity on the Suc-Ala.-Pro-Phe-nlc/v substrate strated a specifie activity on Suc-Ala.-Pro- can only be due to elastase II, which has a Phe-pNA equal to 27 000 VI/mg protein hydrolytic activity towards this substrate and an activity on Suc-Ala.-Pro-Leu-pfê A (DelMar et al, 1980) or possibly to con- substrate equal to 5600 VI/mg protein. ~-Lactoglobulin hydrolysis by elastases 405

Peak estimated to be - 27 500, which was in agreement with the theoretical molecular MW 1 5 234 weight of elastases 1 and II (Shotton and 94000 Hartley, 1970; Kawashima et al, 1987). 67000 When wells were overloaded, two slight 43000 bands migrating as chymotrypsin could be 30000 detected (data not shown) corroborating activity analysis (table II).

Enzymatic hydrolysis of ~.Iactoglobulin 20100 The chromatographie profiles of p-Iac- toglobulin hydrolysis by different enzymes 14400 or mix of enzymes are shown in figures 3 and 4. The percentage of hydrolysis obtained for the various hydrolysates and the residual antigenicity of these hydrolytic products are Fig 2. SDS-PAGE of 80 Ilg of peaks (1 to 5) from Mono-S chromatography on a 14% gel presented in table III. A striking difference stained with Coomassie blue. The molecular was observed in the pattern of peptidic frag- weight markers (MW) are indicated on the left ments formed by the action of each elas- side of the gel. The molecular weight of the elastase. The elution profile of the hydrolysate tases 1 and II was estimated at approximately produced by elastase 1 action showed one 27500. major peak of relatively high hydrophobie- Électrophorèse sur gel de polyacrylamide- bisacrylamide à 14 % en présence de SDS. de ity. In contrast, elastase II hydrolysis indi- 80 J1g de protéines des pics 1 à 5 de la chro- cated a large number of lower hydropho- matographie sur colonne Mono-S. Le gel a été bicity peptides in addition to a major peak of coloré au bleu de Coomassie. Les marqueurs de high hydrophobicity. The percentage of poids moléculaire (MW) sont indiqués sur la par- hydrolysis of ~-Iactoglobulin was 2-fold tie gauche du gel. Le poids moléculaire des élas- higher with elastase II than with elastase 1. tases 1et 11a été estimé à environ 27500. ln contrast, Schmidt and Poli (1991) indicated that only small peptides had been released from ~-Iactoglobulin when hydrol- Therefore, contamination by chymotrypsin ysis was performed with elastase 1. Hydrol- cou Id only be minor. Moreover, the three ysis of ~-Iactoglobulin with pepsin plus pan- substrates were not or insignificantly hydrol- creatic trypsin and chymotrypsin mix greatly ysed by trypsin. The other peaks did not dis- increased the percentage ofhydrolysis (table play such a high enzymatic activity towards III) and the number of peptide peaks (fig 4). elastase 1 and II substrates. These enzymatic hydrolysis resulted in the The purified elastases were then tested production of a greater variety of more for their homogeneity by SOS-polyacry- hydrophylic peptides. The action of pepsin lamide gel electrophoresis (SOS-PAGE) plus pancreatic trypsin and chymotrypsin using a 14% gel under denaturing condi- simultaneously with that of elastase 1 or II tions (fig 2). The migration pattern of peaks did not change the percentage of hydrolysis. 1 and 5 presented a single band. No con- However, the nature and the number of pep- tamination with chymotrypsin, trypsin or tides were modified. other proteins was observed for these peaks. Hydrolysis with elastase 1or II alone was The molecular weight of these bands was effective in reducing the residual antigenicity 406 M Gestin et al

Retention time (min)

Fig 3. Reverse-phase chromatography elution profile of ~-lactoglobulin in control (A) or after 2 h 30 proteolysis by purified elastase 1 (H) or elastase II (C) (E/S ratio was 1:2500). Flow rate was 1 mUmin in 0.1 % TF A solution. A linear gradient of 0-80% acetonitrile/O.I % TF A solution was applied over 1 h. Chromatographie en phase inverse de la f3-lactoglobuline contôle (A) ou après 2 h 30 de protéolyse par l'élastase I (B) ou l'élastase II (C) purifiées (le rapport E/S est de /:2500). La vitesse d'élution est de 1 mUmin avec une solution de TFA à 0,1 %. Un gradient linéaire de 0-80 % d' acétonitrile dans une solution de TFA à 0,1 % est réalisé pendant 1 h. ~-Lactoglobulin hydrolysis by elastases 407

Retention time (min)

Fig 4. Reverse-phase chromatography elution profile of ~-Iactoglobulin after 1 h hydrolysis by pepsin (E/S ratio was 1:5000) followed by 2 h 30 proteolysis with pancreatic trypsin and chymotrypsin mix (E/S ratio was 1:285) alone (A) or in the presence of purified elastase 1 (B) or elastase II (C) (E/S ratio was 1:2500). Flow rate was 1 mL/min in 0.1 % TF A solution. A Iinear gradient of 0-80% acetonitrile/O.I % TF A solution was applied over 1 h. Chromatographie en phase inverse de la {3-lactoglobuline après 1h d'hydrolyse par la pepsine (le rapport E/S est de 1:5000) et 2 h 30 de protéolyse par un mélange de trypsine et de chymotrypsine pancréatiques (le rapport E/S est de 1:285) seul (A) ou en présence d'élastase 1 (B) ou d'élastase II (C)purifiées (le rapport E/S est de 1:2500). La vitesse d'élution est de 1 mUmin avec une solution de TFA à 0,1 %. Un gradient linéaire de 0-80 % d'acétonitrile dans une solution de TFA à 0,/ % est réalisé pendant 1h. 408 M Gestin et al

Table III. Percentage of hydrolysis of p-Iactoglobulin and residual antigenicity in various hydrolysates (means ± SE). Pourcentage d'hydrolyse de la f3-lactoglobuline et antigénicité résiduelle des différents hydrolysats (moyenne ±SE).

Enzyme Percentage of hydrolysis Residual antigenicity"

Control lb 5.6 ± 0.7 100 Elastase 1 33.6 ± 6.3 60.5 ± 2.1 Elastase II 62.6 ± 8.5 42.2±6.0

Control Z'' 5.7 ±0.3 100 Pepsin 26.8 ± 1.2 44.4 ± 3.0 Pepsin + PEMd 86.0 ± 2.4 14.3 ±2.6 Pepsin + PEM + elastase 1 89.2 ± 2.5 9.9 ± 3.7 Pepsin + PEM + elastase II 89.3 ± 1.9 7.1 ± 1.1

"The residual antigenicity was expressed as percentage residual antigenicity of the non-hydrolysed [3-lactoglobulin, which was considered to be 100%; bspecific control carried out under the hydrolytic conditions for pancreatic enzymes without enzyme addition; cspecific control carried out under the conditions for gastric hydrolysis alone or followed by pancreatic hydrolysis without enzyme addition; dpEM: pancreatic enzyme (trypsin and chymotrypsin) mix. "L'antigénicité résiduelle est exprimée en pourcentage de l'antigénicité résiduelle de la {J-lactoglobuline non hydro- lysée qui est considérée égale à 100 % .. "contrôle réalisé dans les conditions d'hydrolyse des enzymes pan- créatiques, sans addition d'enzymes .."contrôle réalisé dans les conditions d'hydrolyse de l'enzyme gastrique seule ou suivie de l'hydrolyse par les enzymes pancréatiques, sans addition d'enzymes ..dpEM: mélange d'enzymes pan- créatiques (trypsine et chymotrypsine).

of hydrolytic products. The most important elastase 1or II was added to pepsin plus the reduction in residual antigenicity was pancreatic trypsin and chymotrypsin mix obtained by treatment with pepsin followed while the residual antigenicity tended to be by hydrolysis with pancreatic trypsin and lower in the presence of elastase II than in chymotrypsin rnix in the presence of elastase the presence of elastase 1. As suggested by II. The specificity of elastases 1 and II was Asselin et al (1989), the relative sensitivi- complementary to that of trypsin and chy- ties of the peptide bonds involved in the anti- motrypsin. Elastase 1 hydrolysis is largely genie sites of ~-Iactoglobulin depend on the limited to Ala-Ala and Ala-Gly bonds specificity of each enzyme. (Gertler et al, 1977). Elastase II preferen- tially cleaves the bonds formed between leucine, phenylalanine or tyrosine with CONCLUSION glycine or alanine (Gertler et al, 1977; DelMar et al, 1980). The presence of elastase ln conclusion, this study clearly demon- 1in pancreatic trypsin and chymotrypsin mix strates the importance and specificity of tended to reduce the residual antigenicity. elastase 1 and particularly of elastase II in The decrease was enhanced in the presence enhancing ~-Iactoglobulin hydrolysis and of elastase II. The percentage of hydrolysis in reducing the residual antigenicity of the of ~-Iactoglobulin was not the only param- hydrolytic products. However, as regards eter influencing the antigenicity. The same industrial application, the intact ~-Iac- percentage of hydrolysis was observed when toglobulin remaining in the hydrolysates ~-Lactoglobulin hydrolysis by elastases 409

should always be removed. This could be ment: a possible strategy for producing a hypoal- achieved by modifying the conditions of lergenic infant milk formula. Br J Nutr 51,29-36 enzyme hydrolysis (E/S ratio, hydrolysis Jakobsson l, Lindberg T, Benediktsson B (1982) ln vitro digestion of cow's milk proteins by duode- time, etc) and by removing ~-Iactoglobulin nal juice from infants with various gastrointestinal via ultrafiltration (Asselin et al, 1988). Nev- disorders. J Pediatr Gastroenterol Nutr 1,183-191 ertheless, this improvement in the hydrol- Jakobsson 1, Borulf S, Lindberg T, Benediktsson B ysis of ~-Iactoglobulin requires further (1983) Partial hydrolysis of cow's milk proteins by human trypsins and elastases in vitro. J Pedi- investigation to determine its efficiency atr Gastroenterol Nutr 2, 613-616 under production conditions for hypoaller- Jost R, Monti JC, Pahud JJ (1987) Whey protein aller- genie infant food formula. It would appear genicity and its reduction by technologie al means. that the hydrolysis of ~-Iactoglobulin and Food Technol41, 118-121 of œ-lactalbumin occurs at a considerably Kahn JM, Mendy F, Roger L (1990) Improvements in or relating to organic compounds. Patent EP lower rate when these are present in crude 0421309 rather than in purified form (Jakobsson et Kawashima l, Tani T, Shimoda K, Takigichi Y (1987) al, 1982 and 1983). Characterization of pancreatic elastase II-cDNAs: two elastase II mRNAs are expressed in human pancreas. DNA 6,163-172 Laemmli UK (1970) Cleavage of structural proteins ACKNOWLEDGMENTS during the assembly of the head of bacteriophage T4. Nature 227, 680-685 Thanks are due to J Quillet for the bibliographi- Largman C (1983) Isolation and characterization of cal information, T Efstathiu, P Lorthioir and rat pancreatic elastase. J Biol Chem 258, 2327- M Piot for technical advice, J Léonil for review- 2333 ing the manuscript and D Fontaine for the English Largman C, Brodrick JW, Geokas MC (1976) Purifi- revision of the manuscript. Financial grants to cation and characterization of two human pancreatic M Gestin from the Conseil régional de Bretagne elastases. Biochemistry 15, 2491-2500 and the Centre National Interprofessionnel Lowry OH, Rosebrough NJ, Farr AL, Randall RJ d'Economie Laitière are gratefully acknowl- (1951) Protein measurement with the folin phenol edged. reagent. J Biol Chem 193, 165-175 Levieux 0 (1980) Heat denaturation ofwhey proteins. Comparative studies with the physical and immuno- logical methods. Ann Res Vet II, 89-97 REFERENCES Mancini G, Carbonara AO, Heremans JF (1965) Immunochemical quantitation of antigens by single Asselin J, Amiot J, Gauthier SF, Mourad W, Hebert J radial immunodiffusion. Immunochemistry 2, 235- (1988) Immunogenicity and allergenicity of whey 254 protein hydrolysates. J Food Sei 53,1208-1211 Mason S (1962) Sorne aspects of gastric function in Asselin J, Hebert J, Amiot J (1989) Effects of in vitro the newbom. Arch Dis Child 37,387-391 proteolysis on the allergenicity of major whey pro- Pahud JJ, Monti JC, Jost R (1985) Allergenicity of teins. J Food Sei 54,1037-1039 whey protein: its modification by tryptic in vitro Bieth J, Spiess B, Wermuth CG (1974) The synthesis hydrolysis of the protein. J Pediatr Gastroenterol and analytical use of a highly sensitive and conve- Nutr4,408-413 nient substrate of elastase. Biochem M ed II, 350- Rigo J, Salle BL, Putet G, Senterre J (1994) Nutri- 357 tional evaluation of various prote in hydrolysate DelMar EG, Largman C, Brodrick JW, Fassett M, formulae in term infants during the first month of Geokas MC (1980) Substrate specificity ofhuman life. Acta Paediatr 83, 100-104 pancreatic elastase 2. Biochemistry 19,468-472 Schmidt DG, Poli JK (1991) Enzymatic hydrolysis of Gertier A, Weiss Y, Burstein Y (1977) Purification whey proteins - hydrolysis of o:-Iactalbumin and and characterization of porcine elastase II and inves- ~-Iactoglobulin in buffer solutions by proteolytic tigation of its elastolytic specificity. Biochemistry enzymes. Neth Milk Dairy J 45, 225-240 16,2709-2716 Shotton DM, Hartley BS (1970) Amino-acid sequence Heppell LMJ, Cant AJ, Kilshaw PJ (1984) Reduction of porcine pancreatic elastase and its homologies in the antigenicity of whey proteins by heat treat- with other serine proteinases. Nature 225, 802-806