chemical methods for the characterization of proteolysis

HAL Id: hal-00929515https://hal.archives-ouvertes.fr/hal-00929515

Submitted on 1 Jan 1997

HAL is a multi-disciplinary open accessarchive for the deposit and dissemination of sci-entific research documents, whether they are pub-lished or not. The documents may come fromteaching and research institutions in France orabroad, or from public or private research centers.

L’archive ouverte pluridisciplinaire HAL, estdestinée au dépôt et à la diffusion de documentsscientifiques de niveau recherche, publiés ou non,émanant des établissements d’enseignement et derecherche français ou étrangers, des laboratoirespublics ou privés.

Chemical methods for the characterization of proteolysisin cheese during ripening

Plh Mcsweeney, Pf Fox

To cite this version:Plh Mcsweeney, Pf Fox. Chemical methods for the characterization of proteolysis in cheese duringripening. Le Lait, INRA Editions, 1997, 77 (1), pp.41-76. �hal-00929515�

https://hal.archives-ouvertes.fr/hal-00929515

https://hal.archives-ouvertes.fr

Lait (1997) 77, 41-76© ElseviernNRA

41

Review

Chemical methods for the characterizationof proteolysis in cheese during ripening

PLH McSweeney, PF Fox

Department of Food Chemistry, University College, Cork, Ireland

Summary - Proteolysis is the principal and most complex biochemical event which occurs duringthe maturation of most cheese varieties. Proteolysis has been the subject of much study and a rangeof analytieal techniques has been developed to assess its extent and nature. Methods for assessing pro-teolysis can he c1assified under two broad headings: non-specifie and specifie techniques, both of whichare reviewed. Non-specifie techniques include the quantitation of nitrogen soluble in various extrac-tants or precipitants and the Iiberation of reactive groups. Specifie techniques include those whichresolve individual peptides or free amino acids, principally chromatography or electrophoresis.Techniques are often combined into a fractionation scheme for cheese nitrogen to facilitate the iso-lation of individual peptides. Methods for the identification of individual peptides are also reviewed.Strategies for assessing proteolysis in various cheese varieties are discussed.

proteolysis / cheese ripening / methodology

Résumé - Méthodes chimiques pour la caractérisation de la protéolyse des fromages durantl'affinage. La protéolyse, qui est le phénomène principal et le plus complexe intervenant dans la matu-ration de la plupart des fromages, a été le sujet de nombreuses études. Toute une gamme de techniquesanalytiques a été développée afin de permettre une meilleure compréhension de ce phénomène. Lesméthodes de suivi de la protéolyse peuvent être classées en deux grands chapitres, traités dans cetterevue: les techniques non spécifiques, et les techniques spécifiques. Les techniques non spécifiquesincluent la quantification de l'azote soluble dans les divers extraits, et s'intéressent aussi à la libérationdes groupements réactifs. Les techniques spécifiques, principalement la chromatographie ou l'élec-trophorèse, sont celles permettant de séparer les peptides ou les acides aminés libres. Ces diversestechniques sont souvent combinées dans la stratégie de purification de la fraction azotée du fro-

Oral communication at the lOF Symposium 'Ripening and Quality of Cheeses', Besançon, France, February26-28, 1996.

42 PLH McSweeney, PF Fox

mage afin de pouvoir isoler individuellememt les peptides. Les méthodes d'identification des pep-tides, ainsi que les différentes stratégies de suivi de protéolyse dans diverses variétés de fromages sontdiscutées dans cette revue.

protéolyse 1 affinage du fromage 1 méthodologie

INTRODUCTION

The biochemistry of cheese ripening is normallyconsidered under three general headings: pro-teolysis, lipolysis and glycolysis. Of these pro-cesses, proteolysis is the most important for mostripened chee se varieties (Fox et al, 1995a). Pro-teolytic agents in cheese originate from fivesources: 1) the coagulant; 2) indigenous andendogenous proteinases from the milk; 3) thestarter; 4) the adjunct starter; and 5) non-starterlactic acid bacteria.

The coagulant

Chymosin and bovine pepsin in the case of calfrennet or recombinant chymosin or fungalaspartyl proteinases in cheeses manufacturedusing rennet substitutes are principally respon-sible for the production of large peptides (Visser,1977; McSweeney etai, 1994a).

lndigenous and endogenous proteinasesfrom the milk

The principal indigenous proteinase in milk isplasmin, but somatic cell proteinases andenzymes from the native microfiora of the milk(particularly heat-stable proteinases from psy-chrotrophs) are also present (Fox et al, 1995a).

The starter

Lactococcus spp, Lactobacillus spp and Strep-tococcus salivarius ssp thermophilus used asstarters for cheesemaking have extensive pro-teinase/peptidase systems (see Exterkate, 1995;

Bockelmann, 1995) which contribute to proteo-lysis in chee se, particularly at the level of theproduction of short peptides and free amino acids(Visser, 1977).

The adijun ct starter

Proteinases and peptidases from the adjunctstarter are particuarly important for proteolysisin certain chee se varieties. Penicillium spp andmicroorganisms from the smear of bacteria! sur-face-ripened cheeses produce pote nt extracel-lular proteinases and peptidases (Gripon, 1993;Reps, 1993).

Non-starter lactic acid bacteria

The microtlora of essentially ail chee ses is char-acterized by the growth of adventitious non-starter lactic acid bacteria, NSLAB (typicallyLactobacillus spp but also Pediococcus spp andperhaps other genera). The contribution ofNSLAB to cheese quality has been the subject ofmu ch recent research (see Peterson and Mar-shall, 1990; Martley and Crow, 1993;McSweeney et al, 1995, for references). Thecontribution of NSLAB enzymes to proteolysisin Cheddar cheese appears to be relatively minorand at the level of the production of short pep-tides and free amino acids (McSweeney et al,1994b; Lane and Fox, 1996).

The extent and nature of proteolysis showconsiderable intervarietal variation caused bydifferences in production protocols and ripen-ing conditions (particularly whether a secondarymicrotlora is encouraged). Proteolysis is very!imited in Pasta Filata varieties (eg, Mozzarella)due to denaturation of enzymes during high-

Methods used to assess proteolysis in chee se

temperature stretching, while in mould-ripenedvarieties (particularly blue mould cheeses) pro-teolysis is extensive due to the presence of patentfungal proteinases.

The products of proteolysis vary in size fromlarge peptides comparable to intact caseins (eg,<XsI-CNf24-199, the primary product of chy-mosin action on <Xsl-casein) to free ami no acidsand their degradation products. Proteolysis hasimportant consequences for the development ofcheese texture due to cleavage of the proteinmatrix (Creamer and OIson, 1982; Lawrence etal, 1983) and f1avour through the production ofthe precursors of a number of volatile f1avourcompounds and directly by the liberation ofamino acids and certain peptides (see Fox et al,1995a).

Because of its central role in cheese ripen-ing, techniques for the assessment of proteolysishave been reviewed (Rank et al, 1985; Fox,1989; IDF, 1991; McSweeney and Fox, 1993;Fox et al, 1995b). These reviews suggest thattechniques for the assessment of proteolysis incheese fall into two general categories: non-spe-cifie and specifie methods. Non-specifie meth-ods include determination of the nitrogen (N)soluble or extractable in various solvents or mea-surement of reactive groups, while specifie tech-niques are those which resolve individual pep-tides, ie, electrophoretic and chromatographiemethods. Non-specifie methods provide ade-quate information on the overall extent of pro-teolysis and the general contribution of the var-ious agents to proteolysis. Such techniques aregenerally simple and are valuable for the rou-tine assessment of cheese maturity. The thor-ough study of proteolysis necessitates the iso-lation and identification of peptides; thereforetechniques which resolve individual peptides,such as electrophoresis and chromatography,have received considerable attention in recentyears. The objective of this review is to assessmethodology for the study of proteolysis, butnot to simply catalogue methods used by spe-cifie authors. In the case of widely-used tech-niques, only a few recent representative refer-ences are supplied. If further references are

43

required, the reader is referred to the abovereviews.

NON-SPECIFIC METHODSFOR ASSESSING PROTEOL YSIS

Extraction / solubility methods

The caseins are insoluble in many solvents, butpeptides produced from them may be solubleand thus the proportion of soluble nitrogen willincrease with proteolysis. This is the principleof a number of widely-used methods for assess-ing proteolysis in chee se during ripening. Suchmethods are particularly useful for determiningthe extent of proteolysis, since peptides of vary-ing size can be precipitated by careful choice ofsol vents. A number of the solvent systemsdescribed below are also used to extract peptidesas the first step in their purification.

Hydraulic pressure/centrifugation

The aqueous phase of cheese ('cheese juice')can he prepared by subjecting a cheese/sand mix-ture to hydraulic pressure (eg, 30 MPa). Prepa-ration of chee se juice by hydraulic pressure hasthe advantage of not altering the ionie compo-sition of the aqueous phase of cheese and hasbeen exploited principally in studies on calciumphosphate in cheese, the buffering capacity ofcheese and in studies on the Iysis of starter cells.Authors who have used this technique includeMorris et al (1988), Wilkinson et al (1992),Lucey et al (1993), Wilkinson et al (1994) andSalvat-Brunaud et al (1995). Preparation of theaqueous phase of cheese by centrifugation hasthe same advantages. Guo and Kindstedt (1995)prepared the aqueous fraction of Mozzarella bycentrifugation at 12500 g for 75 min at 25 oc.

Urea

The proteins and large polypeptides in cheeseare completely soluble in 4 to 6 mol/L urea,which is used to prepare sampi es for analysis bychromatography or electrophoresis (see below).


Water

Water is perhaps the most commonly usedextractant for cheese N. The level of water-sol-uble N (WSN) is a widely-used index of cheeseripening (see Rank et al, 1985; Fox et al, 1995b)and several procedures have been developed(see Christensen et al, 1991). Water is frequentlyused as the primary extractant in the isolationof amino acids and peptides.

Kuchroo and Fox (1982a) compared variousextraction techniques for Cheddar cheese. Ailbut one of the several homogenization tech-niques studied yielded essentially similar results:a stomacher was chosen for routine work.Homogenization temperature (in the range5-40 "C) had \ittle effect on the level ofextractable N, which increased with the ratio ofwater to cheese; a ratio of 2: 1 was recommended,and 90% of the potentially water-extractable Nwas recovered in two extractions. The proce-dure recommended is: grated chee se is homog-enized in a stomacher at 20 "C for 10 min withtwice its weight of water; the slurry is held at40 "C for 1 h, centrifuged and the supernatant fil-tered through glass wool and Whatman no 113filter paper. The residue can be re-extracted toincrease the yield of extract. This and similarprocedures have been used by a number of work-ers (eg, Gonzàlez de Llano et al, 1991; Tiele-man and Warthesen, 1991; Wilkinson et al,1992; Bütikofer et al, 1993; Singh et al, 1995,1996).

The water-soluble extract (WSE) of Ched-dar cheese contains numerous small- andmedium-sized peptides, free amino acids andtheir degradation products, organic acids andtheir salts; extraction with water efficiently sep-arates the small peptides in cheese from pro-teins and large peptides. In many cheeses, theprincipal proteolytic agents responsible for theproduction of WSN are the coagulant (chymosin)and, to a lesser extent, plasmin (Visser, 1977;Fox et al, 1995a). Starter proteinases and pepti-dases play a relatively minor role in the hydro-Iysis of intact caseins and large polypeptides towater-soluble peptides, although they are active

on many of the peptides released by chymosin orplasmin. Water-insoluble peptides in Cheddarchee se have been studied by McSweeney et al(I994a), who found that the principal peptides inthis fraction were produced from u.sl-casein bychymosin or pepsin or from ~-casein by plas-min and therefore that the complementary pri-mary water-soluble peptides must be producedvia the action of these enzymes. This hypothesishas been strengthened by the results of Singh etaI (1994, 1995, 1996) who isolated and identifieda large number of water-soluble peptides fromCheddar cheese; with a few exceptions, thesepeptides corresponded directly to or were in theimmediate vicinity of cleavage sites for lacto-coccal cell envelope-associated proteinase butgenerally did not include the principal cleavagesites of chymosin on U.sl-or plasmin on ~-casein,suggesting that chymosin and plasmin act first.

WSN as a percentage of total N varies withvariety and increases throughout ripening (fig 1);a typical value for mature Cheddar cheese is-25%. Water is a suitable extractant for inter-naI bacterially-ripened cheese varieties, the pHof which changes Iittle during ripening. How-ever, mould and bacterial surface-ripened vari-eties are characterized by an increase in pH dur-ing ripening; and since the extractability of Nvaries with pH, an alternative method should beadopted.

Extraction with solutions containing salts

Extraction and quantification of cheese N solu-ble at pH 4.6 (pH 4.6-SN) is widely used as anindex of proteolysis. Since the pH of many inter-nai bacterially-ripened varieties is -5.2, there islittle difference between the levels extracted bywater or pH 4.6 buffers. However, in the caseof cheeses characterized by an increase in pHduring ripening, WSN may be considerablyhigher than pH 4.6-SN. Like WSN, pH 4.6-SNis produced mainly by rennet (O'Keeffe et al,1976) and increases during ripening (eg, fig 1).Whey proteins and proteose peptones (from plas-min action) are also soluble at pH 4.6, but theircontribution to pH 4.6-SN is relatively srnall;

Methods used to assess proteolysis in cheese

40..----.------------,

30

zt::z 20en;F.

10

0-- -0------0

10 20 30Ripening lime, weeks

Fig 1. Formation of pH 4.6-soluble N (.), 12%trichloroacetic acid-soluble N (e) and 5% phospho-tungstic acid (PTA)-soluble N (....) in an Irish arti-sanal blue chee se and formation of water-soluble N(0) and PTA-soluble N (0) in a commercial Ched-dar cheese. Modified from Folkertsma et al (1996)and Zarmpoutis et al (1996).Formation de N-soluble à pH 4,6 (_), N-soluble dansTCA 12% (e) et N-soluble dans PTA 5% (""), dans unfromage bleu irlandais d'origine artisanale etforma-tian de Nssoluble dans l'eau (0) et N-soluble dansPTA (0), dans un fromage de cheddar commercial.Modifié d'après Folkertsma et al (1996) et Zarmpoutiset al (1996).

y-caseins are insoluble at pH 4.6 (see Rank etal, 1985). According to Kuchroo and Fox(1982a), this method gives slightly lower val-ues for soluble N than extraction with water;and although it is more difficult to perform, itmay be easier to standardize (Fox, 1989). Recentstudies in which pH 4.6-SN has been used asan index of ripening include those by Lau et al(1991), Addeo et al (1992), Bütikofer et al(1993), Picon et al (1994), Yun et al (1995) andZarmpoutis et al (1996).

Precipitation of casein-derived peptides byCaCI2 has been used by sorne investigators.Solutions containing CaCI2 have been used asthe primary extractant or as a method to sub-fractionate WSN (Lau et al, 1991; Addeo et al,1992; Bütikofer et al, 1993). Noomen (I977a)

45

and Venema et al (1987) homogenized cheese in0,037 mol/L CaCI2, adjusted the homogenate topH 7.5, followed by centrifugation. Visser (1977)homogenized cheese in 4% NaCI containing0,55% Ca (as CaCI2). Fat was removed and thepH of the homogenate was adjusted to 7.5, fol-lowed by centrifugation. Noomen (1977b) useda method based on sensitivity to coagulation bypH and heat to fractionate cheese N. An extractwas prepared by homogenizing chee se in aCaCI2 (0.137 mollL)lNaCI (0.684 mollL) solu-tion; the homogenate was adjusted to pH 5.1.The extract was fractionated by acidificationwith 0.25 mol/L HCI to a final pH of -1 ,6 andholding at 55 "C for 30 min, and then overnightat room temperature before filtration.

Kuchroo and Fox (1982a) found that only40% of the WSN was soluble in 0.1 mol/LCaCI2. Increasing the concentration of CaCI2above 0.05% at or above pH 7.0 has Iittle influ-ence on extractability (Noomen, 1977b). Theincrease in CaClrsoluble N correlates with theage of cheese and the extracts contain whey pro-teins, peptides and amino acids, while the CaClrinsoluble fraction con tains caseins and largepeptides, similar to those in the water-insolublefraction (Christensen et al, 1991).

NaCI has also been used to fractionatechee se N (eg, Chakravorty et al, 1966; Gupta etal, 1974), but Reville and Fox (1978) claimedthat> 90% of Cheddar cheese N was soluble in5% NaCl, which was thus not sufficiently dis-criminating, ex ce pt perhaps for very youngcheeses. NaCI (5%)-soluble N and unfractionatedcheese are indistinguishable electrophoretically,indicating that the parent caseins, as weil as pep-tides, are extracted (Rank et al, 1985; Bican andSpahni, 199 J), Inclusion of CaCI2 in the NaCIsolution should reduce the percentage Nextracted and thus increase the selectivity of themethod (see Fox, 1989).

40

Extractionlfractionationwith organic solvents

Harwalkar and Elliott (1971) developed anextraction technique with chloroform/methanol

46 PLH McSweeney; PF Fox

(2: l, v/v). Freeze-dried cheese was extractedusing this solvent, the extract filtered and wateradded. The chloroform layer contained lipidsand was discarded. Removal of methanol fromthe aqueous methanol layer by evaporationresulted in the formation of a precipitate; thesupernatant was very bitter and astringent. Thismethod has also been used by a number ofauthors (eg, Lemieux et al, 1989, 1990; Puchadeset al, 1989a, b, 1990; Tieleman and Warthesen,1991) using different fractions of cheese as thestarting material.

Chloroform/methanol extracts more N thanwater, which is understandable considering therelatively hydrophobie nature of many casein-derived peptides, but chromatograms of bothextracts on Sephadex G-25 were similar (Ranket al, 1985). However, Tieleman and Warthe-sen (1991) found that extracts prepared by themethod of Harwalkar and Elliott (1971) weresignificantly less heterogeneous, as indicated byHPLC, than extracts prepared by the methodsof McGugan et al (1979) or Kuchroo and Fox( 1982a).

McGugan et al (1979) developed a fraction-ation sc he me using methylene chloride:methanol:water which has also been used bySmith and Nakai (1990).

Fractionation of proteins and peptides byethanol is a c1assical technique, and has beenused extensively to fractionate peptides in cheese(Christensen et al, 1991; McSweeney and Fox,1993; Fox et al, 1995b). Ethanol concentrationsused have varied from 30 to 80% (Fox et al,1995b); 70% has been used widely (eg, AbdelBaky et al, 1987). Trichloroacetic acid, TCA(12%, see below) and 70% ethanol yield approx-imately equal extraction levels (Reville and Fox,1978) and soluble fractions contain similar pep-tides (Kuchroo and Fox, 1982b), although dif-ferences are apparent in the peptides in the insol-uble fractions.

An ultrafiltration (UF) retentate (10 kDa) ofa WSE of Cheddar was fractionated by Breenet al ( 1995) using increasing concentrations ofethanol (30-80%); fractionation of the UF reten-tate containing 30% ethanol by stepwise adjust-

ment of pH in the range 5.5-6.5 was found tobe quite effective.

Precipitation with ethanol may be used tofractionate cheese or to sub-fractionate WSEs,and should be preferred to the largely equiva-lent 12% TCA owing to the ease of removal ofethanol by evaporation (Christensen et al, 1991).Fractionation of WSEs of Cheddar using 70%ethanol is used routinely in the authors 'laboratory.

Poznanski et al (1966) fractionated 2 or 12%TCA-soluble extracts with 75% acetone acidifiedto pH 4.6, while Aston and Creamer (1986) usedmethanol to fractionate a WSE of Cheddar.Butan-i-oi has been used to extract bitter pep-tides from casein hydrolysates (McSweeney andFox, 1993) but does not appear to have beenapplied to the fractionation of cheese. Thisextractant may have potential for the isolation ofbitter and/or hydrophobie peptides.

Fractionation using trichloroaceticor trifluoroacetic acids

Trichloroacetic acid (CCI3COOH; TCA) is ac1assical protein 'precipitant which has been usedwidely for this purpose. Concentrations rang-ing from 2 or 2.5% (eg, Reville and Fox, 1978;O'Sullivan and Fox, 1990) to 12% (eg, Bicanand Spahni, 1991; Folkertsma and Fox, 1992;Addeo et al, 1992; Bütikofer et al, 1993; Kat-siari and Voutsinas, 1994; Picon et al, 1994;Yun et al, 1995) have been used, depending onthe degree of fractionation required (Iarger pep-tides being soluble at the lower TCA concen-trations). There is no 'precipitation threshold'relating peptide size to solubility in TCA, butail peptides studied by Yvon et al (1989) con-taining less than seven amino acid residues weresoluble in 12% TCA; peptide solubility is relatedto hydrophobicity. The majority of 12% TCA-soluble peptides isolated and identified fromParmigiano-Reggiano cheese by Addeo et al(1994) were phosphopeptides. Typically, in amature Cheddar about 15% of the total N(Reville and Fox, 1978) and -90% of the WSN(Kuchroo and Fox, 1982b) is soluble in 2.5%


TCA; while 50-60% of a UF retentate (10 kDamembranes) and 100% of the UF permeate(0' Sullivan and Fox, 1990) is soluble in2% TCA.

Rennet is responsible for the production ofsorne of the 12% TCA-soluble N, but starterproteinases and peptidases make a substantialcontribution (Reiter et al, 1969; 0' Keeffe et al,1976) to the formation of 12% TCA-soluble N.Venema et al (1987) reported that the level of12% TCA-soluble N is a better index of maturitythan WSN. The development of 12% TCA-sol-uble N in an Irish artisanal blue cheese is shownin figure 1.

A major disadvantage of TCA for peptidefractionation is the need to remove it prior tofurther analysis of the fractions. Since small pep-tides and free amino acids will be lost on dialy-sis, other methods for removal of TCA, eg,repeated ether extraction or sorne form of chro-matography, are required and are laborious pro-cedures (Fox, 1989). The use of 70% ethanol,which gives similar precipitation levels, is prefer-able because ethanol can be readily removed byevaporation (Fox, 1989).

Trifluoroacetic acid (CF3COOH; TFA) as aprecipitant or extractant for cheese peptides maybe a useful alternative to TCA, and it has themajor advantage of being easily removed byevaporation. Perhaps surprisingly, there are fewreferences to its use in the fractionation ofchee se. Bican and Spahni (1991, 1993) used abuffer containing TFA as an extractant forAppenzeller cheese. The procedure involvedhomogenization at 4 "C in TFA (1%, v/v)/formicacid (5% v/v)/1 % (v/v) NaCI/I mol/L HCI at4 "C, fol1owed by centrifugation and applica-tion to a Sep-Pak Cl8 cartridge. The use ofTFAas an ion-pair reagent in HPLC buffers iswidespread (see below).

Phosphotungstic acid and similar techniques

It is often desirabJe to precipitate ail peptides froma cheese extract in order to quant ify free aminoacids in the cheese since nitrogen soluble in suchprecipitants is a crude measure of free amino

47

acids. Samples for amino acid analysis (see below)must also be rendered free of peptides.

Phosphotungstic acid (tungstophosphoricacid; 12W03.H3P04.xH10; PT A)-H1S04 is avery discriminating protein precipitant; only freeamino acids (apart from lysine and arginine),and peptides less than about 600 Da are solublein 5% PTA (Jarrett et al, 1982). PTA (1,2.5,5,6 or 6.5% )-soluble N has been used widely as anindex offree amino acids in cheese (eg, Wilkin-son et al, 1992; Bütikofer et al, 1993; Picon et al,1994; Guinee et al, 1995). The most widely-used PTA concentration is 5%. The short PT A-soluble peptides in sorne cheese varieties havebeen studied (Jarrett et al, 1982; Bican andSpahni, 1991; Gonzâlez de Llano et al, 1991).The development of 5% PTA-soluble N in Ched-dar and an Irish artisanal blue cheese is shown infigure 1.

5-Sulphosalicylic acid (2-hydroxy-5-sul-phobenzoic acid; SSA), a discriminating pro-tein precipitant, has been used at 3% to prepareextracts of cheese for amino acid analysis (Reiteret al, 1969; Ramos et al, 1987), or as an index ofamino acid N (Cliffe and Law, 1991). However,Kuchroo and Fox (1982a) found that 2.5% SSAprecipitated only 10% of the WSN. The water-soluble peptides in the fractions obtained withSSA do not appear to have been characterized.

Picric acid (2, 4, 6-trinitrophenol) is also avery discriminating protein precipitant (Fox,1989). Reville and Fox (1978) found that of themethods examined 0.85% picric acid-solubleex tracts of cheese contained the lowest levelsof N. It was considered to be the most suitableextractant for amino acids, but small peptidesare also soluble in picric acid (Salji and Kroger,1981). Picric acid interferes with the determi-nation of N by Kjeldahl or spectrophotometricmethods (Reville and Fox, 1978; Fox, 1989).

Free amino acids were extracted from Ched-dar by Hickey et al (1983) using Ba(OH)z/ZnS04' Cheese (lOg) was macerated in 33 mL0.15 mol/L Ba(OH)z and 33 mL 0.14 mol/LZnS04' filtered and residual fat removed byextraction with cholorform/ethanol (1: 1). Theaqueous phase was tïltered and the filtrate freeze-


dried and dissolved in a sodium citrate bufferprior to ami no acid analysis. No data were givenon extraction levels or whether peptides weresoluble. This reagent has not been used exten-sively to fractionate cheese N nor to preparesamples for amino acid analysis.

Miscellaneous techniquesfor fractionating peptides

Phosphopeptides (or other peptides with a highnegative charge) can he precipitated from a waterextract of cheese by addition of BaCI2 to a finalconcentration of 10 to 50 mmol/L and adjust-ing the pH to 6.7-7.0. Ba2+ complex with serinephosphate residues of phosphopeptides (andpotentially other negatively charged groups)which can then be precipitated by adding ethanol(20% v/v) and recovered by centrifugation(3000 g, 30 min, 4 "C), The pellet can be redis-solved in water and adjusted to pH < 2.0 usingH2S04 to precipitate BaS04' This method, basedon that of Peterson et al (1957), has been used inthe authors' laboratory to prepare phosphopep-tides from Cheddar. Kuchroo and Fox (l982b)fractionated a WSE of Cheddar with 0.1 mol/Lethylenediamine tetraacetic acid (EDTA);approximately 30% of the WSN was precipi-tated. However, EDT A does not appear to havebeen used by other authors for this purpose.

Fernândez and Fox (1997) fractionated pep-tides in the water-soluble and -insoluble frac-tions of Cheddar cheese using chitosan (a poly-mer of N-glucosamine). Chitosan is polycationicat acid pH and thus can complex with nega-tively-charged molecules, leading to their floc-culation. Chitosan (0.02%) was unsuitable forthe fractionation of water-insoluble peptides butfractionated water-soluble peptides quite effec-tively at pH 4.0. Chitosan-peptide complexeswere dissociated at pH 7.0.

Fractionation techniques basedon molecular mass

Techniques for the fractionation of peptidesbased on their molecular mass, ie, dialysis, ultra-

filtration and size-exculsion chromatographyhave been used as an alternative to techniquesbased on peptide solubility in various reagents.

Kuchroo and Fox (1982b) found that exhaus-tive dialysis (96 h, four changes) was a simpleand effective method for partitioning water-sol-uble peptides, and that it was suitable for rela-tively large samples. About 50% of the WSNin a mature Cheddar was dialysable; the dif-fusate was completely soluble and the retentate50% soluble in 70% ethanoI. This technique wasincluded by Kuchroo and Fox (1983b) in theirfractionation scheme. Sciancalepore and Alviti(1987) dialysed unfractionated, grated cheese(1 g), dispersed in H20 (1 mL), against H20(10 mL) at 30 "C for 30 min. The A28D of thedialysate correlated weil with age and with 12%TCA-soluble and pH 4.6-soluble N. This methodwas also used by Gonzâlez de Llano et al (1993)to study proteolysis in blue cheese.

While dialysis is a simple and useful tech-nique for peptide fractionation, it has been super-seded by ultrafiltration (UF) which is faster,capable of handling larger samples, uses mem-branes of known molecular mass eut-off andreduces the problem of recovering peptides froma large volume of diffusate. Many authors haveused diafiltration to improve resolution, Mem-branes with nominal molecular weight cut-offsranging from 0.5 to 10 kDa (most commonly)have been used. Most authors have used UF tofractionate the larger peptides in cheese, althoughVisser et al (l983a, b) isolated very low molec-ular weight peptides from a bitter extract ofcheese by UF on 0.5 kDa membranes, and Astonand Creamer (1986) used ultrafiltration with1 kDa eut-off membranes to fractionate WSN.Ali peptides in a 10 kDa UF permeate are solu-ble in 2% TCA, but sorne peptides in the reten-tate are insoluble (O'Sullivan and Fox, 1990); thepermeate contains no peptides detectable bypolyacrylarnide gel electrophoresis (PAGE)when stained using Coomassie blue G250, and40-50% of the WSN is permeable. UF was usedby Singh et al (1994, 1995) as the principal frac-tionation technique for water-soluble peptides.


Rejection of hydrophobie peptides by UFmembranes and aggregation of small peptidesare potential disadvantages of this method. How-ever, UF allows fractionation of large samplesand does not require solvents, both of whichfacilitate taste panel work (Fox, 1989). Despitesorne studies, the potential of UF membraneswith low molecular weight cut-offs « 1 kDa)to fractionate the smaller peptides from cheesehas not been extensively investigated.

The use of size-exc1usion chromatographyto fractionate cheese peptides is discussed below.

Comparison of techniques to quantifynitrogen in fractions

The majority of studies described above havequantified nitrogen in various fractions by themacro-Kjeldahl method (Wallace and Fox,1994). Although this method is highly repeat-able, it is a slow and potentially dangerous tech-nique. Alternative methods have also been usedfor this purpose, most notably by absorbanee ofultraviolet light at 280 nm (Vakaleris and Priee,1959) or the Folin-Ciocalteau reagent. The lat-ter method was used by Citti et al (1963) to mon-itorTCA-soluble peptides. The Lowry method (acombination of the Folin-Ciocalteau and biuretreagents), which is widely used in biochemistryand is simple and rapid, has been used to quan-tif Y peptides in various cheese fractions (eg,Kaminogawa et al, 1986). In the hands of a com-petent analyst, the Lowry method is suitable forthe routine analysis of proteinlpeptides in cheesefractions (Wallace and Fox, 1994). Accordingto Gonzâlez de Llano et al (1993), the Kjeldahland Lowry methods gave similar results for pro-teinlnitrogen in water-soluble extracts from bluecheeses. Protein can also be estimated by theBradford method (which is based on a colourchange in Coomassie blue on binding to pro-tein) or by using erythrosin, but preliminaryresults from this laboratory suggest that thesemethods are unsuitable for analysis of chee sefractions.

49

Methods based on the Liberationofreactive compounds or groups

Most of the techniques described above for thepreparation of various fractions and the quan-tification of N which is soluble therein are quitetime-consuming, and it is desirable to developmore rapid methods to estimate proteolysis. Suchtechniques are based on the liberation of spe-cifie compounds or reactive groups whieh areeasily measured. A wide range of such tech-niques is available, and inc1udes the estimationof Tyr and Trp in cheese fractions by absorbaneeat 280 nm or by the Folin-Ciocalteau reagent,protein-dye binding, reaction of liberated aminogroups with various reagents or the quantificationof ammonia or glutamic acid. Because of theirrelative simplicity, there is a growing demandfor such rapid methods for assessing cheesematurity (Ardô and Meisel, 1991).

Formation of ammonia

Ammonia, formed in cheese on the deamina-tion of free amino acids, is an important productof proteolysis in certain cheese varieties (eg,Camembert or smear-ripened varieties; Fox etal, 1995a) where it contributes to characteristicf1avour and texture (by neutralization of thecheese). Several authors (eg, Furtado and Chan-dan, 1985; Alonso et al, 1987; Zarmpoutis et al,1996) have indirectly monitored ammonia pro-duction by measuring the increase in the pH ofcheeses. However, it is perhaps surprising thatthe development of ammonia per se has not beenused more widely as an index of proteolysis.Ammonia levels in a non-protein fraction ofUlloa chee se were measured by Ordo nez andBurgos (1977) using the reaction of ammoniumsalts with the Nessler reagent (alkaline solutionof K2HgI4) with the formation of a red-browncolloidal precipitate (NH2Hg2I3) which was mea-sured spectrophotometrically; the concentrationof ammonia was - 4 mg g-l cheese dry matterand increased slightly during ripening. Fernan-dez-Salguero et al (1989) quantified ammoniain a number ofblue chee se varieties by isother-mal distillation from a sample dispersed in a


borate buffer, pH 10, followed by trapping inborie acid and back titration. An enzymatic assayfor ammonia is available (Anonymous, 1989)which quantifies NH; by the glutamate dehy-drogenase-catalyzed conversion of 2-oxoglu-tara te to L-glutamic acid, with the concomitantstoichiometric oxidation of NADH to NAD+which is determined spectrophotometrically.However, this technique does not appear to havebeen used in cheese analysis.

Formation of soluble tyrosine and tryptophan

Measurement of the 'soluble' tyrosine and tryp-tophan in cheese is a well-established methodfor assessing proteolysis. These amino acids canbe quantificd by the use of Folin-Ciocalteaureagent (acidic solution of Na tungstate and Namolybdate) or by measurement of absorbanceat 280 nm. lt is necessary to fractionate thecheese (eg, preparation of a water- or TCA-sol-uble extract) before using either of these tech-niques. Hull (1947) used the Folin-Ciocalteaureagent to assess proteolysis in milk. This reagentwas also used by Singh and Ganguli (1972) toquanti l'y peptides in the TCA-soluble fractionof cheese, but it appears to be less sensitive than2, 4, 6-trinitrobenzenesulphonic acid, TNBS(Samples et al, 1984). As described above, Vaka-leris and Priee (1959) measured the A280 of asodium citrate-HeI extract (pH 4.6) of cheese asan index of proteolysis.

Dye-binding

At pH values below their isoionic point, pro-teins have a net positive charge and can inter-act with anionic dyes (eg, amido-black, acidorange 12 or orange G) leading to the forma-tion of an insoluble protein-dye complex. Thiscomplex can be removed by centrifugation orfiltration and the excess dye in the supematant orfiltrate measured spectrophotometrically. Theam ou nt of dye bound by the protein is propor-tional to the concentration of protein (and thusthe excess dye is inversely proportion al to theprotein concentration). However, low molecularweight proteins and peptides react only slowly,

leading to poor separation and turbid filtrates orcentrifugai supernatants.

A decrease in the dye-binding capacity of adispersion of Cheddar cheese with age wasdernonstrated by Ashworth (1966). Kroger andWeaver (1979) reported that a dye-bindingmethod could be used as an index of proteolysisin cheese, but Kuchroo et al (1983) concludedthat the dye-binding method of McGann et al(1972), using amido black, is not suitable forthis purpose.

The Bradford method for protein determina-tion (Bradford, 1976) is based on the colourchange in Coomassie blue G250 when it reactswith proteins. However, Wallace and Fox (1994)found that this technique was not suitable forquantifying water-soluble peptides.

Acid/base titration

The formol titration technique is a simple methodfor estimating amino groups in milk and maybe used to measure the extent and rate of prote-olysis. The method involves titration of a samplewith NaOH to the phenolphthalein end-point,followed by the addition of formaldehyde whiehcon verts free amino groups to less basic sec-ondary and tertiary amines whieh dissociate at alower pH; therefore the solution becomes moreacidic, and it is necessary to re-titrate to the phe-nolphthalein end-point (Davis, 1965). This pro-cedure has been used to estimate proteolysis incheese (eg, Vakaleris et al, 1960; El Soda et al,1981) but il is now considered obsolete owing tovariations in the buffering capacity of cheese(Ardô and Meisel, 1991). Sciancalepore andLongone (1988) compared a dialysis method(Sciancalepore and Alviti, 1987) with formoltitration and the method of lnikhov (1951). Thelatter method involved homogenization of finelyground chee se in water, followed by filtration.Samples of the filtrate were diluted and titratedwith NaOH to thymolphthalein and phenol ph-thalein end points, the index of proteolysis beingthe difference in the titration values.

The buffering capacity of cheese increasesduring ripening owing to the formation of


ammonia, amino, imino and carboxyl groups.This principle was used by Ollikainen (1990) toassess proteolysis in a Swiss-type cheese; it wasclaimed that the method is rapid and convenientand as accurate as colorimetrie methods.Changes in the buffering capacity of Emmentalaround pH 9 have been attributed to proteoly-sis (Lucey et al, 1993).

Free amino groups-colorimetrieand f1uorimetrie methods

The c1eavage of a pepitde bond results in theliberation of a new amino group which can reactwith several chromogenic or fluorogenicreagents. A number of techniques have beendeveloped to assay proteolysis in cheese basedon this principle.

2,4,6- Trinitrobenzenesulphonic acid (TNBS)is one such reagent which reacts stoichiometri-cally with primary amines, producing a chro-

(a)HOOC-rNH'~

"'< 5& - "'\ 5!-.~""'"wNO: NO:

2 ... k6-Trinilro\>t"117ene StliphLllü"A,,;d

(b)0r\OH

~OH

o

Ninhydrin

Chromophore

51

mophore which remains attached to the aminoacid, peptide or protein and ab sorbs maximallyat 420 nm (fig 2a). The reaction is performed atan alkaline pH and stopped by lowering the pH.TNBS also reacts slowly with OH-, a reactionwhich is catalyzed by Iight. Since ammoniacalnitrogen produces only 20% of the absorbance ofamino groups when reacted in equimolar con-centrations with TNBS (Clegg et al, 1982), thisreagent may underestimate proteolysis in cheeseswhich have undergone significant deamination.Clegg et al (1982) concluded that althoughTNBS is not as sensitive as ninhydrin for assay-ing proteolysis in cheese, it is preferable owingto the simple analytical procedure; they pro-posed a correction factor for ammoniacal nitro-gen. A disadvantage of the TNBS method is thatthe dry powder is explosive, and prolonged stor-age leads to high blank values (Ardô and Meisel,1991).

(c)

oAuoresc:lmine (Ilon-tluoresc.onl/

AuorophoreHAHydrolysis Prudu<:ls tnon-tluoresceru r

(d)

»-Phrhaldialdehyde 1SDS. pH '.0

Fig 2. Reaction of (a) 2, 4, 6-trinitrobenzenesulphonic acid, (b) ninhydrin, (c) f1uorescamine and (d) o-phthal-dialdehyde with the œ-arnino group of an amino acid.Réaction de (a) l'acide 2,4,6 -trinitrobenrenesulphonique, (b) de la ninhydrine, (e) de la fluorescamine et (d)de l'o-phthaldialdéhyde avec le groupe a-aminé d'un acide aminé.


Habeeb (1966) and Adler-Nissen (1979) mea-sured the absorbance of the TNBS-amino groupcomplex at 340 nm, whereas in the procedureof Fields (1971) the absorbance of a sul-phite- TNBS-amino group complex was mea-sured at 420 nm, thus avoiding the maximumof absorbance of TNBS itself at 335-340 nm.Barlow et al (1986) reported that the latter ispreferable to the method of Habeeb (1966) inwhich an unstable coloured complex is formedand the procedure is liable to operator error.Alder-Nissen (1979) developed a procedure toassess the degree of hydrolysis of food proteinsusing TNBS; a linear relationship was foundbetween the concentration of œ-amino groupsand A420 but the relationship varied betweenproteins owing to variations in the concentra-tion of lysine.

The method of Hull (1947) was comparedwith the TNBS method by Samples et al (1984),who found that the latter was a better index ofproteolysis in cheese. Jarrett et al (1982) usedthe TNBS assay to quantify free amino acids in5% PTA-soluble fractions of cheese. The TNBSmethod was found to be reproducible for mon-itoring proteolysis in cheese (Kuchroo et al,1983) and while unfractionated cheese could beassayed, it is more sensitive if applied to pH 4.6-or 12% TCA-soluble extracts owing to a lowerbackground colour caused by reaction of TNBSwith e-amino groups. A strong correlation wasfound by Madkor et al (1987) between WSNand the A420 of the TNBS complex for water-soluble extracts of Stilton cheese.

Barlow et al (1986) claimed that analysis ofa water-soluble extract prepared according toKuchroo and Fox (1982a) by the TNBS methodof Fields (1971) would provide a simple rou-tine method for quantifying soluble N in cheese.Who le cheese, dispersed in 0.1 rnol/L Na2B407'0.1 mollL NaOH, pH 9.5, was analyzed by Poly-chroniadou (1988), who found a good correlationbetween the results of the TNBS method andWSN determined by the Kjeldahl procedure;fractionation of the cheese prior to analysis wasconsidered to be unnecessary. The TNBSmethod was also used by Lemieux et al (1990),

who compared a flow-injection technique forthe TNBS and o-phthaldialdehyde (see below)assays with c1assical methods. Bouton and Grap-pin (1994) found a good relationship betweenPTA-soluble N values (by Kjeldahl) and theTNBS method.

Ninhydrin is also widely used to monitor theIiberation of free amino groups in cheese dur-ing ripening. The principle of the ninhydrin reac-tion is the spectrophotometric estimation of achromophore formed when ninhydrin reacts withfree ami no groups (fig 2b). The purple chro-mophore, named Ruhemann' s purple after itsdiscoverer, does not remain attached to the pro-tein or peptide and therefore is not precipitatedduring procedures necessary to c1arify the sam-pIe prior to spectrophotometric analysis at570 nm (Ardë and Meisel, 1991), which is amajor advantage of the technique (Pearce et al,1988). The "max of the chromophore is shifted toa shorter wavelength (-507 nm) when the watercontent of the reaction mixture is reduced (Doiet al, 1981). Ninhydrin reacts with ammoniaalmost as readily as with amino groups andtherefore levels of proteolysis found by ninhydrinassays are consistently higher than those foundby the TNBS procedure, the discrepancy beingdue to ammonia (Clegg et al, 1982). Ninhydrinis more sensitive than TNBS, but the latter waspreferred by Clegg et al (1982) because the ana-Iytical procedure is simpler. Ninhydrin is widelyused to quantify amino groups in chromato-graphie eluates, particularly in amino acid anal-ysis by ion-ex change chromatography followedby post-column derivatization (see below). Cliffeet al (1989) used ninhydrin to monitor fractionsobtained by reversed-phase chromatography.

Pearce et al (1988) developed a Li-ninhy-drin assay for proteolysis in cheese based onthat of Friedman et al (1984): an aliquot ofcheese dispersed in a citrate buffer was mixedwith aqueous solutions of dimethyl sulphoxide,ninhydrin and lithium acetate at pH 5.2, heated,diluted and A570 determined. Total free aminoacids determined by this method correlated weilwith conventional amino acid analyses.


Doi et al (1981) described a number of nin-hydrin-based assays for peptidase activity; twowere modifications of the methods of Mooreand Stein (1948, 1954b) and two were based onthe Cd-ninhydrin assay of Tsarichenko (1966).The Cd-ninhydrin reagent was found to he moreselective for the amino group of free amino acidsthan for the amino groups of peptides or pro-teins and was the most sensitive of several nin-hydrin reagents, inc1uding Sn-ninhydrin. Folk-ertsma and Fox (1992) applied the Cd-ninhydrinassay of Doi et al (1981) to the assessment ofproteolysis in chee se; the reagent was found tobe about five times as sensitive as TNBS for themeasurement of amino acid nitrogen and couldbe performed on citrate-, water- or PT A-solu-ble fractions but not on TCA-soluble fractions,as the latter appeared to interfere with colourdevelopment. The release of total free aminoacids in cheese, as detennined by the Cd-nin-

2.0,-----------------,

1.5

Ec:....0 1.0"'<h.0ct

0.5

10 3020

Ripening lime, weeks

Fig 3. Formation of Cd-ninhydrin reactive aminogroups in Cheddar (.) and an Irish artisanal bluecheese (e) during ripening. Water- (Cheddar) orpH 4.6 (blue)-soluble N fractions were analyzed bythe method of Folkertsma and Fox (1992), modifiedfrom Zarmpoutis et al (1996).Formation des groupes aminés réactifs à la Cd-nin-hydrine dans un cheddar (.) et dans un fromage bleuirlandais d'origine artisanale (e) au cours de l'affi-nage. Les fractions N-solubles dans l'eau (cheddar) ouà pH 4,6 (bleu) ont été analysées par la méthode deFolkerstma et Fox (1992), modifiée à partir de Zarm-poutis et al (1996).

53

40

hydrin technique, is nonnally expressed as mgLeu g-l cheese (by reference to a leucine stan-dard curve) and increases during ripening (fig 3).The Cd-ninhydrin procedure of Folkertsma andFox (1992) is a simple, useful method for esti-mating total free amino acids in WSEs of cheeseand is used routinely in this laboratory; for bestresults, a set of samples should be analyzedtogether.

Fluorimetric reagents are also available forquantifying free amino groups. Such techniquesare more sensitive than the colorimetrie methodsdescribed above but have not been used aswidely. The HPLC separation of fluorescentderivatives of ami no acids and amines (pre-col-umn derivatization) is common (see below).

Fluorescaminc (4-phenylspiro [furan-2 (3H),1'-phthalan]-3,3'-dione) is used to quantify aminoacids and peptides in the picomole range. Fluo-rescamine, introduced by Weigele et al (1972),reacts with primary amino groups to produce afluorophore which is assayed at 390 nm excita-tion and 475 nm emission (fig 2c). It reacts atroom temperature with water or primary amines,the latter reaction being far faster (Udenfried etal, 1972). Fluorescamine has been used to mon-itor the hydrolysis of x-casein by chymosin(Pearce, ] 979; Beeby, 1980) and to quantifyacid-soluble proteins, peptides and ami no acidsin cheese extracts (Creamer et al, 1985). Theseauthors c1aimed that this reagent gave more con-sistent results than TNBS.

o-Phthaldialdehyde (OPA) reacts with 2-mer-captoethanol and primary amines to fonn a fluo-rescent complex (I-alkylthio-2-alkylisoindole;fig 2d), which also absorbs strongly at 340 nm.Church et al (1983), who used OPA to quantifyproteolysis in milk protein systems, reportedthat the method is more accurate than the mea-surement of A280 and is more convenient thanthe ninhydrin, TNBS or fluorescamine proce-dures. Lemieux et al (1990) considered the OPAmethod to be fast and simple for estimating freeamino acids in cheese and if the reaction timewas strictly controlled, was less variable thanthe TNBS method which was considered to betedious, required heating and the reagent was


1ight -sensiti ve. 9-Fluorenylmethoxy-carbony 1(FM OC) may also have potential for use incheese analysis as a fluorescent labelling reagentfor amino groups.

Enzymatic techniques

L-Glutamic acid is an important free amino acidin many cheese varieties. An enzyme assay forGlu using flow injection analysis and glutamatedehydrogenase immobilized on activated glasswas developed by Puchades et al (1989b);enzyme activity was determined based on thereduction of NAD+. These authors measuredfree Glu in Cheddar cheese extracts, preparedby the method of Harwalkar and Elliott (1971),which were freeze-dried, resuspended in a phos-phate buffer, sonicated, filtered and passedthrough a Sep-Pak C'8 cartridge prior to analysis.This technique was reported to be sensitive,rapid and accu rate.

McSweeney et al (1993a) measured free Gluin water extracts of Cheddar using a commer-cially available assay kit containing glutamatedehydrogenase (Boehringer-Mannheim,Mannheim, Germany). Unlike the procedure ofPuchades et al (1989b), in which NADH wasmeasured directly, this procedure uses pig-heartdiaphorase to catalyze the reaction of iodonitrote-trazolium chloride with NADH, forming for-mazan which is assayed spectrophotometricallyat 492 nm. The free Glu content of the chee sesincreased with age and varied between cheeseswith different non-starter populations.

Puchades et al (1990) developed an enzy-matie assay with flow injection analysis for totalfree amino acids in water-soluble extracts ofCheddar cheese, prepared by the method of Har-walkar and Elliott (1971) and filtered throughSep-Pak C'8 cartridges. L-Amino acid oxidase(immobilized on glass beads) acted on freeamino acids, producing stoichiornetric amountsof H202 which then reacted with luminol, achemiluminescent agent, in the presence ofK3Fe(CN)6' emitting light, which was quanti-fied. This technique was reported to be rapidand reproducible and specifie for free amino

acids, since peptides, which react to a greateror lesser extent in most other direct-assay pro-cedures for free amino acids, are not degraded byL-amino acid oxidase. However, this enzyme isnot equally active on ail amino acids; it exhibitshigh activity on Leu, Phe and Trp but His, Ileand Cys are poor substrates (Puchades et al,1990).

TECHNIQUES WHICH RESOLVEPEPTIDES

Although the above non-specifie techniques canprovide valuable information about the extentof proteolysis and the activity of proteolyticagents, they provide little information on whichpeptides accumulate or are degraded duringripening. For this reason, techniques whichresolve individual peptides have received muchattention in recent years.

Electrophoresis

Since only proteins and large peptides can bevisualized by staining, electrophoresis in gelmedia is Iimited to monitoring hydrolysis of theparent caseins and the formation and subsequenthydrolysis of the primary proteolytic productsof caseins and thus it has been used widely tomonitor the primary proteolysis of caseins incheese. More recently, it has been shown thatcapillary electrophoresis (CE) has considerablepotential for resol ving casein-deri ved peptideswith a wide range of molecular masses.

Although c1assical free-boundary techniqueshave been used (eg, Lindqvist and Storgards,1959), the overwhelming majority of elec-trophoretic analyses of cheese and other dairyproducts involve zonal electrophoretic tech-niques in which the electrophoresis buffer is sta-bilized in various media and the proteins andpeptides migrate as discrete bands under theinfluence of an electric field. Zonal elec-trophoresis on paper (eg, Lindqvist et al, 1953;Kuchroo and Fox, 1982b), porous cellulose


acetate or in agar or agarose gels are now rarelyused.

Electrophoresis in starch gels, which wasfirst applied to cheese by Melachouris andTuckey (1966), has been used by sorne work-ers to separate caseins or their degradation prod-ucts in cheese or other products (eg, Vreemanand van Reil, 1990; van den Berg and de Koning,1990; de Koning et al, 1992) and is quite effec-tive. Although starch gel electrophoresis gives farbetter results than earlier zonal techniques andMOIT(1971) c1aimed that it gave superior reso-lution of lower-mobility and cationic peptidesthan polyacrylamide gels, starch gels are brittleand opaque after staining (Creamer, 1991) andhave been largely replaced by polyacrylamidegels.

Electrophoresis in polyacrylamide gels, firstapplied to chee se by Ledford et al (1966), hasbecome the standard technique. The Iiteraturewas reviewed by Shalabi and Fox (1987), Fox(1989), Creamer (1991), Strange et al (1992),McSweeney and Fox (1993) and Fox et al(1995b ). Continuous buffer systems have beenused to separate caseins and are reported to havecertain advantages over discontinuous buffers(see Creamer, 1991). However, nearly ail one-dimensional PAGE techniques used in recentstudies on chee se have employed discontinuousbuffer systems containing urea or sodium dode-cylsulphate (SOS) as a dissociating agent. Non-denaturing buffers are not suitable for caseins,although they are used for the analysis of wheyproteins.

Sample preparation usually involves dis-solving cheese in a buffer (which usually con-tains a reducing agent, usually 2-mercap-toethanol) prior to electrophoresis; the solutionmay be centrifugally defatted, and sucrose orglycerol added to increase the density of thesample to facilitate loading into gel slots(Creamer, 1991).

Direct or indirect staining (followed bydestaining using acetic acid; eg, Laemmli, 1970)using Coomassie blue is usually used. Shalabiand Fox (1987), who compared a number ofstaining techniques for PAGE gels, recom-

55

mended the direct stammg procedure ofBlakesley and Boezi (1977) with Coomassieblue G250 in the presence of TCA, althoughamido black was preferred for larger peptides.However, since only relatively large peptidesstain under these conditions, the procedure islimited to the detection of the primary degrada-tion products in cheese. Q'Sullivan and Fox(1990) found that the peptides in a 10 kOa UFpermeate did not stain with Coomassie blue onurea-PAGE, but the retentate of the WSE con-tained several detectable peptides, as did 2%TCA-soluble and insoluble fractions of the reten-tate. Low molecular mass peptides can be visu-alized using a silver staining technique involvingextensive fixing with glutaraldehyde, althoughsuch stains have not been used widely to studycheese peptides.

Although urea-containing buffers at acid pHhave been used (eg, Creamer, 1991), the major-ity of authors have used urea-containing buffersat alkaline pH. Shalabi and Fox (1987) com-pared seve rai electrophoretic procedures for theanalysis of cheese and strongly recommendedthe stacking gel system of Andrews (1983) inTris buffers (pH 8.9) containing 4.5 mol/L urea.Electrophoresis in alkaline urea-containing gelswith direct staining by Coomassie blue G250 iswidely used in this and other laboratories tomonitor proteolysis in various cheese varieties(fig 4).

Electrophoresis in SOS-containing buffersis a standard technique for protein analysis ingeneral biochemistry; electrophoretic mobility inthe presence of SOS is inversely proportionalto the logarithm of the molecular weight of thepeptide. However, it is less widely used forcheese since the caseins have similar molecu-Jar weights (20 000 to 25 000) and are thereforenot as weil resolved by SOS-PAGE as by alka-line urea-PAGE. Shalabi and Fox (1987) con-cluded thatSOS-PAGE was inferior tourea-PAGE for cheese analysis, and did not rec-ommend its use. However, Creamer (1991) andStrange et al (1992) considered that SOS-PAGEprovides valuable information on cheese ripen-ing, and has been widely used (eg, Marshall et al,


Origin

Fig 4. Alkaline urea-polyacrylamide gel electrophoretograms (Andrews, 1983) of casein (lane 1), Cheddarcheeses made from high heat-treated milk (90 "C x 10 min) at one day and l, 2, 4 and 6 months of age (lanes 2---{)and water-soluble extracts therefrom (lanes 7-11). Direct staining using Coomassie brilliant blue 0250 was bythe method of Blakesley and Boezi (1977); (Folkertsma and Fox, unpubl results).Électraphorégrammes sur gel alcalin urée-polyacrylamide (Andrews, 1993) de caséine (bande 1). deframagescheddar fabriqués à partir de lait traité à température élevée (90 oC x 10 min) à 1 jour et 1. 2, 4, 6 mois (ban-des 2-6) et leurs extraits solubles dans l'eau (bandes 7-11). Fixation directe par du bleu brillant de CoomassieG250 avec la méthode de Blakesley et Boezi (1977); (Folkertsma et Fox. non publié).

1988; Basch et al, 1989; Bhowmik et al, 1990;Tunick et al, 1993). Shalabi and Fox (1987) con-sidered that the 20% acrylamide gel system ofTegtmeyer et al (1975) was more satisfactorythan the more widely used method of Laemmli(1970), in which 12% acrylamide gels are used.However, Creamer (1991) reported that 20%acrylamide gels give poor reso1ution of thecaseins (ail caseins migrate as two bands), but arevery satisfactory for peptides with molecularweights in the region of 10 000.

Two-dimensional electrophoresis has beenused by sorne authors; eg, Trieu-Cuot andGripon (1982) used SOS-PAGE in one dimen-sion and isoelectric focusing in the other to studyproteolysis in Camembert. Although 2-D elec-trophoresis may give good resolution, it is time-consuming and there are difficulties with repro-ducibility and in obtaining quantitative data(Creamer, 1991). Bican and Spahni (1991) sep-arated peptides in extracts from Appenzellercheese by thin-layer chromatography in onedimension, followed by electrophoresis at 90°.

Isoelectric focusing (IEF) is a powerful elec-trophoretic technique for resolving proteins andpeptides based on separation according to dif-ferences in their isoelectric points. It has beenparticularly valuable in studies on genetic poly-morphism in milk proteins (see Creamer, 1991;and Strange et al, 1992, for references). Sorneof the applications of IEF in dairy chemistryinclude the detection of adulteration of ovinemilk cheeses with bovine or caprine milks orthe study non-bovine caseins (Addeo et al, 1983,1990a,b; Conti et al, 1988; Moio et al, 1989,1992; Amigo et al, 1992). Trieu-Cuot and Gripon(1982) used IEF to study the pH 4.6-insolublefraction of Camembert. Kim and Jimenez-Flores(1994) obtained excellent resolution of milk pro-teins by preparative IEF followed by urea- orSOS-PAGE. This technique could easily beadapted for the study of proteolysis in cheese.Electrophoresis of cheese peptides has been facil-itated by the recent introduction of pre-pouredgels and by rapid semi-automated systems suchas the Phast system" (Pharmacia, Uppsala,


Sweden; Strange et al, 1992; Van Hekken andThompson, 1992).

After staining, electrophoretograms are usu-ally recorded photographically, although den-sitrometry or excision and elution of the stainedbands, followed by spectrophotometric quanti-tation have also been used. Difficulty in obtain-ing quantitative data is a serious limitation ofelectrophoresis. Creamer (1991) recommendedthat several control samples should be includedin each gel and that comparisons should he madeonly between samples on the same gel. Heemphasized that band dimensions are critical fordensitometry, and that dye uptake is a function ofthe protein as weil as staining and destaining pro-tocol. However, de Jong (1975) reported thatelectrophoresis in alkaline urea gels (pH 8.9),stained with ami do black lOB and scanned bydensitometry, gave good quantitative results.

Identification of peptides produced duringcheese ripening has posed difficulties. Peptidescan be isolated from PAGE gels by excision ofthe bands or by electroblotting. The latter tech-nique is preferable because of higher recover-ies of protein and because the size of PAGE gelscan change on staining which makes accurateexcision of unstained regions difficult. Elec-troblotted peptides can be identified by N-ter-minai amino acid sequencing (see below), butthey are more difficult to analyze by mass spec-trometry (MS) because they cannot be stainedprior to MS and must he eluted from the blottingmatrix. However, the location of ail the caseinsand sorne major degradation products (eg,y-caseins and asl-CN f24-199) on most PAGEsystems is known (Creamer, 1991; Strange etal, 1992). The majority of the peptides in aurea-PAGE electrophoretogram of Cheddarcheese were partially identified (fig 5) byMcSweeney et al (1994a). Addeo et al (1995)used immunoblotting (rabbit polyclonal anti-bodies raised against the caseins and peroxi-dase-Iabelled immunoglobulins as secondaryantibodies) to detect bands in electrophore-tograms of a number of cheese varieties. Thistechnique permitted the identification of thecasein from which a number of peptides origi-nated.

57

Fig 5. Urea-polyacrylamide gel electrophoretogramsof casein (eN) and the water-insoluble fraction (WISF)of 3-month-old Cheddar cheese made with Lactococ-eus lactis subsp lactis SKII, indicating the positionsof the caseins and the N-terminii of the polypeptidesidentified by McSweeney et al (1994); * undeterminedC-terminus.Électrophorégramme sur gel urée-polyacrylamide decaséine (CN) et de la fraction insoluble dans l'eau(WISF) d'unfromage de cheddar de 3mois fabriquéavec Lactococcus lactis subsp lactis SKI l, avec indi-cation des positions des caséines et des extrémitésN-terminales des polypeptides identifiés parMcSweeney et al (1994) .. * extrémité Csterminaleindéterminée ou non identifiée.

A technique described as high performanceelectrophoresis chromatography was used byGirardet et al (1994) to resolve component 3glycoproteins from bovine milk. Proteins wereseparated in a gel-fi lied column under the influ-ence of an electric field and separated proteinsexiting the bottom of the gel tube were elutedby a continuous flow of buffer and passedthrough a UV spectrophotometric detector. To


date, this technique does not appear to have beenused to study proteolysis but may have applica-tion in the fractionation of large peptides incheese.

Capillary electrophoresis (CE) resolves pep-tides in a buffer-filled capillary under the influ-ence of an electric field (fig 6) and as such is afree-boundary technique although, as describedbelow, zonal CE is also possible. CE separateson the basis of the net charge on the peptides,their mass and Stokes' radius (Young et al, 1992)and sometimes on sorne other property of thepeptide (see below). The use of CE in foodanalysis has been reviewed (Zeece, 1992;Lindeberg, 1996a,b).

Capillary electrophoresis is reported to havegreat potential for the resolution of complexmixtures of peptides. It has a number of advan-tages over traditional electrophoretic techniques,including the choice of running buffer and theuse of automated, high-performance instru-mentation. The composition of running buffercan be changed easily and separation times arerelatively short, although only one sample at atime can be analyzed. Unlike conventional elec-trophoresis, in which proteins and peptides in agel are visualized by staining, CE uses continu-ous monitoring by UV absorbance, and poten-tially by the range of other techniques used fordetection in HPLC. UV absorbance allows thedetection of small peptides which cannot be

visualized by staining and also means that CE isquantitative, unlike other electrophoretic tech-niques. It can be used in a number of modes,including free solution capillary electrophore-sis, FSCE (although problems have been encoun-tered with interations between peptides and thecapillary wall), micellar electrokinetic chro-matography (where electrophoresis is performedin the presence of micelles of SDS and peptidespartition between the micelles and buffer), cap-illary isotachophoresis (where the analyte is'sandwiched' between a leading and terminatingelectrolyte), capillary gel electrophoresis (whichuses a molecular sieving effect), capillary elec-trochromatography (which uses a capillarypacked or coated with a stationary phase) or iso-electric focussing (in which analytes may beeluted from the capillary by a pump after theyhave reached equilibrium at their pl) (Zeece,1992; Lindeberg, 1996a). There are also CEtechniques for resolving racemic mixtures bydiasteriomeric interaction of analytes with a chi-ral mobile or stationary phase (Gordon et al,1988).

The capital cost of CE instrumentation at pre-sent is higher than that of equivalent high per-formance liquid chromatographs (HPLC),although costs are likely to fall. Sample size inCE is extremely small and therefore the tech-nique is not suitable for preparative-scale work.CE may be limited to producing peptide pro-files, although CE coupled with in-line mass

Capillary electrophoretograrn

Capillary

Electrolyte butTer

Fig 6. Schematic diagram of a capillaryelectrophoresis unit.Représentation schématique d'uneunité d'électrophorèse capillaire.


1.S0E-02 T1.30E-02

1.10E-02

Ee 9.00E-03'Ol'...~<Ii 7.00E-03.aex:

S.00E-03

3.00E-03

1.00E-030 2 4 6

59

8 10Time min

12 14 16 18 20

Fig 7. Free solution capillary electrophoretogram of the water-soluble extract (WSE) from a 6-month-old Ched-dar cheese made using Lactococcus lactis subsp cremoris SKII as starter. WSE was dissolved in 1 mL8% (v/v) CH3CN in water containing 0.1 % (v/v) tritluoroacetic acid. Electrophoresis was performed using aBeckman Model P/ACE 2050 capillary electrophoresis unit (Beckman, San Ramon, CA, USA) equipped with a57 cm fused silica capillary (50 cm to detector; internai diameter, 75 um). Electrophoretic buffer was 0.1 mollLNa phosphate buffer, pH 2.5. Electrophoresis was performed al 20 kV and spectrophotometric detection was al214 nm (Singh, unpubl results).Capillaire en solution libre d'un extrait soluble dans l'eau (WSE) de fromage de cheddar de 6 mois fabriqué avecLactococcus lactis subsp cremoris SKI 1 comme levain. WSE était dissous dans 1 ml: de CH3 CN à 8 % (vlv) dansl'eau, contenant 0,1 % (vlv) d'acide trifluoracétique. L'électrophorèse était réalisée avec une unité d'élec-trophorèse capillaire Modèle Beckman PIACE 2050 (Beckman, San Ramon, CA, États-Unis) équipé d'un capil-laire en silice fondue de 57 cm (50 cm jusqu 'au détecteur; diamètre intérieur, 75 pm). Le tampon pour l'élec-trophorèse était un tampon phosphate de Na 0,1 mol/L, à pH 2,5. L'électrophorèse était réalisée à 20 kV et ladétection spectrophotométrique à 214 nm (Singh, non publié).

spectrometry gives much information on theidentity of peptides (Zeece, 1992). It is unlikelythat CE will replace chromatographie and otherelectrophoretic methods, and it is now viewed ascomplementing other separation techniques(Lindeberg, 1996a).

CE has received considerable attention ingeneral food analysis (Zeece, 1992; Lindeberg,1996a,b). To date, FSCE has been the form ofCE most widely used in dairy chemistry. Appli-cations have included ion analysis in milk, milkpowders and cheese (Morawski et al, 1993;Schmitt et al, 1993; Prestwell et al, 1995), mea-surement of hippuric and orotic acids in whey(Tienstra et al, 1992), fractionation of whey pro-teins (Cifuentes et al, 1993; OUe et al, 1994),measurement of sorbate and benzoate in chee se

slices and dips (Pant and Trenerry, 1995), detec-tion of chloramphenicol in milk (Blais et al,1994) and quantification of seleno amino acidsin human milk (Michalke, 1995).

The potential of this technique for isolatingpeptides from cheese was demonstrated by Zeece(1992), who separated peptides in a tryptic digestof native and dephosphorylated asl- and~-caseins. The authors are aware of a numberof laboratories investigating the potential of CEto study proteolysis in cheese, although noreports of capillary electrophoretograms ofcheese peptides appear to have been published todate. A free-solution capillary electrophore-togram of a water extract from a mature Cheddarcheese is shown in figure 7 (Singh, unpublresults).


Chromatography of cheese peptides

Paper chromatography (PC) was used by manyearly workers to quantify free amino acids incheese (eg, Kosikowsky, 1951a; Storgards andLindqvist, 1953; Clemens, 1954) or to studycheese peptides (eg, Q'Keeffe et al, 1978;Kuchroo and Fox, 1982b, 1983a). Paper chro-matography is cheap and straightforward, butsuitable only for assessing the complexity ofsystems containing only small peptides (Ardëand Gripon, 1991) and has been large1y super-seded by other techniques. Thin layer chro-matography (TLC) on silica gel, using differentsolvents, eg, n-propanol/water (70:30, v/v) orn-propanoUacetic acid/water (5: 1:3), has beenused to characterize peptides in cheese fractions(Kuchroo and Fox, 1982b, 1983a, b; Edwardsand Kosikowski, 1983; Visser et al, 1983b). Nin-hydrin is usually used to develop the plates,although UV fluorescence of the spots was usedby Edwards and Kosikowski (1983). Prepara-tive TLC has been used to isolate bitter peptidesfrom Cheddar (Edwards and Kosikowski, 1983).Bican and Spahni (1991) combined TLC on sil-ica gel plates (chloroforrnlammonialethanol assolvent) with electrophoresis in the perpendiculardimension to resolve peptides in Appenzellcheese: plates were developed by spraying withfluorescamine and viewed under UV Iight. TLCis a simple, straightforward method for separat-ing peptides and is normally superior to paperchromatography, although Kuchroo and Fox(1983a) obtained better resolution of peptidesby PC than by TLC. Both paper and TLC chro-matographic techniques are now rarely used.

Column chromatography on silica gels (eg,Visser et al, 1975), metal chelating media (eg,Cu-Sephadex; Ney, 1985; Mojarro-Guerra et al,1991) or on hydrophobie interaction media (eg,Kuchroo and Fox, 1983a; Visser et al, 1983b)have been used to fractionate peptides fromcheese or casein hydrolysates. Mulvihill and Fox(1979) used phosphocellulose to fractionate pep-tides produced from Ust-casein by chymosin, butthis medium does not appear to have been usedfor cheese. A few authors have used the molecular

sieving properties of cation-exchanger Dowex50 resins to fractionate cheese peptides (eg, Tokitaand Nakanishi, 1962; Huber and K1ostermeyer,1974; Biede and Hammond, 1979).

Mooney and Fox (unpubl results) fraction-ated peptides in the water-insoluble fraction ofCheddar cheese by high-performance hydropho-bic interaction chromatography on aPhenyl-Sepharose column in a 0.48 mol/L Naphosphate buffer, pH 6.3, containing 2.5 mol/Lurea; elution was by means of a gradient from0.48 to 0.037 rnol/L Na phosphate (fig 8).

0.4

0.3

= 15

~0.2 il'"

-c50

01

/

Fraction no.

Fig 8. Hydrophobie interaction chromatogram of thewater-insoluble fraction of a 20-week-old Cheddarcheese on Phenyl-Sepharose High Performance"(Pharmacia, Uppsala, Sweden). Elution was by a gra-dient of Na phosphate buffer from 0.48 (buffer A) to0.037 mol/L (buffer B), pH 6.3 to 6.6, containing2.5 mol/L urea and 0.1 % dithiothreitol. Flow rate was1 mL/min and UV spectrophotometric detection wasat 280 nm (Mooney and Fox, unpublished).Chromatogramme d'interactions hydrophobes de lafraction insoluble dans l'eau d'un fromage decheddar de 20 semaines sur Phenyl-Sepharose HighPerformance' (Pharmacia Uppsala, Suède). L'élu-tion était réalisée avec un gradient de tampon phos-phate de Na de 0,48 (tampon A) à 0,037 mol/L (tam-pon B), pH 6,3 à 6,6, contenant 2,5 mol/L d'urée et0,1 % de déthiothreitol. La vitesse d'écoulement étaitde 1 mUmin et la détection en spectrophotométrie UVétait à 280 nm (Mooney et Fox, non publié).

Methods used to assess proteolysis in cheese 61

The most popular forms of chromatographyfor analysis of cheese peptides have been ion-exchange and size-exclusion techniques andreverse-phase high performance liquid chro-matography (RP-HPLC).

Molecular weights may be estimated by size-exclusion chromatography on a calibrated col-umn. The majority of workers have usedSephadex" gels (Pharmacia, Uppsala, Sweden)of various pore sizes. Cheese preparations chro-matographed have included cheese or highmolecular weight peptides (Lindqvist, 1962;Tokita and Hosono, 1968; Creamer and Richard-son, 1974; Foster and Green, 1974; Gripon etal, 1975), water-soluble or similar extracts (Nathand Ledford, 1973; Desmazeaud et al, 1976;O'Keeffe et al, 1976; Santoro et al, 1987), dia-Iyzable or UF permeable fractions from waterextracts (Kuchroo and Fox, 1983b; Singh et al,1994, 1995) or PTA-soluble peptides (Gonzâlezde Llano et al, 1987, 1991).

Numerous eluents have been used, althoughwater and dilute buffers are most corn mon andchromatograms are usually quantified by spec-trophotometry at UV wavelengths. Generally,280 nm is used but for fractions containingsmaller peptides, absorbance of the carbonylgroup in the peptide bond at a lower wavelength,eg, 220 nm (Foster and Green, 1974; Green andFoster, 1974) is preferable, since small peptidesmay not contain aromatic residues. Measure-ment of the amino group by reaction with nin-hydrin (eg, Gripon et al, 1977) or sorne otherreagent is an alternative.

Low-pressure, preparative HPLC techniques(eg, FPLCTM, Pharmacia, Uppsala, Sweden) onsize-exc1usion media have simplified this formof chromatography. Wilkinson et al (1992)resolved press juice from Cheddar by high-per-formance size-exclusion chromatography(HPSEC) on a Superose-12 column; changes inthe peptide profile during ripening were evident.HPSEC on a Superose-12 column is also valu-able for characterizing peptides in various frac-tions from Cheddar (Breen et al, 1995) and issuitable for preparing peptide fractions. Haas-noot et al (1989) fractionated peptides in Gouda

chee se solubilized in urea by HPSEC on TSK3000SW and 2000SW columns in series. Thistechnique gave sorne information about primaryand secondary proteolysis during ripening, butthese authors considered that the size-exc1usioncolumns were not suitable for the separation ofthe components present in water- or TCA-solu-ble fractions. Vijayalakshmi et al (1986) reportedan interesting method for resolving peptides ofmolecular mass between 250 and 6000 Da on aTSK-SW 2000 column using 50 mmollL phos-phate buffer containing 0.1 % trif1uoroacetic acidand 35% methanol as mobile phase. HPSEC hasalso been used to characterize caseins, wheyproteins and peptides derived therefrom (vanHooydonk and Olieman, 1982; Bican and Blanc,1982; Vreeman et al, 1986).

Ion-exchange chromatography is used widelyto fractionate milk proteins but has only hadlimited use in cheese analysis (Fox, 1989).Anion-exchange chromatography on diethy-laminoethyl (DEAE)-cellulose was used to iso-late asl-CN f24-199 (asl-I casein) from Ched-dar (Creamer and Richardson, 1974) and tofractionate water-insoluble peptides (Breen,1992; McSweeney et al, 1994a; Mooney andFox, unpubl results) and the 2% TCA-solubleand insoluble fractions of a 10 kDa UF reten-tate of a water-soluble extract of Cheddar(0'Sullivan and Fox, 1990). High-performanceion-exchange columns (eg, Mono-S" or Mono-QTM,Pharmacia) give better resolution of bovinemilk proteins than c1assical chromatography onDEAE-eellulose (Barrefors et al, 1985; St Mar-tin and Paquin, 1990). Anion exchange chro-matography of a urea solution of Gouda cheeseon a Mono-QTM column was used by Haasnootet al (1989) to study proteolysis. It was con-c1uded that the ratio y- to ~-casein peak areas isa good indicator of the maturity of Gouda cheeseripened for up to about 10 months, whereas theratio of the peak areas representing asl-caseinand ast-CN f 24-199 was a suitable index forvery young cheeses « four weeks). Urea-con-taining buffers have usually been used to dis-solve cheese prior to chromatography and chro-matograms are monitored by A280, although


ninhydrin has been used by sorne workers (Foxet al, 1995b).

ln our experience, anion-exchange chro-matography on DEAE-cellulose or equivalenthigh-performance medium (eg, Mono-QTM) inurea-containing buffers is very suitable for thefractionation of large casein-derived peptides.This form of chromatography is used routinelyin this laboratory to fractionate the water-insol-uble peptides from cheese (fig 9; Mooney andFox, unpubl results). Fractions from the anion-exchange chromatogram can be collected andanalyzed by urea-PAGE.

The introduction of high performance ion-exchange chromatography (HPIEC) and HPSEC

% AU100 i 040

10.30

ml==- --' 0.00

Fig 9. High-performance anion-exchange chro-matogram of the water-insoluble fraction (WISN) ofCheddar cheese on a Mono-Q 5/5 column (Pharmacia,Uppsala, Sweden). WISN (500 ilL of 15 mg WISNmL-1 dissolved in 50 mmollL Tris-HCl buffer, pH 8.0,containing 4.5 mollL urea and 1.2 x 10-3 mol/L dithio-threitol, OTT) was applied to the column and eluted ata flow rate of 1 mL min-I using the above buffer con-taining 8 x 1()-4 mollL OTT and an NaCI gradient (0to 0.5 mollL). UV spectrophotometric detection was at280 nm (Mooney and Fox, unpubl results). .Chromatogramme d'échanges d'anion haute perfor-mance de la fraction insoluble dans l'eau (W/SN) defromage cheddar sur colonne Mono-Q 5/5 (Pharma-cia Uppsala, Suède). W/SN (500 pl: de 15 mg de WISNmL -t dissous dans un tampon Tris-HCI à 50 mmol/L,pH s,a, contenant 4,5 mol/L d'urée et 1,2 x 10-3 mol/Lde dithiotreitol, DIT) était injecté dans la colonne etélué avec une vitesse d'écoulement de 1 mlrmin:', enutilisant le tampon ci-dessus contenant S X /()-4 tnol/LDIT et un gradient NaCI (0 à 0,5 mol/L). Détection enspectrophotométrie UV à 2S0 nm ( Mooney et Fox,non publié).

has greatly reduced the workload involved inthese forms of column chromatography, whileincreasing speed and reproducibility. Thus, itseems likely that these techniques will achievemore widespread application in the future formonitoring proteolysis in cheese.

Reverse phase-HPLC has been used exten-sively to characterize peptides in caseinhydrolyzates (eg, Le Bars and Gripon, 1989;McSweeney et al, 1993b) and the technique isalso very valuable for resolving shorter peptidesin cheese fractions. Water-soluble extracts haveusually been used for RP-HPLC analysis (eg,Cliffe and Law, 1991; Gonzâlez de Llano et al,1991; McSweeney et al, 1993a) but other frac-tions have also been studied, including pH 4.6-soluble and -insoluble extracts (eg, Kamino-gawa et al, 1986; Christensen et al, 1989),10 kDa UF permeate (Singh et al, 1994),70%ethanol soluble and insoluble fractions (Zarm-poutis et al, 1996) and fractions from size-exclu-sion chromatography (Cliffe et al, 1993).

Gradient elution with water/acetonitrile (eg,Amantea et al, 1986) or water/methanol (eg,Cliffe et al, 1993) is usually used; but isocraticconditions using a phosphate buffer as eluenthave also been used by sorne workers (eg, Phamand Nakai, 1984). TFA is the most widely usedion-pair reagent. Detection is generally by UVspectrophotometry, usually at a wavelength inthe range of 200-230 nm (which measures thecarbonyl in the peptide bond), although 280 nmhas been used in cases where larger peptides,which are more likely to contain aromaticresidues, are expected (eg, Christensen et al,1989). Fluorescence detection has also foundlimited use (eg, Gonzâlez de Llano et al, 1991).Reverse-phase chromatography on a semi-preparative CZ/C1s-coated silica was used byCliffe et al (1989) to fractionate cheese peptides.

Although RP-HPLC essentially remains aresearch tool, Smith and Nakai (1990) discussedits potential for the routine assessment of chee sequality. Difficulties remain with the interpreta-tion of RP-HPLC data and the development ofa solvent system which will permit tasting ofthe fractions (Ardô and Gripon, 1991). As var-

0.20

1 0.10

Methods used ta assess proteolysis in chee se

ious columns, elution conditions and detectionwavelengths have been used, comparison ofchromatograms from different studies is diffi-cult. Although RP-HPLC is widely used topurify individual peptides from cheese (eg, Singhet al, 1994, 1995, 1996), the authors cautionagainst the identification of most peptides inchromatograms based solely on their retentiontime. However, two major peptides, as/-CNfI-9 and fI-13 (produced by the action of Lac-tococcus cell envelope-associated proteinase onast-CN fI-23, which is produced rapidly bychymosin) are characteristic of the chro-matograms of Cheddar cheese (fig 10a).

RP-HPLC is a very valuable technique forthe resolution of water-soluble peptides incheese, It hasbeen our experience that water-soluble extracts resolve weil on Cg wide-pore(300 À) columns with a shallow acetonitrilelwater gradient with TFA as the ion-pair reagent(eg, McSweeney et al, 1994b ). However, due tothe complexity of the WSE from many cheeses,il is often desirable to fractionate the WSE. Chro-matography of 10 kDa UF permeates is ade-quate but time-consuming. For this reason, wenow fractionate with 70% ethanol prior toRP-HPLC. Typical chromatograms of theethanol-soluble and insoluble fractions fromCheddar are shown in figure lOa, b (Tobin etal, unpubl results).

Elution of peptides from a C 19-reverse phasecartridge using a stepwise acetonitrile gradientwas used by Singh et al (1994, 1995) to resolvepeptides in a fraction obtained by size-exclu-sion chromatography of a 10 kDa UF permeateof a WSE of Cheddar cheese, and by Bican andSpahni (1993) to sub-fractionate extracts ofSwiss-type cheees. This technique gave consid-erably simpler sub-fractions from which indi-vidual peptides were isolated.

Two-dimensional HPLC (2D-HPLC) wasused by Lagerwerf et al (1995) to fractionatethe peptides in the WSE of Cheddar cheese; theeluate from the first dimension (ion-exchange)was directed to a reverse-phase column (C1g).The resolution obtained was impressive,although increased equipment requirements and

63

long analysis times (- 10 h) are likely to militateagainst the widespread use of 2D- HPLC.

QUANTIFICA TION OFFREE AMINO ACmS ANDTHEIR DEGRADATION PRODUCTS

Ultimately, the hydrolysis of caseins by the corn-bined action of proteinases and peptidases incheese leads to the liberation of free amino acidswhich are considered to be important flavourcompounds in man y cheeses; their concentra-tion varies widely with variety. In addition,volatile compounds formed from amino acidsby decarboxylation, deamination, transamina-tion and other transformations can make sub-stantial contributions to cheese tlavour (see Foxet al, 1995a).

It is generally necessary to deproteinize sam-pIes prior to amino acid analysis to reduce inter-ference from peptides; reagents used haveincIuded sulphosalicylic acid (Reiter et al, 1969;Omar, 1984), ethanol (Visser, 1977; Polychro-niadou and Vlachos, 1979), picric acid (Weaveret al, 1978; Shindo et al, 1980), TCA (Zarm-poutis et al, 1996) or Ba(OH)z/ZnS04 (Hickeyet al, 1983). An internaI standard (eg, norleucine)is normally added to facilitate quantitation ofamino acids.

Early attempts to quantify free amino acids incheese used paper chromatography developed

», using ninhydrin (eg, Kosikowsky, 1951a, b) butsuch techniques were only semi-quantitative.Likewise, classical ion-exchange column chro-matography (eg, Mabbitt, 1955, who usedDowex-50 resin) is now obsolete and has beensuperseded by automated amino acid analyzers,which are less laborious.

Dedicated ami no acid analyzers based onion-exchange chromatography are used widelyand have greatly facilitated the analysis of freeamino acids in chee se, which is now relativelysimple, accurate, quantitative and requires Iit-tle sample preparation. Amino acids are usuallydetected by post-column derivatization, oftenusing ninhydrin. Recent studies in which this

64

0.30

0.25

0.15

0.05

PLH McSweeney, PF Fox

ax

y

0.12

0.16·

0.02-

0.00-

0.'00

b

5.'00

Fig 10. Reverse-phase (Cs) high-perforrnance liquid chromatograms of the 70% ethanol-soluble (a) and insol-uble (b) fractions of the water-soluble fraction of a 9-month-old Cheddar cheese. The locations of the peptidesŒs1-CN fl-9 (X) and fl-13 (Y) are indicated (Tobin et al, unpubl results).Profil obtenu par chromatographie liquide haute performance en phase inverse (Cs) des fractions (a) solublesdans l'éthanol à 70% et (h) insolubles de lafraction soluble dans l'eau d'un/ramage de cheddar de 9 mois. Leslocalisations des peptides asrCN fl-9 (X) etfl-13 (Y) sont signalées (Tobin et al, non publié).


technique was used incIude those by Ardô andGripon (1995) and Zarmpoutis et al (1996).

Reverse-phase HPLC of pre-column-deriva-tized tluorescent-Iabelled amino acids is increas-ingly being used to measure free amino acids incheese, and has the advantage of using standardequipment and being amenable to automatedderivatization. Amino acids have been separatedas their dansyl (eg, Polo et al, 1985), OPA (eg,Ramos et al, 1987; Bütikofer et al, 1991) orFMOC (Bütikofer et al, 1991) derivatives.Bütikofer et al (1991) compared the results ofamino acid analysis of an acid hydrolysate ofcheese pro teins by pre-column derivatizationwith OPA or FMOC, followed by RP-HPLC,with the results obtained using an amino acidanalyzer fitted with an ion-exchange columnand using post-column derivatization with nin-hydrin. HPLC gave rapid, simple and sensitivedetermination of amino acids and yielded nar-rower, better-resolved peaks with shorter reten-tion times and a more stable baseline. Althoughthe HPLC method used had a slightly lowerrepeatability for sorne amino acids, it had goodreproducibility and correlated weil with ion-exchange chromatography.

Amino acids can be also quantified by gaschromatography (GC) after derivatization withheptatluorobutyric anhydride (HFBA) to yieldn-heptatluorobutyryl isobutyl derivatives. Capil-liary GC is most suitable (eg, Wood et al, 1985;Laleye et al, 1987; Bertacco et al, 1992). Thistechnique is reported to give good recovery ofadded amino acids; ail protein amino acids canbe resolved on a single chromatogram, and speedand accuracy are comparable to those of auto-mated amino acid analyzers and at far lowerequipment costs. The technique has good repro-ducibility and high accuracy, although the coef-ficients of variation for Met and Gly are some-what high (Bertacco et al, 1992). However, GChas not been used as widely as alternativetechniques.

Ammonia is produced by deamination ofamino acids, and techniques for its quantificationhave been discussed above. Amines in cheesemay be quantified by RP-HPLC using pre-col-

65

umn derivatization (eg, Bütikofer et al, 1990).Gas chromatography with mass spectrometriedetection (GC-MS) allows the quantificationand identification of volatiles from cheese (eg,Urbach, 1993), incIuding those resulting fromamino acid catabolism. A discussion of GC-MSis outside the scope of this review.

A FRACTIONATION SCHEMEFOR CHEESE PEPTIDES

The peptide system in most cheese varieties isextremely complicated. For example, > 200 pep-tides have been isolated and identified from theWSE of Cheddar (Singh et al, 1994, 1995, 1996).The detailed study of proteolysis in cheese oftenrequires that peptides in the initial extract befractionated by a combination of the techniquesdiscussed above. A number of fractionationschemes have been proposed (Kuchroo and Fox,1983b; Aston and Creamer, 1986; Fox, 1989;O'Sullivan and Fox, 1990; Cliffe et al, 1993;Singh et al, 1994, 1995, 1996; Breen et al, 1995).The scheme used at present in this laboratory isshown in figure 11. This fractionation schemewas developed for Cheddar cheese and may needto be modified for other varieties.

ISOLATION AND IDENTIFICATIONOF INDIVIDUAL PEPTIDES

Individual small peptides are usually isolatedby RP-HPLC. Fractions of eluate may be col-lected and freeze-dried, Il is desirable to checkthe homogeneity of fractions by re-chromatog-raphy under the same or preferably differentelution conditions. Peptides eluted in salt-containing buffers may be desalted by re-chromatography using an acetonitrile ormethanol gradient, the components of which areeasily volatilized.

In our experience, RP-HPLC does not ade-quately resolve larger casein-derived peptides.Such peptides may be resolved by PAGE andisolated by electroblotting onto a suitable mem-


brane (eg, polyvinylidine difluoride; McSweeneyet al, 1994a; Singh et al, 1995). Alternatively,unstained regions of the gel may be excised, andpeptides extracted from the macerated gel piecesby electrophoresis. However, the precise locationof peptides on unstained gels may be difficultsince the y cannot be compared directly with

stained gels, the dimensions of which changeon staining. The N-terminal sequence (seebelow) of electroblotted peptides can be deter-mined readily, but it is necessary to remove pep-tides from the blotting membrane prior to massspectrometrie analysis.

GRATED CHEESE

~---------t.~WSNRe-Exlracl

Fat

Watcr Extraction

Pcllel

(Discard)

WISN

Diafiltrution, 10 kOa membranesor

70'k Ethanol

Ion Exchangc ChromutographyUrea-PAOE

~RETENTATE PERMEATE

orINSOLUBL FRACTION

orSOLUBLE FRACTION

!Scphadcv 025

DEAE

+PAGE,

Electroblotting

~Plides ...............lHPLC

Il III IV V VI VII VIII IX X" JV

Amine Acids

HPLC

Fig 11. Fractionation scheme for chee se N (modified from Fox et al, 1995). WSN: water-soluble N; WISN:waler-insoluble N; DEAE: ion-exchange chromalography on DEAE--œllulose; PAGE: alkaline urea-polyacryl-amide gel electrophoresis; HPLC: reverse-phase (Cg) high performance liquid chromatography.Schéma de fractionnement de l'azote (N) du fromage (modifié à partir de Fox et al 1995). WSN: azote solubledans l'eau .. W1SN: azote non soluble dans l'eau .. DEAE : chromatographie échangeuse d'ions sur celluloseDEAE .. PAGE: électrophorèse sur gel alcalin urée-polyacrylamide .. HPLC: chromatographie liquide hauteperformance en phase inverse (C8).


The amino acid composition of a peptide,determined after hydrolysis, can pro vide usefulinformation as to its identity (eg, Kaminogawaet al, 1986). Computer software can be written(eg, Petrilli, 1982) or is available from com-mercial sources to aid in locating a peptide of agiven ami no acid composition in the proteinchain. However, more than one casein-derivedpeptide may have the same amino acid compo-sition and thus identification of peptides in acomplex mixture from their amino acid compo-sition, alone is not recommended. Identificationof the N-terminal residue or sequence (eg, Visseret al, 1977; Gonzâlez de Llano et al, 1991)improves the probability of correct identifica-tion of peptides of a gi ven composition.

Amino acid sequencing (either manual, eg,Mojarro-Guerra et al, 1991 or automated, eg,McSweeney et al, 1994a; Singh et al, 1994,1995, 1996) by Edman degradation can also beused to identify peptides. However, unless thepeptide is relatively short, it may not be possibleto determine its full sequence. The N-terminalsequence (5-7 residues) is normally adequateto locate the origin of a peptide in the caseins, butother techniques must be used to determine itslength. C-terminal sequencing (eg, Iiberation ofami no acids over time by carboxypeptidase Y,eg, McSweeney et al, 1993b) could be used forthis purpose but is not popular. Mass spectrom-etry (MS) is more convenient for identifyingpeptides of known N-terminal sequence (eg,Singh et al, 1994, 1995, 1996), although massanalysis alone is often insufficient to identifypeptides since different casein-derived peptidescan have the same ami no acid composition (andhence mass). An interesting application of MSinvolves determining the residual mass of a pep-tide after each cycle of a manual Edman degra-dation. The mass of the residue (Mf) re1easedcan be calculated from Mf = Mn-Mn_1, whereMn and Mn-1 are the masses of the peptide beforeand after removal of the N-terminal residue,which can thus be identified. Sequencing of thefirst few amino acids, coupled with its mass,permils precise identification of the peptide.Such an approach has been used to identify pep-

67

tides in Cheddar cheese (Gouldsworthy et al,1996). Most MS techniques used to identify pep-tides involve the production of ions of the intactpeptide. However, if high energies are used, thepeptide is fragmented and the masses obtainedare those of the resulting fragments. Cyclicdipeptides and y-glutamyl peptides have beenidentified using this technique (Roudot-Algaronet al, 1993, 1994), which yields more informa-tion on the structure of the compound understudy than just its mass; il is likely to be usefulfor studying very short peptides or those withan unusual structure.

STRATEGIES FORASSESSING PROTEOL YSIS

The choice of a technique for assessing prote-olysis in cheese depends on a number of fac-tors, including availability of equipment andresources, chee se variety and objectives of thestudy. Since proteolysis is one of the principalbiochemical events during cheese ripening, it isdesirable to include an assay for proteolysis inmost chee se ripening studies. The formation ofsoluble N or liberation of reactive groups may beadequate if a detailed investigation of proteoly-sis is not warranted. However, if a thoroughstudy of proteolysis is deemed worthwhile, it isdesirable to consider the entire process from theinitial hydrolysis of the caseins to the develop-ment of free amino acids and to use techniqueswhich resolve individual peptides.

Certain cheese varieties have characteristicswhich influence the choice of method. Theincrease in pH of mould-ripened varieties meansthat pH 4.6 buffers are more suitable as primaryextractants than water. Likewise, since ammoniais a major product of proteolysis in mould andbacterial surface-ripened varieties, it may bedesirable to quantify ammonia. The rate and/orpattern of proteolysis may be influenced by loca-tion within a cheese, eg, surface-ripened oryoung, brine-salted cheeses. A suitable sam-pling scheme should take account of suchdifferences.


If one or more of the proteolyic agents ismodified in a cheese, then the methods used tostudy proteolysis should be chosen so as toemphasize the level of proteolysis caused bythat agent. For example, if the influence of coag-ulant is studied, the formation of WSN and pep-tide profiles by urea-PAGE are suitable meth-cds. Since the coagulant does not produce freeamino acids, their determination does not directlymeasure the contribution of the coagulant. Incontrast, since the contribution of cheesemicroflora (natural or modified) to cheese ripen-ing is mainly in the production of short peptidesand free amino acids, RP-HPLC and amino acidanalysis are likely to be most effective. How-ever, it is also important to include sorne indexof primary proteolysis (eg, PAGE) to ensure thatit is typical.

ln this laboratory, routine assessment of pro-teolysis in Cheddar cheese during ripeninginvolves the determination of WSN (Kuchrooand Fox, 1982a) and peptide profiles of thecheese and WSEs therefrom by alkalineurea-PAGE (Blakesley and Boezi, 1977;Andrews, 1983). These assays pro vide impor-tant information on primary proteolysis. Water-soluble extracts are fractionated using 70%ethanol and the soluble and insoluble fractionsanalyzed by RP-HPLC (Cg; Singh et al, 1995).The 70% ethanol-insoluble fraction can also beanalyzed by urea-PAGE. Total free amino acidsare determined using the Cd-ninhydrin assay(Folkertsma and Fox, 1992) and individualamino acids by an automated amino acid ana-Iyzer with post-column derivatization using nin-hydrin. This strategy provides information onthe level and type of proteolysis from the initialhydrolysis of the caseins to the Iiberation of freeamino acids; and is suitable for most internai,bacterially-ripened cheese varieties.

ACKNOWLEDGMENT

The authors wish to express their thanks toA Cahalane for assistance in preparingthe manuscript.

REFERENCES

Abdel Baky AA, Abo Ella WM, Aly ME, Fox PF(1987) Improving the quality of Ras-type cheesemade from recombined milk containing high lev-els of total solids. Food Chem 26, 175-188

Addeo F, Chianese L, Di Luccia A, Petrilli P,Mauriello R, Anelli G (1983) Identification ofbovine casein variants by gel isoelectric focusing.Milchwissenschaft 38, 586-588

Addeo F, Moio L, Chianese L, Stingo C, Di Luccia A(1990a) Improved procedure for detecting bovineand ovine milk mixtures in cheese by isoelectricfocusing of para-K-casein. Milehwissensehaft 45,221-224

Addeo F, Moio L, Chianese L, Stingo C, Resmini E,Berner l,Krause D, Di Lucci A, Bocca A (1990b)Use of plasmin to increase the sensitivity ofthe detection of bovine milk in ovine cheeseby gel isoelectric focusing of y2-caseins.Milehwissensehaft 45,708-711

Addeo F, Chianese L, Salzano A, Sacchi R, Cappuc-cio U, Ferranti P, Malorni A (1992) Characteriza-tion of the 12% trichloroacetic acid-solubleoligopeptides of Parmigiano-Reggiano cheese.J Dairy Res 59, 401-411

Addeo F, Chianese L, Sacchi R, Spagna Musso S, Fer-ranti P, Malorni A (1994) Characterization of theoligopeptides of Parrnigiano-Reggiano cheese sol-uble in 120 g trichloroacetic acidIL. J Dairy Res 61,365-374

Addeo F, Garro F, Intorcia N, Pellegrino L, Resmini P,Chianese L (1995) Gel electrophoresis for thedetection of casein proteolysis in cheese. J DairyRes 62, 297-309

Adler-Nissen 1, (1979) Determination of the degreeof hydrolysis of food protein hydrolyzates by trini-trobenzene-sulphonic acid. J Agrie Food Chem27,1256-1262

Alonso L, Juârez M, Ramos M, Martin-Alvarez Pl(1987) Effects of changes during ripening andfrozen storage on the physicochemical and sen-sory characteristics of Cabral es cheeses. lnt J FoodSei Teehnol22, 525-534

Amantea GF, Skura Bl, Nakai S (1986) Culture effecton ripening characteristics and rheological behaviourof Cheddar cheese. J Food Sei 51, 912-918

Amigo L, Ramos M, Calhau L, Barbosa M (1992)Comparison of electrophoresis, isoelectric focus-ing, and immunodiffusion in determination ofcow' s and goal' s milk in Serra da Estrela cheeses.Lait 72, 95-10 1


Andrews AT (1983) Proteinases in normal bovinemilk and their action on caseins. J Dairy Res 50,45-55

Anonymous (1989) Ammonia, UV Methodfor theDetermination of Ammonia in Foodstuffs andOther Materials and for the Determination ofNitrogen After Kjeldahl Digestion. Boehringer-Mannheim, Mannheim, Germany

Ardô Y, Gripon JC (1995) Comparative study of pep-tidolysis in sorne semi-hard round-eyed cheesevarieties with different fat contents. J Dairy Res62,543-547

Ardô Y, Gripon JC (1991) Chromatographie methodsused to measure proteolysis in cheese. Bull lntDairy Fed 261,29-34

Ardô Y, Meisel H (1991) Methods for direct mea-surement of peptide bond cleavage in cheese. BullIntDairyFed261,1O-13

Ashworth VS (1966) Determination of protein in dairyproducts by dye-binding. J Dairy Sei 49,133-137

Aston JW, Crea mer LK (1986) Contribution of thecomponents of the water-soluble fraction to theflavour of Cheddar cheese. NZ J Dairy Sei Technol21,229-248

Barlow lE, Lloyd GT, Ramshaw EH (1986) The rnea-surement of proteolysis in Cheddar cheese: a com-parison of trinitrobenzene sulphonic acid proce-dures. Aust J Dairy Technol41, 79-81

Barrefors P, Ekstrand B, Fagerstam L, Larson-Raznikiewicz M, Scharr J, Steffner P (1985) Fastprotein liquid chromatography (FPLC) of bovinecaseins. Milchwissenschaft 40,257-260

Basch 11, Farrell HM Jr, Walsh RA, Konstance RP,Kumosinski TF (1989) Development of a quanti-tative model for enzyme-catalyzed, time-depen-dent changes in protein composition of Cheddarcheese during storage. J Dairy Sei 72, 591-603

Beeby R (1980) Use of fluorescamine at pH 6.0 tofollow the action of chymosin on x-casein and toestimate this prote in in milk. NZ J Dairy SeiTechnoI15,99-108

Bertacco G, Boschella 0, Lercker G (1992) Gas chro-matographic determination offree amino acids incheese. M ilchwissenschaft 47, 348-350

Bhowmik T, Riesterer R, van Boekel MAlS, MarthEH (1990) Characteristics of low-fat Cheddarcheese made with added Micrococcus or Pedio-coccus species. Milchwissenschaft 45,230-235

Bican P, Spahni A (1993) Proteolysis in Swiss-typechee ses: a high performance liquid chrornato-graphie study. lnt Dairy J 3,73-84

69

Bican P, Blanc B (1982) Milk protein analysis,a high performance chromatography study.Milchwissenschaft 37, 592-593

Bican P, Spahni A (1991) Low molecular mass nitro-gen components in ripening cheese. Lebensm WissTechnoI24,315-322

Biede SL, Hammond EG (1979) Swiss cheese. 1.Chemical analysis. J Dairy Sei 62, 227-237

Blais BW, Cunningham A, Yamazaki H (1994) Anovel immunofiuorescence capillary elec-trophoresis assay for the determination of chi or-amphenicol in milk. Food Agric lmmunol 6,409-417

Blakesley RW, Boezi JA (1977) A new staining tech-nique for proteins in polyacrylamide gels usingCoomassie brilliant blue G250. Anal Bioehem 82,580-582

Bockelmann W (1995) The proteolytic system ofstarter and non-starter bacteria: components andtheir importance for cheese ripening. lnt Dairy J 5,977-994

Bouton Y, Grappin R (1994) Measurement of proteo-Iysis in cheese: relationship between phospho-tungstic acid-soluble N fraction by Kjeldahl and2,4, 6-trinitrobenzenesulphonic acid-reactivegroups in water-soluble N. J Dairy Res 61, 437-440

Bradford M (1976) A rapid and sensitive method forthe quantitation of microgram quantities of pro-tein utilizing the principle of protein-dye binding.Anal Bioehem 72, 248-254

Breen ED (1992) Fractionation and identification ofwater-soluble and water-insoluble cheese nitro-gen. MSc thesis, Nat Univ Ireland, Cork

Breen ED, Fox PF, McSweeney PLH (1995) Frac-tionation of water-soluble peptides from Cheddarcheese. Ital J Food Sei 7, 211-220

Bütikofer V, Fuchs D, Murni D, Bosset JO (1990)Beitrag zur Bestimmung biogener Amine in Kâse.Vergleich einer verbesserten HPLC-mit einer IC-Methode und Anwendung bei verschiedenen Kas-esorten. Mitt Geb Lebensmittelunters Hyg 81,120-133

Bütikofer V, Fuchs D, Bosset JO, Gmur W (1991)Automated HPLC-amino acid determination byprecolumn derivatization with OPA and FMOCand comparison with classical ion-exchange chro-matography. Chromatographia 31, 441-447

Bütikofer V, Ruegg M, Ardo Y (1993) Determinationof nitrogen fractions in cheese, evaluation of a col-laborative study. Lebensm Wiss Teehnol 26,271-275

70 PLH McSweeney. PF Fox

Chakravorty SC. Srinivasan RA. Babbar Il. DudaniAT. Burde SD. Iya KK (1966) Comparative studyof proteolysis during ripening of Cheddar cheese bybacterial milk clotting enzymes and rennet extract.Proc l7th lnt Dairy Congr, Munich.187-198

Christensen TMIE. Kristiansen KR. Madsen JS (1989)Proteolysis in cheese investigated by high perfor-mance liquid chromatography. J Dairy Res 56.823-828

Christensen TMIE. Bech A-M. Werner H (1991)Methods for cru de fractionation (extraction andprecipitation) of nitrogen components in cheese.Bulllnt Dairy Fed 261. 4-9

Church FC. Swaisgood HE. Porter DH. Catignani GL(1983) Spectrophotometric assay using o-phthal-dialdehyde for determination of proteolysis in milkand isolated milk proteins. J Dairy Sei 66.1219-1227

Cifuentes A. de Frutos M. Diez-Masa JC (1993) Anal-ysis of whey proteins by capillary electrophoresisusing buffer-containing polymerie additives.J Dairy Sci 76. 1870-1875

Citti JE. Sandine WE. Elliker PR (1963) Sorne obser-vations on the Hull method for measurement ofproteolysis in milk. J Dairy Sei 46. 337

Clegg KM. Lee YK. McGilligan JF (1982) Trini-trobenzene sulphonic acid and ninhydrin reagentsfor the assessment of protein degradation in cheesesampI es. J Food Technol 17.517-520

Clemens W (1954) Papierchromatographische Unter-suchungen an Kâse, Milchwissenschaft 9. 195-201

Cliffe Al. Law BA (1991) A time course study of pep-tide production in accelerated-ripened Cheddarchee se using reverse phase high performanceliquid chromatography. Food Biotechnol S, 1-17

Cliffe AJ. Reveil D. Law BA (1989) A method forthe reverse phase-high performance liquid chro-matography of peptides from Cheddar cheese.Food Chem 34.147-160

Cliffe AJ. Marks JD. Mulholland F (1993) Isolationand characterization of non-volatile flavours fromcheese, peptide profile of flavour fractions fromCheddar checsc, determined by reverse phase high-performance liquid chromatography. Int Dairy J3.379-387

Conti A. Napolitano L. Cantisani SM. Davoli R. Dall-Olio S (1988) Bovine beta-IactoglobulinH- isolation by preparative isoelectric-focusing inimmobilized pH gradients and preliminary charac-terization. J Biochem Biophys Methods 16.205-214

Creamer LK (1991) Electrophoresis of cheese. BullInt Dairy Fed 261. 14-28

Creamer LK. Richardson BC (1974) Identification ofthe primary degradation product of <Xsl-casein inCheddar cheese. NZ J Dairy Sei Technol S, 9-13

Creamer LK. Oison NF (1982) Rheological evalua-tion of maturing Cheddar cheese. J Food Sei 41.631-636. 646

Creamer LK. Lawrence RC. Gilles J (1985) Effect ofacidification of cheese milk on the resultant Ched-dar cheese. NZ J Dairy Sei TechnoI20.185-203

Davis JG (1965) Cheese. J and A Churchill Ltd,London

de Jong L (1975) A quantitative electrophoretic methodof studying cheese ripening. Neth Milk Dairy J 29.162-168

de Koning PJ. de Wit JN. Driessen FM (1992)Process conditions affecting age-thickening andgelation of sterilized canned evaporated milk.Neth Milk Dairy J 46.3-18

Desmazeaud MJ. Gripon JC. Le Bars D. Bergère JL(1976) Etude du role des micro-organismes et desenzymes au cours de la maturation des fromages.III. Influence des micro-organismes. Lait 56.379-396

Doi E. Shibata D. Matoba T (1981) Modified colori-metric ninhydrin methods for peptidase assay. AnalBiochem 118. 173-184

Edwards J. Kosikowski FV (1983) Bitter compoundsfrom Cheddar cheese. J Dairy Sei 66.727-734

El Soda M. Desmazeaud MJ. Aboudonia S. Kamal N(1981) Acceleration of cheese ripening by theaddition of whole cells or cell free extracts fromLactobacillus casei to the cheese curd.Milchwissenschaft 36.140-142

Exterkate FA (1995) Lactococcal cell envelope pro-teinases: differences. calcium-binding effects androle in cheese ripening.lnt Dairy J 5.995-1018

Fernandez M. Fox PF (1997) Fractionation of cheesenitrogen using chitosan. Food Chem 58. 319-322

Femandez-Salguero J. Andres Marcos Alcala M. Este-ban MA (1989) Proteolysis of Cabral es cheese andother European blue vein cheesc varieties. J DairyRes 56.141-145

Fields R (1971) The measurement of amino groupsin proteins and peptides. Biochem J 124. 581-590

Folkertsma B. Fox PF (1992) Use of Cd-ninhydrinreagent to assess proteolysis in cheese during ripen-ing. J Dairy Res 59. 217-224

Folkertsma B. Fox PF. McSweeney PLH (1996)Accelerated ripening of Cheddar cheese at ele-vated temperatures. Int Dairy J 6. 1117-1134


Foster PMD, Green ML (1974) A quantitative gel fil-tration method for analysis of the proteinaceousfraction of Cheddar cheese. J Dairy Res 41,259-268

Fox PF (1989) Proteolysis in cheese during manufac-turing and ripening. J Dairy Sei 72, 1379-1400

Fox PF, Singh TK, McSweeney PLH (l995a) Bio-genesis of flavour compounds in cheese.ln: Chem-istry of Structure/Function Relationships in Cheese .(Malin EL, Tunick MH, eds) Plenum Publ Corp,New York, 59-98

Fox PF, McSweeney PLH, Singh TK (1995b) Meth-ods for assessing proteolysis in cheese du ring ripen-ing. In: Chemistry of Structure/Function Rela-tionships in Cheese (Malin EL, Tunick MH, eds)Plenum Publ Corp, New York, 161-194

Friedman M, Pang J, Smith GA (1984) Ninhydrin-reactive lysine in food proteins. J Food Sei 49,10-13,20

Furtado MM, Chandan RC (1985) Ripening changesin blue-mold surface ripened chee se from goal' smilk. J Food Sei 50, 545-546

Girardet J-M, Saulnier F, Driou A, Linden G, Cod-deville G, Spik G (1994) High performance elec-trophoresis chromatography of bovine milk com-ponent 3 glycoproteins. J Dairy Sei 77, 1205-1215

Gonzâlez de Llano D, Ramos M, Polo C (1987) Gel fil-tration and high performance liquid chromato-graphie analysis of phosphotungstic acid solublepeptides from blue cheeses. Chramatographia 23,764-766

Gonzâlez de Llano D, Polo C, Ramos M (1991) Pro-duction, isolation and identification of low molec-ular mass peptides from blue cheese by high per-formance liquid chromatography. J Dairy Res 58,363-372

Gonzâlez de Llano D, Ramos M, Martin-Alvarez PJ(1993) Comparative study of different methodsfor evaluating proteolysis in blue cheese.Z Lebensm Unters Forsch 196,29-31

Gordon MJ, Huang X, Petoney SL Jr, Zare RN (1988)Capillary electrophoresis. Science 242, 224-228

Gouldsworthy AM, Leaver J, Banks JM (1996) Appli-cation of mass spectrometry sequencing techniquefor identifying peptides present in Cheddar cheese.Int Dairy J 6, 781-790

Green ML, Foster PDM (1974) Comparison of therates of proteolysis during ripening of Cheddarcheeses made with calf rennet and swine pepsinas coagulants. J Dairy Res 41, 269-282

Gripon JC (1993) Mould-ripened cheeses. In: Cheese:Chemistry, Physics and Micrabiology, vol 2 (ed, PF

71

Fox) Chapman and Hall, London, 111-136,2nd edn

Gripon JC, Desmazeaud MJ, Le Bars D, Bergère JL(1975) Etude du role des micro-organismes et desenzymes au cours de la maturation des fromages.II. Influence de la presure commerciale. Lait 55,502-516

Gripon JC, Desmazeaud MJ, Le Bars D, Bergère JL(1977) Role of proteolytic enzymes of Strepto-coccus lactis, Penicillium roqueforti and Penicil-lium caseicolum during cheese ripening. J DairySei 60, 1532-1538

Guinee TP, Pudja PD, Reville WJ, Harrington D, Mul-holland Eü, Cotter M, Cogan TM (1995) Com-position, microstructure and maturation of semi-hard cheeses from high protein ultrafiltered milkretentates with different levels of denatured wheyproteins. Int Dairy J 5, 543-568

Guo MR, Kindstedt PS (1995) Age-related changesin the water-phase of Mozzarella cheese. J DairySci78,2099-2107

Gupta SK, Whitney RM, Tuckey SL (1974) Brine sol-uble protein of Cheddar and Gouda cheese. J DairySci 57, 540-544

Haasnoot W, Stouter P, Venema DP (1989) High per-formance liquid chromatography determination ofthe extent of proteolysis in Gouda cheese.J Chramatogr483, 319-329

Habeeb AFS (1966) Determination of free aminogroups in proteins by trinitrobenzenesulfonic acid.Anal Biochem 14,328-336

Harwalkar VR, Elliott JA (1971) Isolation of bitterand astringent fractions from Cheddar chee se.J Dairy Sei 54,8-11

Hickey MW, van Leeuwen H, Hillier AJ, Jago GR(1983) Amino acid accumulation in Cheddarcheese manufactured from normal and ultrafilteredmilk. Aust J Dairy Technol 38, 110-113

Huber L, Klostermeyer H (1974) Isolierung und Iden-tifizierung eines Bitterstoffes aus Buuerkâse.Milchwissenschaft 29,449-455

Hull ME (1947) Studies on milk proteins. II. Colori-metrie determination of the partial hydrolysis ofthe proteins in milk. J Dairy Sei 30,881-884

Inikhov GS (1951) Chemical Analysis of Milk Prod-ucts. Part Il. Food Industry Publ, Moscow

IDF (1991) Chemical methods for evaluating proteo-Iysis in cheese maturation. Bull lnt Dairy Fed261

Jarrett WD, Aston JW, Dulley JR (1982) A simplemethod for estimating free amino acids in Cheddar


cheese. Aust J Dairy Technol 37,55-58 (corr1983,38)

Kaminogawa S, Yan TR, Azuma N, Yamauchi K(1986) Identification of 10w molecular weight pep-tides in Gouda-type cheese and evidence for theformation of these peptides from 23-N-terminalresidues of o.sl-casein by proteinases of Strepto-coccus cremoris H61. J Food Sei 51,1253-1256,1264

Katsiari MC, Voutsinas LP (1994) Manufacture oflow-fat Kefalograviera cheese. Int Dairy J 4,533-553

Kim HH, Jimenez-Flores R (1994) Comparison ofmilk proteins using preparative isoelectric focusingfollowed by polyacrylamide gel electrophoresis.J Dairy Sei 77,2177-2190

Kosikowsky FV (1951 a) Paper chromatography ofthe free amino acids in American Cheddar cheese.J Dairy Sei 34, 228-234

Kosikowsky FV (1951 b) The liberation of the freeamino acids in raw and pasteurized milk Cheddarcheese during ripening. J Dairy Sei 34, 235-241

Kroger M, Weaver JC (1979) Use ofprotein dye-bind-ing values as indicators of the 'chemical age' ofconventionally made Cheddar cheese andhydrolyzed-lactose Cheddar cheese. J Food Sei44,304-305

Kuchroo CN, Fox PF (l982a) Soluble nitrogen inCheddar cheese: comparison of extraction proce-dures. Milchwissenschaft 37, 331-335

Kuchroo CN, Fox PF (1982b) Fractionation ofwater-soluble nitrogen from Cheddar cheese: chemicalmethods. Milchwissenschaft 37, 651-653

Kuchroo CN, Fox PF (l983a) Fractionation of thewater-soluble nitrogen from Cheddar cheese: chro-matographic methods. Milchwissenschaft 38,76-79

Kuchroo CN, Fox PF (1983b) A fractionation schemefor the water-soluble nitrogen in Cheddar chee se.Milchwissenschaft 38, 389-391

Kuchroo CN, Rahilly J, Fox PF (1983) Assessmentof proteolysis in cheese by reaction with trini-trobenzene sulphonic acid. Irish J Food Sei Tech-nol7,129-133

Laemmli UK (1970) Cleavage of structural proteinsduring the assembly of the head of bacteriophageT4. Nature 227, 680-685

Lagerwerf FM, van Wassenaar PD, Haverkamp J(1995) Comparison of Cheddar cheeses producedwith recombinant chymosin with calf rennet usingtwo dimensional high-performance liquid chro-matography. J Dairy Res 62, 673-679

Laleye LC, Simard RE, Grosselin C, Lee BH, GirouxRN (1987) Assessment of Cheddar chee se qual-ity by chromatographie analysis of free aminoacids and biogenic amines. J Food Sei 52,303-307,311

Lane CN, Fox PF (1996) Contribution of starter andadjunct lactobacilli to proteolysis in Cheddar cheeseduring ripening. Int Dairy J 6, 715-728

Lau KY, Barbano DM, Rasmussen RR (1991) Influ-ence of pasteurization on protein breakdown inCheddar cheese during aging. J Dairy Sei 74,727-740

Lawrence RC, Gilles J, Creamer LK (1983) The rela-tionship between chee se texture and flavour.NZ J Dairy Sei Technol 18, 175-190

Le Bars D, Gripon JC (1989) Specificity of plasmintowards bovine o.,z-casein. J Dairy Res 56,817-821

Ledford RA, O'Sullivan AC, Nath KR (1966) Resid-ual casein fractions in ripened cheese determinedby polyacrylamide-gel electrophoresis. J DairySei 49,1098-1101

Lemieux L, Puchades R, Simard RE (1989) Size-exclu-sion HPLC separation of bitter and astringent frac-tions from Cheddar cheese made with added Lac-tobacillus strains to accelerate ripening. J FoodSci54,1234-1237

Lemieux L, Puchades R, Simard RE (1990) Free aminoacids in Cheddar cheese: comparison of quanti ta-tion methods. J Food Sei 55,1552-1554

Lindeberg J (1996a) Capillary electrophoresis in foodanalysis. Food Chem 55, 73-94

Lindeberg J (l996b) Addendum to capillary elec-trophoresis in food analysis. Food Chem 55, 95-101

Lindqvist B (1962) Preparative separation of highmolecular weight peptides of cheese. In: Proc 16thInt Dairy Congr, Copenhagen B, 673-678

Lindqvist B, Storgards T (1959) Changes in caseinduring cheese ripening. In: Proc 15th lnt DairyCongr, London 2, 679-684

Lindqvist B, Storgards T, Goransson MB (1953) Elec-trophoresis in paper as a means of studying theripening process in chee se. Proc 13th lnt DairyCongr, The Hague 3,1261-1268

Lucey JA, Gorry C, Fox PF (1993) Changes in theacid-base buffering curves during the ripening ofEmmental cheese. Milchwissenschaft 48, 183-186

Mabbitt LA (1955) Quantitative estimation of theami no acids in Cheddar cheese and their impor-tance in flavour. J Dairy Res 22, 224-231

Methods used to assess proteolysis in cheese 73

Madkor SA, Fox PF, Shalabi SI, Metwally NH (1987)Studies on the ripening of Stilton cheese: proteo-Iysis. Food Chem 25,13-19

Marshall T, Gilliland D, Williams KM (1988) Sodiumdodecylsulphatel polyacrylamide gel elec-trophoresis of milk proteins: application to cheese.Biochem Soc Trans 16, 165-166

Martley FG, Crow VL (1993) Interactions betweennon-starter microorganisms during cheese manu-facture and ripening. lnt Dairy J 3, 461-483

McGann TCA, Mathiassen A, O'Connell lA (1972)Applications of the pro-milk Mk II. III. Rapid esti-mation of casein in milk and protein in whey.Lab Pract 2 l , 628-631, 650

McGugan WA, Emmons DB, Larmond E (1979) Influ-ence of volatile and non volatile fractions on inten-sity of Cheddar cheese flavor, J Dairy Sei 62,398-403

McSweeney PLH, Fox PF (1993) Cheese, methodsof chemical analysis, ln: Cheese, Chemistry,Physics and Microbiology. Vol 1. General Aspects(Fox PF, ed) Chapman and Hall, London,341-388, 2nd edn

McSweeney PLH, Fox PF, Lucey lA, lordan KN,Cogan TM (1993a) Contribution of the indigenousmicrollora to the maturation of Cheddar cheese.Int Dairy J 3, 613-634

McSweeney PLH, OIson NF, Fox PF, Healy A, HojrupP (1993b) Proteolytic specificity of chymosin onbovine as I-casein. J Dairy Res 60, 401-412

McSweeney PLH, Pochet S, Fox PF, Healy A (1994a)Partial identification of peptides from the water-insoluble fraction of Cheddar cheese. J Dairy Res61,587-590

McSweeney PLH, Walsh E, Fox PF, Cogan TM, Dri-nan FD, Castelo-Gonzalez M (1994b) A proce-dure for the manufacture of Cheddar cheese undercontrolled bacteriological conditions and the effectof adjunct lactobacilli on chee se quality. Irish JAgrie Food Res 33,183-192

McSweeney PLH, Lynch CM, Walsh EM, Fox PF,lordan, KN, Cogan TM, Drinan PD (1995) Role ofnon-starter lactic acid bacteria in Cheddar cheeseripening. ln: Proe 4th Cheese Symp, Moorepark,Fermoy, Co Cork, 32-45

Melachouris NP, Tuckey SL (1966) Changes in theproteins in Cheddar cheese made from milk heatedat different temperatures. J Dairy Sei 49, 800-805

Michalke B (1995) Capillary electrophoresis methodfor a c1ear identification of Seleno ami no acids incomplex matrices like human milk. Fresenius JAnal Chem 351, 670-676

Moio L, Di Luccia A, Addeo F (1989) Fast isoelectric-focusing of milk-proteins on small ultrathin poly-acrylamide gels containing urea. Electrophoresis10,533-535

Moio L, Chianese L, Rivemale M, Addeo F (1992)Fast detection of bovine milk in Roquefort cheesewith PhastSystem® by gel isoelectric focusing andimmunoblotting. Lait 72,87-93

Mojarro-Guerra SH, Amado R, Arrigoni E, Solms 1(1991) Isolation oflow-molecular weight tastepeptides from Vacherin Mont d'Or cheese. J FoodSei 56, 943-947

Moore S, Stein WH (1948) Photometrie ninhydrinmethod for use in the chromatography of aminoacids. J Biol Chem 176,367-388

Moore S, Stein WH (1954) A modified ninhydrinreagent for the photometrie determination of aminoacids and related compounds. J Biol Chem 211,907-913

Morawski J, Alden P, Sims A (1993) Analysis ofcationic nutrients from foods by ion chromato-graphy. J Chromatogr 640,359-364

Morr CV (1971) Comparison of protein preparationprocedures and starch versus polyacrylamide gelelectrophoresis for examining casein degradationproducts in cheese. J Dairy Sei 54, 339-343

Morris HA, Holt C, Brooker BE, Banks lM, MansonW (1988) Inorganic constituents of cheese, analysisof juice from a one-month old Cheddar cheese andthe use of light and electron microscopy to char-acterize the crystalline phases. J Dairy Res 55,255-268

Mulvihill DM, Fox PF (1979) Proteolytic specificityof chymosin on bovine asl-casein. J Dairy Res46,641-651

Nath KR, Ledford RA (1973) Growth response ofLactobaeillus easei variety casei to the proteolysisproducts in cheese during ripening. J Dairy Sei56,710-715

Ney KH (1985) N-Endstandige Aminosâuren der Pep-tide aus reifem Cheddarkâse. Milchwissenschaft40,207-208

Noomen A (1 977a) Noordhollandse Meshangercheese, a novel method for research on cheeseripening. 2. The ripening of the cheese. Neth MilkDairyJ3l, 75-102

Noomen A (l977b) A rapid method for the estima-tion of the dissolved and the undissolved nitrogencompounds in cheese. Neth Milk Dairy J 31,163-176

derived by fermentation on Manchego chee seripening. J Dairy Sei 77,16-23

Polo C, Ramos M, Sanches R (1985) Free amino acidsby high performance liquid chromatography andpeptides by gel electrophoresis in Mahon cheeseduring ripening. Food Chem 16,85-96

Polychroniadou A (1988) A simple procedure usingtrinitrobenzene sulphcinic acid for monitoring pro-teolysis in cheese. J Dairy Res 55, 585-596

Polychroniadou A, Vlachos J (1979) Les acides aminésdu fromage Télémé. Lait 59, 234-243

Poznanski S, Habaj B, Rymaszewski J, RapazynskiT (1966) Influence of different starter cultures onthe protein breakdown in Edam cheese. In: Pme17th lm Dairy Congr, Munich, 555-564

Pretswell EL, McGaw BA, Morrison AR (1995) Thecomparison of capillary zone electrophoresis andatomic spectroscopy for the determination of thecation content of a standard reference materialIAEA-A-II milk powder. Talanta 42, 283-289

Puchades R, Lemieux L, Simard RE (1989a) Evolutionof free amino acids during the ripening of Cheddarcheese containing added lactobacilli strains. J FoodSei 54, 885-888,946

Puchades R, Lemieux L, Simard RE (1989b) Sensitive,rapid and precise determination of L-glutamic acidin cheese using a flow injection system with immo-bilized enzyme column. J Food Sei 54, 423-426

Puchades R, Lemieux L, Simard RE (1990) Determi-nation of free amino acids in cheese by flow injec-tion analysis with an enzyme reactor and chemi-luminescence detector. J Food Sei 55,1555-1558,1578


0'Keeffe AM, Fox PF, Daly C (1978) Proteolysis inCheddar cheese: role of coagulant and starter bac-teria. J Dairy Res 45, 465-477

0' Keeffe RB, Fox PF, Daly C (1976) Contributionof rennet and starter proteases to proteolysis inCheddar cheese. J Dairy Res 43,97-107

O'Sullivan M, Fox PF (1990) A scheme for partialfractionation of cheese peptides. J Dairy Res 57,135-139

Ollikainen P (1990) Titration - a rapid method for thedetermination of proteolysis in cheese. J DairyRes 57,149-150

Omar MM (1984) Microstructure, free ami no acidsand free fatty acids in Ras cheese. Food Chem 15,19-29

Ordonez JA, Burgos J (1977) Etude de la variété defromage «Ulloa», 1. Evolution de la flore micro-bienne et des composants azotés au cours de lamaturation. Lait 57,150-163

Otte J, Kristiansen KR, Zakora M, Qvist KB (1994)Separation of individual whey proteins and mea-sure ment of œ-lactalbumin and ~-Iactoglobulin bycapillary zone electrophoresis. Neth Milk Dairy J48,81-97

Pant J, Trenerry VC (1995) The determination of sor-bic acid and benzoic acid in a variety of beveragesand foods by micellar electrokinetic chromato-graphy. Food Chem 53, 219-226

Pearce KN (1979) Use of fluorescamine to determinethe rate of release of the caseino-macropeptide inrennet treated milk. NZ J Dairy Sei Technol 14,233-239

Pearce KN, Karahalios D, Friedman M (1988) Nin-hydrin assay for proteolysis in ripening cheese.J Food Sei 53, 432-435, 438

Peterson RF, Nauman LW, McMeekin TL (1957) Theseparation of amino acid composition of a purephosphopeptone prepared from ~-casein by theaction of trypsin. J Am Chem Soe 80, 95-99

Peterson SD, Marshall RT (1990) Nonstarter lacto-bacilli in Cheddar cheese: a review. J Dairy Sei73,1395-1410

Petri IIi P (1982) A simple BASIC program for pep-tide localization by its amino acid composition.Ital J Bioehem 31, 391-395

Pham AM, Nakai S (1984) Application of stepwisediscriminant analysis to high-pressure liquid chro-matography profiles of water extract for judgingripening of Cheddar cheese. J Duiry Sei 67,1390-1396

Picon A, Gaya P, Medina M, Nuûez M (1994) Theeffect of liposome eneapsulation of chymosin

Ramos M, Caeeres J, Polo C, Alonso L, Juarez M(1987) Effeet of freezing on soluble nitrogen frac-tion of Cabral es eheese. Food Chem 24, 271-278

Rank TC, Grappin R, Oison NF (1985) Seeondaryproteolysis of eheese during ripening. A review.J Dairy Sei 68, 801-805

Reiter B, Sorokin Y, Pickering A, Hall AJ (1969)Hydrolysis of fat and protein in small cheeses madeunder aseptie conditions. J Dairy Res 36, 65-76

Reps A (1993) Bacterial surface-ripened cheeses. In:Cheese, Chemistry, Physics and Microbiology,vol 2 (Fox PF, ed) Chapman and Hall, London,137-172, 2nd edn

Reville WJ, Fox PF (1978) Soluble protein in Cheddarcheese. A comparison of analytical methods. IrishJ Food Sei Teehnol2, 67-76

Roudot-Algaron F, Le Bars D, Einhorn J, Adda J,Gripon JC (1993) Flavor constituents of aqueous


fraction extracted from Comté cheese by liquidcarbon dioxide. J Food Sei 58,1005-1009

Roudot-Algaron F, Kerhoas L, Le Bars D, Einhom J,Gripon JC (1994) Isolation ofy-glutamyl peptidesfrom Comté cheese. J Dairy Sei 77,1161-1166

Salji JP, Kroger M (1981) Proteolysis and lipolysisin ripening Cheddar chee se made with con ven-tional bulk starter and with frozen concentrateddirect-to-the-vat culture. J Food Sei 46, 1345-1348

Salvat-Brunaud D, Maubois JL, Le Graet Y, Piot M,Maillard ME, Corre C, Thierry A (1995) Extractionet analyse de la phase aqueuse de l'Emmental àquatre stades d'affinage. Lait 75,239-249.

Samples DR, Richter RL, Dili CW (1984) Measuringproteolysis in Cheddar cheese slurries, compari-son of Hull and trinitrobenzene sulfonic acid pro-cedures. J Dairy Sei 67, 60-63

Santoro M, Leone AM, La Notte E, Liuzzi VA (1987)Fractionation of soluble nitrogen in Provolonecheese. I.Milehwissensehaft 42, 709-712

Schmitt M, Saulnier F, Malhautier L, Linden G (1993)Effect of temperature on the salt balance of milkstudied by capillary electrophoresis. J Chramatogr640,419-424

Sciancalepore V, Alviti FS (1987) A dialysis methodfor measuring cheese ripening. Milehwissensehaft42,220-221

Sciancalepore V, Longone V (1988) Rapid methods formeasuring the degree of proteolysis as cheeseripening index. Milehwissensehaft 43,357-359

Shalabi SI, Fox PF (1987) Electrophoretic analysis ofcheese: comparison of methods. Irish J Food SeiTeehnolll,135-151

Shindo K, Sakurada K, Niki R, Arima S (1980) Stud-ies on immobilized chymosin. 5. Experiments incheese-making with immobilized chymosin.Milehwissensehaft 35,527-530

Singh A, Ganguli NC (1972) Changes in peptidesand ami no acids in chee se during ripening.Milehwissenschaft 27, 412-416

Singh TK, Fox PF, Hejrup P, Healy A (1994)A scheme for the fractionation of cheese nitrogenand identification of principal peptides. IntDairyJ4,111-122

Singh TK, Fox PF, Healy A (1995) Water soluble pep-tides in Cheddar cheese: isolation and identificationof peptides in the UF retentate of water-solublefractions. J Dairy Res 62, 629-640

Singh TK, Fox PF, Healy A (1997) Isolation and iden-tification of further peptides in the diafiltrationretentate of the water-soluble fraction of Cheddarcheese. J Dairy Res 64 (in press)

75

Smith AM, Nakai S (1990) Classification of chee sevarieties by multivariate analysis of HPLC pro-files. Can Inst Food Sei Teehnol J 23,53-58

St. Martin M, Paquin P (1990) Ion exchange fast pro-tein liquid chromatography, optimization of thepurification of caseins using a non-denaturingdetergent. J Dairy Res 57, 63-68

Storgards T, Lindqvist B (1953) Untersuchungen überdie Kâsereifung. I. Chromatographische Un ter-suchungen über das Vorkommen von Aminosâurenin verschiedenen Kâsesorten. Milehwissensehaft8,5-10

Strange ED, Malin EL, Van Hekken DL, Basch JJ(1992) Chromatographie and electrophoretic meth-ods used for analysis of milk proteins. J Chro-matogr 624, 81-102

Tegtmeyer P, Schwartz M, Collins JK, Rundell K(1975) Regulation of tumor antigen synthesis bysimian virus 40 gene A. J Viral 16, 168-178

Tieleman AE, Warthesen JJ (1991) Comparison ofthree extraction procedures to characterize Ched-dar chee se proteolysis. J Dairy Sei 74, 3686-3694

Tienstra PA, van Reil JAM, Mingorance MD, OliemanC (1992) Assessment of the capabilities of capillaryzone electrophoresis for the determination of hip-puric and orotic acid in whey. J Chromatogr 608,357-361

Tokita F, Nakanishi T (1962) On the distributive pat-terns of lower molecular weight peptides in cheese.Milehwissensehaft 17,198-201

Tokita F, Hosono A (1968) An investigation aboutLimburger cheese ripening by gel filtration onSephadex. Milehwissensehaft 23,758-761

Trieu-Cuot P, Gripon JC (1982) A study of proteoly-sis during Camembert cheese ripening using iso-electric focusing and two-dimensional elec-trophoresis. J Dairy Res 49,501-510

Tsarichenko AP (1966) New ninhydrin reagent forquantitative deterrnination of amino acids in eluatesin column chromatography. Nauchn TI' KrasnodarGos Pedogog Inst 70, 86-88; cited from ChemAbstr (1967) 67,79479c

Tunick MH, Mackey KL, Shieh JJ, Smith PW, CookeP, Malin EL (1993) Rheology and microstructureof low-fat Mozzarella cheese. Int Dairy J 3,649-662

Udenfried S, Stein S, Bohlen P, Darman W Leimgru-ber W, Weigele M (1972) Fluorescamine, a reagentfor as say of amino acids, peptides, proteins andprimary amines in the picomole range. Science178, 871-872


Urbach G (1993) Relations between chee se flavourand chemical composition. Int Dairy J 3, 389-422

Vakaleris DG, Priee WV (1959) A rapid spectropho-tometric method for measuring chee se ripening.J Dairy Sei 42,264-276

Vakaleris DG, Oison NF, Priee WV, Knight SG (1960)A study of the ripening of Dariworld and Ched-dar cheese with special emphasis on proteolysis.J Dairy Sei 43,1058-1067

van den Berg G, de Koning PJ (1990) Gouda cheese-making with purified calf chymosin and microbio-logically produced chymosin. Neth Milk Dairy J44,189-205

van Hekken DL, Thompson MP (1992) Application ofPhast" system to the resolution of bovine milkproteins on urea-polyacrylamide gel elec-trophoresis. J Dairy Scî 75,1204-1210

van Hooydonk ACM, Olieman C (1982) A rapid sen-sitive high-performance liquid chromatographiemethod of following the action of chymosin inmilk. Neth Milk Dairy J 36, 153-158

Venema DP, Herstel H, Elenbaas HL (1987) Deter-mination of ripening time of Edam and Goudacheese by chemical analysis. Neth Milk Dairy J41,215-226

Vijayalakshmi MA, Lemieux L, Amiot J (1986) Highperformance size exclusion liquid chrornatogra-phy of small molecular weight peptides from pro-tein hydrolysates using methanol as mobile phaseadditive. J Liq Chromatogr 9,3559-3576

Visser FMW (1977) Contribution of enzymes fromrennet, starter bacteria and milk to proteolysis andflavour development in Gouda cheese. 3. Proteinbreakdown, analysis of the soluble nitrogen andamino acid fractions. Neth Milk Dairy J 31,210-239

Visser S, Slangen KJ, Hup G (1975) Sorne bitter pep-tides from rennet-treated casein. A method for theirpurification, utilizing chromatographie separationon silica gel. Neth Milk Dairy J 29, 319-334

Visser S, Hup G, Exterkate FA, Stadhouders J (1983a)Bitter peptides in cheese. 2. Model studies on theformation and degradation of bitter peptides byproteolytic enzymes from calf rennet, starter cellsand starter cell fractions. Neth Milk Dairy J 37,169-180

Visser S, Slangen KJ, Hup G, Stadhouders J (I983b)Bitter flavour in cheese. 3. Comparative gel-chro-matographic analysis of hydrophobie peptide frac-tions from twelve Gouda-type cheeses and identi-fication of bitter peptides isolated from a cheesemade with Streptoeoeeus eremoris strain HP.Neth Milk Dairy J 37, 181-192

Vreeman MJ, van Riel JAM (1990) The large-scaleisolation of usrcasein from bovine casein.Neth Milk Dairy J 44,43-48

Vreeman WJ, Visser S, Slangen CJ, van Riel JAM(1986) Characterization of bovine x-cascin frac-tions and the kinetics of chymosin-inducedmacropeptide release from carbohydrate- free andcarbohydrate-containing fractions determined byhigh performance gel permeation chromatogra-phy. Bioehem J 240, 87-97

Wallace JM, Fox PF (1994) Comparison of differentmethods for quantitation of peptideslamino acids inthe water-soluble fraction of Cheddar cheese. In:Proe FLAIR Meeting, Lisbon, 15-18 April 1994,41-58

Weaver JC, Kroger M, Thompson MP (1978) Freeamino acid and rheological measurement onhydrolysed lactose Cheddar cheese during ripening.J Food Sei 43,579-583

Weigele M, de Bernardo SL, Tengi JP, LeimgruberW (1972) A novel reagent for the fluorometricassay of primary amines. J Am Chem Soc 94,5927-5928

Wilkinson MG, Guinee TP, O'Callaghan DM, FoxPF (1992) Effect of commercial enzymes on pro-teolysis and ripening in Cheddar cheese. Lait 72,449-459

Wilkinson MG, Guinee TP, O'Callaghan DM, FoxPF (1994) Autolysis and proteolysis in differentstrains of starter bacteria during Cheddar cheeseripening. J Dairy Res 61, 249-262

Wood AF, Aston JW, Douglas GK (1985) The deter-mination offree amino acids in cheese by capillarycolumn gas-liquid chromatography. Aust J DairyTeehnoI40,166-169

Young PM, Astephen NE, Wheat TE (1992) Effects ofpH and buffer composition on peptide separationsby high-performance liquid chromatography andcapillary electrophoresis. LC-GC 10,26-30

Yun 11, Barbano DM, Kindstedt PS, Larose KL (1995)Mozzarella cheese: impact of whey pH at drain-ing on chemical composition, proteolysis, and func-tional properties. J Dairy Sei 78, 1-7

Yvon M, Chabanet C, Pélissier JP (1989) Solubility ofpeptides in trichloroacetic acid (TCA) solutions.Int J Peptide Protein Res 34, 166-176

Zarmpoutis IV, McSweeney PLH, Beechinor J, Fox PF(1996) Proteolysis in the Irish farmhouse bluecheese, Chetwynd. Irish J Agrie Food Res 35,25-36

Zeece M (1992) Capillary electrophoresis: a new ana-Iytical tool for food science. Trends Food SeiTechnol 3, 6-10

chemical methods for the characterization of proteolysis

Documents