Characterization of casein phosphopeptides prepared using alcalase: Determination of enzyme specificity

ELSEVIER

Characterization of casein phosphopeptides prepared using alcalase: Determination of enzyme specificity Nicholas J. Adamson and Eric C. Reynolds Biochemistry Melbourne,

and Molecular

Biology Unit, School of Dental Science,

University of Melbourne,

Australia

Tryptic casein phosphopeptides containing the cluster sequence-Ser(P)-Ser(P)-Ser(P)-Glu-Gluhave been shown to stabilize amorphous calcium phosphate at neutral and alkaline pH and be anticariogenic in various in vitro, animal and human experiments. Furthermore, metal ion complexes of the casein phosphopeptides (CPPs) have potential as dietetic supplements to increase the bioavailability of calcium, iron, and other essential metal ions. In this study, we have used a Ca’+lethanol selective precipitation procedure to produce a range of phosphopeptides from an alcalase digest of whole casein. The CPPs released by alcalase were truncated relative to those which are released by ttypsin. The peptides could be grouped into those containing the cluster sequence as well as the group of tri-, di-, and monophosphorylated peptides. The two groups contained a number of homologous peptides of varying lengths resulting from the broad specificity of alcalase. Alcalase was observed to cleave peptide bonds on the carboxyl side of Glu, Met, Leu, Tyr, Lys, and Gin; however, of the twenty-six different cleavage sites, seventeen contained a Glu in the P, position and of these, fifteen contained a hydrophobic residue in either the Pi or Pi positions. Furthermore, of the twenty-six cleavage sites identified, twenty-two contained a hydrophobic residue in either the P; or Pi positions. Of the four other sites cleaved by alcalase, two contained a hydrophobic residue in the P; position and one a hydrophobic residue in the P, position.

Keywords: sequencing;

Casein phosphopeptides; alcalase; reversed-phase HPLC; PPLC

subtilisin Carl&erg;

Introduction Based on the published amino acid sequences of the bovine caseins (o,i, os2, l3, and K) reviewed in Swaisgood,’ tryptic digests of whole bovine casein (CN) should contain many different phosphorylated peptides: p-CN-4P( l-25), a,,-CN-5P(59-79), IX,,-CN-4P( l-21), and or,,-CN-4P(4670), all containing the cluster sequence -Ser(P)-Ser(P)Ser(P)-Glu-Glu-; the diphosphorylated peptides o,,-CN126-136), and the 2P(43-58) and CX,*- CN-2P( monophosphorylated peptides @-CN-lP(33-48), a,,-CN-

Address reprint requests to Dr. Eric C. Reynolds, Biochemistry and Molecular Biology Unit, School of Dental Science, University of Melbourne, 711 Elizabeth Street, Melbourne 3000, Australia Received 14 June 1995; revised 12 October 1995; accepted 25 October 1995

Enzyme and Microbial Technology 19:202-207, 1996 0 1996 by Elsevier Science Inc. 655 Avenue of the Americas, New York, NY 10010

characterization;

amino acid analysis;

peptide

lP(106-119), o,,-CN-lP(l38-149), o,,-CN-lP(37-42), and K-CN-lP( 117-169). Those peptides containing the cluster sequence have been shown to stabilize amorphous calcium phosphate at neutral and alkaline pH and to be anticariogenic in various in vitro, animal, and human experiments. 2,3*4The proposed mechanism of anticariogenicity of the cluster peptides is that they localize amorphous calcium phosphate in dental plaque. The amorphous calcium phosphate acts as a buffer of free calcium and phosphate ion species in plaque, thereby preventing tooth enamel demineralization by acid from plaque bacteria. Other workers have shown that casein phosphopeptides (CPP) are formed in viva by normal digestion of casein. Since they are relatively resistant to further proteolytic degradation, they can accumulate in the distal portion of the small intestine.5 It has been proposed that this accumulation together with the ability of the peptides to form soluble complexes with calcium phosphate are responsible for the

0141-0229/96/$15.00

SSDI 0141-0229(95)00232-4

Characterization

of casein phosphopeptides:

enhanced intestinal calcium absorption that has been observed even in vitamin D deficient animals.5-11 CPP-metal ion complexes, therefore, have potential as dietetic supplements to increase the bioavailability of calcium, iron, and other essential metal ions. The unique properties of CPP have led to much interest in the isolation of phosphopeptide fractions and individual phosphopeptides from enzymatic digests of whole casein and purified casein components2*4,‘2-22; however, much of this work has involved the use of analytic trypsin and/or purified casein components. 12-19 For the development of a commercial process for the production of CPP, it would be more cost effective to use an industrial- or food-grade enzyme for the hydrolysis of whole casein. Alcalase, whose main enzyme component is subtilisin Carlsberg, is an industrial- and food-grade enzyme preparation produced by a selected strain of Bacillus licheniformis. Although alcalase has been used extensively in various biotechnological applications, the specificity of this enzyme has not been comprehensively characterized. Subtilisin Carlsberg is an endoproteinase with broad specificity and has been shown to cleave the peptide bonds at GlnQHis5, Ser9-HislO, Leu15-Tyr16, and Tyr26-Thr27 when incubated with the oxidized B-chain of insulin for 4 h.23 Further, the Leu’5-Tyr’6 bond was observed to be cleaved faster than any other bond in the B-chain. Alcalase is considerably less expensive than trypsin and, therefore, would be more cost effective when considering the development of a commercial production process for the CPP; however, the ability to prepare CPP from alcalase digests of casein has not been investigated. In this study, the CPP produced from casein digested by alcalase have been characterized and the specificity of the enzyme determined.

Materials and methods Sodium caseinate was obtained from Murray Goulbum Cooperative (Melbourne, Victoria, Australia). Alcalase Food Grade 2.4L was obtained from Novo Nordisk Bioindustrial PL (product sheet B318, North Rocks, New South Wales, Australia). Ethanethiol, propanethiol, and mesityl oxide were purchased from BDH Chemicals (Melbourne, Victoria, Australia). Trifluoroacetic acid (TFA), phenylisothiocyanate (PITC), amino acid standards, triethylamine (TEA, SequanalTM grade), and constant boiling hydrochloric acid were all obtained from Pierce Chemical Co. (Rockford, IL). Chemicals for sequence analysis were supplied by Applied Biosystems Incorporated (Melbourne, Victoria, Australia). Acetonitrile (HPLC grade) was purchased from Mallinckrodt (Paris, KY) and tetrahydrofuran (THF) was purchased from Millipore-Waters (Melbourne, Victoria, Australia). All other chemicals and solvents were the highest purity analytical grade available.

Casein hydrolysis and selective precipitation

of CPP

Sodium caseinate (86% w/w protein, 100 mg ml-‘) was dissolved in deionized water and adjusted to pH 8.0 by adding 5 M NaOH. Alcalase was added at 2.0% (w/w) of caseinate. Hydrolysis was performed at 50°C for 2 h with the pH maintained at 8.0 + 0.1 by the addition of 1 M NaOH and stopped by the addition of 1 M HCl to pH 4.6. Insoluble material formed was removed by centrifugation (12,OOOg). Casein phosphopeptides were precipitated from the hydrolysate by the addition of 10% (w/v) CaCl, to a 100 mM final

N. Adamson

and E. Reynolds

concentration and ethanol to a 50% (v/v) final concentration. The suspension was centrifuged (12,OOOg)and the supematant discarded. The precipitate

was dried and stored at -20°C.

Analysis of CPP using capillary zone electrophoresis Multiple-phosphorylated casein peptides were analyzed using capillary zone electrophoresis (CZE) (model 270A instrument; Applied Biosystems Incorporated) as described by Adamson er al.‘” Samples were dissolved in 20 mM sodium tetraborate pH 9.2 (running buffer). Separation conditions consisted of 30 kV applied voltage at 30°C. Electric current during electrophoresis under these conditions was 28 p,A. Samples were introduced at the anode into a capillary of length 72 cm (50 pm ID) by creating an intracapillary vacuum (17 kPa) for a specified time (OS-l.0 s). Peptides were detected by UV absorbance at 200 nm using a variable wavelength detector situated 50 cm along the capillary.

Purification, amino acid composition, analysis of CPP

and sequence

A sample of selective precipitate was dissolved in Milli Q water containing 0.1% (v/v) TFA (solvent A) and was applied to a Millipore-Waters RCM p,BondpakTM preparative reversed-phase column ( 10 km C 18,25 mm x 100 mm). The sample then was eluted using an Applied Biosystems Incorporated 4OOA solvent delivery system to generate a linear gradient of O-30% B over 24 min at a flow rate of 44 ml min-‘. Solvent B was 80% (v/v) acetonitrile and 0.1% (v/v) TFA in Milli Q water. The eluant was monitored using an Applied Biosystems Incorporated 1000s diode array detector with the primary wavelength set at 214 nm and the secondary wavelength set at 280 nm. All peaks were manually collected and dried in a Jouan RClO. 10 rotary evaporator, reconstituted in an equivalent volume of Milli Q water, and analyzed by CZE to check for purity. Fractions containing more than one major peak were further purified using anion-exchange (mono Q) fast protein liquid chromatography as described by Reynolds et al. l!’ Anion-exchange liquid chromatography was carried out using a complete Pharmacia-LKB (Melbourne, Victoria, Australia) fast protein liquid chromatography system. Fractions from reversed-phase HPLC were applied to a HR 5/5 mono Q column and then eluted using a linear gradient from O-100% B over 30 min at a flow rate of 1 ml min.‘. Buffer A was 20 mu NH,HCO, pH 8.0 and 50 mM NaCl in Milli Q water and Buffer B was 20 mM NH,HCO, pH 8.0 and 500 mM NaCl in Milli Q water. The eluant was monitored using a UV-M II fixed wavelength monitor at 214 nm with 5 mm path length. Peaks were collected using a FRAC-100 fraction collector. Fractions collected from anion-exchange fast protein liquid chromatography were applied to an Applied Bioslystems Incorporated Brownlee RP-300 Aquapore analytical reversed-phase column (7 km C8, 4.6 mm x 220 mm), and then eluted using an Applied Biosystems Incorporated 140A solvent delivery system to generate a linear gradient from O-40% B over 30 min at a flow rate of 1 ml min-‘. Solvent A was Milli Q water containing 0.1% (v/v) TFA and solvent B was 80% (v/v) acetonitrile and 0.1% (v/v) TFA in Milli Q water. The eluant was monitored at 214 and 280 nm as described. Peaks were collected manually, dried in a Jouan RCl0.10 rotary evaporator, reconstituted in Milli Q water, and analyzed for purity by CZE. Peptides were redried and stored at -70°C for amino acid composition and sequence analyses as described by Reynolds et al. I9 Prior to sequence analysis, the phosphoseryl residues, which are labile to the Edman chemistry, were converted via p elimination to s-ethyl-cysteinyl residues using ethanethiol’” or to s-propyl-cysteinyl residues using l-propanethiol (Adamson, Riley, and Reynolds. 1995. unpublished data).

Enzyme Microb. Technol.,

1996, vol. 19, August

15

203

Papers Results

0.12

The selective precipitation procedure involving the addition of CaCl, and ethanol to a pH 4.6 clarified alcalase hydrolysate of commercial sodium caseinate resulted in a precipitate that was 12.27 + 0.24% (w/v, n = 3-5) of the original casein. The casein hydrolysate was prepared using alcalase for a 2 h hydrolysis at an E/S ratio of l/50 and 50°C. The gravimetric yield of CPP produced from the alcalase casein hydrolysate compared favorably with the previously determined yield of CPP (11.04 + 0.30% w/w) produced using trypsin under identical conditions.21 Similarly, the level of organic phosphorus (4.73 f 0.15% w/w, n = 3-5) of the CPP prepared using alcalase compared well with the 3.47 & 0.06% (w/w) organic phosphorus level of the tryptic CPP.*’ Previous time-course studies25 of casein hydrolysis using alcalase at varying E/S ratios demonstrated that the gravimetric yield of selectively precipitated CPP was optimal after 2 h of hydrolysis at 50°C at an E/S ratio of l/50. Preparative scale reversed-phase HPLC of the CPP selectively precipitated from the casein alcalase hydrolysate is presented in Figure 1. Peaks were collected and checked for purity using CZE. All peaks were subjected to a second stage of chromatography using anion exchange (mono Q) FPLC. Peak Rl was separated into four major fractions (RlQl, RlQ2, RlQ3, RlQ4). Peaks R2, R3, R4, R5, R8, and peak Rl I each yielded only one major fraction. Peak R6 separated into three major fractions (R6Q1, R6Q2, R6Q3) and peaks R7, R9, and RlO each separated into two fractions (R7Q1, R7Q2, R9Q1, R9Q2, RlOQl, and RlOQ2, respectively). These fractions were collected and desalted using analytical scale reversed-phase HPLC. At this step, fraction R7Q1 resolved further to give R7QlR’ 1 and R7QlR’2 and fraction RlOQ 1 resolved further to give RlOQlR’l, RlOQlR’2, and RlOQlR’3. A total of 22 fractions were collected. All fractions were subjected to amino acid composition and sequence analyses. The primary structures of 19 different peptides identified in these purified fractions are listed in Table 1 along with the scissile peptide bonds at the N- and c-termini for each peptide. Amino acid composition and sequence analyses of the fractions RlQlRlQ4 revealed them to be mixtures of free amino acids and short peptides. The two peptides, o,,--CN4P(f64-70) and p-CN-4P(15-21), contained in peak R2 were not resolved at any stage of purification; however, they were detected during sequence analysis. A range of CPP were recovered and are presented, grouped according to their degree of phosphorylation and homology, in Table 2. No peptides containing the diphosphorylated region, -Ser(P)129-Thr130Ser(P)13’- of a,,-casein, or the monophosphorylated region, -Glu147-Ala-Ser(P)-Pro-Glu151of K-casein, were detected in the fractions collected. Alcalase was observed to have broad specificity cleaving at Glu, Met, Leu, Tyr, Lys, and Gln. Of the twenty-six different sites at which hydrolysis occurred, seventeen contained Glu in the Pi position, three contained Lys, two contained Met, two contained Leu, one contained Tyr, and one contained Gln in the Pi position. Of the seventeen hydrolysis sites in which Glu occupied the P, position, fifteen were observed to contain a hydrophobic residue in either the Pi or Pi positions. Furthermore, of the twenty-six different 204

Enzyme Microb. Technol.,

1996, vol. 19, August

R5

Rll

0.10 RIO

I U.08

!I 2 8

R7

R3

R6

R9

0.06

I

9

-e 3

2 0.04

0.02

O.GQ 11

14

17

20

23

26

25,

Retention Time (min) Figure 1 Reversed-phase HPLC of Ca-CPP complexes selectively precipitated from an acid-clarified alcalase casein digest at pH 4.6. Peak RI, free amino acids and short peptides; peak R2, a,,-CN-4P(f64-70)*, p-CN-4P(fl5-21)*; peak R3, a,,-CN-3P(f512)‘; peak R4, a,,-CN-4P(f53-63)*; peak R5, a,,-CN-4P(f6170)*; peak R6, a,,-CN-2P(f43-501, a,,-CN-2P(f43-521, a,,-CN4P(f61-69)*; peak R7, p-CN-lP(f33-42), a,,-CN-2P(f41-501, a,,CN-2P(f41-52); peak R8, fSerWIIcS,-CN-3P(f40-52); peak R9, 5-CN-lP(f32-371, a,,-CN-5P(f52-63)*; peak RIO, CX,~-CNlP(f138-145), u,.-CN-lP(flll-118), 5-CN-4PIfl2-21)*, [Glu3s]9-CN-lP(f32-42); and peak Rll, 9-CN-4Pff7-21)“. Asterisk denotes Ser(P) cluster peptide

sites of hydrolysis, twenty-two were observed to contain a hydrophobic residue in either the Pi or P; position. Of the four other sites cleaved by alcalase, two contained a hydrophobic residue in the Pi position and one a hydrophobic residue in the P, position.

Discussion The Ca*+/ethanol selective precipitation procedure, when performed on an alcalase digest of whole bovine casein, resulted in the recovery of a series of peptides that contained all of the expected phosphorylated regions of the caseins, except the regions -Ser(P)129-Thr130-Ser(p)‘31of o,*casein and -Glu147-Ala-Ser(P)-Pro-Glu151of K-casein. Due to the broad specificity of alcalase, a wide range of phosphopeptides were recovered including homologous peptides of varying lengths (Table 2). Previous studies with the synthetic octapeptide Ac-Glu-Ser(P)-Ile-Ser(P)15

Characterization Table 1 pH 4.6

Assignment

of casein phosphopeptides:

and properties of CPP that were selectively

precipitated

IV. Adamson

and E. Reynolds

using Ca2+ and ethanol from a casein alcalase digest at

Scissile peptide bond

Peak RlsQlR1Q4b R2SlC R2S2 R3 R4 R5 R6Ql R6Q2 R6Q3 R7Q1R’ld R7QlR’2 R7Q2 R8Ql R9Ql R9Q2 RIOQIR’I RlOQlR’2 RlOQ’lR’3 R10Q2 Rll

Free amino acids and short peptides a,,-CN-4P(f64-70) 5-CN-4P(f 15-21) a,,-CN-3P(f5-12) a,,-CN-4P(f53-63) a,,-CN-4P(f61-70) a,,-CN-2P(f43-50) a,,-CN-2P(f43-52) a,,-CN-4P(f61-69) 5-CN-1 P(f33-42) a,,-CN-2P(f41-50) a,,-CN-2P(f41-52) [Ser(W’a,,-CN-3P(f40-52) 5-CN-1 P(f32-37) a,,-CN-4Ptf52-63) CX,,-CN-lP(f138-145) a,,--CN-1 P(fl1 l-l 18) [Glu3*]3-CN-lP(f32-42) 5-CN-4P(f12-21) 5-CN-4P(f7-21)

Table 2 Comparison of homologous CPP that were selectively precipitated using Ca” and ethanol from a casein alcalase digest at pli 4.6

Cluster a,,-CN-4P(f64-70) a,,-CN-4P(f61-69) cx,,-CN-4P(f61-70) 5-CN-4P(fl5-21) 5-CN-4P(f12-21) 5-CN-4P(fl-21) a,,-CN-3P(f5-12) a,,-CN-4Pff53-63) a,,-CN-4P(f52-63) Noncluster a,,-CN-2P(f43-50) a,,-CN-2P(f43-52) a,,-CN-2P(f41-50) a,,-CN-2P(f41-52) u,,-CN-3P(f40-52) 3-CN-1 P(f32-37) S-CN-1 P(f33-42) [Glu3*lf3-CN-lP(f32-42) CX,,-CN-1 P(fl1 l-l 18) (u,,-CN-lP(f138-145)

E 812 E-XL_% M-EHV Y-SIG M-EAE K-DIG K-DIG M-EAE K-FQZ L-SKD L-SKD E-LSK E-KFQ E-YSI K-TVD E-IVP E-KFQ E-WE L-NVP

CIZI;XEE ZLHXHEE EHVXBEE SIGSZZEEBAE EAEZIZZBEE DIGZEZTE DIGZEPTEDQ EAESIXZE FQXEEQQQTE SKDIGZEHTE SKDIGXEBTEDQ LsKDIGHEZTEDQ KFQIZEE YSIGBBZEEZAE TVDMEPTE IVPNZAEE KFQXEEEQQTE IVEXLBPZEE NVPGQEIVEHLZHZEE

‘R refers to preparative reversed-phase HPLC fraction number bQ refers to anion exchange FPLC fraction number “S refers to peptides that coeluted at all stages of purification and were identified dR’ refers to‘analytical reversed-phase HPLCfraction number

Peptide

N-terminal P,-P,‘P,‘P,’

Primary structure C = Ser(P)

Assignment

Primary structure B = Ser(P)

ZIHBZEE EAEHIZHZE EAEXZSSEE ZLZPZEE IVEXLPZSEE NVPGQEIVEZLIXPEE EHVZZBEE SIGZZSEESAE YSIGCZSEEBAE DIGZEZTE DIGZEZTEDQ SKDIGSEHTE SKDIGZEZTEDQ LBKDIGZEZTEDQ KFQXEE FQZEEQQQTE KFQBEEEQQTE IVPNZAEE TVDMEBTE

Ser(P)-Ser(P)-Glu-Glu-NHMe, which corresponds to the fragments c1,I-CN--4P(f63-70) and P-CNAP(f 14-2 1) with a conservative substitution of Leu for IIe in the latter, have

P,-P,‘P,‘P,’

E-IVP E-SlT E-SII E-VAT E-IVP E-DQA Q-AME E-EIV E-DEL E-DQA Q-AME Q-AME E-EQQ E-VAT E-VFT E-RLH E-DEL E-SIT E-SIT

by sequence analysis

shown the peptide to have significant calcium phosphate binding activity and anticaries activity in the rat mode1.26 In the present study, greater than 70% mol mol-’ (based on absorption at 2 14 nm) of the peptides recovered contained at least the full octapeptide cluster sequence indicating that CPP produced using alcalase should have significant biological activity. The CPP released by alcalase were truncated relative to those released by the action of trypsin,” In the previous study with trypsin, l9 phosphopeptides were recovered from an acid-clarified tryptic digest of whole casein using multiple sequential selective precipitation procedures at various pH including 2.0, 3.5, 4.6, and 8.0. At pH 2.0 and 3.5, only the cluster peptides containing the sequence -Ser(P)Ser(P)-Ser(P)-Glu-Gluwere recovered. At pH 4.6, both cluster peptides and the diphosphotylated peptides cx,,-CN2P(f43-58), containing the sequence -Ser(P)-Glu-Ser(P)-Thr-Gluand cl,,-CN-2P( 126-136), containing the sequence -Ser(P)-Thr-Ser(P)-Glu-Glu were recovered. Precipitation at pH 8.0 resulted in the precipitation of all cluster and diphosphorylated peptides as well as the additional recovery of the monophosphorylated pepdde p-CN-lP(f3348) which contains the sequence -Ser(P)-Glu-Glu-. The monophosphorylated peptides cl,,-CN-lP(fl066I 19) VPQLEIVPNXAEER, au,,-CN,,(f138-149) TVDMECTEVFTK, a&N-lP(f37112) VNELCK, and K-CN-lP(fl17-169) TEIPTINTIASGEPTSTPIEAVESTVATLEACPEVIESPPEINTVQVTSTAV where 2 = Ser(P) were not recovered in this previous study, suggesting

Enzyme Microb. Technol.,

1996, vol. 19, August

15

205

Papers that for a monophosphorylated peptide, the minimal motif for calcium/ethanol-induced precipitation is -Ser(P)-GluGlu-. In the present study, however, the peptides B-CNlP(f3242) and related peptides containing the sequence -Ser(P)-Glu-Glu-, o&N-lP(fl38-145) containing the sequence -Glu-Ser(P)-Thr-Gluand a,'-CN-lP(f1 1 l118) containing the sequence -Se@)-Ala-Glu-Gluwere precipitated by Caz+/ethanol from the acid-clarified casein alcalase digest at pH 4.6. This was consistent with the proposition that a Ser(P) residue with neighboring acidic residues is important for precipitation; however, the &r(P) and Glu residues are not required to conform specifically to the contiguous motif -Ser(P)-Glu-Glu in order for Ca2+ cross-linking to occur. Interestingly, a peptide containing the monophosphorylated sequence -Glu’47-Ala-Ser(P)Pro-Glu”‘of K-casein was not recovered, possibly suggesting some restriction on the position of the acidic residues in relation to Ser(P) for effective Ca2+ cross-linking. The monophosphorylated peptides as,-CN-lP(f138149) and ~Y,,-CN-1P(f10~119), found in the precipitate from the alcalase casein digest, were not precipitated on the addition of calcium and ethanol to the tryptic casein hydrolysate described in the previous study.” One possible explanation for this could be the lack of accessibility of trypsin to the required cleavage sites preventing the peptides release. Alcalase, with its broader specificity and preference for hydrophobic residues, might more easily expose hydrolysis sites otherwise buried; however, another possible explanation is that the shorter cluster peptides released by alcalase bound less Ca2+ than the longer tryptic cluster peptides, leaving more Ca2+ available for effective crosslinking of the monophosphorylated peptides for the same level of added CaCl,. The synthetic octapeptide [GluSer(P)-Ile-Ser(P)-Ser(P)-Ser(P)-Glu-Glu] has been shown to bind less calcium per mole than the tryptic peptides (-Ys1-CN-5P(f59-79)a and B-CN-4P(fl-25),b suggesting that the sequences -Ile-Val-Pro-Asn-Ser(P)-ValGlu-Glnand -Glu-Leu-Glu-Glu-Leu-Asn-Val-ProGly-Glu-Ile-Valof a,,-CN-5P(f59-79) and B-CN4P(fl-25) respectively, are necessary together with the Ser(P) cluster sequence [Glu-Ser(P)-Ile/Leu-Ser(P)Ser(P)-Ser(P)-Glu-Glu] for full calcium binding.26 This suggests that the optimum Ca2+ concentration for selective precipitation is dependent upon the peptides released and therefore should be optimized for different enzymes and for hydrolyses which are performed under different conditions. OL,~-CN-2P( 126-l 36) and CY,,-CNThe peptides 2P(126--135), both containing the diphosphorylated region -Ser(P)‘29-Thr’30-Ser(P)131of a,,-casein, have been detected in significant proportions in selective precipitates of CPP from tryptic and pancreatic digests of casein, respectively.19*2’,22 Failure to detect a peptide containing this diphosphorylated region in any of the fractions collected in the present study is most likely attributable to the broad specificity of alcalase resulting in the release of a series of

“Gln-Met-Glu-AlaClu-Ser(P)-Ile-Ser(P)-Ser(~~Ser(~)~lu-Glu-IleVal-PrwAsn-Se@-Val-Glu-Gln-Lys bArg-Glu-Leu-Glu-Glu-Leu-Asn-Val-Pro-Gly-G1u-Ile-Val-GluSer(P)-Leu-Ser(P)-Ser(P)-Ser(P)Glu-Glu-Srg

206

Enzyme Microb. Technol.,

1996, vol. 19, August

peptides which appeared as minor reversed-phase HPLC and FPLC peaks and were not collected. A study of the sequence of ar,,-casein’ revealed a number of potential alcalase cleavage sites, based on the specificity observed in this study, around the diphosphorylated cluster -Ser(P)129Thr’30-Ser(P)‘31-. Alcalase is produced from a selected strain of B. licheniformis. The main enzyme component, subtilisin Carlsberg, is an endoproteinase which has been previously reported to cleave the peptide bonds of the oxidized B-chain of insulin Gln4-His5, Ser9-His’O, Leu”-Va112, Leu15-Tyr16, and Tyr26-Th?7 when incubated with the substrate for 4 h.23 Further, when the initial stages of hydrolysis were investigated in the previous study, the Leu’5-Tyr16 bond was reported to have been cleaved more rapidly than any other bond in the B-chain, The specificity of alcalase observed in our study for Met, Leu, and Tyr in the P, position is not out of character with these previous findings; however, the specificity of the enzyme for Glu in the P, position found in the present study, to our knowledge, has not been reported previously. Interestingly, the hydrolysis of the oxidized B-chain of insulin by alcalase resulted in cleavage at the polar residues Gln-4 and Ser-9 and in both cases a hydrophobic residue occupied the Pi position. In our study, of the seventeen hydrolysis sites in which Glu occupied the P, position, only two did not contain a hydrophobic residue in either the Pi or Pi positions; however, one of these two sites did contain Leu in the Pi position. Furthermore, in the case where cleavage occurred with Gln in the P, position, there also existed a hydrophobic residue in the Pi position.

Conclusion Ca2+/ethanol-selective precipitation of CPP from an alcalase digest of casein resulted in the recovery of a range of variably truncated CPP. Alcalase, therefore, would be a costeffective enzyme for the production of truncated CPP at a commercial scale. A study on the specificity of alcalase revealed a preference for sites containing hydrophobic residues in either the Pi or P; positions, particularly when Glu was in the P, position.

Nomenclature CN = CPP = CZE = PITC = TEA = TFA = THF =

casein casein phosphopeptides capillary zone electrophoresis phenylisothiocyanate triethylamine trifluoroacetic acid tetrahydrofuran

Acknowledgments This work was performed under the tenure of a Dairy Research and Development Corporation Fellowship Award to NJA. The technical assistance of Mr P. F. Riley is gratefully acknowledged.

References 1.

15

Swaisgood, H. E. Chemistry of milk protein. In: Developments in Dain, Chemistry. Vol. 1 Proteins (Fox, P. F., Ed.). Applied Science Publishers, London, 1982, l-70

Characterization 2. 3.

4.

5.

6.

I.

8.

9

10

11

12

13

14

of casein phosphopeptides:

Reeves, R. E. and Latour, N. G. Calcium phosphate sequestering phosphopeptide from casein. Science 1958, 128, 472 Reynolds, E. C. The prevention of sub-surface demineralization of bovine enamel and change in plaque composition by casein in an intra-oral model. J. Dent. Res. 1987, 66, 1120-I 127 Reynolds, E. C. Anticariogenic phosphopeptides, U.S. patent 5,015,628; and University of Melbourne and Victorian Dairy Industry Authority (assignees), May 14, 1991 Meisel, H. and Frister, H. Chemical characterization of a caseinophosphopeptide isolated from in vivo digests of a casein diet. Biol. Chem. Hoppe-Seyler 1988, 369, 1275-1279 Lee, Y. S., Noguchi, T.. and Naito, H. Phosphopeptides and soluble calcium in the small intestine of rats given a casein diet. Br. .I. Nutr. 1980. 43,457-467 Lee. Y. S., Noguchi, T., and Naito, H. Intestinal absorption of calcium in rats given diets containing casein or amino acid mixture: The role of casein phosphopeptides. Br. J. Nurr. 1983, 49, 67-76 Mykkanen, H. M. and Wassermann, R. H. Enhanced absorption of calcium by casein phosphopeptides in rachitic and normal chicks. J. Nurr. 1980, 110,2141-2148 Naito, H.. Kawakami, A., and Imamura, T. In vivo formation of phosphopeptide with calcium-binding property in the small intestinal tract of the rat fed on casein. Agric. Biol. Chem. 1972, 36, 4094 15 Naito, H. and Suzuki, H. Further evidence for the in vivo formation of phosphopeptide in the intestinal lumen from dietary E-casein. Agric. Biol. Chem. 1974, 38, 1543-1545 Sato, R., Noguchi, T., and Naito, H. Casein phosphopeptide (CPP) enhances calcium absorption from the ligated segment of rat small intestine. J. Nutr. Sri. Vitaminol. 1986, 32, 67-76 Peterson, R. F., Nauman, L. W., and McMeekin, T. L. The separation and amino acid composition of a pure phosphopeptone prepared from E-casein by the action of trypsin. J. Am. Chem. Sot. 1958.80, 95-99 Osterberg, R. Isolation of a phosphopeptide as magnesium complex from a trypsin hydrolysate of axasein by anion exchange chromatography. Biochim. Biophys. Acta 1960, 42, 312 Manson. W. and Annan, W. D. The structure of a phosphopeptide derived from E-casein. Arch. Biochem. Biophys. 1971, 145, 16-26

N. Adamson

and E. Reynolds

15.

West, D. W. A simple method for the isolation of a phosphopeptide from bovine a,,-casein. J. Da@ Res. 1977. 44, 313-376

16.

Leonil, J.. Molle, D., and Maubois, J. L. A study of the early stages of @casein hydrolysis. LX Lait 1988, 68, 28 l-294

17.

Juillerat, M. A., Baechler, R., Berrocal, R.. Chanton, S., Scherz, J.-C., and Jost, R. Tryptic phosphopeptides from whole casein. I. Preparation and analysis by fast protein liquid chromatography. _r. Dairy Res. 1989, 56, 603-611

18.

Scanff. P., Yvon, M., and Pelissier, J. P. Immobilized Fe’+ affinity chromatographic isolation of phosphopeptides. J. Chromatogr. 1991. 539,425-432

19.

Reynolds, E. C.. Riley, P. F., and Adamson. N. J. A selective precipitation purification procedure for multiple phosphoseryl-containing peptides and methods for their identification. Anal. Biochem. 1994, 217, 277-284

20.

Brule. G., Roger, L., Fauquant, J., and Piot, M. Phosphopeptides from casein based material. U.S. Patent 4,35$,465; Institut National de la Reserche, Agronomique, Paris, France (assignees). 1982

21.

Adamson, N. J. and Reynolds, E. C. Characterization of tryptic casein phosphopeptides prepared under industrially relevant conditions. Biotech. Bioeng. 1995. 45, 196-204

22.

Adamson, N. J. and Reynolds, E. C. Characterization of multiplyphosphorylated peptides selectively precipitated from a pancreatic casein digest. J. Dairy Sci. 1995, in press

23.

Ottesen, M. and Svendsen, I. The subtilisins. In: Methods in Envymology Vol. 19 ProteoLytic Enzymes (Perlmann, G. E. and Lorand, L.. Eds.). Academic Press, New York, 1970. 199-215

24.

Adamson, N., Riley, P. F.. and Reynolds, E. C. The analysis of multiple phosphoseryl-containing peptides using capillary zone electrophoresis. J. Chromatogr. 1993. 646, 391.-396

25.

Adamson, N. J. Characterization of casein phosphopeptides. thesis, School of Dental Science, University of Melbourne,

26.

Reynolds, E. C., Cain, C. J., Webber. F. L., Black, C. L., Riley, P. F.. Johnson, I. H., and Perich, J. W. Anticariogenicity of soluble calcium phosphate complexes of tryptic casein phosphopeptides in the rat. J. Dent. Rex 1995, 74, 1272-1279

Enzyme Microb. Technol.,

1996, vol. 19, August

15

Ph.D. 1995

207