31

Biochimica et Biophysica Acta, 1077(1991)31-34 © 1991 ElsevierSciencePublishersB.V.0167-4838/91/$03.50 ADONIS 016748389100125W BBAPRO 33864

Influence of pH on the structural changes of fl-lactoglobulin

studied by tryptic hydrolysis J e a n - M a r c C h o b e r t , Mich~le D a l g a l a r r o n d o , Eric D u f o u r , C a t h e r i n e B e r t r a n d - H a r b and Tomasz Haertl6 LEIMA.Institut National de la Recherche Agronomique, Nantes (France)

(Received19June 1990) Key words: ~-Lactoglobulin;Trypsinhydrolysis;Conformation;(Bovine) The attempt to use trypsin in order to monitor pH (7.5-9.0) induced/]-Iactoglobulin co~ormatinn changes has revealed differences in the cleavage of specific sites. The tryptic cleavage of two dibasic X-Lys-Lys-Y sites (Lys 69, 70 and 100, 101) shows slighter predominance of symmetrical cut at pH 7.5 and 8.0. Mostly asymmetrical cleavage yielding two C-terminal lyslnes can be observed at pH 8.5 and 9.0. Atypical cleavage of the Tyr-20-Ser-21 site, which at pH 9.0 is relatively negligible, increases substantially in pH 7.5-8.5. This implies that Tyr-20 probably is the tyroslne reported to be exposed on the surface of the protein during transformation of/]-Iactogiolmlin molecule oecmring in the studied pll range OFanford et al. (1959) J. Am. Chem. Soc. 81, 4032-4036).

Introduction In spite of relatively long acquired knowledge that /]-lactogiobulin interacts strongly with retinol [1], the exact physiological role of/~-lactogiobulin is unknown. This small protein is an abundant component of the milk (whey) of several mammals [2]. It has been recently claimed that /]-lactogiobulin belongs to the "superfamily' of proteins involved in strong interactions with small volatile hydrophobic figands [2-4]. The three-dimensional structures of these proteins overlap in more than 95~ and constitute a hydrophobic pocket inside of an eight-stranded/]-barrel bordered on one side by an a-helix. It has been suggested that this kind of/]-barrel structure might be a general structural device found in animal organisms used to trap and transport the small hydrophobic ligands. /]-laetoglobulin is resistant to pepsin hydrolysis in the stomac [5,6]. There is also evidence of intestinal transloeation of this resistant to the acid proteolysis protein in humans. This can be deduced from reports about the presence of diet-dependent/~-lactogiobulin antigens in maternal milk [7]. All these observations indicate that /]-lactogiobulin might pass, at least partially, across different membrane interfaces. Hence, it is important to investigate the be-

Correspondence':J.M. Chobeftand T. Haertl~, LEIMA-INRA,BP 527,44026Nant*.sCedex03, France.

haviour of this protein and its degradation when it is subject to pH changes as this may simulate, to some extent, the conditions met before transloeation through biological membranes takes place. The structural changes of/]-lactoglobulin under the influence of pH have been studied previously by Tanford et al. [8] who have observed a structural transformation of this prorein exposing at pH 8.0 a tyrosine moiety otherwise hindered at lower pH values. Increase of pH above 8.0 induces further structural changes [8,9] which are becoming irreversible around pH 9.5 when the/]-lactoglobulin tyrosine residues start to ionize [10]. Ample structural and biophysical data available for the /]-lactogiobulin molecule provide the basis for a study of the correlation of its structural transformations with the accessibility of the molecule of this protein to proteolytic hydrolysis. Bovine /]-lactoglobalin (variant B) is small protein composed from 162 amino acids wit~ 17 potential tryptic cleavage sites. Its molecule contains two disulfide bridges as well a s one free cysteine. The studies of the sequence of some cleavage events have been performed and related to the intramolecular transformation of this protein around pH 7-9. Materials and Methods Organic solvents from Carlo Erba (Italy) were used for HPLC. All other reagents were of analytical grade. Buffers and solvents for HPLC were filtered through

32 Millipore 0.45/~m fdters (Millipore, Bedford. MA) and degassed under vacuum before use. Trypsin treated with tosyl phenylalanine chloromethyi ketone [TPCK] (13000-14000 BAEE U/rag) was obtained from Sigma. Bovine ~-lactogiobulin (variant B) was prepared according to the method of Mailliart and Ribadeau Dumas [11] from the whey of a cow homozygous for this protein.

Hydrolysis of [3-1actoglobulin by the TPCK-treated trypsin ~-Lactoglobulin (1 m g / m l - 5 . 5 . 1 0 -5 M) was dissolved in 0.1 M Tris-HCl buffer (pH 7.5, 8.0, 8.5 or 9.0). TPCK-trypsin, previously solubilized in 10 mM HCI (1 mg/ml) was added to the reaction mixture at an E / S ratio of 2.5~. The mixture was incubated at 37°C. Aliquots were taken at intervals (1, 2, 4 and 9 h) and the reaction stopped by the addition of 0.2 M HCI (fmal concentration in 18-lactogiobulin = 2.78-10 -5 M; final concentration in HC1 =0.1 M). The major detected tryptic peptides were numbered according to the 17 potential tryptic cleavage sites (Fig. 1) as previously described 112].

High-performance liquid chromatography (HPLC) Tryptic peptides of bovine/]-lactoglobuiin were separated by reversed-phase HPLC (RP-HPLC) on a Nucleosil C-18 column (4.6 mm i.d.x25 cm, SFCC, Gagny, France) equilibrated in solvent A (0.115~ trifluoroacefic acid (TFA) in H20, pH 2.5). The column was elated with a gradient (Table I) of solvent B (60~ acetonitrile/40~o H 2 0 / 0 . 1 ~ TFA) in solvent A for 62 rain. The temperature of column and solvents was maintained at 00°C. The flow rate was 1 ml/min and the absorbance was recorded at 214 and 280 nm. Results and Discussion

Tryptic hydro(ysis of ~-lactoglobulin The gradual disappearance of the whole /]-lactoglobulin is shown in Fig. 2. After 1 h hydrolysis, 86, 13, 30 and 39% residual /]-lactogiobulin were still present when proteolysis was performed at pH 7.5, 8.0, 8.5 and 9.0, respectively. After 9 h proteolysis at pH 8.0, 8.5 or

TABLE ! Ehaion gradientfor tryptic pepdde purificmion

Tune

Flowrate

(rain)

(ml/min)

0.0 9.0 45.0 52.0 62.0 65.0 65-5

1.0 1.0 1.0 1.0 1.0 1.0 1.0

~A

%B

100 77.5 50.0 40.0 0 0 100

0 22.5 50.0 60.0 100 100 0

-

6 6 6 6 6 6

9.0, ~-lactoglobulin completely disappeared, while 21~ of residual /~-lactogiobulin remained unhydrolysed at pH 7.5. Since/~-lactogiobulin was not reduced, peptides 5 and 16 (Fig. 1) were still linked by a disulfide bridge Cys-66-Cys-160.

Influence of the pH on the yield of main tryptic peptides As shown in Fig. 2, the appearance of peptides 3 (Val-15-Arg-40), 4-5 Lys + 16 (Val-41-Lys-70 + Len149-11e-162) and 10 Lys (Val-92-Lys-101) was very slow when tryptic hydrolysis was performed at pH 7.5, as compared to pH 8.0, 8.5 or 9.0. An atypical cleavage of the Tyr-20-Ser-21 bond of the peptide 3 (Val-15-Arg-40) giving two peptides: peptide 3' (Ser-21-A~40) and peptide 3" (Val-15Tyr-20) has been already observed [12]. As shown in Fig. 3, the disappearance of peptide 3 and the concomitant appearance of peptides 3" and 3" could be easily detected during hydrolysis at pH 8.0 or 8.5. In contrast, at p H 9.0, the cleavage of the Tyr-20-Ser-21 bond seemed very limited even after 9 h of hydrolysis. Apparently the conformational transition, observed in the pH range 7.0-9.0 [8] delays considerably the atypical cleavage at Tyr-20-Ser-21 site. The molecule of i8-1actogiobulin B has two dibasic X-Lys-Lys-Y sites. Their fate may be followed measuring the yield of tryptic peptides 5 Lys + 16 (Trp-61-Lys 69,70 + Len-149-Ile-162), 5 + 16 (Trp-61-Lys-69 + Leu-149-lle-162) and 10 Lys (Val-92-Lys 100,101), 10 (Val-92-Lys-100), respectively. Significant shift of

1

I,¢~-r/e-Val-'l~--Gla-Tl~- Met-Lvs.Gly.Lea-Asv-IIe-Gln.L_vs-V~-AIa-GIy-~-T~T~-L-J-~-~-~1

Curve (Waterssystem)

2

3" •

~-$~-~- ~-$=-~J-

• 3'

3

-:~o-Ah-Gln-$er-Ala-Pro-Leu-mArg-V~l-Tyr-Va]-Gitt-G]u-Leu-L~-Pro-'l'~-Pm~u~y-~-~m-~-~-~m-L~-T~;u-

~.~.~.LÈ.~°.~°.~.c~°.'~.~.L".e.~.~.~.v~.~.L:.~.~..~.~.~.~r~..~.L`~.`.~.~-"'-"~'--"

1:~4

12

"'~"

13

,,

14

9

.....

15

[Thr-Gia~l.e'J-G. Iu-GIu-Gla-Cvs.Hi~.r~OH I

m

,

I

Fig. 1. Trypticpepfidesin ~lactoglobulinB primarystructure.The dashesindicatethat somemoleculesmay have the singlesulphydrylgroup at position119 and othersat 121.

33 0.16 -

A.U. 0

0

A.U.

-s

4

6

0

&U.

2

4

6

.

O

8

~

C

0

8

~

0

2

4

6

I

2

4

6

8

"1

8

0

"I'~IE(hours) TIME(hours) Fig. 2. T'tmeevolutionof 18-1actcglobu~[A],peptide 4--5 Lys+ 16 [B],peptide 3 [C] and peptide 10 Lys (D], studiedin time by RP-HPLCat 214 rim.I pH 7.5.13pH 8.0, 4>pH 8.5 and O pH 9.0. A.U. = arbitrary units normalizedaccordingto the contributionsm the molar absofbancesat 214 nm of oligo~ptide bonds and amino acid aromaticchromophores,expressedper monomer.

selectivity may be observed when two X-Lys-Lys-Y sites are cleaved in various p H values. Whatever the pH of the trypti¢ hydrolysis (from 7.5 to 9.0), strong prefer-

ence for cleavage yielding two C-terminal lysines can be observed (Fig. 4). The symmetrical and asynm~trical hydrolysis of both Lys-Lys bonds is slow at p H 7.5 and

3"

3'

I

-,-I lI o

L 30

03

Stys +la

I

O

"

o

II

I

I

i s°~e

IN

I

I

l

AA ,..,., mlnlItaI

50

60

Fig. 3. Separationof uyptic hydrolysateof B-lactoglobulinby RP-HPLCafter 9 h hydrolysisat differentpH.

34 eJ6 &U.

C

0.04

0

2

,I

6

8

0

2

4

6

$

A.U. 0.e8[

e

2

4 6 8 0 2 4 6 $ ~(tmm) TAaE(Imm) Fig. 4. Asymmetricaland s~ymmen.icalcleavageof peptid~ 5+ 16 and 10; (peptide5-Lys+16 [A],pepfide5+ 16 [B],pepfide 10-Lys[ q and peptide 10 [D]),studied in time by RP-HPLC. same symbolsand A.U. as in Fig.2. 9.0, as compared to pH 8.0 and 8.5. With the progress of enzymatic action, the hydrolysis of these two bonds kept increasing in time at pH 7.5. At pH 9.0, it was limited in time basically ending after its plateau has been reached. The symmetrical hydrolysis of the Lys100-Lys-101 bond appeared to be about twice quicker than Lys-69-Lys-70. The yield of asynnnetrically cleaved peptides: 10 Lys (Val-92-Lys 100,101) and 5 Lys + 16 (Trp-61-Lys 69,70 + Len-149-Ile-162) starts to drop down at pH 8.0 and 8.5 after about 2 h, possibly because of the spfitting of the distal lysines (70 or 101) and the conversion of the peptides into shorter products. Probably this secondary cleavage is contributing to the increase of the yields of symmetrically cleaved peptides 10 (Val-92-Lys-100) and 5 + 16 (Trp61-Lys-69 + Leu-149-11e-162) after 9 h of hydrolysis time at pH 7.5 and 8.0 (Fig. 3 and 4). pH-dependent activity of trypsin on its substrates (96, 100, 91 and 73~ relative activity are observed at pH 7.5, 8.0, 3.5 and 9.0, respectively [13]) could explain the poor cleavage of the Lys-Lys bonds at pH 9.0. However, the lowest trypsin action on fl-lactoglobulin at pH 7.5 (when the relative activity is still 96~ that observed at the optimal pH 8.0) seems in favor of a conformational rearrangement of fi-lactoglobulin molecnle in this pH domain. It has been demonstrated recently [14] that the cleavage at the dibasic sites by serine endoproteinases may be an important processing aspect of the maturation of the immanoacfive proteins. It may be expected that on one side the tryptic processing of the model protein, such as fl-lactoglobulin can shed some additional fight

on the tryptic activity at dibasic sites which may be modified by the micro pH changes certainly highly dependent on the environment, and on the other hand it might ~ve some additional information on the nature of the pH-induced transformation of fl-lactoglobulin in the studied p H range.

Rderenees

1 Futterman,S. and Heller,J. (1972)J. Biol.Chem.247, 5168-5172. 2 Godovac-Timm,-naan.J. (1988)Trendsin ~ Sci. 13, 64-66. 3 Pervaiz, S. and Brew,IC (1985)Sconce228, 335-337. 4 Papiz, M ~ Sawyer,L, ~ EE, North,A~C.T, Fmdlay, J.B.C, Sivapra~'mran, R., Jones, T.A., Newcomer, M.E. and Kraulis, P.J. (1986)Nature 324, 383-385. 50tani, IL (1981)Jpn. J. Zootech. Sci. 52. 689-691. 6 Yvon,M. Van Hille, L P~hsier, J.P. Cmilloteau,P. and Toullec, R. (1984) Reprod. Nun'. D6velop.24, 835-843. 7 Monti, J.C. Mermoud, A.F. and J ~ P. (1989) Experientia45, 178-180. 8 Tanford, C., Bunville, LG. and Nw~l~i~y. (1959) L Am. Chem. Soc. 81, 4032-4036. 9 Pamalonl,D. (1965)Thesis,Paris43rsay, France. 10 Townend, R., Herskovils,T.T. and Tnnasheff, S.N. (1969) Arch. Biochem. Biophys.129, 567-580. 11 Mailliart, p. and Ribadeau Dumas, B. (1988) J. Food ScL 53, 743-746. 12 Dalgalarrondo,M., Chobert,J.M., Dufour,E., Bemand-Elarb,C., Dumont. J.P. and Haertl~ T. (1990) M i l c h ~ f t 45, 212216. 13 Lehninger,A.L (1977) in Binchimie(FlammarlonM~d~ine-Sciences Ed.), p. 192, Paris. 14 FosteL D.C., Sprecher,CA. Holly, ILD., Gambee,J.E, Walker, ILM. and Kumar,A.A.(1990) Biochemistry29, 347-354.