Vol. 49, No. 2APPLIED AND ENVIRONMENTAL MICROBIOLOGY, Feb. 1985, p. 328-3320099-2240/85/020328-05$02.00/0Copyright © 1985, American Society for Microbiology

Partial Isolation and Degradation of Caseins by Cell WallProteinase(s) of Streptococcus cremoris HP

FRED A. EXTERKATE* AND GERRIE J. C. M. DE VEERNetherlands Institute for Dairy Research (NIZO), 6710 BA Ede, The Netherlands

Received 9 July 1984/Accepted 12 November 1984

The cell wall proteinase fraction of Streptococcus cremoris HP has been isolated. This preparation did notexhibit any activity due to either specific peptidases known to be located near the outside surface of and in themembrane or intracellular proteolytic enzymes. By using thin-layer chromatography for the detection ofrelatively small hydrolysis products which remain soluble at pH 4.6, it was shown that 13-casein is preferentiallyattacked by the cell wall proteinase. This was also the case when whole casein or micelles were used as thesubstrate. K-casein hydrolysis is a relatively slow process, and os-casein degradation appeared to proceed at anextremely low rate. These results could be confirmed by using 14CH3-labeled caseins. A relatively fast andlinear initial progress of 14CH3-labeled 13-casein degradation is not inhibited by ae-casein and only slightly byK-casein at concentrations of these components which reflect their stoichiometry in the micelles. Possibleimplications of ,I-casein degradation for growth of the organism in milk are discussed.

Streptococcus cremoris possesses cell wall-associated pro-teinases which are assumed to attack milk proteins initiallyduring growth of the organism in milk (5). By their action,peptides are provided which can diffuse into the cell wall andare furthermore suitable to be used as a source of nitrogenand (essential) amino acids. The subsequent action of pep-tidases at the outside surface of and in the membrane issupposed to play an essential role in nitrogen nutrition (8).At present little is known about the extent to which caseinsare degraded by the cell wall proteinases and whether aspecific milk protein is the main source of those growth-stim-ulating peptides. Although all the caseins, as well as at-lactalbumin and 3-lactoglobulin, are degraded and used bylactic streptococci (14), the rate of hydrolysis and the extentof the degradation, and thus the relative contribution of theindividual proteins to the total protein pool used as thesource of nitrogen, may be significantly different. The acces-sibility of, on the one hand, the casein substrate and, on theother hand, the active site of the proteinase in situ (viz., atthe cell wall periphery or inside the cell wall) may alsodetermine what will be degraded preferentially. Other argu-ments made in favor of a possible role of whey proteins innitrogen nutrition of lactic streptococci are not sound andcan be easily disproved. First, observed but limited growthin whey (9, 10) can partly, if not fully, be explained by thepresence in whey of products of chymosin and starterenzyme action. A report concerning the failure of growth onsodium caseinate (11) may be attributed to Ca2"-deficiencyeither in direct relation to cell wall proteinase production inthe first place (7) or to another hitherto unknown Ca depen-dency (unpublished data). Furthermore, it was found thatwhey proteins are completely resistant to the proteolyticenzymes of rennet and starter during the manufacture oflow-fat semihard cheese (4). Finally, a growth rate compa-rable to that observed in milk could be measured in asynthetic medium supplemented with caseins (unpublisheddata). This paper reports on the problems mentioned aboveand describes some preliminary results with respect to cellwall proteinase action on caseins.

* Corresponding author.

MATERIALS AND METHODS

Growth and harvesting of the organism. Stationary-phasecells of S. cremoris HP were obtained from milk-growncultures as described previously (8).

Preparation of a crude cell wall proteinase fraction. Cellsfrom 12.5 liters of milk culture were washed twice with 0.05M sodium acetate-0.05 M sodium dihydrogen phosphatebuffer (pH 6.5). Care was taken to keep the temperatureduring these washings between 0 and 4°C to prevent therelease of proteinase activity (13). After these washings thecells were resuspended in the same buffer at 25°C, stirredgently for 30 min, and then centrifuged (10 min, 16,000 x g).Resuspension of the cells was repeated several times. Thesupernatants were pooled, adjusted to pH 5.4, and dialyzedovernight against an excess of distilled water. The powderobtained after freeze-drying was extracted with distilledwater, and the precipitate was discarded. The extract wasfreeze-dried again (crude cell wall proteinase fraction, 56 mgof protein) and dissolved in 10 ml of 0.05 M ammoniumacetate buffer (pH 5.4).

Sephacryl S300 column fractionation. Portions of 2.5 ml ofcrude cell wall proteinase were applied to a Sephacryl S300column (30 by 2.5 cm) equilibrated with 0.05 M ammoniumacetate (pH 5.4), and elution was performed with the samebuffer at a rate of 0.65 ml - min-'. Fractions (5 ml) werecollected. Those fractions indicated as proteolytic activewere pooled, dialyzed overnight, and freeze-dried, and thepowder was again extracted with 5 ml of water. The finalpreparation of partially purified cell wall proteinase in watercontained 0.5 mg of protein * ml-'.

Degradation of ['4CH3] casein. The purified cell wall pro-teinase fraction was used to study the breakdown of14CH3-labeled whole casein (specific activity, ca. 34,000cpm * mg-') or 14CH3-labeled 1-casein (specific activity,37,000 cpm * mg-') by following the release of trichloroace-tic acid (TCA) (6% [wt/vol])-soluble products, essentiallyaccording to the method described earlier (5).

Degradation of micelles, whole casein, and the casein com-ponents. Solutions of whole casein (0.2%, [wt/vol]) or cts-, Is-,and K-casein (0.1%, [wt/vol]) in 0.05 M ammonium acetate (5ml) or a suspension of micelles in the same buffer (5 ml)supplemented with 10 mM Ca2+ were incubated with 100 ,ul

328

on July 6, 2018 by guesthttp://aem

.asm.org/

Dow

nloaded from

S. CREMORIS CELL WALL PROTEINASE ACTION ON CASEINS 329

0 10 20 30 60 90 120 150

incubation time (min)

cpm. lo-2 l-'

30

20

10

0 10 20 30 60 90 120 150

incubation time (min)

FIG. 1. Release of proteinase activity from the cell wall of S.cremoris HP (a) and TR (b). Cold (0 to 4°C) collected cells were

suspended in 0.05 M sodium acetate-0.05 M sodium dihydrogenphosphate buffer (pH 6.5) at 25°C. After the indicated time intervals,a sample (0.5 ml) of the cell suspension was centrifuged (10 min;12,000 x g), and the supernatant (0) and cells (@) were resuspendedin cold (0 to 4°C) buffer at pH 5.4 (0.5 ml) and tested for proteinaseactivity at pH 5.4 and 30°C (5). After 10 and 60 min of incubation,collected cells were resuspended in buffer at pH 6.5 and 25°C, andthe proteinase activity in the supernatant (0, A) and that associatedwith the cells (U, A) was determined after 10 min of incubation.Proteinase activities are expressed as cpm * ml-I detectable after 20min of incubation with the substrate 14CH3-labeled whole casein.

(50 ,ug of protein) of the partially purified cell wall proteinasepreparation at pH 6.2 and 30°C. Samples (0.5 ml) were

withdrawn from the incubation mixture at the indicated timeintervals and adjusted to pH 4.6 with 10 pul of an acetic acidsolution to precipitate casein and the larger hydrolysisproducts. The supernatant was freeze-dried and dissolved in25 or 50 ,u of distilled water. Equal amounts of each of thesamples were applied to thin-layer plates (Kieselgel 40 F254;E. Merck AG, Darmstadt, Federal Republic of Germany)and developed with n-butanol-acetic acid-water (24:4:8[vol/vol]). The products were made visible with a ninhydrin

spray consisting of 0.4% (wt/vol) ninhydrin inn-butanol-acetic acid (96:4 [vol/vol]).

Isolation of micelles and caseins. Caseins were isolatedaccording to generally accepted methods (see reference 12).For the isolation of micelles, skim milk was centrifuged for90 min at 100,000 x g and 15°C. The micelles were sus-pended in 0.05 M ammonium acetate buffer (pH 6.2) supple-mented with 10 mM Ca2+ by means of a Potter-Elvehjemtube. After a second centrifugation, the micelles were re-suspended in the same buffer at a concentration equal to themicelle concentration in milk.

Gel electrophoresis and isoelectrofocusing. Sodium dodecylsulfate-8.5% polyacrylamide gel electrophoresis (imidazolebuffer, pH 7.0) and analytical isoelectrofocusing (pH 4 to 6)were performed essentially according to the instructions ofLKB-Produkter AB, Bromma, Sweden (application notes306 and 250, respectively) by using the LKB 2117 multiphor.

Protein quantification. Proteins were estimated accordingto the method of Bradford (2) by using crystalline bovineserum albumin (fraction V; BDH, Poole, England) as thestandard.

RESULTS

Partial purification of the cell wall proteinases. During theincubation of cells of S. cremoris HP at pH 6.5 and 25°C,proteinase activity (PI + PI,) (5, 6) appeared in the superna-tant mainly during the first 30 min (Fig. la). The reduction inactivity measured with the collected cells was much smallerthan the activity appearing in the supernatant. On the onehand this may be due to the variety of the peptide bonds thatare split in whole casein, which gives rise to nonlinear

A2800.5

0.4

0.3

0.2

0.1

0

cpm. 10-2ml-l

.1

11 i

,/I

20

10

0

10 20 30 40tube number

FIG. 2. Gel filtration of crude cell wall proteinase (+15 mg ofprotein) of Sephacryl S300. The A280 was recorded. For details see

the text. Proteinase activity (0) was assayed as described (5) andexpressed as cpm ml-' after 20 min of incubation with the sub-strate 14CH3-labeled whole casein. The arrow marks the voidvolume.

b

*_0

tI

,

VOL. 49, 1985

on July 6, 2018 by guesthttp://aem

.asm.org/

Dow

nloaded from

330 EXTERKATE AND DEVEERAPLENRO.McoI.

1 234 56 78 I 2 3

1 2 3 4 5 8 7 8 1 23

,'.. *

4 5 67 8

4 58 78

FIG. 3. Appearance of hydrolysis products soluble at pH 4.6 followed over a 4-h period of degradation of ot,- (a), P, (b), K- (c), and whole-(d) casein by the action of the cell wall proteinase system of S. cremoris HP. Samples were taken after 15 s (lane 1), 5 min (lane 2), 10 min(lane 3), 20 min (lane 4), 30 min (lane 5), 60 min (lane 6), 180 min (lane 7), and 240 min (lane 8) of incubation. For details see the text.

progress curves (see below), and on the other hand it may bedue to accessibility of the proteinase in situ. Continuedincubation resulted in a further decrease of activity detect-able with the cells, without any increase of proteinaseactivity in the supernatant being detected. No reduction inproteinase activity of these cells upon continued incubationwas detected when they were first resuspended in freshbuffer at pH 5.4. At this pH the permeability of the cell wallfor the proteinase is assumed to be reduced almost com-pletely (13). Since all activity determinations were alsoperformed at a pH of 5.4, the activity measured with cells isnot due to additional released enzyme.

Cells collected after 10 or 60 min and resuspended in freshbuffer at pH 6.5 and 250C showed a rapid additional releaseof activity. This procedure can be repeated with similarresults until the cell wall is exhausted. Such a cascade-likerelease of proteinase activity was observed with all strainsstudied so far, including those belonging to different typeswith respect to their cell wall proteolytic system (6). No lysisof cells, as measured by the release of glucose-6-phosphatedehydrogenase (8), could be detected during the whole washprocedure. Some strains showed only a poor release ofproteolytic activity. Cells of such strains showed no or onlyslight additional reduction of proteolytic activity after max-imum values in the supernatant had been reached (Fig. lb).A typical elution profile obtained with the crude cell wall

proteinase fraction loaded on a Sephacryl S300 column isshown in Fig. 2. Both proteinase activities PI and PI,,

detectable in this strain HP, were found in the same frac-tions. None of the peptidase activities known to be locatednear the outside surface or in the membrane, and whichcontaminated the crude cell wall fractions at detectablelevels, could be detected in the pooled column fractionscovered by the indicated bar. The final preparation showedsix bands after sodium dodecyl sulfate-polyacrylamide elec-trophoresis and analytical isoelectrofocusing. It was used toperform the experiments described in the following para-graphs. Under the conditions of these experiments no de-crease of proteinase activity could be detected. Furtherpurified preparations of the proteinase appeared to showremarkable increased instability.

Degradation of caseins. ots-, Is-, and K-casein, whole casein,and micelles were used as the substrates for cell wallproteinase activity. P-Casein was degraded relatively quicklywhen compared with as~- and K-casein (Fig. 3a through c and4b). In fact, the degradation of a.,-casein proceeded veryslowly under the present conditions. The degradation pat-tern obtained with 13-casein within the first 10 to 15 min (andalready detectable within a few seconds [Fig. 4a]) changedonly gradually over the following period of hours (Fig. 3b).With K-casein and a.,-casein a gradual increase of the inten-sity of the different bands was seen over the whole incuba-tion period. The degradation of whole casein (Fig. 3d and 4b)and micelles (data not shown) is mainly that of 13-casein,since initially the degradation products typical of P-caseindegradation are produced.

APPL. ENVIRON. MICROBIOL.

on July 6, 2018 by guesthttp://aem

.asm.org/

Dow

nloaded from

S. CREMORIS CELL WALL PROTEINASE ACTION ON CASEINS 331

FIG. 4. (a) Appearance of hydrolysis products soluble at pH 4.6 followed over a 15-min period of ,B-casein degradation by the action of --the cell wall proteinase system of S. cremoris HP. Samples were taken after 10 s (lane 1), 30 s (lane 2), 75 s (lane 3), 2 min (lane 4), 3 min(lane 5), 5 min (lane 6), 10 min (lane 7), and 15 min (lane 8) of incubation. (b) Hydrolysis products soluble at pH 4.6 of the different caseinsafter 30 (lanes 1 through 4) and 4 (lanes 5 through 8) h of cell wall proteinase action. Lanes 1 and 5, ao-casein; lanes 2 and 6, ,-casein; lanes3 and 7, whole casein; and lanes 4 and 8, K-casein. For details see the text.

Degradation of 14CH3-labeled caseins. A typical relation-ship between the release of TCA-soluble products from'4CH3-labeled ,-casein by the action of the purified cell wallproteinase system and time is shown in Fig. 5. An initiallylinear progress curve at a relatively high rate was followedby a lower rate of product formation. The initial phaseappeared to approximate, independent of the enzyme con-centration, a mnaximum value (viz., ca. 4,000 cpm * ml-')with a rate higher than that expected for the second phase.In the presence of both ao-casein (0.1% [wt/vol]) and K-casein (0.025% [wt/vol]), the initial-phase degradation of14CH3-labeled 3-casein was inhibited by only 17%. This wasdue to the presence of K-casein, since ao-casein alone had noinhibitory effect, whereas K-casein at the same concentration

(0.1% [wt/vol]) showed a significant inhibition of 14CH3-labeled ,-casein degradation of ca. 40%.

Degradation of 14CH3-labeled whole casein also showedan initially rapid, but nonlinear phase of product formationfollowed by a relatively slow phase (Fig. 6).

DISCUSSIONStudies with respect to proteinases of S. cremoris have

been limited until now. More data have been obtainedconcerning proteinases from S. Iactis (for a review seereference 10a).

cpm.10-2.m-1

cpm .10-2lnl

40 -

30

20 -

16-12

8

4

0 20 40 60 120 180 240 300incubation time (min)

FIG. 5. Appearance of TCA-soluble '4CH3-labeled hydrolysisproducts during incubation of "4CH3-labeled 0-casein (0.1% [wt/vol])at 30°C and pH 6.2 with the following concentrations of the cell wallproteinase of S. cremoris HP: 8.7 .g - ml-' (0), 2.7 .g - ml-' (0),and 1.35 pg * ml-1 (A) of incubation mixture. The buffer was 0.05 Msodium acetate-0.05 M sodium dihydrogen phosphate.

50-

40-

30-

2016-

12-

84

0 20 40 60 120 180 240 300incubation time (min)

FIG. 6. Appearance of TCA-soluble '4CH3-labeled hydrolysisproducts during incubation of '4CH3-labeled whole casein with theindicated concentrations of the cell wall proteinase of S. cremorisHP. Symbols are the same as those described in the legend to Fig.5. For details see Fig. 5 and the text.

VOL. 49, 1985

on July 6, 2018 by guesthttp://aem

.asm.org/

Dow

nloaded from

332 EXTERKATE AND DE VEER

Umemoto and Itoh (19) described some characteristics ofthe proteolytic activity of S. cremoris 11-61, which shows an

optimum activity at 60°C. With respect to its identity and itslocation in situ, the isolation procedure does not warrant theassumption of the authors that the proteolytic activity underconsideration is a "cell-surface proteinase," since isolationwas started with a cell lysate obtained by salt-induced lysisof lysozyme-treated cells (8). Likewise, the procedure fol-lowed by Ohkniya and Sato (15) for the isolation of intracel-lular proteinase from the same organism may have resultedin the release of cell wall proteinase activity (1, 13).Our results with respect to the release of proteinase from

cell walls of lactic streptococci, as originally described byMills and Thomas (13), may give some additional informa-tion. First, the enzymes apparently diffuse through the cellwall into the medium until an equilibrium has been reached.This explains the cascade-like release. Second, in the case of

cells showing a high rate of diffusion of the proteinase(s)through the cell wall (strain HP), a peripheral accumulationof proteinase may occur initially, followed by a redistribu-tion in the cell wall at the moment that the equilibrium hasbeen established. This explains the continuing decrease indetectable activity with such cells (and no decrease or a

smaller one with cells of strains like TR). Substrate diffu-sional limitations, which reduce the effective substrate con-

centration for an increasing amount of proteinase molecules,will be introduced. At pH 5.4, however, the initial situationwill be maintained due to reduced cell wall permeability (13).

3-Casein appears to be the most suitable substrate amongthe caseins for the cell wall proteinase(s) of S. cremoris HP.It is preferentially attacked, apparently also when micellesare offered, although in that case it may well be thatdegradation of extruded, soluble P-casein has been detected(see reference 17). The same order of preference withrespect to caseins is suggested by the results obtained with a

proteinase present in an extract of S. cremoris H-61 (15).However, the kinetic data presented do not give any sub-stantial information due to the possibility of significantdifferences in progress curves, especially the initial veloci-ties.The results obtained with 14CH3-labeled caseins reflect the

visualized initial degradation process remarkably well. Arapid initial degradation of ,B-casein is the most likely mainprocess involved in the occurrence of the initial rapid phaseof TCA-soluble product formation observed during the deg-radation of 14CH3-labeled whole casein. In the presence ofthe other caseins, first-phase degradation of ,-casein isinhibited, but the effect is small at the relative concentra-tions present in whole casein.

Products of ,-casein released during the initial phase may

be essential both with respect to initiation and continuationof growth of S. cremoris HP in milk. This should imply that,-casein alone or in combination with other caseins is a morestimulatory and efficient substrate for growth than any othercasein or combination without ,-casein. In milk the possi-bility of preferential ,-casein utilization for optimal growthwithout limitation due to substrate accessibility and withoutsevere competitive interference of other caseins exists.Growth may be mainly supported by degradation of ex-

truded soluble r-casein (3, 16), leaving micellar caseinunattacked. Since soluble ,B-casein is in equilibrium withmicellar P-casein and additional 1-casein is released from

the micelles so as to reestablish equilibrium upon removal ofsoluble 3-casein (3), it may well be that during growth of theorganism the cells are continuously provided with subse-quently delivered soluble ,-casein. Preliminary results fromgrowth experiments in synthetic media supplemented withdifferent caseins or with milk fractions strongly support anessential role for soluble (P-)casein (unpublished data).

LITERATURE CITED

1. Argyle, P. J., G. E. Mathison, and R. E. Chandan. 1976.Production of cell-bound proteinase by Lactobacillus bulgari-cus and its location in the bacterial cell. J. Appl. Bacteriol.41:175-184.

2. Bradford, M. M. 1976. A rapid and sensitive method for thequantitation of microgram quantities of protein utilizing theprinciple of protein dye binding. Anal. Biochem. 72:248-254.

3. Downey, W. K., and R. F. Murphy. 1970. The temperature-de-pendent dissociation of ,-casein from bovine casein micellesand complexes. J. Dairy Res. 37:361-372.

4. De Koning, P. J., R. de Boer, P. Both, and P. F. C. Nooy. 1981.Comparison of proteolysis in a low-fat semi-hard type of cheesemanufactured by standard and by ultrafiltration techniques.Neth. Milk Dairy J. 35:35-46.

5. Exterkate, F. A. 1975. An introductory study of the proteolyticsystem of Streptococcus cremoris strain HP. Neth. Milk DairyJ. 29:303-318.

6. Exterkate, F. A. 1976. Comparison of strains of Streptococcuscremoris for proteolytic activities associated with the cell wall.Neth. Milk Dairy J. 30:95-105.

7. Exterkate, F. A. 1979. Accumulation of proteinase in the cellwall of Streptococcus cremoris strain AM1 and regulation of itsproduction. Arch. Microbiol. 120:247-254.

8. Exterkate, F. A. 1984. Location of peptidases outside and insidethe membrane of Streptococcus cremoris. Appl. Environ. Mi-crobiol. 47:177-183.

9. Law, B. A., E. Sezgin, and M. E. Sharpe. 1976. Amino acidnutrition of some commercial cheese starters in relation to theirgrowth in peptone-supplemented whey media. J. Dairy Res.43:291-300.

10. Lloyd, G. T. 1971. New developments in starter technology.Dairy Sci. Abstr. 33:411-416.

10a.Law, B. A., and J. Kolstad. 1983. Proteolytic systems in lacticacid bacteria. Antonie van Leeuwenhoek J. Microbiol. 49:225-245.

11. McDonald, I. J. 1956. Utilization of sodium caseinate by lacticstreptococci and enterococci. Can. J. Microbiol. 2:607-610.

12. McKenzie, H. A. (ed.). 1971. Milk proteins, chemistry andmolecular biology, vol. II. Academic Press, Inc., New York.

13. Mills, 0. E., and T. D. Thomas. 1978. Release of cell wall-as-sociated proteinase(s) from lactic streptococci. N. Z. J. DairySci. Technol. 13:209-215.

14. Mills, 0. E., and T. D. Thomas. 1981. Nitrogen sources forgrowth of lactic streptococci. N. Z. J. Dairy Sci. Technol.16:43-55.

15. Ohmiya, K., and Y. Sato. 1975. Purification and properties ofintracellular proteinase from Streptococcus cremoris. Appl.Microbiol. 30:738-745.

16. Rose, D. 1969. Relation between micelles and serum casein inbovine milk. J. Dairy Sci. 51:1897-1902.

17. Schmidt, D. G., and T. A. J. Payens. 1976. Micellar aspects ofcasein. Surf. Colloid Sci. 9:165-229.

18. Thomas, T. D., and 0. E. Mills. 1981. Proteolytic enzymes ofstarter bacteria. Neth. Milk Dairy J. 35:255-273.

19. Umemoto, Y., and H. Itoh. 1981. Characterization of cell-sur-face proteinases of lactic acid bacteria. Neth. Milk Dairy J.35:333-337.

APPL. ENVIRON. MICROBIOL.

on July 6, 2018 by guesthttp://aem

.asm.org/

Dow

nloaded from