Eur. J . Biochem. 55, 369- 373 (1975)
Studies on the Specificity of Action of Bacteriophage T4 Lysozyme David MIRELMAN Department of Biophysics, The Weizmann Institute of Science, Rehovoth Gunnar KLEPPE and Harald B. JENSEN Department of Biochemistry, University of Bergen (Received January 14iMarch 19, 1975)
Lysozyme from bacteriophage T4 was found to digest a soluble, uncrosslinked peptidoglycan which is secreted by cells of Micrococcus luteus when incubated in the presence of penicillin G. Analysis of the enzymatic degradation products shows that T4 acts as an endo-acetylmuramidase capable of cleaving glycosidic bonds only at muramic acid residues that are substituted with peptide side-chains. The results indicate that the secreted peptidoglycan may consist of a mixture of chains, approximately half of which are substituted by peptide side chains on most of their muramic acid residues, while the other half is made up of chains in which the muramic acid moieties are unsubstituted.
Bacteriophage T4 produces a lysozyme which has been identified as an N-acetylmuramidase with a specificity similar to that of hen egg white lysozyme [l,21. A variety of well-defined, small-molecularweight substrates are known for hen and related avian egg white lysozymes, as well as for human lysozyme [3-61. Among these substrates are the bacterial cell wall tetrasaccharide GlcNAcb(1-+4)MurNAcp(l+4)GlcNAcB( 1-4)MurNAc and various b(1-+4)-linked oligosaccharides of N-acetylglucosamine obtained by partial degradation of chitin. Surprisingly, none of these well characterized substrates was found to be degraded by the T4 lysozyme. Consequently, the only available assay for this enzyme is the lysis of insoluble Escherichiu coli cells or cell walls [1,7]. A soluble chemically defined substrate for the enzyme is essential for obtaining better insight into the specificity and mode of action of T4 lysozyme and for comparison with other lysozymes. In this communication we describe the action of T4 lysozyme on a soluble, linear peptidoglycan isolated from the culture medium of Micrococcus luteus cells that were incubated in the presence of penicillin G
PI.
~~
Abbreviation. MurNAc, N-acetylmuramic acid. Enzyme. Lysozyme (EC 3.2.1.17).
Eur. J. Biochem. 55 (1975)
MATERIALS AND METHODS The soluble, uncrosslinked peptidoglycan was prepared from M . luteus cells as described [8], and labeled either with [U-'4C]alanine in the peptide sidechains ( ['4C]alanine-peptidoglycan, specific radioactivity of 16 counts min-' pmol-'), or in both amino sugar constituents ([14C]glycan-peptidoglycan, specific radioactivity of 12 counts min-' pmol-I). Purified bacteriophage T4 lysozyme was prepared from T4D-infected E. coli B cells as described [7]. The p( 1+4)-linked oligosaccharides of N-acetyl-D-glucosamine ( n = 2, 3, 4 and 6) [9] and the cell wall tetrasaccharide GlcNAcj( 1-+4)MurNAcp( 1+4)GlcNAcb (1-+4)MurNAc [3] were prepared as described. Hen egg white lysozyme (3 x crystallized) was from Worthington. Paper Chromatography and Electrophoresis Descending paper chromatograms were run on Whatman 3MM paper with solvent I, n-butanoll glacial acetic acid/water (4: 1 :5 , v/v/v, upper phase) for 40 h. High-voltage paper electrophoresis was carried out on Whatman 3MM paper in pH 6.5 buffer (pyridine 0.2 M, brought to pH 6.5 with dilute acetic acid) at 50 V/cm for 80 min.
3 70
Radioactlve compounds on chromatograms and electrophorograms were detected by exposure of the paper to X-ray film for 24-72 h. Compounds were eluted from paper as earlier described [lo]. Radioactivity was routinely determined by counting aqueous samples in an ethoxyethanol-containing fluid [ l l ] with a counting efficiency of 72%. Samples on paper were counted in a toluene-based fluid (efficiency 74 %).
Enzymic digestion of the labeled soluble peptidoglycan by T4 lysozyine was performed at 24 'C in 0.05 M Tris-NG1 buffer pH 7.0 containing 0.03 M KCI in a final volume of 100 1.11. Enzyme concentrations used were 10, 100 and 500 pgiml; the amount of labeled peptidoglycan in the reaction mixtures was 10.5 nmol ['"Clalanine-peptidoglycan (172000 counts/ min) or 5.2 nmol [14C]glycan-peptidoglycan (62000 countslmin). Aliquots (20 pl) were removed from the reaction mixtures after different incubation times, and : nalysed by paper chromatography (solvent I). Following chromatography and autoradiography, radioactive spots were cut out from chromatograms and counted. Digestion with hen egg white lysozyme was performed in 0.05 M ammonium acetate at 24 C. at enzyme concentrations of 150pg,'ml or 500 pglinl. Incubation was for 45 min or 24 h, respectively. Enzymic digestion of the b(1-+4)-linked N-acetylglucosamine oligosaccharides (n = 2, 3, 4 and 6) was performed both at 22 and 37 'C in 0.05 M phosphate buffer pH 7.2, at substrate concentrations of 15 mM, 50 mM, 250 mM and 0.4 mM, respectively. Enzyme concentrations used were 130 pg/ml for T4 lysozyme and 450 pgiml for hen egg white lysozyme. Aliquots were taken from the reaction mixtures after 2.5 and 24 h and analysed by paper chromatography (solvent I) ; the digestion products were revealed by the fluorescence dip as described [12]. Enzymic digestion of the cell wall tetrasaccharide (5 mgiml) was performed at 37 'C in both 0.05 M sodium acetate/acetic acid buffer, pH 5.2, and 0.05 M phosphate buffer, pH 7.2, using T4 lysozyine concentrations of 0.2 and 0.4 mg/ ml. Aliquots were taken after 2.5 and 6 h and the products analysed as above. Inliihition of Enz.vrnic Activity. Lytic activity of both T4 and hen egg white lysozymes was measured by the turbidimetric assay [7] using as substrate chloroform-treated 15.coli B cells and Micrococcus luteus cells in 0.05 M phosphate buffer pH 7.2. The rate of lysis was recorded as dA4503,/min in the absence and presence of inhibitor (final concentrations, 0.1 M N-acctylglucosamine, 5 mM of the di-, tri- and
Specificity of Action of Bacteriophage T4 Lysozyme
*
* IIisacch or id e
Tetrasacc haride
GP-2
GP-3
I t 1
Origin
Hew1
T4
Fig. 1. Paper cliromarop-ipliy uutorutiiogrtm o/ T4 untl hrtr egg white Iysozyme ( H e w / ) digests o J the linear peptido~qlylnn l u h c l d in its amino .sugars. Markers - disaccharide GlcNAcb( 1 +4)MurNAc and the tetrasaccharide is its corresponding dimer. GP-2 is the disaccharide- hexapeptide GlcNAc[~(1~4)MurNAc-~-Ala-i~-Cl~i-~I~ I.-Lys-D-Ala-D-Ala. Compounds GP-3, I and I1 are oligosaccharide-peptides that have not yet been identified. Chromatograms were run for 40 h in solvent I. See Table 1 for quantitative distribution of rcaction products ~~~
tetrasaccharide, 0.8 mM hexasaccharide and 5 mM of the cell wall tetrasaccharide).
Gel Filtration T4 lysozyme (100 pg/ml) digests of the aminosugar-labeledpeptidoglycan (20.8 nmol, 248 000counts/ min) were performed in a final volume of 0.5 ml for 20 h at 24 "C. The digest was applied to a Sephadex G-100 column (1.9 x 55 cm), elution was carried out with water and fractions of 2.5 ml were collected. Aliquots of 200 pl were taken and counted for radioactivity. RESULTS The main difference found between the digestion products obtained after incubation of the linear bur. I . Biochem. 5.5 (1975)
371
D. Mirelman, G. Kleppe, and H. B. Jensen
Fig. 2. Ejfect of' time on the fornzation ofproducts upon digestion oJ the peptidoglycan polymer by T4 Iysozyme. (A) Glycan-labeled polymer; (B) alanine-labeled polymer. For experimental details, see under Materials and Methods. The results shown are those obtained with 10 pg/ml T4 lysozyme. Products were separated by
paper chromatography and detected by autoradiography. The spots were subsequently counted in a liquid scintillation counter. The values are given as a percentage of total radioactivity applied to the Material which remained at the origin of chropaper. (A---.A) GP-2; (0- -0)I, 11; (O----O) GP-3 matograms; (0 -0)
peptidoglycan labeled in its amino sugar with T4 or hen egg white lysozyme was that no free di- or tetrasaccharide was obtained in the T4 lysozyme digests (Fig.1). Furthermore, it was found that the pattern of radioactive products obtained after T4 lysozyme digestions of the polymer labeled with ['4C]alanine or [14C]glycan were almost identical. In both cases the main product obtained was the disaccharidehexapeptide (GP-2)
enzyme concentration, only traces of free oligosaccharides could be detected (Table 1). In addition, no change in the relative amount of the digestion products could be observed when assays were performed at different ionic strengths (0.03 -0.2 M KCI). In digests of both types of labeled peptidoglycan the relative amount of the disaccharide-hexapeptide (GP-2) increased as a function of time of incubation and enzyme concentration, with a concomitant decrease in the amount of the unidentified oligosaccharide-peptide compounds (Fig. 2, Table 1). Gel filtration of T4 lysozyme digests of the aminosugar-labeled peptidoglycan on a Sephadex G-100 column separated between the low-molecular-weight glycopeptides and high-molecular-weight material (Fig.3). The material in the first peak (tubes 19-25), was isolated and shown to be chromatographically immobile (solvent I). It accounted for approximately 40 of the labeled peptidoglycan and was shown to be resistant to further degradation by T4 lysozyme. Incubation of this compound with hen egg white lysozyme, however, led to its almost complete digestion into low-molecular-weight products which were identified by paper electrophoresis (pH 6.5) and chromatography (solvent I) as the disaccharide (33 "/, of the labeled material), the tetrasaccharide (52 %), the disaccharide-hexapeptide (GP-2, 6 %), and chromatographically immobile material (7 %). The molecular weight of the material in the first peak (Fig. 3), estimated by gel filtration on a Sephadex G-100 column, was similar to that of the intact peptidoglycan ( M , approximately 40000) [8] and was eluted after the same volume (60 ml). Dextran blue
GIcNAcP(1+4)MurNAc L- Ala-D-Glu-Gly
L
L- Lys-D-Ala-D-Ala
and small amounts of three other unidentified oligosaccharide-peptides (GP-3 and compounds I, IT, Fig. 1). Quantitative analysis of the digestion products as a function of time of incubation (Fig.2) and enzyme concentration (Table l), shows that in the T4 lysozyme digests of both types of labeled peptidoglycan, the chromatographically immobile substrate was rapidly degraded and its digestion continued for approximately 4 h (Fig.2). A notable difference between the digestion of the two types of labeled substrate was in the relative proportion of the degradation products. Whereas in the case of the ['4C]alanine-labeledpeptid~ glycan, approximately 95 % of the labeled polymer is converted into small-molecular-weight products (76 as GP-2), exhaustive digestion of the amino-sugarlabeled peptidoglycan left approximately 35 as a chromatographically immobile compound. It was also found that even after prolonged incubation (24 h) of the amino-sugar-labeled peptidoglycan at high bur. J. Biochem. 55 (1975)
-
311-
Specificity of Action of Bacteriophage T 4 Lysoq i l l i
Table 1 . Di,qi,.\tioii 01 /he .rokuhlc, pepptidogl~cunby T4 und hm q , q white Iymzynies Distribution of radioactivity in products after 24-h incubation. For experimental details see under Methods. The iiuinbers refer to the percentagc of total radioactivity applied to the paper chromatogram. A bar means that no radioactivity was detected. HIIW, hen egg white. I . TI, GP-2n and GP-3 are compounds whose chemical structure is as yet unidentified (see also Fig. 1 ) Products
Radioactivity with substrate: ['~~Iglycan-peptidoglycan with lysozyme (pglml) from
Chroma togra phical I > iiniu ohi lc 1 + 11 GP-3 GP-2 CP-2a I'etrasaccharidc Disaccliaridi.
T4(10)
T4(100)
T4(500)
HEW(500)
45.2 21.5 5.7 37.6
37.1 9.6
31.2 11.4
12.4
6.8 31.5
5.7
7.1
43.5
46.0
-
-
4.1 -
-
4.3 -
-
-
30.1 19.2
['4C]a~d~i~ne-pept~dog~ycan with l!so,ryme (pg ml) from
T4(10)
21.8 13.2 7.0 64.9 3.3
T4(100)
8.8
7.5 7.5 70.6 5.6
T4(500)
H E W (500)
1 .-I
3.'
4.7
-
8.5 76.1 3.4
X.9
X5.1
3.4
-
-
-
-
-
-
-
-
tions with hen egg white lysozyme [13] the above oligosaccharides were incapable of inhibiting the activity of T4 lysozyme on I!?. coli or M . luteus cells [7].
DISCUSSION
1
4
-
'0
Although the bacteriolytic enzyme obtained from T4-phage-infected cells has been shown to have the same specificity as hen egg white lysozyme [1,2], it was surprising that none of the chemically defined low-molecular-weight oligosaccharides known to be suitable substrates for hen egg white lysozyme [3- 61 was cleaved by T4 lysozyme. A comparison between the digestion products obtained after incubation of T4 and hen egg white lysozymes with an aniino-sugarlabeled linear peptidoglycan (Fig. I), clearly indicated that a major difference in specificity existed between the two enzymes since no free oligosaccharides were detected in the T4 lysozyme digests. Previously it was concluded that the soluble uncrosslinked linear peptidoglycan [8] is a polymer with an average chain length of 50 disaccharide residues that is substituted with a hexapeptide side chain composed of L-Ala-D-Glu-Gly ~-Ly~-~-Ala-r>-Ala at approximately 50 of the muramic acid moieties. Digestion of this polymer with hen egg white lysozyme produces four major products, the disaccharide GlcNAcfl(l-t4)MurNAc (19.2 its corresponding p( 1+4)-linked dimer the tetrasaccharide (30.1 "~il), the disaccharide-hexapeptide, GP-2 ( 31.5 and smaII amounts of an unidentified oligosaccharide-peptide, GP-3 (6.873. In contrast to the results with hen egg I
( M , > 200000) which was eluted after 55 ml and the cell wall tetrasaccharide ( M , 974) which was eluted after 132 ml. were used as reference compounds in the calibration of the column. The &1-+4)-linked N acetylglucosaniine oligosaccharides ( n = 2 - 6) and the cell wall tetrasaccharide were not digested by the T4 lysozyme under the conditions tested (see under Methods). Furthermore, in contrast to earlier observa-
I:$)
Fur J Riothem 5.5 (1975)
D. Mirelman, G. Kleppe, and H. B. Jensen
white lysozyme, the main digestion product of the same polymer with T4 lysozyme was the disaccharidehexapeptide, GP-2 (46 %). Small amounts of unidentified oligosaccharide-peptides and a chromatographically immobile polysaccharide fraction (Table 1,Fig. l), were also obtained. This fraction was resistant to further digestion by T4 lysozyme, but was readily hydrolysed by hen egg white lysozyme. These results suggest that to serve as a substrate, the T4 lysozyme requires a glycan chain with a peptidesubstituted muramic acid residue at the site where cleavage occurs. Supporting evidence for this was obtained (a) from the fact that P(1+4)-linked Nacetylglucosamine oligosaccharides (n = 2- 6) and the cell wall tetrasaccharide were shown to be neither substrates nor inhibitors of T4 lysozyme, and (b) from the inability of T4 lysozyme to degrade a chemically modified E. coli peptidoglycan [14] from which approximately 90 % of the peptide side chains had been removed by pretreatment with hydrazine hydrate [15] (Kleppe and Jensen, unpublished results). Our present findings on the specificity requirements of T4 lysozyme may indicate that a difference exists between the structure of the active site in hen egg white lysozyme and T4 lysozyme, a fact that has been recently suggested by studies of the tertiary structure of T4 lysozyme [16]. Further studies and comparisons between the two lysozymes may help unravel the interrelation between structure and mode of action in enzymes of similar specificity. Another interesting observation deriving from this work is the information regarding the complex structure of the soluble uncrosslinked, linear peptidoglycan that cells of M . luteus secrete into the culture medium upon incubation in the presence of penicillin [S]. Analysis of T4 lysozyme digestion products indicates that the material may not be as homogenous as previously believed and that it may be a mixture of different types of peptidoglycan strands. It appears that the substitution by peptide side chains on the peptidoglycan strand is not random; in some of the chains virtually all vicinal muramic acid residues are substituted with peptide side chains (or at least to an appreciable extent) and these strands are almost entirely degraded by the T4 lysozyme. In others, most of the vicinal muramic acid moieties in a chain appear to be unsubstituted, and these chains remain intact after incubation with the T4 enzyme. In M . luteus cell walls, the absence of peptide side chains on half of the muramic acid residues of the peptidoglycan is due to the action of an endogenous N-acetylmura-
313
myl-L-alanine amidase [17]. The amount of unsubstituted muramic acid residues and the composition of the soluble peptidoglycan resembles that found in the native insoluble cell walls of M . luteus [IS]. Thus it can be postulated that in the native wall, also, some of the peptidoglycan chains are almost devoid of peptide side chains (and this may occur in selected areas of the wall) whereas in other chains, the opposite situation exists and in these, due to the presence of peptide side-chains and cross-linkages, the chains could be more dense and tightly packed. Investigation is currently in progress to study the effect that different growth and secretion conditions may have on the amount and distribution of unsubstituted muramic acid residues along the M . luteus peptidoglycan chains. These studies may hopefully lead to a better understanding of the role of the amidase in the process of peptidoglycan crosslinkage and growth in the cell wall. We wish to thank Mrs Y . Nuchamovitz for excellent technical assistance. In addition we wish to thank EMBO for the short-term fellowship to G.K. This research was supported in part by a Grant from the Volksivagen Stiftung.
REFERENCES 1. Tsugita, A , , Inouye, M., Terzaghi, E. & Streisinger, G. (1968) J . Biol. Chem. 243, 391 - 397. 2. Tsugita, A. & Inouye, M. (1968) J . Mol. Biol. 37, 201 -212. 3. Chipmdn, D. M. & Sharon, N. (1969) Science (Wash. D.C.) 165,454-465. 4. Sharon, N. & Eshdat, Y. (1974) in Lysozyme (Osserman, E. F., Canfield & Beychok, S., eds) pp. 195-218, Academic Press, New York. 5. Rupley, J. A. (1967) Proc. R. Soc. (Lond.) Ser. B. 167, 416428. 6. Sharon, N., Eshdat, Y., Maoz, I., Bernstein, Y., Prager, E. M. & Wilson, A. C. (1974) Israel J . Chem. 12, 591-603. 7. Jensen, H . B. & Kleppe, K. (1972) Eur. J . Biochem. 26, 305312. 8. Mirelman, D., Bracha, R. & Sharon, N. (1974) Biochemistry, 13,5045 - 5053. 9. Heneine, I. F. & Kimmel, J. R. (1972) J . Biol. Chem. 247, 6589-6596. 10. Eshdat, Y. & Mirelman, D. (1972) J . Chromatogr.65,458 -459. 11. Bray, A. G. (1960) Anal. Biochem. 1,279-285. 12. Sharon, N. & Seifter, S. (1964) J . B i d . Chem. 239, PC 23982399. 13. Sharon, N. (1967) Proc. R. Soc. (Lond.) Ser. B, 167, 402-415 14. Braun, V. & Sieglin, V. (1970) Eur. J . Biochem. 13,336- 346. 15. Ham, S. & Matsushima, Y . (1967) J . Biochem. (Tokyo) 62, 118- 125. 16. Mathews, B. M. & Remington, S. J. (1974) Proc. Nut1 Acad. Sci. U.S.A. 71, 4178-4182. 17. Ghuysen, J. M. (1968) Bacteriol. Rev. 32, 425-464. 18. Mirelman, D. & Sharon, N . (1967) J . Biol. Chem. 242, 34143427.
D. Mirelman, Department of Biophysics, The Weizmann Institute of Science, P.O. Box 26, Rehovoth, Israel G . Kleppe and H. B. Jensen, Biokjemisk Institutt, Universitetet i Bergen, Arstadveien 19, N-5000 Bergen, Norway Eur. J. Biochem. 55 (1975)
Studies on the Specificity of Action of Bacteriophage T4 Lysozyme David MIRELMAN Department of Biophysics, The Weizmann Institute of Science, Rehovoth Gunnar KLEPPE and Harald B. JENSEN Department of Biochemistry, University of Bergen (Received January 14iMarch 19, 1975)
Lysozyme from bacteriophage T4 was found to digest a soluble, uncrosslinked peptidoglycan which is secreted by cells of Micrococcus luteus when incubated in the presence of penicillin G. Analysis of the enzymatic degradation products shows that T4 acts as an endo-acetylmuramidase capable of cleaving glycosidic bonds only at muramic acid residues that are substituted with peptide side-chains. The results indicate that the secreted peptidoglycan may consist of a mixture of chains, approximately half of which are substituted by peptide side chains on most of their muramic acid residues, while the other half is made up of chains in which the muramic acid moieties are unsubstituted.
Bacteriophage T4 produces a lysozyme which has been identified as an N-acetylmuramidase with a specificity similar to that of hen egg white lysozyme [l,21. A variety of well-defined, small-molecularweight substrates are known for hen and related avian egg white lysozymes, as well as for human lysozyme [3-61. Among these substrates are the bacterial cell wall tetrasaccharide GlcNAcb(1-+4)MurNAcp(l+4)GlcNAcB( 1-4)MurNAc and various b(1-+4)-linked oligosaccharides of N-acetylglucosamine obtained by partial degradation of chitin. Surprisingly, none of these well characterized substrates was found to be degraded by the T4 lysozyme. Consequently, the only available assay for this enzyme is the lysis of insoluble Escherichiu coli cells or cell walls [1,7]. A soluble chemically defined substrate for the enzyme is essential for obtaining better insight into the specificity and mode of action of T4 lysozyme and for comparison with other lysozymes. In this communication we describe the action of T4 lysozyme on a soluble, linear peptidoglycan isolated from the culture medium of Micrococcus luteus cells that were incubated in the presence of penicillin G
PI.
~~
Abbreviation. MurNAc, N-acetylmuramic acid. Enzyme. Lysozyme (EC 3.2.1.17).
Eur. J. Biochem. 55 (1975)
MATERIALS AND METHODS The soluble, uncrosslinked peptidoglycan was prepared from M . luteus cells as described [8], and labeled either with [U-'4C]alanine in the peptide sidechains ( ['4C]alanine-peptidoglycan, specific radioactivity of 16 counts min-' pmol-'), or in both amino sugar constituents ([14C]glycan-peptidoglycan, specific radioactivity of 12 counts min-' pmol-I). Purified bacteriophage T4 lysozyme was prepared from T4D-infected E. coli B cells as described [7]. The p( 1+4)-linked oligosaccharides of N-acetyl-D-glucosamine ( n = 2, 3, 4 and 6) [9] and the cell wall tetrasaccharide GlcNAcj( 1-+4)MurNAcp( 1+4)GlcNAcb (1-+4)MurNAc [3] were prepared as described. Hen egg white lysozyme (3 x crystallized) was from Worthington. Paper Chromatography and Electrophoresis Descending paper chromatograms were run on Whatman 3MM paper with solvent I, n-butanoll glacial acetic acid/water (4: 1 :5 , v/v/v, upper phase) for 40 h. High-voltage paper electrophoresis was carried out on Whatman 3MM paper in pH 6.5 buffer (pyridine 0.2 M, brought to pH 6.5 with dilute acetic acid) at 50 V/cm for 80 min.
3 70
Radioactlve compounds on chromatograms and electrophorograms were detected by exposure of the paper to X-ray film for 24-72 h. Compounds were eluted from paper as earlier described [lo]. Radioactivity was routinely determined by counting aqueous samples in an ethoxyethanol-containing fluid [ l l ] with a counting efficiency of 72%. Samples on paper were counted in a toluene-based fluid (efficiency 74 %).
Enzymic digestion of the labeled soluble peptidoglycan by T4 lysozyine was performed at 24 'C in 0.05 M Tris-NG1 buffer pH 7.0 containing 0.03 M KCI in a final volume of 100 1.11. Enzyme concentrations used were 10, 100 and 500 pgiml; the amount of labeled peptidoglycan in the reaction mixtures was 10.5 nmol ['"Clalanine-peptidoglycan (172000 counts/ min) or 5.2 nmol [14C]glycan-peptidoglycan (62000 countslmin). Aliquots (20 pl) were removed from the reaction mixtures after different incubation times, and : nalysed by paper chromatography (solvent I). Following chromatography and autoradiography, radioactive spots were cut out from chromatograms and counted. Digestion with hen egg white lysozyme was performed in 0.05 M ammonium acetate at 24 C. at enzyme concentrations of 150pg,'ml or 500 pglinl. Incubation was for 45 min or 24 h, respectively. Enzymic digestion of the b(1-+4)-linked N-acetylglucosamine oligosaccharides (n = 2, 3, 4 and 6) was performed both at 22 and 37 'C in 0.05 M phosphate buffer pH 7.2, at substrate concentrations of 15 mM, 50 mM, 250 mM and 0.4 mM, respectively. Enzyme concentrations used were 130 pg/ml for T4 lysozyme and 450 pgiml for hen egg white lysozyme. Aliquots were taken from the reaction mixtures after 2.5 and 24 h and analysed by paper chromatography (solvent I) ; the digestion products were revealed by the fluorescence dip as described [12]. Enzymic digestion of the cell wall tetrasaccharide (5 mgiml) was performed at 37 'C in both 0.05 M sodium acetate/acetic acid buffer, pH 5.2, and 0.05 M phosphate buffer, pH 7.2, using T4 lysozyine concentrations of 0.2 and 0.4 mg/ ml. Aliquots were taken after 2.5 and 6 h and the products analysed as above. Inliihition of Enz.vrnic Activity. Lytic activity of both T4 and hen egg white lysozymes was measured by the turbidimetric assay [7] using as substrate chloroform-treated 15.coli B cells and Micrococcus luteus cells in 0.05 M phosphate buffer pH 7.2. The rate of lysis was recorded as dA4503,/min in the absence and presence of inhibitor (final concentrations, 0.1 M N-acctylglucosamine, 5 mM of the di-, tri- and
Specificity of Action of Bacteriophage T4 Lysozyme
*
* IIisacch or id e
Tetrasacc haride
GP-2
GP-3
I t 1
Origin
Hew1
T4
Fig. 1. Paper cliromarop-ipliy uutorutiiogrtm o/ T4 untl hrtr egg white Iysozyme ( H e w / ) digests o J the linear peptido~qlylnn l u h c l d in its amino .sugars. Markers - disaccharide GlcNAcb( 1 +4)MurNAc and the tetrasaccharide is its corresponding dimer. GP-2 is the disaccharide- hexapeptide GlcNAc[~(1~4)MurNAc-~-Ala-i~-Cl~i-~I~ I.-Lys-D-Ala-D-Ala. Compounds GP-3, I and I1 are oligosaccharide-peptides that have not yet been identified. Chromatograms were run for 40 h in solvent I. See Table 1 for quantitative distribution of rcaction products ~~~
tetrasaccharide, 0.8 mM hexasaccharide and 5 mM of the cell wall tetrasaccharide).
Gel Filtration T4 lysozyme (100 pg/ml) digests of the aminosugar-labeledpeptidoglycan (20.8 nmol, 248 000counts/ min) were performed in a final volume of 0.5 ml for 20 h at 24 "C. The digest was applied to a Sephadex G-100 column (1.9 x 55 cm), elution was carried out with water and fractions of 2.5 ml were collected. Aliquots of 200 pl were taken and counted for radioactivity. RESULTS The main difference found between the digestion products obtained after incubation of the linear bur. I . Biochem. 5.5 (1975)
371
D. Mirelman, G. Kleppe, and H. B. Jensen
Fig. 2. Ejfect of' time on the fornzation ofproducts upon digestion oJ the peptidoglycan polymer by T4 Iysozyme. (A) Glycan-labeled polymer; (B) alanine-labeled polymer. For experimental details, see under Materials and Methods. The results shown are those obtained with 10 pg/ml T4 lysozyme. Products were separated by
paper chromatography and detected by autoradiography. The spots were subsequently counted in a liquid scintillation counter. The values are given as a percentage of total radioactivity applied to the Material which remained at the origin of chropaper. (A---.A) GP-2; (0- -0)I, 11; (O----O) GP-3 matograms; (0 -0)
peptidoglycan labeled in its amino sugar with T4 or hen egg white lysozyme was that no free di- or tetrasaccharide was obtained in the T4 lysozyme digests (Fig.1). Furthermore, it was found that the pattern of radioactive products obtained after T4 lysozyme digestions of the polymer labeled with ['4C]alanine or [14C]glycan were almost identical. In both cases the main product obtained was the disaccharidehexapeptide (GP-2)
enzyme concentration, only traces of free oligosaccharides could be detected (Table 1). In addition, no change in the relative amount of the digestion products could be observed when assays were performed at different ionic strengths (0.03 -0.2 M KCI). In digests of both types of labeled peptidoglycan the relative amount of the disaccharide-hexapeptide (GP-2) increased as a function of time of incubation and enzyme concentration, with a concomitant decrease in the amount of the unidentified oligosaccharide-peptide compounds (Fig. 2, Table 1). Gel filtration of T4 lysozyme digests of the aminosugar-labeled peptidoglycan on a Sephadex G-100 column separated between the low-molecular-weight glycopeptides and high-molecular-weight material (Fig.3). The material in the first peak (tubes 19-25), was isolated and shown to be chromatographically immobile (solvent I). It accounted for approximately 40 of the labeled peptidoglycan and was shown to be resistant to further degradation by T4 lysozyme. Incubation of this compound with hen egg white lysozyme, however, led to its almost complete digestion into low-molecular-weight products which were identified by paper electrophoresis (pH 6.5) and chromatography (solvent I) as the disaccharide (33 "/, of the labeled material), the tetrasaccharide (52 %), the disaccharide-hexapeptide (GP-2, 6 %), and chromatographically immobile material (7 %). The molecular weight of the material in the first peak (Fig. 3), estimated by gel filtration on a Sephadex G-100 column, was similar to that of the intact peptidoglycan ( M , approximately 40000) [8] and was eluted after the same volume (60 ml). Dextran blue
GIcNAcP(1+4)MurNAc L- Ala-D-Glu-Gly
L
L- Lys-D-Ala-D-Ala
and small amounts of three other unidentified oligosaccharide-peptides (GP-3 and compounds I, IT, Fig. 1). Quantitative analysis of the digestion products as a function of time of incubation (Fig.2) and enzyme concentration (Table l), shows that in the T4 lysozyme digests of both types of labeled peptidoglycan, the chromatographically immobile substrate was rapidly degraded and its digestion continued for approximately 4 h (Fig.2). A notable difference between the digestion of the two types of labeled substrate was in the relative proportion of the degradation products. Whereas in the case of the ['4C]alanine-labeledpeptid~ glycan, approximately 95 % of the labeled polymer is converted into small-molecular-weight products (76 as GP-2), exhaustive digestion of the amino-sugarlabeled peptidoglycan left approximately 35 as a chromatographically immobile compound. It was also found that even after prolonged incubation (24 h) of the amino-sugar-labeled peptidoglycan at high bur. J. Biochem. 55 (1975)
-
311-
Specificity of Action of Bacteriophage T 4 Lysoq i l l i
Table 1 . Di,qi,.\tioii 01 /he .rokuhlc, pepptidogl~cunby T4 und hm q , q white Iymzynies Distribution of radioactivity in products after 24-h incubation. For experimental details see under Methods. The iiuinbers refer to the percentagc of total radioactivity applied to the paper chromatogram. A bar means that no radioactivity was detected. HIIW, hen egg white. I . TI, GP-2n and GP-3 are compounds whose chemical structure is as yet unidentified (see also Fig. 1 ) Products
Radioactivity with substrate: ['~~Iglycan-peptidoglycan with lysozyme (pglml) from
Chroma togra phical I > iiniu ohi lc 1 + 11 GP-3 GP-2 CP-2a I'etrasaccharidc Disaccliaridi.
T4(10)
T4(100)
T4(500)
HEW(500)
45.2 21.5 5.7 37.6
37.1 9.6
31.2 11.4
12.4
6.8 31.5
5.7
7.1
43.5
46.0
-
-
4.1 -
-
4.3 -
-
-
30.1 19.2
['4C]a~d~i~ne-pept~dog~ycan with l!so,ryme (pg ml) from
T4(10)
21.8 13.2 7.0 64.9 3.3
T4(100)
8.8
7.5 7.5 70.6 5.6
T4(500)
H E W (500)
1 .-I
3.'
4.7
-
8.5 76.1 3.4
X.9
X5.1
3.4
-
-
-
-
-
-
-
-
tions with hen egg white lysozyme [13] the above oligosaccharides were incapable of inhibiting the activity of T4 lysozyme on I!?. coli or M . luteus cells [7].
DISCUSSION
1
4
-
'0
Although the bacteriolytic enzyme obtained from T4-phage-infected cells has been shown to have the same specificity as hen egg white lysozyme [1,2], it was surprising that none of the chemically defined low-molecular-weight oligosaccharides known to be suitable substrates for hen egg white lysozyme [3- 61 was cleaved by T4 lysozyme. A comparison between the digestion products obtained after incubation of T4 and hen egg white lysozymes with an aniino-sugarlabeled linear peptidoglycan (Fig. I), clearly indicated that a major difference in specificity existed between the two enzymes since no free oligosaccharides were detected in the T4 lysozyme digests. Previously it was concluded that the soluble uncrosslinked linear peptidoglycan [8] is a polymer with an average chain length of 50 disaccharide residues that is substituted with a hexapeptide side chain composed of L-Ala-D-Glu-Gly ~-Ly~-~-Ala-r>-Ala at approximately 50 of the muramic acid moieties. Digestion of this polymer with hen egg white lysozyme produces four major products, the disaccharide GlcNAcfl(l-t4)MurNAc (19.2 its corresponding p( 1+4)-linked dimer the tetrasaccharide (30.1 "~il), the disaccharide-hexapeptide, GP-2 ( 31.5 and smaII amounts of an unidentified oligosaccharide-peptide, GP-3 (6.873. In contrast to the results with hen egg I
( M , > 200000) which was eluted after 55 ml and the cell wall tetrasaccharide ( M , 974) which was eluted after 132 ml. were used as reference compounds in the calibration of the column. The &1-+4)-linked N acetylglucosaniine oligosaccharides ( n = 2 - 6) and the cell wall tetrasaccharide were not digested by the T4 lysozyme under the conditions tested (see under Methods). Furthermore, in contrast to earlier observa-
I:$)
Fur J Riothem 5.5 (1975)
D. Mirelman, G. Kleppe, and H. B. Jensen
white lysozyme, the main digestion product of the same polymer with T4 lysozyme was the disaccharidehexapeptide, GP-2 (46 %). Small amounts of unidentified oligosaccharide-peptides and a chromatographically immobile polysaccharide fraction (Table 1,Fig. l), were also obtained. This fraction was resistant to further digestion by T4 lysozyme, but was readily hydrolysed by hen egg white lysozyme. These results suggest that to serve as a substrate, the T4 lysozyme requires a glycan chain with a peptidesubstituted muramic acid residue at the site where cleavage occurs. Supporting evidence for this was obtained (a) from the fact that P(1+4)-linked Nacetylglucosamine oligosaccharides (n = 2- 6) and the cell wall tetrasaccharide were shown to be neither substrates nor inhibitors of T4 lysozyme, and (b) from the inability of T4 lysozyme to degrade a chemically modified E. coli peptidoglycan [14] from which approximately 90 % of the peptide side chains had been removed by pretreatment with hydrazine hydrate [15] (Kleppe and Jensen, unpublished results). Our present findings on the specificity requirements of T4 lysozyme may indicate that a difference exists between the structure of the active site in hen egg white lysozyme and T4 lysozyme, a fact that has been recently suggested by studies of the tertiary structure of T4 lysozyme [16]. Further studies and comparisons between the two lysozymes may help unravel the interrelation between structure and mode of action in enzymes of similar specificity. Another interesting observation deriving from this work is the information regarding the complex structure of the soluble uncrosslinked, linear peptidoglycan that cells of M . luteus secrete into the culture medium upon incubation in the presence of penicillin [S]. Analysis of T4 lysozyme digestion products indicates that the material may not be as homogenous as previously believed and that it may be a mixture of different types of peptidoglycan strands. It appears that the substitution by peptide side chains on the peptidoglycan strand is not random; in some of the chains virtually all vicinal muramic acid residues are substituted with peptide side chains (or at least to an appreciable extent) and these strands are almost entirely degraded by the T4 lysozyme. In others, most of the vicinal muramic acid moieties in a chain appear to be unsubstituted, and these chains remain intact after incubation with the T4 enzyme. In M . luteus cell walls, the absence of peptide side chains on half of the muramic acid residues of the peptidoglycan is due to the action of an endogenous N-acetylmura-
313
myl-L-alanine amidase [17]. The amount of unsubstituted muramic acid residues and the composition of the soluble peptidoglycan resembles that found in the native insoluble cell walls of M . luteus [IS]. Thus it can be postulated that in the native wall, also, some of the peptidoglycan chains are almost devoid of peptide side chains (and this may occur in selected areas of the wall) whereas in other chains, the opposite situation exists and in these, due to the presence of peptide side-chains and cross-linkages, the chains could be more dense and tightly packed. Investigation is currently in progress to study the effect that different growth and secretion conditions may have on the amount and distribution of unsubstituted muramic acid residues along the M . luteus peptidoglycan chains. These studies may hopefully lead to a better understanding of the role of the amidase in the process of peptidoglycan crosslinkage and growth in the cell wall. We wish to thank Mrs Y . Nuchamovitz for excellent technical assistance. In addition we wish to thank EMBO for the short-term fellowship to G.K. This research was supported in part by a Grant from the Volksivagen Stiftung.
REFERENCES 1. Tsugita, A , , Inouye, M., Terzaghi, E. & Streisinger, G. (1968) J . Biol. Chem. 243, 391 - 397. 2. Tsugita, A. & Inouye, M. (1968) J . Mol. Biol. 37, 201 -212. 3. Chipmdn, D. M. & Sharon, N. (1969) Science (Wash. D.C.) 165,454-465. 4. Sharon, N. & Eshdat, Y. (1974) in Lysozyme (Osserman, E. F., Canfield & Beychok, S., eds) pp. 195-218, Academic Press, New York. 5. Rupley, J. A. (1967) Proc. R. Soc. (Lond.) Ser. B. 167, 416428. 6. Sharon, N., Eshdat, Y., Maoz, I., Bernstein, Y., Prager, E. M. & Wilson, A. C. (1974) Israel J . Chem. 12, 591-603. 7. Jensen, H . B. & Kleppe, K. (1972) Eur. J . Biochem. 26, 305312. 8. Mirelman, D., Bracha, R. & Sharon, N. (1974) Biochemistry, 13,5045 - 5053. 9. Heneine, I. F. & Kimmel, J. R. (1972) J . Biol. Chem. 247, 6589-6596. 10. Eshdat, Y. & Mirelman, D. (1972) J . Chromatogr.65,458 -459. 11. Bray, A. G. (1960) Anal. Biochem. 1,279-285. 12. Sharon, N. & Seifter, S. (1964) J . B i d . Chem. 239, PC 23982399. 13. Sharon, N. (1967) Proc. R. Soc. (Lond.) Ser. B, 167, 402-415 14. Braun, V. & Sieglin, V. (1970) Eur. J . Biochem. 13,336- 346. 15. Ham, S. & Matsushima, Y . (1967) J . Biochem. (Tokyo) 62, 118- 125. 16. Mathews, B. M. & Remington, S. J. (1974) Proc. Nut1 Acad. Sci. U.S.A. 71, 4178-4182. 17. Ghuysen, J. M. (1968) Bacteriol. Rev. 32, 425-464. 18. Mirelman, D. & Sharon, N . (1967) J . Biol. Chem. 242, 34143427.
D. Mirelman, Department of Biophysics, The Weizmann Institute of Science, P.O. Box 26, Rehovoth, Israel G . Kleppe and H. B. Jensen, Biokjemisk Institutt, Universitetet i Bergen, Arstadveien 19, N-5000 Bergen, Norway Eur. J. Biochem. 55 (1975)
