THE JOURNAL OF BIOLOGICAL CHEMISTRY Vol. 264, No. 5, Issue of February 15, pp. 2957-2962,1989 Printed in U.S.A.

Serine and Tyrosine Protein Kinase Activities in Streptococcus pyogenes PHOSPHORYLATION OF NATIVE AND SYNTHETIC PEPTIDES OF STREPTOCOCCAL M PROTEINS*

(Received for publication, April 13, 1988)

Thomas M. ChiangSQllII , Jonathan Reizer**, and Edwin H. BeacheyS§SI From the $.Veterans Administration Medical Center, the Departments of §Medicine, TBiochemistry, and $.$Microbiology, University of Tennessee, Memuhis, Tennessee 38104, and **Laboratory of Biochemical Genetics, National Heart, Lung, and Blood Institute, Bethesda, Maryland 20892

Two forms of protein kinase activity were isolated from crude extracts of Streptococcus pyogenes and partially purified by ion exchange chromatography and affinity chromatography. The phosphorylation ac- tivities were shown to be insensitive to CAMP, required the presence of divalent cations, and eluted from a Sephadex G-200 column with approximate molecular masses of 60 and 45 kDa, respectively. Both enzymes were capable of phosphorylating eukaryotic proteins and synthetic polypeptides in addition to endogenous and heterologous prokaryotic proteins at serine and tyrosine residues. Firm evidence for tyrosine kinase activity was obtained by the use of a tyrosine kinase- specific substrate, a 4: 1 g1utamate:tyrosine copolymer. Both protein kinases phosphorylated HPr, a phospho- carrier protein of the phosphotransferase system iso- lated from S. pyogenes and Bacillus stearothermo- philus, but failed to phosphorylate HPr isolated from Escherichia coli, Both also phosphorylated a native polypeptide fragment (pep M24) as well as synthetic peptide copies of M protein, the major virulence deter- minant of group A streptococci. These results indicate that prokaryotic protein kinases are capable of phos- phorylating eukaryotic proteins and suggest that the protein kinases of streptococci may play an important role not only in the phosphotransferase system but also in the virulence properties of these organisms.

The post-translational modification of proteins by reversi- ble phosphorylation has long been recognized as a major mechanism for regulation of cellular activities in eukaryotes (Krebs, 1985; Hunter and Cooper, 1985; Edelman et al., 1987). In contrast, the first conclusive demonstration of protein kinase activity in prokaryotes, excluding phage-infected bac- teria, was obtained only a decade ago (Wang and Koshland, 1978). Since then, a large body of information has emerged, demonstrating intrinsic protein kinases and phosphorylated proteins in diverse prokaryotes including coliform bacteria (Wang and Koshland, 1981; Enami and Ishihama, 1984; Des- marquets et al., 1984), various lactic acid bacteria (Deutscher and Saier, 1983; Reizer et al., 1983, 1984; Deutscher and

* This study was supported by research funds from the Veterans Administration, United States Public Health Service Grants AI- 10085 and AI-13550, and American Heart Association Grant-in-Aid 84-799. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

)I To whom correspondence should be addressed Research Service (151), VA Medical Center, 1030 Jefferson Ave., Memphis, TN 38104.

Engelmann, 1984; Mimura et al., 1987; Reizer et al., 1988), photosynthetic purple bacteria and Cyanobacteria (Holuigue et al., 1985; Allen et al., 1985; Turner and Mann, 1986, Clostridia (Antranikian et al., 1985; Londesborough, 1986), the halophil Halobacterium hulobium (Spudich and Stoeck- enius, 1980), the Archaebacterium Sulfolobus acidocaldarius (Skorko, 1984), Staphylococcus aureus, and Bacillus subtilis (Reizer and and Peterkofsky, 1987). However, many of the bacterial phosphorylated proteins have not yet been identi- fied, and only in a few cases have the physiological roles of such covalent modifications been elucidated. Interestingly, the most extensive studies in this respect bear upon the regulatory role of protein phosphorylation in carbon and energy metabolism. 1) Reversible phosphorylation of isocit- rate dehydrogenase in Escherichia coli and Salmonella typhi- murium modulates carbon flow via the tricarboxylic acid cycle and the glyoxylate shunt (Wang and Koshland, 1982; Nimmo, 1984). 2) Regulation of citrate metabolism in Clostridium sphenoides appears to be controlled by reversible phosphoryl- ation of citrate lyase ligase (Antranikian et al., 1985). 3) The transport of sugars in Gram-positive bacteria is modulated by ATP-dependent phosphorylation of HPr, a general energy- coupling protein of the phosphoeno1pyruvate:glycose phos- photransferase system (Reizer and Peterkofsky, 1987).

The present study stemmed from an earlier discovery of a protein kinase activity in whole cells and in extracts of Strep- tococcus pyogenes (Reizer et al., 1983; Deutscher and Saier, 1983). Although several phosphorylated protein species were recognized in crude extracts of this streptococcus (Reizer et al., 1985), only one phosphoprotein, HPr(SerP), was identi- fied. These observations and our longtime interest in the virulence factors of S. pyogenes prompted us to investigate the potential role of protein phosphorylation in the virulence properties of this important human pathogen. This commu- nication provides evidence for the presence of two distinct protein kinases with broad substrate specificity in S. pyogenes. Both kinases are capable of phosphorylating eukaryotic as well as prokaryotic proteins and synthetic polypeptides at serine and tyrosine residues. In addition, both enzymes phos- phorylate HPrs isolated from Gram-positive bacteria as well as the M protein derived from group A streptococci. Because the latter protein is the major virulence factor of these bac- teria, these findings open new avenues into the investigation of the mechanisms of the pathogenesis of group A streptococ- cal infections.

MATERIALS AND METHODS

Bacterial Growth and Preparation of Crude Extracts-S. pyogenes (Vaughn, type 24) cells were grown in 60-liter batches in Todd-Hewitt broth for 16 h (Beachey et al., 1977). Cells were sedimented by

2957

2958 Streptococcal Protein Kinases centrifugation, washed three times in 20 mM phosphate, 0.15 M NaCl buffer (pH 7.6), and the cell pellet (90 g, wet weight) was resuspended in 450 ml of 50 mM Tris-HC1 buffer (pH 7.6). Cells were ruptured by four cycles of sonication, 2.5 min each, a t 0 "C, and cell debris was removed by centrifugation at 35,000 X g for 30 min.

Protein Purification-The streptococcal protein kinases were pu- rified by ammonium sulfate precipitation and column chromatogra- phy as follows. The crude sonic extract of S. pyogenes was subjected to ammonium sulfate precipitation (30-70% saturation). After cen- trifugation (10,000 X g for 30 min) the pellet, which contained the protein kinase activities, was dissolved in 50 mM Tris-HC1 buffer (pH 7.4, Tris buffer), dialyzed for 24 h at 4 'C against the same buffer, and then applied onto a DEAE-cellulose column (Whatman DE52, 2.5 X 15 cm) equilibrated with Tris buffer (1,000 ml). The column was washed with 250 ml of the same buffer, and proteins were eluted with an 800-ml linear gradient of 0-0.6 M NaCl. Protein kinase a was eluted at about 150 mM NaC1, whereas protein kinase b was recovered at about 240 mM NaC1. Active fractions of each kinase were pooled, concentrated by lyophilization, and after dialysis against Tris buffer were applied to an affinity chromatography column of immobilized poly(G1u:Tyr)-Sepharose 2B (20 ml of packed gel). The column was washed with 60 ml of 20 mM potassium phosphate buffer (pH 7.4) containing 1 mM P-mercaptoethanol, and proteins attached to the matrix were eluted with 60-ml volumes of 0.15 and 1.5 M NaCl in the same buffer. Fractions containing protein kinase activity were pooled, concentrated by lyophilization, dialyzed against Tris buffer, and stored at -80 "C until used. The molecular weights of the partially purified protein kinases were estimated by gel filtration using a calibrated Sephadex G-200 column (Pharmacia LKB Biotechnology Inc.) (2 X 95 cm) equilibrated with the same buffer.

A polypeptide fragment of M protein (pep M24) was purified from limited pepsin digests of whole cells of S. pyogenes as described (Beachey et al., 1977). HPrs from S. pyogenes, Bacillus stearothemo- philus, and E. coli were purified by the procedure of Beyreuther et al. (1977) with the described modifications (Reizer et al., 1984). Plasma protein kinase was purified as previously described (Chiang and Kang, 1985).

Assay of Protein Kinase Activity-Protein kinase activity was de- termined in a standard assay mixture (100-pl final volume) that contained 50 mM Tris-HC1 buffer (pH 7.4), 2 mM MgC12, 2 mM NaF, 10 p~ [y-32P]ATP (0.5 pCi/test), 25-200 pg of the appropriate sub- strate, and 10-50 pg of the kinase preparation. The assay mixture was incubated at 30 "C for 5-10 min before termination of the reaction by the addition of 1 ml of 10% trichloroacetic acid containing 10 mM sodium pyrophosphate (trichloroacetic acid/phosphate) followed by 200 pg of bovine serum albumin. The mixture was held at 4 "C for 30 min, and the precipitate was collected by centrifugation at 2500 X g for 20 min. The pellet was washed twice with 2-ml volumes of cold trichloroacetic acid/phosphate, suspended in Aquasol (10 ml), and the amount of 32P incorporated was determined with a Packard Tri- Carb liquid scintillation counter (model 460C). For assay of tyrosine kinase activity, the standard assay mixture was modified as follows. The concentration of [y3'P]ATP was increased to 50 FM (1.0 pCi/ test), NaF was replaced by vanadate (10 p ~ ) , and 200 pg of the appropriate substrate was employed. Assay of HPr(Ser) phosphoryl- ation was performed in the standard assay mixture that contained fructose-1,6-diphosphate (10 mM). The latter assay and the phospho- rylation reaction of pep M24 were terminated by the addition of an equal volume of SDS' "quench buffer." The samples were then boiled for 2 min, and the proteins were separated by SDS-PAGE (Laemmli, 1970). The gels were stained for proteins with Coomassie Blue (0.025%) in 10% methanol, 7.5% acetic acid, destained with the same solvent, and dried under vacuum. Radiolabeled proteins were detected by autoradiography as previously described (Chiang and Kang, 1985).

Phosphoamino Acid Analysis-The phosphorylated amino acids were identified using the method described by Rubin and Earp (1983). Briefly, samples of radiolabeled proteins or poly(G1u:Tyr) were washed with a mixture of ether/ethanol (l:l), hydrolyzed for 2 h in 6 N HC1 at 110 "C, and lyophilized. The hydrolysates were dissolved in water (50 pl) and applied to thin layer cellulose plates (EM Science, Cherry Hill, NJ) together with authentic phosphotyrosine, phospho- serine, and phosphothreonine. Phosphoamino acids were separated by electrophoresis (for 2 h a t 500 V) in 7.8% acetic acid, 2.5% formic acid buffer (pH 1.9) followed by ascending chromatography in the second dimension with isobutyric acid, 0.5 N NHlOH (5:3, v/v).

The abbreviations used are: SDS, sodium dodecyl sulfate; PAGE, polyacrylamide gel electrophoresis.

Radiolabeled amino acids were detected by autoradiography, and standard phosphoamino acids were visualized by staining with Nin- hydrin. The individual phosphoamino acids were quantified by scrap- ing the cellulose from the appropriate areas of the plate; the cellulose was decolorized with 1 ml of 6% H202, suspended in 10 ml of Aquasol, and then counted in a liquid scintillation counter as described above.

Peptide Synthesis-Peptides with the same amino acid sequence as the amino termini of the M protein of S. pyogenes type 24 (SM24) and type 5 (SM5) were synthesized with a peptide synthesizer ac- cording to the principles of Merrifield (1963) and as described (Beachey et al., 1983).

Protein Determination-Proteins were assayed by the method of Lowry et at. (1951) and by quantitative amino acid analysis (Kang et al., 1969).

Preparation of Poly(G1u:Tyr)-Sepharose 2B Column-Washed packed Sepharose 2B (25 ml) was incubated with 500 mg of poly(G1u:Tyr) in 3 mM phosphate buffer, pH 6.3. After 30 min of stirring, 100 mg of 1-ethyl-3-(3-dimethylaminopropyl)cardodiimide was added and maintained at pH 6.3 for another 60 min (Timkovich, 1977). After stirring the mixture overnight at 4 "C, the gel was washed with 150 ml of 20 mM phosphate, 0.15 M NaCl, pH 7.4, and poured onto a column. The column was kept and used in the cold room at 4 "C.

RESULTS

Purification of Streptococcal Protein Kinases a and b-Two distinct forms of ATP-dependent protein kinase activity were identified in extracts of S. pyogenes and resolved by ion exchange chromatography and gel filtration. The first protein kinase, a, was eluted from the DEAE-cellulose column at 150 mM NaCl, whereas protein kinase b was eluted at 240 mM NaCl (Fig. lA). Size exclusion chromatography on a calibrated column of Sephadex G-200 showed that protein kinase a had a molecular mass of approximately 60 kDa, whereas kinase b had a molecular mass of approximately 45 kDa (Fig. 1B).

Further purification of the enzymes by affinity chromatog- raphy on a poly(G1u:Tyr)-Sepharose 2B column (Fig. 2) re- sulted in a 72- and 41-fold purification of kinases a and b, respectively, using histone as substrate (Table I). SDS-PAGE of the purified enzymes confirmed that the molecular masses of kinases a and b were 60 and 45 kDa, respectively; never- theless, a number of additional contaminating bands were detected in the purified enzyme preparations (see inset of Fig. 2).

Exogenous Substrate Phosphorylation and Phosphoamino Acid Analysis-Both protein kinases a and b were capable of phosphorylating eukaryotic proteins, i.e. histone, protamine, and casein, as well as synthetic polypeptides such as poly(G1u:Tyr). In view of the current controversy regarding the ability of bacterial protein kinases to catalyze the phos- phorylation of eukaryotic proteins (Cozzone, 1984), we have rigorously characterized the phosphorylated products of the kinase-catalyzed reaction.

The phosphorylated products of the acid hydrolysates of casein and histone were resistant to RNase, DNase, and hot trichloroacetic acid treatment (15 min at 90 "C) but were completely digested by Pronase (Table 11). Furthermore, analysis of the phosphorylated product of casein by SDS- PAGE and autoradiography has shown prominent labeling of bands that co-migrated with the major protein band of casein (data not shown). These findings provide compelling evidence for the conclusion that the streptococcal kinases phosphoryl- ate the eukaryotic protein substrates rather than catalyze the transfer of y-phosphate of ATP to nonprotein acceptors such as polyphosphates (Li and Brown, 1973). This conclusion was further substantiated by phosphoamino acid analyses of the labeled proteins.

The stability of the 32P-labeled proteins to hot trichloroa- cetic acid treatment (10% at 90 "C for 15 min) suggested that they contained ester-linked phosphate, i.e. phosphoserine,

Streptococcal Protein Kinases 2959

A

0.6

I

0.4 o a

i 5 I- z w 0 z

0.2 s 2 v)

FRACTION NUMBER

B

I LI U &I D.1 1.1

R l

FIG. 1. Separation and purification of protein kinases a and b by ion exchange chromatography. The dialyzed solution of the (NH4)zSO4-precipitated protein kinase was loaded onto an equili- brated DEAE-cellulose column (2.5 X 15 cm), and unbound material was eluted with 240 ml of 50 mM Tris-HC1 buffer (pH 7.4). Elution of the protein kinase activity was accomplished with a linear gradient (800 ml) of 0-0.6 M NaC1. Fractions (12 ml) were collected, and an aliquot of 50 pl of each fraction was assayed for protein kinase activity using histone type IIA as the protein substrate (A) . The molecular weights of both kinases were estimated with a calibrated Sephadex G-200 column as shown in B.

phosphothreonine, or phosphotyrosine, since amidophos- phates or acyl-linked phosphates are unstable under these conditions. The individual phosphorylated residues were iden- tified and quantified by two-dimensional phosphoamino acid analysis of the radiolabeled proteins that were subjected to acid hydrolysis (6 N HC1 at 110 "C for 120 min). As shown in Table 111, kinases a and b both phosphorylated histone mainly a t serine residue(s) (-90%) and to a lesser extent at tyrosine (-9%). The phosphorylation of threonine residues was only 2% and was of questionable significance. By contrast, the amino acid labeled in poly(G1u:Tyr) was exclusively phospho- tyrosine (Table I11 and Fig. 3).

The kinetic properties of the streptococcal protein kinases provided additional distinctions between protein kinases a and b. Both kinases followed Michaelis-Menten (hyperbolic) kinetics with respect to the protein substrates and ATP; however, kinase a was maximally active with histone, whereas kinase b was maximally active with protamine (Fig. 4). The apparent K,,, values for ATP were 30 and 10 PM when the 60- and 45-kDa enzymes, respectively, were assayed with saturat- ing amounts of their preferred protein substrates.

A

* b C

1 r

$ 301- ,* U

* 20 E

L \

Y " - . : 10 .- ,, K p /

I ' I I 1 - L >L , 0 5 10 15 2 0 2 5 30 35 40 4 5

FRACTION NUMBER

FIG. 2. Purification of protein kinases a and b by affinity chromatography. Five ml of protein kinase a (A) and protein kinase b ( B ) were loaded onto a poly(G1u:Tyr)-Sepharose 2B column (20 ml of packed gel). Proteins were eluted with 60 volumes of (a ) 20 mM potassium phosphate buffer (pH 7.4) containing (b) 0.15 M NaCl and (c) 1.5 M NaCl. Fractions were assayed for protein kinase activity using histone (-) and poly(G1u:Tyr) (---) as substrates. The bound fractions were analyzed with 10% SDS-PAGE and stained with Coomassie Brilliant Blue as shown in the inset. Lane 1, protein kinase a; lane 2, protein kinase b; and lane 3, molecular mass markers: myosin (200,000), @-galactosidase (130,000), phosphorylase b (94,000), bovine serum albumin (68,000), and ovalbumin (45,000).

Effect of Divalent Cations and CAMP on Protein Kinase Activities-Both streptococcal protein kinases a and b failed to phosphorylate histone or poly(G1u:Tyr) without the addi- tion of divalent cations or in the presence of EDTA (2 mM). For example, phosphorylation reactions of histone by kinase a were stimulated (5.0 and 4.4 nmol/min/mg) by either M e or Mn2+ (2 mM), respectively, whereas addition of EDTA completely abolished the stimulatory effect of the divalent cations. Similar results were obtained with protein kinase b. Other divalent cations such as Zn2+, Ba2+, and Ca2+ had little or no effect on either kinase a or b. cAMP (1 PM) had little or no effect on the phosphorylation of histone, protamine, casein, or HPr stimulated by the purified kinases of S. py- ogenes. In contrast to the innocuous effects of cAMP on the kinases of S. pyogenes, two types of protein kinases regulated in a reciprocal mode by cAMP in oral streptococci have been described previously (Khandelwal et al., 1973).

Phosphorylation of HPr by the Streptococcal Protein Kinases-Previous studies have shown that the endogenous product of the 60-kDa kinase-catalyzed reaction is HPr(SerP) (Reizer et al., 1984). Both kinases a and b similarly phospho- rylated purified HPrs of S. pyogenes and B. stearothermo- philus but not of E. coli.

Phosphorylation of S. pyogenes M Protein-Because of our longtime interest in the role of the M protein emanating from the surface of S. pyogenes cells in the virulence of these organisms, we investigated the capacity of the protein kinases of streptococci to phosphorylate this virulence determinant. The activity of streptococcal protein kinase a in phosphoryl- ating M protein was compared with that of a CAMP-depend- ent protein kinase of heart muscle and a plasma protein kinase

2960 Streptococcal Protein Kinases TABLE I

Purification of protein kinases from S. pyogenes The protein kinase activity was assayed as described under “Materials and Methods.”

Fraction Total Specific activity Total activity Recovery protein Histone Poly(Glu:Tyr) Histone Poly(G1u:Tyr) Histone Poly(G1u:Tyr)

mg nmollminlmg nmollmin % Crude extract 609 0.09 0.25 54.8 155.4 100 100 30-70% (NH4)2S04 fraction 134.4 0.36 0.77 48.3 103.5 88.1 66.6 DE52 column

Peak a 3.9 2.8 0.9 10.9 3.51 19.9 2.3 Peak b 10.6 3.6 8.6 38.2 91.2 69.7 58.7

Peak a 0.3 6.5 2.9 1.95 0.87 3.6 0.6 Peak b 1.7 3.7 3.3 6.29 5.61 11.5 3.6

Poly(G1u:Tyr) column

TABLE I1 Effects of Pronase, deoxyribonuclease, ribonuclease, and

trichloroacetic acid treatment on the phosphoproteins The indicated protein substrates (50 pg) were phosphorylated by

the streptococcal protein kinases (20 pg) in a standard reaction mixture as described under “Materials and Methods.” After 5 min of incubation, the phosphorylated proteins were subjected to treatment with hot trichloroacetic acid or the enzymes listed (20 pg) a t 30 “C for 15 min. Residual radioactivity in the proteins was determined by precipitation with trichloroacetic acid as described under “Materials and Methods.”

Radioactivity incorporation

Treatment Kinase a Kinase b

Histone Casein Histone Casein

cpm Control 77,693 65,410 23,045 25,876 Pronase 1,419 1,328 2,032 3,667 DNase 70,480 53,278 24,729 25,236 RNase 77,501 53,285 15,168 17,416 10% trichloroacetic acid 79,325 56,310 19,526 18,981

(90 “C for 15 min)

TABLE I11 Phosphorylation of serine, threonine, and tyrosine residues by

streptococcal protein kinases 200 pg each of poly(G1u:Tyr) (41) and histone were incubated at

30 “C for 10 min with 20 pg of the respective protein kinases and 10 p~ [y-32P]ATP (10 pCi/test). The reaction was stopped by precipi- tation with trichloroacetic acid, and phosphoamino acids were ana- lyzed as described under “Materials and Methods.” The test value obtained with a control of an enzyme alone has been subtracted from each of the test values of radioactivity incorporated. Using poly(Glu:Tyr), the background count for phosphotyrosine (Tyr(P)) was 48 cpm, but using histone, the background counts for phospho- serine (Ser(P)), phosphothreonine (Thr(P)), and Tyr(P) were 406, 155, and 89 cpm, respectively.

Radioactivity

Ser(P) Thr(P) Tyr(P) Substrate Enzyme

cpm1200 pg

Poly(G1u:Tyr) a 0 0 b

4,950 0 0 1,200

Histone a 25,400 630 2,520 b 1,200 90 60 .

previously isolated from human plasma (Chiang and Kang, 1985). A purified polypeptide fragment of M protein (pep M24) extracted by limited pepsin digestion of whole type 24 streptococci was exposed to these enzymes in the presence and the absence of CAMP. The data in Fig. 5 show that the streptococcal protein kinases as well as the eukaryotic protein kinase were capable of phosphorylating the M24 protein; the

t

First Dimension -c

FIG. 3. Identification of phosphotyrosine in the ”P-labeled poly(G1u:Tyr). Phosphorylation of poly(G1u:Tyr) by protein kinase a, acid hydrolysis and separation of the phosphoamino acids by electrophoresis in the first dimension, and ascending chromatography in the second dimension,were as described under “Materials and Methods.” 0, origin. The arrow denotes the position of phosphoty- rosine (P-Tyr), and the heavy circles denote the positions of phospho- serine (P-Ser) and phosphothreonine (P-Thr) standards. A, enzyme control; B, poly(G1u:Tyr) plus enzyme.

M protein did not become phosphorylated in the absence of kinase a or b (data not shown) indicating that the M protein lacked autophosphorylating activity.

The ability of the streptococcal kinases to phosphorylate synthetic peptides copying the amino-terminal regions of two different M proteins and the purified M24 protein was also examined. Both the synthetic peptide of type 24 M protein (amino-terminal 29 residues) and that of type 5 M protein (amino-terminal 35 residues) and the purified M24 protein were phosphorylated by both streptococcal protein kinases (Table IV). The incorporation of phosphate into M24 protein was calculated to be 2.1 and 0.6 mol of phosphate/mol of M24 for enzymes a and b, respectively. An examination of the primary structures of pep M24 (Mouw et al., 1988) and SM5 reveals several possible phosphorylation sites. Thus in pep M24, these sites include an Arg-Ser sequence at positions 4 and 5, and Lys-Thr at positions 102 and 103, 172 and 173, and 207 and 208, whereas the potential phosphorylation sites in SM5 include a Lys-Gly-Thr sequence at positions 4-6, an Asp-Lys-Tyr-Glu at positions 17-20, and a Lys-Thr at posi- tions 29 and 30. The potential phosphorylation sites for SM24 are the same as those for pep M24 at the amino-terminal end of the molecule (see above).

DISCUSSION

Previous studies have shown that unique protein kinases play a pivotal role in the regulation of major metabolic and transport pathways in diverse prokaryotes as discussed in the

Streptococcal Protein Kinases 2961

:“1 8

f-“- I I I

20 40 60 SO

ug of Substrates

FIG. 4. Effect of protein substrate concentration on activity of the streptococcal kinases. Phosphorylation of histone (O), protamine (O), and casein (W) by 10 pg of protein kinase a (A) or protein kinase b ( B ) was as described under “Materials and Methods.” The final concentration of [-y-32P]ATP was 10 p ~ .

A B

I

1 2 3 4 5 6 1

FIG. 5. Phosphorylation of M protein (pep M24) by strep- tococcal and human protein kinases. The phosphorylation reac- tions were performed as described under “Materials and Methods” except that CAMP (1 p ~ ) was included in the mixtures containing the CAMP-dependent protein kinase. Samples were analyzed by SDS- PAGE and autoradiography. Panel A: Coomassie Blue stain of pep M24; panel B: lane I, CAMP-dependent protein kinase without pep M24; lune 2, CAMP-dependent protein kinase with pep M24; lane 3, streptococcal protein kinase a without pep M24; lane 4, streptococcal protein kinase a with pep M24; lane 5, same as lane 4 in the presence of 1 pM CAMP; lane 6, plasma protein kinase alone; lane 7, plasma protein kinase with pep m24. No phosphorylation of M24 alone (without added kinase) was detected (data not shown).

Introduction. In contrast, little information is available re- garding the involvement of this covalent modification device in functions that affect intercellular host-pathogen relations (Ganguly et al., 1985; Wightman and Raetz, 1984). The pres- ence of protein kinase activity in s. pyogenes (Reizer et aL, 1983,1984) and the large body of information concerning the

major virulence determinants of this human pathogen (Lance- field, 1962) prompted us to investigate the involvement of endogenous and exogenous protein kinases in the pathogenic- ity of group A streptococci. The data presented here represent our first efforts toward this goal. We have provided evidence for: 1) the presence of two forms of protein kinase activity in S. pyogenes; 2) the capability of the streptococcal kinases to catalyze the phosphorylation of eukaryotic and prokaryotic proteins; 3) the capability of these enzymes to phosphorylate serine and tyrosine residues; and 4) the phosphorylation of streptococcal M protein by endogenous as well as by human protein kinases.

The distinction between the two kinase activities in S. pyogenes was achieved by column chromatography and was further defined by the converse preference of these enzymes for exogenous protein substrates. Thus, protein kinase a phos- phorylated histone at a higher rate (approximately 10-fold) than protamine, whereas kinase b phosphorylated protamine at a higher rate than histone. Protein kinase a is most likely the 60-kDa HPr-serine kinase that was previously described by Reizer et al. (1984). Nevertheless, by using exogenous protein substrates and a g1utamate:tyrosine copolymer, we have shown that protein kinases a and b can catalyze the phosphorylation of tyrosyl residue in addition to the seryl residues reported previously (Reizer et al., 1984) for endoge- nous protein substrates. The low level of threonine phospho- rylation observed in the present study is of questionable significance.

The unexpected broad specificity of the streptococcal pro- tein kinases does not concur with the strict specificity for tyrosine residues that is frequently shown with protein-tyro- sine kinases of eukaryotes (Hunter and Cooper, 1985). It is not yet known whether the broad specificity of the strepto- coccal kinases is intrinsic, similar to that of casein kinase that possesses protein-tyrosine, -threonine and -serine kinase activities (Lu and Tao, 1986), or whether it is due to a protein- tyrosine kinase distinct from a protein-serine kinase in the partially purified enzyme preparations. Similarly, although the activity of kinases a and b appears to be catalyzed by distinct proteins, we cannot eliminate the possibility that these activities are related due to proteolytic degradation of kinase a. Future studies with homogeneous forms of the streptococcal kinases will address this question.

Previous studies have shown that prokaryotic protein kinases are incapable of phosphorylating exogenous proteins such as histone, protamine, phosvitin, or casein which are commonly used in the assay of their eukaryotic counterparts (Cozzone, 1984). By contrast, the streptococcal kinases readily phosphorylated histone, protamine, casein, and synthetic polypeptides in addition to endogenous and heterologous pro- karyotic proteins. These findings suggest that the restricted substrate specificity and the tight requirement for endogenous protein substrates that were previously shown with the pro- tein kinase activity of coliform bacteria are not properties shared by all prokaryotic protein kinases (Cozzone, 1984). The capability of the streptococcal enzymes to catalyze phos- phorylation of readily available substrates can be most useful in experiments designed to characterize these kinases further since previous studies of bacterial protein kinases have been hampered by limiting quantities of pure endogenous protein substrates.

Of particular interest was the finding that the endogenous streptococcal protein kinases were capable of phosphorylating the M protein isolated from the surface of the streptococcal cells. The ability of the human plasma protein and CAMP- dependent protein kinases to phosphorylate the amino-ter-

2962 Streptococcal Protein Kinases TABLE IV

Phosphorylation of M24 and synthetic peptides by streptococcal protein kinases a and b The radioactivity incorporated into different synthetic peptides was determined in a standardized assay (see

under “Materials and Methods”). The net radioactivity incorporated into the peptides was obtained from the experimental count minus the sum of control counts obtained with the enzyme alone and with each peptide without enzyme. Amino acid sequence of SM5:

1 1 0 Ala-Val-Thr-Lys-Gly-Thr-Ile-Asn-Asp-Pro-Gln-Ala-Lys-Glu-Ala-Leu-

Asp-Lys-Tyr-Glu-Leu-Glu-Leu-Glu-Asn-His-Asp-Leu-Lys-Thr-Asn-Asn-Glu-Gly-Leu-Lys. 20 30

Amino acid sequence of SM24: 1 10 Val-Ala-Thr-Arg-Ser-Glu-Thr-Asp-Thr-Ser-Glu-Lys-Val-Glu-Glu-

Arg-Ala-Asp-Ser-Phe-Glu-Ile-Glu-Asn-Asn-Thr-Leu-Lys-Leu. 20

Radioactivity incorporation

Protein kinase a Protein kinase b cpml5O pg of peptides

Peptides

Poly(G1u:Tyr) 34,400 22,185

14,658 1,602 89,954 24,987

SM5 (1-35) 11,806 1,986 SM24 (1-29) M24

minal ends of these molecules may have considerable biolog- ical relevance because the amino termini of these molecules protrude to the outermost surface of the streptococcal cells where they may influence the interactions with various host products and cells. Indeed, preliminary studies’ indicate that in vivo phosphorylation of the M protein by eukaryotic protein kinases reduces binding of fibrinogen to the surface of the organisms, a reaction previously shown to block the recogni- tion of the organisms by the opsonic third component of complement (Whitnack and Beachey, 1982). The degree to which endogenous protein kinase phosphorylates the surface M protein and the role of exogenous protein kinases present in the plasma and tissue fluids of the host in the pathogenesis of streptococcal infections remain to be elucidated.

Acknowledgments-We wish to thank S. Wathen, V. Woo, and E. Chiang for expert technical assistance and J. Smith and B. Scott for expert secretarial assistance. We are grateful to Dr. M. Kotb for helpful discussions and critical review of the manuscript.

REFERENCES

Allen, J. F., Sanders, C. E., and Holmes, N. G. (1985) FEBS Lett.

Antranikian, G., Herzberg, C., and Gottschalk, G. (1985) Eur. J. Biochem. 153,413-420

Beachey, E. H., Stollerman, G. H., Chiang, E. Y., Chiang, T. M., Seyer, J. M., and Kang, A. H. (1977) J. Exp. Med. 45,1409-1483

Beachey, E. H., Seyer, J. M., Dale, J. B., and Hasty, D. L. (1983) J. Biol. Chem. 258,13250-13257

Beyreuther, K., Raufuss, H., Schrecker, O., and Hengstenberg, W. (1977) Eur. J. Biochem. 75 , 275-286

Chiang, T. M., and Kang, A. H. (1985) Arch. Biochem. Biophys. 2 4 3 ,

Cozzone, A. J. (1984) Trends Biochem. Sci. 9 , 400-403 Desmarquets, G., Cortay, J. C., and Cozzone, A. J. (1984) FEBS Lett.

Deutscher, J., and Engelmann, R. (1984) FEMS Microbiol. Lett. 23,

Deutscher, J., and Saier, M. H. (1983) Proc. Natl. Acad. Sci. U. S. A.

Edelman, A. M., Blumenthal, D. K., and Krebs, E. G. (1987) Annu.

Enami, M., and Ishihama, A. (1984) J. Biol. Chem. 259,526-533 Ganguly, C. L., Dale, J. B., Courtney, H. S., and Beachey, E. H.

193,271-275

530-531

173,337-341

157-162

80,6790-6794

Reu. Biochem. 56, 567-613

(1985) J. Biol. Chem. 2 6 0 , 13342-13346

* E. Whitnack, T. Chiang, and E. Beachey, unpublished data.

Holuigue, L., Lucero, H. A., and Vallejos, R. H. (1985) FEBS Lett.

Hunter, T., and Cooper, J. A. (1985) Annu. Rev. Biochem. 54, 897-

Kang, A. H., Piez, K. A., and Gross, J. (1969) Biochemistry 8 , 3648-

Khandelwal, R. L., Spearman, T. N., and Hamilton, I. R. (1973)

Krebs, E. G. (1985) Biochem. SOC. Trans. 13,813-820 Laemmli, U. K. (1970) Nature 227,680-685 Lancefield, R. C. (1962) J. Zmmunol. 8 9 , 307-313 Li, H.-C., and Brown, G. G. (1973) Biochem. Biophys. Res. Commun.

Londesborough, J . (1986) J. Bacteriol. 165,595-601 Lowry, 0. H., Rosebrough, N. J., Farr, A. L., and Randall, R. J. (1951)

Lu, P. W., and Tao, M. (1986) Biochem. Biophys. Res. Commun. 139 ,

Merrifield, R. B. (1963) J. Am. Chem. SOC. 8 5 , 2149-2154 Mimura, C. S., Poy, F., and Jacobson, G. R. (1987) J. Cell. Biochem.

Mouw, A. R., Beachey, E. H., and Burdett, V. (1988) J. Bacteriol.

Nimmo, H. G. (1984) Trends Biochem. Sci. 9,475-479 Reizer, J., and Peterkofsky, A. (1987) In Sugar Transport and Metab-

olism in Gram-positiue Bacteria (Reizer, J., and Peterkofsky, A., eds) pp. 333-364, Ellis Horwood Ltd., Chichester, Great Britain

Reizer, J., Novotny, M. J., Panos, C., and Saier, M. H., Jr. (1983) J. Bacteriol. 156 , 354-361

Reizer, J., Novotny, M. J., Hengstenberg, W., and Saier, M. H., Jr. (1984) J. Bacteriol. 160 , 333-340

Reizer, J., Deutscher, J., Sutrina, S., Thompson, J., and Saier, M. H., Jr. (1985) Trends Biochem. Sci. 10 , 32-35

Reizer, J., Peterkofsky, A., and Romano, A. H. (1988) Proc. Natl. Acad. Sci. U. S. A. 85,2041-2045

Rubin, R. A., and Earp, H. S. (1983) Science 219,60-63 Skorko, R. (1984) Eur. J. Biochem. 145 , 617-622 Spudich, J. L., and Stoeckenius, W. (1980) J. Biol. Chem. 255,5501-

Timkovich, R. (1977) Biochem. Biophys. Res. Commun. 7 4 , 1463-

Turner, A. M., and Mann, N. H. (1986) J. Gen. Microbiol. 132,3433-

Wang, J. Y. J., and Koshland, D. E., Jr. (1978) J. Biol. Chem. 2 5 3 ,

Wang, J. Y. J., and Koshland, D. E., Jr. (1981) J. Biol. Chem. 256,

Wang, J. Y. J., and Koshland, D. E., Jr. (1982) Arch. Biochem.

Whitnack, E., and Beachey, E. H. (1982) J. Clin. Znuest. 6 9 , 1042-

Wightman, P. D., and Raetz, C. R. H. (1984) J. Biol. Chem. 259 ,

181, 103-108

930

3655

FEBS Lett. 3 1,246-250

53,875-881

J. Biol. Chem. 193,265-275

855-860

33,161-171

170,676-684

5503

1468

3440

7605-7608

4640-4648

Biophys. 2 1 8 , 59-67

1045

10048-10052