Isolation and characterization of a Mitogen characteristic of Group A streptococci (Streptococcus pyogenes)

Zbl. Bakt. 282, 67-82 (1995) © Gustav Fischer Verlag, Stuttgart· Jena . New York

Isolation and Characterization of a Mitogen Characteristic of Group A Streptococci (Streptococcus pyogenes)1 DIETER GERLACH!, ELISABETH GUNTHER, WERNER KOHLER!, STEFAN VETTERMANN 3 , BERNHARD FLEISCHER 2 , and KARL-HERMANN SCHMIDT! ! Friedrich-Schiller-Universitiit Jena, Institut fur Experimentelle Mikrobiologie, D-07745 Jena, Germany 2 Bernhard-Nocht-Institut fur Tropenmedizin, Institut fur Medizinische Mikrobiologie, D-20359 Hamburg, Germany 3 Institut fur Molekulare Biotechnologie, D-07745 Jena, Germany

Received July 25, 1994· Accepted August 17, 1994

Summary It has been supposed for many years that group A streptococci may elaborate more than the three well known erythrogenic toxins A, B or C (ETA, ETB, ETC). The analysis of the culture supernatant of streptococcal strain 27297 carrying neither genes for ETA nor ETC revealed mitogenic activity at pH 7.3 in isoelectric focusing. This mitogen of strain 27297 was purified by hydrophobic adsorption to Phenyl-Sepharose following FPLC chromatography on a Mono S column resulting in two proteins with mitogenic activity called AX and BX, respectively. Both differed in only one aminoterminal residue. The mitogenic activity of BX lacking one amino terminal arginine was found to be about 100 times higher than that of AX. The aminoterminus of BX does not correspond to a predictable cleavage site for signal peptidase. We assume that BX was produced after translation by cleavage of the mature protein or the AX molecule with streptococcal proteinase (ETB) or an arginylaminopeptidase which is detectable on whole cells. The purified proteins BX and AX showed molecular weights of about 27 kDa in SDS electrophoresis and isoelectric points of 8.3 (AX) and 7.3 (BX) in isoelectric focusing, respectively. Both proteins were produced by practically all group A strains tested but not by groups B, C, G or H streptococci. Therefore, AX or BX seem to be proteins characteristic of group A streptococci.

Abbreviations used in this paper: ETA, erythrogenic toxin type A; ETB, erythrogenic toxin type B; ETC, erythrogenic toxin type C; AX and BX mitogen stimulating preferentially V~8 and V~2-bearing T cells; M'X non-mitogenic protein, Tl, T2, T3, T. .. , streptococcal T types; V~, V~-region of T cell receptor; ETA +, ETA-, ETB+, ETC+, ETC-, genotypes of ETA, ETB or ETC; TRIS, Tris(hydroxymethyl)aminomethane; LTT, lymphocyte transformation test.

68

D. Gerlach et al. Introduction

Group A streptococci may elaborate several erythrogenic toxins (types A, B, and C) designated ETA, ETB, and ETC (synonyms: streptococcal pyrogenic exotoxins, SPE-A, SPE-B, SPE-C). All of these toxins belong to the family of mitogenic toxins with superantigenic properties and are involved in streptococcal pathogenesis (1,3, 6,9, 11, 14, 19, 25, 30). These toxins should be involved in the pathogenesis of scarlet fever, erysipelas, and streptococcal toxic shock syndrome (STSS), the latter being clinically characterized by multiorgan involvement. They have been purified, biochemically characterized and most of corresponding genes have been cloned and sequenced (9, 10, 12, 22, 34, 37). One of the typical charactersitics of these toxins is their ability to induce T cell proliferation in vitro. Because Streptococcus pyogenes (group A Streptococcus) has reemerged in recent years as a cause of severe human disease, there has been an increased interest in these toxins. It is noteworthy that virtually all culture supernatants of clinical isolates of Streptococcus pyogenes tested show strong mitogenic activity affecting human T-cells although some of them produce neither ETA, ETC, no SSA (streptococcal superantigen similar to staphylococcal enterotoxin B (26). In addition, we found that the mitogenicity and pyrogenicity commonly associated with ETB was due to a contamination with minute amounts of additional Vf38 stimulator(s) (3, 11). It is very likely that such clinical isolates express novel superantigen(s) uncharacterized until now. This report describes the purification and biochemical characterization of two new mitogens (AX and BX) from a strain of Streptococcus pyogenes, isolated from a scarlet fever patient.

Materials and Methods Streptococcal strains. Strain 27297 (T12, B3264; genotype ETA-, ETB +, ETC-) selected for toxin isolation had been isolated from a scarlet fever patient in 1984. Other strains used came from the culture collection of the Institute of Experimental Microbiology. The bacteria were grown on a large scale at 30 DC for 20 h in yeast dialysate medium (2 kg yeast extract and 200 g pancreatic casein peptone dialysed against 50 L water) at pH 7.2 with automatic addition of glucose and sodium hydroxide solutions (8). Toxin production of the strains listed in Table 3 was screened qualitatively by Western blotting on a small scale only. After growing in batch cultures for 24 hours at 37 C in standard 1 nutrient broth (Merck), 25 ml of culture supernatants were precipitated with 5% (w/v) trichloroacetic acid. The precipitate was dissolved with SDS sample buffer, neutralized with 1 M Tris (final volume 0.5 ml) and used for SDS electrophoresis. Toxin (M'X, AX, BX) purification. After fermentation of strain 27297 in a 50 L culture, the streptococci were killed by addition of 250 m130% (w/v) H 2 0 2 • The culture supernatant was saturated with 40% ammonium sulfate and the mitogen was adsorbed by addition of 1 L Phenyl-Sepharose 6 (Pharmacia, fast flow, high substituted) under stirring. After washing the adsorbent with 40% saturated ammonium sulfate solution the toxin was eluted with 5 L 0.02 M TRIS - 0.01 M HCI buffer (pH 8.1). The combined eluates were precipitated by addition of solid ammonium sulfate (0.5 kg/L). The precipitate was solubilized with water and dialysed against 0.05 M TRIS - 0.0375 M HCI buffer (buffer A, pH 7.65). This material was applied to a Q-Sepharose column (fast flow, Pharmacia, Scm x 10 em, buffer A equilibrated). The column was washed stepwise with buffers A, B (buffer A plus 0.1 M NaCl), and C (buffer A plus 1 M NaCI). The break-through peak (buffer A) D

Streptococcal superantigen

69

containing the majority of the mitogenic activity was concentrated by ammonium sulfate precipitation (50 g/100 ml); the precipitate dissolved in water and dialysed against 0.01 M acetate buffer pH 4.6 was applied to a 10110 Mono S column (FPLC, Pharmacia). A linear gradient from 0 to 1M NaCI in 0.01 M sodium acetate at pH 4.6 was used to develop the column. Two main peaks called AX and BX were obtained. Some activity eluted with buffer B of first batch chromatography on Q-Sepharose was analyzed more intensively on a column of HiLoad 16/10 Q-Sepharose (Pharmacia) using a linear gradient from 0 to 1 M sodium chloride in 0.01 M TRIS-HCI buffer (pH 7.5). Isoelectric focusing in polyacrylamide gel (lEF-PAGE). The lEF-PAGE containing 5% (w/v) acrylamide, 10% (v/v) glycerol, 0.5% Ampholine 3-6, 1% Ampholine 6-8, and 1 % Ampholine 7-9 were polymerized chemically. Focusing and staining of gels were done as described previously (18). For Western blotting, the protein transfer was achieved by diffusion to nitrocellulose membrane for 30 minutes. Immunization. The protein BX from the Mono S column (Fig. 3) used for immunization was purified by lEF-PAGE. After fixation with 5% trichloroacetic acid, the visible bands were cut out and the proteins eluted with 1 M TRIS for 4 hours at room temperature. This eluate was used for immunization of two rabbits. SDS-PAGE and Western blotting. The purity of samples was monitored and the molecular weight (MW) was estimatd by SDS-PAGE with a 12.5% running gel according to Laemmli (23). Blotting onto nitrocellulose membrane (NCM) was done according to the procedure of Towbin et al. (32). The position of the protein bands on the NCM were determined by staining with Ponceau S (0.1 % Ponceau S, 1 M acetic acid) and the destained NCM (0.05 M Tris-HCI buffer, pH 8) was used for immunostaining. After blocking, the blot was incubated for 1 h with specific rabbit antiserum diluted 1 : 10000 and developed with horseradish peroxidase-conjugated anti-rabbit IgG diluted 1: 500 (1 hour). Bands were visualized with 0.025% 3,3'-diaminobenzidine dissolved in 50 mM acetate buffer (pH 5.0), and 0.03% HzO z . Preparative isoelectric focusing. Concentrates of the culture supernatants were analyzed by preparative isoelectric focusing in Sephadex gel (17.5 cm X 3.5 cm X 0.4 cm, 16 h, 300 V) according to Radola (29). The gels contained 1 ml Ampholine (pH 3.5-10), 0.5 ml Ampholine (7-9), 8 ml sample, 10.5 ml water and 1.2 g Sephadex lEF. Assay for proliferative activity. The lymphocyte transformation test (LIT) was performed with mononuclear human blood cells from healthy donors according to Knoll et al. (18). Aminopeptidase assay. Native streptococcal cells of strain 27297 (0.75 mg/ml, dry weight) washed three times were incubated with L-arginine-4-nitroanilide (1 mM/ml) in 0.05 M phosphate buffer (pH 7.2) as described before (4). The enzyme activity was measured at 410 nm and expressed as nMoU(mg*sec) of free 4-nitroaniline. T-cell receptor usage. The stimulation of V~2+ or V~8+ T cells was tested according to Braun et al. (3) using established leukaemic cell lines. Pyrogenicity. The pyrogenicity of purified BX was tested in groups of three rabbits using a dose of 1 flglkg. The rectal temperatures were followed for 5 hours. Amino-terminal sequence determination. The N-terminal amino acids of proteins purified by lEF-PAGE were determined by micro sequence analysis on a pulsed liquid amino-acid sequencer (Applied Biosystems, model 476A, Forster City, USA) following the instructions of the manufacturers. The sample on the micro cartridge was prepared by dot blotting of eluted bands from the IEF-PAGE on PVDV sequencing membrane (Millipore, USA). Preparation of labelled antibodies. The y-globulin from 2 ml of rabbit antiserum was prepared by ammonium sulfate precipitation and conjugated with horseradish peroxidase according to Wilson and Nakane (35).

D. Gerlach et al.

70

pH

Stimulation index

10

8

9

7

8 7

6

6'

15

5

14

4:

j3

2

2

2

3

456

7

8

9

10 11 12 13 14 15 16

o

Fractions

• pH .LTT 1:100 .LTT 1:1000 .LTT 1:10000 Fig. 1. Isoelectric focusing of concentrated culture supernatant of Streptococcus pyogenes strain 29297 (scarlet fever strain, T type 121B3264, genotype ETA-, ETB+, ETC-) in Sephadex gel. The gel was fractionated and eluted with water. The eluted fractions were tested for pH values and mitogenic activity. Results In preparative isoelectric focusing, the culture supernatant of the selected streptococcal strain 27297 (genotype ETA-, ETB+, ETC-) showed only one sharp peak of mitogenic activity at pH 7.3 (Fig. 1). This mitogenic activity was concentrated from large volumes of culture supernatant by hydrophobic adsorption to Phenyl-Sepharose. The following batch chromatography on Q-Sepharose resulted in rwo mitogenically active fractions, the first in the breakthrough (buffer A, 0.05 M TRIS-HCl, pH 7.5) and the second in the eluate with 0.1 M NaCI (buffer B). Rechromatography of buffer B eluate of the first Q-Sepharose separation on the QSepharose high performance column 16/10 (Pharmacia) resulted in a heterogeneous mixture of proteins. Fractions 19 to 25 contained most of the activity (Fig. 2). The protein which was isolated from these fractions was designated M'X, because it was first thought to be an unknown mitogen. It had a molecular weight of 25 kDa (SDS-PAGE) and an isoelectric point of 5.6 (IEF-PAGE) but did not stimulate human lymphocytes. The break-through of the preparative Q-Sepharose contained most of the mitogenic activity. When this material was analyzed by chromatography on the Mono S column (10110, Pharmacia), a complex elution pattern was observed showing three dominant peaks (Fig. 3). These were the main peak (called AX), a smaller peak preceding it (called BX) and one very small peak (C).

Streptococcal super antigen

71

Absorbance STI or mS/sec 0,8,-----------------------------------------------,30

25 0,6

20 15

0,4

Fractions ~ Abs. 280nm

+ mS/cm 2 .. STI 1: 1000

7+ STI 1: 10000

Fig. 2. Ion exchange chromatography on HiLoad 16/10 column of Q-Sepharose. The 0.1 M NaCI eluate from a first batch chromatography on Q-Sepharose was applied. The chromatogram was developed by using a linear salt gradient from 0 to 1 M NaCl in 0.01 TRIS-HCI buffer (pH 7.5). The eluted fractions 16 to 26 contained the majority of mitogenic activity. The protein M'X purified from these fractions by subsequent hydroxyapatite separation was inactive. Its mitogenicity was caused by small amounts of AX or BX protein.

Aba.

I

280nmi

2.0

ex

I

AX

Molll

0.3 1.0

0.2 0.1

o

o

50

100 ml

Fig. 3. Typical elution pattern of Mono S (HR10/1O) separation. The break-through fractions of the first batch separation of Q-Sepharose were applied to Mono S. The column was developed by a salt gradient in 0.01 M acetate puffer (pH 4.6). Three well recognizable peaks of the chromatogram represent mainly mitogens AX (AX), BX (BX), and streptococcal proteinase (C).

72

D. Gerlach et al.

L_

b

c

d

e·

Fig. 4. Isoelectric focusing in polyacrylamide gel. Lane a, protein test mixture 9 (SERVA, Heidelberg); lanes b, c, mitogen BX (strain 27279); lanes d, e, mitogen AX (strain 29297). AX and BX were well separated by isoelectric focusing showing isoelectric points of 7.3 (BX) and 8.3 (AX).

Most of the mitogenic activity was found in the second peak (BX). IEF-PAGE analysis of peaks AX and BX resulted in isoelectric points of 7.3 (BX) and 8.3 (AX) (Fig.4). Western blotting revealed serological identity of proteins AX and BX. In the BX peak, we could detect small amounts of an additional protein with the N-terminal amino acid sequence Asp-Gln-Asn-Phe-Ala-Arg-Asn-Glu-Lys. This demonstrated its identity with the streptococcal proteinase zymogen (Fig. 5). For N-terminal sequencing and the determination of the specific mitogenicity of AX and BX, the Mono S fractions were focused on polyacrylamide gels and after trichloroacetic acid precipitation, the visible bands were cut out. Proteins were extracted with 1 M TRIS and used for sequencing and mitogenicity testing (Tables 1 and 2). The aminoterminal sequencing revealed that AX and BX differed only in one Nterminal amino acid residue. The AX N-terminal sequence was arginine-glutaminethreonine- whereas BX lacked the first arginine (Table 2). In agreement with this, no serological difference between AX and BX could be detected by Western blotting (Fig. 4). The maximal mitogenicity was eluted from fractions of BX although the AX peak contained most of the protein.

Streptococcal superantigen

,

73

->'1

68kDa2 '

e

f

$I ,h

i

Fig. 5. SDS electrophoresis and Western blotting. Lanes a and d, molecular weight standard proteins (Pharmacia); lane i, molecular weight standard proteins (BioRad); lanes b, e, g, BX mitogen; lanes c, f, and h, AX mitogen. Lanes a, b, and c show only the stained SDS-PAGE, the proteins of lanes d, e, f, g, h, and i were transferred onto nitrocellulose membrane, whereby lanes d and i were stained only with amido black lOB. Lanes e and f were developed with antiserum to AX, lanes g and h with antiserum against streptococcal proteinase (Kl19). It is clearly recognizable from lanes g and h that the fractions of AX or BX obtained by FPLC contained small amounts of streptococcal proteinase.

Table 1. Mitogenic activities of isolated AX or BX Substance

Concentration (ng/ml)

Mitogenicity Stimulation ratio

BX-mitogen strain 29297 (isolated band from isoelectric focusing)

3000 300 30 3 0.3

30.4 26.6 8.2 1.3 0.8

AX-mitogen strain 29297 (isolated band from isoelectric focusing)

1500 150 15 1.5 1.15

4.7

1.0 0.7 0.7 0.7

74

D. Gerlach et al.

Table 2. N-terminal amino acid sequences of purified proteins from strain 29297. M'X, protein M'X; AX, protein AX; BX, protein BX; ETB, zymogen of streptococcal proteinase; UX, unidentified protein from a reverse phase separation of strain 27297 concentrate (FPLC; ProRPC-column); BG, Background sequence in fraction 31 of FPLC separation 5

1

10

M'X

Ala-Asn-Ala-Ile-Ile-Glu-Thr-Ala-Lys-Glu-Arg-Phe-Xaa-Gln-GIn-Ala-Ser-

AX

Arg-Gln-Thr-Gln-Val-Ser-Asn-Asp-Val-Val-Leu-Asn-Asp-Gly-Ala-Ser-Lys-ThrLeu-Asn-Gln-Ala-Leu-Ala-

BX

Gln-Thr-Gln-Val-Ser-Asn-Asp-Val-Val-Leu-Asn

ETB

Asp-Gln-Asn-Phe-Ala-Arg-Asn-Glu-Lys

UX

Asn-Val-Ile-Ala-Glu-Ser-Thr-Ile-Ser-Gln-Val-Ser-Val-Glu-Ala-

BG

Ala-Val-Gly-Lys-Leu-Lys-Asn-Phe-Asn-Ala-Glu-Lys-Xxx-Phe-Ala-Phe-Ile-

a b I

Fig. 6. Western blotting of FPLC fractions (Mono S column). Lanes a, c, e, fractions 31132 (AX); lanes b, d, f, fractions 28/29 (BX). Nitrocellulose membrane I was developed with rabbit antiserum against streptococcal proteinase (K40), membrane II with rabbit antiserum against streptococcal pyrogenic exotoxin type B (obtained from Dr. D. W. Watson), and membrane III with human serum (selected by high antibody level to streptolysin 0).

Streptococcal superantigen

75

When we tested the isolated pure BX or AX (after IEF-PAGE), the specific mitogenic activity of BX was about 100 times higher than AX (Table 1). Such big difference in the biological activity of two molecules only differing in one N-terminal amino acid residue was surprising. Because both proteins AX and BX could easily separated by isoelectric focusing (isoelectric points are 8.3 (AX) and 7.3 (BX)) it was possible to detect them in all purification steps. In culture supernatants, both proteins were already present as shown by preparative isoelectric focusing but only at pH 7.3 (the isoelectric point of BX) a peak of mitogenic activity could be detected (Fig. 1). Neither AX or BX showed pyrogenic activity when tested in concentrations of 1 Ilg/ kg in rabbits. Both proteins (AX and BX) stimulated V fl8 T cells, but the AX fraction had a weak action on V132 T cells, too. The C-peak of Mono S separation represented streptococcal proteinase (mostly as zymogen). Some streptococcal proteinase was also detected in the BX and AX fractions of Mono S separation (Figs. 4 and 6). Arginylaminopeptidase could be detected on the streptococcal cells of strain 27297, with a specific activity of 0.35 nMoVsec per mg dry weight. A conversion of AX into BX could not be observed after incubation of AX with whole cells. The analysis of streptococcal strains of different types of group A streptococci and other groups for production of AX or BX has shown that only group A streptococci seem to produce these proteins. Strains of streptococcal groups B, S, G, and H tested by Western blotting (group B, 5 strains, group C, 6 strains, group G j and group H, 5) were negative for production of AX or BX (Table 3). Testing of ou: antisera produced against ETB and ETC has shown that most of these sera contained a.'tibodies against AX and BX, respectively. Also an antiserum against ETB obtained '-,y Dr. Dennis Watson, Minneapolis, in 1980, reacted with AX or BX in immunoblotting (Fig. 6). These results demonstrate the difficulties to separate the streptococcal proteinase and the mitogens AX or BX by biochemical separation methods and it supports the speculation that BX and AX are the mitogenically active materials of some ETB preparations (streptococcal proteinase). TIle mitogenic activity of BX (strain 27297, genotype ETC-) could be neutralized with an antiserum produced against an ETC preparation that, however, was found to react strongly with AX or BX in immunoblotting (Fig. 7). From this antiserum, specific antibodies against AX or BX were prepared by immunoaffinity chromatography (not shown). From two antisera produced by immunization with AX, only one was able to neutralize the mitogenic activity of the culture filtrate of strain 27297. Discussion The pyrogenic exotoxin (erythrogenic toxin) profiles of S. pyogenes isolates have shown that about 50% carry the ETA gene, 100%, the ETB gene, and 25%, the ETC gene (20, 30, 31, 33, 36). About 20% of the isolates carry neither ETC nor ETA gene although such culture filtrates show strong mitogenic activity which is characteristic of bacterial superantigens. Such mitogenic activity may be due to other still unknown and undefined superantigens. We found also that trace amounts of a potent novel V138stimulating activity, not being identical with ETA and ETC, were responsible for the stimulation of V138 + T cells by wild type ETA and ETC as reported previously (3). In

76

D. Gerlach et al.

Table 3. Production of toxins by streptococcal strains of groups A, B, C, G and H. The streptococci were grown in Merck standard broth for 24 hours at 37°C in batch culture. Culture supernatants were concentrated by trichloroacetic acid 50 times. The concentrates were tested for toxin production by immunoblotting. No.

Group

ETA

ETC

AX

M'X

Type; Source

12714 116 27195 27297 35712 36852 37027 37267 37283 37307 37312 37317 37329 37351 37444 37445 37458 37459 37460 37461 37465 37466 37467 37525 A374 AM 12 ATl AT22 B220 C203S C203U M5 NY5 S84 SW-A SW-C SW-D SW-K SW-M SW-p SW-Z Tl8 Tl9 37166 37416 37580 37581

A A A A A A A A A A A A A A A A A A A A A A A A A A A A A A A A A A A A A A A A A A A C C C C

0 0 0 0 0 0 + 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 (+) +++ 0 + +++ +++ 0 +++ +++ 0 0 + 0 0 0 0 0 0 0 0 0 0

0 0 0 0 0 +++ +++ +++ 0 +++ 0 ++ ++ +++ +++ +++ ++ +++ ++ ++ 0 0 0 0 ++ 0 + ++ 0 +++ +++ +++ +++ +++ 0 0 0 0 0 0 0 +++ 0 0 0 0 0

+ +++ (+) ++ + + ++ ++ ++ + + ++ +++ ++ + ++ +++ +++ 0 +++ ++ +++ (+) ++ +++ ++ (+) ++

+

M12; scarlet fever M12 T8; scarlet fever T12; B3264; scarlet fever T2;T2; sepsis n.t.; STSS M12; STSS M1; STSS n.t.; STSS M1; STSS M1; STSS M12; STSS T56; STSS T56; sepsis T56; sepsis M1fT1; STSS n.t.; STSS T2/0F2; STSS T48; STSS Tl; STSS Tl; STSS I4/0F4; STSS n. t.; blood culture T12; nephritis M12;Tl;T22 T8

(±)

++ ++ +++ +++ ++ ++ + ++ ++ + +++ ++ 0 + 0 0 0 0

0 ++ + ++ +++ +++ +++ +++ +++ +++ +++ (+) + + + + (+) 0 + + + 0 ++ + 0 0 ++ ++ + + + ++ ++ ++ ++ + +++ +++ (+)

++ ++ + 0

T3;-

T3;M51strain Manfredo M12, T1O; scarlet fever T3/13 M1; STSS, Sweden MlfT65; STSS, Sweden M1; STSS, Sweden Ml; STSS, Sweden Ml; STSS, Sweden Ml; STSS, Sweden Ml; STSS, Sweden T18; ETC producer strain T19; ETB producer strain Tonsillitis Wound isolates Urine isolate Abscess

Streptococcal superantigen

77

Table 3. Continued No.

Group

ETA

ETC

AX

M'X

H46A 36420 37366 37374 37464 37470 37497 37548 37584 B5230 B5241 B5242 B5244 B5246 V2186 V2190 V2399 V2551 V2697

C G G G G G G G G B B B B B H H H H H

0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0

0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0

0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0

+x (+) + (+) (+)

STSS

=

Type; Source

Erysipelas Blood culture Pharyngitis Blood culture Otitis media Throat isolate Wound isolate Urine isolate IIlC IIIC IIIIR IIIIR lAIC

0 0 0 0 0 0 0 0 0 0 0 0

+ (+)

streptococcal toxic shock syndrome Stimulation index 35

30

25 20

15 10 5 0

1000 500

250

100

50

25

10

5

2.5 medium BX

Antiserum concentration

Fig. 7. Neutralization of mitogenic activity of mitogen BX of strain 27297 (genotype ETA-, ETC-) with rabbit antiserum K121. The antiserum was raised against ETC but was found to react strongly with AX and BX. At the highest antiserum concentration 1000 on the ordinate 100 [-11 of the lymphocyte culture were mixed with 10 [-11 undiluted antiserum. At the lowest antiserum concentration (2.5), 2.5 [-11 of a 1: 100 dilution of K121 was used. A medium control (medium) and a BX sample without addition of antiserum were included.

78

D. Gerlach et al.

addition, we could demonstrate that ETB is not a superantigen and its mitogenic and pyrogenic activity is caused by minor contamination with this (these) unknown mitogen(s) (11). Streptococcus pyogenes strain 27297 produces only one mitogenically active peak when analyzed by isoelectric focusing, showing an isoelectric point of 7.3. This strain produced neither ETA nor ETC (genotypes ETA-and ETC-) and its mitogenic activity seems to have been caused by that unknown mitogen (superantigen) which we could detect in culture supernatants of some streptococcal strains (7, 11). In first experiments, we obtained a protein (called M'X, isoelectric point 5.6, molecular weight 26 kDa) without homology of N-terminal amino acid sequence to known proteins from the EMBL data bank. The rabbit antiserum produced against M'X did not neutralize its mitogenic activity. Western blotting revealed that this protein was produced not only by group A streptococcal strains but also by some group C and group G strains but not by group B strains (Table 3). The mitogen of strain 27297 could be isolated from the break-through fractions of the Q-Sepharose column by ion exchange chromatography on the FPLC column Mono S. Ion exchange chromatography on Mono S resulted in three well marked peaks. The first small peak (called C, Fig.3) mainly corresponded to streptococcal proteinase precursor. It was identified by immunoblotting and amino acid sequencing. The aminoterminal sequence of D-Q-N-F-A-R-N-E-K- was found according to the reported amino acid sequence at positions 28 to 36 for ETB (13) and was identical to the proposed N-terminal amino sequence of the DNA sequence for the mature zymogen of streptococcal proteinase. In addition, immunoblotting revealed bands in the molecular weight range of 30 kDa indicating the existence of active streptococcal proteinase in these fractions, too. The peak (BX) of the Mono S chromatography contained most of the mitogenic activity. The mitogen purified from this peak was called BX protein. The aminoterminal sequence of BX was Q-T-Q-V-S- and identical to the previously isolated mitogen from Streptococcus pyogenes strain NY5 (22,37). A protein with the same N-terminal sequence as BX was copurified with mitogen caned SSA (with high homology to Staphylococcus aureus enterotoxins) by Mollick et al. (26) but there was no reference to its mitogenic activity. The amino terminus of the purified streptococcal superantigen of Mollick et al. (26) was more homologous to the amino termini of staphylococci enterotoxins B, Cl, and C3 than to those of streptococcal pyrogenic exotoxins A, B, C or other streptococcal toxins. It activated human T cells which expressed V[:11,3 or 15 but not V[:12 or V[:18 like BX or AX. The main protein peak (peak AX) of Mono S column chromatography contained the mitogen AX; its N-terminal amino acid sequence was analyzed and found to be R-Q-TQ-V-S. This amino acid sequence was identical to BX, with the exception of one additional N-terminal arginine. This arginine increased the isoelectric point of BX from 7.3 to 8.3 (AX). The existence of this protein AX has not been reported before. Surprisingly, we found now that BX was about 100 times more active in T cell stimulation with a preference to V[:18 bearing T cells, in contrast to AX stimulating preferentially V[:12 bearing T cells. Calculations of the signal sequence cleavage sites of the published DNA sequence showed that it was unlikely for BX to be the primarily expressed protein; the calculated probability was only 0.000016. Also the N-terminus of AX R-Q-T-Q- (positions 42143 of DNA-derived sequence, probability 0.489, calculated molecular weight 25520)

Streptococcal superantigen

79

must not represent the primary cleavage site of the signal peptidase (probability 1 at positions 30/31, calculated molecular weigh of mature protein 26734). Western blotting analysis of streptococcal culture filtrate concentrates showed a weak additional band of about 1 to 2 kDa that was larger than AX. Immunoblotting of purified BX analyzed by IEF-PAGE showed a weak additional band with an isoelectric point of 6.5 but its N-terminal amino acid sequence was found to be identical to BX Q-T-Q-. If there was a fragmentation of this molecule on the Cterminus, it was °not found by analysis. All fractions of BX purified by FPLC chromatography on Mono S columns contained small amounts of streptococcal proteinase if tested by immunoblotting. Therefore, we speculate that the streptococcal proteinase or their cellular form (24) were involved in secondary cleavage of the mature protein to AX and BX generating in this way superantigenic properties of BX because the streptococcal proteinase hydmlyses typical trypsin substrates such as benzoylarginine amide, too (5). A conversion of AX to BX by the arginylaminopeptidase seems to be possible although we have been unable to convert experimentally AX to BX by streptococcal cells or streptococcal proteinase. Although the cleavage of AX to BX by streptococcal proteinase could not be observed, a processing of the primarily expresssed mitogen (calculated aminoterminus V-T-T-V-T-, amino acid positions 31-32-) by the intracellular streptococcal proteinase seems to be more likely (24). Activation of AX to BX by arginylaminopeptidase could be expected only from an intracellular reaction because this enzyme is strongly associated with the cytoplasmatic membrane and only small activities are located on the cell surface (4). If the streptococcal proteinase triggers a superantigenic activity by posttranslational cleavage of less active protein AX (produced by all tested human pathogenic group A strains) to a highly active superantigen BX, this would explain the ETB (streptococcal proteinase) function as a pathogenicity factor as well as the finding that almost all streptococcal strains isolated from toxic shock patients produce large amounts of streptococcal proteinase (15, 31). A similar role of streptococcal proteinase was discussed by Kapur et al. (16) who showed cleavage of inactive human interleukin-l~ precursor to the biologically active interleukin-l~ by this proteinase. Also a cleavage of human fibronectin and vitronectin by streptococcal proteinase was reported (17). In contrast to the streptococcal pyrogenic exotoxins of types A and C (ETA, ETC), neither AX nor BX were pyrogenic for rabbits at concentrations of 1 Ilg/kg. For a nonpyrogenic variant of TSST-l (toxic shock syndrome toxin 1) which had been isolated from an ovine mastitis Staphylococcus aureus strain, a similar effect was reported. This toxin, TSST-ovine (TSST-O), is only weakly mitogenic for T cells and nonpyrogenic, and it does not enhance endotoxin shock (27). As key differences between superantigens, their amino termini and the degree to which a long central helix is covered by surface loops (28) were discussed. This aspect could explain the strong difference in biological activity between AX and BX, which are different only in one N-terminal amino acid residue. The strongly basic arginine residue in the AX molecule did not only change its charge dramatically but also its biological activity in comparison to BX. The first author of a purification protocol for erythrogenic toxin type B (ETB) reported that only one of the three bands detected by isoelectric focusing showed biological activity (2). Because the reported isoelectric point of this active band of 8.4 was in agreement with the value of the new mitogen AX, we were not surprised that the

80

D. Gerlach et al.

antisera against ETB obtained from one of these authors and also the antisera produced by us contained antibodies against AX or BX. That means that BX or AX could be the reported active component of ETB. Because we were able to extract the BX mitogen also from streptococcal-cells, it could be that there is a relation to the T cell mitogen integrated in the cytoplasmatic membrane (21) or to the mitogenic component, released by pepsin-extracted M-proteins from streptococcal cells (7). Acknowledgment. This work was supported by grant Ge684/2-1 from Deutsche Forschungsgemeinschaft.

References

1. Alouf, J. E., H. Knoll, and W. Kohler: The family of mitogenic, shock-inducing and superantigenic toxins from staphylococci and streptococci. In: J. E. Alouf and J. H. Freer (eds.), Sourcebook of bacterial protein toxins. Academic Press. London (1991), pp. 367-414 2. Barsumian, E. L., C. M. Cunningham, P. M. Schlievert, and D. W. Watson: Heterogeneity of group A streptococcal pyrogenic exotoxin type B. Infect. Immun. 20 (19g0) 512-518 3. Braun, M. A., D. Gerlach, U. F. Hartwig, f.-H. Ozegowski, F. Romagne, S. Carrel, W. Kohler, and B. Fleischer: Stimulation of human T cells by streptococcal "superantigen" erythrogenic toxins '(scarlet fever toxins). J. Immunol. 150 (1993) 2457-2466 4. Cowman, R. A. and S. S. Baron: Comparison of aminopeptidase activities in four strains of mutans group oral streptococci. Infect. Immun. 61 (1993) 182-186 5. Elliott, S. D. and T.-Y. Liu: Streptococcal proteinase. Methods of Enzymology 19 (1970) 252-261 6. Fleischer, B., R. Gerardy-Schahn, B. Metzeroth, S. Carrel, D. Gerlach, and W. Kohler: An evolutionary conserved mechanism of T cell activation by microbial toxins. Evidence for different affinities of T cell receptor interaction. J. Immunol. 146 (1991) 11-17 7. Fleischer, B., K-H. Schmidt, D. Gerlach, and W. Kohler: Separation of T-cell-stimulating activity from streptococcal M protein. Infect. Immun. 60 (1992) 1767-1770 8. Fuvessy, I., J. Boszormenyi, and A. Veress: The effect of physicochemical parameters of Streptococcus pyogenes cultures on the formation of O-streptolysin. Acta Microbiol. Acad. Sci. Hung. 14 (1967) 335-343 9. Gerlach, D., H. Knoll, W. Kohler, f.-H. Ozegowski, and V. Hrfbalova: Isolation and characterization of erythrogenic toxins. V. Identity of erythrogenic toxin type Band streptococcal proteinase precursor. Zbl. Bakt. 255 (1983) 221-233 10. Gerlach, D., H. Alouf, L. Moravek, M. Pavlik, and W. Kohler: The characterization of two new low molecular weight proteins (LMPs) from Streptococcus pyogenes. Zbl. Bakt. 277 (1992) 1-9 11. Gerlach, D., W. Reichardt, B. Fleischer, and K.-H. Schmidt: Separation of mitogenic and pyrogenic activities from so-called erythrogenic toxin type B (streptococcal proteinase). Zbl. Bakt. 280 (1994) 507-514 12. Goshorn, S. C. and P. M. Schlievert: Nucleotide sequence of streptococcal pyrogenic exotoxin type C. Infect. Immun. 56 (1988) 2518-2520 13. Hauser, A. R. and P. M. Schlievert: Nucleotide sequence of the streptococcal exotoxin type B gene and relationship between the toxin and the streptococcal proteinase precursor. J. Bacteriol. 172 (1990) 4536-4542 14. Hauser, A. R., D. L. Stevens, E. L. Kaplan, and P. M. Schlievert: Molecular analysis of pyrogenic exotoxins from Streptococcus pyogenes isolates associated with toxic shocklike syndrome. J. Clin. Microbiol. 29 (1991) 1562-1667

Streptococcal super antigen

81

15. Holm, S. E., A. Norrby, A.-M. Bergholm, and M. Norgren: Aspects of pathogenesis of serious group A streptococcal infections in Sweden, 1988-1989. J. Infect. Dis. 166 (1992) 31-37 16. Kapur, V., M. W. Ma;esky, L.-L. Li, R. A. Black, and]. M. Musser: Cleavage of interleukin 1~ (IL-1~) precursor to produce active IL-1~ by conserved extracellular cysteine protease from Streptococcus pyogenes. Proc. Natl. Acad. Sci. USA 90 (1993) 7676-7680 17. Kapur, V., S. Topouzis, M. W. Ma;esky, L. L. Li, M. R. Hamrick, R.]. Hamill,]. M. Patti, and]. M. Musser: A conserved Streptococcus pyogenes extracellular cysteine protease cleaves human fibronectin and degrades vitronectin. Microbiol. Pathogenesis. 15 (1993) 327-346 18. Knoll, H., D. Gerlach, ].-H. Ozegowski, V. Hribalova, and W. KOhler: Purification and characterization of erythrogenic toxins of Streptococcus pyogenes. VI. Mitogenic activity of isoelectrically focused erythrogenic toxin preparations and culture supernatants of group A streptococci. Zbl. Bakt. 256 (1983) 49-60 19. Knoll, H., S. E. Holm, D. Gerlach, O. Kuhnemund, and W. Kohler: Tissue cages for study of experimental streptococcal infection in rabbits. II. Humoral and cell-mediated immune response to erythrogenic toxins. Immunobiol. 169 (1985) 116-127 20. Knoll, H., ]. Sramek, K. Vrbova, D. Gerlach, W. Reichardt, and W. Kohler: Scarlet fever and types of erythrogenic toxins produced by the infecting streptococcal strains. Zbl. Bakt. 276 (1991) 94-106 21. Itoh, T., H. Satoh, N. Isono, H. Rikiishi, and K. Kumagai: Mechanism of stimulation of T cells by Streptococcus pyogenes: Isolation of a major mitogenic factor, cytoplasmatic membrane-associated protein. Infect. Immun. 60 (1992) 3128-3135 22. Iwasaki, M., H. Igarashi, Y. Hinuma, and T. Yutsudo: Cloning, characterization and overexpression of a Streptococcus pyogenes gene encoding a new type of mitogenic factor. FEBS Letters 331 (1993) 187-192 23. Laemmli, U. K.: Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 224 (1970) 361-385 24. Lo, S. S., S. M. Liang, and T. Y. Liu: Intracellular form of streptococcal proteinase: A clue to a novel mechanism of secretion. Anal. Biochem. 136 (1984) 89-92 25. Marrack, P. and]. W. Kappler: The staphylococcal enterotoxins and their relatives. Science 248 (1990) 705-711 26. Mollick,]. A., G. G. Miller,]. M. Musser, R. G. Cook, D. Grossman, and R. R. Rich: A novel superantigen isolated from pathogenic strains of Streptococcus pyogenes with aminoterminal homology to staphylococcal enterotoxins Band C. J. Clin. Invest. 92 (1993) 710-719 27. Murray, D. L., G. S. Prasad, C. A. Earhart, B. A. B. Leonard, B. N. Kreiswirth, R. P. Novick, D. H. Ohlendorf, and P. M. Schlievert: Immunobiologic and biochemical properties of mutants of toxic shock syndrome toxin-I. J. Immunol. 152 (1994) 87-95 28. Prasad, G. S., C. A. Earhart, D. L. Murray, R. P. Novick, P. M. Schlievert, and D. H. Ohlendorf: Structure of Toxic Shock Syndrome Toxin-I. Biochemistry 32 (1993) 13762-13766 29. Radola, B.].: Isoelectric focusing in layers of granulated gels. II. Preparative isoelectric focusing. Biochim. Biophys. Acta 386 (1974) 181-195 30. Reichardt, W., H. Muller-Alouf, ]. E. Alouf, and W. Kohler: Erythrogenic toxins A, B and C: Occurrence of the genes and exotoxins formation from clinical Streptococcus pyogenes strains associated with streptococcal toxic-shock like syndrom. FEMS MicrobioI. Letters. 100 (1992) 313-322 31. Talkington, D. F., B. Schwartz, C. M. Black,]. K. Todd,]. Elliott, R. F. Breiman, and R. R. Facklam: Association of phenotypic and genotypic characteristics of invasive Streptococcus pyogenes isolates with clinical components of streptococcal toxic shock syndrome. Infect. Immun. 61 (1993) 3369-3374

6 Zhl. Bakt. 28211

82

D. Gerlach et al.

32. Towbin, H., T. Staehelin, and J. Gordon: Electrophoretic transfer of proteins from polyacrylamide gels to nitrocellulose sheets: procedure and some applications. Proc. Nat. Acad. Sci. U.S.A. 76 (1979) 4350-4354 33. Tyler, S. D., W. M. Johnson, J. c. Huang, F. E. Ashton, G. Wang, D. E. Low, and K. R. Rozee: Streptococcal erythrogenic toxin genes: detection by polymerase chain reaction and association with disease in strains isolated in Canada from 1940 to 1991. l. Clin. Microbiol. 30 (1992) 3127-3131 34. Weeks, C. R. andJ. J. Ferretti: Nucleotide sequence of the type A streptococcal exotoxin gene from Streptococcus pyogens bacteriophage T12. Infect. Immun. 52 (1986) 144-150 35. Wilson, M. B. and P. K. Nakane: In: Immunofluorescence and related techniques. W. Knapp, H. Holubar, and G. Wick. Elsevier. (eds.). North-Holland, Amsterdam (1978) 215 36. Yu, C. E. and J. J. Feretti: Frequency of the erythrogenic toxin Band C genes (speB and speC) among clinical isolates of group A streptococci. Infect. Immun. 59 (1991) 211-215 37. Yutsudo, T., H. Murai, J. Gonzalez, T. Takao, Y. Shimonishi, Y. Takeda, H. Igarashi, and Y. Hinuma: A new type of mitogenic factor produced by Streptococcus pyogenes. FEBS Letters 308 (1992) 30-33 Dr. sc. nat. Dieter Gerlach, Klinikum der Friedrich-Schiller-Universitiit lena, Institut fur Experimentelle Mikrobiologie, Winzerlaer Str. 10, 07745 lena e-mail: [email protected]. Tel: (03641) 657121, Fax: (03641) 657133