Main

Although the immunologic basis of poststreptococcal acute RF is now undisputed, the precise mechanism and the specific nature of the streptococcal antigens involved in the pathogenesis are still unclear. Since Kaplan(1) demonstrated the cross-reactivity of M protein with myocardial tissue in 1963, M protein is considered to be one of the major antigens in the development of cross-reactive antibodies in RF. More than 80 serotypes of M proteins have been identified thus far. From DNA analyses of several M protein serotypes, it has been determined that M proteins contain several types of tandem repeats, named A, B, and C blocks, and are roughly divided into two parts: the AB regions which are highly variable, and the C regions which are highly conserved among the different M serotypes(2, 3). Bessen et al.(4) showed that the antigenicity of the C region of the M protein is quite similar among streptococci which previously caused RF outbreaks, suggesting that the C region may be an important virulence determinant of pathogenicity in RF.

In this study, we analyzed the immune response to the C region of the M protein in patients with RF. The structural gene for the C region of type 12 group A streptococcal M protein (M12) was cloned and inserted into an expression vector which allowed production of recombinant protein inEscherichia coli. The purified recombinant M protein (C region) was used as an antigen to measure IgG titers against the M protein. The immunodominant domains within the C region in RF were further investigated using overlapping synthetic peptides of the C region.

METHODS

Subjects. Between January 1989 and June 1995, 10 patients (4 boys, 6 girls, mean age 10.2 ± 4.5 y) with acute RF were studied. The diagnosis of acute RF was based on the revised Jones criteria(5). Nine of the 10 patients had carditis, seven had polyarthritis, three had Sydenham's chorea, and one had erythema marginatum as major manifestations of the disease. IgG titers against the C region of M protein were measured during the acute period in nine patients (27 ± 30 d after onset), and after a follow-up period of about 2-4 y in five patients(3.5 ± 0.9 y after onset). Three of the latter five patients showed persistent carditis: two had mitral valve involvement, and one had aortic valve involvement.

Forty-nine healthy children (28 boys, 21 girls, mean age 10.1 ± 4.2 y) and 26 patients with uncomplicated streptococcal pharyngitis (12 boys, 14 girls, mean age 9.8 ± 4.1 y) served as controls. In all patients with streptococcal pharyngitis, IgG titers against the C region were measured on initial presentation and during convalescence (28 ± 4 d after onset).

Bacterial strains and plasmid . Streptococcus pyogenes strain SS95/12 (M type 12) was obtained from the National Institute of Health, Japan. Genomic streptococcal DNA was prepared by a standard method. E. coli strains XL1-Blue and TB-1 were used as hosts for plasmids Bluescript II SK- (Stratagene, obtained via Toyobo Co., Osaka, Japan) and pMAL-c2(New England Biolabs, Beverly, MA).

PCR amplification and DNA sequencing of the C region of M12 gene. A 5′-terminal primer (primer 1) for the C region was taken from the sequence of another Streptococcus M12 protein (CS24), reported by Robbinset al.(3). Its sequence was 5′-GGGGAATTCCAAAACAAAATTTCAGAAGCAAGC-3′(EcoRI site underlined). The 3′-terminal sequence of the M12 gene has not been reported, but it is known that the anchor region of M protein is quite conserved among different serotypes of streptococci(2). Therefore, a 3′-terminal primer (primer 2) of the C region was synthesized by referring the sequence of M type 5 Streptococcus strain Mansfredo(6), and its sequence was 5′-GGGTCTAGATTAATTTTCTTCTTTGCGTTTTAC-3′ (Xba I site underlined).

Amplification of the M12 gene was performed by PCR in a reaction mixture of 50 μL composed of 500 ng of streptococcal DNA, 25 pmol of oligonucleotide primers, 50 mM KCl, 10 mM Tris-HCl (pH 8.4), 1.5 mM MgCl2, 0.01% gelatin, 0.8 mM dNTP, and 1 U of thermostable DNA polymerase (Ampli Taq, Perkin-Elmer Instruments, Norwalk, CT). Thirty cycles of denaturation(96°C, 1 min), annealing (57°C, 30 s), and extension (72°C, 3 min) were performed in a programmable heart block (DNA Thermal Cycler, Perkin-Elmer). The PCR product then was ligated toEco RI/XbaI-cut Bluescript II, and transformed in XL-1 Blue. Six recombinant clones were isolated and sequenced by the dideoxynucleotide chain-termination method(7).

Both strands were sequenced using T7 and T3 primers in the vector, and two additional sequencing primers were synthesized according to the DNA sequence determined in this study. Sequences of the synthesized primers are as follows: primer 3 (5′-GAAGAAGCAAACAGCA-3′) and primer 4(5′-CAAGAGCAGCTAATTTG-3′).

Production of recombinant M protein. PCR products of the C region were cloned into an expression vector, pMAL-c2, which allowed for the expression of the C region as a MBP fusion product.

A 500-mL Luria broth/glucose/ampicillin medium was inoculated with 5 mL of overnight culture containing fusion plasmid. The inoculum was grown at 37°C, to 2 × 108 cells/mL, and 1.5 mL of 0.1 M isopropyl-β-D-thiogalactopyranoside were added into the culture to induce expression of the recombinant M proteins during further incubation for 2 h at 37°C. The cells were collected, resuspended in 25 mL of column buffer (20 mM Tris-Cl, 200 mM NaCl, 1 mM EDTA, 1 mM azide, pH 7.4). The cell suspension was sonicated for 2 min and centrifuged. The supernatant was loaded onto an amylose resin column (2.5 × 8 cm; New England Biolabs). The recombinant protein was eluted with the column buffer plus 10 mM maltose, pooled, dialyzed extensively against PBS, and lyophilized.

SDS-PAGE and immunoblot analysis. Purified recombinant protein was separated on 10% SDS-polyacrylamide gels. Gels were either stained with Coomassie Brilliant Blue or electrophoretically transferred to nitrocellulose. Rabbit antisera specific for the AB region of M12(8), and a mouse MAb 10B6 against the C region of M6(9) were used as primary antibodies for immunoblot analyses. The antibodies were generous gifts of Dr. T. Murai, Toho University, Tokyo, Japan, and Dr. V. A. Fischetti, Rockefeller University, New York, respectively. The bound antibodies were detected with alkaline phosphatase-conjugated anti-rabbit and anti-mouse IgG, respectively (Sigma Chemical Co., Chemical Co., St. Louis, MO). The bands were visualized by a standard method.

Synthesis of peptides and recombinant protein of the anchor region. Amino acid sequences (Fig. 1) were deduced from the DNA sequence in the current study. To determine the immunodominant epitope of the C region in RF, 13 peptides were synthesized on an Excell peptide synthesizer (Milligen, Burlington, MA) by 9-fluorenylmethoxycarbonyl(Fmoc) chemistry and purified by high performance liquid chromatography on a C18 reverse-phase column (0.4 × 15 cm, Nihon Waters Ltd., Tokyo, Japan). The composition of purified peptides was verified by quantitative amino acid analysis as described by Jones et al.(10).

Figure 1
figure 1

Orientation of the various synthetic peptides and recombinant proteins in streptococcal M protein (A), and their amino acid sequences (B). In A, blocks A1, A2, B1, B2, C1, and C2 designate the location of the repeat blocks. Amino acid sequences were deduced from the DNA sequence of M12 SS95/12.

A recombinant protein including 79 amino acids of the anchor region (M12 C-terminal) was finally made with two primers: primer 5, 5′-GGGGAATTCGGAAAAGCATCAGACTCACAAA-3′ (EcoRI site underlined) and primer 2. The method for the production and purification of this recombinant protein was the same as that used for the whole C region.

Detection of antibodies against recombinant M protein by ELISA. ELISA was performed in a microtiter plate with flat-bottomed wells. One-hundred microliters of purified recombinant M protein (the C region), resuspended at 1 μg/mL in 0.05 M carbonate buffer (pH 9.6), were added to separate wells of the microtiter plates. The plates were incubated at 4°C overnight, rinsed three times in washing solution (0.01 M PBS, pH 7.4, containing 0.05% Tween 20), and then incubated at 37°C for 1 h with 100μL of 1% goat serum albumin in PBS. The plates were rewashed, and 100 μL of diluted sample serum (1/100, 1/1000) were added in triplicate and incubated at room temperature for 1 h. The plates were rewashed, and 100 μL of biotinylated anti-human IgG (1/10000 dilution in PBS) were added for 1 h at room temperature (Vector Lab. Inc., Burlingame, CA). After washes, 100 μL of Vectastein ABC reagent (biotinylated alkaline phosphatase and avidin 2,3-dihydroxypropyl) were added and incubated for 30 min. After a final wash, 100 μL of a 1 mg/mL solution of pnitrophenylphosphate (Sigma Chemical Co.) were added. The plates were incubated at room temperature, and the color development was read in a Multiskan bichromatic ELISA reader (Labosystem Co., Tokyo, Japan). Negative controls included wells with antigen but no serum and wells with serum without antigen. The maximal density of negative controls was subtracted as background.

Positive controls were a mixture of sera from three individuals with high IgG titers to the C region. To make a standard curve between IgG titer against the C region and OD, serially diluted sera of positive controls between 1/100 and 1/10000 were included on each plate. The concentration of IgG specific for the C region in these controls was determined as described in the following section.

Measurement of IgG titer specific for M protein in positive controls. Saturated ammonium sulfate solution was added to the sera from positive controls to 50%. The precipitate was dissolved, dialyzed against PBS, and applied to Hitrap affinity columns (Pharmacia LKB, Uppsala, Sweden) coupled with recombinant C region protein. Antibodies specific for the C region were eluted with 0.1 M glycine-HCl, pH 2.8. The concentration of IgG specific for the C region in the eluates was measured in an ELISA by comparing the eluates with serially diluted human IgG of known concentration (Zymed Laboratories, San Francisco, CA). Finally, the serially diluted eluates, in which IgG titer specific for the C region was already known, and the positive control sera were applied to a plate, and IgG titers specific for the C region in the latter samples were determined. Titers of <0.07 μg/mL were considered as negative.

Measurement of antibody titers against synthetic peptides from the C region of M protein. ELISA using synthetic peptides was performed as described by Fischetti et al.(11). Briefly, 2.5 μg of synthetic peptides (50 μg/mL water solution) were added to the wells and evaporated at 37°C overnight in a desiccater under vacuum. One hundred microliters of coating buffer were added to the wells and incubated at 37°C for 2 h. The plates were washed, and the procedures carried out for ELISA against recombinant C region then were followed.

To investigate the B cell epitopes recognized by patients with RF, each peptide was reacted with sera from control (n = 5) and patients with RF (n = 5), and OD was compared between the two groups.

Statistical analysis. The IgG titer against the C region in each group was shown as a median. The reactivity between RF and control groups against each peptide was expressed as mean ± SE. Statistical significance was assessed by Wilcoxon-Mann-Whitney test and Wilcoxon signed-ranks test. A value of p < 0.05 was considered significant.

RESULTS

DNA sequence of the C region of M12 strain SS95/12. PCR amplified 627 bp covering 99% of the C region of M12 protein. Although the carboxy-terminal 35 bp before the termination codon (TAA) was not determined in another M12 strain (M12 CS24)(3), our sequence analysis revealed that the DNA sequence of this region was identical to that of M5(6).

Six nucleotide substitutions were detected in the C region of M12 strain SS95/12 compared with the reported sequence of strain CS24(3). The substitutions are listed inFigure 2. Five of six base substitutions were identical to those found in the M5 gene(6), and four of six were identical to those in the M6 and M24 genes(2, 12).

Figure 2
figure 2

Nucleotide sequence of the C region of M12 strain SS95/12 (A) and its comparison with that of M12 strain CS24(B). In A, bold numbers (1-6) show nucleotide substitutions and double lines show amino acid substitutions observed in M12 SS95/12, when compared with M12 strain CS24. C1 andC2 designate the C repeat blocks. Amino acid numbers of M12 protein, whose signal sequence was eliminated, are shown to the both sides of each line. In B, six nucleotide substitutions are noted. Of these substitutions, five were identical to those in M5 (), and four were found in M6 and M24 ().

Expression of recombinant M protein. The total yield of purified recombinant M protein from 4 L of culture was approximately 60 mg. Two bands of 65 and 63 kD were observed for the MBP-C region protein(Fig. 3a). The observed mobility was close to the expected size deduced from the DNA sequence (23 kD), when the molecular mass of MBP (42 kD) was considered.

Figure 3
figure 3

(a) Coomassie Blue-stained SDS-PAGE of MBP fusion protein of the C region. Molecular mass standard proteins(MW) are phosphorylase b (94 kD), BSA (67 kD), ovalbumin(43 kD), carbonic anhydrase (30 kD), and soybean trypsin inhibitor (20 kD).(b and c) Immunoblot analysis of recombinant proteins. Recombinant C region protein did not react with rabbit anti-AB region antibody(anti-AB), but did react with MAb against the C region of M6(anti-C). Bands were stained using 5-bromo-4-chloro-3-indolyl phosphate as the alkaline phosphatase substrate and nitro blue tetrazolium as the chromophore. The recombinant AB region protein is also shown for comparison.

Immunoblot analysis using antiserum against the AB region of M12 showed no reactivity with recombinant protein of the of C region (Fig. 3b). Conversely, a MAb against the C region of M6 revealed doublet protein bands (Fig. 3c), indicating that the antigenicity of the C region is conserved among different M serotypes(2).

IgG titers against the C region of M protein in patients with rheumatic fever. IgG titers against the C region were measured in nine patients with RF during the acute illness. The titers were remarkably higher in patients with RF than in healthy controls (median: 43 versus 1.5μg/mL, p < 0.01) (Table 1). As shown inFigure 4, there was little overlap between the titers obtained for the RF patients and those of the controls. When the upper limit of normal is defined as “the highest titer that is exceeded by only 20% of a population,” a value of greater than 4.0 μg/mL is considered to be “high.” All patients with RF had high IgG titers against M protein.

Table 1 Antibody titers against Streptococcus in 10 patients with RF
Figure 4
figure 4

IgG titers against the C region of M protein in patients with RF and age matched controls.

In five patients, IgG titers against the C region were reexamined after long-term follow-up. Although titers were decreased compared with the acute titers, all titers were ≥8 μg/mL (Table 1).

When IgG titers in patients with streptococcal pharyngitis were assessed during the acute and convalescent periods, the value during the convalescent period was significantly higher than during the acute period (median: 1.4versus 1.8 μg/mL, p < 0.05) (Fig. 5). However, both values were significantly lower than titers in patients with RF (p < 0.01).

Figure 5
figure 5

IgG titers against the C region of M protein in patients with uncomplicated streptococcal pharyngitis and age matched controls. In the patient group, IgG titers were measured during the acute and convalescent phase.

In patients with RF, IgG titers against the C region of M protein were compared with titers to ASO and ASK, measured from the same sera(Table 1). During the acute phase (n = 9), all nine patients (100%) had high IgG titers against the C region, although ASO was high (>330 IU) in six patients (67%) and ASK (>1280 IU) in four patients (44%). During long-term follow-up (n = 5), IgG titers against the C region remained elevated in all patients (100%), whereas ASO was high in only one patient (20%) and ASK was not elevated in any of the patients(0%). These data show that elevated IgG titers against the C region were maintained for several years in patients with RF, long after ASO and ASK titers had returned to normal.

Recognition of epitopes in the C region by patients with RF. Sera from five patients with RF and five controls, who had relatively high IgG titers against the C region (mean value: 55.6 ± 11.6, 8.0 ± 0.8μg/mL, respectively), were assayed for their reactivity to 13 synthetic peptides and one recombinant protein of anchor region. Higher reactivity was observed in patients with RF than in controls with peptides no. 3 (0.61± 0.13 versus 0.12 ± 0.01, p < 0.05) and no. 7 (0.44 ± 0.13 versus 0.02 ± 0.01, p< 0.05) (Fig. 6). Both peptides were located in the amino-terminal half of the C repeat blocks (C1, C2), indicating that the major epitopes in the C region of M protein that are recognized by B cells of patients with RF are encoded by the C repeat blocks (Fig. 2). When the amino acid sequence of peptide no. 3 was compared with HCMB(13), 53% identity was observed in a 15-amino acid residue of peptide no. 3 and in HCMB beginning at 1164 (Fig. 7).

Figure 6
figure 6

ELISA analysis of sera from controls and patients with RF using overlapping peptides. Sera from both groups were assessed against each peptide listed on the x axis, and their mean reactivity and standard errors in absorbance are shown. Higher reactivity of RF sera against peptides nos. 3 and 7 (asterisks) was observed. Both of these peptides are located in the amino-terminal half of the C repeat blocks. On thex axis, C terminal is a recombinant protein of the anchor region of the C region.

Figure 7
figure 7

Sequence homology between peptide no. 3 from the C region of M protein and the human cardiac myosin β-chain(15). There is 53% homology between the 15 amino acids of the C region beginning at residue 327 and human cardiac myosin residues 1164-1178 (within exon 27 of the DNA sequence).

DISCUSSION

It has been suggested that RF represents an autoimmune disease initiated as a result of cross-reactivity between the components of group A Streptococcus and myocardial tissue; for example, M protein and myocardial sarcolemma(14), protoplast membrane antigen and sarcolemmal membrane of the myocardium(15), and streptococcal carbohydrate and valvular glycoprotein(16). Of these streptococcal antigens, M protein confers to the organism the ability to resist attack by human phagocyte cells and, thus, is considered to be one of the major virulent factors for streptococci.

In previous studies, M proteins had been obtained from the streptococcal cell wall with pepsin or hot acid extraction(17). However, M protein produced by these methods is contaminated by other streptococcal cell components(18). To overcome this problem, we cloned the gene encoding the M protein and produced recombinant M protein. The recombinant C region was visualized as a doublet on an SDS-PAGE gel and by immunoblot analysis, although the larger molecular weight band was the major component. Degradation may occur during the purification process or, alternatively, M protein is susceptible to proteolysis by bacterial enzymes when expressed in E. coli(19).

DNA sequencing of the C region of M12 SS95/12 revealed 6 bp substitutions as compared with M12 CS24, demonstrating that the DNA sequence is not completely conserved among different strains of the same M serotype. However, 5 of 6 base substitutions were identical to those found in other M protein serotypes. This supports the observation by Hollingshead et al.(20) that the C region is highly conserved among different M serotypes.

Bessen et al.(4) have found that all streptococcal serotypes associated with outbreaks of acute RF (M types 1, 3, 5, 6, 12, and 24) had similar antigenicity of the C region and postulated that the C region may be a virulence determinant for RF. The current study demonstrated that IgG titers against the C region of M protein in patients with RF were significantly higher than in controls or in patients with streptococcal pharyngitis. Patients with RF maintained significantly elevated IgG titers against the C region for several years, after ASO or ASK titers had returned to normal. This finding is in line with the observation by Lancefield(21) on the persistence of type-specific antibody after group A streptococcal infection in humans.

A similar phenomenon regarding persistent elevation of antibody levels to another streptococcal antigen, group A carbohydrate, has been reported by Ayoub and co-workers(2224). They found that a RF patient with persistent carditis showed prolonged elevated levels of antibody against group A carbohydrate, whereas this antibody declined sharply to normal levels within 1 y in patients with uncomplicated streptococcosis(impetigo or pharyngitis). In our study, antibody titers against the C region were significantly higher in patients with acute RF than in patients with uncomplicated streptococcosis, whereas peak levels of antibody against group A carbohydrate were not significantly different in the two patients groups during the acute illness. However, our study is small scale and does not address the issue of whether persistent carditis would affect IgG levels against M protein during long-term follow-up. Therefore, further investigation is required.

Fischetti et al.(11) have shown that immunoreactivity to all C region peptides which they generated was weak in the sera of healthy individuals, even though they had the antibodies which reacted with whole C region protein. They speculated that the conformational structure is important for the production of antibodies against the C region. Our study also showed that there was only weak reactivity against each C region peptide in control subjects. However, patients with RF had high reactivity against peptides 3 and 7 located in the amino-terminal half of the C repeat blocks(C1, C2). We speculate that these epitopes are responsible for the high IgG titer against the whole C region of M protein in RF, although we cannot exclude the possibility that there are other B cell epitopes with conformational dependence.

We have shown a sequence homology between the peptide in the C repeat blocks and HCMB. Although it is not clear whether high concentrations of antibodies against the C region play a direct role in the initiation of RF, it is possible that antibodies against the C region play an important role in the pathogenesis of rheumatic heart disease through their cross-reactivity with myocardial HCMB.