Kawasaki syndrome (KS) is the major cause of acquired heart disease in children. Although acute myocarditis is observed in most patients with KS, its pathogenesis is unknown. Because antimyosin autoantibodies are present in autoimmune myocarditis and rheumatic carditis, the purpose of the current study was to determine whether anticardiac myosin Abs might be present during the acute stage of KS. Sera from KS patients as well as age-matched febrile controls and normal adults were compared for reactivity with human cardiac myosin in ELISAs and Western blot assays. A total of 5 of 13 KS sera, as compared with 5 of 8 acute rheumatic fever sera, contained Ab titers to human cardiac myosin that were significantly higher than those found in control sera. Both cardiac and skeletal myosins were recognized in the ELISA by KS sera, although stronger reactivity was observed to human cardiac myosin. Only IgM antimyosin Abs from KS sera were significantly elevated relative to control sera. KS sera containing antimyosin Abs recognized synthetic peptides from the light meromyosin region of the human cardiac myosin molecule and had a different pattern of reactivity than acute rheumatic fever sera, further supporting the association of antimyosin Ab with KS. These Abs may contribute to the pathogenesis of acute myocarditis found in patients with KS.

Kawasaki syndrome (KS)3 is an acute febrile illness that occurs primarily in infants and young children (1, 2). KS and rheumatic fever are the two leading causes of acquired heart disease in children (3). Myocarditis occurs in over half the patients with acute KS (4, 5). In a review of endomyocardial biopsies from 201 patients with KS obtained 1 mo to 11 years after onset of disease, Yutani et al. reported changes consistent with myocarditis (6). Manifestations may include: mild mitral regurgitation, depressed myocardial function, and increased end systolic and end diastolic dimensions, which do not correlate with the presence of coronary artery lesions (7). The myocarditis of KS is associated with clinical features of tachycardia disproportionate to the degree of fever, gallop rhythm, murmur, and electrocardiogram changes, including prolonged PR, QT intervals and diffuse low voltage, ST-T wave changes, as well as the described echocardiographic findings. Although myocarditis is a well-known feature of acute KS, little is known about its pathogenesis (7, 8). Immune activation associated with polyclonal B cell activation as well as an increased number of activated T cells and macrophages is found during the acute phase of KS (9). Acute KS is also associated with the appearance of circulating Abs that are cytotoxic against vascular endothelial cells prestimulated with IL-1β (10), TNF-α (11), or IFN-γ (12) but not against unstimulated endothelial cells. Therefore, it has been postulated that immune reactivity to vascular endothelium may contribute to the vasculitis associated with acute KS.

The etiology of immune activation in acute KS is not certain, although evidence supports an infectious etiology. Microbiologic data suggest a role for staphylococcal and streptococcal superantigens (13, 14). Furthermore, the clinical symptoms of KS share a number of features with staphylococcal and streptococcal diseases, such as toxic shock syndrome and scarlet fever. Superantigens have also been reported to induce the production of autoantibodies such as rheumatoid factor (15).

Autoantibodies against myosin have been implicated in the pathogenesis of rheumatic carditis and autoimmune myocarditis, diseases which also affect heart valve tissue and myocardium (16, 17). Furthermore, cardiac myosin is known to induce myocarditis in animal models (18). To date, there have been no studies examining the occurrence of anticardiac myosin autoantibodies in acute KS. Sera were also obtained from eight children presenting with acute rheumatic fever (ARF). The purpose of the current study was to seek evidence for antimyosin Abs in patients with acute KS.

The study was conducted on sera from 13 children in the acute phase of KS. These patients fulfilled the established clinical criteria for KS (19). These criteria included a fever for ≥5 days and at least four of the five following symptoms: nonexudative conjunctival injection; changes in the oral pharynx, including mucosal erythema, dry fissured lips, and “strawberry tongue”; changes in the extremities, characteristically erythema of the palms and soles, induration of the hands and feet, or perungual desquamation in the subacute phase of the disease; polymorphous rash; and cervical adenopathy (one or more nodes of ≥1.5 cm in diameter). A total of 11 of the 13 KS patients had clinical evidence of myocarditis during the acute phase of their illness. Sera were also obtained from 12 age-matched febrile children, 10 normal adults who had no known illness, and 14 children presenting with ARF. All ARF patients demonstrated positive anti-streptolysin O titers and fulfilled the revised Jones criteria for diagnosis of ARF. Informed consent was obtained from each patient and/or their parents before the study.

Human cardiac myosin was purified as described previously (20); the light meromyosin (LMM) peptides of human cardiac myosin were synthesized using a fluorenylmethoxycarbonyl strategy (21) and purified by HPLC. Sequences of the LMM peptides are indicated in Table I.

Table I.

Sequences of LMM peptides

Peptide Sequence Peptide Sequence 
    
LMM-1 KEALISSLTRGKLTYTQQ LMM-26 EGDLNEMEIQLSHANRMA 
LMM-2 TYTQQLEDLKRQLEEEVK LMM-27 ANRMAAEAQKQVKSLQSL 
LMM-3 EEEVKAKNALAHALQSAR LMM-28 SLQSLLKDTQIQLDDAVR 
LMM-4 LQSARHDCDLLREQYEEE LMM-29 RANDDLKENIAIVERRNN 
LMM-5 EQYEEETEAKAELQRVLSK LMM-30 IAIVERRNNLLQAELEEL 
LMM-6 RVLSKANSEVAQWRTKYE LMM-31 ELEELRAVVEQTERSRKL 
LMM-7 RTKYETDAIQRTEELEEA LMM-32 RSRKLAEQELIETSERVQ 
LMM-8 ELEEAKKKLQRLQEAEE LMM-33 SERVQLLHSQNTSLINQK 
LMM-9 QEAEEAVEAVNAKCSSLE LMM-34 LINQKKKMDADLSQLQTE 
LMM-10 CSSLEKTKHRLQNEIEDL LMM-35 TEVEEAVQESRNAEEKAKK 
LMM-11 EIEDLMVDVERSNAAAAA LMM-36 RNAEEKAKKAITDAAMMA 
LMM-12 AAAAALDKKQRNFDKILA LMM-37 AAMMAEELKKEQDTSAHL 
LMM-13 DKILAEWKQKYEESQSEL LMM-38 TSAHLERMKKNMEQTIKDL 
LMM-14 SQSELESSQKEARSLSTE LMM-39 TIKDLQHRLDEAEQIALK 
LMM-15 SLSTELFKLKNAYEESLE LMM-40 EQIALKGGKKQLQKLEARV 
LMM-16 EESLEHLETFKRENKNLQ LMM-41 LEARVRELENELEAEQKR 
LMM-17 NKNLQEEISDLTEQLGSS LMM-42 AEQKRNAESVKGMRKSER 
LMM-18 EQLGSSGKTIHELEKVRKQ LMM-43 RKSERRIKELTYQTEEDR 
LMM-19 KVRKQLEAEKMELQSALE LMM-44 TEEDRKNLLRLQDLVDKL 
LMM-20 LQSALEEAEASLEHEEG LMM-45 LVDKLQLKVKAYKRQAEE 
LMM-21 EEGKILRAQLEFNQIKAE LMM-46 RQAEEAEEQANTNLSKFR 
LMM-22 NQIKAEIERKLAEKDEEME LMM-47 LSKFRKVQHELDEAEERA 
LMM-23 DEEMEQEKRNHLRVVDSL LMM-48 AEERADIAESQVNKLRAK 
LMM-24 VVDSLQTSLDAETRSRNE LMM-49 KLRAKSRDIGTKGLNEE 
LMM-25 RSRNEALRVKKKMEGDLN   
Peptide Sequence Peptide Sequence 
    
LMM-1 KEALISSLTRGKLTYTQQ LMM-26 EGDLNEMEIQLSHANRMA 
LMM-2 TYTQQLEDLKRQLEEEVK LMM-27 ANRMAAEAQKQVKSLQSL 
LMM-3 EEEVKAKNALAHALQSAR LMM-28 SLQSLLKDTQIQLDDAVR 
LMM-4 LQSARHDCDLLREQYEEE LMM-29 RANDDLKENIAIVERRNN 
LMM-5 EQYEEETEAKAELQRVLSK LMM-30 IAIVERRNNLLQAELEEL 
LMM-6 RVLSKANSEVAQWRTKYE LMM-31 ELEELRAVVEQTERSRKL 
LMM-7 RTKYETDAIQRTEELEEA LMM-32 RSRKLAEQELIETSERVQ 
LMM-8 ELEEAKKKLQRLQEAEE LMM-33 SERVQLLHSQNTSLINQK 
LMM-9 QEAEEAVEAVNAKCSSLE LMM-34 LINQKKKMDADLSQLQTE 
LMM-10 CSSLEKTKHRLQNEIEDL LMM-35 TEVEEAVQESRNAEEKAKK 
LMM-11 EIEDLMVDVERSNAAAAA LMM-36 RNAEEKAKKAITDAAMMA 
LMM-12 AAAAALDKKQRNFDKILA LMM-37 AAMMAEELKKEQDTSAHL 
LMM-13 DKILAEWKQKYEESQSEL LMM-38 TSAHLERMKKNMEQTIKDL 
LMM-14 SQSELESSQKEARSLSTE LMM-39 TIKDLQHRLDEAEQIALK 
LMM-15 SLSTELFKLKNAYEESLE LMM-40 EQIALKGGKKQLQKLEARV 
LMM-16 EESLEHLETFKRENKNLQ LMM-41 LEARVRELENELEAEQKR 
LMM-17 NKNLQEEISDLTEQLGSS LMM-42 AEQKRNAESVKGMRKSER 
LMM-18 EQLGSSGKTIHELEKVRKQ LMM-43 RKSERRIKELTYQTEEDR 
LMM-19 KVRKQLEAEKMELQSALE LMM-44 TEEDRKNLLRLQDLVDKL 
LMM-20 LQSALEEAEASLEHEEG LMM-45 LVDKLQLKVKAYKRQAEE 
LMM-21 EEGKILRAQLEFNQIKAE LMM-46 RQAEEAEEQANTNLSKFR 
LMM-22 NQIKAEIERKLAEKDEEME LMM-47 LSKFRKVQHELDEAEERA 
LMM-23 DEEMEQEKRNHLRVVDSL LMM-48 AEERADIAESQVNKLRAK 
LMM-24 VVDSLQTSLDAETRSRNE LMM-49 KLRAKSRDIGTKGLNEE 
LMM-25 RSRNEALRVKKKMEGDLN   

Purified rabbit skeletal myosin was obtained from Sigma (St. Louis, MO).

Microtiter plates were coated at 4°C overnight with purified human cardiac myosin or LMM peptides at 10 μg/ml in 0.1 M carbonate-bicarbonate coating buffer (pH 9.6). The plates were washed with PBS Tween 20 and blocked with 1% BSA for 1 h at 37°C. The plates were washed again with PBS Tween 20, and 50-μl serum dilutions beginning at 1/250 were added to the microtiter wells. Sera were diluted 2-fold in PBS. Serum dilutions were incubated overnight with Ag at 4°C, and all tests were performed in duplicate. The plates were washed again with PBS Tween 20, and goat anti-human IgM (μ-chain-specific), IgG (γ-chain-specific), and IgA (α-chain-specific) conjugated with alkaline phosphatase (1/250 dilution) was added in 50 μl amounts to the microtiter plates and incubated at 37°C for 1 h. Conjugated Igs were obtained from Sigma. Next, the plates were washed with PBS Tween 20, and 50 μl of the substrate para-phenyl-phosphate (Sigma 104) in diethanolamine buffer was added to the wells. OD was measured at 410 nm in an ELISA plate reader (Dynatech, Chantilly, VA). Titers were calculated from the 0.30-OD endpoint in the ELISA.

Western blots were performed as described previously (17). Briefly, 8 μg of purified human cardiac myosin was loaded into wells of an SDS polyacrylamide gel and electrophoresed until the tracking dye reached the bottom of the gel. The gel was blotted onto a nitrocellulose membrane using a blotting apparatus (Bio-Rad, Hercules, CA). The blot was blocked with 3% milk for 1 h at 37°C. Sera from control subjects and KS patients were diluted 1/1000 in PBS. A section of the blot was used to stain the cardiac myosin heavy chain with a molecular mass of 200 kDa. Nitrocellulose strips containing the human cardiac myosin protein band were cut and reacted separately with each of the patient sera and with rat anti-human cardiac myosin sera at 1/1000. The control sera recognized the 200-kDa myosin heavy chain. The strips were washed five times with PBS Tween 20 and reacted with the secondary Ab conjugate goat anti-human polyvalent Igs conjugated with HRP. The blot was developed using chloronaphthol as indicator with hydrogen peroxide as substrate as described previously (17). The serum reaction with the human cardiac myosin heavy chain was scored as 0–4+ as compared with a 4+ reaction by the positive control anti-rat human cardiac myosin and a PBS-conjugated secondary Ab control that was negative or 0 reactivity.

A primary rat heart cell line (no. CRL-1446, American Type Culture Collection, Manassas, VA) was plated into 96-well tissue culture plates at 1 × 104 cells/well in IMDM with 20% FBS. The heart cells were incubated at 37°C and in 5% CO2. A total of 5 μCi of 51Cr was added per well and incubated at 37°C for 2 h. The cells were washed in the IMDM containing 20% FBS, and 100 μl of IMDM plus 20% FBS was added to each well and incubated at 37°C for 1 h. Sera from patients with KS and control sera were diluted 1/5 in IMDM. Plates were washed with serum-free IMDM three times, and 100 μl of diluted serum was added to wells in triplicate. Negative control wells received 100 μl of IMDM; maximum release wells received 100 μl of 1N HCL. Positive control wells received human and mouse mAbs that were cytotoxic for the heart cell line. Once the test sera were added, plates were incubated at 37°C for 45 min and 100 μl of freshly prepared guinea pig complement was added. After a 1-h incubation at 37°C, the supernatants were harvested by a Skatron harvester (Lier, Norway) and counted on an LKB compugamma counter (Uppsala, Sweden). The formula used to calculated percent lysis is as follows: ([test cpm − spontaneous release cpm]/[maximum release cpm − spontaneous release cpm]) × 100. Spontaneous release was determined as cpm from the IMDM negative control samples without complement.

Means with SDs were calculated for the KS and control sera groups; the means of the sera groups were compared by the unpaired Student’s t test to determine significance, which was calculated as two-tailed p values.

To investigate the possibility that immune responses to cardiac myosin were elevated in KS, sera from 13 patients with KS, 12 age-matched febrile controls, and 10 normal adults were reacted with purified human cardiac myosin by ELISA. Fig. 1 illustrates the stronger reactivity of a group of KS sera with human cardiac myosin in the ELISA as compared with sera from age-matched controls and normal adults. The figure shows a scattergram of individual serum titers when reacted with human cardiac myosin as Ag. Greater than 50% of KS sera had titers greater than the mean antimyosin titers in febrile control patients (Fig. 1). However, only 5 of 13 KS sera tested had higher titers. Titers of normal sera ranged from 500 to 8,000, whereas Ab titers against human cardiac myosin in KS patients ranged from 1000 to 64,000 (Fig. 1). The KS titers to human cardiac myosin as a group were significantly different from age-matched febrile controls (p = 0.047) and normal adult controls (p = 0.04). Only antimyosin IgM Ab titers are shown, because there was little or no difference observed between KS and age-matched control sera in their IgG reactivity to cardiac or skeletal myosins (data not shown). No antimyosin IgA Abs were detected in any of the sera. KS sera, some of which had high titers to human cardiac myosin, also reacted with skeletal myosin but consistently with lower titers. Thus, there is a group of KS patients that responds strongly to human cardiac myosin.

FIGURE 1.

Scattergram of anti-human cardiac myosin Ab titers of sera from acute KS patients and age-matched febrile controls as well as normal adult sera. Endpoints were calculated at an OD of 0.3. Means and p values were calculated using the unpaired Student’s t test. There was a significant difference between the means of the KS and control groups (p < 0.05).

FIGURE 1.

Scattergram of anti-human cardiac myosin Ab titers of sera from acute KS patients and age-matched febrile controls as well as normal adult sera. Endpoints were calculated at an OD of 0.3. Means and p values were calculated using the unpaired Student’s t test. There was a significant difference between the means of the KS and control groups (p < 0.05).

Close modal

For comparison, antimyosin Ab ELISA titers of eight ARF sera are shown in Fig. 2. Five of eight sera demonstrated significantly elevated titers over two controls that had titers of 2000 and 8000. The individual ARF sera titers are, from left to right in Fig. 2, 512,000; 32,000; 16,000; 8,000; 32,000; 32,000; 64,000; and 8,000. ARF sera have been shown previously to react in the Western immunoblot with human cardiac myosin (16, 22)

FIGURE 2.

. Reactivity of 8 ARF sera and 2 normal control sera with human cardiac myosin. The ELISA antimyosin Ab titers as shown in the figure are, from left to right, 512,000; 32,000; 16,000; 8,000; 32,000; 32,000; 64,000; 8,000; and 8,000 for ARF sera and 8,000 and 2,000 for control (Cont) sera.

FIGURE 2.

. Reactivity of 8 ARF sera and 2 normal control sera with human cardiac myosin. The ELISA antimyosin Ab titers as shown in the figure are, from left to right, 512,000; 32,000; 16,000; 8,000; 32,000; 32,000; 64,000; 8,000; and 8,000 for ARF sera and 8,000 and 2,000 for control (Cont) sera.

Close modal

To further demonstrate the antimyosin specificity of the KS sera, the sera were reacted with purified human cardiac myosin (200 kDa) in a Western immunoblot. Table II shows the reactivity of each of the sera with the cardiac myosin heavy chain as compared with strongly positive control rat anti-human cardiac myosin polyclonal sera (4+). The KS sera reactivity (1/1000 dilution) ranged from 2+ to 3+, and the control sera reactivity was negative. The Western blot is shown in Fig. 3. The KS sera were strongly positive in either the blot or the ELISA or both. Further analysis of four additional KS sera demonstrated that they were all 2+ myosin blot positive. Western blot results did not always correlate with the ELISA titers of the KS or control sera (Table II). The control sera were negative in the blot in which serum titers of 8000 were not positive in the blot. The Western blot (Fig. 3) indicated the specificity of the KS sera for human cardiac myosin, whereas the normal sera at a 1/1000 dilution were negative. All of the normal sera tested in the blot had detectable antimyosin titers (1000, 1000, 1000, and 8000) in the ELISA (Table II) but were not reactive in the Western blot. These data suggest that the reactivity of the KS sera was highly specific for human cardiac myosin heavy chain, but that the reactivity of the normal control sera in the ELISA and not in the Western blot was due to 1) recognition of highly conformational epitopes, 2) cross-reactivity, or 3) background in the test system.

Table II.

Reactivity of KS sera with human cardiac myosin heavy chain in the Western immunoblot

KS SeraReactivityControl SeraReactivity
BlotELISA titerBlotELISA titer
KS-1 3+ 32,000 C-1 — 1000 
KS-2 2+ 32,000 C-2 — 1000 
KS-3 2+ 64,000 C-3 — 1000 
KS-4 2+ 8,000 C-4 — 8000 
KS-5 3+ 1,000 Anticardiac myosin sera (+ control) 4+  
KS-6 2+ 2,000    
KS-7 2+ 1,000    
KS-8 3+ 8,000    
KS SeraReactivityControl SeraReactivity
BlotELISA titerBlotELISA titer
KS-1 3+ 32,000 C-1 — 1000 
KS-2 2+ 32,000 C-2 — 1000 
KS-3 2+ 64,000 C-3 — 1000 
KS-4 2+ 8,000 C-4 — 8000 
KS-5 3+ 1,000 Anticardiac myosin sera (+ control) 4+  
KS-6 2+ 2,000    
KS-7 2+ 1,000    
KS-8 3+ 8,000    
FIGURE 3.

Reactivity of KS and control sera with human cardiac myosin in the Western immunoblot. Sera were tested at a 1/1000 dilution.

FIGURE 3.

Reactivity of KS and control sera with human cardiac myosin in the Western immunoblot. Sera were tested at a 1/1000 dilution.

Close modal

To determine whether the KS sera that were highly reactive with human cardiac myosin would recognize peptide epitopes in the human cardiac myosin heavy chain, eight KS sera with elevated titers (titers = 64,000; 32,000; 32,000; 32,000; 16,000; 8,000; 8,000; and 2,000) to myosin were pooled and reacted with 49 LMM peptides in the ELISA. In addition, febrile age-matched control sera (titers = 1000; 1000; 1000; and 8000) were also pooled and reacted with the LMM peptides. The reactivity of the KS sera group with the LMM peptides is shown in Fig. 4,A. LMM peptides 1, 18, 32, 34, 47, and 49 were the most reactive with KS sera after normal age-matched control sera reactivity was subtracted to obtain the final reactivity of the KS sera shown in Fig. 4,A. The reactivity of the control sera ranged from an OD of 0.2 to 0.5 and was subtracted. These data are compared with the reactivity of a group of eight ARF sera reacted with the LMM peptides (Fig. 4,B). The reactivities of the KS and ARF sera with the LMM peptides have similarities, but are also clearly different. The greatest differences are observed for LMM peptides 13 and 25, which are positive only in ARF, whereas peptide 47 was twice as reactive with KS sera. Because results using pooled sera may not reflect the individual serum reactivities, we tested nine KS and six ARF sera individually. Table III shows the results of these tests. Table III compares the reactivity of both KS and ARF sera with the LMM peptide panel. As underlined in the Table III, peptides positive with three or more sera in each group were underlined to highlight the reactivities of multiple sera with a particular peptide. The KS sera reacted with LMM-1, -4, -7, -16, -18, -32, and -43, whereas the ARF sera reactive with LMM-6, -7, -18, -19, -25, -29, -32, -36, and -47. However, normal sera did not react positively with any of the LMM peptides as shown in Table III. Comparison of data from pooled KS sera in Fig. 4,A with the individual serum study indicated that the peptides identified on the pooled sera were similar to those identified in individual sera. Comparison of the ARF sera reacted with LMM peptides (Table III) and with the pooled sera (Fig. 4,B) suggested that the data were quite similar. Some of the other LMM peptides recognized by two of six ARF sera and not considered mainstream were LMM-1, -4, -8, -16, -22, -28, -34, -35, -39, -40, -43, -45, and -49, which were also seen in Fig. 4 B. The conclusion from these data is that both KS and ARF sera recognize different but some overlapping epitopes. In individual and pooled serum studies, LMM-6, -19, and -25 were unique to or seen more often in ARF than in KS. The LMM peptides seen frequently in both diseases were LMM-7, -18, -32, and -47.

FIGURE 4.

A, Reactivity of KS sera with synthetic peptides of human cardiac LMM in the ELISA. A pool of eight anti-human cardiac myosin-positive KS sera and a pool of four age-matched control sera were reacted with the 49 synthetic LMM peptides. Sera were tested at a 1/250 dilution, and control serum reactivity with LMM peptides ranged from 0.2 to 0.5. Control OD (∼0.2–0.5) was subtracted from the KS sera reactivity with each individual LMM peptide. B, Reactivity of a pool of eight ARF sera with synthetic LMM peptides under the same conditions as described in Fig. 3 A.

FIGURE 4.

A, Reactivity of KS sera with synthetic peptides of human cardiac LMM in the ELISA. A pool of eight anti-human cardiac myosin-positive KS sera and a pool of four age-matched control sera were reacted with the 49 synthetic LMM peptides. Sera were tested at a 1/250 dilution, and control serum reactivity with LMM peptides ranged from 0.2 to 0.5. Control OD (∼0.2–0.5) was subtracted from the KS sera reactivity with each individual LMM peptide. B, Reactivity of a pool of eight ARF sera with synthetic LMM peptides under the same conditions as described in Fig. 3 A.

Close modal
Table III.

Comparison of KS with ARF in reactivity of sera with peptides of human cardiac LMMa

PeptideNormal SerumKSARF
LMM-1 0 /3 3 /9 2 /6 
LMM-2 0 /3 1 /9 1 /6 
LMM-3 0 /3 1 /9 1 /6 
LMM-4 0 /3 4 /9 2 /6 
LMM-5 0 /3 0 /9 1 /6 
LMM-6 0 /3 1 /9 3 /6 
LMM-7 0 /3 4 /9 5 /6 
LMM-8 0 /3 1 /9 2 /6 
LMM-9 0 /3 0 /9 1 /6 
LMM-10 0 /3 1 /9 1 /6 
LMM-11 0 /3 1 /9 0 /6 
LMM-12 0 /3 1 /9 1 /6 
LMM-13 0 /3 1 /9 1 /6 
LMM-14 0 /3 0 /9 1 /6 
LMM-15 0 /3 0 /9 1 /6 
LMM-16 0 /3 3 /9 2 /6 
LMM-17 0 /3 1 /9 1 /6 
LMM-18 0 /3 6 /9 3 /6 
LMM-19 0 /3 1 /9 3 /6 
LMM-20 0 /3 0 /9 1 /6 
LMM-21 0 /3 0 /9 1 /6 
LMM-22 0 /3 0 /9 2 /6 
LMM-23 0 /3 0 /9 1 /6 
LMM-24 0 /3 1 /9 1 /6 
LMM-25 0 /3 1 /9 3 /6 
LMM-26 0 /3 0 /9 1 /6 
LMM-27 0 /3 0 /9 1 /6 
LMM-28 0 /3 0 /9 2 /6 
LMM-29 0 /3 0 /9 3 /6 
LMM-30 0 /3 0 /9 1 /6 
LMM-31 0 /3 0 /9 1 /6 
LMM-32 0 /3 3 /9 3 /6 
LMM-33 0 /3 0 /9 1 /6 
LMM-34 0 /3 0 /9 2 /6 
LMM-35 0 /3 0 /9 2 /6 
LMM-36 0 /3 2 /9 3 /6 
LMM-37 0 /3 1 /9 1 /6 
LMM-38 0 /3 0 /9 0 /6 
LMM-39 0 /3 2 /9 2 /6 
LMM-40 0 /3 2 /9 2 /6 
LMM-41 0 /3 0 /9 1 /6 
LMM-42 0 /3 0 /9 1 /6 
LMM-43 0 /3 3 /9 2 /6 
LMM-44 0 /3 1 /9 0 /6 
LMM-45 0 /3 3 /9 2 /6 
LMM-46 0 /3 0 /9 1 /6 
LMM-47 0 /3 5 /9 5 /6 
LMM-48 0 /3 0 /9 1 /6 
LMM-49 0 /3 3 /9 2 /6 
PeptideNormal SerumKSARF
LMM-1 0 /3 3 /9 2 /6 
LMM-2 0 /3 1 /9 1 /6 
LMM-3 0 /3 1 /9 1 /6 
LMM-4 0 /3 4 /9 2 /6 
LMM-5 0 /3 0 /9 1 /6 
LMM-6 0 /3 1 /9 3 /6 
LMM-7 0 /3 4 /9 5 /6 
LMM-8 0 /3 1 /9 2 /6 
LMM-9 0 /3 0 /9 1 /6 
LMM-10 0 /3 1 /9 1 /6 
LMM-11 0 /3 1 /9 0 /6 
LMM-12 0 /3 1 /9 1 /6 
LMM-13 0 /3 1 /9 1 /6 
LMM-14 0 /3 0 /9 1 /6 
LMM-15 0 /3 0 /9 1 /6 
LMM-16 0 /3 3 /9 2 /6 
LMM-17 0 /3 1 /9 1 /6 
LMM-18 0 /3 6 /9 3 /6 
LMM-19 0 /3 1 /9 3 /6 
LMM-20 0 /3 0 /9 1 /6 
LMM-21 0 /3 0 /9 1 /6 
LMM-22 0 /3 0 /9 2 /6 
LMM-23 0 /3 0 /9 1 /6 
LMM-24 0 /3 1 /9 1 /6 
LMM-25 0 /3 1 /9 3 /6 
LMM-26 0 /3 0 /9 1 /6 
LMM-27 0 /3 0 /9 1 /6 
LMM-28 0 /3 0 /9 2 /6 
LMM-29 0 /3 0 /9 3 /6 
LMM-30 0 /3 0 /9 1 /6 
LMM-31 0 /3 0 /9 1 /6 
LMM-32 0 /3 3 /9 3 /6 
LMM-33 0 /3 0 /9 1 /6 
LMM-34 0 /3 0 /9 2 /6 
LMM-35 0 /3 0 /9 2 /6 
LMM-36 0 /3 2 /9 3 /6 
LMM-37 0 /3 1 /9 1 /6 
LMM-38 0 /3 0 /9 0 /6 
LMM-39 0 /3 2 /9 2 /6 
LMM-40 0 /3 2 /9 2 /6 
LMM-41 0 /3 0 /9 1 /6 
LMM-42 0 /3 0 /9 1 /6 
LMM-43 0 /3 3 /9 2 /6 
LMM-44 0 /3 1 /9 0 /6 
LMM-45 0 /3 3 /9 2 /6 
LMM-46 0 /3 0 /9 1 /6 
LMM-47 0 /3 5 /9 5 /6 
LMM-48 0 /3 0 /9 1 /6 
LMM-49 0 /3 3 /9 2 /6 
a

Positive sera were >0.4 OD at 410 nm; negative sera were <0.4 OD at 410 nm. Sera were tested by ELISA at a 1250 dilution.

To further characterize the sera containing antimyosin Abs, the sera were tested for cytotoxicity against a rat heart cell line. KS-8 (ELISA titer = 8000) and KS-11 (ELISA titer = 32,000 and myosin blot 2+) sera produced 31% and 63% cytotoxicity, respectively. The cytotoxicity of normal sera ranged from 2 to 14%. These results were highly reproducible and did not necessarily correlate with anti-human cardiac myosin titers. However, the two patients with cytotoxic Abs had titers above the normal mean (Fig. 1). The strong cytotoxicity of KS-11 was particularly interesting in that this patient was documented to have features of myocarditis clinically at the time the acute sample was obtained. In addition, this patient recognized many of the LMM peptides as shown in Table III (one of nine for many LMM peptides not recognized by the other eight KS sera). Features for this patient included a gallop rhythm, a murmur tricuspid regurgitation, and sinus tachycardia. In addition, the patient had clinical features of congestive heart failure and required diuretic therapy. Clinical laboratory evidence included electrocardiogram changes of low voltage and T wave inversion that resolved in convalescence. The initial two echocardiograms demonstrated biventricular dysfunction, pericardial effusion, and normal coronary arteries. Subsequent echocardiograms demonstrated the development of diffuse aneurysms of the left main, left anterior descending, and right coronary artery. The myocardial function of this patient improved and returned to normal, as did her clinical features of myocarditis. She developed worsening ischemic heart disease over the next 6 mo. The patient went on to cardiac transplantation, and the explanted heart had evidence of coronary stenosis and patchy vasculitis as well as myocyte damage consistent with previous features of myocarditis and ischemic damage. This patient died 9 mo posttransplant after multiple severe episodes of acute rejection. To our knowlege, she is the only reported death posttransplant for a KS patient (23).

The current study was performed to gain further insight into potential mechanisms of acute myocarditis in KS. Our results indicate that, by ELISA, acute KS is associated with an abnormally high level of circulating anti-human cardiac myosin. Western blot analysis confirmed the reactivity of acute KS sera to purified human cardiac myosin. Interestingly, Ab reactivity to cardiac myosin in acute KS sera was localized to the IgM fraction and was not present in the IgG fraction. This contrasts with rheumatic carditis, which is associated with both IgG and IgM antimyosin reactivity (20).

Myosin-reactive sera from acute KS patients were also examined for reactivity with a panel of 49 overlapping peptides that span the human LMM subfragment of cardiac myosin. Sera from acute KS patients had a different pattern of reactivity than sera from ARF patients, although some LMM peptides were recognized by both diseases. These data suggest that acute KS is associated with a specific Ab response to certain epitopes of human cardiac LMM. Strong LMM reactivity by a serum reflected a pattern of autoimmune heart disease (either rheumatic heart disease in ARF or myocarditis in KS). This LMM reactivity may indicate epitope spreading throughout the myosin molecule. Epitope spreading within and among autoantigens has been suggested in other autoimmune states (22). It is possible that the more peptides recognized, the greater the risk of developing pathogenic autoantibodies reactive not only with epitopes within the myosin molecule but with cross-reactive epitopes present at the cell surface extracellular matrix, or basement membrane. The presence of cytotoxic Ab in some of the KS sera suggests that this possibility exists. As for the diagnostic potential of the antimyosin or anti-LMM Abs in KS or ARF, we believe that they at least represent a relative risk factor and could be important in identifying patients predisposed to autoimmune heart disease.

Antimyosin Abs have been associated with a number of autoimmune states, including rheumatic fever and chronic autoimmune myocarditis (16, 18, 20, 24, 25, 26, 27). It is not clear what role these antimyosin Abs play in the pathogenesis of disease. However, it is known that antimyosin Abs deposit in the heart of susceptible animals that develop myocarditis (26), and antimyosin/antistreptococcal mAbs are cytotoxic for heart cells in culture (28, 29). In addition, antimyosin Abs recognize bacterial Ags such as the group A streptococcal M protein and N-acetyl-glucosamine, the dominant epitope of the group A streptococcal carbohydrate (17, 20, 24, 29, 30, 31, 32) as well as viral Ags such as the Coxsackie viral capsid proteins (29, 33). Although antimyosin Ab can be present in the sera of normal individuals (34), it is seen usually at low levels compared with disease (16).

Although depressed myocardial function and wall motion abnormalities may be a manifestation of coronary injury and ischemia, it is unlikely that this is the sole mechanism; only 20% of untreated KS patients develop coronary artery lesions, yet over half have evidence clinically for myocarditis. In addition, the myocardial depression seen is reversible with i.v. γ-globulin in the absence of coronary artery involvement and does not correlate with the presence of coronary artery lesions (7). Cytokines such as TNF-α and IL-6 have both been reported to depress myocardial function (35, 36). They as well as other cytokines are known to be elevated in acute KS and could also contribute to the acute depressed myocardial function.

The mechanism by which anticardiac myosin may contribute to acute myocarditis in KS is not known. Preliminary studies suggest that some of these sera may cause myocardial injury via their cytotoxic activity against heart cells. Cytotoxic Abs against vascular endothelium have been found previously in KS sera, and antimyosin mAbs that recognize cell surface epitopes or Ags such as laminin are cytotoxic for the rat heart cell line used in this study (24). However, currently there is no definitive explanation for the cytotoxicity of KS sera. Further studies are required to determine whether antimyosin Abs are able to depress myocardial function without causing frank cytolysis of cells. Nevertheless, our current studies open up a new avenue for investigation into the mechanisms of myocarditis in KS, one of the most common causes of acquired heart disease in childhood.

We thank Dr. Mark Hemric for purification of the human cardiac myosin, Dr. Kenneth Jackson and the W.K. Warren Molecular Biology Resource Facility at the University of Oklahoma Health Sciences Center for synthesis and purification of the LMM peptides, and Maureen Sandoval for her assistance in the preparation of this manuscript.

1

This work was supported by Grants HL35280 and HL56267 (to M.W.C.) and Grants HL37260 and AR41256 (to D.Y.M.L.) from the National Institutes of Health.

2

Address correspondence and reprint requests to Dr. Donald Y. M. Leung, Department of Pediatrics, National Jewish Medical and Research Center, 1400 Jackson Street, Room K926, Denver, CO 80206. E-mail address: leungd@njc.org

3

Abbreviations used in this paper: KS, Kawasaki syndrome; ARF, acute rheumatic fever; LMM, light meromyosin.

1
Kawasaki, T..
1967
. Acute febrile mucocutaneous syndrome with lymphoid involvement with specific desquamation of the fingers and toes in children: clinical observations of 50 cases.
Jpn. J. Allergol.
16
:
178
2
Meissner, H. C., D. Y. M. Leung.
1995
. Kawasaki syndrome.
Curr. Opin. Rheumatol.
7
:
455
3
Taubert, K., A. Rowley, S. Schulman.
1991
. Nationwide survey of Kawasaki disease and acute rheumatic fever.
J. Pediatr.
119
:
279
4
Hiraishi, S., K. Yashiro, K. Oguchi, S. Kusano, K. Ishii, K. Nakazawa.
1981
. Clinical course of cardiovascular involvement in the mucocutaneous lymph node syndrome: relation between clinical signs of carditis and development of coronary arterial aneurysm.
Am. J. Cardiol.
47
:
323
5
Matsuura, H., T. Ishikita, S. Yamamoto, T. Umezawa, R. Ito, R. Hashiguchi, T. Saji, N. Matsuo, M. Takano.
1987
. Gallium-67 myocardial imaging for the detection of myocarditis in the acute phase of Kawasaki disease (mucocutaneous lymph node syndrome): the usefulness of single photon emission computed tomography.
Br. Heart J.
58
:
385
6
Yutani, C., K. Okano, T. Kamiya, K. Oguchi, T. Kozuka, M. Ota, S. Onishi.
1980
. Histopathological study on right endomyocardial biopsy of Kawasaki disease.
Br. Heart J.
43
:
589
7
Newburger, J. W., S. P. Sanders, J. C. Burns, I. A. Parness, A. S. Beiser, S. D. Colan.
1989
. Left ventricular contractility and function in Kawasaki syndrome: effect of intravenous γ-globulin.
Circulation
79
:
1237
8
Yoshibayashi, M., N. Tamaki, K. Nishioka, M. Matsumura, T. Ueda, S. Temma, Y. Yonekura, K. Yamashita, J. Konishi, H. Mikawa.
1991
. Regional myocardial perfusion and metabolism assessed by positron emission tomography in children with Kawasaki disease and significance of abnormal Q waves and their disappearance.
Am. J. Cardiol.
68
:
1638
9
Leung, D. Y. M., H. C. Meissner, P. M. Schlievert.
1997
. The etiology and pathogenesis of Kawasaki disease: how close are we to an answer?.
Curr. Opin. Infect. Dis.
10
:
226
10
Leung, D. Y. M., R. S. Geha, J. W. Newburger, J. C. Burns, W. Fiers, L. A. Lapierre, J. S. Pober.
1986
. Two monokines, interleukin 1 and tumor necrosis factor, render cultured vascular endothelial cells susceptible to lysis by antibodies circulating during Kawasaki Syndrome.
J. Exp. Med.
164
:
1958
11
Leung, D. Y. M., J. L. Moake, P. L. Havens, M. Kim, J. S. Pober.
1988
. Lytic anti-endothelial cell antibodies in haemolytic-uraemic syndrome.
Lancet
2
:
183
12
Leung, D. Y. M., T. Collins, L. A. Lapierre, R. S. Geha, J. S. Pober.
1986
. Immunoglobulin M antibodies present in the acute phase of Kawasaki syndrome lyse cultured vascular endothelial cells stimulated by γ interferon.
J. Clin. Invest.
77
:
1428
13
Leung, D. Y. M., H. C. Meissner, D. R. Fulton, D. L. Murray, B. L. Kotzin, P. M. Schlievert.
1993
. Toxic shock syndrome toxin-secreting Staphylococcus aureus in Kawasaki syndrome.
Lancet
342
:
1385
14
Leung, D. Y. M., K. E. Sullivan, T. F. Brown-Whitehorn, A. P. Fehringer, S. Allen, T. H. Finkel, R. L. Washington, R. Makida, P. M. Schlievert.
1997
. Association of toxic shock syndrome toxin-secreting and exfoliative toxin-secreting Staphylococcus aureus with Kawasaki syndrome complicated by coronary artery disease.
Pediatr. Res.
42
:
268
15
He, X., J. Goronzy, C. Weyand.
1992
. Selective induction of rheumatoid factors by superantigens and human helper T cells.
J. Clin. Invest.
89
:
673
16
Cunningham, M. W., J. M. McCormack, L. R. Talaber, J. B. Harley, E. M. Ayoub, R. S. Muneer, L. T. Chun, D. V. Reddy.
1988
. Human monoclonal antibodies reactive with antigens of the group A Streptococcus and human heart.
J. Immunol.
141
:
2760
17
Cunningham, M. W., R. A. Swerlick.
1986
. Polyspecificity of antistreptococcal murine monoclonal antibodies and their implications in autoimmunity.
J. Exp. Med.
164
:
998
18
Neu, N., N. R. Rose, K. W. Beisel, A. Herskowitz, G. Gurri-Glass, S. W. Craig.
1987
. Cardiac myosin induces myocarditis in genetically predisposed mice.
J. Immunol.
139
:
3630
19
American Heart Association Committee on Rheumatic Fever, Endocarditis, and Kawasaki Disease. Diagnostic Guidelines for Kawasaki Disease.
J. Am. Med. Assoc.
44
:
1990
1218
20
Adderson, E. E., A. R. Shikhman, K. E. Ward, M. W. Cunningham.
1998
. Molecular analysis of polyreactive monoclonal antibodies from rheumatic carditis: human anti-N-acetyl-glucosamine/anti-myosin antibody V region genes.
J. Immunol.
161
:
2020
21
Carpino, L. A., G. Y. Han.
1972
. The 9-fluorenylmethoxy-carbonyl amino protecting group.
J. Org. Chem.
37
:
3404
22
Lehmann, P. V., T. Forsthuber, A. Miller, E. E. Sercarz.
1992
. Spreading of T-cell autoimmunity to cryptic determinants of an autoantigen.
Nature
358
:
155
23
Checchia, P. A., E. Pahl, R. E. Shaddy, S. T. Shulman.
1997
. Cardiac transplantation for Kawasaki disease.
Pediatrics
100
:
695
24
Cunningham, M. W., J. M. McCormack, P. G. Fenderson, M. K. Ho, E. H. Beachey, J. B. Dale.
1989
. Human and murine antibodies cross-reactive with streptococcal M protein and myosin recognize the sequence GLN-LYS-SER-LYS-GLN in M protein.
J. Immunol.
143
:
2677
25
McCormack, J. M., C. A. Crossley, E. M. Ayoub, J. B. Harley, M. W. Cunningham.
1993
. Poststreptococcal anti-myosin antibody idiotype associated with systemic lupus erythematosus and Sjögren’s syndrome.
J. Infect. Dis.
168
:
915
26
Liao, L., R. Sindhwani, M. Rojkind, S. Factor, L. Leinwand, B. Diamond.
1995
. Antibody-mediated autoimmune myocarditis depends on genetically determined target organ sensitivity.
J. Exp. Med.
187
:
1123
27
Neumann, D. A., N. R. Rose, A. A. Ansari, A. Herskowitz.
1994
. Induction of multiple heart autoantibodies in mice with Coxsackievirus B3- and cardiac myosin-induced autoimmune myocarditis.
J. Immunol.
152
:
343
28
Antone, S. M., E. E. Adderson, N. M. Mertens, M. W. Cunningham.
1997
. Molecular analysis of V gene sequences encoding cytotoxic anti-streptococcal/anti-myosin monoclonal antibody 36. 2.2 that recognizes the heart cell surface protein laminin.
J. Immunol.
159
:
5422
29
Cunningham, M. W., S. M. Antone, J. M. Gulizia, B. M. McManus, V. A. Fischetti, C. J. Gauntt.
1992
. Cytotoxic and viral neutralizing antibodies crossreact with streptococcal M protein, enteroviruses, and human cardiac myosin.
Proc. Natl. Acad. Sci. USA
89
:
1320
30
Shikhman, A. R., M. W. Cunningham.
1994
. Immunological mimicry between N-acetyl-β-d-glucosamine and cytokeratin peptides: evidence for a microbially driven anti-keratin antibody response.
J. Immunol.
152
:
4375
31
Shikhman, A. R., N. S. Greenspan, M. W. Cunningham.
1993
. A subset of mouse monoclonal antibodies cross-reactive with cytoskeletal proteins and group A streptococcal M proteins recognizes N-acetyl-β-d-glucosamine.
J. Immunol.
151
:
3902
32
Shikhman, A. R., N. S. Greenspan, M. W. Cunningham.
1994
. Cytokeratin peptide SFGSGFGGGY mimics N-acetyl-β-d-glucosamine in reaction with antibodies and lectins, and induces in vivo anti-carbohydrate antibody response.
J. Immunol.
153
:
5593
33
Gauntt, C., H. Arizpe, A. Higdon, D. Bowers, M. Rozek, R. Crawley.
1995
. Molecular mimicry, anti-Coxsackievirus B2 neutralizing monoclonal antibodies and myocarditis.
J. Immunol.
154
:
2983
34
Dighiero, G., B. Guilbert, S. Avrameas.
1982
. Naturally occurring antibodies against nine common antigens in human sera: high incidence of monoclonal Ig exhibiting antibody activity against actin and tubulin and sharing antibody specificities with natural antibodies.
J. Immunol.
128
:
2778
35
Torre-Amione, G., S. Kapadia, J. Lee, R. D. Bies, R. Lebovitz, D. L. Mann.
1995
. Expression and functional significance of tumor necrosis factor receptors in human myocardium.
Circulation
92
:
1487
36
Tsutamoto, T., T. Hisanaga, A. Wada, K. Maeda, M. Ohnishi, D. Fukai, N. Mabuchi, M. Sawaki, M. Kinoshita.
1998
. Interleukin-6 spillover in the peripheral circulation increases with the severity of heart failure, and the high plasma level of interleukin-6 is an important prognostic predictor in patients with congestive heart failure.
J. Am. Coll. Cardiol.
31
:
391