The pro-inflammatory cytokines TNFa and LTa have been implicated in the pathogenesis of acute KD. Polymorphisms within the promoter region of the TNFa gene (G-->A, position -308) and within the first intron of the LTa gcnc (A-->G, position +249) are associated with increased levels of cytokine production. An increased prevalence of these high-responder alleles has been associated with poor outcome after certain infections and with increased susceptibility to autoimmune diseases. To determine if these high-responder alleles were more common in Caucasians with KD than controls, we determined the genotypic frequencies at these two loci by PCR amplification and restriction enzyme digestion. These frequencies were compared to the mean values for published Caucasian and Japanese controls.
Caucasian KD (n=39)
0.87 (-308 G/G); 0.13 (-308 G/A); 0.00 (-308 A/A); 0.56 (+249 A/A); 0.39 (+249 A/G); 0.05 (+249 G/G)
Caucasian control (5 studies)
0.69 (-308 G/G); 0.28 (-308 G/A); 0.03 (-308 A/A); 0.45 (+249 A/A); 0.44 (+249 A/G); 0.11 (+249 G/G)
Japanese control (2 studies)
0.97 (-308 G/G); 0.03 (-308 G/A); 0.00 (-308 A/A); 0.47 (+249 A/A); 0.42 (+249 A/G); 0.11 (+249 G/G) LTa genotypic frequencies were similar among the three groups. At the TNFa -308 locus, Caucasian KD patients and Japanese controls had a marked over-representation of thc G allele (low-responder phenotype) as compared to Caucasian controls. Further studies are needed to explore the relationship between the -308 locus and KD susceptibility. Recently reported Japanese mutations associated with TNFa high-responder phenotype are now being sought in our Caucasian KD patients.
There has been much debate on the superantigen hypothesis of the etiology of Kawasaki syndrome (KS). The two caveats in our previous study were the low frequency of isolating S. aureus and its production of uncharacterized superantigens. In this study, we examined the distribution of the staphylococcal superantigen genes among KS patients and age-matched controls including recently characterized enterotoxin genes.
Rectal and throat swabs were obtained from 21 KS patients and throat swabs were obtained from 122 age-matched control childrcn. The presence of S. aureus was determined by isolating the bacteria or by PCR of the coagulase and/or the protein A genes. The seven enterotoxins (sea, b/c, d, e, g, h, i) and the tsst-1 genes were amplified by PCR.
S. aureus was positive in 61.9%, 71.4% of the throat/rectal swabs from KS, respectively, and 73.0% of the throat swabs from controls. The 4 enterotoxins (sea, b/c, g, i) and the tsst-1 genes were detected from both KS and controls with the comparable rates. However, among the control children under 2 y.o., the frequency of the S. aureus-positive swabs was significantly higher than in the older children (p<0.04). Moreover, the tsst-1 and the sei genes were more frequently found among the isolates from the younger children (56% vs. 11%, p<0.001).
The higher frequency of the tsst-1 and the sei positive S. aureus among the children under 2 years may provide a possible link for the higher incidence of KS among ihe younger children.
We have recently reported the unusual finding that IgA plasma cells (PC) infiltrate vascular tissue in U.S. patients with acute KS, with IgG and IgM PC less often present. To confirm and extend these findings, we have studied additional U.S. KS vascular tissues and other non-vascular tissues, and have studied vascular tissue from 2 Japanese KS patients.
To determine whether tissues contain IgA PC, immunohistochemistry was performed on tissue sections using a high temperature citric acid method, followed by incubation with anti-human IgA and staining using an avidin-biotin-horseradish peroxidase detection system. We have now studied 11 U.S. acute KS coronary artery tissues, as well as one aorta and one renal artery. All of these KS tissues were found to contain IgA PC. In addition, IgA PC have been observed in myocardium and epicardium of 2 of these patients. IgA PC were seen infiltrating the renal cortex in one KS patient who had acute cortical necrosis. None of 14 control vascular tissues were found to contain IgA PC, including 2 tissues with leukocytoclastic vasculitis. Tissues from patients with other types of vasculitis are currently being studied.
Coronary artery from 2 Japanese KS patients also show marked infiltration by IgA PC, and adjacent myocardium in one of the 2 is infiltrated with IgA PC. Coronary artery tissue from a U.S. patient who died 10 months following the onset of KS shows marked intimal proliferation and IgA PC in the adventitial layer of the vascular wall, with some areas in the adventitia showing large clusters of lymphocytic infiltrates which contain IgA-producing cells.
Thus, we have described an unusual immunopathologic feature of IgA PC infiltration in KS which appears to be common to cases in the U.S. and Japan.
Among the adhesion molecules, P-selectin exists on platelet and endothelial cell (EC), E-selectin on EC, L-selectin on leukocyte. Thesc selectins play an active role in mediating those cellular interaction in the initial process of inflammation.
Kawasaki disease (KD) is known as a multisystem vasculitis associated with cardiac complications in children. Histologically, vascular lesions in the acute phase of KD are associated with evidence of activation and damage of EC.
Therefore, to investigate the pathophysiological role of adhesion molecules (P-, E-, L-selectin) in the KD, plasma levels of P-, E-, and L selectins were measured with ELISA in 30 patients with KD in the acute phase (days 3-20) and the convalescent phase (days 30-75) and 10 patients with mild congenital heart disease as control.
Patients with KD (n=30, median age 16 months) were divided into the 2 groups (Group 1: 12 patients with coronary artery lesion (CAL), Group 2: 18 patients without CAL). Plasma levels of P- and E-selectin in acute phase in Group 1 were significantly higher than those in Group 2 (P-selectin; 282 +/- 156 vs. 182 +/- 97 .ng/ml, E-selectin; 94 +/- 60 vs. 63 +/-21 ng/ml, p<0.05). However, plasma L-selectin levels in acute phase in Group 1 were significantly lower than those in group 2 (1384 +/- 296 vs. 1595 +/- 59 ng/ml, p<0.05. In the convalescent phase, there was no difference in P- and L-selectin levels between group 1 and 2, whereas plasma E-selectin levels in group 1 remained to be higher than those in group 2 (77 +/-32 vs. 44 +/- 13 ng/ml, p<0.01). Plasma levels of selectins of KD patients in the acute and convalescent phases were significantly altered as compared to those in control.
Thus, present findings indicate that alterations in plasma selectin levels are important clinical molecular markers for the understanding of the pathophysiology in KD patients who may need closer follow-up.
Objective: Endothelial cells (EC) are reported to be present in circulation in several diseases with vascular injury. The aim of the present study is to detect circulating EC in Kawasaki disease (KD).
Methods: Total RNA was extracted from whole blood samples obtained from 19 patients with KD, 5 patients with Henoch-Schonlein purpura (HSP) and 7 healthy children (HC), and they were analyzed with RT-PCR using primers with the specific sequences of E-selectin (CD62E) which is expressed on activated EC.
Results: The expression of E-selectin mRNA was detected in 9 (50%) of 18 samples in the acute phase of KD, 10 (62.5%) of 16 samples in the subacute phase ot KD, and 1 (7.7%) of 13 samples in the convalescent phase of KD. On the other hand, the expression of E-selectin mRNA was detected only one patient each with HSP in the acute phase (20%) and HC (14.3%).
Conclusions: Activated EC were significantly shedding into circulation in the acute and subacute phases of KD. These observations suggested that the detection of circulating EC-derived E-selectin mRNA using RT-PCR developed in the present study may be a marker of EC injury.
Vascular endothelial growth factor (VEGF), also known as vascular permeability factor, is an important regulator of angiogenesis and blood vessel permeability. Kawasaki disease (KD) is characterized by systemic vasculitis with increased vascular permeability, implying a possible role of VEGF in KD. To elucidate the involvement of VEGF in the pathogenesis of KD, we investigated 30 patients with acute KD, comparing the time course of plasma VEGF levels (n=123) with clinical symptoms and laboratory findings.
Compared to control values, the peak levels of plasma VEGF were significantly elevated (38 +/- 26 vs 244 +/- 248 pg/mL, p<0.001). The VEGF levels at the time of skin rash and/or edema of hands and feet were also elevated to 176 +/- 163 pg/mL (p<0.001). In 7 patients (23%), the plasma VEGF levels remained increased after the resolution of the skin rash and peripheral edema. The VEGF levels were independent of gamma globulin therapy and levels of serum albumin and CRP.
We also measured the plasma levels of transforming growth factor-B1 (TGF B-1) and tumor necrosis facior a, both of which can upregulate VEGF in vitro. The plasma levels of VEGF were highly correlated with those of TGF-B1 (n=63, r=0.73, p<0.001) but not with those of tumor necrosis factor a.
These findings suggest that the production of VEGF is increased and may be upregulated by TGF-B1 in acute KD. VEGF may be involved in the hyperpermeability of local blood vessels in acute KD.
| Return to Home Page | Return to Main Research Page
We welcome your suggestions, too. Please notify us by e-mail if you have any questions or ideas, or if you would like to add your name to our mailing list (and leave your mailing address in the message).
Send comments to: kawasaki@compuserve.com We'd love to hear from you. Let us know if it is OK to list you on our "Families" page.
Web design by Greg Wiles
Illustration by Marcel Isler