点击显示 收起
1Department of Pediatrics, Kawasaki Medical School, Kurashiki, 2Department of Medical Genetics, Tohoku University School of Medicine, Sendai, 3Department of Health Promotion and Human Behavior, Kyoto University Graduate School of Public Health, Kyoto, and 4Department of Microbiology and Immunology, Kagoshima University Dental School, Kagoshima, Japan
Received 28 May 2002; revised 12 September 2002; electronically published 6 January 2003.
Since its first description in Japan >30 years ago, Kawasaki disease (KD) has been reported worldwide. Although an infectious etiology is suspected based on the epidemiology and clinical features, a causative agent has not been identified. The disease is more frequent in children of Japanese ancestry, and siblings of children with KD have a significantly greater risk of developing KD than do children of the same age in the general population. This suggests a possible genetic susceptibility to KD. Results of this study showed that allele 1 of the 5 promoter (GT)n repeat in the SLC11A1 (formerly NRAMP1) gene, which endows the gene with a weak promoter activity, was highly represented in patients with KD. This suggests possible explanations for both the infectious etiology of this disease and the genetic risk in the Japanese population.
Informed consent was obtained from the parents of all patients and control subjects in accordance with institutional review board guidelines. Ethics committees of Saiseikai Shimonoseki (former affiliation of K.O.), Yamaguchi University, and Kagoshima University hospitals approved the study.
Reprints or correspondence: Dr. Kazunobu Ouchi, Dept. of Pediatrics, Kawasaki Medical School, 577 Matsushima, Kurashiki, 701-0192 Okayama, Japan ().
Kawasaki disease (KD) is an acute multisystem vasculitis that occurs in infants and children [1]. The diagnostic criteria for KD include fever for 5 days, typically with at least 4 of the following 5 clinical features: bilateral conjunctival injection without purulent discharge; inflammatory changes of the lips and "strawberry" tongue; changes of the peripheral extremities, particularly redness and swelling of the hands and feet with subsequent periungual desquamation; rash, primarily truncal and taking many forms, but nonvesicular; and cervical lymphadenopathy, usually unilateral.
Since 1970, nationwide epidemiologic surveys have been conducted biennially in Japan with the cooperation of medical facilities throughout the country. Three nationwide epidemics of KD occurred in Japan in 1979, 1982, and 1986 [2]. KD is clearly overrepresented in children of Asian background, both in Asia and North America. In 2001, Yanagawa et al. [3] reported an annual attack rate in Japan of 140/100,000 children aged <3 years, which is higher than the rate in US whites and blacks (10/100,000 children aged <3 years) [4]. The KD attack rate for US Asians and for Pacific Islanders in San Diego County and Hawaii is 3 times higher than that in US white children [4, 5]. Siblings of children with KD have a significantly greater risk of developing KD than children of the same age in the general population, which suggests a possible genetic risk. However, the basis of genetic susceptibility remains unclear.
The epidemic nature, seasonal distribution, and clinical features suggest an infectious etiology of this disease. Multiple agents have been implicated, including Streptococcus, Staphylococcus, Yersinia, and Rickettsia organisms, viruses, and environmental chemicals. However, none has been conclusively demonstrated to be causative. The activation of T cells, B cells, and monocytes and/or macrophages is well known in KD and suggests a generalized immune activation. Previous studies suggest that immunoregulatory abnormalities may contribute to its pathogenesis, and one of the significant clinical features of KD is local inflammatory reactivation of a previous bacillus Calmette-Guérin (BCG) inoculation site, a specific and early manifestation of KD [2, 6], which shows an erythematous indurated plaque with activated macrophages at the skin lesion, although no evidence of active mycobacterial infection has been observed.
In laboratory strains of inbred mice, resistance and/or susceptibility to the growth of BCG is controlled by locus Bcg. Subsequently, a gene called SLC11A1 (formerly NRAMP1) was identified in the locus that controls resistance to BCG and other intracellular parasites, and the human homologue was subsequently isolated [7, 8]. SLC11A1 regulates the cascade of gene-inductive events that follow interaction of macrophages with bacterial lipopolysaccharide (LPS) and/or natural killer cell or T cellderived interferon (IFN). The gene has multiple pleiotropic effects, including regulation of interleukin-1, tumor necrosis factor, and major histocompatibility complex class II molecules [9]. Several polymorphic variants have been found in the human SLC11A1 gene promoter, which may change many different aspects of macrophage function and many cytokine responses [10, 11]. KD has unique properties of macrophage and cytokine responses in the acute phase. Here we present evidence that the polymorphism at the human SLC11A1 gene promoter confers susceptibility to KD.
Materials and methods. Blood samples were obtained from 71 Japanese patients with KD and 110 age- and sex-matched healthy Japanese volunteers residing in the same regional area (Yamaguchi Prefecture, Japan). The diagnosis of KD was based on criteria described elsewhere [2].
Leukocyte DNA was extracted by use of the Wizard genomic DNA purification kit (Promega), and 50 ng was used in a polymerase chain reaction (PCR). The 5-promoter (GT)n was genotyped by PCR with primers 5-GTC TTG GAA CTC CAG ATC AAA G-3 and 5-TTG CAT ATT CAT GTC AAT ACC C-3. PCR was done under the following conditions: denaturation at 95°C for 2 min followed by 30 cycles of denaturation at 95°C for 1 min, annealing at 60°C for 1 min, and extension at 72°C for 2 min. PCR was terminated by extension at 72°C for 7 min followed by cooling to 4°C.
Three alleles, 134 bp (designated as "allele 1"), 132 bp ("allele 2"), and 130 bp ("allele 3"), were identified in the Japanese population and imaged by use of an automated DNA sequencer. Exon15, Asp543Asn (D543N), and the deletion of the 3-untranslated region (3-UTR D/I) were genotyped by PCR with the primers 5-GCA TCT CCC CAA TTC ATG GT-3 and 5-AAC TGT CCC ACT CTA TCC TG-3 with the following conditions: denaturation at 95°C for 2 min followed by 35 cycles of denaturation at 95°C for 1 min, annealing at 58°C for 1 min, and extension at 72°C for 2 min. PCR was terminated by extension at 72°C for 7 min followed by cooling to 4°C. Restriction-endonuclease digestions were then done by using AvaII and FokI under conditions recommended by the supplier (New England Biolabs). Restriction enzyme digestion products were resolved by electrophoresis on 12% polyacrylamide gels stained with ethidium bromide. AvaII cleaves the Asp alleles of exon 15 into 3 products of 126, 79, and 39 bp and the Asn alleles into 2 products of 201 and 39 bp. FokI cleaves the -TGTG allele of the 3-UTR region into 211 and 33 bp and does not cleave the +TGTG allele.
We estimated the power of this study by using the formula described by Ohashi et al. [12]. Our study had about 80% power (at the .05 level) to detect an allelic association with a sample size of 71 cases and 110 control subjects, assuming that the frequency of risk allele in control subjects is 0.03 and its relative risk is 3.35. (Under these conditions the incidence of KD is calculated to be 0.0014 in dominant model.) We used the Fisher's exact test to assess the association between the polymorphisms and the disease. Contingency table analysis, odds ratios, 95% confidence intervals, and significance values were estimated by use of computerized methods (SPSS program vers. 10.1J). In the 5-promoter polymorphism, we also estimated empirical P values by use of the Monte Carlo method, using the CLUMP program with 100,000 stimulations [13]. Corrected P values with the number of alleles (Pc) <.02 were considered to be statistically significant.
Results. We conducted a genetic association study in a Japanese population to determine whether variants of SLC11A1 relate to KD. We studied 71 Japanese children (48 boys and 23 girls) who were diagnosed with KD at ages 05 years and 110 healthy children (control subjects) without a history of KD who were matched by age and sex. No significant difference was observed in age or ratio of sexes between patients and control subjects. shows gene frequency of genotypes of the 3-UTR region, exon 15 region, and 5-promoter (GT)n region in children with KD and in control subjects. No association was found between KD and distribution of the variants 3-UTR D/I (P = .834) and exon 15 Asn543Asp (P = .834). However, the association of the distribution of the 5-promoter (GT)n region and KD was significant, and the gene frequency of allele 1 of the 5-promoter (GT)n was significantly higher in patients with KD than in control subjects (P = .0074).
fig.ommitted
Table 1. SLC11A1 gene polymorphisms in Japanese children with Kawasaki disease (n = 71) and in control subjects (n = 110).
Discussion. We investigated genetic associations in a Japanese population to determine whether variants of SLC11A1 relate to KD, a disease for which the etiology and genetic susceptibility remain unclear. To date, 11 variants including 5 coding regions, 3 in the introns, 2 in the 3-UTR, and 1 in the 5 promoter region) have been identified in SLC11A1. Among these, only 2 of the polymorphisms were predicted to cause amino acid substitution in A318V, an alaninevaline substitution, and in D543N, an aspartic acidasparagine substitution [8].
We genotyped the D543N polymorphism because this could affect protein function by substituting a negatively charged amino acid with an uncharged residue in the cytoplasmic carboxy terminal domain and could possibly alter the macrophage function. Another polymorphism on which we focused is a Z DNA-forming polymorphic (GT)n repeat 250300 bp upstream of the transcription start site, designated here as the 5-promoter (GT)n 14, which may be a functional polymorphism at the transcription level [10]. In addition to these possibly functional polymorphisms, a 4-bp deletion polymorphism in the 3-UTR region, designated here as 3-UTR D/I, was genotyped. The physiologic effects of sequence polymorphisms in the 3-UTR are not fully understood, although there are regulatory elements for several genes within the 3-UTR.
D543N and 3-UTR D/I appeared to be in absolute linkage disequilibrium, and no association was found between KD and D543N and 3-UTR D/I. In contrast, in this study, the gene frequency of allele 1 of 5-promoter (GT)n was 0.032 in the control children and 0.113 in the KD patients, which indicates a significant association of the 5-promoter (GT)n with susceptibility of KD. Allele 1 frequency was previously reported to be 0.022 in adults in an area of Japan remote from the region in our study [14]. Elsewhere the allele 1 frequency is 0.001, 0.021, and 0.039 in white persons [11], Brazilians [10], and Koreans [15], respectively. These findings show that Asians have a higher genetic frequency and incidence of KD than persons elsewhere. These results are suggestive of the importance of SLC11A1 expression level in the difference of susceptibility of KD.
Of the 3 (GT)n alleles, allele 3 was observed to drive more than 10-fold higher levels of gene expression than either allele 1 or allele 2 [11]. Addition of IFN- as an exogenous stimulus causes a 12-fold enhancement of the gene expression in all 3 alleles, and bacterial LPS as a second signal results in a further 2-fold enhancement of allele 3driven gene expression. In contrast, no enhancement by LPS was observed in the expression levels of alleles 1 and 2. Moreover, allele 1 differed significantly from allele 2, and its baseline expression level was less than one-half that of allele 2. Therefore, there might be a biologic difference in phenotypic expression between allele 1 and allele 2. The significance of these biologic differences in human disease remains to be investigated.
KD is characterized by monocyte and/or macrophage activation and high circulating cytokine levels of all of the proinflammatory cytokines in response to IFN-, LPS, and other stimuli in the acute phase. It seems inconsistent that allele 1, a poor promoter of SLC11A1, which regulates the cascade of gene-inductive events that follow interaction of macrophages or many procytokines, increased patients with KD than in control subjects. However, no data showed polymorphisms of the SLC11A1 gene; therefore, circulating cytokine levels in homozygous KD patients or control subjects should be analyzed further with each allele of the SLC11A1 gene.
In conclusion, we show that 1 variant of the SLC11A1 gene is associated with KD in a group of Japanese children. The skewed ethnic distribution and seasonality are consistent with the hypothesis that KD is an infectious disease that is influenced by environmental and genetic factors. Our results suggest that KD might be caused or induced by a certain infection with a yet unknown intracellular parasite to which innate resistance of the host is regulated by level of SLC11A1 gene expression. The high incidence of KD in Asians may be due to a high genetic susceptibility to this pathogen.
Acknowledgments
We thank T. Yoshida, M. Fujiwara, T. Matsubara, and S. Furukawa of the Japan Kawasaki Disease Research Group for help with sample collection, clinical information, and critical review of the study proposal.
References
1.
2.
3.
4.
5.
6.
7.
8.
9.
Skamene E. The Bcg gene story. Immunobiology 1994; 191:45160.
10.
11.
12.
13.
14.
15.