Main

KD is an acute systemic vasculitis of early childhood characterized by fever, rash, cervical lymphadenopathy, and inflammation of the mucous membranes with coronary artery damage occurring in 15-20% of affected patients(1, 2). Although it has been widely accepted that KD is an infectious disease, the etiologic agent has not yet been identified. The acute phase of KD is marked by profound immunoregulatory changes that include marked activation of T cells and monocyte/macrophages(3). These characteristic immunologic features suggested a pathogenic role for microbial superantigens in this disease. As it is possible that the GI tract could be one of the primary sites of entry of bacterial toxins in KD patients, we carried out a microbiologic investigation of the small intestine in a preliminary study (Yamashiro, Y, Nagata S, Oguchi S, Shimizu T, manuscript accepted for publication by this journal). In that study, we demonstrated that five kinds of streptococci and two kinds of staphylococci, not detected in control patients, were present in the lumen of the jejunum of KD patients. Staphylococcal enterotoxins and streptococcal exotoxins are known to be superantigens which stimulate large populations of T cells expressing particular TCR Vβ gene segments(4). In the present study, we therefore investigated TCR Vβ expression by T cells in the small intestinal mucosa of KD patients using an immunohistochemical technique.

METHODS

We studied 12 Japanese KD patients (seven boys and five girls, aged 2.0 y to 4.8 y; mean age 3.1 y) who fulfilled established clinical criteria for the diagnosis of KD. Jejunal biopsy specimens were obtained using a sterile pediatric Crosby-type capsule. Jejunal biopsies in the acute phase of KD were performed in the fasting state about 4-6 d after the onset of symptoms, before the fever resolved. Jejunal tissues from eight patients with food-sensitive enteropathies in the convalescent phase (four boys and four girls, aged 10 mo to 4.7 y; mean age 3.0 y) were used as controls. None of the subjects or controls was treated with antibiotics before the jejunal biopsies were performed. All of the families of the patients and the control subjects consented to participate in the study.

The freshly obtained biopsies were oriented, rapidly frozen in liquid nitrogen, and stored at -80°C until use. The specimens were stained with anti-human Vβ MAb by an immunofluorescence technique. Two-micrometer thick frozen sections were cut, fixed in cold acetone, and stained with biotinylated antibodies, followed by avidin-FITC. The specimens were examined under a Zeiss incident-light fluorescent microscope. The mAb used were those against CD3 (Becton Dickinson, Mountain View, CA), TCR Vβ2 (Immunotech, Marseille, France), Vβ 5a, 5b, 5c, 6a, 8a, and 12a (T Cell Science, Cambridge, MA). Because human Vβ genes have more than 20 families, the use of anti-Vβ mAb allows only a limited analysis of the T cell repertoire. We selected the above mentioned anti-Vβ mAb after referring to some previous studies for peripheral blood TCR Vβ gene expression in KD(4, 5). The stained lamina propria cells were counted using a counting grid with 10 × 10 squares, incorporated into a×40 objective lens. The percentage of T cells bearing each TCR Vβ phenotype was expressed as a percentage of total T (CD3+) cells.

Statistical analysis was performed using the t test.

RESULTS AND DISCUSSION

The results of TCR Vβ expression by T cells in the small intestinal mucosa of KD patients and control children are summarized inFigure 1. The occurrence of Vβ2+ T cells was found to be selectively increased in the small intestinal mucosa of patients in the acute phase of KD compared to controls (p < 0.01). None of the other Vβ populations examined, including Vβ 5a, 5b, 5c, 6a, 8a, and 12a was found to be significantly expanded. In both KD and control specimens, Vβ2+ T cells were found more often in the lamina propria than in the epithelium.

Figure 1
figure 1

TCR Vβ expression by T cells in the small intestinal mucosa of KD patients and control children. The occurrence of Vβ2+ T cells was selectively increased in patients in the acute phase of acute KD compared to controls (p < 0.01). None of the other Vβ populations examined, including Vβ 5a, 5b, 5c, 6a, 8a, and 12a, was found to be significantly expanded.

Full size image

Recently by means of the polymerase chain reaction technique and MAb reactivity, it has been found that a selective increase of Vβ2+ T cells in peripheral blood is associated with the acute phase of KD(4–6). Furthermore, new clones of toxic shock syndrome toxin-secreting Staphylococcus aureus and streptococcal pyogenic exotoxin B-producing or C-producing streptococci, which are known to stimulate Vβ2+ T cells, have been isolated by Leung et al.(7) from the throat and rectum of patients with KD.

It seems possible that the GI tract could be one of the primary sites of entry for etiologic agents of KD. In our preliminary study, we suggested that some antigens, which induced a strong delayed-type hypersensitivity reaction in the mucosa, might have invaded the body of KD patients by breaching the barrier of the small intestinal mucosa(8). The GI tract has a greater likelihood to be the primary site of entry of the causative antigen than the throat or the respiratory tract because of its broad exposure to a variety of foreign antigens. It is well known that T cell activation by exotoxins having superantigenic properties requires expression of MHC class II molecules on antigen-presenting cells. Activation of a large proportion of T cells may require an extensively infected lesion and many MHC class II molecules. It seems likely that the small intestine is a superantigen-presenting organ because it has a vast surface area exposed to various foreign antigens and large numbers of MHC class II molecules are present on the surface of antigen-presenting cells such as epithelial cells. The observation by Leung et al.(7) that most patients with KD have toxic shock syndrome toxin-secreting S. aureus or SPEC-secreting streptococci in their pharynx or rectum is consistent with this hypothesis, suggesting the GI tract is the major site of infection.

In a preliminary study, we found that five kinds of streptococci and two kinds of staphylococci, not detected in control patients, were present in the lumen of the jejunum of KD patients (accepted for publication). Staphylococcal enterotoxins and streptococcal exotoxins are known to be superantigens which stimulate large populations of T cells expressing particular TCR Vβ gene segments(4).

In the present study, we found that the T cell repertoire in the small intestinal mucosa of patients in the acute phase of KD was skewed with an increased proportion of Vβ2+ cells among the T cell populations. We found no significant difference in the occurrence of Vβ8+ T cells in the jejunum of KD patients as compared to controls. Our finding is in contradiction with some previous studies(4), however, Leung et al.(7) mentioned in a recent report that an expansion of T cell populations expressing Vβ2, and less so of Vβ8 receptors, is associated with the acute phase of KD. Thus, it is likely that there are several causative antigens and certain superantigens stimulate both Vβ2+ and Vβ8+ T cells.

Although we investigated TCR Vβ expression by T cells in peripheral blood from five of the twelve patients in the acute phase of KD, none of the Vβ populations was found to be significantly expanded compared to controls (data not shown). Indeed previous studies did not demonstrate that all KD patients have a selective increase in occurrence of T cells of any Vβ family in the peripheral blood in the acute phase(4–6). Furthermore, Pietra et al.(9) reported no significant expansion of T cells of any Vβ family in the peripheral blood of KD patients as well. In a recent study by Curtis et al.(6) it was found that the timing of the blood collection was critical for demonstration of Vβ2 expansion. Their data suggest that after expansion of Vβ2+ T cells in the blood there may be rapid migration of such cells into the lymphoid tissues of KD patients. Our current study demonstrating the selective expansion of Vβ2+ T cells in the small intestinal mucosa, but not the peripheral blood, of acute KD patients reinforce the importance of examining the lymphoid tissues (where most of their activated T cells exist) from these patients. We have not proved, as yet, that the bacteria, isolated in our previous study from the lumen of the jejunum of KD patients produce toxins which stimulate Vβ2+ T cells. However, it seems credible that KD may be caused by exotoxins having superantigenic properties produced by bacteria which have invaded the body of KD patients through the small intestinal mucosa. The evidence would be more convincing if we could demonstrate that we can isolate bacteria producing a toxin which induces the selective proliferation of Vβ2+ cells can be isolated from the jejunum of KD patients. We are now investigating the biological activities of bacteria isolated from the microflora of the small intestine of KD patients.