The role of secretory IgA in conferring cross-protective immunity was examined in polymeric (p)IgR knockout (KO) mice immunized intranasally with different inactivated vaccines prepared from A/PR/8/34 (H1N1), A/Yamagata/120/86 (H1N1), A/Beijing/262/95 (H1N1), and B/Ibaraki/2/85 viruses and infected with the A/PR/8/34 virus in the upper respiratory tract (RT)-restricting volume. In wild-type mice, immunization with A/PR/8/34 or its variant (A/Yamagata/120/86 and A/Beijing/262/95) vaccines conferred complete protection or partial cross-protection against infection, while the B-type virus vaccine failed to provide protection. The protection or cross-protection was accompanied by an increase in the nasal A/PR/8/34 hemagglutinin-reactive IgA concentration, which was estimated to be >30 times the serum IgA concentration and much higher than the nasal IgG concentration. In contrast, the blockade of transepithelial transport of dimeric IgA in pIgR-KO mice reduced the degree of protection or cross-protection, in parallel with the marked increase in serum IgA concentration and the decrease in nasal IgA concentration (∼20 and 0.3 times those in wild-type mice, respectively). The degree of the reduction of protection or cross-protection was moderately reversed by the low but non-negligible level of nasal IgA, transudates from the accumulated serum IgA. These results, together with the absence of the IgA-dependent cross-protection in the lower RT and the unaltered level of nasal or serum IgG in wild-type and pIgR-KO mice, confirm that the actively secreted IgA plays an important role in cross-protection against variant virus infection in the upper RT, which cannot be substituted by serum IgG.

We have compared the cytolytic activities and the cellular compositions of the inal lntraeplthellal lymphocyte (i-IEL) populations in three different combinations of conventional (CV) and germ-free (GF) mice. Cytolytic activity of i-EELs exps yS T-l antigen receptors (TCRs) b strain dependent in CV mice (high vs. low), and this straindependent variability b unaltered in the GF condition. Although absolute numbers of yv i-IELs are sightly decrased, the composition of CD8aa+ and CD4-CD8-subsets and the usage of TCR r and 8-chain variable gene segments by yS i-EELs remain the same in GF mice. By contrast, cytolytic activity of ar TCR-expressing i-JELs is uniformly high in CV mice but attenuated sharply in the GF condltion. A conspicuous decrease in the total numbers of afl i-IELs is also noted, and CD8rP+ and CD4+CD8+ subsets are reduced, whereas the CD8aa+ subset is expanded in GF mice. These rults indicate that microbial deprivation preferentially influences the a,B i-EEL population to decrease and become noncytolytic but has little effect on the pd size or characteristics of yS i-EELs. Consequently, cytolytic activity of feshly isolated i-KELs from GF mce b determined by T cellsexp ng y8TCRs and is found to be strin dependent.

T cell receptor α  mutant (TCRα –/–) mice, which spontaneously develop colitis under conventional conditions, did not show any signs of colitis under germ‐free conditions, leaving TCRα –β + cells (β dim cells) and TCRγ δ + cells much reduced. Moreover, TCRα –/– mice with alymphoplastic mutation (aly/aly TCRα –/– mice), which lack Peyer's patches and peripheral lymph nodes, did not suffer from colitis. While both β dim cells and TCRγ δ + cells were present in the colons of aly/aly TCRα –/– mice and aly/+ TCRα –/– mice, cytotoxicity of colonic TCRγ δ + cells in aly/aly TCRα –/– mice was almost abolished. Transfer of TCRγ δ + cells from TCRα –/– mice into scid/scid mice or aly/aly TCRα –/– mice could not induce colitis, but injection of anti‐TCRδ  mAb into TCRα –/– mice prevented colitis from developing. Finally, TCRα –/– mice expressing transgenic (Tg) KN6‐TCRγ δ  hardly developed colitis, accompanied by colonization of non‐cytotoxic Tg TCRγ δ + cells in their colonic mucosa. These results demonstrate that intestinal resident TCRγ δ + cells may be involved in the exacerbation of inflammatory bowel disease in TCRα –/– mice.

Intraepithelial T lymphocytes in the small intestine (IEL) consist of αβ T‐cell receptor (TCR)‐bearing T cells (αβ‐IEL) and γδ TCR‐bearing T cells (γδ‐IEL). Development and cytolytic activation of αβ‐IEL sharply attenuate in germ‐free (GF) mice fed a natural diet (Nat‐GF), but the number and cytotoxicity of γδ‐IEL are comparable between conventional (CV) and Nat‐GF mice. In this report, we compared the properties of IEL in Nat‐GF mice and GF mice fed antigen‐minimized diet (AgM‐GF mice) of C57BL/6 strain to evaluate an influence of gut antigenic load on IEL development. Numbers of αβ‐IEL and γδ‐IEL in AgM‐GF mice were less by 1.9‐ and 1.4‐fold than those in Nat‐GF mice, respectively. Significant decreases in the proportions of CD4+8−, CD4−8αβ+, and CD4+8+ subsets and a resultant increase in the ratio of CD4−8αα+ subset were evident in αβ‐IEL of Nat‐GF mice compared with CV mice, but the subset constitution of αβ‐IEL was similar between Nat‐GF and AgM‐GF mice. In contrast, relative composition of γδ‐IEL was not different between CV, Nat‐GF, and AgM‐GF mice. αβ‐IEL displayed low cytolytic activity in Nat‐GF mice and were almost deprived of their cytotoxicity under the antigen‐minimized condition. While γδ‐IEL were strongly cytolytic in Nat‐GF mice their cytolytic activity was remarkably reduced in AgM‐GF mice. These results indicate that γδ‐IEL are activated independently of microbial colonization in the gastrointestinal tract but their activation occurs in response to the exogenous antigenic substances other than live micro‐organisms.

To evaluate the safety of two probiotic bacterial strains, Lactobacillus casei strain Shirota (LcS) and Bifidobacterium breve strain Yakult (BbY), these probiotics were orally administered to Lewis rats with experimental autoimmune encephalomyelitis (EAE), the experimental model of human multiple sclerosis. We examined three experimental designs by combining different antigen types and probiotic administration periods: (1) EAE was induced with a homogenate of guinea pig spinal cord as the sensitizing antigen, and LcS was orally administered from one week before this sensitization until the end of the experiment; (2) EAE was induced using guinea pig originated myelin basic protein (MBP) as the sensitizing antigen, and LcS was orally administered from one week before this sensitization to the end of the experiment; (3) EAE was induced using guinea pig MBP as the sensitizing antigen, and the probiotic strains (LcS and BbY) were administered starting in infancy (two weeks old) and continued until the end of the experiment. In experiment 1, oral administration of LcS tended to suppress the development of neurological symptoms. Differences in neurological symptoms between the control group and the administration groups did not reach statistical significance in experiments 2 and 3. These results support the notion that neither LcS nor BbY exacerbates autoimmune disease.