Inhaled Bacillus anthracis spores germinate and spread throughout the host only after being phagocytosed. Macrophages, thought to be involved, are not essential for dissemination. Brittingham et al. (p. 5545 ) exposed immature human dendritic cells (DCs) to spores of the fully virulent Ames strain of B. anthracis and to two attenuated strains, one containing only the lethal toxin and the other lacking both lethal and edema toxins. Internalization of spores was observed by confocal microscopy using spore-specific Abs. The method of uptake was determined by transmission electron microscopy to be coiling phagocytosis. Both attenuated spores decreased mRNA expression of tissue-retaining chemokine receptors in DCs but increased expression of maturation markers and of CCR7, required for DC migration to lymphatics. Spore-matured DCs stimulated allogeneic T cells in proliferation assays and induced weak phosphorylation of ERK1/2, p38, and other protein kinases within 30 min of infection. Phosphorylation was detected at 6 h only in DCs infected with the spore lacking toxins. MEK cleavage products were seen at 6 h in DCs infected with the toxin-containing spore; induced inflammatory cytokine mRNAs plateaued at 4 h in these cultures but continued to rise in DCs infected with the spore lacking toxins. The data suggest that B. anthracis spores activate the MEK signaling cascade to direct the engulfing pulmonary DCs to the mediastinal lymph nodes early in infection, but later they interfere with MEK signaling to suppress immune responses.

Although loss of expression of Q9, a nonpolymorphic MHC class Ib molecule, occurs in many murine tumor cell lines, Stroynowski and colleagues have shown that ectopic expression of Q9 in mouse melanoma cells renders them susceptible to antitumor CTL killing. In a continuation of their work, Chiang and Stroynowski (p. 5367 ) transfected Q9 cDNA into mouse lung carcinoma, T cell lymphoma, and hepatoma cells lacking Q9 expression. Few mice injected s.c. with the Q9+ lung carcinoma or lymphoma cells developed tumors compared with mice injected with the Q9+ hepatoma cells or with Q9 tumor cells. CTLs from tumor-free vaccinated mice were cytotoxic in vitro for Q9+ lung carcinoma, T cell lymphoma, and melanoma cells, but not Q9+ hepatoma cells. Nonradioactive Q9+ lung, lymphoma, or melanoma tumor cells or anti-Q9 Ab inhibited lysis of radioactive Q9+ melanoma cells by melanoma-specific CTLs. Mice that survived an initial challenge with Q9+ lung carcinoma or lymphoma cells were resistant to a dose of Q9+ melanoma cells that was lethal in naive mice. No vitiligo or autoimmune destruction of melanocytes was seen during Q9+ melanoma cell challenge of mice immunized with Q9+ tumors or in melanocyte cells added to in vitro CTL reactions. Anti-Q9+ CTL did not kill activated Q9+ splenic T cells. The authors conclude that Q9 presentation of a tumor Ag shared among mouse lung carcinomas, T cell lymphomas, and melanomas, but not hepatomas, renders the tumors susceptible to CTL-mediated killing.

Vaccines that induce tumor-specific Abs are used in therapy of human cancers. Although they are proving to be effective in clinical trials, the mechanism by which they act is not known. Ragupathi et al. (p. 5706 ) compared conjugate vaccines against glycolipid or mucin Ags with regard to their ability to bind to cells, as determined by flow cytometry, and to kill target cells by C-dependent cytotoxicity (CDC). Of sera from patients vaccinated with glycolipid Ags, 48 had strong cell binding and 39 had strong CDC in published trials. In contrast, although most sera from 42 patients vaccinated with mucin Ags exhibited strong cell binding, none had strong CDC. These results correlated well with the results of their own cohort of 100 patients. The patterns were confirmed in their laboratory by retesting 42 of the glycolipid Ag sera and 31 of the mucin Ag sera on MCF-7 breast cancer cells. An additional 15 murine mAbs and 1 human mAb were tested against MCF-7 and LSC (human colon cancer) cells. Five mAbs against glycolipid Ags had strong CDC reactivity against both cell lines, whereas 11 mAbs against mucin Ags had no CDC reactivity. All of the mAbs had comparable cell surface Ig binding; however, only mAbs directed against glycolipid Ags activated C3b and C5b-9 at the cell surface. The authors suggest that differences in molecular architecture between the two classes of molecules are responsible for the different CDC abilities. Glycolipids are small and close enough to the cell surface to permit the C components to be effective, whereas rod-like mucins extend too far from the surface for C-mediated lysis to occur.

Bacterial meningitis, caused by Neisseria meningitidis serogroup B, is increasing in incidence. The variable major outer membrane protein, porin A (PorA), induces T cell-dependent Ab titers; however, the effective PorA T cell epitopes have not been defined. Meiring et al. (p. 5636 ) labeled meningococcal outer membrane vesicles containing PorA with two nitrogen isotopes, 14N or 15N. The labeled outer membrane vesicles were fed to immature homozygous human dendritic cells (DCs) that had been cultured in vitro for 6 days. Two days later, MHC-peptide complexes were purified from the DCs, and eluted peptides were analyzed by nanoscale liquid chromatography electrospray ionization mass spectrometry. Ten peptide ion pairs derived from four regions of the PorA protein were identified. The pairs were expressed at densities ranging from 30 copies to 4000 copies per cell. CD4+ T cells from individuals of the same HLA type as the pulsed DCs proliferated and produced IFN-γ in response to in vitro stimulation with overlapping peptides from two of the four PorA regions. A CD4+ T cell clone specific for one of the two immunodominant regions recognized two length variants of processed PorA. N- and C-terminal truncations and substitutions of single amino acids within the sequence were used to define a minimal core length of 11 aa and an optimal amino acid composition. The experiments show that the use of mass tagging and spectrometry can improve vaccine design by identifying effective Ag-derived MHC class II epitopes of a pathogen, N. meningitidis, taken up by human DCs.

The agents of Lyme disease and relapsing fever, Borrelia burgdorferi and Borrelia hermsii, respectively, are cleared by pathogen-specific IgM early in infection. The Bockenstedt laboratory demonstrated that mice deficient in CD1d, normally expressed at high levels on spleen marginal zone B (MZB) cells, are less able to control B. burgdorferi infection. In a continuation of those studies, Belperron et al. (p. 5681 ) found spirochete binding to wild-type mouse MZB cells and increased B7.1 expression only during the first 16 h after B. hermsii infection, whereas binding to CD1d−/− MZB cells and increased B7.1 expression were detected through 96 h postinfection (p.i.). Up-regulation of syndecan I, a plasma cell marker, occurred only during the first 24 h of infection in wild-type MZB cells, but syndecan I expression was delayed in CD1d−/− MZB cells and remained high through 96 h p.i. Overall, B. hermsii-specific IgM Ab production was impaired in CD1d−/− MZB cells; spirochete-specific IgM Ab serum levels were low, and peripheral blood spirochete levels were high compared with infected wild-type mice. Selective mAb depletion of MZB cells from wild-type mice before infection also resulted in reduced serum titers of B. hermsii-specific IgM Ab and increased pathogen numbers. Infected CD1d−/− mice given immune serum from infected wild-type mice for 72 h p.i. resolved their infections, and expression levels of B7.1 and syndecan I on their MZB cells returned to baseline. The authors conclude that CD1d expression on MZB cells is critical to production of pathogen-specific IgM early in B. hermsii infection in mice.

The frequency of UV-induced skin cancer in humans has increased significantly in recent years. Beissert and colleagues have shown that UV-induced suppressor CD4+CD25+ T cells regulate photocarcinogenesis in mice. In Loser et al. (p. 5298 ), the same group examined factors regulating the function of those cells. UV-treated mice injected with an anti-CTLA-4 Ab developed 50% fewer skin tumors of lower malignancy than mice exposed to UV alone or to UV plus a control Ab. Tumor-free, UV-irradiated, and anti-CTLA-4-treated mice rejected each of two UV progressor tumor lines; naive mice were unable to do so. Mice doubly deficient in CD80 and CD86 had a reduced incidence of UV-induced tumors compared with wild-type or CD80−/− mice, whereas UV-induced tumors appeared earlier in CD86−/− mice. Numbers of CD4+CD25+ T cells in wild-type, CD80−/−, and CD86−/− mice were equivalent before UV exposure and increased after exposure; CD80−/−CD86−/− mice had reduced numbers of suppressor cells before and after UV exposure. CD4+CD25+ T cells from mice treated with UV and anti-CTLA-4 Ab for 4 wk did not suppress stimulated CD4+ T cells in vitro, and anti-CTLA-4 Ab inhibited the in vitro proliferation of stimulated CD4+CD25+ T cells. Mice injected with UV-induced CD4+CD25+ T cells rejected a transplanted UV-induced tumor only when co-injected with anti-CTLA-4 Ab. Dendritic cells from wild-type mice induced more T cell proliferation than those from CD80−/−CD86−/− mice. The data show that involvement of the CTLA-4 pathway in UV-induced skin cancer in mice is mediated via CD80/CD86-CTLA-4 signaling in CD4+CD25+ suppressor T cells.

Autoimmune disease in mice and humans is associated with overproduction of B cell-activating factor belonging to the TNF family (BAFF). BAFF is known to regulate T cell activation; however, its effect on Th1- and Th2-mediated T cell responses has not been determined. Sutherland et al. (p. 5537 ) demonstrated increased paw swelling with erythema and intracellular infiltrate in BAFF transgenic (Tg) mice compared with wild-type mice 48 and 72 h after intradermal immunization and challenge with BSA. Serum levels of IgG1, IgG2a, and IgG2b also were higher in the BAFF Tg mice. BAFF serum levels and the magnitude of delayed-type hypersensitivity responses correlated with degree of paw swelling; the correlation was lost in B cell-deficient BAFF Tg mice. BSA-stimulated BAFF Tg cells from lymph nodes adjacent to the injection site had greater proliferation and IFN-γ production than stimulated cells from immunized wild-type animals. Splenocytes from BAFF Tg mice were enriched in CD44highCD62Llow effector memory CD4+ T cells compared with wild-type and B cell-deficient BAFF Tg controls. BAFF Tg and B cell-deficient BAFF Tg mice exposed to OVA aerosol had reduced eosinophil infiltration in bronchoalveolar fluid and lower in vitro peribronchial lymphocyte proliferation and IL-5 production in response to OVA stimulation compared with controls. However, lymphocytes from immunized BAFF Tg mice had higher OVA-specific recall responses than controls. The authors conclude that high levels of BAFF exacerbate Th1 responses but suppress Th2 responses in mice and that only the Th1 response is dependent on B cells.

Summaries written by Dorothy L. Buchhagen, Ph.D.