fficient responses to different types of pathogens require different mechanisms of immunity. Similarly, for optimal effectiveness and minimal damage to local tissues, the character of the response must be modified to suit particular body compartments. The selection of effector immune functions is controlled by antigen (Ag)-specific T helper (Th) cells, which secrete Th1and Th2-type cytokines at different ratios. Th1-type cytokines [e.g. interferon ␥ (IFN-␥)] promote cellular immunity by activating cytotoxic and phagocytic functions in effector cells, such as cytotoxic T lymphocytes (CTLs), natural killer (NK) cells and macrophages. Interleukin 4 (IL-4), a Th2-type cytokine, supports B-cell production of Ag-specific antibodies, the isotype of which is determined by the cytokine cocktail produced by Ag-specific Th cells 1 .
Matching the character of the response to the nature of the pathogen and the type of infected tissue appears to be a rapid event. Within one week, antibody responses induced within the same lymph nodes show the isotype profiles related to different types of Th-cell responses, depending on the route of infection and the biological activity of the pathogen 2,3 . Indeed, an initial bias to the production of Th1-or Th2-type cytokines can be observed in the lymph nodes within three days of peripheral immunization, depending on the nature of the antigen used.
These data indicate that, at the moment of priming, naive Th cells can receive an initial Th1-or Th2-polarizing signal. This signal, reflecting the nature of the pathogen and the character of the infected tissue, is received despite the fact that naive Th cells do not enter peripheral nonlymphoid tissues and, therefore, are unlikely to meet the pathogen in its native form or to sense the character of the affected tissue. Recent data have suggested that a component of this early polarizing signal can be carried by dendritic cells (DCs) that originate from the site of pathogen entry and are functionally modified by local conditions.