Search

Monocytes have been classically considered essential elements in relation with innate immune responses against pathogens, and inflammatory processes caused by external aggressions, infection and autoimmune disease. However, although their potential to differentiate into dendritic cells (DCs) was discovered 14 years ago, their functional relevance with regard to adaptive immune responses has only been uncovered very recently. Studies performed over the last years have revealed that monocyte-derived DCs play an important role in innate and adaptive immunity, due to their microbicidal potential, capacity to stimulate CD4(+) and CD8(+) T-cell responses and ability to regulate Immunoglobulin production by B cells. In addition, monocyte-derived DCs not only constitute a subset of DCs formed at inflammatory foci, as previously thought, but also comprise different subsets of DCs located in antigen capture areas, such as the skin and the intestinal, respiratory and reproductive tracts.

The prolonged, abnormal immune response patterns produced by many sexually transmitted viruses have been intensively studied. Because normal antiviral immune responses in the vagina are less well-defined, we developed a resolving murine model using vaginal inoculation with the flavivirus, West Nile virus. Infection resulted in 12% mortality, with sterile protective immunity to vaginal or systemic re-challenge. B-cell numbers increased in the vaginal mucosa from day 1-7 after primary infection, while similar increases in B220(+), CD4(+) and CD8(+) lymphocytes in the draining lymph node were delayed by 48 h. By day 4 postinfection, a MHC-II(+) dendritic cell population became depleted from the stroma and formed aggregates below the basement membrane at points of demonstrable epithelial infection. In contrast, primary systemic or intradermal inoculation resulted in 80-90% mortality, but also conferred protective sterile immunity to vaginal West Nile virus re-challenge. Intravaginal and intradermal immunization elicited comparable, accelerated accumulation of larger B-cell numbers in the mucosa and draining lymph node upon intravaginal re-challenge than systemic immunization. However, accumulation of CD4(+) T cells in both sites in the intradermally immunized group was significantly greater than in intravaginally or systemically immunized mice. Accelerated accumulation of dendritic cells occurred at periodic sub-basement membrane sites in the absence of detectable virus 1 day after vaginal re-challenge, irrespective of the route of immunization. These data illustrate the diversity of possible effective immune responses to West Nile virus in the vaginal mucosa. They show primary vaginal inoculation produces effective immunity to flavivirus infection with lower mortality than other routes and suggest a local role for vaginal mucosal dendritic cells in both primary and secondary responses.

Candida albicans is a common opportunistic pathogen, causing both superficial and systemic infection. Clinical observations indicate that mucocutaneous infections are commonly associated with defective cell-mediated immune responses, whereas systemic infection is more frequently seen in patients with deficiencies in neutrophil number or function. Analysis of mechanisms of host resistance against gastrointestinal and oral infection in mouse models has demonstrated an absolute dependence on CD4(+) T cells, although clearance also involves phagocytic cells. Both IL-12 and TNF-alpha appear to be important mediators, but mouse strain-dependent variations in susceptibility to infection may be related to T-cell enhancement of production of phagocytic cells by the bone marrow. In murine systemic infection, the role of innate and adaptive responses is less well defined. Studies in immunodeficient and T-cell-depleted mice suggest that clearance of the yeast may be predominantly a function of the innate response, whereas the adaptive response may either limit tissue damage or have the potential to cause immunopathology, depending on the host genetic context in which the infection takes place.

Recently a new model of vertebrate immunity has been gaining popularity. In this new model it is hypothesized that activation of innate immunity is a prerequisite for an adaptive immune response to an antigen. Following activation the innate system induces key costimulator molecules on APC, which are essential for antigen-driven clonal expansion of T and B cells. The model largely explains the need for adjuvants in the induction of adaptive immunity, provides a possible mechanism for the immune system to perceive the biological nature of a pathogen and thereby produce the most effective immune response, and transfers much of the onus of self-non-self discrimination from the adaptive to the innate immune system. In the present article we highlight two paradoxes raised by the new model. First, by linking adaptive immunity to innate recognition the immune system is unable to take full advantage of the genetic diversity of T and B cell antigen receptors. Thus, the ability of the immune system to combat a pathogen is totally dependent on the efficiency of recognition by the innate system and, therefore, the germ-line mutation rate of the genes involved in the innate response. Second, if signals from the innate system induce costimulatory molecules on APC, then one would expect the accidental clonal expansion of many autoreactive T and B cells. We suggest that one means of resolving the first paradox is to propose that the major reason for the evolution of adaptive immunity was to provide, via immunological memory, resistance to reinfection, rather than simply to combat the primary infection by the pathogen. In the case of autoreactivity we suggest that autodestruction is prevented by immune responses being tightly regulated at the effector T cell level. Finally, we argue that the two paradoxes, rather than undermining the new model of immunity, highlight our lack of understanding of key elements of the vertebrate immune system.

The three modes of antibody production, natural, T independent (TI) and T dependent (TD) are conserved among vertebrate species suggesting an important role for each in protection against pathogens. Here, I use an artificial 'universe' to argue that the three modes of antibody production represent layers that evolved to deal optimally with antigens of different valence. Thus, the apparently more sophisticated TD response has not superseded the natural and TI components of the humoral immune response. Furthermore, the characteristic differences in isotype, somatic mutation and memory displayed by each antibody layer are appropriate for their targeted range of surface structures. It is also suggested that the TD and TI activation arms are at the extremes of a continuum, with signal integration of antigen and T cell-derived signals contributing to B cell decisions about isotype selection, proliferation and secretion that minimize the time to protection.