The 3 talks in this session all deal with aspects of immune response regulation and dysregulation that can be either protective or harmful in brain ischemia. As an introduction to this session, I will review several models of immune response regulation and relate these models to the upcoming talks.1 Under normal circumstances, antigen presenting cells (APCs) such as microglia, macrophages and dendritic cells are in a resting state. In this state, they tend to produce immunomodulatory cytokines such as transforming growth factor-β (TGF-β) and interleukin (IL)-10. TGF-β from the APCs and also from existing regulatory T cells (Treg) can induce naïve T cells to differentiate into TGF-β secreting Treg cells. IL-10 secreting APCs can induce naïve T cells to differentiate into T-regulatory type 1 (Tr1) cells that secrete IL-10 and do not express the transcription factor FoxP3.
The APCs can be provoked by a number of stimuli that include infectious agents, inflammatory mediators, and endogenous ligands. Once activated, the antigen presenting cells are poised to elicit an immune response.
The model for immune response regulation has undergone considerable evolutionary change over the past 40 to 50 years.2 In 1959 Burnet and Medawar proposed a “Self-Non-self Model” for regulation of the immune response in which each lymphocyte was viewed as having a single receptor that would react with a specific antigen or specific antigens. Because of the enormous number of lymphocytes, the vast array of potential antigens was considered to be covered and, when an antigen bound to its cognate receptor, the lymphocyte(s) would proliferate to form a clone that would initiate an immune response. In this model, autoimmunity was prevented by deletion of self-reactive lymphocytes at an early stage. Burnet and Medawar shared a Nobel Prize in 1960 for their work on these mechanisms. In subsequent years, however, it became increasingly clear that the original Self-Non-self Model did not explain accumulating observations. For this reason, there were a series of modifications of this original model. In 1989, Janeway proposed an “Infectious Non-Self Model” that had incorporated several new cells. The APC and the T helper cell were important components of this model. Initiation of the immune response involved activation of germ-line encoded pattern recognition receptors (PRRs) that would respond to pathogen-associated molecular patterns. When these PRRs were activated, antigen would be endocytosed and processed such that it could be expressed in major histocompatibility complex class 2 molecules on the APC surface and presented to T helper cells along with appropriate costimulatory signals. This would then activate an immune response.
However, this model also failed to completely explain various phenomena. For instance, if the inciting stimulus for the immune response had to be an infectious nonself antigen, how could a transplant be rejected, or how would autoimmunity develop. In 1994 Polly Matzinger proposed a new model in which the critical trigger for initiating an immune response was not nonself versus self but was instead a signal indicating that the cells were becoming stressed.2 The model was termed the “Danger Model” and it was predicated on the hypothesis that cells can release endogenous ligands for PRRs and that signaling through these receptors would then activate the immune response by activating APCs. This activation sequence was initially somewhat speculative, but in 1997 the Toll-like receptor was identified as a homologue of the Toll receptor in Drosophila.3 It was subsequently shown that these Toll-like receptors could react to endogenous ligands such as misfolded proteins that had their hydrophobic groups exposed, fragments of degraded matrix, molecules that have been altered by free radical attack, and heat shock proteins (in addition to many others).4 This model is important for immune responses in ischemia because it permits activation of immune mechanisms at a time when the cells are still potentially viable, but are undergoing some form of stress. Activation of the immune system under these circumstances could be protective as Michal Schwartz would affirm or potentially damaging as Kyra Becker’s work would support. In either case, immune system participation could be involved in modulation of ischemic damage rather than being relegated to the role of a garbage detail that cleans up a necrotic mess.
The activated APC can induce differentiation of effector T cells such as TH1 cells and initiate immune responses. It would not be beneficial to the host, however, for immune responses to activate unnecessarily. To hold these responses in check there is constitutive suppression of effector T cell formation by Treg and Tr1 cells. These cells can also deactivate macrophages and restore calm. If the immune-activating stimulus is strong enough, the APC can release IL-6, which is necessary but not sufficient to block the suppression by Treg and Tr1 cells.5 Ulrich Dirnagl will discuss poststroke immunodepression that increases the likelihood of infections in the days after the stroke. This central nervous system injury-induced immunodepression appears to be regulated by the β-adrenergic sympathetic nervous system. Blocking β-adrenoceptors with β-adrenergic blockers can correct this immunodeficiency. These mechanisms do not seem to involve IL-6. Instead they seem to involve direct effects on the APC in which several proinflammatory cytokines are suppressed and expression of an immunomodulatory cytokine, IL-10, is increased. Also, there are direct effects on type 1 T helper cells in which IL-2 and interferon-γ expression is suppressed.6 For Ulrich Dirnagl the activation of immune responses or direct treatment of bacterial infections would seem important to optimize the care of the stroke patient. Kyra Becker will view the problem from a somewhat different perspective. She will discuss the release of brain antigens across damaged blood-brain barrier in a stroke. These antigens are exposed to peripheral lymphoid organs that can then mount an autoimmune response against the ischemically injured brain. Her data indicates that this does not occur unless there is an additional inflammatory stimulus such as lipopolysaccharide to activate brain microglia. This response could be prevented by avoiding infection if possible and by increasing Treg and Tr1 cell immunomodulation if infection supervenes. Michal Schwartz looks at the optimal function of the immune system after stroke in still another way. She will stress that modulated activation of the immune system, with secretion of moderate levels of interferon-γ and other cytokines such as IL-4, can confer a state of “protective autoimmunity” and protect brain by increasing microglial uptake of glutamate and by increasing secretion of neurotrophic factors.
The interrelationships among these 3 presentations will be interesting and provocative.
- Received July 9, 2006.
- Accepted August 6, 2006.