Brain-Immune interactions in sleep

Abstract

This chapter discusses various levels of interactions between the brain and the immune system in sleep. Sleep-wake behavior and the architecture of sleep are influenced by microbial products and cytokines. On the other hand, sleep processes, and perhaps also specific sleep states, appear to promote the production and/or release of certain cytokines. The effects of immune factors such as endotoxin and cytokines on sleep reveal species specificity and usually strong dependence on parameters such as substance concentration, time relative to administration or infection with microbial products, and phase relation to sleep and/or the light-dark cycle. For instance, endotoxin increased SWS and EEG SWA in humans only at very low concentrations, whereas higher concentrations increased sleep stage 2 only, but not SWS. In animals, increases in NREM sleep and SWA were more consistent over a wide range of endotoxin doses. Also, administration of proinflammatory cytokines such as IL-6 and IFN-α in humans acutely disturbed sleep while in rats such cytokines enhanced SWS and sleep. Overall, the findings in humans indicate that strong nonspecific immune responses are acutely linked to an arousing effect. Although subjects feel subjectively tired, their sleep flattens. However, some observations indicate a delayed enhancing effect on sleep which could be related to the induction of secondary, perhaps T -cell-related factors. This would also fit with results in animals in which the Tcell-derived cytokine IL-2 enhanced sleep while cytokines with immunosuppressive functions like IL-4 and L-10 suppressed sleep. The most straightforward similarity in the cascade of events inducing sleep in both animals and humans is the enhancing effect of GHRH on SWS, and possibly the involvement of the pro-inflammatory cytokine systems of IL-1β and TNF-α.

The precise mechanisms through which administered cytokines influence the central nervous system sleep processes are still unclear, although extensive research has identified the involvement of various molecular intermediates, neuropeptides, and neurotransmitters (cp. Fig. 5, Section III.B). Cytokines are not only released and found in peripheral blood mononuclear cells, but also in peripheral nerves and the brain (e.g., Hansen and Krueger, 1997; März et al., 1998). Cytokines are thereby able to influence the central nervous system sleep processes through different routes. In addition, neuronal and filial sources have been reported for various cytokines as well as for their soluble receptors (e.g., Kubota et al., 2001a). Links between the immune and endocrine systems represent a further important route through which cytokines influence sleep and, vice versa, sleep-associated processes, including variations in neurotransmitter and neuronal activity may influence cytokine levels.

The ability of sleep to enhance the release and/or production of certain cytokines was also discussed. Most consistent results were found for IL-2, which may indicate a sleep-associated increase in activity of the specific immune system. Furthermore, in humans the primary response to antigens following viral challenge is enhanced by sleep. In animals results are less consistent and have focused on the secondary response. The sleep-associated modulation in cytokine levels may be mediated by endocrine parameters. Patterns of endocrine activity during sleep are probably essential for the enhancement of IL-2 and T-cell diurnal functions seen in humans: Whereas prolactin and GH release stimulate Thl-derived cytokines such as IL-2, cortisol which is decreased during the beginning of nocturnal sleep inhibits Th1-derived cytokines. The immunological function of neurotrophins, in particular NGF and BDNF, has received great interest. Effects of sleep and sleep deprivation on this cytokine family are particularly relevant in view of the effects these endogenous neurotrophins can have not only on specific immune functions and the development of immunological memories, but also on synaptic reorganization and neuronal memory formation (e.g., Rachal el al., 2001).