Neuroimmunology of Sleep
Jeannine A. Majde and James M. Krueger
Sleep consumes a third of our lives, but its functions are unknown. Excess, inhibited, or fragmented sleep often accompanies psychiatric disorders or their treatments. The fatigue and confusion associated with the abnormal sleep may constitute a major complaint of the patient. Before we can begin to understand the pathological sleep associated with psychiatric disorders, it is essential that we have a better understanding of normal sleep and its functions. Ironically, the major insights we have gained into the regulation of normal sleep come from another class of pathologies, acute infectious diseases. This chapter will summarize what we have learned about sleep regulation through analysis of the molecular changes associated with infections.
Acute infections are generally detected clinically by the manifestations of fever and malaise. Included in the concept of malaise is the subjective feeling of profound fatigue and the overwhelming need to sleep. Sleep is detected and measured primarily through analysis of electroencephalographic (EEG) patterns, as has been described elsewhere in this text. Over the last 20 years we have characterized the EEG changes that occur in response to several acute infections. These changes will be described in more detail later in this chapter, but consistent features are the manifestation of increased slow-wave or non-rapid-eye-movement sleep (NREMS), increased slow-wave amplitudes, and often the reduction of total REM sleep (Krueger and Majde, 1994). These sleep characteristics are seen regardless of whether the stimulus is a purely microbial component or an actual infection. A major breakthrough in our conceptualization of sleep regulation is the realization that the peripheral cytokines produced in response to an infectious or other inflammatory stimulus are responsible for triggering the subjective need to sleep as well as the characteristic sleep changes. Furthermore, these same cytokines are actively involved in regulating physiological sleep.
Cytokines are a large class of protein hormones produced primarily by cells of the immune system or by damaged epithelial cells. Over 100 cytokines have been identified to date. One subclass of cytokines, the chemokines, appears to act primarily in a paracrine fashion to regulate inflammation at the site of tissue damage. However, a large and loosely defined subclass of cytokines, the proinflammatory cytokines, act not only locally but also systemically to trigger all of the characteristic responses to inflammatory challenge, including fever, anorexia, and somnolence. Thus, at least one function of proinflammatory cytokines appears to be signalling the brain that the host is threatened by invading micro-organisms and that adaptive responses are required (Hart, 1988). The best-characterized proinflammatory cytokines are the interleukin-1s (IL1α and ILlβ), IL6, and tumour necrosis factor-α (TNF-α). The array of physiological, behavioural, haematological, and biochemical responses initiated by these cytokines is termed the acute-phase response (APR). In addition to proinflammatory cytokines, growing evidence points to an important systemic role of anti-inflammatory cytokines such as IL10, IL4, IL13, and transforming growth factor-β (TGF-β) in the inhibition of the APR, including the excess sleep component. In the sections to follow, we will outline the evidence that proinflammatory and anti-inflammatory cytokines regulate both pathological and physiological sleep.
In addition to cytokines, several classical hormones, such as growth hormone-releasing hormone (GHRH), corticotrophinreleasing hormone (CRH), and prolactin, have been implicated in sleep regulation. Growth factors such as nerve growth factor (NGF), neurotropins, epidermal growth factor, and fibroblast growth factor are involved as well (reviewed by Krueger and Obál, 1997). The gaseous neurotransmitter nitric oxide (NO) has also been implicated (Kapás et al, 1994; Kapás and Krueger, 1996). The regulation of these hormones and NO synthesis by cytokines may form the basis for the association of cytokines and sleep.
STIMULATION IN THE HEALTHY INDIVIDUAL
It is widely recognized from such conditions as rheumatic heart disease that bacteraemia can occur in response to dental work. Profound immunodeficiency disorders such as AIDS have also demonstrated that the massive microbial flora that lines our mucosal surfaces, particularly the intestine, can escape protective mucosal barriers when normal immune defences are impaired. Much less is known about our day-to-day exposure to these micro-organisms because this exposure is not perceptible. However, studies of the intra intestinal lymphoid tissue, the Peyer's patches, reveal specialized epithelial cells that can transport bacteria to adjacent macrophages for degradation (Owen et al, 1986). Bacteria and their breakdown products accumulating in the Peyer's patches then can enter the portal circulation or the mesenteric lymph (Sartor et al, 1988). Though not quantified, it is thought that millions of bacteria may be cleared by this mechanism daily.
The first insights into the relationship of microbial products to sleep were gained from structural studies on a sleep factor isolated from human urine (Krueger et al., 1982a). This factor proved to share the unique chemical properties of the peptidoglycans found in the cell walls of bacteria (Krueger et al., 1982a). Peptidoglycans comprise 90% of the cell wall of Gram-positive bacteria and 5–20% of the cell wall of Gram-negative bacteria (Krueger and Majde, 1990). Subsequent sleep studies with various natural and synthetic peptidoglycans revealed that the minimally active unit is a glycopeptide consisting of the sugar N- acetylmuramic acid (found only in bacteria) and the dipeptide L-alamne-D-alanine, termed muramyl dipeptide (MDP) (Krueger et al., 1982b). MDP and certain derivatives have immunological