Low levels of circulating inflammatory cytokines—Do they affect human brain functions?

https://doi.org/10.1016/S0889-1591(02)00004-1Get rights and content

Abstract

Animal studies provide consistent evidence for the pivotal role of inflammatory cytokines in inducing sickness behavior during systemic infection and inflammation. Because depression in humans shows a considerable symptomatic overlap with sickness behavior, it has been hypothesized that cytokines are also involved in affective disorders. This view is supported by studies showing that therapeutic administration of inflammatory cytokines can induce typical major depression and by evidence that stimulated cytokine-release during experimental endotoxemia provokes transient deterioration in mood and memory. However, in these conditions, similar to the animal models of acute infections, huge amounts of cytokines produced in the periphery act on the brain. In contrast, during most clinical conditions where depression might involve cytokine actions, such as chronic infection and inflammation, only low amounts of cytokines are circulating. The present paper addresses the question whether and how low amounts of circulating cytokines act on the human brain. Evidence is presented that very low amounts of circulating cytokines are likely to influence brain functions, even under baseline conditions. It is also likely that low levels of cytokines affect the same brain function as high levels do. However, it is uncertain whether these effects go in the same direction. NonREM sleep, for example, is promoted by a slight increase in cytokine levels, but suppressed by prominent increases. Because no comparable data are available for mood and other brain functions, the answer to the question whether and how low circulating amounts of cytokines affect mood depends on further experimental studies.

Introduction

One of the major milestones in psychoneuroimmunology was the finding that peripherally administered interleukin-1 (IL-1) activates the hypothalamo–pituitary–adrenal (HPA) system (Besedovsky, del Rey, Sorkin, & Dinarello, 1986). This outstanding discovery strongly supported the concept developed years before that central nervous system (CNS) host response components of peripheral infection and inflammation are induced in a complex way by inflammatory cytokines such as IL-1, which are humoral factors produced by immunocompetent cells. The number of cytokines involved in brain–immune interactions is huge (see Hopkins & Rothwell, 1995; Rothwell & Hopkins, 1995) and includes such diverse molecules as leptin and nerve growth factor. The present paper is restricted to ‘classic’ cytokines such as interleukins, interferons (IFN), and the tumor necrosis factor (TNF) family because the effects of some of these have been investigated in humans, and the mechanisms underlying effects on complex brain functions have been extensively studied in animals. The induction of fever and neuroendocrine activation, in particular of the HPA system, are the most intensively studied CNS-mediated host responses (Kluger, 1991; Turnbull & Rivier, 1999). Based on these studies, a number of groups have investigated the role of cytokines in complex brain functions, in particular behavior, during experimental models of infection and inflammation. In rodents, it turned out that—mediated by cytokines—a typical pattern of behavioral responses occurs, which has been collectively termed sickness behavior (Dantzer, 2001; Yirmiya et al., 2000) and includes decreases in general activity, in exploratory behavior, social and sexual interaction, food and water intake, preference for sweets (anhedonic behavior), altered sleep, and impaired learning. These behavioral responses are induced by inflammatory cytokines released peripherally, such as IL-1β, IL-6, and TNF-α. For details of the underlying mechanisms, the interested reader is recommended to refer to recent reviews (e.g., Rothwell & Hopkins, 1995), but it should be noted that three major pathways are involved which include (i) passive diffusion at certain areas and active transport of cytokines in the brain, (ii) activation of neural afferents by cytokines, and (iii) self-induction of cytokine synthesis within the brain, mainly by microglial cells.

It can be assumed that these pathways are also functional in humans. Hence, cytokine signals from the periphery are very likely to modulate complex human CNS function. This view is supported by numerous studies showing that the therapeutic administration of cytokines can induce depressive symptomatology, which widely overlaps with the syndrome of ‘sickness behavior’ observed in animal models of infection and inflammation (Yirmiya et al., 2000). Moreover, we showed very recently that cytokine release during experimental endotoxemia in healthy humans correlates with the amount of transient deteriorations in mood, memory, and appetite (Reichenberg et al., 2001). However, in these clinical situations and during acute febrile infections, the amounts of circulating inflammatory cytokines are huge, usually two orders of magnitude or more above baseline levels, e.g., in the range of several picograms for IL-6 or TNF-α (Gudewill et al., 1992; Haack et al., 2000). In contrast, circulating levels are only slightly or moderately increased in the most frequent clinical situations where cytokines might play a role in inducing symptoms of depression, such as chronic infection or inflammation, cancer, cardiovascular disease, and autoimmune disorders (Yirmiya et al., 2000). Independent of these disorders involving definite immunopathology, cytokines have been even hypothesized to be involved in major depressive disorders occurring in otherwise healthy people (Smith, 1991). Data on a presumed increase of circulating cytokine levels are conflicting (for a detailed discussion of this issue see Haack et al., 1999), but if these levels are above those found in healthy people, they clearly remain some orders of magnitude below the levels found during acute infection and inflammation.

The present paper will address the questions whether or not low circulating levels of inflammatory cytokines influence the human brain function. We will argue that there is evidence to suggest that even in healthy individuals cytokines from the periphery affect the brain, in particular sleep–wake behavior, although this evidence is not yet robust enough. We will furthermore argue, again based on data about the influence of cytokines on sleep, that the quantitative relationship between certain brain functions and levels of circulating cytokines depends on the function considered. This relationship may be linear (for example, increasing cytokine levels go along with a linear suppression of rapid-eye-movement [REM] sleep) or complex (for example, nonREM sleep amount shows a bell-shaped relationship to the circulating amounts of cytokines). Unfortunately, there are no data on the quantitative relationship between other brain functions of interest (i.e., mood, anxiety, memory, etc.) to cytokine levels. Hence, it is premature to extrapolate from studies on depressive syndromes during acute experimental host defense activation, or following the administration of therapeutic doses of cytokines that low amounts of circulating cytokines would have qualitatively the same effects. Rather, the present knowledge urgently calls for experimental studies in this field.

Because we attempt to address these questions from a conceptual point of view, we talk about ‘inflammatory cytokines’, although this expression is somewhat imprecise and although we are well aware that IL-1, IL-2, IL-6, interferons, and TNF-α are likely to have differential effects on the brain. Moreover, for the present purpose we disregard any differentiation between direct and indirect effects of cytokines.

Section snippets

Under which circumstances do cytokines appear in the circulation?

Although acute infection and inflammation are the most evident situations involving effects of cytokines on the brain, there are a range of other clinical conditions where peripheral cytokine signals might modulate complex human brain functions. Table 1 lists a selection of such conditions. Conditions involving primarily cytokine production within the brain (e.g., multiple sclerosis and neurodegenerative disorders) are not included because they are beyond the scope of this article. The

What are the effects of low amounts of circulating cytokines on the brain?

In view of the findings that high amounts of inflammatory cytokines impair mood, cognition, sleep and appetite, the most obvious answer to the question heading this section would be: low amounts of cytokines would have the same effects, but to a lower extent, and below a certain threshold no effect would occur at all. Hence, the prediction from the effects of high doses of cytokines or endotoxin on mood would be that low amounts of cytokines induce depressive symptoms to a lower extent, up to a

Conclusion

In conclusion, the evidence presented suggests that inflammatory cytokines influence complex brain functions, even when the circulating levels are very low, possibly even in healthy individuals. Although at present this conclusion relies on a sufficient database only for sleep–wake behavior, it is likely that future studies will allow to generalize it to other complex brain functions such as mood, anxiety, memory, and others. However, the quantitative relationship between circulating cytokines

Acknowledgements

Thomas Pollmächer and Raz Yirmiya have been supported by a grant from the German–Israeli Foundation.

References (35)

  • T. Pollmächer et al.

    Influence of host defense activation on sleep in humans

    Adv. Neuroimmunol.

    (1995)
  • N.J. Rothwell et al.

    Cytokines and the nervous system II: actions and mechanisms of action

    Trends Neurosci.

    (1995)
  • R.S. Smith

    The macrophage theory of depression

    Med. Hypotheses

    (1991)
  • H. Besedovsky et al.

    Immunoregulatory feedback between interleukin-1 and glucocorticoid hormones

    Science

    (1986)
  • S. Gudewill et al.

    Nocturnal plasma levels of cytokines in healthy men

    Eur. Arch. Psychiatry Clin. Neurosci.

    (1992)
  • Haack, M., Schuld, A., Kraus, T., & Pollmächer, T. (2001). Effects of sleep on endotoxin-induced host responses in...
  • M.J. Kluger

    Fever: role of pyrogens and cryogens

    Physiol. Rev.

    (1991)
  • Cited by (179)

    • Stress, aging, and inflammation

      2023, Stress: Immunology and Inflammation: Handbook of Stress Series Volume 5
    • Biological motion during inflammation in humans

      2020, Brain, Behavior, and Immunity
    View all citing articles on Scopus
    View full text