n e w s and vi e w s
Brainless immunity no more Sachin P Gadani & Jonathan Kipnis
I
n this issue of Nature Immunology, Kim et al. demonstrate that the brain can mediate the induction of adaptive immunity after infection and show mechanistically how this is achieved1. The innate immune response to Listeria monocytogenes results in upregulation of tumor necrosis factor (TNF) in the cerebrospinal fluid, which is produced in the brain ventricles and bathes the central nervous system (CNS). Kim et al. show that TNF is detected by the mediobasal hypothalamus (MBH), a circumventricular brain region that is not isolated from the circulation by the blood-brain barrier. The authors demonstrate that signaling through TNF receptors in the MBH sends a message to white-fat tissue via direct sympathetic innervation, which results in lipolysis. The products of this lipolysis—long-chain fatty acids such as palmitate or linoleic acid— increase the number of T cells and B cells both in the fat and in the spleen (Fig. 1). Kim et al. observe that increased amounts of TNF in the cerebrospinal fluid of mice, whether induced by direct injection or as a consequence of systemic infection with L. monocytogenes, lead to an increased number of lymphocytes in the spleen and in epididymal fat1. Furthermore, they find that knockdown of TNF receptors specifically in the MBH nullifies this effect, and re-expression of these receptors in the MBH of TNF receptor–deficient mice restores it. Intracerebral treatment with TNF is not followed by a change in the number of macrophages in the peripheral tissues or in the expression of genes encoding products related
Sachin P. Gadani and Jonathan Kipnis are in the Center for Brain Immunology and Glia, Department of Neuroscience, School of Medicine, University of Virginia, Charlottesville, Virginia, USA. e-mail:
[email protected]
440
to functions of the innate immune system but instead results in the induction of genes encoding lipolytic molecules in the fat and an increase in free fatty acids in the blood. Injection of long-chain fatty acids results in an increase in the number of T cells and B cells in the white adipose tissue and spleen, which is prevented by treatment with cerulenin, an inhibitor of fatty acid synthesis. Finally, the authors find that sympathetic denervation of the white fat diminishes the effect of brain TNF on the induction of T cells and B cells. These experiments clearly suggest a connection among the brain, fat and adaptive immune system and indicate a situation in which the brain controls peripheral immunity through lipolysis. The authors also investigate whether obesity might alter the brain–fat–immune system connection noted above. They find that under basal conditions, the concentrations of mRNA encoding TNF-response factors (Il1b, Il6, Socs3, Nfkb1a, Cxcl1 and Cx3cl1) are higher in the hypothalamus of mice with dietinduced obesity than in that of mice fed normal chow. Furthermore, the increase in white-fat and splenic T cells and B cells in response to intracerebroventricular injection of TNF is attenuated in the obese mice compared with that in normal chow–fed mice. This observation mechanistically links a high-fat diet to impaired adaptive immunity through an unexpected mediator—the hypothalamus—and could have major implications for understanding of the connection between inflammation and obesity2. A published study showing that microglia in the MBH become activated in mice fed a high-fat diet, which produces, among other things, increased TNF3, might also be relevant to these defects in MBH-to-fat signaling described by Kim et al. in obese mice1. This work raises several interesting questions for future research. Are other immune responses dependent on and controlled by the
brain? Does the brain also control immunological reactions in other tissues, including in the brain itself? Is TNF the only molecule of the innate immune system that mediates such an effect on the brain? Which cells in the MBH are involved in detecting TNF? What is the evolutionary advantage of such control? And do autoreactive immune responses also involve the brain? If the answer to the final question is yes, the role of the brain in organ-specific autoimmune diseases and in anti-tumor immunity should be explored further. Moreover, if it turns out that the phenomena described here operate in many inflammatory situations, such findings could point to the MBH as a therapeutic target for specifically dampening or enhancing the adaptive response without affecting the innate response. TNF
Sympathetic innervation White fat
Hypothalamus
Spleen Palmitate Linoleic acid
Lipolysis
Lymphocyte accumulation
Figure 1 The hypothalamus activates adaptive immunity through lipolysis. Kim et al. show that in mice, TNF, whether induced during infection or directly injected, is detected by TNF receptors located in the mediobasal hypothalamus1. This activates direct sympathetic innervation of peripheral white fat that leads to lipolysis. The products of this lipolysis, long-chain fatty acids such as palmitate or linoleic acid, act to increase the number of lymphocytes in the spleen and the white fat.
volume 16 number 5 May 2015 nature immunology
Katie Vicari/Nature Publishing Group
npg
© 2015 Nature America, Inc. All rights reserved.
The mediobasal hypothalamus detects increased amounts of tumor necrosis factor during the early phases of inflammation and relays this information to cells of the adaptive immune system by mobilizing free fatty acids.
npg
© 2015 Nature America, Inc. All rights reserved.
news and views Communication between the brain and immune system is a two-way street and has been documented in many physiological settings4. The brain provides adrenergic input for lymphoid tissues throughout the body and can directly control many aspects of inflammation through neurotransmitter receptors expressed on cells of the immune system. For example, vagus nerve activity dampens inflammation in splenic macrophages, decreasing their expression of genes encoding inflammatory molecules through T cell–derived acetylcholine, and also decreases antibody production by B cells5. Similarly, adrenergic stimulation of the lymph nodes can restrict T cell mobility and suppress inflammation6. GABA, a major inhibitory neurotransmitter, has been shown to control autoimmune responses7. As for the widely studied communication in the opposite direction—effects of the immune system on the brain—perhaps the example best understood is in sickness behavior. Here, peripheral cytokines signal to the brain through afferents of the vagus nerve, or directly to circumventricular organs, causing stereotypical changes in behavior during illness8. Numerous other examples demonstrate beneficial effects of the immune system on the CNS in the healthy state. For example, a normal T cell compartment is needed for proper spatial learning, and mice lacking T cells display cognitive impairment9. T cells mediating such pro-cognitive effects have been found in the meningeal spaces and exert their effects, at least in part, through secreted interleukin 4 (ref. 10). Cells of the immune system have also been shown to participate in maintaining adult neurogenesis, in promoting
appropriate stress responses, and in both protective and destructive and immune responses after injury11. Just as neurotransmitter receptors have abundant expression on cells of the immune system, cytokine receptors are widely expressed in the brain, and many of the brain’s functions depend on them. For example, TNF affects synaptic scaling12, and interleukin 1β has a direct effect on long-term potentiation13. Other molecules typically associated with the immune system, such as major histocompatibility complex class I, are also expressed in the brain and participate in normal development and function of the brain14. The article by Kim et al.1 prompted a flashback to my (J.K.’s) years in graduate school, during which my colleagues and I demonstrated that CNS-derived dopamine alleviates the suppressive function of regulatory T cells, allowing a more efficient effector T cell response15. From those and other studies we began to view the brain as a master organ that controls every aspect of body function, including the immune system. A fanciful experiment that would unequivocally prove the brain’s control over the immune system would be to immunize mice lacking brains and show that this immunization is less efficient than that in a mouse with a brain. Although this experiment is clearly not possible, the system used by Kim and et al. for specific deletion and addition of TNF receptors in the hypothalamus in the context of infection with L. monocytogenes1 is a clever approximation. The brain controls tissues through direct innervation, but less is known about how it regulates immunity. Can the brain distinguish between immune responses or types of
pathogens? For decades, neuroimmune interactions were thought to be inherently associated with pathology, and the conventional understanding that the CNS is an immunologically privileged organ meant that the many functional and bidirectional interactions between the brain and the immune system were overlooked. However, the notion that the brain is completely isolated from the immune system is fading, and the work of Kim et al.1 adds an example to an ever-growing list of physiological neuroimmune communication pathways, supporting the notion that instead of being carefully guarded from the immune system, the brain actually controls it. Competing financial interests The authors declare no competing financial interests. 1. Kim, et al. Nat. Immunol. 16, 525–533 (2015). 2. Gregor, M.F. & Hotamisligil, G.S. Annu. Rev. Immunol. 29, 415–445 (2011). 3. Valdearcos, M., Robblee, M.M., Benjamin, D.I., Nomura, D.K., Xu, A.W. & Koliwad, S.K. Cell Rep. 9, 2124–2138 (2014). 4. Steinman, L. Nat. Immunol. 5, 575–581 (2004). 5. Andersson, U. & Tracey, K.J. J. Exp. Med. 209, 1057–1068 (2012). 6. Nakai, A., Hayano, Y., Furuta, F., Noda, M. & Suzuki, K. J. Exp. Med. 211, 2583–2598 (2014). 7. Bhat, R. et al. Proc. Natl. Acad. Sci. USA 107, 2580–2585 (2010). 8. Dantzer, R., O’Connor, J.C., Freund, G.G., Johnson, R.W. & Kelley, K.W. Nat. Rev. Neurosci. 9, 46–56 (2008). 9. Ziv, Y. et al. Nat. Neurosci. 9, 268–275 (2006). 10. Derecki, N.C. et al. J. Exp. Med. 207, 1067–1080 (2010). 11. Kipnis, J., Gadani, S. & Derecki, N.C. Nat. Rev. Immunol. 12, 663–669 (2012). 12. Beattie, E.C. et al. Science 295, 2282–2285 (2002). 13. Schneider, H. et al. Proc. Natl. Acad. Sci. USA 95, 7778–7783 (1998). 14. Shatz, C.J. Neuron 64, 40–45 (2009). 15. Kipnis, J. et al. J. Neurosci. 24, 6133–6143 (2004).
Cytoplasmic methylation fuels leukocyte invasion Bernhard Wehrle-Haller The methyltransferase Ezh2, an epigenetic regulator associated with tumor-cell metastasis, also methlyates the cytoplasmic integrin adaptor talin. This modification inhibits the binding of talin to F-actin, which enhances the migration and invasion of dendritic cells and neutrophils.
T
he methyltransferase Ezh2, which is part of polycomb repressive complex 2 (PRC2), regulates gene expression through trimethy-
Bernhard Wehrle-Haller is in the Department of Cellular Physiology and Metabolism, University of Geneva, Centre Médical Universitaire, Geneva, Switzerland. e-mail:
[email protected]
lation of histone H3 at Lys27. These epigenetic modifications control normal stem-cell differentiation but are also associated with many tumors, where PRC2 activity stimulates epithelial-to-mesenchymal transitions, migration, invasion and metastatic spread1. PRC2-induced changes in cell adhesion and migration have so far been attributed to Ezh2-mediated transcriptional repression in the nucleus. However, Gunawan et al. now
nature immunology volume 16 number 5 MAy 2015
demonstrate that Ezh2-mediated methyltransferase activity in the cytoplasm controls integrin-mediated adhesion and migration through trimethylation of the integrin adaptor talin, blocking its binding site for the cytoskeletal protein F-actin2. The authors show that this activity reduces cell spreading but enhances the migration and tissue invasion of neutrophils and dendritic cells (DCs) in vivo and in vitro. 441
