Interactions between the mind and the body have sparked the interest of scientists and philosophers for centuries. In ancient Greece, the physician Galen described the spleen as being the source of black bile, which was thought to cause melancholy when secreted in excess. Today, research is uncovering complex ways in which the brain and body interact to affect diverse aspects of health, from mood to immune function. The spleen aids immune defences by functioning as part of the lymphatic system; the organ is a major hub of activities needed to initiate responses in the adaptive branch of the immune system, which handles defences that are tailored to a specific disease-causing agent.
The spleen is a target of top-down control from the brain1. Writing in Nature, Zhang et al.2 have taken our understanding of brain–spleen connections to the next level by revealing an aspect of top-down control that regulates the adaptive immune system.
The spleen’s contribution to immune responses occurs mainly in its white-pulp region, where immune cells that have arrived from elsewhere in the body present peptide fragments called antigens to immune cells called T cells. If a T cell binds to and recognizes such an antigen, which might indicate the presence of an abnormal cell or a foreign invader, this activates the T cell, which in turn activates immune cells called B cells. B cells differentiate to form plasma cells (Fig. 1) that secrete antibodies specific for the antigen presented, and these antibodies are released into the bloodstream to fight infection3.
Spleen activity is controlled by the autonomic nervous system — a part of the nervous system that regulates organs. More specifically, the spleen is controlled mainly by the sympathetic branch of the autonomic nervous system, which is associated with the ‘fight-or-flight’ response4. However, little was known previously about possible upstream brain regions that might connect to the autonomic nervous system in the spleen to control it and, by extension, adaptive immunity. An earlier study in mice5 revealed that stimulation of a brain region called the ventral tegmental area, a part of the brain’s reward circuit, boosts immune responses and protection against harmful bacteria.
Zhang and colleagues developed a surgical technique to remove nerves from the spleen in mice. This mainly removed inputs from the autonomic nervous system and prevented top-down control from the brain to the spleen. After surgery, the animals were injected with an antigen. Plasma cells that made antibodies targeting that antigen arose in abundance in control mice that had undergone a ‘sham’ operation that did not remove nerves. Such an increase did not occur in the denervated mice, indicating that splenic-nerve activity regulates the formation of plasma cells and thus adaptive immunity.
The authors investigated which molecular mechanisms might be needed for plasma-cell formation in this context. They studied the expression of various types of receptor that can bind the neurotransmitter molecule acetylcholine, which is a key signalling component of the autonomic nervous system. Zhang et al. report that B cells express a type of acetylcholine receptor called a nicotinic receptor, and the authors pinpointed protein subunits of this receptor, including one called Chrna9. To test the role of nicotinic receptors containing Chrna9 in plasma-cell formation, Zhang et al. transplanted haematopoietic stem cells, which can generate immune cells, into mice that had undergone a treatment to remove their own haematopoietic stem cells. When the transplanted stem cells came from mice engineered to lack the gene encoding Chrna9, these animals generated fewer plasma cells after an injection of antigens than did animals that received antigen injections and transplants of stem cells with the gene intact. This result indicates that plasma-cell formation requires the presence of nicotinic receptors.
When a type of T cell called a CD4+ T cell is activated by antigen recognition, it secretes acetylcholine in response to the hormone noradrenaline6. The authors reveal that such T cells serve as a ‘relay’ between the release of noradrenaline from the splenic nerve and the subsequent acetylcholine-dependent6 formation of plasma cells (Fig. 1).
To map the neural circuit that connects the spleen and brain, the authors used a method termed retrograde tracing, which relies on monitoring the expression of a fluorescent protein encoded by a virus that can ‘jump’ across the synapses that connect neurons. This enabled Zhang and colleagues to track all upstream inputs to a given nerve cell in the spleen. The authors thereby identified two key brain regions (the central nucleus of the amygdala and the paraventricular nucleus of the hypothalamus) that contain neurons that connect to splenic nerves. These regions are major centres involved in the response to psychological stressors such as fear or threatening situations7, and they have essential roles in regulating the production of neuroendocrine hormones, for example, by a pathway called the hypothalamic-pituitary-adrenal axis8.
One population of nerve cells in these two regions releases the hormone corticotropin, which is thought to have a key role in initiating the body’s response to stress9. To determine whether corticotropin-producing neurons affect the spleen, Zhang et al. stimulated these neurons using a technique called optogenetics, and assessed whether this affected the activation of splenic nerves by monitoring their firing using electrophysiological recording. This provided crucial functional evidence for a brain–spleen connection, because such stimulation increased the firing of splenic-nerve cells. The authors also report that the inhibition or ablation of corticotropin-producing neurons in either of the two brain regions impaired the formation of plasma cells after antigen injection. Conversely, activation of the neurons stimulated such plasma-cell formation.
Although these circuit-based experimental approaches provide key proof for the existence of the brain–spleen axis, the authors also needed to test their model using suitable interventions that activate the ‘stress centres’ in the brain. However, neurons in the central nucleus of the amygdala and the paraventricular nucleus function in a pathway that causes the adrenal gland to secrete the hormone glucocorticoid in response to stress, and glucocorticoids are potentially immunosuppressive10.
The authors therefore considered whether the concentration of glucocorticoids secreted by the adrenal gland might depend on the severity of the stress. To avoid possible glucocorticoid-driven immunosuppression that might interfere with their analysis of antibody production, Zhang et al. studied mice that had been placed on an elevated, transparent platform; this provided a behavioural situation that induced only moderate stress. Following antigen injection, this scenario, but not another set-up that caused more-severe stress, led to the generation of antigen-specific antibodies. The authors showed that this antibody production depends on corticotropin-producing neurons in the brain circuit that they had described.
There is growing evidence that dysregulation of the immune system has a bottom-up role in promoting several behaviours relevant to neuropsychiatric disorders11. Zhang and colleagues’ study provides insights in the other direction — how the brain exerts top-down control of immune-system function. Future research will be needed to investigate whether this particular brain–spleen circuit exists in humans. The authors’ work opens up the exciting possibility that activating certain brain regions (through behavioural interventions or by selective stimulation using neuromodulatory techniques such as transcranial magnetic stimulation) could modulate the immune system. To return to Galen, he was right that the spleen is a key site of connection between the brain and the body, but his ideas about how the spleen induces melancholy now give way to this new perspective on how the mind might modulate resilience-promoting antibodies.
Nature 581, 142-143 (2020)
Mebius, R. E. & Kraal, G. Nature Rev. Immunol. 5, 606–616 (2005).
Zhang. X. et al. Nature 581, 204–208 (2020).
Nutt, S. L., Hodgkin, P. D., Tarlinton D. M. & Corcoran, L. M. Nature Rev. Immunol. 15, 160–171 (2015).
Jung, W. C., Levesque, J. P. & Ruitenberg, M. J. Semin. Cell Dev. Biol. 61, 60–70 (2017).
Ben-Shaanan, T. L. et al. Nature Med. 22, 940–944 (2016).
Rosas-Ballina, M. et al. Science 334, 98–101 (2011).
Davis, M. Annu. Rev. Neurosci. 15, 353–375 (1992).
Smith, S. M. & Vale, W. W. Dialogues Clin. Neurosci. 8, 383–395 (2006).
Peng, J. et al. Front. Neuroanat. 11, 63 (2017).
Coutinho, A. E. & Chapman, K. E. Mol. Cell. Endocrinol. 335, 2–13 (2011).
Cathomas, F., Murrough, J. W., Nestler, E. J., Han, M. & Russo, S. J. Biol. Psychiat. 86, 410–420 (2019).