204 | Nature | Vol 581 | 14 May 2020
Article
Brain control of humoral immune responses
amenable to behavioural modulation
Xu Zhang1,2,3,4,1 6, Bo Lei4,5,1 6, Yuan Yuan6,1 6, Li Zhang1,2,3,4,1 6, Lu Hu^7 , Sen Jin^8 , Bilin Kang4,5,
Xuebin Liao^7 , Wenzhi Sun9,1 0, Fuqiang Xu8,1 1,1 2, Yi Zhong4,5 ✉, Ji Hu6,1 3 ✉ & Hai Qi1,2,3,4,14,15 ✉It has been speculated that brain activities might directly control adaptive immune
responses in lymphoid organs, although there is little evidence for this. Here we show
that splenic denervation in mice specifically compromises the formation of plasma
cells during a T cell-dependent but not T cell-independent immune response. Splenic
nerve activity enhances plasma cell production in a manner that requires B-cell
responsiveness to acetylcholine mediated by the α9 nicotinic receptor, and T cells
that express choline acetyl transferase^1 ,^2 probably act as a relay between the
noradrenergic nerve and acetylcholine-responding B cells. We show that neurons in
the central nucleus of the amygdala (CeA) and the paraventricular nucleus (PVN) that
express corticotropin-releasing hormone (CRH) are connected to the splenic nerve;
ablation or pharmacogenetic inhibition of these neurons reduces plasma cell
formation, whereas pharmacogenetic activation of these neurons increases plasma
cell abundance after immunization. In a newly developed behaviour regimen, mice
are made to stand on an elevated platform, leading to activation of CeA and PVN CRH
neurons and increased plasma cell formation. In immunized mice, the elevated
platform regimen induces an increase in antigen-specific IgG antibodies in a manner
that depends on CRH neurons in the CeA and PVN, an intact splenic nerve, and B cell
expression of the α9 acetylcholine receptor. By identifying a specific brain–spleen
neural connection that autonomically enhances humoral responses and
demonstrating immune stimulation by a bodily behaviour, our study reveals brain
control of adaptive immunity and suggests the possibility to enhance
immunocompetency by behavioural intervention.We developed a surgical denervation protocol by treating splenic nerve
plexuses with alcohol before they enter the spleen along the vasculature
(Extended Data Fig. 1a–c). The tyrosine hydroxylase-containing fibres
were normally seen in the T cell zone, the T–B border and bridging chan-
nels, but disappeared after denervation (Fig. 1a, Extended Data Fig. 1d–g).
Mice with denervated spleens were grossly normal and survived for
as long as did sham-operated mice. Norepinephrine levels in spleen
homogenates were reduced after denervation (Extended Data Fig. 1h).
No abnormal lymphocyte apoptosis was seen in denervated spleens
(Extended Data Fig. 1i). Six weeks after surgery, we immunized dener-
vated and sham-operated mice intraperitoneally with 100 μg NP-KLH
(4-hydroxy-3-nitrophenylacetyl hapten conjugated to keyhole limpet
haemocyanin) in alum plus 1 μg lipopolysaccharide (LPS) and char-
acterized germinal centre (GC) development and splenic plasma cell
(SPPC) formation 7 or 13 days later. The gating strategy (singlets→live
events→total B-lineage cells) is shown in Fig. 1b. To quantify frequencies
of GCs and SPPCs, we used total B-lineage cells including both CD19+
cells and CD138+ plasma cells for normalization. We did not detect sig-
nificant differences in GC formation between the two groups (Fig. 1c).
From day 7 to 13, the SPPC frequency in total B-lineage cells increased
in sham-operated but not denervated mice (Fig. 1d). Notably, when the
same mice were immunized with the T-independent antigen NP-Ficoll,
we found no difference in SPPC formation (Extended Data Fig. 1j, k).
Given the comparably normal tissue anatomy, the GC response, and
T-independent SPPC formation in denervated mice, it is not likely that
defects in the T-dependent response are caused by disrupted organ
physiology. Instead, splenic nerve activity might specifically regulate
T-dependent SPPC formation.https://doi.org/10.1038/s41586-020-2235-7
Received: 7 December 2018
Accepted: 20 March 2020
Published online: 29 April 2020
Check for updates(^1) Tsinghua-Peking Center for Life Sciences, Tsinghua University, Beijing, China. (^2) Laboratory of Dynamic Immunobiology, Institute for Immunology, Tsinghua University, Beijing, China.
(^3) Department of Basic Medical Sciences, School of Medicine, Tsinghua University, Beijing, China. (^4) School of Life Sciences, Tsinghua University, Beijing, China. (^5) McGovern Institute of Brain
Research, Beijing, China.^6 School of Life Science and Technology, ShanghaiTech University, Shanghai, China.^7 School of Pharmacological Sciences, Tsinghua University, Beijing, China.
(^8) Shenzhen Institutes of Advanced Technology, Chinese Academy of Sciences, Shenzhen, China. (^9) School of Basic Medical Sciences, Capital Medical University, Beijing, China. (^10) Chinese
Institute for Brain Research, Beijing, China.^11 Centre for Brain Science, State Key Laboratory of Magnetic Resonance and Atomic Molecular Physics, Key Laboratory of Magnetic Resonance in
Biological Systems, Wuhan Institute of Physics and Mathematics, Wuhan, China.^12 Centre for Excellence in Brain Science and Intelligent Technology, Chinese Academy of Sciences, Wuhan,
China.^13 Co-Innovation Center of Neuroregeneration, Nantong University, Nantong, China.^14 Beijing Key Laboratory for Immunological Research on Chronic Diseases, Tsinghua University,
Beijing, China.^15 Beijing Frontier Research Center for Biological Structure, Tsinghua University, Beijing, China.^16 These authors contributed equally: Xu Zhang, Bo Lei, Yuan Yuan, Li Zhang.
✉e-mail: [email protected]; [email protected]; [email protected]