Painful memories boost protective immunity

How does pain, induced by inflammation, an innate immune response to tissue injury, impact humoral immunity, an adaptive immune response to infection? In a recent study in Cell, Wu et al. show that TRPV1⁺ nociceptive sensory neurons detect the inflammatory lipid mediator PGE₂ and release CGRP to activate the CALCRL/RAMP1 heterodimeric receptor, adenylate cyclase and cAMP signaling on B cells to boost the humoral immune response.

The control of immune responses by short- and long-range neural reflexes and the reciprocal control of synaptic plasticity by local and systemic inflammation are important emerging concepts.¹ Wu et al.² now exploit advances in genetic reporters, lineage tracing, tissue clearing, optical imaging, gene editing, pharmacogenetic and chemogenetic approaches to show that transient receptor potential cation channel subfamily V member 1 (TRPV1)-expressing nociceptive neurons from the thoracic T8–T13 dorsal root ganglia (DRG) innervate the spleen and release the neuropeptide calcitonin gene-related peptide (CGRP) to activate B cells via the CALCRL/RAMP1 heterodimeric receptor. Specificity and memory are two hallmarks of adaptive immunity shared with the nervous system (Fig. 1). This elegant study adds to the growing literature showing how these system-level parallels converge to strengthen antibody responses and reinforce humoral immunity.

Fig. 1: A neuroimmune circuit links inflammation to boost humoral immunity.

The proposed model based on Wu et al. shows how the adjuvant alum may drive local production of PGE₂ in the spleen to activate TRPV1⁺ nociceptors to release CGRP and activate CALCRL/RAMP1 on B cells. This in turn activates adenylate cyclase to generate the second messenger cAMP and enhance B cell activation, GC responses, affinity maturation and the secretion of neutralizing antibodies.

Pain, which is mediated by nociceptive sensory neurons, is one of the four cardinal features of acute inflammation recorded as early as the 1st century AD by Celsus in his Encyclopedia de Medicina. So how is pain induced by inflammation, an innate immune response to tissue injury, related to humoral immunity, an adaptive immune response to infection? Wu et al. first showed that TRPV1-EGFP⁺CGRP⁺ nociceptive nerve fibers enter the hilum of the spleen via the CD31⁺ blood vessels and travel along the neurovascular bundle to innervate the white pulp and B cell follicles. Pharmacogenetic ablation of TRPV1⁺ nociceptors from the DRG with diphtheria toxin led to a small but statistically significant reduction in the splenic B cell response to intraperitoneal immunization with 4-hydroxy-3-nitrophenylacetyl hapten (NP) conjugated to keyhole limpet hemocyanin (NP-KLH). There was a reduction in the number of germinal center (GC) B cells, splenic plasma cells and antigen-specific memory B cells. Importantly, this was associated with lower titers of anti-NP antibodies, impaired affinity maturation and memory recall responses. The authors next injected the retrograde neuronal tracer pseudorabies virus into the spleen to show that the TRPV1⁺CGRP⁺ nociceptive nerve fibers originated from neurons in the left T8–T13 DRG. They performed intraganglionic injection of the plant toxin and ultrapotent TRPV1 agonist resiniferatoxin to abrogate nociceptors in the left T8–T13 DRG and confirmed that the nociceptive nerve originated from there. Injection of resiniferatoxin in the left, but not right, T8–T13 DRG impaired splenic GC and antibody responses. By contrast, specific chemogenetic activation of the TRPV1⁺ nociceptive neurons in the left T8–T13 DRG enhanced GC and antibody responses to NP-KLH.

So what are these nociceptors sensing? Wu et al. showed that TRPV1 channels are activated by the accumulation of prostaglandin E₂ (PGE₂) in the spleen following intraperitoneal immunization. PGE₂ is a pro-inflammatory lipid metabolite of arachidonic acid catabolism by the enzymes phospholipase A2 (PLA2) and cyclooxygenase-1 and -2 (COX-1 and COX-2). The cells that make PGE₂ and what triggers them are unclear, but presumably, PGE₂ production is driven by the adjuvant alum. It is also not clear how PGE₂ is sensed by the nociceptors and which PGE₂ receptors (EP1, EP2, EP3 or EP4), if any, are involved. Regardless, this results in the release of CGRP which binds to the CALCRL/RAMP1 receptor on B cells to activate adenylate cyclase and generate cAMP. In a tour de force, the authors generated mice with B cell-specific deletion of CALCRL and RAMP1 to show that this led to impaired GC and antibody responses. Notably, the rescue of cAMP signaling with the adenylate cyclase agonist forskolin restored antibody responses in CALCRL-deficient B cells. While these studies with model antigens have shown consistent effects on humoral immunity, the magnitude of the differences has not been large. In this regard, the protective effect of this PGE₂-CGRP-CALCRL/RAMP1 circuit in prime-boost vaccination with hemagglutinin against lethal infection with the PR8 strain of influenza A virus was impressive. Wu et al. continued to show that dietary capsaicin, a natural TRPV1 agonist, can also enhance splenic B cell responses and protective immunity against challenges with PR8 infection.

Taken together, these experiments provide convincing evidence of a local neuroimmune circuit involving the peripheral nervous system that integrates pain and inflammatory signals to boost humoral immune responses. An intriguing study by Tynan et al. using optogenetic activation of TRPV1 in the footpad showed that this also enhanced humoral immune responses in the spleen, although the precise neuroanatomic pathways involved are unclear.³ These data complement the data of cholinergic fibers in the splenic nerve that originate from corticotrophin-releasing hormone (CRH)-expressing neurons in the amygdala and paraventricular nucleus that integrate stress responses to boost splenic B cell responses.⁴ They demonstrate the importance of wholistic approaches that capture the in vivo context of dynamic cell-cell interactions in intact model organisms.⁵ While reductionist approaches, such as in vitro assays and 3D organoid cultures, have their place, they do not capture the entirety of the complex interactions between cells and their local microenvironment, many of which are unknown and therefore cannot be modeled ex vivo. The unexpected neuroimmune circuit between TRPV1⁺ nociceptors and B cells described by Wu et al. is a great example of the virtues of these in vivo models and approaches.

What are the implications of this work? Most human vaccines are administered by subcutaneous or intramuscular injection in the deltoids or buttocks, and not by intraperitoneal injection. However, sympathetic and somatosensory innervation of the skin-draining lymph nodes have also been described.^6,7 Could similar local neuroimmune circuits operate in the lymph nodes to enhance vaccination responses? Importantly, would activation of a similar PGE₂-CGRP-CALCRL/RAMP1 circuit, e.g., with dietary capsaicin, also improve the efficacy of these routine vaccines? In this regard, it is worth noting that there are drugs in clinical use that may interfere with the circuit. Steroids inhibit PLA2 and aspirin, and other non-steroidal anti-inflammatory drugs (NSAIDs) inhibit COX-1 and COX-2 enzymes. Interestingly, the use of antipyretics such as NSAIDs to prevent inflammation may be associated with impaired antibody responses to some, but not all, vaccines.^8,9 CGRP inhibitors are in clinical use to treat migraines and the evidence so far does not seem to indicate that they impair antibody responses, e.g., in response to the SARS-CoV-2 vaccine.¹⁰ Nevertheless, the interconnection between the nervous and immune systems is an exciting area that is developing rapidly and we eagerly await the answers to these and many other new questions.