Dopamine and norepinephrine are embracing their immune side and so should we

Abstract

While the history of neuroimmunology is long, the explicit study of neuroimmune communication, and particularly the role of catecholamines in neuroimmunity, is still emerging. Recent studies have shown that catecholamines, norepinephrine, epinephrine, and dopamine, are central to multiple complex mechanisms regulating immune function. These studies show that catecholamines can be released from both the nervous system and directly from immune cells, mediating both autocrine and paracrine signaling. This commentary highlights the importance of catecholaminergic immunomodulation and discusses new considerations needed to study the role of catecholamines in immune homeostasis to best leverage their contribution to disease processes for the development of new therapeutic approaches.

What makes an interaction neuroimmune?

Classically, neuroimmunology studies the interactions between the nervous system and the immune system as they relate to a variety of biological processes including development, response to insult and numerous homeostatic mechanisms. However, as our understanding of neuroimmunology expands to incorporate new systems and communication mechanisms, the definitions of what constitutes neuroimmunology must evolve as well. An example of this is the role of the catecholamine neurotransmitters dopamine (DA), norepinephrine (NE) and epinephrine (Epi) in neuroimmune communication.

Catecholamines were first discovered and defined as neurotransmitters, mediating neuron-to-neuron communication [1–3]. However, these molecules were soon associated with stress and immune activation, and discovery of the cholinergic anti-inflammatory reflex [4] demonstrated specific mechanisms by which vagus nerve endings could regulate immune function. More recent studies show that catecholamines can be produced, released, taken up and/or bound by immune cells. Thus, DA, NE, and Epi can be released and signal from catecholaminergic neurons but can also act exclusively within immune populations via both autocrine and paracrine signaling, with no neurons involved Figure 1. Paradoxically, if a signaling process could involve neurons but does not, or if the mechanism involves neurotransmitters but not neurons themselves, is it still a neuroimmune interaction? And more specifically, does the presence of catecholaminergic neurotransmitters in a signaling cascade make it neuroimmune, or should it just be considered another form of immunoregulation?

Figure 1. Sources and effects of catecholamines in immune regulation.

Catecholamines are generated from a variety of cell types in both the CNS and periphery and have been shown to mediate a multitude of immune functions. Only some of these sources and immunoregulatory effects are represented here. (1) In the CNS, released catecholamines facilitate bidirectional interactions between catecholaminergic neurons and CNS immune cells, such as microglia, that can modulate neuronal function and shape behavior. (2) In the periphery, catecholamine levels can be regulated by direct and indirect interaction with the vagus nerve, inducing changes in immune function, such as the cholinergic inflammatory reflex, in multiple organ systems. (3) Indirect changes in response to vagal stimulation are often mediated by production of catecholamines in non-immune cells, such as enterochromaffin cells or immune cells themselves. These effects can also occur in response to other stimuli. (4) Interaction with catecholamines drives autocrine and paracrine signaling within and between immune populations, regulating multiple immune functions in multiple types of immune cells.

Neuroimmunology and the role of catecholamines

Around 2000 years ago, the Roman scholar Aulus Cornelius Celsus linked the nervous and immune systems when describing the cardinal signs of the inflammatory response. These signs, calor (heat), rubor (redness), tumor (swelling) and dolore (pain), implicitly link stimulation of neuronal nociceptors to the immune response. Despite these implied neuroimmune connections, the foundation for the study of neuroimmune communication was described in the late 19th and early 20th centuries [5], linking sensory nerve innervation to skin inflammation [6] and connecting degenerative changes in the skin caused by varicella-zoster virus infection with neuronal dysfunction [7]. But the nature of neuro-immunoregulation remained enigmatic until the mid-late 20th century, with research into immune-driven neurodegeneration in spinal cord [8,9] and the role of stress in immune activation and susceptibility to infection [10–12]. The term neuroimmunology first appeared in PubMed in 1982, and catecholamines were first shown to be present in immune cells around a decade later [13,14].

The catecholamines, dopamine (DA), norepinephrine (NE) and epinephrine (Epi), are molecules comprised of a catechol ring connected to a terminal amine by an ethyl side chain. These molecules are synthesized from tyrosine, which is converted to the intermediate L-DOPA, which is converted to DA. Dopamine is the precursor of NE, which can then be converted to Epi [15]. In addition to their role in neuron-to-neuron communication, catecholamines, specifically norepinephrine (NE) and epinephrine (Epi), have long been associated a number of peripheral functions, particularly changes in blood pressure, cardiac activity and renal function [16]. In addition, a growing number of studies show the importance of catecholamines, in mediating complex neuroimmunologic regulation [17,18].

The nervous and immune systems work bi-directionally, in discrete neuro-immune units, forming complex, multicellular, tissue- and cell-type specific networks across the body [19,20]. Immune cells express receptors for all catecholamines as well as other neuronal communication factors such as substance P and calcitonin gene-related peptide (CGRP), and neurons express cytokine and chemokine receptors and respond to these substances [17,21–23]. Crosstalk mediated by the release, uptake, and binding of catecholamines and other factors between immune cells and adjacent nerve endings innervating organs “fine tunes” the local immune response. Similarly, long-range, systemic interactions between immune cells and the CNS coordinate multiple organs or the entire body to manage pathogenic insult [17,19].

Over the past two decades, numerous studies have shown novel and surprising roles for catecholamines in functional regulation throughout the immune system (Figure 1). This complexity is exemplified in a recent study using a mouse model of DSS-induced colitis, where systemic release of NE from the adrenal gland increased colonic inflammation, while local release of NE from sympathetic fibers within the colon attenuated clinical symptoms and reduced immune cell abundance. This is not mediated by direct effects on immune cells, but by NE driven downregulation of the MAdCAM-1 receptor on local endothelial cells, blocking immune cell trafficking into the colon [24]. The systemic effect of dopaminergic activation is seen in mice where repeated stimulation of D1-expressing neurons in the nucleus accumbens decreased peripheral tumor volume, potentially through suppression of splenic CD8+ T-cell exhaustion [25]. Similar long-distance effects are mediated by activation of dopaminergic neurons in the ventral tegmental area, enhancing the peripheral antibacterial immune response in both myeloid and lymphoid cells by increasing phagocytosis, bacterial killing, and cytokine release [26]. Chronic increases in Epi can activate of β2-adrenergic receptors (β2AR) on macrophages, which drives IL-6 production, increases neutrophil traffic to wounds and delays wound healing [27]. And NE activation of β2AR also changed inflammatory activity in human leukocytes and LPS-challenged mice, although here activation of β2AR attenuated the production of inflammatory mediators and reactive oxygen species and augmented anti-inflammatory IL-10 [28].

Catecholamines can have both systemic and local effects on immune function, although these effects are not uniform across distinct receptors, cell types or organ systems, supporting the breadth and complexity of their functions. For example, in human neutrophils, activation of β2AR, but not β1AR or β3AR, increases immune activity, driving reactive oxygen species and the formation of neutrophil extracellular traps [29]. In the murine lung, endogenous catecholamine production increases inflammation through activation of macrophage α2AR, but not α1AR or βAR [30]. And in rodent bone marrowederived macrophages, as well as rodent models of LPS-induced endotoxemia and DSS-colitis, NE activation of β2AR blocked secretion of inflammatory cytokines through rapid induction of IL-10 in innate immune populations [31]. These interactions can also allow immune cells modulate neuronal function through catecholaminergic signaling. For example, microglial phagocytosis of dopamine receptor 1 (D1) on nucleus accumbens neurons induces sex-specific differences in social development [32], and inflammatory changes in corticostriatal dopaminergic circuitry can drive changes in motor function and motivation [33]. While these are far from a comprehensive review of the literature, they provide a few examples of both the local and systemic crosstalk between the nervous and immune systems that are mediated by catecholamines.

Catecholaminergic immunoregulation through the vagus nerve

A classic example of complex catecholaminergic immunoregulation is the cholinergic anti-inflammatory reflex [4]. In this process, NE from vagus nerve endings activates adrenergic receptors on cholinergic splenic T-lymphocytes, releasing acetylcholine (ACh) that acts on macrophage α7 nicotinic acetylcholine receptors, reducing TNF-α production [20]. This mechanism of vagus nerve regulation likely occurs in many compartments, as several secondary lymphoid organs are innervated by sympathetic nerves that do not release ACh. In some regions, these effects may be mediated through β2AR [34], as NE activation of β2AR regulates toll-like receptor (TLR)-mediated inflammation in murine macrophages by increasing IL-10 [31], and activation of β2AR on CD4⁺/CD25⁻ regulatory cells T-cells (T_reg) is critical to regulation of serum TNF-α levels in a mouse model of sepsis [35]. And the vagus nerve may not be the only pathway by which adrenergic activity can regulate splenic inflammation. A recent study described an apical splenic anti-inflammatory pathway in which neuronal release of NE directly downregulates inflammatory activity by binding directly to myeloid, rather than T-cell, β2AR [36].

Further, neuroimmune activity driven by the vagus nerve is not restricted to adrenergic signaling, it can also be mediated by DA release. Vagal-stimulated DA production in the adrenal medulla suppresses systemic production of TNF-α, IL-6 and CCL2 in murine sepsis through activation of D1 [37]. The cell types involved in this process are unclear, but activation of D1-like receptors can decrease mast cell degranulation [38] and suppress LPS-mediated NLRP3 inflammasome activity in murine macrophages [39,40]. Along with many other studies, these data demonstrate that multi-step signaling processes initiated by nerve innervations that connect the CNS to different organ systems are key pathways for catecholaminergic neuroimmune regulation.

Endogenous catecholaminergic immunomodulation

Almost two decades ago, multiple groups showed that activation of D1-like receptors (D1 and D5) on human CD4+CD25+ T_reg modulates CD8+ T_eff cell proliferation and activity, increasing proliferation and differentiation of the inflammatory T_h17 CD4+ T-cells. These dopaminergic effects could potentially be mediated by vagally induced DA release, but in these studies T_reg activity was regulated by direct DA treatment or autocrine signaling via the endogenous DA release, modulating secretion of IL-10 and TGF-β [41,42]. These are part of a growing body of research demonstrating the importance of endogenous immune cell catecholamine production and signaling to catecholaminergic immunomodulation.

In T follicular helper (T_fh) cells release of stored DA upon interaction with B-cells maximizes T_fh-B-cell entanglement and accelerates B-cell output from germinal centers [43]. Autocrine stimulation of βAR with endogenous catecholamines increases B-cell IL-10 production, enhancing the suppressive capacity of regulatory B-cells [44]. In human macrophages, LPS-induced DAT-mediated DA efflux promotes autocrine DA signaling that then modulates the inflammatory response to LPS [45]. And in both murine and human macrophages, inhibition of catecholamine synthesis significantly reduces LPS-induced cytokine release mediated by autocrine activation of α1AR [46]. These studies demonstrate that endogenous immune catecholamines likely have a more complex and integral role in immune regulation than previously considered.

The DA and NE mediating these immunologic changes could be released from catecholaminergic neurons, but these effects are also mediated by autocrine and paracrine signaling within immune cells, with no neurons involved. As catecholamines are also produced by non-neuronal cells and are critical to a variety of non-neuronal functions [16,47], there have been suggestions that these molecules are effectively non-traditional cytokines [48]. However, involvement in non-neuronal communication does not override the role of catecholamines in neurotransmission, and many of their immunoregulatory functions involve the nervous system indirectly. Indeed, the same discussion could apply to cytokines, which are classically defined as small, pleiotropic signaling molecules released by immune cells [49]. Both neurons and other non-immune cells release numerous cytokines to modulate neurocircuits, regulate the release of classical neurotransmitters, and drive changes in neuronal function that mediate learning, reward, motor function and behavior, including in neuropsychiatric disease [33,50–52]. Thus, defining whether catecholaminergic immunomodulation should be considered neuroimmune communication is complicated, depends on the mechanism, disease and/or environment being studied. Further, reconsidering and expanding our view of the role of catecholamines in immune function is likely to be more fruitful than defining the specific instances in catecholaminergic immunomodulation is or is not neuroimmune.

Reconsidering catecholaminergic neuroimmune communication

The expansive variety of factors and mechanisms that drive direct and indirect neuroimmune communication suggests the definition of neuroimmunity be broad and feature continual reevaluation of the communication mechanisms and cell types acting within various organ systems. For example, while the gut has long been a site of active immunologic inquiry, the importance of bidirectional gutebrain axis communication in both gastrointestinal and CNS immune response and disease is emerging. Gut-brain crosstalk regulates immune activity that safeguards the maintenance of gastrointestinal homeostasis, and is implicated in higher cognitive functions, affect, motivation and disease states such as Parkinson’s disease [53,54]. Rapid activation of catecholaminergic neurons in the gut in response to luminal infection triggers NE activation of β2AR on muscularis macrophages in the gut lumen, driving neuroprotective functional changes. Notably, these same changes were not seen in nearby lamina propria macrophages, suggesting intra-tissue macrophage specialization and cell-type specific responses to catecholaminergic stimulation [55,56]. Further cell-specific effects in the gut could be mediated by unique catecholamine receptor oligomers, such as the D5:CCR9 heteromer that drive gut inflammation by increasing colonic tropism of CD4+ T-cells [57].

While of these processes can be attributed to cytokine-driven vagal signaling [58], or activation of enteric neurons, another possibility is that the numerous bacterial species in the human gut could directly influence neuroimmune communication via release of cytokines and neurotransmitters, including DA and NE [59–61]. Germ-free mice show reduced gut NE levels that were restored by colonization with a mixture of Clostridia species [62], and clearance of gut microbes that reduced GI dopamine synthesis exacerbated invariant NKT cell mediated hepatitis [60]. The interface between enteric microbiota and vagus nerve activity provides a clear pathway for enterochromaffin-cell signaling to influence gut neuronal circuits, immune response modulation, and other homeostatic functions [53,63]. Thus, within the GI system, there are multiple mechanisms by which the neurotransmitter dopamine could influence gut-associated pathologies, and not all of them involve neuronally produced catecholamines. This is just one example of the type of broader consideration that should be applied to the study of catecholaminergic ?neuro?-immunoregulation moving forward.

Challenges associated with the study of catecholaminergic immunoregulation

Defining the molecular mechanisms that regulate the wide array of catecholaminergic immune processes will require physiologically relevant models of these interactions. This necessitates a number of specific considerations (Table 1), such as evaluation of the concentrations of catecholamine and receptor agonists being used, as well as consideration of cell type specific differences in receptor expression. Catecholamines differentially activate adrenergic or dopaminergic receptors, as each distinct receptor (α1 and 2AR, β1, 2 and 3AR and D1 – 5) has a different affinity for its cognate ligand and different agonists and antagonists. For example, both types of adrenergic receptors have a higher affinity for Epi than NE, and while Epi has similar affinity for β1AR and β2AR, NE has 10-fold selectivity for β1AR over β2AR due to differences in the binding pocket. Epinephrine also has a higher affinity for α-AR over β-AR, and DA has a much higher affinity for D3, D4 and D5 relative to D1 or D2 [64–66]. A higher concentration of catecholamines or their ligands could activate lower affinity receptors, and use of too little might bias the activation toward higher affinity receptors.

Table 1.

Considerations in the study of catecholaminergic immunomodulation.

1	Defining differences in cell-, organ- and species-specific expression of catecholaminergic receptors and transporters
2	Considering how activation of the same catecholamine receptors can produce discrete effects on immune function in different species, cell types or disease states, including in the presence of inflammation
3	Using physiologically relevant concentrations of catecholamines to which immune cells being studied could be exposed
4	Evaluating and discussing how ratios of different types of catecholamine receptors influence immune function
5	Examining how discrete catecholamine concentrations may produce differential effects on the same receptors depending on receptor affinity

The effect of differential concentrations on receptor activation is not well studied. But in ex vivo treatment of peripheral human blood, physiologic concentrations of NE did not affect LPS-induced TNF-α production, but supraphysiologic levels of NE decreased production of this cytokine [67]. Similarly, lower DA concentrations have been suggested to drive inflammation in schizophrenia by selectively stimulating high-affinity dopamine receptors (D3 and D5), while high DA levels stimulate low-affinity dopamine receptors (D1 and D2) to reduce inflammation [68,69]. Supraphysiologic levels of catecholamines could also result in ligand promiscuity, as both NE and Epi can also bind D2-like receptors [70], and higher DA concentrations activate adrenergic receptors [71–73]. Thus, despite the potential selectivity provided by active regulation of catecholamine levels by biogenic amine transporters, dopaminergic and adrenergic signaling may overlap. This could potentially allow different catecholamines to influence each other’s signaling, act in concert and/or influence discrete functions through cross activation of a greater variety of receptors. Indeed, treating macrophages and keratinocytes with the β-AR antagonist propranolol abrogates DA-mediated increases in NF-κB, IL-6 and IL-8, while also blocking DA-mediated IL-12p40 suppression [74,75]. Finally, due to its capacity to form DA quinones and other ROS, higher levels of DA could also non-specifically modulate immune cell function through cytotoxicity and oxidative stress [76].

However, determining the specific catecholamine levels to use is challenging, as concentrations in specific organs, particularly in humans, are not well defined. There are several studies describing the concentrations of NE and Epi in the blood [77,78], but far less research on the levels of these catecholamines in the brain or peripheral organs. Further, concentrations vary widely with detection technique and often do not account for conjugation and metabolism [16]. Our recent review discusses the DA concentrations in peripheral and CNS compartments, but even this comprehensive evaluation of the data shows a wide range of DA concentrations [23]. Cell-type specific differences in receptor expression can influence immune function as much as catecholamines levels and should also be considered during experimental design. Receptor expression, particularly the ratio of expression between different types of receptors, may differ between primary cells and cell lines. For example, in human THP-1 myeloid cells, there are higher relative levels of D4 and lower relative levels of D1 and D5 as compared with primary human macrophages [79,80]. This does not imply that cell lines should not be used, only that the experimental design and interpretation consider their specific catecholamine receptor composition. Using primary cells avoids this issue, but still needs careful consideration, as the human population is heterogeneous and there is substantial variation in immune catecholamine receptor expression between individuals [80]. Further, changes in relative levels of dopamine or adrenergic receptor expression can differentially affect cytokine production and other functions [72,81].

Catecholamine receptor expression and function also varies with environment, particularly exposure to inflammatory stimuli, with disease state, between neurons and immune cells and among immune cell subsets. Both dopaminergic and adrenergic receptors are expressed differentially on distinct T-cell subsets [41,82–84], B- and T-cell dopamine and adrenergic receptor expression changes in response to inflammatory or activating stimuli [44,85,86]. In macrophages under inflammatory conditions, such as exposure to LPS or cardiac injury, activation of D1-like receptors inhibits NLRP3 activity and reduces inflammation [40,87]. However, in unstimulated macrophages, D1-like receptor activation activates NF-κB and primes the NLRP3 inflammasome [88,89]. In human peripheral blood cells, Epi dose dependently increases CXCL8 production in unstimulated cells, but suppresses production of IL-1β, IL-8, and MCP-1 induced by LPS [90]. Variation in receptor signaling is also seen in immune cells relative to neurons, as dopamine receptors in human macrophages do not mediate signaling via cAMP as they do in neuronal systems [91]. Further, while not immune cells, both human and mouse pancreatic α- and β-cells show species-specific differences in the ratios of dopaminergic and adrenergic receptor subtypes, as well as differences in dopamine-mediated changes in glucagon secretion that may result from differences in the ratios of receptors with different DA affinities [72]. These concepts highlight the importance of carefully defining neuroimmune changes using multiple experimental modalities and across several systems to confirm their translational potential.

Conclusions

Defining the complex and multifaceted roles of catecholamines in mediating the neuroimmune response is still in its infancy. A growing amount of data shows that catecholamines are critical immunoregulatory factors, are central to a broad array of inflammatory diseases and hold tremendous therapeutic potential. While DA, Epi and NE are used directly to treat sepsis, anaphylaxis, cardiac arrest, hypotension, and several other acute conditions, we cannot yet effectively leverage vagal stimulation or catecholamine receptor activation to more precisely influence neuroimmunomodulation.

To do this, we need to better understand the nuance in these systems, avoiding issues associated with overstimulation or activation of the wrong receptor. Using a broad range of concentrations for catecholamine or agonists will guard against results stemming from extrap-hysiologic concentrations, while reporting functional results in the context of receptor expression will give critical perspective on the discrete receptors and combinations of receptors that could trigger each function. These can be combined with new technologies enabling better evaluation of dopaminergic and adrenergic receptors and real-time analysis of catecholamine release and uptake to provide a more precise and accurate picture of how much catecholamine is stimulating which receptor. In the context of in vitro systems using organoids or tractable human iPSCs, or with rodent systems using precision opto- and chemogenetic tools, these studies will enable a sophisticated, and accurate understanding of the molecular mechanisms driving catecholamine changes in immune function. This will create a robust foundation for catecholamine-based therapeutics that precisely target specific receptor systems on discrete cell types and organ systems to ameliorate pain, inflammation, and disease.