Cross-Organ Neuroimmunology of Behavior

Abstract

A dynamic, bidirectional dialogue between the nervous and immune systems shapes behavior, with recent discoveries mapping discrete communication axes between the brain and peripheral organs. To unify these findings, we introduce the Cross-Organ Neuroimmunology of Behavior (CONB) Network, which is a systems-level framework that represents a paradigm shift beyond classical psychoneuroimmunology. This framework reconceptualizes behavior as an emergent property of a distributed, whole-body immune-brain network rather than the product of isolated organ-brain interactions. It builds on core molecular and cellular mechanisms of neuroimmune communication, ranging from cytokine-mediated modulation of neural activity and synaptic plasticity to neuroglial-immune cell interactions and neuroendocrine pathways, all of which constitute a shared language enabling cross-organ signaling. We examine how diverse peripheral organs (e.g., the gut, lungs, liver, and bone marrow) function as network nodes translating local immune or physiological changes into systemic signals that modulate brain circuits and behavior. Integrating these axes reveals emergent network properties, including redundant pathways (degeneracy) that confer resilience and vulnerable hub organs (such as the gut and bone marrow) that exert disproportionate influence on network stability. These properties provide a powerful model for understanding complex behavioral phenotypes and multi-system disease crosstalk, reframing neuropsychiatric and neurological diseases as systemic network dysregulations rather than purely brain-centric conditions. Ultimately, the CONB perspective informs precision medicine by leveraging immune biomarkers to identify patient subtypes (for example, distinguishing an “inflammatory” depression biotype) and guiding novel therapeutic strategies, such as bioelectronic neuromodulation (i.e., vagus nerve stimulation), microbiome-targeted interventions, and cytokine-targeted biologics, all aimed at recalibrating cross-organ neuroimmune networks and restoring system-wide homeostasis.

Introduction

The nervous and immune systems maintain a dynamic, bidirectional dialogue throughout the body, enabling the brain to “sense” and respond based on the status of peripheral organs, which in turn secrete factors that shape brain function and behavior profoundly. Classic “sickness behavior”—characterized by fatigue, anhedonia, and cognitive changes—illustrates how peripheral immune activation triggers central symptoms via cytokines, while neural signals emanating from the brain can conversely regulate the immune system. Recent advances have revealed discrete circuits and molecular mediators linking organs such as the gut, lungs, liver, heart, and immune reservoirs like the spleen and bone marrow to the central nervous system (CNS), emphasizing the role of both local and long-range neuroimmune interactions (Wolterink et al., 2022; Y. Zhu et al., 2022).

To unify these diverse discoveries, we propose the Cross-Organ Neuroimmunology of Behavior (CONB) Network—a conceptual framework for a distributed system where behavior emerges as a network-level output. In this framework, nodes are defined as discrete functional units, such as an organ (e.g., liver, lung), a tissue (e.g., bone marrow), or a specific cell population (e.g., microglia). These nodes are connected by edges, which represent the pathways of information flow, including physical connections such as neural circuits and chemical signaling via circulating immune mediators, such as cytokines, metabolites, or hormones. This concept extends classical psychoneuroimmunology by elevating cross-organ immunoregulation to a primary determinant of behavioral state. Within this network, specific nodes function as hubs, characterized by their high degree of connectivity (high “degree centrality”) and their critical role in integrating and distributing information. The gut and bone marrow, for example, represent such hubs; the gut integrates vast inputs from diet and microbiota into systemic immune and metabolic signals, while the bone marrow’s hematopoietic output dictates the immune tone of the entire organism. Perturbation at these hubs can thus disproportionately impact the stability and function of the entire network.

In this review, we first establish the molecular and cellular principles that enable the CONB Network (Foundational Mechanisms of Neuroimmune Communication). We then examine each peripheral organ axis—gut, lung, liver, heart, skin, muscle, bone marrow, and spleen—as functional nodes within this network (The Organ-Wide Neuroimmune Network). Finally, we integrate these axes to illustrate network-level dynamics and discuss emerging therapeutics that target the CONB Network as an integrated whole (Systems-Level Integration and Therapeutic Implications).

Foundational Mechanisms of Neuroimmune Communication

The classical conception of CNS function posits it is an immune-privileged organ operating through a specialized lexicon of neurotransmitters and neuropeptides, largely insulated from immunological insults in the periphery. This view, however, is being supplanted by a paradigm that recognizes a profound and continuous dialogue between the nervous and immune systems, one conducted in a shared molecular language (Castellani et al., 2023; Leunig et al., 2025). This common lexicon challenges the notion of systemic separation, revealing instead a deeply integrated regulatory network where molecules once considered exclusive to one system—such as cytokines, chemokines, and even components of the complement cascade—act directly on neural cells, while classical neural signals reciprocally modulate immune function. The biological meaning encoded by this language is not fixed; rather, it possesses a sophisticated grammar in which the context—including signal concentration, duration, and the specific cellular “listener” determines the ultimate physiological and behavioral output (Dantzer, 2018). Understanding this shared lexicon and its contextual syntax is fundamental to deciphering how peripheral states like infection, metabolic stress, or psychological trauma are translated into adaptive or maladaptive shifts in CNS function and complex behavior.

A primary consequence of this neuroimmune dialogue is the acute and pervasive regulation of brain-wide neurotransmission, a process that functionally reallocates the brain’s resources to orchestrate global shifts in behavioral state, such as the induction of sickness behavior. This acute modulation is remarkably comprehensive, targeting all aspects of neurotransmitter and neuropeptide signaling, from synthesis and release to reuptake and degradation. Pro-inflammatory cytokines, for instance, can systemically deplete the foundational precursor for serotonin synthesis by activating the indoleamine 2,3-dioxygenase (IDO) pathway, which shunts tryptophan toward the production of neuroactive kynurenines, while tumor necrosis factor-alpha (TNF-α) concurrently enhances serotonin reuptake via the serotonin transporter (SERT), a combination that potently contributes to the anhedonic and depressive phenotypes associated with inflammation(Correia & Vale, 2022; Dantzer, 2017). Simultaneously, these immune signals re-tune circuits governing motivation and reward by attenuating evoked dopamine release in structures like the nucleus accumbens (Felger & Miller, 2012). This state of diminished motivation is often compounded by a global shift towards neuronal hyperexcitability. Cytokines such as TNF-α and interleukin-1-beta (IL-1β) can induce glutamate-mediated excitotoxicity through a dual mechanism: stimulating glutamate release from activated microglia and astrocytes while simultaneously impairing its clearance by downregulating astrocytic transporters like EAATs (Felger & Miller, 2012; Takeuchi et al., 2006). The brain’s delicate excitatory-inhibitory balance is further perturbed by context-dependent modulation of GABAergic signaling and the induction of a central cholinergic deficit, driven by inhibited acetylcholine synthesis and enhanced degradation (Miller et al., 2013). This coordinated neurochemical reconfiguration is expanded by neuropeptidergic and neurotransmitter systems governing stress and arousal, such as corticotropin-releasing hormone (CRH) and norepinephrine (NE), which are acutely engaged to orchestrate the broader neuroendocrine components of the sickness response (Black, 2002; Finnell et al., 2019). The regulatory network is tuned by chemokines, which, in addition to their canonical chemotactic roles, function as direct neuromodulators. For example, CCL2 promotes neuronal hyperexcitability and CX3CL1 mediates homeostatic synaptic scaling (Ji et al., 2024; Sheridan et al., 2014; Sowa & Tokarski, 2021). These are not stochastic events, but rather a highly organized, adaptive program initiated by the immune system to inform the brain of systemic danger signals, thus shifting its operational state away from long-term, resource-intensive goals and towards immediate survival.

While the acute effects of immune signals on neurotransmission are critical for immediate behavioral adaptation, a more profound and lasting consequence of this dialogue emerges when communication becomes chronic or overly intense, leading to long lasting structural and functional remodeling of neural circuits. Prominent among these changes are the disruption of synaptic plasticity, long thought of as the cellular basis of learning and memory (Citri & Malenka, 2008). Pro-inflammatory cytokines, like IL-1β, can disrupt hippocampal long-term potentiation (LTP) by activating pathways such as p38 MAPK, which inhibits essential neurotrophic signaling via brain-derived neurotrophic factor (BDNF) and its receptor, TrkB (Tong et al., 2012). Should the inflammatory stimulus persist, the system can further disrupt synaptic function by physically eliminating synapses. This is mediated by the complement cascade, a pillar of innate immunity co-opted in the CNS for circuit refinement. In pathological states, pro-inflammatory signals drive the excessive deposition of complement components, particularly C1q and C3, which “tag” synapses for elimination by microglia (Schafer et al., 2012). This mechanism of synaptic pruning, is pathologically accelerated in neurodegenerative diseases and implicated in the excessive synaptic loss observed in schizophrenia, leading to irreversible structural damage to the brain (Hong et al., 2016; Sekar et al., 2016). Lastly, epigenetic reprogramming mechanisms, encoded by neural and glial cells following immunological insults can establish a persistent memory that changes neural circuit function through regulation of gene expression (Wendeln et al., 2018). Cytokines orchestrate these modifications by influencing key enzymes, such as recruiting histone deacetylases (HDACs) to silence plasticity-related genes or altering DNA methyltransferase (DNMT) activity to repress the expression of genes like BDNF (Karpova, 2014). Concurrently, the regulation of non-coding RNAs, including microRNAs like miR-146a and miR-124, fine-tunes inflammatory and neuronal gene networks, establishing long-lasting cellular states such as microglial priming, where cells become persistently hyper-reactive to subsequent stimuli (R. Yang et al., 2023). This sequence of events, from functional disruption to structural remodeling to epigenetic modifications for long term storage, provides a powerful framework for understanding the transition from acute insult or sickness to chronic neuropsychiatric and neurological disease.

These molecular and structural transformations are not abstract processes but are orchestrated by a dynamic and interacting cellular ensemble, comprising both CNS-resident glia and immune cells recruited from the periphery. Microglia, the primary resident immune cells, act as the first responders, using pattern-recognition receptors to sense danger and release the very cytokines that drive acute neurotransmitter changes. They are also the primary effectors of complement-mediated synaptic pruning, physically engulfing tagged synapses, a function tightly regulated by signals released locally by neurons, such as CX3CL1 and CD200. Astrocytes, which are often cued by microglial signals, act as crucial partners of the neuroinflammatory response, undergoing reactive astrogliosis to adopt context-dependent phenotypes, ranging from pro-inflammatory (A1) to protective (A2) (Ding et al., 2021; Giovannoni & Quintana, 2020). They are central executors of excitotoxicity, contributing to both the release and impaired clearance of glutamate (Mahmoud et al., 2019), and can even upregulate MHC class II molecules to participate in antigen presentation (Rostami et al., 2020). Oligodendrocytes and oligodendrocyte precursor cells (OPCs) are primarily responsible for myelination, which is critical for efficient communication between the brain and the rest of the body. But they also play critical roles in regulating CNS repair and circuit and synaptic plasticity following immune activation. Like other glial cells, OPCs can engulf synapses to regulate synaptic plasticity, or differentiate and remyelinate neural circuits after injury, processes that are regulate by immune-oligodendrocyte interactions (Auguste et al., 2022; Madeira et al., 2022).

In addition to local mechanisms of neuroimmune regulation, systemic immune compartments can also contribute to the processes described above. When a threat overwhelms local CNS defenses, a critical phase shift occurs, where the recruitment of peripheral immune cells acts to amplify the local CNS resident immune responses. Monocytes are recruited from the blood via chemokines like CCL2, across a compromised blood-brain barrier, and differentiate into macrophages that exacerbate neuroinflammation (D’Mello et al., 2009a). The direct behavioral impact of this infiltration is profound, as demonstrated by findings that chronic stress elevates circulating myeloid-derived matrix metalloproteinase-8 (MMP8), which infiltrates the nucleus accumbens to directly drive tissue matrix restructuring, synaptic plasticity, and social avoidance behavior (Cathomas et al., 2024). Lymphoid cells introduce further nuance, particularly at the CNS borders. While certain T cell subsets (Th1/Th17) can drive autoimmunity, others are integral to behavioral regulation; for instance, meningeal T cells producing interferon-gamma (IFN-γ) can directly influence brain circuits that regulate social behavior (Filiano et al., 2016), while meningeal gamma-delta (γδ) T cells producing IL-17 modulate anxiety (Alves de Lima et al., 2020). Recent evidence also reveals that chronic stress can dysregulate B cell homeostasis, leading to the production of brain-reactive autoantibodies, thereby providing a direct mechanistic link between psychological stress and CNS-targeted autoimmunity (Shimo et al., 2023). This layered choreography, from resident glial activation to peripheral immune trafficking and infiltration, underscores the complex cellular dynamics that translate the shared language of neuroimmunology into lasting changes in brain function and behavior. Figure 1 presents these mechanisms into a three-panel roadmap that progresses from peripheral signal entry at the neurovascular interface (Panel A), through glia-mediated transduction (Panel B), to activity-dependent synaptic remodeling (Panel C), thereby paving the way for the network-level analysis that follows.

Figure 1:

A multi-panel illustration of core cellular and molecular pathways in the central nervous system, linking peripheral immune signals to glial transduction and synaptic remodeling that underlie behavioral changes. (Panel A) Peripheral signal entry at the neurovascular interface. Blood-brain barrier disruption showing tight junction (TJ) breakdown by inflammatory mediators. MMP8 degrades extracellular matrix while brain-reactive autoantibodies (IgG) breach the barrier. Monocyte diapedesis and cytokine signaling (IL-1β, TNF-α, IL-6) initiate neuroinflammatory cascades at astrocyte endfeet. (Panel B) Glial activation and signal amplification. Activated microglia release inflammatory mediators (cytokines, ROS, glutamate), triggering astrocyte transformation to the A1 reactive phenotype via IL-1α, TNF-α, and C1q signaling. Reactive astrocytes downregulate EAAT, amplifying glutamate dysregulation. (Panel C) Synaptic alterations and behavioral outcomes. Glutamate excitotoxicity through NMDA receptor overactivation and complement-mediated synaptic pruning (C1q/C3) result in synaptic loss. Reduced neurotransmitter signaling (dopamine, serotonin) and decreased BDNF contribute to anhedonia, anxiety, and cognitive impairment.

Having established these foundational molecular and cellular principles of neuroimmune communication, we now examine how these universal mechanisms are specifically deployed and integrated within the diverse peripheral organ nodes of the CONB Network. In the next section, we will dissect the unique contributions of each central organ axis—from the gut to the spleen—to the overall regulation of behavior.

The Organ-Wide Neuroimmune Network: Peripheral Axes Influencing Behavior

Building on the foundational mechanisms outlined above (Figure 1), we now turn to individual organ axes, examining each as a functional node within the CONB Network. Figure 2 offers a semi-anatomical overview that maps primary and secondary organ nodes, their neural or humoral edges, and the central brain hub—providing readers with a systems-level compass for the subsections that follow.

Figure 2:

Semi-anatomical overview of the Cross-Organ Neuroimmunology of Behavior (CONB) network, mapping neural and humoral connections between major peripheral organs and the brain as an integrated whole-body framework. Panel A (Immune Organs) shows bone marrow, spleen, and lymph nodes as primary immune-responsive organs involved in immune cell production, trafficking, and inflammatory regulation. Panel B (Metabolic Organs) depicts gut, liver, and adrenal glands as key metabolic regulators coordinating neuroimmune signaling through microbial metabolites, hepatic cytokines, and stress hormones. Panel C (Barrier Organs) represents lungs, heart, skin, and skeletal muscle as tissue interfaces that respond to environmental challenges and mechanical stressors. Color-coded arrows indicate communication modalities: blue (neural pathways), red (humoral pathways), and green (immune cell trafficking). Bidirectional arrows emphasize reciprocal organ-brain communication, with pathway thickness reflecting signal strength.

Intestine (Gut-Brain-Immune Axis)

Within the CONB Network, the intestine functions as a primary contextual modulator and a high-traffic hub that continuously translates microbial and dietary signals into systemic immune responses, shaping neuroimmune tone throughout life (Wolterink et al., 2022). The gut microbiota, a key component of this axis, releases metabolites like short-chain fatty acids (SCFAs) that are essential for the homeostatic maturation and function of CNS microglia, thereby influencing neurodevelopment, cognition, and mood (Cryan et al., 2019; Erny et al., 2015; Fung et al., 2017; Leonardi et al., 2022; Matta et al., 2019; Medina-Rodríguez et al., 2024; Ritz et al., 2024; Wang & Pavert, 2022). Perturbations of this hub, such as dysbiosis driven by stress, antibiotics, or even sex hormone fluctuations, can alter systemic immunity via breakdown of the epithelial barriers and increased gut permeability (“leaky gut”). This compromised barrier allows microbial products, such as endotoxins, and immune signaling molecules (e.g., those involving Th17/IL-17) that can reach the brain via circulation or vagal afferents, contributing to neuroinflammation and alterations in behavior (Girolamo et al., 2017; Vanuytsel et al., 2023; Yuan et al., 2023).

This axis is characterized by robust bidirectional communication. Descending neural circuits, such as the stress-activated pathway from the paraventricular nucleus of the hypothalamus (PVN) corticotrophin releasing hormone (CRH+) neurons to the enteric nervous system of the colon, can directly drive intestinal inflammation and barrier disruption, demonstrating top-down control over this peripheral node (Russo et al., 2023). Conversely, ascending signals from the gut have a profound influence on brain states (Margolis et al., 2024; Wallrapp & Chiu, 2024). This positions the gut-brain axis, not merely as a contributor to pathology, but as a critical interface where other contextual factors, like early-life programming, stress and diet, exert their lasting effects on brain health. Consequently, interventions targeting the gut microbiota that effect local immunity (e.g., probiotics, fecal microbiota transplantation, FMT) represent promising strategies for modulating the entire CONB Network and hold therapeutic potential for a range of brain diseases (Cryan et al., 2019; Jordan et al., 2018; Loh et al., 2024).

Adrenal Glands (HPA Axis and Stress Immunology)

The adrenal glands are the primary effector nodes of the central stress response, converting neural signals into systemic hormonal cascades that exert potent but time-dependent control over immunity and behavior. Acutely, elevated GCs typically exert potent anti-inflammatory effects, suppressing pro-inflammatory cytokine production and limiting immune cell activity, thereby protecting the brain from excessive peripheral inflammation—a role highlighted by heightened neuroinflammation in adrenalectomized animals lacking endogenous GCs (Bellavance & Rivest, 2014; Wolterink et al., 2022). Concurrently, catecholamines released from the adrenals during acute stress influence leukocyte trafficking and inflammation (Ince et al., 2018). This neuroendocrine-immune communication is bidirectional, as immune signals also trigger HPA activation, and adrenal hormones like GCs modulate immune cell functions, including T-cell differentiation, which can affect stress resilience (Dantzer, 2018).

However, chronic or severe stress disrupts this finely tuned system, leading to HPA dysregulation, altering GC receptor sensitivity, and impairing immune homeostasis. Sustained high cortisol levels can induce GC resistance, resulting in a paradoxical failure to suppress inflammation and even exacerbation of stress-related neuroimmune pathology (Dantzer, 2018; Nusslock & Miller, 2016). This chronic maladaptation is linked to persistent microglial activation, increased NF-κB inflammatory signaling within limbic circuits, and subsequent anxiety- and depressive-like behaviors (Bellavance & Rivest, 2014; Calcia et al., 2016). Furthermore, early-life stress can impart lasting epigenetic modifications to GC receptor genes, predisposing individuals to lifelong alterations in stress responsivity and increased risk for developing mood disorders (Calcia et al., 2016; Dantzer, 2018; Nusslock & Miller, 2016). Thus, the adrenal glands serve as a critical neuroimmune nexus, a role complemented by other endocrine nodes, such as the thyroid, where autoimmune dysregulation can also drive neuroinflammation, independent of hormonal status.

Lungs (Pulmonary Immune-Brain Interaction)

The lungs act as a key sentinel node in the CONB Network, translating immune challenges at the body’s largest environmental interface—from respiratory infections to airborne pollutants—into systemic inflammatory signals that drive behavioral changes. Chronic respiratory conditions like COPD and severe asthma, which are linked clinically to increased rates of depression, anxiety, and cognitive deficits, exemplify this connection. Experimental models have confirmed that chronic allergic lung inflammation elevates systemic cytokines that induce depressive-like behaviors through the central neuroinflammatory pathways discussed above (Dill-McFarland et al., 2024; Kanaya et al., 2022). Acute respiratory infections, such as bacterial pneumonia or severe viral infections, also have a robust impact on the CNS. Pathogens like Pseudomonas aeruginosa provoke a systemic “cytokine surge” (IL-1β, IL-6, TNF-α) that leads to substantial neuroinflammation and behavioral abnormalities, often mediated by blood-brain barrier (BBB) disruption and endothelial activation, even with limited direct pathogen invasion of the brain parenchyma (Vanderheiden & Klein, 2022; Villalba et al., 2023).

The COVID-19 pandemic has starkly illuminated these lung-brain neuroimmune interactions. Even mild respiratory SARS-CoV-2 infections can trigger persistent neuroinflammation, dysregulation of myelin-producing oligodendrocytes, impaired hippocampal neurogenesis, all of which are linked to enduring cognitive deficits, including memory and attention problems (Fernández-Castañeda et al., 2022; Radke et al., 2024; A. C. Yang et al., 2021). Long COVID frequently involves chronic cognitive impairments (“brain fog”) associated with elevated circulating cytokines (e.g., CCL11) and hippocampal microglial activation, phenotypically resembling neuroinflammatory syndromes seen after chemotherapy (Fernández-Castañeda et al., 2022). The impact of such pulmonary insults is often exacerbated by aging, as the age-related pro-inflammatory state (“inflammaging”) and primed microglia render the brain more vulnerable to the secondary neuroinflammatory consequences of respiratory distress (Demuth et al., 2023). Furthermore, environmental factors play a significant role; the inhalation of particulates, such as silica, triggers pulmonary inflammation that can induce hippocampal inflammation and disrupt synaptic function, ultimately leading to cognitive deficits (Suman et al., 2022). These examples highlight multiple pathways—including circulating cytokines, immune cell trafficking, and neural routes such as vagal afferents—through which pulmonary events influence brain health. Therapeutically, effective management of lung inflammation, such as using anti-inflammatory biologics (e.g., anti-IL-17A) for asthma, correlates with improvements in mood and cognition, underscoring the lung as a key therapeutic target for associated neuropsychiatric conditions (Dill-McFarland et al., 2024; Kanaya et al., 2022). These findings position the lung-brain axis as a key modulator of CONB-mediated sickness behavior and cognitive fog following infection.

Skin (Cutaneous Neuroimmune System and Behavior)

The skin represents a unique CONB Network node where localized neurogenic inflammation, exemplified by the itch-scratch cycle, can escalate to drive systemic consequences for mood and sleep. It often participates in a broader “Gut-Lung-Skin Axis,” a concept reflecting the clinical co-occurrence of inflammatory conditions across these three significant environmental barriers, such as atopic dermatitis (skin), asthma (lung), and inflammatory bowel disease (gut). Richly innervated by sensory neurons and populated by diverse immune cells (e.g., mast cells, dendritic cells, T cells), the skin mediates local sensations like itch and pain and contributes to systemic inflammation. Psoriasis, for example, is associated with heightened systemic inflammatory cytokines (e.g., IL-17, TNF-α), which contribute to comorbid depression by engaging the foundational neuroimmune mechanisms outlined above (Armstrong & Read, 2020; Katamanin et al., 2025). Similarly, atopic dermatitis (AD), characterized by intense itch (pruritus) and chronic inflammation, involves intricate neuroimmune crosstalk where cytokines like IL-31 and neuropeptides, released by immune cells and sensory nerves, perpetuate a cycle of itching and inflammation, severely disrupting sleep and mental wellbeing (Datsi et al., 2021; Steinhoff et al., 2022). Specific signaling networks involving neuronal receptors, such as IL-31R and Mrgprs, are crucial mediators of these chronic itch responses (J. E. Choi & Di Nardo, 2018; Kim et al., 2024).

Neuroimmune communication within the skin involves critical cell-cell interactions. Mast cell-sensory neuron loops, where mast cells release histamine and cytokines, stimulating neurons, and neurons modulate mast cell activity, thereby sustaining allergic and inflammatory symptoms (Bao & Abraham, 2024; Mack & Kim, 2018). Recent findings also suggest that skin-infiltrating neutrophils contribute to chronic itch through CXCR3 signaling (Walsh et al., 2019). Notably, the impact of inflammatory skin diseases extends beyond the cite of local neuroimmune activation in the skin, affecting sleep patterns and cognitive functions through mechanisms that are likely to involve increased circulating cytokine and subsequent neuroinflammation. This is particularly evident in AD patients, where disease severity strongly correlates with sleep disturbances and psychological distress (Cameron et al., 2024). Therapeutically, targeting cutaneous neuroimmune pathways shows significant promise; biologic therapies inhibiting specific cytokines (e.g., IL-17, IL-4/13, TNF-α) effectively treat skin lesions while concurrently alleviating associated mood disorders, underscoring the direct skin-brain neuroimmune link and reinforcing the skin’s integral role in psychoneuroimmunology (Armstrong & Read, 2020; Katamanin et al., 2025).

Heart (Cardio-Neuro-Immune Interactions)

Within the CONB Network, myocardial infarction flips the heart from injured target to inflammatory hub: dying cardiomyocytes alert resident macrophages, while danger signals recruit neutrophils and CCR2⁺ monocytes from circulation that release IL-1β, IL-6 and TNF-α that are linked with post-MI depression and cognitive decline. These cytokines also circulate to blood–brain interfaces to activate microglia, and trigger neuroinflammation that impairs mood and cognitive function (Thackeray et al., 2018). A surge in sympathetic activity post-MI, leads to the release of norepinephrine to mobilize immune cells and enhance inflammation via β-adrenergic receptors, while vagal loss disrupts the cholinergic anti-inflammatory pathway, pushing immune cells toward pro-inflammatory states and reinforcing this cardio-neuro-immune loop (Bellinger & Lorton, 2014; D’Mello et al., 2009b). Aging, a primary risk factor for cardiovascular events, also primes the neuroimmune system for exaggerated responses, creating a scenario where post-myocardial infarction (MI) inflammation leads to more severe neuroinflammatory consequences and behavioral deficits in older individuals. MI serves as a critical example, triggering a potent systemic inflammatory response characterized by the mobilization of monocytes and neutrophils from reservoirs, such as the spleen and bone marrow. These cells infiltrate the damaged heart, releasing cytokines (IL-1β, IL-6, TNF-α) that drive post-MI depression and cognitive decline, consistent with the established effects of systemic inflammation on central neural circuits (Perry & Holmes, 2014; Swirski et al., 2009). Therapeutic targeting of this inflammation, exemplified by minocycline treatment, effectively reduces brain inflammation and depressive symptoms in preclinical models, underscoring the cardio-neuroimmune connection (Jinawong et al., 2021).

In addition to local inflammatory pathways engaged within the heart, recent work has revealed distinct neural pathways connecting cardiac and arterial signals to specific brain areas that govern immune regulation, emotional processing, and stress responses, highlighting a network that integrates multisystem diseases (Mohanta et al., 2023). Chronic heart failure further illustrates these neuroimmune dynamics, with persistent elevations in inflammatory cytokines like TNF-α correlating strongly with cognitive deficits and mood disturbances. Specific mechanisms, such as angiotensin IL-driven neuroinflammation in the hippocampus, have been identified as contributing to cognitive impairment in heart failure, suggesting potential therapeutic targets (Althammer et al., 2023). Adding another layer of complexity, recent research has revealed that sleep plays a crucial role in regulating post-MI cardiac inflammation. Sleep quality, modulated by monocytic TNF signaling in brain nuclei, influences cardiac sympathetic output, thereby impacting myocardial inflammation and recovery. Disrupted sleep exacerbates cardiac injury, emphasizing a vital bidirectional sleep-cardio-neuroimmune interaction (Huynh et al., 2024).

These collective insights pave the way for novel intervention strategies extending beyond traditional cardiovascular care. Treatments targeting neuroinflammation, modulating autonomic function (e.g., via vagal nerve stimulation), and incorporating behavioral factors like sleep optimization offer promising avenues for synergistically improving cardiovascular and mental health outcomes (M. X. Hu et al., 2018; Iseger et al., 2020). In summary, the heart has a significant influence on neuroimmune-driven behaviors through complex, bidirectional pathways that involve systemic inflammation, dedicated neural circuits, and autonomic regulation, thereby establishing its central role within the broader framework of psychoneuroimmunology.

Muscles (Exercise, Myokines, and Brain Function)

Skeletal muscle functions as a powerful health-promoting node within the CONB Network, primarily through the activity-dependent release of anti-inflammatory and neurotrophic myokines during exercise. Physical exercise stimulates the release of numerous myokines, which have profound neuroimmune effects. Notably, muscle-derived IL-6 acts predominantly as an anti-inflammatory signal by enhancing IL-10 and cortisol secretion while suppressing TNF-α, and contributing to the antidepressant and anxiolytic effects of exercise (Pedersen, 2019; Severinsen & Pedersen, 2020). Exercise also activates PGC-1α-dependent kynurenine metabolism within muscle, reducing circulating neurotoxic kynurenine and protecting the brain from stress-induced depression (Agudelo et al., 2018; Cervenka et al., 2017). Furthermore, myokines such as cathepsin B and FNDC5 (which can be cleaved to produce irisin) can cross the blood-brain barrier, stimulate BDNF production, and promote neurogenesis, synaptic plasticity, and cognitive enhancement (Moon et al., 2016; Pedersen, 2019). These mechanisms underline the broad neuroprotective impacts of regular exercise observed in conditions like Alzheimer’s disease or depression, where myokines help counteract pathology and preserve cognition, and contribute to exercise’s robust antidepressant effects and influence on psychological resilience (S. H. Choi et al., 2018; Kandola et al., 2019; Valenzuela et al., 2020).

Conversely, conditions involving muscle atrophy, disuse, or inflammation (e.g., cachexia, chronic diseases) can negatively impact cognitive function. Under these states, peripheral inflammatory signals originating in muscles may propagate to the CNS, contributing to neuroinflammation and subsequent behavioral impairments, highlighting a bidirectional brain-muscle pathway (Liu, 2022; S. Yang et al., 2024). Thus, a major focus of current research is to develop therapeutics that harness muscle-derived signals to promote brain health. Strategies that involves the use of exercise-mimicking pharmacological agents or muscle-targeted treatments, such as electrical stimulation, show promise in preclinical models for enhancing hippocampal neurogenesis and cognitive function (Hendrikse et al., 2017; Sleijser-Koehorst et al., 2023). Collectively, skeletal muscle significantly influences brain health and neuroimmune regulation through dynamic endocrine and immunological pathways, offering novel therapeutic avenues for treating cognitive decline and mood disorders, and reinforcing the muscle’s essential role in psychoneuroimmunology.

Liver (Metabolic-Immune Interface and Brain Impact)

The liver serves as the central immunometabolic hub of the CONB Network, integrating signals from the gut with systemic metabolic state to regulate the inflammatory tone that shapes brain function and behavior. Resident hepatic immune cells, especially Kupffer cells, respond to gut-derived signals like endotoxins by releasing pro-inflammatory cytokines (e.g., IL-6, IL-1β, TNF-α). These mediators can impact the brain via circulation and impair and cognitive and emotional functions. This connection is evident in widespread conditions, such as metabolic dysfunction-associated steatotic liver disease (MASLD), chronic liver inflammation correlates with elevated systemic cytokines, although clinical studies are associative (Kjaergaard et al., 2023; Mikkelsen et al., 2025), whereas experimental MASLD models demonstrate that hepatic inflammation drives peripheral and central cytokine release and consequent neuroinflammation (Kjaergaard et al., 2023). Animal studies directly link diet-induced hepatic steatosis and inflammation to resultant neuroinflammation, brain hypoxia, and anxiety- or depressive-like behaviors, underlining the neurotoxic potential of hepatic inflammation (Hadjihambi et al., 2023).

Advanced liver dysfunction introduces additional neuroimmune challenges, such as hyperammonemia, a metabolic condition marked by increased ammonia levels in the blood that can significantly exacerbate neuroinflammation and cognitive deficits, and involves mechanisms such as increased peripheral cytokine signaling and direct impairment of astrocyte function (Izquierdo-Altarejos et al., 2023). Similar pathways operate in hepatic encephalopathy, where elevated systemic ammonia and inflammation impair astrocyte autophagy, worsening neuronal dysfunction and cognitive decline (Dhanda et al., 2018; Lu et al., 2019). Significantly, the gut microbiota profoundly mediates these liver-brain interactions. Modulating the gut microbiome, for instance through probiotic supplementation (e.g., Akkermansia muciniphila), can exert neuroprotective effects in liver disease models by reducing inflammation and enhancing brain-supportive pathways involving BDNF and serotonin (Kang et al., 2024; Sun et al., 2024). Therapeutic interventions targeting peripheral inflammation, like anti-TNF-α treatment, have successfully reversed cognitive impairments in hyperammonemic models, confirming a causal role for liver-derived inflammation in neuroimmune dysfunction (Izquierdo-Altarejos et al., 2023).

Given these complex interactions, targeting hepatic health offers significant therapeutic potential for associated neurological and neuropsychiatric symptoms. Interventions addressing liver inflammation and metabolic disturbances—ranging from dietary modifications and probiotics to bariatric surgery and specific anti-inflammatory therapies—are crucial for achieving broader neurocognitive benefits. Collectively, liver-driven systemic inflammation, metabolic dysregulation (including factors like ammonia), and modulation via the gut microbiota critically shape brain function and behavior. This metabolic-immune interface is further influenced by the pancreas, where diabetes-associated inflammation and hormones, such as GLP-1, also impact neuroimmune health (Drucker & Holst, 2023; Kellar & Craft, 2020). This reinforces the liver’s prominent position in psychoneuroimmunology research and highlights hepatic health as a crucial strategy for maintaining optimal brain function and emotional wellbeing.

Bone Marrow (Hematopoietic Origins of Neuroimmune Responses)

The bone marrow acts as a primary hematopoietic node of the CONB Network, where central stress signals directly reprogram the production of myeloid cells, thereby shaping the nature and intensity of peripheral immune responses that influence the brain. Chronic stress activates sympathetic neuron inputs that innervate bone marrow, releasing norepinephrine that skews hematopoietic stem cell (HSC) differentiation towards pro-inflammatory myeloid lineages (Cathomas et al., 2024; Heidt et al., 2014; Powell et al., 2013). These inflammatory monocytes subsequently infiltrate the brain, especially during aging or under psychological stress, where they interact with microglia and the extracellular matrix to drive behavioral alterations (Cathomas et al., 2024; H. Hu et al., 2022). This stress-induced shift in myelopoiesis can be programmed by early-life adversity, which establishes a long-term pro-inflammatory bias in hematopoietic output, creating a lasting vulnerability to stress-related psychopathology (Powell et al., 2013). Specific molecular pathways underscore this link: psychological stress elevates vasopressin, which acts on HSCs via the IL-36 G-IL-1RL2 axis to promote the generation of inflammatory monocytes, contributing to depression-like behaviors upon brain homing (Mou et al., 2024). Furthermore, circulating monocyte-derived factors, such as MMP8, can infiltrate specific brain regions (e.g., the nucleus accumbens) to modulate synaptic plasticity, thereby provoking depression-related behavior in stress-susceptible states (Cathomas et al., 2024). The direct influence of the hematopoietic system on emotional states is strongly supported by bone marrow transplant studies, where the transfer of pro-inflammatory marrow transmits susceptibility to anxiety and depression-like symptoms, while healthy marrow normalizes behavior in recipients with impaired behavior (Hodes et al., 2014; Wohleb et al., 2013).

Mechanistically, bone marrow-derived monocytes often utilize chemokine signals, such as CCR2, for trafficking to brain compartments. Under severe situations when these cells infiltrate the CNS parenchyma, they can adopt microglia-like phenotypes, amplifying neuroinflammatory cascades and contributing to stress sensitization and cognitive impairments (Biltz et al., 2022; H. Hu et al., 2022). Recent evidence suggests that the skull bone marrow may be a potentially localized source for immune cell trafficking into adjacent meningeal and brain tissues, as supported by both rodent and human imaging studies that link skull marrow inflammation to markers of depression (Cugurra et al., 2021; Eiff et al., 2025). Therapeutic strategies targeting this bone marrow–brain communication axis have shown promise in preclinical models for reducing anxiety and depressive behaviors, including agents blocking sympathetic signaling to marrow, modulating HSC output towards anti-inflammatory lineages, or inhibiting trafficking molecules like CCR2 antagonists (McKim et al., 2018; Sawada et al., 2014). In summary, the bone marrow serves as a dynamic interface, translating psychological stress into altered peripheral immunity that directly influences neuroimmune tone and emotional behavior, thereby establishing it as a pivotal organ in the field of psychoneuroimmunology.

Spleen (Peripheral Immunity and the Inflammatory Reflex)

The spleen functions as a key neuro-autonomic interface and immune cell reservoir pool within the CONB Network, best exemplified by the cholinergic anti-inflammatory pathway (CAP), which allows the brain to suppress systemic inflammation directly. It receives sympathetic innervation, modulated by efferent vagus nerve fibers, forming the neural basis of the CAP. In this reflex, vagal signals lead to the release of norepinephrine from the spleen, which acts on acetylcholine-producing T cells. The subsequent release of acetylcholine stimulates α7-nicotinic receptors (α7nAChR) on macrophages, thereby suppressing the production of TNF-α and other pro-inflammatory cytokines (Kelly et al., 2022). Activation of this “inflammatory reflex” dampens systemic inflammation, thereby reducing sickness behavior and delirium during infection, while its disruption (e.g., via splenic nerve transection) exacerbates inflammatory pathology. Complementing this anti-inflammatory circuit, distinct neural pathways also mediate behaviorally-driven immunity: CRH-expressing neurons in the amygdala and PVN connect to the splenic nerve to regulate plasma cell formation, with behavioral or pharmacogenetic stimulation enhancing antigen-specific antibody production via α9-nicotinic receptors on B cells (Zhang et al., 2020).

Bidirectional communication is also prominent, with the spleen acting as a crucial reservoir and source of immune cells influencing distant organs. Under conditions of stress or injury (e.g., stroke, myocardial infarction), the spleen mobilizes monocytes that can traffic to the CNS or heart via circulation, where they contribute significantly to sustained inflammation and correlate with worse behavioral outcomes—such as increased anxiety-like behavior and impaired sensorimotor/cognitive function. Consequently, splenectomy can reduce this infiltration and improve recovery in animal models, albeit at the expense of compromising long-term host defense.” Consequently, splenectomy can reduce this infiltration and improve recovery in animal models, albeit at the expense of compromising long-term host defense. Furthermore, the spleen functions as a neuroimmune relay in chronic conditions. Specific neuronal populations (e.g., in somatosensory cortex and amygdala) regulate splenic TH2 immune responses via vagal projections, modulating peripheral inflammation in response to pain states (Zhu et al., 2024). In cardiovascular disease, a heart-brain-spleen reflex involving splenic PlGF secretion regulates adaptive cardiac remodeling following pressure overload (Perrotta et al., 2025).

Recognizing these pathways, the spleen is increasingly targeted for therapeutic modulation. Bioelectronic approaches, such as focused ultrasound stimulation of the spleen (sFUS), activate anti-inflammatory neuroimmune circuits, demonstrating potential in conditions like pulmonary hypertension (Zafeiropoulos et al., 2024). Vagus nerve stimulation (VNS), approved for treatment-resistant depression, stroke rehabilitation, and epilepsy, also leverages splenic pathways to reduce systemic inflammation. In addition, VNS is being investigated for treatment of several other conditions including, Alzheimer’s disease, anxiety disorders, PTSD, heart failure, inflammatory bowel disease, and rheumatoid arthritis. This convergence highlights the spleen as a powerful intermediary between brain states, behavior, and systemic immunity. Overall, the spleen functions dually as both an effector, orchestrating immune tone influenced by the brain, and a sensor/relay, contributing to neuroinflammation, thereby placing it centrally in neuroimmune integration with both protective and pathological implications.

Kidneys (Renal Immune Signals and Neurocognitive Effects)

Within the CONB Network, the kidneys modulate neuroimmune responses primarily via immune-metabolic pathways. Chronic kidney disease (CKD) induces systemic inflammation through uremic toxins and cytokines (e.g., IL-1β) (Bronas et al., 2017; Zimmermann et al., 2024), which activate CNS microglia, impair blood-brain barrier integrity, and contribute to cognitive decline and mood disorders (Bugnicourt et al., 2013; Zimmermann et al., 2024). While kidney-derived factors, such as erythropoietin (EPO), can be neuroprotective, the primary role of this axis in behavior appears to be driven by the consequences of renal dysfunction on systemic inflammation (Thompson et al., 2023). Future work is needed to disentangle the role for kidney signaling to better understand whether the kidneys plays a boarder role within the CONB network as do other organs mentioned above.

Lymph Nodes and Meningeal Lymphatics (Immune Cell Trafficking and Brain Drainage)

The discovery of functional meningeal lymphatic vessels (mLVs) lining the dural sinuses has fundamentally reshaped our understanding of CNS immunity, refuting the concept of absolute immune privilege. By draining cerebrospinal fluid (CSF) and interstitial fluid—along with metabolic waste, soluble antigens, and immune cells—into the deep cervical lymph nodes (dCLNs), these vessels enable ongoing peripheral immune surveillance of the CNS (Mesquita et al., 2018). This outflow operates in tandem with the glymphatic system (for review see “The Glymphatic System – A Beginner’s Guide”(Jessen et al., 2015)) and is modulated by behavioral states such as sleep, influencing the clearance of proteins like amyloid-β (Aβ).

The functional integrity of mLV-dCLN drainage proves vital across a range of neurological conditions, from aging-related neurodegeneration to acute injuries. Impaired drainage facilitates the buildup of toxic aggregates (e.g., Aβ, tau), fueling neuroinflammation and cognitive decline, while also hindering CNS-targeted immunotherapies. Conversely, boosting mLV function (e.g., via VEGF-C) improves clearance and pathology in models of Alzheimer’s disease, stroke, and traumatic brain injury. By processing CNS-derived antigens, the dCLNs can shape systemic immune responses relevant to neurodegenerative conditions like multiple sclerosis. Thus, preserving or restoring meningeal lymphatic flow emerges as a pivotal strategy for maintaining CNS homeostasis and mitigating neuroimmune dysfunction in various disorders. After dissecting the individual functions of these organ-specific nodes, we will now move to the next section where we reintegrate them into a cohesive whole-body framework. This final section will analyze the dynamic, network-level properties that emerge from multi-organ interactions within the CONB framework and explore the therapeutic strategies that target the network as an integrated system.

Systems-Level Integration and Therapeutic Implications

Multi-Organ Crosstalk and Integrated Behavioral Responses

While the preceding sections have examined individual organ axes as distinct nodes within the CONB Network, the true regulatory power of this system lies in its integrated dynamics. Behavior is not the result of a single organ-brain dialogue, but rather it requires multi-organ network integration that is shaped by the constant crosstalk, feedback loops, and signaling convergence between multiple organ systems. This section moves from individual nodes to the edges that connect them, illustrating how multi-organ circuits form complex cycles that drive multi-dimensional behavioral phenotypes. Before we move to a detailed discussion of this multi-organ integration, let’s start with two examples that nicely highlight signaling within the CONB at the edges.

Example 1: The Vicious Cycle of Chronic Stress—A Brain-Adrenal-Marrow-Spleen Axis

Chronic psychological stress provides a well-established example of a maladaptive feed-forward loop within the CONB Network. The cycle initiates in the brain, where perceived stress triggers activation of the HPA axis and the sympathetic nervous system (Kelly et al., 2022). This central signal propagates to key peripheral nodes, triggering the release of glucocorticoids from the adrenal glands, which leads to paradoxical inflammation under chronic exposure. Simultaneously, sympathetic signaling drives pro-inflammatory myelopoiesis in the bone marrow, a process that is amplified by stress hormones, such as vasopressin (Heidt et al., 2014; Mou et al., 2024; Reader et al., 2015). The spleen then acts as a reservoir pool, mobilizing these primed inflammatory monocytes into circulation (Kelly et al., 2022). The loop closes as these monocytes and their products (e.g., MMP8) traffic back to the CNS, possibly compromising the blood-brain barrier in regions such as the nucleus accumbens, which then drives social avoidance behavior and anhedonia, perpetuating the initial stress perception (Cathomas et al., 2024).

Example 2: The Virtuous Cycle of Physical Exercise—A Muscle-Bone-Gut-Brain Axis

In stark contrast, physical exercise initiates a protective multi-organ process across the CONB Network (Khrimian et al., 2017; Qi et al., 2024). The initiating node is skeletal muscle, which releases myokines that promote a systemic anti-inflammatory tone, partly by inducing IL-10 and activating kynurenine-degrading pathways (Pedersen & Febbraio, 2012). These signals propagate through the network, positively modulating the gut microbiome to enhance the production of beneficial metabolites, such as SCFAs. Exercise also stimulates bones to release factors like osteocalcin, which crosses the BBB to improve cognitive function. This multi-organ signaling converges on the brain, where myokines and other factors promote the production of BDNF and neurogenesis, creating a positive feedback loop that enhances both physical and mental resilience (Voss et al., 2011).

Emergent Properties of the CONB Network

These examples reveal several key emergent properties of the CONB Network. First is signaling convergence: disparate peripheral nodes, such as the skin in psoriasis, the lungs in COPD, or the liver in MASLD, can all generate a standard set of inflammatory mediators (e.g., IL-1β, IL-6, TNF-α). These signals converge on the brain to produce a remarkably similar behavioral output—sickness behavior, fatigue, and anhedonia—demonstrating that the brain responds to a generalized “systemic inflammation” signal regardless of its origin. Second, the CONB Network exhibits profound degeneracy, a core principle of complex biological systems, rather than simple redundancy. While redundancy implies the presence of identical, interchangeable components, degeneracy refers to the phenomenon where structurally distinct elements can perform similar functions under specific contexts (Edelman & Gally, 2001; Whitacre, 2010). This property confers enhanced robustness and adaptive potential. Degeneracy is evident at multiple scales within the CONB Network:

At the macro-level, different peripheral organs (structurally distinct nodes) can trigger similar behavioral outcomes (e.g., sickness behavior) during infection, inflammatory flare, or metabolic disease, by releasing functionally convergent, albeit biochemically diverse, combinations of pro-inflammatory cytokines. These represent degenerate pathways leading to a standard functional output.

At the micro-level, the T-cell receptor (TCR) repertoire provides a classic example of degeneracy. A single TCR can recognize multiple different peptide-MHC complexes, while multiple distinct TCRs can recognize a single peptide (Colf et al., 2007). This “many-to-many” mapping ensures robust immune surveillance against a vast array of pathogens with a finite set of receptors.

This property of degeneracy ensures the network’s robustness: if one communication channel is compromised, other structurally different but functionally similar pathways can compensate, ensuring that critical information about peripheral state reaches the brain. Finally, the network exhibits vulnerable hubs, such as organs like the gut and bone marrow, which process and broadcast high-dimensional immune signals, exerting an outsized influence on the entire network’s tone. Perturbations at these hubs, such as gut dysbiosis or myelopoiesis, can have cascading effects that destabilize the entire system. Recognizing these interconnected pathways and network properties necessitates a holistic, system-level perspective for both research and therapy. The traditional approach of targeting a single organ or pathway is insufficient. Interventions must be understood through their network-level effects; for example, VNS does not just target the spleen but re-tunes a brain-spleen circuit (Mota & Madden, 2022). Similarly, diet and exercise are powerful therapeutic modalities precisely because they act on multiple nodes of the network simultaneously (Ashcroft et al., 2024; Townsend et al., 2023). Therefore, managing complex neuroimmune-related behavioral disorders requires a shift from a single-target paradigm to one focused on restoring equilibrium across the entire CONB Network.

Clinical applications and therapeutic interventions

Despite advances in conventional psychopharmacology and therapy, substantial unmet needs persist in treating complex behavioral disorders such as stress-related conditions (i.e., major depressive disorder (MDD), and post-traumatic stress disorder (PTSD)), and inflammation-associated cognitive dysfunction. High rates of treatment resistance, particularly in MDD subgroups exhibiting peripheral inflammation (e.g., elevated CRP, TNF-α, and IL-6), strongly suggest that targeting only central neurotransmitter pathways may be insufficient (Mancuso et al., 2023; Miller & Raison, 2016). The robust mechanistic links established herein—ranging from stress-induced disruption of the blood-brain barrier in specific regions like the NAc allowing peripheral IL-6 influx (Menard et al., 2017), to peripheral myeloid cell-derived factors like MMP8 directly altering the NAc extracellular matrix to drive social avoidance (Cathomas et al., 2024), and stress-induced B cell activation leading to potentially pathogenic brain-reactive autoantibodies (Shimo et al., 2023) provide a compelling rationale for exploring the neuroimmune axis as a source of novel therapeutic targets and biomarkers. Therefore, this section critically evaluates emerging immune-based interventions aimed at restoring neuroimmune homeostasis to alleviate behavioral symptoms and improve mental health outcomes.

Targeting Innate Immunity and Inflammatory Cascades

The clinical need for new psychiatric treatments—especially for treatment-resistant depression (TRD)—has driven a shift toward neuroimmune targets. In recent years, there has been a clear evolution from broad anti-inflammatory agents to more precise modulators of innate immunity. Cytokine-blocking biologics exemplify this. Elevated pro-inflammatory cytokines (TNFα, IL-6, IL-17, CRP) are well-documented in major depression, and meta-analyses confirm that adjunctive anti-inflammatories can reduce depressive symptoms (Köhler et al., 2014; Osimo et al., 2020). However, trials of TNFα inhibitors (e.g., infliximab) or IL-6 blockers in unselected depressed patients have been largely negative (Miller & Pariante, 2020). Crucially, benefits of anti-inflammatory treatments appear confined to an “inflammatory biotype”: patients with high baseline hs-CRP (>5 mg/L) respond selectively to anti-cytokine therapy (Raison et al., 2013), whereas patients with low baseline hs-CRP (<5 mg/L) responded worse. This highlights the importance of patient stratification by immune biomarkers, reframing these drugs as precision treatments for inflammation-driven depression.

Janus kinase (JAK) inhibitors offer another intracellular approach, potently blocking multiple cytokine signals via the JAK–STAT pathway. Agents like tofacitinib and baricitinib have shown potent anti-inflammatory effects, but safety warnings curtail their application in psychiatry. A 2022 NEJM study found JAK inhibitors carry higher risks of major cardiovascular events, cancers, thromboembolism, and serious infections compared to TNF inhibitors. These risks overlap heavily with the typical TRD population (e.g., increased age, smoking, comorbidities). Thus, despite a robust development pipeline, regulatory caution limits JAK use in neuropsychiatric disorders (Ytterberg et al., 2022).

Beyond cytokines, new druggable nodes in innate immunity are emerging. Novel brain-penetrant compounds NLRP3 inflammasome inhibitors (small molecules such as NT-0796, VTX-3232) are entering trials, overcoming prior hurdles with BBB penetration. VTX-3232 and NT-0796, which completed Phase 1, are moving into Phase 2 for neurological indications. Because NLRP3 sits at a convergence point for IL-1β and IL-18 production, these drugs offer a more targeted intracellular blockade than JAK inhibitors (Harrison et al., 2023). Another innovative target is MMP-8. MMP-8 (neutrophil collagenase) has recently been shown to be upregulated in activated microglia and to act as a TNFα-converting enzyme, cleaving pro–TNF to its active form (Lee et al., 2014; Vandenbroucke et al., 2012). Inhibiting MMP-8 in animal models dampens microglial inflammation and preserves the blood–CSF barrier, linking peripheral inflammation to CNS effects (Vandenbroucke et al., 2012). Although still preclinical, MMP-8 exemplifies the search for tractable targets downstream of cytokines. Together, these advances illustrate a precision-medicine paradigm: moving from systemic anti-inflammatory agents toward specific immunologic pathways matched to patient subtypes. I

Modulating Peripheral–CNS Communication

Therapies are also emerging that intervene along peripheral–brain axes rather than targeting the CNS directly. The gut–brain axis has been a significant focus. Dysbiosis of the intestinal microbiome is linked to mood disorders, and clinical trials are now testing “psychobiotics” – probiotics formulated to benefit mental health (Dinan et al., 2013). Additionally, dysbiosis of the gut microbiome has been implicated not only in mood disorders but also in multiple sclerosis, where probiotic interventions and FMT have shown preliminary benefits on disability, fatigue, and inflammatory biomarkers in MS patients (Correale et al., 2022; Tsogka et al., 2023). In a landmark 2025 randomized controlled trial (the ProDeCa study), a multi-strain psychobiotic improved depression, anxiety, and stress scores in high-stress surgical patients compared to placebo (Tzikos et al., 2025). These live microbes are hypothesized to work via local production of neuroactive compounds (e.g., GABA, serotonin), modulation of gut immunity, and vagal signaling (Sampson & Mazmanian, 2015). More radical is fecal microbiota transplantation (FMT), where whole microbial communities are transferred from healthy donors.

Preclinical and early human data indicate FMT can shift psychiatric phenotypes: transplanting microbiota from depressed patients induces depressive-like behavior in recipients, whereas healthy-donor transplants can alleviate symptoms (Chinna Meyyappan et al., 2020). These findings underscore the gut’s central role and align with the discussion in Part I regarding gut-derived metabolites and vagal pathways. However, FMT trials remain small and largely uncontrolled; future work must define which specific microbes are responsible for these effects and the specific pathways from gut to brain that mediate their antidepressant effects. There is growing consensus that larger, well-designed trials are needed to establish the long-term safety and efficacy of therapeutic strains or consortia.

Another peripheral approach is bioelectronic modulation. Vagus nerve stimulation (VNS) – which electrically activates the anti-inflammatory “inflammatory reflex” via the vagus nerve – is FDA-approved as an adjunct for TRD. Recent data have yielded a complex picture (Kelly et al., 2022). Long-term observational data (RESTORE-LIFE study) showed good tolerability and a cumulative therapeutic response of ~53% at one year in patients with treatment-resistant conditions (Young et al., 2020). However, a rigorous sham-controlled RCT found no significant difference in primary depression outcomes between active and sham VNS (Conway et al., 2025). Notably, the RCT revealed improvements in secondary measures (i.e., global clinical status and patient-reported symptoms), indicating that conventional trial endpoints may overlook the full benefits of VNS. One interpretation is that the antidepressant effects of VNS can be slow to emerge and may manifest more in functional recovery and quality of life than raw symptom scores. The dose–response relationship also appears nonlinear, which may impact treatment outcomes. Taken together, these data suggest that future VNS studies should focus on “how, when, and for whom” it works – identifying biomarkers of response and using long-term functional endpoints, rather than simply relying on self-report measures.

Adaptive Immunity and Autoimmune Neuropsychiatric Syndromes

A burgeoning area is the role of adaptive immunity in mental illness. Chronic stress, a key psychiatric risk factor, has been shown to dysregulate B-cell homeostasis directly. In a 2025 mouse study, chronic stress drove aberrant germinal-center B-cell expansion and autoantibody production that targeted the brain (Shimo et al., 2023). Clinically, autoimmune encephalitis (e.g., anti-NMDA receptor encephalitis), an established autoimmune condition marked by overproduction of neuronal autoantibodies against the NMDA receptor, which can cause severe deficits in neuronal function leading to a range of neuropsychiatric symptoms, including psychosis (Kayser & Dalmau, 2016). More subtle forms of neuronal autoimmunity are now being uncovered in broader psychiatric populations. For example, higher rates of anti-thyroid or anti-Ro antibodies have been noted in subsets of depressed or psychotic patients, though some associations may reflect confounds like smoking (Hansen et al., 2020). Circulating antibodies are correlated with symptoms of anhedonia in MDD subjects as well (Shimo et al., 2023). These converging lines of evidence support the concept of an “autoimmune depression” subtype. In such cases, standard antidepressants are unlikely to suffice; instead, treatments approaches used in neurology or rheumatology may be more appropriate (e.g., plasmapheresis, IVIg, B-cell depletion with rituximab). A key implication of this work is the need for routine autoantibody screening in severe, atypical, or treatment-refractory cases, especially when there is a rapid onset or a personal or family autoimmune history. Identifying these patients could open targeted immunotherapies and avoid years of ineffective trials.

Biomarkers for Patient Stratification and Target Engagement

The success of neuroimmune therapies hinges on precise biomarkers. Peripheral blood biomarkers are the most advanced clinically. Simple assays for inflammatory markers (e.g., hs-CRP, IL-6, TNF-α) are already critical for stratifying trials, as they help to identify the “inflammatory biotype” of depression that better responds to cytokine blockers. These markers are sensitive indicators of systemic immune activation but lack brain or disease specificity. Their utility potentially lies in distinguishing responders in RCTs with broad enrollment criteria for immunotherapy trials. By contrast, neuronal injury biomarkers, such as Neurofilament Light Chain (NfL), reflect CNS damage more directly. NfL is elevated in severe mental disorders—albeit less than in frank neurodegeneration—and can distinguish neurodegenerative mimics (e.g., frontotemporal dementia) from primary psychiatric illness. In mood disorders, higher NfL correlates with cognitive impairment and brain atrophy, suggesting it could track “neuroprogression” over time (Bavato et al., 2024; Jy Kang et al., 2025).

In Vivo neuroimaging offers another tools for diagnosis and trial stratification. PET imaging of the 18-kDa translocator protein (TSPO) on glial cells is the leading method for visualizing neuroinflammation in humans. A 2025 meta-analysis of 156 TSPO-PET studies found overall elevated glial activation across CNS disorders, with the most significant effects in Alzheimer’s and MS (De Picker et al., 2023). Mood disorders showed more minor yet significant increases in TSPO binding within limbic brain regions. Notably, the meta-analysis also highlighted extreme heterogeneity driven by methodological factors (e.g., choice of ligand, quantification model, patient TSPO genotype). As a result, TSPO-PET remains mainly a research tool for confirming target engagement in trials, rather than a routine clinical biomarker.

Next-generation biomarker strategies might benefit by assessing multi-dimensional signatures. One promising avenue for which this is being applied is the gut microbiome, as distinct psychiatric disorders have been shown to have unique microbial compositions. A 2025 study used machine learning on gut sequencing data to diagnose MDD, bipolar disorder, and schizophrenia with 80–97% accuracy based on these “dysbiosis” signatures (Ma et al., 2025). If validated, such microbiome profiles could serve as non-invasive diagnostic tools and guide probiotic and fecal microbiota transplant (FMT) therapies. Similarly, autoantibody panels might define an “autoimmune psychiatric” subtype. Discoveries of antibodies against NMDA receptors, muscarinic receptors, or systemic antigens (anti-Ro52, anti-TPO) in subsets of patients suggest that screening could identify those who may benefit from immunotherapy. Finally combining these modalities with imaging tools that can stratify psychiatric “biotypes” (i.e., functional magnetic resonance imaging or TSPO-PET) could further strengthen biomarker predictions. In summary, the emerging paradigm of multimodal biomarker guidance, which combines markers of peripheral inflammation (hs-CRP), neuronal injury (NfL), neuroinflammation imaging (TSPO-PET), microbiome, and autoantibody data, might one day be used to identify unique subtypes of patients with neuropsychiatric or neurological diseases and tailor individualized approaches for treatment. Figure 3 aligns these immunobiotypes with targeted therapies and their anticipated behavioral outcomes, offering a precision-medicine roadmap for restoring CONB network homeostasis.

Figure 3:

Precision-medicine framework based on the CONB model, aligning immune-biotype biomarkers for patient stratification with targeted neuroimmune therapies and anticipated behavioral outcomes. (Panel A) Patient biomarkers and stratification. Three distinct immune-biotypes are identified based on peripheral biomarker profiles. Inflammatory biotype (~35% patients) characterized by elevated high-sensitivity C-reactive protein (hs-CRP >5mg/L), interleukin-6 (IL-6), and tumor necrosis factor-α (TNF-α), representing systemic inflammation-driven neuropsychiatric symptoms. Autoimmune biotype (~20% patients) defined by presence of brain-reactive autoantibodies, anti-NMDA receptor antibodies (Anti-NMDAR), and anti-thyroid peroxidase antibodies (Anti-TPO), indicating autoimmune-mediated CNS involvement. Gut dysbiosis biotype (~20% patients) identified through altered microbiome signatures, reduced short-chain fatty acids (SCFAs), and elevated lipopolysaccharide (LPS) levels, reflecting gut-brain axis dysfunction. (Panel B) Mechanism-driven targeted therapies. Biomarker-guided therapeutic interventions target specific CONB network components. Anti-cytokin biologics including anti-TNF (infliximab) and anti-IL-6 receptor (tocilizumab) agents block systemic inflammatory pathways for inflammatory biotype patients. JAK inhibitors (tofacitinib) simultaneously suppress cytokine signaling and stress-induced myelopoiesis from bone marrow nodes. B-cell depletion therapy (rituximab) eliminates autoantibody-producing cells in autoimmune biotype patients. Microbiome interventions including psychobiotics (Lactobacillus helveticus, Bifidobacterium longum) and fecal microbiota transplantation (FMT) restore gut-brain axis homeostasis. Vagus nerve stimulation (VNS) activates the cholinergic anti-inflammatory pathway, modulating spleen-mediated immune responses across multiple biotypes. (Panel C) Scientific and clinical outcomes. Targeted interventions aim to restore behavioral homeostasis through distinct but complementary mechanisms. Improved mood and reduced depression result from normalized cytokine signaling and restored monoaminergic neurotransmission. Enhanced cognitive function emerges from reduced neuroinflammation, decreased microglial activation, and improved synaptic plasticity. Increased stress resilience follows HPA axis normalization and enhanced cholinergic anti-inflammatory responses. Reduced anhedonia correlates with restored dopaminergic signaling in reward circuits. Decreased anxiety symptoms accompany normalized glutamate homeostasis and reduced amygdalar hyperactivation. Biomarker normalization, including reduced inflammatory markers (CRP, IL-6), serves as objective measures of therapeutic efficacy.

Emerging Technologies and Future Directions

Advances in technology are accelerating psychoneuroimmunology. Single-cell multi-omic platforms now allow mapping of genetic risk to cell-type-specific functions. For example, a 2025 study profiled gene expression and chromatin accessibility in iPSC-derived human neurons under both resting and activated conditions (Liang et al., 2025). This revealed thousands of context-dependent eQTLs and caQTLs, linking psychiatric GWAS variants to specific target genes only evident during neuronal activation (Liang et al., 2025). Such studies pave the way for identifying precise molecular targets within complex genetic loci.

Humanized disease models are also improving. Brain “organoids” (3D cultures) and “assembloids” (fused-region organoids) provide human-specific platforms for studying neuroimmune interactions. Incorporating microglia-like cells into these organoids (“neuroimmune organoids”) creates a mini human brain environment. Recent work has even transplanted these organoids into rodents to achieve vascularization and maturation, showing that human microglia fully mature only within a native neural context (Schafer et al., 2023; Tian et al., 2025). These platforms will enable the testing of neuroimmune drugs in systems that mimic human brain architecture and immune cell interactions.

Finally, overcoming the blood–brain barrier is a significant focus for the delivery of neuroimmune therapies. Novel strategies now harness molecular, cellular, and physical approaches (for review see (Wu et al., 2023)). Molecular nanocarriers (liposomes, polymeric nanoparticles, engineered exosomes) can be decorated with BBB-targeting ligands (e.g., transferrin) to piggyback on endogenous transport routes. Cell-based “Trojan horses,” such as mesenchymal stem cells, exploit natural homing to deliver payloads across the BBB (Mitchell et al., 2021). Physical methods like MRI-guided focused ultrasound (with microbubble contrast) can transiently open tight junctions in a targeted brain region to allow passage of certain drugs too big to otherwise enter the brain (Jolesz, 2009). These convergent technologies promise to enable the delivery of large biologics and gene therapies to the CNS, which was previously impossible.

Conclusion

The evidence synthesized in this review solidifies a paradigm shift in understanding behavior, moving from isolated organ-brain dialogues to the integrated framework we have termed the Cross-Organ Neuroimmunology of Behavior (CONB) Network. By first establishing its fundamental mechanisms and then exploring its constituent organ nodes, we have illustrated how signals from the gut, bone marrow, lungs, and beyond converge to regulate neuroinflammation, mood, and cognition. Viewing neuropsychiatric and behavioral disorders through the lens of the CONB Network reframes them as systemic node-edge dysregulations rather than purely brain-centric conditions. As detailed in our system-level analysis, future interventions, whether microbiome editing, anti-cytokine biologics, or bioelectronic modulation—must be evaluated not for their effect on a single component, but for their capacity to restore equilibrium across the entire CONB network. This approach paves the way for a new therapeutic paradigm where patients are stratified into immunobiological subtypes—such as ‘inflammatory depression’ versus ‘autoimmune depression’—and treated with matched, network-level interventions. Embracing this holistic and personalized framework is crucial for ultimately realizing the promise of neuroimmune medicine and enhancing both mental and physical health.