MECHANORESPONSIVENESS OF THE NEURO-IMMUNE CROSS-TALK IN BONE AND JOINTS: A NARRATIVE REVIEW WITH IMPLICATIONS FOR FORCE-BASED MANIPULATIONS

Abstract

INTRODUCTION:

Recent advances in osteoimmunology have illuminated the significant role of neuroimmune interactions in bone and joint health. The neuroimmune system critically regulates bone cell function through a complex network of molecular mediators - including signals from the sympathetic and parasympathetic nervous systems, pro- and anti-inflammatory cytokines, cortisol and other hormones, and growth factors. These mediators interact within a nonlinear network where each component can influence others reciprocally, with the brain serving as a central coordinator.

AIM:

Our goal in this narrative review is to: 1) Provide an overview of roles the neuroimmune system is involved in bone and joint health; 2) Review which therapeutic interventions delivered by musculoskeletal clinicians might influence bone-related neuroimmune responses (or could be biologically plausible); and 3) Discuss clinical implications and gaps in the literature that need to be filled.

IMPLICATIONS:

Considerable evidence exists for the effects of physical activity on bone and bone-related neuroimmune responses. In contrast, there is a paucity of evidence in the literature for actual changes induced in bone by any force-based manipulation (e.g. acupuncture, massage therapy, manual therapy, spinal manipulation). Yet, a number of early studies examining the responses of bone-related neuroimmune responses to force-based manipulations have found to exert some neuroimmune modulatory effects. Clinicians should tailor physical activity, manual therapy and exercise interventions based on patient-specific factors like age, hormonal status, and comorbidities, which affect neuroimmune responses and bone health.

Introduction

Due to its pivotal function in immune cell production and regulation, bone marrow has been described as the “central immune system” [1]. Structurally and functionally, bone marrow and bone matrix cells are interconnected with the central nervous system, highlighting the integrative nature of bone with the neuroimmune system [2]. Traditionally, mechanical loading [3], growth factors [4] and hormonal regulation [5] have been recognized as primary influencers of bone homeostasis [6]. The responses of bones and joints to mechanical loading are modulated by a multifaceted array of factors that depend on individual factors and comorbidities. Some factors, like biological sex, hormonal status, and age have well stratified effects on bones, e.g., larger and stronger bones in males, loss of bone density in postmenopausal women, peaks in bone density in young adulthood yet declines in older adults [7–9]. Other factors, including emotional state (e.g., psychological stress and depression) and metabolic health co-morbidities (e.g., hypertension, obesity, hypercholesterolemia, and diabetes) [10, 11], also have marked impacts on musculoskeletal health. Recent advances in osteoimmunology have illuminated the significant role of neuroimmune interactions in bone remodeling and health [2]. Neuroimmune interactions in bone are currently being explored both in clinical and basic science contexts, which is an exciting new paradigm for considering musculoskeletal biology.

As an organ where CNS inputs, immune outputs, hormones, and mechanical forces converge, bone is a compelling focal point for understanding neuroimmune regulation across health and pathology. Our goal in this review is to: 1) Provide an overview of roles neuroimmune system plays in bone and joint health; 2) Review which interventions deliverable by musculoskeletal clinicians might influence bone-related neuroimmune responses (or could be biologically plausible); and 3) Discuss clinical implications and gaps in the literature that need to be filled. For objectives 2 and 3, our main focus will be reviewing the literature on touch and force-based interventions, as well as literature on whole body exercise/physical activity. We will use the term force-based manipulation for treatments including gentle human touch, instrument-assisted soft tissue manipulation, massage therapy, manual therapy, acupuncture, osteopathic manipulative therapy, and spinal manipulation treatments, in accordance with the National institutes of Health (NIH) definition [12]. Interest in integrative and nonpharmacologic approaches to pain management has been steadily growing, as has scientific interest regarding the underlying mechanisms of these treatments. This review serves to explore what is currently known and where neuroimmune mechanisms may be leveraged to improve musculoskeletal health [13].

1. Overview of key neuroimmune pathways involved in bone and joint health

Bone is a dynamic tissue that undergoes continuous remodeling throughout life, orchestrated by the coordinated activities of osteoblasts (bone-forming cells), osteoclasts (bone-resorbing cells), and osteocytes (embedded cells responsible for coordinating bone formation and resorption). Mesenchymal stromal cells reside in the bone marrow and give rise to a wide range of musculoskeletal cell types, including osteoblasts, osteoclasts, fibroblasts, adipocytes, and chondrocytes. It is now known that the neuroimmune system critically regulates the function of these bone cells through a complex network of molecular mediators - including signals from the sensory nervous system, autonomic sympathetic and parasympathetic nervous systems, immune system molecules (e.g., pro- and anti-inflammatory cytokines), metabolic mediators, and hormones (summarized in Table 1). These mediators interact within a nonlinear network where each biological system can influence others reciprocally (i.e., crosstalk), with the brain often serving as a central coordinator [14]. Disruptions in the delicate balance of these systems can lead to dysregulation of neuroimmune responses [15], adversely affecting bone remodeling and overall bone health. For example, chronic overactivity of a system (e.g., autonomic, immune, metabolic, or endocrine) can induce a domino effect on the other interconnected systems, leading one or more to overcompensate or become dysregulated. In the case of bone and joints, this can lead to excessive bone loss or growth, and/or synovial membrane and cartilage matrix pathologies. In this section, we cover the major known neuroimmune paradigms that influence bone health.

Table 1.

Neuroimmune modulators that affect bone volume

Modulator	General Effects on Bone Volumea
Promotion (pro-anabolism, bone formation and accrual)
Acetylcholine (ACh, neurotransmitter)	Primarily stimulates osteoblast proliferation and differentiation, leading to bone formation. Yet, ACh receptors (both nicotinic and muscarinic) are located on osteoblasts, osteocytes, and osteoclasts, allowing direct influence on activity of each cell type.
Calcitonin gene-related peptide (CGRP; neuropeptide neurotransmitter)	Mechanical loading stimulates CGRP release from nerves, which inhibits osteoclastogenesis and promotes bone formation. Negatively regulates receptor activator of nuclear factor kappa B ligand (RANKL).
Dopamine (neurotransmitter)	Significant impact on bone metabolism by binding to dopamine receptors on osteoblasts, which modulates osteoblast activity and bone matrix accrual. Concentration dependent: low concentrations reduce bone mineralization; high concentrations modestly increase mineralization by osteoblastic cells. Reduces differentiation of osteoclasts.
Ghrelin (multifunctional metabolic hormone)	Stimulatory effects on many metabolic processes including bone formation.
Insulin like growth factor (IGF-1, growth factor)	Essential for longitudinal bone growth, skeletal maturity, bone mass acquisition.
Interleukin 10 (IL-10; anti-inflammatory cytokine)	Potent anti-inflammatory cytokine that promotes bone formation and inhibits osteoclastogenesis, primarily by reducing production of pro-inflammatory cytokines.
Bone loss through suppression or increased osteoclast activity (pro-catabolism)
Glucocorticoids (multifunctional metabolic hormones, e.g., cortisol or corticotropin-releasing hormone, CRH)	Suppresses osteoblast activity, inhibits collagen synthesis, and increases RANKL expression.
Interleukin 1 alpha or beta (IL-1α or β; pro-inflammatory cytokines)	Local and circulating levels activate osteoclast and macrophage activity.
Interleukin 6 (IL-6; pro- and anti-inflammatory cytokine, and myokine)	An osteoclastogenic cytokine that activates osteoclast and macrophage activity, when in the presence of other pro-inflammatory cytokines.
Noradrenaline/Norepinephrine (neurotransmitter)	Signaling through β2-adrenergic receptors downregulates bone formation and promotes RANKL expression for recruiting osteoclasts; this signaling is also necessary for mechanical loading-induced bone formation.
Neuropeptide Y (NPY, neurotransmitter)	Inhibits bone formation through direct nerve effects (via nerves associated with blood vessels in bone). Overexpression in arcuate nucleus of hypothalamus causes bone loss.
Receptor activator of nuclear factor kappa B ligand (RANKL, osteoclastogenic cytokine)	An osteoclastogenic cytokine that promotes osteoclast differentiation and activity.
Substance P (neuropeptide neurotransmitter)	A nociceptive (pain) neurotransmitter. Induces expansion of macrophage and osteoclast precursor populations (catabolic). Can stimulate late-stage osteoblastogenesis (anabolic).
Tumor necrosis Factor alpha (TNF-α; pro-inflammatory cytokine)	Pro-inflammatory cytokine that activates osteoclast and macrophage activity.
	Site Dependent effects/Other
Leptin (sympathetic metabolic hormone)	Acts on β2-adrenergic receptors. Concentration dependent: High concentrations peripherally play a positive role in bone formation. Lower concentrations in central nervous system play a negative role, i.e., suppression of bone volume.
Osteocalcin (biomarker of osteoblastic activity; metabolic hormone)	Serum and urinary biomarker of osteoblast activity. Participates in body regulation and homeostasis and crosses the blood brain barrier to affect cognitive function.
Serotonin (5-HT, neurotransmitter)	Site dependent: In brain, inhibits sympathetic nervous system output to bone, favoring bone density and formation, reducing bone resorption. In gut, negatively affects bone accrual.
Vasoactive intestinal peptide (VIP; neuropeptide neurotransmitter)	Has an osteo-inductive effect on mesenchymal cells, suggesting it may promote bone formation. Site Dependent: Neural VIP directly inhibits resorption and may modulate RANKL/OPG. Circulating VIP regulates bone metabolism by binding to receptors on both osteoblasts and osteoclasts; thus, able to influence both cell types.

^aMany of these effects depend on context (such as concentration, receptors present, and model system); OPG, osteoprotegerin. References can be found in the main text.

1.1. Crosstalk between Neural Systems and Bone/Joint Tissues

Bone tissue exhibits a complex neural innervation network from somatosensory and sympathetic nervous systems [16]. This neural network is integrally coupled with the extensive vascular system in bone tissues [17]. Sensory and sympathetic nerve axons follow these vessels to terminate in the periosteum, bone marrow cavity, trabecular (cancellous bone), and cortical bone (Figure 1A–D), as well as in synovium membranes, subchondral bone and other joint tissues (Figure 1E). This neural innervation network engages in bidirectional communication with bone and hematopoietic cells, thereby regulating osteogenesis, osteoclastogenesis, bone remodeling dynamics, skeletal adaptation to mechanical and systemic stimuli, and responses during fracture healing [18]. Neurotransmitters made and released from peripheral nerves innervating bone and joint tissues are able to effect the proliferation, differentiation, activation, suppression of bone cells, and repair (Table 1) [2]. Several neurotransmitters made and released by the brain also influence distantly located innervation networks and bone cells (Table 1) [2]. Additionally, these nervous systems and bone cells interact with neuroendocrine pathways in the regulation of bone metabolism.

Figure 1.

Distribution of the nerves within the periosteum, bone, and bone marrow. (A) Scheme of the nerves in the bone. There are mainly two types of nerves: sensory nerves (yellow) and sympathetic nerves (green) that express calcitonin gene related protein (CGRP, sensory), Substance P (Sub P, sensory), nerve growth factor (NGF, sensory), vasoactive intestinal protein (VIP, sympathetic), and Neuropeptide Y (NPY, sympathetic). Modified from [26]. (B–D) Representative images of the nerve fibers distributed in the periosteum (B), cortical bone (C), and bone marrow (D). Symbols: C - cambium layer of the periosteum, CB - cortical bone, F - fibrous layer of the periosteum, and HC - Haversian canals. By immunofluorescence staining, tyrosine hydroxylase immunoreactive sympathetic axons is shown in yellow, CGRP immunoreactive axons shown in green, and endothelial cells of blood vessels shown in red (CD31 immunoreactive). A-D Modified by [26] from Figures 2 and 3 of [140]. (E) Scheme of the nerves in a synovial joint. Created with BioRender.com.

Direct effects of nerves on bones and joints, and their neuromodulators

Sensory nerve fibers in bone can directly contribute to bone anabolism through the release of neurotransmitters from nerve terminals, including calcitonin gene-related peptide (CGRP), substance P, and vasoactive intestinal peptide (VIP, Figures 1 and 2, Table 1) [16]. Sensory nerves in bone and joints are not limited to afferent functions (e.g., signaling pain after fracture and inflammatory conditions). Sensory nerves also secrete neurotransmitters to induce effects in target tissues (i.e., efferent effects) – with sensory nerve secretions primarily involved in bone formation and regeneration processes. For example, CGRP is a neuropeptide produced and released by sensory nociceptive nerves (Figures 1 and 2, Table 1). Mechanical loading of bone stimulates this release of CGRP [16], which then promotes bone formation [19] and inhibits osteoclastogenesis [20]. The underlying mechanism for the latter is that after release, CGRP negatively regulates Receptor Activator of Nuclear Factor κB Ligand (RANKL) expression, an osteoclastogenic cytokine that promotes osteoclast differentiation and activity (Table 1) [21]. The inhibition of RANK shifts the balance to osteoblast activity and thereby bone formation. Substance P, more commonly known as a nociceptive (pain) neurotransmitter [22], is also present in sensory nerves in bone, bone marrow, periosteum, and synovial membranes (Figures 1–3, Table 1). This innervation is highly dynamic, with increased density of substance P immunostained axons (as well as CGRP immunostained axons) in bone after fracture and during callus regeneration [22]. Substance P contributes to fracture repair by regulating osteoblast differentiation in a biphasic manner: increasing cell death (apoptosis) in early stages while reducing apoptosis in later stages of osteoblastic differentiation [23]. Substance P’s efferent functions in bone are also supported by studies showing delayed fracture repair after denervation [24], and that mouse models lacking substance P have impaired fracture repair [25]. Lastly, nerve growth factor’s (Figure 1) release from nerves into bone tissues is required during development for the induction of primary and secondary ossification [26]. Thus, neurotransmitters released from sensory nerves located in bones are primarily involved in promotion of bone volume in a variety of mechanisms in response to mechanical loading and during fracture repair. Figure 2 shows a summary of the many functions and conditions that sensory neurotransmitters play in bones and joints.

Figure 2.

Schematic illustration of roles of sensory nerves and their neuropeptides in regulating the differentiation and formation of different types of bone and joint tissues (and cells) in various bone-related diseases. Semaphorin 3A (Sema3A), a selective repellent for calcitonin gene related protein (CGRP); Substance P (Sub P), immunopositive sensory nerve axons, that can inhibit aberrant sensory sprouting. ↑: upregulation; ↓: downregulation; *: the neuropeptides that have opposite effects in these sensory nerves. Arrows are color coded to the neuropeptide of action: CGRP (light purple), Sub P (brown-orange), Sema3A (green). Modified from [141] and created with BioRender.com.

Figure 3.

Sympathetic nervous system mediated control of bone remodeling. (#1) Neurons within the ventromedial hypothalamus (VMH) and NPY-sensitive neurons in the arcuate nucleus (Arc) excite (+) sympathetic preganglionic neurons within the spinal column through multiple mechanisms, which may include synapsing with intermediate neurons in sympathetic ganglia (#2). Sympathetic postganglionic neurons innervate the bone marrow compartment by following vessels into the bone (#3). Norepinephrine (NE) and NPY are released simultaneously from sympathetic nerve axons in the marrow compartment, which inhibits (−) bone formation by osteoblasts (blue line) and activates (+) bone resorption by osteoclasts (maroon), respectively (#4). Osteocalcin is a product of bone resorption that regulates energy metabolism, neural development and behavior. Osteocalcin and other products of bone resorption may feedback to modulate neural control of bone remodeling (#5). One mechanism through which sympathetic nervous system output to bone is modulated is through leptin-mediated inhibition of serotonergic neurons (blue, in upper left of figure) in the raphe nucleus of the brainstem (#6). Activation of serotonergic neurons inhibits (−) ventromedial hypothalamus (VMH) neurons responsible for sympathetic outflow to bone (#1). Figure from [142]

Sympathetic nerves in bone are associated with blood vessels and release VIP, neuropeptide Y (NPY), and norepinephrine (Figure 3, Table 1) [16]. Nerves that are immunoreactive for VIP can be found in the periosteum, bone marrow, and synovial joint membranes [16]. VIP binds to different types of G-protein coupled transmembrane receptors on osteoblasts, osteogenic cells, and osteoclasts. Although more studies are needed for a full understanding, nerve released VIP acts on bone cells to inhibit bone resorption likely by modulation of RANKL, RANK and osteoprotegerin pathways (RANKL/RANK/OPG; Table 1) [2]. The sympathetic nervous system primarily exerts its effects on bone metabolism through the release of norepinephrine and NPY from sympathetic axons present in bone marrow and periosteum, as shown in Figure 3. Sympathetic neurons innervate the bone marrow by following vessels into the bone. Norepinephrine and NPY are released simultaneously from sympathetic nerve axons into the marrow compartment and effectively activate bone resorption by osteoclasts and inhibit bone formation by osteoblasts, respectively [27] (Figure 3).

Osteocalcin is a product of bone resorption that regulates energy metabolism, neural development and cognitive behavior (Figure 3, Table 1). Osteocalcin and other products of bone resorption may provide feedback to the brain to modulate neural control of bone remodeling, although more investigation is needed.

Related, pharmacological studies have shown that the osteogenic effects of physical activity can be abrogated by β2-adrenergic agonists (a class of sympathomimetic agents), suggesting a potential override of physical activity-induced bone formation by enhanced sympathetic activation [28]. In scenarios where mechanical loading surpasses the beneficial threshold and results in microdamage accumulation, heightened sympathetic tone may shift the balance towards bone catabolism. This can occur through the suppression of loading-induced anabolic signaling and upregulation of osteoclastogenic cytokines, e.g., RANKL and interleukin 6 (IL-6) [29].

Indirect Effects of Autonomic Nervous System Signaling on Bone

The central brain regions involved in regulating the sympathetic nervous system have been identified as the hypothalamic-pituitary-adrenal axis and the sympathetic-adrenal-medullary system (Figure 3) [2, 15]. Neurons within the ventromedial hypothalamus and NPY sensitive neurons in the arcuate nucleus excite sympathetic preganglionic neurons within the spinal column through multiple mechanisms, including synapsing with intermediate neurons in sympathetic ganglia. Sympathetic postganglionic neurons innervate the bone marrow compartment by following vessels into the bone and elicit effects, as described above. The ventromedial hypothalamus neurons are regulated by brain levels of leptin derived from the gut via an indirect central nervous system relay [30, 31]. Thus, the sympathetic nervous system plays a central role in regulating bone volume through a multi-organ crosstalk mechanism.

Chronic overactivation of the sympathetic nervous system can occur during psychological stress-related conditions, which, if prolonged, can reduce bone health (e.g., significant bone loss) [32]. This can occur via heightened activity in the hypothalamic-pituitary-adrenal axis regions shown in Figure 3. Animal studies using leptin-deficient mice confirm the link between leptin and bone metabolism, since these mice demonstrate low sympathetic activity and significant bone loss [33].

The parasympathetic nervous system appears to counterbalance these sympathetic effects. Parasympathetic signaling increases bone mass by promoting osteoclast apoptosis and inhibiting resorption through nicotinic receptors that respond to the neurotransmitter acetylcholine (ACh, Table 1) [34]. Studies show that ACh agonists stimulate bone formation, while antagonists inhibit osteoblast activity. Clinical evidence suggests that preventing ACh degradation reduces hip fracture incidence by 20–30% [35]. The metabolic regulatory mechanisms of ACh are not fully understood in bone. Critically, questions remain about ACh’s effects on osteocytes and whether it originates from parasympathetic or sympathetic sources.

1.2. Crosstalk between the Neuroinflammatory Systems and Bone

A balance of this sympathetic and parasympathetic activity is needed to appropriately regulate the inflammatory/immune system – highlighting the interconnected nature of all these systems. The sympathetic nervous system tends to stimulate pro-inflammatory cytokine production, which would enhance bone and joint catabolism by activating osteoclast differentiation and phagocytic activity, while the parasympathetic nervous system inhibits these responses, which would shift the balance to bone formation [36]. In addition, a balance between pro- and anti-inflammatory responses, both locally in bone, systemically, and in the central nervous system control centers, is needed for bone growth and repair. When that balance is disrupted, bone growth and repair may become dysregulated. In this section we discuss several such examples, including the involvement of persistent local neuroinflammation, chronic systemic inflammation, or brain neuroinflammation in activating the neuroimmune system, which can induce deleterious effects on bone. These conditions can contribute to important clinical maladies like osteoporosis, rheumatoid arthritis and other inflammation-related conditions.

Pro- and Anti-Inflammatory Cytokines

Cytokines are a large group of proteins with divergent effects on the immune system, some stimulating the immune system while others are suppressive. Pro- and anti-inflammatory cytokines are produced by multiple cell types and sources (by nerve, bone, bone marrow cells, and immune cells peripherally, as well as by neurons, glia, and immune cells in the central nervous system) and exert both local and systemic effects [36]. In bone and joints, local immune and injured cell populations produce pro-inflammatory cytokines that recruit immune cells into the injury site. In contrast, anti-inflammatory cytokines can either dampen the production of pro-inflammatory cytokines (e.g., IL-10 inhibition of tumor necrosis factor alpha (TNF-α) production [37]) or enhance tissue repair (e.g., transforming growth factor [38]). Combined, these inflammatory responses can stimulate tissue adaptation, repair, resorption, or further injury by enhancing phagocytic activity, based on the balance of these responses and/or other superimposed processes in the affected tissues.

Pro-inflammatory cytokines that impact bone metabolism include interleukin 1 (IL-1) and TNF-α [as well as IL-6, although it is both pro- and anti-inflammatory depending on other cytokines in the milieu] (Table 1). These cytokines act on osteoclasts and their progenitors to promote osteoclast proliferation, differentiation, and resorption activity, which enhances bone degradation [39]. They can also act on subchondral osteoclasts and macrophages within joints to contribute to cartilage and subchondral bone degradation. Pro-inflammatory cytokines increase the expression of a key osteoclastogenic cytokine, RANKL, by osteoblasts and bone marrow stromal cells, which then promotes osteoclast differentiation and activity (Table 1), as mentioned earlier.

Anti-inflammatory cytokines, e.g., interleukin-10 (IL-10) and transforming growth factor beta 1 counteract the effects of pro-inflammatory cytokines, thereby, promoting bone formation and inhibiting osteoclastogenesis [40] (Table 1). IL-10 reduces bone loss by suppressing the production of TNF-α and IL-1. Studies of rheumatoid arthritis in which IL-10 is blocked by antigen specific antibodies indicate subsequent reduction of rheumatoid arthritis induced inflammation and bone erosion [41]. Another anti-inflammatory cytokine, transforming growth factor beta 1, is critical for osteoblast differentiation and matrix production, while also inhibiting osteoclast activity, thus aiding the balance between bone formation and resorption [42].

Peripheral Neuroinflammation

Substance P not only impacts bone metabolism direct, as outlined above; it also plays a significant role in inflammatory responses, particularly in neurogenic inflammation initiated by the peripheral nervous system. Neurogenic inflammation is a form of inflammation initiated by activation nociceptive neurons rather than by immune system triggers. The release of substance P from axon terminals in affected tissues potentiates inflammatory responses by augmenting the production of pro-inflammatory mediators and cytokines in immune cells and injured cells [43]. Moreover, substance P acts synergistically with transforming growth factor beta 1 in inflammatory processes [44] to extend the duration of substance P signaling and amplify downstream inflammatory cascades. In the context of arthritic conditions, several neuropeptides, CGRP, substance P, neurokinin A, and NPY have also been found in synovial fluid from arthritic joints of patients with rheumatoid arthritis [45] in association with inflammatory joint processes (Figure 4) [46]. These findings suggest that changes in neuropeptide signaling directly contribute to joint catabolism through neurogenic inflammatory processes.

Figure 4.

Inflammatory responses in rheumatoid arthritis. Interleukin 1, 6, and 10 (IL-1, 6 and 10); Matrix metalloproteinases (MMPs); Receptor activator of nuclear factor kappa B ligand (RANKL); Substance P (Sub P); Tumor necrosis factor (TNF). Created with BioRender.com using a template created by Samara Ona, and modified by Barbe, M. (2025).

Role of Central Neuroinflammation

Systemically elevated pro-inflammatory cytokines can also stimulate brain cells and regions, such as the hypothalamic-pituitary-adrenal axis to produce additional cytokines and corticosteroids (to be discussed in the next section), creating a feedback loop that influences inflammatory cytokine levels centrally, peripherally, and reciprocally [15, 47]. These pro-inflammatory cytokines can access the brain by crossing the blood-brain barrier directly or by slow diffusion at specific sites (e.g., ependymal cells, endothelial cells, and circumventricular organs), or by sending signals along nerves or across the blood-brain barrier via second messengers [48, 49]. Cytokine production and signaling, in turn, becomes amplified in neurons and glial elements. Increased cytokine responses in the brain can lead to the enhanced release of inflammatory cytokines from the brain into the blood stream. This results in systemic increases in pro-inflammatory cytokines, which can also result in widespread activation of osteoclasts and enhanced bone resorption. This is highlighted in multiple sclerosis in which pro-inflammatory cytokines are released from the brain into the circulation [50], as discussed next.

Impact of Autoimmune Diseases on Bone and Joint Health

Autoimmune diseases, including rheumatoid arthritis and systemic lupus erythematosus, are associated with significant bone loss due to chronic inflammation and immune dysregulation. Rheumatoid arthritis is a multifactorial inflammatory disease characterized by excessive proliferation of synovial cells, immune cell infiltration, and cartilage and bone destruction (Figure 4). Key pathogenic factors in rheumatoid arthritis include increased synovial membrane production of potent inflammatory mediators, particularly TNF-αand IL-1 [51]. Rheumatoid arthritis is also characterized by T cells and macrophages that produce high levels of TNF-α, IL-1 and IL-6, each capable of promoting osteoclast differentiation and bone erosion (Table 1). Targeted therapies inhibiting TNF-α (e.g., etanercept and infliximab) reduce bone loss in rheumatoid arthritis patients, confirming the central role of this cytokine in bone resorption [52]. Infiltrating T cells and macrophages release other inflammatory cytokines into the synovial fluid, oncostatin M (related to interleukin 6) and additional IL-1. TNF-α and oncostatin M work synergistically to increase production of collagenases (matrix metalloproteinases, Figure 4) by diseased synoviocytes and chondrocytes [51], inducing excessive collagen degradation that can lead to cartilage matrix collapse [53].

Osteoarthritis is also the result of a disruption in the calibrated equilibrium between bone and joint anabolic and catabolic pathways. Pro-inflammatory cytokines, chemokines, and extracellular matrix fragments propagate the elevated synthesis of matrix metalloproteinases and additional inflammatory cytokines [54], culminating in the progressive joint destruction that marks this pathology. In human osteoarthritis, immune cell activity in the bone and joint can be imaged with positron emission tomography and [¹¹C]-PBR28, a radioligand for the inflammatory marker 18-kDa translocator protein (TSPO). This has been shown in patients suffering from knee osteoarthritis, as the elevated signal is specific to the symptomatic knee [55]. While the function and cellular origin of the peripheral TSPO signal is not fully understood, in human rheumatoid arthritis pannus, the [¹¹C]-PBR28 ligand was found to bind to TSPO receptors on activated fibroblast-like synoviocytes, undifferentiated and reparative macrophages and, to a lesser degree, on other immune cells [56]. Similarly, multiple sclerosis patients experience reduced bone density due to immobility and systemic inflammation. Compston [57] reported that individuals with multiple sclerosis had lower bone mineral density than healthy controls, likely due to reduced physical activity and elevated pro-inflammatory cytokine levels that exacerbate bone loss.

These findings combined highlight the need to address systemic inflammation and neuroimmune interactions in managing bone and joint health in autoimmune diseases.

1.3. Crosstalk Between Neuroendocrine Hormones and Bones

Several hormones have neuroimmune effects on bones and joints, including leptin and ghrelin (Table 1). Leptin is a multifunctional neuroendocrine hormone secreted by adipocytes. It signals satiety and is involved in immune responses, energy expenditure, angiogenesis, reproductive processes, and bone formation. Leptin can directly stimulate growth and differentiation of osteoblasts and chondrocytes in bones and joints (Table 1); however, increased levels of circulating leptin can lower bone mass via an indirect central nervous system relay. Specifically, leptin inhibits serotonergic neurons in the raphe nucleus of the brain stem, that in turn signal neurons in the ventromedial hypothalamus to regulate the sympathetic nervous system to decrease bone mass (Figure 3). This means that leptin has a central anti-osteogenic function by altering sympathetic tone. The current consensus is that leptin at high concentrations plays a positive role on bone formation peripherally, yet a suppressive role when at lower concentrations within the central nervous system.

Ghrelin is another a multifaceted hormone with stimulatory effects on food intake, fat deposition, growth hormone release, muscular atrophy, and bone formation [58] (Table 1). Ghrelin is primarily synthesized by cells in the gastrointestinal tract and pancreas mostly, yet can be made by fibroblasts and osteoblasts, with site-specific effects. It is transported through blood circulation to the hypothalamus, where it stimulates growth hormone secretion after binding to the growth hormone secretagogue receptor. Regarding bone function, ghrelin has dual roles in osteoclastogenesis, inhibiting osteoclast progenitors locally yet stimulating osteoclastogenesis when systemic (Table 1). Systemic ghrelin interacts with leptin to regulate osteoclastogenesis and bone metabolism in an age-dependent manner (ghrelin’s systemic osteoclastogenic effect reduces with age). Recent research also shows that ghrelin can enhance blood-brain barrier permeability, potentially modulating cytokine access to the central nervous system [59].

Thus, ghrelin and leptin have opposite effects on bone metabolism, and ghrelin’s systemic osteoclastogenic activity is suppressed by leptin.

Chronic Psychological Stress, Glucocorticoids and Bone Loss

Recent advances in neuroendocrinology have also uncovered direct connections between the hypothalamus and bone tissue, supportive of a central nervous system role in skeletal homeostasis [60]. This finding has led to the hypothesis that neural regulation of bone metabolism may, in turn, influence global energy homeostasis through endocrine feedback mechanisms. For example, the hypothalamic-pituitary-adrenal (HPA) axis is activated in response to psychological stress, leading to the release of glucocorticoids, like cortisol in humans and corticosterone in animals (Figure 5, Table 1). Prolonged exposure high levels of glucocorticoids suppress osteoblast activity, inhibit collagen synthesis, and increase RANKL expression, the latter promoting osteoclast-mediated bone resorption [61]. Chronic psychological stress as well as depression result in lower bone mineral density, increased risk of osteoporosis and fracture [62], and impair tissue healing responses, particularly in individuals with psychological stress-related disorders (Figure 5) [63]. This impact of chronic stress on bone health underscores the importance of neuroimmune regulation in maintaining bone homeostasis.

Figure 5.

Mechanisms associated with the impaired healing response resulting from psychological stress. Hypothalamic pituitary axis (HPA); Sympathetic nervous system (SNS). From [15].

2. Interventions that influence these bone related neuroimmune responses

Because neuroimmune signaling in bone is plastic, clinicians can steer it with modifiable inputs (e.g., mechanical loading, autonomic balance, and systemic milieu) to bias remodeling toward formation and repair. Emerging evidence shows that exercise dosing, manual techniques, and strategies that reduce allostatic load (sleep, stress management, anti-inflammatory nutrition) can shift CGRP and substance P signaling, sympathetic–parasympathetic tone, and RANKL–OPG dynamics relevant to bone turnover. This section reviews the current clinical and preclinical evidence for these interventions, emphasizing documented effects on CGRP and substance P signaling, sympathetic–parasympathetic balance, and RANKL–OPG dynamics, along with safety considerations and populations most likely to benefit. We appraise strength of evidence and key gaps to guide practitioner judgment and identify priorities for future trials.

2.1. Evidence that Physical Activity Can Influence Bone-Related Neuroimmune Mediators and Bone Health

Physical activity is well-known for its osteogenic potential and promotion of bone formation through mechanical loading [3, 64]. Exercise is the deliberate performance of physical activity with the intent to improve performance, health, and/or fitness. Because muscle and bone are biomechanically linked, most types of physical activities are considered beneficial to bone mass or quality, whether sports, planned exercise or household work [65, 66]. Physical activities typically increase bone mineral density and volume by stimulating osteoblast activity and shifting the balance of bone turnover to anabolism [67, 68]. High-impact and whole body aerobic exercises (e.g., jumping, resistance training) are particularly known to exert forces that modulate bone formation and upregulate the release of bone formation markers (e.g., osteocalcin) [3, 64]. It should be noted that there is a limit to the beneficial contribution of physical activity on bone. Several studies have shown that increasing weight bearing loads and muscle loading exercise to excessive levels is associated with diminishing returns in bone mass and quality, and can even lead to increased stress fractures, as well as injuries to tendon and muscle as previously reviewed [69, 70].

Beyond direct skeletal benefits, physical activity promotes favorable neuroimmune pathways, upregulates anti-inflammatory cytokines and enhances systemic resilience [68, 71]. Studies indicate that aerobic physical activity strongly promotes anti-inflammatory processes and suppresses pro-inflammatory processes (locally and systemically), especially the pro-inflammatory cytokines, IL-1 and TNF-α [72]. Positive adaptations have been observed in and around the exercised tissues, including the down regulation of specific pro-inflammatory cytokines in muscle [73], tendons [72], promotion of axonal regeneration in injured peripheral nerves [74], and reduction of adipose tissue, which is known to release a wide-range of adipokines including leptin [75]. These local adaptations likely contribute to the lower levels of systemic inflammation observed in physically fit and active individuals. Many investigators have shown increased circulating levels of IL-6, which is a multifunctional protein (dual role cytokine and myokine), acutely after exercise in humans and animal models [73], while prolonged exercise training is associated with reduced basal IL-6 production [76]. Therefore, the duration of the aerobic physical activity will affect systemic levels of cytokines. Regarding serum, urinary or salivary corticosterone levels, aerobic physical activity is associated with alterations in cortisol levels based on the duration and intensity of the exercise [77], as well as increases if the aerobic physical activity is forced or associated with pain and discomfort [72, 78]. This modulation of cytokines further emphasizes physical activity’s dual role in musculoskeletal health and neuroimmune regulation [71]. Clinical practice can leverage these findings by incorporating graded loading exercises into rehabilitation protocols to promote bone health.

2.2. Evidence that Force-Based Manipulations Can Influence Bone Health

Force-based manipulations are often associated with treatment for pain and mechanical dysfunction; however, there is also some evidence for force-based manipulations having an impact on the physiological function of bone. The applied mechanical pressure and joint movement during force-based manipulations should, in theory, stimulate mechanoreceptors and influence the autonomic nervous system innervation in tissues after treatment, particularly in bone since it is known to respond to mechanical loading [70]. Relevant literature for human subjects is reviewed in this section. Evidence from animal studies is reviewed at the end of Section 2.3.

Several studies examining the effects of force-based manipulations in human subjects have found elevation of serum markers of bone formation after treatment: serum type I collagen C-terminal propeptide (PICP), procollagen type 1 intact N-terminal pro-peptide (P1NP), and osteocalcin. Thai traditional massage in adults has been shown to increase levels of P1NP, particularly in older postmenopausal women with small body builds [79]. A few studies have also shown evidence of improved bone density or quality as a result of treatment [80, 81]. One study utilizing acupuncture as a treatment reported improved total bone mineral density in the femoral bone neck and hip joint (and enhanced serum osteocalcin levels) [80]. A study on preterm infants showed that massage therapy, with or without a physical activity component, led to elevated serum PICP [82], while another small study on preterm infants found that tactile/kinesthetic stimulation elevated urinary osteocalcin levels (Table 2) [81]. In that latter study, the tactile/kinesthetic stimulation also attenuated the decrease in tibial speed of sound (tSOS, m/sec, a surrogate for bone strength, assayed using quantitative ultrasound), observed in untreated preterm infants, matching findings in animal studies using this same treatment, as discussed further in Section 2.3 [83, 84]. A review summarizing evidence of a beneficial role of acupuncture on preserving bone quality postulated that the underlying mechanisms may be linked to regulation of the hypothalamic-pituitary-gonadal (adrenal) axis and activation of the RANKL/RANK/OPG and Wnt/beta-catenin signaling pathways [85]. Future investigations are needed to explore the potential underlying mechanisms, long-term clinical efficacy, and compliance of force-based manipulations in osteoporosis management, for example.

Table 2.

Effects of Force-Based Manipulation (FBM) on known bone neuroimmune modulators in humans^a

Modulator	Population	FBMs	Effect of FBM (↑ = increase; → = no effect; ↓ = decrease)
Factors that promote bone volume (anabolism)
Calcitonin gene-related peptide (CGRP)	Healthy subjects [87]	Acupuncture	→ CGRP (Plasma) [87]
Dopamine	Healthy subjects [86]	Acupuncture	→ Dopamine beta-hydroxylase activity (serum) [86]
	Subjects with depression, pain, immune conditions, stress [review] [98]	Massage Therapy	↑ Dopamine (urine ) [98]
Ghrelin	Obese subjects [31]	Acupuncture	↑ Ghrelin (blood) [31]
	Postpartum women [99]	Massage Therapy	→ Ghrelin (blood) [99]
Interleukin 10 (IL-10)	Subjects with multiple sclerosis [143]	Acupuncture	→ IL-10 (serum) [143]
Factors that suppress bone volume or increase bone loss (catabolism)
Glucocorticoids, e.g., cortisol or corticotropin-releasing	Healthy subjects [86]	Acupuncture	↑ Cortisol (serum) [86]
hormone (CRH)	Healthy subjects [87]	Acupressure massage	→ Cortisol (serum) [87]
	Healthy subjects [88–91, 144]	Massage Therapy	↓ Cortisol (blood) after a single session of treatment [88] ↓ Cortisol (blood) after 5 weeks of treatment [89] → Cortisol (salivary) after single session [91] ↓ Cortisol (salivary) after single session [90]
	Healthy subjects [145]	Osteopathic Manipulative Therapy	↓ Cortisol and CRH (blood) [145]
	Subjects with low back pain [96]	Acupuncture	↓ Cortisol (blood) [96]
	Preterm infants (stable) [101, 106]	Massage Therapy	→ Cortisol (salivary) [106] ↓ Cortisol/CRH (blood) [101]
	Premature infants [102]	Field massage and Gentle Human Touch	↓ Cortisol (urine) [102]
	Subjects with depression [100]	Massage Therapy	↓ Cortisol (urine) [100]
	Subjects with depression pain, autoimmune conditions, job stress, stress of aging, or pregnancy stress [review] [98]	Massage Therapy	↓ Cortisol (saliva or urine) [98]
	Subjects with fibromyalgia [107]	Massage Therapy	→ Cortisol (salivary) [107]
	Oncology subjects [103, 108]	Massage Therapy	→ Cortisol in subjects with breast cancer (blood) [108] ↓ Cortisol in isolated hematological subjects (serum) [103]
	Subjects with cardiac disorders [104, 105]	Massage Therapy	↓ Cortisol (blood) [104, 105]
	Subjects with spinal pain [review] [111]	Spinal Manipulation	Mixed results for cortisol (salivary/serum) in 5 studies; thus, low to very low evidence [111]
Interleukin 1 (IL-1)	Subjects with low back pain [92]	Manual Therapy	→ IL-1 (serum) [92]
	Subjects with spinal pain [review] [111]	Spinal Manipulation	→ IL-1 (serum) [111]
	Asymptomatic subjects [92]	Manual Therapy	→ IL-6 (blood) [92]
Interleukin 6 (IL-6, noting that this is a multi-functional cytokine)	Healthy subject [93]	Spinal Manipulation (low vs high force magnitude)	↑ IL-6 (blood) [93]
	Subjects with multiple sclerosis [143]	Acupuncture	→ IL-6 (serum) [143]
	Subjects with low back pain [92]	Manual therapy	→ IL-6 (plasma) [92]
	Subjects with low back pain [110]	Massage Therapy	↑ IL-6 (blood) [110]
	Subjects with low back pain [146]	Spinal Manipulation	↓ or normalization to control levels of IL-6 (blood) [146]
Receptor activator of nuclear factor kappa B ligand (RANKL)	Subjects with osteoporosis [reviews] [85, 97]	Acupuncture	↑ RANKL/RANK/OPG and Wnt/beta- catenin signaling pathway members [85, 97]
Substance P	Asymptomatic subjects [94, 95, 112]	Spinal Manipulation	↑ Substance P (plasma) [95] ↑ Substance P (plasma) [94] → Substance P (serum) after single SM treatment [112]
	Subjects with fibromyalgia patients [109]	Massage Therapy	↓ Substance P (plasma) [109]
Tumor Necrosis Factor alpha (TNF-a)	Asymptomatic subjects [92]	Manual Therapy	→ TNF-α (blood) [92]
	Subjects with multiple sclerosis [143]	Acupuncture	→ TNF-α (serum) [143]
	Subjects with low back pain [92]	Manual Therapy	↓ TNF-α (blood) [92]
	Subjects with low back pain [114]	Multimodal chiropractic care (that primarily includes spinal manipulation)	↓ TNF-α (urine) [114]
	Subjects with low back pain [113]	Osteopathic Manipulative Therapy	↓ TNF-α (blood) [113]
	Site Dependent effects/Other
Leptin - site specific	Obese subjects [31]	Acupuncture	↓ Leptin (plasma) [31]
	Postpartum women [99]	Massage Therapy	→ Leptin (blood) [99]
Osteocalcin - marker of bone formation	Subjects with osteopenia [80]	Acupuncture	↓ Osteocalcin (serum) [80]
	Postmenopausal women [79]	Massage Therapy	→ Osteocalcin (serum) [79]
	Preterm infants [81]	Tactile/kinesthetic stimulation	↑ Osteocalcin (urinary) [81]
Serotonin (5-HT) - site specific	Subjects with depression, pain, immune conditions, psychological stress [98]	Massage Therapy	↑ 5-HT (urine) [98]
	Subjects with breast cancer [108]	Massage Therapy	→ 5-HT (blood) [108]

^aResults from samples collected post treatment and then tested directly without mitogen or bacterial stimulations. Abbreviations: LPS = Lipopolysaccharide bacterial fragments; OPG = Osteoprotegerin; RANK = Receptor Activator of Nuclear Factor κB; Wnt = Wnt/beta-catenin pathway

2.3. Evidence that Force-Based Manipulations Can Influence Production of Bone-Related Neuroimmune Mediators

In contrast to the scant evidence in the literature for actual changes induced in bone by any force-based manipulation, discussed above, several studies examining the responses to acupuncture, massage therapy, spinal mobilization, and several other force-based manipulations have demonstrated changes in blood, saliva and urine levels of several neuroimmune mediators linked to bone metabolism (focusing on mediators discussed in Section 1 and Table 1). Main findings of these studies are reviewed below and in Table 2 first for humans, followed by evidence in animal studies.

Evidence from studies on human subjects

In healthy or asymptomatic human subjects, studies examining the responses of neuroimmune mediators to acupuncture treatment have observed post-treatment increases in serum levels of cortisol (a glucocorticoid that suppresses bone catabolism when at persistent high levels) [86], yet no increases in plasma levels of CGRP [87] or dopamine beta-hydroxylase activity [86]. Acupressure massage showed no effect on serum levels of cortisol in one preliminary study on healthy subjects [87]. In contrast, several studies examining the effects of massage therapy in healthy human subjects have shown decreases of blood or salivary cortisol [88–90]. However, another study examining the effects of massage therapy in healthy human subjects found no changes in salivary levels of cortisol [91]. Furthermore, blood samples cultured after massage therapy have shown lowered inflammatory cytokine production, IL-1 and IL-10 after mitogen-stimulation, compared to blood samples from untreated subjects [88, 89], studies not included in Table 2 due to methodological differences from other studies in that Table. Manual therapy showed no changes in venous blood levels of IL-6 and TNF-α in human subjects that were asymptomatic for low back pain (these subjects were the control preload-only group) [92]. A study examining the effects of single thoracic spinal manipulation at low versus high force (compared to a control preload only group) found that only the higher force spinal manipulation led to increased plasma levels of IL-6 [93]. Other studies have found increases in plasma levels of substance P in asymptomatic patients following thoracic spinal manipulation [94] or cervical spinal manipulation [95]. In summary, in healthy or asymptomatic subjects, there is some evidence that acupuncture, massage therapy and spinal manipulation treatments can influence bone-related neuroimmune modulators in beneficial directions.

Also shown in Table 2, several studies have examined the influence of force-based manipulations on bone-related neuroimmune modulators in human populations with a variety of clinical conditions. Acupuncture has been shown to: 1) increase blood levels of ghrelin and decrease leptin in obese subjects [31]; 2) lower cortisol levels in subjects with low back pain [96]; 3) enhance serum osteocalcin in subjects with osteopenia in conjunction with improved total bone mineral density in the femoral bone neck and hip joint [80]; and 4) upregulate expression and activation of RANKL/RANK/OPG and Wnt/beta-catenin signaling pathway members in subjects with osteoporosis, as discussed in Section 2.2 [85, 97]. Findings of these studies (albeit more are needed) suggest that acupuncture alters levels of bone-related neuromodulators and thus perhaps alters signaling in bone turnover pathways, as suggested by the improved bone mineral density in the femur and hip joint in the one human subject study that examined bone [80].

A number of studies have examined the effects of massage therapy on levels of bone-related neuroimmune mediators in blood, serum, urine or saliva in subjects with depression, fibromyalgia, immune conditions, and pain, and in postmenopausal and postpartum women, and preterm infants (listed by analyte in Table 2). Massage therapy has been shown to enhance markers of bone anabolism in subjects with depression/pain/immune conditions (urinary dopamine and serotonin (5-HT) increased, reviewed by [98]); and tactile/kinesthetic stimulation of preterm infants elevated urinary osteocalcin levels, compared to untreated preterm infants [81]. However, postpartum women receiving massage therapy showed no changes in blood ghrelin or leptin (metabolic hormones that modulate bone formation) [99], and postmenopausal women receiving massage therapy showed no changes in osteocalcin levels [79]. Massage therapy induces a beneficial effect on cortisol (lowering levels) in 7 out of 10 studies found in the literature. After massage therapy, cortisol has been shown to decrease in urine or saliva in patients with depression, autoimmune issues, or psychological stress [98, 100], in blood and urine of preterm infants [101, 102], in blood of isolated hematological subjects [103], and in blood of subjects with cardiac disorders [104, 105]. However, another study examining salivary levels in preterm infants showed no effects of treatment [106]. Nor were changes in cortisol levels observed post-massage in subjects with fibromyalgia [107] or breast cancer [108]. Studies examining the effects of massage therapy have also shown a potentially benefical decrease in plasma substance P (a neuroinflammatory mediator) in subjects with fibromyalgia [109], and elevated blood IL-6 (a multifunctional cytokine) in subjects with low back pain [110]. A single study using manual therapy found no effects on plasma levels of inflammatory cytokines in subjects with low back pain [92]. Thus, overall, massage therapy treatment shows beneficial effects on bone-related neuroimmune modulators in human subjects, although whether there are any actual effects on bone needs to be examined.

Findings are mixed as to whether spinal manipulation alters levels of salivary or serum cortisol levels in subjects with spinal pain, with low to very low evidence in those that did show a change (reviewed by [111]). Additionally, no change in serum substance P levels has been detected in subjects with low back after spinal manipulation [112]. Yet, osteopathic manipulation and multimodal chiropractic care (that primarily includes spinal manipulation) has been shown to lower blood or urine levels of TNF-α in subjects with low back pain [113, 114]. Several studies examining the effects of spinal manipulation on inflammatory cytokines used whole blood in-vitro evoked methods in which blood samples collected post-intervention were cultured for varying periods of time before bacteria were added to the cultures (e.g., lipopolysaccharide fragments) to evoke an infection-like immune response. Outcome results of these studies are mixed, with some showing no changes in bacteria-evoked IL-1 beta or TNF-α production after spinal manipulation of subjects with spinal pain [111, 115], while others show reduced production [116]. A recent review [117] concluded that spinal manipulation may influence inflammatory and immune markers, although much of the evidence was of low quality, while another review concluded that there is no clear effect [118].

In summary, there is clear variability in the response of neuroimmune mediators to the various force-based manipulations in human subjects (e.g., for cortisol as shown in Table 2), perhaps due to the small number of studies per analyte, differences in the force-based manipulation utilized, techniques performed, lack of standardized treatment delivery parameter, and variability in the healthy subject response versus responses in patients with a variety of clinical conditions. Importantly, while these studies show an effect of some types of force-based manipulations on the production of select neuromodulators, additional research in human subjects is needed to determine if these treatments have any beneficial effect on bone quality since the latter was not usually examined.

Evidence from animal studies

Stronger support for the effects of force-based manipulation on neuroimmune modulators and bone tissues have been reported in several animal studies. Specifically, the literature from animal studies shows strong efficacy for force-based manipulations in preserving or improving bone quality, primarily because these studies included more outcomes, such as additional serum assays, imaging, biomechanical tests, and/or histomorphometry.

The effects of acupuncture on bone biomechanical properties and trabecular bone microarchitecture have been studied in ovariectomized rats [119, 120]. This treatment promoted bone formation and suppressed bone resorption. Specifically, acupuncture offered significant protection against ovariectomy-caused declines in bone strength, elevated trabecular bone volume and thickness, lowered trabecular separation, and promoted serum osteocalcin levels, when compared to sham or model control rats. Acupuncture has been shown to promote bone formation and improve bone architecture in a senescence-accelerated mouse strain (SAMP6) by enhancing testosterone secretion and reducing bone turnover [121]. Another study examining the effects of acupuncture treatment on rats with irritable bowel syndrome observed reduced serum levels of pro-inflammatory cytokines and neuropeptide expression (CGRP and substance P) in their hypothalamus, colon and dorsal root ganglia [122, 123], although effects on bone were not examined. In addition, in a rat model of overuse injury in which weeks of repetitive lever bar pulling at high force loads by untreated rats has been shown to induce bone loss [124–126], twelve weeks of preventive manual therapy lowered circulating and tissue levels of pro-inflammatory cytokines in parallel with elevations in serum osteocalcin and the rescue of trabecular bone volume and density [127]. Bone-specific insulin like growth factor-1 (IGF-1) mRNA are lowered in rat models of neonatal psychological stress by mechanical tactile stimulation with or without kinesthetic stimulation, in parallel with elevations in serum osteocalcin and indices of bone growth [84], or maintenance of bone mineral density (rather than a loss) [83], compared to untreated psychologically stressed neonatal rats. Application of instrument-assisted spinal manipulation to rats leads to increased levels of IL-1 beta and TNF-α in dorsal root ganglia and spinal cord, and increased levels of IL-10 in the spinal cord [128]. Lastly, spinal mobilization in a rat model of low back pain reduces CGRP expression in dorsal root ganglia [129]. Thus, the controlled examination of force-based manipulations in animal models show decreased levels of several bone neuroimmune modulators in serum and/or tissues [122, 123] [81, 84, 127, 130], in parallel with improved bone quality and reduced cartilage pathology [81, 84, 127, 130].

In joint tissues, acupuncture has been shown to downregulate TNF-α and IL-1 beta levels, inhibit inflammatory macrophages, and increase production of anti-inflammatory cytokines, IL-10 and transforming growth factor beta, in rat models of inflammation-induced arthritis (adjuvant-induced) [130–132]. Acupuncture treatment of rats with experimentally induced arthritis showed markedly lower protein levels of IL-1 and higher protein levels of IL-10 in inflamed joint tissues, in parallel with reduced cartilage pathology [130]. In a rabbit model of knee osteoarthritis, acupuncture also reduced inflammatory mediator levels in serum and synovial fluid, in parallel with a reduction in cartilage matrix decomposition [133]. A study examining the effectiveness of joint mobilization in a rat model of Freund’s Adjuvant (CFA)-induced joint inflammation showed no reductions in infiltration of M1-type inflammatory macrophages into joints, yet increased M2 repair type macrophages [134]. While not examined in that study, M2 repair type macrophages are known to produce transforming growth factor beta [135].

3. Implications for Clinical Practice

Physical activity is well-known for its osteogenic potential and promotion of bone formation through mechanical loading [3, 64]. Also, several studies examining the responses to acupuncture, manual or massage therapy, and spinal mobilization in human and animal subjects have also demonstrated beneficial changes in blood, saliva and urine levels of several neuroimmune mediators linked to bone metabolism. The few studies examining the effects of force-based manipulations on bones and joints observed enhanced bone quality or reduced joint inflammation. While considerably more studies are needed, clinicians should consider tailoring physical activity, manual therapy and exercise interventions based on patient-specific factors, like age, hormonal status, and comorbidities, which affect neuroimmune responses and bone health [68, 70]. For example, postmenopausal patients with osteoporosis may benefit from carefully monitored, progressive resistance exercises that can strengthen bone while increasing circulation and stimulating mechanoreceptors in the soft tissue, while individuals recovering from joint restriction, soft tissue injury, and/or inflammation may be unable to tolerate or perform loaded weight-bearing exercises. These patients may need a force-based manipulation that focuses more on local tissues and/or neuroimmune regulation, thereby offering a viable adjunct in support of tissue adaptation, since they can also provide mechanical stimulation and beneficial alterations in one or more bone-related neuroimmune modulators. If replicated, such findings may have implications for managing inflammatory and autoimmune conditions [88]. That said, while force-based manipulations can serve as an intermediate step to improve musculoskeletal functional status, the goal should be to progress toward weight-bearing activities when feasible, given their well-documented benefits for musculoskeletal and systemic health. In addition to neuroimmune responses, responses occurring in mechanoreceptors, bone, soft tissues, and circulatory and lymphatic stimulation are physiologically intertwined - combinations of certain force-based manipulations with therapeutic exercise approaches may best improve bone and overall health of the patient.

A note of caution, the results of many of these studies are limited by small sample sizes, non-standardized duration or frequency of treatments, and often, a lack of sham control groups (which we could not venture into discussing due to manuscript length constraints). Additional gaps in knowledge are listed in the next sections. While the benefits of physical activity on bone is well supported (as long as the beneficial threshold of intensity and duration are not surpassed) [70], many additional studies are needed in order to understand the full effects of force-based manipulations before claims of efficacy or effectiveness can be made. Clinicians should take care to avoid misinterpretation of these findings to suggest that neuroimmune response may contribute to viral immunity or more widespread pseudoscientific claims due to the low level of evidence in this area [118, 136, 137].

4. Implications for Clinical Design and Research Studies

1. Confirmatory studies in larger samples are needed [13, 89].

2. Sham-controlled designs need to be considered for best quality.

3. The influence of the setting, such as auditory input, lighting, environmental pollution, time of day and temperature confounds need to be considered as such factors introduce variability from day to day [71].

4. The expectations of and interactions between participant and practitioner seem to play a role in response to treatment that needs to be clearly defined, verified, and validated [71, 91].

5. There is a current disconnect between the examination of bone and joint health and biomarker production. In some studies, select neuroimmune modulators are assessed but not bone or joint integrity, and vice versa in other studies. This limits full interpretation.

6. Determine the relationship between neuroimmune changes and any short- or long-term clinical outcomes. Specifically, the duration of altered neuroimmune/bone changes needs to be determined.

7. Are induced bone/neuroimmune responses strictly localized/restricted to the anatomical site of delivery? Or, can systemic changes in neuroimmune modulators have beneficial effects?

8. Determine if any specific type/technique or delivery/dosage characteristics of force-based manipulation is more efficacious at eliciting neuroimmune response on bone in comparison to others.

9. Is there a minimum threshold of applied force-based manipulation force/pressure required to elicit bone or neuroimmune response.

5. Additional Gaps in the Literature

1. It is unclear how aging modifies sympathetic and sensory innervation of bone marrow compartments. Longitudinal studies should be performed in young adults, mature and aged humans to capture the effects of changing repair mechanisms, immune responses, hormone levels, changes in metabolism and inflamm-aging (the increase in inflammation that occurs with aging) [138, 139].

2. Future studies should examine whether sensory or sympathetic neuropathy is necessary for reduced bone formation seen with type 1 diabetes mellitus, or if other complications are more direct contributors (e.g., microvascular dysfunction and hyperglycemia).

3. Future studies should consider the potentials of sexual dimorphism (structural, neuroimmune, or hormonal response differences).

Table 3.

A Call to Action

• Clinical practice should leverage these findings by incorporating graded loading exercises with select force-based manipulation treatments into rehabilitation protocols to promote bone health. • Additional research in human subjects is needed to determine if force-based manipulations have any beneficial effect on not only on the production of select neuromodulators, but also on bone quality. • Future investigations are needed to explore the potential underlying mechanisms, long-term clinical efficacy, and compliance of force-based manipulations in management of bone and joint health. • Improved clinical/research designs are needed to understand the full effects of both physical activities and force-based manipulations on underlying bone and joint tissues.

Highlights.

• Responses of bones are modulated by many factors, e.g., neuroimmune mediators

• The neuroimmune system regulates bone function through a complex network of mediators

• Exercise, touch, and force-based treatments increase neuroimmune mediators in humans

• Force-based manipulations may be beneficial for stimulating bone changes