Lumbar astrocytes undergo changes in morphology and increased GFAP expression, indicating a state of increased activation following chronic constriction injury (CCI) of the rat sciatic nerve (29). Since this initial study demonstrating anatomical evidence that spinal glia play a role in pain modulation, later studies documented that spinal glia are closely associated with neural synapses, express many of the same neurotransmitter receptors as neurons, and are known to assume “activated” phenotypes in response to abnormal or heightened neuronal signaling (8, 9, 15, 23, 30–32). Astrocytes and microglia express pain-relevant neurotransmitter receptors, like AMPA (α-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid) receptors, metabotropic glutamate receptors (mGluR), purinergic receptors (e.g., P2X₄R or P2X₇R), or neurokinin 1 receptors (NK1R), which are activated by neurotransmitters released from presynaptic nerve terminals. These neurotransmitters include glutamate, ATP and substance P (9). Following peripheral nerve damage, neurons also release chemokines like CCL2 (33) and CX3CL1(34), and other immune mediators like colony-stimulating factor 1 (CSF-1) (35) and ATP (36, 37). These released factors potently activate astrocytes and microglia (9), which in turn release pro-inflammatory cytokines (e.g., IL-1β, TNF-α, and IL-6) (38), chemokines (e.g., CCL2) (33), ATP, excitatory amino acids (EAAs), and NO (7). As in the periphery, dorsal horn neurons express receptors for many of these factors, including the pro-inflammatory cytokines, IL-1β, TNF-α, IL-6, and IL-17 (12, 15). Activation of neuronally-expressed cytokine receptors modifies neuronal function. For example, the pro-inflammatory cytokines TNF-α and IL-1β enhance excitatory synaptic transmission and suppress inhibitory synaptic transmission in neurons of spinal cord lamina II (31, 39). Whereas, TNF-α enhances spontaneous excitatory post-synaptic current (sEPSC) frequency, IL-6 reduces frequency of the spontaneous inhibitory post-synaptic current (sIPSC). Notably, IL-1β is able to enhance both sEPSC frequency and amplitude while reducing sIPSC frequency and amplitude. Additionally, TNF-α and IL-1β enhance excitatory AMPA- and NMDA-induced currents, and IL-1β and IL-6 suppress inhibitory GABA- and glycine-induced currents (39). Therefore, activation of these cytokine receptors further sensitizes local neurons leading to a feed-forward cycle of nociceptive signaling.

While it is clear that peripheral immune cells migrate to local sites of nerve damage, a growing body of evidence supports that they also traffic to more remote pain-relevant neural tissues and play a role in the development of pathological pain. For instance, macrophages are recruited to sites remote from the initial insult, such as to the spinal cord (40, 41) and DRG (41–43), following peripheral nerve trauma in response to chemotactic factors like CCL2 (42–44) and CX3CL1 (33, 45) derived from injured neurons with their cell bodies or central projections in these distant locations. Macrophages present in the DRG are most likely important for maintenance of chronic pain. This is supported by evidence that depletion of peripheral macrophages partially reverses established paclitaxel-induced or nerve ligation-induced mechanical hyperalgesia and reduces TNF-α expression in DRG (43, 46).

In addition to macrophages, T cells may also be important in the mechanisms that underlie pathological pain. For instance, activated CD4⁺ T cells, specifically Th1 and Th17 cells, migrate to associated lumbar DRG and/or the lumbar spinal cord dorsal horn in animal models of sciatic nerve transection (47–49), spared nerve injury (50, 51), partial sciatic nerve ligation (52, 53), and sciatic nerve chronic constriction injury (54).

Increased T cell trafficking in these animal models of neuropathic pain has been associated with concurrent increases in pro-inflammatory T cell-related cytokine signaling. For example, Costigan et al. identified T cell infiltration of the spinal cord dorsal horn following a spared nerve injury (SNI) in rats with concurrent increases in spinal interferon-γ (IFN-γ) expression (50). IFN-γ is the signature pro-inflammatory cytokine released by activated Th1 cells (55). In a related study, neuropathic pain resulting from spared nerve injury is significantly diminished in both Rag1 knockout (Rag1^−/−; T-cell deficient) mice and IFN-γ receptor null mutant mice as compared to wildtype controls (50).

T helper cell-mediated IL-17 expression also appears to be important for pain pathology. Compared to wild-type mice, IL-17 knockout mice display significantly less mechanical hypersensitivity as well as decreased T cell accumulation at both the site of partial sciatic nerve ligation and within the relevant lumbar DRG (L3-L5) (52). Similarly, increased CD4⁺ T cells were observed in the rat spinal cord in a model of spinal nerve ligation (e.g., peripheral nerve damage), with double-positive CD4⁺/IL-17⁺ T cells located in superficial laminae of the spinal cord dorsal horn by immunofluorescence staining (51). This was accompanied by significant upregulation of IL-17, IL-1β, and IL-6 (51).

Much of what is known about the mechanisms of pain was identified in male animal models. However, recent reports indicate that mechanisms of pain may differ between the sexes (56, 57). While microglial activation is required for preclinical models of pathological pain in males, this does not seem to be the case for females who instead appear to be more vulnerable to the actions of T cells (54, 58–61).

Recent work from Noor et al. demonstrates that males and females exhibit divergent T cell-associated actions within discrete pain-relevant anatomic regions in a rodent model of sciatic neuropathy (54). During neuropathy, females display more profound Th17 specific responses with higher levels of IL-17A than males, both at the injured nerve and in the corresponding lumbar spinal cord. When neuropathic animals were treated with a β2-integrin antagonist, BIRT377, pain-relieved females expressed dramatically reduced IL-17A levels in sciatic and lumbar spinal cord tissues as compared to their male counterparts (for further discussion, see section “Lymphocyte Function-Associated antigen-1” below). This not only supports a role for IL-17A in mechanisms of neuropathic pain, it offers evidence of a T cell differentiation bias toward a pro-inflammatory status that is significantly greater in females than males (54).

There is growing evidence that other cytokines may also be differentially regulated in males vs. females following peripheral nerve injury. For instance, recent RNA-seq analysis of lumbar DRG following rat sciatic CCI identified that the cytokine CSF-1 is 1.7 times higher in females than males (62).

Interleukin-4 (IL-4) is an important regulator of immunity, with diverse roles in processes, such as T-cell proliferation, activated B-cell stimulation, activation of macrophages, chronic inflammation, and wound repair (88). IL-4 is primarily produced by macrophages, T-cells (especially Th2 cells), mast cells, eosinophils, and basophils (89). This cytokine exerts a variety of effects through actions at two receptor subtypes. The IL-4R-I receptor is expressed by T cells, and B cells, monocytes/macrophages, and fibroblasts (90), and is involved in the regulation of maturation and proliferation of Th2 cells, and in IgE synthesis by B cells (88).

IL-4 signaling has been shown to decrease the production and actions of several pain-relevant pro-inflammatory mediators. For instance, IL-4 suppresses NLR protein 3 (NLRP3) inflammasome-dependent caspase-1 activation and the subsequent IL-1β secretion by macrophages (91). NLRP3 serves as a sensor protein capable of initiating assembly of the intracellular inflammasome protein complex, resulting in caspase-1 activation and consequent conversion of pro-IL-1β to mature bioactive IL-1β (92, 93). IL-4 has also been shown to increase expression of endogenous IL-1 receptor antagonist (IL-1ra) mRNA and protein (94), thus diminishing the actions of pro-inflammatory IL-1β. Additionally, IL-4 directly inhibits the induction of nitric oxide synthase (NOS) and levels of cyclooxygenase 2 (COX-2), thereby decreasing the production of NO and PGE2, respectively (95).

In contrast to IL-10, which is predominantly characterized as an anti-inflammatory cytokine, IL-4 exerts some pro-inflammatory actions. In vascular endothelial cells, IL-4 induces expression of CCL2 and IL-6 (96). It also synergistically increases IL-1β, TNF-α, and vascular cell adhesion protein 1 (VCAM-1; a.k.a. CD106) in lipopolysaccharide-stimulated vascular endothelium (96). IL-4-induced IL-6 expression is mediated via a reactive oxygen species (ROS)-dependent mechanism in human aortic endothelial cells (96).

Despite some evidence that it can serves as a facilitator of pro-inflammatory cytokine expression (97–99), IL-4 is an attractive candidate for treatment of pathological pain due to its broad spectrum of anti-inflammatory actions. Additionally, IL-4 deficiency in naïve IL-4 knockout mice exhibit increased tactile allodynia (100). However, IL-4 protein replacement therapy is precluded from use for systemic administration due to IL-4's short protein half-life as well as its pleiotropic effects on immune and vascular function (74). Therefore, gene therapeutic approaches offer the best avenue for application.

IL-4 gene therapy has been examined for control of neuropathic pain in an animal model of spinal nerve ligation. Work by Hao et al. demonstrated that herpes simplex viral vector-mediated expression of IL-4 administered after the onset of mechanical allodynia results in 5 weeks of pain relief with concurrent increases in IL-4 protein expression in DRG (74). Similar effects were observed following a repeat injection. Interestingly, prophylactic treatment at 1 week prior to sciatic nerve ligation appeared to delay the onset of mechanical allodynia by about 5 weeks (74).

IL-4 gene therapy has also been examined in an animal model of painful inflammatory arthritis (101). Intra-articular injection of IL-4 gene therapy using a retroviral vector led to a reduction in paw swelling and decreased radiographic evidence of bone destruction, though differences only achieved significance about 2 weeks post-injection and paw edema measurements were not assessed throughout the entire timecourse of the inflammatory model (101), which can last several months (102), so duration of effect is speculative. Potential sex differences in IL-4 efficacy as a therapeutic to control chronic pathological pain has not been reported.

Gene therapeutic approaches may bypass some of the obstacles faced by application of recombinant TNFSR. When applied subcutaneously to the hindpaw 1 week following sciatic nerve ligation, herpes simplex virus (HSV)-mediated TNFSR is associated with transient improvements in mechanical allodynia and reversal of thermal latency (76). HSV-mediated TNFSR has also been shown to improve pain behaviors in models of diabetic neuropathy, spinal cord injury, HIV gp120-mediated inflammation of the sciatic nerve, and neuropathy secondary to antiretroviral therapy (75, 77, 78).

Interleukin-10 (IL-10) is a powerful anti-inflammatory cytokine that binds to the IL-10 receptor (IL-10R), with IL-10R1 (a.k.a. IL-10Rα) being the high affinity subunit necessary for signal transduction (114). Importantly, the IL-10R1 subunit is expressed by astrocytes, microglia, oligodendrocytes, endothelial cells, and trafficking leukocytes (115). Despite IL-10 R1 having been identified by RNA-sequencing analysis of lumbar spinal and DRG neurons in several reports (16), these findings have not yet been convincingly validated by confirmed neuronal protein expression. Two separate reports provide evidence of IL-10R1 protein expression in embryonic (E16-18) rat neurons in vitro (116, 117). Post-natal rodent protein analyses have been somewhat mixed, with one report identifying IL-10R1 protein expression in DRG neurons (118), and another report failing to identify IL-10R1 in brains (115). Due to ongoing uncertainty of neuronal IL-10 receptor expression, it remains possible that the pain-relieving actions of IL-10 and other anti-inflammatory cytokines may provide therapeutic benefit through actions at neuronal receptors.

IL-10 pleiotropically inhibits the actions of many pro-inflammatory factors by numerous mechanisms briefly described here, but thoroughly discussed elsewhere [for review, see (119, 120)]. IL-10 signaling leads to decreased bioavailability of pro-inflammatory cytokine proteins through upregulation of both IL-1ra and TNF-α decoy receptor (119). This is also accomplished through reduced gene transcription, decreased mRNA stability, and reduced translation for pro-inflammatory cytokines and chemokines, many of which are downstream of toll-like receptor 4 (TLR4) (121, 122). Additionally, IL-10 destabilizes mRNA transcripts for the pro-inflammatory cytokines TNF-α and IL-1β (119, 123). IL-10 signaling also induces suppressor of cytokine signaling 1 (SOCS1) and SOCS3 production, thereby further suppressing pro-inflammatory cytokine production by targeting the p65 NF-κB subunit (a key transcriptional regulator in the inflammatory process) for degradation (124).

The anti-inflammatory effects of IL-10 come about by blocking downstream signaling of TLR4, the actions of which normally promote a pro-inflammatory environment critical for the development of neuropathic pain (125). For instance, IL-10 inhibits translation of MyD88, an adaptor protein used by TLR4 that leads to pro-inflammatory cytokine production and release (126). IL-10 also increases ubiquitination and subsequent proteasomal degradation of MyD88-dependent signaling molecules (127). Interestingly, IL-10 induces microRNA-146b, which negatively regulates TLR4-mediated pro-inflammatory cytokine and chemokine induction, with targets including IL-6, TNF-α, IL-8, CCL3, CCL2, CCL7, and CXCL10 (128). IL-10 signaling also increases transcription of numerous effector proteins that inhibit TLR-induced downstream signaling molecules. For instance, IL-10 increases expression of Abin-3 and DUSP-1, leading to inhibition of p38 MAP kinase (MAPK) phosphorylation as well as increased nuclear translocation and DNA binding of p50/p50 homodimers of NF-κB, a form of NF-κB insufficient for induction of pro-inflammatory cytokine transcription (129).

Several studies indicate that under some chronic pain states, patients suffer from suppressed IL-10 expression (114). Similarly, animal studies also support a link between IL-10 dysregulation and pathological pain. Following peripheral nerve injury, IL-10 is briefly upregulated in what is thought to be a compensatory effort for homeostatic conditions to return. However, as pathological pain develops, IL-10 expression falls below baseline levels in pain-relevant anatomic locations (130–134). Direct application of IL-10 protein to the site of pathology is capable of producing immediate symptom relief, effects are often transient and may require repeated injections (79). These findings support that returning IL-10 expression to near-basal levels may be key to treatment for pathological pain.

Like IL-4, IL-10 has a short half-life, thus making gene therapy a desirable route of administration, particularly in patients requiring local IL-10 expression, such as in individuals with low back radiculopathy. Initial studies utilized viral vectors for lumbar spinal (intrathecal) delivery of the IL-10 gene, including non-replicating adenovirus, adeno-associated virus, and herpes simplex viral vectors. While potent pain relief was observed in animal models of chronic sciatic pain, effects were short-lived (79–82, 135, 136). The transience of the pain relief in these studies may be a result of possible complications associated with the use of a viral vector itself, such as an immune response following exposure to viral antigens (68, 120). To circumvent these possible viral obstacles, non-viral IL-10 gene therapy was explored.

Paradoxically, though non-viral gene therapies are considered the least efficient methods of gene transfer, spinal non-viral IL-10 gene therapy has been repeatedly demonstrated to provide profound and long-lasting pain relief in a variety of animal models (28, 83–87, 114, 137–139). Spinal non-viral IL-10 gene delivery in neuropathic animals produces pain relief through elevated peri-spinal IL-10 production with corresponding reduction of pro-inflammatory mediators (84, 87, 138, 140). The intriguing possibility that the protein derived from the IL-10 transgene could be capable of dimerizing with available endogenous IL-10 protein followed by receptor binding was considered a potential mechanism for the observed enduring efficacy of this unique non-viral approach. Prior work identified a non-silent point mutation in the IL-10 gene being applied in the previously published reports, and there had been some concern that it might disrupt the ability of plasmid-derived IL-10 to engage endogenous IL-10 (28). However, a recent report demonstrates that non-viral IL-10 gene therapy applied to neuropathic IL-10 knockout mice following sciatic chronic constriction injury provides long-lasting relief of allodynia, indicating that endogenous IL-10 is not required for these effects (28). Expression of the IL-10 transgene expression in pain-relieved IL-10 knockout mice was identified in lumbar DRG with concurrent increases in anti-inflammatory transforming growth factor (TGF)-β1 and decreases in pro-inflammatory TNF-α mRNA. While no IL-10 transgene was identified in the lumbar spinal cord, anti-inflammatory changes in the DRG appear to drive concordant decreases in lumbar TNF-α, CCL2, and markers of both astrocytic (GFAP) and microglial (TMEM119) activation (28). This provides compelling evidence that modification of cytokines within the DRG may be sufficient for relief of spinal mechanisms of neuropathic pain following peripheral nerve damage. To date, sex differences in IL-10 efficacy for treating chronic pathological pain have not been reported.

Non-viral IL-10 gene therapy holds significant promise for clinical translation. At this time the non-viral IL-10 gene therapy XT-150 is currently in phase I/II clinical trials (ClinicalTrials.gov Identifiers: NCT03282149; NCT03769662; NCT03477487).

D-mannose is an inexpensive and commonly available dietary supplement (142) that has been assessed via phase III clinical trial for prevention of recurrent urinary tract infections (ClinicalTrials.gov Identifier: NCT01808755) (143). D-mannose is a known mannose receptor agonist. The mannose receptor (MR; CD206) is a c-type lectin receptor that functions in endocytosis and pathogen recognition (144). MR is expressed by macrophages, dendritic cells, and microvascular endothelial cells (145), as well as astrocytes, microglia, Schwann cells, and some neurons (146–148). MR is commonly used as a macrophage marker, though increased expression of MR is often associated with anti-inflammatory macrophages (149).

D-mannose was initially explored for use as an adjuvant due to the hypothesis that it may increase intracellular uptake of naked plasmid (87). Interestingly, D-mannose greatly increases gene therapeutic efficacy allowing for long-lasting pain relief following a single co-injection of D-mannose with plasmid DNA encoding the gene for IL-10 in neuropathic rats (87). Work by Dengler et al. also suggests that D-mannose exposure itself increases anti-inflammatory signaling, possibly through actions of endogenous IL-10. However, work by Vanderwall et al. identifies that D-mannose-mediated anti-inflammatory effects do not require IL-10, as intrathecal D-mannose alone induces transient pain relief in IL-10 knockout mice (28).

Historically MR was not considered to possess signaling ability due to the receptor's absence of a clear intracellular signaling motif. Further, if signaling was identified, it was reported as pro-inflammatory in nature. However, many of these studies were performed under inflammatory states which likely predispose MR toward pro-inflammatory signaling pathways (150–152). New evidence supports that MR activation is also capable of increasing anti-inflammatory cytokines under the right conditions (144, 153). For instance, treatment of dendritic cells with “activating” anti-MR monoclonal antibodies potently increases endogenous IL-10 and IL-1ra while decreasing IL-1β and TNF-α production (154). Furthermore, D-mannose treatment prior to LPS stimulation of RAW 264.7 mouse macrophages reduces release of pro-inflammatory TNF-α, IL-1β, and NO, but increases IL-10 secretion (87).

While D-mannose was initially explored as an adjuvant for gene therapy, its modulation of pain-relevant cytokine signaling pathways make it a novel pain therapeutic in its own right. D-mannose and MR-mediated anti-inflammatory signaling may be clinically relevant for the treatment of a variety of chronic pain conditions due to its engagement of both IL-10-independent and IL-10-mediated anti-inflammation (28, 87). These characteristics may make D-mannose useful for pain relief in patients with known IL-10 gene polymorphisms that would prevent normal IL-10 production in response to other therapeutic approaches. It may also be beneficial for treatment of pain pathologies associated with decreased IL-10 expression, like fibromyalgia (114, 155).