Chronic pain is a leading world health problem. It retards personal and societal productivity, and engenders despair. One person in five suffers the multidimensional consequences of chronic pain. Neuropathic pain is a particularly debilitating subtype of chronic pain that derives from damage to the somatosensory nervous system. Tens of millions globally are affected by neuropathic pain. Many experience persistent, unprovoked burning and shock-like pain for which we currently have minimally effective treatments. There are numerous comorbidities associated with painful conditions, including stress, depression and anxiety. Stress increases the severity of persistent pain and spurs episodic pain. For conditions that can lead to neuropathic pain, (e.g., SCI, stroke, HIV, cancer), stress may hasten the onset and/or increase the severity of pain.
Based on anecdotal and published reports, stress is a significant contributor to pain; despite awareness of this relationship, mechanism(s) has largely been overlooked. Herein, we address this gap to better understand the role of stress and stress hormones in persistent pain.
To test the hypothesis that stress exacerbates the development of neuropathic pain, acute restraint stress was applied immediately prior to peripheral nerve injury and subsequent pain-like behavior was measured. As reported in humans, stress increased allodynia in mice. Stressors elicit numerous physiological changes that could account for enhanced pain. Among them, is activation of the hypothalamic-pituitary-adrenal (HPA) axis that stimulates glucocorticoid (GC) release. Cortisol is the primary stress hormone of the GC family, (NB: corticosterone in rodents). The actions of GCs are initiated by binding the glucocorticoid receptor (GR). The density of GR is variable throughout the nervous system and its relative density in somatosensory circuitry was unknown. Remarkably, we observed that GR abundance in dorsal root ganglia (DRG) exceeded the hippocampus, where it is generally considered most abundant. Even in a more heterogeneous tissue as the dorsal spinal cord (including white matter), GR content was comparable to hippocampus. Thus, the potential for GCs to affect pain processing circuitry is considerable.
To determine if stress exerts its effect on pain via GC-GR activity, mice were treated with a GR antagonist prior to stress. Indeed, GR blockade attenuated stress effects on pain-like behavior after nerve injury. Importantly, besides perceived stressors, there are multiple stimuli that induce GC release, including bodily injury, illness, and exercise. Hence, in broad contexts, exposure to GCs may impact pain. To determine if exposure to elevated GCs is sufficient to produce pain enhancement, mice were treated systemically with corticosterone (CORT) prior to nerve injury (in lieu of stress). Like the stressor, this procedure also increased allodynia in mice. Together, these data indicate that stress-induced potentiation of allodynia is at least partially mediated by GR.
Two broad mechanisms contribute to neuropathic pain: neural plasticity and neuroinflammation. Stress and GCs are known to affect measures of both neural plasticity and neuroinflammation, therefore their impact on pain could be through either mechanism or, as our studies suggest, on both. To investigate structural neuron plasticity associated with pain, we measured the axon growth capacity of DRG neurons derived from acutely stressed mice. Compared to no stress, stress increased measures of axon sprouting and elongation. This in vitro sprouting phenotype is generally compared to the type of axon sprouting observed in painful neuromas; whereas, elongating axons are likened to those that support functional regeneration. In effect, it appears that stress stimulated a growth ‘program’ in DRG neurons that simultaneously promotes both unfavorable and favorable axon phenotypes. In subsequent experiments, these effects were observed to be GR, NMDAR, and mTOR dependent. To verify the growth promoting effects of stress in an in vivo regeneration model, we performed a sciatic nerve crush procedure immediately after acute stress. Mice exposed to the stressor demonstrated increased axon regeneration, as well as increased measures of allodynia. Perhaps stress effects on pain, as seen in two neuropathic models, are related to increased structural neuron plasticity. We suspect that the molecular underpinnings of structural plasticity (e.g., NMDAR, mTOR) are shared by measures of functional neuron plasticity, and that together, these act as substrates for stress effects on pain.
Pro-inflammatory cytokines are effectors of both neural plasticity and neuroinflammation mechanisms of neuropathic pain. Cytokines can directly stimulate neurons, or indirectly, via effects on immune cells. One particular pro-inflammatory cytokine, MIF, is of unique relevance to stress. MIF is induced by stress and GCs, and overrides the anti-inflammatory action of GCs. Its role in pain was unknown, but its constitutive and widespread distribution made it a putative candidate for initiating and propagating injury responses. We hypothesized that stress potentiates pain via MIF. To our surprise, MIF KO mice did not develop pain-like behaviors after nerve injury or CFA-induced paw inflammation, regardless of stress. Likewise, a small molecule inhibitor of MIF reduced allodynia after peripheral nerve injury. Our data demonstrate that MIF plays a critical role in the development and maintenance of persistent pain. To determine if MIF is involved in stress-induced enhancement of pain-like behavior, mice were treated with the MIF inhibitor prior to restraint stress. Stress effects on allodynia were absent in the inhibitor-treated mice, indicating a role for MIF in this context. Supporting its role as a pro-algesic cytokine, intraplantar administration of recombinant MIF (rMIF) alone produced dose-dependent allodynia. Acute restraint stress increased rMIF-induced allodynia. To identify a cellular substrate for MIF, we examined both microglia and neurons. Primary microglia treated with rMIF increased transcript levels of iNOS, IL-1β, and IkBa, indicating the induction of a pro-inflammatory phenotype by rMIF. rMIF also increased measures of both functional and structural sensory neuron plasticity. By targeting both microglial reactivity and neuron plasticity, MIF may play a dual-role in the pathology of persistent pain. These data describe a previously unknown, essential role for MIF in persistent pain.
In summary, we have contributed to the field of neuropathic pain the novel observations that, 1) stress exacerbates the development of neuropathic pain-like behavior, and 2) MIF is essential for persistent pain states. We have identified three putative molecular bases for stress and GC effects on pain: NMDAR, mTOR, and MIF. Of these, MIF is a particularly novel therapeutic target for pain management. We propose that stress and GCs account for some variability in the development and severity of pain in humans.