Central Mechanisms in the Maintenance of Chronic Widespread Noninflammatory Muscle Pain

Abstract

Chronic widespread pain (CWP) conditions such as fibromyalgia and myofascial syndromes are characterized by generalized pain, tenderness, morning stiffness, disturbed sleep, and pronounced fatigue. However, CWP pathophysiology is still unclear. A number of hypotheses have been proposed as the underlying pathophysiology of CWP: muscular dysfunction/ischemia, central sensitization, and a deficit in endogenous pain-modulating systems. This article reviews the current and emerging literature about the pathophysiology and neurobiology of chronic widespread musculoskeletal pain. Widespread musculoskeletal pain results in changes in the central nervous system in human subjects and animal models. These changes likely reflect alterations in supraspinal modulation of nociception, and include increases in excitatory and decreases in inhibitory modulation pathways. These alterations in excitation and inhibition likely drive changes observed in the spinal cord to result in central sensitization, and the consequent pain and hyperalgesia.

Introduction

Chronic pain is an abnormal and nonprotective response. It has been defined as pain that outlasts normal tissue healing time, pain that is out of proportion to the inciting incident, or pain that lasts longer than 6 months. Approximately 14% of the US population suffers from chronic widespread muscle pain (CWP) conditions such as fibromyalgia (FM) [1]. It is estimated that FM affects 6 million Americans (2%−7% of the population), making it a common pain condition [1,2]. FM syndrome is characterized by generalized pain, tenderness, morning stiffness, disturbed sleep, and pronounced fatigue [3]. The diagnostic criteria proposed by the American College of Rheumatology include widespread pain in conjunction with tenderness on palpation of 11 or more of 18 specified tender points [2].

Pathophysiology of CWP in Humans

Several hypotheses have been proposed as the underlying pathophysiology of CWP: muscular dysfunction/ischemia [4,5], central sensitization [1,6,7], or a deficit in endogenous pain-modulating systems [8,9•]. Deficits in the central processing and central sensitization are suggested by studies showing a generalized decrease in mechanical thresholds [10-12], temporal summation to thermal and deep mechanical stimulation [12-14], larger areas of referred pain after infusion of hypertonic saline into the muscle [13], and alterations in descending modulation that are interpreted as decreased endogenous pain inhibition [15]. Temporal summation in patients with FM is enhanced compared with normal control subjects and is followed by exaggerated after-sensations and temporal summation at low stimulus frequencies [16]. Blockade of N-methyl-D-aspartate (NMDA) receptors with systemic ketamine in people with FM reduces the local and referred pain associated with intramuscular infusion of hypertonic saline [10]. However, blockade of NMDA receptors with dextromethorphan in FM subjects produces an effect similar to that in healthy controls, suggesting that enhanced activity at NMDA receptors does not contribute to the differences in central pain processing between FM and healthy subjects [16]. However, other neural mechanisms may contribute to the pain associated with FM. Patients with FM have increased concentrations of substance P and nerve growth factor, and decreases in serotonin in the cerebrospinal fluid [17,18], suggesting that increased excitation and a simultaneous decrease in inhibition can be found in FM. Taken together, there appears to be substantial support for increased central sensitization in people with FM. However, recent evidence also supports a potential role of the peripheral nervous system in the mechanisms underlying development of CWP [19•,20].

Animal Model of CWP

Musculoskeletal pain syndromes such as FM and myofascial pain syndrome can be difficult to treat, and until recently, basic science studies focused on cutaneous pain. Recent works have established both inflammatory and noninflammatory models of muscle pain that result in widespread, long-lasting hyperalgesia [21-23]. Inflammatory models typically involve intramuscular injection of carrageenan, which produces an initial acute inflammatory response that converts to a chronic inflammatory response by 1 week [23]. This model is associated with the development of heat and mechanical hyperalgesia, as well as central hyperexcit-ability of dorsal horn neurons (central sensitization) [23]. However, widespread hyperalgesia without tissue injury is common and likely distinctly different from that associated with muscle inflammation.

To model CWP, our laboratory developed and characterized an animal model of chronic muscle hyperalgesia induced by two intramuscular injections of acidic saline [21]. This model is unique from other muscle pain models because 1) there is long-lasting bilateral mechanical hyperalgesia of the paw, muscle, and viscera; 2) there is no peripheral tissue damage; and 3) hyperalgesia is maintained by changes in the central nervous system (CNS) [21,24-28].

Characterization of the CWP model

Initial characterization of the model showed that repeated injection of pH 4.0 saline into the gastrocnemius muscle produced a long-lasting, widespread mechanical hyperalgesia without motor deficits or tissue damage. This hyperalgesia was dependent on dose and time of injection such that the lowest pH tested (pH 4.0) produced the greatest hyperalgesia, and interinjection intervals of 2 and 5 days (pH 4.0 saline) produced equivalent and significant decreases in mechanical withdrawal threshold bilaterally. However, inter-injection intervals of 10 days did not produce hyperalgesia [21], suggesting there is a critical window for the induction of hyperalgesia by the second injection.

Pharmacologic characterization of the acidic saline model shows similarities to that of people with FM. Specifically, the hyperalgesia is reversed by opioid agonists delivered spinally, spinal NMDA antagonists, as well as systemic pregabalin, NMDA antagonists, potassium channel openers, and sodium channel blockers [27,29,30]. However, there is no effect on the hyperalgesia with cyclooxygenase-2–specific inhibitors, anticonvulsants such as lamotrigine, or anxiolytic drugs such as diazepam [29]. Further, aerobic exercises reverse the hyperalgesia in this model [31] as in human subjects with FM [32•], and muscle fatigue enhances the development of hyperalgesia [33•].

Experimental Models of Muscle Pain in Human Subjects

Decreasing pH increases activity of nociceptors and produces a painful response in humans [34-36]. The infusion of acidic buffer into the tibialis anterior muscle produces pain at the site of infusion as well as referred pain at the ankle, a site distal to the infusion. Further, there are decreases in pressure pain threshold (hyperalgesia) at both sites of infusion (primary hyperalgesia) and at the referred pain site (secondary hyperalgesia) [35]. Similarly, single infusions of hypertonic saline into the anterior tibialis muscle of human subjects produce pain at the site of hypertonic saline infusion and distal to the site of infusion at the ankle; however, hyperalgesia is not present in this model [37-39]. The distal spread was dependent on volume of infusion such that lower volumes did not produce a referred pain area; increasing the volume of infusion increased the referral of distal pain [37]. Thus, in human subjects, stimulation of deep tissue with acidic buffer or hypertonic saline produces local and referred pain, and stimulation with an acidic buffer also produces primary and secondary hyperalgesia.

Spinal Mechanisms in CWP

Increased release of glutamate

Excitatory amino acids (EAAs) glutamate and aspartate play an essential role in nociception transmission through the spinal cord [39,40]. In the model of noninflammatory muscle pain induced by repeated acid intramuscular injections, spinal blockade of ionotropic glutamate receptors, both NMDA and AMPA/kainate receptors, reversed the hyperalgesia [24]. Further, blockade of NMDA (but not AMPA/kainate) receptors during the second injection of acidic saline delayed the onset of mechanical hyperalgesia [24], suggesting a role for increased glutamate release in the development of mechanical hyperalgesia associated with repeated acid injections. In parallel, there was an increased release of the EAAs glutamate and aspartate in the dorsal horn of the spinal cord in response to the second injection of acidic saline, compared with the first injection or controls injected with pH 7.2 saline [25]. In addition, 1 week after the second injection of acidic saline, baseline concentrations of glutamate and aspar-tate were increased in the dorsal horn of the spinal cord, suggesting long-lasting increases in excitatory neurotransmitter release at the level of the spinal cord. The increases in deep dorsal horn EAA concentrations after the second intramuscular injection of acid paralleled the development of mechanical hyperalgesia and central sensitization in this model of noninflammatory muscle-induced hyperalgesia [21,41].

Activation of the cyclic AMP pathway

Activation of the cyclic AMP (cAMP) pathway in the spinal cord and supraspinal pathways plays a crucial role in the transmission of the nociceptive input. Spinal activation of the cAMP pathway produces mechanical hyperalgesia and increases the response of spinothalamic tract neurons to noxious but not innocuous mechanical stimuli [22,42,43]. In mice that lack adenylate cyclase 1 and 8, hyperalgesia does not develop after intramuscular carrageenan [44]. Furthermore, blocking adenylate cyclase or protein kinase A (PKA) prevents the mechanical hyperalgesia and allodynia produced by intradermal, intramuscular, or intraarticular injection of capsaicin, and repeated intramuscular acidic saline injections [22,26,45].

The catalytic subunit of PKA translocates to the nucleus and phosphorylates cAMP-response-element-binding protein (CREB) at serine-133, which is necessary for gene transcription. After repeated intramuscular injections of acidic saline, there is an increase in CREB and phosphorylated CREB (p-CREB), bilaterally in the spinal cord dorsal horn; the increases in p-CREB are prevented by blockade of the cAMP pathway [26]. Although p-CREB is found in spinothalamic tract cells, the increases observed in p-CREB occur in nonspinothalamic tract neurons [45]. The increases in p-CREB after repeated acid injections are time dependent such that increases occur 24 hours but not 1 week after induction of hyperalgesia [26]. Similarly, the effects of blockade of the cAMP pathway are also time dependent such that hyperalgesia is reversed 24 hours but not 1 week after the induction of deep tissue hyperalgesia [22,26]. Thus, activation of the cAMP pathway after deep tissue injury occurs in a time-dependent manner to initiate long-term hyperalgesia potentially through activation of gene transcription by p-CREB.

PKA also phosphorylates the NMDA receptor at serine-897 of the NR1 subunit (p-NR1). Increases in phosphorylation of the NR1 subunit of the NMDA receptor occur after repeated intramuscular acid injection in lamina X spinothalamic tract neurons in the spinal cord [45]. Other protein kinases could also underlie the hyperalgesia associated with repeated acid injections. Numerous studies support a role of protein kinase C (PKC) and the calcium/calmodulin pathway in models of tissue injury [46], and activation of the PKC pathway with phorbol esters in the spinal cord produces mechanical hyperalgesia [47]. However, blockade of PKC has no effect on the hyperalgesia produced by repeated acid injections [47], suggesting that the PKC pathway is not involved at the spinal level in CWP. Involvement of other intracellular mechanisms has not been tested in this unique model of noninflammatory muscle pain.

Glia are not activated

Glial cells in the CNS, particularly the spinal cord, play a critical role in processing of nociceptive information [48,49]. Glia express receptors for many neurotransmitters (including glutamate receptors) and are involved in the clearance of neurotransmitters from the synaptic cleft. Activation of astrocytes and microglia occurs in a number of pain models, including neuropathic and inflammatory [50,51]. However, after repeated intramuscular acid injections, the hyperalgesia is not affected by single or repeated blockade of interleukin-1, a glial metabolic inhibitor, or interleukin-10 [52]. Further, using immunohistochemistry there was no activation of glial cells in the spinal cord after repeated intramuscular injections of acid [52].

Supraspinal Mechanisms in CWP

There is evidence that secondary hyperalgesia is maintained by changes in the CNS [53], which is facilitated and maintained by two distinct areas in the rostral ventral medulla (RVM) that have been implicated in a number of studies on descending inhibition: the nucleus raphe magnus (NRM) and the nucleus gigantocellularis (Gi) [54-56]. In the noninflammatory model of CWP, reversible blockade by ropivacaine (local anesthetic) of the NRM or Gi bilaterally during the second, hyperalgesia-inducing intramuscular injection of pH 4.0 saline prevents the development of primary muscle and secondary cutaneous mechanical hyperalgesia 24 hours later [28]. Moreover, if the hyperalgesia has already developed, ropivacaine microinjected into NRM or Gi reverses the completely developed primary (muscle) and secondary (paw) mechanical hyperalgesia [28]. This suggests that cells in the NRM and Gi (or axons passing through these RVM cell groups) are critical for the development and maintenance of mechanical hyperalgesia after muscle insult [28]. Clinical studies show that deficits in the descending pain inhibitory systems are impaired in chronic conditions such as FM [9•,15,57]. However, it is unknown in these studies if there is an increase in facilitation or a decrease in inhibition. The animal studies suggest an increase in facilitation because blockade of neuronal activity reverses the hyperalgesia. Thus, as in animal models, pain in people with CWP may be due to a shift in the balance between endogenous inhibitory and facilitatory influences, representing a dysfunction within endogenous systems of modulation.

Peripheral Mechanisms in CWP

ASIC3 is important for induction

There are four acid-sensing ion channels found in the mammalian nervous system: ASIC1, ASIC2, ASIC3, and ASIC4. Of these, ASIC3 appears to play a critical role in the development of hyperalgesia after repeated intramuscular acid injection or deep tissue inflammation. In ASIC3 knockout mice, the secondary mechanical hyperalgesia that normally occurs after repeated intramuscular acid injections, muscle inflammation, or joint inflammation does not develop [41,58,59]. However, primary hyperalgesia after paw or joint inflammation still develops, similar to controls [59-61]. Further, ASIC3 expression in peripheral tissues is critical for the development of the hyperalgesia because re-expression of ASIC3 in muscle tissue (but not skin) of ASIC3 knockout mice restores the development of hyperalgesia after carrageenan muscle inflammation [58]. In parallel, central sensitization of dorsal horn neurons, measured as a bilateral spread of receptive fields and increased mechanical sensitivity, does not develop in ASIC3 knockout mice [41]. Further, there is an upregulation of ASIC3 in joint afferents innervating the synovium and their cells bodies (dorsal root ganglion) after joint inflammation [59]. However, in ASIC1 knockout mice, secondary hyperalgesia develops similarly to controls [41]. Together these data suggest that ASIC3 is critical for development of secondary hyperalgesia and central sensitization that occurs after insult to deep tissues.

Neurotrophin-3 in muscle is protective

The neurotrophins are a family of secreted factors (nerve growth factor [NGF], brain-derived neurotrophic factor, neurotrophin-3 [NT-3], and neurotrophin-4/5) that have trophic activity on different classes of dorsal root ganglion neurons. The NGF-related neurotrophins signal via tyrosine kinase (Trk) receptors and bind a common receptor with low affinity (p75), and the subtype TrkC shows high affinity for NT-3 [62]. All of these neurotrophins are expressed in the muscle [63]; most muscle afferents express TrkC and are responsive to NT-3 [64]. Mice that overexpress NT-3 do not develop hyperalgesia after repeated injection of acidic saline [65]. Similarly, injection of NT-3 into muscle during the development of hyperalgesia prevented the onset of hyperalgesia; however, NT-3 was ineffective when delivered after the development of hyperalgesia [65]. Further, intrathecal or systemic delivery of NT-3 had no effect on the development of hyperalgesia, suggesting that activation of TrkC in muscle is protective. Thus, the protective effects of NT-3 are important in induction of hyperalgesia, but it does not have a role after induction.

Conclusions

Widespread musculoskeletal pain results in CNS changes in human subjects and animal models. Peripherally, activation of ASIC3 is necessary for development of widespread muscle pain, and NT-3 prevents development of hyperalgesia. Central changes are also critical for development of widespread muscle pain and likely reflect alterations in supraspinal modulation of nociception; these include increases in excitatory and decreases in inhibitory modulation pathways. These alterations in excitation and inhibition likely drive changes observed in the spinal cord to result in central sensitization and the consequent pain and hyperalgesia.