In addition to dorsal horn neurons, central sensitization also develops in rostroventral medial medulla (RVM) neurons, amygdala neurons, and cingu- late cortex neurons following tissue injury (Urban and Gebhart 1999; Por- reca et al. 2002). Inﬂammation induces phosphorylation of AMPA receptor (GluR1) in the RVM, in a PKC- and CaMK-dependent manner, which may regulate descending facilitation (Guan et al. 2004). The spino–parabrachio– amygdaloid pathway is implicated in nociception. Tissue injury induces an enhanced NMDA receptor function in rat amygdala neurons via PKA activa- tion (Bird et al. 2005). The PKA pathway in the anterior cingulate cortex (ACC) also contributes to inﬂammatory pain, since calcium-dependent adenylate cy- clase AC1 and AC8 are highly expressed in the ACC and inﬂammatory pain is attenuated in AC1 and AC8 knockout mice. The action of AC1 and AC8 appears to be mediated by the cAMP–PKA–CREB pathway (Wei et al. 2002a). Neuronal activity in the ACC could inﬂuence nociceptive transmission in the dorsal horn of the spinal cord by activating the endogenous facilitatory system.
Pain experience includes not only a sensory-discriminative component de- scribing the quality, intensity, and spatiotemporal characteristics of the sensa-
tion, but also an emotional-affective component, referring to the unpleasant- ness or aversion of sensation. Emotional distress could be the most disruptive and undesirable feature of painful experiences. Physiological arousal and hy-
pervigilance to pain cause a negative affect, such as fear, anxiety, anger, worry, aversion, and depression, which in turn could alter pain sensation. While cur- rent studies focus primarily on the sensory component of pain, mechanisms
underlying the affective dimension of pain have recently received more at- tention. ACC is a part of the brain’s limbic system and plays an important role in affect. ACC neurons respond to noxious stimuli with very large, often whole-body receptive ﬁelds, and acquire responses to environmental cues that
predict a painful stimulus (Koyama et al. 1998). Formalin-induced conditioned place avoidance (F-CPA, noxious conditioning) has been established as an an-
imal model to examine affective pain and pain-associated memory. Lesions of ACC in animals produce severe deﬁcits in avoidance conditioning (Johansen et al. 2001).
Molecular mechanisms underlying learning and memory have been inten- sively investigated in hippocampal neurons. It has been demonstrated that PKA, ERK, and CaMK can all converge on CREB phosphorylation, leading to long-term memory via gene transcription (Lonze and Ginty 2002). These mechanisms may also contribute to affective pain and pain-associated mem-
ory in ACC neurons. Activation of NMDA receptors results in Ca2+ inﬂux and is required for formalin-induced affective pain (Lei et al. 2004). Ca2+-dependent
CaMK-IV is required for CREB activation in the ACC by noxious shock and for fear memory, but behavioral responses to acute noxious stimuli or tissue inﬂammation may not require CaMK-IV (Wei et al. 2002b). ERK activation
has been implicated in several forms of learning and memory, including fear conditioning, spatial learning, conditioned taste aversion, etc. We found that pain-related aversion (F-CPA) induced ERK activation in the ACC, and that
microinjection of the MEK inhibitor PD98059 in the ACC signiﬁcantly inhib- ited affective pain as well as CREB phosphorylation following F-CPA (Zhang et al. 2005). These results indicate that the ERK/CREB signaling pathway, which has been implicated in the sensory component of pain in spinal neurons, may
also participate in affective pain in ACC neurons.
Glial and Immuno–Mechanisms
Studies on pathological pain mostly focus on the responses of the neurons and neuronal-speciﬁc mechanisms of hypersensitivity and chronicity. However, glial cells express various receptors for neurotransmitters and neuromodu- lators (Watkins et al. 2001; Ji and Strichartz 2004). There is accumulating evidence indicating a role of spinal glial cells in the pathogenesis of pain. Both microglia and astrocytes are activated in inﬂammatory and neuropathic pain
conditions. Multiple mediators such as proinﬂammatory cytokines (e.g., IL-1β, IL-6, and TNF-α) and inﬂammatory enzymes (e.g., iNOS, COX-1, and COX-2)
are synthesized in spinal astrocytes and microglia (DeLeo and Yezierski 2001; Watkins et al. 2001). There is increased synthesis of these mediators in activated
glia following injuries; all of them are shown to produce pain sensitization. Spinal injection of the glial toxin ﬂuorocitrate, glial modulator propentofylline, or nonspeciﬁc microglial inhibitor minocycline has been shown to reduce in-
ﬂammatory and neuropathic pain (Meller et al. 1994; Watkins et al. 1997; Sweitzer et al. 2001; Raghavendra et al. 2003). In particular, several receptors such as ATP receptor P2X4, chemokine receptors CCR2 and CX3CR1, and Toll-like receptor TLR-4 are expressed in spinal microglia after nerve injury.
Blockade of these receptors results in a reduction of neuropathic pain (Abbadie et al. 2003; Tsuda et al. 2003; Verge et al. 2004; Tanga et al. 2005).
It is not very clear how activation of these receptors in microglia leads to pain sensitization. We have proposed that activation of MAPKs may mediate pro-nociceptive effects of these receptors (Ji and Strichartz 2004). Nerve injury induces a drastic and widespread activation of p38 MAPK in the spinal cord. This activation begins at 12 h and peaks 3 days after nerve injury, in parallel with the time course of microglial activation. Moreover, p-p38 is completely colocalized with OX-42 (CD-11b), a hall marker for microglia (Jin et al. 2003). Inhibition of p38 activation in microglia attenuates nerve injury-induced me-
Fig. 5 Microglial regulation of pain facilitation via p38 MAPK. Inﬂammation and nerve injury release the signaling molecules fractalkine, ATP, and substance P from primary afferents, leading to p38 activation in spinal microglia. p38 activation results in the synthesis of multiple inﬂammatory mediators (e.g., IL-1β, IL-6, TNF-α, PGE2 , NO), enhancing and prolonging pain facilitation via both presynaptic (glutamate release) and postsynaptic (AMPA and NMDA receptor sensitization) mechanisms. Microglia may also synthesize
neuromodulator BDNF to sensitize postsynaptic dorsal horn neurons
chanical allodynia (Jin et al. 2003; Tsuda et al. 2005). p38 activation in microglia is likely to induce the synthesis of inﬂammatory mediators as well as neuro- modulator BDNF. The release of these neuroactive substances could diffuse to synaptic regions and sensitize pain transmission neurons via both presy- naptic and postsynaptic mechanisms (Fig. 5). In addition to p38, ERK is also activated in spinal microglia after nerve lesion and contributes to neuropathic pain (Zhuang et al. 2005).
Compared to rapid activation of microglia, injury induces a delayed but per- sistent activation of astrocytes, suggesting a particular role of astroglia in the persistence of chronic pain. Nerve injury induces persistent activation of JNK
in spinal astrocytes (Ma and Quirion 2002; Zhuang et al. 2006). Importantly, this activation is requried for the maintenance of neuropathic pain (Zhuang et al. 2006). Further, ERK is activated in spinal astrocytes at late times of
nerve injury and contributes to late maintenance of neuropathic pain (Zhuang et al. 2005). Therefore, MAPKs could mediate different phases of chronic pain by regulating the activity of different subtypes of glial cells.
In addition to spinal glial cells, MAPKs can also be activated in glial cells in the sciatic nerve (Schwann cells) and DRG (satellite cells), as well as in inﬂammatory cells inﬁltrating the damaged tissue after injuries, increasing the sensitivity of sensory neurons by producing inﬂammatory mediators (Ji and Strichartz 2004).
Protein Kinase Inhibitors in Clinical Studies
A number of diseases including cancer, diabetes, and inﬂammation are as- sociated with perturbation of protein kinase-mediated signal transduction. Chromosomal mapping has shown that 244 kinases map to disease loci or can- cer amplicons (Manning et al. 2002). Many kinase inhibitors are developed to treat cancer in clinical trials. These include receptor tyrosine kinas inhibitors (e.g., EGF receptor inhibitors) and several MEK inhibitors such as CI-1040, PD0325901, and ARRY-142886, and Raf inhibitor BAY 43-9006 (Cohen 2002; Sebolt-Leopold and Herrera 2004). In the cancer ﬁeld, protein kinase inhibitors are proving to be well-tolerated compared with conventional chemotherapeu- tic treatments. The MEK inhibitor PD184352 was shown to inhibit the growth of colon cancer that was implanted into mice, without causing obvious ad- verse side effects over several months treatment (Sebolt-Leopold et al. 1999). The lack of signiﬁcant side effects of MEK inhibitors is surprising since the ERK cascade has been implicated in many cell processes including cell growth. However, the essential roles of the ERK cascade in proliferation and differen- tiation are only required during development, and this pathway might be far less crucial for normal function in adults. Since cancer pain is so devastating (Mantyh et al. 2002), MEK inhibitors could be assessed for anticancer pain effects while being tested for antitumor effects in clinical trials.
Several p38 MAPK inhibitors including SB281838, BIRB0796, Ro320-1195, and SCIO-469 are in clinical trials for rheumatoid arthritis (Cohen 2002; Watkins and Maier 2003; Nikas and Drosos 2004). Given the important role of p38 MAPK in several pathological pain conditions, these inhibitors should also be tested for clinical pain, if they are well tolerated. JNK pathway is essential for regulating neurotoxicity and apoptosis. CEP-1347, an inhibitor for the JNK upstream kinase MLK (mixed lineage kinase) was in clinical trial for neurode- generative diseases (Cohen 2002). Given the fact that neuropathic pain is also regarded as a neurodegenerative condition and JNK inhibition attenuates neu- ropathic pain (Zhuang et al. 2006), CEP-1347 or JNK inhibitor (e.g., D-JNKI-1) could also be tested for neuropathic pain in clinical studies.
In addition to MAPK inhibitors, wortmannin, a selective inhibitor for PI3K
pathway, is in clinical trials for the treatment of osteoporosis (Noble et al. 2004). Inhibitors for Rho kinase (ROCK) such as HA1077 (AT877, fasudil) are used to treat cerebral vasospasm. Indirubin is a CDK5 inhibitor used to treat neurode- generative disorders (Noble et al. 2004). Since all the kinases are implicated in pain facilitation, these inhibitors could be potentially useful for the manage- ment of clinical pain.
Many speciﬁc protein kinase inhibitors that cannot be used as drugs for reasons of toxicity, pharmacology, or solubility—such as the MEK inhibitors
PD98059 and SB203580—could be useful reagents for basic research. While current efforts on developing protein kinase inhibitors focus on small molecules, membrane permeable peptide inhibitors could provide another efﬁcient way
to block the function of protein kinases. Recently, a peptide JNK inhibitor, derived from JNK binding domain of JNK-interacting protein-1 (JIP-1), was designed to block selectively the access of JNK to c-Jun and other substrates
by a competitive mechanism. A TAT sequence (transporter sequence) is linked to the peptide, making the peptide membrane permeable. This highly speciﬁc peptide inhibitor is an extremely potent neuroprotectant both in vitro and in vivo (Borsello et al. 2003; Borsello and Bonny 2004). It is also highly effec-
tive in suppressing neuropathic pain symptoms after nerve injury (Zhaung et al. 2006). Studies on kinase regulation of pathological pain will greatly beneﬁt from the development of speciﬁc and potent kinase inhibitors.
There are over 500 protein kinases in the human genome. More than 20 protein kinase inhibitors are in clinical trials, and many others have entered clinical trials without their structure being disclosed, and a great many more are still in preclinical studies. Protein kinases are becoming the second largest group of drug targets after GPCRs, accounting for 20%–30% of drug discovery activity in many pharmaceutical companies. There is increasing evidence suggesting that
protein kinases play essential roles in regulating various kinds of pathological pain in animal models. Although many protein kinase inhibitors are in clinical trials for treating different diseases, especially cancer, they are not speciﬁcally being tested for clinical pain.
Animal studies have shown that kinase inhibitors do not affect basal pain perception, indicating that protein kinases are not very active in normal con-
ditions. Although several kinases such as ERK, PI3K, and Cdk5 are important for cell growth during development, they play different roles in the adult by regulating neural plasticity. Importantly, protein kinases are activated after tissue injuries and contribute to the induction and maintenance of patholog-
ical pain by posttranslational, translational, and transcriptional regulation. Therefore, protein kinase inhibitors are different from traditional analgesics such as morphine. Instead of raising pain thresholds, they are antihyperalgesic
and antiallodynic by “normalizing” pain sensitivity.
There is growing interest in MAPK cascades in the ﬁeld of pain research for the following reasons:
– There are speciﬁc antibodies available to detect the activation of ERK, p38, and JNK. Potentially, phosphorylated MAPKs could be used as biomarkers for pathological pain.
– Compared to other kinases, there are relatively speciﬁc inhibitors available to study the function of MAPK pathways.
– MAPK pathways appear to be downstream to many other kinases.
– Multiple MAPK inhibitors are in clinical trials for treating different diseases.
Chronic pain (persistent pathological pain) has affected hundreds of mil- lions people in the world. Although pathological pain is an expression of neural plasticity manifesting as peripheral and central sensitization, the activation of glial cells enhance and prolong neuronal sensitization. Since traditional painkillers were mainly designed to target neurons and are only partially effective, the development of glial- or neural/glial-targeting protein kinase in- hibitors (e.g., MAPK inhibitors) could lead to more effective treatments for chronic pain.