Full-length PKCs in SCDH have previously been shown to contribute to the induction of pain hypersensitivity, or allodynia and hyperalgesia and neuronal sensitization that are driven largely by continued inputs from the peripheral site of injury [4, 5, 8]. The present results, however, indicate the autonomous PKC isoform, PKMζ, plays a specific essential role in sustaining spinal plasticity underlying persistent pain. PKMζ was persistently increased in SCDH by cutaneous injuries (i.pl. formalin and capsaicin), or by spinal stimulation with a glutamate agonist, each of which produces persistent pain or long-lasting pain hypersensitivity that is selectively reversed by inhibition of PKMζ, but not full-length PKCs. Additional shorter-term increases in full-length PKCζ and PKCι/λ 45 min after i.pl. formalin injection are consistent with the behavioral data suggesting a role of full-length PKCs in the induction of persistent pain. However, at later time points (2 and 24 h) after i.pl. capsaicin or i.t. DHPG, only PKMζ is elevated, indicating that SCDH PKMζ specifically is persistently elevated in association with the maintenance of persistent pain.
The specificity of PKMζ effects in sustaining spinal plasticity underlying persistent pain is highlighted by the ability of i.t. treatment with a selective PKMζ inhibitor to reverse established persistent pain or pain hypersensitivity, while treatment with a full-length PKC inhibitor fails to do so, despite its ability to prevent the induction of persistent pain. This transition from the transient role of full-length PKCs to the persistent function of autonomously active PKMζ in pain is reminiscent of similar transitions in LTP and memory consolidation [13–17]. Furthermore, the finding that spinal PKMζ inhibition not only reversed hind paw capsaicin-induced allodynia, but also hind paw capsaicin-induced sensitization of SCDH WDR neurons (evoked by the same mechanical stimuli) provides further support for the role of PKMζ in the maintenance of spinal nociceptive plasticity. Earlier studies demonstrating that pretreatment , but not post-treatment  with NPC-15437 reduced capsaicin-induced sensitization of spinal neurons to mechanical stimulation, is also consistent with our behavioral data showing that post-treatment with NPC-15437 does not reverse persistent pain or allodynia after cutaneous injury or spinal stimulation. Importantly, this earlier study used the same PKC inhibitor we employed, which does not inhibit PKMζ.
Our results are consistent with and extend upon three recent papers that have examined the role of PKMζ in nociception [21, 22, 35]. The first paper showed that PKMζ was increased within the anterior cingulate cortex (ACC) of neuropathic mice, and that PKMζ inhibition in this brain area, but not SCDH, produced anti-allodynic effects in neuropathic mice . A role of PKMζ in nerve-injury induced neuronal plasticity in ACC was further supported since ZIP reduced α-amino-3-hydroxy-5-methyl-4-isoxazole-propionate (AMPA) receptor-mediated excitatory post-synaptic potentials (EPSPs) in ACC neurons of nerve-injured, but not sham mice. Additionaly, PKMζ inhibition reduced spinal neuron AMPA receptor-mediated EPSPs in nerve-injured, as well as sham mice . Importantly, these electrophysiological experiments were performed in brain and spinal cord slices, where there are no longer ongoing inputs from the injured nerves. However, the lack of effect of spinal PKMζ inhibition on allodynia does not rule out a role for PKMζ in spinal plasticity, if the inhibitor is unable to overcome the effects of ongoing inputs from injured nerves. Thus, our results, showing a lack of anti-allodynic effect of spinal ZIP in CCI rats, support the findings of Li et al.  in neuropathic mice. However, we found that i.t. ZIP relieved allodynia in late CPIP rats, in which allodynia depends on central changes (as peripheral pathology has resolved), but not for early CPIP rats, in which allodynia depends on ongoing peripheral inputs . This suggests that while PKMζ may contribute to spinal plasticity in these models, the ability of spinal PKMζ inhibition to relieve allodynia is overshadowed in CCI rats or early CPIP rats when there is significant ongoing peripheral input.
The second paper showed that inhibition of PKMζ in spinal cord prevented the enhancement of allodynia (in response to cutaneous injury) or persistent pain (in response to i.t. DHPG) that occurred following an earlier resolved cutaneous injury (either i.pl. IL-6 injection or plantar incision) . In this case, PKMζ inhibition was shown to prevent the development of enhanced responses to nociceptive stimulation in situations where an earlier injury results in nociceptive priming. Our results support this earlier study's conclusions, extending them to include ZIP's ability to reverse an established persistent pain or pain hypersensitivity. Importantly, Asiedu et al.  did not reverse allodynia, but rather prevented allodynia induced by a second injury. Together the two studies strongly implicate the role of PKMζ in maintaining spinal nociceptive plasticity for both nociceptive priming and persistent pain.
Our data is also consistent with a recent third report published in this journal showing that PKMζ activity (specifically localized in spinal projection neurons) is increased after formalin injury, and that a PKMζ inhibitor reduced nociceptive behaviors, WDR neuronal activity and Fos protein levels in SCDH in response to hind paw formalin injection . However, these investigators only examined the effects of pretreatment with ZIP on formalin-induced nociceptive and dorsal horn neuronal responses. While data here and in this previous study implicate a role of PKMζ in the induction of nociceptive plasticity; the current study, showing a post-treatment reversal of formalin-induced persistent nociception, as well as allodynia and WDR sensitization induced by hind paw capsaicin, and allodynia induced after i.t. DHPG, or in late CPIP rats, also implicates PKMζ in the maintenance of spinal nociceptive plasticity in persistent pain.
Marchand et al.  also showed that spinal PKMζ inhibition was unable to reduce mechanical hypersensitivity in neuropathic animals, although they did show that ZIP post-treatment attenuated mechanical hypersensitivity in rats with complete Freund's adjuvant (CFA)-induced hind paw inflammation. However, the effect on mechanical hypersensitivity was only partial and short-lived (30 min), with slightly better effects on thermal hypersensitivity. These investigators attributed the differential effects of ZIP on neuropathic and inflammatory pain to disparate molecular mechanisms of nociceptive plasticity underlying these two injuries. However, based on the differential effects of ZIP in CCI and early CPIP rats vs. late CPIP rats observed in our study, the disparate results could also be explained by differences in the contribution of ongoing inputs from injured peripheral tissue, which are particularly enhanced in neuropathic animals [32, 33], and we expect are reduced in late CPIP rats when the injured tissue has healed . While it is true that there may be ongoing peripheral inputs in CFA rats, it could also be argued that the high levels of spontaneous C- and Aδ- and A-β fiber activity in damaged peripheral nerves in CCI rats (generated both at the site of injury and at the dorsal root ganglion [32, 33]) may represent greater peripheral drive. Furthermore, peripheral drive from inflamed CFA hind paws may explain the partial and short-lived effects of i.t. ZIP on mechanical and thermal hypersensitivity in these animals.