DISCUSSION
Using in vivo and ex vivo methods, coupled with genetic and pharmacological strategies, this work highlights the involvement of Cav3.2 T-type channels in the development and maintenance of edema and inflammatory hypersensitivity, with a requirement of Cav3.2 channels being required in T CD4+ cells, macrophages and the primary afferent fibers C-LTMRs. From a clinical point of view, these results suggest that T-type calcium channels could be an interesting target for the treatment of inflammatory pain.
We first demonstrated that inhibition of T-type calcium channel by TTA-A2, ABT-639 and Cav3.2 channels deletion (Cav3.2 KO mice), reduced mechanical allodynia and hyperalgesia in the carrageenan and CFA models. These results are consistent with those of previous studies using the same models (François et al., 2015; Kerckhove et al., 2019) and in other inflammatory contexts (Choi et al., 2007; Kerckhove et al., 2014). Interestingly, repeated administration of TTA-A2 accounts for a maintained antihyperalgesic/anti-allodynic action. In parallel, we reported for the first time the involvement of Cav3.2 channels in the inflammation process. We observed an anti-inflammatory effect after genetic and pharmacological inhibition of Cav3.2 channels, suggesting a pathophysiological role of these channels not only in pain but also in the inflammatory process. These results are consistent with others showing a contribution of Cav3.2 to the development of bladder inflammation in cyclophosphamide-induced cystitis (Matsunami et al., 2012). In the CFA model, acute administration of TTA-A2 failed to reduce edema size, suggesting that the dose or duration of treatment was not sufficient in a setting of chronic inflammation. In contrast, their repeated administration significantly reduced edema volume throughout the experiment. The genetic and pharmacological inhibition of Cav3.2 channels systematically induced an almost complete reduction of allodynia/hyperalgesia but only partially inhibited edema volume. These results confirm the well documented dissociation between pain and edema (Lee and Jeong, 2002) and suggest a greater involvement of Cav3.2 channels in the pain process than in inflammation which is regulated by numerous various mechanisms.
Given the marked effect of the inhibition of Cav3.2 channels on inflammatory pain, determination of their functional location to modulate this pathophysiological process is of prime interest. C-LTMRs are specialized subpopulations of cutaneous afferents that modulate mechanical and chemical pain hypersensitivity and consist of two third Cav3.2 channels containing primary afferent neurons lumbar DRG in mice (François et al., 2015). We used Cav3.2Nav1.8 cKO mice to delete Cav3.2 channels solely in C-LTMRs. Strikingly, this specific tool uncovers that pain-like symptoms were absent in Cav3.2Nav1.8 cKO mice submitted to the two inflammatory model studied, corroborating previous finding on formalin pain (François et al., 2015). The same observation was also reported in neuropathic (François et al., 2015) and visceral (Picard et al., 2019) pain murine models further supporting that Cav3.2 channels in C-LTMR are essential to build up nociceptive symptoms. In the present study, the reduction of pain-like inflammatory symptoms after treatment with systemic, intrathecal or intraplantar administration of ABT-639 indicated that Cav3.2 expressed at multiple subcellular levels from the peripheral to the central terminal are involved in the antihyperalgesic action induced by Cav3.2 inhibition. The study of Jarviset al . (2014) also evidenced the analgesic effect of ABT-639 in different murine models of neuropathic pain. However, a clinical trial with ABT-639 in patients with diabetic neuropathy failed to demonstrate any efficacy of the antagonist, whose tolerability was nevertheless good; the low doses used were considered as a possible explanation of this negative result (Ziegler et al., 2015). Clinical evaluations of local intradermal injection of TTA-A2 in a muscle pain model in healthy volunteers and patients (three patients with chronic pain) showed a decrease in mechanical and cold allodynia with no adverse effect (Samour et al., 2015).
In some conditions, inflammation can be controlled by neuronal mediators in a process called neurogenic inflammation (Xanthos and Sandkühler, 2014). To determine the involvement of Cav3.2 channels in this process, two strategies were used. With regard to neuronal Cav3.2 channels, we observed no change in edema size in Cav3.2Nav1.8 cKO mice, which strongly suggests that Cav3.2 channels expressed on C-LTMRs are not involved in subacute or maintained inflammation. Experiments using chimeric mice enabled us to demonstrate for the first time that the absence of Cav3.2 channels in hematopoietic cells was sufficient to reduce edema development and pro-inflammatory mediator release. Interestingly, this reduction was close to that observed in constitutive Cav3.2 KO mice, suggesting that Cav3.2 channels expressed by hematopoietic cells were involved in the control of inflammation. Accordingly, transplantation of hematopoietic cells expressing Cav3.2 channels into constitutive Cav3.2 KO animals completely restored edema development and pro-inflammatory mediator release. Thus, these in vivo experiments provide evidence that inflammation induced by carrageenan and CFA involved Cav3.2 channels expressed in hematopoietic cells. We then explored more specifically the potential contribution of BMDM and CD4+ T cells, two actors that play a crucial role in the inflammatory diseases (Weyand and Goronzy, 2021). Using immunocytochemistry, we demonstrated, for the first time, that these cells expressed Cav3.2 channel protein consistent with a previous demonstration of low level of Cav3.2 transcript expression in murine CD4+ T cells (Jarvis et al., 2014). Genetic deletion of Cav3.2 channels in BMDM showed no morphological signs of activation and lower production of pro-inflammatory mediators (IL-6 and TNF-α) in response to LPS stimulation. This result accounts for the fact that Cav3.2 channels are involved in exocytose not only in excitable cells but also in non-excitable cell (Carbone et al., 2006). This could be related to a reduced calcium recruitment in LPS-stimulated BMDM after Cav3.2 deletion, which lead to a reduced level of pro-inflammatory mediators. A relationship between the calcium signaling pathway and the production of pro-inflammatory mediators has already been established in LPS-stimulated rat peritoneal macrophages (Zhou et al., 2006). Cav3.2 channel deletion was also associated with a lower activation status in spleen APC suggesting an impaired ability of these cells to promote T cell activation. Moreover, Cav3.2 gene invalidation strongly affected T-cell proliferation upon CD3/CD28 stimulation, suggesting the involvement of Cav3.2 channels in early events of lymphocyte activation. In conclusion, Cav3.2 gene deletion strongly impairs the function of APC and CD4+T cells, a process that could explain the involvement of Cav3.2 in edema and inflammatory process observed in our models.
Our study has certain limitations. First, although the spinal and peripheral location of Cav3.2 channels was shown to be involved in inflammatory pain, a supraspinal contribution cannot be ruled out. Indeed, Cav3.2 channels are expressed in the brain (Bernal Sierra et al., 2017) and intracerebroventricular injection of TTA-A2 induced antinociception in the formalin test in mice (Kerckhove et al., 2014). However, the fact that ablation of Cav3.2 channels in C-LTMRs totally suppressed pain-like symptoms in our inflammatory models detracts from this hypothesis. In addition, we cannot definitively conclude that the reduced activation and proliferation of macrophages and CD4+ T cells that were proposed to explain the anti-edematous effect of Cav3.2 inhibition are the only processes involved. It is possible that a default in the fetal development of the hematopoietic lineage in Cav3.2 KO mice or in the recruitment or infiltration process in WT mice also contributes to the observed phenotype. However, after analysis, we observed no significant alteration in the number of the major populations of immune cells in Cav3.2 KO mice. Furthermore, other subpopulations of cells involved in inflammation such as monocytes and osteoblasts could account for the action of Cav3.2 channels inhibition. Elucidation of the mechanistic basis of these different questions will require additional experimental approaches, possibly including tissue‐specific manipulation of the expression of Cav3.2 channels.
In a clinical perspective, together with reports from the literature, our study supports the need of a more thorough clinical evaluation of T-type channels blockers in the treatment of chronic pain, in order to draw firm conclusions on their potential efficacy, especially in patients with inflammatory pain. The findings of our study would permit a rapid translation in patients by performing a clinical study in patients suffering from inflammatory pain.