Discussion
The involvement of adaptive immunity, particularly spinal cord-infiltrating CD4+ T lymphocytes, in the development of nerve injury-induced behavioral hypersensitivity has been well-accepted [3,4]. Several pre-clinical studies have provided evidence, such as the passive transfer of individual T lymphocyte subtypes and the measurement of tissue levels of Th1 cell-producing cytokines [6,8], that suggests that the Th1 subtype of CD4+ T lymphocytes plays a role in peripheral nerve injury-induced sensory hypersensitivity. However, there has been no direct examination of peripheral nerve injury-induced infiltrating CD4+ T lymphocytes. In the current study, we analyzed L5Tx-induced lumbar spinal cord-infiltrating CD4+ T lymphocytes by intracellular flow cytometric analysis via an array of Th1 and Th2 CD4+ T lymphocyte cellular markers. Our results directly show an increase in Th1, but not Th2, CD4+ T lymphocytes in the lumbar spinal cord following L5Tx and suggest that Th1 CD4+ T lymphocytes mediate their pronociceptive effects via multiple cytokines (such as IFN-γ, TNF-α, and GM-CSF). Interestingly, although all of the Th1 cytokines measured have been associated with nerve injury-induced pain behaviors [8,23–25], some of them have also been associated with nerve regeneration and the recovery of nerve function following nerve injury [24–26]. Further investigation of such cytokine-mediated effects will help elucidate the specific role of infiltrating CD4+ T lymphocytes in the development of neuropathic pain, and such knowledge could help in the design of therapies that would target the cytokines/mechanisms responsible for the maintenance of neuropathic pain while avoiding eliminating factors that have neuroprotective effects.
Unlike what was previously reported [8], we did not detect a significant increase in IFN-γ expression in the lumbar spinal cord post-L5Tx. This may be due to the different animal species used (rats vs. mice), the different assays used (polymerase chain reaction vs. ELISA), and/or the existence of multiple cellular sources of IFN-γ (including infiltrating T lymphocytes and macrophages and resident microglia and astrocytes). Nevertheless, we did detect significantly lower levels of IFN-γ in CD4 KO mice compared to WT mice at all times evaluated (including baseline), which could in part explain our previous observation that CD4 KO mice displayed significantly reduced hypersensitivity post-L5Tx. However, the cytokine mediators through which CD4+ T lymphocytes specifically promote the pathogenesis of the maintenance phase but not the initiation phase of neuropathic pain still need further investigation. We suspect that multiple cytokines are likely to be involved in this mechanism and that the cytokines are likely to work in a synergistic manner.
To further identify potential CD4+ T lymphocyte-mediated downstream pathways, we examined the role of T lymphocyte-expressing CD154. CD40 is a cell surface receptor expressed by activated microglia [27–30]. CD40 ligand (CD154) is a surface protein primarily expressed by activated CD4+ T lymphocytes [21]. Although it is known that the CD154-CD40 interaction between Th1 cells and macrophages promotes macrophage responses [21] and that the interaction between microglial CD40 and CNS infiltrating-T lymphocyte CD154 has been linked to the pathogenesis of various CNS diseases [14–19], our data indicate that it is unlikely that lumbar spinal cord-infiltrating CD4+ T lymphocytes mediate their pro-nociceptive effects through CD154-mediated interactions with spinal cord microglial CD40. Similar to our observation, it has been reported that CD40, but not CD154, was required for mounting an optimal response against M tuberculosis infection, and this discrepancy was attributed to an interaction between CD40 and a non-CD154 CD40 ligand [31]. The discrepancy in the development of behavioral hypersensitivity between CD40 KO and CD154 KO mice suggest that such an alternative CD40 ligand may be involved in CD40-mediated pro-nociceptive effects. In addition, we previously did observe a reduction in the L5Tx-induced increase of microglia in the lumbar spinal cord in CD4 KO mice [32]. Thus, further studies are necessary to explore other factors involved in the interaction between CD4+ Th1 lymphocytes and microglia.
It is well-known that glial cells (including both astrocytes and microglia) play critical roles in the development of neuropathic pain [33–35]. Here, to further identify CD4+ T lymphocyte-mediated pathways, we also examined the lumbar spinal cord astrocytic response in CD4 KO mice via the analysis of GFAP expression. Our data suggest that the lack of CD4+ T lymphocytes is associated with a shorter-lasting L5Tx-induced increase of GFAP expression in the lumbar spinal cord. So far, no study has reported a potential interaction between infiltrating CD4+ T lymphocytes and astrocytes in the development of neuropathic pain. We suspect that Th1 CD4+ T lymphocytes could regulate astrocytes by both direct and indirect mechanisms. For instance, it is known that nitric oxide (NO) can induce GFAP expression [36] and that the Th1-related cytokines IFN-γ and TNF-α are known to be involved in NO production in the CNS [37]. Further, our previous observation of a reduction in the L5Tx-induced increase of microglia in the lumbar spinal cord in CD4 KO mice coupled with our GFAP results presented here suggest that CD4+ T lymphocytes may regulate astrocytic responses via microglia-mediated mechanisms [32]. Comprehensive in vitro and in vivo studies could be used to establish the interactions between Th1 CD4+ T lymphocytes and astrocytes in the future.
In summary, our data support that lumbar spinal cord infiltrating Th1, and not Th2, CD4+ T lymphocytes mediate the maintenance of peripheral nerve injury-induced neuropathic pain. These infiltrating Th1 CD4+ T lymphocytes are likely to mediate their effects in mechanisms involving multiple Th1 cytokines and glial activation. Further investigation of CD4+ T lymphocyte-mediated pathways may result in novel, more efficacious treatments for neuropathic pain.