In the present study, we investigated the effect of IL-1β on AMPA- and NMDA-induced currents in SG neurons by using whole-cell patch-clamp recordings. Our study shows the enhancement by IL-1β of both postsynaptic AMPA and NMDA receptor responses which has been reported previously. Moreover, our data reveal that these facilitatory effects of IL-1β on AMPA and NMDA receptors may reflect different mechanisms. Importantly, this is the first study to demonstrate that pretreatment with IL-1ra blocks the IL-1β-mediated potentiation of responses to exogenously-applied AMPA or NMDA. Our data further provide new and important evidence that microglia activation may be involved in the glutamate-receptor response potentiations produced by IL-1β.
The role of ionotropic glutamate receptors (mainly AMPA and NMDA receptors in SG neurons) as targets of analgesics has emerged during the last decade. It has been reported that the activation of NMDA receptors critically contributes to the development of chronic nociceptive hypersentivity following peripheral tissue damage or nerve injuries. Recently, some studies have also shown a contribution of spinal AMPA receptors in the development of both acute and chronic painful conditions[42, 43]. There is still an ongoing controversy whether IL-1β signaling is involved in pain transmission under normal, non-inflammatory states. When injected intrathecally into intact rats, IL-1β is reported to be without effect[20, 44]. However, our study clearly demonstrates that acute application of IL-1β rapidly alters the efficacy of extrasynaptic AMPA and NMDA receptors in SG neurons of spinal cord slices. This may explain the fact that intrathecal administration of IL-1β in healthy rats induces hyperalgesia and allodynia and enhances wind-up activity in dorsal horn neurons. Recently, several groups of SG neurons with distinct electrophysiological properties were characterized[46, 47]. Although we cannot be sure that the excitatory actions of IL-1β in our study are exerted on excitatory interneurons, it is generally believed that most neurons recorded in the SG are excitatory[48, 49]. In addition, IL-1β has been reported to depolarize paraventricular nucleus neurons or to hyperpolarize hypothalamic neurons, but this was not the case in SG neurons, because currents required to hold membrane potentials at −70 or −50 mV were unaffected by IL-1β. In accordance with data showing that IL-1β increases NMDA receptor-mediated rise in intracellular calcium levels in hippocampal neurons and enhances NMDA-induced current in cultured hippocampal or SG neurons, our data also demonstrated an enhancement of NMDA current by IL-1β. Although it was reported that AMPA-induced current was moderately enhanced by IL-1β (P > 0.05), our data demonstrated that it was significantly enhanced by IL-1β. This discrepancy may be due to a longer superfusion time of IL-1β in our studies and indicate that the effects of IL-1β on neurons depend on not only the type of neurons examined but also the duration of time that the neuron is exposed to it. In particular, our results confirm that IL-1β exhibited a persistent facilitatory effect on postsynaptic NMDA but not AMPA receptor function after removal of this cytokine. Thus, IL-1β may activate a postsynaptic process linked to long-term potentiation (LTP) through NMDA receptors which might have functional relevance for hyperalgesia or allodynia. This supports the report that IL-1β produced a thermal hyperalgesia lasting for 24 h after its intrathecal injection.
IL-1RI, which is known to mediate all biological functions of IL-1β, was detected in most of large and small DRG neurons as well as in some glia-like cell types by in situ hybridization staining. Another double immunofluorescence labeling study showed that IL-1RI and NMDA receptor NR1 subunit are co-localized in the spinal dorsal horn. Consistently, acute or chronic administration of IL-1ra, the endogenous antagonist of the IL-1R, which competitively blocks the binding of IL-1β to the receptor without inducing a signal of its own, inhibits a hypernociception induced by IL-1β[16, 19, 55–57]. Thus, inhibition of IL-1β signaling such as blockade of IL-1RI by IL-1ra in spinal dorsal horn neurons could account for its antinociceptive action and decrease the function of excitatory transmission in pain pathway. In agreement with these reports, our present data demonstrate that IL-1ra inhibits AMPA- and NMDA-induced current increases produced by IL-1β in all the neurons tested. Based on reports that NMDA receptors have an essential role in pain hypersensitization, our findings provide new and important evidence that inhibition by IL-1ra of NMDA receptor potentiation produced by IL-1β could prevent plasticity-related phenomena such as central sensitization or excitotoxicity. Actually, IL-1ra prevents the development of LTP, presumably by inhibiting NMDA receptor phosphorylation. An immunohistochemical study demonstrated that IL-1ra inhibited spinal cord phosphorylation of NR1 in a rat model of inflammatory pain. In another study, IL-1ra abolished NMDA-induced intracellular Ca2+ level increases produced by IL-1β, which may be involved in preventing tyrosine phosphorylation of NMDA receptor NR2A/B subunit. Compared to cultured or acutely-isolated neurons, neurons in spinal cord slices used in the present study have the advantage of offering recording under conditions closer to physiological ones. Our electrophysiological experiment data provide a cellular mechanism that IL-1RI may be a target for treatment of inflammatory and neuropathic pain.
Several cellular mechanisms could be involved in the enhancement of postsynaptic AMPA and NMDA receptor responses by IL-1β, e.g., presynaptic transmitter release, the modulation of the receptors in quantity (via a change in trafficking) and efficacy. It has been observed that IL-1β could increase the release of glutamate through an increase in Ca2+ influx in hippocampal neurons. The present study also demonstrates that the facilitatory effect of IL-1β on postsynatpic AMPA receptors was TTX-sensitive and Ca2+-dependent since AMPA-induced current increased by IL-1β was abolished in the presence of TTX or Ca2+-free Krebs solution. Consistent with the report that IL-1β increased the frequency and amplitude of spontaneous excitatory postsynaptic currents (sEPSCs) in SG neurons, our data indicated that the enhancement of AMPA responses may be generated by glutamate-receptor activation and neuronal activity increase, possibly by the facilitatory effect of IL-1β on intracellular Ca2+ level elevation. On the other hand, NMDA receptor response facilitation produced by IL-1β was resistant to TTX or Ca2+-free, indicating a direct action on postsynaptic NMDA receptor. The glutamate-mediated excitatory transmission efficiency is dependent on the number and function of AMPA or NMDA receptors at glutamatergic synapses. It is difficult from our data to know postsynaptic AMPA or NMDA receptor regulation such as receptor trafficking or subunit phosphorylation by IL-1β, but our findings are consistent with previous studies that the former is probably associated with a presynaptic mechanism of glutamate release increase while the latter is mediated by NMDA receptor subunit phosphorylation.
IL-1β is a cytokine released from spinal glial cells in response to pathophysiological changes that occur during different disease states, such as inflammatory and neuropathic pain. Initial reports suggested that IL-1β is an extremely potent hyperalgesic agent when injected systemically, intraperitoneally or intraplantarly in rats. These results may be in accordance with literatures reporting that activation of microglia and astrocyte may be involved in neuropathic pain[59, 60] and strongly suggest signaling via IL-1β between neuronal and glial cells might occur. Our results showed that in the presence of a microglia inhibitor minocycline, AMPA-induced current amplitude increase mediated by IL-1β was inhibited without any changes by minocycline itself. In contrast, NMDA-induced current facilitation produced by IL-1β was not affected by minocycline. Minocycline-sensitive microglia may be activated by neurotransmitters released as a result of an increase in neuronal activity, resulting in an enhancement of AMPA but not NMDA response. This issue remains to be further addressed. Furthermore, we reported, for the first time, that minocycline per se depressed NMDA-induced current. This inhibitory action may be due to an open-channel block by minocycline itself, as shown for GluR2 homomeric channels. Further research is needed to elucidate the molecular mechanism underlying this NMDA receptor inhibition by minocycline in SG neurons. Consistent with others’ data, our study indicates the realization that glia-derived signaling molecules such as IL-1β can contribute to and modulate pain signaling in the spinal cord.