In the present study, we confirmed the previous finding that chronic neuropathic pain induced by SNI challenge impaired cognitive function. Importantly, we showed that chronic neuropathic pain induced excessive GABAARs-α5 expressing on PV and SST interneurons, resulting in weakened inhibitory synaptic transmission and sequentially contributing to cognitive impairment in rodents (Fig. 8).
Chronic neuropathic pain was often caused by lesions or diseases of the somatosensory system, such as major surgeries or traumas, diabetic neuropathy, stroke and herpes zoster [20]. It is estimated that chronic neuropathic pain afflicts 7–10% of the population, which has tightly associated with cognitive impairment [21]. Recently, preclinical studies also proved memory deficits in the animal model of chronic neuropathic pain by using NOR and Y-maze tests [22]. In our study, the SNI model was used to induce chronic neuropathic pain. According to previous finding that right sciatic nerve injury showed more significant cognitive impairment than the left in rodents, we adopted the former method to establish SNI model. The results demonstrated that SNI rats with reduced long-lasting nociceptive threshold presented the deficits of recognition and working memory as evaluated by the NOR and Y-maze tests. Due to SNI surgery did not alter total moving distance of rats, these deficits were unlikely attributed to the injury in locomotor function caused by the surgery. As illustrated in the neurocognitive behavioral tests, rats appeared memory deficits in NOR and Y-maze tests on 21 rather than 7 and 14 days after surgery, which was most partly consistent with previous studies [22, 23]. Specifically, unlike our results, a previous study showed that SNI produced the working memory impairments on 14 days after surgery in Y-maze test [22]. The inconsistency in animal species (mice vs rats) or basal nociceptive response levels might account for these differences.
GABAARs-α5 are most concentrated in hippocampus, accounting for about 25% of all hippocampal GABAARs [24]. GABAARs-α5 are highly sensitive to GABA and show a slow desensitization rate, mainly mediating low concentration GABA-mediated tonic inhibition outside the synapses and inhibitory postsynaptic current in inhibitory postsynaptic membrane [25]. This specific distribution suggests a potential role of GABAARs-α5 in regulating the activity of neurons and hippocampal dependent cognition. In preclinical models, increased expression or activity of GABAARs-α5 contributed to cognitive impairment associated with many neuropsychiatric diseases such as traumatic brain injury, perioperative neurocognitive disorders and stroke, and inhibiting these receptors can mitigate the deficits [13, 26, 27]. Whereas, up to now, there has been little research focusing on whether GABAARs-α5 were involved in cognitive impairment under chronic neuropathic pain conditions. Our study showed that the expression of GABAARs-α5 in hippocampus were up-regulated under the existence of chronic neuropathic pain, and antagonizing GABAARs-α5 by L655,708 improved cognitive function, which consistent with previous reports [15, 28]. This finding suggested that GABAARs-α5 may play a crucial part in chronic neuropathic pain-related cognitive impairment. Additionally, previous studies have demonstrated that chronic pain increased GABAARs-α5 expression in the spinal cord and dorsal root ganglia, this receptor could be a valid pharmacological target to treat chronic pain states [29, 30]. Whether the up-regulated GABAARs-α5 on spinal level are associated with chronic neuropathic-mediated hippocampus memory impairment is unknown and remains to be investigated in future studies. The reverse agonist L655,708 used in this study was carefully selected to preferentially inhibit GABAARs-α5, which shows a preference for GABAARs-α5 receptors and occupy 60–70% of GABAARs-α5 at a dose of 1mg/kg [31]. Moreover, L655,708 did not cause significant off-target behavior effects, such as dyskinesia and sedation [31], which avoided its impact on the accuracy of behavioral results. Additionally, the chronic neuropathic pain caused by SNI surgery was not relieved by L655,708 injection, suggesting that L655,708 alleviated cognitive impairment by directly antagonizing increased GABAARs-α5 rather than relieving pain.
So far, changes in the inhibitory synaptic transmission in hippocampus linked with allodynia and neuropsychiatric aspects of chronic neuropathic pain have been poorly explored. Deficits in inhibitory synaptic transmission may account for underlying mechanism for impaired cognition [32], GABAergic interneurons are of great importance in regulating synaptic plasticity and synchronizing activity in the CA1 region of the hippocampus, both of which are essential in proper cognitive function [33, 34], PV and SST interneurons, two main subsets of inhibitory interneurons, have been reported to have a great impact in coding of neuronal information and the regulation of learning and memory [35, 36]. Interneuron-specific plasticity at PV and SST inhibitory synapses onto CA1 pyramidal neurons shapes hippocampal output [6]. On the one hand, previous studies had found that photogenetic activation of PV interneurons in SNI rats led to mechanical hypersensitivity which were migrated by suppression of PV interneurons [37]. In chronic inflammatory pain models, the decreased function of PV interneurons was observed [38]. In addition, it has been reported that SST interneurons were less active in SNI model [39]. On the other hand, a down-regulation or a disruption of PV interneurons in hippocampus would contribute to sepsis-induced cognitive impairment [40]. Suppressing the function of SST neurons reduces synchronized cellular and neural activity, and resulted in cognitive dysfunction [41]. Taken together, PV and SST interneurons are not only involved in chronic pain but also in cognitive regulation. Therefore, it is reasonable to deduce that interneurons might have a great effect on the development of chronic neuropathic pain-related cognitive impairment.
PV interneurons provide intensive rhythmic inhibition to pyramidal neurons by forming inhibitory synapses, embracing the soma and axon initial segment of pyramidal neurons [42], and SST interneurons regulate local synaptic and dendritic conductances by forming dendritic inhibition to pyramidal neurons [43]. GABAARs-α5 on nonpyramidal cells are the essential effectors controlling plasticity in learning and memory [44]. Our results showed that GABAARs-α5 loaded in PV and SST interneurons was increased in hippocampus of SNI rats, indicating that GABAARs-α5 coordinated SST and PV interneurons in hippocampus to disinhibit pyramidal neurons and damage cognition. These data demonstrated that GABAARs-α5 were essential in cognitive processes by modulating a component of synaptic transmission in hippocampal CA1 region. The accumulating evidence demonstrated that reduced inhibitory synaptic transmission induced by GABAergic interneurons dysfunction would bring about impaired information processing in the hippocampus [45, 46]. Recently, Magnin et al.[47] found that GABAARs-α5 were highly expressed in SST interneurons in stratum oriens/alveus layer and targeted by vasoactive intestinal peptide (VIP) and calretinin to form inhibitory synaptic connections and L655,708 could improve spatial memory in CA1-VIP-off mice. We observed that the sIPSCs mean frequency of pyramidal neurons significantly decreased in SNI rats which likely suggested a selective suppression of GABAergic inhibitory interneurons, resulting in failing to form powerful presynaptic inhibition projecting to pyramidal neurons. Consistent with previous study that the rate of sIPSCs decay was highly sensitive to modulators of the GABAARs [48], L655,708 administration in this study effectively rescued the decrease frequency of sIPSCs. What’s more, glutamate decarboxylase is a key enzyme in γ-aminobutyric acid (GABA) synthesis, in which GAD67 plays a major role. The down-regulated level of GAD67 suggested the deficits in GABAergic activity in this study, and may contribute to damaged cognitive processes. According to previous study, the attenuated inhibitory synaptic transmission may disrupting hippocampus-mediated memory function of everyday‐type memory performance, as assessed on the watermaze delayed‐matching‐to‐place task [49]. The selective inverse agonist at GABAARs-α5 enhanced watermaze delayed‐matching‐to‐place task performance and the induction of long‐term potentiation at hippocampal synapses. Thus, the weakened GABAergic synaptic transmission caused by upregulated GABAARs-α5 in hippocampus was likely the underlying mechanism for cognitive impairment after peripheral nerve injury.
In summary, our behavioral and electrophysiological evidence indicated GABAARs-α5 play a vital role in chronic neuropathic pain-related cognitive impairment, likely through suppressing inhibitory synaptic transmission. It could be a prospective molecular target for the treatment strategy of cognitive impairment under chronic neuropathic pain conditions in clinical practice.
Sources of funding
This work was supported by the National Natural Science Foundation of China (grant numbers: 82071196), the Jiangsu Commission of Health (grant numbers: z201949) and the Basic Research Grant of Southeast University.