Calcitonin Gene-related Peptide Attenuated Discogenic Low Back Pain by Inhibiting Microglia Activation Through the NLRP3/caspase-1 Signaling Pathway

Discogenic low back pain (DLBP) is a multifactorial disease, associated with intervertebral disc degeneration. Calcitonin gene-related protein (CGRP) plays a critical role in pain processing, while the role in DLBP remains unclear. This study aims to investigate the anti-nociceptive role and related mechanisms of CGRP in DLBP. Here we established the DLBP rat and validated the model using histology and radiography. CGRP and minocycline were intrathecally injected respectively and the behavioral test was performed to determine hyperalgesia. Further, BV2 microglial cells and microglial activation agent lipopolysaccharide (LPS) were employed for the in vitro experiment. We observed obvious lumbar intervertebral disc degeneration and hyperalgesia at 12 weeks postoperation in DLBP group, with signicantly activated microglia in the spinal cord. CGRP treatment signicantly inhibited the upregulation of proinammatory cytokines and NLRP3/caspase-1 expression induced by LPS in BV2 cells, whereas treatment with CGRP alone had little effect on BV2 cells. The intrathecal injection of CGRP into DLBP rats relieved mechanical and thermal hyperalgesia, reverted the microglial activation and decreased the expression of NLRP3/caspase-1, similar to the effects produced by minocycline. Our results provide evidence that microglial activation in the spinal cord play a key role in hyperalgesia in DLBP rats. CGRP effectively alleviates hyperalgesia and inhibits microglial activation in the spinal cord. Regulation of CGRP and microglial activation may provide a new strategy for ameliorating for DLBP.


Introduction
Low back pain (LBP) is a multifactorial disease that involves physical and psychological factors and brain changes. It was reported that 60%-80% of the population experiences at least one episode of LBP [1]. Intervertebral disc degeneration (IDD) is considered a signi cant cause of LBP [2]. Discogenic LBP (DLBP) has been used to describe LBP associated with IDD and without herniation or other anatomical deformity. To date, there is still no effective method to treat DLBP.
The upregulated expression of neuropeptides, such as calcitonin gene-related protein (CGRP), and in ammatory cytokines in degenerative disc, has been implicated as a key mechanism of DLBP [3].
However, after patients with DLBP underwent total disc replacement, the pain was relieved but not eliminated [4]. Thus, it was suggested that, in addition to the degenerative disc, there are other factors that contribute to DLBP. It's well known that microglia play a major role in the development and maintenance of neuropathic pain, through central sensitization and long-term potentiation in the spinal nociceptive responses [5][6][7][8]. Miyagi et al [9] observed increased microglia in the spinal dorsal horn in the LBP rats induced by disk injury. These ndings suggested that microglia activation in the spinal cord is closely related to DLBP.
Neuropeptide CGRP involved in both modulation and transmission of pain is widely distributed in both the peripheral nervous system (PNS) and the central nervous system (CNS) [10]. CGRP has been implicated in both pro-and antinociceptive effects in CNS. As reported in a recent study, there were different changes in the levels of CGRP in the spinal cord in different models of neuropathic pain [11].
Studies have shown that CGRP induces an antinociceptive effect in several brain areas [10,12]. whereas the effects of CGRP in spinal cord remains uncertain. It is noteworthy that CGRP was demonstrated to inhibit the microglia activation induced by lipopolysaccharide (LPS) in vitro [13].
Recently it was reported that the degenerative disc might induce the neuroin ammatory microenvironment in the spinal cord and drive activation of microglia [14]. The neuroin ammatory markers released by the activated microglia in the spinal cord, like the interleukin (IL)-1β, could in turn enhance the in ammatory milieu and have been implicated as key mediators of pain. The NOD-like receptor protein 3 (NLRP3) in ammasome has been found to trigger the activation of caspase-1 and induce the maturation of IL-1β [15], which is crucial for the regulation of microglia activation [16,17]. Evidence suggest that the activation of NLRP3/caspase-1/IL-1β pathway in intervertebral disc is involved in the degenerative process and low back pain [18][19][20]. It is still unknown whether the NLRP3/caspase-1/IL-1β pathway in the spinal cord is related to the DLBP. CGRP could reduce the NLRP3 and IL-1β protein expression induced by LPS in murine macrophages [21]. These ndings prompted us to investigate whether CGRP in the spinal cord regulate the microglial NLRP3/caspase-1/IL-1β pathway and play a role in the DLBP.
In this study, the DLBP rat model was established through unbalanced dynamic and static forces. And the intrathecal infusions of the microglial inhibitor minocycline or CGRP were applied to the DLBP rats to investigate the role of microglia and CGRP in DLBP. We hypothesized that CGRP would suppress the microglial activation by inhibiting NLRP3/caspase-1/IL-1β pathway and thereby play an antinociceptive role in DLBP. This study aimed to examine the potential therapeutic effects and underlying mechanism of CGRP in the spinal cord on DLBP.

In vivo experimental designs and animal groups
The surgical procedures were performed on male 8-week-old Sprague-Dawley rats (200-250g). All rats were housed under standard conditions with a 12-h light to dark cycle and ad libitum access to food and water. Ethical approval for performing animal experiments was obtained from the Institutional Animal Care and Use Committee of Shanghai Jiao Tong University, and the experiments were conducted in conformity with the Guide for the Care and Use of Laboratory Animals of the National Institutes of Health.
As shown in Fig. 1, there were two part of animal experiments in this study. Experiment 1: Changes in pain threshold, CGRP expression and microglial activation in DLBP rats Twelve rats were randomly divided into two groups: a sham operation (SHAM) group and a DLBP group. Rats in the DLBP group were subjected to a surgical procedure [22] that could induce unbalanced dynamic and static forces. The paraspinal musculature and ligaments, along with part of the facet joints, were excised to destroy the static and dynamic lumbar stabilizers. For rats in the SHAM group, a posterior skin incision was made. According to previous studies [22,23], the lumbar IDD induced by unbalanced dynamic and static forces was signi cant in the rat from 6 weeks after surgery. Thus the assessment of the pain threshold was done 1 day before surgery (baseline) and at 4, 6, 8, 10 and 12 weeks after surgery. Subsequently, the rats were examined by X-ray radiography (KXO-32R, Toshiba, Japan) and 3.0-Tesla MRI (Siemens, Erlangen, Germany) to assess the IDD. After the radiological examination, all animals were euthanized by an overdose pentobarbital injection. The L5/6 IVDs and the L3-L5 spinal cord segments were harvested for western blot, immunohistochemical and immuno uorescent analysis. Experiment 2: The effects of CGRP and minocycline treatment on hyperalgesia, microglial activation, and the NLRP3/caspase-1 after DLBP A total of 24 rats were randomly divided into four groups: (1) SHAM group; (2) DLBP+vehicle group; (3) DLBP+minocycline group; (4) DLBP+CGRP group. Our preliminary study (Experiment 1) showed that the allodynia and the lumbar IDD were both signi cant at 12 weeks after surgery. Therefore, the DLBP rats were intrathecally treated with drugs or an equivalent volume of saline once daily for 7 days at 12 weeks after the operation. Drugs were diluted with saline and intrathecally injected through the implanted catheter in a 10µl volume of solution followed by 10µl of saline for ushing. In the DLBP+vehicle group, rats were administrated intrathecally with saline. In the DLBP+minocycline group, rats were administrated intrathecally with microglia inhibitor minocycline (100µg/10µl, Selleck Chemicals, USA). And the rats in the DLBP+CGRP group were administrated intrathecally with CGRP (10µg/10µl, Tocris Bioscience, Bristol, UK). Assessment of the pain threshold was done at 1 day before and 1, 4, 7, 14, 21 and 28 days after intrathecal treatment. Then the L5/6 IVDs and the L3-L5 spinal cord segments were harvested for western blot, immunohistochemical and immuno uorescent analysis.

Intrathecal catheter implantation
The rats received Intrathecal catheterization 5 days prior to the drug administration. Brie y, rats were anesthetized with an intraperitoneal injection of pentobarbital (40mg/kg). A polyethylene catheter (PE-10; inner diameter, 0.28 mm; outer diameter, 0.61 mm; Smiths Medical International Ltd., UK) was inserted from L5-L6 spinous processes to the lumbar enlargement, and then the outer part of the catheter was xed onto the skin upon wound closure. An intrathecal injection of 2% lidocaine (10µl) was performed to con rm the success of catheterization.

Behavioral analysis
The mechanical pain threshold was assessed using von Frey laments [24] [25]. Rapidly withdrawal, shrinking or inching of the hind paw was regarded as a positive response. The minimum force required to induce positive responses in at least 3 out of 5 applications was de ned as the paw withdrawal threshold (PWT).
The thermal pain threshold was assessed using the IITC Life Science Inc (Woodland Hills, CA, USA) [26]. Rats were placed on a glass plate in the Plexiglas box, and a beam of radiant heat was applied to the plantar surface of the hind paw. The cutoff of 20 s was set to avoid thermal injury. The time from the commencement of heating to paw withdrawal was recorded. Each rat was measured three times at intervals of 10 min and the paw withdrawal latency (PWL) was calculated as the average of the three repeated measurements.

Histological examination
The L5/6 IVDs were decalci ed in 0.05% EDTA for 4 weeks and embedded in para n. The samples were cut into 4-µm sections in the sagittal plane and stained with hematoxylin and eosin (H&E). The slides were observed using a digital microscope (Olympus, Japan), and the images obtained were used to assess disc degeneration.

Immunohistochemistry
The immunohistochemical observations were assessed according to a previously reported method [27].
Brie y, after overnight incubation at 4°C with the primary antibody, the sections were incubated with a biotinylated secondary antibody for 15 min at 37°C. The primary antibody included anti-NLRP3 antibody

Statistical analysis
Data were analyzed using SPSS 23.0 software (SPSS Inc., Chicago, IL, USA). All data were presented as the means ± standard deviation. The statistical signi cance between two groups was evaluated by the Student's t-test. One-way or two-way analysis of variance (ANOVA) followed by Tukey's post hoc test was used for the analysis of the comparisons among multiple groups. The signi cance level was set at P < 0.05.

Results
Veri cation of the DLBP rat model At 12 weeks postoperatively, the lumbar IDD was assessed through histology and radiography. As shown in Fig. 2A, the disc of DLBP group showed less nucleus pulposus volume, moderate serpentine pattern in annulus brosus and sclerotic in cartilage endplate, where part of cartilage was replaced with new bone formation. Lateral radiographs of the DLBP group showed degenerative changes, including disc height loss and disc space narrowing (Fig. 2B). T2-weighted MRI images displayed a substantial reduction in the signal intensity of the nucleus pulposus in the DLBP group at 12 weeks after surgery (Fig. 2C).The disc height index (DHI) [28] was calculated from lateral radiographs as the ratio of the average anterior, middle, and posterior disc height to the average of the adjacent vertebral body heights. As shown in Fig.  2D, the DHI in the DLBP group was signi cantly decreased when compared with that in SHAM group. The results of the behavioral assessment for mechanical and thermal sensitivity are presented in Fig. 2E. The mechanical PWT and thermal PWL were both decreased from weeks 4 to 12 in the DLBP group, compared with baseline. And compared to SHAM rats, the DLBP rats exhibited a decreased PWT from weeks 6 to 12, and a decreased PWL from weeks 4 to 12. These results indicated that the DLBP model was successfully induced by unbalanced dynamic and static force surgery.
Microglia activation and changed expression of CGRP and NLRP3 in spinal cord of DLBP rats Microglial activation in the spinal dorsal horn was investigated by immuno uorescent labeling of IBA-1.
The expressions of CGRP and NLRP3 in the spinal cord were examined by immunohistochemical analysis. As shown in Fig. 2F, the number of Iba1-positive cells in the spinal cord were signi cantly increased in DLBP rats than that in SHAM rats. The immunohistochemical analysis showed that DLBP rats exhibited reduced expression of CGRP and increased expression of NLRP3 in the spinal cord (Fig. 3A, B).

CGRP treatment in vitro inhibited activation of microglial cells and NLRP3/caspase-1 induced by LPS
LPS is a widely used proin ammatory agent and is known to activate microglia. To investigate whether CGRP can suppress microglial activation in LPS-induced BV2 cells, we examined the expression levels of proin ammatory cytokines using western blotting ( Fig. 4A-C) and Elisa assay (Fig. 4D-F), including IL-1β, IL-6, and TNF-α. The results showed that LPS triggered microglial activation and upregulated the expression of IL-1β, IL-6, and TNF-α. Following CGRP treatment, there was a prominent reduction in the levels of these proin ammatory cytokines. As shown in Fig. 4G, CGRP treatment signi cantly suppressed the upregulated protein expression of NLRP3 and caspase-1 induced by LPS. Intriguingly, CGRP alone showed little effect on the protein levels of proin ammatory cytokines and NLRP3/caspase-1. These ndings suggested that CGRP can inhibit LPS-induced microglial and NLRP3/caspase-1 pathway activation.
CGRP treatment in vivo relieved hyperalgesia and inhibited activation of microglia and NLRP3/caspase-1 in DLBP rats, similar to minocycline This part of experiment was designed to focus on the effects and mechanism of CGRP on the hyperalgesia in vivo. To investigate the role of microglia activation in this process, we employed the minocycline, an inhibitor of microglial activation, as a control. CGRP or minocycline were intrathecally injected once daily for 7 days at 12 weeks after surgery. As shown in Fig. 5A-B, PWT in DLBP rats was signi cantly increased after intrathecal injection of minocycline from day 4, and increased after intrathecal injection of CGRP from day 7. PWL in DLBP rats was signi cantly increased after intrathecal injection of minocycline or CGRP from day 1. After stopping the CGRP or minocycline administration from day 7, the PWT and PWL gradually decreased over time. Importantly, immuno uorescence showed that compared with the DLBP+vehicle group, the expression of IBA-1 in spinal cord was signi cantly decreased in the DLBP+CGRP and DLBP+minocycline group (Fig. 5C). The NLRP3/caspase-1 pathway was examined as well. As shown in Fig. 5D, the expression of NLRP3 in spinal cord was signi cantly decreased in the DLBP+CGRP and DLBP+minocycline group, when compared with DLBP+vehicle group. And western blot analysis showed that the expression of caspase-1 and IL-1β were signi cantly downregulated in the spinal cord after CGRP or minocycline treatment (Fig. 5E). These results demonstrated that CGRP and minocycline could inhibit microglial activation and NLRP3/caspase-1 pathway in the spinal cord, and relieved hyperalgesia in DLBP rats.

Discussion
To date, the mechanisms underlying chronic DLBP are not fully understood. In this study, we investigated the role of CGRP in DLBP induced by unbalanced dynamic and static forces. The current study demonstrated three novel ndings using a DLBP rat model. First, rats with unbalanced dynamic and static force presented signi cant hyperalgesia and IDD, which are well-suited as a model for DLBP study.
Second, microglia were found to be activated in the spinal cord of DLBP rats, with decreased expression of CGRP and increased expression of NLRP3. Third, CGRP relieved mechanical and thermal hyperalgesia in DLBP rats, reverted the microglial activation and decreased the expression of NLRP3/caspase-1 in spinal cord, similar to the effects produced by minocycline. Our results suggested that CGRP played an antinociceptive role in DLBP via the inhibition of microglial activation in the spinal cord.
To investigate the mechanisms of DLBP, we rst employed the unbalanced dynamic and static force rat model, which was usually used in the IDD researches [22]. Compared with other methods, such as disc puncturing, the surgery we used induces IDD via biomechanical disturbance instead of directly injuring the intervertebral disc. The radiological and histological examinations both identi ed the degeneration of the lumbar intervertebral disc in rats at 12 weeks postoperation. On the other hand, the rats received surgery exhibited signi cant mechanical and thermal hyperalgesia. These ndings con rmed that the DLBP was successfully induced by unbalanced dynamic and static force surgery.
Microglia are the resident immune cells of the CNS, which can be activated by stimuli that threaten physiological homeostasis [8]. Activated microglia release various nociceptive mediators including proin ammatory cytokines, such as IL-1β, TNF-α and IL-6. The pro-in ammatory cytokines interact with nociceptive neurons and modulate central sensitization and hyperalgesia [29,30]. Within the activated NLRP3 in ammasome complex, the pro-IL-1β is cleaved by active caspase-1 to be secreted in the active form thereby amplifying the in ammatory response [31,32]. He et al[16] demonstrated that blockade of microglial NLRP3/IL-1β could improve hyperalgesia in migraine model, and inhibit the increase in biomarkers related to central sensitization. The activation of microglial NLRP3/IL-1β has been found to account for the long-term morphine treatment induced analgesic tolerance and hyperalgesia [33]. The NLRP3 −/− mice showed a higher nociceptive threshold and decreased analgesic tolerance induced by morphine [34]. These nding indicated that the microglial NLRP3/IL-1β plays critical roles in the induction of hyperalgesia. In the present study, the microglia in spinal cord were found signi cantly activated in DLBP rats, with the increase of NLRP3, caspase-1 and IL-1β protein expression. Our results suggested a potential role of microglial NLRP3/caspase-1/IL-1β pathway in hyperalgesia in the DLBP.
CGRP is an important neuropeptide that exerts complicated effects in pain modulation and transmission in the PNS and CNS [10]. CGRP expressed in DRG has always been reported to exert pronociceptive effects. Masashi et al. [35] reported that CGRP-IR in DRG neurons increased following disc injury, and the upregulation of CGRP might induce pain. However, in many brain areas, including the nucleus accumbens, CGRP induced an antinociceptive effect [10]. Therefore, the mechanisms underlying regulation of pain by CGRP remain controversial. Researches have shown that CGRP mediated protective effects in in ammation such as allergic airway in ammation [36] and lung injury [37]. Duan et al [21] reported that CGRP inhibited the LPS-induced activation of macrophages by reducing the expression of NLRP3 and IL-1β. And results from a previous study suggested that CGRP exerted a potent inhibitory effect on LPS-induced microglia activation in vitro, and the underlying mechanisms remain unknown [13]. In the present study, we observed the decreased expression of CGRP in the spinal cord of DLBP rats. To further explore the role of CGRP in the microglial activation and the underlying mechanism, we performed in vitro experiments with BV2 microglia cells. We found that CGRP treatment signi cantly inhibited LPSinduced microglia activation in vitro, which is consistent with the previous study [13]. Further, the increase in the protein expression of NLRP3 and caspase-1 induced by LPS was reversed by CGRP as well. Our ndings provide evidence of an inhibitory effect of CGRP on LPS-mediated microglia activation and NLRP3/caspase-1/IL-1β pathway in vitro.
For the validation of the in vitro results, the in vivo study in DLBP rats was designed in the present study. Minocycline is a common microglial inhibitor with strong analgesic and anti-in ammatory properties under a variety of pain models, such as neuropathic pain [38], in ammatory pain [39], and bone cancer pain [40]. Lu et al [41] reported that the pretreatment of minocycline signi cantly inhibited NLRP3 in ammasome activation in BV2 microglial cells. In the current study, the minocycline and CGRP were intrathecally injected into DLBP rats, respectively. Consistent with the other pain models, in the DLBP model, the hyperalgesia was signi cantly alleviated following the administration of minocycline. Additionally, as expected, the minocycline treatment inhibited microglial activation in the spinal cord, and downregulated the expression levels of NLRP3, caspase-1 and IL-1β. Similarly, the intrathecal injection of CGRP alone signi cantly attenuated nociceptive behavior in the DLBP rats. The inhibitory effect on microglial activation and NLRP3/caspase-1 /IL-1β signaling in the spinal cord has also been observed following the administration of CGRP. Notably, the increase in the pain thresholds mediated by CGRP was not as pronounced as for minocycline, but it was still higher than that of the control. In vivo pain modulation is rather complex and often involves multiple co-regulators such as glia, neurons and neuropeptides. Our results suggest that the ability of minocycline and CGRP to alleviate hyperalgesia in DLBP rats may be ascribed, at least in part, to an inhibition in microglial activation and NLRP3/caspase-1/IL-1β signaling in the spinal cord. Yet, further studies are warranted to determine whether other factors like astroglia are involved in this process.
Several limitations to this study need to be acknowledged. First, we examined a wide range of time points for the behavioral tests, while the immunohistochemical and Immuno uorescence analysis was only performed at the last time point during in vivo experiments. Thus the changes of the microglia in the spinal cord could not be observed dynamically and matched with the pain thresholds in this study.
Second, we did not investigate the effect of NLRP3 signaling on the analgesic effect of minocycline and CGRP. Further studies through speci cally regulating the NLRP3 signaling would be warranted. In spite of its limitations, the study certainly adds to our understanding of the DLBP pathogenesis as well as the treatment of this disease.

Conclusions
In summary, our results provided evidence that microglial activation in the spinal cord play a key role in hyperalgesia in DLBP rats. CGRP effectively alleviates hyperalgesia, inhibit microglial activation and NLRP3/caspase-1 pathway in the spinal cord, similar to the effects produced by minocycline. Our study provides several new insights for the development of novel therapeutic strategies for DLBP.

Data Availability
The datasets used and analyses during the current study are available from the corresponding author on reasonable request. The authors declare that they have no competing interests.

Figure 2
Veri cation of the DLBP rat model. A H&E staining showed less nucleus pulposus volume, moderate serpentine pattern in annulus brosus and new bone formation in cartilage endplate in DLBP group (×40). B X-ray images of the rat lumbar spine at 12 weeks after surgery. DLBP group showed disc height loss and disc space narrowing (arrow head). C T2-weighted MRI images of the spine at 12 weeks after surgery. The arrow indicates the "black disc". D Quantitative analysis of the disc height index between groups. E Behavioral analysis for mechanical and thermal pain threshold. F Representative images of immuno uorescence staining with IBA-1 in spinal cord from rats (×100; ×200). The data are expressed as the means ± SD (n = 6 in each group). *P<0.05; **P<0.01; ***P<0.001; ****P<0.0001 versus the SHAM group.

Figure 3
Expression of CGRP and NLRP3 in spinal cord. A Immunohistochemical staining for CGRP in spinal cord (×50; ×100; ×200) and quantitative analysis. B Immunohistochemical staining for NLRP3 in spinal cord (×50; ×100; ×200) and quantitative analysis. The data are expressed as the means ± SD (n = 6 in each group). ****P<0.0001 versus the SHAM group.  CGRP treatment in vivo relieved hyperalgesia and inhibited activation of microglia and NLRP3/caspase-1 in DLBP rats, similar to minocycline A-B Behavioral analysis for mechanical and thermal pain threshold was performed at days 0, 1, 4, 7, 14, 21 and 28 after intrathecal injection. C Representative images of immuno uorescence staining with IBA-1 in spinal cord from rats (×200). D Representative images of immunohistochemical staining with NLRP3 in spinal cord from rats (×200). E Representative blots and quanti cation of caspase-1 and IL-1β in spinal cord from rats. The data are expressed as the means ± SD