Prediction of the position of external markers using a recurrent neural network trained with unbiased online recurrent optimization for safe lung cancer radiotherapy.

BACKGROUND AND OBJECTIVE: During lung cancer radiotherapy, the position of infrared reflective objects on the chest can be recorded to estimate the tumor location. However, radiotherapy systems have a latency inherent to robot control limitations that impedes the radiation delivery precision. Prediction with online learning of recurrent neural networks (RNN) allows for adaptation to non-stationary respiratory signals, but classical methods such as real-time recurrent learning (RTRL) and truncated backpropagation through time are respectively slow and biased. This study investigates the capabilities of unbiased online recurrent optimization (UORO) to forecast respiratory motion and enhance safety in lung radiotherapy. METHODS: We used nine observation records of the three-dimensional (3D) position of three external markers on the chest and abdomen of healthy individuals breathing during intervals from 73s to 222s. The sampling frequency was 10Hz, and the amplitudes of the recorded trajectories range from 6mm to 40mm in the superior-inferior direction. We forecast the 3D location of each marker simultaneously with a horizon value (the time interval in advance for which the prediction is made) between 0.1s and 2.0s, using an RNN trained with UORO. We compare its performance with an RNN trained with RTRL, least mean squares (LMS), and offline linear regression. We provide closed-form expressions for quantities involved in the gradient loss calculation in UORO, thereby making its implementation efficient. Training and cross-validation were performed during the first minute of each sequence. RESULTS: On average over the horizon values considered and the nine sequences, UORO achieves the lowest root-mean-square (RMS) error and maximum error among the compared algorithms. These errors are respectively equal to 1.3mm and 8.8mm, and the prediction time per time step was lower than 2.8ms (Dell Intel core i9-9900K 3.60 GHz). Linear regression has the lowest RMS error for the horizon values 0.1s and 0.2s, followed by LMS for horizon values between 0.3s and 0.5s, and UORO for horizon values greater than 0.6s. CONCLUSIONS: UORO can accurately predict the 3D position of external markers for intermediate to high response times with an acceptable time performance. This will help limit unwanted damage to healthy tissues caused by radiotherapy.

Fentanyl-induced hyperalgesia and analgesic tolerance in male rats: common underlying mechanisms and prevention by a polyamine deficient diet.

Opioids are a mainstay of pain management but can induce unwanted effects, including analgesic tolerance and paradoxical hyperalgesia, either of which leads to increased pain. Clinically, however, the relationship between these two phenomena remains elusive. By evaluating changes in mechanical nociceptive threshold in male rats, we found that in contrast to a purely analgesic control response to a single subcutaneous administration of fentanyl (25 µg/kg), in rats subjected to inflammatory pain 2 weeks previously (Day0), the same test dose (D13) induced a bi-phasic response: initial decreased analgesia (tolerance) followed by hyperalgesia lasting several hours. Both the tolerance and hyperalgesia were further enhanced in rats that had additionally received fentanyl on D0. The dose-response profiles (5 fg to 50 µg/kg) of pain- and opioid-experienced rats were very different from pain/drug-naive rats. At ultra-low fentanyl doses (<5 ng/kg and <500 ng/kg for naïve control and pain/drug-experienced rats, respectively), solely hyperalgesia was observed in all cases. At higher doses, which now produced analgesia alone in naive rats, reduced analgesia (tolerance) coupled with hyperalgesia occurred in pain/fentanyl-experienced rats, with both phases increasing with dose. Transcriptomic and pharmacological data revealed that an overactivation of the spinal N-methyl-D-aspartate receptor-inducible NO synthase cascade plays a critical role in both acute tolerance and hyperalgesia, and together with the finding that the magnitudes of analgesia and associated hyperalgesia are negatively correlated, is indicative of closely related phenomena. Finally, a polyamine deficient diet prevented inducible NO synthase transcript upregulation, restored fentanyl's analgesic efficacy and suppressed the emergence of hyperalgesia.

Prediction of the motion of chest internal points using a recurrent neural network trained with real-time recurrent learning for latency compensation in lung cancer radiotherapy.

During the radiotherapy treatment of patients with lung cancer, the radiation delivered to healthy tissue around the tumor needs to be minimized, which is difficult because of respiratory motion and the latency of linear accelerator (LINAC) systems. In the proposed study, we first use the Lucas-Kanade pyramidal optical flow algorithm to perform deformable image registration (DIR) of chest computed tomography (CT) scan images of four patients with lung cancer. We then track three internal points close to the lung tumor based on the previously computed deformation field and predict their position with a recurrent neural network (RNN) trained using real-time recurrent learning (RTRL) and gradient clipping. The breathing data is quite regular, sampled at approximately 2.5 Hz, and includes artificially added drift in the spine direction. The amplitude of the motion of the tracked points ranged from 12.0 mm to 22.7 mm. Finally, we propose a simple method for recovering and predicting three-dimensional (3D) tumor images from the tracked points and the initial tumor image, based on a linear correspondence model and the Nadaraya-Watson non-linear regression. The root-mean-square (RMS) error, maximum error and jitter corresponding to the RNN prediction on the test set were smaller than the same performance measures obtained with linear prediction and least mean squares (LMS). In particular, the maximum prediction error associated with the RNN, equal to 1.51 mm, is respectively 16.1% and 5.0% lower than the error given by a linear predictor and LMS. The average prediction time per time step with RTRL is equal to 119 ms, which is less than the 400 ms marker position sampling time. The tumor position in the predicted images appears visually correct, which is confirmed by the high mean cross-correlation between the original and predicted images, equal to 0.955. The standard deviation of the Gaussian kernel and the number of layers in the optical flow algorithm were the parameters having the most significant impact on registration performance. Their optimization led respectively to a 31.3% and 36.2% decrease in the registration error. Using only a single layer proved to be detrimental to the registration quality because tissue motion in the lower part of the lung has a high amplitude relative to the resolution of the CT scan images. The random initialization of the hidden units and the number of these hidden units were found to be the most important factors affecting the performance of the RNN. Increasing the number of hidden units from 15 to 250 led to a 56.3% decrease in the prediction error on the cross-validation data. Similarly, optimizing the standard deviation of the initial Gaussian distribution of the synaptic weights σinitRNN led to a 28.4% decrease in the prediction error on the cross-validation data, with the error minimized for σinitRNN=0.02 with the four patients.

Mycolactone displays anti-inflammatory effects on the nervous system.

BACKGROUND: Mycolactone is a macrolide produced by the skin pathogen Mycobacterium ulcerans, with cytotoxic, analgesic and immunomodulatory properties. The latter were recently shown to result from mycolactone blocking the Sec61-dependent production of pro-inflammatory mediators by immune cells. Here we investigated whether mycolactone similarly affects the inflammatory responses of the nervous cell subsets involved in pain perception, transmission and maintenance. We also investigated the effects of mycolactone on the neuroinflammation that is associated with chronic pain in vivo. METHODOLOGY/ PRINCIPLE FINDINGS: Sensory neurons, Schwann cells and microglia were isolated from mice for ex vivo assessment of mycolactone cytotoxicity and immunomodulatory activity by measuring the production of proalgesic cytokines and chemokines. In all cell types studied, prolonged (>48h) exposure to mycolactone induced significant cell death at concentrations >10 ng/ml. Within the first 24h treatment, nanomolar concentrations of mycolactone efficiently suppressed the cell production of pro-inflammatory mediators, without affecting their viability. Notably, mycolactone also prevented the pro-inflammatory polarization of cortical microglia. Since these cells critically contribute to neuroinflammation, we next tested if mycolactone impacts this pathogenic process in vivo. We used a rat model of neuropathic pain induced by chronic constriction of the sciatic nerve. Here, mycolactone was injected daily for 3 days in the spinal canal, to ensure its proper delivery to spinal cord. While this treatment failed to prevent injury-induced neuroinflammation, it decreased significantly the local production of inflammatory cytokines without inducing detectable cytotoxicity. CONCLUSION/ SIGNIFICANCE: The present study provides in vitro and in vivo evidence that mycolactone suppresses the inflammatory responses of sensory neurons, Schwann cells and microglia, without affecting the cell viability. Together with previous studies using peripheral blood leukocytes, our work implies that mycolactone-mediated analgesia may, at least partially, be explained by its anti-inflammatory properties.

Could an endoneurial endothelial crosstalk between Wnt/ß-catenin and Sonic Hedgehog pathways underlie the early disruption of the infra-orbital blood-nerve barrier following chronic constriction injury?

Background: Blood­nerve barrier disruption is pivotal in the development of neuroinflammation, peripheral sensitization, and neuropathic pain after peripheral nerve injury. Activation of toll-like receptor 4 and inactivation of Sonic Hedgehog signaling pathways within the endoneurial endothelial cells are key events, resulting in the infiltration of harmful molecules and immunocytes within the nerve parenchyma. However, we showed in a previous study that preemptive inactivation of toll-like receptor 4 signaling or sustained activation of Sonic Hedgehog signaling did not prevent the local alterations observed following peripheral nerve injury, suggesting the implication of another signaling pathway. Methods: Using a classical neuropathic pain model, the infraorbital nerve chronic constriction injury (IoN-CCI), we investigated the role of the Wnt/ß-catenin pathway in chronic constriction injury-mediated blood­nerve barrier disruption and in its interactions with the toll-like receptor 4 and Sonic Hedgehog pathways. In the IoN-CCI model versus control, mRNA expression levels and/or immunochemical detection of major Wnt/Sonic Hedgehog pathway (Frizzled-7, vascular endothelial-cadherin, Patched-1 and Gli-1) and/or tight junction proteins (Claudin-1, Claudin-5, and Occludin) readouts were assessed. Vascular permeability was assessed by sodium fluorescein extravasation. Results: IoN-CCI induced early alterations in the vascular endothelial-cadherin/ß-catenin/Frizzled-7 complex, shown to participate in local blood­nerve barrier disruption via a ß-catenin-dependent tight junction protein downregulation. Wnt pathway also mediated a crosstalk between toll-like receptor 4 and Sonic Hedgehog signaling within endoneurial endothelial cells. Nevertheless, preemptive inhibition of Wnt/ß-catenin signaling before IoN-CCI could not prevent the downregulation of key Sonic Hedgehog pathway readouts or the disruption of the infraorbital blood­nerve barrier, suggesting that Sonic Hedgehog pathway inhibition observed following IoN-CCI is an independent event responsible for blood­nerve barrier disruption. Conclusion: A crosstalk between Wnt/ß-catenin- and Sonic Hedgehog-mediated signaling pathways within endoneurial endothelial cells could mediate the chronic disruption of the blood­nerve barrier following IoN-CCI, resulting in increased irreversible endoneurial vascular permeability and neuropathic pain development.

Microglial Janus kinase/signal transduction and activator of transcription 3 pathway activity directly impacts astrocyte and spinal neuron characteristics.

After peripheral nerve injury microglial reactivity change in the spinal cord is associated with an early activation of Janus kinase (JAK)/STAT3 transduction pathway whose blockade attenuates local inflammation and pain hypersensitivity. However, the consequences of microglial JAK/STAT3-mediated signaling on neighboring cells are unknown. Using an in vitro paradigm we assessed the impact of microglial JAK/STAT3 activity on functional characteristics of astrocytes and spinal cord neurons. Purified rat primary microglia was stimulated with JAK/STAT3 classical activator interleukin-6 in the presence or absence of a selective STAT3 inhibitor and rat primary astrocytes or spinal cord neurons were exposed to microglia conditioned media (CM). JAK/STAT3 activity-generated microglial CM modulated both astrocyte and neuron characteristics. Beyond inducing mRNA expression changes in various targets of interest in astrocytes and neurons, microglia CM activated c-Jun N-terminal kinase, STAT3 and NF-κB intracellular pathways in astrocytes and promoted their proliferation. Without modifying neuronal excitability or survival, CM affected the nerve processes morphology and distribution of the post-synaptic density protein 95, a marker of glutamatergic synaptic contacts. These findings show that JAK/STAT3 activity in microglia impacts the functional characteristics of astrocytes and neurons. This suggests its participation in spinal cord tissue plasticity and remodeling occurring after peripheral nerve injury. We show that the activity of JAK/STAT3 pathway in microglial cells confers them a specific signaling modality toward neighboring cells, promoting astrocyte proliferation and changes in neuronal morphology. These in vitro data suggest that the early JAK/STAT3 activation in spinal cord microglia, associated with peripheral nerve injury, participates in functional alteration of various cell populations and in spinal tissue remodeling.

Early alterations of Hedgehog signaling pathway in vascular endothelial cells after peripheral nerve injury elicit blood-nerve barrier disruption, nerve inflammation, and neuropathic pain development.

Changes in the nerve's microenvironment and local inflammation resulting from peripheral nerve injury participate in nerve sensitization and neuropathic pain development. Taking part in these early changes, disruption of the blood-nerve barrier (BNB) allows for infiltration of immunocytes and promotes the neuroinflammation. However, molecular mechanisms engaged in vascular endothelial cells (VEC) dysfunction and BNB alterations remain unclear. In vivo, BNB permeability was assessed following chronic constriction injury (CCI) of the rat sciatic nerve (ScN) and differential expression of markers of VEC functional state, inflammation, and intracellular signaling was followed from 3 hours to 2 months postinjury. Several mechanisms potentially involved in functional alterations of VEC were evaluated in vitro using human VEC (hCMEC/D3), then confronted to in vivo physiopathological conditions. CCI of the ScN led to a rapid disruption of endoneurial vascular barrier that was correlated to a decreased production of endothelial tight-junction proteins and an early and sustained alteration of Hedgehog (Hh) signaling pathway. In vitro, activation of Toll-like receptor 4 in VEC downregulated the components of Hh pathway and altered the endothelial functional state. Inhibition of Hh signaling in the ScN of naive rats mimicked the biochemical and functional alterations observed after CCI and was, on its own, sufficient to evoke local neuroinflammation and sustained mechanical allodynia. Alteration of the Hh signaling pathway in VEC associated with peripheral nerve injury, is involved in BNB disruption and local inflammation, and could thus participate in the early changes leading to the peripheral nerve sensitization and, ultimately, neuropathic pain development.

Src family kinases involved in CXCL12-induced loss of acute morphine analgesia.

Functional interactions between the chemokine receptor CXCR4 and opioid receptors have been reported in the brain, leading to a decreased morphine analgesic activity. However the cellular mechanisms responsible for this loss of opioid analgesia are largely unknown. Here we examined whether Src family-kinases (SFK)-linked mechanisms induced by CXCR4 contributed to the loss of acute morphine analgesia and could represent a new physiological anti-opioid signaling pathway. In this way, we showed by immunohistochemistry and western blot that CXCL12 rapidly activated SFK phosphorylation in vitro in primary cultured lumbar rat dorsal root ganglia (DRG) but also in vivo in the DRG and the spinal cord. We showed that SFK activation occurred in a sub population of sensory neurons, in spinal microglia but also in spinal nerve terminals expressing mu-(MOR) and delta-opioid (DOR) receptor. In addition we described that CXCR4 is detected in MOR- and DOR-immunoreactive neurons in the DRG and spinal cord. In vivo, we demonstrated that an intrathecal administration of CXCL12 (1µg) significantly attenuated the subcutaneous morphine (4mg/kg) analgesia. Conversely, pretreatment with a potent CXCR4 antagonist (5µg) significantly enhanced morphine analgesia. Similar effects were obtained after an intrathecal injection of a specific SFK inhibitor, PP2 (10µg). Furthermore, PP2 abrogated CXCL12-induced decrease in morphine analgesia by suppressing SFK activation in the spinal cord. In conclusion, our data highlight that CXCL12-induced loss of acute morphine analgesia is linked to Src family kinases activation.

Gene-based approaches in pain research and exploration of new therapeutic targets and strategies.

Large panel of gene-based techniques is used for many years specifically in the pain research field. From the first identification (cloning) of some "mythic" genes, such as those encoding opioid or capsaicin receptors allowing then the creation of first-generation knockout mice, to the today conditional (time, tissue, cell-type and even pathology-dependent) and regulatable modulation of a gene function, these approaches largely contributed to fundamental leaps forward in our understanding of the function of some proteins and of their interest as possible druggable targets. Perhaps one of the most remarkable evolution in the last years is the passage of these approaches from the bench to the patient; whether it concerns the identification of genes involved in inherited pain insensibility/susceptibility, the search for genetic markers of pain types, the individual pharmacogenomics or even the first gene therapy trials. From many possible variants of gene-grounded techniques used in pain research we focus here on gene knockouts and some recent developments, on viral vectors-based gene transfer and on transgenic models for the tracing of pain pathways. Through these selected examples we attempted to emphasize the immense potential of these approaches and their already well-recognized contribution in both the basic and clinical pain research.

Cellular and subcellular localization of CXCL12 and CXCR4 in rat nociceptive structures: physiological relevance.

Initial studies implicated the chemokine CXC motif ligand 12 (CXCL12) and its cognate CXC motif receptor 4 (CXCR4) in pain modulation. However, there has been no description of the distribution, transport and axonal sorting of CXCL12 and CXCR4 in rat nociceptive structures, and their direct participation in nociception modulation has not been demonstrated. Here, we report that acute intrathecal administration of CXCL12 induced mechanical hypersensitivity in naive rats. This effect was prevented by a CXCR4-neutralizing antibody. To determine the morphological basis of this behavioural response, we used light and electron microscopic immunohistochemistry to map CXCL12- and CXCR4-immunoreactive elements in dorsal root ganglia, lumbar spinal cord, sciatic nerve and skin. Light microscopy analysis revealed CXCL12 and CXCR4 immunoreactivity in calcitonin gene related peptide-containing peptidergic primary sensory neurons, which were both conveyed to central and peripheral sensory nerve terminals. Electron microscopy clearly demonstrated CXCL12 and CXCR4 immunoreactivity in primary sensory nerve terminals in the dorsal horn; both were sorted into small clear vesicles and large dense-core vesicles. This suggests that CXCL12 and CXCR4 are trafficked from nerve cell bodies to the dorsal horn. Double immunogold labelling for CXCL12 and calcitonin gene related peptide revealed partial vesicular colocalization in axonal terminals. We report, for the first time, that CXCR4 receptors are mainly located on the neuronal plasma membrane, where they are present at pre-synaptic and post-synaptic sites of central terminals. Receptor inactivation experiments, behavioural studies and morphological analyses provide strong evidence that the CXCL12/CXCR4 system is involved in modulation of nociceptive signalling.

Antinociceptive effect of peripheral serotonin 5-HT2B receptor activation on neuropathic pain.

Serotonin is critically involved in neuropathic pain. However, its role is far from being understood owing to the number of cellular targets and receptor subtypes involved. In a rat model of neuropathic pain evoked by chronic constriction injury (CCI) of the sciatic nerve, we studied the role of 5-HT(2B) receptor in dorsal root ganglia (DRG) and the sciatic nerve. We showed that 5-HT(2B) receptor activation both prevents and reduces CCI-induced allodynia. Intrathecal administration of 5-HT(2B) receptor agonist BW723C86 significantly attenuated established mechanical and cold allodynia; this effect was prevented by co-injection of RS127445, a selective 5-HT(2B) receptor antagonist. A single application of BW723C86 on the sciatic nerve concomitantly to CCI dose-dependently prevented mechanical allodynia and significantly reduced cold allodynia 17 days after CCI. This behavioral effect was accompanied with a marked decrease in macrophage infiltration into the sciatic nerve and, in the DRG, with an attenuated abnormal expression of several markers associated with local neuroinflammation and neuropathic pain. CCI resulted in a marked upregulation of 5-HT(2B) receptor expression in sciatic nerve and DRG. In the latter structure, it was biphasic, consisting of a transient early increase (23-fold), 2 days after the surgery and before the neuropathic pain emergence, followed by a steady (5-fold) increase, that remained constant until pain disappeared. In DRG and sciatic nerve, 5-HT(2B) receptors were immunolocalized on sensory neurons and infiltrating macrophages. Our data reveal a relationship between serotonin, immunocytes, and neuropathic pain development, and demonstrate a critical role of 5-HT(2B) receptors in blood-derived macrophages.

NOV/CCN3 attenuates inflammatory pain through regulation of matrix metalloproteinases-2 and -9.

BACKGROUND: Sustained neuroinflammation strongly contributes to the pathogenesis of pain. The clinical challenge of chronic pain relief led to the identification of molecules such as cytokines, chemokines and more recently matrix metalloproteinases (MMPs) as putative therapeutic targets. Evidence points to a founder member of the matricial CCN family, NOV/CCN3, as a modulator of these inflammatory mediators. We thus investigated the possible involvement of NOV in a preclinical model of persistent inflammatory pain. METHODS: We used the complete Freund's adjuvant (CFA)-induced model of persistent inflammatory pain and cultured primary sensory neurons for in vitro experiments. The mRNA expression of NOV and pro-inflammatory factors were measured with real-time quantitative PCR, CCL2 protein expression was assessed using ELISA, MMP-2 and -9 activities using zymography. The effect of drugs on tactile allodynia was evaluated by the von Frey test. RESULTS: NOV was expressed in neurons of both dorsal root ganglia (DRG) and dorsal horn of the spinal cord (DHSC). After intraplantar CFA injection, NOV levels were transiently and persistently down-regulated in the DRG and DHSC, respectively, occurring at the maintenance phase of pain (15 days). NOV-reduced expression was restored after treatment of CFA rats with dexamethasone. In vitro, results based on cultured DRG neurons showed that siRNA-mediated inhibition of NOV enhanced IL-1ß- and TNF-α-induced MMP-2, MMP-9 and CCL2 expression whereas NOV addition inhibited TNF-α-induced MMP-9 expression through ß1 integrin engagement. In vivo, the intrathecal delivery of MMP-9 inhibitor attenuated mechanical allodynia of CFA rats. Importantly, intrathecal administration of NOV siRNA specifically led to an up-regulation of MMP-9 in the DRG and MMP-2 in the DHSC concomitant with increased mechanical allodynia. Finally, NOV intrathecal treatment specifically abolished the induction of MMP-9 in the DRG and, MMP-9 and MMP-2 in the DHSC of CFA rats. This inhibitory effect on MMP is associated with reduced mechanical allodynia. CONCLUSIONS: This study identifies NOV as a new actor against inflammatory pain through regulation of MMPs thus uncovering NOV as an attractive candidate for therapeutic improvement in pain relief.

CCL2 released from neuronal synaptic vesicles in the spinal cord is a major mediator of local inflammation and pain after peripheral nerve injury.

CCL2 chemokine and its receptor CCR2 may contribute to neuropathic pain development. We tested the hypothesis that injury to peripheral nerves triggers CCL2 release from afferents in the dorsal horn spinal cord (DHSC), leading to pronociceptive effects, involving the production of proinflammatory factors, in particular. Consistent with the release of CCL2 from primary afferents, electron microscopy showed the CCL2 immunoreactivity in glomerular boutons and secretory vesicles in the DHSC of naive rats. Through the ex vivo superfusion of DHSC slices, we demonstrated that the rate of CCL2 secretion was much lower in neonatal capsaicin-treated rats than in controls. Thus, much of the CCL2 released in the DHSC originates from nociceptive fibers bearing TRPV1 (transient receptor potential vanilloid 1). In contrast, high levels of CCL2 released from the DHSC were observed in neuropathic pain animal model induced by chronic constriction of the sciatic nerve (SN-CCI). The upregulated expression of proinflammatory markers and extracellular signal-regulated kinase (ERK) 1/2 pathway activation (ERK1/2 phosphorylation) in the DHSC of SN-CCI animals were reversed by intrathecal administration of the CCR2 antagonist INCB3344 (N-[2-[[(3S,4S)-1-E4-(1,3-benzodioxol-5-yl)-4-hydroxycyclohexyl]-4-ethoxy-3-pyrrolidinyl]amino]-2-oxoethyl]-3-(trifluoromethyl)benzamide). These pathological pain-associated changes in the DHSC were mimicked by the intrathecal injection of exogenous CCL2 in naive rats and were prevented by the administration of INCB3344 or ERK inhibitor (PD98059). Finally, mechanical allodynia, which was fully developed 2 weeks after SN-CCI in rats, was attenuated by the intrathecal injection of INCB3344. Our data demonstrate that CCL2 has the typical characteristics of a neuronal mediator involved in nociceptive signal processing and that antagonists of its receptor are promising agents from treating neuropathic pain.

Chronic stress induces transient spinal neuroinflammation, triggering sensory hypersensitivity and long-lasting anxiety-induced hyperalgesia.

Chronic stressful events induce biochemical, physiological and psychological changes, resulting in stress-related neuropsychiatric disorders, such as anxiety or depression. Using repeated social defeat as a stressful event model, we show that this preclinical paradigm induces a transient increase in the expression of the genes encoding the pro-inflammatory molecules iNOS and COX-2. We provide the first demonstration that chronic stress affects spinal plasticity through a mechanism involving local neuroinflammation. The functional consequences of such neuroinflammation are associated with a transient decrease in the mechanical nociceptive threshold. Administration of the cholecystokinin(CCK)-2 receptor antagonist, CI-988, directly into the Rostral Ventromedial Medulla reverses the chronic stress-induced decrease in the nociceptive threshold. These data strongly suggest that chronic stress induces a spinal neuroinflammation associated with transient sensory hypersensitivity involving the activation of CCK-dependent nociceptive descending facilitatory pathways. Pharmacological data show that chronic social stress-induced long-lasting state of anxiety is not responsible for maintaining the spinal neuroinflammation and, therefore, for the associated sensory hypersensitivity. Conversely, an evaluation of pain-related behavior in the formalin model indicates that anxiety is directly related to prolonged hyperalgesia prevented by systemic benzodiazepine or CCK-2 receptor antagonist treatments. The present study highlights the adverse effects of chronic stress on spinal neuroinflammation triggering sensory hypersensitivity. Exploration of this phenomenon points out the divergence between pain sensitivity and anxiety-induced hyperalgesia, which is in agreement with clinical observations. Altogether, these data open up new perspectives for clinical research devoted to the evaluation and treatment of pain in anxio-depressive patients.

SOCS3-mediated blockade of JAK/STAT3 signaling pathway reveals its major contribution to spinal cord neuroinflammation and mechanical allodynia after peripheral nerve injury.

Neuropathic pain after peripheral nerve injury, associated with local neuroinflammation in the spinal cord, is a severe incapacitating condition with which clinical treatment remains challenging. Inflammatory molecules signal through various intracellular transduction pathways, activation of which may amplify and cause spreading of the inflammatory response. We showed recently that spinal nerve lesion leads to rapid activation of Janus kinase (JAK)/signal transducer and activator of transcription 3 (STAT3) signal transduction pathway in dorsal spinal cord microglia in relation with enhanced levels of spinal interleukin-6 (IL-6) protein. Here, we selectively inactivated JAK/STAT3 signaling in rat dorsal spinal cord glia through local, lentiviral-mediated production of the suppressor of cytokine signaling SOCS3, a physiologic inhibitory protein of JAK/STAT3, and analyzed its consequences in a preclinical model of neuropathic pain. The targeted blockade of JAK/STAT3 activity prevented the abnormal expression of IL-6, CC chemokine ligand CCL2, and activating transcription factor ATF3 induced in the spinal cord by chronic constriction injury of the sciatic nerve (CCI) and substantially attenuated mechanical hypersensitivity (allodynia) in rats. In naive rats, intrathecal administration of a proalgesic cytokine IL-6 rapidly activated microglial JAK/STAT3 and induced downstream changes closely resembling CCI-evoked alterations. We identified downstream mechanisms through which JAK/STAT3 pathway activation leads to the spreading of neuroinflammation. Our findings reveal that JAK/STAT3 signaling plays a major role in spinal cord plasticity and mechanical allodynia associated with peripheral nerve injury.

Gene-based approaches in the study of pathological pain.

Chronic pathological pain is characterized by extensive plasticity of the systems involved in pain signal transmission and modulation and tissue remodeling in several CNS structures. These long-lasting alterations are mediated by, or associated with, changes in the production of key molecules of nociceptive processing. Gene-based approaches offer the unique possibility of using local or even cell-type specific interventions to correct the abnormal production of some of these proteins, modulate the activity of signal transduction pathways, or overproduce various therapeutic secreted proteins. We showed that certain viral-derived vectors are particularly suitable for mediating gene transfer highly preferential for instance into the primary sensory neurons or into the spinal cord glial cells that represent particularly pertinent targets in the search for new therapeutic strategies of pathological pain.

Differential implication of proinflammatory cytokine interleukin-6 in the development of cephalic versus extracephalic neuropathic pain in rats.

Responses resulting from injury to the trigeminal nerve exhibit differences compared with those caused by lesion of other peripheral nerves. With the aim of elucidating the physiopathological mechanisms underlying cephalic versus extracephalic neuropathic pain, we determined the time course expression of proinflammatory cytokines interleukin-6 (IL-6) and IL-1beta, neuronal injury (ATF3), macrophage/microglial (OX-42), and satellite cells/astrocyte (GFAP) markers in central and ganglion tissues in rats that underwent unilateral chronic constriction injury (CCI) to either infraorbital nerve (IoN) (cephalic area) or sciatic nerve (SN) (extracephalic area). Whereas CCI induced microglial activation in both models, we observed a concomitant upregulation of IL-6 and ATF3 in the ipsilateral dorsal horn of the lumbar cord in SN-CCI rats but not in the ipsilateral spinal nucleus of the trigeminal nerve (Sp5c) in IoN-CCI rats. Preemptive treatment with minocycline (daily administration of 20 mg/kg, i.p., for 2 weeks) partially prevented pain behavior and microglial activation in SN-CCI rats but was ineffective in IoN-CCI rats. We show that IL-6 can upregulate OX-42 and ATF3 expression in cultured microglia and neurons from spinal cord, respectively, as well as in the dorsal horn after acute intrathecal administration of the cytokine. We propose that IL-6 could be one of the promoters of the signaling cascade leading to abnormal pain behavior in SN-CCI but not IoN-CCI rats. Our data further support the idea that different pathophysiological mechanisms contribute to the development of cephalic versus extracephalic neuropathic pain.

JAK/STAT3 pathway is activated in spinal cord microglia after peripheral nerve injury and contributes to neuropathic pain development in rat.

Peripheral nerve lesion leads to the production of interleukin 6 (IL-6)-related neuropoietic cytokines involved in nerve protection and regeneration. This family of cytokines mainly signal through the signal transducer and activator of transcription (STAT) pathway that is locally activated in injured nerves. IL-6 is also involved in pain that frequently arises from peripheral nerve lesion. We investigated the possible activation of this major IL-6 signaling system in the spinal cord after peripheral nerve injury and its role in neuropathic pain. Ligation of L5-L6 spinal nerves (SNL) evoked an accumulation of active, phosphorylated form of STAT3 in microglial cells of dorsal spinal cord mostly in projection areas of injured nerves. SNL resulted also in a massive induction of IL-6 mRNA expression in dorsal root ganglia and increased concentration of IL-6 in dorsal spinal cord. Intrathecal injection of anti-rat IL-6 antibodies prevented the SNL-induced accumulation of phospho-STAT3 in the spinal cord. STAT3 pathway blockade with Janus kinase 2 inhibitor AG490 attenuated both mechanical allodynia and thermal hyperalgesia in SNL rats. These data show that in response to SNL injury Janus kinase/STAT3 system is activated mainly through IL-6 signaling in spinal microglia and that this transduction pathway participates in development of pain associated with nerve alteration.

Spinal CCL2 pronociceptive action is no longer effective in CCR2 receptor antagonist-treated rats.

A better understanding of the mechanisms linked to chemokine pronociceptive effects is essential for the development of new strategies to better prevent and treat chronic pain. Among chemokines, MCP-1/CCL2 involvement in neuropathic pain processing is now established. However, the mechanisms by which MCP-1/CCL2 exerts its pronociceptive effects are still poorly understood. In the present study, we demonstrate that MCP-1/CCL2 can alter pain neurotransmission in healthy rats. Using immunohistochemical studies, we first show that CCL2 is constitutively expressed by primary afferent neurons and their processes in the dorsal horn of the spinal cord. We also observe that CCL2 is co-localized with pain-related peptides (SP and CGRP) and capsaicin receptor (VR1). Accordingly, using in vitro superfusion system of lumbar dorsal root ganglion and spinal cord explants of healthy rats, we show that potassium or capsaicin evoke calcium-dependent release of CCL2. In vivo, we demonstrate that intrathecal administration of CCL2 to healthy rats produces both thermal hyperalgesia and sustained mechanical allodynia (up to four consecutive days). These pronociceptive effects of CCL2 are completely prevented by the selective CCR2 antagonist (INCB3344), indicating that CCL2-induced pain facilitation is elicited via direct spinal activation of CCR2 receptor. Therefore, preventing the activation of CCR2 might provide a fruitful strategy for treating pain.

Lentiviral-mediated targeted transgene expression in dorsal spinal cord glia: tool for the study of glial cell implication in mechanisms underlying chronic pain development.

Activated glial cells in the dorsal spinal cord take an important part in the development of pain after peripheral nerve injury. Our understanding of mechanisms involved in functional changes of spinal glia remains incomplete. Excepting drugs that completely disrupt glial function, pharmacological studies fail to target glia and to modify locally its function in order to really discriminate the role of neuronal versus glial cells in chronic pain. We developed an intraspinal gene transfer approach using pseudotyped lentiviral-derived vector targeting highly preferentially glial cells. Single microinjection of vector expressing EGFP under a CMV promoter control (LV-EGFP) allowed vector diffusion along a rostro-caudal axis but strictly restricted to the grey matter of the ipsilateral dorsal spinal cord. EGFP transgene was mainly expressed in astrocytes and microglial cells whereas less than 9% of cells containing EGFP were neurons. Notably, LV-EGFP administration and EGFP overexpression in glial cells did neither modify glial activity, nor alter animal's nociceptive or locomotor behaviors. Targeted modulation of the expression of gene of interest in glial cells, closely restricted to a particular region of the spinal cord, may thus represent an interesting approach to refine the understanding of mechanisms by which spinal glial cells participate in pain processing.