Skip to main content

Neuronal Panx1 drives peripheral sensitization in experimental plantar inflammatory pain



The channel-forming protein Pannexin1 (Panx1) has been implicated in both human studies and animal models of chronic pain, but the underlying mechanisms remain incompletely understood.


Wild-type (WT, n = 24), global Panx1 KO (n = 24), neuron-specific Panx1 KO (n = 20), and glia-specific Panx1 KO (n = 20) mice were used in this study at Albert Einstein College of Medicine. The von Frey test was used to quantify pain sensitivity in these mice following complete Freund’s adjuvant (CFA) injection (7, 14, and 21 d). The qRT-PCR was employed to measure mRNA levels of Panx1, Panx2, Panx3, Cx43, Calhm1, and β-catenin. Laser scanning confocal microscopy imaging, Sholl analysis, and electrophysiology were utilized to evaluate the impact of Panx1 on neuronal excitability and morphology in Neuro2a and dorsal root ganglion neurons (DRGNs) in which Panx1 expression or function was manipulated. Ethidium bromide (EtBr) dye uptake assay and calcium imaging were employed to investigate the role of Panx1 in adenosine triphosphate (ATP) sensitivity. β-galactosidase (β-gal) staining was applied to determine the relative cellular expression levels of Panx1 in trigeminal ganglia (TG) and DRG of transgenic mice.


Global or neuron-specific Panx1 deletion markedly decreased pain thresholds after CFA stimuli (7, 14, and 21 d; P < 0.01 vs. WT group), indicating that Panx1 was positively correlated with pain sensitivity. In Neuro2a, global Panx1 deletion dramatically reduced neurite extension and inward currents compared to the WT group (P < 0.05), revealing that Panx1 enhanced neurogenesis and excitability. Similarly, global Panx1 deletion significantly suppressed Wnt/β-catenin dependent DRG neurogenesis following 5 d of nerve growth factor (NGF) treatment (P < 0.01 vs. WT group). Moreover, Panx1 channels enhanced DRG neuron response to ATP after CFA injection (P < 0.01 vs. Panx1 KO group). Furthermore, ATP release increased Ca2+ responses in DRGNs and satellite glial cells surrounding them following 7 d of CFA treatment (P < 0.01 vs. Panx1 KO group), suggesting that Panx1 in glia also impacts exaggerated neuronal excitability. Interestingly, neuron-specific Panx1 deletion was found to markedly reduce differentiation in cultured DRGNs, as evidenced by stunted neurite outgrowth (P < 0.05 vs. Panx1 KO group; P < 0.01 vs. WT group or GFAP-Cre group), blunted activation of Wnt/β-catenin signaling (P < 0.01 vs. WT, Panx1 KO and GFAP-Cre groups), and diminished cell excitability (P < 0.01 vs. GFAP-Cre group) and response to ATP stimulation (P < 0.01 vs. WT group). Analysis of β-gal staining showed that cellular expression levels of Panx1 in neurons are significantly higher (2.5-fold increase) in the DRG than in the TG.


The present study revealed that neuronal Panx1 is a prominent driver of peripheral sensitivity in the setting of inflammatory pain through cell-autonomous effects on neuronal excitability. This hyperexcitability dependence on neuronal Panx1 contrasts with inflammatory orofacial pain, where similar studies revealed a prominent role for glial Panx1. The apparent differences in Panx1 expression in neuronal and non-neuronal TG and DRG cells are likely responsible for the distinct impact of these cell types in the two pain models.


Pain sensitivity is amplified and becomes persistent through both peripheral and central sensitization mechanisms [1,2,3]. Chronic inflammatory pain is complex, involving not only tissue damage but also local release of inflammatory signals that aggravate and prolong allodynia. Traditional pharmacological approaches [opioids, cannabinoids, and non-steroidal anti-inflammatory drugs (NSAIDs)] have made great progress in reducing inflammatory pain, but they present potential risks in clinical practice [4, 5]. For instance, long-term analgesic usage may elevate the risk of kidney and cardiovascular disease [6, 7], and pain-relieving opioid and cannabinoid drugs can be addictive [8,9,10]. Thus, it is critical to explore novel drug targets for the development of inflammatory pain therapies.

Peripheral pain sensitization consists of a reduced threshold for painful sensitivity (allodynia /hyperalgesia) [11], and the dorsal root ganglion (DRG) thus offers a target for neuromodulation in the transmission of pain signals to the dorsal horn of the spinal cord in chronic inflammatory pain [12,13,14]. Inflammatory injury intensifies interactions between sensory neurons and satellite glial cells (SGCs) surrounding them, contributing to persistent hyperexcitability through pathophysiological changes in DRG neurons (DRGNs) [15, 16]. These changes offer potential novel therapeutic targets for DRG-based pain modulation in inflammatory pain.

Pannexin1 (Panx1) has been shown to be involved in chronic pain and could thus be a target following inflammatory pain insult [17,18,19,20]. As a non-selective and large–conductance channel, Panx1 provides a conduit for the release of adenosine triphosphate (ATP) [21,22,23,24], which has long been known to enhance neuron excitability and to be activated by both mechanical and chemical signals [25,26,27,28,29,30,31]. Critically, Panx1 is expressed in both neurons and SGCs in sensory ganglia, and its expression increases in the ganglia in chronic inflammatory pain models [17, 20].

In neuropathic pain models [32, 33], Panx1 blockade by gene deletion or small molecule inhibitors has been shown to reduce hypersensitivity. We have previously demonstrated that both glial and neuronal Panx1 in the trigeminal ganglia (TG) contribute to orofacial pain hypersensitivity in an inflammatory mouse model [18, 34, 35], and another study reported that Panx1 channel in immune cells contributes to inflammatory pain originating in the DRG [36]. However, the underlying mechanism by which Panx1 contributes to inflammatory pain is largely unresolved.

In this study, we aimed to investigate how Panx1 controls peripheral pain sensitization after inflammatory stimuli. Panx1 levels in the DRG were found to be increased in a mouse model of chronic inflammation-induced pain, and in mice lacking Panx1, either globally or in neurons, pain hypersensitivity was much attenuated. SGC deletion of Panx1, previously found to appreciably reduce inflammatory orofacial pain, was less effective than neuronal deletion in the hindlimb inflammatory pain model, consistent with the lower SGC than neuronal Panx1 expression we detected in DRG compared to TG. Deletion of Panx1 in DRG and Neuro2a neurons resulted in numerous deficits in morphology and function, including much-reduced neurite extension, decreased excitability, and impaired ATP release. These studies showing major impact of Panx1 on peripheral inflammatory pain may provide a novel therapeutic strategy in future clinical trials and drug development.



The Institutional Animal Care and Use Committee (IACUC) of Einstein approved all animal experiments (NO. IACUCs20151101), assuring compliance with ethical and legal requirements for animal research.

The 8–12-week-old mice (n = 48) were used in the course of the study: wild-type (WT, n = 24), global Panx1 knockout (KO) (Panx1−/−, n = 24), neuron-specific Panx1 KO (NFH-Cre: Panx1f/f, abbreviated NFH-Cre, n = 20), and glia-specific Panx1 KO (GFAP-Cre: Panx1f/f, abbreviated GFAP-Cre, n = 20). Global Panx1 KO and neuron/glia-specific Panx1 KO mice were generated in the C57BL/6 background at the IACUC-approved animal facility of Albert Einstein College of Medicine, as previously reported [21]. Briefly, homozygous global Panx1 KO was generated from hemizygous Panx1tm1a(KOMP)Wtsi mice (KOMP, Davis, CA, USA). Panx1f/f mice were obtained by crossing Panx1tm1a (KOMP)Wtsi with flippase deleter mice (B6.ACTFLPe/J), allowing for neuron/glia-specific Panx1 knockout mice strain generation following crossing with cell-type specific Cre mice (Jackson Laboratory, Bar Harbor, ME, USA). Glia-specific Panx1 KO mice were generated by crossing mGFAP-Cre [B6.Cg-Tg (Gfap-cre) 73.12 Mvs/J] and Panx1f/f mice. Neuron-specific Panx1 KO mice were established by crossing mNFH-Cre [Tg (Nefh-cre) 12 Kul/J] and Panx1f/f mice.

Plantar inflammatory pain model

The plantar inflammatory pain was induced by complete Freund’s adjuvant (CFA, #F5881, Sigma-Aldrich, Germany), following the previously established procedure [18]. For that, 20 µl of CFA solution (1 mg/ml, prepared by mixing 1:1 saline to CFA) was injected subcutaneously into the plantar portions of the right hind paw under anesthesia (isoflurane).

Pain sensitivity analysis

Wild-type (n = 11), global Panx1 KO (n = 11), neuron-specific Panx1 KO (n = 12), and glia-specific Panx1 KO mice (n = 12) were used in the behavioral test. Mouse paws were stimulated with von Frey filaments (Stoelting Co., Dale Wood, IL, USA) to measure tactile sensitivity as previously described [17]. Briefly, the mice were placed in elevated plastic boxes with a wire mesh floor and each filament was applied ten times with 5–30 s random intervals to target the paw. Tactile thresholds (expressed in g) were measured as the calibrated bending forces of the filament that elicited a withdrawal response. Data for 50% pain thresholds were collected on 0, 7, 14, and 21 d, and represented as three-dimensional (3D) surface plots and two-dimensional (2D) heatmaps of the number of responses to 10 stimulus presentations of each intensity using the NSCC software (version 2019, NSCC, Kaysville, UT, USA).

Cell culture

Details regarding the generation of Panx1 CRISPR KO and overexpressing Neuro2a cell lines were previously described [37]. Neuro2a and Neuro2a CRISPR KO Panx1 and overexpressing Panx1 cells were cultured in DMEM (#8,121,513, Gibco, USA) supplemented with 10% fetal bovine serum (FBS; #10,099,141, Gibco, USA). To stimulate neuronal outgrowth, those Neuro2a cell lines above were differentiated by a 5 d treatment with 40 µmol/L retinoic acid (RA; #302-79-4, Merck, Germany) to specifically investigate the function of Panx1 channels.

Sensory neurons were dissociated from L5 DRG following painless death (isoflurane followed by decapitation) using collagenase/dispase digestion [38] from WT, Panx1 mutant, and CFA-treated mice at the same time point. Then, a total of 20,000 neurons were plated in 60-mm dishes coated with poly-d-lysine (PDL; 25 µg/ml, #LPDL001, Sigma-Aldrich, Germany) and then cultured in 5 µg/ml nerve growth factor (NGF; #9061-61-4, Sigma-Aldrich, Germany) for 5 d [39].

To explore the signaling pathways underlying dendritic growth during DRG neuron culture, an agonist (20 mmol/L LiCl, #7447-41-8, Selleck Chemicals, USA) was added to activate Wnt/β-Catenin signaling by limiting β-catenin phosphorylation and degradation [40], Wnt ligand blocker (30 nmol/L IWP2, #686770-61-6, Merck, Germany) was added to prevent early Wnt ligand palmitoylation by targeting the O-acyltransferase porcupine (Porcn) [41], and β-catenin inhibitor (100 nmol/L Triptonide, #38647-11-9, Selleck Chemicals, USA) was added to block the Wnt pathway by binding to β-catenin’s C-terminal Transactivation Domain [42].

Cell transfection

The 3 µg of enhanced green fluorescent protein (EGFP)-labelled plasmid mPanx1-DNA with 10 µl of transit LT1 transfection reagent (#MIR-2304, Mìrus, USA) in 250 µl of Opti-MEM (#31,985,062, GIBCO, USA) were used to re-express Panx1 in Neuro2a Panx1 KO cells. Cell culture medium was replaced with standard growth medium [DMEM (#8,121,513, Gibco, USA) supplemented with 10% FBS (#10,099,141, Gibco, USA)] after 24 h of transfection.

Neurite outgrowth assessment

For quantification of neural differentiation, cells were viewed using phase-contrast microscopy (20×) with a Nikon Eclipse TE300 microscope (Nikon Co., Minato, TY, Japan). Five fields from each cell culture were randomly selected, imaged, and then measured using ImageJ (NIH, Bethesda, MD, USA) [43]. At least 30 cells in each random field of view were studied to calculate the neurite-bearing cell ratio using the following formula:

$$\varvec{Ratio\;(\%)}=(\sum \text{neurite}\_\text{bearing}/\sum \text{neuron})\times100$$

where ∑neurite_bearing is the number of neurite-bearing cells and ∑neuron is the total number of neurons. A cell was defined as neurite-bearing if it had at least one neuronal process that was longer than its soma.

Sholl analysis

Sholl analysis was performed on 8-bit binary images of selected neurofilament-200 (NF-200)-stained DRGNs (neurite length ≥ 20 μm) using the Sholl analysis plug-in (ImageJ, NIH, Bethesda, MD, USA). Dendrite intersections were automatically evaluated at 10-µm radius intervals starting from the soma. Primary and secondary dendrite lengths were determined by measuring the distance of each radius from the soma/branch point.

Ethidium bromide (EtBr) uptake

DRGNs were seeded in monolayers on glass-bottomed dishes (MatTek, USA) at a density of 2000 cells/cm2, and then cultured with DMEM medium (#8,121,513, Gibco, USA) with 10% FBS (#10,099,141, Gibco, USA) for 5 d. After rinsing twice with PBS without Ca2+ and Mg2+ (#21-031-CV, Corning, USA), DRGNs were incubated for 10 min in 5 ml of DMEM with 10 µmol/L EtBr (#E7637, Merck, Germany) and 2 mmol/L ATP (#R0441, Thermo Scientific, USA). The fluorescence intensity of EtBr dye uptake was monitored using a Nikon Eclipse TE300 microscope (Nikon Co., Minato, TY, Japan) and Spot-RT digital camera with fixed gain and exposure duration.

Extracellular ATP release

As previously described [17], the extracellular ATP released from DRG ganglia was determined using the D-luciferin/luciferase assay (#A22066 ATP Determination Kit, Thermo Scientific, USA). Briefly, excised DRG ganglia of mice one week after CFA injection into hindlimb paw were immediately incubated in glass-bottomed dishes (MatTek, USA) containing ACSF (125 mmol/L NaCl, 3.5 mmol/L KCl, 2.0 mmol/L CaCl2, 2.0 mmol/L MgSO4, 1.25 mmol/L KH2PO4, 26 mmol/L NaHCO3, 10 mmol/L D-glucose, pH 7.2) at 37 °C for 40 min. Then, 500 µl of ACSF was collected and evaluated using an ATP Determination Kit according to the manufacturer’s instructions. Finally, the luminescence of ACSF at 560 nm was monitored using a TD 20/20 luminometer (Turner Designs, Sunnyvale, CA, USA). The ATP values released by the DRG were calculated based on the standard curves and normalized to the total protein amount, which was quantified using a bicinchoninic acid assay (#23,225, Thermo Scientific, USA).

Calcium imaging

DRGNs were loaded with the ratiometric Ca2+ indicator Fura-2 acetoxymethyl ester (5 µmol/L; #F1221, Invitrogen, USA) at 37 °C for 30 min, rinsed, and kept in phenol-free α-MEM supplemented with 1% FBS (#10,099,141, Gibco, USA) and 20 mmol/L HEPES (#7365-45-9, Sigma-Aldrich, Germany) throughout each test. Cells were imaged on a Nikon TE 2000 microscope (20 × 0.65 NA objective; Nikon Co., Minato, TY, Japan) with MetaFluor software (Molecular Devices Co., San Jose, CA, USA) using a Hamamatsu camera and Sutter filter changer connected to the microscope via a fiber optic cable. Intracellular Ca2+ was measured at 2 Hz by imaging regions of interest (ROIs) corresponding to individual cell bodies with 380/340 nm dual excitation (Sutter filter changer) and Fura-2 filter cube. As described in our earlier study [44], ratio values were converted to intracellular Ca2+ concentrations using in vitro calibrations in MetaFluor software (version 6.1, Universal Imaging. Corp., USA). Next, 2 mmol/L ATP was added to the bathing solution after 10 s of 200 s tracing to stimulate cellular Ca2+ responses. ROIs on at least 10 cell bodies were used to determine Ca2+ responses in each test. Data were normalized to the average non-ATP treated baseline amplitudes.

RNA extraction and qRT-PCR

RNA samples were extracted from Neuro2a cells, Panx1 KO DRGNs, Wnt/β-catenin agonist/inhibitor-treated DRGNs, and DRG tissue using TRIzol reagent (#15,596,026, Invitrogen, USA) for qRT-PCR with corresponding primers (Additional file 1: Table S1). The cDNA was synthesized from 1 µg of RNA sample using SuperScript™ IV VILO™ Master Mix (#11,756,500, Invitrogen, USA) according to the manufacturer’s instructions.

The qRT-PCR assay was performed using Power SYBR® Green RNA-to-CT™ 1-Step Kit (#4,389,986, Thermo Scientific, USA) and QuantStudio™ 6 Flex system (Life Technologies, Carlsbad, CA, USA). Reaction mixtures were denatured at 95 °C for 10 min followed by 40 PCR cycles. Each cycle consisted of the following three steps: 94 °C for 15 s, 57 °C for 15 s, and 72 °C for 1 min. The ddCT method was employed to determine relative gene-level differences with GAPDH qPCR products used as the control.


Whole-cell patch clamp was utilized to determine the amplitude of voltage-activated membrane currents in Neuro2a and DRG cells before and after the differentiating treatments. Cells were bathed in PBS solution containing 1 g/L glucose at pH 7.2. Patch pipettes with tip resistance of 3–6 MΩ were filled with an internal solution consisting of K-gluconate (140 mmol/L), HEPES (10 mmol/L), MgCl2 (1 mmol/L), EGTA (1 mmol/L), and Na2ATP (4 mmol/L) at pH 7.2. Membrane currents were recorded using an Axopatch-1B amplifier with a digitizer and stored in a computer for analysis using pClamp10 (Molecular Devices Co., San Jose, CA, USA). Leakage subtraction was employed in some recordings using standard protocols. Values reported for inward current amplitudes generally reflected such subtraction. Panx1 blocker probenecid (Pbcd; 1 mmol/L, #57-66-9 Sigma-Aldrich, Germany) was added during cell differentiation in Neuro2a Panx1 KO, Neuro2a WT, Neuro2a Panx1 OE, DRGNs WT, and Panx1 KO DRGNs groups.


The DRGNs on PDL-coated coverslips or DRG tissue Sects. (10–15 μm) were fixed using 4% PFA solution (#158,127, Sigma-Aldrich, Germany) in phosphate-buffered saline (PBS) for 15 min, permeabilized with 0.4% Triton-X 100 (#93,443, Sigma-Aldrich, Germany) in 1× PBS for 10 min, and blocked at room temperature with PBS buffer containing 0.4% Triton-X 100 and 10% goat serum (#16,210,064, Thermo Scientific, USA). The coverslips and sections were stained with primary antibodies targeting NF200 (1:200; rabbit IgG; #N4142, Sigma-Aldrich, Germany) and glutamine synthetase (1:500; mouse IgG, #sc-74,430, Santa Cruz, USA) in the incubating solution (0.4% Triton-X 100 and 2% goat serum) for at least 24 h at 4 °C. Following three gentle rinses with 1X PBS, the samples were incubated with secondary antibodies that included donkey anti-rabbit IgG-Alexa Fluor™488 (1:1,000; #A-21,206, Invitrogen, USA) and goat anti-mouse IgG (H + L) - Alexa Fluor™ 594 (1:1,000, #A-11,032, Invitrogen, USA) at 37 °C for 2 h. After three PBS washes in the dark, the samples were mounted using the mounting medium (#H-1000-10, Vector Laboratories, USA) containing 1% 4’6-Diamidino-2-phenylindole stock (DAPI, 10 mmol/L, #D9542, Sigma-Aldrich, Germany). The coverslips and DRG tissues were photographed using a Zeiss LSM 880 confocal microscope equipped with a Plan-Apochromat 63x/1.40 NA Oil DIC objective (Carl Zeiss AG, Jena, TH, Germany).

β-galactosidase (β-gal) Staining

β-gal staining in TG and DRG sections was performed as previously described [17]. Briefly, tissues were fixed in 4% PFA in PBS for 1 h, rinsed with PBS, washed with the washing buffer (2.0 mmol/L MgCl2, 0.01% deoxycholate, 0.02% NP-40, 97 mmol/L Na2PO4, pH 7.3) for 30 min, and incubated overnight at 37 °C with X-gal reaction solution [1.0 g/L X-gal (#R0941, Thermo Scientific, USA), 5 mmol/L potassium ferrocyanide (#14459-95-1, Merck, Germany), 7 mmol/L potassium ferricyanide]. After rinsing with the washing buffer, tissues were post-fixed overnight in 4% PFA and then immersed in (5, 10, 20, and 30%) sucrose solution in PBS for 24 h each at 4 °C. The tissues were then embedded in an OCT compound and immediately frozen at − 80 °C. Sections 10–15 μm in thickness were cut using a cryostat at − 20 °C. Tissues were examined with a Leica inverted DMi8 SP8 microscope. Leica LASX software was used for image acquisition (Leica Microsystems CMS GmbH, Mannheim, BW, Germany). Images were analyzed using ImageJ software (NIH, Bethesda, MD, USA). The number of stained cells (neurons or satellite glial cells) was normalized to the total number of neurons.

Statistical analysis

Measurement data were expressed as mean ± standard error of the mean (SEM) from at least three independent replicates. Data were analyzed using SPSS (version 22, IBM Co., Armonk, NY, USA) and plotted with Prism software (version 9.5.1, GraphPad, Armonk, NY, USA). T-test was used to compare two groups except for β-gal staining, where unpaired Student’s t-test was utilized. One-way ANOVA was carried out for multiple comparisons. Two-way ANOVA with Bonferroni post hoc test was performed to compare the variables between different genotypes, including 50% pain threshold and β-catenin relative level test. All statistical details are shown in the respective figure legend. A P-value of ≤ 0.05 indicated statistical significance.

For behavioral studies, at least 10 mice were included in each animal group based on the power analysis recommendation [23], and tests were scheduled at their appointed time.


Global or neuron-specific Panx1 deletion decreases pain sensitivity

In this study, the inflammatory agent CFA was injected into the mouse hind paw (Fig. 1a) to test whether Panx1 expression in sensory ganglia was altered during chronic pain. As shown in Fig. 1b, gene expression of Panx1 was significantly progressively increased in L5 DRG from 7 to 21 d after CFA injection. However, there was no increase in Panx1 mRNA in the L5 spinal cord (dorsal and ventral horns; Fig. 1c). Pain threshold values showed a more substantial and significant increase in global Panx1 KO mice, compared to those WT mice (Fig. 1d). Normalized values of 50% pain thresholds (post-/pre-CFA) in WT were strikingly lower than in global Panx1 deletion from 7 d till 21 d (Additional file 1: Fig. S1a). The positive response fraction to each magnitude of von Frey filament applied force at each time point is shown in pseudo-color displays (Additional file 1: Fig. S1c). As shown in the 3D convex topology of WT mice, there are fewer responses to weak stimuli and more responses to moderate and robust stimuli, compared to the planar topology in Panx1 KO group. This difference was apparent as a narrower blue band and a more intense red region in WT in its 2D plots. For Panx1 KO mice, the blue region was particularly pronounced on 14 d and 21 d, while the red intensity was lower at all time points. These results indicated that Panx1 deletion substantially blocks CFA administration-induced peripheral hypersensitivity.

Fig. 1
figure 1

Global or neuronal Panx1 deletion reduces allodynia while GFAP-targeted deletion does not. a Schematic plot of CFA induced peripheral inflammatory pain model. Panx1 mRNA expression in DRG (b) and spinal cord (c) from WT and Panx1 KO mice within 3 weeks after CFA inflammation. d Three-week dynamic 50% pain threshold of WT and Panx1 KO mice. e Three-week dynamic 50% pain threshold of NFH-Cre and GFAP-Cre Panx1 KO mice. f Comparison of pain sensitivity changes among those four genotypes above at each time point. For panels b and c, n = 3 in saline and CFA group; n = 11 in WT group, n = 11 in Panx1 KO group, n = 12 in GFAP-Cre group, and n = 12 in NFH-Cre group. Two-way ANOVA with Bonferroni, interaction F (2, 12) = 26.07 in panel b, F (2, 12) = 0.1176 in panel c, F (7, 68) = 29.66 in panel d, F (7, 87) = 33.44 in panel e, and F (9, 170) = 5.641 in panel f. ns non-significant, *P < 0.05, **P < 0.01. CFA complete Freund’s adjuvant, DRG dorsal root ganglion, NFH-Cre neuro filament H-Cre, GFAP-Cre glial fibrillary acidic protein-Cre, Panx1 pannexin 1, Panx1 KO pannexin 1 knockout, WT wild-type

Interestingly, further allodynia studies showed a difference between neuronal and glial Panx1 deletion mice. As shown in Fig. 1e, the pain sensitivity baseline (pre-CFA) was similar between those two groups. But neuron-specific Panx1 deletion (NFH-Cre) led to a significant decrease in pain sensitivity persisting for 3 weeks after CFA injection, compared to only minor differences in the glia-specific Panx1 deletion (GFAP-Cre) mice. Also, normalized 50% pain thresholds (post-/pre-CFA) in the GFAP-Cre group were markedly less decreased than in the NFH-Cre Panx1 KO mice from 7 to 21 d, indicating that neuronal Panx1 was critical for hypersensitivity attenuation (Additional file 1: Fig. S1b). The 3D plots indicated there were more numerous paw withdrawal responses in GFAP-Cre:Panx1f/f, particularly at the highest stimulus intensities and at longer times after inflammation. The 2D heatmaps of the GFAP-Cre mice showed that the extent and saturation of the red region were much higher than in NFH-Cre mice (Additional file 1: Fig. S1d). These results also confirmed that allodynia was significantly lower in the NFH-Cre group.

For pain threshold baselines, there were no significant differences among WT, Panx1 KO, NFH-Cre, and GFAP-Cre mice before CFA injection (Additional file 1: Fig. S2). Pain sensitivity was dramatically attenuated in global Panx1 KO/NFH-Cre mice, compared to WT and GFAP-Cre groups at all time points (Fig. 1f). The differences were not attributable to gross changes in mouse body weight, which was well maintained in all four genotypes (Additional file 1: Fig. S1e). Therefore, global/neuronal Panx1 KO but not glial Panx1 deletion decreased pain sensitivity in the CFA-induced inflammatory pain model.

Panx1 deletion decreases neurite extension and inward current in Neuro2a

To probe mechanisms underlying the effects of Panx1 on behavioral allodynia, we first established WT, Panx1 KO, and Panx1-overexpressing Neuro2a cell lines to examine the impact of Panx1 on neuronal morphology and cell excitability (Additional file 1: Fig. S3a). Morphologically, Neuro2a cells extended many more neurites in the WT and Panx1-overexpressing cells after 5 d of 40 µmol/L RA treatment compared to the Panx1-deleted group (Fig. 2a). RA did not enhance the percentage of neurite-bearing Panx1-null Neuro2a cells. The fraction of neurite-bearing Neuro2a cells in the absence of RA from 3 to 5 d was low in both WT and Panx1-overexpression groups and much lower in Panx1 KO cells, but it was substantially higher in both WT and especially in Panx1-overexpressing Neuro2a cells after RA treatment (Fig. 2b). To confirm that neurite extension was cell autonomous with regard to Panx1 expression, Panx1 deleted Neuro2a cells were transfected with an EGFP-tagged mPanx1 plasmid. In the same dish, neurite growth was much more elaborate in the EGFP-Panx1 group with RA treatment at 5 d in striking contrast to the surrounding Panx1-KO cells (Additional file 1: Fig. S3b).

Fig. 2
figure 2

Panx1 facilitates neurite extension and promotes cell excitability in Neuro2a neuroblastoma cells. a Representative photomicrograph of Neuro2a WT, Neuro2a Panx1 KO, and Neuro2a Panx1 OE cells after 5 d of RA treatment. Arrows indicate processes of representative cells in each condition. b Neurite extension analysis in WT, Panx1-deleted, and Neuro2a Panx1 OE cells at d 3–5 in culture with or without 40 µmol/L RA (n = 5). c Quantification of peak inward currents in Neuro2a WT, Neuro2a Panx1 KO and Neuro2a Panx1 OE on day 5 with and without RA treatment (n = 15–22 cells in each group). d Representative recordings of whole-cell voltage-clamp currents in Neuro2a WT, Neuro2a Panx1 KO, and Neuro2a WT plus Panx1 blocker 1 mmol/L Pbcd following 5 d treatment with RA. e Current density analysis of Neuro2a cells with 1 mmol/L Pbcd (n = 9–11 in each group). Two-way ANOVA with Bonferroni in b, c, interaction F (2, 24) = 15.04 in panel b, F (2, 56) = 18.62 in panel c; scale bar = 150 μm in panel a and scale bar = 20 μm in panel d. ns non-significant, *P < 0.05, **P < 0.01. Neuro2a WT neuro2a wild-type, Neuro2a Panx1 KO neuro2a pannexin 1 knockout, Neuro2a Panx1 OE pannexin 1-overexpressing neuro2a, Pbcd probenecid, RA retinoic acid

Concerning cell excitability, inward currents were more robust in cells overexpressing Panx1 and markedly weaker in Neuro2a Panx1 KO cells, compared to WT Neuro2a cells. Following 40 µmol/L RA treatment, inward currents were significantly increased in Panx1-overexpressing cells and slightly increased in WT cells, whereas currents were not enhanced by RA in Panx1 KO cells (Fig. 2c). Furthermore, cells were continuously subjected to the Panx1 blocker Pbcd (1 mmol/L) for 5 d of RA treatment to determine whether these differences depended on functional Panx1 channels. Representative recordings (Fig. 2d) and current density analysis (Fig. 2e) revealed the diminished peak currents in Neuro2a WT cells after blocking Panx1 channels. These results indicated that Panx1 channels enhance both neurogenesis and cell excitability and therefore would be expected to impact neuronal differentiation during development and network remodeling.

Panx1 impacts DRG neurogenesis via Wnt/β-catenin pathway

Next, primary cultures of dissociated L5 DRG sensory neurons were established from WT and Panx1 KO mice to probe Panx1’s impacts on neurogenesis and cell excitability. Notably, confocal microscopy imaging of neurofilaments from Panx1 KO DRGNs showed that the cells extended only short processes after 5 d of 5 µg/ml NGF treatment compared to much longer and more elaborate processes seen in WT DRGNs (Fig. 3a). As shown in Fig. 3b, Sholl analysis results revealed a higher number of dendrite intersections with Sholl rings in WT vs. Panx1 KO DRGNs, particularly in the distal dendritic arbor, indicating Panx1 enhanced neurogenesis.

Fig. 3
figure 3

Global Panx1 deletion impacts DRG neurogenesis via Wnt/β-catenin pathway. a cell morphology of NF-200 stained DRG neuron from WT and Panx1 KO mice after 5 d of NGF treatment. b Sholl analysis of intersection numbers in WT and Panx1 KO neurons (WT vs. Panx1 KO group, P < 0.01). c Following 5 d of treatment with NGF, cell morphology of DRG neurons (DRGNs) from WT and Panx1 KO mice was visualized NF200. Exposure to 20 mmol/L LiCl, 30 nmol/L IWP2, and 100 nmol/L Triptonide was compared. d mRNA expression of β-catenin in WT and Panx1 KO DRGNs after 5 d of treatment with NGF. Exposure to LiCl, IWP2, and Triptonide was also compared. e Quantification of cell morphology indicated neurite lengths in WT and Panx1 KO DRGNs following treatment with modifiers of the Wnt/β-catenin pathway (n = 15 in each group). Two-way ANOVA with Bonferroni in d, e, interaction F (7, 40) = 2.595 in panel d, F (3, 16) = 183.1 in panel e. ns non-significant, *P < 0.05, **P < 0.01. Scale bar = 100 μm in panels a, and scale bar = 20 μm in panels c. DRG dorsal root ganglion, IWP2 Wnt production inhibitor 2, LiCl lithium chloride, NF200 neurofilament 200, NGF nerve growth factor, WT wild-type, Panx1 KO pannexin 1 knockout

Since shRNA knockdown of Panx1 was reported to reduce mRNA levels of β-catenin in human melanoma cells, we further explored whether the Wnt/β-catenin signaling was involved in Panx1 KO DRGNs. For that, we applied 20 mmol/L LiCl (Wnt/β-catenin pathway agonist), 30 nmol/L IWP2 (Wnt inhibitor), and 100 nmol/L Triptonide (β-catenin blocker) during 5 d of NGF treatment. As shown in NF200-stained neuron images (Fig. 3c), the dendrite growth was affected in opposite ways by Wnt/β-catenin agonists or inhibitors. The qPCR results showed the mRNA levels of β-catenin were dramatically down-regulated in Panx1 KO compared to WT neurons in basal and LiCl treatment conditions (Fig. 3d). Our quantitative analysis revealed that the average dendrite length of WT neurons was approximately four times that of Panx1 KO neurons (Fig. 3e). After LiCl treatment, dendritic growth was stimulated in both WT and Panx1 KO DRGNs (Fig. 3e), consistent with Wnt signaling activation. However, inhibition of either Wnt or β-catenin almost markedly eliminated dendritic length augmentation in those two groups when IWP2 or Triptonide was added; in contrast, LiCl partially overcame neurite extension inhibition in Panx1 KO DRGNs, consistent with the downstream signaling via β-catenin (Fig. 3c, e). Collectively, these results indicated that Panx1 could modulate neurogenesis via the Wnt/β-catenin pathway in sensory neurons.

Panx1 channels activate DRGNs excitability

Given that pain is transduced by small diameter nociceptors, we assessed the inward currents in DRGNs of various sizes [small (< 20 μm), medium (20–25 μm), and large (> 25 μm)] isolated from ganglia of the CFA treated mice (7d). Representative recordings (Fig. 4a) and quantitative analysis (Fig. 4b) revealed that the amplitudes of the transient inward currents recorded from small DRGNs were significantly lower in Panx1 KO [(3521 ± 512.2) pico ammeter (pA)] than in the WT group [(6411 ± 651.6) pA] after 7 d of CFA treatment. This excitability difference was consistent with the lower pain sensitivity and allodynia in these animals.

Fig. 4
figure 4

Panx1 enhances NGF-induced inward currents, dye uptake, and ATP release. a After 5 d of NGF treatment, representative electrophysiological recordings of different diameter-DRGNs [small (< 20 μm), medium (20–25 μm), and large (> 25 μm)] from WT and Panx1 KO mice were detected and analyzed (b). c Maximum inward currents recordings from WT and Panx1 KO sensory neurons before and 5 d after NGF treatment (Ctrl, 5% serum treated; 5 µg/ml NGF treated; n = 10). d Representative recordings from WT, Panx1 KO and Pbcd inhibited Panx1 groups (n = 10). e Current density analysis of those three types of DRGNs above. e Statistical analysis of c. f EtBr dye uptake images and data analysis (g) between WT and Panx1 KO NGF treated DRGNs with and without 2 mmol/L ATP treatment. h ATP release levels analysis between WT and Panx1 KO DRGNs following CFA injection on day 7. Scale bar = 20 μm in d, f. ns non-significant, *P < 0.05, **P < 0.01 for one-way ANOVA in b, c, g, h. Ctrl control, DRGNs dorsal root ganglion neurons, EtBr ethidium bromide, NGF nerve growth factor, WT wild-type, Panx1 KO pannexin 1 knockout, Pbcd probenecid

To determine the extent to which Panx1 contributed to the heightened neuronal excitability following CFA treatment, we compared inward currents in DRGNs from WT mice with those lacking Panx1 or in which Panx1 function was blocked. As shown in Fig. 4c, the amplitude of inward currents recorded from Panx1 KO DRGNs was smaller than those in WT without CFA treatment, and this difference was markedly enhanced following 5 d in the presence of NGF. Similarly, blocking Panx1 channels by 1 mmol/L Pbcd significantly reduced the inward current amplitude and density of DRGNs (Fig. 4d, e). These results confirmed that functional Panx1 channels greatly augment cellular excitability in DRGNs.

Additionally, we explored the mechanism by which Panx1 channels contribute to pain sensitivity and allodynia. Measurement of the uptake of EtBr dye showed that membrane permeability of WT DRGNs was robustly increased in response to extracellular 2 mmol/L ATP (allogeneic molecule), while Panx1-null neurons did not show appreciably increased EtBr uptake in response to ATP application (Fig. 4f, g). Consistent with these dye uptake results, ATP release levels were much higher in WT DRG than in Panx1 KO ganglia from CFA-treated mice (7 d) (Fig. 4h). These results indicated that ATP-activated Panx1 channels were critical to the up-regulation of pain sensitivity in sensory neurons.

Panx1 in glia impacts exaggerated neuronal excitability

To determine whether ATP sensitivity was different in DRGNs and SGCs in sensory ganglion, intracellular Ca2+ changes evoked by ATP were measured in Fura-2 AM-loaded cells dissociated from DRG ganglion after CFA administration to the hindlimb paw (7 d). As shown in Fig. 5a, applying 2 mmol/L ATP enhanced Ca2+ responses in DRGNs and the SGCs surrounding them. Further calcium imaging analysis revealed there were smaller and slower Ca2+ rises in SGCs than in neurons (Fig. 5b, c). Notably, Ca2+ responses in both cell types were lower in Panx1 KO than in WT cultures (Fig. 5b). This lower ATP sensitivity in SGCs than in DRGNs from Panx1 KO mice indicates that Panx1 in glia also impacts exaggerated neuronal excitability, although its impact is less than in neurons.

Fig. 5
figure 5

Both dorsal root ganglion neurons (DRGNs) and satellite glial cells (SGCs) show heightened sensitivity to ATP in co-cultures from CFA-treated mice. a Ca2+ imaging of Fura2-loaded dissociated WT and Panx1 KO DRGNs and SGCs. Scale bar = 50 μm. b Quantification of calcium responses from DRGNs (blue) and SGCs (red) [n = 5; statistical analysis from either DRGNs or SGCs indicates that there are stronger calcium waves in WT, compared to Panx1 KO group (P < 0.01)]. c Data summarized and replotted from data sets in b (downward-pointing arrow indicates a time of ATP application). ns non-significant, *P < 0.05, **P < 0.01 for one-way ANOVA with repeated measures in c. WT wild-type, Panx1 KO pannexin 1 knockout

Neuronal Panx1 drives neurogenesis and hyperexcitability in DRG following CFA injection in paw

To probe the cell-specific role of Panx1 in DRG following CFA injection, a detailed analysis of dendrite development, cell excitability, and Ca2+ influx was performed in neurons from mice with targeted Panx1 deletion in neurons (NFH-Cre) and glia (GFAP-Cre). Sholl analysis results showed that neurotic arbors were more complex in both NFH-Cre and GFAP-Cre groups compared to global Panx1 KO, with neuron-targeted Panx1 deletion leading to a more severe reduction in intersection number than in GFAP-targeted Panx1 deletion (Fig. 6a, b). Further analysis of dendritic complexity in all genotypes indicated that a lower secondary dendrite length in NFH-Cre prominently affected the total branch length of DRGNs, even though there was no difference in branch numbers and primary dendrite length between GFAP-Cre and NFH-Cre groups (Additional file 1: Fig. S4a, S4b). Interestingly, the β-catenin mRNA level in GFAP-Cre was higher than in the NFH-Cre group, indicating that neuronal Panx1 contributes to neurogenesis (Fig. 6c). As shown in Fig. 6d, the inward current amplitude of GFAP-Cre mice was remarkably higher in smaller DRGNs compared to NFH-Cre mice, implying that neuronal Panx1 deletion evoked greater excitability of the small peripheral nociceptive cells than deletion in glia. Moreover, all Panx1-deleted mice showed lower ATP levels in DRG, but differences among the genotypes were not detected (Fig. 6e). However, intracellular Ca2+ responses of dissociated DRGNs were considerably higher in the GFAP-Cre than in the NFH-Cre group after 7 d post-CFA injection (Fig. 6f). Collectively, these results indicated that neuronal Panx1 contributes substantially to hyperexcitability following CFA injection.

Fig. 6
figure 6

Panx1 contributes to DRG changes in both neurons and glia in chronic inflammatory pain. a Representative morphologies of dorsal root ganglion neurons (DRGNs) from GFAP-Cre and NFH-Cre mice stained with NF200 antibody (green) after 5 d of NGF induction. Scale bar = 100 μm. b Sholl analysis of NGF-treated GFAP-Cre and NFH-Cre DRGNs on day 5. cβ-catenin mRNA analysis of those NGF-treated GFAP-Cre and NFH-Cre DRGNs compared to WT and global Panx1 KO [n = 6; NFH-Cre vs. GFAP-Cre group (P < 0.01), global Panx1 KO vs. NFH-Cre group (P < 0.05), global Panx1 KO vs. GFAP-Cre group (P < 0.01), WT vs. NFH-Cre or GFAP-Cre or global Panx1 KO group (P < 0.01)]. d Quantification of inward currents from GFAP-Cre and NFH-Cre DRGNs (traced with DiI) projecting to CFA-injected hind paw on day 7 (n = 8). e ATP release analysis of WT, Panx1 KO, GFAP-Cre and NFH-Cre DRG ganglia 7 d post-CFA injection (n = 10). f Ca2+ response analysis of GFAP-Cre and NFH-Cre DRGNs after CFA injection on day 7 (n = 5). ns non-significant, *P < 0.05, **P < 0.01 for one-way ANOVA in b, c, d, e, f with repeated measures. CFA complete Freund’s adjuvant, Panx1 pannexin 1, Panx1 KO pannexin 1 knockout, DRG dorsal root ganglion, GFAP-Cre glial fibrillary acidic protein-Cre, NFH-Cre neuro filament H-Cre, NF200 neurofilament 200, NGF nerve growth factor, WT wild-type

The more prominent role of neuronal Panx1 in paw sensitization than in orofacial pain might be due to differential expression in ganglia innervating the sites of inflammation. To test for this, we used staining for β-gal to assess Panx1 levels in sensory ganglia from mice in which glial or neuronal Panx1 was deleted. Results showed the absence of staining in TG and DRG from WT mice, which lack the β-gal transcript; however, in ganglia of both Panx1-nulls, β-gal staining was prominent in brightfield images (Additional file 1: Fig. S5a). SGC staining for glutamine synthetase aided in identifying sensory neurons that they encircled in the merged images (Additional file 1: Fig. S5a). Quantification of nine DRG and TG sections from three mice revealed significantly higher expression of the Panx1 reporter gene in the DRG than in the TG (Additional file 1: Fig. S5b) and significantly higher non-neuronal β-gal staining in the TG than in the DRG (Additional file 1: Fig. S5c). These results indicate that cellular expression levels of Panx1 are different in the TG and the DRG, with Panx1 being more prominent in DRGNs and Panx1 more prominent in TG glia. This difference likely contributes to the higher relevance of SGC Panx1 to orofacial pain and the greater contribution of neuronal Panx1 to hindlimb inflammatory hypersensitivity.


Panx1 contributes to many physiological and pathological processes, such as migraine headaches [45], spinal cord injury [33], sciatic nerve injury [32], cancer pain [46, 47], and inflammatory pain [48]. For example, blockade of Panx1 activation reduced cortical susceptibility to cortical spreading depression in migraine aura pathogenesis [45]; Panx1 blockers (10Panx peptide, carbenoxolone, and Pbcd) decreased mechanical hyperalgesia by inhibiting nocioceptor ion channels [NMDA receptors (NMDAR) and P2X7 purinoceptors (P2X7R)] in a spared nerve injury rat model [33]. These studies provide strong evidence that the Panx1 channel may provide a key element in the process of pain sensitization, both in the brain and in the periphery. The present study revealed that global/neuron-Panx1 deletion in mice rapidly increased pain thresholds in an experimental plantar inflammatory pain. These findings confirmed our previous report that Panx1 deletion provided substantial protection against the development and maintenance of allodynia in an orofacial inflammatory mouse model [17]. In an initial study on rats following sciatic nerve ligation [18], Panx1 was increased in DRG but not in the spinal cord dorsal horn during neuropathic pain development, implying that the locus of sensitization was within the DRG [21, 49]. Our investigation demonstrated that gene expression of Panx1 was dramatically elevated in DRG but not in the spinal cord from 7 to 21 d after CFA injection. Therefore, because cells in the DRG are accessible to drugs delivered systemically, there is enormous potential value in developing therapeutics that target the Panx1 channel in treating inflammatory pain.

Neurite regeneration has a strong correlation with pain sensitization, which is critical for alleviating neuropathic pain [50]. Concerning the role of neurite outgrowth in painful conditions, it is noteworthy that injury induces local branching of nociceptors in the skin [51]. Our study has revealed that neurite extension is retarded when Panx1 expression or function is blocked. However, these results are in contrast with those of other studies that examined the effects of Panx1 deletion or blockade on neurite extension. Prolonged (36 h) probenecid and Panx1 siRNA treatment led to increased neurite numbers in Neuro2a cells and ventricular zone neural progenitors, while transfection of neural progenitors with GFP-tagged Panx1 led to fewer cells with neurites [52]. The hypothesized mechanism for the negative impact of Panx1 on neurite expression involved microtubule cytoskeleton stabilization [53]. A recent study showing higher activity of dendritic protrusions in developing cortical neurons from Panx1-null mice (with rescue by Panx1 expression) was similarly interpreted as indicating increased process stability due to Panx1 binding to cytoskeleton [54]. One possible explanation for different results obtained from Panx1-null mice is that previous publications used transgenic mice whose coding region was only partially deleted and with a genetic background prone to passenger mutations. This effect has been demonstrated to result in the deletion of Caspase-4 and several other gene mutations in this transgenic Panx1 mouse [55]. In contrast, Panx1-null mice used in the present study were generated with the C57BL6 background by crossing mice obtained from the KOMP project [21]. These mice are hypomorphs, expressing deficient mRNA levels and having virtually no Panx1 function.

The changes that we observed in pain sensitivity and neuronal excitability in mice lacking functional Panx1 presumably result from altered signaling pathways linked to Panx1 function. One such pathway is Wnt signaling, where recent study has revealed that pathway components, including Wnt-3a, Frizzled 4, and β-catenin, rapidly increased at both mRNA and protein levels in rat L4–6 DRG in neuropathic pain [56]. Notably, β-catenin levels were found to be lower in periosteal bone from Panx1 KO mice compared to WT, indicating that mechanoresponsive Wnt signaling was impaired in Panx1-null osteocytes [57], and β-catenin expression declined when Panx1 was knocked down in melanoma cells [58, 59]. Consistent with these findings, our results confirmed that the β-catenin level was lower in the Panx1 KO DRGNs, compared to WT mice. Given the critical role of Wnt/β-catenin pathway activation in mediating neuronal-related gene expression, cell survival, and cytoskeletal organization [60,61,62], we tested β-catenin expression using the Wnt signaling agonist (LiCl) and inhibitors (IWP2, Triptonide) on WT and Panx1-null DRGNs. These studies revealed that Panx1 could modulate neurogenesis via activating the Wnt/β-catenin pathway in DRGNs, suggesting that this pathway may represent the downstream effector of neurogenesis changes. Thus, it can be speculated that Panx1 interacts with Wnt signals to enhance pain intensity.

Allodynia depends in part on peripheral sensitization, which involves neuronal hyperexcitability. It is well known that Neuro2a/DRG cells extend neurites and display inward Na+/Ca2+ currents in response to treatment with the differentiating agent RA and NGF [18, 35, 36, 63,64,65]. Our results demonstrated that, in both RA-treated Neuro2a cells and NGF-treated DRGNs, Panx1 was required for neurite extension and robust expression of cellular excitability. ATP, an allogeneic molecule, is released through Panx1 channels [58, 66], which might also contribute to pain sensitivity and allodynia. ATP release from cultured neurons and extracted DRGs was significantly enhanced in the hind paw CFA-induced inflammatory pain model, and this enhancement was absent in Panx1-null DRG and neurons cultured therefrom. Additionally, ATP-elicited Ca2+ elevations were blunted in both neurons and glia from Panx1-null mice, suggesting that ATP signaling might contribute to the phenotypic differences in Panx1-null and WT mice. These results are similar to those obtained in imaging studies of TG, in which submandibular CFA led to heightened Ca2+ responses to both K+ and ATP applications compared to controls [17]. These data indicated that Panx1 channels activate DRG neuron excitability via ATP.

The excitability of primary sensory neurons is pathologically enhanced following tissue injury and inflammation [67]. Recent investigations have indicated that pharmacological inhibition or genetic deletion of Panx1 strongly attenuated the excitability of CA1 pyramidal neurons [68] and ATP release in hippocampal slices [69], suggesting that Panx1 might also reduce the nociceptive threshold by affecting the transmission and persistence of such excitability in peripheral neuralgia. The present study data showed that Panx1 deletion reduced basal inward currents in both Neuro2a cells and DRGNs during neuronal differentiation, while current densities also decreased after Panx1 blockade by 1 mmol/L Pbcd. While supportive of the deletion study, Pbcd is an organic anion transport inhibitor that not only inhibits Panx1 but also affects other cell functions [70]. Generally, smaller diameter DRG cells are nociceptors and thermoreceptors, and larger high threshold mechanoreceptors become recruited into activity in pain sensitization. In this study, it was also observed that electrophysiology of small-diameter DRGNs was improved after Panx1 deletion, indicating that the Panx1 channels particularly impacted nociceptive C- and Aβ-fiber cells that mediate pain sensitization [33, 35, 65].

Our initial behavioral studies showed that pain thresholds were higher in global Panx1-deleted than in WT mice and that neuron-specific deletion also increased pain threshold, albeit not quite as effectively. Because we previously showed that glial-specific deletion effectively raised the pain threshold in an orofacial model [17], it was hypothesized that glial-targeted Panx1 deletion would provide at least partial relief from hypersensitivity. However, it did not significantly raise the threshold in the hind paw inflammatory pain model compared to that in WT mice. Given ATP release of DRGNs was much higher in WT than in Panx1 KO after CFA injection, we further compared neuron and glial sensitivity to ATP. Measurement of Ca2+ signaling in SGCs or neurons revealed that both cell types were hyperactive. Intracellular Ca2+ elevation evoked by ATP in DRGNs was higher and more rapid than in SGCs; however, both were diminished in Panx1-null cells. These findings indicate that reduced participation of glial cells in inflammatory signaling might also contribute to the lesser behavioral hypersensitivity in the knockout mice. Therefore, pain sensitivity has a close association with cell-specific Panx1 in DRG.

Further dendrite growth and cell excitability analysis supported the view that both neuronal and glial Panx1 might contribute to behavioral hypersensitivity in DRG following CFA treatment. On one hand, Panx1 deletion in glia reduced β-catenin expression, and this directly limited Wnt/β-catenin signaling induced neurite remodeling [40, 62, 71], resulting in a decrease in dendritic extension. On the other hand, neuronal Panx1 deletion evoked greater excitability of the small peripheral nociceptive cells than deletion in glia. These interesting data contrast with the striking efficacy of glial Panx1 in TG accompanying inflammatory orofacial pain [17], and indicate that the neuronal Panx1 is more prominent in controlling neurogenesis and hyperexcitability in the DRG than in the TG in inflammatory pain.

This difference likely arises from distinct properties of the neurons that innervate the hind paw and submandibular skin. The sensory neurons innervating these areas are located in the DRG and TG, respectively. Gene expression comparisons have revealed that significant differences in profiles, including a study comparing both transcriptomes of the ganglia and translatomes from Scn10a-positive nociceptors in both DRG and TG [34]. In that study, the overall level of Panx1 transcript was 2.6-fold higher in the DRG than in the TG, whereas translated mRNA was > 5-fold higher in the DRG than in TG nociceptors. Our demonstration of more abundant β-gal reporter activity in DRGNs than in TG neurons is consistent with these findings. Moreover, the higher β-gal reporter activity in non-neuronal cells in TG provides additional support for differential cellular involvement of Panx1 in these ganglia [34]. Although we found that neuronal Panx1 played a prominent role in controlling pain sensitization in inflammatory pain, the mechanisms underlying how neuronal Panx1 impacts synapse formation and nociceptive ion channels are still not known and need to be further studied. Collectively, our results revealed that the prominent role of neuronal Panx1 in neurogenesis and excitability following hindlimb CFA injection may provide a novel target for intervention in pain management.

Despite the dramatically lower pain sensitivity in mice lacking neuronal Panx1 in our study, interpretation is limited because the mechanistic analyses reported here were only performed in cell culture. Future studies should examine whether Panx1-related differences in neurogenesis exist in both DRG and receptive fields within the skin. Our findings also predict that Panx1-targeted therapy might benefit inflammatory pain, and future studies should optimize such pharmaceutical compounds.


The present study demonstrated that neuronal Panx1 is crucial for initiating and maintaining pain hypersensitivity following hind paw inflammatory injury. Mechanisms by which Panx1 facilitates pain sensitivity include enhanced neurite outgrowth, exaggerated cell excitability, and amplified ATP-induced calcium influx. Signaling via the Wnt/β-catenin pathway appeared to underlie Panx1 actions. In conclusion, the present study indicated that Panx1 is a prominent “pain driver” in peripheral hypersensitivity via cell-autonomous effects on neuronal excitability.

Availability of data and materials

The data used to support the findings of this study are available from the corresponding authors upon reasonable request.



Adenosine triphosphate




Complete Freund’s adjuvant




Dorsal root ganglion neurons


Enhanced green fluorescent protein


Ethidium bromide


Glia-specific Panx1 KO mice (GFAP-Cre: Panx1f/f)


Neurofilament 200


Neural-specific Panx1 KO mice (NFH-Cre: Panx1f/f)


Nerve growth factor

Panx1 KO:

Pannexin1 knock-out




pico ammeter




Retinoic acid


Satellite glial cells


Trigeminal ganglia




  1. Meacham K, Shepherd A, Mohapatra DP, Haroutounian S. Neuropathic pain: central vs. peripheral mechanisms. Curr Pain Headache Rep. 2017;21(6):28.

    Article  PubMed  Google Scholar 

  2. Ji RR, Nackley A, Huh Y, Terrando N, Maixner W. Neuroinflammation and central sensitization in chronic and widespread pain. Anesthesiology. 2018;129(2):343–66.

    Article  PubMed  Google Scholar 

  3. Gold MS, Gebhart GF. Nociceptor sensitization in pain pathogenesis. Nat Med. 2010;16(11):1248–57.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  4. Enthoven WTM, Roelofs PD, Koes BW. NSAIDs for chronic low back pain. JAMA. 2017;317(22):2327–8.

    Article  PubMed  Google Scholar 

  5. Morrone AL, Scuteri D, Rombola L, Mizoguchi H, Bagetta G. Opioids resistance in chronic pain management. Curr Neuropharmacol. 2017;15(3):444–56.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  6. Sullivan MD, Ballantyne JC. Understanding the risks of long-term opioid therapy for chronic pain. Am J Psychiatry. 2022;179(10):696–8.

    Article  PubMed  Google Scholar 

  7. Hörl WH. Nonsteroidal anti-inflammatory drugs and the kidney. Pharmaceuticals (Basel). 2010;3(7):2291–321.

    Article  PubMed  Google Scholar 

  8. Darcq E, Kieffer BL. Opioid receptors: drivers to addiction? Nat Rev Neurosci. 2018;19(8):499–514.

    Article  CAS  PubMed  Google Scholar 

  9. Zou S, Kumar U. Cannabinoid receptors and the endocannabinoid system: signaling and function in the central nervous system. Int J Mol Sci. 2018;19(3):833.

    Article  PubMed  PubMed Central  Google Scholar 

  10. Cichewicz DL. Synergistic interactions between cannabinoid and opioid analgesics. Life Sci. 2004;74(11):1317–24.

    Article  CAS  PubMed  Google Scholar 

  11. Gangadharan V, Kuner R. Pain hypersensitivity mechanisms at a glance. DMM. 2013;6(4):889–95.

    Article  PubMed  PubMed Central  Google Scholar 

  12. Krames ES. The dorsal root ganglion in chronic pain and as a target for neuromodulation: a review. Neuromodulation. 2015;18(1):24–32.

    Article  PubMed  Google Scholar 

  13. Mehta V, Snidvongs S, Ghai B, Langford R, Wodehouse T. Characterization of peripheral and central sensitization after dorsal root ganglion intervention in patients with unilateral lumbosacral radicular pain: a prospective pilot study. Br J Anaesth. 2017;118(6):924–31.

    Article  CAS  PubMed  Google Scholar 

  14. Campbell JN, Meyer RA. Mechanisms of neuropathic pain. Neuron. 2006;52(1):77–92.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  15. Noh MC, Stemkowski PL, Smith PA. Long-term actions of interleukin-1β on K+, Na+ and Ca2+ channel currents in small, IB4-positive dorsal root ganglion neurons; possible relevance to the etiology of neuropathic pain. J Neuroimmunol. 2019;332:198–211.

    Article  CAS  PubMed  Google Scholar 

  16. Basbaum AI, Bautista DM, Scherrer G, Julius D. Cellular and molecular mechanisms of pain. Cell. 2009;139(2):267–84.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  17. Hanstein R, Hanani M, Scemes E, Spray DC. Glial pannexin1 contributes to tactile hypersensitivity in a mouse model of orofacial pain. Sci Rep. 2016;6(1):38266.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  18. Zhang Y, Laumet G, Chen SR, Hittelman WN, Pan HL. Pannexin-1 up-regulation in the dorsal root ganglion contributes to neuropathic pain development. J Biol Chem. 2015;290(23):14647–55.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  19. Navis KE, Fan CY, Trang T, Thompson RJ, Derksen DJ. Pannexin 1 channels as a therapeutic target: structure, inhibition, and outlook. ACS Chem Neurosci. 2020;11(15):2163–72.

    Article  CAS  PubMed  Google Scholar 

  20. Spray DC, Hanani M. Gap junctions, pannexins and pain. Neurosci Lett. 2019;695:46–52.

    Article  CAS  PubMed  Google Scholar 

  21. Hanstein R, Negoro H, Patel N, Charollais A, Meda P, Spray D, et al. Promises and pitfalls of a pannexin1 transgenic mouse line. Front Pharmacol. 2013;4:61.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  22. Lohman AW, Isakson BE. Differentiating connexin hemichannels and pannexin channels in cellular ATP release. FEBS Lett. 2014;588(8):1379–88.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  23. Iglesias RM, Spray DC. Pannexin1-mediated ATP release provides signal transmission between neuro2a cells. Neurochem Res. 2012;37(6):1355–63.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  24. Nielsen BS, Toft-Bertelsen TL, Lolansen SD, Anderson CL, Nielsen MS, Thompson RJ, et al. Pannexin 1 activation and inhibition is permeant-selective. J Physiol. 2020;598(2):361–79.

    Article  CAS  PubMed  Google Scholar 

  25. Scemes E, Spray DC. Extracellular K+ and astrocyte signaling via connexin and pannexin channels. Neurochem Res. 2012;37(11):2310–6.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  26. Tressler AB, Penuela S, Laird DW, Dubyak GR. Chemotherapeutic drugs induce ATP release via caspase-gated pannexin-1 channels and a caspase/pannexin-1-independent mechanism. J Biol Chem. 2014;289(39):27246–63.

    Article  Google Scholar 

  27. Sandilos JK, Bayliss DA. Physiological mechanisms for the modulation of pannexin 1 channel activity. J Physiol. 2012;590(24):6257–66.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  28. Thompson RJ, Jackson MF, Olah ME, Rungta RL, Hines DJ, Beazely MA, et al. Activation of pannexin-1 hemichannels augments aberrant bursting in the hippocampus. Science. 2008;322(5907):1555–9.

    Article  CAS  PubMed  Google Scholar 

  29. Billaud M, Chiu YH, Lohman AW, Parpaite T, Butcher JT, Mutchler SM, et al. A molecular signature in the pannexin1 intracellular loop confers channel activation by the α1 adrenoreceptor in smooth muscle cells. Sci Signal. 2015;8(364):ra17.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  30. Boyce AKJ, Swayne LA. P2X7 receptor cross-talk regulates ATP-induced pannexin 1 internalization. Biochem J. 2017;474(13):2133–44.

    Article  CAS  PubMed  Google Scholar 

  31. Billaud M, Lohman A, Straub A, Looft-Wilson R, Johnstone S, Araj C, et al. Pannexin1 regulates 1-adrenergic receptor-mediated vasoconstriction. Circ Res. 2011;109:80–5.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  32. Weaver JL, Arandjelovic S, Brown G, Mendu SK, Schappe MS, Buckley MW, et al. Hematopoietic pannexin 1 function is critical for neuropathic pain. Sci Rep. 2017;7(1):42550.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  33. Bravo D, Ibarra P, Retamal J, Pelissier T, Laurido C, Hernandez A, et al. Pannexin 1: a novel participant in neuropathic pain signaling in the rat spinal cord. Pain. 2014;155(10):2108–15.

    Article  CAS  PubMed  Google Scholar 

  34. Salim M, Pradipta RR, Tavares-Ferreira D, Jamie KM, Ishwarya S, Andi W, et al. Differences between dorsal root and trigeminal ganglion nociceptors in mice revealed by translational profiling. J Neurosci. 2019;39(35):6829–47.

    Article  Google Scholar 

  35. Chang W, Berta T, Kim YH, Lee S, Lee SY, Ji RR. Expression and role of voltage-gated sodium channels in human dorsal root ganglion neurons with special focus on Nav1.7, species differences, and regulation by paclitaxel. Neurosci Bull. 2018;34(1):4–12.

    Article  CAS  PubMed  Google Scholar 

  36. Dai WL, Yan B, Bao YN, Fan JF, Liu JH. Suppression of peripheral NGF attenuates neuropathic pain induced by chronic constriction injury through the TAK1-MAPK/NF-κB signaling pathways. Cell Commun Signal. 2020;18(1):66.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  37. Cibelli A, Scemes E, Spray DC. Activity and stability of panx1 channels in astrocytes and neuroblastoma cells are enhanced by cholesterol depletion. Cells. 2022;11(20):3219.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  38. Cherkas PS, Hanani M, Smith DN, Spray DC, Suadicani SO, Zuckerman J. Bidirectional calcium signaling between satellite glial cells and neurons in cultured mouse trigeminal ganglia. Neuron Glia Biol. 2010;6(1):43–51.

    Article  PubMed  Google Scholar 

  39. Spray DC, Iglesias R, Shraer N, Suadicani S, Belzer V, Hanstein R, et al. Gap junction mediated signaling between satellite glia and neurons in trigeminal ganglia. Glia. 2019;67(5):791–801.

    Article  PubMed  PubMed Central  Google Scholar 

  40. Liu J, Xiao Q, Xiao J, Niu C, Li Y, Zhang X, et al. Wnt/β-catenin signalling: function, biological mechanisms, and therapeutic opportunities. Signal Transduct Target Ther. 2022;7(1):3.

    Article  PubMed  PubMed Central  Google Scholar 

  41. Chen B, Dodge ME, Tang W, Lu J, Ma Z, Fan CW, et al. Small molecule–mediated disruption of wnt-dependent signaling in tissue regeneration and cancer. Nat Chem Biol. 2009;5(2):100–7.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  42. Chinison J, Aguilar JS, Avalos A, Huang Y, Wang Z, Cameron DJ, et al. Triptonide effectively inhibits Wnt/β-catenin signaling via c-terminal transactivation domain of β-catenin. Sci Rep. 2016;6(1):32779.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  43. Eik LF, Naidu M, David P, Wong KH, Tan YS, Sabaratnam V. Lignosus Rhinocerus (cooke) ryvarden: a medicinal mushroom that stimulates neurite outgrowth in PC-12 cells. eCAM. 2012;2012:320308.

    PubMed  Google Scholar 

  44. Thi MM, Suadicani SO, Schaffler MB, Weinbaum S, Spray DC. Mechanosensory responses of osteocytes to physiological forces occur along processes and not cell body and require αVβ3 integrin. PNAS. 2013;110(52):21012–7.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  45. Bu F, Nie L, Quinn JP, Wang M. Sarcoma family kinase-dependent pannexin-1 activation after cortical spreading depression is mediated by NR2A-containing receptors. Int J Mol Sci. 2020;21(4):1269.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  46. Di Cesare Mannelli L, Marcoli M, Micheli L, Zanardelli M, Maura G, Ghelardini C, et al. Oxaliplatin evokes P2X7-dependent glutamate release in the cerebral cortex: a pain mechanism mediated by pannexin 1. Neuropharmacology. 2015;97:133–41.

    Article  PubMed  Google Scholar 

  47. Liu JA, Yu J, Cheung CW. Immune actions on the peripheral nervous system in pain. Int J Mol Sci. 2021;22(3):1448.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  48. Chen C, Huang X, Zhu W, Ding C, Huang P, Li R. TRPA1 triggers hyperalgesia and inflammation after tooth bleaching. Sci Rep. 2021;11(1):17418.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  49. Retamal MA, Alcayaga J, Verdugo CA, Bultynck G, Leybaert L, Sáez PJ, et al. Opening of pannexin- and connexin-based channels increases the excitability of nodose ganglion sensory neurons. Front Cell Neurosci. 2014;8:158.

    Article  PubMed  PubMed Central  Google Scholar 

  50. Wei J, Su W, Zhao Y, Wei Z, Hua Y, Xue P, et al. Maresin 1 promotes nerve regeneration and alleviates neuropathic pain after nerve injury. J Neuroinflammation. 2022;19(1):32.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  51. Wong C, Barkai O, Wang F, Perez CT, Lev S, Cai W, et al. mTORC2 mediates structural plasticity in distal nociceptive endings that contributes to pain hypersensitivity following inflammation. J Clin Invest. 2022;132(15):e152635.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  52. Wicki-Stordeur LE, Swayne LA. Panx1 regulates neural stem and progenitor cell behaviours associated with cytoskeletal dynamics and interacts with multiple cytoskeletal elements. Cell Commun Signal. 2013;11(1):62.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  53. Xu X, Wicki-Stordeur LE, Sanchez-Arias JC, Liu M, Weaver MS, Choi CSW, et al. Probenecid disrupts a novel pannexin 1-collapsin response mediator protein 2 interaction and increases microtubule stability. Front Cell Neurosci. 2018;12:124.

    Article  PubMed  PubMed Central  Google Scholar 

  54. Sanchez-Arias JC, Candlish RC, Slagt E, Swayne LA. Pannexin 1 regulates dendritic protrusion dynamics in immature cortical neurons. eNeuro. 2020;7(4):ENEURO.0079-20.2020.

  55. Berghe VT, Hulpiau P, Martens L, Vandenbroucke RE, Vandenbroucke RE, Wonterghem VE, Perry SW, et al. Passenger mutations confound interpretation of all genetically modified congenic mice. Immunity. 2015;43(1):200–9.

    Article  PubMed Central  Google Scholar 

  56. Zhao Y, Yang Z. Effect of wnt signaling pathway on pathogenesis and intervention of neuropathic pain. Exp Ther Med. 2018;16(4):3082–8.

    PubMed  PubMed Central  Google Scholar 

  57. Seref-Ferlengez Z, Urban-Maldonado M, Sun HB, Schaffler MB, Suadicani SO, Thi MM. Role of pannexin 1 channels in load-induced skeletal response. Ann N Y Acad Sci. 2019;1442(1):79–90.

    Article  CAS  PubMed  Google Scholar 

  58. Freeman TJ, Sayedyahossein S, Johnston D, Sanchez-Pupo RE, O’donnell B, Huang K, et al. Inhibition of pannexin 1 reduces the tumorigenic properties of human melanoma cells. Cancers (Basel). 2019;11(1):102.

    Article  CAS  PubMed  Google Scholar 

  59. Sayedyahossein S, Huang K, Li Z, Zhang C, Kozlov AM, Johnston D, et al. Pannexin 1 binds β-catenin to modulate melanoma cell growth and metabolism. J Biol Chem. 2021;296:100478.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  60. Maguschak KA, Ressler KJ. A role for Wnt/β-catenin signaling in the neural mechanisms of behavior. J Neuroimmune Pharmacol. 2012;7(4):763–73.

    Article  PubMed  PubMed Central  Google Scholar 

  61. Nusse R, Clevers H. Wnt/β-catenin signaling, disease, and emerging therapeutic modalities. Cell. 2017;169(6):985–99.

    Article  CAS  PubMed  Google Scholar 

  62. He CW, Liao CP, Pan CL. Wnt signalling in the development of axon, dendrites and synapses. Open Biol. 2018;8(10):180116.

    Article  PubMed  PubMed Central  Google Scholar 

  63. Wu PY, Lin YC, Chang CL, Lu HT, Chin CH, Hsu TT, et al. Functional decreases in P2X7 receptors are associated with retinoic acid-induced neuronal differentiation of Neuro-2a neuroblastoma cells. Cell Signal. 2009;21(6):881–91.

    Article  CAS  PubMed  Google Scholar 

  64. Phan CW, David P, Naidu M, Wong KH, Sabaratnam V. Neurite outgrowth stimulatory effects of culinary-medicinal mushrooms and their toxicity assessment using differentiating Neuro-2a and embryonic fibroblast BALB/3T3. BMC Complement Altern Med. 2013;13(1):261.

    Article  PubMed  PubMed Central  Google Scholar 

  65. Song XS, Huang ZJ, Song XJ. Thiamine suppresses thermal hyperalgesia, inhibits hyperexcitability, and lessens alterations of sodium currents in injured, dorsal root ganglion neurons in rats. Anesthesiology. 2009;110(2):387–400.

    Article  CAS  PubMed  Google Scholar 

  66. Dahl G, Qiu F, Wang J. The bizarre pharmacology of the ATP release channel pannexin1. Neuropharmacology. 2013;75:583–93.

    Article  CAS  PubMed  Google Scholar 

  67. Li T, Wang G, Hui VCC, Saad D, De Sousa Valente J, La Montanara P, et al. TRPV1 feed-forward sensitisation depends on COX2 upregulation in primary sensory neurons. Sci Rep. 2021;11(1):3514.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  68. Nicholas LW, Peter LT, Roger JT. Anoxia-induced NMDA receptor activation opens pannexin channels via src family kinases. J Neurosci. 2012;32(36):12579.

    Article  Google Scholar 

  69. Shan Y, Ni Y, Gao Z. Pannexin-1 channel regulates ATP release in epilepsy. Neurochem Res. 2020;45(5):965–71.

    Article  CAS  PubMed  Google Scholar 

  70. Silverman W, Locovei S, Dahl G. Probenecid, a gout remedy, inhibits pannexin 1 channels. Am J Physiol Cell Physiol. 2008;295(3):C761–7.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  71. Horton SM, Luna Lopez C, Blevins E, Howarth H, Weisberg J, Shestopalov VI, et al. Pannexin 1 modulates axonal growth in mouse peripheral nerves. Front Cell Neurosci. 2017;11:365.

    Article  PubMed  PubMed Central  Google Scholar 

Download references


We thank Dr. Mia M Thi, for instructions on the analysis of calcium imaging data, Dr. Sean McCutcheon for general advice, Dr. Sandra Veronica Lopez for offering assistance in immunohistochemistry, and Annie James for providing support with animal behavioral tests.


We would like to express our gratitude to all funding agencies that have made this study possible: NIH (R01NS092466), NSFC (U2004201), the Central Plains Thousand People Plan of Henan Province (204200510013), the Henan Overseas Expertise Introduction Center for Discipline Innovation (CXJD2021002), and the Key Special Project of Zhengzhou University Disciplinary Construction (XKZDJC202001).

Author information

Authors and Affiliations



DCS and FXG conceived and designed the experiments. QX prepared reagents and materials and performed most experiments. ES provided mice with global, neuron- and glia-targeted Panx1 deletion and assisted with behavioral analyses. QX and PD prepared and characterized the Panx1 deleted and transfected Neuro2a cell lines. QX, PD, and SM bred, set up mating and cared for mice. DCS, QHL, and AC performed part of the Patch-clamp experiments. QX and AJ performed the behavioral tests. QX, AC, GLY, and SVL performed and analyzed immunohistochemistry. AC, SM, and AJ performed parts of imaging experiments. QX and GLY analyzed the data. QX, FXG, and DCS wrote the manuscript. DCS, ES, AC, and FXG revised the final version of the paper. All authors read and approved the final manuscript.

Corresponding authors

Correspondence to Fang-Xia Guan or David C. Spray.

Ethics declarations

Ethics approval and consent to participate

The Institutional Animal Care and Use Committee (IACUC) of Einstein approved all animal experiments (NO. IACUCs20151101), assuring compliance with ethical and legal requirements for animal research.

Consent for publication

Not applicable.

Competing interests

The authors declare that they have no competing interests.

Supplementary Information

Additional file 1: Table S1

Primer pairs for qRT-PCR. Fig. S1 Detailed pain threshold analysis in WT, global Panx1 KO, NFH-Cre, and GFAP-Cre Panx1 KO mice after CFA inflammation. Fig. S2 Pain threshold baseline studies on all genotypes before CFA injection. Fig. S3 Neural differentiation studies on Panx1 gene-modified Neuro2a cells. Fig. S4 Detailed Sholl analysis on DRGNs from Panx1-deleted genotypes. Fig. S5 β-gal staining in DRG and TG tissue sections.

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit The Creative Commons Public Domain Dedication waiver ( applies to the data made available in this article, unless otherwise stated in a credit line to the data.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Xing, Q., Cibelli, A., Yang, G.L. et al. Neuronal Panx1 drives peripheral sensitization in experimental plantar inflammatory pain. Military Med Res 11, 27 (2024).

Download citation

  • Received:

  • Accepted:

  • Published:

  • DOI: