Humanized anti-P2X4 scFv reduces ATP-induced P2X4 currents and modulates excitability in human DRG neurons

Abstract

Chronic pain affects nearly 100 million adults in the U.S., yet few novel therapeutics have emerged in recent decades. P2X4 receptor (P2X4R), implicated in pain signaling, represents a promising target. We evaluated a humanized single-chain variable fragment (hscFv) targeting P2X4R for its ability to reduce ATP-induced currents and modulate excitability in human dorsal root ganglion (hDRG) neurons. Voltage-clamp recordings confirmed that human P2X4R (hP2X4R) hscFv significantly reduced ATP-evoked currents in HEK-293T cells expressing human P2X4, likely by relocalization of the receptor to the perinuclear region after hscFv treatment. Immunohistochemistry and transcriptomic analyses demonstrated widespread P2X4R (P2RX4) expression across hDRG neuronal subtypes in both male and female donors. Current-clamp recordings revealed that hP2X4R hscFv selectively increased action potential (AP) threshold in multi-firing hDRG neurons, without affecting single-firing neurons. Spontaneous activity at rest and depolarizing spontaneous fluctuation (DSF) amplitude were also reduced. Analysis confirmed consistent effects of hP2X4R hscFv on excitability parameters. These findings suggest that hP2X4 hscFv exerts modest but targeted effects on human sensory neurons, supporting its potential as a novel therapeutic for chronic pain.

Introduction

Chronic pain affects around 100 million individuals in the US and can arise from dysfunction or damage to the somatosensory nervous system.^1,2 Pain management has historically relied on NSAIDS, which can be contraindicated for some individuals, and opioids, which have the risk of addiction, tolerance, severe side effects, and opioid-induced hyperalgesia. ¹ Though a novel pain therapeutic, suzetrigine from Vertex pharmaceuticals, was recently approved by the FDA, prior to this last pain therapeutic approved was celecoxib in 1998.^3,4 With only one new pain therapeutic coming to market in the last 25 years, there is a need for novel pain therapeutics, and understanding the mechanisms of pain is an important step forward in developing patient-specific treatments.

One therapeutic target of interest has been the purinergic P2X family of receptors.^5,6 P2X receptors are a family of non-selective cation channels gated by extracellular ATP and several have been shown to play a role in pain and neuroinflammation.^7–9 P2X3R and P2X7R inhibitors have been tested in preclinical and clinical trials, with some trials reporting analgesic effects.^10–12 P2X4R is shown to increase in both the spinal dorsal horn and dorsal root ganglia (DRG) in many rodent models of pain, making it another attractive target for pain therapeutics.^13–15 Several P2X4R inhibitors have been tested in rodent neuropathic pain models and have been shown to decrease pain behaviors.^16–22

Several single-chain variable fragments (scFv) are used in the clinic or have been in clinical trials for the treatment of cancers, autoimmune disease, and Alzheimer’s. ²³ An scFv against P2X4R has been tested in a mouse chronic orofacial pain model and has shown long-lasting analgesic effects. ²⁴ A humanized version of this scFv, hP2X4 hscFv, has been developed based on the murine scFv.^25,26 Our goal was to test the effectiveness of hP2X4R hscFv on reducing ATP-induced currents of P2X4R and reducing neuronal excitability in human DRG (hDRG) neurons.

HEK-293T cells were stably transfected with human P2X4R (hP2X4R) and treated with hP2X4R hscFv. ATP was applied to these cells, and voltage-clamp experiments showed a decrease in ATP-evoked currents after hP2X4R hscFv pretreatment, showing that the hscFv reduces P2X4R activity. Immunohistochemistry and RNA sequencing (RNA-seq) experiments show that P2X4R protein is expressed in both male and female hDRG neurons. Using patch clamp electrophysiology current-clamp experiments with hDRG neurons, we showed that pretreatment with hP2X4R hscFv reduced neuronal excitability by increasing action potential (AP) threshold in multi-firing cells, decreasing the amplitude of depolarizing spontaneous fluctuations (DSFs), and decreasing the prevalence of spontaneous activity in hDRG neurons.

This study shows that hP2X4R hscFv could be an effective therapeutic in patients with chronic pain. The hP2X4R hscFv has targeted effects on the P2X4R by inhibiting ATP-evoked currents. Given the expression of P2X4R in hDRG neurons, we conclude that hP2X4R hscFv binding to P2X4R reduces neuronal excitability in hDRG neurons.

Methods

Humanized scFv generation

The hP2X4R hscFv was generated from a parental mouse scFv as described previously. ²⁴ Extensive methodologies are published in a patent and a recent publication.^25,26

HEK-293T cell culture

Human embryonic kidney (HEK-293T) cells were cultured in Dulbecco’s modified Eagle’s medium and 10% fetal bovine serum (Neuromics, Minneapolis, MN) containing penicillin, streptomycin, and amphotericin B (Gibco, cat# 15240062, Grand Island, NY). HEK-293T cells stably expressing human His-tagged P2X4R (hP2X4R-HEK-293T) were gifted from the Seguela lab (McGill University, Montreal, Canada). The cells were cultured in DMEM with 10% FBS, penicillin, streptomycin, amphotericin B, and G418 (250 µg/mL) for selection. ²⁷ For electrophysiology and immuno-fluorescence microscopy, cells were seeded on glass coverslips 24–48 h prior to experiments. Cells were either untreated or treated with 10 µg/mL hP2X4R hscFv for 1–3 h prior to recording or fixation.

Immunofluorescence microscopy

hP2X4R-HEK-293T cells were either untreated or treated with 10 µg/mL hP2X4R hscFv for 1–3 h prior to fixation. Cells were fixed with 4% PFA and 2% sucrose in PBS for 20 min. Cells were stained with 2 µg/mL wheat germ agglutin 640R (Biotium, Fremont, CA) in PBS for 20 min. Cells were then permeabilized with 0.1% Triton-X 100 in PBS for 10 min, then blocked with 1% BSA in PBS for 30 min. Cells were stained in blocking solution with 1:1000 anti-P2X4R antibody for 1 h (Alomone labs, Jerusalem, Israel) and 1:1000 Goat anti Rabbit AF488 (Thermo Fisher, Waltham, MA) for 1 h. Stained coverslips were mounted with Fluoromount-G with DAPI (Catalog number 00-4959-52, Thermo Fisher, Waltham, MA) and cured for at least 24 h before imaging. Widefield microscopy was performed on a Leica THUNDER 3D imager (NCI, Brooklyn Park, MN) equipped with an HC PL APO 40X/0.95 CORR objective, LED3 illumination system, GFP ET filter system, and a K5 Microscope camera. The Leica THUNDER is controlled by LAS X imaging software (Wetzlar, Germany). Confocal images were taken on a Fluoview FV1200 confocal microscope (Olympus, Center Valley, PA) controlled by Fluoview imaging software. Images were analyzed in Fiji ImageJ2 v1.54 software (NIH). All fixation, staining, and imaging were performed in parallel using identical microscope acquisition settings across all conditions to ensure comparability.

All statistical analysis was performed using GraphPad Prism v10.0.2 (Boston, MA). Error bars denote mean ± standard error of the mean (SEM) unless otherwise specified. Analysis methods are detailed in the figure texts.

hDRG neuron culture

hDRG neuron culture was done as previously described. ²⁸ hDRG were obtained from consented, recently deceased donors in coordination with New Mexico Donor Services at University of New Mexico Hospital with approval by the Human Research Review Committee and University of New Mexico Health Sciences Center, approval number #23-205. hDRG were minced and processed in NMDG-aCSF and then digested for 1–2 h in a collagenase/papain solution with trituration every 20 min. Digested DRG tissue was then passed through a 200 µm cell strainer and rinsed with hDRG culture media, Neurobasal Plus media (Gibco, cat #A3582901, Grand Island, NY) supplemented with 5% fetal bovine serum (Neuromics, cat #FBS007, Minneapolis, MN), 1% GlutaMAX (Gibco cat # 35050-061, Grand Island, NY), 2% B-27 supplement (Gibco, cat # 17504-001, Grand Island, NY), and 1% Anti-Anti (Gibco, cat # 15240062, Grand Island, NY). Cells were then plated on poly-D-lysine-coated glass coverslips and cultured in hDRG media for up to 11 days in vitro. Donor Demographics are described in Table 1.

Table 1.

Donor demographics.

Donor	Sex	Age	Ethnicity	Use
1	M	21	White	Ephys (n = 21), RNA
2	F	37	White/Hispanic	Ephys (n = 38), RNA
3	M	47	White	Ephys (n = 43)
4	M	51	White	IHC, RNA
5	F	23	White	IHC, RNA
6	M	63	White	RNA
7	M	34	Hispanic	RNA
8	F	22	White	RNA
9	F	51	White	RNA
10	M	23	Hispanic	RNA
11	M	56	Unavailable	RNA
12	M	67	Native American	RNA
13	M	55	White	RNA
14	M	27	Hispanic	RNA
15	M	68	White	RNA
16	M	39	White/Hispanic	RNA
17	M	23	White/Hispanic	RNA
18	F	37	White	RNA
19	M	21	White	RNA

hDRG samples were obtained from the above ethically consented human organ donors.

Bulk RNA sequencing of hDRG tissue

Total RNA was extracted from frozen hDRG tissue derived from donors (n = 14 males, n = 5 females) using the Monarch Total RNA Miniprep Kit (New England Biolabs, Ipswich, MA), following the manufacturer’s protocol. After ensuring RNA quality, libraries were produced from total RNA with rRNA removed or poly-A-captured mRNA (depending on the donor) and sequenced on a NovaSeq X Plus Sequencing System. Raw RNA sequencing reads were assessed for quality and filtered. Additionally, poly-G tails were trimmed before aligning reads to the GRCh38 reference genome using HISAT2. Transcript and gene expression were quantified using StringTie then raw counts underwent filtering by expression to ensure expression amongst most or all samples. Next, batch correction was performed using ComBat-Seq (in the sva Bioconductor package) to adjust for sequencing strategy (library preparation approach and sequencing depth). The following covariates were included in the correction model: sex, age (in 10-year bins), and ethnicity (white and/or Hispanic). TPM was calculated from the corrected transcript counts using transcript lengths derived from BioMart. P2RX4-202 was the only transcript variant that passed the step of filtering by expression, thus it represents P2RX4 expression in these hDRG tissues. The TPM of P2RX4 was compared between males and females using a Mann-Whitney U test. Additionally, the lack of sex differences in P2RX4 expression was confirmed via differential gene expression analysis using the quasi-likelihood generalized linear model approach in edgeR (data not shown).

Whole-cell patch-clamp electrophysiology

Whole-cell patch-clamp electrophysiology was performed as previously described. ²⁹ Recordings were done at room temperature, with the recording chamber perfused with artificial cerebrospinal fluid (aCSF) containing 113 mM NaCl, 3 mM KCl, 25 mM NaHCO₃, 1 mM NaH₂PO₄, 2 mM CaCl₂, 2 mM MgCl₂, and 11 mM D-glucose bubbled with 95% O2/5% CO2. HEK-293T and hP2X4-HEK-293T were identified with differential interference contrast optics connected to an IR-2000 digital camera (Dage MTI, Indiana City, MI) or an Olympus digital camera. Size was measured using Dage MTI camera software or ImageJ (NIH, Bethesda, MD). Voltage clamp recordings were performed using a Multiclamp 700B (Molecular Devices, San Jose, CA). Signals were filtered at 5 kHz, acquired at 50 kHz using a Digidata 1550B converter (Molecular Devices, San Jose, CA), and recorded using Clampex 11 software (Molecular Devices, San Jose, CA). Patch pipettes were made with a Zeitz puller (Werner Zeitz, Martinsried, Germany) from borosilicate thick glass (GC150F, Sutter Instruments, Novato, CA). Electrode resistance was 3 to 7 MΩ. Intracellular solution contained 120 mM K-gluconate, 11 mM KCl, 1 mM CaCl₂, 2 mM MgCl₂, 10 mM HEPES (4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid, 11 mM EGTA (ethylene glycol-bis(β-aminoethyl ether)-N,N,N′,N′-tetraacetic acid), 4 mM Magnesium-Adenosine triphosphate. Cell capacitance was calculated using the whole-cell capacitance compensation circuit in Multiclamp 700B (Molecular Devices). Cells were held at −60 mV. Series resistance compensation (Rs) was applied at 70%.

Continuous voltage-clamp recordings of HEK-293T cells were taken in gap-free mode. A 5-second 3 mL puff of 100 µM ATP in aCSF was applied directly to the cell, with a second pipette close to the patched cell. Current density was calculated by dividing the current (pA) of the cell during the gap-free recordings by the whole-cell capacitance (pF).

$$CurrentDensity(pA/pF)=\frac{Current(pA)}{Capacitance(pF)}$$

Analysis was performed in Easy Electrophysiology v.2.5.1 (London, UK) and Clampfit 11.2 (Molecular Devices, San Jose, CA). All statistical analysis was performed using GraphPad Prism v10.0.2 (Boston, MA). Error bars denote mean ± standard error of the mean (SEM) unless otherwise specified. Analysis methods are detailed in the figure texts.

Current-clamp recordings of hDRG neurons were done as previously described. ²⁹ First, a 500 ms current increasing stepwise from −100 pA in 10 pA increasing increment sweeps was applied to the patched hDRG neurons until the cell inactivated or reached a max current of 4 nA. Resting membrane potential (RMP) was calculated from the first sweep prior to any current injection. Input resistance (R_in) was calculated by finding the ΔSS, the RMP subtracted from the steady-state voltage (the average voltage of the last 100 ms of the −100 pA hyperpolarizing current injection), and dividing by −100 pA. Sag percentage was calculated by taking the minimum value during the first 300 ms of the −100 pA hyperpolarizing minus the steady-state voltage, dividing by ΔSS, and multiplying by 100. Rheobase was measured at the smallest current injection that elicits an action potential (AP), not including any rebound firing or spontaneous activity; thus the smallest possible value for rheobase is 10 pA. Neurons that fired >1 AP during any current injection step were considered multi-firing while neurons that only fired 1 AP were classified as single-firing. FSL was calculated as the time after current injection started on that sweep to the time of the rheobase spike. AP waveform properties were calculated in Easy Electrophysiology using Method II. ³⁰ Spontaneous activity and depolarizing spontaneous fluctuations (DSFs) were assessed in 30 s current-clamp gap-free recordings at rest or with enough current injection to hold the neuron at −45 mV. DSF prevalence, frequency, and amplitude were analyzed using FIBSI software. ³¹ All electrophysiological data analysis was done using pyABF ³² or Easy Electrophysiology v.2.5.1. Statistics and graph generation were done in GraphPad Prism v10.4.0.

Immunohistochemistry of hDRG

hDRG tissues were embedded in optimal cutting temperature (OCT) compound and sectioned at 40 µm on a cryostat. Sections were immediately placed into wells of a spotting plate filled with 1X TBS for free float immunohistochemistry. Sections were washed 3× in 1X TBS before being incubated for 1 h at room temperature in a blocking solution containing 0.3% Triton X-100 and 10% fish gelatin in 1X TBS. The sections were incubated at room temperature in a humidified chamber on a rocking table with the following primary antibodies and dilutions in blocking solution: chicken anti-peripherin (PA1-10012, Invitrogen, 1:2000) and rabbit anti-P2X4 (APR-002, Alomone Labs, 1:500). Primary anti-body solution was gently pipetted out of wells before being rinsed 5× in 1X TBS, followed by a 2-h incubation at room temperature in secondary antibodies in 1X TBS: goat anti-chicken AF555 (AB150170, Abcam, 1:2000) and donkey anti-rabbit AF488 (A21206, Invitrogen, 1:2000). The sections were then washed 5× in 1X TBS and treated with Sudan Black 0.05% in 70% ethanol solution for 30 min to reduce autofluorescence. Sections were then washed 5X in 1X TBS. Sections were transferred to labelled Fisherbrand^TM Superfrost^TM slides (12-550-15, Thermo Fisher) under microscope vision using a paintbrush. Tissue sections were allowed to dry overnight at room temperature in a dark location. Prior to placing coverslips, slides were gently dipped in de-ionized water to remove residual salts from the TBS. Cover slips were applied using Fluoromount-G Mounting Medium with DAPI (00-4959-52, Thermo Fisher). Immunofluorescence was visualized using a Leica THUNDER Imaging System (Leica Microsystems) at 10X magnification and stitched together using the integral tile-scan function within LasX.

Results

hP2X4R hscFv reduces ATP evoked currents in hP2X4R-HEK-293T cells

In order to demonstrate target engagement of the hscFv for hP2X4R, we performed experiments using a stable HEK line expressing hP2X4R. Whole-cell voltage clamp recordings on naked HEK-293T and hP2X4R-HEK-293T exposed to a short 5 s puff of 100 µM ATP show that naked HEK-293T cells do not respond to ATP (Figure 1(a)), but hP2X4R-HEK-293T elicit a large change in current density upon ATP exposure (Figure 1(a)). These data show that the only ATP-dependent channels in hP2X4R-HEK-293T are due to the overexpression of hP2X4R channels in that cell line. hP2X4R-HEK-293T treated with 10 µg/mL hP2X4 hscFv for 3 hours show a reduced current density response when compared to control hP2X4R-HEK-293T (Figure 1(a)).

Figure 1.

hP2X4R hscFv reduces ATP-evoked currents in hP2X4R-HEK-293T cell expression system. (a) Example traces of ATP-evoked currents (grey box) in naked HEK-293T cells, control hP2X4R-HEK-293T, and hP2X4R-HEK-293T treated with 10 µg/mL hP2X4R hscFv. (b) Change in current density of ATP-evoked currents of three consecutive ATP puffs at different treatment times of hP2X4R hscFv.

To further characterize the time course of inhibition, we performed a series of recordings on individual hP2X4R-HEK-293T cells treated with hscFv for 1, 2, or 3 h, each exposed to three consecutive 100 µM ATP puffs. Statistical analysis revealed a significantly greater reduction in current density after 2–3 h of treatment compared to 1 h, with only the first ATP response being affected between treatment times (Figure 1(b)). These findings indicate that a minimum of 2 h of hscFv treatment is required for optimal inhibition.

hP2X4R hscFv redistributes P2X4R channels to perinuclear region

hP2X4R-HEK-293T cells were either untreated or treated with 10 µg/mL hP2X4R hscFv for 1, 2, or 3 h, then fixed and stained with anti-P2X4R antibodies (Figure 2(a)). Cells were imaged with widefield microscopy, and the Mean Fluorescence Intensity (MFI) of the P2X4R antibody signal was compared (Figure 2(b)). No significant difference in total MFI was observed between control and treated cells, indicating that hP2X4R hscFv does not alter the overall abundance of P2X4R channels. Next we took confocal images of the untreated and 3 h hP2X4R hscFv treated cells stained with anti-P2X4R, stacked the confocal planes into a single image, drew a line across the cells from the outside of the cell, across the entire cell including the nuclear region, and measured the fluorescence intensity difference from the perinuclear region defined by the DAPI stain and the cytosolic region, represented by no DAPI (Figure 2(c), example trace Figure 2(d)). There was a significant difference in the change in MFI between the control and hscFv-treated cells (Figure 2(e)), showing that the P2X4R redistributes to the perinuclear region after hP2X4R hscFv treatment.

Figure 2.

hP2X4R hscFv redistributes hP2X4R channels to perinuclear region. (a) Confocal images of hP2X4R-HEK-293T stained with DAPI (blue), anti-P2X4R (green), and WGA (red). (b) Mean Fluorescence Intensity (MFI) of hP2X4R staining at different timepoints. (c) hP2X4R staining in untreated and hP2X4R hscFv-treated cells. (d) Example traces of MFI across the cell. (e) Intensity difference from perinuclear region and cell periphery in untreated and 3-h hP2X4R hscFv-treated cells.

hP2X4R is expressed in male and female hDRG neurons

Immunohistochemistry of hDRG from a male and female donor revealed widespread expression of hP2X4R in neuronal cell bodies (Figure 3(a)). hP2X4R sometimes, but not always, appeared to be co-expressed with peripherin, a marker of small-diameter neurons (Figure 3(a)). Additionally, hP2X4R expression was observed in axons, as indicated by distinct staining of fibers within the section. To confirm that expression of P2RX4, encoding hP2X4R, spans multiple neuronal subtypes, we referenced single-cell RNA-seq data from the Harmonized Cross-Species Atlas, which contains data integrated from several species, including three independent hDRG datasets.^33–36 This transcriptomic data shows that P2rx4 is expressed in multiple DRG neuronal subtypes (Figure 3(b)). ³³ Additionally, bulk RNA-seq analysis of our own donor cohort (n = 14 males, n = 5 females) revealed no significant sex differences in P2RX4 expression levels (Figure 3(c)), supporting the generalizability of P2RX4 expression across sexes.

Figure 3.

P2RX4/P2X4R is expressed in male and female hDRG neurons. (a) Representative immunohistochemistry images of human dorsal root ganglia (hDRG) tissue from a male and a female donor stained with antibodies against hP2X4R (green) and peripherin (magenta), a marker of small-diameter neurons. hP2X4R is broadly expressed in neuronal cell bodies and axonal fibers, with variable co-localization with peripherin. Scale bars: 500 µm. (b) Single-cell RNA-seq data integrated from the DRG of several species, including human, in the Harmonized Cross-Species Atlas confirm P2rx4 expression across multiple DRG neuron subtypes. ³³ Expression units are normalized and log-transformed counts. (c) Bulk RNA-seq of hDRG tissue samples from male (n = 14) and female (n = 5) donors shows no significant sex differences in P2RX4 expression.

hP2X4R hscFv produces differential actions on single- and multi-firing hDRG neurons

Using patch-clamp electrophysiology in current clamp mode, we compared electrophysiological features of control and hP2X4R hscFV-treated hDRG neurons. hP2X4 hscFv did not alter the prevalence of multi-firing or the number of APs elicited in multi-firing hDRG neurons (Figure 4(a) and (b)).

Figure 4.

hP2X4R hscFv reduces excitability in multi-firing hDRG neurons. (a) Representative current-clamp traces from multi-firing hDRG neurons—control (blue) or hP2X4R hscFv-treated (red)—showing every tenth sweep. (b) Maximum number of action potentials (APs) recorded in multi-firing hDRG neurons. (c) Prevalence of single- versus multi-firing hDRG neurons across conditions. (d and e) Rheobase, AP threshold, and AP decay time in multi-firing hDRG neurons. (f–h) Rheobase, AP threshold, and AP decay time in single-firing hDRG neurons. hP2X4 hscFv selectively increases AP threshold in multi-firing neurons, indicating reduced excitability.

However, hP2X4R hscFv treatment did affect specific excitability parameters in multi-firing hDRG neurons. While rheobase was unchanged (Figure 4(c), Table 2), AP threshold was significantly increased in treated cells (–28.69 ± 1.16 mV for control versus –24.16 ± 1.78 mV for hscFv-treated, p = 0.0319 by unpaired t-test) (Figure 4(d), Table 2), indicating reduced neuronal excitability. Treatment also did not have a significant effect on AP decay time in multi-firing cells (Figure 4(e), Table 2). AP threshold is a well-established measure of neuronal excitability, with higher values reflecting reduced excitability.^28,31 These findings suggest that hP2X4R hscFv decreases excitability specifically in multi-firing hDRG neurons by increasing the AP threshold. Importantly, no other electrophysiological features were significantly different after hP2X4R hscFv treatment (Table 2).

Table 2.

Electrophysiological properties of hDRG neurons under control or hP2X4R hscFv-treated conditions.

Multi-Firing hDRG
Electrophysiological property	Control (n = 36)	hP2X4 hscFv (n = 30)	p-value
RMP (mV)	−54.22 ± 1.47	−54.77 ± 1.41	0.6246
Rheobase (pA)	373.6 ± 66.46	608.3 ± 117.7	0.1155
Rin (MΩ)	199.7 ± 30.8	209.6 ± 35.8	0.7389
FSL (ms)	63.50 ± 2.50	36.50 ± 2.50	0.5843
Sag Percentage (%)	14.78 ± 2.16	16.37 ±± 2.49	0.5036
AP Threshold (mV)	−28.69 ± 1.16	−24.16 ± 1.78	* 0.0319
AP Peak (mV)	56.97 ± 1.54	56.19 ± 1.98	0.7956
AP Amplitude (mV)	85.66 ± 1.97	80.35 ± 2.63	0.0899
AP AHP (mV)	−27.38 ± 1.29	−29.21 ± 1.24	0.3446
AP Rise Time (ms)	1.473 ± 0.132	1.550 ± 0.179	0.9324
Max Rise Slope (dVm)	129.6 ± 10.19	116.0 ± 9.425	0.4480
AP Decay Time (ms)	12.62 ± 1.36	11.33 ± 1.73	0.1188
Max Decay Slope (dVm)	−21.79 ± 2.02	−22.51 ± 1.68	0.3612
Single-Firing hDRG
Electrophysiological property	Control (n = 21)	hP2X4 hscFv (n = 15)	p-value
RMP (mV)	−54.78 ± 2.01	−57.21 ± 1.02	0.8245
Rheobase (pA)	1520 ± 209	1173 ± 189	0.2385
Rin (MΩ)	43.4 ± 8.0	112.1 ± 36.1	0.0534
FSL (ms)	54.60 ± 25.44	13.13 ± 0.89	0.2414
Sag Percentage (%)	39.12 ± 4.87	36.42 ± 8.42	0.3567
AP Threshold (mV)	−26.13 ± 1.77	−24.28 ± 1.36	0.3773
AP Peak (mV)	52.37 ± 2.83	53.51 ± 2.13	0.7861
AP Amplitude (mV)	79.75 ± 3.50	77.37 ± 2.91	0.2572
AP AHP (mV)	−29.67 ± 1.55	−32.95 ± 1.64	0.1084
AP Rise Time (ms)	1.100 ± 0.139	1.220 ± 0.192	0.6977
Max Rise Slope (dVm)	144.9 ± 13.8	126.6 ± 15.8	0.3863
AP Decay Time (ms)	7.835 ± 1.556	6.607 ± 1.495	0.7987
Max Decay Slope (dVm)	−30.56 ± 3.64	−31.01 ± 3.55	0.8823

Electrophysiological properties for multi-firing control (n = 36) and hP2X4R hscFv-treated (n = 30) hDRG neurons, single-firing control (n = 21) and hP2X4R hscFv-treated (n = 15) hDRG neurons. ^*p<0.05 by Mann-Whitney test.

In contrast, hP2X4R hscFv treatment had no significant effect on single-firing hDRG neurons in terms of rheobase, AP threshold, or AP decay time (Figure 4(f)–(h)), indicating that the functional impact of hscFv is selective for multi-firing neurons.

hP2X4R hscFv reduces DSF amplitude and spontaneous activity

To assess the impact of hP2X4R hscFv on subthreshold activity, we performed gap-free current-clamp recordings in hDRG neurons. We analyzed depolarizing spontaneous fluctuations (DSFs) and spontaneous action potential firing (representative traces shown in Figure 5(a); blue arrows indicate DSFs, red arrows indicate spontaneous APs).

Figure 5.

hP2X4R hscFv reduces DSF amplitude and spontaneous activity in hDRG neurons.

Spontaneous activity at rest was also affected: 6.5% of control neurons (4/61) exhibited spontaneous APs at resting membrane potential, while none of the hscFv-treated neurons (0/49) did (p = 0.0140) (Figure 5(b)). In contrast, spontaneous activity at a depolarized membrane potential (–45 mV) was not significantly different between groups (Figure 5(c)).

hP2X4R hscFv treatment did not significantly alter the prevalence or frequency of DSFs (Figure 5(d) and (e)). However, DSF amplitude was significantly reduced in treated neurons (2.22 ± 0.07 mV in control versus 1.82 ± 0.03 mV in hscFv-treated cells, p < 0.0001) (Figure 5(f)), indicating a dampening of subthreshold excitability.

Together, these data demonstrate that hP2X4R hscFv reduces DSF amplitude and suppresses spontaneous activity at rest, further supporting its role in decreasing hDRG neuronal excitability.

Discussion

P2X4R has been previously shown to be a potential target for the treatment of pain in various rodent pain models, including diabetic peripheral neuropathy, orofacial pain, spared nerve injury, and chronic constriction injury.^{13,14,16–22,24} Many of the P2X4R therapeutics tested have been short-acting and needed to be administered consistently for long-acting pain relief.^14,16,18,19 Previous work has shown that a single dose of a murine P2X4R scFv was able to block orofacial pain long-term. ²⁴

Here we tested whether a humanized P2X4R hscFv influenced neuronal excitability in dissociated hDRG neurons. First, we validated that the hP2X4R hscFv was able to reduce ATP-induced currents in HEK cells overexpressing hP2X4R. We also showed that P2X4R is expressed hDRG neurons. Immunohistochemical and transcriptomic analyses confirmed P2RX4/P2X4R expression in hDRG neurons. Furthermore, no significant sex difference was noted using bulk RNA-seq of whole hDRG. This is consistent with previous reports of P2RX4 expression in hDRG neurons.^33,35

Electrophysiological recordings demonstrated that hP2X4R hscFv treatment selectively modulates excitability in multi-firing hDRG neurons. Specifically, we observed an increase in AP threshold and a reduction in DSF amplitude and frequency, suggesting decreased neuronal excitability. The effect on AP threshold was not observed in single-firing neurons, indicating a degree of functional specificity. Importantly, these findings were consistent across donors when data were pooled, regardless of sex. While previous studies have reported sexually dimorphic responses to P2X4R inhibition in rodent models,^13,24,37,38 our human data cannot support strong sex-based conclusions due to limited female donor representation and variability in pain history. Thus, while sex differences may exist, further studies with larger and more balanced cohorts are needed to explore this possibility.

Our findings also suggest a potential mechanism by which hP2X4R hscFv may reduce pain: by suppressing DSFs and spontaneous activity, both of which have been associated with chronic pain states in human DRG neurons. ³¹ Although spontaneous activity was infrequent in our donor samples (<10%), hP2X4R hscFv-treated neurons showed no spontaneous firing at rest, supporting its potential to dampen excitability. hDRG neurons derived from pain patients show higher DSF amplitude, suggesting that DSFs may be an electrophysiological property associated with chronic pain. ³¹ Moreover, DRG neurons with spontaneous activity have higher DSF frequency and amplitude compared to cells with no spontaneous activity, linking these DSF features to pain-associated excitability via spontaneous activity. ³¹ P2X4R is an ATP-gated non-selective cation pore, ⁶ and calcium ions, sodium ions, and several ion channels have been shown to have effects on DSF prevalence, frequency, and amplitude. ³⁹ It is well established that P2X4 receptors are cation channels and can for example generate calcium currents that can broadly influence neuronal excitability. ⁴⁰ Taken together, these data suggest that cations may be partially responsible for the reduction in spontaneous activity and DSFs in hDRG neurons after hP2X4R hscFv treatment.

In conclusion, our data support the therapeutic potential of hP2X4R hscFv in reducing the excitability of hDRG neurons, particularly multi-firing neurons. While baseline effects were modest, the selective modulation of excitability parameters suggests that hP2X4R hscFv may be a viable candidate for further development as a pain therapeutic. Given the long-lasting effects observed with murine P2X4R scFv in preclinical models, ²⁴ the humanized version may offer a novel approach for treating chronic pain in humans.