TRPV1 and TRPA1 channels exhibit bifurcated sensing of singlet oxygen and hydrogen peroxide
Yunshen Chen, Gaogao He, Wei Zhang, Jiajie Li, Xiaoxi Li, Sijun Dong, Qinglian Liu, Lei Zhou
TL;DR
TRPV1 and TRPA1 channels react differently to singlet oxygen and hydrogen peroxide, showing unique patterns of activation and inhibition.
Contribution
The study reveals distinct and bifurcated sensing mechanisms of TRPV1 and TRPA1 to two reactive oxygen species.
Findings
TRPV1 is enhanced by singlet oxygen through faster opening and increased current.
TRPA1 is initially activated but then permanently inhibited by singlet oxygen.
TRPA1 is more sensitive to hydrogen peroxide than TRPV1, involving intracellular cysteine residues.
Abstract
By responding to stimuli of diverse physics and chemical nature, transient receptor potential (TRP) channels fulfill important physiological functions in both excitable and non-excitable cells. Capsaicin (CAP) from chili peppers and allyl isothiocyanate (AITC) from mustard are natural and potent agonists for TRPV1 and TRPA1, respectively. Upon exposure to hydrogen peroxide (H2O2), the central molecule in redox signaling pathways, TRPA1 shows robust activation and much higher sensitivity than TRPV1. Singlet oxygen (1O2), the molecular oxygen in electrically excited states, is the least studied reactive oxygen species (ROS). Here we report that both TRPV1 and TRPA1 are sensitive to the modification by 1O2, but they exhibit drastically different responses. 1O2 generated by excited photosensitizers enhances the function of TRPV1 by accelerating its opening kinetics, increasing the current…
Genes, proteins, chemicals, diseases, species, mutations and cell lines named across the full text — each resolved to its canonical identifier and authoritative record.
Click any figure to enlarge with its caption.
Figure 1
Figure 2
Figure 3
Figure 4
Figure 5
Figure 6
Figure 7
Figure 8
Figure 9
Figure 10Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Videos
No videos yet. Explain this paper in a talk, walkthrough, or lecture? Add one.
Taxonomy
TopicsIon Channels and Receptors · Calcium signaling and nucleotide metabolism · Genomics, phytochemicals, and oxidative stress
Introduction
1
The reduction-oxidation (redox) equilibrium plays a versatile, intricate, and still underappreciated role in almost every aspect of cellular physiology [1]. Acting as “redox switch” through protein modifications, reactive oxygen species (ROS) and reactive nitrogen species (RNS) directly impact signal transductions by controlling the function of enzymes, channels, receptors, and transcription factors. Primary ROS include superoxide anion and hydroxyl radicals, which are free radicals and contain unpaired electrons, and non-radicals including hydrogen peroxide (H_2_O_2_) and singlet oxygen (^1^O_2_). It is generally assumed that H_2_O_2_ plays a central role in the redox signaling network [2]. Proteins rich in thiol-containing cysteine residue are prone to H_2_O_2_ modification, which results in both reversible and irreversible products [3]. Extensive research efforts have been put in to investigate H_2_O_2_'s lifecycle, transport, distribution, reaction mechanisms, and its participated physiological functions from embryogenesis, muscle contraction, neuronal activity, to immune responses.
By contrast, the current understanding of ^1^O_2_, particularly regarding its physiological function as a signaling factor in animal cells and the underlying biophysics of protein modification, lags that of other ROS. ^1^O_2_ is the molecular oxygen in electrical excited states [4,5]. ^1^O_2_ does not carry any charge so that steric effects dominate the interaction between ^1^O_2_ and targeted protein residues. Still, ^1^O_2_ is highly reactive and oxidizes a wide range of molecules, including protein, DNA/RNA, and unsaturated lipids in the cell. Low levels of ^1^O_2_ are believed to contribute to cell signaling, while large quantity of ^1^O_2_ is cytotoxic and leads to cell death.
Biological production of ^1^O_2_ can be through either the photodynamic process, which requires photosensitizer, oxygen, and light [6], or metabolic processes involving other ROS and enzymes [7,8]. In cells exposed to sunlight, such as those in the skin and eyes, intracellular compounds including flavins and NADH/NADPH serve as efficient photosensitizers. The ^1^O_2_ produced under sunlight, especially UVA light (320-400 nm), has been linked to aging and diseases in these cells [[9], [10], [11], [12]]. Additionally, enzymatic generation of ^1^O_2_ has been observed in stimulated neutrophils and macrophages, and contributes to the regulation of vascular tone and blood pressure [[13], [14], [15]]. However, several factors make ^1^O_2_ a challenging research target: its volatile chemical nature, the wide range of target molecules, and the heterogeneous distribution of oxygen, photosensitizers, other ROS, related enzymes, and ^1^O_2_ quenchers.
Ion channels, located on the cell surface or within the cell, function as the molecular basis for all electrical activities pertinent to physiology. The function of ion channels can be monitored with high temporal resolution in sub-millisecond timescale, making them ideal models for mechanistic studies. Transient receptor potential (TRP) channels are tetrameric, non-selective cation channels broadly distributed on the cytoplasmic membrane and membrane of intracellular organelles. Each TRP channel subunit contains a transmembrane domain containing six α-helixes, in which S5, S6, and the sequence in between form the ion conducting pore, and intracellularly exposed N- and C-termini [16]. Based on the similarity of intracellular N- and C-termini, TRP channels can be separated into seven subfamilies: TRPA (ankyrin), TRC (canonical), TRPM (melastatin), TRPML (mucolipin), TRPN (NOMPC), TRPP (polycystic), and TRV (vanilloid) [17,18]. Human TRP channels are broadly expressed in both excitable cells, including neurons and muscle cells, and non-excitable cells, especially immune cells. The diverse physiological functions played by TRP channels, from sensations of taste, smell, temperature, pain, to immune responses, are mostly rooted in their calcium permeability. Indeed, through intracellular calcium ions, TRP channels have intricate communications with other members within the calcium signaling pathway.
The opening and closing of TRP channels are under the control of remarkably diverse factors, including membrane potential, chemical compounds like CAP, AITC, and certain lipids, calcium ions and calcium binding proteins, temperature, pH, phosphorylation, and redox modification [[19], [20], [21]]. Several TRP channels, including TRPA1, TRPV1, TRPM2, and TRPC5, are sensitive to the modification mediated by ROS and RNS [[22], [23], [24]]. Among them, TRPA1 channels exhibit robust activation upon exposure to H_2_O_2_, mostly through the oxidation of cysteine residues located in its N-terminus [23,25,26]. In contrast, H_2_O_2_ by itself cannot directly open TRPV1 but potentiate TRPV1's response to other stimulations including heat, CAP, and protons, which is a kinetically slow process and also involves cysteine residues [[27], [28], [29]]. For the TRPM2 channel, a single methionine residue, M215, is critical for the sensitization to heat after H_2_O_2_ exposure [24].
In this study, we investigate the responses of TRPV1 and TRPA1, two well-characterized TRP channels, to ^1^O_2_ -mediated photodynamic modification (PDM) and the differences in their responses to H_2_O_2_. TRPV1 and TRPA1 are sensitive to both ROS but exhibit drastically different changes in channel function. ^1^O_2_ modification exerts profound but totally opposite effects on the functions of TRPV1 and TRPA1, while H_2_O_2_ is a potent agonist for TRPA1 but moderately enhances the function of TRPV1. Our study provides essential insights into the modification of TRPV1 and TRPA1 by ^1^O_2_ and sheds light on the related physiological processes.
Materials and methods
2
Plasmid construction
2.1
The cDNA sequences encoding, hTRPV1 and hTRPA1 channels were obtained from a human cDNA library (Human ORF gene library, Dharmacon, Accession List: BC009731, BC009731.1) and subcloned into vectors for expression in mammalian cells (pCDNA5, pLD2577 kindly provided by Dr. Lin Deng at SZBL, pCDNA3.4). A homologous recombination cloning kit (ClonExpress II One Step Cloning, #C112) and the design of recombination primers were provided by the same company (Vazyme Biotech). A genetically encoded photosensitizer, SOPP3 [30], was subcloned to the N-terminus of hTRPV1 (after L110) or N-terminus of hTRPA1 (after D60) (Fig. S1).
Cell culture and electrophysiology experiments
2.2
Standard cell culture and transfection procedures were used. HEK293T (ATCC, CRL-3216) cells were maintained at 37 °C and 5% CO_2_ and split the day before transfection. When cell confluency reached 70-80%, cells were transfected with PEI (YEASEN Life Science, 40816ES03) with the DNA:PEI ratio approximately 2 μg:4 μl for a 35 mm dish. Electrophysiology recording and calcium imaging experiments were carried out between 48 and 60 h after transfection. Transfected cells were digested using 0.25% Trypsin and then plated onto glass coverslips for channel function assays. After 5-10 min of incubation, the coverslips with attached cells were placed in a chamber filled with extracellular solutions for electrophysiology experiments.
Glass pipettes (JITIAN Bio, JT0062) were prepared using a micropipette puller (P-1000, Sutter Instrument), with the resistance of 3 to 6 MΩ. The configuration of whole-cell patch-clamp recording was used for most macroscopic current recordings. Electrical currents were amplified with Axopatch 200B (Molecular Devices), digitized by a multifunction I/O device (USB 6380, National Instruments), and recorded by the WinWCP program (Dr. John Dempster, University of Strathclyde). The lowpass filter on the amplifier was set at 10 kHz and the data sampling rate was set at 20 kHz. Raw current traces were converted to the Axon ABF V1.6 format and processed by the Clampfit program (Molecular Devices, V10.7).
Calcium imaging
2.3
HEK293T cells were seeded into a 24-well cell culture plate. The plasmids encoding hTRPV1, hTRPV1-SOPP3, and hTRPV1-SOPP3 containing single-point mutations (H65A, H143A, H167A, H207A, H233A, H290A, H321A, H365A, H379A, H411A, H533A, H614A, H787A, C578A, W273A, T307A, Y310A, M335A) were co-transfected with the plasmid for GCaMP6s and expressed for 48 h before measurements. Before imaging, each well was gently rinsed twice with the specified solution: calcium-containing solution (in mM), 150 NaCl, 1 MgCl2, 10 Glucose, 1.5 CaCl2, 10 HEPES, pH 7.4; EGTA calcium-free solution (in mM), 152 NaCl, 1 MgCl2, 10 Glucose, 10 HEPES, 2 or 1 EGTA, pH 7.4. CAP (1 μM) and AITC (100 μM) were prepared using the above solutions.
Calcium imaging was performed on Olympus inverted microscope (IX73) equipped with a 6500K LED light source (Thorlabs, MCWHLP3), a red fluorescence filter set (EX 590/25, EM 610LP, DM 605), and a green fluorescence filter set (EX 490/20, EM 535/45, DM 505), and three objective lenses (Olympus, LCAch N 10 × /0.25, LCAch N 20x/0.4, LCAch N 40 × /0.60). Fluorescence images were acquired using an iXon Life 888 EMCCD camera (Andor). Unless otherwise specified, all calcium imaging experiments were conducted using the 20 × objective lens. Image data was collected using the software Micro-Manager 2.0 with the exposure time set at 500 msec.
Fluorescence images were processed using ImageJ (http://rsbweb.nih.gov/ij/). The background fluorescence intensity was defined as the fluorescence value from a region free of cells and subtracted from measured intensities for all images, ΔFi, with i referring to frame number or time points. The fluorescence intensity of the image collected 5 s after light illumination being turned on was chosen as the baseline fluorescence of GCaMP6s, ΔFo, and used in the calculation of ΔΔFi: ΔΔFi = (ΔFi-ΔFo)/ΔFo. Each experiment was repeated three or four times (100 –200 cells per field), and representative data are presented as mean ± SEM.
Mass spectroscopy
2.4
HEK293 cells transfected with TRPV1-SOPP3 were irradiated with a 455 nm LED light for 5 min, while the cells that received no light treatment were used as control. The excitation light source consists of two arrays of LEDs (100 W, 450-455 nm). A photodiode power sensor (Thorlabs, S121C) and a power meter (Thorlabs, PM100D) were employed for intensity measurement. At the position corresponding to the center of the cell dish, the light intensity was approximately 32.0 mW/cm^2^. After light exposure, cells were washed with pre-chilled PBS and dislodged from the dish by a cell scraper. Then the cell pellet was resuspended with pre-cooled lysis buffer supplemented with PMSF (1 mM; Sangon, A100754) and Protease and Phosphatase Inhibitor Cocktail (1:100, MCE, HY-K0013). The lysis buffer contains 10 mM Tris (pH 7.5), 150 mM NaCl, 0.5% NP-40, and 1% Triton X-100. The cell suspension was incubated on a rotating mixer at 4 °C for 1 h and then centrifuged (20,000×g, 4^o^C 10 min) to remove cell debris. 50 ml supernatants were collected and incubated with pre-washed anti-mCherry affinity agarose (DIA-AN, IP0099S) overnight at 4 °C. After extensive washing, proteins were eluted with pre-chilled acidic elution buffer (0.15 M Glycine-HCl buffer, pH 3.5), followed by neutralization with Tris-HCl (1 M, pH 8.0) and precipitation with methanol and chloroform (methanol:chloroform:sample, 4:1:4) at −20^o^C for more than 2 h. After centrifugation (15000 rpm, 4 °C, 10 min), the pellets were washed with methanol five times and then dried.
The protein pellet was resolubilized and denatured in 8 M urea/25 mM ammonium bicarbonate with sonication. DTT (10 mM, 55^o^C, 40 min) and iodoacetamide (IAA, 15 mM, room temperature, 30 min) were sequentially added to reduce disulfide bonds and alkylate the cysteine residues, respectively. Then, NH_4_CO_3_ (25 mM) was added to reduce the urea concentration. CaCl_2_ (1 mM) and Trypsin (4.7 mg/l) were added, with the mixture being incubated at 37^o^C and 1000 rpm for 12-16 h. Trypsin digestion was stopped by incubation with formic acid (pH 2-3) on ice. After protein concentration determination, samples were desalted with a pre-equilibrated C18 column (Thermo, 60108-302), eluted with elution solution (80% CAN, Sigma-Aldrich, 1.00029.2500), and then air-dried. Finally, the dried peptides were resuspended in 15 μL of 0.1% mass spectrometry-grade formic acid. Peptide suspension at the concentration of 250 ng/μL was analyzed using a Quadrupole-Orbitrap Mass Spectrometer (Thermo, Orbitrap Exploris 480).
Solutions and reagents
2.5
Internal solution (electrophysiology): 150 mM NaCl, 3 mM MgCl_2_, 5 mM EGTA, 10 mM HEPES, pH adjusted to 7.2 with NaOH, at 25 °C.
External solution: 150 mM NaCl, 6 mM CsCl, 1 mM MgCl_2_, 1.5 mM CaCl_2_, 10 mM glucose, 10 mM HEPES, pH adjusted to 7.4 with NaOH, at 25 °C.
Extracellular solution for calcium imaging contained 150 mM NaCl, 1 mM MgCl_2_, 1.5 mM CaCl_2_, 10 mM glucose, and 10 mM HEPES, adjusted to pH 7.4 with NaOH, at 25 °C. EGTA was added as specified (with Ca^2+^ removed) to make calcium free solutions.
All solutions were filtered through a 0.22 μm membrane and sonicated in a 40 Hz water bath for 10 min prior to use. Except for glucose and EGTA (purchased from Sangon Biotech), all chemicals were obtained from Sigma-Aldrich.
Drug preparation: Capsaicin (Sangon Biotech, Cat# A429789-0100) was dissolved in DMSO to prepare a 20 mM stock solution and diluted in the above extracellular solution to the desired working concentration before use. Allyl isothiocyanate (AITC; Supelco, Cat# 36682) was dissolved in DMSO to prepare a 100 mM stock solution and similarly diluted in extracellular solution to the target concentration. rose Bengal (Sigma-Aldrich, Cat# 330000-1G) was dissolved in DMSO (20 mM). Carvacrol (APExBIO, Cat# C6244-50) was dissolved in DMSO (500 mM).
Other reagents used in the experiments included 3% hydrogen peroxide (H_2_O_2_; PYTHONBIO, Cat# AAPR555-I500), potassium iodide (KI; Aladdin, Cat# P128455-1L), Trolox (HarveyBio, Cat# 53188-07-1), and 0.1 mM dilute sulfuric acid solution (Fulin, Cat# A00027-500 ml).
Statistics
2.6
All statistical tests were performed using the programs of OriginPro and Prism. Data are presented as mean ± SEM. Statistical significance was assessed with unpaired or paired student t-test. p < 0.05 was considered statistically significant. ∗, p < 0.05; ∗∗, p < 0.01; ∗∗∗, p < 0.001.
Results
3
Photodynamic modification (PDM) facilitates TRPV1 channel activation
3.1
To examine the PDM effects on hTRPV1, we engineered SOPP3, an engineered protein that utilizes flavin mononucleotide (FMN) (Fig. S2[31][32]) as a photosensitizer to generate singlet oxygen, in the N-terminus of hTRPV1 and before the ankyrin domain (Fig. 1A; structure predicted by AlphaFold3, Fig. S1A; protein sequence, Fig. S1B). Without any photosensitizer, hTRPV1 is not sensitive to light and responds normally to membrane depolarization and CAP (Fig. 1B, top). However, for hTRPV1-SOPP3, light application effectively accelerated channel opening and increased the macroscopic current amplitude (Fig. 1B, bottom; Fig. 1C).Fig. 1PDM of human TRPV1 (hTRPV1) channel concatenated with SOPP3.A. A schematic model showing the construction of hTRPV1-SOPP3 fusion channel. The total energy after the condenser measured at the focal plane was approximately 0.5 mW. An estimation of the illumination circle at the focal plane around 7.1 x 10^−2^ cm^2^ leads to the light intensity of 7.1 mW/cm^2^.B. Current traces of hTRPV1 (top) and hTRPV1-SOPP3 (bottom) in response to a voltage step from 0 to +125 mV. Tail currents were recorded at +60 mV. After light, maximal currents were recorded in the presence of 1 μM CAP.C. Normalized macroscopic current (left) and the time constant of channel activation (right). Channels were activated by a voltage step from 0 to 120 mV. The average value of the current amplitudes before light exposure was used as control to normalize all the raw macroscopic current amplitudes. Data is presented as mean ± SEM (n = 7-10 cells). Statistical significance was determined by an unpaired Student's t-test (∗∗∗p < 0.001).D. Representative current traces of hTRPV1 (top) and hTRPV1-SOPP3 (bottom) in response to a series of voltage steps from −100 to +200 mV with 25 mV intervals. From left to right: before light, after light, after light exposure and in the presence of 1 μM CAP.E. G-V relationship derived from the current traces in D, G = I/(V-V_rev_), The value of V_1/2_ was estimated by visual inspection, G_200 mV_ = I_200 mV_/(V_200 mV_-V_rev_), fitted using the Boltzmann equation: y = A_2_ + (A_1_ – A_2_)/[1 + exp((x – x_0_)/dx)]. From left to right, hTRPV1 (before light, after light, after light with CAP), hTRPV1-SOPP3 (before light, after light, after light with CAP), hTRPV1-SOPP3 recorded with the same procedure but without light exposure (dark 0 s, dark 100 s, CAP 200 s).Fig. 1
For voltage-gated ion channels, the value of V_1/2_ is more closely correlated with the thermodynamics (free energy landscape) of channel activation when the closed and open ensembles (well) are equally distributed, while the activation kinetics is more closely correlated with the energy barrier between two states. To quantify any changes in the response to voltage, we applied a series of voltage steps at a 25-mV interval to activate hTRPV1 (Fig. 1D, top). Notably, in the absence of CAP, the voltage-dependent activation of hTRPV1 is difficult to reach steady-state, even with voltage steps up to +200 mV, and light had no obvious effect on the G-V curve. CAP (1 μM) shifts the midpoint of the G-V curve to the left by about −71 mV and brings the channel activation closer to saturation (Fig. 1E, left; Table S1). For hTRPV1-SOPP3, light pulses shifted the G-V curve to the left by approximately −31 mV (before light, 143.0 ± 9.2 mV, after light, 112.0 ± 12.5 mV, n = 11, p = 0.0003, paired t-test), and interestingly, had almost no effect on the subsequent action of CAP (Fig. 1E, middle). For comparison purposes, G-V curves were also recorded in the same sequence for hTRPV1-SOPP3, but without light exposure (Fig. 1E, right).
PDM of hTRPV1-SOPP3 mediated by 1O2 produced by rose Bengal
3.2
To further elucidate the involvement of ^1^O_2_ in PDM of hTRPV1, we examined the effects of rose Bengal (Fig. S2), a widely employed photosensitizer for producing ^1^O_2_. Based on the excitation and emission spectra of rose Bengal, we chose a 532 nm laser as the light source and evaluated rose Bengal-mediated PDM of hTRPV1. Applications of 1 μM rose Bengal in the bath solution and light pulses resulted in substantial increases in both macroscopic current amplitude and the rate of channel activation (Fig. 2A–B). Analysis of G–V curves showed a negative shift by approximately −34 mV (Fig. 2C–D), consistent with the observations of hTRPV1-SOPP3. To further confirm the involvement of ^1^O_2_, we recorded hTRPV1 currents in the presence of 1 mM Trolox, an effective antioxidant that quenches the generation of ^1^O_2_ (Fig. S2). Indeed, the PDM effects on the function of hTRPV1-SOPP3 were mostly abolished by Trolox (Fig. 2E and F). Furthermore, application of Trolox eliminated photodynamic effects on hTRPV1 current mediated by rose Bengal (Fig. 2G and H).Fig. 2. Identification of ^1^O_2_ as the major player in PDM of hTRPV1.A. Whole-cell currents recorded in HEK293 cells expressing hTRPV1 with 1 μM rose Bengal added to (green) or omitted from (black) the pipette solution, under blue light illumination.B. Whole-cell current recordings from HEK293T cells expressing hTRPV1 in the presence of rose Bengal. Left, normalized macroscopic current; right, activation time constant.C. Representative current traces of hTRPV1 (black) and hTRPV1 with rose Bengal (green) in response to voltage steps from - 100 to + 180 mV (25 mV increments).D. Conductance-voltage (G-V) relationship (left) and the activation time constant (right) of the currents shown in C. Black, hTRPV1 + light; green, hTRPV1 + light + rose Bengal.E. Whole-cell membrane currents recorded from a HEK293 cell expressing hTRPV1-SOPP3, in the absence (black) and presence (red) of blue light illumination. Left to right: No Trolox, with 1 mM Trolox.F. Normalized macroscopic current from the experiments shown in panel E. Currents were elicited by a voltage step from 0 to +120 mV G. Representative current traces of hTRPV1. Left, rose Bengal (1 μM); right, rose Bengal (1 μM) and Trolox (1 mM) added to the pipette solution.H. Normalized macroscopic current from the experiments shown in panel G. Currents were elicited by a voltage step from 0 to +120 mV.Fig. 2
Identification of H167, a histidine residue critical for the PDM of hTRPV1
3.3
To determine residues in hTRPV1 targeted by ^1^O_2_, we systematically mutated histidine residues, which exhibits the highest reaction rate with ^1^O_2_ among 20 amino acids (Fig. S3) [33]. Without light exposure, the responses of hTRPV1/H167A mutant channel voltage and CAP are comparable to the WT channels, including the G-V curve measured by a series of voltage steps (Fig. 3A–C). Light application still increases the current amplitude of hTRPV1/H167A but to a less degree compared to the WT channel, and the response to CAP remains largely unaffected by the H167A mutation (Fig. 3D). Impressively, H167A mostly abolishes the left shift in the G-V curve after PDM (before light, 140.8 ± 8.3 mV, after light, 134.4 ± 11.1 mV, n = 7, p = 0.214, paired t-test), while application of CAP still left shifts the G-V curve of hTRPV1/H167A-SOPP3 by ∼50 mV (Fig. 3E–F).Fig. 3H167 is critical for ^1^O_2_-mediated PDM of hTRPV1-SOPP3.A. Location of H167 in the N-terminus of hTRPV1.B. Current traces of H167A-SOPP3 recorded with the voltage step of 0 to +125 mV C. Normalized macroscopic currents of hTRPV1-SOPP3 (blue), hTRPV1-H167A-SOPP3 (red), hTRPV1 (black), in response to blue light light illumination, and hTRPV1-SOPP3 (grey) without light exposure. Data are presented as mean ± SEM (n = 6-10 cells). Unpaired t-test (hTRPV1-SOPP3 after light vs hTRPV1/H167A-SOPP3 after light, black ∗p < 0.05). Paired t-test (hTRPV1/H167A-SOPP3, before vs after light, red ∗∗p < 0.01).D. Normalized macroscopic currents evoked by CAP in hTRPV1-SOPP3 (blue, pre-illuminated), hTRPV1-H167A-SOPP3 (red), wild-type hTRPV1 (black), and un-transfected cells (solid square). Data are presented as mean ± SEM (n = 7-9 cells).E. Current responses of TRPV1/H167A-SOPP3 activated by a series of voltage steps from −100 mV to +200 mV in a 25 mV interval. Unpaired t-test (∗∗p < 0.01).F. Conductance-voltage (G-V) relationships of hTRPV1-H167A-SOPP3 recorded from −100 mV to +200 mV under three conditions: before illumination (black), after blue light illumination (blue), and after illumination followed by capsaicin (CAP) application. Data are normalized to the peak conductance after light + CAP (left panel) or to the peak before illumination (right panel).Fig. 3
Moreover, the H167 mutation markedly altered the change in V_1/2_ caused by SOPP3 modification (Tables S1 and S2). Mass spectrometry analysis of samples after PDM confirmed that the H167 residue undergoes oxidative modification, collectively identifying H167 as an important site for ^1^O_2_ modification in hTRPV1 (Fig. S4A and S4B). These results suggest that ^1^O_2_-mediated PDM of hTRPV1 is relatively independent from the CAP-dependent gating machinery, which might be due to a spatial separation between structure elements responsible for these two gating mechanisms (Fig. S4C).
Here we focused on H167 but rather other positions because: 1) we recorded robust ionic currents from WT and H167A mutant channels, while the macroscopic currents of H533A and M335A were of relatively small amplitudes. Mutations of H533A or M335A might interfere with surface expression or channel function, making subsequent physiological characterizations more challenging; 2) consistent with electrophysiology recordings, calcium imaging revealed robust responses to capsaicin by H167A (without SOPP3, 1 μM CAP); 3) upon light exposure, swift increases in calcium signals could be observed with H533A-SOPP3, suggesting significant responses to PDM; 4) Oxidation of H167 was identified in our initial mass-spec screening of hTRPV1-SOPP3 with light exposure sample.
To further investigate the alterations in the function of hTRPV1, we employed calcium imaging using GCaMP6s as the sensor for cytosolic calcium (Fig. 4A–B). Impressively, light application by itself effectively elicits a prominent and sustained increase in [Ca^2+^], while H167A significantly reduces the slope and maximal amplitude of the calcium transient (Fig. 4C). Notably, H167A does not have any obvious effects on the maximal response of hTRPV1 to CAP (1 μM), but reduces the sensitivity to CAP (Fig. 4D–S5). Since the ankyrin repeat domain is located far from the transmembrane domain where CAP binds to and activates TRPV1, H167 may allosterically contribute to CAP-dependent activation of TRPV1.Fig. 4. Calcium imaging reveals different responses exhibited by WT hTRPV1 and hTRPV1/H167A mutant channels.A. Schematic diagram of calcium imaging experiments for hTRPV1-SOPP3.B. Representative fluorescence images of cells expressing hTRPV1-SOPP3 or H167A mutant channels (from left to right): mCherry signal showing the expression of channel protein, baseline signal of GCaMP6s fluorescence, fluorescence images collected at peak amplitude, and fluorescence images at the end of the recording.C. Representative traces of GCaMP6s-reported calcium signals in HEK293T cells expressing hTRPV1-SOPP3 (red) and hTRPV1/H167A-SOPP3 (blue). Recordings were performed in extracellular solution containing 1.5 mM Ca^2+^ under blue light illumination. Data points and error bars represent the mean ± SEM (n = 3 wells).D. Representative traces of GCaMP6s-reported calcium signals in HEK293T cells expressing hTRPV1 (green), hTRPV1/H167A (purple), and GCaMP6s (dark yellow). Recordings were performed in extracellular solution containing 1.5 mM Ca^2+^ under blue light illumination. Data points and error bars represent the mean ± SEM (n = 3 wells).E. Statistical summary of the peak amplitude (left) and the rate of calcium signal rise (right) for the indicated hTRPV1 variants. Colors match results shown in C and D. Data is presented as mean ± SEM (n = 3 independent wells). Statistical significance was determined by an unpaired Student's t-test (∗p < 0.05).F. Calcium signaling traces of hTRPV1-SOPP3 and H167A mutant channels with calcium-free extracellular solution containing 1 mM EGTA.Fig. 4
Interestingly, as revealed by calcium imaging as a readout of hTRPV1 function, both the kinetics and the amplitudes of the calcium transients elicited by PDM of hTRPV1-SOPP3 are comparable to the responses to CAP by hTRPV1 and hTRPV1/H167A-SOPP3 channels (Fig. 4E). Furthermore, removal of extracellular Ca^2+^ abolished calcium signals induced by either PDM mediated by ^1^O_2_ or CAP, confirming the hTRPV1 channels on the cell membrane as the carrier for the Ca^2+^ influx from extracellular space (Fig. 4F). Collectively, these results demonstrate that PDM mediated by ^1^O_2_ enhances the function of hTRPV1, reflected in the prominent increases in channel opening and [Ca^2+^] influx under physiological membrane potential, and identify H167 as a critical residue in these processes.
In addition, we examined the role of previously reported regulatory cysteine residues in hTRPV1 in ^1^O_2_-mediated PDM. We constructed a triple-cysteine mutant hTRPV1 channel (C158A/C387S/C767S, termed 3C) and tested its responses to PDM and CAP (Fig. S6A). Calcium imaging experiments confirmed that hTRV1-3C exhibits robust responses to CAP. To examine the responses of WT hTRPV1 and hTRPV1-3C to PDM, we co-expressed a membrane-anchored SOPP3 (SOPP3 with the membrane-anchoring domain from tyrosine-protein kinase Lck, MEM-SOPP3) with hTRPV1 or hTRPV1-3C channels [34]. With MEM-SOPP3 as the photosensitizer, a significant increase in intracellular calcium was observed in cells transfected with hTRPV1 or hTRPV1-3C but not the control cells (Fig. S6B–D). Thus, the three cysteine residues at positions 158, 387, and 767 in hTRPV1 are not essential for ^1^O_2_-mediated PDM of hTRPV1.
Calcium imaging reveals different responses to PDM by hTRPA1 and hTRPV1 channels
3.4
Next, we examined the response to ^1^O_2_-mediated PDM by TRPA1, the only member in TRPA subfamily and is known for its sensitivity towards ROS. We engineered SOPP3 to the N-terminus of the first ankyrin repeat domain of hTRPA1, hTRPA1-SOPP3, analogous to the hTRPV1-SOPP3 channel (Fig. 5A, S1B). We used calcium influx as a more physiological measure of hTRPA1 channel activity. PDM of hTRPA1-SOPP3 results in a transient increase in [Ca^2+^]i, which returns to close to resting levels within minutes, in contrast to the long lasting [Ca^2+^]i increase observed with hTRPA1 + AITC or hTRPV1-SOPP3 after PDM (Fig. 5B-5C-5D). Notably, the calcium transient evoked by PDM of hTRPA1-SOPP3 was much weaker than that of AITC, in both kinetics and maximal amplitude. Importantly, after PDM of hTRPA1-SOPP3 and the return of the calcium transient close to resting levels, AITC failed to elicit any further increases in [Ca^2+^]i (red, Fig. 7B).Fig. 5hTRPA1-SOPP3 shows different responses to PDM compared to hTRPV1-SOPP3.A. Schematic diagram of calcium imaging experiments for hTRPA1-SOPP3.B. Schematic diagram of representative fluorescence signals across experimental groups (left to right): Cellular mCherry signal, GCaMP6s baseline signal, peak Ca^2+^ signal, and Ca^2+^ signal at the end of recording.C. Representative traces of GCaMP6s-reported calcium signals in HEK293T cells. Top, hTRPA1-SOPP3 (red), hTRPA1 stimulated with AITC (green), and GCaMP6s alone stimulated with AITC (100 μM). Data points and error bars represent the mean ± SEM (n = 3-4 wells).D. Representative traces of GCaMP6s-reported calcium signals in HEK293T cells. hTRPV1-SOPP3 (blue) and hTRPV1 stimulated with CAP (purple, 1 μM). Recordings were performed in extracellular solution containing 1.5 mM Ca^2+^ under blue light illumination. Data points and error bars represent the mean ± SEM (n = 3-4 wells).E. Statistics of the peak amplitude (left), the slope of calcium signal rise (middle), and the amplitude near the end of recording (right) as indicated. Data are presented as mean ± SEM (n = 3- 4 wells). Statistical significance was determined by an unpaired Student's t-test (∗p < 0.00.05, ∗∗p < 0.00.01, ∗∗∗∗p < 0.0001).Fig. 5. Fig. 6PDM of hTRPA1/C621S–SOPP3 and hTRPA1-SOPP3 and the responses to a non-electrophilic ligand, carvacrol.A. Calcium signals recorded in cells co-expressing GCaMP6s with either hTRPA1-SOPP3 (red) or hTRPA1 (pink). Arrows in corresponding colors indicate the time of agonist addition. Error bars indicate mean ± SEM (n = 4 wells).B. Calcium signals recorded in cells co-expressing hTRPA1/C621S and GCaMP6s upon addition of carvacrol (orange) or AITC (brown). Error bars indicate mean ± SEM (n = 3-4 wells).C. Calcium signals in cells co-expressing GCaMP6s with either hTRPA1/C621S–SOPP3 (brown) or hTRPA1-SOPP3 (red) in response to light illumination (to generate singlet oxygen) and subsequent addition of carvacrol. Error bars indicate mean ± SEM (n = 3-4 wells).D. Quantification of the maximum calcium signal amplitude (left) and the reciprocal of the time to reach the maximum signal (right) from the experiments shown in panel C. Statistical significance was determined by an unpaired Student's t-test.Fig. 6. Fig. 7Effects of Trolox and DTT on the PDM of hTRPV1-SOPP3 and hTRPA1-SOPP3.A. hTRPV1-SOPP3. Calcium signals upon CAP stimulation at 500 s under control conditions (blue, no Trolox) and following Trolox pretreatment (purple). Recordings were performed in extracellular solution containing 1.5 mM Ca^2+^ under blue light illumination. Data points and error bars represent the mean ± SEM (n = 3-4 wells).B. hTRPA1-SOPP3. Calcium signals upon AITC stimulation at 500 s under control conditions (red, no Trolox) and with Trolox (1 mM) pretreatment (green). Recordings were performed in extracellular solution containing 1.5 mM Ca^2+^ under blue light illumination. Data points and error bars represent the mean ± SEM (n = 3-4 wells).C. Calcium signals of hTRPV1-SOPP3 recorded under conditions of pretreatment with 1 mM DTT for 10 min in the dark (purple), pretreatment with 1 mM Trolox for 10 min in the dark (blue), and control (green). All recordings were performed in extracellular solution containing 1.5 mM Ca^2+^. Data points and error bars represent the mean ± SEM (n = 3 wells).D. Statistics of the peak amplitude (left) and the slope of calcium signal rise (right) shown in C. Data are presented as mean ± SEM (n = 3 wells). Statistical significance was determined by an unpaired Student's t-test (∗p < 0.05, ∗∗∗p < 0.001, ∗∗∗∗p < 0.0001).E. Calcium signals of hTRPA1-SOPP3 under the conditions of pretreatment with 1 mM DTT for 10 min in the dark (dark yellow), pretreatment with 1 mM Trolox for 10 min in the dark (pale red), and control (red). All recordings were performed in extracellular solution containing 1.5 mM Ca^2+^. Data points and error bars represent the mean ± SEM (n = 3 wells).F. Statistics of the peak amplitude (left) and the slope of calcium signal rise (right) shown in E. Data are presented as mean ± SEM (n = 3 wells). Statistical significance was determined by an unpaired Student's t-test (∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001, ∗∗∗∗p < 0.0001).Fig. 7
In TRPA1, several cysteine residues in the channel's N-terminus, including C621, C641, and C665, have been confirmed to be important for the co-valent modification by electrophilic agonists including AITC [35,36]. In addition, K708 also contributes to the sensitivity to AITC [36]. Non-electrophilic agonists, including carvacrol, enhance channel opening by interacting with lipid-binding pockets in the voltage-sensing domain [37]. First, we confirmed the swift response to carvacrol by hTRPA1 (no SOPP3) with the illumination light on (Fig. 6A). Impressively, hTRPA1-SOPP3 after PDM exhibited a steadily increasing response to carvacrol, in contrast to the null response to AITC after PDM. Then we examined the responses to AITC and carvacrol by hTRPA1/C621S (no SOPP3) and observed a fast response to carvacrol and an expected slow response to AITC (Fig. 6B). Importantly, the C621S mutation does not affect the PDM effect: hTRPA1/C621S–SOPP3 and hTRPA1-SOPP3 exhibit comparable responses to PDM and a slow response to subsequently added carvacrol (Fig. 6C and D). Taken together, the null response to AITC and the slower response to carvacrol by hTRPA1-SOPP3 after PDM suggest that in the manifestation of the function phenotype after PDM, important structure elements in the channel protein, including the cysteine residues in TRPA1 N-terminus as well as the carvacrol binding sites near VSD, synergistically affect channel gating through an allosteric mechanism.
Next, we examined the effects of Trolox, an effective quencher for ^1^O_2_, on the PDM of hTRPV1 and hTRPA1 channels. Trolox (1 mM) largely abolished the light-induced calcium influx in both hTRPA1-SOPP3 and hTRPV1-SOPP3 channel but fully restored the responses to AITC or CAP, confirming the important role by ^1^O_2_ in PDM of both channels (Fig. 7A–B). These results further confirm the drastically divergent outcomes of PDM of TRPV1 and TRPA1: ^1^O_2_ induces a sustained calcium response in hTRPV1 but a transient increase with hTRPA1. These differential responses to ^1^O_2_ by TRPV1 and TRPA1 may underlie important roles by both channels under physiological conditions. The much-reduced responses to CAP by hTRPV1-SOPP3 after PDM as shown by calcium imaging are probably due to a combination of factors, including modifications to sites including H167 that allosterically contribute to CAP-dependent channel activation, strong calcium influx during light exposure that saturates the calcium dye and reduces the local cross-membrane gradient of calcium concentration, and the related channel desensitization.
Among all ROS including ^1^O_2_, H_2_O_2_ has received much research attention and it has been well established that TRPA1 is the sensor for H_2_O_2_ [23,38]. Isolated publications reported TRPV1 can also be activated by H_2_O_2_, although at high concentrations in the millimolar range [27,39]. H_2_O_2_ mainly affects the redox environment by promoting the oxidation of residues especially cysteine, while DTT effectively antagonizes or even reverses the H_2_O_2_ effects on both TRPV1 and TRPA1 [25,27,39]. To further clarify the involvement of H_2_O_2_ in PDM of hTRPV1 and hTRPA1, we treated cells expressing hTRPA1-SOPP3 or hTRPV1-SOPP3 with 1 mM DTT prior to calcium imaging. DTT treatment did not affect the slope of the rising phase of [Ca^2+^]i and the peak amplitude of hTRPV1-SOPP3 after PDM (Fig. 7C–D). In contrast, for hTRPA1-SOPP3, DTT significantly enhanced the peak amplitude, while the time to peak remained unchanged. Importantly, for both channels, applying Trolox completely abolished the calcium increases (Fig. 7E–F). These results provide further evidence that the calcium signals elicited by PDM of hTRPA1-SOPP3 and hTRPV1-SOPP3 are primarily mediated by ^1^O_2_ rather than H_2_O_2_. The DTT-induced alterations in peak amplitude are likely attributable to its effect on disulfide bonds formed in the resting state in the channel protein before PDM.
Light-dosage dependent PDM of hTRPV1 and hTRPA1
3.5
To further explore the relationship between the calcium influx and the PDM processes, we examined the light dosage effects by reducing the light intensity from 2 mW/cm^2^ to 0.9 mW/cm^2^. For hTRPV1-SOPP3, reduced light intensity prolonged the time to the peak of calcium signal and decreased the peak amplitude (Fig. 8A–B). Interestingly, reducing the light intensity significantly increased the area under the curve (AUC) of the calcium signal, which is likely due to a delayed but enhanced peak in the hTRPA1-SOPP3 calcium signal.Fig. 8. Light dosage and illumination duration affect PDM of hTRPV1-SOPP3 and hTRPA1-SOPP3.A. Top, calcium signal of hTRPV1-SOPP3-expressing cells. Intensity of blue light illumination: 2 mW/cm^2^ (blue) and 0.9 mW/cm^2^ (light blue). Bottom, hTRPA1-SOPP3-expressing cells stimulated with light illumination with intensities of 2 mW/cm^2^ (red) and 0.9 mW/cm^2^ (light red). All recordings were performed in extracellular solution containing 1.5 mM Ca^2+^. Data points and error bars represent the mean ± SEM (n = 3 wells).B. Statistics of the peak amplitude (up), AUC (middle), and the slope of calcium signal rise phase (down). Data are presented as mean ± SEM (n = 3 wells). Statistical significance was determined by an unpaired Student's t-test (∗p < 0.00.05, ∗∗p < 0.01, ∗∗∗p < 0.001, ∗∗∗∗p < 0.0001)C. Calcium signals of hTRPA1-SOPP3. Following an initial light illumination, the light was terminated at the time points of 40 s (yellow), 60 s (green), or 80 s (blue). Light illumination was restored at the time point of 380 s, followed by application of AITC at 500 s. All recordings were performed in extracellular solution containing 1.5 mM Ca^2+^. Data points and error bars represent the mean ± SEM (n = 3 wells).D. Experiment configuration for calcium imaging with modulated extracellular calcium.E. Calcium signals recorded from cells expressing GCaMP6s + hTRPA1-SOPP3. Ca^2+^-free and with 2 mM EGTA solution was applied first. At 70 s (orange) or 370 s (green), extracellular [Ca^2+^] was increased to 2.5 mM, followed by AITC being added. Error bars represent mean ± SEM (n = 3 wells).F. Calcium signals recorded from cells expressing GCaMP6s + hTRPA1 (without SOPP3).G. Calcium signals recorded from cells expressing GCaMP6s only.Fig. 8
For hTRPV1, as confirmed by both electrophysiology and calcium imaging, PDM directly increases the channel opening. In contrast, PDM of hTRPA1 results in a transient increase than permanently inhibition of channel activity, which is opposite to the sustained increase in calcium upon exposure to H_2_O_2_. Is it calcium-dependent inhibition of channel activity or ^1^O_2_-mediated inhibition of hTRPA1-SOPP3 [40]? For hTRPA1-SOPP3, it is known that sustained calcium influx introduces inactivation of channel opening. To explore whether the calcium transient elicited by PDM of TRPA1 is related to channel inactivation, we examined the light illumination of different durations, from 40, 60, to 80 s, with the aim to halt the PDM process at various time points during the rising phase of the calcium transient (Fig. 8C). Upon reillumination of the system at the time point of 370 s, earlier termination did lead to higher residual calcium signals, indicating intracellular accumulation of calcium didn't play a major role in the shutting down of hTRPA1-SOPP3 during continuous light exposure. Moreover, with the light exposure turned off earlier, the calcium influx continued to stay at an elevated level in the dark. Notably, cessation at 40 s led to a renewed increase (390-410 s) followed by a decrease in the calcium signal, further supporting that the extent of PDM instead of intracellular calcium plays a dominant role in hTRPA1-SOPP3 inactivation during sustained light exposure (Fig. 8C). Responses to subsequently applied AITC were detected only in the experimental groups with light off at 40 or 60 s.
Moreover, we started the calcium imaging (Fig. 8D) experiments from the calcium-free condition. Blue light illumination still elicited a transient calcium signal in cells expressing hTRPA1-SOPP3 (Fig. 8E), probably due to hTRPA1-mediated calcium release from lysosomes as reported previously [41]. Upon replenishment of calcium at 80 s, the calcium signal increased further (orange trace); however, when calcium was added at 380 s, the calcium signal only showed a minimum response (green traces). In both cases, AITC failed to elicit any increases in calcium signal. In contrast, cells expressing hTRPA1 exhibited robust responses to replenishment of extracellular calcium at two timepoints, as well as an additional increase upon AITC application (Fig. 8F). Cells expressing GCaMP6s showed no response to either calcium or AITC (Fig. 8G). In conjunction with the application of light exposure of different lengths, these results demonstrate that ^1^O_2_ dependent modification of hTRPA1 occurs independently of channel inhibition by intracellular calcium, which results in progressive loss of TRPA1 channel function and reduced agonist sensitivity. Notably, calcium imaging experiments, with proper controls, reflect the channel activity under the physiological membrane potential of approximately −50 to −60 mV, while in the absence of agonist like AITC, substantial activation of TRPA1 as monitored by electrophysiology requires high voltage steps, close to or over +100 mV [[42], [43], [44]]. The discrepancy between calcium imaging and electrophysiology has been documented in the literature. As shown by patch-clamp recordings, H_2_O_2_, at the concentration of 10 mM, only sensitizes TRPV1's response to CAP but cannot elicit any obvious increase in currents [27]. In contrast, calcium recordings reveal a dramatic increase in intracellular calcium upon stimulation of H_2_O_2_. Similar observations have been obtained with hemin activation of hTRPA1 [45].
H2O2 preferentially activates hTRPA1 over hTRPV1
3.6
Finally, we examined the difference in the response to H_2_O_2_ by hTRPV1 and hTRPA1. Because of the short shelf life in months from batch-to-batch variations of H_2_O_2_, we started from setting up a quantification of the effective concentration of H_2_O_2_ by the KI-H_2_SO_4_ assay (Fig. S7). Then we applied H_2_O_2_ of calibrated concentrations to hTRPA1 and hTRPV1 and used cytosolic Ca^2+^ signal as readout. Notably, H_2_O_2_ at the concentration 16 μM is sufficient to induce evident increases in Ca^2+^ signals in hTRPA1 (Fig. 9A), whereas ≥300 μM H_2_O_2_ was required to elicit significant responses in hTRPV1 (Fig. 9B). Hill equation fitting of concentration-response curves revealed an EC_50_ of 390.9 μM for hTRPA1 versus 1358 μM for hTRPV1 (Fig. 9C). These results are consistent with publication studies on H_2_O_2_-dependent activation of TRPA1 [25,46,47]. Notably, too high concentrations of H_2_O_2_ might introduce complicated perturbations to TRPA1 activity including desensitization, as well as intracellular signaling pathways, including the effects of intracellular calcium, thus we limit the measurements to 4050 μM H_2_O_2_ or 7100 μM for TRPV1. Taken together, our results illustrated that TRPA1 and TRPV1 exhibit quantitatively and qualitatively different responses to H_2_O_2_ and ^1^O_2_, respectively, and therefore make their unique contributions to ROS sensation.Fig. 9hTRPA1 exhibits robust response and higher sensitivity to H_2_O_2_ compared to hTRPV1.A. Calcium signals of hTRPA1 (top) or GCaMP6s only (bottom). Cells expressing hTRPA1 were treated with different concentrations of H_2_O_2_, while cells expressing GCaMP6s only were challenged with 5 mM H_2_O_2_. All H_2_O_2_ treatments were followed by the addition of AITC (100 μM) at 500 s. Recordings were conducted in an extracellular solution containing 1.5 mM Ca^2+^. Data are presented as mean ± SEM (n = 3 wells).B. Calcium signals of hTRPV1 (top) or GCaMP6s only (bottom). Cells expressing hTRPV1 were treated with different concentrations of H_2_O_2_, while cells expressing GCaMP6s only were challenged with 5 mM H_2_O_2_. All treatments were followed by the addition of 1 μM CAP at 500 s. Recordings were conducted in an extracellular solution containing 1.5 mM Ca^2+^. Data are presented as mean ± SEM (n = 3 wells).C. Dose-response curve of hTRPA1 and hTRPV1 channel activation by H_2_O_2_. AUC vs [H_2_O_2_] in logscale was fitted to the Hill equation (Y = 100/(1 + 10^((LogEC50-X)∗HillSlope))). The R squared for the fits of hTRPV1 and hTRPA1 were 0.7691 and 0.7691, respectively.Fig. 9
Discussion
4
The overall topology and assembly of TRP channels share high similarities to voltage-gated potassium channels, however, they are non-selective, calcium permeable channels and respond weakly to membrane potential depolarization. The opening of TRP channels is driven by a broad range of stimuli of diverse physics and chemistry nature, including the local redox balance encoded by ROS and RNS. Indeed, H_2_O_2_ directly opens TRPA1 and TRPM2 and sensitizes the response of TRPV1 to heat, CAP, and proton. Here we expand our understanding of redox sensing by TRPV1 and TRPA1 to ^1^O_2_, which is still mysterious but may function as an important signaling factor. We discover that through separate molecular mechanisms, TRPV1 and TRPA1 exhibit dramatically different responses to H_2_O_2_ and ^1^O_2_.
PDM of TRPA1 and TRPV1 channels have been reported in the literature. These studies confirmed the involvement of ROS produced under oxidative stress, but did not specifically identify or examine the role of ^1^O_2_. Excited photosensitizers including acridine orange (490 nm) and hypericin (590 nm) increase TRPA1 current amplitude, which are bundled together with effects of UVA and H_2_O_2_ and can be largely reversed by DTT [48]. In addition, blue light in combination with organic photosensitizers (phorphyras) increases the activity of both TRPA1 and TRPV1 channels [49]. Previous studies have identified cysteine residues, especially C621/C641/C665 in TRPA1 and C158/C387/C391/C767 in TRPV1, are important for H_2_O_2_ mediated channel activation or sensitization, and conversely, DTT either reverses or attenuates these modification effects, which are more directly related to oxidative reagents such as H_2_O_2_ and modifications to cysteine [23,25,27,28,45,48,50].
Here to tease out the effects of ^1^O_2_ from other ROS, especially H_2_O_2_, we applied two photosensitizers, rose Bengal and SOPP3, and applied relatively specific quenchers or reducing reagents, such as Trolox for ^1^O_2_ and DTT for H_2_O_2_. Attachment of SOPP3 to the channel protein helps exclude non-specific modifications to lipids or other proteins. Histidine, tryptophan, methionine, and tyrosine show highest reacting rates with ^1^O_2_ [33,51]. Indeed, we identified a histidine residue in hTRPV1, H167, and the alanine replacing mutation largely diminishes the effects of PDM modification. In addition, Trolox, an effective quencher for ^1^O_2_, completely abolishes the PDM effects on the function of TRPV1 and TRPA1, while DTT has minimal impacts on PDM of both channels. These maneuvers help establish the critical role played by ^1^O_2_ in PDM of TRPV1 and TRPA1.
For hTRPV1-SOPP3, DTT has no obvious effects on its response to PDM. In contrast, in the presence of DTT, PDM generated a more robust increase in calcium signal for hTRPA1-SOPP3. Based on the null response to AITC and reduced response to carvacrol, both cysteine residues in the N-terminus as well as the binding site for non-electrophilic ligands appear to allosterically contribute to the changes in channel function after PDM of hTRPA1 (Fig. 6, Fig. 7). It is possible that DTT obliterates certain oxidations to hTRPA1-SOPP3 prior to PDM in the resting state and therefore makes the channel more responsive to PDM, in which ^1^O_2_ is a major player. Notably, in contrast to the effects by DTT, Trolox, an effective antioxidant that quenches the generation of ^1^O_2_, completely abolishes the PDM effects on hTRPV1 and hTRPA1 mediated by SOPP3 and rose Bengal. Thus, DTT and Trolox appear to be able to distinguish the modifications to TRPV1 and TRPA1 by H_2_O_2_ and ^1^O_2_, respectively, despite their volatile chemical nature and the complex, intertwined signaling pathways involving ROS.
Both H_2_O_2_ and ^1^O_2_ are non-radical and do not carry any charges, therefore, their propagation to the targeted residue in protein molecules is not affected by electrostatic forces but rather by simple spatial steric interactions. Given their extremely simple atomic structure but volatile chemical nature, the specificity of the residues and the functional state targeted by H_2_O_2_ and ^1^O_2_ are most likely low. Clearly, for TRP channels, the plethora of gating stimuli, including the sensing of temperature, voltage, and redox balance, relies mostly on a structurally distributed mechanism. H167 is located in the ankyrin repeats in the N-terminus of TRPV1, which has been shown before to be critical for the function of TRP channels [52,53]. In canonical ligand-gated and voltage-gated channels, the domains that sense gating stimuli, such as the ligand-binding domain (LBD) and the voltage-sensing-domain (VSD), and the ion-conduction pore typically are clearly defined and only occupy a small fraction of total volume of the channel protein. The bulk of the channel protein contributes to the allosteric coupling and therefore is more prone to protein modifications. The reaction rate of these residues to ^1^O_2_ and structurally their availability to ^1^O_2_, which might alternate between different conformations, might be key factors determining the likelihood of modification by ^1^O_2_. Notably, H167 in hTRPV1 is a major site being targeted by ^1^O_2_ but clearly not the only one, given the largely diminished but still prominent responses to PDM especially in the aspects of macroscopic current amplitude, channel activation kinetics, and intracellular calcium (Figs. 3C and 4C).
Both TRPA1 and TRPV1 are extensively expressed in primary afferent sensory neurons and carry significant physiological functions. The first mammalian TRPV1 channel was cloned from rat DRG neurons. The TRPV1 channel expressed in the nerve endings of mostly unmyelinated C- and some myelinated Aδ-type nerve fibers of DRG and trigeminal neurons [54]. TRPV1 plays an important role in sensing noxious heat, neuropathic and inflammatory pain, and chemical signals including CAP and pH. On the other hand, TRPA1 has been found in a subset of nociceptive neurons that also expresses TRPV1. In addition to natural agonists including ingredients from mustard and wasabi, TRPA1 contributes to the sensation of hazardous oxidants including H_2_O_2_ and even air pollutants. The interaction between TRPA1 and TRPV1 within the same DRG neurons can be largely summarized as negative regulations but in some cases synergistic or positive impacts on each other's opening, in which intracellular calcium, in conjunction with other signaling pathways including ROS pathways, plays a key role [[55], [56], [57], [58], [59]]. In addition, TRPV1 and TRPA1 co-assemble and form functional heteromeric channels, which exhibit unique biophysical characteristics and make different physiological contributions in nociception and inflammation [[60], [61], [62], [63]]. Here we report that TRPV1 and TRPA1 exhibit mostly opposite responses to ^1^O_2_ and H_2_O_2_. Therefore, in response to the local presence of ^1^O_2_, it is possible that the sustained opening of TRPV1 results in an increase in intracellular calcium which further suppresses the opening of TRPA1 which is also under the regulation of ^1^O_2_. Oppositely, TRPA1 is way more sensitive to H_2_O_2_ than TRPV1, and in turn, the opening of TRPA1 stimulated by H_2_O_2_ results in increases in intracellular calcium concentration which in turn affects the opening of TRPV1. The bifurcated responses to ^1^O_2_ and H_2_O_2_ by two representative TRP channels might form an important mechanism in fine-tuning the responses to different ROS by the corresponding nociceptive sensory neurons.
Notably, the current study of ^1^O_2_-mediated PDM of TRPV1 and TRPA1 contain a number of limitations. The mass-spec analysis is limited by low coverage rates, an incomplete survey of all potential oxidation sites after PDM, and a lack of quantitative estimation of modification percentage, and thus is mainly used to corroborate with functional assays. The identification of H167 in hTRPV1 and N-terminal cysteines in hTRPA1 does not exclude other modification sites, as evidenced by the residual response to PDM by the corresponding mutant channels. The technique of calcium imaging provides a measurement of TRPV1 and TRPA1 channel activity under physiological membrane potential but is limited by short range of linearity and interference from other players in calcium homeostasis. Finally, a combination of different techniques, including patch-clamp recordings of TRPV1 and TRPA1, especially in the inside-out configuration, should be effective in further dissecting the synergistic modulation of channel activity by the plethora factors including voltage, agonist, calcium-dependent activation and desensitization, and ^1^O_2_ modification.
In summary, we demonstrate that TRPV1 and TRPA1, two representative members from the TRP channel family, are highly sensitive to photodynamic processes, and provide evidence that ^1^O_2_ plays a major role. In conjunction with energetic contributions from other gating stimulus, such as voltage and ligand binding, ^1^O_2_-mediated PDM increases the likelihood of channel opening for TRPV1 while having an inhibitory impact on the function of TRPA1 (Fig. 10). The corresponding biophysical mechanism, the chemical nature of the modification process, and importantly, the physiological significance remains to be elucidated for ^1^O_2_ - a ubiquitous but still enigmatic signaling molecule.Fig. 10. Summary of different responses to ^1^O_2_ and H_2_O_2_ by hTRPV1 and hTRPA1.Dashed lines represent calcium signals at the resting level. Dotted lines represent the responses of hTRPV1-SOPP3 and hTRPV1-SOPP3 to PDM in the absence of DTT.Fig. 10
CRediT authorship contribution statement
Yunshen Chen: Formal analysis, Investigation, Software, Writing – review & editing. Gaogao He: Formal analysis, Investigation. Wei Zhang: Formal analysis, Investigation. Jiajie Li: Formal analysis, Investigation. Xiaoxi Li: Formal analysis, Investigation. Sijun Dong: Conceptualization, Supervision. Qinglian Liu: Conceptualization, Writing – review & editing. Lei Zhou: Conceptualization, Funding acquisition, Investigation, Methodology, Project administration, Supervision, Writing – original draft, Writing – review & editing.
Declaration of competing interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
The reference list from the paper itself. Each links out to its DOI / PubMed record.
- 1Sies H.Mailloux R.J.Jakob U.Fundamentals of redox regulation in biology Nat. Rev. Mol. Cell Biol.25920247017193868906610.1038/s 41580-024-00730-2PMC 11921270 · doi ↗ · pubmed ↗
- 2Rampon C.Vriz S.Hydrogen peroxide signaling in physiology and pathology Antioxidants (Basel)123202310.3390/antiox 12030661 PMC 1004500636978909 · doi ↗ · pubmed ↗
- 3Lennicke C.Hydrogen peroxide - production, fate and role in redox signaling of tumor cells Cell Commun. Signal.132015392636993810.1186/s 12964-015-0118-6PMC 4570748 · doi ↗ · pubmed ↗
- 4Ogilby P.R.Singlet oxygen: there is indeed something new under the sun Chem. Soc. Rev.3982010318132092057168010.1039/b 926014 p · doi ↗ · pubmed ↗
- 5Kochevar I.E.Singlet oxygen signaling: from intimate to global Sci. STKE 20042212004 pe 71498310210.1126/stke.2212004 pe 7 · doi ↗ · pubmed ↗
- 6De Rosa M.C.Crutchley R.J.Photosensitized singlet oxygen and its applications Coord. Chem. Rev.2332002351371
- 7Kanofsky J.R.Singlet oxygen production by human eosinophils J. Biol. Chem.263201988969296962838476 · pubmed ↗
- 8Prasad A.Singlet oxygen production in Chlamydomonas reinhardtii under heat stress Sci. Rep.62016200942683121510.1038/srep 20094 PMC 4757480 · doi ↗ · pubmed ↗