IQGAP1 promotes chronic pain by regulating the trafficking and sensitization of TRPA1 channels

Abstract TRPA1 channels have been implicated in mechanical and cold hypersensitivity in chronic pain. But how TRPA1 mediates this process is unclear. Here we show that IQ motif containing GTPase activating protein 1 is responsible using a combination of biochemical, molecular, Ca2+ imaging and behavioural approaches. TRPA1 and IQ motif containing GTPase activating protein 1 bind to each other and are highly colocalized in sensory dorsal root ganglia neurons in mice. The expression of IQ motif containing GTPase activating protein 1 but not TRPA1 is increased in chronic inflammatory and neuropathic pain. However, TRPA1 undergoes increased trafficking to the membrane of dorsal root ganglia neurons catalysed by the small GTPase Cdc42 associated with IQ motif containing GTPase activating protein 1, leading to functional sensitization of the channel. Activation of protein kinase A is also sufficient to evoke TRPA1 trafficking and sensitization. All these responses are, however, completely prevented in the absence of IQ motif containing GTPase activating protein 1. Concordantly, deletion of IQ motif containing GTPase activating protein 1 markedly reduces mechanical and cold hypersensitivity in chronic inflammatory and neuropathic pain in mice. IQ motif containing GTPase activating protein 1 thus promotes chronic pain by coupling the trafficking and signalling machineries to TRPA1 channels.


Behavioural assays
Animals were acclimatized to the testing environment for at least 3h prior to behavioural assays which were typically carried out in the middle of the day.
Hot plate: Mice were placed on a temperature-controlled surface with a transparent plexiglass cylinder restraining the animals (Ugo basile). Latency to nocifensive behaviours such as paw licking, flinching, lifts and jumping was determined.
Von Frey test: The mechanical sensitivity of the mice was assessed using a dynamic plantar aesthesiometer (Ugo Basile). Animals were placed on metal grid pane enclosed by a transparent box (9.6 x 9.6 x 14.5cm) and habituated for at least 30 min prior to experiment. To measure mechanical sensitivity, the plantar surface of mice was stimulated by a rigid von Frey filament with increasing force. The threshold force at which animals withdraw their paws was recorded.
2-3 trials were tested for each measurement with 10min gap between each trial.
Hargreaves test: the thermal sensitivity of mice was measured using a Hargreaves apparatus (Ugo Basile). After thorough habituation of mice on a glass frame plane enclosed by a plastic box, infrared heat was delivered from underneath of the paws. Time latency to paw withdrawal was then automatically recoded.
Acetone evaporation assay: cold sensitivity of animals was assessed using acetone evaporation assy. A drop of 50l acetone was delivered to the plantar surface of the hindpaws of mice, time spent on licking, flinching and biting was recorded.

Cell culture and transfection
HEK293 cells were maintained in DMEM medium (4.5g/l glucose) supplemented with 10% FBS, 2mM L-glutamine ,100 units/ml penicillin and 100g/ml streptomycin in a humidified incubator containing 5% CO2, as described previously 51 . Cells were transfected with TurboFect transfection reagents (Thermo Fisher Scientific) as described previously 51 .
DRG was rapidly isolated from adult mice after sacrifice by cervical dislocation. DRG was then dissociated and cultured as described previously with minor modifiations 51 . Briefly, DRG was treated with Type IV collagenase (Worthington) at 37C for 30min followed by trituration with a 23G needle. Cells were then filtered through a Falcon cell strainer (Corning) to remove debris and pelleted by centrifugation. DRG neurons were resuspended in DMEM medium (1g/l glucose) supplemented with L-glutamine (2mM), 100U/ml penicillin and 100g/ml streptomycin and cultured on coverslips coated with poly-L-lysine (100g/ml) in a humidified CO2 incubator. Nerve growth factor (NGF) was not added to the culture to prevent unexpected effects.
For imaging DRG neurons isolated from mice with CFA inflammatory pain and SNI neuropathic pain, only lumbar DRG (L3-L5) from the contralateral and ipsilateral sides were isolated. Dissociated DRG neurons were plated in droplet on the coverslips coated with 100g/ml poly-l-lysine. Neurons were allowed to adhere for 1h at 37C before adding more DMEM medium containing 2%FBS, 100U/ml penicillin and 100g/ml streptomycin. DRG neurons were then immediately used for calcium imaging within 5h after isolation.

Membrane protein detection and western blotting
Membrane proteins were labelled using biotinylation assay as described previously 50 . Briefly, HEK293 cells expressing TRPA1 and other cDNA constructs were live labelled with 2mM EZlink Sulfo-NHS-LC-Biotin (ThermoFisher) on ice for one hour after stimulation. Cells were then washed with cold PBS containing 100mM glycine followed by solubilization using a lysis buffer consisting of 20mM HEPES (pH 7.4), 150mM NaCl, 1mM EDTA, 1mM EGTA, 1% NP-40 plus protease inhibitor cocktails (Merck). Cell lysate was incubated at 4C for 30min on a mixer with constant rotation followed by centrifuge at 12,000rpm for 10min at 4C.
Supernatant was then incubated with Streptavidin agarose (Pierce) at 4C overnight. Proteins were then thoroughly washed with lysis buffer succeeded by boiling in laemmli buffer. Proteins were next separated in 7.5% SDS-PAGE gel and transferred to a PVDF blot (ThermoFisher).
We also isolated membrane protein using Mem-PER Plus Membrane Protein Extraction Kit (ThermoFisher) from DRG and sciatic nerves in accordance with instructions with mild modifications. Briefly, lumbar DRG (L3-L5) was washed with 200l wash buffer after isolation. They were then added to 200l permeabilization buffer and homogenized using a motor-driven homogenizer followed by incubation at 4C for 10min. Cell suspension was centrifuged at 16,000rpm for 15min at 4C. Cell pellet was next resuspended in the solubilization buffer containing protease inhibitors and homogenized with a homogenizer. Cell solution was then incubated at 4C for 30min with constant mixing prior to centrifuge at 16,000rpm for 15min at 4C. Supernatant was then used for Western blotting as described above. Membrane TRPA1 was detected using anti-TRPA1 antibody (Alomone labs).

Pull down assay, coimmunoprecipitation and proteomics
Nickel bread pull down assay was used to detect interaction between TRPA1 and IQGAP1 in HEK293 cells and was performed as described previously with modificatons 52 . Briefly,

HEK293 cells transfected with TRPA1-V5-Histidine tag and IQGAP1 or other IQGAP1
deletion constructs (CaM-IQGAP1, MK24-IQGAP1) were solubilized in lysis buffer containing 20mM Tris-HCl (pH 7.4), 300mM NaCl, 1% NP-40, 0.4mM EDTA, 20mM Imidazole, 10% Glycerol plus protease inhibitor cocktails. Cell lysate was then incubated at 4C for 30min before centrifuge at 12,000rpm for 10min at 4C. Supernatant of cell lysate was then incubated with 30l Ni-NTA beads (Qiagen) at 4C overnight. Nickel beads were then thoroughly washed in lysis buffer before subjected to boiling in sample buffer. Protein supernatant was then separated in 7.5%SDS-PAGE gel followed by blot transfer and protein detection using anti-IQGAP1(Santa Cruz) and anti-V5 (ThermoFisher).
Nickel beads purified proteins from HEK293 cells expressing TRPA1-V5-6histidine were also used for identification of unknown proteins copurified with TRPA1 ( Fig. 2A). Briefly, after separation in 7.5% SDS-PAGE gel, purified proteins were fixed and stained with silver staining kit (Sigma) in accordance with manufacturer's instructions. Protein bands of interest were excised and processed for LC-MS/MS analysis in the proteomic centre at the University of Aberdeen.
GST pull down assay was employed to delineate the binding regions between TRPA1 and IQGAP1 and was conducted as described previously 52,53 . Briefly, the cytoplasmic tails of TRPA1 coupled to GST tag were purified from HEK293 cells expressing the fusion proteins using GST-agarose (Merck). Purified GST-coupled fusion proteins were then incubated with HEK293 cell lysate containing IQGAP1 or N-IQGAP1 or Flag-C-IQGAP1 at 4C overnight with constant mixing. Bound proteins were then dissociated from beads by boiling in Laemmli buffer followed by protein separation and detection using anti-IQGAP1 and/or -Flag (Merck).
Co-immunoprecipitation was used to detect TRPA1-IQGAP1 interaction in DRG neurons and binding of HA-CaM to IQGAP1 in HEK293 cells. It was performed as described previously 53 .
In Brief, DRG neurons were solubilized in lysis buffer. Cell lysate was then isolated and incubated with monoclonal anti-IQGAP1 or anti-HA (Covance) and Protein A/G PLUSagarose (Santa Cruz) at 4C overnight. Protein-bound beads were thoroughly washed before boiling in Laemmli buffer. Dissociated proteins were next separated in SDS-PAGE gel followed by blot transfer and protein detection using anti-TRPA1 (Alomone) or anti-IQGAP1.

Immunohistochemistry and immunocytochemistry
Mice were transcardially perfused with PBS and 4% paraformaldehyde (PFA). Lumbar DRG (L4-L5), sciatic nerve and skin were then isolated and post-fixed in 4% PFA at 4C overnight with constant rotation followed by cryopreservation in 30% sucrose. Tissues were embedded in OCT medium (Tissue-Tek) and sectioned in a cryostat at 12m thickness. DRG sections were placed onto poly-lysine-coated slides for immunohistochemistry. DRG tissue sections were first blocked in 5% Donkey serum plus 0.03% Triton X-100 at room temperature for 30min. They were then incubated with primary antibodies (mouse anti-IQGAP1(Santa Cruz),

Immunocytochemistry and live labelling
Cultured DRG neurons were fixed in 4%PFA at RT for 20min followed by blocking in 0.1% cDNA was amplified in LightCycler 480 (Roche). All gene expression was normalised against GAPDH and analysed using Ct method.

Statistics
All data are mean ± SEM. Significance between groups was determined using Student's t test or one or two-way ANOVA followed by Bonferroni post-hoc test. P<0.05 was considered to be significant.