Transient Receptor Potential Ankyrin 1 Receptor Stimulation by Hydrogen Peroxide Is Critical to Trigger Pain During Monosodium Urate–Induced Inflammation in Rodents




Gout is a common cause of inflammatory arthritis and is provoked by the accumulation of monosodium urate (MSU) crystals. However, the underlying mechanisms of the pain associated with acute attacks of gout are poorly understood. The aim of this study was to evaluate the role of transient receptor potential ankyrin 1 (TRPA-1) and TRPA-1 stimulants, such as H2O2, in a rodent model of MSU-induced inflammation.


MSU or H2O2 was injected into the hind paws of rodents or applied in cultured sensory neurons, and the intracellular calcium response was measured in vitro. Inflammatory or nociceptive responses in vivo were evaluated using pharmacologic, genetic, or biochemical tools and methods.


TRPA-1 antagonism, TRPA-1 gene deletion, or pretreatment of peptidergic TRP-expressing primary sensory neurons with capsaicin markedly decreased MSU-induced nociception and edema. In addition to these neurogenic effects, MSU increased H2O2 levels in the injected tissue, an effect that was abolished by the H2O2-detoxifying enzyme catalase. H2O2, but not MSU, directly stimulated sensory neurons through the activation of TRPA-1. The nociceptive responses evoked by MSU or H2O2 injection were attenuated by the reducing agent dithiothreitol. In addition, MSU injection increased the expression of TRPA-1 and TRP vanilloid channel 1 (TRPV-1) and also enhanced cellular infiltration and interleukin-1β levels, and these effects were blocked by TRPA-1 antagonism.


Our results suggest that MSU injection increases tissue H2O2, thereby stimulating TRPA-1 on sensory nerve endings to produce inflammation and nociception. TRPV-1, by a previously unknown mechanism, also contributes to these responses.

Gout is the most common cause of painful inflammatory arthritis among men and postmenopausal women. Mainly because of an aging population and lifestyle changes, the incidence and prevalence of gout are steadily increasing ([1, 2]). Poorly controlled gout leads to a limitation of activities and a significant decrease in health-related quality of life ([3]). Signs and symptoms of gout are caused by soft tissue deposits of monosodium urate (MSU) crystals, which trigger episodes of intense articular and periarticular inflammation and excruciating pain ([1, 4]). However, the underlying mechanism of the inflammatory process in gout that results in sensory symptoms and pain is poorly understood. Accordingly, patients who experience acute attacks of gout are undertreated ([1, 2]).

Some members of the transient receptor potential (TRP) family of ion channels expressed on primary sensory neurons, including TRP ankyrin 1 (TRPA-1) and TRP vanilloid channel 1 (TRPV-1), have been labeled thermo-TRPs, because of their ability to sense changes in temperature ([5, 6]). TRPA-1–expressing neurons also contain the neuropeptides substance P and neurokinin A and calcitonin gene-related peptide, which, when released from peripheral terminals, cause neurogenic vasodilatation and edema ([5, 7]). We previously demonstrated that TRPV-1, TRPV-1–positive sensory neurons, and mast cell degranulation contribute to MSU-induced nociceptive and edematogenic responses in rats ([8]).

The observation that high levels of oxidative stress byproducts are present in patients with gout ([9]) and are produced endogenously after MSU challenge in experimental animals ([10-12]) suggests a role of oxidative stress in these conditions. In addition to several food ingredients (e.g., allyl isothiocyanate, which is present in mustard oil) and environmental irritants (e.g., acrolein, a volatile and irritant agent that is present in vehicle exhaust fumes and tear gas), TRPA-1 is activated by an unprecedented series of endogenous compounds generated by oxidative stress. These include H2O2, 4-hydroxynonenal, 4-oxononenal, and other compounds ([7, 13, 14]). Thus, TRPA-1 qualifies as a sensor of oxidative stress. A large body of evidence indicates that TRPA-1 receptor causes inflammatory responses, as well as cold and mechanical hypersensitivity, in models of inflammatory and neuropathic pain ([14, 15]). Therefore, the first aim of the present study was to evaluate the role of TRPA-1 and its activation and sensitization by reactive oxygen species (ROS) in a rodent model of MSU-induced inflammation.

TRPA-1 is usually coexpressed with TRPV-1 in a subset of nociceptive neurons, and several studies have described the synergic action of the 2 channels in different pain conditions ([16-18]). TRPV-1 has already been shown to contribute to MSU-induced hypersensitivity and edema in rodents ([8]). Thus, the second aim of this study was to explore the cooperation of TRPA-1 and TRPV-1 in the mechanism of pain-related behavior and inflammation in a rodent model of MSU-induced inflammation.



Adult male Wistar rats (200–300 gm) bred in-house and wild-type (Trpa1+/+) or TRPA-1–deficient (Trpa1−/−) mice (20–30 gm; C57BL/6 background) ([19]) were used. All protocols were approved by the Ethics Committee of the Federal University of Santa Maria (23081.003640/2009-61) or by the University of Florence (204/2012-B). In addition, the number of rodents and the intensity of noxious stimuli were the minimum necessary to demonstrate consistent effects of the drug treatments, in accordance with current ethics guidelines for the investigation of experimental pain in conscious animals ([20]). The investigators conducting the experiments were blinded to the treatment conditions.


If not otherwise indicated, all reagents were obtained from Sigma and were dissolved in the appropriate vehicle solutions. HC-030031 was synthesized as previously described ([21]).

Preparation and administration of MSU crystals

Synthetic MSU crystals were prepared as previously described ([8]). The crystals were characterized by polarizing light microscopy and showed clinical morphologic characteristics with a mean ± SEM length of 12 ± 2 μm, as described previously ([8]). The preparation was endotoxin-free, as determined by a Limulus amebocyte cell lysate assay (Thermo). MSU crystals were suspended in sterile phosphate buffered saline (PBS) before being injected. MSU suspension (0.25 mg/paw) was administered subcutaneously into the plantar surface of the right hind paws of unanesthetized rats (100 μl) or mice (20 μl), as previously described ([8]).

Evaluation of nociceptive response

The rodents were individually placed in transparent glass chambers, to permit observation of the ongoing nociception. After the acclimation period (20 minutes), the amount of time spent flicking or licking the injected paw following subcutaneous injection of MSU was determined with a chronometer and used as a measure of ongoing nociception ([8, 22]). Moreover, cold-evoked nociception (cold allodynia) was evaluated with an acetone evaporative cooling test ([23, 24]), using the following nociception scores: 0 = no response, 1 = quick withdrawal, 2 = prolonged withdrawal or repeated flicking of the paw, and 3 = repeated flicking and licking of the paw.

Determination of the inflammatory response

Edema formation was observed as an increase in paw thickness, as measured by a digital caliper, and was calculated as the difference between the basal value and the test value (observed after intraplantar injection of drugs) ([8]). Myeloperoxidase (MPO) activity was determined as previously described ([8]). Interleukin-1β (IL-1β) content was assessed using an enzyme-linked immunosorbent assay kit (PeproTech). Hematoxylin and eosin staining and histologic evaluation of infiltrated inflammatory cells (polymorphonuclear leukocytes [PMNs]) was carried out in a randomly selected representative area, using a light microscope with a 20× objective ([25]).

Treatment protocols

HC-030031 (a TRPA-1 selective antagonist; 30–300 nmoles/paw), camphor (a TRPA-1 poorly selective antagonist; 150 nmoles/paw), SB-366791 (a TRPV-1 selective antagonist; 10 nmoles/paw), indomethacin (a cyclooxygenase inhibitor; 280 nmoles/paw), or vehicle solution (0.1% DMSO in PBS; 100 μl/paw) was subcutaneously coinjected with MSU (0.25 mg/paw), vehicle (PBS; 100 μl), or the TRPA-1 agonists allyl isothiocyanate (AITC; 1 nmole/paw) and H2O2 (3 μmoles/paw). In another set of experiments, HC-030031 (300 μmoles/kg) or vehicle (1% DMSO in PBS; 1 ml/kg) was administered orally 1 hour before the subcutaneous injection of MSU (0.25 mg/paw) or vehicle (PBS; 100 μl). Moreover, SB-366791 (0.1 nmole/paw) plus HC-030031 (30 nmoles/paw) or vehicle (0.1% DMSO in PBS; 100 μl/paw) was coinjected with MSU (0.25 mg/paw), H2O2 (3 μmoles/paw), or vehicle (PBS; 100 μl).

In a different set of experiments, catalase from bovine liver (300 IU/paw), dithiothreitol (DTT; 20 nmoles/paw), or vehicle (PBS; 100 μl/paw) was coinjected with MSU (0.25 mg/paw), H2O2 (3 μmoles/paw), AITC (1 nmole/paw, coinjected only with catalase or vehicle), or vehicle (PBS; 100 μl). DTT (20 nmoles/paw) was injected 5 minutes before administration of H2O2, to prevent any ongoing reaction with H2O2. External hind paw temperature was measured before and 10 minutes after the intraplantar injection of H2O2 (3 μmoles/paw), as described previously ([26]).

TRPV-1– and TRPA-1–positive sensory fibers were ablated as described previously ([8, 27]). Briefly, anesthetized rodents were desensitized using a perineural injection of capsaicin (2%; 10 μl) or vehicle (10% ethanol, 10% Tween 80 in PBS) into the nerve sheath, using a microsyringe. Seven days later, MSU (0.25 mg/paw), AITC (positive control; 1 nmole/paw), or vehicle (PBS; 100 μl/paw) was injected subcutaneously. The treatment time and drug doses were chosen based on data from previously published studies as well as the results of pilot experiments using positive controls (data not shown).

Western blot analysis

After 7 days of the desensitization protocol or 0.5 or 6 hours after the MSU injection, the rats were killed, and the right sciatic nerves or hind paw skin, respectively, was quickly isolated and homogenized in lysis buffer containing protease inhibitors. After centrifugation (3,000g for 30 minutes at 4°C), the supernatant was collected. The protein content was determined using bovine serum albumin (BSA) as a standard ([28]). Next, protein samples (30 μg) were mixed with loading buffer and boiled for 10 minutes ([22, 30]). The proteins were separated by electrophoresis on 10% sodium dodecyl sulfate–polyacrylamide gels and transferred to PVDF membranes. The proteins on the membrane were stained with a solution of 0.5% ponceau plus 1% glacial acetic acid in water, as a loading control ([22, 29, 30]). After staining, the membranes were dried, scanned, and quantified. Membranes were then processed using a Millipore SNAP system, blocked with 1% BSA, incubated for 10 minutes with an anti–TRPV-1 or anti–TRPA-1 antibody (1:150; Santa Cruz Biotechnology), washed 3 times, incubated with an alkaline phosphatase–coupled secondary antibody (1:3,000), and visualized with a BCIP/p-nitroblue tetrazolium system. The membranes were dried, scanned, and quantified using the Scion Image (PC) version of NIH Image software. The results were normalized by arbitrarily setting the densitometry of the control group as 100%.

Calcium influx in rat dorsal root ganglia (DRG) neurons

Rat DRG neurons were cultured as previously described ([23]). Intracellular calcium fluorescence was measured in neurons, as previously reported ([21, 23]). Neurons were exposed to uric acid (100–300 μM), MSU crystals (0.003–0.100 gm/liter), H2O2 (10–5,000 μM), acrolein (30 μM), capsaicin (0.1 μM), or their vehicles (buffer solution). The HC-030031 and SB-366791 vehicles (which were used in all of the in vitro experiments) were both 1% DMSO. Results are expressed as the increase in the emission intensity ratios at 340 nm/380 nm excitation (R340/380) over baseline, normalized to the maximum effect induced by ionomycin (5 μM) added at the end of the experiment (percent change in R340/380).

H2O2 production assay

The H2O2 levels in paw skin after subcutaneous injection of MSU were determined by the phenol red–horseradish peroxidase (HRP) method ([31]). Briefly, 0.25–48 hours after administration of MSU (0.25 mg/paw) or vehicle (PBS; 100 μl), and 0.25 hours after administration of MSU or vehicle plus HC-030031 (300 nmoles/paw), SB-366791 (10 nmoles/paw), or catalase (300 IU/paw), the rats were killed, and hind paw skin was removed. Basal values were assessed in the rats that were not injected. The samples were homogenized in 50 mM phosphate buffer (pH 7.4) containing 5 mM sodium azide at 4°C for 60 seconds, and the homogenate was centrifuged (12,000g for 20 minutes at 4°C). The supernatant obtained was used to determine H2O2 levels ([31]). The H2O2 levels were expressed as μmoles of H2O2 on the basis of a standard curve of HRP-mediated oxidation of phenol red by H2O2, corrected by protein content (in milligrams) of the paw skin sample analyzed.

Statistical analysis

The percent inhibition was calculated with the maximum developed responses obtained after injection of MSU, AITC, or H2O2 compared with vehicle-treated rodents (control). Student's t-test was used to determine significant differences between 2 groups. One-way and two-way analysis of variance were also used, when appropriate, to assess statistical significance among >2 groups and among ≥2 groups in time-course curves, respectively, followed by the Bonferroni post hoc test. P values less than 0.05 were considered significant. Median effective dose (ED50) values (i.e., the dose of H2O2 needed to elicit 50% of the response relative to the control value) were determined by nonlinear regression analysis with a sigmoid dose-response equation, using GraphPad version 5.0, and are reported as geometric means and 95% confidence intervals.


Effect of TRPA-1 activation on MSU-induced pain-related behaviors and edema

Subcutaneous injection of MSU (0.25 mg/paw) into rat paws caused a short-lasting increase in ongoing nociception (from 0 to 15 minutes), prolonged cold-evoked allodynia (from 0.25 to 4 hours), and increased paw thickness (from 0.5 to 48 hours) (Figure 1A). According to the time course of the effects produced by MSU administration, time intervals of 0–10 minutes for ongoing nociception, 15 minutes for cold allodynia, and 30 minutes for edema were chosen to investigate their respective mechanisms.

Administration of the poorly selective and selective TRPA-1 antagonists camphor and HC-030031, respectively, decreased the nociceptive and edematogenic responses evoked by MSU. Local coadministration of HC-030031 (300 nmoles/paw) or indomethacin (280 nmoles/paw) also markedly inhibited MSU-induced ongoing nociception (84% and 86% inhibition, respectively), cold allodynia (100% and 100% inhibition at 0.25 hour, respectively), and edema (93% and 87% inhibition at 0.5 hour, respectively) at all time points evaluated (Figure 1A). The local coadministration of camphor (150 nmoles/paw) reduced ongoing nociception, cold allodynia, and edema caused by MSU by 84%, 100%, and 80%, respectively (Figure 1B).

Figure 1.

Transient receptor potential ankyrin 1 (TRPA-1) activation mediates monosodium sodium (MSU) crystal–induced ongoing nociception, cold allodynia, and edema in rodents. A, Effect of local administration of the TRPA-1 selective antagonist HC-080031 (HC) and the cyclooxygenase inhibitor indomethacin (INDO) on MSU-induced ongoing nociception, cold allodynia, and changes in paw thickness. B, Effect of local administration of HC-080031 and the nonselective TRPA-1 antagonist camphor (Cam; 150 nmoles/paw) on MSU-induced ongoing nociception, cold allodynia, and changes in paw thickness. C, Effect of oral (po) administration of HC-080031 on MSU-induced ongoing nociception, cold allodynia, and changes in paw thickness. D, MSU-induced ongoing nociceptive response, cold allodynia, and edema in TRPA-1–deficient mice. In B–D, ongoing nociception, cold allodynia, and edema were measured for 0–10 minutes, 0.25 hour, and 0.5 hour after injection, respectively. Values are the mean ± SEM (n = 5–7 rats or 6–9 mice). ∗ = P < 0.05, ∗∗ = P < 0.01, and ∗∗∗ = P < 0.001 versus vehicle (Veh); # = P < 0.05, ## = P < 0.01, and ### = P < 0.001 versus MSU-treated group (A and B), MSU plus vehicle–treated group (C), and MSU-treated Trpa1−/− mice, by one-way or two-way analysis of variance followed by the Bonferroni post hoc test.

Similar to what was observed with MSU, subcutaneous injection of the TRPA-1 agonist AITC into rat paws induced ongoing nociception, cold allodynia, and paw edema; all of these events were prevented by coadministration with either HC-030031 (300 nmoles/paw) or camphor (150 nmoles/paw) (Table 1). Subcutaneous administration of HC-030031 (300 nmoles/paw), camphor (150 nmoles/paw), or indomethacin (280 nmoles/paw) did not induce nociceptive or edematogenic responses per se (Table 1). When orally administered, HC-030031 (300 μmoles/kg) was also very effective in preventing MSU-evoked ongoing nociception, cold allodynia, and paw edema (93%, 100%, and 75% inhibition, respectively) (Figure 1C). The effects induced by subcutaneous injection of MSU in wild-type mice were markedly reduced in Trpa1−/− mice (79% reduction of ongoing nociception, 100% reduction of cold allodynia, and 95% reduction of edema), further supporting a major role of the TRPA-1 channel (Figure 1D).

Table 1. Effect of control pharmacologic treatments on ongoing nociception, cold allodynia, and edema*
TreatmentNociception time, minutesCold allodynia score, range 0–3Change in paw thickness, mm
  1. Significance was determined by one-way analysis of variance followed by the Bonferroni post hoc test. Values are the mean ± SEM.
  2. aP < 0.001 versus vehicle.
  3. bP < 0.05 versus vehicle.
  4. cP < 0.001 versus allyl isothiocyanate (AITC) alone or monosodium urate (MSU) alone.
  5. dP < 0.05 versus AITC alone or MSU alone.
Vehicle, 100 μl/paw3 ± 10.2 ± 0.20.2 ± 0.03
HC-030031, 300 nmoles/paw4 ± 10.2 ± 0.20.3 ± 0.05
Camphor, 150 nmoles/paw7 ± 10.2 ± 0.20.3 ± 0.05
Indomethacin, 280 nmoles/paw5 ± 10.2 ± 0.20.2 ± 0.05
AITC, 1 nmole/paw41 ± 5a2.3 ± 0.3b1 ± 0.1a
AITC, 1 nmole/paw, plus HC-030031, 300 nmoles/paw9 ± 1c0.2 ± 0.2d0.5 ± 0.1c
AITC, 1 nmole/paw, plus camphor, 150 nmoles/paw13 ± 2c0.2 ± 0.2d0.4 ± 0.03c
SB-366791, 10 nmoles/paw4 ± 10.3 ± 0.30.2 ± 0.05
MSU, 0.25 mg/paw37 ± 4a2.1 ± 0.3b0.8 ± 0.1a
MSU, 0.25 mg/paw, plus SB-366791, 10 nmoles/paw5 ± 4c1.75 ± 0.50.2 ± 0.08c

Because MSU increased TRPA-1 and TRPV-1 expression 6 hours (Figure 2A), but not one-half hour, after intraplantar injection, we explored the role of TRPV-1– and TRPA-1–expressing sensory fibers in MSU-induced pain-related behaviors and edema. The ability of perineural injection of capsaicin to deplete nociceptive fibers was confirmed by a marked reduction in the density of TRPV-1– and TRPA-1–positive sciatic nerve fibers 7 days after treatment (Figure 2B). Capsaicin pretreatment practically abolished ongoing nociception, cold allodynia, and edema induced by AITC and MSU (Figure 2C). These results further support the key role of TRPA-1 channels expressed by TRPV-1–positive sensory neurons in MSU-induced nociception and edema.

Figure 2.

MSU increases TRPA-1 and transient receptor potential vanilloid channel 1 (TRPV-1) expression, and ablation of TRPA-1– and TRPV-1–positive fibers diminishes the MSU-elicited responses. A and B, TRPA-1 and TRPV-1 immunoreactivity in skin samples obtained from hind paw samples, 6 hours after intraplantar injection of MSU (0.25 mg/paw) or vehicle (A) and in right sciatic nerve samples obtained after perineural injection of capsaicin (CPS; 2%) or vehicle (B). C, Effect of subcutaneous injection of MSU, allyl isothiocyanate (AITC; 1 nmole/paw), or vehicle, which was injected 7 days after intraneural injection of vehicle or capsaicin. D, MSU-induced responses following local administration of HC-030031 (30 nmoles/paw), SB-366791 (SB; 0.1 nmole/site), or the combination of HC-030031 and SB-366791. Values are the mean ± SEM (n = 3–4 samples or 6–9 rats). ∗ = P < 0.05 and ∗∗∗ = P < 0.001 versus vehicle (pretreated with vehicle); # = P < 0.05 and ### = P < 0.001 versus MSU or AITC (pretreated with vehicle), by Student's t-test (A and B) or one-way analysis of variance followed by the Bonferroni post hoc test (C and D). See Figure 1 for other definitions.

Synergic action of TRPV-1 and TRPA-1 receptor in MSU-mediated nociception and edema

As previously observed ([8]), local injection of SB-366791 (10 nmoles/paw) significantly reduced MSU-elicited ongoing nociception (98% inhibition) and edema (88% inhibition) but not cold allodynia (Table 1). Furthermore, low doses of HC-030031 (30 nmoles/paw) or SB-366791 (0.1 nmole/paw) did not affect MSU-induced nociception or edema when injected alone (Figure 2D). However, their combination markedly reduced MSU-induced ongoing nociception (78% inhibition), cold allodynia (82% inhibition), and edema (73% inhibition) in rats (Figure 2D).

Reduced MSU-induced inflammatory responses following TRPA-1 blockade

MPO levels were increased only at the 6-hour time point after the subcutaneous injection of MSU. This response was reduced by coadministration of MSU with HC-030031 (300 nmoles/paw) or indomethacin (280 nmoles/paw) (Figure 3A). The histologic analysis showed similar results. The number of PMNs was increased 6 hours after MSU challenge and was reduced by coadministration of MSU with HC-030031 (300 nmoles/paw) or indomethacin (280 nmoles/paw) (Figure 3B). In addition, injection of MSU enhanced the levels of IL-1β 6 hours after treatment, and this effect was reduced by coadministration of MSU with HC-030031 (300 nmoles/paw) or indomethacin (280 nmoles/paw) (Figure 3C).

Figure 3.

TRPA-1 antagonism reduces MSU crystal–induced inflammatory responses. A–C, Increased myeloperoxidase (MPO) levels 6 hours after subcutaneous injection of MSU (A, left), and MPO levels (A, right), numbers of polymorphonuclear leukocytes (PMNs) per high-power field (hpf) (B), and interleukin-1β (IL-1β) levels (C) in response to coadministration of MSU with HC-030031 (300 nmoles/paw) or indomethacin (280 nmoles/paw). D, Top, Representative tracings showing calcium influx in response to MSU (0.1 mg/ml) and uric acid (300 μM) at different time points. Bottom, Pooled data showing that neither MSU nor uric acid produced calcium influx in rat capsaicin (CPS)–sensitive dorsal root ganglia neurons that normally respond to acrolein (ACR) or H2O2. HC-030031 (30 μM) significantly reduced the effect evoked by ACR (30 μM) or H2O2 (500 μM). Bars show the mean ± SEM (n = 5–7 samples or at least 25 neurons). ∗∗∗ = P < 0.001 versus vehicle; ## = P < 0.01 and ### = P < 0.001 versus MSU (pretreated with vehicle), by one-way analysis of variance followed by the Bonferroni post hoc test. ND = not detectable; R340/380 = emission intensity ratios at 340 nm/380 nm excitation; SB = SB-366791 (see Figure 1 for other definitions).

Role of MSU crystals or uric acid in activation of TRPA-1 receptor

Because of the primary role of TRPA-1 in the nociceptive and edematogenic responses induced by MSU in vivo, we sought to determine whether MSU crystals or uric acid could promote calcium influx in rat sensory neurons by activating TRPA-1. Both MSU crystals and uric acid failed to evoke any significant calcium response in 83 of the 134 capsaicin-sensitive DRG neurons that responded to the TRPA-1 agonist acrolein (mean ± SEM 33 ± 3% change in the R340/380 value) (Figure 3D). This finding provides evidence against a direct action of MSU on the TRPA-1 channel expressed in sensory neurons.

Effect of MSU-induced H2O2 production on TRPA-1 stimulation and triggering of nociception and edema

Because MSU stimulates ROS production and TRPA-1 is a sensor of oxidative stress ([6, 32]), we assessed whether ROS were involved in the TRPA-1–mediated responses evoked by MSU. Because catalase decomposes H2O2 to H2O and O2 ([33]), we coinjected the enzyme with MSU. Catalase (300 IU/paw) abolished the development of ongoing nociception (100% inhibition), cold allodynia (100% inhibition), and edema (95% inhibition) (Figure 4A). However, the ongoing nociception, cold allodynia, and edema induced by intraplantar injection of AITC were not reduced by catalase coadministration (data not shown).

Figure 4.

MSU induces H2O2 production to stimulate TRPA-1 and trigger nociception and edema. A, MSU-induced responses following subcutaneous administration of catalase (300 IU/paw). B, H2O2 production in rodent paw skin tissue 0.25–48 hours after subcutaneous administration of MSU. C, H2O2 production following injection of HC-030031 (300 nmoles/paw), SB-366791 (SB; 10 nmoles/paw), and catalase (300 IU/paw). D, Effect of subcutaneous injection of the cell-permeable reducing agent dithiothritol (DTT; 20 mmoles/paw) on MSU- and H2O2-induced ongoing nociception, cold allodynia, and edema. Values are the mean ± SEM (n = 5–7 rats). ∗ = P < 0.05, ∗∗ = P < 0.01, and ∗∗∗ = P < 0.001 versus vehicle (and vehicle pretreated with vehicle); # = P < 0.05 and ### = P < 0.001 versus MSU or H2O2 pretreated with vehicle, by one-way analysis of variance followed by the Bonferroni post hoc test. See Figure 1 for other definitions.

Next, we observed an increase in H2O2 production in the injected tissue 0.25–6 hours after MSU administration (Figure 4B). The H2O2 concentration at 0.25 hour was ∼7-fold greater than baseline values or values measured in vehicle-treated rodent tissue (Figure 4B). Moreover, the administration of catalase in a dose that produces antinociceptive and antiedematogenic effects (300 IU/paw), but not HC-030031 (300 nmoles/paw) or SB-366791 (10 nmoles/paw), prevented the increase in H2O2 levels evoked by MSU. Thus, MSU-induced H2O2 production appeared to be independent of TRPA-1 or TRPV-1 stimulation (Figure 4C).

As previously reported ([32, 34]), H2O2 (10–5,000 μM) produced a concentration-related (EC50 566 μM; mean ± SEM maximum efficiency of 47 ± 4% change in R340/380) calcium influx in sensory neurons that responded to TRPA-1 agonists (Figure 3D). H2O2 (500 μM) evoked a robust calcium influx (in 59 of 112 capsaicin-sensitive DRG neurons), an effect that was significantly prevented by incubation with HC-030031 (30 μM) but not SB-366791 (3 μM) (Figure 3D). HC-030031 and SB-366791 reduced the calcium response evoked by selective agonists of TRPA-1 and TRPV-1 receptors, respectively (Figure 3D). Thus, the H2O2 generated by MSU may act on sensory neurons mainly activating TRPA-1 receptor, thereby causing nociception and edema. In line with this hypothesis, intraplantar injection of H2O2 (3 μmoles/paw) produced a transient ongoing nociceptive response and prolonged cold allodynia and edema in rats, with estimated ED50 values of 2.8 (range 2.1–3.9), 4.7 (range 2.6–8.7), and 1.2 (range 0.8–1.8) μmoles/paw, respectively (Figures 5A and B). Both HC-030031 (300 nmoles/paw) and camphor (150 nmoles/paw) markedly inhibited H2O2-evoked ongoing nociception (71% and 75% inhibition, respectively), cold allodynia (both 100% inhibition), and edema (96% and 94% inhibition, respectively) (Figure 5C). However, the TRPV-1 antagonist SB-366791 (10 nmoles/paw) decreased only ongoing nociception (89% inhibition) without altering cold allodynia or edema induced by H2O2 (Figure 5C).

Figure 5.

H2O2-mediated responses are dependent on TRPA-1 activation. A, Time course of H2O2-induced ongoing nociception, cold allodynia, and edema in rats. B, Effect of different doses of H2O2 (0.3, 1, 3, and 10 μmoles/paw) on ongoing nociception, cold allodynia, and edema. C, Effect of HC-030031 (300 nmoles/paw), camphor (Cam; 150 nmoles/paw), SB-366791 (SB; 10 nmoles/paw) on responses induced by subcutaneous injection of H2O2. D, Effect of local coadministration of HC-030031 (30 nmoles/paw) and SB-366791 (0.1 nmole/site) on ongoing nociceptive, cold allodynic, and edematogenic responses elicited by H2O2 (3 μmoles/paw). Values are the mean ± SEM (n = 5–9 rats). ∗ = P < 0.05, ∗∗ = P < 0.01, and ∗∗∗ = P < 0.001 versus vehicle; # = P < 0.05 and ### = P < 0.001 versus Cam, HC, or SB plus vehicle, by one-way or two-way analysis of variance followed by the Bonferroni post hoc test. See Figure 1 for other definitions.

Similar to data obtained with MSU, low doses of HC-030031 (30 nmoles/paw) or SB-366791 (0.1 nmole/paw), when injected alone, were unable to alter H2O2-induced nociception or edema. In contrast, the combination markedly reduced H2O2-evoked ongoing nociception (81% inhibition), cold allodynia (100% inhibition), and edema (100% inhibition) (Figure 5D). Moreover, intraplantar injection of H2O2 increased external paw skin temperature (from mean ± SEM 28 ± 1°C before treatment to 32 ± 0.8°C 10 minutes after H2O2 injection; P < 0.05 [n = 5–6]), an effect that could contribute to TRPV-1 activation/sensitization ([27, 35]).

It has been demonstrated that reactive TRPA-1 agonists bind to intracellular cysteine residues to activate the channel, an effect that was prevented by the reducing agent DTT, which reverses cysteine disulfide formation and nitrosylation or oxidization of cysteine sulfhydryls ([36, 37]). Local pretreatment with DTT (20 nmoles/paw) decreased both MSU- and H2O2-induced ongoing nociception (71% and 100% inhibition, respectively), cold allodynia (100% and 100% inhibition, respectively), and edema (90% and 100% inhibition, respectively) (Figure 4D), thus supporting the participation of cysteine residues of TRPA-1 receptor in this phenomenon.


Gout is a cause of recurrent acute inflammatory arthritis, which considerably worsens the quality of life of patients ([1, 2]). Despite the vast amount of information about the disease, only limited knowledge on the underlying mechanism of gout pain is available, thus producing an unfavorable impact on current treatments ([1, 38]). In the present study, we obtained biochemical, pharmacologic, and genetic data that suggested a key role of TRPA-1 and oxidative stress in pain-like behaviors and edema in a rodent model of MSU-induced inflammation. Recent evidence has underlined the role of TRPA-1 in different rodent models of neuropathic and inflammatory pain ([7, 14, 23]). Here, we extend previous findings, showing that both pharmacologic inhibition and genetic ablation of the TRPA-1 channel abrogate MSU-induced nociception and edema in rodents.

An acute gout flare usually presents as a painful condition associated with the development of cold allodynia and burning pain ([3, 39]), implying the involvement of the thermoreceptors found in sensory neurons. Accordingly, we previously identified a contribution of TRPV-1 receptor expressed by a subset of primary sensory neurons to the ongoing nociception response and edema induced by MSU in rats ([8]). TRPA-1 receptor is coexpressed in ∼30% of TRPV-1–positive sensory neurons ([40]). TRPA-1–positive neurons also contain neuropeptides, which, upon release from peripheral terminals, mediate neurogenic inflammatory responses. Similar to these previous findings, we observed that ablation of TRPV-1–positive sensory fibers by capsaicin treatment markedly reduced the expression of TRPA-1–positive nerve fibers and inhibited MSU-induced ongoing nociception, cold allodynia, and edema. Although the initial suggestion that TRPA-1 is a sensor of cold temperature has been questioned, several recent studies have proposed the contribution of TRPA-1 in cold allodynia in a wide range of experimental conditions ([15, 23, 41]). In agreement with these conclusions, we observed that TRPA-1 antagonism, but not TRPV-1 antagonism, reduced MSU-induced cold allodynia.

Six hours after injection, MSU induced local infiltration of PMNs and increased IL-1β levels, both of which are hallmarks of acute attacks of gout ([1, 42]). Moreover, MSU also enhanced the expression of TRPV-1 and TRPA-1 receptors. Of note, it was recently demonstrated that IL-1 is able to increase the expression of TRPA-1 in cultured synoviocytes ([43]). Thus, increased PMN infiltration, IL-1β production, and TRPA-1 expression induced by MSU appeared to be related to edema formation (which was greater at this moment [6 hours after intraplantar injection of MSU]) but not to nociception (which was intense at earlier time points). Furthermore, TRPA-1 receptor activation was also important in terms of the leukocyte infiltration and cytokine production induced by MSU. Because blockade of IL-1β has been proposed to be a reliable treatment for acute attacks of gout ([42]), the reduction of IL-1β production by TRPA-1 antagonism is a relevant issue.

It has been demonstrated that MSU crystals or uric acid may directly activate different host cell types, in some cases in a manner independent of crystal phagocytosis ([10, 44, 45]). The hypothesis that uric acid or MSU crystals directly activate sensory neurons by TRPA-1 targeting was excluded by their failure to produce any calcium mobilization in cultured primary sensory neurons. The alternative possibility that uric acid or MSU crystals activate TRPA-1 and sensory neurons via indirect mechanisms is suggested by the kinetics of the response to MSU. In fact, treatment with uric acid or MSU crystals delayed ongoing nociception, which was apparent 5 minutes after stimulus administration, while an almost instantaneous response was observed after injection of AITC or capsaicin (data not shown).

Stimulation of resident or infiltrating proinflammatory cells by MSU crystals and uric acid is known to generate ROS ([10-12]). We observed that MSU injection concomitant to the appearance of nociception and edema induced a remarkable increase in H2O2 levels within the injected tissue. We detected that the increase in the H2O2 concentration induced by MSU peaked at 0.25 hour and was still significantly different from vehicle for up to 6 hours, but at lower levels. Thus, H2O2 levels appear to be pivotal to the early development of nociception but accessory to the maintenance of late edema, which must involve other proinflammatory mediators. H2O2 has been identified as an endogenous TRPA-1 agonist ([32, 34, 46]). Thus, it is possible that following exposure to uric acid or MSU crystals, neighboring cells produce H2O2, which, by targeting TRPA-1 on peptidergic nerve terminals, produces nociceptive and inflammatory responses. Although the TRPA-1 expressed in neuronal cells seems to be predominant in the MSU-induced responses, non-neuronal cells expressing TRPA-1, such as endothelial cells ([47]), could account, at least in part, for the effects of MSU.

Similarly to injection of direct TRPA-1 agonists, injection of H2O2 provoked ongoing nociception, cold allodynia, and edema; all of these phenomena were observed much earlier than the (delayed) effects produced by MSU. To further support the role of H2O2, we demonstrated that the cell-permeable reducing agent DTT, which inhibits channel activation by binding to cysteine residues ([32, 36]), protects against the TRPA-1–mediated pronociceptive and inflammatory responses evoked by MSU. It is worth noting that patients with gout have been described as having an increased content of oxidative substances ([9]).

We previously demonstrated that TRPV-1 contributes to nociception and inflammation in a model of acute gout ([8]). However, the present data provide robust evidence that TRPA-1 also plays a major role in this process. A combination of low doses of TRPA-1 and TRPV-1 antagonists abolished MSU-induced cold allodynia, ongoing nociception, and edema, while administration of each of these antagonists alone had no such effect. Previous studies demonstrated that in mice, H2O2 caused nociception and edema in a manner that was dependent on both TRPA-1 and TRPV-1 ([32, 46]). In accordance with these findings, in the present study we observed that H2O2-elicited nociception and edema were inhibited by a high dose of a TRPA-1 antagonist or by the combination of low doses of TRPA-1 and TRPV-1 antagonists.

TRPV-1 does not appear to be directly activated by H2O2 ([48]). However, it is possible that in vivo TRPV-1 activation/sensitization is produced by mediators/effects evoked by H2O2 or TRPA-1. This hypothesis is supported by the finding that H2O2 injection increased paw temperature by ∼4°C, a phenomenon that, in turn, could lead to the activation of TRPV-1 receptor ([27, 35]). TRPV-1 sensitization by H2O2 ([48]) and the ensuing enhanced stimulation by heat might also exaggerate TRPA-1 activation, as observed in previous studies ([27, 35, 49]) as well as the present study. Thus, it is possible that, as has been shown in other experimental conditions of inflammatory pain ([16-18]), both TRPV-1 and TRPA-1 contribute synergistically to the development of painful inflammatory responses evoked by MSU.

In conclusion, H2O2 production by resident cells and the consequent activation of TRPA-1 receptor in sensory neurons seem to start the process that generates MSU-induced pain and inflammation. From this initial event, additional mechanisms contributing to the overall inflammatory and sensory response are progressively recruited, in a time-dependent manner. Accordingly, early blockade of TRPA-1 in gout might be a reliable pharmacologic approach to completely suppress inflammation and pain in acute attacks of gout.


All authors were involved in drafting the article or revising it critically for important intellectual content, and all authors approved the final version to be published. Dr. Ferreira had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

Study conception and design. Trevisan, Hoffmeister, Rossato, Geppetti, Ferreira.

Acquisition of data. Trevisan, Hoffmeister, Rossato, Oliveira, Silva, Ineu, Guerra, Materazzi, Fusi, Nassini.

Analysis and interpretation of data. Trevisan, Hoffmeister, Rossato, Oliveira, Silva, Ineu, Guerra, Materazzi, Fusi, Nassini, Geppetti, Ferreira.


We are grateful to Prof. David Julius (University of California, San Francisco) for the kind gift of the TRPA-1–deficient mice and Dr. Delia Preti (University of Ferrara, Italy) for providing HC-030031.