Human TRPA1 is a heat sensor displaying intrinsic U-shaped thermosensitivity

www.nature.com/scientificreports OPEN received: 20 March 2016 accepted: 07 June 2016 Published: 28 June 2016 Human TRPA1 is a heat sensor displaying intrinsic U-shaped thermosensitivity Lavanya Moparthi1, Tatjana I. Kichko2, Mirjam Eberhardt3, Edward D. Högestätt4, Per Kjellbom1, Urban Johanson1, Peter W. Reeh2, Andreas Leffler3, Milos R. Filipovic5,6 & Peter M. Zygmunt4 Thermosensitive Transient Receptor Potential (TRP) channels are believed to respond to either cold or heat. In the case of TRP subtype A1 (TRPA1), there seems to be a species-dependent divergence in temperature sensation as non-mammalian TRPA1 is heat-sensitive whereas mammalian TRPA1 is sensitive to cold. It has been speculated but never experimentally proven that TRPA1 and other temperature-sensitive ion channels have the inherent capability of responding to both cold and heat. Here we show that redox modification and ligands affect human TRPA1 (hTRPA1) cold and heat sensing properties in lipid bilayer and whole-cell patch-clamp recordings as well as heat-evoked TRPA1dependent calcitonin gene-related peptide (CGRP) release from mouse trachea. Studies of purified hTRPA1 intrinsic tryptophan fluorescence, in the absence of lipid bilayer, consolidate hTRPA1 as an intrinsic bidirectional thermosensor that is modified by the redox state and ligands. Thus, the heat sensing property of TRPA1 is conserved in mammalians, in which TRPA1 may contribute to sensing warmth and uncomfortable heat in addition to noxious cold. The discovery of TRP ion channels as molecular thermosensors has opened up new avenues for understanding how organisms monitor and adapt to environmental temperature. In contrast to the role of TRPA1 as a heat sensor in non-mammalian species, the temperature-sensitivity of mammalian TRPA1 and its role in thermosensation has been debated ever since TRPA1 was proposed as a noxious cold sensor in the mouse sensory nervous system1. We have recently shown that the purified hTRPA1 is intrinsically sensitive to noxious cold when inserted into lipid bilayers and studied with the patch-clamp technique2, adding direct molecular evidence to the many studies suggesting that mammalian TRPA1 plays a role in noxious cold sensation3. There is, however, no evidence that TRPA1 itself is a heat sensor in mammalians, although being involved in heat detection4–11. It has been speculated that thermosensitive TRP channels are capable of sensing both cold and heat, but experimental evidence is lacking to support such a TRP channel U-shaped thermosensitivity12–17. Here we show that TRPA1 heat sensitivity is conserved in mammalians and for the first time provide experimental evidence of TRP channel inherent U-shaped thermosensitivity. Results and Discussion The purified hTRPA1 inserted into lipid-bilayers responded with single-channel activity when exposed to increasing temperatures from 22 °C to 40 °C, and as previously reported2 to noxious cold (Figs 1 and 2, Table 1). Based on the single-channel mean open probability (Po) (Fig. 2b), we calculated a Q10 value of 6 from the Arrhenius plot (25 °C to 35 °C), which is close to the Q10 value 7.5 of the heterologously expressed TRPM3, a recently identified heat-activated TRP ion channel present in capsaicin-sensitive primary afferents18. At 40 °C, there was still substantial channel activity although Po decreased, possibly indicating channel gating desensitization (Fig. 2b). 1 Department of Biochemistry and Structural Biology, Center for Molecular Protein Science, Lund University, SE-221 00 Lund, Sweden. 2Institute of Physiology and Pathophysiology, Friedrich-Alexander University Erlangen-Nürnberg, Universitätsstrasse 17, 91054 Erlangen, Germany. 3Department of Anesthesiology and Intensive Care, Hannover Medical School, Carl-Neuberg-Strasse 1, 30625 Hannover, Germany. 4Clinical Chemistry and Pharmacology, Department of Laboratory Medicine, Lund University, SE-221 85 Lund, Sweden. 5Department of Chemistry and Pharmacy, Friedrich-Alexander University Erlangen-Nuremberg, Egerlandstrasse 1, 91058 Erlangen, Germany. 6 IBGC, UMR 5095, Universite de Bordeaux, 1, rue Camille Saint Saëns, CS 61390, 33077 Bordeaux cedex, France. Correspondence and requests for materials should be addressed to P.M.Z. (email: ) Scientific Reports | 6:28763 | DOI: 10.1038/srep28763 1 www.nature.com/scientificreports/ Figure 1. The purified hTRPA1 is a warmth receptor. Temperatures above 22 °C evoked steady state outward and inward hTRPA1 single channel currents at a test potential of +60 and −60 mV as shown by representative traces and the corresponding amplitude histograms. Purified hTRPA1 was inserted into planar lipid bilayers and channel currents were recorded with the patch-clamp technique in a symmetrical K+ solution (c indicates the closed channel state). The single-channel mean conductance (Gs) did not increase with increasing temperatures (Table 1), suggesting that the TRPA1 channel pore is negatively affected by heat. As shown at 30 °C, hTRPA1 channel currents were observed at both positive and negative test potentials (Figs 1 and 2c), and the non-selective TRP channel pore blocker ruthenium red and the selective TRPA1 antagonist HC030031 inhibited temperature responses (Fig. 2d) without affecting bilayers (Supplementary Fig. 1). No currents were detected in bilayers without hTRPA1 when exposed to the same test temperatures (Supplementary Fig. 2). Because TRPA1 with its many cysteines is highly sensitive to thiol reactive agents including oxidants3, we asked if changes in redox state could affect the temperature sensitivity of hTRPA1, possibly explaining the many contradictory findings on mammalian TRPA1 and cold3 as well as the lack of mammalian TRPA1 heat responses in heterologous expression systems19–21. As shown by the Cy3-dye disulphide labeling assay, which has been used to study TRPA1 disulphide bond formation22, the purified hTRPA1 used for the bilayer patch-clamp recordings was partially oxidized, a condition that could be rectified by the thiol reducing agent dithiothreitol (DTT) and further oxidized by H2O2 (Fig. 3a). The consequences of changes in cysteine redox state for hTRPA1 channel activity are shown in Fig. 3b–d, where the reducing agents DTT and Tris (2-carboxy ethyl) phosphine (TCEP) inhibited cold and heat responses, and H2O2 evoked hTRPA1 channel activity at 22 °C that was blocked by TCEP. Furthermore, H2O2 decreased hTRPA1 channel activity triggered by 30 °C, and increased cold responses at 15 °C (Fig. 3c,d), possibly indicating different conformational states/gating mechanisms behind hTRPA1 heat and cold responses. Bilayers without hTRPA1 were unaffected by DTT, TCEP and H2O2 (Supplementary Fig. 1). The effects of these thiol modifying agents were studied at concentrations frequently used to explore TRP ion channel redox sensitivity without cytotoxic effects23–27. Taken together, the redox state of purified hTRPA1 determines its coldand heat responsiveness. (...truncated)