Drosophila painless is a Ca2+-requiring channel activated by noxious heat

Abstract

Thermal changes activate some members of the transient receptor potential (TRP) ion channel super family. They are primary sensors for detecting environmental temperatures. The Drosophila TRP channel Painless is believed responsible for avoidance of noxious heat because painless mutant flies display defects in heat sensing. However, no studies have proven its heat responsiveness. We show that Painless expressed in human embryonic kidney-derived 293 (HEK293) cells is a noxious heat-activated, Ca(2+)-permeable channel, and the function is mostly dependent on Ca(2+). In Ca(2+)-imaging, Painless mediated a robust intracellular Ca(2+) (Ca(2+)(i)) increase during heating, and it showed heat-evoked inward currents in whole-cell patch-clamp mode. Ca(2+) permeability was much higher than that of other cations. Heat-evoked currents were negligible in the absence of extracellular Ca(2+) (Ca(2+)(o)) and Ca(2+)(i), whereas 200 nm Ca(2+)(i) enabled heat activation of Painless. Activation kinetics were significantly accelerated in the presence of Ca(2+)(i). The temperature threshold for Painless activation was 42.6 degrees C in the presence of Ca(2+)(i), whereas the threshold was significantly increased to 44.1 degrees C when only Ca(2+)(o) was present. Temperature thresholds were further reduced after repetitive heating in a Ca(2+)-dependent manner. Ca(2+)-dependent heat activation of Painless was observed at the single-channel level in excised membranes. We found that a Ca(2+)-regulatory site is located in the N-terminal region of Painless. Painless-expressing HEK293 cells were insensitive to various thermosensitive TRP channel activators including allyl isothiocyanate, whereas mammalian TRPA1 inhibitors, ruthenium red, and camphor, reversibly blocked heat activation of Painless. Our results demonstrate that Painless is a direct sensor for noxious heat in Drosophila.

Figure 1.

Heat-evoked activation and ion selectivity of Painless. A, A representative image of a Painless-expressing HEK293 cell immunostained with anti-V5-tag antibody. B, Representative fura-2 Ca²⁺ imaging shows [Ca²⁺]_i increase in Painless-expressing HEK293 cells during heating (∼42°C). Pseudocolor indicates intensity of the fluorescence ratio of 340/380 nm. Ionomycin (5 μm) with 10 mm CaCl₂ (Iono) was applied to confirm cell viability. DsRed (red color) indicates Painless-expressing cells. C, Quantitative changes in the fura-2 ratio in Painless-expressing cells during heating. A clear [Ca²⁺]_i increase was observed in the presence [Ca²⁺(+), red bar], but not in the absence [Ca²⁺(−), blue bar] of 2 mm extracellular CaCl₂. Average traces from 111 cells (left) or 73 cells (right) ±SD are shown in the top panels. Bottom panels show temperature changes. D, Heat elicits inward current activation in a Painless-expressing HEK293 cell at −60 mV holding potential in a whole-cell patch-clamp mode (n = 12). A standard bath solution and Cs-Asp/Ca²⁺(−) pipette solution were used. E, Current–voltage relationship of heat-evoked current exhibits dual-rectification with positive reversal potential. Heat-dependent shift of the liquid junctional potentials (ΔJP_H) were not corrected in the plot. The reversal potential was 28.8 ± 1.0 mV after compensation of ΔJP_H (n = 8). A standard bath solution and Cs-Asp/Ca²⁺(−) pipette solution were used. F, Heat-evoked currents exhibit high Ca²⁺ permeability (n = 6–7). NaCl, CsCl, MgCl₂, or CaCl₂ bath solutions and KCl pipette solution were used. Note that *Na⁺ (black trace) was obtained using NaCl bath and KCl/200 nm Ca²⁺ pipette solution. Basal traces were subtracted but ΔJP_H values were not compensated in the plot. After compensation of ΔJP_H values of the reversal potentials, permeability ratios were calculated (P_Na:P_Cs:P_K:P_Mg:P_Ca = 1:0.84:1.13:4.87:41.66).

Figure 2.

Ca²⁺-requiring activation of Painless. A, Painless exhibits only faint currents during heating in the absence of Ca²⁺_o and Ca²⁺_i. Cs-Asp/Ca²⁺(−) pipette solution [(−) in the pipette] and Ca²⁺(−) bath solution [(−) in gray area] were used. Functional Painless expression was confirmed by heat application in standard bath solution (Ca²⁺ in gray area). B, Heat elicits inward currents in the presence of Ca²⁺_o and Ca²⁺_i. After inactivation, Painless shows small currents after a second heating. Cs-Asp/200 nm Ca²⁺ pipette solution (Ca²⁺ in the pipette) and standard bath solution (Ca²⁺ in gray area) were used. C, Heat elicits inward currents in the absence of Ca²⁺_o and the presence of Ca²⁺_i. After inactivation, Painless shows small currents after a second heating. Cs-Asp/200 nm Ca²⁺ pipette solution (Ca²⁺ in the pipette) and Ca²⁺(−) bath solution [(−) in gray area] were used. Currents in A–C are typical examples in a whole-cell patch-clamp mode. D, Ca²⁺_i facilitates activation kinetics of heat-evoked currents. Colored traces correspond to the ones shown in A–C. The arrow indicates the initial points of the currents (0%). One hundred percent of the current means maximal activation. Half of the maximal current is indicated as a gray line. E, Quantification of the time required for 33% (Δt_1/3), 50% (Δt_1/2), and 100% (Δt_max) of maximal activation of Painless by heat. Colored bars correspond to the ones shown in A–C. Data represent mean ± SEM. **p < 0.01 (n = 9–14). F, Heat responsiveness of Painless depends on Ca²⁺_i concentration. Cs-Asp pipette solution including 1–10⁴ nm Ca²⁺ (Ca²⁺ in the pipette) and Ca²⁺(−) bath solution [(−) in the gray area] were used. Maximal values of current density were obtained and fitted to Hill plots. The Ca²⁺_i concentration required for eliciting half of the maximal current was 103.4 ± 6.2 nm (Hill coefficient of 3.4 ± 1.8). Data represent mean ± SEM (n = 8–18). The background currents were taken in each [Ca²⁺]_i by applying heat to mock-transfected HEK293 cells and were subtracted from each point.

Figure 3.

Temperature thresholds of Painless activation. A, A representative temperature-response profile for heat-evoked Painless current in the presence of Ca²⁺_o alone at −60 mV holding potential. The dotted line indicates basal level. B, An Arrhenius plot for heat-evoked Painless current shows a clear flex point on temperature dependency (data in A were converted). The crossing point of the two linear-fitted lines (a flex point) was defined as a temperature threshold for Painless activation. The Q₁₀ value was calculated for each line (see Materials and Methods). C, Temperature thresholds for Painless activation are significantly lower in the presence of Ca²⁺_i. Data represent mean ± SEM. **p < 0.01 (n = 9–14). All of the values were obtained from the same cells analyzed in Figure 2E. D, Temperature thresholds do not depend on membrane potential in the presence of Ca²⁺_o and Ca²⁺_i. Data represent mean ± SEM (n = 9–11). E, A representative trace shows activation currents of Painless elicited by slow heat application (0.2°C/s) in the presence of Ca²⁺_o and Ca²⁺_i. The dotted line indicates a temperature threshold. F, Temperature thresholds do not depend on heat application rate in the presence of Ca²⁺_o and Ca²⁺_i. Painless was stimulated with slow (0.2°C/s) or fast (1°C/s) heat application. Data represent mean ± SEM (n = 12–14).

Figure 4.

Temperature thresholds in repeated exposure to heat. A, Multiple currents are observed in the repeated heating protocol (see Results). Representative heat-evoked currents in the presence of Ca²⁺_o and Ca²⁺_i are shown. Heat application was terminated when the current reached ∼0.5 nA (dotted lines). More than three responses at equivalent currents could be observed without desensitization. B, C, Temperature thresholds are reduced after repeated heat with (B), but not without (C), Ca²⁺_o. Standard (Ca²⁺ in gray area) or Ca²⁺(−) bath solution [(−) in gray area] and Cs-Asp/200 nm Ca²⁺ pipette solution (Ca²⁺ in the pipette) were used. D, E, Raw data of temperature thresholds for the first to third heating cycle are shown in individual cells (n = 10). Standard (D) or Ca²⁺ (−) bath solution (E) and Cs-Asp/Ca²⁺ 200 nm pipette solution were used. Thresholds were calculated as in Figure 2C. F, G, The temperature thresholds are significantly reduced after repeated heating in the presence of Ca²⁺_o and Ca²⁺_i (F), but not in the presence of Ca²⁺_i alone (G). Data represent mean ± SEM. **p < 0.01 (n = 10). All of the colored traces, dots, and bars correspond to the first, second, and third heat applications indicated in F and G.

Figure 5.

Single-channel activation of Painless during heating. A, The top trace shows activation currents in Painless-expressing excised membrane during slow heating (0.2°C/s) in an inside-out patch-clamp mode at +60 mV holding potential (n = 4). The dotted line indicates an initiation point of the currents. Standard pipette solution (Ca²⁺ in the pipette) and Cs-Asp/200 nm Ca²⁺ bath solution (Ca²⁺ in gray area) were used. The basal trace in a and single-channel currents in b are magnified from corresponding lines in the top trace. The dotted line in b indicates the closed-channel level. B, Painless is robustly activated by heat in the presence of Ca²⁺ on the cytoplasmic side (n = 6). Ca²⁺(−) pipette solution [(−) in the pipette] and Cs-Asp/Ca²⁺(−) bath solution [(−) in gray area] or Cs-Asp/200 nm Ca²⁺ bath solution (Ca²⁺ in gray area) were used. Note that movements in basal lines including leak always occurred in the absence of Ca²⁺_o and Ca²⁺_i, but those are clearly different from single-channel currents observed in the presence of Ca²⁺_i. Moreover, small but apparent single-channel currents were evoked as soon as Ca²⁺ was applied to the cytoplasmic side before heat application (arrowhead).

Figure 6.

Effects of N-terminal mutation on heat responsiveness of Painless. A, B, Mutant Painless N363A shows small currents in the presence of Ca²⁺_o alone (A), but exhibits large currents in the presence of Ca²⁺_o and Ca²⁺_i (B). Standard bath solution (Ca²⁺ in gray area) and Cs-Asp/Ca²⁺(−) pipette solution [(−) in the pipette] (A) or Cs-Asp/Ca²⁺ 200 nm pipette solution (Ca²⁺ in the pipette) (B) were used. Current traces are typical examples in a whole-cell patch-clamp mode (n = 10–14). C, Temperature thresholds for mutant Painless N363A are significantly higher than those for wild-type Painless in the presence of Ca²⁺_o alone or in the presence of Ca²⁺_o and Ca²⁺_i. Data represent mean ± SEM. **p < 0.01 (n = 9–14). D, [Ca²⁺]_i sensitivity of mutant Painless N363A was decreased. Cs-Asp pipette solution including 1–10⁵ nm Ca²⁺ and Ca²⁺(−) bath solution were used. Maximal values of current density were obtained and fitted to Hill plots. Gray points and the fitted line indicate wild-type Painless (see also Fig. 2F). [Ca²⁺]_i required for eliciting half of the maximal current was 249.3 ± 59.8 nm (Hill coefficient of 0.9 ± 0.2). Data represent mean ± SEM (n = 9–15). The background currents were taken in each [Ca²⁺]_i by applying heat to mock-transfected HEK293 cells and were subtracted from each point.

Figure 7.

The effects of mammalian TRPA1 agonists and antagonists on Painless. A, B, Painless-expressing cells are not activated by AITC (2 mm; A) or cold stimulation (∼10°C; B) (n = 3–5). Standard bath solution and Cs-Asp/200 nm Ca²⁺ pipette solution were used. Functional Painless expression was confirmed by heat application. C, D, Heat activation of Painless was reversibly blocked by ruthenium red (RuR, 10 μm; C) or camphor (3 mm; D) treatment (n = 6–8). Standard bath solution and Cs-Asp/200 nm Ca²⁺ pipette solution were used. The repeated heating protocol (Fig. 4) was used to validate the function of Painless before and after the treatment with antagonists. Heat termination points are indicated as dotted lines. Cells were treated with each antagonist, followed by application of heated bath solution containing antagonist. The inhibitory effects of ruthenium red were incompletely reversed. Higher concentrations of camphor could not be applied because camphor severely damaged the gigaohm seal.