28.19: Signal Transduction - Temperature

Search Fundamentals of Biochemistry

Learning Goals 

 

(Learning goals written by Claude, Sonnet 4.6, Anthropic)

TREK Channels: Structure, Temperature Dependence, and Role in Thermosensation

• Describe the structural features of TREK family channels (TREK1, TREK2, TRAAK)—four transmembrane segments, two pore domains, dimeric assembly—and explain why their activation by heat produces membrane hyperpolarization and reduced thermoreceptor firing (opposing TRP channel effects), while their inhibition by cold produces depolarization and increased excitability, making them functionally complementary cold sensors.
• Explain the molecular basis of TREK1 thermosensitivity: describe how the C-terminal domain's interaction with inner leaflet phospholipids couples to the C-type gate through the M4 segment, why temperature sensitivity requires cell-attached or whole-cell recording configurations but is lost in inside-out and outside-out patches (implying requirement for intracellular factors or membrane integrity), and how PKA-mediated phosphorylation of the C-terminal region inhibits heat activation.
• Using knockout mouse data, distinguish the specific thermal ranges each TREK family member governs: explain how TREK1/TRAAK cooperate in sensing noxious cold and moderating heat pain thresholds below 50°C, how TREK2 primarily regulates thermoreception at moderate/innocuous cool temperatures (20–25°C) and C-fiber responses below 40°C, and how oxaliplatin-induced neuropathic cold hypersensitivity implicates TREK channels in pathological pain states.

TRP Channels: Thermosensory Specificity, Amplification, and Cold Sensing

• Describe the structural differences between TRP and TREK channels—TRP channels have six transmembrane segments, one pore domain, and assemble as tetramers—and explain why TRP channel activation produces membrane depolarization and increased thermoreceptor excitability (in contrast to TREK channels), noting that thermosensitive TRPs are characterized by extraordinarily high Q10 values (>20), compared to the Q10 ≥ 2–5 criterion for temperature-dependent channels generally.
• Organize the six thermosensitive TRP channels by their temperature activation thresholds and physiological roles: describe how TRPV4 (>27°C, warm, innocuous), TRPV3 (>34°C, warm-hot), TRPV1 (>43°C, noxious heat, capsaicin-sensitive, PCa/PNa ~10), and TRPV2 (>52°C, extreme noxious heat, Q10 ~100) form a heat-sensing series—and contrast with the cold-sensing TRPM8 (<25°C, menthol/WS-12-sensitive, activated by PIP2-dependent voltage-shift mechanism) and TRPA1 (<18°C, cold nociception in visceral neurons, cinnamaldehyde-sensitive, co-localizes with TRPV1 in trigeminal but not DRG neurons).
• Explain how redundancy and cooperation among TRP channel subtypes ensures robust noxious heat detection: describe evidence from triple TRPV1/TRPM3/TRPA1 knockout mice showing that complete elimination of acute noxious heat sensing requires ablation of all three channels, and explain how inflammatory mediators (ATP, bradykinin) lower TRPV1's threshold from 43°C to ~30°C, converting normally innocuous warm stimuli into painful ones.

Integration of TREK and TRP Channels in Thermosensation and Thermal Pain

• Explain how TREK and TRP channels functionally complement each other across the temperature spectrum: describe how TREK channels (potassium channels, hyperpolarizing) counteract TRP channel activation (cation channels, depolarizing) in pain-transducing fibers at elevated temperatures, and how the balance between these opposing conductances sets thermoreceptor firing thresholds—supported by evidence that TREK1/TRAAK double knockout mice show increased heat-responsive C-fiber proportions and action potential frequencies compared to wild-type.
• Describe the proposed molecular mechanisms of TRP channel thermosensitivity—including direct structural rearrangements acting as intrinsic temperature sensors, membrane tension changes, and second messenger generation (with evidence that TRPM8 and TRPA1 require phospholipase C activity for full cold responses in intact cells, yet TRPM8 retains some cold sensitivity in inside-out patches suggesting a partially membrane-delimited mechanism)—and distinguish these from TREK channel thermosensitivity, which requires cell integrity and is lost in excised patches.

This chapter section is taken in its entirety from: Ion Channels and Thermosensitivity: TRP, TREK, or Both? Lamas et al. Int. J. Mol. Sci. 2019, 20(10), 2371; https://doi.org/10.3390/ijms20102371. Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).

Introduction

Mammals and other animals expend considerable energy to maintain a nearly constant body temperature, regardless of ambient temperature. The mechanisms controlling thermal regulation are complex and often rely on negative feedback, in which it is first necessary to determine the body and ambient temperatures. External receptor cells, found mainly in the skin, can sense the environment's temperature. In contrast, body temperature is sensed by internal receptors expressed by cells in several internal organs. Traditionally, only the skin and core thermoreceptors (spinal cord and hypothalamus) have attracted researchers' attention. More recently, some very interesting information has emerged regarding visceral thermal receptors, including in humans. Although a hypothesis conceived many years ago, the terminals of receptor neurons are thought to contain branches of nerve fibers without any apparent structural specialization. Indeed, only recently have we begun to understand the molecular basis of thermoreception by cells.

Many biochemical processes, such as chemical reactions and physical processes, like conformational changes, are extraordinarily dependent on temperature. Although these processes generally occur faster at higher temperatures, the relationships can be very complex. If we consider the nervous system (NS), the effects of temperature on the resting membrane potential (RMP) were the first to be studied, as were its effects on the kinetics and speed of compound and single action potentials, long before ion channels were demonstrated.

All neurons and ion channels are affected by temperature changes, not least because channel gating is generally temperature-dependent. However, only some neurons are thermoreceptors, and very few ion channel types are thermosensors. Only channels with a temperature coefficient (Q10) of 2–5 are considered temperature-dependent. (Q10 is the ratio of a reaction at two different temperatures that differ by 10 °C. See Chapter 32.11 for more details.) Thermoreceptors are sensitive to temperature changes rather than the value of the temperature itself, probably due to their characteristic strong adaptation. These receptors are classified into two groups depending on whether their discharge frequency increases when heated or cooled, as shown in Figure \(\PageIndex{1}\). Based on this classification, it is common to speak of four thermal sensations (cold −10 to 15 °C, cool 16–30 °C, warm 31–42 °C, and hot 43–60 °C), whereby cold and hot are potentially noxious and/or painful.

Figure \(\PageIndex{1}\): Distribution of transient receptor potential (TRP) and TWIK-related potassium (TREK) channels as a function of their temperature threshold. Note that while temperature increases (orange), TREK channels are activated, but TRP channels may also be activated by lowering the temperature (blue).

The modulation of TWIK-related potassium (TREK) channels by temperature has been discussed in several reviews, yet very few have addressed this exciting topic exclusively. Conversely, after transient receptor potential (TRP) channels sensitive to temperature were discovered, they were studied extensively to understand how thermal stimuli were transduced. Such interest led to the publication of positive reviews on this issue. In this review, we will focus on the lesser-known role of TREK channels in thermosensation and compare their behavior with that of TRP channels. Other thermosensitive proteins, such as the Na⁺/K⁺ ATPase and epithelial sodium channels (ENaC), as well as P2X receptors (ion channels activated by ATP), have also been described. While these should also receive attention, we consider this to fall beyond the scope of this review. Indeed, cell thermosensitivity seems to be governed by the interplay of several channel types, as reported in hypothalamic neurons.

TREK Channels

The TWIK-related potassium channel (TREK) subfamily belongs to the two-pore domain potassium channel family (K2P). It comprises three members: TREK1, TREK2, and TRAAK (TWIK-related arachidonic acid-activated potassium channel). These are background potassium channels characteristically modulated by several physical and chemical stimuli, such as membrane stretch, pH, unsaturated fatty acids, general anesthetics, and temperature. TREK channels generally display very weak activity at room temperature and normal pressure, even when overexpressed in heterologous systems. However, their activity increases markedly when several different stimuli are applied, including temperature increases. From a physiological point of view, it is important to note that at 37 °C, all three members of the TREK subfamily respond to stimuli (pH, membrane stretch, or arachidonic acid), much like they do at room temperature. TREK channels may fulfill a dual role in the transmission of thermal pain. Thus, their strong activation by noxious heat results in an outward current that induces membrane hyperpolarization and reduces thermoreceptor firing, thereby relieving heat pain. Conversely, inhibition of TREK channels by noxious cold should depolarize thermoreceptors and increase their excitability, cooperating in the transduction of noxious cold sensations, as shown in Figure \(\PageIndex{2}\).

Figure \(\PageIndex{2}\): Proteins forming TREK channels have four transmembrane segments and two pore domains so that they assemble as dimers. Activation of TREK channels by increasing the temperature reduces the excitability of thermoreceptors (left). In contrast, TRP channels possess six transmembrane segments and only one pore domain, so they are tetramers. Activation of TRPs increases the excitability of thermoreceptors, but some are activated by an increase and others by a decrease in temperature (right).

TREK1

Soon after their discovery, it was shown that TREK1 channels are strongly and reversibly activated by temperature increases when expressed in heterologous systems (cell lines derived from kidney (COS) cells and oocytes). Let's consider that these are mostly voltage-independent channels open at resting potentials. TREK1 channels should function as cold sensors because low temperatures dampen their activity and depolarize these thermoreceptors (see Figure \(\PageIndex{2}\)). Many authors have demonstrated that macroscopic TREK1 currents are strongly outwardly rectifying at room temperature. While the outward current is not evident at 12 °C, it is strongly enhanced at 37 °C and increases progressively as temperature rises. Indeed, the current increases by around 7-fold with a 10 °C increase in the range of 14 to 42 °C, and, importantly, maximal sensitivity (0.9-fold per degree) was reached at nearly physiological temperatures, between 32 and 37 °C. The current induced by heating is also outwardly rectifying, reversing at potentials close to the equilibrium potential for potassium (E_K). In heterologous systems, the activation of TREK1 by temperature may be reversibly inhibited by cAMP, and this inhibition is suppressed by mutation of the C-terminal region that harbors a phosphorylation site for protein kinase A (PKA). Moreover, chicken embryonic atrial myocytes express TREK-like currents and have a resting membrane potential of around −20 mV in culture, which increases to −70 mV when the temperature rises to 35 °C, a change attributed to the activation of TREK1/2 channels. In voltage-clamp recordings, the outward current at +60 mV increased 9-fold. TREK1 and TWIK-related arachidonic acid-activated potassium (TRAAK) channels have been proposed to shut down hippocampal neuronal firing when temperatures rise too high.

Figure \(\PageIndex{a}\) shows an interactive iCn3D model of the mouse temperature sensitive K2P2.1 (TREK-1) potassium channels (6W84)

Figure \(\PageIndex{a}\): Mouse temperature sensitive K2P2.1 (TREK-1) potassium channels (6W84). Click the image for a popup or use this external link: https://structure.ncbi.nlm.nih.gov/i...U28PvTKakHpHW9

The red dots represent the outer leaflet; The gray spheres are potassium ions.

The threshold for the activation of slowly conducting C-fibers by noxious heat (30–50 °C) recorded in a skin-nerve preparation decreases in TREK1 knockout (KO) mice, and the range of activation of these fibers by heat corresponds closely to the range in which TREK1 is activated (30–45 °C). The number of action potentials in response to a heating ramp (30–50 °C) was higher in the KO mice, although the response of C-fibers to a cooling ramp (32–10 °C) was similar in native and KO mice. Indeed, TREK1 KO mice were hypersensitive to thermal pain up to 50 °C but not at higher temperatures (52–56 °C), suggesting that TREK1 channels may be important for the perception of low-threshold, but not high-threshold, thermal stimuli, for which TRP channels may be more important. Accordingly, the proportion of small diameter, cultured, dorsal root ganglia (DRG) neurons that respond to noxious heat (34%) increases in TREK1 KO (64%) and TREK1/TRAAK KO (74%) mice, as does the proportion of heat-responsive C-fibers in nerve-skin preparations from the single and double KO. TREK1 and TRAAK channels may counteract the stimulatory effect of heat-activated TRP channels in pain-transducing fibers when temperatures increase, such that the overall response may reflect a balance of the activity of these two functionally contrasting channel types. The threshold of thermoreceptors should certainly increase in the presence of TREK channels as temperatures increase.

Cooling DRG neurons in culture from 32 to 20 °C induces a depolarization of about 10 mV and often the firing of action potentials, an effect attributed to inhibition of a background potassium current. Accordingly, the inhibition of a native TREK1-like current may underlie the excitation (depolarization and firing) produced by cold in small, cultured trigeminal ganglion (TG) neurons, as seen in Figure \(\PageIndex{2}\). Interestingly, cold induces subthreshold oscillations in cold-sensitive DRG neurons. Transduction seems rather complex, involving dampening a hyperpolarization-activated cationic current and a permissive role for a slowly inactivating potassium current. Interestingly, the TREK1/TRAAK double knockout (KO) mutant exhibits consistent cold allodynia (pain due to a stimulus that does not normally provoke pain), and oxaliplatin, a cancer therapy that causes peripheral nerve neuropathy, exacerbates cold sensitivity in many patients and animals, inducing allodynia to cool temperatures.

Neither the deletion (or knockout - KO) of TREK1 nor TRAAK increases the fraction of small DRG neurons sensitive to noxious cold stimuli (below 20 °C and down to about 10 °C). However, the TREK1/TRAAK KO and the triple TREK1/TREK2/TRAAK KO significantly increased such neurons. Similar results were obtained when recording C-fibers in a skin-nerve preparation, in which case the double KO C-fibers fired more strongly than the single TRAAK KO and wild-type fibers. Oxaliplatin also induced hypersensitivity to noxious cold temperatures, while double and triple KO mice, but not the TREK1 KO mice, are hypersensitive to cold, which is not further affected by oxaliplatin. Hence, deleting two of the three TREK channels appears sufficient to achieve maximal hypersensitivity. The neuroprotective agent riluzole induces an analgesic effect against painful cold in normal animals and in oxaliplatin-pretreated TREK2 KO and TRAAK KO animals. However, riluzole did not affect pain sensitivity in TREK1 KO animals treated with oxaliplatin, in animals treated with the TREK1 inhibitor spadin, or in untreated TREK1 KO mice or triple KO animals. Similarly, a presumed outward TREK leak current recorded in DRG neurons was inhibited by riluzole and fluoxetine at 22 and 30 °C, but not at 14 °C, probably because the current was already inhibited at low temperatures. Together, these experiments suggest that TREK1 channels are essential for perceiving noxious cold and that TREK1 and TRAAK channels work together to sense cold.

Cell-attached patches demonstrated that the basal activity of expressed TREK1 channels is insignificant at room temperature, gradually increasing as temperature rises (17-fold for a 20 °C increase) and with a threshold around 25 °C. The current activated by temperature also displays outward rectification and reverses around the equilibrium potential for K⁺, although the single-channel conductance remains unaffected TREK1-like channels naturally expressed in cardiac ventricular myocytes and DRGs and recorded in cell-attached patches, do not open at 24 °C, yet they are very active at 37 °C  Surprisingly, temperature increases fail to modulate TREK1 activity in outside-out and inside-out patches, but under the same conditions, TREK1 is still strongly activated by arachidonic acid 

TREK1 channels are ideally positioned to act as thermosensors because they are expressed in structures involved in thermosensitivity and thermoregulation, such as dorsal root ganglia (DRGs), the trigeminal ganglion (TG), nodose ganglia (NG), and the anterior and preoptic hypothalamus. 

TREK2

Heterologously expressed TREK2 channels also produce strong outward rectification when recorded in whole-cell configuration at room temperature, which increases greatly at temperatures around 37 °C in several heterologous systems. In COS cells, a small TREK2 current was observed at 0 mV, which increased gradually with temperature up to about 40 °C. Notwithstanding, TREK2's response to abrupt temperature changes was rapid. Importantly, the IVs of the TREK2 current at different temperatures (24 and 37 °C) showed that the effect of temperature was not voltage-dependent: both inward and outward currents increased to the same degree. In this temperature range, the current increased 14-fold per 10 °C, indicating a very strong temperature dependence that was even greater than that of TREK1. Much like TREK1, TREK2 responds to temperature changes within the physiological range, with current activation reasonably well at 37 °C and at the resting membrane potential (RMP). Most experiments on TREK channels have been carried out at room temperature and 0 mV. However, in the future, these currents should be investigated using additional physiological parameters, at a resting potential and 37 °C, to provide a more precise understanding of their role in the behavior of central neurons. 

Figure \(\PageIndex{b}\) shows an interactive iCn3D model of the human two-pore domain temperature-sensitive potassium ion channel TREK2 (K2P10.1)(4BW5)

Figure \(\PageIndex{b}\): Human two-pore domain temperature-sensitive potassium ion channel TREK2 (K2P10.1)(4BW5). Click the image for a popup or use this external link: https://structure.ncbi.nlm.nih.gov/i...wSt9GE2pbDwfq5

Cerebellar granule and DRG neurons expressed native TREK2-like channels with weak activity at 24 °C in cell-attached patches. Yet, when the temperature increased to 37 or 41°C they became very active at all voltages (−80 to +80 mV). Moreover, cultured cortical astrocytes exhibit TREK2-like whole-cell outward currents that are strongly enhanced in the 23–40 °C temperature range. Interestingly, ischemia significantly augmented the outward current provoked by an increase in temperature in these astrocytes In addition, it was recently reported that TREK2 channels contribute about 10 mV to the RMP of DRG neurons at about 30 °C  Furthermore, single TREK2 and triple TREK1/TREK2/TRAAK knockout (KO) mice were more sensitive to warm temperatures (40–42 °C) when tested with the tail-flick reflex.

Using a skin-nerve preparation, it was demonstrated that the proportion of heat-sensitive C-fibers and their activity (the number of action potentials) increased in the TREK2 and triple KO mice when temperatures rose to noxious heat levels (ramped from 30 to 50 °C). In contrast, the temperature threshold for firing decreased. At high temperatures (between 40 and 50 °C), the triple but not the single KO fibers were more active than their wild-type counterparts, indicating that TREK2 regulates C-fiber responses at temperatures below 40 °C. In contrast, other TREK channels participate in these responses at higher temperatures. Both KOs suffered hyperalgesia at temperatures around 45 °C, but only the triple KO showed the same behavior above this temperature.

The withdrawal latency in the tail immersion test was reduced in both the TREK2 KO and the triple KO mice when tested at innocuous cooling temperatures (20–25 °C). As such, the KOs show enhanced sensitivity to temperatures within the normal range and yield similar results in the higher-temperature preference test. The percentage of C-fibers responding to moderate cold (30–21 °C) was higher in single and triple KOs when compared to those recorded from the nerve-skin preparation of wild-type mice. Interestingly, the cold threshold for C-fiber firing (21 °C) was lower in the triple KO (24 °C) but not in the TREK2 KO (23 °C) mice. Moreover, oxaliplatin induces mice to spend more time on a hot plate (30 °C) than on a cold plate (20–25 °C) compared to untreated animals, indicating that neuropathic mice have enhanced sensitivity to moderate cold. It has been suggested that TREK2 is implicated in the neuropathic hypersensitivity induced by this drug, and indeed, oxaliplatin nearly halved TREK2 mRNA levels in DRG neurons. Generally, the data suggest that TREK2 channels may be essential for controlling the C-fiber response to cold at moderate temperatures. The tail immersion test showed that triple KO mice were hypersensitive to noxious cold temperatures (15–5 °C), while the single TREK2 KO mice behaved much like the wild-type mice. Moreover, very similar results were obtained in the nocifensive dynamic cold plate test. Accordingly, it was suggested that TREK2 may not be important in noxious cold sensitivity but might be essential for thermoreception at moderately cool temperatures (25–20 °C).

A clear, fast, and reversible increase in activity was also reported for single TREK2 channels in cell-attached patches held at −40 mV as temperature increased (24 to 37 °C), with a threshold for this increase at 25 °C (from 24 °C) and no effect on conductance. It should be noted that in these circumstances, the activity of TREK2 single channels was very low at 24 °C. Significantly, neither TASK3 nor TRESK2 showed such a temperature dependence. However, as with TREK1, the activity of TREK2 in inside-out patches was not altered by changes in temperature (24 to 42 °C). Finally, it is important to consider that TREK2 channels are expressed strongly in the DRG, TG, and hypothalamus, yet less than TREK1 in the NG.

TRAAK

As with the other family members, TRAAK currents showed a strong open-channel outward rectification when recorded in whole-cell configuration, and these currents increased markedly as temperature rose (24 to 42 °C). Moreover, in culture, the percentage of small-diameter DRG neurons responding to noxious heat is increased in TRAAK and TRAAK/TREK1 knockout (KO) mice. Consistently in skin-nerve preparations, the percentage of fibers responding to heating (30–50 °C) and the number of action potentials in response to a heating ramp also increase, while the firing threshold is reduced. TRAAK and TRAAK/TREK1 KO mice suffer heat hyperalgesia when evaluated in the tail immersion test at 46–50 °C. Moreover, the double, but not the single, KO shows hypersensitivity at higher temperatures (52–56 °C) in the hot plate test.

Knockout of TRAAK did not alter the percentage of DRG neurons in culture that respond to noxious cold (12 °C). Moreover, in the cold plate assay, TRAAK KO mice behave like wild-type mice, whereas TREK1/TRAAK KO mice are more sensitive to cooling in the 10 to 20 °C range. The activity of single TRAAK channels heterologously expressed in COS cells, recorded in cell-attached patches at −40 mV, was very low at 24 °C. Yet, it increased progressively as the temperature rose from 24 to 37 °C. The activity threshold was around 30 °C, slightly higher than that reported for TREK1 and TREK2. However, the behavior of TRAAK channels in inside-out patches mimics that of TREK1 and TREK2, such that their activity was not affected by changes in temperature (from 24 to 42 °C). Native TRAAK-like channels in DRG neurons exhibited little activity at room temperature, but showed clear activity in all cell-attached patches at 37 °C [36, 48]. Finally, TRAAK channels are clearly expressed in the hypothalamus, TG, and DRG, but only weakly in the NG.

Molecular Origin of Thermosensitivity

When first discovered, mouse TREK1 was reported to have four transmembrane segments, two pore domains, and a 370-aa sequence. The activation of heterologously expressed TREK1 currents by increasing temperature is unaffected by the deletion of the cytoplasmic N-terminal region. By contrast, partial deletion of the C-terminal region (Δ103) or replacement of this region with that of TASK1 strongly dampens the activation of TREK1 by heat. The sensitivity of TREK1 channels to temperature can be eliminated by mutating helix 1 of the pore (G137I), suggesting that temperature affects TREK1 and TREK2 channels by modulating the C-type gate. It was suggested that functional coupling between the C-terminal domain and the C-type gate through the M4 segment is crucial for the heat sensitivity of the TREK1 channel. Thus, it is tempting to speculate that increasing the affinity of the C-terminal domain for phospholipids of the inner leaflet would increase the activity of TREK1 by heat, as proposed for other stimuli like stretch, PUFAs, phospholipids, or pH. Conversely, dissociating this domain from the membrane would result in TREK1 inhibition. Surprisingly, replacing the C-terminus of TREK2 with that of TASK3 did not reduce the channel's sensitivity to temperature changes in the range of 24 to 37 °C under similar conditions. However, it became insensitive to pH and arachidonic acid. Heat enhances the activity of TREK1, TREK2, and TRAAK in whole-cell and cell-attached recordings, but not in outside-out and inside-out patches, indicating that cell integrity, and probably also a second messenger, is necessary for this modulation. The contribution of TREK channels to maintaining the RMP has often been questioned; however, this assertion is largely based on experiments conducted at room temperature. Thus, new experiments should be performed at physiological temperatures to ascertain the role of these channels in both the RMP and neuronal excitability.

TRP Channels

Six transient receptor potential (TRP) channels are considered thermosensors, four responding to heat and two to cold. Temperature-sensitive TRP channels (Thermo-TRP) are highly temperature-dependent, exhibiting very high Q10 values (>20).

Heat-Sensitive TRP Channels

Four TRP subtypes are activated by increased temperature (see Figure \(\PageIndex{1}\)). Two of them respond to warm stimuli (TRPV4 Warm >27 °C and TRPV3 Warm >34 °C), and the other two to hot-painful stimuli (TRPV1 Hot >43 °C and TRPV2 Hot >52 °C).

TRPV1s are voltage and temperature-dependent channels that display outward rectification when expressed in human embryonic kidney (HEK) cells.  This is strongly enhanced by heating (to 48 °C) and by capsaicin.

Figure \(\PageIndex{c}\) shows an interactive iCn3D model of human TRPV1 with capsaicin at 48 degrees Celsius in an open state (7LPE)

Figure \(\PageIndex{c}\): Human TRPV1 with capsaicin at 48 degrees Celsius in an open state (7LPE). Click the image for a popup or use this external link: https://structure.ncbi.nlm.nih.gov/i...7Hsf5zpDQoQu99

At room temperature, the current passing through these channels is negligible below 0 mV, but at 42 °C, the channel is active between −100 and +50 mV. These cationic channels are ten times more permeable to Ca²⁺ than to Na⁺ (P_Ca/P_Na ~10) and are thought to be sensors for noxious heat but not activated by innocuous heat. Indeed, the response to noxious heat in mice lacking TRPV1 (KO) or in mice lacking DRG neurons was weaker. However, other channels may also contribute to the perception of noxious thermal stimuli, as heat still elicits receptor activation in several preparations. The NG sensory neurons that innervate the lungs produce an inward current in response to an elevation in temperature (from 23 to 41 °C, with a threshold around 35 °C and a Q10 of about 30 in the range of 35–41 °C) as well as membrane depolarization and action potential firing. This response was attributed to the presence of TRPV1 channels, although the involvement of TRPV2-4 could not be ruled out. We obtained similar results with NG neurons in culture, although these were slightly more complex, as a hyperpolarization was observed before depolarization and firing (unpublished data). Interestingly, inflammatory mediators like ATP and bradykinin strongly reduce the threshold of TRPV1 activation (30 °C) such that warm temperatures become painful. TRPV1 is strongly expressed in small-diameter sensory neurons of the DRG, TG, and NG and in the hypothalamus, sites where it may exert an important role in thermoreception.

TRPV2 is activated at extremely high temperatures (52 °C), although it is not affected by capsaicin and shows an outwardly rectifying IV curve and a P_Ca/P_Na ~3. This channel has a Q10 of around 100, and it is thought that temperatures that activate TRPV2 are more harmful than those that activate TRPV1. These channels are strongly expressed by myelinated medium- to large-diameter DRG neurons (Aδ and Aβ) and in the hypothalamus and the NG.

TRPV3 channels are activated at warm, near-hot temperatures (around 34–39 °C, with a Q10 of around 6), generating currents with pronounced outward rectification and a PCa/PNa ratio of ~12. They are capsaicin-insensitive channels that are stimulated by camphor and are thought to be involved in thermosensation and thermal nociception. Indeed, it has been suggested that TRPV3 channels contribute more to the speed at which mice select a more comfortable temperature than to the choice of the temperature itself. By contrast, TRPV4 channels are more likely to be involved in selecting a preferred temperature within a non-painful range. Interestingly, it was proposed that TRPV3 channels transduce thermal stimuli in skin keratinocytes, which then relay this information to sensory endings. TRPV3 channels are expressed in sensory DRG and NG neurons, but also in the hypothalamus. Interestingly, they co-localize with TRPV1 in DRG neurons.

TRPV4 are cationic (P_Ca/P_Na ~6) channels activated at even lower warm temperatures (around 27 °C, with a Q10 of about 10), generating outwardly rectifying currents and responding dynamically to temperature changes in the physiological range. These channels were proposed to play roles in thermosensation and thermoregulation, although some authors were unable to activate them by increasing temperature. Similarly, some behavioral studies reported a reduced response to temperature changes in TRPV4 knockout (KO) mice, though this effect was less clear in other studies. As with TRPV3, the expression of these channels in keratinocytes has been proposed to play an important role in transmitting thermal information, which probably contributed to the controversy generated. The sensitivity of this channel to temperature is lost in excised patches, suggesting that it requires a soluble intracellular factor. TRPV4 channels are expressed in DRG, TG, NG, and preoptic/anterior hypothalamic neurons. However, in the hypothalamus, they seem to be expressed in terminals rather than in the soma, so their role in thermoregulation is unclear.

TRPM2 (>35 °C), TRPM3 (>40 °C), TRPM4 (>15 °C), and TRPM5 (>15 °C) are channels that can also be activated by warming (see Figure \(\PageIndex{1}\)). Yet, they have received less attention, probably because it was initially thought that somatosensory neurons or keratinocytes did not express them. TRPM2 is voltage-insensitive, shows a P_Ca/P_Na ratio of ~1, activates at 35 °C, and has a Q10 of ~15. TRPM3 is broadly expressed, generates an outwardly rectifying current, has a PCa/PNa between 0.1 and 10, and activates at >40 °C with a Q10 of 7. It is important to note that TRPM3 has been described as part of a triad of TRPs, along with TRPV1 and TRPA1, that transduce acute noxious heat in mice. The combined ablation of these channels (triple KO) was necessary to achieve complete reduction of acute noxious sensing; single- or double-KO combinations resulted in deficits in heat responsiveness, but mice still exhibited vigorous withdrawal responses to noxious heat. Heat activation of TRPM2 and TRPM5 was obtained in inside-out patches, suggesting a membrane-delimited mechanism. Interestingly, TRPM2 activation seems to result from an increase in the IV slope, whereas that of TRPM4 and TRPM5 results from a shift of the activation curve toward more negative potentials. These last two channels are essentially not permeable to calcium.

Cold-Sensitive TRP Channels

Two TRP channels are activated by decreases in temperature (see Figure \(\PageIndex{1}\)). TRPM8 (<25 °C) is active in the cool range, while TRPA1 (<18 °C) senses cold-painful temperatures. Similarly, cool fibers (Aδ and C) have activation thresholds at about 30 °C, and cold fibers (C) have activation thresholds <20 °C. Accordingly, two populations of TG neurons were described by their activation thresholds as temperatures decreased: 30 °C and 20 °C for low and high thresholds, respectively. In general, cold fibers fire continuously at normal skin temperatures, increasing their firing frequency when the skin is cooled and shutting down when the skin is warmed. In addition, cold fibers can adapt to small decreases in temperature.

TRPM8 channels are voltage-dependent cationic channels permeable to Na⁺, K⁺, Cs^-_, and Ca²⁺ (P_Ca/P_Na ~3). When expressed in HEK cells and recorded in the whole-cell configuration, they show a voltage-dependent outwardly rectifying current that increases markedly upon cooling from 30 to 15 °C or upon application of menthol.

Figure \(\PageIndex{d}\) shows an interactive iCn3D model of TRPM8 ion channel in complex with the menthol analog WS-12 and PI(4,5)P2 (6NR2)

Figure \(\PageIndex{d}\): TRPM8 ion channel in complex with the menthol analog WS-12 and PI(4,5)P2 (6NR2). Click the image for a popup or use this external link: https://structure.ncbi.nlm.nih.gov/i...94muQHwYr7eij6

The methanol analog is shown in spacefill and CPK colors. PI(4,5)P2 is shown in spacefill just below the lower cytoplasmic leaflet.

The TRPM8 receptor is a Ca²⁺ cation channel. Cooling compounds like menthol and WS-12 depend on allosteric interactions and on the membrane phosphatidylinositol 4,5-bisphosphate (PIP2).

Importantly, basal and cold-stimulated currents reverse around 0 mV and were almost negligible below this potential. Cooling CHO cells expressing TRPM8 (25 to 15 °C) also increases intracellular calcium, and the Q10 in the 25 to 18 °C range is around 24. The effect of temperature is due to an increase in the open probability and a shift in the conductance–voltage relationships along the voltage axis. Similar results were obtained in inside-out macropatch recordings, although stimulation occurred at lower temperatures, suggesting that cell integrity is important but not indispensable. The role of this channel as a detector of painful cold has been questioned in experiments on KO mice. Still, it is accepted that it is an important cold sensor in vagal, TG, and especially DRG afferents. It was predicted that cold transduction may require the activation and inhibition of several different ion channels (see Figure \(\PageIndex{3}\)), such as TRP, TREK, and ENaC channels. If this were the case, TRP channels would probably be more important in the noxious-cold range, whereas TREK channels might participate more strongly in the cool range of temperatures (see Figure \(\PageIndex{1}\)). TRPM8 is expressed in small-diameter DRG and TG neurons, presumably thermoreceptors, yet it seems not to co-localize with TRPV1.

Figure \(\PageIndex{3}\): Most temperature-modulated channels are expressed in both somatic and visceral afferents. The expression of TREK2, TRAAK, and TRPM8 in visceral ganglia appears to be low compared to other channels. However, investigating the colocalization of these channels in the same neuron is necessary to gain a comprehensive understanding of their relative importance in temperature sensing.

TRPA1 is activated by lower temperatures than TRPM8 (<18 °C), and while it would be expected to be involved in cold nociception, this is not that clear. TRPA1 generates an outward-rectifying cationic current under both control conditions and when cold-activated (about 10 °C), with similar permeability to Ca²⁺ and Na⁺ (P_Ca/P_Na ~1). Cinnamaldehyde can selectively activate currents through this channel in native DRG neurons, as can bradykinin (when co-expressed with BK receptors), strongly suggesting a role in sensing nociceptive stimuli. However, TRPA1 knockout (KO) mice do not seem to have difficulty sensing cold stimuli through the skin, whereas the response of TRPM KO mice to cold is significantly dampened. By contrast, about 50% of NG neurons in culture were activated by cooling (<24 °C), mainly through TRPA1 channel activation (increase in [Ca]_i, depolarization, and AP firing). Interestingly, about 10% of the NG neurons responded to cold through a TRPA1- and Ca-independent pathway. TRPA1 often co-localizes with TRPV1, and this may explain the paradoxical hot sensation elicited by an extremely cold stimulus. Interestingly, most NG neurons are sensitive to cold and heat. TRPA1 is expressed in DRG and TGs, while TRPA1 and TRPM8 are not co-expressed in DRG neurons, but they are in TG neurons. In summary, the data suggest that TRPA1 is the principal ion channel involved in cold sensation in visceral (NG) neurons, whereas TRPM8 appears to play the same role in somatic neurons.

Molecular Origin of Thermosensitivity

The mechanism by which temperature modulates TRP channels is still unclear, yet several hypotheses have been proposed.  Temperature changes could:

• produce a ligand that binds to a receptor and affects the channel;
• produce a structural change in the channel that provokes its opening;
• affect the membrane's structure, causing changes in tension that would, in turn, affect ion channels.

Because capsaicin induces burning pain, it has been hypothesized that capsaicin and heat may use a common mechanism to activate TRPV1 and produce pain. Both stimuli affect excised patches, and it is generally accepted that TRPV1 is directly activated by noxious heat, making it a true heat sensor.

The fact that TRPM8 can be activated by cooling in inside-out patches suggests that the mechanism is membrane-delimited, arguing against the involvement of a second-messenger pathway. Notwithstanding, phospholipase C inhibition strongly dampened the calcium increase provoked by cold stimuli in TRPM8-expressing CHO cells. Cooling activates TRPM8 channels by shifting the voltage dependence of activation toward more negative values; the same mechanism underlies TRPV1 activation by heat, but not TRPM2 activation. It has been proposed that temperature induces large-scale rearrangements of the protein, and that a temperature-sensing domain, or “temperature sensor,” exists in the structure of TRPM8 channels. Much like for TRPM8, inhibition of phospholipase C strongly reduces the increase in calcium provoked by cold stimuli in TRPA1-expressing CHO cells.

Conclusions

There are several important differences between the two main types of thermosensor channels that have been reviewed in this article (see Figure \(\PageIndex{2}\)). First, TREKs are potassium channels with negative reversal potentials, so their activation would reduce thermoreceptor activity. By contrast, as the reversal potential of TRPs (cationic channels) is close to 0 mV, their activation will increase thermoreceptor excitability. Second, the three TREK channels appear to increase their open probability with increasing temperature. At the same time, there are two possible situations in the case of TRPs: one in which their activity increases as the temperature increases, and another in which activity increases when the temperature decreases (see Figure \(\PageIndex{2}\)). Although the activation of these channels generates opposing effects on thermoreceptors (depolarization versus hyperpolarization), the information available to date regarding the participation of TRP and TREK channels in thermosensitivity strongly suggests that both types of channels collaborate and complement each other to generate the sensations of heat, cold, and thermal pain. In support of this hypothesis, TREK, and TRP channels are very often co-expressed in thermoreceptors and other sensory neurons TRP channels are generally accepted as the primary thermosensors; however, several lines of evidence indicate that other channels are necessary to explain the full plethora of mechanisms involved in thermosensation. TREK2 channels appear important in thermoreception at moderate temperatures and sensing innocuous cold but not aversive cold, while TREK1 and TRAAK acting together may be important in sensing painful cold.

Summary

(Summary written by Claude, Sonnet 4.6, Anthropic)

Mammalian thermosensation is mediated by two structurally and functionally contrasting families of ion channels—the two-pore domain TREK potassium channels and the transient receptor potential (TRP) cationic channels—that cooperate across the thermal spectrum to generate the sensations of cool, warm, heat pain, and cold pain.

TREK family channels (TREK1, TREK2, and TRAAK) are dimeric background potassium channels with four transmembrane segments and two pore domains per subunit. They are progressively activated as temperature rises, producing outward K⁺ currents that hyperpolarize thermoreceptor neurons and reduce their firing frequency. This makes TREK channels functionally inhibitory at elevated temperatures—thereby dampening heat pain signaling—whereas their cold-induced inhibition depolarizes neurons and increases excitability, thereby contributing to cold sensing. TREK1 thermosensitivity requires phospholipid interactions of the C-terminal domain coupled to the C-type gate via the M4 transmembrane segment and is lost in excised patches, indicating a requirement for intact cell membranes or intracellular factors. PKA-mediated phosphorylation of the C-terminal region inhibits heat activation, linking TREK1 to cAMP signaling. Knockout mouse studies reveal the specific thermal domains each member governs: TREK1 and TRAAK cooperate in sensing noxious cold and in moderating heat pain thresholds below 50°C (with TREK1/TRAAK double knockout mice showing significant cool allodynia); TREK2 primarily regulates thermoreception at moderate innocuous cool temperatures (20–25°C) and C-fiber responses at temperatures below 40°C, with TREK2 knockout increasing sensitivity to warm temperatures and innocuous cooling. The cancer drug oxaliplatin, which reduces TREK2 mRNA in DRG neurons, induces neuropathic cold hypersensitivity—an effect attenuated by riluzole in a TREK1-dependent manner—thereby illustrating the clinical relevance of TREK channel thermobiology.

TRP thermosensor channels are tetramers with six transmembrane segments and one pore domain that open in response to specific temperature ranges, producing inward cationic currents that depolarize thermoreceptor neurons and increase excitability. Their extraordinary temperature dependence (Q10 values >20, sometimes approaching 100 for TRPV2) reflects highly cooperative conformational transitions that allow sharply threshold-dependent activation. Four heat-sensing TRPs cover the warm-to-noxious range: TRPV4 (>27°C, warm, innocuous range, expressed in keratinocytes and sensory neurons), TRPV3 (>34°C, warm-hot, camphor-sensitive), TRPV1 (>43°C, noxious heat, capsaicin-sensitive, PCa/PNa ~10, expressed in small-diameter DRG/TG/NG neurons), and TRPV2 (>52°C, extreme noxious heat, expressed in myelinated Aδ/Aβ neurons). TRPV1 is the best-characterized noxious heat sensor, and inflammatory mediators—ATP, bradykinin—sensitize it by lowering its threshold to ~30°C, thereby converting innocuous warm stimuli into painful ones. Triple knockout of TRPV1/TRPM3/TRPA1 is required to completely abolish acute noxious heat sensing, demonstrating substantial redundancy among heat-sensing TRPs.

Two cold-sensing TRPs operate at lower temperatures: TRPM8 (<25°C) is the principal cold sensor in somatic DRG and TG neurons, activated by menthol and the analog WS-12 through PIP2-dependent shifts in voltage-dependent activation, with partial cold sensitivity retained in inside-out patches suggesting a partially membrane-delimited mechanism; and TRPA1 (<18°C), which is co-expressed with TRPV1 in trigeminal neurons and appears to be the principal cold sensor in visceral NG neurons, activated by cinnamaldehyde and bradykinin and suppressed by phospholipase C inhibition. The paradoxical burning sensation in extreme cold may reflect TRPA1/TRPV1 colocalization and coactivation.

The overall picture that emerges is one of complementary, overlapping thermosensory systems: TREK channels set inhibitory brakes on thermoreceptor excitability and their contribution is greatest in the innocuous cool-to-warm range; TRP channels provide excitatory drive with extraordinary thermal sensitivity at their specific threshold temperatures; and the balance between these opposing conductances—both often co-expressed in the same sensory neurons—determines the thermoreceptor's ultimate firing pattern, thermal threshold, and contribution to the perception of temperature and thermal pain.