The I_KD current in cold detection and pathological cold pain

Exposure of body surface to temperature reductions evokes a wide variety of cold sensations, from comfortable cooling to intense pain depending on the intensity, speed and duration of the stimulus. In the trigeminal and spinal somatosensory system, cold thermoreceptors (low- and high-threshold) and cold nociceptors are the primary sensory neurons responsible for detecting low environmental temperatures. The underlying neural and molecular machinery of cold sensing is far from simple. Cold transduction and subsequent propagation of action potential result from the concerted action of several classes of transduction and voltage-gated channels that functionally coexist to shape the net response of cold-sensitive nerve fibers. Of these, several members of Kv and K_2P families of potassium channels have been found to play a key modulatory role in cold detection. For example, one of the most critical factors determining cold sensitivity in primary somatosensory neurons is the counterbalance of two functionally antagonist actors: TRPM8 channels, the main molecular sensor responsible for the cold-activated excitatory current, and Shaker-like Kv1.1-1.2 potassium channels, the molecular counterpart underlying the excitability brake current I_KD. This fast-activating and slow-inactivating outward potassium current dampens the depolarizing effect of TRPM8-dependent excitatory current in response to cold stimulation, shifting the thermal threshold of individual neurons to lower temperatures.^1,2

Following peripheral nerve damage, cold sensitivity can be significantly altered as a consequence of dysfunctions of both their transducing and integrating machineries. Abnormal thermal signaling including exacerbated cold-sensitivity is a frequent clinical symptom accompanying neuropathies that follow post-surgical damage of peripheral nerves, post-herpetic neuralgia, ciguatoxin in marine food poisoning, diabetes and chemotherapy, among others.³ Painful hypersensitivity to innocuous cold, or cold allodynia, is a common and debilitating symptom of neuropathic and inflammatory pain induced by the injury of afferent fibers. Our understanding of the molecular and cellular bases of this disabling sensory alteration is still emerging, and they seem to vary among the different diseases considered. At the molecular level, several classes of ion channels, including thermo-TRP channels, voltage-gated and background K⁺ channels, voltage-gated Na⁺ channels and HCN channels have been proposed as important players in development and persistence of pathological cold pain in response to different forms of axonal damage.^3,4 Of these, Kv1 channels responsible for the brake potassium current I_KD arise as interesting candidates.

The I_KD is a thermo-insensitive hyperpolarizing current that exerts its action at membrane potentials below the threshold for the action potential firing. During cold stimuli, the I_KD current can be activated by the cold-induced depolarization, dampening the membrane potential and reducing the response of the neuron to temperature reductions. Individual neurons responding to small temperature drops normally present low levels of I_KD and high functional expression of TRPM8. These neurons correspond to low-threshold cold thermoreceptors signaling pleasant cold. On the other hand, neurons with a prominent I_KD current and a lower expression of the cold-sensitive TRPM8 channel only respond to lower temperatures, and correspond to high-threshold cold-thermoreceptors signaling cold discomfort. Interestingly, the brake current I_KD not only reduces the overall response of individual neurons during temperature drops, but also prevents the unspecific activation by cold of primary sensory neurons of other somatosensory modalities.^1,2 In this scenario, a functional downregulation of the channels underlying the I_KD could not only increase the excitability of different populations of primary afferents, but also could maintain this exacerbated state for long periods of time. This is important, since these changes do not undergo the common desensitization that is observed in depolarizing ion channels, such as thermo-TRP channels. The fact that disturbances in the functional expression pattern of several voltage-gated K⁺ channels in response to axonal damage have been reported, including downregulation of several Kv1 channels, opens the possibility that nociceptive neurons (which are normally insensitive to non-noxious cold) become sensitive to mild temperature drops due to a functional downregulation of I_KD in response to nerve damage, contributing to the different manifestations of pathological cold pain.

In a recent study, we explored the role of the brake I_KD current in cold allodynia induced by peripheral nerve injury, and the contribution of nociceptive neurons in damage-triggered painful hypersensitivity to innocuous cold.⁵ Using a well-known form of axonal damage by chronic constriction of the sciatic nerve that induces cold allodynia in mice, and combining behavioral analysis, calcium imaging, patch-clamping, pharmacological tools and mathematical modeling, we explored the possibility that this sensory alteration may occur as a result of a functional downregulation of this current in response to nerve damage. Using calcium imaging in dissociated neurons, we found that cold allodynia induced by chronic constriction injury (CCI) of the sciatic nerve is related to both, an increase in the proportion of cold-sensitive neurons (CSNs) in dorsal root ganglia (DRG) contributing to this nerve, and also a shift in the temperature threshold of individual neurons to higher temperatures.⁵ Interestingly, we found that the pharmacological suppression of I_KD using 4-AP (or α-DTx) in nerve terminals of intact animals at the hindpaw induces large acute nocifensive responses to innocuous cold in these mice, mimicking the allodynic phenotype of CCI animals. In injured mice, innocuous cold stimulation evokes significant nocifensive responses that are only slightly enhanced by pharmacological intraplantar blockage of I_KD.⁵ Thus, in intact animals, the sensitization induced by this maneuver suggests that a local reduction of I_KD may induce cold-evoked nocifensive behaviors (similar to those observed in mice with chronically injured nerves) in response to innocuous cold stimulation. Similarly, as CCI mice appear to be less sensitive to the pharmacological I_KD blockage, we hypothesize that the functional expression of the molecular target of these blockers (i.e., Kv1 channels underlying I_KD) could be downregulated in injured animals. Accordingly, we found that the shift of the cold threshold of individual CSNs to higher temperatures induced by the pharmacological blockage of I_KD using 4-AP or α-DTx is reduced in a large proportion of CSNs from injured animals.

In CSNs from injured mice, the mean cold threshold is shifted approximately 2°C to higher temperatures compared with sham animals, a difference that is very similar to the shift induced by pharmacological suppression of I_KD in CSNs from control mice. Moreover, since a significantly large subgroup of CSNs from CCI animals appears to be insensitive to 4-AP (and α-DTx), it is reasonable to think that the I_KD current could be functionally downregulated in these neurons in response to injury. In fact, we found that the mean I_KD current density is reduced in CSNs from injured mice, measured at a membrane potential subthreshold to the action potential firing, suggesting that a reduction of the I_KD current density in injured mice contributes to the increased cold sensitivity observed in DRG neurons after nerve damage. This reduction would explain both the shift in thermal threshold and its lower sensitivity to 4-AP and α-DTx.⁵

As we mentioned before, the pharmacological suppression of I_KD induces cold sensitivity in a subpopulation of cold-insensitive primary somatosensory neurons. These transformed neurons (i.e., recruited as cold-sensitive ones by blockage of I_KD) comprise almost 10% of the total neuronal population in DRG in control animals. The I_KD current density in these neurons is greater than that observed in cold thermoreceptors.^1,5 Since 80% of transformed neurons present an electrophysiological profile indicative of nociceptors, these cells would correspond to nociceptive neurons normally damped by the I_KD. We reasoned that, if this current is downregulated in response to injury in this particular neuronal subpopulation, part of these neurons would become cold-sensitive in the non-noxious range in CCI mice. Consistent with this idea, we found that the nociceptive phenotype is more frequently observed in CSNs from injured animals, suggesting that these neurons are now able to respond to mild temperature drops signaling pain.⁵ Thus, the population of cold-insensitive neurons that can be transformed into cold-sensitive ones is reduced in injured animals, and the analysis of the electrophysiological properties and neurochemical profile of CSNs reveals an increase of nociceptive-like phenotype among CSNs from CCI animals compared with sham mice. We concluded that the functional downregulation of I_KD influences the thermal sensitivity of cold-insensitive neurons that have the potential to respond to mild cold due to the reduction of this excitability brake, also shifting the thermal threshold of canonical low- and high-threshold cold thermoreceptors to higher temperatures.⁵

Notably, cold-induced TRPM8-dependent current density in CSNs from both groups of mice was similar, suggesting that changes in the functional expression of TRPM8 are not necessary for cold allodynia in response to this form of nerve damage. Nevertheless, the cold-induced responses in both groups of CSNs were abolished by PBMC, a selective blocker of TRPM8. This observation suggests that TRPM8 is a critical component of the cold-sensitive excitatory machinery in both normal and allodynic mice, and that the response of nociceptive neurons recruited in CCI animals most probably relies on this thermo-TRP channel.⁵

The molecular mechanism behind the functional downregulation of the I_KD in response to nerve damage remains to be clarified. The functional decrease in the Kv1.1-1.2-dependent current could be the result of alterations at different levels, including for instance the reduction in the mRNA of these channels affecting protein production, and/or the modulation by post-translational modifications that could alter their biophysical properties, trafficking and destination of the channels to the plasma membrane, among others. Identifying how nerve damage inducing cold allodynia is producing this reduction would be relevant to design novel and more effective therapies to revert pathological cold pain.

The results of our study were also validated using computer simulation of a conductance-based model of CSNs, including a current resembling the I_KD recorded in cold-thermoreceptors and a depolarizing cold- and voltage-dependent TRPM8-like current. In this model, we can systematically vary the I_KD and I_TRPM8 conductance densities, and to determine basal firing rates, temperature response thresholds and maximum firing rates, among others. A response resembling canonical low-threshold CSNs can be obtained with a high g_TRPM8 and a low g_KD density. At intermediate densities of TRPM8 conductance, the model yields high-threshold CSNs, which upon reduction of g_KD drastically reduces its temperature threshold, i.e., shifts the cold threshold to higher temperatures. On the other hand, with a very low g_TRPM8 density and high g_KD density, the model resembles a cold-insensitive neuron that becomes responsive to temperature reductions into the mild-cold range upon reduction of g_KD density, in coincidence with our experimental findings.⁵

In summary, our results suggest that cold allodynia in this model of peripheral nerve injury is linked to a reduction of I_KD in both high-threshold cold thermoreceptors and nociceptors expressing TRPM8, signaling cold discomfort and cold pain, respectively, providing a molecular and neural mechanism for this sensory alteration induced by nerve damage (Fig. 1). How nerve injury reduces Kv1 channels underlying I_KD and its putative contribution to cold hypersensitivity in other peripheral neuropathies remains to be uncovered.

Figure 1.

Thermal sensitivity of primary sensory neurons and damage-gated cold pain. These illustrations depict our proposal of the cold detection mechanisms operating in primary sensory neurons under physiological conditions (left) and after nerve injury (right). Scales on the peripheral endings depict the counterbalance of the functional expression of excitatory (TRPM8, blue) and brake (Kv1.1-1.2, red) channels involved in the fine tuning of cold detection in three functional subtypes of primary sensory neurons. The temperature at which these neurons would be excited is depicted by the colored thermometers at left of each panel. Note that the temperature values are an estimation of the temperature reached at the nerve endings. The size of the schematic channels reflects their relative functional contribution in the different subclasses of primary sensory neurons. Note that the functional expression level of TRPM8 remains unaltered in this model. LT-CSNs: Low threshold cold-sensitive neurons; HT-CSNs: High threshold cold-sensitive neurons; Recruitable Nociceptor: Nociceptive neurons that can be transformed into cold-sensitive neurons by the damage-gated downregulation of I_KD.