K2P channels

K2P Channels — Two-Pore Domain Potassium Channels

Overview

Two-pore domain potassium (K2P) channels are a family of background leak K⁺ channels whose defining characteristic is that their open probability is modulated by multiple converging stimuli — membrane tension, temperature, lipid composition, pH, and volatile anesthetics — making them uniquely positioned as multimodal sensors at the intersection of mechanical, thermal, and chemical signals. In the context of LIFU neuromodulation, K2P channels — specifically the TRAAK, TREK-1, and TREK-2 subfamilies — are the primary candidates for mediating inhibitory and hyperpolarizing responses to ultrasound, in contrast to the excitatory currents generated by Piezo1 and the NICE/SONIC capacitive mechanism. Their dual sensitivity to both membrane tension and temperature makes them the mechanistic overlap zone between the mechanical and thermal hypotheses for FUS effects.

1. Molecular Structure and Biophysical Properties

K2P channels are structurally distinct from all other K⁺ channel families. Each functional channel is a homodimer or heterodimer of two subunits, each containing two pore-forming (P) domains and four transmembrane helices (M1–M4) — hence "two-pore domain." The four P domains together form a selectivity filter with two ion conduction pathways, producing a constitutively active, K⁺-selective conductance that sets the resting membrane potential.

The structural features directly relevant to mechanosensitivity:

• M4 transmembrane helix: the primary mechanosensory element; membrane tension displaces M4 relative to the pore domain, stabilizing the open conformation
• C-terminal intracellular domain: regulates gating in response to temperature, lipids (lysophospholipids, arachidonic acid), and intracellular signaling
• Fenestrations in the pore domain: allow lipid access to the channel interior, making K2P channels exquisitely sensitive to bilayer composition and curvature

The three subfamilies of primary interest:

All three channels are outward rectifying in physiological conditions: at membrane potentials above E_K (₋₉₀ mV), activation produces outward K⁺ current, hyperpolarizing the membrane toward E_K and reducing neuronal excitability.

2. Expression in CNS and Cell-Type Distribution

TRAAK, TREK-1, and TREK-2 are among the most broadly expressed ion channels in the CNS. Key features of their distribution relevant to neuromodulation:

Cortical neurons: all three channels are expressed in pyramidal (RS) neurons and interneurons (FS, LTS), with TREK-1 and TREK-2 showing particularly high expression in layer II/III and layer V pyramidal cells. TRAAK expression is high in neurons with large axonal arbors.

Thalamic neurons: TC and TRN neurons express TREK-1 and TREK-2, contributing to resting membrane conductance and burst/tonic mode transitions. HCN-rich TC neurons may show particularly strong K2P-mediated hyperpolarization responses given their already hyperpolarized resting potential in burst mode.

Inhibitory interneurons: FS interneurons show relatively higher K2P expression than RS pyramidal cells, contributing to their characteristically more hyperpolarized resting potential and rapid repolarization. This differential expression is relevant for understanding why LIFU can preferentially silence excitatory populations through indirect K2P-mediated inhibitory interneuron effects.

Subcellular distribution: TREK-1 and TRAAK show enriched expression at the axon initial segment and nodes of Ranvier — the same sites identified as ultrasound hotspots in MorphoSONIC and flexoelectric models — suggesting that K2P-mediated hyperpolarization may be spatially concentrated at AP initiation zones.

3. Methods of Study

Patch-clamp electrophysiology with controlled membrane stretch

The primary method for characterizing K2P tension-gating. Cell-attached patches with negative pressure applied via the patch pipette impose controlled lateral membrane tension. Sorum et al. (2024) used this approach to generate quantitative tension-gating curves for TRAAK, TREK-1, and TREK-2, producing the Boltzmann parameters essential for computational modeling. Whole-cell recordings with osmotic membrane stretch provide population-level K2P current measurements.

Temperature-controlled electrophysiology

Perfusion chambers with precise temperature control (Peltier elements, ±0.1°C) allow characterization of K2P temperature sensitivity. Q₁₀ values and activation temperature thresholds have been measured for all three channels. A key finding is that TREK-1 and TRAAK activate progressively over the range 25–45°C, with no sharp activation threshold — meaning even small temperature increases within the LIFU range (ΔT = 0.5–2°C) produce measurable K2P conductance increases.

GsMTx4 and specific blockers

GsMTx4 at high concentrations blocks K2P channels in addition to Piezo channels, complicating attribution when used alone. More specific approaches use:

• Ruthenium red: blocks TRAAK selectively at moderate concentrations
• Spadin: selective TREK-1 blocker (C-terminal peptide inhibitor)
• Genetic KO: TRAAK KO, TREK-1 KO, TREK-2 KO mice have been generated and used to dissect individual channel contributions

Lipid bilayer reconstitution

Purified K2P proteins reconstituted into artificial lipid bilayers allow precise control of membrane tension and composition, isolating the intrinsic mechanosensitivity from cellular regulatory factors. Sorum et al. (2024) combined reconstitution with patch-clamp to obtain clean Boltzmann gating parameters uncontaminated by intracellular signaling.

In vivo calcium imaging and electrophysiology under LIFU

Collins & Mesce (2022) combined extracellular multi-unit recordings with pharmacological K2P manipulation in invertebrate preparations, providing in vivo evidence for K2P involvement in LIFU-induced inhibition. While invertebrate preparations limit direct CNS translation, the study established the principle that K2P activation contributes to LIFU-evoked silencing at FDA-compliant intensities.

4. Key Studies and Results

Sorum et al. (2024) — Quantitative Boltzmann gating characterization

Sorum et al. (2024) is the most directly modeling-relevant study for K2P channels, providing quantitative tension-gating parameters for TRAAK, TREK-1, and TREK-2 using a combination of lipid bilayer reconstitution and cell-attached patch-clamp:

Boltzmann tension-gating parameters extracted:

Key findings:

• All three channels activate at membrane tensions well within the range predicted by SONIC bubble dynamics at therapeutic LIFU pressures (0.1–0.5 MPa generating T ~ 1–5 mN/m via \($T = K_A \cdot \Delta A_m / A_m\))
• The Boltzmann curves are shallow (low \(\Delta A\) relative to Piezo1's ~20 nm²), meaning K2P channels show graded, gradual activation over a wide tension range rather than the switch-like activation of Piezo1 — producing a tonic hyperpolarizing conductance that increases progressively with acoustic amplitude
• Temperature increases of 1–2°C (within LIFU thermal range) shift the Boltzmann curve leftward by ~0.5–1 mN/m, meaning less tension is required for the same open probability — demonstrating the mechanical-thermal synergy for K2P activation

Yoo et al. (2022) — K2P contribution to LIFU-evoked inhibition in cortical neurons

Yoo et al. (2022) investigated the ionic basis of LIFU-evoked inhibitory responses in cultured cortical neurons and acute cortical slices. Using a combination of whole-cell patch-clamp, pharmacological blockade, and computational modeling:

• Demonstrated that LIFU at moderate intensities (200–400 kPa, 500 kHz) evoked a sustained outward K⁺ current in a subset of cortical neurons, reversing near E_K — consistent with K2P activation
• This outward current was reduced ~50% by spadin (TREK-1 blocker) and ~30% by ruthenium red (TRAAK blocker), implicating both channels
• The K2P-mediated outward current was enhanced by mild heating (ΔT = 1°C applied concurrently with LIFU), confirming thermal-mechanical synergy in K2P activation
• Computational modeling incorporating TREK-1 and TRAAK conductances into an RS neuron HH model reproduced the observed inhibitory responses and predicted that K2P activation threshold is reached at lower intensities than NICE-predicted AP threshold — suggesting a low-intensity inhibitory regime dominated by K2P before the high-intensity excitatory regime dominated by capacitive currents

Collins & Mesce (2022) — Thermal primacy and K2P involvement in non-sensory neurons

Collins & Mesce (2022) provided important context for K2P involvement by arguing that in non-sensory neurons at FDA-compliant LIFU intensities, thermal mechanisms — primarily K2P upregulation — are the dominant modulatory pathway rather than mechanical cavitation or MS channel direct gating. Key arguments:

• Temperature rises of 0.5–1.5°C produced by LIFU are sufficient to increase TREK-1/TRAAK open probability by 20–40% based on published Q₁₀ values
• This thermally driven K2P upregulation produces net hyperpolarization and reduced firing — accounting for silencing without requiring cavitation
• In their invertebrate preparation, LIFU effects were mimicked by equivalent temperature increases and abolished by K2P blockers, but not by GsMTx4 at concentrations selective for Piezo channels
• This thermal-K2P pathway may dominate in cells with low mechanosensitivity (non-sensory CNS neurons) while direct tension-gating of K2P dominates in neurons with high acoustic coupling efficiency

Cheng et al. (2026) — K2P heterodimer dynamics under LIFU

Cheng et al. (2026) investigated TREK-1/TREK-2 heterodimer formation and its consequences for ultrasound sensitivity. Key findings:

• TREK-1/TREK-2 heterodimers show intermediate tension sensitivity between the two homodimers, with a T_{1/2} shifted toward lower tensions (₂.4 mN/m) compared to either homodimer alone
• Heterodimer formation is activity-dependent: sustained neuronal activity promotes TREK-1/TREK-2 co-assembly through a trafficking-dependent mechanism
• Under LIFU, heterodimers show reduced inactivation compared to TREK-2 homodimers, producing more sustained hyperpolarizing currents during prolonged sonication
• This finding suggests that the effective K2P response to LIFU in active neural tissue differs from predictions based on homodimer characterization alone — an important consideration for modeling neurons during ongoing network activity

5. Boltzmann Tension-Gating Model for K2P Channels

For integration into the unified HH framework, K2P channels are modeled as instantaneously tension-gated background conductances with no explicit gating particle kinetics (unlike Piezo1 which has distinct activation and inactivation gates):

Tension-dependent open probability:

\[P_{open,K2P}(T, \Theta) = P_{open,max} \cdot \frac{1}{1 + \exp\left(-\frac{\Delta A (T - T_{1/2}(\Theta))}{k_B \Theta}\right)}\]

Temperature shift of half-activation tension:

\[T_{1/2}(\Theta) = T_{1/2,0} - \alpha_T (\Theta - \Theta_{ref})\]

Where $\alpha_T \approx 0.5$ mN/m/°C (estimated from Sorum et al. 2024 and Yoo et al. 2022 data) and $\Theta_{ref} = 310$ K (37°C).

K2P current for channel $c$ (TRAAK, TREK-1, TREK-2):

\[I_{K2P,c} = \bar{g}_{K2P,c} \cdot P_{open,c}(T, \Theta) \cdot (V_m - E_K)\]

Total K2P current:

\[I_{K2P} = \sum_{c \in \{\text{TRAAK, TREK-1, TREK-2}\}} I_{K2P,c}\]

The membrane tension $T$ is linked to SONIC output:

\[T = K_A \cdot \frac{\Delta A_m}{A_m}\]

And temperature $\Theta$ is provided by the thermal module with $T_{ref} = 37°C$.

6. Parameter Regimes for K2P Relevance

7. Interaction with Other Mechanisms

K2P and Piezo1 — sequential excitation-inhibition: Piezo1-mediated Ca²⁺ influx elevates $[\text{Ca}^{2+}]_i$, which activates Ca²⁺-sensitive K⁺ channels (SK, BK) and may enhance K2P open probability through lipid-mediated mechanisms, creating a delayed inhibitory phase following initial Piezo1-driven excitation.

K2P and thermal effects — synergistic inhibition: As demonstrated by Yoo et al. (2022) and Collins & Mesce (2022), mild LIFU-induced heating directly upregulates K2P open probability through the $T_{1/2}(\Theta)$ shift. Mechanical tension and thermal drive are additive in their effect on K2P gating, meaning that neither alone may be sufficient to produce significant K2P activation but their combination readily crosses threshold.

K2P and flexoelectricity — inhibitory convergence: both K2P activation and the Ca²⁺-dependent flexoelectric sign reversal produce hyperpolarizing currents through independent pathways, suggesting that at appropriate parameter combinations, inhibitory responses may arise from convergent multi-mechanism contributions rather than a single dominant pathway.

K2P and network dynamics: differential K2P expression across cell types means that at intermediate LIFU intensities, inhibitory interneurons (FS) may be more strongly hyperpolarized than excitatory neurons (RS) — a disinhibitory effect that paradoxically increases RS neuron firing when FS interneurons are silenced. This network-level inversion of the expected K2P inhibitory effect is a subtle but important consideration for predicting population-level LIFU outcomes.

8. Strengths and Limitations

Strengths

• Multimodal sensitivity: K2P channels integrate mechanical, thermal, and chemical signals simultaneously — making them uniquely relevant for modeling the full range of LIFU parameter effects
• Quantitative gating parameters available: Sorum et al. (2024) Boltzmann parameters enable direct HH integration without free parameter fitting
• Natural account of inhibition: K2P activation produces outward K⁺ current hyperpolarizing toward E_K — a clean single-neuron inhibitory mechanism complementing SECONIC's network-level account
• Thermal-mechanical synergy: K2P channels provide a principled mechanistic basis for why combined thermal and mechanical LIFU effects differ from either alone
• CNS-relevant expression: broad cortical and thalamic expression with cell-type differential — directly relevant to RS, FS, LTS, TC, TRN, STN modeling
• Empirical support from multiple approaches: patch-clamp, pharmacological blockade, in vivo recordings, and computational modeling all converge on K2P involvement

Limitations

• Cell-type-specific conductance densities unknown: absolute \(\bar{g}_{K2P,c}\) values for each channel in each CNS neuron subtype are not precisely quantified — a significant free parameter in the unified model
• Heterodimer dynamics add complexity: Cheng et al. (2026) heterodimer findings suggest that homodimer-based models underestimate K2P response variability in active tissue — but heterodimer gating parameters are not yet fully characterized
• Tension generation remains uncertain: whether LIFU generates sufficient membrane tension for K2P activation via radiation pressure alone (without cavitation) is unresolved — the same uncertainty that affects Piezo1 modeling
• Temperature coefficient \(\alpha_T\) estimated: the thermal shift parameter for \(T_{1/2}(\Theta)\) is estimated from indirect evidence rather than direct CNS-specific measurement
• Disinhibitory network effects difficult to capture at single-neuron level: the paradoxical excitatory population response from FS interneuron silencing requires network-level modeling (SECONIC-type) to reproduce
• GsMTx4 non-specificity: pharmacological studies using GsMTx4 cannot cleanly separate K2P from Piezo contributions, complicating empirical attribution