The human tandem-P domain K⁺ channel, hTREK1, is O₂ sensitive and hypoxia occludes its modulation by arachidonic acid or membrane deformation (Miller et al. 2003). These findings question the neuroprotective role of hTREK1 during central ischaemia proposed by Honore et al. (2002). Here we report the effects of altered intracellular pH (pH_i) on the ability of hypoxia to inhibit this channel.

All experiments employed whole-cell patch clamp recordings from HEK293 cells stably transfected with hTREK1, and data reported are taken from currents recorded at a step potential of +60 mV from a holding potential of -70 mV (see Miller et al. 2003).

Current density measured 20 s following the transition to whole-cell configuration was not dependent upon pH_i. Thus, at pH_i 6.5 (using 11 mM Pipes to buffer pH_i; n = 6) current density was 282 ± 115 pA pF^-1 (mean ± S.E.M.) whilst at pH_i 7.2 (11 mM Hepes; n = 10) and pH_i 7.9 (11 mM Tris; n = 5), current densities were 173 ± 53 pA pF^-1 and 157 ± 46 pA pF^-1, respectively. Subsequently, mean current density of each cohort steadily increased in a pH_i-dependent manner to 654 ± 198 pA pF^-1 (pH 6.5), 410 ± 80 pA pF^-1 (pH 7.2) and 263 ± 46 pA pF^-1 (pH 7.9) after 2-4 min. Thus, hTREK1 whole-cell current amplitude was sensitive to pH_i. Furthermore, acute hypoxia (~20 mmHg) evoked current depression at all three pH_i values with the hypoxia-sensitive current being largest at low pH_i.

In order to test the hypothesis that pH_i and hypoxia are interacting modulatory influences on this important neuronal channel, we investigated the effect of dynamically altering pH_i using extracellular addition of 20 mM sodium propionate. This evoked rapid and significant (Student’s paired t test) channel activation from 228 ± 32 pA pF^-1 to 299 ± 33 pA pF^-1 (P < 0.005, n = 9). In the continued presence of propionate, hypoxia reduced this to 168 ± 18 pA pF^-1 (P < 0.0005). This 42 ± 4 % inhibition was significantly larger than the 26 ± 3 % reduction seen in the absence of weak acid (n = 13, P < 0.002, Student’s unpaired t test). Importantly, the absolute magnitude of the currents in hypoxia alone was not different from that following co-application of weak acid with hypoxia; i.e. acid activation is occluded by hypoxia.

Together with our previous data (Miller et al. 2003), these new observations show that many of the stimuli associated with ischaemia/hypoxia (such as arachidonic acid release, cell swelling, and acid insult) cannot activate hTREK1 when O₂ availability is limited and reinforce our earlier suggestion that the proposed neuroprotective role of this channel during ischaemia/hypoxia is unlikely to occur at PJ{special} values typical in the brain either in health or during pathology.

This work was funded by The British Heart Foundation, The Wellcome Trust and GlaxoSmithKline