TASK-3 and TASK-2 are two-pore domain, background K⁺ channels gated by extracellular pH. TASK-3 dependence upon pH_o is cooperative with K_d of 5.9 ± 0.05 and n_H of 2.1 ± 0.09 (n = 8; all errors quoted as SEM). The pH_o-sensor in TASK-3 is histidine 98 located at the extracellular entrance of the pore and TASK-3-H98N mutant is pH_o-independent (Rajan et al., 2000; Kim et al., 2000). The pH_o-gating of TASK-2 occurs with a pK_1/2 of 8.0, is not cooperative and is mediated by neutralization of arginine 224, so that mutant TASK-2-R224A is pH_o-independent (Niemeyer et al., 2007). To find out whether both pH_o-sensors occurring in these dimeric channels are required in the gating process we have covalently linked the C-terminus of one channel with the N-terminus of the following to form concatenated structures containing either a normal set of pH_o-sensors (WT-WT), mixed structures containing one able and one neutralised pH_o-sensor (WT-Mut, Mut-WT) or two pH_o-sensing disabled channels (Mut-Mut). Analysis was done by transient expression into HEK-293 cells and patch-clamp. Gating by pH_o of TASK-3 WT-WT concatenated channels (n=8) or Mut-Mut (Mut = TASK-3-H98N, n=4) structures was similar to that of their non-concatenated equivalent channels. In contrast, the mixed WT-Mut and Mut-WT followed pH_o in a non-cooperative manner and required more acidification for inhibition to occur. The respective pK_1/2 values for TASK-3 WT-Mut and Mut-WT were 4.9 ± 0.04 (n=5) and 5.1 ± 0.12 (n=6). Assuming that N98 mimics a neutral form of H98, we have used the K_1/2 of these mixed constructs (10^-5 M) as K_d2 in a fit to the data of a simplified model for the gating of TASK-3 by protons in which there are two closed states with two or one charged H98 residues and an open state with both H98s neutralised. This analysis applied to the WT-WT data gave a value for K_d1 = 3.26*10^-7 M consistent with the idea that neutralisation of a first H98 sensor of TASK-3 affects the pK_a of the second making it 30-times more susceptible H⁺ loss. WT-WT TASK-2 concatenated constructs behaved similarly to the non-concatenated channels with pK_1/2 8.2 ± 0.15 (n=4), whilst Mut-Mut (Mut = TASK-2-R224A) TASK-2 constructs lacked pHo-dependence. The mixed WT-Mut and Mut-WT TASK-2 constructs had pH_o-dependences (respective pK_1/2 values of 7.8 ± 0.05, n=6 and 7.9 ± 0.10, n=6) very similar to that of WT and WT-WT channels. Our data show that both pHo-sensors of TASK-3 and TASK-2 channels need to be neutralised for channel opening. There appears to be no interaction between sensors in TASK-2 channels accounting for the lack of cooperativity in the pH_o effect. In TASK-3 however, the data indicate that neutralisation of one sensor profoundly affects the ability of the second sensor to become neutral thus accounting for cooperative, high-gain response to pH_o.