Kir6.2-D323 and SUR2A-Q1336: an intersubunit interaction pairing for allosteric information transfer in the KATP channel complex

Abstract

ATP-sensitive potassium (KATP) channels are widely expressed and play key roles in many tissues by coupling metabolic state to membrane excitability. The SUR subunits confer drug and enhanced nucleotide sensitivity to the pore-forming Kir6 subunit, and so information transfer between the subunits must occur. In our previous study, we identified an electrostatic interaction between Kir6 and SUR2 subunits that was key for allosteric information transfer between the regulatory and pore-forming subunit. In this study, we demonstrate a second putative interaction between Kir6.2-D323 and SUR2A-Q1336 using patch clamp electrophysiological recording, where charge swap mutation of the residues on either side of the potential interaction compromise normal channel function. The Kir6.2-D323K mutation gave rise to a constitutively active, glibenclamide and ATP-insensitive KATP complex, further confirming the importance of information transfer between the Kir6 and SUR2 subunits. Sensitivity to modulators was restored when Kir6.2-D323K was co-expressed with a reciprocal charge swap mutant, SUR-Q1336E. Importantly, equivalent interactions have been identified in both Kir6.1 and Kir6.2 suggesting this is a second important interaction between Kir6 and the proximal C terminus of SUR.

Keywords: ABC transport proteins; KATP; molecular interactions; potassium channels.

Figure 1.. Putative salt bridges between the proximal C terminus of SUR2A and the C terminus of Kir6.1.

(A) Cartoon showing the residues thought to be key in transferring gating information between SUR2A and Kir6.2 subunits. (B) Schematic showing the interaction between SUR2A-E1318 and Kir6.2-K338 identified in our previous study [14] and the putative interaction between Kir6.2-D323 and SUR2A-K1322 or Q1336.

Figure 2.. Kir6.2-D323K mutation yields a constitutively active Kir6.2 subunit that is insensitive to sulphonylureas and ATP.

(A) Cartoon showing the Kir6.2-D323K mutation, disrupting the putative salt bridge to SUR2A. (B) Whole cell current recording held at 0 mV, from HEK293 cells transiently transfected with Kir6.2-D323K/SUR2A-WT, showing constitutive activity that runs down over time (i) and that is insensitive to 100 µM glibenclamide. (C) Mean data from HEK293 cells transiently transfected with Kir6.2-D323K/SUR2A-WT showing a substantial K⁺ current in control conditions compared with K_ATP-WT (Kir6.2/SUR2A) (*** P < 0.0001, Two-Way ANOVA with Holm-Sidak's post test), where there was no difference between glibenclamide treatment and run down of the current (n = >5 for each group). (D) (i) membrane potential (Vm) recordings in K_ATP-WT and Kir6.2-D323K/SUR2A-WT transiently transfected HEK293 cells. (ii) membrane potential recordings in both cell types showing pinacidil-induced hyperpolarisation and glibenclamide-induced depolarisation in K_ATP-WT transfected HEK293 cells. There was no effect on the membrane potential of Kir6.2-D323K/SUR2A-WT transfected cells. (E) Mean data showing the membrane potential in control conditions and in 100 µM glibenclamide in K_ATP-WT and Kir6.2-D323K/SUR2A-WT transfected cells. There was no significant difference in the effect of glibenclamide on the membrane potential of K_ATP-WT or Kir6.2-D323K/SUR2A-WT, however the Kir6.2-D323K mutant caused a significantly more hyperpolarised membrane potential in control and glibenclamide conditions (*** P < 0.0001, Two-Way ANOVA with Holm-Sidak's post test, n > 5 for each group). (F) ATP concentration response data recorded from excised inside-out patches. K_ATP-WT patches had an IC₅₀ for ATP of 23.8 ± 1.2 µM, but there was no effect of ATP in the Kir6.2-D323K/SUR2A-WT mutant (n = 6 for each subunit combination).

Figure 3.. Kir6.2-WT co-expressed with SUR2A-K1322D forms a channel that functionally expresses little current.

(A) Cartoon showing the SUR2A-K1322D mutation, disrupting the putative salt bridge to Kir6.2. (B) Mean whole cell current recording held at 0 mV, from HEK293 cells transiently transfected with Kir6.2-WT/SUR2A-K1322D, showing some sensitivity to pinacidil, metabolic inhibition (with cyanide and iodoacetic acid) and to 10 µM glibenclamide (n = 9). The current produced was too small to record accurate pinacidil, glibenclamide or ATP sensitivity. (C) Mean membrane potential for both K_ATP-WT and Kir6.2-WT/SUR2A-K1322D was not significantly different (t-test, n > 12 for each group), indicative of negligible hyperpolarizing current.

Figure 4.. Kir6.2-WT co-expressed with SUR2A-Q1336E forms a channel that functionally expresses but has reduced sensitivity to glibenclamide and ATP.

(A) Cartoon showing the SUR2A-Q1336E mutation, disrupting the putative salt bridge to Kir6.2. (B) Mean whole-cell current recording held at 0 mV, from HEK293 cells transiently transfected with Kir6.2-WT/SUR2A-Q1336E, showing sensitivity to pinacidil, metabolic inhibition (with cyanide and iodoacetic acid) and to 10 µM glibenclamide (n = 8). The current produced was small but was robust enough to record glibenclamide and ATP sensitivity. (C) Mean membrane potential for both K_ATP-WT and Kir6.2-WT/SUR2A-Q1336E was not significantly different (t-test, n > 12 for each group). (D) Glibenclamide concentration-inhibition data recorded from WT-K_ATP and Kir6.2-WT/SUR2A-Q1336E expressed in HEK293 cells. Inhibition of the Kir6.2-WT/SUR2A-Q1336E complex by glibenclamide was significantly right shifted (IC₅₀ of 4.3 nM to 433 nM, P < 0.002, t-test n = 6 for each group). (E) ATP-sensitivity was also right shifted from an IC₅₀ in control excised inside-out patches of 23.4 µM to 90 µM in the Kir6.2-WT/SUR2A-Q1336E channel complex (P < 0.0001, t-test, n = 6 for each group).

Figure 5.. Co-expression of Kir6.2-D323K with SUR2A-K1322D to try to restore the salt bridge via a charge swop forms a channel that is constitutively active and does not respond to sulphonylureas or ATP.

(A) Cartoon showing the Kir6.2-D323K and SUR2A-K1322D mutations potentially restoring the salt bridge interaction. (B) Mean membrane potential recording from HEK293 cells transiently transfected with Kir6.2-D323K/SUR2A-K1322D showing a hyperpolarised membrane potential comparable with the Kir6.2-D323K/SUR2A-WT (Vm data for Kir6.2-WT/SUR2A-WT and Kir6.2-D323K/SUR2A-WT from Figure 2 shown for comparison). (C) Mean whole-cell recording data, recorded at 0 mV, showing constitutive activity in basal conditions and no significant response to pinacidil, metabolic inhibition or glibenclamide (Repeated measured ANOVA, n = 7). WT-K_ATP activation with pinacidil shown for comparison. (D) ATP concentration inhibition data recorded in excised inside-out patches from WT-K_ATP and Kir6.2-D323K/SUR2A-K1322D showing that ATP had no inhibitory effect on the double mutant channel complex.

Figure 6.. Co-expression of Kir6.2-D323K with SUR2A-Q1336E to try to restore the salt bridge via a charge swap forms a channel that shows some constitutive activity but is sensitive to sulphonylureas and ATP.

(A) Cartoon showing the Kir6.2-D323K and SUR2A-Q1336E mutations potentially restoring the salt bridge interaction. (B) Mean membrane potential recording from HEK293 cells transiently transfected with Kir6.2-D323K/SUR2A-Q1336E showing a partially hyperpolarised membrane potential compared with WT-K_ATP, however not as hyperpolarised as the Kir6.2-D323K/SUR2A-WT subunit combination (Vm data for Kir6.2-WT/SUR2A-WT and Kir6.2-D323K/SUR2A-WT from Figure 2 shown for comparison) (n = 8). (C) Mean whole-cell recording data, recorded at 0 mV, showing some constitutive activity in basal conditions that was enhanced by pinacidil and metabolic inhibition and fully reversed by 10 µM glibenclamide (Repeated measured ANOVA, n = 7). WT-K_ATP activation with pinacidil shown for comparison. (D) Glibenclamide concentration-inhibition data showing that the co-expression of Kir6.2-D323K/SUR2A-Q1336E fully restored glibenclamide sensitivity to the channel complex (IC₅₀ of 3.9 nM and 3.7 nM for WT-K_ATP and Kir6.2-D323K/SUR2A-Q1336E, respectively, t-test, P = 1, n = 8). (E) ATP concentration inhibition data from WT-K_ATP and Kir6.2-D323K/SUR2A-Q1336E showing that the charge swap restored some ATP sensitivity, although this was right shifted from the WT-K_ATP (IC₅₀ 23 µM and 123 µM, n = 6).

Figure 7.. Co-expression of Kir6.2-D323K with a double mutant SUR2A-K1322D/Q1336E forms a channel that is constitutively active and insensitive to sulphonylurea drugs and ATP.

(A) Cartoon showing the Kir6.2-D323K and SUR2A-K1322D/Q1336E mutations potentially restoring the salt bridge interaction. (B) Example whole-cell recordings at 0 mV from HEK293 cells transiently expressing Kir6.2-WT/SUR2A-K1322D/Q1336E showing a very small pinacidil and glibenclamide sensitive current (i) or Kir6.2-D323K/SUR2A-K1322D/Q1336E (ii) which showed a large, sulphonylurea insensitive and constitutively active current. (C) Mean whole-cell recording data, recorded at 0 mV, showing constitutive activity conditions that was unaffected by pinacidil, metabolic inhibition or glibenclamide (Repeated measures ANOVA, n = 13). (D) Mean membrane potential from both groups, showing that expression of the constitutively active Kir6.2-D323K/SUR2A-K1322D/Q1336E complex causes a significantly hyperpolarised membrane potential compared with expression of SUR2A-K1322D/Q1336E mutant with the Kir6.2-WT (*** P < 0.0001, t-test, n = 10). (E) ATP concentration inhibition data from WT-K_ATP and Kir6.2-D323K/SUR2A-K1322D/Q1336E showing that the mutated complex is not ATP-sensitive (n = 4).

Figure 8.. A salt bridge between the equivalent Kir6.1 residue (E332) and SUR2A exists with K1322 rather than Q1336.

(A) cartoon showing the putative interaction with Kir6.1-E332 and SUR2A-K1322. (B) Example trace of whole-cell Kir6.1/SUR2A current activated by pinacidil. (C) Whole-cell trace showing constitutively active Kir6.1 current, enhanced with pinacidil and inhibited with 10 µM glibenclamide. (D) Concentration inhibition curve for Kir6.1-WT/SUR2A-WT and Kir6.1-E332K/SUR2A-WT showing a rightward shift in with the mutated Kir6.1 pore (n = >6 per data point, IC₅₀ values of 6.13 ± 1.2 to 318 ± 11 nM in WT and mutants, respectively). Example traces showing Kir6.1-E332K co-expressed with the charge swap SUR2A mutants SUR2A-K1322D (E), showing constitutive activity and responsiveness to glibenclamide, and SUR2A-Q1336E (F) showing constitutive activity that was unresponsive to 100 µM glibenclamide. (G) concentration response curve for glibenclamide showing no difference between Kir6.1-WT/SUR2A-WT and Kir6.1-E332K/SUR2A-K1322D channels, (n = >6 per data point, IC₅₀ values of 6.1 ± 1.2 to 9.2 ± 3.1 nM in WT and mutant, respectively.

Figure 9.. Truncation of the Kir6.2-D323K mutation to remove the SR-retention sequence yields a channel that is not constitutively active and ATP sensitive.

(A) Cartoon showing the truncation of Kir6.2. (B) Mean whole-cell data recorded at 0 mV from Kir6.2ΔC26 and the Kir6.2ΔC26-D323K mutant showing neither channel is constitutively active, neither responds to sulphonylurea drugs, but both were activated by ATP depletion with metabolic inhibition (cyanide and iodoacetic acid) (n = ≥6 for each group). (C) Mean membrane potential recordings showing no difference in resting membrane potential between the WT and mutated truncation (n = ≥10 for each group). (D) The ATP sensitivity of the Kir6.2ΔC26 truncated channel is shifted 10-fold compared with Kir6.2/SUR2A (IC₅₀ 123 µM compared with 23 µM in Kir6.2ΔC26 compared with Kir6.2/SUR2A, respectively) however the Kir6.2ΔC26-D323K mutant showed no difference in its ATP sensitivity compared with the Kir6.2ΔC26 mutant expressed alone.

Figure 10.. Kir6.2ΔC26-D323K co-expression with SUR2A-Q1336E restores sulphonylurea sensitivity however the channel remains partially constitutively active.

(A) Cartoon showing the truncation of Kir6.2 and the putative interaction with SUR2A between residues D323 of Kir6.2 and Q1336 of SUR2A. (B) Example traces from (i) Kir6.2ΔC26 and (ii) Kir6.2ΔC26-D323K co-expressed with SUR2A-WT. Kir6.2ΔC26-D323K/SUR2A-WT co-expression forms a constitutively active current, whereas the Kir6.2ΔC26/SUR2A-WT combination requires pinacidil to activate the current. Both combinations are inhibited by glibenclamide. (C) Mean membrane potential data from Kir6.2ΔC26 /SUR2A-WT, Kir6.2ΔC26-D323K/SUR2A-WT and Kir6.2ΔC26-D323K/SUR2A-Q1336E, suggesting some constitutive activity with a slightly hyperpolarised membrane potential (*** P < 0.0001, * P < 0.05, one-way ANOVA with Holm-Sidak's post-test, n = >8). (D) Mean current recording from Kir6.2ΔC26/SUR2A-WT, Kir6.2ΔC26-D323K/SUR2A-WT and Kir6.2ΔC26-D323K/SUR2A-Q1336E combinations showing some constitutive activity in the charge swap pairing (n = >8). Kir6.2ΔC26-D323K/SUR2A-WT was constitutively active and not further enhanced by pinacidil. (E) Concentration-response data for glibenclamide showing that co-expression of Kir6.2ΔC26-D323K/SUR2A-Q1336E charge swap mutations had identical inhibition profile to Kir6.2-WT/SUR2A-WT co-expression (n = >6 for each data point, IC₅₀ of 3.8 ± 1.1 and 3.6 ± 0.6 in WT-truncation mutant and double mutant, respectively). (F) ATP sensitivity of the Kir6.2ΔC26-D323K/SUR2A-Q1336E charge swap mutants showing identical inhibition by ATP to the Kir6.2ΔC26 truncation mutation (n = 6 for each data set, IC₅₀ of 113 ± 5 and 114 ± 5 µM for WT-truncation and double mutant, respectively), but not Kir6.2-WT/SUR2A-WT (23.8 ± 1.2 µM, shown in Figure 2).

Figure 11.. Cartoon representation of the interactions between Kir6.2 and Kir6.1 and the SUR2A accessory subunit and the position of the identified residues.

Cartoon representation of the putative interaction between Kir6.2-D323 and SUR2A-Q1336 (A) and Kir6.1-E322 and SUR2A-K1322 (B). Images C–G show the location of Kir6.2-D323 (blue), SUR-K1322 (yellow) and SUR-Q1336 (green) on published K_ATP complex structures showing a core tetramer of Kir6.2 surrounded by four SUR1 Subunits. Structures from Protein Data Bank, (C) PDB code:6C3P [32], (D) PDB code:6C3O [32], (E) PDB code:5TWV [31], (F) PDB code:6BAA [30], (G) PDB code:5WUA [29]. (H) Five model overlay of C–G showing a single SUR2A and Kir6.2 to demonstrate movement of key residues within published structures. * indicates residues that do not move in models 5TWV, 6BAA and 5WUA.

Figure 12.. Kir6.2-D323 is in close proximity to a highly-conserved Lasso region within SUR subunits (residues 193–261) known to be involved in regulation of gating.

(A) Ribbon representation of SUR1-Kir6.2 channel complex in propeller conformation (PDB code: 6C3P [32]) modelled using Pymol software. Red colour indicates the Lasso region (residues 193–261). (B) Expansion of SUR1-Kir6.2 channel complex in ribbon representation, illustrating the distance between D323 and R248 on the lasso region. In this structure; Kir6.2 is in purple, SUR1 is in cream and the lasso region is shown in red. (C) Expansion of SUR1-Kir6.2 channel complex in ribbon representation, illustrating the effect of D323K mutation on the distance between residue 323 and R248 (the lasso region). D323K mutation brings K323 within 3 Å of R248 on the Lasso region. As with part B, in this structure Kir6.2 is in purple, SUR1 is in cream and the lasso region is shown in red.