A mutually induced conformational fit underlies Ca2+-directed interactions between calmodulin and the proximal C terminus of KCNQ4 K+ channels

Abstract

Calmodulin (CaM) conveys intracellular Ca²⁺ signals to KCNQ (Kv7, "M-type") K⁺ channels and many other ion channels. Whether this "calmodulation" involves a dramatic structural rearrangement or only slight perturbations of the CaM/KCNQ complex is as yet unclear. A consensus structural model of conformational shifts occurring between low nanomolar and physiologically high intracellular [Ca²⁺] is still under debate. Here, we used various techniques of biophysical chemical analyses to investigate the interactions between CaM and synthetic peptides corresponding to the A and B domains of the KCNQ4 subtype. We found that in the absence of CaM, the peptides are disordered, whereas Ca²⁺/CaM imposed helical structure on both KCNQ A and B domains. Isothermal titration calorimetry revealed that Ca²⁺/CaM has higher affinity for the B domain than for the A domain of KCNQ2-4 and much higher affinity for the B domain when prebound with the A domain. X-ray crystallography confirmed that these discrete peptides spontaneously form a complex with Ca²⁺/CaM, similar to previous reports of CaM binding KCNQ-AB domains that are linked together. Microscale thermophoresis and heteronuclear single-quantum coherence NMR spectroscopy indicated the C-lobe of Ca²⁺-free CaM to interact with the KCNQ4 B domain (K_d ∼10-20 μm), with increasing Ca²⁺ molar ratios shifting the CaM-B domain interactions via only the CaM C-lobe to also include the N-lobe. Our findings suggest that in response to increased Ca²⁺, CaM undergoes lobe switching that imposes a dramatic mutually induced conformational fit to both the proximal C terminus of KCNQ4 channels and CaM, likely underlying Ca²⁺-dependent regulation of KCNQ gating.

Keywords: M current; X-ray crystallography; biophysical chemical analysis; calcium signaling; calcium-binding protein; calmodulin (CaM); conformational change; isothermal titration calorimetry (ITC); microscale thermophoresis (MST); nuclear magnetic resonance (NMR); potassium channel; potassium voltage-gated channel subfamily Q member (KCNQ).

Figure 1.

KCNQ1–5 contain conserved A and B domains, which are disordered in the absence of CaM. A, representative schematic of a KCNQ protein subunit depicting the S1–S6 transmembrane helices, the intracellular N and C termini, and the A domain (teal) and B domain (gold) within the proximal half of the C terminus, referred to as the RD. The schematic does not show the likely intimate proximity of the RD to the plasma membrane. For reference, the PIP₂ interaction sites are shown as gray circles, and the protein kinase C phosphorylation site conserved between KCNQ2–5 is represented by a black circle. B, sequence alignments of the A domain (left) and B domain (right) of human KCNQ1–5 subunits taken from Uniprot. All sequences represent “isoform 1” of each subunit, with the exception of the KCNQ2 sequences that represent the universally used isoform 4 and those of KCNQ4a. Alignments were performed using the PRALINE alignment tool, and the colors were adjusted using Photoshop Elements. The dots above the alignments indicate the interactions with Ca²⁺/CaM shown in Fig. S3 and Fig. 2. Black dots indicate the residues with strong interactions between the KCNQ4 peptides and CaM that are different from those of KCNQ1, and gray dots indicate strong interactions for which the interacting residues are conserved. Open circles indicate those KCNQ4 residues having nonbonded contacts with CaM, and the star indicates the residues of KCNQ1 that may cause changes of the CaM backbone compared with KCNQ4. C, CD spectra of the peptides used in this study and of CaM. Two dashed vertical lines at 208 and 222 nm indicate the points of deflection typical of proteins such as CaM (black) with high helical content. All of the Q2-4 A and B peptides appear to lack helical content in the absence of CaM.

Figure 2.

The X-ray crystal structure of the Ca²⁺/CaM:Q4A:Q4B complex involves antiparallel A and B helices enveloped by CaM with Ca²⁺ ions in the N-lobe. A, front view (left) and 90° side view (right) of the trimeric co-crystal X-ray structure of Ca²⁺/CaM with the Q4A and Q4B peptides. CaM is shown in pink, with the C-lobe facing the bottom, lacking Ca²⁺ ions, and the N lobe on top, bound by two Ca²⁺ ions, colored dark gray. Q4B (gold) and Q4A (teal) are embraced together by CaM. B, cartoon schematic depicting the crystal structure representing the overall conformation of Ca²⁺/CaM bound to a full KCNQ4 subunit. C and D, the expanded views of the CaM:Q4A:Q4B structure show the interior of CaM (gray) interacting with the side chains of the Q4A and Q4B peptides. The peptide residues are colored to match the conserved color plot shown in Fig. 1. These interactions are clarified in the plot in Fig. S3, and dashed circles in the inset highlight the loci of the interactions. CaM residues are labeled in maroon, Q4B is labeled in black, and Q4A is labeled in teal. E, the backbone C-α alignment of the Ca²⁺/CaM:Q4A:Q4B complex with the Ca²⁺/CaM:KCNQ1AB complex (light gray) from PDB entry 4UMO, in which one of the two asymmetric units of the domain-swapped pair was truncated for clarity. F, the expanded view of the overlaid structures shows the difference in position of the CaM linker as it interacts with Ile⁵³⁹ (pink CaM) or Arg⁵¹⁹ (gray CaM). These structures were rendered using PyMOL.

Figure 3.

The B domain of KCNQ2–4 has a very high affinity for Ca²⁺/CaM, whereas that of the A domain is modest. Isotherms are shown for the peptides (50–100 μm) titrated into 5 mm CaM in the presence of 5 μm Ca²⁺. The A domain peptides are represented in the top row, and the B domain peptides are shown in the bottom row, representing KCNQ2, KCNQ3, and KCNQ4 isoforms. Analysis was performed using the one-site binding model in MicroCal Origin version 7.

Figure 4.

ITC reveals that the A domain must bind CaM first to form a stable Ca²⁺/CaM:Q4A:Q4B trimeric complex. A, the shown isotherm demonstrates no detectable binding between the A and B domains in the absence of CaM. B, the plot of Q4A to the preformed complex of Ca²⁺/CaM+Q4B indicates no detectable interaction. C, the isotherm showing the addition of Q4B to the preformed complex of Ca²⁺/CaM+Q4A revealed a K_d = 0.5 ± 0.2 nm (mean ± S.D., n = 2). Curve fitting was performed using the competitive model in Origin version 7.

Figure 5.

MST analysis of apoCaM affinity for Q4B and Q4A peptides. Titration plots are shown at the bottom for Alexa Fluor 594–tagged apoCaM (200 nm), titrated with Q4B (A), up to 220 μm, which displayed a K_d = 10 μm (confidence interval 6–17 μm, n = 3), compared with Q4A (B), which was too weak to determine an accurate equilibrium constant (n = 2). The normalized fluorograms are shown at the top, with the analyzed time points highlighted in light blue and light red, including Tjump + thermophoresis activity in the analysis. Five traces exhibiting high levels of aggregation were excluded. Data were analyzed using PALMIST software, and the figures were created using GUSSI software. Error bars, S.D.

Figure 6.

HSQC-NMR analysis shows changes in the apoCaM spectrum when combined with Q4B, but not Q4A. The full spectrum representing 150 μm ¹⁵N-labeled apoCaM (in 1 mm EGTA) is shown in A. The spectra in B–D are expanded regions of the boxed region of the full spectrum, comparing apoCaM before (orange spectrum) and after titration with Q4A, Q4B, or both peptides (blue spectra) at a ratio of 1:1.2. E, solution NMR structure of apoCaM (PDB entry 1DMO, conformation 27) with an expanded view of the C lobe. The green, labeled regions represent residues with peak changes greater than 2 S.D. values above the mean peak height after addition of Q4B to apoCaM. F, schematic depicting a possible model of the C lobe of apoCaM (orange), interacting with only the B domain (gold) of a single KCNQ4 subunit, with A and B domains in a nonhelical state.

Figure 7.

TROSY-HSQC-NMR analysis of the relationship between molar ratio of Ca²⁺/CaM or Ca²⁺/CaM+Q4B and alterations in apoCaM residues. Shown is the full NMR spectrum of 50 μm [¹H-¹⁵N]CaM (A) or 50 μm [¹H-¹⁵N]CaM + 62.5 μm Q4B (D), in which the green peaks are from residues of the metal-free protein in ChHBS, purple peaks are those upon the addition of 50 μm Ca²⁺, and orange peaks are from residues upon the addition of 100 μm Ca²⁺. The labeled boxes in the full spectrum images refer to the expanded regions in which B and C correspond to Ca²⁺ titrated to CaM only, and E and F are expanded regions of the Ca²⁺ titrations to CaM+Q4B. Those peaks showing overlapping residues are those that were unaffected by the added Ca²⁺, indicated in color. In contrast, peaks that do not overlap in color indicate a change in the spectral peak of the corresponding residue with the addition of Ca²⁺. G, graphical representation of the total number of peak height changes of the single NMR titration >55% above the mean peak height after each titration of Ca²⁺. Because we could no longer track peaks from the previous titrations at 200 μm, we included all peak changes counted by visual inspection for 200 and 400 μm Ca²⁺. The box at the bottom right shows the expected molar ratios of Ca²⁺/CaM EF-hands in each case.

Figure 8.

Graphical plot of CaM residues displaying changes in spectral peaks over the range [Ca²⁺] titration series of molar ratios of Ca²⁺. HSQC-NMR peak heights that changed >55% from the previous titration of Ca²⁺ to 50 μm [²H-¹⁵N]CaM or 50 μm [²H-¹⁵N]CaM + 62.5 μm Q4B starting in ChHBS, are represented by the gray or black bars. Only peaks that could be tracked from the original apoCaM position are shown. A and B, significant peak changes of peptide-free CaM in response to Ca²⁺ at a ratio of 1:4 EF-hands (i.e. 50 μm Ca²⁺ to 50 μm total CaM in A or 2:4 EF-hands in B). The same analysis was performed for Ca²⁺ titration into CaM + Q4B ranging up to 150 μm Ca² (C, D, and E). Cartoon schematics are shown with each plot, to show the estimated stoichiometry of Ca²⁺ ions (green ovals) with respect to the EF-hands and our speculated movement of CaM (orange dumbbell) with respect to the Q4B peptide (thick gold line). The schematics on the right side of the graph represent metal-free CaM and CaM + Q4B before the addition of Ca²⁺.

Figure 9.

WT or mutant CaM-lobe mutants do not affect the voltage dependence of KCNQ4 currents. A, representative perforated patch voltage-clamp recordings from CHO cells expressing KCNQ4 channels together with either WT or the indicated CaM mutants. The kinetics of activation at 10 mV and deactivation (inset) at −60 mV after the prepulse were quantified by fits to a double and single exponential, respectively. Fits are shown in gray. B, superimposed are the voltage-dependent activation curves for the KCNQ4 + CaM combinations shown in A, assayed as the amplitude of the tail current at −60 mV after the 500-ms prepulse to the indicated voltages. C, comparison of the activation and deactivation time constant values. D, a table summarizes the data shown in A–C. Overexpression of CaM WT or the indicated CaM mutants did not induce significant changes in the voltage dependence of activation or in the kinetics of activation or deactivation. Error bars, S.E., as these are group data.

Figure 10.

Proposed “lobe-switching model” for CaM regulation of neuronal KCNQ channels. Shown schematically is our proposed model of how Ca²⁺ ions direct CaM in interactions with, and regulation of, neuronal KCNQ channels. We here exclude the likely role of Mg²⁺ or other ions, as discussed under “Discussion.” 1, under very low (<10 nm) cytosolic [Ca²⁺] (a physiological state that we cannot determine), apoCaM is prebound to the B domain (gold section of KCNQ subunit), and the NMR and MST data derive a K_d of ∼10–20 μm. During this state, the PIP₂ interaction sites within the proximal C terminus at the S6Jx (pre-A helix) and the A-B domain linker are available to interact with PIP₂. Under such conditions, the A and B domains are likely disordered, not in a helical conformation, and it is still unclear whether this conformational state would represent a functional channel at the plasma membrane, where PIP₂ is located. 2, when [Ca²⁺]_i is in the range of that in cytoplasm in neurons at rest, Ca²⁺ first binds the EF-hands of the C-lobe (indicated by the change in color from orange to pink), displacing CaM from the B domain. 3, upon a rise in [Ca²⁺] in the proximity of the channel, the Ca²⁺-bound C-lobe binds to the A domain with a K_d of ∼400 nm, inducing an α-helical conformation to the A domain (cyan, now shown as a helix). This twisting motion may impose torque on the PIP₂ interaction sites in the proximal C terminus, partially weakening their interactions with PIP₂. 4, in the final step, under a strong [Ca²⁺] signal (such as strong stimulation of certain G_q/11-coupled receptors), the EF-hands of the N-lobe become occupied by Ca²⁺ ions, enhancing its affinity for the B domain, inducing it into a helical formation, retaining C-lobe binding to the A domain (still a helix). This final twisting motion may completely twist or pull away the PIP₂ interaction sites from the inner leaflet of the membrane, severely hindering the ability of the C terminus to bind PIP₂, resulting in maximal Ca²⁺/CaM-mediated inhibition of neuronal M channels. The subnanomolar affinity of the CaM:A+B trimeric complex may allow Ca²⁺ ions to rapidly move in and out of EF-III and -IV while maintaining a stable complex during such elevated [Ca²⁺] conditions; thus, accounting for crystals variably observed to contain Ca²⁺ ions in the C-lobe EF-hands.