Nanobody regulation of C-type inactivation in Kv1.3 channels

Figures

• 

  Fig. 1. Structure of hKv1.3 with NB bound and structures of the ion selectivity filter of the Shaker Kv channel in conducting and C-type inactivated conformations. a, b Side and external views of the structure of hKv1.3 with four NB1.3 nanobodies bound to the external side of the channel and interacting with both the pore domain (PD) and voltage-sensing domains (VSD). Cytoplasmic tetramerization domain (T1). c Cutaway side view of the hKv1.3 model bound to NB1.3, showing interactions with both the VSD and PD. Ion selectivity filter (SF). Front and rear S1–S4 domains and contributions to the T1 domain are omitted for clarity. d Structural model of the ion selectivity filter of the Shaker Kv channel in a conducting conformation (pdb id: 7SIP). e Structural model of the ion selectivity filter of the Shaker Kv channel in a dilated inactivated conformation (pdb id: 8TEO).

• 

  Fig. 2. Structure of hKv1.3 with C-type inactivated and conducting ion selectivity filters. a, b Cryo-EM density and structural models of the ion selectivity filter of apo hKv1.3 in D1 and D2 conformations. c Cryo-EM density and structural model of the ion selectivity filter of hKv1.3 with NB1.3 bound in the D3 conformation. d Cryo-EM density and structural model of the ion selectivity filter of hKv1.3 with ShK-Fab bound in a presumably conducting conformation. e, f Current traces obtained using a three-pulse protocol without leak subtraction for WT hKv1.3 and the H451V mutant in the absence and presence of NB1.3. g Plot of P3/P1 for WT hKv1.3 and the H451V mutant in the absence and presence of 100 nM NB1.3. Smooth curves are Boltzmann fits with V1/2 and z values of −40.8 mV and 4.4 for H451V in control solution, −37.4 mV and 7.1 for H451V in the presence of NB1.3 (n = 4 in 3 independent experiments), −38.4 mV and 3.8 for WT in control solution and −56.1 mV (n = 9 in 7 independent experiments) and 2.2 for WT in the presence of NB1.3 (n = 5 in 4 independent experiments). Error bars represent S.E.M. h Plot of time constants (τ) for inactivation (light gray and orange triangles) or…

• 

  Fig. 3. Interaction of NB1.3 with the S1–S4 voltage-sensing domain and turret of hKv1.3. a Structural model of hKv1.3 illustrating NB1.3 interacting with the S1–S2 loop within the VSD and the turret within the PD. b Cryo-EM density and structural models of the interaction of NB1.3 (orange) with the turret loop of hKv1.3 (purple). c Cryo-EM density and structural models of the interaction of NB1.3 (orange) with the S1–S4 domain of hKv1.3 (pink). d Families of currents elicited by depolarization to voltages between −80 and +20 mV in the absence and presence of 100 nM NB1.3 for WT hKv1.3 and the Y265G, P424G and G427H mutants. Holding voltage was −80 mV, and tail voltage was −60 mV. e Plot of normalized currents at 0 mV (I/I0) for WT hKv1.3 and different hKv1.3 mutants for 100 nM NB1.3 obtained from (d). Values for n and independent experiments were: 3 and 3 for WT, 3 and 3 for Y265G, 5 and 5 for A421S, 4 and 4 for P424G, 5 and 5 for T425V, 3 and 3 for S426G and 3 and 3 for G427H. Error bars represent S.E.M. f Families of currents elicited by depolarization to voltages between −80 and +20 mV of different NB1.3 mutants. Holding voltage was −80 mV and tail voltage was −60 mV. g Plot of…

• 

  Fig. 4. Functional characterization and structure of the P424G mutant of hKv1.3. a Current traces obtained using a three-pulse protocol without leak subtraction from a holding voltage of −100 mV for the hKv1.3 P424G mutant in the absence and presence of 100 nM NB1.3. b Plot of P3/P1 for WT hKv1.3 and the P424G mutant in the absence and presence of 100 nM NB1.3. Smooth curves are Boltzmann fits with V1/2 and z values of −40 mV and 7 for P424G in control solution (n = 4 in 3 independent experiments) and −42.3 mV and 4 in the presence of NB1.3 (n = 3 in 3 independent experiments). Error bars represent S.E.M. P3/P1 is less than 1 at negative P2 voltages due to incomplete recovery from inactivation (see “Methods”). WT data are from Fig. 1g. c Inactivation kinetics of WT hKv1.3 and the P424G mutant in the presence and absence of 100 nM NB1.3. τ values for WT were obtained by fitting a double exponential function in control to current traces in P2 and a single exponential function in the presence of NB1.3. For P424G a single exponential function was fit to current traces elicited during P2 in both control (n = 5 in 4 independent experiments) and in the presence of NB1.3 (n = 4 in 4 independent…

• 

  Fig. 5. Comparison of hydrophobic side chain interactions between residues in the turret and selectivity filter of hKv1.3 for both conducting and C-type inactivated conformations. a, b Cryo-EM map and model showing hydrophobic residue interactions between the pore helix and turret of the H451V mutant of hKv1.3 in a conducting conformation without and with NB1.3 bound. c, d Cryo-EM map and model showing disrupted hydrophobic residue interactions between the pore helix and turret of apo hKv1.3 in D1 and D2 conformations. e Cryo-EM map and model showing disrupted hydrophobic residue interactions between the pore helix and turret of the D3 conformation of hKv1.3 with NB1.3 bound. f Cryo-EM map and model showing disrupted hydrophobic residue interactions between the pore helix and turret of the P424G mutant of hKv1.3. CA distances between W436 and M450 are colored blue. Distances between SD of M450 and CD1 of W436 and between CE of M450 and CD1 of F428 are colored red.

• 

  Fig. 6. The G427H mutant of hKv1.3 stabilizes a conducting state and interferes with NB1.3, promoting inactivation. a Families of currents elicited by depolarization in the absence and presence of 100 nM NB1.3 for the G427H mutant. b Plot of normalized conductance (G/Gmax) obtained from tail current measurements for WT hKv1.3 and the G427 mutant in the absence and presence of 100 nM NB1.3. Smooth curves are Boltzmann fits with V1/2 and z values of −19.8 mV and 2.7 for G427H in the control solution and −19.8 mV and 3 in the presence of NB1.3 (n = 3 in 2 independent experiments). Error bars represent S.E.M. WT data are from Fig. 1j. c Current traces obtained using a three-pulse protocol without leak subtraction for hKv1.3 G427H in the absence and presence of 100 nM NB1.3. d Plot of P3/P1 for WT hKv1.3 and the G427H mutant in the absence and presence of 100 nM NB1.3. Smooth curves are Boltzmann fits with V1/2 and z values of −39.6 mV and 3.7 for G427H in control solution and −31.2 mV and 3.7 in the presence of NB1.3 (n = 3 in 2 independent experiments). Error bars represent S.E.M. WT data are from Fig. 1g. e Cryo-EM density and structural model of the ion selectivity filter of the G427H…

• 

  Fig. 7. F428A mutants of hKv1.3 promote C-type inactivation. a Families of currents elicited by depolarization in the presence of 2 or 100 mM external K+ for the F428A mutant of hKv1.3. b Inactivation kinetics for WT hKv1.3 and the F428A mutant in 100 mM external K+. In both cases, a single exponential function was fit to the data to obtain τ. For F428A, n = 3 in 3 independent experiments and for WT n = 5 in 3 independent experiments. Error bars represent S.E.M. c Family of currents elicited by short depolarization in the presence of 100 mM external K+ for the F428A mutant. d Plot of normalized conductance (G/Gmax) obtained from tail current measurements for WT hKv1.3 and the F428A mutant, both with 100 mM external K+. Smooth curves are Boltzmann fits with V1/2 and z values of −5 mV and 2.8 for F428A in control solution (n = 3 in 3 independent experiments) and −32.9 mV and 4.8 for WT in control solution (n = 8 in 4 independent experiments). Error bars represent S.E.M. e Families of currents elicited by depolarization without leak subtraction in the absence and presence of 100 nM NB1.3 for the F428I mutant. External K+ for (e–l) was 2 mM. f Plot of normalized conductance (G/Gmax) obtained…

• 

  Fig. 8. Structural changes in the interaction of hydrophobic residues between the turret and selectivity filter during C-type inactivation. Cryo-EM density and model of the H451V mutant of hKv1.3 view from the side (a) or from the external side of the membrane (b). Key residues are labeled, hydrogen bonds are indicated by blue dotted lines and hydrophobic interactions by red dotted lines. c Side view of cryo-EM density and model of hKv1.3 with NB1.3 bound with key residues indicated. Note the absence of key hydrogen bonds and hydrophobic interactions illustrated in (a). d Superimposing models for the turret from this study and our previous study5. e Cryo-EM density and model of the P242G mutant of hKv1.3 viewed from the extracellular side of the membrane with key residues indicated. Note the absence of key hydrogen bonds and hydrophobic interactions illustrated in (c).