Fig 1.
Crystal structures of SidK N-terminal domain.
Structures of SidK(16–278) were determined where one (A) and two (B) monomers were in the asymmetric unit. SidK(16–278) is an elongated protein comprised of three α-helical bundles ('Bundle I', 'Bundle II', 'Bundle III'). Bundle I includes two short α-helices that deviate from the main α-helical bundle axis (A, green arrowheads). C, In the crystal structure with two SidK monomers in the asymmetric unit, Bundle I is rotated ~180° with respect to Bundle II. This structural rearrangement allows the interface between Bundle I and Bundle II in the monomer structure (A, red oval and arrowhead) to be maintained in the dimer structure (B, red oval and arrowhead) through domain swapping between the two monomers. Scale bar: 25 Å.
Fig 2.
Cryo-EM structure of the V-ATPase:SidK3 complex.
A, Three SidK proteins (teal) are bound to the V-ATPase in the soluble catalytic V1 region. The SidK density is well-defined in the N-terminal region where it binds the V-ATPase (teal arrowheads) and is poorly-resolved in the C-terminal region closer to the membrane-embedded part of the complex (teal dashed circles). B, SidK binds to the N-terminal region of subunit A of the V-ATPase. The major interaction surface is on SidK α-helical bundle I (blue bracket) and appears to involve residues G24, Y28, F62, S85, and W122. C, The cryo-EM map density (gray surface) reveals a feature resembling a 'hook' at the N terminus of SidK that is missing from the crystal structure. The C-terminal region of SidK is flexibly-tethered to its N-terminal region and is poorly resolved in the cryo-EM density. For clarity, the surface representation of SidK was rendered at a higher density threshold for the N-terminal domain (Threshold 0.04) than for the C-terminal domain (Threshold 0.015). Scale bars: 25 Å.
Fig 3.
Inhibition of the V-ATPase by SidK.
A, SidK, either full length or residues 10–414, inhibits V-ATPase by approximately 40% with 10× excess of SidK to available binding sites. “Baf.” indicates bafilomycin. B, SidK does not inhibit ATPase activity of the F-type ATP synthase, showing that it is a specific inhibitor of V-ATPases. C, Yeast expressing SidK and SidK point mutations show normal growth on medium at pH 5.5. Wild type SidK reduces yeast growth on medium buffered to pH 7.0, indicating V-ATPase inhibition. SidK point mutations F62A and S85E alleviate SidK inhibition of V-ATPase while point mutations Y28A and W122A do not. D, Western blotting shows that all of the point mutations tested are expressed in S. cerevisae at levels similar to wild type (PGK, phosphoglycerate kinase as a loading control), except G24E, which was consequently excluded from the analysis. *, p<0.01.
Fig 4.
SidK reduces the flexibility of the A-subunit and its binding affinity for TNP-ATP.
A, Slices through the cryo-EM maps of V-ATPase and V-ATPase:SidK3 in state 3, low-pass filtered to 10 Å. In V-ATPase without SidK (-SidK), the C-terminal region of the A-subunit (ACT) in the ‘open’ conformation (Aopen) has lower density compared to the A-subunit in the ‘tight’ or ‘loose’ conformations (Atight, Aloose). With SidK bound (+SidK), density for the ‘open’ A-subunit is comparable to other subunits in the V-ATPase:SidK3 map. B, Comparison between the A-subunits in the ‘open’ conformation from V-ATPase (gray surface) and V-ATPase:SidK3 (gray mesh) cryo-EM maps. ACT is better defined with SidK bound. Scale bars: 25 Å. C, The average density of ACT relative to ANT is higher with SidK bound to V-ATPase in all rotational states. D, Representative relative fluorescence values and fitted curves from titration of TNP-ATP into a solution containing V1 alone (blue) or V1 with SidK (pink). E, Dissociation constant Kd values calculated from data in (D) show increased Kd values with SidK treatment, indicating that SidK binding decreases the affinity of the V1 subcomplex for TNP-ATP. *, p<0.01; **, p<0.001.
Table 1.
Summary of data collection and refinement statistics.
The information for the highest resolution shell is given in parentheses.