The conformational landscape of fold-switcher KaiB is tuned to the circadian rhythm timescale

How can a single protein domain encode a conformational landscape with multiple stably-folded states, and how do those states interconvert? Here, we use real-time and relaxation-dispersion NMR to characterize the conformational landscape of the circadian rhythm protein KaiB from Rhodobacter sphaeroides. Unique among known natural metamorphic proteins, this KaiB variant spontaneously interconverts between two monomeric states: the “Ground” and “Fold-switched” (FS) state. KaiB in its FS state interacts with multiple binding partners, including the central KaiC protein, to regulate circadian rhythms. We find that KaiB itself takes hours to interconvert between the Ground and FS state, underscoring the ability of a single sequence to encode the slow process needed for function. We reveal the rate-limiting step between the Ground and FS state is the cis-trans isomerization of three prolines in the fold-switching region by demonstrating interconversion acceleration by the prolyl isomerase CypA. The interconversion proceeds through a “partially disordered” (PD) state, where the C-terminal half becomes disordered while the N-terminal half remains stably folded. We discovered two additional properties of KaiB’s landscape. Firstly, the Ground state experiences cold denaturation: at 4°C, the PD state becomes the majorly populated state. Secondly, the Ground state exchanges with a fourth state, the “Enigma” state, on the millisecond timescale. We combine AlphaFold2-based predictions and NMR chemical shift predictions to predict this “Enigma” state is a beta-strand register shift that eases buried charged residues, and support this structure experimentally. These results provide mechanistic insight in how evolution can design a single sequence that achieves specific timing needed for its function.

. Taken together, this suggests that there could be even up to equal flux to the PD state through the enigma state as directly from the Ground state.The enigma to PD interconversion is underdetermined because of the too low populations.Physically, it would make most sense if both GS and enigma can interconvert with the PD state given their similar structures.These alternate fits rule out the path going exclusively through the Enigma state.

Fig. S1 .
Fig. S1.Major state at 4˚C is a partially disordered state.Chemical shifts of C-terminus (Leu46 onwards) for major state at 4˚C correlate well with random coil chemical shifts predicted by POTENCI 1 .(a) RMSD per residue of N, H, Ca, Cb chemical shifts relative to random coil shifts.(b-e) Measured (black) vs. predicted (grey) for random coil chemical shifts for N, H, C⍺, Cβ atoms.

Fig. S2 .
Fig. S2.Kinetics for Ground to FS interconversion and rate acceleration by CypA.Rate constants and populations fit from KaiB incubated at 4˚C (left) or 40 ˚C (right) and monitored at 20˚C (compare to Fig. 1b).

Figure S3 .
Figure S3.Conservation of P65 and P67 across KaiB phylogenetic tree in ref. 2 .a) Sequences in grey have P65 and P67 conserved.Sequences in magenta have only one of these two prolines, mutations noted on figure.b) Tree colored by pLDDT from closest-10 AF2 predictions and c) Predicted FS (red) or Ground state (blue) from ref. 2 .

Fig. S5 (
Fig. S5 (multi-page figure). 15N CEST data overlaid with global fits for (a) 3-state fit at 4˚C (PD state observed is boxed in blue), (b) 3-state fit at 20˚C, (c) 2-state fit to Ground and Enigma state at 20˚C.

Fig. S6 .
Fig. S6.Evaluating alternate models for fitting CEST data.(a) Minimal models representing data, reproduced from Fig. 3c.(b) Allowing all states to interconvert.(c) Constraining kGround->PD = kEnigma->PD.(d) Removing the Ground-PD interconversion results in notably worse models by reduced Chi 2. Taken together, this suggests that there could be even up to equal flux to the PD state through the enigma state as directly from the Ground state.The enigma to PD interconversion is underdetermined because of the too low populations.Physically, it would make most sense if both GS and enigma can interconvert with the PD state given their similar structures.These alternate fits rule out the path going exclusively through the Enigma state.

Fig
Fig. S7.(multi-page figure, 1 of 2). 15N CPMG data for all assigned Ground state peaks at 25˚C and 35˚C.

Fig
Fig. S7.(multi-page figure, 2 of 2). 15N CPMG data for all assigned Ground state peaks at 25˚C and 35˚C.

Fig. S8 .
Fig. S8.Different minor slow process fit in CEST data at 20˚C, likely due to cis/trans prolyl isomerization of Pro14.a) Populations and kinetics from 2-state fit.b) Location of residues fit to process visualized in red.Prolines P14, P32, P48 shown in black.Prolines P58, P65, P67, which are trans in the Ground and cis in the FS state, are shown in orange.c) CEST data from field strength = 20 Hz overlaid with fits from ChemEx.

Figure S9 .
Figure S9.Difference in hydrophobic packing of core between (a) TE-like and (b) register-shifted, proposed Enigma structure.(c) The hydrogen-bonding and hydrophobic core of the TE-like structure is homologous to the crystal structure of KaiB from Thermosynechococcus elongatus (PDB: 2QKE 3 ).I59, L60, A61 are involved in the dimerization interface in the tetrameric structure in T. elongatus, and this same geometry is predicted for homologous residues in the TE-like structure of R. sphaeroides (compare to (a)).

Fig. S10 ( 2 -
Fig. S10 (2-page figure).NOESY data supporting the TE-like structure as the major ground state, and minor enigma state as register shifted conformation.a) 15 N-edited NOESY supporting inter-strand NOE cross-peaks for both TE-like and register-shifted structure.TE-like NOE cross-peaks are annotated in black and Register-shifted cross-peaks are annotated in purple.b) Scheme of identified cross-peaks supporting TE-like structure.c) Scheme of identified cross-peaks supporting register-shifted structure.In (b) and (c), gray arrows indicate cross peaks where due to overlap, inter-strand NOE cross-peaks are ambiguous while red arrow indicates missing NOE cross-peak.

Fig. S10 ( 2 -
Fig. S10 (2-page figure).NOESY data supporting the TE-like structure as the major ground state, and minor enigma state as register shifted conformation.d) 13 C-edited NOESY annotated with cross-peaks supporting TE-like structure in black.e) 13 C-edited NOESY annotated with cross-peaks supporting register-shifted structure in purple.

Fig. S11 .
Fig. S11. 13C-1 H HMQC 2D collected at 20˚C on 13 C ILVM methyl-labelled KaiB.Ground state assignments are shown.Inset shows zoomed in region where L60 and L85 enigma state peaks are indicated in purple labels.Their corresponding ground state peaks are circled.

Fig. S12 .
Fig. S12. 13C-CH3 CEST data at 20˚C overlaid with global 2-state fit between Ground and Enigma state.This fit obtained kex = 173 ± 8 s -1 and pb = 7 ± 1%, in agreement with the rate and population obtained from 15 N-CEST.

Figure S14 .
Figure S14.AF2 in single-sequence mode predicts the Enigma state.Correlation of 15 N chemical shifts predicted using UCBshift 4 with Ground and Enigma chemical shifts, plotted for all KaiB structure models generated by (a) AF-Cluster 2 , and (b) increased sampling of AF2 5 in single-sequence mode.AF-Cluster models show difference in correlation to either chemical shifts from the Ground or Enigma state, and AF2 single-sequence sampling only samples the register-shifted structure.

Table S1 .
Proline isomerization states in homologous structures of KaiB Ground and FS state."-"= proline not conserved.