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

Abstract

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) states. 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 Cyclophilin A. 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 found two additional properties of KaiB's landscape. First, the Ground state experiences cold denaturation: At 4 °C, the PD state becomes the majorly populated state. Second, 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 relieves buried charged residues, and support this structure experimentally. These results provide mechanistic insight into how evolution can design a single-sequence that achieves specific timing needed for its function.

Keywords: circadian rhythm; metamorphic protein; relaxation dispersion NMR.

Fig. 1.

KaiB interconversion between Ground and FS state is on the hour time regime. (A) Current understanding of the KaiB:KaiC system in R. sphaeroides. Monomeric Ground state and FS state models were generated with AF-Cluster (18) with the fold-switching part colored for the same sequence regions. Structure of KaiB:KaiC complex: 8FWJ (17). (B) Perturbing populations of the Ground and FS state by incubating at 40 °C and jumping to 20 °C followed by real-time NMR spectra to quantify the interconversion rates at 20 °C depicted in the Inset. (C) Comparing ¹⁵N HSQC spectra at 4 °C and 20 °C reveals peaks in the disordered region of the spectrum at 4 °C not present at 20 °C. (D) Predicted backbone order parameters and secondary structure propensity of the major state at 4 °C. The N terminus retains the same fold as found in the Ground and FS states at 20 °C, but the C-terminal domain is more disordered. (E) Representative temperature-dependent peak shifts and population changes of the three states. (F) Populations for 11 residues for which the Ground, FS, and PD states could be assigned in the majority of temperature conditions, calculated from peak volume. The solid lines represent average populations at each temperature.

Fig. 2.

KaiB interconverts between the Ground and FS state via the PD state, and proline cis/trans isomerization is rate-limiting. (A) Incubating at 4 °C and jumping to 20 °C followed by ¹⁵N-HSQC time series at 20 °C reveals that KaiB cold denaturation is reversible, and that interconversion in and out of the PD state at 20 °C is significantly faster than the deadtime of this experiment (7 min). (B) Visualizing location of prolines and their difference between trans and cis conformations in Ground and FS states. Monomeric Ground state and FS state models were generated with AF-Cluster (18). (C) C_β and Cγ chemical shifts of P58, P65, and P67 in Ground, PD, and FS state, colored by cis/trans probabilities predicted with PROMEGA (26). (D) Scheme of proposed mechanism for accelerating interconversion by adding the peptidyl-prolyl cis/trans isomerase CypA. (E) Adding 30 µM CypA to the temperature-jump experiment of 1.2 mM KaiB accelerates the Ground to FS interconversion rates by 2.5-fold. (F) ¹⁵N-HSQC spectra indicate peak broadening for residues in the C-terminus of the PD state upon addition of CypA. Peaks with significant broadening or change in peak location are labeled, as well as the three peaks corresponding to the Ground, FS, and PD state for C-terminal residue Glu90. (G) Change in peak intensities across all assigned states at 4 °C indicates that CypA binds to the C-terminus of the PD state. (H) Single point mutations of prolines P58 or P68 are sufficient to destabilize the FS state.

Fig. 3.

The KaiB Ground state interconverts with a fourth “Enigma” state. (A) Representative ¹⁵N backbone CEST data at 4 °C for Asn84, indicating that the Ground state exchanges with both the PD state and another state. (B) CEST data at 20 °C for Asn84, indicating that at 20 °C, exchange with the Enigma state is still present, and it is not the FS state. (C) Populations and interconversion rates for models of exchange between the Enigma, Ground, and PD states determined by global fits of the number of residues indicated (SI Appendix, Figs. S5 and S6). (D) All residues with signs of exchange at 20 °C are indicated with spheres. Residues fit to one two-state process (Ground to Enigma state) are colored in purple. (E) Representative ¹⁵N backbone CPMG data of residues with small R_ex values (Asn22) and larger R_ex values (Asn84). (F) Estimated R_ex values from CPMG at 25 °C and 35 °C. The majority of residues have greater R_ex with increasing temperature, indicating the CPMG is reporting on slow-intermediate-timescale processes that is consistent with the processes observed by CEST. (G) Estimated R_ex values for all residues with R_ex > 1 s⁻¹ visualized on the structure. The residues with CPMG R_ex > 10 s⁻¹ agree with the residues exchanging with the Enigma state, reported by CEST. Residues with very small R_ex are possibly reporting on a putative cis–trans isomerization (SI Appendix, Figs. S4 and S8).

Fig. 4.

A subtle β1 to β3 register shift predicted by AF2 and experimentally supported by NMR is consistent with the Enigma state. (A) The highest-pLDDT KaiB Ground state produced by AF-Cluster (18) has the same hydrogen bonding as KaiB from T. elongatus [PDB: 2QKE (31)]. We refer to this state in KaiB^RS as “TE-like”. (Left) superposition of TE-like state and 2QKE. (Right) detail of beta-strand hydrogen bonding. (B) AF-Cluster, and AF2 (22) with single-sequence sampling, also predict another ground-like state with lower pLDDT where β3 is register-shifted by two amino acids. Arrows indicate direction of register shift. (Left) superposition of register-shifted state and 2QKE, (Right) detail of altered beta-strand hydrogen bonding. (C) Visual comparison of ¹⁵N chemical shifts for the ground state (green) and enigma state (purple) with chemical shift predictions from three chemical shift predictors [SHIFTX2 (33), SPARTA+ (34), and UCBshift (35)] using the TE-like model (light green) and the register-shifted structure (light purple). (D) Comparing correlations between predicted and measured chemical shifts for three tested chemical shift predictors shows that predictions from the TE-like structure correlate better with the Ground state, and predictions from the Register-shifted structure correlate better with the Enigma state. (E) Methyl–methyl NOESY supporting contacts unique to the register-shifted structure, contacts depicted in F. In the register-shifted structure, Leu82 is on the exterior of the b3 strand and can interact with Ile54, but in the TE-like structure, it is flipped into the interior to form the hydrophobic core (Right), and is not within NOE distance of Ile54. Also, in support of the register-shifted structure is two distinct sets of NOE cross-peaks between L60 and L85, one corresponding to Ground state chemical shifts and the other to Enigma methyl chemical shifts.

Fig. 5.

The conformational landscape of KaiB is unified by a stable subdomain. (A) Calculated free energies of all states and barrier heights, incorporating information from temperature-jump and CEST experiments (green: 20 °C, blue: 4 °C). Models for the Enigma, Ground, and FS state were generated with AF-Cluster (18). The model for the PD state is the N terminus spliced from the Ground state model. (B) The proposed mechanism of fold-switching via a PD intermediate suggests that the N terminus of KaiB should be stable as an independent unit. Indeed, KaiB-∆Cterm [KaiB(1–46)] is stably folded at 4 °C as shown via ¹⁵N-HSQC. (C) Predicted backbone order parameter S² and secondary structure propensity for KaiB-∆Cterm. (D) Calculated structure of KaiB-∆Cterm using CS-Rosetta (38).