Fig 2: Differential rates of FBP dissociation from CUL1.A–C SILAC-based quantitative proteomic profiling of SCF dissociation. Samples were prepared as illustrated in (A). Relative levels of RBX1 and CSN subunits (B) or FBPs (C) that remained CUL1-bound after 60 min of incubation with the SKP1Sortase·SKP2ΔLRR chase are shown. The relative assembly in (B) served as a negative control for the assay. Label-swap was distinguished by dots in different colors. The H/L (or L/H) ratios of CUL1 ligands in each sample were normalized to the H/L (or L/H) ratio of 3xFLAGCUL1, respectively. Data are presented as mean ± SEM; n = 6 biological replicates. D Pearson correlation analysis of exchange vs. relative dissociation levels of FBPs. Average levels of FBP exchange in DKO cell lysates determined via SILAC exchange assays (Fig. 2E–G) were plotted against the average percentage of dissociation for the corresponding FBPs in the SILAC dissociation assay (C). Source data are provided as a Source Data file.

Fig 3: FBP changes the koff but not the kon of CUL1·CAND2.A Binding (kon) of CAND1 or CAND2 to CUL1 preassembled with SKP1·SKP2. kobs of CFPCAND1·CUL1FlAsH or CFPCAND2·CUL1FlAsH association when 5 nM CFPCAND1 or CFPCAND2 was mixed with increasing concentrations of SKP1·SKP2-bound CUL1FlAsH (Supplementary Fig. 7). The linear slope gave the kon. Data are presented as mean ± SEM; n = 3 independent experiments. B–D SKP1·SKP2 increases the koff of CUL1·CAND1 and CUL1·CAND2 to similar degrees. Real-time detection of CFP fluorescence in a stopped-flow apparatus upon the addition of 80 nM SKP1·SKP2 to 5 nM CFPCAND1·CUL1FlAsH (B) or CFPCAND2·CUL1FlAsH (C). D kobs of CFPCAND1·CUL1FlAsH or CFPCAND2·CUL1FlAsH dissociation when increasing concentrations of SKP1·SKP2 were mixed with 5 nM CFPCAND1·CUL1FlAsH or CFPCAND2·CUL1FlAsH (Supplementary Fig. 8). Fitting with hyperbola gave the koff and KM. Data are presented as mean ± SEM; n = 5 independent experiments. Source data are provided as a Source Data file.

Fig 4: Architecture of the CUL1·CAND2 complex.A Composite cryo-EM map of CUL1·CAND2. CUL1, blue; CAND2, green. The color scheme is maintained throughout the work. B Overall structural comparison of CUL1·CAND2 with CUL1·CAND1 (PDB: 1U6G) in cartoon representation. Cartoon (C) and surface (D) representations of the interface between the β hairpin of CAND2 and the first cullin repeat of CUL1. E Structural collision between the CAND2 β hairpin and SKP1. The structure of the CUL1·CAND2 complex was superimposed with the cryo-EM structure of the CAND1·CUL1·SKP1·FBP intermediate complex (PDB: 7ZBW). F The pincer structure in CAND1 and CAND2, which clash with a loop in SKP1. The structure of the CUL1·CAND2 complex was superimposed with the cryo-EM structure of the CAND1·CUL1·SKP1·FBP intermediate complex (PDB ID: 8OR0).

Fig 5: FRET-based kinetic characterization of CUL1·CAND2.FRET assays reporting the interaction between CUL1 and CAND1 (A), and between CUL1 and CAND2 (B). Fluorescence emission spectra after excitation at 430 nm with 100 nM CFPCAND1 (or CFPCAND2), 100 nM CUL1FlAsH, a mixture of the two (FRET), the chase control for FRET, and the buffer alone. Signals were normalized to the peak donor emission. Chase: 1 μM CUL1. Proteins were added in the indicated order and analyzed within 1 min after mixing. C Dissociation kinetics of CUL1·CAND1 and CUL1·CAND2. Relative fluorescence emission of CFP vs. time after the addition of CUL1 chase to preassembled CFPCAND1 (or CFPCAND2) and CUL1FlAsH. Single exponential fit gave the koff. Data are presented as mean ± SEM; n = 3 independent experiments. D Association kinetics of CUL1·CAND1 and CUL1·CAND2. The observed rates (kobs) of CFPCAND1·CUL1FlAsH or CFPCAND2·CUL1FlAsH association at different concentrations of CUL1FlAsH (Supplementary Fig. 6) were plotted. The linear slope gave the kon. Data are presented as mean ± SEM; n = 3 independent experiments, except for kobs of CFPCAND2·CUL1FlAsH association at 20 nM and 30 nM of CUL1FlAsH (n = 4). Source data are provided as a Source Data file.