A folded conformation of MukBEF and cohesin

Abstract

Structural maintenance of chromosomes (SMC)-kleisin complexes organize chromosomal DNAs in all domains of life, with key roles in chromosome segregation, DNA repair and regulation of gene expression. They function through the entrapment and active translocation of DNA, but the underlying conformational changes are largely unclear. Using structural biology, mass spectrometry and cross-linking, we investigated the architecture of two evolutionarily distant SMC-kleisin complexes: MukBEF from Escherichia coli, and cohesin from Saccharomyces cerevisiae. We show that both contain a dynamic coiled-coil discontinuity, the elbow, near the middle of their arms that permits a folded conformation. Bending at the elbow brings into proximity the hinge dimerization domain and the head-kleisin module, situated at opposite ends of the arms. Our findings favour SMC activity models that include a large conformational change in the arms, such as a relative movement between DNA contact sites during DNA loading and translocation.

Figure 1. Folded conformation of MukBEF and cohesin.

(a) Purification of MukBEF. Elution of the MukBEF complex from a Q ion exchange (IEX) column. Peak fractions were separated by SDS-PAGE and stained with Coomassie Blue. An uncropped gel image is shown in Supplementary Data Set 1. (b) SEC of the MukBEF complex, MukB and MukEF. Proteins were separated on Superose 6 Increase. (c) Negative stain EM of native MukBEF. A typical field of view is shown. (d) Particle instances for observed MukBEF conformations are shown on the left. A cartoon highlighting the position of the elbow is shown on the right. (e) Cross-linking of MukBEF with BS³. SEC profiles for native and cross-linked material are shown. (f) Negative stain EM of BS³ cross-linked MukBEF. Typical fields of view for particles from SEC peak 1 and SEC peak 2 are shown. (g) Negative stain 2D class averages for extended (left) and folded (right) conformations, using circular masks of 948 Å and 640 Å, respectively. Data was collected from samples of peak 1 and peak 2 of the SEC shown in (d). (h) Negative stain EM of BS³ cross-linked cohesin. A typical field of view is shown on the left. Class averages using a circular mask of 500 Å are shown in the middle panel.

Figure 2. Elbow positions revealed by cross-linking and mass spectrometry.

(a) Inter-subunit cross-links of MukBEF. Links are colored according to their false discovery rate (FDR). (b) Kernel density estimates for cross-link sites mapped onto the E. coli MukEF subcomplex (PDB: 3EUH). (c) Identification of the MukB elbow region. Long-distance cross-links (d ≥ 100 aa) in a coordinate system along the coiled-coil arm and their midpoints are shown on top. The bottom panel shows the kernel density estimate for the midpoint positions. An inset shows the piecewise interpolation function used to map residue numbers to the arm coordinate system. (d) Kernel density estimates for cross-link sites mapped onto the cohesin Smc3–Scc1 interface (PDB: 4UX3). (e) Identification of the cohesin elbow region as in (c).

Figure 3. Structure of the MukB elbow.

(a) Crystal structure of an E. coli MukB arm fragment. The top panel illustrates the design of the fusion construct used for crystallography. The bottom panel shows the refined atomic model obtained from the X-ray diffraction experiment. (b) Identification of the elbow. Long-distance cross-link midpoint density (see Fig. 2d) was mapped onto the structure. (c) Structure of the elbow. The C-terminal coiled-coil helix is distorted (kinked) close to the conserved Tyr416 on the N-terminal helix. Residues for visual reference between the views are shown in grey. Residues targeted by mutagenesis (Supplementary Fig. 3) are highlighted in black. See also Table 1.

Figure 4. In vivo cross-linking of Pds5 to the Smc1 hinge.

(a) Illustration of the BPA cross-linking experiment (top) and mapping of tested BPA substitutions onto a homology model of the cohesin hinge (bottom). (b) Screen for Smc1(BPA) cross-links to Pds5 and Smc3. BPA was incorporated into the indicated Smc1 positions, cells were treated with UV, cohesin was immunoprecipated via a PK9-tag on Scc1 and products were analyzed by Western blotting. (c) UV-dependent cross-linking of Smc(K620BPA) and Pds5. Cells were either treated or not treated with UV and products were analyzed as in (b). (d) Cross-linking of Smc(K620BPA) and Pds5 depends on Pds5 binding to Scc1. The left panel shows the position of Scc1 V137 in its Pds5 binding site (mapped to the L. thermotolerans structure, PDB: 5F0O). The right panel shows a cross-linking experiment of Smc1(K620BPA) in the presence of Scc1(V137K)-PK9 as in (b). Uncropped blots are shown in Supplementary Data Set 1.

Figure 5. Conservation of the SMC elbow.

Coiled-coil prediction profiles for a diverse set of SMC protein sequences were generated by MARCOIL. Profiles for N- and C-terminal parts of the arms were separately aligned on their center coordinate and averaged. 90 % confidence intervals (purple shading) were estimated by 100 times random resampling with replacement. The Mms21 binding site of Smc5 is highlighted in green. N, number of sequences used for generating the respective aggregate profiles.

Figure 6. Models for conformational changes that involve the SMC elbow.

(a) Model for regulated conformational switching of SMC–kleisin complexes. Transitioning between extended and folded states in either direction might be driven by the ATPase cycle introducing mechanical strain into the SMC arms. (b) Paper model: conversion between extended and folded states is achieved by twisting the arms of the model. (c) Models for DNA translocation and loop extrusion involving a folded state. (left) “Inchworm translocation” using distance changes between two DNA binding sites, one of which might be a topological entrapment device or ring, for example formed by the kleisin and the SMC heads or the kleisin and its associated HAWK or KITE subunits. Folding at the elbow might cause the distance change. (right) Translocation using the segment-capture mechanism that enlarges a loop held in a bottom chamber by merging with a smaller loop captured in a top chamber. Folding at the elbow might drive DNA from top to bottom.