|
Figure 1. Domain architecture and overall structure of the hCAP‐G‐H subcomplex
hCAP‐G is composed of 1,015 amino acids and contains 19 HEAT repeats. hCAP‐H is composed of 730 amino acids and contains 5 conserved motifs (I: hSMC2 binding region, II: hCAP‐D2 binding region, III: DNA‐binding region, IV: hCAP‐G binding region, V: hSMC4 binding region).Scheme of the hCAP‐G‐H subcomplex. Residues 1–478 of hCAP‐G were fused to residues 554–900 of hCAP‐G. hCAP‐G (1–478, 554–900) was co‐expressed in E. coli and co‐purified for crystallography.Cartoon diagram of the crystal structure of hCAP‐G (orange) in complex with a fragment of hCAP‐H (green). Unstructured, disordered regions are indicated by the dots. The 19 HEAT repeats (H1‐H19) and 2 disordered loops (H12 loop and H15 loop) of hCAP‐G, and 4 helices (α2, α3′, and α4) of hCAP‐H are labeled. The N‐ and C‐termini of CAP‐G and CAP‐H are also indicated. A molecule of 4‐(2‐hydroxyethyl)‐1‐piperazineethanesulfonic acid (HEPES) is shown by the orange‐colored stick model. A 90‐degree rotated version is shown on the right.Comparison of the hCAP‐G‐H subcomplex with its related structures. Superimposition of the structures of hCAP‐G‐H (orange), S. cerevisiae YCG1‐BRN1 (PDB ID: 5OQQ; blue), and S. pombe CND3‐CND2 (PDB ID: 5OQR; green) is presented as a Cα‐tracing model.Comparison of the DNA‐bound form with DNA‐free forms. Superimposition of the structures of DNA‐bound YCG1‐BRN1 (PDB ID: 5OQN; red), hCAP‐G‐H (orange), YCG1‐BRN1 (blue), and CND3‐CND2 (green) is presented as in (D).
|
|
Figure EV1. Secondary structures and structure‐based sequence alignment of human CAP‐G and CAP‐H
Secondary structures and structure‐based sequence alignment of human CAP‐G (hCAP‐G), Xenopus laevis CAP‐G (XCAP‐G), and Saccharomyces cerevisiae YCG1. The secondary structural elements of hCAP‐G and YCG1 are drawn above and below the sequence alignments, respectively. Identical and homologous residues are shown on black and gray backgrounds, respectively. The colored circles indicate residues of hCAP‐G that interact with hCAP‐H (green) or bind to HEPES (red). Residues of YCG1 that interact with BRN1 and dsDNA are labeled with light blue and red, respectively. The YC1 and YC2 regions indicate residues essential for DNA binding defined by Kschonsak et al
17.Structure‐based sequence alignment of hCAP‐H, XCAP‐H, and BRN1. The secondary structural elements of hCAP‐H and BRN1 are drawn above and below the sequence alignments, respectively. Identical and homologous residues are shown on black and gray backgrounds, respectively. The colored circles indicate residues of hCAP‐H that interact with hCAP‐G (orange) and residues of BRN1 that interact with YCG1 (purple) or dsDNA (red). BC1, BC2, latch, and buckle regions defined by Kschonsak et al
17, and motifs III and IV of CAP‐H are also shown in Fig 1A.
|
|
Figure EV2. Structure of the hCAP‐G‐H subcomplex
Two molecules of the hCAP‐G‐H subcomplex in the asymmetric unit, shown by orange (hCAP‐G; a) and green (hCAP‐H; b), and blue (hCAP‐G; c) and pink (hCAP‐H; d) ribbon representations. The pink stick model indicates HEPES. Note that HEPES bound only one of the two hCAP‐G molecules (a‐molecule) present in the asymmetric unit.Schematic illustration of the structure and domain organization of hCAP‐G. Two antiparallel helices (A and B helices) comprising each HEAT repeat are colored in orange and light orange, respectively. The binding sites of hCAP‐H and DNA are indicated by the green and red double‐headed arrows, respectively. The H12 loop (residues 479–553) connecting the H12A and H12B helices, and the H15 loop (residues 660–690) connecting the H15A and H15B helices are shown by black loops.Comparison of the b‐factors of the hCAP‐G‐H subcomplex with its related structures. The structures of hCAP‐G‐H (left), S. cerevisiae YCG1‐BRN1 (PDB ID: 5OQQ; middle), and S. pombe CND3‐CND2 (PDB ID: 5OQR; right) are shown as a ribbon model colored by b‐factor. The b‐factors are shown in warm (high b‐factors) to cool colors (low b‐factors).
|
|
Figure 2. Structural details of the interaction between hCAP‐G and hCAP‐H
A. B–E.
|
|
Figure EV3. Structural details of the interaction between YCG1 and BRN1
A. B–F.
|
|
Figure 3. Identification of residues required for interaction between hCAP‐G and hCAP‐H
Zoomed‐in view of site IV. Residues of hCAP‐G and hCAP‐H are shown in orange and green, respectively. The dashed red lines indicate hydrogen bonds.3Q, 5Q, 3A, 5A, 2A, and Δ506–515 mutants of hCAP‐H. Motif IV (residues 461–503) contains amino acid residues highly conserved among eukaryotic species (X, Xenopus laevis; Dr, Danio rerio; Cm, Cyanidioschyzon merolae; Sp, Schyzosaccharomyces pombe; Sc, Saccharomyces cerevisiae; Ec, Encephalitozoon cuniculi). To produce the IV‐3Q, 5Q, 3A, 5A, and 2A mutants, the conserved aromatic amino acid residues (F463, F469, F473, F501, and Y503; labeled in dark blue) were substituted with glutamine (Q) or alanine (A) residues. The secondary structural element of hCAP‐H is drawn below the sequence alignments.Interaction analysis between hCAP‐G and hCAP‐H. Bacterial cell lysates co‐expressing hCAP‐G (residues 1–478, 554–900) and hCAP‐H (residues 394–515), either wild type (WT; lanes 2, 10 and 13), 3Q (F463Q, F469Q and F473Q; lane 3), 5Q (F463Q, F469Q, F473Q, F501Q and Y503Q; lane 4), 3A (F463A, F469A and F473A; lane 5), 5A (F463A, F469A, F473A, F501A and Y503A; lane 6), or 2A (F501A and Y503A; lane 11), or a C‐terminal deletion mutant (506–514 residues were deleted from 394–515; lane 14) were applied to Ni‐NTA agarose resin, and the bound fraction was analyzed by SDS–PAGE. Alternatively, a cell lysate co‐expressing mutant hCAP‐G (D647K) and wild‐type hCAP‐H was examined (lane 7). The uninduced cell lysate was also used as a negative control (lane 9).
|
|
Figure 4. CAP‐H motif IV is required for a physical interaction with CAP‐G
Expression of condensin I subunits in insect cells. The wild‐type (WT) or IV‐5Q mutant CAP‐H subunit was co‐expressed with the other four subunits (GST‐SMC4, SMC2, CAP‐D2, and CAP‐G) in insect cells. Cell lysates were prepared and subjected to SDS–PAGE, followed by immunoblotting with a mixture of antibodies against SMC2 and SMC4 (left panel) or against CAP‐D2, CAP‐G, and CAP‐H (right panel).Purification of the WT and IV‐5Q mutant condensin I complexes. Protein samples purified through glutathione‐affinity chromatography were subjected to SDS–PAGE and analyzed by CBB staining (left panel) or immunoblotting with a mixture of antibodies as described above (middle and right panels).Add‐back assay using the WT and mutant condensin I complexes. Xenopus extracts depleted of endogenous condensin complexes were supplemented with the purified complexes (from top to bottom; WT, IV‐5Q, ΔG, ΔG[IV‐5Q]). The supplemented extracts were then incubated with sperm nuclei to assemble mitotic chromosomes. The samples were fixed and labeled with an antibody against mSMC4 (red). DNA was counterstained with DAPI (blue). The data from a single representative experiment out of two repeats are shown. In the experiment shown here, multiple images were collected for condensin‐depleted extracts supplemented with the WT (n = 17), IV‐5Q (n = 22), ΔG (n = 20), and ΔG(IV‐5Q) (n = 25) complexes. The scale bar represents 10 μm.
|
|
Figure 5. DNA‐binding surfaces conserved between hCAP‐G and YCG1
Purification of hCAP‐G and hCAP‐G‐H subcomplexes: wild type (WT), CAP‐G K60D/R848E double mutant (K60D/R848E), and CAP‐G R168E mutant (R168E). Purified protein samples were subjected to SDS–PAGE and analyzed by CBB staining.Double‐stranded DNA (dsDNA) and single‐stranded DNA (ssDNA)‐binding assay of the hCAP‐G. 30‐bp dsDNA was incubated with no protein (lanes 1) or increasing amounts of WT hCAP‐G (WT; lanes 2–4). 30‐mer ssDNA was incubated with no protein (lanes 5) or increasing amounts of WT hCAP‐G (WT; lanes 6–8).The dsDNA binding assay for the hCAP‐G‐H subcomplexes. 30‐bp dsDNA was incubated with no protein (lanes 1, 5, and 9), increasing amounts of WT hCAP‐G‐H subcomplex (WT; lanes 2–4), CAP‐G K60D/R848E double mutant (K60D/R848E; lanes 6–8), or CAP‐G R168E mutant (R168E; lanes 10–12).The ssDNA binding assay for the hCAP‐G‐H subcomplexes used in panel (C).The molecular surface of hCAP‐G in complex with hCAP‐H. The structural model of hCAP‐G is shown in white. hCAP‐H is shown as a green ribbon model. Identical and homologous residues between hCAP‐G and YCG1 are shown in blue and cyan, respectively.Zoomed‐in view of the HEPES‐binding site. R168 of hCAP‐H interacts with HEPES.
|
|
Figure EV4. Competition between double‐stranded DNA (dsDNA) and single‐stranded DNA (ssDNA) for hCAP‐G‐H binding
30‐mer FAM‐ssDNA was incubated with no protein (lanes 1), WT hCAP‐G‐H (WT; lanes 2–8), or increasing amounts of 30‐bp dsDNA (WT; lanes 2–8).30‐bp FAM‐dsDNA was incubated with no protein (lanes 1), WT hCAP‐G‐H (WT; lanes 2–8), or increasing amounts of 30‐mer ssDNA (WT; lanes 2–8).
|