Yoshida MM et al. (2024), Regulation of condensin II by self-suppression ...

XB-ART-60454

Mol Biol Cell 2024 Feb 01;352:ar21. doi: 10.1091/mbc.E23-10-0392.

Show Gene links Show Anatomy links

Regulation of condensin II by self-suppression and release mechanisms.

Yoshida MM , Kinoshita K , Shintomi K , Aizawa Y , Hirano T .

???displayArticle.abstract???
In vertebrates, two distinct condensin complexes, condensin I and condensin II, cooperate to drive mitotic chromosome assembly. It remains largely unknown how the two complexes differentially contribute to this process at a mechanistic level. We have previously dissected the role of individual subunits of condensin II by introducing recombinant complexes into Xenopus egg extracts. Here we extend these efforts by introducing a modified functional assay using extracts depleted of topoisomerase IIα (topo IIα), which allows us to further elucidate the functional similarities and differences between condensin I and condensin II. The intrinsically disordered C-terminal region of the CAP-D3 subunit (the D3 C-tail) is a major target of Cdk1 phosphorylation, and phosphorylation-deficient mutations in this region impair condensin II functions. We also identify a unique helical structure in CAP-D3 (the D3 HEAT docker) that is predicted to directly interact with CAP-G2. Deletion of the D3 HEAT docker, along with the D3 C-tail, enhances the ability of condensin II to assemble mitotic chromosomes. Taken together, we propose a self-suppression mechanism unique to condensin II that is released by mitotic phosphorylation. Evolutionary implications of our findings are also discussed.

???displayArticle.pubmedLink??? 38088875
???displayArticle.pmcLink??? PMC10881152
???displayArticle.link??? Mol Biol Cell

Species referenced: Xenopus laevis
Genes referenced: cdk1 top2a
GO keywords: mitotic chromosome condensation [+]

???attribute.lit??? ???displayArticles.show???

	FIGURE 1:. Topo II-depleted add-back assay helps to reveal functional features unique to condensin II. (A) Mouse sperm nuclei were incubated with extracts depleted of both endogenous condensins and topo IIα (Δcond I/II Δtopo II extract) that had been supplemented with control buffer (buffer), condensin II wild-type holocomplex (holo[WT]), subcomplexes (ΔD3[WT], ΔG2[WT], ΔD3ΔG2[WT]) or mutant holocomplex containing CAP-D3 that lacks its C-tail (holo[D3-dC201]; formerly referred to as holo[D3-dC]; Yoshida et al., 2022). The condensin II complexes were added at a final concentration of 200 nM, the same concentration used in the standard add-back assay (Yoshida et al., 2022). After 150 min, the reaction mixtures were fixed and labeled with antibodies against mSMC4 and histone H3. DNA was counterstained with DAPI. Shown here is a representative image from over 25 nuclei examined per condition. Different relative exposure times were used wherever indicated (0.5x). Scale bar, 10 µm. (B) Quantification of histone H3-positive (shown in blue) and mSMC4-positive (shown in red) areas in the structure formed by holo(WT), ΔG2(WT) and holo(D3-dC201) in (A). Histone H3 immunofluorescence signals were used instead of DAPI signals to measure the size of whole DNA mass because the corresponding DAPI signals with ΔG2(WT) were too strong to set a proper threshold for binarization in the quantification. The means and SDs are shown (n = 33, 37 and 26 from left to right). (C) Comparison of the ratio of mSMC4-positive to histone H3-positive areas measured in (B). The means and SDs are shown. (D) Mouse sperm nuclei were incubated with Δcond I/II Δtopo II extracts that had been supplemented with the condensin II complexes (holo[WT], ΔG2[WT], or holo[D3-dC201] at 200 nM; Yoshida et al., 2022) or the condensin I complexes (holo[WT], ΔG[WT], or holo[H-III6Q] at 35 nM; Kinoshita et al., 2022) and analyzed as in (A). Shown here is a representative image from over 30 nuclei examined per condition. Scale bar, 10 µm. (E) Quantification of histone H3-positive (shown in blue) and mSMC4-positive (shown in red) areas in the structures shown in (D). The means and SDs are shown (n = 31, 34, 28, 30, 32 and 37 from left to right). (F) Comparison of the ratio of mSMC4-positive to histone H3-positive areas measured in (E). The means and SDs are shown.
	FIGURE 2:. Functional dissection of the D3 C-tail in the standard add-back assay. (A) Schematic representation of the D3 C-tail. Human D3 C-tail contains 11 SP/TP sites (green circles) and three regions that are weakly conserved among vertebrate orthologs (pink boxes). D3-dC201 mutant lacks the entire 201 amino acids of the D3 C-tail. In this study, three additional deletion mutants that lack the last 34, 91, and 163 amino acids (D3-dC34, D3-dC91, and D3-dC163, respectively) were constructed. (B) Mouse sperm nuclei were incubated with extracts depleted of endogenous condensins (Δcond I/II extracts) that had been supplemented with control buffer (buffer) or condensin II holocomplex containing full-length CAP-D3 (WT) or CAP-D3 with progressive deletions (D3-dC34, D3-dC91, D3-dC163, or D3-dC201) at 200 nM. After 150 min, the reaction mixtures were fixed, and labeled with an antibody against hCAP-H2. DNA was counterstained with DAPI. Shown here is a representative image from over 20 chromosome clusters examined per condition. Scale bar, 10 µm. (C) DAPI and hCAP-H2 intensities were measured along lines drawn perpendicular to the chromosomes shown in (B), and line scan profiles for the width of chromosomes were created for each condition (n = 20 chromosomes). The means and SDs were normalized individually to the DAPI intensities (arbitrary unit [a.u.]) at the center of chromosome axes (distance = 0 μm) within each set. Intensities of hCAP-H2 were normalized relative to the central value obtained with holo(WT). The data from a single representative experiment out of three repeats are shown.
	FIGURE 3:. Functional dissection of the D3 C-tail in the topo II-depleted add-back assay. (A) Mouse sperm nuclei were incubated with extracts depleted of both endogenous condensins and topo IIα (Δcond I/II Δtopo II extract) that had been supplemented with control buffer (buffer) or condensin II holocomplex containing full-length CAP-D3 (WT) or CAP-D3 with progressive deletions (D3-dC34, D3-dC91, D3-dC163, or D3-dC201) at 200 nM. After 120 min, the reaction mixtures were fixed, and labeled with an antibody against mSMC4. DNA was counterstained with DAPI. Shown here is a representative image from over 25 nuclei examined per condition. A different relative exposure time (2x) was used where indicated. Scale bar, 10 µm. (B) Quantification of DAPI-positive areas (left), mSMC4-positive areas (middle), and the ratio of mSMC4-positive to histone H3-positive areas (right) in the structures described in (A). The means and SDs are shown (n = 34, 27, 31, 30, and 30, respectively). The p-values listed were assessed by two-tailed Mann−Whitney U test.
	FIGURE 4:. Phosphorylation-deficient mutations in the D3 C-tail decrease chromosome loading of condensin II. (A) Schematic representation of the D3 C-tail. Wild-type D3 C-tail contains 11 Cdk consensus (SP/TP) sites (first line, green circles). D3-T1415A contains a mutation in which T1415 is substituted with alanine (second line, purple circle), whereas D3-C11A contains mutations in which all serines and threonines in the 11 Cdk consensus sites were substituted with alanines (third line, purple circles). (B) Mouse sperm nuclei were incubated with extracts depleted of endogenous condensins (Δcond I/II extract) that had been supplemented with control buffer (buffer), condensin II holo(WT), holo(D3-T1415A), holo(D3-C11A), or ΔD3(WT) at 200 nM. After 150 min, the reaction mixtures were fixed, and labeled with an antibody against hCAP-H2. DNA was counterstained with DAPI. Shown here is a representative image from over 15 chromosome clusters examined per condition. Scale bar, 10 µm. (C) Signal intensities of hCAP-H2 from the experiment described in (B) were divided by DAPI signal intensities and the mean values were normalized relative to the value obtained with holo(WT). The means and SDs are shown (n = 15 chromosome clusters). The p-values listed were assessed by two-tailed Welch’s t test. The data from a single representative experiment out of three repeats are shown.
	FIGURE 5:. The D3 C-tail is a major target of Cdk1-mediated phosphorylation. (A) A metaphase extract depleted of both endogenous condensins was supplemented with the condensin II wild-type and mutant complexes. Aliquots were taken at the indicated time points and analyzed by immunoblotting with the indicated antibodies. Xenopus topoisomerase IIα (Xtopo IIα) was used as a loading control. Asterisks (*) indicate non-specific bands. (B) Recombinant condensin II holo(WT) and holo(D3-C11A) were incubated without (-) or with (+) recombinant cyclin B-Cdk1 (M-CDK) at 25°C for 60 min. The reactions were stopped by adding SDS sample buffer and the samples were analyzed by immunoblotting with the indicated antibodies.
	FIGURE 6:. CAP-D3 has a unique helical structure that is predicted to directly interact with CAP-G2. (A) AlphaFold2 prediction of human CAP-D3. The core region containing the U-shaped HEAT-repeat solenoid (amino acids 1-1172) is shown in orange, whereas the disordered D3 C-tail (amino acids 1277-1498) is shown in blue. N-terminal (N) and C-terminal (C) ends are labeled. Human CAP-D3 contains a non-HEAT helical structure (amino acids 1173-1276) between the HEAT core region and the C-tail (cyan). Prediction was assessed on May 12, 2023. (B) (Left) AlphaFold2-multimer prediction of the interaction between human CAP-D3 (core region in orange; non-HEAT helical structure in cyan; C-tail in blue) and full-length human CAP-G2 (dark red). The non-HEAT helical structure in human CAP-D3 was predicted to be positioned adjacent of a HEAT surface of CAP-G2. Much like the multimer prediction of the human D3 C-tail and CAP-G2 (Supplemental Figure S2B), the D3 C-tail of the full-length CAP-D3 was modeled to interact with CAP-G2 in similar ways. (Right) Close view of the predicted interaction. CAP-D3 K1214 within the helical structure was placed in the vicinity of CAP-G2 K340, which was consistent with the previous chemical crosslinking data reported by Kong et al. (2020). Prediction was assessed on August 4, 2023.
	FIGURE 7:. Deletion of the D3 HEAT docker further accelerates condensin II-mediated chromosome assembly. (A) Schematic representation of human CAP-D3. Full-length D3 has a core region composed of HEAT repeats (orange), the HEAT docker (cyan), and the C-tail (blue). A deletion mutant that lacks the D3 C-tail alone (D3-dC201) and a deletion mutant that lacks the HEAT docker and the C-tail (D3-dC326) are shown. (B) Mouse sperm nuclei were incubated with extracts depleted of endogenous condensins (Δcond I/II extracts) that had been supplemented with a condensin II holocomplex containing full-length CAP-D3 (WT) or CAP-D3 with C-terminal deletions (D3-dC201 or D3-dC326) at a final concentration of 200 nM or 50 nM. After 150 min, the reaction mixtures were fixed, and labeled with an antibody against hCAP-H2. DNA was counterstained with DAPI. Shown here is a representative image from over 20 chromosome clusters examined per condition. The images of hCAP-H2 signal were first captured at a single exposure time (1/3x) and then increased digitally (to 1x) to better visualize signals of complexes apart from holo(D3-dC326) at 200 nM. Scale bar, 10 µm. (C) DAPI and hCAP-H2 intensities were measured along lines drawn perpendicular to the chromosomes shown in (B), and line scan profiles for the width of chromosomes were created for each condition (n = 20 chromosomes). The means and SDs were normalized individually to the DAPI intensities (arbitrary unit [a.u.]) at the center of chromosome axes (distance = 0 μm) within each set. Intensities of hCAP-H2 were normalized relative to the central value obtained with holo(WT). The data from a single representative experiment out of three repeats are shown.
	FIGURE 8:. Models. (A) Cell cycle regulation of condensin II. In the basal state, CAP-D3 and CAP-G2 physically interact with each other. This interaction is initiated by a contact between the D3 HEAT docker and a specific surface of the CAP-G2 subunit and is further stabilized by the disordered D3 C-tail. Upon mitotic entry, Cdk phosphorylation of the D3 C-tail relaxes this interaction, thereby activating condensin II. Phosphorylation-deficient mutations in the D3 C-tail (D3-C11A) greatly impair such activation. Deletion of the CAP-G2 subunit (ΔG2) or of the D3 C-tail (D3-dC201) abolishes this self-suppression mechanism, resulting in artificial hyperactivation. Deletion of the D3 HEAT docker together with the D3 C-tail (D3-dC326) further accelerates condensin II−mediated chromosome assembly. (B) Conversion from condensin I to condensin II, and vice versa. On the one hand, CAP-G2-dependent bulk loading activity of condensin II is intrinsically weaker than CAP-G-dependent bulk loading activity of condensin I (Yoshida et al., 2022). When CAP-G’s function is compromised either by deleting CAP-G (ΔG[WT]) or by introducing point mutations in a region of CAP-H that works together with CAP-G (holo[H-III6Q]), the resulting condensin I mutant complexes become “condensin II-like” (Kinoshita et al., 2022; left, from top to bottom). On the other hand, the self-suppression mechanism involving the direct interaction between CAP-D3 and CAP-G2 (indicated by the blue circle) is unique to condensin II (this study). Thus, if the interaction is destabilized by deleting the D3 HEAT docker (holo[D3-dC326]) and/or the D3 C-tail (holo[D3-dC201]), the resulting condensin II mutant complex becomes “condensin I-like” (right, from bottom to top).

References [+] :

Abe, The initial phase of chromosome condensation requires Cdk1-mediated phosphorylation of the CAP-D3 subunit of condensin II. 2011, Pubmed