Fig 2: BLM interacts with condensin II through a short linear N-terminal motif.a BLM truncations tested in mammalian-two-hybrid (M2H) assay against full-length SMC2. Interaction is indicated by secreted alkaline phosphatase (SEAP) activity as ++ (strong), + (mild), and - (absent). See Supplementary Fig. S1c for measurements of SEAP activity from three experiments. Red box indicates BLM region required for interaction with SMC2. b Alignment of BLM residues 150-184 with BLM from other vertebrates. Residues W154, M157, and F160 were mutated to lysine, referred to as BLM-WMFmut. c SEAP activity detected in M2H assay between two independently generated pVP16-BLM-WMFmut clones (#1, #2) and pM-SMC2. pM3-VP16 is a positive control from the manufacturer expressing a fusion of the GAL4 DNA-binding domain to the VP16 activation domain. Significance of differences between means ± SD was determined by a t-test and is reported as ****p ≤ 0.0001. d Plasmid expressing BLM-WMFmut was stably transfected into BLM-KO cells. Western blot of nuclear extracts from two independently generated BLM-WMFmut clones (#1, #2) and of WT, BLM-KO, and BLM-COMP cells was probed with BLM antibody. PCNA was used as a loading control. Expression levels were analyzed 3 h after release from G1/S arrest. e Co-immunoprecipitations of BLM and BLM-WMFmut #1 were probed with antibodies against SMC2, condensin II-specific subunit NCAPH2, and known BLM-interacting proteins. Whole-cell extracts were prepared from cells 3 h after release from G1/S arrest. f PLA using antibodies against endogenous BLM and NCAPH2 in BLM-COMP, BLM-WMFmut #1, and BLM-KO cells 3 h after release from G1/S arrest. Scale bars, 10 μm. BLM-NCAPH2 PLA foci/nucleus were quantified (BLM-COMP, n = 102 nuclei; BLM-WMFmut #1, n = 130 nuclei; BLM-KO, n = 100 nuclei). Significance of differences was determined by a Mann–Whitney test and is reported as ****p ≤ 0.0001. For additional PLA controls, see Supplementary Fig. S1b.

Fig 3: Model for the effect of the BLM-condensin II interaction on genome stability.a Model of how BLM binding to condensin II increases efficiency of detection and repair of BLM substrates. Current models of loop extrusion posit that the hinge domain reels in new DNA through the SMC2-SMC4 lumen creating an intermediate loop with gate opening potentially allowing the intermediate loop to join the main loop50,72,74,106. BLM, through its interaction with condensin II at the WMF site, can scan this intermediate loop for its substrates such as R-loops, DNA breaks, or G-quadruplexes, leading to their resolution prior to becoming an impediment to the replication or transcription machinery, thus mediating normal DNA transactions and genome stability. The many DNA repair and replication factors that bind to BLM’s disordered tail likely contribute to resolution and repair. In the absence of BLM from condensin II due to the WMF mutation, the DNA lesions and replisome blocks are not resolved efficiently, thus increasing risk of collisions with the replisome or transcriptional machinery, leading to impaired replisome progression and fork stalling, impaired DNA-damage response, delayed DSB repair and, eventually, mitotic chromosome aberrations. Created in BioRender. Rodemoyer, B. (2025) https://BioRender.com/h40u746. b Model of how BLM-condensin II interaction may contribute to suppression of crossovers. Extrusion of adjacent loops may aid BLM helicase-mediated branch migration of Holliday junctions between sister chromatids. BLM-condensin II interaction puts BLM in close proximity to the HR intermediate, increasing efficiency of dissolution as noncrossovers (NCO) in cooperation with interacting proteins. Loss of BLM-condensin II interaction due to BLM-WMFmut reduces dissolution of HR intermediates by BLM complexes and increases resolution via cleavage by structure-specific endonucleases, causing crossovers (CO). Created in BioRender. Rodemoyer, B. (2025) https://BioRender.com/k17g890.

Fig 4: BLM interacts with condensin II.a SMC2 was identified by mass spectrometry in a co-immunoprecipitate of endogenous BLM from nuclear extracts of WT cells 3 h after release from G1/S arrest4. Select known BLM interacting proteins are shown for comparison. Isogenic BLM-KO cells were subjected to the same analysis and yielded zero hits for the listed proteins. Experiment was performed in triplicate. b Reciprocal co-immunoprecipitation of endogenous BLM and SMC2 from nuclear extracts of WT and BLM-KO cells 3 h after release from G1/S arrest. c, Proximity ligation assay (PLA) using antibodies against endogenous BLM and SMC2 in WT, BLM-COMP, and BLM-KO cells 3 h after release from G1/S arrest and quantification of BLM-SMC2 PLA foci/nucleus (WT, n = 111 nuclei; BLM-COMP, n = 103 nuclei; BLM-KO, n = 104 nuclei). Scale bars, 10 μM. Significance of difference between WT and BLM-COMP was determined by a Mann–Whitney test (ns, not significant). For additional PLA controls, see Supplementary Fig. S1a. d Five-subunit condensin I and II complexes (H, NCAPH; D2, NCAPD2; G, NCAPG; H2, NCAPH2; D3, NCAPD3; G2, NCAPG2). e Co-immunoprecipitation of endogenous BLM from whole-cell extracts of WT cells 3 h after release from G1/S arrest was probed with antibodies against condensin I and II subunits. Condensin subunits are color-coordinated with their respective condensin from (d). f Co-immunoprecipitation of endogenous NCAPH from whole-cell extracts of WT cells 3 h after release from G1/S arrest was probed with antibodies against condensin I and II subunits and BLM. g Co-immunoprecipitation of endogenous NCAPH2 from whole-cell extracts of WT cells 3 h after release from G1/S arrest was probed with antibodies against condensin I and II subunits and BLM.

Fig 5: BLM interaction with NCAPH2, SMC2, SMC4, RPA, and TOP2A in S and G2 phases and regulation by phosphorylation.Endogenous BLM was immunoprecipitated after release from G1/S arrest in (a) the absence of λ protein phosphatase (λPPase) and in (b) the presence of λPPase and probed with antibodies against SMC2, NCAPH2, RPA, and TOP2A. Cells were synchronized by double-thymidine, released, and BLM immunoprecipitated every hour for 12 h. IgG was used as a negative control for immunoprecipitation. Endogenous BLM was immunoprecipitated after release from G1/S arrest in (c) the absence of λ protein phosphatase (λPPase) and in (d) the presence of λPPase and probed with antibodies against SMC2, SMC4, and NCAPH2. Cells were synchronized by double-thymidine, released, and BLM immunoprecipitated every hour for 12 h. IgG was used as a negative control for immunoprecipitation. Experiment was performed at least twice with similar results. e Endogenous NCAPH2 was immunoprecipitated from WT, BLM-KO, and BLM-WMFmut #1 cells and probed with antibodies against BLM, SMC2, and SMC4. Cell cultures were arrested at G1/S by double-thymidine treatment, released, and NCAPH2 immunoprecipitated every hour for 7 h. IgG was used as a negative control for immunoprecipitation. Experiment was performed twice with similar results. f Endogenous NCAPH2 and SMC2 was immunoprecipitated from the same WT lysate used for hours 3–7 in (e) and probed with antibodies against NCAPH2, BLM, SMC2, and SMC4. IgG was used as a negative control for immunoprecipitation. NCAPH2 immunoprecipitation was performed three times with similar results and the SMC2 reciprocal immunoprecipitation once.