Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2024 Jul 3;15(1):5604.
doi: 10.1038/s41467-024-49882-x.

DCAF15 control of cohesin dynamics sustains acute myeloid leukemia

Grant P Grothusen #  1 , Renxu Chang #  1 , Zhendong Cao  1 , Nan Zhou  1 , Monika Mittal  1 , Arindam Datta  1 , Phillip Wulfridge  2 , Thomas Beer  3 , Baiyun Wang  4 , Ning Zheng  4 , Hsin-Yao Tang  3 , Kavitha Sarma  2 , Roger A Greenberg  1 , Junwei Shi  1 , Luca Busino  5
Affiliations

DCAF15 control of cohesin dynamics sustains acute myeloid leukemia

Grant P Grothusen et al. Nat Commun. .

Abstract

The CRL4-DCAF15 E3 ubiquitin ligase complex is targeted by the aryl-sulfonamide molecular glues, leading to neo-substrate recruitment, ubiquitination, and proteasomal degradation. However, the physiological function of DCAF15 remains unknown. Using a domain-focused genetic screening approach, we reveal DCAF15 as an acute myeloid leukemia (AML)-biased dependency. Loss of DCAF15 results in suppression of AML through compromised replication fork integrity and consequent accumulation of DNA damage. Accordingly, DCAF15 loss sensitizes AML to replication stress-inducing therapeutics. Mechanistically, we discover that DCAF15 directly interacts with the SMC1A protein of the cohesin complex and destabilizes the cohesin regulatory factors PDS5A and CDCA5. Loss of PDS5A and CDCA5 removal precludes cohesin acetylation on chromatin, resulting in uncontrolled chromatin loop extrusion, defective DNA replication, and apoptosis. Collectively, our findings uncover an endogenous, cell autonomous function of DCAF15 in sustaining AML proliferation through post-translational control of cohesin dynamics.

PubMed Disclaimer

Conflict of interest statement

R.A.G. is a co-founder of RADD Pharmaceuticals and an advisor to Dong-A ST and TrevarX Biomedical, Inc. N.Zheng is a co-founder of and has financial interests in Seed Therapeutics and Molecular Glue Labs Ltd. N. Zheng also serves as a member of the scientific advisory board of SyntheX with financial interests. The remaining authors declare no competing interests.

Figures

Fig. 1
Fig. 1. DCAF15 is an AML-biased E3 ubiquitin ligase dependency.
a Schematic describing the design of CULLIN-RING E3 ubiquitin ligase (CRL) domain-focused sgRNA library. If known, sgRNAs targeted the receptor-adaptor interacting domain of the E3 (indicated with arrow), otherwise targeted first exon of gene. N8 NEDD8, R RING, U Ubiquitin. b Schematic describing workflow for CRISPR dropout screen using CRL sgRNA library. Protein domain Essentiality Score (ES) calculated for each gene as average[log2(final sgRNA abundance +1/initial sgRNA abundance)]. c Heatmap results for selected CRL genes and controls. Genes ranked by AML-biased ES, defined by the difference of ES in AML versus non-AML cell lines. d Overall survival of AML patients using GEPIA platform with low (bottom 75%) or high (top 25%) DCAF15 mRNA expression (TCGA-LAML dataset [https://www.cancer.gov/tcga]). TPM Transcripts Per Million. n = 106. log-rank test, **p-value < 0.01 (p = 0.0069). e DCAF15 mRNA expression in n = 252 AML patient samples compared to n = 6 normal hematopoietic stem cells (HSC) samples according to microarray expression profiling from GSE42519,, probe 91952_at. Welch’s two-sided t-test, *p-value < 0.05 (p = 0.0193). Center line shows the median, box limits show 75th and 25th percentiles, whiskers show minimum-maximum values, with the outlier shown as independent dot (Tukey method). Generated using BloodSpot platform. f Competition-based proliferation assay performed in Cas9 + MV4-11 cell line. sgRNA+ populations monitored over time for mCherry expression by flow cytometry. sgRNA+ proportions normalized to day 3 post lentiviral sgRNA infection monitored over 23 days. n = 3 biologically independent replicates. sgROSA26, negative control; sgPCNA, positive control. g MV4-11 Cas9+ cell lines stably expressing empty vector (EV) or HA-tagged CRISPR-resistant DCAF15[ntG207C] were infected with ROSA26-targeting (negative control) or DCAF15-targeting sgRNAs. Cells treated with DMSO or 3 µM indisulam for 6 h and lysates analyzed by Western blot for indicated proteins. Immunoblots representative of two independent experiments. h Competition-based proliferation assay performed in MV4-11 Cas9+ cell lines stably expressing empty vector (EV) or DCAF15[ntG207C] and infected with ROSA26-targeting (negative control) or DCAF15-targeting sgRNAs. sgRNA+ populations monitored over time for mCherry expression by flow cytometry. sgRNA+ proportions normalized to day 6 post lentiviral sgRNA infection monitored over 24 days. n = 3 biologically independent replicates. Two-way ANOVA (with Bonferroni’s multiple comparisons test), ****padj < 0.0001, *padj < 0.05 (padj = 0.021), ns not-significant. Source data are provided as a Source Data file.
Fig. 2
Fig. 2. DCAF15 loss suppresses AML via activation of p53.
a Differentially-expressed genes (sgDCAF15#2/sgROSA26) for MV4-11 (TP53-WT) cells revealed by RNA-seq. Shown in blue and red, respectively, are mRNAs significantly decreased or increased. n = 3 biologically independent replicates; DESeq2: two-sided Wald test adjusted with Benjamini and Hochberg method for multiple comparisons, padj < 0.05. b Enriched terms from MSigDB_Hallmark_2020 pathways gene set library using Enrichr. Top 5 enriched terms shown for genes up-regulated in MV4-11 sgDCAF15#2/sgROSA26 with padj < 0.05. log10(padj) scale. Dashed line indicates the cut-off for significantly enriched terms (padj < 0.05). c Transcription factors with enriched binding site motifs within genes significantly enriched (padj < 0.05) upon DCAF15-knockout in MV4-11 cells. Ranked using normalized enrichment score (NES). Generated using i-cisTarget platform. d Lysates from Cas9 + MV4-11 cells infected with ROSA26-targeting or DCAF15-targeting sgRNAs, harvested 4 days post-infection, analyzed by Western blot for indicated proteins. Immunoblots representative of three independent experiments. e Same as in (a), except MV4-11 TP53-knockout (sgTP53#6) cells utilized. n = 3 biologically independent replicates; DESeq2: two-sided Wald test adjusted with Benjamini and Hochberg method for multiple comparisons, padj < 0.05. f Competition-based proliferation assays in Cas9 + MV4-11 TP53-WT (sgROSA26) and TP53-knockout (sgTP53#4 and sgTP53#6) cell lines. Plotted are relative sgRNA+ proportions normalized to day 3 sgRNA+ proportions and respective mCherry-co-expressed ROSA26-targeting proportions over 25 days. Error bars mean ± SD, n = 3 biologically independent replicates. g Cas9 + MV4-11 TP53-WT (sgROSA26) and TP53-knockout (sgTP53#6) cells infected with ROSA26- or DCAF15-targeting sgRNAs. On day 3 post-infection, cells stained for Annexin-V for quantification of early-apoptosis induction. Error bars mean ± SD, n = 3 biologically independent replicates. Two-way ANOVA (with Tukey’s multiple comparisons test), ****padj < 0.0001; ns not-significant. h Competition-based proliferation assay performed in 7 different Cas9+ AML cell lines. Plotted are relative sgRNA+ proportions normalized to day 3 sgRNA+ proportions over 24 days. n = 3 biologically independent replicates. Dashed line indicates 50% proliferation defect passed during course of experiment. TP53-WT or TP53-Mutant status is indicated. i Cas9+ AML cells infected with ROSA26- or DCAF15-targeting sgRNAs. On day 3 post-infection, cells stained for Annexin-V for quantification of early-apoptosis induction. Error bars mean ± SD, n = 3 biologically independent replicates. Two-way ANOVA (with Tukey’s multiple comparisons test), ****padj < 0.0001, ns not-significant. Source data are provided as a Source Data file.
Fig. 3
Fig. 3. DCAF15 interacts with the cohesin complex.
a Mass spectrometry analysis of FLAG empty vector (EV) or FLAG-DCAF15 immunoprecipitants in HEK293T cells. Scatter plot shows the ratio between the unique peptides in the FLAG-DCAF15 versus EV samples. Red and blue circles represent DCAF15 interactors enriched by at least 5-fold compared to EV. b Mass spectrometry analysis of biotin-streptavidin (Strep)-affinity purification (AP) from MV4-11 cells stably expressing HA-tagged TurboID-DCAF15 treated with DMSO or 5 μM MLN4924 for 6 h; biotin was added at 50 μM for final 2 h. Scatter plot shows the ratio between the unique peptides in the MLN4924 versus DMSO conditions. Red and blue circles represent the proteins enriched by at least 5-fold in the presence of MLN4924 compared to DMSO. c Venn diagram showing the overlap of DCAF15 FLAG-interactome (a) and proximity-interactome (b). d Heatmap showing number of unique peptides identified in (a) and (b) for the CUL4DCAF15 CRL complex, core cohesin ring subunits, and accessory cohesin-binding proteins. e Lysates from HEK293T cells transiently transfected with FLAG EV or DCAF15, treated with DMSO or 5 μM MLN4924 for final 6 h, were subjected to FLAG-immunoprecipitation and Western blot analysis for the indicated proteins. Immunoblots are representative of three independent experiments. f Lysates from MV4-11 HA-TurboID-EV and HA-TurboID-DCAF15 (D15) cells treated with DMSO or 5 μM MLN4924 for 6 h and 50 μM biotin for final 2 h were subjected to biotin-streptavidin (Strep)-affinity purification (AP) and Western blot analysis for the indicated proteins. Immunoblots are representative of three independent experiments. g Lysates from HEK293T cells stably expressing HA-tagged DCAF15 and transiently transfected with FLAG EV, FLAG-SMC1A-Hinge-Domain-Only (Hinge), FLAG-SMC1A-Head-Domain-Only (Head), or FLAG-SMC1A-WT, treated with 5 μM MLN4924 for 6 h, were subjected to FLAG-immunoprecipitation and Western blot analysis for the indicated proteins. Immunoblots are representative of three independent experiments. h AlphaFold prediction analysis in ChimeraX of DCAF15 (green), SMC1A_C-Terminal (magenta), and SMC1A_N-Terminal (cyan). Source data are provided as a Source Data file.
Fig. 4
Fig. 4. DCAF15 destabilizes cohesin-bound PDS5A and CDCA5.
a Lysates from HEK293T cells stably expressing HA-tagged DCAF15 with the indicated siRNA and/or FLAG-tagged cDNAs, treated with 5 μM MLN4924 for 6 h, were subjected to FLAG-immunoprecipitation and Western blot analysis for the indicated proteins. Immunoblots are representative of two independent experiments. b Lysates from Cas9+ HEL cells infected with lentiviruses encoding ROSA26- or DCAF15-targeting sgRNAs were subjected to immunoprecipitation with the indicated amounts of α-SMC1A antibody and Western blot analysis for the indicated proteins. Immunoblots are representative of three independent experiments. c Same as in (b), except that cells were treated with 5 µg/mL cycloheximide (CHX) for indicated times. Immunoblots are representative of two independent experiments. d Same as in (b), except that cells were treated with 10 µM MG132 (proteasome inhibitor, MG), 5 µM MLN4924 (MLN), or 3 µM indisulam (Ind) for indicated times. Immunoblots are representative of three independent experiments. e Lysates from HEK293T cells transiently transfected with FLAG-DCAF15, HA-SMC1A, HA-PDS5A, and HA-CDCA5 were subjected to FLAG-immunoprecipitation. Following, FLAG-immunoprecipitants were subjected to in vitro ubiquitination assay and Western blot analysis for the indicated proteins. Immunoblots are representative of two independent experiments. f Model of CRL4-DCAF15 binding the head domain of SMC1A and promoting the ubiquitination of PDS5A when bound to the SMC1A-SMC3-RAD21 tripartite ring. Source data are provided as a Source Data file.
Fig. 5
Fig. 5. DCAF15 sustains cohesin acetylation on chromatin.
a Lysates from HEK293T cells transiently transfected with FLAG EV or FLAG-DCAF15 treated with DMSO or the indicated concentrations of PCI-34051 (HDAC8 inhibitor, PCI), and treated with 5 μM MLN4924 for 6 h, were subjected to FLAG-immunoprecipitation and Western blot analysis for the indicated proteins. Immunoblots are representative of three independent experiments. b Lysates from Cas9+ HEL cells infected with lentiviruses encoding ROSA26- or DCAF15-targeting sgRNAs, treated with DMSO or the indicated concentrations of PCI, were subjected to immunoprecipitation with α-SMC1A antibody and Western blot analysis for the indicated proteins. Immunoblots are representative of three independent experiments. c HEK293 FLAG-HA-dTAG-DCAF15[Clone#1] cells were synchronized by double thymidine block (G1/S) followed by Ro-3306 block (G2) or nocodazole block (PM prometaphase), or left asynchronous (Asynch), and treated with DMSO or dTAG-13 (100 nM) for final 6 h prior to harvest. Lysates were subjected to immunoprecipitation with α-SMC1A antibody and Western blot analysis for the indicated proteins. Immunoblots are representative of two independent experiments. d HEK293 FLAG-HA-dTAG-DCAF15[Clone#1] cells were synchronized by double thymidine block followed by Ro-3306 block (3 μM for 24 h) and treated with DMSO or dTAG-13 (100 nM) for final 6 h of block. Following, cells were washed with PBS and released in fresh media containing DMSO or dTAG-13 (100 nM). Cells were harvested at indicated time points, and cell lysates from the cytoplasmic (Soluble) and nuclear (Chromatin) fractions were analyzed by Western blot for the indicated proteins. Immunoblots are representative of two independent experiments. e Venn diagrams showing the overlap of ChIP-seq peak counts for Total-SMC3 (Left) and ac-SMC3 (Right) in Cas9+ HEL cells infected with lentiviruses containing ROSA26- or DCAF15-targeting sgRNAs. f Enrichment plots displaying the average ChIP-seq signals of Total-SMC3 (Left) or ac-SMC3 (Right) in Cas9+ HEL cells infected with lentiviruses containing ROSA26- or DCAF15-targeting sgRNAs. 2500 bp upstream and downstream of peak centers are shown. g Representative genome browser track (produced using UCSC Genome Browser) showing Total-SMC3 (Top) and ac-SMC3 (Bottom) levels in Cas9+ HEL cells infected with lentiviruses containing ROSA26- or DCAF15-targeting sgRNAs. Source data are provided as a Source Data file.
Fig. 6
Fig. 6. DCAF15 loss results in enlargement of chromatin loops.
a Contact probability as a function of genomic distance in Cas9+ HEL cells infected with lentiviruses encoding ROSA26-targeting (negative control) or DCAF15-targeting sgRNAs. Hi-C data derived from n = 2 independent biological replicates in each condition. b Representative region in chromosome 6 (chr6:05,000,000–11,500,000) showing contacts, distribution of ac-SMC3, and sites with differential acetylation of SMC3. Black boundaries delineate TADs called in each condition. Arrows indicate domain boundaries lost in DCAF15−/− cells. c 3D pileup plots of TADs in each condition and log2(Fold Change) upon loss of DCAF15. d Box plot showing length distributions of loops called in DCAF15+/+ versus DCAF15−/− cells. n = 10,662 loops (sgROSA26) and 11,871 loops (sgDCAF15#2). Center line shows median, box limits show 75th and 25th percentiles, whiskers show 1.5X interquartile range (IQR); outliers not shown. sgROSA26 median loop length: 225 kb, sgDCAF15#2 median loop length: 310 kb. Welch’s two-sided t-test, ****p-value < 0.0001 (p = 2.2 × 10−16). e 3D pileup plots of loops categorized by indicated size ranges in each condition and log2(Fold Change) upon DCAF15 loss. f Representative region in chromosome 9 (chr9:14,000,000–18,000,000) showing contacts, distribution of ac-SMC3, and sites with differential acetylation of SMC3. Loops present in control only or in both conditions are indicated in blue. Loops gained in DCAF15-knockout condition only are indicated in red. Arrows indicate example regions of increased contact in DCAF15−/− cells.
Fig. 7
Fig. 7. DCAF15 loss disrupts DNA replication fork integrity.
a Workflow of DNA fiber analysis for quantification of replication fork restart efficiency after 1 mM hydroxyurea (HU)-induced replication fork stalling. Cells cultured in the presence of CldU for 20 min prior to addition of HU. After 60 min, cells washed with PBS prior to adding Idu for 40 min. Representative images of DNA fibers shown for Cas9+ HEL cells infected with ROSA26- or DCAF15-targeting sgRNAs. b Quantification of (a) shows IdU/CldU ratios (replication fork restart efficiency) in individual experimental conditions (sgROSA26: n = 154 fibers analyzed, sgDCAF15#2: n = 278 fibers analyzed). Median IdU/CldU ratios labeled and indicated by horizontal blue (sgROSA26) and red (sgDCAF15#2) bars. Two-tailed Mann–Whitney test, ****p-value < 0.0001. c Cas9 + MV4-11 TP53-WT and TP53-knockout (sgTP53#6) cells infected with ROSA26- or DCAF15-targeting sgRNAs harvested 4 days post lentiviral sgRNA infection. Cytoplasmic (Soluble) and nuclear (Chromatin) protein fractions analyzed by Western blot for indicated proteins. Immunoblots representative of three independent experiments. d Competition-based proliferation assay performed in Cas9 + MV4-11 cells infected with ROSA26- or DCAF15-targeting sgRNAs were grown in presence of 1 nM camptothecin (CPT). sgRNA+ populations monitored over time for mCherry expression by flow cytometry. Plotted values are relative sgRNA+ proportions growing with CPT normalized to matched sgRNA+ growing with DMSO. Error bars mean ± SD, n = 3 biologically independent replicates. Two-way ANOVA (with Bonferroni’s multiple comparisons test), ****padj < 0.0001, ns not-significant. e Same as in (d), except cells grown in the presence of olaparib (Ola) for 7 days. Relative sgRNA+ proportion quantification is shown for cells growing with olaparib normalized to matched sgRNA+ proportion growing with DMSO. Error bars mean ± SD, n = 3 biologically independent replicates. Two-way ANOVA (with Bonferroni’s multiple comparisons test), ****padj < 0.0001, ns not-significant. f HDAC8-mediated cohesin deacetylation enables CRL4-DCAF15 recruitment. DCAF15-mediated destabilization of cohesin-bound PDS5A and CDCA5 enables timely ESCO1/2-mediated cohesin re-acetylation, vital for accurate halting of DNA loop extrusion, efficient DNA replication fork progression, and sustained AML proliferation (Left). In the absence of DCAF15, HDAC8-de-acetylated cohesin is inappropriately hyper-loaded with PDS5A and CDCA5, which precludes ESCO1/2-mediated cohesin re-acetylation and leads to uncontrolled DNA loop extrusion, defective DNA replication fork progression, and activation of apoptosis (Right). Source data are provided as a Source Data file.
See this image and copyright information in PMC

Similar articles

See all similar articles

References

    1. Skaar JR, Pagan JK, Pagano M. SCF ubiquitin ligase-targeted therapies. Nat. Rev. Drug Discov. 2014;13:889–903. doi: 10.1038/nrd4432. - DOI - PMC - PubMed
    1. Duan S, Pagano M. Ubiquitin ligases in cancer: functions and clinical potentials. Cell Chem. Biol. 2021;28:918–933. doi: 10.1016/j.chembiol.2021.04.008. - DOI - PMC - PubMed
    1. Jagannath S, et al. A phase 2 study of two doses of bortezomib in relapsed or refractory myeloma. Br. J. Haematol. 2004;127:165–172. doi: 10.1111/j.1365-2141.2004.05188.x. - DOI - PubMed
    1. Richardson PG, et al. Bortezomib or high-dose dexamethasone for relapsed multiple myeloma. N. Engl. J. Med. 2005;352:2487–2498. doi: 10.1056/NEJMoa043445. - DOI - PubMed
    1. Attar EC, et al. Phase I and pharmacokinetic study of bortezomib in combination with idarubicin and cytarabine in patients with acute myelogenous leukemia. Clin. Cancer Res. 2008;14:1446–1454. doi: 10.1158/1078-0432.CCR-07-4626. - DOI - PubMed

MeSH terms

-