Fig 2: STK35 and PDIK1L bind selectively to the catalytically active form of SCP4(A) Western blot of cytoplasm and nuclear fractions from MOLM-13 cells.(B) Representative FLAG-SCP4 affinity purification western blot analysis for the subsequent mass spectrometry (MS) analysis. Cytoplasm and nucleus fractions from MOLM-13 cells stably expressing empty vector, FLAG-SCP4 WT, or catalytic mutant (D293A). Nuclear fraction was mixed with anti-FLAG M2 agarose overnight. The flow-through was analyzed to ensure efficient binding of the FLAG-tagged constructs (loaded as “unbound”). Following extensive washing, the agarose amount equivalent to the cytoplasm and nucleus fractions loading was boiled in Laemmli buffer and loaded as “FLAG-IP.” The rest was sent for the MS analysis.(C) Venn diagram depicting the overlap between proteins detected by MS and absent in empty vector controls in the 2 independent biological replicates.(D) Total unique peptide counts for SCP4, PDIK1L, and STK35 detected by MS in the 2 independent biological replicates.(E) Domain architectures and homology heat-map of human STK35 and PDIK1L. ATP-BS, ATP-binding site.(F–H) Immunoprecipitation followed by western blotting performed with the indicated antibodies. The whole-cell lysate was prepared from HEK293T 24 h post-transfection with the indicated constructs (F). The nuclear lysates were prepared from the human MOLM-13 cells stably expressing the indicated constructs (G and H). –, empty vector; WT, wild-type FLAG-SCP4; 236–466, FLAG-SCP4236–466; D293A, catalytic mutant FLAG-SCP4D293A; IP, immunoprecipitation. Note: degradation bands appear in the WT and D293A input at ~50 kDa and in D293A at ~40 kDa.See also Figure S4.

Fig 3: SCP4 is an acquired dependency in human AML(A) Competition-based proliferation assays in 23 human cancer cell lines infected with the indicated sgRNAs. PDAC, pancreatic ductal adenocarcinoma; RMS, rhabdomyosarcoma. n = 3.(B) Western blot of whole-cell lysates from MOLM-13 cells on day 5 post-infection with the indicated sgRNAs.(C) Competition-based proliferation assay in MOLM-13 cells infected with the indicated sgRNAs. n = 3.(D) Western blot of whole-cell lysates from K562 cells on day 5 post-infection with the indicated sgRNAs.(E) Competition-based proliferation assay in K562 cells infected with the indicated sgRNAs. n = 3.(F) Quantification of the different cell-cycle stages in MOLM-13 cells on day 5 post-infection with the indicated sgRNAs. n = 3.(G) Bioluminescence imaging of NSG mice transplanted with luciferase+/Cas9+ MOLM-13 cells infected with either sgROSA or sgSCP4.(H) Quantification of bioluminescence intensity. n = 3. p value was calculated by unpaired Student’s t-test. *p < 0.05.(I) Western blot of whole-cell lysates from CD34+ cells electroporated with Cas9 loaded with the indicated sgRNAs on day 8 post-electroporation, with culturing in myeloid conditions.(J) qRT-PCR analysis of indels presence in CD34+ cells electroporated with Cas9 loaded with the indicated sgRNAs over the course of culturing in myeloid conditions. The effects of individual sgRNAs for SCP4 were averaged. n = 4.(K) Quantification of the flow cytometry analysis of myeloid differentiation of CD34+ cells electroporated with Cas9 loaded with the indicated sgRNAs on day 8 post-electroporation, with culturing in myeloid conditions. The effects of individual negative controls and sgRNAs for SCP4 were averaged. n = 4.All bar graphs represent the mean ± SEM. All sgRNA experiments were performed in Cas9-expressing cell lines. “e” refers to the exon number targeted by each sgRNA. The indicated sgRNAs were linked to GFP. ROSA, Mock, and NT, negative controls; PCNA and MYC, positive controls.See also Figure S2.

Fig 4: The catalytic phosphatase function of SCP4 is essential in AML(A) Relative evolutionary conservation score for each residue from 0 — least conserved to 1 — most conserved. Based on data from Chang et al. (2018).(B) Protein disorder prediction score for each residue from 0 — least disordered to 1 — most disordered. Based on data from Meszaros et al. (2018).(C) Running average of log2 fold changes of the CRISPR scan of SCP4 with all the possible sgRNAs. SCP4 protein amino acid numbers are indicated along the x axis.(D) Domain architectures of human SCP4 and the truncated version of SCP4 used in this study.(E) Western blot of whole-cell lysates from MOLM-13 cells stably expressing empty vector or CRISPR-resistant SCP4236–466.(F) Competition-based proliferation assay in MOLM-13 cells stably expressing empty vector or CRISPR-resistant SCP4236–466 infected with the indicated sgRNAs. n = 3.(G) Western blot of whole-cell lysates from MOLM-13 cells stably expressing empty vector (EV; underloaded) or CRISPR-resistant wild type (WT) and catalytic mutants of SCP4. Vertical black dashed lines indicate omitted lanes from the same gel and same exposure.(H) Competition-based proliferation assay in MOLM-13 cells stably expressing empty vector or CRISPR-resistant WT and catalytic mutants of SCP4 infected with the indicated sgRNAs. n = 3.All bar graphs represent the mean ± SEM. ROSA, negative control; PCNA, positive control. See also Figure S3 and Table S2.

Fig 5: STK35 and PDIK1L function redundantly in the same genetic pathway as SCP4(A and B) Competition-based proliferation assays in MOLM-13 (A) and K562 (B) cells co-infected with GFP-linked sgRNA and mCherry-linked sgRNA. Double mCherry+/GFP+ population depletion indicates loss of cell fitness due to genetic redundancy. n = 3.(C) Competition-based proliferation assay in MOLM-13 cells stably expressing EV, WT, or CRISPR-resistant (CR) HA-PDIK1L and HA-STK35 infected with the indicated sgRNAs. n = 3.(D) Western blot of whole-cell lysates from MOLM-13 cells stably expressing WT and CR HA-PDIK1L and HA-STK35 on day 5 post-infection with the indicated sgRNAs.(E) Summary of competition-based proliferation assays in the indicated cell lines. Plotted is the fold change (log2) of sgRNA+/GFP+ cells after 18 days in culture (average of triplicates).(F) Alignment of residues defining ATP binding site for human protein kinases. Residues that are conserved among the kinase sequences are highlighted in light gray. Dark gray indicates the conserved lysine at the ATP binding site. Consensus sequence given; uppercase = conserved, lowercase = nearly invariant. The predicted structural elements of protein kinases are indicated below the consensus sequence.(G) Western blot of whole-cell lysates from MOLM-13 cells stably expressing EV or CR WT and catalytic mutants of STK35 and PDIK1L in MOLM-13.(H) Competition-based proliferation assay in MOLM-13 cells stably expressing EV or CR WT and catalytic mutant versions of HA-STK35 and HA-PDIK1L infected with the indicated sgRNAs. n = 3.(I) Correlation matrix for the global log2 fold changes of gene expression relative to negative controls in each independent experiment across 15,095 genes. All bar graphs represent the mean ± SEM. “dg” refers to the bi-cistronic vector. ROSA, negative control; PCNA, positive control. See also Figure S5.