Fig 2: CRISPR gene tiling scans of INO80 complex members in HCC.(A) Schematic outline of high-density CRISPR gene body scan in Cas9-expressing HepG2 and U87 cells. (B to F) Two-dimensional annotation of CRISPR gene tiling scans for (B) ACTR5, (C) INO80, (D) MCRS1, (E) ACTR8, and (F) YY1. The solid lines indicate the smoothened model of the CRISPR scan score derived from individual sgRNAs (dots) screened in HepG2 (red) and U87 (blue) cells. The median CRISPR scan scores of the positive control (dotted line; defined as −1.0) and negative control (defined as 0.0) sgRNAs are designated. (G) Three-dimensional annotation of ACTR5 CRISPR scan score relative to the AlphaFold structural model of ACTR5 (ID, Q9H9F9). (H) Effect of wild-type (WT)- and ΔA5-ACTR5 expression on the growth competition assay of HepG2-dCas9-Krab cells transduced with sgiACTR5 (n = 3 each group). (I) Silver stain of the TST-purified WT- and ΔA5-ACTR5 protein complexes. The top candidate of each protein band was suggested by LC-MS/MS. (J) Western blot of ACTR5 and IES6 in the TST-purified WT- and ΔA5-ACTR5 protein complexes. *P < 0.01 two-sided Student’s t test.

Fig 3: IES6 stabilizes ACTR5 to support CDK6 signaling in HCC.(A) Growth competition assay of HepG2-dCas9-Krab cells transduced with RFP-labeled nontargeting control (gray lines; n = 2 independent sgiNTC sequences) and IES6-targeting sgiRNAs (blue lines; n = 4 independent sgiIES6 sequences). (B) Two-dimensional annotation of CRISPR gene tiling scans for IES6 in HepG2 (red) and U87 (blue) cells. The solid lines indicate the smoothened model of the CRISPR scan score derived from individual sgRNAs (dots). The median CRISPR scan scores of the positive control (dotted line; defined as −1.0) and negative control (defined as 0.0) sgRNAs are designated. (C) Three-dimensional annotation of IES6 CRISPR scan score relative to the AlphaFold structural model of IES6 (ID, Q6PI98). (D) Western blot of IES6 and ACTR5 in the TST-purified WT-, ΔI1-, and ΔI2-IES6 protein complexes. (E) Effect of WT- and ΔI2-IES6 expression on the growth competition assay of HepG2-dCas9-Krab cells transduced with sgiIES6 (n = 3 each group). (F) Western blot of IES6, ACTR5, CDKN2A, CDK6, Rb, p-Rb, E2F1, and β-actin in HepG2-dCas9-Krab cells transduced with sgiNTC and sgiIES6. (G and H) CellTiter-Glo analysis of the (G) sgiACTR5- and (H) sgiIES6-transduced HepG2-dCas9-Krab cells incubated with dimethyl sulfoxide (DMSO) or ribociclib (n = 4 each group). (I) Model of the ACTR5/IES6 complex supporting CDK6-driven cell proliferation. Data are presented as means ± SEM. *P < 0.01 by two-sided Student’s t test.

Fig 4: CRISPRi screen identifies the essential role of ACTR5 in HCC.(A) Schematic outline of an epigenetic-focused CRISPRi screen in HepG2-dCas9-Krab cells. (B) Volcano plot depicts the change of sgRNA abundance [x axis; log2 (fold change)] and significance [y axis; −log10 (false discovery rate)] of each gene during the 24-day epigenetics CRISPRi screen. (C) Violin plot indicates the median (red lines) and quartiles (blue lines) of the gene dependency score (z score) over 22 HCC cell lines (dots) in the DepMap genome-wide CRISPR screen 20Q2 database. (D) Growth competition assay of HepG2-dCas9-Krab cells transduced with RFP-labeled nontargeting control (gray lines; n = 2 independent sgiNTC sequences) and ACTR5-targeting sgiRNAs (red lines; n = 4 independent sgiACTR5 sequences). (E) Box-whiskers plot of the growth competition assay in nine Cas9-expressing cancer cell models transduced with sgACTR5 (dots; n = 4 independent sgACTR5 sequences). (F) Survival curves of liver HCC patients with high versus low ACTR5 expression. Source: GEPIA database (http://gepia.cancer-pku.cn). (G) Profile plot of tumor volume (in cubic millimeters) and (H) box-whiskers plot of tumor weight (in grams) of the HCC xenograft tumors in NSG mice inoculated with sgiNTC and sgiACTR5-transduced HepG2-dCas9-Krab cells (n = 12 tumor sites per group). Box-whiskers indicate the first and third quartiles (boxes) and the range (whiskers). *P < 0.01 by two-sided Student’s t test.

Fig 5: ACTR5 controls the CDKN2A and CDK6 cell cycle signaling in HCC.(A and C) Western blot of ACTR5, CDKN2A, CDK6, Rb, p-Rb, E2F1, and β-actin in (A) dCas9-Krab–expressing HCC (HepG2 and SNU475) and glioblastoma (U251 and U87) cells, and (C) vector versus ACTR5-TST–expressing HepG2-dCas9-Krab cells transduced with sgiNTC and sgiACTR5. (B) Cell cycle monitored by 5-Ethynyl-2′-deoxyuridine incorporation in HepG2-dCas9-Krab cells transduced with sgiNTC and sgiACTR5 (n = 3). (D) Growth competition assay of vector versus ACTR5-TST–expressing HepG2-dCas9-Krab cells transduced with RFP-labeled sgiNTC and sgiACTR5 (n = 3 each group). (E) Effect of targeting ACTR5 on CDKN2A- and CDK6-triggered cell cycle signaling. Data are presented as means ± SEM. *P < 0.01 by two-sided Student’s t test.