Stearoyl‐CoA Desaturase‐1 Drives Tumor Growth by Interacting With Histone Deacetylase‐2 and Deacetylating Nucleophosmin‐1

Figures

• 

  FIGURE 1. Hypoxia regulates the expression of proteins involved in lipid desaturation and storage in cancer cells. (A and B) Western blot analysis of SCD1 expression in MDA‐MB231 in (A) and HT29 cancer cells in (B) incubated under normoxic or hypoxic conditions and after reoxygenation (Reox). The graphs show the results of the densitometry quantification of the Western blot results. HSC70 served as a loading control. (C) Western blot analysis of SCD1 and HIF1alpha expression in multiple cell lines incubated under normoxic or hypoxic conditions and after reoxygenation (Reox). The graphs show the results of the densitometry quantification of the Western blot results. HSC70 served as a loading control. (D) Pooled densitometric quantification of SCD1 (left) and HIF1α (right) expression by Western blot, normalized to HSC70, under normoxia, hypoxia, and after reoxygenation across all cell lines shown in (A–C). (E) Detection of lipid droplets (LDs) after Bodipy 493/503 staining of MDA‐MB231 and HT29 cancer cells incubated under normoxic or hypoxic conditions. (F) LDs were detected by immunofluorescence staining of PLIN2 in MDA‐MB231 and HT29 cancer cells incubated under normoxic or hypoxic…

• 

  FIGURE 2. PLIN2 and SCD1 inhibition reduces cancer cell proliferation both in vitro and in vivo. (A) Assessment of the migration of HT29 (left) and MDA‐MB231 (right) control cells (expressing the Cas9 enzyme and the non‐targeted gRNA control) and PLIN2‐KO cancer cells incubated under normoxic conditions for 24 h. (B) Assessment of the migration of HT29 (left) and MDA‐MB231 (right) control and PLIN2‐KO cancer cells under hypoxic conditions for 24 h. Scale bars = 100 µm. (C) Proliferation assessment of HT29 (left) and MDA‐MB231 (right) control and PLIN2‐KO cancer cells incubated under normoxic conditions for 24 h. (D) Assessment of the proliferation of HT29 (left) and MDA‐MB231 (right) control and PLIN2‐KO cancer cells incubated under hypoxic conditions for 24 h. (E) Proliferation assessment of the HT29 (left) and MDA‐MB231 (right) control and SCD‐KO cells incubated under normoxic conditions. (F) Tumor growth of HT29 control (Ctrl), PLIN2‐KO1 and PLIN2‐KO2 cells (left graph), and SCD‐KO cells (SCD‐KO1 and SCD‐KO2) (right graph) injected into RAG2−/− mice (n = 6); the HT29 control cells on the left and on the right are identical. (G and H) Growth of tumors formed from MDA‐MB231 control…

• 

  FIGURE 3. Decreased dependency on fatty acid oxidation (FAO) in PLIN2‐depleted cells. (A and B) Western blot of CPT1 in control and PLIN2‐KO HT29 cells (A) and MDA‐MB231 cells (B) incubated under normoxic and hypoxic conditions for 24 h. HSC70 served as a loading control. The results of the densitometry analysis of CPT1 were normalized to HSC70, and the values are shown below the blots. Control and PLIN2‐KO HT29 and MDA‐MB231 cells were preincubated under hypoxia for 24 h, and their metabolic activities were assessed. (C and D) Oxygen consumption rate (OCR) normalized capacity after incubation with the fatty acid oxidation inhibitor (etomoxir) or a combination of pyruvate and the glutamine inhibitors (UK5099/BPTES). (E) Fatty acid oxidation percentage representing capacity (Dependence + Flexibility), dependence (Blue) and flexibility (Orange) in control and PLIN2‐KO HT29 cells. (F and G) Normalized OCR data from dependence (F) or capacity (G) measurements in control and PLIN2‐KO MDA‐MB231 cells. (H) Fatty acid oxidation percentage representing capacity (Dependence + Flexibility), dependence (Blue) and flexibility (Orange) in MDA‐MB231 cells. (I and J) Normalized extracellular acidification…

• 

  FIGURE 4. Inhibition of PLIN2 or SCD1 affects lipid synthesis in cancer cells. (A and B) Western blot analysis of phACC and ACC expression in control and SCD‐KO HT29 cells (A) and in MDA‐MB231 cells (B). (C and D) Western blot analysis of phACC and ACC in control and PLIN2‐KO HT29 cells (C) and in MDA‐MB231 cells (D). HSC70 served as a loading control. The densitometry of phACC/total ACC ratios were normalized to the control, and the values are shown below the blots. (E) Mass spectrometry quantification of the acetyl‐CoA pool, presented as fold changes over control cells, in SCD‐KO and PLIN2‐KO HT29 and MDA‐MB231 cells. (F) Volcano plot of the proteins in acetylome scan analysis of control versus SCD1‐depleted MDA‐MB231 (SCD1‐KD) cells and according to the average ratio of three technical replicates and p values. (G) Graphical representation of pathways involving proteins that are differentially deacetylated by SCD1. (H) Volcano plot of the proteins identified by acetylome scan analysis in control versus PLIN2‐depleted MDA‐MB231 cells according to the average ratio of three technical replicates and p values. (I) Graphical representation of pathways involving proteins that are…

• 

  FIGURE 5. SCD interacts with NMP1 and increases its stabilization. (A) Coimmunoprecipitation of acetylated‐lysines (AcK) and detection of NPM1 by Western blot in control and SCD1‐depleted cells. (B) Western blot analysis of NPM1 in cell lysates of control and SCD1‐depleted cells, with densitometric quantification. (C) Stability assay for NPM1 by Western blot at different time points after treatment with cycloheximide (CHX) (100 µM) in control and SCD1‐depleted cells. HSC70 served as a loading control. (D) The interaction between SCD1 and NPM1 was assessed using Gaussia protein‐fragment complementation assays (gPCA). The normalized luminescence ratio (NLR) for the SCD1–NPM1 interaction is shown. The experiments were performed with n = 5 independent biological replicates. The p value corresponds to p < 0.01. (E) Proximity ligation assay images of HT29 and MDA‐MB231 cells. Red dots represent the proximity between NPM1 and SCD1. (F) Immunofluorescence staining of NPM1 in control, PLIN2‐KO and SCD‐KO MDA‐MB231 cells. NPM1 is shown in red, and DAPI (blue) indicates the nucleus. Scale bars = 10 µm. (G) Correlation analysis of SCD1 gene (SCD) expression with that of other genes revealed a…

• 

  FIGURE 6. SCD1 interacts with HDAC2. (A) Western blot analysis of HDAC1, 2, 3, 4 and 6 protein expression in control, PLIN2‐KO and SCD‐KO HT29 and MDA‐MB231 cells. HSC70 served as a loading control. (B) Immunoprecipitation (IP) of SCD1 and detection of HDAC2 by Western blot analysis in HT29 and MDA‐MB231 cells. (C) Reversed IP after IP of HDAC2 and detection of SCD1 by Western blot analysis in HT29 and MDA‐MB231 cells. (D) Correlation analysis of SCD1 gene (SCD) expression with that of other genes revealed a correlation score of 0.496 with that of HDAC2 in all cancers (p = 0.001005). (E and F) Scatter plot showing the relationships between the genome‐wide correlations of the coexpression of SCD1 (SCD) and HDAC2 in all human cancer tissues (E) and in breast cancer tissues (F). Displayed R values were determined by Pearson correlation. (G) Top differentially (left) and similarly (right) correlated pathways between SCD1 (SCD) and HDAC2 target genes in all cancer types. All experiments were reproduced at least three times (n = 3), and correlation analysis was performed using ARCHS4 v8 (Feb 2020), as shown in Figure 5.

• 

  FIGURE 7. SCD‐KO cells are sensitive to acetylation inhibitors. (A) Proliferation assay with Incucyte at the indicated time points in control, SCD‐KO HT29 and MDA‐MB231 cells treated with DMSO (Ctrl) or a deacetylase inhibitor cocktail. (B) Cell proliferation assessment after 96 h of treatment. (C) Proliferation of cells treated with DMSO (Ctrl) or vorinostat (0.75 µM). (D) Cell proliferation assessment of cells treated as described in (C) at 96 h. (E) Tumor uptake table of MC38 mouse colorectal tumors on Day 7 and Day 17 post‐injection in C57/Bl6 mice with different treatments. (F) Growth of MC38 tumors in C57/Bl6 mice and treated with vehicle, vorinostat, or an SCD inhibitor (SCD1i) alone or with the combination of vorinostat and anti‐SCDi. (G) Tumor weights of MC38 tumors. (H) Growth of E0771 tumors in C57BL/6 mice treated with vehicle, vorinostat, or an SCD inhibitor (SCD1i) alone or with a combination of vorinostat and SCD1i. (I) Growth of HT29 control (Ctrl) or PLIN2‐KO (PLIN2‐KO1) cells treated with vehicle or vorinostat. HT29 tumors were injected into RAG1−/− mice (n = 5). (J) Representative images of tumors from the experiment shown in (I). (K) Tumor weights of HT29 control and…