Stearoyl-CoA Desaturase 1 Protects Ovarian Cancer Cells from Ferroptotic Cell Death

Abstract

Activation of ferroptosis, a recently described mechanism of regulated cell death, dramatically inhibits growth of ovarian cancer cells. Given the importance of lipid metabolism in ferroptosis and the key role of lipids in ovarian cancer, we examined the contribution to ferroptosis of stearoyl-CoA desaturase (SCD1, SCD), an enzyme that catalyzes the rate-limiting step in monounsaturated fatty acid synthesis in ovarian cancer cells. SCD1 was highly expressed in ovarian cancer tissue, cell lines, and a genetic model of ovarian cancer stem cells. Inhibition of SCD1 induced lipid oxidation and cell death. Conversely, overexpression of SCD or exogenous administration of its C16:1 and C18:1 products, palmitoleic acid or oleate, protected cells from death. Inhibition of SCD1 induced both ferroptosis and apoptosis. Inhibition of SCD1 decreased CoQ₁₀, an endogenous membrane antioxidant whose depletion has been linked to ferroptosis, while concomitantly decreasing unsaturated fatty acyl chains in membrane phospholipids and increasing long-chain saturated ceramides, changes previously linked to apoptosis. Simultaneous triggering of two death pathways suggests SCD1 inhibition may be an effective component of antitumor therapy, because overcoming this dual mechanism of cell death may present a significant barrier to the emergence of drug resistance. Supporting this concept, we observed that inhibition of SCD1 significantly potentiated the antitumor effect of ferroptosis inducers in both ovarian cancer cell lines and a mouse orthotopic xenograft model. Our results suggest that the use of combined treatment with SCD1 inhibitors and ferroptosis inducers may provide a new therapeutic strategy for patients with ovarian cancer. SIGNIFICANCE: The combination of SCD1 inhibitors and ferroptosis inducers may provide a new therapeutic strategy for the treatment of ovarian cancer patients.See related commentary by Carbone and Melino, p. 5149.

Figure 1.. SCD1 is upregulated in ovarian cancer.

A. SCD1 mRNA in multiple ovarian cancer cell lines and normal ovarian cells assessed using RT-qPCR (FT-i- immortalized fallopian tube cells; FT-t- transformed fallopian tube cells; (cancer stem cells, see Basuli et. al., 2017) B. Western blot analysis of SCD1 in ovarian cells; N, non-malignant cell lines; C, ovarian cancer cell lines. Histological subtypes represented by these cell lines are as follows: adenocarcinoma, SW626, SKOV3; endometriod, MDAH2774, TOV112D; high grade serous ovarian cancer (HGSOC), COV362, OVCAR4, OVCAR8. C. Expression of SCD1 in GEO (GSE 109071). HGSOC compared to normal fallopian tube epithelium (FTEn). D. Immunohistochemical staining of SCD1 in patient tissues (scale bar =100 μm). Upper panel, representative image of normal ovary and HGSOC; lower panel, staining quantification (normal ovary n=6; HGSOC n=9). Two to three different images were captured per patient slide and quantified. Dot plot depicts means and standard deviations of 17 images of normal tissues and 23 images of HGSOC patient tissues. E. Differences in SCD1 expression in prognostic TCGA subtypes (MES, mesenchymal; IMR, immunoreactive; DIFF, differentiated; PRO, proliferative). Boxes represent first and third quartiles, line is median, whiskers indicate data spread, and outliers are represented by dots. Width of boxes is proportional to sample size. *p<0.05; **p<1.0 E-03; ***p=1.2E-E07

Figure 2.. Blocking SCD1 activity using small molecule inhibitors causes cell death that can be rescued by oleic acid and the ferroptosis inhibitor Fer-1.

Cells were treated with 1 μM MF-438 (MF) or 5 μM CAY10566 (CAY) in the presence or absence of 5 μM fer-1 or 80 μM oleic acid (OA) for 48 to 72 hours and cell viability assessed by calcein-AM (A,C, E); and cell death was assessed using Propidium iodide( B,D). F. C11-BODIPY staining of OVCAR-4 cells following treatment with 1 μM RSL3 in the presence or absence of 2 μM Fer-1 for 2 hours. Experiments are representative of three independent experiments. Shown are means and standard deviation of eight replicates in one representative experiment. *p<2 E-04; **p<6 E-05; ***p<4E-06

Figure 3.. Modulation of SCD1 affects ferroptosis in ovarian cancer cells.

A. COV362 and B. FT-t cells were transfected with siRNA targeted to SCD1 (siSCD1) or non-targeting siRNA (siNTC) for 72 hours in the presence and absence of 5 μM Fer-1 or 80 μM oleic acid (OA). C. Expression of SCD1 in Tet-on constructs analyzed using RT-qPCR and western-blotting (EV, empty vector). D. FT-t cells over-expressing SCD1 or empty vector controls were treated with 5 μM RSL3 in the presence or absence of 10 μM Fer-1 or 80 μM oleic acid (OA) for 24 hours and viability measured. Graphs are representative of four independent experiments. Shown are means and standard deviations of eight replicates in one representative experiment.; *p<2.5E-07; **p<5.3E-11.

Figure 4.. Over-expression of SCD1 protects cells from RSL3-mediated lipid peroxidation.

A. SCD1 mRNA in COV362 cells constitutively over-expressing SCD1 (SCD1) or empty vector control (EV) measured by RT-qPCR. B. Viability of COV362 cells over-expressing SCD1 or empty vector controls following treatment with 5 μM of RSL3 or 5μM erastin in the presence or absence of 10 μM Fer-1 for 48 hours. C. C11-BODIPY staining of COV362 SCD1 and COV362 EV cells following treatment with 5 μM RSL3 in the presence or absence of 2 μM Fer-1 for 4 hours. Blue=DAPI; green=C11 BODIPY. Scale bar =20 μm; D. C11-BODIPY staining analyzed by flow cytometry. E. Quantification of C11-BODIPY positive cells . Means and standard deviation of at least three replicas. *p<5.5E-05; **p<3.7E-9

Figure 5.. Effects of exogenous lipids, death pathway inhibitors, and SCD1 inhibitors on viability and caspase activity in ovarian cancer cells and ovarian cancer stem cells.

A-D. Viability of OVCAR-4 cells treated with 2μM RSL3 in the presence or absence of 80μM of the indicated lipids for 24 hours; E, F. Viability of FT-t cells treated with 80 μM stearic acid (SA) or palmitic acid (PA) in the presence or absence of 10 μM z-VADfmk or 5 μM Fer-1 for 24hrs; G. Viability of FT-t cells treated with 10 μM MF-438 in the presence or absence of 5 μM Fer-1, 10 μM z-VADfmk, or 10 μM necrostatin for 48 hrs. H. Relative caspase 3/7 activity in FT-t cells treated with 10 μM MF-438 for 48 hr or 1 μM staurosporin for 4 hours in the presence or absence of 10μM z-VADfmk. Data shown is representation of at least three independent experiments. Graphs are means and standard deviation of 8 replicas in one representative experiment. *p<0.024; **p<1 E-04; ***p<7E-14

Figure 6.. Inhibition of SCD1 modulates lipid content of ovarian cancer cells.

FT-t Cells were treated in triplicate with vehicle (control) or three chemically unrelated, commercially available SCD1 inhibitors: A939572 (A), CAY10566 (CAY), and MF-438 (MF), for 72hours prior to harvest for untargeted lipid analysis by UHPLC-MS. A. Ratio of saturated/monounsaturated fatty acids in all lipid species; B. Fraction of ceramides with saturated fatty acids; C. Fraction of phospholipids with unsaturated fatty acids; D. levels of coenzyme Q10. Means and standard deviations of triplicate determinations are shown. *p<0.04; **p<0.01

Figure 7.. Blocking the activity of SCD1 with small molecule inhibitors increases lipid peroxidation, ferroptotic cell death, and the anti-tumor effect of a ferroptosis inducer in vivo.

A-C. Viability of FT-t, COV362, or OVCAR-4 cells pretreated with 1μM MF-438 (MF), 1 μM CAY10566 (CAY), or 5 μM A939572 (A) for 24 hours followed by the addition of RSL3 for an additional 24 hours. For each cell type, dose of RSL3 was adjusted to attain modest cytotoxicity in the absence of SCD1 inhibitors (FT-t: 2 μM RSL3;OVCAR-4:1μM RSL3; COV362: 5 μM RSL3). Control cells were treated with SCD1 inhibitor for 48 hours or RSL3 for 24 hours. Data shown is representative of at least three independent experiments each with 8 to 12 replicas D. C11-BODIPY staining of COV362 cells treated with 1 μM MF-438 for 24 hrs followed by 1 μM RSL3 for 4 hrs in the presence or absence of Fer1. E. Viability of FT-t cells treated with A939572 (A) alone, 1μM erastin (ERA) alone, or A939572 followed by erastin. Graphs are representative of at least three independent experiments each with 3 to 8 replicas. F,G. Mice were injected intraperitoneally with FT-t cells and treated for 18 days with either vehicle control, A939572 (A), erastin (E) or the combination of A939572 and erastin. (F) depicts the number of tumors/mouse and (G) depicts total tumor mass/mouse. *p<0.04; **p<0.00027; ***p<2 E-05; ****p<7.2 E-10.