Targeting Squalene Epoxidase Confers Metabolic Vulnerability and Overcomes Chemoresistance in HNSCC

Abstract Cisplatin resistance poses a substantial hurdle in effectively treating head and neck squamous cell carcinoma (HNSCC). Utilizing multiple tumor models and examining an internal HNSCC cohort, squalene epoxidase (SQLE) is pinpointed as a key driver of chemoresistance and tumorigenesis, operating through a cholesterol‐dependent pathway. Comprehensive transcriptomic analysis reveals that SQLE is essential for maintaining c‐Myc transcriptional activity by stabilizing the c‐Myc protein and averting its ubiquitin‐mediated degradation. Mechanistic investigation demonstrates that SQLE inhibition diminishes Akt's binding affinity to lipid rafts via a cholesterol‐dependent process, subsequently deactivating lipid raft‐localized Akt, reducing GSK‐3β phosphorylation at S9, and increasing c‐Myc phosphorylation at T58, ultimately leading to c‐Myc destabilization. Importantly, employing an Sqle conditional knockout mouse model, SQLE's critical role in HNSCC initiation and progression is established. The preclinical findings demonstrate the potent synergistic effects of combining terbinafine and cisplatin in arresting tumor growth. These discoveries not only provide novel insights into the underlying mechanisms of SQLE‐mediated cisplatin resistance and tumorigenesis in HNSCC but also propose a promising therapeutic avenue for HNSCC patients unresponsive to conventional cisplatin‐based chemotherapy.

ANTs across TCGA HNSCC, GSE127165, and GSE37991 datasets.d) SQLE expression patterns in normal tissues, dysplasia tissues, and tumor tissues within the GSE30784 dataset.e) SQLE expression levels in normal tissues, margin tissues, and tumor tissues in the GSE31056 dataset.f-h) Kaplan-Meier analyses of overall survival or disease-free survival for HNSCC patients stratified by SQLE expression levels in TCGA HNSCC and GSE41613 datasets; group differences assessed using the log-rank test.i) Kaplan-Meier analysis of oral cancer-free survival time in patients with oral preneoplastic lesions, grouped based on SQLE expression in the GSE26549 dataset; group differences assessed using the log-rank test.j) IHC scores of SQLE in advanced-stage tumor tissues compared to early-stage tumor tissues in our in-house cohort (n=102).k) Univariate analysis of risk factors for predicting overall survival in HNSCC patients (n=102).Data was presented as mean ± SD. ns (not significant), ***p < 0.001, ****p < 0.0001 for Student's t-test and one-way ANOVA test.

Figure S1 .
Figure S1.Overexpression of SQLE in cisplatin-resistant cell lines and tumor tissues.a) Western blot analysis comparing SQLE expression levels in cisplatin-resistant HNSCC cell lines to their parental counterparts.b,c) Western blot analysis of SQLE expression in cisplatinsensitive (n=48) and cisplatin-resistant (n=48) tumor tissues.Data was presented as mean ± SD. ***p < 0.001 for Student's t-test.

Figure S2 .
Figure S2.Association of SQLE overexpression with unfavorable clinical outcomes in multiple independent HNSCC cohorts.a-c) Comparison of SQLE expression levels in tumor tissues and ANTs across TCGA HNSCC, GSE127165, and GSE37991 datasets.d) SQLE expression patterns in normal tissues, dysplasia tissues, and tumor tissues within the GSE30784 dataset.e)SQLE expression levels in normal tissues, margin tissues, and tumor tissues in the GSE31056 dataset.f-h) Kaplan-Meier analyses of overall survival or disease-free survival for HNSCC patients stratified by SQLE expression levels in TCGA HNSCC and GSE41613 datasets; group differences assessed using the log-rank test.i) Kaplan-Meier analysis of oral cancer-free survival time in patients with oral preneoplastic lesions, grouped based on SQLE expression in the GSE26549 dataset; group differences assessed using the log-rank test.j) IHC scores of SQLE in advanced-stage tumor tissues compared to early-stage tumor tissues in our in-house cohort (n=102).k) Univariate analysis of risk factors for predicting overall survival in HNSCC patients (n=102).Data was presented as mean ± SD. ns (not significant), ***p < 0.001, ****p < 0.0001 for Student's t-test and one-way ANOVA test.

Figure S3 .
Figure S3.Clinical implications of SQLE expression in HPV-positive and HPV-negative HNSCC subtypes, and its relationship with treatment responses and survival outcomes.a,b) SQLE staining intensity in HPV-positive (n=8) and HPV-negative (n=12) HNSCC tissues before and after cisplatin treatment.c,d) Association between SQLE staining intensity and overall survival in HPV-positive (n=21) and HPV-negative HNSCC (n=81) subgroups.e)Relationship between SQLE intensity and therapeutic responses in the HPV-negative HNSCC cohort (n=81).f) Multivariate analysis identifying independent risk factors for the HPVnegative HNSCC cohort (n=81).g) Relationship between SQLE intensity and therapeutic responses in the HPV-positive HNSCC cohort (n=21).h) Multivariate analysis identifying independent risk factors for the HPV-positive HNSCC cohort (n=21).Data was presented as mean ± SD. ns (not significant), *p < 0.05, **p < 0. 01, ***p < 0.001 for Student's t-test and one-way ANOVA test.

Figure S5 .
Figure S5.SQLE depletion enhanced cisplatin sensitivity in HNSCC cells in vivo.a) IHC scores of Ki-67 and CD44 in xenograft tumor tissues formed by SCC-1 cisR cells subjected to the indicated modifications (n=6).b-d) Effects of SQLE depletion on tumor size, weight, and volume in SCC-23 cisR cells with or without cisplatin treatment.Mice received either vehicle control or cisplatin (5 mg/kg intraperitoneally, twice per week) starting 1 week after tumor cell injection (n=6).e,f) IHC scores of Ki-67 and CD44 in xenograft tumor tissues formed by SCC-23 cisR cells under the specified treatments (n=6).Scale bar: 100 μm.Data was presented as mean ± SD. ***p < 0.001 for one-way ANOVA test.

Figure S7 .
Figure S7.SQLE depletion diminished cancer stemness of HNSCC cells in vitro.a,b) Representative ALDEFLUOR™ assays and FACS analysis of control or SQLE-depleted SCC-1 cisR and SCC-23 cisR cells, with or without cisplatin treatment (n=4).c) Impact of SQLE knockdown on the sphere-forming ability of SCC-1 cisR and SCC-23 cisR cells in the presence or absence of cisplatin (n=4).Scale bar: 100 μm.Data was presented as mean ± SD. ***p < 0.001 for one-way ANOVA test.

Figure S8 .
Figure S8.SQLE overexpression promoted tumor initiation capacity of primary HNSCC cells in vivo.a) In vivo limiting dilution analysis of EpCAM + ALDH high tumor cells from Pt. 1, infected with control and SQLE OV lentiviruses (n=10).The frequency of cancer cells capable of initiating tumors at each indicated cell dose was presented.b) Differences in CSC frequencies were calculated using the ELDA software.

Figure S10 .
Figure S10.Strong association between SQLE expression and c-Myc target gene signatures in HNSCC.a) Top enriched signatures in the shCTRL group compared to the shSQLE group, as identified by GSEA analysis.b,c) The c-Myc target signatures MYC_TARGETS_V1 and/or MYC_TARGETS_V2 are enriched in the SQLE-high expression group across multiple HNSCC cohorts.

Figure S11 .
Figure S11.Influence of SQLE modifications on PTEN expression in SCC-1 cisR and SCC-23 cisR cells.a) PTEN expression levels in the specified cell lines.b,c) Effects of SQLE depletion or overexpression on PTEN expression in both SCC-1 cisR and SCC-23 cisR cells.

Figure S12 .
Figure S12.Western blotting analysis of the expression of Gia2 and Flot2 in the indicated fractions isolated with the sucrose gradient ultracentrifugation.

Figure S13 .
Figure S13.Effects of SQLE depletion on c-Myc stability and raft-localized Akt in human dental stem cells.a,b) Impact of SQLE depletion or overexpression on c-Myc protein expression in hDPSCs and hPDLSCs.c,d) Influence of SQLE depletion or overexpression on c-Myc mRNA expression in hDPSCs and hPDLSCs (n=4).e,f) CHX analysis illustrating the effect of SQLE depletion on c-Myc stability in hDPSCs and hPDLSCs.g,h) Impact of SQLE inhibition on the expression of raft-localized Akt in hDPSCs and hPDLSCs.Data was presented as mean ± SD. ns (not significant) for Student's t-test and one-way ANOVA test.

Figure S14 .
Figure S14.Western blot analysis of c-Myc expression in cisplatin-resistant cell lines and their parental cisplatin-sensitive counterparts.

Figure
Figure S15.c-Myc depletion re-sensitizes cisplatin-resistant HNSCC cells to cisplatin and plays a critical role in regulating cancer stemness in HNSCC.a,b) Assessment of colony formation potential in SCC-1 cisR and SCC-23 cisR cells upon c-Myc depletion, with or without cisplatin treatment (n=4).c-Myc knockdown efficiency was evaluated by western blotting.c) Apoptosis rates in SCC-1 cisR and SCC-23 cisR cells following c-Myc depletion, in the presence or absence of cisplatin (n=4).d) Cell viability assessment in SCC-1 cisR and SCC-23 cisR cells subjected to c-Myc depletion and exposed to a gradient of cisplatin concentrations (n=4).e,f) Analysis of c-Myc mRNA and protein expression in ALDH high and ALDH low cells isolated from six HNSCC patients (n=4).g) Western blotting examination of the impact of c-Myc depletion on CD44, BMI-1, SOX2, and KIF-4 expression in ALDH high and ALDH low primary HNSCC cells, obtained from HNSCC patient case #1 (Pt. 1) and case #2 (Pt.2).Data was presented as mean ± SD. ***p < 0.001 for one-way ANOVA test.

Figure
Figure S16.c-Myc serves as a functional downstream target of SQLE.a-c) Effects of c-Myc knockdown on tumor growth in SQLE-overexpressing SCC-23 cisR cells (n=6).d) IHC scores of Ki-67 and CD44 in xenograft tumor tissues subjected to the indicated treatments (n=6).Scale bar: 100 μm.Data was presented as mean ± SD. ***p < 0.001 for one-way ANOVA test.

Figure
Figure S17.c-Myc functions as a downstream target of SQLE, critical for maintaining cancer stemness in HNSCC.a) In vivo limiting dilution analysis of SCC-1 cells subjected to indicated treatments (n=6), demonstrating the frequency of tumor cells capable of initiating xenografts at each injected cell dose.b,c) Differences in CSC frequencies between the indicated groups were calculated using ELDA software.

Figure S18 .
Figure S18.Generation of Sqle cKO mouse model.a,b) Schematic diagram illustrating the generation of the Sqle cKO mouse model.c) Western blot analysis of SQLE expression in tongue tissue from Sqle CTRL and Sqle cKO mice.

Figure S19 .
Figure S19.Terbinafine suppressed the tumorigenesis of HNSCC cells by targeting SQLE.ac) Tumor size, weight, and volume in the indicated treatment groups (n=6).d) SQLE expression in terbinafine-treated xenografts.e) SQLE expression in HNSCC cells treated with terbinafine compared to DMSO-treated control cells.f-g) Expression levels of KSR1, p-AMPK, and AMPK in HNSCC cells treated with terbinafine compared to DMSO-treated control cells.Data was presented as mean ± SD. ns (not significant), ***p < 0.001 for one-way ANOVA test.