Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2019 Aug 13;28(7):1860-1878.e9.
doi: 10.1016/j.celrep.2019.07.027.

p63 and SOX2 Dictate Glucose Reliance and Metabolic Vulnerabilities in Squamous Cell Carcinomas

Affiliations

p63 and SOX2 Dictate Glucose Reliance and Metabolic Vulnerabilities in Squamous Cell Carcinomas

Meng-Hsiung Hsieh et al. Cell Rep. .

Abstract

Squamous cell carcinoma (SCC), a malignancy arising across multiple anatomical sites, is responsible for significant cancer mortality due to insufficient therapeutic options. Here, we identify exceptional glucose reliance among SCCs dictated by hyperactive GLUT1-mediated glucose influx. Mechanistically, squamous lineage transcription factors p63 and SOX2 transactivate the intronic enhancer cluster of SLC2A1. Elevated glucose influx fuels generation of NADPH and GSH, thereby heightening the anti-oxidative capacity in SCC tumors. Systemic glucose restriction by ketogenic diet and inhibiting renal glucose reabsorption with SGLT2 inhibitor precipitate intratumoral oxidative stress and tumor growth inhibition. Furthermore, reduction of blood glucose lowers blood insulin levels, which suppresses PI3K/AKT signaling in SCC cells. Clinically, we demonstrate a robust correlation between blood glucose concentration and worse survival among SCC patients. Collectively, this study identifies the exceptional glucose reliance of SCC and suggests its candidacy as a highly vulnerable cancer type to be targeted by systemic glucose restriction.

Keywords: GLUT1; SGLT2; SOX2; glucose restriction; ketogenic diet; p63; squamous cell carcinoma.

PubMed Disclaimer

Conflict of interest statement

DECLARATION OF INTERESTS

R.J.D. is an advisor for Agios Pharmaceuticals. J.D.M. receives licensing fees for lung cancer cell lines from the NCI and University of Texas Southwestern Medical Center.

Figures

Figure 1.
Figure 1.. Enhanced GLUT1 Expression and Glycolytic Metabolism in SCC
(A) RNA sequencing (RNA-seq) analysis of GLUT1 mRNA expression among 35 tumor types. Each box represents the lower quartile, median, and upper quartile. Whiskers represent the 10th and 90th percentile of the data. Kruskal-Wallis nonparametric ANOVA. TPM, transcripts per million. (B) Representative IHC images (top) and quantification (bottom) of GLUT1 expression in human lung (n = 237), skin (n = 50), oral cavity (n = 43), cervix (n = 198), and esophagus (n = 54) SCC and non-SCC tumor tissue microarray (top). Scale bars, 1 mm. Each box represents the lower quartile, median, and upper quartile. Whiskers represent the 10th and 90th percentile of the data. Mann-Whitney U-test or one-way ANOVA. BCC, basal cell carcinoma; MEL, melanoma; MuCC, mucoepidermoid carcinoma of salivary gland; EnADC, endometrioid ADC; mSCC, metastatic SCC, HSE, hyperplasia of squamous epithelium; SCOC, small cell esophageal carcinoma. (C) H&E staining and IHC images of GLUT1 expression in human lung adenosquamous carcinoma tumor samples. Scale bar, 300 μm. (D) qRT-PCR analysis of GLUT1 mRNA expression in SCC and non-SCC cell lines (n = 3 for each cell line). (E) Immunoblot analysis of ΔNp63 and GLUT1 expression in SCC and non-SCC cell lines. (F) Representative heatmap depicting differential gene expression between the combined TCGA cohorts of NH, lung, cervical, and esophageal SCC (n = 1,372), bladder urothelial carcinoma (BLCA; n = 408), and all non-SCC (n = 7,752) tumors. Extended gene heatmap with clustering information is provided in Figure S2A. All error bars represent the mean ± SEM. ****p < 0.0001, ***p < 0.001, **p < 0.01, *p < 0.05. Two-tailed t test was used unless noted otherwise.
Figure 2.
Figure 2.. p63 Regulates GLUT1 Expression in SCC
(A) IF staining for GLUT1 (green) and p63 (red) in human SCC tissue samples. Scale bars, 300 μm. (B) qRT-PCR (left) and immunoblot analyses (right) of p63 and GLUT1 expression in shScr and shp63 lung SCC HCC2814 cells (n = 4). (C) Representative ICC images (left) and quantification (right) of p63 and GLUT1 expression in shScr or shp63 HCC2814 cells. (n = 3, 5–10 images were captured per group and normalized to nuclei for quantification). Scale bars, 100 μm. (D) qRT-PCR (left) and immunoblot analyses (right) of ΔNp63 and GLUT1 expression in shScr and shΔNp63 HCC2814 and KYSE70 (n = 4). (E) qRT-PCR (left) and immunoblot analyses (right) of ΔNp63 and GLUT1 expression in HCC2814 cells overexpressing EGFP or ΔNp63α (n = 3). (F) Quantification of 2-NBDG uptake in HCC2814 cells overexpressing EGFP or ΔNp63α (n = 3, 8–12 images were captured in each group for quantification). (G) Publicly available ChIP-seq alignment of p63 binding and H3K27ac on the SLC2A1 locus. p63 ChIP-seq was performed in HN SCC JHU-029 (GEO: GSE88859) and ENCODE histone mark ChIP-seq was performed in HeLa-S3 (GEO: GSM733684). Homer analysis (Heinz et al., 2010) identifies enriched p63 binding motifs in peak regions (E1–E3) of the SLC2A1 locus. (H) ChIP-PCR analysis for endogenous p63 and H3K27ac on the potential p63 binding regions in the intronic enhancer cluster of the SLC2A1 in HCC2814 and KYSE70. Values represent the average of triplicates ± SEM in a representative experiment. Data represent a minimum of two independent experiments. (I) Luciferase reporter assay measuring the transcriptional activity of individual enhancers E1, E2, and E3 in shScr or shΔNp63 HCC2814 and KYSE70 cells (n = 3). Luciferase signal is normalized to β-galactosidase activity. (J) GLUT1 and GLUT3 mRNA expression in CRISPR-Cas9-medated genome editing of the E2 p63-binding enhancer region (n = 3). All error bars represent the mean ± SEM. ****p < 0.0001, ***p < 0.001, **p < 0.01, *p < 0.05. Two-tailed t test was used unless noted otherwise.
Figure 3.
Figure 3.. p63/GLUT1 Enhances Anti-oxidative Power in SCC
(A) Schematic representation of uniformly labeled glucose-derived carbons in glucose metabolic pathways. (B and C) Fates of [U-13C] glucose-derived carbons in glycolysis, PPP, and de novo serine biosynthesis in lung SCC HCC95 and HCC2814 and lung ADC A549 (B) and shScr and shΔNp63 HCC2814 (C) cells. Relative 13C abundance of glucose and lactate in the culture media or intracellular glucose-6-phosphage (G6-P), ribose-5-phosphate (R5-P), and serine after 4 h of incubation with [U-13C] glucose were determined by gas chromatography-mass spectrometry (GC/MS). Values represent the average of triplicates ± SEM. Data represent a minimum of two independent experiments. (D) Cell viability of SCC and non-SCC cell lines cultured in increasing vitamin C concentration for 48 h (n = 4). Two-way ANOVA. (E) Increase in intracellular ROS levels measured by H2DCFDA staining in SCC and non-SCC cell lines treated with 1 mM vitamin C for 48 h (n = 3). (F and G) Relative intracellular ROS level by H2DCFDA (F) and DHE (G) staining in shScr, shp63, shΔNp63, and shGLUT1 HCC95 and KYSE70 cells (n = 4). (H and I) Relative intracellular NADPH/NADP+ ratio (H) and GSH/GSSG ratio (I) in shScr, shp63, shΔNp63, and shGLUT1 HCC2814 and KYSE70 cells (n = 3). (J) In vitro proliferation of shScr and shΔNp63 HCC2814 and KYSE70 cells (n = 4). Two-way ANOVA. (K) Soft agar colony formation assays of shScr and shΔNp63 HCC2814 and KYSE70 cells. Images are representative of three independent experiments. Number of colonies was analyzed after 21 days (n = 3). (L and M) In vitro proliferation (L) of shScr, shΔNp63, and shΔNp63 treated with NAC and intracellular ROS levels (M) measured by H2DCFDA staining in HCC2814 and KYSE70 cells (n = 3). Two-way ANOVA. (N and O) Tumor growth (left) and tumor weight (right) (N) and IHC analysis (O) of p63, GLUT1, Ki67, CC3, p-H2A.X, 4-HNE in Tet-inducible shScr (n = 5), shΔNp63 (n = 4), and shΔNp63 treated with NAC (10 g/L) (n = 4) HCC2814 xenograft tumors. Two-way ANOVA. Scale bars, 100 μm. ns, not significant. All error bars represent the mean ± SEM. ****p < 0.0001, ***p < 0.001, **p < 0.01, *p < 0.05. Two-tailed t test was used unless noted otherwise.
Figure 4.
Figure 4.. GLUT1 Rescues Oxidative Stress and Cell Death Induced by p63 Inhibition
(A and B) In vitro proliferation, qRT-PCR, and immunoblot analysis of ΔNp63, GLUT1 and V5-tag expression of shScr and shp63 lung SCC HCC2814 (A) and skin SCC A431 (B) cells ectopically overexpressing EGFP or GLUT1 (n = 3). Two-way ANOVA. (C–F) Relative glucose uptake (C), intracellular ROS (D), intracellular NADPH (E), and GSH/GSSG ratio (F) in shScr and shp63 HCC2814 (left) and A431 (right) ectopically overexpressing EGFP or GLUT1 (n = 3). (G and H) Tumor growth (left) and tumor weight (right) (G) and IHC analysis (H)of p63, GLUT1, Ki67, CC3, p-H2AX, 4-HNE in Tet-inducible shScr (n = 3), shΔNp63 (n = 4), and shΔNp63 overexpressing GLUT1 (n = 4) HCC2814 xenograft tumors. Two-way ANOVA. Scale bars, 100 μm. All error bars represent the mean ± SEM. ****p < 0.0001, ***p < 0.001, **p < 0.01, *p < 0.05. Two-tailed t test was used unless noted otherwise.
Figure 5.
Figure 5.. SOX2 Regulates GLUT1 Expression
(A and B) qRT-PCR (A) and immunoblot (B) analyses of SOX2 and GLUT1 expression in shScr and shSOX2 HCC2814 and KYSE70 cells (n = 3). (C) Quantification of 2-NBDG uptake (left) and extracellular lactate (right) in shScr and shSOX2 HCC2814 and KYSE70 cells (n = 3, 8–12 images were captured in each group for quantification). (D) CoIP analysis of the interaction between endogenous SOX2 and p63 in HCC2814 and KYSE70 cells. Immunoglobulin G (IgG) was used as a negative control. (E) ChIP-PCR analysis for endogenous SOX2 on potential p63 binding regions in the intronic enhancer cluster of the SLC2A1 gene in HCC2814 cells. Values represent the average of triplicates ± SEM in a representative experiment. Data represent a minimum of two independent experiments. (F) Analysis on publicly available ChIP-seq of SOX2 (red bars) and p63 (blue bars) occupancy in the SLC2A1 intronic enhancer cluster in esophageal SCC lines KYSE70 and TT (Watanabe et al., 2014) and lung SCC line HCC95 (GEO: GSE46837). (G) ChIP-PCR analysis for endogenous SOX2 on potential p63 binding regions in the intronic enhancer cluster of the SLC2A1 gene in shScr, shΔNp63, and shSOX2 HCC2814 cells. Values represent the average of triplicates ± SEM in a representative experiment. Data represent a minimum of two independent experiments. (H–J) Relative intracellular NADPH/NADP+ ratio (H), GSH/GSSH ratio (I), and intracellular ROS (J) in shScr and shSOX2 HCC2814 and KYSE70 cells (n = 3). (K) In vitro proliferation of shScr and shSOX2 HCC2814 and KYSE70 cells (n = 3). Two-way ANOVA. (L) Soft agar colony formation assays of shScr and shSOX2 HCC2814 and KYSE70 cells. Images are representative of three independent experiments. Number of colonies was analyzed after 21 days (n = 3). (M) In vitro proliferation of shScr and shSOX2 HCC2814 cells ectopically overexpressing EGFP or GLUT1 (n = 3). Two-way ANOVA. (N) qRT-PCR (left) and immunoblot (right) analysis of SOX2, GLUT1 and V5-tag expression in shScr and shSOX2 HCC2814 cells ectopically overexpressing EGFP or GLUT1 (n = 3). (O and P) Relative 2-NBDG uptake (O), intracellular ROS levels (P) in shScr, and shSOX2 HCC2814 cells ectopically overexpressing EGFP or GLUT1 (n = 3). All error bars represent the mean ± SEM. ****p < 0.0001, ***p < 0.001, **p < 0.01, *p < 0.05. Two-tailed t test was used unless noted otherwise.
Figure 6.
Figure 6.. Ketogenic Diet Suppresses SCC Growth In Vivo
(A and B) Xenograft tumor growth, tumor weight, and blood glucose and plasma insulin levels of lung SCC HCC2814 (NC, n = 6; KD, n = 8) (A) and lung ADC A549 (NC, n = 4; KD, n = 5) (B) fed with normal chow (NC) as control or ketogenic diet (KD). Two-way ANOVA. (C) IHC analysis (top) and quantification (bottom) of Ki67, CC3, p-H2AX, 4-HNE, p-IR, p-AKT, p-S6, and p-4EBP in NC (HCC2814, n = 6; A549, n = 4)- and KD (HCC2814, n = 8; A549, n = 5)-fed xenograft tumors. A total of 5–0 images in each tumor were captured and analyzed for quantification. Scale bars, 100 μm. (D) In vitro proliferation (left) and immunoblot analysis (right) of p-IR, IR, p-AKT, AKT, p-S6, S6, p-4EBP, and 4-EBP expression of HCC2814 cells treated with insulin (0–10 ng/mL) (n = 3). Two-way ANOVA. (E) Tumor growth (left) and tumor weight (right) of HCC2814 xenograft tumors treated with NC alone (NC, n = 5) as control, NC with cisplatin (NC+cisplatin, n = 7), KD alone (KD, n = 6), and KD with cisplatin (KD+cisplatin, n = 8). Two-way ANOVA. All error bars represent the mean ± SEM. ****p < 0.0001, ***p < 0.001, **p < 0.01, *p < 0.05. Two-tailed t test was used unless noted otherwise.
Figure 7.
Figure 7.. Dietary, Pharmacological, and Genetic Glucose Restriction Suppresses KLLuc SCC Tumor Development
(A and B) Representative thyroid transcription factor-1 (TTF-1; ADC marker) and CK5 (SCC marker) IF images (A) and quantification of SCC, adenosquamous, and ADC tumor types determined by histopathological as well as IHC evaluation of TTF-1/CK5 (B) in KLLuc mice fed with normal chow (NC, n = 11), ketogenic diet (KD, n = 7), or canagliflozin (CAG, n = 6). Chi-square test. Scale bar, 2.5 mm. (C) Total tumor burden of KLLuc mice analyzed by in vivo bioluminescence analysis at 11 weeks post intratracheal injection of adenovirus-Cre (NC, n = 11; KD, n = 7; CAG, n = 6). (D) Survival analysis of KLLuc mice fed with NC (n = 11), KD (n = 7), or CAG (n = 6). (E) Blood glucose levels in KLLuc mice fed with NC (n = 11), KD (n = 7), or CAG (n = 6). Two-way ANOVA. (F) Plasma insulin concentration in KLLuc mice fed with NC (n = 11), KD (n = 7), or CAG (n = 6). (G) Representative IHC images (top) and quantification of % area (right) of Ki67, CC3, p-H2AX, 4-HNE, p-IR, p-AKT, p-S6, and p-4EBP in KLLuc SCC tumors fed with NC (n = 11), KD (n = 7), or CAG (n = 6). A total of 5–10 images in each tumor were captured and analyzed for quantification. Scale bars, 100 μm. (H–J) Representative TTF-1 and CK5 IF images(H), quantification of individual tumor types (I), and total tumor burden determined by histological analysis of H&E-stained tumor tissues (J) in wild type (LSL-KrasG12D; Lkb1flox/flox; LSL-Luc, WT, n = 7) and GLUT1 knockout (LSL-KrasG12D; Lkb1flox/flox; LSL-Luc; GLUT1flox/flox, GLUT1-KO, n = 4) KLLuc mice. Chi-square test. Scale bar, 2.5 mm. (K) Comparison of individual SCC tumor size of WT (n = 7) and GLUT1-KO (n = 4) KLLuc mice. (L–N) Kaplan-Meier survival analysis comparing high and low random blood glucose (RGB) levels in the esophageal SCC (n = 65) (L), lung SCC (n = 127) (M), and lung ADC (n = 120) (N) patient cohorts. High and low RGB groups were separated by 120 mg/dL. Significance was determined with the log-rank test. All error bars represent the mean ± SEM. ****p < 0.0001, ***p < 0.001, **p < 0.01, *p < 0.05. Two-tailed t test was used unless noted otherwise.

References

    1. Allen BG, Bhatia SK, Buatti JM, Brandt KE, Lindholm KE, Button AM, Szweda LI, Smith BJ, Spitz DR, and Fath MA (2013). Ketogenic diets enhance oxidative stress and radio-chemo-therapy responses in lung cancer xenografts. Clin. Cancer Res. 19, 3905–3913. - PMC - PubMed
    1. Allen BG, Bhatia SK, Anderson CM, Eichenberger-Gilmore JM, Sibenaller ZA, Mapuskar KA, Schoenfeld JD, Buatti JM, Spitz DR, and Fath MA (2014). Ketogenic diets as an adjuvant cancer therapy: History and potential mechanism. Redox Biol. 2, 963–970. - PMC - PubMed
    1. Bao X, Rubin AJ, Qu K, Zhang J, Giresi PG, Chang HY, and Khavari PA (2015). A novel ATAC-seq approach reveals lineage-specific reinforcement of the open chromatin landscape via cooperation between BAF and p63. Genome Biol. 16, 284. - PMC - PubMed
    1. Bardeesy N, Sinha M, Hezel AF, Signoretti S, Hathaway NA, Sharpless NE, Loda M, Carrasco DR, and DePinho RA (2002). Loss of the Lkb1 tumour suppressor provokes intestinal polyposis but resistance to transformation. Nature 419, 162–167. - PubMed
    1. Bass AJ, Watanabe H, Mermel CH, Yu S, Perner S, Verhaak RG, Kim SY, Wardwell L, Tamayo P, Gat-Viks I, et al. (2009). SOX2 is an amplified lineage-survival oncogene in lung and esophageal squamous cell carcinomas. Nat. Genet 41, 1238–1242. - PMC - PubMed

Publication types

MeSH terms