MUC1-C integrates aerobic glycolysis with suppression of oxidative phosphorylation in triple-negative breast cancer stem cells

Abstract

Activation of the MUC1-C protein promotes lineage plasticity, epigenetic reprogramming, and the cancer stem cell (CSC) state. The present studies performed on enriched populations of triple-negative breast cancer (TNBC) CSCs demonstrate that MUC1-C is essential for integrating activation of glycolytic pathway genes with self-renewal and tumorigenicity. MUC1-C further integrates the glycolytic pathway with suppression of mitochondrial DNA (mtDNA) genes encoding components of mitochondrial Complexes I-V. The repression of mtDNA genes is explained by MUC1-C-mediated (i) downregulation of the mitochondrial transcription factor A (TFAM) required for mtDNA transcription and (ii) induction of the mitochondrial transcription termination factor 3 (mTERF3). In support of pathogenesis that suppresses mitochondrial ROS production, targeting MUC1-C increases (i) mtDNA gene transcription, (ii) superoxide levels, and (iii) loss of self-renewal capacity. These findings and scRNA-seq analysis of CSC subpopulations indicate that MUC1-C regulates self-renewal and redox balance by integrating activation of glycolysis with suppression of oxidative phosphorylation.

Keywords: Cell biology; Molecular biology; Omics; Transcriptomics.

Graphical abstract

Figure 1

Serial passage of TNBC mammospheres selects for MUC1-C-dependent CSC populations (A and B) BT-549 cells growing as monolayers were seeded in mammosphere culture medium. After 10 days, the sphere 1 (S1) cells were isolated and reseeded for selection of S2 cells. Photomicrographs are shown for the serially passaged mammospheres up to S10. Scale bar: 100 μm. (A). The sphere forming efficiency (SFE) was determined by the percentage of cells that formed mammospheres as a function of the number of seeded cells. SFE is expressed as the % (mean ± SD of three determinations) (B). (C and D) MDA-MB-436 cells serially passaged as mammospheres are shown in the indicated S1-S10 photomicrographs. Scale bar: 100 μm. (C). SFE is expressed as the % (mean ± SD of three determinations) (D). (E) BT-549 2D and S8 mammosphere cells (10 × 10⁶) were implanted into left and right flanks, respectively, of NSG mice. Shown are tumors at 5 months after implantation. (F) BT-549/tet-MUC1shRNA 3D cells treated with vehicle or DOX for 7 days were analyzed for sphere formation. Shown are photomicrographs of representative mammospheres. Scale bar: 100 μm. (left). The relative SFE is expressed as the mean ± SD of three determinations as compared to that obtained for vehicle-treated cells (assigned a value of 1) (right). Asterisks represent  ∗∗∗p ≤ 0.001. (G) BT-549 3D cells were treated with vehicle or 2.5 μM GO-203 for 3 days. Shown are photomicrographs of representative mammospheres. Scale bar: 100 μm (left). The relative SFE is expressed as the mean ± SD of three determinations as compared to that obtained for vehicle-treated cells (assigned a value of 1) (right). Asterisks represent ∗∗∗p ≤ 0.001. (H) Lysates from BT-549 cells grown in 2D culture and as serially passage mammospheres (S1-S3) were immunoblotted with antibodies against the indicated proteins. (I) Lysates from BT-549/tet-MUC1shRNA 3D cells treated with vehicle or DOX for 10 days were immunoblotted with antibodies against the indicated proteins (left). Lysates from BT-549 3D cells were treated with vehicle or 2.5 μM GO-203 for 3 days were immunoblotted with antibodies against the indicated proteins (right). (J) NSG mice with established BT-549 3D cell tumors were treated intraperitoneally with PBS or GO-203 (12 μg/gm body weight) each day for 70 days. Tumor volumes are expressed as the mean ± SEM for 6 tumors.

Figure 2

MUC1-C is necessary for activation of glycolysis gene signatures in CSCs (A) Volcano plots showing downregulated (left) and upregulated (right) genes in BT-549 cells grown in S6 3D vs. 2D culture (left; 3070 DEGs; <0.1 padj (Benjamini-Hochberg adjusted p value), log2fold change (FC) > 1.5), with 1,369 upregulated and 1,701 downregulated genes). Volcano plots of downregulated (left) and upregulated (right) genes in BT-549/tet-MUC1shRNA 3D cells treated with DOX vs. vehicle for 10 days (right; 1678 DEGs; <0.1 padj (Benjamini-Hochberg adjusted p value), log2fold change ((FC) > 1.5), with 1,074 upregulated and 604 downregulated genes). (B) Overlap of DEGs upregulated in 3D BT-549 cells and downregulated in 3D BT-549/tet-MUC1 shRNA cells treated with DOX (left). The differential expression of these shared genes (red) is shown for the two conditions (right). Glycolytic genes are labeled within the shared set. (C) GSEA of genes in BT-549 cells grown in 3D vs. 2D culture (left) and BT-549/tet-MUC1shRNA 3D cells treated with DOX vs. vehicle control (right) using the HALLMARK_GLYCOLYSIS gene signature. NES: Normalized Enrichment Score. (D) GSEA lollipop plots of multiple glycolysis-related gene sets for (i) BT-549 cells grown in 3D vs. 2D culture and (ii) BT-549/tet-MUC1shRNA 3D cells treated with DOX vs. vehicle control. (E) Heatmaps of glycolytic genes in (i) BT-549 cells grown in 3D vs. 2D culture, (ii) BT-549/tet-MUC1shRNA 3D cells treated with DOX vs. vehicle, and (iii) BT-549 3D cells treated with GO-203 vs. vehicle. (F and G) MetaPhOR analysis identifying transcriptional dysregulation of metabolic pathways in 3D vs. 2D BT-549 cells (F) and 3D BT-549/tet-MUC1shRNA cells treated with DOX vs. vehicle (G).

Figure 3

MUC1-C is necessary for induction of GLUT1 and HK2 in TNBC CSCs (A) Lysates from 2D and S1-S3 cells were immunoblotted with antibodies against the indicated proteins. (B) Lysates from BT-549/tet-MUC1shRNA S6 3D cells treated with vehicle or DOX for 10 days (left) and BT-549 cells treated with vehicle or 2.5 μM GO-203 for 2 days (right) were immunoblotted with antibodies against the indicated proteins. (C) GSEA of genes in (i) BT-549/tet-MUC1shRNA S6 cells treated with DOX vs. vehicle and (ii) BT-549 cells treated with GO-203 vs. vehicle using the HALLMARK_MYC_TARGETS_V1 gene signature. (D) Lysates from BT-549/tet-MYCshRNA S6 3D cells treated with vehicle or DOX for 10 days were immunoblotted with antibodies against the indicated proteins. (E) Nuclear lysates from BT-549 2D and 3D cells were immunoprecipitated with anti-MUC1-C or a control IgG (left). Nuclear lysates from BT-549 3D cells treated with vehicle or 2.5 μM GO-203 for 3 days were immunoprecipitated with anti-MUC1-C or a control IgG (right). The precipitates were immunoblotted with antibodies against the indicated proteins. (F) Schema of GLUT1 with positioning of the PLS, pELS, and dELS regions. Soluble chromatin from BT-549 2D and 3D cells was precipitated with a control IgG or anti-MYC antibody. The DNA samples were amplified by qPCR with primers for the GLUT1 PLS, pELS, and dELS regions (Table S2). The results (mean ± SD of 3 determinations) are expressed as fold enrichment relative to that obtained with the IgG control (assigned a value of 1). Asterisks represent ∗p ≤ 0.05. (G) Schema of HK2 with positioning of the PLS, dELS1, and dELS2 regions. Soluble chromatin from BT-549 2D and 3D cells was precipitated with a control IgG or anti-MYC antibody. The DNA samples were amplified by qPCR with primers for the HK2 PLS, dELS1, and dELS2 regions (Table S2). The results (mean ± SD of 3 determinations) are expressed as fold enrichment relative to that obtained with the IgG control (assigned a value of 1). Asterisks represent ∗∗p ≤ 0.01. (H and I) Soluble chromatin from BT-549/tet-MUC1shRNA 3D cells treated with vehicle or DOX was precipitated with a control IgG or anti-MYC antibody. The DNA samples were amplified by qPCR with primers for the GLUT1 PLS and pELS regions (H), and HK2 PLS region (I). The results (mean ± SD of 3 determinations) are expressed as fold enrichment relative to that obtained with the IgG control (assigned a value of 1). Asterisks represent ∗p ≤ 0.05, ∗∗p ≤ 0.01. (J) Glucose uptake was measured in BT-549 3D cells treated with vehicle or 2.5 μM GO-203 for 3 days. The results (mean ± SD of 3 determinations) are expressed as luminescence (relative light unit, RLU). Asterisks represent ∗∗∗p ≤ 0.001. (K) BT-549 3D cells were treated with vehicle or 5 μM 2DG for 7 days. The relative SFE is expressed as the mean ± SD of three determinations as compared to that obtained for vehicle-treated cells (assigned a value of 1). Asterisks represent ∗p ≤ 0.05.

Figure 4

MUC1-C is necessary for expression of nuclear genes encoding components of the mitochondrial ETC (A) GSEA of genes in BT-549 cells grown in 3D vs. 2D culture (left) and BT-549/tet-MUC1shRNA 3D cells treated with DOX vs. vehicle control (right) using the WP_ETC_OXPHOS_MITOCHONDRIA gene signature. (B) WP_ETC_OXPHOS_SYSTEM IN MITOCHONDRIA signature heatmaps of ETC encoding genes in (i) BT-549 cells grown in 3D vs. 2D culture, (ii) BT-549/tet-MUC1shRNA 3D cells treated with DOX vs. vehicle, (iii) BT-549 3D cells treated with GO-203 vs. vehicle, and (iv) BT-549/tet-MYCshRNA 3D cells treated with DOX vs. vehicle. The row indicator shows gene origins, nuclear DNA (black) and mtDNA (yellow). (C) Expression of the indicated nuclear genes in BT-549/tet-MUC1shRNA 3D cells treated with DOX vs. vehicle for 7 days was determined by qRT-PCR. The results (mean ± SD of 3 determinations) are expressed as relative mRNA levels compared to that obtained for vehicle-treated cells (assigned a value of 1). Asterisks represent ∗p ≤ 0.05, ∗∗p ≤ 0.01, ∗∗∗p ≤ 0.001. (D–F) BT-549 3D vs. 2D cells and BT-549/tet-MUC1shRNA 3D cells treated with DOX vs. vehicle for 7 days were analyzed for SDHD (D) and CYCS (E) expression. The qRT-PCR results (mean ± SD of 3 determinations) are expressed as relative mRNA levels compared to that obtained for 2D cells or vehicle-treated cells (assigned a value of 1). Lysates were immunoblotted with antibodies against the indicated proteins (F). Asterisks represent ∗∗∗p ≤ 0.001, ∗∗∗∗p ≤ 0.0001. (G) Schema of SDHD with positioning of an E-box in the pELS region. Soluble chromatin from BT-549 2D and 3D cells was precipitated with a control IgG or anti-MYC antibody. The DNA samples were amplified by qPCR with primers for the SDHD pELS region (Table S2). The results (mean ± SD of 3 determinations) are expressed as fold enrichment relative to that obtained with the IgG control (assigned a value of 1). Asterisks represent ∗p ≤ 0.05. (H) Schema of CYCS with positioning of an E-box in the pELS region. Soluble chromatin from BT-549 2D and 3D cells was precipitated with a control IgG or anti-MYC antibody. The DNA samples were amplified by qPCR with primers for the CYCS PLS and pELS regions (Table S2). The results (mean ± SD of 3 determinations) are expressed as fold enrichment relative to that obtained with the IgG control (assigned a value of 1). Asterisks represent ∗p ≤ 0.05. (I and J) Soluble chromatin from BT-549/tet-MUC1shRNA 3D cells treated with vehicle or DOX for 5 days was precipitated with a control IgG or anti-MYC antibody. The DNA samples were amplified by qPCR with primers for the SDHD pELS (I) and HK2 pELS regions (J). The results (mean ± SD of 3 determinations) are expressed as fold enrichment relative to that obtained with the IgG control (assigned a value of 1). Asterisks represent ∗p ≤ 0.05, ∗∗∗p ≤ 0.001.

Figure 5

MUC1-C represses mtDNA gene expression (A) Expression of the indicated mtDNA genes in 3D vs. 2D BT-549 cells was determined by qRT-PCR. The results (mean ± SD of 3 determinations) are expressed as relative mRNA levels compared to that obtained for 2D cells (assigned a value of 1). Asterisks represent ∗p ≤ 0.05, ∗∗p ≤ 0.01, ∗∗∗p ≤ 0.001, ∗∗∗∗p ≤ 0.0001. (B and C) Expression of the indicated mtDNA genes in (B) BT-549/tet-MUC1shRNA 3D cells treated with DOX vs. vehicle for 10 days, and (C) BT-549 3D cells treated with 5 μM GO-203 vs. vehicle for 36 h was determined by qRT-PCR. The results (mean ± SD of 3 determinations) are expressed as relative mRNA levels compared to that obtained for vehicle-treated cells (assigned a value of 1). Asterisks represent ∗p ≤ 0.05, ∗∗p ≤ 0.01, ∗∗∗p ≤ 0.001, ∗∗∗∗p ≤ 0.0001. (D) Lysates from BT-549/tet-MUC1shRNA 3D cells treated with DOX vs. vehicle for 7 days (left) and BT-549 3D cells treated with 2.5 μM GO-203 vs. vehicle for 2 days (right) were immunoblotted with antibodies against the indicated proteins. (E) BT-549/tet-MUC1shRNA 3D cells treated with DOX vs. vehicle for 4 days (left) and BT-549 3D cells treated with 5 μM GO-203 vs. vehicle for 2 days (right) were analyzed for ATP levels. The results (mean ± SD of 3 determinations) are expressed as relative ATP levels compared to that obtained for vehicle-treated cells (assigned a value of 1). Asterisks represent ∗∗p ≤ 0.01, ∗∗∗∗p ≤ 0.0001. (F) BT-549/tet-MUC1shRNA 3D cells treated with DOX vs. vehicle for 5 days were analyzed for TFAM expression by qRT-PCR. The results (mean ± SD of 3 determinations) are expressed as relative mRNA levels compared to that obtained for vehicle-treated cells (assigned a value of 1). Asterisks represent ∗∗∗∗p ≤ 0.0001. (G) Lysates from BT-549/tet-MUC1shRNA 3D cells treated with DOX vs. vehicle for 7 days (left) and BT-549 3D cells treated with 2.5 μM GO-203 vs. vehicle for 2 days (right) were immunoblotted with antibodies against the indicated proteins. (H) Lysates from BT-549/tet-MYCshRNA 3D cells treated with vehicle or DOX were immunoblotted with antibodies against the indicated proteins. (I) BT-549/tet-MUC1shRNA (left) and BT-549/tet-MYCshRNA (right) 3D cells treated with vehicle or DOX were analyzed for mTERF3 expression by qRT-PCR. The results (mean ± SD of 3 determinations) are expressed as relative mRNA levels compared to that obtained for vehicle-treated cells (assigned a value of 1). Asterisks represent ∗∗p ≤ 0.01, ∗∗∗∗p ≤ 0.0001. (J) Lysates from BT-549/tet-MUC1shRNA and BT-549/tet-MYCshRNA 3D cells treated with vehicle or DOX for 7 days were immunoblotted with antibodies against the indicated proteins.

Figure 6

MUC1-C suppresses superoxide production in CSCs (A) BT-549/tet-MUC1shRNA mammospheres treated with vehicle or DOX for 10 days were stained with MitoSOX Red. Shown are representative fluorescence microscopy images. (B and C) BT-549/tet-MUC1shRNA (B) and MDA-MB-436/tet-MUC1shRNA (C) mammospheres were treated with vehicle or DOX for 10 days. MitoSOX Red flow cytometry data (mean ± SD of 3 determinations) are expressed as relative geometric mean fluorescence intensity (gMFI) compared to that obtained for vehicle-treated cells (assigned a value of 1)(left). Trypan blue staining results (mean ± SD of 3 determinations are expressed as the % cell death (right). (C) MitoSOX Red flow cytometry data in MDA-MB-436/tet-MUC1shRNA mammospheres treated with vehicle or DOX for 10 days. The results (mean ± SD of 3 determinations) are expressed as fold-change of gMFI for the mammosphere cells (left). MDA-MB-436/tet-MUC1shRNA mammospheres treated with vehicle or DOX for 10 days were monitored for cell death by trypan blue staining. The results are expressed as the % cell death (mean ± SD of 3 determinations)(right). Asterisks represent ∗p ≤ 0.05, ∗∗p ≤ 0.01, ∗∗∗p ≤ 0.001. (D) BT-549 mammospheres treated with vehicle or 5 μM GO-203 for 3 days were stained with MitoSOX Red. Shown are representative fluorescence microscopy images. (E and F) BT-549 (E) and MDA-MB-436 (F) mammospheres were treated with vehicle or 5 μM GO-203 for 3 days. MitoSOX Red flow cytometry data (mean ± SD of 3 determinations) are expressed as relative geometric mean fluorescence intensity (gMFI) compared to that obtained for vehicle-treated cells (assigned a value of 1). The results (mean ± SD of 3 determinations) are expressed as fold-change of gMFI for the mammophere cells. Asterisks represent ∗∗p ≤ 0.01, ∗∗∗p ≤ 0.001. (G) BT-549 mammospheres treated with vehicle or the indicated GO-203 concentrations for 3 days were stained with JC-1. Fluorescence images are shown for the control and GO-203-treated cells (upper panels). The JC-1 stained mammosphere cells were analyzed by flow cytometry for assessment of JC-1 red vs. green emission as a measure of the MMP (lower panels). The results (mean ± SD of 3 determinations) are expressed as the % depolarized mitochondria (JC-1 monomers). Asterisks represent ∗∗p ≤ 0.01. (H) MDA-MB-436 mammospheres treated with vehicle or the indicated GO-203 concentrations for 3 days were stained with JC-1. The JC-1 stained mammosphere cells were analyzed by flow cytometry for assessment of JC-1 red vs. green emission. The results (mean ± SD of 3 determinations) are expressed as the % depolarized mitochondria (JC-1 monomers). Asterisks represent ∗∗∗∗p ≤ 0.0001. (I and J) BT-549 (I) and MDA-MB-436 (J) mammospheres treated with vehicle or 5 μM GO-203 for 3 days were monitored for cell death by trypan blue staining. The results (mean ± SD of 3 determinations) are expressed as the % cell death. Asterisks represent ∗∗∗∗p ≤ 0.0001.

Figure 7

Single-cell RNA-seq identifies metabolic heterogeneity in mammosphere CSCs that associates with MUC1 expression (A) UMAP visualization of cells and clusters identified. (B) Expression feature plots of select genes associated with glycolysis (HK2, SLC2A1, and PGAM1) and transcriptional regulators of mitochondrial DNA (TFAM and MTERF3) visualized by UMAP representation reveal metabolic heterogeneity within passaged sphere cells. (C) Expression of glycolytic genes (HK2, SLC2A1, PFKL, and PGAM1), mitochondrial transcriptional regulators (TFAM and MTERF3), mitochondrial encoded components of ETC (MT-ND1 and MT-CO2), nuclear encoded components of ETC (SDHD and CYCS), and stemness genes (CD44 and ALDH2) associate with MUC1 status in passaged sphere cells. (D) Metabolic signature scores and MUC1 expression levels across clusters. (E) Comparison of single-cell HALLMARK_GLYCOLYSIS and HALLMARK_OXIDATIVE_PHOSPHORYLATION signature scores within individual cells. Color represents MUC1 expression within individual cells. (F) Correlation of HK2 and MUC1 expression across CTL or MUC1 knockdown passaged sphere cells with color intensity representing HALLMARK_GLYCOLYSIS signature scores. (G) Comparison of TFAM and MUC1 expression in CTL or MUC1 knockdown passaged sphere cells. Color represents overall signature of ETC genes encoded by the mitochondrial genome. (H) Correlation of MTERF3 expression and MUC1 expression with color intensity indicating HALLMARK_GLYCOLYSIS enrichment score.