ASCT2/SLC1A5 controls glutamine uptake and tumour growth in triple-negative basal-like breast cancer

Abstract

Alanine, serine, cysteine-preferring transporter 2 (ASCT2; SLC1A5) mediates uptake of glutamine, a conditionally essential amino acid in rapidly proliferating tumour cells. Uptake of glutamine and subsequent glutaminolysis is critical for activation of the mTORC1 nutrient-sensing pathway, which regulates cell growth and protein translation in cancer cells. This is of particular interest in breast cancer, as glutamine dependence is increased in high-risk breast cancer subtypes. Pharmacological inhibitors of ASCT2-mediated transport significantly reduced glutamine uptake in human breast cancer cell lines, leading to the suppression of mTORC1 signalling, cell growth and cell cycle progression. Notably, these effects were subtype-dependent, with ASCT2 transport critical only for triple-negative (TN) basal-like breast cancer cell growth compared with minimal effects in luminal breast cancer cells. Both stable and inducible shRNA-mediated ASCT2 knockdown confirmed that inhibiting ASCT2 function was sufficient to prevent cellular proliferation and induce rapid cell death in TN basal-like breast cancer cells, but not in luminal cells. Using a bioluminescent orthotopic xenograft mouse model, ASCT2 expression was then shown to be necessary for both successful engraftment and growth of HCC1806 TN breast cancer cells in vivo. Lower tumoral expression of ASCT2 conferred a significant survival advantage in xenografted mice. These responses remained intact in primary breast cancers, where gene expression analysis showed high expression of ASCT2 and glutamine metabolism-related genes, including GLUL and GLS, in a cohort of 90 TN breast cancer patients, as well as correlations with the transcriptional regulators, MYC and ATF4. This study provides preclinical evidence for the feasibility of novel therapies exploiting ASCT2 transporter activity in breast cancer, particularly in the high-risk basal-like subgroup of TN breast cancer where there is not only high expression of ASCT2, but also a marked reliance on its activity for sustained cellular proliferation.

Figure 1

Inhibiting glutamine uptake represses cell growth in HCC1806 basal-like breast cancer cells but not luminal MCF-7 cells. (a) ASCT2 protein (60–75 kDa; Cell Signaling, Beverly, MA, USA; Cat. No. 8057 S) expression was measured by western blotting in normal mammary cells (MCF10A), two luminal (MCF-7, T47D) and six triple-negative breast cancer cell lines including basal-like (HCC1806, HCC1937, MDA-MB-468/MDA468) and claudin-low (CL; MDA-MB-231/MDA231) subsets, with GAPDH (37 kDa; Abcam, Cambridge, UK; Cat. No. ab8245) used as loading control. MCF-7 and MCF10A cells were a kind gift from the Vascular Biology Laboratory, Centenary Institute, Sydney, New South Wales, Australia, and identity confirmed using STR testing. T47D, HCC1806, HCC1937, MDA231 and MDA468 cells were obtained from ATCC, Manassas, VA, USA. HCC1500 and HCC1569 cells were a kind gift from the Tumour Progression Laboratory, Garvan Institute of Medical Research, Sydney, New South Wales, Australia. All cell lines were routinely confirmed to be mycoplasma-free using PCR testing. (b) ASCT2 localization was determined in breast cancer cell lines by immunofluorescence staining with primary antibody as in a, goat anti-rabbit AlexaFluor488 (Life Technologies, Carlsbad, CA, USA; Cat. No. A11034; green) as secondary antibody, and nuclei visualized using 4',6-diamidino-2-phenylindole (DAPI; blue). Scale bar, 30 μm. (c) Normal mammary cells (MCF10A; grey), luminal breast cancer cell lines (MCF-7, T47D; black), and triple-negative breast cancer cell lines (HCC1806, HCC1937, HCC1500, HCC1569, MDA231; red) were treated with l-γ-glutamyl-p-nitroanilide (GPNA; 1 mM; Sigma-Aldrich, St Louis, MO, USA) along with 0.3 mCi [³H]-l-glutamine (PerkinElmer, Waltham, MA, USA) for 15 min to assess uptake as previously described. Uptake data are presented as reduction in raw scintillation counts by GPNA compared with control, whereas MTT data are expressed as % inhibition compared with control. MCF-7 (d) and HCC1806 (e) cells were cultured in 96-well plates (MCF-7: 3 × 10³ cells/well; HCC1806: 1 × 10³ cells/well) for 3 days with or without 1 mM GPNA. MTT assays were conducted every 24 h to assess cell growth according to manufacturer's instructions (MTT Cell Growth Assay Kit; Merck-Millipore, Billerica, MA, USA). Cell cycle progression for MCF-7 (f) and HCC1806 (g) cells was analysed using APC-BrdU Flow Kit according to manufacturer's instructions (BD Pharmingen, San Jose, CA, USA) after 24 h culture with or without 1 mM GPNA. (h) MCF-7 and HCC1806 cells were incubated in GPNA or control media for 6 h. Total and phosphorylated (p-) p70S6K signalling protein (70, 85 kDa; both from Cell Signaling, Beverly, MA, USA; Cat. No. 9205S/9202S) was then detected by western blotting, with GAPDH (37 kDa) as loading control. (i) MCF-7 and HCC1806 cells were cultured in 96-well plates (MCF-7: 3 × 10³ cells/well; HCC1806: 1 × 10³ cells/well) in glutamine-free media (containing 10% dialysed FBS) for 3 days with MTT assays conducted every 24 h to assess cell growth according to manufacturer's instructions (MTT Cell Growth Assay Kit; Merck-Millipore, Billerica, MA, USA). Asterisks denote P-values as follows: *P⩽0.05; **P⩽0.01; ***P⩽0.001; ****P⩽0.0001, NS, not significant; two-way analysis of variance. Data in c–g and i represent mean±s.e.m., n⩾3 experiments. Data in a, b and h are representative of three independent experiments, with blots in H cropped to exclude extraneous lanes. Full blots are in Supplementary Figure 1.

Figure 2

ASCT2 expression is required for HCC1806 cell growth. MCF-7 (a) and HCC1806 (b) cells were transduced with a lentiviral vector (pLKO.1) containing control shRNA (shCont; shRNA sequence and specificity detailed previously) or one of two shRNAs against ASCT2 (shA28: 5′CCGGGCTGCTTATCCGCTTCTTCAACTCGAGTTGAAGAAGCGGATAAGCAGCTTTTTG-3′ or shA63: 5′-CCGGCTGGATTATGAGGAATGGATACTCGAGTATCCATTCCTCATAATCCAGTTTTTG-3′ both from Sigma-Aldrich, St Louis, MO, USA). ASCT2 expression (60–75 kDa) was measured by western blotting at 24, 48 and 72 h post transduction and after long-term puromycin (Puro) selection (a only). (c), MCF-7 cells (Puro) and HCC1806 cells (72 h) were incubated with [³H]-l-glutamine for 15 min as previously described to assess uptake after ASCT2 knockdown. Transduced MCF-7 (d) and HCC1806 (e) cells were cultured for 3 days post selection or post transduction, respectively. MTT assays were conducted every 24 h to assess the cell growth according to manufacturer's instructions (MTT Cell Growth Assay Kit; Merck-Millipore, Billerica, MA, USA). (f), Levels of full-length and cleaved poly ADP ribose polymerase (PARP) protein (116 and 89 kDa; Cell Signaling, Beverly, MA, USA; Cat. No. 9541 S) and LC3B-I/II protein forms (14, 16 kDa; Cell Signaling, Beverly, MA, USA; Cat. No. 2775 S) with GAPDH as loading control (37 kDa) were measured in HCC1806 cells by western blotting at 24, 48 and 72 h post transduction. MCF-7 (g) and HCC1806 (h) cells were transduced with lentiviral vector (pFH1tUTG) that encoded doxycycline (dox)-inducible expression of the control shRNA (shCont; sequence and specificity detailed previously) or the shA63 shRNA (shASCT2; shRNA sequence as above) also encoding an eGFP reporter tag. Transduced cells were sorted for GFP-positive cells using fluorescence-activated cell sorting and then shRNA expression was induced by treatment with 1 μg/ml dox. ASCT2 expression (60–75 kDa) was measured by western blotting at 24, 48 and 72 h ±dox with GAPDH as loading control (37 kDa), (i), Transduced MCF-7 and HCC1806 cells were incubated with 0.3 mCi [³H]-l-glutamine to assess uptake after 72 h dox treatment, as previously described. Transduced MCF-7 (j) and HCC1806 (k) cells were cultured with dox for 3 days and MTT assays were conducted every 24 h to assess the cell growth according to manufacturer's instructions (MTT Cell Growth Assay Kit; Merck-Millipore, Billerica, MA, USA). Asterisks denote P-values as follows: *P⩽0.05; **P⩽0.01; ***P⩽0.001; ****P⩽0.0001, NS, not significant; two-way analysis of variance. Data in c–e, i–k represent mean±s.e.m., n⩾3 experiments. Data in a, b, f–h are representative of three independent experiments. For MTT assays (e), stars under growth curves refer to shCont vs shA28 (top) and shCont vs shA63 (bottom).

Figure 3

ASCT2 knockdown in vivo represses tumour growth and improves survival. (a) HCC1806 cells expressing an additional mCherry-luciferase reporter construct (pHIV1SDm-CMV-mCherry-P2A-luc) and doxycycline-inducible shRNAs (see Figure 2h legend), shControl or shASCT2 (pFH1tUTG) were sorted for mCherry^hi/GFP⁺ cells using fluorescence-activated cell sorting and then 4 × 10⁶ cells suspended in a solution of 50% Matrigel in RPMI (100 μl) were implanted orthotopically into the left and right abdominal mammary glands of 8–12-week-old female athymic nu/nu mice (Animal Resource Centre, Perth, Western Australia, Australia), five mice per group from two separate experiments (n=10). Sample size was estimated on the basis of previous publications with no randomization or blinding. Doxycycline (200 μg/ml) was administered ad libitum in drinking water from Day 0 and tumour growth was measured twice weekly until individual shControl tumours reached ~1000 mm³, when all the mice were humanely euthanized in accordance with ethics approval protocol 2013/030 A from Sydney Local Health District Animal Welfare Committee. (b) Bioluminescence was detected twice weekly using a Xenogen IVIS Lumina II following intraperitoneal injection of 150 mg/kg d-luciferin. At end point, all the mice were euthanized and tumours collected for the measurement of size (c) and weight (d; error bars represent mean±s.d.). (e) Tumour RNA was extracted from shCont and shASCT2 (n=4 each) tumours using TRI Reagent, then reverse transcribed using SuperScript III (Life Technologies, Carlsbad, CA, USA) and random hexamers. Relative gene expression (GAPDH/ACTB) was measured by RT-qPCR using a custom Taqman Low-Density Array (Taqman, Life Technologies) run on a QuantStudio 12 K Flex Real-Time PCR System. Heat map of relative expression (compared with the first shControl lane) for each gene was generated in Gene-E (Broad Institute). Average fold-change between the shControl and shASCT2 groups are indicated to the right of the heat-map, asterisks indicate significant difference (P<0.05, n=4; Mann–Whitney U-test). (f), HCC1806-mCherry^hi/GFP⁺ cells (2 × 10⁶) were injected orthotopically as in a and allowed to establish (tumour size ~100 mm³) before the administration of doxycycline (200 μg/ml). Tumour size was measured twice weekly using digital calipers to calculate mean tumour volume per mouse (g) and individual mice were euthanized when they reached ethical end point (individual tumour size >1000 mm³; h). Data in b represents mean±s.e.m., n=9 (shCont; one mouse was euthanized owing to ulceration at tumour site and is not included in the analyses) or n=10 (shASCT2) mice (with two tumours per mouse), five mice per group in two independent experiments. Data in g represents mean, n=10 (shCont) or n=10 (shASCT2) mice (with two tumours per mouse), five mice per group in two independent experiments. Asterisks denote P-values as follows: *P⩽0.05; **P⩽0.01; ***P⩽0.001; ****P⩽0.0001, NS, not significant; two-way analysis of variance (b), Mann–Whitney U-test (d and e), or log-rank Mantel–Cox test (h). Data in c are representative of two independent experiments.

Figure 4

ASCT2 is highly expressed in human triple-negative breast cancer patient samples. RNA was extracted from 96 formalin-fixed paraffin-embedded triple-negative (TN) patient sample cores (AllPrep DNA/RNA FFPE Kit, Qiagen, Melbourne, Victoria, Australia) obtained from Royal Prince Alfred Hospital and Concord Repatriation General Hospital, Australia, in accordance with ethics approval from Sydney Local Health District Ethics Board with informed patient consent. Unamplified mRNA levels were measured directly using a custom probe set on the NanoString nCounter DX platform according to manufacturer's instructions and normalized to mean expression of six housekeeper genes (CLTC, GAPDH, GUSB, HPRT1, PGK1, TUBB). Six patient samples failed quality control checks and were excluded from subsequent analysis. Heat map (a) was generated in Gene-E (Broad Institute) and shows the global absolute gene expression (log₂, global min and max values shown) after manual clustering into TN breast cancer subtypes according to gene expression patterns previously described (UNS/BL1, unstable/basal-like 1; BL2, basal-like 2; IM, immunomodulatory; M, mesenchymal; MSL, mesenchymal stem-like; LAR, luminal androgen receptor). Log₂ gene expression of ASCT2/SLC1A5, glutamine metabolism-related genes (GLUL, GLS) and transcription factors, MYC and ATF4 (b) across TN subtypes in our patient cohort as defined in a. Correlation between ASCT2/SLC1A5 and transcriptional regulators, MYC and ATF4, was analysed for TN patient subgroups in our cohort (Supplementary Table 1), and subgroups that showed significant correlations (r>0.4, P<0.05; Pearson's correlation statistic; exact P-values shown inset) were grouped and the correlation shown in c. Black line shows the best-fit linear regression (SLC1A5 vs ATF4: r²=0.314; SLC1A5 vs MYC: r²=0.345).