Targeting the Warburg effect in cancer cells through ENO1 knockdown rescues oxidative phosphorylation and induces growth arrest

In the last 5 years, novel knowledge on tumor metabolism has been revealed with the identification of critical factors that fuel tumors. Alpha-enolase (ENO1) is commonly over-expressed in tumors and is a clinically relevant candidate molecular target for immunotherapy. Here, we silenced ENO1 in human cancer cell lines and evaluated its impact through proteomic, biochemical and functional approaches. ENO1 silencing increased reactive oxygen species that were mainly generated through the sorbitol and NADPH oxidase pathways, as well as autophagy and catabolic pathway adaptations, which together affect cancer cell growth and induce senescence. These findings represent the first comprehensive metabolic analysis following ENO1 silencing. Inhibition of ENO1, either alone, or in combination with other pathways which were perturbed by ENO1 silencing, opens novel avenues for future therapeutic approaches.


Supplementary Figure S2: The polyol and pentose phosphate pathways increase the concentration of intracellular reactive oxygen species (ROS) in ENO1-silenced cells. (A) Heat map of differentially-expressed proteins in shENO1 compared
to shCTRL CFPAC-1 cells. Based on the spectra count label-free quantitation approach, LC-MS/MS analysis identified 32 up-regulated (red) and 28 down-regulated (green) proteins. Proteins are classified according to their function: cell cycle/signaling, protein transport, protein biosynthesis, cell adhesion/migration and others/unknown. See also Figure 1A for other identified proteins. (B-C) Analysis of aldose reductase (ALDR) activity measured as the rate of NADPH oxidation (B), and NADPH oxidase activity assessed by the isoluminolchemiluminescence assay (C) in shCTRL and shENO1 PT45 and T3M4 cell lines. Chemiluminescence was expressed as relative luminescence unit (RLU)/mg cell proteins. (D-E) Analysis of ROS concentration measured by DCFDA-AM assay (D) and of [1-14 C] glucose flux through the Pentose Phosphate Pathway (PPP) assessed through 14 CO 2 release (E) after selective inhibition of mitochondrial chain (rotenone), NADPH oxidase (apocynin) and ALDR (zopolrestat) in shCTRL and shENO1 PT45 (left panels) and T3M4 (right panels) cell lines. (F) Analysis of PPP activity, as described above, after selective inhibition of the PPP (DHEA) in CFPAC-1 (left panels), MDA-MB-231 (middle panel) and NCI-H441 (right panel) cell lines. All the graphs illustrate the mean result of three independent experiments ± SEM *p < 0.05; **p < 0.01;***p < 0.001 relative to shCTRL.

Supplementary Figure S5: ENO1 knockdown promotes oxidative phosphorylation. (A-C) Analysis of glutaminase (GLS)
(A), glutamine amidophosphoribosyltransferase (GPAT) (B) and carbamoyl phosphate synthetase II (CPSII) (C) activity in shCTRL and shENO1 PT45 and T3M4 cell lines. (D) Analysis of mitochondrial respiratory chain complex I and complexes II-IV activity in shCTRL and shENO1 PT45 and T3M4 cells, expressed as nmol NAD + /min/mg mitochondrial protein for complex I, nmol Cyt c reduced/min/mg mitochondrial proteins for complexes II-III and nmol Cyt c oxidized/min/mg mitochondrial protein for complex IV. (E) Analysis of ATP production in shCTRL and shENO1 PT45 and T3M4 cell lines. All the graphs illustrate the mean result of three independent experiments ± SEM. *p < 0.05; **p < 0.01;***p < 0.001 relative to shCTRL. (F) Transcript analysis of proteins involved in catabolic pathways/ mitochondrial biogenesis in CFPAC-1 (red), PT45 (orange), T3M4 (purple), MDA-MB-231 (blue) and NCI-H441 (green) shENO1 compared to shCTRL (white) cells. Results are the mean ± SEM of triplicates. *p < 0.05; **p < 0.01;***p < 0.001 relative to shCTRLtransduced cells. and T3M4 (right panel) after shCTRL or shENO1 infection. Cells were starved and counted every 24 hr after serum replenishment. ***p < 0.001 was assessed by two-way ANOVA with Sidak's post hoc test. (B) Cell survival assessed by MTT assay. Cells were starved and MTT solution was added 48 hr after serum replenishment. OD values were measured at 570 nm. (C) Colony forming assay in soft agar. Cells were plated in 0.45% agarose overlaying a 0.9% agar layer. Colonies were counted by optical microscope after 3 weeks. T3M4 cell line was unable to form colonies in soft agar. (D-E) Flow cytometry cell cycle analysis of serum-starved shCTRL and shENO1 NCI-H441 (D) and MDA-MB-231 (E) cells at the indicated time points after serum replenishment. Data are expressed as the percentage of cells at each phase. (F) Senescence-associated β-galactosidase staining. Senescent CFPAC-1 and T3M4 cells were colored blue upon X-gal staining at pH 6. Parental and shCTRL CFPAC-1 cells showed positive β-galactosidase staining even in the absence of a senescent morphology. One representative out of three independent experiments is shown. Results are mean of three independent experiments ± SEM. *p < 0.05, **p < 0.01, ***p < 0.001 relative to shCTRL.

Supplementary
(A) Analysis of aldose reductase (ALDR) activity measured through the rate of NADPH oxidation in shCTRL (white bars) and shENO1 (black bars) cell lines after 24 h treatment with NAC or TROLOX-C. (B) Oxidation of palmitic acid in CFPAC-1, MDA-MB-231 and NCI-H441 cell lines transduced with shCTRL (white bars) or shENO1 (black bars) after 24 h treatment with antioxidants. Results are mean of three independent experiments ± SEM. *p < 0.05, **p < 0.01, ***p < 0.001 relative to untreated cells.  Figure 6G) for the indicated time.

Viral Transduction
Lentiviral infections were performed using HEK293T cells as producers of viral supernatants. HEK-293T cells were grown on 10-cm plates to 70% confluence and co-transfected with shRNA lentiviral DNA and the helper vectors pCMV∆8.74 and pVSV-G (Clonetech) using the calcium phosphate precipitation method. The medium was harvested 24 hr post-transfection and filtered through a 0.45 μm filter. Supernatants from HEK-293T cultures were used to cross-transduce cancer cells in the presence of 8 μg/mL Polybrene (Sigma-Aldrich), and subsequent clones were selected by puromycin (2 μg/mL).
shRNAs Short-hairpin RNAs (shRNAs) targeting the human ENO1 mRNA and cloned into the pLKO.1-puro vector were obtained from the human library MISSION® TRC-Hs 1.0. Five shRNAs targeting ENO1 were purchased as glycerol stocks from Sigma-Aldrich (SHCLNG-NM_001428). The efficiency of each shRNA in decreasing the ENO1 mRNA level was assessed by realtime qRT-PCR in transiently transfected cells. The best shRNA (TRCN0000029324) with the targeted sequence 5′-CCGGCGTACCGCTTCCTTAGAACTTCTCGAGA-AGTTCTAAGGAAGCGGTACGTTTTT-3′ was chosen for the generation of stable cell lines. The control shRNA (SHC002, Sigma-Aldrich) generates a scrambled shRNA sequence that does not target any known human gene.

Quantitative RT-PCR
Total RNA was extracted using the RNeasy Mini kit (Qiagen) and reverse transcription was performed from 2 µg of total RNA using iScript cDNA synthesis kit (BioRad) according to the manufacturer's instructions. Quantitative RT-PCR was performed using SYBR Green dye (Life Technologies) on a Thermal iCycler (BioRad). PCR reactions were performed in triplicate and the relative amount of cDNA was calculated by the comparative C T method using β-actin RNA sequences as a control. Data are represented as mean ± SEM of three or more independent experiments.

Western blot analysis
After SDS-PAGE, proteins were transferred to Hybond-N Nitrocellulose membrane (GE Healthcare Bio-Sciences). Membranes were blocked in Trisbuffered saline (TBS) containing 5% non-fat dry milk and 0.1% Tween 20 (TBS-T), before incubation with the primary antibody overnight at 4°C. Membranes were then washed with TBS-T followed by exposure to the appropriate horseradish peroxidase-conjugated secondary antibody for 1 hr, and then visualized using the enhanced chemiluminescence (ECL) detection system (GE Healthcare Bio-Sciences) by ProXPRESS 2D (PerkinElmer) scanning. The following antibodies were

Trypsin digestion and desalting
Cell pellets were resuspended in 200 µL of 8 M urea, and the protein concentration was measured by the Bradford Assay (BioRad). Proteins were transferred to a 1.5-mL eppendorf tube, reduced by 10 mM dithiothreitol (DTT) for 30 min at 37°C, and then alkylated by 50 mM iodoacetamide for 20 min at room temperature. The concentrated urea in the sample was diluted to a final concentration of 2 M, and the proteins were digested by trypsin at 37°C for 6 hr in a buffer containing ammonium bicarbonate (50 mM, pH 9). The digestion mixture was then acidified by adding glacial acetic acid to a final concentration of 2% and desalted by ZipTip (Millipore).

Tandem mass spectrometry analysis
Peptides were analyzed by highly sensitive reversed-phase liquid chromatography coupled nanospray tandem mass spectrometry (LC-MS/MS) using an LTQ-Orbitrap mass spectrometer (Thermo Fisher) (2). Briefly, the reversed-phase LC column was slurry-packed inhouse with 5 μm, 200 Å pore size C 18 resin (Michrom BioResources) in a 100 µm i.d. × 10 cm long piece of fused silica capillary (Polymicro Technologies) with a laser-pulled tip. After sample injection, the column was washed for 5 min with mobile phase A (0.1% formic acid), and peptides were eluted using a linear gradient of 0% mobile phase B (0.1% formic acid, 80% acetonitrile) to 50% B in 160 min at 200 nL/min, then to 100% B in an additional 10 min for proteomics analysis. The LTQ-Orbitrap mass spectrometer was operated in a datadependent mode in which each full MS scan (60,000 resolving power) was followed by eight MS/MS scans where the eight most abundant molecular ions were dynamically selected and fragmented by collision-induced dissociation (CID) using a normalized collision energy of 35%. The Dynamic Exclusion Time was 30 s, and the Dynamic Exclusion Size was 200. The "FT master scan preview mode", "Charge state screening", "Monoisotopic precursor selection", and "Charge state rejection" were enabled so that only the 1 + , 2 + , and 3 + ions were selected and fragmented by CID.

Mass spectrometry data analysis
Tandem mass spectra collected by Xcalibur (version 2.0.2) were searched against the NCBI human protein database using SEQUEST (Bioworks software from ThermoFisher, version 3.3.1) with full tryptic cleavage constraints, static cysteine alkylation by iodoacetamide, and variable methionine oxidation. Mass tolerance for precursor ions was 5 ppm and mass tolerance for fragment ions was 0.25 Da. The SEQUEST search results of proteomics data were filtered by the criteria "Xcorr versus charge 1.9, 2.2, 3.0 for 1 + , 2 + , 3 + ions; ∆Cn > 0.1; probability of randomized identification of peptide < 0.01". Confident peptide identifications were determined using these stringent filter criteria for database match scoring followed by manual evaluation of the results. The "false discovery rate (FDR)" was estimated by searching a combined forward-reversed database as described by Elias (3). The SEQUEST search results were exported to Excel files and compared. Spectra count label-free quantitation analysis was performed applying the formula: Relative difference (%) = [(Total no. peptides in shENO1)-(Total no. peptides in shCTRL)]/ [(Total no. peptides in shENO1 + Total no. peptides in shCTRL)/2]*100.

Glucose uptake
The uptake of glucose was measured by radiolabeling cells with 0.3 µCi/ml 2-deoxy-D-[ 3 H]glucose, as described earlier (4) insulin increases cyclic GMP production by inducing nitric oxide (NO). The non carrier-mediated glucose uptake was measuring by performing the assay in the presence of 10 µM cytochalasin B, a strong inhibitor of the facilitated glucose uptake. Results were expressed as pmol 2-deoxy-D-[ 3 H]glucose/mg cell proteins.

Lactate
Analysis of lactate level was performed on 2 × 10 6 cells with the L-Lactate Assay Kit (Abcam, Cambridge, MA, USA), following the manufacturer's instructions. Results were expressed as pmol/10 6 cells.

Aldose reductase activity
Cells were washed with PBS, detached by gentle scraping, centrifuged at 13,000 × g for 5 min at 4°C, re-suspended in 0.4 mL of 50 mmol/L Na 3 PO 4 buffer (pH 7.0). A 50 μL aliquot was sonicated and used for determining the cell proteins. The remaining sample was transferred to a 96-well plate, in the presence of 10 mmol/L glucose, dissolved in Na 3 PO 4 buffer. 1.5 mmol/L NADPH was added. The rate of NADPH oxidation was followed for 6 min, monitoring the absorbance at 340 nm with a Packard microplate reader EL340 (Bio-Tek Instruments). Results were expressed as nmol NADP + produced/min/mg cell proteins.

NADPH oxidase activity
NADPH oxidase activity was carried out in vitro by the isoluminol-chemiluminescence assay. A total of 2 × 10 6 cells was washed with PBS, detached with trypsin/EDTA (0.05/0.02% v/v), re-suspended in 1 mL PBS containing 7.5 mmol/L glucose, 0.9 mmol/L CaCl 2 , 0.5 mM MgCl 2 , and incubated for 10 min at 37°C, in the presence of 200 ng/mL phorbol myristate acetate (PMA) to activate the enzyme, 125 μmol/L isoluminol, 25 U/mL horseradish peroxidase. A 50 μL aliquot was sonicated and used for determining the cell proteins. Chemiluminescence of each sample derived from superoxide and lucigenin was detected using a Synergy HT microplate reader (Bio-Tek Instruments) and was expressed as relative luminescence unit (RLU)/ mg cell proteins. For the negative control, in each experiment, one sample was treated with 450 U/mL superoxide dismutase (SOD): the detected chemiluminescence was < 5% of the corresponding sample in the absence of SOD (not shown).

Tyrosine catabolism
Determination of phenylalanine and acetoacetate levels was performed according to the instruction manuals of EnzyChrom TM phenylalanine and ketone body assay kits (BioAssay Systems).

Glutamine catabolism
Glutamine catabolism was measured as reported (5), with minor modifications. Cells were washed with PBS, detached by gentle scraping, centrifuged at 13,000 × g for 5 min at 4°C, re-suspended in 250 µL of buffer A (150 mmol/L KH 2 PO 4 , 63 mmol/L Tris/HCl, 0.25 mmol/L EDTA; pH 8.6) and sonicated. The intracellular protein content was measured using the BCA kit (Sigma-Aldrich). A volume of 100 µL of the whole cell lysates was incubated for 30 min at 37°C in a quartz cuvette, in the presence of 50 µL of 20 mmol/L L-glutamine and 850 µL of buffer B (80 mmol/L Tris/HCl, 20 mmol/L NAD + , 20 mmol/L ADP, 3% v/v H 2 O 2 ; pH 9.4). The absorbance of NADH was monitored at 340 nm using a Lambda 3 spectrophotometer (PerkinElmer). The kinetics was linear throughout the assay. The results were expressed as µmol NADH/min/mg cell proteins, and were considered as an index of the activity of glutaminase (GLS) plus L-glutamic dehydrogenase. In a second series of samples, 20 µL of the GLS inhibitor bis-2-(5-phenylacetamido-1,3,4-thiadiazol-2-yl)ethyl sulfide BTPES (30 µmol/L) was added after 15 min. This concentration was chosen as it produced 100% inhibition of glutaminase activity in our system (not shown). The absorbance of NADH was monitored for 15 min as described previously. The results, considered as an index of the activity of L-glutamic dehydrogenase, were expressed as µmol NADH/min/mg cell proteins. GLS activity was obtained by subtracting the rate of the second assay from the rate of the first one.
Results were expressed as pmol carbamoyl aspartate/min/ mg cell proteins.

Measurement of mitochondrial respiratory chain
Cells were washed twice in ice-cold PBS, then lysed in 0.5 mL buffer A (50 mmol/L Tris, 100 mmol/L KCl, 5 mmol/L MgCl 2 , 1.8 mmol/L ATP, 1 mmol/L EDTA, pH 7.2), supplemented with protease inhibitor cocktail III [100 mmol/L AEBSF, 80 mmol/L aprotinin, 5 mmol/L bestatin, 1.5 mmol/L E-64, 2 mmol/L leupeptin and 1 mmol/L pepstatin (MerckMillipore) 1 mmol/L PMSF, 250 mmol/L NaF. Samples were clarified by centrifuging at 650 × g for 3 min at 4°C, and the supernatant was collected and centrifuged at 13,000 x g for 5 min at 4°C. The new supernatant was discarded, the pellet containing mitochondria was washed in 0.5 mL buffer A and re-suspended in 0.25 mL buffer B (250 mmol/L sucrose, 15 μmol/L K 2 HPO 4 , 2 mmol/L MgCl 2 , 0.5 mmol/L EDTA, 5% w/v bovine serum albumin). A 50 µL aliquot was sonicated and used for the measurement of protein content. The activity of mitochondria respiration complexes was measured according to (7). Results were expressed as nmol NAD + /min/mg mitochondrial protein for complex I, nmol cyt c reduced/min/mg mitochondrial protein for complexes II-III, nmol cyt c oxidized/min/mg mitochondrial protein for complexes IV.

Enolase inhibitor studies
For enolase inhibitor studies, PhAH lithium salt was custom-synthesized by CAGE chemicals. PhAH was dissolved in PBS as a 50 mM stock and stored at −20°C until use.

Enolase enzymatic activity assay
Enolase activity was measured by coupling the reactions of enolase, pyruvate kinase and lactate dehydrogenase, according to (8). After PhAH treatment, cells were washed with PBS, detached by gentle scraping, centrifuged at 13,000 × g for 5 min at 4°C, re-suspended in 0.25 mL of 100 mmol/L Tris buffer (pH 8.0), 10 mmol/L MgCl 2 , 100 mmol/L KCl, 1 mmol/L 2-phosphoglyceric acid, 0.4 mmol/L ADP, 6.8 U/mL pyruvate kinase, 9.9 U/mL lactate dehydrogenase. 0.2 mmol/L NADH was added. The rate of NADH oxidation was followed for 6 min, monitoring the absorbance at 340 nm with a Packard microplate reader EL340 (Bio-Tek Instruments). Results were expressed as nmol NAD + produced/min/mg cell proteins.

Proliferation assay
Cells were seeded in 6-well plates at a density of 1 × 10 5 cells/well. After 24 hr of serum starvation, cells were grown in complete medium and then detached, stained with trypan blue, and counted every 24 hr. Results represent at least three independent experiments. For antioxidant treatment, PT45 cells were cultured for 7 days with NAC (5 or 10 mM, Sigma-Aldrich) or with TROLOX-C (100 or 250 µM, Sigma-Aldrich), adding fresh medium daily. Results represent at least three independent experiments.

Cell survival assay
Cell survival was assessed by MTT assay. Cells were seeded in 96-well plate at 2 × 10 3 cells/well and serum starved for 24 hr. After 48 hr of culture in complete medium, 20 μL of 5 mg/μL MTT solution (Sigma-Aldrich) was added to the medium and incubated at 37°C for a further 4 hr. Medium was removed and the insoluble formazan product was dissolved in 200 µl DMSO (Sigma-Aldrich) for 10 min at room temperature. OD values were measured at 570 nm in an ELISA microtiter plate reader (BioRad). Results represent at least three independent experiments.

Colony forming assay
Briefly, a bottom layer of 2.5 mL culture medium containing 0.9% agarose type VII (Sigma-Aldrich) was initially solidified in a 6-well culture plate. Then, 2 mL of 0.45% agarose solution containing 3 × 10 4 cells was layered on top of each well. Cells were fed twice a week with complete DMEM and incubated at 37°C for 3 weeks. Colonies were counted by optical microscope. Results represent at least three independent experiments.