A2780 were obtained from the European Collection of Cell Cultures (ECACC, Porton Down, UK), IGROV-1 and U251 were from the National Cancer Institute (NCI, Bethesda, MD, USA), and all other cell lines were from American Type Culture Collection (ATCC, Manassas, VA, USA). The cell lines were authenticated as previously described . All cell lines were maintained in a 37°C 5% CO2-95% humidified air incubator. Breast cancer cell lines were cultured in DMEM (Life Technologies, Carlsbad, CA, USA) supplemented with 5% FBS (Sigma-Aldrich, St. Louis, MO, USA). All other cell lines were maintained in R10 medium (Roswell Park Memorial Institute medium (RPMI)-1640 (Life Technologies) supplemented with 10% FBS). MCF10A cells were supplemented with additional cholera toxin (100 ng/mL), hydrocortisone (0.5 mg/mL), insulin (10 μg/mL), and epidermal growth factor (EGF) (20 ng/mL). Stable isogenic cell lines of HCC1937 were generated using LDHA, LDHB, or the non-silencing control Expression Arrest GIPZ lentiviral shRNA particles from Open Biosystems (ThermoFisher Scientific Inc, Fremont, CA, USA). Infected cells were selected and routinely cultured with 1 μg/mL puromycin (Sigma-Aldrich).
Materials and antibodies
Anti-LDHA rabbit monoclonal antibody, anti-cleaved poly (ADP-ribose) polymerase (PARP) (D214) mouse monoclonal antibody, and anti-pyruvate kinase M2 (PKM2) and anti-PARP rabbit polyclonal antibodies were obtained from Cell Signaling Technology (Danvers, MA, USA); anti-LDHB mouse monoclonal antibody was from Abcam (Cambridge, MA, USA); and anti-β-actin mouse monoclonal antibody was from Sigma-Aldrich. All secondary antibodies were from LI-Cor, Inc. (Lincoln, NE, USA); (1,6-13C2)glucose, oligomycin, and rotenone were from Sigma-Aldrich. All other materials were obtained from Life Technologies or from Sigma-Aldrich.
NAD+/NADH concentration determination
Four million HepG2 cells were extracted with 400 μL of NAD+/NADH extraction buffer (BioVision Research Products, Mountain View, CA, USA) and processed with NAD+/NADH quantification kit (BioVision Research Products), per manufacturer’s instructions.
Lactate production assays
Snu398 or HepG2 cells were plated at 7,500,000/mL in 100 μL R10 per well into 96-well plates and allowed to attach overnight. Compound dose responses were prepared in DMSO for nine points in 3-fold dilutions from 30 μM and added to the cells in 100 μL R10 medium. Cells were incubated for 1 h at 37°C to allow for inhibition of LDHA prior to start of the experiment (pre-incubation). After 1 h, the pre-incubation medium was removed, and the cells were washed in PBS and placed in 100 μL glucose-free RPMI medium supplemented with 1.7 mM glucose (low glucose medium) and containing the same compound dilutions. Cells were incubated for 30 minutes at 37°C and 20 μL of conditioned medium from each well was subjected to RapidFire-MS analysis using RapidFire 300 microfluidics system (Agilent Technologies, Santa Clara, CA, USA) and the API 4000 QTrap tandem mass spectrometer (Applied Biosystems, Framingham, MA, USA). Lactate was monitored by the transition of m/z 89 >m/z 43. Data were acquired using Analyst software for Windows Ver. 1.5 (Applied Biosystems), and the RapidFire Integrator 3.4 peak detection and integration software (Agilent Technologies) was used to generate peak areas for the reaction monitoring signals. The concentrations of lactate were calculated from peak areas using a lactate calibration curve. CellTiter-Fluor™ (CTF) cell viability assay (Promega, Madison, WI, USA) was used per manufacturer’s instructions to evaluate cell viability in each well, lactate concentration in each well was normalized to the CTF value, and the lactate/CTF ratio obtained in DMSO-treated cells was set at 100%. For cell lines other than Snu398 and HepG2, the cell numbers varied to obtain approximately 80% confluency and incubation times with the inhibitor varied between 30 and 120 minutes. MDA-MB-453 cells were cultured in normoxic (21% oxygen) or hypoxic (1%) conditions overnight in DMEM supplemented with 5% FBS. Medium was exchanged with physiological DMEM (5 mM glucose, 0.5 mM glutamine, no FBS) containing DMSO or Compound 1 at multiple concentrations and collected after 2 h for hypoxic cells and 6 h for normoxic cells. Lactate concentrations were quantified using a YSI 2900 Biochemistry Analyzer with a lactate oxidase probe (YSI Incorporated, Yellow Springs, OH, USA).
13C nuclear magnetic resonance (NMR) spectroscopy
Snu398 or HepG2 cells were trypsinized and 2,000,000 cells were resuspended in 10 mL R10 containing LDHA inhibitor or DMSO control and incubated for 1 h at 37°C to ensure LDHA inhibition prior to start of NMR data acquisition. After this pre-incubation, cells were washed in PBS and resuspended in 400 μL of glucose-free RPMI supplemented with 10% deuterium oxide, 11 mM D-(1,6-13C2)glucose, and LDHA inhibitor or DMSO. An additional 600 μL of the same medium was gently added to top off the resuspended cells. Five mm NMR tubes with reduced sample depth (Shigemi Inc, Allison Park, PA, USA) were used to ensure that the cells remain inside the active volume of the receiver coil of the probe. The 13C NMR spectra were acquired with a Varian Unity INOVA™ 600 system (Agilent Technologies) operating at 150.8 MHz. The spectra were recorded using a 5-mm broadband probe with 1H decoupling, regulated at 37°C, 32,000 complex data points, sweep width of 37559 Hz, acquisition time of 0.8724 sec/scan, 128 scans and a relaxation delay of 1 sec. Prior to Fourier transformation, sensitivity and resolution were enhanced by applying an exponential line-broadening of 1 Hz over the first 32,000 data points and then zero filled to 256,000 data points. Spectra were analyzed using in-house data analysis software.
Oxygen consumption and extracellular acidification rates measurement
One hundred twenty thousand Snu398 or HepG2 human hepatocellular carcinoma cells per well were plated in XF24 PS or XF96 PET cell culture microplates (Seahorse Biosciences, North Billerica, MA, USA) and allowed to attach and proliferate for 18 h. Cells were equilibrated with DMEM lacking bicarbonate (Seahorse Biosciences) supplemented with glucose at 37°C for 1 h in an incubator lacking CO2. The rate of change of dissolved O2 (oxygen consumption rate, OCR) and rate of change of pH (extracellular acidification rate, ECAR) in medium immediately surrounding the cells was measured in the XF24 Analyzer (Seahorse Biosciences) per manufacturer’s instructions. Compound doses were injected into the wells at the indicated times. For dose response curves, the OCR and ECAR levels obtained in each well before compound addition were subtracted from the values obtained 92 minutes after compound addition.
For XF96 experiments breast cancer cells were seeded in DMEM without bicarbonate at 16,000/well in XF96 V3 PET plates (Seahorse Biosciences) coated with CellTak (BD Biosciences, San Jose, CA, USA). After baseline readings were obtained, multiple concentrations of compound 1 (2.5 to 40 μM) or DMSO, rotenone (1 μM), and antimycin (1 μM) were added sequentially. HepG2 cells were seeded at 16K/well in XF96 V3 PET plates and left overnight for attachment. The medium was exchanged with DMEM without bicarbonate prior to readings on the XF96 Analyzer (Seahorse Biosciences).
Seahorse XF96 permeabilized cells
HepG2 cells were seeded overnight in DMEM, 10% FBS, at 16,000 cells/well in an XF96 V3 PET plate. The next day, the medium was removed, and the wells were rinsed and filled with mitochondrial assay solution (MAS: 70 mM sucrose, 220 mM mannitol, 10 mM KH2PO4, 5 mM MgCl2, 2 mM HEPES, 1.0 mM EGTA) containing 10 mM pyruvate/10 mM malate/4 mM ADP and 2 nM XF Plasma Membrane Permeabilizer (PMP) (Seahorse Biosciences). Compound 1, oligomycin (2 μg/mL), and antimycin (4 μM) were added sequentially. OCR and ECAR readings were determined for three cycles (two minutes mixing, two minutes measuring) after each compound addition.
Ten million Snu398 or HepG2 cells per flask were plated in R10 medium in T150 tissue culture flasks and allowed to attach overnight. The following morning, medium was replaced with medium containing either DMSO or 10 μM LDHA inhibitor, and cells were incubated for 24 h at 37°C. The conditioned medium was collected and cells were washed, lifted by trypsinization and pelleted. Both medium and cell samples were submitted for mass spectroscopy (MS) analysis of over 500 individual metabolites. All samples were prepared in replicates of five.
Pyruvate kinase (PK) activity and PKM2 tetramerization determination
One and a half million cells per well were plated in 6-well tissue culture plates in R10 medium and allowed to attach overnight. The following morning, cells were washed and medium replaced with 2 mL/well of low-glucose medium. DMSO control or the indicated doses of LDHA inhibitor dissolved in DMSO were added to the wells and cells were incubated at 37°C for 6 h. Following the incubation, cells were washed, scraped into 200 μl of distilled de-ionized water and homogenized on ice using OMNI THq homogenizer (Omni International, Kennesaw, GA, USA). Homogenates were cleared by centrifugation and protein concentration measured by bicinchoninic acid (BCA) assay (Thermo Fisher Scientific, Waltham, MA, USA). PK activity was determined using PK Activity kit (BioVision Research Products) per manufacturer’s instructions and normalized to protein concentration. For detection of PKM2 oligomerization, the same protein samples were separated by non-denaturing gel electrophoresis using the NativePAGE™ gel system (Life Technologies), transferred onto polyvinylidene difluoride (PVDF) membranes and subjected to western immunoblotting with anti-PKM2 antibody.
Cell proliferation and survival analysis
Sixty thousand Snu398 cells per well were plated in 6-well tissue culture plates in RPMI-1640 medium supplemented with 2.5% charcoal-stripped FBS (Life Technologies). Cells were allowed to attach overnight and then DMSO control or the indicated doses of LDHA inhibitor dissolved in DMSO were added directly to the wells. After 4 to 8 days of incubation in the indicated oxygen conditions, adherent cells were trypsinized, counted, and had their viability assessed by the trypan-blue exclusion method using the Vi-Cell XR Cell Viability Analyzer (Beckman Coulter, Brea, CA, USA). The same procedure was carried out for HepG2 cells, except non-charcoal-stripped FBS was used. For PARP cleavage analysis, both floating and adherent cells were collected after 24 h and the whole-cell extracts were prepared in radioimmunoprecipitation assay (RIPA) buffer (Sigma-Aldrich) supplemented with protease and phosphatase inhibitor cocktails (Sigma-Aldrich). Solubilized proteins were separated by SDS-PAGE, transferred onto nitrocellulose membrane, and detected with each specific antibody.
Pathway contribution analysis
Gene sets covering selected pathways were identified using published resources  and were as follows: glycolysis - GPI, TPI1, PFKL, ALDOB, PKLR, PGAM1, HK1, PGK1, GAPDH, ENO1; fatty acid synthesis - PTGES3, PRG3, FADS1, PTGS1, MCAT, FADS2, CD74, BRCA1, MIF, OXSM, PTGDS, LTA4H, HPGD, DEGS1; glutaminolysis - GCLC, ALDH18A1, GLUD2, GLUD1, ASL, GCLM, ARG1, PYCR1, GAD2, ALDH4A1, ASRGL1, DDAH2, GAD1, DDAH1; pentose phosphate - ALDOA, TALDO1, ALDOC, PGD, ALDOB, TKTL2, TKTL1, DERA, RPIA, PRPS1L1, PFKL, PFKP, FBP1, TKT, RBKS, PFKM, FBP2, LOC729020, PGM2, GPI, PGLS, RPE, G6PD, H6PD, PGM1, PRPS2, PRPS1. Increases in transcript levels of the enzymes composing these cellular processes were used as measurements of pathway activation. To this end, transcript levels were obtained from data that we previously submitted to the public database (https://cabig-stage.nci.nih.gov/community/caArray_GSKdata/), converted to z-scores, and summed over the pathway for each cell line. The total sum served as a measurement of pathway activity.
All studies were conducted after review by the Institutional Animal Care and Use Committee at GlaxoSmithKline (GSK) and in accordance with the GSK Policy on the Care, Welfare and Treatment of Laboratory Animals. Compound 1 was administered to male CD mice or male Sprague–Dawley rats orally or by intravenous (IV) infusion over 120 minutes into a femoral vein. Arterial blood samples were collected over time and Compound 1 concentration was determined by liquid chromatography (LC)/MS/MS analysis.
Statistical significance was determined using two-way analysis of variance (ANOVA) with contrasts for pairwise comparisons. Results were considered statistically different when P was <0.05.