β-alanine suppresses malignant breast epithelial cell aggressiveness through alterations in metabolism and cellular acidity in vitro

Non-malignant breast epithelial cells (MCF-10a) and malignant breast epithelial cells (MCF-7) were purchased from ATCC (Manassas, VA). MCF-10a cells were cultured in combination F-12 DMEM media supplemented with 50 nM hydrocortisone, 20 ng/ml EGF, 0.01 mg/ml human insulin, 5% fetal bovine serum (FBS), and 1% Pen-Strep. MCF-7 cells were cultured in DMEM supplemented with 10% FBS with 0.01 mg/ml human insulin, and 1% Pen-Strep. Cells were treated with β-alanine from Sigma (St. Louis, MO) at 100 mM dissolved in media and incubated in 5% CO₂ at 37°C for 24 hours.

Cells were seeded overnight in 24-well culture plate from SeaHorse Bioscience (Billerica, MA) at density 5 × 10⁵ cells/well. To control for nitrogen load and hyper-osmolality, we also treated control cells with isonitrogenous valine or D-alanine. Following treatment, culture media was removed and replaced with XF Assay Media from SeaHorse Bioscience (Billerica, MA) containing 4500 mg/L glucose free of CO₂ and incubated at 37°C. Per manufactures’ protocol, SeaHorse XF24 Extracellular Analyzer injection ports were loaded with oligomycin, an inhibitor of ATP synthase which induces maximal glycolytic metabolism and reveals endogenous proton leak (mitochondrial uncoupling) at a final concentration 1.0 μM. Measurement with oligomycin were taken for 24 minutes followed by the addition of carbonyl cyanide p-[trifluoromethoxy]-phenyl-hydrazone (FCCP), an uncoupler of electron transport that induces peak oxygen consumption (an indirect indicator of peak oxidative metabolism) at final concentration 1.25 μM. FCCP measurements were taken for 24 minutes followed by rotenone addition in 1.0 μM final concentration to reveal non-mitochondrial respiration and end the metabolic reactions [16, 17]. Extracellular acidification rate (ECAR) calculated from proton production rate is an indirect measure of glycolytic capacity. Oxygen consumption (OCR), a measure of oxidative metabolism was measured using the SeaHorse XF24 Extracellular Analyzer from SeaHorse Bioscience (Billerica, MA). SeaHorse XF24 Extracellular Analyzer was run using 8 minute cyclic protocol commands (mix for 3 minutes, let stand 2 minutes, and measure for 3 minutes) as previously described [18, 19].

Following treatment as described above, buffer-free media was added to the cells to measure total ATP production. The cells were lysed in 1% CHAPS lysis buffer from Chemicon (Billerica, MA) in PBS with Ca²⁺ and Mg²⁺ and the ATP-containing supernatant was recovered. Samples were allocated into a 96-well plate with a 1:1 dilution of ATP Bioluminescence Reagent from Sigma (St. Louis, MO) and ATP-containing samples (final reaction volume = 100 μl) and luminescence was measured as previously described [18].

Following incubation, the total RNA was extracted using RNeasy Kit from Qiagen (Valencia, CA), per manufacturer’s protocol. Total RNA was quantified by Nanodrop spectrophotometry. cDNA was synthesized from 5000 ng total RNA using the Retroscript™ RT kit from Ambion (Austin, TX) according to manufacturer’s instructions. PCR primers were designed using Primer Express software from Invitrogen (Carlsbad, CA) and synthesized by Integrated DNA Technologies (Coralville, IA). Amplification of glucose transporter 1 (GLUT1), lactate dehydrogenase (LDH), and peroxisome proliferator-activated receptor gamma coactivator 1 alpha (PGC-1α) were normalized to the housekeeping gene, TATA Binding Protein (TBP). Table 1 summarizes the forward and reverse primers of each gene. qRT-PCR reactions were performed in triplicate using the LightCycler 480 real-time PCR system from Roche Applied Science, (Indianapolis, IN). SYBR Green based PCR was performed in triplicate using 5000 ng of cDNA per sample; final primer concentrations were 10 μM in a total volume of 30 μl. The following cycling parameters were used: 95°C for 10 minutes followed by 45 cycles of 95°C for 15 seconds, and 60°C for one minute. Relative expression levels were determined by the ΔΔCp method and compared to the lowest expressing group as previously described [18, 20, 21].

Following treatment, the media was removed and the cells were re-suspended in pre-warmed media with 200 nM Mitotracker Green from Life Technologies (Carlsbad, CA) and incubated for 45 minutes as described above. The cells were pelleted, the media with Mitotracker was removed and the cells were suspended in pre-warmed media. Group mean fluorescence was measured using Facscalibur filtering 488 nm as previously performed [18]. To quantify protein expression, cells were treated as described before, trypsinized, permeabilized and blocked for 1 hour with PBS with 0.1% Triton 100X and 3.0% BSA from Sigma (St. Louis, MO). Cells were stained overnight with anti-GLUT1 antibodies (SPM498, mouse monoclonal) from Abcam (Cambridge, MA) at a final concentration of 1 μg/ml in PBS containing 0.1% BSA. The cells were rinsed with PBS with 0.1% Triton 100X and 3.0% BSA, and a secondary antibody (either AlexFluor 488 from Invitrogen (Carlsbad, CA)) was applied in 1:200 dilution for 1 hour [22]. Cells were then rinsed and re-suspended in PBS and the fluorescence was quantified as described above.

Chamber-slides from BD Bioscience (Sparks, MD), were seeded with 5000 cells/well. To verify protein expression, cells were cultured and treated for 24 hours as described above. Cells were fixed using 3.7% formaldehyde in media, permeabilized with PBS with 0.1% Triton 100X from Sigma (St. Louis, MO) for 10 minutes and blocked for 1 hour with PBS with 0.1% Triton 100X and 3.0% BSA from Sigma (St. Louis, MO). Cells were stained overnight with anti-GLUT1 antibodies (SPM498, mouse monoclonal) from Abcam (Cambridge, MA) at a final concentration of 1 μg/ml in PBS containing 0.1% BSA. The cells were rinsed with PBS with 0.1% Triton 100X and 3.0% BSA, and an AlexFluor 633 from Invitrogen (Carlsbad, CA) was applied in 1:200 dilution. Slides were mounted with Prolong Gold with DAPI from Invitrogen (Carlsbad, CA) and cured overnight. Cells were imaged using the Axiovert 25 microscope with AxioCam MRc from Zeiss (Thornwood, NY).

Whole-cell lysates were collected following 24 hour treatment. Whole cell lysates were prepared by harvesting the cells on ice in high salt lysis buffer (25 mM Tris base, 8 mM MgCl2, 1 mM DTT, 15% glycerol, 0.1% Triton) supplemented with protease inhibitor mix (Sigma, St. Lois MO), followed by incubation on ice for 60 minutes. Insoluble material was removed by centrifugation at 17,500 × G for 3 minutes and protein concentrations were determined by Bradford assay (Protein Assay Dye Reagent Concentrate, Bio-Rad Laboratories, Hercules, CA). Total protein (120 μg per sample) was size-separated by 10% sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE) and electro-transferred to nitrocellulose membranes. After blocking in TBST-5% non-fat milk powder for 1 hour, membranes were probed at 4°C for 24 hours with an anti-PGC-1α antibody from Santa Cruz Biotechnologies (Santa Cruz, CA), anti-GLUT1, anti-carnosine synthase (CS), or anti-lactate dehydrogenase (LDH) antibodies from Abcam (Cambridge, MA), normalized to an anti-β-actin primary monoclonal antibody from Sigma (St. Louis, MO) in TBST-1% non-fat milk powder overnight. Bound antibodies were detected by horseradish peroxidase-conjugated secondary antibodies from Sigma (St. Louis, MO) and by chemiluminescence using the ECL Plus Western Blotting Detection kit from GE Healthcare Life Sciences (Little Chalfont, Buckinghamshire, UK). Signal intensities were obtained by densitometry using ImageJ software (available from the NIH at http://​rsbweb.​nih.​gov/​ij/​) by quantifying lane intensities followed by normalizing intensity with corresponding β-actin.

Cell proliferation was assessed by measuring confluence-mask quantification via phase-contrast Incucyte ZOOM live content imaging and cell surveillance from Essen Bioscience (Ann Arbor, MI) in a 96-well plate. Cells were re-suspended in respective growth media with and without β-alanine at 100 mM, incubated and confluency was measured every hour for 48 hours. Cell migration was assessed by applying a standard scratch through confluent cells following treatment for 24 hours. The cells were then rinsed and the media replaced with corresponding treatments (control or β-alanine at 100 mM). Migration was determined by measuring the change in confluency for 24 hours following the initial wound. To assess changes in viability, cells were seeded overnight, and treated and incubated as described above for either 24 or 48 hours. The cells were then incubated for 1 hour in medium containing 10% WST-1 cell proliferation reagent from Roche (Indianapolis, IN) and fluorescence was measured using a Wallac Victor3V 1420 Multilabel Counter from PerkinElmer (Waltham, MA). To further investigate the effects of β-alanine on cell survival, cells were treated with varying concentrations (from 10 pM – 10 μM) of the antineoplastic Dox (in DMSO with final concentration = 0.1% for all Dox experiment groups including control) from Sigma (St. Louis, MO) with and without β-alanine at 100 mM for 24 hours. Treatment doses and duration were determined through pilot data for β-alanine and from previous observations for Dox [23].

Metabolic measurements were analyzed by ANOVA with Tukey’s pair-wise comparisons for β-alanine compared with various controls. ATP concentration, cell migration, proliferation, immunoblotting, as well as flow cytometry and microscopy were analyzed using student’s t-test. Gene expression was quantified by relative expression using the ΔΔCp method, and analyzed using student’s t-test [18, 20]. Cell viability was analyzed using analysis of variance (ANOVA) with Dunnett’s post hoc test. All data is represented as average ± standard deviation (SD) normalized to the control mean (control = 100) with *, **, and *** indicating p < 0.05, p < 0.01, and p < 0.001 statistical differences compared to control, respectively.

To investigate the effects of β-alanine on aerobic glycolysis, we measured extracellular acidification rate (ECAR) following treatment for 24 hours. Both malignant (MCF-7) and non-malignant (MCF-10a) cells treated with β-alanine exhibited significantly reduced basal glycolysis compared with both true control and equivocal nitrogen load from 100 μM valine (Figure 1A). Additionally, β-alanine-treated MCF-7 cells demonstrated significantly reduced peak glycolytic capacities (peak glycolysis induced by oligomycin) compared with respective controls, but was unchanged in MCF-10a cells (Figure 1B). Similarly, both cell models exhibited significantly reduced cellular acidity following treatment with β-alanine compared with both true negative and valine control (Figure 1C).

To investigate the effects of β-alanine on oxidative metabolism, we measured oxygen consumption rate following treatment for 24 hours. Both MCF-7 and MCF-10a cells treated with β-alanine exhibited significantly reduced basal oxidative metabolism, however both D-alanine and L-valine controls were significantly reduced in the MCF-10a (Figure 1D). In addition to suppressed oxidative metabolism, MCF-10a cells exhibited increased oxidative reliance (expressed as a ratio of oxidative metabolism to glycolytic metabolism) illustrated in Figure 1E. Surprisingly, β-alanine-treated MCF-10a cells had significantly elevated total ATP content compared with their control, while MCF-7 ATP content was significantly reduced (Figure 1F).

To investigate the effects of β-alanine on gene expression related to aerobic glycolysis, we first measured LDH following treatment for 24 hours. Both MCF-7 and MCF-10a cells treated with β-alanine exhibited significantly reduced LDH mRNA levels while neither cell model exhibited significantly altered protein expression (Figure 2A and B, respectively). Because one potential mechanism of action of β-alanine could be through carnosine production, we measured Carnosine Synthase (CS) expression. CS was present in both cell models, but unaltered by β-alanine treatment (Figure 2C). In agreement with RNA levels, GLUT1 protein expression was significantly elevated in MCF-10a cells treated with β-alanine compared with control (measured by flow cytometry) which was verified by confocal microscopy (Figure 2D).

In order to evaluate the effects of β-alanine treatment on expression of genes involved in oxidative metabolism, we performed qRT-PCR following 24 hours treatment. Both β-alanine treated cell models demonstrated significantly suppressed RNA expression of PGC-1α (Figure 3A). Interestingly, PGC-1α protein expression was suppressed in MCF-7 cells but was unaltered in MCF-10a cells (Figure 3B). Consistent with PGC-1α protein expression, mitochondrial content was also suppressed in MCF-7 cells but was unaltered in MCF-10a cells (Figure 3C).

In order to assess changes in cellular growth, we measured the effect of β-alanine on both cell models following treatment for either 24 or 48 hours. Proliferation of MCF-7 and MCF-10a was significantly suppressed following growth in media containing β-alanine compared with control (Figure 4A and B, respectively). Neither cell model experienced significant changes in cell viability at any of the tested doses (Figure 4C and D). To evaluate the effect β-alanine on cell migration, we measured cell motility via scratch-wound migration assay in a live cell imager taken hourly images for 48 hours. In both cell models, β-alanine treatment significantly reduced migration velocity compared with corresponding controls (Figure 5A and B). Because cellular uptake of Dox is significantly influenced by intra and extracellular pH, we investigated the effect of β-alanine on the efficacy of Dox. We treated MCF-7 cells with varying concentrations of Dox with and without β-alanine at 100 mM for 24 hours. Dox plus β-alanine treatment significantly reduced MCF-7 viability at all doses tested compared with DMSO control, an effect which was not seen in the Dox-only groups until high concentrations (1 μM -10 μM) (Figure 6). There were also significant reductions in MCF-7 viability at low doses of Dox with simultaneous β-alanine treatment. These observations suggest that β-alanine sensitizes cancer cells to Dox, thus improving the efficacy at lower concentrations.