The NAD+/NADH ratio is proportional to serine synthesis rate.

(A) α-ketobutyrate (AKB) is an exogenous electron acceptor that promotes the oxidation of NADH into NAD+ through its conversion to α-hydroxybutyrate (AHB). Rotenone inhibits activity of complex I of the mitochondrial electron transport chain (ETC) to block NADH oxidation into NAD+. (B) NAD+/NADH ratio measured in A549 cells cultured without serine and indicated concentrations of AKB (left) and rotenone (right) for 24 hours. NAD+/NADH ratios are normalized to untreated cells cultured without serine, n=3. Pearson correlation coefficients and P-values were calculated by simple linear regression, ***p<0.005, ****p<0.001. (C) Schematic depicting NAD+ requirement to produce serine labeled (M+3) from U-13C-glucose. Under culture conditions without serine, intracellular serine rapidly effluxes from cells into culture media. Thus, cells and media were jointly analyzed at each time point of kinetic tracing to better capture levels of newly synthesized serine. (D) Serine labeled (M+3) from U-13C-glucose over time in A549 cells cultured without serine and with indicated concentrations of AKB or rotenone for 24 hours prior to U-13C-glucose exposure. Serine levels are normalized to internal norvaline standard and cell number measured in indicated conditions, n=3. (E) Relative NAD+/NADH ratios from A549 cells cultured without serine after exposure to 250 μΜ AKB, 1000 μΜ AKB, 20 nM rotenone, or 40 nM rotenone for 24 hours are plotted against the serine synthesis rates from corresponding conditions as shown in (D). Serine synthesis rates are calculated from labeled serine (M+3) after 1 and 15 minutes of U-13C-glucose exposure and normalized to untreated culture conditions without serine, n=3. Pearson correlation coefficients and P-values were calculated by simple linear regression, ****p<0.001. Data shown for all panels represent mean ± SD.

The NAD+/NADH ratio differs between cancer cells upon serine withdrawal and correlates with proliferation in serine depleted conditions.

(A) Proliferation rate (doublings per day) of indicated cancer cells cultured without serine for 72 hours normalized to proliferation rate of corresponding cancer cells cultured with serine, n=3. (B) NAD+/NADH ratios of indicated cells cultured with or without serine for 24 hours. Yellow indicates cells with unaltered NAD+/NADH ratios (redox non-responder cells) and blue indicates cells with elevated NAD+/NADH ratios when cultured without serine (redox responder cells). Values are normalized to the NAD+/NADH ratios of cells cultured with serine, n=3. P-values were calculated by unpaired Student’s t-test, **p<0.01, ***p<0.005. (C) Relative NAD+/NADH ratios of cells cultured without serine are plotted against proliferation rates of corresponding cells cultured without serine. Data points are labeled to denote the different cell lines (Calu6, A549, 8988T, MIA PaCa-2, H1299, and HCT116). Pearson correlation coefficient and P-values were calculated by simple linear regression, *p<0.05. (D) Serine labeled (M+3) from U-13C-glucose over time in A549 and H1299 cells cultured with or without serine for 24 hours prior to U-13C-glucose exposure. Serine levels were normalized to internal norvaline standard and cell number measured in indicated conditions, n=3. (E) Absolute fold change of serine labeled (M+3) from U-13C-glucose in A549 and H1299 cells cultured without serine relative to serine labeled (M+3) from U-13C-glucose in A549 and H1299 cells cultured with serine for 24 hours after 12 minutes of U-13C-glucose exposure, n=3. P-values were calculated by unpaired Student’s t-test, **p<0.01. Data shown for all panels are means ± SD.

Elevated mitochondrial respiration is associated with increased NAD+/NADH ratios and serine synthesis rates.

(A) Oxygen consumption rate (OCR) of indicated cells cultured with or without serine for 24 hours. Values are averages of three repeat measurements, n=14-20. P-values were calculated by unpaired Student’s t-test, *p<0.05, ****p<0.001. (B) OCR of cells (H1299 top, A549 bottom) cultured with indicated concentrations of serine for 24 hours. Values are averages of three repeat measurements, n=3-9. P-values were calculated by unpaired Student’s t-test, ****p<0.001. (C) NAD+/NADH ratios of cells (H1299 top, A549 bottom) cultured with indicated concentrations of serine for 24 hours. Values are normalized to the NAD+/NADH ratio of cells cultured in 400 μM serine (serine concentration in DMEM), n=3. P-values were calculated by unpaired Student’s t-test, **p<0.01, ***p<0.005. (D) Relative OCR plotted against relative NAD+/NADH ratios in H1299 and A549 cells cultured in 0, 25, 50, and 100 μM serine for 24 hours. OCR and NAD+/NADH ratios are normalized to measurements from the 100 μΜ serine condition, n=3. Pearson correlation coefficients and P-values were calculated by simple linear regression, *p<0.05, ****p<0.001. (E) Serine labeled (M+3) from U-13C-glucose over time in cells (H1299 top, A549 bottom) cultured with indicated serine concentrations for 24 hours prior to U-13C-glucose exposure. Serine levels are normalized to internal norvaline standard and cell number measured in indicated conditions, n=3. (F) Relative NAD+/NADH ratios of cells (H1299 top, A549 bottom) cultured with 0, 25, 50, and 100 μM serine for 24 hours plotted against relative serine synthesis rates of cells cultured in corresponding conditions. Serine synthesis rates are calculated from labeled serine (M+3) after 1 and 15 minutes of U-13C-glucose exposure and normalized to untreated culture conditions without serine, n=3. Pearson correlation coefficients and P-values were calculated by simple linear regression, ***p<0.005. Data shown for all panels are means ± SD.

Mitochondrial respiration governs the endogenous cell NAD+/NADH ratio and influences serine synthesis.

(A) Schematic depicting the mitochondrial electron transport chain (ETC) and the protonophore activity of FCCP that uncouples the ETC from ATP synthase activity. CoQ(H2) indicates the oxidized and reduced (H2) forms of coenzyme Q, cytc denotes cytochrome c. (B,C) Proliferation rate (doublings per day) of A549 (B) and H1299 cells (C) cultured with or without serine with indicated concentrations of FCCP for 72 hours, n=3. Increasing doses of FCCP statistically increased proliferation of A549 cells cultured without serine (p<0.001). P-values were calculated by simple linear regression. (D) Serine labeled (M+3) from U-13C-glucose over time in A549 cells without serine and with or without 2 μM FCCP treatment for 24 hours prior to U-13C-glucose exposure. Serine levels are normalized to internal norvaline standard and cell number measured in indicated conditions, n=3. (E) Serine labeled (M+3) from U-13C-glucose over time in H1299 cells without serine, and with or without 2 μM FCCP treatment for 24 hours prior to U-13C-glucose exposure. Serine levels are normalized to internal norvaline standard and cell number measured in indicated conditions, n=3. (F) Oxygen consumption rate (OCR) of A549 cells cultured with or without serine with indicated FCCP treatment for 24 hours. Values are averages of three repeat measurements, n=5-9. P-values were calculated by unpaired Student’s t-test, ****p<0.001. (G) Relative NAD+/NADH ratio of A549 cells cultured with or without serine with indicated FCCP treatment for 24 hours. NAD+/NADH ratios are normalized to the NAD+/NADH ratio in A549 cells cultured with serine, n=3. P-values were calculated by unpaired Student’s t-test, *p<0.05, ***p<0.005. (H) Proliferation rate (doublings per day) of A549 cells cultured with or without serine, 2 μM FCCP, and indicated concentrations of rotenone for 72 hours, n=3. P-values were calculated by unpaired Student’s t-test, *p<0.05. (I) Serine labeled (M+3) from U-13C-glucose over time in A549 cells cultured without serine and with or without 2 μM FCCP or 20 nM rotenone as indicated for 24 hours prior to U-13C-glucose exposure. Serine levels are normalized to internal norvaline standard and cell number measured in indicated conditions, n=3. (J) Relative NAD+/NADH ratio of A549 cells cultured with or without serine, 2 μM FCCP, or 20 nM rotenone as indicated. NAD+/NADH ratios are normalized to the NAD+/NADH ratio in A549 cells cultured with serine, n=3. P-values were calculated by unpaired Student’s t-test, **p<0.01, ***p<0.005. Data shown for all panels are means ± SD.

Lipid depletion enhances cell-specific elevation in mitochondrial respiration and the NAD+/NADH ratio, influencing oxidative citrate and serine synthesis.

(A) Schematic depicting glucose and glutamine oxidative routes for synthesizing citrate, highlighting NAD+ requiring oxidation reactions. (B) Citrate labeled (M+2) from U-13C-glucose over time in A549 and H1299 cells cultured with or without serine for 24 hours prior to U-13C-glucose exposure. Citrate levels are normalized to internal norvaline standard and cell number measured in indicated conditions, n=3. (C) Citrate labeled (M+2) from U-13C-glucose over time in A549 and H1299 cells cultured without serine and with or without 2 μM FCCP for 24 hours prior to U-13C-glucose exposure. Citrate levels are normalized to internal norvaline standard and cell number measured in indicated conditions, n=3. (D) Citrate labeled (M+2) from U-13C-glucose over time in A549 cells cultured with serine and with indicated treatment conditions for 24 hours prior to U-13C-glucose exposure. Citrate levels are normalized to internal norvaline standard and cell number measured in indicated conditions, n=3. (E) Oxygen consumption rate (OCR) in A549 and H1299 cells cultured with or without lipids for 24 hours. Values are averages of three repeat measurements, n=10. Green indicates lipid redox responder cells. Purple indicates lipid redox non-responder cells. (F) Relative NAD+/NADH ratios of A549 and H1299 cells cultured with or without lipids for 24 hours. NAD+/NADH ratios are normalized to the NAD+/NADH ratio of the corresponding cell type cultured with lipids, n=3. P-values were calculated by unpaired Student’s t-test, *p<0.05. (G) Citrate labeled (M+2) from U-13C-glucose over time in A549 and H1299 cells cultured with or without lipids for 24 hours prior to U-13C-glucose exposure. Citrate levels are normalized to internal norvaline standard and cell number measured in indicated conditions, n=3. (H) Serine labeled (M+3) from U-13C-glucose over time in A549 and H1299 cells cultured with or without lipids for 24 hours prior to U-13C-glucose exposure. Serine levels are normalized to internal norvaline standard and cell number measured in indicated conditions, n=3. Data shown for all panels are means ± SD.

Lipid depletion increases mitochondrial respiration and the NAD+/NADH ratio in a cell-specific manner to influence serine synthesis and proliferation in serine depleted conditions.

(A) Oxygen consumption rate (OCR) of A549 and H1299 cells cultured with or without serine or lipids for 24 hours. Values are the average of three repeat measurements, n=14-15. P-values were calculated by one-way ANOVA followed by post-hoc Tukey HSD test, ****p<0.001. (B) Relative NAD+/NADH ratios of A549 and H1299 cells cultured with or without serine or lipids for 24 hours. NAD+/NADH ratio is normalized to the NAD+/NADH ratio in corresponding cells cultured with serine and lipids, n=3. P-values were calculated by one-way ANOVA followed by post-hoc Tukey HSD test, *p<0.05, ****p<0.001. (C, D) Left: Serine labeled (M+3) from U-13C-glucose over time in A549 cells (C) and H1299 cells (D) cultured without serine and with or without lipids for 24 hours prior to U-13C-glucose exposure. Right: Serine synthesis rates of A549 cells (C) and H1299 cells (D) cultured in indicated conditions. Serine synthesis rates are calculated from labeled serine (M+3) after 1 and 15 minutes of U-13C-glucose exposure and normalized to the serine synthesis rates of the corresponding cell line cultured without serine and with lipids. Serine levels are normalized to internal norvaline standard and cell number measured in indicated conditions, n=3. Single data point (H1299, -serine +lipid, t=15 min) removed due to detection error with no signal. (E,F) Left: Citrate labeled (M+2) from U-13C-glucose over time in A549 cells (E) and H1299 cells (F) cultured without lipids and with or without serine for 24 hours prior to U-13C-glucose exposure. Right: Citrate synthesis rates of A549 cells (E) and H1299 cells (F) cultured in indicated conditions. Citrate synthesis rates are calculated from labeled citrate (M+2) after 1 an 15 minutes of U-13C-glucose exposure and normalized to the citrate synthesis rates of the corresponding cell line cultured without lipids and with serine. Citrate levels are normalized to internal norvaline standard and cell number measured in indicated conditions, n=3. (G) Proliferation rate (doublings per day) of A549 cells cultured with or without serine or lipids for 72 hours, n=3. P-values were calculated by one-way ANOVA followed by post-hoc Tukey HSD test, **p<0.01 (H) Proliferation rate (doublings per day) of H1299 cells cultured with or without serine or lipids for 72 hours, n=3. Data shown for all panels are means ± SD.

The cell NAD+/NADH ratio is regulated in a cancer cell-specific manner by changes to mitochondrial respiration in response to nutrient availability, impacting oxidative biosynthetic reactions.

(A) In serine and lipid-replete conditions, cells can acquire environmental serine and lipids to support proliferation. Cells can also oxidize glucose to produce serine, as well as citrate for lipid synthesis. The oxidation reactions involved can be limited by NAD+ availability, which is impacted by mitochondrial respiration. (B) In nutrient depleted conditions, the ability to support oxidative reactions is dependent on the mitochondrial response to the nutrient depletion. For redox responders, environmental depletion of serine or lipids increases the cell NAD+/NADH ratio to support both oxidative citrate and serine synthesis. This ability to increase respiration enables proliferation in conditions lacking serine. (C) In redox non-responder cells, neither mitochondrial respiration nor the NAD+/NADH ratio increases following nutrient depletion. This limits oxidative synthesis of serine (and citrate), subsequently limiting proliferation in environments that lack serine.

Altering NAD+/NADH with α-ketobutyrate (AKB) or rotenone does not change PHGDH protein levels and minimally changes cell viability.

(A,B) Immunoblots assessing PHGDH protein levels in A549 cells cultured with or without serine, 1 mM AKB (A), or 20 nM rotenone (B), as indicated for 24 hours. HSP90 expression is shown as a loading control. (C) Percent of A549 cells dead, measured using dye exclusion 72 hours after exposure to the indicated concentrations of rotenone in serine-replete or serine depleted conditions, n=3. Serine depletion led to a statistically significant increase in cell death compared to serine-replete conditions at all concentrations of rotenone with the exception of 40 nM rotenone, p<0.05. There was no statistical difference in cell death between 0 and 20 nM rotenone in either serine-replete or serine depleted conditions. P-values were calculated using unpaired Student’s t-test. Data shown are means ± SD.

Relationship between PHGDH protein expression, cell NAD+/NADH ratio, and cell proliferation rate following serine deprivation.

(A) Proliferation rates (doublings per day) of cells cultured with or without serine for 72 hours, n=3. (B) Above: Schematic of the serine synthesis pathway. Below: Immunoblot assessing protein levels of PHGDH, PSAT1, and PSPH in the indicated cells cultured with or without serine for 24 hours. HSP90 protein expression is shown as a loading control. These data are representative of three independent experiments. Abbreviations – 3-PG: 3-phosphoglycerate, PHGDH: phosphoglycerate dehydrogenase, 3PHP: 3-phosphohydroxypyruvate, glu: glutamate, αKG: α- ketoglutarate, PSAT1: phosphoserine aminotransferase 1, 3PS: 3-phosphoserine, PSPH: phosphoserine phosphatase. (C) Quantification of PHGDH protein expression for the indicated cells under serine depleted conditions (normalized to HSP90 protein expression) from three biological replicates, correlated with the corresponding relative proliferation rate of indicated cells under serine depleted conditions based on data presented in (A). Cell lines enclosed in the light blue shaded box indicate cancer cell lines with serine synthesis enzyme expression that could not fully predict proliferation in the absence of serine that we chose to examine further. Pearson correlation coefficient calculated by simple linear regression. (D) Cell number over time for either A549 cells (yellow) or H1299 cells (blue) in either serine-replete or serine depleted conditions, n=6. (E) Calculated doublings per day from the data in (D) for the indicated cells and conditions, n=6. P-values were calculated using unpaired Student’s t-test, ****p<0.001. (F) Correlation between the protein expression of serine synthesis enzymes PHGDH, PSAT1, and PSPH across three biological replicates and proliferation rate in serine depleted conditions for the indicated cell lines, n=3. Pearson correlation coefficient and P-values were calculated by simple linear regression, *p<0.05, ***p<0.005. (G) Unnormalized NAD+/NADH ratios measured for the indicated cell lines cultured with or without serine for 24 hours, n=3 per each cell line. P-values were calculated by unpaired Student’s t-test, *p<0.05, ***p<0.005. (H) Fraction of total serine labeled from U-13C-glucose over time in A549 and H1299 cells that were cultured with or without serine for 24 hours prior to U-13C-glucose exposure as indicated, n=3. (I) Left: Total intracellular serine levels over time from the kinetic tracing with A549 and H1299 cells cultured with or without serine for 24 hours prior to U-13C-glucose exposure as described in (H). Right: Average of total intracellular serine levels from all time points of the kinetic U-13C-glucose isotope tracing experiment shown in (H). Serine levels were normalized to internal norvaline standard and cell number. P-values were calculated by unpaired Student’s t-test, ***p<0.005. Data shown for all panels are means ± SD.

Lactate dehydrogenase activity, reliance on the NAD+ salvage pathway, mitochondrial respiration, and PHGDH protein levels following serine deprivation.

(A) Schematic depicting the contributions of lactate dehydrogenase (LDH) and the NAD+ salvage pathway to NAD+ generation. Glucose is consumed by cells and converted to pyruvate via glycolysis. LDH can convert pyruvate to lactate by oxidizing NADH to NAD+. In the NAD+ salvage pathway, nicotinamide can be converted to nicotinamide mononucleotide (NMN) by the enzyme nicotinamide phosphoribosyltransferase (NAMPT), which is inhibited by the small molecule FK866. NMN is subsequently converted to NAD+ via NMN adenylyltransferase. (B) Lactate secretion rate normalized to glucose consumption rate as measured in the indicated cells cultured with or without serine for 72 hours, n=3. (C) Proliferation rate (doublings per day) of A549 (above) and H1299 cells (below) cultured with or without serine and with or without the indicated concentrations of FK866 for 72 hours, n=3. There was a statistically significant difference in the sensitivity of A549 and H1299 cells to FK866 in the serine-replete condition (p<0.001), but not in the serine depleted conditions. P-values were calculated by ANCOVA analysis followed by post-hoc Tukey HSD test. (D) Proliferation rate (doublings per day) of A549 (left) and H1299 cells (right) cultured with or without serine and either NMN (100 μM) or FK866 (10 μM) as indicated for 72 hours, n=3. P-values were calculated using unpaired Student’s t-test, ***p<0.005, ****p<0.001. (E) Oxygen consumption rate (OCR) of cells cultured with or without serine for 24 hours, n=10-20. Rotenone and antimycin A were injected between the third and fourth OCR measurements to define non-mitochondrial OCR. Data shown are normalized to cell number. (F) Proliferation rate (doublings per day) of A549 and H1299 cells cultured in media with the indicated concentrations of serine for 72 hours. Media was replenished every day to avoid serine depletion in lower serine-containing conditions, n=3. (G) Immunoblot assessing PHGDH protein levels in A549 or H1299 cells cultured in media with the indicated concentrations of serine for 24 hours. HSP90 expression is shown as a loading control. Data shown for all panels are means ± SD.

FCCP increases both serine synthesis and proliferation of redox non-responder cells in serine depleted conditions.

(A) Proliferation rate (doublings per day) of redox non-responder cells Calu6 and 8988T cultured with or without serine and the indicated concentration of FCCP for 72 hours, n=3. Increasing doses of FCCP statistically increased the proliferation of Calu6 cells cultured without serine, p<0.001. P-values were calculated by simple linear regression. All doses of FCCP statistically increased the proliferation of 8988T cells cultured without serine but not in a dose-dependent manner, p<0.001. P-values were calculated by unpaired Student’s t-test. (B) Proliferation rate (doublings per day) of redox responder cells MIA PaCa-2 and HCT116 cultured with or without serine and the indicated concentration of FCCP for 72 hours, n=3. (C) Mitochondrial oxygen consumption rate (OCR) of H1299 cells cultured with or without serine and the indicated concentration of FCCP for 24 hours, n=5. Values are averages of three repeat measurements. P-values were calculated by unpaired Student’s t-test, *p<0.05, **p<0.01, ****p<0.001. (D) NAD+/NADH ratio measured in H1299 cells cultured with or without serine and the indicated concentration of FCCP for 24 hours, n=3. P-values were calculated by unpaired Student’s t-test, **p<0.01. (E) Proliferation rate (doublings per day) of redox responder cells H1299 and HCT116 with or without PHGDH overexpression and cultured with or without serine and 2 μM FCCP, as indicated, for 72 hours, n=3. P-values were calculated by two-way ANOVA followed by post-hoc Tukey HSD test, *p<0.05, ***p<0.005, ****p<0.001. (F) Proliferation rate (doublings per day) of redox non-responder cells A549 and Calu6 with or without PHGDH overexpression and cultured with or without serine and 2 μM FCCP for 72 hours, n=3. P-values were calculated by two-way ANOVA followed by post-hoc Tukey HSD test, ***p<0.005, ****p<0.001. (G) Immunoblot assessing PHGDH protein levels in the indicated cell lines without (pLJM1-EGFP) or with (pLJM1-PHGDH) exogenous PHGDH expression. Vinculin expression is shown as a loading control. (H,I) Proliferation rate (doublings per day) of A549 cells (H) and H1299 cells (I) cultured with or without serine, with or without 2 μM FCCP, or the indicated concentration of oligomycin for 72 hours, n=3. P-values were calculated by unpaired Student’s t-test, *p<0.05, ***p<0.005, ****p<0.001. (J) Immunoblot assessing PHGDH protein levels in A549 cells cultured with or without serine and with or without 2 μM FCCP for 24 hours. HSP90 expression is shown as a loading control. (K) Glucose consumption rates of A549 cells cultured with or without serine and 2 μM FCCP for 72 hours, n=3. Data shown for all panels are means ± SD.

Kinetic tracing of labeled carbon from U-13C-glucose into serine following serine withdrawal and FCCP treatment.

(A) Fraction of total serine labeled (M+3 serine) from U-13C-glucose over time in A549 cells cultured without serine, and with or without 2 μΜ FCCP, for 24 hours prior to U-13C-glucose exposure as indicated, n=3. (B) Total serine measured in A549 cells cultured without serine and with or without 2 μΜ FCCP for each time point shown in (A), n=3. (C) Fraction of total serine labeled (M+3 serine) from U-13C-glucose over time in H1299 cells cultured without serine and with or without 2 μΜ FCCP for 24 hours prior to U-13C-glucose exposure as indicated, n=3. (D) Total serine measured in H1299 cells cultured without serine and with or without 2 μΜ FCCP for each time point shown in (C), n=3. (E) Fraction of total serine labeled (M+3 serine) from U-13C-glucose over time in A549 cells cultured without serine and with or without 2 μΜ FCCP and 20 nM rotenone for 24 hours prior to U-13C-glucose exposure as indicated, n=3. (F) Total serine measured in A549 cells cultured without serine and with or without 2 μΜ FCCP and 20 nM rotenone for each time point shown in (E), n=3. All serine measurements were normalized to internal norvaline standard and cell number. Data shown for all panels are means ± SD.

Cytosolic or mitochondrial LbNOX expression does not change the cell NAD+/NADH ratio or proliferation of A549 cells cultured without serine.

(A) Immunoblot for expression of FLAG-tagged cytoLbNOX or mitoLbNOX as indicated. GFP-expressing cells were used as a control. Vinculin expression is shown as a loading control. (B) Relative proliferation rate (doublings per day) of A549 cells expressing either cytoLbNOX or mitoLbNOX and treated with 50 nM rotenone for 72 hours, n=3. P-values were calculated using one-way ANOVA followed by post-hoc Tukey HSD test, ***p<0.005, ****p<0.001. (C) Relative NAD+/NADH ratios of cells expressing either GFP, cytoLbNOX, or mitoLbNOX and cultured with or without serine for 24 hours. NAD+/NADH ratios were normalized to the GFP control cells cultured with serine, n=3. (D) Proliferation rate (doublings per day) of cells expressing either GFP, cytoLbNOX, or mitoLbNOX and cultured with or without serine for 72 hours, n=3. Data shown for all panels are means ± SD.

MDH1 and GOT2 support the increase in serine synthesis observed in H1299 cells following serine withdrawal.

(A) Immunoblots for GOT1 (left), MDH1 (middle), and GOT2 (right) in control cells (NTC, non-targeting control) or cells with knockout (KO) of GOT1, MDH1, or GOT2 as indicated. Vinculin expression is shown as a loading control. (B) Proliferation rate (doublings per day) of the indicated cells from (A) cultured with or without serine for 72 hours, n=3. P-values were calculated by nested ANOVA to compare sensitivity to serine depletion between cell lines, ****p<0.001. (C) Relative NAD+/NADH ratios of the indicated cells cultured with or without serine for 24 hours, n=3. NAD+/NADH ratios were normalized to H1299 NTC cells cultured with serine. P-values were calculated using unpaired Student’s t-test, *p<0.05. (D) Oxygen consumption rate (OCR) of the indicated cells cultured with or without serine for 24 hours, n=6. P-values were calculated by unpaired Student’s t-test, **p<0.01. (E) Serine labeled from U-13C-glucose (M+3) measured over time from the indicated cells cultured with or without serine for 24 hours prior to U-13C-glucose exposure, n=3. Serine levels were normalized to internal standard norvaline and cell number. Data shown for all panels are means ± SD.

Kinetic tracing of labeled carbon from U-13C-glucose and U-13C-glutamine into citrate in cells cultured with or without serine.

(A) Schematic depicting isotopically-labeled carbon contribution to citrate from U-13C-glucose and U-13C-glutamine. Abbreviations: PDH – pyruvate dehydrogenase, CS – citrate synthase, IDH – isocitrate dehydrogenase, SDH – succinate dehydrogenase. (B) Citrate labeled (M+2) from U-13C-glucose over time in Calu6, 8988T, MIA PaCa-2, and HCT116 cells cultured with or without serine for 24 hours prior to U-13C-glucose exposure, n=3. (C-H) Left of each figure panel: Fraction of citrate labeled (M+2) from U-13C-glucose over time in indicated cells cultured with or without serine for 24 hours, n=3. Right of each figure panel: Total citrate levels in the indicated cells cultured with or without serine for 24 hours for each time point shown in left panel, n=3. (I) Citrate labeled (M+4) from U-13C-glutamine over time in A549 and H1299 cells cultured with or without serine for 24 hours, n=3. (J) Citrate labeled (M+5) from U-13C-glutamine over time in A549 and H1299 cells cultured with or without serine for 24 hours, n=3. All citrate measurements were normalized to internal norvaline standard and cell number measured in indicated conditions. Data shown for all panels are means ± SD.

Kinetic tracing of labeled carbon from U-13C-glucose into citrate in cells cultured without serine and in the presence or absence of FCCP and/or rotenone.

(A) Fraction of citrate labeled (M+2) from U-13C-glucose over time in A549 cells cultured without serine for 24 hours, and with or without the addition of 2 μΜ FCCP or 20 nM rotenone as indicated, n=3. (B) Fraction of citrate labeled (M+2) from U-13C-glucose over time in A549 cells cultured without serine for 24 hours, and with or without the addition of 2 μΜ FCCP or 20 nM rotenone as indicated, n=3. (C) Total citrate levels in A549 cells cultured without serine for 24 hours, and with or without the addition of 2 μΜ FCCP or 20 nM rotenone as indicated, n=3. All citrate measurements were normalized to internal norvaline standard and cell number. Data shown for all panels are means ± SD.

Kinetic tracing of labeled carbon from U-13C-glucose into citrate in cells cultured in the presence or absence of lipids.

(A) Fraction of citrate labeled (M+2) from U-13C-glucose over time in A549 and H1299 cells cultured with or without lipids for 24 hours as indicated, n=3. (B) Total citrate levels in A549 and H1299 cells cultured with or without lipids for 24 hours for each time point shown in (A), n=3. (C) Fraction of serine labeled (M+3) from U-13C-glucose over time in A549 and H1299 cells cultured with or without lipids for 24 hours as indicated, n=3. (D) Total serine levels in A549 and H1299 cells cultured with or without lipids for 24 hours for each time point shown in (C), n=3. All citrate and serine measurements were normalized to internal norvaline standard and cell number. Data shown for all panels are means ± SD.

Kinetic tracing of labeled carbon from U-13C-glucose into serine and citrate in cells cultured in the absence of serine, lipids, or both serine and lipids.

(A) Fraction of serine labeled (M+3) from U-13C-glucose over time in A549 and H1299 cells cultured without serine, and with or without lipids for 24 hours as indicated, n=3. (B) Total serine levels in A549 and H1299 cells cultured without serine, and with or without lipids for 24 hours for each time point shown in (A), n=3. (C) Fraction of citrate labeled (M+2) from U-13C-glucose over time in A549 and H1299 cells cultured without lipids, and with or without serine for 24 hours as indicated, n=3. (D) Total citrate levels in A549 and H1299 cells cultured without lipids, and with or without serine for 24 hours for each time point shown in (C), n=3. All citrate and serine measurements were normalized to internal norvaline standard and cell number. Data shown for all panels are means ± SD.