Serine is a natural ligand and allosteric activator of pyruvate kinase M2

Abstract

Cancer cells exhibit several unique metabolic phenotypes that are critical for cell growth and proliferation. Specifically, they overexpress the M2 isoform of the tightly regulated enzyme pyruvate kinase (PKM2), which controls glycolytic flux, and are highly dependent on de novo biosynthesis of serine and glycine. Here we describe a new rheostat-like mechanistic relationship between PKM2 activity and serine biosynthesis. We show that serine can bind to and activate human PKM2, and that PKM2 activity in cells is reduced in response to serine deprivation. This reduction in PKM2 activity shifts cells to a fuel-efficient mode in which more pyruvate is diverted to the mitochondria and more glucose-derived carbon is channelled into serine biosynthesis to support cell proliferation.

Figure 1. Characterisation of PKM1/2-silenced HCT116 cells

a, Protein levels of PKM1 and PKM2 in the indicated cell lines were detected by Western blot. Actin was used as a loading control. b, Quantified intracellular metabolite concentrations and the uptake or secretion of extracellular metabolites in control and shPKM HCT116 cells. For extracellular metabolites, the dashed line indicates the initial levels in the medium, while the graph represents the levels after 24 hours incubation. c, The proliferation rate of the indicated cell lines was measured by an Alamar Blue assay. d, Intracellular ATP levels of the indicated cells normalised to protein concentration in the cell extracts. All results are from 3 independent cultures and are presented as mean ± s.e.m. * = P< 0.05, ** = P<0.01, *** = P< 0.001.

Figure 2. The effect of PKM1/2 silencing on glycolytic flux

HCT116 shCntrl and shPKM cells were incubated with U-¹³C-glucose and the abundance of the main glucose-derived isotopomer of the indicated metabolites was analysed at the indicated time points. The cumulative intensities of each labelled metabolite analysed in each cell line are presented in blue (shCntrl) or red (shPKM). The white and black circles under each metabolite illustrate ¹²C- and ¹³C-labelling, respectively. All metabolic quantifications were performed from 3 independent cultures and presented as mean ± s.e.m. 3-PG, 3-phosphoglycerate; PEP, phosphoenolpyruvate; PKM, pyruvate kinase; PDH, pyruvate dehydrogenase; LDH, lactate dehydrogenase; TCA, tricarboxylic acid. P values at 4 hrs are shown when significant.

Figure 3. Serine and glycine deprivation changes glucose metabolism

a, Parental HCT116 cells were incubated for 12 hours with U-¹³C-glucose in the presence (+SG) or absence (−SG) of serine and glycine. The abundance of the main glucose-derived isotopomer (black bars) and the unlabelled fraction (white bars) of the indicated metabolites were analysed. The circles represent the isotopomer distribution of each individual metabolite. b, PK activity is represented as the ratio between glucose-derived (¹³C₃-) pyruvate and PEP, 30 minutes after labelling with U-¹³C-glucose in the presence or absence of serine and glycine in the indicated cell lines. Results were normalised to shCntrl +SG. PEP, phosphoenolpyruvate. All results are from 3 independent cultures and they are presented as mean ± s.e.m. * = P< 0.05, ** = P<0.01, *** = P< 0.001.

Figure 4. Serine is an allosteric activator of PKM2

a, In vitro activity of recombinant human PKM2 was analysed in the presence of increasing concentrations of FBP (●), serine (○) or glycine (▲).b, Serine binding to PKM2 wild-type or H464A mutant was measured by isothermal titration calorimetry. The K_d of serine for wild-type PKM2 was measured as 200 μM. No serine binding to H464A PKM2 was detected. c, The initial PKM2 reaction (V₀) was measured at different PEP concentrations in the presence of 50 μM FBP (●), 100 mM Serine (○) or vehicle (▲). K_m values for PEP were determined as 1.9 mM, 0.81 mM and 0.19 mM in the presence of vehicle, serine and FBP, respectively. For a and c, results are from 3 independent experiments and are presented as mean ± s.e.m. FBP, fructose 1,6-bisphosphate. d, 2.3Å 2Fo-Fc map (purple) contoured at 1σ for the final, refined, structure of L-ser (orange) bound to PKM2 (green). The side chain of His464, which was subsequently mutated to alanine, is shown in yellow. A detailed schematic representation of the L-ser interactions is shown in Supplementary Fig. 9d. e-f, In vitro activity of PKM2 mutants H464A (e) and S437Y (f) was analysed in the presence of increasing concentrations of FBP (●) or Serine (○). g-h, In vitro activity of PKM1 (g) and PKLR (h) was analysed in the presence of increasing concentrations of FBP (●) or Serine (○). For a and e-h, the signal was normalised to controls containing no enzyme. The basal activity of the studied PK in each panel in the presence of vehicle control is indicated by the dotted line. Data are presented as mean ± s.e.m. of duplicate determinations and are representative of three independent experiments.