Thermal proteome profiling reveals fructose-1,6-bisphosphate as a phosphate donor to activate phosphoglycerate mutase 1

Abstract

Deep understanding of sugar metabolite-protein interactions should provide implications on sugar metabolic reprogramming in human physiopathology. Although tremendous efforts have been made for determining individual event, global profiling of such interactome remains challenging. Here we describe thermal proteome profiling of glycolytic metabolite fructose-1,6-bisphosphate (FBP)-interacting proteins. Our results reveal a chemical signaling role of FBP which acts as a phosphate donor to activate phosphoglycerate mutase 1 (PGAM1) and contribute an intrapathway feedback for glycolysis and cell proliferation. At molecular level, FBP donates either C1-O-phosphate or C6-O-phosphate to the catalytic histidine of PGAM1 to form 3-phosphate histidine (3-pHis) modification. Importantly, structure-activity relationship studies facilitate the discovery of PGAM1 orthostatic inhibitors which can potentially restrain cancer cell proliferation. Collectively we have profiled a spectrum of FBP interactome, and discovered a unique covalent signaling function of FBP that supports Warburg effect via histidine phosphorylation which inspires the development of pharmacological tools targeting sugar metabolism.

Fig. 1. Thermal proteome profiling (TPP) of FBP-interacting proteins.

a TPP workflow for quantitative chemical proteomic profiling of FBP-interacting proteins in HepG2 cell lysates. b Venn diagram for the datasets of proteins identified by TPP strategy from three biological replicates. c Volcano plot of the identified proteins by TPP strategy from HepG2 cell lysates by FBP treatment and control samples. Dashed lines indicate p =  0.05, 1.5-fold change in three replicates from two-sided t-test. Key hits are highlighted for proteins with increased stability (red), proteins with reduced stability (blue) and proteins without stability change (gray). d Molecular function and subcellular distribution analysis of enriched 66 FBP targets. Statistical differences were determined by a one-sided Fisher’s Exact test. e Functional pathways analysis of enriched 66 FBP targets. Statistical differences were determined by a one-sided Fisher’s Exact test. f MS1 chromatographic peaks of the representative peptides from each protein with calculated 57 °C versus 37 °C ratios for cell lysates treated with either FBP or H₂O (Blue traces represent the chromatographic trace of FBP target at 37 °C; red traces represent the chromatographic trace of FBP target at 57 °C; green lines designate the integration range for each peak quantification).

Fig. 2. Validation of the identified FBP targets.

a CESTA workflow for validation of FBP-interacting proteins. HepG2 cell lysates were treated with FBP or H₂O, then subjected to immunoblotting. b Thermal shift curves of FBP targets with increased thermal stability by immunoblotting in HepG2 cell lysates, including PGAM1, PGM1, PMM2 and PRPS1. Blue curves represent FBP treatment and black curves represent H₂O treatment (control). c Thermal shift curves for FBP targets with reduced thermal stability, including GARS, HDAC2, POLR2A and PDXDC1. d Thermal stability changes of recombinant proteins incubated with FBP or H₂O by Coomassie brilliant blue (CBB) staining (n = 3 independent experiments with similar results). All measurements are presented as mean ± SD for three biological replicates. Source data are provided as a Source Data file.

Fig. 3. PGAM1 is phosphorylated and activated by FBP on its catalytic His11.

a Regulations of recombinant PGAM1, PGAM-H11A, PGM1 and PMM2 enzymatic activities by FBP. b Dissociation constants between FBP and recombinant PGAM1, PGM1 and PMM2, measured by microscale thermophoresis (MST). c FBP phosphorylates PGAM1 to form 3-pHis modification by immunoblotting and Phos-tag SDS-PAGE analysis. d Concentration-dependent PGAM1 histidine phosphorylation by 2,3-BPG, PEP and FBP. e Time-dependent PGAM1 histidine phosphorylation by FBP. f LC-MS/MS spectra of histidine phosphorylation at catalytic site His11 of PGAM1. g Immunoblotting analysis and quantification of 3-pHis modification of PGAM1 WT and H11A mutant by FBP treatment. All measurements are presented as mean ± SD for three (a, c, g) or two (b) biological replicates. n = 3 independent experiments with similar results (c, d, e). Statistical differences were determined by a two-sided Student’s t-test. Source data are provided as a Source Data file.

Fig. 4. Molecular mechanism of FBP-PGAM1 interaction through phosphate transfer.

a LC-MS/MS detection of F6P and F1P generation at different time slots after FBP treatment of purified PGAM1. b Schematic illustration of the phosphate transfer from FBP to PGAM1, generating either F6P upon C1-O-phosphate donation or F1P upon C6-O-phosphate donation. c Molecular docking reveals the hydrogen bonding network between FBP and PGAM1 when it donates either its C1-O-phosphate or its C6-O-phosphate. d Immunoblotting analysis and dissociation constants of PGAM1 mutants PGAM1-R10A, PGAM1-R62A, PGAM1-E89A, PGAM1-Y92A, PGAM1-H186A PGAM1-N188A and PGAM1-3A (R10A, E89A and H186A) with key residues involved in the FBP-PGAM1 binding mutated (n = 3 independent experiments with similar results). e Steady state of PGAM1–FBP complex during the process of phosphate transfer from FBP to PGAM1 His11 after 750 ns MD simulation and the probabilities of hydrogen bonding interactions between FBP and different PGAM1 active site residues during the process of either C1-O-phosphate transfer (up) or C6-O-phosphate transfer (down). Orange represents attractive charges, blue represents conventional hydrogen bonds. Size represents the probability of hydrogen bonding interaction between FBP functional group and amino acid residue within the PGAM1 active site. All measurements are presented as mean ± SD from three biological replicates. Source data are provided as a Source Data file.

Fig. 5. Structure-activity relationship studies of FBP-PGAM1 interaction.

a Chemical structures of endogenous sugar phosphates including FBP, F1P, F6P, M1P, M6P, GBP, G1P, G6P and 2,3-BPG. b Immunoblotting and Phos-tag SDS-PAGE analysis of PGAM1 phosphorylation by bisphosphates FBP, GBP, 2,3-BPG compared with monophosphates F1P, F6P, G1P, G6P, M1P and M6P. c Enzymatic assay of PGAM1 treated by various sugar phosphates. d Design of FBP analogues α-MeFBP, β-MeFBP, 4-dexoyFBP, 1-DMeFBP, 6-DMeFBP and 1,6-DMeFBP based on the molecular insight of FBP-PGAM1 interaction. e Immunoblotting and Phos-tag SDS-PAGE analysis of PGAM1 histidine phosphorylation by FBP and its synthetic analogues. f Enzymatic assay of PGAM1 treated by FBP and its synthetic analogues. g IC₅₀ of PGAM inhibition by synthetic FBP analogues 1-DMeFBP, 6-DMeFBP and 1,6-DMeFBP. All measurements are presented as mean ± SD for three biological replicates. Statistical differences were determined by one-way ANOVA followed by Dunnett’s multiple comparison tests. ****P < 0.0001, n.s. P > 0.5. Source data are provided as a Source Data file.

Fig. 6. Chemical synthesis of FBP analogues.

a Chemical synthesis of α-MeFBP and β-MeFBP. b Chemical synthesis of 4-dexoyFBP. c Chemical synthesis of 1,6-DMeFBP. d Chemical synthesis of 6-DMeFBP. e Chemical synthesis of FBP analogue 1-DMeFBP.

Fig. 7. FBP and its analogues regulate PGAM1 activity through histidine phosphorylation in living cells.

a PGAM1 histidine phosphorylation in living HEK293T cells with overexpressed flag-tagged PGAM1 treated by FBP (10 mM) for 30 min in glucose-free medium. b Intracellular FBP concentration and 2-PG/3-PG ratio in HepG2 cells treated by FBP (10 mM) for 30 min in glucose-free medium. c Lactate production and glucose consumption in sgCTRL or sgPGAM1 HepG2 cells treated by FBP (10 mM) for 30 min in 10 mM glucose medium. d Cell proliferation of sgCTRL or sgPGAM1 HepG2 cells treated by FBP (10 mM) for 24 h in glucose-free medium. e Cell viability of HepG2 cells treated by FBP analogues 1-DMeFBP, 6-DMeFBP and 1,6-DMeFBP (1 mM) for 24 h in high glucose (25 mM) medium respectively. f PGAM1 histidine phosphorylation in living HepG2 cells with overexpressed flag-tagged PGAM1 treated by 1-DMeFBP (1 mM) for 24 h. g Lactate production and glucose consumption in sgCTRL or sgPGAM1 HepG2 cells treated by 1-DMeFBP (1 mM) for 24 h in high glucose medium. h Cell proliferation of sgCTRL or sgPGAM1 HepG2 cells treated by 1-DMeFBP (1 mM) for 24 h in high glucose medium. (i) Schematic model depicting the intrinsic intrapathway positive feedback regulation of glycolysis by FBP-PGAM1 covalent signaling or FBP analogue-PGAM1 inhibition. All measurements are presented as mean ± SD for three biological replicates. Statistical differences were determined by one-way ANOVA followed by Dunnett’s multiple comparison tests (a, b, e, f) or two-sided Student’s t-test (c, d, g, h). Source data are provided as a Source Data file.