Acetylation of metabolic enzymes coordinates carbon source utilization and metabolic flux

Abstract

Lysine acetylation regulates many eukaryotic cellular processes, but its function in prokaryotes is largely unknown. We demonstrated that central metabolism enzymes in Salmonella were acetylated extensively and differentially in response to different carbon sources, concomitantly with changes in cell growth and metabolic flux. The relative activities of key enzymes controlling the direction of glycolysis versus gluconeogenesis and the branching between citrate cycle and glyoxylate bypass were all regulated by acetylation. This modulation is mainly controlled by a pair of lysine acetyltransferase and deacetylase, whose expressions are coordinated with growth status. Reversible acetylation of metabolic enzymes ensure that cells respond environmental changes via promptly sensing cellular energy status and flexibly altering reaction rates or directions. It represents a metabolic regulatory mechanism conserved from bacteria to mammals.

Fig. 1

Acetylation of central metabolic enzymes in S. enterica. (A) Global acetylation of S. enterica enzymes of central metabolism. Colored boxes represent both genetic and growth status with arrows inside to indicate changes in acetylation estimated by counting number of positive hits from MS analyses. Comparisons were made for wild-type (WT) cells grown with either glucose (red box) or citrate (white) as the sole carbon source, and for WT cells (yellow) and Δpat (gray), both grown on LB. Black boxes indicate that acetylated peptide was not detected. (B) SILAC quantification of relative acetylation. Acetylation of WT cells grown with glucose (WT_glu) or citrate (WT_cit), Δpat grown with glucose (pat_glu), and ΔcobB grown with glucose (cobB_glu) or citrate (cobB _cit) were quantified by SILAC. Relative acetylation is presented with the top mean value (n = 3) of each pair set as 1 arbitrarily. ND indicates not detectable. Full names of abbreviations used in this figure are available in supporting online material (SOM) text.

Fig. 2

Growth phenotypes of Δpat and ΔcobB. (A) WT (black), Δpat (blue), ΔcobB (red), or Δpat/ΔcobB (green) strains were grown in minimal medium with indicated concentrations of glucose (left) or citrate (right). O.D. indicates optical density. (B) Growth curves of WT S. enterica in minimal medium containing glucose (red) or citrate (blue) with (▲) or without (●) NAD⁺. (C) In vivo metabolic flux profiles in S. enterica during growth on glucose or citrate. Intracellular flux distribution was determined by ¹³C labeling and GC-MS analysis. Arrows indicate the direction of net fluxes, and their widths are scaled to the flux values. (D) Altered metabolic flux ratio of Δpat and ΔcobB. Flux profiles through glycolysis (represented by v_GapA), gluconeogenesis (v_PckA), glyoxylate bypass (v_AceA), and TCA flux (v_ICDH) were quantitated and used for calculating v_GapA/v_PckA and v_AceA/v_ICDH flux ratio. Data shown are mean values of three independent measurements with SD.

Fig. 3

Regulation of central metabolic enzymes by acetylation. (A) Acetylation of metabolic enzymes expressed in ΔcobB and Δpat strains. His-tagged GapA, AceA, or AceK proteins were over-expressed in the WT, ΔcobB, and Δpat strains and purified to homogeneity. Equal amounts of each protein were used and acetylation was determined. SDS PAGE indicates SDS polyacrylamide gel electrophoresis. (B and C) GapA, AceA, and AceK activities by Pat-mediated acetylation and CobB-mediated deacetylation. His-tagged GapA, AceA, and AceK proteins were purified from WT S. enterica and subjected to in vitro acetylation by Pat or deacetylation by CobB. (B) Reciprocal regulation of glycolytic and gluconeogenic activities of GapA by Pat and CobB. (C) Reciprocal regulation of AceA activity and AceK-controlled ICDH activities by Pat and CobB in vitro. Error bars indicate SD of three measurements.

Fig. 4

Differential transcription of pat and cobB in response to metabolic status and various carbon sources. S. enterica cells were incubated in LB and then washed and transferred into minimal medium containing 50 mM glucose or citrate. Cells were sampled at indicated growth phases. (A) Growth curve (a); pat and cobB mRNA amounts normalized against 16S rRNA (b and c); ratio of pat mRNA/cobB mRNA (d). (B) Changes of Pat and CobB protein concentrations at different growth phases in glucose- or citrate-containing medium.