Succinylome analysis reveals the involvement of lysine succinylation in metabolism in pathogenic Mycobacterium tuberculosis

Abstract

Mycobacterium tuberculosis (Mtb), the causative agent of human tuberculosis, remains one of the most prevalent human pathogens and a major cause of mortality worldwide. Metabolic network is a central mediator and defining feature of the pathogenicity of Mtb. Increasing evidence suggests that lysine succinylation dynamically regulates enzymes in carbon metabolism in both bacteria and human cells; however, its extent and function in Mtb remain unexplored. Here, we performed a global succinylome analysis of the virulent Mtb strain H37Rv by using high accuracy nano-LC-MS/MS in combination with the enrichment of succinylated peptides from digested cell lysates and subsequent peptide identification. In total, 1545 lysine succinylation sites on 626 proteins were identified in this pathogen. The identified succinylated proteins are involved in various biological processes and a large proportion of the succinylation sites are present on proteins in the central metabolism pathway. Site-specific mutations showed that succinylation is a negative regulatory modification on the enzymatic activity of acetyl-CoA synthetase. Molecular dynamics simulations demonstrated that succinylation affects the conformational stability of acetyl-CoA synthetase, which is critical for its enzymatic activity. Further functional studies showed that CobB, a sirtuin-like deacetylase in Mtb, functions as a desuccinylase of acetyl-CoA synthetase in in vitro assays. Together, our findings reveal widespread roles for lysine succinylation in regulating metabolism and diverse processes in Mtb. Our data provide a rich resource for functional analyses of lysine succinylation and facilitate the dissection of metabolic networks in this life-threatening pathogen.

Fig. 1.

Anti-lysine succinylation immunoblots showing that different carbon sources altered succinylation profiles of Mtb H37Rv. Coomassie blue staining of protein lysates from A 30 days and B 60 days of cultures. Coomassie blue staining was used for the loading control. Western blotting analysis of lysine succinylation level in protein lysates from C 30 days and D 60 days of cultures with a pan anti-succinyllysine antibody. Specificity of succinylation signals was validated by succinylated or non succinylated BSA assay.

Fig. 2.

Profiling lysine succinylome in Mtb H37Rv. A, Workflow for succinylome analysis of Mtb H37Rv. B, Venn diagram showing the number of identified succinylated peptides. C, Venn diagram showing the number of identified succinylated proteins. D, Distribution of peptide score. E, Distribution of precursor mass deviations.

Fig. 3.

Histogram representations of the distribution of identified succinylated proteins according to their biological processes, molecular functions and cellular localization.

Fig. 4.

Bioinformational analysis of succinylation sites. A, Sequence Logo representation of significant motifs identified by Motif-X software. The motifs with significance of p < 0.000001 are shown. B, Position-specific under- or over-representation of amino acids flanking the succinylation sites. Colors were plotted by using intensity map and represent the log₁₀ of the ratio of frequencies within succinyl-13-mers versus nonsuccinyl-13-mers (blue shows enrichment, yellow shows depletion). C, Distribution of succinylated and nonsuccinylated lysines in protein secondary structures. Probabilities for different secondary structures (α helix, beta-strand and coil) of succinylated lysine were compared with the secondary structure probabilities of nonsuccinylated lysine on all proteins identified in this study. Significance was calculated by Wilcoxon test.

Fig. 5.

The complete interaction network of identified succinylated proteins and their interacting proteins. The succinylated proteins were grouped using functional annotation and interaction network was visualized with Cytoscape. The red node border represents the identified succinylated proteins.

Fig. 6.

Central carbon metabolism network in Mtb H37Rv. The identified succinylated proteins were highlighted in red.

Fig. 7.

Succinylation profiling of Acs of Mtb. A, Schematic illustration of regulation of Acs activity by reversible acetylation and succinylation. B, Identification of the modification sites of Acs. The representation of the corresponding MS/MS spectra of succinylated peptides from Acs annotated with a comprehensive series of b and y fragment ions. Details of succinylation sites were listed in the supplemental Table S1. C, The acetylation/succinylation levels of purified recombinant Acs were determined by immunoblotting using specific anti-acetyl/succinyl lysine antibodies. D, Succinylation of Lys193 and Lys366 affect the enzyme activity of Acs. The Acs and its mutants (Acs-K193R, Acs-K366R, and Acs-K193R/K366R) were expressed in E. coli BL21 (DE3) and the specific activity of Acs was determined. Data are means ± S.D. from three independent assays. E, NAD⁺-dependent deacetylation/desuccinylation of Acs by CobB protein (Rv1151c). Acs was incubated with CobB in the presence of NAD⁺ (0.5 mm) for 10h at 25 °C. Nicotinamide (NAM, 2 mm) was added to the reaction as CobB inhibitor. F, Quantification of acetylated/succinylated Acs in the reactions from E. Each sample was standardized by comparing to the Coomassie blue stained gel. Error bars indicate S.D. of three measurements.

Fig. 8.

Structural modeling of Acs. Final structures of the simulations of Acs were shown. The circular section marks the CoA-binding region A–D. A, Nonsuccinylated (Wt) Acs. Site-directed mutants of B, Acs-K193R; C, Acs-K366R; and D, Acs-K193R/K366R. Acs succinylated at E, K193; F, K366; and G, K193 and K366. H–I, Analysis of the root mean square deviations (RMSDs), root mean square fluctuations (RMSFs) and radius of gyrations (Rgs) for the different Acs systems. (Top) Trajectories of the overall RMSDs of the different Acs systems. (Middle) Radius of gyrations profiles of the different Acs systems. (Bottom) Residue-specific RMSF profiles of the different Acs systems. J–K, Secondary structure evolution of Acs and its mutants. The evolution in secondary structure at each frame was monitored using the dictionary of protein secondary structure (DSSP) algorithm. In the stripes each pixel represents the secondary structure (color-coded) of a residue (185–190 or 355–370, x-dimension) at a given time in simulation (y-dimension). L–M, Time evolution of the solvent accessible surface area (SAS) calculated for the different systems that were simulated.