Accumulation of Succinyl Coenzyme A Perturbs the Methicillin-Resistant Staphylococcus aureus (MRSA) Succinylome and Is Associated with Increased Susceptibility to Beta-Lactam Antibiotics

Abstract

Penicillin binding protein 2a (PBP2a)-dependent resistance to β-lactam antibiotics in methicillin-resistant Staphylococcus aureus (MRSA) is regulated by the activity of the tricarboxylic acid (TCA) cycle via a poorly understood mechanism. We report that mutations in sucC and sucD, but not other TCA cycle enzymes, negatively impact β-lactam resistance without changing PBP2a expression. Increased intracellular levels of succinyl coenzyme A (succinyl-CoA) in the sucC mutant significantly perturbed lysine succinylation in the MRSA proteome. Suppressor mutations in sucA or sucB, responsible for succinyl-CoA biosynthesis, reversed sucC mutant phenotypes. The major autolysin (Atl) was the most succinylated protein in the proteome, and increased Atl succinylation in the sucC mutant was associated with loss of autolytic activity. Although PBP2a and PBP2 were also among the most succinylated proteins in the MRSA proteome, peptidoglycan architecture and cross-linking were unchanged in the sucC mutant. These data reveal that perturbation of the MRSA succinylome impacts two interconnected cell wall phenotypes, leading to repression of autolytic activity and increased susceptibility to β-lactam antibiotics. IMPORTANCEmecA-dependent methicillin resistance in MRSA is subject to regulation by numerous accessory factors involved in cell wall biosynthesis, nucleotide signaling, and central metabolism. Here, we report that mutations in the TCA cycle gene, sucC, increased susceptibility to β-lactam antibiotics and was accompanied by significant accumulation of succinyl-CoA, which in turn perturbed lysine succinylation in the proteome. Although cell wall structure and cross-linking were unchanged, significantly increased succinylation of the major autolysin Atl, which was the most succinylated protein in the proteome, was accompanied by near complete repression of autolytic activity. These findings link central metabolism and levels of succinyl-CoA to the regulation of β-lactam antibiotic resistance in MRSA through succinylome-mediated control of two interlinked cell wall phenotypes. Drug-mediated interference of the SucCD-controlled succinylome may help overcome β-lactam resistance.

Keywords: MRSA; TCA cycle; antibiotic resistance; beta-lactams; succinyl-CoA; succinylome.

FIG 1

Mutation of sucC or sucD increases β-lactam susceptibility in MRSA and impairs growth in MHB. (A) sucCD-encoded succinyl-CoA synthetase catalyzes the conversion of succinyl-CoA to succinate in the TCA cycle. sucAB-encoded α-ketoglutarate dehydrogenase catalyzes conversion of a-ketoglutarate to succinyl-CoA, and sdhAB-encoded succinate dehydrogenase converts succinate to fumarate. (B) Measurement of JE2, NE569, and NE1770 cefoxitin susceptibility by disk diffusion assay. (C) M.I.C.Evaluator measurement of oxacillin MICs for JE2, NE569, and NE1770. (D) Growth of JE2, NE569 (sucC), and NE1770 (sucD) in MHB (no antibiotic supplementation) at 37°C. CFU were enumerated at 1-h intervals for 12 h. The data are the average of three independent experiments, and error bars represent standard deviations. (E) JE2, NE569 (sucC::Tn), and JE2 sucC::Tn transductants 1 and 2 spot-inoculated onto MHA and MHA oxacillin (OX) at 1, 2, 4, and 16 µg/ml. (F) DAR173, NE569, and DAR173 sucC::Tn transductants 1 and 2 spot inoculated onto MHA and MHA OX at 1, 2, 4, and 16 µg/ml. (G) USA300, NE569, and USA300 sucC::Tn transductants 1 and 2 spot inoculated onto MHA and MHA OX at 1, 2, 4, and 16 µg/ml. These assays were repeated three times, and a representative image is shown. (H) Chromosomal organization of the sucCD locus, including the locations of transposon insertions in NE569 (sucC) and NE1770 (sucD). The parts of the sucCD operon carried on psucCD, psucC, and psucD used in complementation experiments are indicated.

FIG 2

sucC suppressor mutation is accompanied by restoration of wild-type colony morphology, oxacillin resistance and growth in MHB and CDMG, but not CDM. (A) JE2, NE569 (sucC), and isolated sucC suppressor 1 grown on MHA for 48 h at 37°C. Red arrow indicates faster growing, more pigmented suppressor mutant of NE569. (B) M.I.C.Evaluator measurement of oxacillin MIC for sucC suppressor 1. Three independent measurements were performed and a representative image is shown. (C to E) Growth of JE2, sucC (NE569), and sucC suppressor 1 in MHB (C), CDMG (D), or CDM (E). Growth was measured by enumerating the number of CFU/ml at 2-h intervals in flask cultures. The data presented are the averages of at least three independent experiments, and error bars represent the standard deviations.

FIG 3

Mutation of sucA, but not sdhA, in the sucC background restores wild-type colony morphology, β-lactam resistance, and growth phenotypes. (A) Colony morphologies of JE2, NE569 (sucC), NE547 (sucA), NE626 (sdhA), sucC sucA, and sucC sdhA strains grown for 24 h on MHA. (B) M.I.C.Evaluator measurement of oxacillin MICs for sucA sucC and sdhA sucC strains. Three independent measurements were performed for each strain, and a representative image is shown. (C) Growth of JE2, NE569 (sucC), NE547 (sucA), and sucA sucC strains. (D) Growth of JE2, NE569, (sucC), NE626 (sdhA), and sdhA sucC strains. Growth experiments were performed in MHB at 37°C, and CFU/ml were enumerated at 1-h intervals for 12 h. All data presented are the average of three independent experiments, and error bars represent the standard deviations.

FIG 4

Mutation of sucC alters the central metabolism in S. aureus. JE2, NE569 (sucC), sucC_comp (NE569 psucCD), and sucC_supp (sucC suppressor 1, which has a SucA Ser₉STOP mutation) strains were grown aerobically in MHB. The cells were harvested in the exponential phase (6 h), and intracellular metabolites associated with the pyruvate node and TCA cycle were analyzed by LC-MS/MS. The metabolites included in the analysis are indicated in the TCA cycle pathway, along with their associated enzymes. The slit arrow indicates a predicted break in the TCA cycle due to mutation of sucC in NE569. n = 3; cps, count per second. The image was created using Biorender.com.

FIG 5

Mutation of sucC perturbs lysine succinylation in the S. aureus proteome. (A) Heatmap depicting the quantifiable succinyl-lysine peptides significantly modulated (ANOVA adjusted P < 0.05) derived from the proteomes of JE2 and NE569 (sucC) collected during early-exponential-, exponential-, and stationary-phase growth. The hierarchical clustering was performed with the Ward method and Euclidean distances. (B) Gene ontology enrichment analysis of succinylated peptides within and shared between heat map clusters. (C) The 40 most succinylated proteins in S. aureus, as indicated by the number of succinylated peptides per protein. (D) Amino acid sequence of Atl highlighting all succinylated lysine residues in gray and lysine residues with significantly increased succinylation in NE569 (sucC) versus JE2 in red.

FIG 6

Mutation of sucC does not affect mecA transcription, PBP2a expression or peptidoglycan structure. (A) Comparison of mecA transcription relative to gyrB measured by LightCycler RT-qPCR in JE2 and NE569 (sucC) grown to exponential phase in BHI or BHI supplemented with 0.5 μg/ml oxacillin. Experiments were repeated at least three times, and standard deviations are shown. Student t test analysis revealed no significant differences between either strain or in the presence of absence of oxacillin. (B) Western blotting of PBP2a in JE2, NE569 (sucC), NE569 psucCD, and MSSA strain 8325-4 (negative control). Total protein was extracted from cells collected during the exponential phase of growth in MHB plus 2% NaCl supplemented with 0.5 µg/ml oxacillin, with the exception of 8325-4, which was grown without oxacillin. A portion (8 µg) of total protein was separated on a 7.5% Tris-glycine gel, transferred to a polyvinylidene difluoride membrane, and probed with anti-PBP2a antibody (1:1,000 dilution), followed by horseradish peroxidase-conjugated protein G (1:2,000 dilution) and colorimetric detection with a BioRad Opti-4CN substrate kit. Three independent experiments were performed, and a representative blot is shown. (C) Relative proportions of cell wall muropeptide fractions based on oligomerization and relative cross-linking efficiency in peptidoglycan extracted from JE2, NE569 (sucC), and sucC suppressor 1 grown to exponential phase in MHB or MHB supplemented with oxacillin at 3 or 32 µg/ml. PG analysis from NE569 (sucC) is shown only at 3 µg/ml oxacillin because 32 µg/ml exceeds its MIC. Each profile shown is a representative of three biological replicates. Significant differences were determined using a Student t test (**, P < 0.01; ***, P < 0.001; ****, P < 0.0001).

FIG 7

Increased succinylation of lysine residues in Atl and Sle1 is associated with reduced autolytic activity. (A) Triton X-100-induced autolysis of JE2, NE569 (sucC), sucC suppressor 1, sucC sucA double mutant, NE569 psucCD, and NE460 (atl) strains. The strains were grown to an OD₆₀₀ of 0.5 in MHB at 37°C before being washed in cold PBS and resuspended in 0.1% Triton X-100 with the OD₆₀₀ adjusted to 1. The OD₆₀₀ was monitored at 30-min intervals, and autolysis was expressed as a percentage of the initial OD₆₀₀. The experiments were repeated at least three times, and error bars represent the standard deviations. (B) Amino acid sequence of Sle1 highlighting the K₂₀₀ lysine residue (within a LysM domain [shaded gray]) that is significantly more succinylated in NE569 (sucC) versus JE2 in red.

FIG 8

Suggested model for succinylome-controlled regulation of two interconnected cell wall-associated phenotypes, namely, autolysis and β-lactam susceptibility, in MRSA. Image created with Biorender.com.