Aspartate deficiency limits peptidoglycan synthesis and sensitizes cells to antibiotics targeting cell wall synthesis in Bacillus subtilis

Abstract

Peptidoglycan synthesis is an important target for antibiotics and relies on intermediates derived from central metabolism. As a result, alterations of metabolism may affect antibiotic sensitivity. An aspB mutant is auxotrophic for aspartate (Asp) and asparagine (Asn) and lyses when grown in Difco sporulation medium (DSM), but not in LB medium. Genetic and physiological studies, supported by amino acid analysis, reveal that cell lysis in DSM results from Asp limitation due to a relatively low Asp and high glutamate (Glu) concentrations, with Glu functioning as a competitive inhibitor of Asp uptake by the major Glu/Asp transporter GltT. Lysis can be specifically suppressed by supplementation with 2,6-diaminopimelate (DAP), which is imported by two different cystine uptake systems. These studies suggest that aspartate limitation depletes the peptidoglycan precursor meso-2,6-diaminopimelate (mDAP), inhibits peptidoglycan synthesis, upregulates the cell envelope stress response mediated by σ^M and eventually leads to cell lysis. Aspartate limitation sensitizes cells to antibiotics targeting late steps of PG synthesis, but not steps prior to the addition of mDAP into the pentapeptide sidechain. This work highlights the ability of perturbations of central metabolism to sensitize cells to peptidoglycan synthesis inhibitors.

Figure 1.

Metabolic pathways relevant to this study. Cofactors such as NAD⁺/NADH are not shown in the reactions for simplicity. Solid lines indicate steps requiring a single enzyme, and dashed lines indicate multiple enzymatic reactions between two compounds. Other processes, such as uptake of cystine or DAP or incorporation of dipicolinic acid into endospore, are indicated with dotted lines. Fosfomycin inhibits the synthesis of UDP-N-acetylmuramic acid (UDP-NAM) from UDP-N-acetylglucosamine (UDP-NAG) by inactivating MurAA. D-cycloserine mimics D-Ala and inhibits both the alanine racemase (Alr) and D-Ala-D-Ala ligase (Ddl), which form D-Ala-D-Ala to be added to complete the pentapeptide side chain. After synthesis at the cytoplasmic membrane, lipid II needs to be flipped to the outside of the membrane by lipid II flippase (the flipping step) and oligomerized into glycan strands (transglycosylation). The glycan strands are then crosslinked by the transpeptidase activity of penicillin binding proteins (PBP) into the mesh-like structure of peptidoglycan (transpeptidation). Vancomycin binds to the D-Ala-D-Ala motif of lipid II and prevent transpeptidation. β-lactam antibiotics such as cefuroxime (CEF) bind to PBPs and prevent transpeptidation. Mg²⁺ can bind to and stabilize the cell envelope even under conditions of reduced peptidoglycan synthesis.

Figure 2.. Dissection of phenotypes caused by ypmB and aspB mutations.

A. Genetic arrangement of the ypmAypmBaspB operon. B. Representative growth curves of strains in liquid LB with or without 1 μg ml⁻¹ moenomycin. C. Colony morphology of strains on DSM plates after 20 hours of incubation at 37°C. D. Representative growth curves of strains in liquid DSM. 1 mM final concentration of IPTG are used to induce P_spac(hy) promoter. All growth curves were measured at least 3 times with similar results. A representative measurement is shown. Strains used were ypmB::mls (HB12259), ΔaspB (HB17401), ypmB::mls P_spac(hy)-ypmB (HB22913), ypmB::mls P_spac(hy)-aspB (HB17062), and aspB::kan P_spac(hy)-aspB (HB17060). Note that indistinguishable phenotypes were observed for the ΔaspB and aspB::kan strains.

Figure 3.. Lysis of the aspB mutant results from depletion of Asp and competition with Glu for GltT. Representative growth curves of aspB mutants in DSM are shown.

A. Effect of supplementation with 10 mM (final concentration) of the following amino acids: Asn, asparagine; Asp, aspartate; Glu, glutamate; Gln, glutamine. B. Effects of mutation of each of the known glutamate/aspartate importers. C. Effects of mutations affecting intracellular Glu synthesis and degradation. D. Effects of mutations in panel C combined with or without gltT. E. Residual concentration of Asp and Glu in DSM medium after growth of 24 hours of indicated bacterial strains, as well as acid hydrolyzed medium. The data is represented as the mean plus and minus one standard error of the mean.

Figure 4.. ysis of the aspB mutant can be suppressed by biochemical or genetic manipulations that favor production of mDAP.

A. Representative growth curves of the aspB mutant in DSM supplemented with or without the following amino acids (final concentration of 10 mM each, except Trp is at 5 mM): DAP, L, L-diaminopimelate; Met, methionine; Lys, lysine; Ile, isoleucine; Thr, threonine; Trp, tryptophan. B. Representative growth curves of mutants in liquid DSM. 1 mM final concentration of IPTG is added to induce P_spac(hy) promoter when present.

Figure 5.. Amino acid and mDAP limitation are physiologically distinct.

A. Growth curve of the aspB mutant growing in DSM supplemented with 1 or 10 mM final concentration of DAP. B. Growth curve of WT and the aspB mutant in DSM supplemented with 10 mM final concentration of DAP, and with or without TCA cycle intermediates (10 mM final concentration). C. Growth curve of mutants in DSM supplemented with 10 mM final concentration of DAP, or 10 mM each of DAP and Glu. D. Growth curve of the aspB mutant in DSM, or MM supplemented with 0.5 mM Asp, 0.5% (w/v) casamino acids, or 0.5 mM Asp and a mixture of 17 amino acids each at concentration of 0.1 mg ml⁻¹. E. Growth rate of aspB mutant in different media. The growth rate is calculated as log₂(OD_600-T2/OD_600-T1)/(T₂-T₁), in which T₁ and T₂ are two time points in the exponential phase of cell growth. When OD₆₀₀<0.3, one doubling is approximated as a two-fold increase in OD₆₀₀ value. The data is represented as the mean plus or minus one standard error of the mean, and statistically significant different samples (Student’s t test, two-tailed P<0.05) are labelled with different letters. All growth curves were measured at least 3 times and yielded similar results. A representative measurement is shown.

Figure 6.. DAP can be imported through cystine uptake systems TcyABC and TcyP.

A. Chemical structure of 2,6-DAP and cystine. B. Growth curve of the aspB mutant growing in DSM with or without supplementation with DAP (1 mM final concentration), Cys (600 μM cystine), and DAP + Cys (1 mM DAP and 600 μM cystine). C. Growth curves of different strains in DSM supplemented with 10 mM DAP. All growth curves were measured at least 3 times and yielded similar results. A representative measurement is shown.

Figure 7.. Limitation of mDAP causes cell wall stress and cell morphology defects.

A. Representative phase contrast microscopic images of WT and the aspB mutant after 12 hours of growth in DSM. Red arrows point to deformed or lysed cells. B. Growth curve of the aspB mutant in DSM with or without supplementation with 10 mM MgSO₄ or 10 mM DAP. C. σ^M activity (relative light unit (RLU)/OD₆₀₀, left axis, closed symbols) and OD₆₀₀ (right axis, open symbols) of WT and the aspB mutant in DSM. The dotted line indicates the time when the aspB mutant starts to lyse, and the dashed line indicates when the luciferase synthesis begins to decrease. Because the luciferase has a short half-life (~4 minutes), OD₆₀₀ and luminescence were measured every 12 minutes to generate near real time gene expression data. D. σ^M activity (relative light unit (RLU)/OD₆₀₀) of the aspB mutant in DSM with or without supplement of 10 mM DAP or 10 mM MgSO₄. All growth curves were measured at least 3 times and yielded similar results. A representative measurement is shown. E. Representative phase contrast microscopic images of WT and the aspB mutant after 24 hours of growth in DSM, with or without supplementing 10 mM DAP or MgSO₄. Red arrows point to deformed or lysed cells. F. Comparison of cell length, width and aspect between WT and the aspB mutant under different growth conditions. n is the number of cells measured per strain per condition. Percentage change of the mean compared to WT is shown above each data plot. The bottom and top of the box are the first and third quartiles, the band inside the box is the second quartile (the median), and the X inside the box is the mean. Whiskers are one standard deviation above or below the mean. Outliers are shown as single dots.

Figure 8.. Limitation of mDAP sensitizes cells to antibiotics targeting late steps in PG synthesis.

A. Sensitivity of WT and the aspB mutant against cefuroxime, vancomycin, D-cycloserine and fosfomycin. Pairwise comparison was made to test if difference between samples are statistically significant (Student’s t test, two-tailed. ***, P<0.01; NS, not significant.) B. Sensitivity of WT and the aspB mutant against cefuroxime when grown in MH, MH+Asp, or MH+Asp+αKG (α-ketoglutarate). C. Sensitivity of WT and the aspB mutant against vancomycin when grown in MH, MH+Asp, or MH+Asp+αKG. Pairwise comparison was made (Student’s t test, two-tailed P<0.05), and samples with statistically significant differences are labelled with different letters. The data is represented as the mean plus or minus one standard error of the mean, with sample number n greater-than or equal to 3.