Identification of a dithiazoline inhibitor of Escherichia coli L,D-carboxypeptidase A

Abstract

The enzyme L,D-carboxypeptidase A is involved in the recycling of bacterial peptidoglycan and is essential in Escherichia coli during stationary phase. By high-throughput screening, we have identified a dithiazoline inhibitor of the enzyme with a 50% inhibitory concentration of 3 microM. The inhibitor appeared to cause lysis of E. coli during stationary phase, behavior that is similar to a previously described deletion mutant of L,D-carboxypeptidase A (M. F. Templin, A. Ursinus, and J.-V. Holtje, EMBO J. 18:4108-4117, 1999). As much as a one-log drop in CFU in stationary phase was observed upon treatment of E. coli with the inhibitor, and the amount of intracellular tetrapeptide substrate increased by approximately 33%, consistent with inhibition of the enzyme within bacterial cells. Stationary-phase targets such as L,D-carboxypeptidase A are largely underrepresented as targets of the antibiotic armamentarium but provide potential opportunities to interfere with bacterial growth and persistence.

FIG. 1.

Structure of the DTZ inhibitor of LdcA.

FIG. 2.

A. Effect of the LdcA inhibitor DTZ on the growth of E. coli MG1655 cultures. An overnight culture was diluted 1,000-fold with fresh medium and incubated at 37°C with shaking for 2 h. Cells (100 μl) were then added to duplicate Bioscreen wells containing 5 μl of DMSO or DTZ at a final concentration of 50 μM and incubated on the Bioscreen for 24 h. The arrows indicate the time of removal of samples from additional replicate wells for determination of CFU. B. CFU from control and DTZ-treated samples at 8 h (open bars) and at 11 h (gray bars).

FIG. 3.

Effect of delayed addition of DTZ inhibitor on E. coli MG1655 growth curves (A and B) and CFU (C and D). DTZ (open circles, 50 μM; solid circles, 200 μM) or DMSO (squares) was added to wells of the Bioscreen at 4 h (panel A) or 8 h (panel B). A representative of 10 replicate wells is shown. At the indicated times, aliquots were removed from wells to which had compound added at 4 h (panel C) or at 8 h (panel D), for CFU determination. Arrows indicate time of addition of DTZ or DMSO.

FIG. 4.

Effect of LdcA overexpression on E. coli growth curves. (A) Protein expression in response to IPTG. E. coli BL21/pLysS cells harboring either pLdcA (lanes 1 to 4) or pMurF (lanes 5 to 8) were treated with IPTG for 3 h prior to polyacrylamide gel electrophoresis as follows: no IPTG, lanes 1 and 5; 6 μM IPTG, lanes 2 and 6; 25 μM IPTG, lanes 3 and 7; 100 μM IPTG, lanes 4 and 8. The arrows at left and right indicate the LdcA and MurF proteins, respectively. Growth of bacterial cells harboring (B) pLdcA or (C) pMurF was monitored on the Bioscreen. Cells without IPTG and without DTZ treatment, solid circles. Similar curves were obtained for IPTG-containing samples lacking DTZ. Cells treated with DTZ (50 μM) and either no IPTG (squares), 6 μM IPTG (triangles), or 25 μM IPTG (open circles). A representative growth curve for each condition, from duplicate curves generated in two independent experiments, is shown.

FIG. 5.

Growth of S. enterica serovar Typhimurium in the presence of DTZ. Colonies of S. enterica serovar Typhimurium strain ATCC 14028 (A) or E. coli strain MG1655 (B) were inoculated into a tube of LB and grown with shaking for 2 h at 37°C. Aliquots (100 μl) of cells were then added to wells of a Bioscreen plate containing either DMSO (squares) or DTZ at a final concentration of 50 μM (open circles) or 150 μM (solid circles). The OD was monitored for an additional 16 h.