Selective pharmacologic inhibition of a PASTA kinase increases Listeria monocytogenes susceptibility to β-lactam antibiotics

Abstract

While β-lactam antibiotics are a critical part of the antimicrobial arsenal, they are frequently compromised by various resistance mechanisms, including changes in penicillin binding proteins of the bacterial cell wall. Genetic deletion of the penicillin binding protein and serine/threonine kinase-associated protein (PASTA) kinase in methicillin-resistant Staphylococcus aureus (MRSA) has been shown to restore β-lactam susceptibility. However, the mechanism remains unclear, and whether pharmacologic inhibition would have the same effect is unknown. In this study, we found that deletion or pharmacologic inhibition of the PASTA kinase in Listeria monocytogenes by the nonselective kinase inhibitor staurosporine results in enhanced susceptibility to both aminopenicillin and cephalosporin antibiotics. Resistance to vancomycin, another class of cell wall synthesis inhibitors, or antibiotics that inhibit protein synthesis was unaffected by staurosporine treatment. Phosphorylation assays with purified kinases revealed that staurosporine selectively inhibited the PASTA kinase of L. monocytogenes (PrkA). Importantly, staurosporine did not inhibit a L. monocytogenes kinase without a PASTA domain (Lmo0618) or the PASTA kinase from MRSA (Stk1). Finally, inhibition of PrkA with a more selective kinase inhibitor, AZD5438, similarly led to sensitization of L. monocytogenes to β-lactam antibiotics. Overall, these results suggest that pharmacologic targeting of PASTA kinases can increase the efficacy of β-lactam antibiotics.

FIG 1

ΔprkA mutants are hypersusceptible to β-lactam antibiotics. Cultures of wild-type (wt), Δprk, or complemented ΔprkA L. monocytogenes bacteria grown overnight were back-diluted and treated with 10-fold serial dilutions of ampicillin (A), ceftriaxone (B), cephalexin (C) vancomycin (D), or kanamycin (E). Antibiotic concentrations are in μg/ml. Growth was analyzed for 12 h at 15-min intervals. Data are representative of at least 3 independent repeat experiments.

FIG 2

Staurosporine sensitizes L. monocytogenes to β-lactam antibiotics. (A to E) Cultures of wild-type L. monocytogenes grown overnight were back-diluted and treated with 10-fold serial dilutions of ampicillin (A), ceftriaxone (B), cephalexin (C), vancomycin (D), or kanamycin (E) in the presence or absence of 10 μM staurosporine (str). (F) Cultures of S. aureus grown overnight were back-diluted and treated with ceftriaxone in the presence or absence of 10 μM staurosporine. Antibiotic concentrations are in μg/ml. Growth was analyzed for 12 h at 15-min intervals. Data are representative of at least 3 independent repeat experiments.

FIG 3

Staurosporine inhibits in vitro PrkA phosphorylation in a dose-dependent manner. Autophosphorylation (arrow) and myelin basic protein (MBP) phosphorylation (asterisk) activities were assayed for PrkA (lanes 1 to 5), Lmo0618 (lanes 6 to 9), and SaStk1 (lanes 10 and 11) in the presence or absence of 1 μM, 10 μM, or 100 μM staurosporine. MW, molecular weight (in thousands).

FIG 4

The activity of staurosporine is dependent on PrkA. (A) Cultures of wild-type or ΔprkA L. monocytogenes bacteria grown overnight were back-diluted and treated with 10-fold serial dilutions of ceftriaxone. (B) Cultures of wild-type or ΔprkA L. monocytogenes bacteria grown overnight were back-diluted and treated with 10-fold serial dilutions of ceftriaxone in the presence of 5 μM staurosporine. Antibiotic concentrations are in μg/ml. Growth was analyzed for 12 h at 15-min intervals. Data are representative of at least 3 independent repeat experiments.

FIG 5

The CDK inhibitor AZD5438 also sensitizes L. monocytogenes to β-lactam treatment through inhibition of PrkA. (A) Cultures of wild-type L. monocytogenes grown overnight were back-diluted and treated with 10-fold serial dilutions of ceftriaxone in the presence or absence of 50 μM AZD5438. Antibiotic concentrations are in μg/ml. Growth was analyzed for 12 h at 15-min intervals. (B) Autophosphorylation (arrow) and myelin basic protein (MBP) phosphorylation (asterisk) activities were assayed for PrkA (lanes 1 to 5) and Lmo0618 (lanes 6 to 9) in the presence or absence of 1 μM, 10 μM, or 100 μM AZD5438. (C) Autophosphorylation (arrow) and MBP phosphorylation (asterisk) activities were assayed for PrkA in the presence or absence of 2-fold serial dilutions of AZD5438. Quantification was performed by using a Typhoon imager and ImageQuant software. Data are representative of at least 3 independent repeat experiments. A.U., arbitrary units.