Involvement of pmrAB and phoPQ in polymyxin B adaptation and inducible resistance in non-cystic fibrosis clinical isolates of Pseudomonas aeruginosa.

The emergence of multidrug-resistant gram-negative organisms and the simultaneous lack of new clinically available antimicrobial agents have led to a resurgence of older compounds such as the polymyxins (6). These agents, including polymyxin B and colistin, have highly potent activity against gram-negative organisms, including Pseudomonas aeruginosa, but were previously abandoned due to a reported high incidence of nephrotoxicity and neurotoxicity (12). Resistance to polymyxin B is predominantly associated with decreased uptake into the bacterial cell resulting from a reduced capacity for initial binding (23). Polymyxin B and other polycationic compounds enter the cell via a process known as self-promoted uptake (10, 11). Polymyxin B binds to outer-membrane lipopolysaccharide (LPS) displacing Mg2+ and disrupting the Mg2+ cross bridges between anionic LPS molecules in the outer membrane, thus leading to membrane destabilization. This leads to an increased permeability of the outer membrane, allowing further uptake of the antibiotic. In P. aeruginosa, the ability of polymyxin to permeabilize outer membranes reflects its ability to bind to LPS with higher affinity than the native cross-bridging cation Mg2+ (20, 22). A polymyxin B adaptive resistance phenotype was first reported by Gilleland and Murray in 1976 (9) when the wild-type PAO1 strain was passaged in minimal medium containing low Mg2+ and exposed to increasing concentrations of polymyxin B. Since this time, many studies have focused on the structural basis of adaptive resistance (7, 8). Under various environmental conditions, P. aeruginosa has been found to synthesize different forms of the lipid A component of LPS (5). In particular, under Mg2+-limiting conditions, P. aeruginosa exhibits lipid A modifications, including the addition of aminoarabinose and palmitate. These modifications have been associated with polymyxin B resistance (17, 18). It is well established that two distinct two-component regulators, PhoP-PhoQ and PmrA-PmrB, respond to limiting Mg2+ conditions, resulting in polymyxin B resistance in P. aeruginosa (14-16). Under Mg2+-limiting conditions, PhoP-PhoQ autoregulates the oprH-phoP-phoQ operon (15), and similarly, PmrA-PmrB autoregulates the PA4773-5-pmrAB-PA4778 operon (17). Furthermore, the arnBCADTEF-PA3559 operon (PA3552-PA3559), which is responsible for the addition of aminoarabinose to lipid A (18), is separately regulated by each of these two-component regulatory systems and is upregulated under Mg2+-limiting conditions (17). Both the PA4773-PA4778 and the PA3552-PA3559 operons have shown independent upregulation in response to various cationic antimicrobial peptides in laboratory mutants (18). Mutations in PmrB and the presence of aminoarabinose have been directly associated with polymyxin B resistance (18). During a comparative study of MIC methods for testing polymyxin B and colistin susceptibility, it was observed that 24 of 243 multidrug-resistant clinical isolates of P. aeruginosa demonstrated skipped wells in the broth microdilution method for either polymyxin B, colistin, or both (a “skipped well” is an isolated well showing no growth of bacteria despite the fact that a well with a higher concentration demonstrates growth). According to Clinical and Laboratory Standards Institute (CLSI) guidelines, one skipped well is acceptable and MIC readings should be taken based on the well with the highest antibiotic concentration exhibiting growth (3). The assumption is that a technical error has occurred. The multidrug-resistant isolates described here skipped one to five wells, and bacteria recovered from wells at the higher polymyxin concentrations (i.e., after the skipped wells) also exhibited a polymyxin MIC profile containing skipped wells. The phenomenon of skipped wells has previously been described and was termed cocarde growth (1, 2, 23). In this study, we set out to determine the potential causes for this type of resistance profile in the identified P. aeruginosa clinical isolates.