The widespread use of antibiotics is selecting for a variety of resistance mechanisms that seriously challenge our ability to treat bacterial infections. Resistant bacteria can be selected at the high concentrations of antibiotics used therapeutically, but what role the much lower antibiotic concentrations present in many environments plays in selection remains largely unclear. Here we show using highly sensitive competition experiments that selection of resistant bacteria occurs at extremely low antibiotic concentrations. Thus, for three clinically important antibiotics, drug concentrations up to several hundred-fold below the minimal inhibitory concentration of susceptible bacteria could enrich for resistant bacteria, even when present at a very low initial fraction. We also show that de novo mutants can be selected at sub-MIC concentrations of antibiotics, and we provide a mathematical model predicting how rapidly such mutants would take over in a susceptible population. These results add another dimension to the evolution of resistance and suggest that the low antibiotic concentrations found in many natural environments are important for enrichment and maintenance of resistance in bacterial populations.

How sublethal levels of antibiotics and heavy metals select for clinically important multidrug resistance plasmids is largely unknown. Carriage of plasmids generally confers substantial fitness costs, implying that for the plasmid-carrying bacteria to be maintained in the population, the plasmid cost needs to be balanced by a selective pressure conferred by, for example, antibiotics or heavy metals. We studied the effects of low levels of antibiotics and heavy metals on the selective maintenance of a 220-kbp extended-spectrum β-lactamase (ESBL) plasmid identified in a hospital outbreak of Klebsiella pneumoniae and Escherichia coli. The concentrations of antibiotics and heavy metals required to maintain plasmid-carrying bacteria, the minimal selective concentrations (MSCs), were in all cases below (almost up to 140-fold) the MIC of the plasmid-free susceptible bacteria. This finding indicates that the very low antibiotic and heavy metal levels found in polluted environments and in treated humans and animals might be sufficiently high to maintain multiresistance plasmids. When resistance genes were moved from the plasmid to the chromosome, the MSC decreased, showing that MSC for a specific resistance conditionally depends on genetic context. This finding suggests that a cost-free resistance could be maintained in a population by an infinitesimally low concentration of antibiotic. By studying the effect of combinations of several compounds, it was observed that for certain combinations of drugs each new compound added lowered the minimal selective concentration of the others. This combination effect could be a significant factor in the selection of multidrug resistance plasmids/bacterial clones in complex multidrug environments.

SummaryAminoglycoside resistance in bacteria can be acquired by several mechanisms, including drug modification, target alteration, reduced uptake and increased efflux. Here we demonstrate that increased resistance to the aminoglycosides streptomycin and spectinomycin in Salmonella enterica can be conferred by increased expression of an aminoglycoside adenyl transferase encoded by the cryptic, chromosomally located aadA gene. During growth in rich medium the wild-type strain was susceptible but mutations that impaired electron transport and conferred a small colony variant (SCV) phenotype or growth in glucose/glycerol minimal media resulted in activation of the aadA gene and aminoglycoside resistance. Expression of the aadA gene was positively regulated by the stringent response regulator guanosine penta/tetraphosphate ((p)ppGpp). SCV mutants carrying stop codon mutations in the hemA and ubiA genes showed a streptomycin pseudodependent phenotype, where growth was stimulated by streptomycin. Our data suggest that this phenotype is due to streptomycin-induced readthrough of the stop codons, a resulting increase in HemA/UbiA levels and improved electron transport and growth. Our results demonstrate that environmental and mutational activation of a cryptic resistance gene can confer clinically significant resistance and that a streptomycin-pseudo-dependent phenotype can be generated via a novel mechanism that does not involve the classical rpsL mutations.

To ensure correct antibiotic treatment and reduce the unnecessary use of antibiotics, there is an urgent need for new rapid methods for species identification and determination of antibiotic susceptibility in infectious pathogenic bacteria. We have developed a general method for the rapid identification of the bacterial species causing an infection and the determination of their antibiotic susceptibility profiles. An initial short cultivation step in the absence and presence of different antibiotics was combined with sensitive species-specific padlock probe detection of the bacterial target DNA to allow a determination of growth (i.e., resistance) and no growth (i.e., susceptibility). A proof-of-concept was established for urinary tract infections in which we applied the method to determine the antibiotic susceptibility profiles of Escherichia coli for two drugs with 100% accuracy in 3.5 h. The short assay time from sample to readout enables fast appropriate treatment with effective drugs and minimizes the need to prescribe broad-spectrum antibiotics due to unknown resistance profiles of the treated infection. Overprescription and extensive use of antibiotics have selected for resistant bacteria at an alarmingly rapid rate, and we are now facing one of the greatest medical challenges of our time (1). Today, both the diagnosis of bacterial infections and determination of antibiotic susceptibility profiles (ASP) are slow and tedious processes. As a consequence, a patient might be given an antibiotic that has no effect on infections caused by resistant bacteria. Thus, there is a considerable need for new techniques enabling quick and specific diagnosis along with characterization of an ASP in order to guide correct treatment, reduce the use of broad-spectrum antibiotics, and slow the development of resistance.In the last few decades, we have seen an amazing development of novel molecular methods to detect bacterial pathogens and their resistance genes and resistance mutations (2-5). These new hybridization/PCR-based methods are generally faster and more sensitive than are the classical phenotypic methods, but they also suffer from serious drawbacks that have often reduced their general use. An intrinsic limitation of all genotypic methods that identify resistance mutations or genes is that they detect only the potential for resistance (i.e., presence of a resistance gene/mutation), whereas phenotypic methods detect the realization of susceptibility (i.e., no growth in the presence of antibiotic). For a clinician, the realization of susceptibility measure is far more relevant as a basis for a therapeutic decision.Padlock probes are oligonucleotides with target-specific ends, which upon perfect target recognition can be enzymatically joined (6). Reacted probes can be amplified by rolling circle amplification (RCA). RCA is a linear amplification technique for the replication of DNA circles, such as reacted padlock probes, and the product is a single-stranded DNA concatemer containing around 1,000 copies of a 100-mer t...

vii