The Widely Used Antimicrobial Triclosan Induces High Levels of Antibiotic Tolerance In Vitro and Reduces Antibiotic Efficacy up to 100-Fold In Vivo

Abstract

The antimicrobial triclosan is used in a wide range of consumer products ranging from toothpaste, cleansers, socks, and baby toys. A bacteriostatic inhibitor of fatty acid synthesis, triclosan is extremely stable and accumulates in the environment. Approximately 75% of adults in the United States have detectable levels of the compound in their urine, with a sizeable fraction of individuals (>10%) having urine concentrations equal to or greater than the minimal inhibitory concentration for Escherichia coli and methicillin-resistant Staphylococcus aureus (MRSA). Previous work has identified connections between defects in fatty acid synthesis and accumulation of the alarmone guanosine tetraphosphate (ppGpp), which has been repeatedly associated with antibiotic tolerance and persistence. Based on these data, we hypothesized that triclosan exposure may inadvertently drive bacteria into a state in which they are able to tolerate normally lethal concentrations of antibiotics. Here we report that clinically relevant concentrations of triclosan increased E. coli and MRSA tolerance to bactericidal antibiotics as much as 10,000-fold in vitro and reduced antibiotic efficacy up to 100-fold in a mouse urinary tract infection model. Genetic analysis indicated that triclosan-mediated antibiotic tolerance requires ppGpp synthesis but is independent of growth. These data highlight an unexpected and certainly unintended consequence of adding high concentrations of antimicrobials in consumer products, supporting an urgent need to reevaluate the costs and benefits of the prophylactic use of triclosan and other bacteriostatic compounds.

Keywords: antimicrobial agents; antimicrobial safety; genetics; urinary tract infection.

FIG 1

Triclosan induces tolerance to multiple antibiotics. E. coli (MG1655) were cultured to an OD₆₀₀ of 0.2, split, and cultured for an additional 30 min with (+T) or without (–T) 200 ng/ml triclosan. The indicated bactericidal antibiotics were then added, and the cells were cultured for an additional 2 to 20 h prior to dilution plating. Each experiment was replicated three independent times, with only representative data shown.

FIG 2

Kinetic analysis of triclosan-induced persistence. E. coli (MG1655) and MRSA (FPR3757) cells were cultured to an OD₆₀₀ of 0.2, split and cultured for an additional 30 min with (black line, closed squares) or without triclosan (gray line, open squares). (A to C) At t = 0, 100 ng/ml (A) or 1,000 ng/ml ciprofloxacin (B) was added to the E. coli cultures, and 50 ng/ml vancomycin was added to the MRSA cultures (C). (D) The relative persistence in the presence of triclosan (CFU+T/CFU–T) was calculated from the 4- and 20-h time points. Values are the means of three independent biological replicates, with error bars representing the SEM. Asterisks represent significant differences between triclosan-treated and nontreated samples calculated using a Student two-tailed t test (*, P < 0.05).

FIG 3

ppGpp is needed for triclosan induced tolerance. (A) Cell viability of MG1655 and ppGpp0 E. coli with (+T) or without (–T) pretreatment with triclosan after challenge with antibiotic. (B) Growth curves of MG1655 (black curve) or ppGpp⁰ (gray curve) in LB with (dashed lines) or without (solid lines) triclosan. (C) Cell viability of MG1655 and ppGpp0 E. coli with (+T) or without (–T) pretreatment with spectinomycin after challenge with antibiotic. Values are means of three independent biological replicates, with error bars representing the SEM. Asterisks represent significant differences between triclosan-treated and nontreated samples calculated using a Student two-tailed t test (*, P < 0.05; **, P < 0.001).

FIG 4

Triclosan reduces ciprofloxacin efficacy up to 100-fold in a mouse UTI model. For each round of the experiment, 15 mice were given water-containing triclosan (100 μg/ml) for 21 days, and 15 control mice received plain water. At 21 days, all mice were infected with E. coli UTI89 (∼5 × 10⁷ CFU). At 24 h postinfection, 10 mice from each group were given 25 mg/kg of ciprofloxacin intraperitoneally. At 48 h postinfection, urine samples (A) were obtained, and then the mice were sacrificed and bladders (B) were harvested. To compare the mice groups, Mann-Whitney U test was used, a P value of <0.05 was considered statistically significant (*, P < 0.05; **, P < 0.005; ***, P < 0.0005; ns, values were not statistically different). The horizontal bar represents the median value. The horizontal broken line represents the limit of detection of viable bacteria. The data are from three independent experiments. (C) Free triclosan levels measured by LC-MS/MS of triclosan in untreated and treated mouse urine. (A total of 45 mice were untreated, and 45 mice were treated; these mice were divided into three independent experiments. Urine samples from three to four mice were pooled for triclosan analysis.) The horizontal broken line represents the limit of quantification of triclosan.