Statins are a family of drugs that lower cholesterol levels by inhibiting 3-hydroxy-3-methylglutaryl-CoA-reductase, a rate-limiting enzyme in the human mevalonate pathway of which cholesterol is the biosynthetic end product.1 Statins also have a range of cholesterol-independent effects, including anti-inflammatory functions and antimicrobial activity. These pleiotropic effects are thought to account for the improved survival observed in statin-treated patients suffering from severe bacterial infections, such as sepsis and pneumonia.2, 3, 4 In order to identify the mechanism involved in the protective effects of statins against infection, research studies focused on the direct effect of statins on bacteria. These studies suggest that statins have bacteriostatic effects on the in vitro growth of clinically important bacterial species, including Staphylococcus aureus and Enterococci,5 Streptococcus pneumoniae and Moraxella catarrhalis,6 and Escherichia coli and Pseudomonas aeruginosa.7 However, the concentrations used in these in vitro studies exceed the concentration detected in human serum during statin therapy,6 suggesting the in vitro bacteriostatic effects of statins are not likely to account for the beneficial outcome of patients suffering from severe bacterial infections.
To date, a single study has examined the effects of statins on bacterial virulence traits: Rosch et al. reported that simvastatin (SIM) could reduce the in vivo attachment of S. pneumoniae to the lung and vascular tissue, but did not affect bacterial toxin production.8 Therefore, our objective was to investigate whether statins could modulate virulence factor behaviour in the human opportunistic pathogen P. aeruginosa. This significant nosocomial pathogen is the most important microorganism associated with chronic respiratory disease in cystic fibrosis (CF) patients.9 P. aeruginosa is also capable of causing serious infections in other sites in the body, including burn wounds, the cornea and urinary tract. The ability of P. aeruginosa to establish such infections is owing to its ability to utilize a range of virulence traits to colonize its host and evade the immune response.10
In this study, P. aeruginosa model strains PAO1 (Holloway et al.11) and PA14 (Liberati et al.12) were cultured at 37 °C in Luria–Bertani broth unless otherwise specified. SIM and lovastatin (LOV) were obtained from Sigma-Aldrich, Dorset, UK, and mevastatin (MEV) was obtained from Calbiochem, Darmstadt, Germany. All statins were resuspended in dimethyl sulfoxide. The inactive prodrug form of each compound, where the lactone ring is intact, was used in all experiments.
To ensure that any effects on virulence factor behaviour occurred independently of growth inhibition, growth in the presence and absence of statins was measured. Bacteria were cultured in 96-well microtitre plates in Mueller–Hinton broth containing a 10-fold dilution series (1 mM–10 nM) of each statin and optical density5 at 600 nm was measured after 24 h incubation. No statin was found to have a significant inhibitory effect on bacterial growth (data not shown). This correlates with a previous observation that statins only decrease the growth of P. aeruginosa at high concentrations.7
The influence of statins on motility, a key factor intrinsically linked to other traits, including biofilm formation13 and quorum sensing,14 was examined. Swarming motility of P. aeruginosa was tested using 0.6% (w/v) Eiken agar (Eiken Chemical, Tokyo, Japan) supplemented with 0.5% (w/v) glucose, while swimming and twitching motility were measured on 0.3% (w/v) and 1% (w/v) agar, respectively. Interestingly, swarming motility of both model strains was decreased by 100 μM concentrations of SIM, LOV and MEV compared with a dimethyl sulfoxide vehicle control (Figure 1a). However, swimming and twitching motility of these strains were not affected by these statin concentrations (data not shown). To further substantiate our finding, the effect of swarming motility of isolates from CF and non-CF patients was investigated. Hundred micromolar of all three statins inhibited swarming motility of both these isolates, indicating the statin effect was not strain-specific (Figure 1b).
Swarming motility is often associated with biofilm formation, a key pathogenic trait in many microorganisms, but particularly in P. aeruginosa as it is associated with chronic infections. Therefore, the effect of statins on attachment, an early stage of biofilm formation, was investigated using 10 and 100 μM SIM. For this, PA14 was incubated at 37 °C for 2 h in 24-well microtitre plates with and without SIM. Plates were washed three times, and attached bacteria were stained using 0.1% (w/v) crystal violet. Excess dye was removed by washing and remaining dye was resuspended in 96% (v/v) ethanol, after which the absorbance at 570 nm was measured to quantify attachment. Hundred micromolar SIM significantly attenuated the attachment of PA14 (Figure 2). Although 10 μM SIM was also found to reduce attachment, it did not have a significant effect. To examine if statins could cause disruption of attached bacteria, PA14 was cultured in 96-well microtitre plates at 37 °C for 8 h, following which unattached cells were removed and fresh media containing 100 μM of SIM was added and re-incubated for 12 h. Attachment was measured as described above. Bacterial attachment was not disrupted by the addition of SIM (data not shown).
In addition to the motility and biofilm virulence assays, the effects of SIM, LOV and MEV on type 3 toxin secretion and quorum-sensing signaling of P. aeruginosa were also examined. Expression of the Type 3 toxin ExoS was measured using a PAO1 exoS-lacZ transcriptional fusion. Production of acylated homoserine lactones by PAO1 and PA14 was examined using the indicator strain Chromobacterium violaceum CV026,15 and expression of the PQS biosynthetic operon in response to statins was examined using a PpqsA-lacZ reporter fusion. However, neither of these virulence factors were altered by 100 μM of SIM, LOV and MEV as all fold changes were lower than 1.1 fold compared with the vehicle control.
Motility and biofilm formation are crucial bacterial virulence factors linked to the establishment of chronic infections and are associated with successful colonization of the lungs of CF patients.16 As well as attenuating these traits, statins have also been shown to lower the levels of P. aeruginosa-induced pro-inflammatory cytokines, such as IL-817 and TNFα,18 in the host and to reduce mucin production in vivo.18 Therefore, the efficacy of using statins in the treatment of chronic P. aeruginosa infection may warrant further investigation. Although circulating statin concentrations are typically lower than the concentrations used in this study, an alternative administrative route, such as inhalation, could lead to higher local concentrations in the lungs and may provide a novel treatment option for the use of statins, particularly in the case of chronic respiratory infection in people with CF. Statins are often administered in the form of a prodrug and are hydrolyzed to an active form in the liver. Interestingly, the prodrug form of each statin yielded the effects described in this study, further supporting inhalation of statins as a potential administrative route. This strategy has recently been used for the antibiotic tobramycin, where inhaled tobramycin powder led to improved lung function in P. aeruginosa-infected CF patients.19, 20 Furthermore, statins were capable of attenuating clinical isolates from CF and non-CF patients to a similar extent, suggesting that statins could also be used in the treatment of non-CF infections.
The mechanism by which statins modulate P. aeruginosa virulence traits remains to be elucidated. While a number of bacteria possess orthologues of the human 3-hydroxy-3-methylglutaryl-CoA-reductase enzyme21 on which statins exert their inhibitory effect, P. aeruginosa does not possess a 3-hydroxy-3-methylglutaryl-CoA-reductase homologous protein. This suggests that the effects of statins on P. aeruginosa is mediated through an alternative novel mechanism and warrants further investigation to elucidate the components involved in this process.
We thank Pat Higgins for excellent technical support and Dr Muireann Ní Chróinín for helpful discussions. This research was supported in part by grants awarded by the European Commission (FP7-KBBE-2012-6, CP-TP-312184; FP7-KBBE-2012-6, 311975; OCEAN.2011-2, 287589; MTKD-CT-2006-042062, O36314), Science Foundation Ireland (07/IN.1/B948; 08/RFP/GEN1295; 08/RFP/GEN1319; 09/RFP/BMT2350), the Department of Agriculture and Food (DAF RSF 06 321; DAF RSF 06 377; FIRM 08/RDC/629), the Irish Research Council for Science, Engineering and Technology (RS/2010/2413; 05/EDIV/FP107), the Health Research Board (RP/2006/271; RP/2007/290; HRA/2009/146), the Environmental Protection Agency (EPA2006-PhD-S-21; EPA2008-PhD-S-2), the Marine Institute (Beaufort award C2CRA 2007/082) and the Higher Education Authority of Ireland (PRTLI3; PRTLI4).