Correspondence: Arsénio M. Fialho, IBB-Institute for Biotechnology and Bioengineering, Center for Biological and Chemical Engineering, Instituto Superior Técnico, Lisbon 1049-001, Portugal. Tel.: +351 21 8417684; fax: +351 21 8419199; e-mail: firstname.lastname@example.org
Burkholderia cepacia complex (Bcc) bacteria are opportunistic pathogens that cause multiresistant pulmonary infections in patients with cystic fibrosis (CF). In this study, we evaluated the in vitro antimicrobial efficacy of eight unsaturated fatty acids against Burkholderia cenocepacia K56-2, a CF epidemic strain. Docosahexaenoic acid (DHA) was the most active compound. Its action can be either bacteriostatic or bactericidal, depending upon the concentration used. The effect of DHA was also evaluated on two others B. cenocepacia clinical isolates and compared with one representative member of all the 17 Bcc species. To test whether DHA could have a therapeutic potential, we assessed its efficacy using a Galleria mellonella caterpillar model of B. cenocepacia infection. We observed that the treatment of infected larvae with a single dose of DHA (50 mM) caused an increase in the survival rate as well as a reduced bacterial load. Moreover, DHA administration markedly increases the expression profile of the gene encoding the antimicrobial peptide gallerimycin. Our results demonstrate that DHA has in vitro and in vivo antibacterial activity against Bcc microorganisms. These findings provide evidence that DHA may be a useful nutraceutical for the treatment of CF patients with lung infections caused by antibiotic multiresistant Bcc microorganisms.
Bacteria belonging to the Burkholderia cepacia complex (Bcc), a group of 17 closely related species, have emerged as highly problematic opportunistic human pathogens in immunocompromised individuals and in patients suffering from cystic fibrosis (CF) (Mahenthiralingam et al., 2005). Bcc strains posses a wide array of virulence factors that are critical for colonization and disease. The virulence of the Bcc members is variable, and Burkholderia cenocepacia and Burkholderia multivorans are the most common species isolated from the respiratory tract of patients with CF (Drevinek & Mahenthiralingam, 2010). They can spread between patients with CF and are exceptionally resistant to many antimicrobial agents (Mahenthiralingam et al., 2005). In a subset of patients with CF, lung infections with these pathogens lead to declining lung function, with necrotizing pneumonia and a rapidly fatal septicemia termed ‘cepacia syndrome’ (Mahenthiralingam et al., 2005).
In an era of increased antibiotics resistance and difficulties in controlling Burkholderia infections in patients with CF, it is imperative to find new nontoxic antibacterial agents effective against this emerging pathogen. Therefore, in this work, we decided to explore the use of long-chain unsaturated fatty acids (LCUFAs) as anti-Burkholderia agents. The microbicidal activity of selected LCUFAs and their derivatives has been reported on various enveloped viruses (Hilmarsson et al., 2007), parasites (Carballeira, 2008) and pathogenic bacteria such as Pseudomonas aeruginosa, Helicobacter pylori, Staphylococcus aureus and Neisseria gonorrhoeae (Desbois & Smith, 2010). These lipids are found in natural products, human skin and body fluids including respiratory secretions, where they play a role in natural host defense against pathogens (Thormar & Hilmarsson, 2007). They exhibit their antibacterial activities through several mechanisms of action, all of which primarily involve the perturbation of the bacterial cell membrane (Desbois & Smith, 2010).
Various LCUFAs omega-3 and omega-6, such as eicosapentaenoic acid (EPA, 20:5n-3), docosahexaenoic acid (DHA, 22:6n-3), linoleic acid (LA, C18:2n-6) and arachidonic acid (AA, C20:4n-6), have been used in the manufacture of lipid-based formulation for nutritional and antimicrobial applications. Among these, DHA is one of the most effective fatty acid compounds. In addition to its documented antimicrobial and antiviral properties, DHA possesses anti-inflammatory activity and inhibits tumorigenesis (Bougnoux, 1999; Calder, 2006; Kang et al., 2010).
Several studies have reported that patients with CF present a deficiency in essential omega-3 and omega-6 fatty acid metabolism, which lead to a lipid imbalance in plasma phospholipids, characterized by a reduced level of DHA and an increased level of AA (Strandvik, 2010). This observation is corroborated through animal models and research in patients with CF where the oral administration of DHA corrects this lipid imbalance and ameliorates the various CF pathological manifestations (Mimoun et al., 2009; Olveira et al., 2010). Moreover, Tiesset et al., 2009 demonstrated that an oral supplementation with DHA could also improve the outcome of pulmonary P. aeruginosa infection in a mouse model of CF. This result corroborates the in vitro studies by Martinez et al., 2009, in which a synergistic antibacterial activity of DHA and lysozyme against a P. aeruginosa strain isolated from the lungs of a patients with CF was demonstrated. Altogether, these results suggest that the administration of DHA affords many benefits to patients with CF, including its antimicrobial action against CF-related opportunistic pathogens.
In view of these findings, we sought to investigate whether LCUFAs including DHA have antimicrobial properties against Burkholderia clinical isolates and therefore might be useful in the treatment of chronic infection in patients with CF caused by this pathogen.
Materials and methods
Bacterial strains, insects and growth conditions
The 19 Bcc isolates used in this study are described in Table 1. Galleria mellonella larvae were reared on a pollen grains diet at 25 °C in darkness. Larvae weighing 250 ± 25 mg were used.
Table 1. Bcc isolates used in this study
Other strain designation
Origin and location
CF-e, strain that spread epidemically among patient with CF.
J. J. LiPuma
J. J. LiPuma
CF-e patient, UK
CF-e patient, Canada
J. J. LiPuma
CF patient, USA
CF patient, Belgium
D. P. Speert
CF patient, USA
J. J. LiPuma
CF patient, USA
J. J. LiPuma
J. J. LiPuma
CF patient, USA
J. J. LiPuma
J. J. LiPuma
Surface soil, Thailand
CF patient, Italy
CF patient, USA
CF patient, Belgium
CF patient, USA
CF patient, USA
CF patient, Belgium
Forest soil, Trinidad
Bacterial overnight cultures were inoculated in 96-well plates with either Luria–Bertani (LB) broth (Conda, Pronadisa) or Müeller–Hinton (MH) (Difco) broth, at 37 °C with orbital agitation (180 r.p.m.). The fatty acids used were purchased from Sigma–Aldrich. Stock solutions of fatty acids (750 mM) were made in ethanol (95%).
Antimicrobial screening of fatty acids
A total of eight LCUFAs were used to evaluate the growth inhibition produced in a liquid culture of B. cenocepacia K56-2. The bacterium was cultured in 96-well microplates with an initial OD640 nm of 0.1, in the presence of each fatty acid at 20 mM. Plates were incubated at 37 °C for 24 h under aerobic conditions, and OD640 nm was followed during the growth, using a microplate reader (Versamax; Molecular Devices). The percentage of inhibition was determined as [(OD640 nm K56-2 − OD640 nm K56-2+fattyacid)/OD640 nm K56-2 × 100)].
Effect of DHA on the growth of B. cenocepacia
Burkholderia cenocepacia K56-2 was grown in a microplate containing LB medium supplemented with different concentrations of DHA (ranging from 2 to 100 mM). Plates were incubated at 37 °C for 24 h under aerobic conditions and OD640 nm and viability were followed during the growth, using a plate reader and determining colony-forming units (CFU), respectively. For CFUs determination, 10 μL of each sample was serially diluted in 0.9% NaCl, plated on LB agar and incubated for 24 h at 37 °C. A negative control was performed using the solvent (ethanol) utilized to solubilize the DHA.
In this study, the in vitro evaluation of the antimicrobial activity of DHA (at a 50 mM concentration) was extended to one representative isolate of each of the 17 Bcc species. In addition, we also included two additional clinical isolates (J2315, AU1054) belonging to the B. cenocepacia species.
Determination of minimum inhibitory concentration (MIC)
The MIC was determined by broth microdilution method recommended by the NCCLS, 1997. Burkholderia cenocepacia K56-2 overnight liquid cultures grown in LB medium at 37 °C were harvested by centrifugation and then resuspended in MH broth (Difco) and diluted to a standardized culture OD640 nm of 0.11. A 96-well plate was inoculated with 190 μL of this cell suspension per well containing 10 μL of DHA in a range of 50–1000 mM (DHA solutions were diluted in MH medium from a stock solution). The microplates were incubated for 24 h at 37 °C, and the OD640 nm was determined using a microplate reader (Versamax; Molecular Devices). The MIC value was achieved as the lowest DHA concentration where no growth was registered (initial OD640 nm). Positive (without DHA) and negative (uninoculated) controls were carried out. Results are expressed as mean values of three independent determinations.
Determination of bacterial hydrophobicity
The cell surface hydrophobicity of Bcc isolates was assessed by measuring the bacterial adhesion to hydrocarbon (BATH), based on the method proposed by Rosenberg et al., 1980, using n-hexadecane as hydrocarbon. Briefly, cells’ growth overnight was harvested by centrifugation, washed twice with phosphate-buffered saline (PBS) and resuspended in a volume of PBS calculated to obtain an OD640 nm of 0.6. Bacterial suspensions (1.5 mL) were mixed with 500 μL n-hexadecane (Sigma–Aldrich) in test tubes, vortexed for 20 s and the phases were allowed to separate for 30 min. After this time, the OD640 nm of the aqueous phase was measured. Results are median values of three independent experiments and were expressed as percentage of hydrophobicity: BATH (%) = (1 − OD640 nm aqueous phase/OD640 nm initial cell suspension)/100)].
Galleria mellonella killing assay and CFU count of B. cenocepacia
Galleria mellonella killing assays were based on the method previous described (Seed & Dennis, 2008). A microsyringe was used to inject 3.5 μL of bacterial suspension (approximately 20 CFU) into each caterpillar via the last left proleg. Following injection, larvae were placed in glass Petri dishes and stored in the dark at 37 °C. For each condition, we used 10 larvae to follow the larval survival over a period of 5 days. Caterpillars were considered dead when they displayed no movement in response to touch. Control larvae were injected with 0.9% NaCl. To determine whether DHA confers a protective effect to Burkholderia-infected larvae, a single dose of DHA (50 mM) was administrated 6 h postinfection.
To determine intracellular bacterial numbers, hemolymph was obtained from three infected larvae by puncturing the larval abdomen with a sterile needle. The out-flowing hemolymph was immediately transferred into a sterile Eppendorf tube containing a few crystals of phenylthiourea to prevent melanization. Then, hemolymph was serially diluted in 0.9% NaCl and plated on LB agar. Colonies were counted after incubation at 37 °C for 24 h.
Quantitative real-time RT–PCR
Three larvae per treatment for each time point (10 and 21 h postinfection) were cryopreserved, sliced and homogenized in 1 mL of Trizol reagent (Sigma–Aldrich). Whole animal RNA was extracted according to the manufacturer's protocol. RNA was treated with Turbo DNase (Ambio, Applied Biosystems) according to manufacturer's recommendations and quantified spectrophotometrically (NanoDrop ND-1000). Quantitative real-time reverse transcription PCR (RT–PCR) was performed with the 7500 real-time PCR system (Applied Biosystems), according to the protocols of the manufacturer. Briefly, cDNA was synthesized from 200 ng of total RNA using Taqman kit (Roche, Applied Biosystems) and then analyzed with Power SYBR Green master mix (Applied Biosystems), using primers to amplify the genes encoding antimicrobial peptides: gallerimycin (Altincicek & Vilcinskas, 2006), galliomycin (Wojda et al., 2009), inducible metalloproteinase inhibitor (IMPI) (Altincicek & Vilcinskas, 2006), lysozyme (Altincicek & Vilcinskas, 2006) and the housekeeping gene actin (Altincicek & Vilcinskas, 2006). All samples were analyzed in triplicate, and the amount of mRNA detected normalized to control actin mRNA values. Relative quantification of genes expression was calculated by using the ∆∆CT method (Livak & Schmittgen, 2001).
All experiments were performed a minimum of three times. Relative comparisons were done between corrected values with anova test for significance. A P-value < 0.05 was considered statistically significant.
Effect of long-chain unsaturated fatty acids on growth inhibition of B. cenocepacia K56-2
The antibacterial activity of eight LCUFAs with various carboxyl lengths (18 carbons or more) was quantitatively evaluated against B. cenocepacia K56-2. The growth was monitored for 24 h at 37 °C in the presence of 20 mM of each LCUFA by measuring the OD640 nm. Of the eight lipids tested, only 3 [Petroselinic acid 18:1 (n-6), DHA 22:6 (n-3) and nervonic acid 24:1 (n-9)] showed growth inhibition effects. DHA caused the greatest growth inhibition (50–60%) compared with the control (Fig. 1), so it was selected for further studies. A control assay with 2.7% ethanol had no effect on the growth of B. cenocepacia K56-2 (Fig. 1).
In vitro antimicrobial effect of DHA against B. cenocepacia K56-2
DHA against B. cenocepacia K56-2 recorded a MIC range of 40–50 mM, determined after 24 h by the reference broth microdilution method. In addition, we have determined the effect of various concentrations (2, 5, 10, 20, 50 and 100 mM) of DHA on the growth of B. cenocepacia K56-2 after 24 h of exposure. As shown in Fig. 2a, DHA exhibits a concentration-dependent bacteriostatic activity. Upon exposure to DHA, B. cenocepacia K56-2 cells aggregated and formed clusters (Fig. 2b). Moreover, the highest concentrations of DHA screened (50 and 100 mM) caused not only a significant growth inhibition (80–90%) but also death of B. cenocepacia K56-2 cells (8 log10-unit reduction of viable B. cenocepacia cells) (Fig. 2c). Therefore, these results indicate that DHA has a bacteriostatic/killing activity against B. cenocepacia K56-2.
In vitro antimicrobial activity of DHA against members of the Bcc
To further confirm the in vitro antibacterial effect of DHA (50 mM), we extended our analysis to one representative strain of each of the 17 Bcc species. In addition, we also included two additional clinical isolates (J2315, AU1054) belonging to the B. cenocepacia species. Figure 3 demonstrates that although there is variation in the extent of the antibiotic effect observed, DHA significantly reduces the growth of all Bcc strains studied (40–100% inhibition). Burkholderia cenocepacia J2315, Burkholderia stabilis LMG14294 and Burkholderia anthinia AU1293 were particularly susceptible to DHA, while Burkholderia vietnamiensis PC259, Burkholderia pyrrocinia BC011 and Burkholderia lata 383 possessed the highest levels of resistance (Fig. 3). To determine whether the observed sensitivity/tolerance of the Bcc isolates to DHA was because of hydrophobic interactions with the bacterial cell membrane, the BATH assay was used (Rosenberg et al., 1980). As shown in Fig. 3, a direct relationship was not observed between the degree of cell surface hydrophobicity and DHA sensitivity/tolerance.
Evaluation of the in vivo antimicrobial action of DHA against B. cenocepacia K56-2
The in vivo antimicrobial efficacy of DHA against B. cenocepacia was examined in a G. mellonella caterpillar model system. To mimic a therapy with DHA, larvae were inoculated with a lethal dose of B. cenocepacia K56-2 followed by the administration of a single dose of DHA (50 mM: 190 mg kg−1), given 6 h after infection. The dose of DHA used was within the limits of dosage used in animal studies (Willumsen et al., 1993; Mizota et al., 2001).
As shown in Fig. 4a, over a period of 5 days, the treatment with DHA, compared with an infected control group, prolonged the survival of G. mellonella caterpillars (P <0.01). Uninfected larvae were also inoculated with 50 mM of DHA, and 100% survival was observed after 5 days (Fig. 4a). We also monitored the growth of B. cenocepacia K56-2 in the hemolymph of infected larvae over a period of 24 h postinfection. We observed a reduced bacterial load (2 log10-unit reduction; P <0.01) in treated group (administration of DHA) compared with control group (Fig. 4b).
Finally, by using quantitative real-time RT–PCR, we determined the expression patterns of four immune-related G. mellonella genes encoding antimicrobial peptides at 10 and 21 h postinfection. We intended to determine whether DHA stimulates the immune response of G. mellonella, thereby enhancing the host defense against B. cenocepacia infection. As shown in Fig. 4c, at 10 h postinfection, the four immune-related genes were either inactive (uninfected larvae treated with DHA) or expressed at very low levels (infected larvae with and without DHA treatment). However, at 21 h postinfection, the mRNAs of gallerimycin, IMPI and lysozyme were found to be induced either in the infected larvae-treated or untreated with DHA. Nevertheless, the total amount of mRNA encoding gallerimycin reached its highest value in the DHA-treated larvae (120-fold). The housekeeping gene actin was used as a reference for relative quantification of mRNA (Fig. 4c).
The antimicrobial property of essential LCUFAs and their derivatives has been recognized for many years (Desbois & Smith, 2010). In the present study, we have investigated for the first time the in vitro antimicrobial properties of eight different LCUFAs against B. cenocepacia, an emerging pathogen in patients with CF. We observed that of the LCUFAs tested, only three fatty acids have anti-Burkholderia activity, namely petroselinic acid, DHA and nervonic acid. The differences in growth inhibition most likely correlates with the geometry and position of the carbon-carbon double bonds as well as the carbon chain lengths of the LCUFAs tested (Huang et al., 2010).
DHA showed the highest level of growth inhibition, albeit with moderate efficacy (millimolar concentrations) (Fig. 1). This is consistent with previous published studies that indicate that DHA exhibits a broad spectrum of in vitro antibacterial activity against various Gram-positive and Gram-negative pathogenic bacteria (Shin et al., 2007; Martinez et al., 2009). The mechanism of action of DHA against B. cenocepacia K56-2 is not known. Possibly, as described for other LCUFAs, DHA primarily affects the integrity of the bacterial plasma membrane, thereby leading to cell damage and cell death (Desbois & Smith, 2010). There are, however, some differences in DHA activity between cell types, whereby DHA has a higher antimicrobial activity against Gram-positive bacteria, which again, probably is a result of structural differences in the cell wall and/or plasma membrane (Shin et al., 2007).
To further extend and confirm the in vitro anti-Burkholderia activity of DHA, a panel of 19 isolates representing all 17 Bcc species was tested. Our results indicated that all Bcc isolates were inhibited by 50 mM DHA, although significant differences in the levels of growth inhibition were observed across all species (Fig. 3). No obvious link was observed between DHA and antibiotic or biocide resistance as previously published (Nzula et al., 2002; Rose et al., 2009). The clinical isolate B. cenocepacia J2315 was found to be more susceptible to DHA than other B. cenocepacia strains, yet J2315 was the strain most resistance to meropenem (Nzula et al., 2002). Conversely, strain B. vietnamiensis PC259 showed the highest level of DHA resistance (Fig. 3), although all strains of B. vietnamiensis were more susceptible to ceftazidime and chloramphenicol than other Bcc species (Nzula et al., 2002).
Similarly, no direct relationship was observed between DHA susceptibility and cell surface hydrophobic properties. Two of the three Bcc strains that were particularly susceptible to DHA (B. stabilis LMG14294 and B. anthinia AU1293) possessed the lowest levels of cell surface hydrophobicity. In addition, the three B. cenocepacia isolates tested have shown identical DHA susceptibility but significant differences in cell surface hydrophobicity (Fig. 3). These findings suggest that the resistance to DHA is not directly correlated with the degree of cell surface hydrophobicity, meaning that other particular cell targets could be relevant. In this regard, Zheng et al. (2005) demonstrated that LCUFAs are selective inhibitors of the Type I fatty acid synthase (FabI), concluding that their antibacterial activity is because of the inhibition of fatty acid biosynthesis.
Martinez et al., 2009 have demonstrated a potent synergistic activity of DHA with lysozyme against a P. aeruginosa strain isolated from the lungs of a patient with CF. Furthermore, the authors highlighted the relevance of this synergistic action and its translation to the clinic as an antipseudomonal therapy for patients with CF. With respect to this finding, we have analyzed whether DHA (50 mM) in combination with two antibacterial proteins [lysozyme (500 mg L−1) and lactoferrin (500 mg L−1)] and one antibiotic (ciprofloxacin at a subinhibitory concentration of 1 mg L−1) can act synergistically, thereby increasing its antimicrobial effectiveness against B. cenocepacia. However, the coaddition of DHA with these three antibacterial molecules does not act synergistically to augment their effects as anti-Burkholderia agents (results not shown).
To assess the in vivo efficacy of DHA against the Bcc, we used a G. mellonella caterpillar model of infection. We conclude that a single administration of 50 mM DHA induced protection against B. cenocepacia K56-2 infection. Additionally, treatment with DHA enhanced the immune response of the larvae, thereby suggesting an intrinsic ability of DHA to modulate the response of G. mellonella to B. cenocepacia infection (Fig. 4). Thus, our data suggest that DHA in vivo exerts both a direct antibacterial activity and an indirect effect via changes in the host immune system.
In summary, our results demonstrate for the first time that the fatty acid DHA has in vitro and in vivo antibacterial activity against Bcc strains. DHA has previously been administrated to humans and animal models in a wide range of daily doses. Furthermore, as reported by Calviello et al., 1997, even high doses of DHA (360 mg per kg body weight day−1) do not cause cytotoxicity or other undesirable effects. Taken together, our preliminary results demonstrate the effectiveness of DHA against B. cenocepacia strains and infections and provide a mandate for further experimentation with either additional animal studies or human clinical trials.
This work was supported by FEDER and Fundação para a Ciência e a Tecnologia (FCT), Portugal (grants: PTDC/QUI/67925/2006, PTDC/BIA-MIC/71453/2006 and PTDC/EBB-BIO/100326/2008) and PhD fellowships to D.M.-H. and N.B. We thank Dr Raquel Seruca from IPATIMUP, University of Porto, Portugal, for her valuable contribution to the present work. We acknowledge Prof. Gerd Döring from University of Tübingen in Germany, Prof. John LiPuma from University of Michigan in USA and Prof. David Speert from University of British Columbia in Canada, who kindly provided Burkholderia strains.