Macrolides decrease the minimal inhibitory concentration of anti-pseudomonal agents against Pseudomonas aeruginosa from cystic fibrosis patients in biofilm
© Lutz et al.; licensee BioMed Central Ltd. 2012
Received: 30 January 2012
Accepted: 31 August 2012
Published: 8 September 2012
Biofilm production is an important mechanism for bacterial survival and its association with antimicrobial resistance represents a challenge for the patient treatment. In this study we evaluated the in vitro action of macrolides in combination with anti-pseudomonal agents on biofilm-grown Pseudomonas aeruginosa recovered from cystic fibrosis (CF) patients.
A total of 64 isolates were analysed. The biofilm inhibitory concentration (BIC) results were consistently higher than those obtained by the conventional method, minimal inhibitory concentration, (MIC) for most anti-pseudomonal agents tested (ceftazidime: P = 0.001, tobramycin: P = 0.001, imipenem: P < 0.001, meropenem: P = 0.005). When macrolides were associated with the anti-pseudomonal agents, the BIC values were reduced significantly for ceftazidime (P < 0.001) and tobramycin (P < 0.001), regardless the concentration of macrolides. Strong inhibitory quotient was observed when azithromycin at 8 mg/L was associated with all anti-pseudomonal agents tested in biofilm conditions.
P. aeruginosa from CF patients within biofilms are highly resistant to antibiotics but macrolides proved to augment the in vitro activity of anti-pseudomonal agents.
KeywordsAntimicrobial susceptibility test Biofilm Azithromycin Clarithromycin
The main cause of morbidity and mortality in cystic fibrosis (CF) is chronic lung disease caused by a vicious cycle of infection and inflammation which leads to progressive deterioration of pulmonary function, respiratory failure, and death . Pseudomonas aeruginosa is the main bacteria associated with pulmonary disease in CF. In vivo and in vitro evidence suggests that P. aeruginosa produce biofilm within the airways of chronic CF pulmonary infection patients,[2–5] which is a protective barrier around the bacterial cells and limits exposure to oxidative radicals, antibiotics, and phagocytes . Bacterial biofilms play a relevant role in persistent infections, which are rarely eradicated with antimicrobial therapy .
Despite the evidence of P. aeruginosa grown in the airways of CF patients in biofilm form, the susceptibility profile of the bacterium is usually evaluated, in vitro, in the planktonic state. However, the planktonic susceptibility profile may not represent the actual susceptibility of the bacteria . To overcome the potential shortfalls of traditional (planktonic) microbiological methods to evaluate susceptibility, biofilm models have been proposed to access susceptibility of P. aeruginosa in vitro.
Macrolide antibiotics are being evaluated for the treatment of chronic lung inflammatory diseases, including diffuse panbronchiolitis, CF, chronic obstructive pulmonary disease, and asthma. Although macrolides have no antimicrobial activity against P. aeruginosa at therapeutic concentrations, there is great interest in the evaluation of treatments of CF patients with these antibiotics, at least as complementary therapy [9–11]. Anti-inflammatory activity of macrolides has been showed in many studies, including clinical trials [12–17]. Macrolides have also proved to present potential effects on inhibition of bacterial biofilm with reduction of bacterial virulence factor when used in sub-inhibitory concentrations . In the present study, we evaluated the in vitro action of macrolides in combination with anti-pseudomonal agents on biofilm-grown P. aeruginosa recovered from CF patients.
Anti-pseudomonal agents in vitro activity against P. aeruginosa (n = 64) in planktonic and in biofilm conditions
Range MIC/ BIC
No. of isolates inhibited by different MIC/BIC values (mg/L) (n=64)
The number of “non-susceptible” (“Resistant” - “R” - plus “Intermediate” - “I”) isolates according to MIC and BIC for each antibiotic was as follows: CAZ 9/64 (14.1%) and 24/64 (37.5%); CIP 19/64 (29.7%) and 23/64 (36%); TOB 13/64 (20.4%) and 30/64 (46.8%); IPM 15/64 (23.4%) and 44/64 (68.8%); MEM 6/64 (9.4%) and 18/64 (28.1%), respectively. There was a statistical significant difference between the susceptibility category of isolates for all antibiotics tested, except for CIP (CAZ: P = 0.001, CIP: P = 0.234, TOB: P = 0.001, IPM: P < 0.001, MEM: P = 0.005).
The macrolide MIC values were tested for all isolates. Both azithromycin (AZM) (range 32 - 4096) and clarithromycin (CLR) (range 128 - 4096) presented a median MIC of 512 mg/L. MIC50 and MIC90 (mg/L) for all isolates were 512 and 1024 for AZM; 512 and 4096 for CLR, respectively.
A total of 19 (29.7%) isolates presented the mucoid phenotype, but no statistical significant differences in the susceptibility profile of mucoid and non-mucoid isolates were found for the antibiotics tested in the different conditions performed in this study (MIC, BIC and MCA).
The repeatability of the assays demonstrated a coefficient of variation (CV) of MIC and BIC for CAZ, CIP, IPM, MEM, and TOB of 10.21 and 9.45, 7.09 and 8.46, 14.74 and 2.13, 7.70 and 3.94, 10.01 and 8.51, respectively. When macrolides were associated, the highest CV was 20.12% for CAZ with 8 mg/L of CLR and the lowest was 0% for TOB with 2 and 8 mg/L of CLR.
Bacteria in biofilm are more prone to resist treatment with antibiotics and to evade the action of immune system cells. The present study observed a significant difference between MIC in planktonic and in biofilm growth conditions. BIC values were considerably higher than the conventional MIC values for all anti-pseudomonal antibiotics tested in our study as also found by Moskowitz and collaborators . MEM proved to be the most active antibiotic regardless the growth condition, CAZ proved to be the second most active antibiotic in planktonic conditions of growth, whereas CIP was the second most active antibiotic in biofilm conditions. In vitro studies have indicated that CIP is one of the most active agents against bacterial biofilm of S. aureus and P. aeruginosa. This is possibly related to the fluoroquinolones ability to penetrate into biofilms killing non-growing bacteria [20–22]. As expected, all isolates were resistant to AZM and CLR.
The principal finding of our study was that non-susceptible P. aeruginosa exposed to macrolides at sub-inhibitory concentrations became susceptible to a variety of anti-pseudomonal agents (CAZ, CIP, IPM, MEM, and TOB) in biofilm conditions. It is of note that in many associations we found a strong IQ between anti-pseudomonal agents and macrolides. The impact of tobramycin/clarithromycin and ceftazidime/clarithromycin co-administration on P. aeruginosa biofilms was also observed in studies of Tré-Hardy and collaborators [23, 24]. Other study showed that the biofilm was strongly affected by the presence of clarithromycin, and, in its presence, amikacin MIC lower than those obtained in the absence of clarithromycin .
In our study, co-administration of AZM at 8 mg/L presented considerable impact when associated with all anti-pseudomonal agents tested (CAZ, CIP, IPM, MEM, and TOB) on P. aeruginosa biofilms from CF patients. Although AZM has no bactericidal effect on P. aeruginosa, it was shown that AZM retards the formation of biofilms and blocks the bacterial quorum sensing involved in the production of biofilms [26–28]. The use of AZM to treat chronic infections of P. aeruginosa in the lungs of CF patients has been gaining favour due to the improved outcome of CF patients treated with this antibiotic [29, 30].
Synergistic and additive activities were noted when AZM and CLR were paired with conventional antimicrobial agents for P. aeruginosa strains in the study of Saiman and collaborators. Overall, combinations were more active against CF isolates than against non-CF isolates and more active against mucoid strains than against non-mucoid strains . However, in our study no significant difference in the macrolides combination assay was observed when we compared mucoid with non-mucoid P. aeruginosa clinical isolates.
Interpretative criteria of susceptibility are not standardized for the combination assay in biofilm conditions and this is the main limitation of our study. Therefore, one must be aware that the biofilm susceptibility testing and the macrolide combination assay proposed in our study need further clinical validation for applying it in microbiology laboratories.
In conclusion, P. aeruginosa clinical isolates from CF patients within biofilms are highly resistant to antibiotics and macrolides may be useful as adjunctive therapy as they proved to augment the in vitro activity of anti-pseudomonal agents.
A total of 64 P. aeruginosa isolates were collected from the sputum of 34 (20 male and 14 female) CF patients attending at the Cystic Fibrosis Centre in Hospital de Clínicas de Porto Alegre, Brazil, from December 2005 to July 2008. The median age of patients was 13 years (range 2 - 30) and the majority of patients presented positive sputum culture for P. aeruginosa for at least 5 years. In most children cases, the sputum was obtained only after respiratory physiotherapy. Sputum samples were cultured quantitatively by standard microbiological methods . Isolates of P. aeruginosa obtained from the sputum culture were stored at −80°C. P. aeruginosa ATCC 27853 was used as quality control for the anti-pseudomonal agents, S. aureus ATCC 25923 was used as quality control for the macrolides agents, and PA01 was used as reference of biofilm-forming bacteria.
Stock solution of antibiotics were prepared following the instructions of the manufacturer (Sigma-Aldrich® Co, St Louis, USA) and stored at −80°C until use. Working solutions were prepared in cation-adjusted Mueller-Hinton broth (CAMHB) (Becton Dickinson, Sparks, MD) at 512 mg/L for CAZ, CIP, TOB, IPM, and MEM. AZM and CLR working solutions were prepared at 8192 mg/L. From these working solutions serial twofold dilutions were prepared in CAMHB and distributed in a 96-well microtiter plate.
Minimal inhibitory concentration (MIC) and biofilm inhibitory concentration (BIC)
MIC values were determined by broth microdilution using the twofold dilution method according to the Clinical and Laboratory Standards Institute (CLSI) guidelines . The antibiotic concentrations tested ranged from 0.5 to 256 mg/L for the anti-pseudomonal antibiotics CAZ, CIP, TOB, IPM, and MEM; and from 2 to 4096 mg/L for the macrolides AZM and CLR.
BIC values were determined as previously described . Prior to testing, the organisms were subcultured in trypticase soy broth with 5% KNO3 and incubated overnight after retrieval from −80°C. Bacteria were re-subcultured in MacConkey agar (bioMèrieux®, France) and incubated overnight. A bacterial suspension in CAMHB containing 5% KNO3 was prepared with an inoculum density equivalent to 0.5 McFarland (Densimat, bioMèrieux®). Afterwards, 100 μL were inoculated into all but the negative control of a flat-bottom 96-well microtiter plate. Plates were covered with lids presenting 96 pegs in which the biofilms could build up, followed by incubation at 37°C for 20 h. Peg lids were rinsed three times with sterile saline to remove non-binding cells, placed onto other 96-well flat-bottom microplates containing a range of antibiotic concentrations and incubated for 18 to 20 h at 37°C. Pegs carrying control biofilms were submerged in antibiotic-free medium. After antibiotic incubation, peg lids were again rinsed three times in sterile saline and incubated in fresh CAMHB in a new microplate and centrifugated at 805 X g for 20 min. The peg lid was discarded and replaced by a standard lid. The optical density (OD) at 650 nm was measured on a microtiter plate colorimeter before and after incubation at 37°C for 6 h (OD650 at 6 h minus OD650 at 0 h). Biofilm formation was defined as a mean OD650 difference ≥ 0.05 for the biofilm control. The BIC values were defined as the lowest concentration without growth. CLSI criteria  were used to classify the isolates as ¨Susceptible¨ (“S”), ¨Intermediate¨ (“I”) or ¨Resistant¨ (“R”).
Macrolide combination assay (MCA) and inhibitory quotient (IQ)
Only isolates with a BIC value in “R” or "I” classification according to CLSI interpretative criteria  for CAZ, CIP, TOB, IPM, and MEM were used in the MCA and IQ.
MCA was performed in a 96-well microplate containing CAZ, CIP, TOB, IPM, or MEM in twofold dilutions in addition to macrolides at sub-inhibitory concentrations . With the purpose to assign activity of AZM and CLR in combination with the antibiotics and to better evaluate susceptibility changing category, we established an inhibitory quotient (IQ). IQ is the quotient of the maximum antibiotic serum concentration and the BIC value of each antibiotic in combination with the macrolide. IQ categorization for CAZ, CIP, TOB, IPM, and MEM to evaluate the activity of macrolides in different concentrations against resistant P. aeruginosa isolates was as follows: strong IQ (IQ ≥ 2, except for CIP, whose IQ was ≥ 1), weak IQ (IQ = 0.5), or non-inhibition (IQ ≤ 0.5). Strong IQ means that there was a reduction in BIC when macrolide combination was tested and the isolates changed their profile from “R” to “S”; weak IQ means that there was a reduction in the BIC value when the isolate profile changed from “R” to “I”; and non-inhibition means no change in the bacteria antibiotic susceptibility profile .
All assays were performed four times. Mean values of the four repetitions, standard deviations, and CV were calculated and the mean value was considered the value which was then used to categorize the isolates as “R”, “I” or “S”.
The susceptibility profile of mucoid and non-mucoid isolates was evaluated under the different conditions performed in this study (MIC, BIC and MCA).
The Wilcoxon signed ranks test was used for statistical analysis of quantitative values of MIC and BIC. McNemar-Bowker test was used to evaluate the categories of the results obtained (“S”, “I” and “R”) by the standard technique and the technique in biofilm. P < 0.05 indicated statistical significance.
The bacterial isolates were obtained from clinical specimens sent for routine culture in the Microbiology Unit of Hospital de Clínicas de Porto Alegre. The information was compiled in order to respect the privacy of patients; written informed consent for participation in the study was obtained from participants or, where participants were children, from a parent or guardian. This study was approved by the Ethics Committee in Research of Hospital de Clínicas de Porto Alegre (project number 06 - 406).
We would like to thank Vania Naomi Hirakata for assistance with statistical analyses.
This work received financial support from FIPE (Fundo de Incentivo à Ensino e Pesquisa do Hospital de Clínicas de Porto Alegre), CNPq (Conselho Nacional de Desenvolvimento Científico e Tecnológico) and ANVISA (Agência Nacional de Vigilância Sanitária).
- Staab D: Cystic fibrosis - therapeutic challenge in cystic fibrosis children. Eur J Endocrinol. 2004, 151 (Suppl 1): S77-S80.PubMedView ArticleGoogle Scholar
- Baltimore RS, Christie CD, Smith GJ: Immunohistopathologic localization of Pseudomonas aeruginosa in lungs from patients with cystic fibrosis. Implications for the pathogenesis of progressive lung deterioration. Am Rev Respir Dis. 1989, 140: 1650-1661. 10.1164/ajrccm/140.6.1650.PubMedView ArticleGoogle Scholar
- Costerton JW, Cheng KJ, Geesey GG, Ladd TI, Nickel JC, Dasgupta M, Marrie TJ: Bacterial biofilms in nature and disease. Annu Rev Microbiol. 1987, 41: 435-464. 10.1146/annurev.mi.41.100187.002251.PubMedView ArticleGoogle Scholar
- Drenkard E, Ausubel FM: Pseudomonas biofilm formation and antibiotic resistance are linked to phenotypic variation. Nature. 2002, 416: 740-743. 10.1038/416740a.PubMedView ArticleGoogle Scholar
- Singh PK, Schaefer AL, Parsek MR, Moninger TO, Welsh MJ, Greenberg EP: Quorum-sensing signals indicate that cystic fibrosis lungs are infected with bacterial biofilms. Nature. 2000, 407: 762-764. 10.1038/35037627.PubMedView ArticleGoogle Scholar
- Hacth RA, Schiller NL: Alginate lyase promotes diffusion of aminoglycosides through the extracellular polysaccharide of mucoid Pseudomonas aeruginosa. Antimicrob Agents Chemother. 1998, 42: 974-977.Google Scholar
- Costerton JW, Stewart PS, Greenberg EP: Bacterial biofilms: a common cause of persistent infections. Science. 1999, 284: 1318-1322. 10.1126/science.284.5418.1318.PubMedView ArticleGoogle Scholar
- Rogers GB, Hoffman LR, Whiteley M, Daniels TW, Carroll MP, Bruce KD: Revealing the dynamics of polymicrobial infections: implications for antibiotic therapy. Trends Microbiol. 2010, 18: 357-364. 10.1016/j.tim.2010.04.005.PubMedPubMed CentralView ArticleGoogle Scholar
- Lopez-Boado YS, Rubin BK: Macrolides as immunomodulatory medications for the therapy of chronic lung diseases. Curr Opin Pharmacol. 2008, 8: 286-291. 10.1016/j.coph.2008.01.010.PubMedView ArticleGoogle Scholar
- Schoni MH: Macrolide antibiotic therapy in patients with cystic fibrosis. Swiss Med Wkly. 2003, 133: 297-301.PubMedGoogle Scholar
- Nguyen T, Louie SG, Beringer PM, Gill MA: Potential role of macrolide antibiotics in the management of cystic fibrosis lung disease. Curr Opin Pulm Med. 2002, 8: 521-528. 10.1097/00063198-200211000-00007.PubMedView ArticleGoogle Scholar
- Shinkai M, Foster GH, Rubin BK: Macrolide antibiotics modulate ERK phosphorylation and IL-8 and GM-CSF production by human bronchial epithelial cells. Am J Physiol Lung Cell Mol Physiol. 2006, 290: L75-L85.PubMedView ArticleGoogle Scholar
- Shinkai M, Lopez-Boado YS, Rubin BK: Clarithromycin has an immunomodulatory effect on ERK-mediated inflammation induced by Pseudomonas aeruginosa flagellin. J Antimicrob Chemother. 2007, 59: 1096-1101. 10.1093/jac/dkm084.PubMedView ArticleGoogle Scholar
- Shinkai M, Tamaoki J, Kobayashi H, Kanoh S, Motoyoshi K, Kute T, Rubin BK: Clarithromycin delays progression of bronchial epithelial cells from G1 phase to S phase and delays cell growth via extracellular signal-regulated protein kinase suppression. Antimicrob Agents Chemother. 2006, 50: 1738-1744. 10.1128/AAC.50.5.1738-1744.2006.PubMedPubMed CentralView ArticleGoogle Scholar
- Parnham MJ: Immunomodulatory effects of antimicrobials in the therapy of respiratory tract infections. Curr Opin Infect Dis. 2005, 18: 125-131. 10.1097/01.qco.0000160901.71813.fe.PubMedView ArticleGoogle Scholar
- Culic O, Erakovic V, Parnham MJ: Anti-inflammatory effects of macrolide antibiotics. Eur J Pharmacol. 2001, 429: 209-229. 10.1016/S0014-2999(01)01321-8.PubMedView ArticleGoogle Scholar
- Schultz MJ: Macrolide activities beyond their antimicrobial effects: macrolides in diffuse panbronchiolitis and cystic fibrosis. J Antimicrob Chemother. 2004, 54: 21-28. 10.1093/jac/dkh309.PubMedView ArticleGoogle Scholar
- Fujimura S, Sato T, Kikuchi T, Gomi K, Watanabe A, Mchami T: Combined efficacy of clarithromycin plus cefazolin or vancomycin against Staphylococcus aureus biofilms formed on titanium medical devices. Int J Antimicrob Agents. 2008, 32: 481-484. 10.1016/j.ijantimicag.2008.06.030.PubMedView ArticleGoogle Scholar
- Moskowitz SM, Foster JM, Emerson J, Burns JL: Clinically feasible biofilm susceptibility assay for isolates of Pseudomonas aeruginosa from patients with cystic fibrosis. J Clin Microbiol. 2004, 42: 1915-1922. 10.1128/JCM.42.5.1915-1922.2004.PubMedPubMed CentralView ArticleGoogle Scholar
- Soboh F, Khoury AE, Zamboni AC, Davidson D, Mittelman MW: Effects of ciprofloxacin and protamine sulfate combinations against catheter-associated Pseudomonas aeruginosa biofilms. Antimicrob Agents Chemother. 1995, 39: 1281-1286. 10.1128/AAC.39.6.1281.PubMedPubMed CentralView ArticleGoogle Scholar
- Gander S, Gilbert P: The development of a small-scale biofilm model suitable for studying the effects of antibiotics on biofilms of gram-negative bacteria. J Antirnicrob Chernother. 2010, 40: 329-334.View ArticleGoogle Scholar
- Dasugupta MK, Shishida H, Salama S, Singh R, Larabie M, Micetich RG: The effect of macrolide and quinolone antibiotics in methicillin-resistant Staphylococcus aureus biofilm growth. Ad Perit Dial. 1997, 13: 214-217.Google Scholar
- Tre-Hardy M, Nagant C, El MN, Vanderbist F, Traore H, Vaneechoutte M, Dehaye JP: Efficacy of the combination of tobramycin and a macrolide in an in vitro Pseudomonas aeruginosa mature biofilm model. Antimicrob Agents Chemother. 2010, 54: 4409-4415. 10.1128/AAC.00372-10.PubMedPubMed CentralView ArticleGoogle Scholar
- Tre-Hardy M, Vanderbist F, Traore H, Devleeschouwer MJ: In vitro activity of antibiotic combinations against Pseudomonas aeruginosa Biofilm and planktonic cultures. Int J Antimicrob Agents. 2008, 31: 329-336. 10.1016/j.ijantimicag.2007.12.005.PubMedView ArticleGoogle Scholar
- Cirioni O, Ghiselli R, Silvestri C, Minardi D, Gabrielli E, Orlando F, Rimini M, Brescini L, Muzzonigro G, Guerrieri M, Giacometti A: Effect of the combination of clarithromycin and amikacin on Pseudomonas aeruginosa biofilm in an animal model of ureteral stent infection. J Antimicrob Chemother. 2011, 66: 1318-1323. 10.1093/jac/dkr107.PubMedView ArticleGoogle Scholar
- Bala A, Kumar R, Harjai K: Inhibition of quorum sensing in Pseudomonas aeruginosa by azithromycin and its effectiveness in urinary tract infections. J Med Microbiol. 2011, 60: 300-306. 10.1099/jmm.0.025387-0.PubMedView ArticleGoogle Scholar
- Gillis RJ, Iglewski BH: Azithromycin retards Pseudomonas aeruginosa Biofilm formation. J Clin Microbiol. 2004, 42: 5842-5845. 10.1128/JCM.42.12.5842-5845.2004.PubMedPubMed CentralView ArticleGoogle Scholar
- Tateda K, Comte R, Pechere JC, Kohler T, Yamaguchi K, Van DC: Azithromycin inhibits quorum sensing in Pseudomonas aeruginosa. Antimicrob Agents Chemother. 2001, 45: 1930-1933. 10.1128/AAC.45.6.1930-1933.2001.PubMedPubMed CentralView ArticleGoogle Scholar
- Equi A, Balfour-Lynn IM, Bush A, Rosenthal M: Long term azithromycin in children with cystic fibrosis: a randomised, placebo-controlled crossover trial. Lancet. 2002, 360: 978-984. 10.1016/S0140-6736(02)11081-6.PubMedView ArticleGoogle Scholar
- Wolter J, Seeney S, Bell S, Bowler S, Masel P, McCormack J: Effect of long term treatment with azithromycin on disease parameters in cystic fibrosis: a randomised trial. Thorax. 2002, 57: 212-216. 10.1136/thorax.57.3.212.PubMedPubMed CentralView ArticleGoogle Scholar
- Saiman L, Chen Y, Gabriel PS, Knirsch C: Synergistic activities of macrolide antibiotics against Pseudomonas aeruginosa, Burkholderia cepacia, Stenotrophomonas maltophilia, and Alcaligenes xylosoxidans isolated from patients with cystic fibrosis. Antimicrob Agents Chemother. 2002, 46: 1105-1107. 10.1128/AAC.46.4.1105-1107.2002.PubMedPubMed CentralView ArticleGoogle Scholar
- Murray PR, Baron EJ, Jorgensen JH, Pfaller MA, Yolken RH: Manual of clinical microbiology. 2003, Washington: ASM Press, 8Google Scholar
- Clinical and Laboratory Standards Institute: Methods for dilution antimicrobial susceptibility test for bacteria that grow aerobically M07-A8. 2009, Wayne: CLSI, 8Google Scholar
- Clinical and Laboratory Standards Institute: Performance Standards for antimicrobial susceptibility testing. M100-S20. 2010, Wayne: CLSIGoogle Scholar
- Ichimiya T, Yamasaki T, Nasu M: In vitro effects of antimicrobial agents on Pseudomonas aeruginosa Biofilm formation. J Antimicrob Chemother. 1994, 34: 331-341. 10.1093/jac/34.3.331.PubMedView ArticleGoogle Scholar
- Moskowitz SM, Foster JM, Emerson JC, Gibson RL, Burns JL: Use of Pseudomonas Biofilm susceptibilities to assign simulated antibiotic regimens for cystic fibrosis airway infection. J Antimicrob Chemother. 2005, 56: 879-886. 10.1093/jac/dki338.PubMedView ArticleGoogle Scholar
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.