Antimicrobial evaluation of selected naturally occurring oxyprenylated secondary metabolites

Abstract This study tested the antimicrobial activity of eight selected naturally occurring oxyprenylated secondary metabolites against Staphylococcus aureus ATCC 29213, S. epidermidis ATCC 35984, Escherichia coli ATCC 8739, Pseudomonas aeruginosa ATCC 9027 and Candida albicans ATCC 10231. Results showed a moderate antimicrobial activity. The most active compounds were 3-(4-geranyloxyphenyl)-1-ethanol (4) and 3-(4-isopentenyloxyphenyl)-1-propanol (5) that were tested on mature and in-formation biofilms of all micro-organisms, moreover the cytotoxic activity was evaluated. Except for S. epidermidis, both compounds reduced significantly (p < 0.05) the microbial biofilm formation at 1/2 MIC and 1/4 MIC, in particular, compounds 4 and 5 at each concentration, inhibited E. coli biofilm formation to a greater extent, the biofilm formation was never more than 44% in respect to the control, moreover both compounds showed a low cytotoxic effect. Oxyprenylated derivatives may be of great interest for the development of novel antimicrobial therapeutic strategies and the synthesis of semi-synthetic analogues with anti-biofilm efficacy. Graphical abstract


Introduction
The growing microbial resistance to commonly used drugs is a global problem (Rossolini et al. 2014). The antimicrobial drugs' failure to treat infections could be also caused by bacteria forming a biofilm, a sessile microbial community embedded in an extracellular polymeric matrix that represents one of the most important causes of recalcitrant infections (Falagas et al. 2009;de la Fuente-Núñez et al. 2013).
During the last decades, natural products as sources of new compounds with antimicrobial and anti-biofilm activities have been extensively studied (Saleem et al. 2010;Nostro & Papalia 2012;Zhou et al. 2012;Jadhav et al. 2014). Oxyprenylated natural products, such as 3,3-dimethylallyloxy-(C 5 ), geranyloxy-(C 10 ), and farnesyloxy-(C 15 ) related compounds, represent a family of secondary metabolites that have been regarded for years as mere biosynthetic intermediates of the more widespread C-prenylated derivatives. These secondary metabolites have been recognised as interesting and valuable biologically active phytochemicals exerting several beneficial pharmacological effects (Epifano et al. 2007) The phytochemistry and pharmacology of such compounds were recently reviewed by our group (Epifano et al. 2007) and to better characterise these compounds, in this work we investigated the in vitro antimicrobial activity of eight selected naturally occurring prenyloxybenzaldehydes, prenyloxyalcohols and phenylpropanoid derivatives ( Figure 1) against a panel of clinically relevant pathogens: Staphylococcus aureus ATCC 29213, S. epidermidis ATCC 35984, Escherichia coli ATCC 8739, Pseudomonas aeruginosa ATCC 9027 and Candida albicans ATCC 10231. For the most active compounds, the cytotoxic and anti-biofilm effects were also evaluated. It is noteworthy to underline that data about the antimicrobial properties of naturally occurring oxyprenylated secondary metabolites are not so numerous in the literature. Naturally occurring parent compounds, namely pOH-benzaldehyde, vanillin, eugenol and isoeugenol of the eight geranyl and 3,3-dimethylallyoxy natural products under investigation are known to have antimicrobial properties (Fitzgerald et al. 2004;Ehab et al. 2008;Devi et al. 2010;Alamri et al. 2012). It is also well known how the addition of a side chain rendering such products more hydrophobic provides more potentialities to disrupt or negatively interact with the functionalities of bacterial cell membrane. Recent examples reported in the literature showed the efficiency of oxyprenylated phytochemicals as antibacterial agents (Curini et al. 2012). Thus, studying such a research topic is of current and growing interest.

Results and discussion
The minimum inhibitory concentration (MIC) and minimum bactericidal/fungicidal concentration (MBC, MFC) results are reported in Table S1. From the data, among tested compounds, the most notable enhancement in the antibacterial activity was for the two alcohols 3-(4-geranyloxyphenyl)-1-ethanol (4) and 3-(4-isopentenyloxyphenyl)-1propanol (5), showing MIC values ranging between 200 and 400 μg mL −1 and MBC values ranging between 400 and 800 μg mL -1 , for all bacterial strains. The other derivatives: p-Isopentenyloxybenzaldehyde (1), geranyloxyvanillin (2), 5-dimethoxy-4-isopentenyloxybenzyl alcohol (3), (2E)-3-(4-((E)3,7-dimethylocta-2,6-dienyloxy)-3-methoxyphenyl) acrylaldehyde (6), 4-isopentenyloxyeugenol (7) and 4-isopentenyloxyisoeugenol (8), possess lower antibacterial activity compared to 4 and 5, except for the geranyloxyvanillin (derivative 2) for S. aureus ATCC 29213 and the derivative 7, 4-isopentenyloxyeugenol, for S. epidermidis ATCC 35984 (Table S1). For C. albicans ATCC 10231, all compounds showed good antifungal activity, the MIC values ranged between 100 and 200 μg mL -1 and the MFC between 3200 and 100 μg mL −1 . As shown in Table S1, also for C. albicans ATCC 10231, compounds 4 and 5 possess a more interesting antifungal activity, together with the isoeugenol derivative 8 that presented, as compound 5, a fungicidal effect at 100 μg mL −1 . Since 4 and 5 derivatives demonstrated higher antibacterial/antifungal activity, they were selected to detect, in addition, their cytotoxic and anti-biofilm activities. The cellular toxicity effect of compounds 4 and 5 was assessed using the hemolytic assay. The results showed that both compounds were not cytotoxic at their MIC and MBC values and compound 4 produced a human red blood cell hemolysis below 50% at all tested concentrations ( Figure S2). The 4 and 5 derivatives' activity was tested on mature and in-formation biofilms. For most of the studied strains, the Biofilm Inhibitory Concentration and the Biofilm Eradication Concentration values were 8-fold or 16-fold greater than the concentration required to inhibit growth in the planktonic phase (Table S2); except compound 4 for S. aureus ATCC 29213 that inhibited the mature biofilm at a concentration 4-fold greater than the MIC value. The mature biofilm of all tested micro-organisms, including C. albicans, was not inhibited by treatment with 4 and 5 compounds, underlying probably the inability to penetrate the microbial matrix and consequently killing the micro-organisms. Significant differences in biofilm formation (p < 0.05) were recorded, in respect to the control, when micro-organisms were treated with sub-inhibitory concentrations of compound 4 and 5 ( Figure S3). In particular, both compounds reduced significantly the biofilm formation of S. aureus ATCC 29213, compound 4 at 1/2 MIC and 1/4 MIC (respectively, 59 and 63% of biofilm formation) and compound 5 at each tested concentration (from 61.4 to 78.4% of biofilm formation). The E. coli ATCC 8739 capability to form a biofilm was inhibited significantly with sub-inhibitory concentrations of 4 and 5 compounds to a greater extent than other bacteria, in fact, in presence of both compounds, the biofilm formation was never more than 44% with respect to the control. Compound 5 reduced significantly the biofilm formation of P. aeruginosa ATCC 9027 at each concentration (from 51.2 to 73.3% of biofilm formation) and on the contrary, no significant reduction in the biofilm formation was recorded for S. epidermidis ATCC 35984. Compounds 4 and 5 reduced significantly the biofilm formation of C. albicans both at 1/2 MIC value (respectively, 63.3% of biofilm formation for compound 4 and 42.8% of biofilm formation for compound 5) and the compound 4 also at 1/4 MIC (56.3% of biofilm formation) and 1/8 MIC values (50% of biofilm formation). The effects of 4 and 5 compounds on the biofilm development were different, they inhibited normal biofilm formation of S. aureus, E. coli, P. aeruginosa and C. albicans, showing no effect on the S. epidermidis biofilm formation. Compound 5 was more active on S. aureus, E. coli and P. aeruginosa; compound 4 was active in reducing the biofilm formation of E. coli and C. albicans. The reason for this action could be attributed probably to the capability of the two compounds to influence the quorum sensing (QS). As known, bacterial biofilm activity is regulated by QS, a system to regulate cooperative activities and physiological processes used by both Gram positive and Gram negative bacteria for the production and release of external signal molecules (Li & Tian 2012). The formation of biofilm by bacteria represents a serious problem in medical and clinical settings (Høiby et al. 2010) and there are some evidences that show that natural compounds are able to influence the QS system in bacteria (Kerekes et al. 2013;Tan et al. 2013;Burt et al. 2014). So, the finding of new compounds able to hinder the biofilm formation is important to design and/or develop new formulations to treat biofilm forming microbial infections.
Results described herein represent the ideal continuation of recently reported studies about the antibacterial activities of prenyloxyphenylpropanoids and biosynthetically related products (Curini et al. 2012). Since oxyprenylated secondary metabolites are widespread in several edible fruits and vegetables, results depicted herein might be of great interest in the near future for the development of novel antimicrobial therapeutic strategies as well as to properly address the synthesis of semi-synthetic analogues of the title compounds with enhanced growth inhibitory activities. Results obtained in the case of C. albicans are preliminary and will be further and better characterised in the near future. In this regard, studies to better define the pharmacological and toxicological profile of the compounds under investigation as well as depicting the mechanism of action underlying the observed effects are now ongoing in our laboratories.

Conclusion
Oxyprenylated derivatives may be of great interest as lead compounds for the development of novel antimicrobial therapeutic strategies as well as to properly address the synthesis of semi-synthetic analogues with enhanced anti-biofilm efficacy.

Supplementary material
Experimental details relating to this article and Tables S1 and S2 and Figures S2 and S3 are available online.