Two different primary oxidation mechanisms during biotransformation ...

Appl Microbiol Biotechnol (2013) 97:1289–1297 DOI 10.1007/s00253-012-4293-8

APPLIED MICROBIAL AND CELL PHYSIOLOGY

Two different primary oxidation mechanisms during biotransformation of thymol by gram-positive bacteria of the genera Nocardia and Mycobacterium Veronika Hahn & Katharina Sünwoldt & Annett Mikolasch & Frieder Schauer

Received: 12 May 2012 / Revised: 6 July 2012 / Accepted: 9 July 2012 / Published online: 25 July 2012 # Springer-Verlag 2012

Abstract Thymol has antibacterial, antifungal, insecticidal, and antioxidative properties which are the basis for the wide use of this compound in the cosmetic, food, and pharmaceutical industries. Although thymol is a ubiquitously occurring substance in the environment, data about its degradation and detoxification by bacteria are sparse. Here, we show the existence of two different pathways for the biotransformation of thymol by Nocardia cyriacigeorgica and Mycobacterium neoaurum which were described for the first time for grampositive bacteria. The first pathway starts with hydroxylation of thymol to thymohydroquinone (2-isopropyl-5-methylbenzene-1,4-diol) with subsequent oxidation to thymobenzoquinone (2-isopropyl-5-methyl-1,4-benzoquinone). The second pathway involves hydroxylation of the methyl group followed by oxidation to 3-hydroxy-4-isopropylbenzoic acid, possibly via the aldehyde 3-hydroxy-4-isopropylbenzaldehyde. It is noteworthy that the branched side chain of thymol was not oxidized. Similarities and differences of these oxidation processes with those of the gram-negative bacterium Pseudomonas putida, fungi, and plants are discussed and, in addition, the toxicity of thymol towards N. cyriacigeorgica and M. neoaurum was tested. The experiments showed a temporary growth inhibition with 0.025 % thymol. This was explained by degradation of thymol and the formation of products which are less toxic than thymol itself. Keywords Biotransformation . Degradation . Environmental pollution . Hydroxylation . Mineral oil . Quinone V. Hahn (*) : K. Sünwoldt : A. Mikolasch : F. Schauer Institute of Microbiology, Ernst-Moritz-Arndt-University Greifswald, F.-L.-Jahnstr. 15, 17487 Greifswald, Germany e-mail: [email protected]

Introduction Thymol belongs to the group of monocyclic monoterpenes (Daniel 2006) and it is part of essential oils from herbaceous plants Thymus vulgaris (Poulose and Croteau 1978), Monarda punctata (Scora 1967) or Origanum vulgare spp. hirtum (Vokou et al. 1993). Thymol has insecticidal (Samarasekera et al. 2008; Pandey et al. 2009), antioxidative (Lagouri et al. 1993; Aeschbach et al. 1994), antifungal (Pinto et al. 2006; Braga et al. 2007), and antibacterial activities against oral (Didry et al. 1994) and food-derived bacteria (Cosentino et al. 1999; Ait-Ouazzou et al. 2011). The range of biological properties and the natural origin of thymol have led to its wide use in cosmetic (Panda 2000; Shrestha et al. 2011) and food industries (Panda 2002; Valero et al. 2006) as well as in pharmaceuticals for human (Derby et al. 2011) and veterinary (Costa et al. 2010) medicine. Though thymol is a ubiquitously occurring compound in the environment, it is not currently clear what kind of products are formed during its degradation by microorganisms in soil and water or even through the bacterial skin flora of humans. The analysis of metabolic products of thymol is therefore of interest to estimate the possible accumulation of metabolic products in the environment and the toxicity risk of the formed metabolites. Up to now, few fungi (Numpaque et al. 2011) and bacteria (Chamberlain and Dagley 1968) as well as plants (Shimoda et al. 2006; Ghasemi et al. 2009) have been examined for their ability to transform thymol. One of the first reports about thymol degradation by bacteria described a pathway involving hydroxylation and meta-cleavage in the gram-negative bacterium Pseudomonas putida (Chamberlain and Dagley 1968). In contrast, plants form thymol glycosides (Shimoda et al. 2006), which is probably a detoxification mechanism for hazardous compounds (Kamel et al. 1992). The strategy to glycosylate a pollutant was also determined for 4-sec-

1290

butylphenol, a compound with branched side chain and a hydroxyl group similar to thymol, by Mycobacterium neoaurum (Hahn et al. not published). M. neoaurum and Nocardia cyriacigeorgica belong to the phylum actinobacteria. M. neoaurum is able to degrade phenylalkanes with aliphatic side chains (Herter et al. 2012) and N. cyriacigeorgica can oxidize phenylalkanes with branched side chains (Nhi-Cong et al. 2010). Because of this, it was assumed that these bacteria are able to degrade thymol in a similar way, possibly at the branched side chain. The aim of the present study was to determine the toxicity of thymol on the gram-positive bacteria N. cyriacigeorgica and M. neoaurum and to analyze the products which are formed during incubation with thymol. These products were structurally characterized by liquid chromatography/mass spectrometry (LC/MS), gas chromatography/mass spectrometry (GC/MS), and nuclear magnetic resonance (NMR) analyses.

Materials and methods Chemicals Thymol was obtained from ABCR GmbH & Co. Kg (Karlsruhe, Germany). Thymobenzoquinone and n-tetradecane were from Sigma-Aldrich Chemie GmbH (Steinheim, Germany).

Appl Microbiol Biotechnol (2013) 97:1289–1297

weight (in grams per liter) was determined. For this purpose, 10 ml of the assay was filtered through a glass fiber filter (GF6, 50 mm, Whatman GmbH, Dassel, Germany) which was subsequently dried for 2 h at 100 °C and weighed. Biotransformation experiments According to the biotransformation experiments of phenylalkanes with M. neoaurum (Herter et al. 2012) and N. cyriacigeorgica (Nhi-Cong et al. 2010) the cells were precultured with tetradecane as sole carbon source on mineral salt medium agar. The mineral salt medium contained 5 g NH4H2PO4, 2.5 g K2HPO4, 0.5 g MgSO4·7H2O, 0.5 g NaCl, 0.46 g K2SO4, 0.07 g CaCl2 (Hundt et al. 1998) and different trace elements 2 mg FeCl3·6H2O, 0.5 mg H3BO3, 0.1 mg CuSO4·5 H2O, 0.1 mg KJ, 0.4 mg MnSO4·5H2O, 0.4 mg ZnSO4·7H2O, 0.2 mg Na2MoO4, 0.1 mg CoCl2 in 1 l A. bidest. (Fritsche 1968; Kreisel and Schauer 1987). For plates, the medium was supplemented with agar-agar. Cells were incubated on these agar plates supplemented with ntetradecane which was evaporated from a sterile filter paper sheet in the plate cover for 2 days at 30 °C (Kreisel and Schauer 1987). For the biotransformation assay, bacterial biomass from three agar plates was suspended in 15 ml mineral salt medium and the suspension was transferred to 500-ml flasks containing 85 ml mineral salt medium and 0.025 % thymol. Controls contained 100 ml mineral salt medium and either no thymol or no cells. Flasks were shaken at 250 rpm and 30 °C.

Microorganisms and culture condition Analysis of biotransformation products by analytical HPLC Experiments were carried out using the two bacterial strains N. cyriacigeorgica SBUG 1472 and M. neoaurum SBUG 109, which are deposited at the strain collection of the Department of Biology of the University of Greifswald (SBUG). The microorganisms were cultivated on nutrient agar plates for 2 days at 30 °C and then used for toxicity tests or biotransformation experiments. Toxicity tests To determine the toxicity of thymol, 500-ml flasks containing 100 ml sterile nutrient broth II supplemented with 0.005, 0.01, 0.025, 0.05, or 0.1 % thymol were used. The nutrient broth II consisted of 15 g nutrient broth II Sifin (Institut für Immunpräparate und Nährmedien GmbH, Berlin, Germany) dissolved in 1 l A. bidest. (pH 7.2). Bacterial biomass from five agar plates was suspended in 25 ml of nutrient broth II and 5 ml of this cell suspension was added to each of the flasks. Control flasks contained either no thymol or no cells. All flasks were shaken at 250 rpm and 30 °C. In the course of incubation (after 24, 48, 72, 96, and 120 h) bacterial dry

Two-milliliter samples of the culture were taken after 0, 6, 24, 72, 96, and 168 h of incubation and centrifuged at room temperature at 5,000 rpm for 8 min (Hettich-Universal 30F, Tuttlingen, Germany) to remove cells from the incubation medium. The supernatant was analyzed on a high-pressure liquid chromatography (HPLC) system LC-10AT VP (Shimadzu, Germany) consisting of a FCV-10AL VP pump, SPD-M10A VP diode array detector, and a SCL-10A VP control unit controlled by Class-VP version 6.12 SP5. The separation of the substances was achieved on an endcapped, 5-μm, LiChroCART® 125-4 RP18 column (Merck, Darmstadt, Germany) run at a flow rate of 1 ml/min. A solvent system consisting of methanol (eluent A) and 0.1 % phosphoric acid (eluent B), starting from an initial ratio of 30 % A and 70 % B and reaching 100 % methanol within 14 min, was used. Liquid–liquid extraction After an incubation period of 168 h the entire biotransformation cultures were centrifuged at 4 °C with 16,300×g for


20 min (Sorvall® RC-5B Refrigerated Superspeed Centrifuge, Du Pont Instruments, Bad Homburg, Germany). The supernatant was filtered through a glass fiber filter (GF6, 50 mm, Whatman GmbH, Dassel, Germany) to ensure complete removal of intact cells. The pH of the culture supernatant was adjusted to pH 9 with sodium hydroxide and extracted with ethylacetate. The resulting solvent phase was named extract 1 (E1). The aqueous phase was then acidified to pH 2 with hydrochloric acid and extracted once more with ethylacetate. The resulting solvent phase was named extract 2 (E2). E1 and E2 were dried separately over anhydrous Na2SO4 and concentrated by rotary evaporation to 1 ml extract (Nhi-Cong et al. 2009). This residual volume was dried by evaporation using a nitrogen stream. For analysis in HPLC and GC/MS, the extracts were dissolved in methanol. Isolation of biotransformation products by semipreparative HPLC After 96 h, 1 l biotransformation assay was centrifuged, filtered, and the cell-free supernatant extracted with ethylacetate at neutral pH. The extract was evaporated using rotary evaporation and a nitrogen stream. For product separation, the extract was dissolved in approximately 1 ml methanol. Product isolation was carried out by preparative HPLC on an Agilent Series 1260 Infinity HPLC system with diode array detector, two separate highpressure pumps and a fraction collector (Waldbronn, Germany). HPLC separation was performed on an endcapped, 5-μm, LiChroCART® 125-4 RP18 column (Merck, Darmstadt, Germany) run at a flow rate of 1 ml/min. A solvent system consisting of methanol (eluent A) and 0.1 % aqueous acetic acid (eluent B), starting from an initial ratio of 30 % A and 70 % B and reaching 100 % methanol within 14 min, was used. The isolated products were structurally characterized by GC/MS, LC/MS, and NMR. Structure elucidation of products by GC/MS GC/MS analyses were carried out using an Agilent gas chromatograph 7890A GC System (Waldbronn, Germany) equipped with a 30 m HP-5ms column (0.25 mm by 0.25 μm film) and linked to a mass selective detector 5975C inert XL EI/CI MSD with a quadrupole mass spectrometer. For separation of products, a temperature program was used, starting with 1 min at 80 °C followed by a ramp from, 80–300 °C at 20 °C/min and finally 5 min at 300 °C. The acid extracts (E2) from liquid–liquid extraction were derivatized by methylation using diazomethane (De Boer and Backer 1956).

1291

Structure elucidation of products by LC/MS and NMR spectrometry The isolated products were characterized on a LC/MS system. The atmospheric pressure ionization (API) mass spectrometry experiments were performed using an Agilent Series 1200 HPLC system with diode array detector and an Agilent 6120 quadrupole mass spectrometer (Waldbronn, Germany). The MS was used with electrospray ionization (API-ES) source (dry and nebulizer gas: nitrogen). HPLC separation was performed on a Zorbax SB-C18 (2.1 × 50mm, 1.8 μm) column (Agilent, Waldbronn, Germany) at a flowrate of 0.4 ml/min. A solvent system consisting of acetonitrile (eluent A) and 0.1 % aqueous ammonium formic acid (eluent B), starting from an initial ratio of 10 % A and 90 % B and reaching 100 % methanol within 7 min, holding this until 13 min, was used. The NMR spectra were obtained at 600 MHz (1H), (HSQC, HMBC) in deuterated methanol on a Bruker Avance 600 instrument (Rheinstetten, Germany). Analytical data of products Thymohydroquinone/2-isopropyl-5-methylbenzene-1,4-diol (P1) Synthesis and isolation as described above. P1 was present in a mixture with P5. Yellow solid. Yield 2.8 % (7.1 mg). 1H NMR: δ 1.16 (d, J06.9 Hz, 6 H, H-8/H-9), 2.08 (s, 3H, H-10), 3.17 (m, J06.9 Hz, 1H, H-7), 6.48 (s, 1H, H-6), 6.57 (s, 1H, H-3). 13C NMR: 15.9 (C-10), 23.3 (C-8/C-9), 27.8 (C-7), 113.6 (C-3), 118.5 (C-6), 123.0 (C-5), 134.3 (C-2), 148.0 (C-1), 149.2 (C-4). HMBC correlations: H-3 (C-1, C-2, C-4, C-5, C-7, C-10), H-6 (C-1, C-2, C-4, C-5, C-7, C-10), H-7 (C-1, C-2, C-3, C-8/C-9), H-8/H-9 (C-2, C-7, C8/C-9), H-10 (C-2, C-3, C-4, C-5, C-6). Rf (HPLC) 5.40 min, UV/Vis (MeOH) λmax 212, 290 nm. LC/MS m/z AP-ESI: pos. ion mode [M + H]+ 167.1 (51); Rf (GC/MS) 7.12 min. GC/MS m/z 39 (5.9), 53 (5.7), 77 (10.4), 79 (8.0), 91 (5.3), 123 (9.8), 133 (5.2), 151 (100), 152 (9.6), 166 (36.4). 5-Hydroxymethyl-2-isopropylphenol (P2) Synthesis and isolation as described above. Brown solid. Yield 2.2 % (5.5 mg). 1H NMR: δ 1.19 (d, J06.9 Hz, 6 H, H-8/H-9), 3.25 (m, J06.9 Hz, 1H, H-7), 4.48 (s, 2 H, H-10), 6.75 (d, J08.3Hz, 1H, H-4), 6.76 (s, 1H, H-6), 7.09 (d, J08.3 Hz, 1H, H-3). 13C NMR: 21.6 (C-8/C-9), 26.4 (C-7), 63.8 (C-10), 113.4 (C-4), 118.0 (C-6), 125.7 (C-3), 134.0 (C2), 139.6 (C-5), 154.3 (C-1). HMBC correlations: H-3 (C-1, C-4, C-5, C-7), H-4 (C-1, C-2, C-3, C-6, C-7, C-10), H-6 (C2, C-4, C-10), H-7 (C-1, C-2, C-3, C-8/C-9), H-8/H-9 (C-2, C7, C8/C-9), H-10 (C-4, C-5, C-6). Rf (HPLC) 7.78 min, UV/

1292

3-Hydroxy-4-isopropylbenzoic acid (P3) Synthesis and isolation as described above. White solid. Yield 5.7 % (14.2 mg). Rf (HPLC) 9.21 min, UV/Vis (MeOH) λmax 217, 247, 298 nm; Rf (GC/MS) 6.82 min. GC/MS m/z 39 (10.7), 41 (14.4), 43 (31.0), 44 (41.0), 45 (49.3), 51 (10.6), 53 (5.8), 55 (7.9), 60 (8.9), 63 (7.1), 65 (9.1), 77 (23.5), 79 (12.3), 91 (28.8), 92 (5.3), 99 (5.3), 103 (11.2), 105 (18.8), 107 (18.2), 108 (6.6), 115 (19.4), 119 (5.7), 121 (8.7), 133 (25.5), 134 (12.2), 135 (13.3), 137 (22.9), 149 (12.2), 151 (17.0), 165 (100), 166 (14.6), 179 (5.0), 180 (32.3). Methylated product: Rf (GC/MS) 7.68 min. GC/MS m/z 39 (7.6), 41 (5.4), 51 (6,8), 53 (4.9), 59 (14.4), 63 (6.0), 65 (11.2), 77 (9.9), 79 (7.9), 91 (31.6), 102 (9.8), 103 (5.0), 105 (7.8), 107 (11.9), 115 (15.4), 119 (5.4), 120 (9.7), 135 (8.0), 147 (18.3), 163 (15.9), 179 (100), 180 (11.2), 194 (29.6). Double methylated product: Rf (GC/MS) 7.12 min. GC/MS m/z 39 (6.5), 41 (7.3), 51 (6.9), 59 (18.6), 63 (6.6), 65 (10.5), 77 (14.4), 78 (6.6), 79 (6.9), 89 (8.7), 91 (37.8), 92 (5.8), 93 (5.0), 102 (6.4), 103 (14.6), 104 (6.2), 105 (9.6), 115 (12.2), 117 (7.8), 119 (7.0), 121 (7.0), 131 (12.2), 133 (5.7), 134 (8.5), 147 (7.9), 149 (8.0), 161 (8.7), 165 (6.0), 177 (11.0), 193 (100), 194 (12.5), 208 (27.8). Thymobenzoquinone/2-isopropyl-5-methyl-1, 4-benzoquinone (P4) Synthesis and isolation as described above. P5 was present in a mixture with P1. Yellow solid. Yield 0.8 % (1.9 mg). 1H NMR: δ 1.13 (d, J06.9 Hz, 6 H, H-8/H-9), 2.00 (d(s), J0 1.4 Hz, 3 H, H-10), 2.98 (m, J06.9 Hz, 1H, H-7), 6.53 (s, 1H, H-3), 6.61 (d(s), J01.4 Hz, 1H, H-6). 13C NMR: 15.3 (C-10), 21.7 (C-8/C-9), 27.8 (C-7), 131.4 (C-3), 134.8 (C-6), 146.7 (C-5), 156.1 (C-2), 188.0 (C-1), 189.8 (C-4). HMBC correlations: H-3 (C-1, C-5, C-7), H-6 (C-2, C-4, C-10), H-7 (C-2, C8/C-9), H-8/H-9 (C-2, C-7, C8/C-9), H-10 (C-4, C-5, C-6). Rf (HPLC) 9.80 min, UV/Vis (MeOH) λmax 254 nm. LC/MS m/z AP-ESI: pos. ion mode [M + H]+ 165.0 (100); Rf (GC/MS) 5.12 min. GC/MS m/z 38 (7.0), 39 (50.1), 40 (25.3), 41 (21.1), 43 (10.0), 50 (9.2), 51 (21.0), 52 (13.2), 53 (51.5), 55 (10.1), 63 (7.5), 65 (17.9), 66 (10.8), 67 (28.4), 68 (28.7), 69 (8.3), 77 (42.7), 78 (8.8), 79 (16.2), 80 (7.0), 81 (14.1), 91 (53.4), 92 (6.9), 93 (93.4), 94 (10.5), 95 (10.3), 96 (12.9), 103 (11.5), 107 (14.8), 108 (28.4), 117 (8.4), 121 (97.2), 122 (10.0), 123 (5.6), 135 (11.8), 136 (77.7), 137 (8.0), 149 (55.2), 150 (5.4), 164 (100), 165 (10.7).

Thymobenzoquinone/2-isopropyl-5-methyl-1, 4-benzoquinone Authentic standard. Yellow solid. Rf (HPLC) 9.80 min, UV/ Vis (MeOH) λmax 254 nm; Rf (GC/MS) 5.08. GC/MS m/z 38 (7.3), 39 (52.5), 40 (26.1), 41 (20.9), 43 (9.8), 50 (8.8), 51 (19.8), 52 (12.8), 53 (47.2), 55 (9.3), 63 (6.8), 65 (16.0), 66 (10.3), 67 (27.6), 68 (28.1), 69 (8.2), 77 (40.9), 78 (8.7), 79 (15.9), 80 (7.0), 81 (14.1), 91 (48.9), 92 (6.4), 93 (89.5), 94 (10.4), 95 (10.0), 96 (12.7), 103 (10.4), 107 (13.4), 108 (25.3), 117 (7.6), 121 (86.0), 122 (9.1), 123 (5.2), 135 (10.6), 136 (67.3), 137 (7.4), 149 (52.4), 150 (5.2), 164 (100), 165 (11.1).

Results Toxicity tests The antibacterial effect of thymol was tested by determining the impact on growth of N. cyriacigeorgica and M. neoaurum of 0.005, 0.01, 0.025, and 0.1 % concentrations (Fig. 1, data for N. cyriacigeorgica). The two lowest concentrations of thymol 0.005 % and 0.01 % had negligible influence on both bacterial strains. In contrast, 0.025 % thymol diminished growth of N. cyriacigeorgica by 96 % (cf. Fig. 1) and of M. neoaurum by 94 % (after 24 h). However, the toxic effects of this thymol concentration are temporary and after a lag-phase of approximately 72 h (M. neoaurum) or 96 h (N. cyriacigeorgica) the bacterial growth resumed. At the two highest thymol concentrations, 0.05 and 0.1 %, bacterial growth was inhibited over the entire culture period of 120 h.

dry weight [g/l]

Vis (MeOH) λmax 212, 275 nm. LC/MS m/z AP-ESI: pos. ion mode [M + H]+ 167.0 (6), [M + Na]+ 189.0 (20); Rf (GC/MS) 7.24 min. GC/MS m/z 39 (5.5), 51 (5.3), 65 (6.2), 77 (19.7), 79 (8.6), 91 (12.4), 95 (11.5), 103 (11.0), 105 (10.2), 107 (8.3), 115 (6.2), 121 (25.1), 133 (10.5), 151 (100), 152 (9.4), 166 (34.9).


2.0 1.8 1.6 1.4 1.2 1.0 0.8 0.6 0.4 0.2 0.0

24 0%

48 0.005%

72 time [h] 0.01%

96

0.025%

120 0.05%

0.1%

Fig. 1 Growth of Nocardia cyriacigeorgica in absence (0 %) and in presence of different concentrations (0.005 %, 0.01 %, 0.025 %, 0.05 %, 0.1 % [w/v]) of thymol (dry weight data of cells at 0 h has been subtracted)


1293

0.25

In addition to the determination of antibacterial activity, the toxicity tests served to determine the optimal thymol concentration for biotransformation experiments—a concentration at which growth inhibitory effect but no cell death is seen. According to our experience, this concentration promises the highest extracellular metabolite yield which is important for product purification and analysis. For our experiments, this concentration was 0.025 % thymol for both N. cyriacigeorgica and M. neoaurum. Over an incubation period of 14 days (336 h), this concentration leads to the formation of four products in case of N. cyriacigeorgica (Fig. 2) which were also produced by M. neoaurum. In case of M. neoaurum, three additional products were formed in very low concentrations which excluded further structural analyses. The main product was P1 and was formed by both bacteria. After 168 h, the concentration of P1 achieved with N. cyriacigeorgica was 44 % higher than that from M. neoaurum. Due to the higher product yield with N. cyriacigeorgica and the similar product pattern of both bacteria, reaction kinetics and isolation of products is further described for this bacterium. In the course of incubation, thymol was not metabolized completely: only 64 % was consumed by N. cyriacigeorgica after 336 h (Fig. 3). The main products were P1 and P2 till 240 h and after which P1 and P2 decreased whereas P3 and P4 increased steadily. These data suggest two possible structural conversions of P1 and P2 to P3 and P4. The four products of the biotransformation experiments with N. cyriacigeorgica were isolated for detailed structural characterization. We were not able to isolate P1 as a pure substance. P1 has been transformed in very low quantity to P4 probably during the isolation procedure. The HPLC data of P1 and P2 showed two absorption maxima at 212 nm (P1/P2) and 290 nm (P1)/275 nm (P2) similar to the substrate thymol (λmax 210 and 275 nm). In particular the absorption maxima of P1 had similarities with 1,4-

0.20

0.15

0.10

0.05

0.00 0

100

200

300

time [h] Fig. 3 Concentration of thymol (dash) and the products P1 (filled square), P2 (open upright triangle), P3 (asterisk), P4 (open square) during incubation of Nocardia cyriacigeorgica on 0.025 % thymol over 336 h

hydroquinone (λmax 220 and 290 nm). LC/MS and GC/MS measurements of both products showed the molecular mass to be 166, indicating a hydroxylation leading to the mass increase of 16. The number of aromatic CH proton signals changed from three—in the reactant thymol—to two signals—in P1. The multiplicity and the CH-couplings of these two signals of P1 indicated a further substituent at the C-4 position. The absent proton signal at C-4 and the 13C NMR signal of C-4 at 149.2 ppm confirm the presence of an electron-withdrawing group such as a hydroxyl group at C-4. The 1H NMR spectrum of P2 has three aromatic CH proton signals such as the reactant thymol and hence the transformation does not take place at the aromatic ring but rather at a substituent. The change of the signal intensity of the aliphatic CH proton signal of H-10 to two-thirds and the 13 C NMR signal of C-10 at 63.8 ppm demonstrated the presence of an electron-withdrawing group such as a

9.21 9

1000

P3

5.40

11.081

750 500

P4

7.787

9.80

P2 250

area [mAU]

thymol P1

0

Fig. 2 HPLC elution profile at 220 nm of the cell-free supernatant after an incubation period of 336 h with Nocardia cyriacigeorgica and 0.025 % thymol

concentration [mg/ml]

Biotransformation experiments and formed products

5

6

7

8

9 time [minutes]

10

11

12

1294

hydroxyl group at C-10. Altogether, these data indicate that thymol was transformed via two different hydroxylation reactions, one on the aromatic ring at the C-4 position and the other on the methyl group C-10. Product 4 was identified in a mixture with P1; no pure P4 was isolated. HPLC data of P4 showed only one absorption maximum at 254 nm as found with the commercially available standard thymobenzoquinone (λmax 254 nm). In addition to the UV/Vis-spectrum, the retention time of P4 and thymobenzoquinone (Rf 9.80 min) were the same. LC/MS and GC/MS measurements showed the molecular mass of P4 to be 164 like that of thymobenzoquinone. NMR data confirmed the quinonoid structure of the hydroxylated and oxidized substrate thymol. Two signals in the range of 180– 190 ppm indicated the quinonoid character of P4. The extraction of the biotransformation assays at different pH-values resulted in identification of P1, P2, and P4 in the alkaline extracts (pH 9), whereas P3 was found in the acid extracts. This was a first hint for an acidic character of P3 in contrast to the other products. HPLC data showed three absorption maxima at 217, 247, and 298 nm. For GC/MS analyses, P3 was methylated and resulted in peaks at m/z 194 for the monomethylated and m/z 208 for the double methylated product. The unmethylated product was found with a peak at m/z 180.

Discussion Antimicrobial activity Thymol is known for its strong antibacterial activity (Didry et al. 1994; Cosentino et al. 1999; Ait-Ouazzou et al. 2011). This effect was also shown by our experiments using N. cyriacigeorgica and M. neoaurum. A thymol concentration of 0.025 % already notably lengthened the lag-phase. However, after 72 h (M. neoaurum) and after 96 h (N. cyriacigeorgica) the growth rate recovered and was 100 % (M. neoaurum) and 52 % (N. cyriacigeorgica) compared with the control without thymol. In contrast to 0.025 %, a concentration of 0.05 % and higher reduced growth significantly, albeit not completely. The concomitant decrease in toxicity and increase in growth with 0.025 % thymol may be caused by the continuous bacterial oxidation and detoxification of thymol and provides an indication for a real detoxification of thymol via the introduced biotransformation steps. Furthermore, it is also an indication that the products formed are less toxic than thymol itself. It is believed that the antimicrobial activity of thymol is due to a permeabilization of the cytoplasmic membrane and the depolarization of the membrane potential, finally leading to cell death (Ultee et al. 2002; Veldhuizen et al. 2006; Xu et al. 2008) and in this respect the phenolic moiety of thymol


may play an important role (Ultee et al. 2002; Veldhuizen et al. 2006). Thus, the transformation of thymol is expected to change this biological activity and thus reduce toxicity. In contrast to the effect of the hydroxyl groups on the benzene ring, aliphatic side chains of thymol may play a less important role for its antimicrobial activity (Veldhuizen et al. 2006; Numpaque et al. 2011). Removal of the side chains of carvacrol, the isomer of thymol, decreased the hydrophobic character, which probably influences the interaction of thymol with the cytoplasmic membrane and this resulted in reduced antimicrobial activity (Veldhuizen et al. 2006). Because of this, we suggest that the hydroxylation of the side chains increases the hydrophilic character of thymol but the precise influence of this structural change on the toxicity is not clear at the moment. In summary, the hydroxylation of the benzene ring is more valuable for detoxification than a modification on the side chains, which fits with our observation of the ratio of P1 and P2 of 2:1 formed after 168 h (P1, 0.06 mg/ml; P2, 0.03 mg/ml). Biotransformation pathways The biotransformation of thymol by N. cyriacigeorgica and M. neoaurum was shown by the accumulation of different products. Four products (P1–P4) were structurally characterized for N. cyriacigeorgica and were also produced by M. neoaurum (Fig. 4). In our study, we show for the first time the biotransformation of thymol by gram-positive bacteria proceeding via two different pathways. The first involves the hydroxylation of thymol in the para-position to the first hydroxyl group forming P1 (thymohydroquinone) with subsequent oxidation to P4 (thymobenzoquinone). The second pathway starts again with a hydroxylation of thymol but this time at the methyl group forming P2 (5-hydroxymethyl-2-isopropylphenol). This alcohol is further oxidized to the 3,4-disubstituted benzoic acid (P3), possibly via the corresponding benzaldehyde. In general, the first step for microbial degradation of phenol-like structures is a second hydroxylation in the ortho-position catalyzed by a monooxygenase (Fritsche und Hofrichter 2005). For phenol, this reaction is the initial step for further metabolization via ring fission by a dioxygenase (Fritsche und Hofrichter 2005). In contrast to phenol, thymol is hydroxylated in the para-position to the first OH-group (Fig. 4). A hydroxylase of the monooxygenase type is proposed to catalyze this step. The resulting thymohydroquinone (P1) was isolated from N. cyriacigeorgica and structurally characterized by LC/MS, GC/MS and NMR whereas for the gram-negative bacterium P. putida Chamberlain and Dagley (1968) only proposed the formation of the thymohydroquinone but did not isolate this compound. The accumulation of thymohydroquinone was also described for thymol

Appl Microbiol Biotechnol (2013) 97:1289–1297 Fig. 4 Biotransformation scheme for thymol by Nocardia cyriacigeorgica and Mycobacterium neoaurum. P1 Thymohydroquinone/2isopropyl-5-methylbenzene1,4-diol; P2 5-hydroxymethyl2-isopropylphenol; P3 3hydroxy-4-isopropylbenzoic acid; P4 thymobenzoquinone/2isopropyl-5-methyl-1,4-benzoquinone which were also formed by other organisms: B bacterium Pseudomonas putida (Chamberlain and Dagley 1968), F fungi Colleotrichum acutatum and Botryodiplodia theobromae (Numpaque et al. 2011), M microalga Oocystis pusilla (Ghasemi et al. 2009); postulated but not accumulated products from Nocardia cyriacigeorgica and Mycobacterium neoaurum are shown in brackets

1295

OH

Thymol 10

5

5

HO

OH

10

6

4

6

4 3

3

OH

1

2

1

2

OH

7

7 8

9

8

9

P2

P1 B,F H

O

10

OH

5

O

6 4 3

O

1

2 7 8

9

P4 F,M

HO

O

OH

P3 degradation by plant pathogenic fungi (Numpaque et al. 2011). Chamberlain and Dagley (1968) proposed a second hydroxylation in ortho-position to the first OH-group which was the substrate for a meta-fisson dioxygenase. Such a reaction mechanism was not evident in our experiments but seems possible as a further bacterial degradation pathway given the phenol-like structure of thymol. In addition to the trihydroxylated compound 6hydroxythymohydroquinone Chamberlain and Dagley (1968) described the corresponding 6-hydroxythymobenzoquinone, which was probably formed by a non-enzymatic oxidation. In contrast, we only found the unhydroxylated thymobenzoquinone, which could be the product of a non-enzymatic oxidation or an oxidation by an unidentified oxidase. The formation of thymobenzoquinone was also described during transformation reactions of thymol with fungi (Numpaque et al. 2011) and with the microalga Oocystis pusilla (Ghasemi et al. 2009). In the microalga O. pusilla (Ghasemi et al. 2009) the

thymobenzoquinone was not further metabolized which is in accordance with our results using gram-positive bacteria. Only for the fungus Botryodiplodia theobromae, Numpaque et al. (2011) tentatively suggested the formation of two partially reduced derivatives of thymobenzoquinone showing ring hydrogenation. In addition to the formation of thymobenzoquinone and further metabolites of thymobenzoquinone by B. theobromae, thymohydroquinone was hydroxylated at the methyl group by this fungus. The product was tentatively described as 2-hydroxymethyl-5-isopropylbenzene-1,4-diol (Numpaque et al. 2011). The second biotransformation pathway involves hydroxylation of the methyl group to the alcohol (P2), followed by formation of the acid (P3) possibly via the corresponding disubstituted benzaldehyde. These oxidation products of thymol have not been described previously for bacteria, fungi or algae. However, a similar reaction mechanism was described for the methylated phenol 2,5-xylenol (2,5-

1296

dimethylphenol), which shares the same basic structure of thymol, by Pseudomonas alcaligenes (Hopper and Chapman 1971; Poh and Bayly 1980). In this case, the corresponding oxidation steps at the methyl group were catalyzed by methylhydroxylase leading to the formation of the alcohol, which by the action of alcohol dehydrogenase and aldehyde dehydrogenase leads to the formation of the acid (Poh and Bayly 1980). It is likely that these enzymes are also involved in the pathway we describe. In addition, the transformation of the thymol-related compound 2,5-xylenol was postulated to involve a subsequent hydroxylation of the acid in para-position to the first OH-group on the benzene ring (Hopper and Chapman 1971; Poh and Bayly 1980). The aromatic ring of this 4methylgentisic acid (2,5-dihydroxy-4-methylbenzoic acid) was then cleaved by a dioxygenase between the newly formed (second) OH-group and the acid group and was further metabolized to citraconic acid (Poh and Bayly 1980). The same reaction mechanisms were described for 3,5-xylenol (3,5-dimethylphenol) and m-cresol (3-methylphenol) via the dihydroxylated acids 3-methylgentisic acid (2,5-dihydroxy-3-methylbenzoic acid) and gentisic acid (2,5-dihydroxybenzoic acid), respectively (Hopper and Chapman 1971; Poh and Bayly 1980). In contrast to these results using 2,5-xylenol, the formation of a dihydroxylated acid during biotransformation of thymol has not been described to our knowledge. The results we present for N. cyriacigeorgica and M. neoaurum suggest an accumulation of the acid without further hydroxylation at the aromatic ring system which is a prerequisite for ring cleavage. Although N. cyriacigeorgica (Nhi-Cong et al. 2010) and M. neoaurum (Herter et al. 2012) are able to degrade phenylalkanes by oxidation of the aliphatic (Herter et al. 2012) or branched side chain (Nhi-Cong et al. 2010), the isopropyl side chain of thymol was not oxidized. In contrast to our experiments, Numpaque et al. (2011) tentatively described the oxidation of the isopropyl group by plant pathogenic fungi. In summary, we show that the biotransformation of thymol by N. cyriacigeorgica and M. neoaurum proceeds via two distinct oxidation pathways involving either the hydroxylation of the benzene ring or of the methyl group. The biotransformation pathways described highlight the mechanisms of thymol oxidation by grampositive soil bacteria. The description of new metabolites broadens our knowledge derived from analyses of thymol biotransformation in P. putida, plant pathogenic fungi, or microalgae. Acknowledgments Robert Jack (Institute of Immunology, University of Greifswald) is gratefully acknowledged for help in preparing the manuscript. We thank M. Lalk (Institute of Pharmacy, University of Greifswald) for providing NMR data.


References Aeschbach R, Lögliger J, Scott BC, Murcia A, Butler J, Halliwell B, Aruoma OI (1994) Antioxidant actions of thymol, carvacrol, 6-gingerol, zingerone and hydroxytyrosol. Fd Chem Toxic 32:31–36 Ait-Ouazzou A, Cherrat L, Espina L, Loran S, Rota C, Pagan R (2011) The antimicrobial activity of hydrophobic essential oil constituents acting alone or in combined processes of food preservation. Inn Food Sci Emerg Technol 12:320–329 Braga PC, Alfieri M, Culici M, Dal Sasso M (2007) Inhibitory activity of thymol against the formation and viability of Candida albicans hyphae. Mycoses 50:502–506 Chamberlain EM, Dagley S (1968) The metabolism of thymol by a Pseudomonas. Biochem J 110:755–763 Cosentino S, Tuberoso CIG, Pisano B, Satta M, Mascia V, Arzedi E, Palmas F (1999) In-vitro antimicrobial activity and chemical composition of Sardinian Thymus essential oils. Lett Appl Microbiol 29:130–135 Costa C, Lodesani M, Maistrello L (2010) Effect of thymol and resveratrol administered with candy or syrup on the development of Nosema ceranae and on the longevity of honeybees (Apis mellifera L.) in laboratory conditions. Apidologie 41:141–150 Daniel M (2006) Medicinal plants: chemistry and properties. Science, New Hampshire De Boer TD, Backer HJ (1956) Diazomethane. In: Leonard NJ (ed) Organic synthesis, vol 36. Wiley, New York, pp 14–16 Derby R, Rohal P, Jackson C, Beutler A, Olsen C (2011) Novel treatment of onychomycosis using over-the-counter mentholated ointment: a clinical case series. J Am Board Fam Med 24:69–74 Didry N, Dubreuil L, Pinkas M (1994) Activity of thymol, carvacrol, cinnamaldehyde and eugenol on oral bacteria. Pharm Act Helv 69:25–28 Fritsche W (1968) Effect of carbon sources on growth rate protein content and enzyme pattern in Candida guilliermondii. Z Allg Mikrobiol 8:91–99 Fritsche W, Hofrichter M (2005) Aerobic degradation of recalcitrant organic compounds by microorganisms. In: Jördening HJ, Winter J (eds) Environmental biotechnology. Wiley, Weinheim, pp 203– 228 Ghasemi Y, Mohagheghzadeh A, Moshavash M, Ostovan Z, RasoulAmini S, Morowvat MH, Ghoshoon MB, Raee MJ, MosaviAzam SB (2009) Biotransformation of monoterpenes by Oocystis pusilla. World J Microbiol Biotechnol 25:1301–1304 Herter S, Mikolasch A, Schauer F (2012) Identification of phenylalkane derivatives when Mycobacterium neoaurum and Rhodococcus erythropolis were cultured in the presence of various phenylalkanes. Appl Microbiol Biotechnol 93:343–355 Hopper DJ, Chapman PJ (1971) Gentisic acid and its 3-methylsubstituted and 4-methyl-substituted homologues as intermediates in bacterial degradation of meta-cresol, 3,5-xylenol and 2,5-xylenol. Biochem J 122:19–28 Hundt K, Wagner M, Becher D, Hammer E, Schauer F (1998) Effect of selected environmental factors on degradation and mineralization of biaryl compounds by the bacterium Ralstonia picketii in soil and compost. Chemosphere 36:2321–2335 Kamel S, Brazier M, Desmet G, Fliniaux MA, Jacquin-Dubreuil A (1992) Glucosylation of butyric acid by cell-suspension culture of Nicotiana plumbaginifolia. Phytochem 31:1581–1583 Kreisel H, Schauer F (1987) Methoden des mykologischen Laboratoriums. Fischer, Stuttgart Lagouri V, Blekas G, Tsimidou M, Kokkini S, Boskou D (1993) Composition and antioxidant activity of essential oils from oregano plants grown wild in Greece. Z Lebensm Unters Forsch 197:20–23

Appl Microbiol Biotechnol (2013) 97:1289–1297 Nhi-Cong LT, Mikolasch A, Klenk HP, Schauer F (2009) Degradation of the multiple branched alkane 2,6,10,14-tetramethyl-pentadecane (pristane) in Rhodococcus ruber and Mycobacterium neoaurum. Int Biodeterior Biodegrad 63:201–207 Nhi-Cong LT, Mikolasch A, Awe S, Sheikhany H, Klenk HP, Schauer F (2010) Oxidation of aliphatic, branched chain, and aromatic hydrocarbons by Nocardia cyriacigeorgica isolated from oilpolluted sand samples collected in the Saudi Arabian Desert. J Basic Microbiol 50:241–253 Numpaque MA, Oviedo LA, Gil JH, Garcia CM, Durango DL (2011) Thymol and carvacrol: biotransformation and antifungal activity against the plant pathogenic fungi Colletotrichum acutatum and Botryodiplodia theobromae. Trop Plant Pathol 36:3–13 Panda H (2000) Herbal cosmetics handbook. Asia Pacific Business Press, Delhi Panda H (2002) The complete technology book on natural products (forest based). Asia Pacific Business Press, Delhi Pandey SK, Upadhyay S, Tripathi AK (2009) Insecticidal and repellent activities of thymol from the essential oil of Trachyspermum ammi (Linn) sprague seeds against Anopheles stephensi. Parasitol Res 105:507–512 Pinto E, Pina-Vaz C, Salgueiro L, Goncalves MJ, Costa-de-Oliveira S, Cavaleiro C, Palmeira A, Rodrigues A, Martinez-de-Oliveira J (2006) Antifungal activity of the essential oil of Thymus pulegioides on Candida, Aspergillus and dermatophyte species. J Med Microbiol 55:1367–1373 Poh CL, Bayly RC (1980) Evidence for isofunctional enzymes used in meta-cresol and 2,5-xylenol degradation via the gentisate pathway in Pseudomonas alcaligenes. J Bacteriol 143:59–69 Poulose AJ, Croteau R (1978) Biosynthesis of aromatic monoterpenes—conversion of gamma-terpinene to para-cymene and

1297 thymol in Thymus vulgaris L1. Arch Biochem Biophys 187:307–314 Samarasekera R, Weerasinghe IS, Hemalal KDP (2008) Insecticidal activity of menthol derivatives against mosquitoes. Pest Manag Sci 64:290–295 Scora RW (1967) Study of essential leaf oils of genus Monarda (Labiatae). Am J Bot 54:446–452 Shimoda K, Kondo Y, Nishida T, Hamada H, Nakajima N, Hamada H (2006) Biotransformation of thymol, carvacrol, and eugenol by cultured cells of Eucalyptus perriniana. Phytochem 67:2256–2261 Shrestha A, Rimal J, Rao A, Sequeira PS, Doshi D, Bhat GK (2011) In vitro antifungal effect of mouth rinses containing chlorhexidine and thymol. J Dent Sci 6:1–5 Ultee A, Bennik MHJ, Moezelaar R (2002) The phenolic hydroxyl group of carvacrol is essential for action against the food-borne pathogen Bacillus cereus. Appl Environ Microbiol 68:1561–1568 Valero D, Valverde JM, Martinez-Romero A, Guillen F, Castillo S, Serrano M (2006) The combination of modified atmosphere packaging with eugenol or thymol to maintain quality, safety and functional properties of table grapes. Postharvest Biol Technol 41:317–327 Veldhuizen EJA, Tjeerdsma-Van Bokhoven JLM, Zweijtzer C, Burt SA, Haagsman HP (2006) Structural requirements for the antimicrobial activity of carvacrol. J Agr Food Chem 54: 1874–1879 Vokou D, Kokkini S, Bessiere JM (1993) Geographic variation of Greek oregano (Origanum vulgare ssp. hirtum) essential oils. Biochem Sys Ecol 21:287–295 Xu J, Zhou F, Ji BP, Pei RS, Xu N (2008) The antibacterial mechanism of carvacrol and thymol against Escherichia coli. Lett Appl Microbiol 47:174–179