Financial support: CSIC, PDT-DINACYT, PEDECIBA and OPCW.

Keywords: Biocatalysis, Monoterpenes, Eucalyptus, Green Chemistry.

The forest industry in Uruguay has grown considerably during the last decade. Eucalyptus plantations account for 74% of the forested land, with Eucalyptus globulus being the most widely distributed species. This industry is dedicated exclusively to the production of wood without exploiting the by-products (leaves and small branches). Eucalyptus leaves are known to contain important amounts of essential oils composed primarily of 1,8-cineole (1,3,3-trymethyl-2-oxabicyclo[2.2.2]octane). In this work, the biotransformation of 1,8-cineole, is achieved using a native bacterium (Rhodococcus sp.) which was isolated from the soil of Eucalyptus forest. A 98% of bioconversion was achieved. Three different optically pure compounds were obtained, and they were identified as 2-endo-hydroxy-1,8-cineole, 2-exo-hydroxy-1,8-cineole and 2-oxo-1,8-cineole.

The development of the forest industry in Uruguay locates it as one of the industrial activities that has increased the most in the last three decades.

Eucalyptus species account for 74% of the forested area, with Eucalyptus globulus being the main cultivar showing an increase of 264.798 hectares in the period 1975-2003 (MGAP, 2004). This industry is dedicated exclusively to the production of wood or pulp for the paper industry. As a result, no use has been planned for leaves and branches, which are left on the ground. Recent estimations indicate that in Uruguay this discarded material reaches 48.000 Kg per hectare, after the logging of the trees.

Eucalyptus leaves are rich in essential oils (FAO, 1995). These oils have a recognized allelopathic action (Romagni et al. 2000a; Romagni et al. 2000b) for this reason its extraction before returning the leaves to the soil would be an ecologically advisable practice. Eucalyptus essential oil, that is commercialized at a price of U$S 3.5 Kg, can be easily obtained from leaves and branches by steam distillation (Clark, 2000).

The main component of Eucalyptus essential oil in most species is 1,8-cineole, representing about 70% of the total oil by gas chromatography (Lawrence, 1997). This monoterpene is easily obtained from the brute essence through precipitation upon cooling (Guenther, 1975). This compound has a market value of U$S 8/Kg. The oxidized derivatives of 1,8-cineole represent a set of compounds of high potential as chiral synthon for organic chemistry. Besides, several oxygenated terpenes have shown wide utility in the scent industry as a consequence of their fragrances (Asakawa et al. 1988; Genta et al. 2002). Therefore, hydroxylation of 1,8 cineole would increase its market value.

Production of these derivatives implies the stereo specific introduction of molecular oxygen in not activated carbon atoms, which continues to be a challenge in organic synthesis (Liu and Rosazza, 1990; Roberts et al. 2002). The use of microorganism that carries out this type of reactions constitutes an interesting alternative. Microbial hydroxylations have advantages over classic organic synthesis procedures since they are carried out in soft conditions, they use biodegradable reagents and they are generally stereo selective resulting in the production of an optically pure synthon (Faber, 1995).

In addition, the products obtained by this type of methodology can be labelled and commercialized as GREEN products.

In this work the results on the biotransformation of 1,8-cineole using a native Rhodococcus sp. strain isolated from soil obtained from beneath Eucalyptus sp. trees are presented.

Bacterial strain

A native Rhodococcus sp. strain was isolated by our laboratory from soil obtained from beneath Eucalytus sp. trees. This strain is maintained in our microbial collection at -70ºC.

Biotransformations

Biotransformations of 1,8-cineole were carried out by fermentation under different conditions.

Optimized methodology for biotransformation: Rhodococcus sp. strain was plated in LB and incubated for 5 days at 30ºC. This culture was used to inoculate 50 mL LB containing 0.1% cineole and the culture was incubated in an orbital shaker (Sanyo IOX400. XX2. C) at 30ºC and 200 rpm for 5 days. The cells were washed with BG-11 minimum media and centrifuged at 5000 rpm and 4ºC for 10 min (this operation is repeated three times). The obtained pellet was used to inoculate the biotransformation. The biotransformation media used was a modified BG-11 containing 0.1% 1,8-cineole. The variables that were studied comprised the inoculum's size (2 x 10⁷, 5 x 10⁷ and 1 x 10⁹ cells/mL) and the reaction time (24, 48 and 67 hrs).

Extraction of metabolites

The cells in the culture media were sonicated for 5 min and extracted with an equal volume of dichloromethane. The extract was dried over anhydrous Na₂SO₄ and concentrated under reduced pressure.

Identification of obtained products

The obtained products were analyzed by TLC, GC, GC-MS, GC-GC.

Gas chromatography was carried out in a HEWLLET PACKARD 5890 serie II equipped with a FID detector and a CARBOWAX capillary column (25 m x 0.25 mm of internal diameter). Temperature program: 60ºC/8 min, 60 to 210ºC at 3ºC/min; injector temperature 240ºC, detector temperature 250ºC; carrier gas: He, 2 mL/min; injection system: split (ratio 1:50). Internal standard: limonene oxide. The data was analyzed with the EZChrome software.

HRGC-MS was carried out in a Shimadzu QP 5050 in the same conditions mentioned above. The identifications were completed by comparison of mass spectra with literature (Adams, 2001).

The optical purity was determined in a GC-GC Shimadzu GC 17A. The first GC is equipped with a SE52 column and the second one with a modified β-ciclodextrin chiral capillary column. Temperature program: 50ºC (6 min), 50-90ºC at 2ºC/min, 90ºC (20 min); 90-180ºC at 2ºC/min, 180ºC (10 min); injector temperature 250ºC; detector temperature 280ºC; carrier gas: He; injection system: split (ratio 1:150).

Reagents

1,8-cineole was provided by the Center of Agroindustrial Technology of Cochabamba, Bolivia.

Limonene Oxide: Aldrich, 97%.

During the initial screening three biotransformation products were detected and identified as 2-endo-hydroxy-1,8-cineole, 2-exo-hydroxy-1,8-cineole and 2-oxo-1,8-cineole (Figure 1).

Chiral GC analysis indicated that all these compounds are optically pure.

In order to optimize biotransformation parameters to obtain the best percent conversion we analyzed two variables. First, the influence of the inoculum's size was analyzed and it was found that the best conversion is achieved for the larger inoculum's size, being 2-endo-hydroxy-1,8-cineole the major product obtained (Figure 2).

We also analyzed the effect of the reaction time on the biotransformation yield and product profile. These studies were conducted using the larger inoculum's size from the previous assay, and following the production of biotransformation products at different reaction times. The results indicate that the better yield is obtained at 24 hrs, with 2-endo-hydroxy-1,8-cineole accounting for 56% of the total product (Figure 3).

Table 1 summarizes the best results obtained with the Rhodococcus sp. strain isolated by our group as well as previous reports with other strains of bacteria, including a Rhodococcus sp. strain. The Rhodococcus sp. strain isolated by our group presents a major percent conversion that the strain described before; although it produces a third metabolite. An important feature of the biocatalyst described in this paper is the stereo selectivity, since only one enantiomer of each product is obtained.

A Rhodococcus sp. strain capable of metabolizing 1,8-cineole was isolated from soil beneath Eucalyptus sp. Three compounds were obtained from the biotransformation of 1,8-cineole with this strain and they were identified as 2-endo-hydroxy-1,8-cineole, 2-exo-hydroxy-1,8-cineole and 2-oxo-1,8-cineole. Chiral GC analysis indicated that these three compounds were optically pure. The biotransformation conditions were optimized to reach 98% bioconversion with this strain, what represents a better percent conversion than those previously reported for the biotransformation of 1,8-cineole with other bacterial strains. Despite this advantage, our strain shows low stereo selectivity since both the 2-endo- and the 2-exo-hydroxy-1,8-cineole are produced and conditions should be optimized to achieve only one stereoisomer.

The authors wish to thank Dr. Carmen Rossini and Dr. Andrés González for technical support. They also want to thank Bach. Daniel Lorenzo and Dr. Eduardo Dellacasa for chiral GC analysis.