SciELO - Scientific Electronic Library Online

Home Pagealphabetic serial listing  

Services on Demand




Related links


Journal of the Chilean Chemical Society

On-line version ISSN 0717-9707

J. Chil. Chem. Soc. vol.50 no.4 Concepción Dec. 2005 


J. Chil. Chem. Soc., 50, N° 4 (2005), págs: 715-718





aFacultad de Ciencias, Universidad del Bío-Bío, Chillán, Chile. E-mail:
bFacultad de Ciencias Químicas, Universidad de Concepción, Concepción, Chile
cInstitut für Organische Chemie und Strukturanalytik, Universität Potsdam, Potsdam, Germany

A new synthesis of 9a-hydroxy-a-agarofuran (6a) is described, using a microbiological hydroxylation a-agarofuran (5) as the key reaction. The stereochemistry of the biohydroxylation was determined on the basis of a NOESY-experiment and GIAO calculations at the B3LYP/cc-pVDZ level. A strong g-effect was observed at C15 of the agarofuran ring which was correctly predicted by the GIAO-B3LYP calculations.

Key words: Agarofuran Synthesis, Celastraceae, Eudesmanes, Microbiological hydroxylation, Sesquiterpenes.


Dihydroagarofurans are polyhydroxy sesquiterpenes which have been isolated from plants of the family Celastraceae. Some of these compounds exhibit insecticidal, insect antifeedant, cytoxic, antitumor, and inmunosuppresive activities [1-6]. Various approaches for the synthesis of hydroxylated agarofurans have been reported in the literature generally giving unsatisfactory yields [7,8]. A major problem is the control of the stereochemistry of the hydroxyl groups. According to Huffman et al. [9,10], 9-oxo-a-agarofuran is prepared from oxycarvone and subsequently reduced to give an epimeric mixture of 9a-hydroxy-a-agarofuran. Synthesis of 9b-hydroxy-a-agarofuran has been reported in a six-step procedure with an overall yield of 15 %. We have shown previously that treatment of 10-epieudesman-4-en-3,9,11-triol with acid gives 6 in 20 % yield [11].

The use of fungi for hydroxylations of terpenoid substrates has been extensively documented in the literature [12,13]. However, nothing has been reported on microbiological hydroxylation of agarofuran. We now report microbiological hydroxylation of a-agarofuran (5) at C-9 by the fungus Rhizopus nigricans.


1H and 13H NMR (300 and 50 MHz, resp.): Bruker AM-300; solvent: CDCl3. The assignments of 13H NMR chemical shifts were made by means of a DEPT experiments, using a flip angle of 135°. Analytical plates (silica gel, Merck 60 G) were rendered visible by spraying with H2SO4, followed by heating to 120°C.

Organism, media and culture conditions

Rhizopus nigricans (LSBPN025) was obtained from Colección de Cultivos, Laboratorio de Microbiología, Facultad de Ciencias de la Salud y de los Alimentos, Universidad del Bío-Bío, Chillán, Chile, and was kept in Hagen medium containing CaCl2 (0.005 %), KH2PO4 (0.0025 %), (NH4)2HPO4 (0.025 %), MgSO4 C 7 H2O (0.015 %), FeCl3 (0.0012 %), malt extract (0.3 %) and glucose (1 %) in H2O at pH 5. Erlenmeyer flasks (250 ml) containing 100 ml of medium, were inoculated with a dense suspension of R. nigricans. The cultures were incubated with shaking (150 rpm) at 28 °C for 5 days, after that ethanol substrate solution was added.

Biotransformation of 5.

The substrate 5 (300 mg) was dissolved in ethanol (3 ml). The substrate solution was added in three equal portions to three Erlenmeyer flask cultures, each containing 100 ml of medium. After incubation in a shaker at 28 °C for 5 days, the cultures were filtered over filter paper and cells were washed thoroughly with EtOAc. The filtrates were pooled and extracted with EtOAc. The combined extract was dried over Na2SO4 and evaporated under reduced pressure to give a mixture of compounds (260 mg). Chromatography on silica gel using n-hexane: AcOEt mixtures afforded 105 mg (35 %) of the starting material 5, 120 mg (40 %) of 6a which crystallized from n-hexane, mp 79-80 °C, and 35 mg of a mixture of hydroxylated compounds which have not been identified yet. Spectroscopic data of 6a: IR (nujol): 3675 cm-1. 1H NMR: d = 5.63 (s, 1 H), 3.92-3.88 (dd, J= 6.32 and 3.2 Hz, 1 H), 2.19-1.41 (m, 9 H), 1.70 (s, 3 H), 1.33 (s, 3 H), 1.21 (s, 3 H), 0.82 (s, 3 H). 13H NMR: see Table 1. MS (EI, 70 eV): m/z (%) = 236 (19) [ M+ ], 218 (74), 221 (74), 203 (23), 175 (75), 160 (77), 145 (58), 121 (100), 105 (26), 93 (33), 79 (14), 55 (16).

Computational aspects

Initial geometry optimizations were done at the HF/6-31G* level. NMR shielding calculations were carried out by the Gauge Invariant Atomic Orbital (GIAO approach) [14,15] at B3LYP level using cc-pVDZ basis set on the optimized HF/6-31G* geometries[16,17]. The calculated 13H NMR chemical shifts are reported with respect to TMS. All calculations were performed with the Gaussian 94 program.


Following a published procedure [11], an epimeric mixture of 10-epieudesm-4-ene-3,11-diol (3) was prepared in four steps starting from (+)-dihydrocarvone (1) (48 % overall yield).

Treatment of 3 with conc. H2SO4 afforded a-agarofuran (5) in 85 % yield (Fig. 1). In a typical aerobically fermentation, substrate 5 was incubated with Rhizopus nigricans for 5 days. Extraction of the reaction mixture followed by chromatography gave 35 % of unconverted 5, 57.4 % of 6 (yield calculated from the amount of 5 converted) and a mixture of other, not yet identified hydroxylation products (ca. 17 %). The hydroxylation at C-9 by Rhizopus nigricans can give two diasteromeric product 9a-hydroxy-a-agarofuran (6a) or 9b-hydroxy-a-agarofuran (6b). The identification of the major product was performed by using both experimental (NOESY) and theoretical (B3LYP-GIAO) methods. Both procedures showed that the 6a was the principal biotransformation product of 5 by the fungi, showing the regio- and steroselectivity of the biotransformation process. Tables 1 and 2 displays the calculated and experimental 13H and 1H NMR spectra for 5, 6a and 6b derivatives respectively. Excellent agreement between the calculated and experimental 13H NMR spectra for 5 was found. All the calculated 13H NMR signals had a deviation of less than 4 ppm with respect to the experimental one. Also, good agreement between calculated and experimental 1H NMR for 5 and 6a were found.

Fig. 1. Chemical and microbiological synthesis of 9a-hydroxy-a-agarofuran (6a). a) KOH,EtOH, reflux 1 h. b) m-CPBA, CHCl3, room temperature. c) LiAlH4, Et2O. d) H2SO4, toluene.

Hydroxylation at C-9 of 5 should induce large shifted at C-9, C-10, C-8 and C-15. Especially, 13H NMR chemical shift of C15 should be very sensitive to the hydroxyl configuration. In the case of 6a derivatives the g-effect, attributed to the steric compression of gauche interactions, should induce the shielding of C15 NMR signal. On the other hand, the 6-b derivative, which posses the hydroxyl group anti respect to C-15 should exhibit a negative g-anti effect due to interactions between the backwards-facing lobes of the C-9-O and C-15-H bonding orbitals [18].

Thus, for the 6a derivative the C-15 calculated chemical shift was 16.08 ppm while for the 6b derivative was 23.17 ppm. The experimental value of C-15 chemical shift was 16.01 ppm, therefore there is a close correspondence between calculated and experimental chemical shift for the 6a derivative. These features are maintained for the other chemical shifts of the 6a derivative. Thus, the theoretical results at B3LYP-GIAO level suggested that the major metabolite of the biotransformation of 5 by R. nigricans was the 6a derivative.

In addition, the 1H NMR spectrum of the biotransformation product showed singlets at d 1.70, 1.34, 1.21 and 0.83 ppm, corresponding to four methyl groups. A singlet at 5.6 ppm corresponding to proton at C-3. The proton at C-9 appears at d = 3.79 ppm with a coupling constants of 10.7 and 5.3 Hz. The calculated chemical shift for the proton at C-9 of 6a derivative is 3.82 ppm, where as H-9 chemical shift for 6b derivative is 3.38 ppm (Table 2). The 13H NMR spectrum confirmed a CH(OH) resonance at d = 73.85 ppm which was confirmed by a HETCOR experiment. The orientation of H-9 of compound 6 was determined by an NOE experiment. Irradiation of the 1H NMR signal at d 3.92-3.88 only showed dipolar coupling with signals of the methyl group appearing at d 1.21 (H-12) suggesting that H-9 is axial (Fig. 1). If H-9 was equatorial, it would show dipolar coupling with H-15. From the above spectral evidence, experimental and theoretical, we concluded that 6 has the structure of 9a-hydroxy-a-agarofuran (6a). This assignment was supported by the low intensity of the signal of the Mz [M - 18]+ ion in the mass spectrum.

In conclusion, we have shown that the microbiological hydroxylation of a-agarofuran (5) gives 6a in higher yield (40%) than a chemical synthesis. This procedure may be a convenient new route to synthesize polyhydroxyagarofurans. In addition, we have shown that GIAO methods at B3LYP/cc-pVDZ level predicted reliable 13H NMR chemical shifts and the substitution of an equatorial hydrogen by hydroxyl group induces a strong g-gauche effect on the C-15 methyl group. This effect produces a shielding of 5.1 ppm at C15 as it was inferred from GIAO calculation. This value is exactly identical to the experimental shielding observed.

Fig. 2. Computer-generated drawing of compound 6a which showed NOE between H-12 and H-9.


We are grateful to Dirección de Investigación de la Universidad del Bío-Bío Proyecto DIUBB 033407-3R y a la Dirección de Investigaciones de la Universidad de Concepción for the financial support. J. Alarcón gratefully acknowledges a stipend from Deutscher Akademischer Austauschdienst (DAAD).



[1] Y.H.Kuo, M. L. King, C. F. Chen, C. H. Chen, K. Chen and K. H. Lee, J. Nat. Prod., 57, 263(1994).         [ Links ]

[2] Y. Takishi, K. Ujita, H. Tokuda, H. Nishino, A. Iwashima and T. Fujita, Cancer Lett., 65, 19(1992).         [ Links ]

[3] Y. L. Zheng, L. Xu and J. F. Liu, Acta Pharm. Sin., 24, 568(1989).         [ Links ]

[4] A. G. González, I. A. Jiménez, A. G. Ravelo, J. Coll, J. A. González, and J. Lloria, Biochem System. Ecol., 25(6), 513 (1997).         [ Links ]

[5] A. G. González, I. A. Jiménez, A. G. Ravelo, X. Bellés, and M. D. Piulachs, Biochem. System. Ecol., 20(4), 311(1992).         [ Links ]

[6] J. Alarcón, J. Becerra, M. Silva, Bol. Soc. Chil. Quím., 43, 65 (1998).         [ Links ]

[7] X. Chen, S. C. Shao, T.L Li and Y. L. Li, Synthesis, 1061 (1992)         [ Links ]

[8] Spivey A.C., Weston M. and Woodhead S. Chem. Soc. Rev. 31, 43-59 (2002), and references therein.         [ Links ]

[9] J. W. Huffman and R. C. Desai, J. Org. Chem., 47(17), 3254(1982).         [ Links ]

[10] J. W Huffman and G. F. Hillebrand, Tetrahedron Suppl., 37, 269(1981).         [ Links ]

[11] J. Alarcón, J. Alderete, M. G. Peter, J. Becerra and M. Silva, Bol. Soc. Chil. Quím., 43, 325(1998).         [ Links ]

[12] H.L. Holland and H.K. Weber. Curr. Opin. Biotech. 11, 547-553 (2000)         [ Links ]

[13] V. Lamare and R. Furstoss, Tetrahedron, 46(12), 4109(1990).         [ Links ]

[14] K. Wolinski, J. F. Hinton and P. Pulay, J. Am. Chem. Soc., 112, 8251(1990).         [ Links ]

[15] R. Ditchfield, Mol. Phys., 27, 789(1974).         [ Links ]

[16] A.D. Becke, J. Chem. Phys., 98, 5648 (1993).         [ Links ]

[17] D. E. Woon and T. H. Dunning Jr., J. Chem. Phys., 98, 1358 (1993) .         [ Links ]

[18] H. O. Kalinowski, S. Berger, S. Braun, Carbon-13 NMR Spectroscopy, p. 278-279, John Wiley & Sons New York (1991).        [ Links ]

Creative Commons License All the contents of this journal, except where otherwise noted, is licensed under a Creative Commons Attribution License