Introduction

Biocatalysis, also known as “white biotechnology,” has been applied in many spheres of activity aiming of more environmentally friendly processes, such as the production of renewable raw materials and clean energy, and the remediation of the environment contaminated by toxic compounds (Gavrilescua and Chisti 2005). It is estimated that by the year 2020, biocatalysis will be responsible for 20 % of global production of all synthetic chemicals (Hans-Peter and Werbitzky 2011).

Biocatalysts, in the form of enzymes or whole microorganisms, are currently established as useful tools in the production of fine chemical (Liu et al. 2004). Factors such as production costs, biological activity, and regulations issues have led to a growing demand for enantiopure products for chiral drugs (Matsuda et al. 2009; Wang et al. 2009). A novel one-pot tandem biohydrogen transfer process for obtaining enantiopure secondary alcohols was described in the literature (Bisogno et al. 2009). There is a number of works on enzymatic reactions involving biocatalysts of terrestrial origin, but little using marine organisms or their purified enzymes (Trincone 2010; Veberlen et al. 2006).

Marine organisms such as seaweeds, sponges, mollusks, and microorganisms have sets of enzymes that present very useful characteristics for biocatalysis, such as stability at high and low temperatures or at extremes pH, and activity at high pressure (Antranikian et al. 2005). Marine fungi produce extracellular enzymes and secondary metabolites different from terrestrial microorganisms, because they are adapted to the specific conditions in marine ecosystems, such as high salinity, which can be used in biocatalytic reactions (Bugni and Ireland 2004; Cheng et al. 2009). The use of whole cells of microorganisms is advantageous, particularly to obtain enantiomerically pure compounds.

Chiral β-azido alcohols are precursors in the synthesis of chiral aziridines and amino alcohols (Besse et al. 1994). There has been growing interest in chiral aziridines and amino alcohols in organic synthesis due to the increasing importance of bioactive molecules containing these functionalities, e.g., enzyme inhibitors, such as salmeterol and albuterol (Pàmies and Bäckvall 2001; Brenelli and Fernandes 2003).

In addition, marine organisms are rich sources of compounds that exhibit significant antitumor, anti-inflammatory, analgesic, antiviral, and antiallergic activities (Newman and Cragg 2004; San-Martín et al. 2008). Recently, we employed marine fungi to catalyze the transformation of several xenobiotic compounds (Rocha et al. 2009, 2010; Martins et al. 2011). In the present study, we report the first screening of marine-derived fungi for asymmetric reduction of α-keto azides 5–8.

Experimental

General Methods

The reagents 2-bromo-1-phenylethanone (1), 2-bromo-1-(4-methoxyphenyl)ethanone (2), 2-bromo-1-(4-bromophenyl)ethanone (3), and 2-bromo-1-(4-nitrophenyl)ethanone (4), and solvents were purchased from Sigma-Aldrich (São Paulo, SP, Brazil). All manipulations involving the fungi (Aspergillus sclerotiorum CBMAI 849, Cladosporium cladosporioides CBMAI 857, Penicillium raistrickii CBMAI 931, Penicillium citrinum CBMA 1186, Mucor racemosus CBMAI 847, Beauveria felina CBMAI 738, Penicillium chrysogenum CBMAI 1199, and Penicillium oxalicum CBMAI 1185) were carried out under sterile conditions in a Veco laminar flow cabinet. Technal TE-421 or Superohm G-25 orbital shakers were employed in the biocatalyzed experiments. The products were purified by column chromatography (CC) on silica gel (230–400 mesh) eluting with mixtures of n-hexane and EtOAc (9:1, 8:2, 7:3). The fractions were monitored by TLC, on pre coated plates with silica gel 60F254 layers (aluminum-backed: Sorbent).

The products were analyzed using Shimadzu LC-10AD® or Shimadzu 20AT® chromatographs with UV detector (190–254 nm) equipped with a Chiralcel® OD-H chiral column (0.46 cm × 25 m; 5 μm); the mobile phase was hexane and 2-propanol (95:5), flowing at 0.5 mL min−1. The Shimadzu LC-10AD model chromatograph was equipped with a photodiode array SPD-M10A detector, a DGU-14A degasser, a SCL-10A control center, and a manual Rheodyne® injector. The Shimadzu LC-20AT model was equipped with a diode array SPD-M20A detector, a DGU-20A5 degasser, a CBM-20A control center, a Sil-20A automatic injector, and a CTO-20A oven. Acquisition and data analysis were performed using the LCSolution/CLASS-VP software. A Shimadzu GC 2010 plus gas chromatography system coupled to a mass selective detector (Shimadzu MS 2010 plus) in electron ionization (EI, 70 eV) mode and equipped with a DB5 fused silica column (J&W Scientific 30 m × 0.25 mm × 0.25) was employed for obtaining the mass spectra. 1H and 13C NMR spectra were registered on a Bruker AC-200 spectrometer (1H at 200 MHz and 13C at 50 MHz), using chloroform-d (CDCl3) as solvent and tetramethylsilane (TMS) as internal standard; the chemical shifts are given in parts per minute. Near-IR spectra were recorded on a Bomem MB-102 spectrometer.

Separations of the enantiomers of β-azidophenylethanols were archived by HPLC analysis on a chiral column. The retention times of the compounds were the following: (R-9, 35.0 min; S-9, 33.0 min), (R-10; 27.0 min; S-10, 23.0 min), (R-11, 24.0 min; S-11, 28.0 min), and (R-12, 21.2 min; S-12, 23.1 min).

Preparation of the α-Azido Ketones 5–8

A mixture of the α-bromoacetophenones 1–5 (14.00 mmol) and sodium azide (1.82 g, 28.00 mmol) in acetone (250 mL) was stirred at room temperature, and the reaction was monitored by TLC (Scheme 1). The mixture was poured into water and extracted with EtOAc (3 × 30 mL); the organic layer was dried with MgSO4, filtered, concentrated under reduced pressure, and purified by crystallization or by column chromatography (Patonay et al. 2002). The spectroscopic data (1H and 13C NMR, MS and IR) of the α-azido ketones 5–6 and 8 were in agreement with those reported in the literature (Patonay et al. 2002; Patonay and Hoffman 1994).

Scheme 1
scheme 1

Synthesis of (±)-β-azidophenylethanols 9–12 from α-bromoacetophenones 1–4

2-Azido-1-(4-bromophenyl)ethanone (7): 1H NMR (400 MHz, CDCl3): δ 4.53 (s, 1H); 7.66 (d, 2H, J = 8.0 Hz); 7.79 (d, 2H, J = 8.0 Hz). IR (KBr) 2102 (N3), 1690 (C = O), 1554, 1228, 1215, cm−1.

Preparation of the (±)-β-Azidophenylethanols 9–12

(±)-β-Azidophenylethanols 9–12 were obtained by reduction of the α-azido ketones 5–8 with sodium borohydride in ethanol (Scheme 1) (Pavia et al. 1999). The spectroscopic data (1H and 13C NMR, MS and IR) of the (±)-β-azidophenylethanols 9–12 were in agreement with those reported in the literature (Kamal et al. 2004a, b; Brenelli and Fernandes 2003; Mesas-Sánchez et al. 2013; Ankati et al. 2008).

Marine Fungal Strains

The marine fungal strain P. raistrickii CBMAI 931 was isolated from the sponge Chelonaplysylla erecta. The strains P. citrinum CBMAI 1186, P. oxalicum CBMAI 1185, and B. felina CBMAI 738 were isolated from a marine alga Caulerpa sp. (De Vita-Marques et al. 2008). The fungal strains A. sclerotiorum CBMAI 849, C. cladosporioides CBMAI 857, and M. racemosus CBMAI 847 were isolated from the Brazilian cnidarian zoanthids, i.e., Palythoa variabilis, Palythoa caribaeorum, and Mussismilia hispida, respectively, collected in São Sebastião City, northern coast of the State of São Paulo, Brazil (Santos et al. 2010). The terrestrial fungal strain P. chrysogenum CBMAI 1199 was donated by Prof. Dra. Mirna Helena Regali Seleghim from the Department of Ecology and Evolutionary Biology (DEBE) at UFSCar, São Carlos, Brazil. The fungi used in this work were identified by both conventional and molecular methods at the Chemical, Biological and Agricultural Multidisciplinary Research Center (CPQBA) at UNICAMP, São Paulo, Brazil (http://www.cpqba.unicamp.br/).

Bioreduction of the α-Azido Ketones 5–8 by Whole Cells of Marine Fungi

The marine-derived fungi were grown in culture media and artificial seawater (Rocha et al. 2009). Small slices of solid medium (0.5 × 0.5 cm) bearing mycelia of the marine-derived fungi (A. sclerotiorum CBMAI 849, C. cladosporioides CBMAI 857, P. raistrickii CBMAI 931, P. citrinum CBMA1186, M. racemosus CBMAI 847, B. felina CBMAI 738, P. chrysogenum CBMAI 1199, and P. oxalicum CBMAI 1185) were cut from the stock solid culture. They were incubated at 32 °C for 9 days in a rotary shaker (130 rpm), filtered using Buchner funnels, and suspended in 250-mL Erlenmeyer flasks containing 100 mL of phosphate buffer solution. The each biocatalytic reactions was carried out with 5.0 g mycelium (wet weight), 0.50 mmol of the α-azido ketones 5–8, and inoculated into 2-L Erlenmeyer flasks containing 1000 mL of the liquid culture medium dissolved in 300 μL DMSO in the buffer solution (0.1 M, Na2HPO4/KH2PO4, pH 7). The mixtures were incubated for 9 days in an orbital shaker at 32 °C and 120 rpm. The progress of the reactions was checked by TLC analysis for 2.0 mL of samples. For compound 6, the reactions were periodically analyzed (1–9-day intervals) by HPLC (Table 1). The samples were extracted with ethyl acetate (1.0 mL), and then centrifuged (6000 rpm, 6 min) and analyzed by GC-MS and HPLC. The products were purified by column chromatography over silica gel to yield the (R)- or (S)-β-azidophenylethanols 9–12 (Table 2).

Table 1 Bioreduction of 2-azido-1-(4-methoxyphenyl)ethanone (6) by marine-derived fungi (32 °C, 130 rpm)
Table 2 Bioreduction of α-azido ketones 5–8 by marine-derived fungi ( 32 °C, 130 rpm, 9 d)

Assignment of the Absolute Configuration

The optical rotations of the β-azidophenylethanols 9–12 were measured in a PerkinElmer (Waltham, MA, USA) 241 polarimeter with a 1-dm cuvette and referenced to the Na-D line. The absolute configurations of the β-azidophenylethanols 9–12 were determined by comparing the sign of the measured specific rotation with those reported in the literature (Kamal et al. 2004a, b; Brenelli and Fernandes 2003; Mesas-Sánchez et al. 2013; Ankati et al. 2008).

(R)-(−)-2-azido-1-phenylethanol (9): [α]D 25 = −20.2 (c 0.82, CH3OH; 90 % ee); Lit. (Ankati et al. 2008) [α]D 22 = +104.9 (c 0.67, CH3OH, 99 % ee, for S-enantiomer). The R-9 was obtained by reduction of ketone 5 with the fungus P. raistrickii CBMAI 931.

(R)-(−)-2-azido-1-(4-methoxyphenyl)ethanol (10): [α]D 25 = −76.0 (c 0.85, CH3OH, >99 % ee); Lit. (Ankati et al. 2008) [α]D 22 = +100.9 (c 1.17, CH3OH, 99 % ee), for S-enantiomer. The R-10 was obtained by reduction of ketone 6 with the fungus C. cladosporioides CBMAI 857.

(S)-(+)-2-azido-1-(4-bromophenyl)ethanol (11): [α]D 25 = +40.3 (c 1.12, CH3OH, 39 % ee); lit. (Ankati et al. 2008) [α]D 25 = +61.3 (c 0.67, CH3OH, >99 % ee). The S-11 was obtained by reduction of ketone 7 with the fungus A. sclerotiorum CBMAI 849.

(S)-(+)-2-azido-1-(4-nitrophenyl)ethanol (12): [α]D 25 = +30.5 (c 0.64, MeOH, 28 % ee); Lit. (Ankati et al. 2008) [α]D 25 = +87.9 (c 0.17, CH3OH, 99 % ee). The S-12 compound was obtained by reduction of ketone 8 with the fungus P. citrinum CBMAI 1186.

Results and Discussion

Bioreduction of α-azido ketones 5–8 was performed for whole mycelia of seven marine fungi (A. sclerotiorum CBMAI 849, C. cladosporioides CBMAI 857, P. raistrickii CBMAI 931, P. citrinum CBMA1186, M. racemosus CBMAI 847, B. felina CBMAI 738, P. oxalicum CBMAI 1185), and terrestrial fungus P. chrysogenum CBMAI 1199. The fungi were screened for the stereoselective reduction of ketones 5–8. The results are shown in Scheme 2 and in Tables 1 and 2.

Scheme 2
scheme 2

Bioreduction of α-azido ketones 5–8 to (S)- or (R)- β-azidophenylethanols 9–12 by marine-derived fungi

For compound 6, the reactions were periodically analyzed (1–9-day intervals) by HPLC (Table 1). After establishing the best reaction time (9 days) for all reactions, the products were analyzed by HPLC on a chiral column, to determine the conversions and enantiomeric excesses (Table 2). The progress of the reactions for compounds 5, 7–8 was followed by TLC analysis of aliquots collected from 3 to 9-day reaction time.

The fungi A. sclerotiorum CBMAI 849 and M. racemosus CBMAI 847 were able to catalyze the reduction of 2-azido-1-phenylethanone (5) to the corresponding (S)-2-azido-1-phenylethanol (9). A. sclerotiorum CBMAI 849 gave S-9 in 98 % conversion and 85 % ee, and M. racemosus CBMAI 847 gave S-9 in 90 % conversion and 72 % ee, in accordance with the anti-Prelog rule. The (S)-azido-alcohol 9 was obtained with 90 and 86 % isolated yields, respectively, after purification by column chromatography. P. raistrickii CBMAI 931 and B. felina CBMAI 738 catalyzed the reduction of the keto azide 5 to the corresponding (R)-2-azido-1-phenylethanol (9) in high conversion and enantiomeric excess following the Prelog rule (Table 2). The (R)-azido-alcohol (9) was isolated with 82 % (P. raistrickii CBMAI 931) and 91 % (B. felina CBMAI 738) yields. C. cladosporioides CBMAI 857, P. citrinum CBMAI 1186, and P. chrysogenum CBMAI 1199 were unable to catalyze the reduction of the keto azide 5.

The fungi A. sclerotiorum CBMAI 849, C. cladosporioides CBMAI 857, P. raistrickii CBMAI 931, and P. citrinum CBMA 1186 catalyzed the bioreduction of 2-azido-1-(4-methoxyphenyl)ethanone (6) to the corresponding (R)-2-azido-1-(4-methoxyphenyl)ethanol (10) with high conversions (c = 68–100 %) and enantiomeric excesses (>99 % ee) in 9-day reaction time, following the Prelog rule (Table 2). The (R)-azido-alcohol 10 was obtained with 62–94 % yield after purification by column chromatography. The fungi M. racemosus CBMAI 847, B. felina CBMAI 738, and P. chrysogenum CBMAI 1199 were unable to catalyze the bioreduction of the keto azide 6.

The fungi A. sclerotiorum CBMAI 849, C. cladosporioides CBMAI 857, P. raistrickii CBMAI 931, P. citrinum CBMAI 1186, and M. racemosus 847 CBMAI catalyzed the bioreduction of 2-azido-1-(4-bromophenyl)ethanone (7) to the corresponding (S)-2-azido-1-(4-bromophenyl)ethanol (11) with different conversions (c = 32–98 %) and enantiomeric excesses (29–70 % ee) in 9-day reaction time, following the anti-Prelog rule. In this case, only the (S)-azido-alcohol 11 was obtained from the fungus P. raistrickii CBMAI 931 with 30 % isolated yield. The fungi B. felina CBMAI 738 and P. chrysogenum CBMAI 1199 did not catalyze the reduction of the keto azide 7 after 9-day reaction time (Table 2). Bioreduction of the α-azido ketones 5–7 by Saccharomyces cerevisiae was reported with 91–94 % isolated yields and 97–100 % enantiomeric excesses (Yadav et al. 2001). The same compounds were reduced by Geotrichum candidum (isolated yield 93–99 and 40–99 % ee), Rhodotorula glutinis (isolated yield 88–97 and >99 % ee) (Fardelone et al. 2006), and Daucus carota root (isolated yield 58–77 and 97–100 % ee) (Yadav et al. 2002).

The 2-azido-1-(4-nitrophenyl)ethanone (8) is unstable under light even at room temperature. Possibly the low solubility in aqueous media made the bioreduction of keto azide 8 harder to perform. However, the fungi P. citrinum CBMAI 1186, P. chrysogenum CBMAI 1199, and P. oxalicum CBMAI 1185 were able to catalyze the bioreduction of the keto azide 8. P. oxalicum CBMAI 1185 and P. citrinum CBMAI 1186 catalyzed the bioreduction of keto azide 8 to the corresponding (S)-2-azido-1-(4-nitrophenyl)ethanol (12) in high conversion and modest enantiomeric excess (Table 2). The (S) and (R)-azido-alcohols 12 were produced with 17–42 % yields after purification by column chromatography (Table 2).

In summary, enantiomerically pure azido alcohols 9–12 were obtained by bioreduction of the corresponding ketones using whole cells of marine-derived fungi. These compounds are precursors of chiral aziridines and amino alcohols used in drugs synthesis. For example (R)-(−)-octopamine a potent chiral drug possessing β-adrenergic activity was prepared from (R)-2-azido-1-phenylethanol (9). The (R)-2-azido-1-(4-methoxyphenyl)ethanol (10) was employed in the synthesis of (R)-(−)-tembamide and (R)-(−)-aegeline used in traditional Indian medicine. It has been shown that these compounds have good hypoglycemic activity (Sadyandy et al. 2005).

Finally, the strains of marine-derived fungi used in this study showed potential for reduction of azido ketones in good conversions and selectivities.

Conclusion

In conclusion, marine-derived fungi demonstrated to be excellent biocatalysts for the reduction of α-keto azides, affording the corresponding azido alcohols 9–12 in good yields and enantiomeric excesses. The reductases of the marine-derived fungi studied showed Prelog and anti-Prelog selectivity in the bioreduction α-keto azides. The fungi A. sclerotiorum CBMAI 849, C. cladosporioides CBMAI 857, P. raistrickii CBMAI 931and, and P. citrinum CBMAI 1186 catalyzed the reduction of the azido ketone 6 to the corresponding (R)-2-azido-1-(4-methoxyphenyl)ethanol (10) with good conversions (68–100 %) and >99 % enantiomeric excess in 9 days following the Prelog rule.