By using the enrichment technique 232 microorganisms from soils or sediments samples were isolated (Table 1). This technique consisted of applying (RS)-1-phenylethanol (1 mM) as a carbon source presented in the culture medium used in the isolation of microorganisms [16]. The use of this chiral alcohol results in the isolation of microorganisms containing new oxidoreductases that can be applied in enantioselective alcohol oxidation. The enzymatic system able to perform the alcohol oxidation can be consisting of a dehydrogenase or an oxidase. However, after the enzyme purification, we will be able to elucidate the enzyme responsible for the alcohol oxidation reaction.

The results described in Table 1 demonstrated that the highest number of microorganisms was obtained from soil samples and culture medium at neutral pH. The culture medium MCFE7 chemical composition was aimed to induce the isolation of fungi. The minimal medium culture MCBE7 was chosen to induce the isolation of yeasts and bacteria. Despite the pH difference between the culture media, it was possible to observe bacteria and fungi growth in both media.

Enantioselective oxidations of alcohols, para-substituted (RS)-1-phenylethanols, were used to evaluate the enzyme activity (oxidoreductases, OXRED) of 232 microorganisms psychrophile/psychrotroph previously isolated (Scheme 1).

From this study, 15 microorganisms presented oxidoreductases that catalyzed the oxidation of the (S)-enantiomer of racemic alcohols to their corresponding ketones. Therefore, the (R)-alcohols remained intact with high enantiomeric excesses (up to >99%). It was considered as good enzyme activity (OXRED), conversions of alcohol to ketone > 10% and enantiomeric excess >60%. Some of the results are shown in Table 2.

As shown in Table 2, the enzyme activities of the microorganisms were very high when the R substituent in alcohols (1–6) was a methoxy group. In these cases, considering a kinetic resolution process, the alcohols conversions to their corresponding ketones were excellent (Entries 3, 9, 21 and 33), and in some cases the enantiomeric excesses were higher than 99% (Entries 3, 9 and 21). It is noteworthy that, in a reaction of kinetic resolution the maximum conversion is 50%.

In reactions mediated by bacterium HK1D10 cells, in which the R substituent was halogen, low conversions (∼1%) and no selectivity were observed (Entries 34 and 35). In general, all microorganisms produce oxidoreductases with preferential oxidation for the (S)-enantiomer from the (RS)-alcohols (1–6)

Based on these results we decided to perform only the identification of microorganisms that showed excellent enzyme activity. For this purpose Flavobacterium sp. (HK1D9) and Arthrobacter sp. (MPS8D3) were selected. These bacteria were isolated from soil and sediment samples, respectively.

The chromatograms from oxidation reaction of the compound (RS)-3 mediated by cells of Flavobacterium sp. and Arthrobacter sp. (Table 2, Entries 3 and 9) are shown in Figure 1, as an example of the chiral analysis using GC employed for this enzyme evaluation step.

According to the chromatogram obtained from oxidation reaction of the (RS)-3 mediated by Arthrobacter sp. (Figure 1b) and Flavobacterium sp. (Figure 1c), it can be seen that the (S)-enantiomer was oxidized, leaving the unreacted alcohol (R)-3 with high enantiomeric excess.

Flavobacterium sp. and Arthrobacter sp. were selected to be applied in the optimization study due to their high enantioselective performance in the oxidation reaction of the alcohols (RS)-1–6. Based on the good results presented in Table 2 (Entries 2, 8, 14, 20) and by the fast GC analysis using a chiral column, we selected (RS)-1-(4-methylphenyl)ethanol as model substrate for this optimization study.

A survey in the literature indicated that psychrophile/psychrotroph microorganisms tolerate temperatures between 0 and 40 °C [17–22]. The temperature corresponds to one of the major environmental factors that can influence the bacterial growth, when the temperature is increased, the chemical and enzymatic reactions in cells tend to become faster, thus accelerating the growth rate [23].

In order to determine the ideal reaction conditions for achieving an excellent kinetic resolution process mediated by Flavobacterium sp. (HK1D9) and Arthrobacter sp. (MPS8D3), the bacterial cells were resuspended in phosphate buffer solution (Na₂HPO₄/KH₂PO₄, pH 7.0, 0.1 mol/L).

The selected temperatures were 5, 10, 15, 20, 25 and 30 °C (each one is tolerated by psychrotroph microorganisms) [18]. These temperatures were used for both bacterial growth and alcohol oxidation reaction. The results are shown in Table 3.

When the bacterial growing and reactions were performed at 15 °C and 25 °C (Flavobacterium sp.) excellent oxidoreductase activities were observed. The best conversion of the (RS)-alcohol 2 to ketone (45%) was obtained after 24 h reaction at 25 °C (Table 3, Entry 13,). Along the reaction, there was a few decrease in ketone concentration and an increase in (R)-1-(4-methylphenyl)ethanol concentration and, consequently its enantiomeric excess (Table 3, Entries 13–15). In an oxidative kinetic resolution process, the ratio of the ketone and a single alcohol enantiomer concentration should be 1:1. The above mentioned behavior indicates the presence of ketoreductase that catalyzes enantioselectively the reduction of the produced 4-methylacetophenone to the (R)-1-(4-methylphenyl)ethanol, resulting in a higher (R)-alcohol concentratrion than 50%. This process can be called as deracemisation (Scheme 2).

The Arthrobacter sp. (MPS8D3) showed good enzyme activity at 15, 20 and 25 °C (24 h), since under these temperatures, the highest enantiomeric excess and concentration of the (R)-1-(4-methylphenyl)ethanol were obtained (Table 3, Entries 25, 28 and 31). On the other hand, it was observed that Flavobacterium sp. and Arthrobacter sp. exhibited very low enzyme activity at 10 °C, since the values of ketone conversions and alcohol e.e. were very low (Table 3, Entries 4–6 and 22–24).

It is noteworthy to mention that there is few deracemisation process described in the literature performed by a single microorganism [3,24–28] or without the aid of chemical agents [29,30]. This phenomenon is extremely interesting to be pursued in bioprospecting process.

We decided to study the oxidation of (RS)-1-(4-methylphenyl)ethanol at different temperatures (5, 10, 15, 20, 25, 30 °C) after growthing the Flavobacterium sp. and Arthrobacter sp. at a specific temperature, 15 or 25 °C. The bacterial growth temperatures were selected due to the good results of conversions and enantioselectivities of the alcohol oxidation (Table 3, Entries 7–9, 13–17 and 31–33). The main goal of this study was to evaluate the enantioselectivity of oxidoreductases produced by bacterial growth at 15 and 25 °C in the oxidation reaction carried out at different temperatures (Tables 4 and 5).

Analyzing the data in Table 4, in which Flavobacterium sp. grew at 25 °C, we found low enzyme activity for the oxidation of (R,S)-1-(4-methylphenyl) ethanol at 30 °C leading to low ketone conversion and enantioselectivity (Entries 16–18). However, this bacterium showed good enzyme activity in other temperatures (Entries 1–15). In general there was an increase in conversion and enantioselectivity with increasing reaction time. However, the best oxidation performance was observed at 10 °C and 72 h of reaction (Table 4, Entry 6). In this case, a perfect kinetic resolution process was achievied, ketone conversion 50% and alcohol e.e. > 99%.

The reactions carried out under several temperatures and mediated by cells of Arthrobacter sp. resulted in moderate to excellent alcohol e.e. and ketone conversion (Table 4, Entries 19–36). For example, oxidation reactions carried out at 15 to 30 °C, there was a decrease in ketone concentration and an increase in alcohol e.e. over the reaction time (Table 4, Entries 25–36). The Flavobacterium sp. has oxidoreductase with high activity at 10 °C and Arthrobacter sp. at 15 and 20 °C.

When the microbial growing temperature was 15 °C (Flavobacterium sp.) a low enzyme activity was observed at 5 and 30 °C. These reactions afforded low values of ketone concentration and enantioselectivies (Table 5, Entries 1–3 and 16–18). However, the best ketone concentration and enantioselectivity were observed for reactions carried out at 15 and 20 °C (Table 5, Entries 9 and 12).

On the other hand, Arthrobacter sp. showed good enzyme activity at 5 and 10 °C after 48 and 72 hours, since it was obtained high enantiomeric excess for the alcohol (R)-2, up to >99% (Table 5, Entries 20, 21, 23 and 24). We also can see that Arthrobacter sp. showed good enzymatic activity at reaction temperatures of 15–30 °C, in which ketone conversions and e.e. for the alcohol (R)-2 were moderate to excellent (Table 5, Entries 25–36). At these temperatures, the value of (R)-alcohol concentration is higher than 50% with increasing the reaction time. As observed for Flavobacterium sp. (Table 3 and Scheme 2), this process indicates the presence of ketoreductase that catalyzes enantioselectively the reduction of the produced 4-methylacetophenone to the (R)-1-(4-methylphenyl)ethanol, resulting a high (R)-alcohol concentration.

In general, it was observed that Arthrobacter sp. and Flavobacterium sp. had excellent enzyme activity with growth temperature at 25 °C.

Therefore, we decided to determine the growth curves of Flavobacterium sp. and Arthrobacter sp. at 15, 20 and 25 °C (Figure 2). By spectrophotometric analysis of optical density (OD) of bacterial culture it was evaluated the turbidity and estimated the relative concentration cell.

According to the Figure 2, it was noted that Flavobacterium sp. and Arthrobacter sp. grew well at 25 °C and exhibited excellent enzyme activity at this temperature (Table 4, Entries 13–15 and 31–33). Then, it was concluded that these bacteria are psychrotroph since it excludes the possibility of psychrophile, in which the maximum temperature for growth is 20 °C. This conclusion was based on literature data about the identification of psychrotroph microorganism [17,31,32].

Acetophenone, 1-(4-methylphenyl)ethanone, 1-(4-methoxyphenyl)ethanone, 1-(4-bromophenyl) ethanone, 1-(4-chlorophenyl)ethanone and 1-(4-nitrophenyl)ethanone were purchased from Aldrich and used without further purification. Thin-layer chromatography (TLC) was performed using pre-coated plates (Aluminum foil, silica gel 60 F₂₅₄ Merck, 0.25 mm). Silica gel (Acros Organics, 0.035–0.070 mm, 240–400 mesh) was used for flash chromatography. Low Resolution Mass Spectra (LRMS) were recorded on a Shimadzu GCMS P5050A (70 eV) spectrometer and GC cappillary column used was a DB-5-MS (30 m × 0.25 mm × 0.25 μm, J&W Scientific), carrier gas-He, 100 kPa): injector 270 °C; detector 270 °C. Temperature program 100–250 °C, rate 10 °C/min. GC analyses were performed in a Shimadzu GC-17A instrument with a FID detector using hydrogen as carrier gas (100 kPa), autosampler AOC20i. It was used a GC chiral cappillary column, Chirasil-Dex CB β-cyclodextrin (25 m × 0.25 mm, Varian), for determination of the conversion and enantiomeric excesses. Sterile materials were used to perform the experiments and the microorganisms were handled in a laminar flow cabinet Fisher Hamilton Class II Biological Safety Cabinet.

To a solution of acetophenones (7–12) (10 mmol) in methanol (20 mL) was added NaBH₄ (11.1 mmol, 420 mg), and the mixture was stirred at room temperature for 4 h. The methanol was removed by vacuum evaporation. Distilled water (20 mL) was added into the reaction mixture and the pH adjusted to 6 with aqueous solution of HCl (1 mmol/L), then the aqueous layer was extracted with methylene dichloride (3 × 30 mL). The organic phases were combined and dried over MgSO₄. The solvent was removed in vacuum and the residue was purified by silica gel column chromatography eluting with a mixture of hexane and ethyl acetate (4:1) to afford compounds (RS)-1–6. The ¹H NMR and ¹³C NMR spectra of these compounds were in agreement with those reported in the literature [34,35].

From each pre-culture obtained previously, a 750 μL aliquot was transferred to microtubes containing 250 μL of glycerol for cryo-preservation (−20 °C). Another 200 μL were transferred to inoculate in a sterile polypropylene microtiter plate containing 2 mL of culture media per well. The plates were kept at 20 °C in an orbital shaker (160 rpm) during 2 or 4 days and then 40 μL of a solution of each alcohol (1–6, 5 mmol/L) in dimethylformamide were added to each well of the plate. The microtiter plates were returned to the same incubation condition as described above for 72 hour. Each well content was extracted by stirring with tert-butyl methyl ether (2.0 mL) followed by centrifugation (6000 rpm, 1 min.). The organic phase was analyzed by GC using a chiral capillary column (see Section 3.5).

GC conditions (carrier gas-H₂, 100 kPa): injector 220 °C; detector 220 °C. Isotherm and t_R (min): retention time of chiral compounds: (RS)-phenyl-ethanol: 107 °C, R-enantiomer 6.22 min, S-enantiomer 6.87 min; (RS)-1-(4-methyl-phenyl)-ethanol: 113 °C, R-enantiomer 6.79 min, S-enantiomer 7.71 min; (RS)-1-(4-methoxy-phenyl)-ethanol: 119 °C, R-enantiomer 6.22 min, S-enantiomer 6.87 min; (RS)-1-(4-chloro-phenyl)-ethanol: 130 °C, R-enantiomer 7.33 min, S-enantiomer 8.19 min; (RS)-1-(4-bromo-phenyl)-ethanol: 135 °C, R-enantiomer 9.54 min, S-enantiomer 10.52 min; (RS)-1-(4-nitro-phenyl)-ethanol: 150 °C, R-enantiomer 19.92 min, S-enantiomer 20.30 min. The enantiomeric excess (ee) was defined as the ratio of {([R] – [S])/([R] + [S])} × 100%, where [R] and [S] are the concentrations of (R) and (S) enantiomers, respectively.

The microorganism strains with the best performance in enzyme assays were stored in a freezer at −20 °C. They were preserved as suspensions in glycerol solution (20%) using two different culture media described in Section 3.3.2.

The bacteria with the best performance in enzyme assays were identified using 16S rDNA sequencing analysis and comparison with GenBank top hits after a BLAST search [36].

Cultures of Flavobacterium sp. and Arthrobacter sp. were separately inoculated under sterile conditions into 20 mL of appropriate culture media in Erlenmeyer flasks (125 mL) and incubated in an orbital shaker during 2 days (160 rpm, 20 °C). After that, the cultures were transferred separately to two Erlenmeyer flasks (2000 mL) containing 600 mL of sterilized culture media and incubated again. Then the cells were harvested by centrifugation at 20 °C on Sorvall RC-5B (Plus rotor) for 15 min at 5000 rpm. The harvested cells were resuspended in 60 mL of sterile phosphate buffer solution (Na₂HPO₄/KH₂PO₄, pH 7.0, 0.1 mol/L), divided equally in 6 Erlenmeyer flasks (125 mL) and an aliquot (200 μL) of substrate solution (RS)-1-(4-methylphenyl)ethanol (1 mmol/L) was added to each flask. The mixture was kept at different temperatures (5, 10, 15, 20, 25 and 30 °C) in an orbital shaker (160 rpm) during 1, 2 or 3 days. The products of the total reaction volume (10 mL) were extracted by stirring with tert-butyl methyl ether (2.0 mL) followed by centrifugation (6000 rpm, 1 min.). The organic phase was analyzed by GC using a chiral capillary column (see Section 3.5).

Growth curve experiments were performed with Erlenmeyer flasks (2000 mL) containing 1000 mL of sterilized culture media and 100 mL of inoculated cultures. The mixture was incubated in an orbital shaker (160 rpm, 15, 20 and 25 °C). Samples of culture (1 mL) were periodically removed at 120 min intervals and the growth was estimated spectrophotometrically (and expressed as optical density [O.D.]) at 600 nm with a Biospectro model sp-220 spectrophotometer using cells with a 1 cm light path.

The replicate values were averaged and the O.D. was plotted against time on linear coordinates.