Chemo-Enzymatic Synthesis of a Multi-Useful Chiral Building Block Molecule for the Synthesis of Medicinal Compounds

Optical resolution of 2-methyl-2-nitrobut-3-en-1-ol has been accomplished using a “low-temperature lipase-catalyzed transesterification” carried out at −40 °C.


Introduction
Chemo-enzymatic reaction protocols are now well recognized as a very useful means to prepare optically active compounds [1][2][3]. 2-Methyl-2-nitrobut-3-en-1-ol (±1) was prepared by a simple method using the nitroaldol reaction for nitroalkenes [4], and it has been expected to become a useful building block for the synthesis of various types of non-natural amino acids (A ~ E) or amino alcohols (F), as illustrated in Scheme 1. However, no one has yet succeeded in preparing optically pure nitro alcohol 1, so preparation of optically active 1 using a practical protocol has thus been strongly desired.
Lipases have wide applicability for various types of substrates [1][2][3]5], however, it is generally not easy to use the lipase-mediated reaction for the kinetic resolution of a primary alcohol like alcohol 1, because the chiral carbon is remote from the reaction point in such a type of compound [2]. Since preparation of chiral compounds that have a quaternary stereocenter is an important challenge for modern organic synthesis, several examples have been demonstrated using enzymatic reactions [6][7][8][9][10][11][12][13].

OPEN ACCESS
Herein, we report the establishment of a protocol that affords both enantiomers of 2-methyl-2-nitrobut-3-en-1-ol (1) using a lipase-catalyzed reaction; the "low-temperature lipase-catalyzed reaction" protocol [14][15][16] was shown to be the key technology to accomplish the desired reaction with sufficient enantioselectivity. Scheme 1. Multi-useful chiral building block for the synthesis of non natural amino acids and amino alcohols.

Results and Discussion
We initially attempted to resolve (±)-1 via lipase-catalyzed transesterification using vinyl acetate as acyl donor in diisopropyl ether (i-Pr 2 O) under standard reaction temperature at 35 °C (Scheme 2); however, after evaluating commercial lipases, we soon recognized that it would be a very tough task for us to find a suitable enzyme, as we were unsuccessful in finding an appropriate enzyme that could convert 1 to the corresponding acetate with acceptable enantioselectivity. Among the 17 types of commercial enzymes tested, only five lipases PS, SL-25, PL, Novozyme 435 and QLM gave the corresponding acetate, but all with insufficient enantioselectivity. Although lipases QLM and SL-25 gave somewhat better results, the E values [17] of the reactions were 7.2 (QLM) and 6.0 (SL-25), respectively. Since lipase QLM gave the best E value, we next attempted to optimize the reaction condition using lipase QLM as a catalyst.
It has long been believed that enantioselectivity of lipase-catalyzed reaction could be explained by the traditional three point attachment rule [2,18]. According to this rule, optimization of a substrate structure or protein engineering of lipases might be essential to control the enantioselectivity of the enzymatic reactions [2].
On the other hand, Ema et al. proposed that the enantioselectivity of the lipase-catalyzed reaction might be determined mainly by kinetic preference due to the conformational requirements and repulsive interaction on the transition state [19,20]. The model allows enhancement of the enantioselectivity of lipase-catalyzed reaction simply by changing the reaction temperature. Sakai and co-workers, in fact, demonstrated that efficient kinetic resolution of primary alcohols was realized using the "low-temperature transesterification method" [14][15][16]. Sakai showed that lipase-catalyzed transesterification of (2,2-dimethyl-1,3-dioxolan-4-yl)methanol in i-Pr 2 O proceeded even at −40 °C; the E value of the reaction at 30 °C was just 9, while it reached 55 when the reaction was conducted at −40 °C [16]. Therefore we decided to apply "the low-temperature method" to our lipase-catalyzed reaction (Scheme 1, and the results are summarized in Table 1). a Determined by HPLC analysis using CHIRALCEL OB-H, hexane/i-PrOH = 19/1, 0.5 mL/min; b Determined by HPLC analysis using CHIRALCEL AD-H, hexane/i-PrOH = 19/1, 0.5 mL/min. c E value was calculated by % ee of (R)-2 (ee p ) and % ee of (S)-1 (ee s ). E= ln[(1 − c (1 + ee p )) / ln[(1 -c (1 − ee p )); here, c means conv. which was calculated by the following formula according to reference [17]: c = ee s / (ee p + ee s ).
Lipase QLM-catalyzed transesterification of (±)-1 proceeded very rapidly, and we obtained acetate (R)-2 in 35% yield with 54% ee, and unreacted alcohol (S)-1 was recovered from the reaction mixture in 28% yield with 75% ee after just 10 min of reaction (entry 1). Enantiomeric excess of the product and unreacted substrate were determined by HPLC analysis using a chiral column. A slightly enhanced enantioselectivity was recorded when the reaction was carried out at −20 °C (entry 4), and it finally reached E = 15 at −40 °C (entry 5). Since the reaction rate was very fast, we obtained (R)-2 with 83% ee when the reaction was stopped at 25 min (entry 5), while 94% ee of (S)-1 was obtained after 26 h of reactions (entry 6); no improved enantioselectivity was recorded when the reaction was conducted at −60 °C. Based on the results, we have developed a protocol providing (R)-2 and (S)-1 with high enantiomeric purities as illustrated in Scheme 3.

Scheme 3.
Protocol of preparation of chiral 1 using "low-temperature lipase-catalyzed transesterification". Racemic (±)-1 was subjected to lipase-QLM-catalyzed transesterification at −40 °C, after being stirred for 3 h, the reaction was stopped and the acetate (R)-2 (72% ee) and alcohol (S)-1 (31% ee) were separated. Enantiomeric purities of (R)-2 and (S)-1 were not sufficient at this stage, so (R)-2 was converted to (R)-1 by acid hydrolysis in 87% yield without any loss of the optical purity. The resulting 72% ee of alcohol (R)-1 was subjected to a second transesterification. After 3 h reaction, optically pure (R)-2 (>99% ee) was obtained in 10% yield (the upper route in Scheme 3). (S)-1 (31% ee) was subjected to a second reaction for 24 h and 94% ee of (S)-1 was obtained in 9% yield (the bottom route in Scheme 3). Although the chemical yield of each reaction was insufficient, this protocol made it possible to provide (R)-2 and (S)-1 with high optical purity. After repeating the process, we succeeded in obtaining multiple grams of (R)-2 and (S)-1 with excellent optical purity (Scheme 3).
Development of efficient means for preparing chiral compounds that have a quaternary chiral center has been a challenging area in the field of synthetic organic chemistry. In particular, it is difficult to achieve this aim by enzymatic reaction because hydrolytic enzymes are usually unable to accept sterically hindered substrates bearing fully substituted quaternary carbons [2]. It should be emphasized that the present "low-temperature lipase-catalyzed reaction" provides a possible solution to this problem.

General Procedures
Reagents and solvents were purchased from common commercial sources and used as received or purified by distillation over appropriate drying agents. Reactions requiring anhydrous conditions were carried out under dry argon, freshly distilled solvents, and magnetic stirring. We tested the following commercial lipases:

Conversion of (R)-2 to (R)-1 by Acid Hydrolysis
To a THF (3.0 mL) solution of (R)-2 (250 mg, 1.44 mmol) was added 1 M aqueous HCl solution (3.0 mL) at rt and the mixture was stirred for 3 h at rt, and then concd. HCl (3.0 mL) was added and the mixture was further stirred for 72 h at rt. The reaction mixture was neutralized carefully with saturated sodium bicarbonate aqueous solution and extracted with dichloromethane. The combined organic layer was dried under MgSO 4 and the solvent removed by evaporation. (R)-1 (164 mg, 1.25 mmol) was obtained in 87% yield after silica gel flash column chromatography (hexane: ethyl acetate = 10:1 to 5:1).

Conclusions
In summary, we established a convenient protocol to prepare both enantiomers of 2-methyl-2nitrobut-3-en-1-ol (1) with over 94% ee using lipase-catalyzed transesterification under low temperature reaction conditions. It was possible to apply the reaction protocol to the multi gram scale preparation and we succeeded in preparing the desired compounds easily. Synthetic application of a medicinal compound using optically active 1 is now ongoing in our laboratory.