Appl Microbiol Biotechnol (1992) 38:209-213

App//ed Microb/obgy

Bioteehnology

© Springer-Verlag 1992

Kinetic resolution of organosilicon compounds by stereoselective dehydrogenation with horse liver alcohol dehydrogenase Toshiaki Fukui, Min-Hua Zong*, Takuo Kawamoto, Atsuo Tanaka Laboratory of Industrial Biochemistry, Department of Industrial Chemistry, Faculty of Engineering, Kyoto University, Yoshida, Sakyo-ku, Kyoto 606-01, Japan Received: 22 April 1992/Accepted: 6 July 1992

Abstract. Stereoselective dehydrogenation of three isomers of trimethylsilylpropanol was carried out w i t h horse liver alcohol dehydrogenase ( H L A D H , EC 1.1.1.1.) and optically active organosilicon compounds were obtained in a water-organic solvent two-layer system with coenzyme regeneration. Furthermore, we examined the effects of the silicon atom on stereoselectivity of H L A D H compared to the corresponding carbon compounds. Substitution of the silicon atom for the carbon atom was found to improve the stereoselectivity of H L A D H . For example, the optical purity of the remaining 1-trimethylsilyl-2-propanol was higher than 99% enantiomeric excess (ee) at 50% conversion, whereas that of the carbon analogue was 84% ee. This phenomenon was probably ascribable to the bulkiness of the organosilicon compounds derived from their longer Si-C bond. Kinetic analysis in an aqueous monolayer system demonstrated that the specific properties of the silicon atom greatly affected the reactivity of these substrate compounds.

H L A D H has proved to be a most versatile enzyme, being promising for the conversion of organic compounds, because it catalyses alcohol/carbonyl oxidoreduction of a broad spectrum of substrates with high degrees of stereoselectivity (Whitesides and Wong 1985). However, no report had been published on the bioconversion of organosilicon compounds with H L A D H until our previous paper, which demonstrated that H L A D H was capable of catalysing the dehydrogenation of organosilicon compounds, (CH3)3Si(CH2)nOH (n = 1-3), and that the silicon atom in organosilicon compounds had a great effect on the activity of H L A D H . As a further step towards the study on bioconversion of organosilicon compounds, here we report the stereoselective dehydrogenation of the three isomers of chiral trimethylsilylpropanol (Fig. 1, 1-3) with H L A D H in a water-organic solvent two-layer system with coenzyme regeneration, and the effect of the silicon atom on the activity and stereoselectivity of H L A D H compared to their carbon counterparts (Fig. 1, 4-6). The results obtained have been rationalized on the basis of the characteristics of the silicon atom and the structure of HLADH.

Introduction An active area of current research in biotechnology is the use of unconventional compounds as substrates for enzymes, to increase the potential and expand the application of enzymes. Organosilicons are very interesting as unconventional target compounds, because of their importance in synthetic organic chemistry (Colvin 1988; Patai and Rappoport 1989) and possible use as new useful materials and drugs (Tacke and Zilch 1986). We have already reported the bioconversion of organosilicon compounds by hydrolases (Kawamoto et al. 1991) and horse liver alcohol dehydrogenase ( H L A D H , EC 1.1.1.1.) (Zong et al. 1991) from this point of view. * Present address." Food Engineering Department, South China University of Technology, Guangzhou, Peoples Republic of China Correspondence to: A. Tanaka

Materials and methods 1H-Nuclear magnetic resonance (NMR) spectra were measured with a Varian Gemini-200 NMR spectrometer and infrared (IR) spectra were obtained with a Japan Spectroscopic IR-810 spectrometer. Gas-liquid chromatography (GLC) analyses were carried out using a Shimazu GC-12A equipped with a flame-ion-

Analyses.

.,2>. I

El = Si

2

El : Si

3

El = Si

4

El :

5

El : C

6

El : C

C

Fig. 1. Structures of three isomers of trimethylsilylpropanols and their carbon analogues used as substrates for horse liver alcohol dehydrogenase (HLADH)

210 ization detector and high performance liquid chromatography (HPLC) analyses were done using a Hitachi L-6000 instrument equipped with an L-6200 UV detector. A Shimazu MPS-200 spectrophotometer was used for kinetic analysis. Specific rotations were determined with a Japan Spectroscopic DIP-140 polarimeter,

Materials. HLADH and 2-oxoglutarate (disodium salt) were purchased from Sigma (St. Louis, Mo., USA) and L-glutamate dehydrogenase (G1DH) from Biozyme (South Wales, UK). ]%Nicotinamide adenine dinucleotide (NAD +) was obtained from Oriental Yeast (Tokyo, Japan). 1-Trimethylsilyl-2-propanol (1) (Davis and Jacocks 1981), 1-trimethylsilyl-l-propanol (2), and 2-trimethylsilyl-l-propanol (3) (Soderquist and Brown 1980) were donated by Nitro Denko (Osaka, Japan). 4,4-Dimethyl-2-pentanol (4) and 2,2-dimethyl-3-pentanol (~) were purchased from Aldrich (Milwaukee, Wis., USA). 2,3,3-Trimethyl-l-butanol (6) was prepared as described below. Other chemicals were also obtained from commercial sources. 2,3,3-Trimethyl-l-butanol (6). 2,3,3-Trimethyl-l-butene (3.5g, 36 mmol) (Tokyo Kasei, Tokyo, Japan) was dropped into a 0.3-1 flask containing 80ml of 0.5M 9-borabicyclo[3,3,1]nonane (Brown et al. 1974) in tetrahydrofuran (Aldrich) under an N2 atmosphere with stirring, followed by stirring at room temperature for 1 h further. Then water (5 ml), 3 M NaOH (17 ml), and 30o70 H202 (17 ml) were added successively at a controlled temperature (less than 50° C) and the aqueous phase was saturated with K2CO3 after refluxing for 1 h. The organic phase, together with the ether extract of the aqueous phase, was dried over Na2SO4, concentrated, and distilled under atmospheric pressure. The 130-140°C b.p. fraction was collected to give 3.0 g (26 mmol, 72% yield) of 6. XH-NMR (CDCI3): J0.87 (s, 9H, C(CH3)3), 0.94 (d, J=6.8 Hz, 3H, CH3), 1.34 (s, 1H, OH), 1.36-1.42 (m, 1H, CH), 3.32 (rid, J=8.7, 10.4 Hz, 1H, CH2) , 3.81 (dd, J=3.9, 10.4 Hz, 1H, CH~). IR (neat): 3350, 2950, 1470, 1380, 1030cm -~.

with HPLC reached 90°70 ee, the remaining (+)-alcohols were purified with silica gel column chromatography. Diastereomeric esters of (+)-1, (+)-4, and their racemic compounds were synthesized with (R)-methoxytrifluoromethylphenyl acetic acid (MTPA) and analysed by 1H-NMR. The configuration of each enantiomer was determined from the observed chemical shifts. The spectra data were as follows: (R)-MTPA ester of (S)-(+)-I. 1H-NMR (CDC13): fi -0.019 (s, 9H; Si(CH3)3), 0.91 (dd, J=9.6, 14.1Hz, 2H; CH2), 1.36 (d, J=6.3 Hz, 3H; CH3), 3.56 (q, J = l . 2 H z , 3H; OCH3), 5.2-5.3 (m, 1 H; CH), 7.4-7.6 (m, 5 H; C6H5). (R)-MTPA ester of (R)-(-)-I. ~H-NMR (CDC13): fi 0.041 (s, 9H; Si(CH3)3), 1.00 (dd, J=10.1, 13.9Hz, 2H; CH2), 1.28 (d, J = 6 . 2 H z , 3H; CH3), 3.54 (q, J = l . 4 H z , 3H; OCH3), 5.2-5.3 (m, 1H; CH), 7.4-7.6 (m, 5H; C6H~). (R)-MTPA ester of (S)-(+)-4. 1H-NMR (CDC13): ~ 0.078 (s, 9H; C(CH3)3), 1.33 (d, J=6.1 Hz, 3H; CH3), 1.39 (d, J=3.4 Hz, 2H; CHz), 3.57 (q, J = 1.5 Hz, 3H; OCH3), 5.1-5.3 (m, 1H; CH), 7.4-7.6 (m, 5H; C6H~). (R)-MTPA ester of (R)-(-)-4. 1H-NMR (CDC13): fi 0.095 (s, 9H; C(CH3)3), 1.26 (d, J=6.1 Hz, 3H; CH3), 1.43 (d, J=4.0 Hz, 2H; CH2), 3.50 (q, J = 1.3 Hz, 3H; OCH3), 5.1-5.3 (m, 1H; CH), 7.4-7.6 (m, 5 H; C6Hs).

Kinetic analysis in an aqueous monolayer system. Assay mixtures (3.3 ml) containing various concentrations of substrate, 1.2 m s NAD ÷, and 1% tetrahydrofuran in 50 mM TRIS-HC1 buffer (pH 8.8) was preincubated at 30° C for 3 min and then HLADH solution (0.4 ml, 3.4 IU.ml -I) was added. The reaction rate was calculated from the increase in absorbance of NADH at 340 nm (Ae34o= 6.22 x 10 -3 M - 1ocm - 1)o Cs . V - 1 versus Cs plots were applied to obtain the values of Km and Vm~, where Cs is the initial substrate concentration and V is the initial reaction rate.

Results Stereoselective dehydrogenation by H L A D H in a two-layer system. Substrate (0.1 mmol) was dissolved in water-immiscible organic solvent (10ml) and mixed with 50mM TRIS-HC1 buffer (10 ml, pH 8.8) containing 10 IU HLADH, 20 mM 2-oxoglutarate (disodium salt), 20 mM ammonium acetate, and 20 IU G1DH. After stirring at 30°C for 30 rain, 0.2 ~tmol NAD ÷ was added to start the reaction and the mixture was stirred at 30 ° C. It was confirmed that only a small amount of the substrate was present in the water phase. The conversion ratio was expressed as the decrease in substrate concentration in the organic phase, measured by GLC using a glass column (1 m) packed with PEG-HT supported on Uniport R (GL Sciences, Tokyo, Japan) (N~ carrier gas, 60ml.min-X; detector and injector temperature, 250°C). n-Pentadecane was used as the internal standard. The optical purity of the remaining substrate was determined with HPLC using a Sumichiral OA-4600 column (4mm i.d. x250mm; Sumika Chemical Analysis Service, Osaka, Japan) after derivatization with 3,5-dinitrophenyl isocyanate (Sumika Chemical Analysis Service) ((3i et al. 1990). The mobile phase was n-hexane/2-propanol (100:1 v/v) for analysis of 1 and 4. For 3 and 6, two columns of Sumichiral OA-4600 in series were used and the mobile phase was n-hexane/ 2-propanol (98:2 v/v). The flow rate was 1.0 ml.min-~ and the eluent was monitored at 254 nm in both cases. The enantiomeric excess (% ee) was calculated from peak areas of both enantiomers in the remaining alcohols, Determination of absolute configuration. Absolute configurations of 1 and 4 were determined by a correlation method with IHNMR (Dale and Mosher 1973). Both (+)-1 and (+)-4 were prepared by stereoselective esterification of racemic 1 and 4 with 5phenylpentanoic acid with lipoprotein lipase Type A (Toyobo, Osaka, Japan). To start the reaction Celite-adsorbed lipoprotein lipase (100 mg enzyme, 100 Ixl deionized water, 250 mg Celite) was added into 2,2,4-trimethylpentane containing 100 mM substrates. When the optical purities of the remaining alcohols monitored

Stereoselective dehydrogenation by H L A D H two-layer system

in a

Stereoselective d e h y d r o g e n a t i o n o f the three isomers o f t r i m e t h y l s i l y l p r o p a n o l (1-3) with H L A D H was tried b y e x a m i n i n g the effect o f the silicon a t o m on the stereoselectivity o f H L A D H c o m p a r e d to their c a r b o n a n a logues (4-6). P r o d u c t i n h i b i t i o n is generally a big p r o b lem in d e h y d r o g e n a t i o n r e a c t i o n s c a t a l y s e d b y a l c o h o l d e h y d r o g e n a s e (Whitesides a n d W o n g 1985). T o solve this p r o b l e m , a w a t e r - o r g a n i c solvent t w o - l a y e r system was a p p l i e d with c o e n z y m e r e g e n e r a t i o n ( o x i d a t i o n o f N A D H to N A D ÷ c o u p l e d with G 1 D H - c a t a l y s e d reductive a m i n a t i o n o f 2 - o x o g l u t a r a t e to I~-glutamate (Lee a n d W h i t e s i d e s 1985)) (Fig. 2). n - H e x a n e was f o u n d to be the best a m o n g the o r g a n i c solvents tested (chlorof o r m , 1 , 2 - d i c h l o r o e t h a n e , ethyl acetate, a n d n-hexane). T h e t i m e - c o u r s e s of~the H L A D H - c a t a l y s e d d e h y d r o g e n a t i o n o f I a n d 3, a n d their c a r b o n c o u n t e r p a r t s (4, 6) were f o l l o w e d in the w a t e r - n - h e x a n e t w o - l a y e r system with c o e n z y m e r e g e n e r a t i o n (Fig. 3). W i t h 1, the react i o n p r o c e e d e d quickly a n d s t o p p e d at 50% conversion, whereas the c a r b o n a n a l o g u e 4 r e a c t e d slowly a n d the c o n v e r s i o n r e a c h e d over 5 0 % . H L A D H d e h y d r o g e n a t e d 3 a n d 6 at high rates a n d the r e a c t i o n c o n t i n u e d a b o v e 50% conversion. I n these cases, the t u r n o v e r n u m b e r o f N A D ÷ was c a l c u l a t e d to be m o r e t h a n 400. T h e dehyd r o g e n a t i o n o f 2 a n d $ was negligible u n d e r the r e a c t i o n conditions employed.

211

)

(±)-Trimethylsilylpropanol

L-Glutamate + H20

:

lOrganic phase[ (-)-Trimethylsilylpropanol +

r~-~_e~phrase}

GIDH

HLADH

+

2_Oxoglutarate

NH4

+

Z..

Aldehyde or ketone

Fig. 2. A two-layer system for stereoselective dehydrogenation of organosilicon compounds with coenzyme regeneration

GIDH: Glutamatedehydrogenase

i Ioo

~

i

//

i

I

I

I

-

80

0

~

I

I

I

40

60

80

I00

8°i

-

E

._ 60

I

IO0

-

-

4o 20 O(

I 8

I //~l 16 _0 Reaction lime (h)

i 48

~ 56

~

Fig. 3. T~m~-cou~scs of ~L~D~-c~t~y~d d~hydEo~cn~fion of th~ o~nos~ficon co~pounds ~nd theft cEbo~ count¢~E~s ~ ~ [wo-]~ycE systemw~th cocnzym~ E~n~[~t~on: ~, ~; ~, 3; i , 4; ~, 6

2.0

Conversion (%)

Fig. 4. Relationship between optical purity of the remaining alcohols and conversion ratio in the HLADH-catalysed dehydrogenation reaction in a two-layer system with coenzyme regeneration: O, 1; A, 3; ~, 4; &, 6

Table 1. Optically active trimethylsilylpropanols and their carbon counterparts obtained by the horse liver alcohol dehydrogenase (HLADH)-catalysed dehydrogenation reaction Alcohol

Reaction time (h)

Conversion (%)

1-Trimethylsilyl-2-propanol (1) 4,4-Dimethyl-2-pentanol (4) 2-Trimethylsilyl- 1-propanol (3) 2,3,3-TrimethyM-butanol (6)

9 41 4 9

51 50 68 71

% ee >99 85 70 59

Optical activity"

Configuration b

(-) (-) (-) (-)

R R

All reactions were carried out in a water-n-hexane two-layer system with coenzyme regeneration, as described in the text: % ee, enantiomeric excess

" Optical activity of the remaining alcohols was determined by HPLC analysis b Configuration of the remaining alcohols was determined by the correlation method with 1H-nuclear magnetic resonance

Stereoselectivity of H L A D H toward organosilicon compounds was examined by measuring the % ee of the remaining alcohols. As illustrated in Fig. 4, stereoselective dehydrogenation of the organosilicon compounds was successfully carried out, especially in the case of 1. The optical purity of the remaining 1 reached higher than 99°7o ee at the conversion ratio of 50%, whereas that of 4 was about 85% ee, although the optical purity was 99°7o ee at 54% conversion. The optical purity of the remaining 3 (34°7o ee) was similar to that o'f 6 (33O/o ee) at 50°7o conversion. As the conversion ratio in-

creased, however, the difference between the stereoselectivity of H L A D H for these two substrates became apparent. At about 70070 conversion, for example, 3 showed 70% ee, while the optical purity of 6 was only 59% ee. The results shown in Table 1 clearly indicate that silicon substitution for the carbon a t o m improved the stereoselectivity of H L A D H . The absolute configurations of remaining ( - ) - 1 and ( - ) - 4 after dehydrogenation by H L A D H were both R (see Materials and methods) indicating that H L A D H was active on the S-enantiomers of 1 and 4. A m o n g the substrates investigated, 1

212 was found to be the most excellent substrate for H L A D H on stereoselective dehydrogenation under the conditions employed because it was converted at the highest rate with the highest stereoselectivity. These phenomena indicate that substitution of the silicon atom for the carbon atom in the substrate breaks through the conventional problem that the more reactive a substrate is, the lower is the stereoselectivity of the enzyme.

Kinetic analysis in an aqueous monolayer system The reaction rate of HLADH-catalysed dehydrogenation in the water-n-hexane two-layer system cannot be discussed thoroughly without information on the diffusion of substrate and product and on the coenzyme regeneration rate. Therefore, kinetic analysis of the dehydrogenation of the organosilicon compounds and the corresponding carbon compounds was performed spectrophotometrically in an aqueous monolayer system with excess coenzyme. Because the difference in stereoselectivity of H L A D H between the silicon compounds and their carbon counterparts was not drastic at low conversion ratios, it would be reasonable to use the racemic alcohols for analysis of the kinetic parameters. As shown in Table 2, 4 was a poor substrate for H L A D H and 5 showed no reactivity, whereas 6 exhibited fairly high reactivity. These results agree with the fact that H L A D H is essentially a primary alcohol dehydrogenase. For the secondary alcohols containing the silicon atom, 1 was the most reactive substrate among the six compounds examined and 2 showed low reactivity, although its carbon counterpart was not a substrate of H L A D H . These results indicate that replacement of the carbon atom with the silicon atom in the secondary alcohols, 4 and 5, improved substrate reactivity. In the case of the primary alcohol, however, the reactivity of the silicon substrate 3 was similar to that of its carbon analogue 6. It is apparent, from the value of Km, that silicon substitution for the carbon atom resulted in a higher affinity of H L A D H towards the substrates.

Table 2. Kinetic parameters of the HLADH-catalysed dehydrogenation reaction Substrate

Km (raM)

l/max l/max/Km (~tmol" ( 1 0 - 4 " 1 • min - 1) min - 1)

Ethanol 1-Trimethylsilyl-2-propanol (1) 4,4-Dimethyl-2-pentanol (4) 1-Trimethylsilyl-1-propanol (2) 2,2-Dimethyl-3-pentanol (5) 2-Trimethylsilyl-1-propanol (3) 2,3,3-Trimethyl- 1-butanol (6)

2.29 2.92 15.7 7.68 -4.62 7.45

0.488 0.461 0.055 0.137 -0.329 0.462

2.13 1.58 0.035 0.178 -0.712 0.620

Kinetic parameters were obtained as described in the text

Discussion Previously (Zong et al. 1991), we reported that the silicon atom increased the reactivity of a substrate having a hydroxyl group on the/?-carbon atom but decreased that of a substrate having a hydroxyl group on the a-carbon atom in the HLADH-catalysed dehydrogenation reaction. Further, it was revealed that 2-trimethylsilylethanol, which had a hydroxyl group binding to the/?-carbon atom, showed smaller activation energy due to the /?-effect (Colvin 1988) of the silicon atom and a smaller frequency factor due to the bulkiness derived from longer Si-C bond than its carbon analogue, 3,3-dimethyl-1butanol. However, the major influence of the activation energy resulted in higher reactivity of the silicon compound. The fact that 1 has much higher reactivity than its carbon analogue (Table 2) can be well understood, as in the case of 2-trimethylsilylethanol. For 3, the/?-effect of the silicon atom should also enhance the reactivity. However, this enhancement seemed not to be great enough to overcome the decrease in reactivity caused by the bulkiness of the trimethylsilyl group, probably because the primary alcohol 3 is more sterically complicated than the alcohol used in the previous study. Accordingly, the reactivity of 3 is similar to that of the carbon counterpart. Compound 2 was supposed to be less reactive than its carbon counterpart, as described previously (Zong et al. 1991), because its hydroxyl group is bound to the a-carbon. However, 2 was a better substrate for H L A D H than the corresponding carbon compound. This phenomenon might be explained by the local decrease in steric hindrance derived from the longer Si-C bond. This would make it easy for H L A D H to attack the c~-carbon atom in the secondary alcohol. These results revealed that the specific characteristics of the silicon atom greatly affected reactivity, although such an effect was dependent on the substrate structure. The higher affinity of the silicon compounds toward H L A D H , shown as the values of Km (Table 2), would be attributable to the higher hydrophobicity of trimethylsilyl group (Fessenden and Fessenden 1980) than that of tert-butyl group. That is, the trimethylsilyl group is favorable for binding to the hydrophobic active centre of H L A D H (Br~ind6n et al. 1973; Eklund et al. 1982). In the two-layer system, accumulation of the dehydrogenated products of 4 and 6 was observed, while those of 1 and 3 were not accumulated, because the products,/?-carbonylsilanes, were decomposed by water into trimethylsilanol and aliphatic carbonyl compounds (Hauser and Hance 1951). Absence of product inhibition derived from this decomposition may be one of the reasons for a higher reaction rate of the silicon compounds observed in the two-layer system. The reason why 2, which became a substrate for H L A D H in the monolayer system, showed no reactivity in the two-layer system is now under exploration. The effect of the silicon atom on the stereoselectivity of H L A D H could be rationally explained based on the structure of the enzyme, that is, the presence of the small and the large alkyl-binding pockets in the active site of H L A D H (Eklund et al. 1982). As a result of the

213 Large olkyl binding pocket

~

~

z,.o,.

'%

tions. This work was supported in part by a Grant-in-Aid for Research from the Ministry of Education, Science and Culture, Japan.

References binding pocket

[A) Secondary alcohol

(B) Primary alcohol

Fig. 5A, B. Hypothetical models of the recognition of enantiomers by HLADH: E1 =Si or C

bigger radius of the silicon atom, the large group in 1 is more bulky than that in the corresponding carbon compound and, therefore, more difficult to fit the small alkyl-binding pocket in the enzyme active site (Fig. 5A). Consequently, the stereoselectivity of HLADH is enhanced by replacing the carbon atom with the silicon atom. The configuration of remaining 1 after stereoselective dehydrogenation with HLADH was determined to be R, the results being consistent with the Prelog rule (Bentley 1970) even in the case of the organosilicon compound. Unlike the secondary alcohol, the chiral centre of the primary alcohols is within the large group of the substrate. The large groups of the two enantiomers might differ from each other in their fit to the large alkyl-binding pocket as shown in Fig. 5B, but this difference is much smaller than that of the secondary alcohol. So HLADH showed rather poor stereoselectivity toward the primary alcohols. However, the greater bulkiness and/or the larger hydrophobicity of the trimethylsilyl group of 3 might facilitate the recognition of the difference of enantiomers by HLADH, accounting for the higher stereoselectivity of HLADH for 3 than for 6. In conclusion, it has been found in this work that HLADH is able to catalyse the stereoselective dehydrogenation of organosilicon compounds and that the silicon atom in the substrates improves the stereoselectivity of HLADH. The effects of the silicon atom could be explained in terms of its specific characteristics. To our knowledge, this work represents the first demonstration of the possibility of constructing useful stereoselective reaction systems for organosilicon compounds with HLADH and is, therefore, of great significance in both basic research on the enzymology of alcohol dehydrogenase and the production of useful chiral organosilicon compounds. Acknowledgements. We are indebted to Dr. T. Omata and Mr. E. Fukusaki, Nitto Denko Co., Ltd., for the generous gift of 1-trimethylsilyl-2-propanol, 1-trimethylsilyl-l-propanol and 2-trimethylsilyl-l-propanol used in this study, and to Prof. Y. Itoh and Dr. K. Tamao, Department of Synthetic Chemistry, Faculty of Engineering, Kyoto University, for their kind help and sugges-

Bentley R (1970) Molecular asymmetry in biology, vol II. Academic Press, New York Br/indOn CI, Eklund H, NordstrOm B, Boiwe T, SOderlund G, Zeppezauer E, Ohlsson I, Akeson A (1973) Structure of liver alcohol dehydrogenase at 2.9-A. resolution. Proc Natl Acad Sci USA 70: 2439-2442 Brown HC, Knights EF, Scouten CG (1974) Hydroboration. XXXVI. A direct route to 9-borabicyclo[3,3,1]nonane via the cyclic hydroboration of 1,5-cyclooctadiene. 9-Borabicyclo[3,3,1]nonane as a uniquely selective reagent for the hydroboration of olefins. J Am Chem Sac 96: 7765-7770 Calvin EW (1988) Silicon reagents in organic synthesis. Academic Press, New York Dale JA, Masher HS (1973) Nuclear magnetic resonance enantiomer reagents. Configurational correlations via nuclear magnetic resonance chemical shifts of diastereomeric mandelate, O-methyl-mandelate, and a-methoxy-a-trifluoromethylphenylacetate (MTPA) esters. J Am Chem Sac 95:512-519 Davis DD, Jacocks HM (1981) Deoxymetalation reactions. The mechanisms of deoxysilylation of mono-trimethylsilyl-and bistrimethylsilyl-substituted alcohols and a comparison to the mechanism of deoxystannylation and deoxyplumbylation. J Organomet Chem 206: 33-47 Eklund H, Plapp BV, Samama JP, Br~tndOnCI (1982) Binding of substrate in a ternary complex of horse liver alcohol dehydrogenase. J Biol Chem 257:14349-14358 Fessenden R J, Fessenden JS (1980) Trends in organosilicon biological research. Adv Organomet Chem 18:275-299 Hauser CR, Hance CR (1951) Preparation and reactions of a-halo derivatives of certain tetra-substituted hydrocarbon silanes. Grignard syntheses of some silyl compounds. J Am Chem Sac 73 : 5091-5096 Kawamoto T, Sonomoto K, Tanaka A (1991) Efficient optical resolution of 2-(4-chlorophenoxy)propanoic acid with lipase by the use of organosilicon compounds as substrate: the role of silicon atom in enzymatic recognition. J Biotechnol 18:85-92 Lee LG, Whitesides GM (1985) Enzyme-catalyzed organic synthesis: a comparison of strategies for in situ regeneration of NAD from NADH. J Am Chem Sac 107:6999-7008 ~i N, Kitahara H, Kira R (1990) Elution orders in the separation of enantiomers by high-performance liquid chromatography with some chiral stationary phases. J Chromatogr 535:213227 Patai S, Rappoport Z (1989) The chemistry of organic silicon compounds. Wiley, New York Soderquist JA, Brown HC (1980) Hydroboration. 56. Convenient and regiospecific route to functionalized organosilanes through the hydroboration of alkenylsilanes. J Org Chem 45 : 3571-3578 Tacke R, Zilch H (1986) Sila-substitution - a useful strategy for drug design? Endeavour, New Series 10:191-197 Whitesides GM, Wang CH (1985) Enzymes as catalysts in synthetic organic chemistry. Angew Chem Int Ed Engl 24:617-638 Zong M-H, Fukui T, Kawamoto T, Tanaka A (1991) Bioconversion of organosilicon compounds by horse liver alcohol dehydrogenase: the role of the silicon atom in enzymatic reactions. Appl Microbial Biotechnol 36:40-43