Enantioselectivity in the reduction of tricyclic hydroaromatic ketones by baker's yeast

Article abstract of DOI:10.1016/S0957-4166(98)00111-6

Three benzo-2-tetralones were hydrogenated to the corresponding alcohols by non-fermenting baker's yeast. Satisfactory yields but modest enantioselectivities were observed. The prevalent enantioform of the benzo- 2-tetralol was found to be in agreement with the predictive abstract model previously proposed for the enzymatic hydrogenation of aromatic ring substituted 2-tetralones.

Full text of DOI:10.1016/S0957-4166(98)00111-6

TETRAHEDRON:
ASYMMETRY
Pergamon
Tetrahedron: Asymmetry 9 (1998) 1381–1387
Enantioselectivity in the reduction of tricyclic hydroaromatic
ketones by baker’s yeast
Gabriele Fontana, Paolo Manitto,∗ Giovanna Speranza and Simona Zanzola
Dipartimento di Chimica Organica e Industriale, Università di Milano e Centro di Studio sulle Sostanze Organiche Naturali,
CNR, via Venezian 21, I-20133 Milano, Italy
Received 9 February 1998; accepted 12 March 1998
Abstract
Three benzo-2-tetralones were hydrogenated to the corresponding alcohols by non-fermenting baker’s yeast.
Satisfactory yields but modest enantioselectivities were observed. The prevalent enantioform of the benzo-2-
tetralol was found to be in agreement with the predictive abstract model previously proposed for the enzymatic
hydrogenation of aromatic ring substituted 2-tetralones. © 1998 Elsevier Science Ltd. All rights reserved.
1. Introduction
The possibility of obtaining optically active 2-tetralols 2 by baker’s yeast reduction of the correspond-
ing ketones has recently been explored in the case of 2-tetralones 1, differing in their hydroxy/methoxy
substitution at the aromatic ring.1 The enantioselectivity of the reaction carried out under ‘non-
fermenting’ conditions was found to be moderate and variable, with the exception of the conversion
of 5-methoxy-2-tetralone (e.e. ≥98%, 87% yield). We showed that enantioselectivity resulted from the
action of a single enzyme rather than from the competition between two or more dehydrogenases, thus
allowing a simple abstract model to be developed relating the prevalent enantioform of the product 2 and
its excess to the substitution pattern of the starting 2-tetralone 1.1
∗
Corresponding author. E-mail: manitto@icil64.cilea.it
0957-4166/98/$19.00 © 1998 Elsevier Science Ltd. All rights reserved.
PII: S0957-4166(98)00111-6

1382
G. Fontana et al. / Tetrahedron: Asymmetry 9 (1998) 1381–1387
Although the microbial reduction of polycyclic hydroaromatic α-oxo compounds has been examined,2
to our knowledge no analogous investigations have been performed on β-oxo derivatives. This situation,
together with the possibility of extending our predictive model to tricyclic hydroaromatic ketones,
prompted us to test the baker’s yeast reduction of compounds 4, 6 and 9.
2. Results and discussion
3,4-Dihydrophenanthren-2(1H)-one 43a and 1,2-dihydrophenanthren-3(4H)-one 63b were prepared by
rhodium(II)-catalyzed decomposition of α-diazoketones 3 and 5, respectively, and subsequent acidic
treatment of the reaction mixture with trifluoroacetic acid (Scheme 1a,b).3 3,4-Dihydroanthracen-2(1H)-
one 9 was conveniently synthesized starting from 1,4-dihydroanthracene 7, prepared in turn according to
Jadot and Roussel4 via the sequence reported in Scheme 1c.
Scheme 1.
The conversion of ketones 4, 6 and 9 into the corresponding alcohols 4a, 6a and 9a by Saccharomyces
cerevisiae was carried out under standard conditions (see Experimental). Benzo-2-tetralols were isolated
by flash chromatography of the ether extract of the reaction medium; the purity of each compound was
checked by TLC (three eluents) and the structure confirmed by NMR spectroscopy. The most abundant
enantioform of each benzo-2-tetralol and its excess were determined through the 19F NMR spectra of
the corresponding (S)-MTPA esters 4b, 6b and 9b, on the assumption that the CF3 signal at higher
field was due to the (R)-form. Although this empirical rule has been widely tested with 2-tetralols,1

G. Fontana et al. / Tetrahedron: Asymmetry 9 (1998) 1381–1387
1383
Table 1
Reduction of ketones 4, 6 and 9 using baker’s yeast
its application to Mosher’s esters of benzo-2-tetralols was confirmed via a non-empirical chiroptical
method.5 Thus, the exciton coupled CD spectra of the benzo-2-tetralols benzoates 4c, 6c and 9c were
compared with those reported in the literature for the corresponding enantiopure compounds.6–8 The
results of all biotransformation experiments are summarized in Table 1.
The prevalent (S)-configuration of the alcohol 4a was predictable on the basis of our model.1
In fact, if one considers the minimal destabilizing interactions visualized when the molecule of the
dihydrophenanthrenone 4, as represented by its van der Waals surface, is lodged in the virtual boundaries
of the dehydrogenase active site (Fig. 1a), the (R)-orientation appears to be disfavoured due to an overlap
of H-10 with the Y-boundary and to counter-clockwise α to α0 rotation.
Analogous considerations led to the forecast of a preference for an (R)-configuration in the case of the
dehydrogenation of compound 6 (Fig. 1b). Finally, the fact that the enantioselectivity of the bioconversion
of the dihydroanthracenone 9 into the alcohol 9a was practically nil can be interpreted as the result of
a very similar extent of destabilization for both (S)- and (R)-orientations of the substrate in the enzyme
active site (Fig. 1c).
3. Experimental
TLC was performed on silica gel F254 precoated aluminum sheets (0.2 mm layer, Merck); eluent
systems: A, hexane:ethyl acetate (2:1); B, hexane:ethyl acetate (3:2); C, chloroform:ether (5:1). Hex-
ane:ethyl acetate (3:1) and hexane:ether (7:3) were used as additional eluents to check the purity of
1
benzo-2-tetralols. Silica gel (40–63 µm) from Merck was used for flash chromatography. H NMR
(200.12 or 300.13 MHz), 13C NMR (50.32 or 75.47 MHz) and 19F NMR (282.40 MHz) spectra were
1
recorded in CDCl3 solution. The solvent signal was used as an internal standard for H NMR and 13C
NMR (7.25 and 77.00 ppm), while CFCl3 was added for measuring 19F NMR chemical shifts. EIMS
spectra were run on a VG 7070 EQ mass spectrometer operating at 70 eV. UV spectra were obtained
in CHCl3 on an HP 8452A spectrophotometer. Optical rotations were measured in CHCl3 at 25°C on
a Perkin–Elmer 241 polarimeter. Circular dichroism curves were recorded in MeOH:dioxane (9:1) at

1384
G. Fontana et al. / Tetrahedron: Asymmetry 9 (1998) 1381–1387
Fig. 1. Schematic representation of tricyclic aromatic ketones 4 [A], 6 [B], 9 [C] (van der Waals surfaces) into the virtual active
site of the baker’s yeast dehydrogenase. See Manitto et al.1 for the construction of the enzyme-pocket model and for the order
of destabilizing interactions

G. Fontana et al. / Tetrahedron: Asymmetry 9 (1998) 1381–1387
1385
room temperature on a Jasco J-500C spectropolarimeter. Baker’s yeast was from Distillerie Italiane (S.
Quirico-Trecasali, Parma, Italy).
3.1. 2,3-Epoxy-1,2,3,4-tetrahydroanthracene 8
To a solution of 7 (1.09 g, 6.05 mmol)4 in 10 mL of CH2Cl2 at −10°C was added a filtered solution
of m-CPBA (2 g, 6.35 mmol) in CH2Cl2 (16 mL) dropwise, and the reaction mixture stirred for 12
h. Excess oxidant was destroyed by adding a 37% aqueous NaHSO3 solution (10 mL). The organic
layer was separated, washed with 5% aqueous NaHCO3, and dried (Na2SO4). Removal of solvent under
reduced pressure and purification by flash chromatography (eluent A) gave 900 mg of pure 8 (4.6 mmol,
76% yield). TLC Rf 0.1 (eluent A); 1H NMR (200 MHz) δ 3.32–3.58 (m, 6H, H2-1, H-3 and H2-4), 7.39
(dd, 2H, J=6.2, 3.2 Hz, H-6 and H-7), 7.55 (s, 2H, H-9 and H-10), 7.73 (dd, 2H, J=6.2, 3.2 Hz, H-5 and
H-8); 13C NMR (50 MHz) δ 30.21 (t), 51.58 (d), 125.22 (d), 126.96 (d), 127.42 (d), 130.47 (s).
3.2. 3,4-Dihydroanthracen-2(1H)-one 9
Compound 8 (886 mg, 4.52 mmol) was transferred to a suspension of LiAlH4 (0.342 g, 9 mmol) in 200
mL of dry ether by means of continuous extraction with ether using a Soxhlet apparatus. After refluxing
for 5 h under N2, the reaction mixture was quenched by adding in sequence 0.5 mL of water, 1 mL of
5 N NaOH, and 0.5 mL of water. The white slurry was filtered and the residue washed with ether. The
combined ether extracts were dried (Na2SO4) and evaporated under reduced pressure to give 790 mg of
1,2,3,4-tetrahydroanthracen-2-ol 9a (3.99 mmol, 88% yield) which was homogeneus by TLC analysis.
TLC Rf 0.27 (eluent C); 1H NMR (300 MHz) δ 1.75 (br s, 1H, OH), 1.82–1.98 (m, 1H, H-3a), 2.10–2.19
(m, 1H, H-3b), 2.91–3.05 (m, 2H, H-1a and H-4a), 3.12 (app dt, 1H, J=16.8, 5.8 Hz, H-4b), 3.29 (dd,
1H, J=16.1, 4.7 Hz, H-1b), 4.20–4.29 (m, 1H, H-2), 7.36–7.42 (m, 2H, H-6 and H-7), 7.57 (s, 2H, H-9
and H-10), 7.69–7.77 (m, 2H, H-5 and H-8); 13C NMR (75 MHz) δ 27.10 (t), 31.83 (t), 38.74 (t), 67.34
(d), 125.13 (d), 125.21 (d), 126.29 (d), 126.98 (d), 127.42 (d), 132.20 (s), 132.27 (s), 133.26 (s), 134.47
(s); UV λmax (log ε): 242 (3.61), 278 (3.48), 288 (3.49).
The above 1,2,3,4-tetrahydroanthracen-2-ol was oxidized by the Swern procedure9 giving rise to a
brown oil, which was flash chromatographed (eluent hexane with increasing portions of ethyl acetate) to
obtain 9 as a white solid (30% overall yield). TLC Rf 0.56 (eluent C); 1H NMR (300 MHz) δ 2.58 (app
t, J=7.0 Hz, 2H, H2-3), 3.21 (app t, J=7.0 Hz, 2H, H2-4), 3.76 (s, 2H, H2-1), 7.41–7.48 (m, 2H, H-6 and
H-7), 7.58 and 7.69 (2×s, 2×1H, H-9 and H-10), 7.73–7.80 (m, 2H, H-5 and H-8); 13C NMR (75 MHz)
δ 28.43 (t), 38.29 (t), 45.92 (t), 125.53 (d), 125.70 (d), 125.82 (d), 126.47 (d), 127.23 (d), 131.36 (s),
132.55 (s), 135.12 (s), 210.22 (s).
3.3. General procedure for biotransformations
A suspension of baker’s yeast (100 g) in preboiled distilled water (1 L) was kept at 37°C for 30
min. Then the substrate (1 g), dissolved in the smallest amount of 1,4-dioxane, was gradually added
and the mixture vigorously stirred for 3–5 days at 37°C. Progress of the reduction was monitored
by TLC analysis. When the reaction was complete, the product was continously extracted with ethyl
ether. The ether extract was dried (Na2SO4), evaporated under reduced pressure and purified by flash
chromatography using the same eluent as for TLC (see Table 1 for yields and enantiomeric ratios). For
the preparation and analysis of MTPA esters, see Manitto et al.1

1386
G. Fontana et al. / Tetrahedron: Asymmetry 9 (1998) 1381–1387
3.4. General procedure for the preparation of benzoyl derivatives
To a solution of alcohol (0.2 mmol) in 0.6 mL of dry CH2Cl2 and 0.6 mL of dry pyridine, was added
50 µL of freshly distilled benzoyl chloride and the reaction mixture was refluxed until completion. After
the usual workup the crude product was purified by flash chromatography using the same eluent as for
TLC.
3.5. (S)-1,2,3,4-Tetrahydrophenanthren-2-ol 4a
25
1
TLC Rf 0.29 (eluent A); [α]D −51.0 (c 0.42, 72% e.e.) (cf. Koreeda et al.7); H NMR (300 MHz)
δ 1.91–2.06 (m, 2H, H-3a and OH), 2.17–2.34 (m, 1H, H-3b), 2.92 (dd, 1H, J=15.6 and 7.8 Hz, H-1a),
3.12 (app dd, 1H, J=17.1, 6.8 Hz, H-4a), 3.21 (dd, 1H, J=15.6, 4.9 Hz, H-1b), 3.35 (app dt, 1H, J=17.1,
5.9 Hz, H-4b), 4.18–4.26 (m, 1H, H-2), 7.18 (d, 1H, J=8.5 Hz, H-10), 7.42–7.53 (m, 2H, H-6 and H-7),
7.64 (d, 1H, J=8.5 Hz, H-9), 7.79 (d, 1H, J=6.8 Hz, H-8), 7.95 (d, 1H, J=8.8 Hz, H-5); 13C NMR (50
MHz) δ 23.68 (t), 31.24 (t), 39.17 (t), 66.89 (d), 122.90 (d), 124.94 (d), 125.98 (d), 126.32 (d), 128.10
(d), 128.35 (d), 130.25 (s), 131.24 (s), 132.00 (s), 132.26 (s); UV λmax (log ε) 288 (3.57), 278 (3.56), 244
(3.65), 324 (2.62); 4a (S)-MTPA ester 4b, TLC Rf 0.48 (eluent B); 19F NMR δ −71.99 (14%), −71.78
(86%).
3.6. (S)-1,2,3,4-Tetrahydrophenanthren-2-ol benzoate 4c
25
TLC Rf 0.52 (eluent A); UV λmax (log ε) 228 (4.94), 280 (3.85); [α]D −74.8 (c 0.72, 72% e.e.);
CD λmax (∆ε) 219 (+13.6), 229 (−25.2) (cf. Koreeda et al.7); 1H NMR (200 MHz) δ 2.24–2.37 (m, 2H,
H-3), 3.14–3.49 (m, 4H, H-1 and H-4), 5.54–5.65 (m, 1H, H-2), 7.21 (d, 1H, J=8.5 Hz), 7.38–7.56 (m,
6H) 7.67 (d, 1H, J=8.5 Hz), 7.83 (d, 1H, J=7.5 Hz) and 8.01 (app t, 2H, J=8.1 Hz) (aromatic H); 13C
NMR (75 MHz) δ 23.16 (t), 27.63 (t), 35.39 (t), 70.01 (d), 122.93 (d), 125.07 (d), 126.10 (d), 126.45 (d),
127.95 (d), 128.29 (d), 128.47 (d), 129.57 (d), 130.23 (s), 130.56 (s), 130.72 (s), 132.10 (s), 132.35 (s),
132.80 (d), 166.20 (s).
3.7. (R)-1,2,3,4-Tetrahydrophenanthren-3-ol 6a
25
1
TLC Rf 0.25 (eluent B); [α]D +13.8 (c 0.40, 30% e.e.) [lit.6 −49, for optically pure (S)-6a]; H
NMR (300 MHz) δ 1.83–1.97 (m, 2H, H-2a and OH), 2.11–2.16 (m, 1H, H-2b), 2.93–3.14 (m, 3H, H2-1
and H-4a), 3.51 (dd, 1H, J=16.7, 2.0 Hz, H-4b), 4.26–4.34 (m, 1H, H-3), 7.21 (d, 1H, J=8.4 Hz, H-10),
7.42–7.53 (m, H-6 and H-7), 7.63 (d, 1H, J=8.4 Hz, H-9), 7.79 (d, 1H, J=9.1 Hz, H-8), 7.93 (d, 1H,
J=8.3 Hz, H-5); 13C NMR (75 MHz) δ 28.02 (t), 31.32 (t), 35.02 (t), 67.67 (d), 122.73 (d), 125.00 (d),
126.12 (d), 126.29 (d), 127.48 (d), 128.49 (d), 128.5 (s), 132.28 (s), 132.31 (s), 133.03 (s); UV λmax
(log ε) 244 (3.63), 282 (3.69), 276 (3.66); 6a (S)-MTPA ester 6b, TLC Rf 0.70 (eluent B); 19F NMR δ
−71.32 (65%), −71.14 (35%).
3.8. (R)-1,2,3,4-Tetrahydrophenanthren-3-ol benzoate 6c
25
TLC Rf 0.67 (eluent A); UV λmax (log ε) 228 (5.00), 280 (3.93); [α]D −8.54 (c 0.57, 30% e.e.);
CD λmax (∆ε) 222 (+21), 229 (−30) (cf. Boyd et al.6); 1H NMR (200 MHz) δ 2.17–2.33 (m, 2H, H2-2),
3.00–3.41 (m, 3H, H2-1 and H-4a), 3.65 (dd, 1H, J=17.1, 5.3 Hz, H-4b), 5.68 (app quint., 1H, J=5.7
Hz, H-3), 7.27 (d, 1H, J=8.4 Hz, H-10), 7.40–7.61 (m, 4H, H-6, H-7, H-20, H-30, H-50 and H-60), 7.69

G. Fontana et al. / Tetrahedron: Asymmetry 9 (1998) 1381–1387
1387
(d, 1H, J=8.4 Hz, H-9), 7.84 (dd, 1H, J=8.3, 1.9 Hz, H-8), 7.94 (d, 1H, J=7.8 Hz, H-40), 8.07 (dd, 1H,
J=7.9, 1.0 Hz, H-5); 13C NMR (50 MHz) δ 27.21 (t), 27.53 (t), 31.35 (t), 70.45 (d), 122.61 (d), 124.99
(d), 126.09 (d), 126.32 (d), 127.42 (d), 128.24 (d), 129.56 (d), 130.59 (s), 132.17 (s), 132.31 (s), 132.82
(s), 166.18 (s).
3.9. (S)-1,2,3,4-Tetrahydroanthracen-2-ol 9a
25
[α]D −7.1 (c 0.42, 12% e.e.) [lit.10 −47, for 90% optically pure (S)-9a]; spectral data matched those
reported above for the racemic compound; 9a (S)-MTPA ester 9b, TLC Rf 0.69 (eluent B); 19F NMR δ
−71.40 (44%), −71.23 (56%).
3.10. (S)-1,2,3,4-Tetrahydroanthracen-2-ol benzoate 9c
TLC Rf 0.66 (eluent A); UV λmax (log ε) 228 (5.00), 280 (3.82), 276 (3.85); [α]D25 +1.95 (c 0.2); CD
λmax (∆ε) 222 (−4.8), 229 (+5.6) (cf. Akhtar et al.8); 1H NMR (300 MHz) δ 2.15–2.28 (m, 2H, H2-3),
3.03–3.13 (app dt, 1H, J=16.6, 6.8 Hz, H-4a), 3.18–3.27 (m, 2H, H-1a and H-4b), 3.42 (dd, 1H, J=16.6,
4.9 Hz, H-1b), 5.52–5.6 (m, 1H, H-2), 7.37–7.74 (m, 8H), 7.99 (dd, 1H, J=6.8, 2.0 Hz) and 8.15 (d,
2H, J=6.8 Hz) (aromatic H); 13C NMR (50 MHz) δ 26.50, 28.22, 34.93, 70.29, 125.19, 125.29, 126.35,
127.02, 127.37, 128.27, 128.82, 129.52, 130.51, 132.15, 132.5, 132.81, 134.47, 162.5.
Acknowledgements
We thank Dr. Diego Monti (CNR, Milan) for running NMR spectra and MURST (Italy) for financial
support.
References
1. Manitto, P.; Speranza, G.; Monti, D.; Fontana, G.; Panosetti, E. Tetrahedron 1995, 51, 11531–11546.
2. Kabuto, K.; Imuta, M.; Kempner, E. S.; Ziffer, H. J. Org. Chem. 1978, 43, 2357–2361.
3. (a) Manitto, P.; Monti, D.; Zanzola, S.; Speranza, G. J. Org. Chem. 1997, 62, 6658–6665; (b) Manitto, P.; Monti, D.;
Speranza, G. J. Org. Chem. 1995, 60, 484–485.
4. Jadot, J.; Roussel, J. Bull. Soc. Roy. Sci. Liège 1954, 23, 69–78 (Chem. Ab. 1955, 49, 10909i).
5. Harada, N.; Nakanishi, K. Circular Dichroic Spectroscopy — Exciton Coupling in Organic Stereochemistry; University
Science Books: Mill Valley, CA, 1983; 350 pp.
6. Boyd, D. R.; Greene, R. M. E.; Neill, J. D.; Stubbs, M. E.; Yagi, H.; Jerina, D. M. J. Chem. Soc., Perkin Trans. 1 1981,
1477–1482.
7. Koreeda, M.; Akhtar, M. N.; Boyd, D. R.; Neill, J. D.; Gibson, D. T.; Jerina, D. M. J. Org. Chem. 1978, 43, 1023–1027.
8. Akhtar, M. N.; Boyd, D. R.; Thompson, N. J.; Koreeda, M.; Gibson, D. T.; Mahadevan, V.; Jerina, D. M. J. Chem. Soc.,
Perkin Trans. 1 1975, 2506–2511.
9. Mancuso, A. J.; Huang, S.-L.; Swern, D. J. Org. Chem. 1978, 43, 2480–2482.
10. Akhtar, N. M.; Boyd, D. R. J. Chem. Soc., Chem. Commun. 1975, 916–917.