Chiral building-blocks by chemoenzymatic desymmetrization of 2-ethyl- 1,3-propanediol for the preparation of biologically active natural products

Article abstract of DOI:10.1016/S0957-4166(99)00020-8

2-Ethyl-1,3-propanediol 1 and its related di-O-acetate 2 were desymmetrized by partial chemoenzymatic acetylation and deacetylation, by Pseudomonas fluorescens lipase (Amano P.; PFL), to (R)-1-O-acetyl-2-ethyl- 1,3-propanediol 3. On treatment of 3 with I₂/Ph₃P/imidazole the related (S)-1-O-acetyl-2-ethyl-3-iodopropanol 4 was obtained and transformed into the corresponding triphenylphosphonium salt 5. Reaction of [(S)-3-acetoxy-2- ethylpropylidene]triphenylphosphorane 6, prepared from 5, with 2,3:4,5-di-O- isopropylidene-β-D-arabino-hexos-2-ulopyranose 7 gave (Z)-3-C-acetoxymethyl- 1,2,3,4,5-pentadeoxy-6,7:8,9-di-O-isopropylidene-β-D-manno-dec-4-ene-6-ulo- 6,10-pyranose 8 which was hydrogenated to 9 and subsequently deacylated to 10. Treatment of 10 with Me₂CO/H⁺ caused a rearrangement to (3R,4R,5S,6R,9R)-9-ethyl-5-hydroxy-3,4-isopropylidenedioxy-1,7- dioxaspiro[5.5]undecane 11, which closely matched the skeleton of the talaromycins.

Full text of DOI:10.1016/S0957-4166(99)00020-8

TETRAHEDRON:
ASYMMETRY
Pergamon
Tetrahedron: Asymmetry 10 (1999) 449–455
Chiral building-blocks by chemoenzymatic desymmetrization of
2-ethyl-1,3-propanediol for the preparation of biologically active
natural products
Isidoro Izquierdo,∗ Maria T. Plaza, Miguel Rodríguez and Juan Tamayo
Department of Organic Chemistry, Faculty of Pharmacy, University of Granada, 18071 Granada, Spain
Received 30 November 1998; accepted 19 January 1999
Abstract
2-Ethyl-1,3-propanediol 1 and its related di-O-acetate 2 were desymmetrized by partial chemoenzymatic
acetylation and deacetylation, by Pseudomonas fluorescens lipase (Amano P.; PFL), to (R)-1-O-acetyl-2-ethyl-
1,3-propanediol 3. On treatment of 3 with I2/Ph3P/imidazole the related (S)-1-O-acetyl-2-ethyl-3-iodopropanol 4
was obtained and transformed into the corresponding triphenylphosphonium salt 5. Reaction of [(S)-3-acetoxy-2-
ethylpropylidene]triphenylphosphorane 6, prepared from 5, with 2,3:4,5-di-O-isopropylidene-β-D-arabino-hexos-
2-ulopyranose 7 gave (Z)-3-C-acetoxymethyl-1,2,3,4,5-pentadeoxy-6,7:8,9-di-O-isopropylidene-β-D-manno-dec-
4-ene-6-ulo-6,10-pyranose 8 which was hydrogenated to 9 and subsequently deacylated to 10. Treatment of 10
with Me2CO/H+ caused a rearrangement to (3R,4R,5S,6R,9R)-9-ethyl-5-hydroxy-3,4-isopropylidenedioxy-1,7-
dioxaspiro[5.5]undecane 11, which closely matched the skeleton of the talaromycins. © 1999 Elsevier Science
Ltd. All rights reserved.
1. Introduction
Scheme 1 clearly shows that chiral moieties derived from 1,3-propanediol are commonly present
in many molecules exhibiting important biological activities, such as β-blockers,1 antifungal agents,2
toxins,3 etc., and hence there is much interest in finding simple synthetic methods for such moieties:
those methods using chemoenzymatic routes have attracted the attention of many groups in recent years.4
On the other hand, the synthesis of the dioxaspiro compound 11 (see Scheme 3) and its C-9 epimer
have been reported5 by our group but this approach was flawed in the stereoselectivity achieved during
formation of branching at the 1,3-diol moiety, and required the use of tedious chromotography. Therein
lies the convenience of exploring new methodologies for the preparation of those molecules from easily
∗
Corresponding author. E-mail: isidoro@platon.ugr.es
0957-4166/99/$ - see front matter © 1999 Elsevier Science Ltd. All rights reserved.
PII: S0957-4166(99)00020-8

450
I. Izquierdo et al. / Tetrahedron: Asymmetry 10 (1999) 449–455
available chirons as starting materials in the stereospecific synthesis of such moieties. The results of this
new approach are reported below.
Scheme 1.
2. Results and discussion
Treatment of 2-ethyl-1,3-propanediol 16 with vinyl acetate in CH2Cl2 under the presence of lipase
from Pseudomonas fluorescens (Amano P.; PFL) at room temperature, caused its desymmetrization to
afford (R)-1-O-acetyl-2-ethyl-1,3-propanediol 3,6 isolated in 72% yield with a 46% ee, determined by
GLC (B) on a β-DEX™ column (see Fig. 1). These findings differed from those recently reported,7
where a lower ee (19%) was obtained and isolation of the products was not described. On the other hand,
when 1,3-di-O-acetyl-2-ethyl-1,3-propanediol (2)6 was subjected to partial hydrolysis catalyzed by PFL,
compound 3 was obtained in 65% yield, after 1.5 h, with a 94% ee (see Fig. 2).
Although partial enzymatically catalyzed acylation of 2-substituted-1,3-propanediols and hydrolysis
of the corresponding diesters by lipases usually display the same prochiral selectivity8 and result in
products of opposite absolute stereochemistry, this was not the case in the present work where only the
R-isomer was obtained. This behaviour has been reported in other cases.2
Treatment of 3, from partial hydrolysis of 2, with I2/Ph3P/imidazole9 gave (S)-1-O-acetyl-
2-ethyl-3-iodopropanol
4
that was straightforwardly transformed into the corresponding
phosphonium salt 5 by reaction with Ph3P in toluene and subsequently into [(S)-3-acetoxy-2-
Figure 1. Kinetic data for enzymatically (PFL) catalyzed monoacetylation of 1 and GLC analysis (ee) on a β-DEX™ column

I. Izquierdo et al. / Tetrahedron: Asymmetry 10 (1999) 449–455
451
Figure 2. Kinetic data for enzymatically (PFL) catalyzed partial hydrolysis of 2 and GLC analysis (ee) on a β-DEX™ column
ethylpropylidene]triphenylphosphorane 6 (Scheme 2). Reaction of 6 with 2,3:4,5-di-O-isopropylidene-
β-D-arabino-hexos-2-ulopyranose5 (7) gave (Z)-3-C-acetoxymethyl-1,2,3,4,5-pentadeoxy-6,7:8,9-di-
O-isopropylidene-β-D-manno-dec-4-ene-6-ulo-6,10-pyranose 8 which was hydrogenated to 9 and
subsequently deacylated to 10 (Scheme 3). Treatment of 10 with Me2CO/H+ caused its rearrangement to
afford (3R,4R,5S,6R,9R)-9-ethyl-5-hydroxy-3,4-isopropylidenedioxy-1,7-dioxaspiro[5.5]undecane (11),
which showed the same physical and spectroscopic data as those previously reported5 and where its
transformation into talaromycin A and B was described.
Scheme 2.
3. Conclusions
The results described above, indicate that the use of the chiral pool (carbohydrates) combined with
chiral catalysts (enzymes), both from natural sources, could be an excellent methodology in order to
overcome many problems in the stereoselective synthesis of complex biologically active natural products.

452
I. Izquierdo et al. / Tetrahedron: Asymmetry 10 (1999) 449–455
Scheme 3.
4. Experimental
4.1. General
Melting points were determined with a Gallenkamp apparatus and are uncorrected. Solutions were
1
dried over MgSO4 before concentration under reduced pressure. The H and 13C NMR spectra were
recorded with Bruker AMX-300, AM-300, and ARX-400 spectrometers for solutions in CDCl3 (in-
ternal Me4Si). IR spectra were recorded with a Perkin–Elmer 782 instrument and mass spectra with
a Hewlett–Packard 5988A and Fisons Platform II and VG Autospec-Q mass spectrometers. Optical
rotations were measured for solutions in CHCl3 (1 dm tube) with a Jasco DIP-370 polarimeter. GLC
was performed on a Hewlett–Packard 6890 gas chromatograph equipped with split/splitless injector,
a flame-ionization detector, and: (A) a capillary HP-5 column (30 m×0.25 mm i.d.×0.25 µm film
thickness) at 110°C; or (B) a β-DEX™ 325 (Supelco™) capillary column (30 m×0.25 mm i.d.×0.25
µm film thickness) 50 min at 80°C, program to 110°C, 10°C/min. The He flow rate was 1.1 mL/min,
the injection port and the zone-detector temperatures were 275°C. TLC was performed on precoated
silica gel 60 F254 aluminium sheets and detection by charring with H2SO4. Column chromatography was
performed on silica gel (Merck, 7734). The noncrystalline compounds were shown to be homogeneous
by chromatography and characterized by NMR and HRMS.
4.2. Partial enzymatic (PFL) acetylation of 2-ethyl-1,3-propanediol 1
To a gently stirred solution of 1 (1.25 g, 12 mmol) in dry dichloromethane (25 mL) and vinyl acetate
(2.5 mL, 27 mmol) was added PFL (Amano P., 600 mg), and the mixture maintained at room temperature
for 4 h. The reaction was monitored by GLC analysis (A). The enzyme was removed by filtering,
thoroughly washed with dichloromethane, and the filtrate and washings concentrated to a residue that
was chromatographed (ether:hexane=1:4 -→ 1:1) to afford first 1,3-di-O-acetyl-2-ethyl-1,3-propanediol
(2, 116 mg), tR 9.57 min (A). The second fraction was (R)-1-O-acetyl-2-ethyl-1,3-propanediol (3, 1 g,

I. Izquierdo et al. / Tetrahedron: Asymmetry 10 (1999) 449–455
453
25
25
72%), tR 5.45 min (A), [α]D +7.5, [α]405 +14.7 (c 1.7), tR 54.18 min (B), ee 46% (see Fig. 1). The
spectroscopic data for 2 and 3 were identical to those previously reported.6 Finally, compound 1 (257
mg), tR 3.72 min (A) was recovered.
4.3. 1,3-Di-O-acetyl-2-ethyl-1,3-propanediol 2
To a cooled solution of 1 (1.5 g, 14.4 mmol), Et3N (6.3 mL, 45 mmol), DMAP (100 mg) in dry
dichloromethane (30 mL), and acetic anhydride (4.3 mL, 45 mmol) were added and the mixture was left
at room temperature for 6 h. Work-up of the reaction mixture as usual gave, after column chromatography
(ether:hexane=1:4), 2 (2.55 g, 94%) whose physical and spectroscopic data were the same as those
previously reported.6
4.4. Partial enzymatic (PFL) hydrolysis of 2
To a gently stirred suspension of 2 (2 g, 10.6 mmol) in 100 mL of a buffered (pH 7) aqueous 0.5 M
phosphate solution (KH2PO4) was added PFL (500 mg). Stirring was maintained for 1.5 h. GLC analysis
(A) of the mixture then revealed the presence of 1 (10.6%), 2 (4%), and 3 (85.1%). The enzyme was
filtered off and the filtrate saturated with sodium chloride and extracted with ethyl acetate. The combined
extracts were concentrated. Column chromatography (ether:hexane=1:2) of the residue gave 3 (1 g, 65%);
23
[α]D +11, [α]40525 +26 (c 1.4), tR 55.38 min (B), ee 94% (see Fig. 2).
4.5. (S)-1-O-Acetyl-2-ethyl-3-iodopropanol 4
To a stirred solution of I2 (2.1 g, 8.3 mmol) in dry dichloromethane (20 mL) was added Ph3P (2.15
g, 8.2 mmol) and then imidazole (1.07 g, 16 mmol) at room temperature. After 5 min, compound 3 (ee
94%) (950 mg, 6.5 mmol) was added and the mixture was stirred for 30 min. GLC analysis (A) then
showed the presence of a new compound (tR 11.39 min). The reaction mixture was washed with aqueous
10% KHSO4, brine and concentrated. The residue was extracted with ether:hexane=1:5 to remove most
of the Ph3PO and the combined extracts concentrated. Column chromatography (ether:hexane=1:9) of
24
25
the residue gave 4 (1.54 g, 93%) as an oil; [α]D +5.5, [α]405 +14 (c 1.8), tR 17.15 min (B at 115°C);
max
film
1
ν
1744 cm−1 (MeCO2). NMR data: H, δ 4.16 (dd, 1H, J1a,1b=11.2, J1a,2=4.8 Hz, H-1a), 3.98 (dd,
1H, J1b,2=7.2 Hz, H-1b), 3.37 (dd, 1H, J3a,3b=10.4, J2,3a=4.6 Hz, H-3a), 3.32 (dd, 1H, J2,3b=5.1 Hz, H-
3b), 2.11 (s, 3H, Ac), 1.57 (m, 1H, H-2), 1.45 (m, 2H, H-10), and 0.98 (t, 3H, J=7.4 Hz, H-20); 13C, δ
170.92 (MeCO), 66.47 (C-1), 40.12 (C-2), 24.25 (C-10), 20.94 (MeCO), 10.97 (C-20), and 10.53 (C-3).
Mass spectrum (CI, CH4): m/z 257 (20.9%, M++1), 255 (9.0, M+−1), 229 (9.8, M++1−C2H4), 197 (23.1,
+
+
M++1−AcOH), 143 (49.4, CH3IH+), 129 (25.5, IH2 ) and 57 (100, C4H9 ).
4.6. [(S)-3-Acetoxy-2-ethylpropyl]triphenylphosphonium iodide 5
A stirred solution of 4 (1.35 g, 5.3 mmol) and Ph3P (1.5 g, 5.7 mmol) in dry toluene (15 mL) was
refluxed for 3 days. After cooling, a crystalline precipitate appeared, the solid was filtered and washed
27
28
with ether to afford crystalline 5 (2.6 g, quantitative): mp: 81–83°C; [α]D −13, [α]405 −30 (c 1.2,
1
methanol). NMR data (d3-MeOD): H, δ 7.92–7.87 and 7.78–7.73 (2m, 15H, 3Ph), 3.98 (dd, 1H,
J
J
3a,3b=11.5, J2,3a=5.9 Hz, H-3a), 3.94 (dd, 1H, J2,3b=4.8 Hz, H-3b), 3.62 (ddd, 1H, J1a,1b=14.3, J1a,2=5.6,
1a,P=15.9 Hz, H-1a), 3.37 (dd, 1H, J1b,2=6.3, J1b,P=15.9 Hz, H-1b), 2.18 (m, 1H, H-2), 2.00 (s, 3H, Ac),
1.43 (m, 1H, H-10a), 1.25 (m, 1H, H-10b), and 0.79 (t, 3H, J=7.5 Hz, H-20); 13C, δ 172.35 (MeCO),

454
I. Izquierdo et al. / Tetrahedron: Asymmetry 10 (1999) 449–455
66.53 (C-3), 35.40 (C-2), 26.61 (C-1), 25.37 (C-10), 20.78 (MeCO), and 10.97 (C-20). Anal. calcd for
C25H28IO2P: C, 57.92; H, 5.44. Found: C, 58.21; H, 5.32. S, 3.21.
4.7. (Z)-3-C-Acetoxymethyl-1,2,3,4,5-pentadeoxy-6,7:8,9-di-O-isopropylidene-β-D-manno-dec-4-ene-
6-ulo-6,10-pyranose 8
To a stirred solution of 2,3:4,5-di-O-isopropylidene-β-D-arabino-hexos-2-ulopyranose5 (7, 1.06 g, 4.1
mmol) in dry diglyme (7 mL) under argon, potassium tert-butoxide (500 mg, 4.5 mmol) was added
at room temperature. Compound 5 (2.2 g, 4.2 mmol) was added and the mixture was left for 30 min.
TLC (ether:hexane=3:2) then revealed the presence of a faster-running compound. The mixture was
diluted with ether (50 mL) and the resulting solution was washed with brine, then concentrated. Column
chromatography (ether:hexane=1:3) of the residue gave 8 (470 mg, 31%) as a colourless oil; [α]D25 −23,
26
[α]405 −57 (c 1.7), tR 6.56 min (A); νmfilamx 1741 (MeCO2), 1384 and 1373 cm−1 (CMe2). NMR data:
1H, δ 5.62 (d, 1H, J4,5=11.6 Hz, H-5), 5.29 (t, 1H, J3,4=11.6 Hz, H-4), 4.57 (dd, 1H, J7,8=2.4, J8,9=7.9
Hz, H-8), 4.20–4.17 (m, 2H, H-7,9), 4.07 (dd, 1H, J1 a,1 b=10.7, J1 a,3=5.7 Hz, H-10a), 3.91 (dd, 1H,
0
0
0
0
J1 b,3=7.0 Hz, H-10b), 3.81 (dd, 1H, J9,10ax=1.9, J10ax,10eq=12.9 Hz, H-10ax), 3.67 (d, 1H, H-10eq), 3.33
(m, 1H, H-3), 2.01 (s, 3H, Ac), 1.50, 1.45, and 1.32 (3s, 12H, 2CMe2), 1.57–1.18 (m, 2H, H-2), and 0.87
(t, 3H, J=7.5 Hz, H-1); 13C, δ 175.39 (MeCO), 135.18 (C-5), 131.16 (C-4), 109.13 and 108.19 (2CMe2),
74.85 (C-7), 70.67 and 70.56 (C-8,9), 67.20 (C-10), 60.83 (C-10), 38.64 (C-3), 26.26, 25.99, 25.03, and
24.80 (2CMe2), 24.38 (C-2), 21.12 (MeCO), and 11.58 (C-1). Mass spectrum (LSIMS): m/z 393.19027
(M++Na). For C19H30O7Na 393.18892 (deviation −3.4 ppm).
4.8. 3-C-Acetoxymethyl-1,2,3,4,5-pentadeoxy-6,7:8,9-di-O-isopropylidene-β-D-manno-dec-6-ulo-6,10-
pyranose 9
A solution of 8 (1.06 g, 2.83 mmol) in dry methanol (20 mL) was hydrogenated over 10% Pd–C (220
mg) at 75 psi for 24 h. GLC (A) then revealed the presence of a new compound (tR 7.60 min). The catalyst
was filtered off, washed with methanol and the combined filtrate and washings concentrated to a residue
which chromatographed (ether:hexane=1:3) to afford 9 (880 mg, 85%) as a colourless syrup; [α]D26 −10
1
(c 1.2). NMR data: H, δ 4.55 (dd, 1H, J7,8=2, J8,9=8 Hz, H-8), 4.20 (bd, 1H, H-9), 4.07 (d, 1H, H-7),
3.98 (m, 2H, CH2OAc), 3.83 (bd, 1H, H-10ax), 3.70 (d, 1H, J10ax,10eq=13 Hz, H-10eq), 2.02 (s, 3H, Ac),
1.90–1.30 (m, 7H, H-2,3,4,5), 1.50, 1.46, and 1.33 (3s, 12H, 2CMe2), and 0.88 (t, 3H, J1,2=7.4 Hz, H-1);
13C, δ 171.34 (MeCO), 108.97 and 107.48 (2CMe2), 104.21 (C-6), 74.93 (C-7), 70.95 and 70.72 (C-8,9),
66.70 (C-10), 61.02 (C-10), 38.77 (C-3), 38.13 (C-5), 26.49, 25.88, 25.19, and 24.20 (2CMe2), 23.98 and
23.62 (C-2,4), 21.04 (MeCO), and 10.90 (C-1). Mass spectrum (LSIMS): m/z 395.20472 (M++Na). For
C19H32O7Na 395.20457 (deviation −0.4 ppm).
4.9. 1,2,3,4,5-Pentadeoxy-3-C-hydroxymethyl-6,7:8,9-di-O-isopropylidene-β-D-manno-dec-6-ulo-6,10-
pyranose 10
To a solution of 9 (480 mg, 1.29 mmol) in dry methanol (15 mL) 0.5 M sodium methoxide (0.25 mL)
was added and the mixture left at room temperature for 24 h. TLC (ether:hexane=2:1) then revealed a
slower-running compound. The mixture was neutralized and concentrated. The residue was partitioned
in ether–water and the organic phase was separated and the aqueous phase extracted with ether.
Concentration of the combined extracts gave a residue which was chromatographed (ether:hexane=1:1) to

I. Izquierdo et al. / Tetrahedron: Asymmetry 10 (1999) 449–455
455
give 10 (310 mg, 73%) as a colourless oil which showed the same spectroscopic data as those previously
reported;5 tR 6.10 min (A at 210°C); [α]D29 −11 (c 0.5).
4.10. (3R,4R,5S,6R,9R)-9-Ethyl-5-hydroxy-3,4-isopropylidenedioxy-1,7-dioxaspiro[5.5]undecane 11
To a stirred solution of 10 (310 mg, 0.94 mmol) in dry acetone (20 mL) was added anhydrous copper
sulfate (250 mg) and p-toluenesulfonic acid (70 mg), and the mixture was kept at room temperature for
70 h. GLC (A at 210°C) then revealed the presence of 11 by comparison with an authentic sample (tR 3.90
min). The reaction mixture was neutralized (K2CO3) filtered and concentrated. Column chromatography
(ether:hexane=1:2) of the residue gave crystalline 11 (147 mg, 58%) which had the same physical
spectroscopic data as those previously reported.5
Acknowledgements
The authors are deeply grateful to the Ministerio de Educación y Cultura for a grant (J. Tamayo) and
Dr. F. Rebolledo and Amano Pharmaceutical Co. for a gift of PFL.
References
1. Terao, Y.; Murata, M.; Achiwa, K.; Nishio, T.; Akamtsu, M.; Kamimura, M. Tetrahedron Lett. 1988, 29, 5173–5176.
2. Morgan B.; Dodds, D. R.; Zaks, A.; Andrews, D. R.; Klesse R. J. Org. Chem. 1997, 62, 7736–7743.
3. Lynn, D. G.; Phillips, N. J.; Hutton, W. C.; Shabanowitz, J.; Fennell, D. I.; Cole, R. J. J. Am. Chem. Soc. 1982, 104,
7319–7322; Hutton, W. C.; Phillips, N. J.; Graden, D. W.; Lynn, D. G. J. Chem. Soc., Chem. Commun. 1983, 864–866.
4. Schoffers, E.; Golebiowski, A.; Johnson, C. R. Tetrahedron 1996, 52, 3769–3826.
5. Izquierdo, I.; Plaza, M.-T. Carbohydr. Res. 1990, 205, 293–304.
6. Gaucher, A.; Ollivier, J.; Marguerite, J.; Paugam, R.; Salaün, J. Can. J. Chem. 1994, 72, 1312–1327.
7. Gill, J.; Parish, J. H.; Herbert, R. B.; Hughes, M.; Monk, D. F. Tetrahedron: Asymmetry 1997, 8, 2227–2230.
8. Boland, W.; Frössl, C.; Lorentz, M. Synthesis 1991, 1049–1072.
9. Sefkow, M.; Neidlein, A.; Sommerfeld, T.; Sternfeld, F.; Maestro, M. A.; Seebach, D. Liebigs Ann. Chem. 1994, 719–730.