Stereochemistry of (3E)-3,7-dimethyl-3-octene-1,2,6,7-tetraol isolated from Passiflora quadrangularis

Article abstract of DOI:10.1016/S0957-4166(99)00461-9

3,7-Dimethyl-3(E)-octene-1,2,6,7-tetraol, a monoterpene recently isolated from Passiflora quadrangularis fruit pulp, has been established to be a 12:42:14:32 mixture of (2R,6R)-, (2R,6S)-, (2S,6R)- and (2S,6S)- stereoisomers, in that order, by HPLC analys

Full text of DOI:10.1016/S0957-4166(99)00461-9

Pergamon
Tetrahedron: Asymmetry 10 (1999) 4313–4319
Stereochemistry of (3E)-3,7-dimethyl-3-octene-1,2,6,7-tetraol
isolated from Passiflora quadrangularis
Coralia Osorio,a Carmenza Duque,a,∗ Takeshi Koami b and Yoshinori Fujimoto b,†
aDepartamento de Química, Universidad Nacional de Colombia, AA 14490, Bogotá, Colombia
bDepartment of Chemistry and Materials Science, Tokyo Institute of Technology, Meguro, Tokyo, 152-8551, Japan
Received 6 September 1999; accepted 6 October 1999
Abstract
3,7-Dimethyl-3(E)-octene-1,2,6,7-tetraol, a monoterpene recently isolated from Passiflora quadrangularis fruit
pulp, has been established to be a 12:42:14:32 mixture of (2R,6R)-, (2R,6S)-, (2S,6R)- and (2S,6S)-stereoisomers,
in that order, by HPLC analysis of the corresponding tri-(R)-MTPA ester in comparison with stereochemically
defined synthetic samples. © 1999 Elsevier Science Ltd. All rights reserved.
1. Introduction
In the course of our studies on flavor chemistry of Colombian fruits, we isolated a polyhydroxylated
monoterpene, 3,7-dimethyl-3(E)-octene-1,2,6,7-tetraol 1.1 The tetraol has two stereogenic centers at the
C-2 and C-6 positions. In our preliminary study this compound was suggested to be an approximately
1:1 diastereoisomeric mixture by 13C NMR analysis.1 However, further stereochemical assessment of
the natural tetraol remained to be established. We have now synthesized authentic samples of four
stereoisomers at the C-2 and C-6 positions of 3,7-dimethyl-3(E)-octene-1,2,6,7-tetraol in the form of tri-
(R)-MTPA esters and succeeded in determining the ratio of the stereoisomers of the natural monoterpene.
∗
†
Corresponding author. E-mail: cduqueb@ciencias.ciencias.unal.edu.co
E-mail: fujimoto@chem.titech.ac.jp
0957-4166/99/$ - see front matter © 1999 Elsevier Science Ltd. All rights reserved.
PII: S0957-4166(99)00461-9

4314
C. Osorio et al. / Tetrahedron: Asymmetry 10 (1999) 4313–4319
2. Results and discussion
Our approach relies on the HPLC profile analysis of the natural tetraol in the form of MTPA esters
in comparison with stereochemically defined synthetic materials. The synthesis of the reference com-
pounds, (2R,6R)-5a and (2R,6S)-5b 3,7-dimethyl-3(E)-octene-1,2,6,7-tetraols, was carried out according
to Scheme 1, starting with the known (2R)-3,7-dimethyl-3(E),6-octadiene-1,2-diol 2a,2 which is readily
obtained from geraniol. The di-(R)-MTPA ester 3a derived from 2a was subjected to Sharpless asym-
metric dihydroxylation3 using AD-mix β to afford (2R,6R)-tetraol diester 4a in a moderate yield. The
(2R,6S)-isomer 4b was prepared by treating 3a with AD-mix α in place of AD-mix β. The tetraol diesters
4a and 4b were converted into the tri-(R)-MTPA esters 5a and 5b, respectively.
Scheme 1. Synthesis of the stereochemically defined tetraols 5a–d. Reagents: (i) (S)-MTPACl, py, (ii) AD-mix β, (iii) AD-mix
α, (iv) LiAlH4. Group M refers to (R)-MTPA

C. Osorio et al. / Tetrahedron: Asymmetry 10 (1999) 4313–4319
4315
Table 1
13C NMR data for (2S,6R)-1a and (2S,6S)-1b (100 MHz in CD3OD)
The other two isomers, (2S,6R)- and (2S,6S)-tetraol tri-(R)-MTPA esters 5c and 5d, were synthesized in
the same manner as described above starting with (2S)-diol 2b.2 Sharpless asymmetric dihydroxylation
of the di-(R)-MTPA ester 3b afforded (2S,6R)-tetraol diester 4c and (2S,6S)-isomer 4d depending on
the reagents, AD-mix β and AD-mix α, respectively. Each isomer was converted into its respective
tri-(R)-MTPA esters 5c and 5d. A part of the esters 4c and 4d was reduced with LiAlH4 to give the
diastereoisomeric tetraols, (2S,6R)-1a and (2S,6S)-1b.
The 1H NMR spectra of the tetraols 1a and 1b were hardly distinguishable from each other. The 13
C
NMR data of 1a and 1b are summarized in Table 1. Differences in the chemical shifts of most carbons
were within the experimental error, although C-9 signal showed the maximum difference, ∆δC 0.06. The
13C NMR data of 1a and 1b confirmed that the natural tetraol is a mixture of the two diastereoisomers.
1
1
In contrast to the very close similarity of the H and 13C NMR spectra of 1a and 1b, the H NMR
spectra (Table 2) of the tri-(R)-MTPA ester derivatives 5a–d were different from each other, and useful
for the identification of the stereoisomers. Taking into consideration the notion of the advanced Moshers
method,4 one may notice the following characteristic H chemical shifts among the isomers. The signals
1
for 8-H3 and 10-H3 of (6R)-isomers 5a and 5c appeared downfield from those of (6S)-isomers 5b and
5d. This proved that Sharpless asymmetric dihydroxylation proceeded with the expected stereochemical
course. The signals for H-5a of (2R,6R)-isomer 5a appeared most upfield, while that of (2S,6S)-isomer
5d resonated the most downfield. This can be reasonably explained by the combined effect of the chiral
(R)-MTPA groups attached at the C-2 and C-6 positions. The signals of the 1-methylene protons of
2S-isomers are expected to be more upfield than those of (2R)-isomers provided that the effect of the 6-
O-(R)-MTPA group is not significant. However, the observed chemical shifts for the methylene protons
at C-1 were not systematic. The interaction of the vicinal O-MTPA groups at C-1 and C-2 positions may
be the reason for this anomalous behavior.
In addition, a careful inspection of the 1H NMR spectra of 5a–d, in particular methyl and methoxy sig-
nals, revealed that (2R,6R)-5a and (2R,6S)-5b were accompanied by ca. 10% of 5c and 5d, respectively,
which are apparently due to the presence of the antipode in starting material 2a. Similarly, compounds
5c and 5d were found to be accompanied by ca. 10% of 5a and 5b, respectively. Thus, the Sharpless
asymmetric dihydroxylation of 3a and 3b was found to be highly stereoselective.
HPLC trace of the tri-(R)-MTPA ester of the natural tetraol isolated from P. quadrangularis is

4316
C. Osorio et al. / Tetrahedron: Asymmetry 10 (1999) 4313–4319
Table 2
1H NMR data for tri-(R)-MTPA esters 5a–d (400 MHz in CDCl3)
Figure 1. HPLC trace of tri-(R)-MTPA ester of the natural tetraol 1. Conditions: solvent MeOH:H2O 6:1, flow rate 0.7 ml/min,
UV detection at 254 nm
illustrated in Fig. 1, which shows three peaks in a 32:26:42 ratio in the order of elution. The authentic tri-
(R)-MTPA esters 5a–d were eluted at 31.82 min for 5d, 34.55 min for 5a and 5c, and 35.88 min for 5b.
Unfortunately, the two isomers 5a and 5c were eluted at the same retention time. To obtain an accurate
ratio of the overlapped (2R,6R)- and (2S,6R)-isomers,5 the second peak in the HPLC was separated and
analyzed by 1H NMR. Based on the intensities of the diagnostic signals of the methyl (δ 1.438 vs 1.618)
and methoxy (δ 3.511 vs 3.453) groups (Table 2), this fraction was revealed to be composed of (2R,6R)-
and (2S,6R)-isomers in a 6:7 ratio. Thus, it has been established that natural tetraol isolated from P.
quadrangularis is a 12:42:14:32 mixture of 5a, 5b, 5c and 5d.
It is interesting to note that the other monoterpene, 3,7-dimethyl-1,3(E)-octadiene-6,7-diol (2,6-
dimethyl-5(E),7-octadiene-2,3-diol), which was isolated together with the tetraol 1 in the same fruit
pulp, is a 4:1 mixture of (6S)- and (6R)-antipodes.1 The ratio of 6S:6R of 1 was 74:26 (=3:1). The fact

C. Osorio et al. / Tetrahedron: Asymmetry 10 (1999) 4313–4319
4317
that (6S)-antipodes are predominant in the two compounds could support the biosynthetic correlation
of the two oxygenated monoterpenes. The closely related olefinic isomer, 3,7-dimethyl-3(9)-octene-
1,2,6,7-tetraol, was also isolated from the fruit of Cnidium monnieri (Umbelliferae)6 and Foeniculum
vulgare (Umbelliferae)7 as a diastereoisomeric mixture whose stereochemical investigation has not yet
been reported.
3. Experimental
3.1. General
1H NMR spectra were obtained on a JEOL JNM LA-400 (400 MHz) spectrometer in CDCl3 solutions
or in CD3OD solutions with tetramethylsilane as an internal reference. 13C NMR spectra (100 MHz)
were recorded on the same spectrometer in CD3OD solutions and chemical shifts are referenced to the
solvent signal (δ 49.0). HPLC was performed on a Shimadzu LC-6A with SPD-6A UV detector equipped
with a Shimadzu Shim-Pack CLC-ODS column (15 cm×6 mm i.d.).
3.2. (2R)-3,7-Dimethyl-3(E),6-octadiene-1,2-diol di-(R)-MTPA ester 3a
(S)-MTPACl (445 µl, 2.38 mmol) was added to a solution of the diol 2a (163 mg, 0.959 mmol) in
pyridine (0.30 ml) at 0°C and the mixture was stirred for 30 min at room temperature. Ice chips were
added and the mixture was extracted with ether. The organic layer was washed with 2N HCl, satd aq.
NaHCO3 and brine, dried over Na2SO4, and concentrated. The crude product was chromatographed on
silica gel (hexane:AcOEt, 10:1) to give 3a (376 mg, 65%) as a colorless oil. 1H NMR δ: 1.55, 1.59, 1.68
(3H each, s, 9-H3, 8-H3, and 10-H3), 2.65 (2H, m, 5-H2), 3.36, 3.49 (3H each, s, OMe), 4.33 (1H, dd,
J=12.1, 8.6 Hz, 1-Ha), 4.57 (1H, dd, J=12.1, 3.1 Hz, 1-Hb), 4.97 (1H, m, 4 or 6-H), 5.44 (1H, t, J=7.2 Hz,
6 or 4-H), 5.53 (1H, dd, J=8.6, 3.1 Hz, 2-H), 7.32–7.47 (10H, m, Ar). Anal. calcd for C30H32F6O6: C,
59.80; H, 5.35. Found: C, 59.85; H, 5.47. The e.e. of 3a was determined to be 76% by 1H NMR analysis.
3.3. (2S)-3,7-Dimethyl-3(E),6-octadiene-1,2-diol di-(R)-MTPA ester 3b
Compound 2b (425 mg, 2.50 mmol) was converted to 3b (898 mg, 60%) as described for 2a.
1
Compound 3b: colorless oil. H NMR δ: 1.60, 1.67, 1.68 (3H each, s, 9-H3, 8-H3, and 10-H3), 2.70
(2H, t, J=7.0 Hz, 5-H2), 3.40, 3.46 (3H each, s, OMe), 4.30 (1H, dd, J=12.0, 7.4 Hz, 1-Ha), 4.54 (1H, dd,
J=12.0, 4.1 Hz, 1-Hb), 5.00 (1H, t, J=7.0 Hz, 4 or 6-H), 5.59 (1H, t, J=7.0 Hz, 6 or 4-H), 5.66 (1H, dd,
J=7.4, 4.1 Hz, 2-H), 7.33–7.45 (10H, m, Ar). Anal. calcd for C30H32F6O6: C, 59.80; H, 5.35. Found: C,
59.67; H, 5.29. The e.e. of 3b was determined to be 74% by 1H NMR analysis.
3.4. (2R,6R)-3,7-Dimethyl-3(E)-octene-1,2,6,7-tetraol 1,2-di-(R)-MTPA ester 4a
AD-mix β (326 mg) was added to a mixture of 3a (140 mg, 0.233 mmol), t-BuOH (1.5 ml) and
H2O (1.5 ml), and the mixture was stirred at room temperature for two days. K2OsO2(OH)4 (0.2 mg)
was added and stirring was continued for another day. Satd aq. Na2S2O3 was added and the mixture
was extracted with AcOEt repeatedly. The organic layer was washed with 2N HCl, satd aq. NaHCO3
and brine, dried over Na2SO4, and concentrated. The crude product was chromatographed on silica gel
1
(hexane:AcOEt, 1:1) to give 4a (48 mg, 32%) as white solid. H NMR δ: 1.17, 1.22 (3H each, s, 8-H3,

4318
C. Osorio et al. / Tetrahedron: Asymmetry 10 (1999) 4313–4319
10-H3), 1.57 (3H, s, 9-H3), 2.14 (2H, m, 5-H2), 3.31 (1H, dd, J=10.0, 2.8 Hz, 6-H), 3.38, 3.48 (3H each,
s, OMe), 4.36 (1H, dd, J=12.0, 8.4 Hz, 1-Ha), 4.58 (1H, dd, J=12.0, 3.0 Hz, 1-Hb), 5.54 (1H, dd, J=8.4,
3.0 Hz, 2-H), 5.59 (1H, t, J=7.2 Hz, 4-H), 7.34–7.49 (10H, m, Ar). Anal. calcd for C30H34F6O8: C, 56.60;
H, 5.38. Found: C, 56.48; H, 5.46.
3.5. (2R,6S)-3,7-Dimethyl-3(E)-octene-1,2,6,7-tetraol 1,2-di-(R)-MTPA ester 4b
AD-mix α (333 mg) was added to the mixture of 3a (140 mg, 0.233 mmol), t-BuOH (1.5 ml), and H2O
(1.5 ml), and the mixture was stirred at room temperature for two days. Satd aq. Na2S2O3 was added and
the mixture was extracted with AcOEt repeatedly. The organic layer was washed with 2N HCl, satd aq.
NaHCO3 and brine, dried over Na2SO4, and concentrated. The crude product was chromatographed on
silica gel (hexane:AcOEt, 1:1) to give 4b (91 mg, 61%) as white solid. 1H NMR δ: 1.16, 1.21 (3H each,
s, 8-H3, 10-H3), 1.56 (3H, s, 9-H3), 2.11 (2H, m, 5-H2), 3.25 (1H, dd, J=8.2, 4.5 Hz, 6-H), 3.38, 3.49
(3H each, s, OMe), 4.34 (1H, dd, J=12.0, 8.6 Hz, 1-Ha), 4.60 (1H, dd, J=12.0, 3.2 Hz, 1-Hb), 5.55 (2H,
m, 2-H, 4-H), 7.33–7.49 (10H, m, Ar). Anal. calcd for C30H34F6O8: C, 56.60; H, 5.38. Found: C, 56.34;
H, 5.47.
3.6. (2S,6R)-3,7-Dimethyl-3(E)-octene-1,2,6,7-tetraol 1,2-di-(R)-MTPA ester 4c
Compound 3b (140 mg, 0.233 mmol) was treated with AD-mix β [576 mg; K2OsO2(OH)4 (0.3 mg)
was added after two days] as described for the conversion of 3a to 4a to give 4c (69 mg, 46%) as white
solid. 1H NMR δ: 1.17, 1.22 (3H each, s, 8-H3, 10-H3), 1.66 (3H each, s, 9-H3), 2.17 (2H, t, J=6.7 Hz,
5-H2), 3.35 (1H, t, J=6.7 Hz, H-6), 3.40, 3.44 (3H each, s, OMe), 4.35 (1H, dd, J=12.0, 7.2 Hz, 1-Ha),
4.55 (1H, dd, J=12.0, 4.0 Hz, 1-Hb), 5.65 (1H, dd, J=7.2, 4.0 Hz, 2-H), 5.73 (1H, brt, J=6.7 Hz, 4-H),
7.35–7.47 (10H, m, Ar). Anal. calcd for C30H34F6O8: C, 56.60; H, 5.38. Found: C, 56.67; H, 5.51.
3.7. (2S,6S)-3,7-Dimethyl-3(E)-octene-1,2,6,7-tetraol 1,2-di-(R)-MTPA ester 4d
Compound 3b (200 mg, 0.332 mmol) was treated with AD-mix α [465 mg; K2OsO2(OH)4 (0.3 mg)
was added after 24 h] as described for the conversion of 3a to 4b to give 4d (159 mg, 75%) as white
1
solid. H NMR δ: 1.17, 1.21 (3H each, s, 8-H3, 10-H3), 1.67 (3H, s, 9-H3), 2.14 (2H, m, 5-H2), 3.29
(1H, dd, J=9.6, 2.8 Hz, 6-H), 3.41, 3.45 (3H each, s, OMe), 4.37 (1H, dd, J=11.9, 7.0 Hz, 1-Ha), 4.53
(1H, dd, J=11.9, 4.1 Hz, 1-Hb), 5.64 (1H, dd, J=7.0, 4.1 Hz, 2-H), 5.69 (1H, t, J=7.2 Hz, 4-H), 7.33–7.47
(10H, m, Ar). Anal. calcd for C30H34F6O8: C, 56.60; H, 5.38. Found: C, 56.64; H, 5.45.
3.8. (2R,6R)-, (2R,6S)-, (2S,6R)- and (2S,6S)-3,7-Dimethyl-3(E)-octene-1,2,6,7-tetraol 1,2,6-tri-(R)-
MTPA esters 5a, 5b, 5c and 5d
(S)-MTPACl (6 µl, 32 mmol) was added to a solution of 4a (5 mg, 7.9 µmol) in pyridine (30 µl) and
the mixture was stirred for 30 min. Ether (200 µl) was added to the mixture and the resulting solution
was directly subjected to p-TLC (hexane:AcOEt, 2:1). The band having Rf=0.6 gave 5a (6 mg, 90%) as
1
a colorless oil. H NMR δ: 7.36–7.57 (15H, m, Ar) and the data for the other protons are summarized
in Table 2. HRFABMS 835.2551 [M+H−H2O]+. C40H40F9O9 requires 835.2529. The other (R)-MTPA
esters 5b, 5c and 5d were prepared in the same manner. The 1H NMR data for these compounds are listed
in Table 2. HRFABMS spectra of these samples displayed [M+H−H2O]+ ion within the experimental
error.

C. Osorio et al. / Tetrahedron: Asymmetry 10 (1999) 4313–4319
4319
3.9. (2S,6R)-3,7-Dimethyl-3(E)-octene-1,2,6,7-tetraol 1a
LiAlH4 (6.0 mg, 0.158 mmol) was added to a solution of 4c (18 mg, 28.3 µmol) in dry THF (2 ml)
under N2, and the mixture was stirred at room temperature for 2 h. A small amount of water was added
and the whole mixture was transferred onto a silica gel column using CHCl3:MeOH (10:1). Elution of
the column with CHCl3:MeOH (4:1) furnished 1a (3.6 mg, 62%) as a colorless oil. 1H NMR (in CD3OD)
δ: 1.15, 1.18 (3H each, s, 8-H3, 10-H3), 1.63 (3H, s, 9-H3), 2.07 (1H, m, 5-Ha), 2.38 (1H, m, 5-Hb),
3.30 (1H, m, 6-H), 3.48 (1H, dd, J=11.2, 6.2 Hz, 1-Ha), 3.54 (1H, dd, J=11.2, 5.2 Hz, 1-Hb), 4.03 (1H,
dd, J=6.2, 5.2 Hz, 2-H), 5.60 (1H, t, J=7.2 Hz, 4-H). The 13C NMR data are listed in Table 1.
3.10. (2S,6S)-3,7-Dimethyl-3(E)-octene-1,2,6,7-tetraol 1b
Compound 4d (40 mg) was reduced with LiAlH4 as described for 4c to give 1b (7.6 mg) as a colorless
1
oil. H NMR (in CD3OD) δ: 1.15, 1.18 (3H each, s, 8-H3, 10-H3), 1.63 (3H, s, 9-H3), 2.07 (1H, m,
5-Ha), 2.38 (1H, m, 5-Hb), 3.30 (1H, m, 6-H), 3.48 (1H, dd, J=11.2, 7.2 Hz, 1-Ha), 3.54 (1H, dd, J=11.2,
5.2 Hz, 1-Hb), 4.02 (1H, dd, J=7.2, 5.2 Hz, 2-H), 5.60 (1H, t, J=6.8 Hz, 4-H). The 13C NMR data are
listed in Table 1.
Acknowledgements
Colciencias and IPICS (Uppsala University, Sweden) grants are greatly acknowledged.
References
1. Osorio, C.; Duque, C.; Fujimoto, Y. Phytochemistry, in press.
2. Murata, S.; Suzuki, M.; Noyori, R. Bull. Chem. Soc. Jpn. 1982, 55, 247–254; Yasuda, M.; Ide, M.; Matsumoto, Y.;
Nakata, M. Synlett 1997, 899–902. Compound 2a was prepared from TBDMS ether of geraniol (2R,3S)-oxide by a slight
modification of Noyori’s method (DBU was not added). This modification afforded 1-TBDMS-2-TMS ether of 2a, which
after desilylation gave 2a (2R:2S, 88:12). The antipode 2b (2S:2R, 87:13) was prepared from TBDMS ether of geraniol
(2S,3R)-oxide in the same manner.
3. Sharpless, K. B.; Amberg, W.; Bennani, Y. L.; Crispino, G. A.; Hartung, J.; Jeong, K.-S.; Kwong, H.-L.; Moriyama, K.;
Wang, Z.-M.; Xu, D.; Zhang, X.-L. J. Org. Chem. 1992, 57, 2768–2771.
4. Ohtani, I.; Kusumi, T.; Kashman, Y.; Kakisawa, H. J. Am. Chem. Soc. 1991, 113, 4092–4096.
5. The 1H NMR spectrum of the crude tri-(R)-MTPA ester of the natural tetraol sample showed the presence of the two major
constituents 5b and 5d, accompanied by the two minor ones 5a and 5c.
6. Kitajima, J.; Tanaka, Y. Chem. Pharm. Bull. 1993, 41, 1667–1669.
7. Ishikawa, T.; Tanaka, Y.; Kitajima, J. Chem. Pharm. Bull. 1998, 46, 1748–1751.