Superacidic cyclization of all-trans-ω-acetoxyfarnesyl benzyl ether

Article abstract of DOI:10.1055/s-2000-6349

The superacidic cyclization of all-trans-ω-acetoxyfarnesyl benzyl ether 2 begins at the internal double bond giving a mixture of diastereomeric monocyclic compounds 5 and 6, prenylated at gem-dimethyl groups.

Full text of DOI:10.1055/s-2000-6349

PAPER
407
Superacidic Cyclization of All-trans-w-Acetoxyfarnesyl Benzyl Ether
Veaceslav Kulci˛ki,a Nicon Ungur,a Pavel F. Vlad,*a Margherita Gavagnin,b Francesco Castelluccio,b Guido Ciminob
a Institutul de Chimie al Academiei de ™iin˛e, str. Academiei 3, MD 2028 Chi∫in„u, Republic of Moldova
Fax +3732-739775; E-mail: vlad@terp.ich.asm.md
b Istituto per la Chimica di Molecole di Interesse Biologico CNR,1 Via Toiano 6, I- 80072, Arco Felice (Na), Italy
Received 15 June 1999; revised 8 September 1999
than two isoprenic residues in their molecules, the cy-
Abstract: The superacidic cyclization of all-trans-w-acetoxyfarne-
clization process could be initiated from the internal dou-
ble bond. In order to check this assumption, the benzyl
syl benzyl ether 2 begins at the internal double bond giving a mix-
ture of diastereomeric monocyclic compounds 5 and 6, prenylated
ether of w-hydroxyfarnesol 1 was synthesized and submit-
ted to superacidic cyclization. Benzylic protection was se-
lected due to its stability under reaction conditions and
ensuring chemical uniformity of the cyclization process
(unlike the acetate group for example). Under standard
experimental conditions, cyclization of compound 1 af-
forded a complex mixture of products, whereas its acetate
2 smoothly reacted with 5 equivalents of fluorosulfonic
acid. The results of this cyclization are reported here.
at gem-dimethyl groups.
Key words: cyclizations, superacids, terpenoids, carbocations
As reported recently,2-6 superacidic low-temperature cy-
clization of aliphatic and partially cyclized terpenoids has
been shown to constitute a general, one step, highly effi-
cient, stereospecific structure- and chemoselective way to
obtain fully cyclized terpenoids. Several types of com-
pounds such as alcohols, hydroxy esters, acids, esters, ox-
ides and lactones have been synthesized according to this
method. However, superacidic cyclization turned out to
be ineffective with epoxyterpenoids having terminal ep-
oxy groups.7,8 In fact, the terminal epoxy groups under-
went isomerization under standard conditions of
superacidic cyclization, giving rise to ketones in the case
of b-terpenoids and, in the case of a-terpenoids, to (E)-al-
lylic alcohols with a w-hydroxy group, and partially to a-
diols and their monofluorosulfonates, whereas carbocy-
clic compounds were not formed.
Compound 2 was prepared starting from farnesol 3 in
three steps (Scheme 1), according to a literature proce-
dure.9 Compound 3 was protected as its benzyl ether 4,
which was oxidized to allylic alcohol 1 with selenium di-
oxide.9 Treatment of 1 with acetic anhydride in pyridine
afforded the target bifunctional compound 2.
Superacidic cyclization of 2 was carried out by treatment
with 5 equivalents of fluorosulfonic acid (-78 °C, 15 min)
affording a 2:1 mixture of compounds 5 and 6 (72%). The
reaction mixture was purified by reverse-phase HPLC
giving compounds 5 and 6.
1H and 13C NMR data (Table) of 5 and 6 were very simi-
lar, indicating that the two reaction products displayed the
This fact led us to the idea that, on interaction of superacid
with terpenoids containing w-hydroxy groups and more
Scheme 1
Synthesis 1999, No. 3, 407–410 ISSN 0039-7881 © Thieme Stuttgart · New York

408
Kulci˛ki, V.
PAPER
same carbon skeleton. In particular, both 1H NMR spectra while a NOE interaction was observed between H-5 and
showed signals attributable to two methyl groups at car- H2-9 in the isomer 5. These results strongly supported a
bon atoms carrying a double bond and two olefinic pro- trans relative stereochemistry at C-4 and C-5 for com-
tons, suggesting that the molecules retained two of the pound 5, whereas 6 was suggested to be the corresponding
three trisubstituted double bonds of the starting compound cis-isomer. According to this assignment, a 2:1 ratio of 5/
2 and, consequently, should be monocyclic. Indeed, the 6 was obtained with predomination of thermodynamically
presence of a singlet methyl signal (H3-8), and of an ABX more favoured trans-isomer. All the proton and carbon
system, assignable to the methylene group (H2-10) linked resonances of 5 and 6 were easily assigned by 2D NMR
to the benzyloxy moiety, further coupled with an angular experiments (1H-1H COSY, HETCOR and HMBC) and
methine (H-5), strongly supported a monocyclic structure are reported in the Table.
for both compounds, bearing a homoacetoxyprenyl chain
The formation of compounds 5 and 6 implies that the cy-
and a methyl at C-4. Being epimers, the two molecules
clization process occurred by protonation of the internal
differed only in the stereochemistry at the chiral centre C-
D6,7-double bond in the starting compound 2. To the best
4. In order to establish the relative configuration for each
of our knowledge, only two reports11,12 describing such
compound, both a series of NOE difference experiments
kind of cyclization have appeared in the literature. This
unusual cyclization could be explained as follows.
and conformational analysis were performed.
First of all, an energy minimization calculation on the two
On treatment with superacid, the w-acetoxy group of 2 is
possible stereoisomers using DMM program,10 revealed
protonated to give the carboxonium-ion 7 (Scheme 2),
that for each molecule the energetically more favoured
which does not allow the protonation of the terminal dou-
conformation exhibited the chain at C-5 in pseudoequato-
ble bond because this would bring about the formation of
rial orientation. Therefore, based on this fixed chiral cen-
an energetically unfavourable 1,3-dicationic system,
tre, the trans-isomer should display an equatorial
whereas the protonation of D6,7-double bond occurs with
homoprenyl chain at C-4, whereas in the cis-isomer the
the formation of the dication 8, transformed into both di-
same homoprenyl moiety should be axially oriented. The
cations 9 and 10. Deprotonation of 9 and 10 led to the final
compounds 5 and 6, respectively. The dication 9 with two
irradiation of the proton H-5 resulted in a diagnostic en-
hancement of the methyl signal at C-4 in the isomer 6,
Table 1H and 13C NMR Spectral Data (CDCl3) of Compounds 5 and 6a
Assignmentsb
5
6
1H NMR
13C NMR
d
mc
1H NMR
13C NMR
d
mc
d, m,c J (Hz)
d, m,c J (Hz)
1
2
3
5.42 m
1.94 m
1.52 dt (13, 8)
1.30 m
–
1.86 m
–
1.72 d (1.5)
0.96 s
1.32 m
1.27 m
3.52 dd (10, 6)
3.43 dd (10, 3)
1.98 m
5.43 m
–
4.44 br s
1.64 s
4.48 s
122.4
22.7
29.9
–
34.4
48.4
133.6
23.2
23.9
38.2
–
d
t
t
–
s
d
s
q
q
t
–
t
–
t
d
s
t
q
t
5.40 m
1.97 m
1.23 m
1.49 dt (13, 8)
–
1.84 m
–
1.74 br s
0.89 s
1.39 ddd (13, 11, 5)
1.26 m
3.51 dd (10, 5)
3.38 dd (10, 3)
2.05 m
5.40 m
–
4.43 br s
1.63 br s
4.46 s
122.1
22.8
30.1
–
34.5
49.1
133.6
23.4
23.0
39.1
–
d
t
t
–
s
d
s
q
q
t
–
t
–
t
d
s
t
q
t
4
5
6
7
8
9
10
70.6
–
70.7
–
11
12
13
14
15
1'
2'
3', 7'
4', 6'
5'
COCH3
COCH3
22.2
130.6
129.3
70.3
13.8
73.0
138.6
127.6
128.2
127.4
21.0
171.0
22.0
131.0
129.3
70.5
13.7
73.1
138.6
127.7
128.4
127.4
21.0
171.1
–
s
d
d
d
q
s
–
s
d
d
d
d
s
7.33 m
7.33 m
7.27 m
2.07 s
7.32 m
7.32 m
7.27 m
2.07 s
–
–
a Assignments were made by 2D NMR experiments [1H-1H cosy, HETCOR, HMBC (J = 10 and 6 Hz)].
b Assignments refer to the numbering of the carbon atoms of the molecule.
c Denotes the multiplicity of the peak.
Synthesis 1999, No. 3, 407–410 ISSN 0039-7881 © Thieme Stuttgart · New York

PAPER
Superacidic Cyclization of All-trans-w-Acetoxyfarnesyl Benzyl Ether
409
Scheme 2
The spectral data (IR, 1H NMR) were identical to those described in
the literature.9
equatorial chains is more stable and consequently com-
pound 5 predominates over its diastereomer 6.
All-trans-w-Acetoxyfarnesyl Benzyl Ether 2
A solution of 1 (650 mg, 1.98 mmol) in anhyd pyridine (5 mL) was
treated with Ac2O (1 mL) and the mixture was kept at r.t. for 2 h.
In summary, the ω-allylic acetoxy group in the farnesol
skeleton deactivates the terminal double bond. The cy-
clization process, starting from the internal double bond, H2O (10 mL) was added carefully to the mixture and the product
was extracted with Et2O (3 × 10 mL). The extract was washed with
10% H2SO4 (2 × 5 mL), H2O (2 × 10 mL), dried (Na2SO4) and con-
centrated. The crude reaction product (727 mg) was chromato-
graphed on a silica gel column (petroleum ether/Et2O, 97:3) to give
675 mg (92%) of 2.
IR (CHCl3): n = 1735, 1455, 1380, 1235, 1065, 1020 cm-1.
1H NMR (300 MHz, CDCl3): d = 1.56 (s, 3 H, CH3-13), 1.60 (s, 3
H, CH3-14), 1.65 (s, 3 H, CH3-15), 2.07 (s, 3 H, OcOCH3), 4.03 (m,
gives rise to monocyclic terpenoids with prenylated gem-
dimethyl groups in the ring. It should be mentioned that
such cyclic compounds have been isolated from natural
sources.13,14 Some of them possess interesting biological
activity, being at the same time hardly accessible by syn-
thesis.
2 H, H2-1), 4.44 (br s, 2 H, H2-12), 4.50 (s, 2 H, CH2Ph), 5.12 (m, 2
H, H-2 and H-6), 5.40 (m, 1 H, H-10), 7.20-7.35 (m, 5 H, Ar-H).
The IR spectra were taken on a Bio-Rad FTS 7 or a Specord 74 IR
1
spectrophotometer. H and 13C NMR spectra were recorded in
CDCl3 on Bruker WM 500, Bruker AM 400 and Bruker WM 300
spectrometers; chemical shifts are reported in ppm, referenced to
CHCl3 as internal standard (d 7.26 for proton and d 77.00 for car-
bon). Mass spectra were recorded on Kratos MSSO and TRIO 2000
instruments, coupled with an INTEL computer. Commercial Merck
silica gel 60 (70-230 mesh ASTM) was used for column chroma-
tography (CC), and Merck precoated Si gel plates were used for
TLC. The chromatograms were sprayed with 0.1% Ce(SO4)2 in 2 N
H2SO4 and heated at 80 °C for 5 min to detect the spots. HPLC pu-
rifications were conducted on a Waters liquid chromatograph
equiped with a UV detector (254 n).
Anal. calc. for C24H34O3: C, 77.80; H, 9.25. Found: C, 77.76; H,
9.22.
Superacidic Cyclization of All-trans-w-acetoxyfarnesyl Benzyl
Ether 2
To a solution of 2 (70 mg, 0.19 mmol) in 2-nitropropane (1.5 mL)
cooled to -78 °C, was added a solution of FSO3H (95 mg, 0.95
mmol) in the same solvent (1 mL) with vigorous stirring. After 15
min of stirring at the same temperature, the mixture was quenched
by adding a 50% excess of Et3N in hexane (1:1, 0.35 mL). H2O (5
mL) was added carefully to the mixture and the product was extract-
ed with Et2O (3 × 5 mL). The extract was washed with 10% H2SO4
(2 × 5 mL), H2O (2 × 10 mL), dried (Na2SO4) and filtered. The sol-
vent was evaporated in vacuo and the crude product (69 mg) was
purified on a silica gel column (1.0 g) (petroleum ether/Et2O, 97:3)
to give 50.2 mg (72%) of a 2:1 mixture of 5 and 6. The mixture of
5 and 6 was separated by HPLC [semipreparative Nova-Pack C-18
column, MeOH/H2O (95:5), flow rate 1.5 mL/min], affording pure
5 (19.5 mg) and 6 (9.2 mg).
All-trans-w-Hydroxyfarnesyl Benzyl Ether 1
To a solution of 49 (2.18 g, 6.99 mmol) in EtOH (4 mL) was added
SeO2 (410 mg, 3.68 mmol) and the solution was refluxed for 2 h.
After removal of the precipitated Se, the solution was concentrated.
Chromatography of the crude product (2.47 g) on silica gel afforded
the starting ether 4 (645 mg, 30%) and the all-trans-w-hydroxyfar-
nesyl benzyl ether 1 (710 mg, 31%):
IR (CHCl3): n = 3400, 2930, 1670, 1450, 1068, 770 cm-1.
Compound 5
IR (CHCl3): n = 1741, 1453, 1378, 1232, 1101, 756 cm-1.
1H NMR (300 MHz, CDCl3): d = 1.59 (s, 9 H, CH3-13, CH3-14 and
CH3-15), 3.83 (m, 2 H, H2-12), 3.90 (d, 7 Hz, 2 H, H2-1), 4.30 (s, 2
H, CH2Ph), 5.10 (m, 2 H, H-2 and H-6), 5.35 (m, 1 H, H-10), 7.20-
7.30 (m, 5 H, Ar-1H).
MS: m/z (%) = 370 (M+, 2), 310 (M+-AcOH, 9), 262 (4), 219 (42),
202 (100), 189 (98), 175 (28), 161 (79), 119 (97), 91(99).
Anal. calc. for C24H34O3: C, 77.80; H, 9.25. Found: C, 77.77; H,
9.21.
Synthesis 1999, No. 3, 407–410 ISSN 0039-7881 © Thieme Stuttgart · New York

410
Kulci˛ki, V.
PAPER
(5) Kulcitki, V.; Ungur, N.; Vlad, P.F. Tetrahedron 1998, 54,
Compound 6
IR (CHCl3): n = 1741, 1452, 1375, 1233, 1096, 741 cm-1.
11925.
(6) Kulcitki, V.; Ungur, N.; Deleanu, C; Vlad, P.F. Izv. Akad.
Nauk, Ser. Khim. 1999, 54, 135.
(7) Ungur, N.D.; Popa, N.P.; Vlad, P.F. Khim. Prir. Soedin. 1993,
691; Chem. Abstr. 1995, 123, 340427.
MS: m/z (%) = 370 (M+, 1), 310 (4), 262 (13), 219 (60), 202 (100),
189 (78), 175 (32), 161 (74), 119 (99), 91 (98).
Anal. calc. for C24H34O3: C, 77.80; H, 9.25. Found: C, 77.83; H,
9.20.
(8) Ungur, N.D.; Popa, N.P.; Kulcitki, V.N.; Vlad, P.F. Khim.
Prir. Soedin 1993, 697; Chem. Abstr. 1995, 123, 34028.
(9) Naruta,Y. J. Org. Chem. 1980, 45, 4097.
(10) Desktop Molecular Modelling, Oxford Electronic Publishing.
(11) Armstrong, R.J.; Harris, F.L.; Weiler, L. Can. J. Chem. 1982,
60, 673.
(12) Polovinka, M.P.; Korchagina, D.V.; Gatilov, Yu.V.;
Bagrianskaya, I.Yu.; Barkhash, V.A.; Shcherbukhin, V.V.;
Zefirov, N.S.; Perutsky, V.B.; Ungur, N.D.; Vlad, P.F. J. Org.
Chem. 1994, 59, 1509.
(13) Nakagawa, M.; Ishihama, M.; Hamamoto, Y., Endo, M.
Abstracts of Papers of the 28th Symposium on The Chemistry
of Natural Products in Japan, Sendai, October 1986, 200;
Chem. Abstr. 1987, 106, 96126.
(14) Hagiwara, H.; Uda, H. J. Chem. Soc., Perkin Trans. 1 1991,
1803.
Acknowledgement
The NMR spectra were recorded at the "Servizio NMR Area di
Ricerca di Napoli". Mass spectra were obtained from the "Servizio
di Spectrometria di Massa del CNR e dell'Universit di Napoli".
We are grateful to Mr. G. Scognamiglio for spectrophotometric
measurements. Mr. R. Turco for graphical work and Mr. S. Zambar-
dino for the NMR measurements. This work was supported partially
by INTAS (grant no. 96-1109).
References
(1) Associated to the National Institute for the Chemistry of
Biological Systems (CNR).
(2) Vlad, P.F. Pure & Appl. Chem. 1995, 65, 1329.
(3) Vlad, P.F.; Ungur, N.D.; Nguen, V.H.; Perutsky, V.B. Russ.
Chem. Bull. 1995, 44, 2390; Chem. Abstr. 1996, 125, 11139.
(4) Vlad, P.F.; Ungur, N.D.; Nguen, V.T. Russ. Chem. Bull. 1995,
44, 2404; Chem. Abstr. 1996, 125, 11140.
Article Identifier:
1437-210X,E;1999,0,03,0407,0410,ftx,en;Z04599SS.pdf
Synthesis 1999, No. 3, 407–410 ISSN 0039-7881 © Thieme Stuttgart · New York