A direct route to conjugated enediynes from dipropargylic sulfones by a modified one-flask Ramberg-Baecklund reaction

Article abstract of DOI:10.1039/b207296n

The reaction of dipropargylic sulfones with dibromodifluoromethane in the presence of alumina-supported KOH in dichloromethane solution results in facile rearrangement affording the corresponding conjugated linear and cyclic enediynes in good yields. This result shows that the direct transformation of a- and a′-hydrogen bearing sulfones assembles enediyne units without resorting to the prior preparation of the a-halo sulfone precursors in a separate step.

Full text of DOI:10.1039/b207296n

View Article Online / Journal Homepage / Table of Contents for this issue
A direct route to conjugated enediynes from dipropargylic sulfones
by a modified one-flask Ramberg–Bäcklund reaction†
Xiaoping Cao,* Yuying Yang and Xiaolong Wang
1
National Laboratory of Applied Organic Chemistry & Department of Chemistry,
College of Chemistry and Chemical Engineering, Lanzhou University, Lanzhou 730000,
P. R. China. E-mail: caoxplzu@163.com; Fax: ϩ86-0931-8912582
Received (in Cambridge, UK) 25th July 2002, Accepted 17th September 2002
First published as an Advance Article on the web 21st October 2002
The reaction of dipropargylic sulfones with dibromodifluoromethane in the presence of alumina-supported KOH
in dichloromethane solution results in facile rearrangement affording the corresponding conjugated linear and
cyclic enediynes in good yields. This result shows that the direct transformation of α- and αЈ-hydrogen bearing
sulfones assembles enediyne units without resorting to the prior preparation of the α-halo sulfone precursors
in a separate step.
contrast to ring closure, the Ramberg–Bäcklund reaction has
Introduction
been put to good use by Nicolaou for the conversion of cyclic
α-chlorodipropargylic sulfones into the corresponding enediyne
arrays.2d We recently reported a new protocol of the Ramberg–
Bäcklund reaction to effect the direct transformation of
α- and αЈ-hydrogen bearing sulfones into alkenes.3 We became
interested in using our procedure to synthesize enediynes 1
(Scheme 2) without resorting to the prior preparation of the
The recent discovery of several potent antitumor antibiotics
containing enediynes or analogues and models of enediynes
has pushed the exploration of conjugated enediynes to the fore-
front of chemical and biological research. Not only a unique
structural feature, but also their most fascinating aspect is the
unprecedented mechanism of the cleavage of DNA that origin-
ates from the chemical reactivity of the enediyne moiety
(Scheme 1).1 The greatly increased demand for structurally
Scheme 1 Bergman cyclization of a typical enediyne moiety.
diverse enediynes has prompted synthetic chemists to seek
novel and more efficient approaches to the enediyne functional-
ity. Many strategies have been developed to construct or modify
enediyne systems.2 Perusal of the numerous reviews on the
chemistry of the enediynes reveals that the field is dominated
by palladium catalyzed cross coupling reactions of terminal
alkynes with vinyl halides, or their equivalents, as a rapid
convergent entry into the hex-3-ene-1,5-diyne moiety. Another
method entails the anchoring of a carbon–carbon triple bond
to each end of a latent olefinic double bond which is sub-
sequently revealed by an eliminative process. The synthesis
of a cyclic enediyne parent system calls for mild conditions
with, at the same time, a strong driving force for the ring closure
reaction. Most often, an acetylide carbonyl addition is
used as the ring closure reaction.2a Danishefsky et al., however,
efficiently obtained highly substituted, 10-membered cyclic
enediynes by two-fold palladium-catalyzed coupling of
iodoacetylene with cis-1,2-bis(trimethylstannyl)ethylene.2b
Jones’ method was based on a sequence of tandem carbenoid
coupling and elimination of two propynylic bromide units.2c In
Scheme 2 Strategy of constructed enediyne.
α-halosulfone precursors in a separate step. In this paper we
extend this methodology to include the synthesis of conjugated
linear and cyclic enediynes as well as full details of their
preparation.
Results and discussion
Scheme 2 outlines the strategy which we pursued in the quest
for stereopure substituted enediyne 1. We wished to prepare it
from dipropargylic sulfone 2. The strategy shown in Scheme 2
was thought to be feasible because of the successful application
of our procedure to the formation of conjugated trienes from
diallylic sulfones, and conjugated tetraenes from allylic dienylic
sulfones.3b,3d Our synthetic method was also applied to the syn-
thesis of natural products containing a triene unit such as
galbanolenes. The procedure reported here started with a diprop-
argylic sulfone 2 with two triple bonds already in place and the
new double bond to be realized by our modified Ramberg–
Bäcklund procedure. This disconnection approach relied on the
† Electronic supplementary information (ESI) available: spectroscopic
data for compound 8, 9o, 10o, 11a–11g, 11o, 12a–12o, 13a–13g and 13o.
For direct electronic access see http://www.rsc.org/suppdata/p1/b2/
b207296n/
DOI: 10.1039/b207296n
J. Chem. Soc., Perkin Trans. 1, 2002, 2485–2489
This journal is © The Royal Society of Chemistry 2002
2485

View Article Online
Scheme 3 Synthesis of enediynes 13a–o via Ramberg–Bäcklund reaction: (a) R1 = n-C5H11, R2 = H; (b) R1 = t-Bu, R2 = H; (c) R1 = Ph, R2 = H;
(d) R1 = n-C5H11, R2 = Ph; (e) R1 = t-Bu, R2 = Ph; (f ) R1 = n-C5H11, R2 = n-C5H11; (g) R1 = Ph, R2 = Ph; (h) R1,R2 = (CH2)4; (i) R1,R2 = (CH2)5;
(j) R1,R2 = (CH2)6; (k) R1,R2 = (CH2)7; (l) R1,R2 = (CH2)8; (m) R1,R2 = (CH2)9; (n) R1,R2 = (CH2)10; (o) R1,R2 = (CH2)11.
synthesis of dipropargylic sulfone 2, which was in turn prepared
from propargylic bromide 4. There are two general approaches
for the assembly of linear or cyclic sulfide 3, from which the
corresponding sulfone 2 could be obtained after oxidation. One
makes use of the coupling reaction between a propargylic
bromide 4 and Na2S, which afforded the symmetrical linear and
cyclic sulfides. The alternative is to couple a propargylic thio-
acetate to a propargylic bromide, thus providing the unsym-
metrical linear sulfides 3. These procedures invariably demanded
the preparation of propargylic alcohol 5, the common inter-
mediate for the requisite bromide 7, 10 and thioacetate 8
(Scheme 3).
chromatography over silica gel. The configurations of the newly
formed double bonds in the unsymmetrical enediynes (13a–13e)
were readily diagnosed by the typical coupling constant of
16.1–16.0 Hz and 10.9–10.0 Hz for the trans- and cis-olefinic
protons, respectively. For the symmetrical enediynes 13f and
13g, the issue of stereochemistry was confirmed by comparison
of their physical and spectroscopic properties to those reported
in the literature.6,7 Furthermore, an X-ray crystallographic
analysis of 13g revealed an (E)-conformation. Similarly
treatment of the 18-membered ring sulfone 12o (n = 9) with
KOH–Al2O3 led to the desired 17-membered ring enediyne 13o
(n = 9) along with the (E)-isomer (2%). Based on the preferen-
tial formation of the (Z)-isomer in all the cyclic enediynes
studied and the fact that they had consistently smaller Rf values
than those of the corresponding (E)-isomer, they were easily
separated by flash column chromatography over silica gel. We
therefore tentatively assigned the faster running isomers the
(E)-geometry (δH = 5.96 ppm), and the slower the (Z)-geometry
(δH = 5.74 ppm). This outcome was similar to the case of linear
enediynes presented previously. Thus, this provides a rapid
route to stereochemically defined linear and cyclic conjugated
enediynes.
We used propargylic alcohols 6 (R1 = H, alkyl or aryl) as
starting materials. These alcohols (R = alkyl or aryl) were either
transformed into the corresponding propargylic bromides 7 by
reaction with CBr4 and Ph3P in nearly quantitative yield, or into
the corresponding thioacetates 8 using the Mitsunobu reaction4
with thioacetic acid in the presence of Ph3P and DEAD (diethyl
azodicarboxylate), respectively. Coupling of propargylic brom-
ide 7 with Na2S in CH3OH furnished the symmetrical linear
sulfide 11 (R1 = R2). Alternatively, in situ cleavage of the acetyl
moiety of 8 with KOH in CH3OH followed by alkylation of the
resulting proparyglic thiol with proparygl bromide, provided
the unsymmetrical linear sulfides 11 (R1 ≠ R2) in good yields
(Table 1). On the other hand, protection of propargylic alcohol
6 (R = H) with DHP (3,4-dihydro-2H-pyran), and conversion
to the corresponding anion via LiNH2 in liquid ammonia,
followed by alkylation of the resulting anion with dibromides
afforded the bis(acetylenic) compounds, which were depro-
tected under methanolic acid conditions leading to diols 9h–o in
good yields (Table 2). The dibromides 10h–o were prepared
from 9h–o by the action of CBr4 and Ph3P. These substrates
were then reacted with Na2S under high dilution conditions to
furnish the cyclic sulfides 11h–o. All of the dipropargylic sul-
fides 11 were oxidized by oxone (2KHSO5ؒKHSO4ؒKSO4)5
in CH2Cl2 to give high yields of the corresponding sulfones 12.
Subjecting the dipropargylic sulfones 12 to our previously
described modified Ramberg–Bäcklund reaction protocol
(CBr2F2, KOH–Al2O3, CH2Cl2),3a they proceeded smoothly to
give a readily separable mixture (approximate 1 : 1) of the linear
(E)- and (Z)-enediynes and cyclic (Z)-enediynes 13. For the
linear enediynes synthesized in this series, the (E)-isomers had
consistently larger Rf values than those of the corresponding
(Z)-isomers and each pair was easily separated by flash column
Conclusion
We have demonstrated the applicability of our protocol of the
Ramberg–Bäcklund reaction in assembling the linear and cyclic
hex-3-ene-1,5-diyne unit from dipropargylic sulfones without
the necessity of preparing the α-halo dipropargylic sulfones
beforehand. The synthesis transformation utilizes readily
available propargyl alcohol and aliphatic bromides as starting
material. The synthetic routes are facile and the reactions can
be performed on molar scales. Direct application of the chem-
istry can be expected both in the design of antitumour agents
and in the preparation of enediyne nanomaterials.
Experimental
Melting points were measured on a Reichert Microscope
apparatus and are uncorrected. IR spectra were recorded on
a Nicolet 205 FT-IR spectrophotometer and reported in
1
wavenumbers (cmϪ1). H and 13C NMR spectra were recorded
on a Bruker Cryospec WM 250 spectrometer or Avance
DRX–200 spectrometer. NMR spectra were recorded in
2486
J. Chem. Soc., Perkin Trans. 1, 2002, 2485–2489

Table 1 Yields (%) of linear sulfides 11, linear sulfones 12 and linear enediynes 13
Yield (%) of (E)-13
δH (J)a
Yield (%) of (Z)-13
δH (J)
Entry
Sulfide 11
Yield (%)
89
Sulfone 12
Yield (%)
>95
Enediyne
a
43
37
6.10, 5.85 (16.1)
5.90, 5.72 (10.9)
b
c
d
e
f
33
82
81
80
75
95
>95
>95
87
33
27
6.07, 5.84 (16.1)
5.90, 5.72 (10.9)
49
45
6.30, 6.04 (16.1)
6.13, 5.88 (10.9)
33
27
6.13, 6.04 (16.0)
5.92, 5.84 (10.0)
87
43
47
6.10, 6.05 (16.0)
5.95, 5.86 (10.7)
90
41
5.88
40
5.73
g
89
40
6.29
30
6.10
a Coupling constants (J) reported in Hz for the newly formed C᎐C bond.
᎐

View Article Online
Table 2 Yields (%) of diols 9, dibromides 10, cyclic sulfides 11, cyclic sulfones 12 and cyclic enediynes 13
Entry
Diol 9
Dibromide 10
Sulfide 11
Sulfone 12
Cyclic enediyne 13
h (n = 2)
i (n = 3)
j (n = 4)
k (n = 5)
l (n = 6)
m (n = 7)
n (n = 8)
o (n = 9)
75
78
88
71
87
67
65
82
85
66
70
80
73
74
76
85
70
67
65
77
66
66
76
61
55
65
70
70
68
60
70
75
50
65
70
73
70
66
71
74
CDCl3, using tetramethylsilane (δH 0.00 ppm) or residual
chloroform (δH 7.26 ppm, δC 77.0 ppm) as internal standard.
Chemical shifts (δ) are reported in ppm and coupling constants
(J) in Hz. Mass spectra (MS) data were obtained on a Finnegan
MAT 95 or VGZAB–HS mass spectrometer. High-resolution
mass spectra (HRMS) were obtained on an APEX II 47e mass
spectrometer. Elemental analyses were carried out by Medac
Ltd, Uxbridge, UK or the Shanghai Insititute of Organic
Chemistry, P.R. China. UV–visible light spectra were recorded
on a Hitachi U-2000 spectrometer. Propargyl alcohol and
aliphatic bromides were purchased from Aldrich Chemical
Company and used without further purification. Oct-2-yn-1-ol,
4,4-dimethylpent-2-yn-1-ol and 3-phenylprop-2-yn-1-ol were
prepared according to literature methods.8–10
in vacuo and the residue was purified by flash chromatography
(hexane–EtOAc = 2 : 1) to give the products 9h–o. The spectral
data (IR, NMR and MS) and mp data obtained for diols 9h–n
were identical with those in the literature.2d Data for heptadeca-
2,15-diyne-1,17-diol (9o) are available as supplementary data.†
Dibromide 10
The individual diol 9 (25 mmol) was added to a solution of
CBr4 (55 mmol, 2.2 equiv.) and Ph3P (55 mmol, 2.2 equiv.) in
C6H6 (50 cm3). After 3 h, the solvent was evaporated in vacuo,
and the residue was purified by flash chromatography (hexane–
CHCl3 = 10 : 1) to give the products 10h–o. The spectral data
(IR, NMR and MS) obtained for dibromides 10h–n were
identical with those in the literature.2d Data for heptadeca-
2,15-diyne-1,17-diyl dibromide (10o) are available as supple-
mentary data.†
Propargylic bromide 7
The individual propargylic alcohol 6 (10 mmol) was added to
a solution of Ph3P (10 mmol) and CBr4 (10 mmol) in C6H6
(10 cm3) at 0 ЊC. After 2 h, the solvent was evaporated in vacuo.
The residue was extracted with hexane and the combined
extracts were washed (saturated NaCl solution), dried
(MgSO4), filtered and concentrated in vacuo to give crude
propargylic bromides 7. The spectral data (NMR and MS)
obtained for 1-bromooct-2-yne and 1-bromo-3-phenylprop-
2-yne were identical with those in the literature.11
General procedure for the preparation of sulfides 11
Method A. For the unsymmetrical linear sulfide, the indi-
vidual propargylic thioacetate 8 (5 mmol) was added to a
solution of KOH (5 mmol) and Na2S2O3ؒ5H2O (10 mg) in
methanol (10 cm3) under nitrogen. The resulting mixture was
stirred at 0 ЊC for 30 min. To this thioate solution was then
added dropwise the individual propargylic bromide 7 (R1 ≠ R2,
5 mmol) over a period of 5 min and stirring was continued for 1 h
at 20 ЊC. The reaction mixture was poured into water (10 cm3)
and the solvent was evaporated in vacuo. The aqueous phase
was extracted with Et2O and the combined extracts were
washed (saturated NaCl solution), dried (MgSO4), filtered
and concentrated in vacuo. Flash chromatography of the
residue over silica gel (hexane–EtOAc = 100 : 1) afforded the
unsymmetrical sulfides 11.
Propargylic thioacetates 8
A solution of DEAD (10 mmol) in dry benzene (10 cm3) was
added to a stirred solution of Ph3P (10 mmol) in C6H6 (50 cm3)
at 0 ЊC. The resulting red solution was kept at 0 ЊC for 15 min
and then a precooled (0 ЊC) mixture of the individual prop-
argylic alcohol 6 (10 mmol) and CH3COSH (10 mmol) in dry
C6H6 (10 cm3) was added in one portion. The mixture was
stirred at rt for 1 h and the solvent was removed in vacuo. Flash
chromatography of the residue over silica gel (hexane–EtOAc =
10 : 1) afforded the following propargylic thioactetes 8: (S)-Oct-
2-ynyl thioacetate, (S)-4,4-dimethylpent-2-ynyl thioacetate and
(S)-3-phenylprop-2-ynyl thioacetate. Data for propargylic
thioactetes 8 are available as supplementary data.†
Method B. For the symmetrical linear sulfides, individual
bromide 7 (10 mmol) was added to a solution of Na2S (5 mmol)
in methanol (10 cm3). The resulting mixture was stirred at 20 ЊC
for 2 h. The reaction mixture was then worked up following the
same procedure as that described in Method A.
Method C. For the cyclic sulfides, solutions of individual
dibromide 10 (50 mmol in EtOH, 100 cm3) and Na2S (55 mmol,
1.1 equiv. in 80% EtOH–H2O, 100 cm3) were added separately
and simultaneously via a funnel to cooled (Ϫ5 ЊC) EtOH (400
cm3) over 4 h under stirring. The resulting mixture was stirred
at 20 ЊC for 4 h. The reaction mixture was then worked up
following the same procedure as that described in Method A.
Linear dipropargylic sulfides 11a–11e were prepared using
method A, 11f and 11g were prepared using method B. Cyclic
sulfides 11h–o were prepared using method C. Data for linear
dipropargylic sulfides 11a–11g and cyclic sulfide 11o are
available as supplementary data.† The spectral data (IR, NMR
and MS) obtained for cyclic sulfides 11h–n were identical with
those in the literature.2d
Diol 9
Lithium amide (120 mmol) was suspended in dry liquid
ammonia (100 cm3) at Ϫ78 ЊC. Tetrahydro-2-(prop-2-ynyloxy)-
2H-pyran (110 mmol, 2.2 equiv.) was added dropwise over
20 min to this suspension with vigorous stirring. This was
followed by addition of the individual dibromide (50 mmol).
The reaction was allowed to warm to rt over night. It was then
added to a solution of saturated NaCl (200 cm3) and EtOAc
(100 cm3). The organic phase was dried (Mg2SO4) and concen-
trated in vacuo. The residue was then dissolved in CH3OH (100
cm3), and p-PTs (pyridinium toluene-p-sulfonate, 5 mmol) was
added. After one day, the reaction mixture was concentrated
2488
J. Chem. Soc., Perkin Trans. 1, 2002, 2485–2489

View Article Online
General procedure for the preparation of sulfones 12
References
A mixture of the individual propargylic sulfide 11 (5 mmol) and
Oxone (25 mmol) in CH2Cl2 (20 cm3) was stirred at 50 ЊC for
2–3 days. The reaction mixture was filtered through a pad of
silica gel and the filtered cake was washed with CH2Cl2–EtOAc
(2 : 1, 50 cm3). The filtrate was concentrated in vacuo. Flash
chromatography of the residue over silica gel (hexane–EtOAc =
10 : 1gradient to 2 : 1) afforded linear and cyclic propargylic
sulfones 12. Data for sulfones 12a–12o are available as
supplementary data.†
1 (a) G. Peiffer, Bull. Soc. Chim. Fr., 1963, 3, 537; (b) T. Böhm-Gössl,
W. Hunsmann, L. Rohrschneider, W. M. Schneider and
W. Ziegenbein, Chem. Ber., 1963, 96, 2504; (c) W. H. Okamura
and F. Sondheimer, J. Am. Chem. Soc., 1967, 89, 5991; (d )
R. R. Jones and R. G. Bergman, J. Am. Chem. Soc., 1972, 94, 660;
(e) K. C. Nicolaou and W. M. Dai, Angew. Chem., Int. Ed. Engl.,
1991, 30, 1387; ( f ) A. G. Myers, Tetrahedron Lett., 1987, 28, 4493.
2 (a) P. A. Wender, S. Beckham and D. L. Mohler, Tetrahedron, Lett.,
1995, 36, 209; (b) M. D. Shair, T. Yoon and S. J. Danishefsky, J. Org.
Chem., 1994, 59, 3752; (c) G. B. Jones, R. S. Huber and J. E. Mathew,
J. Chem. Soc., Chem. Commun., 1995, 18, 1791; (d ) K. C. Nicolaou,
C. Zuccararello, C. Riemer, V. A. Estevez and W. M. Dai, J. Am.
Chem. Soc., 1992, 114, 7360; (e) J. W. Grissom, G. U. Gunawardena,
D. Klingberg and D. H. Huang, Tetrahedron, 1996, 19, 6453; ( f )
K. Burkhard, Angew. Chem., Int. Ed. Engl., 1996, 35, 165; (g)
T. Kaneko, M. Takahashi and M. Hirama, Angew. Chem., Int. Ed.,
1999, 38, 1267; (h) D. S. Rawat and J. M. Zaleski, Chem. Commun.,
2000, 2493.
General procedure for the preparation of linear and cyclic
enediynes 13
A solution of sulfone 12 (2 mmol) in dry CH2Cl2 (10 cm3) was
added to a stirred suspension of KOH–Al2O3 (10 mmol of
KOH) in CBr2F2–CH2Cl2 (1:10, 10 cm3). The mixture was then
stirred for 10 min to 1 h. The reaction mixture was filtered
through a short column of silica gel. The filtrate was concen-
trated in vacuo to give the crude enediyne 13. Flash chrom-
atography of the crude enediyne over silica gel (hexane)
afforded the pure enediynes. Disubstituted enediynes are more
stable and can be stored at Ϫ30 ЊC for days in hexane solution.
The terminal enediynes 13a–13c are less stable and tend to
polymerize even when stored at Ϫ30 ЊC for one day. Hence, we
were unable to obtain elemental analysis or HRMS data of
these compounds. Data for linear enediynes 13a–13g12–14 and
cyclic enediyne 13o are available as supplementary data.† The
spectral data (IR, NMR and MS) for cyclic enediynes 13h–13n
were identical with those in the literature.2d
3 (a) T. L. Chan, S. Fong, Y. Li and T. O. Man, J. Chem. Soc., Chem.
Commun., 1994, 15, 1771; (b) X. P. Cao, T. L. Chan, H. F. Chow and
J. Tu, J. Chem. Soc., Chem. Commun., 1995, 12, 1297; (c) X. P. Cao,
T. L. Chan and H. F. Chow, Tetrahedron Lett., 1996, 37, 1049; (d )
X. P. Cao, Tetrahedron, 2002, 58, 1301.
4 O. Mitsunobu, Synthesis, 1981, 1, 1.
5 B. M. Trost and D. P. Curran, Tetrahedron Lett., 1981, 22, 1287.
6 B. P. Andreini, M. Benetti, A. Carpita and R. Rossi, Tetrahedron,
1987, 43, 4591.
7 (a) M. Noro, T. Masuda, A. S. Ichimura, N. Koga and H. Iwamura,
J. Am. Chem. Soc., 1994, 116, 6179; (b) F. Diederich, D. philp and
P. Seiler, J. Chem. Soc., Chem. Commun., 1994, 2, 205; (c) Y. Hiroto,
T. K. Eijis and N. T. Yoshiki, Orgametallics, 2000, 19, 567.
8 L. Crombie, D. O. Morgan and E. H. Smith, J. Chem. Soc., Perkin
Trans 1, 1991, 3, 567.
9 L. Brandsma, Preparative Acetylenic Chemistry, Elsevier,
Amsterdam, 1988, 81.
10 N. A. Bumagin, A. B. Ponomaryov and I. P. Beletsaya, Synthesis,
1984, 9, 728.
11 (a) M. Hanack, Tetrahedron Lett., 1981, 22, 557; (b) L. Brandsma
and H. D. Verkrnijsse Synthesis of Acetylenes, Allene and
Cumulenes, Elsevier, New York, 1981, 219.
Acknowledgements
The authors gratefully acknowledge the financial support of
the National Natural Science Foundation of China (NSFC, QT
program) and Natural Science Foundation of Gansu Province
(ZS011-A25-003-Z). We would also like to thank Professor
Tze-lock Chan and Professor Hak-Fun Chow, The Chinese
University of Hong Kong, for their helpful guidance.
12 G. Reiner and C. S. Rolf, Polym. Int., 1994, 33, 43.
13 (a) P. J. Lindsell and P. N. Preston, J. Organomet. Chem., 1992, 439,
201; (b) A. J. Jens and M. T. James, Tetrahedron, 1997, 53, 15515.
14 J. D. Miguel, B. D. Vania and M. P. Joseph, Tetrahedron Lett., 2000,
41, 437.
J. Chem. Soc., Perkin Trans. 1, 2002, 2485–2489
2489