Synthesis and metal ion binding studies of enediyne-containing crown ethers

Article abstract of DOI:10.1021/jo961583r

The 3-ene-1,5-diyne crown ether 5 is a novel enediyne-containing crown ether that was designed as a model system for a class of enediynes that might undergo alkali metal ion-triggered Bergman cyclization. We report the preparation of 5 by two different routes. In the shorter and preferable route, a carbenoid coupling reaction is employed to simultaneously construct the enediyne moiety and effect a macrocyclization of an acyclic bis(propargyl)bromide 15 to the 24-membered crown ether 5. Under standard reaction conditions, this carbenoid coupling produces as the major product the isomeric 5-ene-1,3-diyne-crown ethers (Z)-16 and (E)-16. The formation of 5-ene-1,3-diynes from the carbenoid coupling of propargyl bromides is unprecedented. We present evidence that it is the polyether nature of dibromide 15 that leads to the formation of the 5-ene-1,3-diyne-crown ether products. Judicious control of the reaction conditions can be used to produce either 5 or (Z)-16 from 15 in synthetically useful yields. Both enediyne-crown ethers 5 and (Z)-16 bind alkali metal ions, as evidenced by their ability to extract alkali metal picrates into organic solvents. Enediyne-crown ether 5 undergoes Bergman cyclization at 135°C in DMSO/1,4-cyclohexadiene to produce the known o-xylyl crown ether 4. Crown ether 5 represents an enediyne in which molecular recognition of alkali metals might serve as a trigger for Bergman cyclization.

Full text of DOI:10.1021/jo961583r

J . Org. Chem. 1996, 61, 9385-9393
9385
Syn th esis a n d Meta l Ion Bin d in g Stu d ies of En ed iyn e-Con ta in in g
Cr ow n Eth er s
Mark M. McPhee and Sean M. Kerwin*
Division of Medicinal Chemistry, College of Pharmacy, The University of Texas, Austin, Texas 78712
Received August 14, 1996X
The 3-ene-1,5-diyne crown ether 5 is a novel enediyne-containing crown ether that was designed
as a model system for a class of enediynes that might undergo alkali metal ion-triggered Bergman
cyclization. We report the preparation of 5 by two different routes. In the shorter and preferable
route, a carbenoid coupling reaction is employed to simultaneously construct the enediyne moiety
and effect a macrocyclization of an acyclic bis(propargyl)bromide 15 to the 24-membered crown
ether 5. Under standard reaction conditions, this carbenoid coupling produces as the major product
the isomeric 5-ene-1,3-diyne-crown ethers (Z)-16 and (E)-16. The formation of 5-ene-1,3-diynes
from the carbenoid coupling of propargyl bromides is unprecedented. We present evidence that it
is the polyether nature of dibromide 15 that leads to the formation of the 5-ene-1,3-diyne-crown
ether products. J udicious control of the reaction conditions can be used to produce either 5 or
(Z)-16 from 15 in synthetically useful yields. Both enediyne-crown ethers 5 and (Z)-16 bind alkali
metal ions, as evidenced by their ability to extract alkali metal picrates into organic solvents.
Enediyne-crown ether 5 undergoes Bergman cyclization at 135 °C in DMSO/1,4-cyclohexadiene to
produce the known o-xylyl crown ether 4. Crown ether 5 represents an enediyne in which molecular
recognition of alkali metals might serve as a trigger for Bergman cyclization.
Many chemists continue to be fascinated by the ene-
1
diyne natural products, exemplified by calicheamicin
and dynemicin.2 These antitumor antibiotics are both
challenging targets for synthesis and extremely potent
3
cytotoxic agents that possess a unique mechanism of
action.4 The potent cytotoxicity of the enediyne antibiot-
ics is a result of their ability to undergo a triggered
Bergman cyclization to produce DNA-cleaving diradical
species. The challenge of harnessing the potent cytotox-
icity of these compounds for use in cancer chemotherapy
hinges upon the ability to design more selective agents.5
Toward this end, a number of groups have reported on
novel designed enediynes with various means of trigger-
6
ing the Bergman cyclization. Most recently, novel
enediynes in which molecular recognition is coupled with
a triggering mechanism for the Bergman cyclization have
F igu r e 1.
been sought. K o¨ nig and R u¨ tters7 have designed and
synthesized the bis(crown ether) enediyne 1 (Figure 1).
Although 1 binds both sodium and potassium, neither
alkali metal complex of 1 undergoes Bergman cyclization
at temperatures lower than that required for cyclization
X
Abstract published in Advance ACS Abstracts, December 15, 1996.
1) (a) Lee, M. D.; Dunne, T. S.; Siegel, M. M.; Chang, C. C.; Morton,
(
G. O.; Borders, D. B. J . Am. Chem. Soc. 1987, 109, 3464-3466. (b)
Lee, M. D.; Dunne, T. S.; Chang, C. C.; Ellestad, G. A.; Siegel, M. M.;
Morton, G. O.; McGahren, W. J .; Borders, D. B. J . Am. Chem. Soc.
987, 109, 3466-3468. (c) Lee, M. D.; Dunne, T. S.; Chang, C. C.;
Siegel, M. M.; Morton, G. O.; Ellestad, G. A.; McGahren, W. J .; Borders,
8
of the free ligand. Buchwald and co-workers have
1
synthesized 1,2-bis[(diphenylphosphino)ethynyl]benzene
(2) and have shown that addition of palladium(II) chlo-
ride or platinum(II) chloride to 2 results in an over
D. B. J . Am. Chem. Soc. 1992, 114, 985-997.
(2) (a) Konishi, M.; Ohkuma, H.; Matsumoto, K.; Tsuno, T.; Kamei,
H.; Miyaki, T.; Oki, T.; Kawaguchi, H.; VanDuyne, G. D.; Clardy, J . J .
Antibiot. 1989, 42, 1449-1452. (b) Konishi, M.; Ohkuma, H.; Tsuno,
T.; Oki, T.; VanDuyne, G. D.; Clardy, J . J . Am. Chem. Soc. 1990, 112,
3
0 000-fold increase in the Bergman cyclization rate
compared to 2 alone. In contrast, addition of mercury(II)
chloride to 2 results in a slower rate of Bergman cycliza-
3
715-3716.
(3) (a) Shair, M. D.; Yoon, T. Y.; Danishefsky, S. J . Angew. Chem.,
Int. Ed. Engl. 1995, 34, 1721-1723. (b) Hitchcock, S. A.; Chu-Moyer,
M. Y.; Boyer, S. H.; Olson, S. H.; Danishefsky, S. J . J . Am. Chem. Soc.
(6) For examples of triggering mechanisms see: Basak, A.; Khamrai,
U. K. Tetrahedron Lett. 1996, 37, 2475-2478. Nuss, J . M.; Murphy,
M. M. Tetrahedron Lett. 1994, 35, 37-40. Nicolaou, K. C.; Liu, A.; Zeng,
Z.; McComb, S. J . Am. Chem. Soc. 1992, 114, 9279-9282. Nicolaou,
K. C. Maligres, P.; Suzuki, T.; Wendeborn, S. V.; Dai, W.-M.; Chadha,
R. K. J . Am. Chem. Soc. 1992, 114, 8890-8907. Nicolaou, K. C.; Dai,
W.-M. J . Am. Chem. Soc. 1992, 114, 8909-8912. Maier, M.; Brand-
stetter, T. Tetrahedron Lett. 1991, 32, 3679-3682.
1
995, 117, 5750-5756. (c) Nicolaou, K. C.; Hummel, C. W.; Nakada,
M.; Shibayama, K.; Pitsinos, E. N.; Saimoto, H.; Mizuno, Y.; Baldenius,
K.-U.; Smith, A. L. J . Am. Chem. Soc. 1993, 115, 7625-7635. (d)
Nicolaou, K. C.; Hummel, C. W.; Pitsinos, E. N.; Nakada, M.; Smith,
A. L.; Shibayama, K.; Saimoto, H. J . Am. Chem. Soc. 1992, 114, 10082-
1
0084.
4) Nicolaou, K. C.; Smith, A. L.; Yue, E. W. Proc. Natl. Acad. Sci.
U.S.A. 1993, 50, 5881-5888.
5) Nicolaou, K. C.; Dai, W.-M.; Tsay, S.-C.; Estevez, V. A.; Wrasidlo,
W. Science 1992, 256, 1172-1178.
(
(7) K o¨ nig, B.; R u¨ tters, H. Tetrahedron Lett. 1994, 21, 3501-3504.
(8) Warner, B. P.; Millar, S. P.; Broene, R. D.; Buchwald, S. L.
Science 1995, 269, 814-816.
(
S0022-3263(96)01583-6 CCC: $12.00 © 1996 American Chemical Society

9
386 J . Org. Chem., Vol. 61, No. 26, 1996
McPhee and Kerwin
thetic challenge. Although a number of cyclic enediynes
have been reported, there have not been any reported
syntheses of larger than 16-membered ring cyclic ene-
diynes.16 In order to construct crown ether 5, various
routes in which the macrocyclization and enediyne form-
ing steps are either sequential (in either order) or
concurrent were explored. Two routes were found to be
productive in obtaining 5. In the first, sequential route
a Ramberg-B a¨ cklund reaction is employed to install the
enediyne moiety within a macrocyclic precursor. In an
alternative, and more efficient, route to 5 the enediyne
is formed concurrently with macrocyclization through the
recently reported carbenoid coupling strategy of J ones
and co-workers.17 In this latter case, we have found an
unusual product formed in the carbenoid coupling step
and a means to divert the reaction towards the desired
enediyne product.
F igu r e 2.
9
tion. Most recently, K o¨ nig and co-workers have reported
the synthesis of the bipyridyl enediyne 3, which, as the
mercury(II) complex, undergoes Bergman cyclization at
a temperature 100 °C lower than that required for the
cyclization of 3 itself.
We have been interested in DNA-reactive agents that
1
0
are alkali metal ion specific. Alkali metal ion concen-
trations within cells are regulated during the cell cycle
and may be significantly altered in cancer cells versus
normal cells.11 DNA-reactive compounds that are re-
sponsive to these differing alkali metal ion concentrations
could serve as cell cycle specific or cancer cell specific
cytotoxic agents. In addition, there are numerous reports
of DNA-interactive agents that incorporate an alkali
metal ion recognition motif, such as a crown ether
moiety.12 These compounds in general have an increased
affinity for DNA when compared to the analogous com-
pounds lacking the alkali metal ion binding element.
Thus, incorporation of an alkali metal ion recognition
element in an enediyne may result in more potent, as
well as more selective, DNA cleavage agents.
Resu lts a n d Discu ssion
Sequ en tia l En ed iyn e F or m a tion -Ma cr ocycliza -
tion a n d Ma cr ocycliza tion -En ed iyn e F or m a tion
Rou tes. K o¨ nig and co-workers have recently employed
a sequential enediyne formation-cyclization strategy to
construct the bipyridyl enediyne 3 as well as a number
of other enediyne macrocycles.9 In their route, K o¨ nig and
co-workers employ (Z)-1,8-dibromooct-4-ene-2,6-diyne (6)
and an appropriate diphenol, dicarboxylate, or diamine
partner (eq 1). Significantly, these authors note that
more basic nucleophiles fail in the cyclization reaction
due to propargyl-allene tautomerization of the dibromide.
Here we report our design and synthesis of an enediyne
in which molecular recognition of alkali metals might
serve as a trigger for Bergman cyclization. In contrast
to the bis(crown ether) enediyne 1 of K o¨ nig and co-
workers,7 a crown ether ring that incorporates the
enediyne chromophore may more effectively couple the
molecular recognition by the crown to the induction of
strain in the enediyne moiety sufficient to facilitate
Bergman cyclization. This rationale led to a search for
aromatic crown ethers that could formally be derived
from a Bergman cyclization of a precursor enediyne.
Reinhoudt and co-workers have reported a series of
o-xylyl crown ethers, exemplified by compound 4 (Figure
We have attempted a synthesis of 5 that is related to
the above studies of K o¨ nig and co-workers. In this
approach, (Z)-oct-4-ene-2,6-diyne-1,8-diol (7) is allowed
to react with various electrophiles, such as the ditosylate
of pentaethylene glycol (eq 2). However, these reactions
2
M
), that bind alkali metals with high affinity (K
a
> 106
). Crown ether 4 can, at least in a formal sense, be
-
1
13
derived from a Bergman cyclization of the enediyne crown
ether 5. The putative Bergman cyclization of 5 to 4 might
be facilitated by the appropriate alkali metal ion if 5, in
its complexed state, adopts a conformation that is more
strained14 than the complexed form of the putative 1,4-
didehydrobenzene intermediate, whose geometry should
1
5
be similar to the stable complex of 4.
Despite its symmetry and lack of diverse functionality,
the enediyne crown ether 5 presents a significant syn-
afford only intractable mixtures, presumably due to the
instability of the enediyne moiety to the strongly basic
reaction conditions necessary to alkylate the diol. The
inability to synthesize 5 through the enediyne diol 7 and
K o¨ nig and co-workers’ report of difficulties in alkylations
involving the enediyne dibromide 6 and basic nucleo-
philes such as alkoxides led to a search for alternative
routes to 5.
(9) (a) K o¨ nig, B.; Hollnagel, H.; Ahrens, B.; J ones, P. G. Angew.
Chem., Int. Ed. Eng. 1995, 34, 2538-2540. (b) K o¨ nig, B.; Pitsch, W.;
Thondorf, I. J . Org. Chem. 1996, 61, 4258-4261.
(10) Kerwin, S. M. Tetrahedron Lett. 1994, 35, 1023-1026.
(11) Cameron, I. L.; Smith, N. K. R. Magnesium, 1989, 8, 31-44.
(12) (a) Fukuda, R.; Takenaka, S.; Takagi, M. J . Chem. Soc., Chem.
Commun. 1990, 1028-1030. (b) Griffin, J . H.; Dervan,. P. B. J . Am.
Chem. Soc. 1987, 109, 6840-6842. (c) Basak, A.; Dugas, H. Tetrahedron
Lett. 1986, 27, 3-6.
A sequential macrocyclization-enediyne formation
strategy was next attempted. In this approach, macro-
cyclization conditions compatible with functionality suit-
(13) (a) Reinhoudt, D. N.; Gray, R. T.; Smit, C. J .; Veenstra, Ms. I.
Tetrahedron 1976, 32, 1161-1169. (b) Reinhoudt, D. N.; de J ong, F.
In Progress in Macrocyclic Chemistry; Izatt, R. M., Christensen, J . J .,
Eds; Wiley: New York, 1979; Vol. 1, pp 154-218.
(16) Nicolaou, K. C.; Zuccarello, G.; Ogawa, Y.; Schweiger, E. J .;
Kumazawa, T J . Am. Chem. Soc. 1988, 110, 4866-4868.
(17) J ones, G. B.; Huber, R. S.; Mathews, J . E. J . Chem Soc., Chem.
Commun. 1995, 1791-1792.
(
(
14) Snyder, J . P. J . Am. Chem. Soc. 1990, 112, 5367-5369.
15) Lindh, R.; Persson, B. J . J . Am. Chem. Soc. 1994, 116, 4963-
4
969.

Synthesis of Enediyne-Containing Crown Ethers
J . Org. Chem., Vol. 61, No. 26, 1996 9387
able for later elaboration to an enediyne are required.
Previous work by Nicolaou and co-workers demonstrated
that 10- to 16-membered ring macrocyclic enediynes can
be made from the corresponding 11- to 17-membered bis-
14 to form cumulene products.19 The rather low yields
from this route proved even more disappointing: the
2
0
crown ether so obtained proved to be impure by HPLC,
and further purification proved difficult. Given the poor
overall yields and purification difficulties encountered in
this approach, an alternative route to 5 was sought.
Con cu r r en t En ed iyn e F or m a tion -Ma cr ocycliza -
tion Rou tes. Although the strategy of forming the
enediyne during macrocyclization is potentially more
efficient than the sequential strategies discussed above,
it remained to be seen if a method could be found for
forming an enediyne in a macrocyclization reaction. To
our knowledge, there are only two reports of construction
of large-membered ring cyclic enediyne arrays in this
manner. Danishefsky and co-workers have reported on
the cross coupling of a bis-iodo acetylide with (Z)-bis-
(
propargylic)sulfide using a Ramberg-B a¨ cklund ring
1
6
contraction. In the case of the formation of crown ether
, a route analogous to that of Nicolaou required the 25-
membered bis(propargylic)sulfide macrocycle 11 (Scheme
). On the basis of a previous high-yielding synthesis of
5
1
an analogous 19-membered ring bis(propargylic) sulfide
macrocycle,10 a cyclization of the dichloride 10 to mac-
rocycle 11 was envisioned.
The route to sulfide 11 and its conversion to the crown
ether 5 is shown in Scheme 2. The dipropargyl ether of
pentaethylene glycol (8) is easily prepared from the diol
in high yield. Initial attempts to hydroxymethylate 8 by
formation of the diacetylide using n-BuLi or EtMgBr
followed by addition of formaldehyde or paraformalde-
hyde met with limited success, due to modest and highly
variable yields. As it appeared that the solubility of the
diacetylide of 8 was a limiting factor in these reactions,
hydroxymethylation was carried out using n-BuLi as base
in the presence of excess TMEDA. Under these condi-
tions, hydroxymethylation of 8 to diol 9 is accomplished
in reproducible, albeit moderate, yield. The conversion
of diol 9 to dichloride 10 also presented some difficulties,
as dichloride 10 is not stable to chromatography. Treat-
ing diol 9 with excess thionyl chloride in pyridine-
containing solvents under a variety of conditions affords
dichloride 10 contaminated with appreciable amounts of
the corresponding monochloro product. An alternative
procedure that circumvents the problem of incomplete
reaction of the diol was found. Thus, 9 is deprotonated
with n-BuLi in HMPA/THF, and the resulting di-
anion is quenched with thionyl chloride. Decomposition
of the resulting bis(chlorosulfite) is accomplished by the
addition of pyridine and heating the reaction mixture
under reflux. In this way, dichloride 10 is obtained in
high yield and sufficient purity for the key macrocycliza-
tion step.
(
trimethylstannyl)ethylene to form an enediyne analog
2
1
of dynemicin; however, this approach appears not to be
general, as a closely related bis-iodo acetylide failed to
undergo cross coupling. More recently, J ones and Huber
have reported a carbenoid coupling route to cyclic ene-
diynes.17 The generality of this later route has not been
demonstrated; in particular, its use in forming larger
than 11-membered enediyne-containing rings has not
been reported.
All attempts at a palladium-catalyzed macrocyclization
cross-coupling involving the di(propargyl) pentaethylene
glycol 8 and (Z)-1,2-dichloroethylene failed to produce any
enediyne-crown ether 5. These efforts, which included
standard, high-dilution, metal ion-templating, and
pseudodilution conditions employing a polymer-bound
tetrakis(triphenylphosphine)palladium reagent were only
successful in producing bis(5-chloropent-4-en-2-ynyl)-
pentaethylene glycol, the product resulting from 1:2
coupling between 8 and (Z)-1,2-dichloroethylene, even
when limiting amounts of (Z)-1,2-dichloroethylene were
employed.
An alternative route to 5 relies upon an adaptation of
17
the carbenoid coupling route of J ones and co-workers
in which a bis(propargylic)bromide undergoes cyclization
to form a 10-membered cyclic enediyne moiety (eq 3). In
Cyclization of dichloride 10 to the 25-membered mac-
rocycle 11 is accomplished in relatively high yield without
the need for high-dilution conditions using an alumina-
supported sodium sulfide reagent.18 The small amounts
of polymeric material produced in this reaction are easily
separated from sulfide 11 by column chromatography.
Conversion of macrocyclic sulfide 11 to the R-chloro
sulfone 14 proceeds without incident and sets the stage
for the key enediyne-forming Ramberg-B a¨ cklund reac-
tion.
Attempts to convert R-chloro sulfone 14 to enediyne
crown ether 5 produced variable results. A number of
bases were examined for their ability to promote the
Ramberg-B a¨ cklund ring contraction of 14 to 5, but
t-BuOK appeared to be the base of choice. Even so, when
t-BuOK was used as the base, the yields for the formation
of 5 varied between 0 and 13%, depending upon the
initial concentration of 14, the amount of base added, and
the order of addition. In general, a procedure in which
the application of this cyclization to the formation of the
macrocycle 5, competing intermolecular reactions of the
intermediate carbenoid species might lead to oligomeric
products. However, it was reasoned that high-dilution
conditions or the use of a templating metal ion might
overcome this potential shortcoming.
2
equiv of base are added to solutions of 14 (35 mM)
affords the highest yield of 5. The poor yields of 5 may
be due to the facile 1,4 elimination from R-chloro sulfone
(
19) Evidence for the formation of allenic species was obtained from
H NMR of the crude reaction mixtures, which showed a multiplic-
ity of olefinic signals.
20) The crown ether 5 obtained from the Ramberg-B a¨ cklund route
1
the
(
(
18) Czech, B.; Quici, S.; Regen, S. L. Synthesis 1980, 113. (b) Tan,
was 85% pure, as judged from peak areas in the HPLC chromatogram.
(21) Shair, M. D.; Yoon, T.; Danishefsky, S. J . J . Org. Chem. 1994,
59, 3755-3757.
L. C.; Pagni, R. M.; Kabalka, G. W.; Hillmyer, M.; Woosley, J .
Tetrahedron Lett. 1992, 33, 7709.

9
388 J . Org. Chem., Vol. 61, No. 26, 1996
McPhee and Kerwin
Sch em e 1
ratio of 5/(E/Z)-16 was further optimized by the addition
of “templating” alkali metal bromide salts (Table 1, entry
5
). Under these conditions, over 4 equiv of base are
required in order to drive the reaction to completion
(Table 1, entries 6 and 7).
The original purpose of adding alkali metal salts to
dibromide 15 was for templating purposes, in order to
avoid intermolecular reactions leading to dimeric prod-
ucts.23 However, it appears that these salts are playing
a more fundamental role, as the ratio of 5 to (E/Z)-16 is
so markedly affected by the addition of NaBr. Perhaps
the alkali metal bromides serve not only a templating
role but also influence the stability/reactivity of carbenoid
intermediate(s) involved. The stabilizing effect of alkali
2
4
metal halides on carbenoids is well documented. In-
terestingly, initial reports on the carbenoid coupling
method for acyclic and cyclic enediyne formation call for
the addition of 1-10 equiv of HMPA. It is thought that
the HMPA serves to destabilize the carbenoid intermedi-
ates involved and in so doing promote carbene insertion
and enediyne formation, reducing the amount of recov-
ered starting material arising from protonation of the
carbenoid upon workup (eq 3).25 However, addition of
excessive amounts of HMPA is reported to result in
lowered yields of enediyne and the formation of side
products.2 It appears that in the carbenoid coupling
route to enediynes, a balance between carbenoid stability
and reactivity is required.
The required dibromide 15 was prepared in good yield
from the diol 9 (eq 4). Interestingly, whereas the
dichloride 10 proved to be unstable to silica gel chroma-
tography, the dibromide 15 could be subjected to chro-
matography with good recovery. Attempts to effect the
intramolecular coupling of dibromide 15 to the enediyne
crown ether 5 under conditions (2.3 equiv of LiHMDS,
5b
Isomeric 5-ene-1,3-diynes such as (E/Z)-16 have not
previously been isolated from the carbenoid coupling
reactions of propargylic halides or bis(propargylic) ha-
lides. Thus, the formation of (E/Z)-16 as the major
product in the carbenoid cyclization of 15 under certain
conditions stands in strong contrast to previous reports
of acyclic or 10-membered ring cyclic enediyne formation.
The propensity of 15 to form the isomeric enediyne (E/
Z)-16 under carbenoid coupling conditions that produce
exclusively 1,5-diyne products from other bis(propargylic)
2
0 equiv of DMPU, THF, -50 °C) similar to the optimized
conditions reported by J ones and co-workers afforded
very complex mixtures of products (Scheme 3). In
1
addition to remaining dibromide, the H NMR spectrum
of the crude product mixture revealed the presence of
three major components. These products could only be
isolated after repeated chromatographic purification of
the product mixture. The primary component observed
1
in the H NMR spectrum of the crude reaction product
was identified22 as the (Z)-5-ene-1,3-diyne (Z)-16, which
was isolated in 8% yield along with a small amount of
the (E)-isomer (E)-16 (<1% yield). The third major
component of the product mixture was identified as the
desired enediyne 5, which was obtained in 4% yield after
purification.
26
halides (Table 1, entry 2) indicates that the formation
of 1,3-diyne products from the coupling of dibromide 15
is due to structural differences between this dibromide
and other dibromides that have been employed in these
coupling reactions. For example, the polyether nature
of dibromide 15 may affect the reactivity of the initially
formed carbenoid intermediate. It has been recently
shown that a neighboring methoxyethoxyethyl group
both directs the lithiation of a primary halide and alters
the reactivity of the carbenoid intermediate that is
A number of reaction conditions were explored in order
to improve the production of enediyne-crown ether 5 from
the dibromide 15 (Table 1). Specifically, the nature of
the chelator and its inclusion with either the base or the
starting dibromide were varied. The optimized condi-
tions for carbenoid coupling reported by J ones and co-
workers utilize HMPA as the chelator. When DMPU is
replaced with HMPA, the ratio of (E/Z)-16 to 5 obtained
is nearly unchanged (Table 1, entry 2), although the
extent of conversion of dibromide 15 to enediyne products
is improved. The use of the chelator TMEDA, when
present with the starting dibromide 15, results in very
low mass recovery after aqueous work-up, presumably
due to the production of water soluble products (Table 1,
entry 3). However, substituting TMEDA for DMPU, with
the chelator present with the base rather than the
dibromide, results in improved ratios of 5 to isomeric
product (E/Z)-16 (Table 1, compare entries 1 and 4). The
(
23) Although this material appeared homogeneous by 1H NMR, it
contained a small amount (ca. 10%) of a more polar component by TLC.
Isolation of this more polar component and analysis by MS indicated
that it was the dimer:
In preparative reactions, pure 5 was obtained free from this dimer
after column chromatography.
(24) Tarhouni, R.; Kirshleger, M.; Rambaud, M.; Villieras, J . Tet-
rahedron Lett. 1984, 25, 835-838.
(
25) (a) Huber, R. S.; J ones, G. B. Tetrahedron Lett. 1994, 25, 2655-
2658. (b) J ones, G. B.; Huber, R. S. US Pat. 5,436,361.
26) When (trimethylsilyl)propargyl bromide is allowed to react
(
under the conditions of Table 1, entry 1, an approximately 2:1 ratio of
(Z)- and (E)-1,6-[bis(trimethylsilyl)]hex-3-ene-1,5-diyne is formed, with
no observable formation of the corresponding isomeric 1,3-diyne
product.
(
22) The structural identification of (Z)-16 was aided by an HMBC
NMR experiment, which revealed the presence of the 5-ene-1,3-diyne
moiety.

Synthesis of Enediyne-Containing Crown Ethers
J . Org. Chem., Vol. 61, No. 26, 1996 9389
Sch em e 2
Ta ble 1. Effects of Rea ction Con d ition s on th e Ca r ben oid Cou p lin g Cycliza tion of Dibr om id e 15
methoda
chelator
MXb
mass recovery (%)
c
ratio of 15/5 /(E/Z)-16
d
e
f
entry
1
2
3
4
5
6
7
A
A
A
B
B
C
C
DMPU
HMPA
none
none
none
none
KBr
50g
49
4
93
93
64
65
15:25:60h
0:30:70
i
30:30:40
40:40:20
0:60:40
0:80:20
g
TMEDA
TMEDA
TMEDA
TMEDA
TMEDA
KBr
NaBr
a
For all methods, the final concentration of 15 after addition of base was 10 mM. Method A: A solution of 2.3 equiv of LiHMDS in
THF was added to a solution of 15 in THF-containing chelator. Method B: A solution containing 2.3 equiv of LiHMDS and chelator was
added to a solution of 15 in THF. Method C: A solution containing 4.2 equiv of LiHMDS and chelator was added a to solution of 15.
b
c
When present, 50 equiv of alkali metal salt MX was added to the dibromide 15 prior to the addition of base. Mass recovery after
d
aqueous workup, calculated as a percentage of the theoretical yield of 5. Ratio of products and starting material, if present, as determined
1
e
f
by H NMR of the crude reaction mixture after workup. See footnote 23. Compound 16 was produced as a ∼1:3 mixture of E/Z
g
h
stereoisomers. After column chromatography to remove the DMPU or HMPA. Isolation of 5 and (E/Z)-16 after repeated chromatographic
i
1
steps afforded an isolated ratio of 0.32:0.68 of 5 to (E/Z)-16. Only unidentified products and a trace of 5 were see in the crude H NMR.
produced.27 Additionally, the fact that the initial product
of the coupling of 15 is a crown ether may play a role in
the formation of (E/Z)-16. Recent X-ray crystal struc-
tures of crown ether-carbenoid complexes demonstrate
the strong interaction between the metal center of a
carbenoid and the crown ether ring.28 This type of
interaction may affect the reactivity of subsequently
formed carbenoid intermediates. Reaction conditions
that further destabilize the carbenoid intermediates, such
as the presence of a excess of chelator with starting
dibromide 15, result in increased amounts of the 1,3-
diyne (E/Z)-16 at the expense of the 1,5-diyne 5. Con-
(
28) Charette, A. B.; Marcois, J -F.; B e´ langer-Gari e´ py, F. J . Am.
(27) Clayden, J .; J ulia, M. Synlett. 1995, 103-104.
Chem. Soc. 1996, 118, 6792-6793.

9
390 J . Org. Chem., Vol. 61, No. 26, 1996
McPhee and Kerwin
Sch em e 3
Ta ble 2. Alk a li Meta l Ion Bin d in g Ability of
En ed iyn e-Cr ow n Eth er s
%
picrate extracteda
Na+
compound
Li+
K+
4
5
6
2.0 ( 0.3%
7.99 ( 0.03%
9.82 ( 0.03%
6.30 ( 0.03%
9.2 ( 0.3%
9 ( 2%
35.7 ( 0.1%
16.1 ( 0.4%
10.55 ( 0.07%
1
a
Equal volumes of solutions of crown ether (30 mM) in CH2Cl2
and aqueous solutions of alkali metal picrates (3 mM) were mixed,
and the concentration of picrate extracted into the organic layer
was determined spectrophotometrically. The amount of picrate
extracted is expressed as a percentage of the amount of alkali
metal picrate in the original aqueous solution. Values represent
the average of two separate determinations.
extraction assay (Table 2).30 Crown ether 5 appears to
bind potassium ions more strongly than either sodium
or lithium ions, which are bound with comparable affini-
ties. This potassium ion selectivity of crown ether 5 is
analogous to, but somewhat diminished from, that dem-
onstrated by the o-xylyl crown ether 4.13 In contrast,
the 1,3-diyne crown ether (Z)-16 appears to bind all alkali
metal ions examined with near-equal affinity. Surpris-
ingly, the 1,3-diyne crown ether (Z)-16, despite the
elongated nature of the crown ether portion, demon-
strates metal ion extraction ability that is comparable
to that displayed by crown ether 5. Given the near
identical sodium ion binding ability of 5 and (Z)-16, it
appears unlikely that the dramatic effect of added NaBr
on the ratio of 5 to (E/Z)-16 produced in the carbenoid
coupling of 15 is due to selective complexation of 5 as it
is formed in the reaction.
versely, reaction conditions that serve to stabilize the
carbenoid intermediates, such as the addition of alkali
metal bromide salts, give rise to increased yields of the
1
,5-diyne product.
The origin of 5-ene-1,3-diyne (E/Z)-16 is not clear.
Perhaps the 1,3-diyne (E/Z)-16 arises as a result of the
well-established rearrangement29 of a propargylic car-
bene intermediate. An alternative mechanism for the
formation of (E/Z)-16 involving the equilibration of 5 to
(
E/Z)-16 under the reaction conditions is not at work;
when resubjected to the conditions that favor the forma-
tion of 16 (Table 1, entry 1), 5 does not undergo any
measurable isomerization to (E/Z)-16.
Con clu sion s
Under the optimized conditions detailed in Table 1,
entry 7, pure enediyne-crown ether 5 is isolated in 29%
yield after silver-impregnated silica gel chromatography,
for an overall yield of 9% from pentaethylene glycol.
Although this is not as high-yielding as literature ex-
amples of forming 10-membered enediyne-containing
rings via the carbenoid coupling reaction, it does compare
very favorably with the 1.8% overall yield of impure
enediyne 5 from the alternative Ramberg-B a¨ cklund
route (Scheme 2). The (Z)-stereochemistry of 5 was
assigned on the basis of both on the identity with the
material produced through the Ramberg-B a¨ cklund route,
and the facility with which 5 undergoes Bergman cy-
The enediyne-crown ether 5 has been prepared by two
different routes. In the shorter and preferable route, a
carbenoid coupling reaction was employed to simulta-
neously construct the enediyne moiety and effect a
macrocyclization of the acyclic dibromide 15 to the 24-
membered crown ether 5. Under certain conditions, this
carbenoid coupling produces as the major product an
isomeric macrocyclic 5-ene-1,3-diyne (E/Z)-16. The for-
mation of a 5-ene-1,3-diyne from the carbenoid coupling
reaction is unprecedented, and we have presented evi-
dence that it is the polyether nature of dibromide 15 that
leads to the formation of this product, perhaps through
intramolecular complexation of intermediate carbenoid
species produced in the reaction. Work is underway to
probe the nature of the conversion of 15 to (E/Z)-16 in
more detail.
clization. Thus, heating a solution of 5 in DMSO-d
containing a large excess of 1,4-cyclohexadiene at 135 °C
6
1
3
for 45.5 h produces the previously reported o-xylyl
1
crown ether 4 (eq 5), as evidenced by H NMR and MS
analysis of the reaction mixture.
Both enediyne-crown ethers 5 and (Z)-16 bind alkali
metal ions, as evidenced by their ability to extract alkali
metal picrates into organic solvents. While 5 demon-
strates an expected increased extractability for potassium
ions, the crown ether (Z)-16 appears to bind lithium,
sodium, and potassium with near-equal affinity.
Enediyne-crown ether 5 undergoes Bergman cycliza-
tion at 135 °C in DMSO/1,4-cyclohexadiene to produce
the known o-xylyl crown ether 4. Studies of the effect of
alkali metal ions on the rate of this Bergman cyclization
are currently underway and will be reported upon in due
course.
Meta l Ion Bin d in g Stu d ies of En ed iyn e-Cr ow n
Eth er s. The metal ion binding properties of o-xylyl
crown ether 4, (Z)-3-ene-1,5-diyne crown ether 5, and the
(Z)-5-ene-1,3-diyne (Z)-16 were examined using a pricrate
(
29) (a) Noro, M.; Masuda, T.; Ichimura, A. S.; Koga, N.; Iwamura,
H. J . Am. Chem. Soc. 1994, 116, 6177-6190. (b) Padwa, A.; Gareau,
Y.; Xu, S. L. Tetrahedron Lett. 1991, 32, 983-986.
(30) Ouchi, M.; Inoue, Y.; Wada, K.; Iketani, S.; Hakushi, T.; Weber,
E. J . Org. Chem. 1987, 52, 2420-2427.

Synthesis of Enediyne-Containing Crown Ethers
J . Org. Chem., Vol. 61, No. 26, 1996 9391
saturated aqueous NH
Cl. The aqueous layer was extracted
Exp er im en ta l Section
4
with EtOAc (3 × 30 mL), and the combined organic layers were
washed with brine (1 × 40 mL). The residue upon drying and
concentration was purified by flash chromatography on silica
gel (10% MeOH in EtOAc) to afford diol 9 (0.414 g, 50%) as a
pale yellow oil: 1H NMR δ 3.32 (s(br), 2H), 3.52-3.70 (m, 20H),
Gen er a l P r oced u r es. Unless otherwise noted, all materi-
als were obtained from commercial suppliers and were used
without further purification. THF was distilled from sodium/
benzophenone immediately prior to use. CH
2 2
Cl and MeCN
1
3
were distilled from CaH immediately prior to use. Absolute
2
4.14 (t, J ) 1.7 Hz, 4H), 4.16-4.22 (m, 4H); C NMR δ 50.19,
ethanol was dried and stored over 4 Å sieves. Pentaethylene
glycol, pyridine, HMPA, 1,4-cyclohexadiene, and HMDS were
distilled prior to use. TMEDA, n-butylamine, and N,N′-
58.33, 68.67, 70.07, 70.21(4C), 80.52, 85.12; IR 3443, 2122,
-
1
+
1121 cm ; MS 375(MH ), 345, 307; HRMS m/ e calcd for
C H O 375.2019, found 375.2016.
1
8
31
8
dimethylpropyleneurea (DMPU) were distilled from CaH
to use. SOCl and SO Cl were distilled from linseed oil and
stored in glass-stoppered vessels under argon. Paraformal-
dehyde was dried and stored in a vacuum desiccator over P
2
prior
1,14-Bis(4-ch lor o-2-b u t yn -1-oxy)-3,6,9,12-t e t r a oxa -
2
2
2
tetr a d eca n e (10). To a solution of 9 (0.328 g, 0.877 mmol)
in 38 mL of THF was added with stirring HMPA (365 µL, 2.1
mmol), and this solution was cooled to 0 °C in an ice-water
bath. A solution of n-BuLi (2.58M, 750 µL, 1.94 mmol) was
then added dropwise with stirring, and the reaction mixture
was allowed to warm to room-temperature. A solution of
2
5
O .
All reactions were performed under an inert atmosphere of
either argon or nitrogen in oven-dried glassware. Unless
otherwise noted, organic extracts were dried with Na SO ,
2 4
filtered through a fritted glass funnel, and concentrated with
SOCl (450 µL, 6.17 mmol) in 7 mL of THF was then added
2
a rotary evaporator (20-30 mm). Melting points (open capil-
quickly via cannula with stirring, and the reaction mixture
was allowed to stir for an additional 15 min. Pyridine (320
µL, 3.96 mmol) was added dropwise, and the reaction mixture
was heated under reflux for 15 h. After allowing the dark
brown solution to cool to room temperature, water (20 mL)
was added, and the resulting mixture was extracted with
EtOAc (3 x 10 mL). The combined organic layers were washed
3
lary) are uncorrected. IR spectra were determined in CHCl .
1
13
Unless otherwise noted, H and C NMR spectra were
determined in CDCl on a spectrometer operating at 250 and
2.89 MHz, respectively. All mass spectra were obtained by
3
6
chemical ionization using methane as the ionizing gas.
Oct-4-en e-2,6-d iyn e-1,8-d iol (7). In a 100 mL round-
bottomed flask under argon was placed O-(trimethylsilyl)-
propargyl alcohol (2.0 g, 15.6 mmol), prepared from propargyl
alcohol and chlorotrimethylsilane, benzene (10 mL), tetrakis-
with saturated aqueous NaHCO (2 × 20 mL), water (5 × 20
3
mL), and brine (1 × 25 mL). Drying and concentration of the
organic layer afforded dichloride 10 (0.319 g, 89%) as a dark
brown oil. Dichloride 10 was unstable to both alumina and
silica gel column chromatography but was sufficiently pure
to be used without further purification: 1H NMR δ 3.60-3.70
(triphenylphosphine)palladium(0) (0.516g, 0.4 mmol), cuprous
iodide (0.114 g, 0.6 mmol), and n-butylamine (2.2 mL, 22.2
mmol). (Z)-1,2-Dichloroethylene (0.5 mL, 7.2 mmol) was added
to the reaction mixture, which was allowed to stir overnight.
The reaction mixture was diluted with ether (100 mL) and
filtered. The filtrate was concentrated and dissolved in MeOH
1
3
(m, 20H), 4.16 (t, J ) 1.9 Hz, 4H), 4.23 (t, J ) 1.9 Hz, 4H);
C
NMR δ 30.29, 58.52, 69.19, 70.32, 70.50(2C), 70.54, 81.05,
82.49; IR 1135, 1094, 700 cm ; MS 411 (MH ); HRMS m/ e
-
1
+
(75 mL) containing a few drops of AcOH. Concentration and
calcd for C H Cl O 411.1341, found 411.1341.
1
8
29
2
6
purification of the product mixture by flash chromatography
6,9,12,15,18,21-H e xa oxa -1-t h ia cyclop e n t a cosa -3,23-
d iyn e (11). To a solution of 10 (3.06 g, 7.45 mmol) in 318 mL
of 5:1 CH Cl /EtOH was added Na S‚Al O (21.2% w/w, 7.97
g, 21.7 mmol) in one portion, and the heterogeneous reaction
mixture was stirred vigorously overnight at room temperature.
The reaction mixture was filtered through Celite, and the
on silica gel (25% EtOAc in hexanes) afforded diol 7 (0.343 g,
o
1
3
(
9
5%) as an off-white solid: mp 57.5-58 C; H NMR δ 2.10
13
2
2
2
2
3
bs, 2H), 4.47 (s, 4H), 5.87 (s, 2H); C NMR δ 51.50, 82.90,
5.31, 119.47; IR 3274, 2216, 1106 cm-1; MS 137 (MH ); HRMS
m/ e calcd for C 137.0603, found 137.0600.
,14-B i s (2-p r o p y n -1-o x y )-3,6,9,12-t e t r a o x a t e t r a -
+
8
9 2
H O
1
2 2 2 2
solids were washed with CH Cl and 25% EtOAc in CH Cl .
d eca n e (8). To an ice-water bath-cooled stirring solution of
t-BuOK (95% w/w, 8.01 g, 67.8 mmol) in 33 mL of THF was
added via cannula over 1 min a solution of pentaethylene glycol
The combined eluant was concentrated, and the residue was
purified by flash chromatography on silica gel (5% MeOH in
1
EtOAc) to afford sulfide 11 (1.49 g, 54%) as a yellow oil:
H
(
7.18 g, 30.1 mmol) in 13 mL of THF. The resultant solution
NMR δ 3.47 (t, J ) 2 Hz, 4H), 3.63-3.72 (m, 20H), 4.23 (t, J
was allowed to stir for 10 min and was then added dropwise
via cannula over 16 min to an ice-water bath-cooled, me-
chanically stirred solution of propargyl bromide (80% w/w, 20.0
g, 134 mmol) in 46 mL of THF. The reaction mixture was
allowed to stir at 0 °C for 1 h and then at room temperature
for an additional 14 h. The reaction mixture was diluted with
) 2 Hz, 4H); 13C NMR δ 19.17, 58.77, 68.94, 70.51, 70.71, 70.80,
-
1
+
70.87, 79.52, 81.42; IR 1141, 1099 cm ; MS 373 (MH ), 307;
HRMS m/ e calcd for C18
H
29
O
6
S 373.1685, found 373.1682.
Na S‚Al Rea gen t. To Na
2
2
O
3
2
S‚9H O (7.1 g, 0.03 mmol)
2
that had been rinsed with a small amount of distilled,
deionized water and placed in a flask under argon was added
18 mL of warm, distilled, deionized water that had been boiled
1
00 mL of 3:1 saturated brine/water, and the aqueous layer
was extracted with EtOAc (3 × 100 mL). The combined
to remove CO
containing Al
2
. The resultant solution was poured into a flask
(neutral, Brockman activity I, 80-200 mesh,
organic layers were washed with water (3 × 20 mL) and brine
2
O
3
(
1 × 75 mL). The residue upon drying and concentration was
8.7 g, 0.085 mmol), and the water was removed in vacuo with
a rotary evaporator with gentle heating in a warm water bath.
The material was then activated by heating in vacuo (95 °C,
0.1 Torr) for 1.5 h until the material (21.2% w/w Na S) was a
2
free-flowing pink powder. The reagent was stored under argon
and used shortly after it was prepared.
purified by flash chromatography on silica gel (EtOAc) to afford
1
diyne 8 (8.67 g, 92%) as a pale yellow solid: mp 36-37 °C; H
NMR δ 2.39 (t, J ) 2.3 Hz, 2H), 3.54-3.67 (m, 20H), 4.14 (d,
1
3
J ) 2.3 Hz, 4H); C NMR δ 58.23, 68.94, 70.24, 70.43(2C),
-
1
+
7
2
3
0.45, 74.43, 79.52; IR 3259, 2117, 1100 cm ; MS 315 (MH ),
77, 259; HRMS m/ e calcd for C16
H
27
O
6
: 315.1808, found
6,9,12,15,18,21-H e xa oxa -1-t h ia cyclop e n t a cosa -3,23-
d iyn e 1-Oxid e (12). To a solution of 11 (0.062 g, 0.165 mmol)
15.1806.
1
,14-Bis(4-h yd r oxy-2-b u t yn -1-oxy)-3,6,9,12-t et r a oxa -
2 2
in 6.75 mL of CH Cl that had been cooled to -30 °C in a dry
ice/i-PrOH bath was added dropwise with stirring via cannula
over 30 s a solution of m-CPBA (50% w/w, 0.063 g, 0.182 mmol)
tetr a d eca n e (9). To a vigorously stirred solution of 8 (0.697
g, 2.22 mmol) in 100 mL of THF was added TMEDA (6.7 mL,
4
4.4 mmol), and this solution was cooled to -78 °C in a dry
in 2.25 mL of CH
stir at -30 °C for 1.25 h. While still cold, the reaction mixture
was diluted with 20 mL of CH Cl and then washed with
saturated aqueous Na CO
(1 × 20 mL) and saturated aqueous
NaHCO
(1 × 20 mL). The aqueous washes were extracted
with CH Cl
2 2
Cl . The resultant solution was allowed to
ice/acetone bath. A solution of n-BuLi (2.33M, 2.3 mL, 5.36
mmol) was added dropwise, and the resulting solution was
allowed to stir for 5 min. A stirring suspension of paraform-
aldehyde (95% w/w, 1.06 g, 33.6 mmol) in 20 mL of THF was
added quickly via cannula. The heterogeneous reaction mix-
ture was allowed to stir at -78 °C for 15 min, the cooling bath
was removed, and the reaction mixture was heated under
reflux until the solution was as homogeneous as possible (∼20
min). The cooled solution was then quenched with 60 mL of
2
2
2
3
3
2
2
(2 × 12 mL), and the combined organic layers
were washed with brine (1 × 20 mL). The residue upon drying
and concentration was purified by flash chromatography on
silica gel (15% MeOH in EtOAc) to afford sulfoxide 12 (0.048
1
g, 75%) as a colorless solid: mp 50.5-51.5 °C; H NMR δ 3.53-

9
392 J . Org. Chem., Vol. 61, No. 26, 1996
McPhee and Kerwin
1146, 1105, 620 cm- ; MS 501 (MH ), 435, 369; HRMS m/ e
1
+
3
1
5
2
.63 (m, 20H), 3.68 (dt, J ) 15.9, 2.3 Hz, 2H), 3.83 (dt, J )
5.9, 2.3 Hz, 2H), 4.19 (t, J ) 2.3 Hz, 4H); 13C NMR δ 40.99,
8.48, 68.98, 70.27, 70.52, 70.58, 70.64, 74.13, 84.60; IR 2242,
119, 1135, 1095, 1065 cm-1; MS 389 (MH ), 337, 321; HRMS
calcd for C18 499.0331, found 499.0313.
2 6
H29Br O
(Z)-8,11,14,17,20,23-H exa oxa cyclot et r a cos-3-en e-1,5-
d iyn e (5). F r om Ch lor osu lfon e 14. To a solution of 14
(0.085 g, 0.199 mmol) in 6 mL of THF that had been cooled to
-78 °C in a dry ice/acetone bath was added in one portion via
cannula with stirring a dry ice/acetone bath-cooled suspension
of t-BuOK (95% w/w, 0.05 g, 0.423 mmol) in 1 mL of THF.
The dark brown reaction mixture was stirred at -78 °C for
+
m/ e calcd for C18
,9,12,15,18,21-Hexaoxa-2-ch lor o-1-th iacyclopen tacosa-
,23-d iyn e 1-Oxid e (13). To a stirring solution of 12 (0.367
g, 0.946 mmol) in 20 mL of CH Cl was added pyridine (170
µL, 2.1 mmol). The resulting solution was cooled to -78 °C
in a dry ice/acetone bath. SO Cl
29 7
H O S 389.1634, found 389.1620.
6
3
2
2
2
2
(160 µL, 1.98 mmol) was
3
15 min. Solid NaHCO was added, followed by 10 mL of
added in one portion with efficient stirring, and the reaction
benzene. The reaction mixture was allowed to warm to room
temperature and heated in an oil bath for 3 min at 50 °C. The
reaction mixture was then transferred to a separatory funnel
containing 20 mL of brine and 20 mL of EtOAc. The layers
were mixed and separated, and the aqueous layer was
extracted with EtOAc (3 × 30 mL). The residue upon drying
and concentration of the organic layer was purified by flash
column chromatography on silica gel (5% MeOH in EtOAc) to
afford enediyne 5 (0.01 g, 11%) as a colorless oil that was 85%
mixture was allowed to stir for an additional 20 min at -78
°
C. The reaction mixture was diluted with 40 mL of EtOAc.
This solution was transferred to a separatory funnel containing
0 mL of EtOAc and 40 mL of water, the layers were mixed
4
and then separated, and the aqueous layer was extracted with
EtOAc (3 × 40 mL). The combined organic layers were washed
with saturated aqueous NaHCO
3
(1 × 20 mL) and brine (1 ×
1
5 mL). Drying and concentration of the organic layer afforded
chlorosulfoxide 13 (0.383 g, 96%) as a yellow oil that turns
olive on standing. Chlorosulfoxide 13 could not be obtained
analytically pure due to its instability to silica gel column
pure by HPLC (4.5 × 250 mm microsorb SiO
2
, 1.0 mL/min
EtOAc, t ) 9.2 min).
R
F r om Dibr om id e 15. A vigorously stirred suspension of
15 (0.09 g, 0.18 mmol) and pulverized, desiccated NaBr (0.927
g, 9.0 mmol) in 15 mL of THF was cooled to -55 °C in a dry
ice/acetone bath. A room-temperature solution of LiHMDS/
TMEDA, formed by the addition of n-BuLi (2.2 M, 410 µL,
0.902 mmol) to an ice-water bath-cooled solution of HMDS
(190 µL, 0.9 mmol) and TMEDA (550 µL, 3.64 mmol) in THF
(3 mL), was added dropwise over 4 min via addition funnel to
the reaction mixture. The resulting dark green reaction
mixture was allowed to stir for an additional 22 min at -55
°C before being quenched with 10 mL of 2% aqueous HCl. The
mixture was allowed to warm to room-temperature and was
transferred to a separatory funnel containing 12 mL of EtOAc.
The aqueous layer was extracted with EtOAc (3 × 10 mL),
and the combined organic layers were washed with water (3
× 5 mL) and brine (1 × 8 mL). The residue upon drying and
concentration of the organic layer was purified by flash column
chromatography in the dark using silica gel impregnated with
chromatography but was sufficiently pure for immediate
1
subsequent use: H NMR (250 MHz, CDCl
3
, 2:1 diastereomeric
mixture) δ 3.57-3.76 (m, 20H), 3.82-3.97 (m, 2H), 4.26 (t, J
)
1.6 Hz, 2H), 4.34 (d, J ) 1.6 Hz, 1.34H), 4.36 (d, J ) 1.6 Hz,
0
.66H), 5.53 (t, J ) 1.6 Hz, 0.67H), 5.60 (t, J ) 1.6 Hz, 0.33H);
C NMR δ 41.24, 58.42, 58.45, 58.52, 61.13, 61.56, 69.10,
9.38, 70.29, 70.35, 70.60 (2C), 73.85, 73.95, 76.32, 85.12,
5.28, 89.95; IR 2231, 1110 cm-1; MS 423 (MH ), 341, 305;
1
3
6
8
+
HRMS m/ e calcd for C18
,9,12,15,18,21-Hexaoxa-2-ch lor o-1-th iacyclopen tacosa-
,23-d iyn e 1,1-Dioxid e (14). A solution of 13 (0.222 g, 0.525
Cl was cooled to 0 °C in an ice-water
7
H28ClO S 423.1244, found 423.1238.
6
3
mmol) in 13 mL of CH
2
2
bath, and peracetic acid (33% w/w, 0.61 g, 2.65 mmol) was
added dropwise with stirring over 1 min. The reaction mixture
was allowed to stir for 0.5 h at 0 °C and then overnight at
room-temperature. The reaction mixture was diluted with 80
mL of EtOAc and transferred to a separatory funnel containing
3
0 mL of water. The layers were mixed and then separated.
20% w/w AgNO
3
(15% MeOH in EtOAc). The eluant was
The aqueous layer was extracted with EtOAc (3 × 20 mL).
collected into tubes containing 1 mL of brine. The organic
The combined organic layers were washed with saturated
layers in the fractions of interest were combined and evapo-
1
aqueous Na
2
SO
3
(1 × 40 mL) and brine (1 × 20 mL). Drying
rated to afford enediyne 5 (0.018 g, 29%) as a colorless oil:
H
1
3
and concentration afforded chlorosulfone 14 (0.195 g, 85%) as
a pale yellow oil. Chlorosulfone 14 was unstable to silica gel
column chromatography but was sufficiently pure for further
NMR δ 3.55-3.73 (m, 20H), 4.34 (s, 4H), 5.79 (s, 4H); C NMR
δ 58.92, 68.88, 70.33, 70.57, 70.64, 70.71, 83.54, 92.87, 119.53;
IR 3049, 2211, 1106 cm ; MS 339 (MH ); HRMS m/ e calcd
for C18 339.1808, found 339.1809.
-
1
+
1
use: H NMR δ 3.59-3.75 (m, 20H), 4.27 (s(br), 4H), 4.35 (d,
27 6
H O
1
3
J ) 1.7 Hz, 2H), 5.81 (t, J ) 1.7 Hz, 1H); C NMR δ 41.99,
(Z)-8,11,14,17,20,23-H exa oxa cyclot et r a cos-1-en e-3,5-
d iyn e [(Z)-16)] a n d (E)-8,11,14,17,20,23-H exa oxa cyclo-
tetr a cos-1-en e-3,5-d iyn e [(E)-16)]. DMPU (3.1 mL, 25.6
mmol) was added to a stirring solution of 15 (0.642 g, 1.28
mmol) in 107 mL of THF in a 500 mL flask equipped with a
pressure-equalized addition funnel. After the reaction mixture
was cooled to -63 °C in a dry ice/acetone bath, the addition
funnel was charged with a room-temperature solution of
LiHMDS that was prepared by the addition of n-BuLi (2.33
M, 1.25 mL, 2.91 mmol) to an ice-water bath-cooled solution
of HMDS (620 µL, 2.94 mmol) in 21 mL of THF. The LiHMDS
solution was added dropwise with vigorous stirring over 29
min. The reaction mixture was allowed to stir for an ad-
ditional 16 min at -63 °C and was then quenched while still
cold with 36 mL of 1% aqueous HCl. The forest-green reaction
mixture was allowed to warm to room-temperature and was
shaken with 85 mL of water and 100 mL of EtOAc in a
separatory funnel. The aqueous layer was extracted with
EtOAc (4 × 50 mL), and the combined organic layers were
washed with brine (1 × 60 mL). The residue upon drying and
concentration of the organic layer was purified by flash column
chromatography on silica gel (5% MeOH in EtOAc) to remove
the DMPU. All enediyne-containing fractions were pooled,
concentrated, and subjected to flash column chromatography
in the absence of direct light on silica gel impregnated with
5
7
8.30, 59.78, 69.06, 69.34, 70.23, 70.32, 70.49(2C), 70.52(3C),
0.55(2C), 72.36, 74.50, 85.37, 90.05; IR 2235, 1356, 1150 cm
-1
;
+
MS 439 (MH ), 405, 351; HRMS m/ e calcd for C18
39.1193, found 439.1187.
,14-Bis(4-b r om o-2-b u t yn -1-oxy)-3,6,9,12-t e t r a oxa -
tetr a d eca n e (15). To a vigorously stirred, ice-water bath-
cooled solution of PPh (1.98 g, 7.55 mmol) in 36 mL of MeCN
was added via addition funnel over 6.5 min a solution of Br
375 µL, 7.32 mmol) in 2 mL of MeCN. The addition funnel
8
H28ClO S
4
1
3
2
(
was rinsed with an additional 1.2 mL of MeCN, and this was
added to the mixture as well. The reaction mixture was
allowed to stir at 0 °C for 10 min, and then a solution of diol
9
(1.33 g, 3.56 mmol) in 21.5 mL of MeCN was added dropwise
via addition funnel with vigorous stirring over 7 min. The ice-
water bath was allowed to warm to room-temperature, and
the reaction mixture was stirred for an additional 4.5 h at room
temperature. The reaction mixture was transferred to a
separatory funnel containing 175 mL of EtOAc and 175 mL of
water, and the layers were mixed then separated. The
aqueous layer was extracted with EtOAc (3 × 115 mL). The
combined organic layers were washed with saturated aqueous
NaHCO
3
(1 × 130 mL) and brine (1 × 130 mL). The residue
upon drying and concentration of the organic layer was
purified by flash column chromatography on silica gel (10-
0
% hexanes in EtOAc) to afford dibromide 15 (1.2 g, 67%) as
25% w/w AgNO
3
(15% MeOH in EtOAc) with the eluting
1
a colorless solid: mp 31.5-32.5 °C; H NMR δ 3.59-3.70 (m,
0H), 3.93 (t, J ) 2 Hz, 4H), 4.24 (t, J ) 2 Hz, 4H); 13C NMR
δ 14.12, 58.50, 69.10, 70.24, 70.43(2C), 70.45, 81.26, 82.76; IR
fractions being collected into tubes containing 2 mL of brine.
Separation of the organic layers of the fractions containing only
one component afforded 22 mg of pure (Z)-16. Further
2

Synthesis of Enediyne-Containing Crown Ethers
J . Org. Chem., Vol. 61, No. 26, 1996 9393
purification of the earliest eluting mixed fractions [preparative
TLC (1 mm silica gel plate, 5% MeOH in EtOAc), flash
chromatography (silica gel, 5% MeOH in EtOAc), followed by
preparative TLC (1 mm silica gel plate, EtOAc)] afforded an
additional 14 mg of (Z)-16 as a pale yellow solid (36 mg total,
Alk a li Meta l P icr a te Extr a ction . Alkali metal ion bind-
ing was characterized by measuring percent metal picrate
extracted from an aqueous layer into an organic layer via
complexation by the crown ether.31 All glassware was washed
thoroughly with a nonionic detergent to remove traces of alkali
metal ions. Briefly, 150 µL of a 30 mM water-saturated
8
%) and 2 mg of the slightly faster eluting (E)-16, contami-
nated with ∼5% of (Z)-16, as a colorless oil (0.5%). The later
eluting mixed fractions from the AgNO on silica gel flash
CH
CH
2
Cl
Cl
2
crown ether solution was added to 150 µL of a 3 mM
3
2
2
-saturated aqueous alkali metal (lithium, sodium, or
chromatography, upon further purification by flash chroma-
tography on silica gel (5% MeOH in EtOAc), afforded 5 (18
potassium) picrate solution in a small culture tube. The
mixture was mixed for 10 min, allowed to stand for 30 min,
and then centrifuged at high speed for 10 min. A 25 µL portion
mg, 4%) as a colorless oil. Analytical data for (Z)-16: mp 51-
1
5
2 °C; H NMR (250 MHz, DMSO-d
6
) δ 3.45 (m, 20H), 4.22 (d,
of the CH
2
Cl
2
layer was withdrawn and diluted with 25 µL of
Cl /MeCN. This solution was
homogenized, 250 µL of the resulting solution was placed in a
cuvette, and the absorbance at 375 nm was measured versus
J ) 6.9 Hz, 2H), 4.36 (s, 2H), 5.86 (d, J ) 11.3 Hz, 1H), 6.31
MeCN and 200 µL of 1:1 CH
2
2
1
3
(
5
6
dt, J ) 11.3, 6.9 Hz, 1H); C NMR (125 MHz, DMSO-d
6
) δ
8.16, 67.85, 68.78, 69.0, 69.60, 69.74, 69.78, 69.81, 69.90,
+
9.95, 74.50, 77.89, 81.46, 110.10, 144.84; MS 339 (MH );
339.1808, found 339.1801.
a blank of 1:1 CH
2 2
Cl /MeCN. The percent picrate extracted
HRMS m/ e calcd for C18
27 6
H O
into the organic layer was calculated using the extinction
-
1
-1
Analytical data for (E)-16: 1H NMR δ 3.53 (m, 20H), 4.11
coefficients lithium picrate ꢀ375 18 600 M
cm , sodium
-1
-1
picrate ꢀ375 19 000 M cm , and potassium picrate ꢀ375 18 600
(
1
dd, J ) 4.3, 1.9 Hz, 2H), 4.26 (s, 2H), 6.0 (d, J ) 15.9 Hz,
-
1
cm-1.
+
M
H), 6.32 (dt, J ) 15.9, 4.3 Hz, 1H); MS 339 (MH ); HRMS
m/ e calcd for C18
Th er m olysis of 5. To a vacuum hydrolysis tube were
added 5 (280 µL, 51.3 mM in DMSO-d , 14.4 µmol) and 1,4-
26 6
H O 338.1729, found 338.1727.
Ack n ow led gm en t. We are grateful to the donors
of the Petroleum Research Fund, administered by the
American Chemical Society, and to the Robert A. Welch
foundation for partial support of this work.
6
cyclohexadiene (4.25 mL, 44.9 mmol). The solution was
homogenized, and oxygen was excluded by subjecting the
mixture to three freeze-pump-thaw cycles under argon. The
tube was heated in an oil bath at 135 °C for 45.5 h and allowed
to cool, and the solvent was removed in vacuo to afford a green
Su p p or tin g In for m a tion Ava ila ble: 1H NMR spectra for
all new compounds and HMBC NMR of compound (Z)-16 (13
pages). This material is contained in libraries on microfiche,
immediately follows this article in the microfilm version of the
journal, and can be ordered from the ACS; see any current
masthead page for ordering information.
1
oil that was a 2:1 mixture by H-NMR of noncyclized enediyne
5
to Bergman product 4 (3.2 mg, 22%). The identity of 4 was
1
13
verified by comparison to the literature H NMR values and
+
by MS: MS 341 (MH ); HRMS m/ e calcd for C18
41.1964, found 341.1966.
29 6
H O
3
J O961583R