Effects of shape on thermodynamic cyclizations of cinchona alkaloids

Article abstract of DOI:10.1021/jo982496x

Thermodynamic cyclizations of a series of cinchona alkaloid derivatives are investigated. An extended quinine monomer 2:HO-Ce-OMe is prepared and cyclized to produce a mixture of cyclic oligomers. Combining the extended monomer 2:HO-Ce-OMe with the previously reported cinchonidine monomer 1b:HO- Cc-OMe under thermodynamic control produces a library of quinine-derived macrocycles. Two quinidine-derived methyl ester monomers 10:HO-Cd-OMe and 16:HO-Ca-OMe are also reported. Both are preorganized to form cyclic dimers; upon carrying out the thermodynamic cyclization on a mixture of both monomers only a small percentage (5%) of the hetero-dimer is obtained. Thermodynamic cyclization of the corresponding cinchonidine derived methyl ester 1b:HO-Cc- OMe with 10:HO-Cd-OMe results in the self-sorting of the two diastereoisomers.

Full text of DOI:10.1021/jo982496x

5804
J . Org. Chem. 1999, 64, 5804-5814
Effects of Sh a p e on Th er m od yn a m ic Cycliza tion s of Cin ch on a
Alk a loid s
Stuart J . Rowan,*,† Dominic J . Reynolds, and J eremy K. M. Sanders*,‡
Cambridge Centre for Molecular Recognition, University Chemical Laboratory, Lensfield Road,
Cambridge CB2 1EW, U.K.
Received December 22, 1998
Thermodynamic cyclizations of a series of cinchona alkaloid derivatives are investigated. An
extended quinine monomer 2:HO-Ce-OMe is prepared and cyclized to produce a mixture of cyclic
oligomers. Combining the extended monomer 2:HO-Ce-OMe with the previously reported
cinchonidine monomer 1b:HO-Cc-OMe under thermodynamic control produces a library of
quinine-derived macrocycles. Two quinidine-derived methyl ester monomers 10:HO-Cd -OMe and
16:HO-Ca -OMe are also reported. Both are preorganized to form cyclic dimers; upon carrying
out the thermodynamic cyclization on a mixture of both monomers only a small percentage (5%) of
the hetero-dimer is obtained. Thermodynamic cyclization of the corresponding cinchonidine derived
methyl ester 1b:HO-Cc-OMe with 10:HO-Cd -OMe results in the self-sorting of the two
diastereoisomers.
In tr od u ction
binders of the substrate would be stabilized. This then
results in the library modifying its distribution, increas-
ing the concentration of macrocycles that are good
receptors for the substrate. To be able to investigate and
utilize such systems we need to understand the factors
that can influence the library’s diversity. We have shown
that thermodynamically controlled transesterification
reactions can lead to efficient synthesis of oligomeric
macrocycles based on quinine alkaloids10,11 and cholic
acid.4,12 When monomeric building blocks are suitably
predisposed, a single product may dominate even when
other oligomers are kinetically accessible;13 for example,
the thermodynamic cyclization of the quinine methyl
ester 1a :HO-Cq-OMe or cinchonidine methyl ester 1b:
HO-Cc-OMe gives almost exclusively the correspond-
A combinatorial library consists of many members
(M1...Mn ) each of which contains two or more building
blocks (A, B, C...) arranged in a particular way. In a
traditional combinatorial library1 the covalent bonds
between building blocks are fixed by using irreversible
chemistry during synthesis, so that there can be no
postsynthetic interconversion between members such as
ABC and ACB. However, in a dynamic combinatorial
library (DCL)2 the connections between building blocks
are reversible and in flux, continuously being made and
broken; these connections may be covalent bonds such
as imine,2a ester,3,4 disulfide,5 borate,6 or alkene linkages,7
or they may be noncovalent, utilizing metal-ligand8 or
hydrogen-bonding9 interactions. The composition of a
DCL will be dependent upon its environment, allowing
thermodynamic templating to take place. For example,
if a library of macrocycles is produced, then upon addition
of a particular substrate, macrocycles that are strong
† Current address: Department of Macromolecular Science, Case
Western Reserve University, 10900 Euclid Avenue, Cleveland, OH
44106-7202. Fax: (216) 368-4202.
‡ E-mail: jkms@cam.ac.uk.
(1) Wilson, S. R.; Czarnik, A. W. Combinatorial Chemistry. Synthesis
and Application; Wiley: New York, 1997.
(2) (a) Huc, I.; Lehn, J .-M. Proc. Natl. Acad. Sci. U.S.A. 1997, 94,
2106-2110. (b) Brady, P. A.; Sanders, J . K. M. Chem. Soc. Rev. 1997,
26, 327-336. (c) Rowan, S. J .; Sanders, J . K. M. Curr. Opin. Chem.
Biol. 1997, 1, 483-490. (d) Timmerman, P.; Reinhoudt, D. N. Adv.
Mater. 1999, 11, 71-74. (e) Ganesan, A. Angew. Chem., Int. Ed. 1998,
37, 2828-2831.
ing cyclic trimer.10 Predisposition must be carefully
distinguished from preorganization: The latter generally
refers to the ground state of a monomer whose conforma-
tion holds the reactive groups in close proximity, thereby
favoring one pathway over alternatives (Figure 1a).
Preorganization in covalent chemistry is therefore a
kinetic process. Predisposition, on the other hand, should
(3) Some of the extended quinine work has been reported in
communication form: Rowan, S. J .; Sanders, J . K. M. Chem. Commun.
1997, 1407-1408.
(4) Brady, P. A.; Sanders, J . K. M. J . Chem. Soc., Perkin Trans. 1
1997, 3237-3253.
(5) (a) Hioki, H.; Still, W. C. J . Org. Chem. 1998, 63, 904-905. (b)
Tam-Chang, S.-W.; Stehouwer, J . S.; Hao, J . J . Org. Chem. 1999, 64,
334-335.
(6) Giger, T.; Wigger, M.; Aude´tat, S.; Benner, S. A. Synlett 1998,
688-691.
(10) Rowan, S. J .; Brady, P. A.; Sanders, J . K. M. Angew. Chem.,
Int. Ed. Engl. 1996, 35, 2143-2145.
(11) Rowan, S. J .; Hamilton, D. G.; Brady, P. A.; Sanders, J . K. M.
J . Am. Chem. Soc. 1997, 119, 2578-2579.
(12) Brady, P. A.; Bonar-Law, R. P.; Rowan, S. J .; Suckling, C. J .;
Sanders, J . K. M. Chem. Commun. 1996, 319-320.
(13) Rowan, S. J .; Brady, P. A.; Sanders, J . K. M. Tetrahedron Lett.
1996, 37, 6013-6016.
(7) Kidd, T. J .; Leigh, D. A.; Wilson A. J . J . Am. Chem. Soc. 1999,
121, 1599-1600.
(8) Klekota, B.; Hammond, M. H.; Miller, B. L. Tetrahedron Lett.
1997, 38, 8639-8642.
(9) Calama, M. C.; Hulst, R.; Fokkens, R.; Nibbering, N. M. M.;
Timmerman, P. Reinhoudt, D. N. Chem. Commun. 1998, 1021-1022.
10.1021/jo982496x CCC: $18.00 © 1999 American Chemical Society
Published on Web 07/23/1999

Thermodynamic Cyclizations of Cinchona Alkaloids
J . Org. Chem., Vol. 64, No. 16, 1999 5805
Sch em e 1a
F igu r e 1. Schematic representation of a monomer (a) pre-
organized and (b) predisposed to form a cyclic trimer.
be thought of as a strong conformational or structural
preference expressed by the building block once it is
incorporated into a larger structure, giving rise to a
thermodynamic preference for a particular product (Fig-
ure 1b). This means that the final conformation in the
oligomer is not the same as, or dictated by, the ground-
state conformation of the free monomer. We have also
shown that such predisposition can be affected with the
addition of a flexible facilitator14 and thus increase the
size of the DCL. We now report the consequences of
altering the underlying alkaloid shape and geometry on
this predisposition of thermodynamic cyclizations.
a
Reagents and conditions: (a) TsCl, Et3N, CH2Cl2, rt; (b) NaI,
acetone, reflux; (c) methyl-p-hydroxybenzoate, K2CO3, 18-crown-
6, acetone, rt; (d) TBAF, THF, rt.
Resu lts a n d Discu ssion
the formation of macrocyclic over linear oligomers. A
range of macrocycles is observed, the mixture being
dominated by cyclic dimer (Ce2):trimer (Ce3):tetramer
(Ce4) quinine cyclophanes (Scheme 2), with a mass ratio
of 39:47:14, respectively, as observed by HPLC.15 Even
though there is still a preference for cyclic trimer, the
extension unit has allowed access to cyclic dimer (not
previously seen even in kinetic cyclizations of the 1a
derivative) and tetramer under thermodynamic control,
demonstrating the expected relaxation of predisposition.
1H NMR helps to explain this relaxation in distribution.
The 3J H H coupling constant is particularly useful in this
An Exten d ed Qu in in e Bu ild in g Block . Our first
attempt to try to relax the predisposition of quinine3 was
the addition of an arm at the C(11) position, to give the
extended quinine monomer 2:HO-Ce-OMe. The exten-
sion, methyl 4-hydroxybenzoate, serves several func-
tions: it is compatible with our transesterification con-
ditions; it programs into the molecule-added length to
increase the size of any cavity formed and allow access
to cyclic dimer; and it also introduces a little extra
flexibility, while still being rigid enough to prevent
formation of cyclic lactone monomer. Synthesis of 2:HO-
Ce-OMe started from the previously prepared C(11)-
alcohol 3. This was converted into the tosylate 4 with
tosyl chloride and triethylamine (77% yield), which on
heating with NaI in acetone gave 9-O-tert-butyldimeth-
ylsilyl-10-11-dihydro-11-iodo quinine 5 in 86% yield. 5
was converted into 6 (75%) by stirring with methyl
4-hydroxybenzoate, 18-crown-6, and K2CO3, and depro-
tection of the TBDMS group by using TBAF/THF gave
the extended monomer 2:HO-Ce-OMe in 78% yield
(Scheme 1).
The monomer was then thermodynamically cyclized by
using our transesterification conditions.10 The catalyst
(5-10% KOMe/18-crown-6) was added to a solution (5
mM) of 2:HO-Ce-OMe in toluene which was heated to
reflux through 4 Å sieves in a Soxhlet extractor to effect
clean macrocyclization within 10-20 min. The initial
solution of monomer in toluene is heated to reflux for 30-
60 min prior to the addition of the catalyst to ensure that
the reaction is as dry as possible. The Soxhlet extractor
apparatus permits the azeotropic removal of the metha-
nol produced on addition of the catalyst, thereby favoring
8
9
respect, allowing us to examine the gross conformation
of the quinine unit.16-18 3J H in the cyclic dimer (Ce2)
8H9
is 10.2 Hz, indicating that the quinine unit can adopt a
conformation in Ce2 similar to that of the predisposed
3
cyclic trimer (Cq3) in which 3J H
) 10.5 Hz. J H
in
8H9
8H9
Ce3 and the higher cyclic oligomers is 7.8 Hz, suggesting
that the quinine unit, in these molecules, can now adopt
the stable conformation similar to that of linear species18
which display similar J values.
In the thermodynamic (reversible) reaction, once a
product is formed it can be reconverted into other more
thermodynamically stable species, a proof-reading proc-
ess. However, this is not the case in a kinetic (irrevers-
ible) reaction where once a product is formed there is no
going back. We decided, therefore, to investigate the
cyclization of the extended acid alcohol monomer 7 under
kinetic control. 7 could easily be prepared from the
(15) Small amounts of higher oligomers are also observed.
(16) Dijkstra, G. D. H.; Kellogg, R. M.; Wynberg, H.; Svendsen, J .;
Marko, I.; Sharpless, K. B. J . Am. Chem. Soc. 1989, 111, 8069-8076.
(17) Dijkstra, G. D. H.; Kellogg, R. M.; Wynberg, H. J . Org. Chem.
1990, 55, 6121-6131.
(14) Rowan, S. J .; Lukeman P. S.; Reynolds; D. J .; Sanders, J . K.
M. New J . Chem. 1998, 22, 1015-1018.
(18) Rowan, S. J .; Sanders, J . K. M. J . Org. Chem. 1998, 63, 1536-
1546.

5806 J . Org. Chem., Vol. 64, No. 16, 1999
Rowan et al.
Sch em e 2a
a
Reagents and conditions: (a) KOMe, 18-crown-6, toluene, reflux.
Sch em e 3a
Sch em e 4a
a
Reagents and conditions: (a) 1. 2,6-dichlorobenzoyl chloride,
a
Reagents and conditions: (a) LiOH, THF/H2O, rt; (b) TBAF,
Et3N, DMF. 2. DMAP, CH2Cl2, rt, 18 h.
THF, rt.
thermodynamic cyclizations (where mainly cyclic trimer
Cq3 is observed). However, in the extended series 2:HO-
Ce-OMe the difference in macrocycle distribution is not
so stark. These observations imply no significant desta-
bilization of Ce4 and other oligomers relative to Ce3, and
thus no significant increase in the Ce4 reconversion rate
under thermodynamic conditions. This is not the case
with 1a :HO-Cq-OMe, where while the cyclization rates
are comparable for the Cq oligomers relative to Cq3 (this
is demonstrated by the distribution observed in the
kinetic formation of these macrocycles), the stability of
Cq4 and higher oligomers is less. This results in an
unzipping of the Cq higher oligomers (i.e., increased
reconversion rate) and ultimately an increased distribu-
tion of Cq3.
TBDMS-protected extended methyl ester 6 in two steps
(Scheme 3). Ester hydrolysis with LiOH in THF/H2O
followed by deprotection of the TBDMS group with TBAF
in THF affords 7 in a 13% overall yield for the two steps.
Kinetic cyclization of this monomer was carried out, at 5
mM, using a modified Yamaguchi macrolactonization
procedure13 (Scheme 4). As in the thermodynamic cy-
clization, a range of macrocycles is obtained with cyclic
dimer, trimer, and tetramer being the dominant products;
however, the distribution is slightly different (Ce2:Ce3:
Ce4 ratio is 46:43:11. cf. 39:47:14 for thermodynamic
cyclizations at 5 mM). It is interesting, at this point, to
compare the thermodynamic and kinetic cyclizations of
both the quinine 1a :HO-Cq-OMe and extended qui-
nine 2:HO-Ce-OMe derivatives. For 1a :HO-Cq-OMe
there is a large difference observed between the kinetic
cyclizations (where a range of macrocycles is formed) and
The xanthene hydroxy ester 8 is a molecule specifically
designed to have the ester and hydroxyl functional groups
arranged in a parallel fashion on a rigid framework, to

Thermodynamic Cyclizations of Cinchona Alkaloids
J . Org. Chem., Vol. 64, No. 16, 1999 5807
Sch em e 5a
F igu r e 2. (a) Quinidine and (b) quinidine methyl ester 10:
HO-Cd -OMe.
Sch em e 6a
a
Reagents and conditions: (a) KOMe, 18-crown-6, toluene,
reflux.
give a molecule preorganized to form only cyclic dimer.
Under thermodynamic control, mixtures of cinchonidine
monomer 1b:HO-Cc-OMe and preorganized xanthenes
self-sort to give good yields of cyclic trimer and dimer,
respectively, because the homo-products are stabilized.11
To see if the extension unit has any effect on mixed
thermodynamic cyclizations with the xanthene 8, we
submitted the xanthene monomer 8 and the extended
quinine 2 to the thermodynamic cyclization (Scheme 5)
at a total monomer concentration of 5 mM (2.5 mM of
each monomer). Once again self-selection was observed:
only the xanthene dimer and homo-extended quinine
macrocycles were formed. The Ce2:Ce3:Ce4 ratio observed
is 53:39:8, which is similar to the distribution observed
for the cyclization of 2, carried out at 2.5 mM (52:42:6).
This change in macrocycle distribution highlights just
how sensitive to concentration the final product mixture
is; e.g., at 5 mM, cyclic trimer (Ce3) is the slightly more
favored product, but upon reducing the concentration to
2.5 mM, cyclic dimer (Ce2) becomes the more favored
product. Formation of larger amounts of smaller oligo-
mers at lower concentration is predicted by theoretical
considerations.19
Qu in id in e-Der ived Bu ild in g Block s. Having ex-
amined the effect that an extension unit has on the
product distribution of a monomer, we wanted to explore
the effect of a radical shape change within the context of
essentially identical chemistry. We therefore prepared
the methyl ester derivative of quinidine 10:HO-Cd -
OMe. The inverted stereochemistry (Figure 2) at the C(8)
and C(9) positions (but not at the C(3) position) was
expected to lead to a different product distribution.
Synthesis of 10:HO-Cd -OMe was achieved by using
a procedure similar to that reported previously (Scheme
6) for the quinine derivative.18 The C(9)-hydroxyl of
quinidine 11 was protected as the TBDMS ether 12,
prepared in a 97% yield. Hydroboration of 12, followed
by oxidation with trimethylamine N-oxide dihydrate,
gave the alcohol 10 in 81% yield.20 Oxidation of the
terminal alcohol 10 was carried out by employing J ones
reagent. The resulting carboxylic acid 14, formed in 62%
a
Reagents and conditions: (a) TBDMSCl, Et3N, DMAP, DMF,
rt; (b) BH3.THF (5 equiv), diglyme, 0 °C then Me3NO‚2H2O, 100
°C; (c) J ones reagent, acetone, rt; (d) HCl(concentrated), MeOH, rt; (e)
TBAF, THF, rt.
yield, was esterified with MeOH/HCl to give the methyl
ester 15 (92%). Finally, the protecting group was removed
by using TBAF/THF to yield the monomer 10:HO-Cd -
OMe in 67% yield. The overall yield for the five steps is
30%. As in the quinine series,18 the C(9) position in the
quinidines seems to be rather hindered as indicated by
1
the H NMR spectra of the TBDMS ether intermediates:
the TBDMS tert-butyl group resonance is split into two
peaks, indicative of two slowly interconverting conforma-
tions, of relative populations of approximately 3:1. Once
the bulky TBDMS group is removed only one conforma-
tion is observed.
Using a method similar to that used before, the
monomer 10 was thermodynamically cyclized (Scheme
7). The catalyst (5-10% KOMe/18-crown-6) was added
to a solution (5 mM) of 10:HO-Cd -OMe in refluxing
toluene. Electrospray mass spectrometry (ES-MS) was
used to follow the extent of the reaction, and by HPLC
analysis the reaction appears to have yielded almost
exclusively cyclic dimer Cd 2, although up to 5% of cyclic
trimer Cd 3 is observed in both the ES-MS and HPLC.
(19) (a) Ercolani, G.; Mandolini, L.; Mencarelli, P.; Roelens, S. J .
Am. Chem. Soc. 1993, 115, 3901-3908. (b) Ercolani, G. J . Phys. Chem.
B 1998, 102, 5699-5703.
(20) Commercial quinidine from Aldrich or Lancaster contains 10%
dihydroquinidine. The yields reported are based on the actual amounts
of quinidine used.
3
1
The J H
coupling observed in the H NMR spectra
8H9
of 10:HO-Cd -OMe and Cd 2 is less than 3 Hz, which is
indicative of a dihedral angle that does not change

5808 J . Org. Chem., Vol. 64, No. 16, 1999
Rowan et al.
Sch em e 7a
Sch em e 8a
a
Reagents and conditions: (a) KOMe, 18-crown-6, toluene,
reflux.
a
Reagents and conditions: (a) Tosyl chloride, Et3N, rt 4 h; (b)
NaI, acetone, reflux, 16 h; (c) methyl 4-hydroxybenzoate, 18-crown-
6, K2CO3, rt, 16 h; (d) TBAF, THF, rt, 4 h.
Sch em e 9a
3
F igu r e 3. The two different J H values in HO-Cd 2-OMe.
H
8
9
dramatically from around 90° on macrocyclization. In
other words, the conformation that the quinidine unit
adopts in the macrocycle is similar to that in the
monomer, so that it might not seem too surprising then
that the dimer is very much preferred in this case. It
should be noted, however, that there are two different
3J H
values in the linear quinidine dimer HO-Cd 2-
8H9
OMe (Figure 3), which is, presumably, the species which
undergoes the cyclization to form cyclic dimer Cd 2. The
3J H at the terminal alcohol has a value similar to that
8H9
of the quinidine monomer 10:HO-Cd -OMe, whereas
the 3J H
at the central ester has now changed to 6.4
8H9
Hz. This change in conformation of the quinidine unit
containing an ester on the C(9)-hydroxyl is to be expected
a
3
(cf. quinidine acetate J H H ca. 7 Hz).16 This would imply
Reagents and conditions: (a) KOMe, 18-crown-6, toluene,
8
9
reflux.
that upon cyclization a conformational change in one of
the quinidine units is required, which would not be the
case if the quinidine linear dimer HO-Cd 2-OMe was
preorganized.
of the quinoline unit and the TBDMS group. These were
exacerbated by the addition of the tosylate/extension unit,
where the resonance of the C(9) proton becomes so broad
that it nearly disappears in the spectra of 17 and 19;
again, this is believed to be an exchange phenomenon.
As before, deprotection at the C(9)-hydroxyl resulted in
a much sharper proton spectrum, as the previously split
peaks coalesced to give rotationally averaged signals.
Cyclization of 16:HO-Ca -OMe was carried out as
before and the reaction mixture was analyzed by HPLC.
Despite the extra flexibility and size of 16:HO-Ca -
OMe, the cyclization yielded almost exclusively dimer
Ca 2 (>85% of the cyclized material), as shown in Scheme
9; again, the rest of the material is thought to be cyclic
trimer (which is observed by ES-MS). This does not
mirror the results obtained for the quinine derivatives
described above, where the extension unit allowed a large
relaxation of the predisposition properties of 1a :HO-
Cq-OMe. Thus, it would appear that 16:HO-Ca -OMe
is also preorganized to give the dimeric macrocycle; in
this case the phenoxy unit does not confer enough extra
conformational freedom to overcome the strong prefer-
The result of the cyclization of monomer 10:HO-Cd -
OMe implies that this molecule is more preorganized to
give cyclic dimer. In an attempt to create a more
promiscuous monomer, that is, capable of the required
diversity, we added the phenoxy extension group to give
an extended quinidine monomer. The required monomer
16:HO-Ca -OMe was prepared in a way similar to that
used before (Scheme 8), using the terminal alcohol 13 as
a starting point. The primary alcohol was converted into
the tosylate 17, in 77% yield. The tosylate was subse-
quently displaced by iodide to form intermediate 18
(79%). Combination of 18 with methyl 4-hydroxybenzoate
and potassium carbonate gave 19 (98%), which upon
deprotection with TBAF yielded the monomer 16:HO-
Ca -OMe (45%). The overall yield from the quinidine
natural product was 21% for the six steps.
Once again, the NMR spectra of compounds 17-19
were complicated by the appearance of two conforma-
tional isomers, probably resulting from the steric bulk

Thermodynamic Cyclizations of Cinchona Alkaloids
J . Org. Chem., Vol. 64, No. 16, 1999 5809
Sch em e 10a
F igu r e 4. Electrospray mass spectra of the thermodynamic
cyclization of 2b: HO-Cc-OMe and 22:HO-Ce-OMe.
+
Peaks marked (X) are NH4 adducts of molecular ions.
a
Reagents and conditions: (a) KOMe, 18-crown-6, toluene,
reflux.
Sch em e 11a
ence to form dimer, and the product distribution is not
greatly altered. In retrospect, this observation may not
be too surprising since the phenoxy unit acts primarily
as an extension to C(11), adding extra length in the form
of the aromatic ring; there is only one extra degree of
freedom (the C(11)-O bond) in 16:HO-Ca -OMe as
compared to 10:HO-Cd -OMe.
Mixin g Exp er im en ts
We now had four cinchona alkaloid derivatives for
thermodynamic mixing experiments. We have already
carried out a mixing experiment using two cinchona
alkaloids; upon transesterification of 1a :HO-Cq-OMe
and 1b:HO-Cc-OMe, each 2.5 mM, we observe all four
possible cyclic trimers, Cc3, Cc2Cq, CcCq2, Cq3, in the
expected 1:3:3:1 ratio.10 We, therefore, could now examine
the effect that addition of an extension unit and/or change
of shape has on the product distribution by mixing either
1a or 1b with one of the other cinchona alkaloid mono-
mers. Combination of extended monomer 2:HO-Ce-
OMe with predisposed 1b:HO-Cc-OMe allowed us to
examine the effect of the extension unit. Thermodynamic
transesterification of 2:HO-Ce-OMe and 1b:HO-Cc-
OMe, under the usual conditions, led to a combinatorial
library of macrocycles (Scheme 10), as observed by
electrospray mass spectrometry (Figure 4). The mass
spectrum contains two of the three possible dimers, all
possible trimers and small amounts of all the possible
tetramers. However, it is unwise to draw precise quan-
titative conclusions from peak intensities since individual
species have different inherent detectabilities in electro-
spray mass spectra,21 but there are systematic and
interpretable deviations from the expected 2:1 statistical
distribution of hetero:homooligomers. The extended qui-
nine homodimer is seemingly favored over the hetero-
dimer, which is not surprising given that 1b:HO-Cc-
OMe cannot form the homodimer. The trimers show the
expected weighting toward the cinchonidine homotrimer
but the more relaxed monomer 2:HO-Ce-OMe gives
access to two new heterotrimers. To a first approxima-
tion, it appears that the tetramers are formed in statisti-
cal proportions. It is important, however, to note that
a
Reagents and conditions: (a) KOMe, 18-crown-6, toluene,
reflux, 1 h.
from the electrospray data alone we cannot tell the
relative quantities of cyclic dimers, trimers, or tetramers.
These results demonstrate that it is possible to gener-
ate combinatorial libraries of macrocyclic receptors, even
when one of the building blocks is predisposed to produce
just a single product when cyclized alone. By mixing only
two monomers we observe eleven cyclic products, indicat-
ing that large libraries of macrocycles can be obtained
quickly from a small number of monomers.
We also examined the effect that shape change has on
the product distribution by mixing the two diastero-
isomers. When cinchonidine 1b:HO-Cc-OMe and qui-
nidine 10:HO-Cd -OMe monomers are cyclized to-
gether, they self-sorted into the cinchonidine cyclic trimer
Cc3 and quinidine dimer Cd 2 (Scheme 11).
This could be thought of as diastereomeric self-sorting
and emphasizes the importance of shape in thermody-
namic chemistry.
Mixing quinine 1a :HO-Cq-OMe with extended qui-
nidine 16:HO-Ca -OMe allowed us to examine what
would happen to the diversity when an extension unit
was put on the quinidine monomer. The result was that
Ca 2 and Cq3 were the major products, although the
extension allowed access to reasonable amounts of the
(21) Brady, P. A.; Sanders, J . K. M. New J . Chem. 1998, 22, 411-
417.

5810 J . Org. Chem., Vol. 64, No. 16, 1999
Rowan et al.
Sch em e 12a
Cd -OMe and 16:HO-Ca -OMe under the conditions
used for the previous cyclization reactions. The cycliza-
tion reaction involving both 10:HO-Cd -OMe and 16:
HO-Ca -OMe was found to give the two homodimers
as the major product, but about 5% of the heterodimer
was detected by HPLC analysis, as depicted in Scheme
14. The rest of the material is believed to correspond to
the cyclic homotrimers. A similar ratio of homodimer to
homotrimer is observed in both the mixing and single
monomer experiments for each of the quinidine mono-
mers.
In both of our previous cases of self-sorting, the 1b:
HO-Cc-OMe monomer unit with either the xanthene
building block11 or the quindine monomer 10:HO-Cd -
OMe, the individual monomers cyclized to yield different
oligomers: 1b:HO-Cc-OMe yields trimer, while the
xanthene and quinidine monomer both yield dimer. This,
combined with the relative rigidity of the monomers,
results in a greater stability of the two homooligomers
relative to that of the heterodimers or trimers, dictating
that very little of the latter would survive the proof-
reading (or reconverting) process. In the case of 10:HO-
Cd -OMe and 16:HO-Ca -OMe the similarity of the
two monomer units means that both favor cyclic dimer
formation. The slight relaxation of rigidity in 16:HO-
Ca -OMe has reduced the difference in thermodynamic
stability between the homo- and heterodimers, thereby
hindering total self-sorting. However, the different bite-
sizes of these monomers prevent a statistical distribution
being observed.
Reagents and conditions: (a) KOMe, 18-crown-6, toluene, reflux.
Sch em e 13a
The cinchona alkaloid mixing experiments highlight
a range of diversity that can be achieved in the thermo-
dynamic cyclization of two monomers, ranging from self-
sorting (as observed with 1b:HO-Cc-OMe and 10:HO-
Cd -OMe) to complete mixing (as observed, for example,
with 2:HO-Ce-OMe with 1b:HO-Cc-OMe). The de-
gree of mixing will obviously rely upon the conformational
stability of the heterooligomers when compared to that
of the homooligomers. Furthermore, mixing will also be
dependent on the bite-size of the monomer.22 The ques-
tion of bite-size is illustrated in Figure 5. If the monomers
are too rigid and have different bite-sizes, then strain
will be observed in any mixed oligomers, thus rendering
such heterooligomers thermodynamically unfavorable
and inducing self-sorting. In contrast, if the bite-sizes of
the two monomers are similar, then mixing will occur.
One way to overcome the bite-size problem is to increase
the amount of flexibility in at least one of the monomers.
Flexibility helps with diversity in at least two ways. First,
it bestows greater freedom to any heterodimer, allowing
it to adopt a more energetically favorable conformation.
Moreover, flexibility permits the bite-size of the monomer
to alter without paying too much of a penalty in energy,
and in both scenarios mixing is facilitated.
a
Reagents and conditions: (a) KOMe, 18-crown-6, toluene,
reflux, 1 h.
heterodimer and trimers, as well as detectable amounts
of Ca 3 and tetramers (Scheme 12). This might be due to
the fact that one monomer is predisposed to form trimer,
while the other prefers to form dimer. This, along with
the slight relaxation of dimer formation in the extended
quinidine, results in less preference for dimer formation
and allows access to a slightly wider range of macrocycles.
Mixing of the apparently more promiscuous extended
quinine 2:HO-Ce-OMe with the quinidine 10:HO-
Cd -OMe gave access to only a small amount of mixing.
The major products are the two homodimers Ce2 and
Cd 2, and it could be argued that because both monomers
prefer dimer, this limits the stability of other mixed,
nondimeric products. In fact, the other major products
are the heterodimer Cd Ce and the two cyclic trimers that
contain high proportions of the extended quinine, Ce3 and
Ce2Cd (Scheme 13). It is interesting to note that, while
putting the extension on either cinchona alkaloid in-
creases diversity, in both reactions, the degree of mixing
is different. This might be due to the quinidine unit being
too preorganized to encourage diversity.
Con clu sion s
We have now succeeded in showing that the quinidine
derivatives can be synthesized and cyclized under the
transesterification conditions described. The results con-
trast with those previously obtained for the diastereo-
isomer quinine, in that now the predominant cyclic
species formed is the cyclic dimer. Furthermore, and
Finally, we then examined the thermodynamic cycliza-
tion of a mixture of both quinidine monomers 10:HO-
(22) A similar argument based upon bite-angle can also be invoked
here.

Thermodynamic Cyclizations of Cinchona Alkaloids
J . Org. Chem., Vol. 64, No. 16, 1999 5811
Sch em e 14a
a
Reagents and conditions: (a) KOMe, 18-crown-6, toluene, reflux, 1 h.
Exp er im en ta l Section
None of the yields have been optimized. 1H NMR and 13C
NMR were obtained at either 250 or 400 and 63 or 100 MHz,
respectively. Fast atom bombardment (FAB) mass spectra
were obtained using a m-nitrobenzyl alcohol matrix and
positive-ion electrospray mass spectra (ES-MS) were obtained
using 80 V sampling cone voltage (Vc). Samples were intro-
duced into the mass spectrometer source with an LC pump
and a flow rate of 4 mL min-1 of MeCN/H2O (1:1). HPLC
separations were carried out on a µBondapak C18 3.9 × 300
mm reverse phase column using a mobile phase with a flow
rate of 1 mL min-1, pump A 0.05 M n-hexylamine, H3PO4 (pH
) 3), pump B MeCN and detection by a diode array UV
detector.
9-O-ter t-Bu tyld im eth ylsilyl-10-11-dih yd r o-11-tosyl-qu i-
n in e (4). To a solution of the alcohol 3 (3.0 g, 6.8 mmol) in
DCM (20 mL) was added tosyl chloride (1.94 g, 10.2 mmol)
and Et3N (1.75 mL, 12.6 mmol). The solution was allowed to
stir overnight and worked up by adding ethyl acetate, washing
with water (×3), and drying over magnesium sulfate. The ethyl
acetate was removed under vacuum and the remaining oil
purified by flash column chromatography ethyl acetate/
methanol (10:0, 9:1, ..., 6:4) to yield the product (3.09 g, 77%).
TLC ethyl acetate/methanol (8:2) Rf ) 0.51. 1H NMR (400
MHz, CDCl3) δ ) 8.72 (d, J ) 4.4 Hz, 1H), 7.99 (d, J ) 9.2 Hz,
1H), 7.68 (d, J ) 8.2 Hz, 2H), 7.58 (m, 1H), 7.47 (d, J ) 4.5
Hz, 1H), 7.36 (dd, J ) 2.5, 9.6 Hz, 1H) 7.29 (d, J ) 8.2 Hz,
2H), 6.50 (brs, 1H), 4.01 (s, 3H), 3.89 (m, 2H), 3.49 (m, 1H),
3.17 (m, 2H), 2.80 (m, 1H), 2.41 (s, 3H), 1.81-2.35 (m, 5H),
1.80 (m, 1H), 1.57 (m, 2H), 1.41 (m, 1H) 0.98 (s, 9H), 0.27 (s,
3H), -0.37 (s, 3H); 13C NMR (100 MHz, CDCl3) δ ) 159.4 (s),
146.7 (d), 145.3 (s), 144.6 (s), 140.2 (s), 132.5 (s), 131.9 (d),
130.0 (d), 127.8 (d), 125.7 (s), 123.4 (d), 118.9 (d), 100.2 (d),
67.8 (t), 67.3 (t), 60.4 (d), 57.5 (q), 56.1 (t), 42.9 (t), 33.2 (t),
30.3 (d), 25.9 (q), 24.7 (d), 21.7 (q) 18.1 (t), 18.0 (s), -4.7 (q),
-5.2 (q). (Due to different conformations, the C(9) carbon was
too broad to be observed). MS (FAB) 611.29480 (C33H47O5N2-
SiS requires 611.2975).
F igu r e 5. Schematic representation of how incompatible “bite
size” can lead to self-sorting of building blocks in thermody-
namically controlled chemistry: (a) a mixture of large and
small building blocks, (b) mixed dimer exhibiting bond and/or
angle strain, (c) self-sorted mixture of strain-free dimers, and
(d) the use of a small diversifier to facilitate formation of mixed
oligomers.
perhaps more significantly, there appears to be little or
no conformational change on macrolactonization sug-
gesting these quinidine derivatives are more preorgan-
ized, rather than predisposed. We have also demon-
strated that thermodynamic cyclization of cinchonidine
1b:HO-Cc-OMe and quinidine 10:HO-Cd-OMe leads
to self-sorting into cinchonidine trimer Cc3 and quinidine
dimer Cd 2-diastereomeric sorting. Finally, we have
demonstrated that this predisposition can be reduced to
give mixtures of oligomers, with addition of an extension
unit to the quinine molecule (which results in 2:HO-
Ce-OMe) and that libraries of cinchona alkaloid mac-
rocycles can be quickly generated. However, a more
general solution to the diversity problem may lie in the
use of small facilitator molecules.14
9-O-ter t-Bu tyld im eth ylsilyl-10-11-d ih yd r o-11-iod o-qu i-
n in e (5). 4 (2.0 g, 3.28 mmol) was dissolved in acetone (15
mL) in a flask equipped with a condenser under an inert
atmosphere and NaI (1.0 g, 6.67 mmol) added. The solution

5812 J . Org. Chem., Vol. 64, No. 16, 1999
Rowan et al.
was then heated to reflux for 2-3 h. The mixture was worked
up by adding ethyl acetate, washing with water (×3), and
drying over magnesium sulfate. The ethyl acetate was removed
under vacuum and the remaining oil purified by flash column
chromatography ethyl acetate/methanol (9:1) to yield the
product (1.6 g, 86%). TLC ethyl acetate/methanol (8:2) Rf )
0.67. 1H NMR (400 MHz, CDCl3) δ ) 8.71 (d, J ) 4.5 Hz, 1H),
7.99 (d, J ) 9.2 Hz, 1H), 7.47-7.49 (m, 2H), 7.38 (dd, J ) 2.5,
9.2 Hz, 1H), 6.62 (brs, 1H), 4.12 (s, 3H), 4.05 (m, 1H), 3.54 (m,
1H), 3.23-3.40 (m, 2H), 2.97-3.02 (m, 2H), 2.78 (m, 1H), 1.80-
2.40 (m, 5H), 1.68 (m, 2H), 1.20 (m, 1H), 0.97 (s, 9H), 0.38 (s,
3H), -0.38 (s, 3H); 13C NMR (100 MHz, CDCl3) δ ) 159.5 (s),
146.6 (d), 144.7 (s), 143.6 (s), 132.0 (d), 125.6 (s), 123.4 (d),
119.2 (d), 100.4 (d), 67.8 (d), 60.6 (d), 59.3 (q), 56.0 (t), 43.2 (t),
37.1 (t), 34.2 (d), 26.0 (q), 24.8 (t), 24.5 (d), 19.0 (s*), 18.0 (t*),
-3.6 (q), -4.2 (q). (Due to different conformations, the C(9)
carbon was to broad to be observed). MS (FAB) 567.1898
(C26H40O2N2SiI requires 567.19055).
Cycliza tion of Mon om er . 2 (11.2 mg, 2.35 × 10-5 moles)
was added to a round-bottomed flask attached to a Soxhlet
extractor which contained molecular sieves (4 Å) and then
dissolved in toluene (4.7 mL). This was heated to reflux for 30
min to remove all of the water from the system, and then the
KOMe‚18-C-6 catalyst solution (18 µL, 0.066 M, 1.18 × 10-6
moles) was added. To work up the reaction, the mixture was
added to aqueous pH 7 buffer and extracted with ethyl acetate.
The organic solvent was then removed and the resulting
mixture analyzed by HPLC and ES-MS, indicating the pres-
ence of mainly cyclic dimer, trimer, and tetramer. After
purification by column chromatography ethyl acetate/methanol
(10:0, 9:1, ..., etc.) samples of dimer and trimer could be
isolated. HPLC (reverse phase: time/%pump A ) 0/80, 25/30)
Rt ) 12.874, 14.573, 15.762 (main peaks); ES-MS 889 (Ce2,
MH+), 1333 (Ce3, MH+), 1777 (Ce4, MH+).
Cyclic d im er (Ce2): TLC ethyl acetate/methanol (1:1) Rf
) 0.22. HPLC (reverse phase: time/%pump A ) 0/80, 25/30)
1
Meth yl 9-O-ter t-Bu tyld im eth ylsilyl-10-11-d ih yd r oqu i-
n in e-11-(p h en oxy-p-ca r boxyla te) (6). To a solution of 5 (0.5
g, 0.88 mmol) in acetone (5 mL) was added methyl-4-hydroxy-
benzoate (0.269 g, 1.77 mmol), 18-crown-6 (0.466 g, 1.77 mmol)
and K2CO3 (0.244 g, 1.77 mmol), and the mixture was left to
stir overnight. To work up the reaction ethyl acetate was
added, and the solution was washed with water (×3) and then
dried over magnesium sulfate. The ethyl acetate was removed
under vacuum and the remaining oil purified by flash column
chromatography ethyl acetate/methanol (9:1) to yield the
product (0.39 g, 76%). TLC ethyl acetate/methanol (8:2) Rf )
0.53. 1H NMR (400 MHz, CDCl3) δ ) 8.74 (d, J ) 4.5 Hz, 1H),
8.01 (d, J ) 9.2 Hz, 1H), 7.91 (d, J ) 8.9 Hz, 2H), 7.51 (d, J )
4.5 Hz, 1H), 7.37 (dd, J ) 2.4, 9.2 Hz, 1H), 7.18 (brs, 1H), 6.78
(d, J ) 8.9 Hz, 2H), 5.64 (brs, 1H), 4.02 (m, 3H), 3.89 (m, 2H),
3.84 (s, 3H), 1.50-3.50 (m, 13H), 0.94 (s, 9H), 0.18 (s, 3H),
-0.37 (s, 3H). 13C NMR (100 MHz, CDCl3) δ ) 166.8 (s), 162.7
(s), 156.6 (s), 147.4 (d), 147.2 (s), 144.5 (s), 131.9 (d), 131.6 (d),
126.1 (s), 122.6 (s), 121.4 (d), 118.8 (d), 113.9 (d), 100.5 (d),
72.1 (br), 66.5 (t), 60.7 (d), 57.9 (t), 55.4 (q), 51.8 (q), 43.1 (t),
34.1 (t), 32.2 (d), 28.7 (t), 26.0 (q), 25.7 (d), 21.0 (t), 18.0 (s),
-4.7 (q), -5.2 (q). MS (FAB) 591.32540 (C34H47O5N2Si requires
591.3254).
Rt ) 12.87; H NMR (400 MHz, CDCl3): δ ) 8.77 (d, J ) 4.6
Hz, 2H), 8.01 (d, J ) 9.2 Hz, 2H), 8.00 (d, J ) 8.9 Hz, 4H),
7.60 (d, J ) 2.6 Hz, 2H), 7.46 (d, J ) 4.6 Hz, 2H), 7.36 (dd, J
) 2.6, 9.2 Hz, 2H), 6.96 (d, J ) 8.9 Hz, 4H), 6.52 (d, J ) 10.2
Hz, 2H), 4.26 (m, 4H), 3.90 (s, 6H), 3.63 (m, 2H), 3.15 (m, 2H),
3.03 (dd, J ) 9.6, 13.6 Hz, 2H), 2.67 (m, 2H), 2.27 (m, 2H),
1.97-2.25 (m, 4H), 1.86 (m, 2H), 1.451.75 (m, 10H); 13C NMR
(100 MHz, CDCl3): δ )165.6 (s), 162.4 (s), 157.8 (s), 147.5 (d),
144.9 (s), 131.8 (d), 127.8 (s), 122.1 (s), 121.5 (d), 114.6 (d),
102.0 (d), 73.4 (d), 67.4 (t), 60.2 (d), 57.6 (t), 55.6 (q), 42.0 (t),
33.8 (d), 32.8 (t), 28.2 (t), 24.4 (t), 24.4 (d). ES-MS 889 (MH+).
Cyclic tr im er (Ce3): TLC ethyl acetate/methanol (1:1) Rf
) 0.19. HPLC (reverse phase: time/%pump A ) 0/80, 25/30)
1
Rt ) 14.57; H NMR (400 MHz, CDCl3): δ ) 8.74 (d, J ) 4.5
Hz, 3H), 7.99-8.03 (m, 9H), 7.55 (d, J ) 2.6 Hz, 3H), 7.44 (d,
J ) 4.5 Hz, 3H), 7.36 (dd, J ) 2.6, 9.2 Hz, 3H), 6.90 (d, J )
8.9 Hz, 6H), 6.70 (d, J ) 7.8 Hz, 2H), 3.98-4.03 (m, 6H), 3.97
(s, 9H), 3.53 (m, 3H), 3.09-3.20 (m, 6H), 2.68 (m, 3H), 2.36
(m, 3H), 1.44-2.03 (m, 30H); 13C NMR (100 MHz, CDCl3): δ
)165.4 (s), 163.1 (s), 158.0 (s), 147.5 (d), 144.9 (s), 131.8 (d),
127.3 (s), 122.1 (s), 121.8 (d), 114.2 (d), 101.7 (d), 73.4 (d), 66.9
(t), 59.7 (d), 58.1 (t), 55.7 (q), 42.4 (t), 33.6 (d), 32.9 (t), 28.2
(t), 26.7 (t), 24.2 (d). ES-MS 1333 (MH+).
Meth yl 10-11-Dih yd r oqu in in e-11-(p h en oxy-p-ca r box-
yla te) (2). 6 (77 mg, 0.13 mmol), TBAF (60 µL, 1M in THF,
0.6 mmol), and THF (1 mL) were stirred overnight. The THF
was then removed under vacuum, and the resulting oil was
dissolved in dichloromethane and washed with water. Upon
removal of the dichloromethane the remaining oil was purified
by flash column chromatography ethyl acetate/methanol (9:1,
8:2, 7:3) to yield the product (39 mg, 63%). TLC ethyl acetate/
methanol (6:4) Rf ) 0.23. HPLC (reverse phase: time/%pump
10-11-Dih yd r o-11-(p h en oxy-p-ca r boxy)-qu in in e (7). To
a solution of 6 (5 g, 8.47 mmol) in THF/methanol (66 mL/20
mL) was added an aqueous solution of LiOH (1 g in 22 mL
H2O). After the reaction was stirred at room temperature for
24 h, the mixture was carefully neutralized and extracted into
ethyl acetate (×3). The organic solvent was then removed, and
the product could be purified by column chromatography ethyl
acetate/methanol (10:0, 9:1, 8:2, ..., 0:10) to yield a white solid
(2.44 g, 50%) which could be crystallized from chloroform/
methanol mixtures. 1H NMR (400 MHz, CDCl3) δ ) 8.74 (d, J
) 4.0 Hz, 1H), 7.96-8.02 (m, 3H), 7.49-7.53 (m, 2H), 7.34 (m,
1H), 6.73 (m, 2H), 6.27 (brs, 1H), 3.85 (m, 2H), 3.80 (s, 3H),
3.53 (m, 1H), 3.08-3.20 (m, 2H), 2.79 (m, 1H), 1.40-2.20 (m,
9H), 0.97 (s, 9H), 0.20 (s, 3H), -0.37(s, 3H); 13C NMR (100
MHz, CDCl3) δ ) 172.0 (s), 160.8 (s), 158.8 (s), 146.9 (d), 146.0
(s), 144.4 (s), 131.8 (d), 131.4 (d), 128.1 (s), 125.9 (s), 122.5 (d),
118.9 (d), 113.3 (d), 100.4 (d), 69.4 (d), 65.5 (t), 59.8 (d), 56.5
(t), 56.2 (q), 42.1 (t), 34.1 (t), 31.5 (d), 26.4 (t), 26.0 (q), 25.8
(d), 18,9 (t), 18.0 (s), -4.1 (q), -5.2 (q). ES-MS 577 (MH+). The
TBDMS was removed in the usual way. TBDMS-protected
quinine (0.42 g, 0. 73 mmol) and TBAF (2 mL, 1 M in THF, 2
mmol) in THF (10 mL) was stirred overnight. The mixture was
worked up by removal of the THF, followed by trituation with
diethyl ether to yield 7 as a white powder (83 mg, 25%). 1H
NMR (400 MHz, CD3SOCD3, C5D5N) δ ) 8.83 (d, J ) 4.5 Hz,
1H), 8.12 (d, J ) 8.8 Hz, 1H), 8.06 (d, J ) 9.2 Hz, 2H), 7.68-
7.73 (m, 2H), 7.41 (dd, J ) 2.6, 9.2 Hz, 1H), 7.00 (d, J ) 8.8
Hz, 2H), 5.70 (d, J ) 5.4 Hz, 1H), 3.98 (m, 2H), 3.88 (s, 3H),
3.38 (m, 1H), 3.35 (m, 1H), 3.03 (m, 1H), 2.61 (m, 1H), 1.86
(m, 1H), 1.65-1.73 (m, 7H), 1.36 (m, 1H); 13C NMR (100 MHz,
CDCl3) δ ) 168.2 (s), 161.9 (s), 157.2 (s), 147.6 (d), 144.3 (s),
131.4 (d), 127.1 (s), 121.0 (d), 119.1 (d), 113.9 (d), 102.5 (d),
70.5 (d), 66.3 (t), 60.3 (d), 57.1 (t), 55.3 (q), 42.0 (t), 33.6 (t),
31.9 (d), 27.6 (t), 25.8 (d), 22.6 (t). MS (FAB) 463 (MH+).
1
A ) 0/90, 40/60) Rt ) 20.38; H NMR (400 MHz, CDCl3) δ )
8.65 (d, J ) 4.5 Hz, 1H), 7.97 (d, J ) 9.2 Hz, 1H), 7.93 (d, J )
8.9 Hz, 2H), 7.50 (d, J ) 4.5 Hz, 1H), 7.31 (dd, J ) 2.6, 9.2 Hz,
1H), 7.21 (d, J ) 2.6 Hz, 1H), 6.82 (d, J ) 8.9 Hz, 2H), 5.55 (d,
J ) 3.8 Hz, 1H), 3.93 (t, J ) 6.2 Hz, 2H), 3.88 (s, 3H), 3.86 (s,
3H), 3.49 (m, 2H), 3.14 (m, 2H), 2.66 (m, 1H), 2.49 (m, 1H),
1.68-1.80 (m, 5H), 1.47-1.56 (m, 2H); 13C NMR (100 MHz,
CDCl3) δ ) 166.8 (s), 162.6 (s), 157.8 (s), 147.6 (d), 147.4 (s),
144.3 (s), 131.7 (d), 131.6 (d), 126.6 (s), 122.5 (s), 121.5 (d),
118.4 (d), 113.9 (d), 101.2 (d), 72.1(d), 66.4 (t), 59.8 (d), 58.4
(t), 55.7 (q), 51.9 (q), 43.2 (t), 34.2 (t), 32.5 (d), 28.1 (t), 26.0
(d), 21.5 (t). MS (FAB) 477.2369 (C28H33O5N2 requires 477.2390).
Th er m od yn a m ic Cycliza tion of Qu in in e Exten d ed
Mon om er (2). Sa m p le p r ep a r a tion of KOMe ca ta lyst.
KOMe in methanol (0.513 mL, 0.78 M, 0.40 mmol), freshly
prepared from potassium metal and methanol, was added to
18-crown-6 (106 mg, 0.40 mmol). Dried toluene (1 mL) was
then added and the mixture condensed under reduced pressure
to ca. 0.5 mL. More toluene (1 mL) was added, and the mixture
was again condensed under reduced pressure to 0.5 mL. This
was repeated once more to make sure all of the methanol had
been removed azeotropically. The catalyst mixture was then
diluted with toluene (ca.1.5 mL) and the solution filtered under
an inert atmosphere to give a KOMe‚18-C-6 toluene solution
of ca. 0.015-0.03 M as determined by titration.

Thermodynamic Cyclizations of Cinchona Alkaloids
J . Org. Chem., Vol. 64, No. 16, 1999 5813
9-O-ter t-Bu tyld im eth ylsilylqu in id in e (12). To a solution
of quinidine 11 (10 g, 30.8 mmol) in DMF (50 mL) was added
Et3N (8.6 mL, 61.6 mmol), DMAP (0.75 g, 6.2 mmol), and
TBDMSCl (6.97 g, 46.2 mmol). The solution was allowed to
stir overnight and worked up by adding toluene and washing
with water (×3). The toluene was removed under vacuum and
the remaining oil purified by flash column chromatography,
ethyl acetate/methanol (9:1) to yield the product (13.15 g, 97%).
TLC ethyl acetate/methanol (8:2) Rf ) 0.43. 1H NMR (400
MHz, CDCl3) δ ) 8.73 (d, J ) 4.5 Hz, 1H), 7.99 (d, J ) 9.2 Hz,
1H), 7.54 (d, J ) 4.5 Hz, 1H), 7.36 (dd, J ) 2.6, 9.2 Hz, 1H),
7.15 (brs, 1H), 5.9-6.1 (m, 1H), 5.59 (brs, 1H), 5.08 (d, J )
9.0 Hz,1H), 5.03 (d, J ) 15 Hz, 1H) 3.92 (s, 3H), 3.26 (m, 1H),
2.5-3.0 (m, 5H), 2.21 (m, 1H), 2.05 (m, 1H), 1.75 (m, 1H), 1.46
(m, 1H), 1.11 (m, 1H), 0.92 (s, 9H), 0.10 (s, 3H), -0.33 (s, 3H);
13C NMR (100 MHz, CDCl3) δ ) 157.9 (s), 148.2 (s), 147.5 (d),
144.4 (s), 140.6 (d), 131.9 (d), 126.4 (s), 121.5 (d), 118.9 (d),
114.6 (t), 100.5 (d), 73.0 (d), 61.1 (d), 55.7 (q), 50.4 (t), 49.5 (t),
40.4 (d), 28.1 (d), 26.7 (q), 22.4 (t), 21.0 (t), 18.1 (s), -4.6 (q),
-5.2 (q). MS (FAB) 439.27880 (C26H39O2N2Si requires
439.27806).
9-O-ter t-Bu tyld im eth ylsilyl-10-11-d ih yd r o-11-h yd r oxy-
qu in id in e (13). 12 (12.15 g, 27.7 mmol) was dissolved in
diglyme (100 mL) in a flask equipped with a condenser under
an inert atmosphere. The solution was cooled to 0 °C, and BH3‚
THF (1 M in THF, 139 mL, 139 mmol) was added via syringe
and left stirring for 30 min. The mixture was allowed to warm
to room temperature and the THF removed under vacuum.
Triethylamine N-oxide dihydrate (46.17 g, 415 mmol) was then
added, and the mixture was gently heated at 100 °C for 1 h.
Ethyl acetate was then added, to the mixture, and the organic
layer was washed with water (×3) and dried over magnesium
sulfate. The ethyl acetate was then removed to yield an oil
which was purified by recrystallization from DCM/ethyl
acetate to yield white crystals (9.22 g, 81%). TLC ethyl acetate/
methanol (7:3) Rf ) 0.18. 1H NMR (400 MHz, CDCl3) δ ) 8.73
(d, J ) 4.5 Hz, 1H), 8.03 (d, J ) 9.2 Hz, 1H), 7.55 (d, J ) 4.5
Hz, 1H), 7.38 (dd, J ) 2.5, 9.2 Hz, 1H), 7.16 (d, J ) 2.5 Hz,
1H), 5.63 (d, J ) 2.4 Hz, 1H), 3.93 (s, 3H), 3.67 (m, 2H), 2.57-
2.95 (m, 4H), 2.04 (brs, 1H), 2.02 (m, 1H), 1.64-1.78 (m, 5H),
1.44 (m, 2H), 0.94 (s, 9H), 0.12 (s, 3H), -0.32 (s, 3H); 13C NMR
(100 MHz, CDCl3) δ ) 157.9 (s), 148.1 (s), 147.4 (d), 144.3 (s),
131.9 (d), 126.2 (s), 121.6 (d), 118.8 (d), 100.5 (d), 73.3 (d), 61.0
(t), 60.8 (d), 55.8 (q), 51.2 (t) 50.5 (t), 36.2 (t), 32.2 (d), 27.4 (d),
27.1 (t), 26.3 (q), 20.2 (t), 18.1 (s), -4.6 (q), -4.8 (q). MS (FAB)
457.28690 (C26H41O3N2Si requires 457.28863).
7.53 (d, J ) 4.5 Hz, 1H), 7.38 (dd, J ) 2.5, 9.2 Hz, 1H), 7.20
(m, 1H), 5.66 (brs, 1H), 4.03 (s, 3H), 3.69 (s, 3H), 2.75-3.50
(brm, 4H), 2.52 (m, 2H), 2.14-2.40 (m, 2H), 2.01 (m, 1H), 1.52-
2.00 (m, 4H), 0.94 (s, 9H), 0.23 (s, 3H), -0.29 (s, 3H); 13C NMR
(100 MHz, CDCl3) δ ) 173.3 (s), 157.8 (s), 147.4 (s), 147.0 (d),
144.5 (s), 131.9 (d), 125.9 (s), 121.3 (d), 118.8 (d), 100.4 (d),
73.1 (d), 60.4 (t), 60.2 (d), 55.8 (q), 55.5 (q) 51.8 (t), 49.2 (t),
36.7 (t), 26.9 (q), 26.5 (d), 25.8 (d), 18.0 (s), 14.2 (d), -4.6 (q),
-4.7 (q). MS (FAB) 485.2842 (C27 H41O4N2Si requires 485.2835).
Meth yl 10-11-Dih yd r oqu in id in e-11-ca r boxyla te (10).
To a solution of 15 (0.80 g, 1.65 mmol) in THF (15 mL) was
added TBAF (1 M in THF, 3.3 mL, 3.3 mmol). Once this had
stirred overnight, ethyl acetate was added to the solution
which was then washed with water (×3) and dried over
magnesium sulfate. The ethyl acetate was then removed under
vacuum and the resulting oil purified by flash column chro-
matography, ethyl acetate/methanol (10:0, 9:1, ..., 6:4) to yield
a white foam (0.41 g, 67%). TLC ethyl acetate/methanol (6:4)
Rf ) 0.16. 1H NMR (400 MHz, CDCl3) δ ) 8.60 (d, J ) 4.5 Hz,
1H), 7.90 (d, J ) 9.2 Hz, 1H), 7.55 (d, J ) 4.5 Hz, 1H) 7.23
(dd, J ) 2.5, 9.2 Hz, 1H), 7.04 (d, J ) 2.5 Hz, 1H), 5.78 (brs,
1H), 4.71 (brs, 1H), 3.76 (s, 3H), 3.66 (s, 3H), 3.42 (m, 1H),
3.10 (m, 2H), 2.96 (m, 1H), 2.86 (m, 1H), 2.60 (dd, J ) 8.2,
15.8 Hz, 1H), 2.50 (dd, J ) 6.8, 15.8 Hz, 1H), 2.00-2.13 (m,
2H), 1.70 (m, 1H), 1.57 (m, 1H), 1.50 (m, 1H), 1.01 (m, 1H);
13C NMR (100 MHz, CDCl3) δ ) 173.3 (s), 157.8 (s), 147.5 (d),
147.0 (s), 144.0 (s), 131.5 (d), 126.1 (s), 121.7 (d), 118.3 (d),
100.8 (d), 70.9 (d), 59.5 (d), 55.8 (q), 51.7 (q), 50.3 (t), 49.9 (t),
37.0 (t), 31.7 (d), 27.5 (t), 26.2 (d), 19.7 (t). MS (FAB) 371.19350
(C21H27O4N2 requires 371.19707).
Th er m od yn a m ic Cycliza tion of (10), Cyclic Dim er Ca2.
10 (10.3 mg, 2.78 × 10-5 moles) was added to a round-bottomed
flask attached to a Soxhlet extractor which contained molec-
ular sieves (4 Å) and then dissolved in toluene (5.6 mL). The
reaction was heated at reflux for 60 min to remove all of the
water from the system, and then the KOMe‚18-C-6 catalyst
solution (23 µL, 0.06 M, 1.39 × 10-6 moles) was added. To work
up the reaction, the mixture was added to aqueous pH 7 buffer
and extracted with ethyl acetate, and the solvent was removed
to give the product which was immediately submitted for
NMR. TLC ethyl acetate/methanol (6:4) Rf ) 0.22. 1H NMR
(400 MHz, CDCl3): δ ) 8.74 (d, J ) 4.5 Hz, 2H), 8.02 (d, J )
9.2 Hz, 2H), 7.41 (dd, J ) 2.65, 9.2 Hz, 2H), 7.31 (d, J ) 4.5
Hz, 2H), 7.16 (d, J ) 2.63 Hz, 2H), 6.82 (s, 2H), 3.96 (s, 6H),
3.30 (d, J ) 12.7 Hz, 2H), 3.25 (d, J ) 12.7 Hz, 2H), 3.14 (m,
4H), 2.90 (m, 4H), 2.85 (m, 4H), 2.55 (dd, J ) 3.1, 15.2 Hz,
2H), 2.22 (m, 4H), 1.72 (m, 2H), 1.57 (m, 2H), 1.46 (m, 2H),
1.14-1.23 (m, 4H). ES-MS 677 (Cd 2H+).
9-O-ter t-Bu t yld im et h ylsilyl-10-11-d ih yd r oq u in id in e-
11-ca r boxylic a cid (14). 13 (1.0 g, 2.19 mmol) was dissolved
in acetone (100 mL). J ones reagent was then added drop by
drop until the dark brown color persisted (ca.10 mL). The
mixture was then neutralized with saturated NaHCO3, ex-
tracted with ethyl acetate, and dried over magnesium sulfate.
It was then flash-columned ethyl acetate/methanol (8;2, 7:3,
6:4, ..., 0.0:10) to yield a white foam (0.64 g, 62%). TLC ethyl
Lin ea r Qu in id in e Dim er HO-Cd 2-OMe. The linear
quinidine dimer was prepared in the same manner as that
for the previously published linear quinine dimer18 with the
following alterations: quinidine acid 14 (103 mg, 0.22 mmol)
and the quinidine monomer 10:HO-Cd -OMe (81 mg, 0.22
mmol) were reacted with 2,6-dichlorobenzoyl chloride (44 µL,
0.33 mmol), triethylamine (61 µL, 0.44 mmol), and DMAP (6
mg, 0.44 mmol) in DCM (4.6 mL). After stirring overnight, the
reaction was worked up and purified as before, to yield a white
foam (132 mg, 73%). 1H NMR (400 MHz, CDCl3) δ ) 8.70-
8.74 (m, 2H), 8.01 (d, J ) 9.0 Hz, 1H), 8.00 (d, J ) 9.0 Hz,
1H), 7.53 (d, J ) 4.5 Hz, 1H), 7.05-7.40 (m, 5H), 6.48 (d, J )
7.5 Hz, 1H), 5.64 (brs, 1H), 3.94 (s, 3H), 3.89 (s, 3H), 3.62 (s,
3H), 2.42-3.40 (m, 12H), 1.39-2.19 (m, 16H), 0.94 (s, 9H), 0.09
(s, 3H), -0.3 (s, 3H); 13C NMR (100 MHz, CDCl3) δ ) 173.0
(s), 171.9 (s), 157.9 (s), 147.4 (d), 144.8 (s), 144.4 (s), 131.9 (d),
127.0 (s), 126.0 (s), 121.8 (d), 118.7 (d), 101.5 (d), 100.4 (d),
73.8 (d), 58.9 (d), 55.7(q), 55.6 (q), 51.7 (q), 50.0 (t), 49.6 (t),
37.3 (t), 32.5 (d), 32.3 (d), 26.8 (t), 26.6 (d), 26.1 (q), 25.8 (d),
23.6 (t), 19.5 (t), 18.1 (s), -4.7 (q), -5.1 (q). MS (FAB) 823.4171
(C47H62O7N2Si requires 823.41735). The TBDMS group was
removed in a similar way to that used before with the following
amounts of material. The linear TBDMS-protected quinidine
dimer (100 mg, 0.12 mmol) was reacted with TBAF (0.243 mL,
1 M in THF, 0.24 mmol) in THF (1.5 mL). After stirring
overnight, the reaction was worked up and purified as before,
1
acetate/methanol (1:1) Rf ) 0.12. H NMR (400 MHz, CDCl3)
δ ) 8.71 (brs, 1H), 8.03 (d, J ) 9.2 Hz, 1H), 7.49 (brs, 1H)
7.37 (d, J ) 9.2 Hz, 1H), 7.10 (brs, 1H), 5.72 (brs, 1H), 3.93 (s,
3H), 1.24-3.39 (brm, 13H), 0.97 (s, 9H), 0.08 (s, 3H), -0.32
(s, 3H); 13C NMR (100 MHz, CDCl3) δ ) 171.0 (s), 158.2 (s),
148.0 (s), 147.4 (d), 144.4 (s), 132.0 (d), 126.1 (s), 121.5 (d),
118.7 (d), 100.7 (d), 73.0 (d), 61.0 (d), 60.4 (t), 55.9 (q), 55.5
(d), 32.9 (t), 26.7 (q), 26.1 (d), 25.7 (d), 18.0 (s), -4.6 (q), -4.8
(q). MS (FAB) 471.26400 (C26H39O4N2Si requires 471.26789).
Meth yl 9-O-ter t-Bu tyld im eth ylsilyl-10-11-d ih yd r oqu i-
n id in e-11-ca r boxyla te (15). To a solution of 14 (0.84 g, 1.79
mmol) in methanol (40 mL) was added a few drops of
concentrated HCl and left to stir overnight. The solution was
then neutralized with sodium bicarbonate (saturated) and the
methanol removed under vacuum. The resulting oil was
dissolved in DCM, washed with water (×3), and dried over
magnesium sulfate. Once the DCM had been removed, the
compound was purified by flash column chromatography ethyl
acetate/methanol (9:1) to yield a clear oil (0.80 g, 92%). TLC
ethyl acetate/methanol (9:1). Rf ) 0.50. 1H NMR (400 MHz,
CDCl3) δ ) 8.73 (d, J ) 4.5 Hz, 1H), 8.01 (d, J ) 9.2 Hz, 1H),
1
to yield HO-Cd 2-OMe as a clear oil (73 mg, 81%). H NMR

5814 J . Org. Chem., Vol. 64, No. 16, 1999
Rowan et al.
(400 MHz, CDCl3) δ ) 8.61-8.65 (m, 2H), 7.96 (d, J ) 9.0 Hz,
1H), 7.95 (d, J ) 9.0 Hz, 1H), 7.53 (d, J ) 4.5 Hz, 1H), 7.13-
7.32 (m, 5H), 6.48 (d, J ) 6.4 Hz, 1H), 5.59 (s, 1H), 4.48 (brs,
1H), 3.97 (s, 3H), 3.91 (s, 3H), 3.69 (s, 3H), 3.22 (m, 1H), 2.64-
3.00 (m, 9H), 2.42-2.56 (m, 2H), 2.00-2.21 (m, 4H), 1.35-
1.76 (m, 8H), 1.30-1.45 (m, 2H); 13C NMR (100 MHz, CDCl3)
δ ) 173.3 (s), 172.2 (s), 158.0 (s), 157.8 (s), 147.9 (s), 147.6 (d),
147.3 (d), 144.6 (s), 144.2 (s), 143.9 (s), 131.7 (d), 131.6 (d),
126.9 (s), 126.4 (s), 121.8 (d), 121.5 (d), 118.3 (d), 118.1 (d),
101.5 (d), 101.2 (d), 73.7 (d), 72.3 (d), 59.5 (d), 58.9 (d), 55.6
(q), 51.8 (q), 50.6 (t), 50.1 (t), 50.0 (t), 49.7 (t), 37.7 (t), 37.5 (t),
32.0 (d), 32.2 (d), 26.9 (t), 26.8 (t, d), 26.7 (d), 22.9 (t), 20.0 (t).
MS (FAB) 709.3636 (C41H49O7N4 requires 709.3601).
Rf ) 0.48. 1H NMR (250 MHz, CDCl3) δ ) 8.74 (d, J ) 7.45
Hz, 1H), 8.06 (d, J ) 9.2 Hz, 1H), 7.99 (d, J ) 7.22 Hz, 2H),
7.54 (d, J ) 4.5 Hz, 1H), 7.36 (dd, J ) 2.5, 9.2 Hz, 1H), 7.18
(brs, 1H), 6.88 (d, J ) 7.1 Hz, 2H), 5.64 (brs, 1H), 4.02 (m,
2H), 3.95 (s, 3H), 3.87 (s, 3H), 2.54-3.30 (brm, 3H), 1.94 (m,
4H), 1.78 (m, 2H), 1.52 (m, 2H), 1.02 (m, 1H), 0.94 (s, 9H),
0.16 (s, 3H), -0.31 (s, 3H); 13C NMR (63 MHz, CDCl3) δ )
166.8 (s), 162.7 (s), 158.1 (s), 156.6 (s), 147.3 (d), 144.4 (s), 131.9
(d), 131.6 (d), 126.1 (s), 122.6 (s), 121.3 (d), 118.8 (d), 114.0
(d), 100.5 (d), 79.7 (d), 66.5 (t), 60.9 (d), 60.7 (d), 55.4 (q), 51.8
(q), 50.3 (t), 49.7 (t), 49.4 (t), 32.5 (d), 31.9 (t), 26.9 (d), 26.0
(q), 21.0 (d), 19.9 (t), 18.0 (s), -4.7 (q), -5.2 (q). MS (FAB)
591.32550 (C34H47O5N2Si requires 591.3254).
9-O-ter t-Bu tyldim eth ylsilyl-10-11-d ih ydr o-11-tosyl-qu i-
n id in e (17). To a solution of 13 (1.5 g, 3.42 mmol) in DCM
(10 mL) was added tosyl chloride (1.28 g, 6.72 mmol) and Et3N
(0.915 mL, 6.58 mmol). The solution was allowed to stir
overnight and worked up by adding ethyl acetate, washing
with water (×3), and drying over magnesium sulfate. The ethyl
acetate was removed under vacuum and the remaining oil
purified by flash column chromatography ethyl acetate/
methanol (10:0, 9:1, ..., 6:4) to yield the product (1.62 g, 77%).
TLC ethyl acetate/methanol (7:3) Rf ) 0.54. 1H NMR (400
MHz, CDCl3) δ ) 8.73 (d, J ) 4.5 Hz, 1H), 8.01 (d, J ) 9.2 Hz,
1H), 7.82 (d, J ) 8.3 Hz, 2H), 7.53 (d, J ) 4.5 Hz, 1H), 7.36 (d,
J ) 8.2 Hz, 3H), 7.13 (s, 1H), 5.61 (brs, 1H), 4.04 (m, 2H),
3.93 (s, 3H), 2.61-3.04 (brm, 5H), 2.44 (s, 3H), 1.97 (m, 1H),
1.79 (m, 2H), 1.61 (m, 2H), 1.23 (m, 2H), 0.94 (s, 9H), 0.084 (s,
3H), -0.32 (s, 3H); 13C NMR (100 MHz, CDCl3) δ ) 157.9 (s),
147.3 (d), 144.9 (s), 144.4 (s), 133.0 (s), 131.9 (d), 131.5 (d),
129.9 (d), 127.9 (d), 126.0 (s), 121.8 (d), 118.7 (d), 100.3 (d),
79.6 (d), 68.9 (t), 60.8 (q), 55.4 (d), 50.2 (t), 32.1 (d), 32.0 (d),
31.4 (q), 26.4 (d), 25.9 (q), 25.8 (d), 21.7 (d), 18.0 (s), -4.7 (q),
-5.2 (q). MS (FAB) 611.29160 (C33H47O5N2SiS requires
611.2975).
Met h yl 10-11-Dih yd r oq u in id in e-11-(p h en oxy-p -ca r -
boxyla te) (16). Deprotection of the C(9)-hydroxyl was carried
out utilizing a modified procedure for the preparation of 6. 19
(1.17 g, 1.98 mmol), TBAF (5.94 mL, 1 M, 5.94 mmol), and
THF (30 mL) was stirred overnight. Work up was accom-
plished as before with the remaining oil purified by flash
column chromatography ethyl acetate/methanol (9:1, 8:2, 7:3)
to yield the product (0.422 g, 45%). TLC ethyl acetate/methanol
1
(6:4) Rf ) 0.20. H NMR (400 MHz, CDCl3 + drop CD3OD) δ
) 8.50 (d, J ) 4.6 Hz, 1H), 7.83 (d, J ) 8.5 Hz, 2H), 7.79 (d,
J ) 9.5 Hz, 1H), 7.51 (d, J ) 4.6 Hz, 1H), 7.20 (dd, J ) 2.5,
9.5 Hz, 1H), 7.08 (d, J ) 2.5 Hz, 1H), 6.77 (d, J ) 8.5 Hz, 2H),
5.46 (d, J ) 1 Hz, 1H), 3.91 (t, J ) 6.4 Hz, 2H), 3.78 (s, 3H),
3.72 (s, 3H), 3.33 (m, 1H), 2.70-2.84 (m, 3H), 2.66 (m, 1H),
2.02 (m, 1H), 1.84-2.00 (m, 2H), 1.68 (m, 1H), 1.60 (brs, 1H),
1.32-1.43 (m, 2H), 0.80 (m, 1H); 13C NMR (100 MHz, CDCl3)
δ ) 166.8 (s), 162.4 (s), 158.0 (s), 147.1 (d), 144.4 (s), 143.8 (s),
131.7 (d), 131.4 (d), 125.3 (s), 122.9 (s), 122.1 (d), 118.4 (d),
114.1 (d), 99.6 (d), 65.6 (t), 60.0 (d), 56.5 (q), 51.9 (q), 50.0 (t),
49.6 (t), 31.4 (t), 30.8 (d), 25.9 (d), 24.4 (t), 18 0.2 (t). MS (FAB)
477.24260 (C28H32O5N2 requires 477.2390).
Th er m od yn a m ic Cycliza tion of (16), Cyclic Dim er Ca2.
KOMe catalyst was prepared as for the thermodynamic
cyclization of 2. Thermodynamic cyclization of monomer 16
was carried out as for monomer 2 with the following quantities
of reagents: 16 (10.6 mg, 2.22 × 10-5 moles), toluene (4.45
mL), and the KOMe‚18-crown-6 catalyst solution (19 µL, 0.06
M, 1.11 × 10-6 moles). The reaction was worked up as before
and the solvent removed to give the product which was
immediately submitted for NMR. TLC ethyl acetate/methanol
9-O-ter t-Bu tyld im eth ylsilyl-10-11-d ih yd r o-11-iod o-qu i-
n id in e (18). (17) (2.0 g, 3.28 mmol) was dissolved in acetone
(20 mL) in a flask equipped with a condenser under an inert
atmosphere and NaI (2.0 g, 13 mmol) added. The solution was
heated at reflux at 60 °C overnight and then left to cool. The
mixture was worked up by adding ethyl acetate, washing with
water (×3), and drying over magnesium sulfate. The ethyl
acetate was removed under vacuum and the remaining oil
purified by flash column chromatography ethyl acetate/
methanol (9:1) to yield the product (1.47 g, 79%). TLC ethyl
1
(6:4) Rf ) 0.24. H NMR (400 MHz, CDCl3) δ ) 8.73 (d, J )
4.5 Hz, 2H), 8.14 (d, J ) 8.8 Hz, 4H), 8.03 (d, J ) 9.2 Hz, 2H),
7.48 (d, J ) 4.5 Hz, 2H), 7.42 (d, J ) 2.5, 9.2 Hz, 2H), 7.30 (d,
J ) 2.5 Hz, 2H), 7.07 (s, 2H), 6.86 (d, J ) 8.8 Hz, 4H), 3.96 (s,
6H), 3.85 (m, 2H), 3.64 (m, 2H), 3.47 (m, 2H), 3.23 (t, J ) 9.2
Hz, 2H), 3.05 (m, 2H), 2.87 (m, 4H), 2.39 (m, 4H), 2.04 (m,
4H), 1.76 (m, 2H), 1.65 (m, 2H), 1.52 (m, 2H), 1.09 (m, 1H).
ES-MS 889 (Ca 2H+).
1
acetate/methanol (8:2) Rf ) 0.48. H NMR (400 MHz, CDCl3)
δ ) 8.74 (d, J ) 4.5 Hz, 1H), 8.03 (d, J ) 9.2 Hz, 1H), 7.54 (d,
J ) 4.5 Hz, 1H), 7.37 (dd, J ) 2.5, 9.2 Hz, 1H), 7.15 (brs, 1H),
5.64 (brs, 1H), 3.95 (s, 3H), 3.14 (m, 3H), 2.48-2.81 (m, 5H),
2.01 (m, 4H), 1.72 (m, 1H), 1.46 (m, 1H), 0.96 (s, 9H), 0.15 (s,
3H), -0.27 (s, 3H); 13C NMR (100 MHz, CDCl3) δ ) 157.4 (s),
147.4 (d), 147.3 (s), 144.4 (s), 131.5 (d), 126.0 (s), 121.3 (d),
118.8 (d), 100.4 (d), 79.6 (d), 60.4 (d), 55.4 (q), 50.5 (t), 49.4 (t),
36.4 (t), 26.9 (d), 25.9 (q), 25.8 (d), 21.1 (d), 18.1 (s), -4.6 (q),
-4.8 (q). MS (FAB) 567.18580 (C26H40O2N2SiI requires
567.1906).
Ack n ow led gm en t. We thank the EPSRC for gener-
ous financial support and Dr. Darren Hamilton for the
xanthene sample.
Meth yl 9-O-ter t-Bu tyld im eth ylsilyl-10-11-d ih yd r oqu i-
n id in e-11-(p h en oxy-p-ca r boxyla te) (19). To a solution of
18 (0.4 g, 0.70 mmol) in acetone (4 mL) was added methyl
4-hydroxybenzoate (0.21 g, 1.4 mmol), 18-crown-6 (0.37 g, 1.4
mmol), and K2CO3 (0.19 g, 1.4 mmol), and the mixture was
left to stir overnight. To work up the reaction, ethyl acetate
was added, and the solution was washed with water (×3) and
then dried over magnesium sulfate. The ethyl acetate was
removed under vacuum and the remaining oil purified by flash
column chromatography ethyl acetate/methanol (9:1) to yield
the product (0.41 g, 98%). TLC ethyl acetate/methanol (8:2)
Su p p or tin g In for m a tion Ava ila ble: 1H NMR, HPLC,
and UV of 2; 1H NMR 10, 16, Ce2, Ce2 + Ce3, HO-Ca 2-OMe;
1H NMR, ES-MS, HPLC, and UV of the thermodynamic
cyclization of 2 and 10; ES-MS, HPLC, and UV of the
thermodynamic cyclization of 16; 1H NMR, HPLC, and UV of
the thermodynamic cyclization of 10 with 16; ES-MS of the
thermodynamic cyclization of 1b with 10; 1a with 16; 1b with
2 and 10 with 2 (PDF). This material is available free of charge
via the Internet at http://pubs.acs.org.
J O982496X