The cis-2-alkyl-3-oxy-tetrahydropyran unit as a building block for new ionophores with C2-symmetry

Article abstract of DOI:10.1016/j.tetlet.2004.05.036

The use of cis-2-alkyl-3-oxy-tetrahydropyran unit is reported as a novel structure for the design and synthesis of a new type of ionophore with C ₂-symmetry. The synthesis of five macrolides and their complexation properties were investigated.

Full text of DOI:10.1016/j.tetlet.2004.05.036

Tetrahedron
Letters
Tetrahedron Letters 45 (2004) 5215–5219
The cis-2-alkyl-3-oxy-tetrahydropyran unit as a building block
for new ionophores with C2-symmetryq
*
ꢀ
Romen Carrillo, Victor S. Martın and Tomas Martın
ꢀ
ꢀ
Instituto Universitario de Bio-Orgꢀanica ‘‘Antonio Gonzꢀalez’’, Universidad de La Laguna,
C/Astrofꢀısico Francisco Sꢀanchez, 2, 38206 La Laguna, Tenerife, Spain
Received 14 April 2004; revised 6 May2004; accepted 10 May2004
Dedicated to Professor Julius Rebek, Jr. on the occasion of his 60th birthday
Abstract—The use of cis-2-alkyl-3-oxy-tetrahydropyran unit is reported as a novel structure for the design and synthesis of a new
type of ionophore with C2-symmetry. The synthesis of five macrolides and their complexation properties were investigated. The
template effect as an effective tool to improve the selectivityand yield in the macrolactonization process is described.
Ó 2004 Elsevier Ltd. All rights reserved.
In living systems, molecular recognition is at the origin of
highlyselective chemical reactions and transport phe-
nomena. The study of synthetic model systems contrib-
utes to the understanding of these processes and, at the
same time, offers new perspectives for controlling speci-
ficityand reactivityin chemistry. Over the last 30 years, a
great number of functional groups have been incorpo-
rated into the design of new cation receptors. Tetra-
hydropyrans are widely found in known natural
ionophores1–3 and have been frequentlyused as elements
in cation recognition4;5 since theyconfer more rigidityand
new properties to the systems. Thus, mention can be made
of the molecules designed and synthesized by Okada and
co-workers6 and Burke and co-workers,7 which are
chorand-type cyclooligolides that incorporate tetra-
hydropyran units in their structures. On the other hand,
Still and co-workers designed and synthesized a podand-
type system of linked tetrahydropyran rings whose con-
formation is controlled bythe presence of methyl groups
at C-3.8 However, all these receptors use the tetra-
hydropyran rings linked through positions C-2 and C-6.
attention to the search for simple models of fused tetra-
hydropyrans that allows us to establish a relationship
between structure–activityand binding capacityof
polyether toxins. Thus, we focused our attention on the
2-alkyl-3-oxy-tetrahydropyran unit as a masterpiece for
the design of new ionophores. This structural unit is
widelydisplayed in polyether toxins of marine origin,
and is generallyin trans disposition (Fig. 1).
This configuration has the oxygens directed towards the
opposed faces, generating a practicallyflat system, 10 that
maynot be well adapted for the cation recognition.
Alternatively, the cis configuration can preferablybe
found in two conformations, one with the 3-oxygroup
Considering the well-established relationship between
ladder toxins and metallic cations,9 we have directed our
Keywords: Ionophores; Lariat ethers; Template effect.
q Supplementarydata associated with this article can be found, in the
online version, at doi:10.1016/j.tetlet.2004.05.036
* Corresponding author. Tel.: +34-922-318580; fax: +34-922-318571;
e-mail: tmartin@ull.es
Figure 1.
0040-4039/$ - see front matter Ó 2004 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2004.05.036

5216
R. Carrillo et al. / Tetrahedron Letters 45 (2004) 5215–5219
axial (A), such that both oxygens go to the same face,
and the other one with this group equatorial (B), forcing
both oxygens to be located on opposite faces, the first of
these being the one that maygenerate the suitable cur-
vature, so that both oxygens can participate in the rec-
ognition.
On the other hand, it is well known that the best and
easiest wayto maintain the chiralityof a unit is by
transforming this into a molecule with C2-symmetry.
How can molecules that display C2-symmetry be gen-
erated from this unit of recognition? To accomplish this,
we linked the oxygroup of a unit with the alkyl group of
the other, using identical spacers (Fig. 2a). Thus, we
designed the tricyclic diolide 1, structurallycharacter-
ized bythe presence of two cis-tetrahydropyran units,
linked by two 2-oxy-acetyl groups as spacers, forming a
14-membered macrodiolide (Fig. 2).
Figure 3. Retrosynthetic analysis for macrodiolides 1 and 2.
sioned as being synthesized from two suitable hydroxy-
acids 4 or 6 (R1 ¼ R2 ¼ H), byan intermolecular esteri-
fication followed bymacrolactonization. Tetrahdyro-
pyrans 4 or 6 could be obtained from the commercially
available tri-O-acetyl-D-galactal and tri-O-acetyl-D-
glucal, respectively.
The direct dimerization approach of
(R1 ¼ R2 ¼ H), however, is entropicallyunfavourable,
4
or
6
Using the same design, and in order to check the
importance of the configuration of the 2-alkyl-3-oxy-
tetrahydropyran moieties, we decided to synthesize
compound 2, which displays the trans fusion. In prin-
ciple this structure can be considered to be a formal
ether lariat, where the central macroring is a formal 14-
crown-4 and the tetrahydropyran rings are the sidearms
that confer greater conformational rigidityand whose
oxygens can participate in the complexation. From
modelling,11 it can be predicted that the cis fusion could
create a hydrophilic concave face where the oxygen
atoms could act as ligands wrapping the cation guest
and a hydrophobic convex face where the carbon skel-
eton is on the outer surface (Fig. 2b). However, the trans
fusion shows a practicallyflat structure (Fig. 2c).
Additionally, we synthesized the central 14-membered
macrodiolide 3 to verifythat metal complexation does
not take place in the macrocyclic plane.
affording the lactonization product, even using 2-chloro-
1,3-dimethylimidazolinium chloride in presence of
DMAP or NaH or KH in CH2Cl2.12 To overcome this
difficulty, a convergent strategy was applied: iterative
esterification couplings of two suitablyfunctionalized
tetrahydropyran units 4 or 6 (R1 or R2 ¼ H), followed
by macrolactonization of the resulting hydroxy-acid.
Thus, to carryout the synthesis of compound 1 (Scheme
1), the diol 513 was protected, as the bis-silyl ethers,
followed byselective deprotection of the silyl ether of
the primaryalcohol to afford the cis-alcohol 8, which
was alkylated with sodium iodoacetate to provide the
carboxylic acid 9. Esterification of 9 with benzyl alcohol
and further deprotection of the tert-butyl-dimethylsilyl
(TBS) ether afforded the hydroxy ester 10. Coupling of
acid 9 with hydroxy ester 10 gave pseudodimer 11.
Retrosynthetic analyses for macrodiolides 1 and 2 are
shown in Figure 3. Macrodiolides 1 and 2 were envi-
Figure 2. (a) General design of molecules with C2-symmetry. (b)
Molecule 1 in the suitable conformation such that cation recognition
maytake place (U-shaped). (c) Low-energyconformation of molecule
2.
Scheme 1.

R. Carrillo et al. / Tetrahedron Letters 45 (2004) 5215–5219
5217
Cleavage of both alcohol and acid protecting groups
afforded the hydroxy acid 12, which was transformed
into the thioester 13. Unfortunately, 13 under Coreyand
Nicolaou macrolactonization conditions,14 afforded
exclusivelythe macrotetraolide 14.
cis, perhaps as a result of folding of compound 15 over
ion Kþ.
In addition, we observed that the conformation of the
tetrahydropyrans of compound 1 is the same before and
after the complexation. In the free receptor, the NOE
effects (CDCl3) indicate that a large number of mole-
cules have the tetrahydropyran rings in the A confor-
mation (Fig. 1), that implies a partiallypreorganized
system. Thus, when one deals with potassium thiocya-
nate and ultrasound, to form the complex 1 Æ Kþ
(CDCl3),21 the NOE effects continue as before, although
changes take place in the chemical shift of the signals in
the proton (Fig. 4) and carbon NMR spectra. This,
along with the value of the association constant, allows
us to conclude that almost all the molecules adopt the
U-shaped conformation in the complex (Fig. 2b) and
that the change in the chemical shift must be mainlydue
to the change in conformation of the flexible part of the
molecule, that is to say, the 14-membered macrodiolide.
However, when compound 12 was treated under Yama-
guchi and co-workers conditions,15 the macrodiolide 1
was obtained in 20% yield along with 14 in 40% yield.16
The template effect improved selectivityand yield in the
macrolactonization (vide infra). Using the diol 617 and
applying the same sequence of reactions described
above, we obtained the receptor 2 and the macrotetra-
olide 15 (trans isomer of 14), using Yamaguchi condi-
tions in the cyclization step, with 55% and 35% yields,
respectively. The same strategy was applied to obtain
compound 3, using as starting material 3-{[tert-
butyl(dimethyl)silyl]oxy}-1-propanol. The structures
of the macrolides were unambiguouslydetermined
byNMR spectroscopy( 1H and 13C) and FAB mass
spectrometry.18
Finally, in view of the low yields obtained in the macro-
lactonization process, and considering the association
constants observed for 1 and 14, we decided to explore
the possible template effect that could be exerted by
metallic cations in the cyclization step. In our first at-
tempt we used the standard Yamaguchi conditions,
method A (Table 2). The addition of an alkali metal
cation did not improve yield and selectivity, thus the
solvent was changed to a more polar one, dichloro-
methane (method B). In this case the use of K2CO3
The association constants (Ka) of the macrodiolides 1, 2
and 3, and the macrotetraolides 14 and 15 in CHCl3
saturated with water at 23–25 °C were determined by
measurements from the CHCl3 layer using Cram’s pic-
rate extraction method (Table 1).19 The absorption
maxima in CHCl3 (352 < kmax < 359 nm) of the picrate
salts are indicative of 1:1 complexes between metal
picrate and the host macrolides.20 The results of Table 1
show that the onlysubstrate that displasy moderate
association constants is 1, and that in addition it is
þ
slightlyspecific for Na
Kþ/Liþ ¼ 3.5).
and Kþ (Naþ/Liþ ¼ 5.4,
It should be emphasized that the configuration of the
tetrahydropyrans is of extreme importance, since when
this is trans, no extraction was observed. This implies
that the cis-tetrahydropyran oxygen participates in the
recognition process and this along with the null com-
plexation observed bymacrodiolide 3 allows us to
conclude that the complexation in 1 takes place out of
the macrodiolide plane, like a clamp-type rather than a
chorand-type ionophore, despite the fact that 1 is a
macrocycle. This is in agreement with the association
constant values and with the minimized structures
shown in Figure 2b and c. Host 1 wraps the metal cation
and can change the cavitysize according to the metal
cation size, so that it shows comparable binding ability
to all alkali metal cations. Interestingly, in macro-
tetraolides the result is reversed, trans being better than
Figure 4. 1H NMR induced shifts of host 1 bybinding KSCN at 298 K
in CDCl3. (a) Host 1. (b) Complex 1 Æ Kþ.
Table 1. Association constants Ka (MÀ1) of macrolidesa;b
Host
Liþ
Naþ
Kþ
NHþ4
Csþ
1
4.6 Â 103
2.5 Â 104
1.6 Â 104
––
––
2.0 Â 103
2.0 Â 103
2
––
––
––
––
––
––
––
––
––
––
––
––
––
––
––
––
3
14
15
––
2.3 Â 103
a These values are the average of three independent measurements.
b ‘––’ Means that no extraction was observed.

5218
R. Carrillo et al. / Tetrahedron Letters 45 (2004) 5215–5219
Table 2. Template effect
Substrate
Methoda
Solvent (M)
Additive (equiv)
Time
Yieldb (%)
1:14
12
12
13
13
13
13
A
B
C
C
C
C
Toluene (0.001)
CH2Cl2 (0.002)
Toluene (0.001)
CH2Cl2 (0.002)
CH2Cl2 (0.002)
CH2Cl2 (0.002)
––
6 h
22 h
1 d
7 d
9 d
7 d
60
65
50
––
55
95
33:67
87:13
0:100
––
K2CO3 (20)
––
––
Na2CO3 (20)
K2CO3 (20)
65:35
87:13
a Method A: Et3N (2 equiv), 2,4,6-trichlorobenzoyl chloride (1.3 equiv) in THF (0.01 M), then was diluted with toluene (0.001 M) and poured over
DMAP (10 equiv) in boiling toluene. Method B: Et3N (2 equiv), 2,4,6-trichlorobenzoyl chloride (1.3 equiv) in CH2Cl2 (0.01 M), then was diluted
with CH2Cl2 (0.002 M) and K2CO3 (20 equiv) was added and the mixture was warmed up to reflux. Method C: reflux.
b Yields corresponding to compounds 1 þ 14.
enhanced selectivity, but this reaction was difficult to
reproduce and follow byTLC. Then we decided to
explore the macrolactonization using thioester 13
(method C). It is noteworthythat without anycation, no
cyclization product was detected. After several trials, we
References and notes
1. Agtarap, A.; Chamberlin, J. W.; Pinkerton, M.; Steinrauf,
L. K. J. Am. Chem. Soc. 1967, 89, 5737–5739.
2. Berger, J.; Rachlin, A. I.; Scott, W. E.; Sternbach, L. H.;
Goldberg, M. W. J. Am. Chem. Soc. 1951, 89, 5295–5298.
3. (a) Blount, J. F.; Westley, J. W. J. Chem. Soc., Chem.
Commun. 1975, 533; (b) O’Sullivan, A. C.; Struber, F.
Helv. Chim. Acta 1998, 81, 812–827.
4. (a) Shizuma, M.; Adachi, H.; Kawamura, M.; Takai, Y.;
Takeda, T.; Sawada, M. J. Chem. Soc., Perkin Trans. 2,
2001, 592–601; (b) Shizuma, M.; Takai, Y.; Kawamura,
M.; Takeda, T.; Sawada, M. J. Chem. Soc., Perkin Trans.
2, 2001, 1306–1314; (c) Shizuma, M.; Kadoya, Y.; Takai,
Y.; Imamura, H.; Yamada, H.; Takeda, T.; Arakawa, R.;
Takahashi, S.; Sawada, M. J. Org. Chem. 2002, 67, 4795–
4807.
found that better selectivityand yields of compound
1
versus 14 were achieved in boiling dichloromethane with
potassium carbonate, where the Kþ acts as a template,
preorganizing the cyclization precursor for macrolact-
onization, favouring the formation of 1 (95% yield,
1:14 ¼ 6.7) (Table 2). To the best of our knowledge, this
is the first time that the Coreymacrolactonization
reaction is used under alkali metal cation template. It
should be emphasized that the best guest (Naþ) for host
1 is not the best template for its formation. This can be
because the cavitysize of the transition state conducive
to compound 1 is slightlygreater than the cavityof the
final product.
5. For a synthetic transmembrane polyether model see:
ꢀ
ꢀ
ꢀ
Espınola, C. G.; Perez, R.; Martın, J. D. Org. Lett.
2000, 2, 3161–3164.
6. Tajima, I.; Okada, M.; Sumitomo, H. J. Am. Chem. Soc.
1981, 103, 4096–4100.
7. (a) Burke, S. D.; Porter, W. J.; Rancourt, J.; Kaltenbach,
R. F. Tetrahedron Lett. 1990, 31, 5285–5288; (b) Burke, S.
D.; Heap, C. R.; Porter, W. J.; Song, Y. Tetrahedron Lett.
1996, 37, 343–346; (c) Burke, S. D.; Zhao, Q. J. Org.
Chem. 2000, 65, 1489–1500.
8. (a) Iimori, T.; Erickson, S. D.; Rheingold, A. L.; Still, W.
C. Tetrahedron Lett. 1989, 30, 6947–6950; (b) Li, G.; Still,
W. C. J. Org. Chem. 1991, 56, 6964–6966; (c) Wang, X.;
Erickson, S. D.; Iimori, T.; Still, W. C. J. Am. Chem. Soc.
1992, 114, 4128–4137; (d) Erickson, S. D.; Ohlmeyer, M.
H. J.; Still, W. C. Tetrahedron Lett. 1992, 33, 5925–5928;
(e) Li, G.; Still, W. C. Tetrahedron Lett. 1992, 33, 5929–
5932; (f) Li, G.; Still, W. C. Tetrahedron Lett. 1993, 34,
919–922; (g) Tang, S.; Still, W. C. Tetrahedron Lett. 1993,
34, 6701–6704.
In summary, we have used a new structural unit for the
design of ionophores. Host 1 behaves like a clamp-type
ionophore, where the recognition takes place out of the
macrocyclic plane, with assistance of cis-tetrahydro-
furan oxygen. The template effect has been proved to be
a useful tool to improve the yield and selectivity in the
synthesis of macrodiolide 1. Chiral discrimination and
biological activities of these compounds are being eval-
uated. In addition, this design can be used to obtain new
receptors, bysimplychanging the nature and length of
the spacers.
Supplementary materials
9. Yasumoto, T.; Murata, M. Chem. Rev. 1993, 93, 1897–
1909.
10. Candenas, M. L.; Pinto, F. M.; Cintado, C. G.; Morales,
1H, 13C NMR and FAB spectra for compounds 1, 1ÁKþ,
2, 3, 14 and 15; COSY, HSQC, NOEs spectra for
compounds 1 and 1ÁKþ are available.
ꢀ
ꢀ
ꢀ
E. Q.; Brouard, I.; Dıaz, M. T.; Rico, M.; Rodrıguez, E.;
ꢀ
ꢀ
ꢀ
Rodrıguez, R. M.; Perez, R.; Perez, R. L.; Martın, J. D.
Tetrahedron 2002, 58, 1921–1942.
11. MM2 force field as implemented in Chem3DâPro version
8.0.3 was used to determine the minimized structure.
12. These conditions have been described as an excellent way
to obtain macrodiolides: (a) Garcia, D. M.; Yamada, H.;
Hatakeyama, S.; Nishizawa, M. Tetrahedron Lett. 1994,
Acknowledgements
ꢀ
The authors thank Dr. F. Garcıa-Tellado for helpful
€
35, 3325–3328; (b) Furstner, A.; Albert, M.; Mlynarski, J.;
discussion and the MYCT (PPQ2002-04361-C04-02) of
Spain and the CanaryIslands Government, for sup-
porting this research. R.C. thanks the Spanish MEC for
an FPU fellowship. T.M. thanks the Spanish MCYT-
Matheu, M. J. Am. Chem. Soc. 2002, 124, 1168–1169.
13. Snatzke, G.; Raza, Z.; Habus, I.; Sunjic, V. Carbohydr.
Res. 1988, 182, 179–196.
14. Corey, E. J.; Nicolaou, K. C. J. Am. Chem. Soc. 1974, 96,
5614–5616.
ꢀ
FSE for a Ramon yCajal contract.

R. Carrillo et al. / Tetrahedron Letters 45 (2004) 5215–5219
5219
15. Inanaga, J.; Hirata, K.; Saeki, H.; Katsuki, T.; Yama-
guchi, M. Bull. Chem. Soc. Jpn. 1979, 52, 1989–1993.
16. Size exclusion chromatography(Sephadex LH-20 and
MeOH as the mobile phase) was used in the separation.
17. Nicolaou, K. C.; Hwang, C. K.; Marron, B. E.; DeFrees,
S. A.; Couladouros, E. A.; Abe, Y.; Carroll, P. J.; Snyder,
(m, 8H), 2.06 (d, J ¼ 13:4 Hz, 4H), 3.32–3.68 (m, 12H),
3.99 (m, 4H), 4.05 (d, J ¼ 16:8 Hz, 4H), 4.21 (d,
J ¼ 16:8 Hz, 4H), 5.01 (s, 4H); 13C NMR (d, CDCl3):
20.5 (t), 27.6 (t), 68.0 (d), 68.2 (t), 71.4 (t), 77.2 (d), 169.8
(s); IR (film) (cmÀ1): 2954, 2855, 2362, 1751, 1437, 1202,
1132, 1094; MS (FAB) m=z (relative intensity): 711
[MþNa]þ (8), 367 (5), 173 (16), 97 (100), 69 (17); HRMS
J. P. J. Am. Chem. Soc. 1990, 112, 3040–3054.
25
D
18. Compound 1: ½aꢀ À31.2 (c 0.33, CHCl3); 1H NMR (d,
(FAB): calcd for C32H48O16Na [MþNa]þ: 711.2840,
25
D
CDCl3): 1.41 (d, J ¼ 12:6 Hz, 2H), 1.68–1.77 (m, 2H),
1.85–1.97 (m, 2H), 2.10 (d, J ¼ 14:2 Hz, 2H), 3.50 (ddd,
J ¼ 1:8, 1.8, 11.8 Hz, 2H), 3.62 (dd, J ¼ 5:4, 5.4 Hz, 2H),
3.67–3.77 (m, 4H), 4.05 (d, J ¼ 14:2 Hz, 4H), 4.16 (d,
J ¼ 14:3 Hz, 2H), 5.08 (s, 2H); 13C NMR (d, CDCl3): 20.6
(t), 27.2 (t), 68.1 (t), 68.2 (d), 70.4 (t), 70.8 (t), 76.2 (d),
169.4 (s); IR (film) (cmÀ1): 2928, 2854, 1736, 1217, 1095;
MS (FAB) m=z (relative intensity): 367 [MþNa]þ (19), 147
(6), 135 (8), 109 (12), 81 (51), 69 (100); HRMS (FAB):
found: 711.2793. Compound 15: mp 152–155 °C; ½aꢀ
1
þ28.2 (c 1.3, CHCl3); H NMR (d, CDCl3): 1.48 (dddd,
J ¼ 4:6, 12.0, 12.0, 12.0 Hz 4H), 1.68–1.80 (m, 8H), 2.16–
2.20 (m, 4H), 3.34–3.46 (m, 8H), 3.54 (dd, J ¼ 6:0,
10.0 Hz, 4H), 3.73 (d, J ¼ 10:0 Hz, 4H), 3.95 (d,
J ¼ 10:6 Hz, 4H), 4.11 (dd, J ¼ 17:0, 17.0 Hz, 8H), 4.76
(ddd, J ¼ 4:7, 10.4, 10.4 Hz, 4H); 13C NMR (d, CDCl3):
24.8 (t), 29.1 (t), 67.7 (t), 68.6 (t), 68.7 (d), 71.5 (t), 78.7 (t),
169.5 (s), 20.5 (t), 27.6 (t), 68.0 (d), 68.2 (t), 71.4 (t), 77.2
(d), 169.8 (s); IR (film) (cmÀ1): 2949, 2857, 1753, 1456; MS
(FAB) m=z (relative intensity): 711 [MþNa]þ (66), 689
[MþH]þ (11), 173 (14), 136 (39), 97 (100), 73 (28); HRMS
(FAB): calcd for C32H49O16 [MþH]þ: 689.3021, found:
689.2986.
calcd for C16H24O8Na [MþNa]þ: 367.1369, found:
25
367.1387. Compound 2: mp 108–111 °C; ½aꢀ þ44.1 (c
D
1.4, CHCl3); 1H NMR (d, CDCl3): 1.18–1.23 (m, 1H),
1.45–1.75 (m, 6H), 2.26–2.31 (m, 1H), 3.41 (ddd, J ¼ 3:1,
11.4, 11.4 Hz, 1H), 3.48–3.68 (m, 3H), 3.93–3.98 (m, 2H),
4.29 (d, J ¼ 16:9 Hz, 1H), 4.50 (ddd, J ¼ 4:9, 10.3,
10.3 Hz, 1H); 13C NMR (d, CDCl3): 24.6 (t), 29.1 (t),
67.6 (t), 68.8 (t), 69.3 (d), 71.9 (t), 78.7 (d), 169.6 (s); IR
(film) (cmÀ1): 2949, 2863, 1734, 1261; MS (FAB) m=z
(relative intensity): 367 [MþNa]þ (7), 345 [MþH]þ (30),
307 (14), 173 (31), 97 (93); HRMS (FAB): calcd for
C16H25O8 [MþH]þ: 345.1549, found: 345.1558. Com-
pound 3: mp 134–136 °C; 1H NMR (d, CDCl3): 1.97
(dddd, J ¼ 5:6, 5.6, 5.6, 5.6 Hz, 2H), 3.66 (dd, J ¼ 5:9,
5.9 Hz, 2H), 4.11 (s, 2H), 4.38 (dd, J ¼ 5:5, 5.5 Hz, 2H);
13C NMR (d, CDCl3): 28.5 (t), 60.8 (t), 66.8 (t), 69.1 (t),
170.7 (s); MS (FAB) m=z (relative intensity): 255 [MþNa]þ
(4), 233 [MþH]þ (24), 154 (100), 137 (73); HRMS (FAB):
19. Koenig, K. E.; Lein, G. M.; Stuckler, L. P.; Kaneda,
T.; Cram, D. J. J. Am. Chem. Soc. 1979, 101, 3553–
3566.
20. Bourgoin, M.; Wong, K. H.; Hui, J. Y.; Smid, J. J. Am.
Chem. Soc. 1975, 97, 3462–3467.
21. Complex 1ÆKþ: 1H NMR (d, CDCl3): 1.47 (d, J ¼ 13:7 Hz,
2H), 1.76 (m, 2H), 1.96 (m, 2H), 2.12 (d, J ¼ 14:4 Hz, 2H),
3.54 (ddd, J ¼ 2:1, 11.9, 11.9 Hz, 2H), 3.60 (m, 2H), 3.64
(dd, J ¼ 3:9, 9.7 Hz, 2H), 3.82 (dd, J ¼ 5:8, 9.7 Hz, 2H),
4.02 (d, J ¼ 14:3 Hz, 2H), 4.16 (m, 2H), 4.21 (d,
J ¼ 14:3 Hz, 2H), 5.13 (s, 2H); 13C NMR (d, CDCl3):
20.6 (t), 27.5 (t), 68.6 (t), 69.1 (d), 70.5 (t), 71.4 (t), 76.3 (d),
169.0 (s); MS (FAB) m=z (relative intensity): 383 [MþK]þ
(100), 367 [MþNa]þ (10), 154 (14), 136 (15), 95 (14), 69
(36); HRMS (FAB): calcd for C16H24O8K [MþK]þ:
383.11083, found: 383.11134.
calcd for C10H17O6 [MþH]þ: 233.1025, found: 233.1035.
25
Compound 14: mp 135–139 °C; ½aꢀ À42.7 (c 1.4, CHCl3);
D
1H NMR (d, CDCl3): 1.40 (d, J ¼ 13:0 Hz, 4H), 1.57–1.92