A new synthetic approach to mycobacterial cell wall α-(1→5)-D-arabinofuranosyl C-oligosaccharides

Article abstract of DOI:10.1016/S0040-4039(03)00844-X

Three designed arabinofuranose building blocks allowed the diastereoselective synthesis of a C-disaccharide and a C-trisaccharide by Wittig olefination. The latter compound represents the first example of all-carbon linked arabinofuranotriose analogue.

Full text of DOI:10.1016/S0040-4039(03)00844-X

TETRAHEDRON
LETTERS
Pergamon
Tetrahedron Letters 44 (2003) 4067–4071
A new synthetic approach to mycobacterial cell wall
a-(1“5)- -arabinofuranosyl C-oligosaccharides
D
Alessandro Dondoni* and Alberto Marra*
Dipartimento di Chimica, Laboratorio di Chimica Organica, Universita` di Ferrara, Via L. Borsari 46, I-44100 Ferrara, Italy
Received 10 February 2003; revised 31 March 2003; accepted 1 April 2003
Abstract—Three designed arabinofuranose building blocks allowed the diastereoselective synthesis of a C-disaccharide and a
C-trisaccharide by Wittig olefination. The latter compound represents the first example of all-carbon linked arabinofuranotriose
analogue. © 2003 Elsevier Science Ltd. All rights reserved.
The polysaccharidic chains which form the cell wall of
mycobacteria are constituted mainly by -arabinofura-
nose units connected through a-1,5-O-glycosidic link-
ages.1
Since these pentofuranoses and the
O-arabinofuranoside, via nitro-aldol condensation
prompted us to report on our alternative synthetic
D
approach
isosteres.
to
oligoarabinofuranoside
methylene
corresponding glycoconjugates are xenobiotic to
humans, inhibitors of their biosynthesis are potential
therapeutic agents against mycobacterial infections
such as tuberculosis and leprosy.2 Thus, over the past
few years, many efforts have been devoted toward the
preparation of O-oligoarabinofuranosides A (Fig. 1) in
order to elucidate the biosynthesis and evaluate the
immunological properties of the naturally occurring
arabinose-based polysaccharides.3 On the other hand,
despite a large body of knowledge on C-glycoside
synthesis,4 the preparation of hydrolytically and enzy-
matically resistant analogues, i.e. carbon-linked ara-
binofuranosyl oligosaccharides B (Fig. 1), has not been
described until a few months ago. The recent paper5 by
Gurjar et al. dealing with the synthesis of a single
Taking advantage of our experience on the stereoselec-
tive synthesis of C-oligosaccharides by iterative Wittig
olefination,6 we developed new building blocks and a
modified protocol for the straightforward assembly of
C-oligoarabinofuranosides 1 (Scheme 1). While the two
monofunctionalized sugar moieties 2 and 4 were suit-
able precursors to the head and tail of the oligosaccha-
ridic chain, the difunctionalized compound 3 was
envisaged as repeating unit. In particular, the building
block 3 was designed to allow, after the Wittig cou-
pling, the regeneration of the formyl group in a single
step under conditions which do not affect the chemical
and stereochemical integrity of the growing saccharidic
chain. Interestingly, all C-monosaccharides 2–4
appeared to be accessible in a multigram scale from the
same perbenzylated thiazolyl C-arabinoside 5 exploit-
ing the well established thiazole-to-formyl conversion.7
Moreover, di- and oligosaccharide derivatives origi-
nated from the Wittig carbonꢀcarbon bond formation
reaction could be readily converted into the target
compounds 1 by hydrogenation without any other
manipulation of the functional groups.
compound, the methyl 5-(a-
D
-C-arabinofuranosyl)-a-D-
The addition at low temperature of 2-lithiothiazole,
prepared in situ from 2-bromothiazole and butyl-
lithium, to the lactone8 6 (9.6 g, 23 mmol), followed by
the acetylation of the crude thiazolylketose, gave the
1-O-acetyl derivative9 7 as a 4:1 mixture of anomers in
84% overall yield after column chromatography
(Scheme 2). The deoxygenation of 7 was carried out
Figure 1. Natural a-(1“5)-D-arabinofuranosyl oligosaccha-
ride (A) and its methylene isostere (B).
Keywords: arabinogalactan; C-glycosides; sugar aldehydes; thiazole.
* Corresponding author. Fax: +39-0532-291167; e-mail: adn@
dns.unife.it
0040-4039/03/$ - see front matter © 2003 Elsevier Science Ltd. All rights reserved.
doi:10.1016/S0040-4039(03)00844-X

4068
A. Dondoni, A. Marra / Tetrahedron Letters 44 (2003) 4067–4071
Scheme 2. Th=2-thiazolyl. Reagents and conditions: (a) 2-
lithiothiazole, dry Et2O, −75°C, 1 h; (b) Ac2O, Et3N, CH2Cl2,
rt, 48 h; (c) 0.1 M SmI2, (CH2OH)2, dry THF, rt, 30 min; (d)
MeOTf, CH3CN, rt, 15 min; then NaBH4, MeOH, rt, 5 min;
then AgNO3, CH3CN, H2O, rt, 10 min; (e) NaBH4, Et2O,
MeOH, rt, 15 min; (f) I2, PPh3, imidazole, toluene, 80°C, 1 h;
(g) neat PPh3, 120°C, 4 h.
Scheme 1. Retrosynthetic approach to 1,5-C-oligoarabino-
furanosides.
tive 13 by the above mentioned iodination and phos-
phanation reactions (76%, two steps).
with samarium(II) iodide6f in the presence of ethylene
Aiming at preparing the all-carbon linked diarabino-
furanoside 15 (Scheme 4), the formyl C-glycoside 2 was
glycol to afford the desired a-
-C-glycoside10 5 (60%)
D
which was easily separated from the b-anomer (16%).
The anomeric configuration of the latter compounds
was assigned by NOE difference experiments. Upon
irradiation of the H-1 proton of both anomers, a
significant NOE with H-3 was observed only for 5, thus
indicating a cis-relationship between these protons. The
key intermediate 5 was then submitted to the usual
unmasking sequence7,11 involving N-methylation,
hydride reduction, and silver-assisted hydrolysis of the
thiazolidine intermediate to give, without any epimer-
ization, the a-linked formyl C-arabinoside12 2 in good
yield (85%). The second monofunctionalized building
block, i.e. the tribenzylated phosphonium salt13 4, was
obtained from the aldehyde 2 by reduction to alcohol,
iodination14 to 8, and treatment of the latter with neat
triphenylphosphine at 120°C (69%, three steps).
The preparation of the third building block 3 from the
thiazolyl C-glycoside 5 required a longer reaction
sequence, although each step involved high-yielding
transformations of the functional groups and no purifi-
cation of some intermediates was necessary (Scheme 3).
Selective removal of the 5-O-benzyl group by acetolysis
gave the alcohol 9 (82%) which was silylated to 10 and
this converted into the formyl C-arabinofuranoside 11.
This crude aldehyde was protected as diisopropyl acetal
and desilylated to give pure15 12 in 43% overall yield
from 10. The alcohol 12 was finally transformed into
the target phosphonium salt16 3 via the iodide deriva-
Scheme 3. Th=2-thiazolyl. Reagents and conditions: (a)
Ac2O, AcOH, H2SO4, rt, 1 h; then Et3N, MeOH, H2O, rt, 18
h; (b) t-BuPh2SiCl, Py, DMAP, rt, 24 h; (c) MeOTf, CH3CN,
rt, 15 min; then NaBH4, MeOH, rt, 5 min; then AgNO3,
CH3CN, H2O, rt, 10 min; (d) CH(Oi-Pr)3, BF3·Et2O, dry
CH2Cl2, rt, 45 min; (e) Bu4NF, THF, rt, 2 h; (f) I2, PPh3,
imidazole, toluene, 80°C, 1 h; (g) neat PPh3, 120°C, 4 h.

A. Dondoni, A. Marra / Tetrahedron Letters 44 (2003) 4067–4071
4069
,
Scheme 4. Reagents and conditions: (a) BuLi, 3:1 dry THF–HMPA, 4 A MS, −20°C, 2 h; (b) H2, Pd(OH)2/C, AcOEt–MeOH, rt,
3 h; (c) TFA, H2O, THF, rt, 2 h.
allowed to react at −20°C with an equimolar amount of
the phosphorane generated from 4 and butyllithium.
The Wittig coupling was performed in the presence of
construction of higher oligomers by the use of several
units of the difunctionalized building block 3. Hence
the synthesis of other C-oligosaccharides is currently
underway in our laboratory.
,
activated 4 A molecular sieves using a 3:1 mixture of
anhydrous THF and HMPA as the solvent to give a
Z,E mixture of C-disaccharide 14 in 65% isolated yield.
The reduction of the double bond and the removal of
the benzyl groups was quantitatively accomplished by
hydrogenation over Pd(OH)2 to afford the polyol 15
References
1. (a) Chatterjee, D. Curr. Opin. Chem. Biol. 1997, 1, 579–
588; (b) Crick, D. C.; Mahapatra, S.; Brennan, P. J.
Glycobiology 2001, 11, 107R–118R.
1
which was fully characterized as perbenzoate.17 The H
NMR spectrum of the latter compound clearly showed
a C2 symmetric structure, therefore proving that the
2. Kremer, L.; Besra, G. S. Expert Opin. Investig. Drugs
a-
D
configuration of both monosaccharidic reagents 2
2002, 11, 1033–1049.
and 4 was retained during the Wittig reaction.
3. Selected recent references: (a) Sanchez, S.; Bamhaoud, T.;
Prandi, J. Tetrahedron Lett. 2000, 41, 7447–7452; (b)
D’Souza, F. W.; Ayers, J. D.; McCarren, P. R.; Lowary,
T. L. J. Am. Chem. Soc. 2000, 122, 1251–1260; (c)
Gurjar, M. K.; Reddy, L. K.; Hotha, S. J. Org. Chem.
2001, 66, 4657–4660; (d) D’Souza, F. W.; Yin, H.;
Lowary, T. L. J. Org. Chem. 2002, 67, 892–903; (e)
Sanchez, S.; Bamhaoud, T.; Prandi, J. Eur. J. Org. Chem.
2002, 3864–3873.
4. (a) Postema, M. H. D. In C-Glycoside Synthesis; Rees, C.
W., Ed.; CRC Press: Boca Raton, FL, 1995; (b) Levy, D.
E.; Tang, C. In The Chemistry of C-Glycosides; Baldwin,
J. E.; Magnus, P. D., Eds.; Elsevier Science: Oxford,
1995; (c) Beau, J.-M.; Gallagher, T. Top. Curr. Chem.
1997, 187, 1–54; (d) Dondoni, A.; Marra, A. Chem. Rev.
2000, 100, 4395–4421; (e) Beau, J.-M.; Vauzeilles, B.;
Skrydstrup, T. In Glycoscience. Chemistry and Chemical
Biology; Fraser-Reid, B.; Tatsuta, K.; Thiem, J., Eds.;
Springer Verlag: Heidelberg, 2001; Vol. 3, pp. 2679–2724.
5. Gurjar, M. K.; Nagaprasad, R.; Ramana, C. V. Tetra-
hedron Lett. 2002, 43, 7577–7579.
Targeting a higher C-oligoarabinoside homologue, the
aldehyde 2 was coupled with the phosphorane derived
from 3 as described above (Scheme 4). The (Z,E)-C-
disaccharide18 16, isolated in 49% yield, upon treatment
with aqueous trifluoroacetic acid at rt nicely gave
almost pure aldehyde18 17 which was used for the
following step without further purification. When 17
was coupled with the tribenzylated phosphorane pre-
pared from 4, the bisalkene 18 and the unreacted
aldehyde 17 were recovered by column chromatography
in 45 and 25% yield, respectively. Then the C-trisaccha-
ride 18 was hydrogenated over Pd(OH)2 to afford
quantitatively the a-D-arabinotriose methylene isostere
19, characterized as perbenzoate derivative.18 The C2
symmetry of the final product was substantiated by the
NMR spectra, therefore the configuration of the stere-
ocenters next to the reactive functionalities was not
modified through the chain elongation sequence.
In conclusion, our approach to totally carbon-linked
oligoarabinofuranosides appears to be more efficient
than that of Gurjar5 because it requires less steps for
the elaboration of the coupling product and because it
is iteratively repeatable. Although only a C-trisaccha-
ride has been prepared, the method is suitable for the
6. (a) Dondoni, A.; Boscarato, A.; Zuurmond, H.-M. Tetra-
hedron Lett. 1996, 37, 7587–7590; (b) Dondoni, A.; Zuur-
mond, H. M.; Boscarato, A. J. Org. Chem. 1997, 62,
8114–8124; (c) Dondoni, A.; Kleban, M.; Zuurmond, H.;
Marra, A. Tetrahedron Lett. 1998, 39, 7991–7994; (d)
Dondoni, A.; Giovannini, P. P.; Marra, A. Tetrahedron

4070
A. Dondoni, A. Marra / Tetrahedron Letters 44 (2003) 4067–4071
Lett. 2000, 41, 6195–6199; (e) Dondoni, A.; Mizuno, M.;
Marra, A. Tetrahedron Lett. 2000, 41, 6657–6660; (f)
Dondoni, A.; Formaglio, P.; Marra, A.; Massi, A. Tetra-
hedron 2001, 57, 7719–7727; (g) Dondoni, A.; Marra, A.;
Mizuno, M.; Giovannini, P. P. J. Org. Chem. 2002, 67,
4186–4199; (h) Dondoni, A.; Giovannini, P. P.; Perrone,
D. J. Org. Chem. 2002, 67, 7203–7214.
aldehyde from tetra-O-benzyl-D-mannopyranose (six
steps, 15% overall yield) and its spectroscopic characteri-
zation have been recently described, see: Persky, R.;
Albeck, A. J. Org. Chem. 2000, 65, 5632–5638. The
NMR data of 2, prepared by us as shown in Scheme 2,
are in agreement with those reported by Persky and
Albeck. Gurjar and co-workers (see Ref. 5) prepared the
same aldehyde from
D-glucosamine hydrochloride (seven
7. For recent reviews, see: (a) Dondoni, A., Marra, A. In
Preparative Carbohydrate Chemistry; Hanessian, S., Ed.;
Marcel Dekker: New York, 1997; pp. 173–205; (b) Don-
doni, A. Synthesis 1998, 1681–1706; (c) Dondoni, A. Pure
Appl. Chem. 2000, 72, 1577–1588.
8. Known lactone 6 (Meng, Q.; Hesse, M. Helv. Chim. Acta
1991, 74, 445–449) was prepared on a 5–10 g scale in
quantitative yield by oxidation of commercially available
steps, 22%) but no reference to previous syntheses was
given in their paper.
1
13. Compound 4. H NMR (400 MHz, CDCl3): l 7.82–7.71,
7.63–7.57, and 7.36–7.16 (3 m, 30 H, 6 Ph), 4.96 (ddd,
1H, J1a,2=3.0, J1a,1b=15.5, J1a,P=13.6 Hz, H-1a), 4.84
(bs, 1H, H-3), 4.81 and 4.74 (2 d, 2H, J=11.9 Hz,
PhCH2), 4.71 and 4.49 (2 d, 2H, J=11.9 Hz, PhCH2),
4.58 (ddd, 1H, J1b,2=11.3, J2,P=6.0 Hz, H-2), 4.37 and
2,3,5-tri-O-benzyl-D-arabinose (Sigma) with pyridinium
4.34 (2 d, 2H, J=11.5 Hz, PhCH2), 3.96 (bd, 1H, J4,5
=
chlorochromate (Corey, E. J.; Suggs, J. W. Tetrahedron
Lett. 1975, 16, 2647–2650).
1.6 Hz, H-4), 3.86 (ddd, 1H, J5,6a=6.9, J5,6b=6.4 Hz,
H-5), 3.77 (ddd, 1H, J1b,P=11.3 Hz, H-1b), 3.38 (dd, 1H,
1
9. Compound a7. [h]D=+43.9 (c 1, CHCl3). H NMR (300
J
6a,6b=9.6 Hz, H-6a), 3.32 (dd, 1H, H-6b). 31P NMR
MHz, CDCl3): l 7.82 and 7.39 (2 d, 2H, J=3.1 Hz, Th),
7.36–7.26 and 7.04–7.00 (2 m, 15 H, 3 Ph), 4.66 and 4.61
(2 d, 2H, J=12.0 Hz, PhCH2), 4.54 and 4.49 (2 d, 2H,
J=11.7 Hz, PhCH2), 4.53 (d, 1H, J2,3=2.5 Hz, H-2), 4.50
(ddd, 1H, J3,4=4.6, J4,5a=4.7, J4,5b=5.5 Hz, H-4), 4.35
and 4.28 (2 d, 2H, J=11.6 Hz, PhCH2), 4.18 (dd, 1H,
H-3), 3.82 (dd, 1H, J5a,5b=10.8 Hz, H-5a), 3.75 (dd, 1H,
H-5b), 2.15 (s, 3 H, Ac). 13C NMR (75 MHz, CDCl3): l
107.9 (C-1). Compound b7. [h]D=−35.6 (c 0.5, CHCl3).
1H NMR (300 MHz, CDCl3): l 7.82 and 7.38 (2 d, 2H,
J=3.1 Hz, Th), 7.37–7.23 (m, 15 H, 3 Ph), 4.89 and 4.61
(2 d, 2H, J=11.8 Hz, PhCH2), 4.68 and 4.58 (2 d, 2H,
J=11.6 Hz, PhCH2), 4.54 (s, 2H, PhCH2), 4.53 (dd, 1H,
(121 MHz, CDCl3): l 24.6.
1
14. Compound 8. [h]D=−10.0 (c 0.9, CHCl3). H NMR (300
MHz, CDCl3): l 7.40–7.24 (m, 15 H, 3 Ph), 4.60 and 4.55
(2 d, 2H, J=12.0 Hz, PhCH2), 4.58 (s, 2H, PhCH2), 4.55
and 4.50 (2 d, 2H, J=12.0 Hz, PhCH2), 4.35 (ddd, 1H,
J1a,2=J1b,2=6.0, J2,3=3.3 Hz, H-2), 4.24 (ddd, 1H, J4,5=
3.0, J5,6a=8.2, J5,6b=5.5 Hz, H-5), 4.12 (dd, 1H, J3,4=2.0
Hz, H-4), 4.08 (dd, 1H, H-3), 3.61 (dd, 1H, J1a,1b=10.0
Hz, H-1a), 3.56 (dd, 1H, H-1b), 3.38 (dd, 1H, J6a,6b=10.1
Hz, H-6a), 3.30 (dd, 1H, H-6b). This compound has been
previously obtained as a diastereomeric mixture (Reitz,
A. B.; Nortey, S. O.; Maryanoff, B. E.; Liotta, D.;
Monahan, R. J. Org. Chem. 1987, 52, 4191–4202) and as
a pure product (McGurk, P.; Chang, G. X.; Lowary, T.
L.; McNeil, M.; Field, R. A. Tetrahedron Lett. 2001, 42,
2231–2234) but no physical and spectroscopic data were
reported.
J
2,3=7.1, J3,4=6.8 Hz, H-3), 4.40 (d, 1H, H-2), 4.35 (ddd,
1H, J4,5a=4.1, J4,5b=4.5 Hz, H-4), 3.71 (dd, 1H, J5a,5b
=
10.7 Hz, H-5a), 3.62 (dd, 1H, H-5b), 1.99 (s, 3 H, Ac).
13C NMR (75 MHz, CDCl3): l 103.2 (C-1).
10. Compound a5. [h]D=+34.4 (c 0.7, CHCl3). 1H NMR
(400 MHz, CDCl3): l 7.78 (d, 1H, J=3.3 Hz, Th),
7.35–7.24 and 7.13–6.99 (2 m, 16 H, 3 Ph, Th), 5.43 (d,
1H, J1,2=3.1 Hz, H-1), 4.74 and 4.59 (2 d, 2H, J=11.8
Hz, PhCH2), 4.61 and 4.57 (2 d, 2H, J=12.1 Hz,
PhCH2), 4.54 (dd, 1H, J2,3=2.7 Hz, H-2), 4.47 and 4.43
(2 d, 2H, J=12.0 Hz, PhCH2), 4.45 (ddd, 1H, J3,4=4.3,
15. A strong NOE between the CH(Oi-Pr)2 (H-1) and H-3
protons confirmed the a-D configuration of the C-ara-
binofuranoside derivative 12. [h]D=+14.1 (c 0.9, CHCl3).
1H NMR (400 MHz, CDCl3): l 7.39–7.28 (m, 10 H, 2
Ph), 4.67 (d, 1H, J1,2=5.8 Hz, H-1), 4.65 and 4.50 (2 d,
2H, J=11.7 Hz, PhCH2), 4.56 and 4.49 (2 d, 2H, J=11.8
Hz, PhCH2), 4.24 (dd, 1H, J2,3=2.6, J3,4=2.8 Hz, H-3),
4.13 (ddd, 1H, J4,5=5.2, J5,6a=3.3, J5,6b=5.1 Hz, H-5),
4.04 (dd, 1H, H-2), 4.02 (dd, 1H, H-4), 3.94 (qq, 1H,
J=6.2 Hz, CHMe2), 3.90 (qq, 1H, J=6.2 Hz, CHMe2),
3.75 (ddd, 1H, J6a,6b=11.8, J6a,OH=5.7 Hz, H-6a), 3.65
(ddd, 1H, J6b,OH=6.6 Hz, H-6b), 1.99 (dd, 1H, OH),
1.23, 1.22, 1.19, and 1.08 (4 d, 12 H, 4 Me).
J4,5a=5.8, J4,5b=5.5 Hz, H-4), 4.14 (dd, 1H, H-3), 3.68
(dd, 1H, J5a,5b=10.3 Hz, H-5a), 3.65 (dd, 1H, H-5b).
Compound b5. [h]D=+2.4 (c 1, CHCl3). 1H NMR (300
MHz, CDCl3): l 7.84 (d, 1H, J=3.2 Hz, Th), 7.40–7.24
and 7.04–6.99 (2 m, 16 H, 3 Ph, Th), 5.58 (d, 1H,
J
1,2=3.8 Hz, H-1), 4.66 and 4.60 (2 d, 2H, J=12.0 Hz,
PhCH2), 4.53 (s, 2H, PhCH2), 4.31 (ddd, 1H, J3,4=2.7,
4,5a=6.0, J4,5b=6.6 Hz, H-4), 4.29 (dd, 1H, J2,3=1.0 Hz,
1
16. Compound 3. H NMR (400 MHz, C6D6): l 7.77–7.63,
J
7.38–7.30, and 7.18–6.92 (3 m, 25 H, 5 Ph), 6.31 (ddd,
1H, J5,6a=3.0, J6a,6b=J6a,P=15.0 Hz, H-6a), 5.25 and
5.15 (2 d, 2H, J=12.1 Hz, PhCH2), 5.21 (bs, 1H, H-4),
5.04 (ddd, 1H, J5,6b=11.0, J5,P=6.2 Hz, H-5), 4.72 (d,
1H, J1,2=8.0 Hz, H-1), 4.59 and 4.34 (2 d, 2H, J=11.9
Hz, PhCH2), 4.18 (bs, 1H, H-3), 3.86 (bd, 1H, H-2), 3.74
(qq, 1H, J=6.1 Hz, CHMe2), 3.69 (ddd, 1H, J6b,P=11.0
Hz, H-6b), 3.57 (qq, 1H, J=6.1 Hz, CHMe2), 1.05, 0.92,
0.87, and 0.86 (4 d, 12 H, 4 Me). 31P NMR (121 MHz,
CDCl3): l 24.7.
H-2), 4.23 and 4.18 (2 d, 2H, J=12.0 Hz, PhCH2), 4.08
(dd, 1H, H-3), 3.80 (dd, 1H, J5a,5b=10.0 Hz, H-5a), 3.69
(dd, 1H, H-5b).
11. Dondoni, A.; Marra, A. Tetrahedron Lett. 2003, 44,
13–16.
12. Crude aldehyde 2 was previously obtained by oxidative
ring contraction of the tri-O-benzyl-D-glucal (Kaye, A.;
Neidle, S.; Reese, C. B. Tetrahedron Lett. 1988, 29,
1841–1844; Bettelli, E.; D’Andrea, P.; Mascanzoni, S.;
Passacantilli, P.; Piancatelli, G. Carbohydr. Res. 1998,
306, 221–230) but no physical and spectroscopic data
were provided for this product. The synthesis of this
1
17. Compound 15Bz. [h]D=−27.8 (c 0.4, CHCl3). H NMR
(400 MHz, CDCl3): l 8.08–7.99, 7.60–7.50, 7.44–7.39,
and 7.36–7.32 (4 m, 30 H, 6 Ph), 5.67 (dd, 2H, J2,3
=

A. Dondoni, A. Marra / Tetrahedron Letters 44 (2003) 4067–4071
4071
J
J
J
10,11=3.4, J3,4=J9,10=2.1 Hz, H-3, H-10), 5.51 (dd, 2H,
4,5=J8,9=3.6 Hz, H-4, H-9), 4.70 (dd, 2H, J1a,2
11,12a=5.8, J1a,1b=J12a,12b=11.8 Hz, H-1a, H-12a), 4.65
MHz, CDCl3): l 9.57 (d, 1H, J1,2=1.0 Hz, H-1), 7.39–
7.17 (m, 25 H, 5 Ph), 5.81 (dd, 1H, J5,6=8.5, J6,7=11.3
Hz, H-6), 5.74 (dd, 1H, J7,8=8.0 Hz, H-7), 5.12 (dd, 1H,
=
(dd, 2H, J1b,2=J11,12b=4.5 Hz, H-1b, H-12b), 4.53 (ddd,
2H, H-2, H-11), 4.46–4.41 (m, 2H, H-5, H-8), 2.19–2.02
(m, 4 H, 2 H-6, 2 H-7). MALDI-TOF (918.96): 919.7
(M+H), 941.4 (M+Na), 957.5 (M+K).
J
4,5=2.0 Hz, H-5), 4.91 (dd, 1H, J8,9=4.8 Hz, H-8), 4.62
and 4.52 (2 d, 2H, J=11.8 Hz, PhCH2), 4.58 and 4.51 (2
d, 2H, J=11.9 Hz, PhCH2), 4.54 (s, 2H, PhCH2), 4.51 (s,
2H, PhCH2), 4.47 and 4.41 (2 d, 2H, J=11.5 Hz,
PhCH2), 4.46 (dd, 1H, J2,3=1.8 Hz, H-2), 4.24 (ddd, 1H,
18. Compound Z-16. 1H NMR (400 MHz, C6D6): l 7.36–
7.00 (m, 25 H, 5 Ph), 5.89 (dd, 1H, J5,6=7.7, J6,7=11.3
Hz, H-6), 5.85 (dd, 1H, J7,8=8.0 Hz, H-7), 5.31 (dd, 1H,
J
10,11=4.0, J11,12a=5.6, J11,12b=5.8 Hz, H-11), 4.20 (dd,
1H, J3,4=1.5 Hz, H-3), 4.08 (dd, 1H, J9,10=3.3 Hz,
H-10), 3.99 (dd, 1H, H-9), 3.98 (dd, 1H, H-4), 3.57 (dd,
1H, J12a,12b=10.0 Hz, H-12a), 3.54 (dd, 1H, H-12b).
J4,5=5.5 Hz, H-5), 5.26 (dd, 1H, J8,9=4.8 Hz, H-8), 4.72
(d, 1H, J1,2=5.2 Hz, H-1), 4.64 and 4.51 (2 d, 2H,
J=11.9 Hz, PhCH2), 4.62 and 4.47 (2 d, 2H, J=12.0 Hz,
PhCH2), 4.61 and 4.42 (2 d, 2H, J=12.0 Hz, PhCH2),
4.53 (dd, 1H, J2,3=3.0, J3,4=3.3 Hz, H-3), 4.43 (s, 2H,
PhCH2), 4.42 (ddd, 1H, J10,11=4.0, J11,12a=J11,12b=6.2
Hz, H-11), 4.33 (dd, 1H, H-2), 4.25 (s, 2H, PhCH2), 4.25
(dd, 1H, J9,10=3.3 Hz, H-10), 4.12 (dd, 1H, H-4), 4.07
(dd, 1H, H-9), 3.78 (qq, 1H, J=6.1 Hz, CHMe2), 3.76
(qq, 1H, J=6.1 Hz, CHMe2), 3.54 (dd, 1H, J12a,12b=9.8
Hz, H-12a), 3.50 (dd, 1H, H-12b), 1.09, 1.08, 1.04, and
1
Compound 19Bz. [h]D=−25.6 (c 0.9, CHCl3). H NMR
(400 MHz, C6D6): l 8.20–8.16, 8.10–8.03, and 7.10–6.92
(3 m, 40 H, 8 Ph), 5.76 (dd, 2H, J2,3=3.2, J3,4=1.9 Hz,
H-3, H-16), 5.60 (dd, 2H, J4,5=3.3 Hz, H-4, H-15),
5.56–5.54 (m, 2H, H-9, H-10), 4.64 (d, 4 H, J1,2=5.6 Hz,
2 H-1, 2 H-18), 4.39 (dt, 2H, H-2, H-17), 4.34–4.29 (m,
2H, H-5, H-14), 4.25–4.20 (m, 2H, H-8, H-11), 2.23–2.12
and 2.02–1.92 (2 m, 8 H, 2 H-6, 2 H-7, 2 H-12, 2 H-13).
MALDI-TOF (1257.33): 1279.9 (M+Na), 1295.9 (M+K).
1
0.99 (4 d, 12 H, 4 Me). Compound Z-17. H NMR (400