A Total Synthesis of the Methyl Glycoside of Ganglioside GM

Article abstract of DOI:10.1021/jo9912496

The total synthesis of the methyl glycoside of GM₁ (1b) has been accomplished. The key step in the synthesis involves the sulfonamidoglycosidation reaction, which is used to create a β-linkage leading to a GalNAc residue joined to the C4 hydroxyl group of a galactose unit of a C3 sialylated lactosyl moiety. The "proximal hydroxyl" directing effect, which has been postulated before, manifests in this context as well leading to the preponderant formation of the β-glycoside. Together with asialo GM₁ and other substructures, the GM₁ methyl glycoside has been submitted for biological assays as potential ligands for bacterial and viral infection sites.

Full text of DOI:10.1021/jo9912496

144
J . Org. Chem. 2000, 65, 144-151
A Tota l Syn th esis of th e Meth yl Glycosid e of Ga n gliosid e GM1
Samit K. Bhattacharya† and Samuel J . Danishefsky*,†,‡
Department of Chemistry, Columbia University, Havemeyer Hall, New York, New York 10027, and
The Laboratory for Bioorganic Chemistry, The Sloan-Kettering Institute for Cancer Research, 1275 York
Avenue, New York, New York 10021
Received August 5, 1999
The total synthesis of the methyl glycoside of GM1 (1b) has been accomplished. The key step in the
synthesis involves the sulfonamidoglycosidation reaction, which is used to create a â-linkage leading
to a GalNAc residue joined to the C4 hydroxyl group of a galactose unit of a C3 sialylated lactosyl
moiety. The “proximal hydroxyl” directing effect, which has been postulated before, manifests in
this context as well leading to the preponderant formation of the â-glycoside. Together with asialo
GM1 and other substructures, the GM1 methyl glycoside has been submitted for biological assays
as potential ligands for bacterial and viral infection sites.
Ba ck gr ou n d
Bronchial infection by microorganisms, particularly
Pseudomonas aeruginosa, is apparently responsible for
the morbidity and mortality of victims of cystic fibrosis
(CF). Al Awqati and co-workers have found that CF
bronchial and pancreatic epithelia reversibly bind the
microorganism P. aeruginosa.1,2 Strong evidence was put
forward to suggest that asialo GM1 (2a , see Figure 1),
which is bound through a hydrophobic attraction to the
apical membrane of CF epithelia, is a likely binding site
for P. aeruginosa and that its increased abundance
contributes to bronchial bacterial invasion (Figure 1).
These findings stimulate interest in evaluating glyco-
sides of the type 2b where the core carbohydrate is
retained, but the ceramide attachment is replaced by
simple glycosidic linkages. The synthetic carbohydrate
ligands in suitably bioavailable form could serve as
“decoys” to prevent or clear up bacterial infection. The
understanding of how glycolipid “ligands” interact with
protein receptors in infectious bacteria could well benefit
from an insight as to structure-activity relationships
(SAR) of the carbohydrate domain. Efforts toward that
goal have already led to a stereoselective synthetic route
to asialo GM1 and other simpler glycosides.3 For a
thorough investigation of whether these asialoglycosides
can function as potential antagonists of bacterial invasion
in keeping with the findings of Tuomanen,4 Al Awqati,1
and Krivan,5 one needed access to sialylated glycosides
mimicking GM1 for the case at hand.
F igu r e 1.
in our laboratories hoping to apply the lessons learned
in the asialo GM1 synthesis.3 GM1 itself had also been
synthesized previously by Ogawa,7 Hasegawa,8 and
Schmidt.9
In our initial retrosynthetic analysis, drawing upon our
synthesis of asialo GM1, we decided to target the tet-
rasaccharide moiety of asialo GM1 with a suitable pro-
tecting group at C3 of the internal galactose unit. This
acceptor site would then be unmasked prior to coupling
with the sialic acid moiety to provide the penultimate
pentasaccharide glycal, which upon epoxidation and
methanolysis would deliver the protected version of GM1-
methyl glycoside. Thus, we hoped to couple the ABCD
glycal with a sialic acid unit (E) (Scheme 1).
The target being a suitable analogue of GM1 where the
ceramide moiety would be substituted with a simpler
alkyl group, we set out to synthesize the methyl glycoside
applying the glycal methodology6 that had been developed
† Columbia University.
‡ The Sloan-Kettering Institute for Cancer Research.
(1) Imundo, L.; Barasch, J .; Prince, A.; Al-Awqati, Q. Proc. Natl.
Acad. Sci. U.S.A. 1995, 92, 3019-3023.
The ABCD tetrasaccharide glycal would arise from the
coupling of the disaccharide sectors AB and CD. These
disaccharides would ultimately evolve from a combina-
tion of monocyclic glycals. In the case at hand, the
(2) Barasch, J .; Kiss, B.; Prince, A.; Saiman, L.; Gruenert, D.; Al-
Awqati, Q. Nature 1991, 352, 70-73. Barasch, J .; Al-Awqati, Q. J . Cell.
Sci. (Suppl.) 1993, 17, 229-233.
(3) Kwon, O.; Danishefsky, S. J . J . Am. Chem. Soc. 1998, 120, 1588.
(4) Cundell, D. R.; Tuomanen, E. Microb. Pathog. 1994, 17, 361-
374.
(7) Sugimoto, M.; Numata, M.; Koike, K.; Nakahara, Y.; Ogawa, T.
Carbohydr. Res. 1986, 156, C1.
(5) Krivan, H. C.; Roberts, D. D.; Ginsberg, V. Proc. Natl. Acad. Sci.
U.S.A. 1988, 85, 6157-6161.
(6) Danishefsky, S. J .; Bilodeau M. T. Angew. Chem., Intl. Ed. Engl.
1996, 35, 1380.
(8) Hasegawa, A.; Nagahama, T.; Kiso, M. Carbohydr. Res. 1992,
235, C13. Hasegawa, A.; Ishida, H.; Nagahama, T.; Kiso, M. J .
Carbohydr. Chem. 1993, 12, 703.
(9) Stauch, T.; Greilich, U.; Schmidt, R. R. Liebigs Ann. 1995, 2101.
10.1021/jo9912496 CCC: $19.00 © 2000 American Chemical Society
Published on Web 12/12/1999

Synthesis of the Methyl Glycoside of Ganglioside GM1
J . Org. Chem., Vol. 65, No. 1, 2000 145
Sch em e 1
assembly would be comprised of three D-galactal units
and one D-glucal unit (Scheme 2).
a consequence of intramolecular hydrogen bonding be-
tween the free C4 hydroxyl in the AB sector and the ring
oxygen, thus leading to attenuation of the anomeric
control in the coupling event.3
Sch em e 2
Sch em e 3
The C4 PMB group was deprotected by action of DDQ
(Scheme 4). When the resultant tetrasaccharide acceptor
was submitted to glycosylation using sialic acid, choride,
or phosphite based donors, the results were dismayingly
unproductive. The tetrasaccharide was returned unre-
acted. The N-acetylated acceptor (6, R′ ) Ac) also failed
to react under such sialylation conditions. A similar
failure had marked the Schmidt synthesis of GM1. In that
case, an azide had to be substituted in place of the
acetamido group in the glycosyl acceptors.9
The synthesis of the azaglycosyl donor 3 corresponding
to the AB sector disaccharide piece had been described
before.3 For the CD sector disaccharide, we chose to use
the tetrabenzyl lactal compound 4, with a PMB group
masking the C3′ hydroxyl and the axial C4′ hydroxyl
group available to function as a glycosyl acceptor site.
This intermediate was prepared exactly as the similarly
described C3′ benzyl protected lactal, in our asialo GM1
report. Coupling of 3 and 4 under mediation by methyl
triflate under the conditions shown provided the desired
â-glycoside 5 (Scheme 3), as well as a 6% yield of the
R-anomer. The preponderance of the â-glycoside was now
anticipated, on the basis of precedent.3 It is apparently
Sch em e 4

146 J . Org. Chem., Vol. 65, No. 1, 2000
Bhattacharya and Danishefsky
Accordingly, we revised our retrosynthetic analysis
with a view to introducing the sialyl unit at an earlier
stage. With this in mind, we decided to target the
trisaccharide 8 (which is an intermediate in the synthesis
of GM3)10 and carry out the key sulfonamidoglycosidation
with this trisaccharide as an acceptor. Of course, the
reactivity of the axial C4 hydroxyl, with the neighboring
sialyl group in place, in the context of the sulfonami-
doglycosylation was an issue that had hitherto not been
explored and could be cause for serious concern.
However, we were heartened by the work of Ogawa
and Hasegawa, who had carried out similar [3 + 2]
couplings.7,8 Thus, in our revised strategy we planned to
intersect with diol 9 (Scheme 5), which was expected to
be coupled with a suitable sialyl donor.
reactive and was obtained unreacted leading to complica-
tions in the isolation and purification of the resultant
product of such couplings.12 Wong et al. have put forward
the advantages of a benzyl phosphite donor utilizing a
dibenzyl phosphoramidite as starting material.13 We were
disappointed to find that, even with this method of
preparation of the phosphite donor, mixtures of R- and
â-anomers (albeit preponderant in the desired latter)
resulted. Eventually, we discovered that use of the
diisopropyl phosphoramidite as the source of the phos-
phite group provided, reproducibly, an ∼20:1 (â/R) mix-
ture of the anomers in excellent yields (Scheme 7).
Sch em e 7
Sch em e 5
Armed with a clean method to produce the sialyl
phosphite, we carried out the sialylation of the disaccha-
ride acceptor 9. Although other byproducts also resulted,
we obtained the desired trisaccharide 8 in ∼40-50%
yields (Scheme 8).
Sch em e 8
The advantages of a sialyl phosphite donor (10, Scheme
6) have been argued by Schmidt et al.11 This arrangement
works well in sialyl couplings compared to the classical
chloride donors. However, in our hands, preparation of
the desired â-sialyl phosphite was always associated with
varying amounts of the R-phosphite anomer. This result,
by itself, should not pose a problem, as this anomer
should also lead to the common intermediate 11. Trap-
ping of this system by acetonitrile should give a nitrile-
nitrilium conjugate (as postulated by Schmidt11), which
would ultimately be attacked by a free hydroxyl of the
acceptor giving rise to an R-linked sialyl-coupled product.
We now faced the crucial azaglycosidation reaction in
the context of a sialyl-containing acceptor. In the event,
much to our satisfaction, the reaction of the thioethyl
donor 3 and the acceptor 8, mediated by methyl triflate,
provided the desired pentasaccharide glycal 14 (Scheme
9). The yields of 14 tended to be in the range of ∼40-
50%. The process was further complicated by subsequent
reaction of the acetamido group of this pentasaccharide
leading to imino ether formation at that center. The
imino ether 15 was rehabilitated by hydrolysis under
nonaqueous conditions to the parent pentasaccharide,
albeit in modest yields. The intramolecular hydrogen
bonding effect of the C-4 hydroxyl in the thioethyl donor
was manifested in this coupling as well,3 leading to
preponderant amounts of the â-isomer and, on occasions,
negligible amounts of the R-isomer.
Sch em e 6
With the pentasaccharide glycal corresponding to GM1
in hand, we focused on using the glycal to fashion the
(10) For synthesis of GM3 see: (a) Sugimoto, M.; Ogawa, T.
Glycoconjugate J . 1985, 2, 5. (b) Murase, T.; Ishida, H.; Kiso, M.;
Hasegawa, A. Carbohydr. Res. 1989, 188, 71. (c) Ito, Y.; Paulson, J . C.
J . Am. Chem. Soc. 1993, 115, 1603.
(11) (a) Martin, T. J .; Schmidt, R. R. Tetrahedron Lett. 1992, 33,
6123. (b) Martin, T. J .; Brescelo, R.; Toepfer, A.; Schmidt, R. R.
Glycoconjugate J . 1993, 10, 16.
(12) Sames, D. Unpublished results.
(13) Sim, M. M.; Kondo, H.; Wong, C.-H. J . Am. Chem. Soc. 1993,
115, 2260.
Studies carried out with the mixture of phosphites in
our laboratories had shown that the R-anomer was less

Synthesis of the Methyl Glycoside of Ganglioside GM1
J . Org. Chem., Vol. 65, No. 1, 2000 147
Sch em e 9
desired methyl glycoside. To that end, the glycal 14 was
treated with DMDO, and the resultant R-epoxide was
subjected to methanolysis under mediation by anhydrous
zinc chloride to afford a 56% yield of 16 (Scheme 10). It
was critical to use ZnCl2 to accelerate the opening of the
epoxide. Prolonged treatment with methanol alone led
to incomplete conversions and also deacetylations in the
sialyl sector. Although, at this stage, the assignment of
stereochemistry at glycoside 16 was based to a consider-
able extent on well-established analogies rather than on
unambiguous data on the compounds themselves, â-link-
ages between the four saccharide units were corroborated
by the observed coupling constants in the range of 9-11
Hz for all the four anomeric hydrogens (identified by
HMQC analysis of the methyl glycoside 16).
thesis of the asialo methyl glycoside (see structure 2b)
can now be exploited to allow for pinning down of the
role of sialic acid residues in preventing attachment of
bacterial receptors to cell surface carbohdyrate ligands.
Sch em e 11
Sch em e 10
The concluding deprotection sequence commenced with
desilylation. This step was followed by deacetylation and
ester hydrolysis, debenzylation, and peracetylation to
provide a lactone 17 in ∼40-45% yield. Finally, all the
O-acyl bonds were cleaved and after saponification, the
GM1 methyl glycoside (18) was isolated as its sodium salt
(Scheme 11).
Our assignment of the structure of 18 was based on
the NMR analysis of the intermediates en route to the
final structure. The structure of 18 is also supported by
the agreement of its high-resolution mass spectrum with
theory (TOF-MS ES+ calcd for C38H63N2O29Na2 1057.3312,
found 1057.3352).
In summary, the total synthesis of the target system
18 has been accomplished. The key azaglycosidation
coupling step uniting disaccharide 3 with the sialyl-
containing acceptor 8 is promoted by methyl triflate. This
capability, along with our previously accomplished syn-
It is not clear whether the single methyl glycoside in a
structure such as asialo target system 2b will permit
bacterial adherence or whether clustered structures more
akin to the architecture of cell surfaces will be necessary.
In any case, given the ability to synthesize either 2b or
18, the role of the sialic acid in modulating the ligand
receptor recognition can be established in a rigorous way.

148 J . Org. Chem., Vol. 65, No. 1, 2000
Bhattacharya and Danishefsky
Exp er im en ta l P r oced u r es14
CH2Cl2 (1.5 mL) and ether (3 mL). The solution was stirred
for about 10 min at room temperature before being cooled to
0 °C. After ∼15 min, MeOTf (0.11 mL, 0.97 mmol) was added
dropwise and the solution allowed to stir for ∼12 h in a
Cryocool bath temperature of ∼0-5 °C. TLC at this point
showed some starting material left that did not disappear over
time. Therefore, Et3N (0.4 mL) was added to quench the
reaction and the reaction mixture diluted with ether. After
filtration through a Celite bed and evaporation of the filtrate,
the resulting material was chromatographed using EtOAc-
hexane (5 f 20%) to yield the desired major tetrasaccharide
5 (198 mg, 62% based on thioglycoside, 89% based on recovered
alcohol 4). The ratio of 5 (â linkage) and its anomeric isomer
(R-linkage) was found to be 14:1 (isolated).
Mod ified P r oced u r e for P r ep a r a tion of Th ioglycosid e
3. The glycal (250 mg, 0.33 mmol) was azeotropically dried
with anhydrous benzene and placed under vacuum overnight
and thereafter placed under Ar. Freshly dried 4 Å molecular
sieves (300 mg) and benzenesulfonamide (52 mg) were then
added to the glycal in a glovebox. CH2Cl2 (4 mL) was added,
and the solution was stirred at 0 °C for ∼10 min. Iodonium
bis(sym-collidine)perchlorate (IDCP) (201 mg, 0.43 mmol) that
had been weighed in the dark in a glovebox was added quickly
all at once to the reaction mixture and the resultant solution
stirred at 0 °C in the dark (reaction vessel covered with Al
foil). TLC analysis after ∼10 min revealed that most of the
starting material had been consumed. The reaction was stirred
for a further 15 min at 0 °C and then diluted with ether and
filtered through a glass-sintered funnel (pore size M) in the
dark. The filtrate was then washed with aqueous Na2S2O3
solution once followed by water and then vigorously extracted
with aqueous CuSO4 solution (×4), washed with water again,
and dried over anhydrous Na2SO4, and the solvents were
evaporated at a bath temperature of 5-10 °C. The crude
material was taken up in CH2Cl2 and then filtered through a
short pipet column of anhydrous Na2SO4, and the filtrate was
evaporated and then placed under vacuum for 30 min.
An oven-dried vial (Chemglass) was cooled under Ar and
then charged with anhydrous DMF (0.5 mL) and EtSH (122
µL, 1.65 mmol). The solution was cooled to ∼-40 °C, and a
solution of LHMDS in THF (1.0 M, 1 mL) was added followed
by addition of a solution of the crude iodosulfonamide (pre-
pared above) in anhydrous DMF (3 mL) via cannula. The
solution was stirred at -40 °C for ∼1 h and then slowly
warmed to -10 °C in another 1 h. The reaction mixture was
then diluted with EtOAc at this temperature and then quickly
added to ice followed by addition of saturated aqueous NH4Cl
solution. The aqueous portion was then re-extracted with
EtOAc, and the combined organics were washed with brine,
dried over anhydrous Na2SO4, and concentrated in vacuo. The
crude material was chromatographed with EtOAc-hexanes
(5 f 15%) to provide the desired thioglycoside 3 (129 mg, 40%).
For spectroscopic data see ref 3.
1
5: H NMR (500 MHz, CDCl3) δ 7.87 (d, J ) 7.6 Hz, 2H),
7.41-7.18 (m, 27H), 6.89 (d, J ) 8.3 Hz, 2H), 6.51 (d, J ) 6.2
Hz, 1H), 5.64 (d, J ) 4.0 Hz, 1H), 5.19 (s, 1H), 5.07 (d, J ) 9.3
Hz, 1H), 4.95 (m, 1H), 4.88 (dd, J ) 17.9, 10.7 Hz, 1H), 4.76
(d, J ) 10.5 Hz, 1H), 4.64 (m, 2H), 4.58-4.44 (m, 5H), 4.41 (d,
J ) 6.4 Hz, 1H), 4.35 (m, 2H), 4.22 (m, 2H), 4.12 (br unresolved
m, 1H), 4.07 (br unresolved m, 1H), 3.94 (m, 2H), 3.84 (m, 2H),
3.79 (s, 3H), 3.75 (m, 2H), 3.64 (dd, J ) 10.5, 4.1 Hz, 1H), 3.57
(dd, J ) 9.0, 4.8 Hz, 1H), 3.50 (m,1H), 3.44 (m, 2H), 3.38-
3.27 (m, 3H), 3.22 (m 1H), 3.17 (d, J ) 9.7 Hz, 1H), 3.03 (br s,
1H), 1.06, 1.05 and 1.02 (all s and overlapping m, 42H), 0.91
(s, 9H), 0.28 (s, 3H),0.19 (s, 3H). 13C NMR (125 MHz, CDCl3)
δ 183.8, 159.7, 156.1, 144.7, 138.5, 138.2, 130.0, 129.2, 128.3,
128.1, 128.0, 127.7, 127.4, 114.0, 102.8, 101.0, 99.3, 81.5, 80.9,
79.7, 75.7, 75.4, 75.2, 75.0, 74.3, 73.6, 73.2, 72.7, 70.9, 69.7,
69.3, 68.6, 67.8, 65.7, 61.8, 60.9, 56.0, 55.1, 29.7, 25.6, 17.9,
11.9, -4.5, -5.5. (36 unresolved); IR (vmax cm-1) 3529, 3234,
2931, 2891, 2862, 1793, 1509, 1460; [R]25 ) -17.3° (c 1.5,
D
CHCl3); HRMS (FAB) calcd for [C91H129NO22SSi3K]+ 1742.7655,
found 1742.7672.
P r ep a r a tion of Tetr a sa cch a r id e 7. The tetrasaccharide
5 (35 mg, 0.02 mmol) was taken up in CH2Cl2 (1 mL) and H2O
(0.06 mL). To this stirred solution was added DDQ (12 mg,
0.051 mmol). The reaction was monitored by TLC analysis and
after ∼2 h, although a little starting material remained, the
reaction was quenched by addition of saturated aqueous
NaHCO3 solution. The reaction mixture was extracted with
CH2Cl2, and the organics were washed with water and dried
over Na2SO4. The crude material was chromatographed with
EtOAc-hexane (10 f 25%) to provide the alcohol 7 (16 mg,
51%).
7: 1H NMR (500 MHz, C6D6) δ 7.90 (d, J ) 7.7 Hz, 2H),
7.42-7.19 (m, 23H), 6.49 (d, J ) 6.1 Hz, 1H), 5.62 (br
unresolved m, 1H), 5.4 (s, 1H), 5.06 (d, J ) 8.9 Hz, 1H), 4.99
(d, J ) 10.7 Hz, 1H), 4.95 (m, 1H), 4.66 (d, J ) 8.9 Hz, 1H),
4.63-4.32 (m, 10H), 4.25 (dd, J ) 8.7, 5.4 Hz, 1H), 4.17 (br
unresolved m, 1H), 4.08 (br unresolved m, 1H), 4.01 (br
unresolved m, 1H), 3.94 (t, J ) 8.7 Hz, 1H), 3.88 (br unresolved
m, 1H), 3.86-3.79 (m, 3H), 3.76 (t, J ) 9.4 Hz, 1H), 3.66 (dd,
J ) 10.5, 3.9 Hz, 1H), 3.61-3.51 (m, 4H), 3.46 (m, 1H), 3.38
(m, 4H), 3.01 (br s, 1H), 2.84 (br s, 1H), 1.05, 1.04 and 1.01
(all s and overlapping m, 42H) 0.89 (s, 9H), 0.26 (s, 3H),0.19
(s, 3H); 13C NMR (75 MHz, CDCl3) δ 155.9, 144.6, 138.3, 138.2,
138.1, 132.8, 129.2, 128.6, 128.4, 128.1, 128.0, 127.8, 127.7,
127.6, 127.5, 127.4, 102.7, 102.3, 101.2, 99.2, 81.7, 79.6, 75.7,
75.6, 75.4, 75.0, 73.8, 73.6, 73.4, 73.2, 72.8, 70.6, 69.8, 69.6,
68.8, 67.8, 65.8, 61.8, 60.9, 56.3, 29.7, 25.6, 17.9, 17.7, 11.9,
-4.5, -5.3. (36 unresolved); IR (vmax cm-1) 3539, 3234, 2931,
2862, 1793, 1648, 1465; [R]22D ) -19.9° (c 0.99, CHCl3); HRMS
(FAB) calcd for [C83H121NO21SSi3K]+ 1622.7097, found
1622.7045.
P r ep a r a tion of th e Tetr a sa cch a r id e 5. The thioglycoside
3 (183 mg, 0.19 mmol) and the alcohol 4 (142 mg, 0.19 mmol)
were weighed in an oven-dried flask and were then azeotro-
pically dried using anhydrous benzene and thereafter placed
under vacuum overnight. To this mixture were added freshly
activated 4 Å molecular sieves (1.6 g) followed by a mixture of
(14) Gen er a l Exp er im en ta l Meth od s. All commercial materials
were used without purification unless otherwise noted. The following
solvents were distilled under positive pressure of dry argon: tetrahy-
drofuran (THF), diethyl ether (Et2O) from sodium-benzophenone
ketyl, methylene chloride (CH2Cl2) and toluene (C6H5-CH3) from CaH2.
All the reactions were performed under an inert (Ar) atmosphere.
Analytical thin-layer chromatography was performed using E. Merck
silica 60-F254 coated 0.25 mm plates. Compounds were visualized by
dipping the plates in a cerium sulfate-ammonium molybdate solution
followed by heating. Flash column chromatography was performed
using the indicated solvent on E. Merck silica gel 60 (40-63 µm or
10-40 µm).

Synthesis of the Methyl Glycoside of Ganglioside GM1
J . Org. Chem., Vol. 65, No. 1, 2000 149
P r ep a r a tion of th e Sia lyl P h osp h ite 13. The tetraac-
etate-NANA methyl ester 12 (200 mg, 0.41 mmol) was taken
up in THF (6 mL). The 1H-tetrazole (120 mg, 1.71 mmol) was
added, and to this stirred solution was added the phosphori-
midite (0.32 mL, 0.94 mmol) in a dropwise fashion. The
reaction mixture turned cloudy with the addition of the
phosphorimidite. TLC analysis upon completion of addition
showed almost complete consumption of the starting material.
The reaction mixture was stirred for a further 5-10 min before
being diluted with CH2Cl2 and then quenched with ice-cold
0.3 M aqueous HCl. The resultant solution was washed with
saturated aqueous. NaHCO3 solution and ice-water and dried
over anhydrous Na2SO4. After concentration in vacuo, the
material so obtained was chromatographed with EtOAc-
hexanes (80%) to yield the desired â-anomeric phosphite 13
in pure form (201 mg, 67%) and a mixture of anomeric
phosphites (20 mg, 7%) to reflect a combined ratio of ∼20:1 of
the â/R anomers.
P r ep a r a tion of th e P en ta sa cch a r id e 14. The thioglyco-
side 3 (118 mg, 0.12 mmol) and the alcohol 8 (138 mg, 0.12
mmol) were weighed in an oven-dried flask and were then
azeotropically dried using anhydrous benzene and thereafter
placed under vacuum overnight. To this mixture were added
freshly activated 4 Å molecular sieves (300 mg) followed by a
mixture of CH2Cl2 (1 mL) and ether (2 mL). The solution was
stirred for about 10 min at room temperature before being
cooled to 0 °C. After ∼15 min, MeOTf (20 µL, 0.18 mmol) was
added dropwise and the solution allowed to stir for ∼24 h in
a Cryocool bath temperature of ∼5 °C. TLC at this point
showed around 40% product formation. A further equivalent
of MeOTf was then added and the resulting solution allowed
to stir for another 12 h. Although TLC analysis showed
starting material to be still present, since side products
(evinced by other spots) were increasing, the reaction was
terminated by addition of Et3N (0.4 mL) and the reaction
mixture was diluted with ether. After filtration through a
Celite bed and evaporation of the filtrate, the resulting
material was chromatographed using a gradient eluant. With
EtOAc-hexanes (15 f 20%), the imino ether 15 (34 mg, 14%)
eluted first. This was followed by EtOAc-hexanes (40 f 70%)
to give the desired pentasaccharide 14, as a >14:1 mixture
(114 mg, 54% based on thioglycoside) and as a 3:1 mixture of
isomers (14 mg, 6%) and finally EtOAc-hexanes (70 f 100%)
to provide unreacted acceptor 8 (30 mg, 22%). The combined
yield of pentasaccharide products was thus ∼60% based on
thioglycoside and 76% based on recovered acceptor trisaccha-
ride.
For spectroscopic data see ref 13.
P r ep a r a tion of th e Sia lyla ted Tr isa cch a r id e 8. The diol
9 (260 mg) was azeotropically dried with anhydrous benzene
and evacuated overnight. This was put under Ar, freshly dried
4 Å molecular sieves (400 mg) were added (in a glovebox)
followed by anhydrous CH3CN (5 mL), and the resultant
suspension was stirred at room temperature for 30 min. The
phosphite 13 (261 mg, 0.355 mmol) was separately taken up
in anhydrous CH3CN (3 mL) and added via cannula to the
diol solution, and the resultant solution was allowed to stir at
room temperature for a further 30 min. The reaction mixture
was then cooled to ∼-40 °C and stirred for ∼15 min before
addition of TMSOTf (10 µL, 0.056 mmol). After being stirred
for ∼40 min at -40 f -35 °C, the reaction mixture was
quenched by addition of saturated aqueous NaHCO3, warmed
to room temperature. The reaction mixture was then added
to EtOAc and washed with saturated aqueous NaHCO3, and
the organics were dried over anhydrous Na2SO4. The resultant
material was chromatographed with EtOAc-chloroform (70%)
to give a total trisaccharide-product yield of 62% with the
desired trisaccharide 8 as the major isomer (41%). Occasionally
one of the byproductssthe sialic acid-glycal unit formed by
the elimination of the anomeric phosphite, which coelutes with
the mixture of trisaccaharide productsswas separated by a
gel filtration with Sephadex LH-120 using hexanes/CH2Cl2/
MeOH (4:2:1) as eluent followed by a normal chromatography
to separate the different isomers.
1
14: H NMR (500 MHz, C6D6) δ 8.2, (dd, J ) 8.4, 1.2 Hz,
2H), 7.5 (d, J ) 7.1 Hz, 2H), 7.4-6.95 (m, 21H), 6.36 (d, J )
6.3 Hz, 1H), 5.89 (d, J ) 3.6 Hz, 1H), 5.85 (br unresolved m,
1H), 5.59 (m, 1H), 5.41 (dd, J ) 9.4, 1.9 Hz, 1H), 5.1 (m, 2H),
4.9 (dt, J ) 10.8, 6.4 Hz, 1H), 4.9 (m, 3H), 4.79 (d, J ) 9.1 Hz,
1H), 4.61-4.50 (m, 5H), 4.49-3.91 (m, 23H), 4.02 (s, 3H), 3.84
(m, 2H), 3.7 (d, J ) 2.7 Hz, 1H), 3.6 (dd, J ) 8.0, 7.7 Hz, 1H),
3.48 (m, 1H), 3.37 (m, 1H), 2.56 (dd, J ) 14.0, 5.1 Hz, 1H),
2.37 (dd, J ) 13.7, 12.0 Hz, 1H), 2.21 (s, 3H), 1.89 (s, 3H),
1.85 (s, 3H), 1.64 (s, 3H), 1.52 (s, 3H), 1.28 (m, 6H), 1.20-1.10
(all s, 36 H), 0.90 (s, 9H), 0.39 (s, 3H),0.16 (s, 3H); 13C NMR
(75 MHz, CDCl3) δ 170.6, 170.4, 170.2, 169.7, 169.2, 167.9,
160.1, 155.8, 144.5, 138.4, 138.1, 132.7, 129.3, 128.3, 128.2,
128.1, 127.9, 127.7, 127.6, 127.5, 127.3, 102.7, 102.4, 102.1,
101.5, 99.3, 99.3, 81.9, 78.1, 77.6, 75.6, 75.3, 75.2, 74.6, 73.5,
73.4, 73.3, 72.9, 72.6, 72.2, 70.9, 70.1, 69.6, 68.9, 68.7, 67.9,
67.9, 66.6, 65.8, 61.6, 60.9, 56.5, 53.2, 49.3, 35.7, 31.9, 29.6,
25.6, 23.2, 22.6, 20.9, 20.7, 20.7, 20.6, 17.9, 17.7, 14.1, 11.9,
11.8, -4.7, -5.3. (34 unresolved); IR (vmax cm-1) 3546, 3306,
3064, 3031, 2940, 2866, 1793, 1748, 1670, 1541, 1497; [R]20
D
1
8: H NMR (500 MHz, C6D6) δ 7.25 (m, 20H), 6.42 (d, J )
) -26.3° (c 1.5, CHCl3); HRMS (FAB) calcd for [C103H148N2O33
-
SSi3K]+ 2095.8630, found 2095.8728.
6.3 Hz, 1H), 5.38 (m, 1H), 5.28 (dd, J ) 8, 1.5 Hz, 1H), 5.19 (d,
J ) 9.4 Hz, 1H), 4.86 (m, 2H), 4.69 (q, J ) 11.6 Hz, 2H), 4.64
(d, J ) 7.7 Hz, 1H), 4.58 (s, 2H), 4.45 (m, 4H), 4.28 (m, 2H),
4.11 (m, 3H), 4.0 (m, 2H), 3.95 (dd, J ) 12.4, 5.9 Hz, 1H), 3.8-
3.69 (m, 3H), 3.75 (s, 3H), 3.61 (m, 3H), 3.5 (dd, J ) 9.2, 8.0
Hz, 1H), 2.71 (br s, 1H), 2.50 (dd, J ) 13, 4.6 Hz, 1H), 2.08 (s,
3H), 1.99 (m, 1H), 1.97 (s, 3H), 1.91 (s, 3H), 1.85 (s, 3H); 13C
NMR (75 MHz, CDCl3) δ 170.8, 170.6, 170.2, 170.0, 168.5,
160.1, 144.4, 138.9, 138.7, 138.1, 128.3, 128.1, 127.9, 127.6,
127.3, 109.6, 102.6, 99.4, 98.2, 75.8, 75.7, 74.8, 73.5, 73.2, 72.7,
72.6, 71.9, 70.2, 69.0, 68.7, 68.1, 68.0, 67.2, 62.3, 53.4, 53.0,
49.3, 36.7, 29.7, 23.2,21.1, 20.8, 20.7, 20.6 (16 unresolved); IR
(vmax cm-1) 3357, 3031, 2923, 1746, 1663, 1541, 1454; [R]20
)
Da ta for th e im in o eth er 15: 1H NMR (500 MHz, CDCl3)
δ 7.87 (d, J ) 7.5 Hz, 2H), 7.41-7.17 (m, 23H), 6.52 (d, J )
6.3 Hz, 1H), (5.51 (d, J ) 4.2 Hz, 1H), 5.37 (s, 1H), 5.19 (m,
D
-15.2° (c 1.03, CHCl3); HRMS (FAB) calcd for [C60H71NO21K]+
1180.4156, found 1180.4194.

150 J . Org. Chem., Vol. 65, No. 1, 2000
Bhattacharya and Danishefsky
2H), 5.09 (d, J ) 9.1 Hz, 1H), 4.99 (t, J ) 5.1 Hz, 1H), 4.82
(m, 1H), 4.79 (d, J ) 10.1 Hz, 1H), 4.65 (m, 2H), 4.56-4.42
(m, 8H), 4.37 (q, J ) 11.6 Hz, 2H), 4.24 (m, 2H), 4.14 (br
unresolved m, 1H), 4.06 (br unresolved m, 2H), 4.03 (d, J )
3.9 Hz, 1H), 4.01 (br unresolved m, 1H), 3.97 (s, 3H), 3.95 (m,
2H), 3.89 (dd, J ) 9.7, 3.7 Hz, 1H), 3.87-3.61 (m, 11H), 3.58
(br s, 3H), 3.46 (m, 2H), 3.37 (t, J ) 5.5 Hz, 1H), 3.26 (dd, J )
9.6, 7.9 Hz, 1H), 2.96 (m, 2H), 2.49 (dd, J ) 13.5, 4.6 Hz, 1H),
2.1 (s, 3H), 1.96 (s, 3H), 1.95 (m partly hidden, 1H), 1.81 (s,
3H), 1.80 (s, 3H), 1.05, 1.04, 0.99 and 0.98 (all s and overlap-
ping m, 42 H), 0.89 (s, 9H), 0.23 (s, 3H),0.18 (s, 3H); 13C NMR
(125 MHz, CDCl3) δ 170.4, 169.3, 169.1, 168.9, 167.9, 164.8,
160.3, 155.8, 144.5, 138.6, 138.3, 138.1, 132.7, 129.2, 128.8,
128.3, 127.9, 127.7, 127.6, 127.4, 102.6, 102.4, 101.5, 99.7, 99.0,
81.8, 78.2, 75.6, 75.4, 74.2, 73.6, 73.5, 73.3, 72.8, 72.6, 70.3,
70.1, 69.6, 68.7, 68.4, 68.0, 67.9, 67.6, 65.8, 61.6, 60.9, 57.0,
56.6, 53.7, 52.3, 29.7, 25.6, 20.9, 20.6, 20.4, 17.9, 17.7, 15.4,
11.9, 11.8, -4.7, -5.3. (44 unresolved); IR (vmax cm-1) 3540,
3301, 3064, 3031, 2942, 2866, 1794, 1751, 1679, 1650, 1456;
HRMS (FAB) calcd for [C104H150N2O33SSi3K]+ 2109.8787, found
2109.8745.
Hz, 2H), 4.8 (overlapping d, J ) 10.3 Hz, 1H), 4.6 (m, 3H),
4.48 (m, 2H), 4.39 (m, 2H), 4.2 (m, 5H), 4.5-3.83 (m, 6H), 4.92
(overlapping s, 3H), 3.78 (m, 8H), 3.65 (dd, J ) 8.7, 5.1 Hz,
1H), 3.60 (m, 1H), 3.53 (s, 3H), 3.46 (m, 4H), 3.33 (m, 1H),
3.25 (m, 1H), 2.83 (br s, 1H), 2.61 (s, 1H), 2.51 (br s, 1H), 2.30
(dd, J ) 13.4, 4.8 Hz, 1H), 2.19 (d, J ) 12.5 Hz, 1H), 2.10 (s,
3H), 1.97 (s, 3H), 1.95 (s, 3H), 1.88 (s, 3H), 1.87 (s, 3H), 1.01
(m, 6H), 1.04, 1.03, 0.99, 0.98 (all s, 36 H), 0.85 (s, 9H), 0.16
(s, 3H), 0.13 (s, 3H); 13C NMR [125 MHz, CDCl3] δ 170.6, 170.4,
169.7, 169.3, 167.7, 160.1, 155.6, 138.4, 137.9, 132.5, 129.1,
128.4, 128.3, 128.1, 128.0, 127.7, 127.6, 127.5, 127.2, 103.4,
102.9, 102.4, 102.1, 99.7, 83.0, 82.7, 79.4, 78.4, 76.0, 75.8, 75.5,
75.2, 74.4, 73.7, 73.4, 73.2, 72.6, 72.3, 69.3, 69.1, 68.8, 68.1,
67.9, 66.5, 65.7, 61.7, 61.5, 61.1, 56.9, 56.4, 53.8, 53.3, 49.2,
35.5, 29.7, 29.3, 25.5, 23.2, 20.9, 20.8, 20.7, 20.6, 17.9, 17.6,
11.9, 11.8, -4.8, -5.3. (36 unresolved); IR (vmax cm-1) 3546,
3294, 2941, 2866, 2959, 1794, 1750, 1686, 1455; [R]20D ) -16.4°
(c 2.7, CHCl3); HRMS (FAB) calcd for [C104H152N2O35SSi3]+
2127.9100, found 2127.9060.
Con ver sion of th e Im in o Eth er 15 to th e P en ta sa c-
ch a r id e 14. The imino ether 15 (∼20 mg, 0.001 mmol) was
taken up in THF (0.5 mL). Tetrabutylammonium iodide (TBAI)
(∼5 mg) was added followed by Amberlite IR-120 (prewashed
with THF, ∼14 mg). The reaction was monitored by TLC
analysis, more TBAI and Amberlite IR-120 were added, and
the reaction mixture was allowed to stir overnight. Although
starting material remained, on account of an increase of
byproducts formation, the reaction was worked up by dilution
with EtOAc, the Amberlite resin was filtered off, and the
organics were washed first with saturated aqueous NaHCO3
and then with aqueous Na2S2O3 until the yellow color was
P r ep a r a tion of La cton e 17. Tetrabutylammonium fluo-
ride (TBAF) (1 M in THF, 0.51 mL, 20 equiv) was added to
the methyl glycoside 16 (54 mg, 0.026 mmol) in THF (0.5 mL).
The mixture was stirred for 48 h at room temperature and
then concentrated. To the residue was added NaOMe (2%
solution in MeOH, 1 mL) and the solution stirred at room
temperature for another 48 h. To the reaction mixture were
added 1 mL of water and 1 mL of THF and the mixture stirred
for an additional 24 h at room temperature. The reaction
mixture was cooled to 0 °C, and the pH was adjusted to ∼8-9
using Dowex 50X8-400 ion-exchange resin. The resin was
removed by filtration, and the filtrate was concentrated to
dryness. The residue was then dissolved in THF (2 mL).
Meanwhile a two-necked flask fitted with a dry ice condenser
was cooled to -78 °C, and about 15 mL of liquid NH3 was
collected. Na (ca. 4 × 4 × 4 mm) was added to it, and to the
blue solution was added the solution in THF. The reaction was
stirred under reflux of NH3 for ∼30 min. The reaction was
quenched with MeOH (5 mL), and the ammonia was evapo-
rated. The residual mixture was treated with Dowex
50X8-400 ion-exchange resin to pH 10 and then concentrated
in vacuo. The residue was taken up in pyridine (anhydrous, 2
mL), Ac2O (2 mL) was added, and the resultant mixture was
stirred for 48 h at room temperature. The mixture was
concentrated in vacuo, and the residue was chromatographed
first with MeOH-CH2Cl2 (4 f 5%) followed by EtOAc-CH2-
discharged. After
a further brine wash and drying over
anydrous Na2SO4, the solvents were evaporated. The crude
material was chromatographed with EtOAc-hexanes (25 f
60%) to provide the pentasaccharide 14 (8 mg, 40%).
P r ep a r a tion of th e Meth yl Glycosid e 16. The pentasac-
charide 14 (61 mg, 0.03 mmol) was azeotropically dried with
anhydrous benzene and put under vacuum overnight and then
put under Ar. Freshly dried 4 Å molecular sieves (190 mg)
were added (in a glovebox) followed by CH2Cl2 (2 mL), and
the resulting suspension was allowed to stir at 0 °C for ∼20
min. A solution of DMDO/acetone (titrated immediately before
addition, 0.75 mL, 0.045 mmol) was added and the solution
stirred for 30 min at 0 °C. The solvent was then removed by
a positive pressure of Ar followed by evacuation for ∼30 min.
Anhydrous MeOH (2 mL) was then added to the material at 0
°C and the stirred solution cooled to -78 °C. THF (2 mL) was
added at this point followed by a solution of ZnCl2 (45 µL, 0.045
mmol). After 30 min at -78 °C, the reaction mixture was
allowed to warm to room temperature over the course of 5 h
and then stirred for an additional 30 min at room temperature
before diluting with EtOAc and quenching with pH 7.0 buffer.
The reaction mixture was extracted with EtOAc followed by
salting out with NaCl (s), washed with brine, and dried over
anhydrous Na2SO4. After evaporation of solvents, the crude
material was chromatographed through a short column of SiO2
using hexane/EtOAc/acetone (30:20:8 f 30:20:10 f 30:20:15)
to provide the desired methyl glycoside 16 (54 mg, 86%).
Cl2-MeOH (80:20:2 f 80:20:3 f 80:17:3 f 80:16:4
f
80:15:5) to yield the peracetate-lactone 17 as a pure compound
(15 mg, 37%) and as a mixture of isomers (5 mg, 12%).
17: 1H NMR (500 MHz, CDCl3) δ 6.45 (br unresolved m, 1H),
5.48 (m, 2H), 5.38 (d, J ) 3.2 Hz, 1H), 5.29 (d, J ) 2.8 Hz,
1H), 5.22 (m, 3H), 5.05 (d, J ) 7.9 Hz, 1H), 5.03 (dd, J ) 10.3,
7.9 Hz, 1H), 4.92 (dd, J ) 10.4, 3.3 Hz, 1H), 4.88 (dd, J ) 9.4,
8.1, 1H), 4.74 (d, J ) 7.8 Hz, 1H), 4.56 (dd, J ) 10.5, 7.5 Hz,
1H), 4.51 (d, J ) 10.7 Hz, 2H), 4.39 (m, 5H), 4.22-4.0 (m, 7H),
3.94 (dd, J ) 12.0, 7.5 Hz, 1H), 3.9-3.75 (m, 4H), 3.72 (dd, J
) 10.6, 2.5 Hz, 1H), 3.68 (m, 1H), 3.62 (m, 1H), 3.52 (m, 1H),
3.46 (s, 3H), 2.46 (dd, J ) 13.5, 5.4 Hz, 1H), 2.16 (s, 3H), 2.12
(s, 6H), 2.09 (s, 6H), 2.07 (s, 9H), 2.05 (s, 3H), 2.03 (s, 6H),
2.02 (s, 3H), 2.00 (s, 6H), 1.96 (s, 3H), 1.87 (s, 3H), 1.82 (dd, J
) 13.1, 11.8 Hz, 1H); 13C NMR (125 MHz, CDCl3) δ 171.9,
171.1, 170.9, 170.4, 170.3, 170.2, 169.6, 169.5, 163.1, 101.4,
100.3, 99.9, 98.9, 97.9, 75.2, 74.3, 73.8, 73.0, 72.9, 72.7, 72.0,
71.8, 71.4, 70.9, 70.3, 69.2, 68.9, 68.8, 68.4, 68.1, 66.6, 63.5,
1
16: H NMR (500 MHz, CDCl3) δ 7.84 (d, J ) 7.6 Hz, 2H),
7.35-7.10 (m, 23H), 5.6 (d, J ) 3.7 Hz, 1H), 5.29 (dd, J ) 9.5,
1.3 Hz, 1H), 5.2 (m, 2H), 5.15 (d, J ) 10 Hz, 1H), 5.06 (d, J )
9 Hz, 1H), 4.92 (dd, J ) 11.4, 4.8 Hz, 1H), 4.83 (q, J ) 11.6

Synthesis of the Methyl Glycoside of Ganglioside GM1
63.0, 62.5, 62.0, 60.6, 57.0, 54.5, 52.2, 49.3, 38.0, 29.7, 25.1,
J . Org. Chem., Vol. 65, No. 1, 2000 151
of GM1 methyl glycoside 18 (8 mg, ∼80%) after removal of
water via lyophilization.
24.0, 23.1, 20.9, 20.8, 20.7, 20.2 (17 unresolved); IR (vmax cm-1
)
3584, 3390, 2937, 1748, 1673, 1537, 1435;HRMS (FAB) calcd
18: 1H NMR (500 MHz, CD3OD-D2O) δ 4.87 (d, J ) 8.6
Hz, 1H), 4.44 (dd, J ) 8.0,7.6 Hz, 2H), 4.23(d, J ) 7.9 Hz,
1H), 4.15 (m, 2H), 4.03 (m, 2H), 3.92-3.80 (m, 6H), 3.80-3.65
(m, 11H), 3.58-3.37 (m, 10H), 3.52 (s, 3H), 2.71 (dd, J ) 12.7,
4.7 Hz, 1H), 2.01 (s, 3H), 1.99 (s, 3H), 1.92 (t, J ) 11.9 Hz,
1H); 13C NMR (125 MHz, CD3OD-D2O) δ 176.0, 175.3, 175.1,
106.2, 104.9, 104.4, 103.9, 103.2, 82.4, 80.7, 78.8, 76.3, 76.2,
76.1, 75.7, 75.4, 74.8, 74.3, 74.2, 73.4, 72.3, 71.2, 70.1, 69.6,
69.4, 64.8, 62.7, 62.3, 61.7, 61.6, 57.7, 53.5, 52.5, 38.3, 23.8,
for [C66H90N2O42K]+ 1621.4605, found 1621.4552.
22.8 (2 unresolved); [R]25 ) +19.9° (c 0.25, MeOH-H2O);
D
HRMS (TOF, ES+) calcd for [C38H63N2O29Na2]+ 1057.3312,
found 1057.3552.
Syn th esis of GM1 Meth yl Glycosid e (18). To the perac-
etate 17 (15 mg, 0.01 mmol) were added MeOH (1 mL) and
solid NaOMe (∼ 5 mg), and the solution was stirred overnight
at room temperature. To this was added 1 mL of water, and
the mixture was stirred for an additional 24 h at room
temperature. The reaction mixture was cooled to 0 °C, and
the pH was adjusted to ∼8-9 by treatment with Dowex 50X8-
400 ion-exchange resin. The resin was removed by filtration
and the filtrate concentrated and dried by lyophilization. The
residue was subjected to Bio-Gel P2 (fine, 45-90 µm) column
chromatography eluting with water to afford the sodium salt
Ack n ow led gm en t. This work was supported by the
National Institutes of Health (Grant No. HL-25848). We
thank Dr. Ohyun Kwon for many helpful discussions
and Mr. Shingo Nakamura for some preparative work.
Su p p or tin g In for m a tion Ava ila ble: Experimental pro-
cedure for compound 4 and spectral data for all new com-
pounds are included. This information is available free of
charge via the Internet at http://pubs.acs.org.
J O9912496