Synthesis of the 6-O-Methyl-D-glycero-α-Lgluco-heptopyranose Moiety present in the capsular polysaccharide from campylobacter jejuni NCTC 11168

Article abstract of DOI:10.1021/ol202152r

The first synthesis of the 6-O-methyl-D-glycero-α-L-gluco- heptopyranose moiety present in the capsular polysaccharide from Campylobacter jejuni NCTC 11168 is reported. The target (1) was synthesized as the 8-aminooctyl glycoside and then conjugated to bo

Full text of DOI:10.1021/ol202152r

ORGANIC
LETTERS
2011
Vol. 13, No. 19
5290–5293
Synthesis of the 6-O-Methyl-D-glycero-R-L-
gluco-heptopyranose Moiety Present in
the Capsular Polysaccharide from
Campylobacter jejuni NCTC 11168
Wenjie Peng, Anushka B. Jayasuriya, Akihiro Imamura, and Todd L. Lowary*
Alberta Innovates Centre for Carbohydrate Science and Department of Chemistry, The
University of Alberta, Gunning-Lemieux Chemistry Centre, Edmonton, AB T6G 2G2
Canada
tlowary@ualberta.ca
Received August 8, 2011
ABSTRACT
The first synthesis of the 6-O-methyl-D-glycero-R-L-gluco-heptopyranose moiety present in the capsular polysaccharide from Campylobacter
jejuni NCTC 11168 is reported. The target (1) was synthesized as the 8-aminooctyl glycoside and then conjugated to bovine serum albumin (BSA)
for the generation of antibodies recognizing this motif. Heptose 1 was obtained from D-galactose via a series of galactofuranose derivatives.
Campylobacter jejuni is an important food-borne patho-
gen that can cause severe enteritis and is a leading cause of
diarrheal disease.1 Infections by C. jejuni are also linked to
glucuronic acid residue amidated with 2-aminoglycerol, a
methyl phosphoramidate moiety, and a 6-O-methyl-heptose
residue possessing the D-glycero-L-gluco stereochemistry.
ꢀ
GuillainÀBarre syndrome, a neurological disorder that
results in paralysis of varying severity, and which is occa-
sionally fatal.2 The outer surface of all campylobacters is
functionalized with a species-specific capsular polysacchar-
ide (CPS), the production of which is required for virulence.3
The biochemical pathways that assemble the CPS are there-
fore potential sites for therapeutic intervention.
As part of a larger program directed toward understand-
ing polysaccharide biosynthesis in different campylobacter
species,4 we have focused our attention on C. jejuni
NCTC11168 (HS:2), which possesses a CPS with a tetra-
saccharide repeating unit (Figure 1).5 This structure con-
tains a number of unusual motifs including an N-acetyl-
galactosamine residue in the furanose ring form, a
Figure 1. Structure of the tetrasaccharide repeating unit from C.
jejuni NCTC 111168.
(1) Sherman, P. M.; Ossa, J. C.; Wine, E. Curr. Opin. Gastroenterol.
2010, 26, 1–4.
(2) Pithadia, A. B.; Kakadia, N. Pharmacol. Rep. 2010, 62, 220–232.
(3) Guerry, P.; Szymanski, C. M. Trends Microbiol. 2008, 16, 428–
435.
(4) Poulin, M. B.; Nothaft, H.; Hug, I.; Feldman, M. F.; Szymanski,
C. M.; Lowary, T. L. J. Biol. Chem. 2010, 285, 493–501.
(5) Szymanski, C. M.; St. Michael, F.; Jarrell, H. C.; Li, J.; Gilbert,
M.; Larocque, S.; Vinogradov, E.; Brisson, J.-R. J. Biol. Chem. 2003,
278, 24509–24520.
Heptose sugars are present in a number of bacterial
species, particularly in the core structures of lipopoly-
saccharides from Gram-negative bacteria.6 However, the
(6) Kosma, P. Curr. Org. Chem. 2008, 12, 1021–1039.
r
10.1021/ol202152r
Published on Web 08/31/2011
2011 American Chemical Society

CPS from C. jejuni NCTC11168 is the first naturally
occurring glycoconjugate reported to contain a heptose
with the D-glycero-L-gluco stereochemistry.7 It is of interest
to synthesize glycoconjugates incorporating this heptose
for the generation of antibodies specific to this novel mo-
tif, as well as to probe the various biosynthetic enzymes
(e.g., glycosyltransferases) involved in assembling the
full-length glycan. We describe here the first synthesis
of this monosaccharide, functionalized with an aminooctyl
spacer arm (1, Figure 2), and its attachment to a protein
carrier (BSA) for the generation of antibodies.
dimethoxypropane and p-toluenesulfonic acid. This
reaction afforded the expected product 3 in 88% yield.
Benzylation of the two hydroxyl groups using benzyl
bromide and sodium hydride (giving 4) followed by
hydrolysis of the acetal provided 5 in 93% yield over the
two steps. The resulting diol was then protected at
the primary position as a tert-butyldiphenylsilyl ether
affording, in 95% yield, 6.
With a route to secondary alcohol 6 in place, treatment
withmethyl iodide and sodium hydride in DMF resulted in
introduction of the methyl group, and then the silyl ether
was replaced with a benzyl ether via a process involving
desilylation and alkylation. The product 9 was obtained in
92% yield over the three steps from 6. Attempts to prepare
9 in two steps from diol 5 by formation of a stannylene
acetal, selective benzylation of O-6,12 and methylation led
to a lower overall yield of the product than the approach
outlined in Scheme 1. Hydrolysis of the thioglycoside was
achieved by reaction with NBS and water in acetone,
affording hemiacetal 10 in 95% yield.
Figure 2. Retrosynthetic analysis of 1.
The key step in the route was the chain extension of 10,
which was successfully achieved upon reaction with divinyl
zinc, generated in situ from vinyl magnesium bromide with
ZnBr2.13 Divinylzinc addition to 10 led to the formation of
11 in 91% yield with excellent stereoselectivity due to
chelation control. None of the other isomer could be
detected. To explore the importance of the ZnBr2 to
the stereocontrol of the reaction, we also treated 10
with vinyl magnesium bromide alone or combined with
Because heptoses are not widely available from natural
sources, their chemical synthesis has been the topic of a
number of investigations.6,8,9 The majority of heptose
syntheses involve chain extension of an oxidized hexose
derivative, possessing an aldehyde group at C-6, via
Grignard or Wittig reactions. Subsequent functionali-
zation of this homologated product provides the target
molecule. In the case of D-glycero-L-gluco-heptose,
using this approach would require access to significant
quantities of L-glucose, which is not readily available.
Therefore, we investigated a less widely used strategy,
involving chain extension from the C-1 aldehyde func-
tionality of a hexose, in particular D-galactose.
Although the synthesis of heptoses by this route has
been reported as far back as investigations by Fischer
and Kiliani on the synthesis of cyanohydrins,10 the
stereoselectivity of these processes is generally not high
and it is has not been widely used in modern organic
chemistry.6 Mindful of this, we envisioned that the
target, 1, could be synthesized from galactofuranosyl
thioglycoside 2 (Figure 2), which can be obtained on
large scale from D-galactose in two steps,11 provided
that chain extension could be performed in a stereo-
selective manner.
MgBr2 Et2O.14 Both reactions provided the product with
3
poor selectivity and in reduced yield. The structure of 11
1
was proven by H NMR spectroscopy of a subsequently
synthesized cyclic intermediate and the preparation of a
crystalline derivative (see below).
Ozonolysis of the alkene in 11 afforded the heptose
derivative 12, which was O-acetylated and then treated
with p-toluenethiol and BF3 OEt2, yielding thioglycoside
3
14 in 81% yield over three steps from 11. The β-isomer of
14 could be separated by chromatography, and the stereo-
1
chemistry was confirmed by H NMR spectroscopy in
CDCl3; the 3J1,2 was 10.1 Hz, indicative of a trans relation-
ship between H-1 and H-2.
Because the final target, 1, possesses an axially oriented
aglycone, it was necessary to convert the acetate ester on O-2
into a nonparticipating benzyl group. This was achieved in
two steps by deacetylation (yielding 15) and benzylation
resulting in a 91% yield of 16. This thioglycoside is expected
to be a suitable donor for the preparation of glycoconjugates
containing this unusual monosaccharide. We note that 16
was obtained from 2 in a comparable number of steps to
The synthesis of 1 is illustrated in Scheme 1. Thioglyco-
side 2, obtained from D-galactose via a two-step route
involving formation of the corresponding diethyl dithioa-
cetal and subsequent cyclization,11 was first treated with
(7) St Michael, F.; Szymanski, C. M.; Li, J.; Chan, K. H.; Khieu,
N. H.; Larocque, S.; Wakarchuk, W. W.; Brisson, J.-R.; Monteiro,
M. A. Eur. J. Biochem. 2002, 269, 5119–5136.
(8) (a) Hansson, J.; Oscarson, S. Curr. Org. Chem. 2000, 4, 535–564.
(b) Oscarson, S. Top. Curr. Chem. 1997, 186, 171–202.
(9) Ohara, T.; Adibekian, A.; Esposito, D.; Stallforth, P.; Seeberger,
(12) (a) Sugawara, F.; Nakayama, H.; Ogawa, T. Agric. Biol. Chem.
1986, 50, 1557–1561. (b) Gurjar, M. K.; Reddy, L. K.; Hotha, S. J. Org.
Chem. 2001, 66, 4657–4660. (c) Bai, Y.; Lowary, T. L. J. Org. Chem.
2006, 71, 9672–9680.
(13) (a) Boschetti, A.; Nicotra, F.; Panza, L.; Russo, G. J. Org. Chem.
1988, 53, 4181–4185. (b) Nolen, E. G.; Kurish, A. J.; Potter, J. M.;
Donahue, L. A.; Orlando, M. D. Org. Lett. 2005, 7, 3383–3386.
(14) (a) Behr, J. B.; Guillerm, G. Tetrahedron Lett. 2007, 48, 2369–
2372. (b) Seepersaud, M.; Al-Abed, Y. Tetrahedron Lett. 2000, 41, 7801–
7803.
P. H. Chem. Commun. 2010, 46, 4106–4108.
(10) Morrison, J. D.; Mosher, H. S. Asymmetric Organic Reactions;
Prentice-Hall: Washington D. C., 1991; pp 133À141.
(11) Green, J. W.; Pacsu, E. J. Am. Chem. Soc. 1937, 59, 1205–1210.
Org. Lett., Vol. 13, No. 19, 2011
5291

Scheme 1. Synthesis of 1 and its bovine serum albumin conjugate (1-BSA)
that recently reported for the de novo synthesis of heptose
glycosyl donors.9
Further support for the stereochemistry of the heptose
ring substituents was obtained by the preparation of a
crystalline derivative of 16. This was achieved by first
treating 16β with hydrogen over a palladium on carbon
catalyst in ethanol. Unexpectedly, this reaction afforded a
product in which the four benzyl ethers had been cleaved
and the thioglycoside moiety had been replaced by an
ethoxy group, undoubtedly from the ethanol used as the
5292
Org. Lett., Vol. 13, No. 19, 2011

reaction solvent. Subsequent treatment of this product
with p-nitrobenzoyl chloride in pyridine gave the product
17 in 37% yield over the two steps. The product was
recrystallized from ethyl acetateÀhexane, and X-ray ana-
lysis of the resulting crystals confirmed the D-glycero-L-
gluco stereochemistry (CCDC deposition number 838206;
see Supporting Information).
Having unequivocally established the stereochemistry of
16, its reaction with 8-azidooctanol in the presence of NIS
and silver triflate15 was attempted. Unexpectedly, this
reaction led to amixtureofproductsinwhichthe undesired
β-glycoside (1,2-trans-glycoside) predominated. This pro-
blem could be circumvented (Scheme 1B) by conversion of
16 into the corresponding reducing sugar 18 and, in turn,
the trichloroacetimidate 19. Glycosylation of 19 with
8-azidooctanol promoted by TMSOTf16 led to the forma-
tion of 20 (95% yield), as a 4:1 R/β mixture of glycosides,
which could be separated by chromatography.
The conversion of 20 into 1 in a single step by treatment
with hydrogen over a palladium on carbon catalyst proved
problematic. Very low yields of impure product were
obtained in a range of solvents including ethyl acetate,
ethanol, acetic acid, and 1:1 ethanolÀacetic acid. There-
fore, we explored a stepwise approach, which we had used
when a similar problem arose in the course of another
investigation.17 Thus, the azido group in 20was reduced by
hydrogenation over a palladium catalyst poisoned by
pyridine. The resulting amine was converted to the corre-
sponding N-trifluoroacetamide derivative 21 in 95% over-
all yield. Hydrogenolysis of the benzyl ethers in 21
proceeded without difficulty, and the resulting product
22 was obtained in 91% yield. Finally, cleavage of the
amide in 22 was carried out by reaction with sodium
methoxide giving an 83% yield of 1. Subsequent reaction
of 1 with amine-reactive BSA led to the expected glyco-
conjugate with an incorporation of four heptose residues
per protein.
In summary, this report describes the first synthesis of
the D-glycero-L-gluco-heptose monosaccharide present in
the CPS of C. jejuni NCTC11168. The route involved the
chain extension of a readily obtained galactofuranose
derivative (10), which was then converted into the target
1. This approach therefore differs from the usual approach
used for the synthesis of heptoses, where chain extension
proceeds via a C-6 oxidized hexose derivative. Further
investigations on the use of donors such as 16 and 19 in the
preparation of more complex glycoconjugates containing
this heptose moiety are under investigation.
Acknowledgment. This work supported by the Alberta
Innovates Centre for Carbohydrate Science and the Nat-
ural Sciences and Engineering Research Council of Cana-
da. AI thanks the Uehara Foundation for a postdoctoral
fellowship. We thank Dr. Michael Ferguson (University
of Alberta, Department of Chemistry, X-ray Crystallogra-
phy Laboratory) for obtaining the X-ray structure
of 17.
Supporting Information Available. Experimental de-
tails and 1H and 13C NMR spectra for new compounds.
X-ray crystallographic information for 17. This material
is available free of charge via the Internet at http://pubs.
acs.org
(15) Konradsson, P.; Udodong, U. E.; Fraser-Reid, B. Tetrahedron
Lett. 1990, 31, 4313–4316.
(16) Schmidt, R. R.; Kinzy, W. Adv. Carbohydr. Chem. Biochem.
1994, 50, 21–123.
(17) Gadikota, R. R.; Callam, C. S.; Appelmelk, B. J.; Lowary, T. L.
J. Carbohydr. Chem. 2003, 22, 459–480.
Org. Lett., Vol. 13, No. 19, 2011
5293