Total Synthesis of Trisaccharide Repeating Unit of O-Specific Polysaccharide of Pseudomonas fluorescens BIM B-582

Article abstract of DOI:10.1021/acs.orglett.8b02669

The first total synthesis of the trisaccharide repeating unit of the O-specific polysaccharide of Pseudomonas fluorescens BIM B-582 is reported. This efficient synthesis involves consecutive 1,2-cis glycosylations including β-l-rhamnosylation and α select

Full text of DOI:10.1021/acs.orglett.8b02669

Letter
pubs.acs.org/OrgLett
Cite This: Org. Lett. XXXX, XXX, XXX−XXX
Total Synthesis of Trisaccharide Repeating Unit of O‑Specific
Polysaccharide of Pseudomonas fluorescens BIM B‑582
Archanamayee Behera, Diksha Rai, Divya Kushwaha, and Suvarn S. Kulkarni*
Department of Chemistry, Indian Institute of Technology Bombay, Mumbai 400076, India
S
* Supporting Information
ABSTRACT: The first total synthesis of the trisaccharide repeating unit
of the O-specific polysaccharide of Pseudomonas fluorescens BIM B-582 is
reported. This efficient synthesis involves consecutive 1,2-cis glyco-
sylations including β-L-rhamnosylation and α selective coupling of rare
4-deoxy-D-xylo-hexose as the key steps. The synthetic trisaccharide is
equipped with an aminopropyl linker at the reducing end to allow for
conjugation to proteins and microarrays for further immunological
studies.
Pseudomonas fluorescens, a typical representative of the wide-
spread genus Pseudomonas from the family Pseudomonadaceae,
is a Gram-negative rod-shaped bacterium found in soil and
water. P. fluorescens possesses hemolytic activity which may
affect blood transfusion, especially in immunocompromised
patients.1 This pathogen is also one of the important
phytopathogens which causes severe bacteriosis in a wide
range of cultured plants and also causes spoilage of food,
including milk, cheese, and meat.2,3 P. fluorescens is known to
form biofilms,4 which are resistant to a variety of antibiotics.
Thus, alternative methods to eradicate bacteria in the biofilm
form, such as phage treatment, where the O-antigen is the major
target, have been suggested.5,6 Availability of the O-antigen
would enable studies toward delineating the mechanism of the
interaction between the phages and bacteria.
to L-rhamnose at O2, which in turn, is further β linked to the D-
glucosamine moiety at O4. The construction of this
trisaccharide repeating unit would encounter two major
challenges. First, installation of the β-L-rhamnopyranosidic
linkage8 would present a well-known difficulty due to an
unfavorable anomeric effect and steric hindrance from the C2
axial group which also would preclude neighboring group
participation. Moreover, a conformational restriction approach
was ruled out because of its 6-deoxy structure, which prevents
use of a 4,6-O-benzylidene acetal9 to restrict the conformation,
to selectively install the desired β glycosidic bond. Another
important challenge would be synthesis of the 4-deoxy-D-
xyloHexp building block and its α selective coupling with the 2-
OH of L-rhamnose. The consecutive 1−2 cis linkages on the
adjacent 1,2-positions of L-rhamnose make the synthesis of this
trisaccharide repeating unit challenging.
Our retrosynthetic approach for the repeating trisaccharide
unit of P. fluorescens is outlined in Figure 2. The target molecule
1 could potentially be obtained by debenzoylation and
hydrogenolysis of fully protected linker appended trisaccharide
2, which could be assembled by α selective coupling of 4-deoxy
thioglycoside donor 4 and disaccharide acceptor 3. The 4-deoxy
donor 4 could, in turn, be obtained from commercially available
D-galactose. Disaccharide 3 might be synthesized by glyco-
sylation of a suitably protected L-quinovose donor 5 with
glucosamine acceptor 6 followed by a C2′ inversion. Compound
5 could be obtained by C2 epimerization of a semiprotected L-
rhamnose derivative. Alternatively, a direct stereoselective β-
rhamnosylation with a suitably protected L-rhamnosyl donor
with acceptor 6 would afford the disaccharide 3. The D-
glucosamine acceptor 6 might possibly be assembled from
thioglycoside donor 7 and linker derivative 8. Glucosamine
derivative 7 would be obtained via a C2 inversion of
commercially available D-mannose,10 while the linker derivative
In 2011, Knirel and co-workers7 isolated a major O-specific
polysaccharide from the lipopolysaccharide of P. fluorescens BIM
B-582. A unique feature of this polysaccharide is that it contains
a novel and hitherto unknown rare sugar component, 4-deoxy-D-
xylo-hexose (D-4dxylHex), as a lateral substituent in 40% of the
oligosaccharide repeating units. The polysaccharide contains a
disaccharide →3)-β-L-Rhap-(1 → 4)-β-D-GlcpNAc-(1→ back-
bone repeat with some of the β-L-rhamnose units appended at
O2 with the α-D-4dxylHexp. The rare sugar-containing
trisaccharide repeating unit (Figure 1) is crucial for bacterial
specificity and therefore is an important synthetic target.
Structurally, the trisaccharide is composed of a terminal 4-
deoxy-D-xylo-hexose (4-deoxy-D-glucose) unit which is α linked
Received: August 21, 2018
Figure 1. Structure of O-specific polysaccharide repeating unit.
© XXXX American Chemical Society
A
DOI: 10.1021/acs.orglett.8b02669
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
anhydride in pyridine, and the so-formed O-triflate was
displaced by NaN3 to furnish glucosamine derivative 11 in
80% yield over two steps. Reduction of the azide in 11 to the
amine using Zn and AcOH followed by trichloroacetyl
protection furnished thioglycoside donor 7 in 92% yield over
two steps.
Glycosylation of donor 7, with linker derivative 8 using NIS
and AgOTf as a promoter at −15 °C gave β-linked glycoside 12
in 82% yield as a sole product. Reductive benzylidene ring
opening of 12 using TFA and Et3SiH furnished 4-OH
glucosamine acceptor 6 in 62% yield. We first explored the
synthesis of the key disaccharide backbone repeating unit
employing indirect installation of the 1,2-cis rhamnosyl linkage
via a double SN2 inversion. For this purpose, the orthogonally
protected rare L-quinovose derivative 5 was synthesized from L-
rhamnose along the lines of our previously established protocol
(Scheme 2).11
Scheme 2. Synthesis of L-Quinovose Donor 5
Figure 2. Retrosynthetic analysis.
8 could be prepared from commercially available 3-amino-1-
propanol. With these considerations in mind, we began the total
synthesis of 1.
Our synthesis of the reducing end 4-OH glucosamine
acceptor 6 from commercially available and abundant D-
mannose is shown in Scheme 1. The easily accessible 2,3-diol
L-Quinovose is a C2 epimer of L-rhamnose. Alcohol 13 was
obtained by selective 3-O-benzoylation of the easily accessible L-
rhamnosyl 2,3-diol under tin-mediated conditions.12 Triflation
of the 2-OH of 13, followed by SN2 displacement of the formed
C2-OTf by the nitrite anion, smoothly furnished L-quinovoside
derivative 14 in 70% yield over two steps. It should be noted that
the reaction worked well with potassium nitrite even without
addition of a crown ether. The ease of SN2 displacement of OTf
in this case is in accordance with the proposed Hale−Hough−
Richardson rules for nucleophilic displacement of pyranosidic
O-triflates.13 The free 2-OH in 14 was capped with a levulinoyl
group upon treatment with levulinic acid, EDC·HCl, and
DMAP, which afforded fully protected L-quinovoside donor 5 in
87% yield.
Scheme 1. Synthesis of D-Glucosamine Acceptor 6
With all of the desired building blocks in hand, we went ahead
with the coupling of 5 and 6 as outlined in Scheme 3. L-
Quinovoside donor 5 and D-glucosamine acceptor 6 were
glycosylated in the presence of NIS and TMSOTf as a promoter
at 0 °C to deliver the β-linked disaccharide 15 in 97% yield as a
single isomer. The selectivity can be attributed to the
neighboring group participation of the C2 ester group in 5.
Regioselective deprotection of the levulinoyl group using
NH2NH2·H2O afforded disaccharide 16 in 72% yield. The
equatorial 2′-OH in 16 was then epimerized back to axial 2′-OH
to obtain the 1,2-cis-L-rhamno configuration. Accordingly, the
free 2′-OH was subjected to O-triflation using triflic anhydride
and pyridine, and subsequently, the formed O-triflate was
treated with NaNO2, HMPA, and 15-crown-5 to furnish the
desired disaccharide acceptor 3 in 48% yield over two steps with
clean selectivity. Along with the product, surprisingly, we also
observed glycosidic bond cleavage and around 20% glucosamine
acceptor 6 (confirmed from 1H NMR, HRMS, and TLC
910 was first converted to the corresponding tin ketal by
treatment with dibutyltin oxide in toluene at 110 °C and was
further reacted with benzyl bromide and tetrabutylammonium
bromide (TBAB) at 60 °C to afford the 3-OBn derivative 10, in
high regioselectivity, as a single isomer in 81% yield over two
steps. The C2-OH was subjected to triflation using triflic
B
DOI: 10.1021/acs.orglett.8b02669
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
Scheme 3. Synthesis of Disaccharide Acceptor 3
Scheme 5. Synthesis of Disaccharide Acceptor 3
reported a moderate 1:3 α/β selectivity with a less reactive 4-
OH glucosyl acceptor using a 3-O-picoloyl donor.15
standard) was recovered, which could be recycled back for the
coupling reactions.
Depicolylation of disaccharide 20 in the presence of
16
Cu(OAc)2 in MeOH and CH2Cl2 cleanly afforded dis-
Alternatively, we attempted the more challenging direct β-
rhamnosylation route. Recently, Demchenko and co-workers
established a picolinoyl group mediated hydrogen bond aglycon
delivery (HAD) method for creating 1,2-cis glycosidic linkages
with high selectivity.8d Using this method, they report a 1:25 α/
β selectivity in the glycosylation of an L-rhamnose sugar with a
highly reactive 6-OH glucosyl acceptor when the Pico group is
placed at the O3 position. Although in our case we needed to
glycosylate a poorly reactive 4-OH glucosamine acceptor, we
decided to try out their conditions.
Thus, we went ahead with the synthesis of the 3-O-picoloyl
donor 19 as delineated in Scheme 4. Diol 17 was synthesized on
gram scale from commercially available and abundant starting
material L-rhamnose over six steps in good yield.14
accharide 3′-OH derivative 21 in 80% yield. Benzoylation of
the free 3′-OH using BzCl and pyridine afforded disaccharide 22
in 82% yield. PdCl2-catalyzed deallylation of disaccharide 22
cleanly furnished disaccharide 2′-OH acceptor 3 in 78% yield.
The spectral data of 3 were identical to the data of previously
obtained compound 3 synthesized via a C2′ inversion route
(Scheme 3). Disaccharide 21 constitutes the backbone
disaccharide repeating unit of P. fluorescens BIM B-582.
For the synthesis of the rare 4-deoxy sugar (Scheme 6), the
known 2,3-diol 2317 was benzylated using BnBr and NaH to
Scheme 6. Synthesis of 4-Deoxythio Donor 4
Scheme 4. Synthesis of Pico Donor 19
Regioselective allylation of diol 17 using AllylBr and NaH at 0
°C afforded 3-OH derivative 18 in 62% yield (along with 15%
diallylated product). The free OH in 18 was capped with the
picoloyl group via an EDC·HCl-mediated coupling with
picolinic acid to obtain pico donor 19 in 93% yield.
Having both the donor 19 and acceptor 6 in hand, we
proceeded with the picoloyl (HAD)-mediated glycosylation for
the synthesis of our key disaccharide as described in Scheme 5.
Gratifyingly, the glycosylation of pico donor 19 with 4-OH
glucosamine acceptor 6 in the presence of NIS and TfOH in 1,2-
dichloroethane as solvent at −30 to 0 °C delivered the desired
disaccharide 20 in 1:8.4 α/β selectivity and 80% yield. The
minor unwanted α-isomer could be cleanly separated by flash
silica gel column chromatography using a EtOAc/hexane
solvent mixture to obtain the desired β isomer in pure form.
While this work was in progress, Seeberger and co-workers
afford 2,3-di-O-benzyl derivative 24 in 99% yield. Reductive
benzylidene ring opening of 24 using TFA and Et3SiH furnished
4-OH thiogalactoside donor 25 (82%), which upon treatment
with CS2, MeI, and NaH18 afforded xanthate ester 26 (71%).
The so-formed xanthate was deoxygenated by using TBTH and
AIBN to deliver 4-deoxy thio donor 4 in 72% yield.
Assembly of the target molecule, its functional group
transformation, and global deprotection are delineated in
Scheme 7. The 4-deoxy thioglycoside donor 4 was activated
by NIS and TMSOTf promoter in CH2Cl2/Et2O as the
participating solvent and glycosylated with acceptor 3 to cleanly
C
DOI: 10.1021/acs.orglett.8b02669
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
Scheme 7. Assembly of Trisaccharide 2 and Global Deprotection
afford the desired α-linked trisaccharide 2 as a single isomer in
82% yield. Selective removal of the benzoyl group in the
presence of NaOMe and MeOH furnished trisaccharide alcohol
27 in 82% yield. Hydrogenolytic removal of the benzyl groups
and concomitant reduction of NHTCA to NHAc using H2/
Pd(OH)2 and filtration furnished the target trisaccharide 1 in
92% yield.
In conclusion, we have accomplished the first total synthesis
of the trisaccharide repeating unit of the O-specific poly-
saccharide from P. fluorescens BIM B-582. The glycan is
equipped with a β-O-linked aminopropyl linker at the reducing
end to allow for further attachment to carrier proteins. The key
features of the synthesis are efficient synthesis of the challenging
β rhamnosidic disaccharide via double inversion or HAD
mediated direct glycosylation pathways and its further
elaboration into the target trisaccharide via 1,2-cis glycosylation
of a rare 4-deoxy sugar. Trisacccharide 1, which may play
immunodominant roles, is now available for bioevaluation to
assess its immunological potential.
REFERENCES
■
(1) Sperandio, D.; Rossignol, G.; Guerillon, J.; Connil, N.; Orange, N.;
Feuilloley, M. G.; Merieau, A. BMC Microbiol. 2010, 10, 124.
(2) (a) Dogan, B.; Boor, K. J. Appl. Environ. Microbiol. 2003, 69, 130−
138. (b) Del Olmo, A.; Calzada, J.; Nunez, M. Food Control 2018, 86,
359−366.
(3) Delaquis, P. J.; McCurdy, A. R. J. Food Sci. 1990, 55, 898−902.
(4) O’Toole, G.; Kaplan, H. B.; Kolter, R. Annu. Rev. Microbiol. 2000,
54, 49−79.
(5) Sillankorva, S.; Oliveira, R.; Vieira, M. J.; Sutherland, I.; Azeredo, J.
FEMS Microbiol. Lett. 2004, 241, 13−20.
(6) Sillankorva, S.; Neubauer, P.; Azeredo, J. BMC Biotechnol. 2008, 8,
79.
(7) Valueva, O. A.; Rakhuba, D.; Shashkov, A. S.; Zdorovenko, E. L.;
Kiseleva, E.; Novik, G.; Knirel, Y. A. J. Nat. Prod. 2011, 74, 2161−2167.
(8) For β-rhamnosylation, see: (a) El Ashry, E. S. H.; Rashed, N.;
Ibrahim, E. S. I. Tetrahedron 2008, 64, 10631−10648. (b) Lee, Y. J.;
Ishiwata, A.; Ito, Y. J. Am. Chem. Soc. 2008, 130, 6330−6331. (c) Crich,
D.; Li, L. J. Org. Chem. 2009, 74, 773−781. (d) Yasomanee, J. P.;
Demchenko, A. V. J. Am. Chem. Soc. 2012, 134, 20097−20102.
(e) Heuckendorff, M.; Pedersen, C. M.; Bols, M. J. Org. Chem. 2012, 77,
5559−5568. (f) Zhu, Y.; Shen, Z.; Li, W.; Yu, B. Org. Biomol. Chem.
2016, 14, 1536−1539.
(9) (a) Crich, D.; Sun, S. J. Org. Chem. 1996, 61, 4506−4507.
(b) Crich, D.; Sun, S. J. Am. Chem. Soc. 1998, 120, 435−436. (c) Kim,
K. S.; Kim, J. H.; Lee, Y. J.; Lee, Y. J.; Park, J. J. Am. Chem. Soc. 2001,
123, 8477−8481. (d) Baek, J. Y.; Choi, T. J.; Jeon, H. B.; Kim, K. S.
Angew. Chem., Int. Ed. 2006, 45, 7436−7440.
(10) Emmadi, M.; Kulkarni, S. S. J. Org. Chem. 2011, 76, 4703−4709.
(11) Sanapala, S. R.; Kulkarni, S. S. Org. Lett. 2016, 18, 3790−3793.
(12) Sanapala, S. S.; Kulkarni, S. S. J. Am. Chem. Soc. 2016, 138, 4938−
4947.
(13) Hale, K. J.; Hough, L.; Manaviazar, S.; Calabrese, A. Org. Lett.
2014, 16, 4838−4841.
(14) Bourke, J.; Brereton, C. F.; Gordon, S. V.; Lavelle, E.; Scanlan, E.
M. Org. Biomol. Chem. 2014, 12, 1114−1123.
ASSOCIATED CONTENT
* Supporting Information
■
S
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acs.orglett.8b02669.
Experimental details and procedures, compound charac-
1
terization data, and copies of H and 13C spectra for all
new compounds (PDF)
AUTHOR INFORMATION
■
Corresponding Author
*E-mail: suvarn@chem.iitb.ac.in.
ORCID
(15) Emmadi, M.; Khan, N.; Lykke, N.; Reppe, K.; Parameswarappa,
S. G.; Lisboa, M. P.; Wienhold, M. P.; Witzenrath, M.; Pereira, C. L.;
Seeberger, P. H. J. Am. Chem. Soc. 2017, 139, 14783−14791.
(16) Kayastha, A. K.; Jia, X. G.; Yasomanee, J. P.; Demchenko, A. V.
Org. Lett. 2015, 17, 4448−4451.
(17) Podilapu, A. R.; Kulkarni, S. S. Org. Lett. 2017, 19, 5466−5469.
(18) Barton, D. H. R.; McCombie, S. W. J. Chem. Soc., Perkin Trans. 1
1975, 1574−1585.
Suvarn S. Kulkarni: 0000-0003-2884-876X
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
This work was supported by the Science and Engineering
Research Board, DST (Grant No. EMR/2014/000235) and the
ISF-UGC Joint Research Program Framework (Grant No.
2253). A.B. thanks UGC−New Delhi and IIT Bombay for
research fellowships. D.R. and D.K. thank DST for an Inspire
Fellowship and National Post Doctoral Fellowship, respectively.
D
DOI: 10.1021/acs.orglett.8b02669
Org. Lett. XXXX, XXX, XXX−XXX