Palladium-Catalyzed C-S Bond Formation as a Tool for Latent-Active Glycosylation

Article abstract of DOI:10.1021/acs.orglett.0c02090

A high-yielding palladium-catalyzed C-S cross-coupling is presented for utilization in carbohydrate chemistry as a key transformation for attachment of a second chelating sulfur atom that allows the exploitation of a latent-active glycosylation strategy w

Full text of DOI:10.1021/acs.orglett.0c02090

pubs.acs.org/OrgLett
Letter
Palladium-Catalyzed C−S Bond Formation as a Tool for Latent−
Active Glycosylation
Christinne Hedberg, Kamilla S. Jessen, Rikke F. Hansson, Mads Heuckendorff, and Henrik H. Jensen*
Cite This: https://dx.doi.org/10.1021/acs.orglett.0c02090
Read Online
sı
*
Metrics & More
Article Recommendations
Supporting Information
ACCESS
ABSTRACT: A high-yielding palladium-catalyzed C−S cross-coupling is presented for utilization in carbohydrate chemistry as a
key transformation for attachment of a second chelating sulfur atom that allows the exploitation of a latent−active glycosylation
strategy with Cu(OTf)2 as the promoter. The novel approach employs o-Br-benzyl thioglycosides as latent glycosyl donors and o-
SMe-benzyl thioglycosides as the active counterparts.
he transition metal palladium is a highly valued tool
element for catalyzing transformations on especially sp2-
based glycosyl acceptor, which carries an unreactive (latent)
donor functionality that can be converted into an active donor
functionality for activation in a later step. The first latent−
active glycosylation strategies were published in the 1990s by
Roy,18 Fraser-Reid,19 and Boons,20 and later examples include
work from the groups of Kim,21 Huang,22 Wang,23
Demchenko,24 Yu,25 Wan,26,27 and Ragains.28 Various
strategies have been applied for the conversion from a latent
to active donor, including reduction and subsequent
acetylation,18,22 reductive debromination,19 Rh/Ir-catalyzed
isomerization,20,23 hydrogenolysis,21 N-deacylation,24 sulfur
oxidation26,27 and the before-mentioned Pd-catalyzed C−C
cross-couplings (Sonogashira and Suzuki couplings, respec-
tively).25,28 A recent example is given in Scheme 1A.
In oligosaccharide synthesis, the glycosyl donor leaving
group determines the mode of activation, which basically refers
to the choice of Lewis acid (e.g., halophilic, thiophilic, hard or
soft). Based on previous studies with anomeric benzoates as C-
1 leaving groups, we have found that a second coordination
center (o-OMe) significantly increases the glycosyl donor
reactivity.29 We wanted to explore whether a similar trend
would be present among one of the most used glycosyl donor
types, the thioglycosides.
T
hybridized centers. It has perhaps mostly been studied for the
preparation of C−C as well as C−N/C−O bonds, whereas far
fewer examples have been presented for the generation of C−S
bonds, which is probably due to deactivation of the metal
catalyst.1 The main driving force for conducting research into
C−S bond formation chemistry has arguably been to establish
powerful methods for preparing the many sulfur-containing
pharmaceuticals, but the C−S bond has additionally found
application in materials science.2−4
Although carbohydrate chemistry is a subdivision of organic
chemistry, palladium has only rarely ventured into this field
besides being used as a heterogeneous catalyst for O-
debenzylation. We have recently made some use of modern
homogeneous palladium catalysis for protecting group
manipulations5 and protecting group-mediated reactivity
tuning of glycosyl donors.6,7 In this paper we report on our
development of a novel latent−active glycosylation approach
using palladium as a central tool and describe how the
instalment of a sulfur atom is used to facilitate a subsequent
chemoselective step by exploiting the coordinating/chelating
ability of a sulfide.
Oligosaccharides are structurally highly diverse and much
less straightforward to synthesize compared to other
biomolecules such as oligopeptides and oligonucleotides.
Furthermore, due to their complexity, natural isolation has
not offered a way of obtaining adequate amounts for use in
studying their involvement in biological processes.8−13 There-
fore, new methods for the efficient synthesis of complex
carbohydrate structures are still needed and are being
developed.14−17
It was accordingly decided to study the activation of ortho-
(methylthio)benzyl thioglycosides (I and IV, Scheme 1B),
which is a hitherto undescribed glycosyl donor type. A reason
for selecting this donor structure for our investigations was the
Received: June 24, 2020
A powerful strategy used for streamlining oligosaccharide
synthesis is the latent−active method, where a glycosyl donor
is chemoselectively activated and reacted with a carbohydrate-
https://dx.doi.org/10.1021/acs.orglett.0c02090
© XXXX American Chemical Society
Org. Lett. XXXX, XXX, XXX−XXX
A

Organic Letters
pubs.acs.org/OrgLett
Letter
anomeric sulfur atom. We therefore went on to evaluate
various metal triflates as potential promoters hoping the
presence of two chelating sulfur atoms coordinated by the right
metal would facilitate leaving group departure (Conditions B,
Scheme 2). We investigated triflates of Cu(II), Fe(III), Bi(III),
In(III), Mg(II), Zn(II), Ni(II), Al(III), Sc(III), Ag(I), and
Yb(III) and found only Cu(II) and Fe(III) to be suitable for
activation in CH2Cl2 with L-menthol as the acceptor. Bi(OTf)3
also caused some degree of activation of 1, while the other
tested metal triflates only provided less than 10% conversion of
the donor or no activation at all. Cu(II) and Fe(III) triflates
elicited comparable activity, but since Fe(OTf)3 is highly
hygroscopic we chose to proceed with Cu(OTf)2. Only one
previous example of Cu(OTf)2 acting as the single activator of
thioglycosides (SEt) has previously been reported.30
Scheme 1. (A) Recent Example of a Latent-Active
Glycosylation (See Ref 28); (B) This Study Using a Pd-
Catalyzed C−S Cross-Coupling for Activation of the Latent
Donor
Although Cu(OTf)2 provided a large degree of donor
activation of 4 with L-menthol as the acceptor (Table 1),
Table 1. Results for Cu(OTf)2-Promoted Glycosylation with
Donor 4 with Isolated Yields
Entry
Solvent
MeCN
MeCN
MeCN
CH2Cl2/MeCN 2:1 TTBP, MS rt, 8 min
CH2Cl2/MeCN 4:1 TTBP, MS rt, 10 min
Additive
Conditions
rt, 15 min
0 °C, 80 min 80% (1.4:1)
Yield (α/β)
1
2
3
4
5
−
−
75% (1.7:1)
possibility of obtaining such donors from analogous thioglyco-
sides bearing a halide in the ortho-position (as in II and III,
Scheme 1B) through a palladium-catalyzed C−S cross-
coupling. Additionally, providing conditions would exist
where the o-SMe bearing glycosyl donor I could be activated
without affecting the o-halide functionalized donor II, a
potentially useful latent−active approach would be established.
In such a way, with reference to Scheme 1B, the donor
functionality of the product III could subsequent to the first
glycosylation be converted into the active o-SMe functionality
in compound IV, and this donor would be set for another
glycosylation step.
Initially, to investigate the reactivity of thioglycosides of the
type I, various glycosyl donors were synthesized by standard
methods using ortho-(methylthio)benzyl mercaptan and the
desired acetylated carbohydrate substrates (see Supporting
Information). The reactivity of a per-O-benzylated glucosyl
donor (1) of this type was tested against a per-O-benzylated
benzyl thioglucoside (2) in a competition experiment using
NIS/TfOH activation and L-menthol as the acceptor
(Conditions A, Scheme 2 and Supporting Information).
Opposite to our expectations, the ortho-substituted donor 1
was less reactive (ca. 30%) than the nonsubstituted thiobenzyl
donor 2, possibly due to increased hindrance around the
TTBP, MS rt, 7 min
94% (1:2.3)
92% (1:2.0)
94% (1:1.4)
conversion was never complete in CH2Cl2. Switching to
MeCN, which is known to dissolve Cu(OTf)2, resulted in full
conversion in only minutes at ambient temperature and
formation of 75% of the L-menthyl glycoside 5 (entry 1, Table
1). The yield was improved to 80% by performing the reaction
at 0 °C (entry 2), while addition of the hindered base tri-tert-
butylpyrimidine (TTBP)31 and molecular sieves (MS) further
increased the yield to 94% in just minutes at ambient
temperature (entry 3). Mixtures of CH2Cl2 and MeCN also
provided glycosidic product in high yields with a slightly
altered anomeric ratio (entries 4 and 5). Despite these
promising results in terms of yield and reaction rate, activation
was found to require a stoichiometric amount of Lewis acid.
Cu(OTf)2 was found to activate neither benzyl or phenyl
thioglycosides 2 and 6 (Scheme 3) in MeCN at ambient
temperature with L-menthol as the acceptor on the same time
scale as in Table 1. Only a trace of glycosidic product was
observed when the reactions was left overnight, suggesting the
importance of a metal chelating effect of the two sulfur atoms
in the ortho-substituted donor, activating the algycon and
making this donor a privileged substrate under these activation
Scheme 2. General Scheme for Activation of Ortho-
Substituted Benzyl Thioglycosides and Benzyl
Thioglycosides
Scheme 3. Sluggish Activation of SBn and SPh Donors with
Cu(OTf)2
B
https://dx.doi.org/10.1021/acs.orglett.0c02090
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
pubs.acs.org/OrgLett
Letter
conditions. Interestingly, the analogous o-SMe phenyl
functionalized donor possessing a CH2-group less than 1
showed a different activation profile where Bi(OTf)3 was
found to be a far more active promoter than Cu(OTf)2 (data
not shown). This fact underscores the importance of matching
the metal ion size and S−S distance.
Post glycosylation with donor 1, the aglycon could be
isolated as the corresponding disulfide. It has, however, not
been possible to determine whether the formation of disulfide
occurs during glycosylation or following the addition of
triethylamine during the workup procedure.
(7b and 7l) and acid-labile acetals (7c, 7e, and 7i). Acetylated
donors, however, proved too unstable for Cu(OTf)2 activation,
which is in accordance with previous results by us29 and Yu
and co-workers32 using similar promoter systems.
In line with the results from Scheme 3, S-phenyl function-
alized thioglycosides could furthermore be used as glycosyl
acceptors without any activation of these observed (7b and 7l).
It was therefore investigated whether the observed chemo-
selectivity could be exploited to obtain a trisaccharide in a one-
pot fashion (Scheme 5). After full conversion of 8 was
Turning to more relevant carbohydrate-based alcohols in the
context of oligosaccharide synthesis established that these
more demanding acceptors also lead to excellent glycosylation
yields in the time scale of minutes (Scheme 4). The D-
Scheme 5. Synthesis of Trisaccharide 11 in a One-Pot
Fashion
Scheme 4. Glycosylations with Carbohydrate-Based
Acceptors
observed in the first step, the acceptor 10 and NIS were added
to the mixture at rt. There was no need for the addition of any
additional acid for activation of NIS due to the Lewis acid
Cu(OTf)2 already present.33 The trisaccharide was obtained in
a yield of 48% based on the amount of 8, and an even higher
yield would expectedly have been achieved, had it not been for
the troublesome purification of 11.
The key conversion from a latent to the active o-SMe
functionalized glycosyl donor was subsequently explored
through a Pd-catalyzed substitution of an ortho-halide
substituent by SMe. Since an S-benzyl glycosyl donor has
previously proven unreactive toward Cu(OTf)2 activation, we
expected the ortho-halide functionalized donor to be as well
given the bulk and electron-withdrawing nature of the halide.
Literature reveals that bidentate ligands generally are the
ligands of choice for Pd-catalyzed C−S bond formations34−37
and such ligands were therefore explored. After initial
investigations, it was found that ortho-bromo analogues of
the active donor and reaction conditions by Okauchi and co-
workers using Pd2(dba)3 as catalyst, DPPF as ligand, and
DIPEA as base in refluxing toluene with NaSMe as the source
of sulfur would allow the desired transformation (Scheme 6).36
Satisfyingly, the conditions involving both cheap ligand and
catalyst as well as an easily handled protocol without metal/
ligand premixing resulted in the desired transformations in
almost quantitative yields. Furthermore, donors carrying
various protecting groups including ethers, acetals, and silyl
ethers as well as free alcohols were evaluated as substrates.
Other procedures tested included, e.g., the conditions by
Hartwig and co-workers using the Josiphos ligand CyPF-tBu
and Pd(OAc)2 as catalyst together with NaOtBu as base and
octanethiol as the source of sulfur.35 Even though this
procedure has proved successful on simpler chloroarenes, no
reaction was observed on an ortho-chloro analogue of 12.
In addition to the substitutions with NaSMe, we also
established that the reaction could be achieved using NaSiPr as
the source of sulfur providing an alternative access to the
analogous sulfide 13d being a latent donor functionality with
galactosyl donor 4 provided good yields with primary alcohol
acceptors (yielding glycosides 7a, 7b, and 7c), but also with
more hindered secondary acceptors (yielding glycosides 7d
and 7e). The highest yields of the latter two were, however,
obtained at 0 °C. D-Glucosyl, L-fucosyl, and 2-O-Piv-protected
D-galactosyl donors also provided high yields with L-menthol as
the acceptor (7f, 7g, and 7j), and glycosylations with the L-
fucosyl and 2-O-Piv-protected D-galactosyl donors were
achieved with carbohydrate-based glycosyl acceptors in high
yields (7h, 7i, 7k, and 7l). Noticeably, the results in Scheme 4
demonstrate the compatibility of the Cu(OTf)2/TTBP
promoter system with various protecting groups such as esters
C
https://dx.doi.org/10.1021/acs.orglett.0c02090
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
pubs.acs.org/OrgLett
Letter
can be activated by Cu(OTf)2 and that this Lewis acid leaves
thiophenyl, thiobenzyl, and o-Br-benzyl thioglycosides un-
touched. It was also demonstrated that o-SMe/o-Br benzyl
thioglycosides is a novel latent−active glycosyl donor pair for
obtaining oligosaccharides of biological relevance. Two
trisaccharides were prepared in the course of our studies.
One of these demonstrate the use of our novel protocol in
latent−active glycosylation, and the other shows how our
approach can be used in combination with existing methods
for thioglycoside activation in a one-pot two-step fashion. We
believe our present approach is a powerful alternative to
already existing glycosylation methodology, and it will prove
useful in the synthesis of complex oligosaccharides. Potentially,
transition metal catalysis can also be used in other contexts
than glycosylation chemistry for conversion of an unreactive
moiety into a reactive moiety.
Scheme 6. Pd-Catalyzed C−S Cross-Coupling (Yields Given
Are Isolated Yields after Column Chromatography)
respect to activation by Tf2O as recently described by Wan and
co-workers.26,27 The o-SiPr functionalized galactosyl donor
13d was found to undergo activation at a similar rate as the
analogous o-SMe counterpart in the presence of Cu(OTf)2/
TTBP. With L-menthol as the acceptor, the o-SiPr function-
alized donor provided the O-menthyl glycoside 5 in 96% yield
under the conditions optimized for the o-SMe donor. This
offers another option for reactivity tuning during oligosacchar-
ide synthesis as our latent donor has the possibility of being
converted into another active donor (o-SiPr), which is however
inert under the reactions conditions developed by Wan an co-
workers, and can, if desired, be transformed into an active
donor under their glycosylation system.26,27
Finally, to demonstrate our novel latent−active glycosylation
strategy for oligosaccharide synthesis we aimed at the
preparation of trisaccharide 18 (Scheme 7). First, L-fucosyl
donor 14 was used in a glycosylation reaction with the
acceptor/latent D-galactosyl donor 15 providing the dis-
accharide 16 in 83% yield. The two resulting anomers were
completely separated by flash chromatography, and only the α-
anomer (α-16) was used further. α-16 was then converted into
the active donor 17 through the established reaction
conditions from Scheme 6 in 88% yield. Noticeably, the
ester group remains stable under the reaction conditions. To
obtain the trisaccharide, 17 was subsequently coupled with
acceptor 10, providing trisaccharide 18 as a single anomer due
to the participating 2-O-Piv group.
ASSOCIATED CONTENT
■
sı
* Supporting Information
The Supporting Information is available free of charge at
https://pubs.acs.org/doi/10.1021/acs.orglett.0c02090.
Synthetic experimental procedure and full character-
izations including NMR spectra (PDF)
AUTHOR INFORMATION
■
Corresponding Author
Henrik H. Jensen − Department of Chemistry, Aarhus
University, DK-8000 Aarhus C, Denmark; orcid.org/0000-
0003-2738-4502; Email: hhj@chem.au.dk
Authors
Christinne Hedberg − Department of Chemistry, Aarhus
University, DK-8000 Aarhus C, Denmark
Kamilla S. Jessen − Department of Chemistry, Aarhus
University, DK-8000 Aarhus C, Denmark
Rikke F. Hansson − Department of Chemistry, Aarhus
University, DK-8000 Aarhus C, Denmark
Mads Heuckendorff − Department of Chemistry, Aarhus
University, DK-8000 Aarhus C, Denmark
Complete contact information is available at:
https://pubs.acs.org/10.1021/acs.orglett.0c02090
To conclude on our studies, we have shown how palladium
catalyzed C−S bond formation can be used as a tool for the
installation of a chelating site that allows for later chemical
manipulations via chemoselective reactions. We have specifi-
cally shown that o-SMe functionalized benzyl thioglycosides
Notes
The authors declare no competing financial interest.
Scheme 7. Synthesis of a Trisaccharide Using the Latent−Active Glycosylation Strategy
D
https://dx.doi.org/10.1021/acs.orglett.0c02090
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
pubs.acs.org/OrgLett
Letter
(32) Ma, Y.; Lian, G.; Li, Y.; Yu, B. Chem. Commun. 2011, 47,
7515−7517.
ACKNOWLEDGMENTS
■
We are grateful for financial support from The Villum
Foundation (VKR023110).
(33) Krag, J.; Christiansen, M. S.; Petersen, J. G.; Jensen, H. H.
Carbohydr. Res. 2010, 345, 872−879.
(34) Murata, M.; Buchwald, S. L. Tetrahedron 2004, 60, 7397−7403.
(35) Mispelaere-Canivet, C.; Spindler, J.-F.; Perrio, S.; Beslin, P.
Tetrahedron 2005, 61, 5253−5259.
REFERENCES
■
(1) Kondo, T.; Mitsudo, T. Chem. Rev. 2000, 100, 3205−3220.
(2) Li, J.; Yang, S.; Wu, W.; Jiang, H. Org. Chem. Front. 2020, 7,
1395−1417.
́
(36) Fernandez-Rodríguez, M. A.; Shen, Q.; Hartwig, J. F. Chem.
Eur. J. 2006, 12, 7782−7796.
(37) Okauchi, T.; Kuramoto, K.; Kitamura, M. Synlett 2010, 2010,
2891−2894.
(3) Lee, C.-F.; Liu, Y.-C.; Badsara, S. S. Chem. - Asian J. 2014, 9,
706−722.
(4) Ghaderi, A. Tetrahedron 2016, 72, 4758−4782.
(5) Viuff, A. H.; Heuckendorff, M.; Jensen, H. H. Org. Lett. 2016, 18,
5773−5775.
(6) Heuckendorff, M.; Poulsen, L. T.; Jensen, H. H. J. Org. Chem.
2016, 81, 4988−5006.
(7) Poulsen, L. T.; Heuckendorff, M.; Jensen, H. H. ACS Omega
2018, 3, 7117−7123.
(8) Bertozzi, C. R.; Kiessling, L. L. Science 2001, 291, 2357−2364.
(9) Hart, G. W.; Copeland, R. J. Cell 2010, 143, 672−676.
(10) Varki, A.; Gagneux, P. In Essentials of Glycobiology, 3rd ed.;
Varki, A.; Cummings, R. D., Esko, J. D., Stanley, P., Hart, G. W., Aebi,
M., Darvill, A. G., Kinoshita, T., Packer, N. H., Prestegard, J. H.,
Schnaar, R. L., Seeberger, P. H., Eds.; Cold Spring Harbor Laboratory
Press: 2017; pp 77−88.
(11) Rudd, P.; Karlsson, N. G.; Khoo, K.-H.; Packer, N. H. In
Essentials of Glycobiology, 3rd ed.; Varki, A., Cummings, R. D., Esko, J.
D., Stanley, P., Hart, G. W., Aebi, M., Darvill, A. G., Kinoshita, T.,
Packer, N. H., Prestegard, J. H., Schnaar, R. L., Seeberger, P. H., Eds.;
Cold Spring Harbor Laboratory Press: 2017; pp 653−666.
(12) Werz, D. B.; Ranzinger, R.; Herget, S.; Adibekian, A.; von der
Lieth, C.-W.; Seeberger, P. H. ACS Chem. Biol. 2007, 2, 685−691.
(13) Seeberger, P. H.; Werz, D. B. Nature 2007, 446, 1046−1051.
(14) Demchenko, A. V. In Handbook of Chemical Glycosylation;
Demchenko, A. V., Ed.; Wiley-VCH: 2008; pp 16−21.
́
(15) Codee, J. D. C.; Litjens, R. E. J. N.; van den Bos, L. J.;
Overkleeft, H. S.; van der Marel, G. A. Chem. Soc. Rev. 2005, 34, 769−
782.
(16) Yu, B.; Yang, Z.; Cao, H. Curr. Org. Chem. 2005, 9, 179−194.
(17) Kulkarni, S. S.; Wang, C.-C.; Sabbavarapu, N. M.; Podilapu, A.
R.; Liao, P.-H.; Hung, S.-C. Chem. Rev. 2018, 118, 8025−8104.
(18) Roy, R.; Andersson, F. O.; Letellier, M. Tetrahedron Lett. 1992,
33, 6053−6056.
(19) Fraser-Reid, B.; Udodong, U. E.; Wu, Z.; Ottosson, H.; Merritt,
J. R.; Rao, C. S.; Roberts, C.; Madsen, R. Synlett 1992, 1992, 927−
942.
(20) Boons, G.-J.; Isles, S. Tetrahedron Lett. 1994, 35, 3593−3596.
(21) Kim, K. S.; Kim, J. H.; Lee, Y. J.; Lee, Y. J.; Park, J. J. Am. Chem.
Soc. 2001, 123, 8477−8481.
(22) Huang, L.; Wang, Z.; Huang, X. Chem. Commun. 2004, 1960−
1961.
(23) Wang, P.; Haldar, P.; Wang, Y.; Hu, H. J. Org. Chem. 2007, 72,
5870−5873.
(24) Hasty, S. J.; Kleine, M. A.; Demchenko, A. V. Angew. Chem. Int.
Ed. 2011, 50, 4197−4201.
(25) Chen, X.; Shen, D.; Wang, Q.; Yang, Y.; Yu, B. Chem. Commun.
2015, 51, 13957−13960.
(26) Shu, P.; Xiao, X.; Zhao, Y.; Xu, Y.; Yao, W.; Tao, J.; Wang, H.;
Yao, G.; Lu, Z.; Zeng, J.; Wan, Q. Angew. Chem. Int. Ed. 2015, 54,
14432−14436.
(27) Xiao, X.; Zhao, Y.; Shu, P.; Zhao, X.; Liu, Y.; Sun, J.; Zhang, Q.;
Zeng, J.; Wan, Q. J. Am. Chem. Soc. 2016, 138, 13402−13407.
(28) Du, S.; Ragains, J. R. Org. Lett. 2019, 21, 980−983.
(29) Kristensen, S. K.; Salamone, S.; Rasmussen, M. R.; Marqvorsen,
M. H. S.; Jensen, H. H. Eur. J. Org. Chem. 2016, 2016, 5365−5376.
(30) Dondoni, A.; Marra, A.; Massi, A. Angew. Chem. Int. Ed. 2005,
44, 1672−1676.
(31) Crich, D.; Smith, M.; Yao, Q.; Picione, J. Synthesis 2001, 2001,
323−326.
E
https://dx.doi.org/10.1021/acs.orglett.0c02090
Org. Lett. XXXX, XXX, XXX−XXX