A highly versatile convergent/divergent "onion peel" synthetic strategy toward potent multivalent glycodendrimers

Article abstract of DOI:10.1039/c4cc06191h

Both convergent and divergent strategies for the synthesis of "onion peel" glycodendrimers are reported which resulted in one of the best multivalent ligands known against the virulent factor from a bacterial lectin isolated from Pseudomonas aeruginosa. T

Full text of DOI:10.1039/c4cc06191h

ChemComm
View Article Online
View Journal | View Issue
COMMUNICATION
A highly versatile convergent/divergent ‘‘onion
peel’’ synthetic strategy toward potent multivalent
glycodendrimers†
Cite this: Chem. Commun., 2014,
50, 13300
Received 7th August 2014,
Accepted 4th September 2014
Rishi Sharma, Naresh Kottari, Yoann M. Chabre, Leıla Abbassi, Tze Chieh Shiao and
¨
Rene Roy*
´
DOI: 10.1039/c4cc06191h
www.rsc.org/chemcomm
Both convergent and divergent strategies for the synthesis of ‘‘onion balances together with presentation of different epitopes clearly
peel’’ glycodendrimers are reported which resulted in one of the best influenced their potencies as protein ligands. In complement to
multivalent ligands known against the virulent factor from a bacterial this rationally programmed arrangement of branching units, we
lectin isolated from Pseudomonas aeruginosa.
wish to report herein an inverted strategy with multivalent presenta-
tion of different types of ligands around a fixed ‘‘onion peel’’
Dendrimers are well defined, hyperbranched tree like macro- dendritic scaffold. Chemically heterogeneous layers were assembled
molecules which have shown great potential for applications in at each generation by both convergent and divergent strategies using
diverse areas ranging from nanoengineering to medicine.1 Their a combination of orthogonal building blocks and highly efficient
striking architecture leads to excellent properties, but unfortu- chemical reactions such as radical initiated photochemical thiol–ene
nately brings in many synthetic challenges as well. Traditionally, coupling (TEC) reactions,5,9 amidation,10 and copper-catalyzed
their iterative construction emanates from a central core in a azide–alkyne cycloaddition (CuAAc) reactions.11
layer-by-layer fashion using repetitive moieties via most popular
The divergent construction of these novel dendrimers was
divergent2 and convergent3 methods. Both strategies have their initiated with inexpensive commercially available dipentaerythritol
own drawbacks and often require tedious repetitive synthetic 1 serving as a dense A6 core. Per-O-allylation with allyl bromide in
steps, with classically only a slow enhancement in the number of the presence of NaH in DMF provided hexakisallylated G(0) deriva-
peripheral functionalities at each generation. To meet the tive 2 in 80% yield (Scheme 1). Complete allylation was clearly
1
increasing demand of dendrimers for advanced applications, confirmed by H NMR analysis, which showed the characteristic
the scientific focus has been shifted towards their efficient and allylic signals at d 5.90 and 5.34–5.08 ppm and the disappearance of
rapid construction involving a minimum number of reactions OH signals together with its predicted HRMS. Core structure 2 was
and with access to a large number of surface active function- next subjected to a radical TEC reaction with excess of cysteamine
alities. Notably, the introduction of orthogonal building blocks hydrochloride in the presence of photoinitiator 2,2-dimethoxy-2-
and the use of hyperfunctionalized synthons combined with phenylacetophenone (DMPAP, 10 mol%) under UV irradiation
robust and highly efficient chemical reactions have recently at 365 nm in DMF. Water soluble hydrochloride 3 was unevent-
fulfilled these specifications.4–6
fully isolated in 75% yield after dialysis and fully characterized
Glycodendrimers in particular, with their widespread applications7 by H- and 13C-NMR spectroscopy that showed the absence of
as microbial antiadhesins, biosensors, vaccines, drug delivery olefinic signals, and by HRMS. Polyamine 3 was then treated with
vectors, and gene transfection agents, do not depart from this tripropargylated gallic acid derivative 412 by amidation under
efficacy pursuit. In this context, we recently reported a novel classical carbodiimide coupling (72%). Notably, the use of AB3
divergent ‘‘onion peel’’ approach to construct glycodendrimers monomer 4, when combined with our A6 core 2, readily provided
using distinct and orthogonal building blocks at each generation G(1) hypercore 5 already possessing eighteen surface functional
growth.8 Using this strategy, we demonstrated that structural diver- groups. For comparison purposes, PAMAM dendrimers and the
sities could be efficiently and rapidly harnessed at low generations. like, built around AB2 monomers, only reach these values at
Notably, distinct hydrophobic/hydrophilic and rigidity/flexibility the G(2) level. Dendrimer 5 was next treated with peracetylated
b-D-galactopyranosyl azide 613 under classical click reaction
1
conditions (CuSO4Á5H2O, Na-ascorbate in THF/H2O) to afford
Pharmaqam, Department of Chemistry, University of Quebec a Montreal, C.P. 8888,
octadecavalent galactodendrimer 7. The 1H-NMR spectrum showed
Succ. Centre-Ville, Montreal, Quebec, Canada H3C 3P8. E-mail: roy.rene@uqam.ca
the complete disappearance of the propargylic CRCH signals
at d 2.50 ppm and the expected appearance of two distinct
† Electronic supplementary information (ESI) available. See DOI: 10.1039/
c4cc06191h
13300 | Chem. Commun., 2014, 50, 13300--13303
This journal is ©The Royal Society of Chemistry 2014

View Article Online
Communication
ChemComm
Scheme 1 Divergent and convergent synthesis of octadecavalent galactodendrimer 8. Reagents and conditions: (i) NaH, Allyl bromide, DMF, 0 1C to rt,
5 h, 80%; (ii) cysteamineÁHCl, DMPAP, DMF, 365 nm, 3 h, 75%; (iii) EDC, DMPAP, DIPEA, DMF, 60 1C, o.n., 72%; (iv) CuSO4Á5H2O, Na ascorbate, THF/H2O
(1 : 1), 40 1C, 12 h, 81%; (v) MeONa/MeOH, rt, o.n., 88%; (vi) EDC, DMPAP, DMF, rt, o.n., 78%; (vii) CuSO4Á5H2O, Na ascorbate, THF/H2O (1 : 1), 40 1C, 5 h,
84%; (viii) Et3SiH, TFA, 0 1C, 3 h, DCM, 85%; (ix) AIBN, dioxane, 75 1C, 5 h, 53%.
triazole signals integrating in a 2 : 1 ratio at d 8.09 and 8.16 ppm. glycodendrimers without substantial loss of efficiency (5 steps and 24%
Another evidence of the monodispersity of the dendritic structure overall yield from 1 vs. 4 steps and 35% using the divergent method).
was further confirmed by gel permeation chromatography (GPC)
It is well established that the key factors for improving the overall
which showed a narrow and symmetrical Gaussian pattern with a avidity of glycodendritic architectures against bacterial and legumi-
´
PDI of 1.03. Subsequently, de-O-acetylation of 7 under Zemplen nous lectins through multivalent binding processes originate from: (1)
conditions (NaOMe, MeOH) provided the final glycodendrimer 8 the relative accessibility of the sugar ligands at the dendritic surfaces8
having 18 deprotected galactopyranoside moieties in quantitative and (2) the inner scaffold structures/valency themselves.7c,15,16 In
yield (a molecule having 72-OH groups)!
order to further our understanding and for rationalization of these
In order to illustrate the full versatility of this ‘‘onion peel’’ features using the above unique flexible ‘‘onion peel’’ template from
strategy for the rapid access to structurally diversified dendrimers, which galactopyranoside ligands with different aglycones emanated,
we also envisaged the construction of dendrimer 7 by a convergent glycodendrimers with longer penultimate spacers were next con-
approach. This alternative was initiated with S-trityl cysteamine 9 structed. Hence, for lectin’s better accessibility toward the sugar
prepared by a slight modification (ESI,† Scheme S1) of the literature ligands, longer branching residues and the choice of the peripheral
procedure.14 Under classical amidation conditions, 4 (EDC, DMPAP, sugars should constitute an improved design. Toward this goal,
DMF) provided intermediate 10 in 78% yield. A Cu-catalyzed click we synthesised both galactopyranoside and lactoside dendrimers
reaction was then performed in the presence of galactosyl azide 6 to using tetraethylene glycol (TEG) spacers (Scheme 2).
afford wedged glycodendron 11 in 84% yield. Once again, the
apparition of two discrete triazole singlets in the 1H NMR spectrum
with suitable integration (d 8.04 and 8.15 ppm; 2 : 1 ratio), coupled
with the disappearance of propargylic signals confirmed the triple
grafting of the sugar ligand. Chemoselective deprotection of the thiol
group using 5% TFA in the presence of Et3SiH as a cation scavenger
afforded dendron 12 in excellent yield (85%), without any trace of
disulfide side-products. The aromatic protons corresponding to the
trityl group at d 7.46–7.17 ppm completely disappeared. Notably, the
triplet corresponding to CH2 in the a-position of tritylated thiol 11 at
d 2.50 ppm shifted down-field at d 2.75 ppm. Final ligation of thiol
12 with hexakisallylated core 2 was achieved through a thiol–ene
coupling reaction (AIBN, dioxane, 75 1C, 5 h) to provide pure
glycodendrimer 7 in a 53% yield (30% yield under UV/DMPAP).
(b) reference compounds 18 and 19 and (c) a lactoside derivative immobilized
Scheme 2 (a) Syntheses of monomeric azido precursors 16 and 17,
Hence, we clearly demonstrated that the convergent sequence could
be applied toward the construction of functionalized ‘‘onion peel’’
on the chip for SPR studies. Reagents and conditions: (i) BF3ÁEt2O, DCM, 0 1C to
rt, 4 h, 55%; (ii) NaN3, DMF, 90 1C, o.n., 82%.
This journal is ©The Royal Society of Chemistry 2014
Chem. Commun., 2014, 50, 13300--13303 | 13301

View Article Online
ChemComm
Communication
Table 1 IC50 values of glycodendrimers and their monomeric analogs
derived from competitive inhibition SPR studies
Entry
Cpd
IC50 (mM)
R.p.a
R.p./sugarb
bb
Galactoside
1
2
18
8
43 Æ 1.5
1
195
1
11
11
0.22 Æ 0.02
TEG-Galactoside
3
4
19
23
21 Æ 1.5c
2
1162
2
65
32
0.037 Æ 0.005
TEG-Lactoside
5
6
20
25
958 Æ 34
4.2 Æ 0.4
0.05
10
0.05
0.6
11
a
b
Relative potency. Potency enhancement of individual sugars
c
throughout the same family. This value is consistent with the one
previously described for the tri(ethylene)glycol congener.21
This protein constitutes a virulence factor and is involved in the
pathogenesis of the bacteria in cystic fibrosis patients. To suitably
evaluate the beneficial presentation of the multivalent sugar ligands,
monomeric standards 18,19 19 and 20 corresponding to mimetics of
the peripheral saccharidic belt of each conjugate were synthesized.
To this end, CuAAc conditions were applied to glycosyl azides 6 and
16, and peracetylated derivative of 17, respectively, in the presence of
propargylic alcohol, followed by classical de-O-acetylation under the
Scheme 3 Synthesis of glycodendrimers 23 and 25. Reagents and
conditions: (i) CuSO4Á5H2O, Na ascorbate, THF/H2O, 40 1C, 12 h, 76%
22, 77% 24; (ii) NaOMe, DCM, MeOH, pH 9–10, rt, o.n., 90% 23, 86% 25.
´
Zemplen conditions (see the ESI† for protocols).
Treatment of galactopyranose pentaacetate 13 with mono-
For the competitive inhibition studies, the lactoside derivative
tosylated tetraethylene glycol 1417 under Lewis acid-catalyzed 2120 was immobilized onto a commercial SPR sensor chip (CM5)
conditions (BF3ÁEt2O in DCM) afforded compound 15 in 55% yield. following the manufacturer’s procedure. IC50 values (Table 1) were
Substitution of the tosylate in 15 by a terminal azide functionality determined using the pre-incubated mixtures of the PA-IL lectin
was readily accomplished using NaN3 in DMF to give 16 in 82% (1.5 mM) with increasing concentrations of monomers or glyco-
yield. Analogously, coupling of tosylated TEG derivative 14 onto its dendrimers used as analytes over the surface of CM5-bound 21.
peracetylated lactose homolog, followed by substitution with azide
The SPR experiments clearly demonstrated that glyco-
was performed as previously described,17 but better results were dendrimers 8, 23, and 25 exhibited much higher binding affinity
ultimately obtained through the per-benzoylated derivative 17, which compared to their corresponding monovalent derivatives 18, 19,
allowed easier purification and increased yields (see the ESI† for and 20 due to the ‘‘multivalent or glycoside cluster effect’’.22
the protocol). Both azido-terminated sugar ligands 16 and 17 were As expected, monomeric lactoside 20 represented a weaker
coupled onto scaffold 5 via CuAAc to afford glycodendrimers 22 ligand for PA-IL18 while the addition of a TEG linker to the
and 24 in 76–77% yields, which correspond to nearly quantitative galactoside moiety (19 vs. 18) allowed a 2-fold enhancement of
individual coupling (Scheme 3).
the affinity for the lectin. Thus, the additional glucoside residues
Unequivocally, both 1H and 13C NMR spectra indicated complete in lactosides have a detrimental effect which therefore cannot be
disappearance of propargylic signals and sugar incorporation with simply accounted for by a longer linker. Interestingly, galacto-
calculated relative integration. HRMS together with the presence of sylated dendrimer 8 exhibited low micromolar IC50 values
molecular ions and fragmentations corresponding to regular losses (0.22 mM), while most notably galactodendrimer 23 afforded
of carbohydrate moieties gave convincing proofs of the structural one of the best ligands known to date with an IC50 value of
´
integrity. Zemplen transesterification (NaOMe and MeOH) furnished 37 nM that compared well with the results obtained with multi-
two additional water soluble glycodendritic candidates 23 valent conjugates built around flexible or rigid scaffolds.23 These
(90%) and 25 (86%) for comparative inhibition experiments results unambiguously highlight the key role of linkers in the
with a bacterial lectin from Pseudomonas aeruginosa. Note that interactions with lectins, with a counter-balanced entropic cost
dendrimer 25 possesses 126 peripheral OH groups and thus, due to their flexibility. Additionally, tri-dimensional distribution
can serve on its own as an interesting precursor for further of terminal and optimized galactosides crucially contributed to
functionalization and applications.
high potencies since a substantial improvement (32-fold) was
In this context, the relative binding affinities of three novel observed for each ligand in 23, when compared to monomeric
glycodendrimers 8, 23, and 25 were evaluated by competitive reference 19, while weaker individual enhancements were obtained
surface plasmon resonance (SPR) using a galactoside specific with congested (8 vs. 18) or unoptimized (20 vs. 25) conjugates
bacterial lectin from the Gram-negative bacteria P. aeruginosa.6,18 (11-fold).
13302 | Chem. Commun., 2014, 50, 13300--13303
This journal is ©The Royal Society of Chemistry 2014

View Article Online
Communication
ChemComm
9 A. Dondoni and A. Marra, Chem. Soc. Rev., 2012, 41, 573–586; W. Chan,
B. Yu, C. E. Hoyle and A. B. Lowe, Chem. Commun., 2008, 4959–4961.
10 B. Neises and W. Steglich, Angew. Chem., Int. Ed. Engl., 1978, 17,
522–524.
11 H. C. Kolb, M. G. Finn and K. B. Sharpless, Angew. Chem., Int. Ed., 2001,
40, 2004–2021; P. Wu, A. K. Feldman, A. K. Nugent, C. J. Hawker,
In summary, we demonstrated that the structural diversity
in the construction of ‘‘onion peel’’ dendrimers, accessible via
both convergent and divergent routes, represents an additional
strategy for the build-up of dense surface groups at low dendrimer
generations. It also represents clear advantages over existing
approaches by providing versatile hypercore building blocks. More-
over, by not restricting layer-by-layer syntheses using identical
subunits, one can programme the physical/biophysical properties
of the dendrimers, as exemplified here using TEG residues. Of
particular interest in this instance, is the use of underexploited
dipentaerythritol as an A6 core molecule. In fact, work is now in
progress for further application of this useful building block as an
AB5 moiety. The work presented herein will undoubtedly be useful
to generate efficient and programmable multivalent antiadhesive
agents against bacterial infections.7a,24 Rationalization of the
preferential binding mode(s) together with determination of
the precise role of each structural parameter leading to high
avidity ligands such as in compound 23 are under investiga-
tion. Multivalent ‘‘onion peel’’ inhibitors harbouring optimized
sugar epitopes, notably containing aromatic residues, are also
presently under study. Further applications as antiadhesins
towards galectins,17 or as vectors for vaccines or drug targeting
nanomaterials25 are also under investigation.
´
A. Scheel, B. Voit, J. Pyun, J. M. J. Frechet, K. B. Sharpless and
V. V. Fokin, Angew. Chem., Int. Ed., 2004, 43, 3928–3932.
12 S. Zhang and Y. Zhao, Bioconjugate Chem., 2011, 22, 523–528.
13 R. Roy, F. D. Tropper, S. Cao and J. M. Kim, ACS Symp. Ser., 1997,
659, 163–180.
14 K. M. Halkes, A. Carvalho de Souza, C. E. P. Maljaars, G. J. Gerwig
and J. P. Kamerling, Eur. J. Org. Chem., 2005, 3650–3659.
15 Y. M. Chabre, A. Papadopoulos, A. Arnold and R. Roy, Beilstein
J. Org. Chem., 2014, 10, 1524–1535.
16 M. L. Talaga, N. Fan, A. L. Fueri, R. K. Brown, Y. M. Chabre,
P. Bandyopadhyay, R. Roy and T. K. Dam, Biochemistry, 2014, 53,
4445–4454.
17 (a) V. Percec, P. Leowanawat, H.-J. Sun, O. Kulikov, C. D. Nusbaum,
T. M. Tran, A. Bertin, D. A. Wilson, M. Peterca, S. Zhang, N. P. Kamat,
K. Vargo, D. Moock, E. D. Johnston, D. A. Hammer, D. J. Pochan,
´
Y. Chen, Y. M. Chabre, T. C. Shiao, M. Bergeron-Brlek, S. Andre,
R. Roy, H.-J. Gabius and P. A. Heiney, J. Am. Chem. Soc., 2013, 135,
9055–9077; (b) S. Zhang, R.-O. Moussodia, H.-J. Sun, P. Leowanawat,
A. Muncan, C. D. Nusbaum, K. M. Chelling, P. A. Heiney, M. L. Klein,
´
S. Andre, R. Roy, H.-J. Gabius and V. Percec, Angew. Chem., Int. Ed.,
2014, DOI: 10.1002/anie.201403186.
18 (a) J. Rodrigue, G. Ganne, B. Blanchard, C. Saucier, D. Giguere,
`
T. C. Shiao, A. Varrot, A. Imberty and R. Roy, Org. Biomol. Chem.,
2013, 11, 6906–6918; (b) A. Imberty, Y. M. Chabre and R. Roy,
Chem. – Eur. J., 2008, 14, 7490–7499; (c) J.-L. Reymond, M. Bergmann
and T. Darbre, Chem. Soc. Rev., 2013, 42, 4814–4822.
This work was supported by a discovery grant from the National
Science and Engineering Research Council of Canada (NSERC) and
by a Canadian Research Chair in Therapeutic Chemistry.
19 L. Harmand, S. Cadet, B. Kauffmann, L. Scarpantonio, P. Batat,
´
`
G. Jonusauskas, N. D. McClenaghan, D. Lastecoueres and J.-M. Vincent,
Angew. Chem., Int. Ed., 2012, 51, 7137–7141.
20 H. Tamiaki, A. Shinkai and Y. Kataoka, J. Photochem. Photobiol., A,
2009, 207, 115–125.
Notes and references
´
21 S. Cecioni, J.-P. Praly, S. E. Matthews, M. Wimmerova, A. Imberty
¨
1 (a) E. Buhleier, W. Wehner and F. Vogtle, Synthesis, 1978, 155–158;
and S. Vidal, Chem. – Eur. J., 2012, 18, 6250–6263.
(b) A.-M. Caminade, C.-O. Turrin and J.-P. Majoral, New J. Chem., 22 (a) Y. C. Lee and R. T. Lee, Acc. Chem. Res., 1995, 28, 321–327;
2010, 34, 1512–1524; (c) S. Svenson and D. A. Tomalia, Adv. Drug
Delivery Rev., 2005, 57, 2106–2129; (d) M. A. Mintzer and
(b) R. Roy, Curr. Opin. Struct. Biol., 1996, 6, 692–702; (c) J. J. Lundquist
and E. J. Toone, Chem. Rev., 2002, 102, 555–578.
M. W. Grinstaff, Chem. Soc. Rev., 2011, 40, 173–190; (e) M. Sowinska 23 (a) S. Cecioni, V. Oerthel, J. Iehl, M. Holler, D. Goyard, J.-P. Praly,
and Z. Urbanczyk-Lipkowska, New J. Chem., 2014, 38, 2168–2203.
2 (a) D. A. Tomalia, H. Baker, J. Dewald, M. Hall, G. Kallos, S. Martin,
J. Roeck, J. Ryder and P. Smith, Polym. J., 1985, 17, 117–132; (b) N. Launay,
A.-M. Caminade and J.-P. Majoral, J. Am. Chem. Soc., 1995, 117,
3282–3283.
A. Imberty, J.-F. Nierengarten and S. Vidal, Chem. – Eur. J., 2011, 17,
3252–3261; (b) S. Cecioni, S. Faure, U. Darbost, I. Bonnamour,
´
H. Parrot-Lopez, O. Roy, C. Taillefumier, M. Wimmerova, J.-P. Praly,
A. Imberty and S. Vidal, Chem. – Eur. J., 2011, 17, 2146–2159.
24 (a) A. Bernardi, J. Jimenez-Barbero, A. Casnati, C. De Castro, T. Darbre,
F. Fieschi, J. Finne, H. Funken, K.-E. Jaeger, M. Lahmann,
T. K. Lindhorst, M. Marradi, P. Messner, A. Molinaro, P. V. Murphy,
C. Nativi, S. Oscarson, S. Penades, F. Peri, R. J. Pieters, O. Renaudet,
J.-L. Reymond, B. Richichi, J. Rojo, F. Sansone, C. Schaffer, W. B.
Turnbull, T. Velasco-Torrijos, S. Vidal, S. Vincent, T. Wennekes,
H. Zuilhof and A. Imberty, Chem. Soc. Rev., 2013, 42, 4709–4727;
(b) N. Berthet, B. Thomas, I. Bossu, E. Dufour, E. Gillon, J. Garcia,
N. Spinelli, A. Imberty, P. Dumy and O. Renaudet, Bioconjugate Chem.,
2013, 24, 1598–1611.
´
3 C. J. Hawker and J. M. J. Frechet, J. Am. Chem. Soc., 1990, 112,
7638–7647.
4 F. Zeng and S. C. Zimmerman, J. Am. Chem. Soc., 1996, 118, 5326–5327.
5 K. L. Killops, L. M. Campos and C. J. Hawker, J. Am. Chem. Soc.,
2008, 130, 5062–5064.
6 (a) N. Kottari, Y. M. Chabre, T. C. Shiao, R. Rej and R. Roy, Chem.
´
Commun., 2014, 50, 1983–1985; (b) S. Chatani, M. Podgorski,
C. Wang and C. N. Bowman, Macromolecules, 2014, 47, 4894–4900,
DOI: 10.1021/ma501418r.
´
7 (a) Y. M. Chabre and R. Roy, Adv. Carbohydr. Chem. Biochem., 2010, 25 (a) S. Andre, P. J. C. Ortega, M. A. Perez, R. Roy and H.-J. Gabius,
`
´
63, 165–393; (b) Y. M. Chabre and R. Roy, Curr. Top. Med. Chem.,
2008, 8, 1237–1285; (c) Y. M. Chabre and R. Roy, Chem. Soc. Rev.,
2013, 42, 4657–4708; (d) O. Renaudet and R. Roy, Chem. Soc. Rev.,
2013, 42, 4515–4517; (e) M. Gingras, Y. M. Chabre, M. Roy and
R. Roy, Chem. Soc. Rev., 2013, 42, 4823–4841.
Glycobiology, 1999, 9, 1253–1261; (b) D. Giguere, S. Andre,
M. A. Bonin, M. A. Bellefleur, A. Provencal, P. Cloutier, B. Pucci,
R. Roy and H. J. Gabius, Bioorg. Med. Chem., 2011, 19, 3280–3287;
´
(c) S. Andre, B. Liu, H.-J. Gabius and R. Roy, Org. Biomol. Chem.,
´
2003, 1, 3909–3916; (d) N. Ahmad, H.-J. Gabius, S. Andre, H. Kaltner,
8 R. Sharma, K. Naresh, Y. M. Chabre, R. Rej, N. K. Saadeh and R. Roy,
Polym. Chem., 2014, 5, 4321–4331.
S. Sabesan, R. Roy, B. Liu, F. Macaluso and C. F. J. Brewer, Biol.
Chem., 2004, 279, 10841–10847.
This journal is ©The Royal Society of Chemistry 2014
Chem. Commun., 2014, 50, 13300--13303 | 13303