5
082
K. H. Schlick et al. / Bioorg. Med. Chem. Lett. 21 (2011) 5078–5083
times greater than that of methyl
inhibition potency that was 8900 times greater than methyl
-mannoside (see Fig. S1, Supplementary data). The fact that
the inhibition activity with Con A is low for both 12 and methyl
-mannoside relative to the glycodendrimer strongly suggests
that the tris-mannoside unit is not a significantly more effective
a
-
D
-mannoside, while 9 had an
to the surface. Larger dendrimers may only be needed when
additional functionality is desired, such as described by Figure 2c.
In all generations, the dendrimers with the clustered display of
ligands were as effective as the dendrimers with monomeric
ligands distributed across the endgroups.
a
-D
a-D
In conclusion, tris-mannoside clusters have been synthesized
using click chemistry and have been appended to G(3) through
G(6)-PAMAM dendrimers. Varying amounts of tris-mannosides
were used in conjunction with a solubilizing ethoxyethanol unit
to form a series of glycodendrimers. ELISA inhibition experiments
suggest that tris-clustered mannose residues can be used to replace
individual sugars on the dendrimer without loss of function, freeing
up a significant number of the endgroups on the dendrimers. Since
tris-mannoside clustering allows for a redistribution of the dendri-
mers’ surface functionalities, from this chemistry one can obtain
patterned dendrimers that incorporate solubilizing groups (as de-
scribed here), or groups to alter immunogenicity, imaging agents,
prodrugs, and targeting ligands all on the same dendrimer (as de-
picted schematically inFig. 2c). The present study suggests an effec-
tive way to cluster the targeting carbohydrate groups so that
additional groups can be added without loss of binding specificity
or selectivity for the targeted lectin-carbohydrate interaction.
monovalent ligand for Con A than methyl
tails provided in the Supplementary data).
a-D-mannoside is (de-
When IC50 values are examined on a per mannoside basis, all of
the generation three dendrimers (4a–4d and 8) are comparable
inhibitors. On a per dendrimer basis, 4a and 4b have IC50 values
comparable to the value for 8. These three dendrimers also have
approximately the same number of mannosides. Dendrimer 8 is
only 2.6-fold more effective than dendrimer 4c even though 4c
has nine sugars (3.2-fold less) while 8 has 29 sugars on average.
This suggests that, even for small generation dendrimers, a pat-
terned arrangement of ligands on the dendrimer may allow systems
to be as effective in binding events as dendrimers that have the
same number of ligands displayed in an unpatterned presentation
using far more of the dendrimer endgroups. Dendrimer 4d with
only one tris-mannoside on average is a surprisingly effective inhib-
itor, and we cannot rule out the possibility that this activity may be
caused by a small subset of over-functionalized glycodendrimers.
In the generation four series of dendrimers, all of the compounds
again have comparable IC50 values when evaluated on a per manno-
side basis, although dendrimers 5b and 5c, which are in the middle
of the series in terms of amount of mannosides, have slightly lower
IC50 values than those of 9. Dendrimers 5a, 5d, 5e, 5f are all slightly
less effective inhibitors (on a per mannose basis) than 9. On a per
dendrimer basis, 5a, 5b, 5c, and 9 are all comparable. Dendrimer
Acknowledgments
This paper is dedicated to Carolyn R. Bertozzi on the occasion of
her receipt of the Tetrahedron Young Investigator Award. This re-
search was supported by NIH R01 GM62444. J.J.W. was supported
by an undergraduate scholarship from the Beckman Foundation,
and both J.J.W. and M.S.K. were supported by undergraduate schol-
arships from NIH P20 RR016455.
5
d is only 3-fold weaker than 9, 5e is only 6.1-fold weaker than 9,
and 5f is 12.8-fold weaker than 9 (on a per dendrimer basis). The
dramatically reduced number of mannosides that are required for
inhibition (27, 15, and 9 for 5d, 5e, and 5f, respectively compared
to 55 sugars for 9) suggest that patterned architectures can be use-
ful for processes involving higher generation dendrimers. For exam-
ple, 5d shows good inhibition but only nine of the PAMAM amino
groups are functionalized with tris-mannoside, leaving 55 more
amino termini available for functionalization with another group.
In the generation five series of dendrimers, all dendrimers 6a–
Supplementary data
References and notes
1. Astruc, D.; Boisselier, E.; Omelas, C. Chem. Rev. 2010, 110, 1857.
2
3
.
.
Andronov, A.; Frechet, J. M. J. Chem. Commun. 2000, 1701.
Nantalaksakul, A.; Reddy, D. R.; Bardeen, C. J.; Thayumanavan, S. Photosynth.
Res. 2006, 87, 133.
6
c and 10 have comparable IC50 values on a per mannoside basis.
Dendrimer 6a is 2-fold weaker (on a per mannose basis) but has
twice as many sugars as the other dendrimers and is almost cer-
tainly over-functionalized (meaning that some tris-mannosides
may be sterically inaccessible to Con A binding sites once some
Con A/dendrimer interactions have occurred). On a per dendrimer
basis, 6a–6c and 10 all have comparable IC50 values. Thus, adding
the extra 29 tris-mannosides (87 mannosides) to convert 6b to 6a
is not obviously advantageous.
4. Caminade, A.-M.; Delavaux-Nicot, B.; Laurent, R.; Majoral, J.-P. Curr. Org. Chem.
010, 14, 500.
2
5
6
.
.
de Jesus, E.; Flores, J. C. Ind. Eng. Chem. Res. 2008, 47, 7968.
Berger, A.; Gebbink, R. J. M. K.; van Koten, G. Top. Organomet. Chem. 2006, 20, 1
(Dendrimer Catalysis).
7. Dutta, T.; Jain, N. K.; McMillan, N. A. J.; Parekh, H. S. Nanomed. Nanotech. Biol.
Med. 2010, 6, 25.
8
9
.
.
Villaraza, A. J. L.; Bumb, A.; Brechbiel, M. W. Chem. Rev. 2010, 110, 2921.
Heegaard, P. M. H.; Boas, U.; Sorensen, N. S. Bioconjugate Chem. 2010, 21, 405.
In the generation six series, all dendrimers 7a–7d and 11 have
comparable IC50 values on a per mannose basis, and all but 7d have
comparable IC50 values on a per dendrimer basis. Even 7d with nine
carbohydrate clusters (27 mannosides) has only a 4-fold reduction
in effectiveness as an inhibitor of the Con A/surface interaction rel-
ative to 11. Reducing the number of clusters from 66 in 7a (198 sug-
ars) to nine in 7d (27 sugars) causes only a 3.5-fold reduction in the
IC50 value. Thus, the results from the sixth generation dendrimer
series also suggest that dendrimers that are decorated with sugars
in a clustered display, leaving many endgroups available for func-
tionalization with other groups, are as effective as dendrimers
where all the endgroups have been distributed across the dendri-
mer surface. Very little, if any, activity is lost when the carbohy-
drates are clustered.
10. Gupta, U.; Agashe, H. B.; Asthana, A.; Jain, N. K. Nanomed. Nanotech. Biol. Med.
006, 2, 66.
2
1
1. Kannan, R. M.; Perumal, O. P.; Kannan, S. Dendrimers and Hyperbranched
Polymers for Drug Delivery. In Biomedical applications of nanotechnology;
Labhasetwar, V., Leslie-Pelecky, D. L., Eds.; Wiley: Hoboken, 2007. p 105.
2. Svenson, S.; Tomalia, D. A. Dendrimers as nanoparticulate drug carriers. In
Nanoparticulates as drug carriers; Torchilin, V. P., Ed.; Imperial College Press:
London, 2006. p 277.
1
13. Tekade, R.; Kumar, P. V.; Jain, N. K. Chem. Rev. 2009, 109, 49.
1
1
1
1
4. Jevprasesphant, R.; Penny, J.; Jalal, R.; Attwood, D.; McKeown, N. B.;
D’Emanuele, A. Int. J. Pharm. 2003, 252, 263.
6. Swanson, S. D.; Kukowska-Latallo, J. F.; Patri, A. K.; Chen, C.; Ge, S.; Cao, Z.;
Kotlyar, A.; East, A. T.; Baker, J. R. Int. J. Nanomed. 2008, 3, 201.
7. Patri, A. K.; Kukowska-Latallo, J. F.; Baker, J. R. Adv. Drug Delivery Rev. 2005, 57,
2203.
18. Lee, C. C.; Gillies, E. R.; Fox, M. E.; Guillaudeu, S. J.; Frechet, J. M. J.; Dy, E. E.;
Szoka, F. C. Proc. Natl. Acad. Sci. U.S.A. 2006, 103, 16649.
Overall, IC50 values were similar for all generations of dendri-
mers that were studied. This suggests that smaller dendrimers were
as effective as larger dendrimers in inhibiting the binding of Con A
1
2
9. Goodwin, A. P.; Lam, S. S.; Frechet, J. M. J. J. Am. Chem. Soc. 2007, 129, 6994.
0. Antoni, P.; Hed, Y.; Nordberg, A.; Nystrom, D.; von Holst, H.; Hult, A.; Malkoch,
M. Angew. Chem., Int. Ed. 2009, 48, 2126.