panel of related smart dendritic systems has recently been widened to generate promising
glycosylated candidates as imaging agents or vaccines for instance.
In this context, Fukase and colleagues described the multivalent presentation of Nglycans containing or lacking sialic acid termini scaffolded around a dendritic poly-Llysine backbone for the development of in vivo noninvasive imaging devices [32].
Considering the critical role of glycan residues in cell–cell recognition, adhesion, and
signal transduction [33] and their implication as anti-inflammatory agents of immunoglobulin G (IgG) through Siglec interactions [34] with antibody-dependent cellular
toxicity (ADCC) and/or complement-dependent cytotoxicity (CDC), the dynamics of
these multivalent glycans have been systematically studied in vivo. The main goal was to
achieve promising sensors with optimized structures and generating organ-specific
accumulation. To this end, different generations of polyvalent polylysine-based
glycodendrons have been generated by solid-phase synthesis to investigate the cluster
effect. The architectures thus consisted in the mixed presentation of four (1a–e), eight
(2a–e) and sixteen (3a–e) peripheral N-glycan derivatives, attached via CuAAc (coppercatalyzed azide-alkyne cycloaddition or “click chemistry”) and terminal benzylated
histidine. The presence of a focal lysine ε-amino group allowed the subsequent
incorporation of
68
Ga-cryptand (
68
Ga-DOTA) as positron emission tomography (PET)
radiolabel or of Cy5 and NBD as fluorophores through straightforward 6π-azaelectrocyclization (Fig. 3). The administration of the glycoclusters labeled with
68
Ga-DOTA
(1a–3e
68
Ga-DOTA) to the tail vein of BALB/c nude mice allowed determination of the
distribution of the glycosylated derivatives after a whole-body PET scan over 4 h.
Remarkably, the in vivo dynamics and biodistribution appreciably differed from the
different generations of clusters containing the bis-Neuα(2,6)Gal termini (1a–3a).
Prolonged in vivo lifetimes and suitable biodistributions were observed for the larger
derivative 3a, suggesting its plausible efficient uptake by hepatocytes through the
galactose/N-acetylgalactosamine (Gal/GalNAc) lectin receptor.
These observations highlighted the critical role of molecular size, hydrophobic/
hydrophilic balance, and multivalency effects on the in vivo dynamics. While 4- and 8mers (1a and 2a, respectively) were rapidly cleared through the kidney, the larger
congener 3a was retained in the body after 4 h, with a substantial radioactivity detected
in the liver, gallbladder, and blood. Importantly, the 16-mer homologue lacking the
terminal sialic acid residue (3b) or presenting the bis-Neuα(2,3)Gal glycan (3c), were
more rapidly cleared through the kidney, although 3b accumulated in the liver due to
the presence of asialoglycoprotein receptors in this organ. On the other hand, the
variation of the specific sialoside linkage to galactose, i.e., the Neuα(2,6)Gal linkage,
was of a crucial importance in the circulatory residence of N-glycans resulting in the
uptake of 3a in the liver, also avoiding excretion by the biofiltration pathway in the
kidney, as observed for 3c.
A biodistribution study was performed on 16-mer homologues (3a–3e Cy5) using
fluorescence imaging. The importance of at least one specific Neuα(2,6)Gal epitope
was reinforced, with higher in vivo stability observed for Cy5-labeled homogeneous
3a, or heterogeneous 3d and 3e, with fluorescence accumulated both in the liver and
the spleen after 4 h. Considering that the spleen could be involved in the immune
system or in the production of antigen-specific antibodies by the interaction of T cells
302
N. Kottari et al.
Précédent

- 306/349

Suivant