The PD-0332991 cell line latter may serve to “fine tune” their actions in vivo (Schwartz-Albiez, 2012). These results also indicate that this IgG class-dependent cross-reactivity can be reduced by the introduction of PEGs, and this is considered important for the accurate detection of analytes in particular in an artificial array system as the SGA. In order to determine the contribution of non-target binding we assayed P1-, PEG23-, and PEG60-P1 modified beads with fetal calf serum-derived and presumably heterophilic antibodies. The results (Fig. 6B) demonstrate that no substantial binding to all three types of beads was observed for IgM (MFI of around

20) whereas some binding (MFI of around 500) was detected for IgG for the regular P1-beads. This observation is in concordance to the previous experiments (Figs. 5B

and 6A). These IgG signals were reduced for the PEG60-P1 beads to about 100 MFI and for the PEG23-P1 beads to 15 MFI. In summary, unspecific binding was observed almost exclusively when IgGs, but not IgMs were applied as glycan-binding antibodies or as secondary detection antibodies. These data seem to be consistent with existing evidence regarding anti-glycan antibodies. It is known that naturally occurring anti-glycan antibodies are predominantly of IgM class and are produced by CD5 positive B1 cells expressing a distinct pattern of surface markers, but not conventional B cells (Viau and Zouali, 2005, Vollmers and Brandlein, 2009, Griffin et al., 2011 and Bovin, 2013). Despite their polyreactive nature anti-glycan IgMs appear to be highly specific in terms of affinity distinctions. Specific recognition of certain glycan structure strongly GSK1120212 in vitro depends on its natural molecular context (Bovin, 2013). Pentameric IgMs have ten Fab regions and therefore possess a theoretical valency of 10. Multivalent recognition is very important for glycan–protein interaction, providing stable and affine binding to multiple oligosaccharide structures. On the contrary, IgGs are only divalent, their interactions with glycans

may be weaker that is why this antibody class is typically not ascribed to recognize glycans in nature. Due to the same reasons IgGs may be more predisposed to unspecific binding than IgMs upon profiling with glycoarrays. To further exploit the possibility to reduce anti-glycan Parvulin antibody cross-reactivity by using heterobifunctional PEGs, we linked PEG23 and PEG60 to the bead surface, coupled Pk trisaccharide to these beads, and compared the binding of monoclonal human anti-P1 IgM either to P1-coupled beads or to Pk-coupled beads (without or with heterofunctional PEGs) as a function of the antibody dilution (Fig. 7). The results showed that the binding of the anti-P1 IgM antibodies, regardless of the dilution, to Pk-beads was several-fold lower than to P1-beads, indicating that indeed anti-P1 antibodies bind to Pk trisaccharide with much lower affinity than P1 trisaccharide.