05) for all analytes. These results indicate that (i) PEGylation reduces antibody binding to these glycopolymers and that (ii) this decrease is PEG chain length-dependent. This observation can unambiguously be explained by the shielding of the glycan residues by the PEG molecules, which is stronger with longer PEG chains attached to the polymer backbone. However, this shielding effect is likely to affect specific binding and presumably also unspecific binding of the antibodies to these DAPT glycopolymers, and the distinction whether or not unspecific binding of

antibodies occurred to non-glycosylated parts of polyacrylamide backbone was not possible with these kinds of PEGylations. To determine the potential contribution of unspecific binding we assayed the beads modified with the regular or with PEGylated P1-glycoprobes with native ascites fluid and with ascites fluid depleted of anti-P1 antibodies. As expected the results showed (ESM, Fig. 1) substantially lower MFI values in the depleted than in native ascites setting. More importantly, antibody binding decreased with the length of the PEGs in both settings, comparable to the setting for affinity purified Everolimus and plasma anti-glycan antibodies presented in Fig. 3B. The finding that this PEG chain length-dependent decrease in binding occurred

in both settings, i.e. also in the native ascites, indicates that these types of PEGs (different chain lengths) were not sufficient to avoid unspecific antibody binding. The next PEG modification considered was the attachment of biotinylated PEGs (biot-PEG50

and biot-PEG280) Rebamipide to glycopolymer pre-treated beads (see PEGs used for glycopolymer and microbead modifications and Fig. 2A). The idea was that these biot-PEGs may bind streptavidin binding sites that may have been left unbound after the antecedent coupling of the glycopolymers. We assayed the binding of the analytes, i.e. three different human antibodies (commercial anti-P1 monoclonal IgM antibody, affinity purified anti-P1 antibodies, and plasma antibodies) to regular and biot-PEGm-modified P1-conjugated beads. Fig. 4A demonstrates that the MFI values for the regular and the two biot-PEGm-modified P1-conjugated beads were comparable for each of the analytes (differences in MFI values among three types of beads were less than the inter-assay variability (from 8.5 to 18.5%) previously described (Pochechueva et al., 2011a)). This result indicates that the attachment of these two biot-PEGm did not affect the binding of anti-glycan antibodies to P1-beads. Possible explanations are that either all streptavidin binding sites were saturated with biotinylated glycopolymer prior to biot-PEGm coupling or the influence of non-target binding to streptavidin was negligible.