The

following dilutions of affinity find protocol purified antibodies were used: anti-P1 — 1/10 and 1/50; and anti-Btri — 1/100, 1/200 and 1/500. All the other antibodies were used in the dilution of 1/10. Additional depletion of ascites fluid after removal of anti-P1 antibodies was performed as follows. After affinity purification of anti-P1 antibodies, ascites fluid was additionally incubated with P1-adsorbent, taken in a 1/1 ratio (v/v), for 1 h on a shaker at RT, then centrifuged for 10 min, at 13,000 g. The supernatant (diluted to 1:10 with PBS, 1% BSA) was assayed with P1-regular and PEG-modified beads. Statistical analysis was performed with GraphPad Prism 6 software using the repeated-measures ANOVA followed by

the Tukey posttest. Adjusted p-values < 0.05 were considered statistically significant. Expanding on the strategy of PEG linking we designed and investigated several kinds of PEG-modifications in order to mask sites on the glycobeads at which unspecific binding of antibodies may occur: partial PEG substitutions with PEGs of different lengths within end-biotinylated glycopolymers (Scheme 1B); attachment of biotin-modified PEGs to presumably PI3K inhibitor unsaturated streptavidin binding sites next to coupling of end-biotinylated glycopolymers (Scheme 1C); covalent binding of amino-functionalized biotin-PEGs (heterobifunctional PEGs) directly onto the bead surface prior to glycopolymer coupling (Scheme 1D). Glycoconjugates based on linear polyacrylamides (PAAs) with side-attached carbohydrate groups are widely used in bioanalytical research as multivalent glycoprobes (Bovin, 1998 and Bovin, 2003). However, serum antibodies may bind not only to the pendant glycan residues but also to the polymer backbone. The latter effect can be reduced by the substitution of the side groups (e.g. N-(2-hydroxyethyl)) within the non-glycosylated monomer units by PEG. Three types of PEGs were used for this purpose: “short” (m = 4, substitution rate − 80%), “medium” or “long” (m ~ 50 or 280, substitution rate − 5%, see Glycopolymers with end-biotin group and Fig. 1).

Fossariinae In addition, two different glycopolymers were included: Btri belongs to the ABO blood group system and served as a “reference glycan”. P1 trisaccharide is our top candidate as potential ovarian cancer marker ( Pochechueva et al., 2011b and Jacob et al., 2012). The binding of corresponding affinity purified antibodies and healthy donor plasma antibodies (analytes) to these different PEGs with our regular (Scheme 1A) Btri- and P1-glycoprobes was compared with SGA. The results showed that the MFI values, representative for antibody binding, were lower for all three PEGylated compared to the regular glycopolymers. This was true for the Btri (Fig. 3A) as well as the P1 (Fig. 3B) glycopolymers. Even more interesting, the binding of the antibodies to both PEGylated glycopolymers, i.e.