Polymer-tethered glycosylated gold nanoparticles recruit sialylated glycoproteins into their protein corona, leading to off-target lectin binding

Abstract

Upon exposure to biological fluids, the fouling of nanomaterial surfaces results in non-specific capture of proteins, which is particularly important when in contact with blood for in vivo and ex vivo applications. It is crucial to evaluate not just the protein components but also the glycans attached to those proteins. Polymer-tethered glycosylated gold nanoparticles have shown promise for use in biosensing/diagnostics, but the impact of the glycoprotein corona has not been established. Here we investigate how polymer-tethered glycosylated gold nanoparticles interact with serum proteins and demonstrate that the protein corona introduces new glycans and hence off-specific targeting capability. Using a panel of RAFT-derived polymers grafted to the gold surface, we show that the extent of corona formation is not dependent on the type of polymer. In lectin-binding assays, a glycan (galactose) installed on the chain-end of the polymer was available for binding even after protein corona formation. However, using sialic-acid binding lectins, it was found that there was significant off-target binding due to the large density of sialic acids introduced in the corona, confirmed by western blotting. To demonstrate the importance, we show that the nanoparticles can bind Siglec-2, an immune-relevant lectin post-corona formation. Pre-coating with (non-glycosylated) bovine serum albumin led to a significant reduction in the total glycoprotein corona. However, sufficient sialic acids were still present in the residual corona to lead to off-target binding. These results demonstrate the importance of the glycans when considering the protein corona and how 'retention of the desired function' does not rule out 'installation of undesired function' when considering the performance of glyco-nanomaterials.

Fig. 1. Synthesis of polymer-tethered gold nanoparticles. (A) Polymers synthesised in this study. Subscript denotes the targeted degree of polymerisation; (B) size exclusion chromatography of the polymer library; (C) dynamic light scattering of citrate AuNPs (40 nm) and with polymers tethered to their surface.

Fig. 2. SDS-PAGE of protein corona formation on polymer-coated gold nanoparticles. (A) Silver-stained gel showing hard corona proteins released from nanoparticles after incubation with bovine plasma. (B) Densitometry analysis of gel. Polymer codes refer to polymer coatings (Fig. 1) on 40 nm gold particles.

Fig. 3. Glyconanoparticle synthesis. (A) Synthetic scheme for the synthesis of PFP-terminated PHEA, functionalisation with galactosamine and immobilisation onto gold nanoparticles; (B) size exclusion chromatography analysis of PHEA's; (C) dynamic light scattering of polymer-coated particles in buffer and in plasma; (D) representative TEM of polymer-coated nanoparticles. AuNPs = 40 nm in all cases.

Fig. 4. Effect of hard corona on lectin binding capacity to Gal-PHEA₂₅@AuNPs. (A) Schematic of aggregation assay in buffer only; (B) UV-Vis spectra verses SBA in buffer; (C) UV-vis spectra verses WGA in buffer; (D) schematic of hard corona formation and aggregation assay; (E) UV-Vis spectra verses SBA with hard corona; (F) UV-Vis spectra verses WGA with hard corona.

Fig. 5. Impact of corona on lectin binding using biolayer interferometry (BLI). (A) Schematic of buffer-only lectin binding; (B) glyco-nanoparticles binding to SBA in buffer only; (C) glyco-nanoparticles binding to WGA in buffer only; (D) schematic of hard-corona coated particles lectin binding; (E) glyco-nanoparticles binding to SBA with hard corona; (F) glyco-nanoparticles binding to WGA with hard corona.

Fig. 6. Impact of corona on sialic acid binding lectins, using BLI. (A) Glyco-nanoparticles verses MAL-II in buffer alone; (B) glyco-nanoparticles with hard corona verses MAL-II; (C) glyco-nanoparticles verses Siglec-2 in buffer alone; (D) glyco-nanoparticles with hard corona verses Siglec-2.

Fig. 7. SDS-PAGE of protein corona formation on Gal-PHEA_n-coated (n = 25, 50, 75) nanoparticles with and without BSA blocking. (A) Silver-stained gel showing hard corona proteins released from nanoparticles; (B) densitometry analysis of gel shown as a heat map. Polymer codes refer to polymer coatings (Fig. 2) on 40 nm gold particles.

Fig. 8. Western blot analysis of sialic acid contents of the glycoproteins corona after exposure of nanoparticles to bovine plasma. (A) Western blot using SiaFind Pan-Specific Lectenz, with darker regions indicating more sialic acids; (B) densitometry analysis presented as heat map showing change in total sialic acid recruitment to the particles with/out blocking.

Fig. 9. Differential centrifugation sedimentation analysis of particles, in buffer, plasma, BSA, or BSA then plasma. (A) Gal-PHEA₂₅@AuNP₄₀; (B) Gal-PHEA₅₀@AuNP₄₀; (C) Gal-PHEA₇₅@AuNP₄₀.

Fig. 10. Lectin binding by aggregation assay against nanoparticles pre-blocked with BSA, before exposure to plasma. (A) Schematic of experimental process; (B) Gal-PHEA₂₅@AuNP₄₀ verses SBA; (C) Gal-PHEA₂₅@AuNP₄₀ verses WGA; (D) Gal-PHEA₅₀@AuNP₄₀ verses SBA; (E) Gal-PHEA₅₀@AuNP₄₀ verses WGA; (F) Gal-PHEA₇₅@AuNP₄₀ verses SBA; (G) Gal-PHEA₇₅@AuNP₄₀ verses WGA.