Biomimetic Silica Nanoparticles Prepared by a Combination of Solid-Phase Imprinting and Ostwald Ripening

Abstract

Herein we describe the preparation of molecularly imprinted silica nanoparticles by Ostwald ripening in the presence of molecular templates immobilised on glass beads (the solid-phase). To achieve this, a seed material (12 nm diameter silica nanoparticles) was incubated in phosphate buffer in the presence of the solid-phase. Phosphate ions act as a catalyst in the ripening process which is driven by differences in surface energy between particles of different size, leading to the preferential growth of larger particles. Material deposited in the vicinity of template molecules results in the formation of sol-gel molecular imprints after around 2 hours. Selective washing and elution allows the higher affinity nanoparticles to be isolated. Unlike other strategies commonly used to prepare imprinted silica nanoparticles this approach is extremely simple in nature and can be performed under physiological conditions, making it suitable for imprinting whole proteins and other biomacromolecules in their native conformations. We have demonstrated the generic nature of this method by preparing imprinted silica nanoparticles against targets of varying molecular mass (melamine, vancomycin and trypsin). Binding to the imprinted particles was demonstrated in an immunoassay (ELISA) format in buffer and complex media (milk or blood plasma) with sub-nM detection ability.

Figure 1

Schematic representation of the preparation of surface molecularly imprinted silica nanoparticles by the process of Ostwald ripening in the presence of an immobilised template.

Figure 2

(a) Change in diameter of 12 nm silica nanoparticles incubated in 50 mM phosphate buffer, (pH 7.2) with time, according to dynamic light scattering (DLS) measurements. TEM images of silica nanoparticles (b) before and (c) after solid-phase imprinting of melamine by exposure to phosphate buffer for 2 hours.

Figure 3

Competitive binding between free analyte and the corresponding HRP conjugates in competitive ELISA format using immobilized silica NPs in 96 well polystyrene microplates. In (a,c and e) only the linear ranges are depicted, no response was observed for lower analyte concentrations whilst saturation is observed at higher concentrations. (a) melamine with melamine-imprinted silica NPs; (b) melamine with non-imprinted silica NPs; (c) vancomycin with vancomycin-imprinted silica NPs; (d) vancomycin with non-imprinted silica NPs; (e) trypsin with trypsin-imprinted silica NPs; (f) trypsin with non-imprinted silica NPs. All experiments were performed in 50 mM sodium-phosphate buffer, pH 7.0. Error bars represent ±1 standard deviation and are for experiments performed in triplicate.

Figure 4

Cross-reactivity assays. Competitive ELISA between free analyte and HRP conjugates corresponding to the templated species, using immobilised silica nanoparticles in 96 well polystyrene microplates. (a) Desisopropyl atrazine with melamine-imprinted silica NPs; (b) Teicoplanin with vancomycin-imprinted NPs (c) Lysozyme with trypsin-imprinted silica NPs. All experiments were performed in 50 mM sodium phosphate buffer, pH 7.0. Error bars represent ±1 standard deviation and are for experiments performed in triplicate.

Figure 5

Assays in complex sample matrices spiked with the respective templates. Competitive binding between free analyte and corresponding HRP conjugates in ELISA format using immobilised silica nanoparticles in 96 well polystyrene microplates. In all cases the linear ranges are depicted, no response was observed for lower analyte concentrations whilst saturation is observed at higher concentrations. (a) Melamine binding to melamine-imprinted silica NPs in milk; (b) vancomycin binding to vancomycin-imprinted NPs in blood serum; (c) trypsin binding to trypsin-imprinted silica NPs in blood serum. Melamine experiments was performed in 0.5% v/v milk solution and vancomycin and trypsin in 2.5% v/v blood serum solution, in all cases diluted in 50 mM sodium-phosphate buffer, pH 7.0. Error bars represent ±1 standard deviation and are for experiments performed in triplicate.