Comparison between polyethylene glycol and zwitterionic polymers as antifouling coatings on wearable devices for...

Selective capture of disease-related proteins in complex biological fluids and tissues is an important aim in developing sensitive protein biosensors for in vivo applications. Microprojection arrays are biomedical devices whose mechanical and chemical properties can be tuned to allow efficient penetration of skin, coupled with highly selective biomarker capture from the complex biological environment of skin tissue. Herein, the authors describe an improved surface modification strategy to produce amine-modified polycarbonate arrays, followed by the attachment of an antifouling poly(sulfobetaine-methacrylate) (pSBMA) polymer or a linear polyethylene glycol (PEG) polymer of comparative molecular weight and hydrodynamic radius. Using a "grafting to" approach, pSBMA and linear PEG coatings yielded comparative antifouling behavior in single protein solutions, diluted plasma, or when applied to mouse flank skin penetrating into the vascularized dermal tissue. Interestingly, the density of immobilized immunoglobulin G (IgG) or bovine serum albumin protein on pSBMA surfaces was significantly higher than that on the PEG surfaces, while the nonspecific adsorption was comparable for each protein. When incubated in buffer or plasma solutions containing dengue non-structural protein 1 (NS1), anti-NS1-IgG-coated pSBMA surfaces captured significantly more NS1 in comparison to PEG-coated devices. Similarly, when wearable microprojection arrays were applied to the skin of dengue-infected mice using the same coatings, the pSBMA-coated devices showed significantly higher capture efficiency (>2-fold increase in signal) than the PEG-coated substrates, which showed comparative signal when applied to naïve mice. In conclusion, zwitterionic pSBMA polymers (of equivalent hydrodynamic radii to PEG) allowed detection of dengue NS1 disease biomarker in a preclinical model of dengue infection, showing significantly higher signal-to-noise ratio in comparison to the PEG controls. The results of this study will be useful in the future development of a range of protein biosensors designed for use in vivo.