Antifouling Zwitterionic Coating via Electrochemically Mediated Atom

Subscriber access provided by BOSTON UNIV

Article

An antifouling zwitterionic coating via electrochemically mediated ATRP on enzyme-based glucose sensors for long-time stability in 37°C serum Yichuan Hu, Bo Liang, Lu Fang, Guanglong Ma, Guang Yang, Qin Zhu, Shengfu Chen, and Xuesong Ye Langmuir, Just Accepted Manuscript • DOI: 10.1021/acs.langmuir.6b03016 • Publication Date (Web): 18 Oct 2016 Downloaded from http://pubs.acs.org on October 27, 2016

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

Langmuir is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 25

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

An antifouling zwitterionic coating via electrochemically mediated ATRP on enzyme-based glucose sensors for long-time stability in 37°C serum Yichuan Hua,c, 1, Bo Lianga, 1, Lu Fanga, Guanglong Mab, Guang Yanga, Qin Zhua, Shengfu Chen b*, Xuesong Yea,d,*

a

Biosensor National Special Laboratory, Key Laboratory of Biomedical Engineering of Ministry of Education, College of Biomedical Engineering & Instrument Science, Zhejiang University

b

Key Laboratory of Biomass Chemical Engineering of Ministry of Education, Department of Chemical and Biological Engineering, Zhejiang University c d

Zhijiang College, Zhejiang University of Technology

State Key Laboratory of CAD&CG, Zhejiang University Hangzhou, 310027, P. R. China

*Corresponding Author. Tel.: +86 571 87952756. E-mail address: [email protected] (X. Ye) or [email protected] (S. Chen). 1

These authors contributed equally to this work.

ACS Paragon Plus Environment

Langmuir

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 2 of 25

ABSTRACT In this work, a versatile fabrication method of coating enzyme-based biosensors with ultra-thin antifouling zwitterionic polymer films to meet the challenge of the long-time stability of sensors in vivo was developed. Electrochemically mediated atom transfer radical polymerization (eATRP) was applied to polymerize zwitterionic sulfobetaine methacrylate (SBMA) monomers on the rough enzyme-absorbed electrode surfaces, meanwhile a refined overall bromination was developed to improve the coverage of polymers on the biosensor surfaces, and maintain the enzyme activity simultaneously for the first time. X-ray photoelectron spectrum (XPS) and Atom force microscope (AFM) were used to characterize the properties of the polymer layers. The antifouling performance and long-time stability in 37 °C undiluted bovine serum in vitro were evaluated. Results showed that the polymer brush coatings diminished over 99% nonspecific protein adsorption and the sensitivities of evaluated sensor maintained 94% after 15 days. The overall sensitivity deviation of 7% was nearly 50% lower than that of the polyurethane (PU) coated ones, and also much smaller than the current commercially available glucose biosensors. Those results suggested that this highly controllable electro deposition preparation procedure could be a promising method to develop implantable biosensors with long-time stability. Keywords: electrochemically mediated ATRP, zwitterionic hydrogel, nonfouling, biosensor, stability

1

Introduction In the development of implantable biosensors, the long-time stability in vivo maybe is the

most challenging issue, especially when other requirements like sensitivity, linear range,


Page 3 of 25

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

accuracy etc. have been greatly improved over the past few decades [1]. The sensitivities of those sensors have exceeded clinical needs for a long time [2]. Reasons such as enzyme instability, leaching, membrane degradation and delamination, as well as electrode passivation could directly cause the failures of enzyme biosensors in vivo [3]. Moreover, all of those reasons get much worse in vivo because of the foreign body reaction (FBR), where the inflammatory responses may decrease the local oxygen concentration which is essential in the redox reaction to make the oxidase-based biosensors work stably. To eliminate the effect of the decrease of oxygen concentration, Gough developed a glucose sensor with complicated construction integrated with an oxygen sensor worked for over 1 year in pig [4]. Another idea is to prepare robust biocompatible outer membranes to release the FBR. For example, epoxy-polyurethane layer was shown to prolong the lifetimes of glucose sensors in rats [5]. However, few progresses have been achieved in practical application. Up to date, the lifetime of the most durable glucose sensor that has been approved by the U.S. Food and Drug Administration (FDA) is within 7 days and the calibration period of those sensors is less than 12 hours [6]. Besides cell factors release, the nonspecific protein adsorption (biofouling) is one of the first primary steps of FBR during implantation. In proteome investigations, serum albumin and other endogenous protein fragments diminished the diffusion of biochemical analytes to the sensor surfaces [7, 8]. In long-term, the biofouling may induce inflammatory cells invasion and promote subsequent fibrosis [9]. For that reason, numerous materials have been developed to form a protection coating to reduce nonspecific protein adsorption on biosensors, such as monolayer protein [10], Nafion [11, 12], polyethylene oxide (PEO) [13], poly(ethyleneglycol) (PEG) [14], oligo(ethylene glycol) (OEG) [15] and polycarbonate [16]. Among those materials,


Langmuir

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 4 of 25

zwitterionic polymers containing both cationic and anionic groups on the same monomer residue have been proved to be biocompatible and multifunctional [17]. The zwitterionic polymer coatings reduce nonspecific protein adsorption not only from single protein solutions such as BSA, fibrinogen, antibody and lysozyme, but also from complex mediums, such as undiluted blood plasma [18, 19]. Furthermore, the FBR can be resisted on zwitterionic hydrogel implants, which might closely relate with the super-low biofouling of zwitterionic materials [20]. Recently, poly(carboxybetaine) (PCB) functionalized cellulose paper achieved sensitive and rapid glucose detection in undiluted human serum with extremely low biofouling [21]. The oligo (ethylene glycol) (OEG)-based coatings on electrodes were of very low impedance and could limit biofouling [22]. Yang and coworkers prepared protype glucose sensors with photo-polymerized pCBMA hydrogel and results showed that those sensors had long-time stabilities up to 42 days when stored in 4 °C serum [23, 24]. Thus, zwitterionic polymer coatings could be potential candidates of protection layers for implantable enzyme biosensors. However, in 37 °C serum, how the long-time stability of biosensors with this coating is remains to be studied because the stability of the enzyme-based biochemical systems is susceptible to temperature [25, 26]. Most of the zwitterionic polymers mentioned here were polymerized by atom transfer radical polymerization (ATRP) which is conducted by an active equilibrium between lower and higher oxidation states of a transit metal complex. As a universal method, ATRP has been intensively investigated in the recent decades and many new types of ATRP emerge For instance, activators regenerated by electron transfer for atom transfer radical polymerization (ARGET ATRP) facilitates polymerization with high controllability in atmosphere environment. In 2011, Magenau et al. reported electrochemically mediated ATRP (eATRP) for the first time


Page 5 of 25

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

[27]. Applied potential becomes a key force that promotes polymerization [28, 29]. Soon after, electrochemically induced surface-initiated ATRP (e-siATRP) was developed [30], which could facilitate localized polymerization through controlling the applied potential. To minimize the biofouling impact on the implantable neural electrodes performance, ultra-thin pSBMA layers on needle-like gold electrodes were investigated in our previous researches and results demonstrated excellent antifouling performance both in vitro and in vivo [31]. When considering the stability of enzyme-based biosensors, it is desirable that these biocompatible membranes can be coated outside the sensor as to protect the immobilized enzymes. However, whether the zwitterionic coating can be electro polymerized on rough enzyme-based biosensor surfaces is rarely studied so far. When a typical ATRP is applied to the preparation of enzyme biosensors, how to retain the enzyme activity in the harsh reaction conditions, and how to form a complete coating on rough surfaces are challenging. In this work, we presented a new fabrication method of antifouling zwitterionic coating on enzyme-based glucose sensors and the long-time stability was investigated in 37 °C serum to mimic in vivo condition.

2

Experimental methods 2.1

Materials and instruments

N-(3-sulfopropyl)-N-(methacryloxyethyl)-N, N-dimethyl ammonium betaine (SBMA, 97%), bromoisobu-tyryl bromide (BIBB, 98%), polyurethane (PU, average mol. wt. of 100,000) and tris (2-pyridylmethyl) amine (TPMA) were purchased from Sigma-Aldrich (Milwaukee, USA.). Glucose oxidase (GOx) and bovine serum albumin (BSA) were purchased from TCI (Shanghai, China). Copper Chloride (AP), dimethyl formamide (DMF, AP), tetrahydrofuran (THF, AP) and dichloromethane (AP) were purchased from Aladdin (Shanghai, China).


Langmuir

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 6 of 25

Biotin-labeled goat anti-mouse IgG (H+L), HRP-labeled streptavidin, and TMB horseradish peroxidase color development solution were purchased from Beyotime (Shanghai, China). New-born calf serum was purchased from Tianhang Biotech (Hangzhou, China). Platinum wire (99.99%, Φ0.4 mm) was purchased from New-Metal (Bejing, China). Aniline (ANi, AP) chloroplatinic acid (AP), glutaraldehyde (25%) and other regular chemicals used were of analytical reagent grade, purchased from Sinopharm (Shanghai, China). Electro-polymerization and electrochemical measurements were performed with an electrochemical workstation (μAutolab III, Metrohm, Switzerland). The surface morphology of untreated membranes and modified membranes was imaged by AFM (using Multimode SPM from American VEECO Company) by contact mode. The AFM images of 2 μm scans were acquired by scanning the sample in air under ambient laboratory conditions. All electrochemical experiments were carried out at room temperature. 2.2

Preparation of protein-initiator

The initiator used in this paper was prepared following the procedure of Lele’s [32]. In brief, GOx (250 mg) and BSA (100 mg) were dissolved in 30 ml, 0.1 M phosphate buffer (PBS, pH 8.0), then 0.5 ml dichloromethane containing 15 μl BIBB was slowly dropped in under violent magnet stirring of 800 rpm at room temperature for 1 hour. To increase the coverage of initiators on protein surfaces, an excess mole rate of 80:1 for BIBB to protein was chosen. During the modification, the solution pH was maintained at 8 by adding 0.1 M sodium hydroxide in order to increase the nucleophilicity of primary amino group of lysine residues in proteins toward the acylating reagent. After that, proteins were purified by Amicon Ultra ultrafiltration tube (30 kDa) under centrifugation of 5500 g for 40 minutes. Then those GOx-Br were dissolved in PBS (0.1M, pH 7.4) to the concentration of 10 mg/ml.


Page 7 of 25

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

Fig. 1 Mechanism of electrochemically induced ATRP


Langmuir

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

2.3

Page 8 of 25

Preparation of enzyme sensors

The preparation of enzyme sensors is showed in Fig. 2. Firstly, platinum wires of 4 cm in length were thorough rinsed in acetone and anhydrous ethyl alcohol then chemically polished in 1 M sulfuric acid under cyclic voltammetry (CV) from -0.8 to 0.8 V for 20 cycles. A following platinization was performed by immersing those wires into 10 mM H2PtCl6/0.1 M HCl solution and a bias of 100 mV (vs Ag/AgCl) was applied for 10 minutes. Poly-aniline film was electro polymerized in 0.4 M aniline/0.1 M HCl solution at a current density of 0.1 mA/cm2 for 10 min. The GOx-Br prepared previously was physically absorbed on the pANi layer after 2 hour of soaking at 37 °C. After that, the Pt electrodes were taken out and rinsed with deionized water, then dried at room temperature. The proteins immobilized on electrodes were crosslinked in saturated glutaraldehyde vapor at 37°C for 3 hours. 2.4

Overall bromination of enzyme sensors

In order to eliminate possible faults on the surface of electrode during eATRP, an overall bromination was performed after crosslink. Similar to the preparation of protein initiators, Pt-pANi-GOx electrodes were immersed in 30 ml PBS (0.1 M, pH 8.0), then 0.5 ml dichloromethane containing 15 μl BIBB was slowly dropped in under violent magnet stirring of 800 rpm at 20 °C for 1 hour. Then electrodes were taken out, rinsed by deionized water, dried at room temperature. PU-coated electrodes were dipped in 3% (w/w) PU solution of 98% THF/2% DMF (w/w).


Page 9 of 25

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

Fig. 2 The preparation of Pt-pANi-GOx-pSBMA sensor. Red arrows show additional bromination sites of primary amino groups on pANi.

2.5

Conduction of eATRP

eATRP was carried with a typical 3 electrodes system in 10 ml PBS (0.1 M, pH 7.4), with 20% SBMA, 2 mM TPMA, 2 mM CuCl2 (the mole rate of ligand: Cu = 1: 1) and 1 mM glucose (worked as a protector of enzyme activities). Firstly, a cyclic voltammetry (CV) was applied under the potential range of -1.0 V and 0.6 V at 100 mVs-1 (versus Ag/AgCl) for 20 cycle. The applied potential used in final polymerization was determined by the following ELISA assay according to the best antifouling performance. In this experiment, -0.4 V was chosen to perform the eATRP. 2.6

Nonspecific protein adsorption and ELISA assay


Langmuir

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 10 of 25

Prepared Pt-pANi-GOx-pSBMA sensors and other samples were immersed in a 0.1 mg/ml biotin-labeled goat anti-mouse IgG solution for 1 h, at 37 °C. Then, all samples were rinsed five times with PBS (0.05M, pH 7.4) and incubated in BSA (1 mg/mL in PBS, pH 7.4) solution for 90 min at 37 °C to block the areas unoccupied by antibody. After being rinsed five times by the PBS, all samples were treated by HRP-streptavidin for 30 min, at a 1/2000 dilution, then rinsed again. After that, electrodes were transferred to clean micro centrifuge tubes, incubated with 200 µL of 0.01 M PBS containing TMB horseradish peroxidase for 20 min at 37 °C. The enzyme-induced color reaction was stopped by adding 50 µL of 2 M sulfuric acid to the solution in each tube. Finally, the absorbance of light intensity at 450 nm was determined by a micro plate reader. The absorbance from the bare Pt wire was equivalent to 100% for calculating relative adsorption values. 2.7

In vitro stability evaluation in undiluted bovine serum

Newly prepared sensors for long-time evaluations were stored in bovine serum with 1% Proclin (Shinegene, Shanghai) in 37 °C thermostat. The sensors were measured every 3 days in PBS (0.05M, pH 7.4). All amperometric measurements were performed at room temperature (24°C) with magnetic stirring of 280 rpm and the applied potential was +0.65 V (vs. Ag/AgCl).

3

Results and discussion 3.1 The effect of bromination on the sensitivity of glucose sensors To evaluate the effect of bromination to GOx activity, sensors made by brominated GOx

and non-brominated GOx were prepared simultaneously. Both responses to the glucose were invested by constant potential electrochemistry assay. As shown in Fig.3, the brominated GOx


Page 11 of 25

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

maintained about 80-90 % of enzyme activity compared to the non-brominated GOx from the glucose concentration of 1 to 6 mM.

Fig. 3 The sensitivity of Pt-pANi-GOx sensor with/without bromination.

The curves shown in Fig.4 represent the response of Pt-pANi-GOx-pSBMA sensor to successive addition of 0.2 mM glucose. Referred to the data shown in Fig.3, the linear dynamic range of this sensor in PBS is from 0.2mM to 5 mM and the correlation coefficient R is 0.9987. The detection limitation is about 0.05 mM (S/N=3). The polymerization process didn’t show much impact on the activity of enzymes which suggested that the whole preparation procedure of sensors was enzyme-friendly.


Langmuir

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 12 of 25

Fig. 4 Amperometric response of Pt-pANi-GOx-pSBMA sensor to successive addition of 0.2 mM glucose in 50 mM PBS (pH 7.4) at +0.65 V vs. Ag/AgCl, 25°C.

3.2 Electrochemistry performance In a typical ATRP, the reaction system containing the monomer, the ligand, and the initiator will not react until CuI is added in. And the rate of [CuI/Ligand]/[CuII/Ligand] determines the speed of ATRP directly. Commonly, this rate is controlled by [CuI] added into the polymerization systems, the higher the faster. During the propagation of polymer, CuI/Ligand contributes an electron to monomer and turns it to [R·], then converts to CuII/Ligand, preparing for the next cycle. While in eATRP, electrochemistry process is driven by ΔGo=F (Eapp -E0’). Thus, the proper applied potential is a key point in eATRP [33]. When the potential is lower than E0’, CuII reduces to CuI, and CuI/Ligand activates the polymerization [27]. A


Page 13 of 25

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

higher ΔGo provides a larger rate of [CuI/Ligand]/[CuII/Ligand], which pushes a faster propagation on the surface of working electrode. Our previous experiments showed that the peak potential provided a faster polymerization speed than lower potential but the best antifouling performance was achieved under a little bit lower potential between the reductive peak and zero. Therefore, the applied potential of -0.4 V was chosen in following polymerizations.

Fig. 5 Cyclic voltammetry curve (CVA) of 10 mL solution of PBS (0.1M, pH 7.4) at 25 °C containing 20% SBMA, 1 mM TPMA and 1 mM CuI, [ligand] : [CuI] = 1 : 1, scan rate = 50 mVs-1. The red points correspond to the Eapp used in following polymerizations. The arrows represent the reductive peaks of CuII to CuI (a) and CuI to Cu0 (b) respectively.

3.3 Polymerization of SBMA


Langmuir

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 14 of 25

On the basis of the scheme showed in Fig.1, Pt-pANi-GOx-pSBMA electrodes were constructed via eATRP of SBMA. The presence of grafted sulbetaine moieties was ascertained by XPS analysis. As shown in Fig. 6, the S2p peak appeared at the binding energy (BE) of 168 eV after eATRP of SBMA [34]. The peak of Br3d was expected to show at 69 eV. But because of the low proportion in compounds and low atom sensitivity of brominate, it was unable to be identified from the same spectrum.

Fig. 6 XPS of enzyme sensors with and without pSBMA coating. The insertion shows the high resolution spectra of S2p region.

AFM pictures revealed morphological features of different electrodes. In Fig. 7 (A), delicately polished Pt wire was quite flat and the roughness varied within 30 nm. Scratches up to 600 nm could be seen clearly. As shown in Fig. 7 (B), the electrode cross-linked with


Page 15 of 25

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

proteins was much rough which roughness was about 50-100 nm. Co-condensed proteins and pANi formed a block-shaped appearance. After covered with a layer of pSBMA, the surfaces of Pt-pANi-GOx-pSBMA electrodes were full of smaller bumps which was closely packed at 20-30 nm diameter mostly. This suggested that an excellent polymer layer had formed. On the contrary, the surface with PU coating was covered with tiny crinkles which may be caused by pulling during preparation.

Fig. 7 AFM images of different surfaces: bare Pt (A), sensor of Pt-pANi-GOx (B), sensor of Pt-pANi-GOx-pSBMA (C) and sensor of Pt-pANi-GOx-PU (D)

3.4 Antifouling performance Following ELISA assays showed the performances of different coatings. Calculated by the relative OD values vs bare Pt surface, non-pSBMA coated examples absorbed much more protein than coated examples. As has been demonstrated in AFM pictures, the modification of pANi and proteins increased the roughness of electrode surfaces and consequently raised the biofouling by about 40%, as shown in Fig.8. Surfaces with PU appeared closely to bare Pt surfaces in the same assay. In this paper, pSBMA layers were polymerized under the same applied potential. The antifouling capabilities showed clear relationship with time. The best antifouling performance occurred after 3 hours’ polymerization. With shorter or longer


Langmuir

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 16 of 25

polymerization time, protein adsorption increased. This result agreed with our previous research [31]. According to former researches [29, 30], eATRP is a “living” process controlled by both applied potential and polymerization time. As the fundamental factor that triggers the electro-polymerization, higher ΔGo (=Eapp -E0’) leads to faster polymerization rate. Longer polymerization leads to thicker pSBMA polymer brush layer. And the thickness of pSBMA polymer layer was expected to be the most vital parameter that affected antifouling ability. Numerous evidences have demonstrated that 2 features played key roles in antifouling. The first one is the surface hydration. Zwitterionic moieties can bind water molecules strongly via ionic solvation to prevent the surface dehydration of both protein molecules and polymer [35, 36]. The second one is the packing density. The uniform polymer brush helps to achieve higher surface coverage, which deters the protein adsorption on uncovered surface defects [37]. With the increase of the thickness of pSBMA brush layer, the surface hydrated and the coverage increase, which would lead to the decrease of biofouling. However, a too thick pSBMA brush layer could lead to strong dipole interaction between zwitterionic pairs, which could reduce the hydration of the brush and cause the protein adsorption. In our previous research, pSBMA brush layer showed an optimized thickness around 23nm [31]. The ELISA results also showed the effect of overall bromination to enzyme sensors. Without it, the best antifouling performance compelled about 70% protein absorption while with overall bromination, the OD values could reach the detection limitation of ELISA assay (optical density was as low as 0.0001). In preparations of enzyme biosensors, the electro deposition of conductive polymer (CP) is a widely used method. Polymerized CPs formed various nano level structures from particles, wires, to pillars. Nano structures on electrodes


Page 17 of 25

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

provide extremely huge specific surface area and diverse micro environments increasing micro particles adsorption dramatically. Enzymes, no matter physically absorbed or covalently bonded, contribute to increase the roughness of electrodes too. Inconsistent surface augmented the difficulty in fabricating antifouling enzyme sensors. In this research, the proteins, both GOx and BSA, were physically absorbed onto pANi nano fibers under electrostatic interaction. Therefore, it was impossible for the protein molecules to cover all the surfaces. Defects on sensor surfaces without initiators may bring about potential flaws on polymer layers in ATRP [30]. Those flaws would cause non-specific protein adsorption and lower the overall performance of sensors. Thus, it was vital to eliminate all the underlying defects of initiators on sensor surfaces. For this purpose, the overall bromination of the whole electrodes was introduced. ELISA assay showed that proteins adsorbed on electrodes (orange columns in Fig.8) deposited with pANi decreased by about 20% after the bromination suggesting considerable bromination sites on primary amino groups of pANi. On fully prepared enzyme sensors, pSBMA brush layers grafted from non-brominated electrodes (red columns) reduced protein adsorption by 34~80% compared to non-pSBMA coated ones (by 57~73% compared to bare Pt electrodes) after varied polymerization time. By contrast, layers grafted from electrodes with the overall bromination (green columns) increased antifouling performances up to 99% which were very close to those from the compact thiol ester initiators under the same polymerization conditions [31]. Thus, the overall bromination provided additional initial polymerization sites on pANi and gave the sensors a more complete coating to block defections efficiently.


Langmuir

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 18 of 25

Fig. 8 Antifouling performances of different coatings inspected by ELISA. (n =3)

3.5 The effect of pSBMA coating to enzyme sensor stability in complex medium The small drift (the maximum change of the sensitivities) range is very important for the practical use because the sensor can work in a predictable status and the calibration period can be extended. When considering this issue, the stability of the implantable enzyme-based


Page 19 of 25

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

biosensors relays not only on the repress of FBR but also on the concentration of reactants or substrates. For enzyme-based glucose biosensors, because the oxygen and glucose concentration in vivo is time varying in daily life and the oxygen supply will affect the electrochemical reaction [38], the 37 °C serum in vitro instead of in vivo experiments was selected to investigate the impact of biofouling on the stability of the glucose biosensors coated with biocompatible membranes. The GOx enzyme glucose sensors with PU coating and SBMA coating, as well as commercially available Dexcom® G4 glucose sensors were selected to compare the sensor stability in complex medium. In the long-time stability evaluation, all sensors were kept in undiluted bovine serum in 37 °C in vitro. Each one was taken out and current responses to glucose were tested every 3 days under the constant potential at +0.65 V vs. Ag/AgCl at room temperature. The relative sensitivity, the percentage of current sensitivities vs original sensitivities, was chosen as the evaluation criterion. As shown in Fig. 9, for the sensors with PU coating, the relative sensitivity of dropped gradually during the whole experiment and 65% was remained after 15 days. For the Dexcom® G4 glucose sensors, the sensitivity increased in the first day by over 20% and fluctuated from 107% to 130%. Their responses disappeared after 24 days in undocumented data. The overall lifetimes of those sensors were longer than the official data [39]. The responses of sensors without any coating decreased rapidly in the first week and almost disappeared after 9 days. In contrast, the pSBMA layer blocked the protein adsorption and maintained the sensor sensitivity mostly. The sensors with pSBMA of 3 hours’ polymerization showed the best protection ability that 94% relative sensitivities remained after 15 days, nearly 50% higher than PU coated ones,


Langmuir

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 20 of 25

and the drift of those sensors were kept within 7%, which indicated that these sensors were working in a more stable status than the other. In many previous studies of the long-time stability in 37 °C bovine serum in vitro, various ideas were proposed. The glucose sensors with drug-eluting hydrogels outer coating reduced sensitivity drifts and kept 65% initial sensitivity after incubation for 30 days [40]. The pLGA coated glucose sensors left 80% of its initial sensitivity after 44 days’ storage, while the drift of relative sensitivity was about 50% [36]. For the most previous experiments of the application of the zwitterionic hydrogels for biosensors, the enzymes were covalently attached on the outer membrane after ATRP [19, 37], while in this article, eATRP was performed by a “graft from” way and the GOx was packaged inside of the polymer, and it was demonstrated that the highly hydrated zwitterionic chains could provide a protein-friendly micro environment and made the inner GOx stable [41]. Table 1 shows the results of those studies.


Page 21 of 25

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

Fig. 9 The long-time effect of pSBMA coating to sensor relative sensitive (represent for the rate of current response to glucose of day X/day 0) in undiluted new bovine serum. (n = 3)

Table 1 Comparison of the sensor lifetimes in undiluted serum. Sensor structure

Life time

Pt/GOx/pCBMA [23]

21 days

Sensitivity left (%) 60

Pt/0.1% crosslinked CBMA/GOx [24]

42 days

near 100

Pt/pAP/HPU, pVAB [42]

160 h

50

About 20

30 days

65

> 20

pVA/pLGA drug-eluting hydrogels coating [40]

Maximum drift (%) About 10

Human blood serum at 4 °C

Antifouling Zwitterionic Coating via Electrochemically Mediated Atom

Recommend Documents