Fabrication of an Anti-Biofouling Plasma-Filtration Membrane by an

May 23, 2017 - Shengqiu Chen , Yu Du , Xiang Zhang , Yi Xie , Zhenqiang Shi , Haifeng Ji , Weifeng Zhao , Changsheng Zhao. Journal of the Taiwan Insti...
0 downloads 0 Views 4MB Size
Subscriber access provided by CORNELL UNIVERSITY LIBRARY

Article

Fabrication of Anti-Biofouling Plasma-Filtration Membrane by Electrospinning Process Using Photo-Crosslinkable Zwitterionic Phospholipid Polymers Jiae Seo, and Ji-Hun Seo ACS Appl. Mater. Interfaces, Just Accepted Manuscript • Publication Date (Web): 23 May 2017 Downloaded from http://pubs.acs.org on May 24, 2017

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.

ACS Applied Materials & Interfaces 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 27

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

ACS Applied Materials & Interfaces

Fabrication of Anti-Biofouling Plasma-Filtration Membrane by Electrospinning Process Using PhotoCrosslinkable Zwitterionic Phospholipid Polymers Jiae Seo, Ji-Hun Seo* Department of Materials Science and Engineering, Korea University, 145 Anam-ro, Seongbukgu, Seoul, 02841, Korea *Correspondence addressed to: [email protected] KEYWORDS: microfiltration, zwitterion, phosphorylcholine, protein adsorption, platelet adhesion

ABSTRACT: The goal of this study is to fabricate a stable plasma filtration membrane with anti-biofouling properties via an electrospinning process. To this end, a random-type copolymer consisting of zwitterionic phosphorylcholine (PC) groups and ultraviolet (UV)-crosslinkable phenyl azide groups was synthesized. The zwitterionic PC group provides anti-biofouling properties, and the phenyl azide group enables the stable maintenance of the fibrous nanostructure of hydrophilic zwitterion polymers in aqueous medium via a simple UV curing process. To demonstrate the anti-biofouling nature of the PC group, a polymer without antibiofouling PC groups was also prepared for comparison. The successful synthesis of the random-

ACS Paragon Plus Environment

1

ACS Applied Materials & Interfaces

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 27

type copolymers containing phenyl azide groups was proven by H1 nuclear magnetic resonance and Fourier transform infrared spectroscopy, and the fibrous structure of the prepared membranes was observed by field emission scanning electron microscopy. The anti-biofouling properties were analyzed by fluorescein isothiocyanate-labeled bovine serum albumin adsorption and platelet adhesion tests. The experimental results show that membranes containing zwitterionic PC groups exhibited obvious decreases in platelet adhesion and protein adsorption. Platelet-rich plasma solution was filtered using the prepared membranes to test their filtration properties. The sequential filtration process removed 80 % and almost 98 % of the platelets. This finding confirmed that the membrane retained its blood-inert biomaterial surface in a complex medium that included blood plasma and platelets.

ACS Paragon Plus Environment

2

Page 3 of 27

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

ACS Applied Materials & Interfaces

1. Introduction The development of membranes that effectively filter out nano/micro-sized organic matter and microbes is highly desirable for the fields of biomedical and environmental engineering. Bloodfiltration membranes are essential for removing unnecessary blood cells when human blood is used for transfusion, diagnosis, or analysis.1‒3 Based on the size of the target blood cells to be eliminated, nano- and micro-pored filtration membranes have been prepared using poly(ethylene terephthalate), poly(butylene terephthalate), polyurethane, and polyester by adapting various types of processing methods, such as electrospinning, phase inversion, and salt leaching.4‒9 Although various types of blood-filtration membranes have been reported, blood clot formation on the filtration membranes always precludes their safe and long-term application in bloodcontacting environments.10 Because most hydrophobic polymer materials used in blood-filtration membranes interact strongly with plasma proteins, platelet adhesion, downstream activation, and clot formation are readily induced.11 Therefore, the prevention of non-specific protein adsorption and subsequent platelet adhesion (i.e., the creation of an anti-biofouling membrane) has been one of the most important issues in the blood-filtration membrane field. There are two effective ways of preventing membrane fouling: coating anti-biofouling polymers on pre-existing membranes, and fabricating membranes using anti-biofouling polymers. The former method is advantageous in the viewpoint of membrane stability, because pre-existing stable membranes are treated by the anti-biofouling polymers. Physical coating or chemical grafting of anti-biofouling polymers are well used method for preventing membrane fouling. For instance, formation of zwitterionic polymer branches on membrane surface was known as an effective method to provide antibiofouling property on the pre-exist membranes.12,13 However, the small pores in the membranes could be blocked by the aggregation of the coating polymers.14 Moreover, the pressure needed

ACS Paragon Plus Environment

3

ACS Applied Materials & Interfaces

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 27

for the solution to pass through the membrane could be increased, eventually resulting in the deterioration of the filtration efficiency.15 The later methods are relatively free of this problem because the anti-biofouling functional groups are already included in the membrane materials. However, membrane swelling or structural disruption could arise because most anti-biofouling groups are extremely hydrophilic. As a result, such membranes can absorb significant amounts of water molecules resulting in significant dimensional changes of membranes.16 Therefore, maintaining the microstructure of prepared membranes containing hydrophilic anti-biofouling functional groups is a critical issue that must be resolved in the later method. In this study, the feasibility of developing stable anti-biofouling membranes by means of the later method was confirmed by synthesizing a crosslinkable zwitterionic polymer capable of preventing membrane disruption. To this end, a cell membrane mimicking, zwitterionic 2-methacryloyloxyethyl phosphorylcholine (MPC) polymer containing photo-crosslinkable phenyl azide groups was synthesized, and a fibrous membrane was prepared by a conventional electrospinning method. MPC polymer is a well-known anti-biofouling material because of the thick hydration layer that forms around zwitterionic PC polar groups.17‒19 To prevent membrane disruption induced by MPC polymer swelling, phenyl azide-induced crosslinking20 was performed via a photo-curing process. The purpose of this study is to confirm the possibility of preparing hydrophilic MPC polymer membranes that can be used as blood-filtration membranes without inducing platelet adhesion.

ACS Paragon Plus Environment

4

Page 5 of 27

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

ACS Applied Materials & Interfaces

2. Materials and methods 2.1. Materials MPC monomer was purchased from KCI (Gyeonggi-Do, Korea). n-Butyl methacrylate (BMA), 4-azidobenzoic acid powder, 2-hydroxyethyl methacrylate (HEMA), 4-(4,6-dimethoxy1,3,5-tiazin-2-yl)-4-methylmorpholinium chloride (DMT-MM), and glutaraldehyde (ca. 50 % in water) were purchased from TCI (Tokyo, Japan). 2-Aminoethyl methacrylate hydrochloride (AEMA) powder, 2,2'-azobis(2-methylpropionitrile) (AIBN) solution, and Cell Counting Kit-8 were purchased from Sigma-Aldrich (St. Louis, MO, USA). Dulbecco’s phosphate-buffered saline (DPBS, without calcium chloride and magnesium chloride) was purchased from WELGENE Inc. (Daegu, Korea). Bovine serum albumin (BSA) conjugated with fluorescein isothiocyanate (FITC) was purchased from Thermo Fisher Scientific (Waltham, MA, USA). Human platelet-rich plasma (PRP, count range: ~75–200 × 109/mL) was purchased from Zenbio Inc. (Research Triangle Park, NC, USA). All of the organic solvents were purchased from SAMCHUN Chemical (Gyeonggi-Do, Korea).

2.2. Synthesis of terpolymers containing photo-crosslinkable phenyl azide groups A random-type terpolymer composed of 2-methacryloyloxyethyl phosphorylcholine (MPC), nButyl methacrylate (BMA), and 2-Aminoethyl methacrylate hydrochloride (AEMA) was synthesized by free radical polymerization. For the poly(MPC-co-BMA-co-AEMA) (PMBAm), the total concentration of the monomers dissolved in ethanol was 0.1 M, and the MPC/BMA/AEMA molar feed ratio was 3:5:2. The polymerization was performed for 24 h using AIBN as an initiator at 60°C. After the polymerization, the PMBAm solution was precipitated with diethyl ether/chloroform (7/3, v/v) to remove the residual monomers. The

ACS Paragon Plus Environment

5

ACS Applied Materials & Interfaces

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 27

primary amine (-NH2) group in the precipitated PMBAm powder was reacted with 4azidobenzoic acid mediating DMT-MM condensation agent by formation of amide bond.21-23 The reaction was conducted at 60℃ for 18 h, and precipitated by the identical process, yielding white poly(MPC-co-BMA-co-AzAEMA) (PMBA) powder. The chemical structure of the PMBA is illustrated in Scheme 1. The control polymers, poly(BMA-co-AzAEMA) (PBA) and poly(HEMA-co-BMA-co-AzAEMA) (PHBA) were synthesized by the same procedures and the chemical structures are shown in Figure 1.

The size exclusion chromatography (SEC)

measurement was conducted for PHBA and PBA using a Tosoh RI detector containing two connected TSK-GEL Super HZ-M gel column (Tosoh co., Tokyo, Japan) and TSK-GEL SuperHZ-2500 column (Tosoh co., Tokyo, Japan). Polystyrene was used as a standard in tetrahydrofuran as a solvent. The size exclusion chromatography (SEC) measurement was conducted for PMBA using a RI detector containing PL aquagel-OH mixed gel columns (Agilent Technologies co., California, USA). Polyethylene glycol (PEG)/ polyethylene oxide (PEO) were used as a standard in methanol/water = 5:5 as a solvent.

Scheme 1. Overall membrane fabrication and filtration scheme. 2.3. Preparation of membranes by electrospinning and stability evaluation in aqueous medium

ACS Paragon Plus Environment

6

Page 7 of 27

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

ACS Applied Materials & Interfaces

Homogeneous PBA, PHBA, and PMBA solutions were prepared by dissolving PBA in mixed solvent (chloroform/ethanol=2/8) and PHBA and PMBA in pure ethanol. The general electrospinning equipment consisted of a plastic syringe, a needle (24 gauge), an aluminum foilcovered collector, and a voltage power supplier (ESR100D; NanoNC, Korea). A high voltage (30 kV) was applied to the polymer solution, and the flow rate was kept at 5.0 mL/h using a precision fluid metering pump (EP100; NanoNC, Korea). The working distance between the tip and the collector was 25 cm. The structural stability test was conducted by immersing the membranes in distilled water for 1 h. Two types of membranes were prepared for comparison: the as-spun membrane and a membrane irradiated by 254 nm ultraviolet (UV) light for more than 7 min. Subsequently, the membranes were immersed in deionized (DI) water for 1 h. The morphologies of the membranes before and after immersion were observed by field emission scanning electron microscopy (FESEM; S-4300, Hitachi, Japan).

2.4. FITC-labeled BSA adsorption test First, 5 mg/L FITC-labeled BSA was dissolved in PBS, and then, the prepared membrane (10 ×10 mm) was immersed in the solution for 10 min. The membranes were washed three times with fresh PBS and observed by confocal microscopy (LSM 700, Carl-Zeiss, Germany). The fluorescence intensity on the membrane surface was calculated by Image J.

2.5. Evaluation of platelet adhesion The membranes (10 × 10 mm) were fixed on an 8-well plate, and the wells were filled with 5 mL of PBS and left for 1 h. After removing the PBS, each membrane was soaked with 5 mL of

ACS Paragon Plus Environment

7

ACS Applied Materials & Interfaces

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 8 of 27

PRP solution and incubated at 37℃ for 6 or 16 h. Subsequently, the PRP solution was removed, and the membrane was washed three times with fresh PBS. The adhering platelets were fixed with 2 % glutaraldehyde solution (TCI, Tokyo, Japan) for 1 h. The remaining solution was removed, and the membranes were washed twice with PBS and once with DI water. The membranes were freeze-dried, and the adhering platelets were observed by a Hitachi S-4300 FESEM.

2.6. Brunauer-Emmett-Teller (BET) analysis of membranes with attached platelets Membranes of a fixed size (10 × 10 × 1 mm) were placed into sample tubes. For membrane pre-treatment, the sample tubes were maintained at 60°C and exposed to N2 gas for 24 h. Then, the surface areas of the membranes were measured using the BET method, with N2 as the adsorbate gas and a Tristar 3000 instrument (Micromeritics Instrument Corp., Norcross, GA, USA).

2.7. Filtering performance test PRP solution was passed through membranes. Then, the filtered PRP solution was placed into a 96-well plate and incubated at 37℃ for 1 h after adding Cell Counting Kit-8 solution. The degree of filtration was measured by determining the absorbance difference using a Multiskan FC Microplate Photometer (Thermo Fisher, Waltham, MA, USA). The filtering process was repeated with the filtered solution, and the amount of platelets remaining after each step was measured.

3. Results and discussion

ACS Paragon Plus Environment

8

Page 9 of 27

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

ACS Applied Materials & Interfaces

3.1. Synthesis and characterization of the electrospun membranes In this study, a photo-crosslinkable, anti-biofouling zwitterionic polymer (PMBA) was synthesized (Scheme 1, Figure 1). Other types of non-zwitterionic photo-crosslinkable hydrophilic (PHBA) and hydrophobic (PBA) copolymers were also synthesized as positive and negative controls using the same procedures, DMT-MM-mediated condensation reaction.21-23 By adopting this post-polymerization modification method, free amine group was substituted by photo-crosslinkable phenyl azide groups, as verified by the appearance of the phenyl azide peak at 2150 cm‒1 in Figure 2.24 The existence of amide bonds after the formation of the phenyl azide group was confirmed by the peaks at 1312 cm‒1 and 1580 cm‒1, which correspond to N-H bending and C-O stretching vibrations, respectively.25 The final substitution rate of the free amine to photo-crosslinkable phenyl azide groups were 91.0, 87.0, and 88.1% for PBA, PHBA, and PMBA, respectively, and the residual free amine contents were below 1.8% (Table 1).

Figure 1. Molecular structures of the PBA, PHBA, and PMBA polymers Table 1. Molecular profiles of PBA, PHBA, and PMBA

In polymer composition (%) Symbol BMA

HEMA

MPC

AEMA

Substitution ratio of amine to phenyl azide (%) Az AEMA

Mn (×104 g/mol, SEC)

Mw/Mn

ACS Paragon Plus Environment

9

ACS Applied Materials & Interfaces

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 27

PBA

85.4

-

-

1.31

13.29

91.0

6.88

2.61

PHBA

51.8

35.2

-

1.69

11.31

87.0

6.49

2.41

PMBA

53.7

-

31.3

1.78

13.22

88.1

7.13

1.78

Figure 2. Differences in the FT-IR spectra attributable to the phenyl azide (-N3) functional group: (a) PBA, (b) PHBA, and (c) PMBA The synthesized polymers were then electrospun under optimal condition to form fibrous membranes (Figure 3). The electrospinning conditions are critically dependent on the concentrations of the polymer solutions and the type of solvent suitable for the polymer.26 Choosing an appropriate concentration above the critical value enabled the formation of the desired membrane structures. PBA, which comprised hydrophobic monomers, completely dissolved in mixed solvent (chloroform/ethanol=2/8). PBA was observed to have a fibrous membrane morphology was observed when its concentration exceeded 40 wt% (Figure 3c). In contrast, when dissolved in a solution with a concentration of less than 22 wt%, PHBA displayed a bead-like morphology, whereas that dissolved in a solution with a concentration of 22 wt% or higher took on a fibrous morphology.

ACS Paragon Plus Environment

10

Page 11 of 27

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

ACS Applied Materials & Interfaces

Figure 3. SEM images of polymer membranes with different solution concentrations: (a) 20 wt% PBA, (b) 30 wt% PBA, (c) 40 wt% PBA, (d) 5 wt% PHBA, (e) 10 wt% PHBA, (f) 22 wt% PHBA, (g) 5 wt% PMBA, (h) 10 wt% PMBA, and (i) 20 wt% PMBA A similar tendency was observed for zwitterionic PMBA. A bead-like morphology was formed a lower concentrations, and a fibrous membrane morphology was observed when the concentration exceeded 20 wt%. These results indicate that the polymer concentration is the critical factor for the preparation of fibrous polymer networks, regardless of the solvent polarity. Phenyl azide groups are known to form crosslinking networks by reacting with non-specific protons in organic molecules after exposure to light of the appropriate wavelength (254 nm).20 To find the optimized irradiation time for the formation of crosslinking networks of hydrophilic MPC polymers, the disappearance of the phenyl azide group was monitored by FT-IR. Figure 4

ACS Paragon Plus Environment

11

ACS Applied Materials & Interfaces

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 27

shows the changes in the phenyl azide peak (2150 cm‒1) of the fibrous PMBA membrane for different durations of irradiation with 254 nm light. After only 1 min of UV irradiation at 254 nm, around 61.6% of the phenyl azide peak at 2150 cm‒1 disappeared. When the irradiation time exceeded 7 min, around 98.0% of the phenyl azide peak disappeared, which indicates that irradiation for 7 min is sufficient to complete the crosslinking reaction induced by phenyl azide groups. Thus, the downstream membrane preparation involved membrane irradiation with UV light for more than 7 min.

Figure 4. Changes in the phenyl azide peak of PMBA for different durations of irradiation with 254 nm light. To ensure the structural stability of the hydrophilic zwitterionic polymer network in aqueous medium, the prepared fibrous membrane was exposed to 254 nm UV light for more than 7 min. Subsequently, the structural changes in the membranes before and after soaking in water for 1 h were observed via FE-SEM (Figure 5). No morphological change was found for PBA membranes before and after immersion in water (Figure 5a). Because PBA is composed of hydrophobic BMA units, these membranes may not undergo swelling; this phenomenon is thought to be responsible for the structural stability PBA membranes. PHBA also exhibits

ACS Paragon Plus Environment

12

Page 13 of 27

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

ACS Applied Materials & Interfaces

outstanding structural stability in aqueous medium, despite containing hydrophilic HEMA units. HEMA is a well-known hydrophilic monomer, but it becomes slightly hydrophobic when it is polymerized and its molecular weight exceeds a certain point (tens of thousands).27 Because the molecular weight of the PHBA was over 200 k, the PHBA membrane was expected to maintain its fibrous morphology in aqueous medium, regardless of UV light irradiation (Figure 5b). In contrast, the fibrous structure of the PMBA membrane was completely disrupted after immersion in water for 1 h (Figure 5c) because of the extremely hydrophilic nature of zwitterionic MPC polymers. However, when the PMBA membrane was immersed in water for 1 h after UV irradiation, its fibrous morphology was completely preserved. Simultaneously securing both membrane stability and anti-biofouling property in aqueous medium has been a challenging subject because of the hydrophilic nature of the anti-biofouling functional groups. In the present study, around 13% of photo-crosslinkable phenyl azide group was introduced to the zwitterionic polymer, and around 98% of the phenyl azide group was consumed to form the crosslinked network. It has been well verified that around 10% of crosslinking density is enough to preserve microstructures or patterned array of hydrophilic MPC polymers.20 Therefore, the preserved three-dimensional fibrous morphologies of hydrophilic MPC polymers in aqueous medium is thought to be attributed to the formation of the crosslinking induced by phenyl azide group.

ACS Paragon Plus Environment

13

ACS Applied Materials & Interfaces

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 27

Figure 5. Membrane stability test results: (a) PBA, (b) PHBA, and (c) PMBA. Images show the polymers before and after immersion in D.I. water with and without UV irradiation. 3.2. The protein adsorption resistance of the membranes The amount of protein adsorption on the fibrous membranes was evaluated by contacting the as-spun fibrous membranes with FITC-labeled BSA solution. The intensity of the surface fluorescence was observed by fluorescence microscopy and calculated by Image J, as shown in Figures 6 and 7, respectively. In the case of PBA, a significant amount of BSA was adsorbed on the fibrous membrane. In contrast, the amount of BSA adsorption on PHBA was approximately 40 % of that on PBA. Because hydrophobic interactions are the main factor underlying protein adsorption on material surfaces, the increased hydrophilicity of the PHBA surface is thought to be responsible for the decreased protein adsorption. For PMBA, an almost negligible amount of fluorescence was observed on the fibrous membrane. As seen in Figure 6, PMBA exhibited

ACS Paragon Plus Environment

14

Page 15 of 27

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

ACS Applied Materials & Interfaces

conspicuously low fluorescence intensity, confirming the remarkable ability of this membrane to prevent BSA adsorption. Protein adsorption is the primary event that occurs when artificial materials are brought into contact with body fluids, such as blood plasma.28 Previously, the amount of protein adsorption on artificial materials was believed to depend on many physicochemical factors, such as polarity, electrical charge density, and the hydrogen bonding network around the surfaces.29 Among the various types of anti-biofouling materials, zwitterionic MPC polymer has been demonstrated to be an ideal material for eliminating the effects of the above-mentioned factors.30 The present result confirmed the superior properties of the zwitterionic MPC polymer for preventing initial protein adsorption in anti-biofouling applications.

Figure 6. Fluorescence microscopy images of (a) PBA, (b) PHBA, and (c) PMBA obtained after the FITC-labeled BSA adsorption test

ACS Paragon Plus Environment

15

ACS Applied Materials & Interfaces

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 27

Figure 7. Mean fluorescence intensities of the fluorescence microscopy images 3.3. Platelet adhesion on membranes Preventing the adhesion, activation, and coagulation of platelets remaining after a bloodcontacting surface of an artificial filter is in contact with blood for a certain period of time is critical for these devices’ continuous and safe use.31 When platelets adhere to various surfaces, such as injured endothelium, subendothelium, and artificial surfaces, they change from discoid to flat forms as they become activated, and subsequently, they coagulate by spreading prosthetic legs.32 Once platelets are adsorbed and start to be activated, fibrinogen, which is a type of protein in blood plasma, accelerates the coagulation process. Fibrinogen binds to GPIIb/IIIa in the activated platelets and changes the morphology of the extracellular globular head domain.33 As a result, dense granules and α-granules that activate the nearby platelets are released, and a platelet plug is generated.34 Additionally, the platelets activated by this mechanism release thromboxane A2, leading to the crosslinking of the platelets and, thus, the formation of clots. These clots grow

ACS Paragon Plus Environment

16

Page 17 of 27

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

ACS Applied Materials & Interfaces

rapidly and trigger coagulation cascades and the formation of thrombin. The generation of thrombin causes the clot morphology to become strongly meshed by fibrin and rigidly bound.35 This is the main cause of the performance deterioration of blood-filtration membranes. To confirm the platelet-membrane interactions, the membranes were brought into contact with PRP solution for moderate (6 h) and long (16 h) periods, and the changes in the surface morphology and residual free surface area were analyzed by FE-SEM and BET analysis. As a result, large amounts of platelets were observed to become adhered to and activated on both the PBA and PHBA surfaces after 6 h, and after 16 h, and the membrane surfaces were completely blocked by the film (Figure 8). The surface coverage induced by membrane fouling was calculated by ImageJ analysis method. The finally evaluated portion of the flatten surface after 16 h was 86.3, 64.5, and 2.9 % for PBA, PHBA, and PMBA membranes, respectively. These results may be attributable to the coagulation cascade, which produced a rigidly agglomerated morphology within the fibrin network and are representative of the drawbacks associated with the use of blood-filtration membranes. In contrast, almost no adhered platelets were observed on the PMBA membrane surface, even after 16 h of contact with the PRP solution. Indeed, the few adhered platelet after 6 h were insufficient to induce membrane blockage after 16 h. Furthermore, the fibrous PMBA membrane morphology was not significantly changed, even after being in contact with the PRP solution for 16 h. Therefore, the zwitterionic PMBA membrane is suitable for longterm or repeated usage in blood-contacting environments and does not suffer from membrane blockage, which has been a critical drawback of conventional filtration membranes. The porosity of the micro-fiber membrane was experimentally analyzed by N2 gas adsorption and desorption at a cryogenic temperature of 77 K based on BET theory.36 Figure 9 shows the N2 adsorption and desorption isotherms obtained for PBA, PHBA, and PMBA. The open-ended

ACS Paragon Plus Environment

17

ACS Applied Materials & Interfaces

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 27

hysteresis at relatively low pressure arises from deformation because of the swelling of the polymer networks. The S-curve graph can be interpreted based on the measurement principle of the BET method by dividing the graph into three different sections.37 Starting from the leftmost point on the x-axis, which represents the vacuum state, the pressure increases gradually during the experiment. N2 gas is injected until a monolayer is forms on the relatively large pores in the membrane. Infinitesimal increases in the pressure in the low-pressure range result in high N2 uptake because of the large pores, and consequently, a sharp slope is observed. After all of the large pores are filled, additional N2 is adsorbed on the monolayer, leading to multilayer adsorption. Consequently, the slope of the graph decreases gradually until it is almost flat. At the rightmost point of the x-axis, the slope of the graph soars because of the unfilled nano-sized pores. These pores tend to be stable because of capillary condensation, and thus, the volume of the adsorbed N2 gas increases rapidly. Generally, materials showing an S-curve in their N2 adsorption/desorption isotherms possess pore size distributions ranging from macro to nanosized pores.38 Because all of the prepared membranes have S-curves in this study (Figure 9), the membranes were thought to possess macro and nano-sized pores. However, significant differences were observed when the S-curves obtained before and after exposure to PRP solution for 16 h were compared. For the PBA and PHBA membranes, the amount of N2 gas adsorption decreased significantly after contact with PRP solution (red line). In contrast, only a slight decrease was observed for the zwitterionic PMBA membrane. Based on BET theory, changes in the specific surface area were calculated before and after exposure to PRP solution; the results are summarized in Table 2. The specific surface areas of PBA and PHBA decreased to 43.7 % and 50.4 %, respectively. Thus, more than half of the internal pores were blocked by the coagulation cascade. In contrast, 83.1 % of specific surface area was retained by the fibrous

ACS Paragon Plus Environment

18

Page 19 of 27

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

ACS Applied Materials & Interfaces

PMBA membrane after exposure to PRP solution for 16 h. This result is consistent with the FESEM images shown in Figure 8.

Figure 8. Platelet adhesion on membrane exposed to PRP solution for 0 h, 6 h, and 16 h: (a) PBA, (b) PHBA, and (c) PMBA

ACS Paragon Plus Environment

19

ACS Applied Materials & Interfaces

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 27

Figure 9. BET plots of the membranes. ‘None’ corresponds to the bare membrane, and ‘16 h in PRP’ corresponds to the membrane after exposure to PRP. Table 2. BET surface areas of the membranes BET Surface Area (m2/g) None

Contact with PRP for 16 h

PBA

21.3

9.3

PHBA

22.4

11.3

PMBA

23.6

19.6

3.4. Plasma filtration The filtering performance was tested via the top-down filtration of PRP through the fabricated PMBA membrane (2 × 2 cm2). After the first filtration, approximately 80 % of the platelets in the plasma were removed, and after the second filtration, only 2~5 % of the platelets were left (Figure 10). Although 20 % of the platelets remained after the first filtration, this filtration performance is expected to be improved by optimizing the electrospinning conditions and, thereby, reducing the mean pore size of the membrane. In the case of PBA and PHBA membrane, much higher load was required to filter PRP solution resulting in frequent membrane

ACS Paragon Plus Environment

20

Page 21 of 27

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

ACS Applied Materials & Interfaces

failure, which is probably due to the lower polarity of the control samples, or difference in pore size,39 or the significant membrane fouling which is already shown in the present study. Because of these problems, the filtration performance among the prepared samples were not able to be discussed comparatively. To confirm the anti-biofouling properties during the filtration process, each membrane was cut, and the cross-sectional areas were imaged by FE-SEM after filtration. As Figure 11 clearly shows, significant platelet adhesion and membrane blockage were initiated inside the PBA and PHBA membranes, and consequently, the fibrous morphology could not be easily observed. In contrast, the fibrous three-dimensional structure inside of the PMBA membrane was clearly evident, indicating its outstanding anti-biofouling properties. Based on these results, the UVcrosslinked PMBA fibrous membrane can be applied as a plasma-filtration membrane that does not suffer from membrane disruption and blockage, which have been serious drawbacks of previously developed blood-filtration membranes.

Figure 10. Filtering performance test of the PMBA fibrous membrane

ACS Paragon Plus Environment

21

ACS Applied Materials & Interfaces

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 22 of 27

Figure 11. SEM images obtained immediately after the membranes were used to filter PRP solution: (a) PBA, (b) PHBA, and (c) PMBA

4. Conclusions The development of blood plasma-filtration membranes that are able to separate unnecessary blood cells without suffering from membrane blockage has been an important issue in the biomedical field. To this end, anti-biofouling functional groups have been introduced into filtration membranes. However, membrane disruption resulting from the swelling induced by hydrophilic anti-biofouling groups was a critical problem that had to be overcome. In the present study, the direct preparation of membranes using photo-crosslinkable zwitterionic polymers by an electrospinning method was confirmed to be a useful technique for the creation of stable blood-filtration membranes with long-term applicability that do not suffer from membrane blockage.

ACS Paragon Plus Environment

22

Page 23 of 27

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

ACS Applied Materials & Interfaces

AUTHOR INFORMATION Corresponding Author *Tel.: +82-2-3290-3263. Fax: +82-2-928-3584. E-mail: [email protected]

Author Contributions The manuscript was written through contributions of all authors. All authors have given approval to the final version of the manuscript.

Funding Sources This research was supported by Basic Science Research Program through the National Research Foundation of Korea (NRF) (2015R1C1A1A01054022) and Start-up Funding from Korea University Grant (K1505081).

ABBREVIATIONS AEMA, 2-aminoethyl methacrylate hydrochloride; AIBN, 2,2'-azobis(2-methylpropionitrile); BET, Brunauer-Emmett-Teller; BMA, n-butyl methacrylate; BSA, bovine serum albumin; DI, deionized; DMT-MM, 4-(4,6-dimethoxy-1,3,5-tiazin-2-yl)-4-methylmorpholinium chloride; DPBS, Dulbecco’s phosphate-buffered saline; FITC, fluorescein isothiocyanate; FT-IR, Fourier transform infrared spectroscopy; HEMA, 2-hydroxyethyl methacrylate; MPC, 2methacryloyloxyethyl phosphorylcholine; NMR, nuclear magnetic resonance; PBA, [BMA:AzAEMA]n; PBS, phosphate-buffered saline; PC, phosphorylcholine; PHBA, [BMA:HEMA:AzAEMA]n; PMBA, [AEMA:MPC:BMA]n; PRP, platelet-rich plasma; UV, ultraviolet

ACS Paragon Plus Environment

23

ACS Applied Materials & Interfaces

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 24 of 27

REFERENCES (1) Beutler, E.; West, C.; Blume, K. G. The Removal of Leukocytes and Platelets from Whole Blood. J. Lab. Clin. Med. 1976, 88, 328-333. (2) Innerhofer, P.; Klingler, A.; Klimmer, C.; Fries, D.; Nussbaumer, W. Risk for Postoperative Infection after Transfusion of White Blood Cell–filtered Allogeneic or Autologous Blood Components in Orthopedic Patients Undergoing Primary Arthroplasty. Transfusion 2005, 45, 103-110. (3) Novotny, V. M.; van Doorn, R.; Witvliet, M. D.; Claas, F. H.; Brand, A. Occurrence of Allogeneic HLA and non-HLA Antibodies after Transfusion of Prestorage Filtered Platelets and Red Blood Cells: a Prospective Study. Blood 1995, 85, 1736-1741. (4) Renaudie, L.; Le Narvor, C.; Lepleux, E.; Roger, P. Functionalization of Poly (ethylene terephthalate) Fibers by Photografting of a Carbohydrate Derivatized With a Phenyl Azide Group. Biomacromolecules 2007, 8, 679-685. (5) Ma, Z.; Kotaki, M.; Yong, T.; He, W.; Ramakrishna, S. Surface Engineering of Electrospun Polyethylene Terephthalate (PET) Nanofibers towards Development of a New Material for Blood Vessel Engineering. Biomaterials 2005, 26, 2527-2536. (6) Ye, S. H.; Watanabe, J.; Iwasaki, Y.; Ishihara, K. Novel Cellulose Acetate Membrane Blended with Phospholipid Polymer for Hemocompatible Filtration System. J. Membr. Sci. 2002, 210, 411-421. (7) Dzik, S. Leukodepletion Blood Filters: Filter Design and Mechanisms of Leukocyte Removal. Transf. Med. Rev. 1993, 7, 65-77. (8) Higuchi, A.; Yamamiya, S. I.; Yoon, B. O.; Sakurai, M.; Hara, M. Peripheral Blood Cell Separation through Surface‐modified Polyurethane Membranes. J. Biomed. Mater. Res. Part A 2004, 68, 34-42. (9) Zandén, C.; Voinova, M.; Gold, J.; Mörsdorf, D.; Bernhardt, I.; Liu, J. Surface Characterisation of Oxygen Plasma Treated Electrospun Polyurethane Fibres and Their Interaction with Red Blood Cells. Eur. Polym. J. 2012, 48, 472-482. (10) Mann, K. G. The Assembly of Blood Clotting Complexes on Membranes. Trends Biochem. Sci. 1987, 12, 229-233. (11) Tsai, W. B.; Grunkemeier, J. M.; Horbett, T. A. Human Plasma Fibrinogen Adsorption and Platelet Adhesion to Polystyrene. J. Biomed. Mater. Res. Part A 1999, 44, 130-139. (12) Zheng, S.; Yang, Q.; Mi, B. Novel Antifouling Surface with Improved Hemocompatibility by Immobilization of Polyzwitterions onto Silicon via Click Chemistry. Appl. Surf. Sci. 2016, 363, 619-626. (13) Yu, H. Y.; Kang, Y.; Liu, Y.; Mi, B. Grafting Polyzwitterions onto Polyamide by Click Chemistry and Nucleophilic Substitution on Nitrogen: A Novel Approach to Enhance Membrane Fouling Resistance. J. Membr. Sci. 2014, 449, 50-57.

ACS Paragon Plus Environment

24

Page 25 of 27

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

ACS Applied Materials & Interfaces

(14) Bauser, H.; Chmiel, H.; Stroh, N.; Walitza, E. Interfacial Effects with Microfiltration Membranes. J. Membr. Sci. 1982, 11, 321-332. (15) Tang, Y.; Shi, J.; Li, S.; Wang, L.; Cayre, Y. E.; Chen, Y. Microfluidic Device with Integrated Microfilter of Conical-shaped Holes for High Efficiency and High Purity Capture of Circulating Tumor Cells. Sci. Rep. 2014, 4, 6052-6058. (16) Brown, R. H.; Duncan, A. J.; Choi, J. H.; Park, J. K.; Wu, T.; Leo, D. J.; Winey, K. I.; Moore, R. B.; Long, T. E. Effect of Ionic Liquid on Mechanical Properties and Morphology of Zwitterionic Copolymer Membranes. Macromolecules 2009, 43, 790-796. (17) Lewis, A. L. Phosphorylcholine-based Polymers and Their Use in the Prevention of Biofouling. Colloid. Surf. B-Biointerfaces 2000, 18, 261-275. (18) Iwasaki, Y.; Ishihara, K. Cell Membrane-Inspired Phospholipid Polymers for Developing Medical Devices with Excellent Biointerfaces. Sci. Tech. Adv. Mater. 2012, 13, 064101-064114. (19) Seo, J.-H.; Ishihara, K. Diffusion-Induced Hydrophilic Conversion of Polydimethylsiloxane/Block-Type Phospholipid Polymer Hybrid Substrate for Temporal CellAdhesive Surface. ACS Appl. Mater. Interfaces 2016, 8, 21839-21846. (20) Konno, T.; Hasuda, H.; Ishihara, K.; Ito, Y. Photo-immobilization of a Phospholipid Polymer for Surface Modification. Biomaterials 2005, 26, 1381-1388. (21) Montalbetti, C. A.; Falque, V. Amide Bond Formation and Peptide Coupling. Tetrahedron 2005, 61, 10827-10852. (22) Kunishima, M.; Kawachi, C.; Monta, J.; Terao, K.; Iwasaki, F.; Tani, S. 4-(4, 6dimethoxy-1, 3, 5-triazin-2-yl)-4-methyl-morpholinium chloride: An Efficient Condensing Agent Leading to the Formation of Amides and Esters. Tetrahedron 1999, 55, 13159-13170. (23) Kunishima, M.; Kawachi, C.; Iwasaki, F.; Terao, K.; Tani, S. Synthesis and Characterization of 4-(4, 6-dimethoxy-1, 3, 5-triazin-2-yl)-4-methylmorpholinium chloride. Tetrahedron Lett. 1999, 40, 5327-5330. (24) Bhat, V. T.; James, N. R.; Jayakrishnan, A. A Photochemical Method for Immobilization of Azidated Dextran onto Aminated Poly (ethylene terephthalate) Surfaces. Polym. Int. 2008, 57, 124-132. (25) Tang, C. Y.; Kwon, Y. N.; Leckie, J. O. Probing The Nano-and Micro-scales of Reverse Osmosis Membranes—A Comprehensive Characterization of Physiochemical Properties of Uncoated and Coated Membranes by XPS, TEM, ATR-FTIR, and streaming potential measurements. J. Membr. Sci. 2007, 287, 146-156. (26) Doshi, J.; Reneker, D. H. Electrospinning Process and Applications of Electrospun Fibers. J. Electrost. 1995, 35, 151-160. (27) Holly, F. J.; Refojo, M. F. Wettability of Hydrogels I. Poly (2‐hydroxyethyl methacrylate). J. Phys. Chem. A. 1975, 9, 315-326.

ACS Paragon Plus Environment

25

ACS Applied Materials & Interfaces

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 26 of 27

(28) Vaisocherová, H.; Zhang, Z.; Yang, W.; Cao, Z.; Cheng, G.; Taylor, A. D.; Piliarikb, M.; Homolab, J.; Jiang, S. Functionalizable Surface Platform with Reduced Nonspecific Protein Adsorption from Full Blood Plasma—Material Selection and Protein Immobilization Optimization. Biosens. Bioelectron. 2009, 24, 1924-1930. (29) Sun, S.; Yue, Y.; Huang, X.; Meng, D. Protein Adsorption on Blood-contact Membranes. J. Membr. Sci. 2003, 222, 3-18. (30) Ishihara, K.; Oshida, H.; Endo, Y.; Ueda, T.; Watanabe, A.; Nakabayashi, N. Hemocompatibility of Human Whole Blood on Polymers with a Phospholipid Polar Group and Its Mechanism. J. Biomed. Mater. Res. Part A 1992, 26, 1543-1552. (31) Vogler, E. A.; Siedlecki, C. A. Contact Activation of Blood-Plasma Coagulation. Biomaterials 2009, 30, 1857-1869. (32) Siess, W. Molecular Mechanisms of Platelet Activation. Physiol. Rev. 1989, 69, 58-178. (33) Choudhri, T. F.; Hoh, B. L.; Zerwes, H. G.; Prestigiacomo, C. J.; Kim, S. C.; Connolly Jr, E. S.; Kottirsch, G; Pinsky, D. J. Reduced Microvascular Thrombosis and Improved Outcome in Acute Murine Stroke by Inhibiting GP IIb/IIIa Receptor-mediated Platelet Aggregation. J. Clin. Invest. 1998, 102, 1301-1310. (34) Ueno, M.; Kodali, M.; Tello-Montoliu, A.; Angiolillo, D. J. Role of Platelets and Antiplatelet Therapy in Cardiovascular Disease. J. Atheroscler. Thromb. 2011, 18, 431-442. (35) Coughlin, S. R. Thrombin Signalling and Protease-Activated Receptors. Nature 2000, 407, 258-264. (36) Brunauer, S.; Emmett, P. H.; Teller, E. Adsorption of Gases in Multimolecular Layers. J. Am. Chem. Soc. 1938, 60, 309-319. (37) Gelb, L. D.; Gubbins, K. E. Characterization of Porous Glasses: Simulation Models, Adsorption Isotherms, and The Brunauer− Emmett− Teller Analysis Method. Langmuir 1998, 14, 2097-2111. (38) Sing, K. The Use of Nitrogen Adsorption for the Characterisation of Porous Materials. Colloid Surf. A-Physicochem. Eng. Asp. 2001, 187, 3-9. (39) Chang, I. S.; Lee, C. H.; Ahn, K. H. Membrane Filtration Characteristics in MembraneCoupled Activated Sludge System: The Effect of Floc Structure on Membrane Fouling. Sep. Sci. Technol. 1999, 34, 1743-1758.

ACS Paragon Plus Environment

26

Page 27 of 27

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

ACS Applied Materials & Interfaces

TOC

ACS Paragon Plus Environment

27