Dual-Thermoresponsive Phase Behavior of Blood CompatibleZwitterionic Copolymers Containing Nonionic Poly(N-isopropyl

acrylamide)

Yung Chang,*,† Wen-Yih Chen,‡ Wetra Yandi,† Yu-Ju Shih,† Wan-Ling Chu,†

Ying-Ling Liu,† Chih-Wei Chu,§ Ruoh-Chyu Ruaan,‡ and Akon Higuchi‡

R&D Center for Membrane Technology and Department of Chemical Engineering, Chung Yuan ChristianUniversity, Jhong-Li, Taoyuan 320, Taiwan, Department of Chemical and Materials Engineering, National

Central University, Jhong-Li, Taoyuan 320, Taiwan, and Research Center for Applied Sciences, AcademiaSinica 128 Sec. 2, Academia Road, Nankang, Taipei 11529, Taiwan

Received February 17, 2009; Revised Manuscript Received May 29, 2009

Thermoresponsive statistical copolymers of zwitterionic sulfobetaine methacrylate (SBMA) and nonionicN-isopropylacrylamide (NIPAAm) were prepared with an average molecular weight of about 6.0 kDa viahomogeneous free radical copolymerization. The aqueous solution properties of poly(SBMA-co-NIPAAm) weremeasured using a UV-visible spectrophotometer. The copolymers exhibited controllable lower and upper criticalsolution temperatures in aqueous solution and showed stimuli-responsive phase transition in the presence of salts.Regulated zwitterionic and nonionic molar mass ratios led to poly(SBMA-co-NIPAAm) copolymers having double-critical solution temperatures, where the water-insoluble polymer microdomains are generated by the zwitterioniccopolymer region of polySBMA or nonionic copolymer region of polyNIPAAm depending on temperature. Ahigh content of the nonionic polyNIPAAm in poly(SBMA-co-NIPAAm) exhibits nonionic aggregation at hightemperatures due to the desolvation of polyNIPAAm, whereas relatively low content of polyNIPAAm inpoly(SBMA-co-NIPAAm) exhibits zwitterionic aggregation at low temperatures due to the desolvation ofpolySBMA. Plasma protein adsorption on the surface coated with poly(SBMA-co-NIPAAm) was measured witha surface plasmon resonance (SPR) sensor. The copolymers containing polySBMA above 29 mol % showedextremely low protein adsorption and high anticoagulant activity in human blood plasma. The tunable and switchablethermoresponsive phase behavior of poly(SBMA-co-NIPAAm), as well as its high plasma protein adsorptionresistance and anticoagulant activity, suggests a potential for blood-contacting applications.

Introduction

Blood compatibility is highly recommended for blood-contacting materials in important biomedical applications, suchas antithrombogenic implants, hemodialysis membranes, andbiosensors.1-6 However, only a small number of synthesizedbiomaterials are regarded as good blood-compatible candidates.Zwitterionic polymers containing the pendant groups of phos-phobetaine, sulfobetaine, and carboxybetaine have receivedgrowing attention for use in the new generation of blood-contacting materials because of their good plasma proteinresistance.4,7-12 In the last several years, poly(sulfobetainemethacrylate) (polySBMA) with a methacrylate main chain andan analogue of the taurine betaine pendant group (CH2CH2N+-(CH3)2CH2CH2CH2SO3

-) has become the most widely studiedzwitterionic polymer due to its ease of synthetic prepara-tion.4,8,10,12-16 It was reported that the surfaces grafted withpolySBMA reduced fibrinogen adsorption to a level comparablewith the adsorption on poly(ethylene glycol)-grafted films inour previous studies.8,15 We also grafted a dense polymer brushof polySBMA on a gold surface via surface-initiated atomtransfer radical polymerization and suggested that zwitterionic

polySBMA is an effective and stable nonbiofouling material toprovide a surface for use in human blood and implants.4

In general, intelligent biocompatible polymers can representdiverse forms, which might be dissolved as unimers or micellesin an aqueous medium, adsorbed or grafted on aqueous-solidinterfaces, or cross-linked in the form of physical or chemicalhydrogels.17-20 These polymers can undergo large physicalchain conformation changes to small environmental stimuli ofphysical, chemical, or biochemical nature. Poly(N-isopropyl-acrylamide) (polyNIPAAm) is the most widely studied ther-moresponsive polymer.20-27 This nonionic polymer undergoesa sharp hydrophilic-hydrophobic transition in water at 32 °C;this temperature is called the lower critical solution temperature(LCST).25 The solution properties of zwitterionic polymers differconsiderably from those of nonionic polymers. In aqueoussolution, polySBMA, like other zwitterionic polymers, exhibitsan upper critical solution temperature (UCST) that increaseswith the molar content.28 This is attributed to the charge-chargeor dipole-dipole interactions between the betaine groups. Thecombination of the UCST of the zwitterionic polymers withthe LCST of the polyNIPAAm displays intriguing temperature-induced self-assembly behavior of different types of polymericaggregates in aqueous solution.24,29-34

To further develop the zwitterionic-based materials forbiomedical applications, we were inspired to study smartpolymer systems carrying both controllable biocompatibility andstimuli-responsive functions. Recently, some research worksreported physical micellization of synthesized diblock copoly-

* To whom correspondence should be addressed. E-mail: ychang@cycu.edu.tw.

† Chung Yuan Christian University.‡ National Central University.§ Research Center for Applied Sciences.

Biomacromolecules XXXX, xxx, 000 A

10.1021/bm900208u CCC: $40.75 XXXX American Chemical Society

Dow

nloa

ded

by T

AIW

AN

CO

NSO

RT

IA o

n Ju

ly 2

6, 2

009

Publ

ishe

d on

Jul

y 2,

200

9 on

http

://pu

bs.a

cs.o

rg |

doi:

10.1

021/

bm90

0208

u

mers with thermoresponsive and zwitterionic properties.29,30,32

These block copolymers were found to exhibit double ther-mosensitive phase transition of LCST and UCST behaviors inwater. However, these studies did not extend to the use orevaluation of these copolymers as biological or biomedicalmaterials. An early work reported the biocompatible nature ofSBMA-based statistical copolymer coatings as potential anti-bioadherent surface coatings.35 In this work, an interestingcombination of the zwitterionic polySBMA and nonionicthermoresponsive polyNIPAAm was studied as an example ofintelligent biocompatible polymers, especially for their human-blood-contacting properties. The effect of copolymer concentra-tions, solvent polarities, and ionic strengths on the LCST andUCST of poly(SBMA-co-NIPAAm) in aqueous solution withdifferent monomer ratios of SBMA and NIPAAm are discussedin detail. This study also demonstrates the adsorption of plasmaproteins on the surface coated with poly(SBMA-co-NIPAAm)from human blood plasma via surface plasmon resonance (SPR),and the anticoagulant activity of the copolymers in a platelet-poor plasma solution by recalcified plasma clotting tests. Thiswork is aimed at addressing two important issues of poly(SBMA-co-NIPAAm), that is, (i) systematic measurement of the LCSTor UCST characteristics from various copolymer compositionsat different copolymer concentrations, solvent polarities, andionic strengths and (ii) in vitro evaluation of blood compatibilityof the fully coated copolymer surface and copolymer suspensionusing human plasma solution.

Materials and Methods

Materials. [2-(Methacryloyloxy)ethyl]dimethyl(3-sulfopropyl)-am-monium hydroxide (sulfobetaine methacrylate, SBMA) macromonomerwas purchased from Monomer-Polymer and Dajac Laboratories, Inc.,U.S.A. N-Isopropylacrylamide (NIPAAm) from Sigma-Aldrich wasrecrystallized with hexane. Ammonium persulfate (APS), N,N,N′,N′-tetraethylmethylenediamine (TEMED), copper(I) bromide (99.999%),2-bromoisobutyryl bromide (BIBB, 98%), pyridine (98%), 2-hydroxy-ethyl acrylate (97%), 2,2-bipyridine (BPY, 99%), triethylamine (99%),tetrahydrofuran (THF HPLC grade), and ethanol (absolute 200 proof)were purchased from Sigma-Aldrich. 1-Undecanethiol (99+%), (1-mercapto-11-undecyl)tetra(ethylene glycol) (99+%), and 11-mercapto-1-undecanol (99+%) were purchased from Asemblon INC in Redmond,Washington, U.S.A. Fibrinogen (fraction I from human plasma),γ-globulin (fractions II, III, 99%), and human serum albumin (HSA,96-99%) were purchased from Sigma Chemical Co. Acetone andmethanol were of analytical grade, purchased from Sigma ChemicalCo. Deionized water (DI water) used in experiments was purified usinga Millipore water purification system with a minimum resistivity of18.0 MΩ ·m. THF for reactions and washings was dried by sodiumbefore use. ω-mercaptoundecyl bromoisobutyrate was synthesizedthrough a reaction of BIBB using a method published previously by

our group.4,101H NMR (300 MHz, CDCl3): 4.15 (t, J ) 6.9, 2H, OCH2),2.51 (q, J ) 7.5, 2H, SCH2), 1.92 (s, 6H, CH3), 1.57-1.72 (m, 4H,CH2), and 1.24-1.40 (m, 16H, CH2).

Preparation of Poly(SBMA-co-NIPAAm) in Aqueous Solu-tion. A total solid content of 8 wt % for different mass ratios of SBMAand NIPAAm (Table 1) was dissolved in 10.2 mL of DI water, andnitrogen was bubbled through to remove residual oxygen. Thecopolymerization of poly(SBMA-co-NIPAAm) was initiated using 8.0mg of APS and 8.0 mg (0.011 mL) of TEMED. The relative molarratio of [APS]/[TEMED] was 1:2. The reaction was stirred underpositive nitrogen pressure for 6 h at 23 °C. After polymerization, theresulting reaction solution was cooled to 4 °C for 3 h and then addedslowly into acetone and redissolved into DI water repeatedly toprecipitate the polymer out of the reaction solution and to removeresidual chemicals. Finally, the copolymer was dried in a vacuum ovenat room temperature (23 °C) to yield a white powder.

Preparation of Self-Assembled Monolayers on Gold Sur-faces. Two self-assembled monolayers (SAMs) were formed on thesubstrates: (1) methyl-terminated (CH3) and (2) initiator ω-mercap-toundecyl bromoisobutyrate (Br) SAMs. Glass chips were first coatedwith an adhesion-promoting chromium layer (thickness 2 nm) and asurface plasmon active gold layer (48 nm) by electron beam evaporationunder vacuum. Before SAM preparation, the gold-coated glass substratewas cleaned by washing with pure ethanol and DI water in sequence,dried with N2, then left in a UV light cleaner for 20 min at a sourcepower of 110 W, followed by rinsing with DI water and ethanol, andfinally dried again by N2. For preparation of CH3-SAMs, the cleanedchip was soaked in a 2 mM ethanol solution of 1-undecanethiol or(1-mercapto-11-undecyl) and tetra(ethylene glycol) thiols for 24 h toform SAMs on the gold surface, and the chip was rinsed in sequencewith ethanol and water and then dried in a stream of N2. For thepreparation of an initiator SAM on a gold surface, the cleaned chipwas soaked in a 2 mM ethanol solution of ω-mercaptoundecylbromoisobutyrate for 24 h to form Br-SAMs on the gold surface andthen rinsed with pure ethanol followed by THF and dried in a streamof N2.4,10

Preparation of SBMA and NIPAAm Polymer Brushes onGold Surfaces. Dense polymer brushes of polySBMA and polyNIPAAmon an SPR sensor chip were achieved via the surface-initiated ATRPmethod, which were prepared by the following method, as reportedpreviously.4,10,36 Polymer brushes were polymerized on gold substrateswith immobilized initiators of Br-SAMs based on our previousreports.4,10 The reaction solutions of CuBr and BPY were first placedinto a sealed glass reactor in a drybox under nitrogen atmosphere. 200mM of degassed solution (pure water and methanol at a 1:3 volumeratio) with SBMA or NIPAAm monomers was transferred to the reactor,and the gold surface with immobilized initiators was then placed intothe reactor under nitrogen. After polymerization, the substrate wasremoved and rinsed with ethanol and water, and the samples were keptin water overnight. The prepared substrates were usually rinsed withPBS buffer to remove unbound polymers before any experiments. Thethickness of the substrates was measured by ellipsometry. The thickness

Table 1. Characteristic Data of Poly(SBMA-co-NIPAAm) Statistical Copolymers

reaction ratios ofcomonomersa (wt %)

compositions ofcopolymersb (mol %)

characterization ofcopolymersc (g/mol)

critical solutiontemperatured (°C)

sample ID SBMA NIPAAm polySBMA polyNIPAAm Mw Mw/Mn UCST LCST

S100-N0 100 0 100.0 0.0 6336 3.6 27S70-N30 70 30 45.3 54.7 5818 3.4 18S50-N50 50 50 29.0 71.0 6262 3.5 15 41S30-N70 30 70 15.0 85.0 6649 3.5 37S0-N100 0 100 0.0 100.0 6951 3.5 32a Reaction mass ratios of SBMA and NIPAAm monomers used with fixed total monomer mass amount of 0.8 g in the prepared reaction solution. b The

composition of the poly(SBMA-co-NIPAAm) copolymers was estimated by 1H NMR in D2O from the relative peak area of (CH3)2N+ proton resonance ofthe polySBMA side groups at δ ) 3.2 ppm and that of the methyl proton resonance of the polyNIPAAm isopropyl groups at δ ) 1.14 ppm. c Weight-averagemolecular weights (Mw) and molecular weight distributions (Mw/Mn) were estimated by GPC and calibrated with PEO. d UCST and LCST were determinedby reading the absorbance at 230 nm on a UV-visible spectrophotometer.

B Biomacromolecules, Vol. xxx, No. xx, XXXX Chang et al.

Dow

nloa

ded

by T

AIW

AN

CO

NSO

RT

IA o

n Ju

ly 2

6, 2

009

Publ

ishe

d on

Jul

y 2,

200

9 on

http

://pu

bs.a

cs.o

rg |

doi:

10.1

021/

bm90

0208

u

of polySBMA brushes and polyNIPAAm brushes was able to becontrolled from 10 to 15 nm, as measured by ellipsometry. The chemicalcharacterization of polymer brushes is described in detail via X-rayphotoelectron spectroscopy and surface contact angle measurementsin our previous publications.4,10

Characterization of the Copolymers. The structure of poly(SBMA-co-NIPAAm) statistical copolymers was characterized by 1H NMRspectra using a 500 MHz spectrometer and D2O as the solvent. Thecomposition of the poly(SBMA-co-NIPAAm) copolymers was esti-mated by 1H NMR in D2O from the relative peak area of (CH3)2N+

proton resonance of the polySBMA side groups at δ ) 3.2 ppm andthat of the methyl proton resonance of the polyNIPAAm isopropylgroups at δ ) 1.14 ppm. Molecular weights of prepared statisticalcopolymers were determined by aqueous gel permeation chromatog-raphy (GPC) using one column of Viscogel TM GMPWXL K0105 (themolecular weight range was from 500 Da to 800 kDa) connected to aWaters 2414 refractive index detector at a 1.0 mL/min flow rate and acolumn temperature of 23 °C. The eluent was an aqueous solutioncomposed of 0.1 M NaCl at pH 7.4. Poly(ethylene oxide) (PEO)standards from Polymer Standard Service, Inc. (Warwick, U.S.A.) wereused for calibration.

Determination of UCST and LCST for the Copolymers. Thephase transition temperatures (UCST or LCST) of the aqueous solubleand insoluble copolymers at various copolymer concentrations, solventpolarities, and ionic strengths of the solution conditions were determinedby reading the absorbance at 230 nm on a UV-visible spectropho-tometer using SpectraMax M5 from Molecular Devices. A givenconcentration of prepared copolymer was dissolved in an aqueoussolution at 23 °C. The temperature of polymer solution in a well wascontrolled using a heating circulator and a cooler. The temperature wasfirst cooled from 23 to 1 °C and then gradually raised from 1 to 70 °C,and the absorbance value was read for every 1 °C increment of eachsample after a 10 min thermal equilibration. The UCST or LCST in aparticular solution condition was defined as the temperature where themaximum slope for the absorbance versus temperature curve occurs.The confidence in the accuracy of the measured values of UCST orLCST at 230 nm was justified by the phase transition temperatures ofpolyNIPAAm determined at 600 nm. It was shown that the LCST valuesof polyNIPAAm determined by reading the absorbance at 230 and 600nm were identical over a range of concentrations of sodium chloridefrom 0.01 to 2.0 M, consistent with the observation by Hoffman et al.that an increase in the sodium chloride concentration led to a decreasein LCST values of polyNIPAAm.42

Plasma Protein Adsorption. A custom-built SPR biosensor with afour-channel Teflon flow cell, designed by the Institute of Photonicsand Electronics Academy Sciences (Prague, Czech Republic) basedon wavelength interrogation, was used to monitor protein adsorptionon the coated substrate.37 An SPR chip was attached to the base of theprism, and optical contact was established using a refractive indexmatching fluid (Cargille). A protein solution of 1.0 mg/mL humanfibrinogen in a phosphate buffer saline (PBS, 0.15 M, pH 7.4) wasdelivered to the surfaces at a flow rate of 0.05 mL/min at 37 °C. Inthis study, platelet poor plasma (PPP) solution containing plasmaproteins was also tested on the coated substrate. A surface-sensitiveSPR detector was used to monitor protein-surface interactions in realtime. The wavelength shift was used to measure the change in thesurface adsorption amount (mass per unit area). The calibration of thewavelength shift from SPR data associated with the amount of adsorbedprotein was calculated based on equations established by Campbell andco-workers.38 The calibration follows the standard calculation for thesame custom-built SPR system, with a 1 nm wavelength shift resultingin an SPR response equivalent to about 15 ng/cm2 of adsorbedproteins.7,38

Plasma Clotting Time. The anticoagulant activity of preparedcopolymers in this work was evaluated by testing plasma clotting timein human plasma solution. PPP solution was prepared by centrifugationof the human blood at 3000 rpm for 10 min at 23 °C. Next, 160 µL of

the human PPP solution was mixed with the polymer solution (10 mg/mL, 46 µL) in a 96-well plate. The solution was then recalcified byaddition of calcium (1 M CaCl2, 4 µL) and agitation for 30 s. Twoincubation temperatures were tested, 23 and 37 °C. The clotting timeof the plasma was determined at the time when the onset of theabsorbance transition occurs by reading the absorbance at 660 nm usinga PowerWave microplate spectrophotometer with programmed tem-perature control. Each clotting time is an average value of five samplesfrom repeated measurements.

Results and Discussion

Three zwitterionic-based copolymers with differing SBMAmolar contents of copolymers (S70-N30, S50-N50, and S30-N70) were prepared from various monomer compositions in thereaction solution. Molecular weight distributions of the synthe-sized poly(SBMA-co-NIPAAm) copolymers from GPC werecalculated using Empower Pro from Waters, as shown in Table1. All copolymer samples were controlled with a similar averagemolecular weight of about 6.0 kDa and the same broadmolecular weight distribution (i.e., Mw/Mn ) 3.4-3.6). Theincreasing amount of NIPAAm monomers in the reactionsolution increased the molar mass ratio of polyNIPAAm in theprepared copolymer. The composition of the poly(SBMA-co-NIPAAm) copolymers was estimated by 1H NMR in D2O. Atypical spectrum for S50-N50 is shown in Figure 1. Resultsshowed that a pure poly(SBMA-co-NIPAAm) copolymer wasobtained. It is noted that the molar ratio of polySBMA in theprepared copolymers is only 42 mol % even though the amountof SBMA monomers used in the reaction solution is as high as70 wt %, indicating higher polymerized reactivity of NIPAAmmonomers than of SBMA monomers in water. For comparison,two homopolymers of polySBMA (S100-N0) and polyNIPAAm(S0-N100) were also synthesized as references. To determinethe soluble-insoluble phase transition of zwitterionic-basedcopolymers, the optical absorbance of dilute copolymer solutionwas measured using a UV-visible spectrophotometer withprecise temperature control from 1 to 70 °C. All preparedsamples exhibited clear phase transitions associated with thecompositions of copolymers, even though their molecular weightdistributions were poorly controlled by conventional free radicalpolymerization. It is generally acceptable that self-assembledmonolayers (SAMs) presenting hydrophobic methyl groupsusually induce large amounts of protein adsorption. Therefore,a self-assembly method was used to create one dense surfacewith CH3-SAMs as references for the SPR study. To characterizethe blood compatibility of zwitterionic-based copolymers, preparedcopolymers were first physically adsorbed onto the SPR sensorsurfaces covered by CH3-terminated SAMs, followed by the insitu evaluation of plasma protein adsorption on the surfaces withself-assembled poly(SBMA-co-NIPAAm) copolymers. For com-parison, the surface-initiated atom transfer radical polymerization(ATRP) method was also used to create two well-packed graftedsurfaces with polySBMA and polyNIPAAm polymer brushes asreferences.

Phase Transition Temperatures of Poly(SBMA-co-NIPAAm) Copolymers in Aqueous Solution. The phasetransition temperature of the copolymer solution was measuredusing UV-visible spectrophotometer coupled to a temperaturecontroller at 1 °C/10 min. No thermal hysteresis was observedat equilibrium temperature from heating or cooling the copoly-mer solution. The maximum slope of a phase transition curvefrom the change of absorbance (A) signal as a function oftemperature (d2A/dT2 ) 0) was determined by referring to theUCST or LCST that was recorded for each sample. As shown

Blood Compatible Zwitterionic Copolymers Biomacromolecules, Vol. xxx, No. xx, XXXX C

Dow

nloa

ded

by T

AIW

AN

CO

NSO

RT

IA o

n Ju

ly 2

6, 2

009

Publ

ishe

d on

Jul

y 2,

200

9 on

http

://pu

bs.a

cs.o

rg |

doi:

10.1

021/

bm90

0208

u

in Figure 2, zwitterionic polySBMA exhibited a UCST in waterat 27 °C, and nonionic polyNIPAAm exhibited an LCST inwater at 32 °C, as reported previously.29 At temperatures belowthe UCST, polySBMA (S100-N0) is considered to exist as acollapsed coil and precipitates in water due to strong mutualintra- and intermolecular associations of the zwitterionic groupsby electrostatic interaction.40,41 At temperatures above the LCST,polyNIPAAm (S0-N100) chains become more hydrophobic, andthe hydrogen bonds with water molecules weaken, indicatingthe collapse of the polyNIPAAm coils and the precipitationof the polymer.25,43 The soluble-insoluble phase transitions ofcopolymers S70-N30, S50-N50, and S30-N70 in dilute aqueoussolution were studied as functions of temperature to observethe influence of differing molar contents of polySBMA andpolyNIPAAm combinations on the UCST and LCST ofcopolymer solutions. Interestingly, there appear both a UCST(15 °C) and an LCST (41 °C) of a doubly thermoresponsivecopolymer S50-N50 from the combination of 29.0 mol %polySBMA and 71.0 mol % polyNIPAAm in the same

copolymer chain. This indicates that copolymer S50-N50 inaqueous solution is soluble from 15 to 41 °C but insoluble below15 °C and above 41 °C at a copolymer concentration of 5wt %. It should be noted that all prepared samples exhibit aclear shift of phase transitions strongly associated with thecompositions of copolymers, even though their molecular weightdistributions are poorly controlled (Table 1). As illustrated inScheme 1, based on the results from the UV-visible spectro-photometer, the dependence of the solubility and insolubilityof the copolymers on temperature can be explained by intra-and intermolecular interactions between copolymer chains. Inthe case of the homopolymer polySBMA, copolymers S70-N30and S50-N50 also have the ability to exhibit a UCST in water,which decreases with decreasing molar content of polySBMAin the copolymer chain. This is attributed to the interaction levelof mutual electrostatic attraction by ion pairings between theammonium cation and the sulfo-anion of the zwitterionicsulfobetaine groups.42 Thus, the copolymer precipitates with thezwitterionic polySBMA segment associations as temperaturefalls below the UCST. In contrast, in the case of the homopoly-mer polyNIPAAm, copolymers S50-N50 and S30-N70 exhibitan LCST in water that also notably depends on the molar contentof polyNIPAAm in the copolymer chain. It was observed thatthe LCSTs of the copolymers are obviously higher than that ofthe homopolymer polyNIPAAm, as listed in Table 1. As themolar content of polyNIPAAm decreases in the copolymer, thecollapse temperature increases for copolymer precipitates withthe nonionic polyNIPAAm segment associations in water. Thisis attributed to the less hydrophobic interactions betweenpolyNIPAAm side chains induced by the isopropyl groups. Inthe intermediate temperature range, that is, above the UCSTand below the LCST, copolymers were observed to exist assoluble unimers in water, but beyond this range, copolymerswere considered to precipitate as collapsed associations in water.As shown in the simplified model proposed in Scheme 1, webelieve that this doubly thermoresponsive solubility behaviorof the prepared copolymer results from the formation of inter-and intramolecular electrostatic interactions by SBMA segments

Figure 1. 1H NMR spectrum of the copolymer S50-N50 in D2O.

Figure 2. Absorbance of copolymer solutions as a function oftemperature for the various samples of (a) S100-N0, (b) S70-N30,(c) S50-N50, (d) S30-N70, and (e) S0-N100 at the polymer concen-tration of 5 wt %.

D Biomacromolecules, Vol. xxx, No. xx, XXXX Chang et al.

Dow

nloa

ded

by T

AIW

AN

CO

NSO

RT

IA o

n Ju

ly 2

6, 2

009

Publ

ishe

d on

Jul

y 2,

200

9 on

http

://pu

bs.a

cs.o

rg |

doi:

10.1

021/

bm90

0208

u

of zwitterionic sulfobetaine groups and intramolecular hydro-phobic interactions by NIPAAm segments of nonionic isopropylgroups.

We further studied the dependence of UCSTs and LCSTs oncopolymer concentration in aqueous solution at a concentrationof e5 wt %. In Figure 3, UCSTs of homopolymer S100-N0and copolymer S70-N30 were found to be dependent on boththe polymer concentration and molar ratio of polySBMA andpolyNIPAAm. In general, as the concentration increases, thedistance between polymer chains is reduced, which contributesto the formation of more intermolecular electrostatic attractionsof ionic pairings of opposite charges between zwitterionicgroups. Therefore, in the case of a high concentration of polymersolution, a high temperature is required to provide thermal

energy sufficient to break the electrostatic bonding betweeninterchain ionic pairs and to form the soluble polymer chainsdispersed in water. In contrast, curve (d) in Figure 3 shows thatLCSTs of homopolymer S0-N100 are basically independent ofthe polymer concentration. The same tendency in aqueoussolution at a concentration of e1 wt % was also found in theLCST of the homopolymer of polyNIPAAm reported by Narainand his co-workers.44 It is well established that the origin ofthe LCST behavior of polyNIPAAm arises from the release ofordering water molecules and the formation of intrachainhydrophobic interactions associated with the side chain isopropylmoieties as the temperature increases above a critical point.Thus, the results indicate that the intramolecular hydrophobicinteraction by nonionic isopropyl groups in the same polymerchain becomes so dominant that polyNIPAAm precipitationoccurs at a high temperature. Interestingly, it was found thatLCSTs of copolymer S30-N70 exhibit polymer concentrationdependence, decreasing with an increase in polymer concentra-tion. This might be attributed to the zwitterionic sulfobetainegroups contributing to the increase in the hydration capacity ofa copolymer chain while the concentration is below 5 wt %.

Effect of Solvent Polarity and Ionic Strength on theUCSTs and LCSTs of Poly(SBMA-co-NIPAAm) CopolymerSolutions. The effects of solvent polarity on the UCSTs andLCSTs of prepared poly(SBMA-co-NIPAAm) solutions wereinvestigated using the addition of methanol to water as a casestudy to regulate the order of decreasing polarity at a fixedpolymer concentration of 5 wt %. The results are shown inFigure 4. It was found that UCST and LCST strongly dependedon the polarity of the solvent used. UCSTs of S100-N0 andS70-N30 increased with a decrease in solvent polarity (i.e.,increase in methanol content). In general, as the solvent polarityis reduced, the surrounding dielectric property of the solutionmedium decreases, which enhances the formation of strongerintra- and intermolecular electrostatic interactions of opposite

Scheme 1. Simplified Model of the Temperature Dependence of the Copolymer Solubility and Insolubility in Aqueous Solution for the Caseof Copolymer S50-N50

Figure 3. Effect of copolymer content on the UCST or LCST of thepoly(SBMA-co-NIPAAm) solution for the samples of (a) S100-N0, (b)S70-N30, (c) S30-N70, and (d) S0-N100, where UCST or LCST wasobtained from the absorbance transition with temperature.

Blood Compatible Zwitterionic Copolymers Biomacromolecules, Vol. xxx, No. xx, XXXX E

Dow

nloa

ded

by T

AIW

AN

CO

NSO

RT

IA o

n Ju

ly 2

6, 2

009

Publ

ishe

d on

Jul

y 2,

200

9 on

http

://pu

bs.a

cs.o

rg |

doi:

10.1

021/

bm90

0208

u

charges between zwitterionic groups. Thus, the results of curves(a) and (b) in Figure 4 show that a high temperature is requiredto provide high thermal energy to disrupt the intra- andinterchain electrostatic interactions of polySBMA associationsin a less polar environment. This also explains why the solublezwitterionic polymers can be generally achieved by the additionof a stronger polar solvent such as water. In contrast, the LCSTsof S30-N70 and S0-N100 gradually decreased with a decreasein solvent polarity. It was found that the LCST of aqueouspolyNIPAAm solutions shifts to a lower temperature whenmethanol is added, which was observed in previous studies.45,46

This enhanced phase separation of polyNIPAAm in water-methanol mixtures is known as cononsolvency. This propertywas retained at the lower SBMA molar content of S30-N70but was lost at the higher SBMA molar content of S70-N30.On the basis of the hydration of polymer chains, the dependenceof LCST on solvent polarity can be explained by the followingmechanism. The insoluble polyNIPAAm in water occurs as thetemperature increases above a LCST due to the thermallyinduced disruption of H-bonding water molecules hydratedaround isopropyl groups. Thus, it is reasonable to consider thatless hydration around polymer chains in a less polar aqueoussolution results in a decrease in the LCST of the polymersolution.

The effects of the ionic strength on the UCSTs and LCSTsof poly(SBMA-co-NIPAAm) solutions were further evaluatedto investigate the solubility characteristics at a fixed polymerconcentration of 5 wt %. The ionic strength of the aqueousmedium was adjusted by dissolving the ionic salt NaCl into DIwater at appropriate concentrations ranging from 0.01 to 2.0M. The degree of diminution of the UCSTs of homopolymerS100-N0 and copolymer S70-N30 are dependent on both thesalt concentration and the molar content of polySBMA inthe polymer chain, as shown in Figure 5. Below the UCST, theformation of intra- and interchain ionic contacts betweenzwitterionic groups causes the polymer chains to collapse andprecipitate from the solution. The variation in the UCSTs ofthe prepared polymer solution is found to be sensitive to theslight change of low salt concentrations from 0.01 to 0.07 M inDI water. A good linear relationship can be obtained betweenthe critical solution temperature and the sodium chlorideconcentration. PolySBMA has the ability to exhibit an unusualantipolyelectrolyte behavior in the presence of salt ions that

notably increases with the ionic strength in the aqueous solution,as in the case of other zwitterionic polymers.40 The dissolutionprocess of polySBMA upon the addition of electrolytes isattributed to the salt ions screening the net attractive electrostaticinteractions between the zwitterionic polymer chains, whichleads to polymer chain expansion. It can be seen that for eachprepared sample, the LCSTs of the polymer solution sharplydecrease at a certain sodium chloride concentration, which is asimilar tendency found in polyNIPAAm.42 The mechanismresponsible for the salt-induced changes in the soluble-insolubleproperties of nonionic polyNIPAAm solutions has been reportedby Hoffman and his co-workers.42 They showed that thedecreasing trend in the transition temperature was correlated tothe ion-water interaction of salt ions, the so-called salting outeffect, suggesting that the water structure around the polymerchains of polyNIPAAm segments affected by ions results inthe LCST shift.47

Blood Compatibility of Coated Poly(SBMA-co-NIPAAm)Copolymers in Human Plasma Solution. Horbett et al. showedthat the adhesion and activation of platelets from the bloodstreamwas correlated with the adsorption of proteins on surfaces,especially fibrinogen adsorption.5,6,43 For example, for surfacesin contact with blood, even 10 ng/cm2 of adsorbed fibrinogenmay introduce a full-scale blood platelet adhesion and lead tothrombosis and embolism at the blood contact side of implantdevices. The extent of this effect was evaluated by the inspectionof blood compatibility of the surface coated with poly(SBMA-co-NIPAAm) copolymer.

The physical adsorption of the prepared copolymers ontohydrophobic CH3-SAM surfaces was performed to study theantibiofouling characteristics of copolymer-coated surfaces.Nonionic isopropyl groups in the polyNIPAAm chains wereused as the hydrophobic moiety of the copolymers that mediatehydrophobic interaction with CH3-SAM surfaces. It was thenfollowed by the in situ evaluation of plasma protein adsorptionon the surfaces with coated copolymers by SPR measurements.Figure 6 shows a typical SPR sensorgram in the case of theadsorption of S30-N70 copolymers, followed by the in situevaluation of human fibrinogen adsorption. In the first stage ofthe copolymer coated surface formation, the amount of adsorbedcopolymer is defined as the SPR wavelength difference (∆n1,nm) between the two baselines established before and after

Figure 4. Effect of methanol content on the UCST or LCST of thepoly(SBMA-co-NIPAAm) solution for the samples of (a) S100-N0, (b)S70-N30, (c) S30-N70, and (d) S0-N100 at the polymer concentrationof 5 wt %.

Figure 5. Effect of NaCl content on the UCST or LCST of thepoly(SBMA-co-NIPAAm) solution for the samples of (a) S100-N0 and(b) S70-N30 at the polymer concentration of 5 wt %, and (c) S30-N70 and (d) S0-N100 at the polymer concentration of 3 wt %.

F Biomacromolecules, Vol. xxx, No. xx, XXXX Chang et al.

Dow

nloa

ded

by T

AIW

AN

CO

NSO

RT

IA o

n Ju

ly 2

6, 2

009

Publ

ishe

d on

Jul

y 2,

200

9 on

http

://pu

bs.a

cs.o

rg |

doi:

10.1

021/

bm90

0208

u

copolymer adsorption. The saturated adsorbed amounts for eachprepared copolymer on the surfaces can be obtained by thecontrol of mass concentrations of copolymer solution above 0.1mg/mL. In the second stage of the biofouling evaluation of fullycoated copolymer surface, the amount of protein adsorption isdefined as the SPR wavelength difference (∆n2, nm) betweenthe two baselines established before and after protein adsorption.

Real-time adsorption of human fibrinogen in buffered aqueoussolutions and human plasma proteins in blood plasma solutionsonto copolymer coated surfaces was monitored using SPR at37 °C (human body temperature). CH3-SAMs and two ho-mopolymer brushes of polyNIPAAm and polySBMA were usedas references. The polymer brushes were prepared on goldsurfaces via surface-initiated atom transfer radical polymeriza-tion following the method reported previously.4,10 A dilutedsolution containing 10% (v/v, in PBS) plasma proteins fromplatelet poor plasma was used in this measurement to reducethe effects of plasma viscosity in the laminar flow channel. Theadsorption amounts from the SPR for the protein adsorption ondifferent surfaces are shown in Figure 7. It is known thatCH3-SAMs presenting hydrophobic methyl groups usuallyinduce large amounts of protein adsorption,4,8 which can alsobe observed in Figure 7. It was found that polySBMA brushesare highly resistant to nonspecific adsorption for humanfibrinogen and human plasma proteins at 37 °C, while hydro-phobic CH3-SAMs and polyNIPAAm brushes show highprotein adsorption. We observed significant decreases in adsorp-tion of proteins on copolymer coated surfaces with S30-N70,S50-N50, and S70-N30 as compared to those on surfaces ofCH3-SAMs and polyNIPAAm brushes. The copolymer-coatedsurfaces, even with a low molar ratio of 15% polySBMA inS30-N70 copolymers, reduced the protein adsorption to a levelcomparable with the adsorption on the surface grafted withpolySBMA homopolymer brushes. The adsorbed amounts ofplasma proteins on all copolymer-coated surfaces are found tobe less than 5 ng/cm2. On the basis of previous reports fromHorbett et al., it is believed that reducing plasma proteinadsorption levels to below 10 ng/cm2 on biomaterial surfacescan effectively prevent the adhesion and activation of plateletsfrom the bloodstream.6 This result suggests that a surface coated

with a copolymer prepared from a combination of the zwitter-ionic polySBMA and nonionic thermoresponsive polyNIPAAmhas the potential to provide excellent anticoagulant activity inhuman blood.

Poly(SBMA-co-NIPAAm) was directly incubated with humanplasma to inspect the effect of direct contact activation oncopolymer-induced plasma clotting by the evaluation of theirrecalcified plasma clotting time. Normal human PPP solutionwas used in this work to minimize the effects of pro-coagulantactivation. All polymer samples were inserted into the recalcifiedhuman PPP solution in a 96-well plate at a room temperatureof 23 °C and at human body temperature of 37 °C. CommercialPEG with a molecular weight of 4.2 kDa and a polydispersityof 1.1 was used for comparison due to the lack of anticoagulantactivity induced by PEG, as reported in a previous study,39 withthe results shown in Figure 8. It is generally known that thehuman plasma clotting time will be prolonged with a decreasein environmental temperature control, which is consistent withour results, shown in Figure 8. The plasma clotting time forthe recalcified plasma solutions in blank wells was detected tohave an upper limit value of about 40 min at 23 °C and of about

Figure 6. Schematic illustration of the adsorption of S50-N50 copolymers onto a CH3-terminated SAM surface followed by an in situ evaluationof fibrinogen adsorption on the surface with self-assembled poly(SBMA-co-NIPAAm).

Figure 7. Adsorption of 1 mg/mL fibrinogen and 10% human plasmain PBS buffer on CH3-SAMs, surfaces grafted with polyNIPAAmbrushes and polySBMA brushes, and surfaces coated with S30-N70,S50-N50, and S70-N50 at 37 °C. A 1 nm wavelength shift in SPR isequivalent to 15 ng/cm2 adsorbed proteins.

Blood Compatible Zwitterionic Copolymers Biomacromolecules, Vol. xxx, No. xx, XXXX G

Dow

nloa

ded

by T

AIW

AN

CO

NSO

RT

IA o

n Ju

ly 2

6, 2

009

Publ

ishe

d on

Jul

y 2,

200

9 on

http

://pu

bs.a

cs.o

rg |

doi:

10.1

021/

bm90

0208

u

10 min at 37 °C. Almost no change in clotting time wasobserved with the absence or presence of 10 mg/mL PEG at 23°C and at 37 °C, which is in agreement with the reportedresults.39 It was found that S30-N70, S50-N50, and S70-N30copolymers exhibit prolonged plasma clotting time as well asanticoagulant activity in comparison to PEG for the plasmaclotting test at 23 °C, while polyNIPAAm (S0-N100) andpolySBMA (S100-N0) generate faster activation than do thepresence of other polymers or the absence of any polymers inplasma solution. The decrease in clotting time for S0-N100might be associated with the high fibrinogen adsorption on thepolyNIPAAm brush surface by the SPR measurement at 37 °C.We observed that plasma clotting time for S100-N0 is 75%higher than that for blank PS wells at 37 °C, while S100-N0loses its anticoagulant activity in plasma solution at 23 °C. Thisindicates that polySBMA has anticoagulant activity for prevent-ing clotting of plasma from human blood at 37 °C. Thedistinction in the dependence of anticoagulant activity ontemperature can be explained by the UCST of S100-N0 inaqueous solution, as shown in Table 1. Based on aforementionedphase behavior of polySBMA, a possible reason is that S100-N0 at 23 °C, which is below its UCST, exists as a collapsedcoil and precipitates out of polySBMA chains in plasma solutiondue to strong mutual intra- and intermolecular associations ofthe zwitterionic groups by electrostatic interaction. Thus, theloss of anticoagulant activity for polySBMA at 23 °C isattributed to dehydration of the zwitterionic moieties to form ahydrophobic domain, leading to plasma clotting from contactactivation in human plasma solution. However, poly(SBMA-co-NIPAAm) exhibits much higher anticoagulant activity incomparison to PEG and polySBMA, which depended on thetemperature of human blood plasma. The results support that ajudicious combination of the zwitterionic polySBMA andnonionic thermoresponsive polyNIPAAm leads to a newgeneration of blood-compatible biomaterial. As a result, thestudy shows for the first time that zwitterionic-based copolymerscontaining polySBMA and polyNIPAAm can be used to achievehigh blood compatibility while maintaining controllable ther-moresponsive functions.

Conclusions

We have demonstrated the soluble-insoluble phase transitionof poly(SBMA-co-NIPAAm) in aqueous media. We have foundthat the level of electrostatic or hydrophobic intra- and

intermolecular interactions between polymer chains is dominatedby the SBMA content. Both UCST and LCST behavior can beobtained, which depend strongly on the copolymer composition,solution concentration, solution polarity, and ionic strength. Itis worth emphasizing that appropriate control of zwitterionicand nonionic molar mass ratios leads to poly(SBMA-co-NIPAAm) copolymers that exhibit double-critical solutiontemperature in water, which means that the water-insolublepolymer-associated microdomains with zwitterionic or nonioniccharacters can be switched by a thermal stimulus. It was foundthat different ionic strengths, as well as their solution polarities,have different solubility-promoting effects on the preparedcopolymers. Furthermore, results showed that there is a remark-able reduction in the adsorption of plasma proteins from humanblood plasma onto the zwitterionic-based copolymer coatedsurface. Recalcified plasma clotting tests showed that theprepared copolymers exhibit an anticoagulant activity on humanblood plasma, which depends on the composition of copolymersand the temperature of the medium. It is demonstrated thatcopolymers containing a relatively low polySBMA content ofabout 29 mol % can lead to extremely low plasma proteinadsorption and high anticoagulant activity in human bloodplasma. This study suggests that a zwitterionic polySBMAcopolymer containing nonionic polyNIPAAm is a potentialthermoresponsive biomaterial to provide a coated surface withtunable stimuli for use in human blood and biomedical implants.

Acknowledgment. The authors express their sincere gratitudeto the Center-of-Excellence (COE) Program on MembraneTechnology from the Ministry of Education (MOE), R.O.C., tothe project Toward Sustainable Green Technology in the ChungYuan Christian University, Taiwan (CYCU-97-CR-CE), and tothe National Science Council (NSC 97-2120-M-008-002) fortheir financial support.

References and Notes(1) Hoffman, A. S. AdVances in Chemistry Series; American Chemical

Society: Washington, DC, 1982; p 3.(2) Ratner, B. D.; Hoffman, A. D.; Schoen, F. D.; Lemons, J. E.

Biomaterials Science, an Introduction to Materials in Medicine, 2nded.; Elsevier: Amsterdam, 2004.

(3) Ratner, B. D. Biomaterials 2007, 28, 5144–5147.(4) Chang, Y.; Liao, S. C.; Higuchi, A.; Ruaan, R. C.; Chu, C. W.; Chen,

W. Y. Langmuir 2008, 24, 5453–5458.(5) Shen, M. C.; Wagner, M. S.; Castner, D. G.; Ratner, B. D.; Horbett,

T. A. Langmuir 2003, 19, 1692–1699.(6) Kwak, D.; Wu, Y. G.; Horbett, T. A. J. Biomed. Mater. Res. 2005,

74A, 69–83.(7) Chen, S.; Zheng, J.; Li, L.; Jiang, S. J. Am. Chem. Soc. 2005, 127,

14473–14478.(8) Chang, Y.; Chen, S. F.; Zhang, Z.; Jiang, S. Y. Langmuir 2006, 22,

2222–2226.(9) Feng, W.; Brash, J. L.; Zhu, S. P. Biomaterials 2006, 27, 847–855.

(10) Zhang, Z.; Chen, S. F.; Chang, Y.; Jiang, S. Y. J. Phys. Chem. B2006, 110, 10799–10804.

(11) Chen, S. F.; Jiang, S. Y. AdV. Mater. 2008, 20, 335–338.(12) Zhang, Z.; Vaisocherova, H.; Cheng, G.; Yang, W.; Xue, H.; Jiang,

S. Biomacromolecules 2008, 9, 2686–2692.(13) Chang, Y.; Chen, S.; Yu, Q.; Zhang, Z.; Bernards, M.; Jiang, S.

Biomacromolecules 2007, 8, 122–127.(14) Yang, W.; Chen, S.; Cheng, G.; Vaisocherova, H.; Xue, H.; Li, W.;

Zhang, J.; Jiang, S. Langmuir 2008, 24, 9211–9214.(15) Zhang, Z.; Chao, T.; Chen, S. F.; Jiang, S. Y. Langmuir 2006, 22,

10072–10077.(16) Higuchi, A.; Hashiba, H.; Hayashi, R.; Yoon, B. O.; Hattori, M.; Hara,

M. ACS Symposium Series 876; American Chemical Society: Wash-ington, DC, 2004; Chapter 25.

(17) Hoffman, A. S.; Stayton, P. S.; Bulmus, V.; Chen, G. H.; Chen, J. P.;Cheung, C.; Chilkoti, A.; Ding, Z. L.; Dong, L. C.; Fong, R.; Lackey,C. A.; Long, C. J.; Miura, M.; Morris, J. E.; Murthy, N.; Nabeshima,Y.; Park, T. G.; Press, O. W.; Shimoboji, T.; Shoemaker, S.; Yang,

Figure 8. Plasma clotting time of recalcified platelet-poor plasma inthe presence of different polymers at 10 mg/mL. Clotting time for blankPS wells was 40 ( 2 min at 23 °C and 8 ( 0.5 min at 37 °C. Eachclotting time is an average value of five samples.

H Biomacromolecules, Vol. xxx, No. xx, XXXX Chang et al.

Dow

nloa

ded

by T

AIW

AN

CO

NSO

RT

IA o

n Ju

ly 2

6, 2

009

Publ

ishe

d on

Jul

y 2,

200

9 on

http

://pu

bs.a

cs.o

rg |

doi:

10.1

021/

bm90

0208

u

H. J.; Monji, N.; Nowinski, R. C.; Cole, C. A.; Priest, J. H.; Harris,J. M.; Nakamae, K.; Nishino, T.; Miyata, T. J. Biomed. Mater. Res.2000, 52, 577–586.

(18) Qiu, Y.; Park, K. AdV. Drug DeliVery ReV. 2001, 53, 321–339.(19) Jeong, B.; Gutowska, A. Trends Biotechnol. 2002, 20, 305–311.(20) Gil, E. S.; Hudson, S. A. Prog. Polym. Sci. 2004, 29, 1173–1222.(21) Chilkoti, A.; Dreher, M. R.; Meyer, D. E.; Raucher, D. AdV. Drug

DeliVery ReV. 2002, 54, 613–630.(22) Stile, R. A.; Burghardt, W. R.; Healy, K. E. Macromolecules 1999,

32, 7370–7379.(23) Sershen, S. R.; Westcott, S. L.; Halas, N. J.; West, J. L. J. Biomed.

Mater. Res. 2000, 51, 293–298.(24) Schilli, C. M.; Zhang, M. F.; Rizzardo, E.; Thang, S. H.; Chong, Y. K.;

Edwards, K.; Karlsson, G.; Muller, A. H. E. Macromolecules 2004,37, 7861–7866.

(25) Schild, H. G. Prog. Polym. Sci. 1992, 17, 163–249.(26) Pelton, R. AdV. Colloid Interface Sci. 2000, 85, 1–33.(27) Uchida, K.; Sakai, K.; Ito, E.; Kwon, O. H.; Kikuchi, A.; Yamato,

M.; Okano, T. Biomaterials 2000, 21, 923–929.(28) Schulza, D. N.; Agarwala, P. K.; Larabeea, J.; Kaladasa, J. J.; Sonia,

L.; Handwerkera, B.; Garnera, R. T. Polymer 1986, 27, 1734–1742.(29) Arotcarena, M.; Heise, B.; Ishaya, S.; Laschewsky, A. J. Am. Chem.

Soc. 2002, 124, 3787–3793.(30) Virtanen, J.; Arotcarena, M.; Heise, B.; Ishaya, S.; Laschewsky, A.;

Tenhu, H. Langmuir 2002, 18, 5360–5365.(31) Cai, Y. L.; Armes, S. P. Macromolecules 2004, 37, 7116–7122.(32) Maeda, Y.; Mochiduki, H.; Ikeda, I. Macromol. Rapid Commun. 2004,

25, 1330–1334.(33) Weaver, J. V. M.; Armes, S. P.; Butun, V. Chem. Commun. 2002,

2122–2123.

(34) Butun, V.; Liu, S.; Weaver, J. V. M.; Bories-Azeau, X.; Cai, Y.; Armes,S. P. React. Funct. Polym. 2006, 66, 157–165.

(35) Lowe, A. B.; Vamvakaki, M.; Wassall, M. A.; Wong, L.; Billingham,N. C.; Armes, S. P.; Lloyd, A. W. J. Biomed. Mater. Res. 2000, 52,88–94.

(36) Plunkett, K. N.; Zhu, X.; Moore, J. S.; Leckband, D. E. Langmuir2006, 22, 4259–4266.

(37) Boozer, C.; Ladd, J.; Chen, S.; Yu, Q.; Homola, J.; Jiang, S. Anal.Chem. 2004, 76, 6967–6972.

(38) Jung, L. S.; Campbell, C. T.; Chinowsky, T. M.; Mar, M. N.; Yee,S. S. Langmuir 1998, 14, 5636–5648.

(39) Zhang, Z.; Zhang, M.; Chen, S. F.; Horbett, T. A.; Ratner, B. D.;Jiang, S. Y. Biomaterials 2008, 29, 4285–4291.

(40) Lowe, A. B.; McCormick, C. L. Chem. ReV. 2002, 102, 4177-4189.

(41) Dimitrov, I.; Trzebicka, B.; Muller, A. H. E.; Dworak, A.; Tsvetanov,C. B. Prog. Polym. Sci. 2007, 32, 1275–1343.

(42) Park, T. G.; Hoffman, A. S. Macromolecules 1993, 26, 5045–5048.(43) Shen, M. C.; Martinson, L.; Wagner, M. S.; Castner, D. G.; Ratner,

B. D.; Horbett, T. A. J. Biomater. Sci., Polym. Ed. 2002, 13, 367–390.

(44) Housni, A.; Narain, R. Eur. Polym. J. 2007, 43, 4344–4354.(45) Winnik, F. M.; Ringsdorf, H.; Venzmer, J. Macromolecules 1990, 23,

2415–2416.(46) Schild, H. G.; Muthukumar, M.; Tirrell, D. A. Macromolecules 1991,

24, 948–952.(47) Schild, H. G.; Tirrell, D. A. J. Phys. Chem. 1990, 94, 4352–4356.

BM900208U

Blood Compatible Zwitterionic Copolymers Biomacromolecules, Vol. xxx, No. xx, XXXX I

Dow

nloa

ded

by T

AIW

AN

CO

NSO

RT

IA o

n Ju

ly 2

6, 2

009

Publ

ishe

d on

Jul

y 2,

200

9 on

http

://pu

bs.a

cs.o

rg |

doi:

10.1

021/

bm90

0208

u