Cell mimetic lateral stabilization of outer cell ... - Wiley Online Library

Cell mimetic lateral stabilization of outer cell mimetic bilayer on polymer surfaces by peptide bonding and their blood compatibility K. Kaladhar, Chandra P. Sharma Biosurface Technology Division, BMT Wing, Sree Chithira Tirunal Institute for Medical Science and Technology, Thiruvananthapuram, Kerala, India Received 6 August 2005; revised 23 September 2005; accepted 21 October 2005 Published online 6 June 2006 in Wiley InterScience (www.interscience.wiley.com). DOI: 10.1002/jbm.a.30681 Abstract: The biological lipid bilayer membranes are stabilized laterally with the help of integral proteins. We have simulated this with an optimized ternary phospholipid/ glycolipid/cholesterol system, and stabilized laterally on functionalized poly methyl methacrylate (PMMA) surfaces, using albumin, heparin, and polyethylene glycol as anchors. We have earlier demonstrated the differences due to orientation and packing of the ternary phospholipid monolayers in relation to blood compatibility (Kaladhar and Sharma, Langmuir 2004;20:11115–11122). The structure of albumin is changed here to expose its interior hydrophobic core by treating with organic solvent. The interaction between the hydrophobic core of the albumin molecule and the hydrophobic core of the lipid molecules is confirmed by incorporating the molecule into bilayer membranes. The secondary structure of the membrane incorporated albumin is studied by CD spectral analysis. The structure of the altered albumin molecule contains more ␤-sheet as compared to the native albumin. This conformation is also retained in membranes. The partitioning of the different anchors based on its polarity and ionic interactions in the monolayer is studied from the pressure-area (␲-A) isotherm of the lipid monolayers at the air/water interface using Langmuir-Blodgett (LB)

INTRODUCTION The coagulation event, which decides the fate of the blood contacting materials, is initiated by the adsorption and related conformational change of the adsorbed proteins on the blood contacting materials.1,2 Surfaces modified with phosphorylcholine (head group of the phosphatidylcholine) molecules have been explored for blood contacting applications because of its antifouling properties.3,4 These approaches are based on simulating the luminal surface of the Correspondence to: C. P. Sharma; e-mail: sharmacp@sctimst. ker.nic.in Contract grant sponsor: DST. © 2006 Wiley Periodicals, Inc.

trough facility. Such two monolayers are deposited onto the functionalized PMMA surface using LB trough and crosslinked by carbodiimide chemistry. The structure of the deposited bilayer is studied by depth analysis using contact mode AFM in dry conditions. The stabilized bilayer shows stability up to 1 month by contact angle studies. Preliminary blood compatibility studies reveal that the calcification, protein adsorption, as well as blood-cell adhesion is significantly reduced after the surface modification. The reduced adsorption of ions, proteins, and cells to the modified surfaces may be due to the fluidity of the microenvironment along with the contribution of the mobile PEG groups at the surface and the phosphorylcholine groups of the phospholipids. The stability of the anchored bilayer under low shear stress conditions promises that the laterally stabilized supported bilayer system can be used for low shear applications like small diameter vascular graft and modification of biosensors, and so forth. © 2006 Wiley Periodicals, Inc. J Biomed Mater Res 79A: 23–35, 2006 Key words: phospholipids; glycolipid; cholesterol; protein; calcification; blood cells; heparin; polyethylene glycol; supported; poly methyl methacrylate; biomimetic

endothelial bilayer membrane.5,6 The luminal surface of the natural bilayer membrane is enriched with phosphatidylcholines. The cholesterol plays an important role in maintaining the natural membrane integrity.7 The role of other lipids like sphingiomyelin and galactocerebroside, on maintaining the membrane integrity, has also been studied.8,9 The packing of the membrane components is playing an important role in regulating the membrane fluidity, which intern influences the adsorption of ions, proteins, cells, and so forth to these surfaces.10 The integral membrane proteins rich in ␤-sheet exist in the calveolae of the cholesterol rich domains, which anchors the membranes to the cytoskeleton.11 The hydrophobic interaction between the integral proteins and the core of the bilayer membrane holds these integral proteins in the mem-

24

brane bilayer.12,13 These factors enable these membrane surfaces to retain the fluidity, and exchangeability and thus the renewability of its components.14 Such a renewable system may interact more actively with the biological environment and gets easily integrated into the system. Albumin is a flexible passive protein, abundantly present in the blood (3.5– 4 gm %).15 This protein has been found to be nonthrombogenic in its native and altered conformations.16 The membrane proteins are rich in ␤-sheet hydrophobic exterior. This limits the solubility of the integral membrane proteins. Watersoluble globular proteins have rich ␣ helical content in its native conformation changes to molten globular state with more ␤-pleated structure and random conformations in its less-soluble state, under adverse conditions of varied pH, solvents, ionic concentration, and temperature.17,18 These proteins can be incorporated into the model lipid bilayer membranes after its phase inversion to increase its hydrophobicity by treating with organic solvents.19,20 Thus, integrated membrane proteins can be anchored to linkers with functional groups. Heparin is a sulfated polysaccharide and is a natural anticoagulant. Immobilization of heparin on material surface for blood contacting applications as linkers is widely explored.21,22 Polyethylene glycol (PEG) is another antithrombogenic molecule, explored as a linker as well as a surface modifying agent due to its antifouling properties.23,24 Attempts have been made by using such lateral stabilizers could be used for improving the stability of the supported bilayer systems.25–28 We have attempted here a comprehensive cell mimetic approach using major lipid components, as well as proteins, polysaccharides, and PEG, to develop model laterally stabilized lipid bilayer anchored, over functionalized PMMA. The lipids were selected based on the head group structure. The phosphatidylcholine (PC) for phospholipid, galactocerebroside (GalC) for glycolipid and cholesterol (Chol) are the major lipid components of the endothelial luminal cell membrane. We have optimized a outer cell mimetic lipid composition (OCMC) containing PC:Chol:GalC (1:0.35:0.125), based on the air/water interfacial studies.29,30 Here, we have attempted to laterally stabilize a bilayer of this lipid composition onto the surface of functionalized PMMA by forming peptide functional groups. The supported bilayers have been stabilized with the help of a network of albumin (OCMC-A) (PC:Chol: GalC:Alb) (1:0.35:0.125:0.008), albumin and heparin (OCMC-AH) (PC:Chol:GalC:Alb:Hep) (1:0.35:0.125: 0.008:0.052), as well as albumin, heparin and polyethylene glycol (OCMC-AHP) (PC:Chol:GalC:Alb: Hep: PEG) (1:0.35:0.125:0.008:0.052:0.15). The albumin is treated with organic solvents to do phase inversion and incorporated into the lipid solution to get stabiJournal of Biomedical Materials Research Part A DOI 10.1002/jbm.a

KALADHAR AND SHARMA

lized itself. Further we have also incorporated heparin and Diamino-PEG into the lipid solution. The air/ water interfacial behavior of the pure as well as the lipid combinations was studied in detail to know the incorporation of the anchors into the monolayer and its stability during compression, with the help of LB trough. A bilayer of the lipid compositions were immobilized into functionalized PMMA films with the help of LB trough and crosslinked by peptide bonds using carbodiimide chemistry. The surface morphology of the supported lipid systems was studied by using atomic force microscope (AFM). The fundamental interactions that regulate the blood compatibility of these stabilized surfaces have been studied by in vitro calcification, protein adsorption, and blood-cell adhesion studies from washed cells.

EXPERIMENTAL SECTION

Materials Poly methyl methacrylate (PMMA) ((MW: 1,00,000), Polyscience, Warrington) was used as the polymer substrate in these studies. 1-Ethyl-3-(3-dimethylaminopropyl)-carbodiimide (EDC), albumin (Alb) (Human fraction V, 96 –99%), ␥ globulin (Cohn fraction II) and fibrinogen (Human fraction I, over 95% protein clottable), l-␣-Phosphatidylcholine (egg yolk) (PC), and Galactocerebrosides (from bovine brain) (GalC) were from Sigma Chemicals (St Louis, MO). Cholesterol (Chol) procured from Himedia, Bombay, Heparin sodium (Hep) (5000 IU/mL) from Biological E Limited India, and O,O-bis-(2-aminopropyl)-polyethylene glycol (PEG), (MW: 1900) Fluka, was used in this study.

EXPERIMENTAL PROCEDURES Preparation of the polymer substrates The PMMA films were prepared by dry solvent-casting method. The polymer solutions were prepared in chloroform (10 gm % w/v). Films of 0.1 mm thickness were prepared by casting the solution on clean glass plates. An area of 5 ⫻ 5 cm2 were cut and cleaned with 0.1% soap solution (Teepol), rinsed with distilled water, and kept in distilled water for 12 h under stirring. The clean films were then dried in a vacuum oven at 60°C for 4 h. The partial hydrolysis of PMMA films was done by treating in 4M sodium hydroxide for 2 h at 85°C and is then immersed in 10% (w/v) citric acid solution overnight at ambient temperature. The surface-hydrolyzed PMMA was washed with doubledistilled water until the pH of the washing reaches 5.6. Further, the polymer substrate was placed in an aqueous solution containing EDC (0.5M) at ambient temperature for

BLOOD COMPATIBILITY OF LATERALLY STABILIZED CELL MIMETIC BILAYER

48 h to activate the carboxyl groups and incubated in an aqueous solution until deposition of the lipid monolayer.

Pressure-area isotherm studies and the deposition of the monolayer over polymer substrates The optimized ternary lipid combination with and without anchors was studied at the air/water interface at ambient temperature. Lipid stock solutions were prepared by dissolving the preweighed lipids in a known volume of spreading solvent (Water:Acetone:Cyclohexane:n-Butanol) (0.05:0.45:0.357:0.1425 v/v). The albumin, heparin, and PEG were dissolved in acetone/water system and mixed with the cyclohexane/n-butanol system to come to the definite proportions (PC:Chol:GalC:Alb:Hep: PEG) (1:0.35:0.125:0.008:0.052:0.15). The concentrations of the albumin, heparin, and the PEG were selected based on their maximum solubility in the above-mentioned solvent system. The concentration of the components in mixed lipid systems have been expressed in mg wt ratios (w/w), with respect to the concentration of PC. Monolayer experiments were performed on a subphase of pure water (double distilled and passed through a Milli-Q (Millipore, Bedford, MA) purification system to a resistivity of 18 M⍀/cm, at ambient temperature. Plots of surface pressure versus molecular area of monolayers were constructed using a computer-controlled Wilhelmy balance in LB trough (KSV 5000). The working solutions were prepared by mixing appropriate volumes from stock solutions. All mixtures were prepared immediately before use. Lipids were spread in small aliquots of 10 ␮L with a Hamilton syringe, which was rinsed with spreading solvent repeatedly before and after use. The compressions were initiated after a 10-min delay to allow evaporation of the spreading solvent, as well as the area versus barrier position comes to the equilibrium (i.e., after congealing of the monolayer). The compression rate did not exceed 4 Å2/molecule/min. Data were collected with KSV Inbuilt software.29,30 The compressed monolayers of different lipid compositions were deposited on PMMA substrates by the vertical dipping method, using LB double barrier Teflon trough (KSV 5000) at ambient temperature. The surface pressure maintained at 30 mN/m during deposition is higher than the critical bilayer pressure. The surface pressure was continuously monitored using a platinum Wilhelmy plate microbalance. The coated membranes were then immersed in an aqueous solution containing EDC (0.5M) at ambient temperature for 48 h to activate the carboxyl groups of the anchors and to cross-link them.31,32 They were then extensively washed using double-distilled water under static condition to remove the unreacted EDC for 4 days. The samples thus prepared were dried at room temperature and stored in sealed containers. These surfaces were used for further studies.29,30

CD spectroscopy The structure of the membrane-incorporated albumin was studied by CD spectroscopy. For that, albumin was phase

25

inverted as mentioned earlier and vacuum dried along with the respective lipid components. This mixture is reconstituted in PBS 7.4 solution to form the liposomes; they were further extruded using 100 nm polycarbonate filters to form unilamellar liposomes. The albumin concentration in the final solution was 0.075 gm/L. The CD spectra of the solutions were recorded on a Jasco-810 spectropolarimeter, using a 0.1 cm path length quartz cuvette. Acquisition was performed at a 100 nm/min scan rate, 0.5 nm band width, 2 s response, sensitivity 1000 mdeg, resolution 0.2 nm, measuring range 200 –350 nm. Each spectrum was averaged for three runs. The corresponding baseline (buffer, buffer/acetone or lipid buffer solution) was subtracted from each spectrum. The final spectrum is presented in mean ␪ residue ellipticity. ␪ ⫽ ␪/100C rl where ␪ is the measured ellipticity (mdeg), l is the path length of the quartz cell (cm), and Cr is the mean residue molar concentration obtained by the following formulae C r⫽1000nC⬘/Mp where n is the number of amino acids present in the protein molecule (585 for albumin), C⬘ is the protein concentration in the solution (gm/L) and Mp is the molecular weight of the protein (⫽ 67,000 g/mol for albumin). The spectral analysis was done using the algorithm K2d, which allows the estimation of the percentages of the protein secondary structure using a Kohonen neural network with a two-dimensional output layer.33,34 This algorithm allows calculations in the 200 –240 nm wavelength range.35

Atomic force microscope The polymer-supported monolayers of ternary lipid compositions with anchors, OCMC-AH and OCMC-AHP, were studied using AFM (NT-MDT model –SPM solver P47) in contact mode in air. Scanner with maximum scan area of 40 ⫻ 40 ␮m2 and rectangular cantilever of silicon nitride (length 200 ␮m, width 40 ␮m) having force constant of 3 N were employed for the measurement. Several areas were scanned for each sample.29,30

In vitro blood compatibility studies Calcification studies The calcification studies were done using metastable salt solutions.36 In this system, the calcium to phosphate molar ratio was maintained at 1.67, as in hydroxyapatite. The final concentration of calcium (0.20M) (as calcium chloride dihydrate) and phosphate (0.12M) (as potassium dihydrogen orthophosphate) was used. During the studies, the pH was maintained at 5.5 using (0.2M) neutralized phthalate buffer

Journal of Biomedical Materials Research Part A DOI 10.1002/jbm.a

26

(NPB), to reduce the induction of nucleation.37,38 The studies were conducted under static conditions at constant volume and ambient temperature. The samples of 2 ⫻ 2 cm2 were taken and vertically placed in the incubation medium. This is to avoid the error due to the deposition of nucleates on the surface and related crystal growth. After 48 h of incubation period, the substrates were removed and were gently washed three times with the distilled water to remove the excess solution. The films were then oven dried at 60°C for 4 h, and the surface morphology was studied using an optical microscope.29,30 The data are expressed as optical micrograph of the surface of the samples (at 10⫻ magnification, using phase contrast microscope, Labophot-2, Nikon, Japan). The data are the best representative of three consecutive studies.

Protein adsorption-desorption studies The SDS-PAGE was used to separate and evaluate the adsorbed proteins.29,30 The studies were done from a mixture of proteins after desorbing them using Triton X-100. The PMMA and modified PMMA substrates (5 ⫻ 5 cm2) were exposed for 3 h to a protein mixture containing 25 mg % albumin, 15 mg % ␥ globulin, and 7.5 mg % fibrinogen. The films were dipped in buffer and shaken to reduce the air/ water interface, and then the protein mixture was added to make the respective concentrations. After 3 h, each film was taken out and gently rinsed with buffer (pH 7.4) to remove all the loosely adsorbed proteins. The surface adsorbed proteins were desorbed using 0.1M Tris-HCl buffer (pH 8.5) containing 1% Triton X-100. The films were incubated at ambient temperature for 16 h with intermittent shaking during desorption. The proteins were then concentrated using immersible CX-100 ultrafiltration units to 500 ␮L. The samples prepared for SDS-PAGE were similar to that prepared by Chiu et al. He standardized the procedure using radiolabeled proteins.39 Chiu et al. pointed out that it is better to use the SDS-PAGE to separate the fibrinogen and ␥ globulin bands. However, in this procedure, the peptide chains of the protein molecules were not cleaved. Therefore, the individual proteins were separated based on their molecular mass. They have also evaluated the reliability of the data in comparison with the radiolabeled studies. The electrophoresis was carried out with 150 ␮L of the sample. The SDS-PAGE was performed on a 5.6% cylindrical gels containing 1% SDS. Protein samples were prepared by incubating them for 16 h at 37°C in a sample buffer containing 1% SDS, 10% glycerol, 10 mM Tris-HCl, and 1 mM EDTA (pH 8.0). Bromophenol blue was added to the samples to a final concentration of 10 mg/mL as a tracking dye; the electrophoretic run was done at 4°C and at a voltage of 20 V/gel. When the dye front was moved about 0.5 cm from the bottom of the gel, the gels were removed, stained with 0.05% Coomassie blue, destained in 5% acetic acid/10% methanol, and were scanned using a UV–vis spectrophotometer (UV160A Shimadzu) at ␭ max 620 nm. All the experiments were done together at same conditions. The data provided are best representative of three consecutive studies.


KALADHAR AND SHARMA

Blood-cell adhesion studies Platelet adhesion studies. The platelets were separated from a freshly drawn citrated (tri-sodium citrate 3.8% w/v solution) calf blood (at the ratio of (1:9)) by the procedure described elsewhere.29,30,35 In short, the blood was centrifuged at 70g for 10 min; the supernatant was taken and centrifuged at 140g for 10 min to remove the leukocyte button. Then, the supernatant was again centrifuged at 560g for 10 min. The supernatant was discarded, and the platelet button was washed further three times with the tyrode solution. The washed platelet suspension was spread over the polymer films and incubated for 15 min. Then, the films were gently and uniformly washed with 0.1M phosphate-buffered saline (PBS) (pH 7.4). The washed films were then fixed with 2.5% glutaraldehyde solution and stained with Coomassie blue staining solution. The fixed films were counted by light microscopy.

Leukocyte adhesion studies. The leukocytes from the citrated calf blood was separated according to the already reported procedure.29,30,40 The erythrocyte-removed (as above mentioned) citrated calf blood was centrifuged at 140g for 10 min, to remove the leukocyte button, and the cells were washed and suspended in tyrode solution and studied similarly as mentioned earlier.

Erythrocyte adhesion studies. The erythrocyte adhesion studies were done by using the procedure described elsewhere.29,30,40 In short, the erythrocyte was separated from the citrated calf blood by centrifuging at 70g. The supernatant was discarded. The sedimented erythrocyte was washed three times with the 0.1M PBS (pH 7.4), and suspended in PBS and studied similarly as mentioned earlier. All the studies were done separately and within 2 h of collection of the blood. The data of blood-cell adhesion studies are provided as the number of cells adhered/mm2. The data provided are the averages of 10 different regions on the substrate.

Stability studies Stability studies were done under shear conditions using a rotary shaker. The modified surfaces were immersed in 25 mL of PBS (7.4) and shaken at 150 rpm at ambient temperature. Samples were collected at specified intervals, washed with distilled water to remove buffer salts, and dried at room temperature. The change in contact angle was studied by using a goniometer.41

RESULTS The behavior of the monolayer at the air/water interface The surface pressure area (␲-A) isotherm of the lipid monolayers is shown in Figure 1. The optimized


27

Figure 1. ␲-A isotherm of the different lipid compositions. PC (A), OCMC (B), OCMC-A (C), OCMC-AH (D), OCMC-AHP (E).

OCMC lipid mixture29 and with the incorporation of anchors are studied and compared with a PC monolayer at the air/water interface at ambient temperature. The PC [Fig. 1(A)] monolayer shows coexistence

of the liquid-expanded (LE) and liquid-condensed (LC) phases as observed earlier.42 The collapse pressure of the isotherm is at 42 mN/m, and the limiting molecular area of the PC monolayer is at 70 Å2/ Journal of Biomedical Materials Research Part A DOI 10.1002/jbm.a

28

molecule. The ␲-A isotherm of the OCMC is shown in Figure 1(B). With the inclusion of the Chol and GalC, the LE phase of the PC monolayer is reduced, while improving the slope of the ␲-A isotherm. It improves the stability of the monolayer at the air/water interface.29 For laterally immobilizing these lipid layers on polymer substrates, we have selected albumin, heparin, and PEG molecules of proven antithrombogenecity.43 The collapse pressure of the monolayer is reduced after the incorporation of albumin into the monolayer, OCMC-A [Fig. 1(C)], which indicates the interaction at the hydrophobic core of the monolayer. However, the mean molecular area of the monolayer has not reduced here. The pressure area isotherm also has not shown any sharp discontinuity of phase transfer from LE to LC phase during compression. Perhaps, the LE and LC phases are coexisting together. Further incorporation of heparin along with albumin into the monolayer improved the collapse pressure from 38 to 42 mN/m (normal collapse pressure of PC monolayer). This monolayer showed a right shift in the mean molecular area [Fig. 1(D)]. The Figure 1(E) shows the ␲-A isotherm of the monolayer containing albumin, heparin, and diamino-PEG. The LE phase of the monolayer is increased with the incorporation of PEG into the monolayer. The isotherm shows a sharp discontinuity at 12 mN/m surface pressure, after that the monolayer behaved similar to the other monolayers until the collapse.

The structure of albumin in the bilayer by CD spectroscopy The CD spectra of the native as well as the altered conformations of albumin after incorporating into the membranes are shown in Figure 2. The calculated percentages of the secondary structure are given in Table I. The CD spectra demonstrate that the secondary structure of the native albumin is altered during phase inversion. This altered structure is retained irrespective of the lipid composition in all the protein lipid mixtures. The spectral analysis data (Table I) indicates that the secondary structure of the phase-inverted albumin layer was different from the native albumin. The helical content of the phase-inverted albumin is decreased and mainly converted to ␤-sheet during the phase inversion. The denaturation of the native protein structure is reduced in the presence of lipids. Especially, there was little difference in the random coil structure. However, the total dimension of phaseinverted albumin is increased. This indicates that the albumin exists in molten globular state without entirely loosing its entire secondary structure in the lipid membranes. Journal of Biomedical Materials Research Part A DOI 10.1002/jbm.a

KALADHAR AND SHARMA

Figure 2. Structure of the albumin molecule in lipid bilayer. CD spectra of albumin in its native state (A) and altered (B), incorporated into PC (C), OCMC-A (D), and OCMC-AHP (E). [Color figure can be viewed in the online issue, which is available at www.interscience.wiley.com.]

Stability of the laterally stabilized bilayer in phosphate-buffered saline (7.4) Two such monolayers of respective compositions were transferred to the functionalized PMMA surface in the upstroke mode with the help of LB trough. The anchor network formed on this supported bilayer is stabilized by carbodiimide chemistry, where the albumin molecule is perfectly incorporated into the lipid layer. These three anchor molecules cross-link to each other to form a three-dimensional network and stabilize the bilayer (Scheme 1 and 2).

Stability of the laterally stabilized anchored bilayer The water contact angle studies of the modified surfaces revealed that the surface hydrophobicity of the PMMA membranes have decreased after the surface modifications. The contact angle ␪ is reduced from 85° to 20° after the surface modification. The stability studies have been done under low shear stress conditions. Figure 3 shows the stability studies of the modified surfaces represented in terms of the change in contact angle.44 The PMMA surface modified with the ternary lipid layer shows less stability, as compared to the laterally stabilized surfaces. In this surface, the contact angle have been stabilized after a certain period, may be due to the improved stability contributed by the Chol and GalC (condensation effect). However, in the case of PMMA modified with


29

TABLE I Secondary Structure of the Albumin Molecule in Lipid Bilayer (%)

␣ ␤ Random Square distance (Å)

Native

Altered

Incorporated into PC

OCMC-A

OCMC-AHP

31.5 11 57.5 309.54

5.67 46.33 38 734.26

24.73 35 40.27 618.56

18.26 39.74 42 664.25

20.79 38 41.21 667.7

ternary lipids containing anchors [Fig. 3(B,C)], the contact angle has not varied significantly with time. This shows that the anchored lipid layers are much more stable on the polymer surface than the unanchored lipid layers. Therefore, further characterizations of the supported lipid layers are done only for those with improved stability by the lateral stabilization using anchors of albumin, heparin, and PEG.

Structure of the supported bilayer The surface morphology of the modified PMMA surfaces that are stabilized by anchors has been studied using AFM in contact mode in air. Figure 4 shows the contact mode AFM image of the modified PMMA surfaces in air. The surface of the monolayer containing albumin and heparin has [Fig. 4(A)] shown less number of domains and holes. The depth of the holes is found to be 36 – 82 Å. The size of the smaller holes is found to be nearly half to that of the larger holes. This shows that some of the holes have continuity through the bilayer to the polymer surface, while the other holes have not. The smaller holes with depth of 36 Å show the thickness of the upper monolayer, and the depth of the bigger holes reveal the total thickness of the bilayer (the fully extended state of the PTC has been nearly 34 Å). The deviation from the standard bilayer thickness may be due to the inclusion of the polymeric chain of the anchors. Standard deduction of

Scheme 1. Schematic representation of stabilized supported bilayer. [Color figure can be viewed in the online issue, which is available at www.interscience.wiley.com.]

the size of the domains contributed by the polymer chains from the depth of the smaller holes suggests that the monolayers are in less condensed state (26 –30 Å). The surface modified with albumin, heparin, and PEG [Fig. 4(B)] has also shown least number of holes. Many domains with 5 Å have been observed on the PEG incorporated lipid surface, while not in the other surfaces. This confirmed that the smaller domains observed here are due to the polymeric chains. The reduction in holes and smaller domains in the PEGincorporated lipid layers may be due to perfect packing of the monolayer inside the anchor network. The adsorption of ions, proteins, and cells to the surface of the blood contacting materials plays a major role in initiating the thrombosis event, which finally decides the fate of the blood-contacting device in vivo.45 To have a better understanding of the initial adsorption events that regulate blood compatibility, we have studied the adhesion of ions (by calcification studies), proteins (from mixture of proteins), and cells (from washed blood cells) on to these modified surfaces.

Calcification studies to the modified surfaces Calcification data from metastable salt solutions are shown in Figure 5. The overall calcification to the modified surfaces is significantly reduced as compared to the bare PMMA surface [Fig. 5(A–C)]. The size and the density of the crystals formed on the modified surface are reduced as compared to the bare polymer. On bare polymer, flower-type crystals are formed with its nuclei originating from the material surface. These crystals are uniformly formed throughout the surface, with long petals at the cut edge of the surface. This is totally absent in the case of modified polymers.

Scheme 2. Reaction scheme of the components, which laterally stabilizes the bilayer on PMMA surface.


30

KALADHAR AND SHARMA

Protein adsorption to the lipid modified polymer surfaces The protein adsorption data, using mixture of albumin, ␥ globulin, and fibrinogen, is shown in Figure 6. It shows reduced protein adsorption to all the modified surfaces [Fig. 6(B,C)] as compared to the bare PMMA [Fig. 6(A)]. However, the ratio of albumin to fibrinogen and ␥ globulin is higher in the case of bare PMMA surface. This ratio is not maintained in the case of modified polymer surface.

Blood-cell adhesion studies to the modified surfaces

Figure 3. Stability studies under low shear stress conditions. PMMA bare (A), PMMA modified with OCMC (D), OCMC-AH (B), and OCMC-AHP (C).

Figure 7 shows the data of blood-cell adhesion studies to these laterally stabilized lipid bilayer modified surfaces using washed platelets, erythrocytes, and the

Figure 4. AFM image of supported laterally stabilized bilayers. PMMA modified with OCMC-AH (A), OCMC-AHP (B). The depth of the bilayers are calculated from the holes (indicated by arrow), which are visible from the monolayer surface. [Color figure can be viewed in the online issue, which is available at www.interscience.wiley.com.]

Figure 5. Calcification studies. PMMA bare (A), PMMA modified with OCMC-AH (B), OCMC-AHP (C).



Figure 6. Densitometric diagram of the protein adsorption– desorption studies on lipid-modified polymer surfaces, using mixture of proteins. X-axis length in cm (maximum 6 cm), and Y-axis absorbance (maximum value ⫽ 0.5). PMMA bare (A), PMMA modified with OCMC-AH (B), OCMCAHP (C). The peaks are (1) albumin, (2) ␥ globulin, and (3) fibrinogen.

31

sorption, as well as calcification on to the modified surfaces are reduced. This is similar to the results that we have obtained earlier with the OCMC supported hydrophobic polymer substrates. The main difference is that the protein adsorption here is further reduced, may be due to the effect of PEG. The air/water interfacial studies have been utilized extensively to study the individual effect of lipids in a monolayer. This also gives structural information that regulates the self-assembly of the individual molecules. The PC monolayer remained in the fluid phase throughout the compression. It is due to the unsaturation of the hydrophobic chain of the egg PC. The double bond in egg PC is found to be cis,46,47 which reduces further close packing of the monolayer. Earlier, we have observed concentration-depended condensation effect on PC monolayer with the incorporation of Chol.29. This is due to the strong interaction between the hydrophobic cyclo pentano perhydro phenanthrene rings of the Chol, when they are in close proximity. Apart from that, we have also demonstrated that the GalC is maintaining the membrane integrity in reduced cholesterol concentrations.29 The interaction as well as the incorporation of the integral proteins to the monolayer at the air/water interface has been already studied with the help of LB trough.48 They concluded that for the stabilization of the protein molecule into the monolayer strong hydrophobic interaction is needed between the hydrophobic domain of the protein molecule and the monolayer.

leukocytes to the bare PMMA and the modified surfaces. Here, we have done the blood-cell adhesion studies from washed cells, to understand about the direct interaction between the surface and the blood cells. The results show an overall decrease in cell adhesion on to modified surfaces as compared to the bare PMMA surface.

DISCUSSION Here, we have shown cell mimetic lateral stabilization of the bilayer on to the functionalized polymeric substrates using peptide bonds. This has been made possible by incorporating phase-inverted albumin molecule into the lipid monolayer. It appears that they are laterally interconnected at the air/water interface with heparin molecules to improve the stability during compression, along with diamino-PEG decorated at the side of the water subface, during compression. Such two monolayers have been transferred to the functionalized polymeric substrates with the help of LB trough. The monolayers have been subsequently crosslinked by using carbodiimide chemistry with peptide bonds. The overall cell adhesion, protein ad-

Figure 7. Blood-cell adhesion studies on lipid-modified polymer surfaces, using washed cells. PMMA bare (A), PMMA modified with OCMC-AH (C), OCMC-AHP.


32

We have utilized here the flexible nature of the albumin molecule to integrate this globular protein into the monolayer. The amino acid sequence dictates the folding of membrane proteins in above-said solvent. In the case of soluble proteins like albumin, the hydrophobic core is buried inside the protein molecules. During the conformational change for a preferred conformation in a hydrophobic environment, the hydrophobic core is exposed to the exterior.49,50 Earlier it has been observed that the hydrophobic core of the membrane integral proteins interacts firmly with the hydrophobic core of the bilayer membrane. The lipid molecules are acting as molecular chaperons and stabilize the integral membranes in the bilayer membrane,51 and control the folding or refolding of the proteins. The integration of protein molecule in the monolayer has been interpreted from the ␲-A isotherm studies.52 The proper integration of the protein molecule will shift the monolayer to the right, and the mean molecular area is increased accordingly,53 while the loosely packed protein molecules (proteins with hydrophilic exterior, cannot integrate with the hydrophobic tail of the lipids) are found to escape from the monolayer during compression.54 This will lead to a left shift in the mean molecular area. In our studies, the incorporation of the phase-inverted albumin into the lipid layer shows a reduction in the collapse pressure [Fig. 1(C)]. This indicates the interaction at the hydrophobic region of the lipid monolayer. The reduction in collapse pressure may be due to the induction of strong discontinuity in the lipid monolayer under compression by the larger albumin molecule at the air/water interface. This indicates that the albumin molecule is well-incorporated into the monolayer. However, the monolayer stability is decreased here, as evident from the reduction in the collapse pressure [Fig. 1(C)]. Therefore, to further stabilize the monolayer by linking the neighboring albumin molecules, we have added heparin also along with albumin into the lipid solutions. This forms a network of albumin and heparin, at the air/water interface and stabilizes the monolayer during compression. Heparin has been found to be interacting ionically with the albumin molecule,55 which is evident from the increase in collapse pressure of the ␲-A isotherm [Fig. 1(D)]. It also leads to a right shift in the mean molecular area. This is a better indication of incorporation of albumin and heparin into the monolayer and partial interconnection of the albumin molecule with heparin, by ionic interactions. While when the hydrophilic PEG molecules are introduced into the monolayer along with albumin and heparin, there has been a sharp phase change at low surface pressures. This indicates that the PEG is migrating out of the monolayer during compression.56 However, there was no difference in the collapse pressure, indicating the stabilization of albumin into the monolayer by the Journal of Biomedical Materials Research Part A DOI 10.1002/jbm.a

KALADHAR AND SHARMA

heparin molecules [Fig. 1(E)]. However, because of the ionic interactions with the heparin (amino group of PEG and carboxy group of heparin), the PEG molecules cannot escape into the subphase. This ensures that the bulk of the PEG molecule exit out of the monolayer at 30 mN/m, the pressure at which we have transferred the monolayer to the polymer substrates. In these studies, the albumin molecule is mixed with the lipid solution by sequentially treating with the solvents of increasing hydrophobicity. The albumin of specified concentration was dissolved in water, and then mixed with acetone, and the solution is quickly introduced into cyclohexane/n-butanol mixture containing the lipids in definitive proportions. Incorporation of the higher concentration of albumin molecule, into the lipid solution, leads to the precipitation of albumin molecule due to self-aggregation. The self-aggregation is due to hydrophobic interaction between the exposed hydrophobic groups of the albumin molecule.57 But, at this specified concentrations, we did not observed any self-aggregation of the albumin molecule. This may be due to the stabilization of the altered conformation of the albumin molecule, by the lipids. Here, the hydrophobic interaction between the albumin and the lipid tail chain reduces the selfaggregation of the albumin molecule. To understand the solvent-induced phase inversion and albumin lipid interactions, we have done the far UV circular dichroism studies. The CD spectra demonstrate that the secondary structure of the native albumin [Fig. 2(A)] is altered during the phase inversion [Fig. 2(B)]. The altered conformation is maintained as such in the case of albumin-bound liposome with different lipid compositions also [Fig. 2(C–E)]. This is due to the maintenance of the altered conformation. Spectral analysis using K2d algorithm indicates that secondary structure of the phase-inverted albumin layer was different from the native albumin (Table I). The treatment of albumin with organic solvents induces irreversible changes in the albumin molecule. The helical content of the phase-inverted albumin is decreased and mainly converted to ␤-sheet during the phase inversion. Especially, there is little difference in the random coil structure. However, the total dimension of phaseinverted albumin is increased. This indicates that the albumin exists in a molten globular state without entirely loosing its secondary structure in the lipid membranes. This may be due to stabilization of the altered confirmation by the neighboring lipid molecules. Further, we have covalently crosslinked the ligands by peptide bonds based on the carbodiimide chemistry (Scheme 2). Scheme 1 shows the three-dimensional distribution of the three different anchor molecules in and around the bilayer based on the hydrophobic as well as ionic interactions between the anchors. The stability of the modified surfaces is done under low


shear conditions from PBS 7.4 solution. The improved stability of the laterally stabilized bilayer is evident from the contact angle studies [Fig. 3(B,C)]. In natural membranes also, the bilayer membrane is laterally stabilized through transmembrane proteins.58 The lipid–lipid immiscibility between the cholesterol and the phospholipid forms cholesterol rich domains and they form calveolae for the transmembrane proteins in the plasma membrane. Earlier, such domains of larger dimension have been observed in binary phospholipids/cholesterol systems at the air/ water interface, and in unilamellar vesicles, which is correlated with that of the domains or rafts present in biological membranes.59 We have also observed similar effects earlier in simple model lipid mixtures.29 This phenomenon is related with the condensing effect of cholesterol.60 Earlier, we have observed larger domains in the case of inverted29 and direct supported monolayers.30 Here, from the AFM studies (Fig. 4), with the incorporation of anchors the number and size of the cholesterol rich domains formed is reduced. This may be due to reduced lateral movement of the lipids due to the incorporation of albumin into the monolayer. The depth analysis using AFM in the limited holes available in the surface revealed the bilayer structure of the deposited lipid layer. They showed slightly higher thickness than the standard bilayer thickness, might be due to the entangled hydrophilic chains between the substrate and the bilayer, as well as on the lipid surface. Such deviations also have observed earlier. Further, we have attempted to study the adsorption of ions proteins and cells onto these lipid-supported surfaces, as these initial events strongly regulate the blood compatibility of the lipid-modified surfaces. The calcification to the organized organic–inorganic interface in biological systems has been reported.61 The surface-induced nucleation is proposed to be the reason for the calcification, on the surfaces. Therefore, here, we have studied the propensity of the surface toward calcification from metastable salt solutions. The pH of the solution is reduced to pH 5.5 to reduce the induction of nucleation.37,38 The calcification data shows (Fig. 5) that the overall calcification to the modified surfaces is significantly reduced as compared to the bare PMMA surface. This is due to the fluidity of the microenvironment. The loose packing of this stabilized supported bilayer increases the osmotic pressure of the microenvironment and this intern improve the fluid flow in the microenvironment, which leads to the osmotic exclusion of the ions. This has been earlier observed in the case of exclusion of proteins by PEG immobilized surface.62 There also the packing of the PEG influences the effective osmotic exclusion of the proteins from the surface. On the contrary, in natural environment, we can find that the phosphate moieties of the cell membrane are found to be acting as the

33

Scheme 3. Osmotic exclusion of ions, proteins, and blood cells from the modified surface. [Color figure can be viewed in the online issue, which is available at www.interscience.wiley. com.]

nucleation site for calcification.61 There are evidences of calcification on phosphorylcholine-immobilized surfaces also.63 However, here, we find that the calcification is significantly reduced in these laterally stabilized surfaces. This clearly indicates that the fluidity of the microenvironment plays an important role in the adsorption-related phenomena (Scheme 3). Earlier, we have prepared phospholipid-modified surfaces of varying composition, which simulate the natural environment, to see the effect of composition on packing and orientation of this supported monolayers and its influence on adsorption of ions, proteins, and blood cells.29,30 Those results also suggest the same. Protein adsorption is an important phenomenon, which influences the biological events. It has been reported that the protein adsorption to any substrates from a mixture of proteins present in the blood could be directly correlated to that from plasma.64 The protein adsorption to all the modified surfaces [Fig. 6(B,C)] was lower than the bare PMMA surface [Fig. 6(A)]. However, the ratio of albumin to fibrinogen and ␥ globulin was higher in the case of bare PMMA surface. This ratio is not maintained in the case of modified polymer surface. However, different from our earlier studies,29 the protein adsorption to the anchor-incorporated lipid surface was further reduced, may be due to the surface decorated PEG molecules. The adsorbed protein molecules on bare PMMA change their conformation, since it is highly hydrophobic. The increased albumin to fibrinogen ratio of the adsorbed proteins enhances the blood compatibility, and the surfaces preferentially adsorb fibrinogen are found to be thrombogenic. The adsorbed fibrinogen changes its conformation to expose the RGD sequences, and interact with the receptors in the platelet surface,65 which leads to platelet adhesion, activation, and thrombosis.66,67 The phosphorylcholine (the head group of the PC) immobilized surfaces has been found to be less susceptible for protein adsorption.68 This is due to the inertness of the phosphoJournal of Biomedical Materials Research Part A DOI 10.1002/jbm.a

34

KALADHAR AND SHARMA

rylcholine groups in the biological environment. Our earlier results also suggest the same.29,30 Here, we can find that the overall protein adhesion has significantly reduced in the case of modified surfaces as compared to the bare polymer. In addition to the effect of the phosphorylcholine moieties of the bilayer, the flexibility of the brushes of PEG and heparin also helps to reduce the adsorption phenomena. The resistance to protein adsorption of PEG has been attributed to the steric stabilization effect, solution properties, and its molecular conformation in aqueous solution. It has been proposed that competition between steric expulsion forces and van der Waals attraction between the protein and PEG control the protein adsorption process.69 Here, in addition to the effect of PEG, the fluidity of the microenvironment due to the lipid bilayer also might be enhancing the antifouling property of the modified surface. Further, the blood-cell adhesion studies (Fig. 7) were done from washed blood cells to check the specific interaction between the blood cells and the surfaces. The results show an overall decrease in cell adhesion on to modified surfaces as compared to the bare PMMA surface. The reduction in protein adsorption and the blood-cell adhesion reveal that these modified surfaces could considerably reduce the activation of the coagulation pathway. However, further blood compatibility studies on this surface are required for complete understanding of this surface. Such modified blood contacting materials could be used for low shear stress applications like modification of surface properties of biosensors or small diameter vascular grafts. The deposition of the monolayers to the luminal surface of small diameter vascular grafts can be done using LB trough by the controlled rotation at the interface.29 We express our sincere thanks to Dr. K. Mohandas, The Director, SCTIMST and Dr. G. S. Bhuwaneshwar, The Head, BMT Wing for providing the facilities, Dr. Yakhmi JV and Dr. Sipra Choudhury for providing the LB trough facility, Dr. Ajay ghosh. A and Vijayakumar, RRL, Trivandrum for providing the Circular Dichroism facility. Willi Paul, member of the BST division, for discussions and the comments on the manuscript.

4.

5.

6.

7.

8.

9.

10.

11. 12. 13. 14. 15.

16.

17.

18.

19.

20.

21.

22.

References 1.

2.

3.

Sharma CP. Biomaterials and artificial organs: Perspectives towards the 21st century. Transactions of the Xth National Conference of Society for Biomaterials and Artificial Organs (Biomedical Materials and Devices). Trends Biomater Artif Organs 1998;12:39 – 41. Brash JL. Exploiting the current paradigm of blood-material interactions for the rational design of blood-compatible materials. J Biomater Sci Polym Ed 2000;11:1135–1146. Ishihara K, Fujita H, Yoneyama T, Iwasaki Y. Antithrombogenic polymer alloy composed of 2-methacryloyloxyethyl


23. 24.

25. 26.

phosphorylcholine polymer and segmented polyurethane. J Biomater Sci Polym Ed 2000;11:1183–1195. Iwasaki Y, Yamasaki A, Ishihara K. Platelet compatible blood filtration fabrics using a phosphorylcholine polymer having high surface mobility. Biomaterials 2003;24:3599 –3604. Hayward JA, Chapman D. Biomembrane surfaces as models for polymer design: The potential for haemocompatibility. Biomaterials 1984;5:135–142. Durrani AA, Hayward JA, Chapman D. Biomembranes as models for polymer surfaces. II. The syntheses of reactive species for covalent coupling of phosphorylcholine to polymer surfaces. Biomaterials 1986;7:121–125. Anderson TG, McConnell HM. Condensed complexes and the calorimetry of cholesterol-phospholipid bilayers. Biophys J 2003;84:3079 –3086. Ruocco MJ, Shipley GG. Interaction of cholesterol with galactocerebroside and galactocerebroside-phosphatidylcholine bilayer membranes. Biophys J 1984;46:695–707. Samsonov AV, Mihalyov I, Cohen FS. Characterization of cholesterol-sphingomyelin domains and their dynamics in bilayer membranes. Biophys J 2001;81:1486 –1500. Kaladhar K, Sharma CP. Blood compatibility of cell mimetic monolayers. 7th World Biomaterial Conference; 2004, Sydney, Australia. Sheetz MP. Microtubule motor complexes moving membranous organelles. Cell Struct Funct 1996;21:369 –373. Tanford C. How protein chemists learned about the hydrophobic factor. Protein Sci 1997;6:1358 –1366. Tanford C. The hydrophobic effect and the organization of living matter. Science 1978;200:1012–1018. Lenaz G. Lipid fluidity and membrane protein dynamics Biosci Rep 1987;7:823– 837. Curry H, Mandelkow P, Franks BN. Crystal structure of human serum albumin complexed with fatty acid reveals an asymmetric distribution of binding sites. Nat Struct Biol 1998; 5:827– 835. Sharma CP, Sunny MC. Albumin adsorption on to aluminium oxide and polyurethane surfaces. Biomaterials 1990;11:255– 257. Nozaki Y, Reynolds JA, Tanford C. The interaction of a cationic detergent with bovine serum albumin and other proteins. J Biol Chem 1974;249:4452– 4459. Shosheva A, Miteva M, Christova P, Atanasov B. pH-dependent stability of sperm whale myoglobin in water-guanidine hydrochloride solutions. Eur Biophys J 2003;31:617– 625. Cooper A. Heat capacity of hydrogen-bonded networks: An alternative view of protein folding thermodynamics. Biophys Chem 2000;85:25–39. Fernandez A, Kardos J, Goto Y. Protein folding: Could hydrophobic collapse be coupled with hydrogen-bond formation? FEBS Lett 2003;536:187–192. McColl MD, Greer IA. Low-molecular-weight heparin for the prevention and treatment of venous thromboembolism in pregnancy. Curr Opin Pulm Med 2004;10:371–375. Siguret V, Pautas E, Gouin I, Safety profile of different lowmolecular weight heparins used at therapeutic dose. Curr Opin Pulm Med 2004;10:366 –370. Needham D, Kim DH. PEG-covered lipid surfaces: Bilayers and monolayers. Colloids Surf B Biointerfaces 2001;18:183–195. Kim K, Kim C, Byun Y. Preparation of a PEG-grafted phospholipid Langmuir-Blodgett monolayer for blood-compatible material. J Biomed Mater Res 2000;52:836 – 840. Sackmann E. Supported membranes: Scientific and practical applications. Science 1996;271:43– 48. Goennenwein S, Tanaka M, Hu B, Moroder L, Sackmann E. Functional incorporation of integrins into solid supported membranes on ultrathin films of cellulose: Impact on adhesion. Biophys J 2003;85:646 – 655.


27.

28.

29. 30.

31.

32.

33.

34. 35.

36.

37. 38.

39.

40.

41.

42.

43.

44.

45.

46.

47.

Brian AA, McConnell HM. Allogeneic stimulation of cytotoxic T cells by supported planar membranes. Proc Natl Acad Sci USA 1984;81:6159 – 6163. Wong JY, Park CK, Seitz M, Israelachvili J. Polymer-cushioned bilayers. II. An investigation of interaction forces and fusion using the surface forces apparatus. Biophys J 1999;77:1458 –1468. Kaladhar K, Sharma CP. Supported cell mimetic monolayers and their interaction with blood. Langmuir 2004;20:11115–11122. Kaladhar K, Sharma CP. Supported outer cell mimetic model lipid composition: Effect of phosphatidyethanolamine on blood compatibility. J Biomed Mater Res A 2005. Submitted for publication. Nakajima N, Ikada Y. Mechanism of amide formation by carbodiimide for bioconjugation in aqueous media. Bioconjug Chem 1995;6:123–130. Young JJ, Cheng KM, Tsou TL, Liu HW, Wang HJ. Preparation of cross-linked hyaluronic acid film using 2-chloro-1-methylpyridinium iodide or water-soluble 1-ethyl-(3,3-dimethylaminopropyl)carbodiimide. J Biomater Sci Polym Ed 2004;15:767–780. Andrade MA, Chacon P, Merelo JJ, Moran F. Evaluation of secondary structure of proteins from UV circular dichroism using an unsupervised learning neural network. Protein Eng 1993;6:383–390. Merelo JJ, Andrade MA, Prieto A, Moran F. Proteoinitropic future maps. Neurocomputing 1994;6:443– 454. Serro AP, Bastos M, Costa Pessoa J, Saramago B. Bovine serum albumin conformational changes upon adsorption on titania and on hydroxyapatite and their relation with biomineralisation. J Biomed Mater Res 2004;70A:420 – 427. Golomb G, Wagner D. Development of a new in vitro model for studying implantable polyurethane calcification. Biomaterials 1991;12:397– 405. Brown WE, Lehr JR, Smith JP, Frazier AW. Crystallography of octacalcium phosphate. J Am Chem Soc 1957;79:5318 –5325. Heughebaert JC, Zawacki SJ, Nancollas GH. The growth of octacalcium phosphate on beta tricalcium phosphate. J Cryst Growth 1983;63:83–90. Chiu TC, Metcalf LC, Lyman DJ. Electrophoretic analysis of protein adsorbed on polymer surfaces. J Biomed Mater Res 1981;15:781–784. Chandy T, Sharma CP. Changes in albumin/platelet interaction with an artificial surface– due to a antibiotics, pyridoxal phosphate, and lymphocytes. Artif Org 1988;12:143–151. Kim K, Kim C, Byun Y. Preparation of a PEG-grafted phospholipid Langmuir-Blodgett monolayer for blood-compatible material. J Biomed Mater Res 2000;52:836 – 840. Larvor MP, Cerdan R, Gumila C, Maurin L, Seta P, Roustan C, Vial H. Characterization of the lipid-binding domain of the Plasmodium falciparum CTP:Phosphocholine cytidylyltransferase through synthetic-peptide studies. Biochem J 2003;375: 653– 661. Moon HT, Lee YK, Han JK, Byun Y. Improved blood compatibility by sustained release of heparin-deoxycholic acid conjugates in a PCL-PEG multiblock copolymer matrix. J Biomater Sci Polym Ed 2002;13:817– 828. Kim K, Kim C, Byun Y. Preparation of a stable phospholipid monolayer grafted onto a methacryloyl-terminated substrate as blood compatible materialsm. J Biomater Sci Polym Ed 2003;14:887–902. Chandy T, Sharma CP. Protein/platelet interaction with an artificial surface: Effect of vitamins and platelet inhibitors. Thromb Res 1986;41:9 –22. Mattjus P, Bittman R, Slotte JP. Molecular interaction and lateral domain formation in monolayers containing cholesterol and phosphatidylcholines with acyl- or alkyl-linked C16 chains. Langmuir 1996;12:1284. Shah DO, Schulman JH. The ionic structure of lecithin monolayers. J Lipid Res 1967;8:227–233.

48.

49.

50.

51.

52.

53.

54.

55.

56. 57.

58.

59. 60.

61.

62. 63. 64.

65.

66.

67.

68.

69.

35

Bayburt TH, Sligar SG. Self-assembly of single integral membrane proteins into soluble nanoscale phospholipid bilayers. Protein Sci 2003;12:2476 –2481. Dong A, Meyer JD, Kendrick BS, Manning MC, Carpenter JF. Effect of secondary structure on the activity of enzymes suspended in organic solvents. Arch Biochem Biophys 1996;334: 406 – 414. Khmelnitsky YL, Mozhaev VV, Belova AB, Sergeeva MV, Martinek K. Denaturation capacity: A new quantitative criterion for selection of organic solvents as reaction media in biocatalysis. Eur J Biochem 1991;198:31– 41. L. Whmelnitsky Y, Belova AB, Levashov AV, Mozhaev VV. Relationship between surface hydrophilicity of a protein and its stability against denaturation by organic solvents. FEBS Lett 1991;284:267–269. Lahdo R, de La Fournie`re-Bessueille L. Insertion of the amyloid precursor protein into lipid monolayers: Effects of cholesterol and apolipoprotein E. Biochem J 2004;382:987–994. Allen SJ, Curran AR, Templer RH, Meijberg W, Booth PJ. Controlling the folding efficiency of an integral membrane protein. J Mol Biol 2004;342:1293–1304. Maskarinec SA, Hannig J, Lee RC, Lee KYC. Direct observation of poloxamer 188 insertion into lipid monolayers. Biophys J 2002;82:1453–1459. Hattori T, Kimura K, Seyrek E, Dubin PL. Binding of bovine serum albumin to heparin determined by turbidimetric titration and frontal analysis continuous capillary electrophoresis. Anal Biochem 2001;295:158 –167. Marsh D. Elastic constants of polymer-grafted lipid membranes. Biophys J 2001;81:42154 – 42162. Srisailam S, Kumar TK, Srimathi T, Yu C. Influence of backbone conformation on protein aggregation. J Am Chem Soc 2002;124:1884 –1888. Salzer U, Prohaska R. Stomatin, flotillin-1, and flotillin-2 are major integral proteins of erythrocyte lipid rafts. Blood 2001; 97:1141–1143. Mayor S, Rao M. Rafts: scale-dependent, active lipid organization at the cell surface. Traffic 2004;5:231–240. Silvius JR. Role of cholesterol in lipid raft formation: Lessons from lipid model systems. Biochim Biophys Acta 2003;1610: 174 –183. Mann S, Douglas D, Archibald JMD, Douglas T, Heywood BR, Fiona C, Nicholas MFC, Reeves J. Crystallization at inorganic– organic interfaces—-Biominerals and biomimetic syntheses. Science 1993;261:1286 –1292. Arakawa T, Timasheff SN. Mechanism of poly(ethylene glycol) interaction with proteins. Biochemistry 1985;24:6756 – 6762. Goldberg M Septier. Phospholipids in amelogenesis and dentinogenesis. Crit Rev Oral Biol Med 2002;13:276 –290. Sharma CP, Chandy T. Influence of steroid hormones on protein-platelet interaction at the blood–polymer interface. Biomaterials 1989;10:609 – 616. Baier RE, De Palma VA, Goupil DW, Cohen E. Human platelet spreading on substrata of known surface chemistry. J Biomed Mater Res 1985;19:1157–1167. Broberg M, Eriksson C, Nygren H. GpIIb/IIIa is the main receptor for initial platelet adhesion to glass and titanium surfaces in contact with whole blood. J Lab Clin Med 2002;139:163–172. Broberg M, Nygren H. Exposure of fibrinogen-adherent platelets to plasma proteins: A new method for studying protein interactions with platelets. J Biomed Mater Res 2003;66A:403– 410. Ye SH, Watanabe J, Iwasaki Y, Ishihara K. Antifouling blood purification membrane composed of cellulose acetate and phospholipid polymer. Biomaterials 2003;23:43–52. Chung CW, Kim HW, Kim YB, Rhee YH. Poly(ethylene glycol)-grafted poly(3-hydroxyundecenoate) networks for enhanced blood compatibility. Int J Biol Macromol 2003;32:17–22.
