Japanese Society for Biomaterials Award Winner for Junior Investigator in Biomaterials Science 1992 Hemocompatibility of human whole blood on polymers with a phospholipid polar group and its mechanism Kazuhiko Ishihara,* Hiroko Oshida, Yutaka Endo, Tomoko Ueda, Akihiko Watanabe, and Nobuo Nakabayashi lnstitute for Medical and Dental Engineering, Tokyo Medical and Dental University, 2-3-10, Kanda-surugadai, Chiyoda-ku, Tokyo 101, Japan The hemocompatibility of a polymer containing a phospholipid polar group, poly(2-methacryloyloxyethyl phosphorylcholine (MPC)-co-n-butyl methacrylate(BMA)), with human whole blood was evaluated. When human whole blood without an anticoagulant was contacted with polymers, the blood cell adhesion and aggregation on the polymer without the MPC moiety was extensive, and considerable fibrin deposition was observed. This phenomenon was suppressed with an increase in the polymer MPC composition. Thus, the MPC moiety in the copolymer plays an important role in the nonthrombogenic behavior of the copolymer. These results were also confirmed
by the whole blood coagulation time on the polymer surface which was determined by Lee-White method. The adsorption of phospholipids and proteins from human plasma on poly(MPC-co-BMA) was investigated to clarify the mechanism of the nonthrombogenicity observed with the polymer. The amount of phospholipids was increased; whereas, adsorbed proteins were decreased with an increase in the MPC composition. From these results, we concluded that the phospholipids adsorbed o n poly(MPC-coBMA) play the most important role in the nonthrombogenicity of the MPC copolymer. Q 1992 John Wiley & Sons, Inc.
INTRODUCTION
Recently, we have developed new hemocompatible and nonthrombogenic polymers with attention to the surface of the biomembrane as this surface can interact in a compatible manner with biocomponents such as proteins and cells.'-4 We assumed that a polymer having a strong affinity for phospholipids could be used to construct a biomembrane-like structure by adsorbing phospholipids from blood and organizing them on the polymer surface. Phospho*To whom correspondence should be addressed.
Journal of Biomedical Materials Research, Vol. 26, 1543-1552 (1992) CCC 0021-9304/92/121543-10 8 1992 John Wiley & Sons, Inc.
ISHIHARA ET AL.
1544
lipid molecules have a self-organizing property with each other and form a bilayer membrane structure? Therefore, polymers having a phospholipid polar group are expected to have a strong affinity for phospholipid molecules. We prepared methacrylate with a phospholipid polar group in the side chain, 2-methacryloyloxyethyl phosphorylcholine(MPC), and its copolymers with various hydrophobic methacrylates and styrene."' Among these copolymers, poly(MPC-co-n-butyl methacrylate (BMA))s exhibit excellent hemocompatibility as shown by reduction of platelet adhesion and aggregation, and suppression of protein adsorption?" In this communication, the nonthrombogenic properties of poly(MPC-coBMA) with human whole blood are investigated, and the mechanism of nonthrombogenicity observed on the poly(MPC-co-BMA) surface is discussed. EXPERIMENTAL
Materials MPC was synthesized by the method reported previously and was used after recrystallization from acetonitrile.6 Poly(MPC-co-BMA)swere prepared by conventional radical copolymerization using a,a'-azobisisobutyronitrile (AIBN) as an initiator in ethanol. Poly(MPC-co-methyl methacrylate(MMA) was also synthesized in the same manner. The structures of MPC copolymers are shown in Fig. 1. Poly(2-hydroxyethyl methacrylate(HEMA)) and poly(BMA) were synthesized by radical polymerization of the corresponding monomers using AIBN as an initiator. The molecular weights of these polymers were in the range of 1.1-3.1 X lo', which was determined by gel permeation chromatography (co1umn:TSK gel H-type, Tosoh, Tokyo, Japan, eluent:N, N-dimethylformamide) with polystryrene standards (#06476, Tosoh). Acrylic beads (poly(MMA) cross-linked with 1 wt% of ethylene glycol dimethacrylate, average diameter: 350-700 pm) were obtained from Fijikura Chemical Co. Ltd. (Tokyo, Japan). Human whole blood was collected from healthy donors with or without the use of sodium citrate as an anticoagulant. Human plasma was prepared by centrifugation (2800 rpm, 10 min) of the citrated whole blood. The amounts of protein and phospholipid adCH3
CH3
I
+C-CH2
I
c=o ~I
I
ja
(-C-CHzk
0I
I
+
OCH2CH20YOCH2CH2N(CH3)3
c=o
I O(CH2)nH
ij n=l :poly(MPC-co-MMA)
n=4 :poly(MPC-co-BMA) Figure 1. Structure of MPC copolymers.
HEMOCOMPATIBLE PHOSPHOLIPID POLYMER
1545
sorbed were determined with clinical analysis kits, the micro BCA Protein Assay Reagent (#23235, PIERCE, Rockford, IL) and the Phospholipid Test Wako (C-type, Wako Pure Chemical, Osaka, Japan), respectively. Evaluation of blood coagulation on polymer surface LeeWhite test": The test polymers were coated on the inside of glass test tubes (10 mm in i.d., 10 cm in length) by a solvent evaporation method. Five milliliters of human whole blood without anticoagulant was introduced in the polymer-coated test tube, and the samples were shaken gently in a water bath at 37°C. The time was measured until the blood fluidity disappeared. The experiments were carried out three times for each samples, and the mean value (2 standard deviation) on each polymer was indicated. Comparative analyses were determined using analysis of variance and the Student's t test. Microsphere-column method': The polymers tested in this study were coated on acrylic beads by a solvent evaporation method. Diethyl ether was used as a solvent for poly(BMA), ethanol was used that for the other polymers. The surface of the beads was analyzed with an x-ray photoelectron spectroscope (XPS, Shimadzu, ESCA-750, Kyoto, Japan) to confirm the coverage of the beads with the polymers. The MPC mole fraction on the surface was determined from the ratio of phosphorus and carbon atoms (See Table I). The polymer-coated beads (0.52 g) were packed into poly(viny1 chloride) (PVC) tubing (3 mm in i.d., 10 cm in length, Eiken, Tokyo, Japan) equipped with a three-way stopcock to form a column. The packed column was primed with phosphate buffered solution (PBS, pH 7.4,ionic strength; 0.15M) for 24 h to exclude the liquid-air interface and to equilibrate the polymer surface with the physiological environment. After human whole blood was collected without anticoagulant, the whole blood was immediately and continuously infused into the column packed with polymer beads at a flow rate of 0.23 mL/min with the use of an infusion pump for 15 min. Then after the column was rinsed with PBS, PBS containing 2 wt% glutaraldehyde was continuously infused into the column for 10 min to fix the biocomponents adhered on the polymer beads. The beads were rinsed with distilled water and TABLE I Whole Blood Coagulation Time Determined by the Lee-White Method Sample
MPC Mole Fraction
Glass Poly (BMA) Poly(HEMA) Poly(MPC-co-MMA) Poly(MPC-co-BMA)
-
0.18 0.26
Coagulation Time (min)' 8.4 9.6 21 21 28
2 0.46
2 1.3
1.2 0.58 2 2.6
5 5
;qp t Test
> 0.05 c 0.01 **
' 2 value indicates standard deviation and number of samples is 3 for each polymer.
"No significant difference between these values ( p > 0.05). **Significant difference @ < 0.01).
ISIIIHARA ET AI..
1546
freeze-dried. The surface of the beads was observed with a scanning electron microscope (SEM, Comtec, CSM-501, Tokyo, Japan). Analysis of adsorbed protein and phospholipid One gram of polymer-coated beads was packed into PVC tubing (i.d. = 5 mm, length = 11 cm, Sekisui Chemical, Tokyo, Japan) equipped with a three-way stopcock to form a column. The column was filled with distilled water and stored at room temperature (about 26°C) for 12 h to equilibrate the surface of the polymer-coated beads. After the water was evacuated, 1.5 mL of human plasma was introduced into the column and stored at 37°C for 1 h. The plasma was removed and the beads were washed with 6 mL of distilled water to completely remove unadsorbed substances. Then 3 mL of 1 wt% sodium dodecyl sulfonate (SDS) solution was introduced into the column to desorb the proteins and phospholipids adsorbed on the polymer beads. The SDS solution was removed, and the beads were then rinsed twice with 3 mL of distilled water. These solutions were combined and the concentrations of proteins and lipids in the solution were determined by a colorimetric method using analytical kits for clinical use. The amount of proteins and lipids adsorbed on the polymer beads were calculated from the concentration of the eluted solution and the surface area of the beads packed in the column (55.4 an2). The bar graphs represent the mean values of duplicate samples (2 standard deviation) of the polymers. Comparative analyses were determined using analysis of variance and the Student’s t test.
RESULTS
Nonthrombogenicity on the MPC copolymer In Table I, the total coagulation time on the polymer is summarized. The Lee-White test indicated reduced thrombogenicity on the MPC copolymers. The total coagulation time on glass was 8.4 5 0.46 min and that on poly(BMA) (no MPC) was 9.6 2 1.3 min. However, there was no significant difference (p > 0.05j.With the coating of poly(HEMA) and MPC copolymers, the coagulation time was significantly increased compared with glass and poly(BMA) coating case (p < 0.01). On the surface of poly(MPC-co-BMA), the longest times for coagulation, 28 2 2.6 min, were observed. After human whole blood without an anticoagulant was passed through the columns packed with polymer-coated beads for 15 min and the column was rinsed with PBS, the color of the column with poly(BMA) turned red, whereas the poly(MPC-co-BMA) columns was more clear. Figure 2 shows SEM pictures of the polymer-coated beads after contact with human whole blood. On the surface of poly(BMA), a fibrin net completely covered the bead surfaces and many blood cells were adhered. On the other hand, no fibrin de-
HEMOCOMPATIBLE PHOSPHOLIPID POLYMER
1547
ISHIHARA ET AL.
1548
position could be found on poly(MPC-co-BMA)s. On the MPC copolymer with a 0.32 MPC mole fraction, a few blood red cells without deformation were found. Amount of biological molecules adsorbed on the MPC copolymer
Figures 3 and 4 show the adsorbed amounts of phospholipids and proteins on the polymers from human plasma, respectively. The amount of phospholipids adsorbed on hydrophobic poly(BMA) was almost the same as that on hydrophilic poly(HEMA) (p > 0.05). On poly(MPC-co-BMA), a larger amount of phospholipids was adsorbed compared with poly(BMA) (p < 0.01), and the amount adsorbed increased with an increase in the MPC mole fraction of the copolymer. On the other hand, the amount of protein on poly(HEMA) was significantly smaller than that on poly(BMA) (p < 0.01), and it was at the same level as on poly(MPC-co-BMA) with a 0.13 MPC mole fraction. On poly(MPC-co-BMA) with a 0.26 MPC mole fraction, a small amount of protein was adsorbed. DISCUSSION
We have already reported on the nonthrombogenic properties of polymers containing a phospholipid moiety. The platelet adhesion and activation were completely suppressed on the surface of the MPC copolymers when the MPC
%
I Y
Figure 3. Amount of phospholipid adsorbed on poly(MPC-co-BMA) and poly(HEMA) from human plasma. Data are the mean values of duplicate samples (2 standard deviation) of the polymers.
HEMOCOM PATIBLE PHOSPHOLIPID POLYMER
1549
A
%
8 Y
1.o
0.8
0.6 0.4 ) .
0
E
f
0.2
0.0
POly(MPC~-BMA) PW(HEMA) YPC mole fractlon In cowlvmor 0; Poly(BMA) 0.13 0.28
. -
Figure 4. Amount of proteins adsorbed on poly(MPC-co-BMA) and poly (HEMA) from human plasma. Data are the mean values of duplicate samples (? standard deviation) of the polymers.
composition was above 30 m01%.~These good nonthrombogenic properties appeared when the MPC copolymers contacted human whole blood even in the absence of an anticoagulant, as shown in Table I and Fig. 2. The interesting property of the MPC copolymer is its affinity for phospholipids.'*12The amount of phospholipids, diparmitoyl phosphatidylcholine (DPPC), adsorbed on MPC copolymers was larger than that on poly(BMA) and poly(HEMA) and increased with increasing MPC moiety when the MPC copolymers contacted a liposomal solution of DPPC.13This tendency was the same as that of the adsorption of phospholipids from human plasma which is indicated in Figure 3. Thus, the affinity of poly(MPC-co-BMA) for the phospholipids could be observed even in the plasma. The DPPC molecules adsorbed on the poly(MPC-co-BMA) surface assumed an organized structure like that for a bilayer membrane, which was confirmed by differential scanning calorimetric analysis and XPS when the poly(MPC-co-BMA) membrane was immersed in the solution containing DPPC." It is therefore concluded that the MPC copolymers stabilized the adsorption layer of phospholipids on the surface. Little albumin and y-globulins were adsorbed on the MPC copolymer surface pretreated with DPPC, which is in sharp contrast to the fact that on a poly(BMA) surface, pretreatment of DPPC was not effective for suppression of protein adsorption. The difference in protein adsorption on these polymer surfaces reflect the difference in the orientation of the DPPC molecules which cover the polymer surface." The total amount of proteins adsorbed from human plasma on poly(MPCco-BMA) with a 0.26 MPC mole fraction was quite small compared with that on other polymers tested as shown in Figure 4. It is clearly demonstrated that
1550
ISHIHARA ET AL.
x
c
w
0
HEMOCOMPATIBLE PHOSPHOLIPID POLYMER
1551
there are weak interactions between the surface and proteins. The protein composition and distribution on the blood-contacting surface of MPC copolymers from human plasma has been determined by radioimmunoassay and immunogold labeling te~hnique.'~ The surface concentration of all proteins including fibrinogen, coagulation factors, and complement component decreased with an increase in the MPC moiety and appeared to absorb on the surfaces in a uniform and evenly distributed manner. By comparison of the adsorption behavior of phospholipids on MPC copolymers with that of proteins, a mechanism of nonthrombogenicity observed on MPC copolymers is considered as shown in Figure 5. Since the molecular size of phospholipids is smaller than that of proteins and the molar concentration of phospholipids (wt. conc. = 187 2 34 mg/mL, average molecular weight of phospholipids: 7 X lo2, molar conc. = 2.2-3.2 X lo-' mol/ dL) is larger than that of proteins (wt. conc. = 6.5-8.0 g/dL, average molecular weight of proteins: 1.5 X lo5, molar conc. = 4.3-5.3 X mol/dL), the diffusion of phospholipid molecules from plasma to the polymer surface occurs more easily than proteins.I6Moreover, the MPC copolymers have both a strong affinity for phospholipids and protein adsorption-resistant properties in buffered aqueous protein ~olution.'~*'~ From these findings, we conclude that the phospholipids in plasma are adsorbed immediately on the surface of MPC copolymers, undergo self-organization and form a stable adsorbed layer with a biomembrane-like surface. The biomembrane-like surface interacts minimally with proteins and cells and therefore inhibit thrombus formation. Since a part of this study was supported by a Grant-in-Aid for Scientific Research from the Ministry of Education, Japan (03205030) and Terumo Scientific Foundation, one of the authors (K.I.) would like to express his appreciation for this support.
References 1. N. Nakabayashi, K. Ishihara, A. Watanabe, and M. Kojima, "New type nonthrombogenic materials," in Polymers and biomaterials, H. Feng, Y. Han, and L. Huang (eds.), Elsevier Science Publishers, Amsterdam, 1991, pp. 343-351. 2. J. A. Hayward and D. Chapman, "Biomembrane surfaces as model for polymer design: the potential for haemocompatibility," Biomaterials, 5, 135-142 (1984). 3. K. Kano, Y. Ito, S. Kimura, and Y. Imanishi, "Platelet adhesion on to polyamide microcapsules coated with lipid bilayer membrane," Biomaterials, 10, 455-461 (1989). 4. Y. Nozaki, Y. Yamamoto, and T. Akaike, "Blood compatibility of lipid surface," Jpn. J. Artif. Organs, 13, 1147-1150 (1984). 5. R. 8. Gennis, Biomembranes: Molecular structure and function, SpringerVerlag, New York, 1989. 6. K. Ishihara, T. Ueda, and N. Nakabayashi, "Preparation of phospholipid polymers and their properties as polymer hydrogel membranes," P ~ l y mJ., . 22, 355-360 (1990). 7. T. Ueda, H. Oshida, K. Kurita, K. Ishihara, and N. Nakabayashi, "Preparation of 2-methacryloyloxyethyl phosphorylcholine copolymers with alkyl methacrylates and their blood compatibility," in press.
ISHIHARA ET AL.
1552 8.
9. 10. 11. 12. 13.
14.
15.
16.
M. Kojima, K. Ishihara, A. Watanabe, and N. Nakabayashi, "lnteraction between phospholipids and biocompatible polymers containing phosphorylcholine moiety," Bimterials, 12, 121-124 (1991). K. Ishihara, R. Aragaki, T. Ueda, A. Watanabe, and N. Nakabayashi, "Reduced thrombogenicity of polymers having phospholipid polar group," J. Biomed. M i t e r . Res., 24, 1069-1077 (1990). K. Ishihara, "Suppression of protein adsorption on the polymers having phospholipid polar group," J. Jpn. Soc. Biaater. (Seitai Zairyo, in Japanese), 9,296-302 (1991). G.-H. Hsiue, J.-M. Yang, and R.-L. Wu, "Preparation and properties of a biomaterial: HEMA grafted SBS by 7-irradiation," 1. Biomed. Miter. Res., 22, 405-415 (1988). K. Ishihara and N. Nakabayashi, "Specific interaction between watersoluble phospholipid polymer and liposome," J. Polym. Sci.: Part A: P ~ l y mChem., . 29, 831-835 (1991). K. Ishihara, T. Ueda, Y. Endo, A. Watanabe, and N. Nakabayashi, "Effect of phospholipid adsorption on blood compatibility of polymer surface with phospholipid polar group," Polym. Preprints, Jpn., 39, 616 (1990). T. Ueda, A. Watanabe, K. Ishihara, and N. Nakabayashi, "Protein adsorption on biomedical polymers with a phosphorylcholine moiety adsorbed with phospholipid," J. B i a a t e r . Sci.; Polym. Ed., 3, 185-194 (1991). K. Ishihara, N.P. Ziats, B. P. Tierney, N. Nakabayashi, and J.M. Anderson, "Protein adsorption from human plasma is reduced on phospholipid polymers," J. Biomed. Muter. Res., 25, 1397-1407 (1991). T. Akaike, "Control of degradation of biomedical polymers in biological systems," in Biomaterials science, vol. 1, Y. Sakurai and T. Tsuruta (eds.), Kagaku no Ryoiki Zokan No. 134,Nankodo, Tokyo, Japan, 1982, pp. 183-197.
Received April 22, 1992 Accepted May 27,1992
by the whole blood coagulation time on the polymer surface which was determined by Lee-White method. The adsorption of phospholipids and proteins from human plasma on poly(MPC-co-BMA) was investigated to clarify the mechanism of the nonthrombogenicity observed with the polymer. The amount of phospholipids was increased; whereas, adsorbed proteins were decreased with an increase in the MPC composition. From these results, we concluded that the phospholipids adsorbed o n poly(MPC-coBMA) play the most important role in the nonthrombogenicity of the MPC copolymer. Q 1992 John Wiley & Sons, Inc.
INTRODUCTION
Recently, we have developed new hemocompatible and nonthrombogenic polymers with attention to the surface of the biomembrane as this surface can interact in a compatible manner with biocomponents such as proteins and cells.'-4 We assumed that a polymer having a strong affinity for phospholipids could be used to construct a biomembrane-like structure by adsorbing phospholipids from blood and organizing them on the polymer surface. Phospho*To whom correspondence should be addressed.
Journal of Biomedical Materials Research, Vol. 26, 1543-1552 (1992) CCC 0021-9304/92/121543-10 8 1992 John Wiley & Sons, Inc.
ISHIHARA ET AL.
1544
lipid molecules have a self-organizing property with each other and form a bilayer membrane structure? Therefore, polymers having a phospholipid polar group are expected to have a strong affinity for phospholipid molecules. We prepared methacrylate with a phospholipid polar group in the side chain, 2-methacryloyloxyethyl phosphorylcholine(MPC), and its copolymers with various hydrophobic methacrylates and styrene."' Among these copolymers, poly(MPC-co-n-butyl methacrylate (BMA))s exhibit excellent hemocompatibility as shown by reduction of platelet adhesion and aggregation, and suppression of protein adsorption?" In this communication, the nonthrombogenic properties of poly(MPC-coBMA) with human whole blood are investigated, and the mechanism of nonthrombogenicity observed on the poly(MPC-co-BMA) surface is discussed. EXPERIMENTAL
Materials MPC was synthesized by the method reported previously and was used after recrystallization from acetonitrile.6 Poly(MPC-co-BMA)swere prepared by conventional radical copolymerization using a,a'-azobisisobutyronitrile (AIBN) as an initiator in ethanol. Poly(MPC-co-methyl methacrylate(MMA) was also synthesized in the same manner. The structures of MPC copolymers are shown in Fig. 1. Poly(2-hydroxyethyl methacrylate(HEMA)) and poly(BMA) were synthesized by radical polymerization of the corresponding monomers using AIBN as an initiator. The molecular weights of these polymers were in the range of 1.1-3.1 X lo', which was determined by gel permeation chromatography (co1umn:TSK gel H-type, Tosoh, Tokyo, Japan, eluent:N, N-dimethylformamide) with polystryrene standards (#06476, Tosoh). Acrylic beads (poly(MMA) cross-linked with 1 wt% of ethylene glycol dimethacrylate, average diameter: 350-700 pm) were obtained from Fijikura Chemical Co. Ltd. (Tokyo, Japan). Human whole blood was collected from healthy donors with or without the use of sodium citrate as an anticoagulant. Human plasma was prepared by centrifugation (2800 rpm, 10 min) of the citrated whole blood. The amounts of protein and phospholipid adCH3
CH3
I
+C-CH2
I
c=o ~I
I
ja
(-C-CHzk
0I
I
+
OCH2CH20YOCH2CH2N(CH3)3
c=o
I O(CH2)nH
ij n=l :poly(MPC-co-MMA)
n=4 :poly(MPC-co-BMA) Figure 1. Structure of MPC copolymers.
HEMOCOMPATIBLE PHOSPHOLIPID POLYMER
1545
sorbed were determined with clinical analysis kits, the micro BCA Protein Assay Reagent (#23235, PIERCE, Rockford, IL) and the Phospholipid Test Wako (C-type, Wako Pure Chemical, Osaka, Japan), respectively. Evaluation of blood coagulation on polymer surface LeeWhite test": The test polymers were coated on the inside of glass test tubes (10 mm in i.d., 10 cm in length) by a solvent evaporation method. Five milliliters of human whole blood without anticoagulant was introduced in the polymer-coated test tube, and the samples were shaken gently in a water bath at 37°C. The time was measured until the blood fluidity disappeared. The experiments were carried out three times for each samples, and the mean value (2 standard deviation) on each polymer was indicated. Comparative analyses were determined using analysis of variance and the Student's t test. Microsphere-column method': The polymers tested in this study were coated on acrylic beads by a solvent evaporation method. Diethyl ether was used as a solvent for poly(BMA), ethanol was used that for the other polymers. The surface of the beads was analyzed with an x-ray photoelectron spectroscope (XPS, Shimadzu, ESCA-750, Kyoto, Japan) to confirm the coverage of the beads with the polymers. The MPC mole fraction on the surface was determined from the ratio of phosphorus and carbon atoms (See Table I). The polymer-coated beads (0.52 g) were packed into poly(viny1 chloride) (PVC) tubing (3 mm in i.d., 10 cm in length, Eiken, Tokyo, Japan) equipped with a three-way stopcock to form a column. The packed column was primed with phosphate buffered solution (PBS, pH 7.4,ionic strength; 0.15M) for 24 h to exclude the liquid-air interface and to equilibrate the polymer surface with the physiological environment. After human whole blood was collected without anticoagulant, the whole blood was immediately and continuously infused into the column packed with polymer beads at a flow rate of 0.23 mL/min with the use of an infusion pump for 15 min. Then after the column was rinsed with PBS, PBS containing 2 wt% glutaraldehyde was continuously infused into the column for 10 min to fix the biocomponents adhered on the polymer beads. The beads were rinsed with distilled water and TABLE I Whole Blood Coagulation Time Determined by the Lee-White Method Sample
MPC Mole Fraction
Glass Poly (BMA) Poly(HEMA) Poly(MPC-co-MMA) Poly(MPC-co-BMA)
-
0.18 0.26
Coagulation Time (min)' 8.4 9.6 21 21 28
2 0.46
2 1.3
1.2 0.58 2 2.6
5 5
;qp t Test
> 0.05 c 0.01 **
' 2 value indicates standard deviation and number of samples is 3 for each polymer.
"No significant difference between these values ( p > 0.05). **Significant difference @ < 0.01).
ISIIIHARA ET AI..
1546
freeze-dried. The surface of the beads was observed with a scanning electron microscope (SEM, Comtec, CSM-501, Tokyo, Japan). Analysis of adsorbed protein and phospholipid One gram of polymer-coated beads was packed into PVC tubing (i.d. = 5 mm, length = 11 cm, Sekisui Chemical, Tokyo, Japan) equipped with a three-way stopcock to form a column. The column was filled with distilled water and stored at room temperature (about 26°C) for 12 h to equilibrate the surface of the polymer-coated beads. After the water was evacuated, 1.5 mL of human plasma was introduced into the column and stored at 37°C for 1 h. The plasma was removed and the beads were washed with 6 mL of distilled water to completely remove unadsorbed substances. Then 3 mL of 1 wt% sodium dodecyl sulfonate (SDS) solution was introduced into the column to desorb the proteins and phospholipids adsorbed on the polymer beads. The SDS solution was removed, and the beads were then rinsed twice with 3 mL of distilled water. These solutions were combined and the concentrations of proteins and lipids in the solution were determined by a colorimetric method using analytical kits for clinical use. The amount of proteins and lipids adsorbed on the polymer beads were calculated from the concentration of the eluted solution and the surface area of the beads packed in the column (55.4 an2). The bar graphs represent the mean values of duplicate samples (2 standard deviation) of the polymers. Comparative analyses were determined using analysis of variance and the Student’s t test.
RESULTS
Nonthrombogenicity on the MPC copolymer In Table I, the total coagulation time on the polymer is summarized. The Lee-White test indicated reduced thrombogenicity on the MPC copolymers. The total coagulation time on glass was 8.4 5 0.46 min and that on poly(BMA) (no MPC) was 9.6 2 1.3 min. However, there was no significant difference (p > 0.05j.With the coating of poly(HEMA) and MPC copolymers, the coagulation time was significantly increased compared with glass and poly(BMA) coating case (p < 0.01). On the surface of poly(MPC-co-BMA), the longest times for coagulation, 28 2 2.6 min, were observed. After human whole blood without an anticoagulant was passed through the columns packed with polymer-coated beads for 15 min and the column was rinsed with PBS, the color of the column with poly(BMA) turned red, whereas the poly(MPC-co-BMA) columns was more clear. Figure 2 shows SEM pictures of the polymer-coated beads after contact with human whole blood. On the surface of poly(BMA), a fibrin net completely covered the bead surfaces and many blood cells were adhered. On the other hand, no fibrin de-
HEMOCOMPATIBLE PHOSPHOLIPID POLYMER
1547
ISHIHARA ET AL.
1548
position could be found on poly(MPC-co-BMA)s. On the MPC copolymer with a 0.32 MPC mole fraction, a few blood red cells without deformation were found. Amount of biological molecules adsorbed on the MPC copolymer
Figures 3 and 4 show the adsorbed amounts of phospholipids and proteins on the polymers from human plasma, respectively. The amount of phospholipids adsorbed on hydrophobic poly(BMA) was almost the same as that on hydrophilic poly(HEMA) (p > 0.05). On poly(MPC-co-BMA), a larger amount of phospholipids was adsorbed compared with poly(BMA) (p < 0.01), and the amount adsorbed increased with an increase in the MPC mole fraction of the copolymer. On the other hand, the amount of protein on poly(HEMA) was significantly smaller than that on poly(BMA) (p < 0.01), and it was at the same level as on poly(MPC-co-BMA) with a 0.13 MPC mole fraction. On poly(MPC-co-BMA) with a 0.26 MPC mole fraction, a small amount of protein was adsorbed. DISCUSSION
We have already reported on the nonthrombogenic properties of polymers containing a phospholipid moiety. The platelet adhesion and activation were completely suppressed on the surface of the MPC copolymers when the MPC
%
I Y
Figure 3. Amount of phospholipid adsorbed on poly(MPC-co-BMA) and poly(HEMA) from human plasma. Data are the mean values of duplicate samples (2 standard deviation) of the polymers.
HEMOCOM PATIBLE PHOSPHOLIPID POLYMER
1549
A
%
8 Y
1.o
0.8
0.6 0.4 ) .
0
E
f
0.2
0.0
POly(MPC~-BMA) PW(HEMA) YPC mole fractlon In cowlvmor 0; Poly(BMA) 0.13 0.28
. -
Figure 4. Amount of proteins adsorbed on poly(MPC-co-BMA) and poly (HEMA) from human plasma. Data are the mean values of duplicate samples (? standard deviation) of the polymers.
composition was above 30 m01%.~These good nonthrombogenic properties appeared when the MPC copolymers contacted human whole blood even in the absence of an anticoagulant, as shown in Table I and Fig. 2. The interesting property of the MPC copolymer is its affinity for phospholipids.'*12The amount of phospholipids, diparmitoyl phosphatidylcholine (DPPC), adsorbed on MPC copolymers was larger than that on poly(BMA) and poly(HEMA) and increased with increasing MPC moiety when the MPC copolymers contacted a liposomal solution of DPPC.13This tendency was the same as that of the adsorption of phospholipids from human plasma which is indicated in Figure 3. Thus, the affinity of poly(MPC-co-BMA) for the phospholipids could be observed even in the plasma. The DPPC molecules adsorbed on the poly(MPC-co-BMA) surface assumed an organized structure like that for a bilayer membrane, which was confirmed by differential scanning calorimetric analysis and XPS when the poly(MPC-co-BMA) membrane was immersed in the solution containing DPPC." It is therefore concluded that the MPC copolymers stabilized the adsorption layer of phospholipids on the surface. Little albumin and y-globulins were adsorbed on the MPC copolymer surface pretreated with DPPC, which is in sharp contrast to the fact that on a poly(BMA) surface, pretreatment of DPPC was not effective for suppression of protein adsorption. The difference in protein adsorption on these polymer surfaces reflect the difference in the orientation of the DPPC molecules which cover the polymer surface." The total amount of proteins adsorbed from human plasma on poly(MPCco-BMA) with a 0.26 MPC mole fraction was quite small compared with that on other polymers tested as shown in Figure 4. It is clearly demonstrated that
1550
ISHIHARA ET AL.
x
c
w
0
HEMOCOMPATIBLE PHOSPHOLIPID POLYMER
1551
there are weak interactions between the surface and proteins. The protein composition and distribution on the blood-contacting surface of MPC copolymers from human plasma has been determined by radioimmunoassay and immunogold labeling te~hnique.'~ The surface concentration of all proteins including fibrinogen, coagulation factors, and complement component decreased with an increase in the MPC moiety and appeared to absorb on the surfaces in a uniform and evenly distributed manner. By comparison of the adsorption behavior of phospholipids on MPC copolymers with that of proteins, a mechanism of nonthrombogenicity observed on MPC copolymers is considered as shown in Figure 5. Since the molecular size of phospholipids is smaller than that of proteins and the molar concentration of phospholipids (wt. conc. = 187 2 34 mg/mL, average molecular weight of phospholipids: 7 X lo2, molar conc. = 2.2-3.2 X lo-' mol/ dL) is larger than that of proteins (wt. conc. = 6.5-8.0 g/dL, average molecular weight of proteins: 1.5 X lo5, molar conc. = 4.3-5.3 X mol/dL), the diffusion of phospholipid molecules from plasma to the polymer surface occurs more easily than proteins.I6Moreover, the MPC copolymers have both a strong affinity for phospholipids and protein adsorption-resistant properties in buffered aqueous protein ~olution.'~*'~ From these findings, we conclude that the phospholipids in plasma are adsorbed immediately on the surface of MPC copolymers, undergo self-organization and form a stable adsorbed layer with a biomembrane-like surface. The biomembrane-like surface interacts minimally with proteins and cells and therefore inhibit thrombus formation. Since a part of this study was supported by a Grant-in-Aid for Scientific Research from the Ministry of Education, Japan (03205030) and Terumo Scientific Foundation, one of the authors (K.I.) would like to express his appreciation for this support.
References 1. N. Nakabayashi, K. Ishihara, A. Watanabe, and M. Kojima, "New type nonthrombogenic materials," in Polymers and biomaterials, H. Feng, Y. Han, and L. Huang (eds.), Elsevier Science Publishers, Amsterdam, 1991, pp. 343-351. 2. J. A. Hayward and D. Chapman, "Biomembrane surfaces as model for polymer design: the potential for haemocompatibility," Biomaterials, 5, 135-142 (1984). 3. K. Kano, Y. Ito, S. Kimura, and Y. Imanishi, "Platelet adhesion on to polyamide microcapsules coated with lipid bilayer membrane," Biomaterials, 10, 455-461 (1989). 4. Y. Nozaki, Y. Yamamoto, and T. Akaike, "Blood compatibility of lipid surface," Jpn. J. Artif. Organs, 13, 1147-1150 (1984). 5. R. 8. Gennis, Biomembranes: Molecular structure and function, SpringerVerlag, New York, 1989. 6. K. Ishihara, T. Ueda, and N. Nakabayashi, "Preparation of phospholipid polymers and their properties as polymer hydrogel membranes," P ~ l y mJ., . 22, 355-360 (1990). 7. T. Ueda, H. Oshida, K. Kurita, K. Ishihara, and N. Nakabayashi, "Preparation of 2-methacryloyloxyethyl phosphorylcholine copolymers with alkyl methacrylates and their blood compatibility," in press.
ISHIHARA ET AL.
1552 8.
9. 10. 11. 12. 13.
14.
15.
16.
M. Kojima, K. Ishihara, A. Watanabe, and N. Nakabayashi, "lnteraction between phospholipids and biocompatible polymers containing phosphorylcholine moiety," Bimterials, 12, 121-124 (1991). K. Ishihara, R. Aragaki, T. Ueda, A. Watanabe, and N. Nakabayashi, "Reduced thrombogenicity of polymers having phospholipid polar group," J. Biomed. M i t e r . Res., 24, 1069-1077 (1990). K. Ishihara, "Suppression of protein adsorption on the polymers having phospholipid polar group," J. Jpn. Soc. Biaater. (Seitai Zairyo, in Japanese), 9,296-302 (1991). G.-H. Hsiue, J.-M. Yang, and R.-L. Wu, "Preparation and properties of a biomaterial: HEMA grafted SBS by 7-irradiation," 1. Biomed. Miter. Res., 22, 405-415 (1988). K. Ishihara and N. Nakabayashi, "Specific interaction between watersoluble phospholipid polymer and liposome," J. Polym. Sci.: Part A: P ~ l y mChem., . 29, 831-835 (1991). K. Ishihara, T. Ueda, Y. Endo, A. Watanabe, and N. Nakabayashi, "Effect of phospholipid adsorption on blood compatibility of polymer surface with phospholipid polar group," Polym. Preprints, Jpn., 39, 616 (1990). T. Ueda, A. Watanabe, K. Ishihara, and N. Nakabayashi, "Protein adsorption on biomedical polymers with a phosphorylcholine moiety adsorbed with phospholipid," J. B i a a t e r . Sci.; Polym. Ed., 3, 185-194 (1991). K. Ishihara, N.P. Ziats, B. P. Tierney, N. Nakabayashi, and J.M. Anderson, "Protein adsorption from human plasma is reduced on phospholipid polymers," J. Biomed. Muter. Res., 25, 1397-1407 (1991). T. Akaike, "Control of degradation of biomedical polymers in biological systems," in Biomaterials science, vol. 1, Y. Sakurai and T. Tsuruta (eds.), Kagaku no Ryoiki Zokan No. 134,Nankodo, Tokyo, Japan, 1982, pp. 183-197.
Received April 22, 1992 Accepted May 27,1992
