Reduced thrornbogenicity of polymers having phospholipid polar groups Kazuhiko Ishiham,* Runa Aragaki, Tomoko Ueda, Akihiko Watenabe, and Nobuo Na kabayashi* Institute for Medical and Dental Engineering, Tokyo Medical and Denfal University, 2-3-10, Kanda-Surugadai, Chiyoda-ku, Tokyo 101, Japan The thrombogenicity of polymers having a phospholipid polar group, poly(2methacryloyloxyethyl phosphorylcholine (MPC)-co-n-butyl methacrylate (BMA)), was evaluated by a microsphere-column method with attention to the activation and adhesion of platelets on the polymer surface. When citrated platelet-rich plasma (PRP) contacted with the polymers, a large number of platelets adhered and aggregated on poly(BMA). The number of adherent platelets decreased and
deformation and aggregation were suppressed with increasing MPC composition. The same tendency was noted when Ca2+-re-addedPRP came in contact with the polymers. In the case of poly(MPC-coBMA) with 0.320 mole fraction of MPC, activation of platelets and formation of fibrin were completely suppressed. Therefore, MPC moieties in the polymer play an important role in the reduction of thrombogenicity of the polymer.
INTRODUCTION
Biomembranes have many functions such as construction of cell walls, permeation of indispensable substances, rejection of toxins, and, most importantly, excellent biocompatibility. The surface of a biomembrane only mildly interacts with blood proteins and cells and does not adsorb and activate these biological molecules. A biomembrane is a hybrid constituted mainly of two chemical classes, phospholipids and proteins. There is no covalent bonding to bind each molecule; thus, the surface of a biomembrane is heterogenic and dynamic. These features prompted us to propose new concepts for making nonthrombogenic artificial materials. Segmented polyurethanes and some block copolymers having heterogenic microdomain structure have been reported.'" Moreover, a hydrophilic polymer surface covered with grafted hydrophilic chains has the potential for suppressing adsorption of biological molecules due to the mobility of the grafted polymer chain.5 We assumed that if an artificial surface similar to a biomembrane was constituted by an arrangement of phospholipids from blood on the surface, the surface would show excellent nonthrombogenicity. In arranging the *To whom correspondence should be addressed Journal of Biomedical Materials Research, Vol. 24, 1069-1077 (1990) CCC 0021-9304190/081069-09$04.00 0 1990 John Wiley & Sons, Inc.
ISHIHARA ET AL.
1070
phospholipids, a polymer having phospholipid polar groups can be expected to possess an affinity for natural phospholipid molecules. In previous articles, synthesis of a new monomer having a phosphoryl choline moiety, 2-methacryloyloxyethyl phosphorylcholine (MPC), and the biocompatibility of MPC copolymers with methyl methacrylate were These copolymers show mild interactions with cells, however, the function of the MPC was not well studied. In this article, MPC copolymers with n -butyl methacrylate were synthesized and their nonthrombogenicity evaluated by a microsphere-column method' with attention to the MPC composition on the surface of the copolymers. EXPERIMENTAL
Materials 2-Methacryloyloxyethyl phosphorylcholine (MPC) was synthesized by a procedure previously reported and purified by recrystallization with acetonitrile.' The structure is shown in Figure 1. n-Butyl methacrylate (BMA) was purified by distillation under reduced pressure and a fraction of bp 68.5"C/30 mm Hg was used. Poly(MPC-co-BMA)was prepared conventionally using 2,2'-azoisobutyronitrile (AIBN) as an initiator. Poly(MPC-co-BMA) membrane was prepared by a solvent evaporation method to determine the MPC composition in the copolymer by analysis with x-ray photoelectron spectroscopy (XI'S, Shimadzu ESCA-750) and water content of the copolymer membrane. Poly(2-hydroxyethyl methacrylate) (poly(HEMA))was prepared by a conventional radical polymerization method using AIBN in
2-propanol. " Acrylic beads were prepared by suspension copolymerization of methyl methacrylate with triethyleneglycol dimethacrylate as a crosslinker. Diameters of the beads ranged from 200 to 600 pm.
Evaluation of thrombogenicity of the polymer surf ace A microsphere-column method was used for the evaluation of thrombogenicity of the polymer surface.8The detailed procedure was as follows. Polymer coating of acrylic beads was carried out by a solvent evaporation technique using a 0.5 wt% polymer solution. The polymer-coated beads
$=CH2 F=O + OCH,CH,O~OCH,CH,N(CH,),
0-
Figure 1. Structure of 2-methacryloyloxyethyl phosphorylcholine (MPC).
NONTHROMBOGENICITY OF PHOSPHOLIPID POLYMER
1071
were dried in vacuo for 24 h at room temperature. XPS analysis of the polymer coated beads was carried out to determine the MPC composition of the surface of the coated beads. The polymer-coated beads (0.52 g) were closely packed in a poly(viny1 chloride) (PVC) tubing (3.0 mm ID, 10 cm in length) equipped with a stopcock, and the packed column was primed with physiological saline for 24 h to exclude the liquid-air interface and to equilibrate the polymer surface with the physiological environment. Platelet-rich plasma (PRP) was prepared from 10 Japanese white rabbits weighing about 3.0 kg each. The carotid artery was cannulated using PVC tubing and 90 mL of fresh blood per rabbit was collected in a disposable syringe containing 10 mL of a 3.8 wt% aqueous sodium citrate solution. The citrated blood was immediately centrifuged for 15 min at 750 rpm to obtain citrated PRP. The number of platelets of PRP (Po) was adjusted to approximately 1 X lo8 cells/mL by dilution with platelet-poor plasma, which was prepared from the citrated blood by centrifugation for 15 min at 2600 rpm. Calcium ion-re-added PRP was prepared by the addition of 119 pL of 1 mol/liter CaC1, aqueous solution into the citrated PRP. These PRP solutions were continuously infused into the column packed with polymer beads from a disposable syringe at a flow rate of 0.23 mL/min with the use of an infusion pump. A definite volume (50 pL) of the effluent PRP was collected and the number of effluent platelets (P) counted with a Coulter Counter. At least three packed columns were used for evaluation of the thrombogenicity of each polymer. Scanning electron microscopic (SEMI observation After the PRP was passed through the column for 20 min, the beads situated in the middle of the column were placed in a saline solution containing 2.0 wt% glutaraldehyde to fix the adherent platelets. These beads were rinsed with a water-ethanol system, dried with a critical point drying technique, and coated with gold. The morphology of platelets adherent to the bead surfaces was observed with SEM (Comtec CSM-501). RESULTS
Table I lists the MPC composition and water content of the poly (MPC-coBMA) used. X I'S data suggested that the MPC composition on the coated surfaces of acrylic beads was slightly larger than that in the bulk copolymer. Though the water content of poly(BMA) was nearly zero, it increased with increasing MPC composition. Thus, poly(MPC-co-BMA)was a hydrogel." The water content of MB-2 and MB-3 membranes was almost equal to that of the poly(HEMA) membrane. Figure 2 shows a representative elution profile of platelets from a column packed with polymer-coated beads using citrated PRP. Platelets began to
ISHIHARA ET AL.
1072
TABLE I Characterization of Poly(MPC-co-BMA) Used in This Study MPC Mole Fraction Code
In Feed
In Bulk Polymerb
On Coated Surfaceb
Water Content"
0.056 0.077 0.116 0.268 0
0.127 0.186 0.260 0.320 0
0.328 0.367 0.399
-
-
0.381
~~
MB-1 MB-2
0.03 0.05 0.10
MB-3 MB-4
0.40
Poly(BMA) Poly(HEMA)
n.d.' 0
"The values were obtained by measuring the weight of water in polymer membrane swollen in water at 30°C and calculation of the ratio of the weight of water to that of polymer membrane swollen. ?'he values were determined by the XPS analysis. 'The value could not be determined.
elute about 1 min after PRP was injected into the column, which corresponded to the void volume of the column. The elution ratio (P/Po) became constant after 20 min. Figure 3 shows a typical elution profile of platelets from the column when Ca2'-re-added PRP was applied to the column. Figure 3 also shows the result of poly(HEMA) compared to poly(MPC-co-BMA). The elution curves were quite different from those shown in Figure 2. The P/Po on poly(BMA) decreased drastically about 12 min after introduction of the Ca2+-re-added PRP. Elution of platelets from the column did not cease at 17 min. A similar
T / rnin
Figure 2. The representative elution profile of platelets from the column packed with poly(MPC-co-BMA)-coated beads when citrated PRP was injected into the column. Po, number of platelets injected; P, number of platelets eluted from the column. (A) poly(BMA), ( 0 ) MB-1, (0) MB-2, (A) MB-3, (0) MB-4.
1073
NONTHROMBOGENICITY OF PHOSPHOLIPID POLYMER 1 .o
Q ' \
a
0.5
0 0
10 T / min
5
15
20
Figure 3. The representative elution profile of platelets from the column packed with poly(MPC-co-BMA) coated beads and poly(HEMA)-coated beads when Ca*+-re-addedPRP was injected into the column. (A) poly(BMA), (a) MB-1, (0) MB-2, (A) MB-3, (0) MB4, (a) poly(HEMA).
profile was also found in the poly(HEMA) system. Platelet aggregations in the columns were observed in the poly(MPC-co-BMA), MB-1, MB-2, and MB-3 systems, however, their occurrence decreased as the MPC composition increased. On the other hand, with MB-4, P/Po maintained at 1.0 except during the introduction period of PRP. Figure 4 shows the MPC compositional dependence of platelet adhesion on polymer beads coated with poly(MPC-co-BMA) when Ca2+-re-added 0.5 0.4 T
a! \
0.3
a I
a0
0.2
v
0 .I 0
J
0
I
I
I
0.1 0.2 0.3 MPC mole fraction
0A
Figure 4. MPC composition dependence of the adhered platelets ratio on the poly(MPC-co-BMA) coated beads when Ca2'-re-added PRP was infused into the column for 15 min. Open plot, poly(MPC-co-BMA) system; closed plot, poly(HEMA). Number of columns used: poly(BMA), n = 5; poly(HEMA), n = 7; the other copolymers, n = 3. Mean values of the experiments are shown and bars represent standard error mean.
ISHIHARA ET AL.
1074
PRP was passed through the column packed with the beads for 15 min. The adhesion ratio of platelets (Po-P)/Po on the surface of poly(HEMA) is shown. The adhesion ratio decreased with an increase in the MPC composition in poly(MPC-co-BMA). There was a significant difference (P < 0.001) between homopolymers (poly(BMA), poly(HEMA)) and MPC copolymers with high MPC compositions (MB-2, MB-3, MB-4). Thus, it was clearly indicated that MPC moieties in poly(MPC-co-BMA) play an important role in suppressing the adhesion of platelets. Figure 5 shows SEM pictures of platelets adherent to the beads coated with poly(MPC-co-BMA) after 20 min of contact with citrated PRP. On the surface of poly(BMA), numerous platelets adhered with major changes in their morphology and aggregation. The number of platelets adherent to the surface of poly(MPC-co-BMA) decreased with increasing MPC composition. The morphology of adherent platelets was better than on the poly(BMA) and it improved with MPC copolymers. Figure 6 shows SEM pictures of polymer beads coated with poly(MPC-coBMA) after 20 min of contact with CaZ'-re-added PRP. It was found that a fibrin net completely covered the surface of every bead coated with poly(BMA). The density of fibrin fibers decreased with increasing MPC composition; it was difficult to observe any fibrin net on the surface of MB-4. DISCUSSION
A microsphere-column method containing beads coated with test samples is one of the most powerful methods for evaluating the thrombogenic-
Figure 5. SEM picture of poly(MPC-co-BMA) coated beads after contacting citrated PRP for 20 min. (A) poly(BMA), (B) MB-1, (C) MB-2, (D) MB-3, (E) MB4.
1075
NONTHROMBOGENICITY OF PHOSPHOLIPID POLYMER
c )
200pm Figure 6 . SEM picture of poly(MPC-co-BMA)-coated beads a n d poly(HEMA)-coated beads after contacting Caz+-re-addedPRP for 20 min. (A) poly(BMA), (B) MB-1, (C) MB-2, (D) MB-3, (E) MB-4, (F) poly(HEMA).
ity of polymer materials.8The elution profile of platelets reflects the process of thrombus formation, i.e., decrease in the number of platelets eluted from the column indicates activation of the cells and the beginning of thrombus formation. As shown in Figure 2, when citrated PRP was infused into the column, P/Po was almost constant for 20 min in every case, however, P/Po took lower values when the MPC composition was lowered, thus indicating that adherent platelets on the bead surface could not elute. From this, we concluded that the MPC moiety in poly(MPC-co-BMA) could suppress platelet adhesion on the surface. When Caz+-re-addedPRP was injected into the column, the effect of the MPC moiety became clearer. The Ca2' ions were required for the activation of platelets accompanying thrombus formation. Therefore, the activation of platelets was not induced by elimination of Ca2+ions with sodium citrate. The poly(BMA) column was occluded completely by a fibrin net within 20 min. On the other hand, platelet suspension could flow through the poly(MPC-co-BMA) columns. These results confirmed that MPC copolymers suppress not only platelet activation but also platelet adhesion, supported by direct SEM observation of the polymer surfaces after contact with PRP, as shown in Figures 5 and 6. Poly(MPC-co-BMA) is a hydrogel that has an extremely hydrophilic surface. In general, it is considered that the hydrophilic surface gently interacts with biological molecules because of its high surface free energy. l2 However, by comparison of MB-2 and MB-3 with poly(HEMA), which is a typical hydrogel that has almost the same water content as poly(MPC-co-BMA), it is seen that relatively large numbers of platelets adhere to poly(HEMA) (see Figs. 3 and 4).
1076
ISHIHARA ET AL.
Noishiki reported that a polyester vascular prosthesis implanted in the thoracic aorta improved hemocompatibility, because a biomembrane-like structure is constructed by adsorption of phospholipids originating from the living body on the surface of the ve~se1.l~ Since the MPC moiety possesses the same polar groups of phospholipid molecules and poly(MPC-coBMA), which consist of amphiphilic molecules as well as phospholipid molecules, it is considered that the poly(MPC-co-BMA) has an affinity for phospholipids. Moreover, the arrangement of phospholipids adsorbed on the surface of poly(MPC-co-BMA)was strongly indicated by XPS analysis when the surface was treated with a liposome solution of phosph01ipid.l~ From these considerations, phospholipid molecules appear to be adsorbed immediately on the surface of poly(MPC-co-BMA) from PRP. A stable adsorbed layer with a biomembrane-like structure is formed by contact with PRP. The interaction between the adsorbed phospholipid layer and platelets is gentle. We believe that this is a primary factor for suppressing platelet adhesion and activation on the surfaces of poly(MPC-co-BMA).
References 1. J. A. Hayward and D. Chapman, "Biomembrane surfaces as models for polymer design: the potential for haemocompatibility," Biomuteriuls, 5, 135-142 (1984). 2. J. W. Boretos and W. S. Pierce, "Segmented polyurethane, a polyether polymer -an initial evaluation for biomedical applications," J. Biomed. Mater. Res., 2, 121 (1968). 3. D. J. Lyman, K. Kuntson, 8 . Mcneill, and K. Shibatani, "The effects of chemical structure and surface properties of synthetic polymers on the coagulation of blood IV. The relation between polymer morphology and protein adsorption," Trans. Amer. SOC.Artif. lnt. Organs, 11, 49-54 (1975). 4. T. Okano, S. Nishiyama, I. Shinohara, T. Akaike, Y. Sakurai, K. Kataoka, and T. Tsuruta, "Effect of hydrophilic and hydrophobic microdomains on mode of interaction between block polymer and blood platelets," J. Biomed. Muter. Res., 15, 393-402 (1981). 5. Y. Mori, S. Nagaoka, H. Takiuchi, T. Kikuchi, N. Noguchi, H. Tanzawa, and Y. Noishiki, "A new antithrombogenic material with long polyethyleneoxide chains," Trans. Amer. SOC. Artif. lnt. Organs, 28, 459-463 (1982). 6. Y. Kadoma, N. Nakabayashi, E. Masuhara, and J. Yamauchi, "Synthesis and hemolysis test of the polymer containing phosphorylcholine groups," Koubunski Ronbunshu, 35, 423-427 (1978). 7. S. Fukushima, Y. Kadoma, and N. Nakabayashi, "Interaction between the polymer containing phosphorylcholine group and cells," Koubunshi Ronbunshu, 40, 785-793 (1983). 8. K. Kataoka, T. Akaike, Y. Sakurai, and T. Tsuruta, "Effect of charge and molecular structure of polyion complexes of the morphology of adherent blood platelets," Mukromol. Ckem., 179, 1121-1124 (1978). 9. K. Ishihara, T. Ueda, and N. Nakabayashi, "Preparation of phospholipid polymers and their properties as polymer hydrogel membranes," Polym. J., 22, 355-360 (1990). 10. K. Ishihara, N. Muramoto, and I. Shinohara, "Controlled release of organic substances using polymer membrane with responsive function for amino compounds," 1. Appl. Polym. Sci., 29, 211-217 (1984).
NONTHROMBOGENICITY OF PHOSPHOLIPID POLYMER
1077
11. I(. Ishihara, T. Ueda, and N. Nakabayashi, "Drug release from hydrogel membrane having phospholipid structure," Koubunshi Ronbunsku, 46, 591-595 (1989). 12. B. D. Ratner, A. S. Hoffman, S. R. Hanson, L. A . Harker, and J. D. Whiffen, "Blood-compatibility- water-content relationships for radiation-grafted hydrogels," J. Polyrn. Sci. Polyrn. Syrnp., 66, 363-375 (1979). 13. Y. Noishiki, "Biochemical response to dacron vascular prothesis," J. Biorned. Mater. Xes., 10, 759-767 (1976). 14. K. Ishihara, R. Aragaki, J. Yamazaki, T. Ueda, A. Watanabe, and N. Nakabayashi, "Organized adsorption of phospholipid on the polymer surface with phospholipid polar group and its blood compatibility," J. Jpn. SOC.Biomater., to appear.
Received June 1, 1989 Accepted February 21, 1990
deformation and aggregation were suppressed with increasing MPC composition. The same tendency was noted when Ca2+-re-addedPRP came in contact with the polymers. In the case of poly(MPC-coBMA) with 0.320 mole fraction of MPC, activation of platelets and formation of fibrin were completely suppressed. Therefore, MPC moieties in the polymer play an important role in the reduction of thrombogenicity of the polymer.
INTRODUCTION
Biomembranes have many functions such as construction of cell walls, permeation of indispensable substances, rejection of toxins, and, most importantly, excellent biocompatibility. The surface of a biomembrane only mildly interacts with blood proteins and cells and does not adsorb and activate these biological molecules. A biomembrane is a hybrid constituted mainly of two chemical classes, phospholipids and proteins. There is no covalent bonding to bind each molecule; thus, the surface of a biomembrane is heterogenic and dynamic. These features prompted us to propose new concepts for making nonthrombogenic artificial materials. Segmented polyurethanes and some block copolymers having heterogenic microdomain structure have been reported.'" Moreover, a hydrophilic polymer surface covered with grafted hydrophilic chains has the potential for suppressing adsorption of biological molecules due to the mobility of the grafted polymer chain.5 We assumed that if an artificial surface similar to a biomembrane was constituted by an arrangement of phospholipids from blood on the surface, the surface would show excellent nonthrombogenicity. In arranging the *To whom correspondence should be addressed Journal of Biomedical Materials Research, Vol. 24, 1069-1077 (1990) CCC 0021-9304190/081069-09$04.00 0 1990 John Wiley & Sons, Inc.
ISHIHARA ET AL.
1070
phospholipids, a polymer having phospholipid polar groups can be expected to possess an affinity for natural phospholipid molecules. In previous articles, synthesis of a new monomer having a phosphoryl choline moiety, 2-methacryloyloxyethyl phosphorylcholine (MPC), and the biocompatibility of MPC copolymers with methyl methacrylate were These copolymers show mild interactions with cells, however, the function of the MPC was not well studied. In this article, MPC copolymers with n -butyl methacrylate were synthesized and their nonthrombogenicity evaluated by a microsphere-column method' with attention to the MPC composition on the surface of the copolymers. EXPERIMENTAL
Materials 2-Methacryloyloxyethyl phosphorylcholine (MPC) was synthesized by a procedure previously reported and purified by recrystallization with acetonitrile.' The structure is shown in Figure 1. n-Butyl methacrylate (BMA) was purified by distillation under reduced pressure and a fraction of bp 68.5"C/30 mm Hg was used. Poly(MPC-co-BMA)was prepared conventionally using 2,2'-azoisobutyronitrile (AIBN) as an initiator. Poly(MPC-co-BMA) membrane was prepared by a solvent evaporation method to determine the MPC composition in the copolymer by analysis with x-ray photoelectron spectroscopy (XI'S, Shimadzu ESCA-750) and water content of the copolymer membrane. Poly(2-hydroxyethyl methacrylate) (poly(HEMA))was prepared by a conventional radical polymerization method using AIBN in
2-propanol. " Acrylic beads were prepared by suspension copolymerization of methyl methacrylate with triethyleneglycol dimethacrylate as a crosslinker. Diameters of the beads ranged from 200 to 600 pm.
Evaluation of thrombogenicity of the polymer surf ace A microsphere-column method was used for the evaluation of thrombogenicity of the polymer surface.8The detailed procedure was as follows. Polymer coating of acrylic beads was carried out by a solvent evaporation technique using a 0.5 wt% polymer solution. The polymer-coated beads
$=CH2 F=O + OCH,CH,O~OCH,CH,N(CH,),
0-
Figure 1. Structure of 2-methacryloyloxyethyl phosphorylcholine (MPC).
NONTHROMBOGENICITY OF PHOSPHOLIPID POLYMER
1071
were dried in vacuo for 24 h at room temperature. XPS analysis of the polymer coated beads was carried out to determine the MPC composition of the surface of the coated beads. The polymer-coated beads (0.52 g) were closely packed in a poly(viny1 chloride) (PVC) tubing (3.0 mm ID, 10 cm in length) equipped with a stopcock, and the packed column was primed with physiological saline for 24 h to exclude the liquid-air interface and to equilibrate the polymer surface with the physiological environment. Platelet-rich plasma (PRP) was prepared from 10 Japanese white rabbits weighing about 3.0 kg each. The carotid artery was cannulated using PVC tubing and 90 mL of fresh blood per rabbit was collected in a disposable syringe containing 10 mL of a 3.8 wt% aqueous sodium citrate solution. The citrated blood was immediately centrifuged for 15 min at 750 rpm to obtain citrated PRP. The number of platelets of PRP (Po) was adjusted to approximately 1 X lo8 cells/mL by dilution with platelet-poor plasma, which was prepared from the citrated blood by centrifugation for 15 min at 2600 rpm. Calcium ion-re-added PRP was prepared by the addition of 119 pL of 1 mol/liter CaC1, aqueous solution into the citrated PRP. These PRP solutions were continuously infused into the column packed with polymer beads from a disposable syringe at a flow rate of 0.23 mL/min with the use of an infusion pump. A definite volume (50 pL) of the effluent PRP was collected and the number of effluent platelets (P) counted with a Coulter Counter. At least three packed columns were used for evaluation of the thrombogenicity of each polymer. Scanning electron microscopic (SEMI observation After the PRP was passed through the column for 20 min, the beads situated in the middle of the column were placed in a saline solution containing 2.0 wt% glutaraldehyde to fix the adherent platelets. These beads were rinsed with a water-ethanol system, dried with a critical point drying technique, and coated with gold. The morphology of platelets adherent to the bead surfaces was observed with SEM (Comtec CSM-501). RESULTS
Table I lists the MPC composition and water content of the poly (MPC-coBMA) used. X I'S data suggested that the MPC composition on the coated surfaces of acrylic beads was slightly larger than that in the bulk copolymer. Though the water content of poly(BMA) was nearly zero, it increased with increasing MPC composition. Thus, poly(MPC-co-BMA)was a hydrogel." The water content of MB-2 and MB-3 membranes was almost equal to that of the poly(HEMA) membrane. Figure 2 shows a representative elution profile of platelets from a column packed with polymer-coated beads using citrated PRP. Platelets began to
ISHIHARA ET AL.
1072
TABLE I Characterization of Poly(MPC-co-BMA) Used in This Study MPC Mole Fraction Code
In Feed
In Bulk Polymerb
On Coated Surfaceb
Water Content"
0.056 0.077 0.116 0.268 0
0.127 0.186 0.260 0.320 0
0.328 0.367 0.399
-
-
0.381
~~
MB-1 MB-2
0.03 0.05 0.10
MB-3 MB-4
0.40
Poly(BMA) Poly(HEMA)
n.d.' 0
"The values were obtained by measuring the weight of water in polymer membrane swollen in water at 30°C and calculation of the ratio of the weight of water to that of polymer membrane swollen. ?'he values were determined by the XPS analysis. 'The value could not be determined.
elute about 1 min after PRP was injected into the column, which corresponded to the void volume of the column. The elution ratio (P/Po) became constant after 20 min. Figure 3 shows a typical elution profile of platelets from the column when Ca2'-re-added PRP was applied to the column. Figure 3 also shows the result of poly(HEMA) compared to poly(MPC-co-BMA). The elution curves were quite different from those shown in Figure 2. The P/Po on poly(BMA) decreased drastically about 12 min after introduction of the Ca2+-re-added PRP. Elution of platelets from the column did not cease at 17 min. A similar
T / rnin
Figure 2. The representative elution profile of platelets from the column packed with poly(MPC-co-BMA)-coated beads when citrated PRP was injected into the column. Po, number of platelets injected; P, number of platelets eluted from the column. (A) poly(BMA), ( 0 ) MB-1, (0) MB-2, (A) MB-3, (0) MB-4.
1073
NONTHROMBOGENICITY OF PHOSPHOLIPID POLYMER 1 .o
Q ' \
a
0.5
0 0
10 T / min
5
15
20
Figure 3. The representative elution profile of platelets from the column packed with poly(MPC-co-BMA) coated beads and poly(HEMA)-coated beads when Ca*+-re-addedPRP was injected into the column. (A) poly(BMA), (a) MB-1, (0) MB-2, (A) MB-3, (0) MB4, (a) poly(HEMA).
profile was also found in the poly(HEMA) system. Platelet aggregations in the columns were observed in the poly(MPC-co-BMA), MB-1, MB-2, and MB-3 systems, however, their occurrence decreased as the MPC composition increased. On the other hand, with MB-4, P/Po maintained at 1.0 except during the introduction period of PRP. Figure 4 shows the MPC compositional dependence of platelet adhesion on polymer beads coated with poly(MPC-co-BMA) when Ca2+-re-added 0.5 0.4 T
a! \
0.3
a I
a0
0.2
v
0 .I 0
J
0
I
I
I
0.1 0.2 0.3 MPC mole fraction
0A
Figure 4. MPC composition dependence of the adhered platelets ratio on the poly(MPC-co-BMA) coated beads when Ca2'-re-added PRP was infused into the column for 15 min. Open plot, poly(MPC-co-BMA) system; closed plot, poly(HEMA). Number of columns used: poly(BMA), n = 5; poly(HEMA), n = 7; the other copolymers, n = 3. Mean values of the experiments are shown and bars represent standard error mean.
ISHIHARA ET AL.
1074
PRP was passed through the column packed with the beads for 15 min. The adhesion ratio of platelets (Po-P)/Po on the surface of poly(HEMA) is shown. The adhesion ratio decreased with an increase in the MPC composition in poly(MPC-co-BMA). There was a significant difference (P < 0.001) between homopolymers (poly(BMA), poly(HEMA)) and MPC copolymers with high MPC compositions (MB-2, MB-3, MB-4). Thus, it was clearly indicated that MPC moieties in poly(MPC-co-BMA) play an important role in suppressing the adhesion of platelets. Figure 5 shows SEM pictures of platelets adherent to the beads coated with poly(MPC-co-BMA) after 20 min of contact with citrated PRP. On the surface of poly(BMA), numerous platelets adhered with major changes in their morphology and aggregation. The number of platelets adherent to the surface of poly(MPC-co-BMA) decreased with increasing MPC composition. The morphology of adherent platelets was better than on the poly(BMA) and it improved with MPC copolymers. Figure 6 shows SEM pictures of polymer beads coated with poly(MPC-coBMA) after 20 min of contact with CaZ'-re-added PRP. It was found that a fibrin net completely covered the surface of every bead coated with poly(BMA). The density of fibrin fibers decreased with increasing MPC composition; it was difficult to observe any fibrin net on the surface of MB-4. DISCUSSION
A microsphere-column method containing beads coated with test samples is one of the most powerful methods for evaluating the thrombogenic-
Figure 5. SEM picture of poly(MPC-co-BMA) coated beads after contacting citrated PRP for 20 min. (A) poly(BMA), (B) MB-1, (C) MB-2, (D) MB-3, (E) MB4.
1075
NONTHROMBOGENICITY OF PHOSPHOLIPID POLYMER
c )
200pm Figure 6 . SEM picture of poly(MPC-co-BMA)-coated beads a n d poly(HEMA)-coated beads after contacting Caz+-re-addedPRP for 20 min. (A) poly(BMA), (B) MB-1, (C) MB-2, (D) MB-3, (E) MB-4, (F) poly(HEMA).
ity of polymer materials.8The elution profile of platelets reflects the process of thrombus formation, i.e., decrease in the number of platelets eluted from the column indicates activation of the cells and the beginning of thrombus formation. As shown in Figure 2, when citrated PRP was infused into the column, P/Po was almost constant for 20 min in every case, however, P/Po took lower values when the MPC composition was lowered, thus indicating that adherent platelets on the bead surface could not elute. From this, we concluded that the MPC moiety in poly(MPC-co-BMA) could suppress platelet adhesion on the surface. When Caz+-re-addedPRP was injected into the column, the effect of the MPC moiety became clearer. The Ca2' ions were required for the activation of platelets accompanying thrombus formation. Therefore, the activation of platelets was not induced by elimination of Ca2+ions with sodium citrate. The poly(BMA) column was occluded completely by a fibrin net within 20 min. On the other hand, platelet suspension could flow through the poly(MPC-co-BMA) columns. These results confirmed that MPC copolymers suppress not only platelet activation but also platelet adhesion, supported by direct SEM observation of the polymer surfaces after contact with PRP, as shown in Figures 5 and 6. Poly(MPC-co-BMA) is a hydrogel that has an extremely hydrophilic surface. In general, it is considered that the hydrophilic surface gently interacts with biological molecules because of its high surface free energy. l2 However, by comparison of MB-2 and MB-3 with poly(HEMA), which is a typical hydrogel that has almost the same water content as poly(MPC-co-BMA), it is seen that relatively large numbers of platelets adhere to poly(HEMA) (see Figs. 3 and 4).
1076
ISHIHARA ET AL.
Noishiki reported that a polyester vascular prosthesis implanted in the thoracic aorta improved hemocompatibility, because a biomembrane-like structure is constructed by adsorption of phospholipids originating from the living body on the surface of the ve~se1.l~ Since the MPC moiety possesses the same polar groups of phospholipid molecules and poly(MPC-coBMA), which consist of amphiphilic molecules as well as phospholipid molecules, it is considered that the poly(MPC-co-BMA) has an affinity for phospholipids. Moreover, the arrangement of phospholipids adsorbed on the surface of poly(MPC-co-BMA)was strongly indicated by XPS analysis when the surface was treated with a liposome solution of phosph01ipid.l~ From these considerations, phospholipid molecules appear to be adsorbed immediately on the surface of poly(MPC-co-BMA) from PRP. A stable adsorbed layer with a biomembrane-like structure is formed by contact with PRP. The interaction between the adsorbed phospholipid layer and platelets is gentle. We believe that this is a primary factor for suppressing platelet adhesion and activation on the surfaces of poly(MPC-co-BMA).
References 1. J. A. Hayward and D. Chapman, "Biomembrane surfaces as models for polymer design: the potential for haemocompatibility," Biomuteriuls, 5, 135-142 (1984). 2. J. W. Boretos and W. S. Pierce, "Segmented polyurethane, a polyether polymer -an initial evaluation for biomedical applications," J. Biomed. Mater. Res., 2, 121 (1968). 3. D. J. Lyman, K. Kuntson, 8 . Mcneill, and K. Shibatani, "The effects of chemical structure and surface properties of synthetic polymers on the coagulation of blood IV. The relation between polymer morphology and protein adsorption," Trans. Amer. SOC.Artif. lnt. Organs, 11, 49-54 (1975). 4. T. Okano, S. Nishiyama, I. Shinohara, T. Akaike, Y. Sakurai, K. Kataoka, and T. Tsuruta, "Effect of hydrophilic and hydrophobic microdomains on mode of interaction between block polymer and blood platelets," J. Biomed. Muter. Res., 15, 393-402 (1981). 5. Y. Mori, S. Nagaoka, H. Takiuchi, T. Kikuchi, N. Noguchi, H. Tanzawa, and Y. Noishiki, "A new antithrombogenic material with long polyethyleneoxide chains," Trans. Amer. SOC. Artif. lnt. Organs, 28, 459-463 (1982). 6. Y. Kadoma, N. Nakabayashi, E. Masuhara, and J. Yamauchi, "Synthesis and hemolysis test of the polymer containing phosphorylcholine groups," Koubunski Ronbunshu, 35, 423-427 (1978). 7. S. Fukushima, Y. Kadoma, and N. Nakabayashi, "Interaction between the polymer containing phosphorylcholine group and cells," Koubunshi Ronbunshu, 40, 785-793 (1983). 8. K. Kataoka, T. Akaike, Y. Sakurai, and T. Tsuruta, "Effect of charge and molecular structure of polyion complexes of the morphology of adherent blood platelets," Mukromol. Ckem., 179, 1121-1124 (1978). 9. K. Ishihara, T. Ueda, and N. Nakabayashi, "Preparation of phospholipid polymers and their properties as polymer hydrogel membranes," Polym. J., 22, 355-360 (1990). 10. K. Ishihara, N. Muramoto, and I. Shinohara, "Controlled release of organic substances using polymer membrane with responsive function for amino compounds," 1. Appl. Polym. Sci., 29, 211-217 (1984).
NONTHROMBOGENICITY OF PHOSPHOLIPID POLYMER
1077
11. I(. Ishihara, T. Ueda, and N. Nakabayashi, "Drug release from hydrogel membrane having phospholipid structure," Koubunshi Ronbunsku, 46, 591-595 (1989). 12. B. D. Ratner, A. S. Hoffman, S. R. Hanson, L. A . Harker, and J. D. Whiffen, "Blood-compatibility- water-content relationships for radiation-grafted hydrogels," J. Polyrn. Sci. Polyrn. Syrnp., 66, 363-375 (1979). 13. Y. Noishiki, "Biochemical response to dacron vascular prothesis," J. Biorned. Mater. Xes., 10, 759-767 (1976). 14. K. Ishihara, R. Aragaki, J. Yamazaki, T. Ueda, A. Watanabe, and N. Nakabayashi, "Organized adsorption of phospholipid on the polymer surface with phospholipid polar group and its blood compatibility," J. Jpn. SOC.Biomater., to appear.
Received June 1, 1989 Accepted February 21, 1990
