J. BIOMED. MATER. RES. VOL. 11, PP. 283-295 (1977)

Properties of Heparin-Poly (methyl Methacrylate) Copolymers. 11.

D. LABARRE and M. JOZEFOWICZ, Laboratoire de Recherche SUT les Macromolkcules, Centre Scientijique et Polytechnique, Univer- sit& Paris-Nord, 93430 Villetaneuse, France, and M. C. BOFFA, M . D. Inserm Laboratoire d'Hhmostase, Centre National de Trans-

fusion Sanguine, '75015 Paris, Franre

Summary

A heparin-poly(methy1 methacrylate) copolymer in which heparin has been covalently bound was studied. This copolymer showed no release of heparin but presented heparin-like anticoagulant activity when suspended in plasma. After contact with plasma, the copolymer lost part of its antithrombin activity which could be restored by a high ionic-strength medium. Because of its solid form, this copolymer offers new possibilities for producing interesting anticoagu- lant surfaces.

INTRODUCTION

Materials containing heparin have already been prepared in several laboratories by different techniques. When heparin was ionically bound's2 to the material, it was removed by blood or plasma contact. In vicw of durable biomedical application, it was more satisfactory to prepare material with covalently bound heparin. This has been done by isocyanate derivatives of poly~tyrene,~ by cyanogen bromide on agarose,* carbodiimide on sepharose, and glutaraldehyde on nitrocellulose membrane. In a previous paper, we described an original preparation of an heparin-poly(methy1 methacrylate) copolymer (HEP-PMMA) u-it,h a high heparin con- tent.' In the present work, we have investigated in vitro the anti- coagulant properties of the copolymer for possible applications as solid hemocompatible material.

283

@ 1977 by John Wiley &i Sons, Inc.

284 LABARRE, JOZEFOWICZ, AND BOFFA

MATERIALS AND METHODS

Copolymer The crude HEP-PMMA copolymer was prepared by a radical

polymerization of methyl methacrylate initiated by a heparin radical resulting from reaction between cerium IV ions and heparin in nitric aqueous solution, as has already been described.? Three different powder batches have been used for the experiments de- scribed here with similar results. The sodium heparinate supplied by Choay Laboratories (Paris) possessed an activity of 180 units/mg. In order to obtain homogeneous suspensions, the crude copolymer was centrifugated without drying before treatments.

The crude copolymer was washcd several times (20°C, 24 hr) either with 0.15 M sodium chloride or with 0.25 M sodium citrate (1 g of copolymerj250 ml of solution). Each washing was followed by a centrifugation ( 10"C, 15,000 rpm, 20 min).

Each copolymer sample was washed once more before clotting assays, a t which point 5 mg of copolymer were added to 10 ml of 0.013 M sodium citrate in buffer. After stirring a t 37°C and cen- trifugation, the supernatant was controlled by thrombin time measurement.

Reagents Used for Clotting Assays The following buffer was used for dilutions and for control time

measurement: 0.026 M sodium barbital, 0.026 A4 sodium acetate, 0.1 M sodium chloride; pH 7.3. To prepare the copolymer suspen- sion, the buffer was modified by addition of 1 g/l. of an emulsifying agent (Lensex TA 01-NP 40 Shell). As previously described,? the addition of this agent a t this concentration did not modify thrombin time of the controls (containing no copolymer). Human citrated platelet-poor plasma8 collected from two blood donors was stored a t -20°C. Each 2 ml sample was thawed and kept a t 4°C before use. Heparin (Vitrum) was used as reference: 1 international unit (I.U.) = 10 fig; 1% protamine sulfate (Vitrum), thrombin (ISH), and reptilase (Stago) were diluted extemporaneously in buffer and kept a t 4°C.

Clotting Assays The coagulation time of a normal plasma initiated by thrombin

(i.e., thrombin time) was mcasurtd a t 37°C in glass tubes by visual

PROPERTIES O F HEP-PMMA COPOLYMERS. I1 28.5

observation; scattering of results was lower than 10% for thrombin times shorter than 1 min and increased with longer thrombin time. One volume of starting copolymer suspension was added to two volumes of plasma a t 37°C and stirred over a period of time. For direct thrombin time measurement, onc volume of thrombin (10 units/ml) was added to the mixture. To study the properties of the plasma-contacted copolymer or that of the copolymer-contacted plasma, a centrifugation (6000 rpm, 4OC) was performed. The supernatant and the pellet were collected.

The pellets were suspended in buffer in order to obtain the same copolymer content as in the starting suspension. As previously, one volume of this suspension was added to two volumes of plasma before addition of one volume of thrombin. The presence of heparin in the supernatants was tested by suppression of thrombin time delays by addition of protamine sulphate or by using reptilase (a heparin-nonsensitive enzyme) instead of thrombin.

Heparin Acidimetric Titration and Elementary Analysis

Heparin acidimetric titrations were performed as previously de~cr ibed .~ Nitrogen, sulfur, and sodium contents of copolymer samples were measured by elementary analysis. The more repro- ducible results were obtained from sodium content determinations. Calculations were done by taking into account 6 sodium atoms per tetrasaccharide unit of heparin; 1200 was t,aken as molecular weight of one unit.

RESULTS

Effects of Sodium Chloride and Sodium Citrate Treatments of the Crude Copolymer

Two different treatments of the crude copolymer after isolation from the reaction medium were tested as described in Table I. It was noticed that there was no significant difference of the heparin content between the citrate- or chloride-treated copolymer samples, when measured by elementary analysis and acidimetric titration (Table 11). Nevertheless, a small part of heparin was released in the plasma from the chloride-treated copolymer but not from the citrate-treated material.

286 LABARRE, JOZEFOWICZ, AND BOFFA

TABLE I Treatments of the Crude Copolymera

Operations First Procedure Second Procedure

Step 1: 5 washings fol- lowed by centrifuga- tions

Step 8: n washings fol- lowed by centrifuga- tions

Step 3: 1 washing fol- lowed by filtration on Millipore (Cellu- lose 3 p)

With 0.15 M NaCl

With 0.15 M NaCl

I - Supernatants (Sl) - Pellets

With distilled water

1 - Pellets (Pl)

With 0.25 A4 sodium citrate

With 0.15 A4 NaCl

I - Supernatants (S2) - Pellets

With distilled water I - Pellets (P2)

*Copolymer samples (Pl, P2) were dried under vacuum on Pz05. Super- natants (S1, S2) were tested by thrombin time measurements: 0.1 ml of super- natant was incubated with 0.2 ml of plasma (10 min, 37°C). After the addition of 0.1 ml of thrombin, clotting time was measured.

TABLE I1 Analysis of Copolymer Sample HP-10

Heparin Thrombin Elementary Analysis Acidimetric Time of

Copolymer Titration Plasma Super- Sample Na k%) Heparin (g%) (g%) natant*(sec)

P1 0.91 f 0.1 9.0 f 1 10.6 f 1 100

P2 0.87 f 0.1 8.5 f 1 10.1 f 1 25

* Thrombin time: 0.1 ml of HP-10 copolymer suspension (6 mg/ml of buffer) or buffer was incubat,ed for 2 min a t 37°C with 0.2 ml of plasma in a glass tube. After centrifugation (6000 rpm, 4"C), 0.2 ml of supernatant was added to 0.1 ml of buffer and kept for 2 min a t 37°C. After addition of 0.1 ml of thrombin, the clotting time was measured and compared to that of controls: 21 sec.

PROPERTIES OF HEP-PMMA COPOLYMERS. I1 287

0 0) m

300

a E .- 4-

c .- n E 0 2 0 0

.c I-

L

100

0

The thrombin times of the washing supernatants (S1 and 52) measured after each chloride washing in step 2 decreased rapidly (Fig. 1). However, if the chloride-treated copolymer (Pl) was submitted to the second procedure (Table I), the thrombin times measured on supernatants were highly delayed again and shortened progressively with the number of washings as in Figure 1. Free

-

-

-

-

. _ - _ I I

5 1 0

Fig. 1. Thrombin times of successive S1 supernatants. Clotting system: see Table I. Thrombin time of the control: 20 sec.

288 LABARRE, JOZEFOWICZ, A N D BOFFA

heparin was identified as the responsible agent of the observed throm- bin time delay of the supernatants.

These results indicated that : 1) sodium citrate solutions were able to rid the crude copolymer of noncovalently bound heparin; 2 ) sodium chloride solutions were insufficient to obtain the same result; and 3) purified copolymer samples retained a high content of covalently bound heparin.

Anticoagulant Properties of the HEP-PMMA Copolymer As previously described,’ the copolymer had an anticoagulant

effect on plasma. This was related to a hrparin-like activity of the citrate-treated copolymer sample (PZ), 1% hich was responsible for the antithrombin effect measured in the copolymer plasma sus- pension. Indeed, the antithrombin effect was neutralized by prot- amine sulfate (Fig. 2 ) . No delay of the clotting time was observed in the same plasma when using reptilase instead of thrombin (Fig. 3). Consequently, the action on plasma of a typical citrate-washed copolymer (HP-10) was examined by measurements of thrombin times on copolymer suspensions in plasma, on supernatant plasma, or on pellet suspensions in plasma. (Supernatant and pellet were obtained by centrifugation of copolymer suspensions in plasma.)

The influence of different parametcrs on the anticoagulant action of the copolymer has been studied.

Increase of the Copolymer Content. The ratio of copolymer weight in final volume of reactives in plasma suspension markedly delayed thrombin times of the suspension and slightly delayed that of the supernatant obtained by copolymer rrmoval (Fig. 3). Two sets of results were obtained with incubation times of 2 and 20 min, respec- tively, and did not show any significant difference.

Increase of Concentratzon of Added Thrombin . At fixed copolymer content and incubation timc (Fig. 4), increasing thc thrombin concentration caused thrombin times of copolymer suspensions in plasma to decrease sharply.

The influence of this parameter on plasma with copo1ymc.r was very marked when incubation times exceeded 20 min and copolymer contents in the suspensions wTere higher than 0.5 mg/ml (Fig. 5 ) .

Bzological Heparzn-Like Act iv i ty of the Copolymer. This was measured by thrombin time and was calculated referring to Vitrum

Inf luence of the Incubatzon T i m e .

PROPERTIES OF HEP-PMMA COPOLYMERS. I1 289

- - - - - - S o l u b l e h e p a r i n 0.1 u x 0.2 u +

Copolymer s u s p e n s i o n W 5 m g 0 0.5 mg 0

I I , t

I , I

---I+- - - - -

I I

0.5 1.0 P r o t a m i n e s u l f a t e u n i t s

Fig. 2. Neutralization of antithrombin effect by protamine sulfate. Clotting system: 0.1 ml of (-) HP-10 copolymer suspension or (- - -) soluble heparin or buffer was incubated with 0.1 ml of protamine sulfate solution and 0.2 ml of plasma for 5 min a t 37°C. After addition of 0.1 ml of thrombin, clotting time was measured. Clotting time of the control: 22 sec.

290 LABARRE, JOZEFOWICZ, AND BOFFA

Copolymer con t e n t mg/m I

Fig. 3. Influence of copolymer content on thrombin time. Clotting system: 0.1 ml of buffer or of (-) HP-10 copolymer suspension (of increasing content) in buffer was incubated with 0.2 ml of plasma a t 37°C for (X ) 2, 5, or (f) 20 min in a glass tube; 0.1 ml of thrombin was added to 0.2 ml of the incubating mixture [or of its (- - -) supernatant obtained by centrifugation diluted with 0.1 ml of buffer, and incubated for (0) 5 or ( A ) 20 min]. Clotting time was then measured. Thrombin time of the control: 21 sec. Clotting time with (8) reptilase was determined with 2 suspensions of copolymer in plasma. Clot- ting time of the control: 18 sec.

u a v)

3 00

E

i

.- 4-

C n200

x +

100

0

PROPERTIES OF HEP-PMMA COPOLYMERS. I1 29 1

- C o p 0 l y m e r s u s p e n s i o n c o n t e n t 1.25mg/ml o

2.5 0 -.--- C o n t r o l s +

I 1 2 3 4 5

T h r o m b i n a c t i v i t y I - u n i t s

Fig. 4. Variations of thrombin times with the activity of thrombin in the suspension. Clotting system: 0.1 ml of buffer or of HP-10 copolymer suspension in buffer was incubated with 0.2 ml of plasma (2 min, 37°C); 0.1 ml of thrombin of increasing activity was added and the clotting time measured.

292 LABARRE, JOZEFOWICZ, AND BOFFA

300

0 Q 01

100

0

/

us pensi on

0 X +

1.3 A 0.8 a

-.-- - Su pe r natant

lncu bat ion t iine mi n

Fig. 5 . Influence of incubation duration on thrombin time. Clotting system is the same as in Fig. 3 except that thrombin times were measured after different incubation times of the copolymer with the plasma.

heparin solutions (Table 111). The heparin content of the copolymer was 10% of the total weight. Consequently, the heparin-like activity of the copolymer was about 1% of the activity of the total heparin present in the copolymer, if soluble.

PROPERTIES OF HEP-PMMA COPOLYMERS. I1 293

TABLE I11 Percentage of Heparin Activity of Copolymer Sample HP-108

Amount of Amount of Weight of Heparin Thrombin Active Heparin

Copolymer Present Time Heparin Activity ( M d (unit) bet) (unit) (75)

33 0.6 28 0.004 0 .7 50 0.9 30 0.005 0.6 66 1.2 35 0.006 0.5 83 1.5 45 0.010 0 .7

100 1.8 60 0.014 0.8 133 2.4 190 0.045 1.9

a Clotting system: 0.1 ml of HP-10 copolymer suspension (or soluble Vitrum heparin) in buffer was incubated for 2 min a t 37°C with 0.2 ml of plasma in a glass tube; to 0.2 ml of this suspension (or solution) and 0.1 ml of buffer, 0.1 ml of thrombin was added and the clotting time recorded. The amount of hep- arin present was calculated from the weight of copolymer in the suspension. (Copolymer HP-10 contains 10% heparin by wt.) The amount of active heparin was obtained from the thrombin time measurements of heparin solutions at increasing concentration from 0.01 to 0.2 units/ml. The percentage of heparin activity was equal to:

amount of active heparin loo amount of present heparin

Loss and Regeneration of the Anticoagulant Properties By comparison between copolymer suspensions incubated once

(a) or twice ( b ) with plasma, a decrease of the heparin-like activity of the b suspension was observed (Table IV).

The antithrombin activity could be regenerated, by washing the copolymer (pellet a) with high ionic-strength solutions. Indeed, thrombin times of copolymer suspensions in plasma ( e ) were found similar to these of copolymer suspensions a (Table IV).

DISCUSSION AND CONCLUSION

Washings of the crude copolymer have shown that heparin is linked to the polymer by different kinds of bindings. Sodium chlo- ride solutions are able to release the adsorbed heparin, while complex- ing agents such as citrate are able to remove a part of the antithrombin

294 LABARRE, JOZEFOWICZ, AND BOFFA

TABLE IV Loss and Regeneration of the Antithrombin Properties of Copolymer Sample

HP-10

Copolymer Suspension Thrombin Time (sec)

as 150 f 20 bb 57 f 10 CC 71 f 10 d d 84 f 15 ed 135 f 20

a Copolymer suspended in plasma + suspension a. b Suspension a centrifugated + pellet a; pellet a suspended in plasma -+ sus-

pension b. c Pellet a washed three times with 0.5 ml of 0.15 M NaCI, centrifugated, and

suspended in plasma + suspension c. d Pellet a washed twice with 0.5 ml of 0.4 M (or 1.5 M ) NaCl, once with 0.1.5 M

NaC1, centrifugated and suspended in plasma: washing with 0.4 M NaCl -+

suspension d ; washing with 1.5 M NaCl + suspension e . Clotting system: 0.1 ml of copolymer suspension in buffer (1.3 mg/ml) was incubated with 0.2 ml of plasma (2 min, 37°C). After addition of 0.1 ml of thrombin (10 units/ml), clotting times were recorded. Thrombin time of the control: 19 sec.

activity from the chloride washed copolymer. It is highly probable that the citrate solution removes the heparin trapped by cerium I11 ions precipitating out of the preparation of the copolymer. This complcxing elimination permits also to eliminate the possible additive antithrombin activity of these ions.9 After successive citrate, chloridc, and water washings, heparin was no longer released from copolymer by plasma contact. These treatments are necessary to prepare the purified copolymer.

This purified copolymrr possesses a marked antithrombin (heparin- like) activity. Covalently bound heparin is responsible for this activity. As expected, the antithrombin effect on plasma varies x i th the copolymer amount in plasma and with the thrombin con- centration. The increase of thrombin times for incubation times of the copolymer with plasma longer than 20 min has no clear explana- tion. Copolymer-contacted plasma, after removal of the copolymer, does not develop such an increase of antithrombin activity. There- fore, the hypothesis of an activation of antithrombin plasma factors due to the copolymer action must bc considered.

PROPERTIES OF HEP-PMMA COPLOYMERS. I1 295

When the copolymer has been in contact with plasma, the anti- thrombin activity of the copolymer is partly masked but reappears mostly after suitable treatments. The solid covalently bound heparin retains about 1% of the activity of the total heparin if soluble. According to Wolfrom and McNeely,lO this loss of activity can be due to the nitric acid solution used in the preparation of the copolymer.

When compared to gel materials in which heparin is covalently bound either to agarose or ~ e p h a r o s e , ~ , ~ HEP-PMMA copolymer is an interesting material for preparative and analytical investigations of the proteins involved in blood coagulation. The advantage of its solid form suggests a possible application for constituting anticoagu- lant surfaces.

References 1 . V. L. Gott, J. D. Whiffen, and S. M. Valiathan, Ann. New Fork Acad. Sci.,

146, 21 (1968). 2. R. D. Falb, M. T. Takahashi, G. A. Grode, and R. I. Leininger, J. Biomed.

Mater. Res., 1, 239 (1967). 3. B. D. Halpern and R. Shibakawa, Advan. Chem. Ser., 87, 197 (1968). 4. P. H. Iverius, Biochem. J . , 124, 677 (1971). 5. I. Danishefsky and F. Tzeng, Thromb. Res., 4, 237 (1974). 6. W. J. Rea, J. W. Eberle, J. T. Watson, R. K. Ecker, and W. L. Sugg, Surg.

Forum, 22, 188 (1971). 7. D. Labarre, M. C. Boffa, and M. Jozefowicz, in Transformations of Functional

Groups on Polymers ( J . Polym. Sci. Polym. Symp. Ed., 47), C. G. Overberger and B. SedljEek, Eds., Interscience, New York, 1974, p. 131.

8. M. C. Boffa, P. Delori, and J. P. Soulier, Thromb. Czath. Haemorrh., 28, 509 (1972).

9. A. Ducastaing, C. Monyceron, J. L. Azanxa, P. Creac'h, and J. Raymond, C. R. SOC. Biol. Bordeaux, 167, 262 (1973).

10. M. L. Wolfrom and W. H. McNeely, J . Aner. Chem. Soc., 67, 748 (1945).

Received March 12, 1976 Revised May 17, 1976