Login| Sign Up| Help| Contact|

Patent Searching and Data


Title:
HEPARIN FRAGMENT SHOWING COMPLEMENT INHIBITION ACTIVITY
Document Type and Number:
WIPO Patent Application WO/1992/002232
Kind Code:
A1
Abstract:
This invention relates to the use of heparin fragment showing complement inhibition activity and reduced bleeding time when compared to heparin regardless of anticoagulant activity for the manufacturing of medicaments for treating pathological inflammations as well as the method for treatment of pathological inflammations by administration of the heparin fragment. The invention also relates to a method for finding heparin fragments with high therapeutical value by choosing those with short bleeding time and high complement inhibition activity regardless of anticoagulant activity. Another aspect of the invention is the heparin fragment showing complement inhibition activity and reduced bleeding time when compared to heparin regardless of anticoagulant activity and pharmaceutical compositions containing them. The invention also relates to the process for the preparation of heparin fragments.

Inventors:
EKRE HANS-PETER T (SE)
HYDEN PIA (SE)
MATTSSON CHRISTER (SE)
PALM MAUD (SE)
SVAHN CARL-MAGNUS (SE)
WEBER MICHAEL (SE)
Application Number:
PCT/SE1991/000502
Publication Date:
February 20, 1992
Filing Date:
July 19, 1991
Export Citation:
Click for automatic bibliography generation   Help
Assignee:
KABI PHARMACIA AB (SE)
International Classes:
C08B37/00; (IPC1-7): A61K31/725
Domestic Patent References:
WO1990001501A11990-02-22
Foreign References:
US4916219A1990-04-10
US4847338A1989-07-11
Other References:
CHEMICAL ABSTRACTS, Volume 103, No. 1, 8 July 1985, (Columbus, Ohio, USA), SHARATH, MURALI D. et al.: "Small heparin fragments regulate the amplificationpathway of complement", see abstract 4756u; & IMMUNOPHARMACOLOGY 1985, 9(2), 73-80.
CHEMICAL ABSTRACTS, Volume 102, No. 15, 15 April 1985, (Columbus, Ohio, USA), FAREED, JAWED et al.: "Studies on the antithrombotic effects and pharmacokinetics of heparin fractions and fragments", see abstracts 125322n; & SEMIN. THROMB. HEMOSTASIS 1985, 11(1), 56-74.
Download PDF:
Claims:
CLAIMS
1. Use of heparin fragment showing complement inhibition activity and reduced bleeding time when compared to heparin regardless of anticoagulant activity for the manufacturing of medicaments treating pathological inflammations.
2. Use of heparinfragment according to claim 1 in which the bleeding time has been reduced by a factor of at least 3 when compared to heparin.
3. Use according to any of claims 12, characterised by that the heparinfragment have a molecular weight of about 7000 kD or less and have been modified.
4. Use according to claim 3 in which the heparin fragment has a molecular wight of about 7000 kD or less.
5. Use according to claim 4, characterised by that the heparinfragment have been modified by Ndesulphation followed by complete Nacetylation, by oversulphation after Nacetylation or by carboxyl group reduction.
6. Method for finding hepariti fragments with high therapeutical value, characterized by choosing those with short bleeding time and high complement inhibition activity regardless of anticoagulant activity.
7. Method according to claim 6, characterized in that the heparinfragments choosen have a molecular weight of about 7000 kD or less.
8. Method according to claim 7, characterized in that the heparinfragments choosen have been modified.
9. Method according to claim 7 or 8, characterized in that the heparinfragment has a low affinity for antithombin III.
10. Heparinfragment showing complement inhibition activity and reduced bleeding time when compared to heparin regardless of anticoagulant activity.
11. Heparinfragment found by the method according to any of claims 69.
12. Heparinfragment according to claims 10 or 11 , characterised by that they have been modified by Ndesulphation followed by complete 'acetylation, by oversulphation after Nacetylation or by carboxyl group reduction.
13. Pharmaceutical compositions, characterised by that they contain a heparin fragment according to any of claims 911.
14. A process for the preparation of heparin fragments with low affinity for antitrombin III according to any of claims 911 , characterised in modification of heparin by a) nitrous acid depolymerization or periodate oxidation followed by alkaline αepolymerisation b) reduction by excess sodium borohydride and c) separation of the low affinity form by means of affinity chromatography on matrixbound antitrombin.
15. Method for treatment of pathological inflammations, characterized in that a heaparin fragment showing complement inhibition activity and reduced bleeding time when compared to heparin regardless of anticoagulant activity is subcutaneous administrated in a therapeutical effective dose.
16. Method according to claim 15, characterized in a subcutan administration in higher doses than given for heparin, especially higher than 90 mg/person and day.
17. Method for treatment of pathological inflammations, characterized in that a heaprin fragment showing complement inhibition activity and reduced bleeding time when compared to heparin regardless of anticoagulant activity is infused in a therapeutical effective dose.
18. Method according to claim 17, characterized in an infusion in higher doses than given for heparin, especially higher than 3 mg/kg and day.
Description:
HEPARIN FRAGMENT SHOWING COMPLEMENT INHIBITION ACTIVITY

DESCRIPTION

Field of the invention,

This invention relates to the finding that reduced bleeding is an important factor when choosing among heparin fragments with complement inhibition activity. These heparin fragments can be used for treating pathological inflammations,

BacKground

Heparin is sulphate-containing polysaccaride which on a large scale is isolated from intestinal mucus from swine or lung from cattle. It has for several decades been used clinically as an effective αrjg for the prevention and treatment of thromboembolic disorders but it may sometimes cause bleeding complications (Mant M Jet al. Haemorrhagic complications of heparin therapy. Lancet 1977; 1133- 1 135).

A number of factors may contribute to the mechanism of heparin- induced bleedings. The most obvious mechanism is the inhibition of certain activated coagulation factors, e.g. thrombin, and thrombin- induced platelet aggregation. Considerable efforts have therefore been made during the last years to produce heparin-derived compounds with reduced thrombin inhibition but with retained antithrombotic activity. These criteria are fulfilled by certai n low molecular weight heparin (LMWH) fragments (See e.g. Holmer . et al. Anticoagulant and antithrombotic effects of heparin and low molecular weight heparin fragments in rabbits. Thromb Res 1982; 25: 475-485 and EP 14 184).

However, bleeding complications similar to those with heparin have been reported in some cases after clinical administration of LMWH fragments at high doses. (Schmit∑-Huebner U et al, Clinical thrombosis prophylaxis. Klin Wochenschr 1984; 349-353.)

US 4847338, EP 287 477 and WO 88/01280 discloses all different heparin fragments defined by their low anticoagulant activity and having different therapeutic effect.

It has earlier been suggested (Heparin by David A Lane and Ulf Lindahl,1989, page 579) that no simple relationship is found between blood loss and the effects of the preparations on antifactor Xa levels or the APTT measured ex vivo. The relation between the anticoagulant effect and bleeding time has until now been unresolved.

We have now found that a short bleeding time is not necessaraly associated with a low anticoagulant activity and that for finding a heparinfragment with good therapeutical value, e.g. that can be used for treating pathological inflammations without risk for haemorragic complications, the bleeding time is the most important factor and that the anticoagulant effect is of minor importance.

Description of the invention.

This invention relates to the use of heparin fragment showing complement inhibition activity and reduced bleeding time when compared to heparin regardless of anticoagulant activity for the manufacturing of medicaments for treating pathological inflammations. Preferably the bleeding time has been reduced by a factor of at least 3 when compared to heparin. The heparinfragments have normally a molecular weight of about 7000 kD or less and could be modified heparinfragment. They could have a low affinity for antithombin.

The invention also relates to the use of heparinfragment which have -been modified by N-desulphation followed by complete N- acetylation, by oversulphation after N-acetylation or by carboxyl group reduction.

The invention also relates to a method for finding heparin fragments with high therapeutical value, characterized by choosing those with short bleeding time and high complement inhibition activity regardless of anticoagulant activity.

Another aspect of the invention is the heparinfragment showing complement inhibition activity and reduced bleeding time when

compared to heparin regardless of anticoagulant activity and pharmaceutical compositions containing them. The invention also relates to the process for the preparation of heparin fragments. This process includes a) nitrous acid depolymerization or periodate oxidation followed by alkaline depolymerisation of heparin b) reduction by excess sodium borohydride and c) separation of the low affinity form by means of affinity chromatography on matrix- bound antitrombin.

Also the method for treatment of pathological inflammations by administration of the heparin fragment showing complement inhibition activity and reduced bleeding time when compared to heparin regardless of anticoagulant activity is included in the invention. The fragment can be subcutaneous administrated in a therapeutic effective dose, especially higher dose than given for heparin e.g.in higher doses than 90 mg/person and day or infused in a therpeutical effective dose, especially a higher dose than given for heparin e.g.in higher doses than 3 mg/kg and day.

Materials and general preparation methods. Sodium heparin from porcine intestinal mucosa (PM-heparins) (Mr 14 000, Anti-FXa 160 lU/mg, APTT 160 lU/mg sulphur content 10.7%) and from bovine lung (BL-heparin) (Mr 10 000 Anti-FXa 131 lU/mg, APTT 131 lU/mg sulphur content 12.9%) were obtained from Hepar Industries, Franklin, Ohio. Fragments from these heparins were prepared as sodium salts by nitrous acid depolymerization essentially as described by Thunberg et al. (The molecular size of the antithrombin-binding sequence in heparin. Febs Letters 1980; 117: 203-206.), followed by reduction by excess sodium borohydride according to Horton and Philips (The nitrous acid deamination of glycosides and acetates of 2-amino-2-deoxy-D-glucose. Carbohydr Res 1973; 30: 367 374) to convert anhydromannose groups to more stable anhydromannitol groups. A low molecular weight PM heparin fragment (Kabi 2165) was separated into their high (HA) and low (LA) affinity forms by means of affinity chromatography on matrix-

bound antithrombin, as described by Andersson et al (Anticoagulant properties of heparin fractionated by affinity chromatography on matrix bound antithrombin III and by gel filtration. Thromb Res 1976; 9: 575-583).

A totally N-acetylated heparin fragment was obtained by N- desulphation of the pyridinium salt of the heparin fragment dissolved in DMSO and methanol according to Inoue et al. (Selective N-desulfation of heparin with dimethyl sulfoxide containing water or methanol. Carbohydr Res 1976; 46: 87-95) and then acetylated by the method of Danishefsky et al. (Investigations on the chemistry of heparin. I. Desulfation and acetylation. Arch Biochem Biophys 1960; 90: 114-121 ).

The totally N-acetylated heparin fragment was oversulphated basically as described by Ogamo et al. (Binding of heparin fractions and other polysulfated polysaccharides to plasma fibronectin: effects of molecular size and degree of sulfation of polysaccharides. Biochim Biophys Acta 1985; 841 : 30-41) and yielded a totally N-acetylated oversulphated heparin fragment. An O-desulphated-N-sulphated heparin fragment was obtained by O- desulphating a totally N-acetylated heparin fragment according to Nagasawa et al. (Solvolytic desulfation of glycosaminoglycuronan sulfates with dimethyl sulfoxide containing water or methanol. Carbohydr Res 1977; 58: 47-55). The O-desulphated fragment was then N-deacetylated by the method of Dimitriev et al. (Selective cleavage of glycosidic linkages: studies with the O-specific polysaccharide from Shigella dysenteria type 3. Carbohydr Res 1975; 40: 365-372) and subsequently N-sulphated according to Lloyd et al. (Studies on heparin degradation - 1. Preparation of (35S) sulfamate derivatives for studies on heparin degrading enzymes of mammalian origin. Biochem Pharmacol 1971 ; 20: 637-648.). A BL-heparin fragment was reduced twice according to Taylor et al. (Stoichiometric Reduction of Uronic Acid Carboxylic Groups in Polysaccharides in Methods. Carbohydrate Chem VII 1976; 149-151. Ed. Whistler R L, and Be Miller J N. Academic Press New York, San Francisco, London.). The extent of reduction was about 70% after one

reduction and better than 90% after two consecutive reductions when analysed by IR spectroscopy and conductimetric titrations.

EXAMPLES

Bovine heparin (BM) and Porcine heoarin fPM. Sodium heparin from bovine lung was from Hepar Industries Inc., Franklin, Ohio, U.S.A.and Sodium heparin from porcine mucosa was from Hepar Industries

KABI 2165. Fragmin®

This is a fragment of PM-heparin which has been obtained by nitrous acid depolymerization and is manufactured by KABI under the name

Fragmin®.

LA-Kabi 2165 and HA-Kabi 2165

Antitrombin, purified from human blood plasma by Kabi Biopharma, was coupled to CNBr-activated Sepharose 4B (Pharmacia) at a density of ca 5 mg antitrombin/ml gel. 500 ml of this gel packed in a 5 cm x 30 cm column was equilibrated in 0,05 M Tris buffer, pH 7,4, with 0,15 M NaCI. For preparation of LA-Kabi 2165, ca 300 mg of Kabi 2165, dissolved in equilibration buffer, was applied to the column and the unabsorbed Kabi 2165 collected. Retained Kabi 2165 was desorbed by increasing the NaCI concentration to 1 ,5 M. For preparation of HA-Kabi 2165, a Kabi 2165 load of 600-900 mg was applied to the column and the desorbed fraction collected.

Kabi 2226

This is another fragment of PM-heparin which has been obtained by nitrous acid depolymerisation and which has a lower molecule wieght than Kabi 2165, Fragmin®. (Sulphur by elemental analysis: 9.42 %; Nitrogen by elemental analysis: 1.9 %)

KABI 2236

Sodium bovine lung heparin (40g) was dissolved in water (325 ml).

The pH of the solution was adjusted to 1 ,5 with dilute sulfuric acid.

Sodium nitrite (0,75 g) in water (10ml) was added to the acidic heparin solution. After 3 1/2 h, the solution was neutralized with dilute sodium hydroxide solution, and sodium borohydride (0,47 g) dissolved in water (4 ml) was added. After reduction of the anhydromannose residues for 18 h, excess sodium borohydride was destroyed by acetic acid. To this solution ethanol (5 L) was added. The precipitate was collected by filtration and dried. The product was dissolved in water, desalted by ultrafiltration (In an Amicon cell 8400) and freeze-dried. Yield 28 g. (Sulphur by elemental analysis: 12.1 %; N-acetyl groups of total amino groups by H-NMR: <5

%)

Kabi 2236, N-acetylated

A heparin fragment which was obtained as described for Kabi 2236 (6,0 g), was dissolved in water (75 ml) and passed through a cation exchange resin (H+-form, 2,5x26 cm). The eluate (315 ml) was neutralized with pyridine (8,0 ml) to pH 6, and the solution was freeze-dried to give the pyridinium salt (6,45 g). The pyridinium salt (8,0 g) was dissolved in DMSO (380 ml) and heated to 50-52°C and MeOH (20 ml) was added. After 2 h at this temperature the solution was cooled and poured into a 10 % solution of NaOAc in MeOH (1200 ml). The mixture was stirred, centrifugated and the precipitate added to EtOH (650 ml) and water (50 ml). The precipitate was centrifugated and washed with EtOH (2 x 100 ml) and dried in vacuum to give the crude product 208 (7,08 g). This crude product (6,70 g) was dissolved in water (168 ml) and MeOH (17 ml). The solution was cooled to +5°C and the pH was adjusted to 7 with acetic acid (0,2 ml). Acetic anhydride (8.4 ml) and cold 1 M NaOH (125 ml)*were added dropwise over 25 min to keep the pH at 6,5-7. After 2 h in the cold the solution was warmed to 24°C and made alkaline with 1 M NaOH (35 ml). After 35 min the solution was neutralized with acetic acid, desalted by ultrafiltration and freeze-dried. 1 H-NMR spectra showed the presence of O-acetyl groups (2,2 ppm). To remove these groups the product was dissolved in 0,1 M NaOH (120 ml) kept for 40 min at room temperature, neutralized, desalted and freeze-dried to give

the product. (4,8 g). (Sulphur by ele.. antal analysis: 8.45 %; N-acetyl groups of total amino groups by H-NMR: 100 %)

Heparin fragment, partially N-desulphated. N-acetylated By changing the heating of the pyridine salt, described for N- acetylated, to 35°C and 1 hour instead of 50-52°C for 2 h, a partially N-desulphated, N-acetylated heparin fragment was obtained in which the sulphur by elemental analysis was 9.2 % and SO3/COO was 1.58 )

Heparin fragment. N-acetylated. O-oversulphated. N-acetylated heparin fragment (2.14g), was dissolved in water (50 ml) and passed through a cation exchange resin (2,5 x 26 cm. H+- form). The eluate (200 ml) was neutralized with n-Bu3N (5,3 ml) to pH 6,4, freeze-dried and dried ove P2O5 yielding the butylammonium salt (3,50 g). This salt (1 ,47 g) was dissolved in dry DMF (16ml) and Sθ3-pyridine complex (2,87 g) was added.

After 4 h at room temperature the solution was added to 10 % NaOAc in MeOH (75 ml). The precipitate was centrifugated, washed with MeOH (3x80 ml), dissolved in 2 M NaCI and neutralized to pH 6,8 with 0,2 M NaOH. Ultrafiltration and freeze-drying afforded the crude product (1 ,18 g).

This product (0,73 g) was dissolved in water (35 ml) and passed through a cation exchange resin (2x14 cm, H+-form). The eluate was neutralized to pH 7 with 0,2 M NaOH and freeze-dried, yield 0,67 g. (Sulphur by elemental analysis: 13.5 %; N-acetyl groups of total amino groups by H-NMR: 100 %)

Kabi 2236. N/O-Desulphatech N-resulphated. The pyridinium salt (4,0 g) obtained from heparin fragment KABI 2236 was dissolved in DMSO (190 ml) and MeOH (10 ml) and heated to 50°C for 2 h. The solvent was distilled off in vacuum. The residue was dissolved in 2 M NaOAC (30 ml) and 2 M NaOH (4,5 ml) and poured into 95 % EtOH (350 ml) at -10°C. The precipitated was centrifuged, dissolved in 2 M NaOAc (40 ml) and added to EtOH (350 ml) in the cold. The precipitate was centrifugated and washed with EtOH (2x100 ml) and dried in vacuum at 60°C. Yield 2,97 g.

1 ,6 g of this product was dissolved in water (40 ml) and MeOH (2 ml) and cooled to +4°C. The pH was adjusted to 5-6 with acetic acid (0,1 ml). Acetic anhydride (2,0 ml) was added and then 1 M NaOH (31 ml) was added over 40 min to keep the pH at 6,5-6,8. The solution was desalted by ultrafiltration. Freeze-drying and drying in vacuum at 60°C for 3 h yielded 1 ,17 g.

2,54 g of a product, prepared in the same way, was dissolved in water (30 ml) passed through a cation exchange resin (H+-form, 2,5x30 cm). The eluate (300 ml) was neutralized by pyridine (5 ml), concentrated on a rotatory evaporator and the pH was adjusted to 6,1 with pyridine (2 ml). The solution was freeze-dried and the salt dried over P2O5 overnight. This product was dissolved in DMSO (122 ml, dried with molecular sieves) and MeOH (13 ml) and heated at 100°C over 3,3 h. The solvent was distilled off in vacuum. To the remainder was added 2 M NaOH (125 ml) and 2 M NaOH (ca 5 ml) and the solution (pH 8-9) was desalted by ultrafiltration and thereafter freeze-dried. Yield 1.31 g.

1 ,05 g of this product was mixed with 95% hydrazine (25 ml) and hydrazinium sulphate (1 ,25 g) and heated under nitrogen for 6 h at 100°C. The reaction mixture was diluted with water (225 ml) and concentrated in a rotatory evaporator. This procedure was repeted with 2x100 ml of water. The residue was distilled to almost dryness in vacuum and dissolved in 2 M HCI (25 ml) and 2 M NaOH (3 ml). Ultrafiltration and freeze-drying gave 0,81 g. 0,66 g of this product was dissolved in water (12 ml) and the pH was adjusted to 9,4 with 2 M NaOH. The solution was heated to 55°C, sulphur trioxide trimethylamine complex (0,65 g) was added, followed by NaHC03 (0,55 g ) The reaction mixture was stirred at 55°C for 24 h and then cooled to +5°C. A white precipitate was formed. The mother liquor was decanted and the crystals washed with water (2x5 ml). To the combined water solutions NaCI (2,9 g) and 2 M NaCI (30 ml) was added. Ultrafiltration and freeze-drying yielded 0,73 g crude product. This product (0,65 g) was dissolved in water and passed through a cation axchange resin (2x16 cm, H+- form) and eluated with water (100 ml). The eluate was neutralized to pH 8,1 with NaOH and freeze-dried to yield the product (0,58 g).

NMR showed no signal at 2.2 ppm. (Sulphur by elemental aπalyε 6.84 %; N-acetyl groups of total amino groups by H-NMR: <5 %)

Kabi 2236. carbQxyl reduced

Heparin fragment Kabi 2236 (10 g) was dissolved in water (1.5 I). The pH of the solution was adjusted to pH ■ 4.7 with 0.1 M HCI. 1- Ethyl-3-(3-dimethylaminopropyl) carbodiimide (30 g) was added in small portions. 0.1 M HCI was simultaneously added to keep the pH of the solution at 4.7-4.8. After 20 min, solid sodium borohydride (120 g) was added in small portions during 1 h. This solution was warmed at 50°C for 2 h. The pH rose to 11.2. The excess of sodium borohydride was destroyed by 1 M H2SO4. Desalting was carried out by ultrafiltration in Amicon cells 8400 using YM2 filters. The retentate was washed with 0.2 M NaCI and water, concentrated to 20 g and precipitated by ethanol (200 ml). After drying in vacuum for 3 h at 60° C, 6.4 g heparin fragment was obtained. 6 g of this heparin fragment was subjected to a second reduction in the same way as described to give 3.5 g product. 1.5 g of this was dissolved in 10 ml of water and passed through an ion exchange resin (Dowex 50Wx8 H+). The acidic eluate was neutralized by addition of 2 M NaOH. It was concentrated (10 g) and precipitated with ethanol (40 ml). The precipitate was vacuumdried to give the carboxyl reduced heparin fragment (1 ,1 g)

(Sulphur by elemental analysis: 12.9 %; N-acetyl groups of total amino groups by H-NMR: <5 %)

Heparin fragment bv periodate oxidation

PM heparin (1.00 g) was dissolved in 0.05 M phosphate buffer, 0.2 M NaCI, pH 7.0 (50 ml) and heated to 39°C in a dark flask. Sodium periodate (1.07 g) was added and the reaction mixture was stirred at 39°C for 19 h. Ethylene glycol (0.5 ml) was added and the solution was ultrafiltrated and the retentate washed with water. The retentate was dissolved in 0.02 M NaOH (100 ml), after 35 min at 21 ° C the product was rediced by NaBH4 (0.20 g) over 2 h. The excess of NaBH4 was decomposed by HOAc (1.2 ml) and the solution was then neutralised by NaOH, desalted by ultrafiltration and lyophilised to yield 0.56 g. This product was subjected to ion

exchange chromatography on DEAE-Sepharose. The column was washed with 0.3 M NaCI and the eluted with 2 M NaCI. This fraction was precipetaded with EtOH to yield 0.39 g periodate oxidated heparin fragment after drying at 60°C for 3 h. Sulphur by elemental analysis was 12.8 % and SO3/COO was 2.35).

Properties of the heparins, heparin fragments and the chemically modified heparin fragments are summarized in Table 1.

Methods

A template bleeding time test according to Dejana et al. (Bleeding time in laboratory animals. II. A comparison of different assay conditions in rats. Thromb Res 1979; 15: 191 -197) was performed in Sprague-Dawley rats weighing 200-250 g anaesthetized with Mebumal/Stesolid (Dumex A/S Copenhagen). The template device (Simplate, General Diagnostics, Durham, North Carolina) was applied longitudinally to the dorsal part of the tail, taking care to avoid large veins. Blood from the wound was then carefully removed every 30 seconds, with blotting paper. A minimum of six rats were used for each compound and dose. Bleeding times were measured from the moment the tail was incised until the first arrest of bleeding. The bleeding time was recorded with an accuracy of 30 seconds and bleeding times longer than 20 minutes were noted as >20. Two bleeding times were always determined in each rat, viz. 10 minutes before and 10 minutes after drug administration, and the results are expressed as the prolongation of bleeding time.

The molecular weights (Mr) were determined by analytical gel filtration on Sephadex G-75 (Pharmacia Uppsala Sweden) using reference samples with a known degree of polymerization, obtained by nitrous acid depolymerization of heparin.

The anticoagulant activity of heparin and heparin fragments was determined by an APTT assay (Andersson L-0 et al, Anticoagulant properties of heparin fractionated by affinity chromatography on

matrix bound antithrombin III and by gel filtration. Thromb Res 1976; 9: 575-583) and by an anti-FXa assay using the chromogenic substrate S-2222 (Kabi Diagnostica, Sweden) (Andersson L-0 et al, Molecular weight deper.dency of the heparin potentiated inhibition of thrombin and activated FX. Effect of heparin neutralization in plasma. Thromb Res 1979; 15: 531-541 ). All the activities were measured against a house standard which was calibrated against the 4th International Standard preparation of heparin. Elemental analyses (S) were performed by the Analytical Chemistry Department, University of Lund, Sweden, on samples dried for 3 h at 60°C in a vacuum. NMR spectra were recorded in D20 or D20/NaOD on a JEOL GX-400 instrument (400 MHz " I H) at 40°C.

In vitro inhibitory capacity on complement activation was studied by measuring the reduction of C3c formation using crossed immunoelectrophoresis as described (Ekre H-P T. Inhibition of human and guinea pig complement by heparin fractions differing in affinity for antithrombin III or in average molecular weight. Int. J. Immunopharmac.1985; 7, 271-280) as follows. To 100 μl of normal human serum was added 25 μl of r^ivating substance (heat- aggregated human IgG) and 25 μl oi the sample diluted in Dulbecco ' s PBS buffer (D-PBS). The rπxture was vortexed and then incubated for 15 min at 37°C. A five μl sample of the mixture was then immediately applied to wells in an agarose gel for electrophoresis in the first dimension (25 V/cm for ~ 2 h). Electrophoresis was performed using 1 % agarose gels (in Tris-barbiturate buffer supplemented with EDTA to 10 nM, pH 8.6) with rabbit anti-human C3 antibodies in the gels for electrophoresis in the second dimension (7-10 V/cm for ~16 h). Each experiment included references for 100% and 0% inhibition; 100 μl serum + 50 μl D-PBS and 100 μl serum + 25 μl activator + 25 μl D-PBS respectively, and at least three concentrations of each sample at test. The concentration of the samples giving a 50 % inhibition was calculated from the data obtained.

This method was found to correlate very well to the anti- inflammatory effect excerted in vivo by heparin fractions (Ekre H-P T et al. Inhibition of complement dependent experimental inflammation in human skin by different heparin fractions. Immunopharmac. 1986; 8, 277-286)

Results

The pretreatment bleeding time in the anaesthetized rat ranged between 2.0 and 4.0 minutes (2.7 ± 0.6 minutes) and the average prolongation of the bleeding time before/after injection of saline was 0.8 ± 0.7 minutes. A prolongation of less than about 1.5 minutes therefore reflects the variability of the model rather than a true prolongation of the bleeding time.

All results discussed below are referred to table 1. A bleeding time of more than 20 minutes was observed in all rats treated with 2 mg/kg of a PM-heparin (Mr 14 000). Significantly shorter bleeding times were obtained with the same dose of Kabi 2165 (Mr 5000) and Kabi 2226 (Mr 2500), which gave prolongations of the bleeding time of only 3.6±3.2 minutes and 2.7+1.8 minutes, respectively. In fact, the smallest fragment, when administered at a dose of 32 mg/kg, gave the same bleeding time as that of 1 mg/kg of the PM-heparin from which it was prepared. Bovine heparin was on a weight basis, less haemorrhagic than heparin from porcine origin although both heparins had a similar anticoagulant activity. This difference, however, disappeared after nitrous acid depolymerization of BL and PM-heparins to low molecular weight fragments of similar Mr and anticoagulant activity, which both showed a similar effect on the bleeding time.

Kabi 2165, Fragmin® caused a significant prolongation of the bleeding time after a dose of 2,5 mg/kg, and at higher doses the bleeding times were longer than 20 minutes in 16 to 33 per cent of the animals. HA-Kabi 2165 produced a significant prolongation of the bleeding time at as low a dose as 0.5 mg/kg. At doses of 2mg/kg or more most of the animals showed bleeding times that

were longer than 20 minutes. LA-Kabi 2165 caused a short, and none-dose-dependent, prolongation of the bleeding time at 1- 4 mg/kg .

A BL-heparin fragment (Kabi 2236 ) was used as starting material for selective chemical modifications. This fragment caused a considerable prolongation of the bleeding time at a dose of 4 mg/kg. A total conversion of the N-sulphate groups into N-acetyl groups eliminated both the anti-FXa and the APTT activity ( 1 U/mg) as well as all effects on the bleeding time.

Oversulphation of the totally N-c -tylated heparin fragment also resulted in an increased APTT a uvity but no significant change in the already very low anti-FXa activity. The bleeding time of this compound was somewhat longer than that of the N-acetylated starting material, but it was still much shorter than that of the original heparin fragment Kabi 2236 when compared on a weight basis.

N/O-desulphation, N-resulphation of a BL-heparin fragment virtually eliminated anticoagulant activities. This modification also abolished the haemorrh-nic effect, as no prolongation of the bleeding time was seen after a dose of 16 mg/kg.

The twice carboxyl-reduced heparin fragment was practically devoid of APTT activity and also of anti-FXa activity but retained the high sulphur content of the starting BL-heparin fragment. This compound caused only small changes in the bleeding time and a dose of 16 mg/kg had to be admtnistered before any significant prolongation could be demonstrated.

From table 1 it can be seen that even if the anti-FXa (.Kabi 2226) or APTT activity ( Kabi 2236, N acetylated O-oversulphated) is high for some substances, the bleeding time can be short when using high doses. Bleeding time must thus always be controlled and measured when looking for effective drug without adverse effect. This finding is new.

Among the heparins and non-modified heparin fragments tested the complement inhibiting potencies were relatively similar; around 20 μg caused a 50 % inhibition of C3 activation, with the bovine materials showing a slightly higher potency. Chemical modifications of the heparin fragments had definite influences also on the complement inhibition. N-acetylation as well as carboxyl-reduction resulted in reduced, but still appreciable, inhibitory potency (as demonstrated by an increased dose required for 50 % inhibition) whereas O-desulfation (N/O-desulfation, N-resulfation)practically abolished the ability to inhibit complement activation. Oversulfation of N-acetylated fragments, on the other hand, resulted in a very high complement inhibitory potency. Thus, when compared to the corresponding non-modified fragment (Kabi 2236) the N- acetylated, oversulfated fragment had a clear increase in complement activity but also an increase in APTT activity. Surprisingly, this increased APTT-time was not accompanied by an increase but rather a marked reduction in bleeding time.

Discussion

It has been demonstrated in several clinical studies that LMW heparin fragments have an antithrombotic effect similar to that of heparin but they have a weaker effect on lipoprotein lipase and on platelets. Besides, they have a much higher bioavailability and a longer plasma half-life, two properties that make them more convenient both for the patient and for the medical staff. It has also been demonstrated in several precliπical studies that low molecular weight heparin fragments have a weaker haemorrhagic effect than heparin. This effect, however, has not yet been convincingly verified in human studies.

We have now studied the complement inhibition activity. We have also measured template bleeding time in the rat tail after i.v. injections. Heparin from different sources and low molecular weight heparin fragments differing with respect to anticoagulant activity as well as chemical composition were used. All the tested

compounds were characterized by their Mr, their specific anti-FXa and APTT activity and the complement inhibition. The chemically modified heparin fragments were also characterized by degree of N- acetylation and sulphur content and, for the carboxyl reduced fragment, the extent of reduction.

Depolymerization of PM-heparin to fragments of lower Mr was always associated with a reduced bleeding time if the compounds were compared on a weight basis, despite their similar anti-FXa activities.

Although the size of heparin or different heparin fragments seems to play a certain role in their effect on the bleeding time, the surprising finding in the present invention is the lack of correlation between bleeding time and anticoagulant activity.

The APTT activity has in vitro assay been considered to better reflect the in vivo bleeding risk than the anti FXa, but the results presented here, where N-acetylation and oversulphation of heparin fragment 2236 increased the APTT by more than 50 % and yet markedly reduced rather than prolong the bleeding time, shows that bleeding time in vivo cannot be predicted by APTT but has to be measured.

Therapeutic application of heparin for inhibition of complement activation in pathological states of inflammation (e.g. myocardial infarction, respiratory distress syndrome, septicemia, autoimmi-ne diseases) has been hampered by the risk for bleeding. Those heparin derivatives tested here with no or reduced risk on bleeding and with retained (Kabi 2165, LA-Kabi 2165, Kabi 2236), somewhat reduced (N-acetylated, Carboxyl reduced) or even enhanced (N-acetylated, O- oversulphated) complement inhibiting capacity as compared to heparin should thus be useful for treatment of pathological states of inflammation. The N/O desulphated, N-resulphated heparin fragment, on the other hand, is practically without effect on complement activation and would therefor be of negligable interest for therapeutic modulation of complement activation.