TECHNICAL FIELD

The present invention relates to prevention and treatment of thromboembolic diseases by administration of a therapeutically effective dose of an active trisacc- haride comprising an iduronic acid, a glucopyranosyl or glucopyranosylamine unit and a hexuronic acid selected from iduronic acid and glucuronic acid. The specific trisaccharides of the present invention have potent activity towards thrombin and also potentiates the inhibition of FXa, even if the trisaccharides lack sulfate groups. With the present invention, safe and potent anticoagulants with a high oral absorption are potentially achievable, due to the much smaller mole¬ cular size and presence of very few, if any, sulfate groups compared to presently available anticoagulants.

BACKGROUND OF THE INVENTION

Generally, when considering the prevention of thromboembolic processes three groups of substances need to be considered: a) substances which prevent platelet aggregation (platelet antiaggregation agents), b) agents which inhibit the coagulation process (anticoagulants), and c) substances which eliminate fibrinogen from the circulation (defibrinogenating agents).

The most well-known anticoagulant and antithrombotic drugs are unfractionated (standard) heparin and the low molecular weight heparins (LMWHs), the latter compounds being introduced in clinical practice during the last decade. The major anticoagulant effect of standard as well as low molecular weight heparins, is through its interaction with AT III on coagulation enzymes particularly thrombin (factor II a) and factor Xa (FXa).

Anticoagulant drugs that inhibit the synthesis of vitamin K - dependent coagula¬ tion factors, such as warfarin, are also well known. Because warfarin prophylaxes of venous thromboembolism is more complicated to use than fixed low-dose heparin therapy, warfarin is generally reserved for patients at very high risk (J. Hirsh, Oral anticoagulant Drugs, N. Engl. J. Med., 324 (1991) 1865-1875).

Today far more LMWHs than standard heparin are being used in clinical practice. This is because, the LMWHs cause less hemorrhage than standard heparin at doses producing an equivalent antithrombotic effect. Also, the LMWHs are effective and safe for the prevention of venous thromboembolism in surgical and medical patients who are at risk. Furthermore, for heparin numerous side-effects have been documented such as hemorrhage, thrombocytopenia with or without thrombosis, osteoporosis, skin necrosis, alopecia, hypersensitivity reaction and hypoaldosteronism. (J. Hirsh, Heparin, N. Engl. J. Med., 324 (1991) 1565-1574).

The LMWHs are also more convenient to use than standard heparin. Generally, a once daily subcutaneous dose is sufficient for LMWH, whereas 2 to 3 daily doses are required for heparin. At present, however, no formulation is available for oral administration of heparin or LMWHs.

A heparin-like anticoagulant of similar molecular weight (6 kDa) as the LMWHs is the heparinoid Danaparoid sodium (Orgaran ^® , trademark of Akzo of the Nether¬ lands). It consists of a polydisperse mixture comprising sulfated glycosamino- glycans derived from animal mucosa, heparan sulfate (83 % w/w), of which some 4 to 5% has high affinity for antithrombin III (AT III), dermatan sulfate (12% w/w) and a minor amount of chondroitin sulfate (5% w/w).

The major anticoagulant effect of heparin as well as of low molecular weight hepa- rins and heparinoids, are through their interaction with AT III particularly on thrombin and FXa, as already stated above. Heparin molecules with fewer than 18

saccharides (MW less than 5,400 Da) are unable to bind thrombin and AT III simultaneously. Therefore, with less than 18 saccharides heparin molecules are unable to accelerate the inactivation of thrombin by AT III but retain their ability to catalyze the inhibition of FXa by AT III (J. Hirsh and M. N. Levine, Low Molecular Weight Heparin, Blood, 79 (1992) 1-17).

In order for heparin, LMWHs and Danaparoid sodium to bind to AT III and thereby express their anticoagulant effect they must contain a highly sulfated pentasaccharide sequence found in heparin, LMWHs and some heparan sulfates. Synthetic, highly sulfated pentasaccarides which bind to AT III and inhibit FXa and thus exhibit anticoagulant activity have been described. These highly sulfated pentasaccharides correspond to the natural hepta and octasulfated pentasaccha- rides as well as some additional non-natural octasulfated and nonasulfated penta¬ saccharides. Notably, no binding to AT III took place at all when the 3-O-sulfate group of the middle glucopyranosylamine unit of the pentasaccharides was lacking (Nature, 353 (1991) 30-33 suppl.). This shows that the presence of this 3-0- sulfated glucopyranosylamine unit is critical for these pentasaccharides.

Efficient and well-controlled anticoagulants of the prior art derived from heparin or heparan sulfate are highly sulfated and exhibit a molecular weight and size, making them unsuitable for oral administration.

One of the trisaccharides presently claimed for treatment of thromboembolic diseases is known per se. Thus, the structure and synthesis of the claimed trisacc- haride designated in this specification as Tri-3 has been disclosed in M. Nilsson et al, Carbohydrate Res., 246 (1993) 161-172 and J. Westman et al, J. Carbohydrate Chem., 14(1) (1995) 95-113. These references, however, do not disclose any infor¬ mation in general and much less any experimental support showing anticoagulant activity or potential administration by the oral route of the presently claimed trisaccharides.

Further oligosaccharides with few saccharide units and containing idopyranosyl- uronic and /or glucopyranosyluronic acids are known from the prior art. For example, in EP-A-0454220 (assigned to Akzo and Sanofi), carbohydrate derivatives comprising a trisaccharide unit are described. All of the trisaccharides are highly sulfated and have an α-D-glucopyranosyluronic acid at their non-reducing end. Furthermore, no anticoagulant nor antithrombotic activity have been shown. EP-A-0084999 (assigned to Choay), discloses production of mucopolysaccharide products having 2-12 saccharide units for use as antithrombotic agents. There are, however, up till now no disclosures connected to such oligosaccharides showing anticoagulant activity of such importance that administration of an anticoagulant e.g. by the oral route can be envisaged.

Heparin and LMWHs are heterogeneous mixtures of polysaccharide fragments with multiple sites of pharmacological action, which cannot be used successfully by oral administration. Even the above mentioned synthetic pentasaccharides, although being more defined than heparin and LMWHs, are still being too large molecules and much to sulfated for oral administration.

There is thus a need for a new potent and safe anticoagulant for prevention and treatment of thromboembolic diseases. More particularly, there is a need for new anticoagulants with enhanced oral absorption compared to presently used anti¬ coagulants, to make possible efficient administration by the oral route.

DESCRIPTION OF THE INVENTION

One object of the present invention is efficient inhibition of thrombin without the adverse side-effects encountered with standard heparin.

Another object of the present invention is efficient inhibition of thrombin by AT III in the presence of compounds with a molecular weight and size which are sub¬ stantially smaller, more defined and less sulfated than conventional LMWHs.

A further object of the present invention is efficient inhibition of FXa by AT III in the presence of compounds with a molecular weight and size which are substan¬ tially smaller, more defined and less sulfated than conventional LMWHs.

Yet another object of the present invention is to provide potent inhibitors of thrombin and FXa having properties suitable for oral administration to the patients.

These and other objects are fulfilled by the present invention, which relates to a method for prevention and treatment of thrombosis by administration of a therapeutically effective dose of an active trisaccharide of the general formula I

IdoA-(l→4)-GlcX-(l→4)-HexA(l→)-OR. (I)

in which

IdoA = idopyranosyluronic acid (at the non-reducing end)

GlcX = glucopyranosyl unit, where

X = OH or NHY in the 2-position, where Y = hydrogen, acetyl, propionyl or butyryl (straight or branched), and

HexA = hexuronic acid selected from the group consisting of glucopyranosyl- uronic acid (Glc A) and idopyranosyluronic acid (IdoA), and R. = hydrogen, alkyl or aryl, or any pharmaceutically acceptable salt or ester thereof.

In the following iduronic acid or IdoA are used as abbreviations for idopyranosyl¬ uronic acid, idopyranosyluronate or esters of any of these. In the same way, glucuronic acid or GlcA are used as abbreviations for glucopyranosyluronic acid, glucopyranosyluronate or esters of any of these. When Y in NHY equals hydrogen, then a pharmaceutically acceptable salt is suitable, e.g. hydrogen chloride.

The charged groups are compensated by pharmaceutically acceptable counter¬ ions, such as hydrogen, alkali or alkaline-earth metal ions, such as sodium and calcium, respectively, or ammonium ions, which optionally may be substituted. It is also conceivable to use metal ions, primarily copper or zinc ions, as counter¬ ions.

Easily hydrolyzable ester groups can be substituted for one or more of the hydroxyl groups of the trisaccharides according to formula I. This is applicable to all three saccharide units of the trisaccharides. The introduction of easily hydrolyz¬ able ester groups, can also be made in one or both of the carboxylic groups of the uronic acids.

The inventors of the present invention, have found that certain trisaccharides potentiate the anticoagulant activity of AT III. Thus, the trisaccharides according to formula I have potent activity towards thrombin and also potentiates the inhibition of FXa. This is very surprising, since previously it has been believed that at least 18 saccharide units were necessary for binding thrombin and AT III simul¬ taneously. Even more surprisingly is the high activity exhibited by these trisaccha- rides, although they lacked sulfate groups.

The selection of the trisaccharides, has not been made at random. On the contrary, several experiments with di, tri and tetrasaccharides containing iduronic acid, glucuronic acid and /or a glucopyranosyl unit in various combinations, have revealed that the molecular size of a trisaccharide is a preferred size. Furthermore, the sequence of uronic acids and an intermediate glucopyranosyl unit of these tri¬ saccharides is fundamental for the extraordinarily and surprisingly good anticoa¬ gulant activity of the trisaccharides according to formula I.

A representation of trisaccharides suitable for use in the present invention is given as formula II:

in which

A = an iduronic acid (IdoA) linked 1-4 to

B = a glucopyranosyl or glucopyranosylamine unit linked 1-4 to C = a uronic acid, which can be either a glucuronic acid (Glc A) or an iduronic acid

(IdoA), and

R, = H, alkyl or aryl,

R _j = COO ^' and R ₃ = H when C = β-D-GlcA

R _j = H and \ - COO ^' when C = α-L-IdoA, and R ₄ = OH or NHY where Y = hydrogen, acetyl, propionyl or butyryl (straight or branched).

Independently, the IdoA unit A, the glucopyranosyl or glucopyranosylamine unit B, as well as the uronic acid unit C, can be either in their D or L forms. Suitably, the IdoA unit A is in its L form, and the glucopyranosyl or glucopyranosylamine unit B is in its D form. Suitably, the uronic acid unit C is in its D form when the uronic acid is a GlcA unit and in its L form when the uronic acid is a IdoA unit. The combination of L-IdoA (A), D-glucopyranosylamine unit (B) and D-GlcA (C) (see formula II) is a preferred embodiment.

Independently, the IdoA unit A, the glucopyranosyl or glucopyranosylamine unit B, as well as the uronic acid unit C, can be either in their α or β forms. Suitably, the IdoA unit A is in its α form, and the glucopyranosyl or glucopyranosylamine unit B is in its α form. Suitably, the uronic acid unit C is in its β form when the uronic acid is a GlcA unit and in its α form when the uronic acid is a IdoA unit. The

combination of α-IdoA (A), α-glucopyranosylamine unit (B) and β-GlcA (C) (see formula II) is a preferred embodiment.

The combination of α-L-IdoA (A), α-D-glucopyranosylamine unit (B) and β-D- GlcA (C) (see formula II) is a particularly preferred embodiment.

In the NHY group of the glucopyranosylamine unit (B), Y can be an acetyl, propionyl or butyryl group, the two latter of which can be straight or branched. Preferably, Y is an acetyl group and the glucopyranosylamine unit then is a 2- acetamido-2-deoxy-glucopyranosyl unit (GlcNAc).

In order for heparin, LMWHs and some heparan sulfates to bind to AT III and thereby express their anticoagulant effect they must contain a highly sulfated pentasaccharide sequence, as already mentioned above. It is thus a prerequisite of these anticoagulants to contain at least four sulfate groups, and preferably seven or eight groups. In the present invention, one or two OH groups of the trisaccha¬ rides may be sulfated. A suitable position is the 6-position of the glucopyranosyl or glucopyranosylamine unit (B) (see formula II). This is, however, less preferred embodiments, since the presence of sulfate groups decreases the oral absorption. The low sulfate content facilitates administration of an inventive preparation con¬ taining the trisaccharides according to formula I, aimed for prevention and treat¬ ment of thrombosis and treatment of arteriosclerosis or use for manufacture of a medicament for prevention and treatment of thrombosis and treatment of arterio¬ sclerosis. This is further emphasized by the fact that a potentiation of AT III- mediated inhibition of FXa was observed for a trisaccharide according to formula I lacking sulfate groups.

The inventive pharmaceutical compositions containing the trisaccharides according to formula I are suitable for parenteral administration, optionally with one or more pharmaceutically acceptable additives. Parenteral administration here refers to intravenous, subcutaneous or intradermal administration. However,

trisaccharides in general exhibit a much smaller molecular size and weight than conventionally used heparin-derived anticoagulants. This implies a higher proba¬ bility of good oral absorption. The inventive pharmaceutical compositions con¬ taining the trisaccharides according to formula I, optionally with one or more pharmaceutically acceptable additives, are particularly suitable for oral, buccal, sublingual or rectal administration. This is because the present trisaccharides are less sulfated than hitherto known oligosaccharides. A particularly preferred embodiment of the inventive pharmaceutical compositions are those trisaccha¬ rides where sulfate ^τ roups are completely lacking, since this means a particularly low molecular weignt and size and good oral, buccal or sublingual absorption.

The chemical synthesis of one trisaccharide according to formula I has been published. Thus, synthesis of Tri-3 and other trisaccharides, as well as certain di, tri and tetrasaccharides has been disclosed in M. Nilsson et al, Carbohydrate Res., 246 (1993) 161-172 and J. Westman et al, J. Carbohydrate Chem., 14(1) (1995) 95- 113. The trisaccharides may also be prepared by chemical and or enzymatic treat¬ ment of heparin followed by fractionation and purification.

In another embodiment of the present invention, the trisaccharides of formula I are covalently attached to a soluble or insoluble support. In this way use is made of the fact that the said trisaccharides are strongly bound to AT III and /or throm¬ bin and/or FXa. Preferably, the trisaccharide is attached at its non-reducing end, optionally by a spacer between the trisaccharide and the support. A particularly preferred trisaccharide in this embodiment is α-L-IdoA-(l→4)-α-D-GlcNAc-(l→4)- β-D-GlcA-1— = OMe. These supported trisaccharides can be used for purifying heparin and heparan-binding proteins, especially AT III, from fermentation broths, human or animal blood plasma or other mixtures.

The invention will now be illustrated in more detail with the aid of exemplifying, non-limiting examples.

EXPERIMENTAL SECTION

Di, tri and tetrasaccharides having a carbohydrate backbone structure similar to heparin and heparan sulfate were synthesized and tested for anticoagulant activity in the presence of AT III and thrombin (Examples 1 and 2) or FXa (Example 3).

The molecular structure of the synthetic oligosaccharides are evident from Table I.

TABLE I

Structure of the synthetic oligosaccharides used in the experiments. Tri-3 is a tri¬ saccharide used to illustrate the present invention. For comparison, two trisaccharides (Tri-1 and Tri-2), five disaccharides and two tetrasaccharides were also tested.

Structure

A B C D

Disaccharide

Di-1 α-LJdoA-(l→4)-α-D-GlcNAc-l→OMe

Di-2 α-L-IdoA-(l→4)-α-D-GlcNS0 ₃ -l→OMe Di-3 β-D-GlcA-(l→4)-α-D-GlcNAc-l→OMe

Di-4 β-D-GlcA-(l→4)-α-D-GlcNS0 ₃ -l→OMe

Di-5 α-D-GlcNAc-(l→4)- β-D-GlcA-l→OMe

Tetrasaccharide

Tetra-1 α-LJdoA-(l→4)-α-D-GlcNAc-(l→4)- β-D-GlcA -l-(l→3)~β-D-Gall→OMe Tetra-5 β-D-GlcA-(l→4)-6-S0 ₃ -α-D-GlcNAc-(l→4)-β-D-GlcA-(l→4)- α-D-aManOH

Trisaccharide

Tri-1 β-D-GlcA-(l→4)-α-D-GlcNAc-(l→4)- β-D-GlcA-l→OMe

Tri-2 α-L-IdoA-(l→4)-α-D-GlcNS0 ₃ -(l→4)- β-D-GlcA-l→OMe

Tri-3 α-L-taoA-(l→4)-α-D-GlcNAc-(l→4)- β-D-GlcA-l→OMe

Example 1

Experiments were carried out with one trisaccharide according to the invention to illustrate the effect on thrombosis. For comparison, experiments were also carried out with two other synthetic trisaccharides as well as with synthetic di and tetra- saccharides, as well as standard heparin, hexadecasaccharide and sucrose octa- sulfate, the remaining experimental conditions being kept constant.

Experimental conditions:

The reaction mixture contained 150 nM of antithrombin (AT III) and 66 μM of one oligosaccharide in each experiment. The reaction was initiated bv the addition of 2-3 nM of thrombin. The reaction buffer contained 0.15 M of NaCl, 20 mM of Tris- HCl with pH 7.4, and 0.1% of PEG 8000. 50 μl samples were then taken out every 15 seconds for quantification of the residual thrombin activity. Residual thrombin activity was measured by adding the 50 μl aliquots to 250 μl of 300 μM of the chro¬ mogenic substrate S-2238 (obtained from Chromogenix of Mδlndal, Sweden), 5.4 μg/ml polybrene, and 0J2 klU/ml aprotinin. The residual thrombin activity was determined by calculating the rate of hydrolysis of S-2238 over 2 min.

Calculations:

The residual thrombin activity versus the time at which the residual thrombin activity was determined was plotted. The curve obtained was fitted to a single exponential to give a pseudo 1st order rate constant. The pseudo 1st order rate constant was divided by the inhibitor (AT III) concentration to give the 2nd order rate constant.

The 2nd order rate constants are evident from Table II.

TABLE π Determining the inhibition of thrombin by AT III in the presence of various saccharides.

Saccharide 2nd order rate constant, IO ⁵ x (mol ^"1 x min ^"1 )

None 5

Heparin (standard) 5,000

Hexadecasaccharide * 12

Sucrose octasulfate (SOS) 5

Di-1 5

Di-2 5

Di-3 60

Di-4 60

Di-5 5

Tetra-1 5

Tetra-5 5

Tri-1 30

Tri-2 5

Tri-3 900

* Highly sulfated hexadecasaccharide prepared from heparin. It is apparent from Table II, that the inventive trisaccharide (Tri-3) has a consider¬ ably higher ability to inhibit thrombin than the other trisaccharides as well as all of the di and tetrasaccharides tested.

Example 2

Experiments were carried out with one trisaccharide according to the invention

(Tri-3), to study the inhibition of thrombin by antithrombin.

Assays were performed at room temperature with reaction mixtures containing 150 nM of AT III and increasing amounts of Tri-3 in a buffer similar to that

described in Example 1. The reaction was initiated by the addition of 2-3 nM of thrombin. Samples of 50 μl were removed every 15 seconds for quantification of the residual thrombin activity by determining the rate of hydrolysis of S-2238 as described in Example 1. Residual thrombin activity versus time was plotted and values for the first order rate constant (k') were determined by fitting data by non¬ linear least-squares regression to equation 1:

E _t /E ₀ = e " (1)

in which t = time,

E _o = thrombin activity at t = 0,

E _t = thrombin activity at t = t, and k' = pseudo-first-order rate constant.

The correlation coefficient (R) of the fit was greater than 0.96 in all cases. The second-order rate constant of inhibition, k ₂ , was then obtained by dividing the pseudo-first-order rate constant by the inhibitor concentration (I).

For inhibition in the presence of heparin and heparin analogs, thrombins with k, ≤ 1 x IO ⁷ (M ^"1 x min ^"1 ) were assayed as above. Thrombins with , >1 x 10 ⁷ (M ^"1 x min ^"1 ) were assayed by monitoring the progress curves in the presence of 75-150 μM S- 2238, 150 nM antithrombin and 0.5-3 nM thrombin. The hydrolysis of S-2238 to yield p-nitroaniline product (P) was determined continuously at an optical density of 405 nm (OD ₄₀₅ ) over time (t). The data were fitted by non-linear regression (Marquardt algorithm) to equation 2, to obtain estimates of initial velocity (v ₀ ), steady state velocity (v _s ), and the apparent first-order rate constant (k'):

[P] = v _s x t + (v ₀ -v _s )/[k'(l-e ^'k' ')] (2)

No significant substrate depletion of product inhibition was noted over the substrate ranges and time interval determined. Under these conditions, the values of k, is related to k' by equation 3:

k ₂ = k'(l-v /v ₀ )(l + [S]/Km)/[I] (3)

in which

[S] = concentration of substrate (S-2238) Km = Michaelis constant for the substrate [I] = concentration of antithrombin.

Inhibition of thrombin by antithrombin according to the present invention is illustrated in Fig. 1, which is a dose-response curve for Tri-3. In Fig. 1, the second- order rate constant of inhibition, k,, is given as a function of the concentration of Tri-3.

As is evident from Fig. 1, the inhibition of thrombin increases dramatically with the concentration of Tri-3 revealing that Tri-3 is a potent antithrombotic agent.

Example 3

Experiments were carried out to determine inhibition of FXa activity by one natural and various synthetic oligosaccharides. The investigated synthetic saccharides were one trisaccharide according to the invention (Tri-3), and for comparison two trisaccharides (Tri-1 and Tri-2), one disaccharide (Di-1) and one tetrasaccharide (Tetra-1). The structures of these synthetic oligosaccharides are given in Table I. Heparin obtained from Pharmacia Hepar Inc., was used as control. A molecular weight of 15,000 was used to calculate the molar concentra¬ tion of heparin.

In a first experiment, the comparison of the inhibitory effect of different synthetic oligosaccharides upon activity of FXa in the presence of AT III was investigated. The reaction with FXa was followed with synthetic chromogenic peptide substrate Bz-Ile-Glu-Gly-Arg-pNA (S-2222) from Chromogenix of Mδlndal, Sweden.

The assay was carried out in a 0.05 M Tris-HCl buffer with pH 7.5, the buffer containing 0.15 M NaCl, 0.01 M CaCl. and 0.01% Tween 80. The same buffer was used for dilution of the human proteins employed, whereas saccharides were diluted in water. Human FXa was obtained from Enzyme Research Laboratories of South Bend, IL, USA and human antithrombin (AT III) from Chromogenix of Mόlndal, Sweden. FXa was active site titrated with NPBG using the method of Chase and Shaw (Methods in Enzymology, 1970).

25 μl of saccharide (12 μM final concentration for synthetic oligosaccharides and 417 pM for heparin) were added to 25 μl of AT III (142 nM final concentration) in the wells of microtiter plate followed by 100 μl of buffer and 25 μl of S-2222 (0.5 mM final concentration). The reaction mixture was incubated for 10 min at 37°C. The reaction was started by addition of 25 μl FXa (3 nM final concentration) and the release of p-nitroaniline, proportional to the amount of FXa activity, was deter- mined at an optical density of 405 nm (OD ₄₀₅ ) for 15 min. In control experiments, AT III was removed from the assay mixture. The results are given in Table III and expressed as mOD/min determined 1 min after start of the reaction.

TABLE III Comparison of the inhibitory effect of synthetic oligosaccharides on FXa activity in the presence and absence of AT III. FXa is present in all experiments, except the control.

Saccharide FXa activitv, n lOD/min

With AT III Without AT III

Control 53.33 69

Heparin (standard) 44 67

Di-1 57 64

Tetra-1 50 66

Tri-1 58 70

Tri-2 50 68

Tri-3 12 66

As is evident from Table III, Tri-3 gives a significant decrease in FXa activity whereas the other synthetic oligosaccharides show only minor or insignificant effects.

In a second experiment, the dose response of Tri-3 was investigated. The final concentrations of Tri-3 were 316, 105, 35, 12, 3.9 and 1.3 μM. The other experi¬ mental conditions were the same as in the first experimental series. For compari¬ son, one experiment was carried out with 417 pM of standard heparin, as well as one experiment without AT III at the highest concentration of Tri-3. The results of the comparative experiments with standard heparin and without AT III, gave an inhibition of 46 and 66 mOD/min, respectively. The control experiment without Tri-3 and heparin, but with AT III, gave a reaction rate of 67 mOD/min. Inhibition of FXa by different concentrations of Tri-3 is given in Fig. 2 and expressed as mOD/min determined 1 min after start of the reaction.

As is evident from Fig. 2 and the comparative experiments, the inhibition of FXa at a Tri-3 concentration of 1.3 μM is comparable to the inhibition by unf ractionated heparin, albeit the concentration of heparin is much lower.