US20030013682A1

US20030013682A1 - Derivatives of partially desulphated glycosaminologycans endowed with antiangiogenic activity and devoid of anticogulating effect

Info

Publication number: US20030013682A1
Application number: US10/182,185
Authority: US
Inventors: Casu Banito; Torri Giangiacomo; Naggi Anna Maria; Giannini Giuseppe; Pisano Claudio; Penco Sergio
Original assignee: Sigma Tau Industrie Farmaceutiche Riunite SpA
Current assignee: Sigma Tau Industrie Farmaceutiche Riunite SpA
Priority date: 2000-01-25
Filing date: 2001-01-24
Publication date: 2003-01-16
Also published as: US20050107331A1; ES2277911T3; MXPA02007195A; PL357209A1; JP5371167B2; PT1268558E; KR20020080381A; EP1268558A1; HU229509B1; US20050222084A1; KR100812406B1; CZ303030B6; CA2397964C; AU782632B2; CA2397964A1; CN1396930A; DK1268558T3; US7790700B2; SK287566B6; CY1108001T1

Abstract

Partially desulphated glycosaminoglycan derivatives are described, particularly heparin, and more particularly formula (I) compounds, where the U, R and R₁groups have the meanings indicated in the description. Said glycosaminoglycan derivatives are endowed with antiangiogenic activity and are devoid of anticoagulant activity.

Description

The invention described herein relates to partly desulphated glycosaminoglycan derivatives, particularly heparins, to processes for their preparation, to their use as active ingredients for the preparation of medicaments with an antiangiogenic activity, particularly for the treatment of tumours, such as, for example, the metastatic forms, and to pharmaceutical compositions containing them.

State of the Art

The first molecule possessing antiangiogenic activity was discovered in cartilage by Henry Brem and Judah Folkman in 1975. Since that year more than 300 new molecules capable of inhibiting angiogenesis have been discovered.

In the early ‘eighties, with the discovery of interferon (α/β) as an inhibitor of tumour angiogenesis, clinical experimentation was initiated based on this therapeutic approach.

The media caused quite a stir, when on Mar. 3, 1998 the New York Times published the news that two molecules, angiostatin and endostatin, discovered in J. Folkman's laboratories at the Harvard Medical School and Children's Hospital in Boston, were yielding very encouraging results in the struggle against cancer. The high degree of efficacy of these two molecules in inhibiting angiogenesis substantially boosted the search for new compounds. At present, there are about thirty molecules endowed with anticancer activity which act via an antiangiogenic mechanism, which have entered into the clinical trials stage [Phases I-III] and almost the same number of companies and institutions are involved.

In the United States alone it is estimated that there are approximately 9 million patients who could benefit from antiangiogenic therapy. At the present time, at least 4,000 patients have been enrolled into clinical trials using this therapy without any particular unwanted effects being registered.

Within the framework of the general concept of angiogenesis we should distinguish between vasculogenesis, that is to say the formation of blood vessels during embryonic development and angiogenesis in the strict sense of the term, meaning the formation of new blood vessels (capillaries) during the postnatal life starting from pre-existing vessels. The importance of angiogenesis for the growth of solid tumours is amply documented. Over the past three decades it has been reported that tumour growth, as well as the formation of metastases, are strictly dependent on the development of new vessels capable of vascularising the tumour mass.

The inhibition of angiogenesis underlies the formation of necrotic masses within the tumour or the induction of apoptosis in tumour cells.

There are clear-cut differences between neovascularisation in normal tissue and that in tumour tissue. In the former, the vascular endothelium represents a quiescent tissue with a very low mitotic index of its constituent cells (renewal time measured in hundreds of days), and the vascular network is regular, relatively uniform, and suitable for adequately oxygenating all the tissues, without any arteriovenous connection. In tumour tissue, on the other hand, stimulation of the proliferation of endothelial cells gives rise to a high mitotic index in the latter (mean renewal time 5 days), the neovascularisation is distinctly irregular with areas of occlusion, sometimes with closed endings, with arteriovenous contacts at some points, and, lastly, the basal membrane presents gaps, which at some points leads to tissue hypoxia. These differences offer the opportunity of identifying drugs which selectively block tumour neovascularisation.

In a tumour, the neovascularisation does not always coincide with a precise stage in the tumour development; there are, in fact, cases in which angiogenesis begins even before the development of the tumour (for example, carcinoma of the uterine cervix), others in which the two phases are coincident (for example, carcinoma of the bladder and breast), and others in which angiogenesis begins after the neoplasm (for example, melanoma and ovarian carcinoma; see, for example, “Manual of Medical Oncology”, IV ed. (1991) G. Bonadonna et al.

Antiangiogenic therapy presents numerous advantages compared to traditional standard chemotherapy (Cancer Research 1998, 58, 1408-16):

a) specificity: its target is a process, i.e. tumour neovascularisation;

b) bioavailability: its target is the endothelial cells, which can be easily reached without the problems of traditional chemotherapy which acts directly on the tumour cells;

c) chemoresistance: this is perhaps the most important advantage of this therapy; in fact, since endothelial cells, unlike tumour cells, are genetically stable, drug resistance phenomena are unlikely to occur;

d) metastatic spread: blockade of the neovascularisation limits the propagation of the tumour cells to other parts of the body via the bloodstream;

e) apoptosis: blockade of the vascular network in the tumour reduces the supply of oxygen and nutrients to the tumour cells; apoptosis is favoured in these conditions;

f) reduced systemic toxicity: toxic effects, such as myelosuppression, gastrointestinal effects and temporary hair loss, which are almost invariably present with traditional chemotherapy, are not observed with antiangiogenic therapy.

A number of molecular elements are known to be involved in tumour angiogenesis (Oncology 1997, 54, 177-84). Pro- and anti-angiogenic endogenous factors are known to be involved in the biological regulatory mechanism in the formation of new vessels.

Among the angiogenic stimulators we should mention: fibroblast growth factors (FGF), vascular endothelial growth factor (VEGF), angiogenin, transforming growth factor-α, tumour necrosis factor-α (TNF-α), platelet-derived endothelial cell growth factor, transforming growth factor-β, an in-vitro inhibitor, but an in-vivo stimulator, placental growth factor, interleukin-8, hepatocyte growth factor, platelet-derived growth factor, granulocyte colony-stimulating factors, proliferin, the prostaglandins (PGE ₁, PGE₂), GM1-GT1b, substance P, the bradykinins, and nitric oxide.

In contrast, the angiogenesis inhibitors include: the soluble receptor of bFGF, the interferons (α, β, γ), angiostatin, thrombospondin 1, prolactin (16 kDa terminal amino fragment), platelet factor 4 (PF4), the tissue metalloproteinase (TIMP) inhibitors, placental proliferin-related peptide, glioma-derived angiogenesis inhibition factor, the angiostatic steroids, cartilage-derived inhibitor (CDI), the heparinases, interleukin-12, plasminogen activator inhibitor, the retinoids, endostatin, angiopoietin-2, genistein, nitric oxide and GM3.

The integrins are a vast family of transmembrane proteins that mediate cell-to-cell and cell-to-extracellular matrix interactions. All integrins are capable of recognising a common peptide sequence Arg-Gly-Asp (“universal cell recognition site”), though every integrin preferentially recognises a different conformation of this tripeptide. The inhibition of specific subtypes of integrins can also be of great interest from the pharmacological standpoint for the development of angiogenesis inhibitors.

The control of protein kinase-C (PK-C) may also allow regulation of angiogenesis. There are, in fact, classic PK-C inhibitors capable of completely or partially blocking angiogenesis.

Despite the enormous investments and the involvement of large numbers of institutional and private research centres, the cancer problem world-wide is still far from being definitively solved. Though the prognosis of cancer victims has improved, with survival rates rising over the past 30 years on average from 30 to 50%, and the genetic, cellular and biochemical mechanisms involved in the development of a tumour are now well known, the possibility of defeating or at least limiting this type of pathology is still a problem of keenly felt concern and many aspects are still unsolved, such as the likelihood of recurrence, complete remission and metastatic spread of the primary tumour.

Since the late ‘seventies, when Folkman's observations began to be confirmed by the international scientific community, hundreds of molecules endowed with antiangiogenic activity have been isolated from natural sources (plants, fungi, biological fluids) and synthesised in the laboratory.

A number of drugs are already in Phase III, such as Marimastat (British Biotech.), I'AG3340 (Prinomast-Agouron), and Neovastat (Aeterna), all of which act mainly at the pulmonary level (SNCL) with a mechanism involving interference with the metalloproteinases. Also in Phase III are RhuMad VEGF (this is an anti-VEGF antibody by Genetech) and interferon a (commercial), which are active against solid tumours thanks to their interference with pro-angiogenic growth factors, or TNP-470 (TAP Pharm.) which acts directly on the endothelial cells. Lastly, drugs such as CAI (NCI) and IM862 (Cytran) are active as antiangiogenic agents but with a non-specific and poorly known mechanism.

These above-mentioned drugs may, in a few years, become part of the oncologist's therapeutic armamentarium, but there are other molecules which have recently been inserted in Phase I/II clinical trials such as Combretastatin (OxiGene), methoxy-oestradiol and endostatin (EntreMed), which on the basis of preclinical studies are very promising. Companies such as Bristol Myers-Squibb (with BMS-275291), Novartis (with CGS27023A) or Parke-Davis (with Suramin) are involved in this therapeutic strategy.

Heparin

Heparin is a heterogeneous mixture of naturally occurring polysaccharides of various lengths and various degrees of sulphation which possesses anticoagulant activity and is secreted by the connective tissue mastcells present in the liver (from which it was first isolated), in the muscles, lungs, thymus and spleen.

In addition to the regular sequence, a sequence corresponding to the active site for antithrombin activity has been identified in heparin.

The antitumour and antimetastatic activity of heparin and its derivatives is due to its ability to inhibit heparanase, to block growth factors and to regulate angiogenesis.

Heparan Sulphates (HS)

Heparan sulphates (HS) are ubiquitous protein ligands. The proteins bind to the HS chains for a variety of actions from simple immobilisation or protection against the proteolytic degradation action to specific modulations of biological activities, such as angiogenesis.

The carbohydrate skeleton, in both heparin and the heparan sulphates (HS), consists in an alternation of D-glucosamine (GlcN) and hexuronic acids (Glc.A or IdO.A).

In heparin, the GlcN residues are mainly N-sulphated, whereas in HS they are both N-sulphated and N-acetylated, with a small amount of unsubstituted N-.

HS is also on average less O-sulphated than heparin.

The use of heparin in the treatment of angiogenesis disorders, such as tumours, particularly metastases, is substantially limited by the anticoagulant activity of heparin.

Chemical modifications have been made to heparin so as to reduce its anticoagulant capacity, at the same time preserving its antitumour properties.

The opening of a unit of glucuronic acid in the antithrombin site reduces the affinity of heparin for antithrombin: in this way, heparins can be used with reduced anticoagulant effects, but still retaining antiangiogenic properties.

Heparanases

“Heparanases” are enzymes belonging to a family of endoglycosidases that hydrolyse the internal glycoside bonds of the chains of heparan sulphates (HS) and heparin.

These endoglycosidases are involved in the proliferation of tumour cells, in metastases and in the neovascularisation of tumours. This suggests they may also be involved in tumour angiogenesis as a result of the release, from the extracellular matrix, of growth factors bound to heparin, such as aFGF (also called FGF-1), bFGF (also called FGF-2) and VEGF.

These enzymes are biological targets for antiangiogenic activity. In the scientific literature there are a large number of structure/activity correlation studies (see, for example, Lapierre F. et al., Glycobiology, vol.6, (3), 355-366, 1996). Though many aspects have still to be clarified, studies have been reported regarding the inhibition of heparanases by heparin and its derivatives, using specific tests which have led to the emergence of considerations of a structural type which may serve as guides for obtaining new, more selective derivatives.

In the above mentioned work of Lapierre et al., heparin derivatives are described obtained by 2-O desulphation or by “glycol split” (oxidation with periodate and subsequent reduction with sodium borohydride). These derivatives, defined here as “2-O-desulphated heparin” and “RO-heparin”, respectively, have partly maintained the antiangiogenic activity of heparin as assessed by means of the CAM test in the presence of corticosteroids, as reported in Table III (ibid. page 360).

Heparins and FGF

FGFs regulate multiple physiological processes such as cell growth and differentiation, but also functions involved in pathological processes such as tumour angiogenesis.

FGFs are growth factors (a family of more than 10 polypeptides, of which the acid (FGF-1) and basic FGFs (FGF-2) are the ones which have been most studied, which require a polysaccharide cofactor, heparin or HS, to bind to the FGF receptor (FGFR) and activate it.

Though the precise mechanism whereby heparin and HS activate FGFs is unknown, it is known, however, that heparin/FGF/FGFR form a “trimolecular” or “ternary” complex.

One mechanism postulated is that heparin and HS induce so-called sandwich dimerisation of FGF, and the latter, thus dimerised, forms a stable complex with FGFR.

Antimetastatic Activity of Heparin Derivatives

The ability of a primary tumour to generate metastatic cells is perhaps the main problem facing anticancer therapy.

Heparin derivatives with a substantial ability to block heparanase seem to be equally capable of inhibiting angiogenesis both in primary tumours and in metastases.

In addition, the inhibition of heparanase reduces the migration ability of tumour cells from the primary tumour to other organs.

The following table gives an example of structure/antimetastatic activity correlation in the case of heparin:



	% inhibition

	Heparin	97
	N-succinyl-heparin	60
	N-succinyl-RO-heparin	58
	Low MW heparin	86
	Low MW N-succinyl-heparin	61
	Very low MW heparin	83

The data in this table suggest that very short fragments of heparin are still endowed with good antimetastatic activity, while this activity is reduced when the amine group of the glucosamine is bound to succinic acid.

The structural requisites of heparin-like molecules that favour the angiogenesis-inhibiting action can be grouped in two categories on the basis of the target one intends to block:

a) blockade of heparanase, an enzyme that hydrolyses the glycoside bonds of the heparan sulphates, releasing growth factors.

To this end the heparin-like compounds preferably comprise sequences of at least eight monosaccharide units containing N-acetyl-glucosamine-glucuronic acid (or N-sulphated glucosamine (see, for example, D. Sandback-Pikas et al. J. Biol. Chem., 273, 18777-18780 (1998) and references cited).

b) blockade of angiogenic growth factors (fibroblast type: FGF-1 and FGF-2; vascular endothelium type: VEGF; vascular permeability type: VPF).

To this end the heparin-like compounds preferably have sequences at least five monosaccharide units long, containing 2-sulphated iduronic acid and glucosamine N,6-sulphated (see, for example, M. Maccarana et al. J. Biol. Chem., 268, 23989-23905 (1993)).

In the literature small peptides (5-13 amino acids) with antiangiogenic activity (U.S. Pat. No. 5,399,667 of the University of Washington) which act by binding to a thrombospondin receptor, or longer peptides (50 amino acids approx.).

Modified platelet factors are known—(EP 0 589 719, Lilly), capable of inhibiting endothelial proliferation, with IC ₅₀=7 nM.

Oligosaccharide fragments with antiangiogenic activity have also been amply described: it has been found, in fact, that by varying the carbohydrate sequence the interaction selectivity can be increased.

In addition, heparin can be used as a vehicle for substances which are themselves antiangiogenic, such as some steroids, exploiting the affinity of heparin for vascular endothelial cells; see, for example, WO 93/18793 of the University of Texas and Imperial Cancer Research Technology, where heparins are claimed with acid-labile linkers, such as adipic acid hydrazine, bound to cortisol. The antiangiogenic effect of the conjugated molecules is greater than that of the unconjugated molecules, even when administered simultaneously.

In Biochim. Biophys. Acta (1996), 1310, 86-96, heparins bound to steroids (e.g. cortisol) are described with a hydrazone group in C-20 that present greater antiangiogenic activity than the two unconconjugated molecules.

EP 0 246 654 by Daiichi Sc. describes sulphated polysaccharides with antiangiogenic activity with Phase II studies.

EP 0 394 971 by Pharmacia & Upjohn—Harvard Coll. describes hexa-saccharides—heparin fragments—with low sulphation, capable of inhibiting the growth of endothelial cells and angiogenesis stimulated by (FGF-1).

EP 0 618 234 by Alfa Wasserman describes a method for preparing semisynthetic glycosaminoglycans with a heparin or heparan structure bearing a nucleophilic group.

WO 95/05182 by Glycomed describes various sulphated oligosaccharides with anticoagulant, antiangiogenic and anti-inflammatory activity.

U.S. Pat. No. 5,808,021 by Glycomed describes a method for preparing substantially non-depolymerised 2-O, 3-O desulphated heparin with a desulphation percentage in positions 2- of the iduronic acid (I, 2-O) and in position 3 of the glucosamine unit (A, 3-O) ranging from approximately 99 to approximately 75% of the original percentage. This method envisages desulphation conducted in the presence of a cation of a bivalent metal, exemplified by calcium or copper, followed by lyophilisation of the product obtained. The desulphated heparins have antiangiogenic activity.

The aim of the invention described herein is to find optimal glycosaminoglycan structures for generating antiangiogenic activity based on heparanase inhibition and/or FGF growth factor inhibition mechanisms. An additional aim of the invention described herein is to provide a medicament with antiangiogenic activity which is essentially devoid of the typical side effects of heparin derivatives, such as, for example, anticoagulant activity.

ABSTRACT OF THE INVENTION

It has now been found that on subjecting a glycosaminoglycan, such as a heparin-like glycosaminoglycan, heparin or modified heparin, containing glucosamine residues with different degrees of N-desulphation and optional subsequent total or partial N-acetylation, to controlled 2-O-desulphation treatment of the iduronic units up to a degree of desulphation not greater than 60% of the total uronic units, the angiogenic growth factor binding properties are maintained. Surprisingly, heparin 2-O-desulphated to not more than 60% of its total uronic units is markedly antiangiogenic.

The desulphation conducted in the conditions described in the present invention also produces the formation of iduronic units with an oxyranic ring in position 2,3. The opening of the oxyranic ring in the conditions described in the present invention gives rise to L-iduronic or L-galacturonic units.

It is an object of the invention described herein a glycosaminoglycan derivative, particularly desulphated heparin, selectively partially desulphated with a desulphation degree not exceeding 60% of the total uronic units; these desulphation gaps reduce the length of the regular sequences constituted by the succession of disaccharide trisulphate units.

In one particular embodiment, the invention described herein refers to a formula (I) compound

where the U ring may have the following meanings:

X and X′, which can be the same or different, are an aldehyde group or the —CH ₂-D group, where D is hydroxyl or an amino acid, a peptide or a residue of a carbohydrate or oligosaccharide;

R and R ₁, which can be the same or different, are an SO₃or acetyl residue;

n and m, which can be the same or different, may vary from 1 to 40; the sum of n+m ranges from 6 to 40; the m:n ratio ranges from 10:2 to 1:1. Preferably, m is greater than or equal to n. Preferably n ranges from 40 to 60% of the sum m+n. The symbol

indicates that the units marked m and n are statistically distributed along the polysaccharide chain and are not necessarily in sequence.

The compounds which are the subject matter of the invention described herein, have interesting antiangiogenic properties, and are therefore useful as active ingredients for the preparation of medicaments for the treatment of pathologies based on abnormal angiogenesis, and particularly for the treatment of metastases.

Advantageously, the compounds according to the invention present reduced, if not non-existent anticoagulant properties, thus avoiding or reducing the side effects typical of the heparins. A further advantage stems from the fact that the compounds according to the invention can be characterised with instrumental analytical techniques, such as NMR spectroscopy, thus allowing process control which is absolutely desirable from the industrial point of view.

Also in the case of modified heparins, molecular weight (MW) has a very important function when making angiogenesis inhibitors. It is well known, in fact, that a reduction in molecular weight (MW) to values corresponding to penta-saccharide units does not lead to a loss of antiangiogenic activity. On the contrary, it has been established that, beyond a certain length, the heparin chains favour rather than inhibit dimerisation and thus activation of FGF.

DETAILED DESCRIPTION OF THE INVENTION

What is meant by desulphation degree is the percentage of non-sulphated iduronic acids in relation to total uronic acids originally present in the starting heparin. One initial preferred range for the desulphation percentage is from approximately 40 to approximately 60%. Among the formula (I) compounds the preferred compound is:

heparin partially 2-O-desulphated with a molecular weight (MW) of 11200, a polydispersion index D of 1.3, a desulphation degree of 1.99 (expressed as the SO ₃ ⁻:COO⁻ molar ratio), a percentage of modified uronic acids compared to total uronic acids of approximately 50% (hereinafter also called ST1514). Said compound is comprised in formula (I) where, among the other corresponding definitions, m:n=1:1 and the units marked m and n are distributed along the polysaccharide chain in a regular, alternating manner.

The compounds according to the invention described herein are prepared by means of a process comprising:

a) basic treatment at a temperature ranging from ambient temperature to approximately 100° C., preferably from 50 to 70° C., and even more preferably at approximately 65° C., which leads to the elimination of a controlled percentage of sulphate groups in position 2 of the iduronic acid and to the formation of epoxide groups; and, if desired

b) opening of said epoxide ring at approximately pH 7, at a temperature ranging from approximately 50° C. to approximately 100° C., preferably at approximately 70° C., to yield residues of galacturonic acid; or, alternatively

c) opening of said epoxide ring at a temperature ranging from approximately 0° C. to 30° C., preferably at approximately 25° C., to yield residues of iduronic acid; and, if desired

d) oxidation of the diols with sodium periodate, to yield the opening of the glycoside ring and the formation of two aldehyde groups per modified residue;

e) reduction of said aldehyde groups to primary alcohol and, if desired, transformation of the D group to a group other than hydroxyl, as envisaged in the different meanings assigned in formula (I)

f) optional acid hydrolysis to obtain oligosaccharides corresponding to the regular sequences, or alternatively

g) partial enzymatic hydrolysis with an enzyme selected from the group consisting of lyase, heparinase, heparitinase, or equivalent of products obtained in step e) to yield oligosaccharides, preferably tetra- or octa-saccharides, with the non-reducing terminal residue consisting of unsaturated iduronic acid, the reducing residue consisting of an N-sulphoglucosamine and containing at least one residue of open iduronic acid; or, alternatively

h) the compound obtained in step a), or the product obtained in step b) is treated by partial enzyme hydrolysis; and, if desired

i) subjection of the products obtained in one of steps a), b) and e) to partial 6-O-desulphation; or, alternatively,

j) subjection of the starting heparin partially or totally 6-desulphated to steps a), b) and e).

According to the invention described herein, the preferred compound is:

heparin partially 2-O-desulphated, obtainable by the process described above, where step a) is conducted for 45 min at 60° C. and step b) at 70° C. at pH 7, and having a molecular weight (MW) of 11200, a polydispersion index D of 1.3, a desulphation degree of 1.99 (expressed as the SO ₃ ⁻:COO⁻ molar ratio), percentage of modified uronic acids compared to total uronic acids of approximately 50% (hereinafter also called ST1514);

The molecular weights are determined by HPLC-GPC (high performance liquid chromatography—gel permeation chromatography). The desulphation degree is determined by conductimetry and the percentage of modified uronic acids by ¹³C-NMR.

MW is the molecular weight, and D is the polydispersion index expressed as MW/Mn.

According to the invention described herein, the starting products are glycosaminoglycans of various origins, preferably naturally occurring heparins. It is also possible to use chemically modified heparins with a percentage content of N,6 disulphate ranging from 0 to 100%. Starting from products with a different 6-O-sulphated glucosamine content, it is possible to modulate the length of the regular sequences between one open iduronic acid and another. The glycosaminoglycans according to the invention that present opening of the glycoside ring are conventionally called RO derivatives by those skilled in the field, meaning by this that the glycoside ring has been opened by means of an oxidation action, followed by a reduction (Reduction-Oxidation—RO). This opening of the glycoside ring is also conventionally called “glycol split”, so-called because of the formation of the two primary hydroxy present on the open ring. The compounds referred to herein will also be called “RO” or “Glycol Split” derivatives.

In a further embodiment of the invention described herein, the aldehydes and primary hydroxy derived from the opening reaction described above (“glycol split”) also lend themselves to the subsequent functionalisation. Therefore, formula (I) compounds may also bear equal or different groups, as defined above for X and X′, on the primary hydroxy deriving from glycol split, for example, oligosaccharide or peptide groups, ranging from a single saccharide or amino acid to more than one unit of length, preferably 2 or 3 units.

Formula (I) compounds where X and X′ are —CH ₂OH can also be used as vehicles for other types of drugs, by means of suitable binding with the heparin portion which is capable of providing a stable bond in normal conditions of manufacture and storage of a formulated drug. which, however, releases the transported drug in the body, preferably in the vicinity of the target organ. Examples of drugs that can be transported are steroidal and non-steroidal anti-inflammatory drugs, corticosteroids, and other drugs with an antimetastatic action, in which case there will be an advantageous enhancement of the antimetastatic effect as a result of the sum of the separate intrinsic activities of the compounds according to the invention and the antimetastatic agent bound thereto, with the related advantages of a greater target selectivity and lower systemic toxicity. Examples of these drugs are the metalloproteinase inhibitors. Other drugs which can be usefully transported are those that act at the endothelial level. Formula (I) compounds where X and X′ are other than hydroxy or aldehyde can also be used as vehicles for drugs, in which case the X and X′ groups will act as “spacers” between the transported molecule, that is to say the glycosaminoglycan of the present invention and the molecule acting as the vehicle, in those cases where this may be desirable for reasons of pharmacokinetics or pharmacodynamics.

In the case of compounds according to the invention deriving from heparin, these are prepared starting from heparin itself by means of desulphation techniques known to the technical experts in the field. For example, the desulphation is conducted in the presence of alkaline agents, such as sodium hydroxide, at temperatures ranging from ambient temperature to 100° C., preferably from 50 to 70° C., for example at 60° C., for a sufficiently long period to obtain the desired desulphation. The desulphation is controlled by acting on the process parameters, such as the concentrations of reactants, the temperature and the reaction times. One preferred example consists in maintaining constant concentrations of substrate (glycosaminoglycan) at 80 mg/ml and of NaOH at 1 M, a constant temperature of 60° C. and controlling the desulphation with a reaction time from 15 to 60 min. The expert in the field may vary the conditions, for example by raising the reaction temperature and shortening the reaction time, on the basis of normal trial and error in experimental practice and on the basis of his or her general knowledge of the subject.

The treatment with alkaline agents gives rise to an intermediate product characterised by the presence of an epoxide ring on the desulphated unit. In a thoroughly surprising manner, these intermediates have proved to be endowed with antiangiogenic properties similar to those of the formula (I) compounds. Therefore, an further object of the invention described herein is a derivative of partially desulphated heparin, and therefore heparin with a reduced charge, particularly heparin not desulphated more than 60%, characterised by an epoxide ring on the desulphation site. Said compounds characterised by an epoxide ring also belong to the objects covered by the present invention, that is to say the pharmaceutical compositions that contain them and their use for the preparation of medicaments with antiangiogenic activity.

The following compounds are preferred:

heparin partially 2-O-desulphated with a molecular weight (MW) of 12900 D, a polydispersion index D of 1.5, a desulphation degree of 2.05 (expressed as the SO ₃ ⁻:COO⁻ molar ratio), percentage of modified uronic acids compared to total uronic acids: 5% epoxide groups, 29% oxidated and reduced uronic residues (hereinafter also called ST1513);

heparin partially 2-O-desulphated with a molecular weight (MW) of 11000 D, a polydispersion index D of 1.5, a desulphation degree of 2.05 (expressed as the SO ₃ ⁻:COO⁻ molar ratio), percentage of modified uronic acids compared to total uronic acids: 5% epoxide groups, 29% uronic residues oxidated and reduced uronic residues;

heparin partially 2-O-desulphated with a molecular weight (MW) of 9200 D, a polydispersion index D of 1.5, percentage of modified uronic acids compared to total uronic acids: 11% epoxide groups, 27.5% oxidated and reduced uronic residues.

In one of the possible embodiments of the invention described herein, the following are preferred:

heparin partially 2-O-desulphated, obtainable by the process described above, where step a) is conducted for 15 min at 60° C. and step b) at 70° C. at pH 7, and with a molecular weight (MW) of 12900 D, a polydispersion index D of 1.5, a desulphation degree of 2.05 (expressed as the SO ₃ ⁻:COO⁻ molar ratio), percentage of modified uronic acids compared to total uronic acids: 5% epoxide groups, 29% oxidated and reduced uronic residues (hereinafter called ST1513);

heparin partially 2-O-desulphated, obtainable by the process described above, where step a) is conducted for 30 min at 60° C. and step b) at 70° C. at pH 7 (hereinafter called ST1516), and with a molecular weight (MW) of 11000 D, a polydispersion index D of 1.5, a desulphation degree of 1.8 (expressed as the SO ₃ ⁻:COO⁻ molar ratio), percentage of modified uronic acids compared to total uronic acids. 5% epoxide groups, 29% oxidated and reduced uronic residues;

heparin partially 2-O-desulphated, obtainable by the process described above, where step a) is conducted for 60 min at 60° C. and step b) at 70° C. at pH 7, and with a molecular weight (MW) of 9200 D, a polydispersion index D of 1.5, percentage of modified uronic acids compared to total uronic acids: 11% epoxide groups, 27.5% oxidated and reduced (split) uronic residues (hereinafter called ST1515).

Subsequent to the formation of the epoxide ring, the latter is opened, again resorting to known techniques. The percentage of epoxide formed is calculated from the ratio between the areas of the ¹³C-NMR signals at approximately 55 ppm, characteristic of carbons 2 and 3 of the uronic acid ring containing the epoxide and the total number of anomeric signals (C1 of the glucosamine and uronic acid residues). If the opening is conducted hot, a galacturonic acid residue is obtained, whereas, if the opening of the epoxide ring is conducted cold, an iduronic acid residue is obtained. Preferred examples of compounds containing an epoxide ring are those obtainable by the process described above and having epoxidated uronic acid contents of 14% (hereinafter ST1509), 24% (hereinafter ST1525) and 30% (hereinafter ST1526), respectively.

The partially desulphated heparin is then subjected to “glycol-split” (RO for short), according to the process defined above and Smith degradation (SD for short).

Alternatively, formula (I) compounds can also be obtained without passing via the epoxide intermediate, that is to say by direct glycol split and subsequent Smith degradation.

The process described so far leads to formula (I) compounds in which the X and X′ groups are both —CH ₂OH.

For X and X′ other than —CH ₂OH, methods are available to the expert in the field for transforming the hydroxyl group with other groups envisaged in the definitions given above. For example, the conjugation with amino acids or peptides can be done by treating the intermediate aldehyde derived from the glycol-split reaction with a reductive amination reaction (Hoffmann J. et al. Carbohydrate Research, 117, 328-331 (1983)), which can be conducted in aqueous solvent and is compatible with maintenance of the heparin structure.

If desired, and this constitutes a further object of the invention described herein, the formula (I) compounds can be further degraded with acid agents in suitable pH conditions, e.g. at pH 4, to yield a mixture of oligosaccharides that maintain the antiangiogenic properties.

In the same way, objects of the present invention are the compounds obtained by one of the steps g), h), i) and j) of the process described above.

Objects of the invention described herein are pharmaceutical compositions containing as their active ingredient at least one formula (I) compound, alone or in combination with one or more formula (I) compounds, or, said formula (I) compound or compounds in combination with the desulphated heparins described above, e.g. the epoxidated intermediates; the latter can also be used alone as active ingredients in the pharmaceutical compositions. The active ingredient according to the present invention will be in a mixture with suitable vehicles and/or excipients commonly used in pharmaceutical technology, such as, for instance, those described in “Remington's Pharmaceutical Sciences Handbook”, latest edition. The compositions according to the present invention will contain a therapeutically effective quantity of the active ingredient. The doses will be determined by the expert in the field, e.g. the clinician or primary care physician according to the type of disease to be treated and the patient's condition, or concomitantly with the administration of other active ingredients. By way of an example, doses ranging from 0.1 to 100 mg/kg may be indicated.

Examples of pharmaceutical compositions are those that can be administered orally or parenterally, intravenously, intramuscularly, subcutaneously, transdermally or in the form of nasal or oral sprays. Pharmaceutical compositions suitable for the purpose are tablets, hard or soft capsules, powders, solutions, suspensions, syrups, and solid forms for extemporary liquid preparations. Compositions for parenteral administration are, for example, all the intramuscular, intravenous and subcutaneous injectable forms as well as solutions, suspensions and emulsions. Liposome formulations should also be mentioned. The tablets also include forms for the controlled release of the active ingredient whether as oral administration forms, tablets coated with suitable layers, microencapsulated powders, complexes with cyclodextrins, depot forms, for example, subcutaneous forms, such as depot injections or implants.

The compounds according to the invention described herein possess antiangiogenic activity. This makes them suitable for the preparation of medicaments useful for the treatment of objects, generally mammals, and particularly human subjects, suffering from altered angiogenesis. Examples of diseases treated with the medicament which is the object of the present invention are primary tumours, metastases, diabetic retinopathies, psoriasis, retrolenticular fibroplasia, restenosis after angioplasty and coronary by-pass.

Advantageously, the compounds according to the present invention are substantially devoid of the side effects typical of heparin. In particular, the compounds according to the invention are substantially devoid of anticoagulant activity. By substantially devoid of such activity the expert in the field means no or only negligible activity from the point of view of clinical use.

One of the first growth factors found to have an angiogenic role was bFGF (or FGF-2) followed shortly afterwards by aFGF (or FGF-1) (for a review of the subject see Christofori, Oxford University, 1996). Both proteins are members of a class of growth factors characterised by a high degree of affinity for heparin. Other potent angiogenic inductors are VEGF, VEGF-B and VEGF-C. All three VEGF factors, like the FGFs, are expressed ubiquitously in the body. Both of these types of factors, a/bFGF (FGF-1 and FGF-2) and VEGF, bind to specific high-affinity receptors with a transmembrane domain possessing tyrosine-kinase activity. The three receptors for the VEGFs—VEGF-1 (fit-1), VEGF-2 (kdr/flk-1) and VEGF-3 (flt-4)—are expressed specifically on endothelial cells, whereas the four FGF receptors are expressed in numerous organs and tissues (for a review of the subject see Risau and Flame 1995 Annu. Rev. Cell Dev: Biol. 11: 73-91). bFGF binding to heparin or to fragments of heparan sulphate causes their dimerisation and the possibility of binding to their own receptor by activating the transduction pathway of the signal that activates the endothelial cell in both mitogenesis and differentiation.

The inhibition of FGF binding to heparin or to fragments of heparan sulphate thus represents a valid therapeutic target in combating the pathological neoangiogenesis induced by an increased local or systemic level of growth factors.

To this end an in-vitro model was designed capable of assessing the interference of the various different heparin derivatives with bFGF binding to its own receptor.

In particular, ovarian cells of Chinese hamsters (CHO-K1) were used which possess heparan sulphates on their surface, but lack FGF receptor 1. CHO cells called CHO-745flg were also used, which express the FGF receptor 1 but which, unlike the former group, do not possess membrane heparan sulphates. These latter cells were stably transfected with cDNA for green fluorescent protein and for this reason could be observed under the fluorescence microscope with a combination of filters for the detection of green emission. In brief, the technique is based on the possibility that the two types of cells may interact (form stable bonds) only if FGF is added to the two cell types. In fact, in this case, the heparan sulphates of the CHO-745flg cells will bind to FGF, which in turn will bind to the CHO-K1 cells, establishing a bridge. The binding that occurs is detectable as a result of the green fluorescence emitted by the CHO-745flg cells.

In addition, the compounds were assayed for their cell proliferation inhibiting activity. In fact, one of the main effects of FGF is to stimulate cell growth in the absence of serum in cells expressing FGF receptor 1 (FGFR-1).

To this end, two cell lines were used, both expressing FGFR-1 (L6WT1 and bovine aorta cells), the growth of which was evaluated in the presence of FGF with or without the addition of heparin derivatives for which an inhibitory activity was expected.

Inhibition of FGF2-Mediated Intercellular Adhesion

CHO-K1 cells were seeded at a density of 90,000 cells/cm ²in 24-well plates. After 24 hours the cells were fixed in 3% glutaraldehyde in PBS for 2 hours at 4° C. and washed with 0.1 M glycine/PBS. On the fixed monolayers CHO 745/flg cells, transfected with FGFR-1, at a density of 50,000 cells/cm²in DMEM containing 10 mM EDTA, 30 ng/ml FGF-2 and increasing doses of the test compounds. After 2 hours' incubation at 37° C., the adhering cells were counted under the inverted microscope. The results were expressed as percentage of the number of cells adhering compared to those measured in the absence of the test compounds. The compounds were examined in triplicate at 6 concentrations ranging from 1 ng/ml to 100 μg/ml and the ID₅₀was calculated for each compound (Table 1).

TABLE 1


Inhibition of intercellular adhesion (ID₅₀) of CHO-K1
e CHO-745flg cells

	Product	Inhibition (ID₅₀)

	Heparin	100
	RO heparin	8
	ST1513	80
	ST1516	110
	ST1514	105
	ST1515	120
	ST1525	50
	ST1528	75
	ST1507	125

Inhibition of DNA Synthesis in L6 Cells Transfected with FGFR-1

L6-WT1 cells (rat myoblasts transfected with FGFR-1) were seeded at a density of 25,000 cells/cm ²in 48-well plates in DMEM+ 10% FCS. After 24 hours the cells were washed with serum-free medium and incubated for 48 hours with DMEM+ 0.5% FCS. The cells were then incubated for 16 hours with FGF-2 at the concentrations of 15 and 30 ng/ml in the presence or absence of the test compounds (all at 100 μg/ml). At the end of the incubation 3H-thymidine (0.25 μCi/well) was added without changing the medium. After 6 hours, precipitable TCA was measured. Each experimental point is the mean of 3 determinations. The results are given in Table 2.

TABLE 2


Inhibition of DNA synthesis in L6 cells transfected with FGFR-1.

	Incorporation of 3H-
	thymidine

Product	(% vs. control)

Heparin (FGF-2 15 ng/ml)	50
Heparin (FGF-2 30 ng/ml)	74
Heparin RO (FGF-2 15 ng/ml)	41
Heparin RO (FGF-2 30 ng/ml)	80
ST1513 (FGF-2 15 ng/ml)	93
ST1513 (FGF-2 30 ng/ml)	102
G3025B (FGF-2 15 ng/ml)	68
G3025B (FGF-2 30 ng/ml)	71
ST1514 (FGF-2 15 ng/ml)	58
ST1514 (FGF-2 30 ng/ml)	99
ST1515 (FGF-2 15 ng/ml)	79
ST1515 (FGF-2 30 ng/ml)	100

Effect on DNA Syntheses in Bovine Aorta Endothelial Cells

BAEC cells were seeded at a density of 2,500 cells/cm ²in 48-well plates in EGM Bullet Kit complete medium. After 24 hours the cells were cultivated in the absence of serum in EGM medium without bovine brain extract and hFGF. After a further 24 hours, the cells were treated with 30 ng/ml of bFGF in the presence of increasing concentrations of the test compound. After 16 hours, 1 μCi/ml of 3H-thymidine was added to the medium. After 6 hours, precipitable TCA activity was measured. Each experimental point is the mean of 8 determinations. The results are given in Tables 3-6.

TABLE 3


Effect on DNA synthesis in bovine aorta endothelial cells
(BAEC) - ST1514

Incorporation of ³H-

	Concentration	thymidine
	ng/ml	(% vs control)

	1	80
	10	78
	100	42
	1000	33
	10000	22
	100000	5

TABLE 4


Effect on DNA synthesis in bovine aorta endothelial cells
(BAEC) - ST1528

Incorporation of ³H-

	Concentration	thymidine
	ng/ml	(% vs control)

	1	85
	10	73
	100	85
	1000	32
	10000	28
	100000	5

TABLE 5


Effect on DNA synthesis in bovine aorta endothelial cells
(BAEC) - ST1525

Incorporation of ³H-

	Concentration	thymidine
	ng/ml	(% vs control)

	1	75
	10	67
	100	26
	1000	21
	10000	7
	100000	2

TABLE 6


Effect on DNA synthesis in bovine aorta endothelial cells
(BAEC) - ST1507

Incorporation of ³H-

	Concentration	thymidine
	ng/ml	(% vs control)

	1	92
	10	92
	100	45
	1000	40
	10000	18
	100000	12

Test of Cell Proliferation in BAEC

BAEC cells at the 7th pass were seeded at a density of 2,500 cells/cm ²in 96-well plates in EBM medium without bovine brain extract and hEGF. To this medium were added 30 ng/ml of FGF-2 and each of the test compounds at 5 concentrations ranging from 10 ng/ml to 100 μg/ml. After 3 days the cells were fixed and stained with crystal-violet and the optical density was determined by means of an ELISA microplate reader. Each experimental point was done in quadruplicate and the ID₅₀was calculated. The results are given in Table 7.

TABLE 7


Test of cell proliferation in BAEC

Inhibition (ID₅₀)

	Product	μg/ml

Heparin

	100
	ST1509	0.02
	ST1525	10
	ST1526	0.02
	ST1527	100
	ST1528	0.1

Test of Cell Proliferation in Foetal BAEC GM 7373

GM7373 cells were seeded at a density of 70,000 cells/cm ²in 96-well plates in MEM+ 10% FCS medium. After 24 hours, the cells were washed with serum-free medium and treated with 10 ng/ml of FGF-2 in medium containing 0.4% FCS. After 8 hours, the test compounds were added to the medium at 5 concentrations ranging from 10 ng/ml to 100 μg/ml. After a further 16 hours, the cells were trypsinised and counted in a Burker chamber. The ID₅₀was calculated for each compound and the results are the means of 2 experiments in triplicate. The results are given in Table 8.

TABLE 8


Test of cell proliferation in foetal BAEC GM 7373

Inhibition (ID₅₀)

	Product	μg/ml

	Heparin	2
	ST1509	0.3
	ST1525	2
	ST1526	1
	ST1527	20
	ST1528	0.2
	ST1514	0.1

Inhibition of Angiogenesis

For the study of angiogenesis the model used was the chick embryo chorioallantoic membrane (CAM) model (Ribatti e al. Int. J. Dev. Biol., 40, 1189 (1996)).

300 embryonated hens' eggs were incubated at 37° C. in constant humidity conditions. On the 3rd day of incubation, after aspirating 2-3 ml of albumen from the acute pole of the egg so as to detach the CAM from the shell, a window was cut with scissors in the shell. The vessels underlying the CAM were thus exposed. The window was closed with a transparent glass panel and the eggs were put back into the incubator. On the 8th day of incubation, in sterile conditions, the CAMs were treated according to the protocol described below using a technique developed recently (Ribatti et al. J. Vasc. Res. 34, 455 (1997)), which involves the use of sterile gelatin sponges measuring 1 mm ²in size. Each molecule was resuspended in 3 μl of PBS at a concentration of 50 or 100 μg/embryo. As a positive control, FGF-2 was used (1 μg/sponge), for which potent angiogenic activity on the CAM has been demonstrated (Ribatti e al., Dev. Biol. 170, 39, (1995)). The sponges were first rested on the surface of the CAM, and then 3 μl of solution of the test substance were pipetted onto the surface of the sponge. The CAMs were examined daily using a Zeiss SR stereomicroscope equipped with a photographic device. The experiments were discontinued on the 12th day of incubation, when an overall assessment of the angiogenic activity of the molecules was carried out, expressed as percentage inhibition in relation to the number of experiments initiated and completed (Table 9). In addition, for compound ST1514, assayed at the concentration of 100 μg/embryo, a macroscopic quantitative evaluation of the angiostatic activity was also carried out according to the technique proposed by Brooks et al. (Science, 264, 569, (1994)), counting the number of vessels surrounding the sponge. The number of vessels was compared with that of the control sponges treated with PBS (vehicle of the test compound) and with FGF-2 (positive control). The results are given in Table 10.

TABLE 9


Angiostatic activity on CAM model

Concentration

Inhibition

Compound	(μg/embryo)	N. experiments	(%)

N-acetylated	50	10	0
heparin
RO heparin (ox-	50	10	20
red 19.6%)
ST1514 (RO 56%)	50	10	50
ST1514 (RO 56%)	100	10	80

It should be noted that naturally occurring heparin has no activity.

TABLE 10


Angiostatic activity: macroscopic quantitative evaluation
(Brooks)

	N. vessels at
	sponge/CAM

Compound	N. embryos	interface

ST1514 (100 μg/embryo)	10	2 + 1
PBS (3 μl)	5	7 + 2
FGF-2 (1 μg/embryo)	5	50 + 4

It should be noted that naturally occurring heparin has no activity.

Evaluation of Heparin Toxicity in Balb/c Mice.

20 6-week-old female Balb/c mice weighing 20 g (Harlan), divided into groups by casual randomisation, were treated with heparin sodium, as a reference, and with the compound according to the present invention called ST1514. The treatment schedule was of the q2dx5 type, i.e. 5 total administrations at intervals of 2 days, administering 200 μl/mouse of 50 mg/kg/10 ml and 25 mg/kg/10 ml solutions subcutaneously.

The substances were solubilised as follows:

Heparin sodium: the solution is prepared by solubilising 160 mg of powder in 4 ml of Ca ⁺⁺- and Mg⁺⁺-free PBS 1× pH 7.4; it is subdivided into aliquots of 243 μl and stored at −20° C. At the time of treatment, the solution is diluted in Ca⁺⁺- and Mg⁺⁺-free PBS (Dulbecco, modified formula) 1×, pH 7.4 so as to have the substance at final concentrations of 50 mg/kg/10 ml and 25 mg/kg/10 ml.

ST1514: the solution is prepared by solubilising 160 mg of powder in 4 ml of Ca ⁺⁺- and Mg⁺⁺-free PBS 1×, pH 7.4; it is subdivided into aliquots of 243 μl and stored at −20° C. At the time of treatment, the solution is diluted in Ca⁺⁺- and Mg⁺⁺-free PBS (Dulbecco, modified formula) 1×, pH 7.4 so as to have the substance at final concentrations of 50 mg/kg/10 ml and 25 mg/kg/10 ml.

48 hours after the last administration of the test substances a blood sample is taken for full blood count and haematological analysis.

Blood sampling: the mice are placed in a hermetically sealed box, into which is insufflated CO ₂in an amount sufficient to stun the animal. Blood is then taken from the retro-orbital plexus of each animal (approximately 1 ml of blood/mouse) which is distributed in amounts of 0.4 ml of blood/mouse in Eppendorf test tubes containing 20 μl of Vister heparin (5000 U/ml) for carrying out the full blood count and haematological analysis; and the same amount of blood is distributed in other Eppendorf test tubes containing 50 μl of 3.8% sodium citrate in order to perform prothrombin time determinations. After taking the blood sample, each animal is sacrificed by cervical dislocation.

Number of samples: 2 blood samples/mouse are taken, with which two full blood counts are carried out and two slides and a plasma sample are produced to be stored at −20° C.

Full blood count: the appliance (Cell Analyzer 580 A, DELCON) is used according to the standard procedure described in the operating manual. The blood sample (25 μl) is taken by the dilutor and brought up to a volume of 10 ml (dil. 1:400) with isotonic solution (PLTA saline, DELCON). From this solution (called solution A), the dilutor automatically takes a sample of 100 μl and brings it up to a volume of 10 ml (dil. 1:100), thus obtaining solution B. To solution A are added 3 drops of Emosol (DELCON) for lysis of the red blood cells. This solution is used for the WBC reading. Solution B, on the other hand, is used for the RBC and platelet (PLT) readings. Each sample is read in duplicate.

The data obtained will be analysed using ANOVA.

The groups of animals treated with heparin sodium at the doses of 50 mg/kg/10 ml and 25 mg/kg/10 ml present a marked haematoma at the inoculum site; this phenomenon does not occur in the groups submitted to treatment with ST1514. Autopsies conducted in the study animals reveal that the groups submitted to treatment with heparin sodium at doses of 50 mg/kg/10 ml and 25 mg/kg/10 ml present livers with abnormal characteristics, whereas no such phenomenon is detected in the groups treated with ST1514.

The data relating to the haematological analysis show a reduction in red blood cells in both groups treated with heparin sodium at the doses of 50 mg/kg and 25 mg/kg as compared to the control group, whereas no such difference is detectable in the groups treated with ST1514.

The group treated with heparin sodium at the dose of 25 mg/kg presents a significant platelet deficiency compared to the control group. None of the study treatment groups presents significant differences in the numbers of white blood cells as compared to the control group.

The following examples further illustrate the invention.

EXAMPLE 1

1 g of heparin is dissolved in 12.5 ml of NaOH 1N. The solution is heated and stirred at 60° C. for 45 minutes. The reaction is blocked by rapid cooling and neutralisation. The solution is then heated at 70° C. at pH 7 until the epoxide ring is completely open (the reaction trend is controlled by NMR). The desulphated sample (here called G2999H) is dissolved in 20 ml of water and cooled to 4° C. After the addition of 20 ml of a solution of NaIO[0161] ₄0.2 M, the solution is left to stir in the dark for 20 hours, and the reaction is stopped by adding ethylene glycol and the salts are eliminated by tangential ultrafiltration. 400 mg of NaBH₄, subdivided into several portions, are added to the desalted solution (30-40 ml). The solution is left to stir for 3 hours at ambient temperature, then neutralised with diluted HCl and desalted by tangential ultrafiltration. 710 mg of product, here called ST1514, are obtained.
Molecular weight (MW): 11200 D, polydispersion index D 1.3, desulphation degree 1.9 (expressed as SO[0162] ₃—:COO— molar ratio); the percentage of modified uronic acids compared to total uronic acids is approximately 50%. The compound was characterised by means of NMR spectroscopy (FIG. 1).

EXAMPLES 2-4

Adopting the same procedure as in example 1, with the exception that the basic solution was heated for 15, 30 and 60 minutes, respectively, compounds with the following characteristics were obtained: [0163]
ST1513: molecular weight (MW) 12900 D, polydispersion index D 1.5, desulphation degree 2.05 (expressed as SO[0164] ₃—:COO— molar ratio), percentage of modified uronic acids compared to total uronic acids: 5% epoxide groups, 29% oxidated and reduced (split) uronic residues;
ST1516: molecular weight (MW) 12900 D, polydispersion index D 1.5, desulphation degree 1.8 (expressed as SO[0165] ₃ ⁻:COO⁻ molar ratio), percentage of modified uronic acids compared to total uronic acids: 5% epoxide groups, 29% oxidated and reduced (split) uronic residues;
ST1515: molecular weight (MW) 9200 D, polydispersion index D 1.5, percentage of modified uronic acids compared to total uronic acids: 11% epoxide groups, 27.5% oxidated and reduced (split) uronic residues. [0166]

Claims

1) A glycosaminoglycan derivative, particularly a desulphated heparin, with a desulphation degree not greater than 60% of the total uronic units.

2) A derivative according to claim 1, which is a heparin-like glycosaminoglycan.

3) A derivative according to claim 1, which is a modified heparin, containing glycosamine residues with different degrees of N-desulphation and optional subsequent total or partial N-acetylation.

4) A derivative according to claim 1, which has the following formula (I):

where the U ring can have the following meanings:

X and X′, which can be the same or different, are an aldehyde group or the —CH₂-D group, where D is hydroxy or an amino acid, a peptide or a residue of a carbohydrate or oligosaccharide;

R and R1, which can be the same or different, are an SO₃or acetyl residue;

n and m, which can be the same or different, may vary from 1 to 40; the sum of n+m ranges from 6 to 40; the m:n ratio ranges from 10:2 to 1:1, the symbol

5) A derivative according to claim 4, which is partially 2-O-desulphated heparin with a molecular weight (MW) of 11200, a polydispersion index D of 1.3, a desulphation degree of 1.99 (expressed as SO₃ ⁻:COO⁻ molar ratio), percentage of modified uronic acids compared to total uronic acids approximately 50%.

6) A derivative according to claim 4, which is partially 2-O-desulphated heparin with a molecular weight (MW) of 12900 D, a polydispersion index D of 1.5, a desulphation degree of 2.05 (expressed as SO₃ ⁻:COO⁻ molar ratio), percentage of modified uronic acids compared to total uronic acids: 5% epoxide groups, 29% oxidated and reduced uronic residues.

7) A derivative according to claim 4, which is partially 2-O-desulphated heparin with a molecular weight (MW) of 11000 D, a polydispersion index D of 1.5, a desulphation degree of 1.93 (expressed as SO₃ ⁻:COO⁻ molar ratio), percentage of modified uronic acids compared to total uronic acids: 5% epoxide groups, 29% oxidated and reduced uronic residues.

8) A derivative according to claim 4, which is partially 2-O-desulphated heparin with a molecular weight (MW) of 9200 D, a polydispersion index D of 1.5, percentage of modified uronic acids compared to total uronic acids: 11% epoxide groups, 27.5% oxidated and reduced uronic residues.

9) Process for the preparation of the derivatives according to claims 1-8, comprising the following steps:

(a) basic treatment at a temperature ranging from ambient temperature to approximately 100° C., preferably from 50 to 70° C., and even more preferably at approximately 65° C., which leads to the elimination of a controlled percentage of sulphate groups in position 2 of the iduronic acid and to the formation of epoxide groups; and, if desired

(b) opening of said epoxide ring at approximately pH 7, at a temperature ranging from approximately 50° C. to approximately 25° C., preferably at approximately 70° C. to yield residues of galacturonic acid; or, alternatively

(c) opening of said epoxide ring at a temperature ranging from approximately 0° C. to approximately 30° C., preferably at approximately 25° C. to yield residues of iduronic acid; and, if desired

(d) oxidation of the diols with sodium periodate, to yield the opening of the glycoside ring and the formation of two aldehyde groups per modified residue;

(e) reduction of said aldehyde groups to primary alcohol and, if desired, transformation of the D group to a group other than hydroxy, as envisaged in the different meanings according to claim 4;

(f) optional acid hydrolysis to obtain oligosaccharides corresponding to the regular sequences, or alternatively

(g) partial enzymatic hydrolysis with an enzyme selected from the group consisting of lyase, heparinase, heparitinase, or equivalent of products obtained in step e) to yield oligosaccharides, preferably tetra- or octa-saccharides, with the non-reducing terminal residue consisting of unsaturated iduronic acid, the reducing residue consisting of an N-sulphoglucosamine and containing at least one residue of open iduronic acid; or, alternatively

(h) the compound obtained in step a), or the product obtained in step b) is treated by partial enzyme hydrolysis; and, if desired

(i) subjection of the products obtained in one of steps a), b) and e) to partial 6-O-desulphation; or, alternatively,

(j) subjection of the starting heparin partially or totally 6-desulphated to steps a), b) and e).

10)Partially 2-O-desulphated heparin, obtainable by the process according to claim 9, where step a) is conducted for 45 min at 60° C. and step b) at 70° C. at pH 7, and with a molecular weight (MW) of 11200, a polydispersion index D of 1.3, a desulphation degree of 1.99 (expressed as SO₃ ⁻:COO⁻ molar ratio), percentage of modified uronic acids about 50% compared to total uronic acids.

11) Partially 2-O-desulphated heparin, obtainable by the process according to claim 9, where step a) is conducted for 15 min at 60° C. and step b) at 70° C. at pH 7, and with a molecular weight (MW) of 12900 D, a polydispersion index D of 1.5, a desulphation degree of 2.05 (expressed as SO3-:COO— molar ratio), percentage of modified uronic acids compared to total uronic acids: 5% epoxide groups, 29% oxidated and reduced uronic residues.

12) Partially 2-O-desulphated heparin, obtainable by the process according to claim 9, where step a) is conducted for 30 min at 60° C. and step b) at 70° C. at pH 7, and with a molecular weight (MW) of 11000 D, a polydispersion index D of 1.5, a desulphation degree of 1.80 (expressed as SO3-:COO— molar ratio), percentage of modified uronic acids compared to total uronic acids: 5% epoxide groups, 29% oxidated and reduced uronic residues.

13) Partially 2-O-desulphated heparin, obtainable by the process according to claim 9, where step a) is conducted for 60 min at 60° C. and step b) at 70° C. at pH 7, and with a molecular weight (MW) of 9200 D, a polydispersion index D of 1.5, percentage of modified uronic acids compared to total uronic acids: 11% epoxide groups, 27.5% oxidated and reduced uronic residues.

14) Mixtures of oligosaccharides with antiangiogenic activity obtainable from the acid degradation of compounds according to claims 4-8.

15) Mixtures of oligosaccharides with antiangiogenic activity obtainable from step f) of the process according to claim 9.

16) Mixtures of oligosaccharides with antiangiogenic activity obtainable from step g) of the process according to claim 9.

17) Mixtures of oligosaccharides with antiangiogenic activity obtainable from step h) of the process according to claim 9.

18) Mixtures of oligosaccharides with antiangiogenic activity obtainable from step i) of the process according to claim 9.

19) Pharmaceutical compositions containing at least one compound according to claims 1-8 and/or 10-18 as an active ingredient in admixture with pharmaceutically acceptable vehicles and excipients.

20) Use of compounds according to claims 1-8 and 10-18 as medicaments.

21) Use of compounds according to claims 1-8 and 10-18, or according to claim 20, for the preparation of a medicament with antiangiogenic activity.

22) Use of compounds according to claims 1-8 and 10-18, or according to claim 20, for the preparation of a medicament for the treatment of pathological states due to abnormal angiogenesis.

23) Use of compounds according to claims 1-8 and 10-18, or according to claim 20, for the preparation of a medicament for the treatment of tumours, optionally associated with metastases.

24) Use of compounds according to claims 1-8 and 10-18, or according to claim 20, for the preparation of a medicament for the treatment of tumour metastases.

25) Use of compounds according to claims 1-8 and 10-18, or according to claim 20, for the preparation of a medicament for the treatment of diabetic retinopathy.

26) Use of compounds according to claims 1-8 and 10-18, or according to claim 20, for the preparation of a medicament for the treatment of retrolenticular fibroplasia.

27) Use of compounds according to claims 1-8 and 10-18, or according to claim 20, for the preparation of a medicament for the treatment of psoriasis.

28) Use of compounds according to claims 1-8 and 10-18, or to claim 20, for the preparation of a medicament for the treatment of restenosis after angioplasty and after coronary by-pass.