Low Molecular Weight Heparin Derivatives With Improved Permeability

This invention is related to low molecular weight heparin derivatives possessing improved permeability. Thus, said low molecular weight heparin derivatives have an improved ability to pass through cell membranes. More particularly, the invention concerns low molecular weight heparin derivatives containing hydrophobic groups which impart improved permeability to the low molecular weight heparin while maintaining anticoagulant activity.

Heparin is the most widely used agent for immediate management of most thromboembolic disorders, particularly, deep-vein thrombosis and pulmonary and systemic emboli. Treatment times vary depending upon the use. Deep-vein thrombosis and pulmonary embolism are typically treated for 7-10 days. Thromboembolic disorders in pregnancy are typically treated for 2-6 weeks. Coronopathies, myocardiopathies, myocardial infarction and angina pectoris are typically treated for 30 days to many months.

Heparin must be administered by injection or intravenous infusion (parenteral). It is well known that commercial heparin is not capable of crossing the barrier posed by cell membranes such as those found in the intestine and cannot, therefore be an effective therapeutic agent when administered orally or rectally, for example.

The only commercially successful anticoagulant capable of oral administration is warfarin-sodium which is marketed under the trademark "Coumadin, Sodium." See for example U.S Pat. No. 2,999,049. Warfarin-sodium is a widely used rodenticide and is generally recognized as being inferior to heparin as an anticoagulant. Improving the ability of heparin to pass through membranes as, for example, is necessary in the case of heparin administered orally, rectally, transdermally, or topically, has been the subject of numerous patents. However, none of these products and approaches has proven to be commercially or technically successful.

U.S. Pat. No. 3,088,868 teaches the use of an amino acid adjuvant in conjunction with heparin to enable the heparin to be absorbed from the gastro-intestinal tract.

U.S. Pat. No. 3,482,014 teaches the conversion of a portion of the ionic sites of heparin to the acid form. This permits absorption through the walls of the intestinal tract.

U.S. Pat. No. 3,506,642 teaches conversion of the commercially available sodium heparin to the acid form followed by complexing with a suitable amino acid. This results in complexes which can be absorbed through the walls of the intestine. U.S. Pat. No. 3,577,534, which is a continuation in part of U.S. Pat. No. 3,506,642, teaches the use of said complexes in therapeutic compositions wherein the heparin is also absorbed through the walls of the intestine.

U.S. Pat. No. 3,510,561 teaches the preparation of compositions containing heparin and a sulfone. This permits absorption of the heparin through mucous membranes.

U.S. Pat. No. 3,546,338 teaches the combination of heparin, a metabolizable oil, water and a dispersing agent. Said combination is capable of being absorbed in the alimentary canal of mammals.

U.S. Pat. No. 3,548,052 teaches the use of alkyl sulfoxides, such as dimethyl sulfoxide, in conjunction with heparin to promote the absorption of heparin through mucous membranes.

U.S. Pat. No. 3,574,831 teaches the preparation of compositions containing sodium taurocholate and heparin. These compositions can be absorbed through the walls of the alimentary canal when administered orally or rectally.

U.S. Pat. No. 3,574,832 teaches compositions containing heparin and a surfactant selected from sodium lauryl sulfate, dioctyl sodium sulfosuccinate, sodium hexyl sulfate, sodium lauryl sulfonate, sodium cetyl sulfonate and mixtures thereof.

U.S. Pat. No. 3,835,112 teaches the preparation of heparin esters derived from fatty acids having at least 16 carbon atoms which can be administered orally. These esters are prepared by reacting the Hyamine 1622 salt of heparin with fatty acids in the presence of a carbodiimide.

U.S. Pat. No. 4,239,754 teaches the use of liposomes with heparin retained therein or thereon. The preparations of this patent are said to be orally active.

U.S. Pat. No. 4,281,108 discloses a process for obtaining low molecular weight heparin comprising acidification of heparin, depolymerization in the presence of peroxides, and sulfation. The molecular weights of the products are 4,000 to 12,000 daltons. Oral activity is claimed.

U.S. Pat. No. 4,331,697 teaches the preparation of heparin derivatives containing an active carbon-carbon double bond wherein the active carbon-carbon double bond is utilized to bond heparin to a biomedical material.

U.S. Pat. No. 4,440,926 teaches the preparation of selected heparin esters by reaction at the carboxyl sites of heparin. Said esters are prepared by reacting a quaternary ammonium salt or amine salt of heparin with an alcohol or a halide.

U.S. Pat. No. 4,510,135 discloses the use of organic ammonium heparin complexes for oral activity.

U.S. Pat. No. 4,533,549 teaches the depolymerization and fractionation of heparin to obtain derivatives in the molecular weight range of 2,500 to 4,000 daltons. Oral activity is claimed for these derivatives. However, the hydrophilic character of the compounds of this patent would be unsatisfactory for effective permeability.

U.K. Pat. No. GB 2,002,406B teaches the sulfation of a low molecular weight heparin having a molecular weight of from 2,600 to 5,500. Oral activity is claimed for the resulting product.

All of the above approaches suffer from shortcomings with regard to increasing the permeability of heparin. Additives, adjuvants, chemical modifications, and heparin derivatives of the prior art have proven to be unwieldy and ineffective.

It is well known that commercial heparin is not capable of crossing the barrier posed by the cell membranes of the intestine (see for example C. Doutremepuich et al., Path. Biol. 32, 45-48 (1984) or U.S. Pat. No. 3,548,052). The tendency of a substance to pass through cell membranes can be measured in terms of a permeability constant (P) which is linearly related to the partition coefficient (r) between a water immiscible liquid and water itself, and inversely related to the square root of the molecular weight (M) of the substance crossing the membrane (see for example J. Diamond and Y. Katz, J. Membr. Biol. 17, 121-154 (1974); J. Danielli, "The Permeability of Natural Membranes", Cambridge University press, Cambridge (1952); A. Kotyk, Biochim. Biophys. Acta 300, 183 (1973)). This is shown in equation form below. ##EQU1##

Therefore, in order to increase the permeability constant for heparin, it is necessary to increase its partition coefficient (hydrophobicity) and/or decrease its molecular weight. In the preferred embodiment of this invention the partition coefficient is increased and the molecular weight is decreased.

It has now been discovered that the permeability of low molecular weight heparin can be significantly increased by the addition of ester groups to low molecular weight heparin. Surprisingly, ester groups containing as few as 2 carbon atoms are effective at significantly increasing the permeability. Preferably, the ester groups contain 3 or more carbon atoms.

We prefer to prepare low molecular weight heparin derivatives having high permeability by allowing a suitable acid chloride to react with low molecular weight heparin. As is apparent to one skilled in the art, a wide variety of reaction conditions and solvents can be used to effect this reaction. The degree of substitution can be changed by varying the ratio of acid chloride to low molecular weight heparin, by varying the solvent or by using no solvent at all, by varying the reaction time and/or by varying the reaction temperature.

As is apparent to those skilled in the art there are many workup procedures which will allow the isolation of low molecular weight heparin derivatives possessing a permeability significantly greater than heparin. We prefer the use of dialysis in the workup procedure to isolate low molecular weight heparin derivatives possessing high permeability.

Heparin is a mucopolysaccharide composed of amino sugar and uronic acid residues. Heparin is obtained from beef, porcine, sheep, whale, or other mammalian tissue by extraction via procedures known to those skilled in the art. Commercial heparin preparations are now widely available from many sources and are distributed primarily for use as intravascular anticoagulants.

Heparin preparations are known to be heterogeneous on a molecular level. Thus, they exhibit a considerable degree of polydispersity in molecular size, variations in the ratio of glucuronic acid to iduronic acid, alterations in the amount of ester and N-sulfation, and differing extents of N-acetylation. Changes in these parameters have been correlated only to a very limited extent with heparin's anticoagulant potency. Accordingly, it has been widely assumed that its anticoagulant activity is not traceable to a single specific heparin structure, and in any event, no precise relationship between its structure and function has been forthcoming.

The chemical structure of heparin is complex. Heparin is not a single compound, but rather is a mixture of compounds. However, heparin is commonly thought to primarily be a polymeric substance made up of tetrasaccharide repeating units. On the average each tetrasaccharide repeating unit contains approximately 5 free hydroxyl groups and has a molecular weight of approximately 1229. The average molecular weight of commercially available heparin varies from about 10,000 to about 16,000. Thus, on the average, commercially available heparin contains approximately 8 to 13 tetrasaccharide repeating units.

The low molecular weight heparin used in the examples below was prepared following Canadian patent No. 1,195,322. It exhibited a molecular weight of 5187 daltons. Thus, this low molecular weight heparin contained approximately 4.2 tetrasaccharide repeating units.

The heparin used in the examples below is porcine mucosal heparin manufactured by Hepar Industries, Inc., Franklin, OH. It exhibited a molecular weight of 13684 daltons. Thus, this heparin contained approximately 11.3 tetrasaccharide repeating units.

Heparin utilized in the practice of this invention may be derived from porcine intestinal mucosa, beef lungs, and whale tissue as well as from other sources known to those skilled in the art. Synthetically derived heparin and heparin-like substances may also be utilized in the practice of this invention. The preferred sources for use in this invention are porcine intestinal mucosa and beef lungs.

The term low molecular weight heparin is used in the specification and the claims in its broadest sense, in order to designate a low molecular weight fraction isolated from heparin, a product obtained by depolymerizing heparin, or a mucopolysaccharide exhibiting anticoagulant properties that is synthesized from non-heparin sources. Low molecular weight is used to mean a material exhibiting a molecular weight of less than 10,000 daltons.

Without limitation as to the scope of the invention, it is theorized that a preponderance of the ester groups formed by the reaction of an acid chloride with low molecular weight heparin result from the reaction of the free hydroxyl groups of low molecular weight heparin with the acid chloride.

It will be understood that the invention is not limited to the specific examples given herein. Other examples will be obvious to those skilled in the art.

The products of this invention are solids. They can be readily formulated into powders, pills, lozenges, tablets, capsules, ointments, liquids or other suitable forms. Where the compositions are to be swallowed and absorption is to take place in the intestine, the compositions may be given an enteric coating such as cellulose acetate, phthalate, styrene-maleic anhydride copolymers and the like. Enteric coatings are well known to those skilled in the art and are discussed for example in Remington's Practice of Pharmacy or in U.S. Pat. No. 3,126,320.

Preparation of buccal or sublingual tablets and of rectal enemas, suppositories and ointments as well as nasal mists, inhalants and transdermal delivery systems can be easily accomplished.

The following examples are given by way of illustration only and are not to be considered as limiting of this invention.

2 grams of low molecular weight heparin (molecular weight of 5187 daltons, prepared following Canadian Pat. No. 1,195,322) was added to a 250 milliliter round bottom flask protected from the atmosphere by a drying tube. To this was added 24 milliliters of formamide and 24 milliliters of pyridine. The flask was placed in an oil bath maintained at 40.degree. C. 40 milliliters of acetyl chloride was added slowly over a 3-4 hour period with agitation and agitation continued overnight.

50 milliliters of water was then added with agitation. The contents of the flask were then placed in a 2000 molecular weight cutoff dialysis bag (Spectrum Medical Industries, Los Angeles, CA). Dialysis was conducted against a 1% (w/v) sodium chloride solution for 24 hours. The dialysis against 1% sodium chloride was repeated three times. Dialysis was then conducted against water for 24 hours. The dialysis against water was then repeated three times.

The contents of the dialysis bag were then lyophilized to obtain a dry, white powder.

0.1 grams of the product of example I was added to 2 milliliters of deionized water. 3 ml of butanol was added. The mixture was mixed well and allowed to undergo freeze/thaw cycles until the top layer (butanol) was clear. The amount of product in each layer was determined by running the uronic acid assay according to E. V. Chandrasekaran and J. N. BeMiller, "Methods in Carbohydrate Chemistry," vol VIII, pages 89-96 (1980) using heparin (porcine intestinal mucosa) standards. The results indicated a butanol/water partition coefficient of 1.0.times.10.sup.-3.

The number of ester groups per tetrasaccharide unit contained in the product of example I was measured following the method of S. Hestrin, J. BIOL. CHEM, vol 180, pages 249-261, 1949. Butyryl choline chloride was used as the ester standard. A theoretical formula weight of 1229 was used for a tetrasaccharide unit. The results indicated the presence of 4.4 acetyl groups per tetrasaccharide unit.

The molecular weight of the product of example I was estimated by taking the sum of the theoretical formula weight of a tetrasaccharide unit (1229) plus 4.4 times the formula weight of a acetyl group less the formula weight of the hydrogen replaced in the heparin (42) and multiplying said sum by the molecular weight of the starting low molecular weight heparin (5187) and finally dividing by the theoretical formula weight of a tetrasaccharide unit (1229). This gave a molecular weight of 5967 for the product of example I.

The permeability of the product of example I relative to heparin (porcine intestinal mucosa) was determined as follows. The butanol/water partition coefficient of the product of example I was divided by the square root of the molecular weight of the product of example I. A corresponding value was then obtained for heparin. The value obtained for the product of example I was then divided by the corresponding value for heparin to give 15.1 which is the permeability relative to heparin.

The infrared spectrum was obtained on the product of example I. An absorption peak was observed at 1743 cm.sup.-1. This peak is characteristic of an ester group. Said peak was not present in the starting heparin.

2 grams of low molecular weight heparin (molecular weight of 5187 daltons, prepared following Canadian Pat. No. 1,195,322) was added to a 250 milliliter round bottom flask protected from the atmosphere by a drying tube. To this was added 24 milliliters of formamide and 24 milliliters of pyridine. The flask was placed in an oil bath maintained at 50.degree. C. 12 milliliters of propionyl chloride was added slowly over a 3-4 hour period with agitation and agitation continued overnight.

50 milliliters of water was then added with agitation. The contents of the flask were then placed in a 2000 molecular weight cutoff dialysis bag (Spectrum Medical Industries, Los Angeles, CA). Dialysis was conducted against a 1% (w/v) sodium chloride solution for 24 hours. The dialysis against 1% sodium chloride was repeated three times. Dialysis was then conducted against water for 24 hours. The dialysis against water was then repeated three times.

The contents of the dialysis bag were then lyophilized to obtain a dry, white powder.

0.1 grams of the product of example VII was added to 2 milliliters of deionized water. 3 ml of butanol was added. The mixture was mixed well and allowed to undergo freeze/thaw cycles until the top layer (butanol) was clear. The amount of product in each layer was determined by running the uronic acid assay according to E. V. Chandrasekaran and J. N. BeMiller, "Methods in Carbohydrate Chemistry," vol VIII, pages 89-96 (1980) using heparin (porcine intestinal mucosa) standards. The results indicated a butanol/water partition coefficient of 2.0.times.10.sup.31 3.

The number of ester groups per tetrasaccharide unit contained in the product of example VII was measured following the method of S. Hestrin, J. BIOL. CHEM, vol 180, pages 249-261, 1949. Butyryl choline chloride was used as the ester standard. A theoretical formula weight of 1229 was used for a tetrasaccharide unit. The results indicated the presence of 2.9 propionyl groups per tetrasaccharide unit.

The molecular weight of the product of example VII was estimated by taking the sum of the theoretical formula weight of a tetrasaccharide unit (1229) plus 2.9 times the formula weight of a propionyl group less the formula weight of the hydrogen replaced in the heparin (56) and multiplying said sum by the molecular weight of the starting low molecular weight heparin (5187) and finally dividing by the theoretical formula weight of a tetrasaccharide unit (1229). This gave a molecular weight of 5872 for the product of example VII.

The permeability of the product of example VII relative to heparin (porcine intestinal mucosa) was determined as follows. The butanol/water partition coefficient of the product of example VII was divided by the square root of the molecular weight of the product of example VII. A corresponding value was then obtained for heparin. The value obtained for the product of example VII was then divided by the corresponding value for heparin to give 30.6 which is the permeability relative to heparin.

The infrared spectrum was obtained on the product of example VII. An absorption peak was observed at 1739 cm.sup.-1. This peak is characteristic of an ester group. Said peak was not present in the starting heparin.

2 grams of low molecular weight heparin (molecular weight of 5187 daltons, prepared following Canadian Pat. No. 1,195,322) was added to a 250 milliliter round bottom flask protected from the atmosphere by a drying tube. To this was added 24 milliliters of formamide and 24 milliliters of pyridine. The flask was placed in an oil bath maintained at 30.degree. C. 20 milliliters of butyryl chloride was added slowly over a 3-4 hour period with agitation and agitation continued overnight.

50 milliliters of water was then added with agitation. The contents of the flask were then placed in a 2000 molecular weight cutoff dialysis bag (Spectrum Medical Industries, Los Angeles, CA). Dialysis was conducted against a 1% (w/v) sodium chloride solution for 24 hours. The dialysis against 1% sodium chloride was repeated three times. Dialysis was then conducted against water for 24 hours. The dialysis against water was then repeated three times.

The contents of the dialysis bag were then lyophilized to obtain a dry, white powder.

0.1 grams of the product of example XIII was added to 2 milliliters of deionized water. 3 ml of butanol was added. The mixture was mixed well and allowed to undergo freeze/thaw cycles until the top layer (butanol) was clear. The amount of product in each layer was determined by running the uronic acid assay according to E. V. Chandrasekaran and J. N. BeMiller, "Methods in Carbohydrate Chemistry," vol VIII, pages 89-96 (1980) using heparin (porcine intestinal mucosa) standards. The results indicated a butanol/water partition coefficient of 8.2.times.10.sup.31 3.

The number of ester groups per tetrasaccharide unit contained in the product of example XIII was measured following the method of S. Hestrin, J. BIOL. CHEM, vol 180, pages 249-261, 1949. Butyryl choline chloride was used as the ester standard. A theoretical formula weight of 1229 was used for a tetrasaccharide unit. The results indicated the presence of 1.5 butyryl groups per tetrasaccharide unit.

The molecular weight of the product of example XIII was estimated by taking the sum of the theoretical formula weight of a tetrasaccharide unit (1229) plus 1.5 times the formula weight of a butyryl group less the formula weight of the hydrogen replaced in the heparin (70) and multiplying said sum by the molecular weight of the starting low molecular weight heparin (5187) and finally dividing by the theoretical formula weight of a tetrasaccharide unit (1229). This gave a molecular weight of 5630 for the product of example XIII.

The permeability of the product of example XIII relative to heparin (porcine intestinal mucosa) was determined as follows. The butanol/water partition coefficient of the product of example XIII was divided by the square root of the molecular weight of the product of example XIII. A corresponding value was then obtained for heparin. The value obtained for the product of example XIII was then divided by the corresponding value for heparin to give 128 which is the permeability relative to heparin.

The infrared spectrum was obtained on the product of example XIII. An absorption peak was observed at 1736 cm.sup.-1. This peak is characteristic of an ester group. Said peak was not present in the starting heparin.

2 grams of low molecular weight heparin (molecular weight of 5187 daltons, prepared following Canadian Pat. No. 1,195,322) was added to a 250 milliliter round bottom flask protected from the atmosphere by a drying tube. To this was added 24 milliliters of formamide and 24 milliliters of pyridine. The flask was placed in an oil bath maintained at 50.degree. C. 20 milliliters of butyryl chloride was added slowly over a 3-4 hour period with agitation and agitation continued overnight.

50 milliliters of water was then added with agitation. The contents of the flask were then placed in a 2000 molecular weight cutoff dialysis bag (Spectrum Medical Industries, Los Angeles, CA). Dialysis was conducted against a 1% (w/v) sodium chloride solution for 24 hours. The dialysis against 1% sodium chloride was repeated three times. Dialysis was then conducted against water for 24 hours. The dialysis against water was then repeated three times.

The contents of the dialysis bag were then lyophilized to obtain a dry, white powder.

0.1 grams of the product of example XIX was added to 2 milliliters of deionized water. 3 ml of butanol was added. The mixture was mixed well and allowed to undergo freeze/thaw cycles until the top layer (butanol) was clear. The amount of product in each layer was determined by running the uronic acid assay according to E. V. Chandrasekaran and J. N. BeMiller, "Methods in Carbohydrate Chemistry," vol VIII, pages 89-96 (1980) using heparin (porcine intestinal mucosa) standards. The results indicated a butanol/water partition coefficient of 11.0.times.10.sup.31 3.

The number of ester groups per tetrasaccharide unit contained in the product of example XIX was measured following the method of S. Hestrin, J. BIOL. CHEM, vol 180, pages 249-261, 1949. Butyryl choline chloride was used as the ester standard. A theoretical formula weight of 1229 was used for a tetrasaccharide unit. The results indicated the presence of 3.0 butyryl groups per tetrasaccharide unit.

The molecular weight of the product of example XIX was estimated by taking the sum of the theoretical formula weight of a tetrasaccharide unit (1229) plus 3.0 times the formula weight of a butyryl group less the formula weight of the hydrogen replaced in the heparin (70) and multiplying said sum by the molecular weight of the starting low molecular weight heparin (5187) and finally dividing by the theoretical formula weight of a tetrasaccharide unit (1229). This gave a molecular weight of 6073 for the product of example XIX.

The permeability of the product of example XIX relative to heparin (porcine intestinal mucosa) was determined as follows. The butanol/water partition coefficient of the product of example XIX was divided by the square root of the molecular weight of the product of example XIX. A corresponding value was then obtained for heparin. The value obtained for the product of example XIX was then divided by the corresponding value for heparin to give 165 which is the permeability relative to heparin.

The infrared spectrum was obtained on the product of example XIX. An absorption peak was observed at 1739 cm.sup.-1. This peak is characteristic of an ester group. Said peak was not present in the starting heparin.

2 grams of low molecular weight heparin (molecular weight of 5187 daltons, prepared following Canadian Pat. No. 1,195,322) was added to a 250 milliliter round bottom flask protected from the atmosphere by a drying tube. To this was added 24 milliliters of formamide and 24 milliliters of pyridine. The flask was placed in an oil bath maintained at 50.degree. C. 6 milliliters of decanoyl chloride was added slowly over a 3-4 hour period with agitation and agitation continued overnight.

The contents of the flask were placed in a 2000 molecular weight cutoff dialysis bag (Spectrum Medical Industries, Los Angeles, CA). Dialysis was conducted against: 95% ethanol for 24 hours, 95% ethanol for 24 hours, 47.5% ethanol for 24 hours, 47.5% ethanol for 24 hours, 1% sodium chloride for 24 hours, 1% sodium chloride for 24 hours and water for 24 hours. The dialysis against water was then repeated.

The contents of the dialysis bag were then lyophilized to obtain a dry, white powder.

0.1 grams of the product of example XXV was added to 2 milliliters of deionized water. 3 ml of butanol was added. The mixture was mixed well and allowed to undergo freeze/thaw cycles until the top layer (butanol) was clear. The amount of product in each layer was determined by running the uronic acid assay according to E. V. Chandrasekaran and J. N. BeMiller, "Methods in Carbohydrate Chemistry," vol VIII, pages 89-96 (1980) using heparin (porcine intestinal mucosa) standards. The results indicated a butanol/water partition coefficient of 6.8.times.10.sup.31 3.

The number of ester groups per tetrasaccharide unit contained in the product of example XXV was measured following the method of S. Hestrin, J. BIOL. CHEM, vol 180, pages 249-261, 1949. Butyryl choline chloride was used as the ester standard. A theoretical formula weight of 1229 was used for a tetrasaccharide unit. The results indicated the presence of 0.36 decanoyl groups per tetrasaccharide unit.

The molecular weight of the product of example XXV was estimated by taking the sum of the theoretical formula weight of a tetrasaccharide unit (1229) plus 0.36 times the formula weight of a decanoyl group less the formula weight of the hydrogen replaced in the heparin (154) and multiplying said sum by the molecular weight of the starting low molecular weight heparin (5187) and finally dividing by the theoretical formula weight of a tetrasaccharide unit (1229). This gave a molecular weight of 5421 for the product of example XXV.

The permeability of the product of example XXV relative to heparin (porcine intestinal mucosa) was determined as follows. The butanol/water partition coefficient of the product of example XXV was divided by the square root of the molecular weight of the product of example XXV. A corresponding value was then obtained for heparin. The value obtained for the product of example XXV was then divided by the corresponding value for heparin to give 108 which is the permeability relative to heparin.

The infrared spectrum was obtained on the product of example XXXV. An absorption peak was observed at 1739 cm.sup.-1. This peak is characteristic of an ester group. Said peak was not present in the starting heparin.

2 grams of low molecular weight heparin (molecular weight of 5187 daltons, prepared following Canadian Pat. No. 1,195,322) was added to a 250 milliliter round bottom flask protected from the atmosphere by a drying tube. To this was added 24 milliliters of formamide and 24 milliliters of pyridine. The flask was placed in an oil bath maintained at 50.degree. C. 12 milliliters of decanoyl chloride was added slowly over a 3-4 hour period with agitation and agitation continued overnight.

The contents of the flask were placed in a 2000 molecular weight cutoff dialysis bag (Spectrum Medical Industries, Los Angeles, CA). Dialysis was conducted against: 95% ethanol for 24 hours, 95% ethanol for 24 hours, 47.5% ethanol for 24 hours, 47.5% ethanol for 24 hours, 1% sodium chloride for 24 hours, 1% sodium chloride for 24 hours and water for 24 hours. The dialysis against water was then repeated.

The contents of the dialysis bag were then lyophilized to obtain a dry, white powder.

0.1 grams of the product of example XXXI was added to 2 milliliters of deionized water. 3 ml of butanol was added. The mixture was mixed well and allowed to undergo freeze/thaw cycles until the top layer (butanol) was clear. The amount of product in each layer was determined by running the uronic acid assay according to E. V. Chandrasekaran and J. N. BeMiller, "Methods in Carbohydrate Chemistry," vol VIII, pages 89-96 (1980) using heparin (porcine intestinal mucosa) standards. The results indicated a butanol/water partition coefficient of 44.times.10.sup.31 3.

The number of ester groups per tetrasaccharide unit contained in the product of example XXXI was measured following the method of S. Hestrin, J. BIOL. CHEM, vol 180, pages 249-261, 1949. Butyryl choline chloride was used as the ester standard. A theoretical formula weight of 1229 was used for a tetrasaccharide unit. The results indicated the presence of 0.71 decanoyl groups per tetrasaccharide unit.

The molecular weight of the product of example XXXI was estimated by taking the sum of the theoretical formula weight of a tetrasaccharide unit (1229) plus 0.71 times the formula weight of a decanoyl group less the formula weight of the hydrogen replaced in the heparin (154) and multiplying said sum by the molecular weight of the starting low molecular weight heparin (5187) and finally dividing by the theoretical formula weight of a tetrasaccharide unit (1229). This gave a molecular weight of 5649 for the product of example XXXI.

The permeability of the product of example XXXI relative to heparin (porcine intestinal mucosa) was determined as follows. The butanol/water partition coefficient of the product of example XXXI was divided by the square root of the molecular weight of the product of example XXXI. A corresponding value was then obtained for heparin. The value obtained for the product of example XXXI was then divided by the corresponding value for heparin to give 685 which is the permeability relative to heparin.

The infrared spectrum was obtained on the product of example XXXI. An absorption peak was observed at 1737 cm.sup.-1. This peak is characteristic of an ester group. Said peak was not present in the starting heparin.

2 grams of low molecular weight heparin (molecular weight of 5187 daltons, prepared following Canadian patent No. 1,195,322) was added to a 250 milliliter round bottom flask protected from the atmosphere by a drying tube. To this was added 24 milliliters of fprmamide and 24 milliliters of pyridine. The flask was placed in an oil bath maintained at 50.degree. C. 24 milliliters of decanoyl chloride was added slowly over a 3-4 hour period with agitation and agitation continued overnight.

The contents of the flask were placed in a 2000 molecular weight cutoff dialysis bag (Spectrum Medical Industries, Los Angeles, CA). Dialysis was conducted against: 95% ethanol for 24 hours, 95% ethanol for 24 hours, 47.5% ethanol for 24 hours, 47.5% ethanol for 24 hours, 1% sodium chloride for 24 hours, 1% sodium chloride for 24 hours and water for 24 hours. The dialysis against water was then repeated.

The contents of the dialysis bag were then lyophilized to obtain a dry, white powder.

0.1 grams of the product of example XXXVII was added to 2 milliliters of deionized water. 3 ml of butanol was added. The mixture was mixed well and allowed to undergo freeze/thaw cycles until the top layer (butanol) was clear. The amount of product in each layer was determined by running the uronic acid assay according to E. V. Chandrasekaran and J. N. BeMiller, "Methods in Carbohydrate Chemistry," vol VIII, pages 89-96 (1980) using heparin (porcine intestinal mucosa) standards. The results indicated a butanol/water partition coefficient of 663.times.10.sup.31 3.

The number of ester groups per tetrasaccharide unit contained in the product of example XXXVII was measured following the method of S. Hestrin, J. BIOL. CHEM, vol 180, pages 249-261, 1949. Butyryl choline chloride was used as the ester standard. A theoretical formula weight of 1229 was used for a tetrasaccharide unit. The results indicated the presence of 1.03 decanoyl groups per tetrasaccharide unit.

The molecular weight of the product of example XXXVII was estimated by taking the sum of the theoretical formula weight of a tetrasaccharide unit (1229) plus 1.03 times the formula weight of a decanoyl group less the formula weight of the hydrogen replaced in the heparin (154) and multiplying said sum by the molecular weight of the starting low molecular weight heparin (5187) and finally dividing by the theoretical formula weight of a tetrasaccharide unit (1229). This gave a molecular weight of 857 for the product of example XXXVII.

The permeability of the product of example XXXVII relative to heparin (porcine intestinal mucosa) was determined as follows. The butanol/water partition coefficient of the product of example XXXVII was divided by the square root of the molecular weight of the product of example XXXVII. A corresponding value was then obtained for heparin. The value obtained for the product of example XXXVII was then divided by the corresponding value for heparin to give 10,100 which is the permeability relative to heparin.

The infrared spectrum was obtained on the product of example XXXVII. An absorption peak was observed at 1742 cm.sup.-1. This peak is characteristic of an ester group. Said peak was not present in the starting heparin.

0.1 grams of low molecular weight heparin (molecular weight of 5187 daltons, prepared following Canadian Pat. No. 1,195,322) was added to 2 milliliters of deionized water. 3 ml of butanol was added. The mixture was mixed well and allowed to undergo freeze/thaw cycles until the top layer (butanol) was clear. The amount of product in each layer was determined by running the uronic acid assay according to E. V. Chandrasekaran and J. N. BeMiller, "Methods in Carbohydrate Chemistry," vol VIII, pages 89-96 (1980) using heparin (porcine intestinal mucosa) standards. The results indicated a butanol/water partition coefficient of 0.1.times.10.sup.31 3.

The permeability of low molecular weight heparin (molecular weight of 5187 daltons, prepared following Canadian Pat. No. 1,195,322) relative to heparin (porcine intestinal mucosa) was determined as follows. The butanol/water partition coefficient of the product of example XLIII was divided by the square root of the molecular weight of the product of example XLIII. A corresponding value was then obtained for heparin. The value obtained for the product of example XLIII was then divided by the corresponding value for heparin to give 1.6 which is the permeability relative to heparin.

0.1 grams of heparin (porcine intestinal mucosa) was added to 2 milliliters of deionized water. 3 ml of butanol was added. The mixture was mixed well and allowed to undergo freeze/thaw cycles until the top layer (butanol) was clear. The amount of product in each layer was determined by running the uronic acid assay according to E. V. Chandrasekaran and J. N. BeMiller, "Methods in Carbohydrate Chemistry," vol VIII, pages 89-96 (1980) using heparin (porcine intestinal mucosa) standards. The results indicated a butanol/water partition coefficient of 0.1.times.10.sup.31 3.

The above description is for the purpose of teaching the person skilled in the art how to practice the present invention. This description is not intended to detail all of the obvious modifications and variations of the invention which will become apparent upon reading. However, the applicants do intend to include all such obvious modifications and variations within the scope of their invention which is defined by the following claims.