Degradable poly(ethylene glycol) hydrogels with controlled half-life and precursors therefor

ES2 281 925 Τ5

DESCRIPTION

Degradable polyethylene glycol hydrogels with controlled half-life and precursors thereto

FIELD OF THE INVENTION

This invention relates to polyethylene glycol hydrogels, their precursors, methods for making the precursors and hydrogels, and the use of the precursors and hydrogels.

BACKGROUND OF THE INVENTION

In its most common form, polyethylene glycol (PEG) is a linear polymer terminated at each end with hydroxyl groups:

HO-CH₂CH2O- (CH2CH2O) n-CH₂CH2-OH

This polymer can be represented in abbreviated form as HO-PEG-OH, in which -PEG is understood to represent the following structural unit:

-CH2CH2O- (CH2CH2O) n-CH2CH2n typically ranges from about 10 to 2000.

PEG is very useful in biotechnology, and is useful in various applications for drug delivery and surface modification to promote non-stick characteristics "nonfouling", including as hydrogels and for covalent binding to various drugs and surfaces. PEG is non-toxic, does not tend to promote an immune response, and is soluble in water and many organic solvents.

The PEG polymer can be covalently attached to insoluble molecules, to make the PEG-molecule conjugate soluble. For example, Greenwald, Pendri and Bolikal, in J. Org. Chem., 60, 331-336 (1995), report that the water-insoluble drug paclitaxel, when bound to PEG, becomes soluble in water.

Davis et al., In US patent 4,179,337, report that PEG-bound proteins have an increased life of circulation in the blood, due to a reduced rate of renal elimination, and a reduced immunogenicity. The lack of toxicity of the polymer, and its rate of elimination from the body, are important considerations in pharmaceutical applications. Pharmaceutical applications, and many prominent references, are described in Harris's book (JM Harris, Ed., "Biomedical and Biotechnical Applications of Polyethylene Glycol Chemistry", Plenum, New York, 1992).

PEG is commonly used as methoxy-PEG-OH, or mPEG for short, in which one end is the relatively inert methoxy group, while the other end is a hydroxyl group that is susceptible to simple chemical modifications.

CH3O- (CH2CH2O) n-CH2CH2-OH mPEG

PEG is also commonly used in branched forms, which can be prepared by adding ethylene oxide to various polyalcohols, including glycerol, pentaerythritol, and sorbitol. For example, four-branch branched PEG, prepared from pentaerythritol, is shown below:

C (CH2-OH) 4 + nC2H4O ® C [CH2O- (CH2CH2O) n-CH2CH2-OH] 4

Branched PEGs can be generally represented as R (-PEG-OH) n, where R represents the core molecule, which can include glycerol or pentaerythritol, and n represents the number of branches.

It is necessary to use an "activated derivative" of PEG to bind the PEG to a molecule. The hydroxyl group located at the end of the PEG or another group susceptible to simple chemical modifications, is activated by modifying or replacing the group with a suitable functional group to react with a group or other molecule, which includes proteins, surfaces, enzymes, and others. For example, the succinimidyl "active ester" of carboxymethylated PEG forms covalent bonds with amino groups in proteins, as described by K. Iwasaki and Y. Iwashita in US Patent 4,670,417.

The synthesis described in US Patent No.⁰ 4,670,417 is illustrated below, with the reaction of the active ester with amino groups of a protein, in which the succinimidyl group is represented as NHS, and the protein is represented as PRO-NH2:

PEG-O-CH2-CO2-NHS + PRO-NH2 ® PEG-O-CH2-CO2-NH-PRO

ES2 281 925 Τ5

Succinimidyl "active esters", such as PEG-O-CH2-CO2-NHS, are commonly used forms of activated PEG-carboxylic acids, and are prepared by reacting PEG-carboxylic acids with N-hydroxy-succinimide.

Problems have arisen in the art. Some of the functional groups that have been used to activate PEG can result in toxic or otherwise undesirable residues when used for in vivo drug delivery. Some of the linkages that have been devised to attach functional groups to PEG can result in an undesirable immune response. Some of the functional groups do not have sufficient or otherwise appropriate selectivity to react with particular groups on proteins, and may tend to inactivate proteins.

PEG hydrogels, which are gels that swell in water, have been used for wound dressing and drug delivery. PEG hydrogels are prepared by incorporating the soluble hydrophilic polymer into a chemically cross-linked network or matrix, such that the addition of water produces an insoluble swollen gel. Substances useful as drugs are typically not covalently attached to the PEG hydrogel for in vivo administration. Instead, the substances are trapped within the cross-linked matrix, and pass through the interstices of the matrix. The insoluble matrix can remain in the body indefinitely, and control of drug release can typically be somewhat imprecise.

A method for the preparation of these hydrogels is described by Embrey and Grant in US Patent No.⁰4,894,238. The ends of the linear polymer are connected by various strong, non-degradable chemical bonds. For example, linear PEG is incorporated into a cross-linked network by reaction with a triol and a diisocyanate, to form hydrolytically stable urethane linkages that are not degradable in water.

A related method for the preparation of PEG hydrogels has been described by Gayet and Fortier in J. Controlled Release, 38, 177-184 (1996), in which linear PEG was activated as p-nitrophenylcarbonate, and cross-linked by reaction with a protein, bovine serum albumin. The bonds formed are hydrolysis-stable urethane groups, and hydrogels are not degradable in water.

In another method, described by NS Chu in US patent 3,963,805, non-degradable PEG networks have been prepared by random felting of PEG chains with other polymers formed with the use of mixed free radical initiators. with multifunctional monomers. PA King described non-degradable PEG hydrogels in US Patent 3,149,006, which were prepared by radiation-induced crosslinking of high molecular weight PEG.

Nagaoka et al., In US Patent 4,424,311, have prepared PEG hydrogels by copolymerizing PEG methacrylate with other comonomers, such as methyl methacrylate. By this method, polymeric elements that are not substantially PEG are introduced. A vinyl polymerization produces a polyethylene backbone with attached PEG. The methyl methacrylate comonomer is added to provide the additional physical strength of the gel.

European patent application EP 0593284 describes the synthesis and use of polyethylene glycols with photochemically or thermochemically activated crosslinking groups at the chain ends. Exposure to light or heat, respectively, causes these crosslinking groups to randomly react with various bonds in the polymer backbone by alkylation or acylation. A biodegradable peptide can be included in the polymer. These peptides are degradable by enzyme catalyzed procedures.

European patent applications EP 0794211 and EP 0771832 refer to copolymers and blends of polyesters with other components, in which the polymers are formed by polymerization in the high temperature melt. Esters are acetate esters that include the moiety -O-CH2CO2-. These esters would be expected to hydrolyze to produce ethylene glycol, which is a toxic compound.

Sawhney, Pathak and Hubbell, in Macromolecules, 26, 581 (1993), describe the preparation of block copolymers of poly (glycolic acid) or poly (lactic acid) and PEG that are terminated with acrylate groups, as shown below .

CH2 = CH-CO- (O-CH2-CO) nO-PEG-O- (CO-CH2-O) n-CO-CH = CH2

In the above formula, the glycolic blocks are the -O-CH2-CO- units; the addition of a methyl group to the methylene gives rise to a lactic block; n can be a multiple of 2. Vinyl polymerization of acrylate groups provides an insoluble cross-linked gel with a polyethylene backbone.

Elements that are not substantially PEG are introduced into the hydrogel. The poly (lactic acid) or poly (glycolic acid) segments of the polymeric backbone shown above, which are ester groups, are prone to slow hydrolytic decomposition, with the result that the cross-linked gel undergoes slow degradation and dissolution.

ES2 281 925 Τ5

Non-PEG elements tend to introduce complexity into the hydrogel, and degradation and dissolution of the matrix can result in undesirable or toxic components being released into the bloodstream when hydrogels are used in vivo for drug delivery.

It would be desirable to provide alternative PEG hydrogels, which are suitable for drug delivery, and have unique properties that can enhance drug delivery systems.

SUMMARY OF THE INVENTION

The invention provides chemically cross-linked degradable PEG hydrogels as a drug delivery system, for use as a drug, and for the manufacture of a drug capable of degrading in a controlled manner, and methods for manufacturing these PEG hydrogels in the absence of elements that they are not PEG, as defined in the present claims. Weak chemical bonds are introduced into the hydrogel, providing hydrolytic breakdown of the crosslinks and the release of drug molecules that may be entrapped within the matrix. The gels break down into substantially non-toxic PEG fragments, which are typically eliminated from the body. Variation of the atoms near the hydrolytically unstable bonds can provide precise control of the rate of hydrolytic decomposition and drug release. Examples of hydrolytically unstable bonds include carboxylate esters as defined in the present claims, phosphate esters, acetals, imines, orthoesters, peptides, and oligonucleotides. These weak bonds are formed by the reaction of two PEGs that have different end groups, as illustrated below:

-PEG-Z + Y-PEG- ® -PEG-W-PEG In the illustration above, -W- represents the weak bond unstable against hydrolysis, Z- and Y- represent groups located at the ends of the PEG molecule, that are able to react with each other to form weak bonds -W-.

For example, the following pairs of Z and Y groups can be used to form some of the W groups described above:

-PEG-OPO3H2 + HO-PEG ® -PEG-OPO₃(H) -PEG- phosphate ester

-PEG-CHO + (HO-PEG) 2- ® -PEG-CH (O-PEG) 2- acetal

PEG hydrogels can be prepared either by a two-stage method or by a one-stage method. In the one-step approach, two different PEGs are reacted with the appropriate end groups in a single step. A specific example of the one-step approach according to the invention is shown in the following equation, for linking linear PEG acids with a hydroxyl-terminated three-legged PEG. Weak ester bonds are formed.

HO2C- (CH2) nO-PEG-O- (CH2) n-CO2H + CH3C (CH2-O-PEG-OH) 3 ® {CH3C [CH2-O-PEG-O2C- (CH2) nO-PEG-O ( CH2) n-CO2 -] 3} m-H2O

The degree of polymerization is given by m, which refers to "matrix", and is intended to indicate that a cross-linked polymer has been formed as a solid aggregate. N is from about 1 to 10, and can be varied to control the rate of hydrolysis of the gel, usually increasing N to decrease the rate of hydrolysis. It should be understood that the degree of polymerization through the formation of crosslinks is large and indeterminate. The PEG hydrogel that forms is a visible, solid aggregate, which swells in water, in which, in theory, all available crosslinks are formed. However, it is not usually possible to determine the degree of crosslinking that has occurred.

The rate of release of drug molecules trapped within the matrix is ​​controlled by controlling the rate of hydrolytic breakdown of the gel. The rate of hydrolytic decomposition of the gel can be adjusted by controlling the degree of binding of the PEGs that form the hydrogel matrix. A multi-branch PEG that has 10 branches will break down and release drug molecules more slowly than a 3-branch PEG.

Substantially precise control of the rate of hydrolytic decomposition and drug release can be provided by varying the atoms near the hydrolysis-unstable bonds. Typically, increasing the value of n (the number of methylene groups) in the above structure decreases the rate of hydrolysis of the esters, and increases the time required for gel degradation. If n in the example above is 1, then the ester linkages in the gel will hydrolyze with a half-life of approximately 4 days at pH 7 and 37 ° C.

ES2 281 925 Τ5

If η is 2, then the half-life of the hydrolytic degradation of the ester bonds is approximately 43 days at pH 7 and 37 ° C.

Phosphate esters, acetals, and other hydrolytically unstable bonds can be formed in a similar manner, and the rate of hydrolysis can be similarly controlled by controlling the number of methylene groups adjacent to the hydrolytically unstable bond, and by controlling the degree of branching of the PEG.

The degradable hydrogels of this invention can also be made by a two-stage process. In the first stage, soluble, non-crosslinked PEGs are prepared which have hydrolysis-unstable bonds in their backbones. In the second stage, these PEGs with hydrolysis-unstable bonds in their main chains are joined together with other PEGs by hydrolysis-stable bonds. For example, the following PEGs have two hydrolysis-unstable ester bonds in their backbones:

NHS-O2C-CH2-O-PEG-O-CH2-CO2-PEG-O2C-CH2-O-PEG-O-CH2-CO2-NHS

The above PEG is activated at each end with an N-hydroxysuccinimide (NHS) residue, where the active succinimidyl ester residue is NHS-CO2-, and is reactive towards amino groups. When this PEG is attached to a multi-branched PEG-amine, a cross-linked network is formed, which is held together by stable amide bonds, which are formed by the reaction of active esters with amine, and by unstable ester bonds. against hydrolysis already present in the main chain. As in the previous example, the rate of degradation of the gel is controlled by varying the number of methylene groups adjacent to the ester bond.

The two-step method described above for making the PEG hydrogels can be used to form the gel, and to entrap substances in situ, in living tissue, for injectable drug systems. A drug can be combined with a reactive PEG component of the hydrogel, and injected together with another reactive PEG component that will form the gel. The drug is trapped within the matrix that is formed due to its proximity to the reactive system.

Thus, the invention provides, among other things, degradable PEG hydrogels having hydrolysis-unstable bonds, in which the rate of hydrolysis of the unstable bonds can be controlled. The PEG hydrogels of the invention can physically trap drugs, including proteins, enzymes, and a variety of other substances, in the absence of covalent bonds, for precisely controlled in vivo release. Degraded gel can be removed from the body more easily than gels that do not degrade significantly.

The foregoing and other objects, advantages, and features of the invention, and the manner in which they are carried out, will become more readily apparent upon consideration of the following detailed description of the invention, taken in conjunction with the accompanying drawing, which illustrates an illustrative embodiment.

BRIEF DESCRIPTION OF THE DRAWINGS

Figure 1 is a schematic representation of a PEG hydrogel in which the PEGs have three branches.

DETAILED DESCRIPTION

Figure 1 illustrates a polyethylene glycol (PEG) matrix that is held together by hydrolytically unstable or weak W bonds. The PEGs shown in Figure 1 have three branches. The degree of branching can be varied in the hydrogels of the invention to control the physical strength and compressibility of the gels; In general, the greater the degree of branching and the shorter the branches, the greater the strength (resistance to compression or stretching) of the gels. Similarly, higher degrees of branching and shorter branches also provide smaller pores and lower water content.

Degradable PEG hydrogels having hydrolytically unstable PEGs can be prepared in one step, as shown in the following general equation:

Z-PEG-Z + R (CH2-O-PEG-Y) p ® {R [CH2-O-PEG-W-PEG-W-] p} m where m means "matrix", and indicates a degree polymerization such that a cross-linked polymer is formed, which is a solid aggregate. m is large and indeterminate. p is 3 to 10, and refers to the degree of branching, which is the number of branches, of the reacting branched PEG, R (CH2-O-PEG-Y) p. The rate of hydrolysis of the PEG gel is typically prolonged by increasing p. R is a central branching moiety suitable for making multi-branched PEG, and includes moieties selected from the group consisting of glycerol, glycerol oligomers, pentaerythritol, sorbitol, trimethylolpropane, and di (trimethylolpropane). Z and Y are groups that react to form hydrolysis-unstable W bonds. Examples of pairs of Z and Y groups that can react to form hydrolysis-unstable W bonds include pairs selected from the group consisting of alcohol and carboxylic acid, which react to form carboxylate esters as defined in present claim 1 , hydrazide and

ES2 281 925 Τ5 aldehyde that react to form hydrazones, alcohol and phosphate that react to form phosphate esters, aldehyde and alcohol that react to form acetals, alcohols and formate that react to form orthoesters.

It should be noted that the Z groups are shown on a linear PEG, and the Y groups are shown on a branched PEG. However, the reaction will proceed, and the Y groups on linear PEG and Z groups on branched PEG will gel to form the same W weak bonds.

A specific example of the one-step method for making a PEG hydrogel having hydrolysis-unstable carboxylate ester W bonds formed by reacting PEG-carboxylic acid and PEG with hydroxylic groups Z and Y, respectively, is shown in the following equation:

HO₂C- (CH2) nO-PEG-O- (CH2) n-CO₂H + R (CH₂-O-PEG-OH) p ® ® {R [CH2-O-PEG-O2C- (CH2) nO-PEG-O (CH2) n-CO2-] p} m

In the above equation, m, p, and R are as previously characterized. n is from about 1 to 10, and can be varied to control the rate of hydrolysis of the gel. Increasing n typically decreases the rate of hydrolysis.

Note that in this example the hydroxy group is on the branched PEG, while the carboxylic acid groups are on the linear PEG. Alternatively, the hydroxy group could be on the linear PEG, while the carboxylic acid could be on the branched PEG.

The degradable PEG hydrogels can also be prepared in two steps. In the first step, a linear PEG is prepared which has one or more hydrolysis-unstable W bonds in its backbone. Linear PEG has the general formula U-PEG-W-PEG-U, where U represents a reactive terminal residue, and W is the hydrolysis-unstable bond.

In the second stage, the PEG with the hydrolysis-unstable bonds in its main chain is reacted with a second PEG. The second PEG is a branched PEG, as shown in the general formula R (CH2-O-PEG-V) p, where V represents a reactive terminal residue. P is 3 to 10, and refers to the degree of branching, which is the number of branches, of the reacting branched PEG, R (CH2-O-PEG-V) p. The rate of hydrolysis of the PEG gel is typically prolonged by increasing p. R is a central branching moiety, suitable for making multi-branched PEG, and includes moieties selected from the group consisting of glycerol, glycerol oligomers, pentaerythritol, sorbitol, trimethylolpropane, and di (trimethylolpropane).

The functional groups U and V at the ends of the PEG polymer chains, in the first and second PEG, respectively, react to form hydrolysis-stable X crosslinks, as shown in the following equation.

U-PEG-W-PEG-U + R (CH2-O-PEG-V) p ® ® {R [CH2-O-PEG-X-PEG-W-PEG-X-] p} m

Again, m means "matrix", and indicates a degree of polymerization such that a cross-linked polymer is formed, which is a solid aggregate. W is a hydrolysis-unstable group that includes carboxylate esters as defined in the present claims, hydrazones, phosphate esters, acetals, orthoesters, peptides, and oligonucleotides. U and V are groups reactive to each other, which include active esters, which include carbonate esters, which react with amines, isocyanates which react with alcohols, isocyanates which react with amines, epoxides which react with amines, and sulfonate esters which react with amines.

Hydrolysis-stable X bonds, which are formed by the reaction of U and V, include amide from the reaction of active esters with amine, urethane from the reaction of isocyanate with alcohol, urea from the reaction of isocyanate with amine, amine from the reaction of aldehyde with amine and reducing agent, amine from the reaction of epoxide with amine, and sulfonamide from the reaction of sulfonate ester with amine.

A specific example of the two-step method is the preparation of degradable PEG hydrogels that have hydrolysis-unstable carboxylate ester W bonds, and hydrolysis-stable amide X bonds, which are formed by the reaction of active esters. U and amines V, as shown in the following equation.

NHS-O2C- (CH2) nO-PEG-W-PEG-O- (CH2) n-CO2-NHS + R (CH2-O-PEG-NH2) p ® {R [CH2-O-PEG-NHCO- ( CH2) nO-PEG-W-PEG-O- (CH2) n-CONH-] p} m

The symbols n, m, p, and R are as previously described. W is a hydrolysis-unstable ester bond according to the formula -O- (CH2) r-CO2-, where r is from about 1 to 10.

ES2 281 925 Τ5

The amino group V is in the branched PEG, while the active asters U are in the linear PEG. It should be recognized that the two groups could be interchanged, such that the amino group would be present in the linear PEG, while the active aster would be present in the branched PEG.

In a second two-stage method, in a first stage a reactive linear PEG is prepared, which has W bonds unstable against hydrolysis close to the terminal groups of the polymer chain U-R '. In a second stage, the PEG having hydrolysis-unstable W bonds close to the terminal groups of the polymer chain is reacted with a branched PEG having a V reactive residue, to form hydrolysis-stable X crosslinks. .

U-R'-W-PEG-W-R'-U + R (CH₂-O-PEG-V) p ® {R [CH₂-O-PEG-X-R'-W-PEG-W-R'-X] p} m

The symbols m, p, and R are as previously defined. R 'is a small hydrocarbon fragment having from about 1 to 10 carbon atoms. W is a hydrolysis-unstable group that includes carboxylate esters as defined in the present claims, hydrazones, phosphate esters, acetals, orthoasters, peptides, and oligonucleotides, as previously defined. U and V are groups reactive with each other, which include active esters, which include carbonate esters, which react with amines, isocyanates which react with alcohols, isocyanates which react with amines, epoxides which react with amines, and sulfonate esters which react with amines.

The stable bond against hydrolysis formed by the reaction of U and V is X. X includes an amide from the reaction of an active aster with amine, a urethane from the reaction of a carboxylate aster with amine, a urethane from the reaction of an isocyanate with alcohol, a urea from reaction of an isocyanate with amine, an amine from the reaction of an aldehyde with amine and reducing agent, an amine from the reaction of an epoxide with amine, and a sulfonamide from the reaction of a sulfonate ester with amine.

A specific example, shown in the following equation, is the preparation of PEG hydrogels containing hydrolysis-unstable carboxylate aster W groups, and hydrolysis-stable X amides, formed by the reaction of active esters U and V amines, and in which the hydrolytically unstable carboxylate aster W groups have been separated from the U and / or V groups by a small hydrocarbon fragment in the precursor linear PEG.

NHS-O2C- (CH2) i-O2C- (CH2) nO-PEG-O- (CH2) n-CO2- (CH2) i-CO2-NHS + R (CH2-O-PEG-NH2) p ® {R [CH2-O-PEG-NHCO- (CH2) i-O2C- (CH2) nO-PEG-O- (CH2) n-CO2- (CH2) nCONH-] p} m

In the above equation, i is from about 1 to 10, and defines the length of the small hydrocarbon fragment R '. The symbols n, m, p and R are as previously defined. An amino group is shown in branched PEG, while active asters are shown in linear PEG. It should be recognized that the two groups could be interchanged, such that the amino group would be on the linear PEG, and the active aster would be on the branched PEG.

The skilled person should recognize that when referring to a Z moiety that reacts with a Y moiety, or a U moiety that reacts with a V moiety, that additional reagents or steps may be employed, in accordance with commonly accepted chemical procedures and standards, to achieve the desired W or X link as the case may be. There are many possible routes, too numerous to mention here, that could be taken, and which should be readily apparent to the expert. For example, one skilled in the art can hope to understand that when an alcohol and a carboxylic acid are reacted, the acid is typically converted to another form, the acid chloride, before reacting with the alcohol. Various examples are demonstrated in the following examples.

Hydrogels made from the cross-linked PEG polymeric structures of the invention can be used in drug delivery systems and for wound dressings. Wound dressings could be used internally to provide dressings that degrade in the body over time. The hydrogels of the invention could usefully be applied in burn drug delivery systems to apply therapeutic agents to burns. Drug delivery systems can be made so that the rate of hydrolysis of the hydrogel is controlled to provide a controlled release of drug components. By "drug", it is meant any substance that has the purpose of diagnosing, curing, alleviating, treating, or preventing disease in humans and other animals, or of otherwise improving physical or mental well-being. The invention could be used to administer biologically active substances, generally, having some activity or function in a living organism, or in a substance taken from a living organism.

The terms "group", "functional group", "residue", "active residue", "reactive site", and "radical" are all somewhat synonymous in chemical arts, and are used in the art and herein. invention to refer to

ES2 281 925 Τ5 definable, distinct parts or units of a molecule, and units that perform some function or activity, and are reactive towards other molecules or parts of molecules.

The term "bond" is used to refer to groups that are normally formed as a result of a chemical reaction, and are typically covalent bonds. Stable bonds against hydrolysis means that the bonds are stable in water, and do not react with water at useful pH for an extended period of time, potentially indefinitely.Hydrolysis- unstable bonds are those that react with water, typically causing the degradation of a hydrogel and the release of substances trapped within the matrix. The bond is said to be hydrolyzable and hydrolyzable. The time it takes to degrade the cross-linked polymeric structure is referred to as the rate of hydrolysis, and is usually determined in terms of its half-life.

The skilled person should recognize that when referring to a Z moiety that reacts with a Y moiety, or a U moiety that reacts with a V moiety, that additional reagents or steps may be employed, in accordance with commonly accepted chemical procedures and standards, to achieve the desired W or X link as the case may be. There are many possible routes, too numerous to mention here, that could be taken, and which should be readily apparent to the expert. For example, one skilled in the art can hope to understand that when an alcohol and a carboxylic acid are reacted, the acid is typically converted to another form, the acid chloride, before reacting with the alcohol. Various examples are demonstrated in the following examples.

The following examples show the synthesis of various examples of the invention.

EXAMPLES

Example 1

Example 1 shows the preparation of a degradable PEG hydrogel having an ester linkage unstable against hydrolysis. In a 2.54 cm (1 inch) diameter aluminum container, difunctional PEG 2000 acid (600 mg, 0.6 mmol end groups, available from Shearwater Polymers in Huntsville, Alabama), and an equivalent of 8-branch 10,000 PEG (750 mg, Shearwater Polymers) with 30 mg tin (II) 2-ethylhexanoate (Sigma Chemical), and melted. The PEG acids used included PEG-carboxymethyl acid (PEG-OCH2COOH) and PEG-propionic acid (-PEG-OCH2CH2COOH). After a thin film of the melt covered the surface of the container evenly, the container was heated under vacuum to 130^ORC and 13.3 Pa (100 millitorr) for 6-24 hours. A clear and firm gel formed. After cooling in a stream of N2, the gel became translucent, and was cut into fine discs, and purified by the following procedures.

The crude gels were swollen in glacial acetic acid, and washed three times with this solvent over a period of 2-3 days. For hydrogels with a low degree of swelling, swelling was carried out in dioxane prior to washing with glacial acetic acid, to avoid breakage of highly cross-linked gels. After washing, the gels were dried in vacuo. The tin content of the gel was determined by inductively coupled plasma spectroscopy, being less than 60 ppm.

Example 2 (not part of the invention)

Example 2 shows the preparation of a degradable PEG hydrogel having an imine linkage unstable against hydrolysis. Difunctional 3400 PEG-propionic acid aldehyde (100 mg,

58.8 mmol, Shearwater Polymers) and 10,000 8-branch PEG-amine (74 mg, 58.8 mmol) in 1,4-dioxane (Aldrich Chemical). The test tube was heated in a 70 ° C oil bath for at least two hours. The gel was then dried under reduced pressure at room temperature.

The PEG-aldehydes used included PEG-propionaldehyde (-PEG-OCH2CH2CHO), PEG-acetaldehyde (-PEGOCH2CHO), and PEG-benzaldehyde (-PEG-O-C6H4-CHO).

Examples 3 and 4, below, show the preparation of PEG derivatives having hydrolytically unstable bonds, for use in preparing the degradable hydrogel of the invention.

Example 3 (not part of the invention)

Example 3 shows the synthesis of PEG derivatives that have hydrolysis-unstable bonds in the main chain, and active NHS carbonates at each end of the main chain. The PEG derivative can be represented as NHS-OOCO-PEG-W-PEG-OCOO-NHS, where W represents the hydrolytically unstable bond. In a 100 ml round bottom flask, benzyloxy-PEGcarboxymethyl acid 3400 (3.4 g, 1 mmol, Shearwater Polymers) was azeotropically distilled from toluene for two hours, and then cooled to room temperature. A solution of thionyl chloride (2M, 4 ml, 8 mmol, Aldrich) in methylene chloride was injected, and the reaction mixture was stirred under N2 overnight. The solvent was condensed by rotary evaporation, and the syrup was dried under vacuum for about four hours over powdered P2O5.

ES2 281 925 Τ5

Anhydrous methylene chloride (5 mL) and azeotropically dried benzyloxy-PEG 3400 (2.55 g, 0.75 mmol) in toluene (20 mL) were added to the residue. After the benzyloxy-PEG-acyl chloride had dissolved, freshly distilled triethylamine (0.6 ml) was added. The reaction mixture was stirred overnight, the triethylamine salt was filtered off, and the product was collected by precipitation with ethyl ether. It was further purified by dissolution in water and extraction with methylene chloride. The organic phase was dried over anhydrous sodium sulfate, condensed in vacuo, and precipitated from ethyl ether. The precipitate was dried in vacuo. HPLC (GPC) of the product showed that the benzyloxy-PEG had been converted 100% to the PEG-ester, and that approximately 15% by weight of benzyloxy-PEG acid remained.

The mixture was purified by chromatography on an ion exchange column (DEAE-fast flow sepharose, Pharmacia), to remove benzyloxy acid-PEG. Α-benzyloxy-w-benzyloxy-PEG-ester 6800 was obtained with 100% purity. Yield: 4.1 g (80%).

A solution of α-benzyloxy-w-benzyloxy-PEG-ester 6800 (2 g, 0.59 mmol) in 1,4-dioxane (20 ml) was subjected to hydrogenolysis with H2 (202.6 kPa (2 atm) of pressure) and Pd / C (1 g, 10% Pd) overnight. The catalyst was removed by filtration, and the product precipitated in ethyl ether after most of the solvent had been removed on a rotary evaporator. Α-Hydroxy-w-hydroxy-PEG-ester 6800 was collected by filtration, and dried in vacuo. Yield: 1.5 g (75%).

Α-Hydroxy-w-hydroxy-PEG-ester 6800 (1.5 g, 0.44 mmol end group) was azeotropically dried with 100 ml of acetonitrile, and cooled to room temperature. Disucimidyl carbonate (DSC) (0.88 mmol, Fluka) and pyridine (0.1 ml) were added to this solution, and the solution was stirred at room temperature overnight. The solvent was removed in vacuo, and the syrup was dried in vacuo. The product was dissolved in 35 ml of dry methylene chloride, the insoluble solid was filtered off, and the filtrate was washed with acetate buffer saturated with sodium chloride, pH 4.5. The organic phase was dried over anhydrous sodium sulfate, condensed in vacuo, and precipitated from ethyl ether. The precipitate was dried over P2O5 in vacuo. Yield: 1.4 g (93%). NMR (DMSO-d6): (1) benzyloxy-PEG-propionic acid product: δ 3.5 (ma, PEG), 2.55 (t, -OCH2CH2COOPEG-), 4.13 (t, -PEG-COOCH2CH2O -), 4.45 (t, PEGO-CH2CH2OCO-NHS), 2.80 (s, NHS, 4H); (2) Benzyloxy-PEG-carboxymethyl acid product: δ 3.5 (ma, PEG), 4.14 (s, -OCH2COOPEG-), 4.18 (t, -OCH2COOCH2CH2-), 4.45 (t, -PEGO-CH2CH2OCO-NHS), 2.81 (s, NHS, 4H).

Example 4 (not part of the invention)

Example 4 shows the synthesis of PEG derivatives having hydrolysis-unstable linkages in the backbone, and NHS-terminal active esters. The PEG derivative can be represented by the formula NHS-OOC- (CH2) nO-PEG-W-PEG-O- (CH2) n-COONHS, where W is a hydrolytically unstable bond. In a 100 ml round bottom flask, α-hydroxy-PEG 2000 acid (4 g, 2 mmol, Shearwater Polymers) and difunctional PEG-propionic acid 2000 (4 g, 2 mmol, Shearwater Polymers) were azeotropically distilled with 70 ml toluene, low N2. After two hours, the solution was cooled to room temperature, and tin (II) 2-ethylhexanoate (200 mg, Sigma Chemical) was added. The solution was then refluxed under N2 for 24 hours. The solvent was then condensed in vacuo, and a syrup precipitated in 100 ml of ether. The product was collected by filtration, dried in vacuo, and dissolved in a pH 5 sodium acetate buffer. The slightly milky solution was centrifuged, and the upper clear solution was extracted three times with methylene chloride. The organic phase was dried over anhydrous sodium sulfate, filtered, condensed in vacuo, and precipitated from ether. The product was collected by filtration and dried in vacuo. Yield 7g (88%). HPLC: 70% product, 15% reacting diacid, and 15% monoacid. The mixture was further purified by ion exchange chromatography and gel permeation chromatography.¹H-NMR (DMSO-d6): (1) PEG-carboxymethyl acid product: δ 3.5 (ma, PEG), 4.15 (s, OCH2COOCH2-), 4.18 (t, -OCH2COOCH2CH2-); (2) PEG-propionic acid product: δ 3.5 (ma, PEG), 2.58 (t, OCH2CH2COOCH2-), 4.13 (t, -OCH2CH2COOCH2CH2-).

In a round bottom flask, difunctional acid having weak bonds (obtained from the previous step) (2 g, approximately 1 mmol end group) and N-hydroxysuccinimide (NHS) (126 mg, 1.05 mmol) were dissolved in 50 ml of dry methylene chloride. Dicyclohexylcarbodiimide (240 mg, 1.15 mmol) in 5 ml of dry methylene chloride was added to this solution. The reaction mixture was stirred under N2 overnight. The solvent was condensed, and the syrup was redissolved in 15 ml of anhydrous toluene. The insoluble salt was removed by filtration, and the filtrate waters were precipitated in 200 ml of dry ethyl ether. The precipitate was collected by filtration, and dried in vacuo. Yield 1.88g (94%).¹H-NMR (DMSO-d6): δ 3.5 (ma, PEG), 2.8 (s, NHS, 4H), 4.6 (s, -PEG-O-CH2-COONHS) or 2.85 ( t, PEG-O-CH2CH2-COONHS).

Example 5 (not part of the invention)

Example 5 shows the preparation of a degradable PEG hydrogel, from branched PEG-amine and PEG derivatives manufactured according to example 3, in which the PEG derivatives have hydrolysis-unstable linkages in the main chain , and NHS-terminal active carbonates, which can be represented as NHS-OOCO-PEG-W-PEG-OCOO-NhS. In a test tube, 100 mg (4.7

ES2 281 925 Τ5 pmol) of difunctional active PEG-carbonate 6800 (NHS-OOCO-PEG-W-PEG-OCOO-NHS, prepared in example 3) in 0.75 ml of water, and a buffered solution (phosphate 0.1 M, pH 7) of 0.15 ml of 10,000 8-arm PEGamine (250 mg / ml). After rapid stirring, it was allowed to settle, and a gel formed within a few minutes. A suitable buffer pH range was found to be 5.5 to 8.

Example 6 (not part of the invention)

Example 6 shows the preparation of degradable PEG hydrogels from branched PEG-amine and PEG derivatives made according to Example 4, in which the PEG derivatives have hydrolysis-unstable linkages in the backbone, and NHS-terminal active carbonates which can be represented as NHS-OOC- (CH2) nO-PEG-W-PEG-O- (CH2) n-COO-NHS. 100 mg (approximately 50 pmol) of difunctional active PEG-aster (NHs-OOC- (CH2) nO-PEG-W-PEG-O- (CH2) n-COONHS, prepared in Example 4) was dissolved in 0.75 ml of water, and a buffered solution (0.1 M phosphate, pH 7) of 0.25 ml of 10,000 8-branch PEG-amine (250 mg / ml) was added. After rapid stirring, it was allowed to settle, and a gel formed within a few minutes. A suitable buffer pH range was found to be 5.5 to 8.

Example 7

The example shows the synthesis of difunctional PEG-hydroxybutyric acid (HBA), which can be represented as HOOC-CH2-CH (CH3) -OOC- (CH2) nO-PEG-O- (CH2) n-COOCH (CH3) CH2- COOH, for use in preparing the reactive PEGs of Example 8. PEG 2000 acid (2.0 g, 1 mmol, carboxymethyl acid (CM) or propionic acid (PA)) was azeotropically dried with 60 ml of toluene under N2 . After two hours, the solution was cooled to room temperature and thionyl chloride (3 mL, 6 mmol, in CH2Cl2) was added. The reaction mixture was then stirred at room temperature overnight, and the solution was condensed by rotary evaporation. The residue was dried in vacuo for approximately four hours with powdered P2O5. 3-Hydroxybutyric acid (0.30 g, 2.7 mmol) was azeotropically dried with 70 ml of 1,4-dioxane, until approximately 20 ml of solution remained. The solution was then cooled to room temperature under N2, and dried PEG-acyl chloride from the previous step was added. After the PEG dissolved, 0.6 ml of dry triethylamine was injected into the system, and the reaction mixture was stirred overnight. The salt was filtered from the solution, the solvent was condensed on a rotary evaporator, and the syrup was dried in vacuo. The crude product was dissolved in 100 ml of distilled water, and the pH was adjusted to 3.0. The product was extracted three times with a total of 80 ml of methylene chloride. The organic phase was dried over anhydrous sodium sulfate, filtered, condensed in vacuo, and precipitated in 100 ml of ethyl ether. The product was collected by filtration and dried in vacuo. Yield 1.84g (92%).¹H-NMR (DMSO-d6): δ 3.5 (ma, PEG), 2.54 (d, PEGCOOCH (CH3) CH2-COOH), 5.1 (h, PEGCOOCH (CHa) CH₂-COOH), 1.21 (d, PEG-COOCH (CHa) CH2-COOH), 2.54 (t, PEGOCH2CH2COO (PA)), 4.05 (s, PEGOCH2COO (CM)).

Example 8

Example 8 shows the synthesis of the difunctional PEG-HBA-NHS double stanchion, which can be represented as NHSOOC-CH2-CH (CH3) -OOC- (CH2) nO-PEG-O- (CH2) n-COOCH (CH3) CH2COONHS , for use in the preparation of PEG hydrogels of the invention. PEG-3-butyric acid (1 g, approximately 0.5 mmol, prepared in Example 7) and 64 mg of N-hydroxysuccinimide (NHS) (0.53 mmol) were dissolved in 30 ml of dry methylene chloride, followed by the addition of dicyclohexylcarbodiimide (DCC, 126 mg, 0.6 mmol) in 5 ml of dry methylene chloride. The solution was stirred under nitrogen overnight, and the solvent was removed by rotary evaporation. The residue was stirred with 10 ml of dry toluene at 45^ORC, and the insoluble solid was removed by filtration. The product precipitated in 100 ml of dry ethyl ether, and the precipitate was collected by filtration, and dried in vacuo. Yield 0.94g (94%).¹H-NMR (DMSO-d6): δ 3.5 (ma, PEG), 3.0-3.2 (m, -COOCH (CHa) CH2-COONHS), 5.26 (h, -COOCH (CHa) CH₂-COONHS), 1.3 (d, -CO-OCH (CHa) CH2-COONHS), 2.54 (t, -PEGOCH2CH2COO- (PA)), 4.1 (s, -PEGOCH2COO- (CM)).

Example 9

Example 9 shows the preparation of a degradable PEG hydrogel from a branched PEG-amine and the double ester PEG-HBA-nHs from Example 8, which can be represented as NHS-OOC-CH2-CH (CH3) -OOC- (CH2) n-OPEG-O- (CH2) n-COOCH (CHa) CH2COONHS. The PEG-HBA-NHS 2000 double aster (100 mg, approximately 0.1 mmol, Example 8) was dissolved in 0.5 ml of water and a 10,000 8-branch PEG-amine buffered solution (0.5 ml , 250 mg / ml). After rapid stirring, it was allowed to settle, and a gel formed within a few minutes. A suitable buffer pH range was found to be 5.5 to 8.