WO2006125765A2

WO2006125765A2 - Acylated insulin with high purity

Info

Publication number: WO2006125765A2
Application number: PCT/EP2006/062495
Authority: WO
Inventors: Jan Markussen; Svend Havelund; Jes Kristian Jacobsen; Aage Hvass; Ib Jonassen; Georg Wilhelm Jensen
Original assignee: Novo Nordisk A/S
Priority date: 2005-05-26
Filing date: 2006-05-22
Publication date: 2006-11-30
Also published as: WO2006125765A3; EP1888104A2

Abstract

The present invention is related to a highly purified pharmaceutical preparation comprising an acylated insulin where the acyl group is attached via a linker molecule to the ϵ- amino group of the B29Lys residue of the parent insulin moiety. The insulin pharmaceutical formulation comprises less than about 3% weight/weight of an isomer of the acylated insulin.

Description

ACYLATED INSULIN WITH HIGH PURITY

FIELD OF THE INVENTION

The present invention relates to highly purified acylated insulins which are soluble at physiological pH values and have a prolonged profile of action. The invention also relates to methods of providing such acylated insulins, to pharmaceutical compositions containing them and to the use of such acylated insulins in the treatment of diabetes and hyperglycaemia.

BACKGROUND OF THE INVENTION The most intensive therapy of diabetic patients with insulin requires two types of insulin products with markedly different kinetic profiles, a rapidly absorbed insulin to cope with the rise in blood glucose after meals and a long-acting insulin which provides a constant supply of insulin to the circulation to cover the needs for basal insulin in the fasting states between meals. Most conveniently, such long-acting insulin products have a protracted absorp- tion enabling the steady supply of insulin by a single daily injection. One concept of prolongation of the action of insulin products known in the art has been the use of crystalline suspensions of insulin, either co-crystallized with protamine known as NPH insulin, or bovine insulin crystallized with zinc known as Ultralente^® . An insulin product soluble at pH about 4 but insoluble at pH 7, whereby a subcutaneous precipitation occurs after injection, has been intro- duced recently and known as Lantus^®. However, the most predictable and reproducible day- to-day delivery of basal insulin to the circulation is obtained using soluble insulin derivatives that remain soluble after subcutaneous injection. Two mechanisms have been reported whereby insulin derivatives, which remain in solution after injection, become protracted, namely albumin binding and formation of high molecular assemblies of hexamers of the insu- Nn derivatives. An insulin product soluble at pH about 7 which remains in solution after subcutaneous injection have been introduced recently and known as Levemir^®. This insulin derivative is acylated in position B29 with a fatty acid and lacks the B30 amino acid residue, vide EP patent No. 0792290.

WO 99/21888 discloses insulin derivatives prone to aggregate forming high molecu- lar weight assemblies and a gel permeation method (SEC) to access this property. SUMMARY OF THE INVENTION

In one aspect the present invention is related to a pharmaceutical composition com- prising an acylated insulin and an isomer thereof in a weight ratio of greater than 97:3 and pharmaceutically acceptable adjuvants, wherein said acylated insulin has the formula I

G I V

X1 X2 X3

wherein X at position A 18 is Asn or GIn, X1 in position B1 is Phe or deleted, X2 at position B2 is VaI or deleted, X3 at position B3 is Asn or modified to Thr, GIn, GIu or Asp, X4 at posi- tion A21 is Asn or modified to Ala, GIn, GIu, GIy, His, He, Leu, Met, Ser, Thr, Trp, Tyr or VaI, X5 at position B30 may be any codable amino acid residue except Lys, Arg and Cys or is deleted, Acyl is an acyl group derived from the group consisting of mono- or dicarboxylic, unsaturated or saturated fatty acids with a chain length of from about 6 to about 40, lithocholic acids, cholic acid, hyocholic acid, deoxycholic acid, chenodeoxycholic acid, ursodeoxycholic acid, hyodeoxycholic acid and cholanic acid, and wherein the isomer has the formula Il

wherein X-X5 and Acyl have the above meanings.

In one embodiment the weight ratio of the acylated insulin to the isomer thereof is greater than about 98:2.

In another embodiment the weight ratio of the acylated insulin to the isomer thereof is greater than about 99:1 .

In a still further embodiment the weight ratio of the acylated insulin to the isomer thereof is greater than 99.5:0.5.

In a still further embodiment the weight ratio of the acylated insulin to the isomer thereof is greater than 99.8:0.2. In a still further embodiment the weight ratio of the acylated insulin to the isomer thereof is greater than 99.9:01 .

In one embodiment the weight ratio is in the range from 97:3 to 99.9:0.1.

In another embodiment the weight ratio is in the range from 98:2 to 99.9:0.1 .

In another embodiment the weight ratio is in the range from 99:1 to 99.9:0.1 . In another embodiment the weight ratio is in the range from 97:3 to 99.5:0.5.

In another embodiment the weight ratio is in the range from 98:2 to 99.5:0.5.

In another embodiment the weight ratio is in the range from 99:1 to 99.5:0.5.

In one embodiment X is Asn.

In another embodiment X1 is Phe. In another embodiment X2 is VaI.

In another embodiment X3 is Asn.

In a further embodiment X4 is Asn.

In a still further embodiment X5 is deleted or is Thr.

In a still further embodiment X is Asn, X1 is Phe, X2 is VaI, X3 is Asn, X4 is Asn and X5 is deleted. In one embodiment the acyl group is a mono- or dicarboxylic, saturated or unsaturated fatty acid with a chain length of from about 6 to about 24 carbon atoms.

In another embodiment the acyl group is a mono- or dicarboxylic, saturated or unsaturated fatty acid with a chain length from 14-16 carbon atoms. In another embodiment the acyl group is lithocholic acid. In this embodiment the pharmaceutical formulation may contain the insulin derivative Lys^B29(N^ε— lithocholyl-γ- glutamyl)des(B30) human insulin and the isomer thereof is Lys^B29(N^ε-lithocholyl-α- glutamyl)des(B30) human insulin.

In another aspect the present invention is related to a method for producing a solu- tion containing an acylated insulin with formula I and an isomer thereof with formula Il in a weight ratio of greater than about 97:3, said method comprising a) subjecting a solution with a weight ratio of the acylated insulin with formula I and the isomer thereof with formula Il of less than 97:3 to ion exchange or RP-HPLC chromatography under conditions effective to separate the isomer from the acylated insulin with formula I, and b) collecting the fractions from said chromatography containing said acylated insulin and the isomer thereof in a weight ratio greater than 97:3.

The pH may vary under the separation process depending on the insulin compound in question. In one embodiment the pH of the RP-HPLC chromatographic step is be- tween about 3 and about 7. More typically the pH of this step will be between about 4 and about 7, between about 5 and about 6.5, between about 4 and about 6, between about 4 and about 5, between about 3.5 and about 6, between about 4.5 and about 6.5 or between about 4.75 and about 6.5

In one embodiment of the invention the RP-HPLC step is conducted at pH of about 2.5 to about 5 in water-acetonitrile or water-ethanol mixtures.

The temperature of the chromatographic step may also vary but will typically be between about 15 and about 5O⁰C.

In one embodiment the temperature of the chromatographic step is between about 20 and about 45⁰C. In another embodiment the temperature of the chromatographic step is between about 25 and about 45⁰C, between about 20 to about 45⁰C, between about 25 and about 45⁰C, between about 30 to about 45⁰C or between about 35 and about 4O⁰C.

The RP-HPLC step is typically conducted in water-acetonitrile or water-ethanol mixtures. In one embodiment the solvent in the RP-HPLC step will comprise a salt such as Na₂SO₄, (NhU)₂SO₄, NaCI, KCI, and buffer systems such as phosphate, and citrate and maleic acid. The required concentration of salt in the solvent may be from about 0.1 M to about 1 M, preferably between 0.2 M to 0.5 M, most preferable between 0.3 to 0.4 M. In- crease of the concentration of salt requires an increase in the concentration of organic solvent in order to achieve elution from the column within a suitable time.

In another embodiment the principles of RP-HPLC and ion exchange chromatography can be combined, e.g. by using a silica matrix only partially substituted with the organic ligand leaving free silanol sites capable of binding cat-ions. Eluation from such columns us- ing the combination of binding principles typically requires higher concentrations of salts and organic solvent as compared to using the principle separately on separate columns.

The temperature of the chromatographic step will typically be between about 15 and about 5O⁰C.

In one embodiment the temperature will be in the range of about 20 to about 45⁰C. In another embodiment the temperature will be in the range of about 25 to about

45⁰C, or from about 30 to about 45⁰C or from about 35 to about 4O⁰C.

In another aspect of the present invention the pharmaceutical preparation will further comprise a rapid acting insulin

In a still further aspect the present invention is related to a method for treating hy- perglycemia in a patient, said method comprising administering to a patient in need of such treatment an effective amount of a pharmaceutical composition according to the invention.

In one embodiment the invention is related to a method of treating type 1 diabetes, type 2 diabetes and other states that are associated with hyperglycaemia in a patient, comprising administering to the patient in need of such a treatment a therapeutically effective amount of the pharmaceutical preparation according to the invention.

In another embodiment the insulin is a desB30 insulin

In another embodiment the acyl group is a lithocholic acid selected from 5-α litho- cholic acid or 5-β lithocholic acid.

In one embodiment the invention is related to a solution comprising Lys^B29(N^ε-lithocholyl-γ-glutamyl) des(B30) human insulin and Lys^B29(N^ε-lithocholyl-α- glutamyl) des(B30) human insulin in which the content of Lys^B29(N^ε-lithocholyl-α-glutamyl) des(B30) human insulin is less than about 3 % weight/weight as compared to Lys^B29(N^ε-lithocholyl-γ-glutamyl) des(B30) human insulin.

In another embodiment the invention is related to a solution comprising Lys^B29(N^ε-lithocholyl-γ-glutamyl) des(B30) human insulin and Lys^B29(N^ε-lithocholyl-α- glutamyl) des(B30) human insulin in which the content of of Lys^B29(N^ε-lithocholyl-α-glutamyl) des(B30) human insulin is less than 2 % weight/weight as compared to Lys^B29(N^ε-lithocholyl- γ-glutamyl) des(B30) human insulin.

In a further embodiment the invention is related to a solution comprising Lys^B29(N^ε-lithocholyl-γ-glutamyl) des(B30) human insulin and Lys^B29(N^ε-lithocholyl-α- glutamyl) des(B30) human insulin in which the content of Lys^B29(N^ε-lithocholyl-α-glutamyl) des(B30) human insulin is less than 1 % weight/weight as compared to Lys^B29(N^ε-lithocholyl- γ-glutamyl) des(B30) human insulin.

In a further embodiment the invention is related to a solution comprising Lys^B29(N^ε-lithocholyl-γ-glutamyl) des(B30) human insulin and Lys^B29(N^ε-lithocholyl-α- glutamyl) des(B30) human insulin in which the content of Lys^B29(N^ε-lithocholyl-α-glutamyl) des(B30) human insulin is less than 0.5 % weight/weight as compared to Lys^B29(N^ε-lithocholyl-γ-glutamyl) des(B30) human insulin.

In a further embodiment the invention is related to a solution comprising Lys^B29(N^ε-lithocholyl-γ-glutamyl) des(B30) human insulin and Lys^B29(N^ε-lithocholyl-α- glutamyl) des(B30) human insulin in which the content of Lys^B29(N^ε-lithocholyl-α-glutamyl) des(B30) human insulin is less than 0.1 % weight/weight as compared to Lys^B29(N^ε-lithocholyl-γ-glutamyl) des(B30) human insulin.

In a further embodiment the invention is related to a solution of Lys^B29(N^ε-lithocholyl- γ-glutamyl) des(B30) in which the sum of all impurities is less than 1 % based on total protein.

In a further aspect the pharmaceutical preparation according to the invention may be used in mixture with a rapid acting insulin either as a fixed, preprepared mixture or in a treatment comprising administrating the prolonged and the rapid acting insulin separately. If the active components are administered separately, they can be administered at the same time or at separate times.

DETAILED DESCRIPTION OF THE INVENTION

The acyl group attached to the parent insulin molecule may be a lipophilic group containing from 6 to 40 carbon atoms. Examples of such groups are fatty dicarboxylic or monocarboxylic groups having from 6 to 40, from 6 to 36, from 6 to 24, from 6 to 18, from 8 to 36, from 8 to 24, from 8 to 20, from 8 to 18, from 12 to 18 or from 14 to 18 carbon atoms.

In one embodiment the acyl group is selected from the following group: CH₃-(CH₂)_n- CO- , (COOH)-(CH₂)_n-CO-, (NH₂-CO)-(CH₂)_n-CO-, HO-(CH₂)_n-CO- , where 4<n<38. In another embodiment the acyl group is 5-α lithocholic acid or 5-β lithocholic acid.

In another embodiment the acyl group is 5-α or 5-β isomers of cholic acid, hyocholic acid, deoxycholic acid, chenodeoxycholic acid, ursodeoxycholic acid, hyodeoxycholic acid or cholanic acid. In another embodiment the acyl group is fusidic acid, a fusidic acid derivative or gly- cyrrhetinic acid.

The dicarboxylic fatty acid will typically comprise from about 4 to about 26, from 4 to about 18, from about 6 to about 18, from about 8 to about 16, from about 8 to about 22, from about 8 to about 17, from about 8 to about 15, from about 10 to about 18, from about 10 to about 16 and from about 6 to about 17 carbon atoms in the carbon chain.

Examples of the dicarboxylic fatty acids are diacids with the formula HOOC-(CH₂)_r1- COOH, where r1 is 4 to 22

Non limiting examples of dicarboxylic fatty acids are α,ω- tetradecanedioic acid, α,ω- hexadecanedioic acid, and α,ω- octadecanedioic acid. The acyl group is connected to the parent insulin molecule via an amide bond using an amino acid linker having a free carboxylic acid group such as glutamic acid or aspartic acid or a peptide chain of up to four amino acid residues comprising at least one amino acid residue with a free carboxylic acid. The linker amino acid may be in the L-or D-form.

Non limiting examples of acylated insulin derivatives with formula I are: Lys^B29(N^ε-lithocholyl-γ-glutamyl) des(B30) human insulin;

Lys^B29(N^ε-lithocholyl-γ-glutamyl) GlyA21 des(B30) human insulin; Lys^B29(N^ε-lithocholyl-γ-glutamyl) GlyA21 GlnB3 des(B30) human insulin; Lys^B29(N^ε-lithocholyl-γ-glutamyl) GlnB3 des(B30) human insulin; Lys^B29(N^ε-lithocholyl-γ-glutamyl) human insulin; Lys^B29(N^ε-lithocholyl-γ-glutamyl) GlyA21 human insulin;

Lys^B29(N^ε-lithocholyl-γ-glutamyl) GlyA21 GlnB3 human insulin; Lys^B29(N^ε-lithocholyl-γ-glutamyl) GlnB3 human insulin;

N^εB29-(N^α-(HOOC(CH₂)₁₄CO)-γ-Glu) des(B30) human insulin; N^εB29-(N^α-(HOOC(CH₂)₁₅CO)-γ-Glu) des(B30) human insulin; N^εB29-(N^α-(HOOC(CH₂)₁₆CO)-γ-Glu) des(B30) human insulin; N^εB29-(N^α-(HOOC(CH₂)₁₇CO)-γ-Glu) des(B30) human insulin; N^εB29-(N^α-(HOOC(CH₂)₁₈CO)-γ-Glu) des(B30) human insulin; N^εB29-(N^α-(HOOC(CH₂)₁₄CO)-γ-Glu) human insulin; N^εB29-(N^α-(HOOC(CH₂)₁₅CO)-γ-Glu) human insulin;

N^εB29-(N^α-(HOOC(CH₂)₁₆CO)-γ-Glu) human insulin;

N^εB29-(N^α-(HOOC(CH₂)₁₇CO)-γ-Glu) human insulin;

N^εB29-(N^α-(HOOC(CH₂)₁₈CO)-γ-Glu) human insulin; N^εB29-(N-tridecanoyl-γ-Glu) des(B30) human insulin;

N^εB29-(N-tetradecanoyl-γ-Glu) des(B30) human insulin;

N^εB29-(N-decanoyl -γ-Glu) des(B30) human insulin;

N^εB29-(N-dodecanoyl-γ-Glu) des(B30) human insulin;

N^εB29-(N-tridecanoyl-γ-Glu) GlyA21 des(B30) human insulin; N^εB29-(N-tetradecanoyl -γ-Glu) GlyA21 des(B30) human insulin;

N^εB29-(N-decanoyl-γ-Glu) GlyA21 des(B30) human insulin;

N^εB29-(N-dodecanoyl-γ-Glu) GlyA21 des(B30) human insulin;

N^εB29-(N-tridecanoyl-γ-Glu) GlyA21 GlnB3 des(B30) human insulin;

N^εB29-(N-tetradecanoyl-γ-Glu) GlyA21 GlnB3 des(B30) human insulin; N^εB29-(N-decanoyl-γ-Glu) GlyA21 GlnB3 des(B30) human insulin;

Nε^B29 -(N-dodecanoyl-γ-Glu) GlyA21 GlnB3 des(B30) human insulin;

N^εB29-(N-tridecanoyl-γ-Glu) GlnB3 des(B30) human insulin;

N^εB29-(N-tetradecanoyl-γ-Glu) GlnB3 des(B30) human insulin;

N^εB29-(N-decanoyl-γ-Glu) GlnB3 des(B30) human insulin; N^εB29-(N-dodecanoyl-γ-Glu)GlnB3 des(B30) human insulin;

N^εB29-(N-tridecanoyl-γ-Glu) GlyA21 human insulin;

N^εB29-(N-tetradecanoyl-γ-Glu) GlyA21 human insulin;

N^εB29-(N-decanoyl-γ-Glu) GlyA21 human insulin;

N^εB29-(N-dodecanoyl-γ-Glu) GlyA21 human insulin; N^εB29-(N-tridecanoyl-γ-Glu) GlyA21 GlnB3 human insulin;

N^εB29-(N-tetradecanoyl-γ-Glu) GlyA21 GlnB3 human insulin;

N^εB29-(N-decanoyl-γ-Glu) GlyA21 GlnB3 human insulin; N^εB29-(N-dodecanoyl-γ-Glu) GlyA21 GlnB3 human insulin; N^εB29-(N-tridecanoyl-γ-Glu) GlnB3 human insulin; N^εB29-(N-tetradecanoyl-γ-Glu) GlnB3 human insulin; N^εB29-(N-decanoyl-γ-Glu) GlnB3 human insulin, and N^εB29-(N-dodecanoyl γ-Glu) GlnB3 human insulin.

In the synthesized Lys^B29(N^ε-lithocholyl-γ-glutamyl) des(B30) human insulin a contamination of the corresponding α-glutamyl isomer, in the range of 3.0-3.6 %, has been identified.

The Lys^B29(N^ε-lithocholyl-γ-glutamyl) des(B30) human insulin derivative may be syn- thesized by acylation of des(B30) human insulin using a lithocholyl-Glu derivative, in which the α-carboxyl group may be protected in the form of an ester and the γ-carboxyl group may be activated in the form of an active ester or as an active amide. The conditions for the selective acylation of the ε-amino group of the lysine B29 residue of insulin with fatty acids are disclosed in US patent No. 5,646,242 and US patent No. 5,905,140. Table 1 shows that the fraction of high molecular assemblies in mixtures of the α- glutamyl isomer and Lys^B29(N^ε-lithocholyl-γ-glutamyl) des(B30) human insulin is highly dependent of the content of the α-glutamyl isomer when using the SEC method described in PCT WO 99/21888, in this case adapted to a smaller column.

Thus, in a sample of the Lys^B29(N^ε-lithocholyl-γ-glutamyl) des(B30) human insulin derivative that contains 3.5 % of the α-glutamyl isomer as compared to the gamma-glutamyl isomer on a weight/weight basis only 66.7 % of the acylated insulin was found in the high molecular part of the chromatogram, defined as the fraction eluting before aldolase in SEC

Surprisingly, the small content of the α-glutamyl isomer as compared to the gamma- glutamyl isomer has a marked ability to decrease the percentage of the high molecular frac- tion estimated by SEC, much higher than the percentage of α-glutamyl isomer would suggest. For example, the percentage of self-assembled hexamers increases by 10%, from 67 to 77%, when the content of the α-glutamyl isomer in the mixture is lowered only 3.5%, from 3.5 to 0 %.

The increased tendency to form high molecular assemblies in purified Lys^B29(N^ε-lithocholyl-γ-glutamyl) des(B30) human insulin results in increased prolongation of the absorption time after subcutaneous injection. Consequently, as the content of the α- glutamyl isomer is lowered towards zero the possible variation in the day-to-day pharmacokinetics of the absorption, due to variations in the α-glutamyl isomer, will decrease accordingly. The pharmaceutical preparations according to the present invention will comprise less that 3 percent by weight of the unwanted isomer as compared to the desired compound. Typically it will contain less than 2 percent by weight, more typically less 1 percent by weight and the content of the isomer may be as low as 0.5 to 0.05 percent by weight as compared to the desired acylated insulin.

In one embodiment the content of the alpha isomer may even be zero.

The insulin moiety - in the present text also referred to as the parent insulin - of an insulin according to the invention can be a naturally occurring insulin such as human insulin or porcine insulin. Alternatively, the parent insulin can be an insulin analogue. In one group of parent insulin analogues, the amino acid residue at position A21 is

Asn.

In another group of parent insulin analogues, the amino acid residue at position B1 has been deleted. A specific example from this group of parent insulin analogues is desB1 human insulin. In another group of parent insulin analogues, the amino acid residue at position B30 has been deleted. A specific example from this group of parent insulin analogues is desB30 human insulin.

In another aspect the present invention is related to a method for producing a solution containing an acylated insulin with formula I and an isomer thereof with formula Il in a weight ratio of greater than about 97:3, said method comprising a) subjecting a solution with a weight ratio of the acylated insulin with formula I and the isomer thereof with formula Il of less than 97:3 to ion exchange or RP-HPLC chromatography under conditions effective to separate the isomer from the acylated insulin with formula I, and b) collecting the fractions from said chromatography containing said acylated insulin and the isomer thereof in a weight ratio greater than 97:3.

In one embodiment the method according to the present invention may comprise a step wherein the weight ratio between the acylated insulin derivative and the isomer thereof in fractions obtained from step a) is determined and compared to the corresponding weight ratio in the solution before step a) where an increase in the weight ratio in said fractions indicates that the fractions contain a solution of an acylated insulin that exhibits an increased percentage of self-assembled hexamers as compared to the solution of said acylated insulin before step a).

In another embodiment, the method according to the present invention may fur- thermore comprise a step wherein the weight ratio between the acylated insulin derivative and the isomer thereof in fractions obtained from step a) is determined and compared to the corresponding weight ratio in the solution before step a) where an increase in the weight ratio in said fractions indicates that the fractions contain a solution of an acylated insulin that exhibits less day to day variation in pharmacokinetics upon administration to humans as compared to that exhibited by the solution of acylated insulin before step a).

The present invention is also related to a method for lowering blood glucose in a human while minimizing variation of levels of blood glucose in said human from day to day, said method comprising administering to said human the pharmaceutical composition of the invention in an amount effective to lower said blood glucose and where administration of the pharmaceutical composition of the invention produces less variation in levels of blood glucose in said human from day to day as compared to the variation in blood glucose levels in said human from day to day produced by administration to said human of a second pharmaceutical composition wherein the weight ratio of gamma isomer to alpha isomer is equal to or less than 97:3. The starting product for the acylation, the parent insulin or parent insulin analogue or a precursor thereof can be produced by either well-know organic synthesis or by well known recombinant production in suitable transformed microorganisms. Thus the insulin starting product can be produced by a method which comprises culturing a host cell containing a DNA sequence encoding the polypeptide and capable of expressing the polypeptide in a suitable nutrient medium under conditions permitting the expression of the peptide, after which the resulting peptide is recovered from the culture. As an example desB(30) human insulin can be produced from a human insulin precursor B(1 -29)-Ala-Ala-Lys-A(1 -21 ) which is produced in yeast as disclosed in US patent No. 4916212. This insulin precursor can then be converted into desB30 human insulin by ALP cleavage of the Ala-Ala-Lys peptide chain to give desB30 human insulin which can then be acylated to give the present insulin derivatives.

The medium used to culture the cells may be any conventional medium suitable for growing the host cells, such as minimal or complex media containing appropriate supplements. Suitable media are available from commercial suppliers or may be prepared accord- ing to published recipes (e.g. in catalogues of the American Type Culture Collection). The peptide produced by the cells may then be recovered from the culture medium by conventional procedures including separating the host cells from the medium by centrifugation or filtration, precipitating the proteinaceous components of the supernatant or filtrate by means of a salt, e.g. ammonium sulphate, purification by a variety of chromatographic procedures, e.g. ion exchange chromatography, gel filtration chromatography, affinity chromatography, or the like, dependent on the type of peptide in question.

The DNA sequence encoding the parent insulin may suitably be of genomic or cDNA origin, for instance obtained by preparing a genomic or cDNA library and screening for DNA sequences coding for all or part of the polypeptide by hybridisation using synthetic oligonucleotide probes in accordance with standard techniques (see, for example, Sambrook, J, Fritsch, EF and Maniatis, T, Molecular Cloning: A Laboratory Manual, Cold Spring Harbor Laboratory Press, New York, 1989). The DNA sequence encoding the polypeptide may also be prepared synthetically by established standard methods, e.g. the phosphoamidite method described by Beaucage and Caruthers, Tetrahedron Letters 22 (1981 ), 1859 - 1869, or the method described by Matthes et al., EMBO Journal 3 (1984), 801 - 805. The DNA sequence may also be prepared by polymerase chain reaction using specific primers, for instance as described in US 4,683,202 or Saiki et al., Science 239 (1988), 487 - 491.

The DNA sequence may be inserted into any vector which may conveniently be sub- jected to recombinant DNA procedures, and the choice of vector will often depend on the host cell into which it is to be introduced. Thus, the vector may be an autonomously replicating vector, i.e. a vector which exists as an extrachromosomal entity, the replication of which is independent of chromosomal replication, e.g. a plasmid. Alternatively, the vector may be one which, when introduced into a host cell, is integrated into the host cell genome and repli- cated together with the chromosome(s) into which it has been integrated.

The vector is preferably an expression vector in which the DNA sequence encoding the peptide is operably linked to additional segments required for transcription of the DNA, such as a promoter. The promoter may be any DNA sequence which shows transcriptional activity in the host cell of choice and may be derived from genes encoding proteins either homologous or heterologous to the host cell. Examples of suitable promoters for directing the transcription of the DNA encoding the parent insulin in a variety of host cells are well known in the art, cf. for instance Sambrook et al., supra.

The DNA sequence encoding the parent insulin may also, if necessary, be operably connected to a suitable terminator, polyadenylation signals, transcriptional enhancer se- quences, and translational enhancer sequences. The recombinant vector of the invention may further comprise a DNA sequence enabling the vector to replicate in the host cell in question.

The vector may also comprise a selectable marker, e.g. a gene the product of which complements a defect in the host cell or one which confers resistance to a drug, e.g. ampicil- Nn, kanamycin, tetracyclin, chloramphenicol, neomycin, hygromycin or methotrexate. To direct a peptide into the secretory pathway of the host cells, a secretory signal sequence (also known as a leader sequence, prepro sequence or pre sequence) may be provided in the recombinant vector. The secretory signal sequence is joined to the DNA sequence encoding the peptide in the correct reading frame. Secretory signal sequences are commonly positioned 5' to the DNA sequence encoding the peptide. The secretory signal sequence may be that normally associated with the peptide or may be from a gene encoding another secreted protein.

The procedures used to ligate the DNA sequences coding for the parent insulin, the promoter and optionally the terminator and/or secretory signal sequence, respectively, and to insert them into suitable vectors containing the information necessary for replication, are well known to persons skilled in the art (cf., for instance, Sambrook et al., supra).

The host cell into which the DNA sequence or the recombinant vector is introduced may be any cell which is capable of producing the parent insulin and includes bacteria, yeast, fungi and higher eukaryotic cells. Examples of suitable host cells well known and used in the art are, without limitation, E. coli, Saccharomyces cerevisiae, or mammalian BHK or CHO cell lines.

Pharmaceutical preparations according to the invention are typically solutions and may be administered parenterally to patients in need of such a treatment. Parenteral administration may be performed by subcutaneous, intramuscular or intravenous injection by means of a syringe, optionally a pen-like syringe. Alternatively, parenteral administration can be performed by means of an infusion pump. Further options are to administer the insulin preparation nasally or pulmonally, preferably in compositions, powders or liquids, specifically designed for the purpose.

Injectable preparations of the invention can be prepared using the conventional techniques of the pharmaceutical industry which involve dissolving and mixing the ingredients as appropriate to give the desired end product. Thus, according to one procedure, an insulin derivative according to the invention is dissolved in an amount of water which is somewhat less than the final volume of the composition to be prepared. An isotonic agent, a preservative and a buffer is added as required and the pH value of the solution is adjusted - if necessary - using an acid, e.g. hydrochloric acid, or a base, e.g. aqueous sodium hydroxide as needed. Finally, the volume of the solution is adjusted with water to give the desired concentration of the ingredients.

The buffer may be selected from the group consisting of sodium acetate, sodium carbonate, citrate, glycylglycine, histidine, glycine, lysine, arginine, sodium dihydrogen phos- phate, disodium hydrogen phosphate, sodium phosphate, and tris(hydroxymethyl)- aminomethan, bicine, tricine, malic acid, succinate, maleic acid, fumaric acid, tartaric acid, aspartic acid or mixtures thereof. Each one of these specific buffers constitutes an alternative embodiment of the invention.

The preservative may be selected from the group consisting of phenol, o-cresol, m- cresol, p-cresol, methyl p-hydroxybenzoate, propyl p-hydroxybenzoate, 2-phenoxyethanol, butyl p-hydroxybenzoate, 2-phenylethanol, benzyl alcohol, chlorobutanol, and thiomerosal, bronopol, benzoic acid, imidurea, chlorohexidine, sodium dehydroacetate, chlorocresol, ethyl p-hydroxybenzoate, benzethonium chloride, chlorphenesine (3p-chlorphenoxypropane-1 ,2- diol) or mixtures thereof. In one embodiment of the invention the preservative is present in a concentration from 0.1 mg/ml to 20 mg/ml. In a further embodiment of the invention the preservative is present in a concentration from 0.1 mg/ml to 5 mg/ml. In another embodiment of the invention the preservative is present in a concentration from 5 mg/ml to 10 mg/ml. In a further embodiment of the invention the preservative is present in a concentration from 10 mg/ml to 20 mg/ml. The use of a preservative in pharmaceutical compositions is well-known to the skilled person. For convenience reference is made to Remington: The Science and Practice of Pharmacy, 19^th edition, 1995.

The isotonic agent may be selected from the group consisting of a salt (e.g. sodium chloride), a sugar or sugar alcohol, an amino acid (e.g. glycine, L-histidine, arginine, lysine, isoleucine, aspartic acid, tryptophan, threonine), an alditol (e.g. glycerol (glycerine), 1 ,2- propanediol (propyleneglycol), 1 ,3-propanediol, 1 ,3-butanediol) polyethyleneglycol (e.g. PEG400), or mixtures thereof. Any sugar such as mono-, di-, or polysaccharides, or water- soluble glucans, including for example fructose, glucose, mannose, sorbose, xylose, maltose, lactose, sucrose, trehalose, dextran, pullulan, dextrin, cyclodextrin, soluble starch, hy- droxyethyl starch and carboxymethylcellulose-Na may be used. In one embodiment the sugar additive is sucrose. Sugar alcohol is defined as a C4-C8 hydrocarbon having at least one -OH group and includes, for example, mannitol, sorbitol, inositol, galactitol, dulcitol, xyli- tol, and arabitol. In one embodiment the sugar alcohol additive is mannitol. The sugars or sugar alcohols mentioned above may be used individually or in combination. There is no fixed limit to the amount used, as long as the sugar or sugar alcohol is soluble in the liquid preparation and does not adversely effect the stabilizing effects achieved using the methods of the invention. In one embodiment, the sugar or sugar alcohol concentration is between about 1 mg/ml and about 150 mg/ml.

The isotonic agent may be present in a concentration from 1 mg/ml to 50 mg/ml, from 1 mg/ml to 7 mg/ml, from 8 mg/ml to 24 mg/ml, or from 25 mg/ml to 50 mg/ml. The use of an isotonic agent in pharmaceutical compositions is well-known to the skilled person. For convenience reference is made to Remington: The Science and Practice of Pharmacy, 19^th edition, 1995.

Typical isotonic agents are sodium chloride, mannitol, dimethyl sulfone and glycerol and typical preservatives are phenol, m-cresol, methyl p-hydroxybenzoate and benzyl alco- hoi.

Examples of suitable buffers are sodium acetate, glycylglycine, HEPES (4-(2- hydroxyethyl)-1 -piperazineethanesulfonic acid) and sodium phosphate.

The insulin preparation of this invention can be used in the treatment of states which are sensitive to insulin. Thus, they can be used in the treatment of type 1 diabetes, type 2 diabetes and hyperglycaemia for example as sometimes seen in seriously injured persons and persons who have undergone major surgery. The optimal dose level for any patient will depend on a variety of factors including the efficacy of the specific insulin derivative employed, the age, body weight, physical activity, and diet of the patient, on a possible combination with other drugs, and on the severity of the state to be treated. It is recommended that the daily dosage of the insulin derivative of this invention be determined for each individual patient by those skilled in the art in a similar way as for known insulin compositions.

Where expedient, the insulin derivatives of this invention may be used in mixture with other types of insulin, e.g. insulin analogues with a more rapid onset of action. Examples of such insulin analogues are described e.g. in the European patent applications having the publication Nos. EP 214826 (Novo Nordisk A/S), EP 375437 (Novo Nordisk A/S) and EP 383472 (EIi Lilly & Co.).

Examples of pharmaceutical preparations are neutral solutions from pH 6.5 to 8.3 containing from 300 to 4800 nmol/ml of the drug substance, isotonic agents, NaCI, buffers preservatives, zinc and stabilizers. With "desB30" or "B(1-29)" is meant a natural insulin B chain or an analogue thereof lacking the B30 amino acid residue and "A(1-21)" means the natural insulin A chain or an analogue thereof. The C-peptide and its amino acid sequence are indicated in the three letter amino acid code. DesB30,desB29 human insulin is a human insulin lacking B29 and B30. With "B1 ", "A1 " etc. is meant the amino acid residue in position 1 in the B chain of insulin (counted from the N-terminal end) and the amino acid residue in position 1 in the A chain of insulin (counted from the N-terminal end), respectively. The amino acid residue in a specific position may also be denoted as e.g. Phe^B1 which means that the amino acid residue in position B1 is a phenylalanine residue. By "A-chain" is understood the sequence of amino acids in the A-chain of human insulin.

By "B-chain" is understood the sequence of amino acids in the B-chain of human insulin. With "Insulin" as used herein is meant human insulin with disulfide bridges between Cys^A7 and Cys^B7 and between Cys^A20 and Cys^B19 and an internal disulfide bridge between Cys^A6 and Cys^A11, porcine insulin and bovine insulin.

By "insulin analogue" as used herein is meant a polypeptide which has a molecular structure which formally can be derived from the structure of a naturally occurring insulin, for example that of human insulin, by deleting and/or substituting at least one amino acid residue occurring in the natural insulin and/or by adding at least one amino acid residue. The added and/or substituted amino acid residues can either be codable amino acid residues or other naturally occurring amino acid residues or purely synthetic amino acid residues.

The insulin analogues may in one embodiment comprise up to 5 changes compared to the human insulin molecule, more typically up to 4, or up 3 and even more typically 1 or 2 changes compared to human insulin.

The insulin analogues may be such wherein position Asn at position A21 may be modified to Ala, GIn, GIu, GIy, His, He, Leu, Met, Ser, Thr, Trp, Tyr or VaI, in particular to GIy, Ala, Ser, or Thr and in particular to GIy. Furthermore, Asn at position B3 may be modified to Lys or Asp. Further examples of insulin analogues are des(B30) human insulin, insulin analogues wherein one or both of B1 and B2 have been deleted; insulin analogues wherein the A-chain and/or the B-chain have an N-terminal extension and insulin analogues wherein the A-chain and/or the B-chain have a C-terminal extension.

By a "mutation" is understood a substitution with a codable amino acid. By "insulin derivative" as used herein is meant a naturally occurring insulin or an insulin analogue which has been chemically modified by introducing a side chain in one or more positions of the insulin backbone or by oxidizing or reducing groups of the amino acid residues in the insulin or by acylating a free amino group or a hydroxy group

The expression "a codable amino acid" or "a codable amino acid residue" is used to indicate an amino acid or amino acid residue which can be coded for by a triplet ("codon") of nucleotides.

By "insulin hexamer" is meant all the possible conformations of 6 non-covalently associated molecules of insulin, some of which are known as the R6, T6, and R3T3 forms.

By "acylation" is understood the chemical reaction whereby a hydrogen of an amino group or hydroxy group is exchanged with an acyl group. By "isomer" of an insulin derivative is understood a compound differing from the derivative by a change of the position of a chemical bond between two parts of the derivative.

By "prolonged or protracted insulin" is meant an insulin peptide which has a time-action of more than 8 hours in standard models of diabetes. Preferably, the insulin pep- tide has a time-action of at least 9 hours or at least 10 hours. More preferably, the protracted insulin has a time-action in the range from at least 9 to at least 15 hours.

With "rapid acting insulin" is meant an insulin compound which has a faster onset of action than human insulin. One example of rapid acting insulins is analogues wherein the amino acid residue at position B28 is Asp. A specific example from this group of insulin ana- logues is Asp^B28 human insulin. Another example of rapid acting insulin analogues are analogues wherein the amino acid residue at position B28 is Lys and the amino acid residue at position B29 is Pro. A specific example from this group of parent insulin analogues is

|__ysB28p_roB29 _human jnsuljη.

By "fatty acid" is meant any saturated aliphatic monocarboxylic acid of the general formula C_nH_2n+iCOOH.

By "unsaturated fatty acid" is meant a fatty acid featuring one or several double bonds.

By "branched fatty acid" is meant a fatty acid featuring at least one carbon atom bound to 3 or 4 other carbon atoms. Branched fatty acids may be saturated or unsaturated. By "dicarboxylic acid" is meant an acid containing two carboxylic moieties. By an

"α,ω-dicarboxylic acid" is meant a dicarboxylic acid where the carboxyl groups are located in the opposite terminal positions of the main carbon chain.

By "linker" is meant a bridge having the function of binding a substituent to the insulin moiety or forming a bridge between two sites of the insulin derivative. The linker be- comes part of the substituent and may by itself contribute to the properties of the insulin derivative. Examples of linkers are amino acids, dicarboxylic acids, and functionalized PEG (polyethylene glycol) such as amino and/or carboxy PEG.

The expression "an amino acid residue having a carboxylic acid group in the side chain" designates amino acid residues like Asp, GIu and hGlu. The amino acids can be in either the L- or D-configuration. If nothing is specified it is understood that the amino acid residue is in the L configuration.

In the present context the three-letter or one-letter indications of the amino acids have been used in their conventional meaning as indicated the following table. Unless indicated explicitly, the amino acids mentioned herein are L-amino acids. Further, the left and right ends of an amino acid sequence of a peptide are, respectively, the N- and C-termini unless otherwise specified.

Abbreviations for amino acids

Amino acid Tree-letter code One-letter code

Glycine GIy G

Proline Pro P

Alanine Ala A

Valine VaI V

Leucine Leu L lsoleucine He I

Methionine Met M

Cysteine Cys C

Phenylalanine Phe F

Tyrosine Tyr Y

Tryptophan Trp W

Histidine His H

Lysine Lys K

Arginine Arg R

Glutamine GIn Q

Asparagine Asn N

Glutamic Acid GIu E

Aspartic Acid Asp D

Serine Ser S

Threonine Thr T

"ALP" is the Achromobacter lyticus protease which is a lysine specific protease from Acromobacter lyticus.

By "RP-HPLC" is meant reverse phase high pressure liquid chromatography.

By "SEC" is meant size exclusion chromatography.

By "t_50% " is meant the time at which 50% of the injected insulin has disappeared from the subcutaneous injection site.

By "Tris" is meant Tris(hydroxymethyl)aminomethane.

In litocholyl-Glu(ONSu)-OR GIu is acylated in its amino group, the α-carboxyl group is protected as an ester function and the γ-carboxyl group activated as the N- hydroxysuccinimide ester. By ID is meant internal diameter.

By pH in a mixture of water and organic solvent is understood the reading from a pH meter in the aqueous solution using a glass electrode calibrated in an aqueous standard buffer before the organic solvent is added to make the final mixture. All references, including publications, patent applications, and patents, cited herein are hereby incorporated by reference in their entirety and to the same extent as if each reference were individually and specifically indicated to be incorporated by reference and were set forth in its entirety herein (to the maximum extent permitted by law).

All headings and sub-headings are used herein for convenience only and should not be construed as limiting the invention in any way.

The use of any and all examples, or exemplary language (e.g., "such as") provided herein, is intended merely to better illuminate the invention and does not pose a limitation on the scope of the invention unless otherwise claimed. No language in the specification should be construed as indicating any non-claimed element as essential to the practice of the inven- tion.

The citation and incorporation of patent documents herein is done for convenience only and does not reflect any view of the validity, patentability, and/or enforceability of such patent documents.

This invention includes all modifications and equivalents of the subject matter re- cited in the claims appended hereto as permitted by applicable law.

EXAMPLES Example 1

Analytical separation of Lys^B29(N^ε-lithocholyl-γ-glutamyl) des(B30) human insulin and Lys^B29(N^ε-lithocholyl-α-glutamyl) des(B30) human insulin using RP-HPLC in

(NhU)₂SO₄ZH₂SO₄ buffer of pH 2.5 in water-acetonitril.

A 125 X 4 mm I. D. column was packed with a dimethylbutyldimethylsilyl substituted silica having pore size of about 100 A and particle diameter of about 5 μm and equilibrated at

5O⁰C at a flow rate of 1 ml/min with a buffer of 0.075 M NH₄SO₄ adjusted to pH 2.50 with H₂SO₄ in the aqueous buffer containing 7.8% (w/w) acetonitrile, followed by addition of ace- tonitrile to make 28 % (w/w).

A 30 μg sample was applied to the column dissolved in 8 μl of the above mentioned buffer without the acetonitrile. The separation was achieved by isocratic elution using the above mentioned conditions for equilibration. Lys^B29(N^ε-lithocholyl-γ-glutamyl) des(B30) hu- man insulin emerged from the column after about 20 min and Lys^B29(N^ε-lithocholyl-α- glutamyl) des(B30) human insulin after about 23.5 min.

Example 2 Preparative purification of Lys^B29(N^ε-lithocholyl-γ-glutamyl) des(B30) human insulin from Lys^B29(N^ε-lithocholyl-α-glutamyl) des(B30) human insulin using RP-HPLC in (NhU)₂SO₄ZH₂SO₄ buffer of pH 2.5 in water-acetonitril.

A 300 X 50 mm I. D. column is packed with a dimethylbutyldimethylsilyl substituted silica having pore size of about 100 A and particle diameter of about 15 μm is equilibrated at 5O⁰C and at a flow rate of 50 ml/min with a buffer of 0.075 M NH₄SO₄ adjusted to pH 2.50 with H₂SO₄ and comprising 30 % (v/v) of acetonitrile. A sample of 1 g of impure Lys^B29(N^ε-lithocholyl-γ-glutamyl) des(B30) human insulin containing 3.5% of Lys^B29(N^ε-lithocholyl-α-glutamyl) des(B30) human insulin is applied to the column followed by isocratic elution using the conditions of equilibration. The chromatography is monitored by absorption at 280 nm. After about 2.5 h the product emerges from the column. The product in the first half of the peak is isolated by dilution with 2 volumes of water followed by adjustment of the pH to 5.2 whereby it precipitates. Analysis of the purity using the method of example 1 shows a content of 1 % of Lys^B29(N^ε-lithocholyl-α-glutamyl) des(B30) human insulin.

Example 3

Preparative separation of Lys^B29(N^ε-lithocholyl-γ-glutamyl) des(B30) human insulin and Lys^B29(N^ε-lithocholyl-α-glutamyl) des(B30) human insulin using cat-ion exchange in wa- ter-ethanol at pH 4.8.

Approximately 10 mg of Lys^B29(N^ε-lithocholyl-γ-glutamyl) des(B30) human insulin containing 3.5% of Lys^B29(N^ε-lithocholyl-α-glutamyl) des(B30) human insulin dissolved in 60 % ethanol at pH 3.0 was applied to a 150 x 4 I. D. mm column packed with the cat-ion exchanger Source 15 S. The column was equilibrated and eluted with 0.02 M citrate buffer in 60% ethanol, adjusted to pH 4.8 with NaOH at a rate of 0.3 ml/min. The product eluted from about 75 column volumes to about 170 column volumes. The impurity of Lys^B29(N^ε-lithocholyl-α-glutamyl) des(B30) human insulin was absent in the pool eluting between 75 and 142 column volumes. Example 4

Analytical separation of Lys^B29(N^ε-lithocholyl-γ-glutamyl) des(B30) human insulin and Lys^B29(N^ε-lithocholyl-α-glutamyl) des(B30) human insulin using RP-HPLC in Na₂SO₄ZH₃PO₄ buffer of pH 4.0 in water-acetonitrile. A 250 X 4 mm I. D. column packed with an octadecyl substituted silica (e.g. Lichro- sorb RP C18) having particle diameter of about 5 μm was equilibrated at 35⁰C and at a flow rate of 1 ml/min with a buffer comprising 0.04 M Na₂SO₄, 0.008 M H₃PO₄, 34% (v/v) of ace- tonitrile and adjusted to pH 4.0 with NaOH after addition of acetonitrile. A 40 μg sample was applied to the column dissolved in 10 μl of the above mentioned buffer. The separation was achieved by isocratic elution using the above mentioned conditions for equilibration.

Lys^B29(N^ε-lithocholyl-γ-glutamyl) des(B30) human insulin emerged from the column after about 21 min and Lys^B29(N^ε-lithocholyl-α-glutamyl) des(B30) human insulin after about 34 min. The concentration of acetonitrile may be adjusted if needed, as the elution times are sensitive to this parameter.

Example 5

Preparative separation of Lys^B29(N^ε-lithocholyl-γ-glutamyl) des(B30) human insulin and Lys^B29(N^ε-lithocholyl-α-glutamyl) des(B30) human insulin using RP-HPLC in Na₂SO₄/H₃PO₄ buffer of pH 4.0 in water-ethanol. A 250 X 10 mm I. D. column packed with an octadecyldimethylsilyl substituted silica having particle diameter of about 5 μm and pore size about 120 A was equilibrated at ambient temperature and with a flow rate of 2 ml/min with a 40/60 (v/v) mixture of a solvent A comprising 0.1 M Na₂SO₄, 0.078 M H₃PO₄, 10% (v/v) of ethanol and adjusted to pH 4.0 with NaOH and a solvent B being ethanol/water 55/45 (v/v). A sample of 10 mg sample was ap- plied to the column dissolved in 2 ml 20% ethanol adjusting the pH to 4.0. The sample contained 3.5 % of Lys^B29(N^ε-lithocholyl-α-glutamyl) des(B30) human insulin. The separation was achieved by isocratic elution using a 36/64 (v/v) mixture of the above mentioned solvents A and B. The Lys^B29(N^ε-lithocholyl-γ-glutamyl) des(B30) human insulin product emerging from the column after 24-37 min (fraction 1 ) was collected and precipitated at pH 5.2 after 2 fold dilution with water. The impurity of Lys^B29(N^ε-lithocholyl-α-glutamyl) des(B30) human insulin emerged after about 40-43 min fraction 2). The concentration of ethanol may be adjusted if needed, as the elution times are sensitive to this parameter.

Analytical HPLC according to example 1 of the product in fraction 1 showed the product to be 99.7 % pure without any trace of the impurity of Lys^B29(N^ε-lithocholyl-α- glutamyl) des(B30) human. Fraction 2 contained 8.2% of the impurity. Example 6

Preparative separation of Lys^B29(N^ε-lithocholyl-γ-glutamyl) des(B30) human insulin and Lys^B29(N^ε-lithocholyl-α-glutamyl) des(B30) human insulin using cat-ion exchange in water- ethanol in a pH gradient from pH 4 to 6.

Approximately 10 mg of Lys^B29(N^ε-lithocholyl-γ-glutamyl) des(B30) human insulin containing 3.5% of Lys^B29(N^ε-lithocholyl-α-glutamyl) des(B30) human insulin dissolved in 60 % ethanol at pH 3.0 was applied to a 150 x 4 I. D. mm column packed with the cat-ion ex- changer Source 15 S. The column was equilibrated with 0.02 M citrate buffer in 60% ethanol, adjusted to pH 4.0 with triethanolamine at a rate of 0.3 ml/min. Elution was accomplished by a linear gradient from this buffer to the same adjusted to pH 6.0 with triethanolamine over a total of 40 column volumn. The product eluted from pH 4.76 to 4.96. The impurity of Lys^B29(N^ε-lithocholyl-α-glutamyl) des(B30) human insulin was absent in the first fractions from pH 4.76 to 4.91.

Example 7

Analytical separation of Lys^B29( Nε-hexadecandioyl-γ- glutamyl) des(B30) human insulin and Lys^B29( Nε-hexadecandioyl-α-glutamyl) des(B30) human insulin using RP-HPLC in Na₂SO₄/ NaH₂PO₄ buffer of pH 5.9 in water-acetonitril.

A 150 X 4.6 mm I. D. column was packed with a octyldimethylsilyl substituted silica having pore size of about 100 A and particle diameter of about 3.5 μm and equilibrated at 4O⁰C at a flow rate of 1 ml/min with a mixture consisting of 1 : a buffer of 2OmM NaH₂PO₄-H₂O and I OOmmol Na₂SO₄ adjusted to pH 5.9 with NaOH in the aqueous buffer containing 7.8% (w/w) acetonitrile,and 2: acetonitrile solvent containing 42,8% w/w acetonitrile, to make 25 % (w/w).

A 80 μg sample was applied to the column dissolved in 10 μl of the above mentioned buffer without the acetonitrile and pH 7.5. The separation was achieved by isocratic elution using the above mentioned conditions for equilibration. Lys^B29( Nε-hexadecandioyl-γ- glutamyl) des(B30) human insulin emerged from the column after about 20 min and Lys^B29( Nε-hexadecandioyl -α-glutamyl) des(B30) human insulin after about 23.5 min.

Example 8

Preparative purification of Lys^B29( Nε-hexadecandioyl-γ- glutamyl) des(B30) human insulin and Lys^B29( Nε-hexadecandioyl-α-glutamyl) des(B30) human insulin using RP-HPLC in Na₂SO₄/ NaH₂PO₄ buffer of pH 5.9 in water-acetonitril. A 250 X 10 mm I. D. column was packed with a dimethylbutyldimethylsilyl substituted silica having pore size of about 200 A and particle diameter of about 15 μm and was equilibrated at 5O⁰C and at a flow rate of 1 .96 ml/min with a buffer 2OmM NaH₂PO₄-H₂O adjusted to pH 5.9 with NaOH and comprising 16.5% (w/w) of acetonitrile. A sample of 49 mg of im- pure Lys^B29( Nε-hexadecandioyl-γ- glutamyl) des(B30) human insulin containing 0.5 mg (1%) of Lys^B29( Nε-hexadecandioyl -α-glutamyl) des(B30) human insulin was applied to the column followed by isocratic elution with a buffer 2OmM NaH₂PO₄-H₂O adjusted to pH 5.9 with NaOH and comprising 20.2 % (w/w) of acetonitrile. The chromatography was monitored by absorption at 280 nm. After about 2.5 h the product emerges from the column. The product col- lected between OD 0.25 and 0.42 on the leading and trailing edge. Analysis of the purity using the method of example 7 shows a content of 0.02% of Lys^B29( Nε-hexadecandioyl -α- glutamyl) des(B30) human insulin.

Example 9 Preparative purification of Lys^B29( Nε-hexadecandioyl-γ- glutamyl) des(B30) human insulin and Lys^B29( Nε-hexadecandioyl-α-glutamyl) des(B30) human insulin using RP-HPLC in Maleic acid/KCI buffer of pH 5.5 in water-ethanol.

A 250 X 10 mm I. D. column packed with an octadecyldimethylsilyl substituted silica having particle diameter of about 15 μm and pore size about 200 A was equilibrated at 5O⁰C and with a flow rate of 1 .96 ml/min with solvent A comprising 20 mmol/kg maleic acid 250 mmol/kg KCI, 15% (w/w) of ethanol and adjusted to pH 5.5 with NaOH. A sample of 50 ml containing 49 mg Lys^B29( Nε-hexadecandioyl-γ- glutamyl) des(B30) human insulin was applied to the column. The sample contained also 0.5mg of Lys^B29( Nε-hexadecandioyl-α- glutamyl) des(B30) human insulin. The separation was achieved by gradient elution going from 57 to 67%B in 15 column volumes. Solvent B comprising 20 mmol/kg maleic acid 250 mmol/kg KCI, 40% (w/w) of ethanol and adjusted to pH 5.5 with NaOH. After about 2.5 h the product emerged from the column. The product was collected between OD 0.25 and OD 0.5 on the leading and trailing edge was collected. The concentration of ethanol may be adjusted if needed, as the elution times are sensitive to this parameter. Analysis of the purity using the method of example 7 shows a content of Lys^B29( Nε- hexadecandioyl -α-glutamyl) des(B30) human insulin below limit of detection (LOD).

Gel permeation method (SEC test) SEC test of self-assembling into high molecular species of mixtures of

Lys^B29(N^ε-lithocholyl-γ-glutamyl)des(B30) human insulin and Lys^B29(N^ε-lithocholyl-α- glutamyl)des(B30) human insulin. Zinc ions were added in a ratio of 2, 3 or 4 per hexamer of insulin derivative.

Superose 6PC 3.2/30 (Amersham Biosciences) 3.2 x 300 mm (V_τ = 2.4 ml) is eluted by isotonic saline (NaC1 140 mM + Tris/HC1 10 mM + sodium azide 0.01 %, pH 7.4) at 50 μL/min and 37 ⁰C. UV detection is performed at 276 nm, and injection volume of 20 μl_ is used. The samples are compared to a series of molecular weight standards:

Blue dextran (1 .5 MDa); thyroglobulin ( 669 kDa); ferritin (440 kDa); aldolase (158 kDa) albumin dimer (133 kDa); albumin (66.4 kDa); ovalbumin (44.5 kDa); Co(lll)insulinhex. (35.0 kDa); ribonuclease (13.7 kDa); monomeric type insulin ( 5.9 kDa).

Sample preparation: To approximately 4 mg of wet precipitate of is added 200 μl_ water and dissolved by addition of lowest possible amount of sodium hydroxide to about pH 9 and adjusted back to pH 7.7. After transfer to a weighed tube followed by wash with 100 μl_ water the concentration of analogue is determined by HPLC (SEC). The stock solution is then divided in two portions and formulated with Tris buffer, 3 and 4 Zn per θinsulin, phe- nol/m-cresol and finally analysed.

A standard of aldolase, molecular weight of 158 kD was used as the lower limit for self-assembling. The areas under the curves before the elution time of aldolase (-35 min) and before thyroglobulin (-29 min) is indicative of the degree of self-assembling.

Table 1

Claims

1 . A pharmaceutical composition comprising an acylated insulin and an isomer thereof in a weight ratio of greater than 97:3 and pharmaceutically acceptable adjuvants, wherein said acylated insulin has the formula I

5

G I

X1 X2

15

wherein X at position A 18 is Asn or GIn, X1 in position B1 is Phe or deleted, X2 at position B2 is VaI or deleted, X3 at position B3 is Asn or modified to Thr, GIn, GIu or Asp, X4 at position A21 is Asn or modified to Ala, GIn, GIu, GIy, His, He, Leu, Met, Ser, Thr, Trp, Tyr or VaI, X5 at position B30 may be any codable amino acid residue except Lys, Arg and Cys or is deleted, Acyl is an acyl group derived from the group consisting of mono- or dicarboxylic fatty acids with a chain length of from about 6 to about 40, lithocholic acid, cholic acid, hyocholic acid, deoxycholic acid, chenodeoxycholic acid, ursodeoxycholic acid, hyodeoxycholic acid and cholanic acid, and wherein the isomer has the formula Il

wherein X-X5 and Acyl have the above meanings

2. Pharmaceutical formulation according to claim 1 , wherein the weight ratio of the acylated insulin derivative to the isomer thereof is greater than about 98:2.

3. Pharmaceutical formulation according to claim 1 , wherein the weight ratio of the acylated insulin derivative to the isomer thereof is greater than about 99:1 .

4. Pharmaceutical formulation according to claim 1 , wherein the weight ratio is in the range from 99:1 to 99.5:0.5.

5. Pharmaceutical formulation according to claim 1 , wherein the acyl group is derived from a mono- or dicarboxylic fatty acid with a chain length of from about 6 to about 24.

6. Pharmaceutical formulation according to claim 1 , wherein the acyl group is lithocholic acid.

7. Pharmaceutical formulation according to claim 6, wherein and the insulin derivative is Lys^B29(N^ε-lithocholyl-γ-glutamyl)des(B30) human insulin and the isomer thereof is Lys^B29(N^ε-lithocholyl-α-glutamyl)des(B30) human insulin.

8. Pharmaceutical formulation according to claim 1 , wherein X is Asn, X1 is Phe, X2 is VaI, X3 is Asn, X4 is Asn and X5 is deleted.

9. A method for producing a solution containing an acylated insulin with formula I and an isomer thereof with formula Il in a weight ratio of greater than about 97:3, said method comprising a) subjecting a solution the two compounds with a weight ratio of less than 97:3 to ion exchange or RP-HPLC chromatography or a combination thereof under conditions effective to separate the isomer from the acylated insulin with formula I, and b) collecting the fractions from step a) containing said acylated insulin and the isomer thereof in a weight ratio greater than 97:3.

10. A method according to claim 9, wherein pH of the RP-HPLC chromatography step is from about 3 to about 7.

1 1 . A method according to claim 10, wherein the pH of the RP-HPLC chromatography step is from about 4 to about 7, from about 5 to about 6.5, from about 4.0 to about 5.0 4.5 and 6.5 or about 4.75 to about 6.5.

12. A method according to claim 10, wherein the RP-HPLC step is conducted at pH of about 2.5 to about 5 in water-acetonitrile or water-ethanol mixtures.

13. A method according to claim 10, wherein the temperature of the chromatographic steps is between about 15 and about 5O⁰C.

14. A method according to claim 13, wherein the temperature of the chromatographic step is between about 30 and about 45⁰C.