EP1869079A2 - Conjugues di-polymere-proteine et procedes de preparation de ceux-ci - Google Patents

Conjugues di-polymere-proteine et procedes de preparation de ceux-ci

Info

Publication number
EP1869079A2
EP1869079A2 EP06725029A EP06725029A EP1869079A2 EP 1869079 A2 EP1869079 A2 EP 1869079A2 EP 06725029 A EP06725029 A EP 06725029A EP 06725029 A EP06725029 A EP 06725029A EP 1869079 A2 EP1869079 A2 EP 1869079A2
Authority
EP
European Patent Office
Prior art keywords
protein
polymer
peg
conjugate
csf
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Ceased
Application number
EP06725029A
Other languages
German (de)
English (en)
Inventor
Michael Bavand
Horst Blasey
Dietmar Lang
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Siegfried AG
Original Assignee
Siegfried AG
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Priority claimed from PCT/EP2005/002632 external-priority patent/WO2006094530A1/fr
Application filed by Siegfried AG filed Critical Siegfried AG
Priority to EP06725029A priority Critical patent/EP1869079A2/fr
Publication of EP1869079A2 publication Critical patent/EP1869079A2/fr
Ceased legal-status Critical Current

Links

Definitions

  • the present invention refers to di-polymer protein conjugates and processes for their preparation. Further, the present invention refers to the use of such di-polymer protein conjugate, especially di-pegylated protein conjugate, for the manufacture of a medicament for the treatment of disorders.
  • therapeutic proteins have been modified in order to increase their circulation half- life, e.g. by chemically or enzymatically coupling a polymer such as polyethylene glycol (PEG) to the protein.
  • PEG polyethylene glycol
  • the modification of a protein by attaching polyethylene glycol moiety thereto is also known as pegylation.
  • Proteins which were successfully modified by pegylation include erythropoietin (EPO), interferons such as interferon alpha, interferon beta and interferon gamma, granulocyte colony- stimulating factor (G-CSF or GCSF), interleukins such as IL-2 and IL-6, tumour necrosis factor (TNF), various cytokines and synthetic erythropoiesis protein (synthetic EPO).
  • EPO erythropoietin
  • interferons such as interferon alpha, interferon beta and interferon gamma
  • G-CSF or GCSF granulocyte colony- stimulating factor
  • interleukins such as IL-2 and IL-6
  • TNF tumour necrosis factor
  • various cytokines synthetic erythropoiesis protein
  • polyethylene glycol molecules are coupled to the protein via a functional group of the protein.
  • Each functional group in the amino acid chain of the protein which is nucleophilic, i.e. has electron donor ability, may react with a complementary group being attached to a PEG molecule.
  • nucleophilic groups include ⁇ -amino, ⁇ -amino, carboxyl, thiol, hydroxyl, imidazole, and guanidine groups.
  • the amino groups of a protein i.e. the a-amino group (N-terminus) and e-amino groups of lysine residues, are used as the reactive or functional groups for PEG coupling (see F.M. Veronese, Biomaterials 2001, 22: 405-417).
  • US 4,179,337 discloses PEG attachment to a protein via an acylation reaction resulting in an amide linkage between PEG and the protein.
  • EP 0 067 029 acylation of haemoglobin
  • EP 0 098 110 acylation of urokinase, kallikrein or leukocyte interferon
  • EP 0 154 316 reductive alkylation of lymphokines at neutral pH
  • EP 0 154 432 acylation of urokinase
  • EP 0 229 108 acylation of interferon beta, interleukin-2 and immuno toxins
  • EP 0 247 860 acylation of tumour necrosis factor
  • EP 0 335 423 acylation of G-CSF
  • EP 0 372 752 alkylation of CD4 immunoadhesin at neutral pH
  • EP 0439 508 alkylation of, inter alia, GM-CSF or G-SCF with tresylchloride
  • EP 0442 724 acylation of interleukin 6
  • EP 0473 268 acylation of calcitonin
  • pegylated proteins for biopharmaceutical applications preferably have a low number of PEG units per protein.
  • WO 96/11953 describes methods for attaching one polyethylene glycol molecule to the OC- amino group (N-terminus) of a polypeptide under conditions of a reductive alkylation and at a pH sufficiently acidic to selectively activate the a -amino group.
  • di-pegylated proteins Compared to mono-pegylated proteins, di-pegylated proteins, however, have some significant advantages: - A -
  • the maximum number of ethylene oxide units per linear PEG molecule limits the molecular weight to approximately 30 kDa (see Harris and Chess, Nat. Rev. Drug Discovery 2003, 2(3), 214-223). Attaching a second PEG molecule therefore allows increasing the molecular weight of the PEG portion in the PEG protein conjugate. PEG molecules with high molecular weight lead to increased viscosity of the pegylated protein solution. Attaching two PEG molecules of lower molecular weight instead of one high molecular weight PEG molecule can reduce this effect.
  • a more effective shielding of epitopes of the protein can be achieved by linkage of the protein with a second PEG molecule, so that proteolytic sensitivity and immunogenicity are reduced, while the stability of the PEG- protein conjugate is improved. This is particularly advantageous for proteinoids and non-recombinantly synthesised protein fragments that are joined via non- natural linkers, which may potentially increase immunogenicity.
  • WO 01/51510 describes G-CSF mutein conjugates, wherein the G-CSF amino acid sequence differs from the amino sequence of human G-CSF in at least one specifically altered amino acid residue which comprises an attachment group for PEG.
  • WO 01/51510 describes a di-pegylated G-CSF mutant that is pegylated at specifically introduced amino groups by acylation.
  • WO 2004/083242 describes G-CSF mutein conjugates, wherein the G-CSF amino acid sequence differs from the amino sequence of human G-CSF in at least one specifically altered amino acid residue by mutation of cysteine 17 to serine.
  • WO 2004/083242 describes a mono -pegylated G-CSF mutant that is specifically pegylated at the alpha amino group by reductive alkylation.
  • the reaction conditions applied for reductive alkylation will lead to mainly mono-PEG G-CSF mutein conjugates because of the acidic reaction pH, and traces of di- and tri-PEG G-CSF mutein conjugates, which were not used for the preparation of the pharmaceutical product.
  • WO 00/44785 describes an unspecified di-pegylated CrCSF conjugate that is obtained by coupling PEG moietie s to the protein via acylation.
  • Possible PEG attachment sites according to WO 00/44785 are amide linkages to a- and e- amino groups and ester- linkages to primary hydroxyl groups.
  • the pegylation reaction did not yield specifically dipegylated G-CSF molecules.
  • Another object of the present invention is to provide polymer- protein conjugates with two polymer units per protein, in particular di-pegylated protein conjugates, which have a longer circulating half-time and a greater in vivo biological activity than the corresponding unconjugated protein.
  • a pharmaceutical preparation which comprises at least one, and preferably two di-polymer- protein conjugates according to the present invention.
  • a process for the preparation of a polymer- protein conjugate comprises reacting a protein having at least two amino groups with a polymer reagent having a single aldehyde group in the presence of a reducing agent wherein the reaction time is chosen such that polymer- protein conjugates with two nitrogen atoms of amino groups of the protein being conjugated with a polymer unit via an amine linkage are preferably prepared.
  • the present invention relates to the use of a polymer- protein conjugate such as a polyethylene glycol-protein conjugate according to the present invention for the manufacture of a medicament for the treatment of a disorder.
  • the kind of disorder to be treated depends on the kind of protein conjugated to the polymer.
  • the disorder might be characterized by reduced haematopoietic or immune function.
  • the protein is GCSF
  • the disorder is usually neutropenia and/or leukaemia, which are induced by chemotherapy, radiotherapy and/or infections.
  • the protein is erythropoietin
  • the disorder is usually a type of anemia, for example renal anemia.
  • growth hormone also called somatotropin
  • the disorder is usually growth hormone deficiency.
  • the protein is an interferon, the disorder might be an infectious disease.
  • the invention is based on the finding that preferably two polymer units, in particular polyethylene glycol molecules, can be attached to a protein under conditions of a reductive alkylation by selecting a suitable pH, e.g. a pH about 7.
  • a suitable pH e.g. a pH about 7.
  • the reaction time is a crucial parameter for the formation of polymer- protein conjugates with two polymer units per protein, in particular di-pegylated protein conjugates according to the present invention.
  • the formation of pegylated protein conjugates according to the invention occurs sequentially in a time- dependent order from mono-, di-, tri- to higher pegylated protein conjugates (see Figure 3).
  • the molar ratio of protein to polymer and/or the stoichiometric ratio of the number of amino groups within a protein to the number of polymer molecules is crucial for the formation of polymer- protein conjugates with two polymer units per protein, in particular di-pegylated protein conjugates according to the present invention.
  • a lower stoichiometric ratio of polymer moelcules to the amino groups within the protein is required for the formation of polymer- protein conjugates with two polymer units per protein, in particular di-pegylated protein conjugates according to the present invention if the protein is glycosylated.
  • the process yield could be significantly increased by introducing a recycling step in which non- desired reaction products representing non- reacted or partially reacted protein molecules are applied to a second polymer attachment process step such as a second pegylation reaction step yielding further di-pegylated protein conjugate and thus contributing to a higher overall process yield of di-pegylated protein conjugate.
  • a recycling step results in a prominent yield of di-pegylated protein conjugate with high quality, being comparable to the quality obtained in the first pegylation reaction.
  • This principle of recycling non- desired reaction products to obtain a higher yield of polymer- protein conjugates with two polymer units per protein is also applicable to other polymer protein conjugate forming reactions.
  • the method of protein modification according to the present invention allows one to preferably or predominantly attach polymer units at a protein at specific sites of the protein taking advantage of a different reactivity of amino groups of the same type (e.g. ⁇ - amino group of lysine residues) depending on the amino acid residues adjacent to the amino acid with the reactive amino group in the sequence of the protein.
  • the process according to the present invention is preferably performed at a pH, which takes advantage of the pKa difference among e.g. different ⁇ -amino groups, different guanidino groups and/or different imidazole groups of the protein. This leads to a substantially homogenous preparation of di-polymer protein conjugates having the polymer predominantly attached to defined or pre- selectable or pre- selected or predetermined amino groups of the protein.
  • Substantially homogenous within the context of this invention is intended to mean that the preparation contains mainly di-polymer protein conjugates having the polymer predominantly attached to defined or pre- selectable or pre- selected or predetermined amino groups of the protein and to a lesser extent unconjugated protein species or proteins having attached one or more than two polymers or proteins having the polymer attached to other amino groups than the preselected ones.
  • the preparation contains at least 50%, more preferably at least 60, 65, 70 or 75%, even more preferably at least 80, 85 or 90 % and most preferably at least 92, 94, 96, 98 or 99% di-polymer protein conjugates having the polymer attached to the preselected amino groups, in relation to the total amount of proteins present in the preparation.
  • the polymer- protein conjugates according to the present invention are prepared by reductive alkylation.
  • the present invention preferably provides polymer- protein conjugates, wherein two nitrogen atoms of amino groups of the protein are each conjugated with a polymer unit via an amine linkage.
  • amino group includes primary and secondary amino groups and in particular NH- or NH2-groups in side-chains of amino acids such as NH 2 -groups in the side-chain of lysine, NH- or NH 2 -groups in the guanidino group of arginine or NH- groups in the imidazole side- chain of histidine.
  • amino group includes OC- (N-terminal) amino groups and ⁇ -amino groups which are NH 2 -groups in the side- chain of lysine.
  • the polymer unit preferably comprises at least one polymer moiety and a linker moiety, which is between at least the one polymer moiety and the amine linkage.
  • the linker moiety may be linear or branched. If the linker moiety is branched, a polymer unit may comprise more than one polymer moiety.
  • the linker moiety is preferably an aliphatic linker moiety. Suitable aliphatic linker moieties also include substituted alkyl diamines and triamines, lysine esters and malonic ester derivatives. The linker moieties are preferably non-planar, so that the polymer chains are not rigidly fixed. Preferably, the linker moiety includes a multiple -functionalized alkyl group containing up to 18, and more preferably from 1 to 10 carbon atoms. A hetero-atom such as nitrogen, oxygen or sulfur may also be included within the alkyl chain. The linker moiety may be branched, for example at a carbon or nitrogen atom.
  • the linker moiety comprises at least one methylene group attached to the nitrogen atom of the amine linkage, e.g. from 1 to 12, preferably from 1 to 5, more preferably from 1 to 3 and most preferably two methylene groups which are directly attached to the nitrogen atom of the amine linkage.
  • One preferred polymer- protein conjugate according to the present invention has the formula:
  • R is H, lower alkyl, aryl or any suitable protecting group
  • Polymer is a polymer which is suitable to be conjugated with proteins
  • m is an integer representing the number of methylene groups
  • P is a biologically active protein or proteinoid wherein two nitrogen atoms of amino groups of the protein or proteinoid (represented by NH in the above formula) are each conjugated with a polymer unit
  • L 1 is O, N, S and/or a branched or non-branched linker moiety which can be absent or present
  • L 2 is a branched or non- branched linker moiety which can be absent or present
  • y is a integer with the proviso that y is 1 in the absence of L 2 and y is at least 1 in the presence of L 2 .
  • R is preferably a lower alkyl or an aryl such as benzyl.
  • the term lower alkyl includes lower alkyl groups containing e.g. from 1 to 7 carbon atoms, preferably from 1 to 4 carbon atoms, such as methyl, ethyl, propyl, isopropyl, butyl, etc., with methyl being particularly preferred.
  • the linker L 1 which is part of the linkage between at least one polymer moiety and the suitable protecting group and/or the linker L 2 which is part of the linkage between the at least one polymer moiety, such as ethylene oxide residues of a PEG chain, and the nitrogen atom of the protein or proteinoid and, in particular, between the ethylene oxide residues of the PEG chain and the methylene groups, can independently be absent or present and are independently selected from a branched or non- branched linker moiety having any arbitrary structural formulae, such as the linker moieties described above, e.g. (-CO) 2 -CH-, -CO-O-, and/or -CO-NH-.
  • the linker L 1 might also be O, N and/or S.
  • the number m of methylene groups forming a part of the linkage between polymer moiety, such as the ethylene oxide residues of a PEG chain, and the nitrogen atom of the protein or proteinoid, to which the polymer moiety such as a PEG reagent is attached to is in a range from 1 to 12, preferably from 1 to 5, more preferably from 1 to 3 and is most preferably 2.
  • the number y, in the presence of L 2 is in a range from 1 to 10, preferably from 1 to 5, more preferably from 1 to 3 and is most preferably 2.
  • the polymer moiety is usually a substantially non- antigenic or non- immunogenic polymer chain.
  • the polymer moieties used are preferably selected from among water-soluble polymer moieties. This has the advantage that the protein to which the water-soluble polymer moieties are attached or conjugated does not precipitate in an aqueous environment such as a physiological environment.
  • the polymer selected should further have a single reactive aldehyde, so that the degree of polymerization may be controlled as provided for in the present processes.
  • the polymer unit as well as the polymer moiety may be branched or unbranched.
  • the polymer will be pharmaceutically acceptable.
  • One skilled in the art will be able to select the desired polymer moiety based on considerations such as whether the polymer- protein conjugate will be used therapeutically, and if so, the desired dosage, circulation time, resistance to proteolysis and other considerations.
  • the polymer moiety may be selected from the group consisting of polyalkylene glycol moieties, polysaccharide moieties such as dextran and its derivatives, polysaccharide and its derivatives, pyrrolidone moieties such as polyvinyl pyrolidone, cellulose moieties such as carboxymethyl cellulose, polyvinyl alcohol, poly-l,3-dioxolan, poly-l,3,6-trioxane, ethylene -maleic anhydride copolymer, polyaminoacid moieties and/or polyacrylamide moieties and/or other similar non- immunogenic polymer moieties (either homopolymers or random copolymers) and/or derivatives thereof.
  • Such polymers are also capable of being functionalized or activated for inclusion in the present invention.
  • polymer moiety is a polyalkylene glycol moiety.
  • polyalkylene glycol designates polyalkylene glycol radicals or polyalkylene glycol moieties, where the alkylene radical is a straight or branched chain radical.
  • polyalkylene glycol also comprises polyalkylene glycols formed from mixed alkylene glycols such as polymers containing a mixture of polyethylene and polypropylene radicals and polymers containing a mixture of polyisopropylene, polyethylene and polyisobutylene radicals.
  • a polyalkylene glycol moiety in the polymer protein conjugates according to the present invention is a polyethylene glycol moiety or polyethylene glycol residue formed by removal of the two terminal hydroxyl groups.
  • a- substituted polyalkylene oxide derivatives such as methoxy polyethylene glycols (mPEG) or other suitable alkyl substituted polyalkylene oxide derivatives such as those containing mono- or bis -terminal C 1 -C 4 groups.
  • Straight- chained non- antigenic polymers such as monomethyl polyethylene glycol homopolymers are preferred.
  • Alternative polyalkylene oxides such as other polyethylene glycol homopolymers, polyethylene glycol heteropolymers, other alkyl polyalkylene oxide block copolymers and copolymers of block copolymers of polyalkylene oxides are also useful.
  • An especially preferred polymer according to the present invention is polyethylene glycol resulting in PEG- protein conjugate of the formula:
  • PEG has the formula -(CH 2 -CH 2 -O) n with n representing the number of ethylene oxide residues in a polyethylene glycol unit. Further, it is preferred, that L 1 is O, L 2 is absent and y is 1.
  • the nitrogen atom of the amino group to which a polymer unit may be attached according to the present invention is preferably selected from the group consisting of nitrogen atoms of the ⁇ -amino group, ⁇ -amino groups, guanidinium groups and imidazole groups. It is preferred, but not limited, to attach a polymer unit to the nitrogen atoms of a primary amino group.
  • the biologically active protein which can be conjugated in accordance with this invention, may be any of the conventional therapeutic proteins.
  • protein in the context of the present invention the term protein is understood to include proteins and proteinoids.
  • proteinoids are synthetic protein- like molecules which have the identical amino acid sequence as the corresponding native protein, but are chemically synthesised in the form of peptides which are linked together to form the native protein- like structure with a resulting biological activity similar to the native protein. These proteinoids are potential protein therapeutics.
  • Proteins which can be conjugated with two polymer units and especially be di-pegylated according to the present invention include non- mutated and mutated proteins such as, but not limited to, growth factors, antibodies, hormones, in particular therapeutically active proteins such as, but not limited to, erythropoietin, interferon alpha, interferon beta, interferon gamma, consensus interferon, G-CSF, GM-CSF, hemoglobin, growth hormone, interleukins such as interleukin-2 and interleukin-6, tumour necrosis factor, various cytokines as well as immunoglobulins such as IgG, IgE, IgM, IgA, IgD and/or structural and/or functional variants and/or fragments and/or analogues thereof as well as their proteinoids or synthetic protein- like forms such as synthetic erythropoiesis protein (synthetic EPO).
  • non- mutated and mutated proteins such as, but not limited to, growth factors
  • the proteins which can be conjugated with two polymer units may be glycosylated or non- glycosylated. For example, it has been shown for erythropoietin that its activity greatly depends on its glycosylation. Naturally occurring erythropoietin is both N- and O- glycosylated at positions 24, 38 and 83 and 126 of the mature protein, respectively.
  • Glycosylated proteins which are to be used for the conjugation according to the present invention preferably have the same glycosylation pattern as the naturally occurring proteins. However, the present invention also encompasses glycosylated proteins having a different glycosylation pattern than the naturally occurring proteins, but which have essentially the same or even a higher biological activity than the naturally occurring proteins.
  • glycosylated protein is intended to mean that the protein has at least one carbohydrate molecule attached to amino acid chain.
  • proteins useful in the practice of this invention may be of any form isolated from mammalian organisms, a product of prokaryotic or eukaryotic host expression of exogenous DNA sequences obtained by genomic or cDNA cloning or by DNA synthesis or alternatively a product of chemical synthetic procedures or by endogenous gene activation.
  • the protein can be of a natural or recombinant source obtained from tissue, mammalian or microbial cell cultures, plant cell cultures, transgenic animals, yeasts, fungi and/or transgenic plants.
  • Suitable prokaryotic hosts include various bacteria such as E.
  • analogues of wild- type proteins preferably do not comprise additional polymer attachment sites such as additional amino groups as compared to the wild- type sequence. Moreover, analogues preferably do not comprise mutations of the amino acids containing amino groups naturally present in the wild-type amino acid sequence. However, the use of amino acid sequences of the wild- type protein are preferred.
  • the GCSF expression product may also include an initial methionine amino acid residue at position 1.
  • the present invention contemplates the use of any and all such forms of GCSF, although recombinant GCSF, especially E. coli- derived, is preferred.
  • Certain GCSF analogues have been reported to be biologically functional, and these may also be conjugated according to the present invention. These GCSF analogues may include those having amino acid additions, deletions and/or substitutions as compared to the GCSF amino acid sequence according to SEQ ID No. 1. Within the scope of this invention, these analogues do not comprise additional polymer attachment sites such as additional amino groups as compared to the GCSF amino acid sequence according to SEQ ID No. 1.
  • the analogues do not comprise mutations of the amino acids containing amino groups naturally present in the GCSF amino acid sequence according to SEQ ID No. 1.
  • proteins having the amino acid sequence of the wild- type protein according to SEQ ID No. 1 are most preferred.
  • the protein is one having the activity of G-CSF, including mutants of G-CSF, glycosylated G-CSF, non- glycosylated G-CSF and/or otherwise modified structural and/or functional variants of G-CSF.
  • the protein has the amino acid sequence of G-CSF identified in SEQ ID NO. 1 (see Figure 20) which corresponds to recombinant GCSF produced in bacteria, having 174 amino acids and an extra N- terminal methionyl residue.
  • Amino acid sequences of biological active G-CSF which differ from SEQ ID NO. 1 in that they do not contain a methionyl residue at position 1, are also preferred and within the scope of the present invention are also referred to as wildtype proteins.
  • the two amino groups are preferably selected from the group consisting of the ⁇ -amino group (N-terminus) and e-amino groups of lysine residues of G-CSF.
  • N-terminus the ⁇ -amino group
  • e-amino groups of lysine residues of G-CSF e-amino groups of lysine residues of G-CSF.
  • WO 96/11953 it has been shown (i) that mono-pegylation occurred in descending order N-terminus > lysine 35 > lysine 41 » lysines 17, 24 (where pegylation was not significantly detectable) taking advantage of the different reactivity between the ⁇ - and ⁇ -amino groups of the protein based on pKa differences and (ii) that the biological activity of mono-pegylated G-CSF decreased in the order N-terminus > lysine 35 > lysine 41.
  • the two amino groups of the protein, which are coupled to PEG are preferably selected from the group consisting of the N- terminal amino group, the e-amino group of lysine 17 and the e-amino group of lysine 35.
  • the protein might also be a protein of the interferon family, preferably the protein is an interferon ⁇ protein, most preferably it is interferon a 2a. If the protein is interferon a 2a, it has preferably the sequence according to SEQ ID No. 2 (see Figure 21).
  • the present invention contemplates the use of any and all such forms of interferon a 2a, also for example biologically functional analogues of the protein.
  • These analogues may include those having amino acid additions, deletions and/or substitutions as compared to the interferon a 2a amino acid sequence according to SEQ ID No. 2.
  • these analogues do not comprise additional polymer attachment sites such as additional amino groups as compared to the interferon a 2a amino acid sequence according to SEQ ID No. 2.
  • the analogues do not comprise mutations of the amino acids containing amino groups naturally present in the interferon a 2a amino acid sequence according to SEQ ID No. 2.
  • the use of the amino acid sequence of the wild-type protein according to SEQ ID No. 2 is preferred.
  • the two amino groups for polymer attachment are preferably selected from the group consisting of the ⁇ - amino group (N- terminus) and e- amino groups of lysine residues of interferon a 2a. It has been shown that among the potential PEG attachment sites the lysine residues at positions 121, 131, 134, 31 and 49 of the mature protein according to SEQ ID No. 2 are most preferred, because they are accessible to solvent. However, the present invention also relates to the conjugation of polymer groups such as PEG to the ⁇ -amino group of lysine 70, 83, 112, 164, 23 and/or lysine 133 of the mature protein sequence.
  • the protein is erythropoietin, it has preferably the sequence according to SEQ ID No. 3 (see Figure 22).
  • the present invention contemplates the use of any and all such forms of erythropoietin, also for example biologically functional analogues of the protein.
  • the present invention further contemplates the use of both glycosylated and non- glycosylated erythropoietin.
  • These analogues may include those having amino acid additions, deletions and/or substitutions as compared to the erythropoietin amino acid sequence according to SEQ ID No. 3.
  • these analogues do not comprise additional polymer attachment sites such as additional amino groups as compared to erythropoietin amino acid sequence according to SEQ ID No. 3. Moreover, the analogues do not comprise mutations of the amino acids containing amino groups naturally present in the erythropoietin amino acid sequence according to SEQ ID No. 3. However, the use of the amino acid sequence of the wild- type protein according to SEQ ID No. 3 is preferred.
  • the two amino groups for polymer attachment are preferably selected from the group consisting of the ⁇ - amino group (N- terminus) and e- amino groups of lysine residues of erythropoietin. It has been shown that among the potential PEG attachment sites lysine residues at positions 52, 116 and 152 of the mature protein according to SEQ ID No. 3 are most preferred, because they are accessible to solvent. However, the present invention also relates to the conjugation of polymer groups such as PEG to the ⁇ -amino group of lysine 20, 45, 97, 140 and/or lysine 154 of the mature protein sequence.
  • the protein is hGH, it has preferably the sequence according to SEQ ID No. 4 (see Figure 23).
  • the present invention contemplates the use of any and all such forms of hGH, also for example biologically functional analogues of the protein.
  • These analogues may include those having amino acid additions, deletions and/or substitutions as compared to the hGH amino acid sequence according to SEQ ID No. 4.
  • these analogues do not comprise additional polymer attachment sites such as additional amino groups as compared to the hGH amino acid sequence according to SEQ ID No. 4.
  • the analogues do not comprise mutations of the amino acids containing amino groups naturally present in the hGH amino acid sequence according to SEQ ID No. 4.
  • the use of the amino acid sequence of the wild-type protein according to SEQ ID No. 4 is preferred.
  • the two amino groups for polymer attachment are preferably selected from the group consisting of the ⁇ -amino group (N-terminus) and e- amino groups of lysine residues of hGH. It has been shown that among the potential PEG attachment sites lysine residues at positions 140, 145, 38 and 70 of the mature protein sequence according to SEQ ID No. 4 are most preferred, because they are accessible to solvent, whereas the lysine residues at positions 172, 41, 158, 168 and 115 of the mature protein are less preferred, because they are involved in receptor binding (Clark et al.
  • the present invention also relates to the conjugation of polymer groups such as PEG to the ⁇ -amino group of lysine 41, 115, 158, 168 and/or lysine 172 of the mature protein sequence.
  • the molecular weight of a polyethylene glycol moiety attached to a amino group is from 2 to 100 kDa, preferably from 5 to 60 kDa, more preferably from 10 to 30 kDa and most preferably from 12 to 20 kDa.
  • the number n of ethylene oxide residues in a polyethylene glycol moiety is from about 40 to about 2270, more preferably from about 110 to about 1370 and most preferably from about 225 to about 680.
  • compositions of the above- described polymer protein conjugates are provided. Such pharmaceutical compositions may be suitable for administration by injection or for oral, pulmonary, nasal or other forms of administration.
  • the present invention provides pharmaceutical compositions comprising effective amounts of di-polymer- protein conjugates of the present invention together with pharmaceutically acceptable diluents, preservatives, solubilizers, emulsifiers, adjuvants and/or carriers.
  • compositions include diluents of various buffer content, such as Tris-HCl, acetate, phosphate, pH and ionic strength; additives such as detergents and solubilizing agents such as Tween 80, antioxidants such as ascorbic acid and sodium metabi- sulfite, preservatives such as benzyl alcohol and bulking substances such as lactose or mannitol; incorporation of the material into particulate preparations of polymeric compounds such as polylactic acid, polyglycolic acid, etc. or into liposomes.
  • Such compositions may influence the physical state, stability, rate of in vivo release and rate of in vivo clearance of the polymer- protein conjugates according to the present invention.
  • a pharmaceutical preparation according to the present invention comprises at least one di-polymer-protein conjugate according to the present invention. It may be advantageous to prepare a mixture of two di-polymer-protein conjugates in order to select the proportion of the two di-polymer-protein conjugates in the mixture. Furthermore, if desired, one may prepare a mixture of various proteins with various numbers of polymer moieties attached and combine this mixture with a di-polymer-protein conjugate according to the present invention which yields a mixture with a pre- determined proportion of di-polymer- protein conjugate according to the present invention.
  • the preparation according to the present invention comprises at least one di-polyethylene glycol- G-CSF conjugate, also referred to an isoform or di-pegylated G-CSF isoform, selected from the group consisting of G-CSF conjugated with polyethylene glycol moieties at the ⁇ -amino group (N-terminal) and the e- amino group of lysine 17, and G-CSF conjugated with polyethylene glycol moieties at the N-terminal amino group and the e-amino group of lysine 35.
  • Isoforms of di-pegylated protein conjugates may occur if there are more than two attachment sites for PEG, e.g. more than two nitrogen atoms of amino groups in the protein. Both dipegylated G-CSF isoforms show the same bioactivity than the mixture of the di-pegylated G-CSF isoforms.
  • a defined di-pegylated G-CSF conjugate preparation may comprise two isoforms as described above, e.g. an isoform with polyethylene glycol moieties at the N- terminus and the e-amino group of lysine 17 (in the following also referred to as N-terminus + lysine 17), and an isoform with polyethylene glycol moieties at the N- terminus and the e- amino group of lysine 35 (in the following also referred to as N- terminus + lysine 35), in any defined mixed stoichiometric ratio, such as from 0.1 to 10, or from 0.5 to 2 or from 0.75 to 1.33 or about 1.
  • a defined product or preparation of the present invention consisting of a defined mixture of two di-PEG G-CSF isoforms as decribed above [(N-terminus + lysine 17) / (N-terminus + lysine 35) in a mixed ratio of e.g. about 0.76:1] has a physiological activity which is comparable to the physiological activity of mono-pegylated G-CSF (commercially available as Neulasta®) and which even has a higher physiological activity than non- modified G-CSF (commercially available as Neupogen®).
  • the two di-PEG G-CSF isoforms as described above have comparable physiological activity as shown in a further example.
  • each of the two di-PEG G-CSF isoforms as described above has a physiological activity which is comparable to the physiological activity of mono- pegylated G-CSF and even higher than the physiological activity of non-modified G-CSF.
  • This is in contrast to the hypothesis which may be deduced (i) from structural studies (Aritomi et al., Nature (1999) 401:713-717) and (ii) from alanine scanning mutagenesis experiments described by Young et al. (Protein Science (1997) 6:1228-1236) which assumes a negative impact on the biological activity of a PEG G-CSF isoform containing a lysine 17 modification.
  • di-PEG G-CSF isoforms having attached PEG molecules with different molecular weights i.e. 12 and 20 kDa
  • the bioactivity of both isoforms is comparable to the bioactivity of mono-pegylated G-CSF.
  • the protein having attached 20 kDa PEG shows a slightly higher white blood cell count and neutrophil cell counts and a sustained response compared to the protein having attached the 12 kDa PEG.
  • the present invention provides a mixed pharmaceutical preparation of PEG protein conjugates comprising at least one di-pegylated protein conjugate as described- above mixed with multi-pegylated protein conjugates wherein the proportion of said di-pegylated protein conjugate is pre- determined.
  • multi-pegylated protein conjugates means protein conjugates having more than two PEG moieties per protein molecule.
  • the proportion of said at least one di-pegylated protein conjugate within the mixed pharmaceutical preparation ranges from 1% to 99%, preferably from 50% to 99%, more preferably from 60, 70, 80 or 90 % to 99% and most preferably from 92, 94, 96, 98 to 99%.
  • a process for the preparation of a polymer- protein conjugate such as a polyethylene glycol- protein conjugate is provided which, under the conditions of a reductive alkylation, predominantly yields di-polymer- protein conjugates such as di-pegylated protein.
  • this conversion of a polymer reagent such as a PEG reagent with protein to di-polymer- protein conjugates is achieved by selecting reaction conditions which result in an enriched pool of di-polymer-protein conjugates, such as a suitable reaction time, a suitable protein concentration, a suitable molar ratio of polymer to protein and a suitable reaction temperature. It is particularly preferred to provide a process for the predominant preparation of di-pegylated PEG-protein conjugates.
  • a selective or predominant or preferred conversion of polymer with protein to di-polymer-protein conjugates or a selective or predominant or preferred yield of di-polymer-protein conjugates or selectively or predominantly or preferably preparing polymer- protein conjugates having two nitrogen atoms of amino groups of the protein each conjugated with one polymer unit means that at least 20%, preferably at least 25%, more preferably at least 30% and most preferably at least 40% or at least 45% or even at least 50% or at least 60% of the products of the reaction mixture resulting from a process of the present invention are di-polymer-protein conjugates according to the present invention.
  • the percentage of di-polymer- protein conjugates in the reaction mixture is the amount of at least one di-polymer- protein conjugate in relation to the amount of all identified species or products in the reaction mixture determined by cation exchange chromatography using a HPLC system and mass analysis.
  • a selective or predominant or preferred conversion of PEG with protein to di-pegylated PEG-protein conjugates or a selective or predominant or preferred yield of di-pegylated PEG-protein conjugates or selectively or predominantly or preferably preparing polyethylene glycol- protein conjugates having two nitrogen atoms of the amino groups of the protein each conjugated with one polyethylene glycol unit means that at least 20%, preferably at least 25%, more preferably at least 30% and most preferably at least 40% or at least 45% or even at least 50% or at least 60% of the products of the reaction mixture resulting from a process of the present invention are di- pegylated PEG-protein conjugates according to the present invention.
  • the percentage of di-pegylated PEG-protein conjugates in the reaction mixture is the amount of at least one di-pegylated PEG-protein conjugate in relation to the amount of all identified species or products in the reaction mixture determined by cation exchange chromatography using a HPLC system and mass analysis (see also Table 1).
  • the reaction time of each of the two process cycles according to the present invention is from about 6 h to about 48 h, more preferably from 12 h to 32 h.
  • Other preferred reaction times are about 7 h, about 8 h, about 9 h, about 10 h, about H h, about 13 h, about 14 h, about 15 h, about 16 h, about 17h, about 18 h, about 19 h, about 20 h, about 21 h, about 22 h, about 23 h, about 24 h, about 25 h, about 26 h, about 27 h, about 29 h, about, 30 h, about 31 h and about 33h, about 34 h or about 35 h.
  • reaction time is interrelated with other process parameters such as the reaction temperature.
  • Chamov et al. Bioconjugate Chemistry (1994) 5(2): 133- 140
  • increasing the reaction temperature decreased the reaction time at least by a factor of 2.
  • the reaction time also depends on the molar ratio of the polymer to the protein. For example, a doubling of the polymer amount for a given protein concentration may decrease the reaction time for the formation of di-polymer- protein conjugates by a factor of 2.
  • the reaction is performed at a temperature from 2 0 C to 5O 0 C, more preferably at a temperature from 2 0 C to 8 0 C and most preferably about 4 0 C.
  • the selection of a specific temperature may affect the reaction time, which is to be chosen such that polymer-protein conjugates with two nitrogen atoms of amino groups of the protein being conjugated with a polymer unit via an amine linkage are preferably prepared.
  • the reaction time which is preferably from about 6 h to about 48 h, may also be lower than 6 h, e.g. 1 h, 2 h, 4 h, or 5 h, when the reaction temperature is increased.
  • the reaction time is at least 6 h, preferably from 6 h to 12 h or from 8 h to 10 h when the reaction is performed at a temperature from 15°C to 25°C, preferably from 18°C to 24°C and most preferably 22°C and at least 18 h when the reaction is performed at 5°C.
  • the reaction is performed at a protein concentration from 0.5 to 100 mg/ml, more preferably from 1 to 10 mg/ml and most preferably from 3 to 7 mg/ml.
  • the reaction is performed at a protein- to- polymer molar ratio of from 1:1 to 1:400, and preferably from 1:5 to 1:50 and most preferably from 1:15 to 1:30.
  • a protein-to-polymer molar ratio is meant to be the ratio between moles of protein and moles of polymer at a given protein concentration in the reaction sample. Based on the protein to polymer molar ratio the stoichiometric ratio of amino groups within the protein to polymer molecules can be deduced.
  • a stoichiometric ratio of amino groups to polymer molecules is meant to be the number of OC- and ⁇ -amino groups within the protein to the number of protein molecules.
  • the stoichiometric ratio of the number of amino groups to the number of polymer molecules is in a range from 1:1 to 1:80, preferably 1:1.25 :l:80, more preferably from 1:1.25 to 1:50, even more preferably from 1:1.25 to 1:30 and most preferably from 1:2 to 1:10.
  • the stoichiometric ratio of the number of amino groups within the protein to the number of polymer molecules is in a range from 10:1 to 1:50, preferably from 1:0.3 to 1:50, more preferably from 2:1 to 1:25 and even more preferably from 1:0.8 to 1:10. Most preferably, the stoichiometric ratio is 10:1, 5:1, 1:0.3, 1:0.4, 2:1, 10.6, 1:0.7, 1:0.8, 1:0.9 or 1:1. Other preferred stoichiometric ratios are 1:2 to 1:10 or 1:2 to 1:5.
  • the reaction time is from about 6 h to about 48 h, more preferably from 12 h to 32 h.
  • Other preferred reaction times are about 7 h, about 8 h, about 9 h, about 10 h, about H h, about 13 h, about 14 h, about 15 h, about 16 h, about 17h, about 18 h, about 19 h, about 20 h, about 21 h, about 22 h, about 23 h, about 24 h, about 25 h, about 26 h, about 27 h, about 29 h, about, 30 h, about 31 h and about 33h, about 34 h or about 35 h and the protein-to-polymer molar ratio is from 1:1 to 1:400, and preferably from 1:5 to 1:50 and most preferably from 1:15 to 1:30.
  • the reaction time is from about 6 h to about 48 h, more preferably from 12 h to 32 h.
  • Other preferred reaction times are about 7 h, about 8 h, about 9 h, about 10 h, about H h, about 13 h, about 14 h, about 15 h, about 16 h, about 17h, about 18 h, about 19 h, about 20 h, about 21 h, about 22 h, about 23 h, about 24 h, about 25 h, about 26 h, about 27 h, about 29 h, about, 30 h, about 31 h and about 33h, about 34 h or about 35 h and the stoichiometric ratio of the number of amino groups to the number of polymer molecules is in a range from 1:1 to 1:80, preferably 1:1.25 :l:80, more preferably from 1:1.25 to 1:50, even more preferably from 1:1.25 to 1:30 and most preferably from 1:2 to 1:10 for non- glycos
  • the reaction time is from about 6 h to about 48 h, more preferably from 12 h to 32 h.
  • Other preferred reaction times are about 7 h, about 8 h, about 9 h, about 10 h, about H h, about 13 h, about 14 h, about 15 h, about 16 h, about 17h, about 18 h, about 19 h, about 20 h, about 21 h, about 22 h, about 23 h, about 24 h, about 25 h, about 26 h, about 27 h, about 29 h, about, 30 h, about 31 h and about 33h, about 34 h or about 35 h and the stoichiometric ratio of the number of amino groups within the protein to the number of polymer molecules is in a range from 10:1 to 1:50, preferably from 1:0.3 to 1:50, more preferably from 2:1 to 1:25 and even more preferably from 1:0.8 to 1:10 and most preferably, the stoichiometric ratio
  • the reaction time is from about 6 h to about 48 h, more preferably from 12 h to 32 h.
  • Other preferred reaction times are about 7 h, about 8 h, about 9 h, about 10 h, about 11 h, about 13 h, about 14 h, about 15 h, about 16 h, about 17h, about 18 h, about 19 h, about 20 h, about 21 h, about 22 h, about 23 h, about 24 h, about 25 h, about 26 h, about 27 h, about 29 h, about, 30 h, about 31 h and about 33h, about 34 h or about 35 h and the reaction is performed at a temperature from 2 0 C to 5O 0 C, more preferably at a temperature from 2 0 C to 8 0 C and most preferably about 4 0 C and the protein- to-polymer molar ratio is from 1:1 to 1:400, and preferably from 1:5 to 1:50 and most preferably from 1:15
  • the reaction time is from about 6 h to about 48 h, more preferably from 12 h to 32 h.
  • Other preferred reaction times are about 7 h, about 8 h, about 9 h, about 10 h, about H h, about 13 h, about 14 h, about 15 h, about 16 h, about 17h, about 18 h, about 19 h, about 20 h, about 21 h, about 22 h, about 23 h, about 24 h, about 25 h, about 26 h, about 27 h, about 29 h, about, 30 h, about 31 h and about 33h, about 34 h or about 35 h and the reaction is performed at a temperature from 2 0 C to 5O 0 C, more preferably at a temperature from 2 0 C to 8 0 C and most preferably about 4 0 C and the stoichiometric ratio of the number of amino groups to the number of polymer molecules is in range from 1:1 to 1:80, preferably 1:1.25
  • the reaction time is from about 6 h to about 48 h, more preferably from 12 h to 32 h.
  • Other preferred reaction times are about 7 h, about 8 h, about 9 h, about 10 h, about H h, about 13 h, about 14 h, about 15 h, about 16 h, about 17h, about 18 h, about 19 h, about 20 h, about 21 h, about 22 h, about 23 h, about 24 h, about 25 h, about 26 h, about 27 h, about 29 h, about, 30 h, about 31 h and about 33h, about 34 h or about 35 h and the reaction is performed at a temperature from 2 0 C to 5O 0 C, more preferably at a temperature from 2 0 C to 8 0 C and most preferably about 4 0 C and the stoichiometric ratio of the number of amino groups within the protein to the number of polymer molecules is in a range from 10:1 to 1:50
  • the reaction time is preferably from 12 h to 24 h and the reaction temperature is preferably about 4°C and the stoichiometric ratio of the number of amino groups within the protein to the number of polymer molecules is preferably from 1:2 to 1:10.
  • the reaction time is preferably from 12 h to 24 h and the reaction temperature is preferably about 4°C and the stoichiometric ratio of the number of amino groups within the protein to the number of polymer molecules is preferably in the range from 1:1 to 1:10, most preferably it is 1:1.25.
  • the reaction time is preferably from 12 h to 24 h and the reaction temperature is preferably about 4°C and the stoichiometric ratio of the number of amino groups within the protein to the number of polymer molecules is preferably in the range from 1:0.5 to 1:8, most preferably it is 1:0.625.
  • the reaction time is preferably from 12 h to 24 h and the reaction temperature is preferably about 4°C and the stoichiometric ratio of the number of amino groups within the protein to the number of polymer molecules is preferably in the range from 1:1 to 1:8, most preferably it is 1:1.6.
  • the reducing agent used in the reductive alkylation is preferably selected from, but not limited to, NaCNBH 4 or NaBH 4 .
  • the reaction be performed in the vicinity of neutrality, e.g. at a pH from 6 to 8, more preferably at a pH from 6.9 to 7.8, and most preferably at a pH from 7.2 to 7.5, especially if the protein is a protein having the biological activity of &CSF, erythropoietin, hGH or interferon.
  • neutrality e.g. at a pH from 6 to 8, more preferably at a pH from 6.9 to 7.8, and most preferably at a pH from 7.2 to 7.5, especially if the protein is a protein having the biological activity of &CSF, erythropoietin, hGH or interferon.
  • the reactbn may be performed in the presence of a buffer that, e.g., may be selected from a phosphate, acetate, HEPES, MES, and/or bicine buffers.
  • a buffer that, e.g., may be selected from a phosphate, acetate, HEPES, MES, and/or bicine buffers.
  • the PEG reagent used in the process according to the present invention is usually a reagent having the formula
  • R is H, a lower alkyl, aryl or any suitable protecting group
  • n is an integer representing the number of ethylene oxide residues in a polyethylene glycol moiety
  • m is an integer representing the number of methylene groups
  • L 1 is O, N, S and/or a branched or non- branched linker moiety which can be absent or present
  • L 2 is a branched or non- branched linker moiety which can be absent or present
  • y is a integer with the proviso that y is 1 in the absence of L 2 and y is at least 1 in the presence of L 2 .
  • R, m, n, L 1 , L 2 and y are defined as described above for the polymer- protein conjugates according to the present invention.
  • the polymer reagent is a PEG aldehyde, more preferably a PEG acetaldehyde, most preferably methoxy polyethylene glycol acetaldehyde (mPEG acetaldehyde).
  • the polyethylene glycol having a single aldehyde group, which is to be reacted with the protein may be stored in the form of the corresponding acetal.
  • the PEG acetal precursor may be activated to the PEG aldehyde reagent at any time before the conversion with the protein to be conjugated, under the proviso that the activated PEG aldehyde is stable until it is conjugated with the protein according to the process of the present invention.
  • a corresponding polyethylene glycol diethyl acetal may be activated.
  • a process for the long-term storage of a polyethylene glycol acetaldehyde such as methoxy polyethylene glycol acetaldehyde is provided, wherein the polyethylene glycol acetaldehyde is stored as a solid substance under an inert atmosphere such as nitrogen and at a temperature of about -2O 0 C or less.
  • a polyethylene glycol acetaldehyde such as methoxy polyethylene glycol acetaldehyde may be stored for at least 1 month, preferably at least 2 months, more preferably at least 3 months and most preferably at least 4 or 6 months.
  • the process further comprises a separation step or purification step leading to a pool of di- polymer- protein conjugates, preferably di-pegylated polyethylene glycol- protein conjugates, wherein two defined amino groups of the protein are each conjugated with one polymer unit, also referred to as the product pool, and to a pool which contains unreacted protein, unreacted polymer reagent, di-polymer- protein conjugate isoforms other than those of the product pool and/or polymer- protein conjugates with less or more than two amino groups of the protein conjugated with a polymer unit, also referred to as the rest pool.
  • a separation step or purification step leading to a pool of di- polymer- protein conjugates, preferably di-pegylated polyethylene glycol- protein conjugates, wherein two defined amino groups of the protein are each conjugated with one polymer unit, also referred to as the product pool, and to a pool which contains unreacted protein, unreacted polymer reagent, di-polymer- protein conjugate is
  • This separation step or purification step yielding a product pool and a rest pool is preferably performed by ion exchange chromatography.
  • the rest pool is again subjected to the process according to the present invention, e.g. instead of using solely unpegylated protein in the conversion with PEG reagent, the rest pool is reacted with a polyethylene glycol reagent having a single aldehyde group, e.g.
  • reaction time and/or the reaction temperature and/or the molar ratio and/or the stoichiometric ratio is chosen such that polyethylene glycol-protein conjugates having two nitrogen atoms of amino groups of the protein each conjugated with one polyethylene glycol unit are selectively prepared.
  • the reaction time, the reaction temperature, the molar ratio and the stoichiometric ratio are selected as described above for the process of the present invention.
  • di-polymer- protein conjugate such as di-pegylated PEG- protein conjugates according to the present invention.
  • di-pegylated PEG-protein conjugate isoforms having PEG moieties attached to the nitrogen atoms of amino groups at defined positions of the amino acid chain of the protein may be obtained.
  • a di-pegylated protein conjugate broadly includes any protein, which is conjugated with two polyethylene glycol units, wherein both of the polyethylene glycol units are each attached to one amino group of the protein.
  • the di-pegylated protein conjugates of the present invention are preferably obtained by reacting a polyethylene acetaldehyde molecule with the protein under conditions of a reductive alkylation which selectively leads to a di-pegylated PEG protein conjugate having a C 2 linker between the amine group resulting from the amino group of the protein and the chain of ethylene oxide residues of the polyethylene glycol unit.
  • One aspect of the present invention relates to a method for preparing di-pegylated protein conjugates by attaching polyethylene glycol aldehyde moieties to the a- and e- amino groups of proteins under the conditions of a reductive alkylation.
  • a first step in this method can optionally be the provision of the corresponding polyethylene glycol reagent, which is to be converted with the protein.
  • the polyethylene glycol reagent is a polyethylene glycol acetaldehyde, more preferred a methoxy polyethylene glycol acetaldehyde.
  • US 5,252,714 describes that methoxy PEG acetaldehyde is unstable in aqueous solution due to aldol decompositions and oxidation.
  • the process for providing the polyethylene glycol reagent comprises the activation of methoxy PEG di-ethyl acetal which is a stable storage form of the methoxy PEG acetaldehyde, into methoxy PEG acetaldehyde.
  • the PEG di-ethyl acetal precursor may be activated into PEG acetaldehyde, preferably methoxy PEG acetaldehyde, at some hours, e.g. 12 hours or 24 hours or preferably some weeks, e.g. one or two weeks or even some months, e.g. two, three or four months or more before the reaction with the protein, preferably G-CSF.
  • PEG acetaldehyde preferably methoxy PEG acetaldehyde
  • some hours e.g. 12 hours or 24 hours or preferably some weeks, e.g. one or two weeks or even some months, e.g. two, three or four months or more before the reaction with the protein, preferably G-CSF.
  • parameters have been identified which influence the yield and/or the reaction kinetics of the di-pegylation process including the recycling step according to the present invention.
  • These parameters include protein concentration, protein-to-PEG molar ratio, stoichiometric ratio of the number of amino groups within the protein to the number of polymer molecules depending on whether the protein is glycosylated or not, pH, the reducing agent used in the reductive alkylation, temperature, incubation time, co- solvent, stirring speed and method, ionic strength of the buffer.
  • WO 96/11953 teaches that a sufficiently acidic pH selectively leads to mono-pegylation of ⁇ - amino groups in view of the pKa difference between a- and e- amino groups.
  • WO 03/049699 describes that pegylation at a pH in the range from 5.5 to 7.5 only occurred at a- amino groups, which results in mono-pegylated protein.
  • Chamow et al. (Bioconjugate Chemistry (1994) 5(2): 133-140) describe that at pH 7.5 predominantly the a-amino group of CD4-IgG is pegylated, because it is unprotonated, while the e- amino groups remain unpegylated due to their protonation. This results in essentially monopegylated protein molecules.
  • the pegylation reaction according to the present invention is preferably performed in the vicinity of neutrality, e.g. at pH 6.5 to 7.5.
  • neutrality e.g. at pH 6.5 to 7.5.
  • the reaction time is chosen such that at least 20% of the reaction mixture comprises di-pegylated PEG protein conjugates, i.e. polyethylene glycol- protein conjugates, having two amino groups of the protein each conjugated with one polyethylene glycol unit.
  • the reaction time for the di-pegylation reaction is chosen such that at least 25, at least 30, at least 40 or at least 50% of the reaction mixture comprises di-pegylated protein such as di- pegylated GCSF.
  • each mPEG-G-CSF conjugate i.e. mono-, di-, tri- and tetra-pegylated species
  • the yield of each mPEG-G-CSF conjugate i.e. mono-, di-, tri- and tetra-pegylated species
  • the yield of mono-pegylated species reaches a maximum and starts to decline (data not shown) while di- and tri-PEG G-CSF conjugates are increasingly formed.
  • the di-PEG G-CSF conjugate yield reaches and maintains its maximum for several hours before starting to decrease, while the tri-PEG G-CSF conjugate is increasingly formed.
  • tri- and higher pegylated conjugates do not reach their yield maxima.
  • the selection of the defined time frame for the reaction time of a process according to the present invention is essential for the production of specific isoforms or species of pegylated proteins such as G-CSF at high yield.
  • Another critical parameter for the yield of dipegylated proteins is the molar ratio of protein to PEG and the stoichiometric ratio of the number of amino groups within the protein and the number of polymer molecules which differs for glycosylated and non-glycosylated proteins.
  • the process according to the present invention leads to a product mixture of differently pegylated forms, which is highly enriched in di-pegylated protein conjugate, i.e. preferably at least 20%, or more preferably at bast 30% of the reaction mixture comprises di-pegylated protein conjugates.
  • the reaction mixture also contains unreacted protein, unreacted polymer and otherwise pegylated protein conjugates.
  • the process according to the present invention further comprises a purification step in order to separate the di-pegylated species from undesired substances, e.g. by chromatography, preferably by an ion exchange chromatography step, more preferably by a cation exchange chromatography step (C-IEX).
  • the di-pegylated protein conjugates may be separated from unreacted protein, unreacted PEG reagent and/or polyethylene glycol conjugates which have more or less than two polyethylene glycol units per protein.
  • di-pegylated protein conjugate isoforms may be separated which differ in their PEG attachment sites form the desired di-pegylated PEG protein conjugate products.
  • the pool comprising the di-pegylated protein conjugate products is also designated as the product pool, whereas the pool comprising the undesired compounds is also designated as the rest pool, or product- depleted eluate.
  • the overall yield of di-pegylated protein conjugate could be further increased by recyclisation the product- depleted eluate resulting from a purification step such as the C-IEX step, in another pegylation process according to the present invention.
  • the eluate of the chromatography column e.g. a C-IEX column
  • the polyethylene glycol aldehyde reagent preferably an acetaldehyde such as methoxy polyethylene glycol acetaldehyde.
  • the di-pegylated protein conjugates according to the present invention show a high biological activity, which was demonstrated in bioassays for di-pegylated G-CSF isoforms and isoform mixtures according to the present invention.
  • the conjugates according to the present invention showed haematopoietic biological properties of naturally occurring G-CSF.
  • R-O-(CH 2 -CH 2 -O) n VCH 2 -CHO is preferably used for the conjugation of proteins, wherein R is H or lower alkyl such as methyl; n is an integer representing the number of ethylene oxide residues in a polyethylene glycol moiety, preferably from 225 to 680; m is an integer representing the number of methylene groups; L 1 is absent; L 2 is absent; and y is 1.
  • R is H or lower alkyl such as methyl
  • n is an integer representing the number of ethylene oxide residues in a polyethylene glycol moiety, preferably from 225 to 680;
  • m is an integer representing the number of methylene groups; L 1 is absent; L 2 is absent; and y is 1.
  • the skilled person is well aware of protocols and conditions for producing these compounds and other polymer reagents according to the present invention (see, e.g. EP 0 154 316; US 5,990,237; Chamov et al., Bioconju
  • the process of the present invention is started by converting a precursor compound of the polyethylene glycol reagent, such as methoxy polyethylene glycol acetal, to the activated polyethylene glycol acetaldehyde such as methoxy polyethylene glycol aldehyde.
  • This activation is usually performed in an aqueous solution, e.g. for up to 3h, at an acidic pH, e.g. pH 2, which may be achieved by adding any organic and/or inorganic acid.
  • an acidic pH e.g. pH 2
  • the conversion from polyethylene glycol acetal to polyethylene glycol aldehyde is tracked by analysing the components by 1 H-NMR spectroscopy.
  • any natural or recombinant protein or protein variant or protein mutant as obtained from any prokaryotic or eukaryotic cell may be used.
  • Human proteins are preferred.
  • any protein having the activity of the wild- type form, including variants, muteins, structurally or functionally equivalent protein as well as synthetic protein- like forms may be used.
  • the use of a protein having the amino acid sequence of the wild- type protein is preferred.
  • a protein having the activity of G-CSF is used in the process according to the present invention.
  • G-CSF is stored in a suitable buffer, e.g. 10 mM Na acetate buffer at pH 4.0 at low concentration, e.g. about 1 mg/ml.
  • proteins of the interferon family preferably IFN OC 2a, erythropoietin and human growth hormone, having the amino acid sequence of SEQ ID No. 2, SEQ ID No. 3 and SEQ ID No. 4, respectively, and analogues thereof.
  • the pegylation reaction according to the present invention is commonly performed in a suitable incubator, containing the protein such as G-CSF, IFN OC 2a, erythropoietin or human growth hormone, a physiological buffer system, and activated polyethylene glycol aldehyde such as methoxy polyethylene glycol acetaldehyde, and a reducing agent. Further, it is preferred that the reaction mixture is gently stirred at low temperature. The reaction may be stopped by acidification of the reaction mixture.
  • Typical reaction buffers for the pegylation reaction according to the present invention include phosphate, citrate/phosphate, cacodylate, carbonate, HEPES, MES, BES, MOPS, bicine and/or other suitable buffers.
  • the concentration at which the buffers are used may depend on the amount of protein used. Typical concentrations range from 50 to 200 mM buffer such as phosphate buffer. However, the different ionic strengths do not affect the di-pegylated product yield.
  • the concentration of protein used for the pegylation process according to the present invention depends on the physicochemical properties of the protein, such as solubility and aggregation, as well as on process economics.
  • the concentration of protein in the reaction mixture is preferably in the range from 1 to 20 mg/ml, more preferably from about 2 or 2.4 to about 5.7 or 6 mg/ml and most preferred at about 3.2 mg/ml.
  • di-pegylation of proteins according to the present invention may be performed in a wide range of protein to polyethylene glycol reagent molar ratios.
  • molar ratios of polyethylene glycol aldehyde, such as methoxy polyethylene glycol acetaldehyde to protein of about 5:1 to 400:1 are preferred.
  • the molar ratio, especially for the di- pegylation of non- glycosylated GCSF is 10:1 to 50:1, most preferably 15:1 to 30:1.
  • a stoichiometric ratio of the number of amino groups within the protein to the number of polymer molecules in a range from 1:1.25 to 1:50, preferably from 1:1.25 to 1:30 and more preferably from 1:2 to 1:10 is preferred to produce proteins with two polymer groups.
  • glycosylated proteins predominantly di-polymer protein conjugates are produced even when the number of amino groups to be accessible for PEG attachment is in excess to the number of polymer molecules at a given protein concentration.
  • the stoichiometric ratio of the number of amino groups within the protein to the number of polymer molecules ranges for glycosylated proteins from 10:1 to 1:50, preferably from 1:0.3 to 1:50, more preferably from 2:1 to 1:25 and even more preferably from 1:0.8 to 1:10 and most preferably, the stoichiometric ratio is 10:1, 5:1, 1:0.3, 1:0.4, 2:1, 1:0.6, 1:0.7, 1:0.8, 1:0.9 or 1:1.
  • Di-pegylation of proteins according to the present invention may be performed in a wide range of polyethylene glycol reagent molecular weights, i.e. from 2 to 100 kDa.
  • Preferred molecular weights of the polyethylene glycol aldehyde reagents are in the range from 10 to 30 kDa, e.g. from 12 to 20 kDa. Higher molecular weights may lead to lower di-pegylation yields and reaction mixtures may then show an increase in viscosity.
  • the chemical structure of the polyethylene glycol aldehyde reagent is not essential and thus not limited to be linear, but may also be branched.
  • the reaction temperature for the process according to the present invention will be in a range from about 2 0 C to about 5O 0 C, more preferably between 2 0 C and 2O 0 C.
  • the process according to the present invention is performed at lower temperature to minimize non-specific modification such as degradation and/or aggregation of protein by chemical or physical processes.
  • the most preferred reaction temperatures for the process according to the present invention are in the range from about 2 0 C to about 8 0 C, e.g. at about 4 0 C.
  • EP 0 154 316 states that the reducing agent should be stable in aqueous solution and preferably be able to reduce only the Schiff base formed in the initial process of reductive alkylation.
  • Preferred reducing agents may be selected from the group consisting of sodium borohydride and sodium cyanoborohydride and should be added in excess to the protein concentration used. It was found that a low molar excess of a reducing agent, such as sodium cyanoborohydride, i.e.
  • a less than 117- fold molar excess decelerates the formation of di-pegylated G-CSF, but reached the same maximum di-pegylated protein conjugate yield compared to high concentrations of reducing agents such as a molar excess of up to 180-fold molar excess.
  • concentrations of reducing agents such as a molar excess of up to 180-fold molar excess.
  • sodium cyanoborohydride in the range of a 120- fold molar excess, the kinetic formation of di-pegylated protein conjugate does not depend on the concentration of reducing agent used.
  • Ionic liquids have been tested as co- solvent for the regio- selective pegylation, revealing that the final di-pegylated protein conjugate yield decreases with increasing concentrations of a co- solvent such as l-butyl-4-methyl-pyridinium-tetrafluorborate.
  • the reaction mixture is usually subjected to a purification step, preferably by subjecting the reaction mixture to a cation exchange chromatography, after the pegylation reaction according to the present invention is stopped, preferred by acidification.
  • Cation exchange chromatography may be performed with all common and/or commercially available cation exchange matrices.
  • Typical ion exchange resins that may be employed for the purpose of the present invention comprise SP-5PW (Tosohaas Biosciences, Germany), Source S (Amersham Biosciences, Germany), Fractogel SO 3 650 (Merck, Germany), SP Sepharose HP (Amersham Biosciences, Germany) and SP FF Sepharose (Amersham Biosciences, Germany).
  • the purification step is performed at a temperature from 2 0 C to ambient temperature, preferably at room temperature.
  • any buffer may be used which is typically used for ion exchange chromatography, comprising phosphate, sodium acetate, TRIS/HCl, HEPES or other suitable buffers.
  • the buffers are chosen depending on the protein being analysed and are adjusted to pH values from 3.5 to 8.5, preferably from 4 to 6.
  • the dilution, washing, and equilibration buffer contains 15 mM sodium acetate at pH 4.2.
  • the elution buffer may be of the same composition as the equilibration buffer and may contain about 1 M NaCl.
  • the reaction mixture may be diluted, e.g. 3- fold with equilibration buffer in order to reduce the conductivity to approximately 6 mS/cm. This provides conditions for the complete binding of the protein conjugate according to the present invention to the ion exchange column.
  • Purification by C-IEX typically includes the following steps:
  • linear flow rates of about 68 cm/h may be used for all chromatography steps except sample loading which may be done at 17 cm/h.
  • the first pool also designated as product pool I
  • the second pool also designated as rest pool
  • the rest pool may also contain traces of di-pegylated isoforms of the product pool but essentially does not comprise di-pegylated isoforms of the product pool.
  • This rest pool is preferably recycled in another pegylation reaction, which is performed as described above for the process according to the present invention.
  • the rest pool is usually applied to ultrafiltration / diafiltration in order to concentrate the modified protein and to exchange the buffer for a buffer suited to perform a pegylation reaction according to the process of the present invention.
  • the rest pool is then subjected to a pegylation reaction as described above for the process of the present invention, optionally followed by cation exchange purification as described above.
  • the product pool II resulting from this re-pegylation step, optionally purified, may then be combined with the product pool I (see Figure 1).
  • the following examples are provided to illustrate the invention described herein and are not intended to limit it in any way.
  • 150 mg 12 kDa mPEG acetal was dissolved in 80 mM phosphoric acid and hydrolyzed for 3h at 5O 0 C. After cooling down on ice to approximately 2-8 0 C the solution was neutralized to pH 7 by drop wise addition of sodium bicarbonate solution (5%). The mixture was saturated with NaCl and three times extracted with 2 ml methylene chloride. The three collected organic phases were combined, dried by adding disodium sulphate and the volume was reduced to 1.5 ml (nitrogen stream).
  • mPEG acetaldehyde was precipitated by adding 30 ml ice-cold diethyl ether. The precipitate was isolated by vacuum filtration (G4 funnel), dried at room temperature under vacuum and transferred in gas-tight tubes, which were flushed with nitrogen gas before sealing them and stored at -2O 0 C. b) Stability of activated mPEG aldehyde
  • the mPEG acetal was usually activated freshly for each pegylation reaction, or was at maximum stored over one night.
  • Activated mPEG aldehyde appears to be stable for at least 125 days when stored light- protected, in sealed tubing, under N 2 atmosphere and at -20 0 C (see Figure 2).
  • the diluted G-CSF sample was concentrated by ultrafiltration. Depending on the final sample concentration, G-CSF was concentrated to approximately 4-8 mg/ml using an Amicon cell (50 ml) or centricon tubes (2 ml) equipped with YM 10 membranes of 10 kDa molecular weight cut off.
  • 25 mg solid mPEG aldehyde (molecular weight 12 kDa) were added to 3.2 mg/ml G-CSF in 100 mM phosphate buffer, pH 7.5, resulting in a 1:25 molar ratio of G-CSF to mPEG aldehyde.
  • 20 mM sodium cyanoborohydride were added, and the reaction mixture was further incubated by gentle stirring at 2-8 0 C for several hours. The reaction was stopped by acidification with 8 N HCl.
  • the species distribution (mono-, di-, tri-, terra-) in the PEG G-CSF reaction mixture was determined by C-IEX HPLC. Therefore, the stopped reaction mixture was diluted for three times with 15 mM acetate buffer (pH 4.2) and passed through a column packed with SP- 5PW (75 x 7.5 mm, 10 ⁇ m particle size, 3.3 ml bed volume, manufactured by Tosohaas Bioscience), mounted on a Dionex system or Shimadzu HPLC system.
  • the column was equilibrated in 15 mM NaAc, pH 4.2. The flow rate was 0.4 ml/min and typically 200 ⁇ g of protein were injected. The column was run at room temperature; however, the samples were maintained at 5°C in the auto- injector.
  • each pegylated CrCSF conjugated could be allocated to the retention time of an elution peak.
  • the mass of the single pegylated GCSF species determined by MALDI-TOF agreed with the preliminary mass estimation based on the amino acid sequence and PEG moiety (see Table 1).
  • Table 1 Comparison of calculated and experimentally determined molecular weight of different pegylated G-CSF species.
  • a retention time could be allocated to the corresponding PEG G-CSF species.
  • the analytic C-IEX allowed the separation of one mono- PEG G-CSF, three di-PEG G-CSF isoforms, four tri-PEG G-CSF isoforms, and one tetra- PEG G-CSF in dependence of the retention time.
  • converting recombinant human GCSF with mPEG aldehyde in a 1:25 molar ratio to PEG G-CSF conjugates yielded predominantly di-pegylated G-CSF at a reaction time from 18 h to 42 h incubation, reaching the maximum yield at about 30 h incubation time (see Figure 3a).
  • the maximum synthesis yield of the Di-PEG G-CSF conjugate (40%, see figure 3c) is not influenced by the varying stoichiometric ratios of PEG molecules to the number of amino groups within the protein.
  • the temperature could have an additional influence on the incubation time for producing di-PEG G-CSF product
  • di-pegylated product yield di-PEG G-CSF isoform mixture
  • kinetic of product formation two reaction conditions (5.7 mg/ml G-CSF with PEG in either 1:10 or 1:20 molar ratios) were tested at two different temperatures (5°C and 22°C) and analysed according to the invention comparing low (5°C) and high (22°C) temperature influence.
  • the pegylation reaction to produce predominantly di-pegylated protein conjugates can be further accelerated by a factor of at least 2 by doubling the PEG amount for a given protein concentration in the reaction which results in an at least a 2-10 fold excess of PEG molecules in relation to the available amino groups (amino and epsilon) for PEG attachment independent of the selected temperature.
  • the pH of the reaction should be in the vicinity of neutrality (to make use of the different reactivity of the epsilon amino groups of the target protein), and (ii) PEG should be added in at least 2-10 fold molar excess to the available ⁇ - and ⁇ -amino groups in the target protein which in the case of G-CSF corresponds to molar ratios of G-CSF:PEG between approximately 1:10 and 1:50.
  • the average isoform composition in the PEG G-CSF reaction mixture at 30 h incubation was determined (see Table 2).
  • Table 2 Relative percentage of differently pegylated G-CSF species in five PEG G-CSF syntheses at a 30 h reaction time with a single coupling cycle.
  • the rest pool of the preparative C-IEX chromatography (see Figure 4) containing all species including traces of product species according to the present invention, was collected, de- salted and concentrated by ultrafiltration/diafiltration and subjected again to the pegylation reaction (see Example 3), except that PEG modified G-CSF (3.2 mg/ml) was used.
  • Table 3 Relative percentage of differently pegylated G-CSF species in six PEG G-CSF syntheses cycles comprising two coupling cycles of 18 h each.
  • Linear flow rates were approximately 68 cm/h for all chromatography steps except for protein loading which was done at 17 cm/h.
  • C-IEX HPLC Analytical cation exchange high performance liquid chromatography
  • Table 4 Allocation of elution peaks from the preparative C-IEX run with corresponding retention time and allocated PEG G-CSF species (isoforms).
  • di-pegylated GCSF three out of ten possible isoforms could be identified. Two isoforms, Di and Di' PEG G-CSF, eluted in major single peaks. The third minor isoform co-eluates with a tri- PEG G-CSF isoform.
  • di-PEG G-CSF product which comprises two isoforms, di- (peak 2) and di'- (peak 3).
  • PEG attachment sites for these isoforms have been identified to be located at the N- terminus and lysine 17 or the N- terminus and lysine 35, respectively, by using a combination of MALDI-MS/ nano-LC ESI-MS/MS and Edman sequencing (see product characterisation).
  • the defined di-PEG G-CSF product contained at least 90% di-PEG G-CSF species, comprising di and di' isoforms in an approximately 0.76 (di) to 1 (di') ratio and was highly pure (see Table 5)
  • the isolated di-PEG G-CSF product consists of two major (di and di') di-PEG G-CSF isoforms which represent the major active di-pegylated product species.
  • composition and purity of the final di-PEG G-CSF product were analysed by cation exchange and size exclusion chromatography.
  • Buffer A 15 mM NaAc, pH 4,2
  • Buffer B 15 mM NaAc, pH 4,2; 1.000 mM NaCl
  • the isomer composition of the di-PEG G-CSF product was analysed by C-IEX HPLC (see Figure 5) and the composition was within the defined product specification (see Table 6).
  • Table 6 Averaged Di-PEG G-CSF Product definition optimized for the total process.
  • Buffer A 100 mM Pi; pH 6.9
  • the final product consists of > 90% di-pegylated G-CSF, comprising di and di' PEG G-CSF isoforms in a defined mixed ratio.
  • the peaks between 11 and 13 min retention time were not further analysed, but could potentially be higher molecular aggregates (-1.5 %, see Table 7).
  • Table 7 PEG G-CSF isoform distribution in the final di-PEG G-CSF product analysed by SEC analysis.
  • the final di- PEG G-CSF product (sample B) and native unmodified G-CSF (sample A) were analysed by circular dichroism spectroscopy.
  • CD spectra were recorded from 180-260 nm wavelengths using a Jasco J- 720 (Japan) spectropolarimeter.
  • the di-pegylated G-CSF isoforms (di- and di' PEG G-CSF) were analysed by peptide mapping.
  • non-pegylated G-CSF was treated identically.
  • the procedure comprised the steps (i) denaturation of the PEG G-CSF isoforms and capping of the free SH groups by reductive alkylation, (ii) chymotrypsin digestion of the denatured PEG G-CSF isoforms, (iii) rpHPLC peptide separation and (iv) sequence analysis by combination of electrospray tandem mass analysis and MALDI-MS/N- terminal sequencing.
  • Dried samples of di-PEG G-CSF isoforms were reconstituted to a concentration of lmg in ImI 0.1M ammonium hydrogen carbonat buffer, pH 7.8 for digestion.
  • the isoforms were digested with chymotrypsin (enzyme to substrate ratio per weight of 1:100) at 37°C for 3h.
  • Protein digests were injected onto a Vydac C4 column (4.6x 250 mm, 5m particle size, 300 A pore size) and peptides were mapped by HPLC using a linear gradient of acetonitrile in 0.1% TFA (1% acetonitrile increase/min over 90 min). The resulting peptides were collected for peptide sequencing by a combination of nano-LC ESI MS/MS and N-terminal sequencing.
  • peptide mass sequencing with nano-LC ESI-MS/MS identified the sequence from amino acid 1 to 13: M-T-P-L-G-P-A-S-S-L-P-Q-S-F (MW 1470 m/z) having the PEG attachment to the ⁇ -amino group of the N-terminal methionine.
  • N-terminal sequence analysis determined the sequence from amino acid 14 to 40: LLKCLEQ VRKIQGDGAALQEKLCATY having the PEG attachment to the ⁇ -amino group of lysine 17.
  • N-terminal sequence analysis (22 cycles) determined the sequence from amino acid 14 to 40: LLKCLEQ VRKIQGDGAALQEKLCATY having the PEG attachment to the ⁇ -amino group of lysine 35.
  • the Di-PEG G-CSF isoform is di-pegylated at the N-terminus and lysine 17
  • the Di'-PEG G-CSF isoform is di-pegylated at the N-terminus and lysine 35.
  • the di-PEG G-CSF product comprising a mixture of di- and di' PEG G-CSF conjugate isoforms according to the invention was tested in vivo, to show the improved activity versus unmodified G-CSF (Neupogen®) and to compare its biological activity with commercially available mono-pegylated G-CSF (Neulasta®).
  • the in vivo study was an open randomized, parallel- grouped single dose biocomparability study of Neupogen® and Neulasta® with the di-PEG G-CSF product. It was carried out by subcutaneously injecting 0.1 mg/kg to male rats. Six animals per group per time point were subjected to bleeding. Serum samples were subjected to a complete blood count on the same day that the samples were collected. The average white blood cell counts and neutrophil counts were calculated.
  • the Di-PEG G-CSF product comprising a mixture of the di- and di'-PEG G-CSF conjugate isoforms yielded higher WBC and neutrophil counts and longer sustained WBC and neutrophil response compared to unmodified G-CSF (Neupogen®) (see Figure 8 and Figure 9).
  • the di-PEG G-CSF product according to the present invention was at least biosimilar, perhaps potentially better, because of slightly higher maximum WBC and neutrophil counts (see Figure 8 and Figure 9).
  • each single isoform di-PEG G-CSF, di'-PEG G-CSF
  • the isoforms were homogenously purified by cation exchange chromatography. Purity and composition of each isoform product (di-PEG G-CSF, di'-PEG G-CSF) were analysed by size exclusion chromatography and cation exchange HPLC.
  • the defined single isoform products contained at least 95% of the respective di-pegylated G-CSF conjugate isoform (see Table 8) .
  • Table 8 Isoform distribution in the defined single isoform products (di-PEG G-CSF, di'-PEG G-CSF).
  • the single isoform products, di- and di'-PEG G-CSF conjugate products, and the di-PEG G- CSF product mixture comprising both isoforms as described in example 5 were tested in vivo to compare their biological activities among each other.
  • the in vivo study was an open randomized, parallel- grouped single dose biocomparability study of the mixture of both isoforms, Di-PEG G-CSF conjugate mixture as defined product according to the invention with the single isoform products, di- and di'-PEG G-CSF conjugates.
  • the in vivo study was carried out by subcutaneously injecting 0.1 mg/kg of each of said products to male rats. Twelve animals per group per time point were subjected to bleeding. Serum samples were subjected to a complete blood count on the same day that the samples were collected. The average white blood cell counts and neutrophil counts were calculated.
  • Pegylation reaction & Analytics For the pegylation reaction, the educts, 20 kDa mPEG acetal and CrCSF were prepared according to examples Ala and A2. 32 mg solid mPEG aldehyde (molecular weight 20 kDa) were added to either 2.4 mg/ml G-CSF or to 3.2 mg/ml G-CSF in 100 mM phosphate buffer, pH 7.5, resulting in a 1:25 molar ratio of G-CSF (2.4 mg/ml) and in a 1:19 molar ratio of G- CSF (3.2 mg/ml) to mPEG aldehyde, respectively. The reactions were conducted and the samples were analysed by C-IEX HPLC as described in examples 3a and 3b.
  • the analytical C-IEX allowed the separation of one mono-PEG G-CSF - (having a total MW of 39 kDa) -, two di-PEG G-CSF ⁇ (having a total MW of 59 kDa) and three tri-PEG G-CSF conjugates (having a total MW of 79 kDa).
  • the single isoform was highly purified by preparative cation exchange chromatography.
  • the defined product contained at least 95% di'-PEG G-CSF conjugate isoform as described in example A 6, paragraph d)ii).
  • the in vivo study was an open randomized, parallel- grouped single dose biocomparability study of Neupogen® and Neulasta® with the single di'-PEG G-CSF conjugate isoform. It was carried out by subcutaneously injecting 0.1 mg/kg to male rats. Six animals per group per time point were subjected to bleeding. Serum samples were subjected to a complete blood count on the same day that the samples were collected. The average white blood cell counts and neutrophil counts were calculated.
  • the single Di'-PEG 2OkDa G-CSF conjugate isoform according to the present invention yielded i) higher WBC and neutrophil counts and ii) showed a sustained WBC and neutrophil response compared to unmodified G-CSF (Neupogen®) (see Figure 14 and Figure 15).
  • the di'-PEG G-CSF conjugate isoform according to the present invention is bioimproved because of higher maximum WBC and neutrophil counts and its sustained WBC and neutrophil response (see Figure 14 and Figure 15)
  • the in vivo studies were open randomized, parallel- grouped clearance studies of Neulasta® with either the Di-PEG G-CSF product mixture or the single di'-PEG 2OkDa G-CSF conjugate isoform in sham- operated (control group) and nephrectomised rats.
  • the studies were carried out by intravenous injection of 0.005 mg/kg of the conjugate to male rats.
  • Four animals per group per time point were subjected to bleeding.
  • the remaining GCSF plasma concentration was determined in the serum samples and relevant pharmacokinetic data such as the plasma elimination half- life were calculated.
  • an absorbance of 1.05 at 280 nm corresponds to 1 mg/ml protein.
  • the same molar extinction coefficient was applied.
  • the species distribution (mono-, di-, tri-) in the PEG IFN OC 2a reaction mixture was determined by C-IEX HPLC. Therefor, the stopped reaction mixture was diluted three times with 40 mM acetate buffer (pH 4.2) and passed through a column packed with SP- 5PW (75 x 7.5 mm, 10 ⁇ m particle size, 3.3 ml bed volume, manufactured by Tosohaas Bioscience), mounted on a Dionex system or Shimadzu HPLC system.
  • the column was equilibrated in 40 mM NaAc, pH 4.2. The flow rate was 0.4 ml/min and typically 200 ⁇ g of protein were injected. The column was run at room temperature; however, the samples were maintained at 5°C in the auto- injector.
  • the elution peaks of the analytical C-IEX HPLC were mass analysed (MALDI-TOF) or/and analysed by SDS PAGE, which allowed the allocation of each pegylated IFN ⁇ 2a peak to the retention time of an elution peak.
  • Table 10 Comparison of calculated and experimentally determined molecular weight of different pegylated IFN a 2a species.
  • a retention time could be allocated to the corresponding PEG IFN oc 2a species.
  • the analytical C-IEX allowed the separation of mono-PEG-, di-PEG- and tri-PEG-IFN OC 2a isoforms in dependence of the retention time.
  • Table 11 Relative percentage of differently pegylated IFN species at a 30 h reaction time with a single coupling cycle.
  • the molar ratio IFN:PEG had to be at least 1:15 which corresponds to a stoichiometric ratio of at least 1:1.25 of the number of available amino groups within the protein (in total 12 amino groups) to the number of PEG molecules.
  • the PEG concentration (expressed in molar ratio protein to PEG) in dependence to the protein concentration used is another important driving factor of the reaction towards pre-dominant di-PEG IFN OC 2a product formation.
  • Molar ratios of protein to PEG of 1:15 are directing the reactions towards di-PEGylation .
  • di-pegylated IFN ⁇ 2a conjugates were separated from other IFN OC 2a isoforms by preparative cation exchange chromatography as described in example A 5 applying the same linear salt gradient as used for C-IEX HPLC analytics. Each elution peak was analytically re-chromatographed (C-IEX HPLC) to identify the corresponding pegylated IFN oc 2a isoforms (see above).
  • Table 12 Allocation of elution peaks from the preparative C-IEX run with corresponding retention time and allocated PEG IFN OC 2a species (isoforms).
  • di-pegylated IFN alpha 2a For di-pegylated IFN alpha 2a, one major di-PEG IFN ⁇ 2a peak could be identified. However, the shape of the peak lets assume, that potentially more than one out of 66 possible isoforms are eluting at the same time. A potential product for IFN would most likely consist of at least three isoforms because the corresponding peak revealed two shoulders.
  • an absorbance of 0.815 at 280 nm corresponds to 1 mg/ml protein.
  • the same molar extinction coefficient was applied.
  • the stopped reaction mixture was diluted three times with 40 mM acetate buffer (pH 3.8) and passed through a column packed with SP- 5PW (75 x 7.5 mm, 10 ⁇ m particle size, 3.3 ml bed volume, manufactured by Tosohaas Bioscience), mounted on a Dionex system or Shimadzu HPLC system.
  • SP- 5PW 75 x 7.5 mm, 10 ⁇ m particle size, 3.3 ml bed volume, manufactured by Tosohaas Bioscience
  • the column was equilibrated in 40 mM NaAc, pH 3.8 and 20% (v/v) isopropanol.
  • the flow rate was 1 ml/min and typically 200 ⁇ g of protein were injected.
  • a linear gradient elution was done with 1 M NaCl in 40 mM NaAc, pH 3.8 containing 20% (v/v) isopropanol in approximately 20 column volume (see Figure 18c).
  • the column was run at room temperature; however, the samples were maintained at 5°C in the auto- injector.
  • the elution peaks of the analytical C-IEX HPLC were mass analysed (MALDI-TOF) and analysed by SDS PAGE, which allowed the allocation of each pegylated EPO peak to the retention time of an elution peak.
  • Table 13 Comparison of calculated and experimentally determined molecular weight of different pegylated EPO species.
  • a retention time could be allocated to the corresponding PEG EPO species.
  • the analytic C-IEX allowed the separation of mono - PEG -, di-PEG- and tri-PEG EPO isoforms in dependence of the retention time.
  • the PEG concentration (expressed in molar ratio protein to PEG, or better in molar excess to the available amino groups of the target protein for PEG attachment) in dependence to the protein concentration used is another important driving factor of the reaction towards pre- dominant di-PEG erythropoietin product formation.
  • Molar ratios of target protein to PEG of about 1 :5, or better a stoichiometric ratio of the number of amino groups within the protein and the number of PEG molecules of at least 1:0.5 are directing the reactions towards di-PEGylation
  • 12 kDa mPEG acetal was activated for the pegylation reaction as described under example A 1.
  • the lyophilized human growth hormone sample was dissolved in water, dialysed against bicine buffer and finally concentrated by ultrafiltration to a concentration of approximately 4-8 mg/ml depending on the reaction conditions using centricon tubes (2 ml) equipped with YM 10 membranes of 10 kDa molecular weight cut off.
  • an absorbance of 0.72 at 280 nm corresponds to 1 mg/ml protein.
  • the same molar extinction coefficient was applied. 25 mg solid mPEG aldehyde (molecular weight 12 kDa) were added to 6.0 mg/ml hGH in 100 mM bicine buffer, pH 7.5, resulting in a 1:15 molar ratio of hGH to mPEG aldehyde. 20 mM sodium cyanoborohydride were added, and the reaction mixture was further incubated by gentle stirring at 2-8 0 C for several hours. The reaction was stopped by acidification with 8 N HCl.
  • the species distribution (mono-, di-, tri-) in the PEG hGH reaction mixture was determined by C-IEX HPLC. Therefore, the stopped reaction mixture was diluted for three times with 40 mM NaAc buffer (pH 4.2) and passed through a column packed with SP- 5PW (75 x 7.5 mm, 10 ⁇ m particle size, 3.3 ml bed volume, manufactured by Tosohaas Bioscience), mounted on a Dionex system or Shimadzu HPLC system.
  • the column was equilibrated with the same buffer, the flow rate was 0.4 ml/min and typically 200 ⁇ g of protein were injected.
  • the column was run at room temperature; however, the samples were maintained at 5°C in the auto- injector.
  • the elution peaks of the analytical C-IEX HPLC were mass analysed by MALDI/TOF and SDS PAGE which allowed the allocation of each pegylated hGH peak to the retention time of an elution peak.
  • Table 14 Comparison of calculated and experimentally determined molecular weight of different pegylated human growth hormone species. n.d. - not determined
  • a retention time could be allocated to the corresponding PEG hGH species.
  • the analytical C-IEX allowed the separation of mono -PEG -, di-PEG- and tri-PEG hGH isoforms in dependence of the retention time.
  • converting recombinant human hGH with mPEG aldehyde at molar ratios of 1:15 to PEG-hGH conjugates yielded predominantly di-pegylated hGH conjugates from a reaction time of approximately 15 hours onwards (see Figure 19).
  • the molar ratio of 1:15 corresponds to a stoichiometric ratio of the number of amino groups (nine) within the protein to the number of PEG molecules of 1:1.6.
  • Figure 1 Overview of all process steps in the production of di-polymer protein conjugates according to the present invention.
  • Figure 2 Maximum yield of all di-PEG GCSF conjugate isoforms and kinetic of formation as function of time-dependent low temperature storage of activated mPEG aldehyde used for the pegylation reactioa
  • Figure 3a Sequential formation of recombinant G-CSF conjugates with one to four strands of 12 kDa mPEG aldehyde in time- dependent manner. Maximum di-PEG G- CSF yield at 30 h incubation.
  • Figure 3b Kinetic of product formation of di-PEG G-CSF conjugate product being a mixture of two isoforms in dependence of the amounts of protein and PEG used.
  • Various amounts of PEG and protein were tested according to the invention at pH 7.5, 5°C, gentle stirring in 0.5 mL reaction volume.
  • Figure 3d Influence of temperature (22°C versus 5°C) on the kinetic of product formation, di-PEG G-CSF conjugate mixture, during reaction (5.7 mg/ml protein, molar ratios as indicated in 0.5 mL reaction volume).
  • Figure 4 A typical preparative C-IEX elution profile for the separation of di-PEG G-CSF from the pegylation reaction mixture (protein load: 6.1 mg pegylated G-CSF species; column: SP-5PW C-IEX column (Tosohaas, 20 ⁇ m, 10.6 ml bed volume), Buffer A: 15mM NaAc, pH 4.2; Buffer B: 1 M NaCl in 15 mM NaAc, pH 4.2; linear gradient); E- eluat, RP - rest pool.
  • protein load 6.1 mg pegylated G-CSF species
  • column SP-5PW C-IEX column (Tosohaas, 20 ⁇ m, 10.6 ml bed volume)
  • Buffer A 15mM NaAc, pH 4.2
  • Buffer B 1 M NaCl in 15 mM NaAc, pH 4.2
  • linear gradient E- eluat, RP - rest pool.
  • Figure 5 C-IEX HPLC separation of Di-PEG G-CSF product. Chromatogramm recorded at 214 nm. Isoforms are labelled.
  • Figure 6 SEC separation of di-PEG G-CSF product. Chromatogramm recorded at 214 nm. Peak at 23 min corresponds to salt.
  • FIG. 7 CD spectrum of non- modified G-CSF (black), and di-PEG G-CSF product (red).
  • Figure 8 In vivo bioactivity of the di-PEG G-CSF conjugate product comprising a mixture (0.76:1) of di- and di'-PEG G-CSF conjugate isoforms in comparison to unmodified rhG-CSF (Neupogen ®) and mono-pegylated rhG-CSF (Neulasta®) illustrated by the average WBC after single subcutaneous injection (PBS - phosphate buffered saline).
  • Figure 9 In vivo bioactivity of the di-PEG G-CSF conjugate product comprising a mixture (0.76: 1) of di- and di' -PEG G-CSF conj ugate isoforms in comparison to unmodified rhG-CSF (Neupogen ®) and mono-pegylated rhG-CSF (Neulasta®) illustrated by the average neutrophil counts after single subcutaneous injection (PBS - phosphate buffered saline).
  • Figure 10 In vivo bioactivities of the di-PEG G-CSF isoform and the di'-PEG G- CSF isoform in comparison to to di-PEG G-CSF conjugate product being a mixture of both isoforms (0.76:1) illustrated by the average WBC after single subcutaneous injection (PBS - phosphate buffered saline).
  • Figure 11 In vivo bioactivities of di-PEG G-CSF isoform and di'-PEG G-CSF isoform in comparison to di-PEG G-CSF conjugate product being a mixture of both isoforms (0.76:1) illustrated by the average neutrophil counts after single subcutaneous injection (PBS - phosphate buffered saline).
  • Figure 12 Analytical C-IEX elution profiles of reaction mixtures after 30h pegylation of G-CSF (3.2 mg/ml) with mPEG 12kDa (molar ratio 1:25, upper chromatogram) and with mPEG 2OkDa (molar ratio 1:19, lower chromatogram) analysed on a Dionex system equipped with a SP 5PW column under identical conditions (UV 214 nm).
  • the peak profiles are matching, which allows i) the peak identification and ii) peak allocation to the corresponding pegylated G-CSF species (mono- di- and di' isoforms).
  • Figure 13 Sequential formation of recombinant G-CSF conjugates with one to four strands of 2OkDa mPEG aldehyde in time- dependent manner. Predominantly di-PEG G-CSF yield is achieved from 18h incubation onwards. The reaction conditions were 3.2 mg/ml G-CSF with 1:19 molar ratio G-CSF to PEG 20k, 5°C, gentle stirring.
  • Figure 14 In vivo bioactivities of Di'-PEG 20k G-CSF isoform in comparison to unmodified rhG-CSF (Neupogen®) and mono -pegylated rhG-CSF (Neulasta®) illustrated by the average WBC after single subcutaneous injection (PBS - phosphate buffered saline).
  • Figure 15 In vivo bioactivities of Di'-PEG 2OkDa G-CSF isoform in comparison to unmodified rhG-CSF (Neupogen®) and mono -pegylated rhG-CSF (Neulasta®) illustrated by the average neutrophil counts after single subcutaneous injection (PBS
  • Figure 16a Sequential formation of recombinant IFN alpha 2a conjugates with one to three strands of 12 kDa mPEG aldehyde in time- dependent manner. Predominantly di-PEG IFN alpha 2a yield is already achieved after approximately 2Oh incubation, and lasts over the whole investigated time frame of 50 hours.
  • Figure 16b Kinetic of product formation of di-PEG IFN ⁇ 2a conjugate product in dependence of the amounts of protein and PEG used.
  • Various amounts of PEG and protein were tested according to the invention at pH 7.5, 5°C, gentle stirring in 0.5 mL reaction volume.
  • Figure 17 A preparative C-IEX elution profile for the separation of di-PEG IFN alpha 2a conjugate from the pegylation reaction mixture (protein load: 6 mg pegylated IFN alpha 2a after 24h incubation; column: SP-5PW C-IEX column (Tosohaas, 20 ⁇ m, 10.6 ml bed volume), Buffer A: 4OmM NaAc, pH 4.2; Buffer B: 1 M NaCl in 40 mM NaAc, pH 4.2; linear gradient); Pl - Mono-PEG; P2 - Di-PEG; P3
  • Figure 18a Sequential formation of recombinant EPO conjugates with one to three strands of 12 kDa mPEG aldehyde in time- dependent manner. Predominantly di-PEG EPO yield is surprisingly achieved within the first 18h of incubation and lasts for the rest of the incubation
  • Figure 18b Kinetic of product formation of di-PEG erythropoietin conjugate product in dependence of the amounts of protein and PEG used.
  • Various amounts of PEG and protein were tested according to the invention at pH 7.5, 5°C, gentle stirring in 0.5 mL reaction volume.
  • Figure 18c Analytical C-IEX elution profile for the separation of di-PEG EPO conjugate from the pegylation reaction mixture (protein load: 1 mg pegylated EPO mixture after 24h incubation; column: SP-5PW C-IEX column (Tosohaas, 10 ⁇ m, 3.3 ml bed volume), Buffer A: 4OmM NaAc, pH 3.8, 20 % Isoporpanol; Buffer B: 1 M NaCl in 40 mM NaAc, pH 3.8; 20% Isopropanol, linear gradient); Pl - Mono- PEG; P2 - Di-PEG; P3 - Tri-PEG
  • Figure 19 Sequential formation of recombinant hGH conjugates with one to three strands of 12 kDa mPEG aldehyde in time- dependent manner. Pre-dominantly di- PEG hGH yield is achieved after approximately 15h incubation, and lasts.
  • Figure 20 SEQ ID No. 1 (amino acid sequence of G-CSF including methionine at position 1).

Abstract

La présente invention porte sur des conjugués protéine-di-polymère et sur des procédés de préparation de ceux-ci. Cette invention porte également sur l'utilisation d'un tel conjugué di-polymère-protéine, en particulier d'un conjugué de protéine di-pégylée, pour fabriquer un médicament servant au traitement de troubles.
EP06725029A 2005-03-11 2006-03-13 Conjugues di-polymere-proteine et procedes de preparation de ceux-ci Ceased EP1869079A2 (fr)

Priority Applications (1)

Application Number Priority Date Filing Date Title
EP06725029A EP1869079A2 (fr) 2005-03-11 2006-03-13 Conjugues di-polymere-proteine et procedes de preparation de ceux-ci

Applications Claiming Priority (3)

Application Number Priority Date Filing Date Title
PCT/EP2005/002632 WO2006094530A1 (fr) 2005-03-11 2005-03-11 Conjugues de proteines et de dipolymeres et leurs procedes de preparation
PCT/EP2006/060671 WO2006095029A2 (fr) 2005-03-11 2006-03-13 Conjugues di-polymere-proteine et procedes de preparation de ceux-ci
EP06725029A EP1869079A2 (fr) 2005-03-11 2006-03-13 Conjugues di-polymere-proteine et procedes de preparation de ceux-ci

Publications (1)

Publication Number Publication Date
EP1869079A2 true EP1869079A2 (fr) 2007-12-26

Family

ID=38719793

Family Applications (1)

Application Number Title Priority Date Filing Date
EP06725029A Ceased EP1869079A2 (fr) 2005-03-11 2006-03-13 Conjugues di-polymere-proteine et procedes de preparation de ceux-ci

Country Status (1)

Country Link
EP (1) EP1869079A2 (fr)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
EP2205281A1 (fr) * 2007-08-16 2010-07-14 Pharmaessentia Corp. Conjugués protéine-polymère
EP2313457A2 (fr) * 2008-07-31 2011-04-27 Pharmaessentia Corp. Conjugués peptide-polymère

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
See references of WO2006095029A3 *

Cited By (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
EP2205281A1 (fr) * 2007-08-16 2010-07-14 Pharmaessentia Corp. Conjugués protéine-polymère
EP2205281A4 (fr) * 2007-08-16 2014-10-08 Pharmaessentia Corp Conjugués protéine-polymère
EP4129343A1 (fr) * 2007-08-16 2023-02-08 PharmaEssentia Corp. Conjugués protéine-polymère
EP2313457A2 (fr) * 2008-07-31 2011-04-27 Pharmaessentia Corp. Conjugués peptide-polymère
EP2313457A4 (fr) * 2008-07-31 2014-10-01 Pharmaessentia Corp Conjugués peptide-polymère

Similar Documents

Publication Publication Date Title
WO2006095029A2 (fr) Conjugues di-polymere-proteine et procedes de preparation de ceux-ci
EP0822199B1 (fr) Polypeptides monopegylés à l'extrémité N-terminale, et procédé pour leur préparation
JP5336372B2 (ja) Ｇ−ｃｓｆ部位特異的モノコンジュゲート
EP1768700B1 (fr) Conjugues comprenant gm-csf et un polymer
ZA200105932B (en) GCSF conjugates.
CA2617064A1 (fr) Conjugues d'un groupe fonctionnel g-cfs et d'un polymere
EP1834963A1 (fr) Conjugues de proteines et de dipolymeres et leurs procedes de preparation
US6956027B2 (en) N-terminally chemically modified protein compositions and methods
EP1869079A2 (fr) Conjugues di-polymere-proteine et procedes de preparation de ceux-ci
AU2004235682B2 (en) N-Terminally chemically modified protein compositions and methods
WO2012064845A2 (fr) Conjugués g-csf-polymère ayant une liaison libérable
AU741659B2 (en) N-terminally chemically modified protein compositions and methods

Legal Events

Date Code Title Description
PUAI Public reference made under article 153(3) epc to a published international application that has entered the european phase

Free format text: ORIGINAL CODE: 0009012

17P Request for examination filed

Effective date: 20071011

AK Designated contracting states

Kind code of ref document: A2

Designated state(s): AT BE BG CH CY CZ DE DK EE ES FI FR GB GR HU IE IS IT LI LT LU LV MC NL PL PT RO SE SI SK TR

DAX Request for extension of the european patent (deleted)
STAA Information on the status of an ep patent application or granted ep patent

Free format text: STATUS: THE APPLICATION HAS BEEN REFUSED

18R Application refused

Effective date: 20081003