HK1085670A - G-csf conjugates - Google Patents

Description

G-CSF conjugates

Technical Field

The present invention relates to novel polypeptides capable of exhibiting granulocyte colony stimulating factor (G-CSF) activity, to conjugates of polypeptides exhibiting G-CSF activity with non-polypeptide components, to methods of preparing such polypeptides or conjugates and to the use of said polypeptides or conjugates in therapy, in particular in the treatment of neutropenia or leukopenia.

Background

The process of leukocyte growth, division and differentiation in bone marrow is known as hematopoiesis (Dexter and spoonacer, ann.rev.cell.biol., 3: 423, 1987). Each type of blood cell is derived from a pluripotent stem cell. Three types of blood cells are generally produced in vivo: red blood cells (erythrocytes), platelets and white blood cells (leukocytes), the majority of the latter being involved in the host's immune defenses. Proliferation and differentiation of hematopoietic precursor cells is regulated by a family of cytokines including colony stimulating factors (CSF's) such as G-CSF and interleukins (Arai et al, Ann. Rev. biochem., 59: 783-836, 1990). The major biological role of G-CSF in vivo is to stimulate the growth and development of certain white blood cells called neutrophils (Welte et al, PNAS-USA 82: 1526-. When neutrophils are released into the bloodstream, these cells function to fight bacterial and other infections.

Nagata et al, nature 319: 415, 418, 1986 reported the amino acid sequence of human G-CSF (hG-CSF). hG-CSF is a monomeric protein that dimerizes the G-CSF receptor by forming a 2: 2 complex of 2G-CSF molecules and 2 receptors (Horn et al, (1996), Biochemistry35 (15): 4886-96). Aritomi et al, nature 401: 713-717, 1999 describes the X-ray structure of the complex formed between the BN-BC domain of the G-CSF receptor and hG-CSF. They identified the following residues of hG-CSF as part of the receptor binding interface: g4, P5, a6, S7, S8, L9, P10, Q11, S12, L15, K16, E19, Q20, L108, D109, D112, T115, T116, Q119, E122, E123, and L124. Expression of rhG-CSF in E.coli, s.cerevisiae and mammalian cells has been reported (Souza et al, Science 232: 61-65, 1986, Nagata et al, Nature 319: 415-.

Both leukopenia (a decrease in leukocyte levels) and neutropenia (a decrease in neutrophil levels) are diseases that result in increased susceptibility to various types of infections. Neutropenia can be chronic (e.g., in HIV-infected persons) or acute (e.g., in cancer patients receiving chemotherapy or radiation therapy). For critically ill patients with neutropenia (e.g., due to chemotherapy), even very subtle infections can be severe and even life-threatening. Recombinant human G-CSF (rhG-CSF) is commonly used to treat various forms of leukopenia. Thus, commercial preparations of rhG-CSF known by the names filgrastim (Gran * and Neupogen *), lenograstim (neutrogen * and Granocyte *) and nartograstim (Neu-up *) are available. Gran * and Neupogen * are in a non-glycosylated form and can be prepared in recombinant e. Neutrogin * and Granocyte * are glycosylated forms that can be prepared in recombinant CHO cells, and Neu-up * is a non-glycosylated form with 5 amino acids substituted in the N-terminal region of the complete rhG-CSF produced in recombinant E.coli cells.

Various protein engineered variants of hG-CSF have been reported in the literature (US5,581,476, US5,214,132, US5,362,853, US4,904,584 and Riedhaar-Olson et al, Biochemistry 35: 9034-. In order to introduce at least one more sugar chain than the native polypeptide, it has been proposed to modify hG-CSF and other polypeptides (US5,218,092). The amino acid sequence of the polypeptide may be modified by amino acid substitution, amino acid deletion or amino acid insertion to add an additional sugar chain. In addition, polymer modifications of native hG-CSF, including attachment of PEG groups, have been reported (Satake-Ishikawa Cell Structure and Function 17: 157-160, 1992, U.S. Pat. No. 5,824,778, U.S. Pat. No. 5,824,784, WO96/11953, WO95/21629, WO94/20069, EP0612846A1, WO 01/00011).

Bowen et al, Experimental biology 27(1999), 425-432 disclosed a relationship between the molecular mass of PEG-coupled G-CSF muteins and the duration of their activity. Suggesting that there is a clear inverse relationship between the molecular weight of the PEG-conjugated moiety on the protein and the in vitro activity, whereas the in vivo activity increases with increasing molecular weight. It is postulated that the lower affinity of the conjugate has an effect of increasing half-life, since receptor-mediated endocytosis is an important mechanism for modulating hematopoietic growth factor levels.

Commercially available rhG-CSF has a short-term pharmacological effect and must therefore be administered once a day for the duration of the leukopenic state. Molecules with longer circulating half-lives will reduce the number of administrations necessary to alleviate leukopenia and prevent subsequent infections. Another more significant problem with the prior art rG-CSF products is that even when G-CSF is administered to a patient, symptoms of neutropenia occur following chemotherapy. For these patients, it is important to minimize the duration and extent of neutropenia, so as to minimize the risk of serious infections. Yet another problem with existing rG-CSF products is the occurrence of dose-dependent bone pain. Since bone pain is a significant side effect experienced by patients receiving rG-CSF therapy, it would be desirable to be able to provide an rG-CSF product which does not cause bone pain, which product may not have this effect on its own or which is effective in a sufficiently small dose that it does not cause bone pain. Thus, there is a clear need for improved recombinant G-CSF-like molecules.

Increase of protein with respect to half-lifeOne way of ensuring protein clearance, particularly through renal clearance and a decrease in receptor-mediated clearance, is by the circulatory half-life of (c). This can be achieved by coupling the protein with a chemical component that increases the apparent molecular weight, thereby reducing renal clearance and increasing the in vivo half-life of the molecule. Further, linking the protein to a chemical component can effectively block the physical contact of the proteolytic enzyme with the protein, thereby protecting the protein from non-specific proteolytic degradation. Polyethylene glycol (PEG) is one such chemical component that has been used to prepare therapeutic protein preparations. Recently, a G-CSP molecule (Neula sta) with a single 20kDa PEG group attached to the N-terminus^TM) Has been approved for sale in the united states. Although this pegylated G-CSF molecule has an increased half-life compared to non-pegylated G-CSF and is therefore administered less frequently than existing G-CSF products, it does not significantly reduce the duration of neutropenia compared to non-pegylated G-CSF. Thus, there is still considerable room for improvement in the currently known G-CSF molecules.

Therefore, there remains a need to provide novel molecules which exhibit G-CSF activity useful in the treatment of leukopenia/neutropenia, while having improved properties such as increased half-life, in particular, reduced duration of neutropenia. The present invention relates to such molecules.

Brief description of the invention

The present invention relates to specific conjugates comprising a polypeptide exhibiting G-CSP activity and a non-polypeptide moiety, to processes for the preparation of such conjugates and to their use in medical therapy and in the manufacture of medicaments. Accordingly, the first aspect of the invention relates to specific conjugates comprising a polypeptide exhibiting G-CSF activity having the amino acid sequence shown in SEQ ID NO: 1 differs in at least one specifically altered amino acid sequence comprising the amino acid residues of a non-polypeptide moiety linking group, and at least one non-polypeptide moiety is linked to the linking group of the polypeptide. These conjugates have a greatly reduced in vitro biological activity, in addition to an increased in vivo half-life, compared to unconjugated hG-CSF, which we have surprisingly found results in a more rapid recovery of neutrophils.

In another aspect, the invention relates to a polypeptide exhibiting G-CSF activity and forming part of a conjugate of the invention. It is contemplated that the polypeptides of the invention may be used for therapeutic, diagnostic or other purposes, but may also be particularly useful as intermediates in the preparation of the conjugates of the invention.

Another aspect of the invention relates to a polypeptide conjugate comprising a polypeptide exhibiting G-CSF activity, the amino acid sequence of which differs from the amino acid sequence of hG-CSF (the amino acid sequence of which is set forth in SEQ ID NO: 1) in that at least one amino acid residue is an introduced or removed amino acid residue comprising a non-polypeptide moiety linking group, and a sufficient number or type of non-polypeptide moieties provides a conjugate with an increased half-life and/or a more rapid recovery of neutrophils compared to known recombinant G-CSF products.

A particular aspect of the invention relates to a polypeptide conjugate having G-CSF activity comprising a polypeptide that hybridizes to SEQ ID NO: 1, having substitutions K16R, K34R, K40R, T105K, and S159K, a substitution such as R, K or Q at position H170, or a substitution at a position similar to SEQ ID NO: 1 having at least 80% sequence identity, and 2-6 (typically 3-6) polyethylene glycol moieties having a molecular weight of about 1000-10,000Da are attached to one or more attachment groups of the polypeptide. Wherein the substitutions are relative to the substitutions of SEQ ID NO: 1, typically at least about 90% or 95%, such as at least about 96%, 97%, 98% or 99%, of the sequence identity.

Yet another aspect of the invention relates to a method for preparing a conjugate of the invention, comprising a nucleotide sequence encoding a polypeptide of the invention, an expression vector comprising such a nucleotide sequence, and a host cell comprising the above nucleotide sequence or expression vector.

A final aspect of the invention relates to a composition comprising a conjugate or polypeptide of the invention, a method of preparing a pharmaceutical composition, the use of a conjugate or composition of the invention as a medicament, and a method of treating a mammal with the above composition. In particular, the polypeptides, conjugates or compositions of the invention may be used to prevent infection in cancer patients undergoing certain types of radiotherapy, chemotherapy, and bone marrow transplantation, to mobilize progenitor cell accumulation in peripheral blood progenitor cell transplantation, to treat severe chronic or related leukopenia due to whatever cause, and to support treatment of acute myeloid leukemia (acute myeloloid leukemia) patients. In addition, the polypeptides, conjugates or compositions of the invention may be used in the treatment of AIDS or other immunodeficiency disorders, as well as bacterial infections.

Detailed Description

Definition of

In the context of the present application and invention the following definitions apply:

the term "conjugate" refers to a heterologous molecule formed by one or more polypeptides, usually a single polypeptide, covalently linked to one or more non-polypeptide components, especially polymer molecules. Furthermore, the conjugate may be attached to one or more saccharide moieties, especially via N-or O-glycosylation. The term covalently linked refers to a polypeptide and a non-polypeptide component being covalently linked to each other directly or indirectly through one or more intervening components, such as a bridge, spacer arm, or linking component. Preferably, the conjugate is soluble at the respective concentrations and conditions, i.e. in physiological body fluids (e.g. blood). The term "unconjugated polypeptide" may be used to refer to the polypeptide portion of the conjugate.

The term "polypeptide" is used interchangeably with the term "protein" in the present invention.

The term "polymer molecule" is a molecule formed by the covalent attachment of two or more monomers, wherein none of the monomers is an amino acid residue, except where the polymer is human albumin or another high abundance plasma protein. The term "polymer" may be used interchangeably with the term "polymer molecule". The term will include sugar molecules, but generally, the term does not include such sugar molecules attached to the polypeptide by N-or O-glycosylation in vivo (described further below), as such molecules are hereinafter referred to as "oligosaccharide components". Unless the number of polymer molecules is explicitly stated, each reference to "a polymer", "a polymer molecule", "polymer" or "polymer molecule" contained in a polypeptide of the invention or the above-mentioned references as applied in the present invention refers to one or more polymer molecules.

The term "linking group" refers to a group of amino acid residues in a polypeptide that are capable of coupling to a non-polypeptide component of interest. For example, in the case of polymer coupling, especially with PEG coupling, a commonly used linking group is the epsilon amino group or the N-terminal amino group of lysine. Other polymer linking groups include free carboxylic acid groups (e.g., free carboxylic acid groups of the C-terminal amino acid residue or free carboxylic acid groups of aspartic acid or glutamic acid residues), suitably activated carbonyl groups, oxidized sugar components, and sulfhydryl groups. Useful linkers and their paired non-peptide components are shown in the following table.

Linking group	Amino acids	Examples of non-peptide Components	Coupling method/-activated PEG	Reference to the literature
					-NH2	N-terminal Lys, His, Arg	Polymers, e.g. PEG, with amido or imino groups	mPEG-SPATresylatedmPEG	Shearwater Inc. Delgado et al, Critical reviews in therapeutic drug carriers Systems 9(3, 4): 249-304(1992)
-COOH	C-terminal, Asp, Glu	Polymers, e.g. PEG, with ester or amido oligosaccharide components	mPEG-Hz in vitro coupling	Shearwater Inc.

-SH	Cys	Polymers, e.g. PEG, oligosaccharide components with disulfide bonds, maleimide or vinylsulfone groups	PEG-vinyl sulfone PEG-maleimide in vitro coupling	Shearwater Inc. Delgado et al, Critical reviews in therapeutic drug carriers Systems 9(3, 4): 249-304(1992)
					-OH	Ser，Thr，-OH，Lys	The oligosaccharide component has PEG of ester, ether, carbamate, and carbonate	In vivo O-linked glycosylation
-CONH2	Asn as part of N-glycosylation site	Oligosaccharide component polymers, e.g. PEG	In vivo N-glycosylation
					Aromatic residue	Phe，Tyr，Trp	Oligosaccharide component	In vitro coupling
-CONH2	Gln	Oligosaccharide component	In vitro coupling	Yan and Wold，Biochemistry，1984，Jul 31；23(16)：3759-65
					Aldehyde ketones	Oxidized oligosaccharides	Polymers, e.g. PEG, PEG-hydrazine	PEGylation	Andresz et al, 1978, makromol. chem.179: 301, WO92/16555, WO02/23114
Guanidine (guanidine)	Arg	Oligosaccharide component	In vitro coupling	Lundblad and Noyes，ChemicalReagents for ProteinModification，CRC Press Inc.，Florida，USA
					Imidazole ring	His	Oligosaccharide component	In vitro coupling	Same as guanidine

For N-glycosylation in vivo, the term "linker" is used in an unconventional manner to denote the amino acid residues that constitute the N-glycosylation site (with the sequence N-X ' -S/T/C-X ", wherein X ' is any amino acid residue other than proline, X" can be any amino acid residue that is the same or different from X ' and preferably is not proline, N is asparagine, and S/T/C can be serine, threonine or cysteine, preferably serine or threonine, most preferably threonine). Although the asparagine residue of the N-glycosylation site is the site to which the oligosaccharide component is attached during glycosylation, this attachment cannot be achieved unless other amino acid residues of the N-glycosylation site are present. Thus, when the non-peptide component is an oligosaccharide component and coupling is achieved via N-glycosylation, the term "amino acid residue comprising an attachment group for the non-peptide component" in connection with a change in the amino acid sequence of the polypeptide of interest is to be understood as meaning that one or more amino acid residues constituting an N-glycosylation site are changed in such a way that a functional N-glycosylation site is introduced into or removed from the amino acid sequence.

In the present application, the names of Amino Acids and atomic names (e.g., CA, CB, NZ, N, O, C, etc.) are used according to the definitions of the Protein Database (PDB) (www.pdb.org), these names being named according to the IUPAC Nomenclature (IUPAC Nomebrature and Symbilism for Amino Acids and peptides (residual proteins, atom names etc.), Eur.J.biochem.138, 9-37(1984) and their corrections in Eur.J.biochem.152, 1(1985) the term "Amino acid residue" refers primarily to any natural or unnatural Amino acid residue, in particular the Amino Acids of the 20 natural Amino Acids, i.e., alanine (Ala or A), cysteine (Ala or C), aspartic acid (Cys or D), glutamic acid (Glu or E), phenylalanine (Phe or F), glycine (Gly or G), histidine (His or H), isoleucine (His or H), Lys (Lys or L (K or L), lysine (Lys or L) (K or L) (lysine), methionine (Met or M), asparagine (Asn or N), proline (Pro or P), glutamine (Gln or Q), arginine (Arg or R), serine (Ser or S), threonine (Thr or T), valine (Val or V), tryptophan (Trp or W) and tyrosine (Tyr or Y) residues.

The terms used to define the position/substitution of an amino acid are exemplified below: f13 means that position number 13 in the reference amino acid sequence is occupied by a phenylalanine residue. F13K means that the phenylalanine residue at position 13 has been replaced with a lysine residue. Unless otherwise indicated, the numbering of amino acid residues herein is similar to that of SEQ ID NO: 1 corresponds to the amino acid sequence of hG-CSF. Alternative substitutions are indicated with "/", e.g., Q67D/E refers to an amino acid sequence in which glutamine at position 67 is substituted with aspartic acid or glutamic acid. Multiple substitutions are indicated by "+", e.g., S53N + G55S/T refers to an amino acid sequence in which the serine residue at position 53 is substituted with an asparagine and the glycine residue at position 55 is substituted with a serine or threonine residue.

The term "nucleotide sequence" refers to a contiguous stretch of two or more nucleotide molecules. The nucleotide sequence may be of genomic, cDNA, RNA, semisynthetic or synthetic origin, or any combinations thereof.

The term "polymerase chain reaction" or "PCR" generally refers to a known method for amplifying a desired nucleotide sequence in vitro using a thermostable DNA polymerase.

"cell," "host cell," "cell line," and "cell culture" are used interchangeably herein and all such terms are intended to include the progeny of the cell after it is grown or cultured. "transformation" and "transfection" are used interchangeably and refer to the process of introducing DNA into a cell.

"operably linked" means that two or more nucleotide sequences are joined enzymatically or otherwise covalently to form a configuration relative to each other such that the sequences function properly. For example, if the pre-sequence or secretory leader is expressed as a pre-protein involved in secretion of the polypeptide, then the nucleotide sequence encoding the pre-sequence or secretory leader is operably linked to the nucleotide sequence of the polypeptide: a promoter or enhancer is operably linked to a coding sequence if it affects the transcription of the sequence; if the ribosome binding site is located at a position that facilitates translation, the ribosome binding site is operably linked to the coding sequence. Generally, "operably linked" means that the nucleotide sequences being linked are contiguous and, in the case of a secretory leader, contiguous and in reading order. Ligation may be conveniently accomplished at the restriction enzyme site. If such sites are not present, it may be desirable to use synthetic oligonucleotide adaptors or linkers in conjunction with standard recombinant DNA methods.

The term "introducing" refers to introducing an amino acid residue comprising a non-polypeptide moiety linking group, particularly by substitution of an existing amino acid residue, or by insertion of an additional amino acid residue. The term "remove" refers to the removal of an amino acid residue comprising a non-polypeptide moiety linking group, in particular by substitution of that amino acid residue with another amino acid residue, or alternatively by deletion (without substitution) of a pre-removed amino acid residue.

When substitutions are made to the parent polypeptide, the substitutions are preferably "conservative substitutions", in other words substitutions between amino acids having similar characteristics, for example, small amino acids, acidic amino acids, polar amino acids, basic amino acids, hydrophobic amino acids and aromatic amino acids.

Preferred substitutions in the present invention are especially selected from the group of conservative substitutions listed in the following table

Conservative substitution set:

1	alanine (A) Glycine (G) serine (S) threonine (T)
		2	Aspartic acid (D) glutamic acid (E)
3	Asparagine (N) Glutamine (Q)
		4	Arginine (R) histidine (H) lysine (K)
5	Isoleucine (I) leucine (L) methionine (M) valine (V)
		6	Phenylalanine (F) tyrosine (Y) tryptophan (W)

The term "immunogenic" in conjunction with a given agent means that the agent is capable of inducing a response in the immune system. The immune response may be a cell-or antibody-mediated response (see, e.g., Roitt: Essential Immunology (8th Edition, Blackwell)), for further explanation of immunogenicity). Normally, a decrease in antibody reactivity would mean a decrease in immunogenicity. Reduced immunogenicity may be determined by applying any suitable method well known in the art, for example in vivo or in vitro.

The term "functional in vivo half-life" is used in its normal sense, i.e., the time at which 50% of the biological activity of the polypeptide or conjugate is still present in the body/target organ, or the time at which the activity of the polypeptide or conjugate is 50% of the original activity. Another method of determining functional half-life in vivo may be to determine "serum half-life", i.e. the time at which 50% of the polypeptide or conjugate molecules circulate in the plasma or bloodstream before being cleared. Other terms of serum half-life include "plasma half-life", "circulating half-life", "serum clearance", "plasma clearance", "clearance half-life". The polypeptide or conjugate is cleared by one or more of the reticuloendothelial system (RES), kidney, spleen or liver, by receptor-mediated degradation, or by specific or non-specific proteolysis, particularly by receptor-mediated clearance and renal clearance. In general, clearance depends on the size of the protein (cut-off relative to glomerular filtration), charge, attached sugar chains, and the presence of cellular receptors for the protein. The retained functionality is typically selected from proliferative or receptor binding activity. In vivo functional half-life and serum half-life may be determined by any suitable method well known in the art, as further discussed in the materials and methods below.

The term "increased" as used in reference to functional half-life or serum half-life in vivo means that the corresponding half-life of a conjugate or polypeptide is statistically increased relative to the half-life of a reference molecule, e.g., unconjugated hG-CSF (e.g., Neupogen *), as determined under comparable conditions. For example, the corresponding half-life may be increased by at least about 25%, such as at least about 50%, for example at least about 100%, 200%, 500% or 1000%.

The normal meaning of the term "renal clearance" is used to refer to any clearance of the kidney, for example by filtration of the glomeruli, secretion of the tubules or excretion of the tubules. Renal clearance depends on the physical characteristics of the conjugate, including size (diameter), symmetry, shape/rigidity and charge. The reduction in renal clearance can be determined by any suitable assay, for example, one of the existing in vivo assays. Typically, renal clearance is determined by administering a labeled (e.g., radiolabeled or fluorescently labeled) polypeptide conjugate to the patient and measuring the activity of the label in the collected urine of the patient. Under comparable conditions, a decrease in renal clearance can be determined compared to a related reference polypeptide, e.g., a corresponding unconjugated polypeptide, unconjugated corresponding wild-type polypeptide or another conjugated polypeptide (e.g., a conjugated polypeptide not according to the invention). Preferably, the renal clearance of the conjugate is reduced by at least 50%, preferably by at least 75%, most preferably by at least 90% compared to a related reference polypeptide.

Typically, activation of the receptor is associated with receptor-mediated clearance (RMC), such that binding of the polypeptide to its receptor in the inactive state does not result in RMC, whereas activation of the receptor results in RMC. Clearance is due to internalization of the receptor-bound polypeptide and subsequent lysosomal degradation. The RMC can be reduced by designing the conjugate to bind and activate a sufficient number of receptors to achieve an optimal in vivo biological response and to avoid activation of more receptors than is required to achieve the response. This may be reflected in reduced in vitro biological activity and/or increased detachment rate. In a preferred embodiment, the conjugates of the invention have a substantially reduced in vitro biological activity as compared to unconjugated hG-CSF.

In general, a decrease in vitro biological activity reflects a decrease in efficacy/efficiency and/or a decrease in effectiveness, which can be determined by any suitable method for determining any of these properties. For example, in vitro biological activity can be determined in luciferase-based assays ("preliminary assays 2"; see materials and methods). Another method of determining biological activity in vitro is to determine the binding affinity of the conjugates of the invention using a cell-based assay ("second step assay") as described in the materials and methods section.

It has now been found that the in vitro biological activity is lower than that of hG-CSF (SEQ ID NO: 1), which is beneficial for prolonging plasma half-life and for highly stimulating neutrophils. It was also surprisingly found that administration of the G-CSF conjugate of the invention having a lower in vitro biological activity restored neutrophils faster, i.e. the neutrophil count returned to normal levels faster, than administration of hG-CSF. This is an important finding because it is critical to shorten the duration of neutropenia as much as possible in patients with reduced levels of neutrophils due to radiation or chemotherapy, etc. Thus, in a preferred embodiment, the in vitro biological activity of the conjugates of the invention is about 2-30%, preferably about 3-25% of the biological activity of hG-CSF (hG-CSF used herein as the reference polypeptide has SEQ ID NO: 1 and may also have an N-terminal methionine residue; said reference hG-CSF may be, inter alia, Neupogen *, a non-glycosylated Met-hG-CSF) as determined by the luciferase assay described herein, or as determined using a cell-based receptor binding affinity assay ("second step assay"). Thus, the in vitro biological activity of the conjugate is preferably reduced by at least 70%, such as at least 75%, e.g. at least 80% or 85% as compared to the in vitro biological activity of hG-CSF, determined under comparable conditions. In other words, the conjugate has about 2%, typically at least about 3%, such as at least about 4% or 5% of the biological activity of the wild-type polypeptide in vitro. For example, the in vitro biological activity is about 4-20% of hG-CSF, as determined under comparable conditions. When it is desired to decrease the biological activity in vitro in order to reduce receptor mediated clearance, it is clear that sufficient biological activity must be maintained in order to obtain the desired receptor activation, which is why it has been mentioned above that said biological activity must be at least about 2% of hG-CSF, and preferably slightly higher.

It was found that amino acid changes, in particular substitutions, occur in the helical regions of G-CSF, i.e.in the amino acid residues selected from the group consisting of amino acid positions 11-41 (helix A), 71-95 (helix B), 102-125 (helix C), and 145-170 (helix D) (relative to SEQ ID NO: 1), if the resulting polypeptide and polyethylene glycol are presentAlcohol coupling, may result in reduced receptor-mediated clearance and thus increased in vivo half-life. In addition to the discovery of longer half-lives, it was also unexpectedly discovered that administration of such polypeptide conjugates was in combination with administration of commercial G-CSF products Neupogen * and Neulasta^TMIn contrast, it is also possible to stimulate the production of leukocytes and neutrophils at the same or even better levels. G-CSF conjugates having reduced in vitro biological activity may thus be prepared by altering (typically substituting) one or more amino acid residues in the helical region of G-CSF and conjugating the resulting polypeptide to one or more non-polypeptide components, such as polyethylene glycol.

Preferably, the rate of dissociation of the polypeptide conjugate and its receptor will be increased to such an extent as to result in release of the polypeptide conjugate from its receptor before any substantial endocytosis of the receptor-ligand complex occurs. Receptor polypeptide binding affinity can be determined as described in the materials and methods section herein. The release rate can be determined using the Biacore * technique described in the materials and methods section. In vitro RMC can be determined as follows: labeling (e.g., radioactive or fluorescent labeling) the polypeptide conjugate, stimulating cells comprising a receptor for the polypeptide, washing the cells, and detecting the activity of the label. Alternatively, the conjugate is exposed to cells expressing the relevant receptor. After an appropriate period of incubation, the supernatant is removed and transferred to culture wells containing similar cells. The biological response of these cells to the supernatant can be determined in comparison to the unconjugated polypeptide or another reference polypeptide, which is a measure of the extent of RMC reduction.

In general, the reduction in biological activity of the conjugate in vitro may be the result of its modification by a non-polypeptide moiety. However, modifications to the polypeptide portion of the conjugate may also be required to further reduce biological activity in vitro or for other reasons. For example, in one embodiment, at least one amino acid residue located at or near the receptor binding region of the polypeptide can be substituted with another amino acid residue to reduce biological activity in vitro, as compared to a corresponding wild-type polypeptide. The amino acid residue introduced by substitution may be any amino acid residue capable of reducing the in vitro biological activity of the conjugate. Conveniently, the amino acid residues introduced comprise a non-polypeptide moiety linking group as defined herein. In particular, when the non-polypeptide moiety is a polymer molecule such as PEG, the amino acid residue introduced may be a lysine residue.

The term "exhibiting G-CSF activity" means that the polypeptide or conjugate has native G-CSF, in particular having the amino acid sequence of SEQ ID NO: 1, including the ability to bind to the G-CSF receptor (Pukunaga et al, j.bio.chem, 265: 14008, 1990). G-CSF activity can be conveniently measured using the preliminary assay described in the materials and methods below. A polypeptide "exhibiting" G-CSF activity is considered to possess such activity when it exhibits a detectable function, such as a detectable proliferative activity or a receptor binding activity (such as those activities determined by the preliminary detection methods described in the materials and methods). The polypeptides showing G-CSF activity herein may also be referred to as "G-CSF molecules" for the sake of simplicity, even if the polypeptides are in fact G-CSF variants.

The term "parent G-CSF" or "parent polypeptide" refers to a molecule that is to be modified according to the invention. The parent G-CSF is typically hG-CSF or a variant thereof. A "variant" is a polypeptide that differs from the parent polypeptide in one or more amino acid residues, typically 1, 2,3, 4,5, 6, 7, 8, 9,10, 11, 12, 13, 14, or 15 amino acid residues. Examples of rhG-CSF include filgrastim (Gran * and eupogen *), lenograstim (neutrogen * and Granocyte *) and nartograstim (Neu-up *).

Conjugates of the invention

As mentioned above, the first aspect of the invention relates to a conjugate comprising a polypeptide exhibiting G-CSF activity comprising an amino acid sequence that is complementary to the amino acid sequence of SEQ ID NO: 1 in the sequence listing, differs in that at least one amino acid residue comprising a non-polypeptide moiety linking group is specifically introduced or removed. The amino acid residues introduced and/or removed are described further below. It will be appreciated that the conjugate itself also exhibits G-CSF activity.

By introducing and/or removing amino acid residues comprising a non-polypeptide moiety linking group, polypeptides can be directionally engineered to allow easier coupling of the polypeptide molecule to a selected non-polypeptide moiety, to optimize the coupling pattern (e.g., to ensure optimal distribution of the non-polypeptide moiety on the surface of the G-CSF molecule and to ensure that only the linking group to be coupled is present in the molecule), and to obtain new coupled molecules having both G-CSF activity and improved one or more properties compared to existing G-CSF molecules.

The polypeptide may be of any origin, in particular may be of mammalian, preferably human origin, in particular having the amino acid sequence SEQ ID NO: 1.

In a preferred embodiment of the invention, more than one amino acid residue of the polypeptide having G-CSF activity is altered, for example such alteration includes removal and introduction of an amino acid residue comprising a linker group for the selected non-polypeptide moiety.

In addition to the amino acid residue alterations disclosed herein which are intended to remove and/or introduce attachment sites for non-polypeptide components, it is to be understood that the amino acid sequence of the polypeptide of the present invention may, if desired, comprise other alterations, i.e., other substitutions, insertions or deletions, which are not related to the introduction or removal of attachment sites. These alterations may, for example, comprise truncation of one or more amino acid residues at the N-terminus and/or C-terminus, or addition of one or more additional residues at the N-terminus and/or C-terminus, for example addition of a methionine residue at the N-terminus.

The conjugates of the invention have one or more of the following improved properties compared to hG-CSF, in particular compared to rhG-CSF (e.g.filgrastim, lentigritim or artograstim) or known hG-CSF variants: increased ability to shorten the duration of neutropenia, increased functional in vivo half-life, increased serum half-life, decreased renal clearance, decreased receptor-mediated clearance, decreased side effects such as bone pain, and decreased immunogenicity.

It will be appreciated that the amino acid residue comprising the attachment group for the non-polypeptide moiety, whether it is introduced or removed, will be selected according to the nature of the non-polypeptide moiety selected, and in most cases, the method by which coupling is achieved between the polypeptide and non-polypeptide moieties. For example, when polyethylene glycol is coupled to lysine residues, suitable activating molecules are, for example, mPEG-SPA from Shearwater corp, oxycarbonyl-oxy-N-dicarboximide-PEG (US5,122,614), or PEG available from PolyMASC Pharmaceuticals plc. The first of these molecules will be described further below.

To avoid disruption of the structure and function of the parent hG-CSF molecule, the total number of amino acid residues that need to be altered according to the invention, e.g., as described below in the invention, will typically not exceed 15 (as compared to the amino acid sequence shown in SEQ ID NO: 1). The number and type of amino acid residues actually introduced or removed depends, inter alia, on the nature and degree of conjugation desired (e.g., the characteristics of the non-peptide component, how much of the non-polypeptide component is desired or is likely to be conjugated to the polypeptide, the site at which conjugation is desired or should be avoided, etc.). Preferably, the polypeptide part of the conjugate of the invention or the polypeptide of the invention has 1 to 15 amino acid residues, typically 2 to 10 amino acid residues, such as 3 to 8 amino acid residues, such as 4 to 6 amino acid residues, corresponding to the amino acid sequence of SEQ ID NO: 1 are different in comparison with the amino acid sequence shown in fig. Thus, typically the polypeptide part of the conjugate of the invention or the polypeptide of the invention has 1, 2,3, 4,5, 6, 7, 8, 9,10, 11, 12, 13, 14 or 15 amino acid residues other than SEQ ID NO: 1, or a pharmaceutically acceptable salt thereof.

Typically the polypeptide portion of the conjugate will have an amino acid sequence substantially identical to SEQ ID NO: 1, preferably at least about 90%, such as at least about 95%, for example at least about 96%, 97%, 98% or 99% sequence identity. Homology/identity of amino acid sequences can be determined conveniently, for example, by the program ClustalW, version 1.8, 1999, 6.using default parameters from the sequences being aligned (Thompson et al, 1994, ClustalW: Improving the sensitivity of developing multiple sequence alignment, Nucleic Acids Res.22: 4673 4680) or from the PFAM family database version 4.0 (http:// platform. w. /) (Nucleic Acids Res.1999 Jan 1; 27 (1): 260-2) by applying the program version 2.5 (Nichols, K.B., Analysis. B.1997, Analysis. K.1997, N.K.1997 and N.K.1997, Analysis. and N.K.1997, N.1997, N.K..

In a preferred embodiment, the amino acid sequence of the polypeptide and the amino acid sequence of SEQ ID NO: 1 is introduced into the amino acid sequence, preferably by way of substitution, at least one and often a plurality, e.g. 1 to 15, amino acid residues comprising a non-polypeptide moiety linking group. Thus, the polypeptide moiety alters the amount of specific amino acid residues that bind to the selected non-polypeptide moiety, thereby resulting in more efficient, specific and/or broader conjugation. For example, when the total number of amino acid residues comprising the attachment group of the selected non-polypeptide is altered to a preferred level, clearance of the conjugate is typically significantly reduced due to changes in the shape, size and/or charge of the molecule resulting from the conjugation. Further, as the total number of amino acid residues comprising the linking group of the selected non-polypeptide increases, a greater proportion of the polypeptide molecules are masked by the selected non-polypeptide components, resulting in a lower immune response.

The term "a difference" as used in this application is meant to allow for the existence of additional differences. Therefore, in addition to the specific amino acid difference, other amino acid residues can be mutated.

In another preferred embodiment, the amino acid sequence of the polypeptide and SEQ ID NO: 1 is at least one, preferably a plurality, e.g. 1-15, amino acid residues comprising a non-polypeptide moiety linking group are removed from the amino acid sequence, preferably substituted. By removing one or more amino acid residues comprising a non-polypeptide moiety attachment group, conjugation of the non-polypeptide moiety to a site on the polypeptide that results in adverse conjugation, for example to an amino acid residue at or near the functional site of the polypeptide, may be avoided (as conjugation at such site may be affected due to receptor recognition resulting in inactivation of the resulting conjugate or reduced G-CSF activity). The term "functional site" as used herein refers to one or more amino acid residues essential or related to the function or effect of hG-CSF. Such amino acid residues constitute a part of the functional site. Functional sites can be determined by methods known in the art, preferably by analysing the structure of a complex of a polypeptide and a related receptor, such as the hG-CSF receptor (see Aritomi et al, nature 401: 713-717, 1999).

The conjugates of the invention generally comprise a sufficient amount and type of non-polypeptide moiety to increase the ability of the conjugate to shorten the duration of neutropenia as compared to hG-CSF, such as filgrastim, lenograstim or narograstim, preferably to rhG-CSF comprising a single 20kDa PEG moiety attached at the N-terminus. The increase in the ability to shorten the duration of neutropenia can be determined as described in the materials and methods section herein ("measuring the in vivo biological activity of conjugated and unconjugated hG-CSF and variants thereof in rats with chemotherapy-induced neutropenia").

The conjugate of the invention may comprise at least one unconjugated, couplable linking group of the non-polypeptide moiety. The term "conjugatible linking group" as used in the context of the present invention refers to a linking group which is located at a site in a polypeptide where conjugation is possible and which, when conjugated, can be conjugated to the relevant non-polypeptide moiety in the absence of special precautions. For example, such a linker group may be a portion of an amino acid residue that is necessary or relevant for the polypeptide to exhibit its activity. One convenient way to avoid coupling with other couplable linking groups is to mask the linking group with an auxiliary molecule, for example in the manner described under the heading "block functional site" section. It is understood that the number of not yet coupled but couplable attachment groups is determined by the particular G-SCF polypeptide and the position of the couplable attachment groups. For example, the polypeptide conjugate comprises one or two unconjugated but couplable linking groups and at least one, and preferably two or more, conjugated linking groups.

The 4 helices of G-CSF include amino acid residues 11-41 (helix A), 71-95 (helix B), 102-125 (helix C), and 145-170 (helix D) (Zink et al (1994) Biochemistry 33: 8453-8463). It has surprisingly been found that although it is generally believed that modification of a protein helix (e.g.the helical structure of a 4-helix bundle protein such as G-CSF) is at risk of interfering with the function of the protein, advantageous results may be obtained when a non-polypeptide moiety is attached to amino acid residues located in one or more helical regions of G-CSF. In one embodiment, the polypeptide conjugate of the invention thus comprises at least one non-polypeptide component, and the component is linked to an amino acid residue located in one of the helices (in particular B, C or one or more of the D helices).

The conjugate of the present invention, wherein the non-polypeptide moiety is linked to lysine

In general, the polypeptide conjugate may comprise

i) A polypeptide exhibiting G-CSF activity, the polypeptide comprising an amino acid sequence that differs from SEQ ID NO: 1 is different in that it has at least one substitution selected from the following substitutions: t1, P2, L3, G4, P5, A6, S7, S8, L9, PICK, Q11, S12, F13, L14, L15, E19, Q20, V21, Q25, G26, D27, a29, a30, E33, a37, T38, Y39, L41, H43, P44, E45, E46, V48, L49, L50, H52, S53, L54, 156K, P57, P60, L61, S62, S63, P65, S66, Q67, a68, L69, Q70, L71, a72, G73, S76, Q77, L78, S80, F83, Q86, G87, Q90, E93, G94, S96, P97, E98, L99, G100, P101, T102, D104, T105, Q108, Q107, L119, Q127, G87, Q121, Q123, Q170, Q123, S96, P97, Q123, S135, Q123, P170, a 123, Q123, a 123, P123, S123, P170, S123, S

ii) at least one non-polypeptide moiety linked to a lysine residue of the polypeptide, and may also have an N-terminal amino acid residue.

The hG-CSF comprises four lysine residues, of which K16 is located in the receptor binding domain and the others are located at positions 23, 34 and 40, respectively, which are relatively close to the receptor binding domain. To avoid coupling reactions of one or more of these lysine residues (as such coupling may result in loss or severe loss of activity of the resulting conjugate) it may be necessary to remove at least one of the lysine residues, for example two, three or all of these residues. Therefore, in another more preferred aspect the invention relates to a polypeptide conjugate as defined above, wherein at least one amino acid residue selected from the group consisting of K16, K23, K34 and K40 is deleted or substituted with another amino acid residue. Preferably at least K16 is substituted with another amino acid residue.

Examples of preferred amino acid substitutions include one or more of Q70K, Q90K, T105K, Q120K, T133K, S159K and H170K/Q/R, for example two, three or four or five of these substitutions, for example: q70 + Q90, Q70 + T105, Q70 + Q120, Q70 + T133, Q70 + S159, Q70 + H170, Q90+ T105, Q90+ Q120, Q90+ T133, Q90+ S159, Q90+ H170, T105+ Q120, T105+ T133, T105+ S159, T105+ H170, Q120 + T133, Q120 + S159, Q120 + H170, T133 + S159, T133 + H170, S159 + H170, Q70 + Q90+ T105, Q70 + Q90+ Q120, Q70 + Q90+ T133, Q70 + Q90+ T159, Q70 + Q90+ S159, Q70 + T105, Q70 + T105+ T133, Q70 + T105+ S159, Q70 + T105+ H170, Q70 + T120, Q70 + T159 + T120 + T159, Q70 + T120 + T159, Q70 + T133 + T105+ T159, Q70 + T120 + T133 + T159, Q70 + T120 + T159 + T105+ T, Q70 + T159 + T105+ H170, Q70 + T105, Q70 + T120 + T159, Q70 + T159 + T133 + T159, Q70 + T120, Q70 + T159 + T120, Q70 + T159, Q70 + T120, Q70 + T133 + T120, Q70 + T159, Q105 + T159 + T133 + T, Q105 + T159, Q105 + T159, Q70 + T159, Q105 + T133 + T159, Q105 + T, Q70, t105+ S159 + H170, Q120 + T133 + S159, Q120 + T133 + H170, Q120 + S159 + H170, T133 + S159 + H170, Q70 + Q90+ T105+ Q120, Q70 + Q90+ T105+ T133, Q70 + Q90+ T105+ S159, Q70 + Q90+ T105+ H170, Q70 + Q90+ Q120 + T133, Q70 + Q90+ Q120 + S159, Q70 + Q90+ Q120 + H170, Q70 + Q90+ T133 + S159, Q70 + Q90+ T133 + H170, Q70 + Q90+ S159 + H170, Q70 + Q159 + Q105 + S159 + H170, Q70 + T105+ T120 + T133, Q70 + T105+ T159, Q70 + Q120 + H170, Q70 + T105+ T159 + H159, Q70 + T105+ T159 + T133 + T159 + Q105 + T120 + T, Q70 + T159 + T120 + T159, Q70 + T120 + T159 + T133 + Q105 + T, Q70 + T159 + T105+ T159 + T120 + T159, Q70 + T159, Q70 + T105+ T159, Q70 + T159 + T105+ T159, Q70 + T159, Q70 + T105+ T159, Q70 + T159 + T120 + T159, Q70 + T159 + T159, Q70 + T159 + T105, T105K + Q120K + T133K + S159K, T105K + Q120K + T133K + H170K, T105K + Q120K + S159K + H170K, T105K + T133K + S159K + H170K, or Q120K + T133K + S159K + H170K, wherein the substituted H170K may be replaced with H170Q or H170R.

The polypeptide having at least one introduced and one removed lysine preferably comprises at least one, e.g. one, two, three or four substitutions selected from the group consisting of K16R, K16Q, K23R, K23Q, K34R, K34Q, K40R and K40Q, more preferably at least one K16R substitution, in order to avoid coupling of the residues. Preferably, the above polypeptide comprises at least one substitution selected from the group consisting of K16R + K23R, K16R + K34R, K16R + K40R, K23R + K34R, K23R + K40R, K34R + K40R, K16R + K23R + K34R, K16R + K23R + K40R, K23R + K34R + K40R, K16R + K34R + K40R and K16R + K23R + K34R + K40R. In a preferred embodiment, the polypeptide comprises the substitutions K16R + K34R + K40R, but the lysine at position 23 is unchanged. As noted above, each substitution or combination of substitutions listed in this paragraph for removing a lysine residue may be suitably used with any other substitution disclosed herein for introducing a lysine residue, and in particular may be used with the substitutions listed in the preceding paragraph.

In a particular embodiment, the polypeptide comprises the substitutions K16R, K34R, K40R, T105K and S159K and is coupled to 2-6 (typically 3-6) polyethylene glycol moieties having a molecular weight of about 1000-10,000 Da.

In one embodiment, the conjugate of the invention is glycosylated at position T133, i.e. this position is unaltered with respect to wild-type hG-CSF. This is a natural glycosylation site. Alternatively, the conjugate may be non-glycosylated, but glycosylated conjugates are preferred.

In particular, the conjugate may be linked to 2-6 (typically 3-6) polyethylene glycol moieties of molecular weight about 5000 and 6000Da, for example to mPEG of molecular weight about 5 kDa. Preferably, the conjugate has attached thereto 4-5 polyethylene glycol components of molecular weight about 5000-6000Da, such as 5kDa mPEG.

In another embodiment, the conjugates prepared may be linked to only a single number of PEG moieties, e.g., 2,3, 4,5, or 6 PEG moieties per polypeptide, or a mixture of polypeptide conjugates linked to different numbers of PEG moieties, e.g., a mixture having 2-5, 2-4, 3-5, 3-4, 4-6, 4-5, or 5-6 linked PEG moieties. As mentioned above, an example of a preferred coupling mixture also has 4-5 PEG components of about 5 kDa.

It will be appreciated that conjugates having a particular number of attached PEG components, or mixtures of conjugates having a range of numbers of attached PEG components, can be obtained by selecting appropriate pegylation conditions, and that conjugates having the desired number of PEG components can also be isolated by using subsequent purification. Examples of methods for isolating G-CSF molecules with different numbers of PEG components are as follows. The amount of PEG component attached can be determined, for example, by SDS-PAGE. For the purposes of the present invention, a polypeptide conjugate is considered to be linked to a given amount of PEG component if, after separation on an SDS-PAGE gel, the polypeptide conjugate exhibits no bands or only faint bands in addition to the bands corresponding to said given amount of PEG component. For example, a sample of a polypeptide conjugate that has been separated on an SDS-PAGE gel showing bands corresponding to 4 and 5 PEG groups, respectively, and bands corresponding to 3 or 6 PEG groups that are weak or not at all, may be considered to have 4-5 PEG groups attached.

As mentioned above, preferably the non-polypeptide component is a polymer molecule, preferably selected from linear or branched polyethylene glycol or another polyalkylene oxide. Preferred polymer molecules are mPEG-SPA (especially SPA-mPEG 5000) from Shearwater corp, or oxycarbonyl-oxy-N-dicarboximide PEG (US5,122,614).

It will be appreciated that any amino acid alteration (especially substitution) specifically mentioned in this section may be combined with any amino acid alteration (preferably substitution) specifically mentioned in other sections of the invention disclosing specific amino acid modifications (especially the introduction and/or removal of glycosylation sites).

Glycosylated conjugates of the invention

In addition to having introduced and removed lysine residues as described above, the conjugates of the invention may also be glycosylated at the native glycosylation site (T133) and/or the introduced glycosylation site(s) of hG-CSF by expression of the polypeptide in a glycosylated host cell such that the resulting conjugate comprises one or more sugar components. The conjugate may thus be a glycosylated polypeptide having G-CSF activity comprising a sequence different from SEQ ID NO: 1, said difference being that at least one non-natural glycosylation site is introduced into the amino acid sequence by at least one substitution selected from the group consisting of: l3 + P5/T, P5, A6, S8 + P1/T, P10, Q11 + F13/T, S12 + L14/T, F13 + L15/T, L14 + K16/T, K16 + L18/T, E19 + V21/T, Q20 + R22/T, V21 + K23/T, R22 + I24/T, K23 + Q25/T, Q25 + D27/T, G26 + G28/T, D27 + A29/T, A29 + L31/T, A30 + Q32/T, E33 + L35/T, A37 + Y39/T, T38 + K40/T, Y39 + L41/T, P44 + E46/T, E45 + L47/T, E46 + V48/T, V48 + L50/T, L49 + G51/T, T51 + L41/T, T65 + T52/T, P53 + T52/T, P55 + T, P53/T, S52/T, S23 + L35/T, T25 + L35, s66 + A68/T, Q67 + L69/T, A68 + Q70/T, L69 + L71/T, Q70 + A72/T, L71 + G73/T, G73 + L75/T, S76 + L78/T, Q77 + H79/T, L78, S80 + L82/T, F83 + Y85/T, Q86 + L88/T, G87 + L89/T, Q90+ L92/T, E93 + I95/T, P97 + L99/T, L99 + P101/T, P101 + L103/T, T102 + D104/T, D104 + L106/T, T105+ Q107/T, Q107 + D109/T, L108 + V110/T, D109 + A111/T, A111 + F113/T, D112 + A114/T, F113/T, T115/T119/T, T120/T, T118 + W118/T, M121 + M11/T, M121, e122 + L124/T, E123 + G125/T, L124 + M126/T, M126 + P128/T, P128 + L130/T, L130 + P132/T, P132 + Q134/T, T133 + G135/T, Q134 + A136/T, A136 + P138/T, P138 + F140/T, A139 + A141/T, A141 + A143/T, S142 + F144/T, A143 + Q145/T, F144 + R146/T, Q145 + R147/T, R146 + A148/T, R147 + G149/T, S155 + L157/T, H156 + Q158/T, S159 + L161/T, L161 + V163/T, E162, V165 + Y165/T, S164 + R166/T, Y165 + V167/T, R166 + L167/T, R167/T168, T168 + T, T169/T, or S170 + R166/T, T168, T170, T, or T169, t residues are preferred.

It will be appreciated that in order to prepare a conjugate according to this aspect, the polypeptide must be expressed in a glycosylation host cell capable of linking the oligosaccharide component to the glycosylation site, or the polypeptide must be glycosylated in vitro. Examples of glycosylated host cells are given in the section entitled "conjugation to oligosaccharide components" below.

Alternatively, the conjugate according to this aspect comprises a polypeptide having G-CSF activity comprising a sequence different from SEQ ID NO: 1 by at least one substitution selected from the group consisting of P5N, A6N, P10N, P60N, L61N, L78N, F113N and E162N, more preferably from the group consisting of PSN, A6N, PION, P60N, L61N, F113N and E162N, for example from the group consisting of P60N, L61N, F113N, E162N.

Alternatively, the conjugate of this aspect of the invention comprises a polypeptide exhibiting G-CSF activity comprising a sequence different from SEQ ID NO: 1, said difference consisting in at least one substitution selected from the group consisting of: D27N + a29S, D27N + a29T, D104N + L106S, D104N + L106T, D109N + a111S, D109N + a111T, D112N + a114S and D112N + a114T, more preferably selected from D27N + a29S, D27N + a29T, D104N + L106S, D104N + L106T, D112N + a114S, and D112N + a114T, such as selected from D27N + a29S, D27N + a29T, D104N + L106S, and D104N + L106T.

Methods of preparing conjugates of the invention

In the sections "conjugation to polymer molecules" and "conjugation to oligosaccharide components" below, conjugation to specific types of non-polypeptide components is described. In general, the polypeptide conjugates of the invention may be prepared as follows: culturing an appropriate host cell under conditions conducive to expression of the polypeptide, and harvesting the polypeptide, followed by in vitro coupling of the polypeptide to a non-polypeptide moiety. In the case of glycosylated polypeptides containing at least one N-or O-glycosylation site, glycosylation is preferably accomplished using a eukaryotic host cell capable of in vivo glycosylation.

Coupling to polymer molecules

The polymer molecule to be coupled to the polypeptide may be any suitable polymer molecule, such as a natural or synthetic homopolymer or heteropolymer, typically a polypeptide is coupled to a molecule having a molecular weight in the range of about 300-100,000Da, such as in the range of about 500-20,000Da, more preferably in the range of about 1000-15,000Da, even more preferably in the range of about 2000-12,000Da, such as in the range of about 3000-10,000. The term "about" as used herein with respect to polymer molecules refers to a broad range of molecular weights and reflects the fact that there will generally be a defined molecular weight distribution within a given polymer article.

Examples of homopolymers include polyols (i.e., poly-OH), polyamines (i.e., poly-NH 2), and polycarboxylic acids (i.e., poly-COOH). Heteropolymers are polymers that contain different coupling groups, such as hydroxyl and amine groups.

Examples of suitable polymer molecules include polymer molecules selected from polyalkylene oxides (PAOs), including polyalkylene glycols (PAGs), such as linear or branched polyethylene glycols (PEG) and polypropylene glycols (PPG), polyvinyl alcohols (PVA), polycarboxylates (poly-carboxylates), polyvinylpyrrolidone, polyethylene-co-maleic anhydride (polyethylene-co-maleic acid anhydride), polystyrene-co-maleic anhydride (polystyrene-co-maleic acid anhydride), dextran, including carboxymethyl dextran or any other biopolymer suitable for reducing immunogenicity and/or increasing functional half-life and/or serum half-life in vivo. Another example of a polymer molecule is human albumin or another high abundance plasma protein. Generally, polymers derived from polyalkylene oxides are biocompatible, non-toxic, non-antigenic, non-immunogenic, have various water-solubilizing properties, and are readily excreted from living organisms.

PEG is a preferred polymer molecule because it has only a few reactive groups capable of cross-linking compared to polypeptides such as dextran. In particular, monofunctional PEGs, such as methoxypolyethylene glycol (mPEG), are of interest because of their relatively simple coupling chemistry (only one reactive group is available for coupling to a linking group on a polypeptide). Thus, the risk of cross-linking is eliminated, the resulting polypeptide conjugate is more homogeneous and the reaction of the polymer molecule with the polypeptide is easier to control.

To achieve covalent attachment of the polymer molecule and the polypeptide, the hydroxyl-terminated groups of the polymer molecule are provided in an activated form, i.e. as functional reactive groups. Suitable activated polymer molecules are commercially available, for example, from Shearwater Polymers, inc., Huntsville, AL, USA, or from polymac Pharmaceuticals plc, uk. Alternatively, the polymer molecules may be activated by convenient methods known in the art, for example as disclosed in WO 90/13540. Specific examples of activated linear or branched polymer molecules useful in the present invention are described in shearwater polymers, Inc.1997 and 2000 ligands (Functionalized Biocompatible polymers for Research and pharmaceuticals, Polyethylene Glycol and Derivatives, incorporated herein by reference). Specific examples of activated PEG polymers include the following linear PEGs: NHS-PEG (e.g., SPA-PEA, SSPA-PEG, SBA-PEG, SS-PEG, SSA-PEG, SC-PEG, SG-PEG, and SCM-PEG), and NOR-PEG), BTC-PEG, EPOX-PEG, NCO-PEG, NPC-PEG, CDI-PEG, ALD-PEG, TRES-PEG, VS-PEG, IODO-PEG, and MAL-PEG, and branched PEGs such as PEG2-NHS and those disclosed in U.S. Pat. No. 5,932,462 and U.S. Pat. No. 5,643,575, which are incorporated herein by reference. Further, useful polymer molecules and/or pegylated compounds are disclosed in the following publications, which are incorporated herein by reference: U.S. Pat. No. 5,824,778, U.S. Pat. No. 5,32607, EP, US, US, U.S. Pat. No. 5,122,614, U.S. Pat. No. 5,16555, WO/, WO/17039, WO/18247, WO/28024, WO/00162, WO/11924, W095/13090, WO/, WO/00080, WO/18832, WO/41562, WO/, WO/32134, WO/32139, WO/32140, WO/40791, WO/32466, WO/06058, EP439508, WO/, WO/21469, WO/13312, EP921131, U.S. Pat. No. WO, WO 809996, U.S. Pat. No. 5,629,384, WO/41813, WO/, U.S. Pat. No.6,5963, U.S. 5,382,657, EP 183356, EP183503, EP 400503, EP154316, EP.

Conjugation of the polypeptide and the activated polymer molecule may be carried out by applying any convenient method, for example as described in the following references (which also describe suitable methods of activation of the polymer molecule): taylor, (1991), "Protein immunological and applications", Marcel Dekker, n.y.; S.S.Wong, (1992), "Chemistry of protein conjugation and Crosslinking", CRC Press, Boca Raton; hermanson et al, (1993), "Immobilized Affinity Ligand Techniques", Academic Press, N.Y.). The skilled person will be aware of the activation method and/or coupling chemistry to be applied, depending on the linking group of the polypeptide (examples are given above) and the functional group of the polymer (e.g. amine, hydroxyl, carboxyl, aldehyde, sulfydryl, succinimide (succinimide), maleimide, vinylsulfone (vinylsulfone) or haloacetate (haloacetate) — pegylation may be a coupling directed to all available linking groups on the polypeptide (i.e. those which are exposed at the surface of the polypeptide) or may be a coupling directed to one or more specific linking groups, e.g. the N-terminal amino group (US5,985,265) — further, the coupling may be done in a further or stepwise manner (e.g. as described in WO 99/55377).

It will be appreciated that optimal molecules can be prepared by engineering pegylation such that the number, size and form (e.g., linear or branched) of attached PEG molecules, as well as their attachment location within the polypeptide, are optimal. The molecular weight of the polymer used can be selected according to the desired effect. For example, where the primary purpose of conjugation is to achieve a high molecular weight and larger volume of conjugate (e.g., to reduce renal clearance), it may be conjugated to one or a few high molecular weight polymer molecules, or to many smaller molecular weight polymer molecules, to achieve the desired effect. Preferably, however, several relatively low molecular weight polymer molecules are used. This is also true when a greater degree of masking of the epitope is desired. In this case, for example, 2 to 8 polymers, for example 3 to 6 such polymers, with a molecular weight of about 5,000Da can be used. As exemplified below, the use of more polymer molecules having a lower molecular weight (e.g., 4-6 molecules having a molecular weight of 5000) may be more advantageous in improving the in vivo functional half-life of the polypeptide conjugate than the use of less polymer molecules having a higher molecular weight (e.g., 1-3 molecules having a molecular weight of 12,000-20,000), even when the total molecular weight of the attached polymer molecules is the same in both cases, or the like. It is believed that the presence of more polymer molecules of smaller molecular weight may provide a larger radius or apparent shape to the polypeptide than, for example, a single larger polymer molecule, at least when the polymer molecules are relatively uniformly distributed on the surface of the polypeptide.

It has also been found that more favorable results are obtained when at least one major portion of the conjugate of the invention has an apparent size (also referred to as "apparent molecular weight" or "apparent mass") of at least about 50kDa, preferably at least about 55kDa, more preferably at least about 60kDa, for example at least about 66 kDa. We believe this can be attributed to the fact that: for conjugates with a substantial apparent size, renal clearance is essentially eliminated. In this context, the "apparent size" of a G-CSF conjugate or polypeptide is determined by the SDS-PAGE method described in the examples section below.

The use of the term "major part" is related to the fact that the polypeptide conjugate of the invention usually contains a plurality of individual conjugates and that they are associated with different numbers of non-polypeptide components. For example, a given polypeptide may be pegylated under a given set of pegylation conditions to provide a composition in which a majority of individual polypeptide conjugates are linked to, for example, 3-5 PEG groups and a majority of conjugates are linked to 4 PEG groups. It is clear that the apparent molecular weights of these conjugate molecules are different from each other. In this example, if we assume that the G-CSF polypeptide is conjugated to a PEG group with a molecular weight of 5kDa, then those conjugates conjugated to only 3 PEG groups show bands with an apparent molecular weight of less than about 50kDa on SDS-PAGE gels, while those conjugates conjugated to 4 or 5 PEG groups give bands with a significantly larger apparent molecular weight, most likely all of them are greater than about 50 kDa. Thus, in this example, 3 major bands would appear on the SDS-PAGE gel, corresponding to conjugates conjugated with 3, 4, or 5 PEG groups, respectively. Thus, the term "major portion" in the context of the present specification and claims means that at least one of the above-mentioned major bands on the SDS-PAGE gel corresponds to the lowest apparent molecular weight indicated.

Preferably, at least 50% of the individual conjugate molecules have the lowest apparent size as described above. More preferably, at least 60% of the conjugate molecules have the lowest apparent size, still more preferably at least 70%, 75%, 80% or 85%. Most preferably, at least 90% of the conjugate molecules have the lowest apparent size as described above, i.e. at least 50kDa and preferably more, e.g. at least 55kDa or 60 kDa.

It will be appreciated that the apparent size of the conjugate or polypeptide, expressed in kDa, need not correspond to the actual molecular weight of the conjugate or polypeptide. This apparent size is more likely to reflect both true molecular weight and overall mass. Since in most cases the attachment of one or more PEG groups or other non-polypeptide moieties can result in a substantial increase in the mass of the polypeptide to which it is attached, the apparent size of the polypeptide conjugates of the invention typically exceeds the true molecular weight of the conjugate. Thus, in view of renal clearance, the conjugates of the invention can readily exhibit the characteristics of polypeptides with a molecular weight greater than, for example, 66kDa (corresponding to the apparent size), but with a practical molecular weight less than 66 kDa. We believe that this effect on the apparent size is responsible for the following phenomena: the attachment of, for example, 4 PEG groups of 5kDa results in an excess of polypeptides attached to only one PEG group of 20 kDa.

Although it is not preferred to attach only a single polymer molecule to a single attachment group on the protein, in the event that only one polymer molecule is attached, it is generally advantageous if the polymer molecule (linear or branched) has a relatively high molecular weight, for example of about 20 kDa.

In another preferred embodiment, the conjugates of the invention have 1) at least one major moiety with an apparent molecular weight of at least about 50kDa and 2) reduced in vitro biological activity (reduced receptor binding affinity) as compared to hG-CSF as described above. We have found that such conjugates result in lower levels of renal clearance due to larger apparent size and lower levels of receptor mediated clearance due to lower biological activity (lower receptor binding affinity) in vitro. The overall results are very excellent in the following respects: effectively stimulate the neutrophil granulocytes, remarkably increase the half-life in vivo and thus play a role for a long time so as to bring important clinical benefits.

Typically, the coupling of the polymers is carried out under conditions such that as many polymer attachment groups as are available react with the polymer molecules. This can be achieved by means of a suitable molar excess of polymer relative to the polypeptide (number of attachment sites). Typical molar ratios of activated polymer molecules to polypeptide attachment sites are up to about 1000-1, such as up to about 200-1 or up to about 100-1. In some cases, however, for example, where a lower degree of polymer attachment is desired, the above ratio may be reduced, for example, up to about 50-1, 10-1 or 5-1.

The present invention also contemplates the conjugation of polymer molecules to polypeptides via a linker. Suitable linkers are well known to the skilled person. A preferred example is cyanuric chloride (Abuchowski et al, (1977), J.biol. chem.252, 3578-.

After coupling, the remaining activated polymer molecules are blocked by methods known in the art, such as by adding a primary amine to the reaction mixture, and the resulting deactivated polymer molecules are removed by suitable methods (see materials and methods).

In a preferred embodiment, the polypeptide conjugate of the invention comprises PEG molecules, in particular linear or branched PEG molecules, e.g. PEG molecules having a molecular weight of about 1-15kDa, typically about 2-12kDa, e.g. about 3-10kDa, e.g. about 5 or 6kDa, attached to some, most or preferably substantially all lysine residues of the polypeptide that are available for pegylation.

It will be appreciated that depending on the circumstances, such as the amino acid sequence of the polypeptide, the nature of the activated PEG compound used and the particular pegylation conditions, including the molar ratio of PEG to polypeptide, different degrees of pegylation may be obtained, with higher degrees of pegylation generally being obtainable by higher ratios of PEG to polypeptide. However, the pegylated polypeptides produced by any given pegylation procedure will generally comprise a random distribution of polypeptide conjugates with slightly different degrees of pegylation. If desired, such mixtures of polypeptides linked to different amounts of PEG components can be purified using the methods described in the examples below, to obtain products with a more uniform degree of PEGylation.

In another embodiment, the polypeptide conjugates of the invention comprise a PEG molecule attached to a lysine residue that can be used to pegylate the polypeptide, and a PEG molecule additionally attached to the N-terminal amino acid residue of the polypeptide.

Coupling with oligosaccharide Components

The coupling to the oligosaccharide component may be carried out in vivo or in vitro, preferably in vivo. To accomplish the in vivo glycosylation of a G-CSF molecule comprising one or more glycosylation sites, the nucleotide sequence encoding the polypeptide must be inserted into a eukaryotic expression host that can be glycosylated. The expression host cell may be selected from a fungus (filamentous fungus or yeast), an insect, an animal cell, or from a transgenic plant cell. In one embodiment, the host cell is a mammalian cell, such as a CHO cell, BHK cell or HEK cell, such as HEK293, or an insect cell, such as SF9 cell, or a yeast cell, such as Saccharomyces Cerevisiae (Saccharomyces Cerevisiae), Pichia pastoris (Pichia pastoris), or any of the host cells described below. In vitro covalent coupling of amino acid residues of polypeptides using sugars (dextrans) can also be performed, for example, according to the methods described in WO87/05330 and Aplin et al, CRC Crit Rev. biochem., pp.259-306, 1981.

Coupling of oligosaccharide components or PEG to protein-and peptide-bound Gln residues in vitro can be achieved by transglutaminase (TG enzymes). Glutamyltransferase catalyzes the transfer of an amine group of a donor to a Gln residue of a protein-bound type and a peptide-bound type in a so-called cross-linking reaction. The amine groups of the donor may be protein-bound or peptide-bound, such as, for example, the epsilon amine groups in lysine residues, or may be part of a small or large organic molecule. An example of a small organic molecule that can act as an amine group donor in transglutaminase-catalyzed cross-linking is putrescine (1, 4-butanediamine). An example of a larger organic molecule that acts as an amine donor in transglutaminase-catalyzed crosslinking is PEG that contains amine groups (Sato et al, Biochemistry35, 13072-13080).

Generally, transglutaminase is a highly specific enzyme, and not every Gln residue exposed on the surface of a protein can be cross-linked to a substrate comprising amine groups catalyzed by transglutaminase. In contrast, only a few Gln residues may naturally function as a substrate for transglutaminase, but the exact criteria for determining which Gln-residues are good substrates for transglutaminase are not known. Therefore, in order to facilitate the protein to undergo the transglutaminase-catalyzed crosslinking reaction, it is usually a prerequisite to add an amino acid sequence known to function well as a substrate for transglutaminase at a convenient position. Several amino acid sequences are known per se to be or contain excellent natural transglutaminase substrates such as substance P, fast-reacting elastase inhibitory peptides, fibrinogen, fibronectin, α 2 plasmin inhibitors, α casein and β casein.

Functional site blocking

It has been reported that excessive coupling of polymers can result in loss of activity of the polypeptide coupled to the polymer. The method of eliminating this problem may be, for example, removing a linking group located at the functional site, or blocking the functional site before coupling, so that the functional site is in a blocked state at the time of coupling. The latter strategy constitutes a further embodiment of the invention (the first strategy is exemplified above, e.g.by removal of lysine residues adjacent to functional sites). More specifically, according to a second strategy, the coupling between the polypeptide and the non-polypeptide component is carried out under conditions in which the functional site of the polypeptide is blocked by a helper molecule capable of binding to the functional site of the polypeptide.

Preferably, the accessory molecule is capable of specifically recognizing a functional site of the polypeptide, such as a receptor, in particular a G-CSF receptor or a part of a G-CSF receptor. Alternatively, the accessory molecule may be an antibody, in particular a monoclonal antibody recognizing a polypeptide exhibiting G-CSF activity. In particular, the helper molecule may be a neutralizing monoclonal antibody.

The polypeptide may be interacted with an accessory molecule prior to conjugation. This ensures that the functional site of the polypeptide is masked or protected and therefore cannot be derivatized by non-polypeptide components such as polymers. After elution from the helper molecule, the coupling between the non-polypeptide component and the polypeptide can be restored with at least partially preserved functional sites.

Subsequent coupling of the polypeptide having a blocked functional site to a polymer, oligosaccharide component or any other compound is carried out in a conventional manner, as described above.

Regardless of the nature of the helper molecule used to mask the functional site of the polypeptide in the conjugate, it is desirable that the helper molecule contains no or only a small number of groups that bind to the designated non-polypeptide components of the molecule, wherein conjugation to these groups will prevent desorption of the conjugated polypeptide from the helper molecule. Thus, selective coupling to the attachment groups present in the non-masked portion of the polypeptide is possible and the helper molecule can be used repeatedly for repeated couplings. For example, if the non-polypeptide moiety is a polymer molecule such as PEG having lysine epsilon amino groups or the N-terminal amino acid residue as a linking group, it is desirable that the helper molecule be substantially free of couplable epsilon amino groups, preferably free of any epsilon amino groups. Thus, in a preferred embodiment, the helper molecule is a protein or peptide capable of binding to a functional site of the polypeptide, which protein or peptide does not contain any linking groups that are couplable to the specified non-polypeptide component.

In embodiments of the invention where the polypeptide conjugate is prepared from a plurality of nucleotide sequences encoding a polypeptide of interest, the blocking of the functional group is effected in a microtiter plate prior to conjugation, e.g., by adding the expressed polypeptide variant to a microtiter plate comprising an immobilized blocking group, e.g., a receptor, antibody or the like.

In another embodiment, the helper molecule is first covalently attached to a solid phase, such as a column packing material, e.g., sephadex or agarose beads, or a surface, such as a reaction vessel. Subsequently, the polypeptides are loaded onto the cartridge material carrying the helper molecules and then coupled according to methods well known in the art, for example as described above. This method allows separation of the polypeptide conjugate from the helper molecule by elution. The polypeptide conjugate is eluted by conventional techniques under physicochemical conditions that do not result in substantial degradation of the polypeptide conjugate. The liquid phase comprising the polypeptide conjugate is separated from the solid phase, but the helper molecule is still covalently linked to the polypeptide conjugate. The separation can be achieved by other methods: for example, the helper molecule may be derivatized with a second molecule (e.g., biotin) that can be recognized by a specific binding substance (e.g., streptavidin). The specific conjugate may be attached to a solid phase such that the polypeptide conjugate is separated from the helper molecule-second molecule complex when passed through a second helper-solid phase column, and upon subsequent elution, the helper molecule-second molecule complex remains on the solid phase column and the polypeptide conjugate is eluted from the column. The polypeptide conjugate may be released from the helper molecule in any suitable manner. Protection can be released by providing conditions under which the helper molecule is detached from the functional site of the G-CSF to which it is bound. For example, the complex formed between the polymer-conjugated antibody and the anti-idiotype antibody can be dissociated by adjusting the pH to an acidic or basic pH level.

Conjugation of labeled polypeptides

In another embodiment, the polypeptide is expressed as a fusion protein with a marker (i.e., a stretch of amino acid or peptide fragment generally consisting of 1-30, e.g., 1-20, amino acid residues). In addition to enabling rapid and easy purification, the label is a convenient means for achieving conjugation between the labeled polypeptide and non-polypeptide components. In particular, the label may be used to accomplish conjugation in a microtiter plate or to immobilize the labeled polypeptide to another carrier, such as a paramagnetic bead, via the label. Coupling with the labeled polypeptide on the microtiter plate has the advantage that the labeled polypeptide can be immobilized directly from the culture broth on the microtiter plate (in principle without any purification) and coupled. Thus, the total number of manipulation steps (from expression to conjugation) can be reduced. Furthermore, the label may act as a spacer molecule to ensure easier conjugation of the immobilized polypeptide. The conjugation with the marker polypeptide may be to any non-polypeptide moiety disclosed herein, such as to a polymer molecule such as PEG.

The characteristics of the particular marker to be used are not critical, as long as the marker can be expressed with the polypeptide and can be immobilized on a suitable surface or carrier material. Many suitable labels are commercially available, for example from Unizyme manufacturers, Denmark. For example, the marker may be any of the following sequences:

Met-Lys-His-Gln-His-Gln-His-Gln-His-Gln-His-Gln-His-Gln-Gln(SEQ ID NO：5)

His-His-His-His-His-His(SEQ ID NO：9)

Met-Lys-His-His-His-His-His-His(SEQ ID NO：10)

Met-Lys-His-His-Ala-His-His-Gln-His-His(SEQ ID NO：11)

Met-Lys-His-Gln-His-Gln-His-Gln-His-Gln-His-Gln-His-Gln(SEQ ID NO：12)

or any of the following:

EQKLISEEDL (SEQ ID NO: 13) (C-terminal marker described in mol.cell.biol.C 5: 3610-16, 1985)

DYKDDDDK (SEQ ID NO: 14) (C-terminal or N-terminal marker)

YPYDVPDYA(SEQ ID NO：15)

Antibodies against the above markers are also commercially available, for example from ADI, Aves Laband Research Diagnostics.

Convenient methods for PEGylation using labeled polypeptides are described in the materials and methods section below. The tag may then be cleaved from the polypeptide using commercially available enzymes.

Method for preparing a polypeptide of the invention or a polypeptide component of a conjugate of the invention

The polypeptide portion of the polypeptide or conjugate of the invention, optionally in glycosylated form, is prepared by any suitable method well known in the art. The methods include constructing a nucleotide sequence encoding a polypeptide and expressing in a suitable transformed or transfected host. However, the polypeptides of the invention may also be prepared by chemical synthesis (although this is less efficient), or by a combination of chemical synthesis and recombinant DNA techniques.

The nucleotide sequence encoding the polypeptide portion of the polypeptide or conjugate of the invention may be isolated or synthetically encoded parent G-CSF, e.g. having the amino acid sequence of SEQ ID NO: 1, and then altering the nucleotide sequence to introduce (i.e., insert or substitute) or delete (i.e., remove or substitute) the relevant amino acid residue.

The nucleotide sequence may conveniently be modified by site-directed mutagenesis in accordance with conventional procedures. Alternatively, the nucleotide sequence may be prepared by chemical synthesis, for example by using an oligonucleotide synthesizer, wherein the oligonucleotide is designed on the basis of the amino acid sequence of the polypeptide of interest, and preferably those codons which facilitate expression in a host cell in which the recombinant polypeptide is prepared. For example, several small oligonucleotides encoding portions of a polypeptide of interest may be synthesized and then assembled by PCR, ligation or Ligase Chain Reaction (LCR) (Barany, PNAS 88: 189. sup. 193, 1991). Typically, each oligonucleotide comprises a5 'or 3' overhang for complementary assembly.

Additional nucleotide sequence modification methods may be used to prepare high-throughput screened polypeptide variants, for example, methods involving homologous crossing as disclosed in US5,093,257, and methods involving gene shuffling (shuffling), i.e., recombination of two or more homologous nucleotide sequences results in a new nucleotide sequence having multiple nucleotide changes compared to the original nucleotide sequence. Gene shuffling (also known as DNA shuffling) involves the cycling of one or more random fragments and the assembly of nucleotide sequences, followed by screening to select nucleotide sequences encoding polypeptides with desired properties. For homology-based nucleic acid shuffling, the nucleotide sequence-related portions are preferably at least 50% identical, such as at least 60% identical, more preferably at least 70% identical, such as at least 80% identical. Recombination can be performed in vitro or in vivo.

Examples of suitable in vitro gene shuffling methods are described by Stemmer et al, (1994), proc.natl.acad.sci.usa; vol.91, pp.10747-10751; stemmer (1994), nature vol.370, pp.389-391; smith (1994), Nature vol.370, pp.324-325; zhao et al, natureBiotechnology, 1998, Mar; 16(3): 258 to 61; zhao h.and Arnold, FB, 1997, vol.25.no.6 pp.1307-1308; shao et al, Nucleic Acids Research 1998, Jan 15; 26(2): pp.681-83; WO 95/17413. An example of a suitable in vivo shuffling method is disclosed in WO 97/07205. Other techniques for mutating nucleic acid sequences by in vitro or in vivo recombination have been disclosed, for example, by WO97/20078 and US5,837,458. Examples of specific shuffling techniques include "family shuffling", "synthetic shuffling", and "in silico (in silico) shuffling". Family shuffling involves subjecting families of homologous genes from different species to one or more shuffling cycles, followed by screening or selection. Family shuffling techniques have been described by, for example, Crameri et al, (1998), nature vol.391, pp.288-291; christians et al, (1999), nature Biotechnology vol.17, pp.259-264; chang et al (1999), nature Biotechnology vol.17, pp.793-797; and Ness et al, (1999), natureBiotechnology Vol.17, 893-896. Synthetic shuffling involves providing a library of overlapping synthetic oligonucleotides based, for example, on an alignment of the target homologous gene sequences. The synthesized oligonucleotides are recombined and the resulting recombined nucleic acid sequences are then screened for, if necessary, and used in further shuffling cycles. Synthetic shuffling techniques are disclosed in WO 00/42561. In silico shuffling involves a DNA shuffling process which is performed or simulated using a computer system, so manual manipulation of nucleic acids is partially or completely avoided. WO00/42560 discloses a technique of computer simulation shuffling.

Once assembled (by synthesis, site-directed mutagenesis or other means), the nucleotide sequence encoding the polypeptide is inserted into a recombinant vector and operably linked to regulatory sequences required for G-CSF expression in the host cell desired to be transformed.

It will be appreciated that not all vectors and expression control sequences will allow for the equally intact expression of a nucleotide sequence encoding a polypeptide as described herein. All hosts do not perform as well even with the same expression system. However, one technique in the art can select these vectors, expression control sequences and hosts without undue experimentation. For example, in selecting a vector, consideration must be given to the host because the vector must be replicated in the host or be capable of integrating into the chromosome. The copy number of the vector, its ability to regulate the copy number, and the expression of any other protein encoded by the vector, such as antibiotic markers, should also be considered. In selecting the expression control sequence, a number of factors should also be considered. These factors include, for example, the relative length of the sequences, the controllability of the sequences, and the compatibility of the control sequences with the nucleotide sequences encoding the polypeptides, especially with regard to the possible secondary structure of the sequences. The host should be selected with consideration of compatibility with the chosen vector, toxicity of the product encoded by the nucleotide sequence, secretory properties of the host, the ability of the host to fold the polypeptide correctly, fermentation or culture conditions of the host, and the ease of purification of the product encoded by the nucleotide sequence.

A recombinant vector may be an autonomously replicating vector, i.e., a vector which may exist as an extrachromosomal entity, the replication of which is independent of chromosomal replication, e.g., a plasmid. Alternatively, the vector may be one which, when introduced into a host cell, is integrated into the host cell genome and replicated together with the chromosome(s) into which it has been integrated.

The vector is preferably an expression vector in which the nucleotide sequence encoding the polypeptide of the present invention is operably linked to additional segments required for transcription of the nucleotide sequence. Typically, the vector is derived from plasmid or viral DNA. Many suitable expression vectors for expression in host cells as mentioned in the present invention are commercially available or described in the literature. Expression vectors for eukaryotic hosts include, for example, vectors containing expression control sequences from SV40, bovine papilloma virus, adenovirus and cytomegalovirus. Specific vectors are, for example, pCDNA3.1(+) \ Hyg (Invitrogen, Carlsbad, Calif., USA) and pCI-neo (Stratagene, La Jolla, Calif., USA). Expression vectors for yeast cells include the 2. mu. plasmid and derivatives thereof, POT1 (U.S. Pat. No.4,931,373), the pJS037 vector described in Okkels, Ann.New York Acad.Sci.782, 202-207, 1996, and pPICZ A, B or C (Invitrogen). Vectors for use In insect Cells include pVL941, pBG311(Cat et al, "Isolation of the Bovine And Human Genes for Mullerian inhibition of the Substance And Expression of the Human Gene In Animal Cells", cell 45, pp.685-98(1986), pBluebc 4.5 And pMelbac (both available from Invitrogen). Expression vectors for use In bacterial hosts include known bacterial plasmids, such as plasmids from E.coli, including pBR322, pET3a And pET12a (both from Novagen Inc., WI, USA), a broad host range plasmid, such as RP4, phage DNA, such as many derivatives of lambda phage, such as NM989, And other DNA phages, such as M13 And filamentous single stranded phage DNA.

Other vectors for use in the present invention include those which allow amplification of the nucleotide sequence encoding the polypeptide in copy number. Such amplifiable vectors are well known in the art. They include, For example, vectors capable Of amplification by DHFR (see, e.g., Kaufman, U.S. Pat. No.4,470,461, Kaufman and Sharp, "conformation Of A Module draft reaction cDNAgene: Analysis Of Signals Utilized For efficiency Expression", mol.cell. biol., 2, pp.1304-19(1982)), and glutamine synthetase amplification ("GS") (see, e.g., U.S. Pat. No. 5,122,464 and EP338,841).

The recombinant vector may further comprise a DNA sequence enabling the vector to replicate in the host cell in question. An example of such a sequence (when the host cell is a mammalian cell) is the origin of replication of SV 40. When the host cell is a yeast cell, suitable sequences that enable the vector to replicate are the yeast plasmid 2. mu. replication gene REP1-3 and an origin of replication.

The vector may also contain a selectable marker, such as a Gene the product of which complements a defect in the host cell, such as the Gene encoding dihydrofolate reductase (DHFR) or Schizosaccharomyces pombe TPI (P.R. Russell, Gene 40, 1985, pp.125-130), or a Gene which confers resistance to a drug such as ampicillin, kanamycin, tetracycline, chloramphenicol, neomycin, hygromycin or methotrexate. For filamentous fungi, selectable markers include amdS, pyrG, arcB, niaD, sC.

The term "control sequences" as used herein includes all components which are necessary or advantageous for the expression of the polypeptides of the invention. Each control sequence may be native or foreign to the nucleotide sequence encoding the polypeptide. These control sequences include, but are not limited to, a leader, polyadenylation sequence, propeptide sequence, promoter, enhancer or upstream activation sequence, signal peptide sequence, and transcription terminator. The control sequences include at least a promoter.

A wide variety of expression control sequences can be used in the present invention. These expression control sequences that may be used include expression control sequences linked to the structural genes of the aforementioned expression vectors as well as any sequences and various combinations thereof known to control gene expression of prokaryotic or eukaryotic cells or viruses thereof.

Examples of suitable control sequences to direct transcription in mammalian cells include SV40 and the early or late promoters of adenovirus, such as the major late promoter of adenovirus 2, the MT-1 (metallothionein gene) promoter, the human cytomegalovirus immediate early gene promoter (CMV), the human elongation factor 1 α (EF-1 α) promoter, the Drosophila minimal heat shock protein 70 promoter, the Rous Sarcoma Virus (RSV) promoter, the human ubiquitin C (UbC) promoter, the human growth hormone terminator, the polyadenylation signal of SV40 or adenovirus E1b region, and the Kozak consensus sequence (Kozak, M.J Mol Biol 1987 Aug 20; 196), (4): 947-50).

To improve expression in mammalian cells, synthetic introns may be inserted into the 5' untranslated region of the nucleotide sequence encoding the polypeptide. An example of a synthetic intron is the synthetic intron from plasmid pCI-Neo (available from Promega Corporation, Wis., USA).

Examples of suitable control sequences that direct transcription in insect cells include the polyhedrin promoter, the P10 promoter, the Autographa californica polyhedrosis virus basic protein promoter, the baculovirus immediate early gene 1 promoter and the baculovirus 39K delayed early gene promoter, and the SV40 polyadenylation sequence. Examples of suitable control sequences for use in a yeast host cell include the promoter of the yeast α mating system, the yeast Triose Phosphate Isomerase (TPI) promoter, the promoter from the yeast glycolytic or ethanol dehydrogenase gene, the ADH2-4c promoter, and the inducible GAL promoter. Examples of suitable control sequences for use in the filamentous fungal host cell include the ADH3 promoter and terminator, promoters derived from genes encoding the following enzymes: aspergillus oryzae TAKA amylase triose phosphate isomerase or alkaline protease, aspergillus niger alpha amylase, aspergillus niger or aspergillus nidulans (a. nidulans) glucoamylase, aspergillus nidulans acetamidase, Rhizomucor miehei (Rhizomucor miehei) aspartic protease or lipase, TPI1 terminator and ADH3 terminator. Examples of suitable control sequences for use in bacterial host cells include the promoters of the lac system, the trp system, the TAC or TRC system and the major promoter region of bacteriophage lambda.

The presence or absence of a signal peptide depends, for example, on the expression host cell used to produce the polypeptide (intracellular or extracellular) to be expressed, and whether secretion is desired. For use in filamentous fungi, the signal peptide may conveniently be derived from the gene encoding amylase or glucoamylase of an Aspergillus strain, the gene encoding Rhizomucor miehei lipase or protease or Humicola lanuginosa lipase. The signal peptide is preferably derived from a gene encoding Aspergillus oryzae TAKA amylase, Aspergillus niger neutral alpha-amylase, Aspergillus niger acid stable amylase, or Aspergillus niger glucoamylase. For use in insect cells, the signal peptide may conveniently be derived from insect genes (see WO90/05783), such as the lepidopteran Manduca sexta lipotropin precursor (US5023328), bee venom (Invitrogen), ecdysteroid UDP glucosyltransferase (egt) (Murphy et al, Protein Expression and Purification 4, 349-357(1993)) or human pancreatic lipase (hpl) (Methods in Enzymology 284, pp.262-272, 1997). Preferred signal peptides for use in mammalian cells are the signal peptide of hG-CSF or the murine Ig kappa light chain signal peptide (Coloma, M (1992) J.Imm.methods 152: 89-104). For use in Yeast cells, suitable signal peptides have been found to be the Saccharomyces cerevisiae alpha-factor signal peptide (see US4870008), the modified carboxypeptidase signal peptide (see L.A. Valls et al, Cell 48, 1987, pp.887-897), the Yeast BAR1 signal peptide (see WO87/02670) and the Yeast aspartic protease 3(YAP3) signal peptide (see M.Egel-Mitani et al, Yeast 6, 1990, pp.127-137) and the synthetic leader sequence TA57(WO 98/32867). A signal peptide found to be suitable for E.coli is ompA.

The nucleotide sequence of the present invention encoding a polypeptide exhibiting G-CSF activity, whether prepared by site-directed mutagenesis, synthesis or other means, may or may not include a nucleotide sequence encoding a signal peptide. The signal peptide is present when the polypeptide is secreted from a cell expressing the polypeptide. If present, the signal peptide should be recognized by the cell selected for expression of the polypeptide. The signal peptide may be homologous (e.g., typically associated with G-CSF) or heterologous (i.e., derived from a source other than hG-CSF) to the polypeptide or may be homologous or heterologous to the host cell, i.e., is a signal peptide normally expressed by the host cell or is not normally expressed by the host cell. Thus, the signal peptide may be prokaryotic, e.g.derived from bacteria such as E.coli, or eukaryotic, e.g.derived from mammalian or insect or yeast cells.

Any suitable host can be used to produce the polypeptides or polypeptide portions of the conjugates of the invention, including bacteria, fungi (including yeast), plant, insect, mammalian or other suitable animal cells or cell lines, and transgenic animals or plants. Examples of bacterial host cells include gram-positive bacteria such as Bacillus, e.g.Brevibacillus or Bacillus subtilis, or Streptomyces, or gram-negative bacteria, e.g.Escherichia coli or Pseudomonas strains. The vectors may be introduced into bacterial host cellsFor example, it can be carried out by protoplast transformation (see, for example, Chang and Cohen, 1979, Molecular General Genetics 168: 111-. Examples of suitable filamentous fungal host cells include Aspergillus strains, such as Aspergillus oryzae, Aspergillus niger or Aspergillus nidulans, Fusarium or Trichoderma. Fungal cells may be transformed, the transformation process involving protoplast formation, transformation of the protoplasts, and regeneration of the cell wall by means known per se. Suitable methods for transforming Aspergillus host cells are described in EP238023 and US 5679543. In Malardier et al 1989, Gene 78: 147-156 and WO96/00787 describe suitable processes for transforming Fusarium. Examples of suitable yeast host cells include strains of Saccharomyces, such as Saccharomyces, Schizosaccharomyces, Kluyveromyces, Pichia, such as Pichia pastoris or P.Methanolica, Hansenula, such as Hansenula polymorpha or Yarrowia. Yeasts can use Becker and Guarente, In Abelson, J.N.And Simon, M.I., editors, Guide to Yeast Genetics and Molecular Biology, Methods In enzymology volume 194, pp 182-; ito et al, 1983, journal of Bacteriology 153: 163; and Hinnen et al 1978, Proceedings of the national Academy of Sciences USA 75: 1920 and Clontech laboratories, Inc, Palo Alto, Calif., USA (in Yeastmaker)^TMThe Yeast transformation System Kit) for transformation. Examples of suitable insect host cells include lepidopteran cell lines such as Spodoptera frugiperda (Sf9 or Sf21) or Trichoplusia ni cells (HighFive) (US 5077214). Insect cells can be transformed and heterologous polypeptides produced therein as described by Invitrogen. Examples of suitable mammalian host cells include the Chinese Hamster Ovary (CHO) cell line (e.g., CHO-K1; ATCC CCL-61),Green monkey cell lines (COS) (e.g., COS1(ATCCCRL-1650), COS7(ATCC CRL-1651)); mouse cells (e.g., NS/O), hamster kidney (BHK) cell lines (e.g., ATCC CRL-1632 or ATCC CCL-10), and human cells (e.g., HEK293(ATCC CRL-1573)), as well as plant cells in tissue culture. Additional suitable cell lines are known in the art and are available from published collections such as the American type culture Collection, Rockville, Maryland. Methods for introducing exogenous DNA into mammalian host cells include calcium phosphate-mediated transfection, electroporation, DEAE-dextran-mediated transfection, liposome-mediated transfection, viral vectors, and transfection methods using Lipofectamin2000 as described by Life Technologies Ltd, Paisley, UK. Such methods are known in the art, e.g., Ausbel et al (eds.), 1996, Current Protocols in Molecular Biology, John Wiley&Sons, New York, USA. Mammalian cell Culture is performed according to established procedures, such as those disclosed in Animal cell Biotechnology, Methods and Protocols, edited by Nigel Jenkins, 1999, Human Press Inc, Totowa, New Jersey, USA and Harrison MA and Rae IF, General technologies of cell Culture, Cambridge University Press 1997.

In the production methods of the invention, the cells are cultured in a nutrient medium suitable for production of the polypeptide using methods known in the art. For example, the cells may be fermented in the laboratory by shake flask culture, small-scale or large-scale fermentation (including continuous, batch, fed-batch, or solid state fermentations), or industrial fermentation in a suitable medium and under conditions that allow the polypeptide to be expressed and/or isolated. The cultivation is carried out in a suitable nutrient medium containing carbon and nitrogen sources and inorganic salts using methods known in the art. Suitable media are commercially available or can be prepared according to published compositions (e.g., as described in catalogues of the American type culture Collection). If the polypeptide is secreted into the nutrient medium, the polypeptide can be recovered directly from the medium. If the polypeptide is not secreted, it can be recovered from cell lysates.

The resulting polypeptide can be recovered by methods known in the art. For example, the polypeptide may be recovered from the nutrient medium by conventional methods including, but not limited to, centrifugation, filtration, extraction, spray drying, evaporation, or precipitation.

The polypeptide can be purified by a variety of methods known in the art, including, but not limited to, chromatography (e.g., ion exchange, affinity, hydrophobic, chromatofocusing, and size exclusion), electrophoretic methods (e.g., preparative isoelectric focusing), solubility differences (e.g., ammonium sulfate precipitation), SDS-PAGE, or extraction (see, e.g., Protein Purification, j. -c. janson and Lars Ryden, eds., VCH Publishers, new york, 1989). Specific methods for purifying polypeptides having G-CSF activity are described in D.Metcalf and N.A.Nicola The chromatographic column-stimulating factors, p.50-51, Cambridge university Press (1995), and in C.S.Bae et al, appl.Microbiol.Biotechnol, 52: 338-344(1999) and US4,810,643.

The invention relates to a pharmaceutical composition and application thereof

The invention further comprises a composition of a polypeptide or conjugate as described herein and at least one pharmaceutically acceptable carrier or excipient.

The polypeptides, conjugates or pharmaceutical compositions of the invention can be used for the preparation of a medicament for the treatment of certain diseases, in particular for the prevention of infections in cancer patients undergoing treatment with certain types of chemotherapy, radiotherapy and bone marrow transplantation, for mobilizing the precursor cell pool in peripheral blood precursor cell transplantation, for the treatment of severe chronic or related leukopenia patients, for the treatment of acute myeloid leukemia patients, for the treatment of AIDS or other immunodeficiency diseases, and antifungal treatment, in particular for the treatment of systemic or invasive candidiasis.

In another aspect, the polypeptide, conjugate or pharmaceutical composition of the invention may be used in a method of treating a patient with a (general) hematopoietic disorder, including those caused by radiation or chemotherapy, especially neutropenia or leukopenia, AIDS or other immunodeficiency disorder, the method comprising administering to a mammal in need thereof the polypeptide, conjugate or pharmaceutical composition described above. In particular, the methods are directed to increasing the level of neutrophils in a patient with insufficient levels of neutrophils (e.g., due to chemotherapy, radiation therapy, or HIV or another viral infection).

The polypeptides and conjugates of the invention are administered to a patient in a "therapeutically effective" dose, i.e., a dose sufficient to produce the desired effect on the disease being treated. The exact dosage will depend on the disease to be treated and may be determined by one skilled in the art using known techniques. The polypeptide or conjugate of the invention may, for example, be administered in a therapeutic dose similar to rhG-CSF, e.g., Neupogen *. Suitable dosages of the conjugates of the invention are expected to be in the range of about 5-300. mu.g/kg body weight (based on the weight of the protein portion of the conjugate), 10-200. mu.g/kg, e.g., 25-100. mu.g/kg. An effective amount of a polypeptide, conjugate or composition of the invention will depend, inter alia, on the disease, the dose and schedule of administration, whether the polypeptide or conjugate or composition is administered alone or in combination with other therapeutic agents, the serum half-life of the composition, the general health of the patient, and the frequency of administration, as will be apparent to those skilled in the art. Preferably, the polypeptide, conjugate, preparation or composition of the invention is administered in an effective amount, especially in a dose sufficient to normalize the number of leukocytes, especially neutrophils, in the patient in question. It is customary to simply count the leukocytes at regular intervals to determine whether the leukocyte count has reached normal.

The polypeptide or conjugate of the invention is preferably administered in the form of a composition comprising one or more pharmaceutically acceptable carriers or excipients. The polypeptide or conjugate may be formulated into pharmaceutical compositions in a manner as is well known to those skilled in the art to produce polypeptide pharmaceuticals which are storage stable and suitable for administration to humans or animals. The pharmaceutical compositions may be formulated in a variety of forms, including liquids or gels, or lyophilized or in any other form. The preferred form will depend on the particular indication to be treated, as will be apparent to those skilled in the art.

Accordingly, the present invention provides compositions and methods for treating various forms of leukopenia or neutropenia. In particular, the polypeptides, conjugates or compositions of the invention may be used to prevent infection in cancer patients undergoing certain types of radiation therapy, chemotherapy and bone marrow transplantation, to mobilize precursor cell recruitment during peripheral blood precursor cell transplantation, to treat severe chronic or related leukopenia and to support treatment in patients with acute myelogenous leukemia. In addition, the polypeptides, conjugates or compositions of the invention may be used in the treatment of AIDS or other immunodeficiency disorders, in antifungal therapy, particularly in the treatment of systemic or invasive candidiasis, and in the treatment of bacterial infections.

In view of the long in vivo half-life of the polypeptide conjugates of the invention and the discovery that they can shorten the duration of neutropenia and leukopenia by administering a single dose, rather than the daily administration necessary for hG-CSF, the conjugates of the invention are well suited for, e.g., once-a-week administration for the purpose of preventing and/or treating neutropenia. In one embodiment, the polypeptide conjugate or the pharmaceutical composition of the present invention is used for the prevention and/or treatment of neutropenia due to chemotherapy. In the case of chemotherapy administered once a week by intravenous injection or another form of injection (e.g., subcutaneous or intramuscular injection), it is generally sufficient to administer a single dose of the conjugate of the invention before, after, or simultaneously with each chemotherapy. In the case of chemotherapy administered by other routes (e.g., once daily oral administration or infusion by infusion pump over an extended period of time), the conjugates of the invention may be administered in a similar manner, e.g., once a week, or once for each chemotherapy when the number of chemotherapies is less than once a week.

Pharmaceutical forms

The polypeptides or conjugates of the invention may be used "as is" and/or in their salt form. Suitable salts include, but are not limited to, alkali metal or alkaline earth metal salts, such as sodium, potassium and magnesium salts, and, for example, zinc salts. These salts and complexes may exist as a crystalline and/or amorphous structure.

Excipient

"pharmaceutically acceptable" refers to a carrier or excipient that is used at a dose and concentration that does not produce any adverse effects in the patient receiving the drug. Such pharmaceutically acceptable carriers and Excipients are well known in the art (see Remington's Pharmaceutical Sciences, 18th edition, A.R. Gennaro, Ed., Mack Publishing Company [1990 ]; Pharmaceutical formulation development of Peptides and Proteins, S.Frokjar and L.Hovgaard, Eds., Taylor & Francis [2000 ]; and Handbook of Pharmaceutical Excipients, 3rd edition, A.Kibbe, Ed., Pharmaceutical Press [2000 ]).

Mixing of drugs

The pharmaceutical compositions of the present invention may be administered alone or in combination with other therapeutic agents. These formulations may be administered as part of the same pharmaceutical composition, or separately from the polypeptide or conjugate of the invention, simultaneously or according to another treatment schedule. In addition, the polypeptide or conjugate or pharmaceutical composition of the invention may be used as an adjunct to other therapies.

Patient's health

The term "patient" as used herein includes both humans and other mammals. The method is therefore useful for human therapy and veterinary applications.

Route of administration

The formulations of the present invention may be administered by a variety of means including, but not limited to, orally, subcutaneously, intravenously, intracerebrally, intranasally, intradermally, intraperitoneally, intramuscularly, intrapulmonary, intravaginally, rectally, intraventricularly, or in any other acceptable manner. The formulations may be administered by continuous infusion or by bolus injection (bolus injection) using techniques well known in the art. Typically, the formulations are designed to be suitable for parenteral administration, for example by the subcutaneous route.

Parenteral administration of medicaments (Parentals)

An example of a pharmaceutical composition is a solution designed for parenteral administration. The parenteral formulation may also be provided in frozen or lyophilized form, although in many cases the pharmaceutical solution may be provided in a liquid formulation suitable for immediate use. In the former case, the composition must be thawed prior to use. The latter dosage form is typically used to enhance the stability of the active ingredients contained in the composition under various storage conditions, and as is known to those skilled in the art, lyophilized formulations are typically more stable than their liquid counterparts. The lyophilized formulation is reconstituted prior to use by the addition of one or more suitable pharmaceutically acceptable diluents such as sterile water for injection or sterile physiological saline solution.

In the case of parenteral administration, a lyophilized preparation or an aqueous solution is prepared for use, for example, by appropriately mixing the polypeptide having the desired purity with one or more pharmaceutically acceptable carriers, excipients or stabilizers (collectively referred to as "excipients") commonly used in the art, such as buffers, stabilizers, preservatives, isotonic agents, nonionic detergents, antioxidants and/or other various additives.

Buffering agents help to maintain the pH in a range near physiological conditions. They are usually present in a concentration range of about 2mM to 50 mM. Suitable buffers for use in the present invention include organic and inorganic acids and salts thereof such as citrate buffers (e.g., monosodium citrate-disodium citrate mixture, trisodium citrate mixture, monosodium citrate-sodium citrate mixture, etc.), succinate buffers (e.g., succinic acid-monosodium succinate mixture, succinic acid-sodium hydroxide mixture, succinic acid-disodium succinate mixture, etc.), tartrate buffers (e.g., tartaric acid-sodium tartrate mixture, tartaric acid-potassium tartrate mixture, tartaric acid-sodium hydroxide mixture, etc.), fumarate buffers (e.g., fumaric acid-monosodium fumarate mixture, fumaric acid-disodium fumarate mixture, monosodium fumarate-disodium fumarate mixture, etc.), gluconate (gluconate) salt buffers (e.g., gluconic acid-sodium gluconate mixtures, gluconic acid-sodium hydroxide mixtures, gluconic acid-potassium gluconate mixtures, and the like), oxalate buffers (e.g., oxalic acid-sodium oxalate mixtures, oxalic acid-sodium hydroxide mixtures, oxalic acid-potassium oxalate mixtures, and the like), lactate buffers (e.g., lactic acid-sodium lactate mixtures, lactic acid-sodium hydroxide mixtures, lactic acid-potassium lactate mixtures, and the like), and acetate buffers (e.g., acetic acid-sodium acetate mixtures, acetic acid-sodium hydroxide mixtures, and the like). Other possible buffers are phosphate buffers, histidine buffers and trimethylamine salts such as Tris.

Preservatives are added to retard microbial growth, typically in amounts of about 0.2% to 1% (w/v). Suitable preservatives for use in the present invention include phenol, benzyl alcohol, m-cresol, methyl paraben (paraben), propyl paraben, octadecyl dimethyl benzyl ammonium chloride, benzylidene onium halides (such as benzylidene onium chloride, bromide or iodide), hexane diamine chloride, alkyl parabens such as methyl or propyl paraben, catechol, resorcinol, cyclohexanol and 3-pentanol.

Isotonic agents, including polyhydric sugar alcohols, preferably trihydric or higher sugar alcohols, such as glycerol, erythritol, arabitol, xylitol, sorbitol and mannitol, may be added to ensure isotonicity of the liquid composition. The amount of polyhydric alcohol may range from 0.1% to 25% by weight, typically from 1% to 5%, calculated as the relative amounts of the other components.

Stabilizers refer to a broad class of excipients whose function ranges from bulking agents to additives that dissolve the therapeutic agent or help prevent denaturation or adhesion to the container walls. Typical stabilizers may be polyhydric sugar alcohols (listed above); amino acids such as arginine, lysine, glycine, glutamine, asparagine, histidine, alanine, omithine, L-leucine, 2-phenylalanine, glutamic acid, threonine, and the like, organic sugars or sugar alcohols such as lactose, trehalose, stachyose, mannitol, sorbitol, xylitol, ribitol, inositol, galactitol, glycerol, and the like, including cyclic alcohols such as inositol; polyethylene glycol; an amino acid polymer; sulfur-containing reducing agents such as urea, glutathione, lipoic acid, sodium thioglycolate, thioglycerol, alpha-monothioglycerol and sodium thiosulfate; low molecular weight polypeptides (i.e., < 10 residues); proteins such as human serum albumin, bovine serum albumin, gelatin or immunoglobulins; hydrophilic polymers such as polyvinylpyrrolidone; monosaccharides such as xylose, mannose, fructose and glucose; disaccharides such as lactose, maltose and sucrose; trisaccharides such as raffinose, and polysaccharides such as dextran. The stabilizer is generally present in the range of 0.1 to 10000 parts by weight, calculated on the weight of active protein.

Nonionic surfactants or detergents (also referred to as "wetting agents") may be present to help solubilize the therapeutic agent and protect the therapeutic polypeptide from agitation-induced aggregation, which also allows the formulation to be exposed to shear surface pressure without causing denaturation of the polypeptide. Suitable nonionic surfactants include polysorbates (20, 80, etc.), polyoxamers (184, 188, etc.), Pluronic * polyols, polyoxyethylene sorbitan monoethers (Tween * 20, Tween * 80, etc.).

Other excipients include bulking or bulking agents (e.g., starch), chelating agents (e.g., EDTA), antioxidants (e.g., ascorbic acid, methionine, vitamin E), and co-solvents.

The active ingredient may also be encapsulated in microcapsules prepared, for example, by coascervation techniques or by interfacial polymerization, for example, hydroxymethylcellulose, gelatin, or poly (methylmethacylate) microcapsules, in colloidal drug delivery systems (for example, liposomes, albumin microspheres, microemulsions, nanoparticles, and nanocapsules) or in macroemulsions. These techniques are disclosed in Remington's pharmaceutical sciences (supra).

Parenteral formulations used for in vivo administration must be sterile. This process is easily accomplished, for example, by filtration through sterile filtration membranes.

Sustained release preparation

Suitable examples of sustained release formulations include solid hydrophobic polymeric semipermeable materials containing the polypeptide or conjugate, materials having suitable forms such as membranes or microcapsules. Examples of sustained release materials include polyesters, hydrogels (e.g., poly (2-hydroxyethyl methacrylate) or poly (vinyl alcohol)), polylactides, copolymers of L-glutamic acid and ethyl L-glutamate, non-degradable ethylene ethyl acetate, degradable lactic acid-glycolic acid copolymers such as ProLease * technology or Lupron Depot * (injectable microspheres composed of lactic acid-glycolic acid copolymer and leuprolide acetate), and poly-D- (-) -3-hydroxybutyric acid. Polymers such as ethylene ethyl acetate and lactic acid-glycolic acid are capable of releasing molecules for long periods of time, e.g., up to or over 100 days, with some hydrogels releasing proteins in a shorter period of time. When the encapsulated polypeptides remain in the body for a long period of time, they denature or aggregate upon exposure to moisture at 37 ℃, resulting in loss of biological activity and possibly altered immunogenicity. Rational stabilization strategies can be designed according to the mechanism involved. For example, if the aggregation mechanism is found to be intramolecular S — S bond formation by thio-disulfide interexchange, stabilization can be achieved by modification of sulfhydryl groups, lyophilization of acidic solutions, control of moisture content, use of appropriate additives, and development of specific polymer matrix compositions.

All references cited herein are incorporated by reference in their entirety.

The invention will be further described by the following non-limiting examples.

Drawings

FIG. 1: the in vivo half-life of rhG-CSF (Neupogen *) and SPA-PEG 5000-conjugated hG-CSF K16RK34R K40R Q70K Q90K Q120K.

FIG. 2: the in vivo half-life of rhG-CSF (Neupogen *) and SPA-PEG 5000-conjugated hG-CSF K16RK34R K40R Q90K T105K Q159K.

FIG. 3: in vivo biological activity of rhG-CSF (Neupogen *), SPA-PEG 5000-conjugated hG-CSF K16R K34R K40R Q70K Q120K and SPA-PEG 5000-conjugated hG-CSFK16R K34R K40R Q70K Q90K Q120K in healthy rats.

FIG. 4: in vivo biological activity of rhG-CSF (Neupogen *), SPA-PEG 5000-conjugated hG-CSF K16R K34R K40R Q70K Q120K T133K and SPA-PEG 5000-conjugated hG-CSF K16R K34R K40R Q90K Q120K T133K in healthy rats.

FIG. 5: in vivo biological activity of rhG-CSF (Neupogen *), SPA-PEG 12000-conjugated hG-CSF K16R K34R K40R and different doses of SPA-PEG 5000-conjugated hG-CSFK16R K34R K40R Q70K Q90K Q120K in healthy rats.

FIG. 6: in healthy rats, the in vivo biological activity of rhG-CSF (Neupogen *), SPA-PEG 5000-conjugated hG-CSF K16R K34R K40R Q70K Q90K Q120K, SPA-PEG 5000-conjugated hG-CSF K16R K34R K40R Q90K T105K S159K and SPA-PEG 20000-conjugated hG-CSF K16R K34R K40R T105K S159K.

FIG. 7: in vivo biological activity of rhG-CSF (Neupogen *), SPA-PEG 5000-conjugated hG-CSF K16R K34R K40R Q70K Q90KT105K, and SPA-PEG 20000-conjugated hG-CSF K16R K34R K40R Q90K in rats with chemotherapy-induced neutropenia.

FIG. 8: in vivo biological activity (white blood cell count) of rhG-CSF (Neupogen *), SPA-PEG 5000-coupled hG-CSF K16R K34R K40R T105KS159K and SPA-PEG 5000-coupled hG-CSF K16R K34R K40R Q90K T105KS159K in rats with chemotherapy-induced neutropenia.

FIG. 9: in vivo biological activity (absolute neutrophil count) of rhG-CSF (Neupogen *) and SPA-PEG 5000-conjugated hG-CSF K16R K34R K40R T105KS159K in rats with chemotherapy-induced neutropenia.

FIG. 10: in rats with chemotherapy-induced neutropenia, rhG-CSF (Neupogen *), rhG-CSF with a 20kDa N-terminal PEG group (Neulasta)^TM) And in vivo biological activity (white blood cell count) of SPA-PEG 5000-conjugated hG-CSF K16R K34RK40R T105K S159K produced in yeast and CHO cells.

FIG. 11: in rats with chemotherapy-induced neutropenia, rhG-CSF (Neupogen *), rhG-CSF with a 20kDa N-terminal PEG group (Neulasta)^TM) And the in vivo biological activity (absolute neutrophil count) of SPA-PEG 5000-conjugated hG-CSF K16R K34RK40R T105K S159K produced in yeast and CHO cells.

Sequence listing

The attached sequence listing contains the following sequences:

SEQ ID NO: 1: amino acid sequence of human G-CSF.

SEQ ID NO: 2: synthetic DNA sequences encoding human G-CSF have codon usage characteristics that facilitate optimal expression in E.coli.

SEQ ID NO: 3: the amino acid sequence of the OmpA signal sequence.

SEQ ID NO: 4: a synthetic DNA sequence encoding the OmpA signal sequence.

SEQ ID NO: 5: a synthetic histidine tag.

SEQ ID NO: 6: encoding the amino acid sequence of SEQ ID NO: 5, and a synthetic DNA sequence of the histidine tag.

SEQ ID NO: 7: the amino acid sequence of the human G-CSF signal peptide.

SEQ ID NO: 8: encodes human G-CSF comprising SEQ ID NO: 7, having codon usage characteristics suitable for optimal expression in CHO cells.

SEQ ID NO: 9-15: various artificially synthesized markers

Materials and methods

Method for determining amino acids in need of modification

Accessible Surface Area (ASA)

A3D complete picture of the 10 structures determined by NMR spectroscopy (Zink et al, (1994) Biochemistry 33: 8453-8463) was available from Protein databases (Protein Data Bank) (PDB) (www.rcsb.org/PDB /). This information can be entered into the computer program Access (B.Lee and F.M.Richards, J.mol.biol.55: 379-. This method generally uses a probe size of 1.4 * and defines the Accessible Surface Area (ASA) as the area formed by the center of the probe. All water molecules and all hydrogen atoms, as well as other atoms not directly linked to the protein, were removed from the coordinates prior to this calculation.

Fractional ASA of the side chain

The fraction of side chain atoms, ASA, was calculated by dividing the sum of ASA's of the atoms in the side chain by the ASA representative value of the side chain atoms of that residue type in the extended ALA-x-ALA tripeptide. See Hubbard, Campbell & Thornton (1991) J.mol.biol.220, 507-. In this example, the CA atom is considered part of the side chain of the glycine residue but not part of other residues. The following table represents the 100% ASA standard for the side chain:

Ala 69.23 *² Leu 140.76 *²

Arg 200.35 *² Lys 162.50 *²

Asn 106.25 *² Met 156.08 *²

Asp 102.06 *² Phe 163.90 *²

Cys 96.69 *² Pro 119.65 *²

Gln 140.58 *² Ser 78.16 *²

Glu 134.61 *² Thr 101.67 *²

Gly 32.28 *² Trp 210.89 *²

His 147.00 *² Tyr 176.61 *²

Ile 137.91 *² Val 114.14 *²

residues not detected in this structure are defined as 100% exposed, since these residues can be considered to be located in the elastic region.

Determining distances between atoms

The MSI INC can very easily determine the distance between atoms using molecular mapping software, such as InsightII * v.98.0.

General considerations regarding the residues to be modified

As mentioned above, the amino acid residues to be modified according to the present invention are preferably the following amino acid residues: the surface of its side chains is exposed, in particular more than 25% of its side chains are exposed on the surface of the molecule, more preferably more than 50% of the side chains are exposed on the surface of the molecule. Another consideration is that residues on the receptor contact surface are preferably excluded so that their binding or activation to the receptor may interfere or at least minimize interference. It is further contemplated that residues less than 10 * from the nearest lys (Glu, Asp) CB-CB (CA for Gly) should be excluded. Finally, preferred modification sites are in particular those having hydrophilic and/or charged residues, i.e. Asp, Asn, Glu, Gln, Arg, His, Tyr, Ser and Thr, sites having arginine residues being particularly preferred.

Identification of G-CSF amino acid residues in need of modification

The following are examples of factors that should generally be considered in determining the amino acid residue to be modified in the present invention.

Three-dimensional structures of human G-CSF have been reported using X-ray crystallography (Hill et al, (1993) Proc. Natl. Acad. Sci. USA 90: 5167-5171) and NMR spectroscopy (Zink et al, (1994) Biochemistry 33: 8453-8463). As described above, Aritomi et al (nature 401: 713-717, 1999) have identified the following hG-CSF residues as part of the receptor binding interface: g4, P5, a6, S7, S8, L9, P10, Q11, S12, L15, K16, E19, Q20, L108, D109, D112, T115, T116, Q119, E122, E123, and L124. Thus, although these residues may be modified, it is preferred not to modify these residues.

The following residues were identified as ASA with 25% or more by using as input structure 10 NMR structures of G-CSF as determined by Zink et al (1994), and then calculating the average ASA of the side chains: m0, T1, P2, L3, G4, P5, A6, S7, S8, L9, P10, Q11, S12, F13, L14, L15, K16, C17, E19, Q20, V21, R22, K23, Q25, G25, D25, A25, A25, E25, K25, C25, A25, T25, Y25, K25, L25, H25, P25, E25, E25, V25, L25, L25, H25, S25, L25, I25, P25, P25, L25, S25, S25, S25, P25, P25, P25, S25, P72, S25, P167, S25, S25, P167, L167, P165, L159, P165, S124, S165, S124, S165, S113, S124, S165, S113, S124, S113, S165, S113, S124, S165, S113, S123, S113, S165, S113, S124, S123, S117, S113, S117, S165, S123, S113, S123, S117, S123, S117, S123, S, r169, H170, L171, a172, Q173, P174.

Also, the following residues have 50% or more ASA: m0, T1, P2, L3, G4, P5, a6, S7, S8, L9, P10, Q11, S12, F13, L14, L15, K16, C17, E19, Q20, R22, K23, G26, D27, a30, E33, K34, T38, K40, L41, H43, P44, E45, E46, L49, L50, S53, P57, P60, L61, S62, S63, P65, S65, Q65, a 65, L65, Q65, L65, a 65, G65, S65, F65, Q65, G167, P168, L124, P6321, P202, P168, P166, P123, P202, P168, P166, P123, M166, M123, P123, M166, M113, M123, P.

The molecular mapping program InsightII * v.98.0 identified residues with a CB atom (CA in the case of glycine) above the nearest amino group 15 *, such atoms being identified as the NZ atom of lysine and the N atom of the N-terminal residue T1. The following list includes residues that meet this criterion in at least one of the 10 NMR structures: g4, P5, a6, S7, S8, L9, P10, Q11, L14, L15, L18, V21, R22, Q25, G26, D27, G28, a29, Q32, L35, C36, T36, Y36, C36, H36, P36, E36, L36, V36, L36, G36, H36, S36, L36, G36, P36, a36, L36, P36, L36, S36, C36, P36, S36, Q36, a36, Q36, L36, a36, L36, G72, G124, L36122, L36, L36122, L36, L36120, L36, P36, a148, G149, G150, V151, L152, V153, a154, S155, H156, L157, Q158, S159, F160, L161, E162, V163, S164, Y165, R166, V167, L168, R169, H170, L171, a172, Q173, P174.

Molecular mapping program insight ii * v.98.0 was also used to identify residues having a CB atom (CA in the case of glycine) above the nearest acidic group 10 *, identified as the CG atom of aspartic acid, the CD atom of glutamic acid and the C atom of the C-terminal residue P174. The following list includes residues that meet this criterion in at least one of the 10 NMR structures: m0, T1, P2, L3, G4, P5, a6, S7, S8, L9, P10, Q11, S12, P13, L14, T38, Y39, K40, L41, C42, L50, G51, H52, S53, L54, G55, I56, P57, W58, a59, P60, L61, S62, S63, C64, P65, S66, Q67, a68, L69, Q70, L70, a 70, G70, C70, L70, S70, Q70, L70, H70, S70, G70, L70, F72, F149, Y3, G8972, G36142, L124, L152, L124, G168, G124, G143, G168, G143, G168, G143, G168, G143, n 143, M168, G143.

Taken together, and by comparison, it has been found that it is possible to select single amino acid residues for modification, thereby resulting in a limited set of amino acid residues whose modification in a particular G-CSF polypeptide is likely to result in the desired property.

PEGylation method of hG-CSF and variants thereof

PEGylation of hG-CSF and variants thereof in solution

Human hG-CSF and variants thereof were PEGylated in a solution of 50mM sodium phosphate, 100mM NaCl, pH8.5 at a concentration of 250. mu.g/ml. The number of moles of PEG exceeds 50-100 times of the PEGylation sites on the protein. The reaction mixture was placed in a37 ℃ hot mixer at 1200rpm for 30 minutes. After 30 minutes, the reaction was stopped by adding a molar excess of glycine.

Excess PEG, glycine and other by-products were removed from the reaction mixture by cation exchange chromatography. The PEGylation reaction mixture was diluted with 20mM sodium citrate pH2.5 to an ionic strength of 7mS/cm or less. The pH was adjusted to 2.5 with 5N HCl. The mixture was applied to a SP-sepharose FF column equilibrated with 20mM sodium citrate pH 2.5. Unbound material was washed off the column with 4 column volumes of equilibration buffer. The PEGylated protein was eluted in three column volumes with 20mM sodium citrate, 750mM sodium chloride. Pure PEGylated G-CSF was concentrated and buffer exchanged with a VivaSpin concentrator with a molecular weight cut-off (mwco) of 10 kDa.

PEGylation of labeled polypeptides with G-CSF Activity in microtiter plates

The polypeptide carrying the appropriate marker, e.g.any of the markers listed in the general description above, and exhibiting G-CSF activity is expressed and the culture broth is then transferred to one or more wells of a microtiter plate capable of immobilizing the marker polypeptide. When the marker was Met-Lys-His-Gln-His-Gln-His-Gln-His-Gln-His-Gln-Gln, a nickel-nitrilotriacetic acid (Ni-NTA) HisSorb microtiter plate available from QIAGEN was used.

The labeled polypeptide is immobilized to a microtiter plate, and each well is then washed with a buffer suitable for binding and subsequent pegylation, and then incubated with the selected activated PEG. For example, M-SPA-5000 by Shearwater Corp. The molar ratio of activated PEG to polypeptide should be optimized, but is generally greater than 10: 1, e.g., about 100: 1 or higher. After a suitable time at room temperature, typically about 1 hour, the activated PEG solution is removed to stop the reaction. The conjugated protein is eluted from the microtiter plate by incubation with an appropriate buffer. Suitable elution buffers may comprise imidazole, excess NTA or another chelating compound. The coupled proteins were analyzed for the corresponding biological activity and immunogenicity. The label can optionally be cleaved off using methods known in the art, such as using a diaminopeptidase and converting the Gln at the 1-position to pyroglutamyl using GCT (glutamyl cyclotransferase) and finally cleaving it off with PGAP (pyroglutamyl-aminopeptidase) to produce a label-free protein. The method comprises several steps of metal chelate affinity chromatography. Alternatively, the labeled polypeptide may be conjugated.

PEGylation of hG-CSF active polypeptides with blocked receptor binding sites

In order to optimize the pegylation of hG-CSF in a certain way and to ensure that the lysines involved in receptor recognition are not pegylated, the following methods have been developed:

purified hG-CSF was obtained as described in example 4. A homodimeric complex of hG-CSF polypeptide and G-CSF receptor soluble domain with a 2: 2 stoichiometric composition was formed in Phosphate Buffered Saline (PBS) pH 7. The concentration of hG-CSF polypeptide is about 20ug/ml or 1uM and the receptor is present at the same molar concentration.

M-SPA-5000 from Shearwater Corp. was added at three different concentrations, 10, 25 and 50 molar excess over the number of attachment sites in the hG-CSF polypeptide, respectively. The reaction time was 30 minutes at room temperature. After 30 minutes of reaction, the reaction mixture was adjusted to pH2.0 and applied to a Vydac C18 column and then eluted with an acetonitrile gradient essentially as described (Utsumi et al, J. biochem., Vol.101, 1199-one 1208, (1987). alternatively, an isopropanol gradient can be used.

The fractions were analyzed using the primary screening assay described herein, and the active pegylated hG-CSF polypeptide obtained as described above was then stored at-80 ℃ in PBS pH7 containing 1mg/ml Human Serum Albumin (HSA).

Method for identifying conjugated and unconjugated hG-CSF and variants thereof

Determining molecular size of hG-CSF and variants thereof

The molecular weight of the coupled or uncoupled hG-CSF or variants thereof can be determined by SDS-PAGE, gel filtration, matrix assisted laser desorption mass spectrometry (matrix assisted laser desorption) assaymeasurement) or equilibrium centrifugation. As explained, SDS-PAGE provides information about "apparent molecular weight". The true molecular weight can be conveniently determined using mass spectrometry. SDS-PAGE from Invitrogen^TMNuPAGE * kit (high-performance pre-cast gel) by Novex. Mu.l of the sample was loaded onto NuPAGE 4-12% Bis-Tris gel (Cat. Nr. NPO321) and eluted at 200v and 120mA in NuPAGE MES SDS running buffer (Cat. Nr. NPO002-02) for 35 min.

Determination of polypeptide concentration

The concentration of the polypeptide is determined by 280nm densitometry, enzyme-linked immunosorbent assay (ELISA), Radioimmunoassay (RIA), or other such immunoassay techniques well known in the art. Furthermore, the concentration of the polypeptide in the sample can also be detected using a Biacore * instrument using a Biacore * chip coated with antibodies specific for the polypeptide.

Covalent coupling of the above antibodies to Biacore by various chemical methods^*On the chip. Alternatively, the antibody is bound non-covalently, for example by an antibody specific for the Fc portion of an anti-polypeptide antibody. Fc-specific antibodies were first covalently coupled to the chip. The anti-polypeptide antibody is then flowed over the chip and bound directly by the primary antibody. Further, a streptavidin-coated surface (e.g., Biacore sensor Chip SA *) can be used to immobilize biotinylated antibodies (Real-Time Analysis of biomolecular interactions, Nagata and Handa (Eds.), 2000, Springer Verlag, Tokyo; Biacore2000 Instrument Handbook, 1999, Biacore AB).

When the sample flows through the chip, the polypeptide can be bound to the coated antibody and the increase in mass can be detected. A standard curve can be plotted using a preparation of polypeptides of known concentration, followed by determination of the concentration of the polypeptide in the sample. After each sample injection, the sensor chip is regenerated by means of a suitable eluent (e.g. a low pH buffer) capable of removing the bound analyte.

Typically, the antibody used is a monoclonal antibody directed against the wild-type polypeptide. Introduction of mutations or other manipulations (additional glycosylation or polymer conjugation) in the wild-type polypeptide can alter the recognition of the above-described antibody. Further, manipulation to increase the molecular weight of the polypeptide will enhance the cytoplasmic gene resonance signal (plasmon resonance signal). Therefore, it is necessary to establish a standard curve for each molecule to be tested.

Method for detecting in vitro and in vivo activity of conjugated and unconjugated hG-CSF and variants thereof

Preliminary test 1-in vitro G-CSF Activity test

Proliferation of the murine cell line NFS-60 (obtained from Dr. J. Ihle, St. Jude Children's, Research Hospital, Tennessee, USA) was dependent on the presence of active G-SCF in the growth medium. Thus, the in vitro biological activity of hG-CSF and variants thereof can be measured by adding a sample of G-CSF to growth medium and measuring the number of dividing NFS-60 cells after a period of incubation.

NFS-60 cells were cultured in Iscoves DME medium containing 10% w/w FBS (fetal bovine serum), 1% w/w Pen/Strep, 10. mu.g hG-CSF per liter and 2mM Glutamax. Cells were washed twice with hG-CSF free growth medium before addition of sample, and then diluted to 2.2X 10⁵Concentration of cells/ml. Add 100. mu.l of cell suspension to each well of a 96-well microtiter plate (Corning).

Samples containing coupled or uncoupled G-CSF or variants thereof are diluted with growth medium to a concentration of 1.1X 10^-6M to 1.1X 10^-13And M. Mu.l of each sample was added to 3 wells containing NFS-60 cells. Controls consisting of 10 μ l of mammalian growth medium were added to 8 wells of each microtiter plate. The cells were cultured for 48 hours (37 ℃, 5% CO)₂) The dividing cells in each well were then quantified using WST-1 cell proliferation reagent (Roche diagnostics GmbH, Mannheim, Germany). 0.01ml of WST-1 was added to each well, followed by 5% CO₂Was incubated at 37 ℃ for 150min under an air atmosphere. Formazan (f) formed by decomposing tetrazolium salt WST-1 by mitochondrial dehydrogenase in living cellsormazan) which can be quantified by absorbance at 450 nm. Thereby quantifying viable cells in each well.

Based on the above assays, dose response curves were calculated for each conjugated and unconjugated G-CSF molecule or variant thereof, and EC50 values were determined for each molecule. This value is equal to the amount of active G-CSF protein necessary to obtain 50% of the maximal proliferative activity of unconjugated human G-CSF. Thus, EC50 values directly measure the in vitro activity of a given protein, with lower EC50 values indicating higher activity.

Preliminary test 2-in vitro G-CSF Activity test

The murine hematopoietic cell line BaF3 was transfected with a plasmid carrying the human G-CSF receptor gene, the transcriptional regulator, the promoter of fos and, thereafter, the luciferase reporter gene. Stimulation of the above cell lines with G-CSF samples, various intracellular responses elicited stimulation of fos expression, which subsequently led to luciferase expression. The stimulus may be via Steady-Glo^TMThe luciferase assay system (Promega, Cat. No. E2510) was monitored, and the in vitro activity of G-CSF samples was assayed.

BaF3/hGCSF-R/pfos-lux cells were grown to 5X 10 ℃ in a complete medium (RPMI-1640/HEPES (Gibco/BRL, Cat. No.22400), 10% FBS (HyClone, specific), 1 XPenicillin/streptomycin (Gibco/BRL, Cat. No.15140-122), 1 XP-glutamine, (Gibco/BRL, Cat. No.25030-081), 10% WEHI-3 medium (muIL-3-producing) under a humidified 5% CO2 atmosphere at 37 ℃ to grow to 5X 10⁵Density of cells/mL (confluent). About 2X 10 every 2-3 days⁴cells/mL were reseeded.

The day before the assay, cells were plated at 2X 10 log phase⁵cells/mL were suspended in starvation medium (DMEM/F-12(Gibco/BRL, Cat. No.11039), 1% BSA (Sigma, Cat. No. A3675), I.times.penicillin/streptomycin (Gibco/BRL, Cat. No.15140-122), 1. times.L-glutamine (Gibco/BRL, Cat. No.25030-081), 0.1% WEHI-3 medium), maintained in starvation for 20 hours. The cells were then washed twice with PBS and stained with trypan blue viabilityDetecting the activity of the cell. Cells were plated at 4X 10⁶The cells/mL are resuspended in test medium (RPMI-1640 (phenol Red free, Gibco/BRL, Cat. No.11835), 25mM HEPES, 1% BSA (Sigma, Cat. No. A3675), 1 XPenicillin/streptomycin (Gibco/BRL, Cat. No.15140-122), 1 XP-glutamine (Gibco/BRL, Cat. No.25030-081), and 50. mu.l of the cells are added per well in a 96-well microtiter plate (Corning). A sample containing coupled or uncoupled G-CSF or a variant thereof is diluted with test medium to a concentration of 1.1X 10^-7M to 1.1X 10^-12And M. 50 μ l of each sample was added to three wells containing BaF3/hGCSF-R/pfos-lux cells. Negative control 50. mu.l of medium was added to 8 wells of each microtiter plate. The plates were gently mixed and incubated at 37 ℃ for 2 hours. According to Promega Steady-Glo^TMThe luciferase activity was detected by the procedure of (Promega Steady-GIo luciferase detection System, Cat. No. E2510). Add 100. mu.l of substrate per well and mix gently. Luminescence was detected in the SPC (single photon counting) mode on a TopCount luminometer (Packard).

Based on the above assays, dose response curves can be plotted for each conjugated and unconjugated G-CSF molecule or variant thereof, after which EC50 values for each molecule can be determined.

Second step test of the binding affinity of G-CSF or variants thereof to the hG-CSF receptor

The binding of rhG-CSF or a variant thereof to the hG-CSF receptor was investigated using standard binding assays. The receptor may be a purified extracellular receptor domain, a receptor that binds to the plasma membrane of a purified cell, or an entire cell, which may be a cell line that naturally expresses the G-CSF receptor (e.g., NFS-60) or a cell transfected with cDNAs encoding the receptor. The ability of rhG-CSF or a variant thereof to compete with the native hG-CSF for binding sites can be assayed by co-incubation with a labeled G-CSF-analog, e.g., biotinylated hG-CSF or radioiodinated hG-CSF. An example of such an assay is described by Yamasaki et al (drugs. Exptl. Clin. Res.24: 191-196 (1998)).

The extracellular domain of the hG-CSF receptor is optionally coupled to Fc and immobilized onto a 96-well plate. RhG-CSF or a variant thereof is then added, the binding of which can be detected with a specific anti-hG-CSF antibody or biotinylated or radioiodinated hG-CSF.

Determination of the in vivo half-life of coupled and uncoupled rhG-CSF and variants thereof

An important aspect of the present invention is the extension of the biological half-life, which can be achieved by constructing hG-CSF in which the polypeptide is coupled or not coupled to a polymer component. The rapid depletion of hG-CSF serum concentrations makes it important to assess the biological response to treatment with both conjugated and unconjugated hG-CSF and variants thereof. Preferably, the conjugated and unconjugated hG-CSF and variants thereof of the invention also have an extended serum half-life after intravenous (i.v.) administration, enabling e.g.ELISA methods or preliminary screening assays. The in vivo biological half-life is measured as described below.

Male Sprague Dawley rats (7 weeks old) were used. On the day of dosing, animals were weighed (280-310 g per animal). The unconjugated and conjugated hG-CSF samples were injected into three rats via tail vein at 100. mu.g per kg body weight. 1 minute, 30 minutes, 1, 2,4, 6, and 24 hours after injection in CO₂Under anesthesia, 500. mu.l of blood was collected from the eyeball of each rat. Storage of blood samples 1 at Room temperature¹/₂After hours, the serum was centrifuged (4 ℃, 18000Xg for 5 minutes). Blood samples were stored at-80 ℃ until the day of analysis. The active G-CSF in serum samples was quantified by the G-CSF in vitro activity assay (see preliminary experiment 2) after thawing the samples on ice.

Another assay for determining the in vivo half-life of G-CSF or variants thereof is described in U.S. Pat. No. 5,824,778, which is incorporated herein by reference.

In vivo biology of conjugated and unconjugated hG-CSF and variants thereof in healthy rats Activity of

SPF Sprague Dawley rats, M & B A/S from Denmark, tested for in vivo biological activity of hG-CSF and used to assess the biological efficacy of coupled and uncoupled G-CSF and variants thereof.

On the day of arrival rats were assigned randomly in groups of 6. All animals were acclimated for 7 days, and those with poor physical condition or overweight were eliminated. The weight range at the beginning of the acclimation phase was 250- & gt 270 g.

Rats were fasted for 16 hours on the day of administration, followed by subcutaneous injection of hG-CSF or a variant thereof at 100. mu.g per kg body weight. Each hG-CSF sample was injected into a group of 6 randomly assigned rats. Before and 6, 12, 24, 36, 48, 72, 96, 120 and 144 hours after administration, 300. mu.l of blood was taken from the tail vein of rats and stabilized with EDTA. Blood samples were analyzed for the following hematological parameters: hemoglobin, red blood cell count, hematocrit, mean cell volume, mean cell hemoglobin concentration, mean cell hemoglobin, white blood cell count, differentiated white blood cell count (neutrophils, lymphocytes, eosinophils, basophils, monocytes). Based on these assays, the biological efficacy of conjugated and unconjugated hG-CSF and variants thereof can be assessed. Other assays for determining the in vivo biological activity of hG-CSF or variants thereof can be found in US5,681,720, US5,795,968, US5,824,778, US5,985,265 and Bowen et al, Experimental Hematology 27: 425-432(1999).

Determination of conjugated and unconjugated forms in rats with chemotherapy-induced neutropenia In vivo biological Activity of hG-CSF and variants thereof

SPF Sprague Dawley rats were purchased from M & B A/S, Denmark. Rats were assigned randomly as 6 groups on the day of arrival. All animals were acclimated for 7 days, and those with poor physical condition or overweight were eliminated. The weight range of the rats at the beginning of the acclimation phase was 250-270 g.

Rats were injected with 50 or 90mg/kg body weight Cyclophosphamide (CPA) intraperitoneally 24 hours before administration of the hG-CSF sample. The pegylated hG-CSF variant was injected subcutaneously as a single dose on day 0, while the unconjugated hG-CSF was injected subcutaneously as a single dose on day 0 or on a once-a-day schedule. The dose at day 0 when hG-CSF or variant was administered was 100. mu.g/kg body weight. The dose varied when unconjugated hG-CSF (Neupogen *) was administered once daily, see examples below. Each hG-CSF sample was injected with 6 randomly assigned rats. Before administration, and 6, 12, 24, 36, 48, 72, 96, 120, 144 and 168 hours after administration, 300 μ l of blood was collected from the tail vein of rats and stabilized with EDTA. These blood samples were analyzed for the following hematological parameters: hemoglobin, red blood cell count, hematocrit, mean cell volume, mean cell hemoglobin concentration, mean cell hemoglobin, white blood cell count, differentiated white blood cell count (neutrophils, lymphocytes, eosinophils, basophils, monocytes). Based on these measurements, the biological efficacy of conjugated and unconjugated hG-CSF and variants thereof was evaluated.

Determination of receptor-binding affinity (binding and dissociation rates) of a polypeptide

The strength of binding between the receptor and the ligand can be measured using enzyme-linked immunosorbent assays (ELISAs), Radioimmunoassays (RIAs), or other such immunoassay techniques well known in the art. Ligand-receptor binding interactions can also be detected using a Biacore * instrument for cytoplasmic resonance detection (Zhou et al, Biochemistry, 1993, 32, 8193-98; Faegerstranm and O' Sh annessy, 1993, In Handbook of Affinity Chromatography, 229-52, Marcel Dekker, Inc., NY).

The Biacore * technique allows a receptor to bind to a gold surface and a ligand to flow through the receptor. Plasmon resonance detection measures the direct measurement of substances bound to the surface in real time. Using this technique, one can obtain the binding and dissociation rate constants and thus directly determine the ligand-receptor dissociation and affinity constants.

In vitro immunogenicity testing of hG-CSF conjugates

The reduced immunogenicity of the conjugates of the invention is determined by ELISA methods that detect the immunogenicity of the conjugates relative to a reference molecule or preparation. Typically, the reference molecule or preparation is a recombinant human G-CSF preparation such as Neupogen * or another recombinant human G-CSF preparation such as an N-terminally pegylated rhG-CSF molecule described in US5,824,784. The ELISA method was established on the basis of antibodies from patients treated with one of the recombinant G-CSF preparations described above. Immunogenicity is considered to be reduced when the response of the conjugate of the invention in the assay is statistically significantly reduced compared to the reference molecule or preparation.

Neutralization of Activity in G-CSF bioassays

The neutralizing effect of anti-G-CSF serum on hG-CSF conjugates was analyzed using the G-CSF bioanalytical method described above.

Sera from patients treated with G-CSF reference molecules or from animals that have been immunized are used. The above sera were added at a fixed concentration (1: 20-1: 500 dilution (patient serum) or 20-1000ng/ml (animal serum)) or in 5-fold serial dilutions starting from 1: 20 (patient serum) or 1000ng/ml (animal serum). HG-CSF conjugate was added in a total amount of 80. mu.l DMEM medium + 10% FCS at a concentration 7-fold diluted starting from 10nM or at a fixed concentration (1-100 pM). The serum was incubated with hG-CSF conjugate for 1 hour at 37 ℃.

The samples (0.01ml) were then transferred to 0.1ml DMEM medium containing NFS-60 cells in 96 well tissue culture plates. Culture at 5% CO₂Was incubated at 37 ℃ for 48 hours under an air atmosphere. 0.01ml of WST-1(WST-1 cell proliferation agent Roche Diagnostics GmbH, Mannheim, Germany) was then added to the culture and incubated at 37 ℃ for 150 minutes in an air atmosphere of 5% CO 2. The decomposition of tetrazolium salt WST-1 by mitochondrial dehydrogenase in living cells results in the formation of formazan, and this product is quantified by measuring absorbance at 450 nm.

When samples of hG-CSF conjugate were titrated in the presence of a fixed amount of serum, the neutralization effect was defined as Fold Inhibition (FI) and was measured as EC50 (with serum)/EC 50 (without serum). The reduction of antibody neutralization of G-CSF variant proteins is defined as

Example 1

Construction and cloning of synthetic Gene encoding hG-CSF

The following DNA fragments were synthesized according to the general procedure described by Stemmer et al, (1995), Gene 164, pp.49-53:

fragment 1, consisting of the following sequence: a Bam HI digestion site, a sequence encoding a YAP3 signal peptide (WO98/32867), a sequence encoding a TA57 leader sequence (WO98/32867), a sequence encoding a KEX2 protease recognition site (AAAAAAGA), a sequence encoding hG-CSF (SEQ ID NO: 2, whose codon usage characteristics facilitate expression in E.coli), and an Xba I digestion site.

Fragment 2, consisting of the following sequence: bam HI digestion site, the sequence encoding the YAP3 signal peptide (WO98/32867), a sequence encoding the TA57 leader sequence (WO98/32867), the sequence encoding the histidine marker (SEQ ID NO: 5), the sequence encoding the KEX2 protease recognition site (AAAAGA), the sequence encoding hG-CSF (SEQ ID NO: 2, whose codon usage characteristics facilitate expression in E.coli) and the Xba I digestion site.

Fragment 3, consisting of the following sequence: nde I digestion site, OmpA signal peptide-encoding sequence (SEQ ID NO: 3), hG-CSF-encoding sequence (SEQ ID NO: 2, codon usage characteristics favorable for expression in E.coli), and Bam HI digestion site.

Fragment 4, consisting of the following sequence: bam HI digestion sites, Kozak consensus sequence (Kozak, M.JMol Biol 1987 Aug 20; 196 (4): 947-50), sequence encoding the signal peptide of hG-CSF (SEQ ID NO: 7) and sequence encoding hG-CSF (SEQ ID NO: 8, whose codon usage characteristics favor expression in CHO) and Xba I digestion sites.

The DNA fragments 1 and 2 were inserted into the BamHI and Xba I digestion sites in plasmid pJSO37(Okkels, Ann. New York Acad. Sci.782: 202-207, 1996) using standard DNA techniques. This resulted in the plasmids pG-CSFcerevisae and pHISG-CSFcerevisae.

The DNA fragment 3 was inserted into the Nde I and Bam HI digestion sites in plasmid pET12a (Invitrogen) using standard DNA techniques. This gave rise to the plasmid pG-CSFcoli.

The DNA fragment 4 was inserted into the BamHI and Xba I digestion sites of plasmid pcDNA3.1(+) (Invitrogen) using standard DNA techniques. This gave plasmid pG-CSFCHO.

Example 2

Expression of hG-CSF in Saccharomyces cerevisiae and E.coli (E.coli)

Saccharomyces cerevisiae YNG318 (obtainable from American type culture Collection, VA, USA under the accession number ATCC 208973) was transformed with the plasmid pG-CSFcerevisiae or pHISG-CSFcerevisiae, transformants containing either of these two plasmids were isolated, and hG-CSF carrying and not carrying the HIS marker, respectively, was expressed extracellularly, each according to standard techniques described in the literature. Coli BL21(DE3) (Novagen, Cat. No.69387-3) was transformed with pG-CSFcoli, a transformant containing the plasmid was isolated, and hG-CSF was then expressed in the supernatant and the periplasm of the cells, according to the procedure described in pETSystem Manual (8th edition) of Novagen.

Expression of hG-CSF by s.cerevisiae and E.coli can be examined by Western blot analysis using the ImmunoPureultra-Sensitive ABC rabbit IgG staining kit (Pierce) and polyclonal anti-hG-CSF antibody (Pepro Tech EC Ltd.). The protein was found to be of the correct size.

The expression level of hG-CSF with and without an N-terminal histidine tag in Saccharomyces cerevisiae and Escherichia coli can be determined using commercially available G-CSF-specific ELISA kit (QuantikineHuman G-CSF Immunoassay, R)&D Systems cat. No. dcs50) were used for quantification. The values detected are listed below.

Expression system	Expression level (mg G-CSF/L)
		hG-CSF in Saccharomyces cerevisiae	30
hG-CSF with histidine tag in Saccharomyces cerevisiae	25
		hG-CSF in E.coli	0.05

Example 3

Production of Stable CHO-KI G-CSF producers

The day before transfection, the CHO K1 cell line (ATCC # CC1-61) was inoculated into 5ml of DMEM/F-12 medium (Gibco # 31330-038) supplemented with 10% FBS and penicillin/streptomycin in a T-25 flask. One day later (when almost 100% of the confluency), preparations for transfection were started: 90 μ l of DMEM medium without additives was aliquoted into 14ml polystyrene tubes (Corning). Mu.l Fugene 6(Roche) was added directly to the medium and incubated for 5min at room temperature. At the same time, 5. mu.g of plasmid pG-CSFCHO were aliquoted into another 14ml polystyrene tube. After incubation, the Fugene 6 mixture was added directly to the DNA solution and incubated for 15min at room temperature. All liquid was then added dropwise to the cell culture medium.

After one day, the medium was changed to fresh medium containing 360. mu.g/ml hygromycin (Gibco). Selection medium was refreshed daily thereafter until the primary transfection pool reached 100% confluence. The primary transfection pool was recloned by limiting dilution (300 cells were plated into 5 96-well plates).

Example 4

Purification of hG-CSF and variants thereof from Saccharomyces cerevisiae culture supernatant

Purification of hG-CSF was performed as follows:

cells were removed by centrifugation. The cell-depleted supernatant was then sterilized by filtration through a 0.22 μm filter. The supernatant after filter sterilization was diluted 5-fold in 10mM sodium acetate, pH 4.5. The pH of the diluted supernatant was adjusted by adding 10ml of concentrated acetic acid to 5 liters of the supernatant. The ionic strength of the solution should be below 8mS/cm before application to the cation exchange column.

The diluted supernatant was loaded at a linear flow rate of 90cm/h onto a SP-sepharose FF (Pharmacia) column which had been equilibrated with 50mM sodium acetate, pH4.5 until the eluate from the column reached a stable UV and baseline conductivity level. The column is washed with the equilibration buffer until the UV absorbance and conductivity of the column effluent reach a plateau, thereby removing unbound material. Using a linear gradient; 30 column volumes; 0-80% buffer B (50mM NaAc, pH4.5, 750mM NaCl) eluted the bound G-CSF protein from the column at a flow rate of 45 cm/h. Fractions containing G-CSF were pooled according to the results of SDS polyacrylamide gel electrophoresis. Sodium chloride is added until the ionic strength of the solution is above 80 mS/cm.

The protein solution was loaded onto a Phenyl Toyo Pearl 650S column equilibrated with 50mM NaAc, pH4.5, 750mM NaCl. Unbound material was washed off with the equilibration buffer. Elution of G-CSF was performed by applying a stepwise gradient of MilliQ water. The fractions containing G-CSF were pooled. By applying the above strategy of 2-step flushing (down stream), G-CSF of more than 90% purity can be obtained. The purified protein is then quantified by spectrophotometric detection at 280nm and/or by amino acid analysis.

The fractions containing G-CSF were pooled. Buffer exchange and concentration were performed with a Vivaspin concentrator (mwco: 5 kDa).

Example 5

Identification and quantification of unconjugated and conjugated hG-CSF and variants thereof

SDS Polyacrylamide gel electrophoresis

The purified and concentrated G-CSF was analyzed by SDS-PAGE. A single band with an apparent molecular weight of about 17kDa is clearly visible.

Absorbance of the solution

The G-CSF concentration was assessed by spectrophotometry. The concentration of the protein can be calculated by measuring the absorbance at 280nm and applying a theoretical extinction coefficient of 0.83.

Amino acid analysis

The protein is more accurately determined by amino acid analysis. Amino acid analysis of purified G-CSF revealed that the experimentally determined amino acid composition was consistent with the amino acid composition predicted from the DNA sequence.

Example 6

MALDI-TOF mass spectrometry of PEGylated wt G-CSF and G-CSF variants

The number of PEG groups attached to the PEGylated wt G-CSF and the selected PEGylated G-CSF variant was assessed by MALDI-TOF mass spectrometry.

Wt G-CSF contains 5 primary amino groups predicted to be SPA-PEG attachment sites (N-terminal amino groups and epsilon amino groups on K16, K23, K34 and K40). After PEGylation of wt G-CSF with SPA-PEG5000, MALDI-TOF mass spectrometry showed that wtG-CSF variants with 4,5 and 6 PEG groups attached were predominantly present, and furthermore, wt G-CSF variants with 7 PEG groups attached were clearly visible, but in lesser amounts.

G-CSF variants with K16R, K34R, K40R, Q70K, Q90K, and Q120K substitutions also contain 5 primary amino groups (N-terminal amino group and epsilon amino groups on K23, K70, K90, and K120). After SPA-PEG5000 PEGylation of this G-CSF variant, MALDI-TOF mass spectrometry showed that there were mainly G-CSF variants with 4,5 and 6 PEG groups attached, and furthermore, the G-CSF variants with 7 PEG groups attached were clearly visible, but in smaller amounts.

G-CSF with K16R, K34R, and K40R substitutions contains 2 primary amino groups (N-terminal amino group and epsilon amino group on K23). After PEGylation of the G-CSF variant with SPA-PEG 12000, MALDI-TOF mass spectrometry showed that there were mainly G-CSF variants with 2 and 3 PEG groups attached, and furthermore, the G-CSF variants with 4 PEG groups attached were clearly visible but in smaller amounts.

The above observations clearly show that in addition to amino acid residues comprising an amino group, other amino acid residues may sometimes be pegylated under the applied pegylation conditions. The above observations also show that pegylation at the site of amino group introduction is of some importance. This has also been observed in SDS-PAGE analysis of wt G-CSF and G-CSF variants.

Histidine 170 was fully pegylated when SPA-PEG chemistry was applied, as described in example 11. In addition, K23 and S159 were partially pegylated. This explains that there are 1-2 additional PEGylation sites in addition to the primary amino groups already formed in hG-CSF and variants thereof.

Example 7

Peptide mapping of pegylated and non-pegylated G-CSF variants

To map additional SPA-PEG attachment sites on G-CSF and G-CSF variants, the following strategy was applied.

To minimize the number of expected pegylation sites, G-CSF variants with a small number of amino groups were selected. Selected G-CSF variants have K16R, K34R, K40R and H170Q substitutions. This variant contains only one N-terminal primary amine, except that the previous data on K23 show no degree of PEGylation of the epsilon amino group. Thus, the pegylation background on the amine group was significantly reduced in this G-CSF variant. The G-CSF variants were PEGylated using SPA-PEG 5000. After pegylation, the G-CSF variant is denatured, its disulfide bond reduced, the resulting sulfhydryl groups alkylated, and the hexaalkylated and pegylated protein then degraded with a glutamate specific protease. Finally, the formed peptides were separated by reverse phase HPLC.

Non-pegylated G-CSF variants with K16R, K34R, and K40R substitutions were similarly treated in parallel to obtain an HPLC reference chromatogram.

HPLC chromatographic comparison of the degradation of the PEGylated and non-PEGylated G-CSF variants should reveal which peptide disappeared upon PEGylation. Identification of the peptides of the non-pegylated G-CSF variants by N-terminal amino acid sequencing of these peptides indirectly points to the position of pegylation.

In principle, it is preferred to use non-pegylated G-CSF variants with all these substitutions K16R, K34R, K40R and H170Q, but this is not necessary in practice.

More specifically, approximately 1mg of pegylated G-CSF variant with substitutions K16R, K34R, K40R and H170Q and approximately 500 μ G of non-pegylated G-CSF variant with substitutions K16R, K34R, and K40R were dried in a SpeedVac concentrator. Both samples were dissolved in 400. mu.l of 6M guanidine, 0.3M Tris HCl, pH8.3, respectively, and then denatured at 37 ℃ overnight. After denaturation, the disulfide bonds in the protein were reduced by adding 50. mu.l of 0.1M DTT in 6M guanidine, 0.3M Tris-HCl, pH8.3 solution. After 2 hours of incubation at room temperature, the original thiol group was alkylated by the addition of 50. mu.l of 0.6M iodoacetamide in 6M guanidine, 0.3M Tris-HCl, pH 8.3. Alkylation was carried out at room temperature for 30min, and then the buffer of the reduced and alkylated protein was exchanged for 50mM NH using NAP5 column₄HCO₃. After the volume of the sample was reduced to greater than 200. mu.l in the speedVac concentrator, 20. mu.g and 10. mu.g of glutamate-specific protease were added, respectively. The degradation by glutamate specific protease was carried out at 37 ℃ for 16 hours. The resulting peptides were separated by reverse phase HPLC using a Phenomenex Jupiter C18 column (0.2 x 5cm) eluted with a linear acetonitrile gradient in 0.1% aqueous TFA. By MALDI-TOF mass spectrometry analyzes the collected fractions, followed by N-terminal amino acid sequence analysis of the selected peptides.

Comparison of the HPLC chromatograms of the degradation of the PEGylated and non-PEGylated G-CSF variants revealed that only two components disappeared upon PEGylation. N-terminal amino acid sequence analysis of these two components of the non-PEGylated G-CSF variant showed that both peptides were derived from the N-terminus of G-CSF. One of the peptides consists of amino acid residues 1-11 resulting from the accidental cleavage of residues after Gln 11. The other peptide consists of amino acid residues 1-19 resulting from the expected cleavage of the residue after Glu 19.

It is expected that the N-terminal peptide of G-CSF can be identified by this method because the N-terminal amino group is easily PEGylated. However, no other attachment sites for SPA-PEG5000 were identified with this method.

An alternative to indirectly identifying the PEG5000 attachment site is to directly identify the attachment site in the pegylated peptide. However, upon HPLC separation of the degraded pegylated G-CSF variant, the components comprising the pegylated peptide were not separated from each other and failed to separate from several components comprising non-pegylated peptide. Therefore, analysis of the N-terminal amino acid sequences of these components did not yield any useful data except that K23 could be partially PEGylated.

To overcome these problems, two pools of pegylated peptides were prepared from the fractions of the first HPLC separation. The two pools were dried in a SpeedVac concentrator and dissolved in 200. mu.l of freshly prepared 50mM NH₄HCO₃Neutralized and further degraded with 1. mu.g chymotrypsin. The resulting peptides were separated by reverse phase HPLC using a Phenomenex Jupiter C18 column (0.2 x 5cm) eluted with a linear acetonitrile gradient in 0.1% aqueous TFA. The collected fractions were analyzed by MALDI-TOF mass spectrometry, followed by N-terminal amino acid sequence analysis of the selected peptide.

Both K23 and S159 were partially pegylated as determined by the N-terminal amino acid sequence. The actual degree of pegylation at these two positions could not be determined, but since unmodified peptides of K23 and S159 were identified from the initial HPLC separation and sequenced, it was determined that pegylation was only partial.

Example 8

Glycosylation of wt G-CSF and G-CSF variants

When the purified wt G-CSF and G-CSF variants are analyzed by MALDI-TOF mass spectrometry, another fraction is always observed with a mass of about 324Da greater than the mass of the G-CSF molecule analyzed. Since the component with the smallest mass always has the mass of the G-CSF molecule and since the G-CSF molecule has the correct N-terminal amino acid sequence, it can be concluded that the above-mentioned further component is a modified G-CSF molecule carrying two hexose residues. In many cases, the unmodified G-CSF molecule produces the strongest signal but in some cases the signal strength of the modified G-CSF molecule is the strongest.

Peptides prepared to identify additional pegylation sites were analyzed to identify two peptides from each degradation that could be used to identify glycosylation sites.

In both HPLC separations, the two peptides were eluted sequentially and mass difference of about 324Da was shown by MALDI-TOF mass spectrometry. Mass spectrometric data suggest that the peptide spans amino acid residues 124-162. N-terminal amino acid sequence analysis of all 4 peptides showed that this assignment was correct and Thr133 was the only modification site. Thr133 is clearly visible in the sequence in the peptide with the mass of the unmodified peptide, but no amino acid residue is arranged at position 133 in the peptide with the additional mass of 324 Da. Since all other amino acid residues can be arranged in the sequence, it can be concluded that Thr133 is the only modification site. This glycosylation site has previously been reported for expression in recombinant G-CSF in CHO cells, but this glycan is different from the glycan linked in yeast.

Non-glycosylated wt G-CSF is separated from glycosylated wt G-CSF by reverse phase HPLC, in particular using Vydac C₁₈The column (0.21X 5cm) was isocratically eluted with 51% acetonitrile in 0.1% TFA to give a single columnOne fraction containing only non-glycosylated wt G-CSF was shown by MALDI-TOF mass spectrometry.

Example 9

Isolation of G-CSF molecules covalently linked to different numbers of PEG molecules

The G-CSF molecules covalently coupled to 4,5 or 6 PEG groups were isolated as follows. The SPsepharose FF column was equilibrated with 20mM sodium citrate, pH2.5, and then the PEGylated protein in20 mM sodium citrate, pH2.5 was loaded onto the SPsepharose FF column. Any unbound material is washed off the column. Elution was performed with a pH gradient. PEGylated G-CSF begins to elute from the column at about pH3.8 and continues to elute in the fractions from pH3.8 to 4.5.

These fractions were subjected to SDS-PAGE and mass spectrometry. These analyses indicate that the most pegylated G-CSF is present in the "low pH fraction". The less pegylated G-CSF elutes in the "high pH fraction".

Amino acid analysis of the PEGylated G-CSF showed that the theoretical value of the extinction coefficient was very consistent with the experimentally determined value.

Example 10

Construction of hG-CSF variants

Specific substitutions of amino acids in hG-CSF to other amino acid residues, such as those discussed in the general description above, are introduced using standard DNA techniques known in the art. Novel G-CSF variants can be prepared using plasmid pG-CSFcerevisae, which contains the hG-CSF encoding gene but does not carry the HIS marker, as a DNA template in PCR reactions. These variants were expressed in Saccharomyces cerevisiae and purified as described in example 4. Some of the G-CSF variants constructed are listed below (see examples 12 and 13).

Example 11

Covalently coupling SPA-PEG to hG-CSF or variants thereof

Human G-CSF and variants thereof were covalently linked to SPA-PEG5000, SPA-PEG 12000 and SPA-PEG 20000(Shearwater) as described above ("PEGylation of hG-CSF and variants thereof in solution"). The in vitro activity of the conjugates is exemplified in example 13.

Example 12

Identification of SPA-PEG in G-CSF by site-directed mutagenesis and subsequent PEGylation of purified variants Attachment site

SPA-PEG can be attached to amino acid residues other than lysine in G-CSF. To determine whether SPA-PEG is linked to histidine, serine, threonine and arginine, variants were prepared in which these amino acids were substituted with lysine, alanine or glutamic acid. These variants were expressed in Saccharomyces cerevisiae, purified and PEGylated, and then the number of attached SPA-PEG molecules was analyzed on SDS-PAGE. The analysis was done by visual inspection of the SDS-PAGE gels, all containing 3 major bands. From the relative sizes of these bands, the degree of pegylation was estimated for each band with a deviation of closest 5%. The decrease in the amount of attached SPA-PEG after a given amino acid was substituted with glutamic acid or alanine strongly suggests that the amino acid was PEGylated by SPA-PEG and this phenomenon is further supported by the result that the degree of PEGylation does not change after the amino acid was substituted with lysine. The variants analyzed are listed below.

G-CSF variants	Number of PEG groups attached
		hG-CSF	10% 4 PEG, 75% 5 PEG, 15% 6 PEG
K23R	10％4 PEG, 85% 5 PEG, 5% 6 PEG
		H43Q	10% 4 PEG, 75% 5 PEG, 15% 6 PEG
H43K	10% 5 PEG, 75% 6 PEG, 15% 7 PEG
		H52Q	10% 4 PEG, 75% 5 PEG, 15% 6 PEG
H52K	10% 5 PEG, 75% 6 PEG, 15% 7 PEG
		H156Q	10% 4 PEG, 75% 5 PEG, 15% 6 PEG
H156K	10% 5 PEG, 75% 6 PEG, 15% 7 PEG
		H170Q	10% 3 PEG, 75% 4 PEG, 15% 5 PEG
H170K	10% 4 PEG, 75% 5 PEG, 15% 6 PEG
		K16/34R	10% 2 PEG, 75% 3 PEG, 15% 4 PEG
K16/34R R22K	10% 3 PEG, 75% 4 PEG, 15% 5 PEG
		K16/34R R22Q	10% 2 PEG, 75% 3 PEG, 15% 4 PEG
K16/34R S124A	10% 2 PEG, 75% 3 PEG, 15% 4 PEG
		K16/34/40R	10% 1 PEG, 75% 2 PEG, 15% 3 PEG
K16/34/40R S53K	10% 2 PEG, 75% 3 PEG, 15% 4 PEG
		K16/34/40R S53A	10% 2 PEG, 75% 3 PEG, 15% 4 PEG
K16/34/40R S62K	10% 2 PEG, 75% 3 PEG, 15% 4 PEG
		K16/34/40R S66K	10% 2 PEG, 75% 3 PEG, 15% 4 PEG
K16/34/40R S80K	10% 2 PEG, 75% 3 PEG, 15% 4 PEG

K16/34/40R T105K	10% 2 PEG, 75% 3 PEG, 15% 4 PEG
		K16/34/40R T133K	10% 2 PEG, 75% 3 PEG, 15% 4 PEG
K16/34/40R S142K	10% 2 PEG, 75% 3 PEG, 15% 4 PEG
		K16/34/40R R147K	10% 2 PEG, 75% 3 PEG, 15% 4 PEG
K16/34/40R S155K	10% 2 PEG, 75% 3 PEG, 15% 4 PEG
		K16/34/40R S159K	10% 2 PEG, 85% 3 PEG, 5% 4 PEG
K16/34/40R S170K	10% 1 PEG, 75% 2 PEG, 15% 3 PEG

The above data show that SPA-PEG can also be covalently bound to H170 in addition to the N-terminus, K16, K34, and K40. Further, the data show that only 10% of the available K23 amino acid residues are pegylated, and about 10% of S159 is pegylated.

Example 13

In vitro biological Activity of unconjugated and conjugated hG-CSF and variants thereof

The in vitro biological activity of unconjugated and conjugated hG-CSF and variants thereof were tested as described above under "preliminary test 2-in vitro assay for hG-CSF activity". In vitro biology of each variant produced for SPA-PEG5000 conjugation and uncoupling to available PEGylation sitesThe activities (expressed as EC50 measurements) are listed below. These values have been normalized to unconjugated hG-CSF (Neupogen *), i.e.the values in the table represent% activity relative to unconjugated hG-CSF. This value and the value of the variant were each time measured under the same measurement conditions. The EC50 value for hG-CSF in the assay was 30 pM.

G-CSF variants	EC50 (% hG-CSF) unconjugated form	EC50 (% hG-CSF) coupled with SPA-PEG5000
			G-CSF with N-terminal histidine tag	10	Not testing
G-CSF without N-terminal histidine tag	100	0.1
			16R	100	1
16Q	80	1
			23Q	80	0.1
23R	100	0.1
			34R	100	1
34A	80	1
			34Q	70	1
40R	50	1
			K16/23R	100	1
K16/23Q	/0	1
			K34/40R	50	5
K16/34R	100	10

K16/40R	50	5
			K16/23/34R	50	10
K16/23/40R	50	5
			K16/34/40R	35	30
K16/23/34/40R	20	15
			K16/34/40R L3K	50	25
K16/34/40R E45K	Low level	Not testing
			K16/34/40R E46K	10	1
K16/34/40R S53K	5	0.5
			K16/34/40R S62K	10	0.5
K16/34/40R S66K	20	2
			K16/34/40R Q67K	10	0.2
K16/34/40R Q70K	30	20
			K16/34/40R S76	50	20
K16/34/40R Q77	1	0
			K16/34/40R S80K	10	0.2
K16/34/40R Q90K	30	20
			K16/34/40R E98K	Low level	Not testing
K16/34/40R D104K	10	0.9
			K16/34/40R T105K	30	10
K16/34/40R Q120K	30	20
			K16/34/40R Q131K	Low level	Not testing
K16/34/40R T133K	30	10
			K16/34/40R Q134K	10	0.2
K16/34/40R S142K	20	7
			K16/34/40R R147K	20	1
K16/34/40R S155K	20	1
			K16/34/40R Q158	20	5
K16/34/40R S159K	20	3
			K16/34/40R Q70K Q90K	Not testing	20
K16/34/40R Q70K Q120K	25	25
			K16/34/40R Q90K T105K	40	10
K16/34/40R Q90K Q120K	25	15
			K16/34/40R Q90K S159K	45	Not testing
K16/34/40R T105K Q120K	20	8
			K16/34/40R T105K S159K	40	20
K16/34/40R Q120K T133K	20	8
			K16/34/40R Q120K S142K	10	2
K16/34/40R Q70K Q90K T105K	10	4

K16/34/40R Q70K Q90K Q120K	20	12
			K16/34/40R Q70K Q90K T133K	15	5
K16/34/40R Q70K T105K Q120K	10	2
			K16/34/40R Q70K Q120K T133K	15	2
K16/34/40R Q70K Q120K S142K	10	1
			K16/34/40R Q90K T105K Q120K	10	2
K16/34/40R Q90K T105K T133K	10	2
			K16/34/40R Q90K T105K S159K	55	5
K16/34/40R Q90K Q120K T133K	15	2
			K16/34/40R Q90K Q120K S142K	10	1
K16/34/40R T105K Q120K T133K	10	1
			K16/34/40R Q120K T133K S142K	10	1

The above data show that substitution of K23 to arginine does not increase the activity of the conjugated protein. This is due to the fact that only 10% of K23 was pegylated, and that the coupled K23R variant has essentially the same number of PEG group attachments and the same pegylation site with hG-CSF. Removal of the lysines at positions K16, K34 and K40 produced G-CSF variants with significant activity after pegylation. This variant did not significantly reduce activity after conjugation to SPA-PEG5000 compared to the unconjugated variant. Thus, PEGylation of the N-terminus and H170 with SPA-PEG5000 (see example 12) did not reduce the activity of hG-CSF. It was therefore decided to use hG-CSF K16R K34R K40R as backbone for insertion of new PEGylation sites. In this backbone, 24 new pegylation sites were introduced between residues L3 and H159. These residues are distributed in the hG-CSF at sites that do not interact with the G-CSF receptor. The introduction of new PEGylation sites at the L3, Q70, S76, Q90, T105, Q120, T133 and S142 positions resulted in hG-CSF variants that remained significantly active after PEGylation with SPA-PEG 5000. Therefore, some of these new pegylation sites were combined into hG-CSF variants with 2 or 3 new pegylation sites.

In addition, SPA-PEG 12000 and SPA-PEG 20000 can be linked to a selected group of hG-CSF variants. In vitro activity is listed below (Neupogen *%).

G-CSF variants	EC50 (% of hG-CSF) coupled to SPA-PEG 12000	EC50 (% of hG-CSF) coupled to SPA-PEG 20000
			K16/34/40R	10	1
K16/34/40R Q90K	Not testing	7
			K16/34/40R Q70K Q90K	8	Not testing
K16/34/40R Q90K T105K	1	＜1
			K16/34/40R T105K S159K	6	5
K16/34/40R Q90K T105K S159K	1	＜1

Example 14

In vivo half-life of unconjugated and conjugated hG-CSF and variants thereof

The in vivo half-life of unconjugated hG-CSF (Neupogen *), SPA-PEG 5000-conjugated hG-CSF K16RK34R K40R Q70K Q90K Q120K, and SPA-PEG 5000-conjugated hG-CSF K16RK34R K40R Q90K T105K S159K were determined as described above (determination of the in vivo half-life of unconjugated and conjugated hG-CSF and variants thereof). The results are shown in FIGS. 1 and 2. The in vivo half-life of Neupogen * was determined to be 2.01 hours and 1.40 hours, respectively. In an earlier similar experiment (U.S. Pat. No. 5,824,778), the in vivo half-life of hG-CSF was measured to be 1.79 hours. The conclusions of the experiments described herein can therefore be directly compared to this experiment. The in vivo half-lives of SPA-PEG5000 coupled hG-CSF K16R K34R K40R Q70K Q90K Q120K and SPA-PEG5000 coupled hG-CSF K16R K34R K40R Q90K T105K S159K were determined to be 12.15 hours and 16.10 hours, respectively. Thus, the introduction of new pegylation sites in hG-CSF and their coupling to SPA-PEG5000 both resulted in a significant increase in vivo half-life.

In the earlier experiments described above (U.S. Pat. No. 5,824,778), hG-CSF coupled to a larger N-terminally conjugated PEG molecule (10kDa) was determined to have an in vivo half-life of 7.05 hours. Thus, SPA-PEG5000 coupled hG-CSF K16R K34R K40R Q70K Q90K Q120K and SPA-PEG5000 coupled hG-CSF K16R K34R K40R Q90K T105K S159K have significantly increased in vivo half-life compared to Neupogen * and hG-CSF with a 10kDa N-terminally coupled PEG molecule. The SPA-PEG5000 coupled hG-CSF K16R K34R K40R Q70K Q90KQ120K and the SPA-PEG5000 coupled hG-CSF K16R K34R K40R Q90K T105KS159K all have three endogenous PEGylation sites removed and three new PEGylation sites introduced, so the sizes are the same. The only difference between these two compounds was in vitro activity, 12% and 5% of Neupogen *, respectively. This difference resulted in a longer in vivo half-life for SPA-PEG5000 coupled hG-CSF K16R K34RK40R Q90K T105K S159K than for SPA-PEG5000 coupled hG-CSF K16R K34RK40R Q70K Q90K Q120K. Since the in vitro activity is related to the receptor binding affinity of these compounds, it can be concluded that: the receptor-mediated clearance of SPA-PEG5000 coupled hG-CSF K16R K34R K40R Q90K T105K S159K was slower than SPA-PEG5000 coupled hG-CSF K16R K34R K40R Q70K Q90K Q120K. Thus, increasing the size of G-CSF and decreasing its in vitro activity, combined with the previously known G-CSF compounds, results in G-CSF compounds having a significantly longer in vivo half-life than previously known G-CSF compounds.

Example 15

In vivo biological Activity of unconjugated and conjugated hG-CSF and variants thereof in healthy rats

Uncoupled hG-CSF (Neupogen *), SPA-PEG 5000-coupled hG-CSF K16RK34R K40R Q70K Q120K, SPA-PEG 5000-coupled hG-CSF K16K 34K 40RQ70K Q90Q 120K, SPA-PEG 5000-coupled hG-CSF K16R K34K 2K 40R Q70KQ120K T133K, and SPA-PEG 5000-coupled hG-CSF K16R K34R K R Q90KQ 120T 133K, whose in vivo biological activity was assayed as described above (determination of in vivo biological activity of uncoupled and coupled hG-CSF and variants thereof in CSF healthy rats). The results are shown in FIGS. 3 and 4.

No Neupogen * activity was detected at 48 hours after injection of 100 μ g per kg body weight at t 0 hours. After the initial injection, the activity of SPA-PEG5000 coupled hG-CSFK16R K34R K40R Q70K Q120K and SPA-PEG5000 coupled hG-CSF K16R K34RK40R Q90K Q120K T133K and SPA-PEG5000 coupled hG-CSF K16R K R K40RQ70K Q120K T133K could be detected up to 72 hours, while the in vivo activity of SPA-PEG5000 coupled hG-CSFK16R K34R K40Q 70K Q90Q K Q120K was retained up to 96 hours. It can be seen that these conjugated variants have a longer in vivo biological activity than neupogon *, and SPA-PEG5000 conjugated hG-CSF K16R K34R K40R Q70K Q90K Q120K maintained in vivo activity twice as long as neupogon *. SPA-PEG5000 coupled hG-CSF K16R K34R K40R Q70KQ120K T133K and SPA-PEG5000 coupled hG-CSF K16R K34R K40R Q90KQ120K T133K, 2% of Neupogen * (example 13) were active in vitro and did not induce the same level of leukocyte formation seen with Neupogen *, SPA-PEG5000 coupled hG-CSF K16R K34R K40R Q70K Q120K, SPA-PEG5000 coupled hG-CSF K16R K34R K40R Q70K Q120K, SPA-PEG5000 coupled hG-CSFK16R K34R K40R Q70K Q90K Q120K administration within the first 12 hours after administration. Thus, compounds with only 2% in vitro activity or less of Neupogen * were unable to complete stimulation of leukogenesis immediately after administration.

In addition, the in vivo biological activity of Neupogen *, SPA-PEG 12000 coupled hG-CSFK16R K34R K40R and various doses of SPA-PEG5000 coupled hG-CSF K16R K34RK40R Q70K Q90K Q120K were tested as described above ("in vivo biological activity assay for unconjugated and conjugated hG-CSF and variants thereof in healthy rats"). The results are shown in FIG. 5. As was observed earlier, no Neupogen * activity was detected 48 hours after the initial injection at 100. mu.g per kg body weight. Administration of 5 μ g of SPA-PEG5000 conjugated hG-CSF K16RK34R K40R Q70K Q90K Q120K per kg body weight resulted in a slight prolongation of in vivo biological activity compared to Neupogen *, whereas administration of 25 μ g and 100 μ g of the above compounds per kg body weight resulted in hG-CSF activity remaining active up to 72 and 96 hours, respectively, after the initial injection. Thus, the duration of action of the SPA-PEG-conjugated hG-CSF compound can be controlled by increasing or decreasing the standard dosage regimen. SPA-PEG 12000 coupled hG-CSF K16R K34R K40R maintained in vivo activity up to 72 hours after administration of 100. mu.g/kg body weight. As described in example 6, SPA-PEG 12000-coupled hG-CSF K16RK34R K40R is coupled with 2 or 3 SPA-PEG 12000 groups, while SPA-PEG 5000-coupled hG-CSF K16R K34R K40R Q70K Q90K Q120K is coupled with 5 or 6 SPA-PEG5000 groups. Therefore, the molecular weights of the two compounds are 42-54kDa and 43-48kDa, respectively. The in vitro activity of these two compounds was 30% and 12% of Neupogen *, respectively. The in vivo biological activity of SPA-PEG5000 coupled hG-CSF K16R K34R K40R Q70K Q90K Q120K was maintained for a longer period of time than SPA-PEG 12000 coupled hG-CSF K16R K34R K40R of substantially the same molecular weight, indicating that the effect of the G-CSF compound can be further extended only by decreasing its specific activity and thus its receptor-mediated clearance when its size exceeds a specific molecular weight value by PEGylation (example 14).

Furthermore, the in vivo biological activity of Neupogen *, SPA-PEG 5000-coupled hG-CSFK16R K34R K40R Q70K Q90K Q120K, SPA-PEG 5000-coupled hG-CSF K16RK34R K40R Q90K T105K S159K, and SPA-PEG 20000-coupled hG-CSF K16RK34R K40R T105K S159K were determined as described above (in vivo biological activity of unconjugated and conjugated hG-CSF and variants thereof were determined in healthy rats). The results are shown in FIG. 6.

As observed earlier, the effect time of the coupled hG-CSF variant was significantly longer than that of Neupogen *. Each of these three coupled hG-CSF variants, upon administration, resulted in the formation of leukocytes at the same rate and level as observed within the first 12 hours after administration of Neupogen *. SPA-PEG5000 coupled hG-CSF K16R K34R K40R Q70K Q90K Q120K, SPA-PEG5000 coupled hG-CSF K16R K34R K40R Q90K T105K S159K, and SPA-PEG 20000 coupled hG-CSF K16R K34R K40R T105K S159K, which have in vitro activities of 12%, 5% and 5% of Neupogen *, respectively, so that compounds with 5% in vitro activity of Neupogen * are able to adequately induce the formation of white blood cells after administration.

Neupogen *, SPA-PEG5000 coupled hG-CSF K16R K34R K40R Q70KQ90K Q120K, SPA-PEG5000 coupled hG-CSF K16R K34R K40R Q90K T105KS159K, SPA-PEG 20000 coupled hG-CSF K16R K34R K40R T105K S159K, which have apparent sizes of 18kDa, 60kDa, 60kDa and > 100kDa on SDS-PAGE, respectively. The duration of in vivo effect of SPA-PEG5000 coupled hG-CSF K16R K34R K40R Q90K T105K S159K and SPA-PEG 20000 coupled hG-CSF K16R K34R K40R T105K S159K were nearly identical, indicating that the duration of effect was not prolonged by increasing the molecular size of the coupled hG-CSF compound beyond about 60 kDa. Conversely, when the apparent size of the coupled hG-CSF compound exceeds about 60kDa, the increased duration of action may decrease the in vitro activity of the compound and thus decrease the receptor binding affinity. Another example of this (above) can be seen by comparing the in vivo duration of action between SPA-PEG5000 coupled hG-CSF K16R K34R K40R Q70K Q90K Q120K and SPA-PEG5000 coupled hG-CSF K16R K34R K40R Q90K T105K S159K. Both compounds have an apparent size of 60kDa, but the in vitro activity was 12% and 5%, respectively. This difference is directly reflected in the in vivo maintenance time of the action of the two compounds, which is 96 hours and 144 hours, respectively.

Example 16

Uncoupling and coupling in rats with chemotherapy-induced neutropenia In vivo biological Activity of hG-CSF and variants thereof

In vivo biological activity of uncoupled hG-CSF (Neupogen *), SPA-PEG 5000-coupled hG-CSF K16R K34R K40R Q70K Q90KT105K and SPA-PEG 20000-coupled hG-CSF K16R K34R K40R Q90K in rats with chemotherapy-induced neutropenia was measured with CPA at 50mg/kg body weight and a single dose (100. mu.g/kg body weight) of G-CSF as described above (in vivo biological activity of uncoupled and coupled hG-CSF and variants thereof was determined in rats with chemotherapy-induced neutropenia). The results are shown in FIG. 7. The three compounds induced the initial formation of leukocytes at the same rate. It can be seen that only 4% of the in vitro activity of Neupogen * was required for the conjugated hG-CSF compound to adequately stimulate leukocyte formation immediately after administration. After 36 hours, the White Blood Cell (WBC) count in Neupogen * -treated rats decreased to the level of the untreated group (< 3X 10)⁹Cells/liter). At this time, the rats exhibited neutropenia. After 144 hours, WBC reached normal level (9X 10) in both groups⁹Cells/liter).

In the SPA-PEG5000 coupled hG-CSF K16R K34R K40R Q70K Q90K T105K treated group and the SPA-PEG 20000 coupled hG-CSF K16R K34R K40RQ90K treated group, the WBC level is reduced to the minimum value of 4X 10 after 48 hours⁹Cells/liter, then quickly begin to rise back. After 96 hours, WBC levels returned to normal in both groups. Thus, the two coupled hG-CSF compounds were able to alleviate neutropenia and significantly reduce the time required for WBC to return to normal levels(duration of neutropenia), specifically a significant reduction from 112 hours in the Neupogen * -treated group to 48 hours in the SPA-PEG5000 coupled type hG-CSF K16R K34R K40RQ70KQ90K T105K treated group or in the SPA-PEG 20000 coupled type hG-CSF K16R K34RK40RQ90K treated group.

The SPA-PEG5000 coupled hG-CSF K16R K34R K40R Q70K Q90K T105K is more effective in shortening the duration of neutropenia than the SPA-PEG 20000 coupled hG-CSF K16R K34R K40R Q90K. Since both molecules have apparent sizes above 60kDa (60 kDa and 80kDa, respectively), this cannot be explained by a lower renal clearance of SPA-PEG5000 coupled hG-CSF K16RK34R K40R Q70K Q90K T105K than of SPA PEG 20000 coupled hG-CSF K16RK34R K40R Q90K. The in vitro activity of SPA-PEG5000 coupled hG-CSF K16R K34RK40R Q70K Q90K T105K and SPA-PEG 20000 coupled hG-CSF K16R K34RK40R Q90K was 4% and 7% of Neupogen *, respectively. This means that, within the first 48 hours after administration, the SPA PEG 5000-coupled hG-CSF K16R K34R K40RQ70K Q90K T105K had lower receptor binding affinity, and therefore lower receptor-induced clearance, but increased leukocyte levels, than the SPA PEG 20000-coupled hG-CSF K16R K34R K40RQ 90K. Thus, when rats show neutropenia after 48 hours, the in vivo concentration of SPA-PEG 5000-coupled hG-CSF K16R K34R K40R Q70K Q90K T105K is higher than that of SPA-PEG 20000-coupled hG-CSF K16R K34R K40RQ 90K. Since only 4-5% of the in vitro G-CSF activity of Neupogen * was sufficient to fully activate G-CSF receptors on the surface of neutrophil progenitor cells (see above), this higher G-CSF concentration, which occurred after 48 hours, explains the faster increase in WBC levels in the SPA-PEG 5000-coupled hG-CSF K16R K34R K40R Q70KQ90K T105K-treated group. Thus, the conjugated G-CSF compounds of the invention having an apparent size of at least about 60kDa and an in vitro activity of 4% Neupogen * are superior and have a higher in vitro activity in rats with chemotherapy-induced neutropenia compared to compounds of similar size.

Example 17

Purification of G-CSF from Saccharomyces cerevisiae culture supernatant

This example provides an alternative method for purification of hG-CSF and G-CSF variants to that of example 4.

The cells were removed by centrifugation at 5000rpm at 4 ℃ for 10min and the clarified supernatant was filtered through a 0.22 μm filter. The clarified and filtered supernatant was concentrated and diafiltered by Tangential Flow Filtration (Tangential Flow Filtration) using a 10kDa membrane in 50mM sodium acetate, pH4.5

The resulting ultrafiltrate was loaded onto a SP-sepharose column (200ml packed bed) which had been equilibrated with at least 5 column volumes of 50mM sodium acetate. The sample was loaded at a flow rate of about 20 ml/min. The column was washed with the equilibration solution until a stable eluate (effluent) was obtained as determined by measuring the absorbance at 280 nm. G-CSF is eluted at ambient flow rate (ambient flow rate) in 35% buffer using stepwise buffer gradients (e.g.10%, 20%, 30% and 35% buffer) of 750mM NaCI in 50mM sodium acetate.

This one-step procedure yields G-CSF of > 95% purity (as determined by SDS-PAGE).

Example 18

Isolation of multiple-PEGylated G-CSF

Example 9 above describes a method for isolating G-CSF molecules linked to different numbers of PEG groups. This example provides an alternative method of isolating such multiply PEGylated G-CSF in order to obtain a G-CSF product having a desired degree of consistency in the number of PEG groups attached.

The mixture of PEGylated G-CSF described above covalently linked to, for example, SPA-PEG5000 (Shearwater) ("PEGylation of hG-CSF and variants thereof in solution") is diluted with 20mM citrate buffer pH 2.5. The conductivity should be < 3.5 mS/cm. The pH was adjusted to 2.5 with dilute hydrochloric acid as necessary. The following buffers were used for the separation:

and (3) buffer solution A: 20mM sodium citrate, pH2.5 (equilibration and wash).

And (3) buffer solution B: 20mM sodium citrate, pH 2.5; 750mM sodium chloride (elution buffer)

The sample to be separated was loaded at a flow rate of 2ml/min onto an already equilibrated SP-sepharose HP column (7 ml). The column was washed with buffer a until a stable baseline was obtained by monitoring with a 280.

The multiple-pegylated species were isolated as follows: a linear gradient of 0-50% buffer B was applied to the column at a flow rate of 4ml/min for 180 minutes, and 2ml fractions were collected. The collected fractions were analyzed by SDS-PAGE and fractions with the desired number of conjugated PEG groups were pooled. This makes it possible to purify mixtures of pegylated G-CSF comprising those pegylated G-CSF initially coupled with e.g. 3-6 PEG groups, resulting in products with e.g. 4 or 5 coupled PEG groups, or products with only a single number of coupled PEG groups.

Example 19

Peptide mapping

The pegylation pattern of the G-CSF conjugates of the invention was determined by peptide mapping using a method similar to that of example 7, but based on trypsin degradation. In this case, the polypeptides were produced in CHO cells (see example 3) which have the substitutions K16R, K34R, K40R, T105K and S159K in comparison to the sequence of native human G-CSF. PEGylation with 5kDa SPA-PEG as described above gave modified proteins with a majority of 3, 4 or 5 PEG moieties and a minority of 6 PEG moieties. 5 of the 6 possible PEG attachment sites are known, and are the N-terminal amino groups Lys23, Lys105, Lys159 and His 170.

This peptide mapping analysis showed that the conjugated protein was fully pegylated essentially at the N-terminus and Lys105 and Lys159, but Lys23 was only partially pegylated. Although His170 was shown to be partially PEGylated in previous experiments, it is quite unexpected that this result was not found in this experiment. One possible explanation is that the bond between PEG and His170 residues is unstable during sample preparation performed prior to peptide mapping. Possible labile pegylation (as is the case here) can be avoided by substitution of a histidine residue for another residue, in particular for a lysine residue (when more stable pegylation is desired) or for a glutamine or arginine residue (when pegylation is desired to be avoided).

Example 20

In vivo biological activity in rats with chemotherapy-induced neutropenia

The in vivo biological activity of two pegylated G-CSF variants of the invention was tested in rats with chemotherapy-induced neutropenia. The variant has a sequence similar to that of SEQ ID NO: 1 in comparison with the amino acid sequences having the amino acid substitutions K16R, K34R, K40R, T105K and S159K (hereinafter "105/159"), and K16R, K34R, K40R, Q90K, T105K and S159K (hereinafter "90/105/159"), respectively. Both variants were produced in yeast (Saccharomyces cerevisiae), both coupled to SPA-PEG-5000 as described above. The two variants were tested for in vivo biological activity using a single dose, compared to unconjugated hG CSF (Neupogen *) and control (vehicle) using daily doses.

24 hours before administration of the G-CSF sample, rats were administered CPA at 50mg/kg body weight. The pegylated variants of the invention were administered at a single dose of 100 μ g/kg body weight at 0 hours, while Neupogen * was administered at a daily dose of 30 μ g/kg body weight for 5 days (from 0 hours to 96 hours).

In vivo biological activity was measured as described above ("in vivo biological activity of unconjugated and conjugated hG-CSF and variants thereof was determined in rats with chemotherapy-induced neutropenia"). The results are shown in fig. 8 (white blood cell count, WBC) and fig. 9 (absolute neutrophil count, ANC).

As shown in FIG. 8, 105/159, 90/105/159 and Neupogen * all resulted in increased leukocyte levels within the first 12 hours, after which the leukocyte levels were increased by chemotherapyThe drop reaches a minimum after about 48 hours. After 48 hours, the number of leukocytes increased in all three treatment groups, but the rate of increase was significantly greater in the group treated with the two pegylated variants of the invention than in the group treated with Neupogen *. White blood cell levels were normal (more than 10X 10) after 96 hours of treatment with the PEGylated variants 105/159 and 90/105/159⁹/L), whereas the Neupogen * treated group still had a white blood cell level below 10 x 10 after 120 hours⁹And L. Since the last treatment with 5 days Neupogen * was at 96 hours, the leukocyte levels of this group again declined after 120 hours. In contrast, leukocyte levels were relatively stable at 10X 10 from 96 hours up to the end of the 216 hour experiment in both groups treated with a single dose of the PEGylated variant of the invention⁹More than/L.

Similar changes in neutrophil number were seen in fig. 9, which shows that the rate of increase in neutrophil levels was significantly faster in the pegylated variant 105/159-treated group than in the Neupogen * -treated group (no detection of ANC in the 90/105/159 group).

Example 21

In vivo biological activity in rats with chemotherapy-induced neutropenia

Comparison of unconjugated hG-CSF (Neupogen *), hG-CSF with a single N-terminally attached 20kDa PEG group (Neulasta) in rats with chemotherapy-induced neutropenia^TM) And two pegylated G-CSF variants of the invention. These two variants were produced in yeast (Saccharomyces cerevisiae) and CHO cells, respectively, with identical amino acid substitutions relative to hG-CSF: K16R, K34R, K40R, T105K and S159K, and they are coupled to SPA-PEG 5000. The PEGylated variants of the invention, which initially consist of multiple PEGylated species with 3-6 PEG components, are isolated to give a more unitary product with only 4-5 conjugated PEG components. These variants are hereinafter referred to as "G20" (produced by yeast) and "G21" (produced by CHO cells).

G-CSF samples 24 hours after CPA administrationAdministration (90mg/kg body weight). PEGylated variants, Neulasta^TMG20 and G21 were administered in a single dose of 100. mu.g/kg body weight, Neupogen * was administered in a daily dose of 10. mu.g/kg body weight for 7 days.

In vivo biological activity was measured as described above (in vivo biological activity of unconjugated and conjugated hG-CSF and variants thereof was determined in rats with chemotherapy-induced neutropenia). The results are shown in fig. 10 (white blood cell count, WBC) and fig. 11 (absolute neutrophil count, ANC).

Figures 10 and 11 show that all G-CSF compounds induced the initial formation of leukocytes and neutrophils at nearly the same rate over the first 12 hours, followed by a drop in the levels of leukocytes and neutrophils as a result of chemotherapy. After 96 hours, the levels of leukocytes and neutrophils were again elevated in all cases, but the rate of elevation was significantly greater in G20 or G21 treated rats than in Neupogen * or neuasta^TMTreated rats. FIG. 10 shows that white blood cell levels were normal (about 10) after 144 hours in rats treated with G20 or G21⁹/L) via Neupogen * or Neulasta^TMThe treated rats did not reach this level of leukocytes after 168 hours. FIG. 11 shows a similar change in neutrophil count, i.e., G20 or G21 treated rats compared to Neupogen * or Neula sta^TMThe treated rats reached normal levels of neutrophil counts 24 hours earlier. It can thus be concluded that these PEGylated G-CSF variants of the invention are compatible with the existing G-CSF products Neupogen * or Neulasta^TMThe duration of chemotherapy-induced neutropenia can be shortened in rats by approximately 24 hours.

Sequence listing

<110> Makexi root stocks control company (Maxygen Holdings Ltd.)

<120> G-CSF conjugates

<130>0258

<150>US09/904,196

<151>2001-07-11

<150>DK PA 2002 00447

<151>2002-03-22

<150>DK PA 2002 00708

<151>2002-05-08

<160>15

<170>PatentIn version 3.1

<210>1

<211>174

<212>PRT

<213> human (Homo sapiens)

<400>1

Thr Pro Leu Gly Pro Ala Ser Ser Leu Pro Gln Ser Phe Leu Leu Lys

1 5 10 15

Cys Leu Glu Gln Val Arg Lys Ile Gln Gly Asp Gly Ala Ala Leu Gln

20 25 30

Glu Lys Leu Cys Ala Thr Tyr Lys Leu Cys His Pro Glu Glu Leu Val

35 40 45

Leu Leu Gly His Ser Leu Gly Ile Pro Trp Ala Pro Leu Ser Ser Cys

50 55 60

Pro Ser Gln Ala Leu Gln Leu Ala Gly Cys Leu Ser Gln Leu His Ser

65 70 75 80

Gly Leu Phe Leu Tyr Gln Gly Leu Leu Gln Ala Leu Glu Gly Ile Ser

85 90 95

Pro Glu Leu Gly Pro Thr Leu Asp Thr Leu Gln Leu Asp Val Ala Asp

100 105 110

Phe Ala Thr Thr Ile Trp Gln Gln Met Glu Glu Leu Gly Met Ala Pro

115 120 125

Ala Leu Gln Pro Thr Gln Gly Ala Met Pro Ala Phe Ala Ser Ala Phe

130 135 140

Gln Arg Arg Ala Gly Gly Val Leu Val Ala Ser His Leu Gln Ser Phe

145 150 155 160

Leu Glu Val Ser Tyr Arg Val Leu Arg His Leu Ala Gln Pro

165 170

<210>2

<211>525

<212>DNA

<213> Artificial

<220>

<221>misc_feature

<223> DNA encoding hG-CSF having codon usage characteristics of E.coli

<400>2

acccctctgg gcccggccag cagtctgcct cagagttttt tactgaaatg cttagaacag 60

gtgcgtaaaa tccagggcga tggcgcggcc ctgcaggaaa aactgtgcgc gacctataaa 120

ctgtgccatc ctgaagaact ggtcctgtta ggccatagct taggcatccc gtgggcgcct 180

ctgagtagct gcccgagtca ggccctgcag ctggccggct gcctgagtca gttacatagt 240

ggcttatttt tatatcaggg cttactgcag gcgttagaag gcattagtcc ggaactgggc 300

ccgaccctgg ataccttaca gttagatgtc gcggattttg ccaccaccat ttggcagcag 360

atggaagaat taggcatggc gcctgcgtta cagcctaccc agggcgccat gcctgcgttt 420

gcgagtgcgt ttcagcgtcg cgccggcggc gtgttagtgg ccagccatct gcagagcttt 480

ctggaagtga gttatcgtgt gttacgccat ctggcccagc cttaa 525

<210>3

<211>21

<212>PRT

<213> Escherichia coli (Escherichia coli)

<400>3

Met Lys Lys Thr Ala Ile Ala Ile Ala Val Ala Leu Ala Gly Phe Ala

1 5 10 15

Thr Val Ala Gln Ala

20

<210>4

<211>63

<212>DNA

<213> Artificial

<220>

<221>misc_feature

<223> DNA sequence encoding OmpA Signal sequence

<400>4

atgaaaaaga cagctatcgc gattgcagtg gcactggctg gtttcgctac cgtagcgcag 60

gcc 63

<210>5

<211>15

<212>PRT

<213> Artificial

<220>

<221>PEPTIDE

<222>(1)..(15)

<223> synthetic tags

<400>5

Met Lys His Gln His Gln His Gln His Gln His Gln His Gln Gln

1 5 10 15

<210>6

<211>45

<212>DNA

<213> Artificial

<220>

<221>misc_feature

<223> encoding SEQ ID NO: 5 DNA of marker

<400>6

atgaaacacc aacaccaaca tcaacatcaa catcaacatc aacag 45

<210>7

<211>30

<212>PRT

<213> human (Homo sapiens)

<220>

<221>SIGNAL

<222>(1)..(30)

<223>

<400>7

Met Ala Gly Pro Ala Thr Gln Ser Pro Met Lys Leu Met Ala Leu Gln

1 5 10 15

Leu Leu Leu Trp His Ser Ala Leu Trp Thr Val Gln Glu Ala

20 25 30

<210>8

<211>615

<212>DNA

<213> Artificial

<220>

<221>misc_feature

<223> encoding a polypeptide having the sequence of SEQ ID NO: 7 sequence of hG-CSF of Signal peptide

<400>8

atggccggcc ctgccacaca gtcccccatg aagctgatgg ccctgcagct gctgctgtgg 60

cactccgccc tgtggacagt gcaggaggcc acccctctgg gccccgccag ctccctgcct 120

cagtccttcc tgctgaagtg cctggagcag gtgagaaaga tccagggcga cggcgccgcc 180

ctgcaggaga agctgtgcgc cacatacaag ctgtgccacc ctgaggagct ggtgctgctg 240

ggccacagcc tgggcatccc ctgggcccct ctgtccagct gcccctccca ggccctgcag 300

ctggccggct gcctgtccca gctgcactcc ggcctgttcc tgtaccaggg cctgctgcag 360

gccctggagg gcatctcccc cgagctgggc cccacactgg ataccctgca gctggacgtg 420

gccgatttcg ccaccacaat ctggcagcag atggaggagc tgggcatggc ccctgccctg 480

cagcctaccc agggcgccat gcctgccttt gcctccgcct ttcagagacg ggccggcggc 540

gtgctggtgg ccagccacct gcagagcttt ctggaggtgt cctacagagt gctgcggcac 600

ctggcccagc cttga 615

<210>9

<211>6

<212>PRT

<213> Artificial

<220>

<221>PEPTIDE

<222>(1)..(6)

<223> synthetic tags

<400>9

His His His His His His

1 5

<210>10

<211>8

<212>PRT

<213> Artificial

<220>

<221>PEPTIDE

<222>(1)..(8)

<223> synthetic tags

<400>10

Met Lys His His His His His His

1 5

<210>11

<211>10

<212>PRT

<213> Artificial

<220>

<221>PEPTIDE

<222>(1)..(10)

<223> synthetic tags

<400>11

Met Lys His His Ala His His Gln His His

1 5 10

<210>12

<211>14

<212>PRT

<213> Artificial

<220>

<221>PEPTIDE

<222>(1)..(14)

<223> synthetic tags

<400>12

Met Lys His Gln His Gln His Gln His Gln His Gln His Gln

1 5 10

<210>13

<211>10

<212>PRT

<213> Artificial

<220>

<221>PEPTIDE

<222>(1)..(10)

<223> synthetic tags

<400>13

Glu Gln Lys Leu Ile Ser Glu Glu Asp Leu

1 5 10

<210>14

<211>8

<212>PRT

<213> Artificial

<220>

<221>PEPTIDE

<222>(1)..(8)

<223> synthetic tags

<400>14

Asp Tyr Lys Asp Asp Asp Asp Lys

1 5

<210>15

<211>9

<212>PRT

<213> Artificial

<220>

<221>PEPTIDE

<222>(1)..(9)

<223> synthetic tags

<400>15

Tyr Pro Tyr Asp Val Pro Asp Tyr Ala

1 5

Claims (26)

1. A polypeptide conjugate exhibiting G-CSF activity comprising a polypeptide that hybridizes to SEQ ID NO: 1, comprising at least one substitution selected from K16R/Q, K34R/Q and K40R/Q, and at least one substitution selected from T105K and S159K, or at a position that is substantially similar to SEQ ID NO: 1 has the above substitution at the corresponding position in the amino acid sequence having at least 80% sequence identity, and the conjugate has 2-6 polyethylene glycol moieties of molecular weight about 1000-10,000Da attached to the linking group of the polypeptide.

2. The polypeptide conjugate of claim 1, comprising 3, 4, or 5 of said substitutions.

3. The polypeptide conjugate of claim 1 or 2, further comprising the substitution H170K, H170Q, or H170R.

4. The polypeptide conjugate of any one of claims 1 to 3, comprising a substitution selected from the group consisting of: q70 + S159, Q70 + H170, Q90+ S159, Q90+ H170, T105+ S159, T105+ H170, Q120 + S159, Q120 + H170, T133 + S159, T133 + H170, S159 + H170, Q70 + Q90+ S159, Q70 + Q90+ H170, Q70 + T105+ S159, Q70 + T105+ H170, Q70 + Q120 + S159, Q70 + Q120 + H170, Q70 + T133 + S159, Q70 + T133 + H170, Q70 + S159 + H170, Q90+ T105+ S159, Q90+ T105+ H170, Q90+ Q120 + S159, Q90+ Q120 + H170, Q90+ T159, Q90+ H170, Q133 + S159, Q90+ S159 + H170, Q105 + S159 + H170, T105+ H105, Q90+ S120 + S159, Q90+ Q120 + H159, Q90+ T120 + S159 + H170, Q90+ T + S159, Q90+ T + S159 + Q70 + T105+ S159, Q90+ S159 + H170, Q90+ T + S159 + Q70 + S159 + T + S159, Q70 + H + S159, Q90+ S159 + H170, Q90+ S + T + S159 + S + H + S159 + H + S159 + S + H, Q90+ S + H + S, Q70K + T105K + Q120K + H170K, Q70K + T105K + T133K + S159K, Q70K + T105K + T133K + H170K, Q70K + T105K + S159K + H170K, Q70K + Q120K + T133K + S159K, Q70K + Q120K + T133K + H170K, Q70K + T133K + S159K + H K, Q90K + T105+ K + Q105 + S159K + S K + K, Q90K + T K + Q105 + T K + Q K + H170 + S K + T K + H K + T K + H K + T K + H K + 36.

5. The polypeptide conjugate of any one of claims 1 to 4, comprising a substitution selected from the group consisting of: K16R + K23R, K16R + K34R, K16R + K40R, K23R + K34R, K23R + K40R, K34R + K40R, K16R + K23R + K34R, K16R + K23R + K40R, K23R + K34R + K40R, K16R + K34R + K40R and K16R + K23R + K34R + K40R.

6. The polypeptide conjugate of claim 5, comprising the substitutions K16R + K34R + K40R and comprising a lysine residue at position 23.

7. The polypeptide conjugate of any one of claims 1 to 6, wherein the position of the substitution corresponds to a position corresponding to SEQ ID NO: 1, a corresponding position in an amino acid sequence having at least about 90% sequence identity.

8. The polypeptide conjugate of claim 7, wherein the position of the substitution corresponds to a position that is identical to SEQ id no: 1, such as at least about 96%, 97%, 98%, or 99% sequence identity.

9. The polypeptide conjugate of any one of claims 1 to 8, wherein the polyethylene glycol component is attached to at least one lysine residue and may also be attached to the N-terminal amino group.

10. The polypeptide conjugate of claim 1, comprising a polypeptide that hybridizes to SEQ ID NO: 1 comprising a substitution selected from the group consisting of K16R, K34R, K40R, T105K and S159K, and said conjugate has 2-6 polyethylene glycol moieties of molecular weight about 1000-10,000Da attached to the linking group of the polypeptide.

11. The polypeptide conjugate of any one of claims 1-10, which is glycosylated at position T133.

12. The polypeptide conjugate of any one of claims 1 to 11 having 3 to 5,4 to 6, 3 to 4,4 to 5 or 5 to 6 PEG components attached.

13. The polypeptide conjugate of claim 12, having attached 3-6, such as 4-5, polyethylene glycol components with a molecular weight of about 5000-6000 Da.

14. The polypeptide conjugate of claim 13, having attached thereto 4 polyethylene glycol moieties of molecular weight about 5 kDa.

15. The polypeptide conjugate of claim 13, having attached 5 polyethylene glycol moieties of molecular weight about 5 kDa.

16. The polypeptide conjugate of any of claims 1-15, having an in vitro biological activity of about 2-30% of the biological activity of unconjugated hG-CSF, as determined by the luciferase assay described herein.

17. A polypeptide conjugate exhibiting G-CSF activity comprising a polypeptide comprising an amino acid sequence substantially identical to SEQ ID NO: 1 and at least one non-polypeptide moiety attached to the linking group of the polypeptide, the in vitro biological activity of the conjugate, as determined by the luciferase assay described herein, is about 2-30% of the biological activity of unconjugated hG-CSF.

18. The polypeptide conjugate of claim 17, which binds to SEQ ID NO: 1 compared to at least one amino acid change at an amino acid position selected from the group consisting of positions 11-41 (helix A), positions 71-95 (helix B), positions 102-125 (helix C), and positions 145-170 (helix D).

19. The polypeptide conjugate of claim 17 or 18, comprising 2-6 polyethylene glycol moieties of molecular weight about 1000 and 10,000Da attached to the linker group of the polypeptide.

20. A method of making a G-CSF conjugate that has reduced receptor-mediated clearance, and/or reduced duration of neutropenia, as compared to hG-CSF, comprising making a polypeptide having an amino acid sequence substantially identical to SEQ ID NO: 1, and the linker of said polypeptide is attached to at least one non-polypeptide moiety, and the in vitro biological activity of the resulting conjugate, as determined by the luciferase assay described herein, is about 2-30% of the biological activity of unconjugated hG-CSF.

21. The method of claim 20, wherein the desired in vitro biological activity is produced by, relative to the amino acid sequence of SEQ ID NO: 1, at least one amino acid residue, typically a substitution, at an amino acid position selected from the group consisting of positions 11-41 (helix A), positions 71-95 (helix B), positions 102-125 (helix C), and positions 145-170 (helix D), and coupling at least one non-polypeptide moiety to said at least one altered amino acid residue.

22. A composition comprising the polypeptide conjugate of any one of claims 1-19 and at least one pharmaceutically acceptable carrier or excipient.

23. A method of treating a patient having an insufficient level of neutrophils comprising administering to a mammal in need of such treatment a therapeutically effective amount of a polypeptide conjugate according to any one of claims 1-19 or a composition according to claim 22.

24. Use of the polypeptide conjugate of any one of claims 1 to 19 as a medicament.

25. Use of a polypeptide conjugate according to any one of claims 1 to 19 for the preparation of a pharmaceutical composition for the treatment of a deficient level of neutrophils.

26. The use of claim 25, wherein the pharmaceutical composition is for the prevention and/or treatment of neutropenia due to chemotherapy or radiation therapy, or due to HIV or another viral infection.